Sandy Beaches as Ecosystems

Developments in Hydrobiology 19

Series editor
H.J. Dumont

SPRINGER-SCIENCE+BUSINESS MEDIA, B.V. 1983

Sandy Beaches as Ecosystems

Based on the Proceedings of the First International Symposium

on Sandy Beaches, held in Port Elizabeth, South Africa,
17-21 January 1983

Edited by
Anton McLachlan and Theuns Erasmus
Zoology Department, University of Port Elizabeth
South Africa

Editorial Panel:
D.H. Swart (Physical aspects)
G.A Eagle (Chemical aspects)
A McLachlan (Ecology)
AC. Brown (Ecophysiology)
AE.F. Heydorn (Management)
F. Herbst (Language)

SPRINGER-SCIENCE+BUSINESS MEDIA, B.V. 1983

Library of Congress Cataloging in Publication Data

ISBN 978-90-481-8521-4 ISBN 978-94-017-2938-3 (eBook)

DOI 10.1007/978-94-017-2938-3

Cover design: Max Velthuijs

Copyright
© 1983 by Springer Science+Business Media Dordrecht
Originally published by Kluwer Academic Publishers in 1983
Softcover reprint ofthe hardcover lst edition 1983
AlI rights reserved. No part of this publication may be reproduced, stored in a
retrieval system, or transmitted in any form or by any means, mechanical,
photocopying, recording, or otherwise, without the prior written permission of
the publishers,
Springer-Science+Business Media, B.V.
 CONTENTS

Introduction by A. McLachlan and T. Erasmus, Editors ......................................... .

Part One: Physical Aspects

1. Physical aspects of sandy beaches - a review by D.H. Swart .........•..................... 5

2. Sediment reworking on sandy beaches by D.M. Chapman ...................................... 45
3. Beach changes on coasts with different wave cl imates by D.G. Aubrey....................... 63
4. Provenance of bNch sediments in south-eastern Australia by E.C.F. Bird ..... ......... .... 87
5. Properties of logarithmic spiral beaches with particular reference to Algoa Bay by
J.M. Bremner 97
6. Beach and nearshore habitats as a function of internal geometry, primary sedimentary
structures and grain size by B.W. Flemming and A.H. Fricke .............................. . 115
7. Physical variability of sandy beaches by A.D. Short and L.D. Wright ..................... . 133
8. Sediments and structures in beach-nearshore environments, South East Australia by
A.D. Short .............................................................................. . 145
9. A theoretical model of surf-zone circulation and diatom growth by D.F. Winter ........... . 157
10. Holocene coastal development in the NW part of the Netherlands by E.F.J. de Mulder 169
11. Wave-generated water flow through a porous sea bed by D.H. Swart and J.B. Crowley 177
12. Sedimentary aspects of the Mvumase project by J. Nicholson .............................. . 191
13. Physical aspects of sandy beaches - Workshop report by D.H. Swart ....................... . 199

Part Two: Chemical Aspects

14. The chemistry of sandy beach ecosystems - a review by G.A. Eagle ......................... 203
15. Nutrient cycling in sandy beaches by K.B. Pugh .....................................•..... 225
16. The effect of meiofauna and bacteria on nutrient cycling in sandy beaches by
H.F-K.O. Henning, A.H. Fricke and C.T. Martin ............................................ 235
17. Carbon flow and nutrient regeneration from the decomposition of macrophyte debris in a
sandy beach mi crocosm by K. Koop and M. I. Lucas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249
18. The sandy beach area of Kiel Fjord and Kiel Bight (western Baltic Sea) - a structural
analysis of a shallow water ecosystem by M. BHlter and M. Meyer .......................... 263
19. Fouling of the sandy beaches of Nahant Bay (Massachusetts, USA) by an abnormal free-living
form of the macroalga Pilayella littoralis (Phaeophyta). I Habitat characteristics by
A.V. Quinlan, T. Lewis and T.K. Hoyt ................................ ............•...•.... 271
20. Fouling of the sandy beaches of Nahant Bay (Massachusetts) by an abnormal free-living form
of the macroalga Pilayella littoralis (Phaeophyta) II Population characteristics by
R.T. Wilce and A.V. Quinlan 285
VI

21. Effects of fresh water and of pollution from a marine oil refinery on the fauna of a sandy
beach by A.C. Brown .............•......•.•....•.••••.••..•..•.•.•..••.•...••.......•.... 297
22. Donax serra and BuUia rhodostoma - possible bio-indicators of trace metal pollution on
sandy beaches with particular reference to the south-eastern Cape by H.R. Watling and
R.J. Watling •.•.••....•.••...••..........•.....•.•.•.••.....•••.••.•......•.•.•.•..•...• 303
23. Sandy beaches as ecosystems : chemical aspects - workshop report by D.A. Lord and
G.A. Eagle .•••.•....•..••..••.•••...••••.••..••..•..••..••••••......•..•.•.•..••••.•••.• 315

Part Three: Ecology

24. Sandy beach ecology - a review by A. McLachlan 321

25. The role of phytoplankton in surf ecosystems by T. Lewin and C.T. Schaefer 381
26. Factors affecting the distribution of organisms in the intertidal zones of sandy
beaches by R. Bally .•..•••.••.••...••..••..••..••..••..•..••...•.•...•..•••••••.....•.. 391
27. Dynamic zonation of staphylinoid beetles (Coleoptera: Staphylinoidea) on a sandy
beach in east Africa by G. Chelazzi, L. Chelazzi and S. Focardi 405
28. A preliminary account of the ecology of sandy beaches in southern Brazil by N.M. Gianuca. 413
29. Selective microhabitat colonisation by interstitial meiofauna as a function of grain size
by A.H. Fricke and B.W. Flemming •..•.•••.••...•.•••••••••.••.......••..•..•...••...•.. 421
30. Two graphical display methods for ecological data matrices by L.G. Underhill 433
31. Ecological characteristics of sandy beaches in the southern California Bight by
D. Straughan .•........•.......•...•..••.••..••..••....•••••..•.•......•••..••..•....... 441
32. Ecology of beach and surf zone mysid shrimps in the eastern Cape, South Africa by
T.H. Wooldridge ... .••.....•.••... .... ..•. .••..••. .•... ....•.•.• .•.••.... ... .•.••....... 449
33. Community structure of intertidal sandy beaches in New South Wales, Australia by
D.M. Dexter •..•••...•..•..•...•..........••.•..•...••.•••.••...•..•••..•..•••.•••..•... 461
34. The species area relationship on a sandy beach by R.G. Hartnoll 473
35. Interaction between coastal plankton and sand mussels along the Cape coast. South Africa
by L. Hutchings, G. Nelson, D.A. Horstman and R. Tarr 481
36. The impact of surf zone fish communities on faunal assemblages associated with sandy
beaches by T.A. Lasiak ......•......•.......••..•...•.•..•..•...•••.••...•..•.•.•....••. 501
37. Ecological structure and energy requirements of the sandy beach avifauna of
southern Africa by P.A.R. Hockey, W.R. Siegfried and A.A. Crowe .•••......•••.•••.•...•• 507
38 .. Subtidal sandy beachtrophicstructure in the area of Punta Moron, Venezuela by
P. E. Penchaszadeh . . . . • • . • . • . • • . . • • . . • . . . . . . . . • . . . • • . . . • . • • • • . • . • . • • . . • . • . . . • . . • • . • • • • . • 523
39. Food web in the surf zone of an exposed sandy beach along the mid-Atlantic coast of
the United States by J.J. McDermott •.••..••..••.•••..••.••••.•••.....•••..•.••.•••...•• 529
40. The ecology of sandy beaches in the eastern Cape. South Africa by A. McLachlan .••...•.. 539
41. Kelp wrack and the flow of energy through a sandy beach ecosystem by C.L. Griffiths,
J.M.E. Stenton-Dozey and K. Koop 547
42. The fauna associated with kelp stranded on a sandy beach by J.M.E. Stenton-Dozey and
C.L. Griffiths ..•••.••..••..•...••...••..••.•.•...••..••••.•....•..••..••..•.•.•.•....• 557
 VII

43. Sandy beach ecology - workshop report by A. McLachlan and G.C. Bate 569

Part Four: Ecophysiology and Autecology

44. The ecophysiology of sandy beach animals - a partial review by A.C. Brown 575
45. The biology of the genus Donax by A.D. Ansell 607
46. Consumption, assimilation and energy balance in the three-spot swimming crab, OValipes
punctatus (de Haan)(Crustacea; Brachyura) by H.H. du Preez ............•.............•.... 637
47. Population ecology and biology of Dotilla sulcata (Crustacea, Ocypodidae) typical for
sandy beaches of the Red Sea by L. Fishelson ...................••........................ 643
48. Ecology of the sandy beach gastropod Mazatlania aciculata in Quizandal (Carabobo,
Venezuela) by P.E. Penchaszadeh, G. de Mahieu, V. Farache and M.E. Pera ............. ..... 655
49. Production ecology of Haustorius canadensis (Amphipoda: Haustoriidae) in southern Maine by
T.E. Donn and R.A. Croker ...................................................... .......... 661
50. Behaviour and physiological responses of a burrowing bivalve to stress by E.R. Trueman 669
51. Some aspects of the ecophysiology of Scaevola thunhergii, a subtropical coastal dune
pi oneer by N. W. Pammenter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . • . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 675
52. Energetic values ln interstitial isopods and amphipods from sandy beaches as a function
of body size and season (western Mediterranean) by N. Coineau ................•........... 687
53. A synopsis of community respiration studies on exposed sandy beaches by A.H. Dye ......... 693
54. Sandy beach ecophysiology - workshop report by A.C. Brown and R. Bally................... 699

Part Five: Management

55. Management of sandy coastlines - report on review and workshop by A.E.F. Heydom 703
56. Factors influencing beach erosion and accretion: ~ global review of E.C.F. Bird 709
57. Accidental formation and subsequent disappearance of a contaminated beach: a case history
of environmental management by D.V. Ellis 719
58. Monitoring beach and dune advancement and vegetation changes 1937-1977 at the farm
Twinstreams, Mtunzini, Natal, South Africa by P.J. Weisser and A.P. Backer 727
59. Management for survival: a review of the plant ecology and protection of the 'Machair'
beaches of north-west Scotland by R.E. Randall........................................... 733

Part Six: Abstracts

60. Mud accumulation on a microtidal open ocean beach by L.R. Martins 743
61. Beach and river-mouth processes, Natal coast, South Africa by I.L. van Heerden ...... ..... 744
62. Beach barrier sedimentation, Chandeleur Islands, Louisiana by I.L. van Heerden .. .........
• 745
63. Bacteria-meiofauna relationships in a subantartic sand beach (Kerguelen) by
F. de Bovee, G. Cahet, D. Delille and J. Soyer .•................................•........ 746
64. Community structure of Ichthyoplankton off sandy beaches in Algoa Bay, South Africa, by
L. E. Bec k1ey ... . . . . . • . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 747
VIII

65. Features of some sandy estuary systems by J.R. Grindley and Y von Shirnging 747
66. Variation in Sanderling flock size and structure on a southwest Cape sandy beach by
A. Crowe .....................................................•.......................... 748
67. The importance of non-teleost fishes (Elasmobranchs) in the surf zone with special
reference to Rhinobatos annulatus by G.J. Rossouw 749
68. Artificial sandy beaches and environmental impacts due to dumping of copper mine
tailings at Chanaral area, Chile by J.e; Castilla .....•................................. 749
69. Preliminary investigation of surf zone phytoplankton blooms occurring off the Sundays
River beach in Algoa Bay by D.S. Sloff .....................................•............ 750
70. Factors influencing the discontinuous distribution of the lugworm - Arenicola marina (L.)
on the beach of the North Sea (Belgium) by S. Claus and A.F. de Bont ..•................• 751
71. Ecophysiologycal aspects of the genus Donax I. Environmental factors correlated with
absolute and relative densities by L. Neuberger and G. de Mahieu .......•........ ........ 751
72. Ecophysiologycal aspects of the genus Donax II. Filtration rate in Donax denticulatus
(L.) by L. Neuberger and G. de Mahi eu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 752
73. The Gamtoos - an example of beach/estuary interaction by T.J.E. Heinecken and
A.E.F. Heydorn ...................................................•....................... 753
74. The implications of resource partitioning for the structure of a sand-beach meiofauna
community by R.M. Warwick ........................................................•....•. 753
75. Dune: A geobotanical model of foredune development by D.J. Disraeli 754
76. Some aspects of driftsand reclamation in the Republic of South Africa by R. Reyneke
and M. Burns 755
77. The sandy Haifa acre plain as a typical phyto-ecosystem of Israel's coast by
D.E. Tsuriell 756
 INTRODUCTION
What sight is more beautiful than a high-energy beach facing lines of rolling white
breakers? What battleground is more ferocious than where waves and sand meet? What
environment could be more exciting to study than this sandy interface between sea and
land? And yet how much do we know about sandy beaches?
Open sandy beaches are amongst the most neglected fields of scientific study in the
coastal environment. This situation exists despite their great extent along most
temperate and tropical coastlines and their value as recreational areas and buffer
zones against the sea. The traditional oceanographer does not venture into the surf
zone while the terrestrial ecologist stops short at the high water mark. Only a few
coastal engineers have grappled with the problem of sand and sediment movement as it
influences their construction of harbours and pipelines. The marine biologist on the
other hand has regarded estuaries, coral reefs and rocky shores, obviously teeming
with life, as more fruitful areas for study than the apparently poor animal life on
sandy beaches. Sandy beaches have therefore tended to become a scientific no man's
land.
Over the last decade this situation has begun to improve. Recent work on high-energy
beaches has revealed that they may in fact be rich and productive and fertile areas
for study. It has even been suggested that beaches and their adjacent surf zones may
constitute viable marine ecosystems. within this framework the planning for an
international symposium on sandy beaches was initiated. The symposium had as its
major aims (1) to bring together all scientists studying sandy beaches (2) to
encourage a holistic systems approach and interdisciplinary interaction and (3) to
review our present state of knowledge of beaches and provide guidelines for future
research.
The holding of the symposium and publication of these proceedings have been made
possible by the support and assistance of several institutions and many individuals.
In particular, the Council for Scientific and Industrial Research, the Department of
Environmental Affairs and Fisheries and the University of Port Elizabeth, South
Africa, provided administrative and financial support. Further financial support
also came from the South African Nature Foundation, The Anglo American Chairmans
Fund, South African Airways and Avis Car Hire. Amongst many enthusiastic helpers and
participants we would particularly like to thank Mickey Meyer, Andre van der
Westhuysen, Des Lord and the sandy beach research team at the Uni versi ty of Port
Elizabeth.
The symposium was divided into five sessions covering physical aspects, chemical
aspects, ecology, ecophysiology and management topics relating to sandy beaches.
Each session is covered in one part of this volume in the form of an invited review
paper, several plenary papers and a report on a workshop/discussion session.
Abstracts of some papers not published in full are also included. This volume should
fill a big gap in beach literature and will, we believe, become a standard reference
for future work.

ANTON McLACHLAN 18 May 1983, Port Elizabeth, South Africa

THEUNS ERASMUS
 3

PART ONE

PHYSICAL ASPECTS

PHYSICAL ASPECTS OF SANDY BEACHES - A REVIEW
D H SWART (Sediment Dynamics Division, NRIO, CSIR, POBox 320, Stellenbosch 7600, South Africa)
1. INTRODUCTION
1.1 General
Sound management of the coastal zone requires a
sound understanding of all aspects of the eco-
system, namely,
physical
chemical
ecological
ecophysiological
Physical processes should be well understood for
two main reasons,
(1) to allow the judicious management of
physical resources; and
(2) to facilitate the understanding of the
behaviour of chemical/biotic "resources"
to allow, in turn, their proper manage-
ment.
A review of the physical aspects of sandy beaches
is in reality a review of our knowledge, or lack
thereof, of nearshore water and sand movement
and of their interactions on sandy beaches. In
the context of this Symposium it does not mean
that this paper is a review of the state of the
art of all the relevant processes, including ways
in which each process can be quantified, but
rather that all relevant processes are described
qualitatively and that any interrelationships
are pointed out. This review will therefore not
supply coastal engineers or physical oceano-
graphers with all the information they need for
predictive work; it will rather make reference to
and attempt to aid in the understanding of the
5
processes which can form the basis of detailed
study into any given process. As such it will
be of interest to researchers from all disci-
plines who are active in work on sandy beaches,
including also coastal engineers and physical
oceanographers.
1.2 Physical processes
As stated above, physical processes in the near-
shore zone can be categorized into two main
classes, namely, water movement and sand movement.
Water movement
When studying water movement, it is possible to
distinguish between water movement in the water
column and water movement in the sandy bed. The
movement of water in the water column includes
such aspects as surface wave motion, wave-gene-
rated nearshore currents, tide-generated cur-
rents and wind-generated currents. These aspects
will be dealt with in the review and brief mention
will also be made of the water movement in the
vicinity of estuary mouths. Water flow in the
sandy bed is also discussed and a clear distinc-
tion is made between flow in the intertidal zone
due to groundwater seepage and that in the infra-
tidal zone due to wave pumping.
Sand movement
The different modes of nearshore sand movement
are discussed. The mechanisms of longshore and
onshore-offshore wave-ge,herated sand movement are
outlined and it is explained how the motions
cause shoreline reworking, modifications to bay
6

configurations and beach profile reworking. A

brief review of beach control devices will also
SAY -+- 8m
be given. The basic processes will be sum-

r
domInont bftgk.,. line
marized qualitatively and indications given
regarding quantification of the processes with
specific emphasis on variations and/or vari-
ability, whether spatial or temporal.
SAY APPR. - ISm

Aspects which have a bearing on the formation

and behaviour of beaches which are excluded from CROSS-SECTION

this review because they are study fields in their

own right, and because very good literature is mov1no
sand
orolns
available on each of them, are wind and weather
LOW WAVES
in general as well as the large-scale formation
'of bays between headlands.

HIGH WAVES
1.3 Area under consideration

Although it is not recommended that a strict

zonation should be introduced in the study of
LOW WAVES
coastal processes, it is relevant to consider
briefly the area under review.
FIGURE 1. Area under consideration

There are various possible definitions of what

However, another option would be to choose as
belongs to the beach. One possibility is to say
seaward limit the outer edge of any shore-
that the beach is that area which can be ac-
connected .circulation cells. The same argument
tively reworked by wave action. This definition
as to changes to the wave motion offshore of
will coincide with the dynamic swept prism, as
this point apply again.
indicated in Figure 1. On the other hand, it is
quite possible that artificial manipulations of
Whatever definition is adopted, it is essential
the beach profile below the lower limit of the
that all those processes affecting the coast
dynamic swept prism, such as offshore dredging,
be studied.
might sufficiently alter the wave processes so
as, in time, to effect changes in the beach.
A second point which needs consideration is the
Similarly, the shoreward limit of the dynamic
definition of "sandy". In the context of this
swept prism might fall within the primary dune,
review any beach consisting of bed material with
whereas human activity on the landward face of
a median grain size in the range from 50 ~m
the dune might cause breaching which eventually
(0,050 mm, which is exceptionally fine) to
will lead to a different landward extremity of
2 000 ~m (2,0 mm, which is exceptionally coarse)
the dynamic swept prism. The above considera-
will be regarded as a sandy beach. Mud beaches
tions clearly indicate the difficulties in
and gravel beaches are not discussed.
choosing limits for a beach and the inherent
folly in doin~ so.

2. WATER MOVEMENT
2.1 General
Water movement in the context of the present
paper is defined as the motion of surface
gravity waves, secondary currents induced by
these waves, wind-induced currents and tidal
currents. Of these water motions the first two
are the most important.
Water serves as carrier for all sorts of parti-
culate, cohesive or organic matter, chemical
substances and biota. As such the study of the
fate of any of these revolves around the study
of water movement. As a result the study of
water movement as a part of physical and bio-
logical studies is becoming increasingly im-
portant. This section will briefly review the
various elements of water movement in a des-
criptive manner without actually going into
the mathematics governing them. The emphasis
will be mainly on wave motion and wave-induced
phenomena although wind- and tide-induced
currents will be discussed briefly at the end of
the section.
2.2 Factors governing dynamics of regular waves
The stress of the wind on any water surface ge-
nerates waves. These waves are oscillary and
may be defined by their height, length and period.
The wave height R is the vertical distance be-
tween the top of the wave crest and the bottom
of the wave trough; the wave length A is the
horizontal distance between successive crests and
the wave period T is the time taken for successive
crests to pass a fixed point. The wave steepness
is the ratio R/A and the wave speed or celerity C
of the waves is related to the period and the wave
length such that C = A/T. As waves propagate in
water the progressive movement of water particles
is very slight; it is the shape and wave energy
that move forward.
7
The shape and general dynamics of waves are
determined by the conservation of momentum in the
horizontal and vertical planes and by conservation
of mass. To enable the appropriate equations to
be solved, it is customary to make certain sim-
plifying assumptions. For example, it is usually
assumed that only non-breaking waves are con-
sidered, that the water movement is two-
dimensional, that the water depth is constant,
that the flow is frictionless and incompressible,
that surface tension is neglected and that the
waves are periodic and translate with a constant
celerity.
A solution to these governing equations can be
found if three conditions are specified along
the boundaries of the fluid, namely, particles
on the free surface and on the bed remain on the
free surface and bed, respectively, and the
pressure at the free surface just inside the
fluid is atmospheric.
Even the solution of simplified governing equa-
tions obtained in this manner is not simple,
since the boundary conditions which have to be
specified in order to solve the governing equa-
tions need to be defined for the boundaries,
which are the bed and the free surface. The shape
of this latter boundary is unknown since the
whole purpose of solving the governing equations
is to find the wave shape. Therefore an itera-
tive procedure is required to find a solution.
There are numerous ways in which the boundary
conditions can be simplified to ease the way
of solving this wave boundary-value problem.
Each different way leads to a different solu-
tion and because of the iterative nature of the
problem all such solutions are in some way ap-
proximate, although the approximation in some
of the more sophisticated theories can, for all
practical purposes, be neglected.
8
The simplest way of finding a solution to the
wave boundary-value problem is to assume that the
free-surface boundary conditions do not apply at
the free surface but at the mean water surface.
In this manner a wave with a sinusoidal wave
shape is obtained for which the wave crest and
actual wat. lurfac. crllt
Tw...
---~i~t
FIGURE 2. First-order (linear) solution.
wave trough are equidistant from the mean water
surface. It is obvious that this solution is
obtained by means of a very crude assumption,
since the mean and actual water surfaces differ
substantially, especially for waves which are
not of negligible height as compared with the
water depth. Although this sinusoidal wave
theory (commonly called linear wave theory) is
therefore valid for only small-amplitude waves
in deep water it is the wave theory which is
most commonly used in all applications in shallow
and'deep water.
Comparisons between predictions obtained by using
the linear wave theory and actual measurements
have shown that the theory is indeed a good re-
presentation of real conditions in deep water
but that the comparison between theory and data
becomes increasingly poor as the water depth
decreases.
Stokes (1847) used the sinusoidal water surface
obtained for linear theory as the surface along
which to satisfy the free-surface boundary con-
ditions. In this manner he obtained a second-
order correction to the first-order linear wave
theory which consists of another sinusoidal wave
which has a ~ave length which is exactly half
that of the original linear wave solution. The
height of this second-order wave is substantially
lower than that of the linear wave, provided that
the water depth is sufficiently large.
~-- flrat-orderaolullon
+ ..cond-order correction
II actual wat.r lurfac.
FIGURE 3. Second-order solution
The second-order correction causes the wave crests
to become slightly more peaked and the troughs
to become flatter. It is obvious that the dif-
ference between the surface along which the
boundary conditions were assumed to apply and
the actual water surface is less than for the
linear wave theory. However, in very shallow
water for the case of waves with a height of the
same order of magnitude' as the water depth, the
second-order correction "explodes", leading to
unrealistic wave shapes.
In this manner it is possible to increase the
order of the solution step by step by each
time using the water surface of the previous
solution of the free surface which is one order
higher than the previous one. The number of
mathematical manipulations required to do this
increases substantially as the order of the
solution becomes higher. The highest order
theory obtained in this manner was the twelfth-
order theory derived by Wilton (1914).
As the order of the theory increases the dif-
ference between the actual wave shape and the
previous wave shape becomes less while at the

same time the wave crest becomes progressively
more pronounced and the trough becomes flatter.
This group of theories is called the Stokesian
wave theories and although they are relatively
sound for deep and intermediate water depths
(up to a water depth to wave length ratio of
approximately 0.05 to 0.1) they are not recom-
mended for use in shallow water because of the
instability of some of the higher-order terms.
This happens because all Stokes ian wave theories
are in essence small-amplitude wave theories,
which means that the wave height should always
be small compared to the water depth.
Another class of theory is obtained by assuming
that the water depth to wave length ratio (d/A)
and the wave height to water depth ratio (H/d)
are both much less than one. This implies that
small-amplitude waves in shallow water are con-
sidered. In a manner analogous to the Stokesian
theories it is again possible to obtain solutions
of a different order. This is done by truncating
expansions of the governing equations and boundary
conditions in different ways. The first solution
of this type was obtained by Keulegan and
Patterson in 1896. The wave shape is expressed
in terms of elliptical functions and this class
of theory is called the cnoidal theories. The
correspondence of these theories with real data
is best in shallow water but decreases as the
water depth increases. In deep water the cnoidal
solution does not apply since the wave length
becomes imaginary.
With the improvement in the capabilities of
modern computers over the last few decades, it
has become possible to develop a new class of
theory, namely, the numerical wave theories.
One of the first of this type was the Stream-
function wave theory of Dean (1965). In this
class all equations are written as infinite
series and a solution is obtained theoretically
9
in such a way that every condition is exactly
met except one of the two free-surface boundary
conditions, mostly the dynamic free-surface
boundary condftion, that is, the pressure just
within the fluid equals the atmospheric pressure.
The error in the one remaining unsatisfied boun-
dary condition is then minimized by adding more
terms. In the process a solution is obtained
for the unknown parameters. This solution is
complete except that a series of parameters which
define the solution are as yet unknown. In Dean's
theory the number of terms varies from four to
nineteen as the relative water depth (water
depth to wave length) decreases from deep to
shallow water. On the other hand Cokelet (1977)
developed a theory which always contains 110
terms. Because of the approach adopted this
class of theory is very accurate in all relative
water depths for non-breaking waves. Their only
real disadvantages are (1) that they cannot
really be applied by hand (pocket calculator)
due to the inconvenience of adding numerous
terms to obtain the solution and (2) these
theories predict wave breaking at wave height
to water depth ratios (H/d) which do not exceed
0,9 and which cannot be applied in the many
cases in which a higher value of H/d does occur.
Then there is a last class of wave theory in
which the theories are based in some way or
another on knowledge of the definite advan-
tages of certain existing theories. There are
two such types of theory.
In the first type one or more aspects of an ex-
isting theory are modified, either to improve
the accuracy of prediction of the particular as-
pect(s) or to simplify the application of a re-
latively good theory which is difficult to apply
due to a complex formulation. An example of an
improved theory is that of Goda (1964) in which
only the formulation of the horizontal orbital
velocity under the wave crest was modified to
10
bring this property into closer correspondence
with real data. An example of a simplified theory,
on the other hand, is that of Van Hijum (1972),
in which cnoidal theory was simplified by ap-
proximating the elliptical functions with simple
trigonometric functions. Theories of this type
do not usually adhere to the free-surface boun-
dary conditions.
The second type of theory is obtained by using
the knowledge of advantages of existing theories
to reformulate the governing equations and boun-
dary conditions which are then in them solved
from first principles. The best example of this
type of theory is the Vocoidal water-wave theory
(Swart, 1978). Assumptions are made about the
qualitative description of the wave shape, wave
celerity and orbital motion and are then used
to modify the governing equations. A solution
is then found from first principles for three
parameters which quantify the wave shape, wave
celerity and orbital motions in such a way that
all the boundary conditions except for the free-
surface dynamic boundary condition are adhered to
exactly, and the error in this last condition is
kept to an absolute minimum. The resulting theory
is so simple to apply that any wave property can
be calculated by pocket calculator. It also
shows an exceptionally close correspondence with
real data in all relative water depths.
Swart, Loubser (1979) performed the most com-
prehensive comparison of wave theories to data
done to date when they compared the results from
the thirteen wave theories which are used most
frequently with hundreds of sets of data which
represent all basic data available in literature.
This comparison showed that the theories in
closest agreement with data on wave profile,
celerity and orbital motion are Vocoidal theory
and Dean's numerical stream function theory.
It was, however, found that the application of
the latter is restricted due to the limited wave
breaker height predicted by Dean's theory. So,
for example, it was found impossible to compute
properties in Dean's theory for more than twenty
per cent of the available data sets because
of wave breaking.
In their study Swart and Loubser also investi-
gated the adherence of the theories to the boun-
dary conditions. Again it was found that the
two best theories are Vocoidal theory and Dean's
stream function theory, with Vocoidal theory
actually having a better correspondence than
Dean's theory. The difference between these two
was nevertheless small. Cokelet's numerical
theory has a better adherence to the boundary
conditions than both the above-mentioned theories
but could not be tested in the study since it is
so difficult to apply.
Since Vocoidal theory is simple to apply and is
in good correspondence with theory (boundary
conditions) and data for all relative water
depths, efforts are at present being made to
extend the theory to allow it to be applied to
a wide variety of wave-induced phenomena which
can at present be computed only by means of
linear wave theory.
A "water-wave theory" is a theory which for a
given water depth, wave height and wave period
enables a number of basic properties of the wave
motion to be calculated, such as its shape, the
speed of travel of the wave disturbance, the
orbital motions and displacements within the
fluid and the pressure distribution in the fluid.
These properties can in turn be used to formulate
the governing equations for secondary wave-
induced phenomena such as the exact path of wave
travel over a complex underwater topography,
called wave refraction, and the flow of water
into a sandy bed induced by the wave motion
passing overhead.
 11

In the solution of any of these secondary pheno-
mena from first principles the complexity of the
equations to be solved depends largely on the
complexity of the basic wave theory. Similarly,
the accuracy with which a secondary phenomenon
such as wave refraction will be predicted, will
depend largely on the correspondence of predic-
tions with the basic wave theory with real data
on the basic properties such as wave shape and
celerity. Thus, straightforward wave theories
have a better chance of leading to straight-
theory or Dean's stream function theory for
finite amplitude waves in any water depth, a very
systematic pattern will emerge, depending on the
water depth under consideration. Assume for the
moment that a location in deep water is under
consideration. If one could look through a
window into the fluid the following would be
seen:
Brtok., lin. Olplh • l wov. IenOlh

forward formulations of secondary properties

t
I
than complicated theories and good wave theories
lead to an accurate prediction of secondary wave
properties. The first point mentioned above is
most probably the reason for the fact that so-
lutions for secondary wave properties exist
mainly for the very elementary linear wave
theory. Recently there has been a strong ten-
dency to also develop models for secondary wave
properties which are based on the better higher-
order wave theories but in most cases these are
complex models which can be applied only on a
large mainframe computer. An exception to this
is the Vocoidal theory which enables a large
number of secondary wave-induced phenomena to be
calculated with little or no extra effort than
that in linear wave theory.

2.3 Wave shoaling

From the previous section it should be apparent
that it is possible, for a given wave condition,
that is, with a given wave height and wave
period at a given water depth, to calculate
various parameters which quantify the wave
motion at that point. The specific shape of the
water surface variation, the wave speed or wave
celerity, the various orbital motions and the
pressure distribution within the fluid is de-
termined by the wave theory used. However,
assuming that one of the better wave theories
is used, such as the linear wave theory for
small-amplitude deepwater waves, or the Vocoidal
FIGURE 4. Wave kinematics in shoaling waves
The periodic variation of the free surface is
approximately sinusoidal in nature, with the
crests perhaps slightly more than the troughs
for very high waves. Inside the water column
there is a nearly circular orbital motion asso-
ciated with the surface variation. Under the
wave crest the orbital motion is in the opposite
direction under the wave trough. This is illus-
trated in Window 1 in Figure 4. The orbital path
is practically closed with perhaps a small drift
12
in the direction of wave travel.
The intensity or magnitude of the orbital motion
decays exponentially as the point under con-
sideration is situated deeper below the free
surface, that is, the orbital velocities are
lower and the orbital excursion from the mean
position of the particle is less than at the free
surface. If the point under consideration is
deeper into the fluid than about fifty per cent
of the distance between two wave crests, there is
no orbital motion left. This means that the
surface waves exert an influence down to about
fifty per cent of the wave length.
Since the orbital motion is cyclic, any tracer
being a neutrally buoyant particle or a floating
object or whatever, will not move out of the area
with the wave motion but will move forwards and
backwards around its mean position. Since there
could be a relatively insignificant drift pro-
file, which will be directed landwards at the
free surface and less so or even reversed deeper
into the fluid, the tracer might have a very slow
movement away from its mean position.
The pressure within the fluid has a static
component which increases with distance below
the free surface and a dynamic component which
varies cyclically with the wave motion. Pres-
sures under the wave crest are generally higher
than under the wave trough.
The wave surface displacement translates at a
wave speed of 1,56 times the wave period T
(where g/(2n) ~ 1,56; with the wave speed in m/s),
provided that the water depth is more than fifty
per cent of the wave length, which in turn is
equal to the wave speed times the wave period
(or 1,56 x T2 ).
As the wave translates into shallower water,
various changes occur in the behaviour described
above. Assume that a look is taken through a
second window which is situated landwards of the
position where the water depth was exactly equal
to fifty per cent of the wave length.
The wave shape is asymmetrical with respect to
the mean water level, with more pronounced crests
than troughs, as shown in Figure 4, Window 2.
The wave speed (and therefore also the wave
length) is lower than at Window 1. The wave
height will also have changed, but its variation
with depth will be discussed later. The orbital
path is not circular any more but is now ellip-
tical with a slight asymmetry around its hori-
zontal axis, with the top half of the path
bulging out slightly as shown in Figure 4. The
orbital motion is intensified, with higher velo-
cities at all levels. So, for example, the maxi-
mum landward component of the orbital motion
(G fs )' which occurs at the free surface under the
wave crest, is higher than at Window 1 and the
maximum vertical acceleration (a fs ) which reaches
a maximum at the free surface roughly in the
area where the actual surface rises through the
mean water surface, is also higher than at
Window 1. The significance of these two quan-
tities, that is, Gfs and afs ' in relation to
wave breaking will be discussed below. The or-
bital motion now extends down to the bed, with
the elliptical paths of the particles getting
flatter as the mean position goes deeper into the
fluid. At the bed the path is purely cyclic in
the plane of the bed with no component perpen-
dicular to the bed. This is in line with the
kinematic bed boundary condition which stipulates
that no flow should take place through the bed.
The pressure variations are intensified relative
to those at Window 1.
Since the flow extends to the bed now, the os-
cillating flow field and pressure field exert
forces on the sediment particles at the bed. At
Window 1 this was not so. Somewhere between
 13

The cyclic orbital motion at the upper edge of

. the boundary layer generates vortex flows within
the boundary layer, with the little eddies picking

•
....ot ... de9th
lit: wove leooth
up sediment in the troughs of the bed forms and
I carrying them upwards to beyond the ripple crests.
I
It is said that as long as the sediment stays
I
I within the boundary layer all sediment motion
takes place as bed load, which can thus be move-
ment of a particle always in contact with the
bed, movement via the eddies or a saltating type
of motion with the particle returning to the bed
:::..:.......:..
after short hops. However, the high turbulent
I Initiation diffusion associated with the vortex flow in the
BED -+!of n><kI",
FOOMS
~ ~:~
I boundary layer over the ripple on the bed exerts

t ~~M t
an upward force on the sediment particles and

1 some of them are transported through the upper

limit of the boundary layer into the "free flow"
beyond. Some balance is created between the
turbulent upward forces and the downward gravi-
tational pull on the particles and in this way
a vertical profile of sediment in suspension is
created (see Figure 5).

FIGURE 5. Sediment dynamics in shoaling waves
these two windows there is a point at which the
force on the particles becomes sufficiently large
to allow movement of the particles. Initially
this movement is only of a rocking nature, with
no displacement from its mean position, then
cyclic with the particle rolling forwards and
backwards without being displaced from its mean
position and eventually net movement takes place
(see Figure 5, Windows I and 2). As soon as net
movement takes place, bed forms are created.
orbital flow at the bed over the bed forms cannot
be smooth as described above and therefore a bed
boundary layer is formed. The upper limit of
this boundary layer can then be seen as the ef-
fective bed, where the motion as described above
takes place.
This does not mean that the sediment concentration
is constant at any given level in the fluid. In
fact, Brenninkmeyer (1976) (see also Leonard,
Brenninkmeyer, 1978), showed that up to four
major peaks occur in the sediment concentration
near the bed over a single wave period. Fur-
thermore, the mean concentration may vary from
wave to wave for various reasons, for example,
the bed configuration which may be changing on
the time scale of wave periods. Finally, bulk
erosion of the sea bed which occurs apparently
The randomly and which is discussed in more detail
below, tends to introduce large-scale variations
in sediment concentration on the time scale of a
few wave periods to minutes.
Seeing that the fall velocity of the sediment
particles in water decreases as the particles
become finer and since the turbulent intensity
decreases with distance from the bed the finer
14
sediment of the mixture at the bed will be found
higher up in the profile of suspended sediment
than the coarser material. Coupled with the
slight drift profile discussed above this gives
the method for sediment sorting into various
bands in the onshore-offshore direction.
At a third window, just outside the breaker line,
the tendencies observed between Windows 1 and 2
are even more apparent. The wave motion is now
highly asymmetrical with very pronounced wave
crests and long flat wave troughs. The wave
speed has dropped substantially relative to
Windows 1 and 2. The orbital motion is also
strongly intensified relative to that at the
earlier windows with a very marked asymmetry
about the horizontal axis of the orbital path.
The maximum horizontal orbital velocity ti fs and
the maximum vertical orbital acceleration afs are
both substantially greater than in deeper water.
The orbital motion at the bed is much more intense
and the bed forms are more pronounced. The se-
diment suspension profile extends further up into
the fluid and there is a greater tendency for
sediment movement to take place as suspended load
than before.
The pressure fluctuations at the bed which become
important somewhere between Windows 1 and 2 and
which gradually intensify as the wave moves shore-
ward, act as a pump and water is pumped into and
out of the bed continually. Although this is
contrary to the condition of no-flQw through the
bed, the flow is generally low compared with the
orbital motion and it can be assumed that the
solution obtained in the free flow part of the
fluid remains valid. This mechanism therefore
leads to the introduction of "new" water into the
bed on a continuous basis. Swart, Crowley (1983)
have shown that the volume of new water pumped
into the bed per day for a strip of sea bed one
metre wide alongshore and extending from the
mean water line through the surf zone can be as
high as fifty cubic metres/day. To this should
be added the percolation of water through the
sea bed in the intertidal zone which can be
about ten cubic metres per day per running metre
alongshore (McLachlan, personal communication).
The flows in and out of the bed balance out over
the wave period, that is, there is no net influx
or outflow of water into or out of the bed. The
flow initially varies gradually from inflow to
outflow but as the waves become gradually more
asymmetrical the wave-generated pressure variation
with time at the bed also becomes highly asym-
metrical. As a result the flow into and out of
the bed takes place as sudden pulses during the
rising and falling of the wave crest. All flow
could be concentrated in less than ten per cent
of the wave period. Therefore, close to the
breaker line in the area landwards even of
Window 3 the vertical pressure gradients in the
bed as a result of this pumping motion become so
high that bulk erosion of sediment can occur at
the bed due to the bursting motion of water from
the bed back into the fluid. This gives rise to
a different type of suspended sediment profile.
It is not, however, continuous and is partly the
explanation for the fact that suspended sediment
in the nearshore zone is frequently seen as
individual clouds.
The above-mentioned tendencies continue as the
wave shoals landward until wave breaking occurs.
There are two main possibilities for the mech-
anism of wave breaking.
(I) Since the wave celerity decreases with de-
creasing depth while at the same time the maximum
horizontal orbital velocity ti fs increases, a
point is reached in shallow water where the maxi-
mum orbital velocity exceeds the wave celerity
(see Figure 4). This means that the water inside
the fluid is travelling faster under the wave
crest than the wave crest itself is travelling.

The result is a kinematical instability.
wave crest starts toppling and wave breaking is
considered to occur as soon as the front face
of the wave becomes vertical. The wave crest
eventually plunges into the wave trough ahead of
itself as a water jet. Most of the wave energy
is dissipated over a relatively short distance.
This type of wave breaking leads to the so-called
"plunging breakers".
(2) The second type of breaking is associated
with a dynamical instability. The maximum ver-
tical acceleration in the wave motion increases
until a point is reached where it actually ex-
ceeds the downward gravitational acceleration
(see Figure 4). As a result the free surface
boundary conditions are violated and particles
start "popping" out of the free surface.
foaming breakers which gradually lose their
energy are called "spilling breakers".
~ ~
lnc;nWft9 othnllolJon du. 10
b,d friellOt'! and p.rcolotlon
,.fracllng
woy. ror
FIGURE 6. Wave-related processes in the
nearshore zone
15
The The path of a selected point on the wave crest
as the wave moves from deep to shallow water is
called a wave ray (see Figure 6). The variation
in the wave height along a wave ray is determined
by a number of factors. By assuming that the
energy flux in the direction of travel between
any two adjacent wave rays stays constant and by
using Snell's law to define the direction of
travel over a sloping bed (which can be likened
to light travelling through a medium where the
speed of light changes gradually due to a gradual
change in density) it is possible to compute
the wave height and direction of travel at any
given point along the wave ray.
If the contours are smooth and approximately
parallel and if the direction of travel is per-
These pendicular to the contour, the wave height will
increase to a height which may be two to three
times the initial height. The limiting factor
is the onset of wave breaking. This process
of wave height variation due to varying water
depth alone is called wave shoaling. If the
waves approach the shore at an angle (obliquely)
Snell's law will cause the waves to turn in such
a way that the wave tends to approach the con-
tours at right angles. This process is called
refraction. As a result the wave height de-
creases steadily in the shoreward direction
'"t-- .. while the angle between the wave fronts and
the bed contours decreases steadily.
If the shoaling wave approaches a local shoal
waves passing on either side of the shoal tend
to refract towards each other. As a result
focusing occurs with the possibility of very
high (even infinite) wave heights landwards of
a local shoal. Such areas are called caustics.
They occur because the mathematical formulation
used does not permit the flow of wave energy
along a wave front, which will occur in reality
in any area of strongly varying energy along the
front. No adequate method to solve this problem
16
is available at present although some good ap-
proximations are available (see, for example,
Bouws, Battjes, 1982). Mathematical models, so-
called refraction models, are available which
allow the computation of the properties for
waves shoaling and refracting shorewards over an
arbitrary sea bed. To obtain an accurate esti-
mate of the wave height at any given point it is
necessary to include in the computation the at-
tenuating effect of bed friction and percolation,
both of which exist only because wave energy is
used to sustain them.
If a linear wave theory is used for the predic-
tion, the wave shoaling and refraction and the
effects of bed friction and percolation can all
be computed separately and then combined to yield
the ultimate result. It has been shown, however,
that predictions with the linear theory are poor
in shallow water (see Section 2.2). At present
there are only a few refraction models for ar-
bitrary contours which use higher-order theories.
The most advanced of these is one which uses
Vocoidal theory (see Crowley et aI, 1982). In
such higher-order models it is not possible to
separate the various effects as was done for
linear wave theory, and consequently the compu-
tational effort is substantially greater.
A comparison between the results obtained in a
refraction study using linear wave theory and
one using Vocoidal wave theory has shown that
the linear theory underpredicts the wave height
and the angle of incidence in water depths less
than about 20 m. The extent of the under-
prediction becomes substantial for water depths
less than about 10 m to such a degree that the
linear theory should preferably not be used for
this region.
If waves approaching the shore encounter an ob-
struction in their path of travel, such as a
breakwater (see Figure 6), some area is screened
off from wave propagation. Wave energy is trans-
mitted into the shadow area by transfer along
the wave fronts. This process is called dif-
fraction and can be compared with an area of an
otherwise dark passage being lit up by a light
in a room of which the door is open, even though
the area is out of the line of sight of the light
in the room. Techniques do exist for the pre-
diction of diffraction but at present they are
confined to the use of linear wave theory. The
wave height in the lee of a breakwater or groyne
will therefore gradually decrease from undif-
fracted height to zero as one moves through the
diffraction area towards the structure. This
longshore wave height variation generates resi-
dual currents as will be shown in Section 2.6.
If the waves approach the shoreline in the vi-
cinity of a tidal inlet which has a pronounced
outer bar, a local reversal of the approach
direction can occur, as is shown in Figure 6.
It has a strong bearing on the behaviour of the
tidal inlet itself, as will be explained in
Section 3. 8.
Another phenomenon which is restricted mainly to
the shallow-water zone is the occurrence of edge
waves, which are waves which propagate along-
shore. They are forced or generated by the in-
cident surface gravity waves and represent, in
essence, a boundary-layer problem where the
sloping plane beach serves as boundary between
the land and the sea. Edge waves are long-period
waves. Various nodes are possible but the actual
nodes which are excited are determined by the
beach slope. Theoretical treatment of the
phenomenon of edge waves has shown that the edge
wave amplitude decreases exponentially with in-
crease in offshoreward distance and that the
edge wave energy is mostly confined to the
coastal region. The name "edge wave" derives
from this fact.
 17

Stokes in 1846 was the first to treat this sub-
ject. Ursell (1952) extended the first-order
theory of Stokes to a more general edge wave
theory. Sasaki (1975) presented data which
showed that edge waves are generally only
observed when the deep-water wave steepness
Ho/Ao is small compared with the nearshore beach
slope tan 13.
Edge waves play an important part in the study of
nearshore phenomena such as nearshore current
patterns and coastal morphology.
bility to a dynamic instability when the para-
meter (l/(T/gd» has a value of 0,076 in the case
of a horizontal bed. Depending on the wave
period and the water depth at wave breaking, this
represents a value of the deepwater wave steepness
of between 0,015 and 0,032. Above this limit the
dynamic instability occurs first, below it the
kinematic instability predominates. Combination
of this result with that of Kjeldsen and Olsen
provides firm proof that spilling breakers are
caused by a dynamic instability and plunging
breakers by a kinematic instability.

,
2.4 Breaking wave phenomena ,,
Wave breaking
0.06

, ,
,,
Oupwmer
It was shown in the previous section that the "UpnH&

two main types of wave breaking are spilling

, ,
breakers and plunging breakers. Kjeldsen, Olsen ,,
(1972) studied breaking waves and concluded that ,,
the type of wave breaking occurring on plane
0.032
beach slopes is determined by two factors only,
0.022
namely, the deepwater wave steepness Ho/Ao and 0.015

the beach slope a (see Figure 7). From this

I kinematic
tlnstobilit)'
_----suRG\~G
--

--
figure it can be seen that spilling breakers 00 0.05 01

occur primarily for waves of high deepwater t

horizontal
beach slope:

steepness on a relatively flat beach slope, while bed

plunging breakers occur on all slopes but with a

lower wave steepness than for spilling breakers. FIGURE 7. Types of wave breaking (after Kjeldsen,
Olsen, 1972).
Obviously the line on Figure 7 does not represent
a sharp division between spilling and plunging
In most coastal engineering applications it is
breakers. In reality there will be a gradual
important to be able to predict accurately the
transition between these two extremes with waves
initial breaker line. This is done usually by
breaking as part spilling/part plunging in an
using empirically-based techniques. Since the
area around the sharp division shown in Figure 7.
exact breaker point depends on the history of the
A third type of breaker is also shown in the
wave as it travels up to the point of breaking,
figure, namely, surging breakers. These waves do
that is, on the offshore topography, these tech-
not really break but merely rush up the steep
niques are not very reliable and fifty per cent
beach face and are usually associated with strong
error is not uncommon for beach profiles with
wave reflections.
pronounced irregularities. Recently some theo-
retical research has been done on the onset of
Theoretical computations based on Vocoidal wave
wave breaking by following the wave in step by
theory showed that the mechanism which causes
step up to the point of instability (see for a
wave breaking changes from a kinematic insta-
18

very good review of this aspect Longuet-Higgins, present (is generated) in the water body due to
1980). the presence of waves. The best reference in
this field is that by Longuet-Higgins, Stewart
A good first indication of the breaker position (1964) .
can be obtained by making use of the breaker
index, that is, the ratio of breaking wave height
to water depth. For purely spilling breakers
with a very gradual energy loss per distance tra-
velled the breaker index can be as low as 0,4
although in practice it would rarely drop below
0,5. For violently breaking plunging breakers
the theory shows a value of 0,78 although values
of 1,0 occur frequently and a suggested maximum
of 1,3 seems appropriate. Nevertheless, cases
are on record where the breaker index was as high
as 1,8 for waves breaking on a beach with a very
pronounced bar which is in actual fact enhanced
(built up to higher levels) by the flow pattern
set up by the violent plunging breakers. In such
cases this is frequently a rocky ledge or sea wall
on the shoreward end of the breaker zone.
FIGURE 8. A possible mechanism for sustaining
surf diatom blooms over rips
Wave energy is dissipated in the breaker zone.
The rate of dissipation of the energy depends
At the free surface inside the breaker zone there
on the type of wave breaking. Purely spilling
is a mass flow of water in the landward direction.
breakers breaking on a plane beach will have a
In areas where rip currents occur there is a
constant breaker index throughout the breaker
strong undertow, or seaward flow over most of the
zone, that is, the ratio of wave height to water
water depth. It is possible that this system of
depth will remain a constant across the breaker
opposing flows in close proximity, namely, land-
zone. Fully plunging breakers, on the other
ward flow at the free surface and seaward flow
hand, dissipate most of their energy at the outer
just underneath might hold the key to the for-
breaker line. Intermediate breaker types between
mation and maintenance of surf diatom blooms as
spilling and plunging will have dissipation rates
indicated in Figure 8. Recent field work (after
between the two extremes quoted above.
the Symposium) has shown, however, that it is
more likely that the blooms are concentrated in
The dissipated energy sustains a set-up of the
"dead" areas very close to the actual rip current.
mean water level inside the breaker zone and also
This could be either directly landward of the main
in the case of oblique wave attack a longshore
rip channel or just alongshore on the downdrift
current.
side of the rip current.

It is nowadays customary to quantify these effects

by means of so-called "radiation stresses". The
radiation stress is the excess momentum which is

Wave set-up
In the case of perpendicular wave attack the
dissipated energy goes mainly towards sustaining
a higher mean water level within the breaker
zone than outside the breaker zone. The exact
variation of the mean water level within the
breaker zone, called wave set-up or set-down,
depends on the rate at which wave energy is
dissipated, which in turn depends on the breaker
type and the beach profile. A balance exists
between the variations in radiation stress and
the variation in wave set-up. Thus when the
radiation stress decreases abruptly due to waves
breaking as plunging breakers the wave set-up
has to increase abruptly to compensate for this.
brlok., line .. : ..
FIGURE 9. Wave set-up for spilling breakers
Thus the wave set-up profile for spilling breakers
on a smoothly sloping beach will follow a smooth
curve whereas a plunging breaker will be asso-
ciated with a jump in the wave set-up curve.
What happens frequently in nature is that the
waves break initially as plunging breakers on an
outer breaker bar, only to reform and shoal land-
wards as spilling breakers until they break as
plunging breakers again on a secondary breaker
bar. In the case of a wide breaker zone this
process can be repeated a few times. The cor-
responding wave set-up line then exhibits a jump
19
at each of these breaker bars.
The wave set-up is a function of the breaker type
and of the wave height at the outer edge of the
breaker zone. However, Swart (1974) has shown
that the breaker type is not as important as the
initial wave height, and that the maximum wave
set-up at the mean water line will not exceed
twenty per cent of the outer breaker height.
For spilling breakers on a plane beach the on-
offshore gradient in wave set-up is proportional
to the bed slope, with the proportionality factor
being determined by a function of the breaker
index. The maximum wave set-up is, generally
speaking, higher for a higher incident wave
height. This leads to longshore gradients in the
mean water level in areas where for some reason
or another there are longshore variations in
incident wave height. These water-level varia-
tions are the driving force for secondary surf
zone circulations, as will be shown below.
As a result of the wave set-up as outlined above
any tide levels recorded in the breaker zone are
not indicative of the actual sea tide outside
the breaker zone.
Wave-driven longshore currents
During oblique wave attack the dissipated energy
maintains two phenomena, namely, the gradient in
the shore-directed component of the radiation
stress maintains the gradient in wave set-up, as
outlined above, and the on-offshore variation
in the longshore component of the radiation stress
is the driving force for a longshore current
which is confined to an area roughly from the
shoreline to twice the breaker zone width off-
shore.
20
The exact nature of the longshore current is de-
termined by the balance between the driving force
mentioned above and the resistive forces, which
are twofold. The first is the resistance to flow
provided by the sea bed. A quadratic resistance
law is used, which means that the resistance is
proportional to the square of the actual velocity
at the bed, which is made up of a vector product
of the longshore and orbital velocities. The
proportionality constant is a resistance co-
efficient, and although it is based on theory
its absolute value is fixed by correlation of the
theoretical base to data on longshore currents
collected in the laboratory and the field. In
studying the bed resistance it is normally cus-
tomary to make the assumption that the long-
shore current velocity is negligible compared
with the maximum horizontal orbital velocity.
This is usually not the case and the error
introduced in this manner is then compensated
for by the value of the proportionality constant.
Nevertheless, this assumption implies that the
bed resistance is then finally proportional to
the product of the longshore current velocity
and the maximum horizontal orbital velocity.
SPILLING BREAKERS
{plotl. bladll
SH¢RE!..IH(
BARRED
FIGURE 10. Typical longshore velocity profiles
The second resistive term is the lateral on-
offshore mixing, which is a function of the velo-
city gradient in the on-offshore direction.
Sudden changes in the longshore current profile
across the breaker zone will not be tolerated
and will be smoothed out by lateral mixing. For
example, if lateral mixing is neglected the ba-
lance of driving forces to bed resistance for the
case of spilling breakers will yield a trian-
gular velocity profile with zero velocity at the
shore and a maximum velocity at the breaker line.
No flow outside the breaker line will be predicted.
This situation cannot occur in nature and the
effect of the lateral mixing is to create a
smooth longshore current profile which goes from
zero at the shore to a maximum somewhere inside
the breaker zone and down to no flow at some
distance outside the breaker zone (see Figure 10
for typical examples). Strong lateral mixing
will be associated with plunging breakers whereas
spilling breakers will be associated with a
weaker lateral mixing. Longuet-Higgins (1970)
gives a good summary of the derivation of the
theory of longshore currents. In order to obtain
a workable theoretical solution for the case of
spilling breakers Longuet-Higgins assumed that
the lateral mixing increases steadily as the
water depth increases, even outside the breaker
zone. Although this assumption is not really
correct the theoretical solution derived as a
result thereof is in good agreement with actual
data.
In summary, one can in general say that the bed
resistance determines the absolute magnitude of
the current while the lateral mixing determines
the shape of the longshore current profile. The
maximum longshore current velocity is directly
proportional to the incident wave height, the
breaker index, the angle of wave incidence re-
lative to the local contours at wave breaking and
the bed slope and inversely proportional to the
bed roughness (or bed resistance) coefficient.
 21

Fleming and Swart (1982) extended the theoretical
solutions which are readily available to include
the effect of random waves (in a way similar to
an earlier numerical solution by Battjes, 1974)
and Vocoidal theory. It was shown that these de-
velopments improve the accuracy of prediction of
longshore current profiles when compared with the
linear theory version of Longuet-Higgins. A study
of 350 data sets on longshore currents further-
more led to a better definition of the bed re-
sistance coefficient.
The following are a few general tendencies in the
longshore current:
topographical feature, such as an offshore island,
the rip current will persist with time and will
shift alongshore only slightly as the offshore
approach direction changes. If the wave attack
is oblique the rip current will be superimposed
on a longshore current system and the end result
will be a meandering current where the rip cur-
rent is perhaps not defined as well as for the
case of perpendicular attack. If the incident
wave pattern (high-low-high sequence), changes
with time (for example, shifts alongshore as will
be the case when edge waves are the causing fac-
tor), the whole meandering current pattern mi-
grates alongshore.

1. for a given beach slope and angle of in- PERPENDICULAR WAVES

cidence the mean longshore current velocity

HIGH WAVES
increases with a decrease in the grain size;
2. the longshore current velocity is not neces-
LOW WAVES
sarily very low on flat beaches, as was pre-
viously thought, as a result of the inter-
dependence of beach slope and grain size; HIGH WAVES

3. the maximum value of the longshore current

velocity could be as high as 2 mis, occurring
at about two-thirds of the breaker zone width
OBLIQUE WAVES
as measured from shore, although normal maxima
lit
are, however, closer to 0,5 m/s; I IV'k-~ buok.r lin. HIGH WAVES

4. the mean velocity across the surf zone is

approximateiy one-third of the maximum;
=,-'1
line \
~~ r
\I_~

'-ft
LOW WAVES
5. the mean (net) longshore drift of water inside
the breaker zone for any extended period of
{~r l
"
time depends on the wave climate over that I ~ _ HIGH WAVES
direction
period.
t'
..til wohr lint
of wove
attock

Longshore variations in wave set-up

FIGURE 11. Effects of longshore variation in
wave set-up
Figure 11 shows for the cases of perpendicular
and oblique wave attack the effect of a longshore
In the diffraction area in the lee of a break-
variation in the wave height. There is a ten-
water or groyne the wave set-up reduces from the
dency for a mean water flow inside the breaker
shadow line in the direction of the breakwater
zone towards an area of low wave height where a
(see Figure 6). Thus a return .current is gene-
rip current will then occur. If the variation
rated towards the breakwater with as a result a
in incident wave height is caused by an offshore
22

rip current against the breakwater on its lee

side.

The above-mentioned description might create the

impression that the water movement in the shallow-
water region is very much a steady-state situation
which does not vary rapidly. This is not so.
The waves in nature are random and as a result
secondary effects occur which complicate the
above picture significantly. These aspects will
be discussed in the next section.

2.5 Random wave phenomena

It was stated in Section 2.2 that wave motion is
caused by the wind blowing over the water surface.
The wind does not have a constant velocity at all
times and occurs in gusts. As these gusts of
wind blow over the surface of the sea, different
types of waves are generated. The height, .length
and period of any wave in the deep sea is deter-
mined by the velocity of the wind, its duration
and its fetch, that is, the distance for which it
blows over the sea surface. In general, the
longer the fetch of the wind, the higher the wind
velocity and the longer the wind blows the larger
is the wave. If the winds of a local storm blow
towards the shore, the waves will reach the coast FIGURE 12. Composition of random wave surface
(after Visser, Retief, 1979)
in nearly the same form as that in which they
were generated. Under such conditions the waves
Deep water
are steep, that is, the wave height-to-Iength
ratio is about 0,025 to 0,1. Such waves are
Deep water is defined as that area where the
termed seas when observed within their generating
water depth is greater than one-half the wave
area.
length, as was stated earlier. The different
waves generated by the gusts of wind each have
As waves travel away from the generating area they
their own height, period and length. The po-
decay and if the storm centre is far away the
sitions of their crests relative to each other
steep waves will be removed and the wave steep-
can be quite random. The angular difference be-
ness will in general be lower than that of seas.
tween the crests of any two wave crests is called
These waves may travel hundreds or even thousands
the phase (one wave crest in the trough of ano-
of kilometres and have steepnesses which fall
ther wave yields a phase of 180 0 ), which is thus
typically between 0,002 and 0,025. When they are
random for the set of different waves which has
observed outside their generating area they are
been generated. The actual composite wave motion
called swells.

is made up by the linear superposition of these
individual sinusoidal waves with random phase.
A Fourier analysis of the composite wave train,
which extracts sinusoidal waves from the wave
record, will again yield the original set of
individual waves.
+ dition of the individual waves, with their phases
+ posite wave train.
-----------------------------
+ logical, but it will be shown below that it is
•
+8
~~
HEIGHT H
(O'ErMwEl~
11
FREQuENCY F (. lIT )
FIGURE 13. Random waves in deep water
Apart from blowing in gusts and since the direc-
tion in which the wind blows also varies mar-
ginally with time, that is, sudden shifts in wind
direction occur, the waves generated by the wind
do not necessarily all propagate in the same di-
rection. If these waves running in different
directions are added together the resulting wave
pattern will not have long unbroken crests but
will rather be characterized by short-crested
waves (see Figure 12). The results of a Fourier
analysis, which represents only the wave height
spectrum at a single point, are, however, not
affected.
23
As explained in Section 2.2 the shape and general
dynamics of any wave are defined by their ad-
herence to the governing equations and boundary
conditions. For a composite wave train in deep
water with small-amplitude waves it can be shown
readily that the governing equations and boundary
conditions are adhered to to the same level of
accuracy as is the case for each individual com-
ponent wave. This implies that the linear ad-
taken into account, will yield the actual com-
This might sound quite
not so for shallow-water conditions.
Sha HoUJ UJater
It was shown in Section 2.2 that waves in shallow
water are not sinusoidal, but that they have some
sort of higher-order shape. It was also shown
that this higher-order shape is best defined by
wave theories such as the cnoidal theories, Dean's
stream function theory and Vocoidal theory.
last theory was recommended for use because of
This
its simplicity.
In the present context this means that it can be
expected that a random wave train in shallow
water can be built up by linear addition of an
infinite number of individual vocoidal waves with
random phase instead of being built up of indi-
vidual sinusoidal waves.
There are various indications to show that the
components which make up the random wave train in
shallow water in nature are indeed more higher-
order (Vocoidal) than sinusoidal.
1. Waves observed in shallow water in nature
exhibit a definite peakedness which is char-
acteristic of higher-order waves.
24
+ = 5,59
+ f
+
FIGURE 14. Random waves in shallow water
2. Swart et al (1983) showed that the extraction
of vocoidal components from a random shallow
water wave trace leads to a higher represen-
tative wave height than will be obtained if
sinusoidal components are extracted. Swart
(1982) proved that this reduction in wave
height when sinusoidal waves are considered
is characteristic of the spectral decay of
higher-order waves into sinusoidal components.
In other words, if a Fourier analysis is done
on a higher-order wave shape, the wave is re-
constructed to be composed of a number of
sinusoidal waves with periods equal to or
less than the original wave period, all with
heights less than that of the original wave.
If this happens consistently for a whole
spectrum of higher-order waves with random
phase, the result of a Fourier analysis will
be a spectrum of sinusoidal components which
will tend to higher frequencies (lower
periods) than those of the original higher-
order components.
D • 2,om
H rms = 0,92 m
s
=14,625
COMPUTED FOURIER SPECTRUM
INPUT SPECTRUM
FIGURE 15. Fourier analysis of waves in
shallow water
The representative root-mean-square wave
height and the peak period of the spectrum
(where most energy occurs) will both be
underpredicted by a normal Fourier analysis
in such a case. Swart (1982) showed that
similar conclusions apply to visual estimating
techniques such as the Draper analysis for
wave height and period.
3. Should the tendencies outlined in (2) be ob-
served, it would mean that a comparison of
representative wave heights along a wave ray
from deep to shallow water, as calculated by
whatever method of analysis, would indicate
extensive wave decay. In fact, all studies
to date on wave decay due to bed friction
which are based on prototype data have led
to exactly that conclusion, namely, that
theoretical wave energy losses due to bed
friction are lower than those which occur in
nature (see for example Iwagaki, Kakimura,
1967) •
4. In studies on the characteristics of random
waves in shallow water it has been shown re-
peatedly that the wave period seems to de-
crease as the water depth decreases (see
Busching, 1975) which is again indicative of
a method of analysis based on sinusoidal
 25

waves being applied to higher-order waves.
The above outline indicates that there is merit
in assuming that random wave trains in shallow
water consist of higher-order waves with random
phase. Each of these waves would have adhered
to its free-surface boundary conditions exactly,
had they occurred separately. However, since the
individual component waves in shallow water do
not all adhere to the small-amplitude criterion
(wave height is not negligible relative to water
depth) it can be shown readily that the free-
surface boundary conditions in this case will not
be adhered to.
To examine what this implies consider two in-
dividual higher-order waves with random phase
which are superimposed.
I T- lay lIs
surface will drive a wave with a period of 55 s,
which is coupled (bound) to the two individual
component waves. This wave is called a bound
wave. If, during calculations, the bound wave is
not added to the individual component waves, all
orbital motions (velocities and accelerations)
and internal pressures will be in error. If, on
the other hand, the correct bound wave is added,
the composite orbital motions and internal pres-
sures can be found by linear addition of the
values of any of these properties for the two
component waves and the bound wave.
Now consider a whole series of individual corn-
ponent waves with random phase which make up a
random wave train in shallow water. Again, the
free-surface boundary conditions are not adhered
to. The composite fluctuation in the free-
surface boundary conditions will drive a bound
wave train with periods typically in the range
from 30 s to 600 s.

RANDOM
WAVE
TRAIN
I
I
I

•
I BOUND
I WAVE
TRAIN
trror in
KFS Be (or DFSBC) I' T· 55 s

FIGURE 16. Bound wave for two short waves

superimposed

This fluctuation in the error in the dynamic free- I

: 8O..NO
.. 0.033
I CRAVITY Wol\ot:
I
~J-+­
I'lllAVE: $l'ECllll..ll I
surface boundary condition (DFSBC), which as I Sf'EClJlUr,tl
I
I

shown below, will have a period of 55 s, is in

I I
I ,
I

actual fact indicative of a cyclic pressure

fluctuation at the free surface of the composite
wave. In a manner similar to the generation of
wind waves by pressure variations on the water
surface, this fluctuating pressure at the free
FIGURE 17. Bound wave spectra in shallow water
Botes et al (1982) found a clear correlation
between the height of the incident surface gravity
waves and the height of the long waves in shallow
26

water, as can be seen in Figure 18.

'mv, ~ql'll IOtc)rum

II (m)
d • 11.8m

.j
H fml • • . 38 m
2.0
fp o 0.oe99 1_1

0------0 Vocl)ldol Fourier

... - - - - -0 klurltf

I~

:/
:'V
0~--~r----4-----+----~~f-~----+-----t---~

. .:.j.
~ u~----~--~-,---+~~~----~----+-----+----- 0.1 02 03 0.' 0.'

.: '" .' Relot; ... , bCU"od WOlVe ,£!ctrum

~~--~~--~~'~~'+'~'~--r---~-----4-----t----~

:.:,':' :;;~.~ .. f • o.oloe .-1

U· .. I.'~'· 0.0.)0 ,-I
"~----~~~~~-+-----+-----r----+-----+-----

'd::!«. :~~ i 0.1

"~:I':"

.
HS • !~"'nc.A.'lT WIroV[ H(IGIfT (WAVERIOER) 0,1

.• f--~c-'--I--'----t-----+----- Hl • ~:;iR~:~~~~~~~

FIGURE 19. Shallow-water and relative bound

"_':----l-----:':----'------f;---'--I--!;-I---,-1---;;0" wave spectra: Koeberg waverider
H~

FIGURE 18. Relationship between short and

long wave heights
(m)
", 14 ... , . to.",
',I • O.IOU ,-I
f, • o.049t,·1
Examples of spectra of surface gravity waves In
shallow water (after Thompson, 1977 and Bosman,
personal communication) and the corresponding
relative bound wave spectra calculated as out-
00
lined above (Swart et aI, 1983) are given in
Figures 19 and 20. These figures indicate that
the shape of the spectrum of gravity waves strongly ".
(-)

influences the shape of the bound wave spectrum.

It is also apparent that an increase in the width
of the gravity wave spectrum will lead to an in-
crease in the bound wave height. In fact, al-
though the Atlantic City record would appear to
resonate at only one frequency (~180 s), the
height of the bound wave is appreciably higher
FIGURE 20. Shallow-water and relative bound wave
than in the case of the Koeberg record where the spectra: Atlantic City pressure
main resonance frequency is 90 s, even though transducer
 27

the incident wave height at Koeberg was double

that at Michigan City. These two figures indi-
0.20
cate that the scatter in the data on Figure 18 • shallow station
can probably be accounted for by the shape of the
o deep station
incident gravity wave spectrum. The clear cor-
relation between the heights of the short- and
long-wave heights in Figure 18 can be explained
by the fact that the phasing of the frontal wea-
ther systems which lead to high wave action ge- 0.10

nerates wave spectra of roughly similar shape.

During calculations for shallow water conditions

the bound wave spectrum has to be added to the
gravity wave spectrum.
o ~----.-----~----~r-----.-----~

The presence of a bound wave component in random

o 0.1 02 0.3 0.4 Q.5
fi (s -I)
shallow water waves means that the water level in
Relative bound wave spectra
this region will appear to fluctuate with a period
of the order of hundreds of seconds (one to ten MICHIGAN CITY
minutes). The amplitude of this long-period 18· 10 . 1975 : 1020

fluctuation in the water surface will decay

sharply as the water depth increases. These FIGURE 21. simultaneous relative bound wave
spectra in deep and shallow water
bound waves will not propagate freely from one
water-depth to another in shallower water, as
If the incident train of surface gravity waves
happens with gravity waves, but their character-
approaches the coast at right angles, the asso-
istics at each point will be determined by the
ciated bound waves will manifest themselves in
gravity wave spectrum at that point. It would
simultaneous variations in water level in the
therefore not be possible to take the bound wave
infra-gravity range (>~ 30 s) at various locations
train at one location and use- it to calculate the
alongshore where the water depth is the same.
bound wave train at another location where the
Excess water brought into the surf zone by the
water depth is different.
phenomenon would be transported offshore in pul-
sating rip currents. If the approach of the in-
However, relative bound wave spectra calculated
cident wave spectrum is oblique, it would appear
from the observed spectra of surface gravity waves
as if the bound waves are propagating alongshore.
at two stations, one in deep water and one in
The propagation rate alongshore would be closely
shallow water, at Michigan City for the same time
coupled to the directional spectrum of the in-
(see Figure 21) tend to indicate that there may
cident gravity waves, or in other words, to the
be some correspondence between the shapes of the
variation in approach direction of the surface
two bound wave spectra, although there is a shift
8ravity waves.
in the peak frequency with the peak frequency de-
creasing as the water depth decreases.
A system of rip currents which propagates along-
shore would be associated with the bound waves
which move alongshore. Since it will be super-
28

imposed on the normal longshore current, the ob-
served nearshore current system will exhibit a
meandering nature.
The compound water-level fluctuation at the water
line is the result of all the processes described
above. Contributors are the run-up due to short
gravity waves, the bound waves, edge waves and
also tidal variations (see Figure 23).

I
~\
Water line \
(instantaneous)

FIGURE 22. Effect of "oblique" bound waves

Since the bound-wave height also decreases sharply
with increasing depth in the same manner as the
edge waves, it is difficult to distinguish be-
tween these two types of waves. The main dif-
ference between them is that the edge waves will
propagate alongshore for all angles of wave ap-
proach. The relative magnitude of these two types
of long-period shallow-water waves is at present
under investigation (both theoretically and by
field observation).
FIGURE 23. Different types of backwash
The mechanism by which water runs back to sea
is most probably determined by the rate of change
of the water level at the shore. Slow vari~tions
allow water to seep into the bed whereas fast
variations cause air to be trapped in the sand
and the water runs back mainly as surface flow.
2.6 Other water movements

It is often observed that waves in shallow water
appear to be long-crested, although their heights
and periods are still random. It is thought that
this apparent ordering as the waves come from
deep to shallow water could be related in some
way to the bound wave activity. This is, however,
still a gray area which needs considerable re-
search.
Two types of water movement which are not rela-
ted to wave motion but which may under certain
conditions have an appreciable effect on the
overall water movement in the nearshore region
are tidal motion and wind-generated currents.
They will be discussed briefly below.

Tida l motion
The continuous changes in the positions of the
sun and the moon in relation to the earth cause
the sea surface to rise and fall quite regularly
once or twice a day. The tide generated by this
earth-moon-sun system is called the astronomical
tide and is completely unrelated to the sea-level
changes generated by meteorological factors. The
tidal range, which is the vertical distance be-
tween the high- and low-tide levels, varies with
the phase of the moon and reaches maxima just
after both new and full moon (spring tide).
During first- and third-quarter phases of the
moon, neap tides occur which have a low tidal
range.
In deep water tidal currents are negligible. As
the water depth decreases, the tidal currents
increase. Nevertheless, tidal currents on an
open coastline, which are predominantly at right
angles to the coast, are relatively insignificant
and reach their maximum values well outside the
breaker line. It is a different matter altogether
when tidal motion in closed or restricted waters
is considered. The tidal range increases sub-
stantially and the associated tidal currents
become significant with a substantial longshore
component. In places such as the British and
French coasts of the English Channel the tidal
current can even dominate the wave-generated
flows.
Wind-induced currents
The shear stress set up on the sea surface as a
result of a wind blowing over it induces a cur-
rent in the water in the same direction as that
in which the wind is blowing. This wind-generated
flow has a maX1mum velocity close to the free
surface and decays sharply as the distance below
the free surface increases. Various researchers
have shown this maximum surface current velocity
29
to be directly proportional to the wind velocity
10 m above the sea surface.
The factor which relates the wind-generated sur-
face velocity to the wind velocity is usually
referred to as the wind factor. Research by a
number of authors has indicated that the value
of the wind factor varies between 0,01 and 0,04
with a most probable value of 0,018 (Kullenberg,
1974). Especially in cases where very strong
winds blow alongshore (for example, south-
easterly and north-westerly winds at Koeberg), it
is possible that wind-generated velocities of
between 0,25 mls and 0,5 mls can occur within
the surf zone. Such velocities and their
effect on both overall water and sediment move-
ment cannot be neglected.
The tendencies described above are only averages
and it is possible that very complicated water
movemencs can occur as a result of the action of
the wind. Depending on the direction of wave
attack, the wind direction and the alignment of
the shoreline it is quite possible that flow re-
versals can occur, either as a function of dis-
tance offshore or even with vertical distance
from the free surface.
Strong offshore winds cause a set-down of the
water level at the shore whereas the opposite is
true for onshore winds.
3. SAND MOVEMENT
3. 1 General
It was shown in Section 2 that the various modes
of water movement, wave-, wind- or tide-induced
or any combination thereof, set up a shear stress
at the bed which increases in intensity as the
water depth decreases. This shear stress field
has the potential, ~n time, to move material
off the bed and into the main body of the water.
The compound water movement as a result of all
30
the processes described in Section 2 can then
simply carry the material away. This process is
called sediment transport. A brief summary is
given in this section of the various modes of
transport and the ways and means of calculating
the magnitude of this sand movement. For a more
complete review of the techniques used to predict
sand movement specialist papers, to which refe-
rence will be made, have to be consulted.
3.2 The sandy bed
Beach materials consist of sand and gravel
transported to the sea by rivers, of dune sand
eroded by wave action, of the debris from nearby
coastal cliffs and fragmented shells. Wave and
current action repeatedly sort these materials.
At any given location the bed material is made up
of a composition of materials from the various
sources. The grain size composition at any such
location is determined by the sources of material
and the energy regime at that location. If the
median grain size of bed samples along a line
perpendicular to the shore is considered, it will
be seen that wave sorting has led to sharp vari-
ations in the median which can easily amount to
2 to 4 times the average along the line (higher
or lower). The position at which high median
grain sizes occur coincides with areas of high
wave action, such as the plunge point at the outer
breaker line, and with the upper edge of the
summer berm. Apart from these variations in
composition and median grain size, the bed can
also exhibit a very definite micro-structure
which is governed by repeated erosion and de-
position during reworking of the beach. Flemming,
Fricke (1983) in this volume give a very good
account of this phenomenon. It is of particular
importance since it leads to horizontal strati-
fication, which means that the horizontal per-
meability is higher than the vertical permeability.
This in turn restricts the vertical flow of water
into the bed (see Swart, Crowley (1983) in this
volume) .
The grain-size distribution of beach material
was originally determined by sieving the material
through a series of sieves with progressively
smaller mesh sizes. The material has to be pro-
perly dried beforehand and the process takes a
considerable time. Recently the grain-size dis-
tribution has been determined by means of auto-
matic settling tubes in which a wet sample is
allowed to settle in a tube filled with water and
the change in accumulated mass on a scale at the
lower end of the tube is recorded. Conversion to
size distribution is nearly instantaneous by the
use of proper calibration curves. Apart from the
time-saving aspect this latter method is to be
preferred since it gives a more "hydraulic" grain
size; fall velocity is an important parameter in
the determination of sediment transport.
3.3 Incipient motion and bed forms
As the shear stress at the bed increases, a cer-
tain critical depth is reached where the drag on
individual particles becomes sufficiently high to
rock them to and fro. Further shorewards some
particles start rolling forwards and backwards
around their mean position and even later general
movement occurs. In time this movement of par-
ticles caused by the cyclic water movement leads
to the formation of small bed-forms, called sea-
bed ripples. A substantial body of research has
been done into the start of movement of particles
on the sea bed, which is called wave-generated
incipient motion. Silvester (1974) summarized
the results of thirteen different studies into
this phenomenon. There were very large differences
between the results of these studies. This is
partly the result of the application of certain
formulae outside their range of applicability,
but also the high sUbjectivity involved in de-
ciding on whether incipient motion has already
taken place. This last aspect is complicated
further by differences in the definition of in-
cipient motion, which can range between particle
oscillation at the spot and the formation of rip-

pIes. Keeping all these facts in mind Lenhoff
(1982) presented a uniform method which uses all
available data.
The characteristics of the bed forms are deter-
mined by the water movement and the size of the
bed material. It can be stated in general that
the size of ripples under a given energy regime
increases with increasing size of the particles
in the bed. As the orbital velocity at the bed
increases in the region landwards of the in-
cipient motion line the ripple size increases
until a certain critical orbital velocity, which
is of the order of 1 mis, is reached. Landwards
of this point the ripples decrease in size until
the bed is smooth again. This happens frequently,
but not always, in the surf zone.
Inman (1953) did pioneering work on the estab-
lishment of techniques for the prediction of wave-
induced ripples. Swart (1976) summarized all the
techniques, based on all available information,
which supplies a very accurate prediction of bed
form characteristics. Reference can also be made
to Mogridge, Kamphuis (1972).
The orientation of the ripple crests indicates
the type of motion which predominates at the bed.
If the ripple crests are parallel to the shore the
region is wave-dominated. On the other hand, if
the crests are perpendicular to the shore, the
region is dominated by the longshore currents.
This is a relatively rare occurrence and is nor-
mally associated with strong tidal flow in the
vicinity of an estuary mouth or in a confined sea
or with wind-generated currents superimposed on
the wave-driven longshore flow. It is more
usual for combinations of these two extremes to
occur in the shallow-water region, which implies
that both wave and current motions influence the
bed.
31
3.4 Transport modes
Two mechanisms for the transportation of material
into the bulk of the fluid were identified in
Section 2. The oscillating flow over the ripples
sets up little eddies in the lee of the ripple
which explode when the flow is reversed. Material
is ejected beyond the crests of the ripples. Gra-
vity pulls the particles downwards and the turbu-
lent lift forces carry them further into the fluid.
A balance is reached which leads to an equili-
brium profile of material suspended at various
levels in the fluid.
In shallow water, where the wave motion is highly
asymmetrical, the fluctuating pressure at the bed
pumps water into and out of the bed in sharp
bursts. The water bursting out of the bed can
cause bulk erosion of material which would find
its way into the main body of the fluid.
There is also a third mechanism for entraining
sediment, which is associated with the extremely
high turbulence in the breaker zone and specifi-
cally near the plunge point.
Although it is subjective, a distinction is
usually made between material moving in relatively
close (or continuous) contact with the bed and
material moving mainly in the water body. The
definition which is usually used defines bed load
as that part of the total volume of material
moving (i.e. of the total load) which moves
within the turbulent bed boundary layer. Swart
(1976) showed that the magnitude of the boundary
layer can be approximated by the bed roughness
which in turn is related to the ripple geometry.
For example, the boundary layer thickness is five
times the ripple height for ripples with a steep-
ness (ratio ripple height to ripple length) of
0,2 whereas it is equal to the ripple height for
a ripple steepness of 0,04. For a flat bed
Kamphuis (1975) showed that the bed roughness
equals two times the particle size D90 which is
32
that particle size coarser than 90 per cent of
the particles.
Suspended load is then that part of the total
load which is transported above the bed boundary
layer. The concentration of suspended material
in the fluid decreases sharply from a maximum
inside the boundary layer. Depending on the as-
sumptions made regarding the vertical variation
in vertical diffusion it is possible to obtain
different theoretical solutions for the equili-
brium variation in the amount of sediment in
suspension. The results obtained by using these
different techniques are all in fair agreement
with data.
There is usually a tendency for coarse material
to be transported in closer proximity to the bed
than fine material is. This allows the velocity
profile in the vertical, which may even create a
mass transport in opposite directions at dif-
ferent levels within the fluid, to sort beach
materials into size bands parallel to shore.
The complicated water movement in the nearshore
zone together with the suspended sediment con-
centration in this area leads to a complicated
sediment movement which can, broadly speaking,
be classified into a longshore component, which
is called longshore drift or longshore transport
and an onshore-offshore component which is called
onshore-offshore transport. A brief discussion
of these two types of transport is given below.
3.5 Longshore transport
A detailed discussion was given in Section 2.4
of the factors governing the magnitude and dis-
tribution of longshore currents, and the factors
governing the magnitude and distribution of sedi-
ment in suspension in the water column were out-
lined in Section 3.4. The sediment in suspension
is transported away by the longshore current.
The magnitud"e of the resulting longshore sediment
movement depends mainly on the incident wave
height, the longshore current velocity and on the
grain-size distribution of the bed material. The
distribution across the breaker zone, on the
other hand, is strongly related to the on-
offshore distribution of the longshore current
and to the compound suspended sediment profile
which in turn is related to the pattern of wave
breaking.
The volume of material moved alongshore by this
mechanism can vary from virtually nothing during
periods of calm weather or perpendicular wave
attack of the order of tens of thousands of cubic
metres per day under oblique storm-wave attack.
The direction and magnitude of the net drift at
any given site along the coast are determined by
the wave climate with specific reference to the
directionality of the incident waves. The net
drift along the South African coastline which is
directed up the coast in all but a few local ex-
ceptions can typically be as high as a million
cubic metres per year; for example, the net
northerly drift at Durban is 650 000 cubic metres
per year whereas that at Richards Bay is 800 000
cubic metres per year. To visualize this volume
of sand think of 160 000 ten-ton trucks loaded
to capacity and you have an idea of the net long-
shore movement by wave action in a one-year
period. In exceptional cases the transport rate
can be much higher. At Swakopmund, in Namibia,
on the African west coast, the Swakop River comes
down in flood on average every seven years,
bringing with it millions of cubic metres of
material, which forms a large protrusion in the
coastline. It has been reported that more than
ten million cubic metres of sand deposited in
this manner were transported away towards the
north within a few years after a particular
flood. This large transport rate can be attri-
buted to exceptionally large angles of incidence
of the waves at the breaker line.
 33

The prediction of the magnitude of longshore

sediment transport rates is still a relatively
.... V[ 1'I(IC'1T OI$HtlIlUTIOfol
grey area. At present there are two main types
of technique in use, one which predicts only the
total volume of sediment being transported along- ATI....,NTtC
0'''''
shore through anyon-offshore line (see for ex-
ample the SPM method based on energy principles:
SPM, 1973), whereas the second type predicts the
variation of the longshore sediment drift with
distance offshore. This detail-predictor tech-

t
N
nique can be based either on theoretical and ""'dian grQln ~,It':

At .... :. 0,.. • O.I{;O .......

AI 00:· 0... ' 0.800 ......
empirical knowledge of the longshore current and
the quantity of suspended sediment, or on adap-
tations to available predictive techniques for
sediment load in rivers in order to allow the FIGURE 24. Location sketch for typical appli-
cation of package-deal approach
inclusion of the effect of waves on the shear
stresses at the bed.
Volume
chonQe in
Experience has shown that the most reliable control
orea
prediction of longshore transport rates is ob-
1m3)
tained by applying as many of the verified tech- I
I
I
niques as possible and determining a weighted Sept 73
I
Sept 74 Sept 75

answer based on all the techniques used. Swart,

Fleming (1980) explained in detail such a package
deal approach. Typical input for such a compu- FIGURE 25. Prototype data: Volume changes in
tation would include details of waves, the topo- control area
graphy of the area under consideration and the
characteristics of the bed material. Nearshore
wave characteristics would be obtained from
deep-sea data via a wave refraction study; a
detailed knowledge of the topography of the area 1'1. YJ6 m3/yr
t22%
would be needed to allow refraction computations
to be performed as well as to determine the
breaker location and longshore velocity profiles.
For the transport calculations the relative ap-
parent density and the grain-size distribution
of the bed material would have to be determined
2.5.IO'm3/yr
beforehand. t lS %

Figures 24 to 27 show the application of a package

deal approach to a site just north of Cape Town
(Figure 24). The results of prototype measure- FIGURE 26. Prototype application: Results of
ments, over a period of two years are shown in package-deal approach
34
Figure 25, which indicate that the beach in ques-
tion is in overall equilibrium, but that a gross
residual transport rate of 2.9 million cubic
metres per year must have taken place to effect
this. The results of predictions with the package
deal approach are given in Figure 26, which in-
dicates that the beach is in approximate overall
equilibrium but that a gross residual transport
rate of 3.3 million cubic metres per year (plus
or minus 20 per cent) is predicted, which is in
fair (excellent) agreement with the data in
Figure 25. Since the occurrence of waves from
upcoast and downcoast is very seasonal it can be
concluded that both the data in Figure 25 and the
package deal results in Figure 26 indicate a
sloshing mode in the area, with material moving
upcoast in summer and downcoast in winter.
Figure 27 shows the agreement between the results
obtained by the six different methods for pre-
dicting longshore transport which were used in
the package deal approach.
LOCATIONAA: Oowncocal I,onapo,t
0·10
S Idl
S,ot
_ _ h4eon computed yoh...
~
(dl
Stot
chplh dlml
FIGURE 27. Prototype application: Correspondence
of 6 different predictors
3.6 Onshore-offshore transport
It was shown in Section 2 that some sort of re-
circulation of water is set up by the breaking
waves in the nearshore zone which can be either
two-dimensional, that is, restricted to the
vertical plane with, for example, landward net
velocities near the bed and the free surface with
a seaward return flow in the centre of the fluid,
or three-dimensional with seaward return flow in
the form of strong rip currents. Depending on
the nature of the incident waves and the beach
slope the water movement can be either conducive
to a net seaward or a net landward flow of sand.
The factors involved have been schematically de-
picted in Figure 28, where a is the beach slope
in the breaker zone, g is the gravitational ac-
celeration, Ho is the deepwater wave height,
Ao is the deepwater wave length and w is the
particle settling velocity.
_ _ _ _e
EROSION
•
•
ACCRETION
B
• lotal cQn'9Utad 100000'hor.
IfGrnport
FIGURE 28. Tendencies in onshore-offshore
• loncjIthot. Iron'potl
beh"'n Otplh td·O·1) transport
and {eftO·1} """ ..
This figure indicates that the tendency for a net
seaward movement of beach material, that is,
10
beach erosion, will be increased for either
steeper incident waves, or higher waves or
steeper beaches or finer beach material. The
opposite is true for landward sediment movement.
In practice it has been found that the strong
seaward sediment movement during the few per cent
of storm action per year is normally fairly
closely balanced by the landward movement of
sediment at a lower rate during the rest of the
year.
 35

Swart (1971) summarized all data on onshore-
offshore movement under wave action available at
that time and showed that the curve in Figure 28
can be closely approximated by a straight line
with point A occurring at aHo/Ao = 0.065 and with
point B coinciding with 19H<i/w = 78. Data also
indicated that the rate of erosion (or accretion)
was higher as the distance from the equilibrium
Swart (1974) used model and prototype data to
obtain empirical expressions within an analytical
framework to allow the prediction of the limits
as outlined in Figure 29, the equilibrium profile
and the resulting sediment transport at any time
during the profile deformation.
• ______________ ~l~_

... POSSIBLE IroIOV[MENT

curve AB increased, as indicated schematically by
BACKSHORE
the sizes of the dots in Figure 28.

-- ------ ~h!.!!!.!..O_F_

In an extension of the above-mentioned quantita-

tive summary of the behaviour of beach profiles
..:; •..•.'.::.:.. ::..:: .....
Swart (1974) proposed a quantitative method for . ............
the prediction of the direction and magnitude of ~.:..
- - - - - - - - - - - ':.

TRANSITION
...... PROFILE DEVELOPMENT

onshore-offshore sediment transport. This method ZONE

0' • • : . '0

has in the ten years since then been applied to

VISIBLE SEDIMENT
numerous prototype cases with great success. MOVEMENT, NORMALLY
AT ABOUT 15M TO 2I)M
BELOW MSL

The essence of the method can be found in the

three underlying assumptions which were all
verified with real data, namely,
(1) the concept of equilibrium, that is, a beach
profile will deform under a specific wave
condition and will, if the wave condition
persists for a long enough time, come to
equilibrium;
(2) the rate of deformation of the beach profile
will decrease as the profile gets closer to
its equilibrium shape; and
(3) the beach profile can be separated in three
distinct zones, namely, (a) the backshore
above the wave run-up line where erosion
usually takes place due to soil mechanical
reasons, (b) the "developing" or D-profile,
which extends to a water depth equal to ap-
proximately twice the breaker depth and in
which active bed form formation (mobile bars
and ripples) takes place and sediment moves
predominantly as suspended load and (c) the
transition zone in which transport usually
takes place as bed load (see Figure 29).
FIGURE 29. Schematization of typical beach
profile
The method is being used extensively for pre-
dictions on required beachfills, the projected
dynamic swept prism as required for the deter-
mination of the depth of burial of marine pipe-
lines, possible coastline retreat during storms
and the effect of submarine dredging on profile
reworking. It could be rendered more useful by
the addition of features like Eigenfunctions as
discussed by Aubrey to incorporate the effects
of breaker bars, the behaviour of different beach
types under various wave conditions as discussed
by Short and by making more formal use of the
concept of the dynamic swept prism. These indi-
vidual aspects are discussed by the respective
authors elsewhere in this volume (Aubrey, 1983;
Short, 1983; Chapman, 1983), but more thought
should still be given to the welding together of
these various concepts. Baillard (1982) has
recently also developed a method for onshore-
offshore transport predictions, which still needs
36

3.7 Beach types correlation of the parameter ~/(WT) with the

general beach type, as is reproduced here from
Short (1982) as Figure 31, can be coupled with
breaker type as depicted in Figure 7.

REFLECTIVEIINTBi'MEDlAm DISSIPATIVE
IhfliSlio/rl Ihftsltoltl

DISSIPATIVE
~

NH-+
t+-J
I+" .~
kJ lonq',",,"
" bor-Iroujh
:"0:---- f - - - - - 1-- j£ -- ru&lLID
o Madol 5""
~~~m~ N I... G GoolWQ
;I!----
!'t: trwvers.
- - - - I- 1--- NS Nih Sem Mile
MS Mid Seven Mile
~~.!~-- - - - - - 1-.., E Easlern
1-- 1 - - - M MOfuya
"'= ridq.-runnel
...... Iow·tide terrace 1 N Narrabeen
p Palm
C Colloroy--
~ 8 Brocken
REFLECTIVE F Fisherman's
.Y·
0·2 0·5 I 2 3 456 10 20 30

FIGURE 31. Modal beach state and temporal vari-

ability of beach state as functions
of the modal values and temporal
variability of ~/~ T. Horizontal
arrows indicate ave¥age temporal
range of Hb/~ T; vertical arrows in-
dicate averag~ temporal range of
beach state (after Short, 1982)

This has been done in Figure 32 which shows that

dissipative beaches are normally associated with
spilling breakers whereas reflective beaches are
associated with plunging or more generally sur-
ging breakers. Intermediate beaches and plunging
breakers are closely associated.

Deepwater
st.epne ..

FIGURE 30. Plan and profile configurations of

the six major beach states (after
Short, 1982)

Short (1982, and also in this volume) defines

six major beach states, as is shown in Figure
30. These beach states or beach types will not
be discussed in detail here, and reference should
rather be made to Short's excellent paper in this
beach slope
bed
volume (Short, 1983). In the context of this -7

review paper on water and sediment movement it

FIGURE 32. Correlation between beach states and
is, however, interesting to note that Short's breaker types
 37

In order to explain the various circulation pat-

terns observed by Short on intermediate beaches,
one should take into account the shape of the
gravity wave spectrum which inter aZia means that /
ts
Dominant wave direction
ORANGE
the effect of the bound wave spectrum needs to be RIVER
considered. The close correspondence between I Not northerly longahor. IOnd drift
beach states and breaker types as shown in Fi- 't (- 1,4 x 10' m' /yr)
gure 32 should form the basis of further research
Sand accretion duo to blocking by .. owoll
in the next decade.
. ....
. .... . . ':'. st,_ beach .Iopo:
: ••• : :'••• : : Mar. ollahort-dlrected moy• .".nt
3.8 Shoreline changes
; .....
:." :°':0°.: Mor.~in IUlpenllon
It should be clear from the discussion in ; ° 0 •••
~

Sections 3.5 and 3.6 that coastal erosion or GIU-P

accretion can be caused either by onshore/offshore
sediment movement or by longshore variations in
the longshore transport rate. It was shown in
Section 3.6 that the onshore and offshore trans-
port usually balances in the long term, that is,
onshore and offshore transport lead to short-term
changes in shoreline position. In fact, since
the incident wave climate normally contains
seasonal variations at most locations, it is
frequently observed that the shoreline changes
due to onshore/offshore transport exhibited a
seasonal pattern. Shoreline changes due to
longshore variations in the longshore transport,
on the other hand, usually occur over longer
periods of time since the reason for the long-
shore variation in transport" such as major
changes in coastline alignment, headlands or
coastal structures impeding the flow, are forever
present.
A good example of typical shoreline changes due
to the two mechanisms is the shoreline develop-
ment in the vicinity of a man-made seawall of sand
at Oranjemund in Namibia on the African west
coast (see Figure 33 for situation sketch).
To facilitate the mining of diamonds from the
bedrock on the sea floor in the area seawards of
the original high-water mark the mining firm
CDM (Pty) Ltd., which is working in the area,
" Sand accretion a. IOnd in ,"spenslon drape
!~ a. It returna to normol
.,. No 3P1ant Jelly
Location I
FIGURE 33. Location sketch for shoreline changes
pushed a seawall of sand 200 m out to sea. The
wall has a crest level of about 10 m above low-
water level and a seaward slope equal to the
angle of repose of the sand. During periods of
spring tide when the waves can reach higher
levels against the seawall greater erosion occurs
which requires the high maintenance dumping of
sand from the seaward slope of the seawall.
During such times seaward losses of sediment in
front of the seawall are greater than those from
the adjacent coastal areas, as a result of the
steeper beach slopes in front of the wall. The
dominant wave direction is approximately south-
westerly which leads to a net northerly sand drift
of about 1.4 million cubic metres per year. In
front of the seawall where the availability of
suspended sediment is higher than on the adjacent
coast due to the greater offsho~e-directed trans-
port, the longshore transport rate is appreciably
38

higher than 1.4 million cubic metres/year. (June-August) and the biggest seasonal growth at
the end of the summer (December-February). The
long-term average growth rate is about 10 m/yr
I:!O
RELATIVE
over the five-year period plotted.
• • •• AvtraOI thorelint chonOI
ACCRETION
Computed shoreline chono ..
('71_'73)
In m

100
no bypo .. lnQ

o
o~----~~~~~------~--------~~o
9.0

80 G 90 G
o 79 62

Distance alonoshore in km TI ME in YEARS

FIGURE 34. Shoreline changes updrift of seawall

As a result of the situation described above an
unusual phenomenon occurs, namely, the shoreline
grows in the long term on both the updrift and
the downdrift sides of the seawall.
for the growth updrift south of the seawall is
obvious, namely that the seawall serves as a
partial barrier to the northbound drift.
result is shown in Figure 34 which shows the
relative accretion in a two-year period.
data indicates an appreciable sand-wave formation
in the coastline, which is typical of shoreline
development in the vicinity of major human inter-
vention on the coast.
Downdrift of the seawall, on the other hand, long-
term shoreline accretion takes place as the volume
of sand in suspension drops to its normal open-
coast values. This can be seen clearly in Figure
35 which contains a time-history of the waterline
at Location 1 of Figure 33.
nite seasonal variation in the shoreline with the
biggest recession at the end of the winter months
FIGURE 35. Shoreline changes downdrift of
seawall
In the area downdrift of an "offset" tidal inlet
such as shown in Figure 6, where an appreciable
The reason
outer bar exists, erosion of the shoreline takes
place since longshore transport is directed away
from that area in both longshore directions.
The
The IGU Commission on the Coastal Environment
The
have put forward ten reasons for long-term
changes in the shape and plan equilibrium of
the coastline (see Bird, 1983; in this volume),
namely:
1. Variations in fluvial sediment yield;
2. Variations in sediment yield from cliffs and
rocky shores;
3. Variations in aeolian sand supply;
4. Variations in sand supply from the sea floor;
5. Losses of sand onshore, offshore and
alongshore;
Note the very defi-
6. Interception of longshore drift;
7. Losses and gains due to sand mining, dumping,
dredging offshore, etc;

8. Effects of weathering;
9. Changes of sea level relative to the land;
10. Climatic variations.
Bremner (1983, in this volume) gives a good
account of the reasons for the shape of Algoa
Bay, on which Port Elizabeth is situated.
3.9 Aeolian transport
Sand particles washed ashore by wave action can
be dried, if they remain out of reach of the water
long enough, after which the sand can be readily
transported landwards or alongshore by the wind.
Thus sand which was once part of the sediment
balance in the wetted coastal zone can be piled
up to form sand dunes on land and can occasion-
ally cause the blocking of estuary mouths or
excessive sedimentation in harbours or built-up
areas. The purpose of this section is mainly
to point out that windblown sand transport cannot
be left out of a consideration of coastal pro-
cesses at any given site, as it might take on
substantial proportions which are not insignifi-
cant compared with the wave-induced transport in
areas of persistent strong winds. Windblown
transport is one of the mechanisms by which sand
can bypass a pronounced headland, provided of
course that sufficiently strong winds occur.
Research has shown that altho'ugh sand movement
under wind action can occur as surface creep,
saltation and suspended movement, the last mode
is rare due to the large density difference be-
tween sand and air. The shear stress at the sand
surface due to wind is one of the most important
factors determining the magnitude of aeolian
transport. Typically a critical stress exists,
below which the aeolian transport increases ap-
proximately with the shear velocity to the third
power, where the shear velocity is proportional
to the square root of the shear stress at the
surf zone. Other factors which influence the
volume of sand moved by wind are the sand size,
the grain size distribution, the degree of wetness
39
of the sand and the unevenness of the sandy
surface.
3.10 Methods of coastal protection
The type of coastal protection used depends on
the reasons for the erosion, i.e. types of pro-
tection necessary to cope with offshore erosion
will be different, for example, from those used
to control erosion due to a longshore gradient in
longshore transport. A few of the most commonly
used techniques are described briefly below.
(i) Groynes. Where the erosion is caused by a
longshore gradient in the longshore transport,
such as on the downdrift side of a coastal struc-
ture, groynes are frequently used to retard
erosion. It is important to note that groynes
only retard erosion aod do not prevent it. Fur-
thermore, the area of erosion is shifted to
downdrift of the groyne, or in cases where a
whole row of groynes are built, to downdrift of
the groyne row. Once the groyne row has been
completely filled, by-passing sediment again
amounts to the full longshore transport potential
and downdrift erosion is arrested. Groynes can
be either low-level, i.e. to the level to which
the final profile should accrete updrift, or
high-level, with its crest horizontal and above
water. Low-level groynes catch only part of the
longshore transport while high-level groynes in-
tercept a much larger portion of it. For this
reason the erosion downdrift of low-level groynes
is less severe than that downdrift of high-level
groynes.
Permeable groynes are also frequently used, but
their application is more in areas with a dominant
current-induced transport, i.e. areas with little
wave-induced transport. Low-level groynes are
not as unsightly as high-level groynes and are
from a morphological viewpoint preferable above
high-level groynes.
40
(ii) Seawalls. Seawalls are used mostly to pro-
tect some shore structure, such as buildings, a
parking area, etc., against both offshore and
longshore losses. Seawalls can be either sloped
or vertical and are usually placed as far land-
ward as possible, .preferably on or above the
highwater line. Wave reflection from the struc-
ture leads to additional turbulence and thus to
additional erosion in front of the seawall. The
toe of the structure must therefore be protected
adequately. In cases where the erosion is the
result of longshore gradients in longshore trans-
port, the maximum erosion will, as in the case of
groynes, be shifted to the downdrift extremity of
the seawall. Seawalls can be constructed of a
variety of materials, such as sandbags, gabions,
rock, slotted tiles, concrete and dolosse.
(iii) Underwater breakwaters. Underwater break-
waters are constructed at some distance from shore
to reduce incoming wave energy. These break-
waters work effectively only when waves are broken
on the underwater structure. Both the crest ele-
vation and the crest width are important vari-
ables in the determination of the amount of wave
breaking. A sand breakwater, such as the under-
water mound at Durban, reaches an equilibrium
crest water depth of approximately 60 per cent of
the initial water depth to the base of the break-
water. The crest of such a breakwater will there-
fore have to be protected against erosion to
prove effective, because a sand structure cannot
be maintained at a high enough level to be ef-
fective for all wave conditions.
Because underwater breakwaters reduce the in-
coming wave energy they are effective against
both offshore and longshore losses and, in addi-
tion, lead to more onshore sediment transport as
a result of wave height reduction. In cases where
longshore gradients in the longshore transport
cause the erosion, the area of maximum erosion
will again be shifted to the downdrift extremity
of the structure. However, because of the crest
level required, such structures are difficult to
construct and are quite costly.
(iv) Detached breakwaters or T-groynes A de-
tached breakwater is built parallel to shore at
some distance offshore, whereas a T-groyne is
basically an offshore breakwater with a connect-
ing breakwater (or groyne) to shore. Because of
the reduced longshore transport potential in the
lee of the breakwater, sediment accumulates in
this area and can, if the ratio between the dis-
tance offshore and the longshore length is correct,
even reach the breakwater, thereby forming a
natural sand connection to shore. Offshore break-
waters have been used very effectively at Tel
Aviv in Israel. An advantage is that by varying
the breakwater length/opening ratio, conditions
can be created to serve a variety of needs, such
as those of surfers, infant bathers, grown-up
swimmers, etc. The disadvantage is, again, that
erosion is shifted to the downdrift extremity of
the structure(s).
(v) Artificial nourishment. Artificial nou-
rishment is the most direct way of preventing
beach erosion, namely, by replacing eroded native
sand by some borrow material. This implies that
a suitable borrow material must be available,
which has a grain size equal to or greater than
the native material grain size. Because the
sediment cannot be replaced exactly where the
native material was eroded (because of diffi-
culties of placing sand underwater within the
breaker zone) the amount of borrow material
supplied normally exceeds the eroded volume.
Losses can be kept to a minimum by profiling
the nourished material as evenly as possible and
preventing steep seaward slopes. The material
can be supplied to the feeder area either via a
pipeline or directly by heavy vehicles. For a
regular system which requires a continuous supply
it will be most feasible to pump the sand to the
 41

feeder area via a pipeline. If the pipeline be-
comes too long (normally in excess of 1 to 1.5
km) a booster station has to be built. Where the
longshore drift is interrupted by harbour break-
waters (such as at Richards Bay), beach nourish-
ment is a very logical remedy for erosion on the
downdrift beach. The most suitable and logical
borrow area will, in such cases, be the updrift
beach. However, beach nourishment is also ef-
fectively used for beach building in areas where
offshore losses cause a denuded beach. Beach
nourishment is expensive, but is aesthetically
the most pleasing remedy for beach erosion. It
is especially effective when used in combination
with low-level groynes, which even out the rate
of longshore losses.
Water movement: current velocity
Wind-driven currents: Predictive techniques give
only guidelines to the current velocity; the real
flow pattern could be quite comples (see
Kullenberg, 1974).
Wind-driven longshore current: Sweeping sim-
plifications are usually used in the derivations
such as assuming spilling breakers propagating
over a plane beach and the assumption that the
waves adhere to the linear wave theory. Never-
theless, reasonable results can be obtained (use,
for example, Fleming and Swart, 1982).
Water movement: Flow patterns and water quality

3.11 Prediction of coastal processes for Tide-induced flows in shallow bays and estuaries:
design purposes
The flow is two-dimensional and perhaps even three-·
The design of coastal structures often involves
dimensional. Such problems can be solved only
the prediction of water movement and/or sediment
by running large finite element hydrodynamic
movement. The aspects of water and sediment
models on a large mainframe computer (see, for
movement required for design purposes could be
example, CSIR (1980), which uses the Leendertse
either straightforward or complex, depending on
model). The cost of running three-dimensional
the nature of the problem. Straightforward
models is normally prohibitive.
problems, which involve only the prediction of
current velocities or the magnitude of sediment
Effect of co as tal s.tructures on flow: The same
movement, can normally be tackled by someone only
as for the previous topic applies, except that
marginally knowledgeable in the field of coastal
the effects of wave motion should be included in
dynamics, although the interpretation of results
the formulation. These models are still being
obtained by the application of the relevant
developed although there are some "elementary"
equations might require some experience. Diffi-
working versions (such as that by Liu, Mei, 1974)
cult problems which would involve the prediction
which are always two-dimensional.
of flow patterns, water quality and topographical
changes in the coastal region, should always be
Dilution around submerged ocean outfalls: Two-
referred to specialists in the field, who are
dimensional models are available which can be run
both experienced and equipped to provide answers.
on mainframe computers which can be used in dif-
With this background, a guideline to the com-
fuser design (see, for example, Roberts, 1979).
plexity of the various problems and the restric-
tions of simplified techniques are given briefly
below.
42
Sediment movement: Magnitude
Longshore transport: Provided that the appro-
priate package-deal approach is used, reasonably
consistent results can be obtained for both the
total longshore drift and the onshore-offshore
distribution of longshore drift (see Swart,
Fleming, 1980). The main shortcomings in the
methods which are available at present are that
most of them are based on formulae for the pre-
diction of sediment load under uniform-flow
conditions, that wave breaking is mostly not
accounted for in the prediction of the volume
of sediment in suspension (a notable exception
is Nielsen, 1978) and that a plane beach is
assumed. Predictors which supply only the total
load, such as that in SPM (1973), are unsatis-
factory since they give no indication as to the
offshore variation in the longshore transport
rate.
Onshore-offshore transport: The technique deve-
loped by Swart (1974) is recommended for use.
Although the technique can be and has been applied
by using only a pocket calculator, its appli-
cation is simplified tremendously if a computer
model of the technique, which is readily avail-
able, is used. To make its application viable,
the number of wave conditions used should be
limited, although the model can be operated on
6-hourly wave data to cover a one-year period
without being prohibitive.
Beachfill predictors: Various simple predictors
are available to judge the suitability of a
beachfill design, such as those by Krumbein, James
(1965), James (1974) and Dean (1974). However,
since these methods are based almost solely on
the grain size distribution and neglect the effect
of waves, they should be treated with caution and
should preferably not be applied at all.
Sediment movement: TopographicaZ changes
Effect of coastal structures on sediment movement
and estuary mouth dynamics: Traditionally
this type of problem has been studied in physical
movable-bed coastal models. Recently there has
been a tendency to do more and more by mathe-
matical modelling. It is the opinion of this
author that even though mathematical modelling
is an essential step in the understanding of
coastal processes and can aid a speedy solution
to certain coastal problems, as should be abun-
dantly clear from this review, physical models
will always remain part of the technique for
solving coastal problems. This type of problem
should always be referred to specialists.
Beachfill design: The design of a proper beach-
fill should be based on the fundamental mechanisms
which cause sand movement. As such, the method
used by Swart (1976) is to be advocated, where
calculations are done with the relevant onshore-
offshore and longshore transport predictors for
the situation with and without the beachfill. A
typical application was for the design of a
beachfill for denuded beaches downdrift of Durban
harbour. Subsequent monitoring has shown the
results obtained from the Swart method to be
fairly accurate (CSIR, 1977).
Depth of burial of pipeline: As discussed in
Section 3.6, the onshore-offshore transport model
of Swart (1974) is most suited for this type of
prediction and provided the computer model is
available, can be readily applied.

REFERENCES
Aubrey DG (1983) Seasonal beach changes on
coasts with different wave climates. Proc. 1st
Sandy Beaches Symposium, Port Elizabeth, South
Africa.
Bailard JA (1982) Modeling on-offshore transport
in the surfzone. Proc. 18th Coast. Eng. Conf.
South Africa.
Battjes JA (1974) Computation of set-up, long-
shore currents, run-up and overtopping due to
wind-generated waves. Doctoral thesis, Delft
Univ. of Technology, the Netherlands.
Bird ECF (1983) Factors influencing beach
erosion and accretion: a global review. Proc.
1st Sandy Beaches Symposium, Port Elizabeth,
South Africa.
Bouws E and Battjes JA (1982) A Monte Carlo
approach to the computation of refraction of water
waves; J. Geophy. Res. 87, No C8, 5647-5662.
Bremner JM (1983) Long-term beach stability in
Algoa Bay as reflected by its planimetric shape.
Proc. 1st Sandy Beaches Symposium, Port Elizabeth,
South Africa.
Brenninkmeyer BM (1974) Mode and period of
sand transport in the surf zone. Proc. 14th
Coast. Eng. Conf. Copenhagen.
Busching F (1975) Uber die anderung von
wellenperioden im brandungsbereich; Mitteilungen,
Leichtweiss-Institut fur Wasserbau der Technischen
Universitat Braunschweig, Heft 40, 123-164.
Chapman DM (1983) Sediment reworking on sandy
beaches. Proc. 1st Sandy Beaches Symposium,
Port Elizabeth, South Africa.
Cokelet ED (1977) Steep gravity waves in water
of arbitrary uniform depth. Phil. Trans. Roy.
Soc. London, 286, series A, 183-230.
Crowley JB, Fleming CA and Cooper CN (1982) A
computer model for the refraction of non-linear
waves. Proc. 18th Coast. Eng. Conf. South Africa.
CSIR (1977) Management of the beaches in the
Durban bight, 2, Eng. Rep. CSIR Res.Rep.
Stellenbosch.
CSIR (1980) Saldanha Bay water-quality simu-
lation model, Report No.2: a two-dimensional
finite-difference model for hydrodynamical and
water-quality computations. CSIR Report C/SEA
8041.
Dean RG (1965) Stream function representation
of nonlinear ocean waves. J. Geophy. Res. 70,
No.18, 4561-4572.
Dean RG(1974) Compatibility of borrow material
for beach fills. Proc. 14th Coast. Eng. Conf.
Denmark.
Flemming BW and Fricke AH (1983) Selective
microhabitat colonization by interstitial
meiofauna as a function of grain size. Proc.
1st Sandy Beaches Symposium, Port Elizabeth,
South Africa.
Fleming CA and Swart DH (1982) New framework
for prediction of longshore currents. Proc. 18th
Coast. Eng. Conf. South Africa.
43
Goda Y (1964) Have forces on a vertical cir-
cular cylinder. Experiments and a proposed
method of wave force computation. Port and
Harbour Technical Res. Inst. Rep. No.8, Japan.
Inman DL (1957) Wave-generated ripples in
nearshore sands. Beach Erosion Board, T.M. No.100.
Iwagaki Y and Kakimura T (1967) On the bottom
friction factors off five Japanese coasts. Proc.
Coast. Eng. in Japan, Vol.10.
James W R (1974) Beach fill stability and
borrow material texture. Proc. 14th Coast. Eng.
Conf. Denmark.
Kamphuis JW (1975) Friction factor under os-
cillating waves. Waterways, Harbour and Coastal
Eng. Journal, ASCE, Vol.10, WW2.
Kjeldsen SP and Olsen GB (1972) Breaking waves.
Edited film version of project for M SC degree,
Technical University of Denmark.
Krumbein WC and James WR (1965) A lognormal
size distribution model for estimating stability
of beach fill material. Technical Memo. 16;
US Army Coast. Eng. Res. Center, Washington, DC.
Kullenberg GEB (1974) An experimental and
theoretical investigation of the turbulent dif-
fusion in the upper layer of the sea. Inst. Phys.
Ocean., Univ. of Copenhagen Rep (25) 272 pp.
Lenhoff L (1982) Incipient motion of sediment
particles. Proc. 18th Coast. Eng. Conf. South
Africa.
Leonard JE and Brenninkmeyer BM (1978) Storm-
induced periodicities of suspended sand. Proc.
16th Int. Conf. on Coast. Eng. Hamburg.
Liu P L-F and Mei CC (1974) Effect of a break-
water on nearshore currents due to breaking waves.
Ralph M Parsons Laboratory for Water Resources
and Hydrodynamics.
Longuet-Higgins MS and Stewart RW (1964)
Radiation stresses in water waves: a physical
discussion with applications. Deep-Sea Res. 11,
529-562.
Longuet-Higgins MS (1970) Longshore currents
generated by obliquely incident sea waves, I.
J. Geophy. Res. 75, No.33.
Longuet-Higgins MS (1980) The unsolved problem
of breaking waves. Invited lecture; included in
Proc. 17th Int. Conf. on Coast. Eng. Sydney,
1, 1-28.
Mogridge GR and Kamphuis JW (1972) Experiments
on bed form generation of wave action. Proc.
13th Coast. Eng. Conf. Vancouver, 2, 1123-1142.
Nielsen P (1979) Some basic concepts of wave
sediment transport. Series Paper 20, Inst. Hy-
drodynamics and Hydraulic Engineering. Tech. Univ.
of Denmark, Lyngby.
Roberts PJW (1979) Line plume and ocean outfall
dispersion. J. Hyd. Div. Proc. Am. Soc. Civil
Eng. 105, No HY4, 313-331.
Sasaki T (1975) Simulation on shoreline and
nearshore current. Proc. Speciality Conf. on
-Civil Eng. in the Oceans/III, Am. Soc. civil
Engrs. 179-196.
44

Short AD (1983) Sediments and structures in

beach nearshore environments, South-east Australia.
Proc. 1st Sandy Beaches Symposium, Port Elizabeth,
South Africa.
Silvester R (1974) Coastal Engineering, Vol. 2,
Elsevier Scientific Publishing Co., Amsterdam.
SPM (1973) Shore protection manual, Vol.I,
US Army Coastal Eng. Res. Center, US Govt.
Printing Office.
Stokes GG (1847) On the theory of oscillatory
waves; mathematical and physical papers. Cambridge
University Press, London, Vol.1, 314-326.
Swart DH (1971) Die effek van verwringing op
kus-profiele. Skripsie vir Meestersgraad aan
Univ. van Stellenbosch.
Swart DH (1974) Offshore sediment transport
and equilibrium beach profiles. Delft Hydraulics
Laboratory, Publication No. 131.
Swart DH (1976) Sediment transportation in the
coastal environment. Lecture notes for post-
graduate ECOR-course in Coastal Engineering, Port
Elizabeth, South Africa.
Swart DH (1976) Predicted equations regarding
Footnote
coastal transport. Proc. 15th Coast. Eng. Conf.
Hawaii.
In the context of the multidisciplinary approach
Swart DH (1978) Vocoidal water wave theory,
Vol.I: Derivation. CSIR Res. Rep. 357, 137 pp. to research on sandy beaches advocated at this
CSIR, Stellenbosch. Symposium, the following adaptation from the
Swart DH and Loubser CC (1979) Vocoidal water
wave theory, Vol II: Verification. CSIR Res. writings of Leonardo da Vinci (1452-1519) sum-
Rep. 360, 130 pp, CSIR, Stellenbosch.
marizes the required frame of mind for those
Swart DH and Fleming CA (1980) Longshore water
and sediment movement. 17th Coast. Eng. Conf. doing research on sandy beaches, whether it be
Sydney. on physical processes or not :
Swart DH (1982) The nature and analysis of
random waves in shallow water, Part I, text and
figures. CSIR Res. Rep. Stellenbosch. o sandy beach investigator
Swart DH and Crowley JB (1983a) Wave-generated
water flow through a porous sea bed. Proc. 1st do not flatter yourself that you know
Sandy Beaches Symposium, Port Elizabeth, South the things nature performs for herself
Africa.
Swart DH, Russell KS and Botes WAM (1983) but rejoice in knowing the multidisciplinary
Relationships between short and long waves in
purpose of those things designed by your own
shallow water. Submitted to Int. Assoc. for
Hydraulic Research Seminar on "Hydrodynamics of mind.
waves in coastal waters," Moscow.
Thompson EF (1980) Energy spectra in shallow
US coastal waters. US Army, Corps of Engineers,
Coastal Engineering Research Center, Technical
Paper 80-2.
Ursell F (1952) Edge waves on a sloping beach.
Proc. Roy. Soc. London, Series A, 214, 79-97
Van Hijum E (1972) P-golf. Unpublished note.
Visser CJ and Retief G deF (1979) Wave record-
ing and analysis. In ECOR Coastal Engineering
Course, Vol.I, Stellenbosch.
Wilton JR (1914) On deep water waves. Phil.
Mag. 6th series, 27, 385-394.
Wright LD, Short AD and Nielsen P (1982)
Morphodynamics of high energy beaches and surf
zones: a brief synthesis. Dept. of Geography,
Univ. of Sydney~ (referred to as (Short, 1982)
in the text).
 45

SEDIMENT REWORKING ON SANDY BEACHES

DAVID M. CHAPMAN (University of Sydney, 2006, Australia)

A sandy beach is actually a prism of All beaches exhibit dynamic response to

sediment, as shown in Figure 1. The prism is
subjected to reworking at various time scales;
if it were otherwise, the sandy beach would
evolve into a stable vegetated land form.
the impact of extreme events, the prism is
subject to extension at the landward limit to
include the incipient foredune (which is usually
colonized by sand tolerant pioneer species), and
more rarely the established frontal dune, and at
the seaward limit to include the zone which is
normally stable enough to be subject to
bioturbation. The sandy beach thus includes far
more than the recreational, or subaerial, beach
which is merely the berm and swash zone, a very
small proportion of the total sandy prism.
sandy beach is subject to reworking by aeolian,
biological and hydraulic processes, but
principally the latter.
variations in the natural hydraulic conditions:
tides, waves, surges, currents and rainfall all
induce mobility in beach sediments. For any
Under beach the area of the mobile cross section
provides an index of the variability in response
to the extremes of energy input. Knowledge of
the probable dimensions of the prism of a beach
is therefore of considerable val~e in assessing
potential storm erosion limits or potential
accretion and allows foresight concerning the
probable stability of engineering structures
built within the beach zone.
The object of this discussion is to take
a broad scale in both space and time of the
The
reworking of the prism of sand which represents
the total extent of the sandy beach. The
discussion is mainly empirical, and will draw
on a long period of observations, both

G -~ ESTABLISHED FORE DUNE

INCIPIENT FOREDUNE

.~e!:.~..~.. : .. 1 I·
11i~-~---- SUBAERIAL BEACH

SUBAQUEOUS BEACH

FAIR WEATHER/ {~} EXTREME EVENT

MODERA TE STORM ... ..

FIG.l SANDY BEACH SWEPT PRISM CROSS-SECTION.

 120' 140'
-"'
0'

20'-20-0

r--. r- Tp
40°·
r-:- l- I"--
tJ 10-0
............ 8-0 ~ Tz i---
- ........... r- """-
~
g 100
"t ~ I- ,..,
--
~ 120 ~ c::: 6-0 """-
C) ~
--- -
<...:> 1-0..
It>
V)
- - ---
200~ \
r-- r-
D/slance frOfT> shoreline (km) "-
" " " " 4-0 ......... -
~
i"- ~Hs r-- """- ........
..........
..........
~ i'- J I
.,. 2-0 I
............ I"i'. -
V)
It>
~ I I I
It> -
'i'-~ r---.....
1-0
E:: - r- WAVE HEIGHT £. PERIOD,
"- 0-8 - GOLD COAST, .........
-~ PROBABILITY OF EXCEEDANCE -...........
V) 0-6 ...... ......
::t:: ( log- normal curves fitted
0-4 to 5 years of waverider data)
I I I I I I I I I I I
0-5 I 2 5 10 20 50 80 90 95 99 99-5
p (%)
FIG.3 WAVE CLIMATE, GOLD COAST.

Note:Although zero crossing period is often used in discussions of

wave climate, author believes period at energy peak of spectrum to
be a more meaningful parameter.

kms
LEGEND
FIG.2 GOLD COAST ,AUSTRALIA.
@] ~~~g~S~~~~~TeE~!~A:::d~~:eoch The Gold Coast extends from Point Danger to Southport Spit.
ridges, Holocene estuorine sands, muds,
alluvia, backed by Pleistocene beach ridges Inset shows average continental shelf profile.
in places. Cudgen
~ ;~~i~~~I~t!:~r~zi~l,ae;aZ;;:d g~~>;I~~::' wSi~~te, Headland
Teriary basolt.

qualitative and quantitative, of a sandy beach
on the East Coast of Australia which is
considered to be not atypical of many coasts the
world over, and the discussion should be
applicable, perhaps with a change in scale, to
many other locations. Phenomena to be discussed
are related to both onshore/offshore and
longshore movement of sand.
Examination of the swept prism schema
illustrated in Fig. 1 reveals that the beach, as
defined by the recreational user (and indeed by
most lay people), is merely the residuum of
inshore processes. Most of the action takes
place beneath the water line, within the
subaqueous beach, and would not be apparent to
the recreational beach user or resource manager
who was merely concerned with evaluating the
subaerial beach. Hence, many otherwise useful
studies lack predictive and explanatory power
because of reliance of measurements on the
subaerial beach alone or measurements to shallow
wading depths. Since these measurements represent
information on the residuum of inshore zone
processes, sophisticated analyses carried out on
the data represent in effect the analysis of
'noise'. For this reason, considerable emphasis
has been placed in the present study on the
observation of sand movement in the inshore zone.
Techniques include beach and inshore survey as
described by Goetsch and Smith (1978) and
Chapman and Smith (1977), echo sounder traverses,
visuffimonitoring, sea-bed coring and surface
sampling and frequent air-photo sorties.
Repetitive surveys on an array of shore-normal
profile lines have yielded over 1000 surveyed
profiles for analysis.
The maximum and minimum elevations on any
shore-normal profile line which has been
repetitively surveyed over time form the vertices
of an irregular polygon. This polygon is in
turn the cross section of the prism of sand
forming the active beach, and was termed the
47
dynamic swept prism by Chapman and Smith (1980).
The example environment, the Gold Coast,
is a large zeta-form embayment some 30km long,
located at approximately 27 0 S on the east coast
of Australia (Fig. 2). The region is exposed to
moderately high wave energy (Fig. 3) dominantly
in the form of east to south-east swell, for
most of the year. There is no seasonal pattern
of cut and fill such as is reported from many
North American sites. Beach state is normally
intermediate-rhythmic after the classification
of Short (1979, 1980) - variability in the swept
prism would be greater than for the reflective
or dissipative cases. Tropical cyclones
(hurricanes) visit the area at low frequency,
but are highly significant in the coastal regime.
Beach sand is fine and well sorted
(x = 2.01 0 s = 0.44 0). Prior to the
extensive urbanisation which presently ,exists,
morphology could be seen to consist of remnants
of a subdued Pleistocene inner barrier, fronted
by an Holocene outer barrier with several dunes.
Using all surveyed profile lines, carefully
adjusted to a common x-datum, it has been
possible to evaluate long run relative
frequencies (~probabilities) of sediment
disturbance within the swept prism in the example
environment, under the wave conditions
summarised in Fig. 3. Examination of the
resulting plot (Fig. 4) immediately reveals the
great mobility of sand in the inshore zone (the
fine structure of inshore morphology has been
reported on by Wright, et al. (1979) and Short
(1979) and will not be taken up here). The
cross-sectional area of the swept prism, as
evaluated, between the P = 0.05 and P = 0.95
contours, is l618m 2 , which means that 1.62 x 106m3
of sand are mobilised within the prism per km
of beach (4.86 x 10 7m3 for the Gold Coast).
The carpet of activate9 sand extends \km
from the shoreline. Subtle depressions in the
48

-2
RL
(m) -4 -4 RL
-6 (m)
-6

-8 -8

-10 L-__-L__________J -_ _ _ _ _ _ _ _ _ _J -_ _ _ _ _ _ _ _ _ _~_ _ _ _ _ _ _ _ _ _L_~ -10

100m 200m 300m 400m 500m

(Arbilrary Horizonlal Dalum)

FIG.4 SWEPT PRISM,GOLD COAST.

Relative frequency of reworking.

PROBABILITY OF SURFACE REWORKING > 99% LANDWARD

-5 I SWEPT PRISM EFFECTIVELY CLOSES OUT

ZONE LANDWARD LiMIT l, HALLEMEIER (1981l.

-10

~
PRO\BA BI L ::E::O :~R :AFCES u:::~: ~::) :::~.
K

............... ~ TEXTURAL GRADIENT

....... ........ .... . . ...................... .
.. ....... ..... .
...... " . .......... . ..... , . . . . . . . . .......... .
.................
,
.......
.. ..... ......
.... . SEAWARD LIMIT OF LENS OF
...............
.. , ....... .
..... ................... :... ',' ... ........ :.... ,':-:->:-.:.:-:-:-:':-:-:-:-:-:';': : . GREYISH CREAM SAND
...... , ...

. "O'ISTA'N'CE'" 'OF'F'S'HO'R'E"':"'("~'e't~'~'~')""'"
500 600 700 800 900 1000 1100 1200 1300 1400 1500
~-20r-----~-------L------~----~~----~-------L------~------L-----~-------L----~

:!;
c:, -30

-35

_38L-____ ~ ______- L______ ~ ______L -____ ~ _______ L_ _ _ _ _ _ ~ _ _ _ _ _ _L __ _ _ _ ~ ______ ~

4000
______ ~

FIG.5 SWEPT PRISM,GOLD COAST.

Outer limits of reworking.

lower probability contours at x = ca. 200m and
400m reflect the existence from time to time of
storm troughs at these points. Similarly a
relatively high upper limit for sand at 50 Om
reflects the existence from time to time of a
storm bar in this vicinity. The slight depression
in the 0.05 probability contour and in the lower
limit at ca. 200m probably also reflect the
existence of scour in front of boulder walls
which have been located at some parts of the
Gold Coast within what, under normal
circumstances, would be the swept prism. The
phenomenon of toe scour in front of boulder
walls is discussed elsewhere by Smith and Chapman
(1982). The gentle gradient of the probability
contours in the region of the MSL intercept is
the result of both the characteristic rhythmic
topography of the study area, and of longshore
movement of sand in parcels or 'slugs'; the
latter phenomenom is discussed in more detail
hereunder.
The structure of the data follows a log-
normal frequency distribution of disturbance with
depth for those parts of the prism which are
permanently submerged. However the frequency
distribution is Gaussian at x = ca. 200m and
at ca. 100m reveals a split population
THE LIMITS OF THE SWEPT PRISM
The seaward asymptotes of the upper and
lower prism surfaces represent the seaward
limit of sediment exchange between bed and dry
beach, corresponding in broad terms with the
limiting depth of erosion as proposed by
Hallermeier (1977), or the seaward limit of
Swart's (1974) equilibrium profile. The
seaward limit of the prism therefore does not
correspond to surf base (the seaward limit of
incipient sediment motion) which is reached in
considerably deeper water.
The limits of the swept prism, as first
defined by Chapman and Smith (1980), are found
empirically, by survey. However, it is
49
illuminating to compare the effective seaward
limit of the swept prism as shown in Figs. 4
and 5 with the model of profile zonation"
suggested by Hallermeier (1981). Hallermeier's
model recognises three offshore zones
consisting of a littoral zone of maximum
sediment disturbance, a shoal zone which is a
kind of buffer region where expected surface
waves have neither strong nor negligible
effects on the sand bottom during a typical
year, and an offshore zone in which wave action
on bottom sediment is negligible. The seaward
limit of the swept prism as evaluated for the
Gold Coast is at 6.2m water depth (i.e. where
P = 0.95 and P = 0.05 limits become asymptotic
to mean sea bed level); whereas the calculated
limit for Hallermeir's littoral zone, defined
as the maximum depth for erosional cutting of
the near shore by yearly extreme waves," falls
at 7.3m for Gold Coast conditions.
Hallermeier also suggested that the seaward
limit to appreciable bed activity due to surface
waves would be found within the shoal zone.
Gold Coast evidence supports this assertion.
On the basis of wave and sediment entrainment
theory, Chapman (1978) calculated the point at
which sediments had a 50% probability of being
activated within the typical year (14.3m depth);
and moreover, the regularity of seaward-
fining as illustrated in Fig. 6a shows a
threshhold at 14.4m which would seem to indicate
the limit of active sediment sorting by
average surface wave conditions. In addition,
colour variations within seabed cores also
indicate a limit to active sorting of surface
sediments at l5.6m (Fig. 5).
It may also be anticipated that appreciable
bed activity due to surface waves would result
in winnowing of fines from surface sands: on
the Gold Coast, fines are virt~ally absent from
surface sands up to water depths of 10m, but
increase gradually thereafter (Fig. 6b).
50

Outside the prism limits, sorting of sand
according to the wave energy gradient at the sea
bed under normal weather conditions is a
superficial phenomenon: whilst a distinct
relationship exists between wave energy at the
sea bed (which varies logarithmically with water
depth) and sediment grain size, as shown in Fig.
6a, the pattern in cores taken through the surface
and extending to depths of over 1m below the bed
is virtually random.
(a)
Y=I·49-0·2093Inx).0-5636(ln",)2
5 6 7 8 9 10 12 14 /6 1820 25 30
water depth (m)
4
• '0,6
at 35.6m for the same environment. It may be
observed, however, that models alluded to
presuppose that wave drag is interacting with
clean sand; in fact, the sand in water depths
of ca. 30m will contain a proportion of silt,
and also organics (the latter reflecting
bioturbation).
The landward extension of the prism schema
shown in Fig. 1 includes the frontal dune and
foredune. In the case of a beach in its
natural state, landward extension under extreme
conditions will consume that part of the cross
section which is capped by a dune, so that the
event which causes the extension of the prism
is also responsible for the release of a
large volume of stored sand into the active
zone.
40
The landward limits of the example prism,
under the wave climate summarised in Fig. 3,
(b)
are shown in Fig. 7. These limits are reached

~
•
• :
: :.0'· :'"
5 6 7 8 9 10 12 14 !6 1820
water depth (m)
FIG.6 NEARSHORE SEDIMENT PROPERTIES,
GOLD COAST.
(a) mean grainsize of surface samples as a
function of In(depth).
(b) fines content of top O. Sm of core
samples as a function of In(depth).
Chapman (1978) also evaluated the
probability of sediment disturbance on the
entire offshore profile of the Gold Coast, after
Komar and Miller (1973, 1975).
showed (Fig. 5), that Gold Coast sands in water
depths of less than 5.9m have a greater than
.,.~;~:?;,o under moderate storm events, usually associated
... : r'·O'18
" . n-34
with extra-tropical depressions. However,
neither the wave nor survey data cover extreme
25 30
erosional events such as would be associated
with one or a cluster of tropical cyclones with
recurrence interval in excess of, say, 20 years.
Wave penetration under extreme events will
be affected by variables which include height
of storm surge, height of waves, length of time
storm waves are sustained, volume of sand in the
dune, and so on. However, it is possible to
make some estimates of probable cut on the
basis of recorded observations, scientific
judgement, and the position of placer deposits
The model
within the dunes.
The placer deposits (known in Australia as

99% chance of being activated during a typical
year, whereas sands in water depths greater than
32.5m had less than 1% chance of being activated
during a' typical year. In comparison, the seaward
limit of Hallermeier's shoal zone is calculated
seams of heavy mineral sand), winnowed from the
same sediments as those which form present-day
Gold Coast beaches and emplaced at the limit of
wave uprush under extreme storm events, have
traditionally been regarded as 'fossil' deposits.
However, it is possible that they provide evidence
 1948
PRESENT SURVEY 8
""....... ____ < _-_~ \~ ASSUMED 6
l-- . . . _ ...... ; -= -» PRE- 1940's
RL 4 4 RL
(m) 2 2(m)
MSL .......... MSL
o~ o
0:50 -2
-21 I I
=-=-====-=====LL
I
-100m -50m Om 50m 100m

FIG.8 PLACER DEPOSITS,GOLD COAST.

Representative positions before mining.

SECONDARY ARMOUR
BOULDER FACE
CLAY - SHALE FILTER \ I ARMOUR

SAND BEFORE CYCLONE 10(" ) I 0( ) 10( ) I BOULDER WALL BEFORE

""':':"::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::7'<:~7~-,r~~ CYCLONE
Q +'--r~ BEACH BEFORE
EROSION • ,) ~:~.~YCLONE
y-- SCARP cA.<1')<.....:.:-:.:-:.:.::-:::::::::-::::"7::-'77
51 IBEACH DURING AND ./ >-",<r"-'" SAND MS.L.
AFTER CYCLONE ~ ~~~
4 . . . . . . . . . . . . . . . . . .... .................................. .
3 .. ' ~·~·~:.:;6~;;:~·{2;~~~~:;:.:;:::::~:~::::~::::···.·····54·N·D·············:::::::::::::·:·:·:·:·:·:·:·:·:·:·:;:':':':':':':':':';';';':;:::':':':':':':':':':':':':.:.:.:.:.:.:.:.:.:.

2 aD ~:~: O/~Oop"Q1~'Q60:~·ao'O·~~~=:;;:;~~UR AND

0 ~ SHALE AFTER
o I I CYCLONE
o I 2 3 4 5 BOULDERS AFTER CYCLONE SAND
SCALE (metres)

FI G. 9 BOULDER WALL FAI LURE.

Location of debris in swept prism subsequent to event.
(Modified from Smith. Chapman, 1982).
<..n
52

of the inland penetration of erosional events of events which created the features undoubtedly
very low probability, say in the recurrence had effects, including the formation or removal
interval range of 10 2_ 10 3 years. of other features, for which not a trace of
evidence remains.
A representative composite plot of the position
of such heavy mineral seams on the Gold Coast, as Observations on the Gold Coast suggest that
they existed prior to sand-mining, is shown in the lower prism limit under extreme storm condit-
Fig. 8. Data used were collected by Beasley (1948), ions may be underlain by a 'semi-quick' sediment
Gardner (1955), Bayly (1952), and Brooks (1953). layer. Direct measurement during storm conditions
is not possible, but the existence of such a layer
With regard to the seaward-most lens at ca.
is suggested by events such as the collapse of a
50m, it may be noted that erosional cut of this
order would be required locally to breach the boulder wall under extreme storm attack and
subsequent settlement of the armour to a depth of
Southport Spit (N of Surfers Paradise in Fig. 2).
over 1 metre below the sand surface within a matter
Such a breach was actually recorded in 1936.
of minutes. Observation of the event in progress
An inland erosional penetration of over 100m
suggested that the boulders sank through a semi-
appears immense by comparison with the much
quick layer rather than having had the sand
smaller (usually less than 40m) incursions
, scoured from around them by the waves. The result
recorded since the 1930 s. However, comparison
of this event is illustrated in Fig. 9. It is
of recorded observations on erosion, evidence
discussed in more detail by Smith and Chapman (1982).
from old photographs, and known and inferred storm
Drilling through eroded beaches after
characteristics (only one tropical cyclone-
storm attack has also revealed layers of coarse
generated set of storm waves was captured in the
shell and stone particles in layers between 1m
data set used for Fig. 3, but meteorological
and 3m beneath the storm beach profile. At first
observations of cyclone characteristics are more
these coarse layers were thought to be layers of
complete) indicated that dune-cut of ~ 100m might
"armour" indicating antecedent erosion profiles.
be expected with a return period in the range
However subsequent inspections during major storm
100 - 200 years. Some confirmation of the latter
events indicated that many centimetres of sand on
inference has been provided from a Delphi study,
the beach surface were in partial motion and that
using a panel of Australia's leading coastal
"armour" material in surface deposits was not in
practioners, which is in progress (Chapman, in
evidence. It now appears that the "armour"
prep.).
material, having a lower specific surface than the
Parenthetically, it may also be observed that
sand matrix, merely sank to the bottom surface of
evidence from sedimentary bodies represents in a
the swept prism whilst it was in a partially
very real sense a 'single side-band' signal. The
liquefied condition.
FIG.7 SWEPT PRISM,GOLD COAST.
Inner limits of reworking. At the landward limit, modes of cut and
recovery are not complementary. During periods
6
SECOND
RL 4 DUNE 4
(m) 2

o MS/C .. ..."""<~~ _:r':-I

-4
-50m Om 50m 100m
 53

(b) SAND
SLUGS
(PULSES)
~ / l
~T/ME-C7
r-- .

(c) MAJOR ~
EROSIONAL ::;
EVENTS
(TROPICAL ~
CYCLONE) ~
"-
~

---------p
+ 30m
(d) HEADLAND
BYPASSING

-30m TIME .....

------?
~ --~=------------
---=-===-------
TIME-

+30m

( f)
2 4 5
CU M U LAT I VE r--_~~+~~,.LL~C.- __3L--~;::::-----'_+_----L­
EFFECT

-30m
ARBITRARY TIME SCALE. YEARS

FIG.l0 VARIABILITY OF SUBAERIAL BEACH.

Schematic illustration of the effect of various processes.
54
of cut (which may involve recession of the dune
scarp) rapid removal of sand occurs from the
subaerial part of the swept prism; however
during periods of recovery the replacement of
this sand occurs over a longer period and in
addition involves the development of an incipient
foredune. Initial stabilization of this foredune
requires pioneer dune species which are sensitive
to recreational pressure so that the stabilization
of the replaced sand by woody vegetation may be
extremely slow or non-existent on a
recreational beach. In this way a mechanism for
cut without replacement exists.
VARIABILITY IN THE SUBAERIAL SANDY BEACH
A large exposed ocean beach is subject to
a variety of influences acting at widely
different scales of space and time. The result is
a somewhat bewildering set of possible
combinations. Some of the major contributing
factors and their cumulative effects are
illustrated schematically in Fig. 10, where
Fig. lOa illustrates high frequency variations
resulting from minor changes in wave height and
direction, different tidal cycles, local wind
effects, and so on; Fig. lOb illGstrates the
effect of slugs of sand moving through the
littoral drift system; Fig. loc the effect of
major erosional events; Fig. lod the effect of
headland by-passing of sand (which also takes
place in a pulsatory manner); Fig. IDe the
effect of historical sea-level change, and the
cumulative effect of all the. above phenomena,
graphically integrated, is shown in the final
plot, Fig. l[lf.
High frequency fluctuations are introduced
by factors such as diurnal, neap/spring and
other tidal cycles; changes in wave height and
direction; rip cell development and decay; bar
and shoal development and decay, and so on.
The combined effect of such phenemena as
illustrated schematically in Fig. lOa, is to
introduce a great deal of obvious and
measurable change into the subaerial beach and
surf zone bed-forms. However, such changes
activate a very small proportion, possibly 10%
or less, of the total volume of the swept prism.
High frequency fluctuations are perhaps
best exemplified by the variability of High
Water Mark. Even where a consistent definition
is employed (Gibb (1978) found that no less than
seven different definitions of HWM had been used
for cadastral purposes in New Zealand), the
fact that high water intersects a gently
sloping contour within the prism cross section
will create difficulties where HWM is used as
a datum (cf. Fig.ll).
100
HWM
o
o 1000 2000 3000
BEACH LENGTH IN METRES
FIG.11 VARIABILITY OF HIGH WATER MARK
Bi-monthly surveys of debris line on a
section of the Gold Coast.Arbitrary origin.
The tendency for littoral drift to occur
in the form of 'slugs' (or pulses of sand) was
originally detected on the Gold Coast in 1974,
although the significance and ubiquity of the
phenomenon was not appreciated until somewhat
later. During 1975, slug transport was positively
identified by survey, and subsequently
reported by Chapman (1978b).
Typically, a sand slug on the Gold Coast
may produce a wavelength in the subaerial
beach of the order of 3km, with an amplitude of
20 - 50m.
It is important to note that (a) slug
transport creates a pattern of "full" and "lean"

beach segments which is not necessarily related
to purely local conditions of sand supply or
process, and (b) observation over a relatively
long period may be necessary to detect the
pattern of slug behaviour. A beach normally
in the intermediate-rhythmic state (cf. Wright
and Short, op. cit.) may exhibit considerable
increase in the width of the subaerial beach if
it moves into the fully accreted, reflective
state. However, such a change involves merely
the transfer of sand within the prism limits,
whereas the arrival of a slug of sand will
augment both the subaerial and subaqueous zones.
Sand bypassing of headlands or large
groynes extending into "blue water" beyond the
surf-zone is commonplace. Observations on the
Gold Coast have shown that bypassing does not
take place as a continuous flow, but in the form
of slugs. Sand appears to accumulate on the
updrift side of the obstruction to form a slug
which, when activated, may bypass very rapidly.
For bypassing to occur sand on the seabed must
be mobilised by wave energy, and the deeper the
water the larger the waves required to initiate
movement. It would appear that sand accumulates
until the critical combination of wave height
and direction occurs; bypassing them takes place
until wave conditions again become subcritical.
Large parcels of sand are moved on the
Gold Coast under such conditions. In early
1974 a slug estimated at 1 x 10 6 m3 bypassed
Point Danger (see map, Fig. 2) during a period
of high energy conditions: plots of surveys
taken before and after movement are illustrated
in Fig. 12.
FIG.12 HEADLAND BYPASSING.
o o
-10
______ APRIL 1974
-15
RL AUGUST 1974
-20
-25
-30
o 100 200 300 400 500 600 700 800 900 1000 1100
DISTANCE OFFSHORE, m
55
5 3
On a smaller scale, at least 4 x 10 m
of sand was observed to accumulate on the
updrift side of the Currumbin groyne over a
period of nearly four years of mainly mild
conditions; yet over several days during which
HSig = 3m, it was observed that approximately
5 3
1.5 x 10 m of sand bypassed. Shortly
afterward, Foster (pers. comm.) observed a
further 2.5 x 10 5m3 of sand to bypass in less
than a week, adding about 1m to the average
thickness of sand in the nearshore zone on
downdrift Palm Beach.
Initiation of slugs has been observed
to result from the flushing of river estuary
channels during floods, and, in the Gold Coast
case, from intermittent erosion of Pleistocene
beach-ridges located updrift. These Pleistocene
dunes, which during Holocene time have acted
as the principal source for Gold Coast littoral
drift (Chapman, 1981), have, over ten years of
observation, been observed to suffer cut-
without-replacement as a result of infrequent
severe storms, the sand thus removed being
transferred to the nearshore zone where it is
available for longshore transport. Between
storm events, the dunes do not recede, but a
very small incipient foredune develops and
intra-dune sandrock becomes colonised by
tubeworms (Galeolaria sp.).
It is generally assumed that no
significant sea level change has occurred in
late Holocene time, and that no secular sea
level rise is presently occurring, on the East
Australian coast (Chapman et al. 1982).
However, relatively small fluctuations of sea
level, perhaps too small to be of
geomorphological significance, may nevertheless
-10
cause shoreline oscillations which, compounded
-15
RL with other factors described, may be of
-20
engineering or planning significance.
-25
Using the Bruun hypothesis as a guide,
-30
and taking average yearly sea levels from the
56
Fort Denison (Sydney) tide gauge (longest record
in Australia - located on stable Mesozoic strata
singularly free of recent tectonic activity)
it is possible to estimate shoreline variations
on the Gold Coast attributable to variations in
recorded sea level alone. Changes in recorded
sea level are small (up to 8cm from datum over
periods of ~ 20 years), but could result in
apparent "erosion" or "accretion" of up to
0.2m/yr on a low gradient beach.
CONCLUSION SOME IMPLICATIONS FOR COASTAL
MANAGEMENT
Effective management responses demand
knowledge of sand movement within the entire
swept prism. Management schemes based on process
variables and observation of responses such as mean
sea level intercept position or foreshore slope
cannot provide effective guidance, since much of
the beach re~ponse to process input takes place in
that part of the swept prism which is below the
water line.
Reworking of sand within the normal limits
of the swept prism absorbs and dissipates wave
energy impinging on the shore line. If structures
are located within the normal limits of the swept
prism the energy otherwise absorbed by the sand
which is now removed from the limits of the active
prism must be dissipated elsewhere, either by the
structure, by the prism seaward of the structure,
or by reflection of wave energy (cf. Fig. 13, and
Smith, Chapman, 1982).
NATURAL SWEPT PRISM
FIG.13 IMPACT OF STRUCTURE WITHIN
SWEPT PRISM.
(After Smith,Chapman, 1982).
STRUCTURE
AREA ® ~ AREA @
The large proportion of the swept prism
in the subaqueous zone is significant in
accounting for the re-distribution of beach
fills, since re-distribution can involve
substantial losses from the subaerial beach
without material being lost from the prism.
If nourishment material applied is of such a
nature that the upper surface of the prism is
induced to adopt a lower overall gradient, a
large volume of fill material would be
absorbed in order for this redistribution to
take place, but may result in little gain in
width of the subaerial beach. Since the
subaerial beach is merely the residuum of
inshore processes, any "design profile"
adopted for beach nourishment planning which
is inconsistent with the behaviour of proposed
fill material in the total swept prism under
local wave climate will prove impossible to
maintain.
Moreover, it may also be inferred that
nourishment of a sandy beach with a volume
of fill material that is large relative to the
amount already present will induce either a
shift to a dominantly more reflective beach
state (if fill is substantially coarser than
native), or to a dominantly more dissipative
state (if fill is substantially finer than
native (cf. Short, 1979, and this volume).
The day-to-day, week-to-week variability
of the natural sandy beach is the resultant
of many factors, which renders understanding of
 57

.............
................... . . .................
................ . ...
. . . . . . . . . . . . . . . ., ....
................... ...................
.................
............... ...
................

~
~ ........ ..
V)

................. CHANGE
~ ...................................
.................
..................
~ .................
................
................. ..
...............
.................. . .
................
................. ..
~ ................
.................
................. ...
..................
V) ...............
.................
................ .
................
" . , ... , . , . , . , . , .
...., . , ... , . , . , ..... .
, .,
lI... ... .........
, .....
,., .
....... .
. . . . . . . . . . . . . . . . . . ... .... .......... , ...
,
C)
................... ..,., .......
..... ...
... , .. .
,

..................... .,, .... , . , . , .......

'
,., ,

. . ..... . . . . . . .. , ........
,.,

, ... .. , .... ,
, , , , , , ,., ,
,."

~
:::):::::::::- :)c) ..~TEP FUNCTION
~
; :::»»:<:::::::::::: TIME
PRESENT I:. IMMEDIATE
GEOLOGICAL PLANNING
PAST TIME FOR
TIME TIME
WHICH SHORT TERM
SCALE SCALE
DATA ARE AVAILABLE

, . , . , ....... .
.........
........... .. , . , . , .... ., ....
., .... ., ... ,,. ....
. , ..... . ....
...... ., ..
, ....
........
" ,'

............ ,., .
, ,
, , , " ,"

~
~ .................. .
V)

, . , . , . , . , . , .....
CHANGE
~ ...............
...........
......
......... . . . . . ......
. , , .......
,. ,

.... . . . ......
, . , ... , .........
~
,",
,
' , ,

, . , .. ' , . , . , ... .
.. ... .
'
, ,'"
.
....... , ......
,.,
... ... ... , . , . , .. , .
,'" ,-,

~ . ', ,....... ......... ..

,'
,.,

..........
"

,...
,".
,.,.,., .
, ,
,..,,..,,..,, ......
... ... .
,

V) ............
. , .,
...... , ... , . , . , . , .
, . , . , . , ......... .
,
,.,
,

, . , ... , . , ... , . , .
, ... ....... , . , . ,
lI...
,
,

C)

FUNCTION

PRESENT I:. IMMEDIATE TIME

GEOLOGICAL PLA NNING
PAST TIME FOR
TIME TIME
WHICH SHORT TERM
SCALE SCALE
DATA ARE AVAILABLE

FIG.1Q SYSTEM STATES.

Schema showing hypothetical behaviour of a system over time,
and difficulties of extrapolating from data obtained over a
limited time frame.
58
the meaning of observed variability frustrating:
the underlying low frequency phenomena are
clouded by high frequency and sometimes violent
events of little long-term significance. It is
however, the low frequency/high magnitude
events which are most important for planning and
engineering purposes. These events set the
limits to the system under consideration,
defining its long term character and providing
the bounds which must be taken into consideration
by the planner or engineer.
It is possible to distinguish three time
scales of shoreline change. These are long,
intermediate, and short, and refer to the
recurrence interval of events which cause change,
rather than to the scale of the changes them-
selves. Long-term or 'geologic' scale embraces
those changes which have occurred over the last
several thousand years. During the mid to
late Holocene (the stillstand period of sea
level), coastal landforms basically attained the
shape and size seen today. Geologic events
have moulded these landforms by onshore wave-
induced sand transport, wash over processes,
aeolian sand movement, tidal delta formation,
and so on.
At the other extreme, there are short-term
changes associated with contemporary wave, wind
and current processes. These day-to-day,
week-to-week, and even month-to-month changes in
shoreline position and configuration are
related to variations in wave' and wind energy
and tidal amplitude, and are expressed in the
various beach states observed (cf. Short, this
volume) .
In the middle is intermediate or medium-
term change which operates over a time scale
of one to 100 years. This is the time scale
of planners and engineers. It encompasses
aperiodic but major storm events. Although
very relevant to any human activity on the
coast, it can only be understood in the
perspective of long and short term changes,
It is apparent that the combined effects
of the phenomena discussed in preceding
sections may 'produce a state of affairs in
which "noise" at many scales is so great as
to make exceedingly difficult the detection of
any trends at the intermediate scale, such as
progressive decrease in sediment supply, or
secular beach recession, important for planning
purposes. Moreover, the aggregation of the
random effects of a number of different
phenomena, such as those illustrated in Fig.
lOa to Fig. IDe, may be expected to give rise
to quite spurious "trends", if observed over
the short term. Considering the cumulative
plot in Fig. 10f, and arbitrarily assigning
to it a time scale of five years, it may be
seen that extrapolation from detailed
observations made in anyone year merely
produces nonsense.
The problem is, however, of considerable
practical importance for planning and
engineering decisions which impact the
immediate future. These must be made with
reference to an intermediate time-scale
(1-100 yrs.) and depend on (a) a knowledge of
past trends over at least comparable time
spans, and (b) an ability to predict future
trends with a reasonable degree of confidence.
Confident prediction requires the
consideration of intermediate time scales in
a framework of long-term and short-term
knowledge. To ignore this framework is to invite
generation of an erroneous planning base. The
problem confronting the decision-maker can be
illustrated in a more general way as in
Fig. 14, which shows available data sets in
relation to a number of hypothetical system
states spanning long periods of time. In the
present context, "system state" may be
represented by variables such as MHWM or
prism cross-section: tl - t2 is the time
 59

~
2
I Time = 00 hrs
R.L. 0
-I
-2
VE=IO
-~
-4
Rhythmic
-5 (Crescentic Bar)
~

2
1 Time = 51 hrs
R. L. 0
-I
-2
-3 VE=IO
-4
-5
Transition
~

2
I Time = 95 hrs
R. L. 0
-I
-2
-3 VE=IO
-4
Reflective
-5 (CHI467)

o 70 100 150 200 250

DISTANCE SEAWARD

FIG.1S PROGRESSION OF BEACH STATES.

Change from barred to reflective states on
one profile line, Gold Coast.
60
covered by the data set. The earth scientist
may be using this data set in an attempt to
understand development of features in recent
geological time (i.e. extrapolating backward),
whilst the engineer or planner may be using the
same data for decision-making (extrapolating
forward).
In each of the cases illustrated in Fig. 14
the dashed line represents the true "mean"
system state, but the heavy line shows the
extrapolation of the trend as revealed by the
data set in tl - t2. In the case of a "steady-
state" system which has relatively small
fluctuations over time, measurement of the
system state at several points within the small
time frame may give an accurate picture of the
mean state of the system. On the other hand,
where the system is subject to incremental
change, sampling over a small time interval may
either give a misleading picture of the nature
of the trend or, if the data are collected with
the expectation that a "steady state" obtains,
the interpretation of the data may be such that
the trend revealed is considered spurious.
However, this author believes that many
sandy beaches are characterized in both the
short and intermediate time frames by what
might be called "multiple stable state" behaviour,
as shown schematically in Fig. 14. This is akin
to the "step-functional" behaviour considered
by Dury (1980) to apply to certain meteorological
phenomena wherein the system state switches
aperiodically between "high" and "low" mean
states, both significantly different from the
long-term simple mean.
In the short-term time frame, it is
apparent from the analysis of Short (1979) and
subsequent observations by the writer within
Short's schema that, whereas extremes of beach
form may be recognized (reflective and
dissipative), the intermediate forms are not
arranged along a continuum, but fallout into
four recognizably discrete morphologic states.
Whilst beach and inshore morphology may persist
in one of these states for weeks or even
months at a time, transitions between them are
extremely rapid, occurring over one or a few
tidal cycles (cf. Fig. 15).
In the intermediate time frame, it is
suggested that observed behaviour can be partly
explained by considering that a major event
(major storm, say, or arrival or a large slug
of sand) may establish a new stable state, or
equilibrium, around which the system will
fluctuate for some time, rather than returning
to its pre-existing state. Thus there is no
unique "equilibrium" state, but a variety of
possible equilibria at which the sandy beach
may remain relatively stable. Assuming that
the system is not undergoing secular change,
the ultimate limits of the system (as indicated
schematically by the parallel dashed lines
around the step function in Fig. 14) become
significant for the engineer or planner, rather
than the oscillations around temporary
equilibria. The ultimate limits of reworking
for the sandy beach will be represented by
phenomena referred to above, such as heavy
mineral seams, or seaward discontinuities in
sand colour or texture.
ACKNOWLEDGEMENTS
This chapter owes a great deal to my
association with A.W. (Sam) Smith, consulting
engineer and Gold Coast resident, and to
cartographers Johanna de Roder and John Roberts
of the University of Sydney. I am also grateful
to the Gold Coast City Council and the
Queensland Dept. of Harbours and Marine for
assistance in data collection.
REFERENCES
Bayly, M.G. (1952) Beach sand mlnlng in
Queensland. Qld. Govt. Min. J., 53: 742-755.
Beasley, A.W. (1948-50) Heavy Mineral Beach
Sands of Southern Queensland, Part 1 - The
 61

Nature, Distribution and Extent, and Manner Smith, A.W. and Chapman, D.M. (1982) The
of Formation of the Deposits. Part II - behaviour of prototype boulder revetment
Physical and Mineralogical Composition, walls. Proc.18th Int. Coastal Eng.
Mineral Descriptions, and Origin of the Heavy Conf., ASCE (in press).
Minerals. Proc. Roy. Soc. Qld., 59: 109-149 Swart, D.H. (1974) Offshore sediment transport
+ Plates II-IV.61: 59-104 + Plates I-VI and equilibrium beach profiles. Delft
Brooks, J.H. (1953) Mineral sand deposits, Tugun. Hyd. Lab. Pub. 131, 302pp.
Qld. Govt. Min. J., 54: 493-494. Wright, L.D. and Chappell, J. and Thom, B.G.
Chapman, D.M. (1978) Management of sediment and Bradshaw, M. and Cowell, P. (1979)
budget, Lower Gold Coast, Queensland, Ph.D. Morphodynamics of reflective and dissipative
Thesis, University of Sydney, 257pp. beach and inshore systems, southeastern
Chapman, D.M. (1978b) Management of Sand Budget, Australia. Mar. Geol., 32:105-140
Kirra Beach, Gold Coast. Proc. Fourth Australian
Conference on Ocean Engineering, pp. 19-24
(The Institution of Engineers, Australia).
Chapman, D.M. (1981) Coastal erosion and the
sediment budget, with special reference to
the Gold Coast, Australia. Coastal
Engineering, 4: 207-227.
Chapman, D.M. and Geary, M. and Roy, P.S. and
Thom, B.G. (1982) Coastal Evolution and
Coastal Erosion in N.S.W. Sydney, The
Coastal Council of N.S.W., 341pp.
Chapman, D.M. and Smith, A.W. (1977) Methodology of
a large scale tracer experiment. Proc. Third
Australian Conference on Coastal and Ocean
Engineering, pp. 185-189 (The Institution of
Engineers, Australia).
Chapman, D.M. and Smith, A.W. (1980) The
dynamic swept prism. Proc. 17th Int. Coastal
Eng. Conf., ASCE, pp. 1036-1050.
Dury, G.H. (1980) Step functional changes in
precipitation at Sydney. Aust. Geog. Studs.
18.(1); 62-78.
Gardner, D.E. (1955) Beach Sand and Heavy mineral
Deposits of Eastern Australia. Commonwealth
of Australia, Bureau of Mineral Resources,
Bulletin No. 28.
Gibb, J.G. (1978) Rates of coastal erosion and
accretion in New Zealand. N.Z.J. Mar. Freshw.
Res., 12(4) 429~456.
Goetsch, F. and Smith A.W. (1978) Inshore beach
surveys. Proc. Fourth Austral'ian Conference
on Coastal and Ocean Engineering, pp. 176-179
(The Institution of Engineers, Australia.)
Hallermeier, R.J. (1977) Calculating a yearly
limit depth to the active beach profile.
Tech. Pap. 77-9, U.S. Army, CERC.
Hallermeier, R.J. (1981) A profile zonation for
seasonal sand beaches from wave climate.
Coastal Engineering, 4: 253-277.
Komar, P.O. and Miller, M.C. (1973) The threshold
of sediment movement under oscillatory water
waves. J. Sedim. Petrol., 43: 1101-1110.
Komar, P.O. and Miller, M.C. (1975) Sediment
threshold under oscillatory waves. Proc.
14th Int. Coastal Eng. Conf., ASCE,
756-775.
Short, A.D. (1979) Three-dimensional beach
stage model. J. of Geol., 87:553-571.
Short, A.D. (1980) Beach response to variations
in breaker height. Proc. 17th Int. Coastal
Eng. Conf., ASCE, pp. 1016-1035.

BEACH CHANGES ON COASTS t~ITH DIFFERENT WAVE CLIflATES
D. G. AUBREY (Department of Geology and Geophysics, Woods Hole Oceanographic Institution)
SYNOPSIS
Seasonal and longer-term beach
variability is quantified for seven U.S.
beaches exposed to widely varying wave
climates. One U.S. west coast location
(southern California) and six U.S. east
coast locations (from North Carolina to
Massachusetts) form the basis of this
study. Wave exposure varies from complete
exposure to open ocean waves, to partly
sheltered locations, and finally to nearly
complete sheltering where locally-generated
waves dominate. Beach response was
documented with beach profiles distributed
along each of the seven coastal locations,
spanning a minimum of 'five years of observa-
tion. Frequency of measurement was at least
once per month, with periods of more intense
weekly sampling lasting for up to two years
(southern California location). Wave climate
was either measured directly or estimated
from hindcast and/or compilations of ship
observations. Consequently, wave informa-
tion varies in detail from joint statistics
of wave height, frequency, and direction, to
compilations of local storm history (and
hence inferred wave behaviour). Magnitude
of annual beach variability ranged from 3.3 m3
per metre of beach to 0.2 m3 per metre of
beach, with the greatest variability in
regions exposed to open ocean waves (most
63
severe wave climate) and the lowest
variability along protected coasts (least
severe wave climate). All open coast
locations studied had a seasonal variability
whi ch accounted for at 1east 50% of the beach
variability. Protected coastal locations
had less pronounced seasonal signatures.
These seasonal and aseasonal beach responses
mirror corresponding seasonality (or lack
thereof) in wave and storm climates. The
study re-emphasizes the need for careful
measurement or estimation of coastal wave
climate to enable predictive modelling of
shoreline behaviour, and discusses different
analysis techniques for analyzing changes in
beach profiles through time.
I NTRODUC nON
Quantification of spatial and temporal
scales of beach change is vital to a wide
variety of scientific and engineering
investigations of nearshore environments.
Vertical elevation changes of 2.5 metres,
mean shoreline transgressions on the order
of 50 metres, and volume changes on the
order of 10 2 m3/m of beach length can
occur on time scales of hours, drastically
altering the physical and biological
characteristics of beaches (Fig. 1).
Intertidal benthic communities must be able
to respond quickly and efficiently to these
profile readjustments, since habitat, oxygen
levels, nutrient retention, and other
environmental factors can be significantly
64
RANGE 2
, 2880
2 2480
-2
90 60 30 o the residence time of months for oil in
OFFSHORE DISTANCE (m)
1) Beach erosion resulting from a series of
storms battering Santa Barbara,
California, in February, 1980, causing
vertical cuts in the beach of up to 2.5
metres, and horizontal beach retreat of
up to 60 metres.
altered in a short time (Steele, Munro and
Giese, 1970; Parr, Diener, and Lacy, 1978).
The degree of seasonality in these changes
similarly may affect the viability of
nearshore benthic communities, since the
timing of beach changes interacts with the
developmental stage of the benthic
community. The seasonality and magnitude of
beach changes also playa direct role in
retention of hydrocarbons in beach sands
with subsequent impact on biota, a
consideration in many beaches exposed to
naturally-occurring or man-induced
hydrocarbons in the shallow nearshore.
Rapid beach changes of large magnitude will
help rid the beaches of oil naturally;
longer-lived beach hYdrocarbons may limit
benthic diversity or density. An example of
this longer time scale for hydrocarbon
residence was observed along beaches in
Santa Barbara, California, by the author
(unpublished data). A series of major
storms in February, 1980, caused marked
erosion along the beaches in Santa Barbara,
exposing underlying beach material which had
not been disturbed in the preceding decade
(Fig. 1). During the later stages of the
storm, an oil-impregnated horizon which had
been deposited during February, 1969 was
exposed, and eroded from the beachface. In
this instance, the residence time of the oil
was of the order of 1U years, in contrast to
beach sands emanating from local, natural
oil seeps in the Santa Barbara Channel.
Presence of a persistent hydrocarbon horizon
limits the vertical mobility of biota, and
affects the transport of nutrients and
oxygen through normally permeable beach
sands.
The importance of beach variability in
engineering studies is well-known. Seasonal
and aseasonal beach changes can affect the
1 ifetime of coastal structures, and the
design of beach protection devices. Proper
set-back requirements for near-shoreline
development is dependent on long-term trends
in coastal change as well as natural seasonal
fluctuations in beach level. Finally,
quantification of beach variability and its
statistical relationship to driving forces
can serve as useful input to nearshore
sediment transport models, particularly as a
test of variation in beach response as a
function of different sediment types (grain
size, sorting). Empirical guidance for
modellers can also be provided through well-
constructed statistical studies of driving
force/beach response, when constructed using
insight gleaned from dynamical considera-
t ion s (e. g., Au brey eta 1., 198U ) •
The basic problem addressed here is the
quantification of seasonal and aseasonal
patterns of beach change along coasts with

different wave climates, and for beaches
with different sediment characteristics.
Rigorous statistical techniques for
quantifying these changes must be developed
to allow for meaningful comparison of beach
response at different sites, providing a
statistical basis for defining differences
in beach behaviour. The ultimate goal is to
develop a capability for predicting beach
changes on many spatial and time scales, but
this goal is to be achieved only with
careful statistical methods combined with
dynamical (both analytical and numerical)
modell i ng.
Observation of changes in beach planform
have been made for the past century, and
relations between these beach changes and
the driving forces postulated. For instance,
Davies (1964) related beach characteristics
to global patterns of waves (swell coasts,
storm coasts, and protected coasts), using
not direct measurement but compilations of
winds and wave behaviour observed from ships
and shore. Davies (1964) pointed out that
the major drawback in obtaining statistical
relationships between beach behaviour and
driving forces is lack of knowledge of the
driving forces, specifically wave activity.
This is still true at the present, although
progress has been made in the last couple of
decades in measuring nearshore wave
characteristics (e.g., Pawka et al., 1976;
Seymour and Sessions, 1976; Thompson, 1977;
Seymour, 1979).
Beach profile monitoring programmes
generally have had the following character-
istics: limited duration of sampling;
inadequate sample frequency; inadequate
spatial coverage, particularly for beaches
with much longshore variability; inadequate
spatial density of sampling; and poor
documentation of the driving forces.
65
Analysis procedures for most of these
studies have varied considerably, with
little uniformity in treatment of the data.
Consequently we are left with many
observations of beach change, of highly
variable quality, and no capability for
readily comparing changes at one location
with changes at another location. The
resulting lack of comparison leaves us with
a disturbing inability to synthesize these
data into a meaningful set of observations,
which might provide valuable insight into
causes and patterns of beach variability.
Work reported in this paper represents
an attempt to take data from different
coasts of the United States, exposed to
widely different wave climates, with
different sediment types, and synthesize it
in a rigorous fashion to allow quantitative
intercomparison of magnitude of seasonal and
aseasonal beach changes at these different
locations. The work represents a plea for
some uniformity in analyzing beach data to
provide results useful to a variety of
disciplines studying this active nearshore
environment.
STUDY SITES
Seven locations were selected for this
study (Fig. 2), six along the U.S. east
coast (Fig. 3) and one on the U.S. west
coast (Fig. 4). The beaches span a spectrum
of grain sizes, and range from open ocean
beaches, to those partly sheltered by
offshore shoals and islands, to completely
sheltered beaches. A brief description of
each study site follows.
Torrey Pines, California: This southern
California site (Fig. 4) is a long sandy
beach, extending for more than 40 km with no
man-made structures to impede longshore sand
transport. The beach profile locations are
backed by 100 m high sea cliffs, composed of
66

.:.. '1/
LOCATIONS OF STUDY AREAS

:, NEVADA
,
",
",
"
CALIFORNIA "

t-L-O-C-A-Tl-O-N -O-F-T-O-R -E.li.Y~P-ICl.NE-S- I

." BEACH STUDY SITE

t
N

2) Location map for seven beach study sites

distributed along the U.S. east coast
o
(6) and the U.S. west coast (1). 100 wi
500 1000
I
METERS

o 2000 4000
I wi w i ! I
FEET
OEPTH IN FEET

4) Location map for Torrey Pines,

California, with profile line locations.

2.0 m on the average, and 2.5 during spring

80° 78°
39°'---...1..---'---'-----'-----'-----'
3) Location map for six U.S. east coast
study sites, showing beaches in relation
to water bodies over which surface
gravity waves are generated.
well-indurated deposits which add sediment
to the nearshore zone upon collapse. Mean
beach width varies from 40 m to 100 m,
composed of sand with a median grain size of
about 0.20 mm. Tide range is approximately
tides. The site was described in detail by
Nordstrom and Inman (1975), Aubrey et al.
(1976), and Aubrey et al. (1980).
The beach is exposed to a wave field
77°
which is partly sheltered by the islands
offshore on the continental borderland
(Pawka et al., 1976). The nearshore wave
field is dominated by long-period swell from
all offshore quadrants.
Holden Beach, North Carolina: Holden
Beach is a 13 km-long barrier island located
west of Point Fear along an east-west
stretch of coastline (Fig. 3). It is
bordered on the west by Shallotte Inlet, and
on the east by Lockwoods Folly Inlet.
Average beach width is about 250 metres,

with the beach generally narrower near the
ends of the barrier, broader near the
centre. Beach material is a moderately
well-sorted medium sand. Frying Pan Shoals,
the southerly extension of Cape Fear, partly
shelters Holden Beach from waves propagating
from the east and southeast. These waves
often break on Frying Pan Shoals, losing
much of their energy, reducing the impact of
northeasters which are so damaging to the
remainder of the barrier beaches along the
North Carolina shoreline. This site was
described in detail by Miller (1982).
Long Beach Island, NJ: This site
(Fig. 3) is a barrier island along the south
shore of New Jersey, separating the Atlantic
Ocean to the east from salt marshes to the
west. The study area is bounded to the
north by Barnegat Inlet, and to the south by
Beach Haven Inlet, both active inlets. The
island, with its east-southeast exposure,
has a median sand diameter of 0.35 mm
(Ramsey and Galvin, 1977), and stretches
about 32 kilometres. Tides are semidiurnal
with a spring/neap range of 1.5/0.9 metres.
The beach itself is heavily structured, with
110 groins, 83 of which have been built or
rebuilt over the period 1962-1975. The
island is narrow (mean width of about
400-500 m), with a nearly continuous dune of
height 5 to 8 m, MLW. Complex offshore
topography (ridges and swales) imparts
considerable alongshore variability to the
wave field. Peahala Ridge is one ridge
which is presently shore-attached. Except
for the effect of the ridges and swales,
Long Beach Island has an open exposure to
Atlantic Ocean waves. Miller et al. (1980)
describe the study site in more detail.
67
Jones Beach, Long Island, NY: Jones
Beach is a 24-km long barrier beach
separating the Atlantic from Great South Bay
(Fig. 3). It is bounded on the east by Fire
Island Inlet, and on the west by Jones
Inlet. Mean beach width is approximately
150 m, with considerable variability (225 m
near Jones Inlet jetty, to 35 metres on the
eastern third of the study area). Beach
sand is medium to fine grained. Tides are
semidiurnal, with a mean range of 1.3
metres, and spring range of 1.5 metres.
Offshore bathymetry is complicated by ridges
and swales, which (unlike Long Beach Island,
NJ) have no subaerial expression. No struc-
tures interrupt the longshore transport of
sand, with the exception of the jetties
protecting the entrances to Jones and Fire
Island Inlets. The beach with its southerly
exposure is not sheltered from Atlantic wave
conditions. A more detailed description of
the study site is available from Morton et
al. (1982c).
Fairfield/Milford Beaches, CT: Both of
these study sites are located along the
northern shore of Long Island Sound, and are
exposed to the locally-generated waves of
the Sound (Fig. 3). Beach behaviour along
Milford Beach is dominated by the impact of
the Charles Island Bar, a submerged tombolo
or bar. To the east, a sandy beach (Silver
Beach) extends for about half a kilometre,
with no structures in the surf zone, but
backed by a seawall. Beach material ranges
from sand through boulders. To the west of
Charles Island Bar, a series of small beaches
is disrupted by a number of shoreline
structures, segmenting the beach. The beach
is backed by a seawall here, also. Beach
material ranges from medium sands to
boulders. Fairfield Beach is located about
16 km west of Mi 1ford beaches" and stretches
68
1.8 km from Shoal Point to the entrance to
Ash Creek. Along this length, there are no
shoreline structures to interrupt longshore
sand transport. Beach sand here is medium
to coarse in texture.
Tidal range at both these locations, for
spring/mean conditions, is 2.4 m/2.1 m. The
site is described in greater detail in
Morton et al. (1982b).
Misquamicut Beach, RI: Misquamicut
Beach is a low-lying barrier beach located
near the western limit of Rhode Island (Fig.
3). The beach, extending approximately 8.5
km from Watch Hill Point to Weekapaug Point
on the east, is approximately 125 m in
width, varying from about 100 m to 150 m.
Beach material is composed primarily of
fine-to-medium sands, interspersed with
occasional areas of coarser sands and
gravel. Offshore bathymetry is relatively
complicated. Far offshore, a submarine
ridge focuses waves propagating shoreward
from the Atlantic. Nearer shore, a number
of scales of bottom roughness are displayed
which influence, and are in turn influenced
by, nearshore waves. Tides in the area are
semi-diurnal, with a spring/mean range
variation of 0.96/0.78 m. Incident waves
progagate shorewards from the southerly
quadrant from the continental shelf, and are
also locally generated within a restricted
fetch to the western and eastern quadrants.
The eastern part of the study area is
bounded by a tidal inlet (structured).
Other than this inlet, no significant
shoreline structures inhibit the longshore
exchange of sediment. This study site is
described in more detail in Morton et al.
(1982a).
Cape Cod, MA: The Cape Cod study area
(Fig. 3) represents an eroding glacial
feature, backed in the north by sea cliffs
up to 30 m in height, and to the south by a
marsh (the two southern-most lines are on a
barrier beach). Beach material is composed
of unconsolidated drift material derived
from eroding sea cliffs. Grain size ranges
from about 1.0 mm in the north, down to 0.25
mm in the south. No structures exist along
the shore, so longshore exchange of material
takes place unimpeded. Although the range
varies a little along the study area,
spring/mean ranges for this semi-diurnal
tide are 2.5 m/2.0 m. The northern part of
the study area commonly exhibits a longshore
periodic shoreline feature, called a hooked
bar by Aubrey (1980), with length scales of
hundreds of metres. Migration of these
shore-attached features can affect beach
behaviour. The study area is completely
exposed to open ocean waves progagating in
from the Atlantic. This study area is
described in more detail by Miller and
Aubrey (1982).
Summary: These seven study sites
represent a range of beach conditions and
driving forces which can be expected to
provide some insight into different modes
and magnitude of beach change. In partic-
ular, these seven beaches have different
grain sizes; they are exposed to varying
wave climates, ranging from locally
generated seas to distantly-generated swell
conditions; they have variable degrees of
structural development in the surf zone,
ranging from highly developed along Long
Beach Island, NJ, to completely undeveloped
as in Torrey Pines, CA, and Cape Cod, MA;
and they are exposed to different tidal
regimes, with different tidal ranges
(although all have semi-diurnal tides).
This range of conditions makes a semi-
quantitative comparison of beach changes a
useful exercise, since this has not been
done previously.
 69

DATA SETS
Profiles of the beach face from the back- 12.0/2.5

shore out to approximately mean water level HOLDEN BEACH, _ _ 22

~I'·4/1.6
NC

were made on each of these beaches for

LO~~ BEACH ISLAND• • • • • • 32 ~M 10.88/1.15
variable lengths of time (Fig. 5), and for a
JONES BEACH
LONG ISLAND, NY
1 • • • • '8 f0'i;/#M 11.311.5

SAMPLING DURATION
120123

_13
FAIRFIELD-MILFORD. 7
CONN. ~$WM
Wffi
TORREY PINES,
CA. MI~~~AMICUT, • 7 1078/096
HOLDEN BEACH,
NC CAPE COD> 1 2.012.5
MASS
LONG BEACH ISLAND , ! , I , !
NJ
10 20 30
NUMBER OF TIDAL RANGE (m)
JONES BEACH I
LONG ISLAND, NY PROFILE STATIONS

FAIRFIELD - MILFORD
CONN

MISQUAMICUT,
R.I 6) The number of survey lines (left) and
CAPE COD,
MASS. I
approximate tidal range (right) for each
I I I
1960 1965 1970
I
1975 1980 of the seven survey locations making up
this study. Lines were surveyed over
5) Duration of beach profile sampling at the period shown in Fig. 11.
each of the seven study sites. Sampling
was on approximately a monthly basis for NUMBER OF PROFILES ANALYZED

most of the beaches. TORRE~A PINES, 9001

HOLDEN BEACH,
NC

variable number of survey lines (Fig. 6). LONG BE~C~ ISLANDIL-_ _ _ _ _ _ _ _ _ _ _ _ _ _.......:2:.;,;'5~e I
Consequently, the total number of beach
profiles analyzed for each site is variable
(Fig. 7), with a maximum of 2158 profiles
analyzed from 32 lines on Long Beach Island, MISQUAMICUT.

NJ, to a minimum of 378 profiles analyzed

"
CAPE COD,
MASS
from 7 lines at Fairfield/Milford, CT. ! ,

200 600 10DO '400 1800 2200

Duration of sampling ranged from about
thirteen years at Long Beach Island (NJ),
Jones Beach (Long Island, NY), and
Misquamicut Beach (RI), to a short length of
five years at Cape Cod (MA), with other
sample lengths intermediate to these
extremes.
All profiles examined were wading
profiles, taken with engineer's level and
survey rod, extending from an onshore
benchmark along a profile line to the water
line. The different data sets were obtained
with different degrees of accuracy,
reflecting in part climate extremes, wave
7) The number of beach profiles used in the
analysis described in this study varied,
ranging from a low of 378 at the
Fairfield/Milford beaches, to a high of
2158 profiles on Long Beach Island, NJ.
activity, and vested interest of the survey
parties. A more complete description of
survey procedure can be found in Aubrey
(1979). All survey notes were carefully
70
checked for errors, both in the field and
later in the laboratory, to assure high data
standards. The data were checked as part of
this study as well, to minimize outlying
points of dubious validity. Profiles for
all sites except Torrey Pines, CA, were
obtained as part of the Beach Erosion
Programme of the Coastal Engineering
Research Center (USA Corps of Engineers).
The Torrey Pines data set was collected
initially under the aegis of CERC through
Scripps Institution of Oceanography (D.L.
Inman, principal investigator), and later
funded through the Office of Naval
Research. For all profiles, accurate
benchmarks were established to provide both
vertical and horizontal control for
repeatability of surveys.
METHODOLOGY
Analysis of beach profile data sets in
the past has taken a number of different
forms, ranging from heuristic approaches to
much more dynamical approaches. Common
measures of net profile change which have
been used are volume changes on the beach
foreshore, migration of Mean Sea Level (MSL)
intercept or other vertical datum, or beach
stage models. Each of these techniques
provides insight into the behaviour of
beaches at a particular locality; however,
comparison of beach change at different loca-
tions is difficult using these methods of
analysis. Synthesis of changes at beaches
with different characteristics exposed to
different driving forces is difficult
without some common analysis technique.
This paper utilizes a method for quantifying
beach change in a manner amenable to
synthesis at a later time. Al though the
author feels this methodology is useful and
provides insight into patterns of beach
change, others may prefer alternate methods
for intercomparison. One thrust of this
paper is a plea to consider the need for
later synthesis in any analysis scheme, so
concepts of beach change can be determined
not just for a single beach, but rather for
many beaches.
The beach profiles are analyzed here by
empirical eigenfunction analysis (also known
as principal component analysis, empirical
orthogonal function analysis, eigenanalysis,
or factor analysis, a close relative). The
empirical eigenfunction technique has been
used by other investigators to determine the
modes of variability of periodic beach
profile measurements. The method can be
useful in showing the spatial location at
which the major amount of beach variability
occurs along the profile line. Temporal
eigenfunctions also show seasonal or other
periodic trends in the data that may be less
obvious from other methods of analysis.
Properly used in conjunction with other,
more conventional methods of analysis, the
empirical eigenfunction technique provides a
useful tool for understanding beach vari-
ability. Noble and Daniel (1977) provide a
general explanation of the techniques.
Specific applications to the coastal zone
and beaches are provided by Winant, Inman,
and Nordstrom (1975); Vincent, et al.
(1976); Resio, et al. (1977); Aubrey (1978,
1979); and Bowman (1981).
The objective of eigenfunction analysis
is to separate the temporal and spatial
dependence of the data set so that it can be
represented as a linear combination of
corresponding functions of time and space:
n
h(x,t) = l: c (t)e (x) (:\ n n )~ (1)
k =1 k k kx t

where
h(x,t) = a profile sample at any point x
and time t, n = the lesser of nx and
nt (the number of points along each
profile line and the number of times the
profile was measured, respectively),
ck(t) = temporal beach eigenfunc-
tions, ek(x) = spatial beach
eigenfunctions (BEF), Ak =
eigenvalues associated with each
eigenfunction pair (ck,e k).
This representation helps identify processes
responsible for profile changes, assists in
evaluation of their relative importance, and
aids the identification of specific events.
The following properties of empirical
eigenfunction make it a desirable tool for
analysis of beach profile data (Aubrey,
1978) :
(1) Empirical eigenfunctions provide
the most efficient method of compressing
the data; i.e., the most dense
representation of a data set in the
sense that the first n terms in the
expansion represent more of the data
variability than the first n terms of
any other orthogonal expansion.
(2) Since both the spatial and temporal
eigenfunctions are orthogonal sets, each
corresponding set (Ak,ek(x), ck(t))
may be regarded as representing a mode
of variability which is uncorrelated
with any other mode.
(3) The eigenfunction representation is
convenient when using the method of
minimum mean square error estimation.
The eigenfunctions provide a useful ~
priori method for reducing the number of
variables in this estimation theory, and
also provide a means of removing the
noise (or less predictable part of the
data) from the data set.
71
Empirical eigenfunctions, then,
objectively represent the variation in the
beach profile configuration in terms of
distance from fixed data points, and in
terms of temporal changes in the profile
over the period of the study. Comparison of
the variability of eigenfunctions from a
series of profiles taken along the beach may
show differences due to the presence of
structures or change in shorel ine orienta-
tion.
Since the empirical eigenfunctions form
an orthogonal set, they are similar in some
respects to the more familiar Fourier
analysis. In Fourier analysis, a sinusoidal
variation in the data set is assumed, and
one best fits the data to a series of sines
and cosines. This method assumes beforehand
some given form for the orthogonal functions;
in empirical eigenfunction analysis, the
data themselves determine the form of
orthogonal functions which are used in the
analysi s.
Applied to systematic measurements of
beach elevation, the eigenfunction
representation is a concise means of
representing beach profile variability. The
eigenfunction modes can be used to distin-
guish between variability on different time
scales. Though a large number of eigen-
values are determined, Aubrey (1978) found
that more than 99.75 per cent of the mean
square value of his data set could be
accounted for by the three eigenfunctions
associated with the three largest eigen-
values. The second through fourth eigen-
values accounted for approximately 90 per
cent of the variability in 4-year data sets
of beach profiles in southern California.
This concise representation of beach profile
variability is desirable when trying to
compare different locations, especially for
data sets spanning long periods of time.
72
In the empirical eigenfunction
technique, eigenvalues, Ak' provide
information on weights of the eigenfunc-
tions. Each eigenvalue gives the mean
square value of the data (the variance if
the mean has been removed) accounted for by
the eigenfunctions. This provides a
convenient means for ranking eigenfunctions
and assessing the importance of each. This
also provides a convenient means of removing
noise from the data, if it is assumed that a
function accounting for only a small part of
the mean square value of the data is not an
important variable in the data. Eigenfunc-
tions whose eigenvalues are below a certain
value can be neglected in estimation
problems. These screening techniques are
discussed in detail by Preisendorfer et al.
(1981).
Since some of the data considered in
this study is non-uniformly sampled in both
space and time, consideration must be given
to the utility of this somewhat complex
analysis technique for this data. The
primary information derived from the
profiles in this study is magnitude of beach
variability, and seasonality of that
variability. This information will not be
as readily interpretable for a non-uniformly
sampled profile sequence as for a uniformly
sampled beach. However, if the frequency of
sampling is high compared to the time scales
examined, and the beach is observed for a
long period of time, then the statistics
derived from eigenfunction analysis will
approach those derived from uniform
sampling. In order to illustrate this
conclusion, a beach with five years of
relatively uniform and high-frequency
sampling (Torrey Pines, CAl was artificially
resampled at irregular intervals for the
same period, and eigenanalysis results
intercompared. The magnitude of beach
variability (normalized as described below)
was within 10% for the uniformly sampled and
non-uniformly sampled cases, even when large
seasonal sampling discrepancies were
artificially induced. Seasonal beach
signals likewise were apparent in both
cases, with a clear seasonal signature
dominating for the non-uniform sampling as
well. This numerical exercise illustrates
the attraction of using eigenanalysis for
beach profile data, even for non-uniform
sampling. Certain degrees of non-uniformity
in sampling are not going to conform to this
rule; clearly each season must be represented
in the sampling, and sample intervals must
not exceed one or two months, on the average,
if seasonal information is derived. If the
beach is undersampled, aliasing can be a
major problem.
Eigenanalysis determines vectors such
that the maximum variance of observed
variables is described (as opposed to factor
analysis, which optimizes the intercorrela-
tions of all variables, yielding a
correlation-weighted analysis instead of a
variance-weighted analysis; see for example
Joreskog et al., 1976). Since we are
interested in describing beach variability,
eigenanalysis is an obvious choice. As
defined in the equation 1, we have
represented the data by a complete set of
orthonormal functions (Ak,ck,e k).
The mean square value of the data is given
as:
n
MSV =k ~ 1 \ (2)
where the mean is taken in both space and
time, because of normalization by nx and
nt in the governing equation. These

eigenfunctions are referred to as 'mean
eigenfunctions,' since they retain
information about the mean state of the
beach. For each profile line, the
arithmetic time mean can be calculated, and
subtracted from the data prior to analysis,
yielding a new data set h'(x,t), as follows:
h' (x,t) = h(x,t)- h(x) (3)
where h represents the mean (in time) value
of the elevation at a point x. The eigen-
vectors of covariance matrices formed from
h'(x,t) are termed 'de-meaned' eigen-
vectors. In this case, the sum of the
eigenvalues represents the variance of the
data set, instead of the mean square value:
m applying this correction, we have taken care
I: A (4)
k=l k not to include backshore areas with dunes or
This variance estimate provides information
on the variability in the data per spatial
sampling point and per time sample. This
type of normalization is not always amenable
to comparison of one beach with another
beach which has been sampled differently (in
space and/or through time). Because of the
normalization, and desire to compare results
regardless of sampling strategy, a correction
has been introduced into the analysis. The
active beach correction is a factor
multiplying the variance, a 2 to adjust
the results so they reflect only variance
along the active part of the beach. If a
long segment of a beach profile comprises
inactive parts of the backshore, the variance
as defined in equation (4) would be artifici-
ally low. The empirical correction for
active beach width minimizes this problem:
.w (5 )
where W= total length of profile/total
length of active beach. The new variance
73
provides a more consistent basis for
comparison, even though it involves some
definition or assessment of the active
beach. For our work, this definition has
been based on the degree of variability of
the beach at any point along it. After
eigenfunctions have been derived for a given
beach, the first few are summed as in (1)
(to account for a large fraction of the
beach variability), and the weighting for
all points along the beach profile
compared. The segment defining the active
beach is selected as that portion of the
profile which excludes those points
accounting for less than some fraction of
the total variability along the profile. In
seacliffs, since this section of the beach
responds to different forcing than the
foreshore. This was necessary, in
particular, along the Cape Cod (MA) profiles.
RESULTS
Four aspects of beaches and incident
driving forces were examined in this study.
For each location, a wave climate was
formulated from existing data; these wave
cl imates vary from quantitative at Torrey
Pines (CA) to highly qualitative at
Fairfield/Milford (CT) beaches. The
magnitude of beach variability for each site
was quantified. The degree of seasonality
of beach variability was examined. Finally,
as an example of the utility of the method,
the effects of grain size on beach
variability were quantified for the Cape Cod
(MA) site.
WAVE CLIMATE
Documentation of the driving forces is
an essential, but difficult, aspect of
formulating models of beach variability, and
74

in verifying or testing these model s. In

CAPE COD, MA.
situ measurements of wave behaviour is the
60
best way to document wave climate at this x
time, although numerical models of wave ~ 50
>-
<.0
growth, propagation, and shoaling show much a::
w El SPECTRUM METHOD
promise for the future. Even then, local iii 40 o 5MB METHOD

measurements can provide information on the ::; " SSMO

a::
« 30
complete spectrum of the wave climate, w
>-
whereas not all numerical models can handle "-
020
spectral growth and decay. Alternative I-
z
w
techniques for establishing a wave climate ~ 10
w
include hindcasting techniques (primarily (l.

using 5MB techniques and the spectrum o

o
method), visual wave observations from WAVE PERIOD IN SECONDS

shore, and ship observations (primarily the

Summary of Synoptic Meteorological
Observations--SSMO). All techniques have
their limitations and biases which make it
difficult to intercompare results from
different beaches whose wave behaviour has
been derived from different techniques.
As an example of the lack of agreement
between these different techniques, a
comparison was made for Cape Cod (MA)
beaches using the 5MB hindcast technique,
the wave spectrum method, and compiling the
SSMO data for the Boston region (Fig. 8).
The results indicate that the three
techniques yield wave climates which are
about as different as reasonably can be
expected for the open ocean. The spectrum
method yields a modal period of about 5
seconds, the SSMO yields a modal period of
about 11 seconds, while the 5MB technique
yields a peak of about 12.5 seconds. These
results could certainly be improved by using
more up-to-date hindcasting techniques, but
the results indicate the disparity in wave
climates obtained from different approaches.
Consequently, the comparison of wave
climates here can only be qualitative with
no hope at this time for quantitative
8) A graphic comparison of a nearshore wave
climate as determined from three commonly
used sources for wave data. The spectrum
method and 5MB method are both hindcasts,
derived with the same basic meteorolog-
ical data set. The SSMU (Summary of
Synoptic Meteorological Observations)
data are shipboard observations. The
three wave climates differ markedly,
covering the expected variability in
wave climates along any open coast
location. This type of uncertainty in
nearshore wave cliraate emphasizes the
need for either careful in situ
measurements, or rigorously tested
numerical wave hindcasting programs.
intercomparisons. For the following
discussion, more detail on the wave climate
can be found in the studies of each site
referenced earlier.
Torrey Pines, CA: This location has the
best-documented wave climate due to work
performed at the Scripps Institution of
Oceanography. A linear, multi-sensor array
of pressure gauges was deployed off the
 75

beach for a period of about five years

HOLDEN BEACH, NC
(Pawka et al., 1976; Aubrey, 1979).
Four-times daily measurements of directional
wave characteristics have documented the
wave climate to a high degree. This
information was used in previous studies to
relate beach changes to driving forces
(e.g., Aubrey et al., 1980). WAVE HEIGHT ==::::J
The wave measurements show the southern WAVE PERIOD """""""""",

California region to be swell-dominated :t I CT

throughout the year. Locally-generated,

higher frequency waves are more common in 160
10
the winter than summer, as local storms pass 140
E
through. The major difference between ~120
!;:
summer and winter conditions is in the wave '"~IOO o
o
energy, rather than in the period. This in w 6 ffi
0-
turn affects the wave steepness, an ~ 80 W
~
indicator of beach erosion and accretion, f-
~60
4 ;0

according to some studies. Ii:

Z
'"
u; 40
Holden Beach, NC: Wave climate from 2
20
Holden Beach was established from direct
measurements made along a fishing pier along
Holden Beach (Thompson, 1977) and from
Littoral Environmental Observations (LEO)
consisting of visual observations (Miller, 9) Wave climate at Holden Beach, North
1982). The continuous-wire staff measure- Carolina, derived from CERC pressure
ments were made from February 1971 through sensor located off a pier along Holden
February 1975 (with some periods of inactiv- Beach (see Fig. 5). Summary is given in
ity), with 1024 second measurements taken terms of monthly averages of wave period
every 4 hours (Fig. 9). The data show a and wave height. Source for data is
seasonality in wave period (averaged over Thompson (1977).
monthly intervals), with a lower period
generally from April through August. Wave mean yearly period is about 7.5 seconds. LEO
height shows less of a seasonal periodicity. observations provide no additional informa-
As mentioned earlier, the wave field near- tion than the wave staff measurements, other
shore at Holden Beach is significantly than some indication of direction of wave
affected by Frying Pan Shoals to the east, approach.
so the nearshore wave climate is not directly Long Beach Island, NJ: Wave information
reflecting the offshore wave climate. Mean available for this study site include a
yearly significant wave height is about 0.6 m, nearby (35 km to the south) wave staff
located on a pier in Atlantic City, SSMO
data, and local visual beach observations
76

from 1968-1974. For the forty-one month

period from April 1964 through December STEEL PIER, ATLANTIC CITY, NJ

1967, mean significant wave height is 0.81 m, 1962

1963
and mean wave period is 8.2 seconds (Fig. 10). 1964

Since the Steel Pier gauge at Atlantic City 1965

1966
did not have directional capabilities, the 1967
WAVE HEIGHT ==::::J
only directional information available near- 1968 WAVE PERIOD : ....~J

shore are the beach observations, which have 1969 ±Iu t-------I
1970
coarse directional resolution. 197 I

Jones Beach, Long Island, NY: Wave JFMAMJJASOND

information for Jones Beach is compiled from

a variety of diverse sources. Hindcast data 2.0

are presented in Panuzio (1968), as are 14.0

results from occasional wave gauging from

1950-1954 off Fire Island Inlet. Results 15
13.0

from the wave gauging yield an average height 12.0

of 0.4 m for the period of measurement, which

j
E 11.0
was neither continuous nor representative of
l-
the entire year. A surf observation programme X
~
1.0 10.0 -:
UJ Ii
from 1954-1957 yielded a probability distribu- x
UJ 9.0 0
tion for breaking waves in the area (Short >
'"
~
0
0:
Beach Lifeboat Station), indicating waves I-
8.0 "'"-
~ 0.5 UJ
generally less than 1.25 metres. A CERC wave u
u: 7.0
it
~
z
gauge located at Steel Pier, Atlantic City, "
iii
New Jersey, was discussed in the previous 6.0

section; it is the closest CERC gaging o 5.0

location to Jones Beach. Results from that
gage may not be representative, because it is 4.0

located 160 km to the south of Jones Beach. 3.0

Beach Erosion Programme surf observations
were made along the beach from 1968 to 1974
along Jones Beach. Although poorly sampled
through the year (July through September show
almost no observations), they provide an
indication of surf activity. The mean wave
period was 6.5 seconds, and mean significant
wave height was 0.8 m, with some variation in
these numbers month-by-month (Fig. 11).
Waves generally approach the beach from the
southeast.
Fairfield/Milford Beaches, CT: There is
no quantitative information about the wave
field off Fairfield/Milford Beaches. Visual
JFMAMJJASOND mean
10) Wave climate at Atlantic City, New
Jersey, derived from a CERC wave staff
located off Steel Pier, in Atlantic City
(south of Long Beach Island). Summary is
given in terms of monthly averages of
wave period and wave height. Source for
data is Thompson (1977).
wave observations are not available, and
there have been no long-term pressure sensor

deployments within the western part of Long
Island Sound. Waves incident on these
beaches are all locally generated, with
energetic periods limited to about six
seconds or less. Because of a restricted
fetch, wave height is similarly limited.
This area, then, is dominated by locally
generated high-frequency wind waves, which
are significantly modified by offshore
bathymetry and shoreline irregularities (such
as Charles Island Bar).
JONES BEACH, LONG
196ar---
1969r------
1970t----
1971t-----
1973r--
1974t-----
JFMAMJJASOND
Mean heigh! 1m) 0.72 0.60 0.90 0.71 0.83
Mean period Is) 6.8 6.7 6.4 6.1 ELI
11) Wave climate for Jones Beach, Long
Island, New York. Source for data are
Beach Erosion Program Visual Wave
Observations from 1968-1974. Both yearly
mean height and period are given, as well
as monthly means. Coverage of the time
period is shown in main graph,
illustrating lack of observations during
the summers.
Misquamicut Beach, RI: Wave information
consists of visual observations taken as part
of the BEP programme, pressure sensor
measurements taken at nearby Charlestown
Inlet by Raytheon Corporation, and CERC
pressure gauge measurements made at Buzzards
Bay Tower close to the study site. The
Raytheon data, collected over a one-year
77
period, were not available for detailed
analysis; only a summary of results could be
found. The summary indicates that over the
period of study, wave height was less than
1.5 m98%of the time (in 8 m water depth).
Ninety-two per cent of the time significant
wave height was less than one metre (Raytheon
Corporation, 1975). Visual observations were
taken from January 1968 through December
1975, with an average of 22 visual
observations per month. Yearly mean breaker
height was approximately 0.5 metres, with a
ISLAND, NY
mean period of 8.6 seconds. Mean wave
Direction Height
" Iml
Period
( •• c> direction was just east of south. CERC
measurements were made from 23 January 1~64
through 18 April 1975 by pressure gauge.
Mean annual significant wave height (in 19.2
m water depth) was 0.75 m, with a mean period
of about 7 seconds (Fig. 12).
Cape Cod, MA: Wave data for this section
of shoreline consist of two hindcast
techniques, SSMO visual wave observations,
53
and in-situ gauging. The first three of
these have already been discussed in a
previous section (Fig. 8). LEU data, taken
in the format described by Balsillie (1~75),
were taken near the northern and southern
limits of the study area. These were not
used to compile mean periods and heights.
The gauging data consists of some pressure
sensor data obtained off the south end of the
study area by CERC, but the results are not
generally available, and so were not used in
this study. The final set of observations
consist of directional wave estimates made by
the author from 1980 to present, using a
two-axis electromagnetic current sensor with
a pressure gauge (Aubrey, 1980). Although
the data from this study have not been
compiled in a manner similar to other
locations, mean wave periods range from about
8 to 14 seconds, with wave heights generally
78

the data. This value was then normalized to

BUZZARDS BAY TOWER, MASSACHUSETTS
account for the active beach, and to obtain

=j
an annual beach variability. Results for

:~~bt-----
WAVE HEIGHT c=J
WAVE PERIOD c:::::::::J each profile line at each of the seven sites
t I rr
were then averaged to obtain a mean variabi-
~J~F~M~A~M~J~J~A~S~O~N~D
lity for each site (Fig. 13). The three open
2.0
ANNUAL BEACH VARIABILITY
13,0 DE-MEANED EIGENFUNCTIONS

TORREY PINES,
12.0 CA

1.5
HOLDEN BEACH.
11.0 NC

~EJACH 1
~

LONG ISLAND'LI_ _ _ _ _ _ _ _ _ _ _ _2_.5-1

6
10.0 ~
l-
X

g ~g~~sls~~~~~NylL _ _ _ _ _ _ _ _ _ _ _ _2.--1
46 1
(!)

~ 1.0 9.0
0:
'"
~ 8.0
'"
IL
FAIRFIELD-MILFORD,
CONN.
Do .
193

I-
Z
MISQ~~MICUT, 0.7151
""Li:
7.0
0.5
Z
(!)
6.0 CAPE COD,
in MASS.

5.0 o 0'.5 1.0 2.0 3.0

VARIANCE I YEAR 1m2)
0.0 u.;LLLLl..Lll..Ll..u..u...L.l..::'W....J::.l..,u-~-'-"-'-4.0
JFMAMJJASOND

13) Beach variability at each of the seven

12) Wave climate at Buzzards Bay Tower, study sites, where the variability is
Massachusetts, derived from a CERC normalized as in equation 5.
pressure sensor located in 63 m water
depth in the middle of Buzzards Bay. ocean beaches showed the greatest variability
Summary is given in terms of monthly (Cape Cod, MA, Long Beach Island, NJ, and
averages of wave period and wave height. Jones Beach, NY), with an average annual
Source for data is Thompson (1977). variability of 2.76 m2• The two most
closely located beaches, Long Beach Island
of the order of 1 m. The statistics for and Jones Beach, which are separated by only
waves at this location are currently being 100 km, have variances of 2.56 and 2.46 m2 ,
generated. This study area has the most respectively, within4% of each other, as one
energetic wave climate of the seven sites might expect, given similar grain sizes and
examined. similar wave climates. Torrey Pines (CA) and
BEACH VARIABILITY Misquamicut Beach (RI) have lower variances
Using the profiles from each of the study by a factor of four, and represent partially
areas, eigenfunctions were calculated after sheltered coastal reaches. Smallest variance
removing the mean profile from each profile is at Fairfield/Milford Beaches, which are
(yielding de-meaned eigenfunctions). The sum completely sheltered from open ocean wave
of the eigenvalues for each profile line was conditions, and exposed only to locally
determined. to represent the variability in generated seas.

Analysis of beach variance can provide
insight into beach processes on a more local
scale as well. Two examples are given here.
Holden Beach (NC) is a continuous barrier
island, bounded to the west by Shallotte
Inlet, and to the east by Lockwoods Folly
Inlet. Along this beach are distributed 21
profile lines, with line one to the east,
and line 21 to the west. Beach variability
shows a strong longshore trend (Fig. 14),
HOLDEN BEACH, NC
30
W
o
2.0
z
«
a:« 1.5
>
I
~ 1.0
w variability. Tidal range shows a very
ro
0.5
21 20 19 IS 17 16 15 14
STATION
14) Beach variability as defined by equation
5, along the barrier island of Holden
Beach. Variability is much higher near
the active inlets of both ends of the
barrier, reflecting inlet processes
(including sand bypassing, dredging/
spoil disposal, and wave/current
interactions).
with the greatest variance along profile
lines 1-3 and 18-21, which are all within
2.5 km of the two bounding inlets. This
greater beach variability is not related to
grain size, or to longshore wave variability
(according to wave refraction analysis
. discussed in Miller, 1982), except that
scattering behaviour associated with the ebb
tide deltas of the inlets. The variability
is probably due to inlet processes, including
79
ebb delta influence on wave refraction, sand
bypassing across inlets (a periodic event
for some types of bypassing), dredging and
dredge spoil disposal. Beach variability
along the remainder of the barrier island is
fairly constant, reflecting roughly uniform
distribution of driving forces alongshore.
The second example is from Cape Cod
(MA), which has an alongshore gradient in
median grain size, decreasing in size from
north to south (Fig. 15). Beach variability
also has a north-south gradient, with
greater variance in the north than in the
south, mirroring the grain size trend. Wave
refractionperformed along Cape Cod show no
systematic longshore variation in energy or
energy flux (Isaji et al., 1976), suggesting
that a variable driving force is not
responsible for this trend in beach
slight longshore gradient, but incomplete
data does not allow us to rigorously assess
13 12 II 10 9 8 7 6 5 4 :3 2 I its control over beach variability. The
NUMBER
influence of grain size on beach variability
has been suggested before; this example
shows the importance of including this
parameter in modelling studies.
SEASONAL BEACH VARIABILITY
A certain portion of the annual beach
variability can be attributed to a seasonal
cycle in beach change. This seasonality has
long been recognized for U.S. west coast
beaches (e.g., Shepard, 1950), but has been
a matter of debate along the U.S. east
coast, although investigators have shown
some seasonality to exist (e.g., Everts and
Czerniak, 1977; Goldsmith, Farrell, and
Goldsmith, 1974; Dewall, 1977 and 1979; and
Everts et al., 1980). Eigenfunctions will
show seasonal trends if they are energetic
enough, so eigenanalysis has also used to
document seasonality of beach response
(Aubrey, 1979). Computer simulations of the
80

beaches where there is an almost total

CAPE COD, MA
absence of seasonal signature. Weak
VAR' J seasonality can be due to a number of
QL
factors, some of which are related to the
_1_1
physical regime, some of which are due to
inadequate sampling. Structures, such as
02 groins and jetties, can affect the response
of a beach to seasonality in wave climate.
~ This is likely a contributing factor in the
Long Beach Island (NJ) case. Offshore
04
TEMPORAL EIGENFUNCTIONS
Il: SOUTH RANGE
w
[Il05
~-
::J
z
z12-
0
i=
;:!06
en
QL 0 40
w 3"
;j .30

09 '"~ 20

-20
~
1 I I 1 -30,,, I . ," I I , I, I \ I, I, I, I I d I, I, I I" \ I I I ,!, d!,!""" .I
2 3 4 5 1.5 1.0 0.5 JAODFAJAODFAJAODFAJAODFAJAODFAJAOD
1972 1973 1974 1975 1976 1977
(m 2 IYRJ eft -UNITS

16) Temporal beach eigenfunctions for Torrey

15) Beach variability along Cape Cod beaches
as a function of grain size. Independent
studies show the wave climate is
consistent along this length of beach,
so grain size appears to be a dominant
factor in explaining the increasing
beach variability to the north.
sensitivity of eigenanalysis to noise level
is described briefly in Aubrey (1978).
Strong seasonal signals are found in
temporal eigenfunctions from Torrey Pines
(CA) and Cape Cod (MA) (Figs. 16 and 17).
Seasonality was found over some profile
lines along the remainder of the beaches,
with the exception of Fairfield/Milford
Pines Beach, California. Second beach
eigenfunction shows distinct seasonal
trend, and accounts for nearly 80% of the
variability in the data set.
bathymetry may also limit seasonal response;
this is a contributing factor to Holden
Beach, where Frying Pan Shoals severely
modifies the incident wave climate.
Seasonality is also absent where the wave
climate is not seasonal or only weakly so;
this is the case in restricted fetch regions
such as Fairfield/Milford Beaches.
Poor sampling can also affect the
seasonal beach signature. The six U.S. east
coast beaches have a peculiar characteristic
 81

of poor sampling in the summer months, the highest beach variability and the most
parti cul arly July. Thi s type of under- energetic wave climate. Cape Cod has both
sampling in the summer can lead to poor the high incident wave energy and highest
definition of seasonal cycles. beach variability. Long Beach Island and
Jones Beach have the next most energetic
SPA TIAL EIGENFUNCTIONS

PROFILE LINE ~ 4 DATA NOT DEMEANED

wave climate and beach variability, with
0.6~------------------'r7.1
thei r beach vari abil iti es withi n 4% of each
other. This similarity in beach response is
0.4 4.7
encouraging, since the two sites are exposed
0.2 2.4 to nearly the same wave climate. Torrey
0.0
Pines Beach is the next most exposed
location, with the next highest wave
-0.2 -2.4
climate, followed closely by Misquamicut
--I
-0.4 - ___ 2
-----3
-4.7
Beach. These two have similar beach
-0.6 -7.1 variabilities, but are a factor of four less
I
120 130 140 150 160 170
DI5T ANCE
180
(m)
190 200 210 220
than the variability at the more open coast
beaches. Holden Beach is sheltered in large
TEMPORAL EIGENFUNCTIONS part by Frying Pan Shoals to the east, so
PROFILE LINE i 4 OAT A NOT DEMEANED its wave climate is much less than that
along the open ocean beaches close by. Its

w
Q
Q

'"
C'l
0
0.0 ]
~ variability is consequently much lower than
that at more energetic beaches. The most
" -0.5

::J
Co
H
.J
Q. sheltered of all beaches, with the lowest
~
~
[3 ~ wave energy, is the Fairfield/Milford area,
N
H -0,5
exposed only to locally generated wind

"J
;j
L

'"
0
Z Co
~ waves. Its beach variability is an order of
'" o0
magnitude lower than that at open coast
-03 ~
1970 197!
I
1972
I
1973 1974 1975
I
1976
beaches, and a factor of four lower than the
YEAR partly sheltered Torrey Pines and Misquamicut
Beach areas.
17) Seasonal beach changes along Cape Cod, The relationship between beach
Massachusetts, beaches. The second variability and energy has been shown before
temporal eigenfunction displays the by Aubrey et al. (1980) for a single beach.
seasonal response to a seasonal wave It has also been discussed elsewhere for
climate. It accounts for approximately specific beaches, but not quantitatively
75% of beach variability. compared at different sites. The unfortu-
nate fact remains that the wave climate at
DISCUSSION most coastal sites is so poorly known that
The results point out a close relation- even an empirical relationship between wave
ship between the rigorously defined beach climate (suitably represented by wave
variability and the poorly defined wave eigenfunctions, for instance) and beach
climate. Exposed open ocean beaches have change cannot be made at this time. The
82
wave hindcast models now in existence may
give us better data sets in the near future
so we can improve on the qualitative
statements made in this paper (specifically
the WES model should be available in the
time span of a year or so, providing us with
data coincident with the profiling efforts).
Seasonal beach changes, which respond to
seasonal patterns in the driving forces,
have been documented before in many places.
This study shows most beaches have a seasonal
signature unless prevented by one of several
factors. The fetch and/or exposure of a
beach site may be such that the seasonality
in weather patterns may not be reflected in
the wave climate. Examples are Fairfield/
Milford Beaches where the restricted fetch
limits the size of waves and hence the
seasonal differences in wave character-
istics. An example of exposure limiting
wave seasonality is Holden Beach (NC), where
Frying Pan Shoals limits the size of waves
reaching the barrier island. Large waves
will break one or more times on the shoal,
limiting the energy reaching the shore
during large northeast storms which inflict
much of the beach erosion on the more
exposed shoreline bounding the study site to
the north.
Eigenfunction analysis has graphically
shown two beach relationships which might
not otherwise be apparent during routine
analysis of profile data. At Holden Beach
(NC), beach variability is greater near the
inlets bounding the barrier island, than
near the middle of the island. Magnitude of
beach variability near the centre of island
probably represents the part of the beach
variability driven by the incident wave
field, while the outer parts of the barrier
are more affected by inlet behaviour.
Examples of inlet influence include modific-
ation of the nearshore wave field because of
wave refraction around the ebb tide delta
(due both to bottom topography and wave-
current interactions--steepening and
breaking), longshore sand bypassing episodes
along the ebb tide delta imparting large
signatures to beach change, and dredging/
spoil disposal near the inlet channels.
This type of longshore dependence of beach
variance may be reflected in the biological
communities inhabiting these different
areas, although these effects may be
difficult to see in light of expected
differences in response to different
physical and chemical conditions due to
i nl et proximity.
Another observation clarified by
eigenanalysis is the coincidence of beach
variability on grain size, the example here
being Cape Cod beaches. Grain size decrease
from north to south is mirrored by decreasing
beach variability from north to south,
despite no apparent longshore gradients in
energy flux incident on the beach. Grain
size responds to source proximity and
longshore sorting of material; this
difference in grain size is reflected in
markedly different beach slopes alongshore.
The reason for the higher beach variability
in coarser grained beaches is not apparent
at this time. Possibilities include the
influence of greater pore space in coarser
material, increasing permeability, and
transmitting greater fluid pressure to each
grain of sand. This would allow the sand to
respond much more quickly to wave activity
than a less permeable sand. This effect has
not been quantified.

CONCLUSIONS
Eigenanalysis has quantified
relationships between beach change and
driving forces along seven beaches with
markedly different wave climates, spanning a
variety of grain sizes degree of and
structural shoreline modification.
Neglecting long-term beach trends, open
coast beaches have the greatest variability
on an annual basis, while partially
sheltered coasts are lower in variability by
about a factor of four. Beaches nearly
completely sheltered from open ocean wave
conditions (restricted to short, local
fetches) have the least variability, down by
an order of magnitude from open ocean
beaches. A gradation from open coast to
completely sheltered beaches exist, only a
sample of which were analyzed in this
study. Sheltering can result from offshore
islands (Torrey Pines, CAl, convoluted
shorelines (Misquamicut Beach, RI), or
unusual bathymetry (Holden Beach, NC) such
as shoals. Although the relationship
between wave exposure and beach change is
qualitative, improvement in predictive
capability can be expected once improved
wave hindcasting procedures provide us with
realistic nearshore wave climates for the
periods coincident with the beach studies.
Although the profile data examined in this
study was of variable quality (in terms of
both spatial and temporal uniformity), the
major inadequacy of the data set was in the
knowledge of wave climate, which varied from
well-known (Torrey Pines, CAl to poorly
known (Fairfield/Milford Beaches, CT).
Beach variability found in this study is
not easily expressed in terms of a single
morpho 1ogi ca 1 model (such as Short and
Wright, 1983), because this study addresses
83
only total variability, not cross-shore
spatial structure. These profile shape
factors are discussed in a paper presently
in preparation.
Patterns of beach variability along a
single beach provide insight into some
relationships which need to be explained in
more dynamical terms. Along a barrier island
bounded by two tidal inlets (Holden Beach,
NC), annual beach variability was greater
within 2.5 km of the inlets than in the
middle of the barrier. This pattern
reflects the influence of the inlet on beach
processes, particularly through modification
of the incident wave field, sand bypassing
episodes, and dredging/spoil disposal
operations. Beach variability along the
middle of the barrier island was nearly
constant, suggesting that this segment was
undisturbed by inlet behaviour.
Along a beach with a sharp longshore
gradient in mean grain size (Cape Cod, MA),
the magnitude of beach variability increased
with increasing grain size, even though the
wave climate showed no correlative pattern.
This data set suggests the possibility of
testing more dynamical models of beach
change as a function of grain size.
Eigenanalysis has proved to be a useful
tool in synthesizing beach profile data from
a number of different locations, exposed to
different forcing conditions, and sampled
with highly variable uniformity. The
technique allows some quantitative
comparison between beach behaviour at
different locations, which has not been
commonly done in the past. Whether or not
this particular analysis is adopted as a
routine procedure for examining beach
profile data, scientists and engineers must
consider how best to intercompare results
from one beach with results from another
84
beach, rather than concentrate on a single
data set. Insight into dynamics of beach
change, and guidance to much needed
modelling of the process, will occur only
when we can synthesize existing data, and
analyze future data in a manner consistent
with the need for intercomparison.
ACKNOWLEDGEMENTS
Much of the work performed for this
study was funded through the following
agencies: U.S. ArmY Coastal Engineering
Research Center in contract to the Scripps
Institution of Oceanography (D.L. Inman,
Principal Investigator), and to Science
Applications, Inc. at Raleigh, NC (Martin C.
Miller, Principal Investigator); and NOAA
Office of Sea Grant number NA80-AA-D-00077
to the Woods Hole Oceanographic Institution
for the Nearshore Sediment Transport Study.
Martin C. Miller, J. Karpen, R. Morton, and
F. Bohlen contributed to the study of the
east coast beaches. R. Gorski drafted many
of the figures. Woods Hole Oceanographic
Institution Contribution number 5324.
REFERENCES
Aubrey DG (1978) Statistical and
dynamical prediction of sand beaches. Ph.D.
thesis, Scripps Inst. of Oceanography, U.C.
San Diego, 194 pp.
Aubrey DG (1979) Seasonal patterns of
onshore/offshore sediment movement, JGR, v.
84, p. 6347-6354.
Aubrey DG (1980) Our dynamic coast-
lines. Oceanus, v. 23, no. 4, p. 4-13.
Aubrey DG (1981) Field evaluation of Sea
Data directional wave gage, (model 635-9).
Woods Hole Oceanographic Institution
Technical Report, WHOI-81-28,
52 pp.
Aubrey DG, Inman DL and Nordstrom CE
(1976) Beach profiles at Torrey Pines,
California. Proceedings of the 15th Int.
Conf. on Coastal Engineering, Amer. Soc.
Civil Engr., p. 1297-1311.
Aubrey'DG, Inman DL and Winant CD (1980)
The statistical prediction of beach changes
in southern California. JGR, v. 85, p.
3264-3276.
Balsillie JH (1975) Surf observations
and longshore current prediction. U.S. ArmY
C.E.R.C., Ft. Belvoir, VA, Tech. Memo. 58.
Bowman D (1981) Efficiency of
eigenfunction for discriminant analysis of
subaerial nontidal beach profiles. Marine
Geology, v. 39, p. 243-258.
Davies JLD (1964) A morphogenic approach
to world shorelines. Zeitschrift fur
geomorphologie, p. 127-142.
Dewall AE, Pritchett PC and Galvin CJ, Jr
(1977) Beach changes caused by the Atlantic
coast storm of 17 December 1970. U.S. ArmY
C.E.R.C. Technical Report 77-1, Ft. Belvoir,
VA.
Dewall AE (1979) Beach changes at
Westhampton Beach, New York, 1962-1973.
U.S. ArmY C.E.R.C. Misc. Report 79-5, Ft.
Belvoir, VA.
Everts CH and Czerniak MT (1977)
Spatial and temporal changes in New Jersey
beaches. Proceedings of Coastal Sediments
'77, ASCE Conf., p. 444-459.
Everts CH, Dewall AE and Czerniak MT
(1980) Beach and inlet changes at Ludlum
Beach, New Jersey. U.S. ArmY C.E.R.C. Misc.
Report 80-3, Ft. Belvoir, VA.
Goldsmith V, Farrell SC and Goldsmith YE
(1974) Shoreface morphology study, the
south end of Long Beach Island, Little Beach
Island, and the north end of Brigantine
Island. Dames and Moore, Inc., Cranford, NJ.
Helle JR (1958) Surf statistics for the
coasts of the United States. Beach Erosion
Board Tech. Memo No. 108, U.S. ArmY Corps of
Engr's., 22 pp. plus appendices.
Isaji T, Cornillon P and Spaulding M
(1976) Nearshore wave climate for the outer
Cape Cod shore, Part I: Wave Refraction.
Department of Ocean Engineering, U.R.I.,
Kingston, RI.
Joreskog KC, Klovan JE and Reyment RA
(1976) Geological Factor Analysis, Elsevier
Scientiflc Publishing Company, Amsterdam,
178 pp.
Miller MC, Aubrey DG and Karpen J (1980)
Beach changes at Long Beach Island, New
Jersey, 1962-1973. U.S. ArmY C.E.R.C. Misc.
Report No. 80-9, Ft. Belvoir, VA, 289 pp.
Miller MC (1982) Beach changes at Holden
Beach, North Carolina, 1970-1974. Submitted
 85

to the U.S. Army C.E.R.C., Ft. Belovir, VA.

Seymour RJ and Sessions MH (1976)
Miller MC and Aubrey DG (1982) Beach Regional network for coastal engineering
changes on eastern Cape Cod, Massachusetts, data, Proc. 15th Coastal Eng. Conf., Amer.
from Newcomb Hollow to Nauset Inlet, Soc. Civil Eng., p. 60-71.
1970-1974. Submitted to U.S. Army C.E.R.C.,
Ft. Belvoir, VA, 1 volume. Seymour RJ (1979) Measuring the
nearshore wave climate, California
Morton RW, Bohlen WF and Aubrey DG experience. Ocean Wave Climate, Marshall D.
(1982a) Beach changes at Misquamicut Beach, Earle and Alexander Malahoff, eds., Marine
Rhode Island, 1962-1973. Submitted to U.S. Science, v. 8, Plenum Press, NY, p. 317-327.
Army C.E.R.C., Ft. Belvoir, VA, 1 volume.
Shepard FP (1950) Beach cycles in
Morton RW,Bohlen WF and Aubrey DG southern California. U.S. Army Beach
(1982b) Beach changes at Milford and Erosion Board Tech. Memo. 20, 26 pp.
Fairfield Beaches, Connecticut, 1962-1971.
Submitted to U.S. Army C.E.R.C., Ft. Short AD and Wright LD (1983) Beach
Belvoir, VA, 1 volume. systems in high, moderate and low wave
environments. This volume.
Morton RW Bohlen WF and Aubrey DG
(1982c) Beach changes at Jones Beach, Long Steele JH, Munro ALS and Giese GS (1970)
Island, NY, 1962-1974. Submitted to U.S. Environmental factors controlling the
Army C.E.R.C., Ft. Belvoir, VA, 2 volumes. epipsammic flora on beach and sublittoral
sands. J. Marine Biol. Assoc., v. 50, p.
Nordstrom CE and Inman DL (1975) Sand 907-918.
level changes on Torrey Pines Beach,
California, U.S. Army C.E.R.C. Misc. Paper Thompson EF (1977) Wave climate at
No. 11-75, 166 pp. selected locations along U.S. coasts. U.S.
Army Coastal Engineering Research Center
Panuzio FL (1968) The Atlantic coast of Technical Report no. 77-1, Fort Belvoir, VA,
Long Island. 11th Conf. of Coastal 364 pp.
Engineering, p. 1222-1241.
Parr T, Diener D and Lacy S (1978)
Effects of beach replenishment on the
nearshore sand fauna at Imperial Beach,
California, U.S. Army C.E.R.C. Misc. Report
No. 78-4, 125 pp.
Pawka SS, Inman DL, Lowe RL and Holmes LC
(1976) Wave cl imate at Torrey Pi nes Beach,
California, U.S. Army C.E.R.C. Tech Paper
No. 76-5, 372 pp.
Preisendorfer RW, Zwiers FW and Barnett
TP (1981) Foundations of principal
component selection rules. Scripps
Institution of Oceanography, SIO Ret. Series
81-4, La Jolla, CA, 192 pp. .
Ramsey MD and Galvin CJ (1977) Site
analysis of sand samples from southern New
Jersey beaches. U.S. Army C.E.R.C. Misc.
Report 77-3, Ft. Belvoir, VA, 54 pp.
Raytheon Corporation (1975) Charlestown
hydrographic study, April 1974-1975.
Raytheon Corporation, Oceanographic and
Environmental Sciences, Portsmouth, Rhode
1s1 and.
Resio D, Hayden B, Dolan R and Vincent L
(1974) Systematic variations in offshore
bathymetry, Univ. of VA Technical Report No.
9, 28 pp.

PROVENANCE OF BEACH SEDIMENTS IN SOUTH-EASTERN AUSTRALIA
E.C.F. BIRD (Department of Geography, University of Melbourne)
Introduction
In south-eastern Australia, as elsewhere, beach
sediments are mainly derived from eroding coast-
lines, from rivers that supply sediment to the
coast, or from the sea floor, swept shoreward by
wave action. Locally, beach sands are supplied
by dunes spilling on to the shore. Sandy beaches,
consisting of particles with diameters between
0.1 and 2.0 millimetres, are predominant on this
coastline, but shelly, gravelly and bouldery
beaches also occur locally.
Many beaches contain sediment derived from more
than one source, and on coastlines with sectors
of eroding cliff and frequent river mouths it may
be difficult to determine the relative proportions
of coastal, fluvial and sea-floor supply in the
composition of the associated beaches. Some
beaches are still receiving sediment; others are
relict, the sediment source being no longer avail-
able or the process of delivery nO longer effect-
ive; and others are dwindling, losses of sediment
(onshore by deflation to backshore dunes or wash-
over by large waves, alongshore by drifting, or
offshore to the sea floor) exceeding any material
arriving.
Beach sediments also show features produced by
weathering, abrasion and sorting in the course of
their delivery to the coast and subsequent re-
working by shore processes, particularly as they
drift the sediment to and fro along the shore.
As a result there is often considerable variation
in mineralogy and granulometry over the surface
of a beach. The upper beach, above mid-tide level,
is often coarser in texture and steeper in profile
87
than the beach at lower levels because of swash-
and-backwash sorting effects, but this does not
apply where the foreshore is occupied by relict
coarse materials such as cobbles and boulders,
unrelated to shore processes at work on an upper
sandy beach. Further variations in sediment
characteristics accompany minor features such as
berms, beach cusps, swash marks, ripples, rills
and bars.
As well as these spatial variations beaches
change over time in response to tidal movements,
weather conditions and wave regimes. Vertical
sections through a beach often show contrasted
sedimentary horizons, each of which represents a
particular depositional phase related to wave and
current processes accompanying fluctuations of
weather and tide. After a storm an eroded beach
may retain superficial deposits of relatively
coarse material or minerals of high specific
gravity, which are subsequently buried by finer
sediment deposited when wave action is less
vigorous. Laminated sections thus indicate alter-
nations of processes during a period of overall
beach accretion.
There are also lateral variations in beach
materials within embayments, close to rocky
headlands or nearshore reefs, and towards river
mouths. These too may show temporal changes
linked with process fluctuations: for example,
fluvial sediment delivered during an episode of
river flooding may dominate a beach adjacent to
the river mouth until it is dispersed by long-
shore drifting, and shelly debris washed up from
88
a nearshore reef or shoal during a storm may
persist for a time on that part of the shore.
Because of these variations generalisations
about the predominant characteristics of beach
sediments should ideally be based on comprehensive
statistical analyses of grain-size distributions
and properties of mineral constituents obtained
from systematic sampling over many years.
In the absence of such information it is prac-
ticable to base descriptions of beach sediments on
modal characteristics of samples taken at mid-
tide level in central sectors on calm days, there-
by discounting local and short-term variations.
Such reconnaissance studies form the basis of the
present account of beach sediments in south-
eastern Australia.
Nature and origin of beach sediments.
South-eastern Australia is here taken as ex-
tending from Encounter Bay (South Australia)
through Victoria to Sydney in New South Hales
(Fig. 1). The beaches of Tasmania have been de-
scribed by Davies (1978).
The long sandy beach bordering Encounter Bay
shows a transition from quartzose sands, derived
from erosion of the weathering mantle of granitic
rocks in the Victor Harbour region and from streams
bringing similar material from the immediate
hinterland, to carbonate sands composed of shelly
debris and material (notably foraminifera and
bryozoans) washed in from the sea floor. The
proportion of carbonate sand increases from less
than 10 per cent at Victor Harbour to 40 per cent
near Goolwa and almost 90 per cent at Kingston to
the south-east. At Goolwa some of the quartzose
sand may be derived from sediment formerly brought
down by the Murray River and deposited on the sea
floor during Late Pleistocene low sea level epi-
sodes, but at present the Murray flows into coastal
lagoons (Lakes Albert and Alexandrina) behind
barrier islands, and any direct contribution to
beach sediments is prevented by barrage weirs
constructed forty years ago to keep sea water out
of the Murray-mouth lagoons. To the south-east
the sandy barrier seaward of the Coorong Lagoon
is still receiving shelly sand washed in from the
sea floor, the finer fraction being blown inland
as dune formations.
Carbonate sands carried in from the continental
shelf dominate the beaches of south-eastern South
Australia, occupying embayments and coves on an
eroded coastline of Pleistocene dune calcarenite.
Near Cape Northumberland the sandy beach material
is accompanied by worn flint nodules derived from
sea floor outcrops of Oligocene limestone (Bouta-
koff 1963).
The beach sands of Discovery Bay and Portland
Bay in western Victoria are also predominantly
carbonate sands washed in from the sea floor; in
part derived from the marine erosion of submerged
Pleistocene dune calcarenites. Shelly gravels,
extensive in Bridgewater Bay, are locally
comminuted to coarse sand as at Hhites Beach
near Cape Duquesne (Gell 1978). Sediment derived
from volcanic outcrops (mainly basalt and tuff)
is also present in coves on the Portland peninsula,
and near Port Fairy where Quaternary basalts reach
the coast: here, too, there are shelly beaches on
a rocky bouldery shore which affords good habitats
for gastropods.
In the Harrnambool district, beaches of carbon-
ate sand include material derived from the erosion
of cliffs cut into Pleistocene dune calcarenites
(Gill 1967), and similar beach sands extend along
the Port Campbell limestone coast (Baker 1943)
and on towards Cape Otway, with variations re-
lated to material incorporated from Tertiary
rocks (e.g. the gravels at Pebble Point) and
sandstones from the Cretaceous formations of the
Otway Ranges.
East of Cape Otway the beaches occur inter-
mittently along a coastline that faces south-
eastwards, and is thus not directly in receipt
of the prevailing south-westerly ocean swell.
Calcareous sand blown across Cape Otway in
Pleistocene times forms a dune calcarenite which
locally extends down to the lee shore, where it
is eroded to nourish a beach immediately to the
COASTLINE OF SOUTH-EASTERN AUSTRALIA

Wollongong

C Llptrap

N
.'0
'h-----.w 20km' «.
Kingston

Robe
\)
.'>
eachport

'Rea Bluff
C. Northumberland ~'II
~o
..
C Duquesne_
e" Port .'4.
Br-idgew ater ~<:'
---.Campbell
Pebble PI
'Apollo Bay
...._ _t = = =_ _ _==c3200 km C.Olway

FIGURE 1 - South-eastern Australia

00
'"
90
east, but the beaches at Apollo Bay have been
derived by shoreward drifting of sand that has
come round Cape Otway on the sea floor. The Creta-
ceous formations of the Otway Ranges are dominated
by felspathic sediments, many of which weather to
fine silts which are not retained on local beaches.
Near Lorne they give place to Tertiary sandstones,
limestones and clays, and here the beaches are
sandy, largely of local derivation, their com-
position being similar to that of sandy fractions
in coastal outcrops. However, sectors with a
southerly or south-westerly aspect (Point Road-
knight near Anglesea, and at Barwon Heads and Point
Lonsdale) show an increased proportion of carbonate
sands of sea floor derivation, and these are also
dominant from Point Nepean to Cape Schank, west of
Cape Woolamai and Cape Paterson, and in Venus Bay
and Waratah Bay in South Gippsland, as well as in
some of the west-facing bays on the granitic coast
of Wilson's Promontory.
This pattern strongly suggests the receipt of
shelf sands carried shoreward during and since
the Holocene marine transgression to sectors with
a southerly, south-westerly or westerly aspect by
the prevailing ocean swell (Fig.2). The presence
of Pleistocene dune calcarenites on these sectors,
in places cut back into cliffs bordered by shore
platforms, is an indication that this process of
shoreward drifting also took place during earlier
phases of sea level oscillation. The accumulation
of these calcareous sand deposits has been partly
the outcome of onshore movements by wave action,
but some of the dune calcarenites could also be
the result of wind-drifting of sandy material
left on the sea floor during episodes of falling
sea level that accompanied cooling phases of the
Pleistocene, and perhaps also dunes spilling
inland in advance of rising seas during inter-
vening and succeeding marine transgressions
(Bird 1975).
Where these beaches of carbonate sand are
interrupted by rocky headlands and cliffs cut
into formations other than Pleistocene dune
calcarenite, other locally-derived beach materials
have been incorporated. Thus near Flinders and on
the south coast of Phillip Island there are beaches
of black sand and gravel derived from the erosion
of Tertiary basalts; near Kilcunda and Inverloch
Cretaceous sandstones have yielded quartz and
felspar sands; and on Cape Liptrap intricately-
jointed Palaeozoic outcrops are the source of
gravelly beaches (Bird 1972).
Port Phillip Bay has a narrow entrance, and is
only affected to a limited extent by ocean swell,
which has washed carbonate sands in as far as
Swan Island to the west and Portsea to the east.
Beaches of quartzose sand and ferruginous gravel
have been derived from the Tertiary sedimentary
formations which outcrop on the Bellarine Penin-
sula and along the north-east coast (Bowler 1966);
quartzose sand from the weathered mantle of the
Hount Hartha granodiorite to the east (Beasley
1971) and carbonate sand from the Pleistocene
dune calcarenites which extend to the northern
shores of the Nepean Peninsula between Port sea
and Dromana (Beasley 1969). The west coast of
the bay, bordering basalt plains, has beaches
dominated by shelly material generated from
benthic fauna in an environment where the coastal
rock outcrops are not sand-yielding and where the
few streams that drain the hinterland carry silt
and clay, rather than sand, to the coast. As
shore protection works halt cliff erosion and cut
off sand supply on the north-east coast, beaches
have become depleted of quartzose sand and the
proportions of shelly sand and gravel have in-
creased. Hud Islands, an atoll-like sandy island
with a central salt marsh, has been formed by the
deposition of shelly sand washed up from surround-
ing shallows (Bird 1973).
Westernport Bay has beaches consisting of
mixtures of quartz and carbonate washed in from
the ocean and lining the basalt coasts from
Flinders to Somers and Sandy Point to the north,
and the northern shore of Phillip Island past
Cowes to Observation Point (Bird 1981). Elsewhere
 91

VICTORIA ,
LAKES '-.. "" j I
ENTRANCE~--.::-::-~--..-.-/ I
s~ PtHicks / '
,"--"""- ..... - - / /
,,
I

/
I
-,
I
N Promontory \
King •

----i) Flind~rs
"Deal I. :

1
I. ~ __ 1.\ Q
\

, -
~
I
I
<J>\
"5\
'" \
...::;..\~
o I
I "
\ I
" ,
\
I

I
100 I~O I TASMANIA
"

FIGURE 2 - Sectors of carbonate sand on beaches on

the coast of Victoria and islands in Bass
Strait related to shoreward drifting of
marine biogenic carbonates from the con-
tinental shelf by waves from westerly
directions during and since the Holocene
marine transgression.
92
around this bay, beaches are of local derivation,
mainly from shore outcrops of Mesozoic and Tertiary
sandstone (J!arsden, Hallett 1975), with admixtures
of shelly material washed in from tidal mudflat and
bay floor habitats (Gell 1978). Sand blown across
the Woolamai Isthmus has supplied beaches in Clee-
land Bight alongside the eastern entrance to
Westernport Bay.
There are constrasts in the composition of beach
sands occupying the coves and embayments around the
granitic upland of Wilson's Promontory. Squeaky
Beach, in Leonard Bay on the west coast, consists
almost entirely of well-sorted clear quartz grains
of medium sand size, mostly sub-rounded to rounded,
with a fairly high degree of sphericity, the dry
sand emitting a squeak when walked upon. By con-
trast, the beaches in adjacent Picnic Bay to the
north and Norman Bay to the south contain up to 40
per cent of carbonate sand, are generally finer
and less well sorted, and do not squeak (Beasley
1972). Oberon Bay, farther south, has a higher
proportion (50 to 70 per cent) of carbonate sand,
but on the east coast of Wilson's Promontory the
beach sands are again predominantly quartzose,
notably the coarse pale quartz sand in Waterloo
Bay. The quartzose material has been largely de-
rived from the weathered mantle of the Wilson's
Promontory granites, which aprons the lower slopes
and thickens into valley floors. Locally, as in
Little Oberon Bay, this material is being cut back
in low cliffs, yielding quartz sand and some gravel
to the adjacent beach. More generally, these
weathered deposits must have been washed over by
waves during the later stages of the Helocene
marine transgression, when sandy material was
swept shoreward on to existing beaches. On west-
facing beaches this quartzose sand has been
accompanied by varying proportions of carbonate
sand carried in from the sea floor in the manner
mentioned previously, but east-facing beaches,
sheltered from the prevailing south-westerly swell,
have not received carbonate sand. Squeaky Bay is,
nevertheless, an anomaly in terms of this explan-
ation: it may owe its high proportion of quartz
to the interception of shoreward-moving carbonate
sands by offshore reefs and islands.
On the north-east and north-west coasts of
Wilson's Promontory the beach sands are quartzose,
but yellow in colour because of a retained iron
oxide staining. It is possible that the pale or
white beaches on Wilson's Promontory have been
derived from the leached zone of the weathered
mantle of granites, whereas those derived from
underlying horizons, stained by down-washed iron
oxides in the weathering profile, are yellow in
colour. Similar contrasts in quartzose sand
colour have been noted on granitic coasts on
Flinders Island and in north-eastern Tasmania
(Bird 1978a). On the western shores of Corner
Inlet, beaches have been nourished by wind-blown
sand driven across the Yanakie Isthmus from the
ocean coast of Waratah Bay.
East of Wilson's Promontory the barrier islands
off Corner Inlet are predominantly well-sorted
and subrounded to rounded quartzose sand, with
only small proportions (less than 10 per cent)
of carbonates, except for occasional strandline
accumulations of shelly debris. The same is true
of the Ninety ~tile Beach. Apart from quartz and
local accessions of modern shelly debris, the
chief constituents are small quantities (usually
less than 0.5 per cent) of heavy minerals, in-
cluding ilmenite, tourmaline, garnet and monazite
(Tan 1970). These beaches form the seaward
margins of a dune-capped outer barrier, the most
recent of a sequence of Quaternary barriers de-
posited on the coast in such a way as to enclose
the Gippsland Lakes and other lagoons and swamps
(Bird 1978b). The quartzose sand which forms
these barriers must be of marine origin, washed
in from the adjacent sea floor, for apart from
the small Merriman's Creek at Seaspray the streams
draining towards this coast discharge into lagoons,
and thus cannot have delivered sand to the sea.
Moreover, there are no cliff or foreshore outcrops
yielding sand, except for Red Bluff, at the

eastern end of the Ninety Mile Beach, where marine
erosion of Tertiary sandstones contributes some
sand locally. It is deduced that the quartzose
sands of the Ninety Mile Beach sector have been
derived from sea floor deposits, including fluvial
sediment laid down during Pleistocene low sea
level phases and material produced by the weather-
ing of subaerially-exposed granites east of
Wilson's Promontory and Tertiary sandstones on the
Bass Strait shelf. In contrast with shelf areas
west of Wilson's Promontory, this sea floor region
has yielded mainly quartzose sand, a reflection of
the predominance of granitic formations on the
floor of Bass Strait and in the river catchments
of East Gippsland. During alternations of marine
regression and transgression through Quaternary
times these sands have been sorted and transported
landward, to be built into the successively-
prograded coastal barriers of East Gippsland
(Bird 1971, 1976). Under present conditions the
Ninety Mile Beach, in common with most of the
other beaches in south-eastern Australia, is being
cut back by marine erosion, so that the existing
beaches are reincorporating sediment previously
deposited in backshore dunes and beach ridges
(Bird 1983).
East of Red Bluff the Gippsland coast is lined
by quartzose beaches similar to the Ninety Mile
Beach, but interrupted by headlands of granite
and Palaeozoic metamorphic rock. The chief
variation is around the mouth of the Snowy River,
where fluvial sand and gravel are carried into
the sea by occasional floodwater discharge. After
such episodes it is possible to detect accessions
to the beach of angular and subangular quartz,
felspar (which is otherwise rare on the East
Gippsland beaches) and heavy mineral (notably
augite) grains, similar to the material in the
shoal deposits of the Snowy River at and down-
stream from Orbost, but the quantities are very
small compared with the mass of well-sorted, well-
rounded quartzose sand, and these accessions are
soon dispersed laterally by wave action CR. W.
93
NcLennan, unpublished). Concentrations of heavy
mineral sands occur locally along this coasfline
(Baker 1945).
On the south coast of New South Wales (i.e.
from Cape Howe north as far as Sydney) the beaches
occupy embayments separated by rocky and cliffed
sectors, and are also predominantly quartzose,
forming the seaward margins of barriers that have
been nourished by shoreward drifting of sand from
the sea floor (Bird 1967). Some sand has also
been derived from cliffs cut into Devonian sand-
stones which outcrop on the coast south of Tathra,
and Permian sandstones north of Durras, but the
intervening rocky coasts consist of schist, chert,
basalt and crystalline rocks, none of which yields
sandy material, although boulders and gravels have
been produced locally. As in East Gippsland,
most of the rivers drain into lagoons enclosed by
barriers, and are not presently supplying sand to
the coast, but the Towamba, Bega, Moruya ,and
Shoalhaven Rivers have each filled in the lagoons
that formerly existed at their mouths, and when
they flood these rivers do carry sand and gravel
into the sea, to be subsequently deposited on
local beaches. Whereas the beach sands on this
coast are generally well-sorted subangular to
rounded medium sand, largely composed of quartz
with varying proportions of shelly debris and
little felspar, the fluvial sands are less well
sorted, poor in shelly debris and rich in felspars,
having been derived from the deeply-weathered
granite outcrops in the river catchments. Whale
Beach, at the mouth of Towamba River, is a dis-
tinctively coarse beach with sediment character-
istics similar to those of river shoals upstream
at Kiah (Hails 1967, Bird 1967). On a larger
scale, Seven Nile Beach north of the mouth of
Shoalhaven River is frequently in receipt of
flood-supplied fluvial sand containing felspars
as well as quartz, accompanied by small pebbles,
carried northward from the river mouth by pre-
vailing south-easterly wave action (Wright 1970).
Away from these four river mouths the beach sands
94
have come mainly from the adjacent sea floor, and
probably represent re-worked fluvial deposits of
Pleistocene low sea level phases, within which
wave attrition has depleted the less resistant
felspars, leaving a predominance of quartz. In
the Sydney district sands washed in from the
adjacent sea floor are mixed with quartzose sands,
often coarse in texture, derived from weathering
and marine erosion of cliffs and shore platforms
cut in Triassic sandstones.
Conclusions
This account of the characteristics of beach
sediments along the coastline of south-eastern
Australia permits some generalisations concerning
their origin and mode of delivery. The beaches
are predominantly sandy, gravels occurring only
in limited sectors where coastal outcrops are
weathering into suitably coarse rock fragments.
This occurs where the outcrops include harder
layers, such as the calcerate horizons in the
Pleistocene dune calcarenites of South Australia
and western Victoria; where the coastal rocks are
intricately jointed, as in the basalts on Phillip
Island, the granite at Cape Woolamai, and the
metamorphic rocks at Cape Liptrap, and where
coastal outcrops contain harder nodules, such as
the rhizoconcretions found in Pleistocene dune
calcarenites. Beach gravels are not found where
the coastal rocks are hard and homogeneous, as
on the massive granites of Wilson's Promontory,
or soft and homogeneous, as on the Port Campbell
Limestone in western Victoria (Bird 1972). In
general, the correlation between coarseness of
beach sediment and exposure to high wave energy
is weaker than Davies (1978) found in Tasmania:
for example, gravelly beaches are found on low
wave energy shores in Westernport Bay, and fine
sand beaches are not uncommon on exposed oceanic
coasts. The nature of available source materials
is often a more important factor in beach granul-
ometry than variations in wave energy, although
the latter process does add to sorting within
particular beach compartments.
This point is well illustrated by the distrib-
ution and local sorting characteristics of beaches
composed partly or wholly of shell fragments.
Such shelly beaches (gravels as distinct from
comminuted biogenic sands) are localised in re-
lation to nearshore environments where shelly
organisms are abundant, such as rocky reefs,
shallow marine embayments, and estuarine muds.
On exposed coasts they form swash-piled deposits
in coves or behind rocky outcrops, whereas on
sheltered sectors they form strandline accumul-
ations. Gastropods are more abundant behind ex-
posed rocky shores, while pelecypods inhabit
sandy beaches and estuarine fringes (Gell 1978).
Pumice gravels, produced by distant volcanic
eruptions and floated across the oceans, occur
patchily behind sandy beaches in south-eastern
Australia, while artificial materials such as
glass and plastic are becoming increasingly pro-
minent on litter-strewn beaches close to urban
and industrial centres such as Melbourne.
The predominance of sandy beaches is related
to the widespread availability of arenaceous
formations, notably weathered granites and sand-
stones, in coastal rock outcrops, within river
catchments, and on parts of the adjacent con-
tinental shelf. In addition, benthic organisms
and the disintegration of shelly fauna yields
marine carbonates, which have been moved onshore
from the south-west to the coasts of South
Australia and Victoria west of Wilson's Promon-
tory. These calcareous sediments dominate
beaches in the relatively arid sectors where the
sea floor did not receive fluvial sediments during
Pleistocene low sea level phases; they are diluted
on coasts behind shelf sectors that have received
more substantial fluvial deposition: for example
east of Wilson's Promontory and on the south coast
of New South Wales.
The chief source of supply of sediment to
beaches in south-eastern Australia has thus been
the adjacent sea floor, derivation from retreating
 95

cliffs and shore erosion, or from fluvial sources, Port Phillip Survey 1957-63, Mem. Nat. Mus. Melb.,
27: 19-67.
having been limited and localised. It is possible Davies JL (1978) Beach sand and wave energy in
that similar features exist on other coastlines, Tasmania, in J.L. Davies and M.A.J. Williams
(eds.) Landform Evolution in Australasia: 158-167.
for example in South Africa from Port Nolloth Gell RA (1978) Shelly beaches on the Victorian
around to Durban, a sector roughly comparable in coast, Proc. Roy. Soc. Victoria, 90: 257-269.
Gill ED (1967) Evolution of the Warrnambool-
extent, aspect and environmental context with Port Fairy coast, in J.N. Jennings and J.A.
the south-eastern Australian coastline here re- Mabbutt (eds.) Landform Studies from Australia
and New Guinea: 341-364.
viewed, but without the complications of Bass Hails JR (1967) Significance of statistical
Strait and outlying Tasmania. Comparisons of parameters for distinguishing sedimentary en-
vironments in New South Wales, J. Sedim, Petrol.,
beach sediments on these two southern continents 37: 1059-1069.
may prove instructive. Marsden MAR and Mallett CW (1975) Quaternary
evolution, morphology and sediment distribution,
Westernport Bay, Proc. Roy. Soc. Victoria, 87:
References 107-138.
McLennan RW (1972) Beach Sands and the Snowy
Baker G (1943) Features of a Victorian limestone River system. Unpublished thesis, University of
coastline, J. Geol., 51: 359-386. Melbourne.
Baker G (1945) Heavy black sands on some Tan SH (1970) Sediments sorting across the
Victorian beaches, J. Sedim. Petrol., 15: 11-19. Lakes Entrance inlet, Min. and Geol. Journ.
Beasley AlV (1969) Beach sands of .the southern (Victoria): 7: 14-16.
shore of Port Phillip Bay, Victoria, Mem. Nat. Wright LD (1970) The influence of sediment
Mus. Melb., 29: 1-21. availability on patterns of beach ridge develop-
Beasley AW (1971) Safety Beach, Dromana, Vi.c. ment in the vicinity of the Shoalhaven River
Nat., 88: 291-295. delta, Austr. Geogr., 11: 336-348.
Beasley AW (1972) Sands from Squeaky Beach and
adjacent beaches on Wilson's Promontory, Mem.
Mus. Melb., 33: 47-54.
Bird ECF (1967) Depositional features in
estuaries and lagoons on the south coast of New
South Wales, Australian Geogr. Studies,S: 113-
124.
Bird ECF (1971) The evolution of sandy barrier
formations on the East Gippsland coast, in J.A.
Steers (ed.), Applied Coastal Geomorphology:
45-63.
Bird ECF (1972) Beach gravels, Vic. Nat., 89:
180-185.
Bird ECF (1973) Physiographic changes at Nud
Islands, Port Phillip Bay, Vic. Nat., 90: 157-165.
Bird ECF (1975) The shaping of the Nepean Pe-
ninsula, Vic. Nat., 92: 132-141.
Bird ECF (1978a) The nature and source of beach
materials on the Australian coast, in J.L. Davies
and M.A.J. Williams (eds.) Landform Evolution in
Australasia: 144-157.
Bird ECF (1978b) The geomorphology of the Gipps-
land Lakes region. Ministry of Conservation,
Victoria, Publication 186.
Bird ECF (1979) Geomorphology of the sea floor
around Australia, in J.R.V. Prescott (ed.)
Australia's Continental Shelf: 1-21.
Bird ECF (1981) Victorian coastal geomorphology,
Proc. Roy. Soc. Victoria, 92: 19-35.
Bird ECF (1983) Factors influencing beach
erosion and accretion: a global view, Sandy
Beaches as Ecosystems.
Boutakoff N (1963) The geology and gemorphology
of the Portland area. Geol. Surv. Victoria, Mem.
22.
Bowler J (1966) Geology and geomorphology, in

PROPERTIES OF LOGARITHMIC SPIRAL BEACHES WITH PARTICULAR REFERENCE TO ALGOA BAY
J.M. BREMNER (Marine Geoscience Section of the Geological Survey, University of Cape Town, Rondebosch
7700)
1. INTRODUCTION
A hundred and fifty years ago De la Beche
(1833) and later De Beaumont (1845) recognized
that an intrinsic relationship existed between the
angle of wave approach and the pattern of shoreline
curvature (Lewis, 1937). The empirical nature of
this assessment was critically received however,
and geomorphologists during the following hundred
years or so, e.g. Gulliver (1899) and Wheeler
(1902), promoted longshore currents for the forma-
tion of curved shorelines. In 1906, Halligan
formally named assymetrically-curved beaches loca-
ted between headlands, calling them 'zeta-curve'
bays after the letter in the Greek alphabet. Much
later, Silvester (1960) succeeded in producing what
he called 'half-heart shaped bays' in a model wave
tank, and Yasso (1964a), with the aid of a computer,
fitted the outline of natural 'headland-bay beaches'
to segments of a logarithmic spiral. Following the
work of Longuet-Higgins (1970) on wave-induced
currents and Komar's (1971) investigation into the
mechanics of sand transport by these currents, the
early wave versus longshore current controversy
was essentially resolved, and the stage set for
theoretical modelling.
In spite of the log-spiral concept being
critized on grounds that it lacks physical justi-
fication (Tanner, 1976; Walton, 1977), numerical
(LeBlond 1972; Rea, Komar, 1975) and analytical
(Walton, 1977; LeBlond, 1979) models have all
succeeded, using different approaches, to produce
a logarithmic spiral shape. LeBlond (1972) used
the theory of wave-induced longshore currents and
related this to the rate of erosion along a beach
97
oriented obliquely to the incident wave field. His
model eventually led to the formation of hook-like
beaches, but it suffered to some extent from an in-
sufficiencyof data points on the generated curves.
Rea, Komar (1975) took into account the effects of
wave diffraction behind a headland, but were unable
to incorporate changes in the wave field as the
nearshore bathymetry adjusted to increasing in-
dentation of the coast. Walton's (1977) model
was based on an offshore wave-spectrum and a shelt-
ered equilibrium shoreline. With this he produced
curves very close to a logarithmic spiral, but in
his opinion, the good fit was simply a fortuitous
coincidence. LeBlond's (1979) model successfully
employed an empirical relationship between the
interactive variables - wave energy, grain size
and beach slope - to construct synthetic beach
forms of log-spiral shape.
The aim of this paper is not to present an
additional theoretical model, but rather to exam-
ine in depth some of the physical properties of
spiral-shaped beaches that have been proposed by
Silvester (1960, 1970) and Silvester, Ho (1972).
Furthermore, the contemporary outline of Algoa
Bay is evaluated in the light of these findings,
and speculations are made about its geomorphic
development, its long-term stability, and poss-
ible mutations that would occur following alter-
ation of the environment.
2. THE LOGARITffi~IC SPIRAL
An equiangular or logarithmic spiral is
defined by:
p = Re 0cotCi.
98

where p the length of the radius vector
R length of the radius vector at 0 = 0
o the angle between the radius vector
and a chosen azimuth (in this case
the azimuth 0 = 0 is defined as being
parallel to the +x axis on a cartes-
ian grid)
a = the spiral angle, which is constant
for any given log-spiral (measured
between p and a tangent to the curve
at the point of intersection).
These, and other variables, are illustrated in
Fig.1.
For this purpose, the above equation can be
rewritten:
Inp = a0 + b
where constants a(=cota) and b(=lnR) des-
cribe the linear regression lines shown in Fig.9.
3. FITTING SPIRAL CURVES TO EMBAYED COASTLINES
Two different techniques have been used to
find the log-spiral curve that most closely
approximates the outline of an embayed coastline.
One of them involves the use of a computer
(Yasso, 1964; Bremner, LeBlond, 1974) and the
other makes use of overlays on transparent

+y

+x

FIGURE 1. Parameters of a clockeise logarithmic spiral curve. Also shown are the
wave obliquity, the bay dimensions for establishing an indentation ratio, and
the positioning of positive axes for assigning coordinates to data points on the
log-spiral curve.

Since p changes at a much faster rate than
0, it is convenient to plot these curves on semi-
log paper, which also has the effect of reducing
the scatter of data points about the theoretical
straight line.
paper (Silvester, 1970; Gallow, 1980). The
former method is more exacting in that all data
points supplied from the outline of the beach
are taken into account in fitting the log spiral,
but suffers to some extent in being more com-
 99

plicated and time-consuming than the overlay set of logarithmic spiral-curves with spiral
method. angle ranging from say 15° to 80°. An example of
an anticlockwise set is shown in Fig.2.
The programme used here was written by
LeBlond, Bremner. It requires that the following
information be supplied: Coordinates of the
beach data-points (all positive with abscissa
parallel to latitude, and ordinate parallel to
longitude); the spiral direction of the beach
(+ve clockwise, -ve anticlockwise); the coordin-
ates of an arbitrarily-chosen, initial spiral-
centre; and the starting scale of a search square
that would be set up around the initial centre.
The four corners of the search square then serve
as additional centres for testing by least squares
procedures. When the best centre is located, the
search-square scale is reduced to one fifth of
its original size, and the whole procedure re-
peated. This ongoing process is continued until
the scale falls below a predetermined minimum
value.

The absolute difference between the theor-

etical radius vector (p), and the actual radius
vector (r) for each data point is expressed as a
percentage of three times the root mean squared
error (RMSE), thus giving it a confidence level FIGURE 2. Anticlockwise logarithmic spiral curves
with a from 15° to 80° (Gallow, 1980). Drawn on
of 99%:
transparent paper, the set of curves is used as
an overlay for determining the a-angle of spiral-
ABS (p-r) X 100 shaped beaches. Note that Silvester's (1970)
ie % difference
3 X RMSE overlay curves did not progress as far into the
tight portions of the curves as these do.
The spiral angle is determined from the follow-
ing relationship: The procedure is to optically adjust the scale
of the bay (on map or areal photograph) to suit
a arc tan (1/ )
a the scale of the overlay, and then to move the
latter about until the best fit is found. Thus,
Output from the programme includes coordinates of
only one parameter, viz., the spiral angle, is
the best-fitting spiral centre, constants a and b
determined by this method, and uncertainty often
for the best-fitting regression line, the root
exists as to which part of the bay should re-
mean squared error, the spiral angle, coordinates
ceive greatest fitting emphasis. Silvester
of points on the best-fitting spiral that corre-
(1970) attached most importance to the central
spond to the given data-points, the length of
part of the beach, whereas Gallow (1980) tried
radius vectors p and r at consecutive e angles,
to fit the entire length. Often powever, the
and the percentage difference between p and r.
gently-curved, down-coast part of a beach would
The overlay method involves constructing a be far-removed from the best-fitting spiral
100
overlay.
4. THE EQUILIBRIUM SPIRAL BEACH
The planimetric shape of a spiral-shaped
beach is said to be in equilibrium when incident
waves break simultaneously around the length of
the bay (Silvester, 1960; LeBlond, 1979). When
this condition is met, the gently-curved down-
coast part of the beach will be parallel to the
crests of incoming deep-water wave fronts. In
this region, waves are only affected by refract-
ion on the shoaling bottom whereas in the lee of
the headland, they suffer the combined effects
of refraction on the bottom, and diffraction
around the headland. Because this would cause
waves to be smaller here than on the downcoast
part of the beach, the grain-size of sediment
will be finer and the slope of the beach will be
gentler than farther downcoast. Nevertheless,
the wave energy-flux is everywhere normal to the
beach with the result that no longshore currents
or littoral drift will occur. Thus, besides the
log-spiral shape of the beach and the breaking
waves, isobaths in the bay will also assume this
curvature if sufficient unconsolidated sediment
exists.
The steady-state condition described above
is of course ideal. In reality, the seafloor
may be an artefact from lower sea-level stands,
which would have an effect on the relative size
of waves reaching the beach, and in turn, would
influence the outline of the beach. Another
aspect ignored by the model is the existence of
a wave-generated diffraction-current in the lee
of the headland. The current is formed by an
energy transfer from a region of high waves to a
region of low waves; it follows a log-spiral path
and is capable of transporting fine, suspended
sediment out of the bay, past the headland, and
into deep water (Gourlay, 1974).
5. VARIABILITY OF THE SPIRAL PARAMETERS
Silvester (1970) showed, from experiments in
a wave tank, that the initial slope of an arti-
fici.al beach had no effect on the final equili-
brium-shape of the eventual embayment. However,
the steeper the initial beach, the faster it
approached stability. The same principles held
for the incident energy of the waves - the longer
the period (more energy), the faster the beach
eroded to its stable form. This important ob-
servation led him to conclude that the scale of
the operation was not significant, and the model
could represent waves of any size, and the bays
of any magnitude. The one overriding factor in
determining the final shape of an equilibrium
beach however, was the angle of wave obliquity
S. He used this predetermined value therefore
to see how it affected the other spiral para-
meters.
5.1. The spiral angle a
For anyone S angle, the value of a was
found to increase swiftly at the beginning of a
test, and less rapidly as the beach approached
equilibrium. Gallow (1980) obtained a similar
result, both workers using overlays to measure a
at discrete stages in their experiments.
The final a-angles for three test beaches
were plotted against the respective angles of
wave incidence ie., S ~ 30 0 , 45 0 and 60 0 (Sil-
vester, 1970), and found to locate on a straight
line of negative slope. Further refinement, with
points derived from several natural beaches,
gave a similar, but gently upward-sloping curve
(Silvester, Ho, 1972). Gallow's (1980) plots of
artificial and natural beaches on the other hand,
gave rise to a wide scatter of points about a
line with positive slope. The two curves, almost
at right angles to each other, are shown in Fig.3.
Gallow (1980) has attributed Silvester, Ho's
(1972) line to faulty reading of the a angle.
This, he believes, was due to Silvester, Ho's
(1972) overlay curves not going far enough to-
ward the spiral centre (see Fig.2). As mention-
ed earlier however, Silvester (1970) emphasized
the central part of the beach whereas Gallow
(1980) tried to fit the entire coastline.
 101

90
\
\
\ X
\
\
\
80
,
\
\
X X
\
\
\
\
• • 3O-hour beach
70 \
\
'\
....... .,.,...-- _ - GALLOW (1980)

\.
\.
X
o
X
X

0
XAlgoa Bay(S-49)

'"
....
<!) 50
X
Z
<I:

0 0 Ii>.
X X X

40 0 0 0 00 0
X ~ - - - - SILVESTER & HO (1972)
X
0
X
30 0 00 00
0

20 0 00 Ii>. Australia
0
0 Western Cape
and California
00
/
X Southern Cape
I
10 I El Wave tank
I
/
/
/
I
0
0 10 20 30 40 50 60 70

WAVE OBLIQUITY 11 (DEGREES)

FIGURE 3. Plot of the spiral angle a versus wave obliquity 8. Triangles represent Austral-
ian beaches; a measured with overlays (Silvester, Ho, 1972, Table 1). Circles represent
beaches from the Cape Peninsula, the Cape west coast, and from California; a measured with
overlays (Gallow, 1980, Tables 3.1 and 3.2). Crosses represent beaches on the Cape south
coast between Cape Agulhas and Cape Padrone; a determined by computer; and squares re-
present consecutive beaches produced in a model wave tank (Silvester, 1970, Fig.5); a de-
termined by computer.

In an attempt to resolve this conflict, the
writer analysed twelve beaches along the Cape
south coast between Cape Agulhas and Cape
Padrone (Fig.4).
dented by numerous spiral-shaped beaches of
varying size, and between them are resistant
headlands, usually of Table Mountain Group quart-
zite. The spiral angle a was determined by com-
puter from the supplied data-points, and 8 was
measured by drawing a tangent to the beach from
This stretch of coast is in- the down-coast headland, and then measuring the
angle between the tangent and a line joining the
two headlands. Both a and 8, together with other
bay parameters, are shown in Table 1.
102

CAPE SOUTH COA ST

2 0

FIGURE 4. Location of twelve beaches on the Cape south coast that were tested for logarithmic
spiral curvature.

Regarding the supplied data-points, consid- period at 1.4 seconds.

erable uncertainty existed in some cases as to
It is clear from the diagram (Fig.3) that
where the headlands began or the beaches termin-
the trend of points (squares) shows closer alle-
ated, and the lee-side oJ the up-coast headland
giance to Gallow's curve, although still some-
was frequently convex and difficult to include
what removed from it. Initially, a increased
in the beach curvature. Furthermore, the tangent
rapidly with a remaining almost constant, and
to the down-coast part of a beach was not, as
finally, a increased rapidly with a increasing
prescribed for equilibrium spiral-shaped beaches,
only minimally. Between these two stages, that
parallel to the deep-water wave fronts. In fact,
is, approximately midway between the initial
for the beaches studied, the tangent orientations
linear beach and the eventual equilibrium beach,
varied through 52 0 (Fig.5). It is evident
a confused zone exists where a and a vary un-
therefore, that the local nearshore bathymetry
predictably.
still exerts considerable influence on the
In summary then, it appears that a natural
approach-direction of wave-fronts, which in turn
spiral-shaped beach cannot be classified as to
determines the beach alignment.
its equilibrium state based on the location of
From the foregoing, it was not really sur-
its present (a,S) coordinate. Conversely, an
prising that points from the twelve beaches
artificial beach, undisturbed by local irregular-
scattered widely on the a vs. a plot (Fig.5).
ities in bathymetry, or by varying resistance of
In addition, no preference for either the
the shoreline to erosion, will exhibit a pro-
Silvester or Gallow curves was evident. Conse-
gressive change in the coordinate positions of
quently it was decided to test the progressive-
its development stages. The changes reflect
development stages of one of Silvester's (1970)
initial swift erosion of the coast in the lee of
model-tank studies. The a and a angles were
the upcoast headland (increasing a) followed by
determined, as previously, for each stage in the
rapid alignment of the down-coast part of the
development of the spiral-shaped beach where the
beach to incoming wave-fronts (increasing a).
wave obliquity was set at 60 0 , and the wave
 103

TABLE 1. Parameters* of best-fitting logarithmic spiral curves, wave obliquity, and indentation ratios

No.of Spiral centre

Bay name points a X y

Struisbaai 38 36 35 0.25 39.5 145.5

Marcus Bay 23 61 21 0.21 168.6 33.4
Vinbekbaai 65 43 30 0.19 114.4 92.2
St Sebastian Bay 78 44 22 0.19 27,8 -24.8
Stilbaai 46 58 18 0.16 86.0 -27.6
Fish Bay 12 39 37 0.32 99.4 30.8
Flesh Bay(l-40) 40 44 43 0.35 115.8 9.4
" (6-40) 35 51 35 0.25 121.5 4.5
Mossel Bay(l-60) 60 44 45 0.35 87.3 47.1
" (8-60) 35 60 38 0.23 110.6 40.0
Plettenberg Bay(l-80) 80 77 58 0.27 308.9 21.0
" (26-63) 38 77 55 0.11 307.4 -9.4
Slangbaai 29 86 40 0.25 90.2 20.7
St Francis Bay(l-39) 39 32 42 0.29 10.7 74.8
" (14-29) 26 58 23 0.15 111. 1 19.5
Algoa Bay (1-58) 58 47 35 0.34 6506 39.4
" (8-49) 42 53 42 0.25 78.5 26.1
Silvester's (1970) model tank study
1/4 hour beach 18 43 29 0.31 4002 9.8
1/2 19 51 21 0.32 38.8 9.2
1 20 53 29 0.36 37.8 9.9
2 21 67 32 0.36 31.2 13.8
3 23 61 35 0.40 33.6 13.1
4 26 60 30 0.39 33.9 13.2
6 27 67 34 0.41 30.7 17.3
8 31 61 37 0.45 34.0 14.3
10 32 64 42 0.44 32.6 16.3
12 32 65 47 0.45 33.9 16.8
14 32 67 49 0.47 34.0 18.7
18 32 69 53 0.49 33.8 20.1
22 32 68 55 0.49 35.4 19.4
26 32 70 57 0.51 35.4 21.4
30 32 70 59 0.51 36.3 21 8
0

*Determined by computer

FIGURE 5. Orientation of the gently-curved

down-coast portions of twelve beaches lo-
Flesh Bay cated on the Cape south coast between Cape
Agulhas and Cape Padrone. Prevailing wave
W Plettenbe Ba
direction from Harris (1978).
Struisbaai

«0 80 16 of waves
s
104

The next logical step in researching this
aspect of spiral-shaped beaches would be to see
if the equilibrium end-points of artificial
beaches fall along some coherent curve.
would necessitate a series of model tank studies
with incident wave-directions incremented at
say 5° intervals.
Both the Silvester, Ho (1972) and Gallow
(1980) curves are shown in Fig.6 again, together
with the AIB vs. a points for several natural
This beaches. It can be seen that the scatter of
points is less than in Fig.3, and furthermore,
that very few points locate above the Silvester,
Ho curve. Those points below the curve are

5.2.The indentation ratio AIB from beaches that have not receded sufficiently,

Silvester, Ho (1972) recognized that it was this being due to the presence of erosion-

sometimes difficult to position the down-coast resistant materials along the shore, islands and

part of a spiral-shaped beach with overlays, and shoals in the nearshore, etc.

suggested therefore an alternative measure to Points derived from Silvester's (1970)

gauge the equilibrium state of a beach.
was the indentation ratio AlB (Fig.6) which, in
principle, is the inverse of a 'curvature index'
proposed earlier by McLean (1967) for beaches
approximating an arc of a circle.
0.8
This model-tank study coincide fairly closely with
the curve proposed by Silvester, Ho (1972). The
trend is also similar to that seen previously
in that AlB increases initially, although some-
what erratically, followed again by S in the

30-hour beach
0.5 SILVESTER & HO(1972)

_ - -- - GALLOW (1980)

0.4
.....
ID
C
o o
0
~II:
0.3
Z x

.~
0 0 X
;e X
X XAI_ ao,18-49)
0 Western Cape
Z
!OJ and California
C 0
~ t1l ~ X Southern Cape

0 [!] Wave tank

0

0.1 X

o 10 20 30 40 50 60 70
WAVE OBLIQUITY P (DEGREES)

FIGURE 6. Plot of the indentation ratio AlB versus wave obliquity S. Circles represent
beaches from the Cape Peninsula, the Cape west coast, and from California (Gallow, 1980,
Tables 3.1 and 3.2). Crosses represent beaches on the Cape south coast between Cape Agulhas
and Cape Padrone, and squares represent successive beaches produced in a model wave-tank
(Silvester, 1970, Fig.5).
 105

later stages. 6. MORPHOLOGIC DEVELOPMENT AND STABILITY OF

ALGOA BAY
It is tempting, from the foregoing dis-
A 6.1. Geographic setting
cussion, to conclude that IB vs. S is a better
Algoa Bay is the largest and best-formed of
plot than a vs. S for assessing the equilibrium
several spiral-shaped bays along the Cape south
state, or departure therefrom, of a spiral-shaped
coast (Fig.8). At its eastern end, Cape Padrone
beach. This is because AlB is a less-sensitive
marks the boundary between the indented south
parameter to natural deformities in the shore-
coast and a rectilinear east coast, the latter
line curvature than a, and is also easier to
extending northeastwards at least as far as
determine and interpret than a.
Durban.
5.3. The spiral centre (cx, cy)
The embayment occupies the extension of a
From his pioneering work on spiral beaches
rift-type graben that onshore measures roughly
and spit-bar shorelines, Yasso (1964a,b) suggest-
4000 square kilometres (Winter, 1973), and off-
ed that the centre of a best-fitting logarithmic
shore, as defined by the limits of the study
spiral should be located close to the headland.
area, 3100 square kilometres. It is therefore
Bremner (1970), from his study of different
very large compared to the other south-coast
sections of a spiral-shaped beach, felt that the
bays, and extends for about 80 kilometres in an
centre would migrate towards the headland as the
east-west direction, and 40 kilometres towards
beach approached equilibrium. This was later re-
the south.
futed when Bremner, LeBlond (1974) showed by
means of a simple mathematical proof, that the The western end of Algoa Bay is dominated by
spiral centre would migrate away from the head- a low-lying, rocky promontory called Cape Recife,
land as the curve opened up. which protrudes southeastwards into the bay, and
rises to a height of 45 metres above sea level.
In this study, the spiral centre has been
On the leeward side of the headland is the
plotted for each consecutive stage in Silvester's
harbour city of Port Elizabeth, and on the wind-
(1970) S = 60° wave-tank model (Fig.7). It is
ward side is a small, irregular, southward-
evident that the centre locates initially right
facing embayment called Shelly Bay. A small
on the headland and as coastal erosion proceeds,
plume of sand extends across the promontory from
it goes through a series of perturbations in its
Shelly Bay, which is all that remains of a
progress away from the headland, eventually
massive dune field that,during the Wurm inter-
curling around the latter, and into the eroded
embayment. Smoothing its somewhat erratic path stadial, covered most of the southern part of
Cape Recife, and actively supplied sediment to
is best achieved by, of all things, another log-
Algoa Bay (Bremner, 1981).
arithmic-spiral curve! Two speculative infer-
ences drawn from these results are that if the The western half of Algoa Bay, between Cape
centre locates landward of a line joining the two Recife and longitude 26° OO'E, is dissected by
headlands, the beach will be close to, or at, five river systems traversing the coastal plain
equilibrium; secondly, the relative location of (Fig.ll). Two of these rivers, the Sundays and
the spiral centre cannot be used as an index for the Swartkops, are large perennial systems that
the goodness-of-fit of the beach because it is drain extensive catchments. The other three,
just too sensitive a parameter. the Coega, the Papkuils and the Baakens are
small seasonal streams, the former two support-
ing salt-extraction works, and the latter flowing
into Port Elizabeth harbour. East of the
106

22

20

18

6~"
\
\
''
",,
, ',10'
16
' ,,
~, \

"" ,
"" "- "~8
14

,
2~'" .. , WAVE OBLIQUITY
P _50"
I
>
"" " ....... WAVE FRONT

.... ...
12 .... "',
'"
!(
'
"
Z "
is
3
10 ~~
ems

2
~o
ems
POSITION OF FINAL
SPIRAL-SHAPED BEACH

°3~0~------'32C--------3r4--------3T6--------'38--~~---~r--------4T2----

ABSCISSA X

FIGURE 7. Migration of the centre of anticlockwise log-spiral curves

fitted to successive development stages of an equilibrium spiral-
shaped beach. The wave-tank model used (shown in inset) is from
Silvester (1970, Fig.S). Numerals on the beaches represent hours.
Note that the scale of the diagram is lOX that of the inset.

Sundays River, the most easterly of the five
rivers, the drainage is blocked from the sea by
a 2~ kilometre-wide strip of unconsolidated dune-
sand of between 60 and 120 metres elevation.
Bascom (1954) and Davies (1972) have both comment-
ed extensively on the deflection of river out-
lets by high beach-berms. On the gently-curved
down-coast portion of spiral beaches, high berms
of relatively coarse sand are common features.
Behind the berm, the back shore grades gently into
a low-lying coastal plain except in the vicinity
of Woody Cape and Cape Padrone (Fig.8). At these
two locations, the back shore is wooded, and
rises fairly steeply to the crest of the dunes
in the lee of rocky foreshores.
6.2. Morphological development
Diastrophic and eustatic events since early
Tertiary times have led to at least three major
sea-level transgression and regression events in
the Algoa Basin (Ruddock, 1968). Evidence for
these changes onshore is the presence of a
seaward-sloping plateau of marine Caenozoic sedi-
ments lying up to 285 metres above sea level,
and 30 kilometres from the present shoreline.
During the Wurm II regression (10Ka B.P.), indi-
cations are that sea level dropped world-wide by
at least 100 metres (Guilcher, 1969). The
shoreline has therefore experienced a number of
back-and-forth migrations, during which time
unconsolidated sediment in the region was re-
worked, and the shape of the embayment adjusted.
With the commencement of the Flandrian trans-
gression, the present-day floor of Algoa Bay was
blanketed by large volumes of sediment on the
flood-plains of the Sundays and Swartkops Rivers,
and also from dune sand blown in from the coast
westward of Cape Recife (Bremner, 1981). As
these deposits were eroded by the advancing break-
er zone, silt and clay were carried into deep
water below wave-base, fine sand was swept land-
ward by the prevailing waves and wind, and gravel
was left behind as a lag. When sea level reached
about -80 metres, three small spiral-shaped bays
107
were formed: Cape Recife-Riy Bank, Riy Bank-
Bird Island, and Bird Island-Cape Padrone. A
further rise in sea level to -10 metres caused
submer,gence of Riy Bank and isolation of Bird
Island, and the three small beaches were coales-
ced into a single large oneo Shoreline erosion
progressed rapidly with substantial transfer of
sediment taking place in the littoral zone.
About 5Ka ago, when sea level reached its present
height (Clark et aI, 1977), the Algoa Bay shore-
line assumed a log-spiral shape of rather shallow
dimensions. Continued erosion of the shoreline,
with concomitant alongshore transfer of sand
eventually gave rise to the coastal configuration
existent today.
Evidence in support of an equilibrium con-
dition for the Algoa Bay shoreline is provided
by current measurements undertaken over a 5-year
period between Port Elizabeth and Bird Island.
These include a study of 118 ship's drift ob-
servations in which 50% of the currents were
slack, 24% were east-going and 26% west-going
(Harris, 1978, Table 23).
6.3. Deviations from log-spirality
Two beach sections have been examined.
Data for this purpose were obtained by tracing
the shoreline from SAN Chart 126 (scale
1:150 000), and marking points at 1 cm intervals
along its length. The one section, Cape Recife
to Cape Padrone was defined by points 1 to 58
and the other, Port Elizabeth to Woody Cape, by
points 8 to 49 (Table 1). Data points determined
for the best-fitting log-spiral curves to both
beach sections are shown in Figure 8.
The first spiral curve (points 1 to 58) de-
viates markedly from the actual shoreline because
of geographic deformities at the distal ends of
the bay. From Port Elizabeth to Cape Recife,
convexity of the rocky shoreline has forced the
best-fitting spiral onshore in this region, and
offshore between Port Elizabeth and the Sundays
River. Similarly, convexity of the coastal
foreland at Woody Cape has forced the spiral
 o
m

... I ---.~
~A • •
.-a-.- . . . . . .
~ 44

..... ....."" .......... j .. / ........' .........• 'V'\~.~~:..___

........:: .... .

....~t.~~X··,...... .. . S ..... , ..,S•••..-......, ••••,•." ..,••., •.,• ..... , ~/ • ..(' ...........;......... ..

..'I'i"renton Island ./ .rfo \ -::;' ._Bird "Is~~~cf

(/// !(:'::·:,r>'··:'
ALGOA BA Y
<:::.: ..
/\ ..... \ ....::; ............

:iJ}":·~L·i.··iRiY Bank ..........

• ..~spf;~'
~ . ,..., ?entres :... .. ••••.•:
(:: :: ..........
/ ........'" ..... , "." ......." • Cape Recife - Cape Padrone
4 Port Elizabeth - Woody Cape
.....
........ ......[. )
, ........i.
.... r·/· ,· · · ,
FIGURE 8. Planimetric shape of Algoa Bay and two best-fitting log-spiral curves. Circles represent the section Cape Recife to
Cape Padrone (points 1 to 58) and triangles, Port Elizabeth to Woody Cape (points 8 to 49).
 109

onshore between the Sundays River and Woody Cape, seen is a slight progradation of the shoreline

and offshore at Cape Padrone. between the Coega and Sundays Rivers (Fig.8),
and simply reflects the energy lost by waves
The second curve (points 8 to 49), with
being diffracted around the island triad of
minor deviations, is a remarkably good fit to
St Croix - Brenton - Jahleel.
the actual shoreline. This is well demonstrated
by a semi-log plot of the actual data-points and Another departure of the coastline from log-

a regression line representing the best-fitting spirality can be seen as a retrogradation from a

log-spiral curve (Fig.9). One of the deviations point east of the Sundays River to woody Cape.

60
50
40
ALGOA BAY
30 Cape Recife to Cape Padrone
(points 1 to 58)
20

S
Ig
II<:
0
~ ••
U
w 8 • •
> 7
•
.
Vl 6
~
is 5 III to
ct
II<: 4 • "'0 9~
6e
3
•
~

6380
2
•
•
1
0 20 40 60 80 100 120 140 160 180 200 220 240 260
ANGLE BETWEEN RADIUS VECTOR AND AZIMUTH (e)

50

40
ALGOA BAY
Port Elizabeth to Woody Cape
30 (points 8 to 49)
~
II<:
0
~
U
w 20
>
Vl
~
is
ct
II<:

10
9
8
70 20 40 60 80 100 120 140 160
ANGLE BETWEEN RADIUS VECTOR AND AZIMUTH (e)

FIGURE 9. Relationship between supplied data points, and regression lines for the best-fitting
log-spiral curves.
110

The reason for this became evident from an
examination of the local bathymetry (Bremner,
1980) of which only the 30 and 50 metre isobaths
are shown in Fig.8. The percentage difference
between theoretical and actual radius vector
lengths, as listed in the computer output, were
plotted next to the distance of isobaths from
shore for each supplied data-point (Fig.10).
is clear from this diagram that the sea floor is
embayed precisely where the coastline is retro-
grade behind the log-spiral curve.
high energy are therefore able to impinge direct-
lyon the coast in this region, and erosion has
progressed beyond where it would have, had the
isobaths followed smooth spiral-curves.
bathymetry must therefore be an artefact reflect-
ing erosion during a period of lower sea level.
Evidence attesting to this is obtained from on-
shore drilling by the Department of Water Affairs.
These results indicate a broad basement depress-
ion, now in filled with Caenozoic sediments, which
may have been the original course of the Boesmans
River (M. Marker, personal communication).
6.4. Artificial modification of the shoreline
Adjustment of the shoreline to artificial
changes in the natural environment is of major
It
concern to coastal engineers, property developers,
city councils, and so forth. So, as an exercise
to investigate the response of the shoreline to
Waves of
certain environmental modifications, three some-
what drastic measures were considered without
regard to their cost (Fig.11).
The first of these measures involves de-
The
struction of Bird Island, which is made of Table
Mountain Group quartzite, and measures approxi-
mately 1 square kilometre in area. The coastal
foreland at Woody Cape, consisting of calcaren-

100 Sundays River Bird Island

80

p::~!:~geth~~~:;::"a~e60
radius vector (Pl
and actual radius
vector (rl 40

20

Isobath distance
from shore (kml

FIGURE 10. Correspondence between excessive coastal erosion, ie., landward

of the best-fitting log-spiral curve, and embayment of the bathymetry
between Sundays River and Bird Island. See text for details.
 111

-~------

....
--- -
......
.......
--..-.....~ . . . =-=-...-___ ~rone
Cape

... : .................;........... .
... ..
.........

.. .....
',., "
Algoa Bay
Papkuiis
River

Baakens
River centres

FIGURE 11. Equilibrium shoreline curves generated from changes to the natural environment:
1. Destruction of Bird Island, 2. Construction of a sea wall between Woody Cape and Bird Island,
and 3. Construction of a groyne southeastwards from Cape Recife.

ites and pebble-beds of the Alexandria Formation,
would erode back rapidly until more resistant
shale and quartzite beds of the Bokkeveld Group
located at Cape Padrone became the down-coast
headland (curve 1). Material thus eroded would
be transported out of Algoa Bay by eastward-
moving littoral currents to an area beyond Cape
Padrone.
less than 10 metres in depth (Bremner, 1980).
The resulting shoreline (curve 3) differs negli-
gibly from the present one excepting that the
beach immediately north of Port Elizabeth harbour
would prograde by about 1 kilometre. It is
interesting to note that the other two measures
would have the same effect as well (curves 1
and 2).

The next hypothetical change to the environ- 7. CONCLUSIONS

ment involves the construction of an 8-kilometre
sea wall linking Woody Cape to Bird Island.
submerged tombolo in the lee of Bird Island is
dissected midway by a channel 22 metres deep at
its shallowest point, but for the most part,
varies between -15 and -20 metres depth
(Bremner 1980). The stable position of the shore-
line would then be far seaward of its present
location (curve 2), but progradation would not
be able to ensue without artificially injecting
vast amounts of sediment, say from the adjacent
coastal-dune field, into the marine environment.
The last change calls for a southeastward
extension of Cape Recife by a 2-kilometre long
groyne, which at its deepest point, would be
It is reasonable to assume that all natural
The beaches considered in the present study are in
equilibrium with the environment, since at least
5Ka have elapsed from the time sea level was
restored to its present position. However,
because no general relationships between a and S,
A
and between IB and S could be discerned in
plots of these variables, one must accept that
beaches, like people, are often quite similar in
shape, but never have exactly the same features.
In this connection, local peculiarities in the
bathymetry, and particularly in the geometry of
the headlands, are aspects that cause consider-
able deviation from theoretical log-spiral
curvature. Moreover, it should be realized that
112
the equilibrium state of a spiral-shaped beach
is not a static situation, but rather, it re-
presents the mean position of a large, finite
number of configurations brought about by changes
in the environmental climate eg., wave approach
angle, erosion-resistance of headlands, etc.
Artificial beaches on the other hand, do
A
show progressive changes in a, a, /B and (cx,
cy) as they develop towards equilibrium.
Reasons for this are that the headlands normally
used are regular-shaped objects; the initial
straight beach and back shore are built of well-
sorted, fine-grained sand; and the bathymetry
is allowed to adjust itself continuously on the
floor of the wave tank as coastal erosion pro-
ceeds. The next practical step in researching
artificial spiral-shaped beaches would be to
run a series of experiments in which a is incre-
mented at say 5° intervals. Then, by plotting
the spiral parameters against a for each equili-
brium beach so obtained, a set of theoretical
curves may be generated to explain fundamental
aspects of spiral beaches in general.
By reason of the lack of coherent relation-
ships between the spiral parameters of natural
beaches and a, Algoa Bay was investigated simply
to locate departures of the attractive beach
curvature from log spirality, and to seek quali-
tative explanations therefor. Two beach sections
were considered. One of them, which excluded the
headlands Cape Recife and Cape Padrone, yielded
an excellent fit to logarithmic spiral curvature,
and thus afforded easy identification and inter-
pretation of departures therefrom. Finally, it
may be concluded that for all intents and pur-
poses, Algoa Bay may be regarded as having reached
equilibrium with its local environment. Even
drastic alteration of the islands and headlands
would cause minimal but beneficial change to the
beach near Port Elizabeth, although the down-
coast effects, in some instances, would be quite
substantiaL I thank the Director, Geological sur-
vey, for permission to work on and publish these
data. . Trans. Geol. Soc. S. Afr., LXXI, 3, 209-234.
REFERENCES
Bascom W (1954) The control of stream outlets
by wave refraction, Jour. Geol. 62, 600-604.
Bremner JM (1970) The geology of Wreck Bay,
Vancouver Island, unpublished MSc. thesis, Univ.
of British Columbia, 243p.
Bremner JM (1980) The bathymetry of Algoa Bay,
Geol. Surv. Rept., no.1980-0007, 13p.
Bremner JM (1981) Geophysical investigations
in the vicinity of Port Elizabeth harbour, Geol.
Surv. Rept. no.1981-0068, lOp.
Bremner JM and LeBlond PH (1974) On the plani-
metric shape of Wreck Bay, Vancouver Island, Jour.
Sed. Petr., 44, 4, 1155-1165.
Clark JA, Farrell WE and Peltier WR (1977)
Global changes in postglacial sea level: A numer-
ical evaluation, Quat. Res., 9, 265-287.
Davies JL (1972) Geographical variation in
coastal development, Oliver and Boyd, Edinburgh,
204p.
De Beaumont LE (1845) Lecons de geologie
pratique, Paris, p.222.
De la Beche HT (1833) A geological manual,
London, p.79.
Gallow MA (1980) Stabilization of coastlines
using crenulate shaped waves, Unpublished BSc.
thesis, Civil Engineering Dept., Univ. of Cape
Town, 106p.
Gourlay MR (1974) Wave set-up and wave gener-
ated currents in the lee of a breakwater or
headlands, Proc. 14th Conf. on Coast. Engineers,
Copenhagen, 1976-1995.
Guilcher A (1969) Pleistocene and Holocene sea
level changes, Earth-Sci. Rev., 5, 69-97.
Gulliver FP (1899) Shoreland topography, Proc.
Am. Acad. Arts & Sci., XXXIV, p.180.
Halligan GH (1906) Sand movement on the New
South Wales coast, Proc. Limn. Soc., N.S.W.,
31, 619-640.
Harris TFW (1978) Review of coastal currents
in southern African waters, S. Afr. Nat. Sci.
Progr. Rept. 30, C.S.I.R., 103p.
Komar PD (1971) The mechanics of sand transport
on beaches, Jour. Geophys. Res., 76, 713-721.
LeBlond PH (1972) On the formation of spiral
beaches, Proc. 13th Conf. on Coast. Engineering,
Vancouver, 1331-1345.
LeBlond PH (1979) An explanation of the log-
arithmic spiral plan shape of headland-bay
beaches, Jour. Sed. Petr., 49, 4, 1093-1100.
Lewis WV (1938) The evolution of shoreline
curves, Proc. Geol. Ass., 49, 107-127.
Longuet-Higgins MS (1970) Longshore currents
generated by obliquely incident sea waves (1 and
2), Jour. Geoph. Res., 75, 6778-6801.
McLean R (1967) Plan shape and orientation of
beaches along the east coast, South Island,
N.Z. Geographer, 23, 1, 16-22.
Rea CC and Komar PD (1975) Computer simulation
models of a hooked beach shoreline configuration,
Jour. Sed. Petr., 45, 4,866-672.
Ruddock A (1968) Cainozoic sea-levels and
diastrophism in a region bordering Algoa Bay,
 113

Silvester R (1960) Stabilization of sediment-

ary coastlines, Nature, 188, 4749, 467-469.
Silvester R (1970) Growth of crenulate shaped
bays to equilibrium, Jour. Waterways and Harbors
Div., Proc. Am. Soc. Civ. Engineering, 96, WW2,
275-287.
Silvester Rand Ho S-K (1972) Use of crenulate
shaped bays to stabilize coasts, Proc. 13th
Conf. on Coast. Engineers, Vancouver, 1347-1365.
Tanner WF (1976) Discussion: on the planimet-
ric shape of Wreck Bay, Vancouver Island: by
Bremner JM, LeBlond PH, Jour. Sed. Petr., 46,
1, 258-259.
Walton TLJr (1977) Equilibrium shore and
coastal design, Coastal Sediments '77, Am. Soc.
Civ. Engineers, Charleston, 1-16.
Wheeler WH (1902) The sea coast, London,
p.199.
Winter HdelaR (1971) Geology of the Algoa
Basin, South Africa, in: Sedimentary basins of
the African coasts, 2nd part, Ed:Blant G,
I.U.G.S. Comm. for Mar. Geol., 17-48.
Yasso WE (1964a) Plan geometry of headland-
bay beaches, Office of Naval Res., Proj.
NR388-057, Tech. Rept. 7, p.30.
Yasso WE (1964b) Geometry and development of
spitbar shorelines at Horseshoe Cove, Sandy
Hook, New Jersey, Office of Naval Res., Proj.
NR388-057, Tech. Rept. 5, 104p.

BEACH AND NEARSHORE HABITATS AS A FUNCTION OF INTERNAL GEOHETRY, PRH1ARY SEDH1ENTARY STRUCTURES AND
GRAIN SIZE
B.W. FLEMl1ING and A.H. FRICKE (National Research Institute for Oceanology, Council for Scientific
and Industrial Research, P.O. Box 320, Stellenbosch 7700, South Africa).
1. INTRODUCTION
The role of grain size in soft-substrate ecology
has been the subject of much debate. Although
a remarkable correlation has been found in some
cases between individual organisms or even
whole faunal assemblages and the physical
nature of the substrate, the results have in
general not been conclusive. Thus, Jansson
(1971) argues that the importance of grain size
has been over-estimated to the disadvantage of
supposedly more relevant physico-chemical
parameters such as sediment stability, pore
space, light, oxygen availability and temper-
ature, or biological factors such as competition
for food and predation (Gray, 1981). This
view is supported by some studies which have
failed to detect any grain size effects whatso-
ever (e.g. Schmidt, Westheide, 1972; Ravenel,
Thistle, 1981). On the other hand, it has
been demonstrated that some organisms do show
a high preference for specific substrate types
(e.g. Hennig et al., 1982), whereas others
appear to be more cosmopolitan (Johnson, 1971).
Furthermore, it is a well established fact that
many animals are morphologically adapted
either to ingest sediment particles on a size~
selective basis (e.g. Fenchel et al., 19]5;
Deutsch et al., 19]7), or to live in the pore
spaces between sediment grains (e.g. Swedmark,
1964). In both cases textural parameters such
as mean grain size and sediment sorting are
clearly major controlling factors for habitat
selection (cf. Headows, Campbell, 1972).
115
One of the main problems in this debate, it would
seem, is a methodological one. A major con-
straint in quantitative soft-substrate ecological
studies is the requirement of representative
sample sizes (Elliott, 1971). Depending on the
size and the popUlation density of an organism,
the minimum sample volume can vary from several
1000 cm 3 to less than 100 cm 3 of sediment. In
all cases attempts have been made to standardize
the sample size in three dimensions. For example,
characteristic surface areas are 1000 cm 2 for
macrofauna and 35-40 cm2 for meiofauna (Gray,
1981), whereas sampling depths range from a few
centimetres to several decimetres. It is impor-
tant to note, however, that the substrate itself
is rarely homogenous, even though it may appear
so at first sight. This is particularly true
for sandy beach and nearshore environments,whereby
the ecological importance of structural and text-
ural inhomogeneities increases with decreasing
size of the organisms. Unfortunately, not many
biologists have bothered to take a closer look
at the structure of the natural habitat before
embarking on detailed investigations. As will
be demonstrated in the course of this paper,
the microstructures of most depositional environ-
ments are generally several orders of magnitude
smaller than even the smallest sample sizes
required for most quantitative biological analyses.
This methodological problem can be partly over-
come by combining new sampling strategies with
'in situ' preference experiments. The importance
of such experiments has been emphasized by Gray
116

(1981), although very few such studies have in
fact been carried out (e.g. Boaden, 1962; Renaud-
Debyser, 1963; Gray, 1966,1967; Jansson, 1967;
Hicks, 1977; Ravenel, Thistle, 1981). The
present paper makes a renewed attempt at tackling
this problem of organism sediment relationships.
By discussing the effects of physical parameters
characterizing sandy beach and nearshore environ-
ments (especially their morphodynamics and inter-
nal sedimentary structures) on macro, meso and
microscales, the beach habitat is placed into its
proper ecological perspective. Emphasis is
placed on appropriate sampling methodologies by
illuminating the pitfalls of some conventional
approaches, in particular with respect to struc-
tural inhomogeneities and the measure and inter-
pretation of grain size. The points addressed in
this presentation are strongly underlined by the
results of a concurrent study (Fricke, Flemming,
this volume) which deals with a carefully desig-
ned ' in situ' sediment preference experiment.
2. lfORPHODYNAHIC BEACH CLASSIFICATION
Beaches have for a long time eluded a quanti-
tative description, although it has been evident
that they responded quite sensitively to changes
in environmental conditions, especially the
local wave climate (Bascom, 1964; Hayes, 1972).
For example, a systematic relationship was
observed between the grain size of a beach
and its slope as a function of exposure to wave
action (cf. Bascom, 1951; Inman et al., 1963;
Wiegel, 1964). Basically, there is a negative
correlation between grain size measured in phi
or wave action and beach slope measured indegrees.
Thus, for any given sand size a beach slope will
always increase with decreasing wave energy (cf.
Fig. 1). Alternatively, at similar energy
levels a coarser sand will always form a steeper
slope than a finer sand. This relationship is
empirically quantified in Table 1.

Table 1. Empirical formulae for estimating beach slope as a function of grain size and wave
energy and vice versa as illustrated graphically in Figure 1.

Low Wave Energy 3.l7S- 2 . l6 (3.l7/D h.)0.463

or S
p l

Intermediate Wave Energy 2.72S- 2 . 04 or S (2.72/D h.)0.49

p l

High Wave Energy 2.25S- 2 . 0 or S (2.25/D h.)0.5

p l

S beach slope in degrees; mean grain size in phi

 117

Host beach sediments are relatively well size-
sorted and major differences in grain size mostly
reflect corresponding differences in relative
energy levels (cf. Bascom, 1951). Of course
there are exceptions to this rule; for example,
when a localized source supplies coarse material
to a sheltered beach which otherwise consists
of finer sediments or when an exposed beach
receives a surplus of fine sediment (cf. Bird,
this volume).
In addition to this size/slope relationship it
has been established that beach profiles undergo
periodic modifications related to changes in the
seasonal wave climate (e.g. Bascom,1964; King,
1972; Sonu, 1973). A beach is thus a highly
dynamic system which constantly responds to
short and long term fluctuations in energy levels.
One of the shortest cycles of this nature is
induced by the rise and fall of the tide. Guza,
Inman (1975) have described this cycle in terms
of the associated surf zone dynamics by intro-
ducing the terms dissipative and reflective
beaches (d. Fig. 2).

1:250 1:100 1:50 1:25 1:10 1:5 1:2.5 1:1

4

very fine sand

--
~
Q.
3 - - - - - - - - - . . - - - - - - - - - + 0.125 E -
-E
fine sand
w W
N N
(/)
2 .~~~~~~r_------_r----------------rO.25 (/)
z Z
ct c:(
0:: medium sand
C) 0::
C)

Z 1 -------------+ 0.5 Z
c:( c:(
w w
:E coarse sand :E

very coarse sand

-1~----_,r_----~----~--_,r_~~T_rTTT------~--~~~2
0.25 0.5 1 2 4 6 8 10 20 30 45

BEACH SLOPE (degrees)

FIGURE 1. The relationship between wave energy, grain size and beach slope
(recalculated and rearranged after Wiegel, 1964).
118

.... ..
.....
.~5' ::':.....
:':'. +5
.....

o
.::;i\ f0,i!(i iiii ~ ;,.:i i : : ...: ..... : ..,.:,.{;::.... : .:.......- ......."'-...'~.......~-.......--.......- -. ,. " '- ~'....- .....--.......,.
berm .... ::~: :':::':.::.: :'.::.::.:.:.:~. ~. '. '. '. '..
-5 . :... ....
.. ......:... :"....... :.....~ ':..:;;::.::-:-:.;..:. . -5
Inner
trough
..... .: ......
: ...... ~ .............
.............
. ' ........... ......:... ...... -
'.~:
:...... ---
-:-:-. ........
---..~-~
.
......:.. :.:... ......... .
Inner .......... ~:: ..:": ..........:.:....:::::~.~~ ~:-':-, ....~_-..I
-10m bar ......... ::...........: ..::.... :.:-... :-:.:-::::::.. ::::::.:
outer outer
A. trough bar

.....
' ..
+5

':::. -"'"
o

-5 -5
overwash
-10m barf berm
no nearshore bars ..... : ..: ..::..: ......: ..
B.

FIGURE 2. Morphodynamic beach stages. A. Beach and nearshore profile of the high-energy,
dissipative case. B. Beach and nearshore profile of the low-energy, reflective case.

All beach systems constantly adjust between more To distinguish between totally reflective
dissipative and more reflective states. The terms and strongly dissipative wave types these
dissipative and reflective beach profiles are authors have used the surf scaling or
largely synonomous with the more conceptual terms reflectivity parameter E = a.w 2 /gtan 2 S
~

winter and summer profiles (cf. Bascom, 1964).
The work of Sonu (1973) and Guza, Inman (1975)
represent the first steps towards a more quanti-
tative classification of beaches.
has recently been significantly expanded by
Australian scientists, who have identified succ-
essive morphodynamic stages in the process of ad-
justment between the completely dissipative and
totally reflective extremes (e.g.
1979 b, 1981; Wright, 1981, 1982;
1981,1982; Wright et al., 1978,1979,1982).
where a i is the wave amplitude near the
breakpoint, w = 2rr/T where T is the wave
period, g is the acceleration due to gravity
and S is the beach or inshore slope (cf.
Their approach Guza, Bowen, 1975). For total reflectivity
E must be <1, whereas highly dissipative
waves have values of E<33 (cf. Wright et
al., 1979).
Short, 1979 a, Hore recently this morphodynamic classifica-
Wright, tion of beach stage (Bs) has been reduced to
a simple empirical relationship between
 119

breaker height (~), wave period (T) and i.f a sufficiently long record of local wave
the settling velocity of the mean grain size climate were combined with a suitable long~

in m/sec (W s ) such that Bs = ~. / WsT. In shore transport model.

this classification scheme reflective beaches
characteristically score values <1, whereas A different field method for estimating the
dissipative beaches score <6 (Wright, Short, relative exposure of beaches has been proposed
1982). Various intermediate beach stages by McLachlan (1980 a). By rating a number
can be distinguished (Table 2 and Short, of easily determined parameters such as an
this volume). Studies along similar lines estimate of maximum wave height, surfzone
have been carried out by Greenwood, Davidson- width, % very .fine sand, median particle
Arnott (1979), Chappel, Eliot (1979), Gold- diameter, depth of the reducing layer and the
smith et al., (1982) and Bowman, Goldsmith .presence or absence of s table burrows, he
(1982). This work is still in progress and achieves scores between 1 and 20. A score of
will require further quantitative refinements, 1, for example, would represent the very
especially as non-barred, high~energy beaches sheltered beach type, whereas a score of 20
have been described in the literature which, would suggest a very exposed beach. In
at first sight, do not appear to fit the above the absence of more quantitative, process-
morphodynamic model (eg. Clifton et al., 1971; related parameters thi.s approach would appear
Hawley, 1982). A recent study by Aubrey et to be a useful descriptive alternative,
al., (1980) suggests that average weekly although it does not at the same time provide a
beach changes could be successfully predicted rational morphodynamic classification.

. TABLE 2. Morphodynamic beach classification (after Wright, Short, 1982)

Beach Dynamic Wave Beach

stage state energy profile Morphodynamic Beach stage

1 strongly low steep high berm with convex beach face

reflective (H<l m) but smooth

i i i
i
2 welded bar, often with ridge and runnel systems

r r
~ ~ ~
and beach cusps

3 intermediate ~ ~ transverse bars, or anvil-shaped attached

crescentic bars, low-tide terraces with irreg-

1 1 1
ularly spaced rip channels, incipient megacusps

4 ~ ~ ~ transverse bars with evenly spaced rips or un-

1 1 1
attached crescentic bars, megacusps

5 ~ ~ transverse to linear bars with widely spaced

!
~

6
!
strongly high gentle
! rips, cusps

single or multiple parallel bars, sometimes

dissipative (H >3 m) but rhythmic very widely spaced rips
120
3. PROFILE GEOMETRY AND INTERNAL STRUCTURES
From the above discussion it is evident that
the scale and rate of morphodynamic adjustment
of a beach must have a profound effect on the
habitat to which organisms living in the beach
environment have to adapt if they are to
survive. The identification of morpho-
dynamic beach stages, together with their
temporal and spatial variability, is an
important prerequisite for the causal inter-
pretation of beach profile geometry and
related internal sedimentary structures on
various scales (cf. Hawley, 1982). This in
turn is one of the most essential physical
habitat parameters in the study of organism/
sediment relationships.
Internal sedimentary structures are reliable
indicators for depositional environments
and facies (cf. Teichert, 1958; Reading,
1978; Walker, 1979; Reineck, Singh, 1980;
Collinson, Thompson, 1982). In addition,
a large amount of specialized literature has
accumulated which deals with specific
physical and/or biological phenomena
associated with depositional processes under
different environmental conditions, thus
documenting the great diversity of poten-
tial microhabitats available to various
organisms. Summaries of this literature
are given by Schafer (1972), Imbrie, Newell
(1964), Frey (1975), Basan (1978), Reineck,
Sing (1980), Allen (1982). A'brief out-
line of major macro, meso and microstructures
will serve to illustrate the diversity of
habitats found in the beach and nearshore
environment.
3.1 Macrostructures
Macrostructures are here defined as those
structures which characterize specific
morphological units of the beach profile. They
are usually measured in mUltiples of metres and
are best illustrated by a number of typical
cross-sections. Following the morphodynamic
beach classification, the examples include a
parallel-barred or dissipative beach stage
(Fig. 3A), an oblique-barred or intermediate
beach stage (Fig.3B) and a non-barred or
reflective beach stage (Fig. 3C). In each case
the observed facies zonation is illustrated by
cross-sections which also show schematically
the most characteristic internal structures
recorded along the profile. The great variety
of facies-specific structures is immediately
apparent and it will be shown below that this
situation can be further enhanced by variations
in sediment texture.
In the parallel-barred, dissipative case the
sequence of distinct facies and subfacies,
commencing in the offshore, can be summarized
as follows
a nearshore/offshore transitional facies is
followed by a seaward outer-bar slope, an outer-
bar crest, a landward outer-bar slope, an outer
trough, a seaward inner-bar slope, an inner-bar
crest, a landward inner-bar slope, an inner
trough, a swash/trough transitional facies, a
swash zone and a high berm. The oblique-barred
case is in many respects similar, except that
there is now a more pronounced rip-channel facies
and a lateral rhythmic succession of repetitive
cycles. The high-energy, non-barred situation,
on the other hand, is very different. The near-
shore/offshore transitional facies is followed
by an outer rough zone consisting of lunate mega-
ripples, an outer planar zone, an inner rough
zone, an inner planar zone (swash zone) and
finally a high berm. In addition there can be
symmetrical and asymmetrical wave ripples in
most of the zones, whereas small standing waves,
 121

DISSIPATIVE BEACH STAGE

-'IIi1_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ m_8_a_"_S_8_a_-1_8_ve_l_______

SEAWARD IEAWARD
BERM INNER-BAR SLOPE OUTER-BAR SLOPE OUTER-BAR SLOPE

SWASH INNER INNER-BAR OUTER OUTER-BAR

ZONE TROUGH CREST TROUGH CREST OFFSHORE
TRANSITIONAL

FIGURE 3A. Morphodynamic units and associated internal sedimentary structures of a high-energy,
dissipative beach system (modified after Davidson-Arnott and Greenwood, 1976).

INTERMEDIATE BEACH STAGE

mea" sea-level

SEAWARD
ZONE SLOPE BAR SLOPE

BACKSHORE RIDGE SWASH-TROUGH RIP-CHANNEL BAR CREST NEARSHORE-

(RUNNEL) CREST TRANSITIONAL OFFSHORE
TRANSITIONAL

FIGURE 3B. Morphodynamic units and associated internal sedimentary structures of an; intermediate
energy, oblique-barred beach system (modified after Hunter et aZ., 1979).
122

REFLECTIVE BEACH STAGE

mean sea-level
.:.:::~~-----------------

fORE DUNE SWASH INNER ROUGH OUTER OFFSHORE ZONES

ZONE ZONE ZONE

SWASH-BACKWASH OUTER NEARSHORE-OFFSHORE

BERM
TRANSITIONAL PLANAR TRANSITIONAL
(INNER PLANAR) ZONE

FIGURE 3C. Morphodynamic units and associated internal sedimentary structures of a non-barred,
reflective beach system (modified after Clifton et al., 1971.)

antidunes and rhomboidal back-wash patterns
commonly occur in the swash zone. Furthermore,
at the foot of the backwash there is usually
a well-defined step which concentrates the
coarsest material found on the beach and
which migrates across the intertidal zone
in the course of each tidal cycle.
The above examples were chosen simply to
illustrate the morphological and structural
variability and complexity of sandy beach
and nearshore environments.
Numerous other examples could be presented
in similar manner, each representing a
specific morphodynamic beach stage between
the dissipative and reflective extremes found
in different coastal settings and under
different environmental conditions. This
enormous variability emphasises the fact
that organisms living in sandy beaches or
nearshore environments commonly have a
wide choice of distinctly different and
often highly dynamic macro scale habitats
(cf. Sanders, 1958; Mills, 1969; Rhoads,
Young, 1970; Gray, 1974; Komar, 1976;
McLachlan, 1980 b; Vilas et al., 1982).
3.2 Mesostructures
Mesostructures are here defined as specific
sets of internal sedimentary structures that
characterize a particular facies or sub-
facies of a depositional environment. They
form the three-dimensional fabric of the
macrofaunal habitat and are commonly measured
in centimetres or decimetres. Most hydro-
dynamic processes can be recognized by the
unique sets of sedimentary structures which
they produce. The variability of beach and
nearshore environments suggests the presence
of a corresponding variety of process-response
mechanisms which operate on a number of differ-
ent scales. As a result most macro-structures
are built up by more than one set of mesostruc-
 123

un it
of a pa rt ic ul ar
tU re s. The fre qu en cy as ur e
LA N D SEA
ve s some me
os s- se ct io n gi
ob se rv ed in cr a pa rt ic ul ar A
re la tiv e fre qu en cy at wh ich
of th e Th is is we ll
me ch an ism oc cu rs .
on se ... ... . . . . . ... ..'" . .
pr oc es s- re
at ed
sp
in th e sc he m at ic
bo x mo de ls of
Fi g.
. ::..::. :.::::::.' . . : : ... .... ..
'

::::'.:":,:::" : :....: :.....'.'.................. .......

ill us tr ...... .. .. . '" .
3A, Ba nd C. .. "" "" ... ::
. . ...
...
.. :.."" ........ .
'" " '

. ...
to
".
. ... .
w ill su ff ic e .. : .. "". . .. . "
e fu rt he r di sc us si on it
.. : : :. '.': : :: ':: ....... .
............
..... ..'. :....'.... ..............
Fo r th de r to de mo n-
si ng le fa ci es in or ........ "
a
.........
'
co nc en tra te on '
sm s an d
be tw ee n or ga ni
ra te sU bt le re la tio ns hi ps n
st nc lu sio ns dr aw
sU bs tra te . The ge ne ra l co B
th e to al l ot he r
ex am pl e ap pl y more or le ss
fro m th is th e
th re sp ec t to ... ... ... ..
pa rt ic ul ar wi .. ...
. .........
"
si tu at io ns , in is pa rt ic u- ... ... .. .. ........ : ... ........ .
in te rs ti ti al fa un a. The sw as h zo ne
is th e .. '::'::'.': .. .. .. ..
pu rp os e as it ...................... .
la rl y we ll su
ite d fo r th is
. . . . : : ... . .. .. . .. ....
.............................
.. . ::: ': :: :: ::..•................".-0'.'•
..

..
'
th e
ac ce ss ib le an d be st st
ud ie d pa rt of
..- ........... . '. ,

......... .. .... ...... : .. . : ..

mo st
os ys te m .
wi de r be ac h ec
................... .
.

. .. .. .. .. .. .. .. .. .
..
.. . ........ ........ .............. ..
in de po si tio na l un its ob
se rv ed in th e
.. .. .. .. .. .. .. .. .. .. ... ............
The ma
sw as h zo ne , to
ge th er wi th so
me common di p
4. In di vi du
di re c-
al
c
tr at ed in Fi g.
tio ns ar e ill us
or lan dw ar ds in
ei th er se aw ar ds ......
un its ca n di p th e
de pe nd in g on
g se qu en ce s, ............. .. . ..
on or of fla pp in .
tid al cy cl e an
d th e lo ca l wa ve
:' :: :: :. '.. .. .. .. .. .. .. " ........
.. ..
.
st ag e of th e
re de po si te d
wh ich th ey we ......
. ............
....

. . .
......."......... .
.... ""

............ ......
"

cl im at e un de r ......
""

.......
..

.......
. ... .........
"oJ

....."........
""

. .. . ..
.. ..
75 ;
.. ..

.. .. ..'..
..

69 ; Ha rm s, 19 .. ...... .'
.. ..
; Cl ift on , 19 .
(Thompson, 19 37 .. .. .. . ...
. '". .
'"
..
e in di vi al
du .. .. .. .. ... ..~ ...
'"

In al l ca se s th .. .. .. .. .. .. .......' .....
.....
............... ...
.......
.. .........
.. .. . ......
). o ..... o

19 81 ... .. .. .. ..
.
....

,
... ... ... ... ... .. ...
er
• 0

Sa lle ng es .
gl e di sc or da nc
ra te d by lo w- an
un its ar e se pa
re pr es en ts a pa
rt ic ul ar de po - o
EVery se t th us sp ec if ic co mb
in -
ci at ed wi th a .. .. .. .. .. .. .
si tio na l ph as e as so
pl ai ns .. .. .. .. .. .. .. .. ..
on s. Th is ex
nm en ta l co nd iti
at io n of en vi ro in th e ve rt ic al
ch va ri ab ili ty
why th er e is Su . Th is st ru c-
en ce of de po si tio na l un its
se qu d by
he r ac ce nt ua te
ty ca n be fu rt
tu ra l va ri ab ili ja ce nt se ts Th. is
al di ff er en ce s be tw ee n ad
te xt ur re li ef ca st s
ed in Fi g. 5, wh er e th re e
is ill us tr at
co mp ar ed . The
t be aC he s ar e
fro m di ff er en
st s ha s be en
th od s fo r pr ep ar in g SUch ca
me a (1 96 9) . rv ed in
tre at ed by Boum in g ty pe s ob se
ex ha us tiv el y FIGURE 4. The ma in be dd ev en la m in at io n
ts. No te th e
be ac h se di m en es (m od ifi ed
lO w- an gl e di sc or da nc
comes fro m a te d by
se parraTh ompson, 19 37 ).
pl e (F ig . 5A ) af te
The fi rs t ex am
la rg e cr en u-
ca lit y al on g a
se m i-e xp os ed lo e
h A fr ic a) , Th
te ba y (J ef fr ey s Ba y, So ut
la
124
beach slope is approximately 1:50 (1.4 0 ) and
the mean diameter of the sediment is about
2.5 phi (0.18 mm), which is consistent with
the predicted grain-size/beach-slope
relationship illustrated in Fig. 1. At the
sampling site the beach sediment appeared
quite uniform and homogenous, both horizon-
tally as well as vertically. The only
exception was a thin shell layer at a depth
of 15 _cm. No obvious stratification other
than the shell layer was visible in the
field. The epoxy-impregnated relief cast,
on the other hand, reveals a number of
previously undetected features. Firstly,
the sediment column is finely laminated
throughout and secondly, the fine sand is
not entirely homogenous or uniformly
distributed. This is suggested by the
uneven relief of the cast a feature that
will be specifically addressed in the next
section.
The second example (Fig. 5B) also comes from
a semi-exposed beach (Arniston, South
Africa), but this time comprises a con-
siderably coarser material than in the
previous case. The mean grain size of
this sediment is about 1.1 phi (0.45 mm) and
as a result the beach slope is considerably
steeper (~80). This too is consistent with
the model. Furthermore, the entire cast
consists of uniformly seaward dipping
strata which also show slight-inhomogeneities
in their vertical succession.
The third example (Fig. 5C) was chosen to
illustrate the effect of textural variability
which may occur in beach sediments (Blouberg
Strand, South Africa). In this case finer
and coarser layers alternate at irregular in-
tervals, ranging from a few centimetres to a
decimetre or more. The differences in porosity
between these layers are now much more pronounced.
Also note that the slopes of the coarser layers
can vary considerably. Similar slopes for both
coarse and fine laminae suggest major differences
in the energy level under which each one was
deposited (cf. Fig.1). Different slopes, on the
other hand, may simply reflect a change in source
material under otherwise similar energy conditions,
e.g. in the course of the rising and falling tide.
The above examples give some idea of the varia-
bility of beach sediments, both at a single
locality as well as laterally along the shore-
line. For example, dip angles of beach laminae
can differ significantly, depending on the local
wave climate, even when the sediment remains
relatively uniform. This in turn affects the
permeability and hence the filtering capacity of
seawater which plays an important role in supply-
ing nutrients into the beach system (cf. Riedl
et a1., 1972; McLachlan, 1980 c; Lanyon et al.,
1982). In addition, the nature of various
internal structures reveals a great deal about
the depositional history of a particular localit~
including the hydrodynamic conditions under which
they formed. It is thus clear that the variability
of internal structures and grain size as a
function of morphodynamic beach stage and positilln
both perpendicular as well as laterally along the
shoreline requires close attention in the study
of beach ecosystems.
3.3 Microstructures
lficrostructures are here defined as those struc-
tures that are visible down to scales of a few
grain diameters. They are commonly measured in
millimetres or centimetres and include features
such as the composition and orientation of individ-
ual laminae, grain size variations within and
between layers, grain shape effects, grain orien-
tation, grain packing, as well as porosity and
permeability. At the same time the micro-
structures define the physical dimensions of
the meiofaunal and microfaunal habitats.
 5cm
FIGURE SA. Internal sedimentary structures
observed in a semi-exposed beach consisting
of fine sediment. Note the low relief bet-
ween individual laminae and the concentra-
tion and alignment of shell fragments.
5c~
FIGURE SB. Internal sedimentary structures
observed in a semi-exposed beach consisting
of medium sand. Note the high relief
between individual laminae and the alignment
of shell fragments.
5cm
FIGURE SC. Internal sedimentary struc-
tures observed in an exposed beach
consisting of a wide spectrum of grain
sizes. Note the extremely high relief
and the excellent preservation of
individual laminae.
N
In
126
As pointed out above, beach sediments are
neither homogenous nor uniform in their
distribution, even at relatively small
scales. Coarser and finer layers alter-
nate with each other and are usually
separated by relatively sharp boundaries.
This is illustrated in Fig. 6A, Band C.
The images represent strongly magnified
sections of the relief casts shown in Fig.5.
Differences in the packing and orientation
of sand grains are also apparent, a feature
that is strongly accentuated by grain shape.
Less obvious at first sight are differences
within and between laminae which super-
ficially seem to consist of the same sedi-
ment. That subtle inhomogeneities exist
even within apparently uniform sediments
has already been pointed out. The relief
between adjacent laminae of the cast is the
result of differential penetration of the
epoxy, which suggests small differences
in the porosity and hence permeability o~
the sediment at lamina or even sub lamina
level. This phenomenon has recently been
investigated by Grace et al., (1978),
Emery (1978). Both authors have been able
to demonstrate that differences in mean
size and sorting not only occur between
different laminae, but that size grading
occurs even within individual layers. As
illustrated in Fig.7, intra-lamina grading
appears to be the rule rather than the ex-
ception and maximum observed gradients can
span a whole phi interval (0,28-0,55 mm) over
a total sublamina thickness of only 2 mm.
Furthermore, such intra-lamina grading does
not appear to follow any regular trend which
could be attributed to a simple genetic
mechanism. Such trends, for example upward-
fining cycles, can on the other hand be
observed along vertical profiles across
several laminae (e.g. Basumallik, 1966;
Grace et al., 1978).
These observations are significant because they
demonstrate that the structural and textural
variability observed at macro and meso scales can
be extended down to microlevels as small as a
few grain diameters apart. These microstruc-
tures may be of little consequence to the larger
macrofauna, but with respect to the meio and
microfauna they define the very backbone of the
habitat structure. This observation might in
fact explain why major changes in physico-
chemical properties are observed over vertical
distances of only a few millimetres (Anderson,
11eadows, 1978). Similar observations have been
made by Jansson (1966) and Webb (1969).
4. DISCUSSION AND CONCLUSIONS
In the context of this study, which has looked at
potential relationships between organisms and the
physical nature of the habitat, a number of
important points have emerged. Firstly, it is
now clear that beaches are highly dynamic systems
which constantly adjust to fluctuations in local
energy levels. This applies to all kinds of
beaches, irrespective of the apparent uniformity
or heterogeneity of the substrate. Horphodynamic
adjustment proceeds by local erosion and deposi-
tion of sediment, whereby the scale and the rate
of reworking depends on the overall energy
gradients to which a beach system is exposed.
For the interpretation of biological observations
it is therefore important to recognise the
morphodynamic stage of a beach at the time
of the observation or sampling, as well as
the overall rate and scale of potential adjustment.
The higher the rate of sediment reworking is,
the smaller will be the time span for which a
sample is representative. It would thus seem
that the study of beach and nearshore ecology
requires a closer link with the morphodynamics
of the environment.
 2em 2em 2 em

FIGURE 6A. Hicrostructures observed in FIGURE 6B. Hicrostructures observed in FIGURE 6C. Hicrostructures observed
a semi-exposed beach consisting of fine a semi-exposed beach consisting of medium in an exposed beach consisting of a
sediment. sand with a high shell content. variety of grain sizes.

"
128

LAMINA THICKNESS (mm)

3.5 3 2 4 3 2 2 3.5
2.0 0.25

-
.c
Q.
......
.. '
.'.::
'.
'.'
0.3
,....
E
E
1.5
:~
0.35 ......
w
-
N 0.4 w
N
en en
z 1.0 0.5 Z
ct ct
w w
:E 0.8 :E
0.5 0.7
1 5 10 15 20 25
SAMPLING INTERVALS

FIGURE 7. Example of size-grading observed within individual laminae of beach sands (simplified
after Grace et al., 1978).

Secondly, a detailed analysis of internal result produce corresponding differences in

sedimentary structures has revealed an
enormous variability, both perpendicular
as well as parallel to the shoreline.
This variability can be strongly
accentuated by differences in grain size.
The combined effect of stratification
sequences and grain size varia.tions on the
ecology of sandy beaches is still poorly
understood. For example, it could be one
of the underlying reasons for the observation
that the distribution of individual organ-
isms is inherently patchy. As pointed out
above, size grading and orientation of
beach strata greatly influence the gross
permeability of beach sand and hence its
filtering capacity for seawater.
sediments are generally better filters than
coarse ones and the relative abundance of
various layers in the vertical sediment
column must therefore influence local
nutrient enrichment and exchange processes.
The variability of vertical stratification
sequences along the shoreline should as a
biological responses.
Thirdly, it was shown that structural and
textural inhomogeneities persist even at
scales of a few grain diameters. This is of
particular importance in the study of meio
and microfaunal ecology. Since representative
sample volumes are generally several orders of
magnitude larger than the scale of these micro-
structures, it is inevitable that subtle sedi-
ment/organism relationships are obscured by the
vertical mixing of different layers. The
potential effects of such mixing can be illus-
trated in a strongly simplified, theoretical
model (cf. Fig. 8).
Fine
Let us assume that a beach consists of three
different grain size populations which occur
randomly throughout different stratification
sequences along the shoreline. As shown in
Fig. 8A, Population I consists of coarse to
very coarse sand with a mean diameter of
0.09 phi (0.94 mrn). Population II consists
 129

predominantly of medium sand with a mean diameter MEAN GRAIN SIZE (mm)
of 1.3 phi (0.41 mm) and Population III is made 2 1 ~5 ~25 0.125 0.063
up of fine to very fine sand with a mean diameter
of 2.B phi (0.14 mm). Let us further assume that I :n m
a particular organism has a very high preference
for medium sand, i.e. for Population II. In the
course of a sampling programme four random
samples can yield four different mixtures. To
simplify matters each mixture shall consist of
equal proportions. In the first case (Fig. BB)
Population I is mixed with Population II. We
find our organism well represented in a sediment ~ ~~----~~~~~~~~~~~~~~---r------~~
with a mean diameter of 0.7 phi (0.61 mm). In the
second sample (Fig. Be) the organism is equally
well represented in a sediment which now has a
mean diameter of 2.1 phi (0.23 mm). In the third
sample (Fig. BD) Population I and III are mixed
and as a result, the organism is almost totally
absent. Note that in this case the mean diameter
is virtually identical to the optimum preferred by
the organism. In the fourth case (Fig. BE) we
have mixed all three populations. Again the mean
diameter of 1.5 phi (0.35 mm) is near the optimum
and this time the organism is well represented,
although in somewhat smaller numbers than in cases
one and two.
I +m
The evaluation of this data does not reveal the
true relationship between the organism and the i =1.33
sediment, as it appears to occur over almost the
entire size spectrum of the sediment. Furthermore, [)
~----~------~------r-----~----~~
the fact that in one case the organism was present
at a certain mean grain diameter but not in I +1I + m
another, could lead to speculations that are
totally unfounded. That this experiment is not
only valid in theory was demonstrated in an 'in
situ' sediment preference experiment which made use E
of the same grain size populations discussed above
(cf. Fricke, Flemming, this volume). Other effects
-1 o 1 2 3 4
of sediment mixing on grain size parameters are
MEAN GRAIN SIZE (phi)
discussed in greater detail by Tanner (1964), FIGURE B. The effect of random mixing of beach
laminae consisting of different grain sizes on
Syvitski, Murray (1977) and Walton et al.,
the mean diameter of the sediment sample.
(19BO).
130
Finally, we wish to emphasize that the above
discussion of potential organism/sediment
relationships does not imply that only sedimen-
tary parameters play a role in the benthic
ecology of sandy beaches and nearshore environ-
ments. The true relationships with respect to
different parameters, however, can only be
elucidated if each one is properly understood
within its environmental context. In this
respect a careful analysis of the dynamics and
the resulting physical structures of beach and
nearshore environments would appear to be an
important step towards a better understanding
of organism/sediment relationships.
REFERENCES
Allen JRL (1982 Sedimentary Structures.
Their character and physical basis, Elsevier
Publ. Co. Amsterdam, Vol.I, 594 pp Vol. 11,664 pp.
Anderson JG and Meadows PS (1978) Hicro-
environments in marine sediments, Proc. Roy. Soc.
Edinburgh 76B, 1-16.
Aubrey DG, Inman DL and Winant CD (1980) The
statistical prediction of beach changes in
Southern California, J. Geophys. Res. 85,3264-
3276.
Basan PB (1978) Trace fossil concepts,
SEPM Short Course 5, 201 pp.
Boaden PJS (1962) Colonization of graded
sand by an interstitial fauna, Cahiers de
BioI. Marine 3,245-248.
Bascom WN (1951) The relationship between
sand size and beach face slope, Trans. Am.
Geophys. Union 32, 866-874.
Bascom WN (1964) Waves and Beaches, Doubleday
& Co. New York, 267 pp.
Basumallik S (1966) Size differentiation in a
cross-stratified unit, Sedimentology 6, 35-68.
Bird ECF (1983) Provenance of beach sediments
in southeastern Australia (This volume).
Bouma AH (1969) Methods for the study of
sedimentary structures, Wiley-Interscience, New
York, 458 pp.
Bowman D and Goldsmith V (1982) Bar morphology
of dissipative beaches: an empirical model,
Marine Geology 50 (in press).
Chappell J and Eliot IC (1979) Surf beach
dynamics in time and space an Australian case
study and elements of a predictive model,
Marine Geology 32, 231-250.
Clifton HE (1969) Beach lamination: nature
and origin, Mar. Geol. 7,553-559.
Clifton HE, Hunter RE and Phillips RL (1971)
Depositional structures and processes in the non-
barred high-energy nearshore, J. Sediment Petrol.
41, 651-670.
Collinson JD and Thompson DB (1982) Sedimen-
tary structures, George Allen & Unwin, London
194 pp.
Davidson-Arnott RCD and Greenwood B (1976)
Facies relationships on a barred coast, Konchi~
bouguac Bay, New Brunswick, Canada, SEPM Spec.
Publ. 24, 149-168.
Deutsch A, Tietjen JH and Lee JJ (1977) Life-
histories of marine nematodes: a comparison
of the feeding habits and food digestion in
Chromadorina germanica and Diplolimella sp.
Proc. 3rd Int. Meiofauna Conference, Hamburg,
August 1977. Abstracts.
Elliot JH (1971) Some methods for the statisti-
cal analysis of samples of benthic invertebrates,
FBA Scientific Publ. 25,144 pp.
Emery KO (1978) Grain size in laminae of beach
sand, J. Sediment. Petrol. 48, 1203-1212.
Fenchel T, Kofoed LH and LAPpALAINEN A (1975)
Particle size-selection of two deposit feeders :
the amphipod ~orophium volutator and the proso-
branch Hydrobia ulvae, Marine Biology 30,119-128.
Flemming BW (1977) Depositional processes in
Saldanha Bay and Langebaan Lagoon, CSIR Research
Report 362, 215 pp.
Flemming BW (1982) Sediment mixing : its
natural occurrence and textural expression. 11th
International Congress on Sedimentology, Hamilton,
22-27 August 1982. Abstracts, p 81.
Frey RW (1975) The study of trace fossils,
Springer Verlag, New York 562 pp.
Fricke AH and Flemming BW (1983) Organism/
sediment relationships in sandy beaches II.
Selective microhabitat colonization by inter-
stitial meiofauna as a function of grain size
(this volume).
Goldsmith V, Bowman D and Kiley K (1982)
Sequential stage development of crescentic bars
southeastern Hediterranean J. Sediment, Petrol.
52, 233-249.
Grace JT, Grothaus BT and Ehrlich R (1978)
Size frequency distributions taken from within
sand laminae, J. Sediment. Petrol. 48, 1193-
1202.
Gray JS (1966) Selection of sands by Protodrilus
symbioticus (Giard), Veroffenthichungen der
Instituts fUr Meeresforschung in Bremerhaven 2,
105-116.
Gray JS (1967) Substrate selection by the
Archiannelid Frotodrilus rubropharyngeus,
Helgolander j,issenschaftliche Meeresuntersuchungen
15, 253-269.
Gray JS (1974) Animal-sediment relationships.
In Barnes H (ed.) Oceanography and Marine Biology:
An Annual Review, Vol. 12, Allen & Unwin, London
p 223-261.
Gray JS (1981) The ecology of marine sediments.
Cambridge University Press, Cambridge 185 pp.
Greenwood B and Davidson-Arnott RGD (1979)
Sedimentation and equilibrium in wave-formed bars:
a review and case study, Can. J. Earth Sci. 16,
312-332.

Guza RT and Inman DL (1975) Edgewaves and
beach cusps, J. Geophys. Res. 80, 2997-3012
Guza RT and Bowen AJ (1975) The resonant
instabilities of long waves obliquely incident
on a beach, J. Geophys. Res. 80, 4529-4534.
Harms JC (1975) Stratification and sequence
in prograding shoreline deposits. In Harms JC,
Southard JB, Spearing DR and Walker RC (eds.)
Depositional environments as interpreted from
primary sedimentary structures and stratification
sequences, SEPM Short Course 2, 81-102.
HawleY,N (1982) Inter-tidal sedimentary
structures on macrotidal beaches, J. Sediment.
Petrol. 52, 785-795.
Hayes ao (1972) Forms of sediment accumulation
in the beach zone. In Meyer RE (ed) Waves on
beaches and resulting sediment transport, Academic
Press, New York p 297-356.
Hennig HF-KO, Fricke AH, Greenwood PJ and
Eagle GA (1982) Relationships between meiofaunal
population densities and physico-chemical proper-
ties of unpolluted sandy beaches, Environmental
Monitoring and Assessment 1, 337-344.
Hicks GRF (1977) Observations on substrate
preference of marine phytal harpacticoids
(Copepoda), Hydrobiologia 56, 7-9.
Hunter RE, Clifton HE and Phillips RL (1979)
Depositional processes, sedimentary structures
and predicted vertical sequences in barred near-
shore systems, southern Oregon coast, J. Sediment.
Petrol. 49, 711-726.
Imbrie J and Newell ND (eds) (1964) Approaches
to Paleoecology, Wiley & Sons Inc., New York
Inman DL, Gayman WR and Cox DC (1963) Littoral
sedimentary processes on Kanai, a sub-tropical
high island, Pacific Sci. 17, 106-130.
Jansson B-O (1966) Microdistribution of factors
and fauna in marine sandy beaches, Veroffentl.
Inst. Meeresforsch, Bremerhaven, Sonderband 2, 77-86.
Jansson B-O (1967) The importance of tolerance
and preference experiments for the interpretation
of mesopsammon field distributions, Helgol. wiss.
Meeresunters. 15, 41-58.
Jansson, B-O (1971) The "Umwelt" of the inter-
stitial fauna. In Hulings NC (ed.), Proceedings
of the 1st International Conference on Meiofauna.
Smithsonian Contrib. to Zoology 76, 129-140.
Johnson RG (1971) Animal-sediment relationships
in shallow water benthic communities, Har. Geol.
11, 93-104.
King CAM (1972) Beaches and Coasts. Edward
Arnold, London 570 pp.
Komar PD (1976) Beach processes and sedimenta-
tion, Prentice-Hall Inc., New Jersey, 429 pp.
Lanyon JA, Eliot IG and Clarke DJ (1982)
Groundwater-level variation during semidiurnal
spring tidal cycles on a sandy beach, Austr. J.
Mar. Freshwater Res. 33, 377-400.
McLachlan A (1980 a) The definition of sandy
beaches in relation to exposure a simple
rating system, S. Afr. J. Sci. 76, 137-138.
McLachlan A (1980 b) Exposed sandy beaches as
semi-enclosed ecosystems, Mar. Environmental
Res. 4, 59-63.
131
HcLachlan A (1980 c) Volumes of seawater
filtered through Eastern Cape sandy beaches,
S. Afr. J. Sci. 75, 75-79.
Headows PS and Campbell JI (1972) Habitat
selection by aquatic invertebrates, Adv. mar.
BioI. 10, 271-382.
Hills EL (1969) The community concept in marine
zoology, with comments on continua and instability
in some marine commun~t~es: a review, J. Fish.
Res. Board Canada 26, 1415-1428.
Ravenel WS and Thistle D (1981) The effect of
sediment characteristics on the distribution of
two subtidal harpacticoid copepod species.
J. expo mar. Biol. Ecol. 50, 287-301.
Reading HG (ed.)(1978) Sedimentary environments
and facies, Blackwell Publ. Co. London 576 pp.
Reineck HE and Singh B (1980) Depositional
Sedimentary Environments, Springer Verlag, New
York 549 pp.
Renaud-Debyser J (1963) Recherches ecologiques
sur la faune interstitielle des sables. vie et
Milien, Supplement 15, 1-157.
Rhoads DC and Young DK (1970) The influence of
deposit-feeding organisms on sediment stability
and community trophic structure, J. Mar. Res.
28,150-178.
Riedl RT, Haung N and Machan R (1972) The sub-
tidal pump: a mechanism of water exchange by
wave action, Mar. BioI. 13, 210-221.
Sallenger AH (1981) Swash marks and grain flow.
J. Sediment. Petrol. 51: 261-264.
Sanders HL (1958) Benthic studies in Buzzards
Bay. I. Animal-sediment relationships, Limnol.
Oceanogr. 3, 245-258.
Schafer W (1972) Ecology and Paleoecology of
marine environments. Oliver & Boyd, Edinburgh,
568 pp.
Schmidt P and Westheide W (1972) Verteilung
von l1ikrofauna and Hikroflora in einem Strand der
Nordsee-,-Insel Sylt. Proc. 3rd European Marine
Biology Symposium, Vie et Hilien, Supplement 20.
Short AD (1979 a) Wave power and beach-stage
a global model. Proc. 16th CEC (Hamburg)
p 1145-1162.
Short AD (1979 b) Three dimensional beach
stage model, J. Geol. 87, 553-571.
Short AD (1981) Beach response to variations
in breaker height. Proc. 17th CEC (Sydney),
p. 1016-1035.
Short AD (1983) Physical variability of sandy
beaches (this volume)
Sonu CJ (1973) Three dimensional beach changes,
J. Geol. 81, 42-64.
Swedmark B (1964) The interstitial fauna of
marine sand, Biological Rev. 39, 1-42.
Syvitski JP and Hurray JW (1977) Grain-size
distribution using log-probability plots, Bull.
Com. Petrol. Geol. 25, 653-694.
Tanner WF (1964) Modification of sediment size
distributions, J. Sediment. Petrol. 34, 156-164.
Teichert C (1958) Concepts of facies, AAPG
Bull. 42, 2718-2744.
132

Thompson WO (1937) Original structures of

beaches, bars and dunes, Geol. Soc. Am. Bull.
48, 723-752.
Vilas F, Palanca A and Castan C (1982) Activity
of Amphipods in beach sediments and nearshore
environments, 11th Int. Congress on Sedimentology
(Hamilton), Abstracts, 25.
Walker RG (ed.) (1979) Facies liodels, Geoscience
Canada, Reprint Series 1, 24 pp.
Walton EK, Stephenson WE and Shawa HS (1980)
Reading segmented grain-size curves, Geol. Hag.
117, 517-524.
Webb JE (1969) Biologically significant
properties of submerged marine sands, Proc. Roy.
Soc. London 174 B, 355-402.
Wiegel RL (1964) Oceanographic Engineering.
Prentice Hall, New York.
Wright LD (1981) Beach cut in relation to surf-
zone morphodynamics, Proc. 17th CEC (Sydney),
p 978-996.
Wright LD (1982) Field observations of long-
period, surf-zone standing waves in relation to
contrasting beach morphologies, Austr. J. liar.
Freshwater Res. 33, 181-201.
Wright LD and Short AD (1982) High-energy near-
shore and surfzone morphodynamics, JOA (Halifax),
Abstracts, p 86.
Wright LD, Thorn BG and Chappell J (1978) Horpho-
dynamic variability of high-energy beaches, Proc.
16th CEC (Hamburg), p. 1180-1194.
Wright LD, Chappell J, Thorn BG, Bradshaw HP and
Cowell P (1979) l1orphodynamics of reflective and
dissipative beach and inshore systems : South-
eastern Australia, Uar. Geol. 32, 105-140.
Wright LD, Guza RT and Short AD (1952) Dynamics
of a high-energy dissipative surf zone, liar. Geol.
45, 41-62.

PHYSICAL VARIABILITY OF SANDY BEACHES
A.D. SHORT (Coastal Studies Unit, Department of Geography, University of Sydney, NSW, 2006, Australia)
L.D.WRIGHT (Department of Geological Oceanography, Virginia Institute of Marine Science, Gloucester Point
Virginia, 20362, U.S.A.)
SYNOPSIS
This paper examines sandy beach types that exist in
wave environments ranging from low «1m) to high
(>2.5m). Three basic beach types - reflective,
intermediate and dissipative, composed of six beach
states have been identified. Each beach state can
be classified by its wave-sediment characteristics.
In addition, the paper discusses the environmental
conditions required to produce each beach state.
and their associated morphology; surfzone circulat-
ion and dynamics; mode and scale of sediment ex-
change and erosion; and spatial and temporal varia-
bility.
1. INTRODUCTION
Sandy beaches are the product of waves interacting
with a sandy bed at the shoreline. The zone of in-
teraction begins at wave base and extends shoreward
across the nearshore and surfzone to the upper
limit of swash action. The extent and nature of
this zone is dependent on two parameters; the level
of wave energy which controls the depth of wave
base and limit of swash action; and sediment size
which influences sediment transport and beach grad-
ients. By combining these two basic parameters,
wave energy and sand size, it is possible to class-
ify beach types as they exist in low, moderate and
high wave environments. It is the aim of this
paper to use such a classification to describe the
basic sandy beach types that exist in the low and
high wave energy extremes and the intermediate
beach types that result from moderate wave environ-
ments.
The classification is an outgrowth of the impetus
given to beach studies by Hom-ma, Sonu (1962) who
first drew attention to the intermediate,or as they
133
called, "rhythmic", beach types,. Sonn (1973) also
presented one of the first three dimensional models
of beach variability. The reflective and dissipat-
ive beach extremes, which are now used to describe
the low and high energy beaches, were first defined
by Guza, Inman (1975) who introduced the surf scal-
ing parameter (E). Short (1979b) produced a three-
dimensional sequential model of beach changes link-
ing dissipative, intermediate and reflective under
rising and falling wave conditions. The level of
wave energy required to change from one beach type
to another was given by Short (1980). Wright, et
al. (1979) conducted experiments aimed at charact-
erising the surfzone circulation and dynamics
associated with the various beach types. This work
was expanded to include high energy beaches (Wright
et al., 1982a) and macrotidal beaches (Wright, et
al., 1982c). In all Short and Wright conducted
surveys, studies and experiments on 26 separate
sand beach systems around eastern, southern and
western Australia. A synthesis of the results is
published in Wright, et al. (1982d).
Whereas much of the above work relates beach types
solely to levels of wave height or energy, Wright,
et al. (1982d) used Dean's (1973) dimensionless
formulae Q = Hb/Tbws (where Hb is breaker height,
Tb breaker period, and Ws sand fall velocity in
cm sec-I) to combine wave and sand characteristics
to assist beach classification.
This paper is concerned with presenting a general
overview of sandy beach systems based largely on
the work of Short and Wright mentioned above. While
many of the beaches studied are in Australi~ the
work incorporates field and theoretical work fon
elsewhere, particularly North America (see above
134
references) and Japan (Sunamura, 1980a, band
Sasaki, et al., 1976).
The overview will examine each beach type in terms
of the modal environment conditions required for
its formation; its beach-surf zone morphology and
dynamics; the nature and scale of beach-surfzone
sediment exchange including beach erosion; and the
temporal and spatial morphological and shoreline
variability associated with each type. A companion
paper (Short, 1983 this volume) presents results of
investigations into sediment characteristics, bed-
forms and structures of sand beach systems.
2. BEACH TYPES
Beaches can be classified into three basic types -
dissipative, intermediate and reflective (see e.g.
Short, Hesp, 1982; Wrigh4 et al., 1982d). The
intermediate type has four states, making a total
of six beach states. These states were described
morphologically by Short (1979b), dynamically by
Wright, et al.,(1979) and combined into a general
model by Wright, Short (in press). The six states
are listed in Table I along with parameters used by
Wright and Short and others to distinguish between
beach types, and are illustrated in Figure I.
Each of the beach states is now assessed in terms
of - the environmental conditions required for
formation; their morphology, dynamics, mode and
scale of sediment exchange and erosion; and spatial
and temporal variability.
2.1 Dissipative
2.1.1. Environmental conditions. The combinations
of high waves (>2.Sm) and fine sand (mean grain
size (Md<.2mm) result in dissipative beaches
characterised by low beach and surfzone gradients,
high surf scaling parameters (i.e. >20) and like-
wise high values of ~(i.e. 6) (see Table I). They
are consequently found in regions exposed to high
breakers where fine sand is abundant. Such waves
occur year round in the storm wave and west coast
swell environments of southern Africa and Australia.
They occur seasonally in west coastu.s.and period-
ically in areas where high waves persist for
periods of at least several days to weeks.
2.1.2. Morphology. Dissipative beaches are charact-
erised by a wide, low gradient beach face extend-
ing from the foot of the dune to a lower gradient
lower beach face and surfzone (see Fig. la).
Depending on breaker height waves break 200-S00m
seaward of the shoreline (Fig. 2a). Subtle bars
or breaker zones (usually 2 but up to S) may be
present across the surfzone. Seaward of the outer
break point the gradients steepen across the near-
shore zone.
2.1.3. Dynamics. The high incident breakers begin
spilling across the outer breaker zone and decay
progressively across the surf zone. This mode of
wave breaking causes a dissipation of the incident
waves and a shift in energy to infragravity
frequencies (period >30 seconds). The shift
results in low frequency oscillations in water
level· at the shoreline called 'surfbeat'. These
ocsillations take the form of standing waves that
rise and fall in the inner surfzone and provide a
mechanism for first storing water in the surfzone
and then returning it seaward across the breaker
zone (Wright, et al., 1982a). The incoming surf-
beat is usually manifest as an aggradation of
incident bores which ultimately move across the
lower beach face as a single bore with amplitudes
up to third the breaker height (Wright,et al.,
1982a). Circulation is onshore at incident
frequency in the outer surfzone,shifting to infra-
gravity by the beach face. Water accumulates in
the inner surfzone as set-up, followed by a sea-
ward return flow during set-down. In the surf-
zone flow is predominately shoreward toward the
surface and offshore towards the bed (Wright, et
al., 1982a). Under shore normal waves longshore
and rip circulation is weak or nonexistent although
migratory rips may exist.
2.1.4. Mode and Scale of Sediment Exchange and
Erosion. The dissipative extreme represents beach
response to persistent high waves; it is at the
erosional end of the spectrum of beach types, the
 135

m DISSIPATIVE
400

300
:~:):j::l/~~~~RMnt{~~~~It~l )t)~ ),'"
5 FLAT.CONCAVE
I' DISSIPATIVE DOMAIN
HleN ell
'TROUGH' ['~~':'.~..:.~?,H FACEl..OW FREQ. SPILLING BREAKERS
ln~~WASH BORES .... if" ,.......... I"-- MLW

200 ~~rr(jf::'W~M~MMM:m~r:f'~:::~I _;':~g:~ 3 ...:8, t!~':~:~~iA~~!~ifT:~O:U~~~i' ~;~:~i:i;:;~;' ;'i;w,:

100 :::::::::::::::::::::::::: .~.~.~.9.~,::::::::::::::::::::::::: WIDE(i~~~O~5.o':)ROFILE WITH HIGHER WAVE~~C~~~~~~ Y
o ::::N.O .. :~~G.S~?~E. ~~R.I~B!L.I~Y,:::::::::::::::::::::: 50 100 150 200 250 300
INTERMEDIA TE
m LONGSHORE BAR· TROUGH
300 STRAIGHT BAA _ CRESCENTIC BAR
REFLECTIVE DISSIPATiVE
WEAK RIP 1 DDUIN 1 DOMAIN 1
BERM LOW €.b MODERATE ts
3" <eusps) SURGING

b TROUGH •• '.:. BREAKERS

100 o STEEP.

TAN~ .. 0.05-0.20
50 100 150 200 250

100

INTERMEDIATE rc-- HIGH TIDE REFLECTIVE

RIDGE·RUNNEL OR LOW TIDE TERRACE
~ c:tIOW TIDE DISSIPATIVE
MODERATE GsPLlJNGING

Ji~~-;;'~~~==~M:t'LW~__ A'
BREAKER

-3 ~q TIDE REflECTIVE
LOWC.b ~LOW TIDE DISSIPATIVE
~o.,.:.-';';:;"i'i""lli'!";:';:;"":;""'.,..,f""'~~ MOD RATE Es PLUNGING

:-~~~~~B~RE~A:KE~R~=~M~H~W=-
....;. MLW __ I'

............
. .......... ....
........... . .
REFLECTIVE

...............................
::::;:;:::::::;:::::;:;:;:::::;:;:::::::::::::::;:::;::::::::: 50 100 150 200

Figure 1: Plan and profile configuration and basic surf zone

circulation pattern of the six beach states (from Wright and
Short, in press).
136

TABLE 1 SOME CLASSIFICATIONS OF BEACH TYPE

1 2 Hb 4 '6
Beach Type Beach State Surf Hb/ T W 5 Sasaki
3 b s
Scaling

Dissipative Dissipative > 20 > 2.5m > 6 Infra-

gravity

l
Intermediate Bar Trough 20 2-2.5 6 Insta-
bility
Rhythmic bar 1.5-2
& beach
Transverse - 1.5
bar & rip

Ridge & Runnel 2.5 1-1.5 1

Low-tide
Terrace

Reflective Reflective < 2.5 < 1 < 1 Edge

Wave

1. for medium-fine Band beaches only; 2 see Figure 1; 3 from Guza and Inman, 1975; 4
from Short and Hesp, 1982; 5 from Wright et a1 1982b; 6 see Sasaki et a1 1976.

TABLE 2: MORPHOMETRIC PARAMETERS OF SELECTED BEACH PROFILE LOCATIONS

LOCATION GOOLWA N. SEVEN M. SEVEN NARRABEEN COLLAROY FISHERMANS

BEACH TYPE DISSIPATIVE I N T E R M E D I ATE REFLECTIVE

n 3 11 11 94 94 26
Months 60 26 26 78 78 30
Hb(m) 3 1.6 1.6 1-1.5 .5-1 .3
T 12 10 10 10 10 10
Gd(mm) .2 .25 .27 .3 .3 .35
W .02 .32 .036 .04 .036 .05
s
n 9.6 5.10 4.4 3.75 2.1 .60
Gradient 1:33 1:39 1:37 1:15 1:12 1:9
B-S 6 5-6 5 3-4 2-3 1
yb 90 112 112 42 36 27
ayb 10 16 15 14 5
CV .10 .14 .l3 .33 .25 .18
V 120 153 177 180 65 40
crV low 23 30 90 45 12
V!yb 1.3 1.37 1.58 4.3 1.8 1.5

n - sample size; months - length of survey; Hb - modal wave height; T - wave period; Gd - mean grain
size; Ws - mean fall velocity (cm/sec); n = Hb/T Ws; Gradient - subaerial beach slope; B-S - modal
beach state: (Short & Hesp 1982); yb - mean beach ~idth; ayb - standard deviati~n of yb(shore1ine
mobility index);CV - coefficient of variation of yb(backshore mobility index); V - mean subaerial
beach volume and standard deviation oV - (Modified from Short and Hesp 1982) •
 137

Figure 2: Representative beach types in south east Australia. a. High energy

dissipative - Hb = 3m, surfzone is 500m wide; b. Bar-trough, surf zone is
200m wide; c. Transverse bar and rip; and d. Reflective, note absence of
surfzone.
138
end toward which lower energy beaches will strive
during periods of high waves. Consequently they
are relatively stable beaches so long as waves
remain high. Most potentially active sand is
stored in and seaward of the surfzone and will only
move shoreward when waves fall below modal wave
height. Unless this occurs over a prolonged
period, rare in modal dissipative environments,
only slow and minor beach accretion will occur
(see Fig.3, North Seven Mile). An increase in
wave height will produce a relatively minor plan-
ing down of the subaerial beach and seaward move-
ment of the breaker zones on the order of 10's to
100's m. Most severe erosion occurs at the back-
beach-foredune as a result of an increase in stand-
ing wave amplitude which permits large infragravity
bores (1-2m high) to flow right across the beach
and penetrate the back beach where they produce
continuous, parallel foredune scarping (Fig. 4a)
and lor overwash at low points (Short and Hesp,
1982).
2.1.5. Spatial and temporal variability. Modally
dissipative beaches are relatively stable both
spatially, because they are uniform longshore
(see Figs. 1a and 2a); and temporally, because of
the relative minor adjustments to the persistent
high wave conditions, also aided by the fine
grain size (see Fig. 3, North Seven Mile). This
stability is reflected in the low shoreline (oyb)
and back shore (CV) mobility index (see Table 2,
Goo1wa).
2.2. Intermediate beach types.
2.2.1. Modal environmental conditions. The four
intermediate beach types (bar-trough, rhythmic
bar-beach, transverse bar-rip, and ridge and
runnel) represent a transition from dissipative
to reflective; from high wave conditions when
sediment is stored in and seaward of the surfzone,
to low wave conditions when it is stored on the
subaerial beach (Short 1979b). Consequently
sediment usually resides in the surfzone in the
form of bars. Prerequisites for intermediate
types are medium sand and moderately high waves
(1-2.5m) (Short, 1980). The surf scaling para-
meter lies between 2.5 and 20, and n between 1
and 6 (Table 1). Short (1980) found on medium
gtain size beaches that the higher energy bar-
trough system persists when modal waves exceed
2m, the rhythmic bar-beach when waves, are between
1.5-2m, transverse bar and rip with waves 1-1.5m,
and ridge and runnel with waves -1m. Because
waves are rarely stable at the above levels
beaches move from one state to another in response
to varying wave conditions. Consequently the
following refers only to modal conditions, that
is, when a beach is in dynamic equilibrium with
the prevailing wave conditions. Intermediate
beaches commonly occur on coasts exposed to high
deepwater waves but where some energy is lost to
wave shoaling and/or refraction. They also per-
sist on moderate energy east coast swell and
trade wind and monsoonal coasts (Short, 1979a).
On beaches with mUltiple bars (usually 2) the
inner may be intermediate while the outer has
more dissipative features.
2.2.2. Morphology and Dynamics. Morphologically
the intermediate states are characterised by
pronounced longshore variability caused by alter-
nating rip and bar topography and dynamically by
increasing rip circulation.
Longshore bar-trough. The bar-trough system is
distinguished by a reflective beach face, a
relatively wide deep shore-parallel trough and a
pronounced shore-parallel bar (Fig. 1b and 2b).
Waves plunge on the bar,reform in the trough and
surge up the beach face. Limited dissipation
leads to some infragravity standing wave circula-
tion (period approximately four times the incid-
ent wave, Wright and Short, in press). Edge
waves associated with the standing waves impose
weak rip circulation,which in turn leads to the
initiation of onshore migrating lobes of sand
called crescentic bars (Short, 1979b).
 139

E JI~O
3
2 GOOLWA - 60 ITHS 80

1 60
50%
DISSIPA TIVE 0
3 2/83
MLW--
R
BT
....
...
2040 *D
2 NORTH SEVEN MILE LT RB 0
-26 MTHS 50%
1
0
2
k:::-~~~~MLW- LT ~
BT

R BR

MID SEVEN MILE - 26 MTHS

0
5
4
3
2
EASTERN BEACH - 19 MTHS
1
INTERMEDIA TE 0
5
4
3
2
78 MTHS
1

3
2
1

3 %
2 100 R,

1 80 i
0 ____L -__- L__~__~~~~~~L-~MLW ______________ 60 m
REFLECTIVE ~~~~~
50%
40 :
-I
-2 20!L RB D
0···" "BR BT
0 20 60 100 140 180 220
DISTANCE SEAWARD IMETRES)

Figure 3: Sequential beach-surf zone profiles from long term survey sites chosen to be
representative of the three beach types_ The profiles illustrate the beach 'sweep' zone
and morphological variability of each representative type. Months indicate duration of
surveys, actual number of surveys given by n in Table 2. Note the steepening of beach
gradients between dissipative and reflective, and the peak in shoreline mobility and
sediment exchange associated with intermediate beacbe~modified from Short, Hesp, 1982).
The beach state histograms represent the percentage of time each of the seven sites
resides in a particular beach state ( D dissipative, BT bar trough, RB rhythmic bar and
beach, BR transverse bar and rip, LT low tide terrace and R reflective). The asterisk
indicates the modal state for the site.
140

Figure 4: The three modes of beach erosion. (a) Parallel foredune scarping caused by accen-
tuated wave set-up on a moderate to high energy dissipative beach. (b) Rip embayment erosion
(arrowed) on an intermediate beach type, with discrete arcuate scarping in lee of rips and,
(c) continuous parallel berm scarping on a reflective beach.

Rhythmic bar and beach. This state possesses the
most highly rhythmic bar and beach features. As
the crescentic bar migrates shoreward the bar
becomes highly rhythmic in plan and form as bars
alternate with rip channels. As the bar nears
the beach wave attenuation and refraction inside
the surf zone result in erosion and beach scarp-
ing in lee of rip channels and accretion in lee
of crescentic bars, thereby producing rhythmic
shoreline features called megacusp embayments and
horns. respectively (Fig. lc and 2). Spacing is
commonly 100-SOOm on open coast beaches (Short,
1979b). The surf zone dynamics is increasingly
dominated by rip circulations (Fig. 1). Waves
break more heavily over the crescentic bars, with
water flowing across and off the bars into the
channel then converging longshore to return sea-
ward as a rip current. The current pulses at
infragravity frequency indicating wave dissipation
is also contributing to the circulation. At the
beach face the waves alternate from surging in
lee of bars leading to cusp formation on the mega-
cusp horns, to plunging in lee of the rip channels
where less dissipation has occurred (Fig. lc).

Transverse bar and rip. Continued onshore migrat-
ion of the crescentic bars or a period of partial
bar erosion produces the transverse bar-rip state
(Short, 1979b) characterised by alternating trans-
verse bars and rip channels (Fig. ld and 2). In
this state rip circulation is at its strongest
1
and may reach velocities exceeding 100 cm sec
even under relatively low waves (see e.g. Fig. 5).
Figure 5: General morphology and circulation
pattern in a transverse par-rip system (Palm
Beach,NSW). Note the How over arid off the bar
(A) ,into ilnd along the feeder channels (B) and
strong rip current (C) (modified from Wright and
Short, in press).
This is the typical beach condition on many south-
east Australian beaches during summer and causes
many problems and fatalities for inexperienced
swimmers. The bars are relatively shallow,
particularly at low tide, with relatively deep
rips cutting transverse across them-the amount
of skewing depending on wave obliquity. Waves
break on the bar, flow off the bar into the channel
and pulse seaward at infragravity frequency, lead-
ing to multiple rip vortices extending seaward of
the surfzone. The rip channel is heavily rippled
and upper flow regime flow may be attained part-
icularly at low tide. The extreme shoreline pro-
tusions of the previous state become more subdued
as the embayments begin to infill.
141
Ridge and runnel or low-tide terrace. The lowest
energy intermediate state can result from contin-
ued shoreward bar migration and rip infilling
of the previous state, or from erosion of the
reflective beach and formation of an incipient
low-tide terrace (Short, 1979b). Except for
minor details they are identical in general form
and behaviour (Fig. Ie). The bar-terrace lies
at about mean low water. At high tide the low
waves (-1m) may not break on the bar but surge up
the beach face producing more reflective cond-
itions and high tide cusps. At low tide more
dissipative and weak rip circulation prevails
in the largely infilling rip channels, called
'mini-rips' (Short, 1979b).
2.2.3. Mode and Scale of Sediment Exchange and
Erosion. Intermediate beaches are the most mobile
in terms of sediment exchange between the sub-
aerial beach, surfzone and nearshore (Short,1980).
Long-term, multiple surveys of representative
beaches in N.S.W. illustrate the active sweep-
zone on a variety of intermediate beaches (Fig.3).
In all cases they exceed the sweep zone of both
the higher energy dissipative and lower energy
reflective.
The longshore bar-trough state is the most stable
of the intermediate types with sediment exchange
occurring between the beach face and low-tide
terrace, and on the bar as it migrates landward
and seaward in response to falling and rising
waves. Actual sediment exchange between the bar
and beach is relatively minor (Wright, et al.,
1982c) .
The rhythmic bar and beach and transverse bar-and-
rip are the most mobile beach forms as crescentic
bars rapidly migrate shoreward and rip channels
infill under low wave conditions, while seaward
migration of the bar and channel excavation occur
under higher waves. This dynamic mode of equili-
brium combined with the inherent highly variable
morphology between adjacent bars and rips and
142
megacusps (see Fig. lc and d) produces the great-
est spatial and temporal variability in beach
morphology and dynamics. This is reflected in the
high shoreline and backshore mobility indices seen
in Table 2.
The ridge and runnel is also very mobile in that
sediment exchange can occur rapidly between the
bar and beach and conversely. However,because of
the smaller scale of the migration and volume of
sediment involved,they are not as mobile,or as
variable as the preceeding two states.
2.2.4. Intermediate beach erosion. As intermediate
beaches lie between the dissipative and reflective
extremes they possess characteristics of both the
dissipative and reflective (see later) erosion
modes, as well as the distinctive rip embayment
erosion which characterises intermediate systems.
Rip embayment erosion is produced by greater set-
up of water as the rip feeders converge toward the
rip channel; greater wave attack due to less
dissipation across the rip channel; and,proximity
to the rip channel to convey eroded sediment sea-
ward. When the rip is stationary,as in an accret~
ing rhythmic bar-beach state,the erosion produces
embayment scarping (Fig. 4b). However, under
rising wave conditions rip spacing increases,lead-
ing Co lateral shifts or longshore displacement in
the location of rip currents. This leads to a
planing down of the beach and more subdued embay-
ments as the rips stabilize (Fig. 4c) (Short,
1979b). A typical sequence of intermediate eros~
ion is reflective erosion and removal of the beach
face, rip planing of the megacusps and beach, and
ultimately large,widely spaced rip systems (500-
1000m) interacting with wave dissipation to pulse
at infra gravity frequency. Under extreme wave
conditions, particularly in topographically con-
trolled systems (e.g. embayed beaches) very large
'megarips' or 'super rips' may drain an entire
beach system and flow 2-3km seaward (Short, 1979b,
Wright, 1980).
The lower energy transverse bar-rip and ridge and
runnel may not move past the rip formation state.
The bar will remain partially attached to the
beach unless waves exceed 1.5m (Short, 1980).
2.2.5. Spatial and temporal variability. Inter-
mediate beaches are morphologically and dynamic-
ally the most spatially (see Fig. lb,c,d,e) and
temporally (see Fig. 3) variable. Spatial vari-
ability is an inherent characteristic of the bar-
rip topography,while temporal variability is a
response to varying wave conditions and particul-
arly its effects on relocation and/or migration
of rip channels (Wright, and Short, in press).
The degree of temporal variability is expressed
diagrammatically in Fig. 3 and quantitatively by
the shoreline and backshore mobility indices in
Table 2.
2.3. Reflective Beaches.
2.3.1. Modal environmental conditions. Reflect-
ive beaches are produced by low waves «1m, Short,
1980) particularly in areas of coarse sediment
(Md>.6mm). The surf scaling parameter is <2.5,
reflecting both low waves and steep beach grad-
ient, and D is less than 1 (Table 1) also indic-
ating low waves and coarse sand. These criteria
are commonly found in areas of modally low waves
which occur on open coasts in deeply embayed and
protected beach environments where wave shoaling
and refraction significantly lower deep water
wave energy. They are also found in lee of coral
reefs, in estuaries and in the world's low wave
environments particularly tropical, polar and
protected sea coasts (Short, 1979a). On more
exposed coasts they may form following prolonged
periods of low waves (Aubrey, 1979; Short, Wright
1981), and on moderate to high energy coasts
where sediments are composed of gravel and/or
shingle (e.g. McLean, Kirk, 1969) causing D to
remain below 1.
2.3.2. Morphology. Reflective beaches represent
the accretionary extreme in beach types in that
most of the potentially active sand is stored

on the subaerial beach (Short, 1979b, 1980). The
beach is relatively high, usually containing a
berm or high tide cusps, fronted by a steep beach
face which grades into a coarse low tide step
(Fig. lf and 2). Beyond the step extends a lower
gradient relatively deep nearshore zone. There
are no surfzone or bar features.
2.3.3. Dynamics. Incident waves cross the near-
shore zone to surge over the step and up the beach
face. In doing so they produce subharmonic edge
waves which have been correlated with the accret-
ionary beach cusps (Wright, et al., 1979; Inman,
Guza, 1982). Continued growth of the cusps and
steepening of the beach face eventually dampens
the edge waves (Guza, Thornton, 1982) allowing
a straight, steep berm face to dominate. Flow is
onshore-offshore, apart from minor divergence over
cusps (see Fig. 2).
2.3.4. Mode and scale of sediment exchange and
erosion. Reflective beaches are potentially the
most unstable as any increase in wave height will
lead to beach erosion. However,because of the
modally low wave conditions this occurs infrequent-
ly. When erosion does occur it is produced by the
growth of the subharmonic edge waves (Wright, 1980)
which cause accentuated run-up which may overtop
the berm or cusps leading to scouring of the beach
face, or cause scarping of the beach face (Fig. 4).
Either method results in erosion of the berm and
movement of the sediment beyond the step to form
an incipent bar,or low tide terrace (Short, 1979b).
The edge waves may result in larger erosional cusps
(spacing approximately twice accretionary cusps)
being formed. A return to modal wave conditions
results in rapid beach face accretion.
2.3.5. Spatial and temporal variability. Like the
dissipative extreme, the reflective extreme is
relatively uniform alongshore. Except for the
cosmetic effect of cusp horns and swales there is
minor longshore variation in morphology (Fig. If).
Similarly,temporal variability is low because of
the rapidity of beach erosion and recovery, its
infrequency of occurrence and particularly the
143
small volume of sediment involved (see Fig. 3,
Fishermans). The inherent stability is indicated
by the low shoreline and backshore stability in-
dices in Table 2. Consequently reflective beach-
es are the most stable beach types so long as
waves remain below modal wave height (i.e. <1m).
3. CONCLUSION
By combining the two basic parameters that
determine beach form and behaviour, wave charact-
eristics and sediment size, beaches can be class-
ified into three basic types - reflective, inter-
mediate and dissipative. Each type is character-
ised by distinctive beach-surf zone morphologies
and associated circulation patterns and wave-
current dynamics. A further subdivision of the
intermediate type into four states, provides a
range of six beach states from low energy reflect-
ive to high energy dissipative within which all
sandy beaches can be located. Temporal and spat-
ial variations in either.,will result in a shift
in beach state. Consequently regional or spatial
variation in beach state can be explained by
associated spatial variation in wave-sediment
parameters. Temporal variation in state at a
particular site is in turn a response to varying
wave conditions. The full range of potential
response of a particular beach can be plotted as
a histogram of occurrence of the six states
(Fig. 3) called a beach state histogram. If the
breaker wave climate is known it can be plotted
as a histogram of ~ as it ranges from low to high
values and consequently from reflective to
dissipative conditions (Wright,et al., 1982b).
Plots of beach state and ~ can be used to charact-
erise their modal state and temporal variation in
breaker-beach state conditions.
ACKNOWLEDGEMENTS
This study was supported by the Office of Naval
Research, Coastal Sciences Program, Task NR 388-157
Grant N-00014-80-G-001, with additional support
from the Australian Research Grants Committee.
144
The manuscript was typed by W.A.Cooper.
REFERENCES
Aubrey DG (1979) Seasonal patterns of onshore/
offshore sediment movement, Journ. Geophys. Res.
84, 6347-6354.
Dean RG (1973) Heuristic models of sand transport
in the surfzone, Proc. of Conf. on Engineering
Dynamics in the Surf Zone. Sydney, pp.208-214.
Guza RT and Inman DL (1975) Edge waves and beach
cusps, J. Geophys. Res. 80, 2997-3012.
Guza RT and Thornton EB (1982) Swash oscillations
on a natural beach, J. Geophys. Res. 87, 483-491.
Hom-ma and Sonu CJ (1962) Rhythmic patterns of
longshore bars related to sediment characteristics,
Proc. 8th Conf. Coastal Engineering. Mexico City,
pp. 248-278.
Inman DL and Guza RT (1982) The origin of swash
cusps on beaches, Mar. Geol. 133-148.
McLean RF and Kirk RM (1969) Relationships
between grain size, size-sorting and foreshore
slope on mixed sand-shingle beaches, N.Z. Journ.
Geol. Geophys. 12, 138-155.
Sasaki T, Horikawa K and Hotta S (1976) Nearshore
currents on a gently sloping beach, Proc. 15th Conf.
Coastal Engineering. Honolulu, pp.628-644.
Short AD (1979a) Wave power and beach stages: a
global model, Proc. 16th Int. Conf. Coastal. Eng.
Hamburg, pp. 1145-1162.
Short AD (1979b) Three dimensional beach stage
model, J. Geol. 87, 553-571.
Short AD (1980) Beach response to variations in
breaker height, Proc. 17th Int. Conf. Coastal. Eng.
Sydney, pp. 1016-1035.
Short AD (1983) Sediments and structures in
beach-nearshore environments, South East Australia
in McLachlan A and Erasmus T (eds) Sandy Beaches
as Ecosystems. Junk Publishers, The Hague.
Short AD and Hesp PA (1982) Wave, beach and dune
interactions in Southeastern Australia, Mar. Geol.
48, 259-284.
Short AD and Wright LD (1981) Beach systems of
the Sydney Region, Aust. Geog. 15, 8-16.
Sonu CJ (1973) Three dimensional beach changes,
Journ. Geol. 81, 42-64.
Sunamara T (1980a) A laboratory study of offshore
transport of sediment and a model for eroding
beaches, Proc. 17th Int. Conf. Coastal Engineering.
Sydney, pp. 1051-1070.
Sunamara T (1980b) Parameters for delimiting
erosion and accretion of natural beaches, Ann.
Rep. Inst. Geosci. Univ. Tsukuba, No.6 pp. 51-54.
Wright LD (1980) Modes of beach cut in relation
to surfzone morphodynamics, Proc. 17th Int. Conf.
Coastal Engineering. Sydney, pp. 978-996.
Wright LD, Chappell J, ThtJm BG, Bradshaw MP and
Cowell P (1979) Morphodynamics of reflective and
dissipative beach and inshore systems: Southeastern
Australia, Mar. Geol. 32, 105-140.
Wright LD, Guza RT and Short AD (1982a) Dynamics
of a high energy dissipative surfzone, Mar. Geol.
45, 41-62.
Wright LD, Nielsen P, Short AD, Coffey FC and
Green MO (1982b) Nearshore and surf zone morpho-
dynamics of a storm wave environment: Eastern Bass
Strait, Australia, Coastal Studies Unit Tech. Rept.
No. 82/3. Coastal Studies Unit, University of
Sydney, 154 pp.
Wright LD, Nielsen P, Short AD and Green MO
(1982c) Morphodynamics of a macrotidal beach,
Mar. Geol. 50, 97-128.
Wright LD, Short AD and Nielsen P (1982d) Morpho-
dynamics of high energy beaches and surfzones: a
brief synthesis, Coastal Studies Unit Tech. Rep.
No. 82/5, Coastal Studies Unit, University of
Sydney, 64 pp.
Wright LD and Short AD (in press) Morphodynamics
of beaches and surfzones in Australia, in Komar PD
(ed.) Handbook Coastal Processes and Erosion,CRC
Press.

SEDIMENTS AND STRUCTURES IN BEACH-NEARSHORE ENVIRONMENTS, SOUTH EAST AUSTRALIA
A.D.SHORT (Coastal Studies Unit, Department of Geography, University of Sydney, Sydney, N.S.W., 2006,
Australia)
SYNOPSIS
Sandy beach systems consist of wave and current
deposited sand. However, the wide range of sand
sizes and wave conditions produces tremendous
variation in natural beach systems. This varia-
tion extends to their sediment characteristics and
structures. This paper presents the results of
sediment sampling, box coring of structures and
observations of bedforms on several beaches in
southeast Australia located in various wave-
sediment environments. A description of each
sub-environment across the beach systems and a
general classification of beach environments,
sediments and structures is presented.
1. INTRODUCTION
Beach systems are accumulations of wave and current
deposited sand that extend from the limit of spring
-tide,or storm swash, across the surfzone and near-
shore to the depth of modal wave base. Within
this zone the nature of wave-current dynamics and
resultant flow fields, sediment transport, bed-
forms and structures can vary significantly
depending on wave conditions, depth and location
within the beach.
It is the aim of this paper to present the physical
characteristics of the sub-environments that make
up the total beach environment. In particular
the general morphology, flow regimes, bedforms
and associated stratigraphy as well as local
sediment characteristics will be assessed.
The study of boundary layer dynamics in beach
environments and resultant flow regimes, bedforms
and stratigraphy is a relatively recent phenomenon
and reflects a growing interest in both the under-
standing of modern beach sediment dynamics, and
145
the application of this information to paleo-
beach environments. Clifton (1969) and in
particular Clifton et al., (1971) clearly
established the relationship between flow fields,
bedforms and depositional structures (strati-
graphy) in beach environments. Subsequently
Clifton (1976) presented a conceptual model of
wave-formed sedimentary structures in which four
basic flow fields were identified. In the same
paper Clifton discussed the role of variable
levels of wave energy and grain size on the nature
and extent of these fields and associated struct-
ures. Davidson-Arnott, Greenwood (1974,1976) have
shown how the bedforms and structures vary across
barred topography and have used this to interpret
bar formation and dynamics (Davidson-Arnott,
Pember, 1980; Greenwood, Davidson-Arnott, 1979).
The influence of wave energy levels was further
investigated by Howard and Reineck (1981) who
compared the depositional facies on high and low
energy beaches, and Hunter et al., (1979) who
reported on structures on barred and barless
beaches.
While the basic flow controls on bedforms, their
general location across the beach and the assoc-
iated structures are well documented, the role
of variation in beach environmenmin response to
changing wave environments is emphasised in this
paper. Using the beach model of Wright, Short
(in press)whichispresented in this volume (Short,
Wright, 1983) each beach type is systematically
discussed in terms of its morphodynamic and
sedimentary characteristics.
146

2. FIELD DATA
The morpho dynamic approach to beach environments
(see Wright, Thom, 1977) emphasises the role of
variations in environmental parameters in deter-
mining the resultant beach type and associated
characteristics (Short, Wright, in press). This
approach is used here to examine the nature and
extent of certain beach characteristics, and in
particular the variability in these characteristic~

on beaches composed of sand of varying size and

exposed to different levels of wave height. Sever-
al beaches in central New South Wales (NSW) exposed
to modal (or most persistent) low «1m) and moder- +
ate (1 to 2.5m) waves were selected for surveying,
sediment sampling and box coring (see Table 1 and
Fig. 1). In addition morphodynamic and sediment BEACH
BEACH
data from 20 other beach systems around southeast
and southern Australia were incorporated (see + + 34"

Wright, et al., 1982 for location). The NSW and

most other sites were located in temperate (lati- SCALE

tude 30-37 0 S), microtidal (1 to 1.5m spring tide

range), east coast swell, west coast swell and F--·SE.VEN MILE BEACH

pe~iodic storm wave environments (Davies, 1980).

In this paper the results are compressed to give a

description of each enviromental 'type'. For
details of each particular beach system and
associated sediments and structures see Short
(in press).
3. BEACH ENVIRONMENTS
Beaches are composed of three subsystems - the
subaerial beach or foreshore, surf zone and near-
shore, the latter grading into the offshore zone
or inner continental shelf, while on sand beaches
the former can become blanketed by aeolian deposits
The subaerial beach consists of swash-deposited
sediment and may incorporate mesoscale morpholo-
gical features including berms, cusps and mega-
cusps. The surfzone extends from the toe of the
beach face, about mean low water, to the outer
breaker zone. It can contain a wide arrangement
of bars and troughs. The nearshore zone extends
Figure 1: Location of the six beach sites (arrowed)
selected for sediment sampling and box-coring.
The wave-sediment characteristics of each are list-
ed in Table 1.
from modal wave base, where the modal wave will
initiate sand transport, shoreward across the
wave 'build up' zone to the point of wave break-
ing. The actual wave base migrates seaward and
shoreward as waves rise and fall respectively.
The size and extent of the subaerial beach, surf-
zone and nearshore and the nature and arrangement
of micro- and mesco-scale features within a beach
system is dependent on wave and sediment character-
istics (see Short and Wright, in press). In the
following presentation each of the three sub-
systems is discussed in terms of its major morph-
ologic components and their associated flow
 147

TABLE 1: WAVE-SEDIMENT CHARACTERISTICS DF BEACH SITES

LOCATION M. SEVEN GRANTS NARRABEEN HAWKS NEST PEARL FISHERMANS

BEACH TYPE I N T E R ME D I ATE REFLECTIVE

n 25 21 35 14 17 10
Hb(m) 1.6 1.6 1.5 1 .5 0.3
T 10 10 10 10 10 10
Gd(mm) .27 .30 .30 .26 .50 .35
Ws .036 .04 .04 .035 .076 .05
n 4.4 4.0 3.75 2.fl5 .65 .60
Gradient 1:37 1:29 1:15 1:13 1:12 1:9
B-S Bar-Trough Rhythmic Transverse Ri dge-Runne 1 Reflective Reflective
Bar-Beach Bar-Rip

n - number of cores; Hb - modal breaker height; T - Wave period; Gd - mean grain size; Ws - mean fall velocity
(cm sec-I); n = Hb/T Ws. Gradient - subaerial beach slope; B-S - beach state (see figure 2).

regimes, bedforms and stratigraphy, together with
the sediments that prevail in the central NSW
field sites. The beach model of Wright, Short
(in press) which classifies beaches into low
energy reflective, moderate energy intermediate
and high energy dissipative (see Fig. 2) will be
used as a framework for presentation.
3.1. Subaerial Beach and Foredune. The foreshore
consists of sediments deposited by swash uprush
and backwash across the beach face. The height,
width, gradient and morphology are dependent on
grain size (Bascom 19S1) and wave conditions
(Short, 1980) and vary considerably between
beach types (see Figs. 2 and 3 , Table 1). This
variability has been confirmed by Mackaness (1981)
who found beach face sediments and structures
could be used to differentiate statistically
between reflective, intermediate and dissipative
beaches.
3.1.1. Reflective beaches consist of a coarser
grained (mean diameter Md >.6mm) , relatively
narrow steep beach face fronting a berm or cusps.
On very low energy (Hb <.Sm) and consequently
very stable beaches the berm is capped by aeolian
deposits and primary dune vegetation (see Fig.3a).
On higher energy reflective beaches (breaker
height Hb> 1m) where on-offshore sediment exchange
is more prevalent, the berm crest is bare and
slopes gently landward into a runnel where high
tide swash collects before it percolates into the
berm (see Fig. 2f, 3b). The berm crest consists
of horizontal to slightly landward dipping thinly
bedded parallel laminations (Fig. 4a), composed
of relatively uniform, well sorted medium grains,
with occasional coarser (carbonate rich) layers
(Fig. Sa). The steep beach face (l:S to 1:10) is
composed of seaward dipping parallel laminated
beds which coarsen slightly downsiliope. The
beach face terminates at the toe of the beach
where a coarse grained beach 'step' persists.
Waves break and surge over the step and run up the
beach face. Consequently the step is the zone of
highest wave energy and due to selective sorting
is composed of coarsest sediment (Nielsen, in
press). It consists of steeply seaward dipping
thicker parallel beds, composed of carbonate rich
coarser sand (Fig. 4b).
148

m DISSIPATIVE
400,---------------------------

SPILLING BREAKERS

a MLW

b TROUGH
100

TAN@ ~ 0.05-0.20

50 100 150 200 250

INTERMEDIATE .--.-- HIGH TIDE REFlECTIVE

RIDGE·RUNNEl DR LOW TIDE TERRACE
~ c:r-lOW TIDE DISSIPATIVE
MO DERA IE C.SPLUNGING
BREAKER MHW

~~Hl~"'""~_------"M"'L"'-W-----A'

~?':'W"""""">4''"-------''M''L"W----- B'

REFLECTIVE
~F,l!,CTIVE DOMAIN
m ~ ~GH & lOW TIDE)

200 WIDE HIGH LOW Eb lOW t.b

ilERM (CUSPS)

...
3 .'. SURGING

~~~E.~ ~~~~~. :~?~

>:P :~ ~:~:0:~:~:~:~:: -5 ,Y!~~: j'~O~E~:.::~~:. ~~:::;~:~:~:~:;~I~N.T.
100. BERM_CREST. 0 RUNNEL". BREAKER MLW

::::::::::: :9:Y:~:~~:::::::::
a :;:;::::::::;:;:;:;:;:;:;:;:;:;:;:;:;:;:;:;:;:;':;:;:::::: ....
... . .......... ..................
.... .......... .
...........................
............................
....... ......................
....... ...... .............. ..
"
50 100 150 200

Figure 2: The beach model of Wright, Short (in press) illustrating the
range of beach types (dissipative, intermediate and reflective) and states
(a to f) that occur on natural beaches. See Short. Wright 1983, this
volume for full description. The field sites discussed in this paper
represent five of the six beach states.
 149

3.1.2. Intermediate beaches consist of four

(a) ( b)
types (Fig. 2b,c,d and e) and both within and
Foredune
4 ~ FISHERMANS B. between the types possess the most variable
REFLECTIVE
beach face topography. They may possess the
2
characteristic megacusp horns and embayments, as
a well as secondary berms, cusps, erosion scarps
and steps. The beaches are of moderate width
-2 Hb " 0·5 (Fig. 3c, d, e and f) with pronounced spatial
Ws " 0·076
n " 0·65 variability due to the presence of megacusps;
100m a 100 m
(d) and temporal variability due to relocation and
Foredune reformation of megacusps. The megacusps consist
4 HAWKS NEST NARRABEEN BEACH
RIDGE AND RUNNEL TRANSVERSE BAR AND RIP
26·9 ·82 NOV. '81 of both relic and fresh beach face deposits. The
2 relic deposits in lee of the horns and exposed
MLW by the embayment scarp, are usually horizontally-
a
bedded berm deposits. The fresh horn deposits
-2
Hb" 1·0 Hb" 1·5 are moderately to steeply dipping thin parallel
Ws "-0'035 Ws "0·04
n " 2·85 11. "3·75 laminations, with steeply dipping thicker beds
100 m a 100 200m in the embayments. On the horns cusps may be
deposited in the high tide swash zone, with a
GRANTS BEACH
RHYHMIC BAR AND BEACH coarse step at their base. The step usually
4 MAY' 82
continues into the embayments. Sediments are
2. generally slightly finer than reflective beaches,

MLW and locally coarsen toward the step, with coarse,

a
carbonate-rich sediments composing the steeply
-2 seaward dipping beds in the step (Fig. 4b,Sa and
Hb " 1-7
Ws " 0·04 b). Locally, the depth of disturbance increases
n" 4·0
100 200 m between the horns, where waves tend to surge up
the beach face, and embayments where the higher
MID - SEVEN MILE BEACH
BAR-TROUGH waves tend to plunge heavily on the step (Short,
4 12. £, 13 ·11· 81
Wright, in press).
2. 3.1.3. Dissipative beaches are typified by wide,

MLW
low gradient (1:20 to 1:40) rather featureless
a
beach faces, reflecting both the finer sediments

-2 and surfbeat type swash dynamics (Fig. 2a) (Short,

Hb " 1·7
Ws "0·036 Wright, in press). The beach face is uniform
n "4-4
-4
a 100 200 300 m alongshore. Shore normal it consists of an upper
and lower beach face. The upper beach face may
contain low subdued cusps during periods of
Figure 3: Beach-surfzone profiles across the six
accretion. The cusps and upper beach face consist
field sites showing the cross-sectional morphology
and location of box cores (black squares). See of finely laminated slightly seaward dipping
Table 1 for definition of Hb, ws and ~.
parallel beds (Fig. 4d). The lower beach face is
dominated by higher velocity backwash,particularly
the return of infragravity swash bores
150

Figure 4: A selection of resin impregnated box cores from the field sites illustrating some
of the beach structures discussed in the text. All cores are perpendicular to the shore with
shore to left. a. reflective berm; b. intermediate step; c. reflective nearshore; d. dissipative
beach face; e. intermediate trough; f. intermediate bar; g. intermediate rip channel; h. inter-
mediate offshore; i. intermediate nearshore; j. intermediate nearshore.
 151

to incident waves, as well as to even subtle

changes in wave height, period, direction or
(a) SIZE ~ SORTING CARBONATE % (b) SIZE ~ SORTING CARBONATE %
-I 0 I 2 3 05 I 0102030 -I 0 I 2 01020304050 groupiness. In a state of dynamic equilibrium it
5 \ I is continually adjusting to, or striving to
\ \,
1m)
\I adjust to, variation in incident wave character-
/ ~
'.~"? istics. As a result it is the best indicator of

.\ Y \ <:.:: prevailing wave conditions as indicated in Fig.2 .

l.
~ " ~/'/
;:: -5
. -5
\ Surf zones are the most variable beach subsystem,

.)
\
~ \ both in their extent and their morphological
'" \

-10
I -10
\
J complexity.

r \ (\
\ 3.2.1 Reflective 'surfzone'. Reflective beaches
\ 1\ do not contain a surfzone in the true sense. By
",.,~,,} ) -15
\
\:
\\ definition waves move across the nearshore zone

\,!\
PEARL a
\ : 1\

1'1
HAWKS NEST and surge over the step and up the beach face.

"""'" f,4c
GRANTS b - 20
\\
\\
Apart froID the step there are no bars or channels,

i\1 !
and the only circulation is associated with the
MID SEVEN MILE I ~
\
It
GDOLWA
,
I
uprush of the surging wave and backwash. The
MODAL WAVE BASE - 25 backwash will be uniform on a berm face and seg-
regated on a cusped face.
3.2.2. Intermediate surfzones. Intermediate

/
-30
beaches have the most complex surf zone circulation
and morphology. They range from lower energy
welded bars called ridge and runnel or low tide
Figure 5: Sediment characteristics (size, sorting terrace which may be cut by mini-rips(Fig. 2e),
and percent carbonate) across the beach, nearshore
through welded transverse bars with well developed
and offshore zones of the six field sites.
rip channels (Fig. 2d), to crescentic bar systems
(Fig. 2c) and finally a shore parallel bar and
which result in transition and upper flow regimes
wide deep trough (Fig. 2b). The shoaler bars are
being achieved by the backwash, characteristised
dominated by breaking waves and transition flow
by a hydraulic jump and a slowly seaward migrat-
regimes which leads to planar beds and the form-
ing standing wave. A consequence of this is the
ation of horizontal parallel lamination, which
production of small current ripples and assoc-
increasingly dip landward approaching the inner
iated cross-bedding. The alternation of low tide
bar edge (Davis, et al., 1972) (Fig. 4). Sediments
producing current ripples, and high tide producing
are fine to medium (Figs. Sa and b).
planar beds leads to alternating sequences of
parallel beds and usually seaward dipping cross
Shore parallel troughs and channels, particularly
bedding. The base of the beach face is determined
associated with rip feeders, contain both wave
more by low water level than any distinguishable
orbital motions of reforming waves or bores, and
morphological characteristic, as it grades onto
longshore flow feeding the rips. The former
the inner breaker zone. Steps are characterist-
generate asymmetrical shore parallel wave ripples,
ically absent, though grain size does coarsen into
the latter shore normal current ripples, with both
the inner surfzone (see Seven Mile, Fig. 5b).
commonly co-existing. Consequently complex cross-
3.2. Surfzone. The surfzone is perhaps the most
energetic sediment-water interface. It responds bedding of varying scales and coarser sediment
152

dominates (Fig. 4e). seaward directed return flow, particularly at the

Rip channels have the highest flow velocities
which may reach upper flow regimes. The seaward
flow generates seaward migrating current ripples
and megaripples and at times antidunes. The
resulting structures are dominated by larger
scale steeply seaward dipping beds with some
large scale cross bedding (Fig. 4g). Sediments
are relatively coarse and higher in carbonate
(Fig. 5).
During an accretionary sequence (Short, 1979) the
bar sequence will migrate over the channel and
into the rip producing vertical sequence of bar
structures over channel structures (Davis, et al.,
1972; Hunter, et aI., 1979).
3.2.3. Dissipative Surfzone. Dissipative beaches
are by definition fronted by wide,low gradient
surf zones across which the spilling breakers
dissipate their energy. In actuality the energy
is shifted to low frequency infragravity surf-
beat, manifest as a standing wave which alternately
sets-up and sets-down against the beach face
(Wright, et aI., 1982). Because of the energetic
nature and great width of high energy dissipative
surf zones no cores have been obtained in this
study from such surfzones. However, sediment
samples, scuba observations and wave-current
dynamics have been obtained. The surfzone can be
divided into three zones, an inner and outer
breaker zone and an intervening 'trough' (Fig. 2a) .
bed which produces some current ripples. This
is what Clifton et al., (1971) termed the 'inner
rough facies'. The inner breaker zone experiences
both onshore flow during set-up and seaward bed
flow during set-down. Consequently a range of
flow regimes from lower to transition persist,
resulting in planar bed and wave and some current
ripples. Clifton et al., (1971), and Hunter et
al., (1979) who cored such a zone found predom-
inarttley planar bed features with seaward dipping
parallel laminations. Sediments are slightly
coarser (Fig. 5b).
3.3. Nearshore Zone. The nearshore is the zone
where wave orbital motions begin to interact with
the bed sufficiently to initiate sand transport.
The depth where this interaction normally occurs
is called the modal wave base. This interaction
increases shoreward as depth decreases ultimately
resulting in wave breaking in the surfzone.
Morphologically it is the least variable of the
three zones, containing no longshore variation,
although the extent varies considerably between
beach types (Fig. 6). In general, sediments
fine seaward of the surf zone and remain relative-
ly uniform and well sorted with low carbonate
across the nearshore zone (Fig. 5).
NEARSHORE PROFILES
~ FISHERMAN'S

The outer breaker zone is dominated by high spil- ~_ PEARL~ ~-=- __ _

:;: 10 " ~ --..:.:-----------___ -,- __ HAWKS NEST
ling breakers and asymmetrical on-offshore flow.... }"., " '""-- GRANTS - -_ _

--
at the bed. The bed is planar grading into mega- Q ~ ,~
~'- ----- __ --

'--..
-20 " ' . . . ~EVEN MILE --
ripples resulting in a mixture of plane bedding , ~"
~- ____'0.RRABEEN -----....__._._. __ . " ......
and large scale landward dipping bedding. This ---
is what Clifton et al., (1971), called the 'outer
DISTANCE SEAWARD (Km)
planar-outer rough zone'. Sediments are fine
(Fig. 5b). The trough region permits waves to Figure 6: Nearshore profiles across the six
partially reform~assisting oscillatory flow at field sites showing location of nearshore box
cotes (bla'ck squares).
the bed, resulting in wave ripples and at times
megaripples. However setdown leads to phases of

3.3.1. Reflective nearshore zone. Reflective
beaches grade from the beach face - step straight
into the nearshore zone. The steepness of the
beach face continues to a depth of 1 to 5m where
it flattens out and a low gradient linear zone
extends seaward to wave base (see Fig. 6). On
the sloping inner section large wave ripples or
small megaripples may result in moderate scale
cross-bedding and landward dipping beds. In the
flatter nearshore zone parallel crested asymmet-
rical wave ripples dominate resulting in predom-
inantley landward dipping cross-bedding (Fig. 4c).
3.3.2. Intermediate and Dissipative Nearshore
Zone. Both these beach types have essentially
the same sequence of flow regimes, bedforms and
structures , the major variation being their
vertical and horizontal extent (Fig. 6). The
general sequence was developed by Clifton et al.,
(1971) and has subsequently been confirmed by
Clifton (1976), Hunter, et al., (1979), Shipp
(1982) and Short (in press).
Immediately seaward of the outer breaker zone,
strong asymmetrical wave orbital motions result
in shoreward migrating megaripples up to 4-5m
long and 1-1.5m high. They may be subparallel or
lunate and result in high angle landward dipping
beds and large scale cross-bedding (Fig. 4i). As
depth increases, cross-ripple patterns dominate,
resulting in complex cross-bed patterns. These
grade into asymmetrical wave ripples, which are
irregular and sinuous but become more parallel
with depth. They produce cross-bedding structures
with a dominance of low angle landward dipping
beds (Fig. 4j).
Increasing depth across the nearshore also results
in a decreasing depth of disturbance by wave
motions. This enables organisms to increase the
rate and decrease the depth below the surface of
bioturbation . During low waves bioturbated
sediments may approach to within a few centi-
metres of the surface up to halfway across the
153
nearshore zone, on intermediate beaches into
depths of 8-l0m. A similar depth of bioturb-
ation on moderate to high energy beaches was
found by Clifton (1976) in Oregon, and Howard,
Reineck (1981) in Southern California.
3.4. Offshore Zone. In central N.S.W. the trans-
ition from nearshore to offshore is marked by
several distinct changes in sediment character-
istics (Fig. 5). Across the modal wave base
boundary and increasingly into the offshore
sediment grain size increases, percent carbonate
increases, sorting becomes moderate to poor, per-
cent silt increases, and quartz grains become
'frosted' from chemical etching. In addition,
biot'urbation increases and moves to the surface.
In all offshore sites observed and cored,well-
developed asymmetrical-symmetrical parallel-
crested wave ripples were present indicating
prior activity during higher wav~and also in
agreement with Clifton et al., (1971). However,
coarser, poorly sorted sediments and bioturb-
ation mask the structures (Fig. 4h).
4. DISCUSSION AND CONCLUSION
Because of the potentially wide range of sandy
beaches environments,all beach systems cannot be
illustrated by a single diagram,or characterised
by a single set of parameters. Beach systems do
however grade from low to high wave environments
and can be classified and typed along this grad-
ation. Accompanying this gradation are changes
in all associated beach characteristics. In
this paper a classification of beach systems
(Wright, Short, in press) is used to distinguish
the three basic beach types, and to introduce
the sub-environments associated with each. This
paper is primarily concerned with the general
morphology, flow regimes, bedforms and associated
stratigraphy in each beach environment, as well
as the sediment characteristics in the N.S.W.
field sites. The results are summarized in
Fig. 7 which presents composite sequences of the
various beach sub-environments discussed in the
154

REFLECTIVE INTERMEDIA TE
Rhythmic
-I DISSIPA TI VE

Ridge & Transverse bar&

Runnel bar& rip beach Bar-trough
FISHER HAWKS NARRA SEVEN
n -MANS
PEARL
NEST -BEEN
GRANTS
MILE
GOOLWA
0.60 0.65 2.85 3.75 4.0 4.4 9.6
5

:::::::RH~:
>:iiilF'::::,:,},
0
:,BEACH FAC
{SUP
,BF: :::::<:
;}}:, ,::>:; I:,} ::,:\ ::::,:,: ,,} 1::,\:::::,:,
f::
F.:::::: ::::::::: :::::::: ::::,:, ::::a ,:(1 i;:il I { }(~F
:,::::: I~):: :::

: ::::':: ':':':" \m
':i::, }:::,;:,:: ::::,::,r::::l:,
"

~r~M~9~f
::::,:11:::::':5.#
liROuW :,:,:,:,
I,::
\i: iN~) UI ",
»> »>" : : ,: ::: 'inW} :
,:::::,:.:: :',:::::'::':'::
~~ARSHQ~l 1,::>:>< Y~'ul ~~T :~~
5 ,:>:,::", «<><> »><» >:< :,:::,:::::::::::::::::::::::::::
OFFSHORE
i IMMWMF ::::::<>:: :< ",:::::' «><::: ,:,:,:,:«<>
I:: IE,. ><»> ::> >:::: > :»+
>::>«:::
:[ 10 - -
It: :»> <« r»<»
i!: OFFSHORE
::::> »: ,:«::, » ;lm~1 t~ :>:: »
~ <> »> .'.~.~.~.~.~'~~~~:.: «
15 «> «> «»
Cl
1/< <>::: :>,:::,:,:::<", }~HM~Wf
><» >:<»
OFFSHORE «<» <>«: »
<)}+
20 - »»>< ;,:>
><>: <»
OFFSHORE »>< : i «>
25 i»<>< i ••••••••
<»
30 -
Jt.<)Slli OFFSHORE ><)~
ill?
f-- I-

OFFSHORE OFFSHORE

35

Figure 7: Occurence and actual extent of beach subenvironments (shaded) within

the six field sites (from left to right - Fishermans, Pearl, Hawks Nest, Narra-
been, Grants, Seven Mile) and a fully dissipative beach (Goolwa. South Austra-
lia). The corresponding beach state and value of Q is given for each. The
associated morphology for each beach is given in Figs. 2,3, and 6, sediment
characteristics in Fig. 5 and stratigraphy (structures) in Figure 4.

text. The presence and extent of each sub-environ- characterise reflective beaches are non-existent
ment is arranged in sequence from low energy on dissipative beaches. The beach face which is
reflective to high energy dissipative. The most steep and narrow on reflective beaches, and vari-
obvious feature is that as Q increases, primarily able on intermediate, becomes wide and low grad-
due to increasing wave energy, so does the vertical ient on dissipative beaches. Surfzones typified
extent of the entire beach system. Secondly, the by bar-trough forms are non-existent on reflect-
nature and extent of each sub-environment varies ive, whereas they increasingly dominate inter-
between beach types. Berms and steps which mediate and grade into inner and outer surfzones

in dissipative systems.
While this classification is based on field sites
in N.S.W. the value of ~ can be calculated for any
sandy beach environment and will give the position
of that beach relative to those shown in Fig. 7.
Consequently, Fig.7 provides an outline of the
expected nature and range of beach sub-environ-
ments, with the details of each given in other
figures and the text.
AGKNOWLEDGEMENTS
This study was supported by the Office of Naval
Research, Coastal Sciences Program, Task NR 388-
157, Grant N-00014-80-G-000l, with additional
support from the Australian Research Grants
Committee and Australian Marine Science and
Technologies Committee. Essential field assist-
ance was given by P. Cowell, G. Lloyd, J.M.Short
and L.D.Wright. M. Gleeson performed the sediment
analyses. Figures were drafted by J. de Roder,
cores photographed by A. Pritchard and the
manuscript was typed by W.A.Cooper.
REFERENCES
Bascom WM (1951) The relationship between grain
size and beach face slope, Trans. Amer. Geophys.
Union. 321, 866-874.
Clifton HE (1969) Beach lamination-- nature and
origin, Mar. Geol. 7, 553-559.
Clifton HE (1976) Wave formed sedimentary
structures - a conceptual modal, in Davis, RA Jr
and Ethington, RL (eds.) Beach an~Nearshore
Sedimentation. Soc. Econ. Paleo. Mineral. Special
Pub. No. 24, pp. 126-148.
Clifton HE, Hunter RE, and Phillips L (1971)
Depositional structures and processes in the non-
barred high-energy nearshore, Journ. Sed. Pet.
41, 651-670.
Davidson-Arnott RGD and Greenwood B (1974)
Bedforms and structures associated with bar topo-
graphy in the shallow-water wave environments,
Kouchibouguac Bay, New Brunswick, Canada. Journ.
Sed. Pet. 44, 689-704.
Davidson-Arnott RGD and Greenwood B (1976) Facies
relationships on a barred coast, Kouchibouguac Bay,
New Brunswick, Canada, in Davis RA Jr and
Ethington RL (eds.) Beach and Nearshore Sediment-
ation. Soc. Econ. Palaeo. Mineral. Special Pub.
155
No. 24, pp. 149-168.
Davidson-Arnott RGD and Pember GF (1980) Morph-
ology and sedimentology of multiple parallel bar
systems, southern Georgian Bay, Ontario, in
McCann SB (ed.) The Coastline of Canada. Geol.
Survey of Canada. Paper 80-10. pp. 417-428.
Davies JL (1980) Geographical Variation in
Coastal Development, Longman London. 2l2pp.
Davis RA, Fox WT, Hayes MO and Boothroyd JC,
(1972) Comparison of ridge and runnel systems in
tidal and non-tidal environments, Journ. Sed. Pet.
42, 413-421.
Greenwood B and Davidson-Arnott RGD (1979)
Sedimentation and equilibrium in wave formed bars:
a review and case study, Canadian Journ. Earth
SCi., 16,321-332.
Howard JD and Reineck H-E (1981) Depostional
facies of high energy beach to offshore sequence:
comparison with low-energy sequence, Amer. Assoc.
Petroleum Geol. 65, 807-830.
Hunter RE, Clifton HE, and Phillips L (1979)
Depositional processes, sedimentology structures
and predicted vertical sequences in barred near-
shore systems, southern Oregon Coast, Journ.Sed.
Pet. 49, 711-726.
Mackaness J (1981) Microsedimentary structures
of intertidal sand beaches. B.A.Hons. Thesis,
Department of Geography, University of Sydney,
Australia, lllpp.
Nielsen P (in press) Entrainment and distribut-
ion of different sand sizes under water waves,
Journ. Sed. Pet.
Shipp RC (1982) Nearshore depositional facies
of Long Island, New York, U.S.A. Abstract of paper~
11th IntI. Congress of Sedimentology. Hamilton,
Ontario, pp. 103.
Short AD (1979) Three dimensional beach stage
mOdel, Journ. Geol. 87, 553-571.
Short AD (1980) Beach response to variation in
breaker height, Proc. 17th IntI. Conf. Coastal
Engr. Sydney, pp. 1016-1035.
Short AD (in press) Sediment characteristics
and bedforms on modally low, moderate and high
wave beach environments in southeast Australia,
in Greenwood B and Davis RA Jr (eds.) Coastal
Environments Dominated by Waves, Special Pub. Soc.
Econ. Paleont.Mineral. Tulsa, Oklahoma.
Short AD and Wright LD (1983) Physical varia-
bility of sandy beaches. Sandy Beaches as Eco-
systems, Junk Publishers, The Hague.
Wright LD and Short AD (in press) Morphodynamics
of beaches and surfzones in Australia, in Komar PD
(ed.) Handbook of Coastal Processes and~rosion,
CRC Press.
Wright LD, Short AD and Nielsen PN (1982)
Morphodynamics of high energy beaches and surf zones:
a brief synthesis, Coastal Studies Unit Tech. Rept.
No. 82/5. Coastal Studies Unit, University of
Sydney, 64 pp.
Wright LD and Thom BG (1977) Coastal morpho-
dynamics. Progress in Physical Geography, 1,
4, 2-459. '

A THEORETICAL MODEL OF SURF ZONE CIRCULATION AND DIATOM GROWTH
D. F. WINTER (University of Washington, U.S.A.)
1. INTRODUCTION
This paper examines some of the processes and
mechanisms that govern the time-averaged distribu-
tion of diatoms in surf zone environments. To this
end, an idealized mathematical model is developed
of the relative concentration of diatoms resulting
from advection and phostosynthesis in the surf zone.
Segments of exposed, gently sloping beaches with
well-developed surf zones are found along several
coastlines throughout the world. It is character-
istic of these regions that one or two diatoms
species grow vigorously and dominate the surf zone
phytoplankton population. For example, along the
sandy beaches of the Pacific Northwest, the diatoms
Chaetoceros armatum and Asterionella socialis grow
in great abundance, with cell densities as high as
5 6 -1
10 -10 ml during the winter months (Lewin et al.,
1975). Exceptionally high concentrations of
Chaetoceros armatum and Asterionella japonica have
been reported along a portion of the west coast of
New Zealand (Cassie and Cassie, 1960). Along the
east coast of South Africa the surf zone diatom
Anaulus birostratus appears to dominate the phyto-
plankton population (McLachlan and Lewin, 1981).
Investigators who have studied these areas speculate
that competing species are excluded because the
dominant diatoms have adapted to vigorous growth in
the surficial foam produced by breaking waves. The
growth process of the surf diatoms may involve
positive buoyancy as a central feature; however,
such property has not yet been convincingly
demonstrated. Nevertheless, although cells are
157
distributed throughout the water column at all
times due to vertical mixing, during the daylight
hours far greater numbers are found near the sur-
face (Lewin and Hruby, 1973). At any given time
some fraction of the cell mass floats at the sur-
face itself, where filaments of diatoms alter the
surface tension properties of the surface bubbles,
partially stabilizing the surface foam. It is
believed that all the diatoms mentioned either are
capable of stabilizing surface foam or are posi-
tively buoyant. Both mechanisms substantially
increase the nearsurface concentration of cells
and enhance their exposure to available light,
providing the surf zone diatoms with a decisive
advantage over other light-sensitive species.
It is observed that high densities of diatoms are
confined largely to the surf zone and that patchi-
ness often characterizes the population distribu-
tion. Locally high concentrations of cells have
been observed in all three geographical areas
mentioned above, and the horizontal extent of the
patches is often many tens of metres. A schematic
sketch of diatom patches in the surf zone off the
east coast of South Africa has been given in
McLachlan and Lewin (1981); part of that sketch is
shown here as Figure 1. McLachlan and Lewin
proposed the following mechanism for patch
formation: during daytime hours, positive
buoyancy tends to concentrate cells at the surface;
at the same time, there is some movement of diatoms
by longshore and rip currents. Patches of diatoms
will tend to form in the rip currents when there
is a balance between the tendency for the cells to
158

longshore variation in wave amplitude. Waves

approaching the shore tend to break when the ratio
of the wave height to water depth exceeds a certain
critical (constant) value and, in consequence, near
the breaker line, higher breaking waves will result
in higher mean set-up, which ultimately tends to
drive currents into the surf zone. Shoreward of
the breaker line, wave energy is dissipated and
the equations which govern time-averaged water
motion in the surf zone have forcing terms which
produce a flow driven in a gyre pattern as indi-
cated schematically in Fig. 2. Thus there is a

FIGURE 1. Diatom patches in the surf zone along

along the coastline of South Africa. The arrows
OFFSHORE
indicate location of feeder currents of the rip ZONE
channel (from McLachlan and Lewin, 1981).
X=Xo

be advected seaward with the rip currents and to

be carried shoreward by the wave bores. However,
the wave bore contribution to the shoreward advec-
tion of cells was not estimated quantitatively and
it may not be adequate by itself to maintain the
appropriate equilibrium with the rip current. A
second mechanism that may operate to promote patch
development in rip currents is an onshore wind FIGURE 2. Coordinate system used in the analysis.
Mean circulation is indicated by the arrows.
which, if sufficiently intense, may produce a shore-
ward near surface shear layer, thereby effectively
net flow shoreward in the regions of high breaking
retarding the outflow of diatoms from the surf zone
waves, followed by longshore flow toward regions
region. A third possibility is that patchiness is
of low breakers. In the surf zone, longshore
a consequence of wave-induced advection throughout
currents converge toward the latter regions and
the nearshore zone, vigorous photosynthesis in the
flow seaward as rip currents.
main part of the surf zone, and net negative growth
in the vicinity of the breaker line and beyond. A
Since the mean flow occurs in the form of gyres,
mechanism of this sort is most likely to be effec-
water moving shoreward toward the breaker line
tive when the beach slope is small, the surf zone
will be recirculated into the surf zone. Offshore
is broad, and flow throughout the nearshore region
sampling at Copalis Beach, Washington, has shown
is relatively slow. This latter process is
that cells of the surf zone diatom species are
described in more detail below.
extant well beyond the breaker line, although
their concentrations are lower by one or two orders
One of the principal modes of time-averaged flow in
of magnitude than in the surf zone itself (Lewin
the nearshore environment is circulation in the
and Hruby, 1978). Moreover, the cell distribution
form of gyres. Outside the breaker line, undula-
in the vertical shows no evidence of flotation.
tions in depth or wave interactions produce a
Some fraction of the offshore diatoms are
 159

undoubtedly viable (analogously to algal cells at Longuet-Higgins ',and Stewart (1962), Bowen (1969),

great depth in deep estuaries). Hence, as water Bowen and Inman (1969), Komar (1971), and Mei and

moves across the breaker line into the surf zone, Liu (1977) are representative and the reader is

it brings "seed stock" cells into an environment referred to these discussions for appropriate

favorable for vigorous photosynthesis due to the background to the discussion below.

presence of surficial foam and bubbles produced by

breaking waves. The seed stock cells begin to Throughout this analysis, a Cartesian coordinate

grow as they are first advected shoreward and then system is used, with the Ox-axis positive in the

in the longshore direction within the surf zone. offshore direction and y denoting distance in the

The specific growth rate is high and positive longshore direction. The Oy-axis is coincident

throughout the main part of the surf zone where with the stillwater line (see Fig. 2). It is also

the cells are near the surface and foam is in ample convenient to employ a nondimensional shifted off-

supply. When the surf is broad and gently sloping, shore coordinate I; = sex + x s )' where Xs is the
the average current will be relatively slow and mean beach line and 2ITiS is the rip current spac-

the residence time of a water parcel within the ing. The variation of parameters in the longshore

surf zone will be correspondingly long enough to direction is assumed sinusoidal. If xb denotes

allow development of high concentrations of cells the position of the breaker line, then

in the feeder currents of the rip channel. In the I;b = S(xb + xs) is a measure of the width of the

vicinity of the rip channels, production is ex- circulation gyre.

pected to remain high because foam is extant as

the region of breakers is approached. However, Nearshore circulation is assumed to be driven by

as the algal population moves seaward beyond the the momentum flux of waves which approach a

breaker line, it will enter a respiration phase straight coastline at nearly normal incidence.

and the density of viable cells will eventually The bottom topography is assumed to be planar,

decline as the environment no longer provides the except for small perturbations which are periodic

surface bubbles which appear to be essential to in the longshore direction. We express the still-

growth. water depth in the form

(la)

An idealized mathematical model has been formulated where

to test the qualitative description given above. h sx (lb)
o
The assumed processes are (1) advection in the
and
nearshore gyre and (2) growth and respiration at
(lc)
a net specific rate which is generally large and
positive in the surf zone, falling to negative with s representing the mean slope and hb denoting

values near and beyond the breaker line. In the the water depth at the mean breaker line. Here,

next section a few observations are made concerning 6 is a parameter much less than one and the modu-

the relevant aspects of the physical oceanography lation function f is of the order of one at

of the surf zone.

2. NEARSHORE CIRCULATION Suppose that waves of amplitude a o and circular

Since the dynamics of advection in the surf zone frequency w approach a coastline where the bottom

have been discussed by several workers, a detailed topography is represented by eqn. (1). The incom-

account need not be given here. Papers by ing wave train develops an amplitude variation in
160

the longshore direction due to the variable bottom

depth, and also possibly as a consequence of inter- H++
q curl1/!+ it (7)
action with edge waves. It has been observed both
in the field and in laboratory experiments that the Inside the breaker line, circulation is driven by
ratio of the wave amplitude a to the total water wave-induced excess momentum flux expressed in
depth appears to remain approximately constant; terms of radiation stress. The horizontal momen-
i.e. , tum equations express a balance between the hydro-
a = y(h + T)), static pressure gradient, radiation stress, and
where T) is the free surface height above stillwater bottom friction, which is usually assumed to be
+
level and y ~ 0.4. Moreover, it is observed that linearly proportional to the velocity q; i.e.,
+ +
at the position of the breaker line the wave ampli- T = pcfu o q, where P represents the mean density,
tude is related empirically to the water depth by c f is a friction coefficient, and U o is a measure
a = Ybh, where Yb ~ 0.7. The stillwater depth ~ of the orbital velocity in the 'Surf zone. Outside
at the breaker line can be estimated from this last the breaker line, vorticity is conserved and the
relation and linearized shoaling theory: flow is driven primarily by flow inside the
breaker line. The vorticity equation in the surf
zone is a second order linear (by assumption)
(2)
partial differential equation for 1/!-(~,y) with an
inhomogeneous term which accounts for the effects
from which the mean position of the breaker line
of the mean hydrostatic gradient and the normal
can be determined:
radiation stresses Sand S and the effect of
xx yy
(3) ray deflection 8 through the stress component
S (see Mei and Liu (1977)). The vorticity
xy
It can also be shown that the mean beach line is
equation in the surf zone can then be expressed
given approximately by in the form

The equations governing the mean flow are verti-

cally-integrated equations of horizontal momentum
and continuity. A perturbation analysis has been
carried out by Mei and Liu (1977) who derive
linear forms of the relevant equations to the order
0(oy 2 ). The continuity equation has the form With the assumed dependence of hI on ~ and y, and
for nearly normal wave incidence, the deflection
(5) angle is given approximately by

where H+ ho beyond the breaker line and

8(~,y)
H- = a(x + x s ) inside the surf zone, with

3 2- 1 where
a s(l + "2 y ) (6)
~ -3/2 fCO d~.

In the sequel it is convenient to introduce

transport streamfunctions in each region:
 161

If one carries out the indicated operations, the with the Wronskian, W = -2s2. Since~­ o at
righthand side can be written more explicitly as s = 0, the solution for ~ is

r.h.s ~o S ~(s) sin(Sy),

where the constant is given by

~o
(8)
(9a)
and the forcing vorticity factor ~ has the form
Beyond the breaker line, we have simply

- 4 5 sf' _ 2.8 f 1 5
4 (4 + s
2
)g.
(9b)

Here, the prime indicates differentiation with

since ~+ vanishes as s approaches infinity. The
respect to S. The vorticity equation beyond the
constants ~l and c2 are determined by the conti-
breaker line has the same form, except that the
nuity of ~+ and d~+/dt; at s = sb'
righthand side is equal to zero since the forcing
vorticity is negligible and the variable s is to be
The situation considered by Bowen (1969) corre-
replaced by s - Ss = Sx. In order to obtain a
sponds to an extended shoaling zone with ray
circulation description for a particular beach, it
deflection neglected. In this instance the flow
is necessary to evaluate the several parameters on
in the surf zone is seaward in the shoaling
the righthand side of eqn. (8). (The estimation
regions and shoreward in the embayments (deeper
procedure is carried out below for Copalis Beach,
water). Mei and Liu (1977) examine two cases:
Washington.)
one in which f is parabolic and nonzero only in
the surf zone and a second example where
Once the modulation function f is chosen the vorti-
Xo = 2xb and f has the form of an isoceles tri-
city equations are easily solved by first separat-
angle with apex at the breaker line. In each of
ing variables and then, in the case of the surf
these two cases, as sb increases, ray deflection
zone, reducing to quadratures the resulting inhomo-
has an increasing tendency to alter the sense of
geneous ordinary differential equation for ~-(s)
longshore circulation and eventually, for narrow
say. Throughout the entire nearshore region the
gyres, the mean flow is seaward in the embayments.
stream function can be generally represented as

One of the solution forms outlined above has been

applied to the surf zone at Copalis Beach,
Washington. Although certain aspects of the surf
Linearly independent solutions of the homogeneous
zone biology have been carefully studied, almost
ordinary differential equation for ~-
no direct observation of the physical circulation
characteristics at Copalis Beach has been carried
(1 + S) exp(-S)
out. The surf near Copalis Beach, Washington, is

and wide and has an exceptionally gentle slope, with

(1 - S) exp(S) s ~ 0.006. The first breaker line is estimated

to be farther than 250 m from the beach line and
162

the rip current separation appears to exceed the be larger than 0.02. In the present work, we have
surf zone width by a factor of two or three. assigned c f a value of 0.08. With U
o ~/ghb'
Assuming an offshore wave train amplitude a o of cf = 0.08, and 0 = 0.1, the value of ~o is about
1 m and a period of 8 sec, we find hb = 1.8 m from 2.2. The resulting streamline configuration is
eqn. (2) and xb = 290 m from eqn. (3) with exhibited in Fig. 3, where two gyres are shown.
s = 0.006. These values are consistent with
observations at the site and with data on nautical
charts. According to eqn. (4) the location of the
mean beach line is approximately x = -x s = -30 m.
We assume here that S(xb + x s ) ~ 2 or, more con-
veniently, ~b = 2rr/3.

A sandy beach with a gentle slope is often charac-

terized by a bar near the breaker line. To repre-
sent the effect of such a feature at Copalis Beach
we take the modulation function f to be

f (~) o < ~ <b~

~b < ~ < ZSb

StACH LINE
0, ~ > 2~b·

This is one of two examples discussed by Mei and FIGURE 3. Streamline configuration in adjoining
gyres for a beach slope of s = 0.006 and ~b = 2rr/3.
Liu (1977) and the reader is referred to that work
Other relevant parameters are given in the text.
for further details. The solution for ~- has been
expressed in quadrature form convenient for com-
3. DIATOM GROWTH AND DISTRIBUTION
putation (eqn. (9a)), but the integrated expres-
It is in this environment that surf zone diatoms
sion is too cumbersome to be reproduced here.
often proliferate to form locally high concentra-
The solution for ~+ is given by eqn. (9b).
tions of cells. We proceed now to develop a
simple idealized mathematical model for their
Of all the parameter assignments, c f is perhaps
growth and distribution. For simplicity, the
the most uncertain. It is introduced in the
effects of lateral turbulence, nutrient limitation,
analyses of Bowen (1969) and Mei and Liu (1977)
and grazing by secondary producers have been neg-
as an empirically determined bottom friction
lected in the present work. Under these assump-
coefficient and is assigned a value of the order
tions, the vertically-integrated equation for time-
of 0.02 for beaches of moderately steep slope by
averaged diatom density P(x,y) can be written as
Inman ~ al. (1971). However, since horizontal
turbulence effects are not explicitly included, ~ • (H+ q P) = r(x,y)P, (10)
the stress terms in the momentum equations must
where r(x,y) denotes the net specific growth rate.
account for all dissipative forces and, in the
The equation is to be solved in the domain
case of wide, shallow surf zones, the value of
-x s < x < 00, 0 < y <Srr. If we make use of the
the effective friction coefficient c f will likely
transport streamfunctions, eqn. (10) can be written
 163

in the form primarily because of high cell concentration

near the surface. Near the beach line, the
r(x,y)P.
average photosynthetic rate may be low since algal
material is deposited on the beach by the action
Hence, we have
of waves and wind. In the immediate vicinity of
dP dx
the breaker line, vigorous turbulence and vertical
r(x,y)P 31/1
oy mixing probably have the effect of impeding photo-
synthesis by reducing total exposure to light and
From the second of the equalities we have the
possibly interfering with the effective uptake of
obvious result that the streamline passing through
dissolved nutrients. Seaward of the breaker line
the point (x i ,TI/2) has the equation
the cell population may enter a respiration phase
as the environment no longer provides the condi-
1/I(~,Y) = 1/I(~i,TI/2) - ~i (constant), (11)
tions to which it is adapted for rapid growth.
As a working hypothesis, we assume the net specif-
where ~i "identifies" the streamline in question.
ic growth rate to be independent of the longshore
From the first equality, we arrive at a solution
direction and to vary in the offshore direction
of the differential equation in the form of an
in the general manner just described. This
integral of known quantities along a streamline;
assumption is speculative, however, since measure-
in terms of the shifted coordinate, we obtain
ments of the offshore variation of primary

+ '2
P(~,y) -- P. exp [- 1 f 2 r(~,y)2 1/2 d~ 1 (12)
productivity have not been carried out. For
1. (3 r (~ - ~i ) simplicity, we now assume that ~(~) has no· more
than one critical point in (O'~b)' The assumed
behaviour of r(~) then suggests that
where the upper sign is taken for 0 < (3y < TI/2.
Here, r is the streamline identified by the con-
(14)
stant ~i and Pi is a specified value of P at t~e

point (~i,TI/2) on r. In the integral along

will allow reasonable representations of the
1/1 = ~i' it is convenient to choose ~ as the inte-
specific growth rate under a variety of circum-
gration variable. Since the density P is time-
stances. Moreover, with this particular choice
averaged and not influenced by diffusion by
for r(~), the integral constraint is satisfied
assumption, the specific growth rate must satisfy
and the distribution of P(~,y) often may be
the integral constraint
determined analytically. For example, consider
r(~,y) the choice
f~max o (13)
~min
p(~) (15)

where ~max and ~min are minimum and maximum values

where a and b are arbitrary constants which
of ~ on r.
adjust the shape of the growth rate variation,
once ~ has been determined, and Pmax is the
In order to proceed it is necessary to make some
maximum specific production rate divided by the
further assumptions regarding the variations of
maximum value of (a~ + b)-l d~/d~. We introduce
specific growth rate r. When the essential
the notation
nutrient supply in the. water is adequate, photo-
synthesis proceeds rapidly within the surf zone
164

3.0 - 3.5 mg C per mg Chl ~ per hour at 300

P -1/2
footcandles for Chaetoceros. These values were
max
ct< (1 - A2) , o< A < 1, (16a)
"'3 2 obtained during a bloom which was two orders of
P magnitude less than the maximum that may be
max (A 2 _ 1)-1/2,
ct A > 1, (16b)
> reached on that New Zealand beach. Cassie and
b8 2
Cassie note that "the gross instantaneous rate
of primary production must at times reach an
with A = a~i/b. The value of P(s,y) is determined
as follows: point (s,y) lies on a streamline exceptionally high figure." In laboratory
cultures of Asterionella and Chaetoceros, doubling
identified by ~i' which determines a value of A.
If A lies in the interval 0 < A < 1, then P(s.,y) times of 14 hr and 30 hr, respectively, have been

is given by reported by Lewin and Mackas (1972) but growth

rates in the field must be considerably higher.
For the present calculation, we assigned a nominal
P P [1 + A sinBy + cosBy /(1-A2)]+ ct< (17a)
i A + sinBy value of 20 to the carbon-to-chlorophyll ratio
and took the maximum growth rate per unit of
For A > 1, the appropriate expression for P is
chlorophyll to be 3.5, giving a maximum specific
growth rate of approximately 4.86 x 10- 5 sec- l .
- 1 + A sin8 Y]
P Pi exp [ + ct> arcos A + sin8y . (17b) In eqn. (15) for p(~), we set a = 0 and b = 1;
the offshore variation of net specific growth
rate for this choice is shown in Fig. 4. The
In each expression the upper sign preceding ct is
to be taken when By lies in the interval (0,n/2).
1.0
0.8
Diatom growth at Copalis Beach, Washington is used W 0.6
~
to illustrate these determinations. During the a:: 0.4 I BREAKER LINE
I I
0.2 I
winter months the nearshore waters have more than I-
~ I
~ O~~------------------~T-----------------~
an adequate supply of nutrients from the Columbia '"u -0.2
River plume offshore and copious freshwater runoff ~ -0.4
u
w
from adjacent land areas. Large populations of g:, -0.6
- 0.8L---------~7T!;c--------------f27T;;c---------c!;7T~-----'
Asterionella socialis and Chaetoceros armatum have DISTANCE FROM BEACH I';)
been observed in the surf zone over many years.
FIGURE 4. Net specific growth rate (normalized)
It is observed that the diatoms proliferate in
as a function of offshore distance s.
the surf zone, primarily inside the first breaker
line, and that locally high concentrations appear
graph of res) in the figure has been normalized
sometimes in the form of separate "patches" so the maximum value is equal to one; the actual
spaced in the longshore direction (in a manner growth rate used in the calculation is obtained
similar to that illustrated in Fig. 1). by multiplying the ordinate by 4.86 x 10- 5 sec- l .

In the calculation of the surf zone diatom growth The solution for the diatom concentration P is
and distribution in this environment, we make use given by eqn. (17a) with A = O. All that remains
of growth measurements made from samples acquired is the specification of Pi' Since measurements
in the field by Cassie and Cassie (1960). These of relative diatom concentration along a line
workers reported specific production rates of

from the beach to the centre of the circulation
gyre are not available, it is necessary to make
some plausible assumption in this regard: one
possibility is that the total number of cells in
the water column increases from the beach line to
a point near the breaker line in a monotonic
fashion. In the present instance, the function
~-(~) has the appropriate behaviour throughout
most of the surf zone and, for convenience, it was
used to specify Pi:
(IS)
Isopleths of relative diatom concentration are
shown in Fig. 5.
BEACH LINE
FIGURE 5. Isopleths of vertically-integrated
diatom concentration in 4 adjoining gyres for the
standard case of high growth rate and gently slop-
ing beach (r* = 4.S6 x 10- 5 sec- l and s = 0.006).
max
It is apparent from inspection of Fig. 5 that
locally high concentrations of cells are extant
in the vicinity of the rip channel. In fact, the
density there is higher by one or two orders of
magnitude than it is in the embayment regions.
Thus, it seems plausible that the mechanisms of
advection and photosynthesis can together produce
the patchiness observed in the surf zone along
the Washington coastline. In the figure it can
be seen that all lines of equal concentration are
closed within the gyre and that, although high cell
density regions are located in the rip currents,
a narrow band of low density isopleths is concen-
trated on the centre line of the rip channel.
165
This feature is an artifact associated with the
neglect of lateral turbulence. The effect of
horizontal mixing would be to shift the centres
of adjacent gyres toward the rip current and to
produce diffusion of cells across the centre line
of the rip channel.
4. DISCUSSION AND CONCLUSIONS
Several numerical experiments were performed to
examine the dependence of the behaviour of P(~,y)
on various parameters. First, the effect of
diminished growth rate was investigated and, as
expected, the concentration of diatoms everywhere
declined. The consequence of decreasing the
growth rate by approximately 40% without altering
anything else, including the specification of Pi
as given by eqn. (IS), is shown in Fig. 6. As
I
RIP CHANNEL
I
RIP CHANNEL
BEACH LINE
FIGURE 6. Isopleths of vertically-integrated
diatom concentration in 4 adjoining gyres
corresponding to gently sloping beach and some-
what lower maximum growth rate (r*max = 2.7S x
10- 5 sec-I).
expected, the density is somewhat lower and the
regions of relatively higher concentration moved
apart. However, the patches are still extant in
the vicinity of the feeder currents of the rip
channels.
In another numerical experiment, the effect of
changing the beach slope was examined. Since the
phenomenon of proliferating surf zone species
seems to be limited to wide gently sloping
beaches, calculations were performed with larger
166

values of s, all other specifications and In all cases examined in this study, the maxima

parameters being the same. Isopleths of P cor- of the calculated distributions were closer to the

responding to s ~ 0.02 are exhibited in Fig. 7. breaker line than actually observed. There are
two possible reasons for this discrepancy. One
possibility is that the diatoms are less buoyant
I on the average as the population moves toward the
RIP CHANNEL
breaker line; i.e., the flotation mechanism is
less effective in the region of increasingly
intense vertical mixing. A second possibility
relates to the nearshore circulation theories
which predict that the centre of the circulation
gyre is close to the breaker line. As a conse-
BEACH LINE
quence, the greatest accumulation of diatoms occurs
much closer to the breakers than to the beach line.
FIGURE 7. Isopleths of vertically-integrated
diatom concentration in 4 adjoining gyres for the It may be that wide beaches with very gentle
case of high growth rate, but a moderately steep slopes must be characterized by two or more
beach (s ~ 0.02).
breaker lines. If such is the case, not only the
diatom growth model but also the circulation
Clearly, the regions of locally high concentra-
theory would need to be modified appropriately.
tions of cells are gone from the rip channels and
the population density is generally lower. The
The quantitative results of this study support the
isopleths of diatom density now follow the stream-
hypothesis that advection in nearshore gyres,
line configuration and the concentration is as
coupled with high production within the surf zone
high as it is only because the population is
and respiration near and beyond the breaker line,
maintained through specification of Pi. In
is responsible for vigorous growth of surf zone
reality, surf zone diatoms would probably not be
diatoms and the formation of patches in the
competitive at all in this environment because of
vicinity of rip currents. Numerical experimenta-
the narrow surf, rapid advection, and diminished
tion with the model suggests that a broad well-
supply of foam.
developed surf with gentle bottom slope is
prerequisite to the proliferation of the diatom
Two other experimental results will be mentioned
populations. There appear to be two reasons for
here. Increasing the constant a in the equation
this: (1) a wide shallow surf has the necessary
for p(¢) from zero to a positive number greater
supply of surficial foam for cell flotation and
than b has the effect of moving the maximum of the
(2) when advection speeds are low, as will be the
growth rate variation toward the beach and, as a
case for a wide beach with gentle slope, the resi-
consequence, the region of high cell concentration
dence time of cell parcels in the surf is long
in the rip channel becomes elongated in the
enough to permit a substantial increase in the
seaward direction. The effect of altering the
population by the time the parcels arrive in the
variation of P. to a distribution which is not
~
vicinity of the rip currents.
reasonable (e.g., zero at the beach line and at
the breaker, contrary to observation) produces
The numerical model was idealized intentionally in
locally high concentrations, but the distributions
order to allow the development of analytic solu-
are very different from any that have been
tions for both streamline configurations and diatom
reported.

densities. Having analytic solutions available
allows the determination of trends and parameter
sensitivities with comparative ease. However, in
order to retain analyticity it was necessary to
exclude certain factors and processes which also
play some role in the overall growth and distribu-
tion of surf zone diatoms: these are lateral
turbulence and nonlinear advection, nutrient
sources and distribution, wind effects, and sinking
of cells beyond the breaker line. For the most
part, the treatment of these effects would require
the use of numerical methods. Obviously, it would
be highly desirable to supplement the extension of
theoretical treatments with appropriate field data,
including spatial variations of cell concentra-
tions, nutrient distribution, and net specific
primary production rates.
Finally, it should be pointed out that patch
formation by the nearshore advection-net photo-
synthesis interaction may be augmented by wave bore
transport and onshore wind stress, as mentioned in
the Introduction. In any case, these processes
are not mutually exclusive. Moreover, greater in-
sight into these mechanisms and an evaluation of
their relative importance will require the acquisi-
tion of appropriate field data, including
horizontal and vertical flow distributions, cell
concentrations, and wind .measurements.
5. REFERENCES
Bowen AJ (1969) Rip currents 1. Theoretical
investigations, Journal of Geophysical Research
74, 5467-5478.
Bowen AJ and Inman DL (1969) Rip currents 2.
Laboratory and field observations, Journal of
Geophysical Research 74, 5479-5490.
Cassie RM and Cassie V (1960) Primary production
in a New Zealand West Coast phytoplankton bloom,
New Zealand Journal of Science 3, 178-199.
Inman DL, Tait JR and Nordstrom CE (1971) Mixing
in the surf zone, Journal of Geophysical Research
76, 3493-3514.
167
Komar PD (1971) Nearshore circulation cells and
the formation of giant cusps, Bulletin of the
Geology Society of America, 2643-2650.
Lewin J, Hruby T, and Mackas D (1975) Blooms of
surf zone diatoms along the coast of the Olympic
Peninsula, Washington. V. Environmental conditions
associated with the blooms (1971 and 1972),
Estuarine and Coastal Marine Science 3, 229-241.
Lewin J and Hruby T (1973) Blooms of surf zone
diatoms along the coast of the Olympic Peninsula,
Washington. II. A diel periodicity in buoyancy
shown by the surf-zone diatom species,
Chaetoceros armatum T. West, Estuarine and
Coastal Marine Science 1, 101-105.
Lewin J and Hruby T (1978) Blooms of surf zone
diatoms along the coast of the Olympic Peninsula,
Washington. IX. Factors controlling the seasonal
cycle of nitrate in the surf at Copalis Beach
(1971 through 1975), Estuarine and Coastal Marine
Science 7, 173-183.
Lewin J and Mackas D (1972) Blooms of surf zone
diatoms along the coast of the Olympic Peninsula,
Washington. I. Physiological investigations of
Chaetoceros armatum and AsterioneZZa sociaZis in
laboratory cultures, Marine Biology 16, 171-181.
Longuet-Higgins MS and Stewart RW (1962)
Radiation stress and mass transport in gravity
waves, with application to 'surf beats', Journal
of Fluid Mechanics 13, 481-504.
McLachlan A and Lewin J (1981) Observations of
surf phytoplankton blooms along the coasts of
South Africa, Botanica Marina 24, 553-557.
Mei CC and Liu PL-F (1977) Effects of topography
on the circulation in and near the surf zone -
linear theory, Estuarine and Coastal Marine
Science 5.

HOLOCENE COASTAL DEVELOPMENT IN THE NW PART OF THE NETHERLANDS
E.F.J. DE MULDER (Dutch Geological Survey)
INTRODUCTION
In The Netherlands, sandy beaches occur along
almost the entire coastline, including the ex-
posed beaches of the Wadden Islands off the nor-
thern coast. These beaches are invariably backed
by a broad or narrow strip of relatively heavily
vegetated coastal dunes. This natural coastal
defence system protects the hinterland i.e. areas
of reclaimed land (polders), all lying well below
mean sea level.
2 GEOLOGICAL SETTING AND PALEOGEOGRAPHY
The area under discussion is situated in the
large subsiding North Sea Basin, where approxi-
mately 350 metres of moderately to poorly con-
solidated fluvial, marine, glacigenous, and
eolian, predominantly sandy deposits were laid
down during the Pleistocene. Apart from gla-
NAP
ciotectonic displacements, no major tectonic
dislocations have affected the area. The younger
Pleistocene strata are basically in a horizontal
position. The Pleistocene deposits reflect a
periodicity of strongly divergent climatic con-
ditions and indicate an alternation of cold,
glacial, and warm interglacial periods. After
the last glacial period (Weichselian), the cli-
mate ameliorated and ice caps melted, which led
to a marked rise of the sea-level (see Fig. I).
This event marked the start of the Holocene,
10 000 Cl4-years Before Present (BP).
The previously continental, flat North Sea area
FIG. I. Hean sea-level curve for the coastal area
became filled with sea-water and the coastline
prograded rapidly towards the present coast of
169
The Netherlands, which was reached about 7500
Cl4-years BP. Due to the Pleistocene surface
configuration of the area, which reflected a
river-valley system (Fig.2), the sea entered via
the topographically low, former river valley in
the SW. The Atlantic sea could penetrate deeply
land-inward via ENE-directed tidal channels,
corresponding with the configuration of the
Pleistocene surface contour lines. Approxi-
mately at the Atlantic-Subboreal boundary (5000
Cl4-years BP) the coastline reached its eastern-
most extension in the southern part of the in-
vestigated area, about 10 km east of the present
shoreline, while further shifting of the coast-
line in the northern part of the area was blocked
by once emerged Pleistocene areas, several kilo-
metres off the present coast.
so 7000
10
12
14
16
18
20
of The Netherlands (after van de Plassche,
1982).
170

FIG.2. Top of the Pleistocene deposits


The increasing distance between the two roughly
parallel coastlines, the filling up of the
southern'part of the tidal flat in the hinter-
land with clastics, and the presence of an area
with a large potential storage capacity for
clastic sediments in the northern part of the
study area, led to the creation of a vast, deeply
eroded northward-directed tidal inlet, the Bergen
Inlet, about 4900 - 4800 Cl4-years BP (see
Fig. 3). During the next 1800 CI4-years, clastic
sedimentation from the inlet took place in the
northern part of the area by means of predomi-
nantly southward shifting patterns of tidal
channels causing contemporaneous clastic sedi-
mentation in channels and flats and peat growth
in the abandoned areas. The coastal barriers
bordering the inlet, gradually moved up in both
longitudinal and transverse directions, approach-
ing each other and causing the inlet to become
narrower. Shortly before 3000 Cl4-years BP, the
Bergen Inlet was closed.
Except for the development of another inlet
(Oer-IJ) in the SW, the other parts of the study
area did not receive marine sediments during the
next period of more than 2000 CI4-years, and
large quantities of peat could accumulate. In
the last 800 years (since about 1200 A.D.) the
sea has penetrated the coastal area successively
in three, initially small inlets. In large
areas in the west, brackish to marine deposits
were laid down.
The combined effects of early human activities
(peat extraction, reclamation) and natural
conditions (erosion by storms) resulted in the
development of rapidly expanding lakes in the
hinterland. Most of these lakes were pumped dry
during the sixteenth and seventeenth centuries.
Two of the inlets were closed by natural causes.
The embankments of the third inlet (Marsdiep,
north of Den Helder) have been stabilized
artifically.
171
3 COASTAl DEVELOPMENT
The main physical factors which governed the
Holocene coastal development in the western part
of The Netherlands were the post-glacial sea-
level rise, the morphology of the Pleistocene
surface, sand supply, horizontal and vertical
tidal movements, wave and wind induced currents,
the coastal gradient and the effects of micro-
climatic changes. These partially interacting
factors were not only determined but were also
quantitatively approached by means of geological
investigations during the past 20 years. For
details, reference is made to Jelgersma (1961),
Van Straaten (1965), Beets, et al (1981), Van de
Plassche (1982) and De Mulder, Bosch (1982).
The three successive stages in the Holocene
coastal development which are discussed below
are based mainly on the reconstructions given
by van Straaten (1965), (see Figure 4).
I. The post-glacial sea-level rise caused the
coastline to shift landward. No direct evidence
of the occurrence of coastal barriers older than
5000 Cl4-years is left, but changes in the Late
Atlantic/Early Subboreal tidal channel pattern
point to the existence of eroded beach barriers
in the area (De Mulder, Bosch, 1982). Because
the rise of the sea-level was initially rapid,
erosion was concentrated in the intertidal zone
and this resulted in a gentle coastal slope.
Transport of eroded coastal sands was predomi-
nantly performed by longshore tidal currents.
From our data it is apparent that the landward
shift of the coastline proceeded faster in the
southern part of the area than in the north, due
to the presence of the emerged Pleistocene sur-
face in the latter area. Between 5000 and 4700
BP the southern coastline reached its easternmost
position, which lay between 7 and 10 km landward
from the present coast. The coastline in the
northern part of the area was fixed during an
earlier phase at a position several kilometres
off the present coast.
172

Schoorl
o

520

De Rijp
o

3km
J

IT2Eill Sand ridges at or near th. surface

cons;s/ir,g of Older Beach and lor Dune
Sands
"
•
C 74 -doting 3600 -3700 B.P Ibivalved molluscs)
C14 -doting 3750 -3600 BRlbivalved molluscs)

~ Limit of distribution of the Bergen Clay

* C14 -doting 3900 -3750 B.P Ibivalved molluscs)
o C14 -doling 4350 - 4200 BRlbase of peat bed)

FIG.3. Configuration of sand ridges, mainly

representing fossil coastal barriers,
the Bergen Inlet and Cl4-locations
near Alkmaar, (after De Mulder &
Bosch, 1982).
 173

longshore tidal currents. The coastal accretion

led to the development of a flat beach plain with
some subordinate ridges. Accretion rates have
been deduced from Cl4-data for the areas south
of Alkmaar (Beets, et aI, 1981), and north of
The Hague (Van Straaten, 1965), which gave mini-
mal values of 5 mly and 3 mly respectively (see
Fig. 3). This process of accretion continued
until the beach plain became attached to the
landward migrating and accumulating subtidal
sandbar.

After attachment, the beach became fully exposed

again, a relatively high beach barrier developed
and this was followed by the accumulation of
dunes. A new set of subtidal bars developed
several kilometres off the beach as a result
of both wave-induced sand transport towards
the coast and longshore tidal currents.
[ill Dun. and b.ocn send d~llJd ebov. IM'Qfl _-~vwI ~ Tidal cnonn.I d.posits

[JP1.istoc._send

• P-at
mLogoonoI'iOYS

~ Locust",.,. deys
Mineralogical investigations and recent moni-
IlIIIlIIlI Tido! flet ~;t, 5S8 Op"" man'" <kopcsits toring of the sediment movement in the inter-
tidal zone point to a northward-directed net-
FIG. 4. Reconstruction of assumed coastal
development in the Western part of longshore sand transport. Longshore transport
The Netherlands (after van Straaten,
is held responsible for the main part of the
1965).
coastal accretion. This process of coastal
accretion continued until at least 3000 BP.
Beets et al. (1981) assume the end of this ac-
2. Due to the gradual retardation of the sea-
cretion to date from Roman times (1000 BP).
level rise, coastal erosion made place for sub- Reliable data are not yet available for calcu-
marine erosion by wave action. In this phase, lation of the maximum accretion distance off
wave-induced sand transport towards the coast in the present coast, but from historical data a
combination with longshore sand transport caused minimum distance of about 3 kilometres can be
by the dominant northward directed tidal currents,
assumed. Possibly, the southern coastline
resulted in accumulation of subtidal sandbars aligned with the northern coastline.
several kilometres off the coast. After the crests
had reached a considerable height, these bars shel- 3. In the 12th century (i.e. 800 BP), strong
tered the submarine leeward zone from strong wave coastal erosion resulted in a steeper coastal
action and promoted the deposition of fine-grained gradient. The beach became fully exposed once
material in the sheltered zone. This in turn again and a high beach bar developed along a
caused a reduction of the water depth in the shel- strongly retreating coastline. This erosion
tered zone and intensified the coastal accretion phase is attributed to substantially increased
of relatively coarse-grained sands carried by storm frequencies due to micro-climatic changes
(Lamb, 1961).
174
The submarine erosion mobilized very large amounts
of sand which, once transported to the coast, ac-
cumulated as dunes with heights up to 60 metres.
Other factors, such as the reduced supply of sand
from the river mouths, further rise of the sea-
level, and possibly changes in the tidal-current
pattern in the North Sea, might have contributed
to this event. Recent investigations have shown
that the bulk of the sands derived from erosion of
the backshore and dune areas remain in the ad-
jacent intertidal and nearshore areas, thus con-
tributing to a temporarily declining coastal
gradient (Vellinga, 1982).
In contrast to the southern part of the area, the
coastal erosion in the northern part was not
parallel to the present coast.
In the northern area, with its relatively highly
elevated Pleistocene nuclei, at least three
inlets developed from south to north.
Initially, these inlets migrated southward and
again northward in the final stage of their ex-
istence. Due to this process, the coastal area
lost its stability and retreated rapidly. Re-
cession rates of 10 mly have been registered.
4. FUTURE COASTAL DEVELOPMENT
An understanding of the role played by physical
processes in the past will contribute towards
predictions of future developments of the Dutch
coastal area. As described in the preceding sec-
tion, the present beach erosion is due mainly to
the progressive sea-level rise and a steepened
coastal gradient. If we assume climatic con-
ditions comparable to today's, coastal erosion
will decrease and in the long run will eventually
stop because the relative rise of the sea-level
will persistently decrease, as can be inferred
from Fig. I. However, recent measurements in the
Wadden Se~ area (Brolsma, 1982), indicate a re-
lative mean sea-level rise of 0.14 m during the
last 100 years, which might suggest a dis-
crepancy when compared with the relative sea-
level rise of 0.50 m during the last 1500
CI4-years, as derived from the graph published
by Van de Plassche (1982). Besides an expected
retardation of the relative mean rise of the
sea-level, the fact that the advancing shoreline
is approaching more resistant Pleistocene de-
posits in the subsurface of the coastal area
means that another retarding effect will be
exerted on future coastal erosion.
A reversal from erosion to accretion will only
occur if - as reconstructed for the situation
of 5000 Cl4-years ago - a complex of submarine
bars composed of coarse-grained sands develops
off the coast. This would protect the coastal
area against severe storms and wave attacks,
and the present coastal gradient will become
less steep. This can only occur if sufficient
coarse-grained sand is transported coastwards.
In contrast to 5000 Cl4-years ago, this coarse-
grained sand will not be supplied by the
rivers, whose current velocity has meanwhile
been reduced significantly. Furthermore, it is
doubtful whether a supply of such sands from
the present sea floor is a realistic possi-
bility.
On the other hand man must be prepared to
struggle hard to protect the coastal area from
further deterioration by means of sand suppletion
and by the installation of coastal defense con-
structions, such as groins. However, we have
learned from centuries of experience with our
coasts that restriction of coastal defence
constructions to a local scale, will lead to
still more vigorous erosion of other parts of
the coastline. Finally, the present and ex-
pected trends in the physical factors governing
coastal developments in the NW part of The
Netherlands, make it apparent that the coastal
erosion will not terminate in the coming decades -
 175

or even long after that - and that only steps

providing a natural sheltering of the coastal
area from heavy wave action might induce a
possible reversal of the present trend.

REFERENCES

Beets DJ, Th B Roep and J de Jong (1981) Sedi-

mentary sequences of the sub-recent North Sea
coast of the Western Netherlands near Alkmaar.
J. Sedimentology Spec. Publ. I A S 5.
Berg JH van den (1977) Morphodynamic devel-
opment and preservation of physical sedimentary
structures in two prograding recent ridge and
runnel beaches along the Dutch coast. Geologie
en Mijnbou, 56, 185-202.
Brolsma P (1982) Sedimentbalans Waddenzee.
Nota WWKZ-82. Dir. 'Waterhuishouding en Water-
beweging, distrikt Kust en Zee, Adviesdienst
Hoorn.
Jelgersma S, J de Jong, WH Zagwijn and
JF van Regteren Altena (1970) The coastal dunes
of the Western Netherlands; geology, vegetational
history and archaeology. Meded. Rijks Geologische
Dienst, N S 21, 93-167.
Jelgersma, S, E Oele and AJ Wiggers (1979)
History and coastal development in ThE: Nether-
lands and the adjacent North Sea since the Eemian
In: Oele E, et al. (ed). The Quaternary History
of the North Sea, University of Uppsala.
Lamb HH (1961) Atmospheric circulation and cli-
matic changes in Europe since 800 AD. INQUA
Warsaw 2, 291-318, Lodz, 1964.
Mulder EFJ de, JHA Bosch (1982) Holocene stra-
tigraphy, Radiocarbon datings and paleogeography
of Central and Northern North-Holland (The Ne-
therlands). Meded. Rijks Geologische Dienst, 36-3.
Plassche 0, van de (1982) Sea-level change and
water-level movements in The Netherlands during
the Holocene. Meded. Rijks Geologische Dienst,
36-1.
Schoorl H (1973) Zeshonderd jaar water en land.
Bijdrage tot de historische Geo- en Hydrogr~fie
van de Kop van Noord-Holland in de periode + 1150-
1750. Verh. Kon. Ned. Aardr. Gen. 2.
Straaten LMJU, van (1965) Coastal barrier depo-
sits in South- and North-Holland, in particular
in the areas around Scheveningen and IJmuiden.
Meded. GeoI. Sticht. 17, 41-75.
Vellinga P (1982) Rekenmodel voor de verwachting
van duinafslag tijdens stormvloed. Waterloop-
kundig Laboratorium Delft, nota M 1263, IV.

WAVE-GENERATED WATER FLOW THROUGH A POROUS SEA BED
D.H. SWART and J.B. CROWLEY (Sediment Dynamics Division, NRIO)
ABSTRACT
A theoretical model is developed for determining
the flow in a sandy bed which results from cur-
rents induced by wave action. The flow model is
two-dimensional and assumes that the perme-
ability of the bed in the horizontal direction
differs from the permeability in the vertical
direction. Horizontal and vertical components
of the flow velocities within the sandy bed can
be expressed by Darcy's Law, assuming in-
compressible and viscous fluid. Pressure vari-
ations at the sea bottom resulting from wave
action are obtained from vocoidal theory. Vo-
coidal water wave theory was developed for all
non-breaking waves and is most applicable in the
shallow water regions.
The theoretical model is extended to include the
dissipating effects which this flow in the bed
has on the waves as they move shorewards.
INTRODUCTION
As waves approach the shore the fluctuating
pressure at the sea bed due to the passage of
the waves generates a flow of water into the
sandy bed itself. This means that the waves
pump water into the bed which, after some re-
sidence time in the bed, flows back from the bed
into the main water body. This phenomenon is
called percolation and is of importance to
coastal engineers since it leads to an addi-
tional wave height reduction over and above that
177
caused by bed friction and refraction. On the
other hand, the volume of water pumped through
the bed will have some bearing on the nature and
extent of biotic communities living in the sub-
strata. The prediction of percolation was first
treated by Putnam in 1949. Since then various
researchers have done further work. However, the
bulk of the work performed to date has been based
on a simplified sinusoidal wave theory, which is
not applicable in shallow water. This paper dis-
cusses the earlier work, after which a new theory
is developed on the basis of vocoidal water wave
theory which was developed by Swart in 1977 and
which is applicable in all water depths from deep
to shallow water.
2 REVIEW OF AVAILABLE LITERATURE
2.1 General
The work of researchers who have worked on per-
colation can be classified into three groups ac-
cording to the assumptions made regarding the per-
meability of the bed and the viscosity and rigidity
of the bed layer.
Firstly Putnam (1949), Reid, Kajiura (1957), Liu
(1973), Liu (1977) as well as Peregrine, Packwood
(1979) considered a fluid layer overlying a
porous medium of uniform permeability.
Sleath (1970), Moshagen, Torum (1975) as well as
Madsen (1978) assumed that the porous bed is
anisotropic, where the permeability of the bed
in the horizontal direction is not equal to that
in the vertical direction, though both are
178

assumed constant throughout the bed.

Gade (1958) as well as Dalrymple, Liu (1977) de-

veloped their theories in a two-layer fluid
},------
system, with the density and viscosity of the
lower layer greater than the upper layer. In
fluid
addition a non-rigid bed was assumed.

The solutions obtained in the three cases are dif-

ferent and are discussed separately in Sections
2.2 to 2.4. These sections should be read in
combination with Table 1, which summarises the
assumptions made and types of solution for each
of the references studied.

2.2 Group 1 - Isotropic Conditions in Sandy Bed ~~en studying the equations of motion one can
This group of researchers made the following as- see that:
sumptions: (i) Darcy's law is satisfied by all re-
searchers in this group.
(i) Monochromatic small-amplitude waves. (ii) The convective accelerations are ignored
(ii) A horizontal rigid bed. except by Peregrine, Packwood (1979), who
(iii) Incompressible skeleton and fluid in include them.
the bed. (iii) Reid, Kajiura (1957) and Liu (1973)
(iv) Sinusoidal pressure at the bed except that included bed friction.
Peregrine, Packwood (1979) associated theirs
with a solitary wave and a bore. Further details are contained in Table I.
(v) viscosity is neglected in the upper layer
except by Liu (1973, 1977) who considered 2.3 Group 2 - Anisotropic Conditions in
viscosity in the bed boundary layer but not Sandy Bed
in the free fluid above. This implies that The following assumptions were made:
irrotational flow is assumed in the upper
layer except near the boundaries where Liu (i) Monochromatic small-amplitude waves.
(1973, 1977) added a rotational fluid. (ii) A horizontal rigid bed, except Madsen
(vi) The density of the lower layer is greater (1978), who assumed a non-rigid bed.
that that of the upper layer. (iii) Incompressible skeleton, except Madsen
who considered a compressible skeleton.
For incompressible flow and no normal strains in (iv) Compressible fluid in the bed, except
the bed the continuity equation takes the form of Sleath (1970).
the well-known La Place equation: (v) The pressure fluctuation at the bed
is sinusoidal .
... (2.1) (vi) The upper layer is frictionless.
(vii) The density in the lower layer is
where 0 = potential function. greater than that in the upper layer.
 179

The continuity equation for flow in a porous 2.4 Group 3 - Sandy Bed Considered as Two
medium in its most general form reads: Separate Layers

Gade (1958) and Dalrymple, Liu (1978) made the

following assumptions:
(i) Monochromatic small-amplitude waves.
Kx and Kz are the permeabilities in x- and z- (ii) A horizontal rigid bed.
directions respectively. yw is the unit weight (iii) Incompressible fluid and skeleton.
of the pore fluid, n is the porosity of the (iv) The pressure variation at the bed is
porous bed, ~x and ~z are the normal strains sinusoidal.
defined to be positive as elongations, t is time, (v) Gade neglected friction in the upper
S is the compressibility of the pore fluid and layer, whereas Dalrymple and Liu
P is the wave-induced pore pressure. The equa- included it.
tion is treated as follows by the various re- (vi) The density of the lower layer is
searchers: greater than that of the upper layer.
(i) Sleath (1970) did not include the com- (vii) The flow is irrotational except near the
pressibility of the pore fluid, whereas bed.
Madsen as well as Moshagen, Torum (1975)
included it. The equation of continuity is treated by Gade
(ii) Madsen (1978) included the normal strains (1958) and Dalrymple, Liu (1978) as follows:
whereas they were neglected in the two (i) They included the compressibility of the
other papers. pore fluid.
(ii) The normal strains were neglected in both
The equation of motion in the horizontal direc- cases.
tion is given by:
As regards the equations of motion, one can see
u3u -30 32 u 32 w
p (3u + ~+ w3u) 3x + ll(~ + azz-) that for both researchers:
3t 3z
(i) Darcy's law is not satisfied.
u, ware the horizontal and vertical orbital (ii) The convective accelerations are ignored.
velocities respectively, p fluid density, 11 is (iii) The friction term has been included.
the dynamic viscosity of the fluid, 0 the wave-
induced pressure in the water. Further details are contained in Table I.

The following should be noted: 2.5 Discussion

(i) Darcy's law is satisfied by all. Putnam (1949) developed a theory for the loss of
(ii) Moshagen, Torum (1975) included the wave energy as a result of currents induced in a
convective acceleration. permeable sea bed by wave action. Unfortunately
(iii) Sleath (1970) ignored friction. he confused wave height with wave amplitude in
(iv) Madsen (1978) linearised the equation and the description of wave-induced pressure at the
ignored acceleration. bed. The error resulted in an over-estimation
of energy dissipation by a factor of four.
Further details are contained in Table I.
180
Reid, Kajiura (1957) studied damping of gravity
waves over a permeable sea bed using a more
rigorous approach than Putnam. They pointed out
Putnam's error.
Sleath (1970) used slightly different conditions
than Putnam, the permeability of the bed in the
horizontal direction was not equal to that in
the vertical direction. In the special case
when the permeability is equal in all di-
rections his solution reduced to the same
equation as derived by Putnam (after cor-
rection of Putnam's error).Sleath's data
showed considerable deviation from the theory,
especially for the test with coarse bed material
(Keady, 1978). He attributed it to calibration
problems with the measuring equipment. Dal-
rymple, in a discussion of Sleath's work, showed
that Sleath's equation over-predicted the pres-
sure in the bed and that a slight phase shift
should result from the interaction of the wave
field and the underlying porous medium. Sleath
ran several experiments to verify that there
should not be a change in phase. There seemed,
however, to be a trend towards a phase shift with
depth. His conclusion that the stratification of
the bed has a marked effect on the pressure dis-
tribution was verified.
Reid, Kajiura (1957) showed that Darcy's law was
an adequate approximation for unsteady porous
flow. They over-simplified by using an irro-
tational wave solution above the region of vis-
cous porous flow with continuity of pressure and
vertical velocity at the interface. This leads
to a discontinuity in the horizontal velocity
component at the interface. Earlier Putnam
(1949) only specified the continuity of pressure
at the interface and correspondingly he found
discontinuity of both horizontal and vertical
velocity components at the interface. A com-
parison of the results obtained by Putnam (1949)
and Reid, Kajiura (1957) respectively indicate
that this severe over-simplification as intro-
duced by Putnam did not affect the results sig-
nificantly. In a slightly viscous liquid the
potential component of the velocity field domi-
nates in the main interior region whereas the
rotational component dominates within the boun-
dary layer near the interface (see 0 M Phillips,
1966, for a discussion). A rotational field
therefore has to be added to the potential flow
field inside the free surface boundary layer so
as to satisfy the condition of zero tangential
stresses. Liu (1973) attempted to improve on
the earlier theoretical work by applying a non-
slip condition at the water-sediment interface
requiring continuity of pressure and velocity
components as a boundary condition at the
interface.
Moshagen, Torum (1975) developed an alternative
theoretical method for predicting wave-induced
pressures in permeable sea beds by adding the
condition that the bed water is compressible and
the bed material rigid. This implies that the
pressure response in the sea bed changes dras-
tically as a function of the permeability of the
bed. Physical justifications are not clear for
these assumptions (Liu, 1977).
Liu (1977) extended his earlier results (Liu,
1973) by including a permeability stratification
in the bed. He found that the viscous damping
rate is insensitive to permeability stratifi-
cations. The wave-induced pressure and its
gradients, however, are affected significantly
by stratification. Furthermore, the wave char-
acteristics are only slightly influenced by the
presence of the permeable bed as long as Kl/v
is small. The intensities of the velocity and
pressure in the porous bed depend very much on
the permeabilities of the porous media as well
as the thickness of the upper layer. Liu (1977)
gave a good account of the effects of a layered
bed and considered a range of permeabilities.

In one example, for which the lower conductivity
was 10- 5 times that of the upper layer, Liu
found a difference in the vertical velocity at
the free fluid/porous medium interface of the
order 12% compared with the unstratified case.
To ensure the applicability of Darcy's law the
Reynolds number is calculated on the basis of
typical design wave data for various soils. A this motion does not have to be deeper than of
Reynolds number for flow through porous media
is defined as (Bear, 1972): Re = Umax d/v where
d is the average grain diameter, V the kine-
matic viscosity of the fluid and Umax the maximum
horizontal flow velocity in the porous medium.
According to Liu (1973) the maximum velocity
occurs at the interface between the free fluid
and a porous bed is equal to Umax = kag K/V
where K is the permeability of the soil. It is
shown that the largest value of the Reynolds
number corresponding to coarse sand is less
than 10. According to Bear (1972), Darcy's
law is then valid for most practical problems.
Madsen (1978) presented a general theoretical
analysis of flow, pore pressures and the ef-
fective stresses induced in a porous bed by
water waves. The effect of stratification of
the soil was found to have a noticeable effect
on the nature of the wave-induced effective
stresses only for porous beds consisting of
soils coarser than fine sands. One of the major
simplifying assumptions made in the analysis of
percolation was that of treating the waves as
plane progressive waves described by the linear
wave theory, except in the case of Packwood,
Peregrine (1980). They considered the attenu-
ation of both a solitary wave and a bore pro-
gressing over a porous bed of infinite and
shallow depth respectively. This work indi-
cated that any further work relating to non-
linear wave propagation on real beaches should
consider the porosity of the bed as being
equally as important as the dissipation due to
the bottom boundary layers. The effects of
181
stratification of the porous medium have not ge-
nerally been considered, effects which can be
quite pronounced in sedimentary materials.
Gade's paper was an attempt to discuss the ef-
fects of a non-rigid bottom in which the sediment
behaves like a viscous fluid, the penetration of
the order of centimetres. Errors in the experi-
mental data obtained by Gade were caused by:
(i) frictional resistance of the side walls;
(ii) the irregularity of the generated wave
profile and its deviation from a
sinusoidal profile; and
(iii) reflection from the end wall.
Dalrymple, Liu (1978) developed two models for
water waves propagating over a very viscous
fluid, which extended the analysis of Gade
(1957) to deeper water conditions and included
the viscosity in the upper fluid. A boundary
layer model was developed and shown to agree
well with the complete model. The shallow water
model was applied to Gade's model tests and
agreed well with data for wave damping and wave
length, whereas the predicted phase between the
interface and free surface displacements were in
error by about a factor 2, as was Gade's analy-
sis. The conclusion that the principal method
of energy transfer to the lower layer is that
the pressure of the surface wave working on the
lower fluid is drawn, despite the fact that Gade
(1957) neglected the interfacial shear stress
in the upper fluid.
3 NEW DERIVATION USING VOCOIDAL WATER WAVE
THEORY
3.1 General
Vocoidal water wave theory was developed from
first principles by Swart (1977) and is valid
for non-breaking waves on all relative water
depths from deep water to the br~aker point.
Swart and Loubser (1978) compared the theory
182

K
along with 12 other commonly used wave theories x dill
u
s V dX
to more than 500 experimental data sets on various
••. (I )
wave characteristics. It was found that vocoidal
K
z dill
theory was in better agreement with the data than w
s V az
any of the other theories tested. Since perco-
with Kx and Kz the permeability in the x-
lation is of prime importance in the area imme-
and z- direction, V the kinematic viscosity and
diately seawards of the breaker line where the
III the velocity potential.
small-amplitude wave theory which is mostly used
in calculations is not in good agreement with
The continuity equation for incompressible flow
data, a derivation of the flow in a sandy bed
reads:
and the effect thereof on the wave properties
was done for vocoidal theory. A short summary
dU dW
of the derivation follows, after which the ax-s + az
s
o •.• (2)
solution is used to determine the flow of water
in the sandy bed and the effect of this per- From equations (I) and (2) it follows that:
colation on the wave attenuation.

+ o • •• (3)
3.2 Assumptions
The following assumptions are made:
Equation (3) is the governing equation used in
(i) A horizontal bed at z =0 with mono-
the rest of the study.
chromatic finite amplitude vocoidal
waves. The boundary conditions used to solve equation
(ii) The permeability in the horizontal di- (3) are the following :
rection K and the vertical direction
x
Kz is not equal, though both are assumed At z 0, that is at the interface,
constant throughout the bed.
(iii) The motion is two-dimensional p
g(d - z)
p
(iv) The flow in the bed is viscous.
(v)
(vi)
The fluid in the bed is incompressible.
The pressure at the bed varies as in vo-
gn + c
2
(
u2 +
2c 2
W2
- c) Id+n
U
0
••• (4)

coidal theory with time and horizontal At z = -d sb ' that is the lower extremity of the
location. sandy bed,

3.3 Solution of Governing Equations w o ••• (5)

The governing equations for the flow in the sandy
bed are the equations of motion in the hori-
zontal and vertical directions and the contin-
uity equation. When convective accelerations
are neglected, which is a reasonable assumption,
the equations of motion simplify to the so-
called Darcy's law, namely,
III is the velocity potential, P is the pressure,
p is the density of the fluid, g is the accele-
ration due to gravity per unit mass, d is the
mean water depth, n is the wave profile, c the
wave celerity, u and ware the horizontal and
vertical orbital velocities in the water body
and d sb is the thickness of the permeable bed.
 183

By expressing the governing equation (3) as a The total volume of water drawn into the bed
half-range Fourier expansion and by using the during the passage of each successive wave is
boundary conditions (equations (4) and (5)) it equal to the maximum horizontal flow rate from
is found that the velocity potential in the equation (7) above.
sandy bed which satisfies these equations is:

An a i cos~ ki(d sb + z)cos kix Values of qx/(oxzgd) are tabulated for con-
oL z venience of application in Table 2.
R
(6)
i=1
cos~rc kid sb
z As an example of the volume of water &z pumped
In the above equation 0 is the maximum value of into the sandy bed by wave action consider a wave
the velocity potential 0 at the level z and the with a deepwater height Ho of 2 m and a wave
coefficient a i is defined by period of 16 s which is shoaling on an underwater
n slope of 1:100. For the purpose of calculation
o (L a. cos k.x + a )
1- 1- 0 it is assumed that the bed consists of sand with
i=1
a relative uniform grain size of I mm and a
standard deviation o¢ of 0,1. Isotropic con-
3.4 Flow through a Sandy Bed ditions are assumed. The permeability K can
be calculated by using a relationship by Krumbein,
The horizontal flow qx through a porous bed of Monk (1942) which is in good agreement with
thickness d sb ' that is the total horizontal flow measured permeabilities.
from the interface to the lower limit of the
sandy bed, is given by 0
qx =
-d sb
u
s
dz S
with the median grain size Dso in M.
where Us is given by equation (I).

Substitution of 0 from equation (6) into equation

(I) finally results in

An
°xz oi=1L a. tanh (k. ~ d b) Sin k.x
1- Kz s 1-
[;? (7)

Two examples of the time-variation of the total

maximum horizontal flow rate are given below.

.r . . ...
0·.
'.
0-15,. HAl • 0-'

·.Xl ....

O·J XX '0 If..

0-' .. '0 \

~:~:' ......................... ~
CS'"xz gd 0'1 0-' o·J 0-4 0-5 •• O.'...... "O.7...
~ 0·' 0-' "'0 i-
.0-1 ••••••••••• .. T
'0, ..
• 0'1 '0 ..

-0'1 1\ ••••

• 0'4 \ . -

.0'11
Ix l'
' -..l · ·
184

The value obtained for K K K Using these data one can see that the total
x z
volume of water being pumped into the bed inside
All wave properties given below were calculated a say 300 m wide surf zone daily could be of
using vocoidal wave properties as tabulated by the order of 50 m3 /day per metre length of shore-
Swart (1980). The flow rates were obtained from line, or 50 000 m3 /day per kilometre of coast-
Table 3. The results are given in m3 /day/m 2 of line. It could be argued that the method derived
surface area of the bed. herein is not applicable within the surf zone;
on the other hand the whole flow phenomenon
within the bed is governed by the pressure
d(m) T * H/d Aid - /day/m area)
q z (m fluctuation at the sea bed, which is predicted
3 2
c
fairly accurately by vocoidal theory even within
200 3,5 0,01 1,97 0,008
the breaker zone.
100 5,0 0,019 3,72 0,036
50 7, I 0,037 6,17 0,076
3.5 Dissipation of Wave Energy due to
20 11,2 0,099 10,65 0,158
Percolation
10 15,8 0,261 15,84 0,258
The pumping of water into the sandy bed due to
6 20,5 0,622 23,42 0,367
the wave-induced pressure variation at the bed
3,5 26,8 1,244 38,39 0,423
uses wave energy, which in turn leads to a re-
duction in wave height with distance travelled.
!
*T = T(g/d)2 The average rate of dissipation of energy Dp per
c
unit length in the direction of motion for a
For a breaking wave with a breaker index unit crest width is
(H/d)B = 0,6 the total flow into the bed, again
expressed in m3 /day/m 2 area, is given as follows acp + q acp) dzdx
(q x ax (8)
z az •••
across the breaker zone:
Substitution of the relevant expressions into
H/d 0,6. equation (8) and the use of the orthogonality
properties of the Fourier expansion leads to
T d(m) Aid q (m 3 /day/m 2 area) the following expressing for the average
c z
rate of energy dissipation:
20 6,28 22,61 0,356
25 4,02 28,52 0,262
a - 11
E ia~ tanh (ki~ ~ drKJ
30 2,79 34,41 0,213 -P"x"7f xz
,h2
'I' sb •• (9)
i=1 z
40 1,57 46,18 0, lSI
If the sandy beach exceeds a thickness of
50 1,00 57,95 0,123
30 per cent of the wave length (that is,
Idsb/AI > 0,3), the dissipation reaches
its maximum value and does not increase sig-
nificantly with increasing thickness of the
sandy bed. Therefore the maximum rate of energy
dissipation Dp can be written as

-7fpa
xz A
i2 epb ••. (10)
 185

where given below. The wave properties were again

calculated from Swart (1980) and the decay
II
ep b = L:
i=1
i a:~ ... (II) coefficients a p were calculated from Table 2.

Equations were obtained by curve-fitting for e pb

in terms of the boundary conditions T, d and a T2
d(m) T H/d A/d ~
H (see Appendix A). c 0-
xz

200 3,5 0,01 1,97 0,26 1,000

To determine the change in wave height due to
100 5,0 0,019 3,72 1,43 0,998
percolation the energy balance over one wave
50 7, I 0,037 6,17 6,38 0,994
length per unit wave crest is used:
20 11,2 0,099 10,65 18,69 0,985
10 15,8 0,261 15,84 38,70 0,976
dE D
dt + P
° ••• (12)
6 20,5 0,622 23,42 54,40 0,970
3,5 26,8 1,244 38,39 48,0 0,966
The wave energy E is given by E = pe t gH 2 (e t is
an energy coefficient). It was found that it
It is, therefore, obvious that the effect of
is a good approximation to assume that the wave
percolation on wave height is only relatively
height H decays exponentially with time as a
mild, if one takes into account that the wave
result of percolation, that is,
in the above example lost only 3,4 per cent of
height over a distance of nearly 20 km. On the
H .•. (13)
other hand, more than half of the reduction in
wave height took place over the last 2 km and,
where HI is the initial wave height and a p is a
in fact, inside the surf zone the wave would
decay coefficient.
lose even more energy. For example, if the
wave in the above example broke as a spilling
Substitution of the relevant quantities into
breaker with a breaker index (H/d)B = 0,6 it
equation (12) and taking averages over a time
would lose 1,3 per cent of its height over a
step which is small with respect to the wave
distance of just 60 m in a water depth of I m.
period T, leads to the following expression
for a p
4 SUMMARY AND CONCLUSIONS

0- ... (14)
xz
A new framework was developed on the basis of
vocoidal wave theory for the prediction of
For convenience of application values of percolation and its effects, which is valid for
(a p T2)/0- are tabulated in Table 2. both non-breaking and breaking waves. The re-
xz
sults obtained from the theory are consistent
Using the same deepwater wave with a height of with results of earlier researchers and can form
2 m and period of 16 s as an example, decay the basis of new theoretical developments in the
rates as influenced by percolation losses are field of the refraction of higher-order (vocoidal)
waves as well as of research into the behaviour
of biotic communities in the substrata.
186

5 REFERENCES APPENDIX A

Bear J (1972) Dynamics of fluids in porous

media. Elsevier, Amsterdam, 764 pp.
Dalrymple RA,and Liu P L-F (1978) Waves over
soft muds: a two-layer fluid model. J. Phys.
with
Oceanog.8, pp 1121-1131.
I
Gade HG-(1958) Effects of a non-rigid im- neff = (1.0 - - -- +
8xr3P 128xrgp)
permeable bottom on plane surface waves in Irrxr 3P)
shallow water. J. Mar. Res., 16, pp 61-82.
Keady, DM (1978) Dynamics of-Wave-induced a = -0.6314 for r3P 2 2.97.
currents in sediments. Nat. Tech. Info.
b 5.2558
Service, U S Dept. of Commerce, Springfield,
Va, 22161.
Krumbein WC and Monk GD (1942) Permeability as
a = -.0004
a function of the size of parameters of uncon- for 2.97 < r3P < 31.0
solidated sand. Amer. Inst. of Mining and b 3.2472
Metallurgical Engineers, Tech. Pub. No.1492.
Liu P L-F (1973) Damping of water waves over
porous bed. J. Hydraul. Div. ASCE, 99, a = -0.0067
pp 2263-2271. -- for r3P > 31.0
b 3.3104
Liu P L-F (1977) On gravity waves propagated
over a layered permeable bed. Coastal Eng.,
I, pp 135-148.
and
- Madsen, OS (1974) Stability of a sand bed under
A
breaking waves. Proc. Conf. Coastal Eng. 14th For d 2:. 20
Copenhagen, Vol.2, chapter 45, pp 776-794.
Moshagen H and-To rum A (1975) Wave-induced pres-
sures in permeable seabeds. J. Waterways Har-
r3P = al+ bl( Hid) for Hid < 0.53
bour and Coastal Eng. Div. pp 49-57.
Packwood AR and Peregrine DH (1979) The propa- with
gation of solitary waves and bores over a A
4.75 - 0.25(d)
porous bed. Elsevier, Amsterdam, Coastal Eng.
A 2 A
3, pp 221-242. 0.1698 (d) + 0.97 (d) -353
- Phillips OM (1966) The dynamics of the upper
r3P = a2 + b2 (Hid - 0.53) for Hid> 0.53
ocean. University Press, Cambridge, UK.
Putnam JA (1949) Loss of wave energy due to with
percolation in a permeable sea bottom. Trans.
Am. Geophys. Union, 30, pp 349-356. a2 (r3P)H/d = .53
Reid RO and Kajiura-K (1957) On the damping of
gravity waves over a permeable seabed. Trans b2 = -1.147 (~- 12)1>536
d
Am. Geophys. Union, 38, pp 662-666.
Sleath JFA (1970) Wave-induced pressures in
beds of sand. J. Hydraul. Div., ASCE, ~, For ~d < 20
pp 662-666.
A)b 3
Swart DH (1978) Vocoidal water wave theory a3 (d
Volume I, derivation. CSIR Research Report 357,
Council for Scientific and Industrial Research,
where
Stellenbosch.
Swart DH and Loubser CC (1979) Vocoidal water 0.00033 (Hid) -1.6459 for H/d<0.117
wave theory Volume 2, verification. CSIR Re-
search Report 360, Council for Scientific and a3 = 0.00073 (H/d)-1.2762 for 0.117< H/d2 0.538
{
Industrial Research, Stellenbosch. 0.5989 (H/d_0.5)1.808 for Hid> 0.538
Swart DH (1980) Tables for the application
of vocoidal theory. CSIR Report T/Sea 8010, b3 = .03338 In (~)
a3
Council for Scientific and Industrial research,
Stellenbosch.
c - (51.82 (Hid) - 0.25 for Hid < 0.33
~27.2146 - 29.1589 (Hid - 0.53) for Hid> 0.53
For r3P < I, r3P
~
t-<
I:'J

Continuity Hotion

""
. ::
""
D
EE -, K.x a 2p a 2p YC«l~ ~ _ ywa, cx+Ez) Cau2 a2 w
""
D
. · ~ K ~ + a;t - K at - -at U a;z+~
>- .~ ,.. .~
U "00
">o PoPo~
§ §
. .. >-~ o
• 0 .~ ~ "
. 7.l f.:
.~k
."D
:1"""
'D ~ o
~ .. ~ e...~ N • .~
.. .....~
•
.::.~
0 U""
u.o
." OU o o -~
0-:; ...
e u '" ~
~
.~ " o "
~ i:§ E~ " u
Po" ~~
. t!
"
. . . .
.~ i"
o ~
,,~

'~E Po.
e .. ~)C~ -
,~

o u
·· .
! i .. " .. ""
o"" ~ : 3~
. ." Q
QlJ, ~ ~
I. .~ ".. f... iH
" .e :ll
'"
Putnam 19491 I HSAWI I I I

Reid con-
19571 I HSAW 1 X stant I I I I I
Kajiura

Liu 19731 I HSAWI I I I I I I

Liu 19771 I IKx.K z HSAW 1 I I I I

Packwood SW SW I I bed
19791 I B B I ignored I I I
Peregrine

Sleath 19701 I IKx.K z HSAWI I I

Hashagen density
19731 I IKx.Kz very I I I I
forum small

"Hadsen 19781 I I Kx.K z HSAW I I I I I I I

No
Gede 19581 I HSAW urfacel I I I I
pres-

Dalrymple
19781 I HSAWI I I I I
liu

Compressib iIi t y: I incompress ib 1 e Type of wave: HSAW monochromatic rIow: irrotational except near the bed
C compressible small ampl i tude
Pressure: 5 sinicoidal waves
SW - solitary SW sol i tary Nye

B bore B bore

00
....,
188

TABLE 2
PERCOLATION DECAY PARAMETER
_rTl__
0(

6"'2-

.....................................................................................................
1.0 • 0310
.0 I • n?
.ooon
.05 •In
.0000
.15 .20 .OJ
.25
.000 I
• O. .35 • <0 • JO

2.u .09"1 .0000 .0000 .0000 .0000 .0000

3.0 .2Ztt8 .0122 .012' .0125 .0127 .0129 .0143 .01 bl .0202
<.0 .42'-111 .52l. .SZJ4 .~2Jb .5239
.5244 .5295 .535S .54~O .StlOJ .5&1J • bOlta .6)93
5.0 I 6~b4 Z.6908 Z.btUU 2.6858 2.6632 2. ",806 2.bb9Q Z. 650 1 Z.6219 2.6009 2.5872 2.5803 2.S 79,

··
•• 0 o 7~)9 6.1T4 b .166 6.15b 6. J 46 h.135 '.077 b.OOO S.H8l S.76tS 5.65'1 5.55. 5. ,,58
7.0 .8114 10.115 10.097 10.077 10.055 10.0l2 9.901 9.739 9.492 9.243 8.9'ib 8.755 B.520
8.0 • 'thRi9 14.059 14.029 13.995 13.958 13.91& 13.'93 13.437 13.020 12.'07 12.195 Il.82S 11.52"
9.0 .Cllb4 IT .596 17.560 17.522 17. ,83 17,; 443 17.234 19.551 19.940 19.9Z7 19.928 20.0'0 ZO.l45
10.0 .91))9 21.'35 ?1.5A2 21.52' 21 _,,68 21.410 ?J.730 23 .... 5'1 ~2. 9&9 22.'45 22.572 22.7.5 23.225
11.0
· l. 04 lit
·
• 25.94>1 25.001 2S. b 7& 25.550 2h.1b3 27.308 26.tiSO 2b.67C6 26.644 26.1:11:15 27 .17' 27.192

··
Il.O 101721 29.734 29.58' 29.437 29.925 30.92 I 3 I. 01< 31.161 31.643 31.'95 30.915 30."69 30.400

·
13.0 1.3"05 • 33.44b JJ.2~0 32.924 34.1 H 34.552 35.317 36.~Z2 36.899 3~. 761 34.7'iJ 3< 01 38 34.089

·
14.Q 1.3730 37.086 36.905 37.09. J7.689 JA.087

···
40.489 42.257 41.S86 40.299 39.17' J8.229 J7.7I1

·
15.0 1.3hbO • 40.662 J9.JJ~ 40.31t0 41.111 '1.871 '6.420 47 .259 46.711 .5.278 43.68b 42.J01 41.416
16.0 I.J595 44.185 43.,.49 4J.582 44.70S 46.012 52.156 52.751 52.438 50.528 48.412 46.521 45.204
17.0 1.)~J4 • 47.660 46.tlOI 46.tt"7 48.571 S.0.554 58.270 58.7JO 58.JJ8 56.0J7 5J.333 50.866 49.083

··
18.0 1.3<77 ·51.094 49.809 50.263 52.148 55.526 64.242 65.199 64.637 61.808 S8.439 55.J42 53.069
19.0 I.H2J • 54.493 52.712 ~J.d84 57.260 60.952 70.682 71.539 71.264 67.825 63.727 59.957 57.180
20.0 1.3H2 • 58.1'2 55.6J9 57.140 62.1 25 66.860 77 .59S 78.716 78.205 14.080

··
69.19' 64.121 61.4J3
22.0 1.3278 62.34 61.81 65.'2 72.61 80.19 92.88 9J.89 91.J2 87.28

·
24.0 80.69 14.75 70.43
I. JI9J 68.22 68.56 14.15 8,.95 96.14 108.2' 110.'0 106.79 10 I. 41 92.97
26.0 85.51 80.29
I. JII5 tJ .90 75.99 ts4.28 99.56 II 0 .81 125.86 128.18 123.32 116.48

···
106.10 97.28 91.03
28.0 I. J 0' 3 79.76 8'.IJ 96.00 116 •.81 127.11 144.88 1"7.18 14 o. 90 132.54 120.J7 109.95 102.65
30.0 1.2977 85.97 9J. OJ 109.56 135.77 145.10 165.24 167.38 159.55 149.76 135.63 123.58 115.17
32.0 1.2915 92.61 102.12 125.09 153.J4 164.8. 186.90 168.75 179.26 168.16 151.93

··
138.15

··
34.0 128.57
1.2"57 99.72 113.23 1'3.0 I 17 2.52 180.93 209.83 211.30 200.J7 1.7.70 169.27 153.70 142.87
36.0 1.2'03 107,33 12'.60 16) .59 19~. J5 201.62 234.00 2J5.04 2n.66 .2 0&.4 a

··
187.67

·
170.20 158.05
38.0 1.2752 115.'6 136.84 187.2 I 208.9J 223.53 259.41 260.03 246.16 230.27 207.14 187.65 114.11
40.0 1.2703 12' .1' 150.00 214.10 230.25 246.65 2~6.0? 286.49 270.89 253.31 227. 6' 206.06 191.02
42.0
·· 1.2658
··
133.39 16'.09 233.66 252.70 270.95 313.85 Jilt .2_ 296.85 277.53 2'9.23

·
H.O 225.'0 208.79
1.2'14 143.22 179.15 250.61 276.29

···
296.41 342.89 343.28 324.05 302.93 271.86 2'5.67 227.40

···
1t6. (I 1.2573 153.6' 195.19 272.62 300.98 323.00 J1J .16 37J.63 352.49 J29.52 295.52 266.88 246.85
'8.0 1.25JJ I" .68 212.25 295.6' 32,.75 J50.71 1t04.81 405.31 J82.19 357.27 320.23 288.99 267.12
50.0 1.2'95 176.3< 2JO. J5 J19.71 J53.60 379.53 437. ~3 438.31 413.13 386.2 I J45.97 312.02 288.ll

···
52.0 1.2059 188.65 247.68 3U.77 J81.51 .. 09.43 472.J6 472.65 445.JJ 416.37 372.75 J J5. 9 6
310.11
54.0 1.2424 • 201.61 273.72 370.8J 410.450 440.42 508.10 508.33 478.79 4J8.J6 400.55 360.19 332.80
56.0 1.2391 • 21S.25 J02.'0 397.86 440.43 472. 4 tt 545.16 545.36 513.50 469.87 429.38 386.52

·
356.29
58.0 1.2359 • 229.57 3J3.99 425.85 411.42 505.60 583.55 583.14 549.47 502.50 459.23 <13.13 J8 0.56
60.0 1.2328 • 244.59 36'.19 454.80 503.42 539.80 623.28 623.48 586.70 536.25 490.09 440.63 405.62

.....................................................................................................
1.0 • OJIO
.40 .45 .50 .55 .60 .70 .80 .90 1.00 1.10 1.20 1.30

2.0 .0961
3.0 .2288
5.0 .6964 • 2.5794 2.5636 2.5984 2.6099 2.6282
6.0 .7539 5.458 5.367 5.282 5.200 5.139 24.770

·
7.0 .8114 8.520 8.294 8.07, 5.920 9.087 24.614 27.732
8.0 .8689 11.524 11.231 10.948 16.650 17.103 23.978 21.681
9.0 .926' • 20.345 20.762 20.921 22.021 22.286 22.796 19.622 19.385
10.0 .9639 • 23.225 24.1 02 25.028 24,220 23.803 24.467 21.462 19.086

··· ·
1\ .0 1.0'" • 27.192 27. '93 27.926 2S.645 25.709 26.025 22.507 19.566 18.J74

····
12.0 1.1 721 30.400 30.872 31.295 27.733 28.216 28.025 24.0J9 20.480 19.128 lK.

··
Ob~
13.0 1.3605 31t.089 J4 .2JI J<.523 30.065 J 0.966 JO.JH 25.880 21.'81 20.161 18.988 18.084 17.J97
14.0 1.3730 37.711 J7.623 37.784 32.50J J3.949 3J. 0 17 27.9'7 23.095 21.394 20.099 19.119 18.J88

··
15.0 I. J"O 41.416 41.051 4 0.955 35.1 30 37.161 35.921 JO.266 24.687 22.785 21.35. 20.286 19.50'

·
16.0 1.3S95 <5.20' ".532 ".225 37.942 1t0.603 39.064 32.758 26.435 2'.314 22.731 21.561 20.718

··
17.0 1.3534 49.083 48.094 47.580 40.9U 44.277 '2.4J2 35.429 30.083 25.96d 24.217 22.934 20.551

··
18.0 1.3477 53.069 51.760 51.05J 41t .140 48.185 46.016 J8.27J 32.381 27.741 25.809 24.400 19.048

·
19.0 1.3'2J 57.180 55.556 54.669 47.533 52.3JI 1t9.R09 41.285 34.819 29.630 27.503 25.956 20.241
20.0 I.J372 61.4J3 59.501 SH.443 51.16. 56.764 53.84J 44.461t J7.395 3 I. 63' 29.300 27.60J 21.50 I

··
22.U 1.3278 70.43 67.91 6'.72 59.12 66.43 h?.1100 51.42 42.99 35.9d 33.20 JI.l6 24.22
24.0 1.3193 80 .29 77.32 75.80 b7.9Z 77.11 72.21 59.07 49.21 40.80 35.14 27.24

··
3~.52

··
26.0 1.3 II 5 91.03 • 7.50 85.72 17.57 88.81 82.67 b7.1t 3 5h.04 46.14 42.29 39.51 30.57
28.0 1.3043 102.65 98.55 96.1t 7 88. O. 101.53 94.02 76.53 'J.51 52.02 '7.52 44.32 34.24

· ··
30.0 1.2917 115.17 108. US
110. '5 99.39 115.27 10t-.29 86.42 71.66 Sfi.46 53.26 49.45
32.0 1.2915 128.57 123.20
120.45 111.56 130.02 119.52 97.14 80.57 65.5 I 59.56 55.04

· ··
J4t.0 1.2857 1'2.87 136.7.
I JJ.'5 124.56 1415.80 133.18 108.17 90.29 7J.23 66.4H 61.85
36.0 1.2803 158.05 IS I. 21
156.71 138. <0 162.61 ]49.13 121.38 100.92 81.'8 14.16 68.91

·
J8.0 1.215? 114.11 166.44 17&.9q 153.08 1 RO ."5 165.66 13'.07 Ill.55 90.9h ft2.bl 76.71
40.0 1.2703 191.02 182.48 203 •• 0 1'8.60 199.32 1I~3.4tS 149.t,l3 125.30 101011 9 I. 96 85.3b

··· ··
42.0 1.2"8 • Z08.79 199.31 230.0 I 184.95 219.23 202.01 1.6 oil 139.31 112.42 102.32 94.97
44.0 1.2." 2Z7.40 216.93 235.42 202.1' 2'0.1 A 22J.26 183.74 154.76 124.86 113.85 105.70

···
46.0 1.2573 246.85 235.31 250.97 220.18 2.2.19 24S.~5 203.00 171.85 138 •• 7 126.72 117.73
44.0 1.2533 207.12 254.<5 266.98 239.06 285.25 269.64 22'.1 0 190.83 154.06 141.19 IJI.32

·
50.0 1.2<95 • 288.21 214.35 283.43 258.78 309.37 295.7Z 247.27 212.01 171.30 157.52 1'6.74

· ···
52.0 1.2'59 310011 294.QA 300.32 279.3' 33' .57 324.01 272."2 235.77 1 'jIO. 71 176.08 164.41
54.0 1.24Z' 332.80 316.35 324.07 300.79 360.84 354.15 301.10 2.2.57 212.70 197.35 184.9b

··
56.0 1.2391 356.29 338.44 345.11 323.07 388.19 388.l7 332.53 293.01 237.81 221.92 209.17

·
. 58.0 1.2359 380.56 361.2' 36 •• 72 34'.22 416.b4 424. &9 3b7.b4 327.86 266.70 250.61 2J7 .81
60.0 1.2328 <05.62 38'.7. 388.89 370.23 .46.19 465.05 407.08 368.11 30U.Z1 28'.51 272.20
 189

TABLE 3
TOTAL MAXIMUM HORIZONTAL FLOW THROUGH POROUS BED
Cix.
(WITH DSB/LAMBDA = 1) O'-"2.S ct
........................................................................................................
1.0 .0 J 1 0
•01

.0000 .0000
• O? .03

• OODO
.0. .0, .10 .IS .20 .25 .30 .J5 .40

Z.O .0901 .0000 10000 .0000 .0000 • 0000

3.0 .2lt1H • 000 1 • 0003 .000" • ooo~ .OOOb .OU14 .00l2 .0032
'.0 • .. ZYI .ooot:! .001 b .0025 .0033 .0041 IOOAl .0 1 ~5 .0168 .0214 .0203 .03H .0310
5.u .6"10" .0018 .0036 • DOS .. .0072 .OOaq .0174 .0269 .0360 • u4~1 .05«'3 .0636 .0734
6.0 • 1~]9 .0026 .0052 .0078 .0103 • 012. .02SQ .03tH:J .0518 .0641:S .on • .0910 .10Ol
7.0 .lill" • 0032 • 0063 • U095 .0126 .OlSH .031 , .0474 • ObJ~ .0790 • D94u .1107 .li60
R.O .81i1j<;l • 0036 .0071 .0107 .0142 .0178 .O3~6 .05) .. .0712 .OHIlO .1006 • 1246
9.0 .926_ .143 •
.0038 .0071 .011 :) .0154 .01.2 ,031'19 .0626 .OM!:!3 .130_
10.0 .0041
.107t:1 .1553 .1 HI4
• 9~39 .OORI .0122 .0162 .0203 .0440 .0662 .OHtH .11 uJ .1327 .1552 .1782
II. V
·· 1.010'14 .0042 .0084 .0127 .0169 .0216 .0455 .061:13 .0906 01115 .1307 01470 01620

·
12.0 1.1721 .00lt3 .0087 .01)0 .OIFi .0229 .04t'1Q .070 I .0911 01 OijO 01 531

···
13.0 oil " .1t198
1.3'05 .00lt. .OOA9 .0130 .0 1 ~6 .0235 .04B2' • 07UB .0880 01203 0146)
14.0 1.3130 .0045
.135~ 010 . .
.0090 .011100 .0189 .02)9 • Olt9 1 .0bvJ .09.1H 01127 .122" .144.6 .1575
15.0 1. )b60 .0046 .0092 .0141 .0191 .0243 .0494 .0977
• Ob~3 01 04J .122' ,140b .1589

···
10.0 1.3'>'5 .0046 .0094 .011t3 .0194 .0247 .04RR .0807 .0899 01070 oIlZ2 .1368 01523
17.0 1.3534 .0047 .00qf, .0144 .0196 .0251 .0473 .0731 .0941 01 065 .1212 01337 01473
16. V 1. 3477 .0010 7 .0096 .0 tlt5 .019M .025,> .0553 .0703 .0916 01 0'5 .120, 01 3 14

·
19.0 1.3423 .14Z6
.0047 .009f1 .0145 .0201 .025b .05b9 .0702 .04.4 01 053
20.0 .1199 .12ijij .100 I
1.3372 • OOIoR .0097 .0146 .0203 .0260 .052' .0747 .0918

··
01 002 .1148 .1236 .1339
22.0 1. Je78 .0049 .0097 .Ot47 .0206 .0262 .04Ql .0730 .0920

···
.1043 .1124 .1196 01 JOb
24.0 1.3193 .0049 .0097 .0147 .0209 .0261 .0540 .0752 .0891 .1010 .1 076 .1162 .1250
26.0 1.3115 .0049 .0097 .0148 .0212 .0253 .0517 • 074 .. .0875 .0983 .1062 .1128 01201
28.0 1.3043 .0049 .0097 .0150 .0213 .02'2 .0525- .071 0

·
.OHb6 .0980 .1045 01 098 .1178
30.0 1.2<;77 .0049 .0097 .0152 .0210 .0324 .0713 .0865

···
• 0534 .0951 .102b .1087 .1155
32.0 1.2915 .0049 .0097 .0155 .0204 .0262 .OSI8 .0717
34.0
.0l;45 .0937 .0995 .1049 .1108
1.2~57 .0049 .0097 .0156 .0235 .0264 .0528 .Ob48 .0653 .0914
30.0 .0":194 .I 036 .1123
1.2H03 .0049 .0097 .01bl .0234 .0284 .0516 .0689 .0625 .0922 .0967 .I 022

·
38.0 1.2752 .0049 .11 08
.0097 .0162 .0217 .0283 • o~>l 0 .0705 .0811 .0903 • 0942 .I 01 • .1094
40.0 1.2703 .0049 .0097 .0161 .0218 .0281 .05<2 .0801 .0880 .0966 .I 030

···
• 04'>9h .1080
42.0 1 .2b~8 .0049 .0097 .0157 .0220 .0280 .0515 .066 I .0812 .0861:1 .09b1 .1019 .1081
44.0 1.2b14 .0049 .0097 .0185 .0222 .0294 .067S

···
.051' .0bOO .0869 .OY56 .1007 .1001
46.0 1.2573 .0049 .0097 .0192 .0237 .0291 .0514 .0672 .0787
48.0
.0870 .0949 .0994 .1074
1.2533 .0049 .0097 .0176 .0237 .0290 .0513 .0666 .0765 • v870 .0942 .0961 .1086
50.0 1.2-'5 .0049 .0097 .0159 .0236 .0289 .0512 .06b4 .0768 .0670 .0933 .0995

·
S2.0 1.24r,9 .0049
.1070
.0099 .0162 .0237 .0286 .0506 .06,9 .0770

···
.0868 .0909 .1007 .1081
54.0 1.2lt24 .0049 .0101 .0177 .0230 .0267 .0505 .0653 .0773 .0863 .0909 .0995 .1089
56.0 1.2391 .00lt9 .0104 .0179 .0236 .028' .0491 .0775
56.0 • Oft~5 .0657 .0923 .1006 .1096
1.2359 .0049 .0107 .0180 .0237 .0279 .0492 .0641 .0775
60.0 .OH49 .0937 .1016 .1099
1.2J2R .0049 .010q .0 I" I .022A .02fiO .04 ..... 3 .0647 .0774 .0839 .0949 .1023 .1084

.....................................................................................................
I.° • a 310
•• 0 • '5 .50 .55 .00 .70 .00 .90 1.00 1010 1.2u 1.30

2.0 .0961
3.0 .22M8
•• 0 .4291 .0370
5.0 .6~b'+ .0734 .0" 3Z .0932 .1034 01139
6.0 .75J9 01 042 01175 .1309 .1444 01580 .26'2
7.0 .8114 .1266 .1426 01 586 01747 .?.41102 .2993 .34,+ 5
6.0 .8bdQ .1434 01023 01816 .239'+ .2733 .2935 .3845
9.0 .921)'+ .1814 .2089 .23b'+ .2515 .2564 .3617 .3295 .J655
10.0 .9039 .17B2 .2011 .2237 .2280 .2969 .3121 .3532 .3953

·
··
11.0 1.0414 .1620 .2153 .2537 .23.5 .2563 .30bO .3'+/j2
12.0 .3927 .4335
1.17"1 oj 696 .1775 .2014 .2361 .2596 .3167 .3447
13.0 .3'10 .4391 .4747
1.3805 .1 b44 01" I" .20 I 0 .2208 .2475 .3114 .3434 .3906 .4219 .4631

·
1'.0 1.3730 •• 669 .6277
01575 .I" 13 . j 970 .2191 .2497 .30M5 .3437 .3752 .4298
15.0 1.3660
.4SbO .4911 .62~0
.15fi9 01710 .I 6. 7 .2169 .2417 .2946 .3309 .3636 .4191

··
16.0 1.3S95 •• 73' .4956 ,b291
.1523 • I b2S .1826 .2064 .2362 .2953 .3342 .3654 .4040
17.0 .4.481 ,4908 .4d1"
1.353. .1473 01563 .1746 .200' .2322 .2968 .3237 .3558 .J9n ,44tH

··
18.0 1. 31t 7 7 .1426 .'921 .4931
01517 01689 .1944 02210 .2862 .3241 .3b04 .3921 .450:1 .4983
19.0 1.3423 .4393
.1'0 I .1483 .1654 .1935 .2227 .2f.J09 .3190 .3529 .3893 .4542 .4767 .4J65
20.0 1.3372 .1339 .1471 .1596 .la74 .2173 .2Bb4 .3156 .3503 .3Sbd .4393

··
.4572 .4J41
22. u 1.3278 .1306 .1395 .1528 .1801 .2109 .2627 .3}44
24.0
.3472 .3616 ,4300 .4560 .4301
1.3193 01250 .135' .1_6,+ 01732 .2045 .2771 .3096 .3390 .3842

·
26.0 .4209 ,4959 .4270
1.3115 .120 I 01313 .1416 .1672 01967 .2717 .3056 .3380 .:J8S4

···
2a.0 1.304) .4313 .4861 • "Z45
.1178 .1276 .1370 .1643 01961 .2601 • 2'oH 7 ,3280 .3703
30.0 .4229 ,4199 ._Z24
1. Z ~7 7 .1155 .1217 .1375 .1648 .1974 .2648 .3262
32.0
.2f.J42 .3730 ,4140 .4262
1.2Y15 oIloa .1229 .1346 .1610 01935 .2676 .2912 .3166 .3645

·
34.0 .4237 .4251
1.28":)7 .1123 .120A .IJ48 .1604 01933 .2635 .2952 .3235 .3565 .417a

··
36.0 .417 J
1.2"03 01108 0121' 0131. 01563 .1869 .2591 .2913 .3210 .3554 .4092 .470b
38.0 1.2752 .10f.JIt .1191 .I 26 I 015.1 01 a60 .2671
.2542 .3163 .351. .4107 .4242
40.0 1.2103 .1080 .116A 01303 .1593 01 67a .21.00 .2b22 .3151 .3506 .4040 .4540
42.0
· 1. Z!)5A .10al .1153 oIJIO .1601 .1894 .2628 • Z80li .3162

·
.34Q1 .3\113 .4653
.4.0 l.llJ 14 .I 061 .1166 .1316 .1606 .}94d .2650 .2917 .3106 .3470 .3992 .46H9
1t6.0 1 • 2~73 .1074 01177 .1320 .1606 01952 .2668 .2944 .3150 .3506 .3905 .4539
48.0 1.2'33 .1086 oil A7 .I J22 .1606 01919 .2679 .2966 .ll8a .3461 .3661 .4601
50.0 1. 24~5 .1070 .1195 .1320 01600 .1916 .2663 .2922 .3Z20 • J4 08 .3a75 .4524

··
~2. 0 1. Zlt59 .1 06 1 .1200 01315 .15b9 01911 .2619 .2935 .3245 • J'+5 7 .3Ab5 .4491
54.0 1.2424 01 069 01202 .1290 .1557 .1898 .2b33 .2939 .3200 .3501 .3'tH7 .4509
56.0 1.2)Ql .1 096 .120? .1280 .1540 .18~O .2618 ,c935 .3215 .3539 .3893 .,"'32

·
58.0 1.23159 .1099 .j 1"3 .1267 0151 a .1855 .2595 .2921 .3222 • )494 .3847 .4447
bO.O 1.2328 .1084 .117A .j250 .1492 • PHS .25t.4 .2b9H .3219 .J~22 .37111 .4445
 191

SEDIMENTARY ASPECTS OF THE MVUMASE PROJECT

J. NICHOLSON (National Research Institute for Oceanology, Council for Scientific and Industrial
Research, Stellenbosch, South Africa)

1. INTRODUCTION

The Mvumase project consists of the Mvumase and Sunbury Dams which are to be constructed on the
lower Tugela River 56 and 32 km, respectively, upstream of the river mouth (see Fig. 1).

RIVER

MVUMASE
DAM

TUGELA RIVER o 5 10 15 20km

!

FIGURE 1. The Mvumase Project.

This paper describes the results of a study which concerned the following aspects of the project:
(i) the pre-d~m sediment dynamics of the Tugela coastline;
(ii) the post-dam sediment dynamics of the Tugela coastline;
(iii) the pre-dam sediment dynamics of the Tugela estuary mouth;
(iv) the operating policy for the dams.
192

2. PRE-DAM SEDIMENT DYNAMICS OF THE (Swart, Serdyn, to be published) and the breaker
TUGELA COASTLINE
line wave parameters (Krb ; 8b ) were derived from
a refraction study. The resulting value for the
The coastline adjacent to the Tugela River mouth
net annual longshore transport rate was 1,1.10 6
is supplied with material by two processes,
m3 /year.
namely, fluvial sediment transport and longshore
sediment transport.
The estimated volume of sand arriving at the
mouth of the Tugela during a typical year is
The average mass of solids transported by the
thus 0,6.10 6 m3 due to fluvial transport and
Tugela River is between 7,6.10 6 and 9,5.10 6
1,1.10 6 m3 due to net northbound longshore
tonne/year (Rooseboom, 1982a). Thus, using the
transport, making a total of 1,7.10 6 m3 /year.
higher, and therefore the more conservative, of
these two values and converting it into a volume
Offshore surveys carried out on the Tugela
discharge, assuming a void ratio of 40 per cent,
coastline (CSIR, 1978) indicate that the bar at
the volumetric rate is 5,0.10 6 m3 /year. It was
the mouth of the river grew at a rate of approxi-
also established (Rooseboom, 1982b) that 12 per
mately 1,1.10 6 m3 /year between 1913 and 1978.
cent of this load consists of sand. The re-
This accumulation, however, consisted of sediment
sulting fluvial sediment transport rate is there-
with a wide spectrum of grain sizes, and it is
fore 0,6.10 6 m3 /year.
therefore necessary to separate out the sand
fraction. That portion of the fluvial sand load
The longshore transport rate was computed by
which accumulates on the bar does so on the land-
means of a modified form of the Shore Protection
ward side of this structure, the seaward limit
Manual formula (Swart, 1976). This formula was
of deposition being dependent on the fall velo-
applied to each wave condition in turn and the
city of the smallest grains (60V diameter), the
results then summed to give the total transport
river flow velocity and the local bathymetry.
rate. Details of the formula are as follows:
The fall velocity of the smallest grains is
S = K(D).f .• T .H2 .K2~sin 28 b
x ~ p os ro 0,003 m/s (Graf, 1971) and an estimated value
where S longshore sediment transport rate
x for the flow velocity is I m/s. These two velo-
(m 3 /year);
cities, in conjunction with the available bathy-
K(D) grain size parameter
metric survey (CSIR, 1978), indicate that the
91.10 4 • log (0,00146/Dso)(m/(s.year));
fluvial sand load is deposited landwards of the
Dso median grain size (m);
15 m depth contour. As this is also the limit-
f. fractional occurrence of a given wave
~ ing depth for most of the longshore sand trans-
condition;
port, it can therefore be concluded that the
T wave period corresponding to the peak
P section of the bar landwards of the 15 m depth
of the energy spectrum(s);
contour consists mainly of sand. On this basis,
significant deep-water wave height (m);
the sand deposition rate on the bar is approxi-
refraction coefficient at the breaker
mately 0,4.10 6 m3 /year. This deposition rate,
line;
in combination with the supply rate of 1,7.10 6
angle between the wave crest and the
m3 /year, thus indicates that the longshore trans-
local bed contour at the breaker line.
port rate immediately to the north of the river
mouth is 1,3.10 6 m3 /year in a northerly direction.
The deep-water wave parameters (f i ; Tp; Hos)
were obtained from voluntary observing ships
 193

Aerial photographs of shoreline movement at the 3. POST-DAM SEDIMENT DYNAMICS OF THE

TUGELA COASTLINE
Siaya River mouth (see Fig. I) show that the
local beach width grew at an average rate of
Assuming that the Mvumase and Sunbury Dams trap
approximately 3 m/year between 1937 and 1977.
all the fluvial sand load, then the expected
A similar study of the beach width at the Nyoni
post-dam conditions along the Tugela coastline
and Mlalazi estuaries (see Fig. I) suggests that
will be as follows:
the Siaya growth rates are representative of the
(i) the bar opposite the Tugela mouth will
whole stretch of coastline between the Tugela
be eroded;
and Durnford Point. Taken in conjunction with
(ii) the longshore sand transport rate immedi-
the length of the coastline, this information
ately to the north of the Tugela mouth
thus points to a mean volumetric growth rate of
will decrease until it eventually equals
0,6.10 6 m3 /year between the Tugela and the Siaya
the rate immediately to the south of the
and 0,4.10 6 .m 3 /year between the Siaya and Durn-
Tugela mouth, namely, 1,1.10 6 m3 /year;
ford Point. Furthermore, it also follows that
(iii) the progression rate for the Tugela-Siaya
the net northerly longshore transport rate of
coastline will ultimately fall by ap-
1,3.10 6 m3 /year immediately to the north of the
proximately 33 per cent to 2 m/year;
Tugela has decreased to (1,3 - 0,6).10 6 or
(iv) annual fluctuations in the Tugela - Siaya
0,7.10 6 m3 /year at the Siaya and is (0,7 - 0,4).10 6
coastline progression rate will be less
or 0,3.10 6 m3 /year at Durnford Point.
marked due to the removal of the highly
variable Tugela sand load;
The pre-dam conditions along the Tugela coastline
(v) the first kilometre or so of coastline
can be summarised as follows:
immediately to the south of the bar will
(i) fluvial sand load arriving at the Tugela
become more stable due to the removal of
mouth is 0,6.10 6 m3 /year;
(ii)
fluctuations in the blocking effect caused
longshore sand load arriving at the Tugela
by the variations in the shape of the bar;
mouth from the south is I, I. 10 6 m3 /year;
(vi) little change will take place along the
(iii) sand deposition rate at the Tugela mouth
section of coastline further to the south
is 0,4.10 6 m3 /year;
of the Tugela mouth;
(iv) longshore sand transport rate immediately
(vii) the change from the pre-dam to the post-
to the north of the Tugela mouth is
dam conditions will be gradual (over
1,3.10 6 m3 /year;
decades).
(v) volume of sand deposited between the Tugela
and the Siaya is 0,6.10 6 m3 /year, equiva-
4. PRE-DAM SEDIMENT DYNAMICS OF THE
lent to a progression rate of 3 m/year; TUGELA ESTUARY MOUTH
(vi) volume of sand deposited between the Siaya
and Durnford Point is 0,4.10 6 m3 /year,
The stability of an estuary mouth is dependent on
equivalent to a progression rate of
factors which tend to flush the mouth out and
3 m/year;
factors which tend to block it. There are two
(vii) contribution of the Tugela sand load to
flushing factors, namely, tidal flow and fluvial
the accretion between the Tugela and the flow, and two blocking factors, namely, longshore
Siaya is effectively (0,6 - 0,4)/0,6 or sediment transport and fluvial sediment transport.
33 per cent.
It is, moreover, the peak values of these para-
meters which have most influence on the estuary
194

stability. where 'Sfm' is the fluvial monthly sediment trans-

port rate, 'Q ' i s the peak monthly flow rate
pm
A quantitative field study of estuary mouth sta- and the value of 1,8 for the exponent was based
bility using the ratio of the flushing factors on previous field work (Oliff, 1960). The spring
to the blocking factors as a stability criterion tidal prism was estimated to have a value of
has been made (Bruun, 1974). However, fluvial 5.10 5 m3 based on an estuary plan area of approxi-
flow and fluvial sediment transport were neglected mately 5.10 5 m2 (Begg, 1978) and an average es-
with the result that the stability criterion com- tuarine spring tidal range of 1 m. The spring
prised the ratio of the spring tidal discharge in tidal prism was assumed to remain constant through-
m3 , '~s', to the gross longshore transport rate out the year.
in m3 /year, 'sxg '. The findings of this study
were as follows: The resulting '~/S 'ratios (see Table 1) have
s xg
~ /S > 50 estuary mouth perma- a minimum value of 6 in August and a maximum
s xg
nently open; value of 55 in March. These'~ /S 'values in-
s xg
20 ..; ~ /S ..; 50 estuary mouth kept open, dicate that the estuary mouth is stable during
s xg
if at all, by floods; the summer months when the large river flows flush
~ .Is
s xg
< 20 estuary mouth perma- out the sand deposits. However, during the win-
nently closed. ter months, the river flows are low and most
In the case of the Tugela estuary, the fluvial material arriving at the estuary mouth remains
flow and the fluvial sediment transport cannot there, thus reducing the cross-sectional area.
be neglected. The stability analysis was under- During the latter period, the mouth is kept open
taken on a monthly basis and the parameter '~s' by the relatively weak tidal flushing action sup-
was therefore taken to be the average monthly plemented by the small river flows; the estuary
flood discharge, or 'fluvial spring tidal prism', only closes during periods of severe drought.
plus the actual spring tidal prism. The para-
meter'S 'was taken to be the sum of the gross The maximum (summer) and minimum (winter) values
xg
longshore transport rate and the fluvial sediment for the cross-sectional area of the estuary mouth
transport rate. The latter two parameters were are estimated to be as follows:
expressed in m3 /year even though they were based Maximum (summer)
on the transport rates occurring during a parti- The cross-sectionally averaged, maximum, spring
cular calendar month. tidal velocity in an estuary mouth is approxi-
mately 1,0 m/s (Bruun, Gerritsen, 1960). Neg-
The average monthly flood discharges or 'fluvial lecting the small tidal flow and assuming that
spring tidal prisms' (see Table 1) were derived the maximum, monthly fluvial flood discharge of
from the available hydrological data. The monthly 1176 m3 /s, which occurs in February, is equivalent
gross longshore transport rates (see Table 1) to the maximum, spring tidal discharge, this cri-
were computed as described in Section 2. The terion yields
monthly fluvial sediment loads (see Table 1) were A '" 1200 m2
max
derived by assuming that this parameter is a where 'A ' i s the maximum, cross-sectional area
max
function of the peak monthly flow rate, the of the estuary mouth below mean sea level. In
relationship being addition, a value of 1000 m has been quoted for
S ex: Ql'B the maximum width of the estuary mouth (Begg,
fm pm
1978), although this is probably associated with

extreme floods. Hence, assuming a typical, maxi-
mum width of 600 m, say, the corresponding average
depth of the estuary mouth below mean sea level
is 2 m.
Minimum (winter)
Application of the 'ns /S xg stability criterion
I during each wet season.
to the minimum estuary mouth conditions involves
the tidal flow, as well as the fluvial flood dis-
charge, because the former quantity cannot be
neglected. The tidal flow is estimated to gene-
rate a maximum, spring tidal discharge of 36 m3 /s,
based on a spring tidal prism of 0,5.10 6 m3 dis-
charging sinusoidally, and the minimum, monthly,
fluvial flood discharge, which occurs in June,is
34 m3 /s. The effective, minimum, monthly spring
tidal discharge is therefore 70 m3 /s and hence
the Bruun criterion gives
A. 2 Sunbury Dam of 125.32.10 3 or 4,0.10 6 m2 , the
ml.n '" 70 m
where Amin minimum, cross-sectional area of the
estuary mouth below mean sea level.
A minimum value of 50 m is quoted for the width
of the estuary mouth (Orme, 1974). The corres-
ponding average depth of the estuary mouth below
mean sea level is therefore approximately 1,5 m.
5. OPERATING POLICY FOR THE DAMS
The primary aim of the operating policy for the
proposed dams, from an engineering standpoint, is
to maintain the existing morphology of the Tugela
River mouth. Ideally, therefore, compensation
water should be released from the dams so that
the cross-sectional area of the estuary mouth
varies between a minimum value of approximately
70 m2 relative to mean sea level during the dry
winter (June) and a maximum value of approxi-
mately 1200 m2 relative to mean sea level during
the wet summer (February). It is also desirable,
from an aesthetic standpoint, that there should
be a small, perennial river flow downstream of
the lower dam site.
195
In order to achieve the objectives set out above,
the operating policy for the dams should consist
of two separate parts, namely,
(i) the release of a minimum, perennial flow;
(ii) the release of one or more "floods"
The minimum perennial flow must be large enough
to counteract losses due to evaporation; in ad-
dition, an extra 50 per cent allowance is recom-
mended for seepage losses. According to avail-
able hydrological data, the evaporation is lowest
between February and May, when it has a mean
value of 44,6 mm/month, and is highest between
June and January, when it has a mean value of
91,7 mm/month. Using an evaporation area of
4,5.10 6 m2 derived from the estuary area of
0,5.10 6 m2 and a river area downstream of the
resulting minimum perennial flows are as follows:
February to May 0,1 m3 /s;
June to January 0,2 m3 /s.
The total quantity of water involved in the main-
tenance of these discharges is 5,2.10 6 m3 /year
or less than 0,2 per cent of the total annual
runoff.
The simulated floods should be large enough to
ensure that the estuary mouth remains open through-
out the year. This, however, is an impractical
goal because it involves the release of an un-
acceptably large proportion of the total annual
runoff (in excess of 10 per cent). A suggested
operating procedure, therefore, involves the re-
lease of the peak monthly discharge in October
and April for a duration commensurate with an
'n s /S xg I value of approximately 20. October and
April releases were chosen because they coincide
with the beginning and the end of the wet season
and are spaced more or less equally throughout
the year and a 'n s /S xg value of 20 was chosen
I
because it is the limiting value for permanent
closure (see Section 4). Details of the result-
ing operating procedure are as follows:
196

REFERENCES

Begg G (1978) The estuaries of Natal, Natal

DISCHARGE Town and Regional Planning Report, Vol.41.
MONTH RATE DURATION QUANTITY Bruun P (1974) Regime equations and tidal
(I OG. m3) inlets (discussion), Proc. ASCE, Vol. 100,
(m 3 / s) (days)
No. WW4.
Bruun P and Gerritsen F (1960) Stability of
Oct. 200 2,0 35 coastal inlets, North-Holland Publishing Co.,
Amsterdam.
Apr. 200 1,6 28
CSIR (1978) Mandini sea outfall, CSIR Report
C/SEA 7811, Stellenbosch.
Graf WH (1971) Hydraulics of sediment transport,
McGraw-Hill, New York.
The total quantity of water released, 63.IO G m3 , Oliff WD (1960) Hydrobiological studies on the
is equal to 2 per cent of the total annual runoff. Tugela River system. Part I. The main Tugela
River, Hydrobiologica, Vol XIV, No.3-4.
Orme AR (1974) Estuarine sedimentation along
6. CONCLUSIONS the Natal Coast, South Africa, Office of Naval
Research, Tech. Report No.5, Arlington, Va.
Rooseboom A (1982a) Interim report on expected
The main conclusions of this study are: sediment-related changes due to the Mvumase
Scheme, submission to Department of Environment
Affairs.
(A) Pre-dam coastline: Rooseboom A (1982b) Addendum I to interim re-
port on expected sediment-related changes due to
(i) the deposition rate at the bar opposite
the Mvumase Scheme, submission to Department of
the Tugela estuary mouth is 0,4. lOG Environment Affairs.
Swart DH (1976) Predictive equations regarding
m3 /year; coastal transports, Proc.Fifteenth Coastal Engrg.
(ii) the Tugela-Siaya coastline progresses Conf., Hawaii.
Swart DH and Serdyn JdeV (to be published)
at a rate of 3 m/year; Statistical analyses of visually observed wave
(B) Post-dam coastline: data from voluntary observing ships for South
African East Coast. Vol. 17, CSIR Report,
(iii) the bar opposite the Tugela estuary Stellenbosch.
mouth will disappear;
(iv) the Tugela-Siaya coastline will pro-
gress at a reduced rate of 2 m/year;
(C) Pre-dam estuary mouth:
(v) the estuary mouth has minimum and
maximum cross-sectional areas of 70
and 1200 m2 respectively;
(D) Operating policy for the dams:
(vi) the release of a perennial flow of
between 0, I and 0,2 m3 /s;
(vii) the release of flood discharges in
October and April of each year.
 197

TABLE I. Estuary mouth stability

rI S
s xg

'Fluvial Actual Gross

Spring Spring Longshore Fluvial
Month rI/S xg
Tidal Tidal Transport Transport
Prism' Prism Rate Rate
(J06 m3 ) (10 6 m3 ) (10 6 m3 /yr) (J06 m3 /yr)

Oct. 38,9 0,5 1,5 0,2 23

Nov. 56,9 0,5 1,3 0,3 36
Dec. 74,1 0,5 I, I 0,5 47
Jan. 151,9 0,5 1,3 1,9 48
Feb. 203,2 0,5 I, I 3,3 46
Mar. 86,9 0,5 0,9 0,7 55
Apr. 42,7 0,5 1,2 0,2 31
May 17,6 0,5 1,6
° II
Jun. 5,9 0,5 0,9
° 7
Jul. 9,3 0,5 1,5
° 7
Aug. 7,6 0,5 1,4
° 6
Sep. 17, I 0,5 1,4
° 13

PHYSICAL ASPECTS OF SANDY BEACHES - WORKSHOP REPORT
D H SWART (Sediment Dynamics Division, NRIO, CSIR, POBox 320, Stellenbosch 7600, South Africa)
Thinking back over the discussions during the
Symposium, not only those on physical processes,
bht'on all five topicS, there are basically
three questions which were discussed either
extensively or which come to mind as a result
of the discussions, namely,
(1) Why is research on the physical processes
taking place on sandy beaches being done?
or WHY?
(2) Which processes need to be studied as a
matter of most urgent priority over the
next five years? or WHAT?
(3) How should these properties be studied
or HOW?
WHY?
There is general agreement that the Sandy Beaches
Symposium has been a resounding success. To my
mind this is because of a new awareness, cul-
tivated during the week, of the fact that
researchers in any given disclpline are not
working in isolation, but can, and most probably
do, work in parallel with those in other disci-
plines. In this respect those researchers
studying the physical processes in the coastal
zone are perhaps the most affected, since a
sound understanding of these processes is basic
to the understanding of the behaviour of chemi-
cals, nutrients and biota and holds the key to
the judicious management of the coastal zone.
\,ith management in mind, physical processes are
thus studied, first, because one wants to obtain
a better understanding of these processes and,
second, because researchers in other disciplines
199
need the information for their research. A good
example here is the research on phytoplankton
blooms being done at the Sundays estuary here in
Algoa Bay. Coastal engineers are observing surf
circulation patterns and are measuring incident
wave heights and water-level fluctuations both
spatially and temporally, in order to under-
stand more fully the driving force behind the
circulation patterns. It is expected that answers
will be provided to questions such as what are
the physical reasons for bloom formation and
maintenance, the rate of exchange between the
surf zone and beyond, and the seaward limit of
the exchange cells.
WHAT?
The review on physical aspects of sandy beaches
has emphasized the fact that this field of re-
search is basically still in its infancy and
that useful research should still be done on
numerous topics. However, the discussions
during the week of the symposium have high-
lighted a few key areas which need to be consi-
dered as a matter of priority and urgency over
the next, say, five years. These are listed
briefly below, not necessarily in order of
priority.
Water circulation: Since water serves as the
carrier for both biota and nutrients it can, in
a sense, be seen as the driving mechanism for
biotic communities. All aspects of water circu-
lation should be studied, namely, water circu-
lation in the water column, spatial water cir-
culation patterns, the driving mechanisms for
these circulations and also the flow of water
200

through the sandy bed in the infratidal and as yet unknown is the extent of stratification
intertidal zones. in general and its effect on permeability.

Sediment movement: Not much attention was given HOW?

to this aspect during the week. Nevertheless,
the understanding of the movement of sand in the
coastal zone holds the key to the correct manage-
ment of physical coastal processes. More research
should thus be directed to sediment transport by
wave, current and wind action. Specific attention
should be given to the interaction between sand
dune areas and the beach area itself.
Morphodynamics: This aspect is basically an
encompassment of the first two aspects and covers
the whole dynamic behaviour of beach systems as
a result of wave, current and wind action. Items
of specific interest are beach states, modes of
transport and interactions between water and
sediment movement. Research should be aimed not
only at qualitative descriptions, as is mostly
the case in this field, but also at finding
quantitative descriptions.
Although it was decided at the Symposium that
it is not really important in the context of the
Symposium to define the boundaries of the beach,
it is very important to do so from a management
point of view. Not only are the limits impor-
tant, but also their behaviour over time and in
space. When planning any coastal development it
is important to know how the boundaries of the
coastal zone are going to vary over the following
ten, twenty to fifty years.
Research can be done on various levels, namely,
purely theoretically, or by measurements in the
laboratory or in the field. Efforts at these
levels should be properly coordinated. Ulti-
mately research in the field will be needed. The
most important single factor which has retarded
progress on this front has been the difficulty of
gaining access to the hostile surf. Priority
should be given to research aimed at improving
the accessibility to the surf zone and at
improving the actual measuring techniques.
Coupled with this is the need to properly
analyse and represent the data which will require
a strong emphasis on statistical techniques as
well as, again, a proper understanding of the
physical processes, since one cannot properly
analyse a process if one does not know what it
is all about.

Microstructure within the sandy bed: A point made

during the week, which is well taken, is that we
should look very carefully into the effect of
microstructures in the bed on the whole flow
field. This will imply that research will have
to be done on the permeability of stratified
sandy beds. There are available theories which
will all?w the use of the information gathered
in this way for actual predictions but what is
 201

PART TWO

CHEMICAL ASPECTS

THE CHEMISTRY OF SANDY BEACH ECOSYSTEMS - A REVIEW
G.A. EAGLE (National Research Institute for Oceanology, Republic of South Africa)
1. INTRODUCTION
l1uch of the chemical information on sandy beaches
which can be found in the literature has been
gained incidentally as part of biological or
ecological studies. As such it is widely dis-
persed and fragmentary. There have been very
few studies on the chemistry per se of sandy
beaches. This immediately forces us to ask the
question, "What is the role of the chemist in
sandy beach ecology?". Indeed we may even ask
the more basic question, "Does the chemist have
a role to play in sandy beach ecology?".
In summarising existing knowledge of the chem-
istry of sandy beach ecosystems, an attempt will
therefore be made in this review to highlight
the key role which the chemist can play in
understanding these systems.
No attempt is made here to define a sandy beach.
Nevertheless it is important to set out clearly
in the beginning what will be considered in this
discussion, or more simply, what will not be in-
cluded. Since sandy beaches are largely well
oxidised, this review contains very little dis-
cussion of anaerobic systems. Therefore, al-
though the discussion includes oxygen-stressed
beaches, those systems which are frequently
anoxic, such as salt marshes, tidal flats,
estuaries and offshore sediments are excluded.
A large amount of literature on these systems
does, however, exist.
Much of the chemical work which has been done
203
on sandy beaches has been related to some aspect
of pollution. Since these are special cases and
mostly do not fall within the context of "Sandy
Beaches as Ecosystems", they are also not included.
In the context of ecosystems, the most important
aspect of chemical investigation is the cycling
and recycling of nutrients, and therefore this
forms the basis of this review. By far the grea-
test amount of existing information concerns the
two elements nitrogen and carbon, with smaller
inputs dealing with oxygen and phosphorus. These
four elements are therefore first considered
separately and thereafter an attempt is made in
the discussion to relate them to each other.
This should reveal some of the gaps in the
existing knowledge.
Some of the earliest studies of the cycling of
nutrients in coastal systems were carried out by
Waksman and co-workers in the 1930's (Waksman,
Carey, 1933; Waksman et al., 1933a, 1933b, 1933c).
Although these studies dealt with coastal water
and not sandy beaches, they pointed, even at that
stage, to the close link between chemist and bac-
teriologist in the study of nutrient cycling. The
views on remineralization and recycling of nu-
trients have changed considerably over the past 50
years and this has been described in a recent re-
view by Nixon (1981). He has traced the histo-
rical development of remineralisation cycles, as
they have evolved over the past three decades,
and emphasized the fact that these, views are still
changing. For example, in a review of nutrient
204
cycling written fourteen years ago, Johannes
(1969) concluded that "it has been demonstrated
beyond reasonable doubt that most of the nitrogen
and phosphorus incorporated into aquatic plants
is usually regenerated by processes other than
direct bacterial action". Thirteen years later
in a complete turnabout, Koop et al. (1982b)
have concluded that bacteria are primarily res-
ponsible for the decomposition and recycling of
nutrients from kelp debris cast up on the shore.
On the beach there is a complex interrelation-
ship between many physical, chemical, geo-
logical and biological factors, none of which
can be controlled and all of which vary. A com-
mon characteristic of all sandy beaches is high
turbulence (Reidl, 1971; Reidl, McMahan, 1974;
McLachlan, 1979). For this reason, far fewer
studies have been made of sandy beaches than in
many other types of situation. Working in the
surf zone probably remains the single most dif-
ficult problem of research in the coastal zone.
This has forced many workers to turn to the la-
boratory, with studies of sand cores (McIntyre
et al., 1970; Boucher, Chamroux, 1976; Munro et
al., 1978; Wormald, Stirling, 1979; McLachlan
et al., 1981b; Mevel, Chamroux, 1981) or model
sandy beaches (Pugh et al., 1974; Pugh, 1975,
1976; Vosjan, Olanczuk-Neyman, 1977). The success
of these experiments has amply demonstrated
that, although refinements are required, models
have a very valuable role to play in the future.
In the discussion which follows, the four ele-
ments, oxygen, nitrogen, carbon and phosphorus,
have been considered separately. This in no way
implies that they can be separated in nature.
Indeed, examination of the literature immedi-
ately reveals that the cycles of these elements
are inextricably linked not only to each other,
but also to physical and biological processes.
2. OXYGEN
There can be no doubt that oxygen plays a central
role in sandy beach ecosystems since it has a
direct influence on both the chemical and bio-
logical reactions which take place. The level
of oxygen controls the redox equilibria main-
tained by some other elements, including ni-
trogen, sulphur, iron and manganese (Aller, 1980)
and therefore influences the cycles of these
elements. Thus the discussion of oxygen is
closely linked to that of the redox potential
Eh. Fenchel, Riedl (1970) and McLachlan (1978)
have shown the importance of the Eh in sandy
beaches. Eh measurements were found to be a more
sensitive measure of the interstitial climate
than was oxygen availability, but this may have
been due to the methods of measurement. The
redox potential discontinuity (RPD) has been
shown to be an area where significant changes in
the oxygen cycle occur (McLachlan, 1978). This
is "the layer where oxidizing processes become
displaced by reducing processes" or "the flattest
part of the Eh curve". It was also found to be
the region of highest organic matter.
Eh values were also found to vary with tides, and
had a clear influence on the distribution patterns
of the different forms of nitrogen. Nitrate was
dominant in oxidized layers, while ammonia was
more abundant in reducing layers. Nitrite was
found in appreciable concentrations in the layers
between.
Most of the chemical processes which take place
in natural systems are redox processes, involving
the transfer of electrons. Many of these are
biologically mediated and in the process energy
is released. Although oxygen is energetically
the most favourable electron acceptor, other
electron acceptors are readil y avail abl e for this
purpose, in the well-known sequential order 02,
N0 3 - , 50 4 2- and C02 (Brezonik, 1977). It should

however be recognised that reduction of nitrate
to nitrogen does not require, nor result in,
highly reducing conditions, as do reduction of
sulphate to sulphide or of carbon dioxide to
methane.
Oxygen availability is relatively high in the
surface layers of most beaches, but it drops
within the deeper layers. A characteristic of
some sandy beaches is the dark grey or black
layer in the region of stagnation and oxygen
deficiency, resulting from the formation of iron
and other sulphides under reducing conditions.
Its depth beneath the surface is related to the
penetration of oxygenated water into the inter-
stitial spaces of the porous system, and ranges
from some metres in coarse sand and shell subject
to heavy surf, to a few millimetres in more pro-
tected areas with fine sand (Riedl, McMahan, 1974;
Pugh et al., 1974; Dye, 1981).
Another factor which contributes to the oxygen
levels in interstitial water is the filtration
distance, i.e. the distance which this water
travels within the interstices (Riedl, 1971;
McLachlan, 1982). Finer sands have greater
surface areas for microbial colonisation and are
also more efficient detritus traps. This, to-
gether with lower permeability leads to greater
oxygen uptake and mineralisation than in coarser
sands. This is also a common and well-known
phenomenon in freshwater systems (e.g. Hargrave,
1972) •
In most exposed South African beaches, the per-
colation of oxygenated surf water is so ef-
ficient that oxygen is supplied in excess of
organic matter, so keeping the sediments com-
pletely aerobic despite the greater 'digestive'
activity that accompanies this (McLachlan, 1979).
An exception to this has been found in Hout Bay
on the Cape Peninsula (Fricke et al., 1979).
similar situation has been found in Nahant Bay,
205
Massachusetts (Quinlan et al., 1983). Both of
these beaches are subject to deposition of large
amounts of algae.
In their experimental sand column study, McIntyre
et al. (1970) found that oxygen data provided the
most detail~d picture of events within the column.
There appeared to be a gradual increase of ac-
tivity within the top 5 cm, whereas all layers
below this showed a decrease. Munro et al. (1978)
used small (5 cm) sand columns for short periods
(2-3'days) for measuring the rate of oxygen con-
sumption. These showed that sediment samples
from the region of the permanent water table had
considerably lower rates of respiration than those
from surface layers. This respiration was shown
to be microbial. Increases in flow rates through
the column increased the rate of respiration.
This was probably due to increases in the amount
of organic material received by the sand column,
but may also have been assisted by removal of
inhibiting waste products (Dye, 1980).
A similar result was obtained in a model tidal
flat (Vosjan, Olanczuk-Neyman, 1977) although
the oxygen consumption in the tank was much
higher than that in the natural environment.
The oxygen level in beaches is also dependent on
position on the beach. This was found in a model
beach (Pugh, 1976) where the dissolved oxygen
levels at higher beach stations were higher than
at those lower down. This is probably due to a
greater filtration distance (McLachlan, 1982).
There was also a seasonal variation, with lower
levels in summer (Pugh et al., 1974; Pugh, 1976).
This is due to greater microbial action and ni-
trification.
In the field, oxygen consumption in sandy beaches
has been studied extensively by Dye (1979, 1980,
A 1981). He has suggested that experimental systems
almost invariably overestimate the oxygen demand
206
since they rely on diffusion only. In sandy
beaches oxygen is supplied primarily by water
movement and this occurs at a much faster rate
(Dye, 1979).
Tidal fluctuations were found to influence bio-
logical oxygen demand (BOD) (Dye, 1980). This
was highest during, or slightly after, high tide
with the lowest BOD at low tide. These fluc-
tuations were found to be greatest near the
sediment surface and at higher tidal levels.
Similar results were found by Munro et al.
(1978). A significant relationship was found
between water content of the beaches and oxygen
consumption but no rel ationship between tem-
perature or number of microorganisms and BOD
could be found (Dye, 1980).
Dye (1980) worked out oxygen budgets for two
beaches with different exposures. At the more
exposed beach interstitial systems were found
to account for 43% of the oxygen consumed, with
macrofauna consuming 57% of the oxygen. In the
more sheltered beach, macrofauna consumed only
3% of the oxygen budget, the rest being ac-
counted for by interstitial organisms.
The role of oxygen can therefore be summarised
as follows: oxygen enters the ecosystem pri-
marily through dissolution in the surf zone,
although photosynthesis by primary producers is
another major source. This oxygen is used as
an electron acceptor in redox'processes, during
which energy is rel eased. Although other elec-
tron acceptors exist, oxygen is energetically
the most favourable and is therefore used pre-
ferentially. The level of oxygen therefore
influences other redox reactions which may take
place. The level of oxygen within the beach
depends to a large extent on physic& factors,
such as turbulence, sediment grain size, fil-
tration distance and depth in sediment.
3. NITROGEN
Nitrogen possibly provides the simplest perspec-
tive from which to view the material balance in
a marine ecosystem (Roman, 1980). Whereas cellu-
lar nitrogen is primarily bound in structural ma-
terial, both carbon and phosphorus are largely
involved in cellular metabolic activities in ad-
dition to their structural role (McCarthy et al.,
1975). The chemical and biological processes of
the sea are sufficiently versatile to account for
nitrogen in all its possible oxidation states,
from -3 to +5. The most important reduced forms
in sandy beaches are as ammonia* and organic ni-
trogen compounds. According to Vaccaro (1962),
they represent about 35% of the combined nitrogen
in the ocean. The most important oxidised form
of nitrogen is nitrate, although nitrite plays
a somewhat lesser role.
On the basis of these most important forms of
nitrogen, we can draw up a very simplified scheme
of nitrogen transformations in beaches (Figure I).
8H3 nitrification S
• NO •
nitrification (5)
• NO;_
runoff
E reduction
FIGURE I. Simplified nitrogen scheme.
*Throughout this review the term ammonia will be
used to signify the equilibrium between ammonia
and the ammonium ion.

This scheme contains none of the biological in-
tricacies of nitrogen cycling and will be con-
sidered from a purely chemical viewpoint.
3. I Nitrogen fixation and assimilation
As is the case on land, most marine plants are
unable to break the high energy N=N bond in the
nitrogen molecule. However, certain blue-green
algae, notably Trichodesmium spp. have been shown
to fix nitrogen on a large scale (Dugdale et al.,
1964; Goering et al., 1966; Dugdale, Goering,
1967). These organisms obtain their energy for
this process from sunlight and therefore nitrogen
fixation does not occur in the darkness. The pro-
cess is complex biochemically but evidently depends
on a metal-containing enzyme in which both iron and
molybdenum take part (Bailey et al., 1978). This
leads to ammonia which is immediately incor-
porated into amino acids in the plant.
Nitrogen from the atmosphere is fixed by phyto-
plankton and algae, and is frequently regarded
as being the critical limiting factor to photo-
synthetic activity (e.g. Ryther, Dunstan, 1971;
Lewin et al., 1979; Nienhuis, 1981; Nixon, 1981;
Quinlan et al., 1983). However, in some cases
phosphorus has been shown to be limiting (Chiaudani,
Vighi, 1982). Nitrate depletion is attributed to
phytoplankton growth and activity during spring
and summer. During this time nitrification may
occur but if it does the rate of nitrate pro-
duction is exceeded by its rate of uptake, re-
sulting in depletion. The N:P ratio in phyto-
plankton is commonly regarded to be about 10:1
while that in oceanic water is considered to be
about 15:1. Therefore at first sight the fact
that nitrogen becomes depleted more rapidly
than phosphorus may seem surprising, and merits
further discussion.
First, while the ratio of N:P in ocean water may
be about 15:1, this ratio does not necessarily
hold for coastal water (Ryther, Dunstan, 1971;
207
Lewin et al., 1979; Pugh, 1976; Nixon, 1981).
This can be attributed partly to land runoff of
phosphorus and to nitrogen utilisation.
Second, this points to a mechanism whereby nitro-
gen and phosphorus are recycled in the coastal
zone, including sandy beaches (HcLachlan, 1980-81).
In this process phosphorus is recycled at a rela-
tively greater rate than is nitrogen (assuming
an N:P assimilation ratio of 10:1) and this en-
sures that phosphorus is always present in excess
(Ryther, Dunstan, 1971). In Narragansett Bay on
the east coast of the USA, the input of total
nitrogen was estimated at 1.4 times the measured
recycling of nitrogen, while in the case of
phosphorus and silicate the input amounts were
80% and 14%, respectively, of the recycled
amounts (Nixon, 1981). It was therefore calcu-
lated that, without nutrients being recycled, the
input could support, at most, some 24-50% of the
annual production.
A third process which may contribute to the de-
pletion of nitrogen is the loss to the atmosphere
as N2 by the process of denitrification (Nixon,
1981) (see later).
The dominant source of nitrogen in sandy beaches
is the ocean. Fixed nitrogen is assimilated
through photosynthesis by the primary producers.
For this the nitrogen must be in the form of
ammonia, nitrite or nitrate. Although all three
forms of nitrogen can be utilised, it has been
shown that ammonia is taken up preferentially,
sometimes accounting for more than 90% of the
nitrogen ration (HcCarthy et al., 1975).
3.2 RNH2 - NH3 interconversions
The interconversion of nitrogen from the organic
form to ammonia does not involve a change in the
oxidation state of nitrogen. These processes are
therefore not controlled chemically by the oxi-
dation status of the beach.
208
Once organic nitrogen reaches the beach it can be
processed via a number of pathways. Billen (1978)
gives a schematic representation of how this
occurs in coastal sediments from the North Sea.
The regeneration of organic nitrogen starts with
the deamination of native marine proteinaceous
substances and leads to the formation of ammonia.
This process, ammonification, can be initiated
by heterotrophic bacteria. Once available, the
ammonia is a suitable source of nitrogen for
autotrophic bacteria, leading ultimately to
nitrate. While the nature of the oxidative pro-
cess remains uncertain, its rapidity suggests
that both chemical and biological mechanisms
contribute to this oxidation (Vaccaro, 1962).
The biological forms of nitrogen are numerous
and therefore some discussion on this aspect is
warranted. In plants and animals, nitrogen is
primarily bound in structural material in the
form of proteins and amino acids. The principal
amino acids of the proteins of phytoplankton
are glutamic and aspartic acids, alanine and
leucine (Ri ley, Ches ter, 1971). This plant
tissue nitrogen is assimilated into the eco-
system in a number of ways. Surf zone phyto-
plankton and algae may excrete extracellular
metabolites, contributing to the dissolved
organic nitrogen pool of the surf zone. Alter-
natively they may be grazed by a wide variety
of animals, discussion of which is beyond the
scope of this review. Faecal material from
these animals may be in the form of urea or
ammonia itself. Phytoplankton blooms may die
giving rise to a large amount of particulate
organic material, which may be deposited on
sandy beaches. Both dissolved and particulate
material, in its suspended and sedimented forms,
is subject to heterotrophic attack, which again
falls outside the scope of this review. However
autolysis of this bacterial tissue will also give
rise to ammonia.
3.3 Ammonia - nitrate interconversions
The conversion of ammonia to nitrate involves
the oxidation of nitrogen from the -3 to +5
oxidation state. Thus, this process can take
place only in an oxidising environment. On the
other hand, the reverse process can take place
only under reducing conditions. Nitrate re-
presents the highest oxidation state of nitro-
gen and therefore can be formed only through
an oxidative process. The concentration of
nitrate ~easured at any time therefore repre-
sents the balance between the rate of forma-
tion of nitrate and the rate of nitrate assimi-
lation. The oxidation of ammonia takes place
through bacterial nitrification. Some nitrite
is also formed. Both nitrite and nitrate are
used in the photosynthetic process of phyto-
plankton, although, as stated, uptake of ammonia
is preferred.
The process is reversible, and the reductions
also normally take place through bacterial
action under conditions of oxygen stress, where
the reduction of nitrate provides an energy
source for the bacteria.
3.4 Denitrification
Denitrification is an anaerobic bacterial
process whereby nitrate is used as the terminal
electron acceptor and reduced primarily to N2
which is lost to the atmosphere (Brezonik, 1977).
It is regarded as a four-step sequence:
N03 + N02 + NO + N20 + N2 , during which the
oxidation state of nitrogen is reduced from +5
to zero. Although this process would not take
place in a well-oxidised sandy beach, it has
nevertheless been shown to take place in an oxi-
dised zone (S~rensen, 1978), where anaerobic
microniches occurred in the sediment. In ex-
periments in which 15 N was used as a tracer,
Brezonik (1977) found that all the nitrogen in
denitrified N2 arose from nitrate, and that none

came from the partial oxidation of ammonia or
any other forms of reduced nitrogen. In contrast,
Goreau et al. (1980) have suggested that where
oxygen concentrations are very low, bacterial
nitrification of ammonia to N20 may contribute
significant quantities of the latter to the atmos-
phere. According to Nixon (1981), coastal areas
are strong sinks of nitrogen in the marine envi-
ronment because of denitrification.
In sandy beaches the nitrogen reactions taking
place are therefore governed to a large extent
by the redox conditions pertaining at the time.
Under oxidising conditions, high nitrate con-
centrations are likely to be found. Such oxi-
dising conditions almost invariably occur in the
water of the surf zone. In the interstitial
water oxidising conditions are likely to be
found in situations with a high water percolation
rate, which in turn is governed by the physical
characteristics of the beach. On exposed coasts,
truly reducing conditions are rarely found, al-
though they can exist in more sheltered envi-
ronments.
3.5 Nitrogen cycling
With this as a background, the chemical work
dealing with nitrogen in sandy beaches can be
reviewed. In addition to work in the natural
environment, much useful information has been
obtained by laboratory simulation experiments,
either sand cores or model beaches. An attempt
will be made to describe this research in the
context of the discussion above.
The ratios of the different forms of nitrogen
are variable. The proportions of NH 3 :N02-:N03
were 9:4:1 in the Baltic (Bolter et al., 1981)
while Orren et al. (1981a) found that N03 :N02
concentrations on exposed sandy beaches in South
Africa were in the ratio of about 30: 1. In the
latter case no bacterial or ammonia measurements
were made but no correlations between nitrate
209
or nitrite and meiofauna could be found. This
supports the suggestion by McIntyre et al. (1970)
that interstitial fauna do not make direct use
of dissolved inorganic nutrients. Their main
source of energy seems to be bacterial flora
which are, in turn, maintained by 'soluble'
organic material.
Laboratory studies have been conducted using
experimental sand columns or model beaches to
observe the behaviour of nitrogen under control-
led conditions. In most of these, the flow-
through rate of water ensured that the columns
were well oxygenated. Thus oxidative reactions
have dominated most of these experiments. In
one of the first of this type of experiment,
McIntyre et al. (1970) studied large sand columns
(29 cm diameter) for a period of about one year.
However, these sand columns were kept continu-
ously inundated with seawater and therefore,
over the period of study, the biota took on a
more subtidal than intertidal nature. The nu-
trient concentrations in both ingoing and
emerging water were monitored. Concentrations
of ammonia and nitrite were found to be unaltered
by the sand columns. However, there was a
marked increase in nitrate concentrations, in-
dicating that the microbial population of the
column oxidised organic nitrogen, the end result
being nitrate.
Wormald, Stirling (1979) conducted a study using
columns based on the design of McIntyre et al.
(1970), with modifications. This was undertaken
primarily to observe the behaviour of phosphorus
in the columns (see later) but some observations
of nitrate utilisation were also carried out.
Some columns were enriched with nitrate, others
with sewage, and these were compared with controls.
It was found that although the nitrate concen-
trations in influent water fluctuated markedly,
effluent nitrate concentrations were much more
consistent. The percentage of nitrate removed
210
was greatest when concentrations in the influent
were highest, while there was production of these
nutrients when the influent water was not en-
riched. A surprising result was that removal of
bacterial populations from the columns by the
addition of antibiotics had no effect on nutrient
removal. This was possibly due to incomplete re-
moval of the bacteria.
McLachlan et al. (1981b) developed the idea of
sand columns further, in that the water, instead
of being made to drain through the column con-
tinuously, was added in 20-minute pulses every
12 hours, to simulate tidal flow. The rate of
through-flow was also regulated to give realistic
flow volumes and rates. These were based on
calculations of the average residence time of
water in sandy beaches (Riedl, 1971; McLachlan,
1979). The experiments were run for four months.
After two months, the organic nitrogen in the
inflowing water was boosted by the addition of
glycine. This clearly resulted in greater ac-
cumulation of organic nitrogen, ammonium and
nitrite in the columns and increased the gene-
ration of nitrate, detected in effluent water.
This is indicative of greater microbial activity
and an increase in microbial biomass. However,
because the amino acids are more easily utilised
than the organic matter naturally present in
seawater, they cannot be regarded as being re-
presentative of the range of organic compounds
normally present in seawater. A similar result
was obtained by Mevel, Chamroux (1981) in a
nitrification study.
About 35% oxidation of the inflowing organic
nitrogen occurred in a 50 cm sand column at the
measured levels of organic nitrogen, (McLachlan
et al., 1981b). A lower flow-through rate in-
creased this to about 45% while a greater flow
rate decreased it to 18%. Halving the column
length (to 25 cm) had a negligible effect on the
nitrogen mineralisation, confirming that most
interstitial activity occurs in the upper layers
of the column. Therefore, since the most readily
oxidised nitrogen compounds are utilised the
fastest, the rate of mineralisation should de-
crease down a sand column.
These results were extended to a simple mathe-
matical model for sandy beaches in which the
amounts of inorganic nutrients generated could be
predicted for a range of beach types and con-
ditions (McLachlan, 1982). This indicated that
nutrient regeneration followed essentially the
same pattern as that for water filtration. Rates
were lowest with flat beach slopes and small tides
and increased towards intermediate situations
with small tides and steep slopes or with larger
tides and flatter slopes. Mineralisation rates
were higher in finer sediments, since there is a
greater particle surface area available for
microfauna colonisation. However, in spite of
this, it was found that more nutrient regene-
ration occurred in 300 ~m sands than 200 ~m
sands, because of the greater filtered volumes in
the coarse sand. This emphasizes the importance
of the effect of filtered volumes on minerali-
sation rates. This is different from the result
obtained for phosphorus by Wormald, Stirling
(1979), who found that doubling of the flow rates
had no effect on phosphorus concentrations in
the effluent water. However, the latter case
can be attributed to some form of "buffering"
mechanism for phosphorus (see later).
Mevel, Chamroux (1981) have studied nitrification
processes in experimental tanks. These appeared
to be related to the introduction of organic
matter and ammonia (added as ammonium sulphate)
and, as expected, were favoured by oxygenation.
The carbon source used was sodium succinate. The
rate of disappearance of ammonia depended on the
succinate concentration, as well as on the con-
centration of the ammonia itself.

Another approach which has been used to study
nutrient cycling in the laboratory is the con-
struction of model sandy beaches (Pugh et al.,
1974; Pugh, 1975, 1976; Vosjan, Olanczuk-Neyman,
1977). The model of Pugh (1975, 1976) was main-
tained for one year and was fed with seawater
directly from Menai Strait, North Wales. The
nitrate concentration in this inflowing water
followed the normal pattern of low (or undetect-
able) levels in summer and maximum levels in
winter, with other minor variations. Inter-
stitial water from the model beach contained
approximately double the nitrate concentrations
of the inundating seawater, again with vari-
ations, particularly in summer when interstitial
water was relatively richer in nitrate. The
nitrification process in sediments was rapid,
since the interstitial water was found to have
double the concentration of nitrate as that
in the overlying water, after being covered for
only three hours. Pugh et al. (1974) and Pugh
(1976) found some differences between their
model and nearby sandy beaches. The ammonia
concentrations in the interstitial and surface
waters of the model were about the same, while
nitrite concentrations were invariably lower
in interstitial water. However, the concen-
trations of nitrate and ammonia in the inter-
stitial water of beaches were found to vary with
coarseness of substratum and position on the
beach. Ammonia concentrations in groundwater
in a fine beach were up to twelve times those in
overlying water, and ammonia was the dominant
form of nitrogen. This is similar to the results
of Bolter et al. (1981) working in the Baltic.
On the other hand, nitrate dominated a coarser
beach, although the overall nitrogen status was
only about 25% of that in the fine beach.
This can be explained in terms of the redox
status of the beach. In a fine beach, percolation
of the water would be much slower and hence mi-
crobial action would lead to the depletion of
oxygen. This would favour reduction reactions
211
with ammonia as the product. The opposite
situation could be expected to arise in a coarser
beach.
Koop et al. (1982b) studied cycling in a micro-
cosm on a beach subjected to large amounts of
washed-up kelp. While the beach in this study
was therefore a special case, it nevertheless
provided some insight into the nitrogen path-
ways in beaches. Heterotrophic microorganisms
play the key role in the degradation of organic
nitrogen compounds in sandy beaches, although
in some areas macrofauna also play an important
part, e.g. Donax on Eastern Cape, South Africa,
beaches (McLachlan, 1980-81) and razor clams
on the Pacific coast of the USA (Lewin et al.,
1979).
In the experiment of Koop et al. (1982b), high
concentrations of ammonia accumulated in the
microcosm, but particularly immediately under
the pile of kelp, while nitrate and nitrite con-
centrations were low. Thus nitrogen leaching to
the sea was dominated by ammonia. This would
imply an oxygen deficiency immediately under the
kelp. A similar result was obtained by Raine,
Patching (1980) in a shallow, semi-enclosed inlet
of average depth 8 m, again, a low energy situ-
ation. In the latter case the flux of nitrogen
across the sediment-water boundary was measured
by a bell-jar technique. Koop et al. (1982b)
found that the very large majority of the nitro-
gen in the kelp was converted to microbial bio-
mass, while only a small fraction was returned
to the sea, mostly in the form of ammonia. From
their nitrogen budget they calculated that leach-
ing could supply only 0,09% of the nitrogen re-
quirements of the adjacent kelp bed, and that the
latter must therefore depend mainly on in situ
regeneration of nutrients and upwelling. The
sandy beach ecosystem thus retains the nitrogen
in the kelp debris cast up. Ultimately however,
the bacterial biomass must also be remineralised
212
and the nitrogen recycled.
This underlines the uncertainty of how effective
sandy beaches are at recycling nutrients back to
the surf zone. This problem was addressed by
McLachlan (1979) on some exposed phytoplankton-
dominated South African beaches. On the as-
sumption that only half the particulate matter
was fine enough to enter the beach pore system,
he calculated the amount of nitrogen which could
be generated for each cubic metre of water
filtered. He suggested that the entire nitrogen
pool would be regenerated every 24 days (or 15
times per annum), through the mineralising ac-
tivity of the intertidal interstitial fauna and
macrofauna (McLachlan, 1980-81).
From his model, McLachlan (1982) estimated that
under average conditions, nitrate regeneration
would amount to about 0,9 g nitrate per metre of
shoreline per day. This is sufficient to re-
plenish the nutrient pool of the surf zone every
17 days. This conclusion is different from that
of Koop et al., (1982b) and at this stage must
be attributed to the completely different primary
production systems in the surf zone, the one
being macrophyte-based while the other is phyto-
plankton-based. This problem has also been ad-
dressed by Hennig et al. (I 983), who suggest that
the nutrient regeneration time is much longer
than that put forward by McLachlan. These dif-
ferences will have to be resolved by further
research.
Lewin et al. (1979) have examined the role of
macrofauna, particularly razor clams, in the re-
cycling of nitrogen on beaches along the north-
western coast of the USA. At these beaches the
nitrate concentrations in the surf are often
critically low for long periods. On the other
hand, annnonia in surf samples during sunnner is
often present at higher levels than is nitrate
and is obviously an important nitrogen source.
The presence of high clam densities led to the
conclusion that annnonia regeneration could be
significant to the overall nitrogen cycle of the
surf zone. This occurs particularly during
summer when nitrate concentrations are low
(Lewin et al., 1975).
Highly reducing conditions have been dealt with
in a large number of publications, particularly
those concerning salt marsh, estuarine or off-
shore systems (e.g. Painter, 1970; Webb, 1981;
Nedwell, '1982; Nishio et al., 1982). Since
such conditions do not arise frequently on sandy
beaches, they will not be dealt with in detail
here. However, they should be considered briefly.
Working with 15N-Iabelled nitrate, Buresh,
Patrick (1981) have investigated the reduction
of nitrate to ammonium and organic nitrogen com-
pounds in a laboratory study. Sediments were
covered by relatively large volumes of water and
were incubated in unsealed containers open to
the air. Nevertheless, they were anaerobic.
After a month, less than 1% of the added nitrate
remained in the sediments as nitrate and only a
small fraction was recovered as ammonia. Con-
siderably more of the recovered 15N was present
as organic nitrogen. However, more than 70% of
the added nitrogen was not recovered and was
lost from the system, presumably by denitri-
fication. The recovery of added nitrate as both
ammonia and organic nitrogen increased markedly
as the redox potential decreased. The quantity
of labelled organic nitrogen increased con-
tinually with time, whereas the labelled ammonia
initially increased more rapidly but then de-
creased. This confirms that the ammonia was an
intermediate in the production of organic
nitrogen compounds.
S~rensen (1978) investigated the reduction of
nitrate to nitrogen and ammonia in coastal
marine sediment. He concluded that the capacity

for denitrification depended on the presence of
bacterial denitrifiers in the sediment and that
the process is restricted to the top few centi-
metres of sediment.
The nitrogen cycle in sandy beaches can there-
fore be summarised as follows: organic nitrogen
compounds imported from the surf zone are broken
down rapidly by heterotrophic action to ammonia.
On the one hand this can be re-exported to the
surf zone or, on the other hand, can be either
oxidized to nitrate under oxidizing conditions
or reconverted to organic nitrogen compounds by
autotrophic action under reducing conditions.
The nitrate itself can be used as an electron
acceptor or it may be reduced to N2 in a de-
nitrifying process and lost from the system.
4. CARBON
The carbon cycle in the ocean begins and ends
with carbon dioxide. Although the oxidation
state of carbon remains unaltered in sandy beaches,
oxidation and reduction of organic compounds
occur. These transformations are almost entirely
biologically mediated. Since the biological
communities which exist are very largely con-
trolled by oxygen availability, the influence of
the redox conditions is again important.
Carbon compounds represent the source from which
the ecosystem derives much of its energy. In con-
trast to the inorganic pool of the oceans, how-
ever, where more than 99% of the total material
can be characterised with eleven fairly invariant
constituents, about 90% of the organic material
remains uncharacterised and the remainder is, at
best, quite variable.
The carbon in the ocean is divided traditionally
into "particulate" and "dissolved" constituents.
By convention, these are usually defined as the
fractions retained by, or passing through, a
213
0.45 ~m membrane respectively. However, it is
well known that there is no clear distinction
between dissolved and particulate material. Be-
tween the two is a large grey area of colloidal
material, which has yet to be characterised. For
convenience, the terms "particulate" and
"dissolved" carbon will be retained here and used
in the conventional sense. McIntyre et al. (1970)
have however stated that "while the division of
organic material in seawater into particulate and
soluble fractions does define the large reser-
voir of non-living organic matter less than about
1 ~m, it is rather artificial and does little to
help understand the complex processes involved in
the turnover of this material". However, Farke,
Riemann (1980) have sugges ted that colloids may
play a role in animal nutrition that cannot be
neglected, and that carrier-mediated concen-
tration phenomena with involvement of colloids
should be considered in future.
Primary productivity of the sea is controlled by
a number of intricate environmental factors, in
which physical considerations play an extremely
important role. Undoubtedly the most important
removal processes for dissolved organic carbon
are those in which the breakdown and removal
of this material provides energy and cellular
carbon for living organisms. Bacteria can only
assimilate dissolved substrates; solid substrates
are first hydrolysed by extracellular enzymes
before being assimilated. Dissolved carbon com-
pounds, which are probably of little value to
marine animals, are thus converted into bacterial
tissue which can be utilised by them. However,
much of the dissolved matter, e.g. humic material,
"gelbstoffe", is resistant to bacterial attack
and is therefore biologically unavailable.
The amount of dissolved organic matter (DOM) in
the sea usually exceeds the particulate fraction.
It has been suggested that th~ concentration of
dissolved organic matter is about an order of
214
magnitude greater than that of particulate organic
matter (PaM) in seawater (McIntyre et al., 1970)
and that free dissolved monosaccharides form less
than 10% of DaM. In contrast, Meyer-Reil et al.
(1980) have found that DaM formed only about 1%
of the total organic matter of sandy beaches in
the Baltic and that less than 19% of this was
made up of free dissolved monosaccharides. The
significance of this material is difficult to
determine; however, its potential as a food
source is large. An added complication in this
respect is that a low concentration of a par-
ticular compound does not necessarily imply that
it is unimportant as a nutrient - in fact it
could indicate quite the opposite (Koop, Carter,
1982).
The dissolved organic material is made up of an
almost infinite number of compounds. A few spe-
cific classes have been looked at in detail. For
example, Vaccaro et al. (1968) have shown that
the concentration of glucose in certain produc-
tive areas of the Atlantic could be correlated
with phytoplankton productivity.
The soluble small molecular weight fraction in
seawater can be removed by the permease systems
of bacteria which have been shown to work even
in the low concentrations prevailing in seawater
(McIntyre et al., 1970). This leads to a fast
turnover time of this fraction. However, it may
be that the larger soluble molecules are not
accessible to heterotrophic action because of
the large energy requirement necessary to pro-
duce the extracellular enzymes necessary to
break these down.
Temperature does not seem to be a major factor in
controlling primary productivity (Nienhuis, 1981).
Rates of primary productivity in the open ocean
have been found to be of comparable magnitude
in widely differing latitudes. In a comparative
study between tropical and temperate beaches
Ansell et al. (1978) attributed the higher turn-
over rate of carbon in the tropical beach to a
higher input rate and not to temperature.
Very little work has been done on the role of de-
tritus in sandy beach ecosystems. There are two
important aspects, the time scale over which it
is deposited and the patchiness with which it
occurs (M. Bolter, personal communication). The
detritus can be divided into (a) the food source
and potential faeces directly available to the
microheterotrophs, particularly low molecular
weight compounds such as monosaccharides and
amino acids, and (b) potential food sources such
as higher molecular weight substances not directly
available, e.g. humic compounds. The ratio of
high molecular weight organic material (charac-
terised by a high C:N ratio) to low molecular
weight organic material is therefore of great
importance, particularly in a nitrogen-deficient
environment, where organic nitrogen must be
utiliseable by microheterotrophs. Many of the
problems associated with this type of study are
methodological ones.
Degradation of detritus starts with hydrolytic
cleavage of the particulate material into smaller
molecules which can be assimilated by the bacteria.
The organic matter first loses nitrogen, phosphorus
and oxygen, (Sharp, 1975; Cauwet, 1981; Froelich
et al., 1982). The end products of extracellular
hydrolysis are amino acids, mono- and disac-
charides and long chain fatty acids. These are
taken up directly by heterotrophic bacteria. A
fraction of the detritus is never completely re-
mineralised but accumulates mainly within the
anoxic environment. Here it is gradually trans-
formed into organic complexes which are re-
fractory to microbial attack.
The measurement of particulate organic carbon
(PaC) does not normally take into account the
difference between detritus and living matter.

This POC normally makes up about 75-80% of the
total suspended matter of oceanic water. However,
in the surf zone of sandy beaches the proportion
is likely to be much lower. Although some par-
ticulate organic matter may be brought into the
surf zone through land runoff via rivers or from
the atmosphere, the major part seems to originate
from in situ production of living organisms. Of
the particulate organic material, only a small
proportion consists of living cells, although
it must all ultimately have been produced by
living organisms.
Sandy beach ecosystems derive their carbon almost
entirely from the sea with very little input from
the land (McLachlan et al. 1981a). The quantity
of organic material reaching a beach from the sea
varies widely with location and environmental
conditions. It frequently also fluctuates sea-
sonally. For example, on some South African west
coast beaches which are adjacent to kelp beds,
winter storms may cause the deposition of large
amounts of kelp on the shore. Estimates by Koop,
Field (1980) suggest that between 1 200 and
1 800 x 10 3 kg wet mass of Ecklonia maxima are
cast up annually along a 1 km strandline from a
kelp bed at Kommetjie. This represents an aver-
age of 206 g C.m-I.d- I (Koop et aI., 1982b).
By far the majority of this occurs in the two
months of July and August. On the other hand,
McLachlan et al. (1981a) have estimated that on
eastern Cape beaches, where there are no macro-
algae, the total carbon input (made up of phyto-
plankton, detritus and carrion) is 46 g C.m-I.d- I
The amount of particulate organic material trapped
within a beach depends on the physical structure
of the sand particles. Finer beaches would be
expected to trap more particulate matter. This
would also be expected to have a longer residence
time in the beach, making it more susceptible to
microbial attack.
215
This has been found in a comparison of the carbon
distribution in a coarse and fine beach in North
Wales (Pugh et al., 1974). A seasonal variation
in carbon occurred, particularly on the fine beach.
A maximum carbon concentration occurred during
summer and autumn. In the fine beach there was
a reduction in carbon with depth, a feature not
observed on the coarser beach. The amount of
carbon associated with the fine beach was higher
than that in the coarse, an observation which
has also been made elsewhere (e.g. Orren et al.,
1981a).
The interstitial food webb starts with dissolved
and particulate matter flushed into the pore
spaces by tides and waves (Riedl, 1971; McLachlan,
1980). This organic matter is predominantly
attached to the sand grains. In a sandy beach
in Loch Ewe, Scotland, Steele, Baird (1968)
found that the distribution of POC in the top
2 cm of sand showed a close relationship with
chlorophyll a. However, this relationship did
not hold when the vertical distribution in the
sand was examined. At all the deeper stations
there was a rapid decrease in chlorophyll a with
depth but a much slower change in carbon. This
may be due to a more even bacterial distribution
with depth, as was observed by Koop, Griffiths
(1982).
In a tropical Indian beach (Munro et al., 1978)
the carbon depth profiles were reasonably uniform
but there was a definite trend with carbon in-
• creasing by a factor of 2 from high- to mid-tide
levels on the beach and thereafter remaining
constant.
As with nitrogen a variety of forms of experi-
mental sand columns have been used effectively
in studying the carbon cycle in beaches. In
the sand column experiment described above,
McIntyre et al. (1970) found greater activity
in the top 5 cm of their column suggesting that
216
the available organic material is easily as-
similated by bacteria in the surface layers.
Added glucose further stimulated oxygen intake
in the top 4 cm of sand indicating enhanced
bacterial activity.
Boucher, Chamroux (1976) used an amino acid
mixture as a carbon source for their experi-
mental sand columns since they found that this
was a better organic support than glucose, which
is not assimilable by a large number of bacteria.
There is however some disagreement as to whether
this is a suitable substrate to use, since it is
more easily assimilable than the range of carbon
compounds normally found in the marine environment
(McLachlan, 1981b). As a result it was shown
that rapid and total mineralisation of amino
acids occurred, suggesting that amino acids can
be one of the limiting factors in bacterial
growth (Boucher, Chamroux, 1976). The highest
bacterial density was again in the upper 2 cm of
sediment. Similarly Munro et al. (1978) found a
four-times higher respiration rate in the top
3.5 cm of experimental column. The ingoing sea-
water was rich in POCo
A 14C method has been developed for measuring
net uptake and respiration by bacteria in cores
(Meyer-Reil, 1978). A similar method has been
used to measure in situ rates of 14C glucose
uptake in a tidal flat in New Zealand (Gillespie,
MacKenzie, 1981). It was found that the net
uptake and mineralisation in undisturbed cores
were so rapid that short incubation times
(1-7 minutes) were necessary in order to measure
uptake (Meyer-Rei I , 1978). Uptake rate was
further increased when cores of disturbed or
mixed sediment were used, and the respiration
rate was higher in aerobic cores.
An attempt to study the carbon flow through a
sandy bea~h microcosm, described above, has been
undertaken by.Koop et al. (1982a). Clearly the
biomass of bacteria was related to the concen-
tration of organic matter, leached from the kelp.
This is known to contain high concentrations of
mannitol, laminarin, alginates and other carbo-
hydrates (Newell et al., 1980). The bacterial
component in the microcosm was capable of rapidly
mineralising much of this carbon from the kelp
debris before it was returned to the sea.
D-mannitol (which may reach 5-11% of the dry
mass) was very rapidly used and did not even
appear in the initial leachates.
The bacteria were therefore using both complex
structural carbon compounds in the leachates
as well as simpler soluble carbohydrates, and
leaving the inorganic components to be returned
to the kelp bed by drainage. The carbon flow
analysis for this decomposer-dominated system
suggested that 27% of the kelp carbon was as-
similated into bacterial carbon, while some 69%
was mineralised, presumably lost to the atmos-
phere as C02 (Koop et al., 1981b). This was
compared with a nearby consumer-dominated beach
(Griffiths, Stenton-Dozey, 1981). Even when the
direct consumption of kelp by grazers was high,
much of this was returned to the ecosystem as
faeces. The total bacterial carbon was found
to be 23% of kelp carbon while the remaining
amount was mineralised.
Koop, Griffiths (1982) found that very strong
gradients of DOM occur below the kelp drift line.
The organic matter is used within the first metre
or so of sand. These beaches are perhaps a
special case in view of the large input of kelp
debris.
In general, organic material has been considered
collectively as "carbon" and there has been
little work on individual classes of compounds.
Apart from the few cases already mentioned,
Bolter et al. (1981) have measured glucose,
fructose, ribose and total carbohydrates, as well

as uptake rates (among other things) in seawater
and sediments of sandy beaches in the Baltic Sea
near Kiel. Correlation studies of all these
parameters led them to suggest that glucose plays
a major role and may be regarded as a good re-
presentative for organic nutrient sources. It
amounts to 0.3% of the dissolved organic carbon.
The carbon cycle can therefore be summarised
briefly as follows: carbon, in the form of car-
bon dioxide, is fixed by primary producers.
Secondary production results from the consumption
of these primary producers. Destruction and
mineralisation of faecal material and dead or-
ganisms restores carbon dioxide and nutrients
to the water. This cycle is not perfect and
some organic matter escapes as soluble or par-
ticulate matter.
5. PHOSPHORUS
Phosphorus is an essential nutrient in most eco-
systems, being required for biological synthesis
and energy transfer processes. In solution,
phosphorus occurs predominantly in the +5 oxi-
dation state, as the tetrahedral phosphate units
H2P04- and HP042-. It is in these forms that
phosphate is taken up by organisms. Natural
waters also contain organophosphorus compounds
which make up a significant proportion of the
soluble phosphorus content. These are generally
of unknown composition. Although phosphorus
compounds in a number of other oxidation states
are known, they are generally unimportant under
environmental conditions.
The sorption and release of phosphorus is af-
fected by, among other factors, the oxidation-
reduction status of sediments. This has been
studied extensively in estuaries (e.g. Butler,
Tibbitts, 1972), and the ocean (Froelich et al.,
1982). A paper dealing with the behaviour of
phosphorus in soils is relevant here (Patrick,
217
Khalid, 1974). Although the mechanism by which
phosphorus is removed from solutions by sediments
is not clearly understood, it is thought to be a
sorption rather than a precipitation process.
This is attributed to a gel complex consisting
of hydrated iron oxides. Anaerobic sediments
have up to a hundred times more iron in solution
than aerobic sediments. The oxidation state of
iron compounds affects the phosphorus equilibrium
between solid and solution. Phosphate, which is
co-precipitated or occluded in ferric oxyhy-
droxide in an aerobic sediment, does not exchange
with the solution as readily as in an anaerobic
sediment. Ferric oxyhydroxide is capable of
binding orthophosphate ions more firmly than the
ferrous form. Changes in the hydrated iron ox-
ides can therefore be expected to change the
concentration of dissolved orthophosphate. This
has been found in a laboratory study (DeLaune
et al., 1981).
With this as background, the dearth of studies
concerning the role of phosphorus in sandy beach
ecosystems comes as something of a surprise and
forces one to consider whether phosphorus is
important in sandy beaches. There is no doubt
that phosphorus is one of the most important
elements in freshwater systems, to the extent
that models are sometimes based on a currency of
phosphorus, (e.g. Edmondson, 1970). Some possible
reasons for this low status of phosphorus in
sandy beach studies will be discussed briefly.
As mentioned earlier, there is usually a surplus
of phosphorus relative to nitrogen in coastal
water (Ryther, Dunstan, 1971; Lewin et al., 1975).
This means that phosphorus seldom becomes limit-
ing in the coastal zone. In a situation where
there is an excess of phosphorus in an ecosystem
which is, however, limited by other elements,
the role of phosphorus in that ~ystem becomes
less important.
218
In a study of sandy beaches in Kiel Fjord and
Kiel Bight, Bolter et al. (1981) studied the
interaction of 33 parameters (physical, chemical,
planktological and microbiological). Of these,
only two, including phosphate, did not correlate
with any others. This is a further indication
that phosphate was not limiting in this case.
In an experiment to examine the fate of phos-
phorus in experimental columns (Wormald, Stirling,
1979) the levels of phosphorus in the effluent
water were found to remain fairly constant in
spite of large fluctuations in the influent
concentrations. The percentage removal, i.e.
sediment uptake, of phosphate was greatest when
the influent concentrations were highest, while
there was production of phosphate when influents
were not enriched. This suggests that there is
a reversible physical adsorption or "buffering"
mechanism for phosphorus in the sand columns.
A similar "buffering" mechanism has been proposed
for estuaries (Butler, Tibbitts, 1972; Bray et
al., 1973; Liss, 1976). Phosphate removal by
adsorption seems to occur when the influent
concentrations exceed some equilibrium value,
while below this, production of phosphate may
be caused by desorption. If this is the case,
the adsorption capacity of a sediment would
depend on the clay mineralogy and would obviously
also depend on the redox conditions described
above.
In their sand columns Wormald, Stirling (1979)
found no change in effluent nutrient concen-
trations after removal of bacteria by addition
of antibiotics. This is a surprising result
but perhaps this was due to some type of physical
adsorption reaction of the nature described above,
or else incomplete removal of bacteria, as stated
earlier.
In general the phosphate concentrations of inter-
stitial water have been found to be about double
or three times those of overlying water (Pugh
et al., 1974; Orren et al., 1981a). This has
also been found in sand columns and model beaches
(Wormald, Stirling, 1979; Pugh, 1975, 1976).
In summary, phosphorus is known to be an essen-
tial nutrient in sandy beaches. However, it does
not normally appear to be limiting in the eco-
system, probably as a result of some form of
"buffering" mechanism. This ensures that the
concentration of dissolved phosphorus does not
fall below a certain level. The mechanism of
this process is probably related to the formation
of hydrated iron oxides, the solubility of which
are governed by the redox conditions of the beach.
However, such processes are at present unquan-
tified and remain an uncertainty in the phosphorus
cycle.
6. OTHER CONSIDERATIONS
A few other factors will be considered briefly.
The pH of water is known to affect a number of
variables, e.g. dissolved organic carbon, am-
monia concentration, as well as some other ions,
including iron and manganese (DeLaune et al.,
1981). However, while the pH is likely to be of
importance in estuaries or in freshwater, it
rarely falls outside the limits of 7.8 - 8.2
in the ocean. Immediate control of the pH of
seawater is exerted by the carbonic acid buf-
fering system, and to a lesser extent by the
boric acid equilibrium. The only beaches likely
to be subjected to large pH fluctuations are
highly polluted ones which are not being con-
sidered here.
Similarly the salinity of sandy beaches is un-
likely to vary markedly enough to cause appre-
ciable chemical changes, particularly in the
sand where salinity changes are far slower than
those in overlying water (Chapman, 1981). The
large amount of published literature dealing

with chemical changes in estuaries, related to
changes in salinity, will not be considered here.
However it should be mentioned that in their
wide ranging study in Kiel Fjord and Kiel Bight,
Bolter et al. (1981) found that salinity had a
marked effect on other parameters including
microbial activity. The low salinity of the
Baltic Sea however, makes this a special case.
Sulphur is an element which is important under
highly reducing conditions such as are found in
salt marshes, some estuaries and offshore sedi-
ments. However, in aerobic sandy beaches
sulphur is less likely to play an important role.
The sulphur cycle in anoxic sediments has been
dealt with elsewhere (e.g. Fenchel, Riedl, 1970;
Jorgensen, 1977) and is not included here.
7. DISCUSSION
From the above discussion of the cycles of
selected elements, it is immediately clear that
the recycling of these elements is of funda-
mental importance to the ecological processes
operative in beaches. The main source of energy
for exposed sandy beaches is organic matter
originating from the sea, either in the form of
phytoplankton (Lewin et al., 1975, 1979;
McLachlan, 1980-81; McLachlan ~t al., 1981a),
macrophytes (Koop et al., 1981a, 1981b), or
dissolved. It is also clear that individual
parameters cannot be regarded as major control-
ling factors in a complex ecosystem (Bolter et
aI., 1981).
Some factors are common to most beaches. Con-
centrations of inorganic nutrients are invariably
found to be higher (usually by between 2- and
10-fold) in interstitial water than in the over-
lying seawater (Oliff et al., 1967a, 1967b, 1970;
Pugh et al., 1974; Vosjan, Olanczuk-Neyman, 1977;
McLachlan, 1977; Zeitzschel, 1979; Meyer-Reil et
al., 1980; Rutgers van der Loeff, 1980; Rutgers
219
van der Loeff et al., 1981; Orren et al., 1981a;
and many others). Dissolved inorganic nutrients
in interstitial water are frequently linked to
each other (Meyer-Reil et al., 1980; Orren et
aI., 1981a).
Another common property is the high biological
activity in the sediment surface layers. The
depth of this highly active layer varies from
1 or 2 cm (e.g. Steele, Baird, 1968) to more
than 50 cm (McLachlan, 1982). The main factors
affecting this layer are the degree of oxygena-
tion, sediment particle size and water flow-
through rate (McLachlan, 1979), desiccation
(Dye, 1980), organic content (Koop et al.,1981b)
and others.
The oxygen status of a beach is an extremely
important factor in the cycling of all other
elements. Some attempts have been made to re-
late this to both nitrogen and carbon cycles.
It appears that there is good correlation between
the amount of inorganic nitrogen released and
the oxygen taken up (Nixon, 1981). Generally,
oxygen levels regulate decomposition, conversion
and energy transfer from detritus to consumers.
Hanson (1982) has attempted to relate the or-
ganic nitrogen and caloric contents of different
forms of detritus in detrital food chains. His
experimental results predict that the rate of
detrital enrichment of the benthos is an impor-
tant variable in the decomposition of most sea-
weeds, but that available calories, along with
organic nitrogen, limit decomposition. He was
able to show therefore that microbial biomass
and activity are functions of both organic ni-
trogen content and available calories.
On the basis of total interstitial respiration,
McLachlan et al. (1981a) estimated the amount
of organics which must be mineral'ised by the
interstitial fauna. This was done on the as-
220
sumption that the interstitial systems of sandy
beaches are closed systems in terms of predation.
Therefore total respiration must equal total
assimilation. This was also studied in sand
columns (McLachlan et al., 198Ib). The results
indicated that the measured rates of oxidation
of organic nitrogen and nitrate production ac-
count for only about 60% of the oxygen uptake by
the columns, compared with 60-90% obtained by
McIntyre et al. (1970), and 100% obtained by
Boucher, Chamroux, (1976). These differences
may be accounted for by variations in the assumed
C:N ratio.
Physical factors are known to have significant
influences on the chemistry. There is a highly
significant negative correlation between bac-
terial biomass and mean grain size (Dale, 1974).
This includes both aerobes and anaerobes. This
is probably based on the greater surface area
available for colonization in fine sediments.
Particle size also has a definite influence on
the rate of percolation of water through the
sediment interstices. This influences the rate
of supply of organic material and oxygen.
McLachlan et al. (198Ia) have suggested that
one of the most significant features of sandy
beach ecosystems is the extensive production of
nutrients which are ultimately exported to the
sea in repayment of organic imports. They have
estimated that the total amount of nitrogen in
the surf zone of eastern Cape beaches (average
200 m wide and I m deep) is regenerated by the
beach every eight days. Of this, macrofauna
account for 37% and interstitial fauna for 63%.
Their findings support the results of Lewin et
al. (1975), who have suggested that the popu-
lations of surf zone diatoms are not nutrient
limited. They found however that nitrogen may,
on occasions, be in short supply, particularly
in summer when nitrate concentrations fall to un-
detectable levels. At these times, ammonia ge-
nerated by macrofauna has been shown to be the
most significant nitrogen source.
Koop et al. (1982b) have also suggested that
nitrogen leaching to sea from piles of kelp wrack
is dominated by ammonia and that this is there-
fore a potential source of nitrogen for nearshore
primary producers. However, their work indicated
that from the decomposition of kelp wrack on the
beach, very little is returned in the form of
nutrients to the surf zone and that in the short
term most of the kelp was converted into bac-
terial biomass. This bacterial biomass must,
however, ultimately be remineralized and returned
to the sea. Hennig et al. (1983) have also sug-
gested that the quantity of nutrient returned to
the surf from the beach is insignificant.
A number of problems will have to be addressed
in the future. One of these is the role which
experimental sand columns can play in chemical
research related to sandy beaches. This review
has shown that a large proportion of the avail-
able chemical information has been obtained from
such experimental systems. Whether this is sim-
ply because this type of study has been under-
taken in preference to field studies in view of
the difficulties associated with the latter, is
an open question. Nevertheless the fact remains
that experimental systems are capable of supply-
ing a large amount of useful information in
both long- and short-term experiments. The de-
sign of sand columns has evolved considerably
over the last decade. Some of the first were
those of McIntyre et al. (1970) in which large
columns were subjected to continuous water flow.
It is perhaps to be expected that this type of
system will, over time, become more subtidal in
nature. These have given way to the more na-
tural system of McLachlan et al. (198Ib) in
which the water is passed through the column in

pulses, simulating tides.
The fact that most of the biological activity
has been detected in the top few centimetres of
experimental columns leads one to speculate
whether columns will not become very much
smaller in the future, similar to those of
Munro et al. (1978). While such columns may
not fulfil biological requirements, they are
likely to supply relatively quick answers to
some chemical questions (e.g. Meyer-Reil, 1978).
If the columns are to be really useful, they
must be of such a scale that edge effects are
not significant. In the case of the chemist,
this is likely to permit smaller columns.
However, in spite of their considerable poten-
tial, there is no doubt that results obtained
with any type of model will require field veri-
fication before they are truly acceptable.
Limitations to experimental columns continue to
be lack of mixing of the surface layers by the
swash zone.
Most studies conducted so far have been related
to some aspect of nitrogen cycling, followed
fairly closely by carbon studies. These are
both linked very closely to the oxygen status
of the beach, although where the beach is well
oxygenated, oxygen assumes a .less important
role. The nitrogen and carbon cycles cannot
be separated since the majority of nitrogen
in sandy beaches is supplied in the form of
organic compounds and nitrogen must first be
released through heterotrophic bacterial action.
On the other hand, carbon remains the universal
currency on which most energy calculations are
based. Up to the present all carbon compounds
have very largely been considered simply as
"carbon". The future of research in this area
probably lies in the separation of carbon into a
few classes of the more important carbon com-
pounds. The work of Meyer-Reil (1978), Meyer-
221
Reil et al. (1980), Bolter et al. (1981) on
glucose and other monosaccharides and of Koop et
al. (1981a, 1981b) on mannitol give the lead
in this regard. The role of colloid material
also merits further attention (Farke, Riemann,
1980).
Phosphorus has yet to be shown to be limiting in
sandy beach ecosystems. This is because of its
rate of supply and recycling in relation to up-
take in sandy beaches, and also because of the
buffering mechanism referred to above. This
insures a minimum supply of phosphate to the
system. For the present, therefore, phosphorus
must be relegated to a position of lesser im-
portance.
Throughout the work discussed in the present
paper, one point has emerged very clearly, and
that is the role of chemical analysis in eco-
logical work of this nature. Herein lies the
value of the chemist. The fact that elements
have simply been lumped together as "carbon" or
"nitrogen" represents a relatively simplistic
approach to the chemistry of ecosystems. Even
at this level many analytical problems remain to
be solved, e.g. the determination of organic
carbon in the presence of carbonate. With in-
creasing sophistication, ecological research
will require an increasingly sophisticated
chemical approach. There has been no work to
date on chemical processes in sandy beaches.
Key factors are the rates of appearance and dis-
appearance of nutrients and other compounds, not
simply their concentrations at any time. There-
fore in addition to becoming indispensable to
the ecologist, the analytical chemist will find
a myriad of chemical problems in the coastal zone
of interest to himself.
222

8 ACKNOWLEDGEMENTS DeLaune RD, Reddy CN and Patrick WH Jr (1981)

Effect of pH and redox potential on concentration
of dissolved nutrients in an estuarine sediment.
I would like to thank Drs D. Lord and A McLachlan J. Environ. Qual. 10(3), 276-279.
for many useful comments on this paper and Mrs Dugdale RC, Goering JJ and Ryther JH (1964)
High nitrogen fixation rates in the Sargasso Sea
S Macintosh for typing the manuscript. and the Arabian Sea. Limnol. Oceanogr. 9, 507-510.
Dugdale RC and Goering JJ (1967) Uptake of new
and regenerated forms of nitrogen in primary pro-
9 REFERENCES duction. Limnol. Oceanogr. 12, 196-206.
Dye AH (1979) Measurement of biological oxygen
Aller RC (1980) Diagenetic processes near the
demand in sandy beaches. S. Afr. J. Zool. 14,
sediment-water interface of Long Island Sound.
55-60.
II. Fe and Mn. Adv. in Geophys. 22, 351-415.
Dye AH (1980) Tidal fluctuations in biological
Ansell AD, McLusky DS, Stirling A and
oxygen demand in exposed sandy beaches. Estuar.
Trevallion A (I978) Production and energy flow
Coast. Mar. Sci. 11, 1-8.
in the macrobenthos of two sandy beaches in
Dye AH (1981) A study of benthic oxygen con-
south west India. Proc. Roy. Soc. Edinburgh 76B,
sumption on exposed sandy beaches. Estuar. Coast.
269-296.
Shelf Sci. 13, 671-680.
Bailey RA, Clark HM, Ferris JP, Krause Sand
Edmondson WT (1970) Phosphorus, nitrogen and
Strong RL (1978) Chemistry of the environment.
algae in Lake Washington after diversion of
New York, Academic Press, p 371.
sewage. Science 169, 690-691.
Billen G (1978) A budget of nitrogen recycling
Farke H and Riemann F (1980) Dissolved organic
in North Sea sediments off the Belgian coast.
carbon in littoral sediments: concentrations and
Estuar. Coast. Mar. Sci. 7, 127-146.
available amounts demonstrated by the percolation
Bolter M, Meyer-Reil L-A, Dawson R, Liebezeit
method. Veroff. Inst. Meeresforsch. Bremerh.
G, Wolter K and Szwerinski H. (1981) ~tructure
18, 235-244.
analysis of shallow water ecosystems: lnter-
Fenchel TM and Riedl RJ (1970). The sulfide
action of microbiological, chemical and physical
system: a new biotic community underneath the
characteristics measured in the overlying waters
oxidized layer of marine sand bottoms. Mar.
of sandy beach sediments. Estuar. Coast. Shelf
BioI. 7, 255-265.
Sci. 13, 579-589.
Fricke AH, Eagle GA, Gledhill WJ, Greenwood PJ
Boucher G and Chamroux S (1976) Bacteria and
and Orren MJ (1979) Preliminary pollution surveys
meiofauna in an experimental sand ecosystem. I.
around the south-western Cape coast. Part 3,
Material and preliminary results. J. expo mar.
Hout Bay. S. Afr. J. Sci. 75, 459-461.
Biol. Ecol. 24, 237-249.
Froelich PN, Bender ML, Luedtke NA, Heath GR
Bray JT, Bricker OP and Troup BD ~1973) .Phos-
and De Vries T (1982) The marine phosphorus
phate in interstitial water of anOXlC sedlments:
cycle. Am. J. Sci. 282(4),474-511. .
oxidation effects during sampling procedures.
Gillespie PA and MacKenzie AL (1981) Autotrophlc
Science 180, 1362-1364.
and heterotrophic processes on an intertidal mud-
Brezonik PL (1977) Denitrification in natural
sand flat, Delaware Inlet, Nelson, New Zealand.
waters. Prog. Wat. Tech. 8 (4/5), 373-392.
Bull. Mar. Sci. 31(3). 648-657.
Buresh RJ and Patrick WH Jr (1981) Nitrate
Goering JJ, Dugdale RC and Menzel DW (1966)
reduction to ammonium and organic nitrogen in an
Estimates of in situ rates of nitrogen uptake by
estuarine sediment. Soil Biol. Biochem. 13,
Trichodesmium sp. in the tropical Atlantic Ocean.
279-283.
Limnol. Oceanogr. 11, 614-620.
Butler ET and Tibbitts S (1972) Chemical survey
Goreau TJ, Kaplan WA, Wofsy SC, McElroy MB,
of the Tamar estuary. I. Properties of the waters.
Valois FW and Watson SW (1980) Production of
J. mar. biol. Ass. U K 52, 681-699.
N02- and N20 by nitrifying bacteria at reduced
Cauwet G (1981) Non-living particulate matter.
concentrations of oxygen. Appl. Environ.
In: Duursma EK, Dawson R, eds., Marine organic
Microbiol. 40(3), 526-532.
chemistry, pp 71-89, New York, Elsevier.
Griffiths CL and Stenton-Dozey J (1981) The
Chapman PM (1981) Measurement of the short-term
fauna and rate of degradation of stranded kelp.
stability of interstitial salinities in subtidal
Estuar. Coast. Shelf Sci. 12(6), 645-653.
estuarine sediments. Estuar. Coast. Shelf Sci.
Hanson RB (1982) Organic nitrogen and calorie
12, 67-81.
content of detritus. II: Microbial biomass and
Chiaudani G and Vighi M (1982). Multistep
activity. Estuar. Coast. Shelf Sci. 14, 325-336.
approach to identification of limiting nutrients
Hargrave BT (1972) Aerobic decomposition of
in northern Adriatic eutrophied coastal waters.
sediment and detritus as a function of particle
Water Res. 16(7) 1161-1166.
surface area and organic content. Limnol. Oceanogr.
Dale NG (1974) Bacteria in intertidal sediments:
17(4), 583-596.
factors related to their distribution. Limnol.
Oceanogr. 19(3), 509-518.

Hennig HF-KO, Fricke AH and Martin CT (1983)
The effect of meiofauna and bacteria on nutrient
cycles in sandy beaches. This volume.
Johannes RE (1969) Nutrient regeneration in
lakes and oceans. In: Droop MR and Wood EJF,
eds. Advances in the microbiology of the sea,
Vol. 1. pp 203-212, New York, Academic Press.
Jorgensen BB (1977) The sulphur cycle of a
coastal marine sediment (Limfjorden Denmark).
Limnol. Oceanogr. 22(5), 814-832.
Koop K and Carter RA (1982) Mannitol-fermenting
bacteria as evidence for export from kelp beds.
Limnol. Oceanogr. 27(5), 950-954.
Koop K and Field JG (1980) The influence of
food availability on population dynamics of a
supra-littoral isopod, Ligia dilatata Brandt. J.
expo mar. BioI. Ecol. 48, 61-72.
Koop K and Griffiths CL (1982) The relative
significance of bacteria, meio- and macrofauna
on an exposed sandy beach. Mar. BioI. 66, 295-300.
Koop K, Newell RC and Lucas MI,(1982a) Bio-
degradation and carbon flow based on kelp (Eck-
lonia maxima) debris in a sandy beach microcosm.
Mar. Ecol. Progr. Ser. 7, 315-326.
Koop K, Newell RC and Lucas MI (1982b) Micro-
bial regeneration of nutrients from the decom-
position of macrophyte debris on the shore. Mar.
Ecol. Progr. Ser. 9, 91-96.
Lewin J, Eckman JE and Ware GN (1979) Blooms of
surf-zone diatoms along the coast of the Olympic
Peninsula, Washington. XI. Regeneration of am-
monium in the surf environment by the Pacific
razor clam Siliqua patula. Mar. BioI. 52, 1-9.
Lewin J, Hruby T and Mackas D (1975) Blooms of
surf-zone diatoms along the coast of the Olympic
Peninsula, Washington. V. Environmental conditions
associated with the blooms (1971 and 1972).
Estuar. Coa.st. Mar. Sci. 3, 229-241.
Liss PS (1976) Conservative and non-conservative
behaviour of dissolved constituents during es-
tuarine mixing. In: Burton JD and Liss PS, eds.
Estuarine chemistry, pp. 93-130, London,
Academic Press.
McCarthy JJ, Taylor WR and Taft JL (1975) The
dynamics of nitrogen and phosphorus cycling in
the open waters of Chesapeake Bay. In: Church
TM, ed. Marine chemistry in the coastal environ-
ment, pp 664-681, Washington, American Chemical
Society.
McIntyre AD, Munro ALS and Steele JH (1970)
Energy flow in a sand ecosystem. In: Steele JH,
ed. Marine food chains, pp 19-31, Edinburgh,
Oliver and Boyd.
McLachlan A (1977) Studies on the psammolittoral
meiofauna of Algoa Bay, South Africa. I. Physical
and chemical evaluation of the beaches.
Zool. Africana 12(1), 15-32.
McLachlan A (1978) A quantitative analysis of
the meiofauna and the chemistry of the redox
potential discontinuity zone in a sheltered sandy
beach. Estuar. Coast. Mar. Sci. 7, 275-290.
McLachlan A (1979) Volumes of sea water filtered
through eastern Cape sandy beaches. S. Afr. J. Sci.
75, 75-79.
223
McLachlan A (1980) The definition of sandy
beaches in relation to exposure: a simple rating
system. S. Afr. J. Sci. 76, 137-138.
McLachlan A (1980-81) Exposed sandy beaches as
semi-enclosed ecosystems. Mar. Environm. Res. 4,
59-63.
McLachlan A (1982) A model for the estimation
of water filtration and nutrient regeneration by
exposed sandy beaches. Mar. Environm. Res. 6,
37-47.
McLachlan A, Dye A and Harty B (1981b) Simu-
lation of the interstitial system of exposed
sandy beaches. Estuar. Coast. Shelf Sci. 12,
267-278.
McLachlan A, Dye AH and van der Ryst P (1979)
Vertical gradients in the fauna and oxidation of
two exposed sandy beaches. S. Afr. J. Zool. 14,
43-47.
McLachlan A, Erasmus T, Dye AH, Wooldridge T,
van der Horst G, Rossouw G, Lasiak TA and
McGwynne L (1981a) Sand beach energetics: an
ecosystem approach towards a high energy inter-
face. Estuar. Coast. Shelf Sci. 13, 11-25.
Mevel G and Chamroux S (1981) An experimental
study of the regulation of nitrification in
marine sediments. Oceanologica Acta 4(4),
457-463.
Meyer-Reil L-A (1978) Uptake of glucose by bac-
teria in the sediment. Mar. BioI. 44, 293-298.
Meyer-Reil L-A, Bolter M, Dawson R, Liebezeit G,
Szwerinski H and Wolter K (1980) Interrelation-
ships between microbiological and chemical para-
meters of sandy beach sediments, a summer aspect.
Appl. Environ. Micro. 39(4), 797-802.
Munro ALS, Wells JBJ and McIntyre AD (1978)
Energy flow in the flora and meiofauna of sandy
beaches. Proc. Roy. Soc. Edinburgh 76B, 297-315.
Nedwell DB (1982) Exchange of nitrate, and the
products of bacterial nitrate reduction, between
seawater and sediment from a UK saltmarsh.
Estuar. Coast. Shelf Sci. 14, 557-566.
Newell RC, Lucas MI, Velimirov Band Seiderer
LJ (1980) Quantitative significance of dissolved
organic losses following fragmentation of kelp
(Ecklonia maxima and Laminaria pallida). Mar.
Ecol. Progr. Ser. 2, 45-59.
Nienhuis PH (1981) Distribution of organic
matter in living marine organisms. In: Duursma EK
and Dawson R, eds. Marine organic chemistry,
pp 31-69, New York, Elsevier.
Nishio T, Koike I and Hattori A (1982) Denitri-
fication, nitrate reduction, and oxygen consump-
tion in coastal and estuarine sediments. Appl.
Environ. Micro. 43(3), 648-653.
Nixon SW (1981) Remineralisation and nutrient
cycling in coastal marine ecosystems. In: Nielson
BJ and Cronin LE, eds. Estuaries and nutrients,
pp. 111-138, Humana Press.
Oliff WD, Berrisford CD, Turner WD, Ballard JA
and McWilliam DC (1967a). The ecology and che-
mistry of sandy beach and nearshore submarine
sediments of Natal. I. Pollution criteria for
sandy beaches in Natal. Water Res. 1, 115-129.
224
Oliff WD, Berrisford CD, Turner WD, Ballard JA
and McWilliam DC (1967b). The ecology and che-
mistry of sandy beaches and nearshore submarine
sediments of Natal. II. Pollution criteria for
sandy beaches in Natal. Water Res. I, 131-146.
Oliff WD, Gardner BD, Turner WD and Sharp JB
(1970) The chemistry of the interstitial waters
as a measure of conditions in a sandy beach.
Water Res. 4, 179-188.
Orren MJ, Eagle GA, Fricke AH, Gledhill WJ,
Greenwood PJ and Hennig H F-K 0 (198Ia) The
chemistry and meiofauna of some unpolluted sandy
beaches in South Africa. Water S.A. 7(4), 203-210.
Painter HA (1970) A review of literature on
inorganic nitrogen metabolism in micro-organisms.
Water Res. 4, 393-450. . seawater. Limnol. Oceanogr. 13, 356-360.
Patrick WH and Khalid RA (1974) Phosphate re-
lease and sorption by soils and sediments: effect
of aerobic and anaerobic conditions. Science 186,
53-55.
Pugh KB (1975) A model beach system. J. expo
mar. BioI. EcoI. 18, 197-213.
Pugh KB (1976) An annual cycle, at constant
temperature, in a model sandy beach. I. Nutrient
chemistry. J. expo mar. BioI. Ecol. 22, 179-192.
Pugh KB, Andrews AR, Gibbs CF, Davis SJ and
Floodgate GD (1974) Some physical, chemical and
microbiological characteristics of two beaches of
Anglesey. J. expo mar. Biol. Ecol. 15, 305-333.
Quinlan AV, Lewis T and Hoyt JK (1983) Fouling
of the sandy beaches of Nahant Bay (Massachusetts,
USA) by an abnormal free-living form of the ma-
croalga PiZayeZZa ZittoraZis (Phaeophyta). I.
Habitat characteristics. This volume.
Raine RCT and Patching JW (1980) Aspects of
carbon and nitrogen cycling in a shallow marine
environment. J. expo mar. Biol. Ecol. 47, 127-139.
Riedl RJ (1971) How much seawater passes
through sandy beaches? Int. Revue ges. Hydrobiol.
56(6), 923-946.
Riedl RJ and McMahan EA (1974) High energy
beaches. In: Odum HT, Copeland BJ and McMahan EA,
eds. Coastal ecological systems of the United
States, pp. 180-251, Washington, The Conservation
Foundation.
Riley JP and Chester R (1971) Introduction to
marine chemistry. London, Academic Press, p 204.
Roman MR (1980) Tidal resuspension in Buzzards
Bay, Massachusetts. III: seasonal cycles of
nitrogen and carbon:nitrogen ratios in the seston
and zooplankton. Estuar. Coast. Mar. Sci. II, 9-16.
Rutgers van der Loeff MM (1980) Time variations
in interstitial nutrient concentrations at an
exposed subtidal station in the Dutch Wadden Sea.
Neth. J. Sea Res. 14(2), 123-143.
Rutgers van der Loeff MM, van Es FB, Helder W
and de Vries RTP (1981) Sediment water exchanges
of nutrients and oxygen on tidal flats in the
Ems-Dollard Estuary. Neth. J. Sea Res. 15(1),
113-129.
Ryther JH and Dunstan WM (1971) Nitrogen, phos-
phorus and eutrophication in the coastal marine
environment. Science 171, 1008-1013.
S~rensen' J (1978) Capacity for denitrification
and reduction of nitrate to ammonia in a coastal
marine sediment. Appl. Environ. Micro. 35(2),
301-305.
Sharp JH (1975) Gross analysis of organic matter
in seawater: why, how, and from where. In:
Church TM, ed. Marine chemistry in the coastal
environment, pp 664-696, Washington, American
Chemical Society.
Steele JH and Baird IE (1968) Production ecology
of a sandy beach. Limnol. Oceanogr. 13, 14-25.
Vaccaro RF (1962) The oxidation of ammonia in
seawater. J. du Conseil 27(1), 3-14.
Vaccaro RF, Hicks SE, Jannasch HW and Carey FG
(1968) The occurrence and role of glucose in
Vosjan JH and Olanczuk-Neyman KM (1977) Vertical
distribution of mineralization processes in a
tidal sediment. Neth. J. Sea Res. 11(1), 14-23.
Waksman SA and Carey CL (1933) Role of bacteria
in decomposition of plant and animal residues in
the ocean. Proc. Soc. Exp. BioI. Med. 30, 526-527.
Waksman SA, Carey CL and Reuszer HW (1933a)
Marine bacteria and their role in the cycle of
life in the sea. I. Decomposition of marine plant
and animal residues by bacteria. Biol. Bull.
65(1), 57-79.
Waksman SA, Hotchkiss M and Carey CL (1933b)
Marine bacteria and their role in the cycle of
life in the sea. II. Bacteria concerned in the
cycle of nitrogen in the sea. BioI. Bull. 64,
137-167.
Waksman SA, Reuszer HW, Carey CL, Hotchkiss M
and Renn CE (1933c) Studies on the biology and
chemistry of the Gulf of Maine. III. Bacterio-
logical investigations of the sea water and marine
bottoms. BioI. Bull. 64 (2), 183-205.
Webb KL (1981) Conceptual models and processes
of nutrient cycling in estuaries. In: Neilson BJ
and Cronin LE, eds. Estuaries and nutrients,
pp. 25-46, Humana Press.
Wormald AP and Stirling HP (1979) A preliminary
investigation of nutrient enrichment in experi-
mental sand columns and its effect on tropical
intertidal bacteria and meiofauna. Estuar. Coast.
Mar. Sci. 8, 441-453.
Zeitzschel B (1979) Sediment water interactions
in nutrient dynamics. Belle W Baruch Library in
Marine Science, No.ll, Marine Benthic Dynamics
Meeting, Georgetown, USA, 1979. 195-218.
 225

NUTRIENT CYCLING IN SANDY BEACHES

K.B. PUGH (North East River Purification Board, Woodside House, Persley, Aberdeen, U.K.)

INTRODUCTION The tides also impose their own cycles of inun-

The statement (Pugh 1975) that marine sandy
beaches are, for the most part a chemical and mic-
robiological "no man's land" remains true in 1983,
for still the majority of marine chemists and mic-
robiologists have stopped short at the low water
mark, and their soil science colleagues have not
ventured beyond the sand dunes. Between these
boundaries lies a complex inter-relationship of
uncontrollable and continuously varying factors,
e.g. sediments, tides, temperature, light, nut-
rients, etc. Perhaps it is this inherent varia-
bility that has dispelled the courage, and dam-
pened the exploratory urge, of so many chemists
and microbiologists.
FACTORS AFFECTING NUTRIENT CYCLES
If observations and measurements are made at this
fringe the data may be expected to reflect cyclical
influences (see Fig. 1). At the lowest level are
the chemical and microbiological transformations -
the textbook cycles for carbon, nitrogen, phos-
phorus, etc. - fuelled by inputs of detritus,
excreta and man's polluting wastes from the sea.
dation and exposure, not uniformly, but with sub-
cycles of spring and neap tides to change the
length of the intertidal area, to construct and
destruct beach profiles from day to day. Tidal
and diurnal cycles are not always synchronised,
and so emergence occurs in the light and heat of
day for a period, but gradually shifts to the
dark and cold of night, and back again. On a
longer time scale there are annual cycles, the
passage of months through the increasing day
length and air and water temperatures of spring
to the maxima of summer and back again through
the fall to the chill of winter. Seasonal changes
in weather patterns bring storm and calm to re-
arrange beach topographies. And beyond these
there may be cycles of longer time period.
Unrecognised cycles?
"UTUM"

Here on hectares of percolating filter, nature's I Annual (seasonal) cycles

unwanted dissolved and particulate materials are
incubated and, by digestion, fermentation and
metabolism, are converted into useful basic build-
ing materials, both nutrients and biomass, to start
the ascent of the ecological ladder once again,
via protozoa, meiofauna, grazers and consumers.
Superimposed on these cycles are the diurnal ones
of day and night, of light and darkness, the heat
of the noonday sun, the cool of the evening shade. FIGURE 1. Nutrient cycles in sandy beaches.
226
At anyone point in time the beach system will be
found at a different point in each of the cycles.
The cycles will operate continuously as wheels
within wheels. It matters therefore when, and for
how long, a beach is investigated if its nutrient
cycles are to be fully understood, e.g. nitrate
production during an ebbing neap tide at midday
in summer may be quite different from that during
a flooding spring tide at midnight in winter.
IN SITU STUDIES OF NUTRIENT CYCLES IN SANDY
BEACHES
Whilst it is reasonable to expect sandy beaches
to exhibit nutrient cycles their in situ measure-
ment is beset by difficulties. Fig. 2 illustrates
the profiles measured at Newborough Beach,
Anglesey, UK. (For full details see Pugh et al.
1974.) Like most sandy beaches, this one was not
static during the study period and being exposed
experienced extensive movements. Sand waves were
observed to migrate shorewards, their amplitude
being damped as they approached the upper beach.
There was evidence of a long term cycle to these
topographical changes supporting the suggestions
made earlier (see Fig. 1). Such instability may
have accounted for the failure to discern a clear
seasonal pattern of nutrient change (see Fig. 3);
Beach
Fall
\-------'~8.2.71
19_8.71
Metres
Dale
100 200
Distance from Datum - Metres
300
FIGURE 2. Beach topography at Newborough, North
Wales. (See Pugh et al. 1974)
the vigorous reworking of each site between visits
disturbed any long term equilibrium that may have
been established so that in essence return visits
were to sites with new characteristics and con-
tinuity had been lost. The only distinct pattern
amongst the results occurred for the nitrogen con-
centration of the seawater and even that could
have been anticipated on the basis of other work
(Ewins, Spencer 1967). With the springtime devel-
opment of the phytoplankton bloom in this area
the winter concentration of nitrogen was depleted.
The bloom's demand for nutrients kept the seawater
concentrations low until its dieback in the
autumn, whereupon first the ammonium nitrogen
concentration increased, as degradation products
were released into the water column. Then after
an upsurge of nitrifying activity, ammonium con-
centration decreased slightly and nitrate nitrogen
became more abundant.
Throughout the work it was assumed that, since
most life processes are interface activities,
microbial activity producing chemical change would
occur on the surface of sand grains and in the
associated interstitial water (Anderson, Meadows
1969, 1978). Groundwater was therefore viewed as
a resultant of the activity in all the sand grain
microenvironments. Clearly activity occurred and
nutrients were cycled in the sand column but the
nutrient pattern in the groundwater was not seen
to follow the cycle in the seawater. Fig. 3 only
illustrates the situation at two of the eighteen
sites investigated on the beach, but always the
groundwater was nutrient enriched with respect
to the inundating seawater. Analysis was always
of filtered water samples so that the enhanced
nutrient concentration must have reflected solu-
bilisation of particulate material, organic and
inorganic, and its microbial degradation, excre-
tion from beach macrofauna and exudation from
macroflora. Until proven otherwise there remains,
for any beach system, the possibility of sub-
surface nutrient input via freshwater infiltration.
 227

Nitrogen fig at /1
20

10

Phosphate )J9 at/I

Inorganic

Ol~. Or9~nic

M A M A 5 ONOJFMAM

1972 1973

30 l Ni"ogen

20

Ammonium
10

Nitrate

Phosphate ).19 at /1

Inorganic

Organic

1972 1973

SEAWATER

Nitrate

Ammonium

Phosphate }J9 at /1

o l~~~::=:~~~~~~~==~~~S;;:~~~~~::~~__-L~ln~O~,~ga;n~iC Organic

M A M A 5 o N D M A M

1972 1973

FIGURE 3. The nutrlent characteristics of groundwater and

seawater at Newborough Beach. (From Pugh et a1. 1974)
228
There are no other literature reports of attempts
to study such nutrient cycles in situ known to this
author.
THE SHORTCOMINGS AND DIFFICULTIES OF STUDYING
NUTRIENT CYCLES IN SITU
Evidently the approach adopted at Newborough was
not the way to recognise nutrient cycles though it
did give a slightly improved view of beach chem-
istry than was avallable thereto. Most literature
reports involve beach surveillance over very short
periods of time, or even on the basis of a single
visit to one or only a few sites. Such 'snapshots'
may be misleading, for it will be difficult, if
not impossible, to place the observations into a
context of cycles. At the time of such a visit is
a determinand value in its ascendancy, descendancy
or constancy? How frequently should observations
be made? The Newborough study was only a series of
'snapshots'. With hindsight it was powerless to
reveal 'sources' and 'sinks' of various nutrients
and effects on populations because of the beach's
dynamic physical nature. Even the cycles of a much
more stable environment (Llanddona Beach; see
Pugh et al. 1974) were not unveilled by such cur-
sory glances even wlth the ald of statistics.
Ideally continuous short term sampling (say over
tidal cycles) at regular intervals may be more
revealing; but there are logistic difficulties
to such suggestions, not least that the site is
flooded for a part of the experimental period,
and experimental varlables cannot be held constant.
Is the study site chosen representative of the
entire beach? The groundwater at site K at
Newborough was dominated by the ammonium nitrogen
form and yet only 120 metres seaward along the
studied transect the dominant form was nitrate.
Throughout the study the groundwater at site K
exhibited a markedly lower oxygen concentration
relative to all the other sltes. Heterogeneity is
common, e.g. more recently Harty and McLachlan
(1982) hgve shown that nitrate generation capacity
of the sand On King's Beach near Port Elizabeth
varies widely between high, mid and low tide pos-
itlons and within the profile.
THE IMPORTANCE OF NUTRIENT CYCLES AND THE NEED
FOR THEIR STUDY
Slow sand filtration is a device man has been
using to purify his drinking water since the be-
ginning of the nineteenth century (Huisman, Wood
1974). Every cubiC metre of sand of average dia-
meter 0.25 and 38% porosity provides 1.5 hectares
of surface area for bacterial activity and biolog-
ical filtration. The marine fringe provides a vast
area of cleansing filter for the oceans which has
been used since the advent of micro-organisms.
Sandy beaches are "great digestive and incubating
systems" (Pearse et al. 1942) with "self purifying
and self regulating" (Oliff et al. 1970) abilities.
There is presumably a Ilmit, as yet unknown and
undefined, to their self-regulation and a limit to
their ability to purify water trickling through
them. The filtering capacity needs preserving
especially as the chemical insult of the ocean con-
tlnues; the limlts need definition. Before any
damage, real or potential, can be assessed it is
clearly necessary to have knowledge of the mechan-
isms and chemical status of beaches.
Not only does the sandy beach act as a cleansing
filter it provides the base for an important ecol-
ogical niche whose study has captivated many. The
interface receives organics from the sea, sees
their conversion into microbial biomass and lnor-
ganic nutrients, the utilisation of the former and
the return of the latter to the sea via interstit-
ial and groundwater. How much organic matter is
received? How much is converted in one pass through
the sand column? What are the by-products and end-
products? Despite some answers, ecologists are
still having to base their estimates of beach ener-
getics on assumptions rather than direct infor-
mation (McLachlan et al. 1981a).
 229

IN VITRO STUDIES OF NUTRIENT CYCLES IN SANDY

SYSTEMS
In a bld to answer these and related questions,
and in the face of logistic problems, some of which
have been intimated above, researchers have turned
to in vitro studies. Sand columns, disturbed and
"undisturbed", retained in various devices, have
been incubated and irrigated with stored and some-
times filtered seawater, wlth and without nutrient
(organic and inorganic) enrichment, with and with-
out tidal simulation, and nutrient concentrations
have been monitored with a view to understanding
cycles at the transformation level of the scheme
~IGURE 4. The model sandy beach. (See Pugh 1975)
in Fig. 1 (e.g. McIntyre et al. 1970, Johnston 1970
and more recently Wormald, Stirling 1979,
McLachlan et al. 1981b and contributions to this
~ Upper Ii mit of swash

:tt==~~~==~:H=W~M==(n~o=w=~~:e=S=)~~~~~~~LW:=M~~=
symposium). Such experiments have been of Ilmited
duration and have lacked a wave component which
some (McLachlan et al. 1981a, 1981b) consider to Cm
be of underestimated importance. It is not certain o 20 40 60 80 100 120 140 160 180 200 220 240 260

how reliably the results of these experiments can em

be extrapolated to the natural beach environment
and so provide a measure there. They are best FIGURE 5. Dlagram of the model sandy beach showing
the sampling points (.) and the sulphide system
estlmates. (stippled). (From Pugh 1976)

With a view to investigating longer term changes,
larger models (see Fig. 4) containing a large wedge
of sandy material, rather than just a small dia-
meter core, were constructed and using sand from
Newborough Beach proved to be a successful mimic
of that beach system, but without its gross in-
stability (Pugh 1975). A model was supplied with
non-cycled and essentially non-stored (maximum
holding time, 10 min), unfiltered seawater and was
operated under controlled conditions of tide and
wave action. Since the nutrient chemistry of
beaches is affected in part by microbial agency,
and since the bacterial population of stored sea-
water is several orders of magnitude greater than
that of non-stored (Moebus 1972), a continuous
supply of non-stored seawater was deemed advisable.
The water supply came to the model at ambient tem-
perature but was maintained in the model at 5°C
(~ 1°C). Fig. 5 illustrates the experimental system
and shows the permanent position of the stainless
steel capillary tubing used for gravity sampling
of the beach groundwater.
Fig. 6 illustrates the annual cycle of nitrate
nitrogen in the groundwater and seawater in the
model. The pattern of concentration in the sea-
water showed a remarkable similarity to that obser-
ved two years earlier at Newborough (see Fig. 3),
and for the same reason. The beach groundwater
sampled from the high and mid tide positions was
approximately twice as concentrated as the inun-
dating seawater with respect to nitrate nitrogen
throughout much of the year. The trends, whether
increasing or decreasing were gradual, suggesting
a dynamic equilibrium between input and utilisation.
During the summer months however the groundwater
was relatively very rich in nitrate nitrogen.
Within a week, 10-17th June, the concentrations
230

20 1/20/10

10

10

20
....
(J)
.....
-...,.. 30
o N03 - N in the seawater

iii N03 - N gain in the sand profile

I
Ol
::t.
20 1/120/10

z 10
(J)

~
.....
0
z
10

20

Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr
1974 1975

FIGURE 6. The annual cycle of nitrate-nitrogen concentration in the groundwater and seawater of the
model beach, also showing the nitrate gain in the sand profile. For frequency of sampling see Fig. 7.
(From Pugh 1976)

10 1/20/10 15.S

~ 5
--,...
iii 0
Ol
I
10 1/120/10
::t

5
Z

E 0
.~
c 10 ! t
0 20.0 27.3
E
E 5
<!

0
Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr
1974 1975

FIGURE 7. The annual cycle of ammonium-nitrogen concentration in the groundwater and seawater of the
model beach. Frequency of sampling is shown along the top of the diagram. (From Pugh 1976)
 231

-
e
0
til,
Ol 2 1/120/10
::\.
v-
a..
al 0
til
~
a. 1/20/10
2
'"0
~
a..

0
Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr
1974 1975

FIGURE 8. The annual cycle of inorganic phosphate concentration in the groundwater and seawater of the
model beach. Frequency of sampling is shown along the top of the diagram. (From Pugh 1976) ,

increased from less than 1 to more than 24 ~g-at
-1
N0 3 -N 1 ,and then between June and mid August
the pattern was very erratic, suggesting a lack
of balance between'input and utilisation. By mid
August the concentration at both stations had de-
-1
creased to 3-4 ~g-at N0 3-N 1 and subsequently a
pattern analagous to that for seawater was observed. counterparts (see e.g. Spencer 1956).
There was obviously a potential for nitrate pro-
duction in the sand profile throughout the year,
but the figure emphasises the mid summer burst of
activity in response to the ingress of the degrad-
ation products of the Phaeocystis bloom. That the
primary substrate for this activity must have been
dissolved nitrogen-rich organic material was sug-
gested by the very low concentrations of ammonium-
(see Fig. 7) and nitrite nitrogen (the latter
-1
rarely exceeded 0.2 ~g-at 1 and usually exhibited
lower concentrations in the groundwater than in the
inundating seawater).
That nitrification is enhanced in the sandi
groundwater system relative to the seawater system
(as indicated by the amount of nitrate present) is
possible evidence for the requirement by nitrify-
ing organisms of surfaces at which the oxidation
of ammonia to nitrite can take place, a require-
ment well known for terrigenous organisms (see
e.g. Burges 1958), but debated for their marine
The annual cycle of phosphate in the model was
also measured (see Fig. 8) as was that of dissolved
oxygen (Pugh 1976). The microbial biology was also
investigated (Andrews et al. 1976).
Oliff et al. 1970 consider beaches to be in equi-
librium with the nutrient supply from the oceans
and warned against the possible overloading with
organic material, with the subsequent breakdown
of aerobic processes, the onset of anaerobic con-
ditions and the consequent unaesthetic conditions.
Work with the model confirmed their doubts - an
increased input of organic material (and this only
the natural seasonal degradation products of a
232
phytoplankton bloom, let alone a polluting dis-
charge) gave rise to a new equilibrium with higher
concentrations of nutrients. The trend towards
anaerobiosis was also observed, especially where
sediment reworking was not achieved.
This was the first chemical study of a marine sandy
beach held under controlled conditions of temper-
ature, waves and tide and allowed the first view
of the longer term, annual nutrient cycles. Much
remains to be done.
CONCLUSIONS
The concentration of nutrients in natural sandy
beaches is the resultant of a complex inter-
relationship of micro- and macro- biological,
chemical and physical factors. None of these can
be controlled; all are continuously varying. It
is difficult, if not impossible, therefore to un-
ravel the effects of the individual factors. How-
ever sandy beaches are an important ecological
niche and appear to have a vital role as natural
cleansing filters of seawater especially in the
context of marine pollution. Despite the frequent
hostility and mobility of this environment its
study demands more attention than it has hitherto
received. In turn that will demand determination
and facilities, for controlled simulation is pro-
bably the only route to a useful understanding of
them. The way has been paved with in vitro studies.
The models described, perhaps with improved under-
drainage, offer the ability to control at will,
for a given sediment, wave activity, tides (with
or without the spring/neap cycle), temperature
(with or without diurnal and seasonal cycles) and
light (with or without the day/night cycle) and
unravel the cycles of beach chemistries (nutrients,
organic carbon, biological and chemical oxygen
demands, chlorophyll, dissolved oxygen, salinity,
pH, Eh, metals, specific organics) and possibly
biologies (micro-, meio- and macro- faunal and
floral activities) with or without seawater en-
richment (including catastrophic or chronic pol-
lutions).
Acknowledgement. This work was financed by the
Natural Environment Research Council and performed
at the Marine Science Laboratories, University
College of North Wales, Menai Bridge, Angelesey,
UK.
REFERENCES
Anderson JG and Meadows PS (1969) Bacteria on
intertidal sand grains, Hydrobiologia, 33, 33-46.
Anderson JG and Meadows PS (1978) Micro-
environments in marine sediments, Proceedings of
the Royal Society of Edinburgh, 76B, 1-16.
Andrews AR, Floodgate GD and Pugh KB (1976) An
annual cycle, at constant temperature, in a model
sandy beach, II, Microbial biology, Journal of
Experimental Marine Biology and Ecology, 24, 61-72.
Burges A (1958) In: Microorganisms in the soil,
London, Hutchinson University Library, 139-141:
Ewins PA and Spencer CP (1967) The annual cycle
of nutrients in the Menai Straits, Journal of the
Marine Biological Association of the United
Kingdom, 47, 533-542.
Harty B and McLachlan A (1982) Effects of water-
soluble fractions of crude oil and dispersants
on nitrate generation by sandy beach microfauna,
Marine Pollution Bulletin, 13, 287-291.
Huisman L and Wood WE (1974) In: Slow sand
filtration, World Health Organisation, Geneva,
122 p.
Johnston R (1970) The decomposition of crude oil
residues in sand columns, Journal of the Marine
Biological Association of the United Kingdom,
50, 925-932.
McIntyre AD, Munro ALS and Steele JH (1970) Energy
flow in a sand ecosystem. In: Steele JH, ed. -Marine
food chains, pp. 19-31. Edinburgh, Oliver and Boyd.
McLachlan A, Erasmus T, Dye AH, Wooldridge T,
Van der Horst G, Rossouw G, Lasiak TA and
McGwynne L (1981a) Sand beach energetics: An
ecosystem approach towards a high energy inter-
face, Estuarine, Coastal and Shelf Science, 13,
11-25.
McLachlan A, Dye A and Harty B (1981b) Simulation
of the interstitial system of exposed sandy
beaches, Estuarine, Coastal and Shelf Science, 12,
267-278.
Moebus K (1972) the Influence of storage on anti-
microbial activity of seawater, I, Experiments
with seawater stored at 18°C, Marine Biology, 13,
346-351.
Oliff WD, Gardner BD, Turner WD and Sharp JB (1970)
The chemistry of the interstitial water as a
measure of conditions in a sandy beach, Water
Research, 4, 179-188.
 233

Pearse AS, Humm HJ and Wharton GW (1942) Ecology

of sand beaches at Beaufort, N.C., Ecological
Monographs, 12, 135-190.
Pugh KB (1975) A model beach system, Journal of
Experimental Marine Biology and Ecology, 18,
197-213.
Pugh KB (1976) An annual cycle, at constant
temperature, in a model sandy beach,I, Nutrient
Chemistry, Journal of Experimental Marine Biology
and Ecology, 22, 179-192.
Pugh KB, Andrews AR, Gibbs CF, Davis SJ and
Floodgate GD (1974) Some physical, chemical and
microbiological characteristics of two beaches
of Anglesey, Journal of Experimental Marine
Biology and Ecology, 15, 305-344.
Spencer CP (1956) The bacterial oxidation of
ammonia in the seawater, Journal of the Marine
Biological Association of the United Kingdom,
35, 621-630.
Wormald AP and Stirling HP (1979) A preliminary
investigation of nutrient enrichment in experi-
mental sand columns and its effects on tropical
intertidal bacteria and meiofauna, Estuarine and
Coastal Marine Science, 8, 441-453.

THE EFFECT OF MEIOFAUNA AND BACTERIA ON NUTRIENT CYCLES IN SANDY BEACHES
H.F-K.O. HENNIG, A.H. FRICKE and C.T. MARTIN (National Research Institute for Oceanology, P.O. Box 320,
Stellenbosch, South Africa)
1. INTRODUCTION
Little is known about the flux of carbon and
nutrients from surf water through exposed beaches
and back into the sea. Nonetheless it is clear
that such rates of flux depend greatly on the
quantity of water input into the beach (l1unro
et al. 1978). So far few field studies have been
made to determine percolation rates and seawater
volumes passing through sandy beaches (Fenchel,
Riedl, 1970. Steele, l1unro, 1970; Riedl, 1971;
Riedl, l1achan, 1972; I1cLachlan, 1979, 1982).
Recently various aspects of the nutrient cycles in
beaches with high kelp input have received
attention (Koop, Griffiths, 1982; Koop et al.
1982 a,b; Griffiths et al. 1983). However the
vast majority of exposed South African beaches
are "clean" or have no kelp input. Hence an
understanding of nutrient flux in these exposed
"clean" beaches would greatly enhance our
knowledge.
Due to the dynamic and unpredictable nature of
sandy beaches, experimental sand systems have
been used to study interstitial ecosystems in
terms of metabolism (I1cIntyre et al. 1970;
Vosj an, Olanczuk-Neyman, 1977; l1unro et al.
1978), faunal changes (Chamroux et al.1977),
nutrient chemistry (Pugh, 1976; Boucher,
Chamroux, 1976; l1evel, Chamroux, 1981;
I1cLachlan et al. 1981), responses to oil
pollution (Johnson, 1970; Harty, HcLachlan,
1982) and sewage (Wormald, Stirling, 1979). In
most cases it is difficult to assess how far the
235
experimental sand columns did simulate in situ
conditions. A review of the construction factors
and limitations of these experimental columns was
given by HcLachlan et al. (1981).
It was hoped that most of these limitations would
not apply to experimental runs of relative short
duration. Hence an experimental sand column was
developed to represent an "instant" moment in
the time of a beach (90 min) rather than trying
to model long-term interstitial ecosystems.
With this sand column we attempted to determine
whether meiofauna or bacteria contribute more
towards mineralization and which of the two
trophic elements utilized mainly particulate
matter and which dissolved organic material.
Answers were sought by percolating fixed
volumes of filtered and unfiltered seawater
through undisturbed sand columns containing
their indigenous flora and fauna and ones in
which the bacteria had been eliminated with
antibiotics.
Furthermore this study re-examines the literature
reports of the flux of surfwater through beaches
in conjunction with other field data and focusses
on the controversy: can biological regeneration
of nutrients by the interstitial fauna fuel surf
phytoplankton blooms?
236

2. 11ATERIAL AND METHODS Analyzer II. Standard methods of analysis were

Sand cores were collected from Llandudno beach,
Cape, South Africa, with a perspex corer 15 cm
long and 19.6 cm2 cross section. Undisturbed
wet sand cores were corked at both ends and
returned to the laboratory. Salinity was measured
with an optical refractometer and temperatures
were taken. Sediment particle size was deter~
mined using the hydraulic method using a
settling tube (Flemming, Thurn, 1978).
Meiofauna were removed from core samples by two
methods:
1. For purpose of counting and identification
animals were extracted in a modified Oostenbrink
apparatus (Fricke, 1979), lightly stained and
counted (Eagle et al. 1977). Different taxa
were handpicked, washed, dried on glass slides
and their carbon content determined by combustion
in a Heraeus CRN elemental analyser.
used throughout. Some of the field data results
have been published elsewhere (Orren et al. 1981).
The experimental apparatus is shown in Figs. 1 &2.
Undisturbed sand cores were held in a perspex
tube on a fine mesh (45 ~m) nylon gauze and
covered by coarse gauze at the top. The latter
prevented disturbance of the top sediment layers.
The sand core was coupled to another perspex tube
of similar length. This tube was fitted with an
overflow to regulate a seawater flow rate of
between 0.3 and 0.4 mm sec- l through the sediment
core.
ST

2. During tests of the effect of antiobiotics,

the live meiofauna were extracted with seawater
ice as described by Uhlig et al. (1973).

Bacterial densities were estimated by processing

3 g sediment (Koop, Griffiths, 1982) and the
formula in Uazure (1978). Some total cell counts
were determined using the method of Hobbie et al.
(1977) . WL

For the appropriate experiments natural and

experimental sea water samples were filtered
through a 0.2 Vm Nuclepore membrane filter (47 mm
B beaker PT perspex
diameter). Sand column samples (10-15 g) were NG nylon gauze S sediment
taken up with equal amounts of distilled water P pump ST stand
PC perspex core SW seawater
and sonicated for 10 minutes. This was repeated PD petri dish T tubing
twice and the resulting sand effluent was combined. PF perspex feet WL water level
Sand effluent, surfwater and column effluents Fig. 1. Experimental apparatus used in all four
experiments to percolate surf water through a
were stored frozen prior to analysis. Samples sand column.
were analysed for total, inorganic and organic
carbon, silicate, phosphate, total nitrogen, The sand core stood on small perspex feet in a
nitrite and nitrate using a Technicon Auto petri-dish to e~sure eve~ flo~·~ rate ar.d to
 237

prevent surface tension suction by the eluate. minutes. The viability of meiofauna and
Seawater was recirculated for 90 minutes to bacteria was checked, the former extracted with
ensure oxygen supply to the animals and to seawater ice and the latter by plating on seawater
maintain the stated flow rate to allow 500 ml agar (Mazure, 1977).
of seawater to percolate through the column.

2.1. Experiments
To study the mineralization due to meiofauna and
bacteria the following approach was used. Surf
water (500 ml) was allowed to percolate through
ST the sand column, both seawater and sand core were
analysed for nutrients before and after percolation.
Any gain in nutrients by the effluent was con-
sidered as mineralization, any loss would be due
raise
to "nutrient fixing" by organisms in the sand,
RES
that is nutrients removed from the system by
the meiofauna as well as bacteria and used for
"growth" or respiration. Data were collected
from four triplicate experiments as set out in
Table 1. These results were normalized to
comparable core volumes, meiofauna and bacterial
density.

3. RESULTS AND DISCUSSION

NG nylon gauze SI&A seawater &
PC 3.1. Background
perspex core antibiotic
PF perspex feet ST stand The physical and chemical characteristics of
RES reservoir T tubing
Llandudno beach sediment and surf water
S sediment
unfiltered and filtered are summarised in
Fig. 2. Experimental apparatus to sterilize the
sand column by raising and lowering the reservoir. Table 2. These values are very similar to those
reported for ten "clean" beaches along the south
The sediment was "sterilized" by bathing the
and south-west coasts of South Africa (Table 3)
sediment core in antibiotic solution by raising
and hence are representative of the major beach
the reservoir as shown in Fig. 2. The anti-
type of that part of the coast.
biotic cocktail consisted of 100 ~g ml- 1 each
of choramphenicol (Sigma), erythomycin (Sigma)
Numbers and biomass of meiofauna and bacteria
and oxytetracyline (Sigma)dissolved in auto-
from Llandudno beach cores are shown in Table 4.
claved filtered seawater (Robb et al. 1979). A
It is interesting to note that meiofauna make
saturation speed of about 0.3 - 0.4 mm sec- 1
up 64% of the total biomass. This is in sharp
was maintained to prevent animals from dis-
contrast to kelp enriched beaches where bacteria
lodging. Once the antibiotic solution had
make up 75% of meiofauna/bacteria standing stock
covered the sediment, the reservoir was lowered.
(Koop, Griffiths, 1982). The mass of the
This was repeated several times until the centre
individual animals of the different taxa was very
of the core had been submerged for over 60
similar to that found by McLachlan (1977, 1978).
238

The carbon content of the various taxa did not

differ much, as shown in Table 4.

TABLE 1. FLOW DIAGRAM OF FOUR EXPERIMENTAL SETUPS FOR NUTRIENT CYCLES IN AN EXPERIMENTAL CORE

Unfiltered Seawater (500 m1)

---------~~~--~~~----~
1. Core 2. Core and antibiotics
~ ~
Meiofauna Meiofauna
and Bacteria
.+
Part~cu1ate Nutrients "fixed"
and dissolved nutrients by meiofauna alone
"fixed"

Filtered Seawater (500 ml)

---------~~.--~~~~~
3. Core 4. Core and antibiotics
+
Meiofauna
+
Heiofauna
and Bacteria

Mainly dissolved
... 1
Dissolved nutrients
nutrients "fixed" "fixed" by meiofauna alone

TABLE 2. PHYSICAL AND CHEl1IeAL CHARACTERISTICS OF LLANDUDNO BEACH SEDllmNT AND SURF WATER

Sediment grain size 264 ~m (! 98) n 10

Beach slope About 1 18
Surf water temperature 13,7°C
Salinity 34,8 0 /00
SURF WATER

Unfiltered (s .d.) Filtered (s. d.)

-1
Total carbon (mg 1 ) 24,45 (! 2,55) 21,12 (! 2,84)
-1
Inorganic carbon (mg 1 ) 20,55 (! 2,55) 17,39 (! 2,99)
Organic carbon (mg 1- 1 ) 3,90 (! 1,77) 3,73 (! 0,62)
-1
Phosphate (~mo1e 1 ) 1,42 (! 0,30) 1,18 (! 0,14)
Silicate (~mo1e 1-1) 7,00 (1,24) 6,74 (! 1,41)
-1
Total nitrogen (~mole 1 ) 7,64 (! 0,91) 7,16 (! 1,00)
Nitrite (~mo1e -1) 0,53 (! 0,26) 0,47 (! 0,47)
Nitrate (~mo1e -1) 7,11 (! 0,78) 6,69 (! 0,93)

n = 17
s.d. = standard deviation
Sampling period = 1982.08.24 1982.10.14.
 239

TABLE 3. l1EAN VALUES OF l1EASURED CHEllICAi CHARACTERISTICS IN TEN "CLEAN" BEAHCES

SOUTH COAST WEST COAST

X s.d. X s.d.
Surface Water

Salinity (0/00 ) 35,1 0,2 34,7 0,2

_1
Total phosphorus ()lmol 1 ) 1,2 0,6 1,7 0,4
_1
Silicate ()lmol 1 ) 5,2 2,5 13,6 6,0
_1
Nitrite ()lmol 1 ) 0,2 0,2 0,3 0,2
_1
Nitrate ()lmol 1 ) 5,8 3,0 11,7 4,4
_1
Dissolved oxygen (mg 1 ) 6,7 1,0 7,7 1,4
_1
OA (mg g ) 0,004 0,001 0,006 0,003
pH 8,2 0,1 7,9 0,2

Interstitial Water
_1
Total phosphorus ()lmol 1 ) 3,2 1,5 7,6 5,7
-1
Silicate ()lmol 1 ) 18,3 10,5 23,3 17,2
-1
Nitrite ()lmol 1 ) 0,2 0,2 0,2 0,1
-1
Nitrate ()lmol 1 ) 27,7 19,8 10,4 8,7
-1
Dissolved oxygen (mg 1 ) 3,9 2,4 4,2 2,3
-1
OA (mg g ) 0,004 0,003 0,006 0,002
pH 8,1 0,1 7,8 0,2

(From Orren et al. (1981))

TABLE 4. NUl1BERS AND BIOHASS OF 11EIOFAUNA AND BACTERIA AT LLANDUDNO BEACH, CAPE, SOUTH AFRICA,
HEASURED IN A 15 cm CORE TAKEN FROH THE HID-TIDE POSITION

Number per Biomass % Total

100 ml sand (mg dry wt/100 ml sand) %C Biomass

Meiofauna 666 0,154 ht.,3

Nematodes 188 0,073 39,7
Harpacticoids 41 0,021 27,5
Oligochaetes 1 0,003 39,9
Flatworms 86 0,057 33,3
Protozoa 350

Bacteria 6,47 x 10 6 0,055 35,7

It was found that live meiofauna extracted with
sea water ice showed no ill effects after 12
hours of antibiotic solution recycling, although
the copepods appeared more sluggish.
experimental runs (90 min) with antibiotics
the meiofauna showed no behavioural difference
to that of the control.
Sand grains and surfwater were found to be
bacterial free after 60 minutes antiobotic treat-
ment (complete submerging of sand grains). After
After plating neither treated surfwater nor sand grains
showed any bacterial growth for 3 days when the
experiment was terminated.
240
The method of enumeration of viable bacteria
adopted in the present study leads to an under-
estimation and gives no indication of bacterial
types. It is assumed, however, that the results
are at least an index of the total bacterial
population. Elimination of bacteria by anti-
biotics in experiments 2 & 4 was very likely
complete. Unfortunately negative results obtained
by our seawater agar plating could mean that only
substrate yich bacteria had been eliminated
(M. Bolter, pers. comm., see also this volume).
Davis et al. (in press) found for instance that
only 17% of total interstitial bacteria were
platable.
The results of the 4 experiments (Table 1) are
summarized in Table 5. They are derived as
follows: (Nutrient of undisturbed sand core +
nutrient of surfwater (500 ml) + nutrient of
interstitial water within the core (23.2 ml»
(Nutrient of sand cere after percolation + nutrient
of percolated water (effluent) + nutrient remaining
interstitial water after the experiment) equals
to nutrient "fixing" (positive value) or
mineralization(negative value).
It was found that the nutrient concentration of
sand cores before and after experimental runs did
not change beyond experimental errors. Hence
the data in Table 5 represents the amount of
nutrient "fixed" by the sand fauna plus particulate
organic matter (POM) trapped in the column pores.
The latter would not necessarily be ingested
immediately by the meiofauna and bacteria, but
replenishes matter taken up during the experiments.
These results (Table 5) suggest that more
nutrients are "fixed" by meiofaunaand bacteria
than by meiofauna alone.
Particulate matter is an important source of
nutrients ingested by meiofauna, for instance
of the total carbon "fixed" (3.41 mg), 40.5%
originated from the particulate fraction (3.41 mg ""
2.03 mg = 1. 38 mg) (Table 5 & 6). Similarly
the portion of particulate matter utilized for
"fixing" total phosphorus is 45.7% and nitrate
58.3% (Table 6). In our system, on the other
hand, organic carbon was mineralized if meio-
fauna alone was present represented by the
negative values: - 2.66 mg & - 2.30 mg in Table 5.
Without the bacteria, organic carbon could not
be retained.
3.2. Carbon
Although POM makes up only about 14% of the total
-1
carbon found at Llandudno (3.33 mg 1 from
24.45 mg 1- 1 , Table 2), meiofauna were respon-
sible for most of the inorganic carbon retention,
whereas bacteria were responsible for virtually
all the organic carbon retention (Table 5).
lieiofauna regenerated minor amounts of organic
carbon, probably from predation on other meio-
fauna or bacteria ingesting.
-
In this context it should be noted that for
instance the nematode Rhabditis marina's diet
consists of bacteria. About ~O%of its body size
is gut and it consumes 100 times its own body-
weight per year and has a lifespan of about one
month. While Oncholaimus brachycercus preys
on bacteria eating R. marina and consumes 10% of
its body mass per day, living about 6 months to
one year. (R. Warwick, pers. comm.). Usually
a 10% efficiency per food chain link is assumed.
Hence large amounts of nutrients are necessary
when considering a complex food chain like
particulate and dissolved matter bacteria
meiofauna predator meiofauna, so that
population of O. brachycercus an;! ot],er
predators can be maintained. Therefore
nutrients are seldom accumulated in beaches,
after an equilibrium is reached between food
supply and animal density, as indicated by the
averages and small standard deviations (s.d.)
presented in Table 3 (see also Orren et al.198l) .
The bacteria attacked and fixed only about 10%
TABLE 5. AMOUNTS OF NUTRIENTS RETAINED AND l1INERALIZED m THE SAND COLUMNS DURING PERCOLATION OF 500 ml OF SURF WATER

Total Inorganic Organic Total Total

Exp. Carbon Carbon Carbon l'hosphorus Silicate Nitrogen Nitrite Nitrate
Surf Water set-up Organisms mg mg mg * )lmol )lmol )lmol )lmol )lmol

1 M& B 7.99 6.64 1.35 0.68 5.57 2.88 0.71 2.57

Unfiltered
2 M 3.41 6.08 -2.66 0.46 4.01 2.30 0.24 2.06

3 M& B 4.66 4.10 0.56 0.36 6.30 2.52 0.10 2.4.

Filtered
4 M 2.03 4.33 -2.30 0.25 4.88 2.27 0.10 2.17
~

M Meiofauna B Bacteria -, Calculated from Total carbon - Inorganic carbon values

TABLE 6. RELATIVE CONTRIBUTIONS OF MEIOFAUNA AND BACTERIA TO NUTRIENT RETENTION

Total Inorganic Organic Total Total

Carbon Carbon Ca.rbon Phosphorus Silica.te Nitrogen Nitrite Nitrate
Organisms % % % % % % % %
M 42.7 91.6 67.6 72.0 79.9 77 .4 80.2
Unfiltered
B 57.3 8.4 32.4 28.0 20.1 22.6 19.8

M 43.6 105.6 69.4 77 .5 90.1 10.00 90.0

Filtered
B 56.4 30.6 22.5 9.9 10.0

Heiofauna &
Bacteria P * 41.7 38.3 58.5 47.1 0.0 12.5 67.7 6.2

Heiofauna P * 40.5 28.8 13.5 45.7 0.0 1.3 58.3 0.0

U Meiofauna B Bacteria P * Particulate material is utilized for x% "fixing" of nutrients

'"
N
-I'-
242

of the retained particulate inorganic carbon

(38.3- 28.8 = ~.5%, Table 6) while most of thp.
organic carbon "fixed" by bacteria came from the
dissolved fraction.
"
N
H
o
o o
3.3. Phosphorus
The amount of total phosphorus "fixed" (0.68 mg,
Table 5) originated to about half (47.1%; 0.68 mg N N
co o
- 0.36 mg = 0.32 mg, Table 5) from POl1. Further- (V)
'"
N

more nearly all phosphorus was retained, from

0.71 mg phosphorus (seawater percolating through
column, Table 2) 0.68 mg were "fixed", which is
o I
II
about 95.8% (Table 7). If phosphorus is linked '"
(V)
II
II
II
to POM, bacteria rely on the excretion by II
II
II
meiofauna and amounts of dissolved phosphorus in II
II
II
the surf water. The pathway of new phosphorus via II
II
co II
POM meiofauna bacteria appears to be long, -n II co
o
7 7
II

hence it is not surprising that marine bacteria ~H o II

II
II '"a-
o II
store large amounts of phosphorus: 26.9% dry u II
II
II
weight of bacteria is phosphorus. This was found II
II
II
after bursting and leaching bacteria with fresh II
II
II
water. II
II
II
II
II
II
3.4. Silicate II
II
co II
N
All the silicate "fixed" appeared to come from II
o
H
II
II
the dissolved matter: surprisingly about 25% II
II
II
(28.0% and 22.5%, Table 6) of this silicqte was II
II
II
fixed by bacteria. It is possible that this II
II
(V) a- II
really is incorporated by diatoms and not by N a- II
II
N II
bacteria as indicated in Table 6. Nearly 80% H II
II
II
(79.7%, Table 7) of silicate input was retained II
II
II
within the column. II
II
II
II
II
II
3.5. Nitrogen II
II
II
Meiofauna were responsible for 80% (79.9%, II
II
II
Table 6) of all nitrogen retained in the column II
II
II
but only a small proportion (1.3%) of this II
II
II
retained nitrogen came from P0l1 (2.30 mg - II
II
II
2.27 mg = 0.03 mg, Table 5). This result is II
II
II
perhaps not so surprising since meiofauna make II
II
w II
up 64% of the biomass (Table 4). By contrast
"
II
II
§< II
Koop et al. (1982) found that 94% of nitrogen H II

is taken up by the bacteria. Our system is very


much less nutrient-rich and as Koop et a1 (1982)
also mentioned under lean conditions more carbon
is used to extract the required amount of
nitrogen. That would explain why in our system
bacteria "fixed" such relatively large amounts
of carbon compared to their biomass.
4. CONCLUSIONS
Meiofauna on average retained 66% (42.7% total
carbon + 67.6% total phosphorus + 72.0% silicate
+ 79.9% total nitrogen, X 65.6%, Table 6) of
all nutrients in our column. Particulate matter
contributed 22% (40.5% total carbon + 45.7%
total phosphorus + 0% silicate + 1.3% nitrogen,
x 21.9%, Table 6) and 78% of the nutrients
were absorbed as dissolved organic matter.
Carbon and phosphorus were about equally "fixed"
from particulate and dissolved fractions by
meiofauna (40.5% & 45.7%, Table 6) and bacteria
(59.5% & 54.3%, difference of percentage fixed
by meiofauna/bacteria - meiofauna, Table 6), while
silicate only came from the dissolved fraction.
If it is now assumed that our experimental sand
column represents an "instant" moment in a
beach, our data suggests that 77% (65.3% total
carbon + 95.8% total phosphorus + 79.7% silicate
+ 66.0% total nitrogen, X 76.7%, Table 7)
of all nutrients are "fixed" by meiofauna and
bacteria in a beach system like L1andudno beach.
Similar results have also been obtained by
Boucher, Chamroux (1976) in a moderately enriched
-2 -1
environment (61 g nitrogen m g ) and by Koop
et a1. (1982) in a highly enriched environment
-1 -1
(4344 g nitrogen m g ). Hence in these systems
studied so far, only about 23% of nutrients can
reach the sampling depth of 15 cm. In terms of
the presented field data (Table 2 , sampling
depth one metre), even more nutrients are "fixed"
while percolating to the interstitial water at
one metre depth. Nutrient concentrations at that
depth will therefore depend greatly on the
quantity of water percolating vertically through
a beach and its initial nutrient concentration.
243
The rate of entry of water into a beach has been
_1
measured as 5 cm d ; that is each m2 of sand
introduced 0.049 m3 of water into the beach per
day (Steele, Munro, 1970). On the other hand,
Fenchel, Riedl (1970); Riedl (1971); Riedl,
l1achan (1972) and I1cLach1an (1979), all using
Riedl's method of calculating water flux,
reported huge volumes of surf water being pushed
3 _1
through the interstitial beach (10 m d
200 m3d- 1 per running metre of beach). Using
the data obtained from our column together with
field data (Table 2) and the literature flow
rates, some thought provoking points emerge.
In the case of either low or high surf water
input rates, equal volumes of water must drain
back to the sea (Fig. 3). Riedl (1971)
described a water particle drainage path as
"following gravity first, leading steeply down-
Fig. 3. Drainage path of surf water: for
equal volume of water entering the beach an
equal volume drains back to the sea. (See text
for description).
wards and then bending deep into the outflow
pattern toward the sea" as illustrated in
Figs. 3 to 5. In terms of particulate and
dissolved nutrients a simple model can be
postulated as follows:
Referring to Fig. 4, if one litre of surfwater
containing 10 units of nutrients enters the
beach and percolates to the interstitial pool,
then one litre of water returns to the sea
244

f- I

D
10 UNITS

OJ
AT 15cm
77%,"f;xtd"

' \ UNIT e- I

[]~IJ
Fig. 4. Traditional model of nutrient flux Fig. 5. Traditional model of nutrient flux
observed in the field. (See text for description). incorporating field observations and experimental
data (See text for description).
despite the time taken. However the nutrient
concentration measured in the interstitial pool Thus to balance input and drainage, the following
is between 2 to 5 times higher than in surf model incorporating our results is proposed:
water (Table 8). Similar results have also been Fig. 6 10 units nutrient per litre surfwater
obtained by Pugh (1976). Hence more nutrients enter the beach sand with the waves. Host of the

TABLE 8. CONCENTRATION OF NUTRIENTS IN

INTERSTITIAL WATER COl1PARED WITH
SURF WATER OF THE SAHE BEACHES AS
TABLE 2 (Percent)

South Coast West Coast

Total phosphorus 267% 447%
Silicate 351% 171%
Nitrite 100% 66%
Nitrate 478% 89%

_1
(50 units 1 ) would be drained to the sea than
_1
entering the interstitial pool (10 units 1
untenable situation that would soon drain the
beach of all nutrients. In view of our column
results and literature values the situation is
even more complex. In Fig. 5, 10 units
per litre of surf water enter the beach; of
which 77% is "fixed" by meiofauna and bacteria.
Hence only nutrients of the order of 2 unit 1- 1
will reach the interstitial pool, but the
interstitial pool should lose 50 units of
nutrients for every litre of water entering
the system (Fig. 5). The situations
schematized in Fig. 4 and 5 violate conservation
of mass for the nutrients so a new model is
needed.
), an
Fig. 6. New model of nutrient flux balancing
input, drainage and experimental result (See
text for description).
nutrients are filtered (particulate fraction)
and adsorbed (dissolved fraction) within the
first few centimetres (Wormald, Stirling, 1979),
while most of the cleaned water (800 ml in our
case) drains back to the sea just below and just
above the sand-water interface. Again 77% of
the nutrients are "fixed", but at the same time
more water (160 ml) is lost either in draining
through microscopic channels as mentioned by
Riedl, Hachan (1972) or along the sand striations
observed by Hennig, Fricke (1980), Fricke et al.

(1981) and Flemming, Fricke (1983). Hence, due
to the filtration effect and sideways drainage
of the water, nutrients are concentrated in a
small volume of water (40 ml). This (2 units in
40 ml) in turn equals the drainage rate back to
the sea (Fig. 6).
The rate of 18 m3 of water percolating into the
beach per year proposed by Steele, l1unro (1970)
appears to fit our experimental and field data
better than the huge drainage amounts proposed
by Riedl (1971) and McLachlan (1979).
To establish the potential significance of our
model and data, let us assume particulate organic
matter in the inshore water to average about
-1
4.8 mg 1 (Griffiths, 1980). If only half the
particulate matter was fine enough to enter the
beach pore system, then 2.4 g organic material
would enter the beach with every cubic metre of
water filtered. At a nitrogen content of 5% and
with 23% mineralization, this would result in
0.27 g nitrogen being regenerated per cubic metre
of water filtered. Average concentrations of
total nitrogen in the surf zone is about 105 ]Jg
-1
1 If the average beach has a surf zone 200 m
wide with an average depth of 1.5 m, this
represents a nutrient pool of 31.5 g nitrogen per
metre of shore. Using filtration rates proposed
by Riedl (1971) and McLachlan (1979) of 10 m3d- 1
this entire nutrient pool would then be regenera-
ted by the beach microfauna every 117 days. Thus
if it takes four months for the nutrient pool to
be regenerated the interstitial water at one metre
depth can be regarded as a nutrient sink.
Similar conclusions have been drawn by Griffiths
et al. (1983) who found that kelp enriched beaches
rely upon imported debris to support internal
food chains with limited quantities of nutrients
returning to the sea by drainage. These
conditions also appear to govern "clean" exposed
beach as seen in this study.
245
One conclusion to be drawn from this is that the
biological regeneration of nutrients by the inter-
stitial fauna of these types of beaches cannot
contribute to the development and maintenance
of surf phytoplankton blooms.
Recently the possible contribution of beach
nutrients to the maintenance of surf phyto-
plankton blooms has received some attention
(McLachlan, 1979, 1982; McLachlan et al. 1981;
Lewin, Schaefer, 1983; Quinlan et al. 1983).
From our data and model, three conditions of surf
wa,ter flux and remineralization can be identified
and are summarised as follows:
1. Low volume water flux + elevation of inter-
stitial nutrient concentration above surf
concentrations (x 2-5)'+, low volume drainage
+ no contribution to surf bloom.
2. High volume water flux + interstitial
nutrient concentration equal to that in
surf + high volume drainage + no contribution
to surf bloom.
3. Low volume water flux + interstitial nutrient
concentration 1 000 times higher than surf +
low volume drainage + conjectured contribu-
tion to surf bloom.
The last situation is rather a special case
described by Quinlan (pers. comm.) in that the
alga Pilayella littoralis carpets the seabed
and hence can absorb any low volume, high
nutrient drainage from the beach.
ACKNOWLEDGEMENTS
Financial support from the National Programme
for Environmental Sciences is gratefully
acknowledged.
246
REFERENCES
Boucher G and Chamroux S (1976) Bacteria and
meiofauna in an experimental sand ecosystem. 1.
Material and preliminary results, J. expo mar.
BioI. Ecol. 24, 237-249.
Chamroux S et al. (1977) Etude experimentale
dun ecosysteme sableux. II. Evolution des
populations de bacteries elde meiofaune.
Helgolander wiss. Meeresunters 30, 163-177.
Eagle GA et al. (1977) Camps Bay Beach:
A pollution survey, S. Afr. J. Sci. 73, 342-345.
Fenchel TI1 and RIEDL RJ (1970) The sulfide
system: a new biotic community underneath the
oxidized layer of marine sand bottoms, Mar. Bio.
7, 255-268.
Flemming BW and Thum AB (1978) The settling
tube a hydraulic method for the grain size
analysis of sands, Kieler Meeresforsch Sonderheft
4, 82-95.
Flemming BW and Fricke AH (1983) Organism sedi-
ment relationship in sandy beaches: inter-
stitial microhabitats as a function of internal
geometry, primary sedimentary structures and grain
size. This volume.
Fricke AH et al. (1981) Relationship between
oil pollution and psammolittoral meiofauna density
of two South African beaches, Mar. envir. Res.
5, 59-77.
Griffiths RJ (1980) Ecophysiology of the
black mussel ChoromytiZus meridionaZis (Krauss).
Cape Town, University of Cape Town, Department
of Zoology, 1980 (Ph.D. thesis) 269 pp.
Griffiths, et al. (1983) Kelp wrack and the
flow of energy through a sand beach ecosystem.
This volume.
Harty B and McLachlan A (1982) Effects of
water-soluble fractions of crude oil and dis-
persants on nitrate generation by sandy beach
microfauna, Mar. Pollut. Bull. 13, 287-291.
Hennig HF-KO and Fricke AH (1980) Recovery
potential and chronic oil pollution on the
South African coast, Mar. Pollut. Bull. 11, 30l~·
302.
Hobbie JE et al.(1977) Use of Nuclepore
filters for counting bacteria by fluorescene
microscopy, App 1. environ. Microbiol. 33, 1225-
1228.
Johnston R (1970) The decomposition of crude
oil residues in sand columns, J. mar. bioI. Ass.
U.K. 50, 927-937.
Koop K et al. (1982a) Biodegradation and
carbon flux based on kelp (EckZonia m=ima)
debris in a sandy beach microcosm, Mar .. Ecol.
Prog. Ser. 7, 315-326.
Koop K et al. (1982bj. Microbial regeneration
of nutrients from the decomposition of macrophyte
debris on the shore, Mar. Ecol. Prog. Ser. 9,
91-96.
Lewin J and Schaefer CT (1983) The role of
phytoplankton in surf ecosystems. This volume.
Mazure HG (1977) Bacteria associated with
kelp fronds in the Benguela upwelling system and
their ecological importance, Cape Town, University
of Cape Town, Department of Zoology. (Ph.D. thesis)
157 pp.
}1clntyre AD et al. (1970) Energy flow in
a sand ecosystem. In Steele JH, ed. Marine Food
Chains, pp. 13-31. Edinburgh, Oliver & Boyd.
McLachlan A (1977) Composition, distribution
abundance and biomass of the macrofauna and
meiofauna of four sandy beaches, Zool. Afr. 12,
279-306.
HcLachlan A (1978) An ecophysiological study
of the meiofauna of the Swartkops Estuary 2.
The l1eiofauna composition, distribution,
seasonal fluctuation and biomass, Zool. Afr.
13, 19-32.
McLachlan A (1979) Volumes of sea water
filtered by East Cape sandy beaches, S. Afr.
J. Sci. 75, 75-79.
HcLachlan A et al. (1981) Simulation of
the interstitial system of exposed sandy beaches,
Estuar. Coastal Shelf Sci. 12, 267-278.
McLachlan A (1982) A model for the estimation
of water filtration and nutrient regeneration by
exposed sandy beaches, Mar. Environ. Res. 6,
37-47.
Hevel G and Chamroux S (1981) An experimental
study of the regulation of nitrification in
marine sediments, Oceanol. Acta, 4, 457-463.
Munro AL et al. (1978) Energy flow in the
flora and meiofauna of sandy beaches, Proc. roy.
Soc. Edinb. 76·B, 297-315.
Orren MJ et al. (1981) The chemistry and
meiofauna of some unpolluted sandy beaches in
South Africa, Water S.A. 7, 203-210.
Pugh KB (1976) An annual cycle at constant
temperature, in a model sandy beach. 1. Nutrient
Chemistry, J. expo mar. BioI. Ecol. 22, 179-192.
Riedl RJ (1971) How much seawater passes
through sandy beaches? Int. Revue. ges. Hydrobiol.
56, 923-946.
Riedl RJ and Machan R (1972) Hydrodynamic
pattern in lotic intertidal sands and their
bioclimatological implications, Mar. BioI. 13,
179-209.
Robb F et al.(1979) Cellulolytic bacteria as
primary colonizers of Potamogeton pectinatus L.
(Sagopond weed) from a brackish south-temperate
coastal lake, Hicrob. Ecol. 5, 167-177.
Quinlan AV et al. (1983) Fouling of the sandy
beaches on Nahant Bay (Massachusetts, U.S.A.) by
an abnormal free-living form of the macroalga
PiLayeLLa LittoraZis (Phaeophyta). 1. Habitat
Characteristics. This volume.
Steele JH and Hunro ALS (1970) Environmental
factors controlling the epipsammic flora on
beach and sublittoral sands, J. mar. bioI. Ass.
U.K. 50, 907-918.
Uhlig G et al. (1973) The quantitative
separation of meiofauna, Helgolander wiss.
l1eeresunters, 25, 173-195.
Vosjan JH and OLANCZUK-NEYlL\N KM (1977)
Vertical distribution of mineralisation processes
in a tidal sediment, Neth. J. Sea Res. 11, 14-23.
Wormald AP and Stirling HP (1979) A
preliminary investigation of nutrient enrichment
in experimental sand columns and its effect on
 247

tropical intertidal bacteria and meiofauna,

Estuar. Cstl. Shelf. Sci. 8, 441-453.

CARBON FLOW AND NUTRIENT REGENERATION FROM THE DECOMPOSITION OF MACROPHYTE DEBRIS IN A SANDY BEACH
mCROCOSM
K. KOOP* and M.I. LUCAS (Zoology Department, University of Cape Town, Rondebosch 7700, South Africa,
*Present address: School of Biological Sciences (A12), University of Sydney, NSW 2006, Australia)
INTRODUCTI ON
Recent studies on the extensive kelp beds of the
west coast of South Africa have provided a good
deal of information on the standing stocks
(Vel imirov et al., 1977; Field et al., 1980),
primary production (Field et al., 1977; Dieck-
mann, 1978, 1980; Mann et a1 ., 1979; Brown,
1980, 1981; Borchers, Field, 1981; Jarman,
Carter, 1981; Carter, 1982; for review see
Newell et al., 1982) and the ecological energet-
ics of the consumer organisms in this system
(Greenwood, 1977, 1980; Fitzgerald, 1979;
Griffiths, King, 1979a b; Shafir, Field, 1980a,
b; Pollack, 1981). Much of the prima ry
production of the two dominant phaeophytes
Lamin~a pattida and EcRfonia maxima enters the
water column after fragmentation and is utilised
by filter feeders which dominate the system
(Field et al., 1977, 1980). Newell et al. (1982.)
have, however, pointed out that about 6% of the
annual seaweed primary production is broken free
in storms and is exported either to nearby
beaches or fragments in ~itu in the swash zone
much as has been reported from other parts of the
world (e.g. Hayes, 1974).
Until recently little was known of the fate of
this cast-up material apart from an early study
by Backlund (1945). Estimates by Koop and Field
(1980) and Jarman and Carter (1981) suggest that
between 1200 and 1800 x 10 3 kg wet weight of
EcRfonia maxima are cast up annually along a 1 km
strandline from a kelp bed of 700 ha at Kommetjie
on the west coast of the Cape Peninsula (see also
249
Koop et al., 1982a, b). Muir (1977), Koop and
Field (1980, 1981), Stenton-Dozey and Griffiths
(1980) and Griffiths and Stenton-Dozey (1981)
have shown that much of this material on the
surface may be consumed by the amphipod Tafo~­
chv.,;U.a capeVil>~, the isopod Ugia cUtata.ta and
larvae of the kelp fly Fuc~ capeVil>~. Little
is known, however, of the fate of material voided
as faeces or buried beneath sand, nor of the piles
of kelp debris that may build up on the shore.
This paper summarises results of studies made on
the relative rates of utilisation of cast kelp by
the dominant consumers and by microheterotrophic
decomposer organisms. This allows us to make
some quantitative estimates of the relative
significance of consumer organisms and microhetero-
trophic pathways in the energetics of the sandy
beach community. It also allows us to follow the
decomposition products of kelp debris through a
sandy beach adjacent to a kelp bed of known
primary production in nearshore waters.
MATERIALS AND METHODS
Study Site
A sample of 26.35 kg wet weight of freshly-cut
kelp (EcRfonia maxima Osbeck) was placed at high
water of spring tides on a shore with sand
patches at Kommetjie, Cape Peninsula, South
Africa (34°08'S : 18°18'E) (Fig. 1). Kelp was
put in a cleft with a flat rock floor which
drained into a small depression of approximately
10 £ volume (Pool 1). An overflow of excess
fluids from this pool drained through a sand
250

patch of 1 m length and 50 cm width with a depth
of approximately 12 cm on an impervious rock
floor. Finally, material from the sand could
drain into a larger rock pool of 30 £ volume
(Pool 2). Fluids from Pool 1 were used to assay
the chemical composition and microbiology of the
initial kelp leachates. The sand strip allowed
us to study the role of the interstitial micro-
fauna in the degradation of dissolved and
particulate matter entering the beach from the
kelp while analyses of organic matter entering the
larger Pool 2 represented material returned to the
sea following processing activities in the sand
(Fig. 2).
i AUantic
I Ocean
The contents of the pools were first drained and
replaced with clean seawater, then the sand was
replaced with clean sand from the adjacent dunes
so that the amount of leachates from the kelp and
the microbial communities associated with them
could be followed quantitatively. Microbial
densities, numbers of consumer invertebrates and
the chemical composition of the leachates and the
kelp itself were measured at daily intervals for
8 d until Pool 2 was inundated by the next spring
high tide. All measurements were corrected for
evaporative losses from the pools using
evaporation dishes. Desiccation of the kelp pile
was prevented through the rapid formation of a
superficial dry layer and at the end of the study
all but this layer had disappeared through a com-
bination of grazing by inverbrates and bacterial
decomposition (Koop et al., 1982a, b). The
chemical composition of the dried out kelp
appears to be remarkably stable - the C:N ratio
remained unchanged over the study period.

FakJe Bay
Invertebrate Consumer Species
c..,._ It was recognised that removal of invertebrates
from the kelp pile to assess consumer densities
34°5- would affect the rate of consumption of kelp.
Densities of invertebrate consumers associated
Indian Ocean with kelp piles were therefore estimated in a
subsequent experiment. Eight mesh bags, each
FIGURE The study areas at Oudekraal and containing 2 kg of fresh EcRfonia max~a were
Kommetjie on the west coast of the Cape Peninsula, placed on the study site immediately after the
South Africa.
initial 8 d experiment. One bag was removed
HWS each day and the numbers and biomass of the
I associated invertebrates recorded. Biomass was
expressed as dry weight, after oven drying at 60°C
to constant weight.

Microbial Communities
Microbial communities associated both with the
kelp debris Pool 1 sands~ Pool 2
26.35 kg 10 L interstitial water 8 l 30 L initial degradation of the freshly cut kelp and
with the subsequent utilisation of dissolved and
FIGURE 2 Detail of the study site at Kommetjie,
Cape Peninsula, South Africa, showing the differ- particulate matter in the leachates were assayed
ent components of the microcosm. Average depth quantitatively at daily intervals. To estimate
of the sand strip was 12 cm.

the numbers and biomass of bacteria on the sur-
face of the kelp, discs of 1 cm diameter were
punched in fronds and fixed in 1.5% glutaralde-
hyde. These were prepared for scanning electron
microscopy by a combination of methods described
by Todd and Kerr (1972), Paerl (1975), and Bowden
(1977). The discs were desalinated, dehydrated
through an ethanol series, then fixed in osmium
tetroxide before critical point drying and sputter
coating with gold palladium. Samples were then
viewed on a Cambridge S180 scanning electron
microscope.
The numbers of bacteria on the surface of the
fronds were estimated from counts on 20 random
fields. The biomass was then calculated from
measurements of the cell dimensions and values of
the specific gravity of bacterial cells (Luria,
1906) using the following expression (Linley
et al., 1981; see also Rodina, 1972):
N • V • SG
Wet biomass
10 6
where N number of cells mm- 2 (x 10 6 );
V mean volume of the cells (~m3);
SG specific gravity of the cells.
The dry biomass and carbon equivalent of the wet
biomass was then calculated using the coefficients
of 0.2 and 0.1 respectively (Luria, 1960; Troitsk~
Sorokin, 1967; Sorokin, Kadota, 1972).
The numbers of bacteria in the water column were
estimated by acridine orange direct counts (AODC)
following filtration onto 0.2 ~m Nuclepore poly-
carbonate filters (Hobbie et al., 1977; Linley
et al., 1981). Cell dimensions were estimated
from scanning electron micrographs of discs of
the filters prepared as described by Linley et
al. (1981) and biomass was estimated as before.
Numbers of bacteria in the sand column were
assayed following three repeated sonications of
the sand with DAWE Sonicleaner Type 6442A. The
detached bacteria were then filtered from the
eluted sample. The sum of these two values was
251
then used to estimate the numbers of bacteria in
the microcosm as a whole.
CHN Analysis
The total carbon and nitrogen content of three
subsamples of 100 to 200 ~g of freeze-dried
material from both the kelp debris and the liquid
material from Pool 1, the interstitial water, and
from Pool 2 were measured by combustion in a
Carlo Erba elemental analyser (Model 1106) fitted
with an integrator (see also Lucas et al., 1981).
The nitrogen and carbon contents of five blank
samples of tin cups alone and of triplicate
experimental samples were then calculated from
the mean peak areas of at least ten weighed
samples of approximately 1.5 to 2.0 mg cyclohexa-
none (Carlo Erba) containing 20.14% Nand
51.79% C. Inorganic carbon was measured with a
Beckman TOC Analyser (Model 915A). The values
were generally low, making up no more than 30%
of the total carbon in the pools during the early
phases of the experiment and decreasing to as
little as 3% towards day 8. Inorganic carbon
values were somewhat higher in the interstitial
water, accounting for about 70% on the first 2 d
but also dropping to insignificant levels (0.06%
of total carbon) on day 8.
Nutrient Analyses
Samples of 40 me were taken daily from initial
leachates in Pool 1, from the interstitial water
of the sand strip and from the final drainage
Pool 2. The water was filtered through 25 mm
Whatman GF/C filters and the filtrate frozen
prior to analysis. Samples were analysed for
ammonia, nitrite, nitrate and phosphate on a
Technicon Autoanalyzer II. The methods used are
those of Armstrong et al. (1964), Strickland
and Parsons (1972) and summarised in Koop et al.
(1982b) .
252

RESULTS AND DISCUSSION Microbial Colonisation of Kelp Debris

Numbers and Biomass of Invertebrate Consumers on the Strandline
The freshly cut kelp was almost immediately in- In all cases initial colonisation and subsequent
vaded by large numbers of the amphipod TatOh~h~­ development of microbial communities followed a
t1a ~apen6~, which reached a peak biomass of similar pattern. The freshly cut kelp (d zero)
approximately 16.7 g dry weight kg- 1 original wet was already sparsely conolised by patches of
weight of kelp on day 2. At the same time the coccoid bacteria (Fig. 3a) much as has been
kelp was colonised by smaller numbers of the described for the surface of living kelp and other
isopod Ligia ditatata, which reached a peak bio- macrophytes (Sieburth, Thomas, 1973; Laycock,
mass of approximately 10.5 g dry weight kg- 1 1974; Cundell et al., 1977; Mazure, Field, 1980;
original wet weight of kelp. Larvae of the kelp E.A.S. Linley, pers. comm.). In the first phase
fly Fu~eLtia ~apen6~ appeared on day 1. In con- of decomposition, which starts after approximate-
trast to the crustaceans, however, their biomass ly 24 h, distinctive colonisation patterns (Fig.
remained almost constant at approximately 230 mg 3b) are formed, with bacteria aligned along super-
dry weight kg- 1 wet weight of kelp during much of ficial cell borders. This may result either from
the period of decomposition of the kelp. the migration of bacteria initially present, or
from their replacement by others. Subsequently,
Results are summarised in Table 1. They general- on d 3, epidermal lesions develop along the cell
ly support those obtained by Griffiths and boundaries and the contents of the cells swell to
Stenton-Dozey (1981) for colonisation of kelp form rounded protruberances. The cocci border-
debris by Fucettia cape~h~ and Tatohchehtia ing the cells, which are either responsible for,
cape~h~ on an adjacent open sandy beach. Since or develop along, the lesion zones can be clearly
the period egg to 1st instar of F. cape~h~ is as seen in Fig. 3c .
short as 24 to 32 h (Stenton-Dozey, Griffiths,
1980) it is evident that kelp-fly eggs are laid This phase is followed by shedding of much of the
almost immediately after kelp is deposited on the epidermal layer of the kelp, revealing the honey-
strandline. comb-like cells of the underlying parenchymal
tissues (Fig. 3d). It is probable that this phase
TABLE 1 Numbers (N) and biomass (B, mg dry is associated with a major release of dissolved
weight) of consumer invertebrates per kg original and particulate organic matter from the cell con-
fresh weight of kelp E~hlo~ maxima on a mixed
substrate strandl ine. tents, which are known to contain high concentra-
tions of mannitol, laminarin, alginates and other
Day Talorchestia Ligia Fucellia carbohydrates (Jensen, Stein, 1978; Stephen, 1979;
capensis dilatata capensis
N B N B N B
Newell et al., 1980). The cell contents and
leachates are then colonised by dense communities
0 0 0.00 0 0.00 0 0.00
1 730 6789.01 3 65.34 58 104.98 of rod-like bacteria which often appear to be
2 1825 16698.83 8 131.58 89 153.97 attached to the kelp surface by fibrils. Occa-
3 1179 11825.49 56 1051.68 97 188.18
4 760 9196.00 32 617.60 125 248. 75 sional yeasts and fungi also occur (Fig. 3e). A
5 150 2055.03 5 69.87 110 234.30
6 89 1066.28 19.31 113 240.69
very similar sequential colonisation of kelp
7 32 457.64 0 0.00 98 194.04 exudates by cocci and subsequently by larger rods
8 11 189.89 1 21.53 101 190.89
under culture conditions has been described
recently by Linley and Newell (1981) and Linley
et al. (1981).
 253

a - 3pm b -10pm

c -10pm d 30 pm

e -3pm f -3pm
FIGURE 3 A selection of stages in the microbial colonisation of kelp (E~~onia max~a) debris.
(a) Day 0: small bacterial cocci attached to the surface of freshly cut kelp. Note that fimbriate
attachment structures radiate from the point of contact (see text).
(b) Day 2: mucilage release and bacterial colonisation along superficial cell borders.
(c) Day 3: lysis of the cell surface. The colonisation of cell borders can be clearly distinguished.
(d) Day 6: epidermal lesion leading to exposure of the underlying honeycomb-like cells. This phase
is associated with major loss of dissolved organic components by leaching.
(e) Day 6: colonisation of the underlying cells by bacterial rods and some yeasts.
(f) Day 8: massive population of bacterial rods which dominate the kelp surface during later phases
of decomposition.

This sequence of epidermal lysis, followed by process of decomposition took nearly 10 times as
bacterial invasion of the cell contents has al so long as in kelp. Howard-~Iill iams et al. (1978)
been reported in leaves of the mangrove Rh~zo­ and Robb et al. (1979) have also found that
rhona mangte (Cundell et al., 1979), although the primary microbial decomposition of the brackish
254

sago pond weed Potamogcton pectinatuh is associa- TABLE 3 Biomass (mg £-1) of bacteria colonising
ted with bacteria rather than with fungi, which initial leachate from kelp debris (Pool 1) , inter-
stitial water and associated sand grains, and
are generally regarded as the initial decomposers final drainage water (Pool 2) as a function of
of allochthonous material in terrestrial and time; ( ) no estimate available
freshwater habitats (Gessner et al., 1972). Day Pool 1 Interstitial Pool 2 Total
community
water sand total
Numbers and Biomass of Bacteria
Numbers and biomass of bacteria colonising the 0 0.28 0.00 0.63 0.63 0.44 1.35
5.33 0.33 5.65 5.98 1.45 12.76
surface of the kelp debris are shown in Table 2. 2 18.46 0.38 4.37 4.75 2.37 25.56
3 23.01 1.49 3.37 4.86 3.27 31.14
4 1.94 5.17 7.11 (
TABLE 2 Numbers (N mm- 2 ) and biomass (B, ~g dry 5
( )
27.30 8.91 3.77 12.68
(
1.76
)
41.74
)

weight mm- 2 ) of bacteria colonising freshly cut 6 31.85 5.98 7.75 13.73 1.42 47.00
EcRtonia max~a debris on the strandline as a 7 28.60 4.57 9.19 13.76 1.84 44.20
function of time 8 7.25 1.60 10.61 12.21 1.63 21.09

Day Numbers Mean vol. Biomass

(.10 6 ) (11m3 )

0 0.23 0.18 9.11 bacteria is related to the concentration of organ-

0.22 0.18 8.71
2 0.43 0.21 19.87 ic matter which decreased from the leachate
3 0.35 0.20 15.40 through the sand column and into Pool 2 (see
4 0.50 0.21 23.10
5 0.60 0.21 27.72 below). After d 6 the bacterial biomass in the
6 0.90 0.20 39.60 leachate of Pool 1 declined, probably reflecting
7 1.03 0.23 52.12
8 1.24 0.41 111.85 the final phase of release of dissolved organic
exudation from the kelp debris. The bacterial
biomass in the drainage Pool 2, however, remained
at a relatively constant lower value of
It should be emphasised that because of the
1.4 - 1.8 mg £-1 from d 5 onwards, suggesting that
structural complexity of the cell surface follow-
much of the organic matter from the leachate had
ing cell lysis from d 3 onwards, estimates are
been processed in Pool 1 and during its passage
likely to be minimal. By d 8, bacteria became
through the sand column. The increase in bac-
so dense both within the cells and over the whole
teria in the whole microcosm amounted to a total
surface of the debris (Fig. 3f) that accurate
of 622.2 g dry weight, of which 619.8 g were
counts were not possible. The counts do, however,
associated with the surface of the kelp itself,
show a regular increase in bacterial numbers and
1.4 g with the leachates in Pool 1, 0.6 g with
biomass which reached 112 ~g dry weight mm- 2 kelp
the interstitial water of the sand strip and
surface by d 8.
0.4 g with the larger volume of Pool 2.

The biomass (mg dry weight £-1) of bacteria
colonising the initial leachate in Pool 1, the
interstitial water and the sand grains, and the
final drainage Pool 2 are shown in Table 3. The
most obvious feature of the data is that bacterial
biomass reached up to 31.85 mg £-1 on d 6 in the
initial leachate, but only 1.42 mg £-1 in the
lower drainage Pool 2. Clearly, the biomass of
Removal of Organic Material by Grazing
The amount of kelp debris grazed directly by
Ta£ohche~tia capen~~, Ligia diiatata and Fuce£-
lia capen~~ may be calculated from the biomass
(Table 1) and energy budget equations for these
organisms (Muir, 1977; Stenton-Dozey, Griffiths,
1980; Koop, Field, 1981) which are summarised in
Table 4. The table shows that some 9% of the

kelp was consumed by the grazers in the microcosm
while 7% was returned as faeces for further decom-
position within the system. The dominant T. Qapen-
~~ removed 71% of the material consumed by the
grazing organisms.
TABLE 4 Consumption (C), faeces production (F)
and production (p) and, by difference,
respiration (R) in g dry weight of the 3 inverte-
brate consumer species for the 8-d experiment
5,270 g dry weight of kelp were originally
present. Compiled from the numbers of inverte-
brates counted in the microcosm, their mean
individual weights and the following energy
budget equations (C = P+R+F+U): Ligia ditatata
C = 0.8%+4.2%+71.7%+(); Koop and Field (1981).
TatonQh~tia Qape~~ C = 2.3%+7.4%+81.9%+5.5+;
Muir (1977). FUQeLtia Qape~~ C = 17%+22%+61%+
(); Stenton-Dozey and Griffiths (1980). () U
not estimated. All data for C and Wwere conver-
ted to dry weight from published values.
Species F p R
C In each case, however, the dominant pathway is
Ligia dilatata 60.71 46.72 0.49 13.50
Talorchestia capensis 333.85 273.42 7.65 52.78
Fucellia capensis 73.79 45.01 12.54 16.24
Total 468.35 365.15 20.68 82.52
% of kelp in microcosm 8.89 6.93 0.39 1.57
Regression equations for consumption vs weight were as
follows:
L. dilatata C = 1.15 . WO. 56 (C = g wet wt 24 h- I ;
W = g AFDW of individual)
T. capensis C = 0.90 . WO. 23 (C = mg dry wt 24 h- I ;
W = mg dry wt of individual)
F. capensis C = 0.27 . WO. 95 (C = mg dry wt 24 h- I ;
W = mg wet wt of individual)
These results contrast with those of Griffiths
and Stenton-Dozey (1981) who found that 74% of
the kelp on the surface of the adjacent open
sandy beach was consumed by amphipods and
dipteran larvae. However, in the open beach
environment, the biomass of the grazers was
approximately four times the biomass of grazers
recorded in the microcosm and accounts to some
extent for the increased percentage of the kelp
removed. All the grazers are, however, character-
ised by high consumption rates and low absorption
efficiencies (Table 4), a feature which may reswt
in a high energetic gain in the presence of
abundant food resources (Newell, 1980; Newell,
Branch, 1980).
255
It is likely that high consumption rates such as
these are characteristic of superficial algal
debris which is periodically wetted by the tide
or rainfall. Much of the kelp debris is
deposited in large piles often covered by sand,
making it both anoxic and largely unavailable to
surface-dwelling organisms. Under such conditions
decomposition rather than consumption by inverte-
brates is likely to be of importance. It is
interesting in this context that, in contrast to
bacteria in interstitial and nearshore waters, a
large proportion (92.5%) of those encountered
among rotting kelp are facultative anaerobes able
to ferment kelp exudates under anoxic conditions
(Davies et al., 1983). Rather variable rates of
direct consumption of kelp debris are thus to be
anticipated, depending largely on local conditions.
via the heterotrophic decomposer organisms since
up to 80% of the material consumed by the·
grazers is returned as faeces and is subsequent-
ly decomposed by bacteria.
Carbon, Nitrogen and Phosphate in the Microcosm
The concentrations of carbon, nitrogen and phos-
phate analysed in the initial leachate pool be-
neath the kelp (Pool 1), in the interstitial
water of the sand column, and in the final
drainage Pool 2 are summarised in Table 5. It is
evident that very high concentrations (up to
221 .33 ~g-at C.10 3 e- 1 corresponding to
5640 mg c·e- 1) occurred in the leachate zone
immediately beneath the decomposing kelp debris
and that this material was mainly utilised
during passage through the 1 m sand zone. Only
some 3-10% appeared in the final Pool 2 towards
the end of the experimental period of 8 d.
The high concentrations of dissolved organic
carbon which occur beneath the decomposing kelp
are in marked contrast to values found
I
in the
kelp bed or open oceanic water. R.A. Carter
(cited in Newell et al., 1980) reports a maximal
value of 20 mg c.e- 1 in the kelp bed water
256

TABLE 5 Nutrient values (~g-at £-1) in leachates from the kelp EQQto~ maxima (Pool 1), in the
interstitial water of a 1 m sand strip and in a subsequent drainage pool (Pool 2). The carbon values
were recalculated from Koop et al. (1982a). All data have been corrected for evaporation

Day C'103 NH3 NO z Total inorganic N

Pool 1
o 6.05 32.56 0.73 0.05 33.34 0.42
17.22 210.84 0.94 0.09 211.87 15.14
2 52.78 1535.60 0.94 0.06 1536.60 115.12
3 104.17 3443.00 0.98 0.06 3444.04 60.78
4 128.06 5372.64 1.49 0.06 5374.19 468.69
5 158.89 5809.34 2.63 0.12 5812.09 14.67
6 107.22 6853.69 3.61 0.13 6857.43 17.26
7 221.33 7280.60 4.71 0.16 7285.47 11.02

Interstitial
o 13.33 198.00 0.73 0.11 198.84 0.85
26.39 534.00 108.53 757.55 1400.08 1.50
2 22.33 697.00 89.28 361.60 1147.88 2.35
3 20.00 936.00 78.43 452.00 1466.43 1.34
4 20.25 906.00 21.37 1.72 929.09 2.35
5 25.25 1170.00 1.19 0.18 1171.37 8.57
6 14.17 992.00 6.75 0.93 999.68 2.35
7 13.89 1201.00 28.88 4.11 1233.99 1.79

Pool 2
o 8.06 31.52 1.09 0.18 32.79 0.68
6.39 255.65 13.43 1.11 270.19 3.02
2 6.39 939.40 1.07 0.10 940.57 6.67
3 11.39 1185.53 1.15 0.06 1186.74 9.46
4 9.42 1395.26 1.97 0.11 1397.34 6.30
5 8.58 1810.35 1.27 0.12 1811.74 12.76
6 8.58 2023.04 3.75 0.13 2026.92 13.38
7 10.56 2600.43 1.48 0.13 2602.04 16.21

during periods of strong wave action, when frag-
mentation of kelp fronds is greatest, falling to
less than 5 mg C.£-1 within 24 h with the onset
of calm weather. Clearly, the bacterial compo-
nent in the microcosm is capable of rapidly
oxidising much of the carbon in the leachates
from kelp debris before it is returned by drain-
age to the sea.
Very high concentrations of ammonia accumulated
over the experimental period after which time 92%
of the kelp initially supplied to the microcosm
had been decomposed. In contrast, nitrite, and
to a greater extent nitrate, were low except in
the interstitial water when grazing amphipods in-
vaded the sand for a period of 3 d. Phosphate
initially accumulated in the leachates but subse-
quently declined with time and with passage
through the system, a maximum of only
16.21 ~g-at P0 4 £-1 appearing in the final drain-
age pool.
In order to estimate the decomposition products
accumulating in the microcosm, it is necessary
to multiply the values shown in Table 5 by the
volumes of initial leachate, interstitial water
and the water in the final drainage Pool 2. The
values attained at the end of the experiment are
shown in Table 6. These represent products which
had not been utilised by microbial activity or
consumers and which are thus available for export
to the adjacent coastal waters. It is apparent
that approximately 95% of the carbon is used up
during passage through the sand column. Although
 257

there is some evidence of utilisation of nitrogen TABLE.7 A carbon and nitrogen mass balance for
in the sand, nitrogen accumulation in the 1ea- the decomposition of kelp EeRtonia maxima on the
strand1ine. Wet weight:dry weight and C:N
chates pool is similar to that draining from the ratios of 6.5:1 and 17.2:1 respectively were used
sand column to the sea (Pool 2). for kelp conversions. Percentage loss from
microcosm represents material not incorporated
into biomass or returned to the sea and is
TABLE 6 Total amounts (~g-at.103) of material probably lost to the atmosphere
from decomposing kelp EeRtonia maxima accumulat-
ing in 1eachates (Pool 1), interstitial water of Source Carbon ~itrogen
a 1 m sand strip, and final drainage Pool 2. (g) (g)
Data derived from amounts of material accumulated
on the final day of experiment (Table 5) multi- Total amount deposited 1300.5 75.8
plied by the volumes of water in each of the 3 Net removal by invertebrates
sections of the microcosm amount removed by grazing 115.7 6.7
amount returned as faeces 89.7 5.2
NO, N0 3 Total N P0 4 difference 26.0 1.5
Source Carbon NH3
% incorporation 2.0% 2.0%
Pool 1 4639.5 72.48 0.07 0.002 72.550 1.30 Residues in microcosm
Interstitial 0.0 8.02 0.04 0.004 8.064 0.01 dried kelp 104.0 6.1
Pool 2 233.1 77.07 0.01 0.000 77.080 0.46 dissolved matter in Pool 1 55.7 1.0
dissolved matter in interstitial 0.0 0.1
dissolved matter in Pool 2 2.8 1.1
Total residues 162.5 8.3
Material available for microbial 1112.0 66.0
Nitrogen leaching to the sea is dominated by
decomposition
ammonia. This is in agreement with the results Biomass of bacteria 311.1 62.2
of Raine and Patching (1980) and thus potentially Bacterial conversion efficiency 28.0% 94.2%
significant as a source of nitrogen for nearshore Percent returned directly to sea 0.2% 1.5%
Percent loss from microcosm 69.8% 2.3%
primary producers (Lewin et a1., 1975; McLachlan
et al., 1981). In order to arrive at a mass
balance for the reminera1isation of carbon and
nitrogen, however, it is necessary to express the Table 4 shows that 8.9% of the kelp carbon is
residual components leaching to the sea in terms directly removed by grazing invertebrates while
of the amounts contained in the material initially 6.9% is returned as faeces. This yields a net
cast ashore. loss of 26.0 g C (=2%) from the microcosm by
grazing. The residual dried kelp remaining at
Mass Balance for a Sand Beach Microcosm the end of the experiment amounted to 8% of the
It is now possible to calculate a mass balance carbon initially supplied and together with the
for carbon and nitrogen during decomposition of sum of the residual carbon in the water (Table 7)
the kelp debris and to estimate the quantitative this gives 162.5 g C. The carbon available for
significance of mineralisation products leaching microbial decomposition is thus 1300.5
to the sea during the time course of this experi- (26.0 + 162.5) = 1112.0 g. The total bacterial
mental microcosm. The results are summarised in biomass in the microcosm has been estimated to be
Table 7. Since the wet weight:dry weight ratio 622.2 g dry weight which yields a carbon equiva-
of Eeklonia maxima is 6.5:1 (Newell et a1., 1980) lent of 311.1 g (Luria, 1960). The carbon con-
and the carbon and nitrogen were found to be version efficiency by the bacteria in the micro-
32.08% and 1.87% of the dry weight respectively, cosm is thus 28.0%, a value which accords well
the initial 26.35 kg wet kelp supplied to the with that of 31% recorded for similar bacteria
microcosm corresponds to 1300.5 g-C and 75.6 g-N. by Fenche1 and Blackburn (1979). It is also
clear that carbon leaching from the microcosm to
258
the sea (Pool 2) amounts only to some 0.2% of
that originally supplied in the debris while that
lost, presumably as CO 2 to the atmosphere, is
100 - (2 + 28.0 + 0.2) = 69.8% (Fig. 4).
KELP DEBRIS
CARBON
Leachates
to sea
FIGURE 4 Mass balance of carbon from decomposing
kelp (EQR{onla maxima) on the strandline. The
figures are expressed in g carbon derived from an
initial 100 g carbon in kel p; data in paren-
theses represent the weight (g) of carbon flowing
through the system derived from the decomposition
of 100 g dry weight of kelp.
In much the same way, the net exchange of nitro-
gen within the microcosm and the significance of
drainage to the sea can be estimated from the
difference between the amount supplied in the
kelp and that incorporated into grazers plus
bacteria and that remaining in the water of the
microcosm (Table 7). In this case the nitrogen
removed by grazing amounted to 6.7 g while the
nitrogen equivalent of the faeces was 5.2 g
yielding a net loss of 1.5 g N to grazers. The
total residual nitrogen in the microcosm at the
end of the study was 8.3 g (Table 7) leaving
66.0 g N for microbial utilisation. The nitro-
genequivalent of the bacterial biomass, using a
C:N ratio of 5:1 (Fenchel, Blackburn, 1979) is
62.2 g. The nitrogen conversion efficiency of
the bacteria in the microcosm was thus as high as
94.2%. Although the concentrations of ammonia
returned to the sea as leachates were high, it is
apparent that these comprise only 1.5% of the
nitrogen supplied in the kelp debris. Nitrogen
loss to the environment amounted to only 100 -
(.1.5 + 94.2 + 2) 2.3% (Fig. 5).
KELP DEBRIS
INITROGEN I
Leachates
to sea
FIGURE 5 Mass balance of nitrogen from decompos-
ing kelp (EQR{onia maxima) on the strandline.
The figures are expressed in g nitrogen derived
from an initial 100 g nitrogen in kelp; data in
parentheses represent the weight (g) of nitrogen
flowing through the system derived from the
decomposition of 100 g dry weight of kelp.
The mass balance for the decomposition of kelp
thus clearly shows that microbial incorporation
of nitrogen from the kelp debris is as much as
94.2% efficient and is accompanied by oxidation
of approximately 70% of the carbon. However,
because there is no net accumulation in sandy
beaches with successive cycles of decomposition,
it is obvious that the material incorporated
into bacteria will eventually be returned to the
environment.
Recent work on carbon conversion efficiencies of
bacteria on natural mixed substrates under
environmental conditions (Linley et al., 1981;
Linley, Newell, 1981; Lucas et al., 1981;
Newell, Lucas, 1981; Newell et al., 1982) has
shown rather low carbon conversion efficiencies
of about 10%. In similar experiments Stuart et
al. (1981, 1982) have, however, recorded carbon
conversion efficiencies of approximately 30%,
much as in the present study. A feature of both
these studies is that high concentrations of
nitrogen were available to the microbial communi-
ty. In a study on degradation of Sp~na debris

Newell et al. (1983) have tested bacterial carbon
conversion efficiencies under conditions of in-
organic nitrogen enrichment. They recorded a
range of carbon conversion efficiencies between
10% and 35% with highest values being recorded
when most nitrogen had been added. The efficien-
cies could be further increased by enrichment
with both inorganic nitrogen and phosphorus which
agrees with results obtained by Fenchel and
Harrison (1976). This indicates that there is a
complex interaction between inorganic nutrients
and the bacterial conversion of detritus and
highlights the need to investigate not only flows
of energy or carbon but also the fluxes of inor-
ganic nutrients when studying such ecosystems.
CONCLUSION
Using the data cited above for the carbon and
nitrogen exchange within the period of the
experimental sand beach microcosm, it is possible
to estimate the annual fluxes on the strandline
and the potential significance of ammonia regene-
rated from microbial activity in the intertidal
zone for primary production in the adjacent
coastal waters. Koop et al. (1982a) have
estimated that 206.1 g carbon d- l are cast onto
each metre of strandline from a kelp bed at
Kommetjie, whose production is known. Since the
C:N ratio of kelp is 17.2:1, the nitrogen input
to the beach is 12.0 g N m- l of strandline d- l .
From Table 7 we know that 0.2% of the carbon and
1.5% of the nitrogen in the microcosm is regene-
rated and returned to the sea by leaching. The
carbon returned to the sea is thus 0.4 g m- l of
strandline d- l , while the nitrogen is
0.2 g m- l d- l
Newell et al. (1982) have summarised the primary
production of Cape coast kelp beds and have
reported that the combined production of kelp
bed phytoplankton, understorey algae and kelp
plus epiphytes amounts to 58,700 kJ m- 2 yr- l .
For a kelp bed of 1,000 m width seawards, such as
259
that found at Kommetjie, the carbon primary
production would thus amount to 3,710 g C for a
1 m wide strip d- l • Equally, nitrogen product-
ion would amount to 215.7 g N d- l in the same
1 m x 1,000 m strip. Since, as shown above,
nitrogen regeneration from the microcosm is
0.2 g m- l of beach d- l , it follows that leaching
could supply only 0.2:215.2 = 0.09% of the nitro-
gen requirements of the adjacent kelp bed. Even
if all the nitrogen deposited on the beach as
kelp was ultimately returned to the sea, it is
clear that this could meet only 12.0:215.7 =
5.6% of the nitrogen requirements of the adjacent
kelp bed.
It is also possible to estimate the potential
significance of microbial regeneration of ammonia
to surf zone phytoplankton. Primary production
by the kelp bed phytoplankton is estimated·as
32,800 kJ m- 2 of kelp bed yr- l (Carter, 1982).
McLachlan et al. (1981) have defined a surf zone
of 1 m width and depth x 200 m seawards as
characteristic of exposed sandy beaches. The
nitrogen requirements of such a zone would amount
to 50 g N d- l using a C:N ratio of 6:1 for phyto-
plankton and the primary production figures cited
above. Leachates from the microcosm could thus
supply only 0.2:50 = 0.4% of the estimated phyto-
plankton requirements even from a strandline
where organic input is high. If all the nitrogen
cast up onto the strandline as kelp was ultimately
returned to the sea, this could meet 12:50 = 24%
of the estimated phytoplankton requirements,
assuming the nitrogen was in a suitable form for
incorporation by phytoplankton. However, on open
exposed beaches where less kelp is deposited on
the strandline and where primary production by
the surf zone diatoms is high (Lewin et al.,
(1975), the contribution of nitrogen fromthis
debris is likely to be even less than 0.4% of the
phytoplankton requirements.
The sand beach ecosystem thus receives an organic
260
input from the adjacent coastal zone and largely
conserves nitrogen in the short term from the
debris cast up onto the strandline. While as
much as 70% of the carbon is oxidised by the
microbial community, combined losses of nitrogen
account for only 3.8% of the original amount
supplied to the shore. A similar oxidation of
carbon and immobilisation of nitrogen appears to
occur in a benthic kelp bed community (see Newell,
Field, 1982). Although there is no information
to date on the further fate of the material
immobilised by bacteria, it is clear that some or
most of the nitrogen may leach back to the sea
after the bacteria themselves have been
oxidised. This would indicate that as much as
95% of the nitrogen cast up onto the strandline
may ultimately be returned to the sea, though
there may be further losses to the atmosphere at
this step. However, even if a very large propor-
tion of the nitrogen input returns to the near-
short water, this could still meet only a rather
small fraction of the nitrogen requirements in
that habitat. Primary production in the kelp bed
community must therefore be dependent mainly on
in ~~u regeneration of nutrients and on local
upwelling. Although the high concentrations of
ammonia regenerated from the decomposing kelp may
have strictly localised significance in the
immediate subtidal zone, the mass balance for the
decomposition of macrophyte debris shows clearly
that leachates from the strandline could meet
only a small proportion of the nitrogen require-
ments of the primary producers in the adjacent
nearshore ecosystem.
REFERENCE
Armstrong FAG,Sterns CR and Strickland JDH
(1967) The measurement of upwelling and subsequent
biological processes by means of the Technicon
Autoanalyzer and associated equipment, Deep Sea
Res. 14, 381-389.
Backlund HO (1945) Wrack fauna of Sweden and
Finland. Ecology and chorology, Opuscula
Entomologica (Supplementum) 5, 1-236.
Borchers P'and Field JG (1981) The effect of
kelp shading on phytoplankton production,
Botanica mar. 24 ,89-91.
Bowden WB (1977) Comparison of two direct-count
techniques for enumerating aquatic bacteria, Appl
Env. Microbiol. 33, 129-132.
Brown PC (1980) Phytoplankton production
studies in the coastal waters off the Cape Penin-
sula, South Africa. MSc thesis, Zoology Depart-
ment, University of Cape Town, Rondebosch 7700,
South Africa 98pp.
Brown PC (1981) Pelagic phytoplankton, primary
production and nutrient supply in the southern
Benguela region. Proceedings of the Symposium
on Upwelling Processes in the Southern Benguela
Region, Trans. R. Soc. S. Afr. 44, 347-356.
Carter RA (1982) Phytoplankton biomass and
production in a southern Benguela kelp bed
system, Mar. Ecol. Prog. Ser. 8, 9-14.
Cundell AM, Brown MS, Stanford R and Mitchell R
(1979) Microbial degradation of Rh~zophona mangle
leaves immersed in the sea, Estuar. coast. mar.
Sci. 9, 281-286.
Cundell AM, Sleeter TO and Mitchell R (1977)
Microbial populations associated with the surface
of the brown alga A6eophyllum nodo~um, Microb.
Ecol. 4, 81-91.
Davies C, Koop K, Muir DG and Robb FT (1983)
Bacterial diversity in adjacent kelp dominated
habitats, Mar. Ecol. Prog. Ser. (submitted to
press) .
Dieckmann GS (1978) Aspects of growth and
production of Lam~n~ p~da (Grev.), J. Agr.
off the Cape Peninsula. MSc thesis, Botany
Department, University of Cape Town, Rondebosch
7700, South Africa, 98pp.
Dieckmann GS (1980) Aspects of the ecology of
Lamin~ p~da (Grev.) J. Agr. off the Cape
Peninsula (South Africa). I. Seasonal Growth,
Botanica Mar. 23, 579-585.
Fenchel T and Blackburn TH (1979) Bacteria and
mineral cycling, Academic Press, London, 225pp.
Fenchel T and Harrison P (1976) The signifi-
cance of bacterial grazing and mineral cycling
for the decomposition of particulate detritus.
In: The role of terrestrial and aquatic organisms
in the decomposition processes (JM Anderson,
A Macfayden, eds), Blackwell, Oxford, 285-299.
Fitzgerald B (1979) A preliminary investigation
of the filtration rates of Pyuna ~~olo~6eha
(Heller) under laboratory conditions. Unpubl.
Honours Project, Zoology Department, University
of Cape Town, Rondebosch 7700, South Africa, 19pp.
Field JG, Jarman NG, Dieckmann GS, Griffiths
CL, Velimirov Band Zoutendyk P (1977) Sun, waves,
seaweed and lobsters: the dynamics of a west
coast kelp bed, S. Afr. J. Sci. 73, 7-10.
Field JG, Griffiths CL, Griffiths RJ, Jarman NG,
Zoutendyk P, Velimirov B and Bowes A (1980)
Variation in structure and biomass of kelp
communities along the south west Cape coast,
Trans. R. Soc. S. Afr. 44, 145-203.
Gessner RV, Goos RD and Sieburth JMcN (1972)
The fungal microcosm of the internodes of
Spantina ~eh~6lona, Mar. Biol. 16, 269-273.
Greenwood PJ (1974) The population dynamics and

ecological energetics of Pan~~hi~Uh a~gU£o~u~ at
Robben Island and in False Bay, South Africa.
MSc thesis, Zoology Department, University of
Cape Town, Rondebosch 7700, South Africa, 71pp.
Greenwood PJ (1980) Growth, respiration and
tentative energy budgets for two populations of
the sea urchin Pan~~hi~Uh a~gU£o~Uh (Leske).
Estuar. coast. mar. sci. 10, 347-367.
Griffiths CL and King JA (1979a) Some relation-
ships between size, food availability and energy
balance in the ribbed mussel AU£a~omya cUVL, Mar.
Biol. 51, 141-149.
Griffiths CL and King JA (1979b) Energy
expended on growth and gonad output in the ribbed
mussel AU£a~omya cUVL, Mar. Biol. 53, 217-227.
Griffiths CL and Stenton-Dozey J (1981) The
fauna and rate of degradation of stranded kelp.
Estuar. coast. Shelf Sci. 12, 645-653.
Hayes WB (1974) Sand beach energetics: import-
ance of the isopod Ty{o~ pu~~tat~. Ecology 55,
838-847.
Hobbie JE, Daley RT and Jasper S (1977) Use of
Nuclepore filters for counting bacteria by
fl uorescence microscopy, Appl. Env. Microbiol. 33,
1125-1128.
Howard-Williams C, Davies BR and Cross RHM
(1978) The influence of periphyton on the surface
structure of a Potamogeto~ p~~~~cUUh L. leaf
(an hypothesis), Aquatic Botany 5, 87-91.
Jarman NG and Carter RA (1981) The primary
producers of the inshore regions of the Benguela.
Symposium on Upwelling Processes in the Southern
Benguela Region, Trans. R. Soc. S. Afr. 44, 321-
326.
Jensen A and Stein JR (eds) (1978) Proceedings
of the international seaweed symposium 9, Science
Press, Princeton, USA, 634 pp.
Koop K and Field JG (1980) The influence of
food availability on population dynamics of a
supralittoral isopod, Ligia ditatata Brandt, J.
expo mar. Biol. Ecol. 48, 61-72.
Koop K and Field JG (1981) Energy transforma-
tion by the supralittoral isopod Ligia ditatata
Brandt, J. expo mar. Biol. Ecol. 53, 221-233.
Koop K, Newell RC and Lucas MI (1982a) Bio-
degradation and carbon flow based on kelp
(E~~{o~ maxima) debris in a sandy beach micro-
cosm, Mar. Ecol. Prog. Ser. 7, 315-326.
Koop K, Newell RC and Lucas MI (1982b) Microbial
regeneration of nutrients from the decomposition
of macrophyte debris on the shore, Mar. Ecol.
Prog. Ser. 9, 91-96.
Laycock RA (1974) The detrital food chain based
on seaweeds. I. Bacteria associated with the
surface of Lami~~ fronds, Mar. Biol. 25,
223-231.
Linley EAS and Newell RC (1981) Microhetero-
trophic communities associated with the
degradation of kelp debris, Kieler Meeresforsch.
5, 345-355.
Linley EAS, Newell RC and Bosma SA (1981)
Heterotrophic utilisation of mucilage released
during fragmentation of kelp (E~~o~ maxima
and Lami~~ p~da). I. Development of micro-
261
bial communities associated with the degradation
of kelp mucilage, Mar. Ecol. Prog. Ser. 4, 31-41.
Lucas MI, Newell RC and Velimirov B (1981)
Heterotrophic utilisation of mucilage released
during fragmentation of kelp (E~~o~ maxima and
Lami~ania p~da). II. Differential utilisation
of dissolved organic components from kelp muci-
lage, Mar. Ecol. Prog. Ser. 4, 43-55.
Lewin J, HrubyT and Mackas D (1975) Blooms of
surf zone diatoms along the coast of the Olympic
Peninsula. Washington. V. Environmental con-
ditions associated with the blooms (1971 and
1972), Estuar. coast. mar. Sci. 3, 229-241.
Luria SE (1960) The bacterial protoplasm:
composition and organisation. In: Gunsalus IC
and Stanier RY (eds). The bacteria, Vol. 1,
Academic Press, New York, pp.1-34.
Mann KH, Jarman NG and Dieckmann GS (1979)
Development of a method for measuring the
productivity of E~~o~ia maxima (Osbeck) Papenf,
Trans. R. Soc. S. Afr. 44, 27-42.
McLachlan A, Erasmus T, Dye AH, Wooldridge T,
van der Horst G, Labiak TA and McGwynne L (1981)
Sand beach energetics: an ecosystem's approach
towards a high energy interface, Estuar. coast.
Shelf Sci, 13, 11-25.
Mazure HGF and Field JG (1980) Density and
ecological importance of bacteria on kelp fronds
in an upwell ing region, J. expo mar. Biol. Ecol.
43, 173-182.
Muir DG (1977) The biology of Ta{Oh~h~~
~ap~~~ (Amphipoda : Talitridae), including a
population energy budget. MSc thesis, Zoology
Department, University of Cape Town, Rondebosch
7700, South Africa, 94pp
Newell RC (1980) The maintenance of energy
balance in marine invertebrates exposed to
changes in environmental temperature. In: Gilles
R (ed.) Animals and environmental fitness,
Pergamon Press, Oxford and New York.
Newell RC and Branch GM (1980) The influence of
temperature on the maintenance of metabolic
energy balance in marine invertebrates, Adv. mar.
Biol. 17, 329-396.
Newell RC and Lucas MI (1981) The quantitative
significance of dissolved and particulate organic
matter released during fragmentation of kelp in
coastal waters, Kieler Meeresforsch 5, 356-367.
Newell RC and Field JG (1983) The contribution
of bacteria and detritus to carbon and nitrogen
flow in a benthic community, Mar. Biol. Letters
4, 23-36.
Newell RC, Field JG and Griffiths CL (1982)
Energy balance and the significance of micro-
organisms in a kelp bed community, Mar. Ecol.
Prog. Ser. 8, 103-113.
Newell RC, Linley EAS and Lucas MI (1983)
Microbial production and carbon conversion based
on saltmarsh plant debris, Estuar. coast. mar.
Sci. (in press).
Newell RC, Lucas MI, Velimirov Band Seiderer
LJ (1980) Quantitative significance of dissolved
organic losses following fragmentation of kelp
(E~~o~a maxima and Lami~~ pa{{ida). ~1ar.
262

Ecol. Prog. Ser. 2,45-59. Il.CKNO~lL EDGEII'nITS

Paerl HW (1975) Microbial attachment to We are grateful to Ms Colette Weimar of the
particles in marine and freshwater ecosystems,
Microb. Ecol. 2, 73-83. Pollution Monitoring Unit, National Research
Pollock DE (1981) Carnivores of the inshore Institute for Oceanology, and Mr W.R.T. Hemsted
region of the Benguela upwelling system. Proceed-
ings of the Symposium on Upwelling Processes in of the Organic Chemistry Department, University
the Southern Benguela Region, Trans. R. Soc. S. of Cape Town, for carrying out some of the
Afr. 44, 341-345.
Robb FT, Davies BR, Cross R, Kenyon C and analyses. The work was supported by funds from
Howard-Williams C (1979) Cellulolytic bacteria as the South African National Committee for Oceano-
primary colonizers of Potamogeton peeti~ L.
(Sago pond weed) from a brackish south-temperate graphic Research (SANCOR) through the Benguela
coastal lake, Microb. Ecol. 5, 167-177. Ecology Programme at the University of Cape Town.
Rodina AG (1972) Methods in aquatic microbiolo-
gy, University Park Press, Baltimore, USA.
Shafir A and Field JG (1980a) Population
dynamics of the isopod C~olana irnpo~~
(Crustacea: Malacostraea) in a kelp bed,
Crustaceana 39, 185-195.
Shafir A and Field JG (1980b) The importance of
a small carnivorous isopod in energy transfer,
Mar. Ecol. Prog. Ser. 3,203-215.
Sieburth JMcN and Thomas CD (1973) Fouling on
eelgrass (Zo~teka m~na L.), J. Phycol. 9,46-50.
Sorokin VI and Kadota M (eds) (1972) Techniques
for the assessment of microbial production and
decomposition in fresh waters, I. BP Handbook No.
23, Blackwell Sci. Publ., Oxford, 112pp.
Stenton-Dozey J and Griffiths CL (1980) Growth,
consumption and respiration by larvae of the
kelp-fly Fueellia eape~i¢ (Diptera : Anthomyii-
dae), S. Afr. J. Zool. 15,280-283.
Stephen AM (1979) Plant carbohydrates. In:
Pirson A and Zimmermann MH (eds), Encyclopedia of
plant physiology 8, Springer-Verlag, Heidelberg,
pp.555-584.
Strickland JDH and Parsons TR (1972) A
practical handbook of seawater analysis, 2nd ed.
Bull. Fish. Res. Bd Can. 167.
Stuart V, Lucas MI and Newell RC (1981)
Heterotrophic utilisation of particulate matter
from the kelp Lamin~ pallida, Mar. Ecol. Prog.
Ser. 4, 337-348.
Stuart V, Newell RC and Lucas MI (1982)
Conversion of kelp debris and faecal material
from the mussel Aulaeomya at~ by marine micro-
organisms, Mar. Ecol. Prog. Ser. 7,47-57.
Todd RL and Kerr TJ (1972) Scanning electron
microscopy of microbial cells on membrane
filters, Appl. Microbiol. 23,1160-1162.
Troitsky AS and Sorok in VI (1967) On the methods
of the calculation of the bacterial biomass in
water bodies, Trans. Inst. Biol. Inland Waters
Acad. Sci. USSR 19, 85-90.
Velimirov B, Field JG, Griffiths CL and
Zoutendyk P (1977) The ecology of kelp bed
communities in the Benguela upwelling system.
Analysis of biomass and spatial distribution,
Helgolander wiss. Meeresunters, 30, 495-518.
 263

THE SANDY BEACH AREA OF KIEL FJORD AND KIEL BIGHT (WESTERN BALTIC SEA) - A STRUCTURAL ANALYSIS OF A
SHALLOW WATER ECOSYSTEM

MANFRED BOLTER+ and MICHAEL MEYER++ (+Institut fUr Polarokologie, Universitat Kiel, 0-2300 Kiel 1,
F.R.G.; ++Institut fUr Meereskunde an der Universitat Kiel, 0-2300 Kiel 1, F.R.G.)

1. INTRODUCTION
parameter interactions and the structuring
The ecosystem of sandy beaches shows a of this environment.
high complexity of interrelationships We would like to thank our colleagues, Drs. W.
between physical, chemical and biological Balzer, R. Dawson, G. Graf, G. Liebezeit, I.-A.
properties and may be regarded as a spe- Meyer-Reil, S. Schnack, B. Schneider, H. Szwe-
cial biotope at the boundary between the rinski and K. Wolter who provided us with the
pelagic environment, the sediments and data of the individual analyses. Valuable tech-
adjacent terrestrial zones. A review on nical assistance is acknowledged to H. Tiedtge,
such interactions with special regard to W. Schmidt, I. Rei6mann and U. Oest. Mr. W.
biological features has been given recent- Dzomla prepared the graphs. The investigation
ly by McLachlan et al. (1981), lending was financed by the Deutsche Forschungsgemein-
support to the description of these schaft (SFB 95) and carries the contribution
niches as autonomous systems. number 407 of the Sonderforschungsbereich 95 at
The study presented here is part of a Kiel University.
joint research programme carried out
during 1977/1978 in order to get detailed
information on the interactions between 2. PROCEDURE
the lower organisms, in particular bac- 2.1. Material and methods
teria, and their environment. Recent pub-
A total of 53 parameters was measured at
lications (Meyer-Reil et al., 1980 ; Bolter
12 stations in the overlying waters of
et al., 1981) have shown this in detail
sandy beaches in Kiel Fjord and Kiel
with special regard to a summer aspect by
Bight during 1977/1978 in summer (July),
describing the mutual interrelationships
spring (March) and late autumn (November).
and establishing a diagram of carbon flow
Concomitantly, 62 parameters were ana-
in beach sediments. Combining these re-
lysed in the sediments, including some
sults with further analyses at two other
additional properties of this environ-
seasons (spring and late autumn) by an
ment. The data set comprises physical,
exploratory data analysis, the effects
chemical, microbiological and planktonic
of locality and seasonality in this sys-
characteristics of the system. Detailed
tem became more evident. Thus, this
lists of the parameters and descriptions
approach is intended to give an overall
of the analytical methods are given in
view on the interacting items, concerning
previous reports (Meyer-Reil et al.,
both the aspects of the (Qualitative)
1980; Bolter et ~l.,1981). The locations
264
are depicted in Fig. 1.
Fig. 1. Map of the sampling stations in
the area of Kiel Fjord and Kiel
Bight (Western Baltic Sea).
The statistical treatment on the data set
was carried out according to Bolter et
al. (1980) and Meyer, Bolter (1981).
complete linkage clustering was chosen
because this method refers to a strong
grouping, i.e. all members of a cluster
show at least the similarity (correla-
tion coefficient) of the given threshold
(p> 5 %). Further, the stability of the
groups was tested by using the average
linkage and single linkage procedure.
The algorithm of the cluster analysis
follows the formulation given by Lance,
Williams (1967) and refers to distance
measures. The original similarity index
is based on the Spearman rank correlation
coefficient which is modified in the
follwing way: a) A correlation coeffi-
cient ~f 1.0 is transferred to -3.0; b)
an integer value of -2 is added to each
correlation coefficient which holds true
on the %- significance level; c) corre-
lation coefficients which hold true on the
5 %- level are transferred to -1.
This procedure causes a change of the ori-
ginal sign of the correlation coefficient
but became necessary in order to include
the different degrees of freedom into
the cluster analysis. Thus, the individual
data sets may contain "missing values".
2.2. Structure of the analysis
The original data are given into a n x m
-matrix, where n = number of the indivi-
dual data per parameter (here: 12 x 3),
m = number of the parameters analysed
(here: 115, i.e. 53 in the overlying
water + 62 in the sediments).
Different subdivisions of this matrix are
possible by extracting individual rows or
columns of the complete matrix.
First, we analysed the mutual interrela-
tionships of the parameters with respect
to their evaluation, season or location,
water or sediment. Next, we divided the
The complete data files of the water and se-
diment into two groups concerning the
geographical aspect of Kiel Fjord (sta-
tions A, B, C, G, H, J) and Kiel Bight
(stations D, E, F, K, L, M) (see Fig. 1) •
During this analysis we ignored the sign
of the correlation coefficient and fo-
cused only on their absolute values.
After this, the matrix was normalized and
transposed. The subsequent analysis shows
a division into seasonal and local sub-
systems. Here we used the sign of the
correlation coefficient as an indicator
of possible intersections of the system
analysed due to seasonal or local effects.
All computer operations were carried out
at the computer centre of Kiel University.

3. RESULTS
3.1. Analysis of the individual periods
(July, March, November), overlying water
The complete linkage clustering of the
data set of July 1977 shows the unity of
components of the dissolved organic
matter (DOM), glucose and fructose, with
parameters describing microbial activity
(uptake parameters, microbial production
and remineralization rates). The average
linkage clustering adds the number of
active cells (estimated by microauto-
radiography) to this group. These para-
meters show no significant correlations
to particulate matter, inorganic nutrients,
environmental factors and total micro-
bial biomass. The latter parameters form
separate clusters. Total microbial bio-
mass and the total number of bacteria
are connected to water temperature.
Filamentous cells are combined with the
biochemical oxygen demand (BOD) and the
particulate matter. Saprophytic bacteria
are connected with salinity and inorganic
nitrogen sources.
This pattern of interactions changes for
the results concerning the investigation
in March 1978. However, it should be
borne in mind that we added some new
parameters during this analysis (ATP,
energy charge and total nucleotides; for
details see Graf, 1979). Filamentous
cells were not observed. The structure
found shows a rather different pattern
as described for July. As such, the
parameters of microbial activity are not
as closely related to its sources, glu-
cose and fructose. The determinations of
saprophytic bacteria form a separate
cluster. Inorganic nutrients, including
phosphate (the latter one remained un-
clustered in July), show close relation-
ships to DOM and the standing stock of
265
microheterotrophs. The active number and
biomass remain unclustered.
The analysis carried out in November 1978
results In a pattern which is rather
similar to that of March. The clusters
obtained by the different algorithms are
generally identical. Again, the data
concerning microbial activity are sepa-
rated from those of organic and inorga-
nic nutrients and environmental factors,
only chlorophyll shows a connection to
the latter parameters. All inorganic
nutrients are combined in a separate
cluster. The microbial standing stock
is related to glucose, whereas fructose,
saprophytes and other parameters remain
unclustered.
3.2. Analysis of the total file "over-
lying water"
This data file comprises 28 parameters,
since only those parameters entered into
this analysis which were monitored during
all individual studies. The cluster
analysis reveals 4 groups, only phosphate
and the ratio glucose:fructose remain un-
clustered.
The first cluster combines all data of
microbial activity, ammonia and DOM.
Cluster 2 comprises the inorganic nitro-
gen compounds (nitrite, nitrate), tempe-
rature, pH, oxygen and the microbial
standing stock. Cluster 3 combines sali-
nity, chlorophyll, saprophytes and the
ratio microbial biomass:number. Cluster
4 represents fructose, glucose and its
percentage of the DOM.
3.3. Partitioning of , the system into
"Kiel Fjord" and "Kiel Bight"
The analysis on the parameters measured
at the stations of Kiel Fjord shows 4
groups of closely interacting parameters:
266
i) The parameters of microbial biomass
and number are connected with DOM and the
ratio of inorganic N:P, the microbial In-
corporation of glucose and microbial bio-
mass production; ii) inorganic N-nutrients
are combined with temperature, pH, and
microbial remineralization rate; iii)
saprophytic bacteria are related to sali-
nity; iiii) the DOM-components glucose
and fructose. Unclustered remain chloro-
phyll, oxygen, phosphate and the ratio
glucose: fructose.
The parameters of the station Kiel Bight
show more homogeneity: the above mentioned
clusters i and ii are combined, no deci-
sion can be performed between the micro-
bial standing stock, the turnover of
glucose and fructose, temperature, N-
nutrients, oxygen and DOM. The second cluster
contains glucose and the microbial respi-
ration; the third one ammonia and the
actual uptake of glucose, its reminera-
lization rate and the biomass production.
Another cluster shows again the connection
of salinity and saprophytic bacteria. Un-
clustered remain pH, chlorophyll, phos-
phate, and the ratios glucose:fructose
and microbial biomass:number.
The view on these results gained by the
different approaches in structuring the
various interrelationships shows - on
one hand - the variability of interactions
within the system. On the other hand, it
points to some homogeneity of interacting
subsystems, as expressed by the stability
of the clusters with regard to the diffe-
rent cluster algorithms.
3.4. Analysis of sediment parameters
The corresponding analysis of the sedi-
ment parameters shows a division into
numerous and smaller clusters. As such,
group 1 in July combines the data of
microbial uptake parameters and chloro-
phyll; group 2 combines the inorganic
nutrients. The organic content of the
sediment is connected with the amount of
the microbial standing stock. The other
clusters (in total 9) describe several
mutual interrelationships, however, many
of these interactions seem to be random
at this state of the analysis.
This pattern holds also true for the
analysis of March: The cluster analysis
reveals in 11 groups and many "pairs" of
interacting characteristics. The exami-
nation reflects relations between the
grain size, primary production and micro-
bial standing stock. Further, parameters
concerning inorganic nutrients, ATP,
microbial respiration, dissolved carbo-
hydrates and the turnover of glucose are
connected In several clusters. The
microbial biomass measured in the pore
water is connected to DOM, nitrite and
phosphate; saprophytes of the sediment
and of the pore water are related to the
organic content of these environments and
to the amount of free amino acids in the
pore water.
The situation In November underlies a
similar pattern of diversification. Only
a few small groups could be established
with respect to the given threshold. As
such, the microbial standing stock of
the pore water is connected to the total
inorganic nitrogen; the saprophytes of
the sediment are related to nitrate,
glucose, microbial biomass production
and the remineralization rate.
This division into small groups of inter-
acting items at low similarity levels
points to the different structure of the
interactions in water and sediment. The
overall analysis of the sediment para-
meters supports this statement. The
 267

interaction of environmental factors with

biological properties is only evident by Water
the relation between saprophytic bacteria A
and DOM in Kiel Fjord and salinity in Kiel J
G
Bight. Further, in the latter area the C
B
microbial standing stock is related to o
M
March
glucose.
H
Thus, it becomes difficult to establish K
E
general conclusions on stable interacting F
L
subsystems in this environment. However,
the combination of both water and sedi-
ment, makes it possible to suspect that
the inorganic nutrients serve a central
position in these systems, as shown by November
close relationships to microbial activity,
in particular with characteristics which
describe the remineralization rate, and
saprophytic bacteria and the total stan-
ding stock. Further, salinity seems to
dominate the saprophytes, indicating a
relationship to the location of the
sampling areas. July
The varying ecological conditions concer-
ning the different stations are obviously
influenced by environmental factors. Similarity Index
j , , , ,

Thus, the general decision of the total 2 o -2 -21. -2.6 -2.8 -3

area into the groups of Kiel Fjord and
Kiel Bight needs a better validation
with respect to the individual stations
and their environmental characteristics. Fig. 2. Dendrog:mm of the similarity matrix
derived from the transposed matrix of the
parameters measured in the overlying waters
The stations are coded (A-M) as in Fig. 1 .
3.5. Analysis of the transposed matrix

At first glance, the analysis of the
system "overlying water" results ln a
clear separation between the three samp-
ling times, see Fig. 2, and an overall
division of the seasons summer (July),
spring (March) and late autumn (November)
becomes possible. With regard to the
stations of the individual periods, a
decision becomes feasible at rather
different boundaries than proposed above.
As such, in July the stations A, B, G,
and H form a cluster which reflects the
environment "Kiel Fjord", omitting the
stations C and J at the border to the
area of Kiel Bight. The latter stations
are more closely connected to this area
but show a separate position, underlining
their intermediate character between the
268

geographical areas Kiel Fjord and Kiel

Bight. Sediment
The situation found ln March lS quite
different from that ln July: there is no
clear separation into "Fjord" and "Bight".
The clusters combine stations of both
areas, indicating an indifferent biologi- March
cal pattern. Effects which might be due
to an increasing eutrophication and/or
stagnation of water masses in Kiel Fjord
seem to be of mlnor importance. Further,
the separations of the individual clusters
are without any significance, i.e. the
stations are connected at a high simila-
November
rity level (1 %).
In November, it is possible to establish
three groups of stations: the stations
F and M are separated by their lower cor-
relations to other ones. They represent
the outside stations of the investigation
area. Another cluster combines the sta-
tions C, D, and E which are located July
at the western site of Kiel Bight. An
analogy, however, cannot be found for the
corresponding area at the opposite site.
Similarity Index
Here, the stations are melted into one
2 o -2 -2.4 -2.6 -2.8 -3
cluster which allows no clear interpre-
tation, except that they may represent a
similar ecological feature.
Fig. ;). Drendrogram of the similarity matrix
The corresponding analysis on the sediment derived from the transposed matrix of the
parameters measured in the sediments of the
environment reveals a more complex pattern stations A-M (c.f. Fig. 1).
as shown in Fig. 3. The overall division
of the clusters is not basically related ln March show some homogeneity, however,
to the three periods of investigation. the stations C, F, G, K, and J represent
The separation of this environment re- a miKture of stations of Kiel Fjord and
sults ln groups independent of the sea- Kiel Bight. They have no significant
sons. As such, the cluster "July" is in- interactions to the remaining stations
consistent with regard to this prediction. of the period March.
Station G is excluded from this data set With regard to this clustering, the
and shows a more closer connection to some "November"-group shows the most hetero-
stations of "November" which, in turn, geneous structure. The stations analysed
are independent of the bulk of the sta- during this period are divided in two major
tions of this season. The data analysed groups, separated at a low similarity
 269

level. This special kind of structure In

the sediment environment accounts for the Sediment
different patterns of the parameter cluste-
ring in water and sediment as shown above.
The compari son of the groups show evidently
that there are only a few correlations between
these two systems. Further, it points to
an independence of both systems - with
regard to the parameters analysed and it
should be borne in mind that this analysis
is restricted to these individual snap-
shots. Thus, it is impossible to get an
insight into the complete dynamics of the
sediment - pore water - overlying water
interactions.
An illustration of the results of the
cluster analysis is given in Fig. 4. This
graph depicts the different local aspects
based on "Gnc lliscussion above and the Figs.
2 and 3.
The system "overlying water" shows a do-
minant structure with regard to the sea-
sons. During summer, the effect of eutro-
phication and water stagnation in Kiel
Fjord may explain the clear separation
into Kiel Fjord and Kiel Bight. In spring, Fig. 4. "Ecological areas" of the systems hlater
and sediment at three different seasons.
an indifferent situation seems to charac- The dots refer to the stations A - M,
terize a "waiting system", which is still see Fig. 1. Thick lines comprise clusters
of high similarity level (1%), thin lines
influenced by low biological activities represent a lOhler level (5%), compare
and the exchange of water masses seems to Figs. 2 and 3.
prevent a stabilization of characterizable
states. The late autumn still seems to be 4. CONCLUSIONS
influenced by the processes of the summer,
but at this time, the system seems to be Though one may argue that the separations
in a state of disintegration. Such a based on the correlation coefficients and
characteristic becomes impossible with the significance levels may be arbitrary,
regard to the sediment. Some possible the results give some insight into the
reasons for this different ecological ecological structure of the system and
pattern will be discussed below. its variations at this trophic level, i.e.
the microheterotrophs. The different
patterns of the interacti~ns in the
systems water and sediment reflect their
independence. These differen~ natterns
270
resulted In the overall ecological struc-
tures and may be due to the following
reasons: There are different time scales
of the biological activities and chemical
flux rates in these systems. Therefore,
a balance between the systems water and
sediment cannot be accomplished and re-
flected by this "snapshot" analysis. The
amounts of nutrients which are present
and/or available for microbial use are
quite different, even with regard to their
composition. Thus, different adaptions of
the biological components become obvious.
Large amounts of organic nutrients may be
"inactive" or "inattractive" in the sedi-
ment. Further, this environment seems to
react as a buffer with respect to its
inputs and the storage capacity. The micro-
bial system in the overlying water of
beach sediments is closely connected to
the pelagic environment which causes very
short exchange rates. Different internal
cycles of nutrient fluxes and species
successions are able to cause different
interactions within these systems. Thus,
it seems unlikely that there will be a
direct representation of the processes
occurring in the water by the processes
in the sediment and vice versa.
In summary, this qualitative approach In
structuring a comprehensive data set may
be used to depict the various types of
ecological patterns in this complex
environment influenced by seasonal and
regional impacts. Many detailed results
of this exploratory data analysis need
further investigation. This study was
only able to highlight some figures of
the interactions between water and sedi-
ment. Based on these results, the quanti-
fication"of several individual interre-
lationships between physical, chemical
and biological properties may be feasible.
Thus, this structuring may be regarded
as a parsing phase in systems analysis,
the choice of relationships bet¥een
interacting entities which are of inter-
est in complex models.
5. REFERENCE
Bolter M, Meyer M and Probst B (1980)
A statistical scheme for structural
analysis in marine ecosystems, Ecol.
Model. 9, 143-151.
Bolter M, Meyer-Reil L-A, Dawson R,
Liebezeit G, Wolter K and Szwerinski H
(1981) Structure analysis of shallow
water ecosystems: Interaction of micro-
biological, chemical and physical charac-
teristics measured in the overlying
waters of sandy beach sediments, Estuar.
Coast. Shelf Sci. 13,519-589.
Graf G (1919) Energy-Charge-Messungen
im marinen Benthos. Ph D Thesis, Kiel,
pp. 1-121.
Lance GN and Williams WT (196 1 ) A gene-
ral theory of classificatory sorting
strategies.I. Hierarchical systems,
Comput. J. 9, 313-380.
McLachlan A, Erasmus T, Dye AH,
Wooldridge G, van der Horst G, Roussow G,
Lasiak TA and McGurynne L (1981) Sand
beach energetics: An ecosystem approach
towards a high energy level, Estuar.
Coast. Shelf Sci. 13, 11-25.
Meyer M and Bolter B (1981) Programm-
block zur Strukturanalyse von Okosyste-
men, Rep. SFB 95, Kiel, pp. 1-4 1 .
Meyer-Reil L-A, Bolter M, Dawson R,
Liebezeit G, Szwerinski H and Wolter K
(1980) Interrelationships between micro-
biological and chemical parameters of
sandy beach sediments, a summer aspect,
Appl. Environm. Microbiol 39, 191-802.
 271

FOULING OF THE SANDY BEACHES OF NAHANT BAY (HASSACHUSETTS, USA) BY AN ABNORMAL FREE-LIVING FORM OF THE
HACROALGA PILAYELLA LITTORALIS (PHAEOPHYTA). I. HABITAT CHARACTERISTICS

ALICIAN V. QUINLAN (Duke University, Durham, NC 27706, USA)

TH10THY LEWIS (University of Rhode Island, Narragansett, RI 028.82, USA)
JOSHUA K. HOYT (HIT/WHOI Joint Program, Woods Hole, HA 02543, USA)

1. INTRODUCTION
Nahant Bay is a 15 sq km rectangular embayment beaches devoted almost exclusively to residential
located 17 km northeast of Boston on the and recreational usages. The sandy shoreline,
Hassachusetts coast (Fig.l). It does not re- totalling 8 km in length, consists of five beaches,
ceive substantial freshwater inflows and thus namely, Little Nahant (Short), Nahant-Lynn (Long),
is coastal marine, rather than estuarine, in Kings, Fishermans, and Whales (Fig. 2). Stretching
character. Its shoreline consists of a com- inland from these beaches are large year-round
bination of rocky outcroppings and white sandy residential populations housed in high-rise

.............'-"'"'-------;-------' depth at MLW

42°N

0+W
N

5
E

I 20km I t ~~
LEGEND
XXX FOULED BEACHES I 1/21Cm
71·W 70·W

FIGURE 1. Hap showing location of Nahant Bay on
the Hassachusetts coast (source: R.T. Wilce;
drawn by H.R. Halverson).
FIGURE 2. Hap of Nahant Bay showing locations
of the chronically fouled beaches, the Swampscott
Sewage Treatment Plant outfall, and the fifteen
water quality sampling stations (drawn by ~1.R.
Halverson) •
272

--buildings and homes overlooking the bay. The

two largest beaches, Nahant and Kings, typically
attract warm-weather recreational crowds of N
25,000 per day on weekends and 10,000 per day
on weekdays (C.P. Wilkinson, personal communi-
cation).
/
All of Nahant Bay's sandy beaches are fouled
sporadically throughout the year by drifting
masses of the filamentous brown alga Pilayella
littoralis and associated organisms (Wilce
et al., 1982) (Figs. 2-4). Depending on wind
and wave conditions, hundreds of tonnes
(1 tonne 10 3 kg) of this material may wash
ashore and become stranded in sludge-like mats
covering several hectares of sandy beach to a
depth of up to half a metre. In warm weather,
nauseating gases somet-imes emanate from portions REGION I : Th. beach, .Irand.d and buried alga •.
or the fouled beaches. Onshore breezes can
REGION 2: Surl- ,0lIl, nea"hare region oaluralld wilh alga•.
transport these foul~smelling gases up to a
REGION 3: Dllp walor, claud-Ii. forma"an., u...., Iwa to
kilometre inland across heavily travelled roads tin em thick at .Idim,n'·wat., interface.
A- lighl accumulation
into densly populated neighbourhoods. The most B- hMW)' accumulation
severely impacted roads and neighbourhoods are
in the City of Lynn (Fig. 2). This problem has FIGURE 3. Horizontal distribution of free-living
Nahant Bay Pilayella littoralis in vicinity of a
persisted throughout at least the present fouled beach (drawn by H.R. Halverson).
century (Anonymous, 1903; T. Bassett, personal
free until it comes into contact with toxic de-
communication), but according to the local
composition products conveyed to the surface by
population, it has considerably worsened during
tidal waters draining out of the buried deposits
the past decade (T. Bassett, personal communi-
of rotting algal biomass. The nature of the toxic
cation). The stench problem is not exclusive
chemicals is not known.
to Lynn Beach, but also plagues the northern
To eliminate the stench, the standing crop of
corner of Little Nahant (Short) Beach (Figs.2
free-living Pilayella littoralis must be
and 3).
drastically reduced in the surf zones of the
The stench emanating from these beaches is
fouled beaches. The alga exists as small feathery
caused by anaerobic wet decomposition of algal
balls averaging less than 1.5 cm in diameter
biomass buried chiefly during the annual late
(Wilce et al., 1982). During beach build-up,
winter to early spring beach build-up process.
these algal balls are commingled with the deposit-
Especially foul-smelling organic sulfides are
ed sand grains (Fig. 4). As long as hundreds of
produced by the buried algae as they decompose
tonnes of free-living ~ littoral is accumulate
(J. S. Hovis, unpublished). These sulfides are
in the surf zone, substantial quantities of algae-
introduced to the air either directly through
laden sand will be buried each year during the
gas bubbles or indirectly through tidal drain-
beach build-up cycle, and the stench will inex-
age. Algal biomass stranded on the beaches
orably follow each summer.
during summer usually remains viable and odour-

SURGE - ZONE
.-SURF-ZONE.
HI GH TI DE: Algae transported onshore
LOW TIDE: Algae left stranded onshore
FIGURE 4. Vertical distribution of free-living
Nahant Bay Pilayella littoralis along transect
perpendicular to fouled beach (source: Wilce
et al., 1982; drawn by M.R. Halverson).
The enormous standing crop of free-living
Pilayella littoralis in Nahant Bay has long been
blamed on sewage. At the turn of the century,
some people thought the algal biomass was garbage
drifting in from Boston (Anonymous, 1903) (Fig.
1). For Boston's sewage to reach Nahant Bay,
coastal currents would have to flow northward
out of Boston Harbor. However, dye studies per-
formed by the Massachusetts Division of Public
Health in 1965 showed coastal currents flow south-
ward from Cape Ann to Boston Harbor (A.
Cooperman, personal communication) (Fig.l).
Computer simulation studies also suggest coastal
currents flow southward from Nahant Bay toward
Boston Harbor (Heureux, 1980). So, sewage from
Boston or any other place south of Nahant Bay is
not likely to affect the standing crop of the
fouling alga.
273
More recently, the problem has been blamed on
effluents from the Swampscott Sewage Treatment
Plant (T. Bassett, A.F. Ferullo, personal
communication). Its outfall is the main point
source of growth-stimulating algal nutrients in
Nahant Bay. It is located at a depth of 15 m
(MSL) about half a kilometre south of Phillips
Point (Fig.2). It discharges roughly 80-235 kg
nitrogen and 4-66 kg phosphorus every day at the
ocean boundary of Nahant Bay (Massachusetts Div-
ision of Water Pollution Control Westborough
Laboratory, unpublished data). This is only a
few percent of the nutrients conveyed across the
ocean boundary into Nahant Bay on flood tides
each day. The only other potentially significant
source of nutrients is microbial and inverte-
brate remineralization of nutrients incorporated
in the algal biomass.
If effluents from the Swampscott Sewage Treat-
ment Plant could be shown to cause the exces-
sive abundance of the fouling alga in Nahant
Bay, then restrictions could be placed on them
to solve the fouling problem. However, should
they be exonerated, then a detailed scientific
investigation of the growth and reproduction
mechanisms used by the alga would be required
to find a long-term solution to the problem.
Therefore, a research programme was undertaken
to evaluate how effluents from the Swampscott
Sewage Treatment Plant affect the standing crop
of free-living Pilayella littoral is in Nahant
Bay.
2. METHODS
Levels of the macronutrients nitrogen and phos-
phorus, as well as other water quality
variables, were measured in the outfall boil, in
bay waters, and in the biomass and interalgal
waters of the Pilayella drift community. In
addition, the amounts of nitrogen and phosphorus
contributed by the treatment plant were compared
to estimates of the amounts contributed by tidal
exchange with Massachusetts Bay (Fig. 1).
Furthermore, since trace metals also affect
274
algal growth and reproduction, levels of copper,
cadmium, and lead were measured in the outfall
boil and in bay waters. Finally, dye releases
along with computer simulations were used to
determine whether substantial quantities of
nutrients could be conveyed from the outfall to
the nearshore areas where the fouling alga accumu-
lates.
2.1 Standing Crop. The clouds of free-living
Pilayella littoralis stranded on the sandy
beaches and drifting in the surf zone and shallow
sublittoral form the basis of a community of
marine life (Wilce et al., 1982). The amount of
biomass in this drift community was the only
easily accessible measure of the standing crop of
~ littoralis, and since the drift community was
almost a monoculture by weight of free-living
~ littoralis (Wilce et al., 1982), it was a good
measure. Aerial photographs were used to map
and calculate the areal coverage of both the
stranded and the floating portions of the
Pilayella drift community at low tide. The areal
densities of the biomass stranded and floating at
low tide were estimated from field samples
collected during the aerial surveys. The standing
crop was calculated by multiplying the areal den-
sities by the corresponding areal coverages.
2.2 Biomass Analyses. Dry weights were obtained
by washing algal biomass samples free of sand and
salt and then drying them at 85°C to a constant
weight. For nitrogen determinations, finely
ground dried biomass was digested and distilled
according to Standard Method 421 (Standard Methods,
1976), except 100 ml of digestion reagent was
used. Also, after the flasks were clear of gas,
each flask was rotated, and digestion was contin-
ued for 2 h. The nitrogen content of the distil-
late was determined by the acidometric method
(Standard Method 4l8D). Neutralization of the
digestion reagent required 100 ml sodium
hydroxide-sodium thiosulfate reagent. For phos-
phorus determinations, finely ground dried biomass
was digested with concentrated nitric and sulfuric
acids according to the method of O'Shaughnessy
and McDonnell (1973). After acid digestion, a
persulfate digestion was carried out according
to Standard Method 425C III, and the phosphorous
content of each sample was determined as ortho-
phosphate by the stannous chloride method
(Standard Method 425E).
2.3 Water Quality Analyses. A standard -10°C
to +50°C thermometer enclosed in a ventilated
metal tube was used to measure surface water
temperatures. Reversing thermometers attached
to a Nansen bottle were used to record bottom
water temperatures. Surface seawater samples
were collected from the stern of the research
vessel with a 10-litre polyethylene bucket.
Bottom seawater samples were obtained with either
a 3-litre teflon-lined Nansen bottle or as-litre
van Dohrn bottle. All samples were preserved
with 0.5 ml chloroform and stored at ambient
temperature in acid-washed polyethylene bottles.
With the exception of samples analyzed for total
Kjeldahl nitrogen (TKN) and total phosphorus (TP),
all samples were filtered at 40-50 cm Hg vacuum,
first through Whatman GF/C glass fibre filters
and then through Millipore 0.45W HA filters.
Before use, all filters were soaked and rinsed
with distilled water. The samples analyzed for
TKN and TP were split into two equal portions,
one filtered as indicated, the other unfiltered.
All water analyses were usually completed within
two days of collection. When this could not be
done, samples were refrigerated until analyzed.
Salinities were determined by the low precision
method 1.2 of Strickland and Parsons (1972).
The standard sodium chloride solution was after
Standard Method 209C. Nitrate was determined by
the reactive nitrate method II. 6 of Strickland
and Parsons (1972). The cadmimum-copper reduc-
tion columns were regenerated and repacked before
each batch analysis. According to method 11.7
of Strickland and Parsons (1972), nitrite was
determined in the same manner as nitrate except
the reduction steps were omitted and a 10-cm cell
 275

was used. Ammonia was determined by the distilla- lations applied the quasi-steady vertically
tion method of Jenkins (1967) and the phenate averaged coastal circulation model CAFE-l to
colorimetric method of Zadorojny et al. (1973). Nahant Bay (Wang, Connor, 1975; Heureux, 1980).
A 500 ml sample was adjusted to pH 9.5 with a This model was used to study how effluents re-
sodium carbonate buffer and a sodium hydroxide leased at the start of flood tide are affected
solution. A distillate volume of 250-400 ml was by the topography of Nahant Bay, the semidiurnal
recovered in boric acid. A boric acid solution tidal range of 2.8 m, the 0.5 cm/min N latitude
matching the boric acid concentration of the tidal tilt, and the prevailing summertime SW
sample distillates was used for each standard winds and wintertime NVJ winds (Heureux, 1980).
curve determination. A 2-cm cell was used.
3. RESULTS
Filtered and unfiltered TKN were determined
3.1 Standing Crop. At low tide on 23 June and
according to the digestion and distillation steps
16 August 1980, attempts were made to estimate
of Standard Method 421 and the phenate colorimet-
the amount of biomass in the Pilayella drift
ric method of Zadorojny et al. (1973). Orthophos-
community. The results of these surveys are
phate was determined by the stannous chloride
summarized in Tables 1 and 2. On 23 June 1980,
method (Standard Hethod 425E). Filtered TP deter-
the total standing crop of the Pilayella drift
minations followed the per sulfate digestion
community was found to be about 1,102 tonnes live
method with a 30-min autoclave step (Standard
weight. Of this, 92.9% was accumulated south of
Hethod 425C III). After digestion, filtered TP
Red Rock along Lynn Beach (Fig. 2), and another
was determined as orthophosphate. Unfiltered TP
5% was found just north of it along Kings Beach.
determination required two successive digestions:
On 16 August 1980, the total standing crop was
first, a nitric-sulfuric acid digestion (Standard
roughly 1,855 tonnes live weight. Of this, 97.1%
Hethod 425C II); then, a persulfate digestion
was found along the northern portion of Lynn Beach
(Standard Hethod 425C III). After the second
just south of Red Rock.
digestion and pH adjustment, unfiltered TP was
During both surveys, the offshore sublittoral
determined by the orthophosphate method. Copper,
accumulations covered a much greater area than
cadmium, and lead were determined by the chelator-
the stranded and surf-zone accumulations, but
concentration, solvent-extraction, and atomic
they accounted for only a small part of the total
absorption spectrophotometric methods of Lewis
standing crop (Tables 1 and 2). For example, the
(1980) .
June field survey showed the sublittoral drift
2.4 Effluent Trajectories. Dye releases together
community covered about 0.33 sq km and contained
with computer simulations were used to determine
8.1% of the total standing crop. In contrast,
effluent plume trajectories for various tidal
the beach and surf-zone coverage was only 0.043
states and wind conditions. The dye studies re-
sq km, but 91.9% of the total standing crop was
leased Rhodamine WT either directly into the out-
concentrated there at low tide. The August data
fall boil at the start of flood tide or into the
showed the sublittoral drift community covered
exit main of the treatment plant at a time that
0.56 sq km and contained 12.2% of the total stand-
would allow the dye to emerge in the outfall boil
ing crop at low tide. The remaining 87.8% was
at the start of flood tide. The flood tide trajec-
concentrated in 0.041 sq km of beach and surf zone.
tories of the dyed effluent were recorded by
During winter, much less biomass was ob-
aerial photography and on one occasion by ship-
served on the beaches and in th';ir surf zones.
board fluorometry synchronized with Loran-C read-
It could not be determined whether or not this
ings from HIT's R/V Edgerton. The computer simu-
was the entire standing crop because wintertime
 N
.TABLE 1. Distribution and standing crop of the Pi1aye11a TABLE 2. Distribution and standing crop of the Pi1aye11a
drift community in Nahant Bay at low tide on 23 drift community in Nahant Bay at low tide on 16
"cr-
June 1980 August 1980

Sq km Dry wt Wet wt* Per cent Sq km Dry wt Wet wt* Per cent
Location covered (10 3 kg) (10 3 kg) total wt Location covered (10 3 kg) (10 3 kg) total wt

Long Beach Long Beach

Stranded 0.012 7.3 285 25.9 Stranded 0.032 34.7 1360 73.1
Surf zone 0.031 18.6 727 66.0 Surf zone 0.009 7.0 270 14.1
Sub1ittora1** 0.095 0.3 12 1.0 Sub1ittora1** 0.39 4.4 170 9.3
Sub total 0.138 26.2 1024 92.9 Sub total 0.431 46.1 1800 97.1

Kings Beach Kings Beach

Stranded 0 0 0 0 Stranded 0 0 0 0
Surf zone 0 0 0 0 Surf zone 0 0 0 0
Sub1ittora1** 0.147 1.4 55 5.0 Sub1ittora1*'~ 0.09 0.3 12 0.6
Sub total 0.147 1.4 55 5.0 Sub total 0.09 0.3 12 0.6

Short Beach Short Beach

Stranded 0 0 0 0 Stranded 0 0 0 0
Surf zone 0 0 0 0 Surf zone 0 0 0 0
Sub1ittora1"'* 0.088 0.6 23 2.1 Sub1ittora1'b~ 0.08 1.1 43 2.3
Sub total 0.088 0.6 23 2.1 Sub total 0.08 1.1 43 2.3

All beaches All beaches

Stranded 0.012 7.3 285 25.9 Stranded 0.032 34.7 1360 73.1
Surf zone 0.031 18.6 727 66.0 Surf zone 0.009 7.0 270 14.7
Sub1ittora1** 0.330 2.3 90 8.1 Sub1ittora1*'~ 0.56 5.8 225 12.2
Total 0.373 28.2 1102 100.0 Total 0.601 47.5 1855 100.0

*Wet weight ~ (dry weight)/0.0256 '~Wet weight ~ (dry weight) /0.0256

*'~Offshoreof breaking waves but inshore of 6 m MLW depth **Offshore of breaking waves but inshore of 6 m MLW depth
contour contouc
 277

TABLE 3. N/P ratios (by wt) in the biomass of the Pilayella drift community and in the semvater
within and above the algal clouds

Range Nedian Mean SD n

Biomass N/P 2.68 - 13.14 6.74 7.45 2.70 44
Seawater DIN/DIP;'
Interalgal 0.09 - 19.10 1. 40 2.32 2.97 56
Extraalgal 0.76 - 18.00 2.43 3.99 4.67 23
Seawater TDN/TDP";'
Interalgal 0.91 - 22.00 4.48 5.07 3.62 54
Extraalgal 1. 94 - 15.00 4.33 4.92 3.29 14

;'DIN = Dissolved Inorganic Nitrogen ammonia-N + nitrate-N + nitrite-N

DIP = Dissolved Inorganic Phosphorus = orthophosphate-P
,HTDN Total Dissolved Nitrogen = TKN-filtered + nitrate-N + nitrite-N
TDP Total Dissolved Phosphorus TP-filtered

field conditions were too severe for the research
SCUBA divers to take surf-zone and sublittoral
samples. However, the amounts of free-living
Pilayella littoralis cast on and embedded in shore-
bound sea ice seemed to be on the order of tonnes
during mid-winter (J.A. Simpson, field notes).
Also, extensive portions of the shallow « 6 m !fUn
sublittoral remained covered by the Pilayella
drift community throughout the whole winter
(J. S. Hovis, aerial slides). It remains to be
determined whether the biomass of the drift com-
munity actually declines during winter, or is
merely conveyed offshore into the deep sublittoral
under the influence of the prevailing winter winds,
or more likely, experiences some combination of
the two.
3.2 Nitrogen and Phosphorus Content of the
Pilayella Drift Community. During the summers
of 1979 and 1980, nitrogen (N) and phosphorus (P)
levels were measured in 44 samples of the Pilayella
drift community. These samples had an average of
3
21.5 kg N/IO kg dry wt (SD = 11.9) and 2.98 kg
3
P/IO kg dry wt (SD = 1.57).
Given these biomass Nand P concentrations, the
3
roughly 28.2 x 10 kg dry wt in the drift communi-
ty on 23 June 1980 (Table 1) would have contained
about 606 kg Nand 84 kg P. Similarly, the rough-
3
ly 47.5 x 10 kg dry wt on 16 August 1980 (Table
2) would have contained about 1020 kg Nand 142 kg
P.
3.3 N/P Ratios. During the summers of 1979 and
1980, Nand P concentrations were also measured
in 56 interalgal and 23 extraalgal seawater
samples collected off Lynn Beach and Little
Nahant Beach (Fig. 2). These concentrations to-
gether with the 44 biomass Nand P concentrations
were used to calculate N/P ratios (Table 3). The
biomass had a mean N/P ratio (by wt) of 7.45.
This ratio exceeded the mean seawater DIN/DIP
and TDN/TDP ratios (by wt) found in the seawater
within (interalgal) and above (extraalgal) the
algal clouds.
3.4 Effluent Trajectory. Project divers inspect-
ed the outfall terminus and the effluent jet
issuing from it. The terminus consists of a
partially buried solid pipe, 0.5 m in diameter,
at a depth ranging from 13.4 m (MLW) to l6.4m
(Hilln. The effluent issues from a meter long
elbow inclined 45° upward from the horizontal.
The effluent jet is buoyant and surface-incident
with considerable entrainment of seawater. Dye
released into the exit main of the treatment plant
confirmed these jet characteristics.
Computer simulations as well as dye releases into
the exit main and the outfall boil indicated the
effluent plume could travel no further shoreward
on flood tide than roughly the 4 m (MLW) depth
contour before being swept offshore by the sub-
sequent ebb tide (Fig.5). This depth contour is
well offshore of the shoal nearshore region where
27S

collected 12 October 1979 (Raytheon Company, 1971,

1972, 1973; Frankel, Pearce, 1973). The anomalous-
ly low salinities reported then most likely stem
SIMULATED DEPTH - AVERAGED TRAJECTORIES OF EFFLUENTS
FROM THE SWAMPSCOTT SEWAGE TREATMENT PLANT
from a systematic error, as an impossible large
volume of freshwater would have been required to
SWAMPSCOTT
produce such low salinities uniformly throughout
the bay. Excluding these anomalous values, the
~~~=-. OUTFALL
LYNN mean salinity ranged from 30.03 ppt to 32.45 ppt.
More often than not, the salinity of the surface
\

\j
, and bottom waters of the outfall boil was less
;' \ than the corresponding values for the ocean boun-
I '
WINTER, ._ ',SUMMER
(
\
dary station and the mean. As salinity more
\ " strongly affects seawater density than temperature

"','" (Knudsen, 1903), the low salinities in the outfall

boil counteract the low temperatures there and
LEGEND tend to make the effluent plume buoyant so it
_CALM

___ NW 20 KNOTS
_____ SW 10 KNOTS
Figure 5. Comparison of the computer-simulated
effects of sustained winter and summer prevailing
winds on the depth-averaged trajectories of ef-
fluents released from the Swampscott Sewage Treat-
ment Plant at low water slack (after Heureux, 19S0;
drawn by M.R. Halverson).
most of the fouling alga was found.
3.5 Water Quality. Between 19 April 1979 and
IS August 19S0, surface and, when possible, bottom
water samples were collected monthly from the out-
fall boil and up to fourteen other locations
throughout Nahant Bay (Fig.2).
The mean water temperature for all stations
sampled excluding the boil (Station 4, Fig. 2)
ranged from a low of 1.9°C on 15 February 19S0 to
a high of lS.9°C on 31 July 19S0.
temperatures in the outfall boil were 0.4°C cooler
on the average than the corresponding values at
the ocean boundary (Station 2, Fig. 2),
averaged 0.6°C less than the corresponding mean
surface temperatures.
spreads across the water surface rather than in
a lens below it.
Because the effluent jet was surface-incident and
buoyant, only surface macronutrient levels had to
be considered to evaluate the influence of the
effluents on the standing crop of the fouling alga.
The observed values of surface TDN and TDP are
summarized in Figures 6 and 7, respectively.
These figures show that 53% of the TDN and 29% of
the TDP levels observed in the outfall boil were
above the 95% confidence interval (Cl) for all
'other surface stations sampled in Nahant Bay.
Even so, the highest TDN and TDP levels found in
the outfall boil are less than those symptomatic
of eutrophication.
Finally, the seawater copper, cadmium, and lead
concentrations found were typical of those report-
The surface
ed from areas not experiencing such a fouling
problem (Lewis, 19S0). Thus, it seemed these
trace metals were not contributing to the fouling
They also
problem.
4. DISCUSSION AND CONCLUSIONS

Salinity measurements made throughout Nahant Bay 4.1 Macronutrient Controlling the Fouling Prob-

(Table 4) were within the range of values previous- lem. The Nip ratios given in Table 3 indicate

ly reported for Massachusetts Bay and Nahant Bay, dissolved inorganic nitrogen (DIN) is, on the

with the exception of measurements made on samples average, in short supply relative to dissolved
inorganic phosphorus (DIP). The same conclusion
TABLE 4. Summary of salinity data (ppt) TABLE 5. Effluent characteristics of the Swampscott Sewa~e Treatment
Plant
Outfall
Date Depth boil Mean* SD n Flow Chlorides NH 3-N N0 3 -N TKN TP
Date (ll day) (ppt) (ppb) (ppb) (ppb) (ppb)
19 Apr 79 S: 30.13 30.37 0.17 10
B: 30.30 30.41 0.09 3 26-27 Ju1 76 7.8x10 6 0.060 X X X 8,500
17 May 79 S: 30.82 31.14 0.45 10 27-28 Ju1 76 7.9x10 6 0.092 19,000 400 X X
B: 31.17 31.40 0.50 4 100 X X
28-29 Ju1 76 7.0x10 6 0.050 16,000
13 Jun 79 S: 30.84 30.74 0.20 13 _5.2x10 6 8,600 ~O X 3,000
21 Sep 77 X
B: 30.49 30.92 0.15 7
ll-12 Sep 78 6.4x10 6 X 19,000 700 X 650
12 Ju1 79 S: 30.01 30.65 0.38 12
B: 30.16 30.34 0.51 6 12-13 Sep 78 6.3x10 6 X 300 18,000 X 10,000
14 Aug 79 S: 30.12 31.03 0.78 14 13-14 Sep 78 6.4x10 6 X 100 16,000 X 5,600
B: 30.87 30.98 1.20 8 11,000 300 17,000 750
22-23 Apr 80 4.6x10 6 X
14 Sep 79 s: 30.18 30.52 0.30 12
B: 30.18 30.71 0.22 3
X No value reported
12 Oct 79 s: 26.47 27.80 1.19 10
B: X X X 0
16 Nov 79 S: 30.46 30.54 0.33 11
B: X 30.46 X 1
15 Dec 79 S: 30.66 30.66 0.28 11
B: 30.98 30.55 0.10 3
TABLE 6. Amounts of nitrogen and phosphorus discharged per day by
14 Jan 80 S: 31.15 31.31 0.36 11 the Swampscott Sewage Treatment Plant
B: X 31.86 X 1
15 Feb 80 s: 31.18 31.86 0.22 11
B: 32.01 31.90 0.08 4 NH -N NO -N TKN TP
Date (kg (day) (kgrday) (kg/day) (kg/day)
20 Mar 80 S: 32.03 32.10 0.41 11
B: 31.86 32.45 0.32 5 X X 66
26-27 Ju1 76 X
17 Apr 80 S: 31.11 31.11 0.37 13 3 X X
27-28 Ju1 76 150
B: 31.51 31.25 0.47 6
28-29 Ju1 76 ll2 -1 X X
15 May 80 S: 31.04 31.47 0.46 12
B: 32.61 31.69 0.38 7 21 Sep 77 131 -0 X 46
9 Jun 80 S: 30.11 30.60 0.38 14 ll-12 Sep 78 122 5 X 4
B: 30.27 30.69 0.35 2 12-13 Sep 78 2 ll3 X 63
31 Ju1 80 S: 29.75 31.33 0.53 14 102 X 36
13-14 Sep 78 ·1
B: 31.17 31.21 0.59 8
22-23 Apr 80 51 -1 78.2 4
18 Aug 80 S: 29.65 30.09 0.45 14
B: 30.13 30.03 0.29 8
X No value reported
* Calculated for all stations sampled except
the outfall boil
N
....,
X No value reported
'"
280

300
TOTAL DISSOLVED NITROGEN
X X
NAHANT BAY SURFACE WATERS

250 LEGEND

X OUTFALL BOI L

0 LYNN BEACH
200
95% CI (BOIL
TDN DATA EXCLUDED) X

(ppb) X
150
x x
x 0

100

1 r ~
t tt r
+ t t
x
50 X + x

~
o APR MAY JUN JUL AUG SEP OCT NOV DEC JAN FEB MAR APR MAY
19 17 13 12 14 14 12 16 15 14 15 20 17 15

• 1979 1980
FIGURE 6. Comparison of TDN values measured in the Swampscott Sewage Treatment
Plant outfall boil and at Lynn Beach, and their relation to the 95% confidence
interval calculated for all surface TDN data excluding those for the outfall boil
(drawn by M.R. Halverson).

150 I I I
TOTAL DISSOLVED PHOSPHORUS
NAHANT BAY SURFACE WATERS
125 - LEGEND -
X OUTFALL BOIL

0 LYNN BEACH

100 - I 95% CI (BOIL

DATA EXCLUDED
-
0

TDP 75 c-
(ppb) ·X

50 e-

t t
X

tf
X

25 X

q>
1r ~ X

6
+
t ~

o APR MAY JUN JUL AUG SEP

I
OCT NOV DEC JAN FEB
I
MAR APR MAY JUN
I
JUL AUG
19 17 13 12 14 14 12 16 15 14 15 20 17 15 9 31 18
--------------1979-------------------------1980------------
FIGURE 7. Comparison of TDP values measured in the Swampscott Sewage Treatment
Plant outfall boil and at Lynn Beach, and their relation to the 95% confidence
interval calculated for all surface TDP data excluding those for the outfall
boil (drawn by M.R. Halverson).

holds for total dissolved nitrogen (TDN) and total
dissolved phosphorus (TDP). In other words, based
on an average biomass Nip ratio of 7.45 by weight,
the fouling alga, in producing new biomass,
would tend to consume all the dissolved nitrogen
in the interalgal and extraalgal seawaters before
depleting the dissolved phosphorus. Thus, nitro-
gen seems to be the macronutrient that most likely
limits the amount of biomass produced by the
fouling alga in the nearshore areas of Nahant
Bay. Consequently, as schematized in Fig. 8, the
availability of dissolved nitrogen nutrients
appears to drive the algal fouling problem in
terms of both the amount of algal biomass produced
and the stench rising from its anaerobic decay in
the beach sands.
4.2 Influence of Sewage Effluents. The Swamp-
scott Sewage Treatment Plant outfall is the
largest point source of algal nutrients in Nahant
Bay. The original sewage disposal facilities,
built during 1902-1903 (Clinton Bogert Associates,
1968), post-date the algal fouling problem
(Anonymous, 1903). This fact alone indicates
Swampscott Sewage Treatment Plant effluents did
not initially cause the problem.
The effluent consists almost entirely of domestic
wastes (Massachusetts Division of Water Pollution
Control Westborough Laboratory, unpublished data).
Since March 1974, these effluents have received
primary treatment. Flow, salt, nitrogen, and
phosphorus levels observed in the effluent since
1976 are given in Table 5 (Massachusetts Division
of Water Pollution Control Westborough Laboratory,
unpublished data). These effluent characteristics
are normal for primary-treated domestic sewage
from a community the size of Swampscott, with the
exception of the apparent nitrification occurring
in September 197b. The low ammonia and high nitrate
levels reported then are puzzling because the
plant is a through-flow system with a hydraulic
residence time too short to allow detectable
oxidation of ammonia to nitrate (Clinton Bogert
Associates, 1968).
281
NAHANT BAY ECOSYSTEM
FIGURE 8. Conjectured coupling of the nitrogen
and sulfur cycles in the Nahant Bay ecosystem
through the biomass of the fouling alga. Note
the availability of dissolved nitrogen nutrients
driving the production of nauseating volatile
sulfides (drawn by M.R. Halverson).
Daily discharge rates for nitrogen and phosphorus
(Table 6) can be estimated from the data given
in Table 5. If NH 3-N and TKN are assumed to ex-'-
ist in the same ratio as observed on 22-23 April
1980, then the largest daily discharge of nitro-
gen would have been 235 kg on 27-28 July 1976.
The largest daily discharge of phosphorus report-
ed was 66 kg on 26-27 July 1976.
The dilution experienced by the effluent as it
rises from the outfall terminus to the water
surface may be estimated from the salinity
values given in Table 4. The discharged effluent
contains a negligible amount of salt relative to
seawater (cf. Tables 4 and 5), so the percentage
of sewage effluent in the boil water may be
simply estimated by the following equation:
% sewage effluent = 100 (Sm - Sb)iSm (1)
where Sm - mean of all surface salinity measure-
ments in Nahant Bay excluding that of
the boil (Table 4)
Sb = outfall boil surface salinity (Table 4).
The percentages of sewage effluent in the outfall
boil, as estimated from Equation 1, are listed
in Table 7. The sewage content of the boil
water would appear to have ranged from virtually
undetectable (0%) to 5%, thereby indicating that
the effluent was substantially diluted by the
282

:time it reached the surface and began to spread. TABLE 7. Dilution of sewage effluent in outfall
boil.
The amount of dilution occurring in the rising
Percent Dilution
plume may be estimated by taking the reciprocal Date Sewage Factor
of the decimal fraction (as opposed to percentage) 19 Apr 79 0.8 125
of sewage effluent in the outfall boil, namely, 17 Hay 79 1.0 100
dilution factor (2) 13 Jun 79 0.0
12 Jul 79 2.1 48
100/(% sewage effluent).
14 Aug 79 2.9 34
The dilution factors thus estimated are also 14 Sep 79 1.1 91
listed in Table 7. The effluent seems to have 12 Oct 79 4.8 21
experienced dilutions ranging from 1:20 up to 16 Nov 79 0.3 333
1: 00 , with 1:100 being the median dilution experi- 15 Dec 79 0.0 00

enced. 14 Jan 80 0.5 200

Combined with an expected lateral dilution of 15 Feb 80 2.1 48
1:50 (Clinton Bogert Associates, 1968), the 20 Mar 80 0.2 500
dilutions in the outfall boil (Table 7) would 17 Apr 80 0.2 500
produce total dilutions at Lynn Beach ranging 15 May 80 0.7 143
from at least 1:1000 to 1: 00 , with a median of 9 Jun 80 1.6 63
1:5000. As a consequence, the sewage plume may 31 Jul 80 5.0 20
be considered virtually dispersed long before it 18 Aug 80 1.5 67
reaches the nearshore area off Lynn Beach where
alga grows seems well supported by the nutrient
the fouling alga concentrates.
data.
To check the conclusions derived from the salinity
As well-substantiated as this conclusion may appear,
data, the expected lateral dilution of 1:50 was
it is nonetheless tentative. As noted earlier,
applied to the· TDN and TDP concentrations attribu-
53% of the TDN and 29% of TDP levels observed in
table to sewage effluents in the outfall boil,
the boil were above the 95% confidence interval
and the results were compared with the correspond-
for Nahant Bay waters (Figs. 6 and 7). After
ing TDN and TDP concentrations measured at Lynn
subtracting out amounts supposedly attributable to
Beach (Figs,. 6 and 7 ). For example, on the date
sewage (% sewage, Table 7), the corrected TDN and
having the lowest dilution (31 July 1980), 5%
TDP values for the outfall boil still fall above
(or 3 and 0.6ppb, respectively) of the TDN and
the 95% CI. This throws some doubt on the accura-
TDP in the outfall boil might be attributed to
cy of the dilution factors reported in Table 7,
the sewage effluent (Table 7, Figs. 6 and 7).
but not on the overall conclusion of negligible
After a 1:50 dilution of the se~age effluent in
nutrient enrichment at Lynn Beach. If the excess-
the boil, only 0.06 ppb (or 0.05%) of the 120 ppb
es of outfall boil TDN and TDP over the upper
TDN and 0.012 ppb (or 0.05%) of the 25 ppb TDP
limit of the 95% CI are used instead of the esti-
observed that day at Lynn Beach can, in turn, be
mated sewage content, the sewage effluents would
attributed to the Swampscott Sewage Treatment
be expected to produce at most 1-2% enrichments
Plant. Similar analyses of the data obtained on
at Lynn Beach. Such enrichments would still be
the other 16 sampling days produced similar
negligible.
results. Thus, the conclusion that effluents
Finally, the estimates of nutrient enrichment at
from the Swampscott Sewage Treatment Plant negli-
Lynn Beach have all been based on the assumption
gibly enrich the nearshore waters where the foul-

that the effluent plume could reach the beach.
However, the results of both the dye studies and
the computer simulations indicated the plume
travels no further shoreward than roughly the
4 m (MLW) depth contour. In other words, tidal
and wind-induced currents in Nahant Bay do not
appear to convey sewage effluents into the near-
shore area where most of the fouling alga accumu-
lates.
In summary, no evidence has been found that
supports the conjecture that Swampscott Sewage
Treatment Plant effluents either originally
caused the algal fouling problem or presently
affect it in any way. The fouling problem exist-
ed years before the first outfall was built; so,
sewage effluents released from it could not have
originally caused the problem. Present-day
water quality measurements and plume trajectory
studies have all consistently shown the effluent
plume to be virtually undetectable long before it
reaches the nearshore areas where the fouling
alga is found. Moreover, the concentrations of
nitrogen and phosphorus throughout the bay are
far less than those symptomatic of eutrophication.
So, it is also unlikely that S"mmpscott's sewage
effluents have contributed to the problem since
the outfall was built. As a consequence, clues
to the causes of the excessive amount of algal
biomass in Nahant Bay must be sought by examining
the growth and reproductive peculiarities of the
fouling alga itself. These peculiarities are
discussed in a following paper (Wilce, Quinlan,
1983) .
4.3 Influence of Tidal Exchange with Nassachu-
setts Bay. Based on a tidal prism of 4.5 x 10 10
litres (15 sq km surface area, 3 m tidal range)
and the mean values (respectively, 96 and 22 ppb)
of the TDN and TDP observed at the ocean boundary
of Nahant Bay (Station 2, Fig. 2), an average of
roughly 4,320 kg Nand 990 kg P might be convey-
ed into Nahant Bay on a flood tide. These values
are, respectively, 74 and 60 times the largest
discharges of Nand P expected during a flood
283
tide. Hence, tidal exchange appears to dominate
the,'Swampscott Sewage Treatment Plant as a source
of algal nutrients at the ocean boundary of Nahant
Bay.
4.4 Influence of In-Situ Nutrient Regeneration.
In addition to tidal exchange, the other likely
major sources of algal nutrients are microbial
and invertebrate regeneration of the nutrients in-
corpora ted in the biomass of the fouling alga.
Time and funds did not permit a quantitative ana-
lysis of the contributions of biologically mediated
regeneration processes. However, the following
conjectures and future research priorities have
been generated.
Nicrobial decomposition and invertebrate scaveng-
ing of the algal biomass stranded on the beaches
and buried in their sands probably are the domi-
nant sources of nutrients in the surf zone where
the fouling alga accumulates. In this regard, the
stranded and buried deposits of algal biomass may
function in the Nahant Bay ecosystem as leaf litter
does in a forest ecosystem by sequestering and re-
cycling nutrients. Given the potential signifi-
cance of this source of nutrients, the nitrogen
balance of the beach sands merits considerable
future investigation.
In the algal clouds themselves, several species
of zooplankton, mostly amphipods and copepods,
have been observed to feed on free-living Pilayella
littoralis. Upon digestion of the ingested plant
material, the zooplankton can be expected to ex-
crete nitrogen in the form of urea and ammonia,
which may be quickly assimilated by the remaining
viable plant material to regenerate lost biomass.
Thus, the quantitative contributions of zooplank-
ton predation and nutrient regeneration deserve
more investigation.
Finally, it is conjectured that the nearshore
beach and surf-zone areas of Nahant Bay function
as a self-sustaining nutrient-recycling ecosystem
whose productivity is limited by the availability
of the macronutrient nitrogen, and furthermore,
the microbial and invertebrate populations in
284

these areas probably play major roles in control- REFERENCES

ling nitrogen supply rates. Future research ef- Anonymous (1903) Deposit from the sea nauseates
residents, Lynn, MA, Lynn Daily Evening Item,
forts should quantitatively determine the flux
19 August 1903.
rates for nitrogen nutrients across the interti- Clinton Bogert Associates (1968) Report on
sewage treatment facilities: Town of Swampscott,
dal sand/water interface, and how water motion,
Hassachusetts, Fort Lee, NJ.
beach contours, sand porosity, bioturbation, and Frankel SL and Pearce BR (1973) Determination
of water quality parameters in the Massachusetts
seasonal beach erosion influence these rates.
Bay (1970-1973), report no. MITSG74-8, Cambridge,
MA, Sea Grant College Program, M.LT.
ACKNmvLEDGEMENTS. The work reported in this pa-
Heureux AJ (1980) An analysis of the hydrodyna-
per was carried out while all three authors were
mic circulation in Nahant Bay, M.S. thesis,
associated with the Department of Mechanical En-
Cambridge, MA, Department of Mechanical Engineer-
gineering, M.LT., Cambridge, MA, USA. lIuch help
ing, M. LT.
was received from student researchers at M.I.T.
Jenkins D (1967) Estuarine water analysis, J.
Jeff Hovis characterized the chemicals causing
>Vater Poll. Control Fed. 39, 159-180.
the stench associated with the decomposing algal
Knudsen M, ed. (1901) Hydrographical tables,
biomass. Jeff, alternating with Matt Thompson,
London, Hi11iams and Norgate.
also served as pilot and photographer during the
Lewis T (1980) Analysis of trace metals in the
aerial surveys. Alan Heureux carried out the com-
waters of Nahant Bay, its beaches, and the bloom
puter simulations of the trajectories of Swamp-
alga, M.S. thesis, Boston, MA, Department of
scott Sewage Treatment Plant effluents. Bruce
Chemistry, Northeastern University.
Kirch, as well as Alan, conducted field tests of
O'Shaughnessy JC and McDonne1 AJ (1973) Criteria
the computer results. Anna Piccolo, administra-
for estimating limiting nutrients in natural
tive secretary to AVQ, took care of logistic and
streams, publ. no. 75, University Park, PA, Insti-
personnel problems. The research staff made ex-
tute for Research on Land and >Vater Resources,
tensive use of the boat and laboratory facilities
Pennsylvania State University.
at Northeastern University's Edwards Marine
Raytheon Company (1971) Lynn Harbor-Nahant Bay
Science Institute on East Point, Nahant Bay. All
ecological survey: July-November 1970, New
the biomass and water quality analyses were con-
London, CT, Raytheon Environmental Research Labor-
ducted by faculty and students from Northeastern's
atory.
Boston campus. Jim O'Shaughnessy and Fred Blanc
Raytheon Company (1972) Lynn Harbor-Nahant Bay
(Department of Civil Engineering) supervised the
ecological survey: December 1970-December 1971,
biomass, salinity, and macronutrient analyses.
New London, CT, Raytheon Environmental Research
Tom Copeland (Department of Chemistry) supervised
Laboratory.
the trace metals analyses. John Simpson (Depart-
Raytheon Company (1973) Lynn Harbor-Nahant Bay
ment of Geology) served as boatman, gear tender,
ecological survey: January-December 1972, New
and laboratory technician.
London, C~ Raytheon Environmental Research Labor-
A1 Cooperman of the vJestborough Laboratory of the
atory.
Massachusetts Division of >Vater Pollution Control
Standard Methods for the Examination of >Vater
(Department of Environmental Quality Engineering)
and >vastewater (1976) 14th edn. IJashington, DC,
provided personnel, equipment, and data that were
American Public Health Association.
critically needed to evaluate the influence of
Strickland JDH and Parsons TR (1972) A practi-
effluents from the Swampscott Sewage Treatment
cal handbook of seawater analysis, bulletin 167,
Plant.
2nd edn. Ottawa, Fisheries Research Board of
The Rockefeller Foundation supported the prepara-
Canada.
tion of the original research proposal through a
Hang JD and Connor JJ (1975) Mathematical model-
Fellowship Grant in Environmental Affairs awarded
ing of near coastal circulation, technical report
to AVQ. The research project itself was support-
no. 200, Cambridge, MA, Ralph M. Parsons Labora-
ed with funds provided by the U.S. Department of
tory for Water Resources and Hydrodynamics, M.I.T.
Commerce through the 11. I. T. Sea Grant College
Vlilce RT and Ouin1an AV (1983) Fouling of the
Program (Grant No. NA 79 AA-D-00101), by the
sandy beaches of Nahant Bay (Massachusetts, USA)
Commonwealth of Massachusetts through the Metro-
by an abnormal free-living form of the macroa1ga
politan District Commission, by the Henry L. and
Pi1aye11a 1ittora1is (Phaeophyta). II. Popula-
Grace Doherty Charitable Foundation through the
tion characteristics, in this book.
Henry L. Doherty Professorship in Ocean Utiliza-
>vi1ce RT, Schneider C>v, Quinlan AV and Vanden
tion awarded to AVQ, and by a grant from the
Bosch K (1982) The life history and morphology
DuPont Fund of the M.LT. Department of Mechani-
of free-living Pi1aye11a 1ittora1is (L.) Kje11m.
cal Engineering.
(Ectocarpaceae, Ectocarpa1es) in Nahant Bay,
This paper was prepared for publication by Leslie
Massachusetts, Phyco1ogia 21, 336-354.
Regan (M.I.T.) and Bonnie Kissell (Duke).
 285

FOULING OF THE SANDY BEACHES OF NAHANT BAY (MASSACHUSETTS, USA) BY AN ABNORMAL FREE-LIVING FORM OF
THE MACROALGA PILAYELLA LITTORALIS (L.) KJELLM. (PHAEOPHYTA). II. POPULATION CHARACTERISTICS

ROBERT T. WILCE (University of Massachusetts, Amherst, MA 01003, USA)

ALICIAN V. QUINLAN (Duke University, Durham, NC 27706, USA)

1. INTRODUCTION reported that some people thought it was sewage

Every summer this century the sandy beaches and conveyed by coastal currents from Boston into
nearshore waters of Nahant Bay (Lat. 42 0 27' N, Nahant Bay.
Long. 70 0 55' W) have been infested by drifting Today the identity and much of the biology of
masses of a filamentous brown alga (Fig. 1) the fouling alga has been determined (Wilce et
(Anonymous, 1903; Wilce et al., 1982; Quinlan al., 1982). However, its origin and the causes
et al., 1983). The earliest description of this of the infestation still remain unknown. Sewage
infestation found so far is a local newspaper effluents have recently been ruled out both as
report published on 19 August 1903 (Anonymous, origin and as cause (Quinlan et al., 1983).
1903). The article aired "complaints .... brought
SWAMPSCOTT
to the Board of Health regarding the deposit of
brown colored matter upon parts of Lynn Beach."
This material decayed to produce a "stench, sour
and disgusting." The problem substance was con-
sidered "an annual summer nuisance" of uncertain
identity, origin, and cause. It was blown onshore
by prolonged southerly and southeasterly winds, 20m depth at MLW

and offshore by westerly breezes. Then, as today,

the most severely infested beach area extended
from Red Rock southward a third of a kilometer
•
along Nahant-Lynn (Long) Beach (Fig. 1). The
same material was also often found in the surf
zone at both ends of Kings Beach and along the EGG
ROCK~_
1{j;!
beach in a "thick brown porridge-like belt at the
surf edge, which the bather must swish through in
order to gain the clear water further out, which
is sometimes a long way to reach, especially at
high tide, often obliging one to get into deep
water." The reporter thought it might be the
"off-shoot of some low growing sea plant, which
is loosened by the action of the water and car-
ried in by the wind," but realized "this may not
FIGURE 1. Map of Nahant Bay showing shore and
[have been] the right definition, and the local
sublittoral collection stations.
naturalists [were] sounded in vain." He also
286

Instead, newly discovered (Wilce et al., 1982) features is distinctive and intricately tied to
peculiarities in how the fouling alga grows and the development of the free-living Pilayella
reproduces offer important clues to its origin population and, ultimately, to the massive
and to the causes of the infestation. These pe- fouling problem, not one seemed sufficiently
culiarities and the clues derived from them are distinctive to set the fouling alga taxonomically
described and discussed in this paper. apart as a new variety or a new species.
3.2.1. Standing Crop - The standing crop of at-
2. METHODS
tached Pilayella littoralis in Nahant Bay would
With two exceptions, the methods used in this
normally be expected to measure in tens of kilo-
study have been previously described (Wilce et
grams live weight, with virtually all of this
al., 1982). First, the measurements reported
biomass confined to the rocky intertidal as epi-
earlier (Wilce et al., 1982) for cell length/width
phytes on fucoids or as epiliths (Chock, Mathieson,
ratios of both branches and axial regions were not
based on seasonally and spatially identified col-
unpubl. mscr.). The standing crop of attached X.
littoralis was the expected size. However, the
lections. Measurements have now been made on
selected regions of three plants collected monthly
total standing crop of 1. littoralis in Nahant Bay
was measured in hundreds of tonnes (~ 10 3 kg) live
from each station over 13 months. These regions
weight (Quinlan et al., 1983). The excess stand-
were located three cells distal from branch initi-
ing crop consisted of the subtidal free-living
ation and three cells basipetal from a branch tip.
ball form that fouls the nearshore waters and
Width, length and frequency of multiseriate regions
sandy beaches of the bay. The vast difference
were also determined. Second, unialgal clonal
between the standing crops of the two forms sug-
cultures of free-living Nahant Bay Pilayella
gested that the free-living form has a vastly
littoralis were used to determine the effect of
superior ability to exploit the nearshore habitats
light intensity on growth and ball structure.
of Nahant Bay.
Innocula consisted of tip segments not more than
3.2.2. Perennating Population - Typical attached
a few mm long. The cultures were grown at lSoC
plants of Pilayella littoralis in the shore
+ 2 0 C and at five light intensities (2S30, lSOO,
regions of Nahant Bay, whether epiphytic or epi-
11SO, 800, 6S0 lux) in 0.4S ~m filtered and heat-
lithic, are late winter-spring annuals with a
sterilized sea water enriched with 2% (by volume)
short lifespan of not more than 4-S months (Wilce
Alga-Gro nutrient solution (Carolina Biological
et al., 1982). Temperature and light intensity
Supply Co.).
may be important in the control of lifespan.
3. RESULTS
Sublittoral attached populations persist longer
3.1. IDENTITY OF THE FOULING ALGA
into summer than do their intertidal counter-
The brown matter fouling the beaches of Nahant Bay
parts, but they, too, eventually diminish owing
was found to be virtually a monoculture (by weight)
to high light and temperature levels of summer.
of a previously unreported free-living ball form
By contrast, the free-living populations of
of the normally attached filamentous brown alga
Pilayella littoralis persisted year-round, i.e.,
Pilayella littoralis (L.) Kjellm. (Ectocarpaceae,
perennated, throughout the sublittoral of Nahant
Ectocarpales) (Wilce et al., 1982).
Bay. Viable free-living plants were always found
3.2. PECULIAR POPULATION CHARACTERISTICS
in collections obtained weekly from the sandy
Seven population characteristics distinguished the
beaches and nearshore waters during 1979-1980
free-living fouling alga from the typical attached
and monthly from selected sublittoral sites
form of the species. Although each of these
(3-20m MLW) with the aid of SCUBA between June

1978 and July 1979 (Fig. 1). Aerial surveys
during 1979 and 1980 also showed an extensive
year-round abundance of the Pilayella drift
community over the sandy sublittoral bottom in-
shore of the 6m (MLW) depth contour (Fig. 1)
(J. S. Hovis, aerial slides).
3.2.3. Planktonic/Free-living Life History -
Populations of unattached Pilayella littoralis
are the most common filamentous algae in the
northern areas of the North Atlantic (R. T. Wilce,
pers. obs.). Nonetheless, f. littoralis has only
occasionally been cited as a member of loose-lying
and free-living populations in the sublittoral
(Kjellman, 1883; Rosenvinge, 1898; Jonsson, 1904).
Accompanying comments in these previous investi-
gations are generally terse, e.g., unattached,
drift on the bottom, or merely, loose-lying.
The individual plants in these drift populations
are largely vegetative and invariably possess the
basal/distal polarity indicative of plants ini-
tially from attached populations.
Large loose-lying populations of Pilayella
littoralis were also found in several Massa-
'chusetts coastal areas (e.g., the Cape Cod
National Seashore, Revere Beach, Nantasket
Beach). In each instance, the plants were
clearly derived from attached populations with
features characteristic of attached plants.
Although their longevity in the water column
was unknown, the period of drift was sufficient
to begin morphological change: (1) strict basi-
petal-distal organization was lost or distorted;
(2) the plants frequently had fragmented; and
(3) short, spine-like branches were commonly
present. Fragmentation of laterals and the
apparent inability to initiate numerous new
branch initials resulted in the gradual break-
down of the entire filamentous axis. Despite
occasional new growth on these plant fragments,
their presence in the water column seemed on the
whole to be a short-lived summertime/autumn phe-
nomenon.
The free-living forms, whether the commonplace
287
loose-lying Pilayella populations of the north
or those from the Massachusetts coast, are dis-
tinct from the ball form of f. littoralis from
Nahant Bay. The differences between the detached
forms and the ball form were dramatic and equally
as distinctive as the ball form was when compared
to attached plants of this species.
Most particularly, field and laboratory data sug-
gested that the life cycle of the free-living
Nahant Bay ball form of Pilayella is completely
independent of attachment to any form of solid
substratum thereby making it truly planktonic.
Thousands of field specimens were examined and
no sign of a holdfast was ever seen, although
rhizoids emerged from wounded cells and from
peripheral cells of the central axes. These
rhizoids appeared similar to the horizontal
rhizoids of the attachment organ of X. littoralis
except that they invested the main axes and did
not attach the Pilayella balls to the substratum.
Similar structures were common in old attached
plants, but they, too, rarely functioned in sup-
porting or contributing to the holdfast.
Furthermore, two unialgal isolates from the
free-living population were maintained in
culture for 2 1/2 and 1 1/2 years respectively
without any portion of the plants adhering to
the walls of the culture vessels. When cultured
under identical conditions, germlings of unialgal
isolates of attached X. littoralis typically ad-
hered within a short time to the culture vessel
walls and developed holdfasts consisting of an
extensive system of prostrate filaments. There-
fore, the lack of any attachment capability in
the Nahant Bay ball form is one character that
distinguishes it from the other reported free-
living forms of this species, which are all
attached during some portion of their life cycle.
3.2.4. Non-polar Ball Shape - As a result of
their planktonic character, free-living Pilayella
plants rotate freely in respon,se to water move-
ment. Thus, all lateral filaments have an equal
probability of receiving light and nutrients for
288
growth, and the plant therefore assumes a non-
polar ball, or aegragropiloid, form (Fig. 2)
(Wilce et al., 1982). The laterals during the
early development of the ball are short, stiff
in appearance and arranged roughly radially on
the central axes. Only the lateral branches of
the Nahant Bay Pilayella ball have polarity; the
main axes do not have acropetal/basipetal
FIGURE 2. Nahant Bay free-living Pilayella
littoralis. (Source: Wilce et al., 1982).
Lower right: Photograph of mounted herbarium
specimens of summer balls; collected 3 August
1979. These specimens have the approximate
maximum diameter of the Pilayella ball. !££
~: Camera lucida drawing of an entire
winter ball; collected 1 December 1979. Note
irregular branching and the broad multiseriate
central axis. The diameter of the central axis
was appreciably increased by the green algal
epiendophyte Endocladia wittrockii which
commonly ensheaths old axes.
organization. Branches from the main axes are
numerous and frequently initiated from adjacent
cells. They are mostly of equal length and have
a similar degree of reb ranching , all of which
gives the impression of a small puff of plant
tissue. The average diameter of the free-living
Nahant Bay plants ranges from a little less than
1 mm in winter to not quite 1.5 cm in summer.
The degree of branching present varies seasonally,
but for most of the year the plants are distinctly
spherical in outline.
The ball form generally disappeared in old, non-
aerated, quiescent cultures of these plants. The
ball form, however, was readily reformed from
excised branch tips of old culture material after
being placed in fresh medium under low light and
temperature conditions characteristic of the
plant's sublittoral habitat. By contrast, only
linear structural organization developed from
excised branch tips from attached plants of this
species. Therefore, on the basis of these ex-
perimental results, evidence exists that the
ball form may have a genetic component as well
as being environmentally influenced.
3.2.5. Apparent Infertility - One of the char-
acteristic features of truly free-living forms
of algae is the lack of reproductive organ
development. The general lack of both gamete
and spore production, or at best the very low
frequency of their occurrence, in free-living
algae has been attributed to many aspects of
the plants' environment and its attendant form
(Gibb, 1957; Coleman, 1961; Dring, 1974; Bird,
McLachlan, 1976; Oliveria, Fletcher, 1980;
Terry, Moss, 1980). Few of these hypotheses
have been tested experimentally with satisfac-
tory results. Norton and Mathieson review the
literature of the biology of free-living marine
and freshwater algae and discuss the relevant
problems concerning their asexual character and
the solutions that have been proposed (T. A.
Norton, A. C. Mathieson, unpubl. mscr.).
 289

In field specimens of free-living Nahant Bay Pilayella littoralis. Vegetative periclinal cell
plants, reproductive organs characteristic of divisions occur frequently in the central axial
those formed on attached plants of this species cells (Wilce et al., 1982). As a consequence of
have been seen, but only r~rely (Wilce et al., repeated periclinal divisions at a given locus,
1982). The discovery of either unilocular or the central axes of each ball may contain several
plurilocular reproductive organs in a given many-ceIl-wide, or multiseriate, chain segments
sample of free-living Pilayella is thus entirely (Fig. 3). These segments may be up to 50 or more
fortuitous. Discharge of motile reproductive
cells has not been observed, although a few
mature and partially discharged reproductive
organs have been discovered in preserved field
material (Wilce et al., 1982).
In contrast, plants from the attached Pilayella
population in Nahant Bay and those from unialgal
cultures of attached plants regularly produced
reproductive organs, often in large numbers.
When unialgal isolates of the attached form and
the free-living form were grown under identical
laboratory conditions, the attached strain de-
veloped reproductive organs, but the free-living
form did not. Hence, the absence of reproductive
organs in free-living Pilayella littoralis seemed
to be more than a phenotypic response to its
free-living state and habitat conditions. This
provides additional evidence for genetic differ-
entiation between the two populations.
3.2.6. Multiseriate Central ~ - In normal
attached Pilayella littoralis; all plant axes
are almost entirely filamentous, i.e., a chain
of cells, with intercalary cell divisions largely
responsible for increases in filament length
(Knight, 1923; Rosenvinge, Lund, 1941; Russell,
1963; West, 1967; Wilce et al., 1982). The cell
terminating a filament also has the potential for
vegetative cell division but only during those FIGURE 3. Nahant Bay free-living Pilayella
littoralis. Camera lucida drawings of a winter
portions of the year when the filament is not
ball; collected 1 December 1979. (Source: Wilce
terminated by a pseudohair (Wilce et al., 1982). et al., 1982). Lower Right: Entire winter ball,
scraggly in character and lacking basal-distal
Periclinal cell divisions, i.e., divisions along
polarity. ~ Center: Fragment of winter ball
the long axis of a cell, occur rarely in the enlarged c twenty times. Note multiseriate
condition is irregular in its occurrence and
cells of attached plants and then only in nodal
limited to the central axis, also the single
cells of axial filaments. early infection of a young P. littoralis cell
by Eurychasma dicksonii.
Quite the opposite is true concerning the struc-
ture of the central axes of the free-living
290
cells long and seven or more cells across, with
each multiseriate segment only a normal cell-width
wide. Multiseriate portions of axial filaments
are separated, sausage-like, by filamentous seg-
ments of variable length, sometimes with as few
as several cells.
Lateral branches of free-living balls remain
filamentous except occasionally at their point
of insertion with the axial rows; at times the
most basal cell of a branch may show several
periclinal divisions (Wilce et al., 1982).
3.2.7. Reproduction - Attached Pilayella
littoralis reproduces only by motile cell ger-
mination and development on some form of solid
substratum (Knight, 1923; Rosenvinge, Lund,
1941; West, 1967). Asexual zoospores liberated
from both unilocular and plurilocular repro-
ductive organs are the principal means of
attached g. littoralis population increase
(Dammann, 1930; West, 1967; Nygren, 1975).
Sexual stages have been reported, but doubt
remains to both their occurrence and their
frequency (West, 1967; Nygren, 1975).
In contrast, the free-living ball form of
Pilayella littoralis appears to reproduce
primarily by vegetative rather than asexual
or sexual methods (Wilce et al., 1982). Al-
though some motile cells are formed they are
likely to be insignificant in the development
of the Nahant Bay free-living population of
this species since they are so few in number
and solid substratum is absent in the area
where they are liberated. Instead, vegetative
propagules are abundantly formed by two novel
processes involving plant fragmentation (Wilce
et al., 1982). Each propagule can develop into
a new free-living ball at an explosive rate
under favorable habitat conditions.
One vegetative means of ball population increase
exploits the fragmentation of the central axes
into a number of multiseriate portions (Wilce
et al., 1982). In late summer, uniseriate
branches and filamentous regions of the central
axes of the free-living ball senesce concurrent
with increased summer water temperatures. But
the more heat resistant multiseriate portions
persist to form the overwintering seed stock of
the population. Subsequent lateral branch de-
velopment from the multiseriate axial segments
is favored by the low light/temperature and
relatively nutrient rich conditions of the surf
and shallow sublittoral zones in mid-winter.
Numerous branches develop then from each multi-
seriate axial segment, at times from adjacent
axial cells (Figs. 4-5). The early winter plant
is initially ragged (Fig. 3) but rapidly becomes
a ball form. Branches are at first short and un-
branched and spine-like but soon elongate and
rebranch profusely (Figs. 4-5). In this novel
manner, multiseriate axial segments function as
vegetative propagules from which new free-living
balls can develop. The resulting early winter
ball is largely filamentous and consists mostly
of newly formed cells.
In late winter and early spring, a second popu-
lation explosion occurs, facilitated by the
parasitism of a marine unicellular fungus,
Eurychasma dicksonii (Wright) Magnus (Wilce
et al., 1982). As new branches emerge from
the multiseriate propagules, this holocarpic
marine fungus invades numerous newly formed
cells of the filamentous lateral branches
(Fig. 6). Cells of the multiseriate axes re-
main inviolate except on what appears to be
rare occasions. Fungus infected cells die, and
following sporangial discharge by the fungus,
their walls rupture, producing many filamentous
algal segments. These segments range in size
from entire branch systems to short chains of
cells. Habitat conditions at this time seem
favorable for massive amounts of new growth.
Thus, each segment has the potential to develop
into a complete free-living ball and thereby
also serve as a vegetative propagule. This is
the first report of a marine algal population
enhanced as a result of fungal infection.

FIGURE 4. Hahant Bay free-living Pilayella
littoralis. Camera lucida drawing and photo-
micrograph illustrations of laboratory grown
material 14 days after innoculation. -Note the
abundance of multiseriate regions, numerous in-
cipient lateral branches and the occurrence of
rhizoids. Bar on photomicrograph = 100 ~m.
To learn what stimulates the immense amount of
early winter and spring adventitious branch
formation, attempts were made on several occa-
sions to develop new Pilayella balls in unialgal
culture. Innocula for the initial experiments
were filamentous segments of variable length
from free-living balls collected in early August
1979 (R. Alberty, pers. comm.). Little growth
occurred and new balls were not formed in the
cultures owing perhaps to an 'early stage of
temperature-induced summer senescence in the
innocula. On the other hand, filamentous
291
portions of branches excised from late winter
balls all showed large numbers of incipient
lateral branches after an initial growth lag
(Fig. 4). Periclinal cell divisions occurred
in a large number of the axial cells from which
the laterals emerged.
Ball formation developed under a wide range of
light conditions at l5 0 C + 2 0 C. Loose scraggly
balls were formed at 2530, 1500 and 1150 lux.
These balls developed long unbranched laterals
that terminated in pseudohairs of great length.
Tight balls with a multiseriate axis and numer-
ous short, relatively unbranched laterals oc-
curred under 800 and 650 lux. The light regimes
used in these experiments approximate the ambient
summer light of the inshore and deepwater (10 m
MLW) environments where Pilayella was routinely
collected.
3.3. DISTRIBUTION
Attached Pilayella littoralis is known from both
hemispheres (Wilce et al., 1982). It occurs in
temperate to polar waters, and at least in the
North Atlantic and northern polar seas, it is the
dominant filamentous brown alga (Wilce et al.,
1982; R. T. Wilce, pers. obs.). On the Massa-
chusetts coast E. littoralis is the most success-
ful of the brown epiphytic flora and outcompetes
all other filamentous forms as a low intertidal
rock cover (Wilce et al., 1982; R. T. Wilce,
pers. obs.).
292
FIGURE 5. Nahant Bay free-living Pilayella
littoralis. Camera lucida drawing and photo-
micrograph illustrations of laboratory grown
material 28 days after innoculation. Note
multiseriate central axes and the increased
length of the spine-like lateral branches.
The large number and relative unifurm length
of these laterals and their potential for re-
branching profusely ultimately results in new
ball formation. Bar on photomicrographs =
100 ~m.
Free-living surface drift populations of detached
Pilayella littoralis, invariably with a short
longevity in the water column, occur seasonally
wherever the attached plants are successful
(R. T. Wilce, pers. obs.). The species may also
occur in various forms of free-living bottom
dwelling populations, often constituting a large
portion of the unattached loose-lying populations
of the smaller algae characteristic of many arctic
and subarctic embayments (R. T. Wilce, pers.
obs.). Plants of the loose-lying arctic popu-
lations of E. littoralis, despite an often
bizarre morphology, have distal/basipetal
organization suggestive of an initially attached
origin. Although reproductive organs are oc-
casionally encountered in these unattached and
free-living arctic populations, these diverse
forms of R. littoralis are largely vegetative
and perennate for unknown periods.
The free-living ball form of Pilayella littoralis
has been found only in Nahant Bay. It was col-
lected from most areas of the bay from the surf
zone to c 20m (MLW). It occurred mostly at the
sandy bottom/water interface, occasionally
throughout the water column, and more rarely
in surface windrows during late summer. Despite
careful searching, the ball form of R. littoralis
was not discovered in any of the sandy beach em-
bayments to the north or to the south of Nahant
Bay despite the apparent similarity of some of
these habitats with Nahant Bay embayments.
4. DISCUSSION ~ CONCLUSIONS
4.1. ORIGIN AND LOCALIZATION OF THE BALL FORM
Although the ball form, the unusually well-
developed multiseriate central axes, and the
general lack of reproductive organs are peculiar,
most features of the ball form of Pilayella
littoralis are present in the attached forms of

FIGURE 6. Nahant Bay free-living Pilayella
littoralis. Camera lucida drawings of the alga
showing stages of infection by the marine uni-
cellular.fungus Eurychasma dicksonii. Note
portions of P. littoralis axes distal to fungal
infections are incipient vegetative propagules,
simultaneous occurrence of young and old (dis-
charged) fungal sporangia, and the potential of
algal filament tip regeneration even while re-
tained within a discharged sporangium.
the species in Nahant Bay. So, the initial
genetic stock was probably fragments of attached
]:. littoralis from coastal Massachusetts, most
likely from the shores of Nahant Bay. But if so,
then why hasn't evidence of ball formation from
fragments of attached plants of this species been
seen in the natural environment or in the culture
laboratory? The peculiarities of the free-living
form of Pilayella in Nahant Bay suggest genetic
divergence from the attached forms of the species
(Russell, 1963; Wilce et al., 1982). Then, owing
to the suggested lack of both meiosis and genetic
exchange through sexual reproduction in the free-
living form, its distinct genotype may have been
293
perpetuated. Moreover, whatever genetic diver-
gence may have occurred seems sufficient to
produce life history modifications that success-
fully adapt the free-living plant to the peculiar
features of Nahant Bay and quite possibly make it
unique to Nahant Bay.
The possibility of an adaptive genetic divergence
raises many exciting research questions. Finding
answers to these questions is the object of con-
tinuing research. The genetic distinctiveness
of attached and free-living forms of Pilayella
littoralis from Nahant Bay is being examined.
Also, unialgal isolates of both forms are being
used to determine and compare the effects of
irradiance, temperature and nutrient levels on
their morphology and growth rate.
The uniqueness of the ball form and its locali-
zation in Nahant Bay for at least the last eight
decades further suggest that some features of the
Nahant Bay environment are critical for the con-
tinuance of the population as well as the curious
ball form. What is it about Nahant Bay that
originally allowed the initial free-living form
to survive and thrive, continues to nurture it,
and retains the aberrant form within its confines?
It doesn't appear to be peculiarities in water
chemistry (Quinlan et al., 1983). It more likely
is some peculiar hydrodynamic attribute of the
ball in relation to the circulation regimes pre-
vailing in the bay.
A physical model of the free-living Pilayella
phenomenon in Nahant Bay might be seen as biomass
on a one-way conveyor line driven by currents and
waves. Water entering the bay from the north
around Phillips Point flows southward through the
bay with varying degrees of onshore thrust de-
termined largely by wind and tidal direction, and
it exits the bay through a deep channel between
Egg Rock and East Point (Fig. 1) (Quinlan et al.,
1983). Seed stock of the Pilayella population
boards this conveyor somewhere offshore and
gradually moves shoreward where it becomes
trapped by wind and wave induced onshore surf
294
zone currents.
Loss of portions of the Pilayella drift community
must occur in the water mass that exits the bay to
the south, but the ball form of this species has
not been discovered in the sandy beach embayments
south of East Point. As a consequence, the major
loss of biomass from the Pilayella population
occurs in the form of shore cast, resulting in
the Pilayella problem.
The model proposed above was physically tested
using tufts of cotton wool dyed with Rhodomine WT
to track the trajectory of the Pilayella drift
community in Nahant Bay (D. F. Brown, field notes).
4.2. CAUSE OF THE INFESTATION
The obvious cause of the fouling problem is the
free-living ball form of Pilayella littoralis,
but the nutrient source stimulating its growth
is not so obvious.
Eutrophication is immediately suspected when one
is confronted with so much biomass, in this in-
stance hundreds of tonnes yearly, in a relatively
constricted coastal area (Quinlan et al., 1983).
Furthermore, before this project, the local
communities blamed the Swampscott Sewage Treat-
ment Plant outfall in outer Nahant Bay as the
nutrient source. There is no evidence to support
such a conjecture (Quinlan et al., 1983). It was
merely the simplest one to conceive, and as it
turned out, the one most readily eliminated.
Nonetheless, the periodically overwhelming mass
and pervasiveness of the Nahant Bay Pilayella
community, roughly only 1,000 kg of organic
nitrogen is tied up in the mass of the fouling
alga at its summer peak of abundance (Quinlan
et al., 1983). If this nitrogen were evenly
distributed throughout the waters of Nahant Bay,
it would raise its nitrogen level by roughly
only 10 ppb N, or .11% of the observed mean ni-
trogen level of c 90 ppb N (Quinlan et al., 1983).
This is certainly not symptomatic of eutrophica-
tion.
The remaining alternatives for a nutrient source
are the twice daily tidal exchanges with Massa-
chusetts Bay and in situ regeneration of nutrients
in the nearshore waters of Nahant Bay and its
beach sands (Quinlan et al., 1983). Of these,
the most plausible is in situ regeneration, i.e.,
recycling of algal biomass nutrients by its as-
sociated flora and fauna. The fouling alga grows
mostly in negatively-buoyant clouds drifting just
above the inshore sand/water interface. In this
respect, the Pilayella drift community is similar
to loose-lying leaf litter on the forest floor,
but unlike leaf litter, growth occurs in the drift
community in the shallow sublittoral « 6m MLW)
immediately offshore of the breakers. Therefore,
it follows that the requisite nutrients must occur
in the nearshore environment. Nutrients released
from decomposing algal biomass cast on and buried
in the intertidal sands is the most obvious
concentrated source in the nearshore waters.
Also, within the drift community itself, inver-
tebrate remineralization of ingested biomass and
microbial remineralization of extracellular or-
ganics are two plausible sources of recycled
nutrients.
The self-sustaining nutrient recycling system as
suggested here is a testable hypothesis. Such a
system would be maintained by nutrient-rich
interstitial water pumped from the sandy sedi-
ments by wave and invertebrate action. Nutrients,
especially nitrogenous compounds, are added to the
interstitial environment from decay of sand-buried
Pilayella and animal excrement. Furthermore, dis-
solved organics of unknown volume are also re-
leased from the drifting masses of the Pilayella
community with the possibility of subsequent
direct recycling by developing plants. The
proximity of these nutrient sources fo the
growing masses of the fouling alga is an attrac-
tive aspect of this hypothesis.
5. SUMMARY
The success of free-living Pilayella littoralis
in Nahant Bay is a result of several biological
and physical features. Genetic divergence of
the free-living ball form from the attached form

of the species has probably occurred and is sup-
ported by several phenotypic differences. The
free-living ball form in concert with a marine
fungus has developed novel and very successful
methods of vegetative propagation to increase
population numbers. The physical peculiarities
of Nahant Bay, including the sandy beach eco-
system and the prevailing onshore thrust of the
local currents, permit continuous generation and
subsequent transport of the free-living Pilayella
balls from offshore to inshore waters and in situ
nutrient recycling. Lastly, the ball form and
the small size of the plant readily enhance its
movement and ultimate concentration into masses
by the slightest water current; in this respect
it is much like snow flakes blown into drifts.
ACKNOWLEDGEMENTS. We acknowledge the professional
assistance of several phycologists including G.
Russell, G. T. Boalch and C. W. Schneider; their
participation in this project was critical to its
successful completion. Many others have con-
tributed to the development of this project; a
full citation acknowledging their assistance is
given in Wilce et al. (1982). A. N. Davis and
S. Miller assisted in the final review of the
manuscript.
The research project was supported with funds
provided by the U. S. Department of Commerce
through the M.I.T. Sea Grant College Program
(No. NA 79 AA-D-0010l), by the Commonwealth of
Massachusetts through the Metropolitan District
Commission, by the Henry L. Doherty Professor-
ship in Ocean Utilization awarded to A.V.Q.,
and by a grant from the DuPont Fund of the
Department of Mechanical Engineering, M.I.T.
REFERENCES
Anonymous (1903) Deposit from the Sea Nauseates
Residents, In: Lynn Daily Evening Item, 19 August
1903, Lynn, MA.
Bird NL and McLachlan G (1976) Control of for-
mation of receptacles in Fucus distichus L. subsp.
distichus (Phaeophyceae), Phycologia 15, 79-84.
Coleman AW (1961) Role of modern culture methods
in the study of algal life cycles, Quart. Rev.
Biol. 36, 247-253.
Dammann H (1930) Entwicklungsgeschichte und
Zytologische Untersuchungen an Helgolander
Meeresalgen. Helgolander Wiss. Meeresunters.
18(4), 1-36.
Dring MJ (1974) Reproduction, p. 814-837 In:
WDP Stewart (ed.), Algal Physiology and Bio~
chemistry, Blackwell Sci. Publ., Oxford.
2~
Gibb DC (1957) The free-living forms of
Ascophyllum nodosum (L.) Le Jol, J. Ecol. 45,
49-83.
Jonsson H (1904) The marine algae of East
Greenland, Medd. om Gr6nl. XXX, 1-73.
Kjellman FR (1883) The Algae of the Arctic Sea,
K. Sven. Vetenskapsakad. Handl., 20, 1-349.
Knight M (1923) Studies in the Ectocarpaceae.
I. The life history and cytology of Pilayella
littoralis Kjellm, Trans. Roy. Soc. Edinb. 53(2),
343-61.
Nygren S (1975) Life history of some Phaeo-
phyceae from Sweden, Bot. Mar. XVIII(3), 131-141.
Oliveria Filho EC de and Fletcher A (1980)
Taxonomic and ecological relationships between
rocky-shore and salt marsh populations of Pelvetia
canaliculata (Phaeophyta) at Four Mile Bridge,
Anglesey, U. K. Bot. Mar. 23, 409-417.
Quinlan AV, Lewis T and Hoyt JK (1983) Fouling
of the Sandy Beaches of Nahant Bay (Massachusetts,
USA) by an Abnormal Free-living Form of the Macro-
alga Pilayella littoralis (Phaeophyta). I. Habitat
Characteristics, In: Symposium on Sandy Beaches
as Ecosystems, Port Elizabeth, South Africa, 17-
21 January 1983.
Rosenvinge LK (1898) Deuxieme m~moire sur les
algues marines du Groenland - Medd. om Gr6nl.
20, K6benhavn.
Russell G (1963) A study in populations of
Pilayella littoralis, J. mar. biol. Ass. U.K
43, 469-483.
Terry LA and Moss BL (1980) The effect of
photoperiod on receptacle initiation in Asco-
phyllum nodosum (L.) LeJolis, Br. Phycol~
15, 291-301.
West J (1967) Pilayella littoralis f. rupincola
from Washington: The life-history in culture,
J. Phycol. 3, 150-153.
Wilce RT, Schneider CW, Quinlan AV and Vanden-
Bosch K (1982) The life history and morphology
of free-living Pilayella littoralis (L.) Kjellm.
(Ectocarpaceae, Ectocarpales) in Nahant Bay,
Massachusetts, Phycologia 21(3), 336-354.

EFFECTS OF FRESH WATER AND OF POLLUTION FROM A MARINE OIL REFINERY ON THE FAUNA OF A SANDY BEACH
A.C. BROWN (Department of Zoology, University of Cape Town, South Africa)
1. INTRODUCTION
The factory of Marine Oil Refiners of Africa Ltd
is situated at the foot of Dido Valley, near
Simonstown on the coast of False Bay, South
Africa. Between the factory and the sea lie the
road to Simonstown and an intertidal sandy beach
somewhat less than 400 m in length and flanked
to the north and south by rocky shores. For many
years, until August 1979, the factory effluent
was discharged into a small stream which flowed
down Dido Valley, through a drainage pipe under
the road, and escaped onto the open beach above
high watermark. Some of the water/effluent mix-
ture sank into the sand, especially when the
tide was low, the rest flowing down the beach to
the sea through a shallow channel which it had
cut in the sand. The effluent was discharged
intermittently, mainly during the day.
Between 3 000 and 5 000 litres of effluent were
discharged monthly, although peak figures of up
to 10 000 £ were reached from time to time. The
effluent consisted largely of glycerol and fatty
acids, together with emulsified oils and solid
fats, and was virtually lacking in nitrogen. The
only metal present in concentrations significant-
ly higher than in the local tap water was nickel,
at about 0.55 mg £-1. Chemical oxygen demand
was in the region of 2 500 mg £-1 , occaSlona
. 11 y
reaching much higher values. The pH varied from
8.6 to 9.4. The effluent was normally discharged
at a temperature a few degrees above that of the
stream, although on occasion it entered the
stream at 70 to 75 0 C, causing thermal pollution
of the area.
297
The work reported below constituted part of a
project undertaken to determine the effects of
this discharge on the fauna and flora of Dido
Valley Bay.
2. PRELIMINARY OBSERVATIONS
In May 1975 and again in April and August 1976,
a team from the National Research Institute for
Oceanology conducted chemical and physical anal-
yses of surface and interstitial waters along
the 1ength of the beach (Fi gs. 1 and 2). These
surveys showed that phosphate and silicate levels
were elevated in the vicinity of the outfall,
while oxygen levels were depressed. The oxygen
absorbed by samples was considerably increased
both in surface and interstitial waters. Salinity
measurements showed the decrease expected when
sampling near a fresh water stream and salinity
was significantly depressed even at a depth of
10 cm below the sand surface (Fig. 3).
Experiments conducted in the laboratory showed
that the effluent had a very low short-term
toxicity (Brown, 1979). These tests are not
reported here in detail but it may be noted that
the animals used in reaching this conclusion
included gametes and embryos of the sea urchin
PMec.rul1U)., al1giLfoJ.>u)", adul t GMtIlO!.>ac.c.uJ.> pJ.>ammo-
dyteJ.> , the sandy-beach whelk B~ dig~
and a variety of fish larvae. The effects of a
range of effluent concentrations on the heart
rate of the black mussel Cnohomyt{tu)" mehidio-
11~ were also considered. No effects below a
5% concentration of effluent in sea water could
be detected for any of these animals or prepar-
298

arations, and in the case of B~ no adverse

effects could be demonstrated at a 10% concen- 0.04 r--___________..:.OU:.;,tf..:.81..:.1_ _ _ _---,
tration in an experiment lasting several weeks.
Oxygen absorbed
Although these were all relatively short-term (m • • -1)
0.03
tests, it was concluded that - bearing in mind
the composition of the effluent - any damaging 0.02
effects of the outfall on the sandy beach eco-
system were less likely to be due to actual toxic- 0.01
ity than to reduced oxygen tensions, particularly
in the interstitial spaces. These were possibly O~====~=====L--~~~--~
combined with physical changes in the texture of 10.0 Dissolved oxygen
(m. 1-1)
the substratum due to the effluent's tendency to
7.5
cause the grains of sand to coalesce.
Field observations seemed to confirm this view,
5.0
interstitial oxygen levels being low at all seas-
ons of the year, with marked black layers near 2.5
the sand surface except when clean sand

Phosphate
0.03 Oxygen absorbed (mg 9_1 )
15 (U9- at 1- 1 )
(surfaCe water)
0.02
10
0.01

0.9 Oxygen absorbed (mg g-1)

(Interstitial water)
0.6
0.3L_----------
Silicate
_1 (uo-at 1-1)
Dissolved oxygen (mo 0 )
10 (surface water)
80

8 60

9 Phosphate (Ug- at 1-1 ) 40

(surface watel)
6
3 20~
(ug-at 1-1 )
90
o
Silicate
(surface water) 100 200 400
60
Distance in meters from southern
30
end
100 200
FIGURE 2. Chemical measurements in 1976.
Distance in meters from southern end (Open circles: surface water samples. Closed
circles: interstitial water samples.)

FIGURE 1. Some chemical measurements along Dido was deposited on the beach during strong south-
Valley Beach obtained in May 1975. easterly winds in summer. Evidence was also
gained that much of the effluent sinking into
 299

the sand formed a layer on top of the water table mid-tide level along the length of the beach,
and acted as a barrier to the penetration of using a stainless steel corer penetrating to a
oxygen (see Brown, 1979, 19S0, for details). depth of 30 cm. Three replicate cores were taken
No aquatic macrofauna was apparent on or in the from each sampling site, mixed together and then
intertidal sand, or in the surf, although staphyl- preserved in 5% formaldehyde. In the laboratory,
inid beetles were in evidence above high water- sub-samples were taken and the meiofaunal species
mark. The absence of aquatic macrofauna was not counted and identified as far as was possible.
thought to constitute evidence of severe pollution
as there are several sandy beaches in the Cape 4. RESULTS AND DISCUSSION
Peninsula area which, although relatively unpoll- The results of some of the total meiofaunal
uted, also lack aquatic macrofauna (Brown, 1971). counts up to August 1979 are shown in Fig. 4.
Densities at the southern end of the beach, while
not high, fell within the normal range for sand
35 of medium to fine particle size fairly exposed
JAN.1977 to wave action (Fricke, pers. comm.). A marked
34
AUG.1981
decline was, however, apparent towards the out-
0.
fall, at which site the meiofaunal density was
-
C.
33
>.
c
reduced to about 5% of that at the southern end.
32 The same low density was recorded up to about 10
'"
CI)
m north of the outfall. Proceeding further north,
31
100 400 some recovery in density was apparent, although
Distance in meters from southern end
the numbers never reached those encountered at
the southern end of the beach. Exactly the same
FIGURE 3. The salinity of interstitial water at trends were found in the flatworms, nematodes
a depth of 10 cm along the length of the beach and polychaetes considered separately.

In August 1979, the source of pollution was

60r-----------------------------------~
removed by rerouting the factory effluent through
a pipe running across the rocks north of the
sandy beach. The present investigation was con-
tinued until the end of 19S2, in order to study
the possible recovery of the fauna.

3. METHODS
As there was no macrofauna to study, attention
was concentrated on sampling the meiofauna,
although a search was conducted for macrofaunal
species on every visit to the site.
Meiofaunal densities and distribution were assess-
ed on a number of occasions, using the methods
and apparatus described by Dye (197Sa,b) and by Distance in 'meters from southern end
Fricke et al. (19S1), so that the results could
be compared with those reported from other South FIGURE 4. Total meiofaunal densities along the
length of Dido Valley Beach prior to August 1979.
African beaches. Sand cores were obtained from
300
It was tentatively concluded that, while the
absence of aquatic macrofauna was not necessarily
a result of pollution, the meiofauna was being
damaged by the effluent, this damage being due
largely or even entirely to reduced oxygen
tensions.
The same methods were employed to sample the
meiofauna after August 1979, when pollution of
the beach by this effluent ceased. Results obtain-
ed in November 1979 and February 1980 showed no
significant recovery in meiofaunal densities and
it seemed possible that the effects of years of
effluent discharge were still in evidence. How-
ever, as more sampling was undertaken through
1981 and into 1982 (see Fig. 5), this explanation
became less and less likely until it was necessary
to conclude that the meiofaunal distribution was
not, and never had been, related to the outfall.
45
~
'0
::.30
"'E
~
(J)
Q.
~
(J)
15
.0
E species.
:::J
Z
o 100 200 300
Distance in meters from southern end
FIGURE 5. Total meiofaunal densities along the
length of Dido Valley Beach after August 1979.
Meiofaunal distribution did, however, correlate
with the salinity of the interstitial water, for
the stream which had formerly carried the efflu-
ent continued to run across the beach. In
addition to the stream, seepage of fresh water
was apparent along the beach and especially to
the north of the stream. Severe reductions in
meiofaunal densities attendant on reduced salin-
ities have in fact been reported from other sandy
beaches (Ganapati, Rao, 1962; Bush, 1966;
Govindankutty, Nair, 1966; Jansson, 1968; Pollock,
1971; Fricke, pers. comm.).
Another surprise caused a complete reversal of
the initial indications. When the beach was
visited in January 1980, some five months after
the rerouting of the effluent, the wash zone
was found to have a flourishing population of the
cirolanid isopod E~ydZQC lo~g~QOh~, maximum
densities being about 150 animals per m2 . This
is quite a high density for this species on Cape
Peninsula beaches (Brown, 1973). This animal
was noted on all subsequent visits to the beach.
In August 1980, the mysid G~tho~aQQu~ p~ammo­
dy;t~ was encountered and this species, too,
has been found on all subsequent visits, although
not in large numbers.
It is thus necessary to conclude, in contrast to
my previous hypothesis, that meiofaunal densities
Aug.19BO
on Dido Vally Beach were, and are, controlled
largely by interstitial salinities, to an extent
Feb.1982
Aug,1981 which masked any damage caused by pollution; on
the other hand, it appears that the effluent was,
until the end of 1979, effective in preventing
the colonisation of the beach by macrofaunal
This brief account is presented as a warning of
the caution that must be exercised in drawing
400
conclusions from field data.
5. ACKNOWLEDGEMENTS
I am grateful to Marine Oil Refiners of Africa
Ltd for their financial support of this project
and for their willing co-operation. Messrs A.H.
Fricke and R. Bally assisted me with meiofaunal
sampling and Miss F.M. da Silva prepared the
figures. The National Research Institute for
Oceanology gave permission to use their data on

chemi ca 1 analyses of the water of Di do Valley
Beach.
6. REFERENCES
Brown AC (1971) The ecology of the sandy beaches
of the Cape Peninsula, South Africa. Part 1:
Introduction. Trans. Roy. Soc. S. Afr. 39, 247-
279.
Brown AC (1973) The ecology of the sandy beaches
of the Cape Peninsula, South Africa. Part 4:
Observations on two intertidal Isopoda, Eunydi~~
longi~o~~ ~StUder) and Exo~pha~oma tAun~ati­
~~on Barnard. Trans. Roy. Soc. S. Afr. 40
381-404. ' observatlons on the interstitial fauna of the
Brown AC (1979) The effect of the effluent from
Marine Oil Refiners (pty) Ltd on the marine flora
and fauna of Dido Valley Bay during the period
1974-1979. Report in the library of the Zoology
Department, University of Cape Town, 68 pages.
Brown AC (1980) The effect of the effluent from
Marine Oil Refiners of Africa Ltd: a second report.
Re~ort ~n the library of the Zoology Department,
Unlverslty of Cape Town, 28 pages.
Bush LF (1966) Distribution of sand fauna in
beaches at Miami, Florida. Bull. mar. Sci. 16
58-75. '
301
.Dye AH \1978a) Diurnal vertical migrations of
melofauna ln an estuarine sand flat. Zool. afro
13, 201-206.
Dye AH (1978b) Seasonal fluctuations in the
vertical distribution of meiofauna in estuarine
sediments. Zool. afro 13, 207-212.
Fricke AH, Hennig H and Orren MJ (1981)
The relationship between oil pollution and psammo-
littoral meiofauna density of two South African
sandy beaches. Mar. Environm. Res. 5, 59-77.
. Gana~a~i PN and RAO GC (1962) Ecology of the
lnterstltlal fauna inhabiting the sandy beaches
of Waltair coast. J. mar. biol. Ass. India 4
44-57. '
Govin~akutty AG and Nair NB (1966) Preliminary
south-west coast of India. Hydrobiologia 28
101-12.2. '
Jans~on BO (1968) Quantitative and experiment-
al studles of the interstitial fauna in four
Swedish sandy beaches. Ophelia 5,1-71.
Pollock LW (1971) Ecology of intertidal meio-
benthos. Smithsonian Contrib. Zool. 76, 141-148.

DONAX SERRA AND BULLIA RHODOSTOMA: POSSIBLE BIOINDICATORS OF TRACE METAL POLLUTION OF SANDY
BEACHES, WITH PARTICULAR REFERENCE TO THE SOUTH-EASTERN CAPE
H.R. WATLING and R.J. WATLING (Zoology Dept., University of Port Elizabeth, Cape).
1. INTRODUCTION
A great variety of molluscs occur around the South
African coast, extending as it does from the sub-
tropical environment of Natal to the temperate
environment of the Cape. The potential of many of
these molluscs as bioindicators has been discussed
in general terms on the basis of the reported use
of related species (Darracott and Watling, 1975)
and certain of these, among them the bivalve
Donax serra and the gastropod Bullia rhodostoma,
have been included in the National Marine
Pollution Monitoring Programme (Cloete and
Watling, 1981). However, few data on the
accumulator ability of these species or on their
tolerance to metals are available.
The present studies were undertaken a) to
determine the metal concentrations in D. serra
and B. rhodostoma growing along the southern
African coast, supplementing data from sediment
and water sampling surveys of the same region;
and b) to determine in laboratory studies whether
these molluscs accumulate metals, thus meeting
some at least of the criteria for monitoring
and indicating organisms.
2. BIOINDICATORS AND MONITORING ORGANISMS
The trace metal composition of marine organisms
has received attention by the scientific
community since the widely publicised discovery of
hazardous levels of mercury in certain fish and
shellfish from Minamata Bay, Japan. While most
303
of the attention has been focused on determining
whether trace metals can accumulate in organisms
to levels which are potentially detrimental to
human health, increased attention has also been
given to determining the cycling mechanisms of
trace metals in the marine environment and the
possible damage inflicted by metals on the biota
in this environment.
2., t Biological Monitors
Some biological accumulators can be used to
monitor pollution levels quantitatively. The ideal
characteristics for a monitoring organism have
been discussed briefly by several authors (Bittel
and Lacourly, 1968; Haug et al., 1974; Phillips,
1977) and have been the subject of two inter-
national workshops (Butler et al., 1971;
Portmam et al., 1975). Goldberg et al., (1978)
place particular emphasis on the practical
application of the use of monitoring organisms,
proposing that mussels and oysters are the most
useful species for a world-wide survey. The
"ideal characteristics" for monitoring organisms
as suggested by these authors are summarised in
Table 1.
Assuming that these requirements for a
monitoring organism can be met, any monitoring
programme must still be carefully planned. Data
must be intercomparable and wherever possible,
should be obtained for a specific purpose. Mere
measurement, especially on a large scale, is
wasteful of time and resources (Portmann e1; al.,
1975). For a full understanding of anyone
304
TABLE 1. Characteristics of monitoring organisms
1 - The organism must be able to accumulate the
pollutant without being killed by the levels
encountered; accumulation should be
proportional to the levels in the environment
and, ideally, independent of variations in
environmental parameters.
2 - The organism should be sedentary in order to
be representative of the study area.
3 - The organism should be abundant and
sufficiently long-lived to allow the sampling
of more than one year class, if desired.
4 - The organism should have a broad distribution,
both ecologically and geographically; this
condition facilitates comparison between
geographically separated areas, but such
comparisons must be supported by laboratory
data on the similarity of behaviour and
physiological responses of individuals from
different populations.
5 - The organism should be hardy and adaptable
so that it can be transferred into a new
locality. Euryhaline species such as
oysters and clams are particularly useful in
locating sources of industrial effluents
which typically enter the estuary in low
salinity areas near river mouths; alternative-
ly, it is often desirable to transfer the
organism to an aquarium where it can purge
its intestinal contents before being killed
for analysis.
6 - The organism should be of reasonable size so
that adequate tissue is available for
analysis.
7 - There should be sufficient knowledge of the
biology of the species to recognise its most
sensitive life stage, its position in the
trophic web and its reaction to environmental
changes other than those induced by man.
8 - It should be possible to investigate the
physiological effects of the suspected
pollutants in controlled laboratory studies;
the rates at which the organism accumulates
and eliminates the pollutants should be
studied in relation to the levels of
contamination in the environment; these
results should facilitate the interpretation
of field data
environmental situation it is advisable, if not
necessary, to measure the pollutant levels in
sediment and water samples as well as those in the
biota. It is also desirable to use a range of
organisms. Hetals can be accumulated from three
different sources, from the ingestion 6f food,
from solution, and from the ingestion of metal-
rich particulate matter (Phillips, 1977). Not all
indicator types will reflect all three trace metal
loads equally.
The use of biological species to monitor
metal pollution in the marine environment has the
advantage that accumulation will reflect the
presence of "available" metals over a period of
time. Such organisms can be sampled at convenient
intervals independent of the intermittent
addition of pollutants to a body of water because
they provide an integrated measure of the metal
load in the water. For example, severe cadmium
pollution of industrial origin was traced to its
source, not as a result of the sediment and water
surveys of the estuary, but because of the extreme
accumulation of this metal by oysters placed in
the estuary (Thornton et aZ., 1975; Boyden,
1975) .
2.2 Biological Indicators
The ability of many species to accumulate
pollutants has been used with varying degrees of
success, to indicate the presence of pollution
(e.g. Hajori and Petronio, 1973; Bryan and
Hummerstone, 1977; Reish, 1970). Indicator
organisms are used primarily to identify rather
than to measure environmental changes whose cause
may be unknown or which may be the result of a
variable mixture of pollutants. For example,
increased numbers of some species may indicate
the presence of toxic pollutants or changes in
the overall species composition may occur over a
period of time as an indication of increased
pollution (Dills and Rogers, 1974; Emmerson and
Watling, 1983; Halcrow et aZ., 1973). Bryan
(1971) in his review on the effects of selected

metals on marine and estuarine organisms, stated
that "the concentrations of metals in some sessile
organisms, such as brown seaweeds and some
molluscs, tend to reflect the concentrations in
the water and might, after further study, be used
as indicators of chronic water pollution".
Clearly the criteria for biological
indicator organisms are far less stringent than
those for monitoring organisms where the
intention is to quantify the degree of pollution
and measure changes occurring with time.
3. THE BIOLOGY AND ECOLOGY OF DONAX SERRA AND
BULLIA RHODOSTOMA WITH PARTICULAR REFERENCE
TO THE SOUTH EASTERN CAPE
Two species of Donax, D. sordidus and the sand
mussel D. serra occur on Eastern Cape sandy
beaches (McLachlan 1977a, 1977b, 1980). The
former is small and undergoes tidal migration,
while D. serra is the largest species in the
genus (De Villiers, 1975b) and only exhibits a
semilunar pattern of movement up and down the
intertidal zone (McLachlan, Wooldridge and Van
der Horst, 1979). Day (1969) records D. serra
from Luderitz to Port Elizabeth below mean tide
level and migrating with the tides. Numerous
live D. serra have since been collected on the
southern Transkei coast, thus extending their
geographical range considerably. Further, their
spring-tidal position has been shown to be above
the mid-tide level in the Eastern Cape, where
they do not undergo normal tidal migration
(McLachlan and Hanekom, 1979). D. serra
develops vast populations on some South African
shores (De Villiers, 1975b; McLachlan, 1977b).
The results of studies on the reproduction
and growth of D. serra on west coast shores have
been reported (De Villiers, 1975a, b) and
McLachlan and Hanekom (1979) have described the
general distribution of D. serra in the eastern
Cape, its population, structure growth,
305
production, aspects of its biology and seasonal
changes in the biochemical composition,
Plough shells of the genus Bullia are well
represented on South African sandy shores with
five intertidal and eight subtidal species
(Brown, 1971). They are carnivorous scavengers
and feed on stranded organisms, particularly
coelenterate medusae and siphonophores. Some
aspects of the general biology and physiology of
Bullia species in the Cape peninsula have been
studied by Brown (1961; 1971; 1982a) .
B. rhodostoma is the common intertidal species of
the south and east coasts of South Africa (Day,
1969; McLachlan, 1980) and is the dominant
macrofaunal organism on many beaches of the
Eastern Cape (McLachlan, 1977a). From their study
on the growth and production of B. rhodostoma
McLachlan, Cooper and Van der Horst (1979)
conclude that this species is a successful
scavenger on exposed sandy beaches in the Eastern
Cape where it appears to be well adapted to a
physically controlled environment and an erratic
food supply. It is a long-lived species that has
a slow growth rate and produces relatively few
eggs that are well cared for by the females
(Brown, 1971). It reaches a length of about
10 mm after one year and 40 mm after 10 years
(McLachlan, Cooper and Van der Horst, 1979).
Most production by adults goes into reproduction,
particularly in the females which grow larger
than the males.
Further comparative ecological studies with
particular reference to the role of D. serra and
B. rhodostoma on Eastern Cape sandy beaches have
been carried out at UPE and the results of these
are reported by McLachlan and co-workers (Ansell
and McLachlan, 1980; McLachlan, Dye and Van der
Ryst, 1979; McLachlan et al., 1981; McLachlan,
Wooldridge and Dye, 1981; McLachlan and Van der
Horst, 1979).
306
4. FIELD STUDY
D. serra and B. rhodostoma were collected from
sites along the South African coast (Fig. 1)
during several surveys in the period August 1978
to August 1979.
4.1 Materials and Methods
Living specimens were suspended in clean seawater
for up to four days. The wet tissues were then
removed from the shells and frozen inside glass
vials.
The frozen specimens were thawed, weighed
into clean dry flasks and oven-dried at 90°C for
24 h. The dried samples were reweighed,
dissolved in 25 ml of redistilled analar grade
nitric acid and the sample solution boiled and
evaporated to near dryness. The residue was
dissolved in 25 ml of a 4:1 nitric:perchloric
acid mixture. This solution was fumed to dryness
at about 120°C. The white residue was redissolved
in 10 ml of 10% v/v nitric acid and the metal
concentrations in this solution determined by
atomic absorption spectrometry.
Composite standards containing zinc, cadmium,
copper, lead, iron, manganese, nickel, cobalt and
chromium in the range 0.1 to 20 ~g/ml, in the
presence of sodium, potassium, calcium and
magnesium in the range 100 to 5 000 ~g/ml were
prepared in 10% v/v nitric acid. A Varian
Techtron AA5 with AA6 readout module and BC6
background corrector was used for all measurements.
Background correction was applied to the
determination of zinc, cadmium, lead, nickel and
cobalt and the slotted tube (Watling, 1978) was
used to increase the sensitivity of the lead
determination. The instrument was calibrated
using the composite standards.
,4.2 ResuLts and Discussion
Metal concentrations recorded for D. serra and
B. rhodostoma are listed in Tables 2 and 3
respectively. The major portion of the coast
from which samples were collected is undeveloped
and it is expected that the concentrations
determined represent near background levels for
these species. Concentrations in oysters and
mussels from this coast indicate that the area
is essentially unpolluted with the exception of
a few isolated places which are almost always
associated with urbanization (Watling and Watling,
1979; 1982). Therefore, the data from the
present survey will serve as a baseline for the
future monitoring of the sandy beaches region.
Considerable variations in metal
concentrations in D. serra and B. rhodostoma are
apparent. These are often accounted for by
differences in the size of the samples, the
higher levels occurring in smaller individuals.
Hornung and Oren (1981) also reported higher
concentrations of zinc, cadmium, copper and
lead in smaller individuals of Donax truncutus
from Haifa Bay. Zinc and cadmium in B. rhodostoma
follow this pattern, but in the case of cadmium,
the concentrations are slightly higher for a
given size of individual in the western part of
the survey area, as has also been found for
oysters (Watling and Watling, 1982). K.C. Davies
(unpublished data) has assayed the concentrations
of both cadmium and zinc in a number of Buttia
populations on the west and south coasts of South
Africa. She found that Buttia digitatis from
the west coast have higher tissue levels of
cadmium and zinc (B. digitatis from Ou Skip
x 30 ~g/g Cd in wet tissue (Cuthbert et at.,
1976)) than do south coast populations of
B. digitatis and B. rhodostoma, and that the
levels of metals accumulated do not increase
with increasing size of adult individual. These
findings tend to support those of the present
survey. The results of a survey to study seasonal
variations in metal concentrations were inconclu~
sive but led us to believe that differences due to
size were masking any variations due to seasonal
changes. As far as is practicable, individuals
 307

TABLE 2. Metal concentrations in D. serra collected from sites on the South African coast

Wet Dry
Sample Location llg metal / g wet tissue
No mass mass Zn
(refer Fig. 1) Cd Cu Pb Fe Mn Ni Co Cr
(~) (~)

2 Dana Township 40 x 8.50 1.45 13.0 0.07 0.82 0.34 81 1.33 0.47 0.19 0.65
(new development) s 2.3 0.02 0.12 0.05 29 0.64 0.13 0.04 0.29
3 Diaz Beach 4 x 5.62 0.96 19.3 0.12 1. 20 0.06. 79 1. 39 0.23 0.09 0.82
6 Glentana 3 x 3.43 0.62 15.6 0.14 1. 18 0.03 84 1. 15 0.43 0.04 0.26
9 Buffalo Bay 30 x 8.93 1. 99 15.2 0.03 1.14 0.02 128 1. 36 0.46 0.06 0.61
s 1. 41 0.28 3.7 0.01 0.15 0.01 26 0.46 0.05 0.01 0.16
14 Keurboomsrivierstrand 10 x 8.15 1. 79 14.3 0.03 1. 35 0.01 126 1.58 0.50 0.07 0.45
16 Oyster Bay 28 x 9.08 1. 64 14.6 0.03 1.01 0.02 71 1. 73 0.39 0.05 1. 26
s 2.49 0.51 3.6 0.01 0.26 0.01 21 0.90 0.08 0.04 0.51
17 Kromme River mouth 4 x 8.62 1.77 12.8 0.03 1. 00 0.03 229 1.47 0.41 0.04 0.63
19 1 x 7.50 1. 59 15.6 0.06 1. 16 0.08 209 2.30 0.36 0.08 0.65
20 5 x 5.74 1.12 12.9 0.02 0.88 0.08 222 1. 99 0.33 0.04 0.59
21 18 x 10.30 1. 97 12.7 0.03 0.92 0,04 235 2.10 0.23 0.05 0.71
s 4.34 0.90 2.2 0.01 0.15 0.02 69 0.58 0.09 0.02 0.28
23 Gamtoos River mouth 3 x 4.34 0.81 11.3 0.04 0.82 0.09 230 1. 56 0.25 0.09 0.76
24 8 x 7.73 1. 54 15.9 0.02 1. 08 0.07 453 3.69 0.32 0.10 1.03
25 14 x 10.60 2.07 12.6 0.02 0.93 0.04 291 2.34 0.31 0.07 0.65
26 14 x 10.24 2.04 13.4 0.05 1. 07 0,05 264 2.28 0.32 0.06 0.64
27 11 x 11.34 2.37 14.7 0.04 1. 01 0.03 202 1. 90 0.34 0.06 0.56
28 21 x 10.01 2.14 13.9 0.06 1.14 0.03 181 1. 60 0.45 0.07 0.68
s 3.10 0.81 2.8 0.02 0.23 0.02 69 0.59 0.29 0.05 0.24
29 Maitland River mouth 35 x 8.50 1. 73 11.6 0.04 1. 02 0.03 199 1. 62 0.40 0.06 0.67
s 3.36 0.79 3.6 0.02 0.23 0.04 102 0.85 0.06 0.02 0.18
37 2 x 4.36 0.85 10.9 0.05 0.87 0.03 81 1. 48 0.28 0.03 0.55

DURBAN
N

t
PORT ELiZABET
,250km,

35
38

,25 km

FIG. I LOCATION OF SAMPLE SITES

308

TABLE 3. Metal concentrations in B. rhodostoma collected from sites on the South African coast

Wet Dry
Sample Location ]1g metal / g wet tissue
No mass mass
(refer Fig.1) Zn Cd Cu Pb Fe Mn Ni Co Cr
(g) (~)
Vlees Point 21 x 1. 96 0.47 32.1 3.9 1.36 0.17 84 1.88 0.18 0.14 3.9
s 0.46 0.12 7.7 1.6 0,40 0.11 38 0.44 0.07 0.08 1.3
2 Dana Township 18 x 2.02 0.50 36.3 5.1 1. 89 0.13 93 1. 66 0.20 0.08 2.4
(new development) s 0.41 0.11 11.3 2.5 0.49 0.06 39 0.53 0.07 0.05 0.8
3 Diaz Beach 18 x 3.32 0.58 26.0 2.4 0.90 0.11 24 1. 24 0.13 0.08 1.9
s 0.36 0.10 10.4 0.6 0.27 0.10 12 0.87 0.04 0.04 0.5
4 Hartenbos 30 x 1.54 0.41 50.0 3.6 1. 28 0.12 72 1.56 0.22 0.11 1.6
s 0.81 0.25 21.1 0.9 0.36 0.08 29 0.46 0.15 0.07 0.9
5 Tergniet 15 x 2.94 0.75 31.3 3.7 1. 41 0.06 49 1. 33 0.19 0.08 2.8
s 0.36 0.22 12.1 1.2 0.32 0.03 23 0.17 0.09 0.03 1.2
6 Glentana 30 x 0.80 0.19 46.6 8.5 2.23 0.26 80 1. 78 0.19 0.11 3.7
s 0.31 0.06 9.5 1.1 0.92 0.07 20 0.28 0.08 0.09 1.1
7 Herolds Bay 18 x 2.79 0.67 26.5 4.1 1.42 0.11 63 1. 31 0.17 0.06 2.5
s 0.47 0.13 7.3 1.8 0.61 0.16 25 0.24 0.18 0.04 0.9
8 Walker Point 10 x 1. 85 0.40 43.6 5.2 2.17 0.04 27 1. 00 0.11 0.04 0.61
s 0.50 0.18 7.1 1.0 0.43 0.03 16 0.23 0.07 0.02 0.22
9 Buffalo Bay 1 17 x 2.95 0.64 41.1 4.5 2.23 0.05 38 1. 04 0.11 0.05 0.65
s 0.42 0.08 6.4 0.9 0.39 0.03 12 0.11 0.09 0.02 0.21
10 Brenton-on-Sea 16 x 2.47 0.51 38.1 4.3 2.19 0.04 39 1. 03 0.11 0.04 0.5
s 0.31 0.07 10.7 0.4 0.50 0.02 13 0.16 0.05 0.01 0.16
11 Noetzie 18 x 2.06 0.50 52.5 5.3 4.42 0.05 22 0.82 0.15 0.03 0.81
s 0.17 0.05 14.5 1.3 1. 26 0.03 6 0.20 0.03 0.01 0.10
12 Cape Seal (west) 25 x 2.09 0.55 51.5 5.4 1. 98 0.13 50 1. 19 0.32 0.14 3.3
s 0.67 0.15 20.3 2.4 0.69 0.08 16 0.18 0.13 0.08 1.6
13 Cape Seal (east) 19 x 2.92 0.74 44.4 4.2 3.74 0.04 27 1. 10 0.23 0.03 0.68
s 0.60 0.16 15.4 1.2 0.96 0.02 10 0.41 0.29 0.01 0.23
14 Keurboomsrivierstrand 20 x 2.56 0.44 50.6 8.8 2.36 0.09 55 0.75 0.17 0.08 2.4
s 0.82 0.15 17.8 2.7 0.77 0.08 18 0.16 0.07 0.05 1.0
15 Nature's Valley 30 x 2.19 0.49 46.8 7.1 1. 67 0.03 29 0.90 0.15 0.03 0.78
s 0.27 0.07 13.6 6.3 0.47 0.02 8 0.12 0.10 0.02 0.29
16 Oyster Bay 40 x 2.32 0.48 37.5 7.2 1. 34 0.08 47 0.85 0.11 0.03 2.1
s 0.47 0.10 11.8 2.4 0.41 0.03 21 0.13 0.06 0.02 0.6
17 Kromme River mouth 20 x 2.34 0.48 32.1 4.0 1. 58 0.11 42 1. 49 0.15 0.08 3.6
s 0.60 0.13 12.6 2.2 0.63 0.12 19 0.21 0.09 0.04 1.1
18 25 x 2.31 0.50 34.2 3.4 1.05 0.06 24 1.16 0.17 0.04 0.70
s 0.44 0.13 7.2 0.9 0.41 0.03 5 0.13 0.08 0.02 0.21
19 14 x 1. 72 0.32 37.2 6.8 1. 23 0.18 47 1. 60 0.13 0.19 2.8
s 0.47 0.09 13.8 2.0 0.50 0.08 16 0.26 0.08 0.10 1.1
20 Jeffreys Bay 17 x 2.08 0.55 57.4 3.7 1. 72 0.09 33 1. 36 0.25 0.08 1.6
s 0.54 0.16 22.7 1.5 0.47 0.04 11 0.19 0.17 0.04 0.7
21 25 x 2.30 0.47 31.4 2.6 1.73 0.09 21 1. 37 0.22 0.05 0.87
s 0.53 0.14 11.7 0.9 0.66 0.04 10 0.16 0.12 0.03 0.25
22 8 x 1.77 0.43 49.6 5.0 2.05 0.14 50 1. 58 0.14 0.12 2.9
s 0.69 0.19 11.0 1.5 0.99 0.05 38 0.41 0.13 0.07 0.9
23 Gamtoos River mouth 15 x 1. 80 0.37 53.3 4.2 2.20 0.06 29 1. 40 0.18 0.07 1. 25
s 0.6 0.15 10.2 1.1 0.77 0.02 8 0,32 0.06 0.05 0.49
24 25 x 1. 82 0.37 50.6 5.3 2.10 0.13 38 1. 82 0.10 0.08 3.6
s 0.62 0.17 23.8 3.1 1. 09 0.05 19 0.72 0.06 0.05 2.6
25 15 x 2.01 0.38 37.7 3.1 1. 40 0.08 33 1. 11 0.19 0.06 1. 22
s 0.63 0.16 15.6 1.3 0.61 0.06 13 0.35 0.07 0.04 0.41
26 14 x 1. 58 0.29 37.2 3.7 1. 35 0.11 35 1. 21 0.33 0.08 1. 29
s 0.92 0.23 14.7 1.0 0.29 0.05 10 0.20 0.14 0.04 0.53
27 30 x 1. 93 0.40 44.4 4.8 1.44 0.09 49 1. 33 0.22 0.05 1.5
s 0.71 0.14 15.2 1.5 0.48 0.05 22 0.31 0.16 0.03 0.6
28 35 x 1. 78 0.39 28.6 5.1 1. 44 0.06 37 1. 17 0.19 0.07 0.87
s 0.96 0.24 7.7 1.8 0.37 0.03 10 0.33 0.09 0.04 0.33
 309

TABLE 3. (continued)

Wet Dry
Sample Location Vg metal / g wet tissue
No mass mass
(refer Fig.1) Zn Cd Cu Pb Fe Mn Ni Co Cr
(g) (g)

29 Maitland River mouth 20 x 2.72 0.57 36.4 4.3 1.51 0.07 38 1.15 0.20 0.12 0.81
(1978) s 0.82 0.21 11.9 0.9 0.32 0.04 26 0.16 0.16 0.05 0.28
( 1979 28 x 2.56 0.68 21.4 2.8 1. 07 0.08 27 1. 67 0.12 0.04 1.6
s 1. 32 0.32 8.3 0.7 0.68 0.06 7 0.45 0.13 0.01 0.9
30 Kings Beach 15 x 0.90 0.17 43.1 3.7 2.01 0.11 54 1. 65 0.17 0.11 1.84
s 0.21 0.06 13.0 0.8 0.31 0.04 17 0.48 0.09 0.06 0.72
31 Fishwater Flats 12 x 1.44 0.30 30.6 4.7 2.06 0.09 45 1. 26 0.22 0.14 0.81
s 0.24 0.06 2.9 0.9 0.26 0.03 13 0.07 0.18 0.09 0.16
32 Hougham Park 8 x 1. 70 0.36 35.5 3.6 1. 89 0.14 38 1. 52 0.15 0.10 1. 33
s 0.85 1. 20 11.8 0.7 0.32 0.04 22 0.30 0.06 0.05 0.72
33 13 x 2.15 0.38 42.3 2.1 1. 74 0.09 52 1.67 0.15 0.12 2.4
s 1.04 0.20 12.7 0.6 0.55 0,03 15 0.52 0.05 0.07 0.4
34 16 x 0.86 0.15 44.7 2.3 1.48 0.27 51 1. 37 0.20 0.13 2.6
s 0.19 0.04 9.0 0.5 0.37 0.22 18 0.23 0.08 0.10 0.6
35 34 x 1. 17 0.28 47.6 3.5 1.53 0.06 40 1. 61 0.10 0.06 1.8
s 0.34 0.08 14.3 1.0 0.26 0.04 6 0.16 0.05 0.03 0.5
36 15 x 0.85 0.20 47.1 3.0 1. 94 0.29 64 1. 71 0.22 0.16 4.1
s 0.20 0.05 15.3 0.8 0.47 0.06 23 0.30 0.09 0.12 1.5
37 5 km west of Woody 20 x 2.51 0.53 47.5 3.3 1. 94 0.11 44 1.55 0.17 0.08 1.9
Cape s 0.55 0.13 8.3 2.9 2.18 0.03 14 0.29 0.07 0.03 0.4
38 Woody Cape 25 x 1. 25 0.25 59.6 4.1 1.46 0.12 46 1. 36 0.11 0.05 1.8
s 0.32 0.07 16.8 1.4 0.29 0.04 14 0.22 0.07 0.03 0.4
39 Cannon Rocks 3 x 0.89 0.20 58.3 3.7 1. 51 0.11 38 1. 28 0.11 0.04 1.7

of similar size should be collected at each site sand and 40 1 of seawater. The water was aerated
to minimise concentration variations caused by continuously throughout each 3-week experiment.
size differences. Aliquots of 5 000 Vg/ml stock solutions of the
metal chlorides were added to achieve the
5. LABORATORY STUDY appropriate concentrations in each tank. The
experimental solutions were renewed daily at
This study is concerned with the ability of
whicn time tne molluscs were dislodged from the
D. serra and B. rhodostoma to accumulate metals
sand and placed on its surface so that burrowing
without being killed by the levels encountered,
benaviour could be observed.
in such a manner that the rates of accumulation
can be related to the average metal concentration
Metal loss from these solutions was tested
in the surrounding water.
under the conditions of the experiment, but in
the absence of the molluscs. No losses were
5.1 Materials and Methods
observed for cadmium, nickel, cobalt or chromium
D. Serra and B. rhodostoma were collected from an
in the 24-h period. However, zinc concentrations
unpolluted beach in St Francis Bay to the west
decreased by 5% at 50 Vg/l and 10% at 10 Vg/l
of Port Elizabeth. They were transferred
in the 24-h period; copper concentrations
immediately to the laboratory and allowed to
decreased by about 30% at both 50 and 10 Vg/l
acclimatise in tanks containing.sand and seawater,
levels and lead losses of 30% at 50 Vg/l and
the latter being renewed each day.
50% at 10 Vg/l were observed in the 24-h period.

Comparative accumulation studies were carried

out in polythene tanks containing 5 1 of clean
310

TABLE 4. Cadmium accumulation by Donax serra and Bullia rhodostoma

Donax serra Bullia rhodostoma

Treatment Wet Dry Wet Dry
lJg/g wet tissue lJg/g wet tissue
(lJ gil) mass mass (g~ss (g~ss
Zn Cd Cu Zn Cd Cu
(g) (g)

Control x 18.2 3.1 11.1 0.10 0.60 2.21 0.43 39 5.5 2.6
s 0.72 0.15 9 0.9 1.2
25 X 16.1 3.2 11.5 0.54 0.64 2.60 0.49 38 5.7 2.4
s 0.83 0.18 8 1.6 0.5
50 x 15.1 2.4 11.0 1. 05 0.64 2.70 0.51 39 6.5 2.3
s 0.97 0.17 11 1.2 0.5
100 x 16.4 2.6 11.5 2.19 0.67 2.15 0.41 40 11.4 2.8
s 0.79 0.14 10 2.6 0.7

TABLE 5. Cadmium accumulation in Donax serra tissues (expressed as lJg/g wet tissue)

Control 25 lJg/l 50 lJg/l 100 lJg/l

Mantle 0.08 0.30 0.70 1.18

Gill 0.27 3.38 8.08 19.67
Digestive Gland 0.08 0.42 1. 02 1.82
Syphon 0.15 0.30 0.46 0.73
~Muscle 0.14 0.65 1. 31 2.97
Gonad 0.11 0.49 0.75 1.11
Foot 0.07 0.18 0.24 0.47

TABLE 6. Comparative uptake of seven elements by Donax serra and Bullia rhodostoma
(expressed as lJg/g wet tissue)

Treatment Rate of
Metal Accumulation
Control accumulation
20 lJg/l factor
(lJg/g/day)
Donax serra
Zn 11. 1 + 1. 3 13.8 2: 3.8 1.2 0.13
Cd 0.17 + 0.04 0.32 2: 0.04 1.9 0.01
-
Cu 0.83 + 0.20 1. 75 2: 0.61 2.1 0.04
Pb 0.04 + 0.03 0.65 +
- 0.17 6.2 0.03
Ni 0.31 + 0.11 1. 18 2: 0.40 3.8 0.04
Co <0.02 0.66 + 0.18 33 0.03
Cr 1.0 + 0.6 2.6 2: 1.4 2.6 0.08
Bullia rhodostoma
Zn 33 -+ 11 33 2: 9 1.0 0
Cd 4.3 + 1. 7 7.7 + 4.4 1.8 0.16
Cu 1.2 + 0.4 1.9 2: 0.8 1.6 0.03
Pb 0.09 + 0.06 1.2 2: 0.4 13.3 0.05
Ni 0.18 + 0.08 0.57 -+ 0.17 3.2 0.02
Co <0.02 0.71 + 0.24 35 0.03

Cr 0.6 + 0.2 2.7 + 1.2 4.5 0.10


The wet tissues of individuals were removed
from their shells and frozen prior to chemical
analysis. The oxidative dissolution and trace-
metal analysis of these samples was carried out "
as described for the field survey specimens.
5.2 Results and Discussion
In the first experiment specimens (20 per treat-
ment) were exposed to cadmium in the range 0 to
100 ~g/l; the water temperature was between
14-16°C throughout the three-week experiment. All
individuals survived the experiment, the results
of which are summarised in Table 4. In this
case D. serra individuals were dissected and the
separate tissues analysed for cadmium content.
The mean concentrations for individuals in each
treatment were calculated from the tissue
analyses. The smaller B. rhodostoma individuals
were analysed whole and the mean concentration
and standard deviation for each suite of results
was calculated.
Tissue cadmium concentrations have increased
with increasing concentration in the water for
both species, but particularly in D. serra where
a 20-fold increase is observed. This is mostly
a reflection of the initial low cadmium concentr-
ation of this species as compared with that of
B. rhodostoma. If instead the rate of cadmium
accumulation (~g/g/day) is calculated for the
100 ~g/l treatment, it is seen that the mean rate
of accumulation of B. rhodostoma (0.3 ~g/g/day) is
somewhat greater than that of D. serra (0.1 ~g/g/
day) for this 3-week experiment. For comparison
zinc and copper concentrations were also deter-
mined in each sample but the concentrations of
both these elements remain relatively constant
regardless of the treatment.
Cadmium accumulation by the various tissues
of D. serra is shown in Table 5. Concentrations
in each tissue increase relative to the solution
cadmium concentration, the greatest accumulation
occurring in the gill tissue. In terms of
311
accillmulation factors these data may be summarised
as
gill»>digestive gland~uscle>mantle>gonad~
foot~syphon
The comparative accumulation of seven metals
was also investigated. Specimens were exposed to
20 ~g/l of one of the elements zinc, cadmium,
copper, lead, nickel, cobalt and chromium for
three weeks. There were 20 individuals in each
tank and the water temperature varied between
21-23°C during the experimental period. All
individuals survived the experiment. The whole
tissues were analysed and the mean concentration
and standard deviation for each suite of results
was calculated. These data are summarised in
Table 6.
It is possible to calculate an accumulation
factor (the ratio of the mean concentration of the
study element in the tissues of the treated
individuals to the mean concentration in the
tissues of the control individuals) and these
factors can be used to indicate whether signifi-
cant metal accumulation has occurred relative to
the normal tissue-metal concentration. Clearly,
the ability to distinguish between contaminated
and uncontaminated samples is essential if we
intend to use molluscs as bioindicators.
On the basis of these factors, we suggest
that it would be unwise to use D. serra and
B. rhodostoma to indicate the presence of zinc
contamination; however, the accumulation factors
determined for the remaining elements are great
enough to indicate that metal accumulation is
taking place (from the point of view of a marine
pollution survey).
It is also possible to calculate the rate of
accumulation (~g/g/day) of each element by each
species under these particular experimental
conditions; these values can be used to compare
the accumulation rates of a particular metal by
different species or the accumulation rates of a
312
number of metals by the same species. It is
interesting to note that the rates of accumulation
of lead and cobalt, the elements with the highest
accumulation factors (Table 6 ), are very similar
to the rates of accumulation of most other
elements. Indeed, too few measurements are
available to assign a rank order according to
the rates of accumulation.
It must be remembered that these laboratory
experiments were carried out under largely
controlled conditions which would not necessarily
exist in the natural environment. The elements
were added as metallic chlorides but need not
remain in that form. The problem of losses of
metals from solution has already been mentioned,
in so far as the concentrations to which the
experimental animals were subjected is not known
but may be lower than was intended. However, some
preliminary studies on the behaviour of metals
under similar conditions suggest that the greater
quantity of "missing" metal becomes adsorbed onto
surface sand grains, which could mean that the
experimental animals are in fact sUbjected to
higher than expected concentrations at the sandi
water interface. Alternatively, the metal
adsorbed onto the sand grains may then be less
"available" to Donax and BuZZia.
The question remains as to whether the
comparative' lack of accumulation of certain
elements is due to a chemical mechanism in the
solution whereby the element is rendered unavail~
able to the mollusc (e.g. precipitation,
. 'complexation, adsorption) or to some form of
discrimination, whether passive or active, on the
part of the mollusc.
6. SUMMARY
The main purpose of this study is to assess the
potential of sandy beach molluscs as biological
indicators of metal contamination of Eastern
Cape sandy beaches. The two most obvious choices
for such a study are D. serra and B. rhodostoma.
They are the largest abundant species and their
biology and ecology have been investigated,
with particular reference to the Eastern Cape.
Equally obviously it would be premature to suggest
that these species can be used as biological
monitoring organisms. Indeed certain of their
biological characteristics and behavioural
responses (e.g. mobility) are not compatible with
the criteria for monitoring organisms, as listed
in Table 1.
Brown (1982a; b) has summarised the published
data on both the biology and the effects of a
wide range of pollutants on BuZZia, mainly
B. digitaZis. His compilation of the results of
both field and laboratory studies indicates that
BuZZia is suitable as a test organism and readily
surveyed in the field. However, laboratory
studies on the accumulation of metals from
experimental solutions are not reported.
Comparative data on metal levels in Donax or the
effects of pollutants on Donax are rare. Van As
et aZ., (1973; 1975) reported the levels of a
few metals in D. serra collected from a number
of sites on the south and west coasts of South
Africa.
From the results of the field survey and
laboratory experiments, it is concluded that
D. serra and B. rhodostoma would accumulate metals,
were they present as pollutants in the sandy beach
environment. The metal levels tested were not
apparently toxic during these short-term experi-
ments, but were nevertheless greater than would
be expected in a sandy beach environment. Only
inorganic metal chlorides were tested. Almost
certainly the form of the metal in the environment
will affect the rate at which it is accumulated
and as nothing is known about the types of
pollutants likely to occur on our sandy beaches,
or on their possible rates of accumulation, it
is impractical to suggest that D. serra and
B. rhodostoma should be used as monitoring

organisms. Nevertheless, these species may make
a useful contribution to the coastal monitoring
programme as a whole, supplementing data from
water and sediment sampling surveys. The data
from the present study concerning metal concentr-
ations in these two molluscs growing at a number
of sites along the south-eastern Cape coast
will serve as a baseline for the future monitoring
of the sandy beaches of this region.
7. REFERENCES
Ansell AD and McLachlan A (1980) Upper temper-
ature tolerances of three molluscs from South
African sandy beaches, J. expo mar. BioI. Ecol.
48, 243-251.
Bittel Rand Lacourly G (1968) Discussion sur
Ie concept de facteur de concentration entre les
organisms marins et l'eau en vue de l'interpret-
ation des mesure~ Rev. into Oceanogr. Med. 11,
107-128.
Boyden CR (1975) Distribution of some trace
metals in Poole Harbour, Dorset,Mar. Pollut. Bull.
6, 180-187.
Brown AC (1961) Physiological-ecological
studies on two sandy-beach Gastropoda from South
Africa : Bullia digitalis Meus~hen and Bullia
laevissima (Gmelin), Z. Morph. Okol. Tiere 49,
629-657.
Brown AC (1971) The ecology of the sandy beaches
of the Cape Peninsula, South Africa. Part 2.
The mode of life of Bullia (Gastropoda:Proso-
branchiata),Trans.Roy. Soc. S. Afr. 39, 281-319.
Brown AC (1982a) The biology of sandy-beach
whelks of the genus Bullia (Nassariidae),Oceanogr.
mar. BioI. Ann. Rev. 20, 309-361.
Brown AC (1982b) Pollution and the sandy-beach
whelk Bullia, Trans. Roy. Soc. S. Afr. 44, 555-561.
Bryan GW (1971) The effects of heavy metals
(other than mercury) on marine and estuarine
organisms, Proc. Roy. Soc. (Lond.) Series B. 177,
115-136.
Bryan GW and Hummerstone LG (1977) Indicators
of heavy-metal contamination in the Looe estuary
(Cornwall) with particular regard to silver and
lead, J. mar. BioI. Ass. U.K. 57, 75-92.
Butler PA, Andren L,Bonde G, Jernelev A and
Reish DJ (1971) Monitoring organisms. IN FAO
Technical conference on Marine Pollution and its
Effects on Living Resources and Fishing. Rome
1970. Supplement 1: Methods of detection,
measurement and monitoring pollutants in the
marine environment, pp. 101-112. Rome, FAO.
Cloete CE and Watling RJ (1981) South African
marine pollution monitoring programme 1979-1982,
Pretoria, National Programme for Environmental
Sciences Report No 51.
313
Cuthbert KC, Brown AC and Orren MJ (1976)
Cadmium concentrations in the tissues of Bullia
digitalis (Prosobranchiata) from the South
African west coast, S. Afr. J. Sci. 72, 57.
Darracott A and Watling HR (1975) The use of
molluscs to monitor cadmium levels in estuaries
and coastal marine environments, Trans. Roy. Soc.
S. Afr. 41, 325-338.
Day JH (1969) A guide to marine life on South
African shores. Cape Town, AA Balkema.
De Villiers G (1975a) Reproduction of the white
sand mussel Donax serra Reding. Cape Town,
Department of Industries, Sea Fisheries Branch
Investigational Report No 102.
De Villiers G (1975b) Growth, population
dynamics, a mass mortality and arrangement of
white sand mussels Donax serra Reding on beaches
in the southwestern Cape Province. Cape Town,
Department of Industries, Sea Fisheries
Investigational Report No 109.
Dills G and Rogers DT (1974) Macroinvertebrate
community structure as an indicator of acid mine
pollution, Environ. Pollut. 6, 239-262.
Emmerson WD and Watling RJ (1983) Effects of
two sewage outfalls in Algoa Bay, Water SA 9:
(in press).
Goldberg ED, Parker PL, Bowen VT, Risebrough RW,
Farrington JW, Robertson W, Harvey G, Schneider E,
Martin JH and Gamble E (1978) The mussel watch,
Environ. Conserv. 5(2), 101-125.
Halcrow W, Mackay DW and Thornton I (1973) The
distribution of trace metals and fauna in the
Firth of Clyde in relation to the disposal of
sewage sludge, J. mar. bioI. Ass. U.K. 53, 721-
739.
Haug A, Melsom Sand Omang S (1974) Estimation
of heavy metal pollution in two Norwegian fjord
areas by analysis of the brown Alga Ascophyllum
nodosum, Environ. Pollut. 7, 179-192.
Hornung H and Oren OH (1981) Heavy metals in
Donax trunculus L. (Bivalvia) in Haifa Bay,
Mediterranean (Israel), Mar. Environ. Res. 4,
195-201.
Majori L and Petronio F (1973) Marine pollution
by metals and their accumulation by biological
indicators (accumulation factor), Rev. into
Oceanogr. Med. 31-32, 55-90.
McLachlan A (1977a) Studies on the psammolittoral
meiofauna of Algoa Bay, South Africa. II. The
distribution, composition of the meiofauna and
macrofauna. Zool. Afr. 12: 33-60.
McLachlan A (1977b) Composition, distribution,
abundance and biomass of the macrofauna and meio-
fauna of four sandy beaches, Zool. Afr. 12, 279-
306.
McLachlan A (1980) Intertidal zonation of macro-
fauna and stratification of meiofauna on high
energy sandy beaches in the Eastern Cape, South
Africa, Trans. Roy. Soc. S. Afr. 44, 213-223.
McLachlan A, Cooper C and Van der Horst G (1979)
Growth and production of Bu~lia rhodostoma on
an open sandy beach in Algoa Bay, S. Afr. J. Zool.
14, 49-53.
314

McLachlan A, Dye AH and Van der Ryst P (1979)

Vertical gradients in the fauna and oxidation of
two sandy beaches, S. Afr. J. Zool. 14, 43-47.
'McLachlan A, Erasmus T, Dye AH, Hoodridge T,
Van der Horst G, Rossouw G, Lasiak TA and McGwynne
L (1981) Sandy beach energetics: an ecosystem
approach towards a high energy interface, Estuar.
Cst1. Shelf Sci. 13, 11-25.
McLachlan A and Hanekom N (1979) Aspects of the
biology, ecology and seasonal fluctuations in
biochemical composition of Donax serra in the
Eastern Cape, S. Afr. J. Zoo1. 14~ 183-'193,
McLachlan A and Van der Horst G (1979) Growth
and reproduction of two molluscs from an exposed
sandy beach, S. Afr. J. Zool. 14, 194-201.
McLachlan A, Wooldridge T and Dye AH (1981) The
ecology of sandy beaches in Southern Africa, S.
Afr. J. Zoo1. 16, 219-231.
McLachlan A, Wooldridge T and Van der Horst G
(1979) Tidal movements of the macrofauna on an
exposed sandy beach in South Africa, J. Zool.
(Lond.) 188, 433-442.
Phillips DJH (1977) The use of biological
indicator organisms to monitor trace metal
pollution in marine and estuarine environments -
a review, Environ. Pollut. 13, 281-317.
Portmann JE, Mandelli E, Pentreath RJ, Lee RF,
Addison RF, Jensen S, Reish DJ and Yoshida T
(1975) Draft outline of the guideline manual on
the use of bioaccumulators. FAO Fisheries Report
No 160. Annex V, pp 9-18.
Reish, DJ (1970) A critical review of the use
of invertebrates as indicators of varying degrees
of marine pollution, FAO FIR: MP/70/R9.
Thornton I, Watling Hand Darracott A (1975)
Geochemical studies in several rivers and
estuaries used for oyster rearing, Sci. Tot.
Environ. 4, 325-345.
Van As D, Fourie HO and Vleggaar CM (1973)
Accumulation of certain trace elements in marine
organisms from the sea around the Cape of Good
Hope. In Radioactive contamination of the marine
environment, proceedings of a symposium, Seattle,
1972.pp 385-401, Vienna, IAEA.
Van As D, Fourie HO and Vleggaar CM (1973) Trace
element concentrations in marine organisms from
the Cape west coast, S. Afr. J. Sci. 71, 151-154.
Watling HR and Watling RJ (1979) Metal
concentrations in Perna perna from the Southern
African coast. S. Afr. J. Sci. 75, 371-373.
Watling HR and Watling RJ (1982) Metal
concentrations in oysters from the Southern
African coast, Bull. Environ. Contam. Toxicol.
28, 460-466.
Watling RJ (1978) The use of a slotted tube
for the determination of lead, zinc, cadmium,
bismuth, cobalt, manganese and silver by atomic
absorption spectrometry, Anal. chim. Acta 97,
395-398.

SANDY BEACHES AS ECOSYSTEMS CHEMICAL ASPECTS - WORKSHOP REPORT
D.A. LORD1and G.A. EAGLE 2 (lCooperative Scientific Programmes, C.S.I.R. P.O. Box 395, Pretoria, South
Africa, 2National Research Institute for Oceanology, P.O. Box 320, Stellenbosch, South Africa)
The review paper by Dr. G. Eagle emphasized that
the symposium had selectively directed itself to
high energy beaches which were unpolluted, and
therefore well oxygenated. Consequently, the
important chemical considerations in such beaches
are the reactions of major nutrients, carbon,
nitrogen and phosphorus. The workshop was
designed to complement this concept, and to
investigate further specific items which had
arisen and which were considered to be of
importance concerning the chemistry of sandy
beaches. To accomplish this, four discussion
questions were posed, these being:
1. Are analyses for the right chemical moeities
being conducted?
2. Are we able to look at individual classes of
organic materials in sandy beach ecosystems?
3. What is the flux of nutrients returned to
the sea from beaches?
4. What laboratory models can be used to
simulate chemical processes in sandy beaches?
1. Are analyses for the right chemical species
being conducted?
Considered here were specifically the compounds
of nitrogen and phosphorus. Carbon compounds
were to be addressed in question 2. Also in~
cluded in the concept of 'analysis' were
techniques used for the collection and
preservation of samples, as well as the inter-
pretation of analytical results.
1.1 Sampling
Sampling techniques and strategies received
315
limited discussion. It was recognized that, in
the sand column of a beach with its inherent
discontinuities, microniches do occur where
substantial gradients in parameters such as pH
and pE will exist. Sampling in the sand column
particularly in interstitial waters, must
recognize these variables.
1.2 Analysis
Analysis of the nutrients nitrogen and phosphorus
is becoming more standardized although different
techniques are still in use. For nitrogen, the
analysis for nitrate (and nitrite) is now well
standardized (although spurious results for
Kjeldahl nitrogen were reported). Analytical
procedures for ammonia are standard; however,
sample collection and preservation techniques
are critical. Little is currently known on the
organic forms of nitrogen, and their distribution.
For phosphorus, the various forms in which this
element can exist complicates analytical require-
ments.
1.3 Interpretation
At this stage in the study of sandy beaches the
interpretation of nutrient data from sandy
beaches has been limited to comparing total
concentrations of specific chemical forms.
Little is known about the rates and equilibria
involved in the various reactions. There are
two important considerations complicating this
type of study. Firstly, in the surf zone,
nutrient levels fluctuate widely, with pulses of
316
higher concentrations probably having
significant ecological influence. Under such
conditions a description of the system using
mean values is limited as a means of describing
the functioning of the system, as well as the
productivity of such a system.
Secondly, it is well accepted that both nitrogen
and phosphorus can be rapidly recycled (as can
other nutrients) such that a plankton bloom can
be sustained for a considerable period of time
with the introduction of minimal amounts of these
nutrients. Consequently, at any time, the
amounts of these 'free' materials in the water
column may be very small, despite the importance
of the material in sustaining the productivity.
2. Are we able to look at individual classes of
organic materials in sandy beach ecosystems?
Traditionally, measurements involving organic
carbon concentrate on the separation of dissolved
and particulate material at a specific size
(namely 0.45 ~m), followed by a measurement of
the total organic carbon content of each fraction.
These .methods still only result in total concen-
trations of specific chemical compounds being
measured and do not account for interconversion
between, for example, organic and inorganic
carbon.
To date, virtually no work has been reported on
the nature of this organic material with regard
to sandy beaches. For other marine ecosystems
(such as kelp beds), it has shown that techniques
for the extraction, isolation and identification
of specific organic compounds such as mannitol
are not adequately developed for quantitative
assessment of these materials. However, the
problem can be partially overcome by isolating
and enumerating bacteria which utilise the
specific organic compounds of interest.
3. What is the flux of nutrients returned to
the surf zone from beaches?
There was some disagreement as to whether the
flux of nutrients from the beach back to the surf
zone is significant. On the one hand it is felt
that beach ecosystems process large amounts of
organic material, and return considerable
quantities of nutrients to the surf. Other
delegates on the other hand felt that this model
is incorrect and that nutrient flux from beach
to surf contributes an insignificant amount to
coastal zone requirements.
This question was ultimately considered too
limited in extent. The flux of nutrients to the
sea is dependant on the input of nutrients to the
beach from both the surf zone and from groundwater,
and on the retention time of the water percolating
through the beach.
It was felt that this was a question requiring
further research.
4. What laboratory models can be used to
simulate chemical processes in the beach?
Laboratory models can be extremely useful for
studying chemical processes occurring in sandy
beaches, although laboratory models have limitations
associated with them. However, provided these can
be accounted for either in the design or inter.-
pretive phase, clearly their uses are substantial.
Discussion on the use of sand columns concluded
that these are normally too simple to replicate
properly all sandy beach processes of importance,
particularly the lateral flow of water. However,
strong support exists for their use in certain
experimental situations as it is possible to
control accurately many of the physical variables
within these columns; such that the use of a
large number of columns in multi-variate
experimental programmes can be of considerable
value. Using sand columns it is relatively easy
 317

to control the important features of grain size

distribution, and tidal cycles. Results obtained
from some of the original sand column work where
the sand columns remained inundated should be
viewed with caution.

Models of greater complexity can be devised, where

for example, wave action is included. In models
of this kind, a major concern is to scale features
such as wave movement and boundary effects. Of
particular importance for chemical studies is the
proper representation of sand porosity. In
addition, dynamic models of this type must ensure
that a sufficiently deep layer of sand is available
(equivalent to at least 30% of wave height) to
ensure that spurious boundary effects are not
generated within the model.

Models could, in fact, be made even more complex,

simulating more of the natural processes.
However, the point will be reached when it is
more practical to work on the beach itself.
 319

PART THREE

ECOLOGY
 321

SANDY BEACH ECOLOGY A REVIEW

ANTON McLACHLAN (Zoology Department, University of Port Elizabeth, P.O. Box 1600,
Port Elizabeth 6000, South Africa)

1. INTRODUCTION open to the sea. This does not include

Sandy beaches dominate most temperate sand flats in estuaries or closed
and tropical coastlines where they lagoons but only open marine beaches.
represent both important recreational Nevertheless, such beaches may differ
assets and buffer zones against the considerably in their degree of exposure
sea. In some areas they are very to wave action. On a 20 point exposure
productive and are exploited scale, developed for studies on
commercially. However, they have been intertidal fauna (HcLachlan, 1980a) ,
regarded as marine deserts by many beaches scoring 11 to 18 are considered
biologists and were largely neglected exposed and are characterised by
until Remane -(1933) began studies on the continuous, often heavy, wave action,
coasts of Germany. The work of Pearse the absence of silt, a mobile fauna and
et al (1942) was also pioneering and a high degree of oxygenation of the sand.
represented the first qualitative Sheltered beaches (scores 5-10) have
attempt to evaluate a whole beach less, often intermittent wave action,
system. Since then sandy beach ecology fauna which may construct semi-permanent
has advanced considerably, though it has bur rows and reduced layers present and
always lagged behind other aspects of sometimes close to the surface.
coastal marine ecology in the attention Beaches scoring belm; 5 points on this
it has enjoyed. During this period rating scale are not open ocean beaches
biological research on beaches has and fall outside the scope of this
spread from early taxonomic and review. vlright, Short (1982) have
qualitative studies through quantitative developed a six point scale classifying
ecology and physiology of important beaches from dissipative to reflective
species towards a more holistic systems extremes based on morphodynamics.
approach today.
As the sandy beach occupies a dynamic
The term sandy beach has been loosely interface position between sea and land,
used in the literature to cover a range its boundaries with the adjacent
of environments from high energy open terrestrial and marine environments are
ocean beaches to extremely sheltered not always clear and the functional
estuarine sand flats. For the purposes extent of the beach itself has seldom
of this review a sandy beach is been discussed. 'l'he sand beach is
considered to be a sandy littoral area considered here to be part of a system
322

comprising i) the sand uody from the been written on sediments, wave action
highest drift line near the dune/beach and relationships between sedimentary
boundary out to beyond the break point parameters, beach slope and wave action,
of the waves and ii) the moving water considerably less attention has ueen
envelope of the surf zone to the outer focused on the important processes
limit of surf circulation cells where relating to water movement in the surf
they exist. This dynamic environment, zone. 1'he significance of surf
which I call the beach/surf zone circulation cells, rip current systems
ecosystem, is of great extent in both etc., has still to be realized in ueach
tropical and temperate areas. It does ecological studies.
not include the dunes typical of many
sandy coasts, although dune/beach Biological work to date has concentrated
interactions are important. Its on the fauna of the intertidal sand uody
landward boundary is clearer than its and most other aspects of ueach ecology,
seaward boundary where, in exposed including the surf zone as a whole, have
situations, surf circulation cells, ueen neglected. The intertidal sand
driven by breaking waves, mix with body of open sandy ueaches is
inshore water. The basic topography and characterized uy a mobile sUbstratum and
features of this system are illustrated the absence of attached plants.
in Fig. 1. Normally the substratum provides two

DUNES
r----------=~~~----------_"ri--~~~~---,i
BEACH NEARSHORE
------------------~~~~~~--------------_,
.1

ST ABLE DUNES
BACK FORE INNER OUTER
ACTIVE DUNES SHORE SHORE TURBULENT ZONE TURBULENT ZONE
r------J~----~'~~rl------~~~--------llr-----------~A------------~

FIGURE 1. Profile of a typical sandy beach environment, showing areas referred to in

the text.

The physical featUres of this system habitats suitable for faunal

will not be examined here. Besides population. These are i) the macrofauna
other papers in this volume, descript- habitat consisting of the sand surface
ions of physical
features of beaches and upper layer of sediment and ii) the
important to ecologists may be found in interstitial habitat consisting of the
many publications, including Bascom porous system of the sand uody. The
(1964), King (1972), Davies (1972) and former, particularly on exposed ueaches,
Inman, Brush (1973) . Though much has is a uniform but dynamic and unstable,
 323

essentially two dimensional system. The environment. Thus although meiofauna-

latter is a more stable and complex macrofauna interactions may occur
habitat with clear vertical gradients (Hummon et al. , 1976; Reise, 1979)
and is essentially a three dimensional these are generally of negligible
system (McLachlan, 1977a). It has great significance in terms of energy flow on
vertical extent on exposed beaches or exposed sandy beaches.
high energy 'windows' where strong
hydrodynamic forces introduce oxygen The only review on sand beach ecology as
deep into the sediment. Elsewhere its a whole is that of Hedgpeth (1957),
base consists of reduced layers although some aspects of beach ecology
(Fenchel, Riedl, 1970). are covered uy Eltringham (1971). Both
these works are outdated and there is a
The macrofauna of sandy beaches and the distinct need for a synthesis of
meiofauna (and microfauna) of their information on sandy beaches. The aim
interstices comprise two entirely of this review is to outline our
separate faunal components with knowledge of the ecology of individual
virtually no overlap or exchanges of components of sandy beach biotas and the
energy. This is because the meiofauna structure and function of beach
are extremely small interstitial forms ecosystems as a whole. It progresses
while the macrofauna are several orders through the interstitial fauna and
of magnitude larger (McIntyre, 1971; macrofauna of the sand body,
McIntyre, Murison, 1973; McLachlan, phytoplankton, zooplankton and fishes of
1977a) . Schwinghamer (1981) disting- the surf zone and birds of the beach and
uished three biomass peaks in intertidal dune margin to functional aspects within
sediments corresponding to bacteria, the framework of energy flow and
interstitial meiofauna and macrofauna. nutrient cycling.
As sediments become finer below 200 urn
mean grain size, large burrowing forms 2. THE INTERSTITIAL SYSTEM
of meiofauna become increasingly The porous system between the sand
important and tend to br idge the size grains generally averages about 40% of
gap between macrofauna and meiofauna. the total sediment volume. Most
This makes separation into two physical properties of this system are
components less marked in finer directly determined by the sediment
sediments and facilitates better energy properties which in turn are related to
flow between these components. Energy the wave and current regime as well as
flow between these components also the geological history of an area.
increases because the meiofauna are Grain size, shape and sorting are most
concentrated near the surface and important in fixing porosity and
macrofauna deposit feeders increase in permeability which influence drainage.
abundance in finer sediments. In Drainage is critical in determining the
coarser sediments such intermediate moisture content, oxygen and organic
forms are absent, probably because they input and the depth of 'reduced layers.
represent a size range selected against permeability increases with coarser
by the physical forces of the beach substrate and better sorting and
324

drainage also increases on steeper

beaches. ']'he interactions between these
factors are well documented and covered
in several papers (Webb, 1958; Hulings,
Gray, 1971; Crisp, Williams, 1971).

The major process involving the

interstitial system on open beaches is
the filtration of sea water. 'J:'hi s water
is introduced into the sediment either
by flushing by waves and tides in the
intertidal (Riedl, 1971; Riedl, Machan,
1972 ; McLachlan, 1979a) or the pumping
effects of waves in the subtidal (Steel
et al., 1970a; Riedl et al., 1972). In
exposed situations filtered volumes may o Yellow layer" oxygenated sand
• Gray layer = RPD zone
average 5-10 m3 per metre strip of • Black layer = reduced sand

intertidal beach per day and be several FIGURE 2. Schematic representation of

times on very steep, coarse grained vertical chemical gradients in a
beaches. In the subtidal surf zone off shel tered and an exposed beach. (After
open beaches wave pumping may filter Fenchel, Riedl, 1970; McLachlan et al.,
0.05-5.00m 3 m- 2 .d- l . This flushing 1979a)
or pumping concentrates dissolved and
particulate organics in the sand. Under moisture content and surface temperature
high energy conditions, however, the (Pollock, Summon, 1971; 11cLachlan et
input of oxygen and the high drainage al. , 1977a) while in very sheltered
rates maintain this system fully situations it can result in sharp
oxygenated and there are no steep changes in chemical gradients, for
vertical gradients in oxygen and example pH changes coupled to
oxidation states of sulphur or nitrogen temperature changes (Gnaiger et al. ,
(Riedl, MacMahan, 1969; McLachlan et 1978). In sheltered beaches with
al., 1979a). permeability and flushing reduced layers three vertical zones
generally decrease with increasing (Fig. 2) may be uistinguished, namely
shelter until drainage is so poor that clean (oxygenated) sand, a gray (or RPD)
the sand is constantly saturated. transition zone and black (reduced)
vertical chemical gradients are then conditions (Fenchel, Riedl, 1970) .
steep and reduced layers occur close to Horizontal zones may also be distin-
the surface (Fig. 2). guished down the beach from the
relatively constant conditions of the
The interstitial system is subject to subli ttoral to the backshore with large
cyclic changes related to tidal and temperature and salinity fluctuations
diurnal cycles and the seasons. In high (Salvat, 1964, 1967; Schmidt, 1970;
energy beaches this mainly results in Mielke, 1976). On exposed beaches,
fluctuations in the water table, pore because of its greater vertical extent
 325

and better drainage, the interstitial

system lends itself to subdivision into
strata. pollock, Hummon (1971 )
recognized five strata in a coarse
grained (Hd 660 urn) Hassachusetts
beach which they could relate back to
Salvat's (1964) horizontal zones.
HcLachlan (1980b) distinguished four
strata on East Cape beaches (Hd 250
urn) which closely parallel those of
Pollock and Hummon. In both cases the
strata range from surface dry sand at
the top of the beach down through moist FIGURE 3. Stratification of intertidal
layers to the permanently saturated zone sand based on interstit'ial moisture
below the water table (Fig. 3 and Table content during low tide for a beach in
1) • Hassachusetts and the East Cape, South
Africa. (After Pollock, Hummon, 1971;
McLachlan, 1980). See also Table 1.

Table 1. Summary of stratification of exposed sandy beaches based on interstitial

water content (Hodified from pollock, Hummon 1971; HcLachlan, 1980). See also Fig.3.

Moisture Content & Water Degree of

Zone/Stratum Circulation Oxygenation Temperature

Zone of dry sandi LOW, loses capillary water High Highly variable
Dry sand stratum

Zone of drying/ LOW, loses capillary water High Variable

Dry sand stratum

Zone of retention/ Loses gravitational water but High Little variation

Hoist sand stratum retains capillary water

Zone of resurgence/ Gravitational water from zone Moderate Stable

Water table stratum of retention drains through to low
here during ebb tide

Zone of saturation/ permanently saturated, little Low to Very stable

Low oxygen stratum water circulation very low
326

Large numbers of microscopic organisms cells per gram dry sand and increase
occupy the interstices: fungi, algae, with finer sediment and greater surface
bacteria, protozoans and metazoans. area (Dale, 1974; Meyer-Reil et a1.,
Amongst the animals, metazoans that pass 1978; Mazure, Branch, 1979). Khiyama,
undamaged through a lmm screen are Makemson (1973) recorded 362 bacterial
considered meiofauna and the rest strains in beach sand, many associated
microfauna. Individual dry weights are symbiotically with metazoans. Those on
10-5_ l0 -8 g for meiofauna, 10- 6 sand grains made up 54-78% of the total
-lO-l1 g for protozoans and 10- 10 numbers of 10 7 -10 8 per gram of wet
_1012 g for bacteria (Fenchel, 1978; sand. Hicks (1974) demonstrated the
Warwick, pers. comm.) Fungi in sandy presence of Azotobacter in marine sand
beaches are generally concentrated at beaches while Rheinheimer (1977)
higher tide levels and have been little isolated 31 strains from beach sand, of
studied (Upadhyay et a1., 1978). A which the majority showed salinity
flora of diatoms or flagellates only optima at 30-50 0 / 00 and \{ere gram
develops in sheltered beaches and is negative rods that could decompose
absent on very exposed beaches (Steel et proteins, carbohydrates and fats.
a1., 1970a; Brown, 1971a). This also Andrews et al., (1976) studied microbial
appears to be more characteristic of development in a model sandy beach,
temperate than tropical beaches (Munro recording greatest abundance near the
et a1., 1978). In sheltered sands the high and low tide marks.
diatom flora may be locally important
and, due to mixing by wave action, may Organic carbon and nitrogen in the sand
occur to some depth in the sand (Steele, have been found to correlate with
Baird,1968; MCIntrye,1977; Anderson, bacterial biomass and abundance (Dale,
Meadows, 1978) • Munro, Brock (1968) 1974; Anderson et al., 1981; Bolter et
recorded numerous living bacteria and al., 1981) which can be up to four times
diatoms attached to sand grains up to the biomass of the meiofauna
10cm below the sand surface and Round (Meyer-Reil, Faubel, 1980). Munro et
(1979) described a diatom assemblage al. (1978) estimated microbial
living below the sand surface. Amspoker production at l5gc.m- 2 for a temperate
(1977 ) described the intertidal beach and 72gc.m- 2 for a tropical
distribution of epipsammic diatoms on a beach, the higher latter value due to
Californian beach. Where sand is greater water percolation. fiJazure,
permanently saturated diatoms generally Branch (1979) estimated an average
occur closer to the surface. Pearse et bacterial biomass of l4.2gm- 2 dry
al. , (1942) recorded Chrysophyta, weight in the sands of Langebaan
Chlorophyta, pyrrophyta and Cyanophyta Lagoon. Dye (1979a) showed that
on sandy beaches. disturbance caused greatly increased
bacterial activity in beach sand.
Bacteria are abundant and important in McLachlan et al., (1979a) recorded
beach sediments, mostly being attached bacterial numbers around 10 8 per gram
to the sand grains (Meadows, Anderson, dry sand down to more than 1m in the
1966, 1968). Numbers range 10 8 _10 10 sediment of an exposed beach. Dye
 327

(l980a) was able to show tidal 1977; Fenchel, 1978; Hartwig, 1982).
fluctuations in oxygen uptake by beach Small protozoans have also been
sand. Maximum uptake coincided with enumerated on exposed beaches in the
maximum water percolation through the Eastern Cape (Dye, 1979b; McLachlan et
sand and maximum fluctuation occurred at al., 1979a; Dye, 1981). Their numbers
higher tide levels. Desiccation of the 1
ranged 10 -10 3 per gram dry sand and
sand caused a severe drop in oxygen they were found to more than 1m depth in
uptake or bacterial activity. Dye a well oxygenated beach. Their
(1981) also attempted to partition total contribution to total benthic oxygen
benthic oxygen uptake on exposed East consumption was significant, accounting
Cape beaches and concluded that bacteria for 15-25% of interstitial oxygen uptake.
were responsible for most of the oxygen
co~umption except where there was a Meiofauna have received considerable
very rich macrofauna. Meyer-Reil et attention on sandy beaches and several
al. , (1980) estimated microbial turn- reviews cover aspects of meiofauna
over time of 100h with daily microbial ecology (Delamare-Deboutteville, 1960;
carbon production of 43mg.m- 2 in Swedmark, 1964; Fenchel, Riedl, 1970;
sheltered beaches in the Baltic. They Hieser, 1975; Giere, 1975; Fenchel,
concluded that 50% of microphytobenthos 1978) .
primary production was fixed by micro-
organisms. Koop, Griffiths (1982) found Small metazoans occurring on sandy
a high bacterial biomass on an exposed beaches may be either temporary
beach receiving high kelp input on the meiofauna, i.e., larval forms of
west coast of South Africa, bacteria macrofauna, or permanent meiofauna. As
being found to 1.2 m in the sediment. we are mainly dealing with interstitial
Koop et al., (1982a, b) estimated that meiofauna here, the temporary component
bacteria were responsible for over 90% is generally not of importance on open
of carbon utilization of these kelps. beaches. 'J'he dominant components of
Annual turnover (P/B) of bacteria in sandy beach meiofauna are nematodes and
this system was estimated to be 30 times harpacticoid copepods with several other
(Koop, Griffiths, 1982; Stenton-Dozey, taxa of var iable importance. These
Griffiths, 1983). Despite this include turbellarians, oligochaetes,
information we still know nothing about ostracods, mystacocarids, gastrotichs,
the ecology of individual species of halacarid mites, tardigrades, gnathosto-
micro-organisms on sandy beaches. mulids, hydrozoans and bryozoans. There
is a well established relationship
All groups of protozoans may be present between the relative proportions of
in the interstitial system but ciliates nematodes and harpaticoids and grain
and foraminiferans have been best size. Nematodes tend to dominate in
studied as they are usually relatively finer sediments, harpacticoids in
large and quite abundant, especially on coarser sediments and in sediments wi th
sheltered beaches with fine sands mean grain size around 3,00-350 urn they
(Swedmark, 1964; Fenchel, Jansson, 1966; are about equally important (Gray, 1971;
panikkar, Rajan, 1970; Hartwig, parker, McLachlan et al., 1981b; Raffaelli,
328

Mason, 1981; Hockin, 1982). Fricke, related to differing tolerances to

Fleming (1983) have demonstrated reducing conditions (Fenchel, Riedl,
experimentally the neamtode preference 1970). vertical distribution is
for finer sediments. generally shallower in the surf zone
than on the beach but high numbers may
In sands above 200 urn the meiofauna is still occur 10-30 crn into the sediment
usually entirely interstitial while (NcLachlan et al., 1977b). Dean (1981)
below 200 urn burrowing forms become looked at the effects of wave action on
increasingly important due to pore size meiofauna abundance.
restrictions (Wieser, 1959). Inter-
stitial harpacticoids may occur down to Horizontal distribution may take the
160 urn (McLachlan et al., 1977b; Moore, form of layers, horizontal zones,
1979a) while nematodes can pursue an bathymetrical steps and geographical
interstitial mode of life down to regions (Fenchel, Riedl, 1970).
125-100 urn (Wieser, 1959; MCIntyre, Horizontal zones of meiofauna
Murison, 1973; Fenchel, 1978). In distribution have been described from
noncapillary sediments nematodes some beaches, particularly under
dominate (Fenchel, 1978). As most open relatively sheltered conditions (Coull,
sandy beaches have grain sizes in the 1970; Schmidt, 1972a, b; Harris,
range 200-500 urn, interstitial nematodes 1972a; Moore, 1979b) where the
and harpacticoids are almost always meiofauna is close to the surface and
dominant. MCIntyre, Murison (1973) the distribution tends to take on a more
considered sands of 230 urn to be optimum two-dimensional character (Fig. 4).
for the development of interstitial
fauna while Gray, Rieger (1971 ) Seasonality has been documented in
suggested that meiofaunal diversity several cases in temperate areas with
increases as sands get finer. In South the meiofauna occurring in lower
Africa richest meiofauna occurs in sands abundance and moving deeper into the
of 250-350 urn (McLachlan et al., 1981b). sediment in winter (Renaud-Debyser,
1963; Ha r r is, 1 9 7 2b ; Schmidt, 1972a,
On a vertical basis meiofauna distri- b,; Hulings, 1974; Mielke, 1976;
bution is related to the degree of Faubel, 1976; Nodot,1976; Feder,
drainage and oxygenation of the sediment paul, 1980). In warmer areas
(see Fig. 3 and Table 1). Meiofauna seasonality is less clear (McLachlan,
abundance drops off drastically in the 1977b) and in sublittoral areas it may
RPD and reduced layers (Fenchel, Riedl, be more complex (Emberton, 1981).
1970; McLachlan, 1978) and a deep Vertical migrations other than seasonal
vertical distribution therefore only have been recorded as response to
occurs on exposed beaches where reduced factors such as heavy rains (Bush,
layers are absent (Renaud-Debyser, 1966), wave disturbance (Boaden, 1968),
1963; Fenchel, Riedl, 1970; McLachlan tidal factors (Rieger, Ott, 1971;
et a1., 1979a) (see also Fig. 2). Meineke, Westheide, 1979) and changes in
Different species and taxa show well moisture and oxygen over the tidal cycle
defined vertical distribution patterns (McLachlan et al., 1977a).
 329

(a) SHERTALLAI BEACH, INDIA

~ Arenopontia orienta lis }
illIill Arenopontia acantha INTE~~~ITIAL
FORMS
1m Psammopsyllus operculatus
'WAT~J'+~i~~iii_~
(b) A TYPICAL MANX BEACH
3 EURYHALINE UPPER SHORE SPECIES (ALL INTERSTITIAL)
3 EULITTORAL SPECIES (1 EPIBENTHIC
2 INTERSTITIAL)

1 6 SUBLI.TIiT.OiiRA.LiliFiiR.IN.G.E.SPECIES
l~=q.I~~ms;00~~::~;;~(EPIBENTHIC)

FIGURE 4. Comparison of vertical and horizontal distribution patterns of

harpacticoid copepods in a well drained, exposed beach (a) and a sheltered beach (b)
where distribution patterns tend to be very three-dimensional and rather two-·
dimensional respectively. (After (a) Munro et al., 1978 and (b) Moore, 1979b). In
all cases the copepods are concentrated above the low tide water table. On the
sheltered beach interstitial fauna are mostly concentrated high on the shore where
some drainage occurs. Nevertheless zonation in this case is still mostly across the
shore and not vertically in the sediment.

seasonality are becoming clearer as a feeders, non-selective deposit feeders,

result of ecophysiological studies on diatom feeders, carnivores and even
responses, preferences and tolerances of filter feeders. Various studies have
important meiofaunal species (Jansson, shown aggregation of meiofauna in
1967, 1968a; Wieser et al., 1974; response to specific bacterial strains
Wieser, 1975; Hartwig et al., 1977; (Gray, Johnson, 1970) or where there are
Wieser, Scheimer, 1977). high levels of organics generally or
localized small scale concentrations
A wide range of food items has been (Lee et al., 1977; McLachlan et al.,
recorded for meiofauna including 1977b; Moore, 1979a; Hogue, Miller,
diatoms, bacteria, protozoa, detritus, 1981 ) or concentrations of prey
other meiofauna and dissolved organics (Fenchel, 1978). Feeding rates have
(Jansson, 1968b; MCIntyre, 1969; been estimated directly for a copepod
Fenchel, 1978; Giere, 1975, 1982) . (Rieper, 1978). There have also been
Feeding categories include bacterial studies on growth, production and
330

population dynamics of important species supply adequate dissolved and

(pawlak, 1969; Lasker et a1., 1970; particulate organic food. Meiofauna are
Hall, Hessler, 1971; McLachlan 1977c; generally considered to subsist on the
Feller, 1980). Most species have year microfauna which in turn are largely
round reproduction with generation times fuelled by dissolved and particulate
ranging 1-3 months (McIntyre, 1969; organics flushed into the sand (McIntyre
Lasker et al., 1970; Giere, 1975; et a1., 1970). Measurements of benthic
Bartsch, Schmidt, 1979). metabolism have attempted to partition
oxygen uptake between these different
Over the past decade ecological studies components of the interstitial fauna
on sandy beach meiofauna have advanced (Munro et al., 1978; Dye, 1980a; 1981).
from the more general surveys of the Rates of mineralisation of organics by
sixties. More attention has been paid the interstitial system have also been
to the complex of factors composing the measured in experimental sand columns in
interstitial system and results have an attempt to understand these processes
been better analysed statistically better and estimate nutrient cycling
(Gray, Rieger, 1971; Hulings, 1971; (McIntyre et al. , 1970; Boucher,
Hulings, Gray, 1976; McLachlan, 1978; Chamroux, 1976; McLachlan et al., 1981c).
Hogue, Miller, 1981). The basic pattern
of meiofaunal response to different Vernberg, Coull (1974) estimated the
sediment types is well documented and ratio of metabolic activities of
responses to chemical features of the ciliates to Hleiofauna to macrofauna as
interstitial system are clearer. More 1:0.5:2.1 for a sandy sublittoral
work is needed on biotic interactions sediment and 1:0.54:0.04 for an exposed
including competition, feeding and beach. Dye (1981) partitioned benthic
predation. metabolism between microfauna, meiofauna
and macrofauna on two exposed beaches in
With regard to energetics and benthic the ratios 1:0.23:0.04 and 1:0.27:1.85,
metabolism the interstitial system has the latter beach having an exceptionally
still largely been regarded as a black rich macrofauna. /llcLachlan et al.,
box and few authors have progressed (1981a) partitioned benthic metabolism
beyond the estimation of abundance or on East cape beaches amongst the
biomass. On exposed beaches, and in interstitial fauna and macrofauna as
most other sediments, meiofaunal numbers 1:0.4. On a west Cape beach receiving a
fluctuate within an order of magnitude high kelp input, biomass ratios of
of 106m-2 and dry biomass values macrofauna: meiofauna: bacteria were
range 20-4400 mg. m- 2 . Some figures 3.5:1:1.5 (Griffiths et a1., 1983) while
are compared in Table 2. these ratios for productivity were
1: 1: 4.7.
Clearly biomass tends to be higher in
the intertidal on exposed beaches It may be concluded that our knowledge
because the great vertical extent of of the interstitial system of open sandy
these systems provides greatest habitat beaches has advanced considerably over
availability and high percolation rates the past two decades. He have a ':lood
 331

Table 2. Abundance and dry biomass values for meiofauna recorded from marine sands.

Source Locality Abundance Biomass

vlieser (1960) Buzzards Bay sublittoral 100 - 600 mg.m- 2

106. m-2

Coull (1970) Bermuda sublittoral 33 - 259 mg.m- 2

106. m-2

McIntyre, Murison (1973) South Africa exposed 279 - 1 092 mg.m- 2

intertidal
106. m-2

McLachlan (1977a, b) South Africa exposed 20 - 3 360 mg.m- 2

intertidal
l06. m-2

McLachlan et al., (1977b) South Af rica subli t toral 350 - 1 045 mg.m- 2
106. m-2

Fenchel (1978) sublittoral sand 50 - 1 700 mg.m- 2

105_ l0 6. m-2

McLachlan et al., (1981b) South Africa exposed 400 - 4 000 mg.m- 2

intertidal
10 6 . m-2

Stenton-Dozey, Griffiths South Africa exposed 1 000 - 44 000 mg.m- 2

(1983 ) intertidal
106_ l0 7. m-2

general idea of the distribution patterns of meiofauna and most other

patterns of interstitial fauna in interstitial processes in intertidal
relation to many features of inter- sands (e.g. Figs. 3 and 4a). This water
stitial climate. However, in this flow is the superparameter controlling
respect the excellent work of Riedl and interstitial climate and the distri-
co-workers in describing water flow bution of interstitial fauna in beach
through beach sand has not received the sands. Both llIeiofauna and bacteria
attention it deserves. The patterns of therefore tend to concentrate in the
water input, percolation and pulsing moist sand at and above NTL and below
disturbance described by Riedl, Machan the sand surface where there is high
(1972) (Fig. 5) add a considerable water input, percolation and aeration,
degree of comprehension to distribution minimum stagnation and pulsing currents
332

are not too severe. Under sheltered all organic materials it receives,
conditions the aerated strata become whether these be dissolved or
much flatter, the interstitial fauna particulate. In this process bacteria
concentrates close to the sand surface are most important, with meiofauna
and the overall dimensions of the system mostly at the top of the interstitial
take on a more two dimensional nature. food chain. There is little exchange
with higher trophic levels, especially
MAXIMUM REACH OF STRONG -MHWS on exposed beaches. By the removal and

CU~R~R:E:N~T~S~~~~~~~~rt
mineralisation of organic materials and
INPUT ••• -MTL the release of inorganic nutrients the
micro- and meiofauna are responsible for
-MLWS the 'purification' of large volumes of
sea water flushed through the
. -MHWS
MAXIMUM AERATION AND interstitial system.
INPUT CURRENTS~ ..;::;:::;gg~!.~§~~?:~'
..- -MTL
3. THE MACROFAUNA

-MLWS
Hhile not usually as quantitati vely
important in energy flow as the
interstitial fauna, the macrofauna of
-MHWS
MAXIMUM INPUT AND MINIMUM
PULSING DISTURBANCE sandy beaches are often abundant and, in
some cases, attain exceptionally high
densities. 'I'heir main feature on open
beaches is the lligh degree of mobility
displayed by all species. 'I'hese animals
-MHWS may vary from a few mm in length to IDcm
MAXIMUM PERCOLATION AND
MINIMUM PULSING and consequently a variety of methods
DISTURBANCE -MTL
have been used to sample them. Generally
sand is passed through aD. 5mm or Imm
-MLWS
screen, though in areas of coarse sand,
screens of 2mm or even 4mm have been
MAXIMUM INPUT -MHWS
used. 'I'hus not all workers have sampled
MAXIMUM AERATION AND ~~:r:
MINIMUM DISTURBANCE~~~~~?3 the same components of the community.
-MTL

-MLWS The macrofauna community consists of

those organisms too large to move
between the sand grains and generally in
FIGURE 5. Diagrammatic representation of the size range Img-2g dry tissue mass.
some important interstitial parameters Such species make up shifting popu-
related to water flow in an intertidal lations in the intertidal and surf zones
high energy beach. (After Riedl, Machan, of open beaches. Hhen evaluating them
1972) • it is best to consider the population
across an intertidal strip of beach as a
The interstitial system appears to be whole, this unit contracting and
very efficient at mineralising virtually expanding as beach profile changes and
 333

tides pass from springs to neaps.

•
Consequently all further discussion of SHELTER EXPOSURE
'lIIII
abundance and biomass of macrofauna
populations will consider metre wide CRUSTACEANS
strips of beach and not square metres,
unless stated otherwise. All authors
should present their results in this
way, at the same time stating the width POLYCHAETES
of the beach.

The macrofauna of sandy beaches includes FIGURE 6. Responses of the major

most major invertebrate taxa although it invertebrate groups to the exposure
has long been recognised that molluscs, gradient on sandy beaches.
crustaceans and polychaetes are the most
important (Rees, 1939; Hatkin, 1942; (Stephen, 1928, 1931, 1932; Rapson,
Southward, 1953; Schuster-Diedricks, 1952, 1954; Holme, 1954; Figueras,
1956; Sourie, 1957; pichon, 1967; 1956; Edgren, 1959; Ansell, 1961;
Dexter, 1969, 1972) . There is a Hade, 1967; MCIntyre, 1970; Smith,
I
tendency for crustaceans to be more 1971; Brown, 1971b; Ansell et al.,
abundant on tropical beaches or more 1972b; Irwin, 1973; de villiers,
exposed beaches and molluscs to be more 1974; McLachlan, Hanekom, 1979;
abundant on less exposed and/or McLacblan, van der Horst, 1979 ;
temperate beaches (McIntyre, 1968, McLachlan et al. , 1979b; Leber,
1970; Seed, Lowry, 1973; Dexter, 1981) 1982b). Brown (1982) reviewed work on
although there are many exceptions to whelks of the genus Bullia and Ansell
this and polychaetes are sometimes more (1983) bas covered the genus Donax.
abundant than either of these taxa
(Brown, 1981; McDermott, 1983) . There bave been some general accounts of
Indeed, Dexter's (1983) work suggests crustaceans on sandy beaches (MCIntyre,
that crustaceans dominate the most 1963; Epelde-Aguire, Lopez, 1975;
exposed beaches and polychaetes the most Kamihira, 1979) and many stUdies on
sheltered beaches with molluscs reaching individual species or groups. Amphipods
maximum abundance in intermediate have been extensively studied on the
situations (Fig. 6) • In terms of east coast of the United states (Croker,
biomass, however, molluscs are usually 1967a, 1967b, 1968,1970; Dexter, 1967,
most important (Trevallion et al., 1971; Sameoto, 1969a, 1969b;
1970; Dexter, 1974, 1976; Eleftheriou, Bousefield,1970; Craig,1973; Croker
MCIntyre, 1976; McLachlan, 1977a; et al., 1975; Scott, Croker, 1976;
Ansell et al., 1972a; Bally, 1981; Holland, Polgar, 1976; Donn, 1980) and
Shelton, Robertson, 1981). elsewhere (Williamson, 1951; Barnard,
1963; Fincham, 1971, 1974, 1977;
There have been many studies on the Kamihira, 1981). Isopods, both air- and
biology and ecology of dominant bivalves water-breathers, have also been widely
and gastropods of sandy beaches studied (Fish, 1970; Jones, 1971, 1974,
334

1979; Dexter, 1977a; K1apow, 1972; ranges from one (Gau1d, Buchanan, 1956;
Fish, Fish, 1972; Kens1ey, 1972, 1974; McLachlan et a1., 19 81b) to 82 (Vohra,
Ho1dich, Brown, 1973; Ferrara, 1974; 1971), this number increasing with
Glynn et al., 1975; Johnson, 1976a, decreasing exposure. Bally (1981)
1976b; Che1azzi, Ferrara, 1978; divided beaches into three degrees of
Eleftheriou et a1. , 1980; Ho1anov, exposure and, based on a literature
Hendrickson, 1980). Mysids have been survey of 105 beach studies, listed
investigated by Fishe1son, Loya (1968), average numbers of species, abundances
Brown, 'l'a1bot (1972) and Woo1drigde and dry biomass values for these (Table
(1981) and cumaceans by pike, Le Sueur 3).
(1958) and Corey (1970).
TABLE 3. Summary of mean macrofauna1
Hippid crabs have been extensively abundance and dry biomass values based
stUdied. Efford (1976 ) reveiwed on 105 beach surveys. (After Bally, 1981).
distribution in the genus Emerita while
other stUdies include A1ikunhi (1944), High Medium Low
Matthews (1955), Thomassin (1969), energy energy energy
Efford (1965, 1966, 1972) , Di11ery, No. spp. 11 17 30
Knapp (1970), Wenner (1977) , Henner, Abundance m- 2 400 752 1710
Fusaro (1979) and Subramoniam (1979). Abundance m- 1 20045 34571 2797867
Ocypodids have also been widely studied Biomass g.m- 2 2.26 1. 97 6.23
(Crane, 1941; Geo rge, Knott, 1965 ; Biomass g.m- 1 871 170 63
Fellows, 1975; Jones, 1972 ; Hill , Sand particle
Hunter, 1973; Vannini, 1976; Wolcott, diameter urn 310 257 238
1978; MCLachlan, 1980c; Robertson,
Pfeiffer,1981; Robertson, Pheiffer,
1981; Hill , 1982; Wada, 1982; Hails, While diversity and abundance decrease
Yaziz, 1982) and Caine (1974) and du with exposure (Angus, 1979; McLachlan
Preez (1981 ) have studied portunids. et al., 1981b) individual size
There have also been studies on increases, yielding high biomass values
cnidarians (Kastendiek, 1982) , even at lower abundances in some cases.
echinoderms (Buchanan, 1966; Laurence, Though most classifications of beach
Ferber, 1971 ; Scheib1ing, 1982; exposure are very subjective, the above
Dexter, Ebert, 1975; Dexter, 1977b), trends remain clear. 'l'he highest
po1ychaetes (Clark et al. , 1962; biomass values come from very exposed
Longbottom, 1970; Mock, 1981; Wilson, dissipative beaches, e.g. 6621 g ashfree
1981; Brown, 1982) and insects (New, dry mass.m- 1 in the East Cape
1968; Craig, 1970; Orth et al., 1977; (McLachlan, 1977a) and 25735g dry mass
Stenton-Dozey, Griffiths, 1980; Che1azzi m. -1 in Peru (penchaszadeh, 1971).
et al., 1983). These high biomass values are, without
exception, due to filter feeders (Donax,
Bally (B81) summarised the results of Emerita) . Exposed Leaches are thus not
all beach macrofauna surveys to 1981. necessarily sparsely inhabited.
The number of species found on a beach
 335

Species diversity normally increases McIntyre, 1970; Lawrence, Ferber,

from high to low tide marks (McLachlan, 1971; Alheit, 1978; Bally, 1981,
1977a) and may decrease again around the McLachlan et al., 1981b; Lopez-Cotelo
break point (McGwyyne, 1980) and then et al., 1982) beach slope (McLachlan et
increase offshore (Field, 1971; Day et al., 1981b) sand moisture (Hayes, 1977;
al., 1971; Ch ri s tie, 1976). However, Salvat 1964, 1966, 1967; Bally, 1981;
on a beach dominated by kelp input, Hithers, 1977) food in the surf water
maximum macrofauna diversity occurred (Brown, 1964; Rapson, 1954; Wade,
around the drift line at the top of the 1968; MCLusky et al., 1975; Ansell et
beach (Stenton-Dozey, Griffiths, 1983). al.,1972a; Nair, 1978; McLachlan et
al., 1981a; Hutchings et al., 1983) and
The distribution of macrofauna along the dynamic changes such as due to storms
beach is generally patchy, the combined (Scott, 1960; Brown, 1971a). Effects of
result of movement and sorting by the dredging and beach nourishment have also
swash, localized food concentrations or been investigated in the inter- and sub-
biological aggregations of species tidal (Reilly, Bellis, 1978; 'l.'urbeville,
(Loesch, 1957; Thurn, Allen, 1975; Marsh, 1982; Culter, Mahadevan, 1982;
Achutankutty, 1976; Moueza, Chessel, Saloman et al., 1982), while effects of
1976; Hayes, 1977; Saloman, Naughton, disturbance predation have been assessed
1978; Fusaro, 1980; Brown, 1982) . experimentally in a sheltered sand flat
Bally (1981) investigated patchiness in by Woodin (1981). Many authors have
detail on west Cape beaches and showed monitored organic or chlorophyll levels
it to be a common occurrence in all in the surf without veing able to
species at all tide levels, often having demonstrate correlations with faunal
proportions of lOl-102m. This abundance. Undoubtedly, however,
suggests that line transect sampling veaches with high viomass must receive
strategies may be unreliable, high inputs of particulate organics and
particularly if not replicated. Dauer, it is noticeable that on most veaches
Simon (1975), however, showed that by where very rich filter feeding
pooling the results of a whole line populations occur, phytoplankton vlooms
transect, error due to patchiness can be have been recorded in the surf (New
reduced by 10-15%. Hartnoll (1983) has Zealand Rapson, 1954; Hashington
quantified the species area relationship Lewin et al., 1979a; South Africa
on a sheltered sandy beach. He found McLachlan et al., 1981a; South America
little increase in diversity as sample Gianuca, 1983) . Alternately, rich
size increased above 0.25m 2 and populations develop in upwelling
suggested that for
accurate sampling an regions, probably feeding on
aggcegate area of 0.5m 2 is adequate. phytoplankton advected shorewards after
upwelling cycles (Hutchings et al.,
Distribution and abundance of organisms 1983) .
on the beach has been related to many
factors. Factors said to influence the Of the physical factors, wave action and
macrofauna include sand grain size andl particle size have generally been
or organic content (Longbottom, 1970; considered the most important although
336

20~----------------------------------r4000
Eleftheriou, Nicholson (1975) show that (a)
grain size alone can not characterize a
115
,....
I
beach. As particle size, beach slope 3000 I
I
and wave action are closely related, rJ)
w
U I
defining the former two fixes the latter w E
(Davies, 1972; MCLachlan, 1980a) . 55 10 2000 rn
cr
U. w
McLachlan et al. , (198lb) obtained o CD
cr :E
significant correlations between w ::l
CD 5 1000 Z
macrofauna species numbers and abundance :E
::l
Z
and both grain size and beach slope, but
not with wave action estimates for
beaches on the south and east coast of BEACH SLOPE
South Africa (Fig. 7). This suggests
20~--------------------------------~4000
that it is not wave action but rather (b)
steep slope and coarse sands that limit
I
the fauna. very exposed beaches (in 3000 I
I
terms of wave height) often have richer
~

I
faunas than less exposed beaches where E
2000 rn
the latter have coarse grains and steep cr
w
CD
slopes. Finer sands result in flatter :::E
::l
slopes as does heavier wave action. 'l'he 1000 Z
flatter the slope of a beach the more
evenly wave energy is dissipated in the
surf zone and intertidal and this is the 100 500 1000
crucial factor. Thus a very exposed Md (jJm)
beach can support a rich macrofauna if FIGURE 7. 'I'he relationships between
it is dissipative as opposed to macrofaunal species diversity and abun-
relective (e.g. Maitland beach, dance and the slope (a) and particle
McLachlan, 1977a; Copalis beach, Lewin size (b) of beaches in South Africa.
et al., 1979a). Indeed, high energy (After McLachlan et al., 1981b).
dissipative conditions seem optimal for
the development of a high biomass of 1970, 1971, 1972; Dexter, 1978; Hill,
filter feeders. HUnter, 1979; Shin, 1981, 1982;
Kastendiek, 1982). The zone inside the
Much less is known of macrofauna ecology break point has been termed the inner
below the intertidal. Fauna is turbulent zone and the area outside this
generally absent around the break point (out to where wave effects on the bottom
but increases in abundance and diversity are negligible) the outer turbulent
onshore and offshore in accordance with zone. Biomass values are seldom high.
less turbulent conditions, less coarse Masse (1970, 1971) recorded number of
substrates and higher organic levels 1048 10031 m- 2 and biomass 2.5-l1.7g
(Clark, Milne, 1955; Morgans, 1962; dry mass m- 2 at various localities in
Barnard, 1963; Field, 1971; Day et the Mediterranean at depths 1,5 5m.
al. , 1971 ; Christie, 1976; Masse, ~1cIntyre, Eleftheriou (1968) recorded an
 337

average dry biomass of 1.3g.m- 2 in the the midshore in New Zealand. In South
intertidal and 3.7g.m- 2 in the Africa Dahl's scheme, with modifi-
subtidal with 62 species on the beach cations, has been found acceptable
and 116 off the Scottish shore. The (McLachlan, 1980b; McLachlan et al.,
poorest zone was just below LHS. 1981b) .

Much has been written on intertidal

distribution patterns. 'l'he earliest
WATER BREATHERS
serious attempt at providing a zonation
scheme for sandy beaches was that of
Dahl (1952) (Fig.8). This was very
similar to an earlier suggestion by
Davenport (1903). Based on the distri-
bution of crustaceans in northern
temperate and South American beaches he
proposed three zones essentially
equivalent to the three zones of rocky
shores. These zones were: the
subterrestrial fringe characterized by
ocypodid crabs in warm areas and ELWS

talitrid amphipods in cold areas; the

midlittoral zone characterized by FIGURE 8. Diagrammatic representation of
cirolanid isopods (although these may be schemes of macrofauna zonation on sandy
absent in cold temperate areas); and the beaches.
sublittoral fringe with a mixed fauna
often characterized by hippid crabs in Sal vat (1964, 1966, 1967) described an
the tropics and haustorid and other alternate scheme based on physical
amphipods in temperate areas. Dahl's factors. He delineated four zones on
zones were thus defined biologically. the beach, based on water saturation
(Fig. 8). pollock, Humon (1971) used
Subsequently many authors have tried this system for a Massachusetts beach
this scheme with varying results. they were studying, except that they
Gauld, Buchanan (1956) found an overlap subdivided the zone of dryin'J and they
of talitrids and ocypodids on west extended their :cones into strata
African shores and also recorded covering the whole interstitial system
Excirolana in this zone. Lower zones (see Table 1 and Fi'J. 3). Salvat's
showed less correspondence. Philip system has been found reliable in
(1972) in India, Vohra (1972) in describing zonation of macrofauna,
Singapore and Jaramillo (1978) in Chile particularly isopods (Hithers, 1977;
found Dahl's scheme useful and Escofet Bally, 1981) . Botb these authors
et al., (1978) found it fairly suitable admitted, however, that boundaries are
for beaches in Brazil, Uruguay and not sharp and that zones 'Jrade into each
Argentina. Hood (1968), however, found other. Eleftheriou, Jones (1976)
ci rcolanids replaced by a sphaeromid on pointed out that differences in inter-
338

tidal distribution vetween different the larger one in the eulittoral (e.g.
areas and species make eurydicid isopods Amphidesma spp. in New Zealand, Rapson,
unreliable indicators of zonation as 1952; Donax spp. in India, Ansell et
proposed by Dahl. They did not, however, al., 1972v; Donax spp, in the Bast
attempt to use Salvat's scheme. Cape, MCLachlan, 1980b).

Views on zonation of macrofauna on sandy Besides zonation patterns at community

beaches are thus highly conflicting, a level, intraspecific zonation has Leen
situation not surprising in a dynamic recorded in several species. This
environment with a highly mobile fauna. generally takes the form of );.onation of
It may safely be said that zonation, in size classes and has veen recorded in
the classical sense, has never been most taxa typical of open beaches,
proved on a sandy beach, i.e. sharp including molluscs and crustaceans
boundaries have not been demonstrated. (Alikunhi, 1944; Glynn et al., 1975;
Further, individual species show much McLachlan et al., 1979b; Hooldridge,
clearer zonation than the fauna as a 1981; Ansell, Lagardere, 1980; Brown,
whole. Indeed, Brown (per s. comm.) is 1982; Haley, 1982). This may ve due to
of the opinion that only two zones are differential sorting of the sizes in the
distinguishable on sandy shores air swash or to active mi9ration to areas
breathers in the supralittoral and water differentially sui table to the or9anism
breathers below them. However, two at different life stages.
boundaries are clearly and indisputably
visible on open veaches, namely the Zonation as the researcher records it
drift line and the edge of the saturated is, however, only the distribution of
sand where the water table reaches the the psammolittoral fauna at an instant
surface. These correspond to Dahl's in time and most species undergo tidal
boundaries. However, although the migrations of some sort. Typically this
boundary between Salvat's zones of involves a simple movement up and down
retention and resurgence are not obvious the veach with the tides allowing the
there is considerable evidence that animal to stay in the swash zone where
meaningful subdivision of the conditions for feeding are opitmal and
midlittoral can be made on many predation minimal. In molluscs this
beaches. These zonation schemes are movement involves no endogenous rhythms
summarized in Fig. 8. Within these but simply a series of responses to
zones most species show subzones and changing physical conditions, most
there is often blurring of boundaries. notably the degree of saturation and
Isopod species usually exhibit a clear thixotropy of the sand (Mori, 1938;
zonation near the top of the shore and Turner, Belding, 1957; Ansell,
midlittoral (Withers, 1977; Eleftheriou, Trevallion, 1969; Trueman, 1971). In
Jones, 1976; Bally, 1981; McLachlan et some molluscs it may be more complex,
al., 198~b; Dexter, 1983). Bivalves, e.g. Donax serra in the Bast Cape, where
also exhibit zonation, often with two there is not normal tidal migration but
species occupying a beach, the smaller rather semilunar movement up and down
species in the sublittoral fringe and the shore to occupy a position near mean
 339

tide level during springs and near low compress so that at high tide most of
tide during neaps (McLachlan et aI, the invertebrate population on the beach
1979c). Movements related to periods of may be compressed into a narrow strip
storm and calm have also been recorded with considerable overlap between
(Leber, 198 2a) . species. Further, in addition to normal
tidal migration, movement into the
In crustaceans these rhythms are subtidal bas been recorded in winter in
generally endogenous and more complex. Donax parvula (Leder, 1982b).
They have been shown to include entry
into the plankton at night (e.g. mysids, Temporal changes in beach macrofauna
anphipods and isopods), increased have been recorded both in whole
activity during springs as opposed to communities (Sanchez et al., 1982) and
neaps, and orientation behaviour (papi, in individual species populations. 'l'he
pardi, 1963; Enright, 1963, 1965, 1972; most dramatic of these are those assoc-
Hammer et al., 1968, 1969; cubit, 1969; iated with the monsoon in India where
Fincham, 1970, 1973; Alheit, Naylor, most of the fauna disappears during this
1976; MacQuart-Moulin, 1977; McLachlan period (Ansell et al., 1972a; Dwivedi et
et al., 1979c; Marsh, Branch, 1979; al., 1973; McLusky et al., 1975).
Hager, Croker, 1980; Fusaro, 1980; Longer term fluctuation spanning several
Jaramillo et al. , 1980; Scapini, years have also been recorded (Rapson,
Ugolini, 1981) . The advantage of 1954; Coe, 1955; Davis, Von Blaricom,
increased activity during springs is 1978). Dramatic changes in the long
that it prevents the animals being term may also take the form of occas-
stranded as the tides retreat. ional mass mortalities, e.g. Donax spp.
(Orton, 1929; Fitch, 1950; Loesch, 1957;
Ansell, Trueman (1973) calculated the Johnston, 1966, 1968; Grindley, Nel,
energy costs of migration in Donax and 1968; de Villiers, 1974) such as related
Emerita and concluded that it was more to poisonous dinoflagellate blooms.
profitable for the animals to migrate Seasonal changes have ueen reported in
than to attempt to maintain position. most species studied and for several
Tidal migratory behaviour has several communities.
advantages; it allows animals to stay in
the zone optimal for feeding, Le. the In an unstructured and physically
edge of the swash zone; it keeps them controlled habitat like a sandy beach,
too shallow for many fish predators but most species tend to be unspecialized
just immersed enough not to be fully generalists with board niches (Brown,
open to birds; it helps to prevent Talbot, 1972; Brown, 1982; MCGwynne,
animals being stranded on the shore; in 1980; Bally, 1981). Because of this,
supralittoral forms, e.g. Tylos, it the physical nature of the environment
enables them to move down the beach and the mobility of the fauna,
during the low tide to feed on debris. competition amongst the sandy beach
fauna is unlikely, though the
This migration naturally has marked possibility of exploitation competition
effects on zonation and all zones can not be ruled out. Wilson (1981)
340

demonstrated negative interactions in a predators.

dense assemblage of deposi t feeding
polychaetes, but this was in a very As the bulk of the fauna are usually
sheltered beach. Trophic groups among filter feeders, the size of beach
the macrofauna include scavenger/ populations is probably closely related
predators, filter feeders (suspension) to the richness of inshore waters in
and, on less exposed shores, deposit particulate organic material (Ansell et
feeders. Filter feeders usually al., 1972a; Hutchings et al., 1983). In
dominate and consist mostly of some areas large amounts of macrodebris
bi val ves. The tellinaceans of high may be cast ashore, particularly
energy beaches, Donax spp., are, unlike adjacent to kelp beds. In such
most other members of the family, situations vast populations of supra-
suspension feede~s (Ansell , 1983) . littoral scavengers develop (Hayes,
Deposit feeding is not common in sandy 1974; Griffiths, Stenton-Dozey, 1981).
beach filter feeders, except in very In other areas carrion input may be
sheltered situations. In cases where high, e.g. in the form of cnidarians,
the fauna is impoverished, however, such and in such cases midlittoral scavengers
as on very coarse, steep beaches, such as Bullia reach high abundance
supralittoral scavengers may dominate as (McGwynne, 1980; Brown, 1982).
a result of the virtual absence of other
species (Gauld, Buchanan, 1956; Dye et Macrofaunal food webs have been given
al., 1981; Wooldridge et al., 1981). On for beaches in South America (Koepcke,
sheltered beaches deposit feeders, e.g. Koepcke, 1952; Gianuca, 1983), the Cape
Arenicola, Callianassa and Scolelepis, peninsula (Brown, 1964; Griffiths et
may be very important or even dominate al., 1983) and the Bast cape (McLachlan
(e.g. Saloman, Naughton, 1978; McDermott, et al., 1981a). The main predators on
1983) . the macrofauna as listed by several
authors are birds, fishes and crabs
The tropic structure of the beach macro- (Coe, 1955; Loesch, 1957; Masse, 1963;
fauna is therefore normally dominated by Wade, 1967; Penshaszadeh, Olivier,
mobile filter feeders. Even where other 1975) . The former two will be discussed
forms may be more abundant, filter later. Crabs, being part of the
feeders usually dominate biomass benthos, represent a form of predation
(Dexter, 1979; Eleftheriou, MCIntyre, resident in the beach/surf system. Du
1976). As there is little or no primary Preez (1981) has made a detailed study of
production on the beach, the macrofauna the swimming crab Ovalipes on East Cape
is dependent on food imports from beaches and shown it to be an important
adjacent systems, i.e. the land and the predator on Donax and Bullia. Several
sea (Brown, 1964; Ansell et al., 1972a; other authors mention the importance of
McLachlan et al., 1981a; Griffiths et both ghost crabs and swimming crabs as
al., 19~3). The sea is the most predators on beaches (Loesch, 1957;
important and supplies particulates for Koepcke, 1953; Ansell et al., 1972a).
filter and deposit feeders and carrion Virnstein (1977) showed the importance
and macrodebris for scavengers and of crabs in controlling macrofauna
 341

communities by predation in subtidal 4. PHYTOPLANKTON

sand in Chesapeake Bay. Naticid Blooms of diatoms in surf zones,
gastropods are important predators on persistent or sporadic, now appear to ue
burrowing bivalves but are not generally a typical feature of many sandy beach
found on open ocean beaches. surf zones. Rich surf phytoplankton
blooms have been reported from the
Besides predation, "natural" mortality Washington coast, U.S.A., (Thayer, 1935;
can at times be important on sandy Lewin, Norris, 1970), the Gulf coast of
beaches. This includes not only mass the U.S.A., (Gunter, 1979; Gunter,
mortalities due to poiso~ing (de Lyles, 1979), New Zealand (Rapson, 1954;
Villiers, 1974) but also being cast Cassie, Cassie, 1960; Lewin, Norris,
above the shore by storms (Brown, 1971a; 1970), South Africa (McLachlan, Lewin,
Penshaszadeh, Olivier, 1975; McLachlan 1981) and the east coast of South
et al., 1981a). McLachlan, Young (1982) America (Gianuca, 1983). These blooms
suggested the importance of upwelling are mostly composed of masses of cells
and sudden temperature drops in of one or two species uelonging to the
retarding mobility as a factor contri- genera Chaetoceros, Asterionella,
buting towards this. Ansell et al., Aulacodiscus and Anaulus.
(1978) have shown that where the
mortality rate is fairly constant, the The blooms described by Gunter (1979)
mortality coefficient, longivity and and Gunter, Lyles (1979) are Chaetoceros
production are related. Ansell (1983) dominated strips hugging the ueaches.
lists cases of commercial exploitation Heavy rains leach nutrients from the
of Donax where heavy mortality may be land and these, ueing retained near the
caused by man. beach by calm seas, apparently cause the
blooms. These blooms take the form of
It may be concluded that the distri- strips 5-6m wide for miles along the
bution and diversity of beach macrofauna shore. All the other reported surf zone
is largely determined by physical blooms occur as patches of phytoplankton
factors, prime among which are wave within well developed surf zones. Of
action, particle size and beach slope. these, those off the Washington coast
Food input probably determines abundance have been most intensively stUdies and
of both scavengers and filter feeders. reviewed by Lewin, Schaefer (1983).
The most pronounced feature of
macrofauna animals is a high degree of Lewin, Norris (1970) first described the
mobility and tidal migrations on exposed similar i ties uetween blooms on the
beaches. The main rale of the ~iashington and New ~ealand coasts where

macrofauna is to process food imports, Chaetoceros and Asterionella were

coming mainly from the surf, and in turn dominant. Lewin, Hruby (1973) described
to be consumed by terrestrial and marine the diel periodicity in buoyancy of
predators. They are thus in the middle Chaetoceros, this species dispersing at
of the food chain. night and rising to the surface dUring
the day. Floating at the surface, they
create a staule foam which is then
342

retained in the surf by wave action and Subsequently more attention has been
often stranded on the beach. This paid to the ecology and distribution of
clearly appears to be a mechanism for the blooms. Jijina, Lewin (in prep. a,
retention in the surf zone. Lewin b) described physical features of the
(1974) described long term changes in coastal zone in relation to blooms at 13
species composition of the blooms wi th beaches in ~Iashington and Oregon.
Aulacodiscus littorii dominant before Blooms were best developed on long flat
1950 and Chaetoceros armatum dominant beaches while steeper beaches and
after 1950. Other studies looked at beaches near rocks has poorer blooms.
light utilization (Lewin, Mackas, 1972), Low air and water temperatures were
nitrate reductase activity (Collos, associated with increased diatom
Lewin, 1974), daily periodicity (Lewin, densities while nutrient concentrations
Rao, 1975), C:N ratio (Collos, Lewin, were inversely proportional to
1976), respiration (Robertson, Lewin, densities. ':i.'hey concluded that the type
1976), chemical composition (Lewin et of beach and its physical characteris-
al., 1979a) and the clay coat (Lewin et tics were very important in bloom
al., 1980) of the surf diatoms development.
Chaetoceros and Asterionella.
McLachlan (198 Od) suggested that blooms
Lewin et al., (1975) analysed environ- form where surf circulation cells
mental conditions associated with the develop and trap beach generated
blooms over two years, finding the nutrients, pointing out that this is
diatom populations relatively constant most likely on long beaches of flat
over this period. Lewin (1978) slope (i.e. dissipative beaches).
evaluated factors controlling the McLachlan, Lewin (1981) described blooms
seasonal cycle of nitrate in the surf. of Anaulus birostratus from the East
Nitrate concentrations fluctuated widely Cape, South Africa, and showed that the
in relation to season, upwelling and blooms were positioned over rip currents
phytoplankton growth and ammonium seemed by day. All surf bloom forming species
to be an important nitrogen source, show a daytime buoyancy and this results
particularly during summer. Lewin et in the collection of diatom cells as a
al., (1979b) measured ammonia excretion foam or scum over rip currents. Presum-
by Siliqua and concluded that the razor ably the opposing forces of incoming
clam populations of Washington beaches waves and outgoing rip current keep them
could regenerate significant amounts of in position. While diatom cells may
ammonium into the surf zone. Lewin travel all around the surf circulation
(1977) summarised knowledge of these cells, the rip current area may act as a
blooms, pointing out the interdependence bottle-neck for buoyant material.
of the diatoms and the razor clams.
Diatoms are the main food of the razor McLachlan, Lewin (1981) also showed that
clams while the clams in turn regenerate blooms developed in water not very high
inorganic nitrogen which is utilized by in nutrients. Though nutrient concen-
the diatoms. tration is low, supply is constnat as
inorganic nutrients are continually
 343

draining out of the beach. Nutrients species commonly occur in well developed
are generated on the beach both by the surf zones on gently shelving beaches.
filtration of water and mineralisation Diurnal buoyancy of the diatoms results
of organics by the interstitial fauna in their concentration over rip currents
(McLachlan, 1980a, 1982; McLachlan et by day. Nutrients are derived from many
al., 1981a) and by excretion of the sources, but the beach is probably most
macrofauna (Prosch, McLachlan, 1983). important. Inorganic nitrogen generated
Both of these sources are important and as ammonium by macrofauna and
have been shown to generate significant zooplankton excretion and as nitrate by
amounts of nitrogen on East Cape the interstitial fauna is constantly
beaches. Subsequently wind has been draining into the surf and passing
shown to be of overriding importance in through rip currents where the blooms
controlling short term changes in blooms are situated. In this way the blooms
in the surf (Romer, Sloff, pers. comm.). are positioned in a site of steady
nutrient supply. Hind has a pronounced
Estimates of primary production in surf influence on Lloom development by
zones off beaches have been made by controlling wave action and surf
Cassie, Cassie (1960) and Edwards circulation patterns. He need to know
(1973a). The latter obtained figures of more of the interaction between wind,
10495 and 5597 kcal.m- 2 .y-l for waves, nutrients and blooms. He also
phytoplankton primary production in the need more reports on the sources of
shallow off two sheltered venezuelan primary production in surf zones that do
beaches. Estimates of pr imary not support these diatom blooms.
producti vi ty in surf phytoplankton
blooms are now under way (Lewin, 5. ZOOPLANKTON
Schaefer, 1983) and we should soon have Studies on surf zone zooplankton are
figures for biomass and production of virtually absent from the literature.
these rich algal communities. Hhile s~veral taxonomic papers list
species taken off sandy beaches, few
Recently Wilce et al., (1982), Hilce, were sampled in the surf zone or include
Quinlan (1983) and Quinlan et al. (1983) ecological notes. Brattegard (1969 and
reported on masses of free living subsequent works) descr ibed mysids from
ball-like algae, pilayella, on sand shallow water off the Bahamas including
bottoms and in the surf off beaches in some from sandy habitats and similar
Nahant bay, Massachusetts. This species accounts are given by Higley, Burns
reached peak biomass in summer and (1971) for the Atlantic coast of the
proliferated by fragmentation and re- U.S.A. and Bacescu (1975) for Tanzania.
generation. It's nutrient source is
uncertain but
nutrient recycling within Bowman (1971) described copepods off the
the beach/surf system is considered southeast coast of the U.S.A. and
important. recognised a 'coastal association'
dominated by one'coastal' and one
It may be concluded that rich surf 'estuarine' species. Outside this lay
diatom blooms consisting of one or two 'shelf' and 'oceanic' associations
344

dominated by four and seven species outer limit of distribution near the
respectively. Sander, Moore (1978) outer limit of rips. He postulated that
studied inshore and offshore copepods detrital food in suspension decreases
around the west Indies as shallow as 10m outside this zone and that zonation was
and recorded average numbers of 100- mainly related to the nearshore,
500.m- 3 with other groups making up wave-induced circulation ~atterns.

11-22% of total iooplankton numbers.

Youngbluth (1979) studied zooplankton Several studies have recently been done
within lkm of the coast of puerto Rico on surf zone zooplankton in the East
at about 10m depth. He recorded numbers Cape. In most of this work the use of
of 41-7568.m- 3 of which 65-84% were large nets has attempted to avoid the
copepods. Settled volume biomass values normal loss of large forms such as
averaged 0.086ml.m- 3 and highest mysids, a problem often not appreciated
numbers were recorded at night. by workers using small nets. Cockcroft
(1979) sampled inside the breakers at 1m
Moran (1972) found the benthic mysid, depth, Just outside at 3m and further
Gastrosaccus sanctus, to occur in shoals out at 10m off a moderately exposed
along the shore in the top lcm of sand, beach. Greatest biomass occurred just
but moving into the top I-2m of water at outside the breakers (although netting
night in the Mediterranean. It efficiency was very low in the surf) .
exhibited a clear zonation offshore. Abundance increased considerably after
Fincham (1970), studying amphipods in dark when dry biomass values ranged
Britain, Macquart-Moulin (1977), 6-980mg.m- 3 . Mysids dominated biomass
studying two benthic mysids and an (21% at 1m, 83% at 3m and 84% at 10m;
isopod in the Mediterranean, and mean 63%) with copepods averaging only
Wooldridge (1981), studying a benthic 13%. At the 1m site the prawn,
mysid in the East Cape, found similar Macropetasma africana, was very
patterns. Night time planktonic important although poorly sampled. 'l'he
activity of benthic crustaceans, dominant species overall was the mysid,
particularly mysids, amphipods and Mesopodopsis slabberi. Biomass was
isopods, is well known (Alheit, Naylor, distinctly higher than offshore in Algoa
1976) . Bay and about the same as recorded in
local estuaries, indicating the general
The most detailed account of surf zone richness of zooplankton in the surf zone.
zooplankton yet published is the work of
Clutter (1967) at La Jolla. He studied Subsequently Cockcroft (1983) has made a
the nearshore zonation of four benthic detailed study of Macropetasma, which
and five pelagic mysids out to 17m depth. occurs in vast shoals in East Cape surf
All species formed schools and occupied zones. Growing up to 7cm in length,
clear zones, occurring on or near the this prawn is really a nektonic form.
bottom. A large species dominated It exhibits a clear zonation with
inside the surf while the most abundant juveniles inshore and adults out as far
speciesi Metamysidopsis elongata, peaked as 20m depth. They shoal near the
where rip c'urrents dispersed and had its bottom by day and disperse at night,
 345

being most abundant inside the breakers. 1000.m- 3 just behind the breakers and
Numbers recorded averaged 21-37.m- 3 10.m- 3 fUrther offshore. It even
giving a dry biomass of 0.9-1.8g.m- 3 . reached densities exceeding 15000.m- 3
This species uses the surf zone as a (mainly juveniles, dry biomass
nursery area and, being an opportunistic 1.02g.m- 3 ) behind the breakers at
omnivore, it feeds on detritus, times. Other mysid species seldom
phytoplankton and small crustaceans. totalled more than lO.m- 3 . Dry
Shoals tend to move inshore at night and zooplankton biomass averaged about
to concentrate around phytoplankton O. 25g. m- 3 behind the breakers and
blooms by day. O.015g.m- 3 offshore, an order of
magnitude lower.
Romer (pers. comm. ) is investigating
food chains around surf phytoplankton Mesopodopsis formed vast shoals which
blooms, including small and large virtually disappeared by day (presumably
zooplankton. He finds that zooplankton moving to deeper water) and then became
is concentrated around blooms by day but abundant at night. This inshore
more dispersed at night, this resulting population consisted largely of
in a slight decrease in biomass after juveniles and breeding occurred at sea
dark. crustaceans (Macropetasma, even though large numbers often moved
mysids, copepods) account for about 85% into estuaries. This study indicates
of biomass and siphonophores, that the inshore area and surf zone bas
chaetognaths, and others for the a mysid dominated zooplankton community
remainder. Mean total numbers in a of very high biomass. Abundance and
bloom during the day are about 2000 biomass in this region are an order of
zooplankters.m- 3 with a dry biomass of magnitude higher than offshore and even
0.8g.m- 3 . However, higher values, up higher than recorded in most estuar ies.
to 3
750g. m- , have been recorded in r-lany authors have probably
dense swarms of plankters, particularly underestimated the contribution of the
where prawns were abundant. larger and more mobile zooplankton
through using small nets.
Hooldridge (1983) studied the zoo-
plankton of Algoa Bay from just behind Clearly our knowledge of surf zone
the breakers to about 4km offshore (20m zooplankton is abysmal. Few papers go
depth) using a net of 1.5m diameter. beyond species lists or counts of
Hhile salps, medusae, ctenophores and abundance. While it is difficult to
others were often common, crustaceans extrapolate from the small amount of
dominated, making up about 80% of hard data currently available, three
biomass. Fifteen species of mysids points seem clear: i) the zooplankton of
accounted for more than 90% of exposed surf zones may be very rich with
crustacean biomass, with copepods second high lJiomass values, ii) large
in importance. A single species of crustaceans (e.g. mysids, penaeids)
mysid, Mesopodopsis slabberi, was by far domina te the lJiomass (as they can cope
the dominant species, especially close wi th the turbulence of these areas) but
inshore, with average numbers around are difficult to sample, iii) there are
346

clear offshore zonation patterns and recruitment, growth and feeding of

diurnal changes in abundance. Much work juveniles in the shallow subtidal off
remains to be done in this field, in sandy beaches (Riley, Corlett, 1966;
particular relating distribution and Macer, 1967; Edwards, ~teele, 1968;
abundance to wave induced circulation Lockwood, 1974; Pox ton et al., 1982).
patterns and available food in the near- Among these fish, plaice (Pleuronectes
shore zone. pIa tessa), dabs (Limandi a limandia) and
turbot (Scophthalmus maximus) are the
6. FISHES most important. In summer, after
Though far better studied than metamorphosis, large numbers move into
zooplankton, we still know comparatively fine sand !Jays where the water is less
little of the ecology of fishes from than 3-5m deep. There they feed on
surf zones off open beaches, largely demersal plankton and benthos, including
because of the difficulties of sampling mysids and the siphons of Tellina and
in these environments. pearse et al., tentacles of polychaetes. Nortali ty is
(1942) list fish caught by seine net off high and the populations of these
North Carolina beaches and more juveniles generally decline rapidly
extensive reports of fishes in surf through autumn until they move into
zones of the United States are given by deeper water again at the onset of
Gunter (1958) et al. (1960), McFarland winter.
(1963), Schaeffer (1967, 1970), Cupka
(1972) and McDermott (1983). Around the In the United States Gunter (1958)
coasts of Britain and Europe much sampled surf fish off the ~l'exas (;oast,
attention has been paid to juvenile recording fewer species off open beaches
flatfish using beaches as nursery than inside the barrier. He mostly
grounds (Macer, 1967; Edwards, Steele, caught small and juvenile fish which
1968; Edwards et al., 1970; Steele, were most abundant in spring and
Edwards, 1970; Steele et al., 1970b; summer. McFarland (1963) found that
Jones, 1973; Gibson, 1973; Thijssen et zooplankton \{as the primary food source
al., 1974; Lockwood, 1974; Poxton et for the surf fish at Mustang Island,
al., 1982). In South Africa Rossouw Texas. Planktivores were the most
(1983) has studied elasmobranchs from abundant trophic group and even mullett
surf zones with particular emphasis on (Mugil cephalus) and benthic feeders
the sandshark Rhinobatos annulatus, took large amounts of zooplankton at
Lasiak (1982, 1983) has made a detailed times. He recorded 47 species with an
investigation of surf zone ichthyofauna average of four in winter (biomass
communities and Romer, (pers. comm.) has 25.81b/acre) and 16 in summer (biomass
investigated food chains around surf 103.21b/acre). High primary production
phytoplankton blooms. in the surf, he concluded, was more than
sufficient to support the fish community.
In the cold temperate waters of Britain Moddle, Ross (1981) also recorded
and Europe shallow sandy bays are midwater planktivorous fish dominating
important as nursery areas for flatfish the surf zone in the northern Gulf of
and several authors have studied Mexico. Armitage, Alevizon (1980)
 347

recorded the diet of the Florida pompano Edwards (1973b) sampled with a beach
(Trachinotus carolinus) and found Donax seine net off two Venezuelan beaches and
to be an important item. recorded 49 species (31 at one beach and
32 at the other) which he divided into
Carlisle et al., (1960) recorded 71 plankton feeders, fish carnivores,
species off Californian beaches with an benthos feeders and herbivores (Mugil).
average of 284 specimens per beach seine Fish biomass was estimated at 5g.m- 2
haul. The surfperch, Amphistichtus dry mass at the unpolluted beach and
argenteus, and the orbina, Menticirrhus Ig.m- 2 at at a polluted beach.
undulatus, were dominant but flatfish, plankton feeders and benthos feeders
skates, rays and sandsharks (Rhinobatos) were most important. Penchaszadeh
were also common. Best catches were (1983) found mysids and other motile
taken on the low tide. They made a crustaceans to be the main prey of
detailed study of the biology of the fishes off venezuelan beaches.
surfperch including reproduction, growth
and feeding and found the mole crab, In Australia Lenanton et al. (1982)
Emerita analoga, to be the main item in described the utilization of surf zone
the diet. accumulations of macrophytes and their
associated amphipods as nursery feeding
Schaefer (1970) studied feeding of the areas for 0+ year classes of three fish
striped bass, Morone saxatilis, in the species.
surf at Long Island. Amphipods and
mysids were the dominant foods though In the Eastern Cape, South Africa,
fishes, particularly anchovies, were Rossouw (1983) has studied elasmobranchs
also important. During sampling 71 with particular reference to the biology
species of fish were recorded. of the sandshark, Rhinobatos annulatus.
This species comes into the surf to give
McDermott (1983) sampled the fauna of a birth in summer. Juveniles stay in the
barrier beach in New Jersey by means of surf but adults move offshore in
cores and seine netting. The dominant winter. 'J.'hey feed on mysids, small Donax
benthic organism and main food for the and other benthic prey. Amongst various
fish was the polychaet, Scolelepis other elasmobranchs, duckbill rays,
squamata, which reached densities up to Myliobatis spp., come into very shallow
40000.m- 2 (=50g dry mass.m- 2 ) in a water to prey on Donax spp. 'l'he se,
band approximately 20m wide. Of 26 fish together with other large rays are very
species only seven were common and one important predators on the benthos. As
resident. This species, Menidia menidia, in the case of flatfishes in Luropean
the silverside, made up 85% of fish waters, the great success of skates and
numbers and 40% of fish mass and rays in these environments is probably
Scolelepis formed 70-80% of its diet. due to their flattened shape allowing
Leber (1982a) recorded four fish species them to forage in very shallow water,
feeding on beach macrofauna, including thus being able to take even the
the silverside. Donax and Emerita were macrofauna that undergo tidal migrations.
the main prey items.
348

Romer (pers. comm. ) is investigating controlled by winds.

food chains around surf phytoplankton
blooms. He has found that mullett, Liza Lasiak (1982) distinguished six feeding
richardsoni, grazed directly on the categories, namely: benthic feeders,
phytoplankton and are taken in turn by planktivores, detritivores, ~iscivores,

larger predatory fishes. The mullett as herbivores and omnivores. Herbivores

well as other juvenile fish in the surf owed their occurrence to the ~resence of
also prey extensively on zooplankton. some rocky reefs near the sampling site
at the less exposed beach, and were
Lasiak (1982, 1983) has made the most virtually absent at the more exposed
detailed study to date of surf fish beach. The mullet, Liza richardsoni,
communities, seining on two Algoa Bay was the only detritivore, but in
beaches over a period of three years. Lasiak's study fed mainly on zooplankton
At the less exposed beach she recorded and may thus be considered a planktivore.
66 species and 29 at the more exposed Her feeding groups may thus be
site with teleosts dominating numbers simplified to include, benthic feeders,
and elasmobranchs dominating biomass. planktivores, piscivores and omnivores.
Seven common species were resident while At both beaches motile organisms
other species were intermediate or (mysids, prawns, fish, cephalopods and
sporadic in occurrence. The community zooplankton) dominated the food base and
showed no overall seasonal trends but a these foods were taken by all feeding
very high degree of short term groups including benthic feeders.
variabili ty which seemed to be largely Almost all species employed oppor-
coupled to the effects of wind on the tunistic feeding strategies. Table 4
surf zone as several biological summarises the biomass, composition and
parameters were closely correlated to food consumption of this surf zone fish
wind direction and intensity prior to assemblage.
sampling. Studies over 24h showed no
clear trends though maximum abundance, He still know very I i ttle of surf zone
biomass and diversity often tended to ichthyofauna and the danger of sampling
occur around twilight. Sampling on in heavy surf will probably keep the
consecutive days proved to be very situation that way. To really
variable. She concluded that the surf understand resource partitioning and
zone fish assemblage was a highly niche structure of surf zone fishes some
variable community whose structure and more accurate means of sampling is
dynamics were probably dominated by necessary whereby the actual location of
abiotic factors while biological fish within uifferent parts of the surf
interactions were of secondary can be made. perhaps SCUBA counts are
importance. As most food organisms were the solution. In this way the distri-
highly motile, their availability would bution and feeding of fish in the surf
be strongly influenced by surf could be better understood in terms of
conditions which in turn are determined spatial partitioning of this zone in
by wind effects on the wave regime. three dimensions. However, zooplankton
Hence short term variability is largely as food for fish llill be more important
 349

Table 4. Biomass, composition and food consumption of fishes in less exposed surf
zones in the East Cape. (After Lasiak, 1982).

FISH TOTAL PREY TAKEN

Number of Species
Feeding group Biomass g.m- l Amount consumed
(dry mass) Dominants Other Species or group g.m-l.y-l

Benthic feeders 78 3 6 Macrobenthos 269

Planktivores 60 4 5 Mysids/prawns 657
piscivores 10 2 4 Zooplankton 200
Omnivores 10 2 0 Fish 672
Total 158 'I'otal 1798

at night and studies not employing after there are few published reports on the
dark sampling will miss this. ecology of birds on sandy beaches. Most
papers simply record counts of waders
Surf zone tish communities are highly and other birds along sandy shorelines.
dynamic assemblages with few resident
species. They are variable in space and Fitch (1950) recorded gulls dropping
time and short term variations in wind, clams to break them and Brunton (1978)
controlling surf conditions, are recorded gull predation of toheroa
probably a major factor controlling clams. veitch (1980) recorded dead
these communities. Most species show a birds on beaches in New Zealand and
high degree of opportunism in feeding, similar records have Leen made in
and consequently all feeding types may Belgium (Kuyken, 1978) and Britain
predate heavily on zooplankton and (Cadbury, 1978). Prater, Davies (1978)
larger schooling crustaceans when they recorded sanderling uverwintering on
are locally abundant. Planktivores, and British beaches, while Schneider,
benthic feeders with opportunistic Harrington (1981) demonstrated depletion
planktivore tendencies, are therefore a of benthic prey at a migratory stopover
major component of surf zone for shorebirds. Bur<jer et al. (1977)
ichthyofauna. Fish are important as compared the utilization of an open
energy transformers in surf zones and, beach, a sheltered Leach and a salt
because they are highly motile, are also marsh by migrating birds. ~L'here are
important in the export of energy from several reports of birds breeding on or
the beach/surf zone system. adjacent to beaches, particularly terns
(Fuchs, 1977; Davies, 1981; Randall,
7. BIRDS MCLachlan, 1982) and plovers (Summers,
Hhile some authors have made brief Hockey, 1980).
mention of birds as predators on sandy
beach benthos (Koepcke, Koepcke, 1952; In South Africa there are several count
Brown, 1964; Leber, 1982; Vader, 1982) records for the western Cape. pringle,
350

Cooper (1977) conducted counts over 18 amounts of insects blowing onto the
months which revealed eight migrant and beach and car r ion cast ashore. On kelp
seven resident species. Summers et al. dominated beaches on the west coast
(1977) recorded higher bird numbers on birds are the main predators and were
the west coast than the south coast of estimated to take 40% of herbivore
the western Cape, presumably due to standing stock (Griffiths et al.,
greater food abundance in the former 1983). Hockey et al. (1983) summarised
area. They recorded 1267 birds over the r61e of shorebirds on sandy beaches
99,6km of sandy beaches and 8319 birds in South Africa. They found diversity
over 120,4km of mixed shores. Twelve and abundance of shorebirds to increase
wader species occurred on the beaches with latitude. Birds consumed 10-49% of
with sanderling (Calidris alba), white- macrobenthic production in different
fronted sandplovers (Charadrius areas and return significant amounts of
marginatus) and black oystercatchers faeces, feathers and carcases to the
(Haematopus moquini) most important. beach system.
Resident species averaged 7,3km- l and
migrants 15,2km- l in summer. Summers, Moran, Fischelson (1971) recorded two
Cooper (1977) recorded fewer oyster- plovers, Charadrius hiaticula and C.
catchers on beaches than rocky shores alexandrius, feeding on the mysid
and Summers, Waltner (1979) recorded Gastrosaccus sanctus. 'l'hey found these
mass changes in waders, showing most plovers to hunt at random, regardless of
species to be heavier in winter. On the the size of the mysids, preferring areas
Natal coast Joubert (1981) made counts where mysids were densest. However,
over 10 months, recording 14 species and Schneider (1981) reanalysed their data
densities of 39km- l , with waders to show that c. hiaticula selectively
making up only 0,52km- l • Terns takes larger liIysids. Schneider (1982)
roosted on the beach in relatively high studied the foraging behaviour of
numbers. turnstones (Arenaria interpres) feeding
on Donax variabilis on a florida beach
In the Eastern Cape McLachlan et al. and Pienkowski (1982) described diet and
(1980) counted birds on three beaches feeding of two plovers on a sand flat in
over 12 months, recording 17 species Northumberland. ~hey fed during day and
with average numbers of 18,9km- l of night and polychaet worms and crus-
which waders accounted for 47%. They taceans dominated in the diet.
estimated bird biomass and also food
consumption from standard metabolic Myers et al. (1980) analysed the
rates for four dominant species which predation of sanderlings on beach
made up 95% of numbers. From this they crustaceans in detailed laboratory
estimated that birds were taking 32% of experiments. ~hey showed how several
available macrobenthic production from variables affect a sanderling's capture
these beaches each year. This was rate. capture is determined by prey
mainly- in the form of sand mussels density, the sampling rate and sampling
(Donax spp.) , mussel siphons, and efficiency. Sampling rate depends on
mysids. Birds also took significant penetrability of the substrate and
 351

sampling efficiency depends on prey flow between them. The macrofauna are
size, depth and substrate part of a larger food web including
penetrabili ty. Prey risk varied zooplankton, fishes and birds, while the
inversely with prey depth, 60% of prey interstitisal fauna constitute their own
10mm or shallower being taken and very food web in the sand (MCIntyre, Murison,
much less below 10mm. Larger prey were 1973; McLachlan, 1977a; McLachlan et
more vulnerable than smaller prey but al., 1981a). The only links between
small prey were never rejected. Prey these two food webs is the common input
risk was higher where prey organisms of particulate (and probably dissolved)
occurred close together. capture rate organics to both and the possible
could be predicted on the basis of prey washing of micro-organisms from the sand
size, density and sand penetrability. (Munro et al., 1978) to provide food for
For prey it was best to be small, deep filter feeders. In more sheltered
in the substratum and not near other situations, however, deposit feeders and
individuals. Prey species had no effect sand swallowers may take in large
(Excirolana vs. Emerita). However, this amounts of interstitial fauna. Oliver
was all done with dead prey. Most et al., (1982) recorded phoxocephalid
feeding was idone without visual cues amphipods feeding on a variety of foods
although sanderling may use visual cues including nematodes and Croker (1967b)
at times, e.g. when feeding on beach recorded a baustorid amphipod feeding on
wrack insects. Although probe location meiofauna, both cases being in sbeltered
was random it could also be affected by sands. On a lagoon beach Alheit,
the feeding patterns of nearby birds. Scbeibel (1982) recorded Juvenile
demersal fisb feeding on harpacticoid
Hading birds are important predators on copepods. However, as these exchanges
sandy beaches, being efficient croppers are mostly negligible on open beaches I
of the macrobenthos during periods of look separately at the lilacrofauna food
low tide. On beaches where high inputs web and interstitial energetics.
of macrodebris concentrate macrofauna
along the drift line above tbe reach of 8.2 Nacrofauna
marine predators, birds may be the most Several authors have mentioned aspects
important predators. Even birds that of predation or food chains relating to
only roost on the beaches may be sandy beacbes (Pearse et al., 1942;
important where the faecal deposits of Bonnet, 1946; Koepcke, Koepcke, 1952;
large flocks may add significant amounts Hedgpeth, 1957; Brown, 1964; Steele et
of nutrients locally. al., 1970b; Philips, 1970; Edwards,
1973a, 1973b; Reilly, Bellis, 1978;
Ansell et al., 1978; Reilly, 1979;
8. FOOD VlEBS AND ENERGY FLOH McLachlan et al., 1981a; Leber, 1982;
Gianuca, 1983). IIJacrofauna food chains
8.1 Introduction start and end in the sea, ·witb tbe land
The macrofauna and interstitial fauna of playing only a negligible r61e, mainly
open sandy beaches form two separate in the form of insects blowing onto the
communities with little or no energy beach and land crabs, birds and mammals
352

moving onto the beach to feed. Open herbivorous fishes. Ansell et al. ,
beaches are characterized by the absence (1978) distinguished two 9roups of
of attached plants (Brown, 1964) and macrofauna, (1 ) herbivores and
primary production by the benthic detritivores and ( 2) carnivores.
microflora is generally negligible. On HcLachlan et al., (1981a) separated only
a sheltered temperate beach, however, filter feeders and scavengers/predators
Steele, Baird (1968) found viable on East Cape beaches, where deposit
diatoms mixed to 20cm depth in the sand feeders were of negligible importance.
and primary production low but On kelp dominated beaches Griffiths et
measurable. They considered vertical a1. (1983) uistinguished herbivores,
mixing due to wave action to be the main carnivores and filter feeders. 'l'he
limiting factor. In the shallow possible food sources and trophic 9roups
subtidal off two venezuelan beaches, of macrofauna on sandy Leaches are
Edwards (1973a) recorded significant listed in ~able 5.
benthic pr imary production due to both
micro- and macro-algae (2943 and 6500 Zooplankton, which ure extremely import-
kcal.m- 2y-l at two beaches) but ant in surf water, although largely
these areas are hardly typical of open ingored to date, would fall in cate-
beaches and the latter area was subject 90ries (1 ) and (4). On most sandy
to heavy pollution. Primary production beaches food source (1) is by far the
in the water column may be very most important, with food sources (3)
important, particularly where surf and (4) Leing next. On sheltered
phytoplankton blooms occur. Unfor- shores, however, particulate detritus
tunately primary production figures for and living microflora and fauna may be
a surf phytoplankton bloom have yet to of most importance once incorporated in
be published. Edwards (1973a) estimated the sediments, and under such conditions
primary production in the water column deposit feeders such as Arenicola,
off venezuelan beaches as Callianassa and Scolelepis may be
2934mgc.m- 2 .d- l • abundant.

Different authors have partitioned food While inputs of particulate matter to

sources for sandy beaches in a variety Leaches have yet to be hleasured, some
of ways. Brown (1964) distinguished an estimates are available based on the
erratic supply of macrodebris and a requirements of the I,lacrofauna. In
regular supply of particulates while India particulate feeders accounted for
Hayes (1974) recognised dissolved and 56141 kJ.m- 2y-l and carnivores for
microparticulate organics, stranded 4291 kJ .m. -2y -l on Shertallai beach
plankton and large organic masses. and 38765 kJ .m-l.y-l and 235kJ.
Edwards (1973a) distinguished three main m-ly-l at Cochin beach respec-
pathways by which energy travels from tively. Particulate feeders thus
primary foods to fishes; (1) via zoo- accounted for 91% of the food
plankton to plankton feeding fish, (2) consumption on average (Ansell et al. ,
via benthos to demersal fishes and (3) 1978) . In the East Cape, filter feeding
directly from benthic algae to macrofauna were estimated to assimilate
 353

Table 5. Food sources and trophic groups of macrofauna on open sandy beaches.

FOOD SOURCE FEEDING GROUP

1. particulate organics including phytoplankton Filter feeders

2. Deposited detritus, benthic diatoms, interstitial fauna Deposit feeders

3. Stranded debris - plants Scavengers - herbivores

animals - carnivores

4. Other fauna predators

93% of the total and fly larvae (Muir, 1977; Koop, Field,
for the beach macrofauna, with 1980, 1981; Stenton-Dozey, Griffiths,
scavenger/predators making up the 1980; Griffiths, Stenton-Dozey, 1981;
remaining 7% (McLachlan et al., 1981a). Koop et al., 1982a; Griffiths et al.,
/
This does not include zooplankton 1983). Biodegradion of macrophytes by
organisms which are mainly filter the macrofauna is thus very important.
feeders and very abundant. Scavengers
are only important where there are large In sheltered localities or where large
inputs of debris or where filter feeders amounts of particulate organics are
are absent because of unstable physical deposited, deposit feeders may reach
conditions or the absence of food (e.g. high l.Jiomass values (Eleftheriou,
East Coast of South Af rica, Dye et al., McIntyre, 1976; McDermott, 1983). On
1981; Wooldridge et al., 1981). most beaches organic content of the sand
is very low, e.g. 0,2-0,4mg.g- l sand
Huge inputs of macrophytes characterize in Scotland and India (Steele, 1976) and
beaches adjacent to kelp beds or similar undetectable amounts in the West Cape
areas. In southern California Hayes and East Cape (Brown, 1971a; McLachlan,
(1974) estimated these inputs to amount 1977a) .
to 473kg wet (ca.78kg dry) algae
m-l.y-l (~jacrocystis), while on Hest Predators on the other benthos include
Cape beaches adjacent to kelp beds crabs, starfish and gastropods. These
values range around 2000kg wet Ecklonia normally occur in relatively low
m-l.y-l (Koop, Kield, 1980; Koop et abundance, making up 5-10% of biomass.
al., 1982a; Stenton-Dozey, Griffiths, As many are supralittoral forms, e.g.
1983; Griffiths et aI, 1983). Hhile Ocypode, they may be the only macrofauna
Hayes (1974) found Tylos punctatus to where swash conditions are too dynamic
consume only 4-5% of the edible kelp for intertidal macrofauna to survive
input, estimates for macrofauna (Gauld, Buchanan, 1956; Dye et al.,
consumption in the west Cape range 9-74%, 1981). 'l'hey generally take both living
this being taken by isopods, amphipods benthic prey and stranded carrion.
354

Energy flows through different species These food chains end in fishes, birds,
and trophic efficiencies vary widely. crabs and other invertebrate predators
Turnover (P/B) ranges from 0.3 for and man. very tew studies have
Tellina tenuis on a temperate beach estimated the importance of these
(Trevallion, 1971) to 10.3 for Donax different predators. In Scotland
spiculum on a tropical Indian beach juvenile flatfish are the most import-
(Ansell et al., 1972b). Ansell et al., ant predators, cropping siphons of
(1978) found turnover on these tropical bivalves, palps of polychaetes and whole
beaches generally ten times higher than prey. Bivalves may then put more energy
on cold temperate Scottish beaches. into regeneration of siphons and less in-
Edwards (1973a) recorded average P/B's to reproduction. Trevallion (1971 )
of 2.8-3.5 for the macrobenthos, estimated that up to 1.3% of total
McLachlan et al., (198la) used P/B' s of annual energy expenditure in Tellina
0.5-5.0 for somatic production of tenuis may go into regenerating
molluscs and crustaceans, and Koop, siphons. In India crabs were considered
Field (1981) estimated the P/B for Ligia the most important predators, with
dilatata at 3.75 times per year. fishes and man less important and birds
virtually absent (Steele, 1976; Ansell
Assimilation efficiencies (A/ C X et al., 1978). Edwards (1973a)
100%) vary widely, commonly being very estimated fish to account for 71-89% of
low for scavengers/grazers feeding on production at his two beaches, starfish
abundant macroalgae, e.g. 64% for Tylos accounting for 13% on one and crabs for
(Hayes, 1974), 5.5% for Ligia (Koop, 13% on the other beach. McLachlan et
Field, 1981) and 30-50% for Talorchestia al., (198la) partitioned predation on
and fly larvae, (Stenton-Dozey, east Cape ueaches. Birds took 32% of
Griffiths, 1980). Koop et al., (1982a) intertidal macrofauna production
estimated an overall average (McLachlan et al. , 1980), fishes were
assimilation efficiency of 22% for estimated to take 55% and uenthic
invertebrate kelp grazers while Ansell predators (mainly crabs, Ovalipes) 10%
et al., (1978) and Edwards (1973a) with man accounting for less than 3%.
assumed assimilation efficiencies of 80% Du Preez (1981) has subsequently made a
for carnivores and 60% for herbivores. detailed investigation of the predation
Ansell et al. , (1978) estimated of Ovalipes on Donax and Bullia and
efficiencies of transfer of energy Lasiak and Rossouw (1983) have studied
through the food chain (p/ A X 100%) fish predation. On kelp dominated
as 15-16% for herbivores and beaches of the West Cape, however, birds
detritivores and 25-30% for carnivores are the major predators (Griffiths et
in India while Edwards (1973a) estimated al., 1983).
Pic X 100% as 15% for benthos, 17%
for demersal fish and 11-15% for other Evaluating our knowledge of sandy beach
fish at his unpolluted site. Ecological food chains as a whole is made difficult
efficiencies calculated from McLachlan by not only the small number of
et al. (l'98la) were 9-19% for the quantitative studies, but also the
macro-fauna with a mean of 17%. absence of data on zooplankton. The
 355

macrofauna cannot be considered in iso- To conclude this section on macrofauna

lation when describing these food energetics it appears that five main
chains. Surf zone zooplankton beach types can be recognised in terms
production, also untimately falling prey of macrofauna energy flow patterns
to fishes and being exported from the (Table 7). Information on surf zone
system, is probably more important in zooplankton in different zoogeographic
te rms of ene rgy flow. In the East Cape regions is unfortunately not available
for example, where we have some idea of to allow incorporation into this scheme.
surf zone zooplankton biomass, a It may ue expected, however, that as
different pattern emerges if larger zoopolankton and macrobenthic filter
mobile crustaceans, such as mysids and feeders utilize the same food resources,
prawns, are included with zooplankton their total biomass would ue roughly
out to 500m from the beach (Table 6). similar in many cases.

Table 6. Summary of approximate faunal 8.3 Interstitial Fauna

dry biomass for an average beach and Less attention has been paid to
surf zone in the East Cape. (From interstitial energetics, although as
McLachlan, 1983) early as 1942, Pearse et al., proposed
that the interstitial systems of sandy
BIOMASS beaches had digestive and incubating
CATEGORY <j.m- l functions, the microfauna mineralising
organic materials in the sand. Little
attempt was made subsequently to
Macrobenthos 1500 evaluate or quantify this until quite
Zooplankton 1100 recently. This system consists
Birds 5 typically of bacteria, protozoans and
Fishes 300 meiofauna constituting their own food
web in the sand.

The food chain is centred around filter Dissolved and fine particulate organics
feeding benthos and zooplankton feeding form the base of this food chain,
on particulate detritus and phyto- although in sheltered situations the
plankton and being taken in turn by benthic microflora might also play a
fishes, birds and crabs. As zooplankton rene. McIntyre et al., (1970) were the
turnover is 6-7 times that of the first to demonstrate that interstitial
macrofauna, the centre of gravity of the fauna could subsist on dissolved
food chain is clearly in the surf and organics. Using two experimental sand
not on the beach. McFarland (1963) came columns, they maintained viable micro-
to a similar conclusion, showing that and meiofaunal populations in the second
plankton was the predominant f ish food column on the effluent leaving the first
off Texas beaches, greatly exceeding the column. They suggested that bacteria
benthos as an available food source. The were the prime utilizers of dissolved
surf zones he studied were characterized organics, utilizing these organic mole-
by high primary productivity in summer. cules when they adsorb onto sand <jrains
356

Table 7. Major beach types in terms of macrofauna energetics.

predominant Food Macrofaunal Dominant

Type Latitude Sources Biomass Trophic Group Predators

1- High turnover, Tropical Particulates Low to Filter Invertebrates

particulate- moderate Feeders Fishes
based food
chain

2. Low turnover, Temperate particulates Low to Filter Fishes, Birds

particulate- high Feeders Invertebrates
based food
chain.

3. MacrodeLris- 'I'emperate Stranded Hoderate Scavengers/ Birds

based food macrophytes Grazers
chain.

4. Sediment-based 'l'emperate Deposited detritus Low to Deposit Birds, Fishes

food chains in or microorganisms and high Feeders Invertebrates
sheltered Tropical benthic microflora
beaches.

5. Carrion-based Temperate Stranded carrion Low Scavengers/ Birds, Crabs

food chains in or Predators
extremely ex- Tropical
posed steep
beaches.

where they can be digested with extra- some meiofaunal species may feed
cellular enzymes. The meiofauna in turn directly on detritus, microflora or
fed mainly on these bacteria, accounting dissolved organics, in a Lroad sense
for approximately 5% of bacterial con- meiofauna do form the top of the inter-
sumption. stitial food weL as most species feed on
bacteria, protozoa or other meiofauna
Fenchel (1972) considered bacteria the (McIntyre, ~lurison, 1973; McLachlan et
only primary decomposers of particulate al., 1981a). 'l'he position of protozoans
plant detr itus with their decomposition in this system has Leen little studied
rate limited by oxygen and nutrients. in open beaches; while some forms may
Consequently he suggested that higher have a similar trophic status to
trophic levels in the interstitial food bacteria, others are even predatory on
web are dependant on initial processing meiofauna. However, in a Lroad sense
of plant detritus by bacteria. Although they probably occupy a position inter-
 357

mediate between, and overlapping with, (1977) and Dye (1983). Another approach
bacteria and meiofauna in interstitial has been the study of experimental sand
food chains. columns in the laboratory.

Three trophic types may be recognised Several authors have maintained

within the meiofauna, predators, laboratory sand microcosms for extended
herbivores and bacterivores, with only periods (McIntyre et al., 1970; pugh,
the Hydrozoa and Turbellaria exclusively 1976; Boucher, Chamroux, 1976; Chamroux
predatory (Tietjen, 1980). Swedmark et al., 1977; Hormald, stirling, 1979;
(1964) however, listed four categories, McLachlan et al., 19S1c). A closed
predators, diatom and epigrowth feeders, circuit system has been successfully
detritus eaters and suspension feeders. maintained for 16 months on soluble
Besides direct grazing of meiofauna on amino acids (Boucher, Chamroux, 1976)
the microfauna, Tietjen (1980) con- confirming the utilization of dissolved
sidered that meiofauna stimulate organics by microorganisms and their
bacterial activity in the interstitial subsequent consumption by the meiofauna.
system by (1) mechanical breakdown of However, changes in meiofaunal communi ty
detritus, making it more susceptible to structure (decreased diversity)
bacterial colonisation, (2) excretion of indicated that the utilization of
nutrients for microbial use, (3) soluble organics via uacteria was not
secretion of mucous thereby attracting the only trophic 1Jath to the lIIeiofauna.
bacterial growth and (4) movement In this case uacteria and meiofauna were
conveying nutrients and oxygen. estimated to account for 95% and 5% of
Assimilation (AI
efficienciesX the carbon input res1Jecti vely, the same
C
100%) for meiofauna have been estimated proportions as found by r-lcIntyre et al.
in the range 6-26% while transfer (1970) •
efficiencies (PIA X 100%) were
79-97% for three nematode species There have been few detailed studies of
(Tietjen, 1980). (P /B )
For turnover interstitial energy budgets in situ on
a ratio of 10 has been widely used for open Leaches. The three cases that
meiofauna (Gerlach, 1971). warrant attention concern a comparative
study of Scottish and Indian beaches,
Because of the difficulty of scale when studies on high energy beaches in the
working with the minute inhabitants of Eastern cape and studies on beaches
the interstitial fauna, investigations receiving high kelp inputs in the
of interstitial energetics have western Cape.
generally taken the form of I black box I

studies of the system as a \{hole. This r-lunro et al. ·(1978) made a detailed
has commonly been done by measuring comparative investigation of inter-
interstitial or benthic oxygen consump- stitial dynamics on tropical Indian
tion in situ and then attempting to beaches (Md 150-190u!l1) and Scottish
partition this. Benthic oxygen consump- beaches (Md 250um) using small and large
tion measurements have been discussed sand columns and in situ measurements.
and reviewed by Lassere (1976), Pamatmat The tropical beach lacked epipsammic
diatoms which were common on the temper- organic levels were 500-1000 ug C.l- l
ate beach. Dissolved organics were in Scotland and 900-3700 ug.l -1 in
estimated to make up 80% of the input on India. However, surf flushing of the
the Scottish beach and 39% on the Indian Indian beaches may remove significant
beaches. Interstitial community respir- quantities of bacteria, thus making them
ation on the tropical beach (164gC. available to macrofauna filter feeders
m- 2 .y-l) was nine times the winter and decreasing the quantity available to
values and twice the summer values for the meiofauna.
temperate beach (mean 42gc.m- 2 .
with microbial production In the Last Cape the interstitial fauna
estimated to be 72 gc.m- 2 .y-l in the consists essentially of meiofauna,
former and an annual average of protozoans and bacteria to considerable
15g.c.m- 2 .y-l in the latter (using depth in the sand (Dye, 1979a; McLachlan
respiration figures based on an assumed et al., 1979a). Dye (1980) showed tidal
conversion efficiency of 45% and the fluctuations in benthic oxygen consump-
assumption that the contribution of tion as a result of tidally induced
meiofauna was negligible). Meiofauna changes in pore moisture and interstitial
biomass was, however, much higher on the water flow, lowest values being recorded
Scottish beach (273-523mg. m- 2 ) than when the sand dried out. Dye (1979a)
the Indian beach (24-60mg.m- 2 ) (Table investigated biological oxygen demand in
8) • They postulated that more vigorous beach sand, using intact sand cores in
flushing of the sand on the tropical the laboratory as these gave higher
Indian beach both boosted respiration values than in situ techniques. He
and striped bacteria from the sand showed that disturbance caused a
grains. significant increase in oxygen uptake.

The tropical beach received more In a more detailed study Dye (1981 )
organics and had a higher interstitial later partitioned this oxygen uptake
oxygen consumption rate than the between bacteria, protozoans and
Scottish beach as a result of increased meiofauna on two beaches (Table 9).
water flow by wave flushing. Surf

Table 8. Biomass, production and respiration of interstitial organisms on Scottish

and Indian beaches. (After Munro et al., 1978). All values in gc.m- 2 .y-l.

INDIA SC01'LAND
Respiration Production Biomass Respiration Production Biomass

Algae o o o 7
Microfauna 160 72 31 15
Meiofauna 4 2 0.02 4 2 0.2
 359

Table 9. Partitioning of interstitial oxygen uptake on two East Cape Deaches.

(After Dye, 1981).

MODERATELY EXPOSED BEACH VERY EXPOSED BEACH

Biomass % O2 uptake % Biomass % O 2 uptake %

Meiofauna 75 18 79 21
Protozoa 19 18 17 24
Bacteria 6 64 4 55

In the west Cape very rich interstitial rapidly. 'l'he great significance of
populations develop on beaches with high these microbes indicates why Hayes
kelp inputs. Koop, Griffiths (1982) (1974) recorded such little kelp
recorded dry biomass of 241, 200 and 663 breakdown in California where he looked
g.m- l of macro- rneio- and microfauna at only one grazer, Tylos punctatus.
respectively. Bacteria were concen-
trated at lower tide levels, probably 8.4 Conclusions
subsisting on small particles of broken It may be concluded that intertidal
down kelp resulting from feeding and sandy beaches are important in the
faeces of macrofauna and meiof auna. processing of organics .vhether these be
KOOp et al., (1982a) quantified carbon as DaM, paN or larger animal and plant
flow through this system. Bacteria on remains. These foods generally come
the stranded fronds were responsible for from the sea and in turn most production
much initial degradion. Breakdown of returns there. The interstitial fauna
the kelp resulted in high concentrations are always more important than the
of leachates (up to 5640 mgC.l- l ) in macrofauna in this mineralisation of
the sediment below the kelp. Over 90% organic materials, even where a rich
of this was utilized by bacteria after macrofauna occurs (Table 10).
seeping through 1m of sand and 23-27% of
carDon in the kelp was converted to The interstitial system is driven by
bacterial carbon. The ratio of increase inputs of soluble and particulate
in bacterial biomass to kelp carbon used organics, the main path being via
was 58.7%, giving a conversion bacteria attached to sand grains to
efficiency of 29.4% if bacterial biomass meiofauna. The intermediate rOle of
is 50% carbon (Koop et al., 1982a). protozoans is uncertain. CJn exposed
Much of the faeces of invertebrate beaches there is little exchange between
grazers were also converted to bacterial this sytem and the macrofauna although
carbon as these grazers have low the importance of microorganisms flushed
assimilation efficiencies. The from the sand as food for the macrofauna
remaining 73-77% of the kelp was needs further investigation. The
mineralised by the microbes quite concentrations of meiofauna ilushed trom
360

Table 10. Approximate partitioning of total benthic assimilation between macrofauna

and interstitial fauna in four geographic areas based on data in Munro et al. (1978)
Ansell et al. (1978), McLachlan et al. (l981a) and Koop, Griffiths, (1982).

ASSIMILATION [kJ.m-l.y-l (%) I

LOCALITY AND BEACH Macrofauna Interstitial BEACH WIDTH

TYPE fauna (m)

Scotland, temperate 30 000 (23) 99 000 (77) 60

India, tropical 427 000 (38) 703 000 (62) 60

East Cape, warm temperate 116000 (37) 236 000 (63) 60

West Cape, temperate 42 000 (3) 1599 000 (97) 63

with high kelp input

the sand of exposed beaches were 9. NUTRIENT CYCLING

measured by McLachlan (19 77a) and found It has long been speculated that beaches
to be negligible. On sheltered beaches may be of importance in recycling
a greater size spectrum of organisms may nutrients by mineralising organic
occur, filling the gap between inter- materials coming from the sea (Pearse et
stitial fauna and larger macrofauna and al., 1942). 'l'h i s is effected by the
thereby facilitating energy flow by fauna consuming organic nutrients
predation. In such environments deposit including nitrogen an<l phosphorous and
feeders may also ingest significant then excreting this in inorganic form,
quantities of interstitial fauna. e.g. NH 4 , P0 4 . In some areas
Inshore food chains in beach environ- groundwater seepage can, however, also
ments are however centred in the surf supply considerable amounts of inorganic
zone rather than the intertidal, even nutrient via the beach to the surf zone.
where rich intertidal macrofaunas occur. This can take the form of both artesian
Our area of least understanding at the aquifers and general seepage around the
moment is these dynamic food chains of area where the permanent water table
the surf zone and in particular the role meets the sea. Johannes (1980) found
of zooplankton and larger motile this to be very important in "estern
crustaceans. Australia where submarine groundwater
 361

discharge ranged l-Sm.y-l and con- nitrogen (Chamroux, 1965; McIntyre et

-1 al. , pugh, 1976; Boucher,
tained up to 380 ug at 1 N0 3 -N. 1970;
Chamroux, 1976; Cahet, 1976; Hormald,
Nutrient regeneration by the macrofauna, Stirling, 1979; Munro et al., 1978;
zooplankton, birds and fishes has been McLachlan et al. , 1981c) . Generally
virtually ignored. Lewin et al., nitrification is the dominant activity
(1979b) and Prosch, McLachlan (1983) in Leach sand and organic nitrogen is
respectively investigated ammonia efficiently mineralised to nitrate.
excretion in bivalves of the genera Increasing organic loading tends to
Siliqua and Donax on beaches in raise equilibrium levels and it is
Washington and the East Cape, South usually only at very high levels that a
Africa. In both cases very large beach can not cope and anoxic conditions
populations of these bivalves develop and ammonia appear (Oliff et al., 1970;
and nutrients regenerated by them could PU g h , 1976). Boucher, Chamroux (1976)
be of significance in supplying recorded total mineralisation of amino
phytoplankton requirements in the surf acids added to their sand columns while
zone. Hayes (1974) found Tylos to Munro et al., (1978) recorded 70%
consume only 4-S% of kelp input on mineralisation of natural organics,
Californian beaches while Griffiths, mostly in the top Scm of their columns.
Stenton-Dozey (1981) and Koop et al. McLachlan et al. (1981c) found
(1982a) estimated that 9-74% of stranded mineralisation of natural organics in
kelp was consumed by the macrofauna. SOcm sand columns and this increased to
Not all of this is assimilated, however, 6S% when amino acids were added.
and low assimilation efficiencies may
result in a comparatively small Most of these sand column studies have
prop6rtion of kelp nitrogen being been somewhat artificial as water flow
regenerated directly by the macrofauna. has been unrelated to in situ flow
The contribution of other faunal com- rates, there has been continuous submer-
ponents must also be important in this gence, and inputs have been constant
regard but have yet to be investigated. rather than tidal or pulsing. Except-
ions are pugh (1976) and McLachlan et
Generally the interstitial system of al. (lS81c), although even these experi-
sandy beaches has been thought more ments did not incorporate pulsing
important in nutrient regeneration than flows. Filtration rates have ranged
the macrofauna. Water filters through 2 l
2S-1000 1.m- .d- , covering a variety
the sand, as a result of tides and waves of flows Loth intertidal and subtidal.
both in the intertidal and the subtidal,
and DOM and POM are removed Ly the 1,1cLachlan (1982) followed up earlier
interstitial fauna and mineralised. sand column work with a model of water
filtration and nutrient regeneration by
Several studies of experimental systems sandy Leaches. This was based on a
of sand columns or model beaches have regression model predicting filtration
looked at mineralisation and/or nutrient volume as a function of sand particle
regeneration, mostly concentrating on size, beach slope and tide range.
362

Coupled to this \Jas an estimate of tile to rocks and subject to larye kelp
degree of mineralization of organics in inputs are interesting for comparison.
relation to sand particle size and the
filtration distance. The average It may therefore be concluded that sandy
filtration distance is about 35% of the beaches can mineralize most oryanics
horizontal intertidal distance (Riedl, they receive, the Lulk of this Lein,=!
Machan, 1972) and average mineralisation done very efficiently (80-100%) Ly the
was estimated to be at least 60-80%. At interstitial fauna. 'I'his is
a filtration rate of 1041 . m-l. d -l accomplished Ly the llrocess of water
and an organic nitrogen level of 0.15mg. filtration through the Leach and will be
1-1 this predicted regeneration of greatest where steep beaches and short
0.9g N0 3 -N m-l.d- l . An attempt to period waves result in high filtration
estimate total nutrient production by volumes (e.g. in the tropics). Where
beaches in the East Cape was made by vast macrofauna llopulations occur they
McLachlan et al. (1981a). They estimated may also contriLute significantly to
nitrogen excretion by the macrofauna and nutrient regeneration. Zooplankton,
interstitial fauna of the beach and surf birds and fishes must also be very
zone at 350gN.m- l .y-l and 591gN. important.
m-l.y-l respectively. Thus, even on
beaches such as these with very high Most organic nitrogen and phosphorous
macrofauna biomass, the interstitial supplied to a beach will rapidly be
fauna are more important in nutrient returned to the surf as nitrates,
regeneration, accounting for 63%. While ammonia and phosphates. 1'hereafter the
only approximate, the above figures fate of these nutrients will depend on
suggest that the mineralising activities physical conditions in the surf zone.
of the benthos alone can replenish the Where cellular circulation patterns occur
nutrient pool of the surf zone in a there may be a high degree of retention,
matter of days or weeks. while on irregular coasts or reflective
beaches, exchange wi th the open sea may
Koop et al. (1982b) quantified micro- be much more rapid. As lilany surf zones
bial regeneration of nitrogen from are characterized by high primary
stranded kelp over an eight day cycle. productivity, utilization of beach
In this time the microbes incorporated generated nutrients in the surf zone may
94% of the kelp nitrogen, which normally be very important and is in need Gf more
strands at an average input of attention. Hayes (1974) and Koop et al.
12gN.m- l .d- l . However, in the long (19 82b) consider such nutrients of
term the beach can not act as a nitrogen little importance in the surf while
sink and this must return to the sea. McLachlan et al. (1981a) suggest that
In this case this could supply roughly these nutrients may be sufficient to
one quarter of the nitrogen requirements maintain surf phytoplankton blooms on
of phytoplankton in the adjacent kelp long open beaches.
beds. While hardly typical of open
beaches, even on the west coast (Bally, 10. TOWARDS A MORB HOLISTIC APPROACH
1981), these figures for beaches adjacent This review has attempted a functional
 363

look at our present knowledge of the Energy flow and nutrient cycling in a
ecology of beach/surf zone environments long open beach typical of the Bast Cape
as a whole, incorporating all aspects of is illustrated in Fig. 9. This is a
fauna and flora, energy flow and rich system ,;ith considerable flows of
nutrient cycling. Unfortunately, most energy and llutrients. It includes
published studies of beach environments producers (phytoplankton), consumers
have not attempted such a holistic (benthos, zooplankton, fishes) and
approach and have tended to investigate decomposers (interstitial fauna).
one or more components in isolation. Energy flows within this system appear
to be greater than across its
In the East Cape research on beach boundaries. Bowever, although exchange
environments has been unique in across the dune/beach boundary is
attempting a holistic approach towards negligible, exchanges across the outer
the quantification of energy flow and boundary of the surf cells remain to be
nutrient cycling under the concept of a quant if ied. 'l'he main trophic pathway is
beach/surf zone ecosystem. This is from fine organic materials (POM and
based on two contentions (McLachlan, phytoplankton) via zooplankton and
1980c): (1) that the sand body of the filter-feedins macrofauna to fishes.
beach together with the water envelope The interstitial fauna is also extremely
of the surf zone form an ecosystem where important in mineralising POM and DOM.
internal energy flows are greater than 'l'he total ",mount of inorganic ni trogen
exchanges across boundaries and (2) that regenerated by the beach and surf zone
this system has definable boundaries, fauna is not yet known, but must be sub-
these being the drift line on the stantial as the Jdacrofauna and inter-
landward margin and the outer limit of stitial fauna alone can regenerate
surf circulation cells on the seaward enough to replenish the surf zone
margin. The central concept is that the nutrient pool within a few weeks to
circulation cells that typify the surf days. Although not indicated in Fig. 9,
zones of open beaches (Inman, Brush, wave action probably represents the most
1973) imply a certain measure of important factor controlling this system,
integrity of surf zone water and as it is responsible for transporting
retention of materials within this water, sediment and biological
zone. This is in opposition to the idea materials, driving surf circulation
that beaches are purely high energy cells and pumping water through beach
interfaces between sea and land. sand.
Clearly, however, this can not apply to
pocket beaches, beaches associated with Clearly, not only biological and
rocky shores or beaches dropping chemical information is needed to
directly into deep water where there is understand beach/surf ecosystelLls. vlater
no real surf zone. The kelp input movement, the key to all dynamic
beaches of the South African west coast, processes occurring here, is the realm
for example, function as interfaces of the physical oceanographer. The
between adjacent kelp beds and the land. discreteness of surf circulation
boundaries and exchanges across them
364

SEA SURF BEACH DUNES

I
EXPORT OTHER
PREDATERS I
1837 886 1
3684
I 2468
45170 I

600 IEXPORT 512Q

2155
I
I
I
I

o
497

I I

,
, :6125
I
1
1'------'" t
I
I I 1
'
, \
89269 \ \ ' ,
I , ,,,
1

I II
~
,
,
IMPORT!
1
\
\
\
236325 ..
I

I
I \ I
I \ I
\
I
I
EXPORT',
.----_.- -""'-- - - -- -+-----
I DRIFT LINE
OUTER LIMIT OF SURF
CIRCULATION CELLS

FIGURE 9. Energy flow (solid lines) and nutrient cycles (broken lines) for a typical
East Cape beach, largely after McLachlan et al. (1981a). All values in
kJ.m-l.y-l except nutrient values in g inorganic N.

need to be quantified. It is therefore ecosystems, and this can only be

essential that physical oceanographers, evaluated in the presence of a detailed
biologists and chemists work together. knowledge of the physical and chemical
Beaches in different geographic areas environment.
need to be studied in this fashion.
Biologists in particular must not only REFERENCES
study the intertidal fauna, but must
Achutankutty CT (1976) Ecology of
venture out into the surf zone. In the sandy beach at Sancoale, Goa: Part 1.
dynamics of fine organic materials, Physical factors influencing production
of macrofauna, Indian J. mar. Sci. 5,
phytoplankton and zooplankton lies the 91-97.
key to a more realistic understanding of Alikunhi KH (1944) The zonal distri-
bution of the mole crab (Emerita
the function of sandy beach/surf zone
 365

asiatica) on the Madras coast, J. Bombay the macrofauna, Mar. BioI. 17, 38-62.
Nat. Hist. Soc. 45,94-96. Ansell AD, Sivadas P, Narayanan Band
Alheit J (1978) Distribution of the Trevallion A (1972b) 'l'he ecology of two
polychaete genus Nephtys: a stratified sandy beaches in South Hest India III:
random sampling survey, Kieler Observations on the populations of Donax
Meeresforsch 4, 61-67. incarnatus and Donax spiculum, Mar.
Alheit J and Naylor E (1976) BioI. 17, 318-332.
Behavioral basis of intertidal zonation Armitage TM and Alevizon WS (1980) The
in Eurydice pulchra Leach, J. exp.mar. diet of the Florida pompano (Trachinotus
BioI. Ecol. 23, 135-144. carolinus) along the East Coast of
Alheit J and Scheibel W (1982) Benthic Central Florida, U.S.A. Fla. Sci. 43,
harpacticoids as a food source for fish. 19-26.
Mar. BioI. 70, 141-147. Bacescu M (1975) Contributions to the
Amspoker, A.M. (1977) The distribution knowledge of the mysids (Crustacea) from
of intertidal epipsammic diatoms on the Tanzanian water, Univ. Dar es Salaam
Scripps Beach, La Jolla, California, Sci. J. 1, 39-6l.
U.S.A. Bot. Mar. 20, 227-232. Bally R (1981) The ecology of three
Anderson JG and Meadows PS (1979) sandy beaches on the west coast of South
Micro-environments in marine sediments, Africa, Ph.D. thesis, University of Cape
Proc. Roy. Soc. Edinburgh 76B, 1-16. Town.
Anderson JG, Boonruang P and Itleadows Barnard LJ (1963) Relationship of
PS (1981) Interralationships between benthic amphipoda to invertebrate
cholrophylls, carbon, nitrogen and communities of inshore sublittoral sands
heterotrophic bacteria in an intertidal of Southern Carolina, Pacific Naturalist
sediment transect, Mar. Ecol. Prog. Ser. 1963, 3. 439-467.
6, 278-284. Bartsch I and Schmidt P (1979)
Andrews AR, Floodgate GD and pugh KB Distribution and ecology of some
(1976) An annual cycle at constant halacarid mites (Halacaridae) in sandy
temperature in a model sandy beach. beaches of the Baltic, Kiel bay, the
Part 2. Microbial bioI. J. expo Mar. North Sea, Sylt, Hest Germany and the
BioI. Ecol. 24, 61-72. Norwegian Sea Tromso area Norway,
Angus IS (1979) The macrofauna of Mikrofauna Meeresbodens 0(74), 163-197.
intertidal sand in the outer Hebrides, Bascom W (1964) Waves and beaches, New
Proc. Roy. Soc. Edin. 77B, 155-171. York, Doubleday & Co Inc.
Ansell AD (1961) Reproduction, growth Boaden PJS (1968) Water movement - a
and mortality of Venus striatula (Da dominant factor in interstitial ecology,
Costa) in Kames Ba~illport, J.mar. Sarsia 34, 125-136.
bioI. Ass. U.K. 41, 191-215. Bolter /II. Meyer-Reil L, Dawson R,
Ansell AD (1983) The biology of the Liebezeit G, "olter K and Szwerinski H
genus Donax. In McLachlan A and Erasmus (1981) Structure analysis of shallow
T,,/ edS:-Sandy beaches as ecosystems, water ecosystems: interaction of
The Hague, Junk. microbial, chemical and physical
Ansell AD and Trevallion A (1969) characteristics measured in the over-
Behavioral adaptations of intertidal lying waters of sandy beach sediment,
molluscs from a tropical sandy beaches, Estuar. cstl shelf Sci. 13, 579-589.
J. expo mar. BioI. Ecol. 4, 9-35. Bonnet DD (1946) The Portuguese
Ansell AD and Trueman ER (1973) The man-of-war as a food source for the sand
energy cost of migration of the bivalve crab (Emerita pacifica), Science 130,
Donax on tropical sandy beaches, Mar. 148-149.
Behav. Physiol. 2, 21-32. Boucher G and Chamroux S (1976)
Ansell AD and Lagardere F (1980) Bacteria and rneiofauna in an
Observations on the biology of Donax experimental sand ecosystem I. Material
trunculus and D. vittatus at Ile--d' and preliminary results, J. expo mar.
Ole ron (French Atlantic Coast), Mar. BioI. Ecol. 24, 237-249.
BioI. 57, 287-300. Bousefield EL (1970) Adaptive
Ansell AD, McLusky DS, Stirling A and radiation in sand burrowing amphipod
Trevallion A (1978) Production and crustaceans, Cheseapeake Sci. 11,
energy flow in the macrobenthos of two 143-154.
sandy beaches in South west India, Proc. Bowman 'l'E (1971) 'l'he distribution of
Roy. Soc. Edinburgh 76B, 269-296. calanoid copepods off the south eastern
Ansell AD, Sivadas P, Narayanan B, United States between Cape Hatteras and
Sankaranarayanan VN and Trevallion A Southern Florida, Smithsonian Contrib.
(1972a) The ecology of two sandy beaches zool. 96, 58pp.
in South West India I. Seasonal changes Brattegard T (1969) Mysidacea from
in physical and chemical factors and in shallow water in the Bahamas and
366

Southern Florida. Part I. Sarsia 39, biopedologiques, Vie Milieu 26, 1-25.
17-105. Caine EA (1974) Feeding of Ovalipes
Brown AC (1964) Food relationships on guadulpensis (Saussure) (Decapoda:
the intertidal sandy beaches of the Cape Brachyura: Portunidae) and morpholog ical
Pe n ins u 1 a" S. Af r. J. Sci. 60 , 35 - 4l. adaptations to a burrowing existence.
Brown AC (1971a) The ecology of the BioI. Bull. 147, 550-559.
sandy beaches of the Cape Peninsula, Carlisle JG, Schott JH and Abrahamson
South Africa Part I: Introduction, NJ (1960) The barred surfperch in
Trans. Roy. Soc. S. Afr. 39, 247-279. southern california, Calif. Dept. Fish
Brown AC (1971b) The ecology of the Game. Bull. 109.
sandy beaches of the Cape peninsula, Cassie Rr.1 and Cassie V (1960) Primary
South Africa Part 2: The mode of life production in a New Zealand west coast
of Bullia (Gastropoda; prosobranchiata), phytoplankton bloom, New Zealand J. Sci.
Trans. Roy. Soc. S. Afr. 39, 281-319. 3, 173-199.
Brown AC (1973) The ecology of the Chamroux S (1965) contribution a
sandy beaches of the Cape peninsula, l' etude du pouvoir deni trif iant des
South Africa Part 4: Observation on two sediments marins, Insti t. Pasteur. Anal.
inter-tidal isopoda, Eurydice 109, 424-442.
longicornis (Studer) and Exosphaeroma Chamroux S, Boucher G and Bodin P
truncatitelson (Barnard), Trans. Roy. (1977) Etude experimental d'un ecosystem
Soc. S. Afr. 40, 381-404. sableux. II. Evolution de populations
Brown AC (1982) The biology of sandy de bacteries et de meiofaune,
beach whelks of the genus Bullia Helgolander wiss. Meeresunters. 30,
(Nassariidae), Oceanogr. Mar. BioI. Ann. 163-177.
Rev. 20, 309-361. Chelazzi G and Ferrara F (1978)
Brown B (1982) Spatial and temporal Research on the coast of Somalia. The
distribution of a deposit-feeding poly- shore and dune of Sar Uanle. 19.
chaete on a heterogeneous tidal flat. Zonation and activity of terrestrial
J.exp. mar. BioI. Ecol. 65, 213-227. isopods (Oniscoida). Monitore Zool.
Brown R (1981) Saccocirridae ital. Suppl. 9, 189-219.
(Annelida: Archiannel ida) f rom the Chelazzi G, Chelazzi Land ];'ocardi S
central coast of New South Hales, Aust. (1983) Dynamic zonation of staphylinoid
J. Mar. Freshwat. Res. 32,439-456. beetles (Coleopatera: Staphylinoidea) on
Brown AC and Talbot MS (1972) The a sandy beach in East Africa. In
ecology of the sandy beaches of the Cape McLachlan A and Erasmus T, eds. Sandy
Peninsula, South Africa Part 3: A study Beaches as Ecosystems, The Hague, Junk.
of Gastrosaccus psammodytes Tattersall Christie ND (1976) A numerical
(Crustacea: Mysidacea), Trans. Roy. anaylsis of the distribution of a
Soc. S. Afr. 40, 309-333. shallow sublittoral sand macrofauna
Brunton PM (1978) Toheroa predation by along a transect at Lamberts Bay, South
black-backed gulls on Dargaville beach Africa, Trans. Roy. Soc. S. Afr. 42,
North Auckland, New Zealand, Notornis 149-172.
25, 128-140. Clark RB, Alder JR and r.1cIntyre AD
Buchanan JB (1966) The biology of (1962) The distribution of Uephytys on
Echinocardium cordatum (Echinerdermata: the Scottish coast, J. Anim. Ecol. 31,
Spatangoidea) from different habitats. 359-272.
J. mar. bioI. Ass. U.K. 46,97-114. Clark RB and Milne A (1955) -:Llhe
Burger J, Howe MA, Caldwell D and sublittoral fauna of two sandy bays on
Chase J (1977) Effects of tide cycles on the Isle of Cumbrae, Firth of Clyde, J.
habitat selection and habitat mar. BioI. Ass. U.K. 34, 161-180.
partitioning by migrating shore birds, Clutter RJ (1967) Zonation of
Auk 94, 743-758. nearshore mysids, Ecology 48, 200-208.
Burger J, Haan DC and Chase J (1979) Cockcroft A (1979) Observed trends in
Aggressive interaction in mixed species the surf zone zooplankton of Kings
flocks of migrating shore birds, Anim. Beach, Unpubl. Honour project,
Behav. 27, 459-469. University of Port Elizabeth.
Bush LF (1966) Distribution of sand Cockcroft A (1983) Aspects of the
fauna in beaches at Miami, Florida, biology of the swimming prawn
Bull. Mar. Sci. 16, 58-75. Macropetasma africanus (Balss). M.Sc
Cadbury CJ (1978) The beached bird dissertation, University of Port
survey and other sea bird surveillance, Elizabeth.
Ibis 120, 119-120. Coe ,iC (1955) Ecology of the bean clam
Cahet G (1976) Biogeochimie du carbone Donax gouldi on the coast of Southern
et du soufre en sediments litteraux California, Ecology 36, 512-514.
Mediterraneens. (Golfe du Lion): Mod~lles Collos Y and Lewin J (1974) Blooms of
 367

surf zone diatoms along the coast of the Florida, u.S. Army Corps Engineers,
Olympic peninsula, Washington. IV Nitrate Misc. Rep. 82-2, 94pp.
reductase activity in natural popu- Cupka DM (1972) A Survey of the
lations and laboratory cultures of icthyofauna of the surf zone in South
Chaetoceros armatum and Asterionella Carolina. S. Carolina Hildlife Mar. Res.
socialis, Mar. BioI. 25, 213-22l. Dept. Tech. Rep 4, 14pp.
Collos Y and Lewin J (1976) Blooms of Dahl E (1952) Some aspects of the
surf-zone diatoms along the coast of the ecology and zonation of the fauna on
Olympic Peninsula, Washington 7: Var- sandy beaches, Oikos 4, 1-27.
iation of the carbon to nitrogen ratio Dale NG (1974) Bacteria in intertidal
in field samples and laboratory cultures sediments: factors related to their
of Chaetoceros armatum, Limnol. distribution, Lirunol. Oceanogr. 19,
Oceanogr. 21, 219-225. 509-518.
Corey S (1970) The quantitative Dauer DM and Simon JL (1975) Lateral
distribution of Cumacea (Crustacea: or along-shore distribution of the
peracarida) in Kames Bay, Scotland, Can. polychaetons annelids of an intertidal
J. Zool. 48, 925-930. sandy habitat, Mar. BioI. 31, 363-370.
Coull BC (1970) Shallow water Davenport CB (1903) The animal ecology
meiobenthos of the Bermuda Platform, of the Cold Spring sand spit, with
Oecologia (Berl) 4, 325-357. remarks on the theory of adaptation,
Craig PC (1970) Behaviour and Univ. Chicago Decennial Publ. IG,
distribution of the intertidal sand 157-176.
beetle Tinopinus pictus (Coleoptera: Davies JL (1972) Geographical
Staphylinidae). Ecol. 51, 1012-1017. variation in coastal development,
Craig PC (1973) Behaviour and distri- London: Longman.
bution of the sand beach amphipod Davies S (1981) Development and
Orchestoidea corniculata, Mar. BioI. 23, behaviour of little tern chicks, British
101-109. Birds 74, 291-298.
Crane J (1941) On the growth and Davis N and Von Blaricom GR (1978)
ecology of brachyuran crabs of the genus Spaticll and temporal heterogeneity in a
Ocypode, Zoologica 26, 298-309. sand bottom epifaunal community of
Crisp DJ and Williams R (1971) Direct invertebrates in shallow water, Limnol.
measurement of pore-size distribution on Oceanogr. 23, 417-427.
artificial and natural deposits and pre- Day JH, Field J and Montgomery MP
diction of pore space accessible to (1971) The use of numerical methods to
interstitial organisms, Mar. BioI. 10, determine the distribution of the
214-226. benthic fauna across the continental
Croker RA (1967a) Niche diversity in shelf of North Carolina. J .anim. Ecol.
five! sympatric species of intertidal 40, 93-125.
amphipods (Crustacea: Haustoriidae), Dean HK (1981) Effects of the de'Jree
Ecol. Monogr. 37, 173-199. of wave action upon intertidal
Croker RA (1967b) Niche sepcificity of meiofaunal abundances, Am Zool. 21, 972.
Neohaustorius schmitzi and Haustorius Delamare-Demoutteville C (1960)
sp. (Crustacea: Amphipoda) in North Biologie des eaux souterraines
Carolina, Ecology 48, 971-975. littorales et continentales, Paris:
Croker RA (1968) Distribution and Hermann 740 pp.
abundance of some intertidal sand beach De Villiers G (1974) Growth,
amphipods accompanying the passage of population dynamics, a mass mortality
two hurricanes, Chesapeake Sci. 9, and arrangement of white sand mussels,
157-162. Donax serra, Roding, on Leaches in the
Croker RA (1970) Intertidal sand South-Hestern Cape Province, Invest.
macrofauna from Long Island, New York, Rep. Div. Sea Fish. S. Afr. 109, 1-31.
Chesapeake Sci. 11, 134-137. Dexter DM (1967) Distribution and
Croker RA, Hager RP and Scott KJ niche diversity of haustoriid amphipods
(1975) Macroinfauna of northern new in North Carolina, Chesapeake, Sci. 8,
England marine sand II. Amphipod 187-192.
dominated intertidal communities, Can. Dexter DM (1969) structure of an
J. Zool. 53, 42-51. intertidal sandy beach community in
Cubit J (1969) Behaviour and physical North Carolina, Chesapeake Sci. 10,
factors causing migration and 93-98.
aggregation of the sand crab Emerita Dexter D/'J (1971) Life history of the
analoga (Stimpson), Ecology 50, 118-123. sandy beach amphopod Neohaustorius
Culter JK and Mahadevan S (1982) Long schmitzi (Crustacea, Haustoriidae), /'Jar.
term effects of beach nourishment on the BioI. (Berl.) 8, 232-237.
benthic fauna of Panama City beach, Dexter DM (1972) Comparison of the
368

community structures in a pacific and an Dye Ail (1980) Tidal fluctuations in

Atlantic panamanian sandy beach, Bull. biological oxygen demand in sandy
mar. Sci. 22, 449-462. beaches, Estuar. cstl. mar. Sci. II, 1-8.
Dexter DM (1974) Sandy beach fauna of Dye AH (1981) A study of benthic
the Pacific and Atlantic coasts of costa oxygen consumption on exposed sandy
Rica and Columbia, Rev. BioI. 1'rop. 22, beaches, Estuar. cstl. Shelf. Sci. 13,
51-66. 671-680.
Dexter DM (1976) The sandy beach fauna Dye AH (1983) A synopsis of Lenthic
of Mexico, Southwest Naturalist 20, oxygen consumption stUdies on salidy
479-485. beaches. In ~jcLachlan A Clnd Erasmus 'J.',
Dexter DM (1977a) Natural history of eds. Sandy Leaches as ecosystems, The
the pan-American sand beach isopod, Hague, Junk.
Excirolana braziliensis (Crustacea: Dye AH, McLachlan A and Hooldridge 'J.'
Malacostraca), J. Zool. Lond. 183, (1981) The ecology of sandy beaches in
103-109. Natal, South Africa, S. Afr. J. Zoo I.
Dexter DM (1977b) A natural history of 16, 200-209.
the sand dollar Encope stokesi L Agassiz Edgren RE (1959) Coquinas (Donax
in Panama, Bull. Mar. Sci. 27, 544-551. variabilis) on a Florida Leach, Ecology
Dexter DM (1978) The infauna of a 40, 498-502.
subt idal, sand-bot tom communi ty in Edwards RRC (1973a) Production ecology
Imperial Beach, California, Calif. Fish of two Caribbean marine ecosystems. II.
Game 64, 268-279. Metabolism and energy flow, Estuar.
Dexter DM (1979) Community structure cstl. mar. Sci. I, 319-333.
and seasonal variation in intertidal Edwards RRC (1973b) Production ecology
panamanian sandy beaches, Estuar. cstl. of two Caribbean marine ecosystems. 1
mar. Sci. 9, 543-558. Physical environment and fauna, Estuar.
Dexter DM (1981) Intertidal fauna of cstl. mar. Sci. I, 303-318.
Israeli and Sinai sandy beaches of the Edwards RRC and Steele JH (1969) 'l'he
Mediterranean and Red Sea, Bull. Mar. ecolo~y of O-group plaice and common
Sci. 31, 812. dabs at Loch Ewe. I. population and
Dexter DM (1983) Community structure food, J. expo mar. BioI. Ecol. 2,
of intertidal sandy beaches in New South 215-238.
Wales, Australia. In McLachlan A and Edwards RRC, Steele JH and Trevallion
Erasmus T, eds. Sandy Beaches as A (1970) The ecolo~y of O-group plaice
Ecosystems, The Hague, Junk. and common dabs at Loch Ewe. III.
Dexter DM and Ebert TA (1975) A Prey-predator experiments with plaice,
natural history study of Encope grandis J. expo mar. BioI. Ecol. 4, 156-173.
and Mellita grantii, two sand dollars in Eiford IE (1965) Aggregation in the
the northern Gulf of California, Mexico, sand crab Emerita analoga (Stimpson) J.
Mar. BioI. 32, 397-407. Anim. Ecol. 34, 63-75.
Dillery DG and Knapp LV (1970) Long- Efford IE (1966) Feeding in the sand
shore movements of the sand crab Emerita crab Emerita analoga (Decapoda, Hippidae)
analoga (Decapoda, Hippidae). Crustaceana 10, 167-182.
Crustaceana 18, 233-240. Efford IE (1972) The distribution of
Donn TE (1980) population dynamics of the sand crabs, Hippa strigillata
the sandy beach amphipod Haustorius (Stimpson) and Hippa PaCifica (Dana) in
canadensis in Southern Maine U.S.A., Am. the i:astern pacific ucean (Decapoda,
Zool. 20, 812. Anomura), Crustaceana 23, 119-122.
Du Preez HH (1981) The biology of the Efford IE (1976) Distribution of the
three-spot swimming crab, Ovalipes sand crabs in the genus Emerita
punctatus (de Haan) (Brachyura: (Decapoda, Hippidae), Crustaceana 30,
Portunidae) with special reference to 169-183.
feeding. M.Sc. dissertation, University Eleftheriou A and Nicholson I-ID (1975)
of Port Elizabeth. The effects of exposure on beach fauna,
Dwivedi SN, Ayyappan S and Rahim A Cah. Bio!. Mar. 16, 695-710.
(1973) Ecology and production of Eleftheriou A and MCIntyre AD (1976)
intertidal macrofauna during monsoon in The intertidal fauna of sandy beaches -
a sandy beach at Calangute, Goa, J. Mar. a survey of the Scottish Coast, Scottish
BioI. Ass. India 15, 274-284. Fisheries Research, Report No.6. 1-61.
Dye AH (1979a) Measurement of Eleftheriou A and Jones DA (1976) 'l'he
biological oxygen demand in sandy genus Erydice on the west coast of
beaches, S.Afr. J. Zool. 14,55-60. India, J. Zoo I. Lond. 178, 385-394.
Dye AH (1979b) Quantitative estimation Eleftheriou A, Holdich DM and Harrison
of Protozoa from sandy substrates, K (1980) The systematics and ecology of
Estuar. cstl. mar. Sci.8, 199-204. a new ~enus of Isopod (Sphaeromatidae)
 369

from the west coast sandy beaches of along a transect across False Bay, South
India, Est. cstl. mar. Sci. 11, 251-262. Africa, J. EXp. Mar. BioI. Ecol. 7,
Eltringham SK (1971) Life in mud and 215-253.
sand, London, English Universities Press. Figueras A (1956) Moluscos de las
Emberton KC (1981) Season depth playas de la Ria de Vigo. 1. Ecologia y
relations in subtidal meiofauna of Cape distribution, Inv. pesq. 5, 51-88.
Cod Bay, Estuaries 4, 121-126. Fincham AA (1970) Amphipods in the
Enr ight JT (1963) The tidal rhythm of surf plankton. J. mar. bioI. Ass. U. K.
activity of a sand beach amphipod, z. 50, 177-198.
vergl. Physiol. 46, 276-313. Fincham AA (1971) Ecology and
Enright JT (1965) Entrainment of a population studies of some intertidal
tidal rhythm, Science 147, 864-867. and sub-littoral sand-dwelling
Enright JT (1972) A virtuoso isopod: amphipods, J. mar. bioI. Ass. U.K. 51,
circa-lumar rhythms and their tidal fine 471-488.
structure, J. compo Physiol. 77, 141-162. Fincham AA (1973) Rhythmic swimming
Epelde-Aguire A and Lopez MT (1975) behaviour of the t-Iew Zealand sand beach
Zonacion en el sustrato arenoso de Playa isopou pseudaega punctata, J. expo mar.
Blance, Bahia de Coronel y observaciones BioI. Eco!' 11, 229-237.
sobre crustaceos poco f recuentes, BioI. Fincham AA (1974) Intertidal sand-
Soc. bioI. de Concepcion 49, 161-170. dwelling );>aracarid fauna of stewart
Escofet A, Gianuca N, Maytia Sand Island, N.Z. J. mar. Freshwat. Res. 8,
Scarabino V (1979) Playas arenosos de 1-14.
Atlantico Sudoccidental entre los 29 0 y Fincham AA (1977) Intertidal sand
4JoLS: Consideraciones generates y dwelling peracarid fauna of North
esquena biocenologico, Unesco seminar, Island, New Zealand, N.Z. J. mar.
9-12 May 1978, pp 245-258. Freshwat. ,Res. 11, 677-696.
Faubel A (1976) population dynamics Fish S (1970) The l>iology of Eurydice
and life cycles of interstitial Acoela pulcha (Crustacea: Isopoda) , J. mar.
and Macrostomida (Turbellaria). bioI. Ass. U.K. 50, 753-7585.
Investigations of a middle lotic sandy Fish JD and Fish S (1972) The swimming
beach of the Island of Sylt, Microfauna rhythm of Eurydice pulchra Leach and a
Meeresbodens 56, 1-107. possible explanation of intertidal
Feder HM and paul AJ (1980) Seasonal mi<jration, J'. expo mar. BioI. Ecol. 8,
trends in meiofaunal abundance on two 195-200.
beaches in Port valdaz, Alaska, Syesis Fishelson Land Loya Y (1968)
13, 27-36. preliminary observations on a population
Feller RJ (1980) Quantitative cohort of Gastrosaccus sanctus (Mysidacea,
analysis of a sand-dwelling meiobenthic Gastrosaccinae) on a Mediterranean sandy
harpacticoid copepod, Est. Cstl. mar. beach of Israel, Crustaceana 15, 149-152.
Sci. 11, 459-476. Fitch JE (1950) The pismo clam, Calif.
Fellows DP (1975) On the distribution Fish Game 36, 285-312.
of the Hawaian ghost crab Ocypode laevis Fricke AH and Fleming B (1983) Select-
Dana, Pac, Sci. 29, 257-258. ive microhabitat coolonisation by inter-
Fenchel T (1972) Aspects of decomposer stitial meiofauna as a function of grain
food chains in marine benthos, Deutsche size. In McLachlan A and Erasmus T eds.
Zoologische Geselschaft Verhandlungen Sandy Beaches as Ecosystems, The Hague,
65, 14-23. Junk.
Fenchel TM (1978) The ecology of Fuchs E (1977) Predation and
micro- and meiobenthos, Ann. Rev. Ecol. antipredator l>ehaviour in a mixed colony
Syst. 9, 99-12!. of terns Sterna sp. and Black-headed
Fenchel T and Jansson B-O (1966) On gulls Larus ridibundus with special
the vertical distribution of the reference to the Sandwich tern Sterna
microfauna in the sediments of a sandvicensis Ornis Scand. 8, 17-32.
brachish-water beach, Ophelia 3, 161-177. Fusaro C (1980) Diel distribution
Fenchel TM and Riedl RJ (1970) The differences in the sand crab, Emerita
sulfide system: a new biotic community anagola (Simpson) (Decapoda, Hippidea),
under the oxidized layer of marine sand Crustaceana 39, 287-300.
bottoms, Mar. BioI. 7, 255-268. Gauld DT and Buchanan JB (1956) The
Ferrara F (1974) Researches on the fauna of sandy beaches in the Gold
coast of Somalia. The shore and the Coast, Oikos 7, 293-301.
dune of Sar Uanle 3: Terrestrial George RW and Knott ME (1965) The
isopods, Monitore Zool. ital., Suppl 5, Ocypode ghost crabs of western Australia
191-220. (Crustacea, Brachyura), Trans. Roy. Soc.
Field JG (1971) A numerical analysis west. Austr. 48, 15-21.
of changes in the soft-bottom fauna Gerlach SA (1971) On the importance of
370

meiofauna for benthos communities. Gunter G and Lyles CH (1979) Localized

Oecologia (Berl) 6, 176-190. plankton blooms and jubilees on the Gulf
Gianuca NM (1983) A preliminary Coast, Gulf Res. Rep. 6, 297-299.
account of the ecology of sandy beaches Hager RP and Croker RA (1980) The
in Southern Brazil. In McLachlan A and sand-burrowing amphipod Amphiporeia
Erasmus T, eds. Sandy Beaches as virginiana Shoemaker 1933 in the tidal
Ecosystems, The Hague, Junk. plankton, Can. J. zool. 58, 860-864.
Gibson RN (1973) The intertidal Hails AJ and Yaziz S (9182) Abundance,
movements and distribution of young fish breeding and growth of the ocypodid crab
on a sandy beach with special reference Dotilla myotiroides (Milne-Edwards) on a
to the plaice Pleuronectes plateisa, J. west malaysian beach, Estuar. cstl shelf
expo mar. BioI. Ecol. 12, 79-102. Sci. 15, 229-230.
Giere 0 (1975) population structure, Haley SR (1982) Zonation by size of
food relations and ecological r61e of the pacific mole crab, Hippa pacifica
marine oligochaetes, with special Dana (Crustacea: Anomura: Hippidae), in
reference to meiobenthic species, Mar. Hawi i, J. exp. mar. BioI. Ecol. 58,
BioI. 31, 139-150. 221-231.
Giere 0 (1982) Habitat conditions and Hall JR and Hessler RR (1971) Aspects
distribution pattern of the gutless in the population dynamics of
01 igochaete Phallodr illus leukodermatus, Derocheilocoris typica (Mystacocorida,
Mar. Ecol. Prog. Ser. 8. 291-299. Crustacea), Vie Mileu 22, 305-326.
Glynn PH, Dexter DM and Bowman TE Hammer ,m, Smyth ~j and Mulford 1:D
(1975) Excirolona braziliensis and a (1968) Orientation of the sand beach
Pan-Amer ican sand beach isopod: isopod Tylos punctatus, Anim. Behav. 16,
taxonomic status, zonation and distri- 405-409.
bution, J. Zool. Lond. 175, 509-52l. Hammer WM, Smyth M and Hulford ED
Gnaiger E, Gluth G and Hieser H (1969) The behaviour and life history of
(1978), pH fluctuation in an intertidal a sand-beach isopod 1'ylos punctatus,
beach in Bermuda, Limnol. Oceanogr. 23, Ecology 50, 442-453.
851-857. Harris RP (1972a) Horizontal and
Gray JS (1971) The effects of vertical distribution of the intertidal
pollution on sand meiofauna communities, harpacticoid colepods of a sandy beach,
Thalassia Jugoslovica 7, 79-86. J. Mar. BioI. Ass. U.K. 52, 375-387.
Gray JS and Johnson RM (1970) The Harris RP (1972b) Seasonal changes in
bacteria of a sandy beach as an the meiofauna population of an
ecological factor affecting the intertidal sand beach, J. Mar. BioI.
interstitial gastrotich Turbanella Ass. U.K. 52, 382-403.
lyalina, Schultse. J. expo mar. BioI. Hartnoll RG (1983) The species urea
Ecol. 4, 119-123. relationship on a sandy Leach. In
Gray JS and Rieger RM (1971) A quanti- McLachlan A and l:rasmus 1', eds. Sandy
tative study of the meiofauna of an Beaches as Ecosystems, The Hague, Junk.
exposed sandy beach, at Robin Hood's Hartwig 1: (1982) Effects of
Bay, Yorkshire, J. Mar. Biol. Ass. U.K. temperature and oxygen on the
51, 1-19. distribution of Diophrys scutium
Griffiths CL and Stenton-Dozey J (Ciliophora, Hypotrichida) on a
(1981) The fauna and rate of degradation temperate sand beach, an ecophysio-
of stranded kelp, Estuar. cstl shelf, logical study. Cah. BioI. Mar. 23,
Sci. 12, 645-653.' 227-242.
Griffiths CL, Stenton-Dozey J and Koop Hartwig E and Parker JG (1977) On the
K (1983) Kelp wrack and energy flow systematics and ecology of interstitial
through a sandy beach. In McLachlan A ciliates of sandy beaches in North
and Erasmus T eds. Sandy Beaches as Yorkshire, J. mar. bioI. Ass. U.K. 57,
Ecosystems, The Hague, Junk. 735-760.
Grindley JR and Nel E (1968) Mussel Hartwig E, Gluth T and Wieser H (1977)
poisoning and shellfish mortality on the Investigations on the ecophysiology of
west coast of South Africa. S. Afr. J. Geleia migriceps Kahl (Ciliophora,
Sci. 64, 420-422. Gymnostomata) inhabiting a sandy beach
Gunter G (1958) population studies of in Bermuda, Oecologia 31, 159-175.
shallow water fishes of an outer beach Hayes WB (1974) Sand-beach energetics:
in S. Texas, Texas Inst. Mar. Sci. Publ. importance of the isopod Tylos
5, 186-193. punctatus, Ecology 55, 838-837.
Gunter G (1979) Notes on sea beach Hayes ,m (1977) Factors affecting the
ecology'~ Food sources on sandy beaches distribution of Tylos punctatus (Isopoda,
and localized diatom blooms bordering Oniscoideo) on beaches in Southern
gulf beaches, Gulf Res. Rep 6, 305-307. California and Northern /!lexico, Pacific
 371

Science 31, 165-187. fauna on tidal and atidal beaches, Mar.

Hedgpeth JW (1957) Sandy beaches, In BioI. 34, 77-83.
Hedgpeth JHW, ed. Treatise on Marine Hutchings L, Nelson G, Hortsman D and
Ecology and Palaeociology Geol. Soc. Tarr R (1983) Interactions between
America Memoir 67 1, 587-608. coastal plankton and sand mussels along
Hill GW (1982) Orientation of ghost the Cape coast, South Africa. In
crab Ocypode quadrata, burrows as an McLachlan and Erasmus T, eds. Sandy
indicator of shoreline position and wind Beaches as Ecosystems, The Hague, Junk.
direction. Tex. J. Sci. 34, 23-34. Hummon WD, F'leeger JW and Hummon MR
Hill GW and Hunter RE (1973) Burrows (1976) Meiofauna macrofauna
of the ghost crab Ocypode guadrata interaction 1. Sand beach meiofauna
(Fabricius) on the barrier islands, affected by maturing Limulus eggs,
south-central Texas coast, J. Sedim. Chesapeake Sci. 17, 297-299.
Petrol. 43, 24-30. Inman DL and Brush BM (1973) The
Hill GW and Hunter RE (1976) Coastal challenge, Science 181, 20-32.
Interaction of biological and geological Irwin 'l'H (1973) The intertidal
processes in the beach and nearshore behaviour of the bean clam, Donax
environments, Northern Poole Island, gouldii Dall 1921, veliger 15, 206-2~
Texas, In Davis and Ethrin, eds. Beach Jansson BO (1967) The significance of
Nearshore Sedimentation. 169-187. grain size and pore water content for
Hockey PAR, Siegfried WR, Crowe AA and the interstitial fauna of sandy beaches,
Cooper J (1983) Ecological structure and Oikos 18, 311-322.
energy requirements of the sandy-beach Jansson BO (1968a) ~uantitative and
avifauna of Southern Africa. In experimental studies of the interstitial
McLachlan A and Erasmus T, eds. Sandy fauna in four Swedish sandy beaches,
Beaches as Ecosystems, The Hague, Junk. Ophelia 5, 1-71.
Hockin DC (1982) The effects of Jansson BO (1968b) Studies on the
sediment particle diameter upon the ecology of the interstitial fauna of
meiobenthic copepod community of an marine sandy beaches, Rep. Monogr. non
intertidal beach. A field and Serials 16.
laboratory experiment. J. Anim. Ecol. Jaramillo E (1978) zonacion estructura
51, 555-572. de la comunidad macrofaunistica en
Hogue EW and Miller CB (1981) Effects player de arena de Sur de Chile (Mehuin,
of sediment microtopography on valdi via), Studies on Neotropical r'auna
small-scale spatial distributions of and Environment 13, 71-92.
meiobenthic nematodes, J. exp. mar. Jaramillo E, Stotz H, Bertran C.
BioI. Ecol. 181-191. Navarro J, Roman C and Varela C (1980)
Holanov SH and Hendrickson JR (1980) Actividad locomotrix de Orchestoidea
The relationship of sand moisture to turberculata (Amphipoda, Talitidae)
burrowing depth of the sand beach isopod sobre la superficie de una playa arenosa
Tylos punctatus, J. expo mar. BioI. del sur de Chile (Mehuin, provincia de
Ecol. 46, 81-88. Valdivia), Studies on Neotropical Fauna
Holdich DM and Jones DA (1973) The and Environment 15, 9-33.
systematics and ecology of a new genus Jijina JL and Lewin J (in prep, a)
of sand beach isopod (Sphaeromatidae) persistent blooms of surf diatoms along
from Kenya, J. Zool. Lond. 171, 385-395. the pacific cost, USA. 1. Physical
Holland AF and polgar TT (1976) characteristics of the coastal region in
Seasonal changes in the structure of an relation to the distribution and
intertidal community, Mar. BioI. 37, abundance of the species.
341-348. Jijina JL and Lewin J (in prep, b)
Holme NA (1954) The ecology of the persistent blooms of surf diatoms along
British species of Ensis, J. mar. bioI. the pacific coast, USA. II. Use of
Ass. U.K. 33, 145-17-2-.--- regression analysis to assess the
Hulings NC (1971) A comparative study relationships between environmental
of the sand beach meiofauna of Lebanon, variables and cell populations along
Tunisia and Morocco, Thalassia Jugosl. Oregon and ~Iashington beaches (1977 and
7, 117-122. 1978).
Hulings NC (1974) A temporal study of Johannes RE (1980) 'I'he ecological
Lebanese sand beach meiofauna, Cah. significance of the submarine discharge
BioI. Mar. 15, 319-335. of groundwater, Mar. EcoI. Prog. Ser. 3,
Hulings NC and Gray JS (1971) A manual 365-373.
for the study of meiofauna, Smithsonian Johnson ~lS (1976a) Biology and
Contrib. Zool. 78, 1-84. population dynamics of the' intertidal
Hulings NC and Gray JS (1976) Physical isopod Cirolana harfordi Mar. BioI. 36,
factors controlling abundance of meio- 343-350.
372

Johnson HS (1976b) population and reproductive cycles in the sand

energetics of the intertidal isopod beach isopod Excirolana chil toni, BioI.
Cirolana harfordi, Mar. BioI. 36, Bull. (Hoods Hole) 143, 568-591.
351-357. Koepcke HII (1953) Contribucion al
Johnston, PT (1966) Mass mortality in conocimiento de la forma de vida de
a bivalve mollusc. Limnol. Oceanogr. 1, Ocypode gaudichaudi I·alne Edwards Y
429-431. Lucas (Decapoda, Crust):, publ. Mus.
Johnston PT (1968) population crashes Rist. Nat. 'Javier Prado' ser A 13.
in the bean clam, Donax gouldi, and Koepcke HN and Koepcke N (1952) Sobre
their significance to the study of mass el processo de transformacion de la
mortality on other marine invertebrates. materia organica en las playas arenosas
J. Invest. Pathol. 12, 349-358. del Peru, Revista de Ciencias Univ. Nac.
Jones A (1973) The ecology of young San Marcos 54, 5-29.
turbot Scopthalmus maximus at Barth, Koop K and Field JG (1980) The
Cardiganshire, Hales, J. Fish BioI. 5, influence of food availability on
367-383. population dynamics of a supralittoral
Jones DA (1971) The systematics and isopod, Ligia dilatata Brandt, J. expo
ecology of some sand beach isopods mar. BioI. Ecol. 48, 61-72.
(Crustacea, Eurydicidae) from the coast Koop K and Field JG (1981) Energy
of Kenya, J. Zool. (Lond). 165, 201-227. transformation by the supralittoral
Jones DA (1972) Aspects of the ecology isopod Ligia dilatata Brandt, J. expo
and behaviour of Ocypode ceratophthalmus mar. BioI. Ecol. 53, 221-233.
(pallas) and O. kuhlii (de Haan) Koop K and Griffiths CL (1982) The
(Crustacea, Ocyp-odidae), J. expo mar. relative significance of bacteria, meio-
BioI. Ecol. 8, 31-43. and macrofauna on an .exposed sandy
Jones DA (1974) The systematics and beach, Mar. BioI. 66, 295-300.
ecology of some sand beach isopods Koop K, Newell, RC and Lucas MJ
(family Cirolanidae) from the coasts of (1982a) Biodegradation and carbon flow
Saudi Arabia, Crustaceana 26, 202-211. based on kelp debris (Ecklonia Maxima)
Jones DA (1979) The ecology of sandy in a sandy beach microcosm, f.1ar. Ecol.
beaches in penang Malaysia with special Prog. Ser. 7, 315-326.
reference to Excirolana orientalis, Koop K, Newell RC and Lucas MI (1982b)
Estuar. cstl. mar. Sci 9, 677-682. Microbial regeneration of nutrients from
Joubert CSW (1981) The density of the decomposition of rnacrophyte debris
shore birds on a Natal sandy beach, on the shore, ~lar. Ecol. Prog. Ser. 9,
Ostrich 52, 190-192. 91-96.
Kamihira Y (1979) Ecological stUdies Kuyken L (1978) Beached bird surveys
of macrofauna on a sandy beach of in Belgium, Ibis 120, 122-123.
Hakodate Japan. 2. Distribution of Lasiak TA (1982) Structural and
peracarids and the factors influencing functional aspects of the surf-zone fish
their distribution, Bull. Fac. Fish, community in the Eastern cape, Ph.D.
Hokkaido Univ. 30, 133-143. thesis, University of Port Elizabeth.
Kamihira Y (1981) Life history of sand Lasiak TA (1983) The impact of surf
burrowing amphipod Haustorioides zone fish communities on faunal
japonicus (Crustacea, Dogielinotidae) , assemblages associated with sand
Bull Fac. Fish. Hokkaido Univ. 32, beaches. In llcLachlan A and Erasmus T,
338-348. eds. Sandy beaches as Ecosystems, The
Kastendiek J (1982) Factors Hague, Junk.
determining the distribution of the sea Lasker R, Wells JBJ and MCIntyre AD
pansy, Renilla kollikeri, in a subtidal (1980) Growth, reproduction, respiration
sandbottom habitat, Oecologia 52, and carbon utilization of the
340-347. sand-dwelling harpacticoid copepod,
Kensley BF (1972) behavioural Asellopsis intermedia, J. mar. bioI.
adaptations of the isopod Tylos Ass. U.K. 50, 147-160.
granulatus Krauss, Zool. afro 7, 1-4. Lassere P (1976) tletabolic activities
Kensley BF (1974) Aspects of the of benthic microfauna and meiofauna.
biology and ecology of the genus Tylos, Recent advances and review of suitable
Ann. S. Afr. Mus. 65, 401-471. methods of analysis, In New York, 'The
Khiyama HM and Makemson JC (1973) Sand Benthic Bounday Layer'. ed. T.N. McCave
beach bacteria: numeration and Plemmum Press.
characterization, Appl. Mircobiol. 26, Lawrence JM and Ferber J (1971) Sub-
293-297. strate particle size and the occurrence
King CAM (1972) Beaches and coasts, of Lovenia elongata (Echinodermata:
London, Arnold. Echinoidea) at Taba, Gulf of Elat (Red
Klapow LA (1972) Fortnightly moulting Sea), Israel J. Zool, 20, 131-138.
 373

Leber KM (1982a) Seasonality of Blooms of the surf-zone diatoms along

macroinvertebrates on a temperate, high the coast of the Olympic Peninsula.
wave energy sandy beach, Bull. Mar. Sci. Hashington V: Environmental conditions
32, 86-98. associated with the Llooms (1971-1972),
Leber KM (1982b) Bivalves (Tellinacea: Est. cstl mar. Sci. 3, 229-24l.
Donacidae) on a North Carolina beach: Lewin J, Chen C and Hruby T (1979a)
contrasting population size structures Blooms of surf-zone diatoms along the
and tidal migrations, Mar. Ecol. Prog. coast of the Olympic peninsula,
Ser. 7, 297-301. Hashington X: Chemical composition of
Lee JJ, Tietjen JH, Mastropaola C and the surf diatom Chaeterceros armatum and
Rubin H (1977) Food quality and the its major herbivore, the pacific
heterogeneous spatial distribution of razor-clam, siliqua patula, Mar. BioI.
meiofauna, HelgoUinder wiss. 51, 259-265.
Meeresunters. 30, 272-282. Lewin J, Colvin JR and McDonald KL
Lenanton RCJ, Robertson AI and Hansen (1980) Blooms of surf-zone diatoms along
JA (1982) Nearshore accumulations of the coast of the Olympic peninsula,
detached macrophytes as nursery areas Hashing ton. XI I. The clay coast of
for fish, Mar. Ecol. prog. Ser. 9, 51-57. Chaetoceros armatum 'l'. Hest. BioI. Mar.
Lewin J (1974) Blooms of surf zone 23, 333-34l.
diatoms along the coast of the olympic Lewin J, Eckman JE and Ware GN (1979b)
Peninsula, Hashington. II 1. Changes in Blooms of surf zone diatoms along the
the species composition of the blooms coast of the Olympic peninsula,
since 1925, Nova Hedwigia 45, 251-256. Hashington XI. Regeneration of ammoniulll
Lewin J (1977) persitant blooms of in the surf environment Ly the Pacific
surf diatoms along the Northwest Coast, razor clam, Siliqua patula, Mar. Biol.
The Maring Plant. In R Krauss, ed. 52, 1-9.
Biomass of the pacific NW Coast, 81-92. Lockwood SJ (1974) 'l'he settlement,
Lewin J (1978) Blooms of surf-zone distribution and movements of O-group
diatoms along the coast of the Olympic plaice Pleuronectes platessa (L. ) in
Peninsula, Hashington IX: Factors Filey Bay, Yorkshire, J. Fish BioI. 6,
controlling the seasonal cycle of 465-477.
nitrate in the surf at Copalis Beach Loesch HC (1957) Studies of the
(1971 through 1975), Est. cstl. mar. ecology of two species of Donax on
Sci. 7, 173-183. Mustang lsland, 'l'exas, TexaSUni v.
Lewin J and Norris RE (1970) Surf-zone Instit. Mar. Science PuLl. 4, 201-227.
diatoms of the coasts of Hashington and Longbottom HR (1970) The distribution
New Zealand (Chaetoceros arma tum) T. of Arenicola marina (L.) with particular
Hest and Asterionella spp., Phycologia reference to the effects of particle
9, 143-149. size and organic matter of sediments. J.
Lewin J and Mackas D (1972) Blooms of expo mar. BioI. Ecol. 5, 138-157.
surf-zone diatoms along the coast of the Lopez-Cotelo J. Vieitez JM and Diaz-
Olympic peninsula, Hashington. J. Pineda F (1982) Tipos de comunidades
Physiological investigations of bentonicas de la Playa de Punta 1 (Bahia
Chaetoceros armatum and Asterionella de Santander), Cah. BioI. Mar.23, 53-69.
socialis in laboratory cultures, Mar. l1acer C'l' (196 7) 'l'he tood webs in Red
BioI. 16, 171-181. Wharf Bay (N. Wales) with particular
Lewin J and Hruby T (1973) Blooms of reference to young plaice, Helgolander
surf zone diatoms along the coast of the wiss. Meeresunters. 15, 560-573.
Olympic Peninsula, Hashington. II. A Macquart-l1oulin C (1977) Le contr61e
diel periodicity in buoyancy shown by de l'emergence et des rages nocturned
the surf zone diatom species, chez les peracarides des plages de
Chaetoceros armatium T. West, Estuar. Mediterranee. Eurydice affinis Hansen
cst 1. rna r. Sci. 1, 101-105 . (Isopoda), Gastrosaccus mediterraneus
Lewin J and Rao VNR (1975) Blooms of Bacescu, Gastrosaccus spinifer (Goes)
surf-zone diatoms along the coast of the (Mysidacea), J. expo mar. BioI. Ecol.
Olympic Peninsula, Hashington VI: Daily 27, 61-81.
periodicity phenomena associated with Marsh BA and Branch GM (1979)
Chaeteceros armatum in its natural Circadion and circatidal rythms of
habitat, J. Phycol. 11,330-338. oxygen consumption in the sandy beach
Lewin J and Schaefer CT (1983) The isopod 'l'yles granulatus Krauss. J. expo
r61e of phytoplankton in surf mar. BioI. Ecol. 37, 77-89.
ecosystems. In I1cLachlan A and Erasmus Masse H (1963) Quelques donnees sur
T, eds. Sandy Beaches as Ecosystems, The l'economie alimentaire d'une biocenose
Hague, Junk. infralittorale. Rec. 'l'ran. St. l1ar. End.
Lewin J, Hruby T and Mackas D (1975) 31, 153-166.
374

Masse H (1970) Contribution a l'etude distribution, abundance and biomass of

de la macrofauna de peuplements des the marcofauna and meiofauna of four
sables fins infralittoraux des Cotes de sandy beaches, Zool. Afr. 12, 279-306.
provence. 1. La Baie de Bandol, Tethys McLachlan A (1977b) Studies on the
2, 783-820. psammolittoral meiofauna of Algoa Bay II.
Masse H (1971) Etude quantitative de The distribution, composition and
la macrofaune de peuplements des sables biomass of the Toleiofauna and macrofauna
fins infralittoraux II. La Baie du Prado communities, Zool. Afr. 12, 33-60.
(Golfe de Marseille), Tethys 3, 113-158. McLachlan A (1977c) Tbe larval
Masse H (1972) Quantitative investi- development and population dynamics of
gations of sand-bottom macrofauna along Derocheilocaris algoensis (Crustacea:
the Mediterranean north-west coast, Mar. Mystacocarida) Zool. Afr. 12, 1-14.
BioI. 15, 209-220. McLachlan A (1978) A quantitative
Matthews DC (1955) Feeding habits of analysis of the meiofauna and the
the sand crab Hippa pacifica (Dana), chemistry of the redox potential
Pac. Sci. 9, 382-386. discontinuity 2.one in a sheltered sandy
Mazure HGF and Branch GM (1979) A beach, Estuar. cstl. mar. Sci. 7,
preliminary analysis of bacterial 275-290.
numbers and biomass in Langebaan Lagoon, McLachlan A (1979a) Volumes of sea
Trans, Roy. Soc. S. Afr. 44, 43-54. water filtered through Eastern cape
McDermott J (1983) Food web in the Sandy beaches, S.Afr. J. Sci. 75, 75-79.
surf zone of an exposed sandy beach McLachlan A (1979b) Growth and
along the mid-Atlantic coast of the production of Donax sordidus Hanley
United States. In McLachlan A and (Mollusca, Lamellibranchia) on an 0pen
Erasmus T, eds. Sandy beaches as sandy beach in Algoa Bay 1., S. Af r. J.
ecosystems, The Hague, Junk. Zool. 14, 61-66.
McFarland WN (1963) Seasonal changes McLachlan A (l980a) ']'he definition of
in the numbers and biomass of fishes sandy beaches in relation to exposure: a
from the surf at Mustang Is. Texas, simple rating system, S.Afr. J. Sci. 76,
Publ. Inst. Mar. Sci. 9, 91-105. 137-138.
McGwynne LE (1980) A comparative McLachlan A (1980b) Intertidal
ecophysiological study of three sandy zonation of macrofauna and stratifi-
beach gastropods in the Eastern Cape, cation of meiofauna on high energy sandy
M.Sc. dissertation, University of Port beaches in the I::astern Cape, South
Elizabeth. Africa, 'l'rans. Roy. Soc. S.Afr. 44,
MCIntyre AD (1968) The macrofauna and 213-223.
meiofauna of some tropical beaches, J. McLachlan A (198 Oc) Occu r rence of
zool. Lond. 156, 377-392. ghost crabs Ocypode spp., in the eastern
MCIntyre AD (1969) Ecology of marine Cape, S. Afr. J. Sci. 15, 57-58.
meiobenthos, BioI. Rev. 44: 245-290. McLachlan A (1980d) Exposed sandy
MCIntyre AD (1970) The range of beaches as semi-closed ecosystems, l'lar.
biomass in intertidal sand with special Envi ron. Res. 4, 59-63.
reference to the bivalve, Tellina McLachlan A (1982) A model for the
tenuis, J. mar. bioI. Ass. U.K. 50, estimation of water filtration and
561-575. nutrient regeneration by exposed sandy
McIntyre AD (1971) Control factors on beaches, Mar. environ. Res. 6, 37-47.
meiofauna populations, Thalassia HcLachlan A (1983) The ecology of
Jugoslavia 7, 209-215. sandy beaches in the Eastern Cape, South
MCIntyre AD (1977) Sandy forshores, In Africa. In l'lcLachlan A and I::rasmus T,
Barnes RSK, ed. The Coastline, 31-47. eds. Sandy beaches as Ecosystems, The
MCIntyre AD and Eleftheriou A (1968) Hague, Junk.
The bottom fauna of a flatfish nursery HcLachlan A, I::rasmus '1' and Furstenberg
ground, J. mar. bioI. Assn. U.K. 48, JP (1977a) Migrations of sandy beach
113-142. meiofauna, Zool. Afr. 12, 257-277.
McIntyre AD, Munro ALS and Steele JH HcLachlan A, "inter PI::D and Botha L
(1970) Energy flow in a sand ecosystem, (1977b) Vertical and horizontal
In JH Steele, ed. Marine Food Chains, distribution of sub-littoral meiofauna
19-31. in Algoa Bay, South Africa, Har. BioI.
McIntyre AD and Murison DJ (1973) The 40, 355-364.
meiofauna of a flatfish nursery ground, HcLachlan A and Hanekom N (1979)
J. mar. bioI. Ass. U.K. 53, 93-118. Aspects of the biology, ecology, and
McIntyre RJ (1963) The supralittoral seasonal fluctuations in biochemical
fringe of New Zealand sand beaches, composi tion of Donax serra in the East
Trans. roy. Soc. N.Z. Gen. 1, 89-105. Cape, S. Afr. J. Zoo I. ~83-193.
McLachlan A (1977a) Composition, McLachlan A and van der Horst G (1979)
 375

Growth and production of two molluscs bacteria, Mar. Ecol. Prog. Ser. 3,
from an open sandy beach, S. Afr. J. 251-256.
Zoo1. 14, 194-210. Meyer-Reil L-A, Dawson R. Liebezeit G
McLachlan A, Dye AH and Van der Ryst P and Tiedge H (1978) Fluctuations and
(1979a) Vertical gradients in the fauna intractions of bacterial activity in
and oxidation of two exposed sandy sandy beach sediments and overlying
beaches, S.Afr. J. Zool. 14, 43-47. waters, Mar. BioI. 48, 161-171.
McLachlan A, Cooper C and van der Meyer-Reil L-A, Bolter M, Dawson R,
Horst G (1979b) Growth and production of Liebezeit G, Szwerinski Hand Vlolter K
Bu11ia (1980) Interrelationships between micro-
rhodostoma on an open sandy beach in biological and chemical parameters of
Algoa Bay, S.Afr. J. Zool, 14, 49-53. sandy beach sediments, a summer aspect,
McLachlan A, Hooldridge T and van der Appl. Env. Microbiol. 39, 797-802.
Horst G (1979c) Tidal movements of the Mielke vv (1976) Copepod ecology of a
macrofauna on an exposed sandy beach in sandy beach on the Isle of Sylt Hest
South Africa, J. Zoo1. Lond. 188, Germany, Mikrofauna Meeresbodens 59,
433-442. 1-86.
McLachlan A. Hooldridge T, Schramm M Mock H (1981) Diurodrilus subterraneus
and KUhn M (1980) Seasonal abundancej (Polychaeta: Dinophylidae) from sandy
biomass and feeding of shorebirds on beaches of the North Sea island of Sylt,
sandy beaches in the Eastern Cape, South Hest Germany, Helgol. Hiss.
Africa, Ostrich 51, 44-52. Meeresunters. 34, 329-336.
McLachlan A, Erasmus T, Dye AH, Moddle T and Ross ST (1981) Season-
Hooldridge T, van der Horst G, Rossouw ality of tishes occupying a surf zone
G, Lasiak TA and McGwynne L (1981a) Sand habitat in the northern Gulf of !-lexico,
beach energetics: an ecosystem approach Fish. Bull. U.S. 78, 911-922.
towards a high energy interface, Estuar. Moore CG (1979a) The distribution and
cstl shelf Sci. 13, 11-25. ecology of psammolittoral meiofauna
McLachlan A. Hooldr idge T and Dye AH around The Isle of Man, Cah. de BioI.
(1981b) The ecology of sandy beaches in Mar. 20, 383-415.
Southern Africa, s. Afr. J. Zool. 16, Moore CG (1979b) The zonation of
219-231. psammolittoral harpacticoid copepods
McLachlan A, Dye AH and Harty B around ~he isle of Man, J. mar. bioI.
(1981c) Simulation of the interstitial As s. U. K • 5 9, 711- 7 2 4 .
system of exposed sandy beaches, Estuar. Moran S (1972) Ecology of distribution
cstl. shelf Sci. 12, 267-278. of the sand-dwelling mysid Gastrosaccus
McLachlan A and Lewin J (1981) Obser- sanctus (Van Beneden, 1961) along the
vation on surf phytoplankton blooms Mediterranean sandy shore of Israel,
along the coasts of South Africa, Bot. Crustaceana Suppl. B., 357-361.
Mar. 24, 553-557. Moran Sand Fishelson L (1971)
McLachlan A and young N (1982) Effects Predation of a sand-dwelling mysid
of low temperature on burrowing rates of crustacean Gatrosaccus sanctus by plover
four sandy beach molluscs. J. expo mar. birds (Charadriidae), Har. BioI. 9,
BioI. Ecol. 65, 275-284. 63-64.
McLusky DS, Nair SA, Stirling A and Morgans JFC (1962) The benthic ecology
Bharoava R (1975) The ecology of a of False Bay. Part ~. Soft and rocky
Central Hest Indian beach with bottoms observed L>y diving and sampled
particular reference to Donax incarnatus, by dredging, and the recognition of
Mar. BioI. 30, 267-270. ----- grounds, Trans. Roy. Soc. S.Afr. 36,
Meadows PS and Anderson JG (1966) 288-334.
Microorganisms attached to marine and Mori S (1938) Donax semigranosus Dkr.
freshwater sand grains, Nature 212, and the experimental analysis of its
1059-1060. behaviour at the flood tide, Zoo1. ~lag •
Meadows PS and Anderson JG (1968) 'I'okyo 50, 1-12.
Microorganisms attached to marine sand Mou~za M and Chessel D (1976)
grains, J. mar. bioI. Ass. U.K. 48, Contribution a l'etude de la biologie de
161-175. Donax trunculus L. (Mollusque:
Meineke T and Hestheide H (1979) Tide Lomellibranche) dans I' Algero's: analyse
dependent migrations of the interstitial statisque de la dispersion Ie long d'une
fauna in a sand beach of the Island of plage en baie de bon Ismail, J. EXp.
Sylt, North Sea, Hest Germany, Mikrofauna Mar. BioI. Ecol 21, 211-221.
Meeresbodens 0(75) 203-236. Muir DG (1977) The biology of
Meyer-Reil L-A and Faubel A (1980) Talorchestia capensis (Amphipoda:
Uptake of organic matter by meiofanua Talitridae), including a population
organisms and interralationships with energy budget, M.Sc thesis, University
376

of Cape Town. Pearse AS, Humm HJ and Wbarton GVl

Munro ALS and Brock TD (1968) Distinc- (1942) Ecology of sand l>eaches at
tion between bacterial and algal Beaufort, N.C., Ecol. Monogr. 12,
utilization of soluble substances in the 135-190.
sea, J. gen. Microbiol. 51, 35-42. Penchaszadeb PB (1971) Observaciones
Munro ALS, Wells JBJ and McIntyre AD cuantitatives preliminares en playas
(1978) Energy flow in the flora and arenosas de la coast central del Peru,
meiofauna of sandy beaches, Proc. Roy. con especial referencia a las
Soc. Edinburgh 76B, 297-315. poblaciones de Muy-Muy (Emerita analoga)
Myers JP, Williams SL and Pitelka FA - (Crustacea, Anomura, Hippidae), Contr.
(1980) An experimental analysis of prey Inst. bioI. mar. Mar. de Plata 177, 3-19.
availability for sanderlings (Aves: Penchaszadeh P (1983) Subtidal sandy
Scolopacidae) feeding on sandy beach beach trophic structure in tbe area of
crustaceans, Can. J. Zool. 58, 156401574. punta Moron, Venezuela. In McLachlan A
Nair A (1978) Production ecology of a and Erasmus ~, eds. Sandy Beaches as
tropical sandy beach at Benaulim, Goa, Ecosystems, The Hague, Junk.
Ind. J. Mar. Sci. 7, 163-167. Penshaszadeb PE and Olivier SR (1975)
New TR (1968) The dispersal of insects Ecologia de una poblacion de "l>erberecbo"
across sandy beaches with special (Donax hanleyanus) en Villa Gesell,
reference to the larvae of Neuroptera, Argentina, Malacologia 15, 133-146.
Entomol. gaz. 19, 119-125. Philip KP (1972) The inter-tidal fauna
Nodot C (1976) Biological cycles of of the sandy beaches of Cochin, Proc.
some species of psammic harpacticoid Ned. Sci. Acad. India, 38B, 317-328.
copepods, Tethys 8, 241-248. Philips T (1970) Notes on the
Oliff WD, Gardner BD, Turner WD and predation of the asteroid Luidia on
Sharp JB (1970) The chemistry of the Donax navicula at the Plaza Almejas,
interstitial water as a measure of Baja California del Norte, ~jexico,
conditions in a sandy beach, Water Res. Tabulata 2, 16-18.
4, 179-188. Pichon M (1967) Contribution to the
Oliver JS, Darden JM and Slattery PN study of populations of tropical sandy
(1982) Phoxocephalid amphipod beaches in the vicinity of 'l.'ulear,
crustaceans as predators on larvae and Madagascar, Recl. Trav. Sta. Mar.
juveniles in marine softbottom Endoume Suppl 7, 57-100.
communities, Mar. Ecol. Prog. Ser. 7, pienkowski ~jH (1982) Diet and energy
179-184. in take of grey and ringed plover s,
Orth RE, Moore J and Fisher TW (1977) Pluvialis squatarola and Charadius
Year round survey of Staphylinidae of a hiaticula, in the non-breeding season,
sandy beach in Southern california, J. Zool. Lond. 197, 511-549.
Wasmann J. BioI. 35, 169-195. pike HB and Le Sueur RF (1958) The
Orton JH (1929) Severe environmental shore zonation of some Jersy Cumacea,
mortality among Abra (=Syndosmya) alba, Ann. Mag. Nat. Hist., Ser ~3, 1, 515-523.
Donax vittatus and other organisms of Pollock LW and Hummon HD (1971) Cyclic
the Lancashire coast. Nature Lond. 124, changes in interstitial water content,
91l. atmospheric exposure and temperature in
Pamatmat MM (1977) Benthic community a marine beach, Limnol. Oceanogr. 16,
metabolism: a review and assessment of 522-535.
present status and outlook, In B.C. Poxton MG, Eleftheriou A and HCIntyre
Coull, ed. Ecology of Marine Benthos. AD (1982) The population dynamics of
University of South Carolina Press, O-group flatfish on nursery grounds in
89-111. the Clyde Sea area, Bstuar, cstl, shelf.
panikkar BM and Rajan S (1970) Sci. 14, 265-282.
Observations on the ecology of some Prater AH and Davies M (1978)
sandy beaches of the southwest coast of Wintering sanderlings in Britain, Bird
India, Proc. Indian Acad. Sci. (B) 71, Studies 25, 33-38.
247-260. Pringle JS and Cooper J (1977) Hader
papi F and pardi L (1963) On the lunar populations (Cbaradrii) of tbe marine
orientation of sandhoppers. BioI. Bull. littoral of the Cape peninsula, South
124, 97-105. Africa, Ostrich 48, 98-105.
Pawlak R (1969) On the systematics and Prosch Hand NcLacblan A (1983)
ecology, life cycles and population Regeneration of surf zone nutrients by
dynamics of the turbellarian genus the sand mussel, Donax serra. In prep.
paromalostomum. Investigations at the pugh KB (1976)An annual cycle, at
sandy beach of the North Sea Island constant temperature, in a model sandy
Sylt, Helgolander wiss. Meeresunters 19, beach. I Nutrient chemistry, J. expo
417-454. mar. BioI. Ecol. 22, 179-192.
 377

Quinlan AV, Lewis T and Hoyt JK (1983) Hydrodynamic patterns in lotic

Fouling of the sandy beaches of Nahant intertidal sands and their bio-
Bay (Massachusetts, USA.) by an abnormal climatological implications, Mar. BioI.
free-living form of the macroalga 13, 179-209.
pilayella littoralis (Phaeophyta). I. Riedl RJ, Huang N and Machan R (1972)
Habitat characteristics. In McLachlan A The subtidal pump: a mechanism of
anQ Erasmus T, eds, Sandy Beaches as interstitial water exchange l.Jy wave
Ecosystems, The Hague, Junk. action, Har. BioI. 13, 210-221-
Randall R and McLachlan A (1982) Rieger R and Ott T (1971) Gezeiten-
Damara terms Sterna balaenarum in bedingte Wanderungen von ~rubellarien
breeding the Eastern Cape, South Africa, und Nematoden eines Nordadriatischen
Ostrich 53, 50-51. Sandstrandes, Vie Hilieu 22 (Suppl)
Rapson AM (1952) The Toheroa, 425-447.
amphidesma ventricosum Gray Riley JD and Corlett J (1966) The
(Eulamellibranchiata), development and number of O-group plaice in Port Erin
growth, Aust. J. Mar. Freshwat. Res. 3, Bay, 1964-1966. Dept. mar. bioI. Stn.
170-198. Po r t E r in, 78 , 51- 5 6 .
Rapson AM (1954) Feeding and control Rol.Jertson JR and Pfeiffer WJ (1981)
of Toheroa (Amphidesma ventricosum Gray) Deposit feeding by the ghost crab
(Eulamellibranchiata) populations in New Ocypode quadrata, J. expo mar. BioI.
Zealand, Aust. J. Mar. Freshwat. Res. 5, Ecol. 56, 165-178.
486-512. Robertson S and Lewin J (1976) Blooms
Rees CB (1939) Notes on the ecology of of surf-zone diatoms alonCj the coast of
the sandy beaches of North Donegal, the Olympic peninsula, Vlashington. VII.
Proc. roy. Irish Acad. 45, 215-229. Effect of temperature and oxygen concen-
Reilly FJ and Bellis VJ (1978) A study tration of respiration of Chaeteceras
of the ecological impact of beach armatum in the natural environment, Int.
nourishment with dredged materials on Revue, ges. Hydrobiol. 61, 201-210.
the intertidal zone. U.S. Army Corps. Rossouw G (1983) The importance of
Eng. Ins. Cstl. and Mar. Res. 'Tech. Rep non-teleost fishes in the surf zone with
4, 107pp. special reference to :H.hinobatos
Reilly FJ (1979) Energy flow between annulatus. In /IlcLachlan A and Erasmus
biological systems on exposed sandy T, eds, Sandy Beaches as Ecosystems, The
beaches, J. Elisha Mitchell Soc. 95, 125. Hague, Junk.
Reilly FJ and Bellis VJ (1978) Trophic Round FE (1979) A diatom assemblage
dynamics of sandy beach macrobenthos, living below the surface of intertidal
Assoc. Southeast. BioI. Bull. 25, 34-35. sand flats, Har. BioI. 54, 219-223.
Reise K (1979) Moderate predation on Sal oman CH and NauCjhton SP (1978)
meiofauna by the macrobenthos of the Benthic macroinvertebrates inhabiting
Wadden Sea, Helgolander wiss. the swash zone of Panama City Beach,
Meeresunters 32, 453-465. Florida, Northeast Gulf Science 2, 65-72.
Remane A (1933) Verteilung und Saloman CH, Naughton SP and 'Taylor JL
Organisation der benthonischen (1982) Benthic community response to
Mikrofauna der Kieler Bucht, Wiss. dredging borrow pits, Panama City Beach,
Meeresunters. 21, 161-221. Florida, U.s. Army Corps l:ngineers,
Renaud-Debyser J (1963) Recherches Hisc. Res. 82-3, 138 pp.
ecologiques sur la faune interstitielle Salvat B (1964) Les conditions
des sables-Bassin d' Arcachon lIe de hydrodynamiques interstitielles des
Bimini, Bahamas, Vie ~lilieu, Suppl. 15, sediment H1enbles intertidaux et la
1-157. repartition verticale de la jenne
Rheinheimer G (1977) Investigations on endogee, C.R. 259, 1576-1579.
the bacterial ecology in sand beaches of Sal vat B (1966) Eurydice pulchra Leach
the North Sea and the Baltic, Bot. Mar. 1815 et Eurydice affinis Hansen 1905
20, 385-400. (Isopodes: Cirolanidae). Taxonomie,
Rieper M (1978) Bacteria as food for ethologie, repartition verticale et
marine harpacticoid copepods. Mar. cycle reproduction, Acta Soc. Limn.
BioI. 45, 337-345. Bordeaux A 193, 1-77.
Riedl RJ (1971) How much seawater Sal vat B (1967) La macrofaune
passes through sandy beaches? Int. Rev. carinologique endogue des sediments
ges. Hydrobiol. 56, 923-946. meubles intertidaux (Tanaidaces,
Riedl RJ and MacMahan E (1969) High isopodes et AHiphipodes) : ethologie,
energy beaches In Odum and Copeland, bionomie et cycle biologique, Hem. Hus.
eds. Coastal Ecological Systems of the natn. Hist. nat. Ser. A 45, 1-275.
United States. Sanchez H, Castilla JC and Mena 0
Riedl RJ and Machan R (1972) (1982) Summer and >linter variations of
378

the macrofauna of a sandy beach in macrofauna am sandingen Brandungsstrand

Morrillas Norte-Chico, Chile (In von Ll Salvador, Senckenberg. bioI. 37,
Spanish) studies on Neotropical Fauna 1-56.
and Environment, 17, 31-50. Schwinghamer P (1981) Characteristic
Sameoto DD (1969a) Comparative size distributions of integral benthic
ecology, life histories, and behaviour communities, Can. J. Fish. Aquatic Sci.
of intertidal sand-burrowing amphipods 38, 1255-1263.
(Crustacea: Haustoriidae) at Cape Cod., Scott A (1960) The fauna of the sandy
J. Fish. Res. Bd. Canada 26, 361-388. beach, village Bay, St. Kilda: a dynamic
Sameoto DD (1969b) Some aspects of the relationship, Oikos 11, 153-160.
ecology and life cycle of three species Scott KJ and Croker RA (1976) Macro-
of subtidal sand-burrowing amphipods infauna of northern New I.;ngland marine
(Crustacea: Huastoriidae), J. Fish. Res. sand. II I. The ecology of psammonyx
Bd. Canada 26, 1321-1345. nobilis (Stimpson), 1853 (Crustacea:
SanderF and Moore E (1918) A Amphipoda), Can. J. Zool. 54, 1519-1529
comparative study of inshore and Seed R and Lowry BJ (1973) The
offshore copepod populations at intertidal macrofauna of seven sandy
Barbados, West Indies, Crustaceana 35, beaches of County Down, Proc. Roy. Irish
225-240. Acad, B. 73, 217-230.
Scapini F and Ugolini A (1981) Shelton CR and Robertson PB (1981)
Influence of landscape on the Community structure of intertidal
orientation of Talitus saltator macrofauna on two surf exposed Texas
(Crustacea, Amphipoda), Monit. Zool. sandy beaches, Bull. l'Iar. Sci. 31,
Ital. 15, 324-325. 833-842.
Schaeffer RH (2967) Species Shin PKS (1981) Temporal variations in
composition, size and seasonal abundance a shallow subtidal community in North
of fish in the surf waters of Long Bay, Inner Galway Bay, \Iest Coast of
Island, N.Y. Fish Game J. 1, 1-46. Ireland, Ir. Nat. J. 20, 321-324.
Schaeffer RH (1970) Feeding habits of Shin PKS (1982) Multiple discriminant
striped bass (Morone saxatilis) from the anaylsis of macrobenthic infaunal
surf waters of Long Island, N.Y. Fish assemblages, J. expo mar. BioI. Ecol.
Game J. 17, 1-17. 59, 39-50.
Scheibling RE (1982) Habitat Smith DAS (1971) polymorphism and
utilization and bioturbation by Oreaster population density in Donax rugosus
reticulatus (Asteroidea) and Meona (Lamellibranchiata: Donacidae), J. <:.ool.
ventricosa (Echinoicea) in a subtidal Lond. 164, 429-441.
sand patch. Bull. mar. Sci. 32, 624-629. Sourie R (1957) Etude ecologique des
Schmidt P (1970) Zonation of the plages de la eote Senegalaise aux
interstitial polychaete Stygocapiletta environs de Dakar, Ann. Ec. sup. Sci.
subterranea (Stygocapitellidae) in Dakar 3, 1-110.
European sandy beaches, Mar. BioI. Southward AJ (1953) 'l'he fauna of some
(Berl.) 7, 319-323. sandy and muddy shores in the south of
Schmidt P (1972a) Sonierung und the Isle of Man, Proc. Trans. Liverpool.
jahreszeitliche fluktuationen de bioI. Soc. 59, 51-71.
interstitiellen fauna in sandstranden Steele JH (1976) Comparative studies
des gebiets von Tromso (Norwegen), beaches, Phil. Trans. Roy. Soc.
Mikrofauna Meeresbodens 12, 81-164. Edinburgh B 274, 401-415.
Schmidt P (1972b) Zonation and annual Steele JH and Baird JE (1968)
fluctuation of the mesopsammon in the Production ecology of a sandy beach,
sandy beach of Schilksee Bay of Kiel, Limnol. Oceanogr. 13, 15-25.
West Germany, Mikrofauna Meeresbodens Steel JH and Ldwards RRC (1970) 'l'he
10, 3-60. ecology of O-group plaice and common
Schneider AC (1981) Size selective dabs in Loch Lwe. IV. Dynamics of the
predation on mysids by birds (Plovers), plaice and dab populations, J. expo mar.
Mar. Ecol. Prog. Ser. 5, 223-224. BioI. Ecol. 4, 174-187.
Schneider D (1982) Predation by ruddy Steele JH, Munro ALS and Giese GS
turnstones Arenaria interpres on a (1970a) Environmental factors
polymorphic clam Donax variabilis at controlling the epipsommial flora on
Sanibel Island, Florida, Bull. Mar. Sci. beach and sub-littoral sands, J. mar.
32, 341-344. bioI. Ass. U. K. 50, 907-918.
Schneider DC and Harrington BA (1981) Steele JH, MCIntyre AD, Edwards RRC
Timing of shorebird migration in and 'l'revallion A (1970b) Interrelations
relation to prey depletion. Auk, 98, of a young plaice population with its
80l-8ll. invertebrate food supply. In Watson,
Schuster-Diederichs 0 (1956) Die A., ed. Animal populations in Relation
 379

to their food Resources. British Thurn AB and Allen JC (1975)

Ecological Society Symposium 10, Distribution and abundance of the lamp
375-388, Oxford, Blackwell. urchin Bchinolampas crassa (Bell) 1980
Stenton-Dozey J and Griffiths CL in False Bay, Cape ~rans. Roy. Soc.
(1980) Growth, consumption and S . Af r. 41, 359 - 373 .
respiration by larvae of the kelp fly Tietjen JH (1980) Microbial-meiofauna
Fucellia capensis (Diptera: interrelationships: a review, Micro-
Anthomyiidae), S. Afr. J. Zool. 15, biology, 1980, 335-338.
280-283. Trevallion A (1971) Studies on 'l'ellina
Stenton-Dozey J and Griffiths CL tenuis da Costa. III. Aspects of general
(1983) The fauna accosiated with kelp biology and energy flow, J. expo mar.
st randed on a sandy beach. In McLachlan BioI. Ecol. 7, 95-122.
A and Erasmus T, eds. Sandy Beaches as Trevallion A, Ansell AD, Sivadas P and
Ecosystems, The Hague, Junk. Narayanan B (1970) A preliminary account
Stephen AC (1928) Notes on the biology of two sandy beaches in South Hest
Tellina tenuis da costa J. mar. bioI. India, Mar. BioI. 6, 268-279.
Ass. U.K. 15, 683-702. Trueman L;R (1971) 'l'he control of
Stephen AC (1931) Notes on the biology burrowing and the migratory behaviour of
of certain lamellibranchs on the Donax denticulatus (Bivalvia:
Scottish coast, J. mar. bioI. Ass. U.K. Tellinacea), J. Zool. Lond. 165, 453-469.
17, 277-300. Turbeville DB and Marsh GA (1982)
Stephen AC (1932) Notes on the biology Benthic fauna of an offshore burrow area
of some lamellibranchs in the Clyde Sea in Broward Country, Florida, U.S. Army
area, J. mar. bioI. Ass. U.K. 18, 51-68. Corps Bngineers Misc. Reo. 82-1, 42pp.
Subramoniam T (1979) Some aspects of Turner JH and Belding DL (1957) ~he
reproductive ecology of a mole crab tidal migrations of Donax variabilis
Emerita asiatica Milne Edwards, J. expo (Say), Lirnnol. Oceanogr.~20-125.
mar. BioI. Ecol. 36, 259-268. Upadhyay RS, Singh DB and Bharat Rai
Summers RW and Cooper J (1977) The (1978) Bcology of rnicrofungi in a
population ecology and conservation of tropical coastal sand belt, Indian J.
the Black Oystercatcher Haematopus Mar. Sci. 7, 187-190.
moquini, Ostrich 48, 28-40. Vader ,/ (1982) pied wagtails catching
Summers RH, Cooper JZ and Pringle JS young ghost crabs, Ostrich 53, 205.
(1977) Distribution and numbers of vannini M (1976) Research on the coast
coastal waders (Charadrii) in the south of Somalia. The shore and the uune uf
western Cape South Africa, summer Sar Uanle. Part 10. Sandy beach
1975-76, Ostrich 48, 85-97. decapods, Moni t. "'001. 1 tal. Suppl. 8,
Summers RW and Waltner M (1979) 255-286.
Seasonal variations in the mass of veitch CR (1980) Sea birds tound dead
waders in southern Africa, with special in New Zealand in 1970, 1971 and 1972,
reference to migration. Ostrich 50, Notornis 27, 369-385.
21-37. vernberg WB and Coull BC (1974)
Summers RW and Hockey PAR (1980) Respiration of an interstitial ciliate
Breeding biology of the white-fronted and benthic relationships, Oecologia
plover Charadrius marginatus in the (Berl.) 16, 259-264.
south western Cape, South Africa, J. Virnstein RW (1977) The importance of
Nat. Hist. 14, 433-446. predation by crabs and fishes on benthic
Swedmark B (1964) The interstitial infauna in Chesapeake Bay, Bcology 58,
fauna of marine sand, BioI. Rev. 39, 1199-1217.
1-42. Vohra FC (1971) ",onation on a tropical
Thayer LA (1935) Diatom water-blooms sandy shore, J. Anim. Lcol. 40, 679-708.
on the coast of Hashington. Proc. Hade BA (1967) Studies on the biology of
Louisiana Acad. Sci. 2, 67-71. the West Indian beach clam Donax
Thijssen R, Lever AJ and Lever J denticulatus Linne. 1. L;cology, Bull.
(1974) Food composition and feeding mar. Sci. 17, 149-174.
periodicity of O-group plaice Hade BA (1968) Studies on the biology
Pleuronectes platessa in the tidal area of the West Indian beach clam, Donax
of a sandy beach, Neth. J. Sea Res. 8, denticulatus (Linne) 2. Life-history,
369-377. Bull. Mar. Sci. 18, 877-899.
Thomassin B (1969) Indentification, Hade K (1982) Handering in Scopimera
variabilite et ecologie des Hippidea globosa (Crustacea, Ocypodidae) Publ.
(Crustacea, Anomura) de la region de Seto. Mar. BioI. Lab. 26, 447-454.
Tulear, S.H. de Madagascar, Rec. Trav. Watkin EE (1942) The macrofauna of
Stn. Mar. Endoume-Marseille Fasc. Aovs. intertidal sand at Kames Bay, Millport,
Ser. Suppl. 9, 135-177. Buteshire, Trans. Roy, Soc. Bdinburgh
380

60, 543-561. polychaete Abarenicola pacifica. J. mar.

Webb JE (2958) The ecology of Lagos Res. 39, 735-748.
Lagoon V: Some physical properties of vHthers RG (1977) Soft-shore
Lagoon deposits, Phil. Trans. Roy. Soc. macrobentos along the South liVest Coast
Lond. Ser B 241, 393-419. of Wales, Estuar. cstl. mar. Sci. 5,
li~enner AM (1977) Food supply, feeding 467-484.
habits, and egg production in Pacific liiolcott TG (1978) Ecological r61e of
mole crabs (Hippa pacifica Dana) Pac. ghost crabs Ocypode quadrata (Fabr.) on
Sci. 31, 39-47. an ocean beach: scavengers or predators.
Wenner AM and Fusaro C (1979) An J. expo mar. BioI. Ecol. 31, 67-82.
analysis of population structure in Hood DH (1968) An ecological study of
pacific mole crabs (~pacifica Dana), a sandy beach near Auckland, New
BioI. Bull. 157, 205-220. Zealand, 'l'rans. Roy. Soc. N.Z. Zoo 1. 10,
Wicks SR (1974) Presence of 89-115.
Acotobacter in marine sand beaches, Fla. Hoodin SA (1981) Disturbance and
Sci. 38, 167-169. community structure in a shallow water
Wieser W (1959) The effect of grain sand flat, Ecology 62, 1052-1066.
size on the distribution of small Wooldridge T (1981) Zonation and
invertebrates inhabiting the beaches of distribution of the beach mysid,
puget Sound, Limnol. Oceanogr. 4, Gastrosaccus psammodytes, J. Zool. Lond.
181-194. 193, 183-189.
Wieser W (1975) The meiofauna as a Wooldridge T (1983) The ecology of
tool in the study of habitat beach and surf zone mysids in the
heterogeneity: ecophysiological Eastern cape, South Africa. In
aspects. A review, Cah. BioI. Mar. 16, McLachlan A and Lrasmus T, eds. Sandy
647-670. beaches as Ecosystems, The Hague, Junk.
Wieser W (1960) Benthic studies in Wooldridge 'j.', Dye AH and McLachlan A
Buzzard's Bay II: The meiofauna, Limnol. (1981) 'l'he ecology of sandy beaches in
Oceanogr. 5, 121-137. Transkei, S. Afr. J. Zool. 16, 210-218.
Wieser Wand Scheimer F (1977) The Hormald AP and Stirling HP (1979) A
ecophysiology of some marine nematodes preliminary investigation of nutrient
from Bermuda: Seasonal espects, J. expo enrichment in experimental sand columns
mar. BioI. Ecol. 26, 97-106. and its effect on tropical intertidal
liHeser W, Ott J, Schiemer F and bacteria and meiofauna, Lst. cstl. mar.
Gnaiger E (1974) An ecophysiological Sci. 8, 441-453.
study of some meiofauna species Hright LD and Short AD (1982)
inhabiting a sandy beach at Bermuda, High-energy nearshore and surf zone
Mar. BioI. 26, 235-248. morphodynamics. JOA (Halifax) Abstracts
Wigley RL and Burns BR (1971) p 86.
Distribution and biology of mysids Youngbluth MJ (1979) ':l'he variety and
(Crustacea, Mysidacea) from the Atlantic abundance of zooplankton in the coastal
coast of the United States in the NMFS waters of Puerto Rico, Northeast Gulf
Woods Hole collection, Fish. Bull. 69, Sci. 3, 15-26.
717-746.
Wilce RT and Quinlan AV (1983) Fouling
of the sandy beaches of Nahant Bay
(Massachusetts, USA) by an abnormal
free-living form of the macroalga
pilayella littoralis (Phaeophyta) II.
population characteristics. In
McLachlan A and Erasmus T, eds., Sandy
Beaches as Ecosystems, The Hague, Junk.
Wilce RT, Schneider ZW, Quinlan AV and
Van den Bosch K (1982) The life history
and morphology of free-living pilayella
littoral is (L.) Kjellm. (Ectocarpaceae,
Ectocarpales) in Nahant Bay,
Massachusetts, Phycologia 21, 336-354.
williamson DI (1951) On the mating and
breeding of some semiterrestrial
amphipods. Dept. Dove Marine Lab. 1949,
49-62.
Wilson WH (1981 ) Sediment-mediated
interactions in a densely populated
infaunal assemblage: The effects of the
 381

THE ROLE OF PHYTOPLANKTON IN SURF ECOSYSTEMS

JOYCE LEWIN AND C.T. SCHAEFER (School of Oceanography, University of Washington, Seattle, WA, U.S.A.
98195)

1. INTRODUCTION
A widely-held view concerning primary
production in the littoral and inner sublittoral
zones of the ocean is that high productivity can
occur only in areas having a stable substratum
to which benthic plants can attach. Consequent-
ly, exposed sandy beaches, where the shifting
substratum precludes attachment of macroalgal
holdfasts, are commonly regarded as zones of low
primary production. If little plant material is
produced within the sandy beach environment,
then the fauna of such beaches must rely heavily FIGURE 1. View to the north along Copalis
Beach, Washington, showing masses of diatoms
on food produced in other environments and
deposited on the sand during ebb tide.
subsequently transported to the beach. The food
is believed to come chiefly from the ocean
In 1970, when we began our studies of the
beyond the surf zone and, in lesser quantities,
persistent diatom blooms along the Pacific
from the land (Brown, 1964). In short, the
northwest coast, U.S.A., we were studying a
general rule is that the sandy beach biota is
coast that had been the subject of earlier
"subsidized" to a high degree (McLachlan,
investigations 50 years ago (Becking et al.,
1980). This paper concerns a widespread
1927; Thayer, 1935a,b), giving useful data for
phenomenon that constitutes an exception to that
comparative purposes. Also, papers had been
rule.
published reporting on the surf diatom blooms
Our studies have focused on exposed coasts occurring along the New Zealand coast (Rapson,
that have wide, gently-sloping sandy beaches 1954; Cassie, Cassie, 1960). Other than these,
(Fig. 1) and well-developed surf. In certain there were no additional published accounts of
regions of the world, the surf zone along such which we were aware. Recently (within the past
beaches has a persistent dense accumulation of five years), we have been contacted by several
phytoplankton, imparting a rich brown colour- beach researchers who have reported to us their
ation to the water. In all instances known to observations of brown water along other coasts
us, the algae forming these blooms are (South Africa, Brazil, Costa Rica), so the list
diatoms. Usually one or two diatom species of reported sightings has now grown considerably
predominate over all others. (Fig. 2 and Table 1). The list will probably
382

75°'~~~'-~~--~--~~~~------------~~V-----~~~--"---------Y~~'-----~~-----'7r~--~~

L-~~-f~+-~~--~-------------------------------+------~~-------------+------~~----------~O°
,. C~~.~

c:J
6o°L-----~L_ ______________L_______________ ~ _______________ L_ _ _ _ _ _ _ _ _ _ _ _ _ __ L_ _ _ _ _ _ _ _ _ _ _ _ _ __ L_ _ _ _ _ __"60 0

FIGURE 2. Locations of known persistent blooms of surf diatoms.

grow even longer as more remote beaches are Because of the absence of attached plants
visited and studied. For example, we have found on exposed sandy beaches, the basic food supply
very brief descriptions of surf diatom blooms for beach fauna is furnished by the microscopic
that were once observed at beaches in the Congo plant forms (chiefly diatoms, but including
(now Zaire) (Van Heurck, 1896) and Nicaragua other unicellular algae) and detritus. Along
(Thayer, 1935b), but we know of no published beaches in the areas mentioned (Table I), the
research from either of these localities. Dense surf diatom populations constitute a highly con-
blooms are known to occur sporadically along centrated, reliable food source, which in turn
some California beaches (U.S.A.), but these have supports extensive standing stocks of a wide
not been studied intensively, and the degree of variety of beach and surf fauna. For the most
regularity of their occurrence is unknown to part, filter-feeding bivalve molluscs appear to
us. A recent collection of one of the principal be the predominant herbivores grazing directly
surf diatom species, Chaetoceros armatum T. on the surf diatoms along those beaches we are
West, on the Australian coast suggests the pos- considering. The principal bivalve species
sibility of surf blooms in Australia similar to include the razor clam Siliqua patula Dixon
those already reported from New Zealand and (U.S.A.), the toheroa Amphidesma ventricosum
Tasmania. Also summarized in Table 1 are the Gray (New Zealand), and the clam Donax serra
major species comprising the well-documented Roding (South Africa); each of these species
blooms. It should be noted that all of the strongly dominates the macrofauna in its partic-
important species found thus far can be assigned ular locality (Lewin et al., 1979; Rapson, 1954;
to four genera: Chaetoceros, Asterionella, McLachlan et al., 1981). At least two species
Aulacodiscus, and Anaulus. of bivalves, Mesodesma mactroides Deshayes and

TABLE 1. Summary of known persistent blooms of surf diatoms.
Locale
1) Northwest coast, U.S.A.:
Washington and Oregon
43° - 48°N
2) North Island of New Zealand:
Westcoast beaches
34° - 41°S
5 Eastcoast beaches
34° - 36°S
3) South Africa, south coast:
Port Elizabeth to Cape Town
33° - 34°S
4) Tasmania, west coast:
42°S
5) Brazil, southeast coast:
29° - 34°S
6) Pacific coast, Central America:
Costa Rica and Panama
8°N
Major species
comprising the blooms
50 years ago
Au1acodiscus kittonii Arnott
Asterione11a socia1is Lewin & Norris
today
Chaetoceros armatum T. West
Asterione11a social is Lewin & Norris
Chaetoceros armatum T. West
Asterione11a glacia1is Castracane
(-.!. japonica Cleve ex Gran)
Chaetoceros armat.um T. West
Au1acodiscus kittonii Arnott
Chaetoceros armatum T. West
Au1acodiscus kittonii Arnott
Anau1us sp.
(birostratus Grunow or
mediterraneus Grunow)
Au1acodiscus petersii Ehrenberg
Chaetoceros armatum T. West
Anau1us sp.
(birostratus Grunow or
mediterraneus Grunow)
Asterione11a glacia1is Castracane
(- A. japonica Cleve ex Gran)
Au1acodiscus africanus Cottam
383
Reference
Thayer, 1935b
Lewin, Norris, 1970
Rapson, 1954;
Cassie, Cassie, 1960
S.R. Stido1ph, personal
communication (1982)
M.J. Kindley, personal
communication (1982)
McLachlan, Lewin, 1981
Romer, 1981
J. Lewin from samples
collected by D.P.
Thomas and provided by
R. Horner, 1981
Gianuca, in press
R. Holmes and J. Lewin
from samples collected
by J.A. Sherfy, 1982
384

Donax hanleyanus Philippi, are major consumers (outside the surf), from river dis-
of surf diatoms on beaches in Brazil (Gianuca, charge, from rainfall, and through
in press). Other kinds of herbivores are recycling from excretion products
important on some coasts. For example, on the released by beach interstitial fauna
Pacific coast of Central America, deposit- and by beach and surf macrofauna. The
feeding ghost crabs appear to be the predominant latter source appears to be of prime
grazers on surf diatoms, although the possible importance (Lewin, 1978; Lewin et al.,
importance of bivalve populations known to exist 1979; McLachlan, 1980, 1982; McLachlan,
there has yet to be evaluated (J.A. Sherfy, Lewin, 1981), although the proposed
personal communication). Mullet and prawns feed role of interstitial fauna in nutrient
directly on the diatom scum at the water surface regeneration has recently been disputed
along the African coast (described in detail by by Hennig et aL (in press), who hy-
Romer, 1981). pothesize that the interstitial system
is actually a nutrient sink.
2. FACTORS CONTROLLING SURF DIATOM BLOOMS iv) Rainfall- Most of the locations of
For several years our studies have been known surf blooms are in regions of
aimed at pinpointing the factors that must be high rainfall (up to 2,500 mm per year,
considered to understand how diatom blooms are based on 10-year averages). On the
developed and maintained within the surf habi- South African coast, where rainfall is
tat. Our own results, plus data from earlier relatively low, evidence suggests that
works, and now new observations from those who the richest blooms might follow heavy
are currently working along sandy beaches have rains (McLachlan, Lewin, 1981), a
led us to the following list of most significant pattern which has also been observed
factors: in some of the high-rainfall areas
i) Topography of the coas t- A broad shal- (Becking et al., 1927; Thayer, 1935b;
low surf zone is required (Garver, Gianuca, in press). The apparent im-
Lewin, 1981), where wave-induced circu- portance of rainfall probably derives
lation cells can develop which, in from its role in the drainage of
turn, act to retain both nutrients and nutrient-rich interstitial water out of
diatom populations (McLachlan, 1980; the sand and back into the surf
McLachlan, Lewin, 1981). (McLachlan, Lewin, 1981).
H) Winds- The densest blooms accompany v) Physiological adaptations of the surf
onshore winds (Becking et al., 1927; diatom species- All known surf species
Rapson, 1954; Romer, 1981; Gianuca, in float at the water surface, either
press; M.J. Kindley, personal commu- attached to rather large surf bubbles
nication). Strong offshore winds or forming a stable foam consisting of
apparently blow diatoms out to sea many small bubbles (Becking et al.,
(Rapson, 1954; M.J. Kindley, personal 1927; Thayer, 1935b; Lewin, Mackas,
communication), removing them from the 1972; Lewin, Hruby, 1973; Lewin et al.,
cellular circulation. 1975; McLachlan, Lewin, 1981; N.M.
Hi) Nutrient supply- A source of nitrogen Gianuca, personal communication; J.A.
must be readily available. Nitrogen Sherfy, personal communication).
can come from ocean waters offshore
 385

Topography is, of course, a rather constant again in the later afternoon (before sunset)
feature of a given coast; however, winds and (Lewin, Hruby, 1973; Lewin, Rao, 1975;
rainfall are highly variable and often sea- McLachlan, Lewin, 1981; N .M. Gianuca, personal
sonal. Where winds and rainfall are strongly communication), although at least one species,
seasonal, we would expect, in turn, to see Asterionella social is Lewin & Norris, appears to
corresponding seasonal changes in the abundance show flotation capability both day and night
of the diatom populations. Along the Pacific (McLachlan, Lewin, 1981).
northwest coast, U.S.A., a definite seasonality
This accumulation of cell masses at the
in the diatom populations is observed (Lewin
water surface during the daytime means that cell
et al., 1975; Lewin, 1977; Garver, Lewin,
numbers in daytime surf samples are higher than
1981). Both surf species are in highest abun-
those in night samples (Lewin, Hruby, 1973;
dance during those seven months of the year
Lewin et al., 1975; Lewin, Rao, 1975; Lewin,
(October through April) when there are strong
1977). The ability to float at the surface is
prevailing winds blowing from the southwest and
an adaptation that would benefit the species in
when rainfall is at its maximum. Similarly,
two obvious ways: i) Flotation maintains the
along the New Zealand west coast, the surf
cells within the surf habitat and decreases the
blooms occur at their maximum during the strong
likelihood that they will be washed out to sea.
westerly winds of the winter months (Rapson,
Objects that float are carried by wave action
1954), with August apparently being the month
toward the beach. On a receding tide in the
when populations reach their peak. Easterly
daytime when the cells are floating, they are
winds commencing in the spring bring about a
deposited in large masses on the beach (Fig. 1),
decline in the surf blooms (Rapson, 1954).
often forming a deposit several centimetres
Along the south coast of South Africa, on the
thick and extending for several kilometres
other hand, where onshore winds prevail through-
(Lewin, Hruby, 1973). Diatom deposits 50 cm
out the year (Romer, 1981), there is little
thick were recorded at Copalis Beach, Washington
evidence of seasonality of the diatom blooms
in earlier studies (Thayer, 1935a). Experiments
(McLachlan, Lewin, 1981).
have shown that the photosynthetic viability of
surf diatoms stranded on the beach for several
3. FLOTATION - A UNIQUE ATTRIBUTE OF SURF
hours may be essentially equal to that of cells
DIATOMS
collected directly from the surf (J. Lewin, un-
We have been interested in possible
published). Provided the cells remain healthy,
physiological adaptations that surf diatoms
deposition on the beach is actually advanta-
might possess that would allow them to thrive in
geous; removing the diatoms from circulation
such a rigorous environment. Do they have
during part of the day increases their chances
special characteristics or adaptations which
of remaining within the confines of the surf and
make them different from other diatom species?
beach ecosystem. ii) Flotation places the cells
The ability to float at the surface of the
in a position where maximum utilization of
water either as a stabilized foam or attached to
available light for photosynthesis can occur.
larger surf bubbles constitutes a feature unique
Cells floating at the surface of the water and
to the surf diatoms; other diatom species lack
those deposited on the beach are exposed to full
this capability. Some of the surf species
intensity sunlight. Maximum absorption of
appear to rise to the water surface in the early
available light would give the surf species an
morning hours (before sunrise) and to sink out
advantage, particularly during winter months
386

_1 _1
when light is diminished in intensity and (5-8 gCogChla oh as in winter (3-4
_1 ::l
duration. The usual phytoplankton community of gCogChl~ oh) (Fig. 3). Thus, Pmax exhibits
coastal temperate waters virtually disappears a seasonal pattern that is the reverse of the
during winter months, due to the fact that they pattern for standing stock.
are mixed into deeper waters where light is

Ir
limiting for photosynthesis.
8,-"--,-,,,'-T'--'-"'--T'--~"'~I~'-'-'-'
.--

II
I
An important distinction should be made .r:.
6 -
between flotation and buoyancy. All available
evidence indicates that surf diatoms are not .r:. I
innately buoyant even when they are floating. '6, 4- -
I I
If removed from the surf and kept in quiescent U
0>
I -
water, the diatoms sink to the bottom of a
x 2- -
container within a few minutes; this observation o
has been made in the U.S.A. (Lewin, Rao, 1975) l -
and in South Africa (D.S. Sloff, personal O~O~N~~D~~iJ-LF~~M~~A~~M-L~J~~J~A~~S~
communication), and it is presumed to apply to 1981 1982
surf diatoms in general. Apparently diatom
FIGURE 3. Midday (between 1000h and 1400h)
flotation is completely dependent on bubbles values for light-saturated chlorophyll-specific
photosynthesis(Pma~) measured throughout the
produced by breaking waves.
year from October 1981 through September 1982.
Bars represent mean + 2 s.e.
4. PHOTOSYNTHETIC PRODUCTION OF SURF DIATOM
SPECIES For comparing actual growth of phytoplank-
We have recently undertaken a study
ton populations at different seasons, the
(Schaefer, 1983) of the photosynthetic capa-
specific growth rate (~) is more meaningful than
bilities of the surf diatom populations at
Pmax • Constant proportionality between ~ and
Copalis Beach, Washington, using the carbon-14
Pmax occurs only if the carbon-to-chlorophyll
method. One of the major objectives of this
ratio (C/Chl) does not change over time.
study was to determine whether the decrease in
Although ~ is very difficult to determine for
diatom standing stock that occurs each summer
natural phytoplankton populations, an approxi-
reflects a decrease in photosynthetic rates.
mation, the specific carbon incorporation
Another objective was to determine whether surf
rate (~', as defined by Li, Goldman, 1981), can
diatom species possess special photosynthetic
be calculated based on carbon-14 photosynthetic
adaptations to account for their proliferation
measurements and particulate organic carbon
during winter, when light limitation might be
analyses. Exact equality of ~' to ~ is unlikely
expected. The data obtained enable us to draw
(for reasons discussed by Li, Goldman, 1981) ,
some conclusions regarding both of the stated
but for the present discussion the assumption is
objectives.
made that the proportionality of ~' to ~ is
Our results show no evidence of decreased essentially constant over the year for the surf
photosynthetic rates by surf diatoms in summer diatoms. Unlike Pmax ' ~~ax for the surf diatoms
in comparison to winter rates. In fact, when shows no obvious seasonality, and most values
normalized to chlorophyll , light-saturated for ~~ax fall within the comparatively narrow
_1 _1
photosynthetic rates (Pmax ) were found to range of 0.09-0.13 gCogC oh (Fig. 4).
be approximately twice as high in summer Since ~~ax is a light-saturated daytime rate,
 387

however, it is likely that V over the light~dark In light of these results, we concluded
cycle is higher in summer than in winter because that the decline in abundance of surf diatoms in
of longer daylight periods and higher light summer could not be explained by any decrease in
intensities in summer. Consequently, the photosynthetic capabilities. The explanation
seasonal pattern (if any) for V appears to be must lie elsewhere; two possibilities are in-
the reverse of the pattern for standing stock. creased grazing pressure and greater seaward
losses of diatoms from the surf zone due to
prevailing northerly winds in summer.

I
.12
I I teristics
Our measurements of photosynthetic charac-
and chemical properties of the surf
species indicate that they do not differ
.08
I
U significantly from other coastal diatom
0>
species. For instance, the relationship between
><
o photosynthetic rate and light intensity (Fig. 6)
E .04
-:::l does not differ in any major respect from that
determined for other diatoms. A saturating
o 0 N D A M J J A S light intensity of 200 vEinom
_2
0 s
_1
does not
1981 1982
differ from that measured for two other diatom
FIGURE 4. Midday values for light-saturated species (see Chan, 1980). C/Chl, and
specific carbon incorporation rate (v~ax). See
carbon content related to cell volume are
text for explanation of v~ax
comparable with values measured for other
diatoms.
The discrepancy between the seasonal trends
for Pmax and v~ax results from a seasonal
change in C/Chl. Much higher values of C/Chl 81-
_1
were observed in summer {50-75 gCogChl~ ) than I-
~

_1 I
in winter (20-30 gCogChl~ ) (Fig. 5). The .!:: 61-

fact;..- that v~ax remains relatively constant de-

spite large changes in C/Chl indicates that the
chlorophyll functions more efficiently in summer.

80r-,,--,-,'1--'-"1--'-"1--'-"-'--'-' Y = 0.0318 X-0.057

a.. r2 = 0.9955
1 1 1 1
- 300 400 500 600 700 800 900
..c I (fLEin· m- 2 • 5- 1 )
~
U
40J I I- FIGURE 6. Chlorophyll-specific photosynthesis
0>
(P) versus light intensity (I) measured in
- I I August. Bars represent mean + 2 s.e. Results
I
..c I
of least-squares linear regression for the
U 20- - light-limited portion are shown •
......
U - -
O~~~~I~~I~I~~I~~~~~~~ Therefore, it appears that the ability to
ON DIJ F M AM J J AS
float and thereby to take maximum advantage of
1981 1982
available light, rather than any unusual photo-
FIGURE 5. Midday values for carbon-to-
synthetic properties, explains the success of
chlorophyll ~ ratio (C/Chl).
388

surf species during winter in temperate Lewin J (1977) Persistent blooms of surf
diatoms along the northwest coast. In Krauss R,
latitudes. ed. The marine plant biomass of the Pacific
northwest coast, pp. 81-92. Corvallis, OR,
Oregon State University Press.
5. CONCLUSIONS
Lewin J (1978) Blooms of surf-zone diatoms
The idea that exposed sandy beaches are along the coast of the Olympic Peninsula,
Washington. IX. Factors controlling the
unproductive, highly subsidized environments
seasonal cycle of nitrate in the surf at Copalis
clearly is not universally applicable. Some Beach (1971 through 1975), Estuarine Coastal
Mar. Sci. 7, 173-183.
beaches, together with their surf zones, can
Lewin J, Eckman JE and Ware GN (1979) Blooms
be regarded as semi-closed ecosystems (as pro- of surf-zone diatoms along the coast of the
Olympic Peninsula, Washington. XI. Regeneration
posed by McLachlan, 1980) with high primary
of ammonium in the surf environment by the
production provided by specially-adapted surf Pacific razor clam Siliqua patula, Mar. BioI.
(Berl.) 52, 1-9.
phytoplankton.
Lewin J and Hruby T (1973) Blooms of surf-zone
diatoms along the coast of the Olympic
Peninsula, Washington. II. A diel periodicity
ACKNOWLEDGMENTS
in buoyancy shown by the surf-zone diatom
These studies were supported by funds species Chaetoceros armatum T. West, Estuarine
Coastal Mar. Sci. 1, 101-105.
provided by Department of Energy Contract DE-
Lewin J, Hruby T and Mackas D (1975) Blooms of
AT06-EV-75026. This is Contribution No. 1314 surf-zone diatoms along the coast of the Olympic
Peninsula, Washington. V. Environmental
from the School of Oceanography, University of
conditions associated with the blooms (1971 and
Washington. 1972), Estuarine Coastal Mar. Sci. 3, 229-241.
Lewin J and Mackas D (1972) Blooms of surf-
zone diatoms along the coast of the Olympic
REFERENCES Peninsula, Washington. I. Physiological
investigations of Chaetoceros armatum and
Becking LB, Tolman CF, McMillin HC, Field J
Asterionella socialis in laboratory cultures,
and Hashimoto T (1927) Preliminary statement
Mar. BioI. (Berl.) 16, 171-181.
regarding the diatom "epidemics" at Copalis Lewin J and Norris RE (1970) Surf-zone diatoms
Beach, Washington, and an analysis of diatom of the coasts of Washington and New Zealand
oil, Econ. Geol. 22, 356-368. (Chaetoceros armatum T. West and Asterionella
Brown AC (1964) Food relationships on the spp.), Phycologia 9, 143-149.
intertidal sandy beaches of the Cape Peninsula,
Lewin J and Rao VNR (1975) Blooms of surf-zone
S. Afr. J. Sci. 60, 35-41. diatoms along the coast of the Olympic
Cassie RM and Cassie V (1960) Primary
Peninsula, Washington. VI. Daily periodicity
production in a New Zealand west coast phenomena associated with Chaetoceros armatum in
phytoplankton bloom, N. Z. J. Sci. 3, 173-199. its natural habitat, J. Phycol. 11, 330-338.
Chan AT (1980) Comparative physiological study
Li WKW and Goldman JC (1981) Problems in
of marine diatoms and dinoflagellates in rela-
estimating growth rates of marine phytoplankton
tion to irradiance and cell size. II. Relation- from short-term 14C assays, Microb. Ecol. 7,
ships between photosynthesis, growth, and car-
113-121.
bon/chlorophyll a ratio, J. Phycol. 16, 428-432.
McLachlan A (1980) Exposed sandy beaches as
Garver JL and iLewin J (1981) Persistent blooms
semi-closed ecosystems, Mar. Environ. Res. 4,
of surf diatoms along the Pacific coast, U.S.A.
59-63.
I. Physical characteristics of the coastal
McLachlan A (1982) A model for the estimation
region in relation to the distribution and
of water filtration and nutrient regeneration by
abundance of the species, Estuarine Coastal
exposed sandy beaches, Mar. Environ. Res. 6,
Shelf Sci. 12, 217-229. 37-47.
Gianuca NM (in press) A preliminary account of
McLachlan A, Erasmus T, Dye AH, Wooldridge T,
the ecology of sandy beaches in southern Brazil.
Van der Horst G, Rossouw G, Lasiak TA and
In McLachlan A and Erasmus T, eds. Sandy beaches
McGwynne L (1981) Sand beach energetics: an
as ecosystems. The Hague, W Junk.
ecosystem approach towards a high energy inter-
Hennig HF-KO, Fricke AH and Martin CT (in
face, Estuarine Coastal Shelf Sci. 13, 11-25.
press) The effect of meiofauna and bacteria on
McLachlan A and Lewin J (1981) Observations on
nutrient cycles in sandy beaches. In McLachlan surf phytoplankton blooms along the coasts of
A and Erasmus T, eds. Sandy beaches as eco-
South Africa, Bot. Mar. 24, 553-557.
systems. The Hague, W Junk.
 389

Rapson AM (1954) Feeding and control of toheroa

(Amphidesma ventricosum Gray)(Eu1ame11ibranchiata)
populations in New Zealand, Aust. J. Mar.
Freshwater Res. 5, 486-512.
Romer GS (1981) An evaluation of the grazing
food chain associated with phytoplankton blooms
on an eastern Cape Beach. B. Sc. Hons Project
in Zoology, University of Port Elizabeth.
Schaefer CT (1983) Productivity of surf
diatoms at Copalis Beach, Washington, and its
relation to standing stock. M.S. Thesis,
University of Washington.
Thayer LA (1935a) Some experiments on the
biogenetic origin of petroleum. Ph.D. Thesis,
Stanford University, 137 pp.
Thayer LA (1935b) Diatom water-blooms on the
coast of Washington, Proc. La. Acad. Sci. 2,
68-72.
Van Heurck H (1896) A treatise on the
Diatomaceae (translated by WE Baxter). London,
William Wesley & Son.

FACTORS AFFECTING THE DISTRIBUTION OF ORGANISMS IN THE INTERTIDAL ZONES OF SANDY BEACHES
R. BALLY (Zoology Department, University of Cape Town, Private Bag, Rondebosch 7700, South Africa)
1. INTRODUCTION
Our understanding of the factors that affect the
distribution of organisms on sandy beaches has
gone through two stages. The process of learning
has been especially hampered by the fact that even
the largest beach organisms are usually buried
in the sand during the period when the intertidal
zone is most accessible to researchers. In the
first phases of our understanding, researchers
observed that sandy beach organisms were distri-
buted into zones, similar to those found on rocky
shores (Stephen, 1929, 1930; Elmhirst, 1931;
Newcombe, 1935; Rees, 1939; Watkin, 1942;
Brady, 1943; Southward, 1953).
Following the proposal for a universal zonation
scheme for rocky shores (Stephenson, Stephenson,
1949), a similar universal scheme was suggested
for sandy beaches by Dahl (1952). Like the
system for rocky shores, Dahl's scheme was
defined in terms of the type of fauna living in
each zone, of which Dahl recognized three. This
zonation scheme has been widely accepted and used
for exposed beaches allover the world (see
Bally, in press, for review), although frequently
with local modifications.
The next stage in the development of our under-
standing occurred when Salvat (1964) proposed a
new zonation scheme based on the interstitial
hydrodynamic conditions of the beach that are
responsible for the zonation shown by organisms.
Although it seems at first sight merely to be a
more sophisticated zonation scheme, it is, in
391
fact, a substantial advance in the way' we look
at the distribution of beach organisms. It
involves a departure from the simple recording
of the intertidal position of organisms and is
an attempt to understand the underlying causes
for these distributions.
Salvat (1964, 1966, 1967) proposed four zones,
the existence of which was confirmed, in essence,
by Pollock, Hummon (1971). The only studies to
date (other than those of Sal vat) making use of
this zonation scheme have been Withers (1977),
Koop, Griffiths (1982) and Bally (1981, in press).
The zonation schemes discussed so far have dealt
exclusively with macrofauna, or organisms re-
tained by a 1 mm mesh sieve. Mare (1942), how-
ever, defined two other components of the benthic
fauna; the meiofauna (passing through a 1 mm
sieve, but retained by a 0.1 mm sieve) and the
microfauna (passing through a 0.1 mm sieve).
These two components are also present on sandy
beaches, but, because of their small size, work
started on them only recently. One of the first
studies on the distribution of meiofauna was that
of Delamare Deboutteville (1954) on a mystaco-
carid. On exposed or high-energy beaches, where
anoxic conditions only occur at some depth or not
at all, the meiofauna may occur down to depths of
1 m or more below the surface, and thus there is
both a vertical and a horizontal succession of
species on the beach. The first studies to
demonstrate this clearly were those of McLachlan
392
(1977a, b, 1980), from which it appears that
organisms follow humidity conditions through the
beach, regardless of the depth from the surface.
This represents a fundamental difference between
the macro- and meiofauna in that the former must
remain close to the surface since they depend
more directly on the incoming tides and overlying
waters for their sources of food than do the
meiofauna. In essence, therefore, McLachlan's
studies have shown that the meiofauna exhibit a
similar zonation to the macrofauna, but that
this is not restricted to the surface layers of
sediment.
The distribution of microfauna on sandy beaches
is as yet only poorly understood. The studies
by Koop, Griffiths (1982) and Koop et al.
(1982a, b) have shown that their distribution can
depend to a large extent on the presence of wrack
beds. Because of the difficulty of identifying
organisms and because of their ability to form
resistant forms for long periods until conditions
become suitable, it is difficult to relate the
distribution of microfauna to purely physical
factors.
The amount of research carried out on the macro~,
meio- and microfauna and our consequent under-
standing of the ecology of these three groups is,
therefore, directly related to the size of the
organisms and the ease with which they are
studied. Most of the studies mentioned above
have only dealt with differences across the shore,
while only a few studies have been carried out
specifically on long-shore distributions. These
are the studies by Moran (1972) on the micro-
distribution of G~~o~aQQ~ ~anctU6, Dauer,
Simon (1975) on the long-shore distribution of
polychaetes, Moueza, Chessel (1976) on the long-
shore distribution of Vonax tnunQul~, Kamihira
(1979) on the distribution of peracarids, Bally
(1981) on both long-shore and across-shore
distributions of macrofaunal species and meiofaun~
and Hockin (1982) on the spatial structure of a
harpacticoid copepod community. All these
studies, with the exception of Dauer, Simon
(1975), showed distinct patchiness in the distri-
bution of organisms, confirming a number of
earlier observations both on macrofauna (e.g.
Pichon, 1967; Fielder, 1971; Gray, 1971) and
meiofauna (e.g. Gray, Reiger, 1971; Gerlach,
1977; Platt, 1977; Giere, 1979). It should be
noted that Dauer, Simon (1975) only looked at
three transects, situated 0.8 and 1.6 km apart.
Patchiness in the distribution of organisms is a
well-known phenomenon in both terrestrial and
marine environments. The superficial appearance
of many sandy beaches, however, tends to convey
the idea of uniformity to the observer. This is
probably the main reason why the possibility of
patchiness in distributions has been considered
so rarely in earlier studies.
This paper reports on a detailed study which
follows earlier findings by Bally (1981). The
extent of patchiness shown by the distributions
of sandy beach organisms are examined, together
with the intertidal distributions of a number of
physical and biotic factors. The relationships
between the distributions of the organisms and
the physical and biotic parameters are then
explored and discussed.
2. MATERIALS AND METHODS
The site studied was at Yzerfontein beach
(33°20'S, 18°08'E), a 20 km long sandy beach
lying 80 km to the north of Cape Town on the
west coast of South Africa (see Fig. 1). This
is a high energy beach, with a mean sediment
diameter of 2.30 phi and the beach does not
possess a black or reducing layer (Bally, 1981).
A sampling grid was used, consisting of 56
stations in an 8 by 7 grid. Station intervals
were 8 m across the beach and 15 m along the
beach, within the intertidal zone at low water

spring tide. The grid thus covered an area
measuring 105 by 48 m.
Capo
Yzerfonteln
Cape of Good Hope
FIGURE 1. The position of ths study site at
Yzerfontein, on the west coast of South Africa.
At each station, a quadrat of 33 cm 2 was excavated
to a depth of 30 cm, deep enough to sample about
95% of the macrofauna (Bally, 1981). The exca-
vated sand was then passed through a 1 mm mesh
sieve to remove the macrofauna. Four cores, mea-
suring 10 cm 2 in cross section and 30 cm long,
were taken from the top 30 cm of sediment, where
at least 85% of the meiofauna occurs on this beach
(Bally, 1981). These were mixed on site and a
300 m£ sub-sample taken. The meiofauna were
separated from sand by a modified Oostenbrink
extractor, with an efficiency of about 90%
(Fricke, 1979).
Bacteria were sampled by collecting 10 m£ of
sediment which was immediately mixed with 10 m£
of 10% glutaraldehyde solution and stored at 5°C
prior to processing. The samples were then
sonicated three times for 10 minutes in glutaral-
dehyde solution, then filtered onto 0.2 ~m
393
nucleopore membrane filter and counted by the
acridine-orange direct count method (Hobbie et
al., 1977; Linley et al., 1981).
Moisture content of the sediment at each site was
measured by collecting approximately 26 m£ of
sediment with a small corer and sealing this in
a container. The sediment was then weighed in
the laboratory, dried to constant weight and
re-weighed. Samples for grain size analysis
were also collected and, after dialysis and dry-
ing, were passed through a settling column in
order to determine their characteristics (Flem~n~
Thum, 1978). The results were analysed by a
computer programme and mean diameters, sorting,
kurtosis and skewness values were obtained for
each sample. The settling column is a consider-
ably more accurate method than the traditional
nested sieve system, and has both higher
resolution and reproducibility (Flemming, Thum,
1978). A third group of sediment samples were
heated for 3 hours at 450°C in a muffle furnace
to obtain the weight of organic carbon, while a
fourth group were placed in concentrated hydro-
chloric acid to determine the proportion of
calcium carbonate in the samples.
Relative permeability of the sediment was
measured by dropping a 164 g, 10 mm diameter
stainless steel rod down a 1 m tube. The depth
to which the rod penetrated into the sediment
was measured four times at each station, and the
mean was taken. A similar method has been used
by Griffiths, Griffiths (in press).
Nutrients were collected by thoroughly washing
600 m£ of sediment in 400 m£ of distilled water.
The supernatant was then decanted into containers
and stored frozen. The samples were later
passed through an autoanalyser to determine
concentrations of nitrate, nitrite, phosphate
and silicate. While the absolute values
obtained by this method should be viewed with
394
some caution, they are nevertheless useful for
comparison between the samples collected.
The distributions of the values obtained for the
parameters discussed above were mapped using a
SACLANT computer programme, originally obtained
from the NATO SACLANT Anti-submarine Warfare
Research Unit, La Spezia, Italy. In interpret-
ing these figures, it should be remembered that
the programme interpolates between sampling
points. It is conventional to interpolate in
this manner when analysing the results of inter-
tidal transects, and the SACLANT programme
extends this interpolation to stations along the
beach as well as across it. The actual sampling
points are marked as dots in the figures.
3. RESULTS AND DISCUSSION
To determine the interrelationships of the
physical, chemical and biological parameters of
the beach, their distributions were mapped.
These distributions are discussed in the follow-
ing sections.
3.1 - Physical parameters
3.1.1 Mean sediment diameters. These showed
considerable variation both across the beach and
along it, with no definite trends visible. One
sample of coarse sand was found in the zone of
resurgence (Salvat, 1964), but this may simply
have been an anomaly. In general, there seem
to be patches of fine sand alternating more or
less randomly with coarser areas (Fig. 2).
o "..
o 30 60 90
FIGURE 2 Spatial distribution of mean sediment
diameters (phi) in the intertidal zone on Yzer-
fontein beach. Values on axes are distance (m)
while contour values represent the mean sediment
diameters (phi).
3.1.2 Sorting. Sorting values tended to decrease
upshore, indicating better sorted sand towards
the landward edge of the beach. There was also
a fair amount of variation, both across and along
the beach, except for the uppermost zone (the
zone of drying) which appeared uniformly well-
sorted (Fig. 3).
48,t--~-
FIGURE 3. Spatial distribution of sorting values
(QO). Axes as in Fig. 2.
3.1.3 Skewness and kurtosis. These measures
showed considerable variation both across and
along the beach. All the sand samples were
negatively skewed, although about 20% were
classified as 'almost symmetrical'. Some 25% of
the samples were mesokurtic, while the remainder
were leptokurtic. No clear zonational or long-
shore trends in these two parameters were
observed.
3.1.4 Relative penetrability. Relative
penetrability showed a certain amount of long-
shore variability, whereas distinct zones
occurred across the beach (Fig. 4). These tended
to involve very high penetrability at the high-
water mark, low penetrability lower down (in the
zone named 'zone of retention' by Salvat, 1964),
and a higher penetrability in the lower, water-
saturated regions of the beach.
3.1.5 Moisture content of sand. The collection
method used for this parameter was, unfortunately,
LWS
found to affect moisture contents above 26% of
sediment dry weight. This means that the results
obtained are of little use for the lower portion
of the beach. They do show, however, a decrease

for eight beaches on the South African west and
south coasts.
3.3.3 Bacterial distributions. The abundance of
bacteria was highest at the low water mark and
decreased fairly rapidly towards the high tide
level. There were considerable variations, how-
ever, in numbers, both across and along the shore
(Fig. 11). The intertidal distribution of
bacteria is the reverse of that found by Pugh
et al. (1974) and Anderson et al. (1981) but
similar to that observed by Koop, Griffiths (1982~
FIGURE 11. Spatial distribution of bacteria
(No.10 6 ,m£-1 sediment).
3.4 Interrelationships between parameters
There can be no doubt that animal distributions
are affected by sediments, as has been shown by a
considerable body of literature. Most of the
evidence, however, deals with subtidal communities
(e.g. Sanders, 1958, 1960; McNulty et al., 1962;
Driscoll, 1975; Warwick, Davies, 1977;
Biernbaum,1979; James, Gibson, 1980). Inter-
tidal evidence published to date is less informa-
tive. Carr (1976) showed some correlation
between particle sizes and faunal distributions,
although she noted that most of these effects
were probably due to differences in amounts of
shelter at the various sites she studied. A studies is that of Hummon et al. (1976) who found
similar conclusion was reached by Eleftheriou,
Nicholson (1975) working on Scottish beaches. In
New Zealand, Grange (1977) found that deposit-
feeders favour the finest sediments of inter-
tidal flats, carnivores and scavengers prefer
intermediate grades, and suspension-feeders are
more abundant in coarser sediments. Conrad (1976)
found that sand grain angularity affected
395
colonization of sediments by meiofauna and that
selection of sediments differed according to
taxon. Hulings, Gray (1976) found that on tidal
beaches, sorting and median diameter of the
sediment were related to meiofaunal abundance.
In addition, Dale (1974) reported that bacterial
densities were highly correlated with grain
size and other granulometric properties. Some
studies have looked at individual species and
found individual preferences (e.g. Meadows,
1964a; Gray. 1971; Brown, Talbot, 1972; Brown,
1973; Muir, 1977; Holanov, Hendrickson, 1980;
Khayrallah, Jones, 1980; Bally, 1981; Griffiths,
Griffiths, in press).
It has also been recognized that grain size and
sorting can affect both the amount of pore water
and the permeability of the sand, and that the
distribution of infauna may be related directly
to these factors and only indirectly to grain
size and sorting (e.g. Harrison, Wass, 1965;
Jansson, 1967).
In this study the only animal-sediment size
relationship observed was that Vonax ~~ and
Scolelep~ ~quamata were both associated with
areas of finer sand in the zone of retention.
This was not a very strong correlation, however,
and other relationships are discussed below.
Comparatively few attempts have been made to
relate distributions of sandy beach fauna to
factors other than those related directly to the
sediments or exposure of beaches. Among the few
that differential aggregations of meiofauna on a
beach in Delaware were associated with maturing
eggs of the horseshoe crab L~ul~, while Gerlach
(1977) suggested that in some instances the
patchiness of meiofauna might be caused by
attraction to decaying organisms. Similarly
Lee et al. (1977) found that meiofauna can detect
small patches of different food types and
396
in moisture content corresponding to the area of
rapidly increasing relative penetrability near
the top of the shore.
FIGURE 4. Spatial distribution of relative
penetrability of the sediment. Axes as for Fig.
2.
3.2 Chemical parameters
The nutrient values recorded here are to be
viewed with caution. Most published data deals
with nutrient values of interstitial water,
collected from the water tables of beaches. The
values reported here, however, come from the top
10 cm of sand, in many cases well above the water
table. It is these values which are most likely
to be related to the distributions of the
organisms collected in this study, rather than
samples from the water table which may be as much
as 2 m below the surface. The results given
below are, therefore, a reflection of concentra-
tions in the water of retention, i.e. the surface
film of water on the sand grains in the case of
samples from the upper half of the beach, and
these values are diluted by increased moisture
content of the sand towards the low water mark.
Secondly, the values obtained are also affected
by the period each site was exposed prior to
collection; they are, therefore, cumulative
values that have accrued in the period between
exposure and collection. Because of these two
factors, it is difficult to compare the results
with those in the literature. Nevertheless,
comparative results are given, in order to place
the results into some form of context, although
admittedly vague. The results are of greatest
use, however, as an indication of distributions
of nutrient values at low tide.
3.2.1 Nitrate levels. Nitrate 1eve 1s were
highest at the high-water mark (in the zone of
drying) . Relatively high values were also
measured at the low water mark, in the zone of
saturation. Most noticeable, however, was a
'trough' of low values coincident with the zone
of retention (Fig. 5). The range of values
(1.34 - 67.61 ~g-at.e-1) were an order of magni-
tude higher than the values found by Rodriguez et
al. (1980) on beaches in Spain. On the other
hand, they fall within the lower portion of the
range of values recorded by Koop et al. (1982b)
in the interstitial water of an experimental
beach microcosm and the ranges observed by Pugh
(1976) in a model sandy beach, Oliff et al.
(1970) in Natal, Pugh et al. (1974) in two Welsh
beaches, Orren et al. (1981) on 20 Cape beaches
and Stenton-Dozey (1983) at Kommetjie beach, near
Cape Town.
Figure 5. Spatial distribution of Nitrate
(~g-at.e-1) in the sediment. Axes as for Fig. 2.
3.2.2 Nitrite levels. Nitrite values were
distributed in four peaks in the study area, two
being in the zone of saturation and two around
the mid-tide level, in the zone of retention
(Fig. 6). The highest values measured were
around 2.5 ~g-at.e-1 at the low water mark, while
intertidally two peaks of 1.2 ~g-at.e-1 were
recorded. These values too, fall within the
lower portion of the range of values found by
Koop et al. (1982b) in their microcosm experiment,
but agree with values recorded by Pugh et al.
(1974), Pugh (1976), Orren et al. (1981) and

Stenton-Dozey (1983).
48 HWS
FIGURE 6. Spatial distribution of Nitrite
(Vg-at.£-1) in the sediment at Yzerfontein beach.
Axes are for Fig. 2.
3.2.3 Ammonia levels. Very high ammonia levels
were encountered in the zone of retention (with
one peak in excess of 600 ~g-at.£-1). Concentra-
tions fell away sharply, however, both above and
below this zone (Fig. 7). These results are
within the same order of magnitude as those
measured by Koop et al. (1982b) and Stenton-Dozey
(1983). The results are between one and two
orders of magnitude higher than Pugh (1976) found
in his model beach, although the lower ones are
similar to values found by Oliff et al. (1970)
and Pugh et al. (1974).
481r-----~Q~.~-====-~--------~------~
32
16
.~
OJ!===~====~========~====~==~~~LWS
o ~ 00 00
Figure 7. Spatial distribution of Ammonia
(~g-at.£-1) in the sediment. Axes as for Fig. 2.
3.2.4 Silicate levels. Silicate values showed
the least variation of the nutrients analysed,
both along and across the study area. A small
peak occurred in the zone of retention. Silicate
concentrations ranged from 13 to 27 ~g-at.£-1 and
these values are substantially higher than those
recorded by Stenton-Dozey (1983), but similar to
findings by Orren et al. (1981).
3.2.5 Phosphate levels. Phosphate levels reach
397
their highest concentrations at the mid-tide
level (Fig. 8). The area studied had two con-
centration peaks at this level. The range of
values obtained are one order of magnitude higher
than those found by Pugh et al. (1974) in Wales
and Rodriguez et al. (1980) on Spanish beaches
but fairly similar to those found by Orren et al.
(1981), Koop et al. (1982b) and Stenton-Dozey
(1983) •
O~==~~~~===T====~==~~==~==~LWS
a 30 60
FIGURE 8. Spatial distribution of Phosphate
(~g-at.£-1) in the sediment. Axes as for Fig. 2.
3.2.6 Organic carbon content of sediment.· There
was a general decrease in organic content of the
sediment upshore with a certain degree of
variation both along and across the beach. The
amounts recorded ranged from 0.56 to 1.03% of
sediment dry weight and these values are similar
to those recorded previously by Bally (1981) from
the same beach.
3.3 Biological parameters
3.3.1 Macrofaunal distributions. Sixteen macro-
faunal species were recorded from the study area
during this survey. These were the beetles
Acantho~c~ nu6ico~~ and Pachyphatekia
capen6~, the amphipod Tato~ch~tia capeno~, the
isopods Tylo~ g~nul~, Ex~olana natateno~,
Pontogeloidu latipu, EWLydice longicM~ and
Exo~pha~oma ~ucatLt~on, the mysid G~~o~acc~
p~ammodyzu, the cumacean Cumop~~ ~ob~za, the
nemertean C~eb~~ 6~c~, the bivalve Vonax
~~, the gastropod Bullia digiZ~ and the
polychaetes Scolelep~ ~quamaZa, Sigalion capenoe
and NephZy~ capeno~. t10st of these species
were only found at a few sampling sites, and in
398

low numbers. By far the most important were P. All the species showed irregular distributions
latipv>, E. .[0 Ylg-LCCotlvU.6 , V. -6VULa and S. -6quamata, along the beach as well. The diversity of macro-
both in terms of numbers and biomass. It is faunal species was also highest in the zone of
interesting to note that all four species were retention, although this is a function of the
found predominantly in the mid-tidal regions, abundance of the most important species in this
although there were slight differences (Fig. 9). region, and the very low numbers of the species
occurring in the zone of saturation.
a
48,~----~r-------~~-------------------;HWS 3.3.2 Meiofaunal distributions. Meiofaunal
abundances were relatively low in the zone of
saturation and showed a single high peak in the
zone of retention (Fig. lOa). Above this,
-~ ___ o.oo
densities declined, only to increase again in the
zone of drying. When the logarithm of the
OJ~0====~====~30~==~====~6~0====~~~9~0~===~LWS nematode/copepod ratios are plotted, a similar
pattern appears (Fig. lOb), although the ratios
b are highest in the zone of drying. Since
48~-----------------------===~--------~HWS
harpacticoid copepods occur in very low numbers
indeed in the upper half of the intertidal zone,
the presence of only one or two more animals in
[;00.00-
a sample can affect the ratio radically. In
• • .~. -;,oo-'-~ order to minimize this effect, the rat i os have
0,00· • • • • •
0.00 _ _ _ _
_ been expressed as logarithms. The ratios from
~ !LWS
30 60 90 the lower half of the intertidal zone are
similar to those found by Hennig et al. (1982)
481~C~--------~====----------==------~--~HWS
48 a

32

16

C).~§
16 .)~~.
g
.
0,00

OJE=====~===7~==~====~====~==~Oi,OO~==~LWS
o 30 60 90

FIGURE 9. Spatial distribution of the major

(b
(d
l Scco'[e..£eph.
macrofaunal species (No.m- 2 ). (a) VoYtax -6VULa,
quamata. (c)
-6
PoYttoge..£oidv> .[atipv>.
EMycUCCe..[O YlgiccoltrU..6.
FIGURE 10. Spatial distribution of meiofauna.
(a) Meiofaunal densities (No.l00 mt sediment- l ).
(b) Log. nematode/copepod ratios.

aggregate around them. Meiofaunal amphipods,
archiannelids, copepods and the macrofaunal poly-
chaete SQofeLep~ 6ufig~~o~a have been found to
show preferences for sands containing favourable
species of bacteria (~eadows, 1964b; Gray, 1966,
1967a, 1967b, 1968, 1971; Ravenel, Thistle,
1981). Tunnicliffe, Risk (1977) found that
densities of the bivalve Maeoma batthiea was
positively correlated with the density of bacteria
in the sediment, and Meyer-Reil, Faubel (1980)
found high meiofaunal biomasses coincided with
low bacterial masses. Rieper (1982) demonstrated
feeding preferences for particular bacterial
species in harpacticoid copepods. Finally Bally
(1981) suggested that competition, predator-prey
interactions and population structure may also
affect faunal distributions.
This study shows that some additional factors
could also influence distributions, although it
is necessary to be cautious in interpreting the
results. Of the non-biological parameters
measured, a number show distributions similar to
each other. Thus ammonia, nitrate, phosphate
and, to a lesser extent, nitrite, all show a
close correlation to the distribution of relative
penetrability levels. It is likely, therefore,
that it is complex factor combinations that
affect the distribution of organisms rather than
just one or two.
The high levels of ammonia and phosphate in the
zone of retention suggest that there is some
correlation between these and the distributions
of Vo~ax and SeofeLep~. On the other hand,
these two species are also associated with the
areas of lowest penetrability (see Fig. 4, 7,8,
11). The landward cut-off points of the distri-
butions of Vo~ax and SeofeLep~ also coincide with
lowered moisture contents of the sand. It is
unfortunate that the higher moisture values are
suspect, since Harrison, Wass (1965) found good
correlations between sediment water content and
399
frequency of infaunal species even from subtidal
samples.
Furthermore, there are fairly good correlations
between meiofaunal abundances and the nutrients
ammonia, nitrate and phosphate, and consequently
also with Vo~ax and SeofeLep~. This is to be
expected, since the vast majority of marine
animals are ammonotelic, i.e. their main excretion
product is ammonia (Kormanik, Cameron, 1981).
Finally, the sand in the zone of retention
contains slightly more organic matter in areas
associated with the distributional peaks of these
two species.
The "valley" of low meiofaunal numbers and low
logged nematode/copepod ratios is closely
correlated with the highest numbers of the carni-
vorous isopod EUAyd{ee fo~g~eo~~. This implies
that these isopods are preying on meiofauna and
having a noticeable effect in this region due to
the large numbers of EUAyd{ee, although Bell
(1980) has pointed out that macrofauna may also
cause a decline in meiofaunal numbers due to
disturbance of the sediment by activities such as
burrowing.
There is always the possibility that patchiness
is a random occurrence, or simply the consequence
of the irregularity of the swash across the inter-
tidal zone, resulting in organisms being deposited
differentially. This must undoubtedly playa
certain role in the distribution of the more
mobile organisms, and especially with the smaller
forms. The importance of this process is not
known, however, although Grant (1981) concluded
that faunal distribution in bedforms on a high-
energy sandflat was related to active preference
by the organisms concerned rather than to
passive hydrodynamic sorting.
One feature that needs resolving is the behaviour
of patches over time, since the results presented
400
here merely represent the distribution during one
period of low tide. The longevity of patches
would appear to vary both with species and,
presumably, with environmental conditions. For
example, Gray (1971) reported that patches of
Seolelep~ 6uliginoca change their locality on the
beach frequently, although he did not state how
frequently. Muir (1977) found that aggregations
of Talonehectia eapenc~ changed daily and were
largely the consequence of the distribution of
stranded food items. Efford (1965), on the other
hand, concluded that aggregations of Em~
analoga on Scripps Beach may keep their identity
for at least four months and that they do not mix
with neighbouring aggregations. This is despite
the fact that the aggregations migrate up and
down the beach with the tide.
It is clear that the distribution of organisms on
sandy beaches is influenced by complexes of
factors (cf Vernberg, Coull, 1981). Many of these
we have not even begun to measure, while we do
not yet properly understand the effects of others.
The microenvironments of sandy beach sediments
are also very complex (e.g. Anderson, Meadows,
1978; see also Flemming, Fricke, 1983) and still
poorly understood. One thing that emerges clearly
from the literature, however, is that different
species are affected in different ways by their
environment (e.g. Ravenel, Thistle, 1981). This
has long been acknowledged for the macrofauna,
but is becoming increasingly apparent for the
meiofauna as well, and will probably also be
found to apply to the bacteria and other micro-
fauna (Koop, 1982). Moreover, many species have
different tolerances at different stages of their
life cycles (e.g. Bally, 1981, in press).
When biological interactions are considered as
well, it is easy to imagine how involved and
convoluted the factors affecting distribution in
the field may. become. For example, the
distribution of a, prey species may be ultimately
influenced by factors affecting the predator
species where predation is an important factor.
In situations where there is a marked longshore
drift, aggregations of prey and predator species
may travel at different rates along the beach,
resulting in occasional periods of high predation
alternating with periods of reduced predator
activity. Temporal studies are needed to
elucidate this point.
Finally, little is understood of processes that
occur on sandy beaches at high tide. It is quite
possible that conditions prevailing during this
period exert a strong influence on the low-tide
distribution of animals, since many members of
the swimming macrofauna bury themselves in the
sand before the beach becomes exposed with the
ebbing tide.
4. CONCLUSIONS
It is clear from the evidence presented here,
that there are very many factors affecting the
distribution of organisms on sandy beaches. It
also appears likely that the effect of some
factors is modulated by other factors, resulting
in a very complex set of interactions. Much
remains to be investigated before the causes
underlying the distribution of beach organisms is
fully understood.
5. ACKNOWLEDGEMENTS
I would like to thank Rod Baxter, Daryl Birkett,
Suzanne Painting and Elaine Rumbak for assistance
in the field. I am also indebted to Mary Armour,
Daryl Birkett, Suzanne Painting and Jeanie
Stenton-Dozey for their help and advice, and to
Leonora Fox for her flawless typing. The nutrient
analyses were undertaken by the Sea Fisheries
Institute, whose kind help is acknowledged.
REFERENCES
Anderson JG, Boonruang P and Meadows PS (1981)

Interrelationships between chlorophylls, carbon,
nitrogen and heterotrophic bacteria in an inter-
tidal sediment transect, Mar. Ecol. Prog. Ser. 6,
277-283.
Anderson JG and Meadows PS (1978) Microenviron-
ments in marine sediments, Proc. Roy. Soc.
Edinburgh 76B, 1-16.
Bally R (1981) The ecology of three sandy
beaches on the west coast of South Africa, PhD
thesis, University of Cape Town.
Bally R (In press) Intertidal zonation on
sandy beaches of the west coast of South Africa,
Cah. Biol. mar.
Bell SS (1980) Meiofauna-macrofauna interactions
in a high salt marsh habitat, Ecol. Monogr. 50,
487-505.
Biernbaum CK (1979) Influence of sedimentary
factors on the distribution of benthic amphipods
of Fishers Island Sound, Connecticut, J. expo
mar. Biol. Ecol. 38, 201-223.
Brady F (1943) The distribution of the fauna
of some intertidal sands and muds on the
Northumberland coast, J. Anim. Ecol. 12, 27-41.
Brown AC (1973) The ecology of sandy beaches
of the Cape Peninsula, South Africa. Part 4:
Observations on two intertidal Isopoda:
Eunydic~ long~con~ (Studer) and Exo~pha~oma
tAuncatit~on Barnard, Trans. Roy. Soc. S.
Afr. 40, 381-404.
Brown AC and Talbot MS (1972) The ecology of
sandy beaches of the Cape Peninsula, South
Africa. Part 3: A study of G~tAo~acQ~
p~ammodyt~ Tattersall (Crustacea; Mysidacea),
Trans. Roy. Soc. S. Afr. 40, 309-333.
Carr CE (1976) Distribution pattern of the
intertidal macrofauna at Langebaan Lagoon,
Unpublished Project, Zoology Department,
University of Cape Town.
Conrad JE (1976) Sand grain angularity as a
factor affecting colonization by marine
meiofauna, Vie Milieu 26B, 181-198.
Dahl E (1952) Some aspects of the ecology and
zonation of the fauna of sandy beaches, Oikos 4,
1-27.
Dale NG (1974) Bacteria in intertidal
sediments: factors related to their distribution,
Limnol. Oceanogr. 19, 509-518.
Dauer OM and Simon JL (1975) Lateral or along-
shore distribution of the polychaetous annelids
of an intertidal, sandy habitat, Mar. Biol. 31,
363-370.
Delamare Deboutteville C (1964) L-ecologie du
mystacocaride V~och~oc~ n~an~ f.
b~cay~~~ Delamare sur les cotes du Golfe de
Gascogne, Vie Milieu 5, 310-329.
Driscoll EG (1975) Sediment-animal-water inter-
action, Buzzards Bay, Massachusetts, J. mar.
Res. 33, 275-302.
Efford IE (1965) Aggregation in the sand crab
Em~a analoga (Stimpson), J. Anim. Ecol. 34,
63-75.
Eleftheriou A and Nicholson MD (1975) The
effects of exposure on beach fauna, Cah. Biol.
mar. 16, 695-710.
401
Elmhirst R (1931) Studies in the Scottish
marine fauna - the Crustacea of the sandy and
muddy areas of the tidal zone, Proc. Roy. Soc.
Edinburgh 51, 169-175.
Fielder DR (1971) Some aspects of distribution
and population structure in the sand bubbler
crab Scop~~ ~n6lata Milne Edwards, 1873
(Decapoda, Ocypodidae), Aust. J. mar. Freshwat.
Res. 22,41-47.
Flemming BW and Fricke AH (1983) Selective
microhabitat colonization by interstitial
meiofauna as a function of grain size. In
McLachlan A and Erasmus T, eds. Sandy beaches
as ecosystems. pp , The Hague, Junk
Publishers.
Flemming BW and Thurn AB (1978) The settling
tube - a hydraulic method for grain size analysis
of sands, Kieler Meeresforsch., Sonderh. 4,
82-95.
Fricke AH (1979) Meiofauna extraction
efficiency by a modified Oostenbrinck apparatus,
Helgolander wiss. Meeresunters. 32, 436-443.
Gerlach SA (1977) Attraction to decaying
organisms as a possible cause for patchy
distribution of nematodes in a Bermuda beach,
Ophelia 16, 151-165.
Giere 0 (1979) Some apparatus for preference
experiments with meiofauna, J. expo mar. Biol.
Ecol. 41, 125-131.
Gray JS (1966) The attractive factor of inter-
tidal sands to Pnoto~~ ~ymb~o~cu~, J. mar.
biol. Ass. U.K. 46, 627-645.
Gray JS (1967a) Substrate selection by the
archiannelid Pnoto~~ nubnophanyng~~,
Helgolander wiss. Meeresunters. 11, 253-269.
Gray JS (1967b) Substrate selection by the
archiannelid Pnoto~~ hypol~uc~ Armenante,
J. expo mar. Biol. Ecol. 1, 47-54.
Gray JS (1968) An experimental approach to the
ecology of the harpacticid L~pta~tac~
con~~ct~ Lang, J. expo mar. Biol. Ecol. 2,
278-292.
Gray JS (1971) Factors controlling population
localizations in polychaete worms, Vie Milieu
Suppl. 22, 702-722.
Gray JS and Johnson RM (1970) The bacteria of
a sandy beach as an ecological factor affecting
the interstitial gastrotrich Tunban~a hy~na
Schultze, J. expo mar. Biol. Ecol. 4, 119-133.
Gray JS and Rieger RM (1971) A quantitative
study of the meiofauna of an exposed sandy beach
at Robin Hood's Bay, Yorkshire, J. mar. biol.
Ass. U.K. 51, 1-19.
Griffiths CL and Griffiths RJ (In press)
Biology and distribution of the littoral rove
beetle P~amathobl~~~ pun~~~~ (Le Conte)
(Coleoptera: Staphylinidae), Hydrobiologia.
Grange KR (1977) Littoral benthos-sediment
relationships in Manakau Harbour, New Zealand,
N.Z. J. mar. Freshwater Res. 11,111-123.
Grant J (1981) Sediment transport and
disturbance on an intertidal sandflat: infaunal
distribution and recolonization, Mar. Ecol.
Prog. Ser. 6, 249-255.
402
Harrison Wand Wass ML (1965) Frequencies of
infaunal invertebrates related to water content
of Chesapeake Bay sediments, Southeast. Geol. 6,
177-187.
Hennig HF-KO, Fricke AH. Greenwood PJ and Eagle
GA (1982) Relationships between meiofaunal
population densities and physico-chemical
properties of unpolluted sandy beaches. Envir.
Monit. Assess. 1, 337-344.
Hobbie JE, Daley RT and Jasper S (1977) Use of
Nucleopore filters for counting bacteria by
fl uorescence mi croscopy, App 1. env. tli crob i 01 •
33, 1225-1228.
Hockin DC (1982) The spatial population
structure of a harpacticoid copepod community in
spring, Hydrobiologia 96, 201-209.
Holanov, S.H. and Hendrickson, J.R. 1980. The
relationship of sand moisture to burrowing depth
of the sand beach isopod Tylo¢ pun~. J. expo
mar. Biol. Ecol., 46, 81-88.
Hulings NC and Gray JS (1976) Physical factors
controlling abundance of meiofauna on tidal and
atidal beaches, Mar. Biol. 34, 77-83.
Hummon WD, Fleeger JW and Hummon MR (1976)
Meiofauna-macrofauna interactions: 1. Sand beach
meiofauna affected by maturing Limulu¢ eggs,
Chesapeake Sci. 17, 297-299.
James CJ and Gibson R (1980) The distribution
of the polychaete Cap~ella cap~ (Fabricius)
in dock sediments, Estuar. coast. mar. Sci. 10,
671-683.
Jansson B-O (1967) The significance of grain
size and pore water content for the interstitial
fauna of sandy beaches, Oikos 18, 311-322.
Kamihira Y (1979) Ecological studies of macro-
fauna on a sandy beach of Hakodate, Japan. II.
On the distribution of peracarids and the factors
influencing their distribution, Bull. Fac. Fish.,
Hokkaido Univ. 30, 133-143.
Khayrallah NH and Jones AM (1980) The ecology
of Bathypo~ela pilo¢a (Amphipoda:
Haustoriidae) in the Tay Estuary. II. Factors
affecting the micro-distribution, Proc. Roy. Soc.
Edinburgh 78B, s121-s130.
Koop K (1982) Fluxes of material associated
with the decomposition of kelp on exposed sandy
beaches and adjacent habitats, PhD thesis,
University of Cape Town.
Koop K and Griffiths CL (1982) The relative
significance of bacteria, meio- and macrofauna on
an exposed sandy beach, Mar. Biol. 66, 295-300.
Koop K, Newell RC and Lucas MI (1982a)
Biodegradation and carbon flow based on kelp
{Ecklonla maximal debris in a sandy beach
microcosm, Mar. Ecol. Prog. Ser. 7,315-326.
Koop K, Newell RC and Lucas MI (1982b) Micro-
bial regeneration of nutrients from the
decomposition of macrophyte debris on the shore,
Mar. Ecol. Prog. Ser. 9, 91-96.
Kormanik GA and Cameron IN (1981) Ammonia
excretion in animals that breathe water: a
review, Mar. Biol. Lett. 2, 11-23.
Lee J,z, Tietjen JH, Mastropaolo C and Rubin H
(1977) Food quality and heterogeneous spatial
distribution of meiofauna Helgolander wiss.
Meeresunters. 30, 272-282.
Linley EAS, Newell RC and Bosma S (1981)
Heterotrophic utilisation of mucilage released
during fragmentation of kelp {Ecklonla maxima
and Larn-LrtaJUa palUdal. 1. Development of
microbial communities associated with the
degradation of kelp mucilage, Mar. Ecol. Prog.
Ser. 4, 31-41.
Mare MF (1942) A study of a marine benthic
community with special reference to the micro-
organisms, J. mar. biol. Ass. U.K. 25, 517-554.
McLachlan A (1977a) Studies on the psammo-
littoral meiofauna of Algoa Bay, South Africa.
II. The distribution, composition and biomass
of the meiofauna and macrofauna, Zool. Afr. 12,
33-60.
McLachlan A (1977b) Composition, distribution,
abundance and biomass of the macrofauna and
meiofauna of four sandy beaches, Zool. Afr. 12,
279-306.
McLachlan A (1980) Intertidal zonation of
macrofauna and stratification of meiofauna on
high energy sandy beaches in the eastern Cape,
South Africa, Trans. Roy. Soc. S. Afr. 44,
213-223.
McNulty JK, Work RC and Moore HB (1962) Some
relationships between the infauna of the level
bottom and the sediment in South Florida, Bull.
mar. Sci. Gulf Caribbean 12, 322-332.
Meadows PS (1964a) Substrate selection by
Co~oph.i..um species: the particle size of
substrates, J. Anim. Ecol. 33, 387-394.
Meadows PS (1964b) Experiments on substrate
selection by Co~oph.i..urn species: films and
bacteria on sand particles, J. expo Biol. 41,
499-511.
Meyer-Reil L-A and Faubel A (1980) Uptake of
organic matter by meiofaunal organisms and
interrelationships with bacteria, Mar. Ecol.
Prog. Ser. 3, 251-256.
Moran S (1972) Ecology and distribution of the
sand-dwelling mysid Ga¢tno¢accu¢ ¢anciu¢ (van
Beneden, 1961) along the Mediterranean sandy
shore of Israel, Crustaceana Suppl. 3,357-361.
Moueza Mand Chessel D (1976) Contribution a
1 'etude de la biologie de Vonax tnunculu¢ L.
(Mollusque: Lamellibranche) dans 1 'Algerois:
analyse statistique de la dispersion le long
d'une plage en Baie de Bou Ismail, J. expo mar.
Biol. Ecol. 21, 211-221.
Muir DG (1977) The biology of Talo~che¢t.i..a
capen¢~ (Amphipoda, Talitridae), MSc thesis,
University of Cape Town.
Newcombe CL (1935) Certain environmental factors
of a sand beach in the St Andrews region, New
Brunswick, with a preliminary designation of the
intertidal communities, J. Ecol. 33, 334-355.
Oliff WD, Gardner BD, Turner WD and Sharp JB
(1970) The chemistry of the interstitial water
as a measure of conditions in a sandy beach,
Water Res. 4, 179-188.
Orren MJ, Eagle GA, Fricke AH, Gledhill WJ,
Greenwood PJ and Hennig HF-KO (1981) The
 403

chemist'ry and meiofauna of some unpolluted sandy marine fauna. Additiona.1 observations of the
beaches in South Africa, Water SA 7, 203-210. fauna of the sandy and muddy areas of the tidal
Pichon M (1967) Contribution a 1 'etude des zone, Trans. Roy. Soc. Edinburgh 56, 521-535.
peup1ements de 1a zone intertida1e sur sables Stephenson TA and Stephenson A (1949) The
fins et sables vaseux non fixes dans 1a region universal features of zonation between tide-
de Tu1ear, Rec. Trav. Sta. mar. Endoume, Supp1. marks on rocky coasts, J. Eco1. 37, 289-305.
7, 57-100. Tunnic1iffe V and Risk MJ (1977) Relationships
Platt H~·~ (1977) Ecology of free.,living nematodes between the bivalve Maeoma batthiea and bacteria
from an intertidal sand flat in Strangford Lough, in intertidal sediments: Minas Basin, Bay of
Northern Ireland, Estuar. coast. mar. Sci. 5, Fundy, J. Mar. Res. 35, 499-507.
685-693. Vernberg WB and Coull BC (1981) t·1eiofauna.
Pollock LW and Hummon WD (1971) Cyclic changes In Vernberg FJ and Vernberg ~JB, eds. Functional
in interstitial water content, atmospheric adaptations of marine organisms, pp.147-177, New
exposure and temperature in a marine beach, York, Academic Press.
Limno1. Oceanogr. 16, 522-535. \~arwick RMand Davies JR (1977) The distribu-
Pugh KB (1976) An annual cycle, at constant tion of sublittoral macrofauna communities in
temperature, in a model sandy beach. I. Nutrient the Bristol Channel in relation to the substrate,
chemistry, J. expo mar. Bio1. Eco1. 22, 179-192. Estuar. coast. mar. Sci. 5, 267-288.
Pugh KB, Andrews AB, Gibbs CF, Davis SJ and Watkin EE (1942) The macrofauna of the
Floodgate GD (1974) Some physical, chemical and intertidal sand at Kames Bay, Millport,
microbiological characteristics of two beaches of Buteshire, Trans. Roy. Soc. Edinburgh 60,
Anglesey, J. expo mar. Bio1. Eco1. 15,305-333. 543-561.
Ravenel WS and Thistle D (1981) The effect of Withers RG (1977) Soft-shore macrobenthos along
sediment characteristics on the distribution of the south-west coast of Wales, Estuar. coast1
two subtidal harpacticoid species, J. expo mar. mar. Sci. 5,467-484.
Bio1. Eco1. 50,289-301.
Rees CB (1939) Notes on the ecology of the sandy
beaches of North Donegal, Proc. Roy. Irish Acad.
45B, 215-229.
Rieper M. (1982) Feeding preferences of marine
harpacticoid copepods for various species of
bacteria. Mar. Eco1. Prog. Ser. 7, 303-307.
Rodriguez V, Ibanez M and Rodriguez J (1980)
Eco10gie des anne1ides po1ychetes de quelques
p1ages de 1a Baie d'A1geciras, Vie Milieu 30,
131-138.
Sa1vat B (1964) Les conditions hydrodynamiques
interstitie11es des sediments meubles intertidaux
et 1a repartition vertica1e de 1a fauna endogee,
C. R. Acad. Sci. Paris 259, 1576-1579.
Sal vat B (1966) E~ydiee pufehna Leach 1815 et
E~ydiee a66i~ Hansen 1905 (Isopodes: Ciro-
1anidae). Taxonomie, etho10gie, repartition
vertica1e et cycle reproducteur, Acta Soc. linn.
Bordeaux A 193, 1-77.
Sal vat B (1967) La macrofaune carcino10gique
endogee des sediments meubles intertidaux
(Tanaidaces, Isopodes et Amphipodes): etho10gie,
bionomie et cycle bio10gique, Mem. Mus. natn.
Hist. nat. Ser. A 45, 1-275.
Sanders HL (1958) Benthic studies in Buzzards
Bay. 1. Animal-sediment relationships, Limno1.
Oceanogr. 3, 245-258.
Sanders HL (1960) Benthic studies in Buzzards
Bay. 3. The structure of the soft-bottom
community, Limno1. Oceanogr. 5, 138-153.
Stenton-Dozey J (1983) Stranded kelp - its
fauna and influence on sandy beach energetics.
MSc thesis, University of Cape Town.
Stephen AC (1929) Studies on the Scottish marine
fauna: the fauna of the sandy and muddy areas of
the tidal zone, Trans. Roy. Soc. Edinburgh 56,
291-306.
Stephen AC (1930) Studies on the Scottish
 405

DYNAMIC ZONATION OF STAPHYLINOID BEETLES (COLEOPTERA: STAPHYLINOIDEA) ON A SANDY BEACH IN EAST AFRICA

G. CHELAZZI 1 , L. CHELAZZI 2 and S. FOCARDI 3 (lInstituto di Zoologia dell'Universita di Firenze, Italy,

2Centro di Studio per la Faunistica ed Ecologia Tropicali del Consiglio Nazionale delle Ricerche,
Firenze, Italy, 3 Inst ituto Nazionale di Biologia della Selvaggina, Bologna, Italy)

1. INTRODUCTION
The ecology of East African sandy beaches is poorly known in comparison to other coastal systems of
subtropical and temperate areas (Pardi, 1976). For this reason the Centro di Studio per la Faunistica
ed Ecologia Tropicali of the Italian National Council for the Scientific Research promoted a study on
the sandy shore and dune of Sar Uanle (0 29'48"S-42 25'30"E) along the southern Somalia coast. Besides
the faunistic inventory and structural analysis of the beach-dune system, the aim of the programme is
the study of the short- and long-term dynamics of species distribution and the interactions in space
and time between the different organisms living on a tropical sandy littoral.
The first analyses on migratory and rhythmic behaviour were made on crustaceans: Amphipoda (Pardi et
al., 1974), Decapoda (Vannini, 1976 a-b) and terrestrial Isopoda (Chelazzi, Ferrara, 1978).
The present study deals with the behavioural ecology of the species of the superfamily Staphylinoidea
which constitutes an important fraction on the littoral fauna of all the sandy and rocky shores in
temperate as well as in tropical and subtropical regions (Doyen, 1976; Moore, Legner, 1976). Despite
their importance in the littoral environment, the behaviour of staphylinids and other related small
sandy beach beetles is poorly known and mostly limited to short etho-ecological notes in faunistic
inventories.

2. MATERIALS AND METHODS
The collection of animals on the Sar Uanle beach
and dune was made during different seasons, using
a system of integrated traps which gave informa-
tion on the temporal and spatial patterns of ani-
mal activity (Chelazzi et al., 1976) and was ac-
companied by the recording of macro- and microcli-
matic parameters along the sea-land axis (Messana
et al., 1977; Vannini et al., 1978) and by botan-
ical surveys.
The present study was based on data collected dur-
ing October-November 1971 (dry season) and June-
July 1973 (wet season), using directional pitfall
traps (cross-traps).
The daily activity rhythm of the different spe-
cies was obtained by averaging the number of ani-
mals collected each hour of the day throughout
the entire study period. The tidal activity curve
was obtained by dividing the synodic month into
segments comprised between two successive high
tides and overlapping all these segments of time
made equal to one, independent of the actual
length of the HT-HT time which varies throughout
the spring-neap cycle.
For both the daily and tidal curves of activity
we computed the resultant vector whose direction
represents the mean angle of phase (i.e. the mean
time of activity) and whose length is proportional
to the clustering in time of the distribution of
the captures (i.e. to the width of the activity
phase of each species). The vector length tends
to one when all the specimens are collected at
the same time and to zero if the activity of the
species is completely asynchronous.
The design of the cross-trap (see also Tongiorgi,
1968) allowed an analysis of the directional com-
ponents of the activity of the species considered.
The frequency of the specimens collected in each
of the four cups of the pitfall trap were used to
compute the resultant vector, whose direction in-
dicates the mean orientation of the flow of ani-
mals crossing the zone of the beach where the
trap was placed. The length of this vector de-
pends on the clustering of the circular distribu-
tion around the mean direction, reaching unity
when all the animals head in the same direction.
Multiplying the vector length for the sample size
and projecting the vector on the land-sea axis
406

gave information on the movement normal to the
shore line. This orientation index assumes in-
creasingly positive values as the landward flow
increases in strength, reversing to negative val-
ues when the movement is mainly seaward. It ap-
proaches zero when the two tendencies are balanc-
ed or when no animals are collected.
Distribution of the animals collected in the four
cups at each daily and tidal time was tested a-
gainst randomness using the "v-test" (Batschelet,
1981) assuming as zero direction both the landward
and seaward direction.
The Staphylinidae and ptiliidae were identified
by P.M. Hammond of the British Museum (N.H.), Lon-
don and Cl. Besuchet of the Museum d'Histoire Na-
turelle, Geneve, respectively.
The taxonomy of the Histeridae and Pselaphidae of
Sar Uanle were studied by Therond (1974) and Ca-
stellini (1974).
Philontus cf. cinctipennis Fauvel, 1875
Philontus cliens Eppelsh., 1890
Philontus cf. igacus Tottenham, 1955
Cilea cf. floralis Cameron, 1953
? genus Aleocharinae Sharp, 1833
Aleochara trivialis Kraatz, 1859
+ Arena Fauvel, 1862 sp.
? Atheta Thoms, 1859 sp.
Atheta (Acrotona) Thoms, 1861 sp.
+ Cameronium flavipenne Cameron, 1944
Cordalia Jacobs, 1924 sp.
+ Heterota pictipennis Fauvel, 1905
Pselaphidae
+ Halorabyxis pardii Castellini, 1974
Afroplectus Jeannel, 1952 sp.
The total number of specimens, as well as the fre-
quency of each species depends strictly on season
(Fig. 1); histerids (namely Halacritus algarum) are

3. RESULTS
The Staphylinoidea collected on the Sar Uanle DRY WET
beach belong to the families Histeridae, ptilii- N =2710 6009
dae, Pselaphidae and Staphylinidae (Table 1).
While the first three families are almost mono- HISTERIDAE
specific, 26 species of staphylinids were collect-
PTiLIiDAE
ed, eight of which were sufficient in number to
permit a quantitative study on the dynamics of STAPHVLlNIDAE
their distribution and activity. PSElAPHIDAE

TABLE 1. Staphylinoid taxa collected at Sar Uanle.

60 40 20 0 20 40 60 %
The asterisk (+) indicates the species considered I I I I I I I
in the present study
N = 868 2227

Histeridae
Caflus eorallieola
+ Halacritus algarum (Schmidt, 1893)
C. ef. filum
Chalcionellus aeneovirens (Schmidt, 1890)
Hypocacculus ascendens Reichardt, 1932 C. font leola STAPHVLININAE
Ptiliidae
C. nautleus
+ Actinopteryx fucicola Allib., 1844
Staphylinidae C. ragazzli

? Actocharis Fauvel, 1869sp. Arena sp.

Acanthoglossa Kraatz, 1859 sp.
Cameronium ALEOCHARINAE
Astenus cf. bimaculatus Er., 1839
Oedichirus cf. dollmani Brnhaver, 1928 Heterota pietlpennis
Paederus sabaeus Er., 1839 alii
Pinophilinus Eickelbaum, 1908 sp.
Scopaeus punctatellus Fauvel, 1905
+ Cafius corallicola Fairm., 1949 FIGURE 1. Frequencies (expressed in percent) of the
+ Cafius cf. filum Kiesw., 1849 four families of Staphylinoidea and of the eight
+ Cafius fonticola Er., 1839 species of Staphylinidae collected during the dry
+ Cafius nauticus Fairm., 1849 and wet seasons by the cross traps on the Sar Uanle
+ Cafius ragazzii Gestro, 1885 shore.
Gabrontus Tottenham, 1955 sp.
Philontus bisignatus Boh., 1848
 407

the most frequent during the dry season followed

by staphylinids, pselaphids (Halorabixis pardii) 12 hr
and ptilids (Actinopteryx fucicola). During the wet
season A. fucicola prevails, followed by the sta-'
phylinid complex. Among the relatively constant
staphylinids important variations emerge in the

,I
I
N
3000

2000
06
1000
!
500
f 100

~ __-U~-U~__~~____~~~-L__~__-+O
-9 -8 -7 -6 -5 -4 -3 -2

~ 24
EHWS --------------
AHTL -------------
MTL -----------
MLWN i SUPRA HT
!-LlTTORAL.--:----- DUNE - - - - - - B

-40 -30 -20 -10 o +10 +20 +30

m

FIGURE 2. Average (wet-dry seasons) zonation of

I
each species. The length of each segment is propor- ~

tional to the total number of collected specimens. ~...

Below, profile of the beach and dune showing the III
Q
'i
Gl
average tidal levels. The metric scale on abscisse j:

"~
refers to the seaward (-) and landward (+) distance
from the upper limit of the beach. Species symbols
as in Fig. 3. -;.

m4

A. fuclcola •
l-
Ll.
:;: H. pardii.
3
.
III
LT
Q C. nauticus AH. algarum
c
~
~C. corallicola
c FIGURE 4. Daily (A) and tidal (B) components of the
~ 2
..
rhythmic activity of each species during wet sea~
> C. flavlpenne son. The direction of resultant vectors indicates
Q CD the average time of activity, while the vector

•
Arena sp. C H. pictipennls
I- length is proportional to the degree of activity
W
~ synchronization. Species symbols as in Fig. 3.
@C.cf. filum
C. ragazzii

@C. fonticola different species: during the dry season the Cafius
O+-~--.-~--.-~--.-~--.-~--.-~--.-~--~
-9 -8 -7 -6 -5 -4 -3 -2 spp. are dominant, while during the wet season they
WET ZONATION m are outnumbered by the three species of the subfam-
ily Aleocharinae.
FIGURE 3. Extent of the seasonal shift in zonation
The 11 species considered in this study inhabit the
of each species (ordinate) plotted in function of
proximal belt of the supralittoral zone (Fig. 2),
their zonation during the wet season (abscisse).
just seaward of the vertical step which constitutes
Metric scale on the abscisse as in Fig. 2.
the natural border between the sandy beach and the
408

dune. Only 1.6% of the animals were collected land- positive peak in the afternoon (landward migra-
ward of this border, on the dune. The average (dry- tion). The effect of these moments on the average
wet seasons) zonations of the different species are hourly zonation of the species is evident (Fig. 7,
very close and overlap considerably. below). Plotting the same index in function of the
Nevertheless, a gradient is evident between the
most proximal ptilids and the distal staphylinids DRY WET
of the Cafius genus. Moreover, crowding species in N = 2710 6009
this narrow belt is reduced by the seasonal varia-
tion in faunal composition.
Each species of this system shows a distinct sea- 50 %
sonal change in zonation, consisting in a seaward
40
shift between wet and dry season, which is larger
for the upper species (Fig. 3). This produces a 30
clustering of species toward the lower supralittor- >
c 20
al during the dry season, which is balanced in c
terms of local density by the net reduction of ac- 10
tive fauna during this part of the year.
Hourly inspection of the pitfall traps throughout
the entire synodic month (wet season), and distinc- 10

tion between day and night collections (dry sea- ...:z: 20

son), permitted an analysis of the temporal activ- CI
ity pattern of each species. Concerning the daily z 30
rhythm, the species are clustered in a nocturnal 40
and diurnal group (Fig. 4, A). The diurnal group
comprises the largest number of species, whose ac- 50 %
tivity is maximum during the first part of the day.
Among the most abundant species, two are nocturnal
(H. algarum and Arena sp.) and two diurnal (A. fu- FIGURE 5. Overall diurnal and nocturnal activity
cicola and H. pictipennis). (expressed in percent) of all the Staphylinoidea
The tidal diagram (Fig. 4, B) shows that with the during dry and wet season. Sample size is shown
exception of two species which are active mainly above the histograms.
during high tide (Cafius ragazzii) and full low
tide (H. pardii), the others again form two groups
which are respectively active during outgoing and D W D W D W
N= 163 24 191 16 56 41
incoming tides.
The overall daily activity on the beach depends a-
gain on season (Fig. 5). During dry periods noctur-
nal activity prevails, while during rainy seasons
50%
the trend is reversed. The change of the overall >
daily pattern is mainly due to the season-dependent c
c
variation of the species: the diurnal A. fucicola
nearly disappears and the nocturnal H. pardii in- 0
creases during the wet season. Moreover, some Ca-
fius spp. (C. fonticola, C. nauticus and C. ragaz-
...:z:
CI
zii) show seasonal variation in their daily pat-
z 50%
tern, becoming nocturnal during the dry season
(Fig. 6).
The directional analysis of data showed that the
vast majority of species perform zonal migrations C. fontlcola C. nautlcus C. ragazzii
along the shore. Plotting the index of orientation
in function of the time of the day provided the FIGURE 6. Diurnal and nocturnal activity (express-
daily migrational component of each species. In ~ ed in percent) of three Cafius spp. during dry and
pictipennis (Fig. 7) a negative morning peak (sea- wet season.
ward migration) is followed by a period of rela-
tively non-polarized movement, and then by a minor
 409

tidal phase shows (in Arena sp.) the tidal compo-

nent of the migration and consequent zonal varia-
tion (Fig. 8). Fig. 9 shows the result of the "V-
test" analysis applied to the four most abundant
species during the wet season. Both nocturnal spe-
cies showed a landward migration during the rise of
activity following sunset. This migration reverses

+20 LAND N = 900

,
10 t
0
10

20
z SEA

....
!:!
c
40
30

z
- 50
0:
0
60

70

80

-90
6
24 6 12 18 24 hr
z

m
I I I I I I I I I I I I I I I I I I I I I I I I I
.
o
c 7
z
0
o
N
-1
8
z 2 ---------
..!:!
c
3
9
z
0 4
N
5
FIGURE 8. Tidal variation of the orientation index
6
relative to Arena sp. Below is shown the tidal va-
riation of the average zonation of the same spe-
8 cies; metric scale as in Fig. 2.

FIGURE 7. Daily variation of the orientation index
relative to Heterota pictipennis. Below is shown
the hourly variation of average zonation of the
same species; metric scale as in Fig. 2.
around sunrise, at the end of the activity phase.
The two diurnal species move seaward when becoming
active (in the first part of the morning), but
while A. fucicola soon reverses its movement, ~
pictipennis remains seaward oriented throughout
the morning phase and returns to the proximal part
of its belt during the afternoon.
The tidal rhythm of migration of the four species
(Fig. 9, right) is more differentiated: only Arena
sp. simply follows the tide with its movement,
while H. algarumand A. fucicola show a landward
movement when the tide recedes and seaward orien-
tation only during full low tide. In the case of
H. pictipennis the dominant movement is always
seaward, probably because the cross-traps failed
to reveal their quick landward return made by fly-
ing at a height above the directional arms of the
device. Of the eight most abundant species analyz-
ed, only H. pardii never showed significant move-
ment normal to the shore line throughout either
the daily or tidal cycle.
4. DISCUSSION
The 11 species considered in this study are typic-
al of the littoral environment, since they were
almost never collected in the extralittoral. Never-
theless, they cannot be considered as marine spe-
cies in the sense proposed by Doyen (1976) since
all seem to actively avoid submergence, in contrast
410

to such littoral beetles as the staphylinid Ble- the species present at Sar Uanle and the staphyli-
'dius spectabilis (Evans et al., 1971). Burrowing nid fauna of the southern California sandy beaches
in the sand during high tide, such as observed in (Moore, Legner, 1976).
Tinopinus pictus (Craig, 1970), also seems to be The results of this dynamic investigation suggest
absent in the present species. that the sympatry of this group of species is per-
Concerning their zonation, it is evident that the mitted by a niche apportionment based mainly on
results of a dynamic study cannot be easily com- different strategies of space and time exploita-
pared to the static representations offered by tion. For the more frequent species is evident that
other authors. Moore, Legner (1976) divided the the zonal distance between synchronous species is
larger than that between the species which move in
DAILY RHYTHM TIDAL RHYTHM different phases (Fig. 10). Temporal shifts be-
tween species which compete for space and resources
A is a well documented phenomenon among terrestrial
N=1375 beetles (Williams, 1959), but assumes a particular
adaptive value in the littoral environment where
resources are packed in a narrow belt and stress
B sources and resources fluctuate in time in a com-
N=846 plex but somehow predictable manner, following the
short-term (diurnal-tidal) and long-term (synodic)

r--------.------------------~------__r-1 m
C
N=2276

H. pardll

•
-3
A. fueleols
D
N=900 z
H. slgsrum o
~
C
-5 Z
oN
C. nsutleus
C. eorallleois
24 0.8 12 18 24
hr C. ragsllil

•
C. flsvlpenne
FIGURE 9. Curves of daily and tidal activity of -7
H. pletlpennls
the four more abundant staphylinoids, recorded C. ef. filum
C. font leola
during the wet season. Above each curve are indi- Arena sp. -11• •_
cated the times of statistically significant (V
test: ~<0.01) landward (upward arrows) and sea-
ward (downward arrows) migration flow. A) Halacri- 12 18
tus algarum; B) Arena sp.; C) Actinopteryx fuci- TIME hr
cola; D) Heterota pictipennis.
FIGURE 10. Separation in time and in space between
the different Staphylinoidea during the wet season.
staphylinids of the Pacific coast of North Ameri-
The extension of the phases of most intense activ-
ca into three main ecological groups, each inhab-
ity (black horizontal bars) is plotted in function
iting a distinct subzone of the sandy beach. Such
of the average zonation recorded during the same
a clear distinction cannot be recognized between
phases. The width of the bars indicates the abun-
the staphylinoids of Sar Uanle, since all the most
dance of each species. For the metric scale on or-
abundant species - with the exception of H. pardii
dinate see Fig. 2.
- perform rhythmic changes of zonation. For in-
stance, the Cafius spp. whose average zonation is
rhythmicity. In fact, all the species investigated
in the supralittoral, can reach the lower eulit-
showed an integrated activity rhythm with distinct,
toral during some activity phases, Nevertheless,
specific diurnal and tidal components. Similar in-
there is a gener,al agreement between zonation of
tegrated activity rhythms are widespread among

littoral organisms (Enright, 1975; Palmer, 1976;
Saunders, 1976) and are also found in intertidal
beetles (Evans, 1976; Treherne, Foster, 1977). En-
right (1970) discussed the ecological importance
of endogenous timing for beach animals, which can
initiate and stop their zonal migrations in the
appropriate daily and tidal phase, regardless of
occasional stimuli. Evans (1976) showed the pres-
ence of circadian and circa tidal rhythms in the
carabid Thalassostrechus barbarae, which persisted
under constant conditions for seven and three days,
respectively. Without direct experiments we cannot
speculate on the presence of endogenous timing in
the species investigated. It would be interesting
to study the temporal pattern of the same species
on other sandy beaches, in order to verify if the
daily and tidal components of their activity are
constant and species-specific or whether they may
vary in different ecological conditions.
The different modulation of activity in space and
time can be accompained by other mechanisms of
niche-apportionment. All the species considered
frequent wrack mounds on the beach and it is evi-
dent that the most marine species (such as Cafius
spp. and Aleocharinae) utilize the daily input
from the sea, while the more terrestrial ones
(Histeridae, Ptiliidae, Pselaphidae) frequent the
older decaying mounds whose entomological associ-
ation is more complex. Special food requirements
such as predation and parasitism, well known among
rove beetles are not probable in this group of
abundant species, but are possibly present among
the other less frequent staphylinid species col-
lected on the same beach.
The migratory species probably depend on different
mechanisms (Enright, 1978) in performing their
zonal movements, ranging from beach slope orienta-
tion to anemotactic responses, which were discov-
ered in other sandy beach zonal migrators (Erco-
lini, Scapini, 1974; Vannini, 1975). But the most
probable mechanism for the diurnal species remains
sun compass orientation which works in sandy beach
carabids and tenebrionids (Papi, 1955; Pardi,
1955) and also in riparian staphylinids of temper-
ate areas \Ercolini, Badino, 1961; Ercolini, 1963;
Ercolini, Scapini, 1976).
ACKNOWLEDGEMENTS
We wish to thank Prof. L. Pardi for his criticism
of the manuscript. We are also indebted to all the
members of the Centro di Studio per la Faunisti-
ca ed Ecologia Tropicali of the Italian National
Council for the Scientific Research (C.N.R.) and
of the Istituto di Zoologia dell'Universita, Flor-
ence who participated in the research missions
411
on the Sar Uanle beach. We are grateful to Dr P.
M. Hammond for the identification of the staphy-
linids.
REFERENCES
Batschelet E (1981) Circular statistics in
biology. London, Academic Press, 371 pp.
Castellini G (1974) Ricerche sul litorale del-
la Somalia. La spiaggia e la duna di Sar Uanle.
2. Pselaphidae (Coleoptera). Monitore zool. ital.
(N.S.) Supp1. 5, 183-189.
Chelazzi G and Ferrara F (1978) Researches on
the coast of Somalia. The shore and the dune of
Sar Uanle. 19. Zonation and activity of terrestri-
al isopods (Oniscoidea). Monitore zool. ital. (N.
S.) Suppl. 9, 189-219.
Chelazzi L, Chelazzi G, Ercolini A, Ferrara F,
Messana G, Messeri P, Pardi Land Vannini M (1976)
Un programma di studio ecologico del litorale sab-
bioso in Somalia. 3. Tecniche di raccolta. Boll.
Zool. 43, 411-412.
Craig P (1970) The behaviour and distribution
of the intertidal sand beetle Tinopinus pictus
(Coleoptera: Staphylinidae). Ecology 51, 1012-
1017.
Doyen JT (1976) Marine beetles (Coleoptera ex-
cluding Staphylinidae). In Cheng L, ed. Marine
insects, pp. 497-519. Amsterdam, North-Holland.
Enright JT (1970) Ecological aspects of endo-
genous rhythmicity. Ann. Rev. Ecol. Syst. 1,
221-238.
Enright JT (1975) Orientation in time: endo-
genous clock. In Kinne 0, ed. Marine ecology,
Vol. 2, pp. 917-944. London, Wiley J and Sons.
Enright JT (1978) Migration and homing of
marine invertebrates: a poutpourri of strategies.
In Schmidt-Koenig K and Keeton WT, ed. Animal
migration, navigation and homing, pp. 440-446.
Berlin-Heidelberg, Springer-Verlag.
Ercolini A (1963) Ricerche sull'orientamento
astronomico di Paederus rubrothoracicus Goeze
(Coleoptera-Staphylinidae). Monitore zool. ital.
71, 416-429.
Ercolini A and Badino G (1961) L'orientamento
astronomico di Paederus rubrothoracicus Goeze
(Coleoptera-Staphylinidae). Boll. Zool. U.Z.I.
28, 421-432.
Ercolini A and Scapini F (1974) Sun compass
and shore slope in the orientation of littoral
amphipods (Talitrus salta tor Montagu). Monitore
zoo1. ita1. (N.S.) 8, 85-115.
Ercolini A and Scapini F (1976) Fototassia
negativa e orientamento astronomico solare in due
specie di Stafilinidi ripari (Paederus rubrotho-
racicus Goeze e Stenus bipunctatus Erichson).
412
Redia 59, 135-153.
Evans WG (1976) Circadian and circatidal loco-
motory rhythms in the intertidal beetle Thalasso-
strecus barbarae' (Horn): Carabidae. J. expo mar.
Biol. Ecol. 22, 79-90.
Evans PD, Ruscoe CNE and Treherne JE (1971) Ob-
servations on the biology and submergence behav-
iour of some littoral beetles. J. mar. biol. Ass.
U.K. 51, 375-386.
Messana G, Chelazzi G, Chelazzi L, Ercolini A,
Ferrara F, Messeri P, Pardi Land Vannini M (1977)
Researches on the coast of Somalia. The shore and
the dune of Sar Uanle. 12. Physical environment:
microclimate and soil. Monitore zool. ital. (N.S.)
Suppl. 9, 147-181.
Moore I and Legner EF (1976) Intertidal rove
beetles (Coleoptera: Staphylinidae). In Cheng L,
ed. Marine insects, pp. 521-551. Amsterdam, North-
Holland.
Palmer JD (1976) An introduction to biological
rhythms. New York, Academic Press, 375 pp.
Papi P (1955) Orientamento astronomico di alcu-
ni Carabidi. Atti Soc. tosc. Sci. nat., Memorie
62 (B), 83-97.
Pardi L (1955) Orientamento solare in un Tene-
brionide alofilo: Phaleria provincialis Fauv.
(Coleoptera). Boll. Ist. Mus. Zool. Univ. Torino
5, 1-39.
Pardi L (1976) Researches on the coast of So-
malia. The shore and the dune of Sar Uanle. Intro-
duction. Monitore zool. ital. (N.S.) Suppl. 8,
179-193.
Pardi L, Ercolini A and Ferrara F (1974) Ritmo
di attivita e migrazioni di un Crostaceo Anfipodo
(Talorchestia martensii Weber) sul litorale della
Somalia. Atti Accad. naz. Lincei Rc. (Cl. Sci.
fis. mat. nat.) 55, 609-623.
Saunders DS (1976) Insect clocks. Oxford, Per-
gamon Press, 279 pp.
Therond J (1974) Histerides d'Afrique Orientale
et Sud-Orientale. Monitore zool. ital. (N.S.) Sup-
pl. 5, 101-109.
Tongiorgi P (1968) Ricerche ecologiche sugli Ar-
tropodi di una spiaggia sabbiosa del litorale tir-
renico. 1. Caratteristiche generali dell'ambiente
e metodi di studio. Redia 48, 165-177.
Treherne JE and Foster WA (1977) Diel activity
of an intertidal beetle, Dicheirotrichus gustavi
Crotch. J. Anim. Ecol. 46, 127-138.
Vannini M (1975) Researches on the coast of So-
malia. The shore and the dune of Sar Uanle. 4. Ori-
entation and anemotaxis in the land hermit crab,
Coenobita rugosus Milne Edwards. Monitore zool.
ital. (N.S.) Suppl. 6, 57-90.
Vannini M (1976) Researches on the coast of So-
malia. The shore and the dune of Sar Uanle. 7.
Field observations on the periodical transdunal
migrations of the hermit crab Coenobita rugosus
Milne Edwards. Monitore zool. ital. (N.S.) Suppl.
7, 145-185.
Vannini M (1976) Researches on the coast of So-
malia. The shore and the dune of Sar Uanle. 10.
Sandy beach decapods. Monitore zool. ital. (N.S.)
Suppl. 8, 255-286.
Vannini M, Chelazzi G, Chelazzi L, Ercolini A,
Ferrara F, Messana G, Messeri P and Pardi L (1978)
Researches on the coast of Somalia. The shore and
the dune of Sar Uanle. 13. Physical environment:
geomorphological notes, climate and tides. Moni-
tore zool. ital. (N.S.) Suppl. 9, 249-271.
Williams G (1959) Seasonal and diurnal activ-
ity of Carabidae, with particular reference to
Nebria, Notiophilus and Feronia. J. Anim. Ecol.
28, 309-330.
 413

A PRELIMINARY ACCOUNT OF THE ECOLOGY OF SANDY BEACHES IN SOUTHERN BRAZIL.*

N.M. GIANUCA (Departamento de Oceanografia, Fund. Universidade de Rio Grande, C.P. 474, Rio Grande,
RS, Brazil)

1. INTRODUCTION.
Sandy beaches constitute by far the most long shore currents and also influence markedly
important littoral habitat along the southern the slope of the beach. In accordance with the
coast of Brazil, running almost without definition and rating system proposed by
interruption for nearly 700 kilometers (Fig.1). McLachlan (1980a) these beaches should be
Although it forms the habitat of two commercially classified as exposed.
important species of molluscs no detailed study
r -______~----------------------~~lat.S
of their ecology has yet been conducted.
Previous biological studies were concerned with
U
the taxonomy of Foraminifera (Closs, Barbarena,
1960) and Polychaeta (Orensanz, Gianuca, 1974).
More recently, Escofet et al. (1979) presented
the results of joint bionomical observations of
the sandy beaches of southern Brazil, Uruguay
and northern Argentina, indicating zonation
patterns, some of the biogeographical changes
and the importance of the Rio de la Plata as an
effective ecological barrier. The present paper
summarizes some of the preliminary results of an
on-going investigation describing the general
ecology of southern Brazilian beaches.
Uruguay

2. STUDY LOCALITY
The beaches in the area are very open and 52 51 50 49
subjected to moderate to strong wave action,
FIGURE 1. Map showing the distribution of sandy
having fine to very fine well sorted quartz sand.
beaches and the transect site.
The slope is gentle (1/30 - 1/50) but the inter-
tidal width only 30 m to 60 m since the tidal
Seasonal variations of temperature and salinity
range is very small (0.5 m). Winds can be
in the surf zone waters were significant with
considered an important factor, because they can
maximum values of 26.5°C and 37%0 and minimum
cause abnormally high or low water levels, strong
values of 12.5°C and 22%orespectively. The
maximum temperatures were recorded in February
*Investigation conducted at the Department of
and the minimum in J{lly, whilst the maximum
Oceanography, University of Southampton, U.K.
414

salinities were registered in March 1982 and the mortality and residual biomass method described
minimum in November 1980 and July 1981. Fig.2 by Crisp, (1971). For this purpose only the
shows the monthly averages for temperature and central band of the distribution of each species
salinity during the period of study. population was considered (Fig. 3).
Qualitative samples of infratidal, supratidal and
upper beach organisms were also taken.
A record of birds, fishes, crustaceans and
T.(OC) 5'(%0)
gastropods feeding upon the intertidal populations
28 33
was made and stomach contents were studied with
the purpose of identifying the main predator/
prey relationships.

250 9 AFDW m- 2 250

200 200
I
I
I
I

150 I 150
I
I
I
FIGURE 2. Seasonal changes in temperature
(dashed line) and salinity (solid line) during 100 100
____ E. brasiliensis
the period October 1980 - March 1982.

_ _ _ M. mactroides
3. METHODS
After preliminary observations a permanent 50 I ............ D. hanleyanus 50
transect site was established covering the inter- I

tidal zone in an area considered fairly represent-

............ ·· .. ·10
ative of the beach ecosystem of southern Brazil ..............
(Fig.1). A series of 10 quantitative samples was F M A M ~ ~ A 5 0
obtained monthly at the transect site for a
FIGURE 3. Cumulative production of each of the
period of 18 months. A quadrat-metallic marker main intertidal suspension-feeders in the period
(1/25 m- 2 ) and an mm sieve were used. It was October 1980 - October 1981.

impossible to use a 0.5 mm sieve due to the large

amount of small broken shells that obstructed it. 4. RESULTS AND DISCUSSION

After being sorted into species the organisms 4.1. Supratidal.

were measured and the flesh was dried to constant The upper fringe of the supratidal zone is

weight at 70°C. Ash-free dry weight was characterised by the presence of Ocypode

determined by loss of weight on ignition at 550°C. quadrata (Ocypodidae), and the lower fringe by

Net secondary production of the three major inter- Orchestoidea brasiliensis (Talitridae),

tidal species was estimated in terms of ash-free representing a transition between the subtropical

dry weight per square meter per anum using the and temperate areas of Dahl (1952) as stated by
 415

Escofet et al. (1979). Both species are mainly (Spionidae)

scavenger/detritivores but O. quadrata occasion- Nevertheless, it 1S in the lower intertidal,
ally preys upon the juveniles of the intertidal coinciding most of the time with the swashing
bivalves Mesodesma mactroides (Mesodesmatidae) zone, where the greatest number of species and
and Donax hanleyanus (Donacidae). Possibly this organisms are concentrated. Typical of this
behaviour represents a response to the temporary level are the sand crab E. brasilienBis, the
scarcity of carrion and detritus on that specific amphipods Phoxocephalopsis zimmeri (Haustoriidae)
part of the beach. and Bathyporeiapus biBetoBus (Oedicerotidae),
Two other conspicuous supratidal scavengers are the polychaetes Hemipodus olivier~ (Glyceridae)
the insects Phaleria brasiliensis (Tenebrionidae) and Sigalion cirriferum (Sigalionidae) and,
and Scapteriscus acletus (Gryllotalpidae). A around the LWS level, Diopatra viridis (Onuphidae),
number of other insect species, chiefly coleop- Macrochiridotea giambiagiae (Idoteidae) ,
terans (Staphylinidae and Cicindelidae), have MetamysidopBis elongata atlantica, Bowmaniella
been recorded. (Coifmannie Ua) brasiliensis (Mys idae) and
Some halophytes live in the upper supratidal, an CallianaBsa mirim (Callianassidae).
area only covered by the sea during storms. The
main species are Philoxerus portulacoides
(Amaranthaceae), Paspalum vaginatum, Spartina
ciliata and Panicum racemosum (Gramineae) and
Hydrocotile bonariensis (Umbelliferae).

4.2 Intertidal.
The intertidal zone was the area of main interest
during this study and it is here that the bulk of
the macrofaunal biomass is concentrated. About
99% of that biomass is represented by only four
species: three suspension feeders, M. mactroides
(Fig. 4), D. hanleyanus and Emerita brasiliensis
(Hippidae) and a scavenging isopod Excirolana
armata (Cirolanidae). The E. armata distribution
agrees with the cirolanid belt proposed by Dahl
FIGURE 4. Tidal migration of Mesodesma
(1952) but it is wider, spreading to the swashing
mactroideB, showing its high abundance.
zone where most of the juveniles tend to
concentrate. The cirolanid belt is shared 1n the
4.3. Trophic relationships.
upper part of the beach with adults of M.
Such a high diversity of intertidal species
mactroides, a deep burrower, and in the lower part
together with the high abundance and biomass of
with D.hanleyanus, a superficial burrower. The
the four main species (Fig. 6, 7, 8, & 9),
distribution of these bivalves tends to follow
attract a large number of predators. When the
the pattern described by McLachlan (1980b) for
tide is in, several surf-zone fishes such as the
Donax serra and D. sordidus in the Eastern Cape,
abundant TrachinotuB marginatuB (Carangidae),
South Africa. Other species sharing the upper
MenticirrhuB littoraliB (Sciaenidae) and
fringe of the cirolanid belt are the polychaetes
Oncopterus darwini (Pleuronectidae), feed on
EuzonuB furciferus (Opheliidae) and Spio gaucha
these populations being accompanied by the
 416

portuni d crabs Arenaeu s cribari us and Calline ctes

sapidus and several gastrop ods. Observ ations
suggest the occurre nce of a well defined
gAFDW
separat ion of trophic niches between the three m- 2
20775
most abundan t gastrop od species ; Olivan cillaria
auricul aria (Olivid ae) the larger one (Fig. 5)
5600
·t,
I
I
I
140

I
preying mainly upon the sand crab E. brasili ensis,
Olivan cillaria uretai preying mainly upon
D.hanle yanus, and Buccinanops duartei (Buccin idae)
100
feeding mainly on the juvenil es of M.mact roides. I

,
f
l
I
f
3000 f
f
'I
f
f
I
f I
\
\ f
I
50

I
f

-'
400 10
ONDJ FMA MJJA SOND J FM

FIGURE 6. Abundan ce (dashed line) and biomass

(solid line) of Mesodesrna rnactro ides during the
period of study; October 1980 - March 1982.

FIGURE 5. Olivan cillaria auricul aria holding a

capture d Mesodesma rnactro ides in its foot.
gAFDW
18250 m- 2
5600
96t~5
As the tide retreat s birds become the main , 140
I I
predato rs. Some of them such as Haernatopus I I
162/ '
palliat us (Haema topodid ae), Charad rius collari s
(Chara driidae ), Larus dorninicanus and Larus maculi -
pennis( Laridae ) are permane nt residen ts in the 100
region, while others are tempora ry visitor s during I I
, I
their migrati ons north and southwa rds. Norther n
300 I
migran ts are numeri cally the most importa nt I
visitor s. For these species the large expanse of
almost deserte d beaches with plenty of food must
,
1\
represe nt an ideal site to spend the boreal winter f

and to lay down reserve s prepari ng for the

," \
return migrati on. The most conspic uous norther n f
f
migrant s are Calidri s fuscico llis, by far the 400 f
10
mo_st abundan t. Calidri s alba, Calidri s canutus .. - ... -- f

FMAM J FM
(Scolop acidae) and Pluvia lis dominic a
FIGURE 7. Abundan ce and biomass of Donax
(Chara driidae ).
hanleya nus.
 417

The austral winter is the season of the southern

migrants, the most common of which are Charadrius
gAFDW falklandicus and Zonibyx modestus (Charadriidae).
9900 152 -2

t~
m Results of production studies regarding the three
,,to,
5600 140 main intertidal suspension-feeders revealed for

13900 the period October 1980 - October 1981 very high

values (Fig.3). The biomass produced by these
species is an important food source shared by most
100 of the above predators as revealed by the study
of the stomach contents.

3000 Lewin's papers (Lewin 1977, 1978, Lewin et al,

1979a, 1979b) on the relationship between the
Pacific razor-clam Siliqua patula and the surf
50
blooms of the diatoms Chaetocerus armatum and
Asterionella socialis in northern Oregon and
southern Washington beaches, drove our attention

;, to the possibility that the same sort of inter-

400 ' , 10
,~ action could take place in the southern Brazilian
beaches. In fact we have been observing for
several years on these beaches regular seasonal
FIGURE 8. Abundance and biomass of Emerita
brasi liens is . surf blooms of the diatom Asterionella glacialis
(= A. japonica), a species that very often
constitutes the bulk of the stomach contents of
gAFDW
7075 m- 2 M. mactroides and D. hanleyanus and sometimes

t
5950

,,t,,
also of E. brasiliensis.
5600 8.4
According to our records the A. glacialis blooms
I'
,, ' are much more conspicuous from late summer,
throughout the autumn and winter, tending to
disappear by mid-spring. Observations also
6.0
"
I, indicate that they tend to occur after rainy
II
, I
periods and southerly winds, being probably
,,
, I
I
3000 I associated with the resuspension of sediments and
I release of nutrients and with the discharge of
I 3.6
I many fresh water streams that are formed on such
\
\ occasions and run from behind the sand dunes.
The diatoms appear to be aggregated in the top
centimeter of the water, forming a brown scum
often in association with bubbles in the surf foam.
400 We have seen extensive diatom patches moving along
ONDJ FMAMJJASONDJ FM the beach either northwards or southwards,
FIGURE 9. Abundance and biomass of Excirolana depending on the prevailing long shore currents
armata. within the breakers. When the tide recedes it is
very common to find these diatom patches
418
deposited on the beach (Fig. 10).
In future, more detailed studies will certainly
be necessary in order to elucidate the origin of
the nutrients and other factors upon which the
blooms are dependent, as well as its relation
with the high intertidal production of the
suspension-feeding species. It could well be
that surf blooms and intertidal macrofauna and
meiofauna are interacting in the same sort of
semi-closed ecosystem proposed by McLachlan
(1980-81), where surf phytoplankton are the
primary producers, beach macrofauna the consumers
and interstitial fauna the decomposers.
FIGURE 10 AsterioneZZa gZaciaZis deposited on
thebea-ctl by the receding tide.
Finally, it must be said that despite the normal
high abundance and the observed rapid rate of
repopulation (Gianuca, 1982), M. mactpoides
almost disappear at the end of each summer
season around every important seaside resort
where it is usually consumed as food and
utilised as bait. Another threat that affects
the intertidal populations, especially the
juveniles and superficial burrowers, is the use
of the wet and most compacted part of the beaches
by vehicles of all kinds, including fishermen's
lorries.
5. ACY~Ov~EDGEMENTS
This research has been supported by a grant from
CAPES (Ministerio da Educacao e Cultura, Brazil)
and made pclssible by leave of absence from
Fundacao Universidade de Rio Grande, Brazil. We
are grateful to the following scientists who
have helped us regarding the taxonomic identi-
fication of some species: Dr E. Nonato (Polychaeta),
Dr M.S.A. Prado Pohr (Mysidacea) and Dr Y.Wakabara
(Amphipoda), all from the Instituto Oceanografico,
Universidade de Sao Paulo, Brazil, Dr D.K.McE.
Kevan and Dr V.R. Vickery (Gryllotalpidae) from
the Lyman Entomological Museum, McGill University,
Canada. Special thanks are due to Dr. M. Sheader
of the Department of Oceanography, University of
Southampton, U.K., for valuable comments on the
manuscript and to Conselho Nacional de Pesquisas
(CNPq), Brazil, which provided funds for the
attendance of this Symposium.
REFERENCES.
Closs D and Barbarena MC (1960) Foramin1feros
recentes da Praia do Cassino (Rio Grande, RGS),
Bol. Esc. Geologia UFRGS 5, 1-50.
Crisp DJ (1971) Energy flow measurements. In
Holme NA and McIntyre AD, ed. Methods for the
study of marine benthos, pp. 197-279. Oxford,
Blackwell Scientific Publications.
Dahl E (1952) Some aspects of the ecology
and zonation of the fauna on sandy beaches,
Oikos 4, 1-28.
Escofet A, Gianuca NM, Maytia S, Scarabino V
(1979) Playas arenosas del Atlantico Sudoccidental
entre los 29° y 43°L S.: consideraciones generales
y esquema bioceno16gico. In UNESCO, Memorias del
seminario sobre ecologia bentonica y sedimentacion
de la plataforma continental del Atlantico Sur,
pp. 245-258. Montevideo.
Gianuca NM (1982) Repovoamento de urn trecho de
praia arenosa afetado por deposi~~o de lama no
litoral do Rio Grande do SuI. In Resumos IX
Congresso Brasileiro de Zoologia, pp. 51-52.
Porto Alegre.
Lewin J (1977) Persistent blooms of surf
diatoms along the Northwest Coast. In Krauss R,
ed. The marine plant biomass of the Pacific
Northwest Coast, pp 81-92. Corvallis, Oregon
State University Press.
Lewin J (1978) The world of the razor-clam
beach, Pacific Search 12, 12-13.
Lewin J, Ching-hong Chen, Hruby T (1979a) Blooms
of surf-zone diatoms along the coast of the
Olympic Peninsula, Washington. X. Chemical
 419

composition of the surf diatom Chaetoceros

armatum and its major herbivore, the Pacific
razor clam Siliqua patula, Mar. BioI. 51, 259-265.
Lewin J, Eckman JE, Ware GN (1979b) Blooms of
surf-zone diatoms along the coast of the Olympic
Peninsula, Washington. XI. Regeneration of
ammonium in the surf environment by the Pacific
razor clam Siliqua patula, Mar, BioI. 52, 1-9.
McLachlan A (1980a) The definition of sand
beaches in relation to exposure: a simple rating
system, S. Afr. J. Sci. 76, 137-138.
McLachlan A (1980b) Intertidal zonation of
macrofauna and stratification of meiofauna on
high energy sandy beaches in the Eastern Cape,
South Africa, Trans. R. Soc. S. Afr. 44, 213-223.
McLachlan A (1980-81) Exposed sandy beaches as
semi-closed ecosystems, Mar. Envir. Res. 4, 59-63.
Orensanz JM and Gianuca NM (1974) Contribui~~o
ao conhecimento dos anel{deos poliquetas do Rio
Grande do SuI, Brasil. I. Lista sistematica
preliminar e descrir~o de tres novas especies,
Comun. Mus. Ci. PUCRGS 4, 1-37.

SELECTIVE HICROHABITAT COLONIZATION BY INTERSTITIAL IIEIOFAUNA AS A FUNCTION OF GRAIN SIZE
A.H. FRICKE and B.W. FLEl1UING (NRIO, Stellenbosch)
1. INTRODUCTION
The trend toward focussing in on ever smaller scale
physical features and biological processes is
becoming increasingly evident in the study of
organism/sediment relationships. In this area of
research, meiobenthologists and sedimentologists
have much to gain from co-ordinated studies.
It is the physical dimension, the shape and
composition of sand grains that constitute the
fundamental abiotic setting for small organisms,in
other words, the micro-habitat. Water circulation,
oxygenation and the ability to trap detritus are all
dependent on the sediment parameters; they
constitute the 'world' in which this species rich
suite of organisms lives. The relative importance
of some physical factors was reviewed by Wieser
(1975), who pointed out that while interstitial
species are generally adapted to a particular
habitat, they nevertheless live close to the
limits of their capability in some respects, while
enjoying a wider 'safe' margin of functioning in
others. Eco-physiological research in this field
is very recent and carried on by a small community
of specialists. Giere (1979) designed apparatus
to test the response of oligochaets to physical
gradients such as salinity, temperature and
humidity in the laboratory. Hockin, Ollason (1981)
used cluster analysis to study colonization of
shallow trays containing sterile sediment in the
field, by means of shallow sediment boxes. These
authors demonstrated a significant difference
between the two approaches. For this reason, we
decided to conduct a natural experiment, thereby
allowing the mUltiple biotic interactions and
421
environmental influences to proceed unimpeded. We
concentrated mainly on nematodes, which constituted
the most diverse (at least 34 species) and
numerically abundant taxon, among the colonizing
meiofauna.
Linley et al. (1981) gave a visual illustration
(S.E.U. and photomicrography) of the small scale
micro-niches presented by individual grains of
sand. These were occupied by a succession of
different bacteria and fungi. It appears that
not only is the degree of rounding or chipping
(micro-cavities) important, but the type of
material constituting the sediment, for example
whether quartz or shell carbonate. Minute
features, such as surface smoothness influence the
so-called 'Aufwuchs' organisms by stimulating or
inhibiting attachment to the substrate. These
together with detritus constitute diet items for
meiofauna.
Flemming, Fricke (c.f. this volume) pointed out
that beaches, as a sedimentary province, display a
complex internal geometry, characterised for
example by mUltiple low angle discordances, such
as are formed in the swash zone of a beach.
Additional internal bedforms further complicate the
setting and may be of sufficient permanence in
terms of meiofaunal lifespans - to allow the
development of specific communities differentiating
from the total community; in other words selective
colonization of specific, very local habitats. For
example a given 10 cm plug of sediment (which is a
popular coring interval) may be traversed by a
sloping band of sand, quite different in texture -
422

LANGEBAAN LAGOON
x i"
o
I I If" I
1
I
2km

mean diameter(phi)

Fig. 1.

and hence organism content - from the main bulk of
sediment. The same band may not pass through a
replicate of a 10 cm plug close by, and may suggest
an unrealistic patchiness of fauna density in the
analysis of the samples. Before any detailed field
sampling strategy can be started, it is therefore
obviously important to become acquainted with the
internal sedimentary geometry. This can be done
efficiently by taking epoxy resin relief casts of
a particular sediment. Such casts give a clear
picture down to grain size level of internal
geometry and sediment texture (see also Bouma,
1969).
The aim of this study was to ascertain whether
these inhomogeneities with respect to grain size in
natural sediments elicit a measurable response in
colonizing meiofauna in the short term. Such a
response may lead to significant differences in
animals (of a particular taxon) and species numbers
in the three grain sizes tested in this experiment.
2. STUDY SITE
The mid-tide level in a west coast lagoon was
selected for our experiment, as the zone of highest
faunal density (McLachlan, 19S1). The alternative,
namely a study site on an open beach, although more
typical, was not realized, because of the disturbing
factors in such a situation, for example beach
slope instability. The study sites were located in
the southern part of Langebaan lagoon (33 0 9'S -
lSo03'E), one on a sand bank a few hundred metres
off-shore and the other on the south-eastern flank
near Stofbergfontein (Fig. 1). The lagoon is
virtually wholly marine, through the absence of
any significant fresh water inflow (low rainfall
and small catchment area). There are no muds,
except in the salt marsh region at the southern tip
near Geelbek. Tidal height here is much reduced.
The study sites 2 and 3 are densely populated by
sand shrimps (Callianassa kraussi) and hermit crabs
(Diogenes sp.).
 423

3. HETHODS
The experimental containers were core boxes
(300 x 150 x 125 mm) made of stainless steel
(AISI 316) constructed as a pair of fitting trays
with handles (Fig. 2a) held together in a frame.
A close fit is essential to avoid loss of sediment.
Three grain size fractions were used. These were
25mm
phi interval apart i.e. 1-0,5 rum, 0,5-0,25 rum and
do.
0,25-0,125 mm sand wet sieved. The sand was acid
washed, remaining organics removed with hydrogen
peroxide (20 volume strength) and finally auto-
claved. The boxes were charged with alternating
layers, each 25 mm thick, of fine, medium and 150m m
coarse sand (Fig. 2b). Each box contained four Grain size: fine = 125LJ diam.13 <1»
medium=250"" 12")
such series, starting with fine sand. Before coarse 500" 11")
exposing the sediment, it was conditioned in sea
water by immersion of the entire box. After
placing these boxes into the lagoon sediment, Fig. 2. Experimental core box.
flush with the surface and parallel to the water A - assembled, B - deployed
line, one metre apart, the surrounding sediment
was allowed to settle firmly against the boxes.
The trays were then drawn out and the frame left sand around the periphery which might have been
in situ with the sand column freely exposed all disturbed by handling.
round. The sites were marked by buried nylon
lines leading to the experiment, in addition to Extraction of the large number of samples (156) was
surveyed land beacons. Experiments one and two achieved by means of the method of Oostenbrink
lasted for six weeks, while experiment three (Fricke, 1979). The fauna was then lightly
entailed exposure of seven core boxes, individually stained in rose bengal and counted under a
for doubling periods of time, i.e. 1, 2, 4, 8, dissecting microscope. Further identification
16, 32 and 122 days (the last one by error!) was done by mounting all the nematodes from
representative samples of the three size grades in
Rapid fixation was important in this experiment anhydrous glycerine and examination under a
since we had to avoid positional changes of the compound microscope. A total of 650 nematodes
contained organisms at all costs. The exposed were evaluated in this manner.
boxes were allowed to drain for a short while,
after which they were immersed in 4% formaldehyde The oxygenation status of the lagoon sediment near
made up in fresh water. the study site (Sites 2 & 3) was obtained by
measuring redox potentials of several vertical
This method (Riemann, 1979) (see also Schiller, cores of natural sediment at L.W.S. and neap tide
1930) preserves nematodes very well, due to a (Table 3). The measurements were made at 25 mm
solvent action of formaldehyde in the absence depth intervals (the thickness of the individual
of salts. The cores were then split into their layers) by inserting the probes close together
original size layers, after removal of some 50 mm into the sediment.
424

Representative splits were taken of all layers of
the cores, dried, powdered and combusted in a Carlo
Erba Elemental Analyser (Uod. 1106 fitted with
integrator). The sediment carbon values were well
above the minimum detection limit of 7 ~g,
did not need to filter off the organic fraction,
but used the whole sediment, despite the high inert
(quartz) to organic ratio.
4. RESULTS AND DISCUSSION
4.1. Experiment 1
The time chosen for the first experiment was
arbitrary, because we had no idea of how long it
would take interstitial organisms to invade a
totally sterile sediment. We wanted sufficient
numbers of nematodes without blurring of effects
by excessive bioturbation. The results of nematode
counts obtained in the first experiment are shown
Nematodes were much more abundant (Fig. 4) and
were found to prefer the medium-fine layers. The
coarse fraction was noticeably less well
colonized. The counts for the control, while
thus we resembling those found in the background (survey)
as done near Churchhaven (Fig. 6), are never-
theless puzzling, and typical of sediments
composed of a mixture of grain sizes. Flemming
and Fricke (cf. this volume) showed in a
theoretical model of progressive grain size
mixtures, that by mixing differing portions of
70 r- Nematodes
r-
- r- Con.

~
50 r C 01.1
in Figure 3. The error bars represent 95% f
.; Col.2
r-
confidence limits for the means of the fine, medium
't:I
>

c:
- l- f
f
and coarse sets of layers, each of which was
f
calculated from two replicates. The third core '0.30 ~
0 f
served as control and contained the sand of the c: m
environment (ca 125 median diamter). To make m
~m

comparison easier, the control was subdivided

~
-
into three sections equivalent in position (depth c
10 ~
in the core) to the layered cores. Nematodes c
showed a significant (XL, P < 0,05) for the fine
fraction. The discrepancy in nematode totals sediment
(whole core) between the control (8 725) and the
layered cores (6 098 and 2 855) emphasises this Fig. 3. Experiment 1.
Nematode densities found in three
point.
sedrment grades.

Harpacticoids were present in very low numbers at
this site, hence no data are shown. The same
applied to other taxa including oligochaets,
turbellarians, archiannelids and acarines, which
were equally sparse, a phenomenon which may be
seasonal (winter).
4.2. EJ.!;periment 2
A very different picture was obtained in the
second experiment (sand median diameter ca 300
coarse and fine sand, a seemingly optimal grain
size mean for a given taxon may be obtained
(eg. 284 ~m) for nematodes as shown by Hennig
et al. (1981), but lacking the actual animal
den~ities expected for such a grain size category.
Stated differently, the sediment suggested by
this mean may in fact be absent, being instead
represented by a coarser and finer fraction.
Previous knowledge of small-scale texture before
~m). sampling is therefore important for unambiguous
 425

results. The standard procedure of coring to 300nnn

Nematodes
in three 100 nnn sections may therefore be 70 r-
questionable in terms of doing justice to inhomo-
geneities on the grain size scale. I-

It is not innnediately apparent, why the layered 50 >-

m
sediments harbour many more nematodes than the
uniform core. It is possible that a uniform core . > -
f

e
may be more easily clogged by percolating detritus '0 m
c

mt~
and micro-organisms. A blocking of interstitial :: 30 l- e
0
spaces may lead to an oxygenation problem. That
0
this actually does take place periodically, was c
~
- m
m
shown by two sets of Redox measurements (Table 3)
which are discussed below. 10 -
Figure 5 shows the results obtained for harpacti-
Col.1 Col.2 Con.
coids in the second experiment. In contrast to the
first, these animals are now numerous (autumn?) and
Fig. 4. Experiment 2
show a clear preference for the coarse sediment Nematode densities found in three
sediment grades. Surrounding sand
fraction, except in the third core (Col. 3) A medium.
reason for this may be that this core suffered some
bioturbation by sand shrimps and the separation of
Harpacticoids in 3 sed. grades
the original size fractions was thus less accurate. 30
Furthermore, a large, possibly spurious value
(dotted line) distorts the total for the fine ;20
c
T
fraction of Col. 3. Eliminating this value brings
the histogrannne more into agreement with Col.l & 2 ~10 m c m c
0 m c m
c
(Fig. 5). The single column-on the right hand
~ 0
side (Con.) shows harpacticoid numbers in a core Col.1 Col.2 Col.3 Con.
packed with the medium sand of the environment.
Fig. 5. Experiment 2
It seems that colonisation of homogeneous sediments Harpacticoid densities found in three
differs from stratified ones on the millimeter sediment grades. Surrounding sand medium.
size scale. Harpacticoid numbers in the control our sediments had different cut offs and were very
(Con.), a medium sediment, reflect densities in well sorted by artificial means.
the field during the study period. The opposing
trend of nematodes and harpacticoids emerges The large standard deviations are a well known
clearly. McLachlan et al. (1981) found the same property of meiofauna distribution. It was
reverse abundance pattern between nematodes and impossible to reduce this by scaling up the
harpacticoids in his field data from the South experiment (more replications, larger sand volumes)
African coast. He suggests that sands with a f6r practical reasons, However, the trends are
median diameter of 1,58 phi (334 ~m) contain nevertheless consistent. The p~rpose of the
equal proportions of the two taxa (38%). We could controls was to show nematode or harpacticoid
not apply this finding to our situation, because densities at equivalent levels with respect to
426

110 Churchhaven

100

iii 80
>
"C N
0::
H
60
'0
c:i
0::

~ 40

20

Fig. 6. Experiment 2.
Summary of nematode and harpacticoid densities compared with
background numbers (Churchhaven).

the test cores, namely in a sediment natural to the DeEth of sediment level medilUTI
~ ~
environment.
25 mm
50 2 2
Reference to Table 1 clarifies this point. I t can 75 3 3
be argued, that the depth effect, i.e. the drop 100 4 4
125 5 5
off in animal numbers vertically, interferes with
150 6 6
the preference behaviour we wanted to test. This
175 7 7
is theoretically correct. The effect is consistent 200 8 8
however, since the relationship of the different 225 9 9
grain size sets remains constant towards one 250 10 10
275 11 11
another; furthermore, we found the depth effect to
300 12 12
be .insignificant within a given grain size cycle.
Previous work by us on sandy beaches showed that
Position of experimental grain size sediments in core boxes
the 10 cm division of sample cores (to a total
depth of 30 cm) was rather coarse in terms of
revealing the depth effect but our subdivisions Table 1.
were on a smaller scale.

fraction is very convincing. The top layer shows

Table 2 lists the counts and basic statistics for very high numbers because we have the greatest
the second experiment, which gave very consistent mobility of both fauna and detrital material here.
results. The data are shown graphically in Fig.? It is the one level which is exposed to somewhat
The preference of nematodes for the fine sediment different influences than the ones below. Further-
 427

core boxes in a time study is shown in Figs. 8 & 9.

ST.D.
Nematod es invaded our first core box in appreci able
Core 1 Core 2 Mean
Sediment type

numbers within two tidal cycles, i.e. 24 hours.

657 2761 1709 1487
fine (Top)
1157 775 966 270
medium
235 465 350 162
The similar ity of the slopes of the three graphs
(Fig. 9) indicat e similar colonis ation rates of
641 562 111
fine 484
208 558 383 247
medium
coarse 128 184 156 39 the three sedimen ts up to day sixteen , when
fine 593 450 521 101
immigr ation increas es sharply for the fine and
211 50 130 113
coarse sand. This trend continu ed till the
medium
108 38 73 49

fine 131 119 125 fourth month (not shown).

42 57 22
medium 73
26 15 20
coarse
Sed, Nemat odes/ 100 m!

I~
Table 2.
81illtill
Nematode counts normalised to lOOml found in stratified
sediment cores deployed
1;:;:;:;:;:;:;:;:;::1
......
were exposed to coloniza-
in an environmen t of "medium size sediment. The cores
tion for a period of six weeks during autunm. I···········j
!::::::::::::::::::::::~:::1

I::;:;:;:;:;:;:;:;:;j
~
more, sedimen t from the surroun di,ngs was invaria b ly
.....
I',•••••••••••~ ~

Tt,
found to have contam inated the top l~yer. The Key:-
inheren t patchin ess of the fauna is seen in the (::::::::::::::::::::::::::J Dill 3 phi
large differe nces between the two cores amounti ng ~:;:;:;:;:;:;:;:;:;:I ~ 2"
to more than 100% in some instanc es. 1 "
core1
The results obtaine d for the two taxa tested, i,e. I::::::::::::::::::::::::::J 2
nematod es and harpac ticoids are, summar ised in Fig. 6
1:;:;:;:;:;:;:;:;:;:1 I f--i 1 m eon + s.d.
togethe r with backgro und data from the area for
compar ison. Figures of a typical COarse sedimen t I ~, I I I I I
I I
beach are also given (Arnist on). Deploym ent for a 30 20 10 0 10 20 30 40 50
V-no. of indiv.
period of six weeks clearly does not allow animal
numbers to reach their maximum density typical for
Fig. 7. Experim ent 2
the surroun ding environ ment. Nematod es amounte d to Nematod e densiti es in two experim ental
somewh at more than half, harpac ticoids to about one cores showing respons e to grain size
and depth.
third of the figure found for the equival ent medium
sedimen t. The reverse prefere nce with regard to
grain size lS seen. We added Arnisto n beach as an Almost invaria bly, animal numbers were found to be

example of an impove rished biotope due to its
coarsen ess (X; 0,15 ~) related to exposur e. Nema-
todes are sparse, harpac ticoids relativ ely more
abundan t, but both taxa very much sparser than the
lagoon referen ce site (Church haven).
highest on the top, i.e. nearest the surface . This
can be explain ed by driftin g, which is a reporte d
means of meiofau na dispers al (Bell ,Herman , 1980;
Palmer, Brandt, 1981). This was not always so
howeve r. Burrowi ng and inters titial wriggli ng are
mechani sms by which vertica l penetra tion is

4.3 Experim ent 3 achieve d by differe nt species . In the first two

cores there was no evidenc e of bioturb ation by
The progres sive coloniz ation of six stratif ied core
 428

sand shrimps or polych aetes, i.e. mechan ical means

which could have carried numbers of nematod es nema tode s
50
deeper into the sedimen t. We found both burrowe rs
and inters titial wriggle rs among nematod es and
harpac ticoids in our microsc ope prepara tions.
The rapidit y with which these slow moving
organis ms occupie d an empty substra te was
surpris ing. What is the reason for this rapid 30
respons e? If the attract ion was food, this should
til
be reflect ed in the organic carbon content of the >
differe nt layers. We determi ned percent age carbon "0

-
s::::
for all samples i.e. all cores (Fig.10 ). The
values instead of rising to some maximum value, 0
10
fluctua ted widely until all three grain size layers 0
s::::
were virtual ly equal after four months. Through an ~
error the seventh core box was left in the field
much longer than the rest. Thus the value for the 1 2 4 8 16 32
Days
fine fractio n had dropped , while the coarse
Fig. 8. Coloniz ation rate of sedimen ts
fractio n had built up carbon from very low initial in core boxes 1-6 in time.
levels.

Further analysi s of this data suggest that carbon Nemat odes

enrichm ent of sand in this localit y occurs down- 30
ward in pulses. While this is occurri ng the carbgn
is being utilise d by both meio~ and microfa una. 20
The 'four-m onth' core may represe nt an equilib rium,
in that reduced inters titial circula tion due to
10
blocked pore space, stops further enrichm ent.
20 m
Reduced drainag e has been correla ted with meio~
faunal abundan ces in shelter ed localit ies (Dean,
1981). Clearly , the fine layers in our stratif ied
cores would become blocked more readily than the
coarser ones. After a period of four months the .....
o
entry of organic substan ces into the sedimen t takes
o
place more easily via numerou s sand shrimp burrows c20 c
and polycha ete holes than through the body of the >
sand. We counted over 100 Callian assa burrows 10
within 1 m2 , i.e. 2-3 burrows per core surface
area. The nematod e and harpac ticoid count for
the four-mo nth core was 14 971 and 1 160 respect - Days 1 24 8163 2
ively, the highest counts found, while the organic
Fig. 9. Coloniz ation rate of sedimen ts in
carbon content of this core (exclud ing meiofau na core boxes 1-6 in time, plotted
biomass ) was on the lower side compare d to the rest. separat ely for three grain sizes.
 429

It seems then that carbon content (food) does not

Effect of tidal state upon Redox Potential in lagoon sediment (corrected readings),
have a simple relationship with meiofauna content.
and normalised counts of background meiofauna (per lOOmI) on the same site.

Salinity and temperature were not considered

- 25 nun + 228 mV 1 824 340 20 56 12
relevant variable in the context of this study and
- 50 + 209 684 72 21 23
were therefore not measured.
- 75 + 188 + 12 660 44 50 13

-100 + 208 + 42 895 92 77 102

Table 3 lists a set of redox measurements done at -125 + 198 + 62 866 64 26 124

L.W.S. and during neap tide together with fauna -150 + 210 + 62 738 69 67 20 44

counts of a control core taken nearby. I t is -175 +220 ~ 713 30 13 11 44

-200 + 189 - 78 638 77 26 14 29
interesting to note that the sand remains
-225 + 198 - 18 518 64 14 52
reasonably oxygenated during a period of spring
-250 + 168 - 18 536 27 20 27
tides, when the water level undergoes maximum -275 + 242 - 98
vertical fluctuations. The sediment is alter- -300 + 226 - 18

natively flooded and drained and then draws in air,

which escapes through a maze of holes, similar to Table 3.
those due to the sand shrimp. During neap tides the_
o

F=
sed. 5 10 15%tot.Carbon
study site remains water saturated and redox
potentials drop steeply. The sediment in fact
'::::::::::::::::::, +- Xper core
becomes oxidizing-anaerobic (Eh between + 300 and
-200 m V). (See also Billen et al. 1974). Since
.:.:.:.:.:.:.

~
this fluctuation is a transient effect, meiofauna
2 ::::::::::::::::::.
may adjust to it by migration. The redox status
............
,
eo- ._._._.~

of a sediment may be important in the design of

~
sampling strategy.
4
Experiment 3 served the study of meiofauna

b~"~
colonization of a sterile sediment in time (Fig.S).
Through an oversight, replicates were not taken. 8
This meant that for statistical analysis we had to

~
obtain the error factor from corresponding grain
size sets, which functioned as replicates.
added the 'depth' effect to the natural inherent
This
.......
.....
. e e.e __ • •

variability of the material. It was not certain

that the populations were either normally
distributed or the variances equal. For this
reason, we used Friedman's test, a non-parametric
equivalent of the 2-way Anova where indicated.
A significant numerical increase was found for
nematodes in the six staged (results for core 7 not days deployed
shown) cores (X 2 , P < 0,05). Grain size preference Fig. 10. Change of total organic carbon of
by nematodes only just failed to emerge as core box sediments during staged
experiment. Meiofauna excluded.
significant (X2, P < 0,1 > 0,05) possible as a
result of our 'substitute' replicates.
430
Harpacticoid copepods showed a significant
preference for medium to coarse sand (X 2 ,P < 0,05)
supporting evidence shown earlier.
The overall drop of animal numbers with increasing
depth was significant, i.e. there was a significant
difference (X 2 , P<0,05) between the layers of
sets 1-3, 4-6, 5-9 and 10-12, which represents
the difference between the fine and the coarse
layers in each case. To test whether there were
specific grain size preferences by different
species of nematodes, we first compared the
spectrum of forms found in the natural sediment
(Md. 2 ~) with the nematodes found in the experi-
mental cores taken as a whole. All the species
found in the natural sediment were found in our
cores; no significant difference could be
detected (X2, P>O,I). It seems therefore that
there is no avoidance of the artificial sand
columns.
We then tested the abundances of nematode species
in the three size grades. A statistically
significant (X 2 , P < 0,05) difference could be
shown for the following eight species (thirty-
three identified): Desmodora, Sabatiera,
Paramonhystera, Oncholaimellus, Spirininae
(Chromaspirina?), Bathylaimus, Trissonchulus
janetae, Nudora (Monoposthia?).
For example Trissonchulus janetae Inglis 1961
(ironidae), Nudora (Monoposthia?) Sphaerolaimus
and Anoplostoma show a strong preference for fine
sediment, while Sabatiera (Comesomatidae) prefers
medium and coarse sand. Warwick reports (pers.
comm.) this form from muddy off-shore sediments.
The complicated interactions ego ratio of feeding
types according to Wieser (1954) between these
organisms obviously warrant much fuller treatment
of the presence-absence data than was possible
in the present context.
5. CONCLUSION
This field study has shown the very important role
played by a purely dimensional parameter, namely
particle size, in the behaviour of the meiofauna
of sandy environments. We feel that grain size
is indeed the cardinal factor upon which virtually
all the other biotic and abiotic factors are
dependent in varying degrees. The study requires
further refinement to answer the many questions
left open at this stage, for example the relation-
ship between carbon flux through the sediment in
time and its effect on animal abundance and
composition. Clearly accurate prior assessment
of sedimentary structure is a prerequisite before
commencing any detailed investigation of small
scale distribution. It may not always be
practical, indeed necessary to allow for the
multitude of fabric variations found in natural
situations, which are subjected to constant
change, but a knowledge of the complexity of a
particular sediment body goes a long way in
assisting the interpretation of data.
6. ACKNOI.JLEDGEMENTS
The authors owe much gratitude to Mrs. C. Martin
and lfiss Susan Brundrit for painstaking effort
spent in processing the material.
The taxonomic expertise of Dr. Sievert Lorenzen
(University of Kiel) and Dr. Franz Riemann
(Institut fur Meeresforschung, Bremerhaven) was
invaluable in the identification and confirmation
of species names. Thanks also to Dr. Richard
Warwick (of the Harine Biological Association
in Plymouth) for valuable discussion pertaining
to this study.
 431

REFERENCES
Bell S and Herman ~1S (1980) A field
investigation of meiofaunal dispersal tidal
resuspension and implications, Har. Ecol. Prog.
Ser. 3, 245~249.
Billen, G et al. (1974) Bacterial methyl
mercury~ineralizing activity in river sediments,
Water Res. 4 part 8, 2l9~225.
Bouma AH (1969) Hethods for the study of
sedimentary structures, Wiley~lnterscience,
New York 458 pp.
Flint RW et al. (1982) Benthos investigations
Sediment boxes on natural bottom, Bull. Environ.
Contam. Toxicol. 28, 257~265.
Fricke AH (1979) l1eiofauna extraction
efficiency by a modified Oostenbrink apparatus,
Helgolander wiss. Heeresunters, 32, 436~443.
Giere 0 (1979) Some apparatus for preference
experiments with meiofauna, J. expo mar. BioI.
Ecol. 41, 125-131.
Hennig HF-KO et al. (1982) ReLationships
between meiofaunal population densities and
physico-chemical properties of unpolluted
beaches, Environ. Honit. & Assessment 1, 337-344.
Hockin DC (1981) An apparatus to study the
colonisation of sediments by the meiofauna, Est.
Cstl. Shelf Sci. 12, 119-120.
Hulings NC and Gray J (1971) A Hanual for
the study of meiofauna, Smithsonian Contributions
to Zoology 78, 1-83.
Linley EAS et al. (1981) Heterotrophic
utilisation of mucilage released during fragmen-
tation of kelp (Ecklonia maxima mmLaminaria
pallida.) I. Development of microbial
communities associated with the degradation of
kelp mucilage, Mar. Ecol. Prog. Ser. 4, 3l~4l.
McLachlan A et al. (1979) Vertical gradients
in the fauna and oxidation of two exposed sandy
beaches, S.A. J. of Zool. 14, 43-47.
McLachlan A (1981) The ecology of sandy
beaches in Southern Africa, S. Afr. J. of Zool.
16(4), 2l9-23l.
Palmer H and Brandt RR (1981) Tidal
variations in sedimentary densities of marine
benthic copepods, Mar. Ecol. Prog. Ser. 4, 207-
212.
Riemann .. F (1979) Nematoden aus dem Brackwasser
des Weser-Astuars and Beschreibung von drei
Monhysteroidea.Veroff. lnst. Meeresforsch,
Bremerh. 17, 213-223.
Schiller W (1930) Gewebsfixierung unter
Erhaltung der basischen Kernfarbung,
Z. Zellforsch. mikroskop, Anat. 11, 63-178.
Wieser W (1953) Die Beziehung zwischen
Mundhohlengestalt, Ernahrungsweise und Vorkommen
bei freilebenden marinen Nematoden, Arkiv for
Zoologi Serie Band 4 NR 26, 439-484.
Wieser W (1975) The Heiofauna as a tool in
the study of habitat heterogeneity, Ecophysio-
logical aspects. A Review (1). Cah. BioI.
Mar. 16, 647-670.

TWO GRAPHICAL DISPLAY METHODS FOR ECOLOGICAL DATA MATRICES
L.G. UNDERHILL (Department of Mathematical Statistics, University of Cape Town)
There has been a growing interest in graphical
display techniques for various kinds of multi-
variate data matrices over the past 25 years.
The most well known of these include multidimen-
sional scaling (Kruskal 1964a,b), classical
scaling (Torgerson 1958), cluster analysis
(Sneath 1957) and Chernoff's faces (Chernoff
1973). Recent reviews of scaling (ordination)
and cluster analysis are Greenacre and Underhill
(1982) and Hawkins et al (1982) respectively.
In this paper, we discuss two graphical display
techniques: correspondence analysis, developed
during the 1960's by a French school of data
analysts, led by J-P Benzecri, and not well
understood by the English speaking world
(Benzeri, J.P. 1973), and a technique, developed
fairly recently and not yet widely known, for
scaling a skew-symmetric matrix (Gower 1977,
Constantine and Gower 1978).
Although these two methods are very different,
they both have as the central feature of their
computational procedures the calculation of the
singular value decomposition of a transformation
of the original data matrix. They thus join a
wide variety of seemingly disparate graphical
display techniques which can readily be incor-
porated into a single computer program (Green-
acre and Underhill 1982, Underhill 1983). These
methods include principal components, canonical
correlation, discriminant analysis, and the re-
latively new family of techniques known as
biplots (Gabriel 1971, 1972, 1981, Bradu and
433
and Gabriel 1978).
CORRESPONDENCE ANALYSIS
Correspondence analysis was originally designed
for contingency table analysis. With care
however, its use may be extended to data mat-
rices which have all their elements positive
(Greenacre 1977, p110). The 15 6 matrix of
densities (birds/km) of 6 species of waders at
15 coastal localities (Table 1) is typical of
the type of ecological data matrix that may
profitably be analysed by correspondence
analysis.
The algorithm details are available elsewhere
(Greenacre 1977, and Greenacre and Underhill
1982) and will not be discussed here.
The output from correspondence analysis consists
of graphical displays (as in Figure 1), and two
tables - a table of "inertias" on the "axes"
(Table 2) and a table of "relative" and "abso-
lute" contributions of each species(row) and of
each locality (column) to the inertia on each
axis (Table 3). The concept "inertia" which
plays a key role in correspondence analysis,
is a generalization of the concept of variance.
In the interpretation of Table 2, the chief
point of interest is the percentage of the
total inertia explained by each successive axis.
Thus, in Table 2, the first and second axes
account for 46,8% and 29,9% of the inertia re-
spectively, and thus the plane of the first
434

pair of axes accounts for 76,7% of the inertia.

TABLE 1. Wader densities (birds/km) at 15 sections of shoreline in St Helena Bay
(Underhill and Cooper 1982), and shoreline classification
Locality Black Oystercatcher Turnstone White Grey Plover Curlew Sanderling
number and Haemo.topM Alte.ncvUa. fronted Pluv.ia..W Sandpiper CaLldJrM
classification moqtUJU in.teJl.PJtu Plover I.l qua..ta.Jtola CaLldJrM alba
Cha.Jta.dJUUI.l 6e.JtJtUgine.a
ma.Jtgina..tUl.l
S036 Sandy 1,1 0,0 2,9 0,0 0,0 5,6
S041 Mixed 2,1 23,8 12,6 10,9 98,5 4,1
S043 Rocky 4,0 22,8 1,2 5,1 7,4 1,6
S044 Sandy 2,2 67,4 9,1 2,6 8,7 109,6
S046 Mixed 4,7 114,0 10,0 1,3 36,7 28,0
S047 Sandy 1,2 3,9 5,8 0,6 42,1 70,0
S048 Mixed 11 ,8 59,4 10,0 12,4 12,4 11,2
S049 Mixed 3,0 184,5 12,0 7,5 22,0 99,0
S050 Sandy 1,2 0,0 7,9 0,5 0,0 5,8
S052 Sandy 2,0 0,0 4,0 0,0 0,0 5,0
S053 Mixed 5,7 93,1 17,7 1,7 109,7 4,0
SO 58 Rocky 0,4 12,2 3,3 0,2 0,2 1,3
S059 Rocky 2,9 51,8 4,9 0,5 26,9 5,3
S060 Rocky 2,1 4,1 0,6 0,0 0,3 0,0
S061 Rocky 8,3 38,3 13,3 5,0 3,3 10,0

The coordinates for the graphical display in
this plane (Figure 1) are obtained from Table 3,
columns 4 and 7. Notice that the 15 localities
and 6 species are all plotted on the same pair
of axes. This is justified by means of the
"transition formulae" (Greenacre 1977, Greenacre
and Underhill 1982). The transition formulae,
translated into words, state that the localities
get drawn away from the origin in the direction
of those species for which they have relatively
high densities. Thus examining Figure 1, we
see that Sanderling is the characteristic
species of sandy localities, Black Oystercatcher,
Turnstone and Grey Plover are characteristic of
rocky and most of the mixed localities, whilst
Curlew Sandpiper is the characteristic species
of certain mixed localities.
The remaining columns of Table 3 are interpreted
as follows. The mass (column 1) is the weight-
ing given in the analysis to each locality and
species (and is here proportional to the total
density). It is often desirable to reweight
the rows and/or column masses so that they all
have equal weight.
The quality (column 2) is a measure (based on
a geometrical concept) of how well the locality
or species is fitted in the first two axes.
Thus locality S044 is well fitted (1,00) but
locality S050 is very poorly fitted (0,06).
This depth of information is unavailable in
conventional non metric multidimensional scaling
analyses.
The inertia (column 3) refers to the proportion
of the total inertia attributable to each
locality (adds to one) and the proportion attri-
butable to each species (also adds to one).
Thus localities S036 and SO 41 contribute 2%
 435

TABLE 2. The inertias on the axes

Axis Squared basic value Percentage of total Cumulative
inertia percentage
0,278 46,8 46,8
2 0,177 29,9 76,7
3 0,091 15,3 92,0
4 0,035 5,8 97,8
5 0,013 2,6 100,0
l: 0,593

and 18% of the inertia, whilst Black Oyster-
catcher and Turnstone contribute 8% and 14%
respectively. Those rows and columns which have
the largest inertias are the ones which most in-
fluence the analysis.
These contributions to inertia are further
broken down into relative and absolute contri-
butions for each locality and species on each
axis (columns 5, 6, 8 and 9). In brief, these
localities and species which account for a
large proportion of the relative and/or absolute

FIGURE 1. Correspondence analysis display of the data of Table 1.

axis 2 (29,9%)

0 S047

Curlew Sandpiper. @
·Sanderling S041

~36
S044

@S053
®S0520 S050
axis 1 (46,8%)

"Whitefronted Plover
0 8059
@8049
~8046
Turnstone· .Grey Plover

8048@ 0 8043
0
S058 ·B1ack Oyster catcher
® sandy
0 S061
@ mixed

(0 rocky
0 8060
436

TABLE 3. The relative and absolute contributions to the inertia

Axis 1 Axis 2
Name Mass Qual ity Inertia Coord. Rel. Abs. Coord. Rel. Abs.
A The rows (shore section)
S036 0,01 0,35 0,02 -0,84 0.31 0,02 0,30 0,04 0,00
S041 0,09 0,95 0,18 0,91 0,71 0,28 0,53 0,24 0,15
S043 0,03 0,58 0,03 0,28 0,13 0,01 -0,52 0,44 0,04
S044 0,12 1,00 0,15 -0,80 0,91 0,29 0,24 0,09 0,04
S046 0,12 0,44 0,03 0,03 0.01 0,00 -0,23 0,43 0,04
S047 0,08 0,97 0,13 -0,42 0,17 0,05 0,90 0,80 0,35
S048 0,07 0,60 0,06 0,10 0,02 0,00 -0,53 0,58 0,12
S049 0,20 0,76 0,08 -0,38 0,62 0,11 -0,18 0,14 0,04
S050 0,01 0,06 0,06 -0,47 0,06 0,01 0,06 0,00 0,00
S052 0,01 0,13 0,03 -0,60 0,13 0,01 0,04 0,00 0,00
S053 0,14 0,91 0,11 0,64 0,87 0,21 0,14 0,04 0,02
S058 0,01 0,63 0,13 -0,08 0,01 0,00 -0,65 0,62 0,03
S059 0,06 0,64 0,20 0,32 0,48 0,02 -0,18 0,16 0,01
S060 0,01 0,39 0,02 0,15 0,01 0,00 -0,99 0,38 0,02
S061 0,04 0,69 0,07 0,17 0,03 0,01 -0,78 0,66 0,15
B The columns (species)
Black Oys-
tercatther- 0,03 0,35 0,08 0,16 0,02 0,00 -0,69 0,33 0,09
Turnstone 0,42 0,78 0,14 -0,02 0,00 0,00 -0,38 0,76 0,35
White fron-
ted Plover 0,07 0,03 0,10 0,09 0,01 0,00 -0,13 0,02 0,01
Grey Plover 0,03 0,21 0,07 0,37 0,10 0,02 -0,38 0,11 0,02
Curlew Sand-
piper 0,23 1,00 0,30 0,72 0,68 0,43 0,50 0,32 0,32
Sanderling 0,22 1,00 0,32 -0,84 0,80 0,55 0,42 0,20 0,21

contribution to the inertia of the axis are im-
portant in determining its position, and there-
fore in its interpretation. For example, ex-
amination of the relative and absolute contri-
butions on the first axis (columns 5 and 6) and
the signs of the coordinates (column 4) reveals
that the position of the first axis is princi-
pally a contrast between relatively high densi-
ties of Sanderling at localities S036, S044 and
S049 and relatively high densities.of Curlew
Sandpiper at localities S041, S053 and S059.
The second axis contrasts localities S043 and
S048, S058, S060 and S061 with high Turnstone
and Black Oystercatcher densities with locali-
ties S041 and·S047 which have high densities of
both Curlew Sandpiper and Sanderling. A more
detailed discussion of absolute and relative
contributions to inertia and their interpre-
tation may be found in Greenacre (1977).
THE GRAPHICAL ANALYSIS OF SKEW-SYMMETRY
Skew-symmetric matrices have the property that
the element in row i column j is equal to the
element in row j column i, but has opposite
sign; in symbols, t ij = -t ji or Tt =-T.
Skew-symmetric matrices occur in ecology as
follows. Let the nxn matrix R = (r ij ) be
such that r ij is the number of individuals

which migrate from zone i to zone j in a given
time period, or alternatively let r.·
lJ
be the
number of times behaviour pattern i is followed
by behavi our pattern j. In both cases,
r ij f r ji , and R is asymmetric. However, R
may be represented as R = S+T where S is
symmetric and T is skew-symmetric. (Put
S = ~(R+Rt) and T = ~(R-Rt).) S contains the
symmetric information in R and may, if appro-
priate, be handled by conventional multidimen-
sional scaling methods. T, which contains the
skew-symmetric information in R, is of parti-
cular interest, focussing attention on the
difference between r ij and r 1i - in the con-
text of the above examples, the researcher
would be interested in the difference in the
direction of migration between zones, or in the
sequential arrangement of behaviour patterns.
The singular value decomposition of the skew-
symmetric matrix T is now computed. A property
of skew-symmetric values is that their singular
values occur in equal pairs. The graphical dis-
play method presented here depends on the first
pair of singular values being substantially
larger than the rest. If this is the case, use
the first two columns of the matrix of singular
vectors to give coordinates for each object in
two dimensions. It is quite easy to prove
(Gower 1977, Constantine and Gower 1978) that
the skew-symmetry t ij is then approximated by
the area of the triangle formed by points i,j
and the origin. The area is taken to be posi-
tive if 0ij goes around the triangle clockwise,
and negative if it goes around anticlockwise.
This is the manner in which the skew-symmetry
is accounted for.
We have the following guidelines for the inter-
pretation of this unusual space (Gower 1978).
1. Consider the jth column of T and draw a
line through object j and the origin. All
objects i which have t ij positive ought to
437
lie in one of the half-planes defined by the
line, and all objects i having t ij negative
ought to lie in the opposite half-plane.
2. As too increases, object i should lie
lJ
at an increasing perpendicular distance from
the line through j and the origin.
3. If t·· = t k · then the areas of the assoc-
1J J
iated triangles in the display should be
equal. Since the triangles have the same
base, their heights must be equal, implying
that a line through points i and k must lie
parallel to the line through j and the origin;
As an example suppose the data of Table 4 re-
present the numbers of animals observed to
migrate in between eight zones (labelled A to
H). Call this matrix R. Then T = ~(R-Rt)
is a skew-symmetric matrix containing the in-
formation on the differences in the direction
of migration between zones (Table 5).
The largest skew-symmetries (Table 5) are be-
tween zones Hand G, and Hand F, the corres-
ponding triangles (Figure 2) HGO and HFO have
the largest areas. Note, too, that A has small
skew-symmetries with F and G, and that the
areas of the triangles AFO and AGO are relati-
vely small. Drawing a line through H and the
origin, check that the points below the line
(A, B, C and D) have negative skew-symmetries
with H, and that G and F are above the line and
have positive skew-symmetries, whilst E which
has a small negative skew-symmetry (-0,5) is
slightly misplaced; it ought to lie just below
the line. Our interpretation of Figure 2 is
that the overall pattern of migration is
clearly away from zones A, B, C and 0 towards
zones E and H, and away from these two zones to
zones G and F.
438

TABLE 4. Hypothetical migration of animals from row zone to the column zone

Zone A B C 0 E F G H
A 107 48 10 3 1 1 6 39
B 35 90 48 12 6 4 20 27
C 7 43 83 25 10 8 36 15
0 3 9 18 95 39 21 33 14
E 0 8 9 23 110 48 20 5
F 0 3 7 11 25 130 41 12
G 5 16 27 16 13 55 91 21
H 25 12 13 7 4 36 53 73

TABLE 5. The skew-symmetric information contained in Table 4

Zone A B C 0 E F G H
A 0 6,5 1 ,5 0 0,5 0,5 0,5 7,0
B -6,5 0 2,5 1 ,5 - 1 ,0 0,5 2,0 7,5
C -1 ,5 -2,5 0 3,5 0,5 0,5 4,5 1 ,0
0 0 -1,5 -3,5 0 8,0 5,0 8,5 3,5
E -0,5 1 ,0 -0,5 -8,0 0 11 ,5 3,5 0,5
F -0.5 -0,5 -0,5 -5,0 -11 ,5 0 -7,0 -12,0
G -0,5 -2,0 -4,5 -8,5 -3,5 7,0 0 -16,0
H -7,0 -7,5 -1,0 -3,5 -0,5 12,0 16,0 0

FIGURE 2. Graphical display of the skew-symmetric information contained in the data

of Tables 4 and 5. 84% of the information is displayed.

@
®

+
®

©
@
@®
 439

CONCLUSION
Graphical display methods may be regarded as a
visual summary that represents, in some sense,
as much of the information contained in the data
as possible. They are invaluable aids to under-
standing the relationships within the data in
all disciplines where researchers grapple with
multivariate data, including ecology.

REFERENCES
Benzeri JP (1973) L'Analyse des Donnees (Tome
2). L'Analyse des Correspondences, Paris, Dunod.
Bradu D and Gabriel KR (1978) The biplot as
a diagnostic tool for models of two-way tables,
Technometrics 20, 47-48.
Chernoff H (1973) Using faces to represent
points in k-dimensional space graphically,
Journal of the American Statistical Association
68, 361-368.
Constantine AF and Gower JC (1978) Graphical
representation of asymmetric matrices, Applied
Statistics 27, 297-304.
Gabriel KR (1971) The biplot-graphical dis-
play of matrices with application to principal
components analysis, Biometrika 58, 453-467.
Gabriel KR (1972) Analysis of meteorological
data by means of canonical decomposition and
biplots, J. Appl. Meteor. 11, 1071-1077.
Gabriel KR (1981) Biplot display of multi-
variate matrices for inspection of data and
diagnosis. In Barnett V, ed. Interpreting
multivariate data, Chichester, Wiley.
Gower JC The analysis of asymmetry and ortho-
gonality. In Barra JR et al, eds. Recent
developments in statistics, pp.l09-123.
Amsterdam, North-Holland.
Greenacre MJ (1978) Some objective methods
of graphical display of a data matrix, Pretoria,
Department of Statistics and Operations Re-
search, University of South Africa.
Greenacre MJ, Underhill LG (1982) Scaling a
data matrix in a low dimensional Euclidean
space. In Hawkins DM, ed. Applied multivariate
analysis, pp.183-268. Cambridge, Cambridge
University Press.
Hawkins DM, Muller MW, Ten Krooden JA (1982)
Cluster analysis. In Hawkins DM, ed. Applied
multivariate analysis, pp.301-356. Cambridge,
Cambridge University Press.
Torgerson WS (1958) Theory and methods of
scaling. New York, Wiley.
Underhill LG (1983) Instructions for the use
of BSD, the basic structure display program.
Cape Town, Department of Mathematical Statistics.
Underhill LG, Cooper J, (1982) Counts of
waterbirds on the shoreline in southern Africa
1978-1981. Cape Town, Western Cape Wader Study
Group and Percy FitzPatrick Institute of
African Ornithology.
 441

ECOLOGICAL CHARACTERISTICS OF SANDY BEACHES IN THE SOUTHERN CALIFORNIA BIGHT

DALE STRAUGHAN" (7015 Marcelle St •• Paramount, Ca. 90723)

INTRODUCTION extended over the eastern North Pacific Ocean as

The Southern California Bight (Fig. 1) is an well as the Southern California Bight.
extremely complex hydrographic and bathymetric
area. The resulting complexity of oceanic circu- These hydrographic changes may be reflected
lation patterns combined with changing habitat, by obvious events such as the intertidal strand-
complicates interpretation of patterns in sandy ing of the Mexican red pelagic crab, Pleuroncodes
beach biota. planipis, as far north as the mainland shores of
the Santa Barbara Channel when the warm Davison
current extended northwards in winter 1973. Less
SANDY BEACH SURVEY SITES
1969 -1980 obvious changes include the flow of discrete
CALIFORNIA
parcels of cold water southwards (e.g. summer
3.'
1977 when water temperatures dropped up to SoC in
a matter of hours (Straughan, Stephenson, 1980»
SOUTHERN CALIFORNIA BIGHT and formation and movement of oceanic eddies as
SAN NICOLAS L
'Ill>
revealed by thermal satellite imagery (Bernstein
t
DUTCH HARIOR et aI, 1977).
33"
PACIFIC
OCEAN
IUlQMITlU

~ The Southern California Bight is composed of

a number of deep basins, near surface ridges,
oceanic canyons, mainland shelf, and island areas.
FIGURE 1. Map of Southern California Bight These restrict circulation patterns (e.g. there is
showing location of sampling site.
only relatively shallow water flow to the south
and east out of the Santa Barbara Channel result-
Four distinct water masses characterize the
ing in some isolation of the Santa Barbara Basin
surface waters of the California Current system
area).
(Reid, et al). These come from the north
(California Current), west, south, and below (due
The offshore islands provide shelter to some
to upwelling). Changes in circulation patterns
coastal areas. Storms in the form of wave action
occur frequently and these changes are reflected
and surge only, or accompanied by wind and rain
in sea surface temperature, salinity, nutrient
approach from the ocean. In general, winter is
concentrations, and sea level. For example, at
characterized by cold rainy storms from the north-
the end of 1976, there was a decrease in upwell-
west which may be augmented by warmer storms
ing and extremely clear water (Cayan, 1980;
accompanied by rain moving in from the west-south-
Mearns, 1980; Tont, Delistraty, 1980). This
west. In summer, there is usually no rain and
442

the storm surge moves in from the south-southwest. revealed over 500 categories of animals which
Areas in the northern and central section of the included identification of over 280 species.
Southern California Bight are generally exposed The larger number of categories is due to diff-
to winter storms while areas in the southern and iculty in obtaining positive identification of
central areas are exposed to summer storms. species of some groups (e.g. insects) and diffi-
culty in identification of damaged specimens.
The coastline is generally eroding and this
is one of the main sources of intertidal beach
The most abundant and widespread species is
sand. Rainfall is generally low (average for
the sand crab, Emerita analoga. Other widespread
Los Angeles is 28 cms/year) so that sediment
crustaceans include the isopod, Excirolana
supply from rivers is not high. Many rivers
chiltoni, and a group of beach hoppers in the
have been dammed or concreted and this further
genus Megalorchestoidea. The second most
reduces this sediment source. The presence of
abundant group of organisms are the polychaetes.
nearshore canyons along the coastline and a pre-
The most common of these are Euzonus mucronata,
dominance of a southerly longshore drift results
~. dillonensis, Hempidous borealis, H.
in the formation of a number of littoral sand
californiensis, Lumbrineris zonata, Nephtys
cells along the coast (Habel, Armstrong, 1978).
californiensis, Scolelepis bullibranchia, ~.
As some sandy beach species are believed to be
squamata. The molluscs, Tivela stultorum, Donax
transported with sand, this could have a semi-
gouldii, and Olivella biplicata, the echinoderm,
isolating force among at least the "sessile"
Dendraster excentricus, are found in low inter-
stages in the life cycle of some species. Hence
tidal areas on some beaches. Sheltered beaches
the Southern California Bight is a very dynamic
provide habitats for crustaceans such as
habitat for the sandy beach biota on an oceanic,
Callianassa spp. and the mollusc Transennella
metereologic, and sedimentary basis.
tantilla.

This is an area of overlap of the Califor-

There are several dominant beach types. One
nian Zoogeographic Province which extends south-
type is backed by steep eroding cliffs and has a
wards and the Oregonian Zoogeographic Province
shallow underlying beach rock platform. Coal Oil
which extends northwards. This Transition Zone is
Point is typical of this type of beach. In many
also characterized by the presence of a group of
cases the sand is periodically eroded away in
short-range species (California Endemics) whose
portion of the intertidal area to reveal the rock
distribution is centred at Point Conception
platform. These sites are usually characterized
(Newman, 1978).
by Megalorchestoidea at the top of the beach;
The combination of habitat variability with isopods~. chiltoni) and bloodworms (~.
this complex zoogeographic pattern results in a mucronata) in the middle of the upper beach; the

high level of biotic variability on the beaches. sand crab (~. analoga) extends from slightly
This complicates data interpretation because, above the middle of the beach to below low tide;
while this data base is composed of over 900 another polychaete (~. zonata) occurs in the
surveys, these are still only surveys at isolated lower intertidal areas. At sites such as Coal

points in space and time and are not continuous Oil Point, Q. biplicata may occur at lower

monitoring of a site or sites. intertidal levels. At sites such as Scripps,

where there is no beach rock exposure, D.
Surveys conducted from 1969 through 1980 gouldii occurs in the lower intertidal areas.

A second type of beach is naturally backed
by low sand dunes and is usually in areas where
a river has developed a flood plain. The beaches
are usually more steeply sloping with coarser
sediments than those overlying bench rock plat-
forms. There are fewer species and specimens
than on the finer grained beaches. They can be
characterized by the presence of ~. analoga and
H. californiensis. They may also provide a
habitat for some beach hoppers, Megalorchestoidea
spp., and~. chiltoni, as well as the sand
dollar D. excentricus in low intertidal areas.
A third type of beach is usually short,
steep, with coarse sediments. All of the sediment
may be lost during storms. This beach is fre-
quently located at exposed rocky points or near
the upper end of submarine canyons. It is not
uncommon to collect no macro-invertebrates on
these beaches. E. ~naloga is the most characte-
ristic inhabitant.
There is only one area that provides a
natural example of a sheltered coastal beach in
the Southern California Bight. This is the Cat
Harbor area on Santa Catalina Island. This
intertidal area (Twin Harbors) is long, gently
sloping, with fine sandy sediments. A large
number of species and specimens occur here
including the ghost shrimp, Callianassa spp. and
the bivalve, Transennella tantilla, which do not
occur at open coastal sites.
DATA BASE AND ANALYSIS
The data base is composed of over 900 field
surveys conducted from 1969 through 1980 at 60
sites located on mainland and island shores of
the Southern California Bight. A number of sam-
pling methods were employed to obtain data to
answer different questions. These sampling
methods were evaluated and their limits defined
(Straughan, 1982). Briefly, the Patterson method
employed a stratified random quadrat sampling
strategy for biota, combined with a single line
443
sampling strategy for abiotic data to obtain an
overall measure of the distribution and abundance
of species and an approximate description of the
habitat for the beach. Data obtained by the
Patterson method form the basis for these analy-
ses. These data are limited to the macro-infaunal
component of the biota (Straughan, 1982).
Data collected from repeated surveys conduct-
edby the Patterson method between July 1975 and
March 1978 were analysed by a series of multi-
variate techniques. Classification of the
surveys on the basis of all specimens collected,
using species maximum and square root transfo~­
mations, clustered .the surveys on the basis of
similarity in species composition and abundance.
Abiotic data from these groups of surveys were
then analysed by multiple discriminant analysis
to determine which abiotics were most important
in delineating these clusters.
RESULTS AND DISCUSSION
Fig. 2 is a summary of the sites and areas
where each cluster is located (see Straughan,
1979 for details). Fig. 3 shows the location of
these survey/site clusters. For example, Area A
surveys dominant survey clusters 3 and 6 while
Area B surveys dominate survey clusters 1 and 2.
The survey clusters (4,7,8,9,10) which cover Area
C are different from the other survey clusters
in that they are site specific and do not contain
data for more than two survey areas in each
cluster. Data from three sites in Area C (Corona
del Mar, Little Harbor I, and Little Harbor II.
see Fig. 1) do not fit this pattern. All three
are located at the mouth of a creek and undergo
large changes after rain. Cluster 5 which is
not shown on Fig. 3 contains a mixture of surveys
from different sites on which few, if any, animals
were collected. These surveys were all conducted
after a major disturbance on the beach.
444

Survey Survey This clustering of surveys would suggest that

...... Decreasing Similarity- Group Location
200
, I !
100
I , ! , ,
0
,
the sandy beaches of the Southern California

Area A Bight can be divided into three areas; the north-

2 Area B
3 Area A
ern or western area, A, which is composed of
mainland and island sites from the Santa Barbara
Channel and Pacific Ocean; the central area, B,
which extends along the mainland shore from the
southern end of the Santa Barbara Channel to the

San Clemente Is. Palos Verdes Peninsula; and the southern area, C,
4 and
San Nicolas Is.
which includes offshore islands and the mainland
coast south of Palos Verdes Peninsula.
5

The biota at each location in area A is vari-

6 Area A able in that the cluster analysis did not form

site specific clusters but each single survey may

7 Santa Catalina Is. yield up to 24 species; the biota at each location

(The Isthmus)
Point Lorna in area B is variable because again site specific
8 and
Outer Cabrillo clusters were not formed by classification but in
9 Scripps general the biota are sparse and most surveys
Santa Catalina Is. yield 0 to 6 species; the biota in area C, except
10 (Twin Harbor)
for those sites exposed to catastrophic changes,
form site specific groups and are generally abun-
FIGURE 2. Survey clusters formed by classifi-
dant (up to 42 species at Twin Harbors).
cation of biological data collected on surveys
between July 1975 and March 1978. See Fig. 3
for location of survey groups and areas A,B.
Multiple discriminant analysis of the abiotic
data based on these site groupings indicated that
• 1975 -1976 only grain size of sediments and moisture content of
... 1975 -1978
. £ 1976 - 1978 only sediments were the most important parameters in
6. .Di.turb.d Sit••
the formation of these site groups. The range of
these parameters for areas A and B, and for each
site specific cluster in area C, were plotted
(Fig. 4). This was predicted to describe the
general habitat features for sandy beaches in
these areas except except at sites subject to
PACIFIC OCEAN major habitat changes (e.g. those at estuaries).

FIGURE 3. Location of Survey groups 1 through
10, Areas A,B,C and sampling sites used in
cluster analysis.
The remainder of the data base was compared
with the prediction to test its validity. Note
that while grain size data were recorded on most
surveys, data on the moisture content of sediments
are only available for surveys after mid-1974.
 445

normally backed by sand dunes and may have a

3.0
relatively flat area in the low intertidal zone.

2.5
This resulted in a significant enlargement of the
habitat description (Fig. 4). This does not in-
'&. 2.0 clude sites such as Port Hueneme where dredged
c
•
g
material is periodically deposited. However,
:IE 1.5
sandy beaches in area A still have finer sedi-
IIJIIIIII] ~ Predicted A
1.0 ~-Amended A ments with a higher overall moisture content than
~ - Predicted B
ELZ2LZ1 - Amended B beaches in area B.
0.5+------,--------,--------,-----,
5 10 15 20 25
" Moisture Content of Sediments Most of the available data for area C were
3.0 recorded for the predictive data base. However,
Group C
sites such as Huntington Beach, and Dana Point
2.5
exhibit site specific trends. Doheny Beach

,&2.0
near the mouth of Doheny Creek has a wider
c range of habitat characteristics on different
•
g

:IE 1.5 surveys. With the exception of Survey Group 7

(Santa Catalina Island, The Isthmus) sediments
1.0 ffiITffi : ~>Ma;nland
~-Dlsland
were generally sandy with an average mean W
0.5+-----,--------,---------,--------- greater than 1.5. The average moisture content
5 10 15 20 25
" Moisture Content of Sediments was usually above 15%. Therefore the abiotic

FIGURE 4. Range of average mean wand % moisture
content of sediments per survey for Areas A and B
and for the site specific groups in Area C. . is
the mean value.
In area A, data from most surveys fell with-
in the predicted range. The exceptions were
5 surveys from sites immediately after impact by
flooding from an adjacent creek. In addition
there were two records of finer sediments (Average
Mean W= 2.59, 2.63) from North Carpinteria, and
one record of drier sediments (12.8%) from
Government Point. This may indicate that the
habitat description will be enlarged slightly by
additional data. The present description is
based on 188 surveys in area A.
The prediction for area B was not as accurate
as that for area A. This was partly due to the
small data base used for the prediction (only 10%
of the total surveys, 244) but mainly because one
beach type found in the area was not really
represented in the predictive sample.
characteristics of areas A and C are similar.
It could perhaps be interpreted that the
sandy beaches of the Southern California Bight
are ecologically divided into three by a central
area (area B) which is more exposed to oceanic
conditions and storms and thus has a more
rigorous habitat than the areas on either side
of it. The areas on either side are more
sheltered, have finer wetter sediments and more
species. This is only part of the explanation
as indicated by comparison of the species
characteristic of each area and the inclusion
of Pacific coast island sites with mainland
sites in Area A.
Species characteristic of area A include E.
~naloga, !. chiltoni, !. mucronata, 10. zonata,
~. californiensis, ~. squamata while species
characteristic of area B include!. analoga, E.
chiltoni, and H. californiensis. In area C,
site specific characteristics are more important
This is than area characteristics so that no species
 446
could be considered generally characteristic of
the area. Emerita analoga, ~. chiltoni,
dillonensis are characteristic of the offshore
sites on San Nicolas and San Clemente Islands;
~. borealis is characteristic of the Isthmus
site on Santa Catalina Island; numerous species
including!. tantilla and Callianassa spp. are
characteristic of the survey site at Cat Harbor
on Santa Catalina Island. Emerita analoga,
californiensis, and~. bullibranchia are gener-
ally characteristic of the mainland sites in
area C.
Some of the more common species in the
Southern California Bight are not characteristic
of major areas because they have specific habitat
requirements that are found at isolated locations
or at few locations in more than one area.
species was characteristic of all areas but E.
analoga was recorded on most surveys.
Comparison of characteristic species from
area A and C, showed that four species characte-
ristic of area A (~. analoga, ~. chiltoni,
zonata, !. californiensis) were also characte-
ristic of at least one site specific grouping in
site C. The distribution of closely related
species is perhaps of more interest and probably
indicates a subtle change between the two areas.
Euzonus mucronata and E. dillonensis co-exist in
both areas but E. mucronata is characteristic of
area A while E. dillonensis is characteristic of
San Nicolas and San Clemente sites in area C.
Scolelepis bullibranchia and~. squamata occur
together but~. squamata is characteristic of
area A and S. bullibranchia is characteristic of
mainland sites in area C.
A less subtle change in closely related
species occurs between areas Band C.
borealis and H. californiensis rarely occur on
the same surveys. Hemipodus californiensis is
characteristic of area B while H. borealis is
characteristic of the Isthmus site on Santa
Catalina Island in Area C. These changes could
and~. be related to changes in water temperature or to
ocean current patterns. However these are diff-
icult to interpret due to annual changes in these
patterns. For example, the warm water sand crab,
Lepidopa californica only extended north to Coal
Oil Point in the warm winters of 1973 and 1977
(Straughan, 1983).
!.
CONCLUSION
In summary, the sandy beaches of Southern
California Bight divide into three ecological
subregions based on a complex interaction
oceanic, meterologic, geologic, and zoogeo-
graphic factors. Area A which includes the
Santa Barbara Channel and the Pacific Ocean
shores of the Channel Islands, is characterized
No
by relatively fine grain sandy beaches (Average
Mean ~ = 1.6 to 2.6) with a moderate to high
moisture content (Mean moisture content 13 to
21.5%). Most beaches are backed by steep cliffs
and have an underlying rock bench which may be
periodically exposed during natural cycles of
~.
sand erosion and accretion. These beaches are
usually only impacted by winter storms. The
beaches are characterized by eight species of
crustaceans and polychaetes.
Area B includes the mainland shores from the
southern end of the Santa Barbara Channel to
Palos Verdes. The beaches are characterized by
relatively coarse sediments (Average Mean ~ =
0.95 to 2.0) and relatively dry sediments (Mean
moisture content = 9 to 14.5%). The beaches are
impacted by both summer and winter storms. While
some of the beaches are backed by steep cliffs
with an underlying rock bench, most of these
beaches are backed by sand. Two crustaceans and
one polychaete are regarded as characteristic of
Hemipodus
the area. However, no macro-infauna were collect-
ed on some surveys in this area.
Area C includes all mainland and island

shores south and west of the Palos Verdes Penin-
sula. This area is composed of several site
specific areas. However, excluding the unique
site at the Isthmus on Santa Catalina Island,
the sediments are generally fine (Average Mean
= 1.6 to 2.8) with medium to high moisture
levels (average moisture content = 12 to 24.5%).
In most years, this area is only exposed to
summer storms. While most of the' mainland is
composed of beaches backed by steep cliffs, ex-
posure of an underlying rock platform is less
frequent than in area A. The sites surveyed on
the offshore islands were generally not backed
by steep cliffs and ranged from very long (350
m) sheltered sites that merged into the land to
shorter (30 m) steeper sites with clear defi-
nition between land and intertidal areas. The
island sites are each characterized by specific
species while the coastal sites are characterized
by a different group of species.
Sites located at the mouth of watersheds or
are periodically replenished by dredged material
do not fit the model.
ACKNOWLEDGEMENTS
I wish to thank Southern California Edison
for support, D. Hadley for preparing the figures
and B. Allen and P. Woodland for preparing the
photo-ready copy.
REFERENCES
Bernstein R1, Breaker 1 and Whritner R (1977)
California current eddy formation: ship, air,
and satellite results. Science 195, 353-359.
Cayan DR (1980) Regimes and events in recent
climatic variables. CalCOFI Rep. xxi, 90-101.
Habel S and Armstrong GA (1978) Assessment and
atlas of shoreline erosion along the California
coast. Pub. State of California, Department of
Navigation and Ocean Development, Sacramento,
Ca. 95814. 277 pp.
447
Mearns AJ (1980) Changing coastal conditions:
1979 compared to the past 25 years. In Coastal
Water Research Project Report 1979-1980, 273-
284.
~
Newman WA (1978) California transition zone:
significance of shortrange endemics. In Gray J
and Boueot AJ, eds. Historical biogeography,
plate tectonics, and the changing environment.
J. Gray and A.J. Boueot. pp. 399-416. Oregon
State University Press.
Reid J1, Roden GI and Wyllie JG (1958) Studies
of the California current system. CalCOFI Rep.
1, 27-56.
Straughan D (1979) Distribution of tar and re-
lationship to changes in intertidal organisms
on sandy beaches in Southern California. In
Proc. 1979 Oil Spill Conference (Prevention,
Behavior, Control, Clean-up). Sponsored: API,
EPA, USCG 10s Angeles, California. 591-601.
Straughan D (1982) Inventory of the natural re-
sources of sandy heaches in Southern California.
Tech. Repts. Allan Hancock Foundation 6, 447 pp.
Straughan D (1983) Sandy Beach Communities ex-
posed to natural oil seepage. In Proc. 1983 Oil
Spill Conference (Prevention Behavior, Control,
Clean-up). Sponsored: API, EPA, USCG San
Antonio, Texas: 485-490.
Straughan D and Stephenson W (1980) Benthic
physical and rhemical stability in King Harbor
1974-1978. In Report to Southern California
Edison, Rosemead, Ca. 80-RD-113. 53 pp.
Tont SA and Delistraty DA (1980) The effects of
climate on terrestrial and marine populations.
CalCOFI Rep. xxi, 85-89.
 449

ECOLOGY OF BEACH AND SURF-ZONE MYSID SHRIMPS IN THE EASTERN CAPE, SOUTH AFRICA

T.H. WOOLDRIDGE Wepartment of Zoology, University of Port Elizabeth, P.O. Box 1600,
Port Elizabeth 6000, South Africa)

INTRODUCTION As part of a major programme on energy

Mysid shrimps are an important component
of the surf-zone biota along sandy
shores in the eastern Cape, South
Africa. Rich concentrations occur in
these inshore waters and they have been
shown to be important sources of food
for many fish species (Lasiak 1983,
Rossouw 1983). Such concentrations of
mysids are also known from other parts
around the world and are predated upon
by a great variety of organisms (see
Mauchline 1980). Despite their apparent
significance, the biology and ecology of
mysids in general have been inadequately
investigated, and for most species only
fragmentary information is available.
flow through the biotic components
associated with eastern Cape sandy
beaches (McLachlan 1983), it was
necessary to investigate and quantify
the role of mysids in detail. This work
is ongoing and only completed aspects
are reported on here.
STUDY AREAS
Investigations were carried out in Algoa
Bay and St Francis Bay, South Africa
(Fig. 1). For the purposes of this
study two zones were delimited:
1. An inner zone between the highest
level of the swash on the beach
and the furthest breakpoint of the

SUNDAYS
RIVER

ALGOA BAY

1
Scale: 1Okin'

FIGURE 1. Algoa bay and St Francis bay showing the location of the sampling sites.
450
waves. This represented a horizon-
tal distance of ca 50-80m off Kings
Beach; a horizontal distance of ca
100-300m for the stretch of beach
between Swartkops estuary mouth and
2km beyond Sundays estuary and a
horizontal distance of ca 500 m off
Maitlands beach. On the latter
beach surf action is extreme and
consequently dangerous to sample.
On this beach no sampling was done
beyond 50m from the lowest level of
the swash. A full description of
these beaches is given by McLachlan
(1977a, 1977b, 1980).
2. An outer zone between the outermost
breaker line and the 20m depth con-
tour. Since the sampling programme
was largely concerned with shallow
marginal areas and areas directly
influenced by surf circulation cells
(see McLachlan 1983) the strip of
water within the 20m depth contour
(ca 3km wide) was considered to meet
these requirements. Along the 20m
depth contour a rocky outcrop (St
Croix Island) and a fixed marker
buoy allowed for orientation and
station location.
SAMPLING METHODOLOGY
Mysids were collected quantitatively
using a variety of gear.
A benthic sledge 50cm wide was used
monthly over one year along a 40m
transect perpendicular to the water
line. Sampling was done during daylight
on King's and Maitland beaches (Fig. 1).
Along each transect four consecutive 10-
metre tows were made, the first located
just below the lowest level of the swash
line. The sledge sampled the upper 1,5cm
layer of sand which filtered through the
stainless-steel mesh floor (1,5mm mesh
aperture) while mysids and other benthic
organisms were swept backwards into a net
(500m mesh) fixed behind the sledge.
Filtration was enhanced by commencing on
a backwash and moving shorewards during
each tow. The degree of surf turbulence
influenced sampling efficiency, so that
sampling was carried out on days when
surf conditions were considered to be
relatively calm. Such conditions were
most likely prevalent before IlhOO around
low tide and when wind was absent or
minimal.
The sledge was also used to determine
spatial distribution of psammophilous
mysids on a number of occasions from the
lowest level of the swash out of 6-7m
depth off Kings beach. This represented
a horizontal distance of about 250m and
extended well beyond the zone of breaking
waves. Sampling was done using SCUBA and
a small boat equipped with an outboard
motor.
Beyond the breaker line and out to 20m
depth in Algoa Bay, it was possible t
sample the water column with a large
conical plankton net (Diameter 1,5m,
length 6,5m and mesh aperture of 500~m).
Eleven series of samples, each series
representing eight stations, were sampled
at approximate intervals of two months.
Stations 1 and 4 (Fig. 1) were situated
off sandy beaches which may generally be
described as "exposed" (McLachlan 1980).
In contrast, Station 7 off Kings beach
experiences relatively calm surf
conditions. Stations 2 and 5 are
opposite estuary mouths while Station 8

is located in relatively calm water
where fairly extensive rocky reefs
occur. These "inshore" stations are
often hazardous to sample and are
located in water of 5-7m depth just
behind the breaker-line. The remaining
sampling sites (stations 3 and 8) are
located in deeper water (18-20m depth)
and allow for comparison with "inshore"
stations.
A rope attached to the net and a large
luminescent buoy allowed for regulation
of maximum sampling depth as the gear
was towed 40-50m behind a ski-boat. The
net sampled from near-bottom (Zlm above
substrate) to the surface at inshore
stations and from 10m to the surface at
the two deeper stations. A Kahlsico 005
WA 130 flowmeter fitted to the net
allowed for quantitative sampling.
During each tow approximately 400-500m 3
of water was sampled.
Sampling in the water column was
extended into the shallow inner zone
using a 57cm WP2 net. These samples
were not taken on a regular basis, but
were used to supplement data collected
using the other types of gear.
It was also necessary to investigate
specific aspects of mysid ecology over
relatively short time periods. Studies
undertaken were:
a) Spatial re-distribution of psammo-
philous mysids relative to tidal
changes in the inner zone over 24
hrs. The methods employed are de-
scribed in McLachlan et al. (1979).
b) The possibility of psammophilous
mysids in the inner zone becoming
451
pelagic after dark was also invest-
igated. These methods are described
by Wooldridge (1981) and are only
briefly summarized here. A transect
20m in length; perpendicular to and
beginning just below the swash line,
was sampled using the sledge. At
Station 2 in water of 1m depth, a WP2
net was towed by hand. At deeper
stations plankton tows were taken
from a boat; Station 3 was located
behind the breaker line ca 80m from
the shore (depth 3-4m) and Station 4
ca 250m from the beach (depth 6-7m).
c) Mesopodopsis slabberi, a shoaling
species, is also present in waters of
reduced salinity (see Tattersall,
Tattersall 1951). In the Swartkops
estuary (Fig. 1) movement between the
estuary and Algoa bay occurs. Two
separate 24h sampling programmes were
therefore carried out in the mouth of
the Swartkops estuary to monitor
movement of M. slabberi and to
investigate whether the estuary is
dependent upon the bay as a major
source of recruitment. Samples were
collected with a WP2 net in the
estuary mouth every 1,5 hrs from near
the water surface and just above the
bottom (depth 2-3m). Current direct-
ion and velocity were recorded using
a calibrated current meter (06AWA
SEIKI MODEL aSK 861) after securing
the boat to a buoy attached to a
heavy anchor.
d) A further 24h study was undertaken to
establish whether M. slabberi
remained in shallow bay waters
immediately behind the breakers
during the day and night. A fixed
station was located 100m off Kings
452
beach (depth Z 4 m) and sampled in
oblique tows every 3 hrs using a WP2
net.
Laboratory determinations of com-
position and abundance are detailed
in Wooldridge, Bailey (1982). On
most occasions it was possible to
analyse entire sledge samples.
Others were analysed from at least
3-5 sub-samples. Dominant mysid
species were grouped into 6 classes
which relate to degree of maturity,
sex and in the case of brooding
females, brood development stage.
Standing stock was calculated with
the aid of a computer after estab-
lishing the relationship between
mass and length by regressing loglO
dry mass (determined at 60°C for 24
hr) on loglO length.
RESULTS AND DISCUSSION
Fourteen species comprising eight
mysid genera were recorded. Repre-
sentation in the inner zone was
almost exclusively due to Gastrosac-
~ psammodytes, while in the outer
zone, all fourteen species were
present. Of these, Mesopodopsis
slabberi was by far the most
abundant. Density and standing
stock of Q. psammodytes and .t!.
slabberi are shown in Tables 1-3,
while the abundance of other species
of mysids recorded in the outer zone
are given in Table 4. These did not
exceed 300 m- 3 on any occasion at
any station. Some of these species
were also recorded in outer zone
sledge samples, but were taken in
very low numbers.
Gastrosaccus psammodytes is bentho-
pelagic. During the day the population
burrows into the surface sand in the
inner zone. Sledge sampling and obser-
vations using SCUBA indicate that it does
not extend beyond the general area where
waves first begin to break as they
approach the shoreline. Within this zone
of distribution, the population exhibits
clear intraspecific zonation (Wooldridge
1981). Brooding females are closest
inshore, while non-brooding females,
males, immatures and juvenile animals
become more abundant in the turbulent
water nearer the wave break-point.
Brooding females exhibit a further
well-defined pattern of specific
distribution. In mysids, larval develop-
ment takes place entirely within the
brood pouch or marsupium. These larvae
are all at the same stage of development,
although in some species replacement has
been observed (Wittmann 1978) which
results in different stages being present.
Examination of brooding Q. psammodytes
indicates nearshore zonation of females
in different stages. Females with
embryos newly-extruded (egg-shaped) are
furthest from the swash line and as
larvae continue to develop in the pouch,
parent females move progressively inshore.
Females with larvae which are ready to
emerge (similar in appearance to the
adult and with pigmented eyes), occupy a
band just behind the swash line on the
beach. Q. psammodytes is also a tidal
migrant (McLachlan et al.) and animals
emerge periodically from the substrate
and move up or down the beach as tides
rise and fall. Thus observed patterns of
zonation are constantly maintained.
 453

TABLE 1. Density and standing stock of Gastrosaccus psammodytes along a 40m transect (O,Sm wide) on
Maitland Beach, St Francis Bay. Sampling monthly over 12 months using a benthic sledge.

J F M A M J J A S 0 N D
Abundance
(Numbers 20m- 2 ) 572 136 525 83 99 273 167 112 223 201 556 1172

Standing: stock
(dry mass in mg 20m- 2 ) 951 948 872 268 434 709 1292 800 1568 1299 1780 3565

Mean abundance per transect = 343

Mean standing stock per transect = 1 207mg

TABLE 2. Temporal and spatial variation in abundance (numbers m- 3) of Mesopodopsis slabberi at eight
stations in Algoa Bay. Sampling at night in the water column using a conical net of 1,Sm diameter.
N.S. indicates no sample. L.N. indicates lost net.

Date Station
1 2 3 4 5 6 7 8
Jan. 1980 N.S. N. S. N.S. 2 215 744 6 74 2
May 1980 1 164 4 98 2 5 13 7
July 1980 100 14 14 15 246 13 1 445 11
Sept. 1980 2 849 1 787 33 2 238 33 10 527 313
Dec. 1980 329 3 932 38 2 225 L.N. L.N. L. N. L. N.
Feb. 1981 1 045 276 4 109 433 4 34 4
Apr. 1981 1 5 1 164 2 0 49 2 265
June 1981 3 5 47 19 961 43 602 31
Aug. 1981 0 130 3 2 721 625 24 13 3
Oct. 1981 0 26 2 21 1 3 0 0
Jan. 1982 215 593 15 2 76 1 0 0
454

TABLE 3. Temporal aDd s£atial variation. in staDdinibstock (mg dry mass m- 3) of Mesopodopsis slabberi
at stations sampled ln A goa Bay. Detalls as ln Ta le 2. .
Date Statlon
1 2 3 4 5 6 7 8
Jan. 1980 N.S. N.S. N.S. 169,0 143,8 0,5 5,1 1,6
May 1980 0 5,9 5,2 7,6 0,1 0,3 1,4 0,5
July 1980 8,5 8,7 13,5 1 264,5 0,5 0 67,3 1,6
Sept. 1980 599,2 314,2 9,1 711,3 3,0 0,4 51,0 22,7
Dec. 1980 10,3 240,0 1,2 131,6 L.N. L.N. L.N. L.N.
Feb. 1981 755,1 74,8 0,2 8,8 93,2 1,0 11,8 1,0
Apr. 1981 0 0,7 0,1 39,4 1,7 0 2,8 1 913,0
June 1981 0,3 1,6 48,7 19,8 727,9 33,0 327,8 1,3
Aug. 1981 0 36,3 2,0 127,1 23,3 3,0 5,4 1,7
Oct. 1981 0 11,6 0,1 18,7 0 0,1 0 0
Jan. 1982 90,8 112,8 0,4 0,1 50,6 0 0 0

TABLE 4. In addition to Mesopodopsis slabberi, The diurnal benthic distribution of G.

thirteen species of mysids were recorded in psammodytes contrasts with a nocturnal
outer zone samples taken with the 1,5m net.
Included is Gastrosaccus psammodytes which pelagic phase when part of the populat-
becomes pelagic at night. ion is present in the water column (see
Table 4). At night the animals may dis-
Species Maximum abundance m- 3 perse seawards and are taken in plankton
recorded on any samples relatively far from the shore
occasion and the zone occupied during the day
Acanthomysis sp. 28 (Wooldridge 1981). Similar patterns of
Doxomysis sp. 84 a diurnal benthic distribution and noc-
Gastrosaccus turnal pelagic activity have been shown
brevifissura 2 for other beach mysids (Bacescu 1934,
Gastrosaccus Moran 1972, Macquart-Moulin, 1977).
psammodytes 6 However, the degree of pelagic activity
Gastrosaccus is differentially variable, dependent on
sanctus 1 age, sex and within females, breeding
Gastrosaccus sp. 6 status. This is related to the observed
Mysidopsis bispinosa 1 pattern of intraspecific zonation.
Mysidopsis major 277 During the day, males, immature and
Mysidopsis schultzei 1 juvenile animals are located furthest
Mysidopsis similis 1 from the shore where excessive turbu-
Nouvelia natalensis 5 lence around the wave-breakpoint results
Rhopalophthalmus in a relatively unstable substrate.
terranatalis 1 constant re-location of mysids in this
Sirie11a sp. 1 zone is necessary as the sand is con-
tinuously re-worked through turbulence.

Inshore, females become less active and
as they move progressively nearer the
beach they spend an increasing pro-
portion of their time remaining in the
sand. Here pelagic activity is related
more to tidal migration than to re-
location in the same area due to sand
disturbance. When sand movement becomes
too severe due to excessive crashing of
the waves, the population becomes con-
centrated in a narrower band inshore.
Similarly, strong ebb and flow currents
on the beach result in animals moving
further offshore. The pattern of dif-
ferential pelagic activity is also
reflected in night samples. Q. psammo-
dytes taken behind the breaker line are
almost exclusively juveniles, immatures,
males and non brooding females (Wool-
dridge 1981). The pattern was again
evident in samples taken behind the
breaker line (Table 4).
Reduced swimming activity in females
permits a greater proportion of the
energy budget to be directed into egg
production. Q. psammodytes breeds
continuously after sexual maturity has
been attained, any individual female
producing a number of consecutive
broods. Young are released from the
brood pouch after 18-20 days in
mid-winter and 8-10 days in mid-summer.
Within 24 hrs after release of young,
the parent female moults, copulates and
extrudes a further batch of eggs.
Fecundity increases with increasing body
size and largest individuals produce up
to 80-90 eggs. Thus continuous all-year
cohort production by the population and
retention and protection of a relatively
large brood coupled with specific dis-
tribution and behavioural patterns con-
455
tribute to the success with which G.
psammodytes is able to cope with harsh
physical conditions encountered in the
surf zone. By remaining close inshore,
loss from the brood pouch as a result of
abrasion, water turbulence and violent
movements of the parent can also be re-
duced. This loss has been shown to be as
high as 25% (Wooldridge 1981) while in
eight species of British mysids the loss
was only 10% (Mauchline 1973).
Whereas Q. psammodytes was the most abun-
dant mysid encountered in the inner zone,
Mesopodopsis slabberi was the dominant
mysid seawards of the breaker line. It
formed dense concentrations in the water
column and was encountered out to 20m
depth which was the maximum depth
sampled. Numbers and standing stock
(Tables 2 & 3) were highly variable
between stations, probably due to its
gregarious nature. Consistent however,
was its greater abundance at stations
just behind the breaker line (depth 5-7m)
compared to stations in deeper water (18-
20m depth). M. slabberi showed no
association with a specific substrate
type and no consistent concentration in a
particular region around the bay,
although on occasions only a specific age
class was represented in samples. Such
monospecificity is well known in mysids
(see Clutter 1969, Wittmann 1977,
Mauchline 1980) and may be partly due to
differences in swimming speed between age
classes so that animals of a similar size
grouped together (Clutter 1969). Maximum
abundance and standing stock at shallow
stations (5-7m depth) exceed 15 000 and
1 900 mg dry mass m- 3 resp~ctively, while
at deeper stations values did not exceed
50 and 6 mg m- 3 water on any occasion.
456
Maximum numbers (IS 246 m- 3 ) recorded
in July 1980 at station 4 were ex-
clusively due to individuals of a
single age class (juveniles) which had
a standing stock of 1 26Smg m- 3 .
Maximum standing stock (1 913mg m- 3 in
April 1981 at Station 8 - Table 2) was
associated with individuals of all age
and breeding classes. On this occasion
abundance was only 2 26Sm- 3 .
Although Mesopodopsis slabberi has a
wide geographical distribution and is
also recorded from waters of reduced
salinity (Tattersall, Tattersall 19S1),
little quantitative data are available
on its abundance from different parts
of its range. Pillai (1968) notes that
M. slabberi occurs in "large numbers"
in coastal waters of the Mediterranean
Sea, while in the Black Sea and Sea of
Azov, Zakutskiy (1970) recorded
concentrations of up to 4 SOOm- 3 •
Standing stock was 3 270mg m- 3 ,
although this is interpreted as wet
mass. M. slabberi is also abundant in
estuarine waters in southern Africa,
for example in the Sundays estuary
where concentrations up to 1 900m- 3
have been recorded (Wooldridge, Bailey
1982). In this estuary specific
patterns of behaviour aid retention of
the resident and breeding population
(Wooldridge, Erasmus 1980). In the
adjacent Swartkops estuary considerable
movement between estuarine and bay
waters occurs, while the resident
population is relatively low in abund-
ance. Figure 2 illustrates flux
through the estuary mouth, with ~.
slabberi entering on the flood tide and
moving out again on the ebb. This was
only evident at night and animals were
mostly present in bottom-water samples
(depth 3m). On the other occasion when
flux was monitored, low numbers of M.
slabberi were encountered. Movement is
probably depended on the presence of
dense concentrations of M. slabberi in
the bay opposite the estuary mouth
coinciding with the time of flood tide.
Such occurrences are illustrated in Table
S. Data given are the first 12 months of
a 24-month night-sampling study of
zooplankton in the estuary. Stations
near the mouth were occupied around low
water and the start of the flood tide.
On one occasion (January 1977), almost
20 000 individuals m- 3 were recorded and
all were of a single age class (juven-
iles). Abundance higher up the estuary
(11 stations sampled) never exceeded 2S0
m- 3 on any occasion during the same
period of study. Thus the dense con-
centrations of M. slabberi in the lower
Swartkops are not resident in the estuary
for any length of time and the bay is not
a major source of recruitment into the
estuary.
The appearance of ~. slabberi at night
only (Figure 2) may reflect a general on-·
shore movement from deeper bay waters.
This is supported by the 24hr study off
Kings Beach (Table 6) which shows ~.
slabberi to be abundant only during dark-
ness. This aspect however, requires
further investigation.
Considering the abundance of M. slabberi
off eastern Cape sandy beaches, it is not
unexpected that they form a major prey
item of teleost fishes feeding in the
surf zone. Approximately SO% of 66
species recorded by Lasiak (1983) pre-
dated upon mysids to a greater or lesser
 457

~
CURRENT VELOCITY - FLOOD TIDE
I 0,6
o
Q)
!I)
0,4
E
0,2·~~~
~() \

o..J
W
>
I-
Z
W • SURFACE
a: 0,6 °
a: BOTTOM
:::> CURRENT VELOCITY - EBB TIDE
()
1000
Mesopodopsis slabberi /
800

__
NUMBERS m-3 ON FLOOD TIDE

600

(')
I

(J)
E
400

200
IO~o ~
a: ~-----------------------.-=-~-~~~--------.~-----~'~----------~~~~==~------------~
w 0
!Xl
:i!
:::>
z

• SURFACE
800 Mesopodopsis slabberi
NUMBERS m- 3 ON EBB TIDE
° BOTTOM
1000 NIGHT 0 _______

12.00~----~----~--------~-----.----~----------~
15.od 18.00 21.00 24.00 03.00 06.00 09.00 12.00
TIME

FIGURE 2. Movement of Mesopodopsis slabberi through the mouth of the Swartkops estuary (lower section)
over 24h in relation to ebb and flow tides shown in the upper section of the figure.

TABLE S. Occurrence of M. slabberi (numbe~s m- 3) over 12 months in the lower Swartkops estuary.
Data reflects presence of mysids during night-sampling on a flooding tide.

Distance from the mouth up the estuary (krn)

Month: 0,5 3,0 4,0 Month: 0,5 3,0 4,0
N 1976 1 080 169 1 M 1 0 0
D 1 0 0 J 0 0 0
J 1977 2 856 19 875 57 J 26 80 21
F 107 31 0 A 0 0 0
M 378 0 0 S 0 0 0
A 388 1 0 0 46 294 0
458

TABLE 6. Abundance (numbers m-3) of M. slabberi over Z4hrs 100m off Kings beach. Oblique tows taken
at three-hourly intervals using a WPZ-p1ankton net in water of ca Sm depth. (-) indicates no animals
recorded.

Sexually mature Immature males

Time males and females and females Juveniles Total
18hOO
21hOO 417 116 4 497 5 030
MIDNIGHT 655 525 1 255 2 433
03hOO 256 713 141 1 110
06hOO 4 3 17 24
09hOO 2 2
12hOO
15hOO 1 1

degree, the proportion increasing to ca during the day it is present in the

70% when numbers were considered. How-
ever, the degree to which mysids con-
tribute to the diet of fish in the surf
zone (recorded by Lasiak, (1983) varied
considerably between sampling occasions.
Variation is to be expected, considering
the gregareous nature of ~. slabberi
(see Tables 2and 3). The occurrence of
mysid shoals in the proximity of the
sampling site at the time of fish-
sampling, will then be reflected in fish
stomachs. As in the case of mysids,
there appeared to be a general dearth of
fish abundance during daylight in the
surf zone, the fish only appearing in
samples in significant numbers after
sunset (Lasiak 1983). Such short-term
changes in time and space of both
predator and prey species can easily be
overlooked so that day and night sampling
is essential and unavoidable.
Although Gastrosaccus psammodytes is
present in the water column at night, low
numbers (maximum 6m- 3 , Table 4) preclude
it as a ~ignificant food item for fish
feeding in the water column. In contrast
substrate in relatively high numbers
(maximum recorded during this study was
1172 in 20 m- 2 ). Here, specialized
bottom feeders such as the sandshark
Rhinobatus annulatus feed on mysids
(Rossouw 1983). Predation is largely by
juvenile sandsharks which feed close
inshore where brooding Q. psammodytes
predominate. In the swash zone G.
psammodytes is also predated upon by
foraging birds, particularly Ca1idris
alba (McLachlan, et ale 1980). Predation
by birds was estimated to be 16% of
annual production.
Although production estimates are not yet
available for ~. slabberi, some
indications of their role in energy flow
through the various biotic components in
the surf zone can be made. Mean standing
stock of M. slabberi was calculated to be
135mg m- 3 for inshore stations around
Algoa Bay. The nocturnal distribution of
M. slabberi in relation to depth has not
yet been investigated, but during the
duration of the programme, the general
zone in which the 1,5m net was operated

at inshore stations was ca 350 m wide and
had an average depth of 9m. Total water
volume in a 350m wide and 1 m long strip
with a mean depth of 9m was therefore
3 150m 3 . This gave a total standing
stock of 425gm M-l strip (M-l = per metre
strip of shoreline and the general unit
used for benthic and pelagic populations
in the surf zone; see McLachlan 1983).
Production estimates for Rhopalophthal-
mus terranatalis, an estuarine mysid
common in the eastern Cape give an
annual plB ratio (somatic and
reproductive production) for this mysid
of 8,66 y-l (Wooldridge, unpublished).
M. slabberi has a smaller mass than R.
terranatalis and both are continuous
breeders producing a number of
generations per annum. M. slabberi
produces four generations per annum in
the Black Sea and Sea of Azov {Zakutskiy
1970) while ~. terranatalis produces
three. If a plB estimate of 8 is con-
sidered for M. slabberi in the eastern
Cape, then annual production is
estimated to be 3 400gms M-l. Although
M. slabberi has a a lower standing crop
than other components in the surf zone
(see McLachlan 1983) the relatively high
plB ratio indicates the key role it must
play in energy flow through the surf
zone community.
Further attention must be addressed to
trophic interactions in this outer zone.
Lasiak (1983) sampled closer inshore
where M. slabberi does not generally
occur in abundance and therefore her
data may have underestimated inter-
actions between fish predators and prey
species in the water column. Similarly,
the role of M. slabberi as a consumer
organism requires investigation. Mysids
459
are generally described as omnivores;
available data suggests that M. slabberi
does not deviate from this pattern. In
this respect the possibility of ~.
slabberi feeding on phytoplankton derived
from rich inshore concentrations, a
feature of eastern Cape high energy
beaches, also requires investigation.
Phytoplankton cells may be circulated
beyond the breaker line via rip currents
where they become available to mysids.
Such a pattern has been suggested for
Metamysidopsis elongata occurring behind
the breaker line off southern California
by Clutter (1967).
Finally, intensive quantitative night-
sampling in coastal environments similar
to those described in the eastern Cape
will possibly disclose rich concentrations
of mysids which play an important role in
energy flow through the respective surf-
zone communities.
REFERENCES
Bacescu M (1934) Contribution a
l'etude des Mysides de la Mer Noire
ainsi que des limans et des lacs en
relation avec la mer ou avec la Danube
Annls scient. Univ. Jassy 19: 331-338.
Clutter RI (1967) Zonation of near-
shore mysids, Ecology 48: 200-208.
Clutter RI (1969) The microdistribution
and social behaviour of some pelagic
mysid shrimps, J. expo mar. Biol. Ecol.
3: 125-155.
Lasiak TA (1983) The impact of
surf-zone fish communities on faunal
assemblages associated with sandy
beaches : a review. In "Sandy Beaches
as Ecosystems" eds. McLachlan A and
Erasmus T, Junk Publishers, The Hague.
Macquart-Moulin C (1977) Le controle
de l'emergence et des nages nocturnes
chez les Peracarides des plages de
Mediterranee. Eurydice affinis Hansen
(Isopoda), Gastrosaccus mediterraneus
Bacescu, Gastrosaccus spinifer (Goes)
(Mysidacea). J. expo mar. B10l. Ecol.
27: 61-81.
Mauchline J (1973) The broods of
British mysidacea (Crustacea), J. mar.
biol. Ass. U.K., 47: 801-817.
460

Mauchline J (1980) The biology of Wooldridge T (1981) Zonation and

mysids and euphausiids. In: Advances distribution of the beach mysid, Gastro-
in Marine Biology, 18: 1-681. eds. saccus psammodytes, J. Zool., Lond. 193:
Blaxter, J.H.S., Russell, Sir Frederick, 183-189.
S. and Young, Sir Maurice. Academic Press, Wooldridge T and Erasmus T (1980)
London. Utilization of tidal currents by
McLachlan A (1977a) Studies on the estuarine zooplankton, Estuar. Coastl.
psammolittoral meiofauna of Algoa Bay, Mar. Sci. 11: 107-114.
South Africa. 1 Physical and chemical Wooldridge T and Bailey C (1982)
evaluation of the beaches, Zoologica Euryhaline zooplankton of the Sundays
afro 12: 15-32. estuary and notes on trophic relation-
McLachlan A (1977b) Composition, ships, S. Afr. J. Zool. 17: 151-163.
distribution, abundance and biomass of Zakutskiy VP (1970) Some biological
the macrofauna and meiofauna of four features of mysids of the hyponeuston in
sandy beaches, Zoologica afro 12: the Black Sea and Sea of Azov,
279-306. Gidrobiologicheskii Zhurnal 6: 17-22.
McLachlan A (1980) The definition of
sand beaches in relation to exposure : a
simple rating system, S. Afr. J. Sci.
76: 137-138.
McLachlan A (1983) The ecology of
sandy beaches in the eastern Cape, South
Africa. In "Sandy Beaches as Ecosystems"
eds. McLachlan A and Erasmus T, Junk
Publishers, The Hague.
McLachlan A, Wooldridge T and van der
Horst G (1979) Tidal movements of the
macrofauna on an exposed sandy beach in
South Africa, J. Zool. Lond. 187:
433-442.
McLachlan A, Wooldridge T, Schramm M
and Kuhn M (1980) Seasonal abundance,
biomass and feeding of shore-birds on
sandy beaches in the eastern Cape, South
Africa, Ostrich 51: 44-52.
Moran S (1972) Ecology and
distribution of the sand dwelling mysid
Gastrosaccus sanctus (van Beneden, 1861)
along the Mediterranan sand shore of
Israel, Crustacean a Suppl. 3: 357-361.
Pillai NK (1968) A revision of the
genus Mesopodopsis (Czemiavsky) Crustacea
Mysidacea, J. Zool. Soc. India 20: 6-24.
Rossouw GJ (1983) The importance of
non-teleost fishes (Elasmobranchs) in the
surf zone with special reference to
Rhinobatus annulatus. In "Sandy Beaches
as Ecosystems" eds. McLachlan A and
Erasmus T, Junk Publishers, The Hague.
Tattersall WM and Tattersall OS (1951)
The British Mysidacea 460pp. Ray Society
Monograph.
Wittmann KJ (1977) Modification of
association and swarming in north
Adriatic Mysidacea in relation to habitat
and interacting species. In "Biology of
benthic organisms": 605-612. eds. Keegan
BF, Ceidigh PO and Boaden PJS, Pergamon
Press, Oxford.
Witmann KJ (1978) Adoption, replacement
and identification of young in marine
Mysidacea (Crustacea), J. expo mar. BioI.
Ecol., 3~: 259-274.

COMMUNITY STRUCTURE OF INTERTIDAL SANDY BEACHES IN NEW SOUTH WALES, AUSTRALIA
DEBORAH M. DEXTER (San Diego State University, California, U.S.A. 92182)
INTRODUCTION
Sand beaches comprise approximately 60% of
the intertidal shoreline of New South Wales.
However, there are no published accounts of the
sandy beach fauna in New South Wales, or for
that matter, for any sandy beaches in
Australia. The shoreline, particularly in the
vicinity of Sydney, is exposed to a heavier
wave regime than that of most other coasts
throughout the world (Short, Wright 1981,
Wright 1976, Wright et al. 1979). Hence the
infauna of exposed sandy beaches inhabit an
especially rigorous physical environment.
The purpose of this study was to examine
qualitatively and quantitatively a large number
of sand heaches along the New South Wales
coastline. Of the physical and biological
factors considered to be of significance in
influencing faunal composition and structure of
sandy beach communities, the parameters related
to exposure seem most important. Exposure to
wave action (Croker et al. 1975, Eleftheriou,
Nicholson 1975, Withers 1977), sediment
oxidation (McLachlan et al. 1979), and other
sediment characteristics (Driscoll 1975,
Howard, Dorjes 1972, Weiser 1959) have been
recognized as determinant factors on sandy
beaches. The interaction of wave action, slope
of beach, sediment particle size, depth of
reduced layers and other related parameters is
well known (McLachlan et al. 1981). These
factors have been combined into an overall
exposure scale for sandy beaches (McLachlan
461
1980). Presumably, the fauna of Australian
sand beaches respond similarly to these
factors. Consequently, sand beaches in New
South Wales were selected along a complete
gradient of exposure to wave action. Faunal
composition, density, zonation, and other
aspects of community structure were examined in
relation to the exposure gradient.
METHODS
From June 1980 through July 1981 a total
of 84 N.S.W. intertidal sandy beaches located
between Trial Bay and Bateman's Bay (Fig. 1)
were examined for species presence, and in the
vicinity of Broken Bay to Port Hacking 27
beaches were sampled quantitatively for faunal
densities by species.
In quantitative studies beaches were
divided into 5 strata, from low elevation with
frequent tidal inundation to high elevation and
frequent exposure, over the exposed area of
beach during spring tides. Within each stratum
four sites were randomly selected. At each
site a stainless steel coring device which
sampled 0.01 m2 surface area was pushed into
the sediment. Three cores were combined to
form a single sample at each site. In a
preliminary survey at four beaches along the
complete exposure gradient (exposed,
semi-exposed, protected clean sand, and
protected gray sand), there was no significant
difference between either the number of
individuals or the number of species collected
462
Port Jackson
o f/
~Bay
/
FIGURE I Location of study area and quantita-·
tive beach sites in New South \';ales.
in core samples taken to a depth of 10 cm in
the sediment versus those taken to a depth of
20 cm. Accordingly, a standard core depth of
10 cm was used for subsequent samples. On some
beaches a replicate set of 3 cores was taken at
each site. The overall sampling design at each
location thus included 5 strata, 20 sites, 20
to 40 samples (60 to 120 cores), for a total
surface area sampled of 0.6 to 1.2 m per
beach.
The sand was seived through a 500 micron
sieve and the remaining residue was returned to
the laboratory. All organisms were removed
from the sediment, then preserved in 5%
formalin and later identified to species and
counted. These data provided information on
zonation, density, number of species, and
dominance for each site.
The physical environment was examined to
determine the degree of exposure to wave
action, and beaches were specifically selected
that had a wide exposure gradient. Field
observations were made on the slope of the
beach, the height of the waves, the width of
the surf zone, the presence or absence of
stable burrows, and the depth of the reduced
layer, if present, within the top 20 cm of the
intertidal beach. A small sediment core was
used to sample sand from high (stratum 1), mid
(stratum 3), and low (stratum 5) elevations.
Particle size, mean particle size, and sorting
was measured using an Emery settling tube and
percent of shell present in the sand was
analyzed through acid treatment (Briggs 1977).
Some of this information was combined to rank
the beaches on the basis of exposure using
McLachlan's (1980) exposure scale for sandy
beaches with one modification. The scale for
the presence of the reduced layer in this study
was defined as: 4 when a reduced layer was not
present within the top 20 cm at any tidal
level; 3 when a gray layer occurred at the low
tidal level. but at a depth greater than 10 cm;
2 when a gray layer occurred only at low tidal
level but at a depth less than 10 cm; 1 when a
gray layer was present at both mid and low
tidal levels at a depth greater than 5 cm; and
o when a gray layer was present at mid and low
tidal levels at depths less than 5 cm. Six
factors: wave action, surf zone width, slope
and median particle diameter, percent of very
fine sand, presence of stable burrows, and
depth of reduced layer, allow a single
composite exposure score ranging from 0 to 20.
In this study beach exposure scores ranged from
4 to 19 (highest exposure).
Beaches along the open exposed coast can
be categorized into two main types: reflective
and dissipative (Wright et al. 1979). A
reflective beach has a steep intertidal slope

with a narrow surf zone so that surging waves
break directly on the beach resulting in coarse
grained sediments. A dissipative beach has a
flat intertidal slope with fine grained
sediments, a wide surf zone frequently with
sand bars, such that wave energy dissipates
before reaching the intertidal beach. Sand
beaches occurring at inlets to bays are
subjected to variable wave activity, from high
to low, and can be characterized as
semi-exposed sites. Within the protected
reaches of a bay, beaches can be divided into
four types: protected clean sand beaches,
beaches which do not have a gray layer present
in the upper 10 cm of the intertidal zone;
protected gray sand beaches, beaches in which a
gray layer is present in the upper 10 cm of the
intertidal beach; sand flats, beaches in which
a gray layer is present, the slope is a very
flat gradient, and epifaunal organisms are seen
on the surface of the sand; and lagoons.
Exposure to wave action declines across this
gradient of protected sand beaches and lagoons
are not subjected to any wave action.
Multivariate analysis of the data was done
using the C.S.I.R.O. TAXON library of programs
on a Cyber computer. All samples from each
intertidal beach were combined to represent
that location. The beaches were classified
into groups on the basis of species composition
and abundance using the Canberra metric
dissimilarity coefficient (Clifford, Stephenson
1975). In this analysis, joint absences do not
contribute to the data analysis. In a zero/non
zero match, zero counts were replaced by 0.1 to
reduce the contribution of rare species. A sites where the number of individuals exceeded
flexible sorting system was used in which beta
= -0.25 (Stephenson, Williams 1971). Since the
number of individuals collected at each
location ranged over 4 orders of magnitude, the
count data were transformed to percent
composition of the total fauna at each beach.
463
This standardization by beach site was used for
normal classification. An inverse
classification of species in relation to their
distribution at sites was also done. The
species were standardized by transforming the
species counts to a percentage occurrence of
each species between the sites thus giving each
species equal weight.
Generally only the more abundant species,
or those with a high frequency of occurrence,
are useful in classifying communities. Of the
78 species collected in the quantitative
studies of 27 sites, 18 were used in the
analysis. Nemertea and Nematoda were
identified only to the phylum level and were
excluded. Also excluded were 33 species
collected at only one location, 14 species
collected at only 2 locations and contributing
less than 2% to the total composition at each
site, and 11 species which occurred at only 3
of the 27 locations and which contributed less
than 1% to the total fauna at each site. The
18 species which occurred at 4 or more sites,
or contributed more than 1% at 3 sites or more
than 2% at 2 sites, were included in the
computer analysis.
The Canberra metric dissimilarity values
(determined using CANMAR) were used in classi-
fication (MULCLAS), to construct a minimum
spanning tree (MINSPAN) (Gower, Ross 1969), and
for ordination of the sites (GOWER) (Gower
1969). Additional statistical treatment of the
data included determination of species diver-
sity (H') and species evenness (J') for those
200 (Pielou 1977). Spearman rank correlations
among various aspects of community structure
were also calculated. The relationships
between the number of species and the number of
individuals partitioned by tidal level and
beach exposure were examined by ANOVA.
464
RESULTS
From a total sample area of 19.2 m2 at the
27 beaches, 16,778 individuals were found,
representing at least 78 species. Other
quantitative and qualitative studies during the
year brought the total number of individuals
collected from sandy habitats in New South
Wales to ove'r 60,000 with a total species
composition of at least 118 species (Dexter, in
press). Of these, 65 species (55%) were
crustacea, 30 (25%) were polychaetes, 16 (14%)
were molluscs, and 7 (6%) were other taxa.
Crustacea, particularly amphipods, were
abundant in all habitats, while the abundance
and diversity of polychaetes increased with
increasing protection from wave action. A South Australia to New South Wales (Barnard,
summary of several descriptive features and
characteristics of structure for each beach at
which quantitative studies were conducted is
given in Table 1.
Marked zonation patterns were evident on
all beaches studied with different species
abundant in the upper versus lower tidal
levels. The upper levels were always dominated
by crustaceans, especially cirolanid isopods.
Amphipods were usually most abundant in the
middle zones, and peracarid crustaceans shared
dominance with polychaetes, especially
spionids, at the low tide levels. Generally
more species and higher densities were found at
the lower tidal levels.
Community structure
The results of Spearman rank correlations
among several aspects of community structure
are shown i.n Table 2. There were significant
correlations among exposure, number of species,
density, relative abundances of polychaete
individuals, and the ratio of number of species
of crustacea/number of polychaete species.
With increasing protection from exposure, more
species occur, the density of the infauna
increases, polychaetes comprise a greater
percent of the numerical composition, and the
number of species of crustacea decline in
relation to the number of species of
polychaetes.
Species composition
Six faunal groupings of species were
identified from the inverse classification.
The first group consists of Tittakunara katoa,
Gephyrocuma pala, and Glycera tridactyla. The
platyischnopid amphipod!. katoa and the
bodotriid cumacea Q. pala were found only on
exposed reflective beaches. Both have a wide
geographical range in Australia occurring from
Dummond 1979, Hale 1929). The polychaete Q.
tridactyla occurred in a variety of sand
habitats, from exposed to protected beaches,
but it contributed a much larger percent of the
fauna at reflective beaches. The density of
this species was, however, very low at all
sites.
The second group occurred in semi-exposed
habitats. It consists of Pseudolana concinna,
Exoediceros maculosus, and Scolelepis
carunculata. The cirolanid isopod R. concinna
is a very widely distributed species
(Bruce 1980) occurring from Queensland to
Western Australia, and was characteristic of
reflective, dissipative, and semi-exposed
beaches. The oedicerotid amphipod !. maculosus
reached its highest densities in semi-exposed
inlet beaches, but occurred in low numbers at
more exposed sites. Its geographical range is
from South Australia through Victoria and New
South Wales (Sheard 1937, M.M. Drummond,
personal communication). The spionid
polychaete~. carunculata occasionally occurred
in low densities in protected sand beaches, but
had higher densities at semi-exposed and
dissipative beaches. It occurs from Queensland
 465

TABLE I
Characteristics of the quantitatively sampled sand beaches of New South Wales.
# LOCATION 2 3 4 5 H' J' DOMINANT SPECIES

1 Coogee R 16 1.67 113 3 Exoediceros maculosus 91 %

2 Pearl R 16 1.76 31 5 Glycera tridactyla 48%
Gephyrocuma pala 38%
3 Maroubra R 18 1.40 35 4 G. pala 52%, E. maculosus 38%
4 Palm R 17 1.53 90 5 Tittakunara katoa 33%, G. pala 26%
5 Narrabeen R 19 0.98 13 5 Exoediceros sp. 38%, T. katoa 25%
6 Garie R 18 1.48 22 6 G. tridactyla 46%, T. katoa 15%
7 Patonga R 17 1.02 12 2 Actaecia pallida 86%
8 La Perouse SE 14 1.67 936 5 0.5 0.22 E. maculosus 93%
9 Bomborah SE 14 1.60 153 3 E. maculosus 96%
10 Bundeena SE 10 2.43 113 4 E. maculosus 91 %
11 Manly Cove SE 14 1.45 600 6 1.0 0.42 E. maculosus 74%
12 Snapperman SE 8 2.10 657 14 1.3 0.37 E. maculosus 72%
13 Balmoral SE 13 1.65 33 6 E. maculosus 50%
14 Deeban Spit SE 13 1.65 33 8 E. maculosus 46%, E. fossor 21 %
15 Ocean D 13 2.12 198 11 2.2 0.64 Urohaustorius metungi 44%
Spio pacifica 33%
16 Towra Point PG 9 1.83 846 23 2.6 0.58 Exoediceros fossor 55%
17 Towra Flat SF 10 1.66 592 17 2.7 0.67 Scoloplos simplex 29%, E. fossor 29%
18 Towra Flat SF 8 1.82 1553 16 1.5 0.38 E. fossor 75%
19 Gunnamatta Bay SF 10 1.45 1110 27 2.3 0.50 S. pacifica 56%
Australonereis ehlersi 17%
20 Towlers Bay SF 9 1.60 873 17 2.7 0.67 A. ehlersi 43%, U. metungi 15%
21 Dolls Point PC 10 1.72 202 10 1.6 0.50 U. metungi 58%, S. pacifica 31%
22 Port Botany SF 10 1.57 6817 13 1.0 0.27 S. pacifica 79%
23 Clontarf PG 9 1.70 397 19 2.7 0.65 E. fossor 47%, S. pacifica 15%
24 Quibray Bay SF 9 2.15 832 21 2.7 0.63 A. ehlersi 40%, S. pacifica 27%
25 Ettalong PG 9 1.53 360 3 0.7 0.43 E. fossor 88%
26 Bells Point SF 8 2.03 1845 21 1.5 0.34 A. ehlersi 77%
27 Narrabeen Lagoon L 4 1.75 5787 19 2.2 0.53 Caraziella victoriensis 53%

Beach type 2 Exposure scale 4 Density #/m'

R = reflective PG = protected gray 3 Median grain size 0 5 Number of species
SE = semi exposed SF = sand flat
D = dissipative L = lagoon
PC = protected clean
466

TABLE 2

Spearman rank correlation on community structure of sandy beaches.

r critical P of 0.05 0.38; P 0.01 0.49; P 0.001 0.61. N 27.

s

Exposure Number of Density %Polychaete # Species

species composition crustacea
polychaetes
Exposure x -0.72 -0.85 -0.48 +0.50
Species x +0.69 +0.69 -0.42
Density x +0.40 -0.61
% x -0.41
Ratio x

south around the Australian continent to
Western Australia (Blake, Kudenov 1979).
The third group, composed of Urohaustorius
metungi, Mictyris platycheles, Exoediceros
fossor, and Pseudolana towrae, is character-
istic of dissipative and protected sand
beaches. The urohaustoriid amphipod ~. metungi
and the soldier crab M. platycheles contributed
a larger percent of the fauna and attained
higher densities on dissipative and protected
clean sand beaches, while the oedicerotid
amphipod ~. fossor and the cirolanid isopod P.
towrae were much more abundant in protected
gray sand beaches. U. metungi, which also
occurs ,in Victoria and Queensland (Barnard,
Drummond 1982a, Fearn-Wannan 1968b) had its
highest densities in clean sand, such as Dolls
Point in Botany Bay and Ocean Beach in Broken
Bay. M. platycheles occurred in clean sand,
but was found together with~. longicarpus in
gray sand substrates. ~. platycheles is found
from Queensland to Tasmania (McNeill 1926).
E. fossor was one of the most abundant species
in protected gray sand beaches, but had a very
wide range of distribution occurring in low
numbers in semi-exposed habitats, ranging
throughout protected habitats and into truely
estuarine conditions (Smiths Lake, Georges
River, etc.). E. fossor is found in salinities
from 16.4 to 28.0 0/00, with a zoogeographical
distribution from South Australia to New South
Wales, including Tasmania (Barnard, Drummond
1982b, Fearn-Wannan 1968b). ~. towrae occurred
in protected sand beaches and was found only
occasionally at very low densities in inlet
beaches.
The oniscoid isopod Actaecia pallida is
partially isolated from other species because
its distribution was sporadic and unpredict-
able. It is found in the uppermost layers of
the beach, as a semi-terrestrial isopod, from
South Australia to New South Wales (Hale 1929),
from reflective beaches to sand flat beaches.
In this study it was most abundant at Towra
Point.
Another group, composed of four species of
polychaetes, Spio pacifica, Capitella capitata,
Nephtys australiensis, and Barantolla lepte, is
characteristic of protected sand beaches. Only
N. australiensis was collected on reflective
beaches and its distribution extends from
Queensland around the southern part of the
continent to Western Australia (Paxton 1974).

The capitellids!. lepte and Q. capitata were
restricted to gray sands, sand flats, and
lagoons. !. lepte is found only in New South
Wales (Hutchings 1974). C. capitata has a
world wide distribution but its taxonomic
position is highly questionable based on recent
work (Grassle, Grassle 1976). Spio pacifica
was one of the most abundant species occurring
on dissipative beaches, in protected clean sand
beaches,and protected sand beaches. It occurs
in Queensland, New South Wales, and Victoria
(Blake, Kudenov 1979). This spionid often was
the numerically dominant species in the fauna
of protected beaches, especially during the
summer months.
The last group, composed of Australonereis
ehlersi, Scoloplos simplex, and Sanguinolaria
donaciodes, is most characteristic of very
protected sites where the reduced layer is
quite evident. The nereid polychaete !.
ehlersi occurred in almost the entire range of
sand habitats surveyed, being absent only from
reflective beaches, and had its highest
densities in the most protected sand habitats.
This species is also abundant in muddy and
estuarine habitats and is distributed from
Western Australia to Queensland (Hartman 1954).
The orbinid polychaete~. simplex, a New South
Wales species (Day, Hutchings 1979), occurred
in all protected sand habitats as well as
sediments higher in silt and clay content.
bivalve~. donaciodes, found also in South
Australia (Cotton, Godfrey 1938), was collected
only from protected sand habitats, but usually
only juveniles were collected intertidally.
Its abundance increased with increasing
protection from wave action.
Site classification and ordination
The 27 beaches can be grouped into three
or four similar site groups, closely related to
exposure regime, using the classification,
467
ordination, and minimum spanning tree
techniques. The ordination technique (Fig. 2)
produced a classification of sites in which the
exposed reflective beaches were placed into one
Vector2--------~,~~----------------~'.\
·12
........... .11 .10-
.............~~~:~..:.?..........i
..... exposed
1111111 semi-exposed
_ protected to very protected
FIGURE 2 Ordination of beach sites. Vector 1
and 2 account for 23% of the variation.
group, the semi-exposed beaches into another
group, and the protected beaches into a third
group. In the classification (Fig. 3) and
minimum spanning tree (Fig. 4) analyses,
protected beaches could be separated into two
groups: protected and very protected
locations. The semi-exposed and exposed
beaches were distinguishable from each other
but formed looser groupings.
The species which occur on the reflective
beaches were predictable, their relative
The abundances were not, and their densities were
always very low. Consequently, the reflective
beaches did not form close groups. The
characteristic species are Gephyrocuma pala and
Tittakunara katoa. The densities on these
beaches near Sydney were usually less than
100/m2 • Typical reflective beaches in the
Sydney area are Bondi, Coogee, Garie, Maroubra,
Narrabeen, Palm Beach, and Pearl Beach. On the
south coast of New South Wales both Racecourse
Beach in Ulladulla and Hyams Beach in Jervis
Bay are reflective beaches, but densities of
468

Great Mackerel Beach in Pitt Water, Chinaman's

1.6 Beach, Manly Cove, and Balmoral Beach in Port
1.4 Jackson, La Perouse and Bomborah Beach in
1.2 Botany Bay, Bundeena Beach and Jibbon Beach in
Port Hacking, Pretty Beach at Kiola, and Green
Patch Beach in Jervis Bay.
Dissipative beaches, protected clean sand

0.4
beaches, and protected gray sand beaches can be

0.2
separated from the very protected sand flat and
lagoonal habitats. The species present at the
_ _ _ _ _ 11111 _ Qallli till
1618172314 25 1 9 7
protected sites were generally the same, but
_ protected
W..If")};! very protected
....... exposed
11111111 semi-exposed
the relative abundances of the dominant species
differed. In the fauna at Ocean Beach, Dolls
Point, and Trial Bay, Urohaustorius metungi and
FIGURE 3 Dendogram produced by normal classi- Spio pacifica were the dominants. In the
fication showing groups of beach sites. protected gray sand beaches Exoediceros foss or
was especially abundant. These sites include
three locations at Towra Point in Botany Bay,
Clontarf Beach in Middle Harbour, and Ettalong
Beach in Broken Bay. Other sites with similar
fauna are Sandringham in Botany Bay, Brooklyn
in the Hawkesbury River, Spit Bridge Beach in
Middle Harbour, Gladesville Bridge Beach, Shell
Cove, and Balls Head Bay in the Parramatta
River, Chowder Bay in Port Jackson, Clareville
Beach and Gruyan Point in Pitt Water, and
_ exposed
Swallow Rock in Port Hacking. Deeban Spit,
1111111 semi-exposed
_ protected
located near a channel within the inlet to Port

~ very protected Hacking, and characterized by a mixed faunal

composition, affiliates with these sites rather
than the semi-exposed beaches.

FIGURE 4 Minimum spanning tree of beach sites.

The fauna of sand flat and lagoonal habi-
tats, the most protected locations, was charac-
the infauna there are much higher than occur on terized by Exoediceros fossor, Australonereis
Sydney beaches. ehlersi and Sanguinolaria donaciodes. Sites
included here are Gunnamatta Bay and Bells
Most of the semi-exposed inlet beaches Point in Port Hacking, Port Botany and Quibray
form a single group. The semi-exposed beaches Beach in Botany Bay, Narrabeen Lagoon, and
are subjected to moderate wave action and are Towlers Bay in Pitt Water. Other sites with
dominated by Exoediceros maculosus and similar fauna are Kogarah and Kyle Bay in the
Pseudolana concinna. Included among Georges River, Rose Bay in Port Jackson, and
semi-exposed sites are Snapperman's Beach and Shoalhaven Heads Inlet.

One additional site group, which appeared
in the qualitative sampling, consisted of
beaches that were intermediate in exposure
between reflective and dissipative beaches.
Beaches along the north coast of New South
Wales such as Crowdy Head, Lighthouse Beach at
Seal Rocks, Windy Wappa at Hawks Nest, Zenith
Beach at Nelson Bay, and the Entrance, and
south coast beaches such as Surfside Beach at
Bateman's Bay, Collingwood Beach and Callala
Beach at Jervis Bay, and Seven Mile Beach were
dominated by Scolelepis carunculata, Pseudolana
concinna, Urohaustorius gunni, Exoediceros n.
sp., and Zobracho canguro. Other species which
may be present at these sites include some
characteristic of reflective beaches such as
Gephyrocuma pala and Tittakunara katoa.
The influence on the number of species and
the number of individuals as partitioned by the
four site groups (exposed: 7 sites, semi-
exposed: 7 sites, protected: 7 sites, and very
protected: 6 sites) and tidal levels (1-5; high
to low) was examined. The mean number of
species was significantly different according
to both tidal level and exposure (Table 3A).
The mean number of individuals was signifi-
cantly different according to exposure, but not
tidal level (Table 3B). Most of the residual
variance in this analysis was contributed by
the very protected beaches.
DISCUSSION
The various techniques used to examine
community structure all showed a similar
general pattern of faunal groupings related to
exposure gradients. Some species were
restricted to reflective beaches: Tittakunara
katoa and Gephyrocuma pala while several were
restricted to protected habitats: Exoediceros
fossor, Urohaustorius metungi, Pseudolana
towrae, Spio pacifica, Scoloplos simplex,
and Sanguinolaria donaciodes. Species
469
characteristic of semi-exposed habitats, such as
Exoediceros maculosus, Pseudolana concinna, and
Scolelepis carcunculata often were present in low
numbers in both exposed and protected habitats.
The observed distributional changes in faunal
composition along the exposure gradient is
predictable. The interaction of wave intensity,
sediment particle size, sediment oxygen and water
content, slope of beach, and their effects on
faunal abundance and composition is well document-
ed in the literature. On the more exposed sand
beaches of New South Wales the platyischnopid
amphipod Tittakunara katoa shares dominance with
the cumacean Gephyrocuma pala. Other character-
istic species are the cirolanid isopod Pseudo lana
concinna, the spionid polychaete Scolelepis
carunculata, and the bivalve Donax deltoides.
Species of Donax are characteristic of exposed
sandy beaches throughout the world as are
cirolanid isopods (Dahl 1952). On the more
protected beaches of New South Wales the two most
abundant amphipods are the oedicerotid
Exodiceros fossor and the urohaustoriid
Urohaustorius metungi. The most abundant
polychaete is the spionid Spio pacifica. Other
characteristic organisms include the bivalves
Mesodesma elongata and Eumarcia fumigata, the
polychaetes Scoloplos simplex and Australonereis
ehlersi, and the crab Mictyris platycheles. The
family Urohaustoriidae is characteristic along
Australian temperate sand beaches, the family
Urothoidae is the most abundant amphipod family
along the warm temperate beaches of the Sinai
(Dexter 1981), and the family Haustoriidae
dominates most of the sand habitats along the
eastern U.S. coast (Croker 1977). There is marked
convergence in morphology of these families (E.L.
Bousfield, personal communication) and they share
similar life histories.
In the most protected sand habitats of New
South Wales, a wide variety of species of
amphipods and polychaetes is present.
470

TABLE 3

Analysis of the variance of the number of species (A) and the

number of individuals (B) partitioned by tidal level and beach exposure.

A B
Number of species Number of individuals
Sources of SS df F P SS df F P
variation
Tidal 108.5 4 8.47 0.002 124,134 4 1.91 0.17
level
Exposure 157.2 3 16.36 0.0002 360,899 3 7.39 0.005
Error 38.4 12 195,444 12

E. fossor is still the dominant amphipod, while
A. ehlersi is the dominant polychaete.
Epifaunal organisms, particularly species of
Nassarius and Polinices are evident, as is the
starfish Astropecten polyacanthus. The burrows
and sand manipulations of the infaunal soldier
crabs Mictyris longicarpus are very obvious.
Several species of bivalves are present as is
the ghost shrimp Callianassa arenosa.
Several generalizations can be made
concerning community structure of sandy beaches
along the New South Wales coastline. The
number of species and the density of individ-
uals is lowest at the most exposed sites, and
increase significantly with a reduction in
exposure. Exposed beaches show marked
dominance by peracaridans (amphipods and
isopods), but with increasing protection from
wave action, polychaetes become more important,
both in terms of number of species and in
absolute densities. An increase in polychaete
abundance with decreasing exposure was also
shown for northern U.S. beaches (Croker 1977).
Comparative studies from Australia do not
exist, but a survey of similar habitats along
the coast of Wales (Withers 1977) showed
similar results. The faunal composition
changed with degree of exposure to wave action
and correlated variables, and number of species
was lowest at the most exposed sites. The
range in species diversity (R') seen in sand
habitats in New South Wales is the same as that
reported for boreal, temperate, and tropical
sand beaches sampled with similar techniques
(Croker 1977, Dexter 1979).
Of the 118 species collected in this
study, at least 50% are presently undescribed.
Of the 18 common species used in my analysis, 5
have been described within the past 8 years,
giving further indication of the limited
knowledge of the sand fauna. All but 4 of
these 18 species (little information is avail-
able for Barantolle lepte, Scoloplos simplex,
Glycera tridactyla, Pseudolana towrae), are
reported from New South Wales and two addi-
tional states. One third of the species occur
in at least 5 Australian states, typically from
Queensland south around the coast to Western
Australia. Given the limited sampling of sand
habitats, it is likely that most of the sand
species will show larger geographical distribu-
tions than are currently known. Such broad
zoogeographical distribution patterns, ranging
from tropical through warm and cold temperate
latitudes is typical of many species of sand
beach invertebrates (Dexter 1977).

ACKNOWLEDGEMENTS
Facilities for this research were provided
by the University of Sydney. The fauna was
identified by the following taxonomists: Dr.
Niel Bruce, University of Queensland (cirolanid
isopods): Margaret Drummond, National Museum
of Victoria (Amphipoda), Dr. Pat Hutchings, The
Australian Museum (Polychaeta), Dr. James
Lowry, The Australian Museum (Crustacea),
Hannalore Paxton, MacQuarie University
(nephtyid polychaetes), Dr. Gary Poore,
National Museum of Victoria (Isopoda),
Dr. Winston Ponder, The Australian Museum
(Mollusca), and Suzette Talbot, University of
Sydney (Mysidacea). Dr. Alan Jones, The
Australian Museum, introduced the author to the
TAXON library, and Dr. Peter Rothlisberg,
C.S.I.R.O., Cleveland, provided access to the
Cyber compvter. Special appreciation is
expressed to Dr. James Lowry for his
enthusiastic interest in this research, and to
Professor Donald Anderson and the faculty,
staff, and students of the Department of
Zoology, University of Sydney for their
marvelous hospitality. This manuscript has
benefi tted from critical review by Dr. Richard
Ford, Dr. Alan Jones, and Dr. Joy Zedler.
LITERATURE CITED
Barnard,JL and Drummond MM (1979) Gammaridean
Amphipods "Of Australia Part IV, Smithsonian
Contributions to Zoology, 269, 1-66.
Barnard JL'and Drummond MM (1982a)
Gammaridean Amphipods of Australia Part V,
Superfamily Haustorioidea, Smithsonian
Contributions to Zoology, 360, 1-148.
Barnard JL and Drummond MM (1982b)
Redescription of Exoediceros fossor (Stimpson
1856), an Australian marine fossorial amphipod,
the type-genus of the new family
Exoedicerotidae, Proc. Biol. Soc. Wash. 95(3),
610-620.
Blake JA and Kudenov JD (1979) The Spionidae
(Polychaeta) from southeastern Australia and
adjacent areas with a revision of the genera,
Mem. Nat. Mus. Vic. 39, 171-280.
Briggs D (1977) Sources and methods in
geography, sediment, London, Butterworths.
Bruce NL,(1980) The Cirolanaidae (Crustacea:
Isopoda) of Australia: The Genus Pseudolana
from the Queensland Coast with description of
three new species, Pac. Sci. 34(2), 153-164.
Clifford HT and Stephenson W (1975) An
introduction to numerical classification, New
York, Academic Press.
Cotton BC and Godfrey FK (1938) The Molluscs
of South Australia Part 1, The Pelecypoda,
Adelaide, Government Printer.
Croker RA, Harger RP and Scott KJ (1975)
Macroinfauna of northern New England marine
sand, II. Amphipod-dominated intertidal
communities, Canadian J. Zool. 53, 41-51.
471
Croker RA (1977) Macroinfauna of northern New
England marine sand, long-term intertidal
community structure. In Coull, BC ed. Ecology
of Marine Benthos, pp. 439-450. Columbia, South
Carolina, Univ., South Carolina Press.
Dahl E (1952) Some aspects of the ecology and
fauna of sand beaches, Oikos., 4(1), 1-27.
Day JH and Hutchings PA (1979) An annotated
check-list of Australian and New Zealand
Polychaeta, Archiannelida and Myzostomida, Rec.
Austr. Mus., 32(2), 80-161.
Dexter DM (1977) Natural history of the
Pan-American sand beach isopod Excirolana
brasiliensis (Crustacea: Malacostraca), J.
Zool., London, 183, 103-109.
Dexter DM (1979) Community structure and
seasonal variation in intertidal Panamanian
sandy beaches, Est. and Coastal Mar. Sci., 9,
543-558.
Dexter DM (1981) Intertidal fauna of Israeli
and Sinai sandy beaches of the Mediterranean
and Red Sea, Bull. Mar. Sci. 31, 812.
Dexter DM (In Press) A guide to sandy beach
fauna of New South Wales, Wetlands.
Driscoll EG (1975) Sediment-animal-water
interactions, Buzzards Bay, Massachusetts, J.
Mar. Res. 33, 275-302.
Eleftheriou A and Nicholson MD (1975) The
effects of exposure on beach fauna, Cah. Biol.
Mar. 16, 695-710.
Fearn-Wannan HJ (1968a) Littoral amphipoda of
Victoria Part 1, Proc. Roy. Soc. Vict. 81,
31-56.
Fearn-Wannan HJ (1968b) Littoral amphipoda of
Victoria Part 2, Proc. Roy. Soc. Vict. 81,
127-135.
Gower JC (1969) A survey of numerical methods
useful in taxonomy, Acaralogia 11, 357-375.
Gower JC and Ross GJS (1969) Minimum spanning
trees and single linkage cluster analysis,
Appl. Statis. 18, 54-64.
Grassle JP and Grassle JF (1976) Sibling
species in the marine pollution indicator,
Capitella capitata (Polychaeta), Science 192,
567-569.
Hale HM (1929) The Crustaceans of South
Australia, Adelaide, Government Printer.
Hartman 0 (1954) Australian Nereidae
including descriptions of three new species and
one genus together with summaries of previous
records and keys to species, Trans. Roy. Soc.
S. Austr. 77, 1-41,
Howard JD and Dorjes J (1972) Animal-sediment
relationships in two beach-related sand flats:
Sapelo Island, Georgia. J. Sed. Petrol. 42,
608-623.
Hutchings PA (1974) Polychaeta of Wallis
Lake, New South Wales, Proc. Linn. Soc. N.S.W.
98(4), 175-195.
McNeill FA (1926) Studies in Australian
Carcinology, No.2., A revision of the family
Mictyridae, Rec. Aust. Mus. XV(l), 100-131.
472

McLachlan _A: -( 1980) The defini tion of sandy

beaches in relation to exposure: A simple
rating system, S. Afr. J. Sci. 76, 137-138.
McLachlan -A, Dye AH and Van der Ryst P (1979)
Vertical gradients in the fauna and oxidation
of two exposed sandy beaches, S. Afr. J. Zool.
14, 43-47.
McLachlan A, Woolridge T and Dye AH (1981)
The ecology of sandy beaches in southern
Africa, S. Afr. J. Zool. 16, 219-231.
Paxton H (1974) Contribution to the study of
Australian Nephtyidae (Polychaeta), Rec. Aust.
Mus. 29(2), 197-208.
Pielou, EC (1977) An Introduction to
Mathematical Ecology, New York, Wiley.
Sheard K,(1937) Amphipods from a South
Australian- Reef, Part 1., Rec. S. Aust. Mus.
5(4), 445-455.
Short AD and Wright LD (1981) Systems of the
Sydney Region, Austr. Geographer. 15, 8-16.
Stephenson Wand Williams WT (1971) A study
of the benthos of soft bottoms, Sek Harbour,
New Guinea, using numerical analysis, Aust. J.
Mar. Freshwater Res. 22, 11-34.
Weiser W (1959) The effect of grain size on
the distribution of small invertebrates
inhabiting the beaches of Puget Sound, Limnol.
Oceanogr. 4, 181-193.
Withers RG (1977) Soft-shore macrobenthos
along the southwest coast of Wales, Est. and
Coastal Mar. Sci. 5, 467-484.
Wright LD (1976) Nearshore wave-power
dissipation and the coastal energy regime of
the Sydney-Jervis Bay region, New South
Wales----A comparison, Aust. J. Mar. Freshw.
Res. 27, 633-640.
Wright LD, Chappell J, Thom BG, Bradshaw MP
and Cowell P (1979) Morphodynamics of
reflective and dissipative beach and inshore
systems: Southeastern Australia, Mar. Geol. 32,
105-140.
 473

THE SPECIES-AREA RELATIONSHIP ON A SANDY BEACH

R.G. HARTNOLL (Department of Marine Biology, University of Liverpool, Port Erin, Isle of Man, U.K.)

1. INTRODUCTION (mean particle size 0.2 mm), and since a general

On hard intertidal substrates the marked local account of the physical features of the beach and
variations in slope, texture, aspect and drainage its fauna is already available (Southward, 1953),
of the rock surface emphasize the physical hetero- further details will not be given here. Samples
geneity of the environment, and this is reflected were collected from the three station detailed
in the obviously uneven distribution of many below.
important species in the community. In marked Station no. Level Sieve mesh No. of guadrats
contrast, rleposi ting substrates usually appear MLWN 2 mm 32
homogeneous from the surface, and there is a 2 MTL mm 10
natural tendency to assume that this is accompanied 3 MLWN mm 12
by an even distribution of the burrowing fauna. A square quadrat of 25 cm side was used for all
The practical implication of this would be that samples. It was randomly located, dug out to a
depositing shores could be effectively investigated depth of 25 cm and sieved through a mesh of the
with the use of much smaller sample areas than size indicated above. The contents were preserved,
rocky ones. However, this may not be true. The and subsequently all organisms identified and
surface uniformity may in fact mask sub-surface counted. For stations 2 and 3 the aggregate weight
variations in the physical environment, and the of each species was determined after drying at
presence of one species may modify the distribution 70°C to constant weight. The various numerical
of others (Reise, 1981a, 1981b). Such physical treatments of the data are described where
and biological factors will tend to generate patchy appropriate in the following section.
distributions, and reduce the effectiveness of a Comparisons are made with other shores in the Isle
given sampling format. of Man : a muddy one at Derbyhaven (Ordnance Survey
In this study selected aspects of species distrib- reference SC 286677) and a rocky one at Port St
ution patterns and community structure are Mary (Ordnance Survey reference SC 212671) (see
analysed for a sandy shore of moderate wave Hawkins, Hartnoll, 1980).
exposure in the Isle of Man. The results are
compared with those from neighbouring rocky and 3. RESULTS
muddy shores, and the implications for sampling 3.1. Species-area analysis
strategies are considered. The simplest technique for this analysis is to
take the quadrats in random sequence, and then
2. STUDY AREA AND SAMPLING METHODS plot cumulative species number against increasing
The sandy beach investigated was at Derbyhaven in sample size. The problem is that the result
the Isle of Man (Ordnance Survey reference SC depends upon the arbitrary order of the quadrats,
289673). It is a sheltered beach of clean sand as discussed by Hawkins, Hartnoll (1980). This is
474

demonstrated very well by lines A and ~ in Fig. 1,

which derive from two different sequences of the
same quadrats for station 1. The solution is to 30
consider all possible sequences of the quadrats, :v
'0
Q) B
a.
'"
'0 20
30
0
C
., '
'"
Q)
Z
'0
Q)

~ 10
20
'0
d 2 5 10 20 30
Z No. of quadrats
10
FIGURE 3. Cumulative species number (calculated
by the probability theory method) plotted against
log number of quadrats sampled, with the fitted
o 10 20 30 regression lines described in the text. A,
No. of quadrats Derbyhaven station 1. B, Derbyhaven station 2.
C, Derbyhaven station 3.
FIGURE 1. Cumulative species number plotted
against number of quadrats sampled for Derbyhaven
of cumulative species number on log sample size
station 1. A and B for two different sequences
of quadrats. C using the probability theory (Gleason, 1925; Williams, 1944, 1950; Holme,
method described in the text.
1953; Ursin, 1960; Hawkins, Hartnoll, 1980). The
data for the three Derbyhaven stations are plotted
'"
Q) in this fashion in Fig. 3. Clearly all points
'0
Q)
a.
for stations 2 and 3 lie very close to straight
'" 20 lines, and for station there is a good fit if
'0
d sample sizes of one to four quadrats are excluded.
Z
This divergence from a linear relationship at
o 5 10 small sample sizes is a common feature in species-
No. of quadrats area analyses (see Hawkins, Hartnoll, 1980).
FIGURE 2. Cumulative species number (calculated Linear regressions of species number on loge
by the probability theory method) plotted against
quadrat number were calculated (for station 1
number of quadrats sampled. A, for Derbyhaven
station 2. B, for Derbyhaven station 3. sample sizes of less than 5 quadrats were omitted)
with the following parameters, and the lines are
and a method which effectively does this by using drawn on Fig. 3.
probability theory has been developed (Hawkins,
coeff. corr. slope intercept
Hartnoll, 1980). The calculation is only
Station 0.999 7.35 5.34
practicable by computer, and a listing of the
Station 2 0.997 6.27 19.9
BASIC programme for the operation is given in
Station 3 0.999 7.74 15.7
Appendix 1. The results for the three stations
Once this linear relationship has been established
produced in this way are presented as line C
it is possible to calculate the 'Index of
in Fig. 1, and lines A and ~ in Fig. 2.
The species-area data are more conveniently Diversity', ~, of Williams (1950). This is a
measure of species richness in the community which
analysed if they can be presented as a linear
is independent of sample size.
relationship, and the most generally useful
transformation to achieve this is the regression
 475

a = increase in sp. number from area 1 to area 2 area, both for all species together, and also for
log (area 2/ area 1)
e selected individual species. For the whole
The value obtained will depend upon the base of community the full sets of samples produce the
the logarithms used, and here the base ~ is used following estimates.
in accordance with general practice to give the -2 g dry wt. m-2
Station No. m g wet wt. m-2
following:
2193
Station - 7.36 2 16926 83.3 25.2
Station 2 - 6.35 14.4
3 12969 53.8
Station 3 - 7.65
The much larger numbers of specimens taken in
3.2. Species proportions
stations 2 and 3 is due to the use of a finer
In addition to the complement of species in a
mesh sieve, and to the greater abundance of
community, measured by the above procedures, the
juveniles in the summer. The effect of increasing
community structure may be evaluated in terms of
sample size on these estimates was investigated
the proportions of the various species of which it
for the data on number of specimens. The quadrats
is composed. Accordingly informational diversity
were taken in the same random sequence as for the
indices were calculated: the 'Shannon-Weiner
Shannon-Weiner calculation, and the 95% confidence
Measure', ~', and the 'Evenness',~. See Poole
limits determined for samples of increasing size
(1974) for a full discussion of these measures
(Table 2). The distribution is obviously clumped,
and their calculation. Since they might well be
with variance much greater than the mean in nearly
affected by the increased number of species in
all instances, and a full discussion of the
larger samples, this was considered. The quadrats
application of confidence limits to clumped
were taken in random order, and~' and ~ calculated
distributions is provided by Elliott (1977). For
on the basis of the number of individuals of
each species for increasing numbers of quadrats the present analysis a simple logarithmic
(Table 1). transformation was considered adequate, or if
zero values were present, a log (x + 1)
transformation.
3.3. Confidence limits of sampling
The same technique was used to calculate confidence
The final aspect of community structure to be
investigated is the number and biomass per unit limits for selected species, mainly the more common
ones, from each station (Table 3).
TABLE 1. Values of the Shannon-Weiner measure (H') and Evenness (J) calculated for increasing numbers
of randomly ordered quadrats. The number of species present is also listed.

4 8 12 16 20 24 28 32
Station 1
H' 1.49 1.52 1.61 1.65 1.67 1.71 1.70 1.72 1.73
J 0.64 0.58 0.54 0.5 1 0.5 1 0.52 0.51 0.50 0.50
Species 10 14 20 25 26 26 28 30 31
Station 2
H' 0.83 1.03 1.01
J 0.30 0.32 0.30
Species 16 26 32
Station 3
H' 1.84 1.80 1.78 1.78
J 0.64 0.55 0.50 0.50
Species 18 27 34 35
476

TABLE 2. The 95% confidence limits for the number of organisms m- 2 for increasing numbers of quadrats.
Method of calculation described in the text. The value in parenthesis is the upper confidence limit
expressed as a percentage of the lower.

Number of guadrats Station 1 Station 2 Station 3

4 1310 - 3656 (279) 8572 - 34701 (405) 8784 - 14691 (167)
8 1789 - 3411 (191) 12989 - 21399 (165) 10380 - 14509 (140)
10 13499 - 19769 (146)
12 1796 - 2964 (165) 11253 - 14414 (128)
16 1565 - 2735 (175)
20 1499 - 2471 (165)
24 1352 - 2320 (172)
28 1455 - 2359 (162)
32 1538 - 2365 (154)

TABLE 3. The 95% confidence limits for individuals m- 2 of selected species for increasing numbers of
quadrats. Method of calculation described in the text. Value in parenthesis is the upper confidence
limit expressed as a percentage of the lower.

Station 1
Number of Eurydice Ampelisca Bathyporeia Urothoe Clymene
quadrats pulchra brevicornis pelagica brevicornis oerstedii
4 306 - 1482 (484) 34 - 197 (580) 400 - 1205 (301) 296 - 1116 (377) o
8 506 - 1052 (208) 43 - 171 (397) 320 - 1315 (411) 419 - 955 (228) o- 39
12 489 - 843 (172) 51 - 158 (310) 416 - 1026 (247) 440 - 820 (186) 2.5 - 41
16 433 - 774 (178) 52 - 143 (275) 401 - 960 (239) 351 - 743 (212) 3.1 - 31
20 415 - 704 (174) 45 - 121 (269) 386 - 728 (189) 325 - 658 (203) 6 - 32
24 406 - 687 (169) 46 - 110 (239) 345 - 656 (190) 243 - 591 (243) 5.5 - 27
28 409 - 657 (161) 52 - 110 (212) 379 - 667 (176) 274 - 613 (224) 7.9 - 27
32 412 - 665 (161) 56 - 112 (201) 394 - 653 (166) 300 - 615 (205) 7.4 - 24
Station 2
Number of Pygospio Eurydice Bathyporeia Urothoe
quadrats elegans Capitellid sp. pulchra pelagica brevicornis
4 6280 - 27640 (44) o- 1690 13 - 3100 1450 - 2440 (169) 54 - 169 (312)
8 9550 - 16620 (174) 53 - 549 (1030) 56 - 650 (1160) 1610 - 2360 (146) 63 - 148 (234)
10 9780 - 15170 (155) 79 - 455 (576) 81 - 595 (730) 1660 - 2430 (146) 67 - 167 (250)
Station 3
Number of Clymene Pygospio Eurydice Ampelisca
quadrats oerstedii elegans pulchra brevicornis
4 82 - 1310 (1600) 964 - 5210 (540) 2650 - 3900 (147) 164 - 435 (265)
8 156 - 556 (357) 1790 - 4250 (237) 2990 - 4070 (136) 213 - 369 (173)
12 235 - 562 (240) 2310 - 4380 (190) 3130 - 3960 (127) 240 - 363 (151)
Bathyporeia Urothoe Corophium
pelagica brevicornis crassicorne
.4 2890 - 3350 (116) 553 - 1075 (194) 662 - 1100 (166)
8 2630 - 3340 (127) 723 - 1042 (144) 844 - 1200 (143)
12 2530 - 3140 (124) 777 - 1013 (130) 846 - 1170 (138)
 477

4. DISCUSSION An additional application of the linear relation-

4.1. Species-area analysis ship is to estimate the total species within a
The species-area curves in Figs. 1 and 2 are of community, provided that the total area occupied
the same form, initially steep, becoming flatter, by the community is known (see Ursin, 1960;
but not levelling off within the area sampled in Hawkins, Hartnoll, 1980). Such an extrapolation
this study - an area which is large in relation to must be done cautiously though, and the original
that usually worked at a single station in most samples must be a proper stratified random sample
surveys. Thus the idea of a 'minimal area', namely of the community.
an area which needs to be sampled to include all The Index of Diversity provides a measure of
of the species in the community, is not applicable species richness in the community which is
(see Hawkins, Hartnoll, 1980, for discussion and independent of the area sampled. The values
references to minimal area). However, the obtained in this study can be compared with some
existence of a close linear relationship between other intertidal and subtidal communities.
species number and log sample area (Fig. 3) enables
Intertidal
the effect of increasing sample size on species
Derbyhaven mud MTL 5.4
number to be evaluated for areas both within and
Port St Mary MTL 6.6 Hawkins,
greater than that sampled (Table 4). This is done rock MLWN 15.9 Hartnoll,
MLWS 16.1 1980
in terms of the addition of species both in
absolute terms, and as a percentage increment to Subtidal
those already collected. Such information does not Plymouth, soft Bigbury 6.5 Holme,
provide any absolute criteria for sampling formats, Whitsand 9.6 1953
Station B5 16.7
but does enable sampling to be planned
Northumberland, March 24.8 Buchanan,
realistically in relation to the objectives of the soft Sept. 29.1 et al, 1978
study. If Table 4 is compared with a comparable
In comparison with the above the values for the
analysis for a rocky shore in the Isle of Man at
Derbyhaven sand of 6.3 to 7.7 are certainly at
MTL and MLWN (Hawkins, Hartnoll, 1980) it is clear
the lower end of the range, though comparable with
that the sandy shore can not, in terms of species
mid-tide mud and rock. However it would be
complement, be sampled effectively in a very much
dangerous to generalise too much, since there is
smaller area than the rocky shore.
considerable variation in the level of identific-

TABLE 4. The effect of increments in sample area ation, and the extent to which sampling was
on the absolute and relative increase in predicted restricted to a single environment.
species number.
Station Station 2 Station 3
4.2 Species proportions
Abs. % Abs. % Abs. %
inc. inc. inc. inc. inc. inc. From Table 1 it is seen that the general trend

0.25 15.5 28.6 26.4 with increasing sample area is for H' to increase

0.5 5.1 4.3 15 20 and for ~ to decrease, but in both cases the
33 5.4
0.75 3.0 changes are relatively modest after an area of
15 2.5 7.7 3.1 9.7 2
1.0 2.1 1.8 5.0 2.3 6.6 0.25m has been sampled. The effect of the
9
1.25 1.7 6.4 increased number of rare species is countered by

1.5 1.3 4.7 the further addition of specimens of the common

1.75 1.1 species. Relatively modest sample areas seem

3.9
2.0 1.0 4.3 11.6 adequate to determine these indices of community
3.3 5.3 14.3
4.0 5.1 17 structure on this sandy shore. However on a rocky
478

shore at Port St Mary H' was quite variable until upper 95% confidence limit still generally exceeds
-2
an area of 0.5 - 0.75 m had been sampled (S.J. the lower limit by 50 - 100%. The figures for
Hawkins, unpublished). individual species (Table 3) are generally
There is not space here to discuss the values of estimated with even less precision than the
~' and i at length in relation to other communities, total complement of individuals, though within
but a few points will be considered. One is to a community there is considerable differe~ce

see how the values derived from the number of between species. It is also clear that the
specimens compare with those derived from dry distribution of a particular species may be more
weight. even in one community than another. As in the
case of species content, these figures do not
H' J Sp. no. enable an ideal sampling format to be designated,
Derbyhaven Station 2
but they give an indication of the accuracy to be
Number 1.04 0.30 34
Dry wt. 1.50 0.43 34 expected from a given format.
Derbyhaven Station 3
Number 1.78 35 ACKNOWLEDGEMENTS
Dry wt. 2.07 35
I am grateful to S.J. Hawkins and J.A. Steventon
The species in these two samples at least are more
for permission to use unpublished data. M.T.
evenly distributed in terms of biomass than in
Burrows and F.M. Ellard collected, sorted and
number of individuals. The second comparison
identified much of the material, and this study
involves looking at values for other intertidal
would not have been possible without their help.
communities in the Isle of Man - a muddy shore,
The Nuffield Foundation generously provided
and the animals (the algae produce rather different
financial support for this work.
results) of a rocky shore (S.J. Hawkins,
unpublished) •
REFERENCES
H' J Sp. no. Buchanan JB, Sheader M and Kingston PF (1978)
Derbyhaven mud Sources of variability in the benthic macrofauna
(number) off the south Northumberland coast, 1971-1976,
1.63 0.49 28
J. mar. bioI. Ass. U.K. 58, 191-209.
Port St Mary rock Elliott JM (1971) Some methods for the
(dry wt.) statistical analysis of samples of benthic
MTL 1.14 0.38 20 invertebrates, Scient. PubIs. Freshwat. bioI.
MLWN 1.14 0.30 43 Ass. 25, 1-160.
MLWS 0.93 0.24 46 Gleason HA (1925) Species and area,
Ecology 6, 66-74.
The values for the muddy shore are similar to Hawkins SJ and Hartnoll RG (1980) A study of
those for the sandy shore. The values for the rock the small-scale relationship between species
number and area on a rocky shore, Estuar. cstl
are rather lower than those for the dry weight mar. Sci. 10, 201-214.
Holme NA (1953) The biomass of the bottom
values from the sand, indicating a more uneven fauna in the English Channel off Plymouth, J. mar.
distribution between the species. bioI. Ass. U.K. 32, 1-49.
Poole RW (1974) An Introduction to Quantita-
tive Ecology, Tokyo. McGraw-Hill Kogakusha.
4.3. Confidence limits of sampling Reise K (1981a) Gnathostomulida abundant
alongside polychaete burrows, Mar. Ecol. Prog.
The figures for the total number of specimens
Ser. 6, 329-333.
(Table 2) demonstrate that these sandy shore Reise K (1981b) High abundance of small
communities are certainly not homogeneous, and zoobenthos around biogenic structures in tidal
sediments of the Wadden Sea, Helgolander wiss.
that at this scale of sampling (1/16 m2 ) there is Meeresunters. 34, 413-425.
considerable inter-sample variation. Even after Southward AJ (1953) The fauna of some sandy
and muddy shores in the south of the Isle of Man,
an aggregate area of 0.5 m2 has been sampled the Proc. Trans. Lpool bioI. Soc. 59, 51-71.
 479

Ursin E (1960) A quantitative investigation

of the echinoderm fauna of the central North Sea,
Meddr Danm. Fisk. og Havunders. 2, 1-204.
Williams CB (1944) Some applications of the
logarthmic series and the index of diversity to
ecological problems, J. Ecol. 32, 1-44.
Williams CB (1950) The application of the
logarithmic series to the frequency of occurrence
of plant species in quadrats, J. Ecol. 38, 107-138.

APPENDIX 1. BASIC programme for the calculation

of predicted species number in different numbers
of quadrats. Modified from Hawkins, Hartnoll
(1980).

10 REMARK SPECIES/AREA ANALYSIS

20 INPUT "NUMBER OF SPECIES IN SAMPLES"jS
30 INPUT "NUMBER OF QUADRATS IN SAMPLES" i M
40 INPUT "NUMBER OF QUADRATS TO BE PREDICTED"iN
50 LET A=0
60 LET B=0
70 DIM K(S)
80 DIM R(S)
90 REMARK INPUT QUADRATS CONTAINING EACH SPECIES
100 REMARK ONE SPECIES PER LINE
110 FOR 1=1 TO S
120 INPUT K( I)
130 LET R(I)=K(I)/M
140 IF K(I)<>1 THEN 160
150 LET A=A+1
160 IF K(I)<>2 THEN 180
170 LET B=B+1
180 NEXT I
190 PRINT "NUMBER OF PREDICTED CORRECTED"
200 PRINT "QUADRATS SPECIES PREDICTION"
210 FOR Q=1 TO N
220 LET T=0
230 FOR 1=1 TO S
240 LET T=T+(1-R(I))fQ
250 NEXT I
260 LET Y=S-T
270 LET X=Y+(1-(1-1/M)fQ)*(A/(1-(1-1/M)1M)-A)
+(1-(1-2/M)fQ)*(B/(1-(1-2/M)tM)-B)
280 LET Y=INT(Y*100+.5)/100
290 LET X=INT(X*10@+.5)/100
300 PRINT" "iQi" "iYi" "iX
310 NEXT Q
320 END

INTERACTIONS BETWEEN COASTAL PLANKTON AND SAND MUSSELS ALONG THE CAPE COAST, SOUTH AFRICA
L. HUTCHINGS, G. NELSON, D.A. HORSTMAN and R. TARR (Sea Fisheries Research Institute, Cape Town,
South Africa)
INTRODUCTION
The white sand mussel Donax serra Roding is an
important and often dominant component of the
sandy beach macrofauna around the southern
African coastline from Namibia through to the
Transkei, with very large populations on the west
c.oast and in the Algoa Bay region (de Villiers,
1973, 1974; McLachlan and Hanekom, 1979; . Bally,
1981). It is commonest on exposed beaches and
filter feeds on phytoplankton and detritus. East
of False Bay it coexists with Donax sordidus
Hanley, which occupies a niche lower down the
shore than D. serra (McLachlan, 1977; Ansell,
1981). On the West Coast adult D. serra occupy
this infratidal zone. In both areas, high
variability in population size occurs between
adjacent beaches, which may be due to differences
in food availability within the surf zone, which
is in turn related to wave action, source of
nutrients, sediment size and beach slope
(McLachlan and Hanekom, 1979; Bally, 1981).
Even on scales of a few hundred metres there is
considerable variation in the distribution of
mussels (Bally, 1981). Between the eastern and
western Cape beaches very much smaller
populations of mussels occur (Table 1).
The population structure and vertical distribution
of mussels appears to be very different on
beaches north of Cape Point compared with beaches
further east (e.g. Fig. 2 of de Villiers, 1974;
Fig. 4 of McLachlan and Hanekom, 1979), although
Ba lly (1981) has observed both types of verti ca 1
distribution of mussels on West Coast beaches.
481
Generally, on West Coast beaches juvenile mussels
are most abundant high up the beach, with
progressively larger mussels lower down. On
eastern Cape beaches few juveniles are found
intertidally, and the bulk of the adults occur at
the mid-tide level, moving up and down the beach
between spring and neap tides.
There are also differences in the maximum size
attained and in the reproductive cycle. While
West Coast mussels reproduce throughout the year
with only slightly increased activity in spring
and autumn (de Villiers, 1973), on the East Coast
'a better defined reproductive cycle is apparent,
with two spawnings per year (McLachlan and
Hanekom, 1979). These reproductive differences
may be attributed to the seasonal temperature
cycles in the two areas, with upwelling along the
West Coast offsetting the increased solar
radiation during summer months.
The differences in population structure and
vertical distribution and the apparent paucity of
mussels on southern Cape beaches may be due to
differences in the particulate food sources
around the Cape coast. On the West Coast
upwelling caused by southerly winds occurs
frequently, resulting in high concentrations of
phytoplankton. Upwelling is less frequent east
of Cape Point but does occur sporadically along
the South Coast, usually associated with
headlands (Schuman et al., 1982). Surf-zone
diatoms occur commonly on South and East Coast
beaches and may provide a sufficient source of
482

TABLE 1
Biomass estimates of sand mussels along the Cape coast

Beach Biomass Mil dmil

Source Comments
g dry wt.m- l NO'm- Z g.m- z
West Coast
El ands Bay 8 846 Hors tman 1981, 5 x 106 adult mussels of 60 mm
this report length washed up following red
water mortality
Rocherpan 76 1316 19 Bally 1981 Mean of 16 sample grid, 6 x
over 15 months, intertidally
Brittania Bay/ 6 900 75-288 173-662 This report Mean of 5 transects over 4 km
Pa ternos ter on two beaches from 1 m depth
to L.W.S. 12 x 10 6 adult
mussels, 60 mm
Yzerfontein 370 1748 65 Bally 1981
Me 1kbos
-170
- 1700 68 Bally 1981
Mean 3272
South Coast
Strui sbaai 83 McLachlan Series of 2 - 4 .25 m2 quadrats
Still Bay 1 et al. 1981 on a transect of 6 - 7 sample
Wil derness 18 sites - includes biomass of
Keurbooms 16
-- D. sOY'didus
Mean 29
East Coast
Maitland River 6 524 250 315 McLachlan Mean of summer and winter
Sardinia 33 (1977 ) transects
St Georges 88
Mean 2 179

food for sand mussels, with recycled nutrients
maintaining the fertility of the surf zone in
circulation cells which persist during onshore
wind conditions (Lewin et al., 1975; McLachlan,
1980). In the absence of any notable exchange
of energy with coastal waters, these east Cape
beaches may well function as semi-enclosed
ecosystems, but on the West Coast some very close
interactions between coastal plankton and sand
mussels are shown together with some possible
reasons for the variations in mussel abundances
around the coastline.
DISTRIBUTION OF MUSSELS AROUND THE
SOUTH AFRICAN COAST
Estimates of mussel abundance on beaches is
fraught with errors associated with the state of
the sea, seasonal or long-term changes, spatial
patchiness and method of collection (de Villiers,
1973, 1974; McLachlan and Hanekom, 1979; Bally,
1981) .
A comparison of mussel biomass on Cape beaches
(Fig. 1) is available from the work of McLachlan
(1977), McLachlan et al. (1981) and Bally (1981)
- see Fi g. 2, Tab 1e l. Th i s compari son shows
that high biomasses (and high variability) of
mussels occur in both the eastern and western
Cape, but biomass levels at Maitland River Beach
in the eastern Cape are exceptional, being an
order of magnitude higher than almost any other
published beach biomass. However, two further
biomass estimates of white mussels on the West
Coast have been made recently.

32°
CAPE PROVINCE
Fig. 1 Location of sandy beaches, Cape Province, South Africa
The fi rs t was reported by Hors tman (1981) who
estimates that some 5 million adult mussels were
washed up over 1,3 kilometres of beach at Elands
Bay in March 1980 during a red-water outbreak.
From the biomass conversion of McLachlan and
Hanekom (1979) for 60 mm long mussels, a biomass
of some 8 800 g dry weight per metre of beach
results.
A second estimate of infra tidal mussel biomass
at Paternoster and Britannia Bay on the l~est
Coast was made recently by the Sea Fisheries
Research Institute in response to a request to
harvest mussels commercially at these beaches.
A series of 0,25-m2 samples excavated by hand on
a number of transects below LWS along 4 km of
beach produced an estimate of 12 x 106 adult
mussels, again giving a biomass estimate of the
same order as Maitland River Beach. It would
appear that the adult mussel populations on West
Coast beaches have been severely underestimated
by conventional transect sampling within the
intertidal zone,' and comparable mussel
populations occur on both eastern and western
Cape beaches. In fact, excluding Maitland River
483
beach, published biomass estimates of mussels on
eastern Cape beaches are more modest.
FOOD AVAILABILITY FOR MUSSELS IN THE
SOUTH-WEST CAPE
Both de Villiers (1974) and Bally (1981) allude
briefly to upwelling creating a rich supply of
food for mussels on the West Coast. In this
section, the mechanisms whereby food generated
by a coastal divergence can return particulate
matter in dense concentrations to the nearshore
zone are illustrated. During vigorous upwelling
low concentrations of particulate matter occur
close inshore (Andrews and Hutchings, 1980;
Barber and Smith, 1981). Uniform upwelling
along a coastline would then produce a near-
shore zone poor in phytoplankton, unless the
rate of. upwelling was sufficiently slow to allow
phytoplankton to develop during the ascent.
However, upwelling is seldom uniform, and off
the south-west Cape it has been shown (Fig. 3)
to occur in localized areas (Nelson and
Hutchings, in press; Shannon et aZ., 1980;
Lutjeharms, 1981a).
484

WEST COAST
I SOUTH COAST I EAST COAST

10000
I I
t;--;-, I I
~ I I r:-
5000
I I
I I
I I
_ 2000 I I
"
,£
I I
11
~ I I
'0 1000 I
E I
Q;
a. I
500 .:
'j
>-
I
.;; I
'"
1 I
I
I
~
g 200
.2 r,-
I
g
~

iii
90
80 ,........
I
I
'"
70 r.-
60 I
50
40
.. I
30 I
20
.... I
~
10
~
I
I

F"7'l
I
I n...
j
>- E 8 >-

"""
..( .0
...-.;
~ ~
'c
~
~
j
0 0 0
~ .§
.n ~
lID lID
Ii
~
--
0 0
'c 'c
J
~
-tl '3 :2 0
~

.!! ~ ~ ~ ~ ~ ]
~
~
0 ~ ..: ~ .g
iIi ,j;
""
Q<
~ ~ ~
'"
BEACHES ALONG COAST

Fig. 2 Relative abundance of sand mussels on various beaches as g ash-free

dry wei ght per running metre of beach

North of these sites upwelling is less intense
and countercurrents set up during upwelling
events can bring mature upwelled water close to
the coast. Similarly a change in wind to the
west can transport maturing upwelled water
shorewards, as indicated by the drift of a
drogue set in newly upwelled water in December
1979 (Fig. 4), as the peak in phytoplankton
concentration occurs (Fig. 5).
Even a relaxation in wind stress can cause
downwelling at the coast as the sea level
readjusts (Andrews and Hutchings, 1980). A
drogue set in newly upwelled water in February
1981 first travelled northwards and offshore,
and then southwards and onshore as the wind
stress stopped (Fig. 6). Recurvature of south-
easterly winds north of the Cape Peninsula can
also result in onshore flow of surface waters.
Progressive vector diagrams of annual wind at
four sites between Olifantsbos and Nuwedam are
shown in Fig. 7, where onshore components occur
at Bokpunt, Cape Columbine and particularly
Nuwedam.
 485

(a)

30°

25 November 1980

(b)

23 November 1980

(e)

11 January 1980

16° 17° 18° 19° E

Fig. 3 A montage of active upwelling plumes at various sites in the S.W. Cape during
strong southerly winds (A.R.T. data, °C)-from Nelson and Hutchings (in press)
486

PLANKTON DYNAMICS 7 - 11 DECEMBER 1979

DROGUE TRACK

S ~--------------~----~~--~---,~-------------,
Drogue"'relrieved
D29 .,.
PD28
2,5nm PD27

DAY 3

3:f50'

DAY 1 Table Bay

e----..-PD16

1----41 20m/s

Jl!f1l!/;M!IIU&I,lliiJiJII4.!.)l/ilkJ;
DAY 1 I DAY 2 I DAY 3 I DAY 4 I DAY 5 I I
8 9 10 11 12 13

Fig. 4 Track of a drogue set at 10 m depth in newly upwelled water in

Dec. 1979, with wind stick vectors
 487

-.....
'
C)
:::t
PhaseI
y
Phase II

-20
o 16
.....
.....
>- 12

~ 8
9J: 4
U
NITRATES

SILICATES
........................................
...............

Time 12hOO 24hOO

Date 7.12.79
12hOO 24hOO
8.12.79
12hOO 24hOO
9.12.79
12hOO 24hOO
10.12.79
12hOO
11.12.79
Fig. 5 Variation in mean integrated chlorophyll 'a' and nutrients in the euphotic zone
-
during the Dec. 1979 drogue cruise

A similar picture is obtained from wind data
collected by ships at sea. These winds produce
a transport of near-surface water roughly in the
direction of the wind and if the onshore
component is frequent, plankton will be
transported up to the surf zone. Once in the
surf zone, rip currents and wave motion play the
dominant role in transporting particulate matter
in the shallow water. For instance the onshore
flow compensating an offshore rip may equal in
magnitude the onshore transport caused by wind-
induced surface transport, thus doubling the net
onshore flow at localized sites.
Examination of the progressive vector plots shows
that wind reversals and relaxations occur
suffi ci ently frequently (every 5 - 15 days) to
ensure high phytoplankton concentrations close
inshore, except at the upwelling sites themselves.
For instance, Noordhoek beach, some three
kilometres long on the Cape Peninsula, has a
much smaller mussel population than beaches
further north and P. Brown (pers. comm.) shows
consistently more phytoplankton present at Robben
Island than at Oudekraal during summer months.
East of Cape Point upwelling is less vigorous
and less frequent due to changes in bottom
topography (Tromp and Horstman, in prep.) and
wind stress curl (Hutchings and Nelson, in
prep.). This decline in upwelling activity
results in less phytoplankton in the nearshore
zone east of Cape Point, as can be seen in the
average chlorophyll 'a' concentrations in the
upper 20 m at a series of inshore stations (Fig.
8) which were sampled monthly during 1977/78
(Fig. 9). There is a steady decline in
concentrati ons from the West to the South Coasts,
 488

PLANKTON DYNAMICS 4 - II FEBRUARY 1981

DROGUE TRACK

S~~----~----r-----r-----r----'----~~--~----~----~

5nm
35'

40'

45'

50'

55'

Drogue retrieved
1------11 20m/s

I I I I I I I I I I I~ ~'~\\lll"". . . I\I~f-JI/JIJ,I!I!JlJit1,';.v\\\
DAY 1 I DAY 2 I DAY 3 ~
I DAY 41 I DAY 5 I DAY 6 I DAY 7 I
5 6 7 8 9 10 11

50' 55' 5' 10' 15' 20' 25' 30' E 35'

Fig. 6 Track of a drogue set at 10 m depth in newly upwelled

water in Feb. 1981, with wind stick vectors

with distinct minima at the upwelling centres,

and a marked decrease east of Cape Agulhas.
Along the southern Cape coast upwelling is
infrequent and is linked to capes and strong
easterly winds, although the frequency and
biological implications of this upwelling are
not yet known (McLachlan and van der Horst, 1979;
Schuman et al., 1982). Some idea of the food
availability for surf-zone organisms can be
obtained from the distribution of catches of
mullet (harders) around the Cape coast
(de Vi lliers, 1976). These fish are interface
feeders capable of consuming plankton and
detritus at the surface and the bottom in the
surf zone and are likely to compete with white
mussels for the particulate matter present.
Table 2 shows that some 60% of the fish are
landed in the St Helena Bay area, followed by a
further 20% in the False Bay area and many fewer
east of Cape Agulhas. Reasons for this may be
availability or marketability of fish, or wave
exposure limiting seining or perhaps increased
predatory activity of large fish in warmer
waters east of Agulhas, or they may indicate
real changes in the carrying capacity of surf
zones.
 489

PROGRESSIVE VECTOR OF
OLiFANTSBOS
14/3/79 - 29/4/80

10/3

1/3

10

Scale of displ~cement 8

;;-
S
6 x
E
-"
o
IX

4 ~
J:
Ii<
o
2 Z

10 8 6 4 2
WESTWARD (km x 10 4 )

14/3/79

Fig. 7a Wind progressive vector diagram at Olifantsbos from 14/3/79 to 29/4/80

491)

PROGRESSIVE VECTOR OF BOKPUNT

20/3/79 - 28/1/80

28/1

10

C
Scale of displacement IX

4~
:I:

8z
2

10 8 6 4 2
WESTWARD (km x 104 )

Fig. 7b Wind progressive vector diagram at Bokpunt from 20/3/79 to 28/1/80

 491

PROGRESSIVE VECTOR Of STOMPNEUS

22/12/78 - 8/2/79 8/2/79

30/1

20/1

10/1

10

1/1/79
Scale of displocement

o
'"~
....
J:
4 ~
22/12/78 Z

10 8 6 4 2
WESTWARD (km x 10 4 )

Fig. 7c Wind progressive vector diagram at Stompneus from 22/12/78 to 8/2/79

492
PROGRESSIVE VECTOR OF NUWEDAM
2/8/79 - 5/2/80
1/1/80
1/12
1/11
10
Fig. 7d Wind progressive vector diagram at Nuwedam (Elands Bay) from 2/8/79 to 5/2/80
OUTBREAKS OF RED WATER AND MUSSEL MORTALITIES
One consequence of the enrichment due to
upwelling along the West Coast is the occurrence
of outbreaks of toxic dinoflagellates (Grindley
and Sapeika 1969; Grindley and Nel, 1970;
Popkiss et aZ., 1979; Horstman, 1981), which
requi re amongst other conditions a calm stable
water column with little horizontal dispersion
and some form of organic conditioning (Barber,
1973) derived from mature phytoplankton blooms
or river discharge. Because outbreaks of red
water are usually only detected close inshore,
they may be considered as characteristic of the
land-sea interface. However, offshore sightings
of red water by the air force and patrol boats
(SFRI, unpublished data) indicate that outbreaks
5/2/80
1/2
10
Scale of displacement
;;-
S
6 x
1:
Cl
4 ~
r
§
2 z
8 6 4
WESTWARD (km x 10 4 )
can occur at least up to 30 kilometres offshore.
Further, many common coastal diatoms have been
identified in samples collected during red-water
outbreaks (Table 3). These data show that
considerable intermingling of water occurs
between the "surf zone" and the "coastal zone".
Good conditions for normal phytoplankton
development can also lead to conditions
necessary for red-water outbreaks, with dense
phytoplankton close inshore in calm conditions
following upwelling (e.g. Fig. 8 of Horstman,
1981). Ignoring for the moment problems of
seeding and activation of resting spores, it is
in areas such as St Helena Bay, Elands Bay and
Yzerfontein - Melkbos that most toxic outbreaks
occur, and these can have tremendous effects on
 493

TABLE 2
Permi ts, permit-hol ders and harder catches per geographical region
(from de Villiers 1976)

Length of Permits Permit Numbers reported by Percentage

Geograph i ca 1 coastline on ho 1ders permit-ho 1ders of total
(nauti ca 1 record catch
miles) 1975 1975 1974 1975 1974 1975
South West Africa 840 28 25 5 074 55 223 0,1 0,9
Orange River to
Lambert s Bay
I 230 66 55 212 153 98 936 4,1 1,7
Elands Bay to
Langebaan 120 1 043 337 3 235 124 3 655 471 61,9 62,8
Yzerfontein to
Cape Point 60 59 58 220 670 423 051 4,2 7,3
Cape Point to
Kleinmond 60 132 115 967 324 867 300 18,5 14,9
Kleinmond to Agulhas 50 83 74 492 903 597 385 9,4 10 ,3
Agulhas to Still Bay 100 51 43 83 206 111 067 1,6 1,9
Mossel Bay to Knysna 100 8 8 1 615 3 023 0,1 0,1
East of Knysna 400 5 5 8 881 5 273 0,2 0,1
Whole coastline 1 960 1 475 720 5 226 950 5 816 729 100,0 100,0
Purse seine catches:
Elands Bay to Langebaan 700 000 325 000 11,8 5,3
TOTAL 5 926 950 6 141 729

populations of sand mussel (de Villiers, 1974,
1979; Horstman, 1981). The Elands Bay mussel
population has been monitored for 18 years and
examination of the changes are of interest.
In 1974 de Villiers demonstrated a severe
mortality of adult mussels at Elands Bay, caused
by a bloom of Gonyaulax grindleyi (Grindley and
Nel, 1970). Two adult year classes of mussels
were killed, and two juvenile year-classes
higher up the beach survived (Fig. 3 of
de Villiers, 1979). This had a long-term effect
on recruitment of young mussels with two years
of low recruitment alternating with a single
year of good recruitment for a further nine
years, before recruitment increased considerably
(Fig. 4 of de Villiers, 1979). Part of this
delayed recovery may have been due to recurring
mortalities caused by red water in December
1967, December 1968 and September 1974
associated with outbreaks of another toxic
dinofiagellate, Gonyaulax eatanella (Horstman,
1981). De Villiers (1979) stated that the
numerical strength of the population had already
recovered, and that the biomass levels would be
fully recovered to their pre-1966 levels by
1981. However, in March 1980, a further toxic
outbreak of G. eataneZZa caused another
population crash (Fig. 10), so it appears that
good feeding conditions also create some
drawbacks.
These recurring mortalities on the West Coast
may account for the much higher-estimated
mortality rate of West Coast mussels (de Villiers,
1974) compared with the assumed low mortality
494

5
Line
8·

12·

3'1' 16·

20·

24·

33'
8·

32·

36·

34°
40
San Sebastian .: .. :-.. :;.

44

E
Fig. 8 Map of C.E.L.P. grid, showing location of stations sampled monthly in
1977/78 - data from the inshore station on each line (~02), 5 km off=
shore, was used to plot the chll 'a' distribution shown in Fig. 9
rates of adult mussels on East Coast beaches macrophyte-derived material is unimportant on
(Mclachlan and Hanekom, 1979). Underestimation most of the West Coast sandy beaches because of
of the adult population on the West Coast would a limited proportion of rocky coastline compared
also account for this conclusion. with beaches (less than 10 - 20%) and a lack of
washed-up material on most extensive beaches.
DISCUSSION
Despite a lack of research aimed specifically at Surf-zone diatoms appear to be important when
feeding of white mussels, it is clear that there onshore winds persist (Lewin et al., 1975) and
are strong interactions between coastal when cell-like surf circulation patterns are
phytoplankton and surf-zone organisms on the maintained for some time (McLachlan, 1980).
West Coast. Alternative food sources are Blooms of surf-zone diatoms are commonly
detritus from macrophytes (Koop et al., 1982) observed from False Bay to the eastern Cape on
and surf-zone diatoms (Lewin et al., 1975; extensive beaches (McLachlan, 1980; McLachlan
McLachlan, 1980). Bally (1981) suggests that et al., 1981; Bally, 1981). On the West Coast
 495

_9
<?
E ---- Excluding Jan. 1979
cil
.5 8
Z
0
~ 7
Z
'"
....
w
u 6
Z
0
u ,, Agulhas
5
a Columbine
,,
:::I
>-
J: 4
"-
0
~
I 3 Peninsula
u
z
w 2
..(

:l<

12 16 20 24 28 32 36 40 44 48 52 56 60 64 68 72 76 80
C.E.L.P. TRANSECT NUMBER

Fig. 9 Mean chll 'a' concentrations in the upper 20 m inshore around the SW. Cape coast

7
.. - .. Number of mussels (x 1 000)
JL Biomass (shell weight) x 10 kg ..
I
,
"-
.... I "-

'"
I
I
6 I
\
\
I \
I
0 I I
I \
mO

,
\
-" 0 5 1 I
0 -
x __
1 I
\
\
- x I
\
....Iu:l
<J")
4
1
1 I
I
\
\
0<J") I I
_<J")
I I \
w ~ \
1 I
~ :l< I I \
--' u... 3 I I \
u:lO 1 I
\
\
I<J")
'"
W
«l
1
I I
I \
\
:l< 2 I I \

---r/
~ 1 I \
z 1 I \
\
1
" ~---- \
~

~ !*--- /.,e------_e_ / ~

Inn
'ti
~
~ ~ ~ ----t----;" , ~
1966 67 68 69 70 71 72 73 74 75 76 77 78 '79 80 81 '82
YEAR

Fig. 10 Relative population size of mussels in terms of numbers and biomass (three
transects combined) at Elands Bay 1966 - 1982
496

TABLE 3 MARCH 1980. Gonyaulax catanella.

Diatoms associated with red-water outbreaks Elands Bay
(* diatoms common throughout the coastal zone) Co-occurred with Dinophysis accuminata

DECEMBER 1966/JANUARY 1967. Gonyaulax gl'indleyi. Diatom species present

Elands Bay * Chaetocel'os constrictus
5
Diatoms 5 x 10 cells·£-l 10% of total. Pleul'osigma dil'ectum
Gl'ammatophol'a angulosa
Species
Actinoptychus undulatus
* Nitzschia pungens val'.
atlantica
Biddulphia pulchellum * Chaetocel'os didymus
* Chaetocel'os decipiens C. lol'enzianus
Cocconeis s p. * Thalassiosil'a decipiens
Col'ethl'on criophilium
* Coscinodiscus gigas val'. Speci es
pl'aetexta
Coscinodiscus sp.
* Col'ethl'on criophilium
Licinophol'a abbl'eviata
* Pleul'osigma dil'ectum
Navicula sp.
* Chaetocel'os cul'visetus
* Nitzschia closterium
* Nitzschia seriata
* N. delicatissima
* Coscinodiscus pel'fol'atus
* N. pungens Val'. atlantica NOVEMBER 1977. Mesodinium rubrum (non toxic).
Pal'alia sulcata Elands Bay
Planktoniella sol Diatoms 95% of total nos.
Rhizosolenia hebetata form
semispina Species
R. styliformis Skeletonema costatum
* Thalassionema nitzschioides (38 x 10 6 cells·£-l -
Thalassiosil'a sp. highest concentration observed)
* Chaetocel'os compl'essus
DECEMBER 1968. Gonyaulax catanella. * C. l'adicans
Elands Bay - Lamberts Bay * C. debilis
Diatoms 1 x 10 6 cells·£-l 50% of total. * C. didymus
Species * C. affinis
* Skeletonema costatum * Thalassiosil'a nol'denskoldii
* Nitzschia sel'iata * Nitzschia pungens Val'.
* Chaetocel'os spp. atlantica

* Astel'ionella japonica MAY 1978. Gonyaulax catanella.

Stephanopyxis tul'l'is Saldanha Bay - Blaauwberg
* Thallasionema nitzschioides Diatoms "common" in sample.
* Nitzschia clostel'ium
Species
SEPTEMBER 1974. G. grindleyi. * Skeletonema costatum
Elands Bay * Thalassiosil'a decipiens
Diatoms 5% of total. * Nitzschia seriata

* Rhizosolenia alata
* Chaetoceros lorenzianus
* Coscinodiscus radiatus
* C. centralis
no surf-zone diatoms have been identified.
Although they could have been missed by being
unfamiliar it is likely that the preponderance
of coastal diatoms would obscure their presence.
In addition, Hutchings and Nelson (in prep.)
show that winds favourable for upwelling persist
north of Yzerfontein, a fact which would limit
the development of extensive blooms of surf-zone
diatoms (Lewin et al., 1975; McLachlan, 1980).
It would appear that surf-zone diatoms are a
more important food source east of Cape Point
than along the West Coast.
A third source of food is organic matter
discharging from rivers. McLachlan (1977)
postulates that the very high biomass of mussels
on Maitland River beach may be the result of the
supply of organic matter from three rivers in
the vicinity, and higher mussel populations
occur east of river mouths than elsewhere on
east Cape beaches. On the West Coast, river
discharge is minor for most of the year and this
source of food can be discounted.
The very high biomass levels on Maitland River
beach appear somewhat anomalous compared with
other beaches in the eastern Cape and it is
possible that the food supply to this beach is
particularly rich and consistent. Schuman et
al. (1982) demonstrate upwelling off Cape
Recife, with the tongue of cool water extending
westwards into St Francis Bay, which may enrich
these waters considerably. Therefore a
combination of surf-zone diatoms, riverine
organic matter and phytoplankton from upwelling
events, coupled together with a nearshore
circulation system, produce a mussel biomass
497
which appears to be higher than most other
reported beach faunas worldwide (McLachlan, 1977;
Bally, 1981). These biomass values are only
equalled by the recent measurements of adult
mussel biomass at Elands Bay, Britannia Bay and
Paternoster (this report), on the Cape west
coast, and exceeded by Emerita biomass in Peru
(Penchaszadeh, 1971), interestingly both in
upwelling zones. On this basis one would expect
a very high biomass of filterfeeders on
upwelling zone beaches in N.W. Africa as well.
One cannot neglect the importance of the coastal
zone as a potential source of particulate matter
for eastern Cape beaches. Algoa Bay appears to
be a productive area and Zoutendyk (1970)
demonstrates high zooplankton volumes in the
area. Lutjeharms (1981b) describes edge effects
which begin at Port Elizabeth as the Agulhas
Current diverges from the coast, which may be
responsible for nutrient enrichment of the near-
surface coastal waters. The abundance of
anchovy, pilchard and gannets also suggest a
rich pelagic zone in the Algoa Bay region
(Batchelor, 1982).
The variation in the size distribution of mussels
on eastern and western Cape beaches is more
puzzling. Ansell (1981) demonstrates that
D. serra in the eastern Cape is better adapted
morphologically to intensive bouts of feeding on
high food concentrations than D. sordidus, which
lives lower down in the intertidal zone. Yet,
on the West Coast, D. sordidus is absent and
adult D. seY'L?Q occupy thi s zone, where food
availability is consistently high. McLachlan
and Hanekom (1979) consider the mid-tide
distribution to be caused by the inability of
D. serra to maintain itself in a rapidly eroding
substrate lower down the shore in the eastern
Cape, yet on the West Coast, where similar wave
action prevails, large mussels are extremely
common.
 498
One can speculate that very high concentrations
of particulate material are required by D. serra
to meet the energy requirements of filtering as
well as to provide scope for growth and
reproducti on. On eastern .Gape beaches, where
temperatures are higher, this requirement may
impose further stress on the animals. If
particulate food is scaroe on southern and
eastern Cape beaches, any physi cal concentrati on
of food particles would favour the net energy
intake of mussels. Sandy beaches are known to
filter large volumes of water (McLachlan, 1979)
resulting in particulate matter being trapped in
the upper layers of sand, parti cul arly as the
tide ebbs. On the rising tide, much of this
. material would be resuspended, enriching the
swash zone considerably. At a mid-tide level,
optimum feeding conditions would arise as a
compromise between food concentration in the
swash zone and time for filtering.
Much experimentation on the scope for growth of
juvenile and adult mussels at different
temperatures and food concentrations is clearly
needed to explain differences in mussel
distribution. Predation pressure may also
impose different distribution patterns, but
little is known about mortality rates of
different aged mussels in the eastern and
western Cape.
Sand mussels may also occur beyond the surf zone
in the larval stages, which may last for 2 - 3
weeks (de Vi11iers, 1973, 1974; McLachlan and
Hanekom, 1979), with a lag of some months
between the peak in reproductive activity and
the appearance of young mussels in the
intertidal zone. De Villiers (1974) showed
that, at very low parent stocks, recruitment at
Elands Bay was much reduced, indicating very
localized seeding ability of mussel populations
i.e. a very short pelagic phase with little
"coastal" influence. At high parent stocks,
however, there was considerable variability in
recruitment, indicating strong environmental
influence. Knowledge of the dispersal and
duration of the pelagic phase of sand mussel
larvae may indicate how much interlinking there
is between coastal and surf zones.
CONCLUSIONS
Very high mussel popUlations are apparent on
eastern Cape beaches (in particular, Maitland
River beach) and along the Cape west coast, with
much lower populations on beaches in between.
On the Wes t Coast upwell i ng provi des a ri ch
source of coastal phytoplankton which is
transported onshore by countercurrents, wind
reversals and recurvature and by readjustments
of sea level. On the east Cape beaches a
combination of river discharge, surf-zone diatoms,
infrequent upwelling events and possibly edge
effects of the Agulhas Current combine to
provide a rich source of particulate organic
matter on certain beaches. On South Coast
beaches indications are that large mussel
popUlations cannot develop because of a limited
supply of particulate matter, which could be at
least three times less than on the West Coast.
Beaches in the Algoa Bay region appear to be
more "independent" than West Coast beaches,
largely because of the lack of upwelling-
favourable conditions on this coastline. West
Coast beaches have an extremely rich mussel
biomass over extensive areas owing to rich food
supply, but these populations are subject to
periodic mass mortalities caused by toxic
dinoflagellates, which are also linked to the
upwelling system.
REFERENCES
ANDREWS, W.R.H. and L. HUTCHINGS (1980).
Upwelling in the southern Benguela Current.
Prog. Oceanogr. 9(1): 1-81.
ANSELL, A.D. (1981). Functional morphology and
feeding of Donax serra Roding and Donax

sordidus Hanley (Bivalvia: Donacidae).
J. moZZusc. Stud. 47: 59-72.
BALLY, R. (1981). The ecology of three sandy
beaches on the west coast of South Africa.
Unpublished Ph.D. Thesis, University of
Cape Town: 404 pp.
BARBER, R.T. (1973). Organic ligands and
phytoplankton growth in nutrient-rich sea
water. In: Trace MetaZs and MetaZ-Organic
Interactions in NaturaZ Waters. Singer,
P.C. (ed.). Ann Arbor Scientific
Publications: 321-338.
BARBER, R. T. and R.L SMITH (1981). Coastal
Upwelling Ecosystems. In: AnaZysis of
Marine Ecosystems. Longhurst, A.R. (ed.).
Academic Press: 31-68.
BATCHELOR, A.L. (1982). The diet of Cape
Gannets SuZa capensis breeding on Bird
Island, Algoa Bay. Unpublished M.Sc.
Thesis, University of Port Elizabeth.
DE VILLIERS, G. (1973). Reproduction of the
white sand mussel Donax serra Reding.
InvestZ Rep. Sea Fish. Brch 102: 33 pp.
(Afrikaans 1973; English 1974).
DE VI LUERS, G. (1974). Growth, popul ati on
dynamics, a mass mortality and arrangement
of white sand mussels Donax serra Reding
on beaches in the south-western Cape
Province. InvestZ Rep. Sea Fish. Brch
109: 31 pp. (Afrikaans 1974;
English 1975).
DE VILLIERS, G. (1976). Exploitation of harders
along South African shores. S. Afr. Shipp.
News Fishg Ind. Rev. 31(5): 57, 59, 61.
DE VILLIERS, G. (1979). Recovery of a
population of white mussels Donax serra at
Elands Bay, South Africa, following a mass
mortality. Fish. BuZZ. S. Afr. 12: 69-74.
499
GRINDLEY, J.R. and E.A. NEL (1970). Red water
and mussel poisoning at Elands Bay,
December 1966. Fish. BuZZ. S. Afr.
6: 36-55.
GRINDLEY, J.R. and N. SAPEIKA (1969). The cause
of mussel poisoning in South Africa. S.
Afr. Med. J. 43: 275-279.
HORSTMAN, D.A. (1981). Reported red-water
outbreaks and their effects on fauna of the
west and south coasts of South Africa, 1959
- 1980. Fish. BuZZ. S. Afr. 15: 71-88.
HUTCHINGS, L. and G. NELSON (in prep.).
Envi ronmenta 1 factors affecting pel agi c
fish populations in the southern Benguela
region. Proceedings of the BEP Anchovy
Workshop, August 1982, Cape Town.
KOOP, K., NEWELL, R.C. and LUCAS, M.I. (1982).
Biodegradation and carbon flow based on
kelp debris (EckZonia maxima) in a sandy
beach microcosm. Mar. EcoZ. Frog. Ser.
LEWIN, J., T. HRUBY and D. MACKAS (1975). Blooms
of surf zone diatoms along the coast of the
Olympic Peninsula, Washington. ·5.
Environmental conditions associated with
the blooms. Estuar. coastZ. mar. Sci.
3: 229-241.
LUTJEHARMS, J.R. E. (1981a). Satellite studies
of the South Atlantic upwelling system. In:
Oceanography from Space. Gower., J. F. R.
(ed.). Mar. Sc. CoL. 13, Plenum Press,
New York.
LUTJEHARMS, J.R.E. (1981b). Features of the
southern Agulhas current circulation from
satellite remote sensing. S. Afr. J. Sc.
n: 231-236.
MCLACHLAN, A. (1977). Composition, distribution,
abundance and biomass of the macrofauna and
meiofauna of four sandy beaches. zooZ.
500
Afr. 12(2): 279-306.
MCLACHLAN, A. (1979). Volumes of sea water
filtered through eastern Cape sandy beaches.
S. Afr. J. Sci. 75: 75-79.
MCLACHLAN, A. (1980). Exposed sandy beaches as
semi-enclosed ecosystems. Mar. environ.
Res. 4: 59-63.
MCLACHLAN, A. and N. HANEKOM (1979). Aspects of
the biology, ecology and seasonal
fluctuations in biochemical composition of
Donax serra in the East Cape. S. AIr. J.
Zool. 14: 183-193.
MCLACHLAN, A. and G. VAN DER HORST (1979).
Growth and reproduction of two molluscs
from an exposed sandy beach. S. Afr. J.
Zool. 14: 194-201.
MCLACHLAN, A., T. WOOLDRIDGE and A.H. DYE (1981).
The ecology of sandy beaches in southern
Africa. S. Afr. J. Zool. 16: 219-231.
NELSON, G. and L. HUTCHINGS (in prep.). The
Benguela Current region. Frog. Oceanogr.
PENCHASZADEH, P. E. (1971). Pre 1i mi na ry
quantitative observations of sandy beaches
of the Peru central coast with special
reference to Emerita analoga populations
(Crustacea, Anomura, Hippidae). Contr.
Inst. Biol. mar., Mar del Plata 117: p.16.
POPKISS, ~1.E.E., D.A. HORSTMAN and D. HARPUR
(1979). Paralytic shellfish poisoning: a
report of 17 cases in Cape Town. S. Afr.
Med. J. 55: 1017-1023.
SCHUMAN, E.H., L.A. PERRINS and I.T. HUNTER
(1982). Upwelling along the Cape south
coast. S. Afr. J. Sci. 78: 238-242.
SHANNON, L.V., G. NELSON and M.R. JURY (1980).
Hydrological and meteorological aspects of
upwelling in the southern Benguela Current.
In: Coastal Upwelling. Richards, F.A. (ed.).
Coastal and Estuarine Sciences I: 146-159.
TROMP, B.B.S. and D.A. HORSTMAN (in prep.). A
seasonal study of the western Agulhas Bank
region.
ZOUTENDYK, P. (1970). Zooplankton density in
the south western Indian Ocean.
Unpublished pape~ CSIR Symposium
Oceanography in South Africa, Durban,
Aug. 1970.

THE IMPACT OF SURF-ZONE FISH COMMUNITIES ON FAUNAL ASSEMBLAGES ASSOCIATED WITH SANDY BEACHES
T.A. LASIAK (Department of Zoology, University of Transkei, Umtata)
INTRODUCTION
Studies on fish assemblages associated with
surf-exposed beaches are scarce. The most
notable (Pearse et al 1942; Gunter 1945, 1948;
Carlisle et al 1960; McFarland 1963; and
Schaefer 1967) in common with other studies on
nearshore fish assemblages, have concentrated
on "community structure". Cailliet, et al
(1978) pointed out that few attempts have been
made to stress the functional relationships
between fish assemblages and their food sources.
Based on the estimation of assimilation and
importation of organic carbon to a surf beach at
Mustang Island, Texas, McFarland (1963) related
productivity of plankton to fish biomass.
The present paper synthesizes data from studies
of "community structure It of the surf-fish
assemblage at King's Beach, Algoa Bay with in-
formation on the feeding habits of the
assemblage (Lasiak 1982). The impact of the
surf-fish assemblage on other biotic components
of the beach/surf ecosystem is discussed in
terms of energy flow and transformation, as a
result of predation.
METHODS
The surf-zone fish assemblage at King's Beach
was sampled monthly between September 1978 and
October 1980, by means of seine netting.
Details of the procedure are given in Lasiak
(1982). Based on the monthly catch statistics,
biomass and energy equivalents the standing crop
of each species was estimated in terms of
501
-3 -1
kJ m y . The feeding habits of each species
were studied by means of wet and dry gravimetric
analyses of stomach contents. The frequency of
occurrence of prey items was also noted.
Attempts were made to obtain laboratory
estimates of food consumption rates. However,
the dynamic nature of the fish assemblage meant
that considerable resources were required to
obtain sufficient animals for experimentation at
regular intervals. Experimental studies there-
fore had to be abandoned. This Itdynamicismlt
also affected the success of 24h studies in
providing data on the quantity of food con-
sumed by each species on a daily basis. To
overcome this problem the minimum estimates of
daily food intake rates are based on the
maximum feeding intensities (stomach contents
are expressed as a percentage of the total body
mass on both a dry and wet mass basis) observed
during the course of the study. The annual
food intake is given by:
Total intake = 365 x M x F (P 1 + P2 + P3 + •. Pn )
100
where: M is the mean biomass of each trophic
group.
F is the maximum feeding intensity
(P . ••••• P ) is the average percentage
1. n
contribution of each prey item to the
overall diet of that trophic group.
Using calorific values from Joubert and
Hanekom (1980) the contribution of each
was expressed in kJ m- 3 y-1.
502

RESULTS AND DISCUSSION The 59 species of fish recorded were divided

Infaunal beach organisms made only a small con- into six trophic groups based on their feeding
tribution (12% dry mass) to the food base habits. The majority of species consumed a
utilized by the fish. Motile organisms, in the wide variety of prey items. Opportunism was
form of mysids, prawns and other fish, con- prevalent which suggests that many species are
stituted the primary food source (58,6% dry mass) capable of utilizing IIsuperabundantll prey items
(Fig. 1). An additional food source was the when they are available. The King1s Beach
epi- fauna and flora associated with the nearby ichthyofauna was dominated either by plankti-
harbour wall and scattered offshore reefs. This vores or benthic feeders. McFarland (1963)
accounted for 15,5% of the food consumed. found that the most abundant species at Mustang
Island were planktivorous.

o 5 10 15 20

FISH/CEPHALOPODA
MACROPETASMA
MOTILE FAUNA MYSIDS
MEGALOPAS
SMALL CRUSTACEA
MOLLUSCA
EPI FAUNA ECHINODERMS
CRUSTACEA

FLORA ALGAE
POLYCHAETES
MOLLUSCA
INFAUNA ECH INODERMS
MINOR PHYLA
CRUSTACEA
OTHERS UNIDENTI FIE D
SAND

FIGURE 1 Total trophic spectra for King's Beach. Values are the total percent by mass
(wet) of each prey category for all fish species.

Food consumption estimates.
The major representatives of each trophic group
plus details of their observed feeding
intensities are given in Table 1. The maximum
feeding intensities observed amongst benthic
feeders, detritivores, herbivores, omnivores,
piscivores and planktivores were 3,06%; 4,65%;
5,32%; 7,17%; 8,45%; and 10,86% respectively.
Use of these values as an indication of the
minimum daily.intake rate is subject to
criticism. However, Pandian (1967) obtained
a maximum daily consumption rate equivalent to
5% of total body mass in Megalops cyprinoides
under experimental conditions. Being of a
similar order of magnitude the use of maximum
feeding intensities as an estimate of minimum
daily intake may be justified.
Comparison of standing crop and food intake
estimates.
Figure 2, based on Odum1s (1975) energy circuit
language, represents the estimated energy flow
from the major prey items to each trophic group.
Only food items contributing to more than 1% of
the food base are indicated.
Beach macrofaunal production has been estimated
6 8 kJ m-2 y -1 at King1s Beach ( McLachlan
at 10,5
1977). Preliminary estimate of sublittoral
macrofaunal production for King1s Beach is
approximately 61, 25kJ m-2 y-1 (A. Cockroft,
personal communication). Feeding habit studies
indicated that benthic feeders consume a
-2 -1
minimum of 69,78 kJ m y. Thus the loss of
macrofaunal production by fish predation amounts
to 83,2% of the total production.
It has been estimated that the p/B ratio for the
mysid, Mesopodopsis slabberi, is unlikely to
exceed eight, based on this ratio and the
standing crop .. estimate (T. Wooldridge, personal
communication) predation by surf-zone fish re-
moves at least 3,32 times the ~. slabberi
production annually. Similarly the p/B ratio
for the swimming prawn Macropetasma africana is
503
unlikely to exceed four (A. Cockroft, personal
communication) • On this basis it would appear
that the surf-fish assemblage consumes 96,2% of
the available production.
McFarland (1963) concluded that the recruitment
of plankton from net productivity and from im-
portation was at least several orders of
magnitude in excess of the food requirements of
the surf-fish at Mustang Island, Texas. He
also pointed out that beach fauna would make
only a small contribution to the food base,
since they probably represented less than 5 -
10% of fish biomass.
Differences in the efficiency of sampling
methods used to estimate both fish and plankton
stocks may account for the anomalies between
estimates of consumption and annual production
of prey items at King1s Beach. However, the
patchy occurrence and immigration of motile
organisms are likely to be the major factors
involved. Estimates of both food consumption
rates and standing crop will be influenced by
the patchy distribution of both predators and
prey. Considerable fluctuations were noted in
the contribution of five major prey items to the
total food base utilized by surf-fish throughout
the year (Table 2).
The large variation (0 - 96%) in the contri-
bution made by the mysid~. slabberi may be
attributed to the presence or absence of large
shoals in the sampling area. On occasion the
biomass of M. slabberi just outside the surf
- -3
reached 1,9g.m whereas the mean value used in
estimating annual production was only 0,135.m- 3
(T. Wooldridge, personal communication).
Differences in the proportional representation of
the trophic groups also affects the contribution
of various prey items to the food base. This
variation reflects the dynamic nature of the
surf-fish assemblage. The two prominent trophic
504

TABLE: 1

Number of stomachs examined, frequency of occurrence of empty stomachs and feeding intensity of the
dominant species from each of the trophic groups represented at Kings Bea.ch.

%empty FEEDING INTENSITY

TROPHIC GROUP SPECIES N
stomachs min. max. mean

BENTHIC FEEDERS Lithognathus lithognathus 88 31 ,8 0,005 3,06 0,40

Lithognathus mormyrus 118 32,2 0,003 2,37 0,38
Umbrina capensis 99 23,2 0,003 2,47 0,34
DETRITIVORES Liza richardsoni >20 g 273 63,4 0,004 7,01 1,09
HERBIVORES Sarpa salpa >5 g 361 8,9 0,004 4,65 1,45
OMNIVORES Diplodus sargus >10 g 358 15,6 0,003 7,17 0,83
Rhabdosargus holubi 65 38,5 0,01 2,61 0,43
PISCIVORES Argyrosomus hololepidotus 147 27,9 0,01 5,39 0,67
Pomatomus saltatrix 89 36,0 0,02 8,45 0,87
PLANKTIVORES Diplodus sargus < 10 g 184 9,8
Liza richardsoni ( 20 g 89 33,7
Lithognathus mormyrus ( 10 g 217 7,8
Monodactylus falciformis 299 28,4 0,005 5,21 0,31
Pomadasys olivaceum 729 24,0 0,01 10,86 1,66
Rhabdosargus globiceps ( 10 g 150 29,3
Sarpa salpa ( 5 g 113 6,2
Trachurus capensis < 10 g 119 41,2

SURF ZONE

KINGS BEACH

WES EXPORT

ALGOA lAY
~
FIGURE 2 En~rgy circuit diagram illustrating fish predation on major food categories.
Values in kJ/m /yr. 3
 505

TABLE 2
Variability in the percentage contribution of five major prey items to the diet of the fish assemblage.

%Contribution
Prey Item
x s.d. min. max.

Meso)2odo)2sis slabberi 9,6 21,8 0,0 96,1

Macro)2etasma africana 12,5 18,4 0,0 59,0
Algae 21,9 42,7 0,0 92,7
Beach macrofauna 3,4 5,4 0,0 24,2
Fish 21,8 20,2 0,0 65,7

groups, benthic feeders and planktivores, showed

the greatest variability (Table 3).

TABLE 3
Variability in the proportional representation of the five major trophic groups within the fish
assemblage.

Biomass
Trophic Group x s.d. Coeff. of Var.
(g) (g) (g)
Piscivores 2811,7 2507,7 522,9
Benthic feeders 18640,0 28812,1 5650,5
Planktivores 7551,1 7881,5 1545,7
Omnivores 1752,7 2177,5 427,1
Herbivores 2601,8 6516,9 1278,1

The impact of surf-zone fish on other biota.
The fate of energy and matter gained from prey
items ingested needs to be examined in order to
assess the impact of predators. Depending on
the residence time, diel activity patterns and
digestion rates of the predator, material lost
in faeces and excreted as nitrogenous wastes may
be lost or retained in the sandy beach/surf zone
system. McLachlan et al (1981) suggest that as
a result of turbulence and the absence of sedi-
mentation of organic matter, all macrofauna
faeces and pseudofaeces are immediately returned
to the detritus pool in the surf zone. This
will also apply to fish remaining in the environ-
ment for a sufficient period of time for
digestion and evacuation of wastes to take place.
A species such as the herbivore Sar)2a sal)2a,
which moves into the surf zone at night for
shelter (Lasiak, 1982), may contribute directly
to the detritus pool by bringing in material
from outside the system (ie. reefs where they
feed).
Winberg (1956) considers the energy content of
waste products to have an average value equi-
valent to 20% of that of the food consumed.
Consequently the fish assemblage may have a
significant effect on the recycling of energy
and material within the beach/surf ecosystem.
Energy going into reproduction and growth may
eventually be lost. This is a result of high
larval mortality, the stochastic processes
506

determining the distribution of surviving off-

spring and migration patterns. Energy put into
activity, digestion and metabolism will be lost
to the system as heat.
The relationship of the surf-zone fish
assemblage to other biotic components within the
beach/surf ecosystem clearly revolves around
their role as energy transformerso The in-
stability of the surf-fish "community structure"
suggests that a great deal of energy and material
gained by predation within the surf-zone will be
ultimately exported. The residence time of
individual fish has a significant .effect on
macrofaunal and planktonic production.
Elsevier, Amsterdam.
Pandian TJ (1967) Transformation of food in
the fish Megalops cyprinoides II. Influence
of quantity of food, Mar. BioI. 1, 107-109.
Pearse A, Humm HJ and Wharton GW (1942)
Ecology of sand beaches at Beaufort, North
Carolina, Ecol. Monogr. 12, 135-190.
Schaeffer RH (1967) Species composition, size
and seasonal abundance of fish in the surf
waters of Long Island, N.Y. Fish Game J. 1,
1-46.
Winberg T (1956) Rate of metabolism and food
requirements of fishes, Fish. Res. Bd. Canada
Transl. Ser. No 194, 1960.

REFERENCES
Cailliet G, Antrim BS and Ambrose DS (1978)
Trophic spectrum analysis of fishes in Elkhorn
Slough and nearby waters. In Fish Food Habit
Studies 2nd Pac. Northwest Workshop Tech.
pp. 1 Hl-128.
Carlisle JG, Schott JW and Abramson NJ
(1960) The barred surf perch in Southern
California, Calif. Dept. Fish. and Fish. Bull.
109.
Gunter G (1945) Studies on marine fishes of
Texas, Publ. Inst. Mar. Sci. Univ. Texas 1,
1-190.
Gunter G (1948) Population studies of the
shallow water fishes of an outer beach in
Southern Texas, Publ. Inst. Mar. Sci. Univ.
Texas 5, 186-193.
Joubert CSW and Hanekom PB (1980) A study of
feeding in some inshore reef fish of the Natal
coast, South Africa, S. Afr. J. Zool. 15,
262-274.
Lasiak TA (1982) Structural and functional
aspects of the surf-zone fish community in the
Eastern Cape, PH. D. thesis, Univ. Port
Elizabeth.
McFarland WN (1963) Seasonal change in the
number and the biomass of fishes from the surf
at Mustang Island, Texas, Publ. Inst. Mar. Sci.
Univ. Texas 9, 91-105.
McLachlan A (1977) Composition, distribution,
abundance and biomass of the macrofauna and
meiofauna of four sandy beaches, Zool. Afr. 12,
179-206.
McLachlan A, Erasmus T, Dye AH, Wooldridge T,
van der Horst G, Roussouw GJ, Lasiak TA and
McGwynne L (1981) Sand beach energetics: an
ecosystem approach towards a high energy
interface, Estuar. cstl. Shelf Sci. 13, 11-25.
Odum HT (1975) Marine ecosystems with energy
circuit diagrams. In Nihoul JCJ, ed.
Modelling of marine systems pp. 127-151.

ECOLOGICAL STRUCTURE AND ENERGY REQUIREMENTS OF THE SANDY BEACH AVIFAUNA OF SOUTHERN AFRICA
P.A.R. HOCKEY, W.R. SIEGFRIED, ANNA A. CROWE and J. COOPER (Percy Fitzpatrick Institute of African
Ornithology, University of Cape Town, Rondebosch 7700, South Africa)
1. INTRODUCTION
Shorebirds, primarily waders of the sub-
Order Charadrii, are conspicuous and
common members of the intertidal fauna of
sandy beaches in many parts of the world
(Hale 1980). However, in general, the
potential importance of the birds as
predators and as distributors of nutrients
is often ignored or overlooked by
investigators of sandy beaches as
ecosystems (e.g. Ansell et al. 1978)
Edwards et al. (1982) draw attention to
the effects on structure of intertidal
communities of such highly mobile,
intermittent predators as crabs, fishes
and birds, and remark that they (the
effects) "are hard to demonstrate and easy
to overlook or misinterpret, but assuming
them to be unimportant may perpetuate
oversimplified models of community
interactions".
Sandy beaches dominate the marine littoral
of southern Africa, accounting for some
68~ of the 4 000 km coastline (Fig. 1).
Sufficient information on the
distribution and abundance of the coastal
avif~una of this region is now available
for a preliminary assessment to be made
of some of the ecological roles of the
birds in these ecosystems. This paper
presents a quantitative description of
the spatial and temporal distribution,
and abundance, ecological structure and
507
energy requirements of the avifauna of
sandy beaches from northern Namibia to
northern Natal.
2. STUDY AREA AND METHODS
The study area extends from the Cunene
River (17° 15'S, 10° 53'E) in the west to
Kosi Bay (26° 52'S, 32° 53'E) in the east.
The coastline was divided into 21 sections,
each of approximately 200 km. The
relative proportions of sand and rock in
each section were determined from
1:50 000 topographical maps of the South
African Department of Surveys and Mapping
( Fig. 1).
Information on the distribution and
abundance of birds on beaches was
obtained during a series of surveys
conducted by the Percy FitzPatrick
Institute of African Ornithology and the
western Cape Wader Study Group between
1978 and 1982. Surveys were made during
summer when maximum numbers of migrant
shorebirds are present (pringle, Cooper
1977) . Censuses were carried out at
low tide, either from a moving vehicle
or on foot. Habitat descriptions were
made at the time. No surveys were
carried out in soutl;1ern Namibia between
Sandwich Harbour (23° 10'S, 14° 15'E) and
the Orange River (28° 20'S, 16° 15'E),
since access to this diamond-rich area
is normally prohibited.
508

150 300km
I I

••

FIGURE 1. A map of the southern African coast, including the 200-km sections (numbered
1-17) used in this study, the relative proportions of sand (open circles) and rock
(shaded) in each section, and the relationship between the proportion of sand and
latitude South.

Twentyfive species of shorebirds (Table 1),
dependent largely or entirely on the
intertidal zone for food, were used
analyses, including five species of
resident non-waders, five species of
resident waders and 15 species of migrant
waders. All of these species prey on
intertidal invertebrates. However,
certain species, such as the Whitefronted
Plover (see Table 1 for scientific and
vernacular names of species), also eat
terrestrial invertebrates, and gulls also
feed by scavenging.
Density, biomass and energy requirements
of the birds are expressed according to
length (km) of shoreline. Walsberg's
(1980) equation was used to calculate the
birds' daily energy expenditure (DEE).
in Bird weights were taken from Hockey (1981)
for the African Black Oystercatcher and
from Siegfried (1981) for other species.
Seasonal variation in avian abundance was
computed for the following areas: the
southwestern Cape (A.A. Crowe,
unpublished), the eastern Cape
(McLachlan et al. 1979) and Natal
(Joubert 1981).
Correspondence analysis (Greenacre 1978)
was used to analyse ecological structure,
the original suite of 25 species being
reduced to 10. Rare and erratically
distributed species, as well as certain
TABL~ 1, Mean + S,D, densities of shorebiras (birds Km- 1 ) in four geographical areas of the southern African
coast

Geographical area 17 200-km sections Northwest Southwest Southeast Northeast

along the coast (1-4) (5-9) ( 10-13) (14-17)

African Black Oystercatcher Haematopus moquini <0,01:::0,01 0,88+0,81 0,88+0,44

European Oystercatcher H, ostralegus <0,01:::0,01
Turnstone Arenaria interpres 2,80:::2,65 2,92:::2,79 0,14:::0 ,10 0,13:::0,21
Ringed Plover Charadrius hiaticuZa 0,05:::0,06 0,18:::0,38 O,16:!:O,30 0,06:::°,07
Whitefronted Plover C, marginatus 2,59:::1,62 5,48+1.84 1,95+0,83 1,64:::0,35
Kittlitz's Plover C, pecuarius 0,04:::0,04 0,01:::0,02
Threebanded Plover C, tricoZZaris 0,01:::0,01 <0,01:::0,01
Great Sandplover C, leschenaultii <0,01:::<0,01
Grey Plover Pluvialis squatarola 1,62+0,97 0,41:::0,27 0,33:::0,53 0,01:::0,01
Curlew Sandpiper Calidris ferruginea 0,54:::0,40 4,53+2,96 0,02:::0,02 0,05:::0,09
Little Stint C. minutus 0,03+0,03 0,06:::0,09 O,Ol:!:O,03
Knot C. canutus 0,52+0,52 0,14:::0 ,15 0,02:::0,03
Sanderling C. alba 17,18:::5,41 15,99+8,27 4,22+3,59 3,19+1,96
Ruff Philomachus pugnax 0,31:::0,66
Common Sandpiper Tringa hypoZeucos 0,01+0,01 0,02:::0,02 0,03+0,04 0,04:::0,04
Greenshank T. nebuZaria 0,01:::0,01 0,05+0.03 0,02:::0,02 0,03:::0,04
Bartailed Godwi t Limosa Zapponica 0,06+0,06 <0,01:::<0,01
Curlew Numenius arquata <0,01:::<0,01
Whimbrel N. phaeopus 0.09+0,16 0,15:::0,19 O,03:!:O,Ol 0,80:::0,06
water Dikkop Burhinus vermiculatus <0,01:::0,01 <0,01:::<0,01
Kelp Gull Larus dominicanus 2,20:::3,10 13,98:!:12,60 3,48::: 1 ,81 0,42:::0,49
Greyheaded Gull L. cirrocephalus <0,01:::< 0,01 0,05:::0,09 0,01:::0,02 0,20:::0,35
Hartlaub's Gull L. hartlaubi 0,45:::0,67 4,67:!:4,19
Pied Wagtail MotaciZZa aguimp <O,Ol:!:O,Ol
Cape Wagtail M. capensis 0,01:::0,01 0,98:::0,54 0,30:::0,27 0,08:::0,08

en
o
1.0
510

.
species not believed to be obligate radical fluctuations.
intertidal foragers, were excluded. Four NW : I SW i BE I HI
COAST I : COAST I COAST 1 cO •• y
principal feeding classes were recognized 1 2 3 4: '5 6 7 B 9:10111213'141
i • • 11
,, ,,
[ _ J _

-------
,, ,
AFRICAN BLACK OVSTeRCATCtER
on the basis of bill length and eUROPEAN OVSTERCATCHER I SCALE ,

~ [" :
behaviour: deep probers; African Black :
TURNSTONE
,- ~ I
Oystercatcher, Curlew Sandpiper and RINGED PLOVER
--- I birds per
,km
i • ----1 --1-1,--
Whimbrel, medium probers; Grey Plover WHITEFRONTED PLOVER
,, I , ~
and Knot, shallow probers; Sanderling,
K,TTLlTZ'S PLOVER
THREEBANDED PLOv,eR ,,, '-
, '----! ,
:, :-1
f----- , ,
GREAT SANDPLOVER
and surface peckers; Turnstone, Ringed ......,
GREY PLOVER
, ~
,
Plover, Whitefronted Plover and Cape CURLEW SANDPIPER
~ • • .J
I r - :
--I--
Wagtail. Invertebrate production was LITTLE STINT
KNOT
- - - ', ,
f--------
,----- ---,
,---'
, '
------.J
derived from published and estimated ,I ,I
,,,
:
, ,
PIB ratios of individual species or ,
classes of invertebrates. Somatic PIB, : '
rather than reproductive PIB, was used
SANDERLING
RUFF ,,, ,,
for Donax spp. COMMON SANDPIPER
---, -,,
GREENSHANK -- , , ---l
,,
BARTAILEO GODWIT -----1
,, ,
L-
CURLEW
,
I-- ,,,
3. RESULTS AND DISCUSSION WHIMBREL ~
,, ,
3.1. Geographical distribution, species
WATER DIKKOP
,, ----1
,
,
,, ,
,
,,,
richness and diversity ,
,
Based on the geographical distribution of , ,,
,,
avian species and their abundance, at
least two major zoogeographical zones
KELP GULL
GREYHEADED GULL
.-JI,, -----J
,,
may be visually distinguished
Table 1):
(Fig. 2,
a west coast zone (sections
HARTLAUB'S GULL
PIED WAGTAIL
......:,
,
,
~
,
,
,
,
1-9) and a south and east coast zone CAPE WAGTAIL ,,
, I - I
(sections 10-17). The west coast zone
may be further subdivided into northwest
(sections 1-4) and southwest (sections
FIGURE 2. Mean densities (birds km- 1
5-9) sectors. Key species in
of shoreline) of 25 species of shorebirds
distinguishing these sectors are African on sandy beaches in 17 200-km sections
around the southern African coast.
Black Oystercatcher, Grey Plover, Curlew
Sandpiper and Hartlaub's Gull. The
south and east coast zone apparently Most species were most abundant in the
also contains two sectors, with species southwest sector (Fig. 3). Densities
richness being greater in the southeast of all species of resident waders peaked
sector (sections 10-13) than in the in the two southern sectors, whereas an
northeast sector (sections 14-17). appreciable proportion (27 %) of migrant
Species richness generally is greater in waders occurred at highest density in the
the west coast zone (Fig. 2). Species northwest sector. Fifteen of the 25
diversity, as determined by the Shannon species (five residents and 10 migrants)
Index of "General Diversity (Shannon 1949), did not exceed a mean density of 0.5 birds
follows a similar pattern but with fewer km- 1 of shoreline in any of the four
 511

sectors (Table 1).

RESIDENT MIGRANT
100
.~ MIGRANT WADERS

o
POSSIBLE GEOGRAPHICAL TRENDS IN

IC
BIRD DENSITY AROUND THE SOUTHERN
AFRICAN COAST
RESIDENT WADERS

NONE NONE
~ RESIDENT NON-WADERS

NW SW SE NE

I~
NONE COMMON SANDPIPER

NW SW SE NE
NW SW SE NE
GEOGRAPHICAL AREA TURNSTONE

I~
GREY PLOVER
KNOT
NONE
SANDERLING
BARTAILED GODWIT

FIGURE 3. The relative frequency of 1

1/\
maximum summer densities (birds km of NW SW SE NE
shoreline) of resident non-waders, AFRICAN BLACK OYSTER- RINGED PLOVER
CURLEW SANDPIPER
CATCHER
resident waders and migrant waders on WHITE FRONTED PLOVER LITTLE STINT
sandy beaches in four geographical THREEBANDED PLOVER
WATER DIKKOP
sectors around the southern African KELP GULL
coast. NW = northwest, SW ~ southwest, CAPE WAGTAIL

SE = southeast, NE = northeast). NW SW SE NE

There are five possible geographical
trends in avian abundance (Fig. 4).
Eight rare species and Hartlaub's Gull
cannot be ascribed readily to anyone of
the trends, but the remaining 16 species
may be classified as belonging to three
of the five possible trends;
in bird density with increasing latitude
was the most common trend (Fig. 4).
Classic ecological theory predicts that
faunal diversity will increase as
latitude decreases (Pianka 1966).
However, the opposite trend is exhibited
by the sandy beach avifauna of southern
Africa (Fig. 5). A significant
difference in the relationship of
diversity and latitude exists between
the west and east coasts (F ig. 5).
Diversity is consistently lower on the
east coast and increases linearly with
increasing latitude (H 0.06L - 0.57,
r = 0.97, p<O.OOl). The best
I~
NONE NONE
NW SW SE NE
FIGURE 4. Geographical trends (northwest
to ~ortheast) in the densities (birds
km -1 of shoreline) of resident and
an increase mi8ratory shorebirds on sandy beaches
around the southern African coast.
mathematical description of the
relationship on the west coast (H = 1.31
In L - 2.84, r = 0.80, p<O.Ol), implies
that the rate of decrease in diversity
accelerates near the tropics. However,
close inspection of the distribution of
points (Fig. 5) suggests that the rate
of decrease in diversity at lower
latitudes on the west coast is more rapid
than suggested by the logarithmic
relationship, and that the avifauna is
relatively homogeneous (in' terms of
diversity) until tropical latitudes are
reached. Data are required from
 512

2.0
.= WEST COAST
•
"'=
.-- -. •
EAST COAST
1.8
> )------
I-
rJ)
1.6
•....... ..........
a:
•, -,
,-

w 1-4 ,- ",""
> ,/ WORLD
0 1.2 / :A

,, - •
,

-.
3

. .....
rJ)

...-
60
W 1.0 , • ... •• I ./.
(,)
W
Q.. 18
I

,---
,- ,- -- 'N ..,'
/
>~/-· ..
.
, 50

--
rJ) I
I ,- ,-
1.6 I
/
~/
,-
//.
! " -- --
40

,,
I

.. ""
12 14 16 18 20 22 24 26 28 30 32 34 36
30

LATITUDE (Ll" S
,,

.
20 • • • .,

,7'
FIGURE 5. The relationship between
species diversity of the sandy beach
avifauna and latitude in southern Africa.
Dots = west coast, open triangles = east
10 ,.....
... :
.l . . _1 _.

coast. west-coast line (1) is fitted !, INVERTEBRATE SPECIES RICHNESS

by eye, west-coast line (2) is a
,,, 20 30 40 50 60 70 80
logarithmic fit, and east-coast line (3)
,,,
is a linear fit. 10 , 28
\-, SOUTH AFRICA
,,
southern Namibia, between latitudes 20 ,, 30
's ,
23°S and 2SoS, to test this hypothesis. .I"
30 ' . I 32
Despite this gap in the current data set, J~....
0
it is apparent that the "tropical state" I·
0°
40 34
°°
extends farther south on the warm east
coast than on the cool west coast.
2 8 14 20
Projections of lines 1 and 3 of Fig. 5
INVERTEBRATE SPECIES RICHNESS
indicate that the avifaunas of the east
and west coasts could have similar FIGURE 6. The relationship between
invertebrate species richness and
diversities (H ± 0.3) at between
latitude on sandy beaches in South Africa,
14°S and 15°S. No data exist to and on a world scale (based on data from
Achuthankutty (1976), Bally (19S1),
examine this prediction.
Brown (1971), Croker et al. (1975),
Dexter (1972, 1974, 1976), Dye et al.
(1981), Eleftherion, Nicholson (1975),
The reasons for the increase in diversity
Holland, Polgar (1976), Jones (1979),
of the avifauna with increasing latitude McIntyre (1970), McLachlan (1977a,b),
McLachlan et al. (19S1a), Penchaszadeh
are obscure, and are unlikely to be the
(1971), Penchaszadeh, Olivier (1975),
result of only one factor. Invertebrate Seed, Lowry (1973), Trevallion et al.
(1970), Vohra (1971), Webber (1979),
production in beaches in the south of
Withers (1977), Wooldridge et al. (19S1)
South Africa generally is higher than in
the north, as is invertebrate biomass fragmented at higher latitudes in South
(McLachlan 1977a,b, Bally 1981, Dye et Africa, and it may be that rocky shores,
al. 19S1, McLachlan et al. 19S1a, which generally support a different
Wooldridge et al. 19S1). Also the avifauna, are influencing assemblages of
sandy beach habitat tends to be more birds at adjacent sandy beaches. There

is a tendency for sandy beaches at higher
latitudes to support greater numbers of
invertebrate species, both on a world
scale and in South Africa (Fig. 6), and
greater avian species richness at higher
latitudes in southern Africa may reflect
this trend.
3.2. Energy requirements and consumption
of invertebrate pr~
The calculation of gross energy consumed
by birds in the sandy intertidal zone
poses several problems. The theoretical
maximum daily energy consumption (assuming
all birds present derive all their energy
from the intertidal zone during summer)
is contained in Table 2. However, the
figures are not realistic, since not all
species are obligate intertidal foragers.
For instance, the intertidal dependency
of gulls, a major component of the avian
biomass, is unknown, and probably varies
locally according to the abundance of the
sand mussel Donax serra and the
availability of food resulting from man's
activities. If the data in Table 2 were
used in an unmodified form, gulls would
account for up to 86 % of the energy
consumption locally.
A more realistic estimate of the total
annual energy consumption in the study
area (excluding southern Namibia) is
based on the following assumptions:
migrant waders are present for 150 days
(Pringle, Cooper 1977), intertidal
invertebrates account for 20 % of the
daily existence energy of Kelp Gulls,
60 % for smaller gulls and 50 % for Cape
Wagtails, and the study area contains
2 200 km of sandy beach. Thus, the
total annual energy consumption by birds
is about 2.15 X 10 9 kJ which, assuming a
mean energy content of 24 kJ g -1 (ash
513
free dry mass) for invertebrate food
(Bally 1981), translates as 90 t ash free
dry mass ( c.720 t wet mass) of "flesh".
The assimilation efficiency of waders
(African Black Oystercatchers) is about
73 % (Hockey in press). Hence, 5.8 X
8
10 kJ of the avifauna's gross consumption
of 2.15 X 10 9 kJ is returned to the
system by means of excretory products.
This figure is an underestimate of the
total amount of energy potentially
contributed by birds to the sandy beach
ecosystem, because large populations of
certain species, such as cormorants
Phalacrocorax spp. and terns Sterna spp.,
which feed at sea, roost on beaches and
thus provide a net energy gain to the
intertidal zone, not only through faeces,
but also by way of feathers and carcasses.
Invertebrate biomass ranges from 37 g
AFDW m- 1 in Natal (Dye et al. 1981) to
1 709 g AFDW m- 1 in the eastern Cape
where the large sand mussel D. serra
predominates (McLachlan 1977b). Annual
consumption by birds of the invertebrate
standing crop ranges from 2 % to 65 %,
and consumption of production from 10 %
to 49 % (Table 3). Only in the southern
Cape is more than 25 % of the annual
invertebrate production removed. However,
the calculated figure (49 %) for this area
is probably an overestimate, resulting
from invertebrate surveys having been
made on relatively unproductive beaches.
McLachlan et al. (1979) estimated that
birds consumed 8 % of the standing crop
of invertebrates from beaches in the
eastern Cape (comparable figure in this
study is 2 %), but they assumed that
invertebrates (Donax) supplied 35 % of
the DEE of Kelp Gulls throughout the
year. They estimated the annual
TABLE 2. Daily existence energy (DEE) (kJ km- 1 ) of all species of shorebird on sandy beaches in 17 200-km <.n
sections around the southern African coast during summer '"

Section 1 2 3 4 5 6 7 8 g 10 11 12 13 14 15 16 17

Resident Waders
African Black
Oystercatcher 8 103 1259 643 820 493 406 977 379
Whitefronted Plover 46 284 321 465 311 855 534 658 588 332 121 174 214 222 170 130 186
Kittlitz's Plover 3 9 4 5 4
Threebanded Plo~er 1 1 1 2 1
Water Dikkop 6 3

All resident waders 46 292 321 465 312 962 1809 1307 1413 825 527 1151 597 222 170 134 186

Migrant waders
European Oystercatcher 7
Turnstone 2 726 294 1224 1291 378 510 4 520 19 7 39 51 88 14 4
Ringed Plover 1 7 15 6 95 4 68 7 1 17
Great Sandplover 1
Grey Plover 87 678 461 725 76 135 203 9 195 28 335 34 4 2 4
Curlew Sandpiper 5 110 50 104 1155 447 646 215 343 2 6 23
Little Stint 151 1 2 3 15 2 4
Knot 92 284 95 1 31 89 16 24 6 13
Sanderling 1300 1863 2803 2457 2273 2156 3395 1104 872 127 164 1007 770 567 136 624 237
Ruff 9 247
Common Sandpiper 2 4 5 6 1 11 2 9 10 3
Greenshank 7 2 4 14 15 17 25 6 12 18 17
Bartailed Godwit 33 5 43 3 2 3
Curlew 5 7
Whimbrel 18 3 147 25 15 84 5 211 10 8 21 17 29 5 75 39
All migrant waders 1488 3733 3729 4706 4879 3243 5133 1370 2288 84 192 1425 978 714 169 750 300

All waders 1534 4025 4050 5171 5191 4205 6942 2677 3701 1009 719 2576 1575 936 339 884 486

Resident Non-Waders
Kelp Gull 173 311 1042 5162 1184 4213 24350 16514 6897 2946 3218 3774 650 856 249 177 9
Greyheaded Gull 2 76 2 18 15 2 4 270 28
Hartlaub's Gull 140 524 1598 539 3792 2655 11
Pied Wagtail 1
Cape Wagtail 1 62 131 34 68 46 27 44 5 6 13 3 5
All resident non-waders 173 311 1185 5686 2920 4883 28178 19255 6954 2973 2362 3794 658 869 254 450 42

All resident birds 219 603 1506 6151 3232 5845 29987 20562 8367 3798 2889 4945 1255 1091 424 584 228

All birds 1707 4336 523510857 8111 9088 35120 21932 10655 3982 3081 6370 2233 1805 593 1334 528
 515

TABLE 3. Estimated annual energy consumption by birds in relation to invertebrate

standing crop biomass and production, as calculated from PIB ratios.
All data are per m of beach. Invertebrate data from Bally (1981) (western Cape),
McLachlan et al. (1981a) (southern Cape), McLachlan (1977) (eastern Cape), Wooldridge
et al. (1981) (Transkei), and Dye et al. (1981) (Natal). The mean energy value of
intertidal invertebrates was taken as 24 kJ g-1 AFDW (Bally 1981)

Geographical area W. Cape S. Cape E. Cape Transkei Natal

17 200-km sections of the coast 5-8 9-11 12 14 15-17

Invertebrate standing crop (g) 366 51 1709 46 37

Invertebrate standing crop (kJ) 8795 1227 41011 1102 882
Invertebrate production (kJ) 9049 1622 9523 1418 1008
% consumption of standing crop by waders 11 38 1 17 14
% consumption of standing crop by all birds 26 65 2 23 17
% consumption of production by waders 10 29 7 13 12
% consumption of production by all birds 25 49 10 18 15

consumption of D. serra production to be 29.9 % respectively (Fig. 7). The high

36 %, of D. sordidus to be 12 % and of the proportion of energy ingested by resident
mysid GastroBaCCuB psammodyteB to be 16 %. waders in the eastern Cape is explained
At Langebaan Lagoon, in the southwestern by high densities of African Black
Cape, Curlew Sandpipers alone consumed Oystercatchers exploiting large
approximately 12 % of the gross annual populations of D. Berra. This source
production of zoobenthos living on the of food is largely unavailable to
surface or in the upper 100 mm of the migrant waders, since none of the migrant
substratum (Puttick 1980). On The Wash, species has an appropriate beak
in eastern England, waders removed between morphology for opening Donax. The
14 % and 43 % of invertebrate biomass protruding siphons of D. Berra do,
during the winter period in favoured however, apparently constitute an
feeding areas (Goss-Custard 1977). important food item for Sanderlings in
the eastern Cape (McLachlan et al. 1979)
Counts of birds on a seasonal basis have
been made at three sandy beach localities Overall, energy consumption by waders at
on the South African coast. Considering the eastern Cape site is approximately
only waders, which are largely obligate an order of magnitude lower than in the
intertidal feeders, there is a consistent southwestern Cape, and, similarly,
seasonal pattern in energy demand at the consumption by waders in Natal is an
three sites, peaking in summer when large order of magnitude lower than in the
numbers of migrant~ are present (Fig. 7). eastern Cape (Fig. 7).
On an annual basis, residents in the
southwestern Cape account for only 8.6 % 3.3. Ecological structure of the avifauna
of the total annual energy consumed by all Fig. 8 depicts simultaneously four
waders. In the eastern Cape and Natal, principal feeding classes (the variables)
the corresponding figures are 58.1 % and of 10 species of birds, and the 17 200-km
516

250 NATAL

150

50

09

35 000 SOUTHWESTERN CAPE

, i
i\
\
2 000

/ \ I I
25000 ,'\ I \
! .," : 1 000
'I \
\
,r' \

,f \
, I,
09 12 03

./ ,.
·:-t;, ..
b_.tor··tl
,/ \
4
''b-b-b-b-
\
\

09 12 03 06

TOWN ELIZABETH

FIGURE 7. Seasonal patterns of the daily energy requirements of waders at three sandy
beach localities on the South African coast. Pie diagrams depict overall annual
energy partitioning between resident (hatched) and migrant (open circles) waders.
Triangles trace graphs for resident and dots for migrant waders. DEE (kJ km- 1 of
shoreline) on y axis, and months on x axis.

sections (the objects) of the southern
African coastline.
similarity between objects or variables
is related to their separation in the
graphical display.
the same direction from the origin as the
variables for which they have the highest
"scores", i.e., a coastal section with a
high proportion of shallow probers would
be plotted close to the point for shallow
probers.
The results of this analysis indicate
The degree of that the ecological structure of the
avifauna varies in relation to latitude,
and generally support the recognition of
Object points lie in four geographical sectors (Fig. 8), as
identified from avian species abundance
(Fig. 2, Table 1). The northwest and
northeast sectors are clearly
identifiable, the northwest sector being
characterized by a relatively high
proportion of medium probers, and the
northeast by a high proportion of shallow
 517

probers (Table 4). In the two southern

sectors, sections 10 and 11 are SURFACE
apparently discrete, being characterized PECKERS
X
by a high proportion of deep probers
(African Black Oystercatcher) Sections
12 and 13, on the other hand, fall within
the "area" of the southwest sector in the
graphical plot and appear ~o form a
transition between the extreme south AXIS 2
(sections 10 and 11) and the northeast ORIGIN ~.~
sector (sect ions 14-17). --P:XIS 1

The most important factor separating the

sectors is the relationship between the
proportion of medium and deep probers,
which accounts for 64 % of the "inertia"
(Greenacre 1978) in the analysis.
FIGURE 8. Correspondence analysis
(Greenacre 1978) of four principal
feedin9 classes of the sandy-beach
avifauna in 17 200-cm sections around
Medium the southern African coast.

TABLE 4. Proportionate (%) mean daily energy consumption in summer by four feeding
classes of bird in four geographical areas of the southern African coast

Geographical area Northwest Southwest Southeast Northeast

17 200-km sections of the coast 1-4 5-9 10-13 14-17

Deep probers 3.0 26.4 39.2 6.6

Medium probers 16.5 3.4 7.0 0.5
Shallow probers 57.4 43.3 35.0 59.9
Surface peckers 23.1 26.9 18.8 33.0

Energy consumption ( kJ km- 1 ) 3669 4528 1479 652

probers are prominent in the northwest summer (Table 4).

sector, whereas deep probers account for
more than 10 % of the energy consumed It is suspected that a correlation
only in the two southern sectors (Table 4). exists between the ecological structure
There is a strong negative association of the avifauna and the distribution of
between the proportion of deep and shallow sand particle size. For example,
probers in the four regions (% shallow densities of Whitefronted Plovers (D,
probers 61.71 - 0.68 % deep probers, in birds km- 1 ) are correlated
r ~ 0.98, p<O.Ol). Shallow probers significantly with median particle size
predominate consistently in all (0), (D ~ 2.400-0.15, r ~ 0.55, d.f. ~

geographical areas except the southeast 18, p<0.02) (W.R. Siegfried, unpublished)
area, in terms of energy requirements in This species, a surface pecker, is most
518

abundant on beaches with fine sand. of Walvis Bay), and one temperate (south
Insufficient data exist to attempt of the Orange River). Whether the
similar correlations for other species. region between Walvis Bay and the Orange
River falls within the tropical or
3.4. Zoogeography of the sandy beach temperate areas is not known, but the
avifauna relationship of species diversity to
The geographical classification of the latitude (Fig. 5) suggests that it will
southern African coastal biota remains be temperate in nature. The analysis
uncertain, and differing opinions have of the ecological structure of the sandy
been proffered by workers based on beach avifauna supports the existence of
results obtained by analysing different a discrete area north of Transkei, a
taxa and habitat types (see Brown, region identified by Stephenson,
Jarman 1978, Siegfried 1981). Stephenson (1972), McLachlan et al.
(1981a) and Siegfried (1981) as being
Intertidal floral ard faunal studies zoogeographically distinct.
(Stephenson, Stephenson 1972) suggest
that there are three biogeographical Based on the sandy beach avifauna, the
zones ("provinces") in southern Africa: area from Cape Agulhas to Transkei is
a west coast zone extending from Walvis not clearly identifiable as discrete; a
Bay to Cape Agulhas, a south coast zone conclusion similar to that of Siegfried
from Cape Agulhas to Transkei, and an (1981) . It could be argued that the
east coast zone extending north of area warrants status as a distinct
Transkei. Similar conclusions were zoogeographical zone, but its avifauna is
reached by McLachlan et al. (1981a) far from homogeneous. The position of
based on the invertebrate fauna of sandy points 9-13 in the correspondence
beaches. Siegfried (1981) analysed the analysis (Fig. 8), relative to the X axis
estuarine avifauna of the subcontinent (the first factor), confirms this
and concluded that a distinct west coast observation. with regard to factor 1
zone existed, extending from southern (deep probers VB med ium probers), sect ion
Angola to approximately Cape Point, and 9 is more similar to sections 12 and 13
that there was an east coast zone than to sections 10 and 11. Sections 10
extending from Transkei northwards. The and 11 appear to be clearly discrete from
intermediate area, between Cape Point the sections 14 - 17, but sections 12 and
and Transkei, he consider~d as a zone of 13 fall midway between the two,
overlap. suggesting transitional status.
Although sections 12 and 13 fall within
Very little is known of the biota of the the "area" of the southwest sector in the
southern African coast north of Walvis graphical plot (Fig. 8), it must be
Bay, but the analysis presented herb remembered that this figure is in fact a
(Fig. 8) suggests that the west coast two-dimensional representation of a
zone, identified by Siegfried (1981) as "cloud" of points in multidimensional
extending homogeneously as far north as space. If the third dimension (not
~uthern Angola, does in fact contain plotted) is considered, points 12 and 13
at least two areas: one tropical (north are separate from the cloud of points

representing the southwest sector.
In summary, analysis of the sandy beach
avifauna suggests that there are at least
three zoogeographical zones around the
southern African coast: a tropical west
coast zone, a temperate southwest zone,
and a subtropical east coast zone. There
might be a fourth, much smaller, zone on
the south coast (sections 10-11), and an
area of intergrading between this zone
and the subtropical east zone.
3.5. Directions for future research
To improve this picture censuses are
required of the shorebirds between the
Ora~ge River and Walvis Bay. Secondly,
quantitative sampling of the intertidal
invertebrate macrofauna in both southern
and northern Namibia is needed, and,
thirdly, studies are required of the diets
of the birds and the extent of their
dependence on the intertidal zone to
satisfy their food requirements.
There is a need for long-term monitoring
of shorebird populations at designated
localities on a regular (monthly) basis.
Although within-year seasonal variations
have been described, very little is known
about annual fluctuations. Robertson
(1981) found appreciable between-year
variations in populations of migrant
waders at Langebaan Lagoon. Such
fluctuations in sandy beach avifaunas
need to be determined before all the
assumptions of McLachlan et al. (1981b)
in analysing the flow of energy in sandy
beach ecosystems can be accepted.
4. SUMMARY AND CONCLUSIONS
Sandy beaches comprise 68 % of the southern
African coastline and the proportion of
sandy beaches in the coastal habitat of
519
the region increases with decreasing
latitude. Of the 25 species of
shorebirds which forage intertidally on
sandy beaches in the region, six are
numerically dominant. The most frequent
trend in sandy beach shorebird
distribution is one of increasing
abundance with increasing latitude.
The west coast supports a greater
richness and diversity of shorebirds than
do the south and east coasts. Diversity
increases with increasing latitude; and,
a tropical avifauna occurs at a hi9her
latitude on the warm east coast than on
the cool west coast.
Total energy consumption by the sandy
beach avifauna along 2 200 km of the
southern African coastline is about
9 -1
2.15 x 10 kJ year ,or approximately
720 t (wet mass) of invertebrates. More
8
than 5.8 x 10 kJ is returned to the
system each year as faeces, feathers and
carcasses. Shorebirds consume between
10 % and 49 % of the annual invertebrate
production of sandy beach ecosystems in
five areas studied. Energy consumption
by waders on sandy beaches shows marked
seasonal fluctuation. A higher
proportion of the over,all consumption by
waders is accounted for by resident
species in the eastern Cape than
elsewhere, due to high densities of
African Black Oystercatchers
Haematopus moquini exploiting large
beds of the sand mussel Donax serra.
Analysis of the sandy beach avifauna in
terms of principal feeding classes
(ecological structure) suggests that
there are at least three zoogeographical
zones.
5. ACKNOWLEDGEMENTS
The nature conservancies of South West
520
Africa/Namibia, the Cape Province and
Natal provided logistical and financial
support for coastal bird surveys.
Additional financial support was received
from the South African National Committee
for Oceanographic Research (SANCOR), the
University of Cape Town and the South
African Nature Foundation. The Western
Cape Wader Study Group provided
information. Professor L.G. Underhill
advised on statistical matters.
6. REFERENCES
Achuthankutty CT (1976) Ecology of sandy
beach at Sa~coale, Goa: Part 1 - Physical
factors influencing production of
macrofauna. Indian J. Mar. Sci. 5, 91-97.
Ansell AD, McLusky DS, Stirling A and
Trevallion A (1978) Production and energy
flow in the macrobenthcs of two sandy
beaches in south west India. Proc. Roy.
Soc. Edin. 76, 269-296.
Bally R (1981) The ecology of three
sandy beaches on the west coast of South
Africa. Ph.D. thesis, University of Cape
Town.
Brown AC (1971) The ecology of the
sandy beaches of the Cape Peninsula,
South Africa. Part 1. Introduction.
Trans. Roy. Soc. S. Afr. 39, 247-280.
Brown AC and Jarman N (1978) Coastal
marine habitats. In Werger MJA and
van Bruggen AC, eds. Biogeography and
ecology of southern Africa. The Hague,
W Junk.
Croker RA, Hager RP and Scott KJ (1975)
Macroinfauna of northern New England
marine sand II: amphipod dominated
intertidal communities. Can. J. zool.
53, 42-51.
Dexter DM (1972) Comparison of the
community structures in a Pacific and
Atlantic Panamanian sandy beach. Bull.
Mar. Sci. 22, 449-462.
Dexter DM (1974) Sandy beach fauna of
the Pacific and Atlantic coasts of Costa
Rica and Colombia. Revta. BioI. Trop. 22,
51-66.
Dexter DM (1976) The sandy beach fauna
of Mexico. Southwest. Nat. 20, 479-485.
Dye AH, McLachlan A and Wooldridge T
(1981) The ecology of sandy beaches in
Natal. S. Afr. J. Zool. 16, 200-209.
Edwards DC, Conover DO and Sutter F
(1982) Mobile predators and the structure
of marine intertidal communities. Ecology
63, 1175-1180.
Eleftherion A and Nicholson MD (1975)
The effects of exposure on beach fauna.
Cah. BioI. Mar. 16, 695-710.
Goss-Custard JD (1977) The ecology of
The Wash III: density-related feeding
behaviour and the possible effects of a
loss of feediny grounds on wading birds
(Charadrii). J. Appl. Ecol. 14, 721-739.
Gray JS (1981) The ecology of marine
sediments. Cambridge, Cambridge
University Press.
Greenacre MJ (1978) Some objective
methods of graphical display of a data
matrix. Ph.D. thesis, University of
Pierre and Marie Curie, Paris. Published
as a special report, University of South
Africa, Pretoria.
Hale WG (1980) Waders. London, Collins.
Hockey PAR (1981) Morphometrics and
sexin9 of the African Black Oystercatcher.
Ostrich 52, 244-247.
Hockey PAR in press Growth and
energetics of the African Black
Oystercatcher Haematopus moquini. Ardea.
Holland AF and Polgar TT (1976)
Seasonal changes in the structure of an
intertidal community. Mar. BioI. 37,
341-348.
Jones DA (1979) The ecology of sandy
beaches in Penang, Malaysia, with special
reference to Excirolana orientalis (Dana).
Estuar. Coastal Mar. Sci. g, 677-682.
Joubert CSW (1981) The density of
shorebirds on a Natal sandy beach.
Ostrich 52, 190-192.
McIntyre AD (1970) The range of biomass
in intertidal sand, with special reference
to the bivalve Tellina tenuis. J. Mar.
BioI. Ass. U.K. 50, 561-575.
McLachlan A (1977a) Studies on the
psammolittoral meiofauna of Algoa Bay,
South Africa II. The distribution,
composition and biomass of the meiofauna
and macrofauna. Zool. Afr. 12,33-60.
McLachlan A (1977b) Composition,
distribution, abundance and biomass of
the macrofauna and meiofauna of four
sandy beaches. Zool. Afr. 12, 279-306.
McLachlan A (1980) The definition of
sandy beaches in relation to exposure:
a simple rating system. S. Afr. J. Sci.
76, 137-138.
McLachlan A and Hanekom N (1979)
Aspects of the biology, ecology and
seasonal fluctuations in biochemical
composition of Donax serra in the east
Cape. S. Afr. J. Zool. 14, 183-192.
McLachlan A and van der Horst G (1979)
Growth and reproduction of two molluscs
from an exposed sandy beach. S. Afr. J.
Zool. 14, 194-201.
McLachlan A, Wooldridge T, Schramm M
and Kuhn M (1979) Seasonal abundance
biomass and feeding of shore birds i~ the
east Cape, South Africa. Ostrich 51, 44-52.
 521

McLachlan A, Wooldridge T and Dye AH Webber HH (1979) The intertidal and

(1981a) The ecology of sandy beaches in shallow subtidal benthos of the west
southern Africa. S. Afr. J. Zool. 16, coast of Whidbey Island, spring 1977 -
219-231. winter 1978. N.O.A.A. Tech. Mem. E.R.L.
McLachlan A, Erasmus T, Dye AH, MESA 37, 1-108.
Wooldridge T, van der Horst G, Rossouw G, Withers RG (1977) Soft-shore
Lasiak TA and McGwynne L (1981b) Sandy macrobenthos along the south-west coast
beach energetics, an ecosystem approach of Wales. Estuar. Coastal Mar. Sci. 5,
towards a high energy interface. Estuar. 467-484.
Coastal Mar. Sci. 13, 11-25. Wooldridge T (1980) Studies on the
Penchaszadeh PE (1971) Observaciones biology and ecology of estuarine
cuantitativas preliminares en playas Copepoda, Mysidacea and the beach mysid
arenosas de la costa central del Peru, Gastrosaccus psammodytes in the eastern
con especial referencia a la poblaciones Cape, South Africa. Ph.D. thesis,
de Muy-Muy (Emerita analoga) (Crustacea, University of Port Elizabeth.
Anomura, Hippidae). Contr. Inst. Siol. Wooldridge T, Dye AH and McLachlan A
Mar. Mar del Plata 177, 3-19. (1981) The ecology of sandy beaches in
Penchaszadeh PE and Olivier SR (1975) Transkei. S. Afr. J. Zool. 16,210-218.
Ecologia de una poblacion de "Serberecho"
(Donax hanLeyanus) en villa Gessel,
Argentina. Malacologia 15, 133-146.
Pianka ER (1966) Latitudinal gradients
in species diversity: a review of
concepts. Am. Nat. 100, 33-46.
Pringle JS and Cooper J (1977) Wader
populations (Charadrii) of the marine
littoral of the Cape Peninsula, South
Africa. Ostrich 48, 98-105.
Puttick GM (1980) Energy budgets of
Curlew Sandpipers at Langebaan Lagoon,
South Africa. Estuar. Coastal Mar. Sci.
11, 207-215.
Robertson HG (1981) Annual summer and
winter fluctuations of Palaearctic and
resident waders (Charadrii) at Langebaan
Lagoon, Sout~ Africa. In Cooper J, ed.
Proc. Symp. Birds of the Sea and Shore,
1979, pp. 335-345. Cape Town, African
Seabird Group.
Seed R and Lowry BJ 1973. The
intertidal macrofauna of seven sandy
beaches of County Down. Proc. Roy. Irish.
Acad. Sect. B 73, 217-230.
Shannon CE (1949) The mathematical
theory of communication. In Shannon CE
and Weaver W, eds. The mathematical
theory of communication. Urbana,
University of Illinois Press.
Siegfried WR (1981) The estuarine
avifauna of southern Africa. In Day J,
ed. Estuarine ecology, pp. 223-250.
Cape Town, AA Balkema.
Stephenson TA and Stephenscn A (1972)
Life between tide-marks on rocky shores.
San Francisco, Freeman & Co.
Trevallion A, Ansell AD, Sivadas P and
Narayanan B (1970) A preliminary account
of two sandy beaches in south west India.
Mar. BioI. 6, 268-279.
Vohra FC (1971) Zonation on a tropical
sandy shore. J. Anim. Ecol. 40, 679-708.
Walsberg GE (1980) Energy expenditure
in free-living birds: patterns and
diversity. Proc. Int. Orn. Congr. 17,
300-305.
 523

SUB-TIDAL SANDY BBACH TROPHIC STRUCTURE IN 'l'HE AREA OF PUN1'A MORON,

VENEZUELA

P.E. PENCHASZADBH (Instituto de Technologia y Ciencias Marinas and Departamento de

Estudios Ambientales, Universidad Simon Bolivar, Caracas, Venezuala)

1. INTRODUCTION food web of the benthic-demersal trawl

Coastal food chains have been studied fishing bottoms in Golfo Triste.
mainly in relation to the structure of Cervigon ( 1966) provides considerable
communities under direct or indirect information on the diet of several
changes due to human activity (pollution fishes. In the framework of an
and fisheries). Although a great deal ecological study of the marine coastal
of information is available for the environment of Punta Moron (penchaszadeh,
carribbean, specially for Florida 1979) the trophic structure of the sandy
(U.S.A.), this is highly dispersed. In bot tom communi ty ( 0-10 m depth) was
Snedaker (1976) a revision of the know- studied in detail quantifying the
ledge of trophic chains in Biscayne Bay different components of the diet of the
is presented; Odum, Heald (1982) can be organisms, especially that of fish
cited among others, with studies in related to the demersal layers and
particular environments like the trophic benthos. Punta Moron is of <jreat
analysis of the mangrove community in interest because of Planta Centro's
Florida and Brook (1977) on a Thalassia thermoelectric activity.
testudinum community, also in Florida.
An important paper which compiles a 2. MBTHODS
great deal of previous information is The stomach contents of 1,160 fish were
Randall (1967) on the food habits of analized. The animals were captured on
coral reef fish. In Buesa (1969,1970) sandy bottom areas using two different
valuable information is given on the types of net. The first, a beach seine,
trophic interrelations and energy flow 100m long with a central portion of l,2cm
in mangrove lagoons, Thalassia beds and mesh size; the second, a trawl net 4,5m
the food of Lutianus synagris in Cuba. long, 5,8r.l wide and 2,Om tall, with a
Nevertheless not much has been published 2,4 cm mesh size in the cup. It was
on sandy beaches and sandy bottoms of trawled at a depth of 4-10m with a boat.
littoral environments. In Venezuela the
following studies on trophic relations 'l'he animals were fixed in 10% formalin
can be cited: Edwards (1973) on sandy after making an incision on the ventral
beaches and upper infralittoral in two surface. The stomach and intestinal
oriental localities, Bastida, et al. contents were analyzed at the laboratory
(1977) on a coral reef of cubagua island under a stereoscopic microscope and the
and Penchaszadeh (unpublished) on the prey identified to the lowest sistematic
524

level. primary consumer of organic detritus,

sedentary algae and microrganisms. The
Besides the species listed, the stomach sand dollar Mellita quinguiesperforata
contents of several others were is a typical sediment feeder.
investigated but not considered in the
treatment because they appeared only Several arbitrary groupings of some
occasionally or in low numbers, or organisms have been made, as in many
because their stomachs were always found cases it "as not possible to determine
empty. specifically the material in the stomach
content; as an example of this we have
3. RESULTS AND DISCUSSION polychaetes "hich "ere codified as
Amphipods are generally regarded as sediment-feeders or primary consumers
omnivores and a great diversity of although several of the identified
feeding habits can be found in this group species in the benthic study of the area
(Hargrave, 1970; Odum, Heald, op. cit). belonged to carnivorous families. In
Odum, Heald (op. cit.) found two isopods the same "ay, the group assembled as
in Florida that are detritivorous and "zooplankton" were also considered
feed on microalgae, benthic diatoms as primary consumers for schematic purposes.
well as on sediment particles, thus
confirming other observations. The bivalve Donax is a filter-feeder of
Cumaceans are sediment-feeders and phytoplankton and particles in suspension
detritivorous (peres, 1961; Odum, Heald, in the water (Wade, 1969). ,"-'he rest of
op. cit.). The great majority of the bivalves were grouped as suspension-
Mysidaceans are suspension feeders feeders although some species could be
(Barnes, 1963), ingesting small sediment feeders. The Gastropoda have
particles like diatoms and organic been arbitrarily considered as secondary
detritus. Dall (1968) made a revision consumers, although llerbivorous species
of the alimentary habits of shrimps were registered dUring the benthic study
(penaeidae) and concluded that they are (penchaszadeh, op. cit.).
oportunistic omnivores that obtain food
from the organic elements in sediments, Although the data were obtained by
most of the energy being provided by sampling over a period of one year, no
bacteria, small algae, protozoans, seasonal differences were observed in
harpacticoid copepods and nematodes. the diets. This, plus the fact that the
Portunids can be regarded as primary as number of or<janisms in each monthly
well as secondary consumers as they feed sample ,las not high, allows the combi-
on organic detritus, plant and animal nation of data assembled over a year
remains, molluscs and other small crus- (Odum, Heald, op. cit.). Figure 1 shows
taceans (Darnell, 1958). Other the benthic demersal tood chains of
brachyura are fundamentally Punta Horon littoral; they are simpli-
detritivorous and eat plant and animal fied and correspond to the sandy bottoms
remains (Odum, Heald, op. cit; Olivier, between 0 and 10m depth, excluding a
et. al., 1968). Olivier, et. al., (op. small Thalassia testudinum bed (which
cit.) found the sergestid peisos to be a represents 5% of the total area studied).
 525

For purposes of a better systematization consumers that feed on alloctonour plant

of the results we have adopted the basic remains, superior algae or ~halassia:
system (trophic binomial) given by Ubrina, Hemiramphus, Anisotremus and
Angelescu (1965). Five alimentary bases Archosar~us. The Penaeidae, as discussed
have been characterized: a) phyto- earlier, were considered primary con-
plankton plus suspended organic sumers. He can distinguish six
detri tus; b) alochthonous plant remains fundamental elements in the ~eneral

of terrestrial origin that are borne to trophic web. Almost the entire Vieb
the sea by fresh water flows; c) depends on its nutritive value. '1'hese
superior algae removed by coastal are the Brachyura, polychaetes, shrimps
currents from their place of origin; d) (mysidaceans, sergestids and penaeids),
organic detritus in sediments and e) bi val ves and stomatopods. Crabs are
remains of Thalassia blades (ingested in part of the diet of the fish Lutianus,
the Thalassia bed or removed by coastal Rhinobatos, Dixonina, polydactilus,
currents). The term "organic detritus" Arius, Menticirrhus, Umbrina and
as used here is the one defined in Trachinotus. Shrimps are components of
Edwards (1973). The problem of detriti- the diet of Lutianus, Polydactilus,
vorous organisms in a determined trophic Urotrygon, Dixonina, pomadasys,
level is that their food does not Oligopl ites, Eucinostornus, Conodon,
consist entirely of plant and animal Ar ius, belar (=Vomer), Trachinotus,
organic material but also includes Bagre, Dasyatis, Larimus, Rhinobatos and
organisms of different trophic Caranx. Small crustaceans (isopods,
hierarchies such as bacteria and fungi, amphipods and cumaceans) are part of the
protozoans, benthic diatoms and small diet of Lutianus, polydactilus,
invertebrates, herbivores and Diapterus, pomadasys and the sea star
carnivores. For this reason and the Astropecten. Anchoa, Iiarengula and
purpose of schematization of the food juvenile Selar are zooplankton fish.
chains, organic detritus and the micro- The bivalves are ingested by gastropods,
organisms associated with them are the sea star Luidia and Astropecten and
considered here together with the by the fish Anisotremus, Diapterus,
producers. pomadasys, Trachinotus, Umbrina,
Menticirrhus and Arius. The fish
Among suspension-feeders we can Dixonina, Polydactilus, Menticirrhus, and
distinguish the bivalve molluscs notably Thachinotus feed on anomurans (Emerita
Donax denticulatus, and anomurans and Lepidopa). Stomatopoda are food for
Emerita brasiliensis and Lepidopa Syacium, Larimus, Polydactilus and
richmondi, medium and small zooplankton, Menticirrhus. The tertiary consumers
mysidaceans and sergestids. Among the (secondary carnivores) are mainly repre-
primary consumers that feed on organic sented by icthiophagous fishes such as
detritus in the sediments we find small polydactilus, Lutianus, Larimus,
crustaceans (isopods, amphipods and Urotrygon, Oligoplites, Diapterus,
cumaceans), polychaetes, crabs, the sand Dixonina, Arius, Trachinotus, Pomadasys,
dollar, stomatopods and the teleostean Conodon, and Eucinostomus. '1'0 establ ish
Mugil. The following fishes are primary The trophic l.iche for these animals is
526

00+"tJ
~.a
:+gcno
g>-!
c _.'C 'C
",C"I~g
c.
CD_
...
0.0
::I

Fish - eating
Fishes

a,~~~~~
Tylosurus
Rhizoprionodon
Sphyrna
Caranx latus
C.bartholomaei
Elops

FIGURE I. Sub-tidal sandy beach trophic relatioships in the orea of

Punta Moron.
 527

difficult as the material of teleostean belonging to the third level of con-

origin was not easily identified in the sumers and presumably to the fourth or
stomach contents, but they fall into at fifth.
least the third level of consumers
because their diet includes other 4. CONCLUSIONS
elements from the level of secondary The trophic structure studied in the
consumers (Table 1). littoral of punta Moron is uased on five

TABLE 1
til r-l
. .
Diet of the fishes of Punta Moron (in % of the represented volume)
!-;
.gJ
til
Q)
Q) ~

~
Q) ~
PREY r-l tIi til tIi ~
e
Q) C)
~
~

e~ . . .
til tIi til til 'd Q) +' til r-l til Q)
~~
Q) ~
g1 'd F-l ro 'd <11 ;>
g
tIi Q) tIi 'M C) Q) 'M 'M
~0
'M
'd P
&'dtil 'd 'd P-l
r-l C) tIi ~ 0 'M tIi til
i-f'M r-l ~ ~ 'M til ~ 0 ,.q til til
,tt l; ~
~~

e
FISH SPECIES ~;> >< F-l Q) F-l 0 'M 1M 'd ,.q Q) 'tli ~ 0 Q) i-f tIi ~
'M ro Q) ;> Q) PI
~
'M ,F-l 'M C) ro ~ S 'M F-l C) ;> r-l C)
S'd ~- ~
0
,.q 'M
r-l :S 0
til .~ ~ ,~
0 ~
tIi ~ !-; 0 ,.q ~
til
r-l
0
til F-l
til :;j
til
,.q r-l
0
.
Q) Q) C) 'M
I~'~ ~ ~
Q)~ ~ ~
r-lH A 0 H 0 H P-l r£i <11 P-l P-l r-l til 8 til

;
0 0 Cf) Cf)

-b.nisotremus sur~namens~E ~t; ~~ 8 ~~ .ll

~~ 60
- It;
Archosargus rhomboidali! 26-29
Arius spixii
Bagre marinus
27
14
20-31
28-32
50 13 120 P 2C
2 3
8)
Caranx bartholomaei 2 6-8
C. hippos 3 6-7 ~~
C. latus 29 5-21 JOC
Conodos nobilis 29 7-26 3 9 29 2 7 2 4 4~
Dasyatis guttata 2 100-105 ~((!
Diapterus rhombeus 17 12-22 24 21 21 2 lC 12 lC
Dixonina nemoptera 21 11-16 5 34 ~3 2 10 8 18
Elops saurus 6 21-46 leX
Eucinostomus argenteus 4 9-11 25 75
E. melanopterus 10 8-13 28 l~ 7 4C
Harengula clupeola 33 7-13
H. jaguana 134 9-14 ~2 7 6 6
Hemiramphus brasiliensi 24 18-26 JOe
Larimus breviceps 23 6-23 ~l 7 2
Lutjanus griseus 2 9-23 14 5 ~l 2C
Menticirrhus littoralis 22 11-12 5 33 28 4 5 l~
Mug!i:l curema 28 11-30
Oligoplites saurus 42 9-21 30 lC ~5 1":
Polydactilus virginicus 67 10-34 7 70 15 5 3 5
Pomadasys curvinaeformi 19 8-18 4 4 4 430 lC 37
Rhizoprionodon porosus 11 43-46 4 {o:;
Sphyrna lewini 3 60-63 IX:
Syacium papillosum 1 14 io:;
Trachinotus carolinus 14 7-11 9C 10
T. falcatus 18 6-16 23 7 ~8 32
T. goodei 58 5-27 5~ 15 6 6 2 1 15 4 2 4
Umbrina coroides 21 4-25 2 6 5 4 5E ~O 3 4
Urotrygon venezuelae 7 21-29 ::>0 80
Rhinobatus percell ens 1 49 101,
Selene setapinnis 25 7-12 5 90 5
Tylosurus raphidoma 14 30-34 -IDC

Fish like Tylosurus, Rhizoprionodon, components of the trophic binomial

Sphyrna, Carnax latus, C. bartholomaei Producers + Urganic Detritus, of which
and Elops are exclusively icthiophagous. the Phytoplankton and the organic
For this reason they could not be detritus in suspension and in the
codified in their trophic level, sed iments are most important. I t is a
528

highly diverse community that presents a Buesa RJ (1970) Energy flow at Jababo
Port. Cent. Invest. Pesq. 2, pp. 1-53.
complicated trophic web with a great Buesa RJ (1974) population and bio-
diversification of feeding niches. The logical data on turtle grass (Thalassia
testudinum, Konig, 1805) on the
scarcity of Thalassia as food for the northwestern cuban shelf, Aquaculture
fish is a result of the area where the 4(2), 207-226.
Cervigon P (1966) Los peces marinos de
sampling took place. The allocthonous Venezuela. Fund. La Salle, Caracas,
material of terrestrial vegetable origin Vol. 1 and 2. 1-951.
Dall W (1968) Food and feeding of some
could have a greater importance during Australian penaeid shrimp. Proc. Horld
the rainy season, although differences Soc. conf. BioI. and CuI t. of !:>hr imps
and Prawns. Fish Rep. F.A.O. 57(2),
were not detected through the year, and 251-258.
seem not to be very important for the Darnell R~l (1958) Food habits of
fishes and larger invertebrates of Lake
diet of the fish. Its importance would Pontchartrain, Louisiana, an estuarine
probably increase in a study of micro- community. publ. Inst. Mar. Sci. Univ.
Texas 5, 353-416.
organisms and invertebrates. Edwards RRC (1973) Product ion exology
of two Caribbean Marine Ecosystems. 1.
the physical environment and the fauna.
A major part of the primary and second- Estuar. cstl. mar. Sci., 308-318.
ary consumers, which form the basis of Green J (1968) The biology of estua-
rine animals. Univ. of Hash. Press,
the diet of the majority of the fish, Seattle, 1-401.
are planktonic crustaceans larger than Hargrave BT (1970) The utilization of
benthic microflora by Hyalella azteca
5mm in length, especially mysidaceans (Amophipoda). J. Amin. Ecol. 39, 427-437.
and sergestids. An analysis of the Odum WE and de la Cruz AA (1963)
Detritus as a major component of
plankton samples obtained in the same ecosystems. A.I.B.S. Bull. 13(3), 39-40.
area revealed that those two groups of Odum WE and Heald EJ (1972) Trophic
analysis of an estuarine mangrove com-
crustaceans represented up to 92% of the munity, Bull. Nar. Sci. 11(3), 671-738.
total number obtained (with a 505 micron Olivier SR, Bastida R and Torti MR
(1968) sobre el ecosistema lie las aguas
net)(Scelzo and Pellegrini, personal com- litorales de Mar del plata. Miveles
munication) • troficos v cadenas alimentarias pelagico-
demersales y bentonico denersales.
Servo I-lidr. Nav. Buenos Aires, Ii. 1025,
REFERENCES 1-46.
Angelescu V (1965) Cadenas alimentarias Penchaszadeh PE (1979) Ecologia del
y niveles troficos en el ecosistema ambiente marino costero de Punta Moron,
marino, in CUrso Av. Latinoam, de Vene zue la . Ed. INTr;C~lAR, Univ. Simon
Plancton. (UNESCO) Mar del plata, 12, Bolivar, Caracas, 1-343.
1-16. Peres JM (1961) Oceanographie Bilogique
Barnes RD (1963) Invertebrate Zoology. et Biologie Marine. 1. La vie benthique.
W.B. Sanders Co., Phil. 1-428. Presses Univ., paris, 1-541.
Bastida R, Vervigon F and Horeno C Randall JE (1967) Food habits of reef
(1977) Aspectos ecologicos de la fishes of the \lest Indies. Stud. 'l.'rop.
ictiofauna de un arrecife de Acropora Oceanogr., Miami, 5, 655-847.
palmata (Isla de Cubagua, venezuela). Snedaker !:>C (1976) Ecology and the
Res. IV Simp. Latinoam. Oceanogr. BioI., food web of Biscayne Bay. In Biscayne
Guayaquil, 73-74. Bay: Present and Future. Univ. Miami Sea
Brook 1M (1977) 'i:'rophic relationships Grant Special Rep. No.5, 27-233.
in a seagrass community (Thalassia Wade B (1969) Studies on the biology
testudinum) in Card Sound, Florida. Fish of the Hest Indian beach clam, Donax
diets in relation to macrobenthic and denticulatus L. 3. Functional l'lorph-
cryptofaunal abundance. Trans. Am. ology. Bull. Mar. Sci. 19, 306-322.
Fish. Soc. 106(3), 219-229.
Buesa RJ (1969) La bijaiba y su trama
alimentaria. Mar y Pesca, La Habana,
Cuba, 54, 24-29.
 529

FOOD WEB IN THE SURF ZONE OF AN EXPOSED SANDY BEACH ALONG THE MID-ATLANTIC COAST
OF THE UNITED STATES

J.J. McDERMOTT IDepartment of Biology, Franklin and Marshall College, Lancaster,

Pennsylvania 17604 USA)

1. INTRODUCTION 2. MATERIALS AND METHODS

To the casual observer it appears 2.1 The study site
that the exposed sandy beach of temper- All studies were conducted (pririlariry
ate latitudes is, except for the con- from 1977-79) on the outer beach of a
spicuous migrating shore birds, a rather barrier island that was free of man-made
desolate and abiotic region of the sea. obstructions and thus subjected to the
We know, however, that this air-sand- full force of the surf (exposed according
water interface is also inhabited by to MCLachlan's (1980) rating). The beach
less conspicuous shore-zone fishes and was of uniform topography, thus the col-
invertebrates. Each group is relatively lecting site at Avalon was representative
well-known, has a low diversity and has of a 6.4 km section (Fig. 1). A site
evolved a variety of unique adaptations protected by two groins, located about 2
required for such an existence. Their km from the southern tip of the island,
interactions as parts of a dLstinct com- was sampled occasionally.
muni ty, however, have been intensively
studied in only a few parts of the world
(e.g. in South Africa, McLachlan et al.
1981). The low diversity of the exposed
beach is a factor which tends to sim-
plify progress toward an understanding
of the interactions, but the physical
harshness of the region tends to impede
collection of required biological
information dealing with the fauna.
Repeated observations of apparent
SEVEN MILE
feeding activity by fishes within the BEACH
surf of exposed sandy beaches prompted
this study. It was my purpose to gather
quantitative baseline information on the
occurrence and distribution of the
macro-benthic invertebrates and fishes
FIGURE 1 Location of the study site at Avalon,
in the surf, and to interpret the roles New Jersey (39 0 04' 43"N, 74 0 44' OS''W), situated
of the two groups in the food web of the betweem Townsend and Hereford Inlets. Depth
curves (MLW) are given in meters.
community.
530

Sampling encompassed the intertidal

northerly direction against the prevail-
zone and the shallow subtidal area with-
ing inshore drift, for a distance of 50
in the breaker zone. The gradual slope
m. Fish were sampled on a monthly basis
of the foreshore (about 2°) was par-
from September 1977 through October 1979;
ticularly responsible for a prominent
additional collections were made in 1976
surf zone, between the swash and breaker
and 1980. A total of 293/50 m seine
zones (Ingle 1966), where a southerly
hauls were made. Each species of fish
directed longshore current was generally
from each collection was weighed; most
present. Benthic sampling points along
fish were measured and representative
transects in the surf were located in
samples were preserved in 20% sea water
relation to a fixed marker at the base
formaldehyde for later analyses of their
of the foredunes. The mean HW and LW
diets.
locations were approximately 50 m apart.

Seining also provided important

Mean air temperature was 13.8°C
information on the occurrence of the
(-9.5 to 32.3°C), and mean water tem-
larger crustaceans, which were treated in
perature was 14.4°C (-2.0 to 25.7°C).
the same manner as the fish.
Mean salinity was 30.9 0 / 00 (28.8 to
33.7 0 / 00 ). The yearly patterns of
2.4 Food analysis
temperature and salinity were very
Each fish was measured and weighed,
similar to those recorded by Allen et
the stomach contents were flushed into a
al. (1978) in the Hereford Inlet
dish, and the food items were identified
estuary.
and counted. The same or similar items
(e.g. a variety of amphipod species) from
2.2 Benthic sampling
each collection were pooled into indivi-
Cores were taken to a depth of 20
dual tared dishes and dried to constant
cm with a coring device that sampled an
weight. Stomachs of crustaceans were
area of 46 cm 2 . Two cores were obtained
removed and similarly studied.
at each point along a transect (5-10 m
apart) and pooled into a 4 litre wide-
3. RESULTS
mouth plastic jar to which were added
3.1 Occurrence of benthos
Rose Bengal and formaldehyde. Organisms
Of the 37 species of invertebrates
retained by a 0.5 mm mesh screen were
identified in cores taken in the surf
identified and counte.d; some were
zone, only 12 were common inhabitants.
weighed. During the period of this
Among the latter, the polychaete
study 700 cores were taken from 355
Scolelepis squamata, the bivalve Donax
points along 31 transects.
variabilis, the amphipods
Acanthohaustorius millsi, Amphiporeia
2.3 Collection of fish
virginiana, Haustorius canadensis and
A 1. 5 x 15.0 m nylon beach seine
Pseudohaustorius borealis, and the deca-
wi th a 6.5 mm mesh was used to sample
pods Crangon septemspinosa, Emerita
the shore-zone fishes. Seine hauls
talpoida andOvalipesocellatus, occurred
were taken parallel to the beach in a
abundantly or frequently enough to be of
 531

TABLE 1. Dry weights (mg) per 0.1 m2 of species occurring along a transect (18 September 1981) to
show a rather typical distribution of some of the principal fauna in the surf zone of seven Mile
Beach. Original transect also included samples at 5 m intervals, but are not shown. MLW is at
the 50 m mark.

Distance from MHW mark (m)

Organisms 10 20 30 40 50 60 70 80 Totals
Micrura leidyi 261 261
Nephtys bucera 11 11
Scolelepis squamata 3718 6229 4381 16 24 8 8 14384
Spionid A 8 8
Donax variabilis 500 217 315 98 22 11 1163
Amphiporeia virginiana 1 65 196 5 267
Acanthohaustorius millsi 22 22 44
Haustorius canadensis 185 44 229
Neohaustorius schmitzi 2 1 3
Parahaustorius longimerus 1 1
Protohaustorius deichmannae 2 1 3
Pseudohaustorius borealis 22 33 55
Haustoriinae (unidentified) 2 2 2 2 8
Emerita talpoida 44 6 109 174 333

Totals 233 4271 6449 4870 487 312 65 83 16770

importance in the food web (Crangon and

Ovalipes were sampled almost exclusively
NOV. I %
I by seine).

~
*-
• • I~

-II"
I
I Although the Amphipoda (at least 16
MAR.
species, mostly haustoriids) was the

-I
tt-:
MAY I
I most diverse group, the dominant inver-
I tebrate in terms of numbers, weight and
AUG. I frequency of occurrence was the spionid
~ Scolelepis. It dominated the mid-inter-
H L H L H L
region, forming a 20 m wide dense band
SCOLELEPIS AMPHIPOREIA AMPHIPODA along the whole shoreline (McDermott
(1979). Table 1 and Fig. 2 illustrate
this sharp distributional pattern. The
FIGURE 2 Distribution of Scolelepis squamata, numbers of worms/O.l m2 corresponding to
Amphi oreia virginiana and Amphipoda (other the dry weights given at the 20-40
specles in the surf zone at Seven Mile Beach at
four seasons of the year (Nov. 1978, Mar., May intervals in Table 1, were 4185, 8566,
and Aug. 1979). Mean high (H) and low (L) water and 7196, respectively. Conservative
locations are shown on the 10 m interval scale.
Each kite diagram shows the %distribution of extrapolation from a mean of 4000
organisms along the transect (scale of 10% worms/O.l m2 (approx. 5 g dry wt based
intervals in upper right).
532

on overall data, not those in Table 1) pattern also applied to the mean
in the 20 m "Scolelepis zone," suggests wt/sO m-haul.
6.5 t dry wt (45.5 t wet wt) of the
worms in the 6. 4 km of exposed beach.
The population exhibited a distinct
offshore movement in the winter months
(Fig. 2).
0 30
Amphiporeia, a species which spends )( 20

~
much time off the bottom apparently --.J
::>
« 10 f- -
moving in and out with the tide, I
:;:
exhibited a variable zonation (Fig. 2). 0
'""-ci

r
The zonation patterns of the other haus-
tori ids were usually distinct, e.g. z
z
Haustorius and Neohaustorius were near «
the HW mark, Pseudohaustorius near LW,
w
:;: n r
~~
/---r:rI1- '-
Acanthohaustorius below LW and S SDMJSDMJS M S
'76 '77 '78 '79 '80
Bathyporeia parkeri beyond the latter
(Table 1 and Fig. 2 show some of these
trends).
FIGURE 3 Mean numbers of fish/SO m seine haul
collected in the surf at Seven Mile Beach from
Two of the more conspicuous in- Sept. 1977 to Oct. 1979, including collections
in Sept. 1976 and Mar. and Sept. 1980.
habitants of the surf, Donax and Rectangles below the abscissa represent <10
Emerita, which usually move in and out fish/haul.
with the tide, showed substantial
seasonal and annual fluctuations and Twenty-six species of fishes be-
very patchy distributions. longing to 18 families were collected in
the Avalon surf. The major i ty of
The two portunid brachyurans species (17 of 26 = 65.4%) occurred only
Ovalipes and Arenaeus cribrarius and the as juveniles while some (5 of 26
caridean Crangon were collected season- 19.2%) occurred as juveniles and adults.
ally in the seine. The predominantly The 10 most abundant species, accounting
juvenile population of Ovalipes moved for 99.6% of the total, are listed in
inshore in July and remained through Table 2. Only six species occurred in
October (N=>800). Juveniles of Arenaeus more than 5 of 28 collections; Menidia
occurred in the fall (N=83). Crangon (Atlantic si1verside) the most frequent,
was found occasionally throughout the occurred as juveniles and adults. It
year, but the peak was from October to was the only endemic species in the
December (N=750). surf, appearing every month except
February.
3.2 Occurrence of fishes
Yearly peaks of fish abundance Menidia was by far the dominant
occurred in the fall (Fig. 3); a similar species, accounting for 85.1% by number
 533

and 39.5% by weight (Table 2). Juve- TABLE 2 Numbers, weights and incidence of ~ishes
niles of Leiostomus (spot), Mugil (white in 28 collections from the surf at Seven Mlle
Beach, (1976-1980)
mullet), Menticirrhus (northern king-
fish) and Pomatomus (bluefish) ranked
next in total numbers. Juvenile wet Inci-
Paralichthys ( summer flounder) and Species Number weight dence
Sphaeroides (northern puffer) and adult (g)
and juvenile Anchoa (bay anchovy) Menidia menidia 9195 25593 23
occurred infrequently, but Sc.ophthalmus Leiostomus
(windowpane), while never abundant, was xanthurus 603 18683 7
more frequent. Young Fundulus (mummi- Mugil curema 530 11590 8
chog) was a straggler from nearby Menticirrhus
estuaries. Juvenile Trachinotus saxatilis 103 10
carolinus (Florida pompano) and Mustelus Pomatomus
canis (smooth dogfish) are shallow surf saltatrix 99 5343 5
species, but in this location they were Paralichthys
sparse and infrequent. In other exposed dentatus 98 1573 1
beaches south of Avalon I found them Anchoa mitchilli 58 58 3
more frequently, and Thomas, Milstein Sphaeroides
(1973) have reported the pompano to the maculatus 42 717 1
north. Scophthalmus
aquosus 17 223 6
Thus juvenile fishes such as the Fundulus
spot mullet, kingfish and bluefish, and heteroclitus 16 25 4
perhaps the windowpane, pompano and dog- Other species
fish, are regular seasonal influents in (N=16) 42 595 1-4
the community. Juveniles of the floun- Totals 10803 64837
der, puffer and others appear to be
incidental seasonal influents. was overwhelmingly dominant (67-80%) in
the diet of Menidia at all seasons.
3.3 Diet of fishes and crustaceans Amphipods were common in the diet
(Table 3, Fig. 4) Al though cope- throughout the year, and recruits to the
pods and amphipods were abundant (35.5 Emerita population were utilized in the
and 35.6%, respectively) in the diet of summer and fall. Copepods were dis-
Menidia, they were overshadowed by Sco- tinctly more prevalent in the diet in
lelepis on a dry weight basis. Amphi- early winter when some of the benthic
poreia was the only other taxon that crustaceans, normally common in the diet,
accounted for more than 10% of the diet. had migrated to deeper water. Insect
Crustaceans, in general, accounted for utilization occurred primarily in the
21.8% of the total. Twice as many summer and early fall. All of the
species (32) were identified in the diet abovementioned trends in food utiliza-
of Menidia as compared to its nearest tion have been amply confirmed by stomach
competitor in the nekton. Scolelepis analyses of an additional 355 fish
534
collected in the same area from 1976-78
for. which only numbers and incidence of
food items were determined.
Scolelepis and Donax were the prim-
ary food items in the diet of juvenile
spot (71.5 and 24.B%, respectively)
crustaceans amounted to only 3.6% of
total. The data for Donax included
shell, and shell is about 92% of
total dry weight of the animal.
northern kingfish, which belongs to
same family as the spot, fed heavily on
Scolelepis, but Donax was never found in
its stomach. It consumed about four
times more crustaceans than the spot.
The diet of juvenile bluefish, ap-
parently the top predator in the shallow
surf consisted primarily of fish (76.0%)
Scolelepis was found only in f ish less
than 130 mm in length. Mugil is a her-
bi vore and apparently feeds little in
the surf. Juvenile Paralichthys (n;37),
collected only on two occasions in
September 1977, fed primarily on Neo-
mysis, Crangon and other crustaceans
(96.2%). The other flatfish,Scoph-
thalmus, (N;16) also fed primarily on
crustaceans (67.2% ), particularly
Amphiporeia, but Scolelepis also figured
prominently in its diet (34.8%). Anchoa
(N;54) fed primarily on Neomysis and
other zooplankton. Eight juvenile pom-
panos, collected on four occasions, were
feeding on Donax (54.7%), Scolelepis
(31.7%), Emerita (9.2%) and amphipods
(4.4%). My previous observations of
pompanos elsewhere in the New Jersey
surf indicated that Donax and Emerita
were the principal items in the diet of
this southern straggler. Juvenile dog-
fish (N;5) fed on Ovalipes and Emerita.
Scolelepis, Donax and a variety of
crustaceans were the important food items
in the diet of Ovalipes (N;289; based on
frequency of occurrence). A dry weight
analysis of the food from 52 crabs yield-
ed 20.9% polychaetes~ 75.2% Donax and
and 3.9% crustaceans (these data include the
the Donax shell). Other less quantitative
the data from the same location indicated
the that Emerita may be one of the more
The important items in the diet of Ovalipes.
the Arenaeus(N;45) appeared to be primarily
a crustacean feeder (Amphiporeia, Emerita
and Ovalipes).
Polychaete remains were found in
41. 0% and crustacean fragments in 33.3%
of 78 Crangon. Nearly all of the poly-
chaete material appeared to be from
Scolelepis. Only Amphiporeia was iden-
tified among the crustacean material,
which was nearly all amphipods. A dry
weight analysis of the stomach contents
of 33 shrimp yielded 79.6% polychaete
and 20.4% crustacean material. Twenty
shrimp with food in their stomachs were
selected from a large collection made
during an extremely low spring tide well
offshore of the "Scolelepis zone". On a
dry weight basis over 95% of the stomach
contents consisted of amphipods,
primarily Amphiporeia.
4. DISCUSSION
Both the benthos and the nekton of
this exposed sandy beach are of extreme-
ly low diversity (Shannon-Weaver index
(H) values are <0.1). This factor, as
indicated earlier, would tend to reduce
the complexities of the interactions and
simplify analyses, and basically this
was the case. The benthos is dominated
by the surface deposit-feeding spionid
 535

TABLE 3. The percentaRes of food organisms (based on d~ wt) found in the stomachs of four species of
fishes collected in tfie surf zone of Seven MIle Beach (1977-80). MM = MenIdia menidia,
LX = Leiostomus xanthurus, MS = Menticirrhus saxatilis, PS = PomatolIlus saltatIlx
Percent of stomach contents
Food organisms MM LX MS PS
N=760 N=236 N=103 N=49

Polychaeta
NeEht;t:s bucera 2.1
ScoleleEis squamata 72.4 71.5 82.2 7.4
Heteronereids 3.0
Mollusa
Donax variabilis 24.8
Crustacea
Copepoda 1.9
Neom;t:sis americana 0.5
Mysidacea and Cumacea 0.5 0.7 0.9
Edotea triloba 0.2
AmEhiEoreia virginiana 10.5 2.7
Amphipoda 3.5 1.3 5.4 0.3
Crang:on seEtemsEinosa 0.7 0.8 6.5 5.1
Emerita talEoida 4.3 0.6 1.3 0.2
OvaliEes ocellatus 4.1
Other crustaceans 0.3 0.2 0.5
Insecta 4.8
Pisces
Anchoa mitchilli 4.6
Menidia menidia 22.9
Other unidentified 48.5
Miscellaneous 0.9 0.1 1.0 0.6

ScoleleEis sguamata, a species known to is the most typical species, maintaining

inhabit the foreshore and in some cases a large population with constant zona-
to be relatively abundant along the east tion except for the predictable seaward
and gulf coasts of the united States movement in the winter. The haustoriid
(Croker. 1970, Hill, Hunter 1976, amphipod AmEhiEoreia virg:iniana is an-
Howard, Dorjes 1972, Saloman, Naughton other fairly abundant species, inhabi t-
1978, Shelton, Robertson 1981). Nowhere ing a wider intertidal range tuan
else, however, has it been shown to ScoleleEis.
exert such dominance in the surf zone
communi ty, overshadowing groups that Menidia menidia is the nektonic
usually are considered to characterize counterpart of ScoleleEis in the shallow
this zone. In the present location, it surf; consisting of juveniles and ~dults,
536
INSECTA +M CRUSTACEA
\ \
MENIDIA N = 760 LEIOSTOMUS N = 236
CRUSTACEA +M SCOLELEPfS
\ \----.-
MENTICIRRHUS N = 103 POMATOMUS
FIGURE 4 Diets of Menidia menidia, Leiostomus
xanthurus, Menticirrhus saxatilis and Pomatomus
saltatrix in the surf at Seven Mile Beach, based
on the per cent dry weight of food items.
M = miscellaneous.
it is the only endemic species, and
competes with the influx of various
juvenile fishes in the fall, during
which time its opportunism may be an
asset. For the remainder of the year it
has few competitors.
Scolelepis is particularly available
to fishes, as well as crabs and shrimp,
during the period when the waves are
rolling up the foreshore over .the zone
inhabited by the worms. While feeding
at the. ends of their loosely-built tubes
many .tend to be dislodged and carried up
the foreshore by the wave, and as the
water drains seaward again they burrow
rapidly back into the bottom. Large
numbers of worms therefore become inter-
+M mittently planktonic (the resident tem-
porary meroplankton of Williams, Bynum
1972) • I.t is during this period that
small fishes, such as Menidia and juven-
iles of larger species, are manoeuvring
and feeding on the worms in the highly
productive "Scolelepis zone". These
fishes, in turn, are relatively protec.t-
ed from predation by the larger species
which are unable to manoeuvre in the
very shallow water (Merriman 1947). It
+M
is not surprising, therefore, that
Scolelepis .is so prominent in the diet
of the more opportunistic .f ishes such as
Menidia, Leiostomus, Menticirrhus and
some others. Crabs and shrimp, moving
seasonally into the surf zone, also
appear to be taking advantage of the
N=49
Scolelepis production. That Amphiporeia
is the most common amphipod in the diet
of various fishes is also not surprising
because it is the most intermittently
planktonic haustoriid.
Benthic species such as Emerita and
~, while always important food
sources for the nekton, were not as
predictable an overall food source as
Scolelepis because of their temporal and
spacial discontinuities. Little inform-
ation is presently available on the
importance of mysids and larval Crangon
in the shallow surf, but both occur
periodically in the diets of some of the
fishes.
There is much evidence seen for par-
titioning of food within the nektonic
component of the community. Donax is
utilized by only one of the two sciaenids
while both heavily utilize Scolelepis;
Paralichthys tends to concentrate on
 537

small epibenthic crustaceans; .Anchoa, lands Institute, Lehigh University,

Bethlehem, Pennsylvania, 138 p.
being a zooplankton feeder, seems to Chao LN and Musick JA (1977) Life
compete little with the benthic species; history, feeding habits, and functional
morphology of juvenile sciaenid fishes
smaller bluefish capable of manoeuvring in the York River estuary, virginia,
in the shallows inhabited by other F ish. Bu 11., U. S. 75, 657 -7 0 2 •
Croker RA (1970) Intertidal sand
j uveni les and Menidia, become the main macrofauna from Long Island, New York,
top predator in the system. The factors Chesapeake Sci. 11, 134-137.
Hill GW and Hunter RE (1976) Inter-
involved in this niche separation action of biological and geological
concern behaviours imposed by adaptive processes in the beach and nearshore
environments, northern Padre Island,
morphological limitations as well as the Texas. In Davis RA Jr and Ethington RL
habits of the prey species involved eds. Beach and nearshore sedimentation,
pp. 169-187. Society of Economic
(Chao, Musick 1977, Roelofs 1954). Paleontologists and Mineralogists, Spec.
Publ. No. 24, Tulsa, Oklahoma.
Howard JA and Dorjes J (1972) Animal-
The relative simplicity of this sediment relationships in two beach-
community is its distinctive feature, related tidal flats; Sapelo Island,
Georgia, J. Sed. Petrol. 42, 608-623.
and thus the major pathways within the Ingle JC (1966) The movement of beach
benthos-nekton part of the food web have sand. Elsevier Publishing Co., N.Y.
McDermott JJ (1979) The role of a
been eluc ida ted . The details and dis- spionid polychaete in the food web of an
cuss ion of the more complete web will be intertidal community on an exposed sandy
beach, Am. Zool. 19, 865 (Abst).
elaborated in a subsequent paper. McLachlan A (1980) The definition of
sandy beaches in relation to exposure: a
simple rating system, S. Afr. J. Sci.
ACKNOWLEDGEMENTS 76, 137-138.
I am indebted to the following from McLachlan A, Erasmus T, Dye AH, Wool-
dridge T, Van der Horst G, Rossouw G,
Franklin and Marshall College: H.B. Lasiak TA and McGwynne L (1981) Sand
Breneisen and H. Hayden for their help beach energetics: an ecosystem approach
towards a high energy interface, Estuar.
in fabricating equipment; R.A. Fluck and Coast. Shelf Sci. 13, 11-25.
C.R. Shearer, Jr. for aid with photo- Merriman D (1947) Notes on the mid-
summer ichthyofauna of a Connecticut
graphy; a large number of students, beach at different tide levels, Copeia
particularly C.J. Collette, P.J. de 1947, 281-286.
Roelofs EW (1954) Food studies of
Bruyn, M.H. Hassel, J.R. Orchardo, J.D. young sciaenid fishes, Micropogon and
Pickle, D.A. Russell and R. Triol, for Leiostomus, from North Carolina, Copeia
1954, 151 153.
help in the field. W.F. Smith-Vaniz of Saloman CH and Naughton SP (1978)
the Philadelphia Academy of Natural Bentbic invertebrates inhabiting the
swash zone of Panama Ci ty Beach,
Sciences kindly identified some of the Florida Northeast Gulf Sci. 2, 65-74.
juvenile fishes. I am thankful for Shelton CR and Robertson PB (1981)
Communi ty structure of intertidal
financial aid from Franklin and Marshall macrofauna on two surf-exposed Texas
College that made much of this study sandy beaches, Bull. Mar. Sci. 31,
833-842.
possible. I appreciate the many kind- Thomas DL and Milstein CB (1973) Eco-
ness rendered to me by the staff of the logical studies in the bays and other
waterways near Little Egg Inlet and in
Wetlands Institute of Lehigh University. the ocean in the vicinity of the pro-
REFERENCES posed site for the Atlantic Generating
Allen DM, Clymer JP III and Herman SS Station, New Jersey. Prog. Rept. for
(1978) Fishes of the Hereford Inlet the Period January December 1972,
estuary, southern New Jersey. The Wet- Parts 1 and 2, Ichthyological Asso-
ciates, Ithaca, New York, 1065 p.
538

Williams AB and Bynum KH (1972) A

ten-year study of meroplankton in North
Carolina estuaries: amphipods, Chesa-
peake Sci. 13, 175-192.

THE ECOLOGY OF SANDY BEACHES IN THE EASTERN CAPE, SOUTH AFRICA
ANTON McLACHLAN (Zoology Department, University of Port Elizabeth, P.O. Box 1600,
Port Elizabeth 6000, South Africa)
1. INTRODUCTION
The eastern Cape coastline is
characterized by extensive and productive
sandy beaches fringed by dynamic surf
zones and large dunefields. These beaches
of St Francis Bay and Algoa Bay (Fig. 1),
total 120 km, and all receive continuous
moderate to heavy swell. South facing
beaches are most exposed and 6 m breakers
are not uncommon, (McLachlan,1980a).
Maximum spring tide range is 2.1 m, the
intertidal zones average 50-80 m wide with
slopes generally 1/ 20 - 1/30 and sands are
fine to medium quartz with high calcium
carbonate content (McLachlan, 1977a, b).
Surf zones are 150-400 m wide, with well-
defined rip systems. The considerable
wave energy released in these zones is
responsible for driving surf circulation
patterns, pumping water through the inter-
stices of the sand body, transporting
sediment and facilitating the transport
and exchange of many biological materials.
Basic ecological research on these
systems started a decade ago at U.P.E.
with quantitative faunal surveys and has
subsequently progressed to encompass
studies on energetics and food webs,
nutrient cycling, oil pollution and
taxonomy. This paper aims at describing
these systems as we understand them at
present.
539
2. MACROFAUNA
The macrofauna consists of two
trophic groups, filter feeders and
scavenger/predators. Of five species
of filter feeders two bivalves of the
genus Donax are most important and
make up more than 95% of the biomass
(McLachlan 1977b, c). Thirteen
scavenger/predator species occur on
the beach and in the shallow subtidal.
Three species of whelk of the genus
Bullia are the most important of these.
Abundance and biomass have been
expressed per metre shoreline rather
than per square metre, in order to
avoid the problems that occur as a
result of changing beach profile
during rough and calm conditions.
Biomass values recorded on these
beaches (dry tissue mass) range <100-
>700 g. m- l with a mean of 1500 g. m- l
McLachlan 1977b, McLachlan et al.,
1981a). Surveys of sublittoral macro-
fauna are nearing completion.
Several aspects of the biology of
important mollusc species have been
studied. zonation (McLachlan, 1980b)
and tidal migrations (McLachlan ~. al.,
1979a) have been described and growth
and production has been quantified for
different populations (McLachlan, 1979
a; McLachlan et al., 1979b; McLachlan
540

van der Horst, 1979; McLachlan & Hanekom,
1979). All of the macrofauna are highly
mobile and, with the single exception of
Donax serra, undergo tidal migrations.
the crustaceans this includes a planktonic
phase at night. Donax serra exhibits semi
lunar movement up and down the beach with
spring and neap tides. Studies on respira-
tion and production have been used to
compile energy budgets for these species
(Dye, 1979a, 1980a; Dye, McGwynne, 1980)
and other physiological studies have inves-
tigated burrowing, feeding, temperature
tolerances, enzyme activity and ammonia
excretion (Ansell, 1981; Ansell,
McLachlan, 1980; Blackstock, Ansell, in
press; McGwynne, 1980; McLachlan, Young,
1982; Prosch, Mclachlan, 1983).
Wooldridge (1981) has investigated the
population biology of the mysid, Gastro-
saccus psammodytes, Sieben (unpublished)
has studied the giant air-breathing isopod,
Tylos capensis, du Preez (1983 a, b, c; Du
Preez, McLachlan 1983 a, b, c, d) investi-
gated ecology and feeding in the crab
Ovalipes and Cockcroft (1983) has studied
the biology and ecology of the prawn
Macropetasma. Ocypodid crabs have also
been recorded (McLachlan 1980c).
In
3. THE INTERSTITIAL FAUNA
The average beach has 10 7 I sea water
flushed through its interstices per kilo-
metre shoreline per day (McLachlan, 1979b).
Purification of this water occurs by mine
ralization of fine particulate and dis-
solved organics by the interstitial fauna.
This fauna consists of meiofauna (10 6 m- 2 ),
protozoans (10 4 cells. cc- l ) and bacteria
(10 8 cells. cc- l ) occurring to consider-
able depth in the sand body (McLachlan et.
al., 1979c) and constituting their own food
chain. Distribution of the mieofauna has
been studied in some detail and has been
shown to be stratified (McLachlan, 1977b,
c, 1980a; McLachlan, Furstenberg, 1977).
Greatest abundance and diversity occur
above the permanent water table between
the mean and high tide levels, as this is
where greatest percolation of intersti-
tial water occurs. This fauna can keep
pace with changing beach profile and
vertical migrations (McLachlan et al.,

"'"
Bird
Island
ALGOA BAY

34°S
~

A
Jeffries
Bay Receife ~

ST. FRANCIS BAY (j~

o
'ND\~~ o~,__~1~p ____2~p____3~p____4~p__~50 km
SANDY BEACHES ~

FIGURE 1. Map of study area.


1977a) in rhythm with the tidal cycle and
coupled to changes in interstitial
moisture and oxygen have been described.
Meiofauna distribution in the subtidal
has also been investigated (McLachlan et
al., 1977b) and there are also several
papers on meiofaunal systematics and
biology (McLachlan, 1977d, 1978a, 1978b,
1979c, McLachlan, Grindley, 1974,
McLachlan, Moore, 1978).
Nematodes and harpacticoid copepods
dominate the meiofauna, nematodes where
the sediment is finer and interstitial
oxygen lower (e.g. below the permanent
water table) while harpacticoids, being
more sensitive to low oxygen tensions,
are more abundant in coarser sands where
drainage is better and where the inter-
stitial water is well flushed.
The quantitative role of the inter-
stitial fauna in energy flow and nutrient
cycling in these beaches has been inves-
tigated by measuring total benthic oxygen
demand and attempting to partition this
between meiofauna, protozoans and
bacteria (Dye, 1979b, c, 1980b, 1981).
Though a black box approach, this has
yielded some useful results.
Besides studying the macro- and meio-
fauna of East Cape beaches, some surveys
have also been done further afield along
the east and south coasts of South Africa
(Dye et al. 1981; McLachlan et al., 1981a;
Wooldridge et al., 1981). Taking the coast
as a whole, significant relationships were
found between faunal diversity and abun-
dance and the slope and particle size of
beaches.
4. FOOD WEBS AND ENERGY FLOW
Most food chains associated with the
541
beach start and end in the sea, though
not the sea in general but rather the
surf zone in particular. Primary food
sources for macrofauna food chains are
living and dead particulate organics
from the surf zone and carrion, mostly
coelenterates, fishes, birds and mammals
from the open sea. Particulastes make
up about 95% of food input and carrion
about 5% (McLachlan et al., 1981b).
Windblown insects reaching the beach
are important but not yet quantified.
Predators on the macrofauna fall into
three categories, birds, fishes and the
swimming crab, Ovalipes. Of twenty
species of birds recorded on these
beaches four are important predators on
the macrofauna and consume about 30% of
macrofauna production (McLachlan et al.,
1980). These are the Kelp gull, Larus
dominicanus, the Black oystercatcher,
Haematopus moquini, the Sanderling,
Calidris alba, and the white-fronted
sandplover, Charadrius marginatus.
Terns and cormorants feed on mullett
in the surf and other waders occasion-
ally also prey on the macrofauna.
Sixty-six species of fishes occur
in these surf zones (Lasiak, 1981, 1982,
1983a, b). They may be divided into
six feeding categories, viz. detriti-
vores, herbivores, planktivores, pis-
civores, omnivores and benthic feeders.
Herbivores, omnivores and detritivores
are not typical of open beaches and the
three major groups are those that feed
on benthos, plankton and fish. Among
the benthic feeders, elasmobranchs are
the most important, particularly the
sands hark Rhinobatos annulatus and the
ray Myliobatus aquila (Rossouw 1983).
These benthic feeders are estimated to
542
take about 50-60% of macrofauna produc-
tion.
piscivores prey on other fishes,
particularly mullett, and planktivores
include a number of species feeding
mainly on zooplankton and in particular
the mysid, Mesopodopsis slabberi, and the
prawn, Macropetasma africana. These two
prey items are extremely abundant in our
inshore waters. Included in plankton
feeding must be the important phenomenon
of mullett feeding on surf phytoplankton
blooms. Direct grazing of mullett on
these blooms is an important form of
telescoping of the surf zone food chain.
The three-spot swimming crab,
Ovalipes punctatus, is an important
predator on Donax and Bullia (du Preez
1983c). It enters the intertidal on the
high tide and has been shown to be highly
efficient at locating, selecting and
opening the shells of its prey. An
energy budget has been estimated for this
species and it is estimated to take about
10% of macrofauna production.
The details of energy flow through
the intertidal have already been put
together in the familiar energy circuit
language (McLachlan et al., 1981b) but we
are still far from being able to do this
for the surf zone.
5. NUTRIENT CYCLING
One of the main functions of sandy
beaches is the return of inorganic
nutrients to the sea. The beach can
produce nutrients in a number of ways:
excretion by the macrofauna, birds, fishes
and plankton; mineralization by the
interstitial fauna; and the seaward
seepage of ground water rich in nutrients.
Ammonia excretion in Donax ~ (Prosch,
McLachlan, 1983) has been measured.
The interstitial system is important in
nutrient regeneration through the process
of mineralization of dissolved and
particulate organics~ Besides earlier
theoretical estimates of this (McLachlan
et al., 1981c), coupling these measure-
ments to a model of water filtration by
the beach it was possible to attempt some
'accurate' estimates of nutrient regenera-
tion rates (McLachlan, 1982).
It appears that these beaches regene-
rate significant amounts of inorganic
nutrient for the surf zone. The intersti-
tial system is most important in this
process, followed by the zooplankton,
macrofauna, and nekton.
6. SURF PHYTOPLANKTON BLOOMS AND THE
BEACH/SURF ZONE AS AN ECOSYSTEM
An important feature of most of these
beaches is the regular occurrence of blooms
of a surf zone diatom, Anaulus birostratus,
accompanied in low numbers by the larger
species Aulacodiscus petersii (McLachlan,
Lewin, 1981). These blooms form over rip
currents by day as a result of buoyancy of
the cells and are important as a food
source for the filter feeding macrofauna,
zooplankton and mullett. Their presence
in the surf is thought to be due to 1)
their ability to maintain position over rip
currents during the day and 2) the steady
supply of inorganic nutrients regenerated
by the beach. Though nutrient levels in
the surf never become very high, their
continual production by the faina and the
interstitial system results in a constant
supply and this is believed to allow the
build-up of the blooms.
The normal circulation pattern in the

surf of open beaches is of a cellular
nature, implying that nutrients will tend
to be retained in this. zone. as will the
phytoplankton. This coupling between the
beach and surf zone (and discreteness of
the outer boundary of the surf zone in
the form of surf circulation cells)
suggested the idea of a beach/surf zone
ecosystem. Under this hypothesis the
sand body of the beach and the water
envelope of the surf zone are considered
a viable and semi-closed system with the
drift line and the outer limit of circu-
lation cells being the boundaries.
Within these boundaries phytoplankton,
macrofauna and zooplankton and
interstitial fauna may be considered to
represent the producers, consumers and
decomposers of a 'textbook ecosystem'
(McLachlan, 1980a; McLachlan et al.,
1981b).
This hypothesis does not ignore the
importance of exchanges with the dunes
and open sea, it merely encourages a more
holistic approach towards the beach and
surf zone.
7. DUNE/BEACH INTERACTIONS AND OIL
POLLUTION
Amongst the dunes bordering these
beaches, the Alexandria dunefield,
between the Sundays River and Woody Cape,
is the most extensive, totalling >100 km 2
and with dunes >150 m in height. Its
sand is derived from the beach and it forms
part of an aeolian conveyor belt of
littoral sand up our east coast. Various
biotic interactions occur between the dunes
and beach, both in the movements of mobile
predators and in the supply of primary food
sources. Insects blowing onto the beach
represent an important food source at
times. Further, air breathing isopods and
543
amphipods, which formn large populations
in the dunes, move onto the driftline to
feed at night. Many birds which nest or
roost in the dunes move into the inter-
tidal to feed and several mammals from
the dunefield also patrol the beach for
food. Several aspects of this interface
have been investigated (McLachlan et al.,
1982; Randall, McLachlan, 1982) and work
is continuing.
Local effects of ore dust pollution,
which occurs near the Port Elizabeth
harbour, have been studied on supralittoral
meiofauna (McLachlan, 1977e). Effects of
strandings of crude oil on beach intersti-
tial fauna and water filtration processes
have also been evaluated (Harty, McLachlan
1982, Mclachlan, Harty, 1981, 1982).
Because of their high energy nature these
beaches are resilient to oil pollution and
recover within months from even the
heaviest spills.
8. PRESENT AND FUTURE WORK
The emphasis of recent work has
shifted from the intertidal to the surf
zone. While most macrofauna biomass is
concentrated in the intertidal, large
concentrations of phytoplankton, zoo-
plankton and fishes occur in the surf
zone (Table 1).
The lack of attention these mobile
components have received in the past
was largely due to their inaccessibility
and the difficulty of working in the
surf zone. The high turnover of zoo-
plankton (P/B = 6-7) make them an
extremely important component of the
beach/surf zone biota. We have become
increasingly aware of the quantitative
importance of the surf zone and several
projects have been initiated in this
544
TABLE 1. Summary of approximate macrofaunal
dry biomass in a metre wide transect through
a typical East Cape beach and surf zone.
Category Biomass (g.m- l -)
Macrobenthos 1500
Zooplankton 1100
Fishes 300
Birds 5 Dye AH (1980a) Aspects of the respira-
area. Fishes (Lasiak, 1982) elasmobranchs
(Rossouw, 1983) shrimps and prawns (Cock-
croft, 1983; Wooldridge, 1983) have been
extensively studied and work is continued
on surf zone benthos, larger zooplankton,
food chains associated with phytoplankton
blooms (Romer, in prep.) and the ecology
and primary productivity of the phyto-
plankton itself (Sloff, Talbot, in prep.).
The beach/surf zone is a wave driven
system and full understanding and
modelling of this system depends on a
knowledge of both the biological and the
physical structure and dynamics of this
zone. An interdisciiplinary approach is
thus essential. We are therefore
confident that the collaborative program
that we started with physical oceanogra-
phers in 1982 will lead to more complete
understanding of these exciting and
dynamic ecosystems.
9. REFERENCES
Ansell AD (1981) A note on the biology
of Donax serra Roding and Donax sordidus
Hanley (Bivalvia : Donacid~J. Moll.
Stud. 47, 59-72.
Ansell AD and McLachlan A (1980) Upper
temperature tolerances of three molluscs
from South African sandy beaches, J. expo
mar. BioI. Ecol. 48, 243-251.
Blackstock J and Ansell AD (in press)
Maximal glycolytic enzyme activities in
muscles of bivalves of the genus Donax in
relation. to their behaviour and ecology.
In European Society for Compo Physiol.
Biochem. Oxford, Pergamon pp.132-133.
Cockcroft A (1983) The biology of
Macropetasma africa, M.Sc. thesis, Uni-
versity of Port Elizabeth.
Dye AH (1979a) The effect of acute and
long term temperature changes on the
respiratory rate of two sand-dwelling
bivalves, Compo Biochem. Physiol. 3B,
405-409.
Dye AH (1979b) The measurement of bio-
logical oxygen demand in sandy beaches,
S. Afr. J. Zool. 14, 55-60.
Dye AH (1979c) Quantitative estimation
of protozoa from sandy substrates,
Estuar. cstl. mar. Sci. 8, 199-204.
tory physiology of Gastrosaccus psammody-
tes Tattersall (Crustaces: Mysidacea),
Compo Biochem. Physiol. 65A, 187-191.
Dye AH (1980b) Tidal fluctuations in
biological oxygen demand in exposed sandy
beaches, Estuar. cstl. mar. Sci. 11, 1-8.
Dye AH (1981) A study of benthic oxygen
consumption on exposed sandy beaches,
Estuar. cstl. Shelf Sci. 13, 671-680.
Dye AH and McGwynne L (1980) The
effect of temperature and season on the
respiration of three psammolittoral
gastropods, Compo Biochem. Physiol.
66A, 107-111.
Dye AH, McLachlan A and Wooldridge
T (1981) The ecology of the sandy
beaches of Natal, South Africa, S.
Afr., J. Zool. 16, 200-209.
Du Preez HH (1983a) The effects of
temperature, season and activity of the
respiration of the three-spot swimming
crab, Ovalipes punctatus, Compo Biochem.
Physiol. (in press).
Du Preez HH (1983b) Consumption,
assimilation and energy balance in the
three spot swimming crab, Ovalipes
punctatus (de Haan) In A. " McLachlan &
T Erasmus eds. Sandy Beaches as
Ecosystems, Junk, The Hague.
Du Preez HH (1983) Feeding in the
three-spot swimming crab, Ovalipes
punctatus (de Haan) in the East Cape.
Sumbitted.
Du Preez HH and Mclachlan A (1983a)
Biology of the three-spot swimming crab,
Ovalipes punctatus (de Haan). I.
Morphometrics and relative growth,
Crustaceana (in press).
Du Preez and McLachlan A (1983b)
Biology of the three-spot swimming crab,
Ovalipes punctatus (de Haan). II.
Growth and moulting, Crustaceana (in
press).
Du Preez HH and McLachlan A (1983c)
Biology of the three-spot swimming crab,
Ovalipes eunctatus (de Haan). III.
Reproductlon, fecundity and early larval
development, Crustaceana (in press).
Du Preez HH and McLachlan A (1983d)
Seasonal changes in biochemical
composition and energy content of the

three-spot swimming crab, Ovalipes
punctatus (de Haan), J. expo mar. BioI.
Ecol. (in press).
Harty B and Mclachlan A (1982) Effects
of water soluble fractions of oil and
dispersants on nitrate generation by
sandy beaches, Mar. Pollut. Bull., 13,
287-29l.
Lasiak TA (1981) Nursery grounds of
juvenile teleosts: evidence from the
surf zone of Kings Beach, Port
Elizabeth, S. Afr. J. Sci. 77, 388-390.
Ladsiak TA (1982) Structural and
functional aspects of the surf zone fish
community in the East Cape, Ph.D.
thesis, University of Port Elizabeth.
Lasiak TA (1983a) Recruitment and
growth patterns of juvenile marine
teleosts caught at Kings Beach, Algoa
Bay, S. Afr. J. Zool. 18, 25-30.
Lasiak TA (1983b) The impact of surf
zone fish communities on faunal
assemblages associated with sandy beaches
In A Mclachlan & T Erasmus eds. Sandy
Beaches as Ecosystems, Junk, The Hague.
McGwynne L (1980) A comparative eco-
physiological study of three sandy beach
gastropods in the Eastern Cape, M.Sc.
thesis, University of Port Elizabeth.
McLachlan A (1977a) Studies on the
psammolittoral meiofauna of Algoa Bay. 1.
Physical and chemical evaluation of the
beaches, Zool. afro 12, 15-32.
McLachlan A (1977b) Composition, dis-
tribution, abundance and biomass of the
macrofauna and meiofauna of four sandy
beaches, Zool. afro 12, 279-306.
Mclachlan A (1977c) Studies on the
psammolittoral meiofauna of Algoa Bay.
II. The distribution, composition and
biomass of the meiofauna and macrofauna,
zool. afro 12, 33-60.
McLachlan A (1977d) Larval development
and population dynamics of Derocheiloca-
ris algoensis (Mystacocarida), Zool. afro
12, 1-14.
McLachlan A (1977e) Effects of ore dust
pollution on the physical and chemical
features and on the meiofauna and micro-
fauna of a sandy beach, Zool. afro 12,
73-88.
McLachlan A (1978a) Sediment particle
size and body size in meiofaunal harpac-
ticoid copepods, S. Afr. J. Sci. 74,
27-28.
McLachlan A (1978b) A temporal study of
niche breadth and overlap in two sympa-
tric species of Mystacocarida (Crustacea)
Zool. afro 13, 351-353.
McLachlan A (1979a) Growth and produc-
tion of Donax serra Hanley on an open
sandy beach in Algoa Bay, S. Afr. J.
Zool. 14, 61-66.
Mclachlan A (1979b) Volumes of sea
water filtered by East Cape sandy
545
beaches, S. Afr. J. Sci. 75,75-79.
McLachlan A (1979c) The identity of
Derocheilocaris remanei katesae
(Mystacocarida) from South Africa,
Crustacean a 37, 223.
McLachlan A (1980a) The definition of
sand beaches in relation to exposure: a
simple rating system, S. Afr. J. Sci.
76, 137-138.
McLachlan A (1980b) Intertidal zona-
tion of macrofauna and stratification of
meiofauna on high energy beaches in the
Eastern Cape, South Africa, Trans. roy.
Soc. S. Afr. 44, 213-233.
McLachlan AS (1980c) The occurrence of
ghost crabs, Ocypode spp. in the East
Cape, S. Afr. J. Zool. 15, 57-58.
Mclachlan A (1980d) Exposed sandy
beaches as semi-closed ecosystems,
Marine Environ. Res. 4, 59-63.
McLachlan A (1982) A model for the
estimation of water filtration and
nutrient regeneration by exposed sandy
beaches, Mar. Environ. Res. 6, 37-48.
Mclachlan A and Grindley JR (1974) A
new species of Mystacocarida (Crustacea)
from Algoa Bay, South Africa, Ann. S.
Afr. Mus. 66, 169-175.
McLachlan A and Furstenberg JP (1977)
Studies on the psammolittoral meiofauna
of Algoa Bay. III. A quantitative
analysis of the nematode and crustacean
communities, zool. afro 12, 61-72.
McLachlan A, Wooldridge T, Schramm M
and Kuhn M (1980) Seasonal abundance,
biomass and feeding of shore birds on
sandy beaches in the East Cape, South
Africa, Ostrich, 51, 44-52.
McLachlan A, Wooldridge T and Dye AH
(1981a) The ecology of sandy beaches in
Southern Africa, S. Afr. J. Zool. 16,
219-2132.
McLachlan A, Erasmus T, Dye AH,
Wooldridge T, van der Horst G, Rossouw G,
Lasiak TA and McGwynne L (1981b) sand
beach energetics: an ecosystems approach
towards a high energy interface, Estuar.
cstl. shelf. Sci. 13, 11-25.
McLachlan A, Dye AH and Harty B (1981c)
Simulation of the interstitial system of
exposed sandy beaches, Estuasr. cstl.
Shelf Sci. 12, 267-278.
McLachlan A and Harty B (1981) Effects
of oil on water filtration by exposed
sandy beaches, Mar. Poll. Bull. 12,
374-378.
McLachlan A and Lewin J (1981) Observa-
tions on surf phytoplankton blooms along
the coiasts of South Africa, Bot. Mar.
24, 553-557.
Mclachlan A and Harty B (1982) Effects
of crude oil on the supral{ttoral
meiofauna of a sandy beach, Mar. Environ.
Res. 7, 71-79.
McLachlan A and Young N (1982) Effects
546

of low temperature on the burrowing rates

of four sandy beach molluscs, J. expo
mar. Biol. Ecol. 65, 275-284.
McLachlan A, Sieben P and Ascaray C
(1982) Survey of a major coastal
dunefield in the Eastern Cape. U.P.E.
Zoology Department Research Report No.
10. 70pp.
Prosch R and Mclachlan A (1983) Ammonia
excretion in Donax serra and its role in
the regeneration of surf zone nutrients.
(In prep).
Randall RM and McLachlan A (1982)
Damara terns Sterna balaenarum breeding
on the south-eastern Cape coast, Ostrich
53, 50-51.
Rossouw GJ (1983) Studies on elasmo-
branchs in Algoa Bay, with special
reference to the sandshark, Rhinobatos
annulatus, Ph.D. thesis, University of
Port Elizabeth.
Wooldridge T (1978) Two new species of
Gastrosaccus (Crustacea, Mysidacea) from
sandy beaches in Transkei, Ann. S. Afr.
Mus. 76, 309-327.
Wooldridge T (1981) Zonation and
distribution of the beach mysid,
Gastrosaccus psammodytes, J. Zool. Lond.
193, 183-189.
Wooldridge T (1983) Ecology of beach
and surf zone mysid shrimps in the
Eastern Cape, South Africa In A
McLachlan & T Erasmus eds. Sandy Beaches
as Ecosystems, Junk, The Hague.
Wooldridge T, Dye AH and Mclachlan A
(1981) The ecology of sandy beaches in
Transkei, S. Afr. J. Zool. 16, 210-218.

KELP ~JRACK AND THE FLOW OF ENERGY THROUGH A SANDY BEACH ECOSYSTEM
C.L. GRIFFITHS, J.M.E. STENTON-DOZEY and K. KOOP (Zoology Department, University of Cape Town,
Rondebosch 7700, South Africa)
1. INTRODUCTION
One of the most distinctive and frequently cited
features of exposed sandy beaches as ecosystems
is their almost complete lack of in ~itu primary
production (e.g. Brown, 1964; Munro et al., 1978;
McLachlan et al., 1981a). This results directly
from the mobility of the sediments, which renders
them unsuitable for the attachment of macrophytes
or the development of dense benthic diatom
communities. In the absence of these potential
food resources macrofaunal organisms on exposed
beaches must obtain their nutrition from imported
materials. These may take the form of finely
divided detritus or phytoplankton particles,
many of which may originate from rich blooms of
phytoplankton in the surf zone (McLachlan et al.,
1981b), or large items of carrion. The macro-
fauna associated with these food resources
normally comprises two major trophic elements,
namely filter feeders, including bivalves (such
as Vonax) and crustacea (such as Em~a and
G~tno~aee~)j and predator/scavengers (e.g.
Bullia, Oeypode, various isopods and polychaetes).
The proportion of these two trophic elements
differs widely, depending on the relative
availability of plankton and detritus versus
carrion (McLachlan et al., 1981a), but filter
feeding bivalves are frequently the overwhelming
dominants in terms of biomass (e.g. McLachlan
et al., 1981b).
Certain beaches along the west coast of South
Africa deviate from this generalized scheme in
that they receive vast energy subsidies in the
547
form of macrophytes uprooted from the highly
productive kelp beds that dominate rocky shores
along the coast (Field et al., 1980; Newell et
al., 1982). This material provides a rich food
supply that is concentrated high up the beach
(whereas the food supply of most intertidal
organisms is more available lower down the shore)
and hence profoundly affects both the nature and
distribution of the fauna. Various aspects of
the ecology of beaches with high kelp input have
recently received attention (Muir, 1977; Koop,
Field, 1980; Stenton-Dozey, Griffiths, 1980;
Griffiths, Stenton-Dozey, 1981; Koop et al.,
1982 a,b), making it possible to compute various
aspects of energy flow through such a system.
In this study we aim to synthesize the available
information on high kelp-input beaches and to
compare the composition, biomass and distribution
of the fauna, and the energy flow pattern, with
those described for open sandy beaches in the
Eastern Cape by McLachlan et al. (1981b). These
authors suggest that the beach and surf zone may
represent a more or less closed system in which
surf zone phytoplankton, the main producers, are
fed upon by a rich macrofauna dominated by filter
feeding bivalves. Nutrients regenerated by the
interstitial fauna and by the.macrofauna are
returned to the sea "in repayment of organic
imports" and in turn support the surf zone
phytoplankton.
2. RESULTS AND DISCUSSION
2.1 Description of study site
548
The data presented here were all collected from
one of a number of short beaches interspersed
between rocky headlands at Kommetjie (34°08'S,
18°19'E) on the west coast of the Cape Peninsula,
South Africa. The profile and sediment
characteristics of one such beach are described
by Koop and Griffiths (1982) and the fauna and
flora of the adjacent kelp beds by Field et al.
(1980). Wave characteristics of a nearby site
are given by Velimirov et al. (1977). On
average the intertidal zone is some 45 m wide,
with a slope of about 1:20, and the median grain
size of the sand varies between about 240 ~m and
280 ~m. Wave height is normally between 1 and
4 m and water temperatures between 8 and 15°C.
During our monthly visits to the study site air
temperatures varied from 26°C (Jan.) to 13,7°C
(July) with a mean of 19,3°C. The temperature
range beneath the wrack was lower, with a maximum
of 21°C, a minimum of 12°C and a mean of 16,5°C.
2.2 Rates of food supply to the beach
Rates of wrack deposition at Kommetjie beach have
been estimated by Koop and Field (1980) and by
Stenton-Dozey and Griffiths (1983). Two main
factors complicate these measurements. Firstly
the wrack is extremely unevenly deposited along
the length of the beaches, and secondly an estimate
of residence time is required if standing stock
estimates are to be converted into annual rates
of input.
Koop and Field (1980) measured the wet mass of
wrack lying on a randomly chosen 5 m wide strip
of beach at monthly intervals for 17 months.
Minimum values of about 2 kg m- l occurred in
summer and maxima of over 40 kg m- l during winter.
Using an 8 day residence time Koop et al. (1982a)
have used these figures to calculate the yearly
input of kelp as 1200 - 1800 kg wet mass m- l y-l.
In an attempt to verify these figures and to
correct for the extreme patchiness of wrack
deposits a further 12 months data have been
collected by Stenton-Dozey and Griffiths (1983)
at an adjacent site. In this case overall
estimates were made of the volume of wrack
deposited over a 300 m stretch of beach and
supplemented by weighings at selected sites. The
results show a similar seasonal cycle of deposi-
tion, with high values in autumn and winter and
lower ones in summer. The mean standing stock
of kelp on the beach was 25,07 kg m- l • Using
a residence time of 14 days, based on degradation
experiments conducted by Griffiths and Stenton-
Dozey (1981), this gives a total deposition rate of
2179 kg wet mass m- l y-l. This may be converted
to energy equivalents on the basis of energy
values derived from Newell et al. (1982),
assuming the wrack to comprise equal proportions
of the kelps Laminania patt£da and Eekionia
maxima, and gives a value of 4,07 x 10 6 kJ
deposited per running metre of beach each year.
2.3 Nature, distribution and biomass of the
macrofauna
The intertidal biota of Kommetjie beach has been
surveyed by Koop and Griffiths (1982) and the
macrofauna within the wrack beds described by
Griffiths and Stenton-Dozey (1981) and Stenton-
Dozey and Griffiths (1983). The macrofaunal
distribution patterns obtained by Koop and
Griffiths (1982) are typical and are summarized
in Table 1.
Of the fifteen species (7 Coleoptera, 2 Diptera,
2 Amphipoda, 2 Isopoda, 1 Oligochaeta, 1
Polychaeta) recorded overall, all but three were
directly associated with kelp wrack on the
driftline. As a result there was a marked
increase in species diversity as one moved up
the beach from the low water mark, which is the
reverse of the normal pattern. Recent results
by Stenton-Dozey and Griffiths (1983) have
increased the number of species recorded amongst
kelp wrack at Kommetjie to 35 (of which 22 are
insects), suggesting that the increase in diversity

TABLE 1 Species richness, abundance and biomass of macrofaunal organisms along a transect across
Kommetjie beach (from Koop and Griffiths, 1982).
Zone Saturation Resurgence
Distance from LWS (m) o - 12 12 - 28
No. of species recorded 2 0
Abundance (Nos. m- 2 ) 8 0
Biomass (g dry mass m- 2 ) 0,31 0.0
upshore may be even more marked than shown here.
It is significant to note that many of the resi-
dents typical of lower tidal reaches on adjacent
beaches of similar sediment type, such as Vonax,
Butfia and G~~o~accu¢ have not been found in
these surveys. In terms of biomass a similar
trend is evident, with exceptionally high values
occurring amongst damp wrack along the recent
driftline and much smaller ones low on the beach.
While these figures are useful for comparative
purposes they were based on a 'once-off' survey
taken at the height of summer (Dec - Jan). A mean temperature beneath the wrack is 16,5°C.
subsequent year-long survey (Stenton-Dozey.
Griffiths, 1983) has shown that the biomass of the
macrofauna varies greatly during the year, with
maximum values during winter, so that the mean
values are considerably higher than previously
supposed. Mean annual standing stocks of the
major macrofaunal groups derived from these data
are given in Table 2. The Amphipoda recorded
during this survey consisted almost entirely of
drift-line species of the genus TafohChehtia, and
these made up over 90% of the total macrofauna.
Since the insects were also intimately associated
with drift algae the wrack fauna comprised over
97% of the total intertidal macrofaunal biomass.
2.4 Consumption by primary consumers
Energy budgets are available for the two most
important herbivores on Kommetjie Beach - the
amphipod TafohChehtia capen6~ (Muir, 1977) and
549
Retention Current Storm
driftline driftline
28 - 35 35 - 49 49 - 63
1 9 11
6 1894 156
0.01 15.37 1.57
the larvae of the kelp-fly FuceteIa capen6~
(Stenton-Dozey, Griffiths, 1980).
The consumption rates of TafOhChehtia capen6~
given by Muir (1977) vary considerably both with
temperature and animal size. At 17°C, for
example, a 1 mg dry mass amphipod consumes 50% of
its body mass in dry kelp per day, while one of
20 mg takes less than 20%. Equivalent figures
at 21°C would be 70% and 25%. The mean mass of
individual T. capen6~ on Kommetjie beach is about
8 mg (J.M.E. Stenton-Dozey, unpublished) and the
Using these figures we have read off a mean
consumption rate of 20% dry body mass d- 1 • Given
the energy equivalent for TafMChehtia (~luir,
1977) of 18,13 kJ g-l, the standing stock of
amphipods on the beach thus has an energy value
of 37765kJ m- 1• The consumption of kelp by
this population would be 417 g dry kelp per day
or 152 kg per year. Using an energy equivalent
for kelp of 14,12 kJ g-l dry mass (Newell et al.,
1982) this is equal to 2,146 x 10 6 kJ m- 1 y-1.
From the energy budget equation of Muir (1977)
this is apportioned by the amphipods as shown in
Table 3.
A similar analysis may be made for larvae of the
kelp fly FuceteIa capen6~ from data given by
Stenton-Dozey and Griffiths (1980). Fly larvae
have an energy value of 25,8 kJ g-l and consume
1,8 times their dry mass in kelp per day. Since
the mean population of larvae is 63 g m- 1 (or
550

1625,4 kJ m- 1), total consumption is 63 x 1,8 x organisms eat their own dry flesh mass in kelp
365 g or 41,4 kg dry kelp m- 1 y-1, which is per day. Assuming this to be 10% of the whole
equivalent to 0,584 x 10 6 kJ (27% of the figure mass (flesh + shell) this would be 1,1 x 365 g or
for amphipods). Ingested energy is expended as 402 g dry kelp m- 1 y-1 for the herbivores and
shown in Table 3. 292 g dry mass for the filter feeders, giving a
total of 694 g or 9799 kJ m- 1 y-1. For the
TABLE 2 Mean annual standing stocks of macro- energy value of small molluscs we have taken the
faunal organisms on Kommetjie beach
(after Stenton-Dozey, Griffiths, 1983). value of 1,77 kJ g-1 whole dry mass given for
mixed small molluscs by Fiel d et al., 1980, which
Mean Annual Biomass gives a value of 34 kJ m- 1 beach.
Macrofaunal Group (g dry mass m- 1)
2.5 Consumption by carnivores
Herbivores
The herbivorous macrofauna on Kommetjie Beach are
Amphipoda 2083
preyed upon by a variety of wading and non-wading
Dipteran larvae 63
birds, by carnivorous isopods (mainly Eunydi~e
Herbivorous Coleoptera 28
fongi~on~ and Exo~phaenoma tnun~atZtef~on) and
Molluscs 11
by a variety of carnivorous Coleoptera of the
Filter feeders families Staphylinidae, Histeridae, Tabaridae and
Bivalve molluscs 8 Carabidae.
Carn i vores The distribution and numbers of coastal waders
Isopoda 42 found along the beaches of the south-western Cape
Carnivorous Coleoptera 21 are given by Summers et al. (1977) and from their
Total 2256 unpublished data we have extracted the figures
specifically for Kommetjie beach. Equivalent
counts for non-wading birds have been obtained
We -do not have equivalent data for the herbivorous from unpublished records of the Western Cape Wader
Coleoptera, which have a biomass of 28 g m- 1 of Study Group. The feeding ecology, mass, and
beach (1,3% of total herbivore biomass). For energy requirements of each species are given by
energy budgeting purposes we have simply assumed Hockeyet al. (1983). From these figures the
that they have an energy equivalent of 15,0 kJ g-1 annual energy requirements of the avifauna can be
and eat their own body weight of kelp per day calculated. A number of assumptions must be made
i.e. 28 x 365 g = 10 kg m- 1 y-1 or as regards the lengths of time migratory species
141 200 kJ m- 1 y-1. feed on South African beaches and the dependency
of non-waders on intertidally collected foods, and
The mollusc fauna has also not received attention, Hockey et al. (1983) have been followed in this
since it makes up less than 1% of total consumer regard. The numbers, biomass, and energy
biomass (in contrast to the situation in the areas requirements of the avifauna of Kommetjie beach,
studied by McLachlan et al., 1981b, and others). as derived from the above sources, are given in
The only molluscs recorded were small « 1 mm) Table 4. The total of 357 birds found per km of
bivalves provisionally identified as Neogaimand{a beach have a biomass of 54738 g (wet). Given a
Rowie~~, and minute herbivorous gastropods, dry to wet mass conversion of 0,4:1 and energy
mostly Eatonietia nigna. For the purposes of value of 22,5 kJ g-1 dry mass (McLachlan et al.,
this analysis we have again i "d these 1980) this is equal to 21,9 g dry mass m- 1 or
 551

TJ),BLE 3 Energy budgets for TatotLc.huua c.ape.M-u' and Fuc.e.LUa Mpe.M-u' on Kommetjie beach (after
Muir, 1977 and Stenton-Dozey, Griffiths, 1980).

C Pg + Pr + R + F + U

TatotLc.hu.:Ua c.ape.n~-u'
kJ m- 1 y-1 x 10 3 4293 81 + 26 + 378 2095 1713
% 100 1 ,9 0,6 8,8 48,8 39,9
Fuc.e1w c.ape.M-u'
kJ m- 1 y-1 x 10 3 584 105 123 356
% 100 18 21 61

493 kJ m- 1 beach. The total energy requirements
of these birds, derived from Table 4, is 16094
kJ m- 1 y-1 which is 33 times their standing stock
and represents 40% of the invertebrate biomass or
16% of production, assuming a P~ ratio of 2,5
(McLach 1an, 1977; Koop, Griffi ths, 1982).
Since the assimilation efficiency of wading birds
is approximately 73% (Hockey et al., 1983) they
probably return some 4000 kJ m- 1 y-1 to the beach
in the form of faeces.
No energy budget data are available for the
carnivorous isopods ElLfLydic.e. long~c.otLn-L6 or
Exo~phae.!Loma ttLunc.atLt~on, although various
aspects of their ecology have been described by
Brown (1973). Both species are known to feed
on TalotLc.huua and other amphipods, dead or
dying molluscs and virtually any other available
animal material. The only South African
carnivorous isopod for which consumption estimates
can be derived is the subl ittoral C~olana~po~Ua,
which has been studied by Shafir and Field (1980).
These authors give the following energy budget
for a population of mean standing stock
394 kJ m- 2 (26,8 g dry mass) at Oudekraal, on the
Cape Peninsula.
C Pg + Pr + R
kJ m- 2 y-1 9899 1891 + 338 + 1822
% 100 19 + 3 + 18
Note that the consumption estimate in this equa-
tion exceeds the combined total for energy
expenditure by a factor of two, suggesting that C
was considerably overestimated, a common failing
in laboratory feeding experiments (Shafir, Field,
1980). In the light of this probable overesti-
mate and Johnson's (1976) estimate that C in an
intertidal species of C~olana is only 6,78 x B
we have taken annual consumption rate to be ten
times the standing stock, rather than the 25 x
calculated from Shafir and Field (1980). Given
the biomass of carnivorous isopods on the beach
(42 g m- 1) and using the energy equivalent for
C. ~po~Ua given by Shafir and Field (1980),
the standing stock of isopods is 617 kJ m- 1 and
total annual consumption is 6166 kJ m- 1 y-1, which
is equivalent to 15% of the standing stock of
primary consumers on the beach or 6% of their
annual production. We consider this to be a
maximum estimate since both the major isopod
species undoubtedly obtain an appreciable percent-
age of their food in the form of carrion washed
up on the beach and from prey which are of
sublittoral origin.
+ F + U
+ 977 +-
+ 10 + (50 - by subtraction)
552

TABLE 4 Abundance, biomass and energy requirements of the avifauna of Kommetjie Beach. Abundance
of wading birds after Summers et al. (1977) and for non-wading birds from unpublished data
of \Jestern Cape ~iader Study Group. Biomass and daily energy requirements from Hockey et
al. (1983). Only birds with an abundance of more than one individual per km are included
in the analysis

Energy Annual
No. birds Biomass km- 1 expenditure
Species per km (g wet mass) per bird expenditure
" (kJ day-1) Kj km- 1 y-1 x10 3

Resident waders
Black oystercatcher 4,8 3 321 641,8 124
White fronted plover 9,3 418 107,5 365

Mi gran t waders
Ringed plover 2,0 97 110,6 33
(x150 days)
Curlew sandpiper 58,2 3 259 124,0 082
(x150 days)
Sanderling 143,2 7 876 122,6 2 633
(x150 days)
Common sandpiper 1,0 57 125,5 19
(x150 days)
Non-waders
Hartl aubs gull 133,8 39 604 367,9 10 780
(x 60%)
Cape wagta il 4,6 106 69,4 58
(x 50%)

Total 356,9 54 738 16 094

No data are available for the carnivorous Coleop-
tera, other than descriptive accounts of the diets
of various species (Backlund, 1945; Cheng, 1976).
We have thus assumed their energy equivalent to be
15 kJ g-1 dry mass, as for herbivorous forms, and
their consumption rate to be 10 times their stand-
ing stock per year. This gives a standing stock
1
of 315 kJ m- and consumption estimate of
3150 kJ m- 1 y-1.
2.6 Energy flow through the macrofaunal community
From the results presented above it is possible
to draw up an energy flow diagram for the high
kelp-input beach at Kommetjie (Fig. 1). As in
other beaches the macrofauna relies entirely on
material imported from the sea to meet its
nutritional requirements. The most important
component of these imports at Kommetjie is
undoubtedly kelp wrack, which has an energy
equivalent of over 4 x 10 6 kJ m- 1 y-1. This is
more than 30 times the combined input of detritus,
phytoplankton and carrion reported for a beach
near Port Elizabeth by McLachlan et al. (1980b).
Although small amounts of carrion (largely dead
sponges and ascidians) were observed on Kommetjie
beach we did not record surf zone phytoplankton
blooms there. It thus seems reasonable to assume
that the enormous influx of kelp accounts for at
least 95% of the food supply of the macrofauna in
th i s a rea.
Since the major food resource, kelp wrack, is
 553

1----- ------------
I
I

z
Q KELP WRACK
c: 1 198 800
c:
~
4 070 000

0
I-
L1J
..J
III
<I:
::! <I:
<I: Z
> :::l
<I: <I:
II.
a:
0 ..J
<I:
<I:
L1J I-
(J) I-
(J)
0 a:
I- L1J
Q I-
L1J ~
Z
a:
:::l
I-
L1J
a:

BIRDS

CARNIV. COLEOPTERAr =====~::4~9~3::~========~==~

315 F
ISO PODS
I
617 1- __________ ._______ I

FIGURE 1. Energy flow diagram for Kommetjie beach. Figures in boxes are mean annual
standing stocks in kJ m- 1 , or in the case of kelp wrack, annual input in
kJ m-1 y-1, Arrows show annual flows in kJ m- 1 y-1.

deposited high on the beach, the macrofauna is
concentrated around the drift line and is composed
largely of terrestrial or semi-terrestrial
species. For this reason species diversity, as
well as biomass, tend to decline towards the low
water mark. Amphipods of the genus Talo~ch~t1a
are the major herbivores (95% by energy) and con-
sume 52,7% of the kelp deposited on the beach, as
opposed to 14,5% for kelp fly larvae and 3,5% for
herbivorous Coleoptera. Total calculated
consumption for the herbivores thus amounts to
70,7% of annual kelp deposition. The remaining
29,3% is thought to be degraded by bacteria and
of the material they eat is also returned to the
beach in the form of faeces or excretory products.
The organic input into the sand column is thus
relatively insensitive to variations in the
proportions of wrack decomposing or being eaten
by macrofaunal herbivores (Koop et al., 1982a).
Three principal groups of macrofaunal predators
occur on the beach - birds, isopods and
carnivorous Coleoptera. All three groups have
similar standing stocks, but the birds are the
most important predators because of their high
energy requirements. vJe estimate that the

to either wash back into the sea at high tide or
enter the sand column in dissolved or particulate
form. The assimilation efficiency of both the
major herbivores is, however, low, so that much
avifauna removes some 40% of herbivore standing
stock and the Coleoptera about 8%. Although the
energy requirements of the isopods are equivalent
to 15% of herbivore standing stock, much of this
may be met by consumption of carrion deposited in
554
the intertidal zone. We have not considered
predation by fish, which we think is low because
the fauna is so concentrated around the driftline.
It may, however, become available to marine
predators when flushed out of the wrack during
storms. Since the assimilation efficiency of
carnivores is high compared to that of herbivores
little of the food consumed by them finds its way
back into the beach in the form of faeces.
This situation is very different from that
depicted for open beaches in the Eastern Cape by
McLachlan et al. (1981b). Here detritus and
phytoplankton comprise over 90% of the food supply
and support a fauna dominated by filter feeders
such as Vonax and Ga4tno~aQQU6 (97,8%) and the
scavenger B~ (1,7%). These in turn are
preyed upon by fish, birds and crabs. Neither
amphipods, isopods nor insects form significant
components of the fauna and very few macrophytes
are deposited on the beaches.
2.7 Interstitial fauna and energy flow
The standing stocks and distribution patterns of
meiofauna and bacteria on Kommetjie beach are
given by Koop and Griffiths (1982) and Stenton-
Dozey and Griffiths (1983). The meiofauna is
dominated by nematodes and oligochaetes and
concentrated in the vicinity of wrack banks at
higher tidal levels. Bacteria show the reverse
distribution, with highest concentrations at low
water, possibly because of grazing pressure in
areas of high meiofaunal density. The mean
annual standing stock of meiofauna is 624 g m- 1
and that of bacteria 961 g m- 1 (since the biomass
of macrofauna is 2256 g m- 1 this gives biomass
ratios of macrofauna:meiofauna:bacteria of
approximately 3,5:1:1,5).
The energy supply to interstitial systems is in
the form of particulate or dissolved organic
matter that percolates into the sand, often in
association with tidal or wave action or rainfall.
'In a high kelp-input beach this material consists
predominantly of decomposed algae or the faeces
and excretory products of its consumers. The
rates of production of these materials are given
in Fig. 1. Although an unknown proportion of
this energy may be swept back into the sea during
high tides (the first waves to inundate wrack
banks often run brown with detritus), we bel i eve
most of it ultimately enters the sand column.
The total amount of organic material entering the
sand is estimated as 3,5 x 10 6 kJ m- 1 y-1, or 85%
of the total input to the beach. The fate of
this material has been investigated by Koop et
al. (1982a & b). who used a microcosm technique
to show that over 90% of the carbon leaching from
decomposing kelp was utilized by bacteria and/or
meiofauna during drainage through a 1 m long sand
column. By comparing rates of utilization of
carbon derived from the kelp material with
simultaneous increases in bacterial biomass in
their microcosm, they calculate that 28% of the
carbon entering the sand is converted into
bacterial carbon. Most of this carbon is
respired by the bacteria and very little of it
percolates back into the sea.
Extrapolating to our field survey we found the
mean annual standing stock of bacteria on the
beach to be 961 g m- 1 dry mass. At 28% conver-
sion efficiency, production of this biomass of
bacteria would require an input of 961 x 10~28 or
3432 g kelp debris, which is equivalent to
48462 kJ. The actual estimated input is
6
3,5 x 10 kJ, so that if this were all utilized
by bacteria sufficient energy would be available
for them to have a production to biomass (p/ s)
ratio of E~:§ or 70 times per year. This should
be taken as a maximum figure, since bacteria may
in fact suffer periods of food shortage inter-
spersed with ones of surfeit, and meiofauna may
also compete with them for the available food
resources. A large biomass of bacteria may,
moreover, be associated with the surfaces of
decomposing kelp itself (Koop et al., 1982a),

whereas we only counted the bacteria in the sand
column. Our bacterial biomass may thus be sub-
stantially underestimated; if the biomass were
larger, the available energy would support fewer
turnovers per year. Our maximum estimate of a
PIB of 70 for bacteria thus correlates fairly well
with the figure of 30 proposed by Koop and
Griffiths (1982) from literature values.
The meiofauna are thought to feed principally upon
bacteria and have been estimated to have an annual
food requirement of 50 times their biomass
(Gerlach, 1978). Given the standing stock of
meiofauna on Kommetjie beach (624 g m- 1) this
gives an annual consumption rate of 31.2 kg
m- 1 y-1 bacteria, which could be supplied by a
bacterial turnover of 319~~0 = 32 times per year.
This agrees well with our estimates of bacterial
turnover and suggests that the interstitial
environment is a relatively closed system in
which bacteria absorb almost all of the available
organic input and bacterial production is almost
entirely consumed by the meiofauna.
3. CONCL US IONS
1) The main source of energy for sandy beaches
in the vicinity of kelp beds along the south
west coast of South Africa is kelp wrack.
Over 2 metric tons (wet mass) or 4 x 106 kJ
of this material may be deposited on each
metre of beach per annum.
2) Since this food resource is concentrated
along the drift line, this is the zone of
maximum macrofaunal species richness and
biomass. The macrofauna consists predomin-
antly of semi-terrestrial amphipods and
insects and has a standing crop of over
2 kg m- 1 (dry mass), of which Tal itrid
amphipods comprise over 90%.
3) Calculated kelp consumption by herbivores on
the beach amounts to 71% of deposition. The
remaining 29% of the wrack is degraded by
555
batteria and either enters the sand column or
is washed back into the sea. Much of the
material eaten by herbivores is also returned
to the beach in the form of faeces or of
excretory products.
4. ~1acrofaunalherbivores are consumed by preda-
tory birds, isopods and Coleoptera. These
are thought to take 40%, 15% (maximum) and 8%
of overall herbivore standing stock per year,
respectively, or a total of 63% of biomass,
but since the production to biomass ratio of
the macrofauna is approximately 2,5 this is
only 25% of macrofaunal production.
5. Dissolved or particulate kelp debris that
enters the sand column is rapidly utilized by
bacteria and by a rich meiofauna that is
dominated by nematodes and oligochaetes and is
concentrated beneath the driftline.
6. Bacteria can convert kelp detritus into
bacterial biomass with up to 28% efficiency
and, given the available energy resources,
may have a PIB ratio of 30 - 70. Meiofauna
may also be able to absorb dissolved organic
material, but the inverse relationships
between meiofaunal density and bacterial
biomass indicates that they feed primarily on
bacteria.
7. Meiofaunal food requirements total approximate-
ly 32 times the bacterial standing stock of
961 g m- 1 , indicating that they consume
virtually all the bacterial production.
8. Most of the carbon in wrack deposited on the
beach is ultimately dissipated as CO 2 through
respiratory losses by bacteria or meiofauna.
Nitrogen is efficiently retained by the
bacteria and may thus ultimately find its way
back to the sea. The amounts of nutrients
returned to the nearshore zone are, however,
insignificant in relation to the nutrient
demands of the macrophytes and phytoplankton.
556
9. Beaches of this type are thus not closed eco-
systems (cf. McLachlan et al., 1981b), but
energy sinks, which rely upon imported debris
to support internal food chains and which
return only limited inorganic nutrients to the
sea.
ACKNOWLEDGEMENTS
Financial support for this project was provided
through the Benguela Ecology Programme of the
South African National Committee for Oceanographic
Research. We are indebted to P.A.R. Hockey for
providing us with, and helping us to locate
ornithological data.
REFERENCES
Backlund HO (1945) Wrack fauna of Sweden and
Finland. Ecology and chorology, Opuscula
Entomologica (Suppl.). 5, 1-236.
Brown AC (1964) Food relations on the intertidal
sandy beaches of the Cape Peninsula, S. Afr. J.
Sci. 60, 35-41.
Brown AC (1973) The ecology of the sandy beaches
of the Cape Peninsula, South Africa. Part 4:
observations,on two intertidal Isopoda, E~ydiee
long~eo~~ (StUder) and Exo~pha~oma tnllneatitel-
~on Barnard,' Trans. Roy. Soc. S. Afr. 40, 381-404.
Cheng L (Ed.) (1976) Marine Insects, Amsterdam,
North-Holland Publishing Company.
Field JG, Griffiths CL, Griffiths RJ, Jarman
N, Zoutendyk P and Bowes A (1980) Variations in
structure and biomass of kelp communities along
the west coast of South Africa, Trans. Roy. Soc.
S. Afr. 44, 145-203.
Gerlach SA (1978) Food-chain relationships in
subtidal silty sand marine sediments and the role
of meiofauna in stimulating bacterial productivit~
Oecologia (Berl.) 33, 55-69.
Griffiths CL and Stenton-Dozey J (1981) The
fauna and rate of degradation of stranded kelp,
Estuar. Coastal Shelf Sci. 12,645-653.
Hockey PAR, Siegfried WR, Crowe AA and Cooper J
(1983) Ecological structure and energy require-
ments of the sandy beach avifauna of southern
Africa. In McLachlan A and Erasmus T, eds
Sandy beaches as ecosystems, pp.507 - 522. The
Hague, Netherlands, W. Junk.
Johnson WS (1976) Population energetics of the
intertidal isopod C~lana ~6o~di, Mar. Biol.
36, 351-357.
Koop K and Field JG (1980) The influence of
food availability on population dynamics of a
supralittoral isopod L~g~a ditatata Brandt,
J. expo mar. Biol. Ecol. 48, 61-72.
Koop K and Griffiths CL (1982) The relative
significance of bacteria, meio- and macrofauna on
an exposed sandy beach, Mar. Biol. 66, 295-300.
Koop K, Newell RC and Lucas MI (1982a) Bio-
degradation and carbon flow based on kelp
(Eeklo~ max~a) debris in a sandy beach micro-
cosm, Mar. Ecol. Prog. Ser. 7,315-326.
Koop K, Newell RC and Lucas MI (1982b) Micro-
bial regeneration of nutrients from the decomposi-
tion of macrophyte debris on the shore, Mar. Ecol.
Prog. Ser. 9, 91-96.
McLachlan A (1977) Composition, distribution,
abundance and biomass of the macrofauna and meio-
fauna of four sandy beaches, Zool. Afr. 12,
279-306.
McLachlan A, Wooldridge T and Dye AH (1981)
The ecology of sandy beaches in southern Africa,
S. Afr. J. Zool. 16, 219-231.
McLachlan A, Erasmus T, Dye AH, Wooldridge T,
van der Horst G, Rossouw G, Lasiak TA and McGwynne
L (1981) Sand beach energetics: an ecosystem
approach towards a high energy interface, Estuar.
Coastal Shelf Sci. 13, 11-25.
McLachlan A, Wooldridge T, Schramm Mand KUhn ~1
(1980) Seasonal abundance, biomass and feeding of
shore birds in the east Cape, South Africa,
Ostrich 51, 44-52.
Muir DG (1977) The biology of Talo~eh~~
eape~~ (Amphipoda : Talitridae) including a
population energy budget, M Sc Thesis, University
of Cape Town.
Munro ALS, ~Jells JBJ and McIntyre AD (1978)
Energy flow in the flora and meiofauna of sandy
beaches, Proc. Roy. Soc. Edinburgh 76B, 297-315.
Newell RC, Field JG and Griffiths CL (1982)
Energy balance and significance of micro-
organisms in a kelp bed community, Mar. Ecol.
Prog. Ser. 8, 103-113.
Shafir A and Field JG (1980) Importance of a
small carnivorous isopod in energy transfer, Mar.
Ecol. Prog. Ser. 3,203-215.
Stenton-Dozey JME and Griffiths CL (1983) The
fauna and rate of degradation of kelp wrack.
In McLachlan A and Erasmus T, eds. Sandy beaches
as ecosystems, pp. 557-568. The Hague,
Netherlands, W. Junk.
Stenton-Dozey JME and Griffiths CL (1980)
Growth, consumption and respiration by larvae of
the kelp-fly Fue~ eape~~ (Diptera :
Anthomyiidae), S. Afr. J. Zool. 15, 280-283.
Summers RW, Cooper J and Pringle JS (1977)
Distribution and numbers of coastal waders
(Charidrii) in the south-western Cape, South
Africa, Summer 1975-76, Ostrich 48, 85-97.
Velimirov B, Field JG, Griffiths CL and
Zoutendyk P (1977) The ecology of kelp bed
communities in the Benguela upwelling system.
Analysis of biomass and spatial distribution,
Helgolander wiss. Meeresunters. 30, 495-518.
 557

THE FAUNA ASSOCIATED WITH KELP STRANDED ON A SANDY BEACH

J.M.E. STENTON-DOZEY and C.L. GRIFFITHS (Zoooogy Department, University of Cape Town, Private Bag,
Rondebosch, 7700)

principally the amphipod TaZorchestia capensis

1. INTRODUCTION
[Muir 1977), the isopod Ligia diZatata [Koop,
Seventy percent of the southern African coastline
Field 1980,1981) and larvae of the kelp fly
consists of sandy beaches (McLachlan et al. 19S1),
FuceZlia capensis [Stenton-Dozey, Griffiths 1980).
but these received little attention until Brown's
A one month survey of the macrofauna associated
publications [1964, 1971a,b) on the general
with stranded kelp and the rate at which this
ecology of beaches around the Cape Peninsula.
material was degraded was also undertaken by
Since then McLachlan [19na-c, 19S0), Dye (1979),
Griffiths and Stenton-Dozey [19S1). Koop and
McLachlan et a1. [1979, 19S1), Dye et a1. [19S1)
Griffiths (1982) studied the relative significance
and Wooldridge et a1. [19S1) have investigated
of the macro-, meio-fauna and bacteria and
the physical parameters and fauna of eastern and
recently the fluxes in material arising from
southern coast beaches and Bally [19S1), those of
decomposing wrack has been investigated (Koop
sandy beaches along the west coast north of the
et a1. 1982a,b).
Cape Peninsula. All these are clean open beaches
which receive only erratic deposits of macro- This paper presents results of a year-long survey
phytes. The on ly form of primary produ ction conducted at Kommetjie to establish the seasonal
arises from offshore blooms of phytoplankton with pattern in composition, distribution, abundance
the occasional stranding of carrion. and biomass of macro-, meio-fauna and bacteria
and their relative contributions to the beach
By contrast the west coast of the Cape Peninsula
economy in terms of standing stock and productivity.
has extensive offshore kelp beds dominated by
Fluctuations in the fauna are correlated with the
Ecklonia maxima and Laminaria pallida. Newell
deposition rate of kelp material.
et a1. [19S2) have reviewed the available
information on primary productivity, standing
2. MATERIALS AND METHODS
stocks and ecological energetics of consumer
2.1 Beach zonation and collection of wrack
organisms inhabiting these beds. Much of the
In zoning the beach, three stages of kelp
kelp material is ultimately uprooted and stranded
degradation were recognised at the time of low
on nearby shores, providing a rich source of
spring tide, namely those in which deposits were
energy for the intertidal fauna. One particular
old, in the process of decay and fresh. The
area along this coastline, Kommetjie [34 0 08'S,
position of these corresponded to HWS, MW and LWS
18 0 19'E), has been a focus of study for the past
respectively and this zonation generally followed
few years. The beach is approximately one kilo-
the pattern suggested by Dahl (1952), but in this
metre in length with sandy pockets bordered by
case it was subjected to the condition of surface
rocky outcrops. Several of the primary consumers
wrack rather than using indicator species of the
of stranded kelp have been investigated,
558

fauna to determine zones. silicon glass cover slips which were oven-dried

To establish the quantity of material cast ashore, at 60 0 C to constant weight. The mean individual
each month five 1m2 quadrants were removed from dry mass was used to convert numbers to biomass.
fresh deposits along the 300m stretch of beach. 2.4 Bacteria
A large area was covered to compensate for patchy Bacterial densities were established quarterly
deposition. The mean wet weight of these samples, from four random sand cores extracted at 30cm
times 300m was then regarded to represent the depth intervals to the water table in each zone.
total quantity of new material present on the day From the cores, which were mixed for a sampling
of sampling. This value was later increased by site, a 10ml subsample was preserved in 10ml of
an estimated rate for the total replacement of 10% formalin in sterile seawater. Bacteria
stranded kelp in order to reach a monthly figure. were firstly removed from sand grains by shaking
the samples in a DAWE Sonicleaner type 6442A,
2.2 Macrofauna
and then stained with acridine orange and counted
Four random 0,2m 2 quadrants were collected monthly
under an epifluorescent microscope (Hobbie et al.
from each zone at the time of low spring tide.
1977; Linley et al. 1981). Bacterial densities
Since many species, notably kelp flies and
were calculated using the formula in Mazure
beetles, are very motile, each quadrant was
(1978) . The proportion of rods to cocci and
initially enclosed with a plastic tank, sprayed
the dimensions of these cells were determined
with insecticide and left for 10 minutes, where-
from scanning electron micrographs. Volume was
after kelp and sand to a depth of 15cm were
converted to wet weight assuming a specific
transferred to plastic bags. Sandpiles from
gravity of 1,1 and wet weight to dry weight using
each zone were pooled and the fauna identified to
a ratio of 0,23 (Luria 1960).
species, counted and oven dried at 60 0 C.
3. RESULTS AND DISCUSSION
2.3 Meiofauna
3.1 Kelp deposition
Once every three months, four random sand cores
In attempting to determine the quantity of kelp
were extracted from each zone using a stainless
cast ashore, an estimate of turnover rate is
steel corer 30cm in length and 10cm2 in cross
required. Koop and Field (1980) and Koop et al.
section to depths of 30cm and 60cm. It was not
(1981a) estimated an eight day cycle of replace-
considered necessary to sample beyond this depth
ment in which period approximately 80% of the
as Koop and Griffiths (1981) found that 97% of
kelp present passed through the grazer and
the meiofauna associated with a nearby wrack bed
microheterotrophic pathways. Although an
at Kommetjie was concentrated in the upper 60cm.
initial rapid loss in kelp mass was also observed
A 200ml subs ample was removed from the pooled
by Griffiths and Stenton-Dozey (1981), total
cores and fixed in 5% formalin. Animals were
degradation was only completed after two weeks.
separated from the sand in a modified Oostenbrink
It is therefore believed that the total replace-
apparatus (Fricke 1979), stained with rose bengal
ment of kelp has a fourteen-day cycle, coinciding
and counted under a dissecting microscope, a
with spring tides and this has been used to
distinction being made between the major taxonomic
estimate the quantity of kelp cast ashore
groups. Counts were increased by 10% to account
annually.
for Bxtraction loss (Fricke 1979). Biomass for
a taxon was determined by placing a hundred Maximal kelp depoffition occurred in winter (Table
representative individuals on each of three 1), a feature common along the Cape west coast
 559

(Muir 1977 ; Koop, Field 1980J when large offshore species is common in wrack beds and responsible
swells uproot whole plants and drive them ashore. for most of the primary consumption of surface
The mean standing stock calculated from Table 1 material (Griffiths et al. 1983J. Other amphipods
-1
was 83,5kg wet mass m ,with a total deposition present in order of abundance were Talorchestia
-1 -1
rate of 2179kg m yr • a value very similar to quadrispinosa, Paramoera capensis and Gitanopsis
that established for a nearby rocky shore, namely pusilla. The isopods, Exosphaeroma truncatitelson
-1 -1
1200-1800kg m yr (Koop, Field 1980). The and Eurydice longicornis were restricted to the
offshore kelp bed at Kommetjie is approximately waters' edge, where they feed on organic matter
700ha in area (Koop et al. 1982aJ. It has a and prey on small living animals such as
standing stock of 23 030 tonnes of Ecklonia Talorchestia (Brown 1973). Their numbers declined
maxima and 17 284 tonnes of Laminaria pallida in winter [Fig. 1) when the swash zone became
(Field et al. 1980), of which 10% (Simons, inhospitable due to heavy wave action.
Jarman 1981) and 15% (N .Jarman, personal commun-
icationJ respectively may be exported annually to
the adjacent 3km shore. This represents 4 890
tonnes of stranded kelp or 1 630kg m- 1 yr- 1 This
value is also similar to that recorded above and
supports our estimate of turnover time.
-1
TABLE 1. The wet mass of kelp (kg m ) sampled
at monthly intervals from the swash
zone along a 300m stretch of beach at
Kommetjie, Cape Peninsula.

Month -1
Mass of kelp (kg m )
January 36
February 80
March 284
April 43
May 78
June 240
July 116
August 38
J M A M J A SON D
September 19
T(monlhs)
October 76
KEY
November 38 II Amphipoda
December 55 WJ Isopod a
k8 Mollusca
3.2 Macrofauna
o
• Diptera
Coleoptera
Thirty-five macrofauna I forms were recorded,
comprising 4 species of amphipods, 2 isopods, 7
molluscs, 4 dipterans and 18 coleopterans. The
talitrid amphipod, Talorchestia capensis was FIGURe. 1. The taxonomic composition of macro-
fauna (percentage abundanceJ
dominant for most of the year and accounted for associated with kelp deposited on
over 90% of macrofauna I numbers (Fig. 1). Kommetjie beach, Cape Peninsula.
This
560

Of the seven mollusc species, two were common, a
brown bivalve, 1mm in diameter (7 Neogaimardia
kowiensis) and a small grazing gastropod
{Eatoniella sp.} both of which were abundant to
a depth of 10cm at all tidal levels.
organisms were the dominant forms during
September, comprising 85% of the total macrofauna
(Fig. 1). Other molluscs present were mainly
juveniles of sublittoral forms such as
Choromytilus meridionalis and Burnupena sp. which
were probably washed ashore with Kelp.
--'Macrofauna
'-'Kelp
forms, Aleochara salsiipotens (StaphylinidaeJ and
.4critus lightfooti (Histeridae). The most
common herbivore was Cercyon maritimus
(Hydrophilidae), while other typical intertidal
These two forms were Pac;"iyphaleria capensis, Bledius sp.,
Omalium sp. and Cafius xantholoma. The
abundance of beetles through the year was erratic
with no seasonal pattern. Like the isopods,
their contribution towards total numbers seldom
exceeded 10% (Fig. 1). Both orders are not
permanent residents of the intertidal habitat,
the isopods entering from the sublittoral
environment, while the beetles migrate down from
the landward sand-dunes.
250

KEY
~ Nematoda
~Oligochaeta
200
o Harpacticoida
";E ";E • Turbellaria
~
~
~
5 ~
~
~
o Others

E 150 E
,.. ;;
-C 4
~
Ol Ol

~
~ ~
~ 3 100
E'
"
'""

j\J \
~

~
"
/
~
::;; 50

.~.

FMAMJJASOND
T (months)

FIGURE 2. Monthly fluctuations in total b~~mass

of the macrofauna (g dry mass m ) in
response to the quantity of Kclp cast
ashore at Kommetjie ueach, Cape
JAN APRIL JULY OCT
Peninsula.
T(months)
The adults of the two most common dipterans
Fucellia capensis and Coelopa africana were FIGURE 3. The taxonomic composition of meio-
always present, but exceptionally high numbers fauna in terms of percentage abundance
present on Kommetjie beach, Cape
were recorded in summer and autumn, coinciding Peninsula for different seasons of the
with the presence of many larvae (Fig. 1). This year.

may reflect a summer/autumn breeding peaK, this The macrofauna as a whole were concentrated
possibility being reinforced by the virtual between mid tide and HWS, while the lower inter-
absence o~ larvae from July to October. tidal remained sparsely colonised throughout the

The Coleoptera were dominated by two carnivorous year (Table 2). Although the highest abundance

(4 173 x 103m-1) and biomass (5 BB5,67g dry mass
-1) were recorded in July at HWS, no seasonal
pattern was evident for either MW or the upper
intertidal, Species diversity generally
declined down the beach, a phenomenon which is the
reverse of the usual pattern (McLachlan 1977c),
A similar observation was made by Koop and
Griffiths (1982) and they attributed this to
increased food availability in the form of
stranded Kelp at higher beach levels and to the
fact that many of the species at HWS are
essentially of terrestrial origin.
Monthly fluctuations in standing stocKs of macro-
faurla corresponded closely to the pattern of Kelp
deposition recorded during the year (Fig. 2).
Overall biomass fluctuations were controlled by
variations in the biomass of amphipods which in
turn are a function of the lifecycle of T.
capensis as described by Muir (1977).
3.3. Meiofauna
Insufficient taxonomic data on marine meiofauna
in southern Africa is available for specific
identification to be made.
Nevertheless some
distinct trends in basic taxonomic composition
are evident. Fig. 3 compares the percentage
abundance of the four major taxonomic groups,
the less frequent meiofauna being grouped under
'others'. The meiofauna were dominated by
nematodes and oligochaetes (their joint contribu-
tion never being less than 90%) followed by
harpacticoid copepods and turbellarians (both
less than 3%). Other forms such as gastrotrichs,
acarines, archiannelids and polychaetes were
found in extremely low numbers. Although nematodes
generally contributed 50% or more of numbers the
relatively large size of individual oligochaetes
(mean 3,4~g dry mass) made this group equal to
or more important in terms of biomass.
Seasonal fluctuations in total numbers and
biomass with respect to vertical and horizontal
distribution are shown in Table 3. In summer and
autumn the meiofauna were concentrated at HWS,
561
especially in the first 30cm below the surface.
Densities declined dramatically towards the sea,
although similar values were obtained for the
mid tide zone and the lower intertidal. During
winter and spring the concentration of organisms
shifted to the middle of the beach where again
densities were greatest 30cm into the sand.
From this area the decline towards HWS was more
gradual than towards the sea, indicating a more
even distribution between MW and HWS than found
in summer and autumn.
On the east and south coast beaches of southern
Africa, a similar distribution of meiofauna
exists as that observed during winter and spring
at Kommetjie (McLachlan 1977c; Dye et al. 1981;
McLachlan et a1. 1981; Wooldridge et a1. 1981).
The shift away from MW to HWS during summer and
autumn was liKely to be in response to the
extremely anoxic conditions which were prevalent
in surface wracK and 40cm below the surface in
the mid tide area at this time.
3.4. Bacteria
12 -1
Bacterial densities ranged from 25.10 m to
12 1 6 6
6 386.10 m- or 14.10 to 734.10 per ml of dry
sand (Table 4) of which approximately 63% were
of coccoid form. This range is slightly greater
than found by Koop and Griffiths (1982) in a
nearby wracK bed and McLachlan et a1. (1979) for
east coast sandy beaches devoid of wracK.
Numbers varied through the year but were
consistently high to the depth of the water
table.
In summer and autumn numbers and biomass were
highest at the mid-tide level with a marKed
decline towards LWS and HWS (Table 4). In
winter bacteria were concentrated at HWS whereas
in spring the greatest biomass existed at MW and
numbers peaKed at HWS. This seasonal
distribution pattern was the reverse of that
observed for meiofauna (see Table 3) possibly
indicating the influence of grazing pressure by
meiofauna. This was especially evident in
562

TABLE 2. Monthly estimates of the numbers and biomass of macrofauna pres8nt on

Kommetjie beach, Cape P8nin~~la, with r8sp8ct to tidal distribution.
Th8 data can be 8xpr8ssed m by dividing the values by the width of the
resp8ctive zon8S. (N LWS c zon8 of fresh Kelp deposition at low water
springs] o MW c zone of decaying kelp at mid-tide] o HWS c zon8 of old
Kelp deposits at high water springs).

Zon8s and their r8S- Biomass

Month pective widths in Numb~r -1 g dry mass Total numb8r
metr8s x 10m m- 1 of sp8ci8s

January N LWS 15 52,BB 35,15 9

0 MW 10 9,00 8,57 7
0 HWS 15 100,88 146,06 19

F8bruary N LWS 17 37,12 23,55 9

0 MW 15 403,75 1171,39 17
0 HWS 6 169,95 664,96 16

March N LWS 20 10,83 7,90 6

0 MW 13 809,87 582,92 14
0 HWS 27 273.3B 193,B7 13

April N LWS 20 147,83 68,25 7

0 MW 13 271,27 228,37 15
0 HWS 10 72,50 170,30 19
May N LWS 20 74,67 49,07 8
0 MW 29 276,48 337,39 9
0 HWS 14 414,40 1956.12 11

Jun8 N LWS 14 63,12 20,58 6

0 MW 14 3424,40 4588,44 6
0 HWS 16 1203,60 1592,36 B

July N LWS 21 0,00 0,00 0

0 MW 14 615,35 875,30 10
0 HWS 6 4173,83 5885,67 11

August N LWS 16 411,87 137,76 3

0 MW 14 46,67 26,15 7
0 HWS 17 1513.85 1342,52 4

S8pt8mb8r N LWS 38 96,00 26,83 2

0 MW 10 1, DB 2,52 5
0 HWS 7 18,64 36,56 5

October N LWS 17 327,82 371,53 6

0 MW 9 583,80 2419,49 11
0 HWS 8 1241,60 4532,54 10
 563

TAClLE 2. continued:

Zones and their res- Biomass

Month pective widths in Number -1 g dry mass Total number
metres x 10 3 m m- 1 of species

November N LWS 16 66,27 49,06 12

D MW 12 123,10 312,93 17

° HWS 13 37,36 76,66 14

December N LWS 17 56,95 19,17 9

D MW 13 111,69 140,30 15

° HWS 14 9,22 16,66 17

autumn, when bacterial biomass reached its peak
at 2 644g m- 1 when meiofauna was at its minimum
(see Tables 3 and 4). It has been reported
elsewhere that meiofauna may control bacterial
densities (Jansson 1966: Giere 1975: Gerlach
1976) .
3.5. Standing stocks
The greatest range of monthly standing stocks
occurred in the macrofauna (66 - 7 324g dry mass
m- 1 ) as compared to 431 - 759g m- 1 and 115 -
2 644g m- 1 for meiofauna and bacteria
respectively. As most of the macrofauna are
restricted to surface habitats, they are more
susceptible to environmental fluctuations and
hence a wide range in biomass can be expected
OVer the year. By contrast, the interstitial
fauna occupies a more stable environment,
relatiVely free from external influences.
Comparison of the mean standing stock of macro-
fauna (2 257g m- 1 ), meiofauna (624g m- 1 ) and
bacteria (961g m- 1 ) shows a biomass ratio of
3,6:1:1,5. Thus whereas the biomass of meio-
fauna and bacteria per metre of beach are
similar, together they make up some 42% of the
total biomass, with macrofauna accounting
for the remaining 58%. Bacterial standing
stocks are, however, minimal values since
estimates were based on bacteria associated
with the surface of sand grains only. The
population colonising the surfaces of stranded
kelp plants was not included although Koop et al.
(1982a) found that 99% of the increase in
bacterial biomass in a microcosm experiment over
an eight day period was associated with the
surface of the kelp itself. Although the
equivalent figure in an open sandy beach would be
far lower (the kelp in the microcosm was placed
on rock draining into a shallow basin of sand),
it does emphasise that kelp surfaces may support
a significant bacterial biomass.
3.6. Productivity
As standing stocks only provide information on
the quantity of the different biotic components
present on the beach at anyone time, it is more
meaningful to establish the significance of each
fauna in terms of productivity. McIntyre (1969)
suggests a turnover rate of 2 for macrofaunal
species living one year and longer, and 5 for
shorter-lived forms. McLachlan (1977b,c) used
values between 2,5 and 3,5 for species from
South African east coast beaches, while Koop and
Griffiths (1982) applied a mean Pis ratio of 2,5
to the macrofauna of a wrack bed. This latter
value was used in this study to convert the mean
standing stock of the macrofauna to an estimate
of productivity.
The life histories of different species of meio-
fauna are very diversified but since most
authors have accepted a PI B ratio of 10
(McIntyre 1964; Gerlach 197B; McLachlan 1977b,
c; Koop, Griffiths 1981), this value has been
followed here.
564

TABLE 3. The seasonal abundance (numbers x 10 6 m- 1 ) and biomass (g dry mass m- 1 ) of meiofauna
with respect to different tide levels and vertical gradients on Kommetjie beach,
Cape Peninsula (see Table 2 for the abbreviations and widths of the different zones).

Sampling depth (cm)

0 30 30 60
Season Total Total
Zone Numbers Biomass Numbers Biomass
Sampled Number Biomass

Summer N LWS 9,54 13,32 3,96 4,10 13,50 17,42

0 MW 6,57 9,00 8,54 10,45 15,11 19,45
0 HWS 243,83 437,96 104,36 138,47 348,19 576,43

376,79 613,29

Autumn N LWS 7,71 9,35 16,38 22,05 24,09 31,40

0 MW 15,39 18,43 4,02 4,62 19,40 23,05
0 HWS 109,25 220,32 95,36 156,15 204,60 376,47

248,09 430,92

Winter N LWS 12,00 18,30 3,18 5,79 15,81 24,10

o MW 163,23 292,36 206,98 319,33 370,21 611 ,69
o HWS 44,68 67,02 27,07 56,64 71,75 123,67

457,77 759,45

Spring N LWS 18,36 31, 34 15,22 31,29 33,58 62,63

o MW 192,25 449,52 35,03 57,34 227,29 506,86
o HWS 46,54 101,30 14,27 20,03 60,80 121 ,33

321, 68 690,82

As pointed out by Gerlach (1978) it is not possible the Pia ratio of 30 was applied.
to estimate bacterial productivity from laboratory
From the biomass and Pia ratios for each component
cultures as conditions are optimum, resulting in
of the biota annual production estimates have been
astronomical figures. In the field turnover rates
calculated and are presented in Tab Ie 5. In
are influenced Dy many parameters (Dale 1974),
contrast with the biomass figures, productivity
while interactions with larger fauna can stimulate
estimates show that bacteria are of paramount im-
productivity [Fenchel 1970, 1972; Lopez et al.
portance, accounting for some 71% of productivity
1~77; Rooertson, Mann 1980). At Kommetjie, a
of the beach, while macrofauna and meiofauna
turnover of 70 times per year has been estimated
contribute equally to the remaining productivity.
on the basis of the conversion efficiency of
The importance of bacteria would have been greater
available organics [Griffiths et al. 1983).
if samples penetrated beyond the depth of the
However, as already mentioned, bacterial standing
water table and if Kelp surfaces were analysed.
stOCKS at this locality have probably been under-
3.7. Trophic interrelationships of the fauna
estimated and therefo,e this value should be reduced
At Kommetjie, the major part of the macrofaunal
probably to the range of 30 to 40, which is close
biomass is composed of primary consumers, which
to that suggested by Gerlach [1978). .To arrive at
are responsib~e for 72% of the surface kelp lost
a conservative estimate of bacterial productivity,
 565

12 -1 -1
TABLE 4. The seasonal abundance (numbers x 10 m ) and biomass (g dry mass m ) of bacteria
at different tide levels and sampling depths to the water table on Kommetjie beach.
Cape Peninsula (see Table 2 for abbreviations and widths of the respective zones).

Sampling depths (cm)

0 30 30 60 60 90
Total Total
Season
Zone Numbers Biomass Numbers Biomass Numbers Biomass Number Biomass
Sampled
Summer N LWS 190,99 24,22 190,99 24,22
0 MW 217,42 14,09 726,01 47,12 239,52 19,08 1236,95 80,29
0 HWS 25,54 0,86 215,82 7,38 41,25 1,40 282,61 9,65

1710,55 114,16

Autumn N LWS 2010,04 540,36 2010,04 540,36

0 MW 6385,62 968,66 3328,56 504,95 9714,18 1 463 ,61
0 HWS 1580,12 220,42 1779,96 248,26 2030,21 171,20 5390,29 639,88

17114,51 2643,85

Winter N LWS 760,62 52,27 760,62 52,27

0 MW 635,57 55,31 595,83 51 ,82 1231,40 107,13
0 HWS 1070,25 64,05 1117,19 66,91 750,33 44,93 2937,77 175,89

4929,79 335,29

Spring N LWS 1972,34 96,53 1972,34 96,53

0 MW 1080,49 196,24 1078,51 195,88 2159,00 392,11
0 HWS 1364,03 133,77 446,03 43,72 848,06 83,15 2658,12 260,64

6789,46 749,28

via grazing and fragmentation (Griffiths, Stenton- eight days after stranding. However, this study
uozey 1981; Griffiths et al. 1983). The took place in the absence of natural densities of
surface biota thus plays an important role in macrofauna and meiofauna. The role of bacteria
fragmenting whole kelp plants, thereby providing in the initial decomposition of kelp lying on an
leachates and small organic particles to the open beach is probably not as pronounced, their
interstitial fauna. As the dominant macro- impact being shifted to the final degradation of
faunal organisms, the amphipod T. capensis and kelp fragments.
larvae of the kelp fly F. capensis are
Beoides the role of decomposing kelp, bacteria
characterised by low assimilation efficiencies
probably constitute the major food source for the
(Muir 1977; Stenton-Dozey, Griffiths 1980),
meiofauna, as detrital residues on their own
faeces production would greatly increase this
provide a poor source of nitrogen and other
input.
essential minerals [McIntyre 1969; Fenchel 1970).
In contrast to the findings cited above, where There is evidence that this food source is
macrofauna were responsible for most of the supplemented by direct absorption of dissolved
initial loss of kelp mass, Koop et al. (1982a) organic matter (eg. Chia, Warwick 1969; Meyer-
concluded from their microcosm study that Reil, Faubel 1980). In addition certain groups
bacteria degraded 90% of this material within such as nematodes, turbellarians and acarines are
566

TABLE 5. The productivity of the components of the sandy beach biota at Kommetjie,
Cape Peninsula, estimated from standing stocks and annual turnover rates.

Mean annual Annual Annual Percentag8

standing stoc~1 turnover production -1 of total
(g dry mass m ) (P/ B) (g dry mass m ) production

Macrofauna 2256,97 2,5 5642.4 14 ,3

Meiofauna 623,62 10,0 6236,2 15,2
Bacteria 960,85 30,0 28619,5 70,4

pr8datory on other meiofauna (Giere 19751 Cox REFERENCES

1976) .
Griffiths et al. (1983) have quantified the energy
flow through the Kommetjie wrack ecosystem.
a qualitative basis, it appears that two isolated
trophic levels are in existence, a feature not8d
in most intertidal sandy beach8s (eg. McIntyre
1969. Gerlach 19711 Mclntyr8, Murison 1973).
The macrofauna occupy a two dimensional environ-
ment, their stability being influenced by trophic
interactions from within and by kelp supply and
predation by birds from the outside. The infauna
are relatively free from external factors and
their only connection with the surface
environment is related to the amount of nutrients
arising from the wrack beds. The major impact
of this isolated interstitial ecosystem on the
economy of the beach is its consumption of organic
carbon and production of waste products such as
faeces and CO 2 and the remineralization of
ni trogen and phosphoru s. The beach appears
therefore to act as an 'energy sink' which is
fuelled by vast quantities of imported kelp
and from which little is returned to the sea.
4. ACKNOWLEDGEMENTS
The authors thank Dr Andre Prins of the South
African Museum for identifying insects and
Or Richard Kilburn of the Natal Museum for
identifying molluscs. Funding was provided by
the Benguela Ecology Programme through SANCOR
(South African National Committee for
Oceanographic Research), for which we are
ext~emely grateful.
Bally R [1981) The ecology of three sandy
beaches on the west coast of South Africa, Ph.D.
thesis, University of Cape Town, Rondebosch 7700,
On South Africa, 404 pp.
Brown AC (1964) Food relationships on the
intertidal sandy beaches of the Cape Peninsula,
S. Afr. J. Sci. 60, 35-41.
Brown AC [1971a) The ecology of the sandy
beaches of the Cape Peninsula, South Africa,
Part I: Introduction, Trans. R.Soc. S. Afr. 39,
247-279.
Brown AC (1971b) The ecology of the sandy
beaches of the Cape Peninsula, South Africa,
Part II: The mode of life of Bullia [Gastropoda
Prosobranchiata), Trans. R. Soc. S. Afr. 39,
281-319.
Brown AC [1973) The ecology of sandy beaches
of the Cape Peninsula, South Africa, Part IV:
Observations on two intertidal Isopoda :
Eurydice longicornis (Studer) and Exosphaeroma
truncatitelson (Barnard), Trans. R. Soc. S. Afr.
40, 381-404.
Chia F-S and Warwick RM (1969) Assimilation of
labelled glucose from seawater by marine
nematodes, Nature, Lond. 224, 720-721.
Cox JL (1976) Sampling variation in sandy
beach littoral and nearshore meiofauna and
macrofauna, Ft. Belvoir Va: U.S. Coastal Engineer-
ing Research Centre. Technical paper No. 76-14,
72pp.
Dale NG (1974) Bacteria in intertidal
sediments: factors related to their distribution,
Limnol. Oceanogr. 19, 509-518.
Dahl E (1952) Some aspects of the ecology and
zonation of the fauna on sandy beaches, Oikos 4,
1-27. and
Dye AH, McLachlan A,/Wooldridge T (1981) The
ecology of sandy beaches in Natal, S. Afr. J.
Zool. 16, 200-209.
Fenchel T (1970) Studies on the decomposition
of organic detritus from the Turtle grass
Thalassia testudinum, Limnol. Oceanogr. 15, 14-20.
Fenchel T (1972) Aspects of decomposer food
chains in marine benthos, Verh. Ot. Zool. Ges. 65,
14 c 23.

Field JG, Griffiths CL, Griffiths RJ, Jarman N,
Zoutendyk P, Velimirov B and Bowes A (1980)
Variations in structure and biomass of kelp
communities along the south-west Cape coast, Trans.
R. Soc. S. Afr. 44. 145-203.
Fricke AH (1979) Meiofauna extraction
efficiency by a modified Oostenbrink apparatus,
Helgolander wiss. Meeresunters 32, 436-443.
Gerlach SA (1971) On the importance of marine
meiofauna for benthos communities, Oecologia
[Berl) 6, 176-190.
Gerlach SA (1978) Food chain relationships in
subtidal silty sand marine sediments and the role
of meiofauna in stimulating bacterial productivity,
Oecologia (Berl) 33, 55-69.
Giere 0 (1975) Population structure, food
relations and ecological role of marine
oligochaetes, with special reference to meio-
benthic species, Mar. BioI. 31, 139-156.
Griffiths CL and Stenton-Dozey J (1981) The
fauna and rate of degradation of stranded kelp.
Est. coastl. Shelf Sci. 12, 645 -653.
Griffiths CL, Stenton-Dozey J and Koop K
(1983) Kelp wrack and the flow of energy through
a sandy beach ecosystem. in McLachlan A and
Erasmus T, eds. Sandy beaches as Ecosystems,
pp. Junk Publications, The Hague,
Netherlands.
Hobbie, JE, Daley PT and Jasper S (1977) Use
of Nucleopore filters for counting bacteria by
fluorescence microscopy, Appl. env. Microbiol.
33, 1225-1228.
Jansson B-O (1968) Quantitative and experimental
studies of the interstitial fauna in four Swedish
sandy beaches, Ophelia 5, 1-71.
Koop, K and Field JG (1980) The influence of
food availability on the population dynamics
of a supralittoral isopod, Ligia dilatata [Brandt),
J. expo mar. BioI. Ecol. 48, 61-72.
Koop K and Field JG (1981) Energy transformation
by the supra littoral isopod Ligia dilatata
[Brandt), J. expo mar. BioI. Ecol. 53, 221-233.
Koop K and Griffiths CL (1982) The relative
significance of macro-, meio- and microfauna
on an exposed sandy beach, Mar. BioI. 66,
295-300.
Koop K, Newell RC and Lucas MI [1982a) Bio-
degradation and carbon flow based on kelp
{Ecklonia maxima} debris in a sandy beach
microcosm, Mar. Ecol. Prog. Ser 7, 315-326.
Koop K, Newell RC and Lucas Ml [1982b) Microb-
ial regeneration of nutrients from the
decomposition of macrophyte debris on the shore.
Ma~ Ecol. Prog. Ser. 9, 91-96.
Linley EAS, Newell RC and Bosma SA (1981)
Heterotrophic utilization of mucilage released
during fragmentation of 1<I31 p (Ecklonia maxima
and Laminaria pallida) 1. Development of
microbial communities associated with the
degradation of kelp mucilage, Mar. Ecol. Prog.
Ser. 4, 31-41.
Lopez GR, Levinton JS and Slobodkin LB (1977)
The effect of grazing by the detritivore
Orchestia grillus on Spartina litter and its
5{)}
associated microbial community, Oecologia [Berl)
30, 111-127.
Luria SE (1960) The bacterial protoplasm :
composition and organisation. In Gunsalus IC and
Stanier FY, eds. The bacteria Vol. 1, pp. 1-34.
New York, Academic Press.
Mazure HGF (1978) The seasonal cycle of marine
bacteria in a west coast kelp bed, Trans. R. Soc.
S. Afr. 43, 119-124.
McIntyre AD (1964) Meiobenthos of sublittoral
muds, J. mar. BioI. Ass. U.K. 44, 665-674.
McIntyre AD (1969) Ecology of marine meio-
benthos, BioI. Rev. 44, 245-290.
McIntyre AD and Murison DJ(1973) The
meiofauna of a flatfish nursery ground, J. mar.
BioI. Ass. U.K. 53, 93-118.
McLachlan A [1977a) Studies on the psammo-
littor~l meiofauna of Algoa Bay, South Africa.
I. Physical and chemical evaluation of the
beaches, Zool. Afr. 12, 15-32.
McLachlan A [1977b) Studies on the psammo-
littoral meiofauna of Algoa Bay, South Africa.
II. The distribution, composition and biomass of
the meiofauna and macrofauna, Zool. Afr. 12,
33-60.
McLachlan A [1977c) Composition, distribution,
abundance and biomass of the macrofauna and
meiofauna of four sandy beaches, Zool. Afr. 12,
279-306.
McLachlan A (1980) Intertidal zonation of
macrofauna and stratification of meiofauna on
high energy sandy beaches in the eastern Cape,
South Africa, Tr9ns. R Soc. S. Afr, 44, 213-223.
McLachlan A, eye AH and van der Ryst P (1979)
~rtical gradients in the fauna and oxidation of
two exposed sandy beaches, S. Afr. J. Zool. 14,
43-47.
McLachlan A, Wooldridge T and Dye AH (1981) The
ecology of sandy beaches in South Africa, S. Afr. J.
Zool. 16, 219-231.
Meyer-Reil L-A and Faubel A (1980) Uptake of
organic matter by meiofauna organisms and inter~
relationships with bacteria, Mar. Ecol. Prog. Ser.
3,251-256.
Muir DG (1977) The biology of Talorchestia
capensis [Amphipoda; TalitridaeJ, including a
population energy budget, M.Sc. Thesis, University
of Cape Town, Rondebosch, 7700, South Africa, 94 pp.
Newell RC, Field JG and Griffiths CL (1982)
Review: Energy balance and significance of micro-
organisms in a kelp bed community, Mar. Ecol. Prog.
Ser. 8, 103-113.
Robertson AI and Mann KH (1980) The role of
isopods and amphipods in the initial fragmentation
of Eelgrass detritus in Nova Scotia, Canada, Mar.
BioI. 59, 63-69.
Simons RH and Jarman NG (1981). Subsommercial
harvesting of a kelp on a South African shore. In
Levring T, ed. 10th International Seaweed
Symposium. pp. 731-736. Walter de Gruyter & Co.,
Berlin, New York.
Stenton-Dozey J and Griffiths CL (1980) Growth,
consumption and respiration by larvae of the kelp
fly FUcellia capensis [Diptera: AnthomyiidaeJ,
568

S. Afr. J. Zool. 15, 280-283.

Wooldridge T, Dye AH and McLachlan A (1981)
The ecology of sandy beaches in the Transkei,
S. Afr. J. Zool. 16, 210-218.
 569

SANDY BEACH ECOLOGY WORKSHOP REPORT

A. McLACHLAN and G.C. BATE (Zoology and Botany Departments, University of Port
Port Elizabeth, P.O. Box 1600, Port Elizabeth 6000, South Africa)

Workshop discussions were divided into functional with respect to the

two main sessions - particular system being studied.
1. General beach ecology
2. Surf diatom and algal ecology 2. What are the main biological com-
ponents of beaches and what are
GENERAL BEACH ECOLOGY the gaps in our knowledge of
Discussions centred around six ques- these?
tions: The general consensus was that, where
possible, biological work should be
1. What is a beach? concentrated at the level of indivi-
The earlier review had emphasised the dual species. Components that had
importance of looking at both the beach been neglected were zooplankton and
and surf zone together in order to phytoplankton in the surf, meio- and
understand the function of the whole microfauna in the sediment and
system. The beach had been defined as driftline vegetation at the top of
an area from the drift line to the outer the beach. The main biological gaps
limit of surf circulation cells. During in our knowledge of sandy beaches
discussion there was a feeling, however, were due to either taxonomic diffi-
that a rigid beach definition was not culties (e.g. micro- and meiofauna)
satisfactory since beaches represented a or practical sampling problems (e.g.
continium along a gradient of exposure zoo- and phytoplankton). The con-
and sand particle size. Also, each clusion, therefore, was that interest
beach had unique features which would should focus at the individual
represent differing degrees of species level in the different cata-
importance to different research gories and that every effort should
workers. The conclusion was that no be made to overcome problems of
absolute definition of a beach existed taxonomy and quantitative sampling.
and that workers should define their In particular our knowledge of sandy
particular beach system, including its beach flora is behind our knowledge
physical boundaries, substratum, of the fauna.
geographical factors, food inputs,
energy flow and degree of exposure. 3. What major biological processes
Therefore the definition should be need study?
570

The question was posed whether energy superparameter controlling all beach
flow was the only biological processes. There was general agreement
process warranting study. It was with this but sedimentary parameters
felt that biological interactions were also considered very important.
such as predation and competition Besides the small scale effects of
needed more study. Species-species sediments on meiofauna, larger scale
interactions had been thought to be effects of morphodynamics of the bed on
unimportant on high energy beaches. surf circulation patterns were con-
However, predation, may be a major sidered important. Morphodynamics of
reason for high mobility and the the bed. and water circulation should
migration of sandy beach organisms. therefore be studied in conjuction.
Consequently, the conclusion was that Furthermore, both short and long term
while energy flow and nutr ient cycl ing studies were required, including
were prime pro.cesses in beach environ- periods greater than ten years.
ments, biological interactions affecting Since episodic events were ·of ;Jreat
community structure rather than energy importance on beaches, the conclusion
flow should also be investigated, was that studies of water flow in
especially in relation to exposure beaches on all time and spatial
gradients. scales were needed and that these
studies need to be coupled to studies
4. What chemical processes are import- on the bed and sediment dynamics.
ant in ecological studies?
The question was posed whether nutrient 6. Is a systems approach valid in
cycling was the only important process. beach studies?
Other factors affecting nutrient inputs The suggestion was made that beaches
were considered important, including need not be seen as interfaces and
those of rain and groundwater inputs and individual components studied in
it was felt that currencies other than N isolation, but that whole beach/surf
and C should be examined, especially P. zone environments should be studied
Studies should progress beyond a simple as ecosystems. This approach
monitoring of chemical concentrations to attempted to understand the structure
a study of processes and environmental and function of such environments.
sensitivity. Nutrient cycling was seen In this context a top-down approach
as the major chemical process needing was considered essential starting
study in sandy beaches, but specific with major questions or nypotheses
attention should be paid to rates of and working down into the system.
change and transfer rather than only to Studies of processes were needed,
levels of concentration. particularly in the surf zone. It
is, however, extremely difficult to
5. What physical processes are import- quantify many processes operating in
ant in ecological studies? these dynamic environments. Never-
The review had stressed the importance theless, in such systems approach
of water flow., on all scales, as the studies at least the major processes
need to be identified.
 571

There was' some scepticism regarding currents. In addition, there were dis-
the systems approach since it was cussions on experimental problems
considered too broad. Others were relating to algal ecophysiology and
not even sure that beaches/surf zones pr imary production estimates. Mostly,
could be considered ecosystems, since questions were raised and few answers
where winds blow predominantly off- were provided. For example:
shore, these environments may be very Is vertical .movement of the cells
open. Exchange acr.oss .the seaward related to photosynthesis?
boundary was thus very important. Is the rising active and physiologi-
Nevertheless the general conclusion was calor simply the passive result of
that a systems approach was considered a cells attaching themselves to
logical beginning to understanding the bubbles?
structure and. function of beach systems. How might cells physically attach
This may not provide the type of data on themselves to bubbles?
which predictions could. be based, but Are exudates important?
would improve the intuitive .understand- Is a 'bloom' a zone of production or
ing of the systems. a zone of accumulation?
Why do only some surf circulation
BENTHIC SAMPLING cells support blooms?
In addition to the six questions posed What is the residence time of diatom
above. a small group discussed benthic cells in the surf?
sampling. problems on sandy beaches. A Is the bloom a new population each
variety of techniques were described day?
including dredges, suction samplers, Where do 'seed' phytoplankton for the
corers and quadrat frames. All existing blooms come from?
sampling methods were considered accept- How can biomass and producti vi ty be
able provided details were given. Samp- quantified?
ling depth was important since sublit- What is the rate of water .exchange
toral macrofauna has at times been found between rip currents and the area
to a depth of 60 cm in the sediment. outside the breakers?
The depth distribution of the fauna Are nutrients present in excess or
should therefore be ascertained prior to are other factors the cause of the
sampling. Agreement was reached that blooms?
all results should be expressed per
linear meter of beach rather than per Practical problems were discussed,
metre squared. The width of beach including those associated with working
should, however, also be given. in very dynamic (high energy) environ-
ments. These included maintaining
SURF ALGAL ECOLOGY equipment in the surf dur ing measure-
Discussions on surf algae mainly covered ments of primary production, the
problems relating to the r'che of cell necessity to dilute dense diatom
buoyancy and water circulation in the concentrations, incubation times and
maintenance of bloom position over rip estimates of biomass. The use of
572

plastic bags for incubations rather than

rigid containers, as well as the possi-
bilityofincubating samples for primary
production on the beach r.ather than in
the surf were. also discuss.ed.• It was
agreed . that great accuracy was not
possible under surf conditions.

While no unifying. answers were obtained,

there was general agreement on the need
for more detailed investigations of surf
zone algae. There also appeared to be a
need for conceptual and simulation
models of plant development and accumu-
lation in surf zones.

CONCLUSION
The study of ecology of sand beach
.environments is still in its infancy
and basic components and procedures,
well understood or satisfactorily
developed in other environments, are
barely known in ocean beaches. There is
vast scope for research, in particular
on the dynamic processes occurring
within surf zones.
 573

PART FOUR

ECOPHYSIOLOGY AND AUTECOLOGY


THE ECOPHYSIOLOGY OF SANDY BEACH ANIMALS - A PARTIAL REVIEW
A.C. BROWN (Department of Zoology, University of Cape Town, South Africa)
1• I NTRODUCTI ON
There is clearly no sharp distinction between
ecology and ecophysiology, and definitions of
the latter will vary with personal taste and
training. For the purposes of the present review
I take ecophysiology to relate primarily to exper-
imental studies on individual sany-beach species,
the results of which help to elucidate ecological
findings or to predict ecological changes. How-
ever, although laboratory and field experiment-
ation forms an essential aspect of ecophysiology,
it is convenient and appropriate to include under
this heading the entire study of behaviour, quite
regardless of whether the behaviour has been
induced experimentally or simply observed in the
field. This may distort our definition somewhat
but that is preferable to an artificial division
into experimental and non-experimental aspects.
Partly because of limitations on space, the
present review is concerned largely with those
aspects of ecophysiology which are peculiar to
or at least particularly marked in sandy-beach
animals. This means laying special stress on
forms from high energy sandy beaches, for it is
here that the problems of instability of the
substratum and shortage of food are most apparent.
By contrast, the problems encountered on low
energy sandy beaches differ from those of muddy
shores mainly in degree, so that it is not very
surprising to find both types of shore often
inhabited by the same species.
I have been unable to trace any previous review
of the ecophysiology of sandy-beach organisms,
575
although there are many publications in which
certain aspects are reviewed, often in the wider
context of soft substrata in general or of marine
biology as a whole. For example, Trueman (1975)
has reviewed the locomotion of soft-bodied
animals, Cruetzberg (1975) the orientation of
marine animals and Kinne(1970, 1971) the effects
of temperature and salinity. A valuable source
for the present review has been the Biotogy 06
i~~dat anim~ by RC Newell (1979), a book
in which sandy-beach forms are accorded a gener-
ous amount of space. Eltringham's Li6e in mud
and ~and (Eltringham, 1971) is still worth
consulting, while the texts by MacGinitie,
MacGinitie (1968) and Vernberg, Vern berg (1972)
contain considerable relevant information. There
are also pub 1 i ca ti ons such as Li6e in Mndy ~hOft~
by AE Brafield (1978) which, while not detailed,
extensively referenced reviews, nevertheless
expound some of the main principles involved
with admirable clarity and insight. There have,
in addition, been two very recent reviews dealing
with particular sandy-beach genera - one on the
whelk B~ (Brown, 1982a), the other on the
bivalve Vonax (Ansell, this volume).
The ecophysiology of sandy-beach organisms has
not, perhaps, aroused as much interest as that
of rocky-shore forms, possibly because fewer
species are involved. The study is, however,
just as old, dating back to the early 1930s,
with a few still earlier papers worthy of note.
Many of these early papers dealt with burrowing
or with larval settlement, or simply recorded
576
observed behaviour. One of the first principles
to emerge was the wide-spread occurrence of
tidally linked activity rhythms.
2. MIGRATORY TIDAL RHYTHMS
Interest in tidal rhythms of activity was first
stimulated by the work of Gamble, Keeble (1904)
on an innate rhythm of vertical migration in the
turbellarian Convoluta ~o~co66enh~, a rhythm
which persisted for a week in the laboratory in
undisturbed sand. Since then, so many sandy-beach
animals have been shown to undertake tidal migrat-
ions, commonly leaving the shelter of the sand to
migrate up and down the beach or to enter the water
column, that the phenomenon may be considered
characteristic of this type of ecosystem. These
migrations appear always to be associated with
max imi sing food resources, although other benefits,
such as avoidance of predators, may also accrue
in some species.
Among Crustacea, such tidal excursions are display-
ed by the isopods E~yd{ce(Jones, Naylor, 1970;
Hastings, 1981), Ex~olana(Enright, 1972; Dexter,
1977), P~eudaga (Fincham, 1973) and Tylo~(Kensley,
1974), the amphipods T~~(Papi, 1955; Feather-
ston, MacIntyre, 1957; Bregazzi, Naylor, 1972),
O~ch~:U.a (RUppell, 1967; Wildish, 1970), Talon-
ch~:U.a (Muir, 1977), O~ch~~oidea(Craig, 1973),
M~nogamm~~ (Fincham, 1971), B~hypo~eia
(Fincham, 1970) and Co~ophium(Morgan, 1965), the
mysid G~~o~acc~ (Brown, Talbot, 1972) and
decapods such as Em~ (MacGinitie, 1938; Efford,
1965), C~cin~ (Naylor, 1958; Arudpragasam,
Naylor, 1964; Atkinson, Parsons, 1973), Ocypode
(Cott, 1929; Magnus, 1960) and VotiUa (Magnus,
1960; Fishelson, this volume). Tidally migrant
Mollusca include various species of Bullia (Brown,
1961,1971; Ansell, Trevallion, 1969), T~ebM
(Kornicker, 1961) and Vonax (Mori, 1938, Ansell,
Trevallion, 1969).
Tidal rhythms of activity are, of course, by no
means limited to high energy sands and are
displayed by animals inhabiting all types of
intertidal shore (Newell, 1979); however, it
is on sandy beaches exposed to heavy wave action
that such rhythms become immediately obvious and
often spectacular. In many of these species, an
endogenous rhythm has been demonstrated or assum-
ed, triggered by factors.related to the ebb and
flow of the tide. In most of the Crustacea that
have been studied in this regard, activity rhythms
continue to manifest themselves in the laboratory
in the complete absence of tidal clues and this
has not unreasonably been taken as evidence of
an internal clock. Molluscs, on the other hand,
have not been found to continue their rhythmic
activity in the laboratory. Nevertheless, Mori
(1950) suggested that the emergence of the
bivalve Vonax on the rising tide might be due
to an intrinsic rhythm, although Jacobsen (1955)
and Wade (1967) pointed out that emergence could
simply be due to the animals being washed out of
the sand by the incoming waves. This theory
does not seem to be true in general, however,
as the emergence of Vonax is usually an active
process, the foot being pressed down strongly
into the sand, as is the case also in the whelk
Bullia (Brown, 1971, 1982a). In fact emergence
appears to be response to increasing liquefaction
of the sand as the tide rises (Trueman, 1971), a
factor which may trigger the emergence of other
psammophiles as well (Cubit, 1969). The emergence
of Vonax from their upper positions on the slope
is triggered by increasing periods of non-satur-
ation between waves (Turner, Belding, 1957; Wade,
1967; Trueman, 1971) and similarly does not of
necessity imply an endogenous rhythm. Indeed,
it now seems clear that such innate rhythms are
not found in the genus Vonax and that the
migratory responses of the animals are dictated
entirely by changing conditions as the tide moves
up and down the slope (Ansell, this volume).
Similar conclusions have been reached in the case
of the whelk Bullia (Brown, 1961; McLachlan et
al., 1979).

The transport of molluscs such as Vonax, ButtLa
and T~ebna up and down the beach (surfing) is
greatly facilitated by extension of the foot,
which presents a broad surface to the waves, acting
as an underwater sail (Ansell, Trueman, 1973;
Kornicker, 1961; Brown, 1961, 1971). Such foot
extension also ensures .that the animal is ready to
burrow at the end of an excursion; this is of some
importance if the animal is to be able to dig its-
elf in before being washed away again by the waves.
More complex than the behaviour of sandy-beach
molluscs is that of the cirolanid isopods exempli-
fied by E~ydiee, a genus which has been studied
extensively since the pioneering work of Watkin
(1942). Fish (1970) showed that the distribution
of E~ydiee pulehna varies according to a semi-
lunar tidal rhythm, while Jones, Naylor (1970)
found an endogenous rhythm of swimming activity
to be present. This swimming rhythm occurs in
the laboratory both with and without a sandy
substratum. Fish, Fish (1972) showed that if
sand was provided the animals swam in the over-
lying water to a maximum extent at the times of
high water a few days after new and full moon -
in other words on waning spring tides. In the
field such behaviour will ensure that the isopods
remain in the water long enough to be carried
down the slope, where they will be covered by
subsequent neap tides. In contrast, very little
swimming takes place during neap tides in the
presence of sand (Fish, Fish, 1972; Alheit, Naylor,
1976), although a circatidal swimming rhythm is
seen at both spring and neap tides if no sand is
provided (Jones, Naylor, 1970). The animals
swim more readily at night than during the day.
It may be that the diurnal swimming rhythm and the
circa-semilunar rhythm of emergence from the sand
are distinct from one another, a suggestion which
highlights the potential complexity of the respons-
es of sandy-beach Crustacea to changing conditions.
Similar responses have been described for Exe~o­
lana (Enright, 1972), as well as for the amphipod
M~nogamm~U6, which has a .particularly marked
577
diurnal rhythm due to photonegativity (Fincham,
1971) •
Jones (1970) has shown that mechanical disturbance
of the sand is an important trigger in the emerg-
ence of E~ydiee pule~, just as it is in Ex~­
olana, SyneheLLdium, EmvUta, Vonax and ButtLa.
The difference in this respect between animals
from high energy and relatively low energy sandy
shores has been highlighted by Brown (1973), who
demonstrated that water currents of sufficient
strength to disturb the sand provoked emergence
in E~ydiee longieo~~ from high energy Cape
Peninsula beaches, but induced burrowing in the
sheltered-shore Exo~pha~oma tnuneatit~on.
These contrasting reactions are also shown by·
ButtLa dig~ and B. ~hodo~toma from surf-
swept beaches as compared with B. laev~~ima, a
predominantly subtidal species which is only
intertidal on very sheltered shores (Brown, 1961,
1982a).
The types of rhythmic behaviour outlined above
are not limited to aquatic forms but are also
found in semi terrestrial species, such as the
isopod Tylo~. Most species of this genus are
nocturnal, emerging from their burrows above
high watermark on the ebb tide, feeding on any
stranded material which may be present lower
down the slope, and then returning. Kensley
(1974), working on T. gnanulatUh, and T. eape~~,
demonstrated a 24.8 hour diurnal rhythm of emerg-
ence and burial, the isopods thus emerging later
each night. After some 15 nights, when mainten-
ance of the cycle would result in it growing
light while the animals were still on the surface
of the sand, the cycle is switched back a tide,
so that nocturnal emergence and foraging is
maintained. These animals thus resemble many of
the aquatic psammophiles in possessing a nocturn-
al, semi-circatidal rhythm upon which is super-
imposed a semi-circalunar rhythm. Orientation up
and down the beach is related to a number of
factors. Both T. l~eillei and T. punetatU6
578
orientate to light, including moonlight (Pardi,
1954, 1955; Hamner et al., 1968) but they also
respond to the slope of the beach, in the absence
of light, even when the slope is only 1.50 • More-
over, the latter response reverses according to
whether the animals are on wet or dry sand. In
the South African species of Tylo~ the juveniles,
unlike the adults, allow themselves to be carried
up the beach by the waves, rolling into a ball
with an air bubble between their legs (Kensley,
1974) •
Rhythmic behaviour similar to that of Tylo~ is to
be found in semi terrestrial amphipods, including
TaLi;ttw-6, Of1.c.hu:Ua, Of1.c.hu:to'[dea and TalOf1.c.hu:Ua
(Featherston, MacIntyre, 1957; RUppel 1 , 1967;
Wildish, 1970; Craig, 1971; Benson, Lewis, 1976;
Muir, 1977; Williams, 1979). It seems that the
strength of the endogenous tidal rhythm in these
amphipods is correlated with the extent to which
the species migrates down the shore (Bregazzi,
1972; Bregazzi, Naylor, 1972; Newell, 1979). The
mechanisms of orientation of TatitnU6 in its
migrations up and down. the shore have received
much attention (Williamson, 1951, 1953; Pardi,
Papi, 1952, 1953; Papi, Pardi, 1953; Pardi, Grassi,
1955; Pardi, 1960; Williams, 1980).
TatitnU6 ~attatOf1. can orientate to the sun, to
the moon and to polarised light. The experiments
in which Pardi, Papi (1952) transferred animals
from one coast to another and showed that they
still orientated to the sun, despite the fact that
they then moved in the "wrong" direction across
the beach, are justifiably famous. Eggs removed
from T. ~attatOf1. females and reared without any
view of the sky, developed into hoppers showing
the same orientation as the parent. The orient-
ation is thus either genetically determined or it
is aquired by the early embryo while carried up
and down the beach in the marsupium. Incredibly,
the animals can correct for the movement of the
sun or moon across the sky, implying the possess-
ion of an internal clock distinct from that
concerned with the tides themselves. This
suggestion has been confirmed by Pardi, Grassi
(1955) .
The responses of TatitnU6 are complicated by
other factors, such as whether the sand is wet
or dry (Pardi, Papi, 1961). Williamson (1953)
further claimed that T. -6attatOf1. from British
shores did not require a view of the sun or moon
in order to orientate correctly. He devised
experiments which indicated that distinctive
landward outlines, breaking the horizon, were
sufficient orientational clues. Brown (unpubl.)
was unable to gain similar results with either
TalOf1.c.hu:Ua c.apeYlJ.>M or T. quadfLMpbwM; on the
other hand such abilities are not foreign to some
decapod Crustacea (see below).
Activity rhythms in ocypodid crabs have been
studied by Barrass (1963), Hughes (1966) and
Jones (1972), while their behaviour patterns have
been reviewed by Herrnkind (1972) and by Creutz-
berg (1975). In both Oc.ypode c.eJLatophthalmU6
and O. Q~ there is a semi-circalunar rhythm
synchronised with a 14-day spring/neap cycle, as
well as with a diurnal sequence and with the
ebb and flow of the tide. However, Jones (1972)
has shown that neither species has a very strong
endogenous rhythm, emergence being related primar-
ily to the amount of water in the burrow. The
photonegativity of some ocypodids may also have
been overstressed and at least in some areas their
predominantly nocturnal behaviour may be partly
due to human activities during the day (Hughes,
1966).
Herrnkind (1968) .has demonstrated that some
ocypodids, although normally using the sun or the
moon to orientate, may, under overcast conditions,
orientate towards prominent landmarks, while the
possibility exists of actual recognition of some
such landmarks. Some semi terrestrial psammophiles
may even show a homing ability based on a memory
of past movements, a sort of kinasthetic memory
(Altevogt, Von Hagen, 1964; Von Hagen, 1967).

Herrnkind (1968) has rightly stressed the adaptive
value of being able to modify responses to visual
cues according to prevailing conditions.
Certainly it is not only the rhythms themselves
that are complex in sandy-beach animals, but also
the interactions of the responses to factors that
control them. In general non-directional stimuli,
such as disturbance of the sand or its liquefact-
ion, changes in temperature or pressure, act as
releasing factors, while directional stimuli such
as light and water currents are orientational cues
(Creutzberg, 1975). Many of the factors acting
as releasers or orientational cues are not in
doubt; for example the importance of light, includ-
ing polarised light, to both aquatic and semi-
terrestrial forms is well established (Schone,
Schone, 1961; Daumer et al., 1963; Creutzberg,
1975), as is the value of changes in hydrostatic
pressure as a synchronising factor or releaser in
animals such as Synchetidium, C~clnU6 and E~yd­
ice (Enright, 1962, 1963; Williams, Naylor, 1969;
Fish, Fish, 1972). However, it is now clear that
in general a number of environmental factors act
together to produce a response and that reports
of single factors producing single responses,
while they may certainly occur in the laboratory,
are simplistic as far as field realities are con-
cerned. Thus Cubit (1969) considered that, while
liquefaction of the sand is one important factor
in triggering the emergence of Emenita, the mole-
crab repeatedly moves to its optimal feeding zone
on the beach by reference to a combination of
stimuli derived from currents, depth of water,
slope of beach and light intensity.
Another approach has been used by Naylor et al.
(1971), who investigated how the crab C~clnU6
phases its rhythm. Using arhythmic animals, they
found that a persistent rhythm could be entrained
artificially in the laboratory ny a cycle of
immersion and exposure to air, by a temperature
cycle, or by alternations of hydrostatic pressure.
Clearly in the natural environment all three
579
variables, and possibly others as well, act to-
gether and reinforce one another.
The essential role of environmental cycles in
phasing and maintaining endogenous rhythms has
been demonstrated on numerous occasions and with
many different species by keeping the animals
in the laboratory under constant conditions.
After some time the activity cycle invariably
declines in intensity and gradually becomes out
of phase with the environmental cycles present
on the shore. It is also reported that psammo-
philes such as C~clnU6 and Cnangon enangon, as
well as some cirolanid isopods, which move off-
shore during winter, lose much of their rhythmic-
ity of behaviour during this period (Edwards,
1958; Naylor, 1962; Naylor et al., 1971). The
gradual alteration of the rhythmic period under
constant conditions is thought to be due to a
slight imbalance between endogenous factors which
tend to advance or delay the phase of the rhythm.
Under natural conditions such opposing tendencies
would allow the rhythm to adapt rapidly to alter-
ations in environmental phase; the mechanism has
been termed "autophasing" (Brown, 1972).
It may be convenient at this point to state what
I hope is the main thread which is to hold this
review togther. It is that, although it is
convenient to consider separately the physical
and chemical conditions pertaining to sandy
beaches, the distribution and ecology of the
organisms inhabiting them and the physiological
and behavioural peculiarities of the individual
species, the system can really only be fully
understood as a whole. For example, the highly
complex and sophisticated sensory physiology of
many sandy-beach animals is dictated by their
need to undertake tidally-linked migratory
excursions, and this in turn is necessary because
of poor or erratic food resources - and food is
scarce largely due to the instability of the
substratum and the consequent lack of primary
production on the beach itself, and the instabil-
580
ity of the substratum is dictated by current
patterns and wave action. One should not lose
sight of this thread of cause and effect if one
is to reach a broad understanding of those aspects
of the physiology of sandy-beach animals which
are related to their ecology. In this light the
central fact of ecophysiological considerations
with regard to the macrofauna of high energy inter-
tidal sands resides in the tidal excursions under-
taken by the animals.
While some aspects of this rhythmicity have now
been studied in quite considerable detail, little
attention has been paid to the importance of these
rhythms to the physiological well-being of the
animals. Dalley (1980) found that the zoeae of
the shrimp Cnangon survived markedly better under
circadian light-dark regimes than under non-circad-
ian conditions, thus supporting the hypothesis
that the influence of rhythmic cycles is important -Gray (1966c, 1967a, b) studied the responses of
in co-ordinating physiological processes within
the animal. On the other hand, the whelk Buliia
will survive almost indefinitely under constant
conditions (Brown, 1961, unpubl.). It may well
be that this difference is to be correlated with
the absence of an endogenous rhythm in these whelks. importance. However, it was clear that this
Unlike the macrofaunal species, the sandy-beach
meiofauna tends not to make tidal excursions
above the sand but remains within it, migrating
vertically through the substratum in phase with
the tides and seasons of the year. These move-
ments appear to be commonly linked to preferences
in temperature, salinity and moisture (see Section
7).
3. CHOICE OF HABITAT
At least once during their lives, the majority
of intertidal psammophiles must choose their
habitat. Such choice may be critical at the end
of larval development, at metamorphosis, or, as
is the case with the macrofauna of high energy
beaches, the choice may have to be made repeatedly
as their migratory behaviour carries the animals
to less favourable or unsuitable sites. The
choice may also have to be made seasonally for
those animals which migrate off-shore for part
of the year.
Until the 1930s it was generally assumed that the
settlement of larvae occurred more or less at
random, with only those individuals settling in
suitable areas having a chance of survival. It
has since been shown that a number of sandy-beach
species explore the substratum before settling
and may even delay metamorphosis until suitable
conditions are found. The classic work of Wilson
(1948, 1952, 1953a, b, 1954, 1955) demonstrated
that chemical cues were of great importance in
the settlement of larvae of Ophelia biQon~ and
that this attraction depended largely on the film
of micro-organisms coating the sand grains.
Pnoxo~U6 ~ymbio~Qu~ to sand grains whose
surface had been modified by various procedures.
He was able to show a correlation between numbers
of bacteria and attractiveness of sand and also
that the type of bacteria present was of great
attractiveness was a result of the organic film
developed by the bacteria rather than of the
bacteria themselves. The results of Meadows
(1964) on Conophium are consistent with this
finding. The presence of particular species of
bacteria, rather than total bacterial numbers,
has also been shown to be of importance in the
distribution of the psammophilic harpacticoid
Lepxo~taQU6 QO~~~ and the interstitial
gastrotrich Tunbanetta hyalina (Gray, Johnson,
1970).
Day, Wilson (1934) discovered that the larvae of
SQo{eQo{ep~ 6u{gino~a could delay metamorphosis
for several weeks until a suitable substratum was
located; they also concluded that this suitability
depended partly on the grade of the deposit and
partly on its organic content. Even earlier,

Mortensen (1921) found that when larvae of the
echinoid MettLta ~exi~p~6o~ata which were ready
to metamorphose were placed in vessels containing
sea water and natural sand, metamorphosis occurr-
ed within a few days, while larvae deprived of
sand failed to metamorphose.
Habitat choice is not always critical, however,
and some species appear to show little preference
for particular substrata. For example, although
Nab~~U0 ob~otetU0 veligers display a preference
for substrata suitable for adult life, N. vibex
has no distinct habitat preferences (Scheltema,
1961). Such conclusions, must, however, be
treated with some caution. Vo~ax ~~a spat
settles annually on the northern stretch of beach
at Noordhoek, in the Cape Peninsula, despite the
fact that adult bivalves of this species are not
to be found there and that winter storms severely
erode the beach each year (Brown, unpubl.). It
might appear that these larvae are settling in a
quite unsuitable habitat and do not survive a
season. This is not the only possible explanat-
ion, however, and it may be that they settle on
the beach only temporarily before being resuspend-
ed in the water column for transport to areas
more suitable for the adults (Trueman, pers.
comm.).
Responses of adult psammophiles to a wide variety
of stimuli enable them to recognise and stay in,
or return to, those areas best suited to them.
Some of these have been mentioned in the previous
section, in relation to direction-finding up and
down the shore. Swimming forms on high energy
beaches commonly show a strong positive rheotaxis,
in contrast to forms from quiter waters (Brown,
Talbot, 1972; Brown, 1973, 1976). This response
to water currents undoubtedly helps to maintain
the animal's position. Nevertheless, they can-
not help but be carried to some extent along the
shore, where the structure of the sediments may
be different. Animals such as E~ydiQe can
discriminate between different grades of sand
581
by trial burrowing and will not burrow into
grades which are unsuitable (Jones, 1969; Brown,
1973). The mysid Gab~o~aQQu~ p~ammody~~ can
also choose among substrates differing in particle
size distribution, preferences becoming less
marked as the animal increases in size (Brown,
Talbot, 1972); that preferences among juveniles
of this species should be more acute than those
of the adults is apt in that newly-hatched indiv-
iduals are released from the marsupium some way
off-shore and, showing no immediate tendency to
burrow, swim until they reach the shore, when a
correct choice of substratum becomes critical.
Adult C~a~go~ display a somewhat different method
of habitat selection, sinking through the water
at a rate dependent on temperature and salinity,
so that they reach the substratum fastest under
optimal conditions (Spaargaren, 1980).
Like macrofaunal psammophiles, the distribution
of some species of the meiofauna, such as the
oligochaete M~o~~a, is governed partly by a
choice of particle size range (Locy, 1977). It
should be noted, however, that grain size and
factors dependent upon it may interact with
variables such as oxygen tension and pore space
to determine the distribution of the meiofauna.
Indeed the apparent preferences of many species
for particular grades of sand may be due mainly
to other, related factors, such as water circul-
ation and the availability of oxygen (Jansson,
1966, 1967; Salvat, 1967). On the other hand,
Hockin (1982), studying the colonisation of
artificial monometric sediments by harpacticoid
copepods, found that different species colonised
sediments of different particle size, endopsammic
forms colonising the coarsest matrices, mesop-
sammic species the intermediate, and epipsammic
the finest. In the absence of any other abiotic
differences, the non-random distribution of the
species was attributed directly to differences
in interstitial pore diameter.
The ability to choose between different grades
582
is not universal, however, and some species, such
as the whelk BuLUa, appear to have no particle
size preferences, even although particle size
distribution affects the speed of burrowing
(Brown, 1961, 1982a).
Field observations, such as the contention that
nematodes tend to dominate the finer sediments,
harpacticoid copepods the coarser sands (Gray,
1971; McLachlan et al., 1981; Hennig et al.,
1981), require to be studied in the laboratory
before it can be stated with any confidence which
factors associated with coarse and fine sands
are important in these distributions.
4. NUTRITION
That feeding periods are linked to the tidal
cycle is not a peculiarity of sandy shores but
appears to be true of all intertidal environments
(see Newell, 1979). However, as noted above,
feeding excursions and migrations up and down the
shore are far more marked on high energy sandy
beaches than on any other type of shore. Food
resources are maximised as a result, although an
obvious requirement is that. the energy gained by
undertaking the excursion must exceed its cost.
The absence of primary production on exposed inter-
tidal sands dictates a predominance of filter feed-
ers and scavengers among the resident intertidal
community. The physiological and behavioural
adaptations of the scavengers are largely linked
to a highly erratic food supply, while the filter
feeders rely on a more constant but poorer diet.
There are thus very few highly specialised feeders
on these beaches, opportunism being the order of
the day. An exception is provided by the mole
crab Em~ talpoida, which can only filter feed
in a current of water, although E. anaLoga can
feed also in quiet water by waving and twisting
its antennae (Marshall, Orr, 1964). In contrast,
most sandy-beach animals appear to have several
methods of feeding. For example, ocypodid crabs
are both scavengers and predators, displaying
a remarkable ability to locate, dig up and capture
their prey (Robertson et al., 1980, 1981; Hughes,
1966); yet they also feed by scooping up the top
few millimetres of sand by means of their chelae,
the sand being transferred to the mouthparts,
which sort it into fine and coarse fractions, the
latter emerging as pseudofaecal pellets (Hughes,
1966; Robertson, Newell, 1982; Robertson ,Pfeiffer,
1982) . OQljpode Qe.Jw..toph.thaLmUJ.J uses th i s method
of feeding to ingest the meiofauna in particular
(Hughes, 1966; Jones, 1972), while O. gaudiQhau~
appears to ingest primarily the microfauna which
encrusts the grains (Crane, 1941; Tweedie, 1950).
O. quadAa..ta is said to be able to ingest up to
70% of the available algae present in the harvest-
ed substratum. Vo~a also displays a wide
range of diet and removes small organisms from
sand grains rotated between the mouthparts (Alte-
vogt, 1957). This crab extracts between 10 and
30% of the total organic matter available in the
sand (Fishel son, this volume).
The mysid G~tho~aQQu~ p~ammodljte~ also has sever-
al methods of feeding. It is both a suspension
and a deposit feeder, and can in addition scavenge
soft animal material (Brown, Talbot, 1972).
BuLUa, a typically opportunist, scavenging whelk,
will turn predator after a period of enforced
starvation and can adopt several methods of ob-
taining its prey (Brown, 1964b, 1982a). Even
species which seem at first sight to be highly
specialised feeders appear in a different light
after more detailed study. For example, Vonax
~~a is both a suspension feeder and, given the
chance, a deposit feeder (Brown, unpubl.), while
the sandy-shore amphipod Hau~to~~ ~en~
has different feeding mechanisms for dealing with
large and small particles (Dennell, 1933). A
wide range of food items has been recorded for
meiofaunal species (McIntyre, 1969; Fenchel,
1978; Giere, 1982), but some at least of the
meiofaunal species appear to be specialised
feeders (Warwick, this volume).

Rates of feeding and assimilation efficiencies
have in general received far less attention than
the type of food eaten and methods of feeding,
although there is some information on the bivalve
Vo~ax (see Ansell, this volume) and on ocypodid
crabs (Robertson, Newell, 1982; Robertson,
Pfeiffer, 1982). Buttia can ingest up to a third
of its own tissue weight of food in a single meal,
the time taken to perform this feat depending on
the nature of the food eaten (Brown, 1964b, 1982a).
However, such scavengers have the opportunity of
feeding only very erratically, so that it is
difficult to estimate the total amount of food
ingested over a given period of time, although
minimum requirements are known in principle (Brown,
1981, 1982a). Assimilation efficiencies in such
carnivorous scavengers are expected to be high.
Koop et al. (1982) have estimated an average
assimilation efficiency for invertebrate sandy-
beach kelp grazers of 22%. This is very much
lower than the figures of 80% for carnivores and
60% for herbivores presented by Edwards (1973b)
and Ansell et al. (1978). However, the range of
measured values is certainly very great - an
assimilation efficiency of only 5.5% has been
measured for Ligia feeding on kelp (Koop, Field,
1981), while Tyto~ is said to have an efficiency
of 64%. TatohQh~tia and kelp-fly larvae display
intermediate values of 30 to 50% (Stenton-Dozey,
Griffiths, 1980).
Another aspect which has received little attention
is the pattern of responses which lead to feeding.
An exception is provided in the case of surfing
species of the whelk Buttia (Brown, 1971). This
animal can detect extremely low concentrations of
substances emanating from potential food, and
particularly trimethylamine and related compounds,
by means of its osphradium (Brown, Noble, 1960).
If not recently fed, it reacts to this stimulus
by emerging from the sand if buried and, if not
already in the wash zone, surfing ashore. Once
in the wash, the whelk crawls towards the food,
583
the straightness of its path depending on the
degree of interference from the wash. The siphon
is held horizontally, so that water from the
surface film overlying the sand continues to be
sampled and passed over the osphradium even when
the wash has retreated. Constant movements of
the siphon from side to side permit an accurate
orientation towards the food which would not
otherwise be possible using a single organ of
chemoreception. Thrusts of the everted proboscis
are initiated on contact with the food and the
whelk may also feel it with the cephalic tentacles.
In order for ingestion to take place, the food
must contain substances which add a further stim-
ulus, possibly acting on the sub-radula organ,
although this is not well developed in Buttia
(Brown, 1982a). Amino acids are effective in
stimulating ingestion, while neither sugars nor
the volatile amines which served to attract the
animal will initiate feeding (Brown, 1971).
The possible uptake and metabolism of dissolved
organic matter, including amino acids, by sandy-
beach invertebrates is a phenomenon which might
well repay further study. That such material
may represent a valuable food source has been
shown by Stephens (1967, 1968), Reisch (1969)
and others, while McIntyre et al. (1970) appear
to have been the first workers to demonstrate
that the interstitial meiofauna could exist large-
lyon dissolved organic material. Colclough,
Brown (in press) have recently shown a marked
ability on the part of Buttia dlg~ to take
up dissolved amino acids from extremely low
concentrations of these substances in sea water.
Uptake was not only through the gill but to an
equal extent through the surface of the foot and
occurred in quantities which could make a very
considerable contribution to the nutrition of
the adult animal. It may be even more significant
in the case of young juveniles.
584
5. LOCOMOTION
There are no truly sessile animals on open sandy
beaches and locomotion, particularly with regard
to tidal excursions, plays a much bigger part in
the lives of the animals than it does on other
types of shore. A number of different forms of
locomotion are in evidence, the one type which
all sandy-beach animals have in common being
burrowing. In addition, swimming is common among
the Crustacea, while crawling, running, leaping
and surfing also occur.
The mechanisms of burrowing into soft substrata
have received much attention since the pioneer-
ing work of Wells (1944, 1954) and Chapman,
Newell (1947) on burrowing in the polychaete
Anenicala, work which has been extended by such
experimenters as Trueman (1966a, b) and Seymour
(1971). In most cases burrowing involves two
distinct phases: penetration of the substratum
followed by burrowing movements. Burrowing is
facilitated to a greater or lesser extent by the
thixotropic properties of the sand, which result
in the abil ity to penetrate the substratum with-
out the application of large forces. Initial
penetration may be slow on low energy sands -
from 20 seconds in Anenicala m~na up to 15
minutes in the anemone Peachla (Ansell, Trueman,
1968; Trueman, Ansell, 1969) - but has to be far
more rapid on surf-swept beaches, where the
danger of being carried away by the waves is
constantly present. In such cases, larger forces
may be applied in penetration, as they are in
certain bivalves (Trueman et al., 1966; Trueman,
Ansell, 1969), in the whelk B~ (Trueman,
Brown, 1976) and in the mole crab Em~ (True-
man, 1970).
All species of Vanax are rapid and efficient
burrowers, achieving burial in only a few digging
cycles (Trueman et al., 1966; Nair, Ansell, 1968;
Ansell, Trevallion, 1969; Trueman, Ansell, 1969).
The times required for burial in tropical species
are only five to six seconds, and even less in
Vanax ~emigkana~U6 (Mori, 1938; Trueman, 1971).
Temperate species take somewhat longer (Ansell,
Trevallion, 1969; McLachlan, Young, 1982). The
animals also make powerful upward movement when
too deeply buried, pressing firmly downwards
with the foot. They are capable to taking down-
ward steps into the sand in response to wave
crash (Ansell, this volume; Brown, unpubl.).
In B~ burrowing is essentially an adaptation
of surface locomotion, the propodium being insert-
ed into the sand as a mobile, freely progressing
wedge (Brown, 1971; Trueman, Brown, 1976). Burrow-
ing takes longer than it does in Vanax, being
completed in about 10 digging cycles lasting up
to 45 seconds (Trueman, Brown, 1976; Brown,
1982a) .
In most soft-bodied psammophiles, burrowing move-
ments subsequent to penetration are very similar.
They depend essentially on the development of a
penetration anchor against which a protrusible
organ such as a foot (in the case of a mollusc)
or a proboscis (in the case of an annelid) can
be extended through the sand. This is followed
by the establishment of a terminal anchor, after
which the penetration anchor is released and the
animal drawn through the sand by the contraction
of longitudinal muscles. In bivalves and in the
whelk B~, the release of water from the
mantle cavity at intervals serves to liquefy the
sand and so facilitates burrowing (Trueman, 1968;
Trueman, Brown, 1976). Nevertheless, the forces
exerted and the internal pressures developed
during burrowing appear always to be greater
than those measured during the initial penetrat-
ion (Wells, 1954; Trueman, 1968, 1975; Trueman,
Brown, 1976).
In Crustacea, burrowing mechanisms rely less on
hydrostatic mechanisms and the contraction of
longitudinal muscles than on digging movements
of the pereiopods. A great variety of methods
is employed, from the backward burrowing of some
forms, such as Ovalip~ and Em~ (Pearse et

al., 1942; Trueman, 1970; Caine, 1974) to the
side burrowing of ocypodids (Vanini, 1980), the
virtually horizontal burrowing of G~tno~aee~
and Exo~pha~oma (Brown, Talbot, 1972; Brown, 1973)
and the head-first burrowing of Tylo~ and E~yd­
ice ~Brown, 1973; Kensley, 1974).
Crawling and running on the surface of the sand
are typical of the semi-terrestrial crustacean
fauna. It is less common in the aquatic forms,
although sandy-beach whelks crawl in the wash
zone in search of their food; in Bu.U.ia., crawling
and burrowing movements are similar. However,
much lower forces and internal pressures are
developed during crawling (Trueman, Brown, 1976),
with correspondingly less energy expenditure
(Brown, 1981,1982a).
The utilisation of the surf to migrate up and
down the beach, which has been observed in some
species of Vonax (Ansell, Trevallion, 1969; Ansell,
Trueman, 1973), in Emenita (MacGinitie, 1938;
Efford, 1965; Cubit, 1969), in the young of some
species of Tylo~ (Kensley, 1974), and in sandy-
beach whelks (Kornicker, 1961; Brown, 1961, 1971,
1982a; Ansell, Trevallion, 1969), has been studied
in detail only in Bu.U.ia. (Trueman, Brown, 1976;
Brown, 1979a, 1981). Surfing in this animal
begins in response to water currents and involves
maximum turgour of the foot, internal haemocoelic
pressures being comparable with those developed
during burrowing and more than five times those
associated with crawling. Energy costs per unit
time are also comparable with those of burrowing,
although they are cheap in terms of distance
travelled (Brown, 1979a, 1981, 1982d).
Some species of Vonax can perform leaping move-
ments over the sand (Stoll, 1937; Ansell, 1969;
Ansell, Trevallion, 1969). A sudden leaping
movement has also been reported for G~tno~aee~
p~ammody~~ in response to pressures in the sand
associated with the approach of a potential
predator (Brown, Talbot, 1972).
585
6. RESPIRATION, METABOLISM AND ENERGETICS
Respiratory mechanisms and processes in marine
animals, as well as the factors affecting oxygen
uptake, have been reviewed on a number of occas-
ions, most recently and notably by Newell (1979).
It is clear that the physiological adaptations
in the respiration of aquatic animals inhabiting
low energy sandy beaches are different from those
found on surf-swept beaches or indeed on rocky
shores, and are associated primarily with the
incidence of low oxygen tensions, particularly
during ebb tide. The lugworm A4enieola provides
a classic example of tidally-linked adaptations
to reduced oxygen availability (Wells, 1949a, b;
Dales, 1958; Borden, 1931; Bancroft, 1934; Wolver-
kamp, Vreede, 1941; Fox, 1955; Jones, 1963;
Mangum, 1977). Adaptations to low oxygen levels
may by behavioural or purely physiological or
both. For example, Callia~~a exposed to poorly
oxygenated water can either increase its ventil-
ation rate or increase its efficiency of extract-
ing oxygen from the water or may employ both
methods simultaneously (Thompson, Pritchard,
1969). Such adaptations are shared by many of
the invertebrates of muddy shores, a fact which
is undoubtedly partly responsible for the ability
of many species to colonise both.
On high energy, surf-swept intertidal sands, a
high proportion of the macrofaunal species are
very much more active than those of low energy
shores, this higher activity being especially
associated with tidal feeding excursions. Yet
the amount of food available is generally lower
than is found on sheltered shores. It is thus
not surprising to find that the accent here
appears to be on reduction of metabolic rate
whenever possible and on other ways of conserving
energy. Adaptations to low oxygen tensions are
much less in evidence.
In those animals displaying endogenous activity
rhythms, coincidental rhythms of oxygen uptake
can be demonstrated and it would appear that the
586
amplitude of these rhythms bears a relationship
to the strength of the innate activity rhythms.
Chandraschekaran (1965) convincingly demonstrated
a strong respiratory rhythm in Emenita, while
Marsh, Branch (1979) showed a marked circadian
and circatidal respiratory rhythm to be present
in Tyio~, with the suggestion of a semi-lunar
rhythm in addition. The rate of oxygen consumpt-
ion of T. gnanutat~ during its three-hour period
of activity proved to be at least six times that
of its 21 hours of quiescence and was in some
cases nearly ten times higher. The lower rates
maintained for most of the day represent a saving
of some 1.4 cal per Kcal of tissue. In some
species this suspension of activity, with its
attendent saving in energy, may be very long-term,
as in the case of Tyio~ pun~, which hiber-
nates during the winter at depths of up to 70 cm.
During starvation, its rate of oxygen uptake
drops to about a sixth of its initial value and
a comparable or even greater decline presumably
occurs during hibernation (Hayes, 1969).
Tidal rhythms of oxygen consumption have been
noted in Canein~ (Arudpragasam, Naylor, 1964;
Aldrich, McMullan, 1979), as well as in the mysid
G~~o~aee~ (Dye, 1980) and some cirolanid
isopods (Brown, unpubl.). The aquatic crab
Ovat£pe¢, which displays a circadian rather than
a tidal rhythm of activity has a corresponding
rhythm of oxygen consumption, the rate being
eight times higher in the dark than during day-
light (H. Du Preez, unpubl.).
Those invertebrate sandy-beach animals which do
not appear to possess endogenous activity rhythms
do not display rhythms of oxygen uptake either;
this is true of most of the molluscs, for example
Vonax (Ansell, this volume) and Buttia (Brown,
1982a). Moreover, the difference in rates of
oxygen consumption between active and quiescent
states in these animals is not as marked as in
those showing endogenous rhythms. The active
rate in Buttia dig~~ is not much more than
twice our best estimate of quiescent rate (Brown,
1979a, 1981). Nevertheless, behavioural and
metabolic control mechanisms keep the energy
expenditure of such animals as low as possible
and Brown (1981, 1982b) has shown that the cost
of free existence of an individual of B. dig~
of 750 mg dry tissue weight is only about 52 cal
over a 24 hour period.
Such low values are not encountered in all species
of Buttia, however, and weight-for-weight the
tropical Indian B. meianoide¢ has a routine rate
of oxygen uptake about an order of magnitude
higher than B. dig~ at local sea-water
temperatures (Brown et al., 1978). The non-
acclimation which this implies has been confirmed
for B. dig~~ by Brown, da Silva (1979),
although B. nhodo~toma and B. puna do show some
metabolic acclimation to long-term temperature
changes (Dye, McGwynne, 1980). The tropical
bivalve Vonax ineann~ also respires much
faster at its environmental temperature of 30 0 C
than does the temperate V. v~ at 15 0 C,
possibly representing another case of non-acclim-
ation (McLusky, Stirling, 1975).
An important method of conserving energy is to
reduce the effects of acute increases in environ-
mental factors particularly where water and/or
air temperatures may fluctuate rapidly. This
may be achieved by avoiding the higher temperat-
ures, as in the case of Tyio~, which spends the
daylight hours deep within its burrow (Marsh,
Branch, 1979). It may also be achieved by a
physiological flattening of the metabolic rate:
temperature curve. A noteworthy example of the
latter strategy occurs in the whelk Buttia digi-
~, where the Ql0 for the routine rate is only
about 1.1 over the range of temperatures normally
encountered by the animals in the field (Brown,
da Silva, 1979). This relative temperature
independence is, moreover, in evidence at all
levels of activity, unlike other marine inverte-
brates that have been studied (Newell, Branch,

1980), although tissue homogenates display a
temperature dependence (Brown, da Silva, in press~
The flattening of the curve is of particular
benefit to B. ~g~ on the west coast of
southern Africa, where rapid and marked fluctua-
tions in water temperature occur due to spasmodic
upwelling (Brown, Jarman, 1978). B. ~hodo~toma
and B. puna from the south coast have a much
steeper acute rate:temperature curve but, unlike
B. ~git~, show some seasonal acclimation to
temperature (Dye, McGwynne, 1980).
The rate of oxygen consumption associated with
activity in Vonax V~U6 is also markedly
dependent on temperature (Ansell, 1973). In fact,
there is a fairly substantial literature on resp-
iratory rates in the genus VoVtax ('Rao, Kutty,
1969; Ansell, 1973; Ansell, Sivadas, 1973; McLusky,
Stirling, 1973; Edwards, 1973a; Mane, Talikhedkar,
1976; Ansell et all., 1978; Dye, 1979a). This is
reviewed by Ansell (this vOlume), who points out
that there appear to be significant differences
in the metabolic adaptations of Vonax from the
cooler waters of the northern and southern hemis-
pheres. Temperature adjusted rates of respiration
in Vonax adults lie close to the standard rates
for poikilotherms in general (Hemmingsen, 1960),
although there is somewhat less variation than
would be predicted by size alone. The higher
metabolic rates of the juveniles of some species
is consistent with their migratory behaviour.
It is quite common for the metabolic energy expen~
iture of small individuals of a species to be
less affected by temperature changes than that of
large individuals; among sandy-beach invertebrates
this has been demonstrated in Tafo~ch~tLa, in
Em~, and in ocypodid crabs (Rao, Bullock,
1954; Vernberg,1959). An interesting observation
is that the rate of oxygen consumption of the
mys i d GM~OMCC(L6 p~ammodyt~ is more tempera ture
sensitive in winter than in summer (Dye, 1980a)
and that light has no effect in this regard.
Most sandy-beach invertebrates in fact show
587
temperature acclimation and Em~ is often
cited as a classic example of this phenomenon
(Edwards, Irving, 1943; Rao, Bullock, 1954;
Bullock, 1955; Newell, Branch, 1980).
The effects of salinity changes on oxygen uptake
have been less studied than the effects of temp-
erature, although some attention in this regard
has been given to psammophiles such as C~angon
vuLg~ (Hagerman, 1970), C~cin(L6 (see Newell,
1979) and Buttia (Brown, Meredith, 1981). In
Buttia ~g~ both increased and decreased
salinities result in a lowering of oxygen consump-
tion. This is not usual among marine invertebrates
(Vernberg, Vernberg, 1972) and constitutes a
further example of energy conservation under sub-
optimal conditions.
Few attempts have been made to calculate energy
or activity budgets for intertidal sandy-beach
animals, although there is some relevant data,
as for example in the case of VoVtax (Ansell, this
volume). Brown (1979a,b, 1982c) has undertaken
experiments to determine as far as possible the
energetic costs of the various activities of
Buttia dig~, and, combining the results with
field data, has constructed a time-energy (activ-
ity) budget and calculated the cost of free
existence for adult individuals (Brown, 1981,
1982a). Burrowing was found to be by far the
most costly activity in terms of distance trav-
elled, some 60 cal .kg- 1 .m- 1 as compared with only
36 cal .kg- 1.m- 1 for crawling over wet sand. Both
burrowing and crawling in this animal cost much
less than the energy expended in crawling by
terrestrial slugs, however, the latter value
being in the region of 215 cal.kg- 1 .m- 1 (Denny,
1980). Surprisingly, the cost of surfing in B.
~g~ appears to be comparable with the cost
of burrowing in terms of unit time but is, of
course, much cheaper in terms of distance actually
travelled. The animals take every opportunity
to rest, buried in the sand, so that despite
their tidal excursions, involving repeated surfing
588
and burrowing, the cost of free existence averages
only about 52 cal per 24-hour day for an animal
of 750 mg dry tissue weight (Brown, 1981). It is
calculated th~t to supply this energy the animal
must feed at least once a fortnight if it is lucky
enough to find stranded mussels or other high
energy food, but much more frequently if feeding
on the cnidaria which form its staple diet in
most localities (Brown, 1982a). This calculation
does not, however, take into account the calorific
value of dissolved organic material absorbed
directly from the surrounding sea water (Colclough,
Brown, in press). It also makes the assumption
that oxygen consumption represents the total
metabolism, an assumption which is clearly more
or less untrue.
Characteristically, scavengers, including Buttia,
often ingest large amounts of food at one time
but have long periods of fasting between meals.
Some adjust their metabolic rates to these circum-
stances, increasing metabolic rate after a meal
and then allowing it to decline during fasting
(Marsden et al., 1973; Wallace, 1973). In cold
and cold-temperate lattitudes, feeding may de-
crease during winter, so that the effect of low-
ered temperatures on metabolic rate is intensified
(Atkinson, Parsons, 1973; Wallace, 1973).
R. Bally (in prep.) has recently measured the
rates of oxygen consumption of several intertidal
isopods from South African sandy beaches. In
Exc»tolana l1CLta£en6-w and in Pontoguo-td~ la.Up~
he has found that the slope of the acute rate:
temperature curve decreases above about 17.5 0 C.
This may be seen as an adaptation to survival in
shallowly buried individuals when the tide receeds,
under which circumstances a temperature of 17.5 0 C
may well be exceeded. Reduction in respiration
will not only conserve energy but al so reduce
water loss, vital under these conditions. On
west coast beaches, Exc»tolal1a l1CLta£en6-W, Ponto-
guo-td~ la.Up~ and EunydiQe lOl1g-tQOhn-W show
differences in the effects of temperature, body
weight and activity on metabolism, which may be
correlated with the zones of the beach they
inhabi t.
In contrast to the measurement of oxygen uptake
in individual species of the macrofauna, meiofaun-
al and microfaunal respiration has usually been
measured in terms of the community as a whole
(Dye, 1979b, 1980, 1981; McLachlan et al., 1979).
More work needs to be done along these lines and
techniques need to be standardised before any
substantial generalisations can be offered with
confidence (Dye, this volume). There are, how-
ever, some interesting pointers, such as the
finding of McLachlan et al. (1979) that a linear
relationship appears to exist between the flow
rate of water through the sand column and total
oxygen consumption.
It has been customary for marine biologists to
equate oxygen consumption with metabolism. This
ignores the anaerobic component of respiration,
however. Most studies on anaerobiosis have been
conducted on bivalves, including VOl1ax (Alyakrin-
skaya, 1972), although AheniQola mahil1a, CahQ-tI1Ul,
and C~l1alla are also known to undertake such
respiration and may be dependent upon it when
the tide is out (Dales, 1958; Zebe,1975; Burke,
1979; Swinbanks, Murray, 1981). Indeed, for the
animals inhabiting low energy beaches some anaer-
obic ability would appear to be essential. The
necessity for anaerobiosis on high energy sands
would at first sight appear to be less apparent,
yet there is good evidence that the surf-loving
species of Buttia can respire anaerobically
(Brown, 1979c, 1982a). It may be that a switch
to anaerobic metabolism allows a lower metabolic
rate to be maintained than would otherwise be
possible. Even semi-terrestrial Crustacea such
as OQypode may undertake anaerobic respiration,
particularly while in their burrows (Brafield,
1978), although Votitla and some other ocypodids
commonly trap an air bubble in the burrow during
flood tide, in which the animal continues to

respire (Fiedler, 1970; Hartnoll, 1973).
The ocypodid crabs are, in any case, not true air
breathers, air being circulated through water
retained in the branchial cavity or water from
the cavity being directed over the carapace,
where gaseous exchange occurs (Magnus, 1960;
Edney, 1961; Griffin, 1968; Jones, 1972; Fishelson;
this volume). As in ocypodids, many sandy-beach
decapod Crustacea show a reversed respiratory
current, this modification being of value to the
animal while buried (Caine, 1974; Arudpragasam,
Naylor, 1966; Taylor, Butler, 1973).
Gaseous exchange is not always the obvious process
that may be supposed and Brown (in prep.) has
demonstrated that, while the animal is above the
surface of the sand in oxygen-saturated water,
Buttia can absorb as much oxygen as it requires
for even its most energetic activities through
the surface of the expanded foot and that, under
certain circumstances, the gill plays little part
in gaseous exchange.
Anaerobic pathways in bivalve molluscs have been
reviewed by De Zwaan (1977). End products include
alanine and succinate, with smaller quantities of
lactate, glutamate and aspartate. In the whelk
Buttia, also, alanine appears to be an important
end product (Brown, 1982a). The position is, how-
ever, complicated, with some organs being much
more adapted to anaerobiosis than others, and with
metabolic pathways shifting with time, so that the
end products may appear sequentially (see Newell,
1979). In many Crustacea, anaerobic respiration
results in oxygen debt (Brafield, 1978; Bridges,
Brand, 1980), while in Buttia no such debt has
been deomstrated (Brown, 1979c, 1982a).
Many of the animals inhabiting low energy shores
are tubiculous and, like ~enico£a, display
rhythms of burrow irrigation (Wells, 1949a, b,
1950; Wells, Albrecht, 1951a, b). The irrigation
response in A~enico£a is not a simple response to
589
decreased oxygen tension, and closure of the
circulation as the tide ebbs actually leads to
a rapid reduction in irrigatory activity. Irrig-
ation is thus tidally linked and the responses
so adjusted that water at unsuitable temperatures
or salinities is not drawn into the burrow. By
contrast, very few aquatic animals on exposed
sandy beaches live in permanent or semi-permanent
burrows, although some of the semi-terrestrial
fauna may face similar problems.
It is clear that in sandy-beach animals in gener-
al, whether aquatic or semi-terrestrial, the
control of metabolism by behavioural or physiolog-
ical means, or both, is of vital importance. The
whelk Buttia dig~ presents an ideal model
for studying these control mechanisms in the
absence of endogenous metabolic rhythms. It has
been established that, while ATP levels vary quite
considerably in this animal (Brown, Gedye, in
press), values fall within the range of ATP levels
of the bivalves studied by Ansell (1977). The
turnover of ATP in B. dig~ represents a
weight of ATP equal to about 40% of the animal's
wet tissue weight in 24 hours (Brown, 1982a).
Values for the adenyl ate energy charge vary from
just below 1.0 to 0.5, the latter figure being
close to the lower limit of physiological well-
being for invertebrates in general (Brown, Gedye,
in press). However, in the laboratory low AEC
values are rapidly restored to the normal high
values, ADP being converted to ATP at the expense
of carbohydrate reserves if the animals are not
fed (da Silva, unpubl.; Brown, 1982a). Present
investigations of metabolic control in Buttia
dig~ are being conducted at a mitochondrial
level and are centred on the control of the
supply of substrate to these organelles under
various conditions (da Silva, Brown, in prep.).
7. ENVIRONMENTAL TOLERANCE AND PREFERENCE STUDIES
No experimental procedure in marine biology has
been more fashionable or more productive in terms
of numbers of publications than has the determin-
590
ation of the concentration, salinity, temperature
or other variable causing 50% mortality. Early
experiments in this regard were carried out,
usually in small beakers, with no regard for the
natural habitat of the animal, its normal behav-
iour or its life history. It was mandatory that
only one factor be varied at a time, a condition
almost never occurring in nature. Even later
work, while frequently somewhat more sophisticated,
tended to repeat may of the practical and concept-
ual errors of previous studies, partly perhaps
through ignorance but in some cases at least,
deliberately, so that the new results could be
compared with data already existing in the liter-
ature for other species.
In view of the artificiality of such experiments,
it may be found surprising that a fair number of
conclusions can, in fact, be drawn from them,
particularly if the normal habitat and behaviour
of the animals is also known. For example, it is
clear that in the majority of cases intertidal
animals have tolerance levels of natural variables
which far exceed those necessary for survival in
their particular habitats (Kinn~ 1970, 1971;
Newell, 1979). Exceptions do, of course, occur,
on sandy beaches as on rocky shores, and espec-
ially among psammophiles living near the top of
the slope. Ocypodid crabs, for instance, may
regularly encounter conditions close to their
upper lethal limits (Edney, 1961, 1962; Brown,
unpubl.; Fishelson, this volume). In such cases
a knowledge of the behaviour and physiology of
the animal is essential to any understanding of
how they manage to survive and even flourish.
Descent into the burrow is an obvious way of
escaping high temperatures; in addition, evapor-
ative cooling occurs in ocypodids, the water in
the branchial cavity being replaced from time to
time by entering the burrow, by plunging into the
sea, or by absorption from the substratum (Wilkens,
Fingerman, 1965; Fishelson, this volume). Evapor-
ative cooHng may also take place in some aquatic
Crustacea, such as C~c{n~, which may actively
emerge from the water and take up positions in
which such evaporation is facilitated. C~c{nu~
survives high temperatures much better under such
conditions than where evaporation cannot occur
(Ahsanullah, Newell, 1977). Of course the price
such an animal pays for keeping cool is loss of
water and it is significant that C~c~n~ can
tolerate water loss of up to 25% of its body
weight (Ahsanullah, 1969).
Meiofauna living in the upper layers of the beach
may also on occasion be near their upper limits
of tolerance (Wiesser, 1975; Wieser, Schiemer,
1977), although in most cases migration downward
through the sand will relieve the situation.
Indeed many aspects of the distribtion of the
meiofauna are becoming clearer as a result of
ecophysiological studies on the responses,
preferences and tolerances of individual species
(Jannson, 1962, 1967, 1968; Wieser et al., 1974;
Wieser, 1975; Hartwig et al., 1977; Wieser,
Schiemer, 1977).
Another conclusion to be reached from simple
tolerance studies is that in general intertidal
species are more tolerant of environmental ex-
tremes than are infratidal species. The essential-
ly intertidal Emekita talpo~da, for example,
survives salinity changes much better than does
the predominantly subtidal E. anaioga (Gross,
1957; Bursey, Bonner, 1977). Bursey (1978) sub-
jected E. talpo~da to temperature-salinity combin-
ations within the range 5 to 35 0 C and 15 to 65
ppt salinity. The entire range of salinities
was easily tolerated at temperatures below 20 0 C,
while the optimal salinity for survival at higher
temperatures was 40 ppt. The salinity tolerances
of zoea larvae of E. talpo~da proved to be greater
than required under nearshore conditions (Stirts,
Turner, 1981). The differing salinity-temperature
tolerances of Ckangon Ckangon and C. ~epzem~p~no~a
also highlight the greater resistance of inter-
tidal forms (Allen, 1966; Haefner, 1969, 1970)
and could explain why C. ~epzem~p~no~a migrates

offshore in winter, when temperatures drop. The
tolerances of C. QAangon also accord well with
its patterns of migration (Havinga, 1930; Vannini,
1980). Orr (1955) studied the effect of exposure
time on heat death in a number of species and was
able to show that intertidal animals had a greater
thermal tolerance than subtidal species.
Differences in tolerance levels are displayed not
only between intertidal and infratidal forms but
also between animals inhabiting different zones
on the shore. Indeed the correspondence between
zonational level, the relative importance of
aerial as opposed to aquatic respiration, and
resistance to desiccation, temperatures and sal-
inity changes, is now as clear for sandy beaches
as it is for rocky shores. For example, Brown
(1961) and McGwynne (1980) were able to show that
thermal to 1erance in BuLUa i s.corre 1a ted with
the zonation of the various species. Ansell,
McLachlan (1980) carried out more sophisticated
temperature tolerance studies on B. ~hodo~toma
and demonstrated that large individuals have a
higher thermal tolerance than small individuals,
which commonly occur lower down the shore. B.
~hodo~toma also displays a higher thermal toler-
ance than the more consistently and more deeply
buried bivalves of the genus Vonax inhabiting the
same beaches.
Differences in thermal resistance of meiofaunal
organisms are also to be expected in relation to
tidal level (Johnson, 1965; Tweedie, 1973; Von
Oertzen, 1973). Jansson (1968) demonstrated a
correlation between upper limits of temperature
tolerance and environmental temperatures in
populations of turbellarians, while Lasserre
(1971) showed a similar correlation for the oligo-
chaete M~onina. The thermal tolerance of
P~oto~Uh ~ymbiotiQUh, near high watermark, is
over 34 0 C, as compared with only 25 0 C for T~ban­
ctfa hyatina which occurs at low tidal levels
(Boaden, Erwin, 1971).
591
Vernberg, Vernberg (1972) have reviewed the zoo-
geographical significance of temperature and have
shown that in general species from low latitudes
can survive higher temperatures than those from
high latitudes.
It is hardly surprising that most of the work
relating to rates of water loss and resistance
to desiccation has been performed on rosky-shore
animals or on semi-terrestrial forms. Among the
exceptions is the work of Brown (1961) and Mc-
Gwynne (1980; reported in Brown, 1982a) on species
of BuLUa. These studies were, in fact, erroneous
in failing to take into account the fact that
when these whelks come ashore they bring with
them a considerable amount of water in the free
space, mantle cavity and aquiferous system (Brown,
1964a), a potentially most important buffer
against desiccation of the tissues. The animals
were subjected to the experiments with their feet
retracted, whereas in the field retraction of the
foot is extremely uncommon (Brown, 1982a). The
rate of water loss from a retracted whelk must
clearly be less than from an expanded one, al-
though there is far less water to be lost so that
death may occur much sooner. Nevertheless, in
spite of the fact that no reliance can be placed
on the actual data obtained in absolute terms,
relative values correlate extremely well with
the zonation of the different species on the
shore.
Preserving the body temperature at the cost of
water loss in semi-terrestrial species such as
OQypoda and Vo~ is clearly disadvantageous
for oxygenation as well as for thermoregulation
As mentioned above, only by a critical timing of
activities can these hazards be overcome (Bliss,
Mantel, 1968; Fishelson, this volume).
Surprisingly, sandy-beach talitrid amphipods are
poorly adapted to low humidities, despite their
semi-terrestrial status (Williamson, 1951a; Muir,
1977). The survival times of TalLtnUh ~aftato~,
O~Qhe~tia gamm~etta and Tato~Qhe~tia de~hay~ii
592
range from 9 to 16 hours at 95% relative humidity
but only from 1 to 1.75 hours at 36% relative
humidity (Williamson, 1951a). Survival times
increased, however, if the animals were allowed
to feed on moist wrack, In an experimental humid-
ity chamber they spent most of their time in the
most moist air presented to them.
Other tolerance tests include those carried out on
the mysid G~tAo~aQQ~ with regard to salinity
(Pora, Bacescu, 1939; Brown, Talbot, 1972) and the
upper lethal temperature studies on Vonax (Bodoy,
Masse, 1977; Ansell et al., 1980; Ansell, McLach-
lan, 1980; reviewed in Ansell, this volume).
Interesting and original work has been undertaken
by Zhirmunsky, Chu Li-chun (1963), who related the
thermostability of various Vonax tissues to zonat-
ion on the shore. The genus has an upper lethal
temperature range of 29 to 40 0 C. Salinity toler-
ance studies have also been undertaken on Vonax -
V. v~b~ transferred to salinities below
normal is relatively intolerant of the change,
with a median lethal salinity of only about 19.5
ppt. However, if acclimated at intermediate
salinities, this value is lowered to 10 or 11 ppt
(Castagna, Chanley, 1973).
One of the difficulties seldom faced squarely in
attempting to correlate laboratory tolerance
studies with environmental situations is the
importance of the period of exposure. Wieser et
al. (1974) have shown that in some nematode worms
the LT50 after 10 hours is only a little lower
than after 1 hour's exposure, while in Nannofaim-
oidC6 the difference, tested under the same con-
ditions, is over 50 C. The latter must avoid high
temperatures by downward migration. Many sandy-
beach meiofaunal species do, in fact, avoid
extreme conditions by migrating down through the
sand or to lower tidal levels and it is by no
means surprising to find that their preferred
conditions tend to lie midway between their upper
and lower lethal extremes, as has been shown, for
example with regard to temperature preferences in
the archiannelid P~oto~~ (Gray, 1965).
Vertical migrations of meiofauna have in fact
been recorded as a response to heavy rain (Bush,
1966), wave disturbance (Boaden, 1968), tidal
factors (Rieger, Ott, 1971; Meineke, Westheide,
1979) and changes in the moisture and oxygen
content (McLachlan et al., 1977) as well as to
tempera ture.
In most of the work undertaken on tolerance limits,
only one factor has been considered at a time.
This is unrealistic not only because such a condit-
ion is seldom provided in the field but also the
responses to different variables may show an inter-
action with one another. For example, the energy
cost of osmoregulation frequently increases with
increasing temperature and is compounded by the
decreasing oxygen tensions. There is thus
no doubt that the tolerance limits of intertidal
animals are best defined and presented in terms
of three-dimensional multi-variable response
surfaces, rather than relationships to a single
factor (Newell, 1979). However, sandy-beach
animals are singularly absent from the list of
organisms to which this approach has been applied.
My own view is that, however sophisticated the
work and its presentation, short-term tolerance
studies are of limited interest and should now
be considered archaic. They must always be
treated with extreme caution, whether they involve
natural variables or pollutants (Brown, 1976).
Much more realistic are statements such as the
finding that in general subtropical meiofaunal
species require temperatures below 33 to 35 0 C for
successful completion of their life cycles
(Hopper et al., 1973; Wieser, 1975). Mostly,
however, one looks in vain for any realisation
that it is the effect of prevailing conditions on
the reproductive potential of the population which
is of prime ecological importance, whether that
potential is impaired through the death of adults,
reduction in scope for growth, a decrease in the
number of gametes produced, failure of the cells

to develop, larval mortality or an increase in the
number of abnormalities. This concept is at last
bearing fruit in the field of pollution studies,
due to work such as that of Bayne et al. (1982)
but appears not yet to have influenced the sandy
beach ecophysiologist to any extent.
It is also necessary to realise the fact that
conditions which may be tolerated in the laboratory
may be indirectly lethal in the field. Any factor
which upsets the normal functioning of the animal
with regard to its migratory excursions, its
seasonal pattern of movements, its orientation,
its responses to food or other reactions is likely
to result in its death. Fortunately for inter-
tidal psaamophiles, biological rhythms appear to
be relatively independent of temperature, an obvi-
ous advantage, indeed a necessity, under field
conditions (Naylor, 1963). Locomotion, on the
other hand, is usually temperature dependent and
low temperatures may jeopardise the success of
migratory excursions, and prove dangerous to the
life of the animal, by reducing activity even
though the temperature does not approach the
lethal value as determined in the laboratory.
McLachlan, Young (1982) have shown that low temp-
eratures significantly decrease the rate of burrow-
ing in both bivalves and sandy-beach gastropods,
increasing the chances of the animals being swept
away by the waves before burial is complete. Low
temperatures may also prevent migratory behaviour
altogether (Trueman, pers. comm.). More work
along these lines is essential for an understand-
ing of the effects of environmental variables on
the lives of sandy-beach animals.
8. REPRODUCTION
Most of the published work on the reproduction of
sandy-beach invertebrates is descriptive, record-
ing the numbers of eggs produced, the anatomy of
the reproductive organs, the morphology of egg
cases, times of breeding, mating behaviour or
developmental stages (e.g. Williamson, 1951c;
Buchanan, 1966; Griffin, 1968; Brown, 1971, 1982a;
593
Fish, 1975; Muir, 1977). In a few cases sections
of gonads have been used to elucidate the sequence
of events associated with gamete production but
the actual physiology of reproduction has been
sadly neglected.
One of the problems facing intertidal animals is
when to reproduce. It may, for example, be
critical to coincide breeding, egg production or
release of young with a particular phase of the
tidal cycle. It is thus not surprising that fort-
nightly or monthly rhythms of reproductive
activity are common. For example, the release of
juveniles from the brood pouch of the aquatic
isopod Ex~ofana chittoni follows the same
pattern as moulting, being accomplished at a
particular stage of the lunar cycle when the tides
are getting higher on successive days; in this
way stranded juveniles, like newly-moulted adults,
will be covered by the next tide (Klapow, 1972).
Other psammophiles must make an annual decision
to reproduce and in many cases changes in temper-
ature may provide the necessary cues although
blooms of phytoplankton on which the larvae can
feed may be more important in some instances
(Himmelman, 1975). The influence of temperature
may also be indirect in influencing the animal's
energy budget, so that gamete production is only
possible at certain times of year. Seasonal food
shortages may also preclude breeding at certain
times of year.
The minimum breeding age and the number of times
breeding should occur during the life-span of the
animal are other problems to be faced. The
advantages and disadvantages of being once-breed-
ing (semelparous), as opposed to perennial (itero-
parous), have been hotly debated (Cole, 1954;
Calow, 1973; Pianke, 1976; Stearns, 1976). Clear-
ly the animal's best bet is to maximise the sum
of present and future surviving offspring. Most
of the sandy-beach macrofauna is iteroparous, the
number of offspring produced at anyone time
594
being carefully geared to the limited food supply.
It may be noted that, in general, reproductive
costs increase with successive breeding seasons
(Browne, Russell-Hunter, 1978).
McKillup, Butler (1979), working on Nahl~Uh
paupenatUh from Australian intertidal sands, have
reported a fascinating example of the adaptability
of egg production to the food supply. As food
availability decreases, these whelks produce more
eggs and more egg capsules per female, but less
eggs in each capsule. This may be explained in
terms of the relative advantages and disadvantages
of semel parity and iteroparity in different envir-
onments. Moreover, individuals from the same
population can vary their patterns of egg product-
ion in response to different levels of food supply.
This is, in fact, yet one more example of the
tendency of sandy-beach animals to be opportunist
in every way and to adapt appropriately to every
changing condition. Brown (in prep.) has also
discovered a remarkable adaptive variation in
egg case production in South African ButtZa, al-
though the significance of this variation remains
to be investigated.
As far as development is concerned, there is a
latitudinal gradient in the proportion of species
with pelagic larvae, which declines from the
equator to the poles. This may be correlated
with a parallel decline in both food availability
and temperature (Thorson, 1950). Vance (1973)
has concluded that planktotrophy is more efficient
than lecithotrophy if there is plenty of suitable
food, so that development can be rapid, and also
if planktonic predation is low. The retention of
eggs in a brood pouch (as in amphipods and isopods)
II}
may also be seen as possible in intertidal animals
only where temperatures are low enough to permit
optimal development. This may be a major reason
why talitrid amphipods on temperate beaches are
replaced by ocypodid crabs as the dominant supra-
littoral forms on tropical and subtropical beaches,
these crabs having planktonic larvae. However,
the "choice" of type of development is clearly
a complex one and has to take into account many
factors, including energetic considerations, dis-
persal of the species and maximum survival of
young. Reproductive strategies have been well
reviewed by Stearns (1976), although sandy-beach
animals are not prominent in his arguments or
among his examples. Indeed reproductive physio-
logical considerations represent a major gap in
our knowledge and this gap must be filled before
we can claim any deep insight into the factors
governing the sandy-beach community.
9. OTHER ASPECTS OF ECOPHYSIOLOGY
There are, of course, many other aspects of the
ecophysiology of sandy beaches, which I have
relegated to the end of this review either because
so little is known about them or because they are
relevant to only a few species or to particular
beaches.
One of these aspects concerns the mechanisms
underlying the apparently gregarious behaviour
of many - perhaps the majority - of sandy-beach
species. They tend to form aggregations which
lead to a markedly discontinuous distribution
along the shore (Brown, 1971; Bally, 1981; Mc-
Lachlan, this volume). Little attention has
been paid to the way in which such aggregations
come about, although in those species which
undertake feeding excursions above the sand, the
convergence of water currents in particular areas
may obviously playa major part (Cubit, 1969;
Brown, 1971). In animals feeding on fairly large
but unevenly distributed food masses, attraction
to such food will also lead to the formation of
aggregations or will maintain groups that already
exist, as in the case of ButtZa (Brown, 1971;
Ansell et al., 1972).
Dillery, Knapp (1969) followed individuals of
Em~ visually and concluded that aggregations
of this animal are passively maintained by the
physical properties of the beach. However, Perry
(1980) has demonstrated peak periods of aggregat-

ion coinciding with the maximum abundance of
reproductive females. Thus although physical
conditions may playa major part, at least in
some species, other phenomena, such as visual
cues or the release of pheromones may be of some
importance either continuously or at certain times
of year. It may be noted that chemically mediated
gregariousness has been reported in several inter-
stitial psammophiles, including the archiannelid
P~ozodk1iU6 ~ub~op~y~geU6 (Gray, 1967a), the
polychaete worm Seoletep~ 6ulgi~oha (Gray, 1971)
and the gastrotrich T~ba~ella (Boaden, Erwin,
1971). Brown (1971) was unable to demonstrate
any gregarious behaviour on the part of Bullia,
in the laboratory, but did not dismiss the idea
of pheromonal attraction during the brief breeding
season of the animal.
In contrast to the gregariousness - passive or
active - of many aquatic psammophiles, some semi-
terrestrial species avoid forming aggregations
by being more or less territorial. Indeed, the
complex territorial activity and social behaviour
of the ocypodid crabs represent far more advanced
patterns than those of the majority of sandy-beach
inhabitants. Ritualistic defence of territory
as well as courtship displays, involve learning
as well as simple responses to releasing and
orientational factors (Lisenmair, 1967; Herrnkind,
1972; Creutzberg, 1975; Hyatt, 1977; Vannini, 1980). characteristics are dictated by cycles of environ-
A phenomenon which is of some importance to the
survival of intertidal invertebrates is the abil-
ity to regenerate limbs (in the case of Crustacea)
or other body extremities. This appears to be of
particular significance in the case of bivalve
molluscs, whose siphons are commonly, and some-
times repeatedly, bitten off by fishes or other
predators. Among the Donacidae at least, such
removal of the siphons neither prevents feeding
nor affects ~urvival. The process of regeneration
in VOMX heMa has recently been studied by
Hodgson (1982).
595
The physiological and behavioural responses to
pollution displayed by sandy-beach animals have
deliberately been omitted from this review. Never-
theless, it may be noted that pollution studies
quite often provide clues with regard to normal
reactions and responses. Furthermore, pollution
stress often illicites the same pattern of
responses as do changes in various environmental
parameters. For example, it is now well known
that pollution, environmental changes and alter-
ations in physiological state can all result in
reversals of the sign of orientation (Creutzberg,
1975). Thus in some sandy-beach animals, such
as Gah~OhaeeU6 and cirolanid isopods which are
normally strongly and positively rheotaxic, un-
suitable water conditions lead to a weakening or
cessation of this response, regardless of whether
the unsuitability is conferred by pollution,
lowered salinity or other factors (Brown, 1976,
unpubl.). In Bullia, the normal tendency to
burrow is reversed under unsuitable conditions,
so that the animals emerge from the sand and,
like the isopods and mysids referred to above,
may be carried to more equitable areas (Brown,
1976, 1982a, 1982d).
10. CONCLUDING REMARKS
To sum up the ecophysiological characteristics of
sandy-beach animals, I would say that these
mental conditions which are probably more complex
than those faced by any other community in any
other habitat; waves are themselves a cyclic
phenomenon and in addition the intensity of wave
action tends to be cyclic; there are tidal cycles,
cycles of lunar periodicity, diurnal cycles,
seasonal cycles, cycles of erosion and deposition,
cycles of water movements through the sand,
nutritional cycles as far as food supply is con-
cerned; and through all these cycles, which are
to a large extent predictable, runs the constant
threat of sudden change, change which is unpred-
ictable by the sandy-beach fauna although it may
596
predictable through the sophistications of modern
science.
The aquatic animals inhabiting high energy sandy
beaches thus face, and must adapt to, unique
physical conditions. They constitute one of the
few animal communities that can do nothing to
modify the environment to their own advantage
but must accept it as it stands, with all the
changes it may impose upon them. There is nothing
they can do to ameliorate the harshness, the
relentlessness, the instability of prevailing
conditions. Unlike Voltaire's heroes and heroines,
unlike some limpets, unlike some of the nematodes
studied by Warwick (this volume), they cannot
find peace by cultivating their gardens. Not for
them the luxury of a permanent burrow or the
insurance of a quiet backwater in old age. These
are certainly animals which never need to send to
know for whom the bell tolls. They must indeed
keep pace until they die not only with the times
but also with the tides - and their built-in
clocks, neurological tide-tables, sun-compasses
and other navigational aids, count for nought
unless they can also adapt rapidly and without
too much physiological stress to changing, less
predictable circumstances. It is not perhaps
surprising, in this light, that there are so few
sandy-beach species, but rather that there are
any such species at all.
The apparent ease with which these animals can
adapt, both physiologically and behaviourly, to
changing circumstances is indeed remarkable.
This was brought home to me many years ago, through
the work of two scientists in Britain, both work-
ing on the isopod Eun&d{~~ pU£~~a. The results
reported on the behaviour of this species were
very different and I imagined that one or other
author was mistaken. It proved not to be so; the
animals were behaving differently at different
sites, their behaviour being adapted to different
circumstances. This is true also of the surfing
species of the whelk Butcia. Researchers in the
Port Elizabeth area found that B. ~hada4~ama, B.
pUka and B. d{gital~ follow the tide up and
down the beach throughout the tidal cycle (McLach-
lan et al., 1979a), whereas Brown (1961, 1971),
working on Cape Peninsula beaches, found that the
same species tended to burrow into the sand once
the tide had passed the mid-point of its flow.
Brown (1982a) has suggested that differences in
beach slope may playa major role in these diff-
erences in behaviour. Whether this suggestion is
valid or not, the fact remains that the behaviour
of the species is not uniform throughout their
range. The ability to make such adjustments in
behaviour and physiology is clearly essential in
such a variable and unstable environment.
Not only may certain aspects of the physiology
and behaviour of a species differ from one site
to another, but significant differences may occur
between groups within a single population. For
example, gravid Ga4~a4a~~U4 p4ammad&~~ tend to
remain buried when the rest of the population
emerges from the sand (Brown, unpubl.). A rather
similar phenomenon is found in some cirolanid
isopods; Fraenkel (1961) showed that individuals
of Eun&d{~~ pU£~~a attracted to a bright light
at night later became photonegative in the labor-
atory, while Jones, Naylor (1970) demonstrated
that both photopositive and photonegative animals
were present in single natural populations, photo-
taxic responses being related to the nutritional
state of the animals. Ovigerous females of both
E. pU£~~a and E. iangi~o~~ appear, on the other
hand, to be consistently photonegative (Jones,
1970; Brown, 1973).
By and large, invertebrate ecophysiologists do
not look for differences within a population,
sometimes acknowledging males and females or
juveniles and adults as constituting different
groups but failing to recognise the possibilities
of more subtle groupings determined by such factors
as nutritional state, environmental stresses, or
indeed any aspect of the animal's immediate past

history. On the contrary, experimental design
more often than not imposes an assumed uniformity
on the population, predicting a normal distribut-
ion curve for the results which may in fact be
falacious.
I have attempted to outline in this review not
only our knowledge of sandy-beach ecophysiology
but also some of the gaps in out knowledge and
some possible errors in our approach to the sub-
ject. In conclusion I would like to stress two
areas in which our ignorance prevents us from
acquiring a deep appreciation of the ecosystem
as a whole. One of these is the virtual absence
of ecophysiological work on the microfauna. This
is a particularly serious gap in view of work
such as that of Koop, Griffiths (1982) and Koop
et al. (1982) which demonstrates the importance
of bacteria in the energy flow on sandy beaches.
The situation with regard to the meiofauna is
only a little better, for although meiofaunal
ecology has received considerable attention in
recent years, it is still mostly of the 'black
box' variety (McLachlan, this volume), the meio-
faunal community being treated as a whole. This
approach extends to experimental studies as well,
despite the term 'ecophysiology' sometimes appear-
ing in the titles of published work. There is,
of course, knowledge to be derived from community
studies, some of it very valuable, but it is not
ecophysiology according to our definition and it
can never be a substitute for experimental work
on individual species.
Lastly, I would like to make a plea for ecophysio-
logical work to be extended into the surf zone.
This is critical in view of concepts now emerging
which, for a number of reasons, include the surf
zone as an essential part of the sandy-beach
ecosystem (see McLachlan, this volume). Such
studies should include not only the fish and the
crabs but also the surf zooplankton, such as
mysids and shrimps, whose influence on the eco-
system may well prove to be incomparably greater
597
than ever previously imagined.
11. ACKNOWLEDGEMENTS
I did not have time to discuss this review with
colleagues or to ask anyone to read it in draft.
I am, however, grateful to Dr Anton McLachlan for
pursuading me to write it and to Prof E.R. True-
man and Dr A.D. Ansell for valuable conversations
on the subject over the years. Prof. G.M. Branch
and Mr N. Young brought to my attention a number
of valuable references while the review was being
written.
12. REFERENCES
Ahsanullah M (1969) A comparative study of the
effects of desiccation and thermal stress on the
intertidal crab C~e1nUh mae~ (L) and the sub-
littoral crab Pontun~ m~04eUh Leach. Ph D
thesis, University of London.
Ahsanullah Mand Newell RC (1977) The effects
of humidity and temperature on water loss in
C~e1nUh mae~ (Ll and POntunUh m~MeUh (Leach),
Compo Biochem. Physiol. 56A, 593-601.
Aldrich JC and McMullan PM (1979) Observations
on non-locomotory manifestations of biological
rhythms and excitement in the oxygen consumption
rates of crabs, Compo Biochem. Physiol. 62A, 707-
709.
Alheit J and Naylor E (1976) Behavioural basis
of intertidal zonation in EU4ydice pulch4a Leach,
J. expo mar. Biol. Ecol. 23, 135-144.
Allen JA (1966) Dynamics and interrelation-
ships of mixed populations of Caridea off the
north-east coast of England. In Barnes H, ed.
Some contemporary studies in marine science, pp.
45-66. George Allen & Unwin, London.
Altevogt R (1957) Beitrage zur Zoologie und
Ethologie von Vo~ blan604di Alcock und
Vo~ myctLto~d~ (Milne Edwards) (Crustacea ,
Decapoda), Z. Morph. Dkol. Tiere 46, 369-388.
Altevogt R and Von Hagen H (1964) Uber die
Orientierung von Uca tang~ Eydoux im Freiland,
Z. Morph. Okol. Tiere 55, 641-655.
Alyakrinskaya JO (1972) Biochemical adaptat-
ions of aquatic molluscs to an air environment,
Zool. Zh. 51, 1630-1636.
Ansell AD (1969) Leaping movements in the
Bivalvia, Proc. malac. Soc. Lond. 38, 387-399.
Ansell AD (1973) Oxygen consumption by the
bivalve Vonax vittatUh (da Costa), J. expo mar.
Biol. Ecol. 11,311-328.
Ansell AD (1977) The adenosine triphosphate
content of some marine bivalve molluscs, J. expo
mar. Biol. Ecol. 28, 269-283.
Ansell AD (this volume) The biology of the
genus Vonax. In Sandy Beaches as ecosystems,
Junk, The Hague.
Ansell AD and McLachlan A (1980) Upper
598
temperature tolerances of three molluscs from
South African sandy beaches, J. expo mar. Biol.
Ecol. 48, 243-251.
Ansell AD and Sivadas P (1973) Some effects
of temperature and starvation on the bivalve
Vonax v~Uh (da Costa) in experimental laborat-
ory populations, J. expo mar. Biol. Ecol. 13, 229-
262.
Ansell AD and Trevallion A (1969) Behavioural
adaptations of intertidal molluscs from a tropical
sandy beach, J. expo mar. Biol. Ecol. 4, 9-35.
Ansell AD and Trueman ER (1968) The dynamics
of burrowing of the sand dwelling anemone PeaQhia
ha~tata, J. expo mar. Biol. Ecol. 2, 124-127.
Ansell AD and Trueman ER (1973) The energy
cost of migration of the bivalve Vonax on tropical
sandy beaches, Mar. Behav. Physiol. 2, 21-32.
Ansell AD, Barnett PRO, Bodoy A and Masse H
(1980) Upper temperature tolerances of some
European molluscs II. Vonax v~, V. ~em~~­
atUh and V. thunQuiU6, Mar. Biol. 58, 41-46.
Ansell AD, McLusky OS, Stirling A and Trevallion
A (1978) Production and energy flow in the macro-
benthos of two sandy beaches in South West India,
Proc. Roy. Soc. Edinburgh 76B, 269-296.
Arudpragasam KD and Naylor E (1964) Gill vent-
ilation volumes, oxygen consumption and respirat-
ory rhythms in CanQinUh maen~ (L), J. expo Biol.
41, 309-321.
Arudpragasam KD and Naylor E (1966) Patterns
of gill ventilation in some decapod Crustacea, J.
zool. Soc. Lond. 150,401-411.
Atkinson RJA and Parsons AJ (1973) Seasonal
patterns of migration and locomotor rhythmicity
in populations of CanQinU6, Nethl. J. sea Res. 7,
81-93.
Bally R (1981) The ecology of three sandy
beaches on the west coast of South Africa. Ph D
thesis, University of Cape Town.
Barcroft J (1934) Features in the architecture
of physiological function. Cambridge Univ. Press.
Barrass R (1963) The burrows of OQypode Qenat-
ophthatmU6 (Pallas)(Crustacea, Ocypodidae) on a
tidal wave beach at Inhaca Island, Mozambique, J.
animo Ecol. 32, 73-85.
Bayne BL, Widdows J, Moore MN, Salkeld P,
Worral CM and Donkin P (1982) Some ecological
consequences of the physiological and biochemical
effects of petroleum compounds on marine molluscs,
Phil. Trans. Roy. Soc. Lond. 292B,
Benson JA and Lewis RD (1976) An analYSis of
the activity rhythm of the sand beach amphipod
TatonQh~tia quoyana, J. Compo Physiol. 105, 339-
352.
Bliss DE and Mantel LH (1968) Adaptations of
crustaceans to land: a summary and analysis of new
findings, Amer. Zool. 8, 673-685.
Boaden PJS (1968) Water movement - a dominant
factor in interstitial ecology, Sarsia 34, 125-136.
Bodoy A and Masse H (1977) Etude sur la resist-
ance a la temperature de quelques mollusques marins
des cotes de Provence, Bull. Ecol. 8, 91-101.
Borden MA (1931) A study of the respiration
and of the function of the haemoglobin in Planonb~
QOnnCU6 and AneniQola m~na, J. mar. biol. Ass.
U.K. 17, 709-738.
Brafield AE (1978) Life in sandy shores.
London, Edward Arnold.
Bregazzi PK (1972) The effects of low temper-
ature on the locomotor activity rhythm of T~U6
~a£taton(Montagu)(Crustacea: Amphipoda), J. expo
Biol. 57,393-399.
Bregazzi PK and Naylor E (1972) The locomotor
activity of T~U6 ~a£taton (Montagu)(Crustacea:
Amphipoda), J. expo Biol. 57, 375-391.
Bridges CR and Brand AR (1980) The effect of
hypoxia on oxygen consumption and blood lactate
levels of some marine Crustacea, Compo Biochem.
Physiol. 65A, 399-410.
Brown AC (1961) Physiological-ecological
studies on two sandy-beach Gastropoda from South
Africa: ButtZa d{g~~ Meuschen and B~ laev-
~~~a (Gmelin), Z. Morph Okol. Tiere 49, 629-657.
Brown AC (1964a) Blood volumes, blood distrib-
ution and sea-water spaces in relation to expans-
ion and retraction of the foot in B~ (Gastro-
poda), J. ex p. Bi 01. 41, 837 -8 54.
Brown AC (1964b) Food relationships on the
intertidal sandy beaches of the Cape Peninsula,
S. Afr. J. Sci. 60, 35-41.
Brown AC (1971) The ecology of the sandy
beaches of the Cape Peninsula, South Africa. Part
2: the mode of life of B~ (Gastropoda), Trans.
Roy. Soc. S. Afr. 39, 281-319.
Brown AC (1973) The ecology of the sandy
beaches of the Cape Peninsula, South Africa. Part
4: observatiohs on two intertidal Isopoda, Eunyd-
{Qe long{Qon~ (StUder) and Exo~phnaenoma tnunQat-
~~on Barnard, Trans. Roy. Soc. S. Afr. 40, 381-
404.
Brown AC (1976) Toxicity studies on marine
animals, S. Afr. J. Sci. 72, 197-199.
Brown AC (1979a) Oxygen consumption of the
sandy-beach whelk B~ d{g~~ Meuschen at
dufferent levels of activity, Compo Biochem. Phys-
iol. 62A, 673-675.
Brown AC (1979b) The energy cost and efficiency
of burrowing in the sandy-beach whelk B~ d{g{-
~ (Dillwyn), J. expo mar. Biol. Ecol. 40, 149-
154.
Brown AC (1979c) Respiration and activity in
the sandy-beach whelk B~ d{g~ (Dillwyn),
S. Afr. J. Sci. 75,451-452.
Brown AC (1981) An estimate of the cost of
free existence in the sandy-beach whelk B~
d{g~~ (Dillwyn) on the west coast of South
Africa, J. expo mar. Biol. Ecol. 49, 51-56.
Brown AC (1982a) The biology of sandy-beach
whelks of the genus B~ (Nassariidae), Oceanogr.
mar. Biol. Ann. Rev. 20, 309-361.
Brown AC (1982b) Metabolic regulation in the
whelk ButtZa (Nassariidae), J. moll. Stud.
Brown AC (1982c) Pollution and the sandy-
beach whelk B~. Trans. Roy. Soc. S. Afr. 44,
555-562.
Brown AC (1982d) Towards an activity budget
for the sandy-beach whelk B~ d{g~~ (Dill-
wyn), Malacologia 22, 681-683.
Brown AC and da Silva FM (1979) The effects of
temperature on oxygen consumption in B~

dig~~ Meuschen (Gastropoda; Prosobranchiata),
Compo Biochem. Physiol. 62A, 573-576.
Brown AC and da Silva FM (in press) r~etabolic
rate:temperature relationships for intact and
homogenised Bu1Lia dig~ (Gastropoda, Nassar-
iidae), Trans. Roy. Soc. S. Afr. 45.
Brown AC and Gedye (in press) ATP levels and
adenyl ate energy charge in the whelk Buttia digi-
~ (Dillwyn), Compo Biochem. Physiol.
Brown AC and Jarman N (1978) Coastal marine
habitats. In Werger MJA, ed. Biogeography and
ecology of southern Africa, pp. 1239-1277. The
Hague, Junk.
Brown AC and Meredith FL (1981) Effect of sal-
inity changes on oxygen uptake in the sandy-beach
whelk Bu1Lia dig~, Compo Biochem. Physiol.
69A, 599-601.
Brown AC and Noble RG (1960) Function of the
osphradium in Buttia (Gastropoda), Nature, Lond.,
188, 1045.
Brown AC and Talbot MS (1972) The ecology of
the sandy beaches of the Cape Peninsula, South
Africa. Part 3: a study of GahtnolaccUh plammo-
dy:tru, Tattersall (Crustacea: Mysidacea), Trans.
Roy. Soc. S. Afr. 40, 309-333.
Brown AC, Ansell AD and Trevallion A (1978)
Oxygen consumption by Buttia (Vo~a~um) mela~oidru,
(Deshayes) and Buttia dig~~ Meuschen (Gastro-
poda; Prosobranchiata) - an example of non-acclim-
ation, Compo Biochem. Physiol. 61A, 123-125.
Brown FA jr (1972) The 'clocks' timing biolog-
ical rhythms, Am. Scient. 60, 756-766.
Browne RA and Russ~ll-Hunter WD (1978) Repro-
ductive effort in molluscs, Oecologia 37, 23-27.
Buchanan JB (1966) The biology of Echi~oc~dium
condatum (Echinodermata; Spatangoidea) from
different habitats, J. mar. biol. Ass. U.K. 46,
97-114.
Bullock TH (1955) Compensation for temperature
in the metabolism and activity of poikilotherms,
Biol. Rev. 30, 311-342.
Burke EM (1979) Aerobic and anaerobic metabol-
ism during activity and hypoxia in two species of
intertidal crabs, Biol. Bull. 156, 157-168.
Bursey CR (1978) Temperature and salinity
tolerances of the mole crab, Em~ tatpoida
(Crustacea, Anomura), Compo Biochem. Physiol. 61A,
81-83.
Bursey CR and Bonner EE (1977) Osmotic regul-
ation and salinity tolerance of the mole crab,
Em~a :talpoida (Say)(Crustacea, Anomura), Compo
Biochem. Physiol. 57A, 207-210.
Bush LF (1966) Distribution of sand fauna in
beaches at Miami, Florida, Bull. mar. Sci. 16,
58-75.
Caine EA (1974) Feeding of Ovalipru, guadutpen-
l~ (Saussure) (Decapoda: Brachyura: Portuni dae) ,
and morphological adaptations to a burrowing
existence, Biol. Bull. 147,550-559.
Calow P (1973) The relationship between fecund-
ity, phenology, and longevity: a systems approach,
Am. Nat. 197, 559-574.
Castagna M and Chanley P (1973) Salinity toler-
ance of some marine bivalves from inshore and
599
estuarine environments in Virginia waters on the
western mid-Atlantic coast, Malacologia 12, 47-96.
Chandraschekaran MK (1965) Persistent tidal
and diurnal rhythms of locomotor activity and
oxygen consumption in Em~ ah~ca (M.-Edw.),
Z. vergl. Physiol. 50, 137-150.
Chapman G and Newell GE (1947) The role of
the body fluids in relation to movement in soft-
bodied invertebrates, I. The burrowing of Ane~c­
oia, Proc. Roy. Soc. Lond. 134B, 431-455.
Colclough J and Brown AC (in press) Uptake of
dissolved organic matter by a marine whelk, Trans.
Roy. Soc. S. Afr. 45.
Cole LC (1954) The population consequences of
life history phenomena, Quart. Rev. Biol. 29, 103-
137.
Cott HB (1929) Observations on the natural
history of the racing crab Ocypode cenatoph:thalmul
from Beira, Proc. zool. Soc. Lond. 4, 755-765.
Craig PC (1971) An analysis of the concept of
lunar orientation in Onchru,:toidea con~cuta:ta,
Anim. Behav. 19, 368-374.
Craig PC (1973) Behaviour and distribution of
the sand beach amphipod Onchel:toidea con~cuta:ta,
Mar. Biol. 23, 101-109.
Crane J (1941) On the growth and ecology of
brachyuran crabs of the genus OcypodC,Zoologica,
N.Y. 26,297-310.
Creutzberg F (1975) Orientation in space:
animals. In Kinne 0, ed. Marine ecology 2: phys-
iological mechanisms, pp. 556-665. London, Wiley
Inter-Science.
Crocker RA (1977) Macro-infauna of northern
New England marine sand: long term intertidal
community structure. In Coull BC, ed. Ecology of
marine benthos. South Carolina, Belle WBaruch
Library in Marine Science.
Cubit J (1969) The behaviour and physical
factors causing migration and aggregation in the
sand crab, Em~ a~aloga (Stimson), Ecology 50.
118-123.
Dales RP (1958) Survival of anaerobic periods
by two intertidal polychaetes, Ane~coia m~~a
(L.) and Owe~ 6Uhinonm~ Delle Chiaje, J. mar.
biol. Ass. U.K. 37, 521-529.
Dalley R (1980) The survival and development
of the shrimp Cna~gon cna~gon reared in the lab-
oratory under non-circadian light-dark cycles, J.
expo mar. Biol. Ecol. 47, 101-112.
Daumer K, Jander R and Waterman TH (1963)
Orientation of the ghost crab Ocypode in polarised
1ight, Z. vergl. Physiol. 47, 56-76.
Day JH and Wilson DP (1934) On the relation
of the substratum to the metamorphosis of Scoie-
coiep~ nutgi~ola (Claparede), J. mar. Biol. Ass.
U.K. 19, 655-661.
Dennell R (1933) The habits and feeding mech-
anisms of the amphipod Haul:to~ ~en~l
Slabber, J. Linn. Soc. (Zool.) 38, 363-388.
Denny M (1980) Locomotion: the cost of gastro-
pod crawling, Science, 208, 1208-1212.
Dexter DM (1977) Natural history of the Pan-
American sand beach isopod Ex~oiana bnaz~e~l~
(Crustacea: Malacostraca), J. Zool. Lond. 183,
600
103-109.
Dillery DC and Knapp L (1969) Longshore move-
ments of the sand crab Em~ anataga (Decapoda:
Hippidae), Crustaceana 18, 233-240.
Du Preez H (1981) Biology of the three-spot
swimming crab Ovalip~ pun~. M.Sc. thesis,
University of Port Elizabeth.
Dye AH (1979a) The effect of acute and long
term temperature changes on the respiration of two
sand-dwelling bivalves, Camp. Biochem. Physiol.
63A, 405-409.
Dye AH (1979b) Measurement of biological oxy-
gen demand in sandy beaches, S. Afr. J. Zool. 14,
55-60.
Dye AH (1980a) Respiratory physiology of GahX-
no~aee~ p~ammodyx~ (Crustacea, Mysidacea), Compo
Biochem. Physiol. 65A, 187-192.
Dye AH (1980b) Tidal fluctuations in biologic-
al oxygen demand in sandy beaches, Estuar. cstl
mar. Sci. 11,1-8-
Dye AH (1981) A study of benthic oxygen consump-
tion on exposed sandy beaches, Estuar. cstl Shelf
Sci. 13, 671-680.
Dye AH (this volume) A synopsis of community
respiration studies on exposed sandy beaches. In
Sandy beaches as ecosystems.
Dye AH and McGwynne LE (1980) The effect of
temperature and season on the respiratory physiol-
ogy of three psammolittoral gastropods, Camp.
Biochem. Physiol. 65A, 187-191.
Edney EB (1961) The water and heat relation-
ships of fiddler crabs (Uea spp.), Trans. Roy.
Soc. S. Afr. 36, 71-91.
Edney EB (1962) Some aspects of the temperature
relations of fiddler crabs, Biometeriology, 79-85.
Edwards RL (1958) Movements of individual
members in a population of the shore crab CiV1.c).n~
maenah (L) in the littoral zone, J. animo Ecol.
27, 37-45.
Edwards RRC (1973a) Rates of oxygen consumption
of marine fish and invertebrates from Venezuela,
Bol. Inst. Oceanogr. Cunana Venezuela.
Edwards RRC (1973b) Production ecology of two
Caribbean marine ecosystems, II: metabolism and
energy flow, Estuar. cstl mar. Sci. 1,319-333.
Edwards GA and Irving L (1943) The influence
of temperature and season upon the oxygen consump-
tion of the sand crab, Em~ xalpo~da Say, J.
cell. camp. Physiol. 21, 169-182.
Efford IE (1965) Feeding in the sand crab
Em~ analoga (Stimson) (Decapoda, Anomura),
Crustaceana 10, 166-182.
Eltringham SK (1971) Life in mud and sand.
London, English Univ. Press.
Enright JT (1962) Responses of an amphipod to
pressure changes, Camp. Biochem. Physiol. 7, 131-
145.
Enright JT (1963) The tidal rhythm of activity
of a sand beach amphipod, Z. vergl. Physiol. 46,
276-313.
Enright JT (1965) Entrainment of a tidal
rhythm, Science, 147, 864-867.
Enright JT {1972) A virtuoso isopod: circa-
lunar rhythms and their tidal fine structure, J.
camp. Physiol. 77, 141-162.
Featherston DW and MacIntyre RJ (1957) Learn-
ing and daily activity in a sand hopper, Nature,
179, 381-382.
Fenchel T (1978) The ecology of marine meio-
benthos II. The food of marine benthic ciliates,
Ophelia 5, 73-121.
Fiedler DR (1970) The feeding behaviour of the
sand crab Seop~ena ~n6Lata (Decapoda, Ocypodidae),
J. Zool. Lond. 160,35-49.
Fincham AA (1970) Rhythmic behaviour of the
intertidal amphipod BaXhypaneia petag~ea, J. mar.
biol. Ass. U.K. 50, 1057-1068.
Fincham AA (1971) Ecology and population stud-
ies of some intertidal and sublittoral sand-dwell-
ing amphipods, J. mar. biol. Ass. U.K. 51,471-
488.
Fincham AA (1973) Rhythmic swimming behaviour
in the New Zealand sand beach isopod P~eudaga
pun~ Thomson, J. expo mar. Biol. Ecol. 11,
229-237.
Fish JD (1975) Development, hatching and brood
size in BaXhypaneia p~a~a and B. petag~ea, J.
mar. biol. Ass. U.K. 55, 357-368.
Fish JD and Fish S (1972) The swimming rhythm
of Euny~ee pufehna and a possible explanation of
intertidal migration, J. expo mar. Biol. Ecol. 8,
195-200.
Fish S (1970) The biology of Euny~ee pufehna
(Crustacea: Isopoda), J. mar. biol. Ass. U.K. 50,
753-768.
Fishelson L (1978) Tracks on the beach: life
in the tidal zone, Israel Land and Nature 3, 105-
112.
Fishelson L (this volume) Population ecology
and biology of Voti£la ~ufeaXa (Crustacea, Ocyp-
odidae) on sandy beaches of the Red Sea. In
Sandy beaches as ecosystems.
Fox HM (1955) The effect of oxygen on the
concentration of haem in invertebrates, Proc. Roy.
Soc. Land. 143B, 203-214.
Fraenkel G (1961) A new type of negative photo-
taxis observed in a marine isopod Euny~ee pufehna,
Physiol. Zool. 34, 228-232.
Gamble FW and Keeble F (1904) The bionomics
of ConvoluXa no~eo66e~~ with special reference
to its green cells, Quart. J. microsc. Sci. 47,
44-57.
Giere 0 (1982) Habitat conditions and distrib-
ution patterns of the gutless oligochaete Phalla-
dnltt~ leukodenmaX~, Mar. Ecol. Prog. Ser. 8,
291-299.
Gray JS (1965) The behaviour of Pnaxa~~
~ymb~a~e~ (Giard) (Archiannelida) in temperature
gradients, J. animo Ecol. 34,455-461.
Gray JS (1966) Selection of sands by Pnoxo-
~~ ~ymb~a~eu~ (Giard), Veroff. Inst. Meeres-
forsch, Bremerhaven 2, 105-116.
Gray JS (1967a) Substrate selection by the
archiannelid Pnoxo~~ nubnophiV1.ynge~ Jager-
sten, Helgol. wiss. Meeresunters. 15, 253-269.
Gray JS (1967b) Substrate selection by the
archiannelid Pnoxa~~ hypaleueu~ Armenante,
J. expo mar. Biol. Ecol. 1, 47-54.

Gray JS (1968) An experimental approach to
the ecology of the harpacticoid LeptahtaQu~ Qon-
~tniQt~ Lang, J. expo mar. Biol. Ecol. 2, 278-292.
Gray JS (1974) Animal-sediment relationships,
Oceanogr. mar. Biol. Ann. Rev. 12, 223-261.
Gray JS and Johnson RM (1970) Bacteria of a
sandy beach an an ecological factor affecting the
interstitial gastrotrich Tunbane££a hyatina
Schultze, J. expo mar. Biol. Ecol. 4, 119-133.
Griffin DJG (1968) Social and maintenance
behaviour in two Australian ocypodid crabs (Crus-
tacea: Brachyura), J. Zoo 1. Lond. 156, 291-305.
Gross WJ (1957) An analysis of response to
osmotic stress in selected decapod Crustacea,
Biol. Bull. 112,43-62.
Haefner PA (1969) Temperature and salinity
tolerance of the sand shrimp, Changon ~eptem~p~n­
o~a Say, Physiol. Zool. 42, 388-397.
Haefner PA (1970) The effect of low dissolved
oxygen concentrations on the temperature-salinity
tolerance of the sand shrimp, Changon ~eptem~p~n­
OM Say, Physiol. Zool. 43, 30-37.
Hagerman L (1970) The oxygen consumption of
Changon vuig~ (Fabricius) (Crustacea, Natantia)
in relation to salinity, Ophelia 7, 283-292.
Hamner WM, Smyth M and Mulford ED (1968) Orien-
tation of the sand beach isopod Ty£o~ pun~~,
Anim. Behav. 16, 405-409.
Hartwig E, Gluth T and Wieser W (1977) Investi-
gations on the ecophysiology of Ge£~a m~g~Qep~
Kahl (Cil iophora, Gymnostomata) inhabiting a sandy
beach in Bermuda, Oecologia 31, 159-175.
Hastings MH (1981) Semi-lunar variations of
endogenous circatidal rhythms of activity and
respiration in the isopod Euny~Qe puiQhha, Mar.
Ecol. Prog. Ser. 4, 85-90.
Hastings MH and Naylor E (1980) Ontogeny of an
endogenous rhythm in Euny~Qe puiQhha, J. expo
mar. Biol. Ecol. 46, 137-145.
Havinga B (1930) Der Granat (Changon vuig~
Fabr.) in den hollandischen Gewassern, J. cons.
perm. into Explor. Mer. 5, 57-87.
Hayes WB (1969) Ecological studies on the high
beach isopod Ty£o~ pun~~ Holmes &Gray. Ph.D.
thesis, University of California.
Hennig HFKO, Eagle GA, Fiedler L, Fricke AH,
Greenwood P and Orren MJ (1981) Ratio and popul-
ation density of psammolittoral meiofauna as a
perturbation indicator of South African beaches,
Sci. Total Environ.
Herrnkind WF (1968) Adaptive, visually-direct-
ed orientation in UQa pU9~atoh, Am. Zool. 8, 585-
598.
Herrnkind WF (1972) Orientation in shore-living
arthropods, especially the sand fiddler crab. In
Winn HE and Olla BL, ed. The behaviour of marine
animals, pp. 1-59. New York, Plenum Press.
Himmelman JH (1975) Phytoplankton as a stimulus
for spawning in three marine invertebrates, J. expo
mar. Biol. Ecol. 20, 199-214.
Hockin DC (1982) The effects of sediment part-
icle diameter upon the meiobenthic copepod commun-
ity of an intertidal beach: a field and laboratory
experiment, J. animo Ecol. 51, 555-572.
Hodgson AN (1982) studies on wound healing
601
and regeneration of the siphons of the bivalve
Vonax ~~ (Roding), Trans. Roy. Soc. S. Afr.
44, 489-498.
Hopper BE, Fell JW and Cefalu RF (1973) Effect
of temperature on life cycles of nematodes assoc-
iated with the mangrove (Rhizophoha mang£e)
detritus system, Mar. Biol. 23, 293-296.
Hughes DA (1966) Behavioural and ecological
investigations of the crab OQypode Qehatophtha£m~
(Crustacea, Ocypodidae), J. Zool. Lond. 150, 129-
143.
Hyatt GW (1977) Field studies of size-depend-
ent changes in waving display and other behaviour
in the fiddler crab, UQa pug~atoh (Brachyura,
Ocypodidae), Mar. Behav. Physiol. 4, 283-292.
Jacobsen MK (1955) Observations on Vonax
60~~Oh Say at Rockaway Beach, N.Y. Nautilus 68,
31-32.
Janssen GM and Kuipers BR (1981) Tidal migrat-
ions in the shrimp ChaVlgcn cAangon, Nethl. J. Sea
Res. 14, 339-348.
Jansson B-O (1962) Salinity resistance and
salinity preference of teo oligochaetes Akte~~
mono~pehm~Q~ Knoller and M~o~na pheQ~e££O­
QhaQta n.sp. from the interstitial fauna of marine
sandy beaches, Oikos 13, 293-305.
Jansson B-O (1966) Microdistribution of
factors and fauna in a marine sandy beach, Veroff.
Inst. Meeresforsch. Bremerhaven Sonderbd. 3, 77-
86.
Jansson B-O (1967) The importance of tolerance
and preference experiments for the interpretation
of mesopsammon field distributions, Helgol. wiss.
Meeresunters. 15, 41-58.
Jansson B-O (1968) Quantitative and experi-
mental studies of the interstitial fauna in four
Swedish sandy beaches, Ophelia, 5, 1-71.
Johnson RG (1965) Temperature variation in
the infaunal environment of a sand flat, Limnol.
Oceanogr. 10, 114-120.
Jones DA (1969) The genus Euny~Qe (Crustacea:
Isopoda) in the Aegean Sea, including E. £ong~­
~p~na sp. nov., Cah. Biol. mar. 10, 15-29.
Jones DA (1970) Factors affecting the distri-
bution of the intertidal isopods Euny~Qe puiQhha
Leach and E. a66~~ Hansen in Britain, J. animo
Ecol. 39, 455-472.
Jones DA (1972) Aspects of the ecology and
behaviour of OQypode Qehatophtha£mu~ (Pallas) and
O. ku~ de Haan (Crustacea: Ocypodidae), J. expo
mar. Biol. Ecol. 8, 31-43.
Jones DA and Naylor E (1970) The swimming
rhythm of the sand beach isopod Euny~Qe puiQhha,
J. expo mar. Biol. Ecol. 4, 188-199.
Jones JD (1963) The functions of the respira-
tory pigments in invertebrates, Probl. Biol. 1,
11-89.
Kensley B (1974) Aspects of the biology and
ecology of the genus Ty£o~ Latreille, Ann. S. Afr.
Mus. 65, 401-471.
Kinne 0 (1970) Temperature - Invertebrates.
In Kinne 0, ed. Marine ecology, 1 (1): environ-
mental factors, pp. 407-514. London, Wiley Inter-
science.
Kinne 0 (1971) Salinity - Invertebrates. In
602
Kinne 0, ed. Marine ecology 1 (2): environmental
factors, pp. 821-995. London, Wiley Interscience.
Klapow LA (1972) Fortnightly molting and
reproductive cycles in the sand beach isopod
Ex~olana ehLttoni, Biol. Bull. 143,568-591.
Koop K and Field JG (1981) Energy transform-
ation by the supra-littoral isopod Lig~ dLtatata
Brandt, J. expo mar. Biol. Ecol. 53, 221-233.
Koop K and Griffiths CL (1982) The relative
significance of bacteria, meio- and macrofauna on
an exposed sandy beach, ~1a r. Bi 01. 66, 295-300.
Koop K, Newell RC and Lucas MI (1982) Micro-
bial regeneration of nutrients from the decompo-
sition of macrophyte debris on the shore, Mar.
Ecol. Prog. Ser. 9, 91-96.
Kornicker LS (1961) Observations on the behav-
iour of the littoral gastropod T~ebna ~alleana,
Ecology 42, 207.
Lasserre P (1971) Donnees ecophysiologique sur
la repartition des oligochaetes marins meiobenth-
iques. Incidence des parametres salinite-temper-
ature sur le metabolisme respiratoire de deux
especes euryhalines du genre Makionina Michaelson
1889 (Enchytraedae, Oligochaeta), Vie Milieu 22,
523-540.
Lee JJ, Tietjen JH, Mastropaolo C and Rubin H
(1977) Food quality and the heterogeneous spatial
distribution of meiofauna, Helgol. wiss. Meere-
sunters. 3D, 272-282.
Lisenmair KE (1967) Konstruktion und Signal-
funktion der Sandpyramide der Reiterkrabbe Oeypode
~onatan Forsk. (Decapoda, Brachyura: Ocypodidae),
Z. Tierpsychol. 24, 403-456.
Locy S (1977) Some preference and tolerance
experiments using Makionina ~ub~~anea, a common
interstitial oligochaete, from the Pacific coast,
Proc. 3rd into Meiofauna Conf. Hamburg (Abstract).
MacGinitie GE (1938) Movements and mating of
the sand crab (Ern~ analoga) , Amer. Midl. Natur.
19, 471-481.
MacGinitie GE and MacGinitie N (1968) Natural
history of marine animals. New York, McGraw-Hill.
Magnus D (1960) Zur Okologie des Landeinsid-
lers Coenob~ jo~~eaurnei Bouvier und der Krabe
Oeypode aegyp~ea Gerstacker am Roten Meer, Ver.
deutsch. Zool. Ges. 316-329.
Mane UH and Talikhedkar PM (1976) Respiration
of the wedge clam Vonax eun~, Indian J. mar.
Sci. 5, 243-246.
Mangum CP and Towle D (1977) Physiological
adaptation to unstable environments, Amer. Scient.
65, 67-75.
Marsden ID, Newell RC and Ahsanullah M (1973)
The effect of starvation on the metabolism of the
shore crab CaAei~'rnaen~, Compo Biochem. Physiol.
45A, 195-213.
Marsh BA and Branch GM (1979) Circadian and
circatidal rhythms of oxygen consumption in the
sand-beach isopod Tylo~ gnan~~ Krauss, J.
expo mar. Biol. Ecol. 37, 77-89.
McGwynne LE (1980) A comparative ecophysiolog-
ical study of three sandy-beach gastropods in the
Eastern Cape. M.Sc. thesis, University of Port
El i zabeth.
Marshall SM and Orr AP (1964) Feeding and nutrit-
ion. In Waterman TH, ed. The Physiology of Crus-
tacea I: metabolism and growth, pp. 227-258. New
York, Academic Press.
~1cIntyre AD (1969) Ecology of marine meioben-
thos, Biol. Rev. 44, 245-290.
McIntyre AD, Munro ALS and Steele JH (1970)
Energy flow in a sand ecosystem. In Steele JH,
ed. Marine food chains, pp. 19-31.
McKillup SC and Butler AJ (1979) Modification
of egg production and packaging in response to
food a va i 1ab i 1i ty by M~M)UU~ paup~~, Deco 1-
ogia 43,221-231.
McLachlan A and Young N (1982) Effects of low
temperature on the burrowing rates of four sandy
beach molluscs, J. expo mar. Biol. Ecol. 65, 275-
284.
McLachlan A, Erasmus T and Furstenbero JP
(1977) Migrations of sandy beach meiofa~na, Zool.
afro 12, 11-89.
McLachlan A, Dye AH and Van der Ryst P (1979a)
Vertical gradients in the fauna and oxidation of
two exposed sandy beaches, S. Afr. J. Zool. 14,
43-47.
McLachlan A, Wooldridge T and Van der Horst G
(1979b) Tidal movements of the macrofauna on an
exposed sandy beach in South Africa, J. Zool. Lond.
188, 433-442.
MCLachlan A, Dye AH and Harty B (1981) simulat-
ion of the interstitial system of exposed sandy
beaches, Estuar. cstl shelf Sci. 12, 267-278.
McLachlan A (this volume) Sand beach ecology:
review paper. In Sandy beaches as ecosystems.
The Hague, Junk.
McLusky DW and Stirling A (1975) The oxygen
consumption and feeding of Vonax ineaAn~ and
V. ~pieulurn from tropical beaches, Compo Biochem.
Physiol. 51A, 942-947.
Meadows PS (1964) Experiments on substrate
selection by Co~ophiurn species: films and bacteria
on sand particles, J. expo Biol. 41, 499-511.
Meineke T and Westheide W (1979) Tide depend-
ent migrations of the interstitial fauna in a sand
beach of the Island of Sylt, North Sea, West Germ-
any, Mikrofauna Meeresbodens 0 (75), 203-236.
Morgan E (1965) The activity rhythm of the
amphipod Co~ophiurn vol~o~ (Pallas) and its
possible relationship to changes in hydrostatic
pressure associated with the tides, J. animo Ecol.
34, 731-746.
Mori S (1938) Vonax ~ernig~ano~~ Dkr. and the
experimental analysis of its behaviour at the
flood tide, Zool. Mag. Tokyo 50, 1-12.
Mori S (1950) Characteristic tidal rhythmic
migration of a mussel, Vonax ~ernig~ano~~ Dkr.
and the experimental analysis of its behaviour,
Zool. Mag. Tokyo 59, 88-89.
Mortensen TH (1921) Studies on the develop-
ment and larval forms of echinoderms. Gad.
Ki1lbenhavn.
Muir DG (1977) The biology of Talo~ehe~~
eape~~ ~Amphipoda, Talitridae). M.Sc. thesis,
University of Cape Town.
Nair NB and Ansell AD (1968) Characteristics

of penetration of the substratum by some marine
bivalve molluscs, Proc. malac. Soc. Lond. 38,
179-197.
Naylor E (1958) Tidal and diurnal rhythms of
locomotory activity in C~~nUh ma~n~ (L.), J.
expo Biol. 35, 602-610.
Naylor E (1962) Seasonal changes in a populat-
ion of C~~n~ ma~n~ (L) in the littoral zone,
J. animo Ecol. 31, 601-609.
Naylor E (1963) Temperature relationships of
the locomotor rhythm of C~~n~. J. expo Biol.
40, 669-679.
Naylor E, Atkinson RJA and Williams BG (1971)
External factors influencing the tidal rhythm of
shore crabs, Proc. 2nd interdisc. Conf. Cycle Res.
2, 173-180.
Newell RC (1979) Biology of interidal animals,
3rd ed. Haversham, Kent, Marine Ecological Surveys.
Newell RC and Branch GM (1980) The influence
of temperature on the maintenance of metabolic
energy balance in marine invertebrates, Adv. mar.
Biol. 17,329-396.
Newell RC, Ahsanullah M and Pye VI (1972) Aerial
and aquatic repiration in the shore crab C~~n~
ma~n~ (L.), Compo Biochem. Physiol. 43A, 239-252.
Orr PR (1955) Heat death I. Time-temperature
relationships in marine animals, Physiol. Zool.
28, 290-294.
Papi F (1955) Experiments on the sense of time
in T~Uh ~attato~, Experimentia 11, 201.
Papi F and Pardi L (1953) Ricerch sull'orient-
amento di T~Uh ~attato~ (Montagu) (Crustacea,
Amphipoda) II. Sui fattori chez regolano le variaz-
ione deu'angolo di orientamento nel corso del
giorne. L'orientamento diurno di altre popolazione,
Z. vergl. Physiol. 35, 490-518.
Pardi L (1954) Uber die Orientierung von Tylo~
latncittei (Aud. &Sav.) (Isopoda, terrestrial, Z.
Tier-psychol. 11, 175-181.
Pardi L (1955) L'orientamento diurno di Tylo~
iatncittei (Aud. and Sav.), Boll. Inst. Mus. Univ.
Torino 4, 167-196.
Pardi L (1960) Innate components of the solar
orientation of littoral amphipods, Cold Spring
Harb. Symp. quant. Biol. 25, 395-401.
Pardi L and Grassi M (1955) Experimental modi-
fication of direction finding in T~Uh ~attato~
(Montagu) and Talo~Qh~~ d~hay~ei (Aud.)(Crus-
tacea, Amphipoda), Experimentia 11, 202-203.
Pardi Land Papi F (1952) Die Sonne als Kom-
pass bei T~Uh ~attato~ (Montagu) (Amphipoda,
Talitridae), Naturwiss. 39, 262-263.
Pardi Land Papi F (1953) Richerche sull'orient-
amento di T~Uh ~attato~ (Montagu) (Crustacea,
Amphipoda) I. L'orientamento durante il giorno in
una popolazione del litorale tirrenico, Z. vergl.
Physiol. 35, 459-489.
Pardi Land Papi F (1961) Kinetic and tactic
responses. In Waterman TH, ed. The Physiology of
Crustacea, II, pp. 365-399. New York, Academic
Press.
Pearse AS, Humm HJ and Wharton GW (1942) The
ecology of sandy beaches at Beaufort, Ecol. Monogr.
12, 135-190.
Perry OM (1980) Factors influencing aggregation
603
patterns in the sand crab, Em~ analoga (Crust-
acea: Hippidae), Oecologia 45, 379-384.
Pianka ER (1976) Natural selection of optimal
reproductive tactics, Am. Zool. 16, 775-784.
Pora E and Bacescu M (1939) Sur la resistance
du Myside G~tho~aQQUh ~anQtu~ (van Beneden) de
la Mer Noire, aux variations de salinite du milieu
ambient, Ann. sci. Univ. Jassy 25,259-271.
Rao KP and Bullock TH (1954) Q10 as a function
of size and habitat temperature in poikilotherms,
Amer. Nat. 87, 33-43.
Rao SR and Kutty m,~ (1969) Resistance to
desiccation and oxygen debt in wedge clams, Proc.
Symp. Moll., Mar. Biol. Ass. India II, 595-606.
Reisch DJ (1969) Uptake of organic material
by aquatic invertebrates. V. Tije influence of
age on the uptake of glycine-C by the polychaete
N~anth~ ~~naQ~od~ntata, Mar. Biol. 3, 352-355.
Ride WDL (1962) In Bernier and Dorre Islands
2, pp. 1-131. W. Austr. Fish. Dept. Fauna Bull.
Robertson JR and Newell SY (1982) Experimental
studies of particle ingestion by the sand fiddler
crab UQa pugilato~ (Bosc), J. expo mar. Biol.
Ecol. 59, 1-21.
Robertson JR and Pfeiffer W (1982) Deposit-
feeding by the ghost crab OQypod~ qua~ata, J.
expo mar. Biol. Ecol. 56, 165-177.
Robertson JR, Bancroft K, Vermeer G and Plais-
ier K (1980) Experimental studies on the foraging
behaviour of the sand fiddler crab UQa pugilato~
(Bosc, 1802), J. expo mar. Biol. Ecol. 44, 67-83.
Robertson JR, Fudge J and Vermeer G (1981)
The feeding responses of the sand fiddler crab
UQa pugilato~ (Bosc) to live and chemical food
sources, J. expo mar. Biol. Ecol. 53, 47-64.
RUppell G (1967) Zur Lokomotionsaktivitat des
Amphipoden O~Qh~~ plat~n~ih im Freiland und im
Laboratorium, Helgol. wiss. Meeresunters. 15, 272-
280.
Salvat B (1967) La macrofaune carcinologique
endogee des sediments meubles intertideaux (tanai-
dacea, isopodes et amphipodes), ethologie, bionomie
et cycle biologique. Thesis, Fac. Sci. Paris.
Scheltema RS (1961) Metamorphosis of the
veliger larvae of N~~~Uh ob~ol~ (Gastropoda)
in response to bottom sediment, Biol. Bull. 120,
92-109.
Schone H and Schone H (1961) Eyestalk move-
ments induced by polarised light in the ghost crab
OQypod~ qua~ata, Science 134, 675-676.
Seymour MK (1971) Burrowing behaviour in the
European lugworm ~~n{Qola m~na (Polychaeta;
Arenicolidae), J. Zool. Lond. 164, 93-132.
Spaargaren DH (1980) The significance of sea-
water viscosity for the tidal transport of common
shrimps, C~angon ~angon, Ophelia 19, 145-153.
Stearns SC (1976) Life-history tactics: a
review of the ideas, Quart. Rev. Biol. 51, 3-47.
Stenton-Dozey J and Griffiths CL (1980) Growth,
consumption and respiration by larvae of the kelp-
fly FUQellia Qap~mih (Diptera: Anthomyiidae), S.
Afr. J. Zool. 15, 280-283.
Stephens GC (1967) Dissolved organic matter
as a nutritional source for marine and estuarine
invertebrates. In Lauff GH, ed. Estuaries, pp.
604
367-373. vJashington D.C., Amer. Ass. Adv. Sci.
Stephens GC (1968) Dissolved organic matter
as a potential source of nutrition for marine
organisms, Amer. Zool. 8, 95-106.
Stirts HM and Turner RL (1981) Effects of
salinity on survival and moulting of larval
Em~a tatpoida (Crustacea: Anomura), Compo Bio-
chem. Physiol. 69A, 125-128.
Stoll E (1937) Beobachtungen Uber die Fortbe-
wegung bei einigen grabenden muscheln, Rev. suisse
Zool Geneve 44, 383-390.
Swinbanks DO and Murray JW (1981) Biosediment-
ological zonation of Boundary Bay tidal flats,
Fraser River Delta, British Columbia, Canada,
Sedimentol. 28, 201-238.
Taylor EW and Butler PJ (1973) The behaviour
and physiological responses of the shore crab
c~ci~Uh mae~~ during changes in environmental
oxygen tension, Nethl. J. Sea Res. 7, 496-505.
Theede H (1973) Resistance adaptations of
marine invertebrates and fish to cold. In Wieser
W, ed. Effects of temperature on ectothermic organ-
isms, pp. 249-269. Heidelberg, Springer-Verlag.
Thompson RK and Pritchard A\>J (1969) Respiratory
adaptations of two burrowing crustaceans, Cattia-
~~a ca£i6on~e~~ and Upogebia pug~e~~
(Decapoda, Thalassinidae), Biol. Bull. 136, 274-
287.
Thorson G (1950) Reproduction and larval ecol-
ogy of marine bottom invertebrates, Biol. Rev. 25,
1-45.
Trueman ER (1966a) Observations on the burrow-
ing of Ane~cola mani~a (L), J. expo Biol. 44, 93-
118.
Trueman ER (1966b) The mechanism of burrowing
in the polychaete worm Ane~cola mani~a (L.),
Biol. Bull. 131,369-377.
Trueman ER (1968) The burrowing activities of
bivalves, Symp. zool. Soc. Lond. 22, 167.
Trueman ER (1970) The mechanism of burrowing
of the mole crab, Em~a, J. expo Biol. 53, 701-
710.
Trueman ER (1971) The control of burrowing
and the migratory behaviour of Vo~ax denticulatD4
(Bivalvia: Tellinacea), J. Zool. Lond. 165,435-
461.
Trueman ER (1975) The locomotion of soft-
bodied animals. London, Edward Arnold.
Trueman ER and Ansell AD (1969) The mechanisms
of burrowing into soft substrata by marine anim-
als, Oceanogr. mar. Biol. Ann. Rev. 7, 315-366.
Trueman ER, Brand AR and Davis P (1966) The
dynamics of burrowing of four common littoral
bivalves, J. expo Biol. 44,469-492.
Trueman ER and Brown AC (1976) Locomotion,
pedal retraction and extension, and the hydraulic
systems of B~a (Gastropoda: Nassaridae), J.
Zool. Lond. 178, 365-384.
Turner HJ and Belding AL (1957) The tidal
migrations of Vo~ax v~b~ Say, Limnol. Ocean-
ogr. 2, 120-124.
Tweedie MWF (1950) Notes on grapsoid crabs
from the Raffles Museum, Bull. Raffles Mus. 23,
310-324.
Vance RR (1973) On reproductive strategies in
marine benthic invertebrates, Am. Nat. 107,339-
352.
Vannini M (1980) Notes on the behaviour of
Ocypode nydeni (Crustacea, Brachyura), Mar. Behav.
Physiol. 7, 171-184.
Vernberg FJ (1959) Studies on the physiolog-
ical variation between tropical and temperate
zone fiddler crabs of the genus Uca, II. Oxygen
consumption of whole organisms, Biol. Bull. 117,
163-184.
Vern berg WB and Vernberg FJ (1972) Environment-
al physiology of marine animals. Berlin, Springer-
Verlag.
Von Hagen H (1967) Nachweis einer kinHstheti-
schen Orientierung bei Uca napax, Z. Morph. Okol.
Tiere 58, 301-320.
Von Oertzen JA (1973) Abiotic potency and
physiological resistance of shallow and deep water
bivalves, Oikos (suppl) 15,261-266.
Wade BA (1967) Studies on the biology of the
West Indian beach clam, Vo~ax denticulatu~ Linne,
I. Ecology, Bull. mar. Sci. Gulf Caribb. 17, 149-
174.
Watkin EE (1942) The macrofauna of the inter-
tidal sand of Kames Bay, Millport, Buteshire,
Trans. Roy. Soc. Edinb. 60, 543-561.
Wallace JC (1973) Feeding, starvation and
metabolic rate in the shore crab, C~ci~u~ mae~~,
Mar. Biol. 20, 277-281.
Wells GP (1944) Mechanism of burrowing in
Ane~cola mani~a (L.), Nature 154, 396.
Wells GP (1949a) Respiratory movements of
Ane~cola mani~a (L.). Intermittent irrigation
of the tube, and intermittent aerial respiration,
J. mar. biol. Ass. U.K. 28, 447-464.
Wells GP (1949b) The behaviour of Ane~cola
mani~a (L.) in sand, and the role of the spontan-
eous activity cycles, J. mar. biol. Ass. U.K. 28,
465-478.
Wells GP (1950) Spontaneous activity cycles
in polychaete worms, Symp. Soc. expo Biol. 4, 127-
142.
Wells GP (1954) The mechanism of proboscis
movement in Ane~cola, Quart. J. micro Sci. 95,
251-270.
Wells GP and Albrecht EB (1951a) The integrat-
ion of the activity cycles in the behaviour of
Ane~cola mani~a L., J. expo Biol. 28, 41-50.
Wells GP and Albrecht EB (1951b) The role of
the oesophageal rhythms in the behaviour of Ane~­
icola ecaudata Johnson, J. expo Biol. 28, 51-56.
Wieser W (1975) The meiofauna as a tool in
the study of habitat heterogeneity: ecophysiolog-
ical aspects, a review, Cahiers de Biol. Mar. 16,
647-670.
Wieser Wand Schiemer F (1977) The ecophysiol-
ogy of some marine nematodes of bermuda: seasonal
aspects, J. expo mar. Biol. Ecol. 26, 97-106.
Wieser W, Ott J, Schiemer F and Gnaiger E
(1974) An ecophysiological study of some meio-
fauna species inhabiting a sandy beach at Bermuda,
Mar. Biol. 26, 235-248.
Wildish OJ (1970) Locomotory activity rhythms

in some littoral On~heh~ (Crustacea: Amphipoda),
J. mar. biol. Ass. U.K. 50, 241-252.
Wilkens JL and Fingerman M (1965) Heat toler-
ance and temperature relationships of the fiddler
crab U~a pug~aton, with reference to body color-
ation, Biol. Bull. 128, 133-141,
Williams JA (1979) A semi-lunar rhythm of
locomotory activity and moult synchrony in the
sand-beach amphipod T~uo ~attaton. In Cyclic
phenomena in marine plants and animals, Proc. 13th
Eur. mar. Biol. Symp., pp. 407-414. New York,
Pergamon Press.
Williams JA (1980) The effect of dusk and
dawn on the locomotor activity rhythm of T~uo
~attatM U~ontagu), J. expo mar. Biol. Ecol. 42,
285-297.
Williams JA (1982) A circadian rhythm of oxy-
gen consumption in the sand beach amphipod T~uo
~attaton (Montagu), J. expo mar. Biol. Ecol. 57,
125-134.
Williams BG and Naylor E (1969) Synchronisation
of the locomotor tidal rhythm of Can~inuo, J. expo
Biol. 51, 715-725.
Williamson DI (1951a) Studies in the biology
of Talitridae: effects of atmospheric humidity,
J. mar. biol. Ass. U.K. 30, 73-90.
Williamson DI (1951b) Studies on the biology
of the Talitridae (Crustacea, Amphipoda). Visual
orientation in T~uo ~attaton, J. mar. biol.
Ass. U.K. 30, 91-99.
Williamson DI (1951c) On the mating and breed-
ing of some semi-terrestrial amphipods. Rep. Dove
605
mar. Lab. 1949, 49-62.
Williamson DI (1953) Landward and seaward
movement of the sand hopper, T~uo ~altaton.
In How animals find their way about, pp. 71-73.
Repts. Brit. Ass. adv. Sci. (Section D).
Wilson DP (1948) The relation of the substrat-
um to the metamorphosis of Oph~!ia larvae, J. mar.
biol. Ass. U.K. 27, 723-760.
Wilson DP (1952) The influence of the nature
of the substratum on the metamorphosis of larvae
of marine animals, especially the larvae of
Oph~a bi~on~ Savigny, Ann. Inst. Oceanogr.
Monaco 27, 49-156.
Wilson DP (1953a) The settlement of Oph~
bi~on~ Savigny larvae, the 1951 experiments, J.
mar. biol. Ass. U.K. 31, 413-438.
\~i 1son DP (1953b) The settl ement of Oph~
bi~on~ Savigny larvae, the 1952 experiments, J.
mar. biol. Ass. U.K. 32, 209-233.
Wilson DP (1954) The attractive factor in the
settlement of Oph~ bi~on~ Savigny, J. mar.
biol. Ass. U.K. 33,361-380.
Wilson DP (1955) The role of micro-organisms
in the settlement of Oph~ bi~on~ Savigny, J.
mar. biol. Ass. U.K. 34, 513-543.
Wolverkamp HP and Vreede MC (1941) On the gas
binding properties of the blood of the lugworm,
M~niw{'a m~na L., Arch ~!eerl. Physiol. 25, 265-
276.
Zebe E (1975) In vivo-Untersuchungen Uber den
Glucose-Abbau bei An~ni~o{a m~na (Annelida,
Polychaeta), J. compo Physiol. 101,133-147.
Zhirmunsky AV and Chu Li-chan (1963) The cell
thermostability of sympatric species of Vonax in
relation to the temperature conditions of their
habitat, Acta zool. Sinica 15, 21-27.

THE BIOLOGY OF THE GENUS DONAX
A.D. ANSELL (Dunstaffnage Marine Research Laboratory, P.O. Box No.3, Oban, Argyll, Scotland)
1. INTRODUCTION
The exposed sandy beaches of the world
represent for their fauna high energy environments
dominated by the physical processes of tides and
wave action. A few genera of molluscs, crustaceans
and polychaetes dominate such environments, each
adapted to exploit actively the apparently severe
conditions rather than passively survive their effects.
In so doing the successful genera derive benefits such
as refuge from predation, reduced competition, or
ready availability of food resources, which more than
offset the costs incurred in increased activity or in
the risks of stranding and desiccation. In
consequence they may reach very high population
densities in some areas.
Among molluscs on a world-wide basis, species
of Donax form by far the most dominant group in
such environments. Although restricted by their
specialisation to beaches and to some subtidal high
energy environments, adaptive radiation has ensured
their spread to occupy most available niches within
these constraints. In this paper, in reviewing the
published literature on this one genus, I shall examine
the results of such adaptive radiation in terms of the
geographical and habitat distribution of present day
species; I shall try to identify the major adaptive
trends which have led to the success of the group;
and I shall speculate on the balance of costs and
benefits involved in some of the adaptive solutions
they present. Such a review, though, must inevitably
reflect the gaps in our present knowledge, for several
important groups of species remain largely unstudied,
and there are large gaps in our detailed knowledge of
all species, in spite of the fairly broad geographical
607
interest in the group shown in the literature.
The genus Donax is included in the family
Donacidae within the superfamily Tellinacea, with
two other genera, Egeria and Iphigenia. Both these
genera are restricted in their distribution to tropical
and subtropical waters, and are typically inhabitants
of brackish lagoons or rivers (Purchon 1963; Kwei,
1966; Hiroki, 1971; Narchi, 1972, 1974; Yoloye, 1977).
The genus Hemidonax, previously included in the
family Donacidae, is more correctly placed in the
Cardiidae (Boss, 1971) or as a separate family in the
Cardiacea (Ponder et aI, 1981). These genera will not
be further discussed here.
The genus Donax itself contains some 64
species. As in other bivalve groups this number
represents only a small proportion of the total
number of species names assigned to the genus, and
taxonomic works abound. This is not the place for a
taxonomic review, however, and such works will not
be reviewed. Any taxonomic interpretations included
here are based on a current personal assessment of
which are valid species. Information on distribution
is based on personal examination of material in
European museums, supplemented only where such
material is limited, by reliance on published records,
but is only summarized here, and the numerous
works referring only to the occurrence of Donax
species will not be dealt with.
2. GEOGRAPHICAL DISTRIBUTION
Donax is predominantly tropical in its recent
distribution. Approximately 77% of living species
are found in tropical waters, 22% are mainly warm
temperate in their distribution, and only 5% extend
608

their distribution into cold temperate areas; there apparent as a major feature leading to diversity in
are no high latitude species in either the Arctic or the Indo-West-Pacific region. Distinctive groups of
Antarctic. endemic species characterise each of the fringe areas
The limits of distribution for individual species, of this region where the relatively uniform conditions
as far as they are known, conform closely with the of the tropical region gi ve way to subtropical
major faunal boundaries defined by Ekman (1953). conditions, or where other environmental gradients
They reflect the general importance of temperature occur. The areas of Japan and Indonesia, Madagascar
in determining distribution limits for intertidal and and West Africa, Australia, and the north Arabian
shallow subtidal marine invertebrates, and indirectly Sea, Red Sea and Persian Gulf, each support
the importance of the impingement of major ocean distinctive endemic tropical or subtropical species
currents in determining marine climate zones along which share the coastline with, or replace, widely
major continental coastlines. All individual distributed Indo-West-Pacific species. In these
distributions of Donax species are confined within instances speciation appears to have proceeded by
distinct zoogeographical regions and there are no the successive separation of populations at the
examples of more cosmopolitan distributions. boundaries of very widespread distributions, and the
The numbers of species decline with increasing relationships of the locally distributed endemic
latitude along each major north-south continental species with co-existing Indo - Pacific species of
coastline. Generally only one species to the north widespread distribution are clearly apparent.
and one to the south on each coastline extends in
distribution into the cold temperate region, although 3. HABIT A T DIVERSITY
in some areas where such extensions occur, for Where more than one species of Donax occurs
example, in Europe and on the Atlantic coast of N. within an area there is generally little habitat
America, there may be a progression of closely- overlap between species. Three main factors are
related forms which share the range from warmer to important in the separation of species in such
colder conditions. In such cases speciation along an instances; bathymetric depth, wave energy and
environmental temperature gradient appears to have sediment composition, although these are also clearly
been important in extending the range of the genus interrelated. Other factors such as tolerance of
into cooler waters, but there is still some temperature extremes and air exposure may also be
disagreement regarding the status of some of the involved. Typically, geographically-sympatric
species which have been distinguished in such species of Donax occupy three distinct habitats; the
situations. intertidal zone of moderately-exposed or exposed
Between different regions there are significant beaches with clean well-sorted sand, the shallow
differences in the numbers of species found. This is sublittoral zone of such beaches, and an offshore
clear especially within the tropics; the greatest zone where coarse sediments occur under the
number of species are found in the areas of the influence of strong currents. Further habi tat
Philippine-Indonesian Archipelago and on the Pacific diversity is provided by differences between beaches
coast of the Americas, while the areas of lowest in shelter and wave energy and in sediment
species diversity are the coasts of W. Africa and the characteristics, and on coasts subject to continuous
Caribbean. Some other groups of predominantly surf action, by dynamic conditions which provide a
tropical distribution show a similar phenomenon separate niche for species which migrate tidally
(Ekman, 1953; Crane, 1975). through all or part of the tidal cycle, rather than
The trend toward speciation along remaining distributed in a constant zone. Tidal
environmental gradients seen elsewhere is also migration has been reported, or may be inferred to
 609

occur, in many Donax species from most tropical and 4.1 Predation
some warm temperate regions, and is a major feature Donax are often the main primary consumers in
of the adaptive radiation of the genus. such communities, sharing this role with other
bivalves, and in many cases with the mole crab
4. FAUNAL ASSOCIA nONS AND ECOLOGY Emerita. In turn, Donax are subject to predation by a
Donax species are the dominant bivalves on wide variety of predators, including other
many tropical and subtropical beaches but are invertebrates, fish, birds and mammals.
progressively replaced by other genera on beaches The main invertebrate predators are crabs and
which are more sheltered and composed of finer gastropo~s. Most Donax populations are preyed upon
particles, on beaches consisting of sands of biotic by portunid crabs (Loesch, 1957; Wade, 1967; Mass~,
rather than terrigenous origin, and on beaches at 1963; Moueza, 1975; McLachlan et ai, 1979a). The
higher latitudes. Donax species are particularly portunid Ovalipes punctatus preys on Donax serra and
characteristic of highly productive areas, only D. sordidus on Eastern Cape beaches of South Africa,
reaching their greatest population densities where and Du Preez (1981) has made a detailed study of
eutrophic conditions lead to high levels of predation by this species on Donax and on the
phytoplankton production. In some species, major gastropod Bullia on the same beaches. In choice
populations may then be located near the mouths of experiments, Ovalipes from the two beaches showed
major rivers (Wade, 1967); while in others the distinctive preferences for different molluscs,
development of surf-blooms (McLachlan, Lewin, apparently selecting those species to which it had
1981; McLachlan, 1980) may be an important factor least access in the field.
in the maintenance of large standing stocks Portunid crabs are found mainly in the
(McLachlan et ai, 1981). sublittoral and may attack Donax, often nocturnally,
Donax species characterise certain zones within when the bivalve population is exposed or submerged.
well-defined beach communities, and brief reviews In contrast, ghost crabs of the genus Ocypode, which
discussing parallels between such faunas in different are also important predators in tropical areas, live in
areas have been given by Dahl (1953), Gauld, burrows near or above high water mark and consume
Buchanan (1956), Hedgepeth (1957) Pichon (1967) Donax which they dig from exposed sand at low
Trevallion et al (1971) and Olivier et al (1971). Some water, often also exhibiting nocturnal behaviour.
attempts have been made to summarise the complete Loesch (1957), Hughes (1966), Wade (1967), Ansell et
trophic relationships of the fauna on such beaches. al (1972a), and Wolcott (1978) report predation by
Koepke and Koepke (1952) illustrate a food web for Ocypode species. Smith (1975) who studied predation
Peruvian beaches, but give only qualitative by Ocypode ceratophthalma in East Africa found that
information on the interrelationships of all species this species selected the different colour morphs of
(see also Hedgepeth, 1957). Brown (1964) provides D. faba differentially under certain conditions.
similar information for South African beaches near Selection was found to be stabilizing at high densities
Cape Town, and Ansell et al (1978) provide a partial and apostatic at low densities. Many Donax species
food web for Indian sandy beaches. The most show a similar colour polymorphism to that of D.
complete such study, with quantitative estimation of faba (Nolte, 1954, 1958; Moment, 1962; MacNae,
energy flows between all the main components, has Kalk, 1962; Wade, 1968; Smith, 1971), as do other
been made on Eastern Cape beaches of South Africa, sandy beach invertebrates with similar habits,
summarized by McLachlan et al (1981). notably the gastropod Umbonium on Indo-Pacific
beaches. There seems no doubt that colour
polymorphism is adaptive. In the case of D.
610

variabilis, Moment (1962) considered that colour McLachlan et ai, 1980) among others (Loesch, 1957;
polymorphism is important in preventing the major Penchaszadeh, Olivier, 1975). The sanderling
predators (birds) forming searching images (Crocethina alba) has been observed feeding on Donax
(Tinbergen, 1960). Smith (1975) suggests that colour on Texas beaches (Loesch, 1957), in the West Indies
polymorphism provides crypsis for juvenile Donax, (Wade, 1967), in East Africa (Smith, 1975), and in
and in adults provides visual predators (including South Africa (McLachlan et ai, 1980), in the latter
Ocypode) with intraspecific choice as an alternative area taking only the siphon tips of the large D. serra.
to interspecific options, and concludes that colour In common with other bivalves which are subject to
polymorphism is especially important in tidal migrant such siphon predation, D. serra is quickly able to
species which are particularly vulnerable to visual regenerate the tissues lost in this way, including the
predators during migration. complex fringing tentacles (Hodgson, 1982a,b).
Shells drilled by gastropod predators are Perhaps the most unusual predator is found
common and the relative frequencies of such bored among the mammals. Wade (1967) reports that in the
shells in collections have been used to estimate the West Indies the pig is a frequent predator on D.
contribution naticid and other borers make to denticulatus. "Walking along the beach, it digs its
mortality of Donax populations (George, 1965; Masse, snout into the sand and scoops out hundreds of clams
1963, 1971, 1972a,b; Negus, 1975). Experimental as it goes along. Periodically it stops to crush and
studies by Lever et al (1961, 1964; Lever, Thijssen, swallow the clams which have been scooped up, and
1968), however, show that such estimates must be then moves on again. The pigs' appetite for Donax is
treated with caution since both dead shells and immense, and it seems that one pig in a day may
drilled valves may be differentially sorted on devour more than what all other predators consume
beaches. Other gastropod predators include Terebra in a month".
(Wade, 1967), OIivancillaria (March-Marchaud, Donax are also frequently subject to human
personal communication, Penchaszadeh, Olivier, predation, in some areas on a commercial basis, but
1975) and Bullia (Trevallion et ai, 1971; J. Taylor this aspect of predation will be dealt with in a later
personal communication). Echinoderm predators are section.
relatively uncommon in Donax communities, but may
occur. Phillips (1970) found that of 10 individuals of 4.2 Parasitism
Luidia foliata collected intertidally, 9 contained D. Donax species may be infested by a variety of
navicula, the total of Donax ingested being 51. parasites, including a species of Rickettsia (Camps,
Fish seem always to be important predators on Raimbault, 1978), and gregarine sporozoans and
Donax populations. Some species ingest the whole nematodes (Moueza, 1975).
animal, Jater ejecting the shell whole or crushed, In most areas, trematode larvae are by far the
while others feed by cropping siphons. Teleosts, rays most important parasites and references to such
and sharks all contribute to such predation in many infestations are numerous, although few of the
areas (Koepke, Koepke, 1952; Coe, 1957; Loesch, species have been studied in detail. Hopkins (1958)
1957; Moueza, 1975; Penchaszadeh, Olivier, 1975; reviewed the earlier literature, and Wade (1967) also
Armitage, Alevizon, 1980; McLachlan et ai, 1981). lists some earlier references. Many of the trematode
Birds also prey on Donax during the period when parasites found are found in the visceral mass, and
the bivalves are exposed on beaches or when just cause parasitic castration in the infested bivalve
covered by the tide. Bird predators include (Pelseneer, 1896; Loesch, 1957; Hopkins, 1958;
especially gulls (Cae; 1957; McLachlan et ai, 1979b) Lucas, 1965; Wade, 1967, Moueza, 1975; Johnson,
and oyster catchers (Penchaszadeh, Olivier, 1975; 1968). The main hosts of the trematodes from Donax
 611

are fish and birds. Sculpture is generally restricted to patterns of

Some authors have sought to associate the concentric and radiating ridges especially on the

occasional mass mortalities of certain Donax species posterior faces of the shell, but a few species also

with parasitic infestations (Coe, 1956; Beeson, develop stronger shell sculpture and in extreme cases
Johnson, 1967; Johnson, 1968) but no clear strong spines (D. scortum). Oblique sculpture is also
association with a particular parasite has been a feature of the distinctive shell patterns of many
established. species, and in some cases this may be developed as
strong shell ridges (D. madagascariensis).
4.3 Epibionts Partly aided by the streamlined shell shape,
The shells of living Donax are used as a Donax species are all rapid and efficient burrowers,
substratum for attachment by a variety of organisms. achieving complete burial using only a relatively few
The hydroid Clytia is commonly found on D. gouldi digging cycles (Trueman et aI, 1966; Nair, Ansell,
and D. variabilis (Coe, 1957; Loesch, 1957; Johnson, 1968; Ansell, Trevallion, 1969; Trueman, Ansell,
1966b) and similar hydroids are found on several 1969; Trueman, 1968, 1971). The time from initial
other Donax species. Green algae, such as penetration to complete burial may be very short in
Enteromorpha, and occasionally brown algal tropical species like D. denticulatus (5-6 secs;
sporelings, may also attach near the posterior Trueman, 1971) or D. semigranosus 0-6 secs; Mori,
margin, sometimes growing to considerable size and 1938) but is proportionally longer in species living at
constituting a hazard to the stability of the Donax in lower environmental temperatures (Ansell,

the sand (McKay, 1972). In the Mediterranean, the Trevallion, 1969; McLachlan, Young, 1982).

small nassarid gastropod Cyclonassa attaches its egg
capsules to the posterior margins of the shells of
Donax trunculus (H. Masse, personal communication).
5. BEHAVIOURAL AND SENSORY ADAPTATION
5.1 Burrowing and Recovery
The most significant common characteristic of
all habitats occupied by Donax species is that they
are high energy environments with strong wave
action or high current speed causing
disturbance of the sediment. Donax species are well-
adapted to maintain their position in such conditions.
The shell shape is adapted to provide easy
penetration of the sediment. Individual shell shapes
in the genus vary from the extreme posteriorly-
Donax species also make powerful recovery
movements to regain the surface when accidentally
buried too deeply by sand movements and can move
over the surface of the sand by leaping (Stoll, 1937,
1947; Ansell, 1969; Ansell, Trevallion, 1969;
Trueman, 1971).
A major factor in the success of Donax in high
energy environments is their ability to co-ordinate
these movements to maintain their normal position in
frequent the sediment in the face of wave disturbance, and to
respond to the physical changes in the environment
caused by wave and current action.
In some species, notably those from higher
latitudes and more sheltered beaches, these abilities
are expressed only in the integration of burrowing

truncated wedge shape which gives the group one of and recovery movements in response to disturbances
its common names (wedge shells), to a more as the animals are covered or uncovered by the tide
equilateral shape with both posterior and anterior (Trueman, 1967). In others they lead to
margins extended and rounded. Extreme wedge characteristic changes in zonation on the beach
shapes are characteristic of intertidal, or migrating, which may follow a semilunar tidal rhythm
species while the more streamlined equilateral shape (McLachlan et aI, 1979) or be related to changes in
is more characteristic of subtidal forms, especially wave energy resulting from periods of storms or calm
those occupying offshore coarse sands and gravels. (Leber, 1982). In their most fully-developed form,
612

however, these responses are efficiently co-ordinated beaches, and between species, can be adequately
to form the basis for tidal migration. interpreted within a common framework of
interaction between sensorally-mediated behavioural
5.2 Tidal migration responses, and cyclically changing physical
Tidal migration was first reported for the conditions.
Japanese species D. semigranosus by Mori (1938, The benefits the animals derive from tidal
1950) who described the typical sequence of migration remain the subject of speculation. Tidal
emergence, transport by the wave, and reburial, migration may serve to optimise feeding, to minimize
which results in the population moving landward on exposure to environmental stress, or to minimise
the flood tide, and seaward on the ebb. Mori predation pressure. None of these alternatives can
attempted to analyse this behaviour by means of be adequately tested, however, or their advantages
simple in situ experiments and concluded that the expressed quantitatively.
changes in behaviour between tide states might be Against these possible advantages may be set
partly accounted for by differences in the the disturbance to feeding associated with wave
environment, but also suggested that they might be action in this zone and the energy costs of migration.
regulated by some intrinsic physiological rhythm. The former are offset by the rapid reopening
Tidal movements have since been recorded for responses shown by the siphons (Ansell, Trevallion,
Donax species from most tropical and subtropical 1969) together with the elaborate systems of
surf beaches; in California (Hedgepeth, 1957; straining tentacles which prevent the entry of sand
Johnson, 1966b), the Caribbean (Wade, 1965, 1967; grains into the inhalant siphon (Wade, 1967, 1969;
Trueman, 1971), the Gulf of Mexico (Hedgepeth, Narchi, 1978; Ansell, 1981) ensuring that interruption
1953; Loesch, 1957; Tiffany, 1971), the Atlantic of the feeding current is minimal. Ansell and
coasts of N. America (Pearse et ai, 1942; Jacobsen, Trueman (1973) attempted to estimate the energy
1955; Turner, Belding, 1957; Morrison, 1971), and S. costs of the locomotory movements involved in
America (Penchaszadeh, Olivier, 1975), Madagascar maintaining migration for D. incarnatus and D.
(Pichon, 1967; Roman, 1974), the Seychelles (Taylor, denticulatus, and concluded that, together, the
1968), East Africa (Smith, 1975), South Africa maintenance of position and migration might add
(McLachlan et ai, 1979), West Africa (Smith, 1971) about one third to the daily maintenance
and India (Ansell, Trevallion, 1969). requirements, with a 2cm Donax requiring 6 cal/day
In each case the principal components of for this activity. However, the assumption of a value
migration behaviour are the same, although there are of 20% 'efficiency' in these calculations has been
differences both in the various authors' descriptions criticised by Brown (1969) on the basis that it does
and in their interpretations of the events involved. In not take into account both mechanical efficiency and
addition to Mori (1938, 1950), Turner and Belding metabolic efficiency. Brown showed that in the
(1957), Ansell and Trevallion (1969) and Trueman sandy beach gastropod Bullia digitalis overall
(1971) have all attempted an experimental approach efficiency of the burrowing process was only 6%.
to analysis of this behaviour, and Tiffany (1971) Brown (1982) has also shown that the activity of
transplanted populations of O. variabilis and observed surfing in Bullia, an activity similar in many respects
that they followed the local pattern of migration. to transport in the surf during migration of Donax,
Most authors now agree that no intrinsic mechanism consumes oxygen at approximately twice the rate
is required for the control of migration in Donax shown by resting animals. If similar considerations
species, and that the observed variations in migratory apply to Donax locomotion, the actual costs incurred
behaviour within the same species on different in migration may be much greater than has been
 613

thought. It is likely, however, that the methods used 1979). D. serra populations on Eastern Cape beaches
to measure rates of oxygen consumption for some show a semilunar rhythm of movement, and the
tidal migrant species of Donax (Ansell et aI, 1978; observations of Irwin (1973) suggest a similar pattern
McLusky, Stirling, 1975; Dye, 1979) actually measure of movement may occur for D. gouldi on California
an active rate, so that part of the cost of migration beaches. Donax species, together with other sandy
may already be discounted in such measurements. beach invertebrates, also often show a patchy long-
shore distribution which may affect different year
5.3 Vertical Distribution groups differentially (Loesch, 1957; Degiovanni,
Not all Donax species migrate tidally, and not Moueza, 1972; Moueza, Chessel, 1976; Bally, 1981).
all tidal migrant species migrate through the whole All these examples illustrate the extreme flexibility
tidal cycle, at all times of the year, or at all times shown by Donax species in maintaining dynamic
during their lifespan. Differing interactions between distribution patterns in response to changing physical
the physical characteristics of different beaches, and conditions.
the behavioural adaptations of the animals, result in The behavioural bases for the significant
Donax species displaying a variety of distribution distributional differences which occur where two or
patterns. Some remain at a characteristic level of more species occupy the same beach remain
the beach with little variation, such as D. vittatus on unknown. Differences in burrowing abilities,
European beaches (Ansell, 1972; Guillou, Le Moal, especially in response times, may be a factor in
1978; Ansell, Lagardere, 1980), and D. faba on Indo- determining the levels on the beach at which
Pacific beaches (Pichon, 1967; Jones, 1979). Species individual species can maintain themselves, but other
which normally migrate may at times become factors, including differences in environmental
'stranded' during low water periods at different levels tolerances may also be involved. In some cases, like
of the beach, a phenomenon first noted by Mori that of D. vi ttatus and D. trunculus on European
(1938) for D. semigranosus, but also seen in D. Atlantic beaches, where the newly-recruited
variabilis (Aldrich, 1959; Edgren, 1959; Mikkelsen, juveniles of the two species occur at widely
1981; Leber, 1982), or they may move offshore to separated levels of the beach (Ansell, Lagardere,
subtidal habitats, as in D. variabilis in winter, and D. 1980) there is a strong presumption of differences in
parvula in summer (Leber, 1982). Juveniles may settlement behaviour, but these have not been
migrate actively, while adults remain at a fixed tidal demonstrated.
level as observed for D. faba in Madagascar (Roman,
1974). Other species show a differential distribution 5.4 Sensory control of movements
of different size classes on the beach, with the older, Trueman (1971) has shown that D. denticulatus
larger, individuals separated from the juveniles. Such reacts in the laboratory to acoustic stimulation by
size sorting at different levels has been recorded for activity, confirming the conclusions reached from
D. trunculus on European Atlantic beaches experiments carried out in the field with several
(Amouroux, 1972; Guillou, Le Moal, 1978; Ansell, species (Mori, 1938, 1950; Turner, Belding, 1957;
LagardBre, 1980) although not in the Mediterranean Loesch, 1957). The relatively high sensitivity to
(Moueza, 1972; Geldiay, Uysal, 1972); for D. faba on external stimuli shown by Donax species is associated
Indian beaches (Alagarswami, 1966); for D. with morphological adaptation of some sensory
denticulatus in the West Indies (Wade, 1967); and for structures. In many species, the middle fold of the
D. serra on beaches on the west coast of South Africa mantle edge, which has a sensory function (Yonge,
(De Villiers, 1975b; Bally, 1981) and on Eastern Cape 1957; Frenkiel, 1980) is greatly developed; it is
beaches of South Africa (McLachlan, Hanekom, duplicated in some species (Frenkiel, 1982) and in its
614

most extreme development bears groups of primary, in L T50 and BT 50 values of only 2 to 3 DC, D.
secondary and tertiary tentacles with varying semistriatus having the greater thermal tolerance
complexities of pinnate branching (Ansell, 1981). In consistent with its more southerly distribution.
common with other Tellinaceans, Donax species also Acclimation to higher temperatures increases
have a special sense organ, the sense organ of the thermal tolerance, and in all cases the burrowing
cruciform muscle (Von Ihering, 1900; Graham, response is more sensitive to changes in acclimation
1934a,b; Yonge, 1949; Frenkiel, 1982) which contains temperature.
two types of sensory cells, one having sensitivity to The two South African species each have
vibration, the other to mechanical stimulation similar thermal tolerances, although juveniles of D.
(Moueza, Frenkiel, 1976b; Frenkiel, Moueza, 1977; serra show a slightly lower tolerance. The thermal
Pichon et aI, 1978; Pichon et aI, 1980; Frenkiel, tolerance of these two southern warm temperate
1982). Unlike the situation in most Tellinaceans forms compares most closely with the two SUbtidal
where the sensory capsule of this organ opens into European species, D. vittatus and D. semistriatus.
the siphonal cavity, in Donax the capsule is closed, There are no comparable data on thermal
giving the possibility of greater sensitivity, and it is tolerance available for Donax from other regions,
tempting to speculate that this additional sensitivity including the tropics, but Zhirmunsky and Chu Li-
is important in the development of the more complex chun (1963) have measured the thermostability of the
co-ordinated movements shown by tidal migrant ciliated epithelial cells and of the foot muscles of
species. Although a complete understanding of the three species of Donax from the South China Sea, D.
co-ordination of these movements remains elusive, dysoni, D. cuneatus and D. semigranosus. The
Frenkiel's (1982) observations show clearly that it is temperature causing thermonarcosis in 1 min (t 1 ) or
the cruciform muscle sense organ which is 10 min (tID) was used to characterise cell
responsible for the perception of the acoustic and thermostability (Zhirmunsky, 1963). D. dysoni and
mechanical stimuli described by Trueman (1971) from D. cuneatus, living higher in the intertidal zone,
the laboratory, and by others from the field. showed higher tl and tID values than D. semigranosus
which is a tidal migrant in the wash zone (Mori,
6. ENVIRONMENT AL TOLERANCE 1938).
6.1 Temperature Extrapolation from the cell thermostability
Bodoy and Masse (1977) and Ansell et aI, curve would indicate 24 h L T 50 values for these
(198ob) have determined the upper temperature tropical species of 38 - 40 DC compared with values
tolerance limits, expressed as median lethal of 29 - 31 DC found for D. vittatus or D. semistriatus,
temperatures (L T50) or median burial temperatures of 33 DC for D. trunculus and 31-33 DC for D. serra
(BT 50)' for the European species D. vittatus, D. and D. sordidus, when acclimated at 20 DC. A range
semistriatus and D. trunculus,' and Ansell and of 29 to 40 DC therefore probably represents the
McLachlan (1980) the upper temperature tolerances maximum range of adaptation in upper lethal
of D. serra and D. sordidus from South Africa using temperature found in the genus.
similar methods. The thermal tolerances of the There are no records of extensive mortality of
European species reflect differences in their ecology Donax species caused by high temperature, but
and geographical distribution. D. trunculus has the Johnson (1966a) reported unusual mortality of D.
greatest thermal tolerance, consistent with its gouldi caused by exposure to hot (7o DC) fresh water
distribution in shallower water; the closely-related from a warm spring during a low tide in ·winter.
species, D. vittatus and D. semistriatus show lower The effects of low temperature on Donax
and much closer thermal tolerance with differences species have been almost entirely neglected. No
 615

investigation of low temperature tolerances has been individuals may burrow initially at lower salinities
made, but McLachlan and Young (1982) have recently than those at which they eventually survive.
investigated the effect of low temperatures on In D. vittatus from European waters, 24-h L T 50
burrowing performance of D. serra and D. sordidus and BT 50 values were significantly reduced at
from South Africa in relation to the effects of sudden salinities of 20 and 17.5 0 /00 compared with values at
temperature depression in summer caused by normal salinities. D. trunculus from the
upwelling. Under certain conditions, temperatures Mediterranean Sea showed little difference in L T 50
may drop from 2o-25 DC down to 14-16 DC and under and BT 50 values for salinities in the range 22-30%,
exceptional conditions as low as lo DC. Burrowing but the effect of lower salinities has not been
time is increased at lower temperatures with Q 10 investigated (Ansell et ai, 1980). Stefan (1962) found
values of 3.47 for D. serra and 5.01 for D. sordidus, that the normal rhythm of shell movements in D.
indicating a greater thermal sensitivity for the lower julianae (= D. trunculus) showed maximum activity
shore tidal migrant species, but comparison of the at c 15 0 /00.
burrowing times at different temperatures with the
time available for burial under different wash 6.3 Air exposure
conditions suggests that only extreme conditions of Of two common species of Donax found
upwelling would retard burrowing in either species to intertidally on Indian beaches, D. faba is more
a level where individuals were in danger of being tolerant of air exposure than D. cuneatus as indicated
swept away and stranded. Mortality of Donax by by 50% survival times at 30 DC of 94 hand 69 h
stranding does occur (Penchaszadeh, Olivier, 1975; respectively (Rao, Kutty, 1969). The relative amount
McLachlan et ai, 1981) but has not been shown to be of mantle cavity fluid was higher in D. faba than in
coupled to upwelling. Orton (1929) reported large D. cuneatus, and D. faba accumulated a smaller
numbers of recently dead D. vittatus along the shore oxygen debt during air exposure. The survival time
following a storm in November of the north-west of D. cuneatus in air was similar at 17DC to that at
coast of Britain. Similar mortalities are not 30 DC, but was significantly reduced at 12DC.
infrequent in northern winters (lingwood, 1976) and
most probably result from the failure of the animals 7. LIFE HISTORY ADAPT A nON
to maintain position when severe disturbance 7.1 Reproduction
coincides with low temperature. Aspects of the reproductive cycle have been
studied for D. vi ttatus (Ansell, 1972) and D.
6.2 Salinity trunculus (Moueza, Frenkiel-Renault, 1973; Lucas,
The tolerance of Donax to factors other than 1965; Badino, Marchionni, 1972) in European waters;
temperature has received less attention. D. for D. serra (De Villiers, 1975a; McLachlan,
variabilis tolerates only relatively small changes of Hanekom, 1979) in South Africa; D. hanley anus
salinity when transferred directly into lower than (Penchaszadeh, Olivier, 1975) in South America; D.
normal salinities (Castagna, Chanley, 1973), and faba (Alagarswami, 1966, 1967) and D. cuneatus (Rao,
experimental results indicate a median lethal salinity 1967) in India; and for D. denticulatus (Wade, 1968)
to this species of 19-20 0/00. This relatively narrow in the Caribbean. The studies reveal no significant
salinity limit is extended appreciably if the bivalves differences in sexuality or basic reproductive biology
are first acclimated at intermediate salinities, and within the genus, with the exception of differences
the median lethal salinity for acclimated animals is which are mainly environmentally mediated.
10-11 0/00. D. variabilis burrows and filters In all species the sexes are separate.
normally, at all salinities at which it survives, and Unusually, there are no records of even rare
616

hermaphrodite individuals being found (Loesch, 1957; species D. trunculus and D. vittatus, there are
De Villiers, 1975a; Moueza, Frenkiel-Renault, 1973; particularly clear annual cycles of reproduction, with
Lucas, 1965; Badino, Marchionni, 1972; the gonads of most animals regressing during the
Nagabhushanam, Talikhedkar, 1977a). The sex ratio winter to an indeterminate state (Ansell, 1972; Poli,
is normally close to equality, although some authors 1972; Badino, Marchionni, 1972; Moueza, Frenkiel-
report a slight preponderance of males (Lucas, 1965; Renault, 1973). In both, the gonad proliferates
de Villiers, 1975a). rapidly in the spring, and is mature throughout the
De Villiers (1975a) found that in D. serra sexual summer until regression begins again in the autumn.
differentiation followed soon after development of There is a close synchrony between individuals in the
the first gonadal alveoli, and there was no early population in the timing of these events, and the
hermaphrodite phase. D. serra reaches maturity timing varies only slightly between populations at
relatively later than other species at c 2 years on the different latitudes (Ansell, et al 1980).
west coast (De Villiers, 1975a) or slightly earlier in As in other bivalves the gonadal cycle in Donax
the warmer waters of the Eastern Cape (McLachlan, is subject to some neurosecretory control. Badino
Hanekom, 1979). Most other species mature early. and Marchionni (1972) showed that there is an annual
The tropical species become sexually mature neurosecretory cycle of the cerebral and visceral
generally within a few months of settlement (Nayar, ganglia paralleling the annual reproductive cycle of
1954; Alagarswami, 1966; Wade, 1968; Trevallion et D. trunculus from the Mediterranean.
ai, 1971). In D. hanley anus sexual differentiation is Nagabhushanam and Talikhedkar (1975) identified two
evident at an age of 2 months in males and 4 months types of neurosecretory cells in the cerebral and
in females (Penchaszadeh, Olivier, 1975). Even in the visceral ganglia of D. cuneatus collected in
cold temperate species D. vittatus, first sexual September, but did not convincingly demonstrate any
development usually occurs in the first year following functional link with the reproductive cycle or
that in which settlement occurred (A.D. Ansell, spawning.
unpublished), and this is also the case for D. trunculus
in European waters (Lucas, 1965; Moueza, Frenkiel- 7.2 Spawning
Renault, 1973; Badino, Marchionni, 1972). Attempts to observe spawning in Donax species
Following first maturity, the gonads in some in the field, and to determine by laboratory
species remain active continuously without a period experimentation what factors trigger spawning, have
of repose; in D. hanley anus (Penchaszadeh, Olivier, met with little success. Stimuli which successfully
1975) there is some synchrony between individuals in induce epidemic spawning at suitable times in other
the development of oocytes, producing two main bivalve species, like Venus striatula, fail to trigger
spawning periods each year; in D. serra (De Villiers, spawning in ripe D. vittatus (A.D. Ansell,
1975a) there is little synchrony and no clearly defined unpublished). Wade (1968) with D. denticulatus,
cycle, although active and spawning stages rise to Chanley (1969) with D. variabilis, and Frenkiel and
higher than average levels during two periods of each Moueza (1979) with D. vittatus and D. trunculus
year. In the other species which have been examined, report similar experiences. De Villiers (1975a) noted
more or less clear seasonal cycles of reproductive that in D. serra populations, individuals with empty
activity occur, although these are generally less gonadal alveoli, indicating complete release of
distinct, and with less synchrony between individuals gametes, were rare, and he concluded that spawning
in tropical species than in those from higher latitudes in this species takes the form of a number of
(Wade, 1968; Alagarswami, 1966, 1967; Nayar, 1955, incomplete releases over an extended period. This is
Rao, 1967, Talikhedkar et ai, 1976). In the European also the case with D. vittatus, where spawning may
 617

occur at intervals through the summer while the spring phytoplankton blooms, when reserves are
animals have mature gonads. A similar pattern of renewed, growth resumes, and rapid gametogenesis
repeated incomplete spawnings by individuals over an and gonadal proliferation take place; and 3) the
extended spawning season, probably represents the reproductive period during the summer, when somatic
basic pattern of spawning activity for Donax species. growth, gonadal growth, and spawning all proceed
It seems to offer the advantage of spreading the together. Later in the summer, there is a progressive
potential for spawning over a longer period than in shift from gonadal regeneration towards storage of
species which show epidemic spawning triggered by reserves, so that in autumn, when the gonad is
environmental stimuli and co-ordinated by chemical regressing the reserves build up to a peak. High
releasers present with the sperm or eggs, and ensures levels of carbohydrate (representing mainly the
that part, at least, of the reproductive effort will stored reserves of glycogen) characterise the periods
produce larvae which find the conditions needed to when stored reserves are high, while lipid levels rise,
complete development successfully and return to the especially in females, when gonadal proliferation
beach environment. predominates.
In its main features the seasonal cycle for D.
7.3 Seasonal changes in tissue weight and trunculus from the Mediterranean closely resembles
biochemical composition that of D. vittatus, although the changes in tissue
In the European species, D. vittatus and D. weight and in composition are less extreme,
trunculus, the clear cycles of reproduction and indicating a lesser importance of the seasonal
growth, interspersed by periods of relative inactivity accumulation of large metabolic reserves (Ando et ai,
in winter in response to decreased temperature and 1976; Bodoy and Masse, 1979; Ansell and Bodoy, 1979;
low levels of food availability, are associated with Ansell et ai, 1980a).
large seasonal fluctuations in tissue weight and in The cyclical fluctuation in reproduction and in
biochemical composition, which show a close the storage and utilisation of reserves shown by these
synchrony in individuals within the population. The species is an adaptation to the predictable seasonal
detailed changes occurring during this seasonal cycle fluctuation in temperature and food availability
were described for D. vi ttatus by Ansell (1972) and found in northern temperate and boreal waters. It
Ansell and Sivadas (1973), who also studied the effect contrasts markedly with the situation found in the
of temperature and starvation on the utilisation of southern warm temperate species D. serra. The
reserves. Bodoy and Masse (1979) and Ansell, et al environment of D. serra shows much less seasonality
(1980a) described the features of the seasonal cycle in temperature and food availability (de Villiers,
for Mediterranean populations of D. trunculus, and 1975a; McLachlan, Hanekom, 1979) and reproductive
Ansell and Bodoy (1979) have compared some cycles are much less distinct. In D. serra, tissue
characteristics of the cycles of these two species. mass decreases in winter, probably as a result of
The cycle in D. vi ttatus has many features in spawning, builds up again during the southern spring,
common with those of other bivalves in the same to a peak in June, and then falls to September,
area (Ansell, Trevallion, 1967). The main features of coinciding with winter spawning. Highest levels of
the cycle are 1) a period of inactivity during the carbohydrate coincide with the June peak in tissue
winter months when reserves stored in various tissues weight, but there is no extended period of nutritional
are drawn upon to supply the metabolic demands of stress to deplete reserves, and no marked cycle of
the animal which are not fully met by the available storage and utilization (McLachlan, Hanekom, 1979).
food supply at this time; 2) a short period following The only tropical species for which the seasonal
the renewal of activity in the spring, and fuelled by cycle has been studied in any detail, D. incarnatus
618

(Ansell et ai, 1972b, 1973) shows relatively little variabilis between 275 and 340 11m (Chanley, 1969b).
seasonal fluctuation in tissue weight and biochemical Young benthic stages attach by means of byssus
composition. The values for composition found for D. secretion, but the young retain the ability to detach
incarnatus, and for D. spiculum for the same beaches, and resecrete the byssus, and may pass through a
compare most closely with values for D. vittatus bysso-pelagic phase in which the byssus filament is
found when res~rves are at their mimimum during the used as a float, allowing secondary resuspension and
annual cycle, indicating that there is no build-up of transport (Sigurdsson et ai, 1976; Frenkiel, Moueza,
high levels of reserves. The limited information 1979). Observations such as those of Williams and
available on other tropical species (Chari, Mukundan Porter (1971) indicate that redistribution of even
Unny, 1947; Velloso et ai, 1951; Rahaman, 1965, later benthic stages (up to 6 mm shell length) may
1966; reviewed by Ansell et ai, 1973) supports this also occur in areas of strong currents, and Irwin
conclusion. In D. faba, changes in tissue weight (1973) has observed post-larvae of D. gouldi invading
follow the seasonal reproductive cycle with the the beach in large numbers "attached to pelagic
highest values being found when the gonads are in full stages of red algae".
condition (Alagarswami, 1966). The main changes in
percentage biochemical composition shown by D. 7.5 Recruitment
cuneatus (Nagabhushanam, Talikhedkar, 1977b) are Recruitment patterns are highly variable both
also related to the gonadal cycle. The lack of build- between different species, and in the same species, in
up of stored reserves in these species may be related different locations, or in different years. Extreme
to the generally close balance between the metabolic cases of such variability have led to their description
demands of the animal and the food available from as "resurgent populations" (Coe, 1953, 1956;
the water throughout the year in the tropics. Mikkelson, 1981). Variability in recruitment in part
reflects the generally extended periods over which
7.4 Larval development spawning may take place, and in part the importance
The eggs of D. vittatus, and probably of other of unpredictable local hydrographic conditions in
Donax species, have a very thin chorion and are determining settlement patterns. Only in areas
unprotected (Frenkiel, Moueza, 1979). After where there are strong seasonal factors does
fertilization, which takes place in the water, they settlement show a more consistent pattern, and even
measure 80 x 80 11m. Larval development is then there are great differences in settlement
planktotrophic, probably for all species (Rees, 1950; density from year to year. Under exceptional
Zakhavatkina, 1959; Wade, 1968; Chanley, 1969b; conditions recruitment to local beaches may be
Frenkiel, Moueza, 1979), and the larval period lasts dependent on the local density of mature spawning
for 3 - 4 weeks. Full details of larval development, stock (De Villiers, 1975b), but it is probably more
and of larval shell and hinge structures are given by normal for populations to recruit from a common
Chanley (1969b) and Frenkiel and Moueza (1979), larval pool to which several beach populations may
while Frenkiel (1982) has studied the structure and contribute (Wade, 1968).
ultrastructure of the cruciform muscle and its Recruitment may be more or less continuous in
associated sense organ during development of D. some species (Wade, 1968; de Villiers, 1975a) but
vittatus and D. trunculus. most species which have been studied show one or
Settlement of Donax juveniles takes place at a more major periods of settlement each year,
relatively small size; D. vittatus metamorphoses at producing major cohorts within the population. In
250 350 11m; (Frenkiel, Moueza, 1979), D. tropical species settlement patterns may be
denticulatus at 245 - 330 11m (Wade, 1968) and D. influenced by major environmental changes, such as
 619

those resulting from monsoon conditions (Ansell et ai, high settlement densities in exceptional years, and
1972b; McLusky et ai, 1975; Nair, 1978) or from other are distinct from mortalities attributable to specific,
periods of heavy rainfall (Wade, 1968). unusual events such as the mortali ty of D. serra on
the west coast of South Africa caused by a bloom of
7.6 Mortality and Population Structure the dinoflagellate Gonyaulax grindleyi (Grindley, Nel,
Mortality rates of Donax species are often high, 1968, 1970; De Villiers, 1975b,1979).
especially in the tropical species. Several studies
have shown more or less linear survivorship curves 7.7 Longevity and Growth
(Ansell et ai, 197Zb; McLachlan, 1979; McLachlan, The majority of tropical and subtropical species
Hanekom 1979; Ansell, Lagardere, 1980). Values of of Donax have a short lifespan of 1 to Z years (Nayar,
the mortality coefficient range from 1.9 x 10- 3 for 1954; Coe, 1955; Wade 1965, 1968; Ansell et ai,
long-lived species like D. vittatus to 6Z.1 x 10- 3 for 1972b; McLusky et ai, 1975; T alikhedkar et ai, 1976;
short-lived species like D. spiculum. Where the McLachlan, 1979). The anomalous interpretation of
mortality rate is constant, the mortality coefficient, Smith (1971) of a 7-year life span for D. rugosus
longevity, and production are related in a general from tropical West Africa requires further
relationship which may be used to show general confirmation. At higher latitudes the lifespan may
trends in these features (Ansell et ai, 1978). be similar to that of the tropical species or may be
Populations with linear survivorship curves, more extended. D. variabilis and its related forms
when stable, often show a preponderance of young have a lifespan of 1 - Z years throughout their
individuals, but fluctuations in settlement density geographical range on the east coast of N. America
usually result in one or more cohorts dominating the (Loesch, 1957; Chanley, 1969a; Morrison, 1971). The
population structure for varying periods. Other South American species D. hanleyanus lives for c 3
populations show relatively low mortality rates for years near the southern end of its range in Argentina
much of their post-larval life, except for a period (Penchaszadeh, Olivier, 1975). The Mediterranean
immediately following settlement, so that the species D. semistriatus, and D. venustus live for 1 -Z
population structure is dominated by relatively large years (Masse, 1968, 1971, 1972a,b). D. vittatus and
individuals, and such populations may show a high D. trunculus in Europe have a short lifespan (c 3
mortality following spawning (Penchaszadeh, Olivier, years) at the southern end of their geographical range
1975; McLachlan, 1979). (Ansell, Lagardere, 1980; Guillou, Le Moal, 1980), but
In D. vittatus increased mortality may occur in longevity increases at higher latitudes and may reach
the spring (Ansell, Sivadas 1973; Warwick et ai, 1978) 6 - 7 years in northern populations of D. vittatus
when reserves in the animal reach their lowest levels (Ansell, 1972). The large South African species D.
for the year, and Ansell and Sivadas (1973) suggest serra may also live for more than 5 years (De Villiers,
that such mortalities during periods of stress 1975b; McLachlan, Hanekom, 1979).
represent a cost associated with the lack of All the tropical and subtropical species of
temperature acclimation of metabolic processes Donax which have been studied show a similar growth
shown by this species. Tropical species may be pattern. Shell growth is rapid, more or less
particularly vulnerable to similar stress-induced continuous and steady during the initial growth period
mortalities because of their generally low levels of of c 1 year, but slows or stops after sexual maturity
reserves and high metabolic rates. In extreme cases is reached, although there may be short periods of
mass mortalities may occur. These appear to be a renewed growth thereafter in response to
regular feature of the biology of some species, exceptionally favourable conditions (Wade, 1968;
(Fitch, 1953; Coe, 1953, 1956) associated with very Nayar, 1954). The longer lifespan of species from
620

higher latitudes is reflected in slower growth rates. of high productivity, selection seems to have
Although the maximum rates of growth recorded for favoured a large size, as in the case of D. serra in
short periods may be similar to those of the tropical South Africa and D. deltoides in Australia. Both
populations, seasonal differences in rate are more these adaptive trends have been accompanied by
pronounced, and the duration of the growth period is compensating changes in the time of first maturity,
reduced (Dorjes, 1979; Ansell, Lagardere, 1980; in longevity, and in metabolic rate when compared
Guillou, Le Moal, 1980; Bodoy, 1982). Where the with the basic life history pattern.
growth curve can be d!;lrived in detail it can be seen
that there is an early period of logarithmic growth, 8. PRODUCTION ECOLOGY
followed, after an inflection in the growth curve, by 8.1 Individual Production
a period when growth rate decreases with increase in Production comprises three elements: somatic
size (Ansell, Lagard~re, 1980). For this latter period growth, gonadal growth, and, in those species like D.
Ford - Walford plots of growth increments for short vittatus which show a cycle of reserve storage and
periods of time (c 1 month) show a considerable utilisation, growth of reserves. In general, only
uniformity of slope indicating little variation in the somatic production by individuals contributes
constant 'k' in the Von Bertalanffy growth equation, significantly to net annual production which is
and reflecting the similarity in longevity between retained within the population since production
species. increments associated with reproduction and with
All species show variations in the maximum storage of reserves, are balanced on an annual basis
size reached in different locations, related to by losses due to spawning and to utilisation.
differences in the local productivity (Nolte, 1958; For most Donax species, somatic production is
Wade, 1965, 1967, 1968; Ansell et ai, 1972b; McLusky the only element in total production which is easily
et ai, 1975; Nair et ai, 1978; Ansell, Lagardere, assessed, i.e. from data on rate of growth. Absolute
1980). Such differences between populations affect production due to somatic growth at first increases
mainly the constant Leo in the Von Bertalanffy rapidly with size and age, reaches a maximum
equation while 'k' is relatively little affected, corresponding to the point of inflection of the
indicating that although the maximum size is altered, weight/age growth curve, and thereafter declines.
the rate at which it is approached is not. Such Relative production is initially high, but declines
differences in growth characteristics between progressively; thus production: biomass (P/B) ratios
populations may be large, and consequently obscure based on somatic growth increments decline with
latitudinal and other trends in growth. increasing age (Warwick et ai, 1978; McLachlan, Van
Modification of maximum size has also been a der Horst, 1979; McLachlan, 1979; McLachlan,
major adaptive feature in evolution for certain Hanekom, 1979; Ansell, Lagard~re, 1980).
groups of Donax species. Most species have a Gonadal production and spawning, and
maximum tissue weight of 200-500 mg although production of tissue reserves and their utilisation,
associated with a range of shell lengths because of being dynamic processes, are difficult to assess in
differences in shell shape between species. Two natural populations. The ripe gonad may account for
adaptive trends seem to have favoured departures at least 50% of the total mass of mature animals, and
from this 'normal' size. Firstly, in environments some attempts have been made to assess gonadal
subject to irregular disturbance, selection seems to production and spawning from changes in tissue
have favoured reduction in size, and several small weight (Ansell et ai, 1972b; McLachlan, Hanekom,
species of Donax are found in such habitats, notably 1979). These are not satisfactory, however, unless
in the tropics. Secondly, in southern temperate areas there is a very close synchrony between individuals in
 621

the population, and a discrete and relatively short Where mortality follows an exponential model the
spawning period involving complete spawning of mortali ty coefficient provides an estimate of E/B
individuals. Most Donax species fail to meet these (Ansell et ai, 1978).
criteria, and the contribution of gonadal production Analyses of production for individual cohorts
to total production is probably generally have been made by empirical methods for D. vittatus
underestimated. In bivalves generally, production from British waters (Warwick et ai, 1978), for D.
associated with gonadal growth increases in absolute trunculus and D. vittatus from European coasts
terms with increase in size and a similar relationship (Ansell, Lagardere, 1980), for D. incarnatus and D.
exists in D. vittatus (A.D. Ansell, unpublished). In spiculum from India (Ansell et ai, 1972b, 1978) and
relative terms, gonadal production increases to a for D. sordidus (McLachlan, Van der Horst, 1979;
plateau in older animals. While somatic growth McLachlan, 1979) and D. serra (McLachlan, Hanekom,
constitutes the major elements in production during 1979) from South Africa. Production due to growth
the early part of the life cycle of individuals, progressively declines during the lifespan of the
therefore, in mature animals production is cohort, while elimination rises to a maximum and
represented almost entirely by growth of the gonad then declines. Relative production (P/B) is initially
and reserves. high and thereafter declines, while relative
The maximum potential rate of production, elimination (E/B) remains constant reflecting the
whether by somatic or reproductive contribution, is linear mortality of the populations studied. The
shown in most species for only relatively short biomass of survivors rises to a peak value at the time
periods when food is available in suitable quantity when P/B and E/B are equal, followed by a gradual
and quality, and when other conditions are decline. The general form of these relationships are
favourable. In D. vittatus such conditions occur in similar in all species, although the details depend on
the early summer, in response to spring blooms of growth, mortality and longevity. In the long-lived D.
phytoplankton. They result in maximum rates of serra for example, the maximum biomass of survivors
production, comprising somatic growth, gonadal is reached at around 3 - 4 years of age (De Villiers,
proliferation and increase in stored reserves. During 1975b); in D. trunculus and D. vittatus on the French
this period only, growth efficiency is independent of Atlantic coast, at 2 - 3 years and c 2 years
body size (A.D. Ansell, unpublished). respectively (Ansell, Lagardere, 1980) and for D.
incarnatus from the Indian coast at c 6 months
8.2 Cohort Production (Ansell et ai, 1972b, 1978).
Assessment of cohort production requires The differences in productivity of individual
information on survivorship for the cohort as well as beaches which are responsible for differences in the
for growth of individuals. Production may be maximum size reached by Donax in different
estimated by summation of losses from the populations, result in differences in the mean biomass
population over a period, a measure often referred to of the population and in productivity. Provided
as elimination (E) (Petrusewitz, 1966), or by longevity and mortality remain the same, however,
summation of production gains (P) by the cohort over P/B and E/B ratios remain unchanged when compared
a period. For a given cohort, P = E over the whole for animals of the same age (Ansell, Lagardere,
lifespan, but during the lifespan of the cohort there is 1980). Geographical or other differences in lifespan,
first a period when P>E and the biomass increases, on the other hand, change the time base of the
while later E>P and biomass decreases. The ratios of changes in biomass and productivity so that animals
P and E to mean biomass i.e., p/B and E/B ratios are of similar age show different P/B and E/B ratios.
extensively used for comparing production rates.
622

8.3 Population Production are not in steady state, population PIB ratios are
The contribution of production by Donax highly dependent on population structure, and such
populations to the overall energy flow of the differences are responsible for much of the variation
community has been assessed in only four instances. in PIB ratios observed in natural populations.
Warwick et al (1978) estimated production by the Nonetheless some geographical trends can be
European species D. vittatus as 2.8% of total distinguished, both within and between species. In D.
macrobenthos production in a Venus community in trunculus in European waters, production, elimination
the Bristol Channel, England, this low value and mean biomass all decrease progressively from
reflecting the subsidiary role of Donax in such north to south in the geographical range, while there
northern European sand communities, which are is an increase in plB and EIB ratios. For D. vittatus,
dominated by other species of bivalve. In contrast, there is no clear trend in production and elimination,
McLachlan et al (1981) estimated that production by but there is a similar trend of increase in plB and
D. serra populations represents c 94% of total EIB ratios from north to south (Ansell, Lagardere,
macrobenthos production, on the Eastern Cape 1980). Bivalves generally show a trend of increase in
beaches of South Africa, with D. sordidus population PIB and EIB ratios from arctic to tropical
contributing a further 2.5%. On two tropical Indian waters, which Donax species follow, although there
beaches Ansell et al (1978) estimated that production are some indications of differences between different
of two Donax species, D. incarnatus and o. spiculum tropical regions (Ansell et aI, 1978).
together represented 56% and 61% of total
macrobenthos production, although the relative 9. METABOLIC ADAPTATION
contributions of the two species differed at the two 9.1 Respiration
sites. For a beach at San Luis, Venezuela, studied by Ansell (1973) and Ansell and Sivadas (1973)
Edwards (1973, a,b,c) production by bivalves, of measured oxygen consumption for D. vittatus under a
which D. denticulatus was the dominant species, variety of conditions, and Ansell and Masse
accounted for only c 5% of total macrobenthos (unpublished data) compared oxygen consumption
production. Clearly, however, there is a possibility rates for three of the European species, D. vittatus,
of a complete range of such values, depending on the O. trunculus and D. semistriatus. Dye (1979) made a
relative dominance of Donax in particular situations, comprehensive study of the effects of acute and
while total annual production may similarly show a long-term temperature change on oxygen
very wide range depending on overall population consumption by D. serra and D. sordidus from South
density and population structure. Africa. Studies of tropical species have been made
Estimates of PIB and EIB ratios for complete for D. incarnatus and D. spiculum from India
Donax populations range from 0.2 for D. serra (McLusky, Stirling, 1975; Ansell et aI, 1978) and for
(McLachlan, Hanekom, 1979) to 5.9 for D. incarnatus D. denticulatus from the Caribbean (Edwards, 1973c);
(Ansell et aI, 1978). Higher values were found by the Rao and Kutty (1969) give some information for D.
latter authors for D. spiculum, but since this species faba and D. cuneatus, and Mane and Talikhedkar
was present for only part of the year, there is some (1976) investigated the effect of various factors on
doubt on the validity of the estimates. When respiration of D. cuneatus.
populations are in steady state, and the mortality For D. vittatus, oxygen consumption rates are
curve follows an exponential model, population plB temperature dependent; Q10 declines with
and EIB ratios are inversely proportional to lifespan temperature in the range 2.9 - 20 0 C (Ansell, 1973).
(Ansell et aI, 1978). When the mortality rate changes D. trunculus and D. semistriatus also show similar
through the lifespan, however, or when populations relationships within their environmental range (A.D.
 623

Ansell and H. Masse, unpublished). For warm- information for D. faba and D. cuneatus, but their
acclimated D. vittatus, oxygen consumption is data are not sufficiently complete for detailed
depressed at low temperatures and increased at comparison.
higher temperatures, relative to cold-acclimated Comparison of the rate-temperature
animals, i.e., there is a slight anticlockwise rotation relationships for the European temperate species,
of the rate-temperature curve (Ansell, Si vadas, with the Indian tropical species, suggest that no
1973). There are no marked seasonal changes in lati tudinal temperature compensation occurs
rates, and the exponent 'b' in the mass-metabolism (McLusky, Stirling, 1975; Steele, 1976). Similar
equation is not affected by temperature or by conclusions were reached in studies of sandy beach
seasonal factors. gastropods (Brown et ai, 1978). In this respect the
In contrast, in D. serra (Dye, 1979), summer invertebrates of sandy beaches contrast with tropical
respiration rates are significantly higher than winter flatfish, which show a plateau of temperature
rates, and the exponent 'b' is significantly changed by independence at higher temperatures (Edwards et ai,
seasonal temperature acclim ation. Q 10 is dependent 1970).
on body size in summer-acclimated animals only, and The respiration rates of those species which
acclimation causes a shift to the right in the have been studied to date may be compared more
metabolism-temperature curve for winter-acclimated readily if the measured rates are first adjusted for
animals. D. sordidus also shows significant effects of temperature effects. Adequate adjustment may be
seasonal acclimation. made using a Q10 value of 2, to calculate equivalent
These differences in temperature response rates at 20°C. For most species, the adjusted rates
between the European and South African species for adult animals fall close to standard rates for
indicate significant differences in the metabolic poikolothermic animals (Hemmingsen, 1960). Rates
adaptation of these two groups from cooler waters. for D. spiculum, however, whose maximum size is c
For D. incarnatus and D. spiculum, there was a 10 times smaller than most species, lie above the
ten times increase in oxygen consumption for open, standard line, while those for D. serra, which is some
actively filtering animals, compared with closed 10 times greater in maximum size, lie below the line,
individuals, with values for spontaneously active so that there is less variation overall than would be
animals intermediate between them (McLusky, predicted from size alone. Because of differences in
Stirling, 1975). Only the open, active animals showed the exponent of the mass-metabolism equation
a significant effect of body weight on oxygen between species, however, rates for smaller
consumption. Temperature had relatively little individuals of D. serra, D. sordidus and D. incarnatus
effect on rate, but measurements were only made at also fall above the standard line, comparing closely
25 and 30°C. The relatively small number of with the rates found for D. spiculum. The higher
measurements made for these species by Ansell and than average metabolic rates for juveniles of these
Trevallion (Ansell et ai, 1978) fall within the range of species seems consistent with their migratory habits
McLusky and Stirling's results. (the juvenile D. serra used by Dye (1979) are found in
In D. cuneatus (Mane, Talikhedkar, 1976), the rate of the wash zone), and suggests that the measurements
oxygen consumption increases with temperature made with these species include, at least in part, a
between 24 and 34°C, but decreases again at 34°C. contribution of active metabolism.
Oxygen consumption in this species is increased
following periods of exposure, but decreased by 9.2 Feeding
reductions in salinity and following periods of In contrast to most members of the superfamily
starvation. Rao and Kutty (1969) also give some Tellinacea, which comprises predominantly deposit-
624

feeding bivalves (Vonge, 1949), Donax species rely on relative rates, fail to measure the actual rates of
the collection of suspended particles from the water water transport, because not all the neutral red
for food (Wade, 1965, 1967, 1969; Pohlo, 1967, 1969; transported through the mantle cavity is retained.
Narchi, 1978; Moueza, 1976; Ansell, 1981). The McLusky and Stirling (1975) measured the rate of
contents of the stomach usually consist of pigmented clearance of unialgal cultures of Tetraselmis chui by
plant material and diatoms with relatively few sand D. incarnatus from Indian beaches and Ansell (1981)
grains (Pohlo, 1967; Mori, 1938; Moueza, 1976). measured the rate of clearance of the flagellate
Krishnamurthy et al (1966) showed that D. cuneatus Monochrysis lutheri, by D. serra and D. sordidus from
would ingest yeast, bacteria, and algal cells, but their South Africa. Both show interesting anomalies which
experiments are somewhat inconclusive in relation to require further investigation.
relati ve food value. Pohlo (1967) concluded that For D. incarnatus, the measured filtration rate
Donax is an unspecialised or transitional type of fell off very rapidly with time; rates measured after
tellinacean, but Wade (1969) argued that the 20 mins were significantly less than those measured
presence of morphological features associated with 10 mins after transfer to the experimental
deposit-feeding points to the Donacidae having conditions. A similar pattern was found for D.
evolved from a deposit-feeding stock. sordid us by Ansell (1981); filtration rates following
The species of Donax for which details of transfer were initially high and then fell. D. serra,
functional morphology have been described; D. however, showed a completely different response. In
vittatus (Atkins, 1937a,b; Vonge, 1949), D. trunculus this species (Ansell, 1981) there was a relatively low
(Moueza, 1975,1976; Moueza, Frenkiel, 1974, 1976a, rate for the first period after transfer followed by a
1978), D. denticulatus (Wade, 1969), D. striatus higher rate which was then maintained. Both D.
(Wade, 1967), D. gouldi (Pohlo, 1967) and D. serra and incarnatus and D. sordidus are tidal migrants, while
D. sordidus (Ansell, 1981) differ in some aspects of D. serra may spend up to half of each tidal cycle in
their feeding morphology. Differences include the the exposed intertidal. These limited results suggest
relative sizes of ctenidia and palps, the degree of that in the tidal migrant forms, which open the
plication of the gill, the ciliation of the mantle inhalant siphon to feed in the water current draining
cavity and visceral mass, and the direction of some from the beach after each wave, the ciliary
of the main ciliary tracts on the ctenidia. There are mechanisms of the gill may be 'switched on' by
also differences in the length of different regions of sensory stimulation from the regular disturbance;
the gut, and in the structure of the stomach of while species which maintain a fixed position, like D.
different species (Purchon, 1958, 1960; Nakazima, serra and D. vittatus feed more continuously.
1965; Pohlo, 1967; Reid, 1965; Wade, 1969, Ansell, Further quantitative studies of feeding by other
1981). For the two South African species, Ansell species are required if such differences related to
(1981) attempted to relate such differences to behaviour are to be confirmed.
differences in distribution and behaviour, but details The maximum rates of water filtration reached
are known for too few species for wider conclusions for both D. serra and D. sordidus showed a clear
to be drawn. relationship to body mass, with exponents of 0.834 -
Published records of rates of filtration by 0.937 in the relationship between log filtration rate
Donax species are sparse. Ansell and Sivadas (1973), and log dry tissue mass. (Ansell, 1981). Rates for
and Talikhedkar and Mane (1977) compared relative individuals of the same size were similar for the two
filtration rates for experimental animals using the species. For D. incarnatus (McLusky, Stirling, 1975)
neutral red method of Cole and Hepper (1954), but there was, unusually, little effect of body weight on
such measurements, although useful for comparing filtration rate.
 625

When the measured rates of filtration for D. by bivalves may include protein-linked
serra and D. sordidus are compared with measured mucopolysaccharides secreted as part of the normal
rates of oxygen consumption for the same species feeding activities of the animal (Allen, Garrett, 1971)
(Dye, 1979) they provide an estimate of c 10% for and amino-acids which have leaked through the body
removal efficiency of oxygen for larger animals. wall (Hammen, 1968). Hammen (1968) showed that in
Because of the differences in the exponents for the D. variabilis the proportions of identified excretory
relationships for filtration rate and respiration rate, products were; ammonia 74.6%, urea 0%, amino acids
the apparent efficiency of removal for smaller 24.6% and uric acid 0.8%. The rates of ammonia
animals is greater. The greater activity of small D. excretion, and of amino-acid loss for D. variabilis and
serra (which are found in the wash zone) and of D. D. vittatus compare closely with values for other
sordidus apparently requires a higher removal of bivalves which have been measured (Hammen, 1968;
oxygen from the water pumped through the mantle Allen, Garrett, 1971; Bayne et ai, 1976; Lewin et ai,
cavity. The reduction in filtration rate with time in 1979).
D. sordidus and D. incarnatus appears to have the
effect of reducing the rate of water flow to a level 9.4 Metabolism
just sufficient to satisfy the estimated oxygen Ansell (1977) measured the adenosine
requirments. triphosphate (ATP) content of D. vittatus, D.
trunculus and D. semistriatus from European waters
9.3 Excretion and found values of 0.78, 0.75 and 0.62% of dry tissue
Ansell and Sivadas (1973) studied the rates of weight respectively. High maintained levels of ATP
excretion of NH 3-nitrogen for D. vittatus using in bivalves are associated with the ability to use
individuals freshly collected from field populations, energy rapidly for short periods, for example in
and animals which had been kept under experimental escape responses, usually reflecting high levels of the
conditions, to study the effects of temperature and phosphagen, phosphoarginine, in muscle tissue. Tidal
food deprivation on metabolism. Acutely-measured migrant species may be expected to show even higher
NH 3-N excretion from freshly collected animals was levels. Preliminary observations of the maximal
temperature dependent with a Q10 of c 2 over the activities of glycolytic enzymes in D. vittatus from
range 10 -20 DC. NH 3 -N excretion showed a Europe, and in D. serra and D. sordidus from South
significant correlation with tissue dry weight with an Africa (Blackstock, Ansell, 1981) show very high
exponent of 0.639 in the rate mass equation. The levels of octopine dehydrogenase (ODH) in the two
ratio of oxygen consumed : NH 3-N excreted (O:N South African species. In muscles having high ODH,
ratio) for freshly collected animals varied from 19.6 - ATP for muscle contraction is obtained mainly by the
45.7 with a mean of 36.4 ± 2.5 (SE). breakdown of arginine phosphate (Zamitt and
When acclimated to constant temperature, but Newsholme, 1976), the arginine formed condensing
without feeding, so that the animals were utilizing with pyruvate produced by glycolysis to form
body tissue to provide energy-yielding substrates, the octopine. High ODH activity in the South African
O:N ratio decreased to 17.74 reflecting the species seems to be related to the greater activity
significantly increased utilisation of protein as a needed to maintain zonation in these species on
respiratory substrate. Under these circumstances exposed surf beaches compared with the more
nitrogen loss from the tissues by means other than sheltered conditions in which D. vittatus is found.
excretion of ammonia also increased, such losses
perhaps accounting for up to 75% of the total
nitrogen loss. Nitrogenous products lost in this way
626

10. HUMAN IMPACT of this material.

10.1 Pollution In polluted areas, Donax species may
Few studies have dealt with the impact of accumulate metals (Stenner, Nickless, 1975; Mauri
pollution on Donax populations, although clearly their et aI, 1978; Hornung, Oren, 1981). In D. trunculus,
habitats and habits render them particularly zinc and copper are more concentrated in the
susceptible to pollution from coastal industry, and digestive gland, and iron and manganese in the kidney
from oil spillage at sea, or from major tanker (Orlando, Mauri, 1978; Mauri, Orlando, 1979; Mauri,
accidents. Those studies which have been made 1980). Metals are accumulated in the kidney as solid
indicate that Donax populations may indeed be highly granules which in D. trunculus are rich in manganese,
susceptible to the effects of such pollution. and their accumulation may be affected by other
D. denticulatus from Venezuela shows a lower factors such as season and sex (Mauri, Orlando, 1982).
LD50 when exposed to light crude petroleum than the
mangrove oyster Crassostrea rhizophorae from the 10.2 Commercial exploitation
same area (Mahieu et aI, 1981). Burrowing behaviour Few Donax species are exploited commercially
is affected at lower sublethal concentrations, a on an intensive scale, but many species are utilized
factor which could be of greater significance than locally for human consumption or for bait. In Europe,
direct mortality in this tidal migrant species. D. D. trunculus is harvested in most countries bordering
trunculus in Europe is also more susceptible than the the Mediterranean Sea, and in some areas special
mussel Brachidontes variabilis, from the same area, dredges are used for this purpose dragged by man or
to crude oil, to the oil dispersant Corexit 7664, and animal power (Moueza, 1975; Aleem, 1969). In Egypt,
to mixtures of these (Avolizi and Nuwayhid, 1974). Aleem (1969) records the annual catch of D.
The time to LC 50 decreases with increasing trunculus from populations near the mouths of the
concentration and is lower with oil-dispersant river Nile as 3000 - 5000 tons with a value of 150,000
mixtures than with either oil or dispersant alone. - 200,000 Egyptian pounds. Most species are utilized
The respiratory rate is also decreased by exposure to only locally, often by holidaymakers, as in the case of
oil, to dispersant and to mixtures. D. denticulatus in Trinidad (Bacon, 1970). Talavera
Field studies have confirmed that Donax and Faustino (1933) record D. radians (= D. faba)
populations are susceptible to oil pollution. among the edible molluscs of Manila. Alagarswami
Following the grounding of the tanker Amoco-Cadiz (1966) records that this species is among the few
on the Brittany coast of France, D. vittatus was shellfish used by fishing communities in India,
among the species severely affected by stranding of although Talikhedkar et aI, 1976 state that only D.
the resultant oil spill on beaches. D. vittatus cuneatus and D. scortum are valued as food in India,
disappeared completely from affected beaches where and then only for poorer people. Coe (1953) and
it had been previously abundant (Chasse, Guenole- Johnson (1968) mention development of plants for
Bauder, 1981), but the species has recently begun to canning broth made from D. gouldi in California, a
reappear in the same areas (C. Chasse, personal use which the massive population fluctuations of this
communication). species quickly rendered uneconomic. In South
Neff et al (1981) showed that D. variabilis Africa, the large species D. serra is subject to
texasiana was very sensitive to the layered solid intensive exploitation by bait dealers and anglers on
phase (LSP) resulting from used drilling mud, but not the west coast (De Villiers, 1975a) although this use
to the suspended solids phase. Sensitivity to LSP may represents only a minor loss to the population (3% of
have resulted, however, from depletion of oxygen in total predation losses) on the Eastern Cape coast
the experimental aquarium caused by the high BOD (McLachlan et aI, 1981).
 627

Donax species have been recorded as Donax species are found most abundantly in
accumulating toxins from red tide organisms, eutrophic areas with high plankton productivity,
sometimes to very high levels (Cummins et ai, 1971) where their contribution to secondary production may
and may themselves be killed in large numbers as a reach high levels. All species are suspension feeders,
result of red tides (Grindley, Nel, 1968; De Villiers, and metabolically they are resource exploiters rather
1975b) with full recovery requiring as much as 14 than conservers. This facili tates their efficient
years (De Villiers, 1979). utilisation of periods of high levels of productivity,
but may render them particularly susceptible to
11. CONCLUDING SUMMARY periods of nutritive stress, and lead to high mortality,
The genus Donax consists of a morphologically which may appear catastrophic where it occurs at
discrete group of species which has its origins in times when population density is high.
tropical waters and still shows its greatest species Donax species show little variability in basic
diversity in the tropics. Individual species, and life history. All are gonochoristic, with equality of
groups of species characteristically dominate sandy sex ratio, larval development is planktotrophic, and
beaches in well-defined zoogeographical areas, and primary settlement occurs at relatively small size,
there are no cosmopolitan species. The genus has although later bysso-pelagic redistribution may
extended its distribution towards higher latitudes in occur. Most Donax are short-lived, with a lifespan of
Europe, North and South America, South Africa, and 1 - 2 years, with rapid growth to maximum size, and
Australia, but only few species are represented in early maturity. Reproduction, following maturity, is
such cooler waters. The genus is extremely potentially continuous with repeated partial
successful and forms a major dominant in the fauna spawnings, but growth, reproduction and settlement
of exposed sandy beaches in tropical and subtropical may all show seasonal responses to environmental
waters. change.
Donax species show considerable uniformity of Adaptive trends in these characteristics within
adaptation to habitat. All are found in relatively the genus express themselves mainly in modifications
high energy environments such as the intertidal and of maximum size, longevity and metabolic rate. In
shallow subtidal zones of clean sand beaches, or in the tropics, these trends have led toward smaller
coarse sediments in current-swept situations size; selection in environments subject to strong
offshore. A major feature of the success of the disturbance seems to have favoured more
genus is the ability to co-ordinate their movements in opportunistic species, with shorter lifespan, very
response to changes in physical conditions resulting early maturity, great mobility and high metabolic
from wave and current action. This is expressed in rate. Spread of the genus into cooler waters has
the variety of dynamic patterns of spatial involved other changes. In Europe, where strong
distribution found in different species and reaches its seasonal fluctuations in temperature and productivity
most complex expression in the habit of tidal are major features of inshore marine environments,
migration, a habit shared with some gastropods and seasonality of growth and reproduction have become
crustaceans from the same habitats, but unique to important features of the life cycle; metabolic rate
Donax among bivalves. Consistent with the relative shows little seasonal acclimation and a distinct
uniformity of their geographical and habitat annual cycle of storage and utilisation of reserves
distribution, the total range of adaptation of allows the animal to survive a prolonged period of
environmental tolerance between different Donax nutritive stress during the winter at the expense of
species is rather limited, while sensitivity to earlier periods of high productivity. Annual growth is
pollution is high. reduced, and the lifespan is increased. In contrast, in
628
the South African species D. serra, which lives in
waters with much less seasonal fluctuation and with
high year-round productivity, a longer lifespan is
accompanied by a late maturity and growth to a
much larger maximum size. Seasonality in growth
and reproduction is less marked; the metabolic rate
shows seasonal acclimation; and the storage and
utilisation of reserves has much less importance in an
annual cycle with longer periods of nutritive stress.
Many gaps remain in our knowledge of this
interesting group. In functional morphology (not
reviewed in detail here) insufficient detail
available to evaluate relationships between groups of
species which have been proposed as subgenera on the
basis of shell characters. The biology of many
tropical species remains unknown in spite of the
importance of Donax in the intertidal ecology of
tropical beaches, and unfortunately this is
particularly for those areas of the Indo-Pacific region
where the greatest species diversity is found.
physiology of the tropical species is even less well
studied, and our present information on respiration
and feeding contains anomalies which only more and
careful investigation will resolve. Preliminary
observations also suggest that there are important
metabolic adaptations associated with
differences in distribution, environmental tolerance,
and behaviour of different species, which deserve
further study. The active research interest in the
group which exists in Europe, in North and South
America, Japan, India and South Africa, for its
intrinsic interest, its importance as a dominant in
beach communities, and increasingly as a sensitive
indicator of the impact of man's activity on the
beach environment, should continue to fill these gaps.
REFERENCES
Alagarswami K (1966) Studies on some aspects of
biology of the wedge clam Donax faba from
Mandapam coast in the Gulf of Mannar, J. Mar. bioi.
Ass. India. 8, 56-75.
Alagarswami K (1967) Studies on some aspects of
biology of the wedge shell Donax faba Gmelin from
Mandapam coast in the Gulf of Mannar, Adv. Abstr.
Contr. Fish. Aquat. Sci. India. 1, 11-12.
Aldrich JW (1959) Activities of coquina clams,
Atlantic Nat. 4(1), 41-43.
Aleem AA (1969) Marine resources of the United
Arab Republic, Stud. Rev. gen. Fish. Coun. Mediterr.
No. 43: 1-22.
Allen JA and Garrett MR (1971) The excretion of
ammonia and urea by Mya arenaria L. (Mollusca:
Bivalvia), Compo Biochem. Physiol., 39A, 633-642.
Amouroux JM (1972) Donnees sur la structure et I'
instabilite des peuplements infra-littoraux de la cote
du Rousillon, These de 30 cycle, Universite Paris VI.
Ando T Moueza M and Ceccaldi HJ (1976)
Variations des lipides et des sterols chez Donax
trunculus L. (mollusque, Lamellibranche) durant les
mois d'automne et d'hiver, Compte r. Seanc. Soc.
Bioi. 170(1), 149-153.
Ansell AD (1969) Leaping movements in the
is Bivalvia, Proc. malac. Soc. Lond. 38, 387-399.
Ansell AD (1972) Distribution, growth and seasonal
changes in biochemical composition for the bivalve
Donax vittatus (da Costa) from Kames Bay, Millport,
J. expo mar. Bioi. Ecol. 10(2), 137-150.
Ansell AD (1973) Oxygen consumption by the
bivalve Donax vittatus (da Costa), J. expo mar. Bioi.
Ecol. 1l(3), 311-328.
Ansell AD (1977) The adenosine triphosphate
true content of some marine bivalve molluscs, J. expo
mar. Bioi. Ecol., 28, 269-283.
Ansell AD (1981) Functional morphology and
The feeding of Donax serra Roding and Donax sordidis
Hanley (Bivalvia: Donacidae), J. moll. Stud. 47, 59-
72.
Ansell AD and Bodoy A (1979) Comparison of
events in the seasonal cycle for Donax vittatus and
Donax trunculus, In Naylor E and Hartnoll RG, eds.
Cyclic phenomena in marine plants and animals. pp.
191-198. Oxford. Pergammon Press.
Ansell AD and Lagardere F (1980) Observations on
the
the biology of Donax trunculus and D. vittatus at Ile
d'Oleron (French Atlantic Coast), Mar. Bioi. 57, 287-
300.
Ansell AD and McLachlan A (1980) Upper
temperature tolerances of three molluscs from South
African sandy beaches, J. expo mar. Bioi. Ecol., 48,
243-25l.
Ansell AD and Sivadas P (1973) Some effects of
temperature and starvation on the bivalve Donax
vittatus (da Costa) in experimental laboratory
populations, J. expo mar. Bioi. Ecol. 13, 229-262.
Ansell AD and Trevallion A (1967) Studies on
Tellina tenuis (da Costa). I. Seasonal growth and
biochemical cycle, J. expo mar. Bioi. Ecol., 1, 220-
235.
Ansell AD and Trevallion A (1969) Behavioural
adaptations of intertidal molluscs from a tropical
sandy beach, J. expo mar. Bioi. Ecol., 4(1): 9-35.
Ansell AD and Trueman ER (1973) The energy cost
of migration of the bivalve Donax on tropical sandy
beaches, Mar. Behav. Physiol. 2(1), 21-32.
Ansell AD Frenkiel Land Moueza M (1980)
Seasonal changes in tissue weight and biochemical
composition for the bivalve Donax trunculus, J. expo
mar. Bioi. Ecol. 45, 105-116.

Ansell AD Sivadas P and Narayanan B (1973) The
ecology of two sandy beaches in South-west India IV.
The biochemical composition of four common
invertebrates, Spl. Publ. mar. bioi. Ass. India. 333-
348.
Ansell AD Barnett PRO Bodoy A and Masse H
(1980) Upper temperature tolerances of some
European molluscs II. Donax vittatus, D.
semistriatus and D. trunculus, Mar. Bioi., 58, 41-46:-
Ansell AD Sivadas P Narayanan B Sci. Tech. CNEXO: Acta Colloq., 7, 753-766.
Sankaranarayanan VN and Trevallion A (1972a) The
ecology of two sandy beaches in south-west India I.
Seasonal changes in physical and chemical factors
and in the macrofauna, Mar. Bioi., 17, 38-62.
Ansell AD Sivadas P Narayanan Band Trevallion A
(1972b) The ecology of two sandy beaches in south-
west India III. Observations on the population of
Donax incarnatus and D. spiculum, Mar. Bioi. 17(4)
318-332.
Ansell AD Mclusky OS Stirling A and Trevallion A
(1978) Production and energy flow in the
macrobenthos of two sandy beaches in South West
India, Proc. R. Soc. Edinb. (Sect.B). 76, 269-296.
Armitage TM and Alevizon WS (1980) The diet of
the Florida pompano (Trachinotus carolinus) along the
East Coast of Central Florida, U.S.A., Fla. Sci. 43(1)
19-26.
Atkins 0 (1937a) On the ciliary mechanisms and
inter-relationships of lamellibranchs. Part II. Sorting
devices on the gills, Quart. J. Micr. Sci., 79, 339-
373.
Atkins 0 (1937) On the ciliary mechanisms and
inter-relationships of lamellibranchs. Part III. Types
of lamellibranch gills and their food currents, Quart.
J. mar. Sci., 79, 375-421.
Avolizi RJ and Nuwayhid M (1974) Effects of
crude oil and dispersants on bivalves, Marine Pollut.
Bull. 5(10), 149-153.
Bacon PR (1970) Studies on the biology and
cultivation of the mangrove oyster in Trinidad with
notes on other shellfish resources, Trop. Sci. 12(4),
265-278.
Badino G and Marchionni V (1972) Neurosecretion
and gonad maturation in a popUlation of Donax
trunculus L. from Leghorn (Italy), Bolettino zool.
39(3), 321-326.
Bally R (1981) The ecology of three sandy beaches
on the west coast of South Africa, PhD. Thesis.
University of Cape Town. 404pp.
Bayne BL Bayne CJ Care foot TC and Thompson RJ
(1976) The physiological ecology of Mytilus
californianus, Conrad, Oecologia (Berlin). 22, 211-
228.
Beeson RJ and Johnson PT (1967) Natural
bacterial flora of the beach clam Donax gouldi, J.
Invertebrate Pathol. 9, 104-110.
Blackstock J and Ansell AD (1981) Maximal
glycolytic enzyme activities in muscles of bivalves of
the genus Donax in relation to their behaviour and
ecology, Abst. 3rd Congress European Society for
Comparative Physiology and Biochemistry.
Noordwijkenhout, Netherlands. 132-133.
629
Bodoy A (1982) Croissance saisonnaire du bivalve
Donax trunculus en Mediterranee nord-occidentale
(France), Malacologia. 22(1-2) 353-358.
Bodoy A and Masse H (1977) Etude sur la
resistance a la temperature de quelques mollusques
marins des cotes de Provence, Bull. Ecol. 8, 91-101.
Bodoy A and Masse H (1979) Quelques parametres
permettant de suivre la production organique d'un
mollusque bivalve au cours d'un cycle saisonier, Publ.
Boss KJ (1971) Familial affinities of Hemidonax
(Bivalvia), Nautilis, Philad. 85: 9-12.
Brown AC (1964) Food relationships of the
' intertidal sandy beaches of the Cape peninsula, South
African J. Sci. 60. No.2. 35-41.
Brown AC (1979) The energy cost and efficiency
of burrowing in the sandy-beach whelk Bullia digitalis
(Dillwyn) Nassariidae, J. expo mar. Bioi. Ecol., 40,
' 149-154.
Brown AC (1982) Towards an activity budget for
the sandy beach whelk Bullia digitalis (Dillwyn),
Malacologia, 22, 681-683.
Brown AC Ansell AD and Trevallion A (1978)
Oxygen consumption by Bullia (Dorsanum)
melanoides (Deshayes) and BulliaCITgTtalis Meuschen
(Gastropoda Nassariidae) - an example of non-
' acclimation, Compo Biochem. Physiol., 61A, 123-125.
Buchanan JB (1954) Marine Molluscs of the Gold
Coast West Africa, J. West African Science Assoc.
1, 30-45.
Castagna M and Chanley P (1973) Salinity
tolerance of some marine bivalves from inshore and
estuarine environments in Virginia waters on the
western mid-Atlantic coast, Malacologia 12(1), 47-96.
Chanley P (1969a) Donax fossor: a summer range
extension of Donax variabilis, Nautilis. 83(1): 1-14.
Chanley P (1969b) Larval development of the
coquina clam Donax variabilis Say, with a discussion
of the structure of the larval hinge in the Tellinacea,
Bull. mar. Sci., 19, 214-244.
Chari ST and Mukundan Unny M (1947) The food
value of two common molluscs found near Pamban
and the chemical composition of their shells, Current
Sci. 16, p.294.
Chasse C and Guenole-Bouder A (1981)
Comparaison quantitative des populations benthiques
des plages de St Efflan et St Michel-en-Grieve avant
et depuis Ie naufrage de L'Amoco Cadiz. in: Amoco
Cadiz: Consequences d'une pollution accidentelle par
les hydrocarbures, Actes Colloque Int. Publ. CNEXO,
Paris, p. 347-357.
Coe WR (1953) Resurgent populations of littoral
marine invertebrates and their dependence on ocean
currents and tidal currents, Ecology. 34, 225-229.
Coe WR (1955) Ecology of the bean clam Donax
gouldi on the coast of Southern California, Ecology.
36, 512-514.
Coe WR (1956) Fluctuations in populations of
littoral marine invertebrates, J. mar. Res. 15: 212-
232.
Cole HA and Hepper BT (1954) The use of neutral
red solution for the comparative study of filtration
rates of lamellibranchs, J. cons. perm. into Explor.
Mer. 20, 197-204.
630
Comps M and Raimbault R (1978) Infection
rickettsienne de la gland digestive de Donax
trunculus Linne, Science Peche. No.281, 11-12.
Crane J (1975) Fiddler crabs of the world.
Ocypodidae genus Uca, 1-736. Princetown,
Princeton University Press.
Crichton MD (1942) Marine Shells of Madras, J.
Bombay Nat. Hist. Soc. 62, pp. 323-341.
Cummins JM Jones AC and Stevens AA (1971)
Occurrence of toxic bivalve molluscs during a
Gymnodinium breve "Red Tide", Trans. Amer. Fish.
Soc., 100(1): 112-116.
Dahl E (1952) Some aspects of the ecology and
zonation of the fauna of sandy beaches, Oikos, 4, 1-
27.
Degiovanni C and Moueza M (1972) Contribution a
I'etude de la biologie de Donax trunculus L.
(Mollusque Lamellibranche) dans I'Algerois. Ecologie
en baie de Bou Ismail, Tethys. 4, 729-744.
Dexter DM (1969) Structure of an intertidal sandy-
beach community in North Carolina, Chesapeake Sci.
10(2): 93-98.
Dorjes J (1979) Zur populations dynamik des
Sagezahnchens, Donax vittatus, Natur. und Museum.
109(9), 305-311.
Du Preez HH (1981) M.Sc. Thesis, University of
Port Elizabeth, 231 pp.
Duval DM (1963) The comparative anatomy of
some lamellibranch siphons, Proc. malac. Soc. Lond.
35: 289-295.
Dye AH (1979) The effect of acute and long term
temperature changes in the respiration of two sand-
dwelling bivalves, Compo Biochem. Physiol. 63A,
405-409.
Edgren RA (1959) Coquinas (Donax variabilis) on a
Florida beach, Ecology. 40, 498-502.
Edwards RRC Blaxter JHS Gopalan VK Mathew CV
and Finlayson DM (1970) Feeding, metabolism and
growth of tropical flatfish, J. expo mar. Bioi. Ecol.
6, 279-300.
Edwards RRC (1973a) Production ecology of two
Caribbean marine ecosystems. I. Physical
environment fauna, Estuarine coastal marine Science,
I, 303-318.
Edwards RRC (1973b) Production ecology of two
Caribbean marine ecosystems. II. Metabolism and
energy flow, Estuarine coastal marine Science, I,
319-333.
Edwards RRC (1973c) Rates of oxygen
consumption of marine fish and invertebrates from
Venezuela, Bol. Inst. Oceano gr. Cunana Venezuela.
Ekman S (1953) Zoogeography of the sea, Sidgwick
and Jackson Ltd. London. 417pp.
Fitch JE (1953) Common marine bivalves of
California, Calif. Dept. Fish Game, Fish Bull. No.
90, 1-102.
Frenkiel L (1980) Les organes sensoriels chez les
lamellibranches, Ann. Bioi., 19(2), 147-162.
Frenkiel L (1982) L'organe sensoriel du muscle
cruciforme des Tellinacea (mollusques
lamellibranches), Structure-ontagenese-role. These
de Docteur en Sciences Naturelles Universite Paris 6,
149pp 51pl.
Frenkiel Land Moueza M (1977) Le
fonctionnement du muscle cruciforme des Tellinacea
J. Moll. Stud., 43, 189-191. '
Frenkiel Land Moueza M (1979) Developement
larvaire de deux Tellinacea, Scrobicularia plana
(Senelidae) et Donax vittatus (Donacidae) Mar. Bioi.
55, 187-195. "
Gauld DT and Buchanan JB (1956) The fauna of
sandy beaches in the Gold Coast. Oikos. 7 II 293-
301. ' ,
Geldiay Rand Uysal H (1972) Ege Denizi
Satillerinde Tesbi t Edilen Donacilla (Donacilla)
cornea (PolO ve Donax (Serrula) trunculus un
Populasyon Dinamig;-ve Ekolojik Ozellikleri
Hakkinda, Sci. Rept. Fac. Sci., Ege, Univ. 135, 1-8.
George CJ (1965) The use of beached valves of the
Lamellibranch molluscs Glycimerus glycimeris (L.)
Donax semistriatus Poli and Donax trunculus L. fo;
the determination of percentage mortality by Natica
spp, Doriana. 4, 164, 1-8. ---
Graham A (1934a) The cruciform muscle of
Lamellibranchs, Proc. Roy. Soc. Edinb. 54, 1, 17-30.
Graham .A (1934b) The structure and relationship
of LameliIbranchs possessing a cruciform muscle
Proc. Roy. Soc. Edinb. 54(2), 158-187. '
Grindley JR and Nel EA (1968) Mussel poisoning
and shellfish mortality on the west coast of S. Africa,
S. Afr. J. SCI. 64: 420-422.
Grindley JR and Nel EA (1970) Red water and
mussel poisoning at Elands Bay, December 1966, Fish.
Bull. S. Afr. 6, 36-55.
Guillou J and Le Moal Y (1978) Variabilite spatio
temporelle des popUlations de Donax en baie de
Douarnenez, Haliotis Paris. 9(1), 77-80.
Guillou J and Le Moal Y (1980) PopUlation
dynamics of Donax vittatus and Donax trunculus in
the bay of Douarnenez, France, Ann. Inst. Oceanogr.
56(1), 55-64.
Hammen CS (1968) Amino transferase actvities
and amino acid excretion of bivalve molluscs and
brachiopods, Compo Biochem. Physiol., 26, 697-705.
Hedgepeth JW (1953) An introduction to the
zoogeography of the north-western Gulf of Mexico
with reference to the invertebrate fauna, Pubis. Inst.
Mar. Sci. Univ. Tex. 3,110-224.
Hedgepeth JW (1957) Sandy beaches. Chapter 19 in
Treatise on Marine Ecology and Paleoecology, Geol.
Soc. America, Memoir 67, Vol. 1, 587-608.
Hemmingsen AM (1960) Energy metabolism as
related to body size and respiratory surfaces and its
evolution, Rep. Steno. Mem. Hosp. Copenh. 9 1-
110. " ,
Hiroki K (1971) Fisioecologia de invertebrados
marinhos. Resistencia a anoxia, Boletim zool. Bioi.
mar. Univ. Sao Paulo (N.S.) No.28, 315-341.
Hod~son ~N (1982a) Studies on wound healing, and
a.n estimation of the rate of regeneration, of the
Siphon of Scrobicularia plana (da Costa), J. expo mar.
Bioi. EcoJ. 62, 117-128.
Hodgson AN (1982b) Studies on wound healing and
regeneration of the siphon of the bivalve Donax serra
(Roding), Trans. Roy. Soc. S. Afr. 44, 489-498.

Hopkins SH (1958) Trematode parasites of Donax
variabilis at Mustang Island, Texas, Publ. Inst. mar~
Sci. Univ. Tex. 5: 301-31l.
Hornell J (1951) Indian Molluscs, The Bombay
Natural History Society.
Hornung H and Oren OH (1981) Heavy metals in
Donax trunculus (Bivalvia) in Haifa Bay,
Mediterranean (IsraeJ), Mar. Environ. Res. 4(3): 195-
202.
Hughes DA (1966) Behavioural and ecological
investigations of the crab Ocypode ceratophthalmus
(Crustacea:Ocypodidae), J. Zool. Lond., 150, 129-143.
Von Ihering H (1900) The musculus cruciformis of
the order Tellinacea, Proc. Acad. Nat. Sci.
Philadelphia. Vol.52, 480-48l.
Irwin TH (1973) The intertidal behaviour of the
bean clam, Donax gouldii Dall, 1921, Veliger. 15(3),
206-212.
Jacobson MK (1955) Observations on Donax fossor
Say at Rockaway beach, New York, Nautilus, Philad.
68, 73-77.
Johnson PT (1966a) Mass mortality in a bivalve
mollusc, Limnol. Oceanogr. 1: 429-43l.
Johnson PT (1966b) On Donax and other sandy
beach inhabitants, Veliger. 9: 29-30.
Johnson PT (1968) Population crashes in the bean
clam, Donax gouldi, and their signficance to the
study of mass mortality on other marine
invertebrates, J. invert. Pathol. 12(3), 349-358.
Jones DA (1979) The ecology of sandy beaches in
Penang, Malaysia, with special reference to
Excirolana orientalis (Dana), Estuarine and coastal
marine Science, 9, 677-682.
Koepke HW and Koepke M (1952) Sobre el proceso
de transformacion de la materia organica en las
playas avenosas del Peru. Revista de Ciencias Univ.
Nac. San Marcos, 54, (479-480), 5-29. Publ. Mus. Hist.
Nat. "Javier Prado" Ser. A. Zoologica. No.8, 1-25.
Krishnamurthy S Seshadri Rand Ramamurthy UD
(1966) Role of microbes in the nutrition of some
estuaries of marine bivalve, J. mar. bioi. Assoc.
India. 8(2), 231-235.
Kwei EA (1966) The spawning and the growth of
the Volta oyster, Egeria radiata (Lam.), Ghana J. Sci.
5: 150-160.
Leber KM (1982) Bivalves (Tellinacea; Donacidae)
on a North Carolina beach: Contrasting population
size structures and tidal migrations, Mar. Ecol. Proj.
Ser., 7, 297-30l.
Lebour MV (1907) Larval trematodes of the
Northumberland coast, Newcastle Trans. Nat. Hist.
Soc., Ser. 2, 1, (437-453).
Lever J and Thijssen R (1968) Sorting phenomena
during the transport of shell valves on sandy beaches
studied with the use of artificial valves, Symp. zool.
Soc. Lond. 22: 259-27l.
Lever J Kessler A van Overbeeke AP and Thijssen
R (1961) Quantitative beach research. II. The "hole
effect". A second mode of sorting of lamellibranch
valves on sandy beaches, Neths. J. Sea Res., 1, 339-
358.
631
Lever J Van Den Bosch M Cook H T Van Dijk AJH
Thiadens SJ and Thijssen R (1964) Quantitative
beach research. III. An experiment with artificial
valves of Donax vittatus, Neths. J. Sea Res., 2, 458-
492.
Lewin J Eckman JE and Ware GN (1979) Blooms of
surf-zone diatoms along the coast of the Olympic
Peninsula, Washington. XI Regeneration of
ammonium in the surf environment by the Pacific
Razor Clam Siligua patula, Mar. Bioi., 52, 1-9.
Lingwood PF (1976) Moribund macrofauna of the
S. W. Lancashire coast 1971-1975, Publ. Lancashire
and Cheshire fauna Soc. 69, p6-7.
Loesch HC (1957) Studies of the ecology of two
species of Donax on Mustang Island, Texas, Publ. Inst.
mar. Sci. Univ. Texas 4 No.2: 201-227.
Lucas A (1965) Recherche sur la sexuali te des
mollusques bivalves, Bull. bioi. Fr. Belg. 99(2): 115-
247.
Macnae Wand Kalk M (1962) The fauna and flora
of the sand flats at Inhaca Island, Mozambique, J.
animo Ecol. 31, 93-128.
Mahieu G de Mascitti G and Jaffe K (1981) Effecto
del petroleo crudo sobre los moluscos comerciales
litorales Donax denticulatus y Crassostrea
rhizophorae in Venezeula, Proc. Gulf. Carib. Fish.
Inst., 33, 125-139.
Mane UH and Talikhedkar PM (1976) Respiration
of the wedge clam Donax cuneatus, Indian J. mar.
Sci. 5(2), 243-246.
Masse H (1963) Quelques donnees sur I'economie
alimentaire d'une biocoenose infrali ttorale, Rec.
Trav. St. Mar. End. Bull 31 Fax 47,153-166.
Masse H (1968) Etude des variations de la
biomasse dans une biocoenose infralittorale de
substrat meuble, Rec. Trav. St. Mar. End. Bull. 43
Fax 59. 25-3l.
Masse H (1971) Contribution a I'etude de la
macrofaune de peuplements des sables fins
infralittoraux des c6tes de Provence. II La baie du
Prado (Golfe de Marseille), Tethys 3, 113-158.
Masse H (1972a) Contribution a I'etude de la
macrofaune de peuplements des sables fins
infralittoraux des Cotes de Provence V. La cote de
Camargue, Tethys 3, 539-568.
Masse H (1972b) Contribution a I'etude de la
macrofaune de peuplement des sables fins
infralittoraux des Cotes de Provence. 6 -Donnees sur
la biologie des especes, Tethys 4(1) 63-84.
Mauri M (1980) Formazione di nefroliti contenenti
manganese in Donax trunculus L. (Bivalvia), Mem.
Bioi. Marina and Oceanogr. Suppl X. 235-240.
Mauri M and Orlando E (1979) Accumulo di Fe,
Mn, Cu, Zn, nei tessuti di Donax trunculus L.
(Bivalvia), Quad. Lab. Tecnol. Pesca, 3(1 suppl.) 643-
648.
Mauri M and Orlando E (1982) Experimental study
on renal concretions in the wedge shell Donax
trunculus L, J. expo mar. Bioi. Ecol. 63,47-57. - - -
Mauri M and Orlando E and Falcone M (1978) Dati
preliminari sull' accumulo di metalli pesanti in Donax
trunculus L. (Bivalvia), Boll. Zool. 45(2), 226. - - -
632
McKay OW (1972) Mass mortality of Donax
vittatus (de Costa) at Sandend Bay, Banffshire,
Conchologist Newsl. No.43 284-285.
McLachlan A (1979) Growth and production of
Donax sordidus Hanley on an open sandy beach in
Algoa Bay, S. Afr. J. Zool., 14, 61-66.
McLachlan A (1980) Exposed sandy beaches as
semi-enclosed ecosystems, Marine Environ. Res. 4,
59-63.
McLachlan A and Hanekom (1979) Aspects of the
biology, ecology and seasonal fluctuations and
biochemical composition of Donax serra in the East
Cape, S. Afr. J. Zool., 14, 183-193.
McLachlan A and Van Der Horst G (1979) Growth
and production of two molluscs from an exposed
sandy beach, S. Afr. J. Zool., 14, 194-20l.
McLachlan A and Lewin J (1981) Observations on
surf phytoplankton blooms along the coast of South
Africa, Botanica Marina, 24, 553-557.
McLachlan A and Young N (1982) Effects of low
temperature on the burrowing rates of four sandy
beach molluscs, J. expo mar. Bioi. Ecol. 65, 275-284.
McLachlan A Cooper C and Van Der Horst H
(1979a) Growth and production of Bullia rhodostoma
on an open sandy beach in Algoa Bay, S. Afr. J.
Zool., 14, 49-53.
McLachlan A Wooldridge T and Van Der Horst G
(1979b) Tidal movements of the macrofauna on an
exposed sandy beach in South Africa, J. Zool. Lond.
188, 433-442.
McLachlan A Wooldridge T Schramm M and Kuhn
M (1980) Seasonal abundance, biomass and feeding of
shore birds on sandy beaches in the eastern cape,
South Africa, Ostrich, 51, 44-52.
McLachlan A Erasmus T Dye AH Wooldridge T Van
Der Horst G Rossouw G Lasiak T A and Mcgwynne L
(1981) Sand beach energetics. An ecosystem
approach towards a high energy interface, Estuarine,
Coastal and Shelf Sci., 13, 11-25.
McLusky 0 and Stirling A (1975) The oxygen
consumption and feeding of Donax incarnatus and
Donax spiculum from tropical beaches, Compo
Biochem. Physiol. A 51(4), 943-947.
McLusky 0 Nair SA Stirling A and Bhargava R
(1975) The ecology of a central West Indian beach
with particular reference to Donax incarnatus, Mar.
Bioi. Berlin. 300), 267-276.
Mikkelsen PS (1981) A comparison of two Florida
populations of the coquina clam, Donax variabilis
Say, 1822. (Bivalvia: Donacidae) I. Intertidal
density, distribution and migration,. Veliger, 23, 230-
239.
Moment GB (1962) Reflexive Selection: a possible
answer to an old puzzle, Science, N.Y. 136, 202-203.
Mori S (1938) Characteristic tidal rhythmic
migration of a mussel, Donax semignosus Dunker, and
the experimental analysis of its behaviour at the
flood tide, Zool. Mag. Tokio. 50, 1-12.
Mori S (1950) Characteristic tidal rhythmic
migration of a mussel, Donax semignosus Dunker and
the experimental analysis of its behaviour, Zool. Mag.
Tokio. 59, 87-89.
Morrison JPE (1971) Western Atlantic Donax,
Proc. bioi. Soc. Wash. 83, 545-568.
Moueza M (1972) Contribution a I'etude de la
biologie de Donax trunculus L. (mollusque
lamellibranche) dans I'Algerois. Ethologie en baie de
Bou-Ismail, Tethys 4(3), 745-756.
Moueza M (1975) Contribution a l'etude de la
biologie de Donax trunculus L. (mollusque
lamellibranche) dans I'Algerois, These Doctor es
Sciences Nat., Univ. Aix-Marseille, 196 pp.
Moueza M (1976) Contribution a I'etude de la
biologie de Donax trunculus L. (mollusque
lamellibranche) mechanisme de la nutrition, Journal
moll. Stud. 42(1), 17-30.
Moueza M and Chessel 0 (1976) Contribution a
I'etude de la biologie de Donax trunculus L.
(mollusque lamellibranche) dans l'Alg~rois: analyse
statistique de la dispersion Ie long d'une plage en baie
de Bou Ismail, J. expo mar. Bioi. Ecol. 21(3), 211-
22l.
Moueza M and Frenkiel L (1974) Contribution a
l'etude des structures palleales des Tellinacea.
Morphologie et structure de manteau de Donax
trunculus L, Proc. malac. Soc. Lond., 41, 1-20. - - -
Moueza M and Frenkiel L (1976) Contribution a
I'Btude de la biologie de Donax trunculus L.
(moll usque lamellibranche) morphologie fonctionelle
- anatomie microscopique - histologie de l'appareil
digestif, J. moll. Stud. 42(1), 1-16.
Moueza M and Frenkiel L (1976) Premiers donnees
sur l'ultrastructure de I'organe sensoriel du muscle
cruci forme de Donax trunculus L. (Moll usque
lamellibranche Tellinacea), Archives zool. expo gen.
117(4) 485-492.
Moueza M and Frenkiel L (1978) Le systeme
circulatoire et Ie jeu des siphons chez Donax
trunculus L. (moll usque lamellibranche), Malacologia
17(1), 117-124.
Moueza M and Frenkiel-Renault L (1973)
Contribution a I'etude de la biologie de Donax
trunculus L. (mollusques lamellibranches) dans
I'Algerois: la reproduction, Cahiers BioI. mar. 140)
261-283.
Nagabhushanam Rand Talikhedkar PM (1975)
Neurosecretory cells in the cerebral and visceral
ganglia of Donax cuneatus, Bull. Dept. Mar. Sci.
Univ. Cochin, VII(2), 447-453.
Nagabhushanam Rand Talikhedkar PM (1977a)
Reproductive biology of the wedge clam Donax
cuneatus, Indian J. mar. Sci. 6(1), 35-38. ---
Nagabhushanam Rand Talikhedkar PM (1977b)
Seasonal variations in proteins, fat and glycogen of
the wedge clam Donax cuneatus, Indian J. mar. Sci.
6(1), 85-87.
Nair A (1978) Production ecology of a tropical
sandy beach at Benaulim, Goa. Indian J. mar. Sci.
70), 163-167.
Nair A Dalal SG and Ansari ZA (1978) Growth of
the bean clam Donax incarnatus Gmelin from a sandy
beach at Benaulim, Goa, Indian J. mar. Sci. 7(3),
197-199.
Nair NB and Ansell AD (1968) Characteristics of
penetration of the substratum by some marine
bivalve molluscs, Proc. malac. Soc. Lond., 38, 179-
197.
 633

Nakazima M (1965) On the differentiation of the Petrusewitz K (1966) Suggested list of more
crenated S fdds in the midgut-gland of important concepts in productivity studies
Eulamellibranchia. (IIII) crenated fold in Donax (definitions and symbols) in Secondary productivity of
semigranosus, Venus 23- ~18-222. terrestrial ecosystems, 1, 51-58. Warszawa-Karkow
Narchi W (1972) L' the biology of Iphigenia R.W.N.
brasiliensis Lamarck, 1~18 (Bivalvia, Donacidae), Phillips T (1970) Notes on predation of the
Proc. malac. Soc. Lond. 4U(2), 79-91. asteroid genus Luidia on Donax navicula at Playa
Narchi W (1974) Aspects ecologicos e adaptativos Almejas, Baja California del Norte, Mexico,
de alguns bivalves do litoral poulista, Papeis Avuls. Tabulata. 3(3): 16-18.
Zool. S. Paulo 27(19), 235-262. Pichon M (1967) Contribution a l'etude des
Narchi W (1978) Functional anatomy of Donax peuplements de la zone intertidale sur sables fins et
hanley anus Philippi 1847 (Donacidae-Bivalvia), sables vaseux non fixes dans la region de Tulear, Rec.
Boletim Zool. Univ. S. Paulo 3, 121-142. Trav. Stn. mar. Endoume (fasc. hors. ser. suppI). 7,
Nayar KN (1954) Studies on the growth of the 57-100.
wedge clam, Donax (Latona) cuneatus Linnaeus, Pichon Y Moueza M and Frenkiel L (1978)
Indian J. Fish, 2, 325-348. Physiologie de l'organe sensoriel de Donax trunculus
Neff JM Carr RS and McCulloch WL (1981) Acute L. (MolI.Lam. Tellinacea), J. Physiol, 74, 4, 9A.
toxicity of a used chrome lignosulfonate drilling mud Pichon Y Moueza M and Frenkiel L (1980)
to several species of marine invertebrates, Mar. Mechanoreceptor properties of the sense organ of the
Environ. Res. 4(4), 251-266. cruciform muscle in a tellinacean lamellibranch,
Negus M (1975) An analysis of boreholes drilled by Donax trunculus: An electro-physiological approach,
Natica catena (da Costa) in the valves of Donax Mar. BioI. Lett. 1(5), 273-284.
vittatus (da Costa), Proc. malac. Soc. Lond. 41(4), Pohlo RH (1967) Aspects of the biology of Donax
353-356. gouldi and a note on evolution in Tellinacea
Nickles M (1950) Manuels Ouest-African. (Bivalvia), Veliger. 9: 330-337,
Mollusques Testaces marins de la Cote occidental Pohlo RH (1969) Confusion concerning deposit
d'Afrique, Paris, Lechevalier. feeding in the Tellinacea, Proc. Malacol. Soc. Lond.,
Nolte DJ (1954) Polymorphic coloration in Donax, 38, 361-364.
Caryologia 6 Suppl. Atti IX Congr. Internat. Genet. Poli G (1972) Cycle reproducteur de Mollusques
pt.2. 889-890. Lamellibranches et Gasteropodes du Bassin
Nolte DJ (1958) Genetic drift in populations of a d'Arcachon, These Doct. Specialite, Univ. Paris VI,
sand mussel, Publ. Univ. Pretoria N.R. No.7. 12-15. 230pp.
Olivier SR Capezzani DDA Carreto JJ Christiansen Ponder WF Colman PH Yonge CM and Colman MH
HE Moreno VJ Moreno JEA and Penchaszadeh PE (1981) The taxonomic position of Hemidonax, with a
(1971) Estructura de la comunidad, dinamica de la review of the genus (Bivalvia, Cardiacea), J. Malac.
poblacion y biologia de la almeja amarilla Soc. Aust. 5(1/2), 41-64.
(Mesodesma mactroides Desh. 1854) en Mar Azul Purchon RD (1958) The stomach in the
(Pdo. de Gral. Madariaga, Bs. As., Argentina), Eulamellibranchia: stomach type IV, Proc. Zool. Soc.
Projecto Desar. Pesquero, F.A.O. Ser. info Tec. 27, 1- Lond., 131,487-525.
90. Purchon RD (1960) The stomach in the
Orlando E and Mauri M (1978) Accumulation of Eulamellibranchia: stomach type IV and V, Proc.
manganese in Donax trunculus L. (Bivalvia), IV es Zool. Soc. Lond., 135, 431-489.
Journees. Etud. Pollutions, pp279-299. Purchon RD (1963) A note on the biology of Egeria
Orton JH (1929) Severe environmental mortality radiata Lam. (Bivalvia: Donacidae), Proc. malac.
among Abra (= Syndosmya) alba, Donax vittatus and Soc. Lond. 35: 251-271.
other organisms of the Lancashire coast, Nature. Rahaman AA (1965) Nitrogen content of the
London. 124 1929 p.911. lamellibranch Donax cuneatus Linnaeus, Current Sci,
Pearse AS Humm HJ and Wharton GW (1942) 34, 217-218.
Ecology of sand beaches at Beaufort, North Carolina, Rahaman AA (1965) The chemical composition of
Ecol. Monogr., 12, 135-190. the Lamellibranch Donax cuneatus Linn, Proc. Indian
Pelseneer P (1896) Un trematode produisant la Acad. Sci. 62B: 188-194.
castration parisitaire chez Donax trunculus, Bull. Sc. Rao KS (1967) Annual reproductive cycle of the
Fr. Belg., 27, 357-363. wedge clam Donax cuneatus Linnaeus, J. mar. bioI.
Pelseneer P (1906) Trematodes parasites de Ass. India 9(1), 141-146.
mollusques marins, Bull. Sci. France. Belgique. Paris. Rao SR and Kutty MN (1969) Resistance to
40, 116-186. dessication and oxygen debt in wedge clams, Proc.
Penchaszadeh PE and Olivier SR (1975) Ecologia Symposium of Mollusca Marine Biological Ass. India.
de una problacion de "berberecho" (Donax hanley anus) Pt. II, 595-606.
en Villa Gesell, Argentina, Malacologia 15(1), 133- Rees CB (1950) The indentification and
146. classification of Lamellibranch larvae, Hull. Bull.
mar. Ecol. 3, 73-104.
Reid RGB (1965) The structure and function of the
stomach in bivalve molluscs, J. Zool., 147, 156-184.
634
Roman M-L (1974) Etude d'une plage d'arriere-
mangrove dans la region Tulear (sud-ouest de
Madagascar) Migrations tidales des especes
caracteristiques de I'etage mediolittoral, Tethys, 5(2-
3),449-467.
Sigurdsson JB Titman CW and Davies PA (1976)
The dispersal of young post-larval bivalve molluscs by
byssus threads, Nature, Lond., 262, 386-387.
Smith DAS (1971) Polymorphism and population
density in Donax rugosa (Lamellibranchiata:
Donacidae), J. Zool. Lond., 164, 429-441.
Smith DAS (1975) Polymorphism and selective
predation in Donax faba Gmelin (Bivalvia,
Tellinacea), J. expo mar. BioI. Ecol. 17(2), 205-219.
Steele JH (1976) Comparative studies of beaches,
Phil. Trans. R. Soc. (B) 274(934) 401-415.
Stefan M (1962) Recherches sur I'activite
periodique motrice de quelques Lamellibranches
marines, Stud. Cercet. BioI. Acad. R.P.R. Ser. BioI.
Anim. 14: 241-251.
Stenner RD and Nickless G (1975) Heavy metals in
organisms of the Atlantic coast of south west Spain
and Portugal, Mar. Pollut. Bull., 6(6), 89-92.
Stoll E (1937) Beobachtungen uber die
Fortbewegung bei einigen grabenden muscheln, Rev.
suisse. Zool. Geneve. 44, 1937, 383-390.
Stoll E (1947) Sur Ie mode de locomotion des
quelques mollusques marins, Trav. sta. bioI. Roscoff,
Paris No.16, 1938, 5-29.
Talavera F and Faustino LA (1933) Edible mollusks
of Manila, Philipp. J. Sci. Manila. 50, 1, 1-48.
Talikhedkar PM and Mane UH (1977) Rate of
water filtration by the wedge clam, Donax cuneatus,
Marathwada Univ. J. Sci., (Biol.Sci.) 16(19), 117-122.
Talikhedkar PM Mane UM and Nagabhushanam R
1976 (1978) Growth rate of the wedge clam Donax
cuneatus at Miriya Bay, Ratnagiri, Indian J. Fish.
23(1-2), 183-193.
Taylor JD (1968) Coral reefs and associated
invertebrate communities (mainly molluscs) around
Mahe, Seychelles, Phil. Trans. R. Soc. (B). 254: 129-
206.
Tiffany WJ (1971) The tidal migration of Donax
variabilis Say (Mollusca: Bivalvia), Veliger 14(1) 82-
85.
Tinbergen L (1960) The natural control of insects
in pinewoods. I. Factors influencing the intensity of
predation by song-birds, Archs. need. Zool., 13, 265-
236.
Trevallion A Ansell AD Sivadas P and Narayanan B
(1971) A preliminary account of two sandy beaches
in South West India, Mar. BioI. (Berlin) 6(3): 268-279.
Trueman ER (1967) Activity and heart rate of
bivalve molluscs in their natural habitat, Nature,
Lond. 214: 832-833.
Trueman ER (1968) A comparative account of the
burrowing process of species of Mactra and of other
bivalves, Proc. Malacol. Soc. Lond. 38(Pt.2) 139-151.
Trueman ER (1971) The control of burrowing and
the migratory behaviour of Donax denticulatus
(Bivalvia, Tellinacea), J. Zoo I. Lond., 165, 453-469.
Trueman ER and Ansell AD (1969) The
mechanisms of burrowing into soft substrata by
marine animals, Oceanogr. Mar. BioI. Ann. Rev., 7,
315-366.
Trueman ER Brand AR and David P (1966) The
dynamics of burrowing of some common littoral
bivalves, J. expo BioI. 44: 4ti9-492.
Turner HJ Jnr and Belding DL (1957) The tidal
migrations of Donax variabilis Say., Limnol.
Oceanogr. 2: 120-124.
Velloso KP Beck Hand Wiedermann RO (1951)
Molluscos Comestiveis da costa riograndense Donax
hanleyana, Anais. assoc. quinn. Brazil. 10, 216-2~
de Villiers G (1975a) Reproduction of the sand
mussel Donax serra Roding, Invest. Rep. Sea Fish.
Brch. Repl. S. Afr. 102, 1-33.
de Villiers G (1975b) Growth, population dynamics,
a mass mortality and arrangment of white sand
mussels, Donax serra Roding, on beaches in the
south-western Cape Province, Invest. Rep. Sea Fish.
Brch. Repl. S. Afr. 109, 1-31.
de Villiers G (1979) Recovery of a population of
white mussels Donax serra at Elands Bay, South
Africa, following a mass mortality, Fish. Bull. S.
Afr. 12,
Wade BA (1965) Notes on the ecology of Donax
denticulatus (Linne), Proc. Gulf Caribb. Fish Inst.
17: 36-4l.
Wade BA (1967) Studies on the biology of the West
Indian beach clam, Donax denticulatus Linne. l.
Ecology, Bull. Mar. Sci. 17: 149-174.
Wade BA (1967) On the taxonomy, morphology and
ecology of the beach clam, Donax striatus Linne,
Bull. Mar. Sci. 17: 723-740.
Wade BA (1968) Studies on the biology of the
West-Indian beach clam, Donax denticulatus Linne. 2.
Life history, Bull. Mar. Sci. 18: 876-90l.
Wade BA (1969) Studies on the biology of the West
Indian beach clam, Donax denticulatus Linne. 3.
Functional morphology, Bull. Mar. Sci., 19, 306-322.
Warwick RM George CL and Davies JR (1978)
Annual macrofauna production in a Venus community,
Estuar. coastal mar. Sci., 7, 215-241.
Williams AB and Porter HJ (1971) A ten-year
study of meroplankton in North Carolina estuaries:
occurrence of postmetamorphal bivalves, Chesapeake
Sci. 12(1): 26-32.
Wolcott TG (1978) Ecological role of ghost crabs,
Ocypode guadrata (Fabricius) on an ocean beach:
Scavengers or predators, J. expo mar. BioI. Ecol., 31,
67-82.
Yoloye V (1977) The biology of Iphigenia truncata
(Monterosato) (Bivalvia, Tellinacea), Malacologia
16(1), 295-301.
Yonge CM (1949) On the structure and adaptations
of the Tellinacea deposit-feeding Eulamellibranchia,
Phil. Trans. London. 234B, 609, 29-76.
Yonge CM (1957) Mantle fusion in the
lamellibranchia, Publ. Staz. Zool. Napoli, 29, 151-
171.
 635

Zakhvatkina KA (1959) Lichinki duustvorchatykh

molliuskov sevastopol'skogo rainona Chernogo Moria.
(larvae of bivalves mollusks of the Sevastopol Region
of the Black Sea), Akad. Nauk. S.S.S.R. Trudy
Sevastopol'skoi Biologicheskoi Stantsii, 11, 108-152.
Zammit VA and Newsholme EA (1976) The
maximum activities of Hexokinase, phosphorylase,
phosphofructokinase, glycerol phosphate
dehydrogenase, lactate dehydrogenase, octopine
dehydrogenase, phosphoenolpyruvate carboxykinase,
nucleoside diphosphokinase, glutamate-oxyaloacetate
transaminase and arginine kinase in relation to
carbohydrate utilization in muscles from marine
invertebrates, Biochem. J. 160,447-462.
Zhirmunsky AV (1963) A comparative study of
cellular thermo-stability of marine invertebrates in
relation to their geographical distribution and
ecology, In Troshin AS and Prosser Cl, eds. The Cell
and Environmental Temperature. london. Pergamon
Press.
Zhirmunsky AV and Chu l-c (1963) The cell
thermostability of sympatric species of Donax in
relation to the temperature conditions Dftheir
habitat, Acta zool. Sinica 15: 21-27.

CONSUMPTION, ASSIMILATION AND ENERGY BALANCE IN THE THREE-SPOT SWIMMING CRAB,
OVALIPES PUNCTATUS (DE HAAN) (CRUSTACEA~
H.H. DU PREEZ (Department of Zoology, University of Port Elizabeth, P.O. Box 1600,
Port Elizabeth 6000, South Africa)
INTRODUCTION
The swimming crab Ovalipes punctatus (de
Haan) is commonly found on the sandy
beaches of the south and east coasts of
South Africa. These crabs prey exten-
sively on the bivalve and gastropod
populations which occur on these
beaches. An energy flow diagram has
been constructed for the sandy beach
macrofauna of the eastern Cape but
mollusc production removed by the
swimming crab Ovalipes was only roughly
estimated (McLachlan et al., 1981).
Since the oxygen consumption and
reproductive cycle have been determined
(Du Preez 1982a, Du Preez, McLachlan
1982c) it is possible to estimate growth
production if the consumption and
assimilation efficiency of the crabs are
knov.'
w
o
LO
C\I
FIGURE 1. Map of study area showing the position of Kings Beach.
637
BRACHYURA)
Energy flow through different species
populations provides a quantitative
basis for studying the role of each
species within a community. The energy
gain from the food consumed (C) is used
for metabolism (R) and production (P),
while excretory products (including
exuviae) (U) and the non-digested food
(F) are lost to the organism (Crisp,
1971). These values are represented in
the energy budget equation:
C = P + R + U + F
and can be applied to an individual
animal, a single species population or
to all organisms in a multihabitat eco-
system (Crisp, 1971).
In this study the food consumption,
faecal production, assimilation effi-
638
ciency and "scope for growth" for Q.
punctatus were investigated and an
energy budget for an average sized crab
was derived.
MATERIAL AND METHODS
During February 1980, 26 crabs were
collected from Kings Beach (Fig. 1) and
kept in SO 1 glass tanks with a bottom
filter and a sandy substrate. Two
animals were kept separated by a glass
partition in each tank. The water was
kept constant at 19 + O.SoC and a
salinity of 3S o / 00 .
For the consumption experiments the
crabs were starved for 4Bh so that all
crabs would be in approximately the same
state of hunger. The crabs were then
fed on live sand mussels, Donax ~,
of 3S + Smm total length. Each day the
number of D. serra eaten was noted and
if any soft flesh remained its dry mass
was determined (24h at 90°C). For the
duration of the experiment the number of
D. serra available was kept at seven.
After seven days the experiment was
terminated and the carapace width (CW)
of each crab was measured.
Using 1) the following equation to cal-
culate dry tissue mass of ~. ~:
LoglO [mass (mg)] = 3.07 10glO [length
(mm)] -2.1 •.••• (1)
2) an energy content of 19.14kJ/g
for ~. ~ (McLachlan, Hanekom, 1979)
and 3) calculating the dry mass of the
crabs using the following equation (Du
Preez, McLachlan, 19B2a). Dry mass (g)
= 4.74 x 10-S (CW (mm» 3.09 ..•... (2)
the consumption in kJ/day was calculated
for a crab of a given mass.
To determine the assimilation effi-
ciency, crabs (size range 40-60mm CW)
were starved for 4Bh and then given a
freshly opened~. serra. After each
crab had finished eating, the Donax was
removed, measured and the dry mass of
the remaining flesh determined (24h at
90°C). The crab was then removed and
placed in a 10 1 glass tank with sea
water (IS + SoC, 3S o / 00 ). The faeces
produced were removed every 6h until 4Bh
had elapsed. The faeces were placed on
pre-weighed filter paper and dried (24h
at 90°C) and re-weighed.
To determine the energy (microbomb
calorimeter) and ash (Sh combustion in a
muffle furnace at 4S0°C) content of the
faeces, the crabs were fed on D. serra
and the faeces collected and dried (24h
at 90°C) until a sufficient amount was
obtained.
The assimilation efficiency was calcu-
lated using:
1) the method described by Conover
(1966) and Griffiths, King (1979):
assimilation efficiency (%) =
Fl - El
(1 - E2) Fl x 100 .........•.• (3)
where Fl = organic fraction of food
and El = organic fraction of faeces.
2) the method described by Crisp (1971):
assimilation efficiency (%) =
ACWC - AFWF
cWc x 100 (4 )
where Ac is the energy content and Wc
the weight of food ingested and AFWF the
corresponding values for the faeces
produced.
The respiration rates at 19°C were cal-
culated using the following equation
 639

from Du Preez (1982a): log M ( g02/g/h)

= 1. 05 lono temp. (OC) - 0.44 lono
mass (g) + 1.37 •.•.••.••.••.••..•• (5)
and were expressed as kJ/day using the
following equations from Du Preez,
McLachlan (1982a)
wet mass (g) O. 70e O• 075 (CW( mm))) .. (6)
Dry mass (g) 4.74xlO- 5 (CW(mm))3.09 (7)
and assuming 1 1 02 respired = 20.06kJ 20
consumed ..•..••.•••.•••••••.•.••• (8)

The mean consumption rate at 19°C

expressed in kJ/day is given in Fig. 2 15
and described by:
consumption (kJ/day) 1.50 (Dry mass
(g) of crab)0.815 (r 0.99; n = 18; I

p<0.005) •••...•.•••••••••••••••.•• (9) >- 10

~
"0

RESULTS
Of the food consumed, 5% was egested as
faeces. The faeces had energy and ash 5
contents of 12.75 (kJ/g) and 61% respec-
tively. Using these values and the
energy (19.14 kJ/g) and ash (17%) values
given for Q. ~ (McLachlan, Hanekom, O2
4 6 8 10 12 14 16 18
1979) the assimilation efficiencies were
calculated as 96.6% using the method of
Crab dry mass (9)
Conover (1966), Griffiths, King (1979)
FIGURE 2. Mean consumption (~ SD) of D. serra
and 94.4% using the methods described by by an individual O. punctatus of a range ~
Crisp (1971). The assimilation effi- sizes in captivity.
ciencies of crabs over a wide mass range
are very similar with no significant during respiration by a crab of a given
differences. dry mass at 19°C the equation:
Respiration (kJ/day) 0.48 (dry mass
Using the average (95%) of the calcu- (g) of crab)0.585 (r 0.99; n = 18;
lated assimilation efficiencies and the p<0.005) •.•.•.••••••.••.••..•••.•• (11)
mean consumption values (from Fig. 2) was derived from data given by Du Preez
the following equation was derived: 1982a) .
assimilation (kJ/day) = 1.45 (dry mass
(g) of crab)0.812 (r = 0,93, n = 18, The above equations are graphically
P<0.005) •..••..•...•.....••.••.•. (10) represented in Fig. 3 where the shaded
area represents the "scope for growth".
For the amount of energy (kJ/day) spent The term "scope for growth" describes
 640
the nett change in energy content over a
time period and is determined by sub-
tracting the metabolic energy expenditure
from the assimilated energy and thus
gives .an indication of the energy avail-
able for production without specifying
the type of production (Griffiths, King,
1979) .
20
1
...
I 90% assimilation efficiency. This value
>0.
~ 1
..,
~
5
O+---~-----r----~---r----~--~----~--~
4 8 10 12 14 16
Dry mass (g)
FIGURE 3. The consumption ( -.. - ) , assimila-
tion (--------), oxygen consumption (----)
and "scope for growth" in O. punctatus expressed
as a function of crab dry mass.
Ouring feeding, food particles may be
lost to the water, while leaching may
occur from the food and the faeces.
This influences .thees.timation of
assimilation efficiency (Clark, 1979).
However, Q. ~. has a low ash content
(17%) and thus would have a high
digestability. A high assimilation
efficiency may therefore be expected.
Assimilation efficiencies greater than
90% have been reported for other
Crustacea. These include the marine
isopod Glyptonoeus antarticus(95%)
(Clarke, 1979) and the crab, 1979). How-
ever, Klein Breteler (1975) recorded that
Carcinus maenas digested only 44% of the
food consumed.
CONCLUSIONS
Specimens of Q.punctatus which occur in
the surf zone have a mean carapace width
of 49. Smm. The energy budget for .an
average surf zone crab of this size can
be calculated using the appropriate
equations. Consumption was calculated
from equation (9) and assimilation and
faecal losses were calculated assuming
was used because it is felt that 94-96%
efficiency is an overestimation due to
difficulties in collecting all the
faeces, leakage from faeces and over-
estimation of consumption (because some
of the food is wasted when devoured).
The amount of energy lost to excretion
was calculated, assuming that 1% of the
energy consumed is lost as nitrogen
1 (Klein Breteler, 1975) and that the
crabs moult twice a year, the energy of
the exuviae being 2.94kJ/g (Ou Preez,
Mclachlan 1982b, d).
The energy expended on reproduction was
calculated from foregoing data, first
using the following equation from Ou
Preez, McLachlan (1982c):
Brood wet mass (g) = 0,001 (CW(mm»0.80
(r = 0.89, n = 30, p<0,005) ..•..• (12)
to estimate brood mass. By correcting
for 44% water and using an energy
content of 14,17kJ/g dry mass (Ou Preez,
McLachlan, 1982c) the energy in one
brood was estimated. Reproductive
output was then calculated, assuming
that two broods are produced during a
year by each female, but that there is

no reproductive output by males and that
67% of the animals on the beaches are
females (Ou Preez, McLachlan, 1982c).
Metabolic expenditure was calculated
using equations (5),..(8) and assuming
that the oxygen consumption experienced
in the field was three times higher than
the mean measured oxygen consumption
(i.e. approximately the same as routine
oxygen consumption). The energy
expended on growth was calculated by
difference. However, growth expenditure
can also be calculated as the difference
in energy content of crabs one moult
larger and one moult smaller than 49.5mm
carapace width using the growth and
energy data given by Ou Preez, McLachlan
(l982b, d). By this method the
estimated energy expended on growth is
90 kJ.crab-l.y-l which is much lower
than the value of 803 kJ.crab-l.y-l
calculated. This latter high value is
probably mainly due to accumulated expe-
rimen.tal errors in determining the
various parameters of the energy budget
as it is unlikely that Pg could be much
higher than 110 kJ.crab-l.y-l.
The consumption studies were performed
under optimal feeding conditions in the
laboratory after the crabs were starved
for 48h. The actual consumption value
of 3033 kJ.crab- l .4- l may be as much as
25% too high. Furthermore, predation
by the crabs on the beach is affected by
the prey species availability and the
population structure of the prey species
present (Ou Preez, 1982b). It is known
that Ovalipes prefer the prey species
least available in situ (Q. ~ at
Kings Beach, ~. rhodostoma at Maitlands
River Beach), thus also limiting con-
641
sumption. The values calculated for F
(303kJ. crab-l.y-l) are probably under-
estimates due to the methods used.
5--1"---('
22 7 128
227
88
1797
FIGURE 4. Energy flow diagram for an average size
o. punctatus (49,5 rom carapace width) in the surf
zone. All values in kJ.crab.- 1y-l.
Fig. 4 may be considered to be a better
estimation of the energy flow through an
2. punctatus population per metre shore-
line, assuming no mortality and one
crab per running metre (McLachlan et al.,
1981). There is however, mortality in
situ and sandsharks, rays and fish are
known to take Ovalipes. The energy
circuit diagram of McLachlan et al (1981)
showed that the production on sandy
beaches is mostly exported to birds (32%)
fish (55%) with a minor loss to man (3%),
benthic scavengers and "natural
mortality" (10%). Using the same pro-
duction values for the different prey
642
species, it was calculated that Ovalipes
consume 34% of production, thus
suggesting that the percentage of
sumption. The values calculated for F
scavengers was underestimated. However
the production of the other Bullia
species is not known and Ovalipes also
use them as prey. Because Ovalipes is
the most important benthic scavenger on
the beaches and is resident to the surf
zone, it may be concluded that O.
punctatus is capable of effectively
cropping a considerable amount of the
mollusc producti~n on these beaches.
ACKNOWLEDGEMENTS
I thank Dr A. Mclachlan for a stimulat-
ing discussion and useful suggestions on
the manuscript and the staff of the
Zoology Department, University of Port
Elizabeth for their assistance. I
gratefully acknowledge a post-graduate
bursary from the Council for Scientific
and Industrial Research. I thank Mrs A.
Gerber for typing and Miss M. Maree for
preparing the figures.
REFERENCES
Clark A (1979) Assimilation efficiency
of the Antartic marine isopod Glyptono-
tus antarciticus, Mar. BioI. 52,
157-160.
Conover RJ (1966) Assimilation of
organic matter by zooplankton, Linol.
Oceanogr. 11, 338-354.
Crisp OJ (1971) Energyflow measure-
ments In methods for the study of marine
benthos I.B.P. Handbook, 11, pp.197-299
Oxford and Edinburg, Blackwell Scien-
tific Publ.
Du Preez HH (1982a) The effects of
temperature, season and activity on the
respiration of the three-spot swimming
crab, Ovalipes punctatus (De Haan), In
Press.
Du Preez HH (1982b) Feeding in the
three-spot swimming crab Ovalipes
punctatus (De Haan) in the east Cape,
South Africa, In press.
Du Preez HH and A MCLachlan (1982a)
Biology of the three-spot swimming crab,
Ovalipes punctatus (De Haan) I. Morpho-
metrics and relative growth, In press.
Du Preez HH and A McLachlan (1982b)
Biology of the three-spot crab, Ovalipes
punctatus (De Haan) II. Growth and
moulting, In press.
Du Preez HH and A McLachlan (1982c)
Biology of the three-spot swimming crab,
Ovalipes punctatus (De Haan) I.II.
Reproduction, fecundity and egg develop-
ment, rn press.
Du Preez HH and A McLachlan (1983d)
Seasonal changes in Biochemical composi-
tion and energy content of the three-
spot swimming crab, Ovalipes punctatus
(De Haan) (Crustacea, Brachyura), In
press.
Griffiths CL and JA King (1979) Some
relationship between size food avail-
ability and energy balance. in the ribbed
mussel, Aulacomya alter, Mar. BioI. 51,
141-149.
Klein Breteler WCM (1975) Food con-
sumption, growth and energy metabolism
of juvenile shore crabs, Carcinus maenas,
Neth. J. Sea. Res. 9, pp.255-272.
McLachlan A and N Hanekom (1979)
Aspects of the biology, ecology and
seasonal fluctuations in biochemical
composition, of Donax serra in the East
Cape, S. Afr. J. Zool. 1.4, pp.183-193.
Mclachlan A, TErasmus, AH Dye, T
Wooldridge, G van der Horst, G Rossouw,
TA Lasiak and LE McGwynne (1981) Sand
beach energetics; an ecosystems approach
toward a high energy interface, Estuar.
Cstl. Mar. Sci. 13, pp.347-356.
Snedecar GW and WG Cochran (1967)
Statistical Methods, 593pp •. Ames Iowa:
Iowa State University Press.

POPULATION ECOLOGY AND BIOLOGY OF DOTILLA SULCATA (CRUSTACEA, OCYPODIDAE) TYPICAL FOR SANDY BEACHES OF
THE RED SEA
LEV FISHELSON (Professor of Zoology, George S. Wise Faculty of Life Sciences, Tel Aviv University, Tel
Aviv, Israel)
INTRODUCTION: In the course of evolution, the
adaptation of aquatic organisms to terrestrial life
has caused many changes in their morphology,
physiology and behaviour. As a result of the inva-
sion of land, large groups of species evolved
entirely independently of the aquatic environment.
Among the aquatic invertebrates a small group of
decapods crustaceans, living where the sea and land
merges, are unique in that they are really amphi-
bious, living part of their lives on the land and
the other part in the water. Some grapsid crabs
belong here in addition to several species of the
ghost crabs (Ocypode), species of Uca, Scopimera
and Dotilla, hermit crabs (Coenobita and Virgo)
and others. Within the Red Sea many of them are
found along sandy beaches and mangrove zones, and
thus characterize widespread communities
(Fishelson, 1971).
Figure 1. Dotilla sulcata
D. sulcata is a small ghost crab (Fig. 1) first
mentioned by Paulson (1875) from the Red Sea as
Dita sulcata. Here they form extensive popula-
tions (Magnus, 1960; Fishelson, 1971; Lewinsohn,
643
1977). This distribution forces Q. sulcata, a
terrestrially adapted marine crab, to face the
extreme conditions of a land-locked sea, in which
the temperature, rates of evaporation and salini-
ties are among the highest known from natural
marine systems.
The purpose of this study was to analyse the life
history of Dotilla with special emphasis on the
feeding dynamics, to assess their environmental
adaptations, their roles as food consumers and
sediment movers in this very specialised habitat.
MATERIALS AND METHODS: Field observations were
made with large binoculars (50 x 70); still photos
were taken with a Nikon F. camera. Films of 10
second duration taken at 3 minute intervals with a
16 mm Bolex camera recorded the activities at
various times and documented behavioural events.
The temperature in air and water as well as at
various depths of Dotilla burrows was measured
with a Rostrac multichannelled telethermometer.
Transect and quadrant techniques were applied to
measure the density of the populations, and sie-
ving the sediment and counting the collected ani-
mals provided additional data. Measuring the dia-
meter of the burrows gave an accurate estimation
of the maximum carapace width of its inhabitants
and these measurements were used to analyse the
distribution and composition of various popula-
tions. For observations on behaviour and circa-
dian rhythms in the laboratory, the animals were
kept for 18 months in sand-shore imitating boxes;
delicate threads were glued to their carapaces,
and the other ends glued to the writing levers of
644
a kymograph. Such crabs, isolated in a constant
temperature room, moved in their cages without
limitation and did not appear disturbed.
Experiments on desiccation were carried out at the
site where the crabs were collected, or in the
laboratory. Measurements of organic matter were
made by burning the sand sampled in an oven at
450 0 C.
HABITATS: We studied soft-bottom intertidal habi-
tats of three main types:
1. Sandy shores that stretch for many miles along
the sea, or form soft-bottom enclaves among and
between the rocky conglomerates. The sediments
on these shores are mainly terrigenous, formed by
degradation of Precambrian mountains or younger
sedimentary formations (Fishelson, 1977; Gvircman,
1978).
2. Mangrove flats are intertidal regions, mainly
with Avicenia ~ facing the sea and Salvadora
persica and other halophytes growing above the
supratidal. Such habitats are very common in the
southern Red Sea (Lewinsohn, Fishelson, 1967),
especially on the Dahlak Archipelago and around
Massawa in Ethiopia. In more northern localities
such habitats were investigated at Ras-Muhammad,
Nabeq and Shurat-al-Mankata of the Sinai Peninsula.
3. Sharems are bays usually found where dry river
beds (wadis) open toward the sea. Winter floods
occurring in these wadis carry and deposit large
amounts of silt and sand, forming a fan-like for-
mation with assorted size sediment.
Along these habitats the intertidal is modulated
by two main forces: the tidal excursion of the sea
and winds blowing along the shores. The semi-
diurnal tide ranges between 90 - i10 cm maximum,
but this predicted model is modified by pressures
produced by the Indo-Pacific monsoon winds.
During the winter months (December-April) these
winds force large amounts of water from the
Arabian Se~ into the Red Sea and its northern
Gulfs, causing a rise of about 20-25 cm above
average sea level. Contrary to this, during
summer (June-October) the monsoon trade winds blow
in the opposite direction and push large amounts
of water from the Red Sea through the Bab-el-Mandab
straits. This "escape" of water masses, together
with low tides and northern local winds, fre-
quently causes a drop in the average water level
of 25-30 cm. Due to these movements, the inter-
tidal in one part of the year moves upwards in-
vading the supratidal, and in the next half of the
year sinks to expose, for a longer period, a part
of the former infratidal, invading the subtidal
(Fishelson, 1973, 1977).
HUMIDITY AND TEMPERATURE: The Red Sea is in its
northern part situated between desert countries
of high temperature and very low precipitation.
Here the humidity along the sea-shore is usually
less than 25%, and in some cases close to 10%.
This causes high evaporation and salination of the
land. Toward the south, the Red Sea extends into
the sub-tropical belt of winter and summer wet
seasons. Here along the Ethiopian shores, humi-
dity can attain 85%-90%. Also the temperatures
along the Red Sea are different in the northern
and southern parts. In the south, these are more
stable and do not drop below 23 0 _25 0 C. Along the
northern Gulfs of the Red Sea the summer air
temperatures rise to 42 0 -46 0 C, but during the
winter (January-February) they may fall at night
to 10 0 or even 80 C, a range that must be very
harsh for many intertidal organisms.
ASSEMBLAGES OF ORGANISMS ON THE SAND SHORES: The
surface of the sand flats supports a dense popula-
tion of microforms, such as nematodes, polychaetes,
diatoms and foraminiferans. There are also many
harpacticoid copepods, especially their copepodids.
The highest number of organisms observed in 1 m2
was 20.266 of which 10.540 were nematodes; 9.935
crustaceans (mainly harpacticoid copepods and
ostracods); 1.100 micropolychaets and more than
3.000 forams. Stomach contents of many Dotilla
showed that all these groups serve as food items.
The sandy beaches are also inhabited by a variety
of macro-organisms and the most common are the
 645

crustaceans: Ocypode saratan, Talorchestia hide in burrows, closed by a sand bulge produced
martensi and Coenobita scaevola. Within the belt by the crab. Below this prop an air bubble is
occupied by Dotilla are Ebalia abdominalis, preserved in which the crab continues to be partly
Diogenes ~, in the vicinity of the mangrove, active. A few minutes after the water recedes,
various species of Uca. The specific composition emergence from the holes begins. The first crabs
of these assemblages varies according to geogra- emerge 3 to 4 minutes after the sea recesses; the
phical regions. most massive emergence occurs 7 to 8 minutes later,
All along the sites studied, an increase in ~. passing the peak at 15-16 minutes of the exposure.
sulcata population density and a decrease in First, the crab pushes upwards from below the
overall dimensions of the majority of crabs occurs prop that closes the hole and so raises the
in a cline from north to south (Table 1). As we sediment. Consequently, the anterior part of the
will see from figures on feeding, this is not due animal becomes visible. Moving up, the crab turns
to a problem of resource availability. slightly around along his body axis. After emer-
gence, the animal starts to dig out and renew the
TABLE
used burrow. The crab holds the chelae almost
Population density per m2 of D. sulcata, the di- vertically and pushing them down into the sand,
mension of individual crabs (;m) of the majority
and the largest specimen observed (*Southern Red separates portions of it. These sand portions are
Sea; **Gulf of Suez; *** Gulf of Aqaba). then pressed by the second pair of legs towards
the sternum and so, between it and the legs, a
Carapace width
sand bulge is formed. This is then carried a
n Number lar- distance from the hole and with one quick thrust,
per m2 majority gest Locality
thrown away. The time consumed for digging is
2 320 3-5 6 Massawa* closely correlated with the depth of the burrow
4 400 3-5 6 Isola Verde*
which in turn is correlated with the slope of the
4 310 3-5 6 Melita Bay*
5 168 4-6 7 Melita Bay* surface and distance from the low tide mark. As
10 240 3-4 6 Dahlak*
a rule, the steeper the slope, the deeper the
6 280 4-5 7 Musseri*
8 160 4-5 8 Dahab*** burrow, then the animal spends more time digging
5 205 3-4 6 Dahlak*
or renewing it after each ebb. Thus, the closer
4 420 3-4 6 Dahlak*
4 200 5-6 9 Dahab the animal is to the lowest water line, the less
6 180 5-6 13 Magrash***
it digs. Crabs found along the lowest water line,
6 160 5-6 14 Marsa Murakh***
5 140 6-7 12 A-Tur** do not dig at all: they simply push aside with
6 75 5-7 10 Nabeq***
their legs the sediment and pressing their body
7 86 5-6 9 Nabeq
3 170 4-6 12 Marsa Murakh backwards, dive into it. A question arises, why
don't most of the crabs live along or close to
HABITAT AND BEHAVIOUR OF DOTILLA: On all sites the water line and so save the energy of digging?
investigated, ~. sulcata forms populations of As this is not the case, then what factors regu-
various densities, dimensions of individual crabs late the distribution of Dotilla populations in
and exploited width of surface. As in several various localities?
other species of amphibious decapods (Altevogtt,
FEEDING: After emerging and clearing the burrow,
1957a; Fiedler, 1970; Quinn, 1980), also in D.
and often simultaneously with these activities,
sulcata we are dealing with a species of precise
the Dotilla start to feed. Using their chelae
tidal rhythmicity of activities (Magnus, 1960;
they collect sand grains and pass them towards and
Fishelson, 1978). During high tides, the Dotilla
between the maxillipeds. There a rotary movement
646

of the grains is initiated to form a sand ball as at each site: The maximal exposure of a Dotilla on
described for other crabs (Altevogtt, 1957a). The the outermost part of the colony will be almost 6
dimensions of these balls varies between 2 and 4 rom hours. This part will be elevated 30-45 cm above
and correlates with the diameter of the sand gra- the main seasonal sea-level. At the same time the
nules and dimension of the crab. A formed sand minimal exposure will be for Dotilla living along
ball is moved up between the maxillipeds to the the line of the lowest water level. Between these
frontal regions, where it is picked up by one of two lines of tidal fluctuations, activity gradients
the chelae and disposed on the ground. The dis- of the crabs correlate to the exposure time.
posal is performed by the left or right chela but Calculations show (Table 2) that during 24 hours,
individual crabs tend to be either right or left- the time for outside activities, inclusing feeding,
handed. The place of the sand ball deposition is is almost 11 hours for extreme landward situated
usually laterally, adjacent to the acting chelum, Dotilla, and 40-70 minutes for the most seaward
but it can also be ahead of the moving crab or situated ones.
even between the legs posteriorly. TABLE 2
As deposition takes place, new particles of sand Duration of activity (in minutes) of Dotilla at
are handed from the chelae towards the crab's various distances from lowest water line, during
two tide cycles (July 1979, Marsa-Murakh).
mouth producing a constant flow of sand from the
substrate to the maxillipeds and as balls towards Distance from
the ground. The speed of this process at the the lowest tide Activity Organic material
line. time. in sand.
highest observed was 40 balls per minute of acti- (m) (minutes) (mg/gr sand)
vity; this was in summer, when the air temperature
was 23 0 C. At an air temperature of 19 0 _20 0 C, it 1 46 160
3 142 162
takes 6 to 7 seconds for a crab to produce a ball. 5 223 180
Below 15 0 C Dotilla produced 3-5 balls per minute 7 310 130
9 375 128
and at 10 0 _12 0 C this activity stops altogether. 12 465 112
16 575 96
As the crabs continue to feed, they move sideways, 20 632 80
leaving an elongated, shallow pathway-groove, from 24 710 58
which the feeding material is collected. Along
The feeding area of the most landward situated
this groove a raised bar of the placed sand pellets
populations of Dotilla differs during Winter-Spring
if formed. These feeding lines extend straight
and Summer-Autumn and correlates with the sea
from the hole for a few centimetres up to 30 cm.
levels in those months. In the Winter-Spring sea-
In such a manner, radially extending feeding spurs
son, when the average water level is in the N. Red
are formed on the sand, sometimes covering 95 0 to
Sea at its highest, the tide fluctuations cover
180 0 of the circle (Fig. 2). During feeding, the
also most of the upper part of the inter-tidal,
crabs water the sand granules by expelling water
where the Dotilla are situated, and renew food re-
from the gill-chamber and some fluid from their
sources daily. Then the crabs feed close to their
mounth.
burrows. Beginning in June, however, the average
INTENSITY OF FEEDING: Ebb and the following flood
water level drops and the tidal fluctuations do
control the onset and cessation of on-ground acti-
not cover the most landward situated Dotilla.
vities, including feeding. As the populations of
Around these holes the layer of used surface sand
Dotilla form wide belts within the tidal zone, the balls is sometimes 3 to 4 cm thick. In these
overall duration of activity varies along the slope
months, the Dotilla situated here start to make

Figure 2. Sand pellets produced by D. sulcata (the larger one from digging).
feeding migration towards the sea, sometimes aggre-
gating in high numbers on the water line. As the
tide rises these crabs escape back to their living
quarters on the opposite side of the colony.
Observations of marked individuals, proved that
60% to 75% of crabs return to their original
burrows, whereas 25% to 40% penetrate foreign
empty burrows or dig new ones.
FOOD EXTRACTION: The amount of food extracted by
populations of Dotilla of various sizes is in
accordance with the amount of food available, dura-
tion of feeding activities and density of popula-
tion. Comparing the amount of organic material per
gram sediment in used food balls and in sediments
freshly exposed and not disturbed,some figures
could be extrapolated. At Marsa Murakh (Gulf of
Aqaba), the organic content of the surface 1.0 cm
layer of sand, within the centre of Dotilla popu-
lations, varies between 57 mg/gram sediment on
sites with low concentration to 160 mg/gram on
sites with a high content of interstitial organism
and detritus. From this supply the crab popula-
tion extracts once or twice a day, 10% to 30% of
the actual organic resources available.
Measurements show that crabs of 3.0 - 4.0 mm in-
gest 8 to 10 mg organic material during one period
of feeding; crabs of 11 to 13 rom use 100 to 160
647
mg of food during a similar period (Table 3).
Considering the varying amounts of organic mate-
rial on the investigated sites, the crabs living
on various sites would need different times to
collect their food. For example, if a small 3-4
mm crab, lives on sediment with 57 mg organic
matter/per gram sand and it is able to extract 10%
of the available material, then on an average to
obtain 9 mg food, it must process some 1.6 gram
of sand. If the same crab lives on a substrate
with 160 mg organic matter/gram sand, it needs to
process less than one half gram of sand. If the
crab is 13 rom and lives on grounds with 57 mg
organic material/gram sand, then utilising 30% of
the stock, he needs to screen some 5 grams sand,
and on sands with 160 mg organic material/gram of
sand, less than 2 grams of sediment.
At the southern Red Sea, at sites where the den-
2
sity of the crabs is 400 per m and organic mate-
rial 57 mg/gram sand, the amount of sediment moved
by all the animals will be approximately 6 kg/m2 /
day. This ignores the sediment carried out to the
surface during digging. These numbers show how
important a large population of Dotilla can be as
food utilisers and how much sediment is moved by
them during their activities.
648

TABLE 3
Food utilisation and processing of sand by D. sulcata on grounds with 57 mg.org. mat/gram sand and 160
mg.org. mat/gram sand. Lowest efficiency of food extraction is 10% and the highest 30% of the organic
material available.

2
Data on Crabs How much sediment turned around in grim
n Food If 57 mg org/gr sand If 160 mg org/gr sand
Dimension Weight ratio
mm mg day/mg One crab 200 crabs One crab 200 crabs
6 3 - 4 50 - 70 9.0 1.6 - 0.45 320 - 90 0.55 - 0.18 llO - 36
5 5 - 6 80 - 95 15.0 2.4 - 0.80 480 - 160 0.96 - 0.25 192 - 50
6 8 - 10 210 - 240 39.0 6.8 - 2.50 l360 - 500 3.45 - 0.70 790 - 140
6 12 - l3 520 - 580 97.0 16.0 - 4.60 3200 - 920 5.20 - 1.60 1040 - 320

CLOSING THE BURROWS: As mentioned by Fiedler ENVIRONMENTAL PHYSIOLOGY: The microclimatic con-
(1970) and Hartnell (1973)jseveral species of ditions of the habitat used by Dotilla are a func-
intertidal crabs close their burrows with the on- tion of interaction between humidity and tempera-
set of floods. In D. sulcata this activity begins ture. As in other terrestrial crabs (see Bliss,
5 to 25 minutes prior to the arrival of the advan- 1968), Dotilla digs to escape the extremes. The
cing water front. Those living very close to the temperature regimes in such burrows differ greatly
lowest water line simply sink into the watery from ambient air and water temperatures (Fig. 3).
sediment and push sand over their bodies until they
disappear. In such a way they remain very close
_._.-. Air
to the surface and can be easily collected from - - - - 6 em deep
..... 10 ..
their hideouts. Those living higher within the _ _ 1411
/~.-<>- . ....q,
intertidal react to the approach of the tidal 32 Water ,/ \.
/
.p-.-o-.~'
front by collecting sediment around their holes. <1/
.'
They produce a pile of sand, sometimes larger than
the diameter of the burrows. Then embracing it in 30 ;>--0- -0..
f/ ",,<-
a manner as described for digging, they move in, p/ ",,<-
/ ,
descending into the hole with the free chela first. P ,0.. 'b--<>

The crab disappears and within and above the hole UJ 28 /if''' .p.'--<f.' 'b,.
0:

,,
:::l I
p--.if/ ,;:/ ..'o- ... .Q
constructs a sand prop. The advancing flood pene- <0: ,(>-.-i
i
I
P--'.<f'-

,I
W
trates the lower parts of the burrows before the "- ./
::;
upper parts are submerged, and rises within them
UJ
>- 26 )0 ,<I'

~-</
to press the enclosed air towards the sand prop. I
I

I
In such a manner an air bubble is trapped within
I
l
24
the burrow, producing some kind of submerged inter- I
//0 ,0"
face of air-water. d' .(>-····(l

o.····a'
Experimenting with crabs attached to kymographs
22
revealed that they really rest at this level, i.e.
800 900 1000 1100 1200 1300 1400 1500 1600 1700
they remain in air. The actographs obtained in TIME
these experiments also show that the crabs con-
tinue to be active while in the burrow during high Figure 3. Temperature in burrows of Dotilla
tide, showing limited vertical movements. during August.
 649

Descent into these burrows allows Dotilla to

escape the high midday extremes as well as to ab-
sorb humidity or even water.
BREATHING AND TEMPERATURE REGULATION: Although
Dotilla are semiterrestrial animals, their oxygen
supply is provided mainly through gills surrounded
by water within the gill-chamber. Feeding above
ground during ebb, the animals use a large amount
of water from the gill-chamber and cause a deple-
tion around t ·n e gills. To renew the oxygen con-
tent in the ((ill- water, and possibly also to aid
thermoregulation, a special mechanism of reflu-
shing water over the body has been developed that
involves special morphological structures of the
exoskeleton. From exhalant openings situated
close to the antennule bases, the crabs expel two
Figure 4. Grooves on carapace 0-£ D. sulcata.
streams of water that spreading over the carapace,
flow along wulci (Grooves), fenced by lines of
cirri (Figs. 4 and 5). On both sides of the body,
these grooves unite toward the inhalant openings
situated at the bases of the 4th and 5th thoracic
legs. The reoxygenated water, before being
absorbed into the inhalant openings frequently
spreads out for a second or two among the leg
bases as a fine film enabling a better enrichment
with oxygen. The mechanism of flushing water over
the body may also serve for evaporative cooling.
Hints for this are provided by the fact that
during the hours with high temperatures, the fre-
quency of water-film formation is 2 or 3 times
a minute. At lower air temperatures, this flu-
shing can be once in 4- 5 minutes. During tests,
microthermocouples implanted within the animal's
Figure 5. Direction of water flow on D. sulcata.
body, enabled to follow the body's heat regime as
compared to the ambient. Below 25 0 the animal's maintenance of body temperature around 29 0 _32 0 e
body temperature follows ambient, and the sprea- breaks down at an ambient temperature of 39 0 _42°e.
ding of water from the gill-chamber oter the body In summer Dotilla finds itself in ambient
occurs once in 1 or 2 minutes. As the temperature temperatures of 36°e to 43 0 C, in localities where
rises above 25 0 e, the body temperature diverges even the water temperature rises to 29 0 e-32 0 e.
from that of the ambient, remaining 20 e to 30 e Experiments in the field showed that when crabs
lower. Then the spreading of gill water over the were prevented from renewing the water, they lost
body is 2 to 3 times per minute, so that the cara- weight and entered into a coma, not able to move.
pace remains wet almost constantly . The At ambient temperatures of 32 0 e death occurs
 650

after 30 to 40 minutes during which they lose 40% Preserving body temperature at the cost of water
to 50% of their body weight. Crabs of carapace loss is extremely disadvantageous for oxygenation
width of 4-6 mm put on sand surface and at the as well as for thermoregulation. In natural habi-
same temperature, remain alive for 80 to 130 tats the animals get round these critical situa-
minutes, but finally death occurs from dehydra- tions by timing the activities and renewing their
tion. The behaviour of the larger animals, of water reserves. One of the behaviours observed
11-12 mm carapace width and collected from the in Dotilla during high temperature is frequent
higher parts of the community is different (Figs. visiting of burrows. There it is not only cooler,
6 and 7). These are much more resistant to but also more moisture is present in the sand.
adverse situations and their ability to control Contact with the ground was also later observed
dehydration is much greater. frequently during feeding and possibly serves for
water absorption as well as for conductive cooling.
25
WIDTH OF DOTILLA POPULATIONS: The extension of
the crab populations in the studied areas is land-
20
wards limited by the availability of humidity and
'-'
::;
seawards by the slope steepness of the intertidal.
~
(f)
(f) 15 Humidity in the ground is closely related to pene-
'3 tration of sea water during high tides as well as
'>-"
w
the rate of evaporation.
'"
;: 10
CARAPAX
8- 8.5
As for slope of the intertidal, the gentler the
bottom descent, the wider is the space occupied
by the population. In some places the potilla
30 90 150 210 270
DESICCATION TIME. MIN. population spans 100 m and more within the inter-
tidal, whereas, on shores with a steep slope, the
Figure 6. Water loss in D. sulcata during belt occupied by Dotilla is very narrow, some-
desiccation.
times comprising a single row of burrows. Within
the populations the largest dominate the landward
CARAPAX
WIDTH
part of the habitat, while smaller animals occur
60
(MM)
near the sea (Table 4).
~
>- TABLE 4
J:
':2 Diameter of crabs along 15 transects at Marsa-
UJ
;: Murakh (Gulf of Aqaba), May 1980.
>
Cl
0
Ol
u..
Distance from Diameter (mm)
0 LT-line
(/)
(/) (m) 3 5 7 9 11 13
0
...J

2 40 62 0 0 0 0
3 66 46 29 0 0 0
6 20 48 72 2 0 0
9 0 30 60 38 0 0
DESICCATION TIME 12 0 16 48 52 18 0
MIN
15 0 0 43 66 22 6
18 0 0 10 45 48 18
Figure 7. Loss of body weight (%) with 21 0 0 0 47 50 34
desiccation. 24 0 0 0 12 60 68
 651

The females are as a rule smaller than the males marine species found in habitats like
(Fig. 8) and they dominate sites closer to the this.
lowest part of the tidal range (Fig. 9). This There are at least two drawbacks to life along
structuring is very clear on shores with a gentle the high-tide line:
but clear slope of 1 0 or 1.5 0 grade descent. On
First - The higher Dotilla are on the shore the
sand flats that are almost horizontal, such as
more they are exposed to extreme
around mangroves of Nabeq (Sinai Peninsula), this
temperatures and evaporation.
pattern disappears and a mixture of larger and
smaller animals is observed. Second - The higher on the shore the shorter the
flood period, so that fewer microscopic
animals and detritus are deposited and
the food density is low.
MELITA For example, at Marsa-Murakh (Gulf of Aqaba) orga-
N=308
nic content per gram surface sediment was, in the
8
E higher parts, around 57 mg/gr sand, whereas in
E
~
the lower intertidal it averaged 160 mg/gr sand.
o CONCLUSION AND DISCUSSION: The ability of animals
to conquer habitats bears on some kind of com-
MARSA
MURAKH promise between their biological qualities and the
N=250
acting environmental parameters. According to
this, the more extreme these parameters are, the
E closer they are to the limits of biological
E 6
~ utilisation, therefore, the population using them
is would be more specialised. By occupying the bare
2
2 intertidal soft-bottom habitats, Dotilla sulcata
~ and related species conquered a part of an eco-
system with a few animal species and with the
lowest interspecific competition possible. The
Figure 8. Dimension of sexes and their per cent
in various age-classes at Melita Bay populations of D. sulcata in the Red Sea; Q.
(Ethiopia) and Marsa-Murakh (Gulf of fenestra in South Africa (Hartnell, 1973);
Aqaba) •
Scopiomera inflata (Fielder, 1970) and Myctiris
There are at least three benefits of inhabiting longicornis (Quinn, 1980) of Australia, as well
high-intertidal sites: as several more species, are the most successful
First - The higher above the water level, the users of such extreme intertidal environments.
more time is available for feeding The basic and primary reaction to such conditions
(Table 3). is an avoidance behaviour, such as digging burrows
Second - Higher on the slope, the water remains in which to hide and escape from overheating and
shallow even at the highest tide, and dehydration. As observed in D. sulcata the
digging predators (fish) are less able higher the ambient temperature, the more fre-
to penetrate this area. quently do the crabs visit their burrows.

Third - The higher in the intertidal, crabs Smaller individuals are especially sensitive to

experience less competition with other overheating and, partly because of this, their
 652

14 ~----,------<'

Width 16
of
carapax
(mm) No
of
specimens

5 7 9 11
Distance from low tide mark (m)

Figure 9. Distribution of Dotilla accroding to dimension, sex and distance from the low-tide mark.

burrows are much closer to the low water line than
those of larger animals. According to Altevogtt
(1957b) Q. blanfordi wander around on their sites,
dig several burrows a day and use foreign holes.
D. sulcata uses one burrow for a long time, pro-
tects it from intruders and acts around its
opening. This makes them less vulnerable to the
severity of the sites. In the Red Sea the summer
temperature may rise to 43 0 C - 45 0 C, surpassing
the upper physiological limit which Dotilla and
particularly their small individuals are able to
withstand. Despite this, these crabs form popu-
lations of up to 420 individuals per square
metre, comparable only to crab densities in deep
seas where the environmental parameters are
almost invariant. The important adaptation is
the ability to preserve the internal aquatic
environment needed for respiration and cooling.
One such mechanism, first described by Vervey
(1930) for Sesarma taeniolata and ~. nudulifera,
and later observed for ocypodid crabs, including
Dotilla, is the recycling of gill-chamber water
over the carapace. In Q. sulcata this water,
flowing over the front, streams along the cara-
pace in a special set of sulci that direct it
toward the inhalant openings. The higher the
temperature, the more frequent is this flushing.
This serves for oxygenation as well as for
cooling the body by evaporation. This was also
observed in Uca (Edney, 1962). To replace the
coolant, the crabs living close to the water
simply plunge into the sea, as is done by species
of Ocypode as well as the hermit crab, Coenobita
scaevola (pers. observ.). Dotilla living far
from the water line, absorb water from the sedi-
ment by means of tufts usually found around their
leg-bases. These crabs press their body towards
the sediment, inserting the tufts into it. In
this way, a capillary flow is activated and the
water, moving along the setae, ends within the
gill chamber. This behaviour was described for
several other species (Rao, 1968; Hartnell,
1973; Walcott, 1976; Quinn, 1980). This type of
water absorption called "sponging" by Fiedler
 653

(1970) is more intensive during high temperatures oxidation in the usually anaerobic layers.
or intense feeding. The closing of the burrows
Comparisons of dimension of Dotilla in popula-
by sand props before the sea covers them, is an
tions of the southern and northern Red Sea show
additional adaptation that leads to trapping of
that those found in the northern part are, on
air bubbles below them, which enables the sub-
the average, larger than those of the south.
merged crabs to continue to use atmospheric
This can be correlated to the usually more dense
oxygen.
populations occurring in the southern part. If
After solving the physiological problems, a this is so, then the maximal growth is density
central problem of food resources availability dependent. This may also be the result of
remains. For D. sulcata and related species, selective predation by animals feeding on ghost
there are numerous unicellular organisms found crabs. There are several bird species breeding
on sand surfaces as well as many interstitial in the southern Red Sea and feeding on Dotilla.
organisms living between the sand granules. Of these, the mangrove heron (Egretta gularis)
This kind of food is constantly renewed by the and larger plover birds (Dromas) were actually
tides so that the investment for food collection observed collecting Dotilla. Stomach contents
is a low one. Extraction of this food is per- of two Dromas contained a large number of these
formed by combing the sand granules with the very crabs, all of them 5-7 rom in width. This means
hairy mouth appendages. During this activity that the predator selects the larger individuals
water is extruded on the food grains in a pro- and so possibly prevents aging.
cess called flotation (Altevogtt, 1957a;
ACKNOWLEDGEMENTS: I would like to thank all
Fiedler, 1970). The food organisms are then
those who helped at various stages of this study,
separated and devoured, the water is resucked to
especially Dr. D. Popper, Mrs. N. Gunderman and
the gill chamber or foregut and the used sand is
Mr. N. Sharon. Mr. S. Sheffer, R. Kalif and
deposited on the sand surface in the form of a
A. Shuw kindly helped in the preparation of the
baIlor pellet (see also Silar and Sankarankutty,
figures, and many others helped in the logistics
1967). ~. sulcata is able to extract 10% to 30%
and techniques during several expeditions.
of the overall organic content of the substrate
Thanks to Dr. Linn Montgomery for critically
and hence in order to fulfil their food demands,
reading and improving the manuscript. Special
they must process large amounts of sand.
gratitude to my son Zvi and wife Luba, who
Calculations show (Table 3) that during a day of
helped me during many burning hours on the
feeding a population of some 350-400 crabs on one
exposed shores of the Sinai Peninsula.
square metre will move some 6 kg of sediment.
Tons of sand are, therefore, moved on shores with
extensive populations of Dotilla. This activity
totally changes the texture and layering of the
upper 1.0-2.0 cm of sediment, the result of
which will be a decrease in organic content and
an increase in oxidatory decomposition. This
beneficiation process also extends deeper into
the substrate as these crabs dig to a depth of
16-17 cm to produce their burrows. Air pene-
trating into the burrows possibly also enhances
 654

REFERENCES:
Altevogtt R (1957a) Untersuchungen zur Biologie, Paulson 0 (1875) Studies on Crustacea of the
Okologie und Physiologie indischer Winkerkraben. Red Sea, Part I (English Translation by IPST,
Z. Morph. Okol. Tiere 46, 1-110. Jerusalem 1961), 164 p.
Altevogtt R (1957b) Beitrage zur Zoologie und Quinn RH (1980) Mechanism of obtaining water
Ethologie von Dotilla blanfordi Alcock und for flotation feeding in the soldier crab,
Dotilla myctiroides (Milne Edwards) (Crustacea, Mictyris longicarpus Latreille, 1906 (Decapoda,
Decapoda) Z. Morph. Okol. Tiere 46, 369-388. Mictyridae). J. expo mar. BioI. Ecol. 43, 49-
60.
Bliss DE (1968) Transition from water to land
in Decapod crustaceans. Am. Zoologist, 8, 355- Rao KR (1968) The pericardial sacs of Ocypode
392. in relation to conservation of water, molting
and behaviour. Am. Zool. 8, 561-567.
Edney EB (1962) Some aspects of the temperature
relation of fiddler crabs ~ spp.) in: Silas EG and Sankarankutty C (1967) Field
Biometerology (ed. S.W. Tromp), 79-85, Pergamon observations on the shore crabs of the Gulf of
Press, London. Mannar and Palk Bay, with special reference to
the ecology and behaviour of the pallet crab,
Eibl-Eibesfeldt I and Hass H (1969) Dotilla Scopimera proxima Kemp. Proc. Symp. Crustacea
sulcata - Fressen und Graben-10 min mov., (Ernakulum, India) Part III, 1008-1025.
Encyclopedia Cinematographica, Gottingen.
Verwey J (1930) Einiger uber die Biologie
Fiedler DR (1970) The feeding behaviour of the ostindisher Mangrove-Krabben. Treubia 22, 169-
sand crab Scopimera inflata (Decapoda, 261.
Ocypodidae). J. Zool. Lond. 160, 35-49.
Wolcott ThC (1976) Uptake of soil capillary
Fishelson L (1971) Ecology and distribution of water by ghost crabs. Nature (Lond.) 264, 756-
the benthic fauna in the shallow waters of the 757.
Red Sea. Mar. BioI. 10, 113-133.
Fishelson L (1973) Ecological and biological
phenomena influencing coral species composition
on the reef tables at Eilat. Mar. BioI. 19,
183-196.
Fishelson L (1977) Stability and instability of
marine ecosystems, illustrated by examples from
the Red Sea. Helgolander wiss. Meeresunters.
30, 18-29.
Fishelson L (1978) Tracks on the beach: Life in
the tidal zone. Israel Land and Nature 3(3),
105-112.
Fishelson L (1980) Marine reserves along the
Sinai Peninsula (northern Red Sea) Helgolander
Meeresunters. 33, 624-640.
Hartnoll RG (1973) Factors affecting the distri-
bution and behaviour of the crab Dotilla
fenestrata on East African shores. Estuar.
cstl. mar. Sci. 1, 137-152.
Macnae Wand Kalk M (1958) A natural history of
Inhaca Island. Macambique. Witwatersrand Univ.
Press. Johannesburg.
Magnus D (1960) Zur Okologie des Landeinsidlers
Coenobita jousseaumei Bouvier, und der Krabe
Ocypode aegyptica Gerstacker, am Roten Meer. Ver.
Deutch. Zool. Ges. 316-329.
 655

ECOLOGY OF THE SANDY BEACH GASTROPOD MAZATLANIA ACICULATA IN QUIZANDAL (CARABOBO,

VENEZUELA) .

P.E. PENCHASZADEH, G. DE MAHIEU, V. FARACHE and M.E. LERA (Instituto de Technologia

y Ciecias Marinas Universidad Simon Bolivar, Apartado 80659, Cracas, Venezuela)

1. INTRODUCTION American Pacific and Atlantic, but does

Mazatlania aciculata is a small snail, not give precise localities; vermeij
very common on Venezuelan sandy beaches. (1978) supports this hypothesis
The presence of M. aciculata was including M. aciculata in the list of
recorded in different localitites living species Vlith an amphioceanic
including the area of Puerto Cabello distribution after the definitive
(Carabobo), the Bay of Turiamo, on the closing of the marine connection about
littoral of the Distrito Federal, the 3.5 million years ago.
peninsula of paria and also on the
Island of Margarita. Its abundance, 2. MATERIAL AND METHODS
especially in the oriental coast of Samples Vlere taken twice monthly from
Venezuela, is very high and the shell is July, 1976 to February, 1978, using two
used for artistic purposes. This different methods. The first used a
abundance contrasts with the very scarce canvas net wi th an opening 40 x 10 cm
information about the species we have supported by an iron frame, which was
found in the malacological literature pulled towards the shore from a depth of
for the Caribbean area. 1.5 m. The sediments collected Vlere
screened through a 1 mm mesh; this
Mazatlania aciculata is a characteristic method allowed us to determine the size
species of the late Pliocene Mare structure of the population as well as
Formation of Districto Federal, the community composition. The second
Venezuela (Weisbord, 1962) • Through method used a 18.5 cm diameter cylinder
access to the Gibson-Smith collection in which ,las introduced into the sediment
Caracas we were able to confirm that the to a depth of 20 em. 'l'his method was
fossil material is identical to the used for obtaining information on
living animals we studied in Quizandal. vertical distribution and population
Other records for M. acicilata include densities. All living material >las
those of Coomans (1958) for the Island fixed in 7% formalin. From ~jarch, 1976
of Margarita, Olsson, McGinty (1958) for to January, 1978, surface temperature of
Isla Bocas del Toro in Panama and the beach water was measured daily at
Kaufmann, Gotting (1970) for Rodadero in 08hOO. In March and September 1977,
Colombia. Radwin (1977) cites M. temperatures were recorded for a 24 hour
aciculata as being of amphioceanic cycle. Live material was collected for
distribution in both the Central experimental purposes uetween 1976 and
656

1981. Once in the laboratory they Vlere 1977) . The maximum absolute temperature
acclimated for 10 days at 27 0 C and registered (08hOO) Vias 29.8 0 C in
36 0 / 00 salinity, then placed in an October 1976, anc1 the absolute minimum
aquarium. Temperatures was 24.5 0 C in March 1977. There is
experiments ranged from also a typical daily cycle in the
and observations were made for up to 96 coastal water temperature Vlith the
hours of exposure. Animals were lowest temperature at O&hOO and the
considered dead when they no longer highest at 15hOO anc1 a c1aily range of
responded to pinching the foot. '1'0 approximately 2,5 0 C.
distinguish the animals under thermal
shock from those really dead the The Community.
individuals were placed in recuperation In Quizanc1al veach a defined
chambers at 27 0 C for 24 hours. vertical zonation can be recognised.
The upper fringe of the intertidal and
3. RESULTS supralittoral is characterized by the
Habitat brachyuran Ocypode quadrata and the
Mazatlania aciculata lives in sandy isopod Excirolana braziliensis. '1'he mic1-
beaches from the low tide level to 1.5 m littoral zone is uccupied by the
of depth in the surf area. It is a anomurans Emerita brasiliensis and
superficial burrower, occurring in the Lepidopa richmonc1i and scarce adult
first 5-10 mm in the substrate; when individuals of the molluscs Donax
resting only the siphon remains denticulatus, Terebra cinerea, Polinices
exposed. Sediments of the Quizandal hepaticus and Olivella verreauxi, and
beach contains a very low percentage of the polychaetes Glycera ~. and
silt, the rest being fine sand. Sthenelais limicola. On the upper
Carbonates are low ( 6,9%) and the gran- infralittoral Mazatlania aciculata is
ulometry indicates a predominant (55%) the characteristic species, associated
grain of 0.125mm. There is no signifi- with o. verreauxi, D. denticulatus, the
cant variation in the grain composition isopod Ancimus brasiliensis, juveniles
pattern throughout the year. The of the sand-dollar Mellita quinquies-
maximum value above mean sea level can perforata latiambulacra, '1'. cinerea, P.
reach 0.75 m due to meteorological hepaticus, the marginellid Prunum
tides; the waves in this part of the prunum, the portunid crab Aranaeus
Caribbean are produced mainly by the cribarius, the polychaetes Glycera sp.,
Trade winds with a predominantly S. limicola, Onuphis eremita, Dispio
east-northeast direction and by local uncinatus, Ammotrypane sp., and Nephtys
storms (INTEVEP, 1980). There is a sp., the mysid Bowmaniella sp., the
marked seasonal cycle in the Quizandal lancelet Branchiostoma caribaeum and the
coastal water temperature. Minimum mean fish Gillelus grayae. This association
monthly temperature taken at 08hOO reaches a depth of 1.50-2.00 m. The
occured in March (25.5 0 C in 1976 and middle infralittoral zone is character-
25.4 0 C in 1977) and the maximum mean ized by adult quinquiesperforata
temperatures were registered in October latiambulacra , the gastropods Persicula
(28.4 0 C in 1976 and 28.0 0 C in interruptolineata, Conus puncticulatus,
 657

Oliva reticular is, O. verreauxi and 2 months= 5.6 mm 6 months=ll.l mm

Turbonilla sp, the bivalves Strigillia 3 months= 7.5 mm 7 months=11.6 mm
producta and juvenile D. denticulatus, 4 months= 9.3 mm 8 months=12.1 mm
hermit crabs (pagurus ~.) utilizing 5 months=10.4 mm
Mazatlania shells, ophiuroids, number of individuate I m2
o 0
TY
sipunculids,
the
individuals
anthozoan
stomatopods,

of
Renilla
Mellita
Amphipoda
sp. Isolated
sexiesperforata
and

Ocypode
II 1 1 .

500 20 50200
I

----------1
:::: __Meilita adult I
- _ _ _ _ _ _ .J

and Tivela mactroides were also

o
recorded, but they are very rare in .'''~~:~7.··~~..·~~~;~:]
....... - ... - ... - ... - ..
~

Quizandal beaches. Polychaeta i

:c . _ . _ . __ . . .:~.;.- ... _ ... _ ... J
I- <MYSld~~.~:"::::=- .
population density. 0..
W
Mazatlania aciculata density in the 0 2
Quizandal beach varied in relation to
beach morphology, wave action, tidal
currents and other factors which often
resulted in a patchy distribution, some- Berm 10 20 30 40 50 60
times with high densities. The highest DIST ANCE em)
density recorded in Quizandal was 407 FIGURE 1. Sandy beach community zonation
.m- 2 . Minimum mean specific density at Quizandal, Venezuela.
recorded was 23 .m- 2 (census frequency
25.0%) and the maximum mean density was The maximum length recorded in Quizandal
207 .m- 2 (presence 100%). The density beach was 16.4 mm. The L t calculated
of M. aciculata is low compared with by the Ford-Halford method was 14.2 mm
some measurements made in eastern and the Bertalanffy constant k = 0.227.
Venezuelan beaches. In Playa Guacuco,
paria peninsula, we recorded a mean As a result of the high mortality
density of 1480 .m- 2 . observed from 11 mm of length, the
larger classes are poorly represented.
Growth. In the deeper adjacent level the pagurus
The growth rate for the Quizandal ~. population utilizes the shells of H.
population of M. aciculata was studied aciculata with sizes between 7 and 13 mm
by means of modal displacements through in length.
time of size-frequency distributions.
Four well-defined periods of recuitment Reproduction.
occur in February, May, July and The females attach their egg capsules to
September, 1977. Growth is fast. If we the shells of other individuals of the
assume 3 mm length for the first month population of more than 9.0 mm in
of benthic life, the length-ages for the length, which represents four months of
first eight months are as follows: benthic life (Fig. 2). The egg capsules
are attached, occupying most of the
surface of the shell during the breeding
season. Its shape is ovoidal measuring
658

(199 capsules) between 1.4 and 1.7 mm in percentage is o. The reproductive cycle
length and 1.0 to 1.4 mm in width. The of M. aciculata is charact~rized by a
number of eggs per capsule varies very long active season which starts in
between IS and 62 (mean 34). The egg November and ends in August. During
diameter varies between 180 and 220 m this period the gonads contain both
(mean 209 m). There is no supple- mature and immature elements.
In one
mentary food for the embyros and all reproductive season different generations
arrive at the eclusion stage as free are involved, some of them born in the
swimming veligers. The larval shell at same period.
hatching measures 300 m; the veligers
escape through an exit aperture in the Temperature tolerance.
top of the egg capsule. The reproductive M. aciculata has a moderate tolerance to
season is related to water temperature. high temperature. The values (Fig. 3)
indicate a median lethal temperature
LTSO of o
34.S C for 72 hours of
exposure and 33.S o C for 96 hours.
Median burial temperature was always less
than the value of LTSO for each time
of exposure, supporting the observations
that emergence occurs always before the
death of the animal (BT SO of 32.6 o C
for 72 hours and 32.S o C for 96 hours
of exposure.)

35

34

33

FIGURE 2. Mazatlania aciculata. A and B, 32

lateral and apical views of an egg 31

capsule; C, apical view of an empty egg-
capsule; D and E, detail of the apex of 36 12 24 48 72 96
an adult shell. HOURS

FIGURE 3. Mazatlania aciculata. Change

The highest percentage of shells with in L'1'SO and B'1'SO with increasing
attached egg capsules (44%) occurs in time of exposure at experimental
March, the coldest month in Quizandal. temperatures (shell length between 9 and
In October, the warmest month, the l6mm) •
 659

4. DISCUSSION with the values recorded for Bullia

Mazatlania aciculata is the dominant rhodostoma of similar size (lO-lSmm) in
species in the upper infralittoral of South Africa (Ansell, McLachlan, 1980),
Quizandal beach. Its presence in the even taking into consideration that the
South Caribbean is confirmed from the acclimation temperature in that case was
late Pliocene and it coexists with other 20 0 (instead of 27 0 C for M.
endemic gastropods such as Prunum prunum, aciculata) there is a slightly higher
persicula interruptolineata, Conus temperature tolerance for ~j. aciculata.
puncticulatus and Terebra cinerea. Both are, however, related to the
maximum temperatures which could be
The Columbellidae present a great registered in each locality. Bullia
diversity in the egg capsule rhodostoma would be living at 5-10 0 C
characteristics; Thorson (1940) found under its tolerance limit, while
five different types for the genus Mazatlania aciculata would be living
Columbella alone and D'Asaro (1970 ) only at 5. SoC under its tolerance
described some other patterns for the limit, assuming a maximum acclimation
genus Anachis. The egg capsule most temperature in natural
close to M. aciculata in the family is condition (the maximum mean temperature
the one of Columbella flava (petit, in Quizandal).
Risbec, 1929) and C. rustica (Bacci,
1947) . 'J.'he habit of attaching the egg Lower temperatures in the South
capsules to the shell of individuals of Caribbean compared with the rest of the
the same species is known for several Caribbean basin and low percentage of
other species, i.e. Hydrobia ulvae carbonates in the sediments appear to be
(Thorson, 1946). key factors in determining the
distribution of Mazatlania aciculata.
The high densities of Mazatlania
aciculata in the Venezuelan sandy 5. ACKNOHLEDGEMEN'J.'S
beaches, as well as its wide To Drs. Hinnifred and Jack Gibson Smith
distribution in the South Caribbean of Universidad Central de Venezuela who
could be related to the presence of a provided valuable comments on the fossil
free swimming veliger and the fact that and recent distribution of the species,
coastal currents would not take away the which was most appreciated. The writers
stock from coastal environments. Thorson wish to thank also Mr. Luis Jos& Gonzalez
(1940) found that direct development is and Mr. Alexis Ojeda for their assistance
the most common type in Columbellidae on the sampling, to Mr. Hector Patino for
but D'Asara (op.cit.) states that in his help in the Laboratory and to Mr.
Anachis hatching takes place as veliger Eduardo Perez Perez for the illus-
larvae. The same was found for C. trations.
rustica (Bacca, op. cit. ) and for
Nitidella laevigata (Bandel, 1974).
G. REFERENCES
Ansell A and McLachlan A (1980) Upper
temperature tolerances of three molluscs
If we compare the results obtained on from South African sandy beaches. J. expo
mar. BioI. EcoI. 48, 243-25I.
temperature tolerance (LTSO and BTSO) Bacci G (1947) Le capsule ovigere di
660

Columbella rustica (L.) e di Fasciolaria

l~gnaria (L.) (prosobranchia Stenoglossa).
Boll. Zool. Torino 14, 75-81.
Bandel K (1974) Spawning and
development of some Columbellidae from
the Caribbean Sea of Colombia (South
America). veliger 16, 271-282.
Coomans H(1958) A survey of the
littoral Gastropoda of the Netherlands
Antilles and other Caribbean Islands.
Stud. Fauna Curacao 8(31), 42-111.
D'Asara CN (1970) Egg capsules of some
prosobranchs f rom the Pacif ic coast of
Panama. veliger 13(1),37-43.
INTEVEP (1980) Distribucion de las
masas de agua y sus vinculaciones
dinamicas con el sector Centrol-
occidental Veneqolana, Mar. Caribe;
Caracas, 1-81.
Kaugmann Rand Gotting KJ (1970)
Prosobranchia aus dem litoral der
Karibischen Kuste Kolumviens. Helg.
Wiss. Meeres. 21, 333-398.
Olsson AA and McGinty TL (1958) Recent
marine mollusks from the Caribbean coast
of Panama with the description of some
new general and species. Bull. Am.
paleont. 39, 1-52.
petit G and Risbec J (1929) Sur la
ponte de quelques Casteropodes
Prosobranches. Bull. Soc. Zool. Fr. ;
Seance 22 Octobre 1929, 54, 564-570.
Radwin GE (1977) The family
Columbellidae in the Western Atlantic.
Veliger 19, 403-417.
Thorson G (1940) Studies on the egg
masses and larval development of
Gastropoda from the Iranian Gulf. Danish
Scientific Invest. in Iran, Pt. 2,
159-238.
Thorson G (194G) Reproduction and
larval development of Danish marine
bottom invertebrates with special
reference to the planktonic larvae in
the Sound (Oresund). Medd. Kom. Danm.
F i s k . Ha r. P1 a n k t. 4, 1- 5 23 .
Vermeij GJ (1978) Biogeography and
Adaptation. Patterns of Marine Life.
Harvard University Press, Cambridge,
1-332.
Weisbord NE (1962) Late Cenozioc
gastropods from northern Venezuela.
Bull. Am. Paleont. 42(193), 1-672, pIs.
1-48.

PRODUCTION ECOLOGY OF HAUSTORIUS CANADENSIS (AMPHIPODA:
T.E. DONN, JR and R.A. CROKER (Department of Zoology and Jackson Estuarine Laboratory, University of
New Hampshire, Durham, NH 03824 U.S.A.)
1. INTRODUCTION
Several studies (Croker, 1977; Dexter,
1969; Samaeoto, 1969; Shelton, Robertson, 1981)
have shown that the sandy beach communities of
the east coast of North America are dominated
by haustoriid amphipods. The sandy beach
communities in northern New England consist of
approximately 11 species, predominantly
peracarid crustaceans, and are dominated by the
haustoriid amphipods Haustorius canadensis and
Acanthohaustorius mi11si and Amphiporeia
virginiana (Croker, 1977). Although much work
has been done on these communities, to date no
attempt has been made to describe the patterns
of energy flow or secondary production
occurring in these systems.
The purpose of this study was to examine
the seasonal dynamics of production in the
amphipod Haustorius canadensis on a sandy beach
in southern Maine, U.S.A. In New Hampshire and
Maine ~ canadensis remains within the
intertidal throughout the year; however,
seasonal migrations are evident, with the
amphipod moving upward on the beach during the
summer and toward MLW during winter (Croker,
1977; Donn, pers. obs.).
1Contribution number 147 of the Jackson
Estuarine Laboratory Contribution Series.
2This paper represents a portion of a
dissertation submitted to the University of New
Hampshire by T. Donn in partial fulfillment of
the requirements for the Ph.D. in Zoology.
by placing them in
661
HAUSTORIIDAE) IN SOUTHERN MAINE l ,2
2. METHODS
Long Sands, located in southern Maine,
U.S.A. (43 0 07'46" N, 70 0 37'12" W) is an exposed,
13 on McLachlan's (1980) exposure scale, fine
sand beach. The beach is approximately 160m
wide from the base of the seawall to MLW, with a
mean tidal amplitude of 2.6m. Long-term studies
(Croker, 1977) have shown the macroinfaunal
community to be relatively stable over time.
A population of Haustorius canadensis was
sampled along a permanent transect using 0.04m 2
quadrats in a stratified random design (5
levels, 3 replicates) at monthly intervals from
November, 1978 until August, 1981. During the
reproductive seasons of 1979 and 1980 the samp-
ling frequency was increased to biweekly to
obtain a better understanding of the reproduc-
tive patterns of H. canadensis. The amphipods
in the samples were sorted, counted and the H.
canadensis were retained. Densities were
corrected for the changing width of the distri-
bution, and population size is expressed as
abundance (number per metre of beach front). Sex
and length, base of rostrum to end of pleosome,
of all H. canadens is in the samples was
subsequently determined. The length-frequency
histograms obtained were separated into individ-
ual cohorts (Harding, 1949; Hasselblad 1966).
Determination of the length-weight
relationship and caloric content was made on
animals brought back live from the field.
Within 24 hours the animals were anaesthetized
662
70% ethanol for approximately 15 minutes, and
then diluting the solution by a factor of about
5. Length-weight determinations were made on
30 animals per month from August 1980 to July
1982. Animals were measured, sexed and then
dried for 24 hours at 60 0 C, weighed, ashed for
4 hours at 450 0 C and reweighed. Individual
functional regressions (Ricker, 1973) of log
weight on log length were determined for each
month, and the slopes were tested for equality
using the method of Clarke (1980). Of the 24
months for which the length-weight relationship
was determined only 2 showed significant
differences from the rest, therefore all data
were pooled and the functional regression;
10g(AFDW) -5.01 + 3.15 log(length)
n = 614 r = .89
was used in subsequent calculations. Caloric
content was determined on Haustorius canadensis
collected between June 1981 and July 1982.
amphipods were grouped by sex and 1mm size
classes, dried 24 hours at 60 0 C, ground in a
mortar and pestle, formed into pellets, redried
and weighed. Between 1 and 5 pellets from each
group were combusted in a Parr Semi-micro Bomb
Calorimeter depending on the amount of material
available. Mean caloric content was determined
for each size class and these values used for
the calculation of a mean monthly caloric
content.
Four methods of calculating production
were compared, the Removal Summation, Increment
Summation , and Instantaneous Growth methods
(Crisp, 1971; Gillespie, Benke, 1979), and the
Size Frequency method (Hamilton, 1969). The
first three methods, along with the Allen Curve
method, have been shown to be equivalent
(Gillespie, Benke, 1979). The Removal Summa-
tion, Increment Summation and Instantaneous
Growth methods require that individual cohorts
be distinguished. However, they allow the
seasonal patterns of production to be observed.
The size frequency method does not require
individual cohorts to be distinguished, using
instead an "average" cohort, but will only yield
estimates of total production by the cohort,
which must be corrected for the lifespan of the
cohort (Cushman et al., 1978; Waters, 1979).
3. RESULTS
In southern Maine, Haustorius canadensis
has a 2 year life span, with a reproductive
period extending from mid-May through August.
Juveniles hatch and are released from the brood
pouch at a length of 1.8mm to 2.0mm, and grow to
a maximum adult length of 8mm to 9mm.
The separation of the length-frequency
histograms into their component cohorts allowed
individual cohort abundances (Fig. 1) and, using
the above equation, mean ash-free weight of each
cohort (Fig. 2) to be determined. The vari-
ability in cohort abundances is most probably
due to the spatial dynamics of the population on
The the beach as well as to sampling error.
The mean caloric content (± 95% confidence
interval) of Haustorius canadensis is 4400 + 77
cal·gm- 1 dry weight. H. canadensis has a mean
ash content of 22.8%, resulting in an ash-free
caloric value of 5700 ± 100 cal·gm-1. When mean
monthly caloric content (dry weight) is plotted
against time (Fig. 3) a distinct seasonal
pattern emerges; a minimum value during winter
and a peak caloric content at the end of the
reproductive season in late August.
Annual production was calculated using the
four methods described above. A comparison of
the results from these methods (Table 1)
indicates that the removal summation, increment
summation, and instantaneous growth methods
yielded very similar results. The size-
frequency method overestimated the others by
60%. This result is similar to that obtained by
Waters (1981, Table 1) and is greater than
 663

30000

"
:c 25000
'-
0
z:
20000
w
U
z:
<I 15000
0
z:
::::l
(])
<I 10000
I-
0:::
0
I 5000
0
U

0
o N0 J F M A M J J A SON 0 J F M A M J J A SON 0 J F M A M J J A S
1978 1979 DATE 1980 1981

FIGURE 1. Mean abundances for the 1977 through 1981 cohorts.

the 30% predicted by Cushman et al.'s (1978) time by the mean ash-free caloric content
computer simulation. determined from Fig. 3 and using these data in the
The results from the instantaneous growth production model. Calculating production in terms
method were chosen for further analysis to of calories did not alter the pIn ratios obtained
enable comparisons with Waters' (1981) results. (Table 2) or change the seasonal patterns of
Production was also calculated in terms of production. Therefore only production in terms of
calories by multiplying the biomass at a given ash-free dry weight will be discussed further,
bearing in mind that

10
C!l
!:

8
l-
I
C!l

w 6
:3

W
W
0:::
lJ... 4
I
I
(J)
<I
2
z:
<I
w
:c
0
o N0 J F M A M J J A SON 0 J F M A M J J A SON 0 J F M A M J J A S
1978 1979 DATE 1980 1981

FIGURE 2. Mean ash-free dry weights for the 1977 through 1981 cohorts.
664

TABLE 1. Annual Production (gm m- l ) TABLE 2. Comparison of Caloric and AFDW

Production Determinations - Instantaneous
Nov. 78- Nov. 79- Aug. 80- Growth Method
Nov. 79 Nov. 80 Aug. 81
Removal P(m-lyr- l ) B(m- l ) pIB
Summation 64.05 77 .14 66.01 Nov 78-79
AFDW (gm) 62.26 39.04 1.59
Increment kcal (AF) 363.4 222.7 1.63
Summation 64.05 76.57 66.00 Nov 79-80
AFDW (gm) 79.57 54.82 1.45
Instantaneous kcal (AF) 460.9 269.0 1.71
Growth 62.26 79.57 71. 75 Aug 80-81
AFDW (gm) 71.75 50.18 1.43
Size- kcal (AF) 421.9 286.8 1.47
Frequency 99.97 125.35 134.29
caloric production follows a similar pattern. Production rates of individual cohorts (Fig. 5)
Production rates were calculated as grams per show a small rise in production during late
metre per month and these data were then summer as the cohort is recruited, followed by a
smoothed (4253H, twice; Velleman, Hoaglin, decline during the winter months; maximum cohort
1981) (Fig. 4) to eliminate noise and to production occurs during the first summer after
discern the underlying seasonal patterns. recruitment, falls again during the winter and
Production rate peaks during the summer months has the possibility of a third peak during the
at an average of 9.4 gm'm-umonth- l as compared early part of the last summer. Production and
to 2.7 gm·m-l.month- l during the remaining 8 biomass data can be converted to a per square
months, with the result that two-thirds of the metre basis by dividing the values by the mean
total annual production occurs during summer. width of the distribution of

5500

CAL = 4391 - 250 Sin('WT) - 11.5 cos (~T)

I-
:3 r= .75
5100
>-
ll<:
a
I: 4700

I
<I
ll<:
(!)

ll<:
W 4300
Cl..
(J)
W
3900
ll<:
Q
-l
<I
U
3500
M J J A S 0 N 0 J F M A M J J A
1981 1982
DATE
FIGURE 3. Mean monthly caloric content and 95% confidence intervals for Haustorius canadensis, with
fitted periodic regression based on means. Number of size classes for each month ranged from 8 to
11.
 665

40
35
D 30
I:
'- 25
I:
'- 20
I:
(!) 15
10
W
I- 5
([
Q::
0
z: -5
D
-10
I-
U -15
:::)
0 -20
D
Q::
a.. -25
-30
aND J F M A M J J A saN D J F M A M J J A saN D J F M A M J J A s
1978 1979 DATE 1980 1981

FIGURE 4. Population production rate. Solid dots are observed values. Open circles and line are
smoothed data.
Haustorius canadensis (70m); yielding an
2 1
average production of 1.02 gm·m- ·yr- and mean
biomass of 0.69 gm·m- 2 •
4. DISCUSSION
Percy (1979) investigated the biochemical
composition of the amphipod Onisimus affinis
and compared its caloric content (3290 - 4180
ca1.·gm- 1 ) with those of other amphipods, which
12
ranged from 2319 cal.·gm- 1 to 5807 cal.·gm- 1 dry
weight. The caloric content of Haustorius
canadensis (4400 cal.·gm-1) lies near the middle
of this range. Hastings (1981) gave ranges of
3110 - 4665 cal.·gm- 1 for Ampe1isca brevicornis,
and 4785 - 6100 cal.·gm- 1 for Urothoe
brevicornis in the U.K. Cederwall (1977) gave a
caloric content of 5346 cal.·gm- 1 for
Pontoporeia affinis. Williams and Robins

W
I-
([
Q::
o
z:
D

I-
U -4
:::l
o
D
Q::
a..
-8
aND J F M A M J J A SON D J F M A M J J A SON D J F M A M J J A S
1978 1979 DATE 1980 1981

FIGURE 5. Smoothed production rates for the 1977 through 1980 cohorts.
666
(1979) gave a caloric content for the hyperiid
amphipod Parathemisto gaudicaudi of 5138
ca1.·gm- 1 ash-free, while Strong and Daborn
(1978) gave an ash-free value of 5303 ca1.·gm- 1
for the isopod Idotea baltica.
Haustorius canadensis shows a marked
seasonal pattern in mean monthly caloric
content. Other authors (Hastings, 1981; Percy,
1979; Strong, Daborn, 1978; Tyler, 1973) have
observed seasonal differences in caloric
content for several crustaceans, and have
attributed these changes to reproductive
activity. In contrast, peak caloric content
occurs at the end of, rather than prior to, or
early in the reproductive season in H.
canadensis, which suggests that the observed
pattern is not due to reproduction. An alter-
native hypothesis relates to the quantity and
quality of food available during the year.
Haustoriid amphipods feed by filtering out
particulate matter from between the sand grains
(Croker, 1967; Sameoto, 1969). At Long Sands,
the primary food source for H. canadensis is
pennate diatoms (Donn, pers. obs.). If the
diatoms present show an increase in abundance
andlor nutritional quality during the summer
months, this will explain the observed pattern
of caloric content.
The ratio of annual production to mean
biomass offers a convenient method for
comparing relative productivities of differing
populations or communities (Ansell et a1.,
1978). The mean pin ratio of 1.49 for
Haustorius canadensis falls within the range
observed for other amphipods with two year
1ifespans. In the Baltic (Cederwal1, 1977),
Pontoporeia affinis and ~ femorata, both with
two year lifespans, have pin ratios of 1.90 and
1.43, respectively. In Nova Scotia, where ~ seasonal cycling of caloric content did not
femorata has an annual life history, Wild ish
and Peer (1981) calculated plB ratios of 3.64
to 4.78. Ampelisca brevicornis, with an annual
life history, has pin ratios of from 3.05 to 4
(Hastings, 1981; Klein et a1., 1975). Hastings
(1981) also listed values of 0.94 to 1.58 for
Urothoe brevicornis with a two year life cycle.
The relationship between pin ratios and lifespan
has also been documented for other animal groups
(Ansell et al., 1978; Robertson, 1979; Waters,
1979).
Haustorius canadensis composes 20 - 25% of
the total macrofaunal standing crop of 4.3 gm·m-
2 at Long Sands (Croker, 1977). If we assume that
the amphipods Acanthohaustorius millsi and Amphi-
poreia virginiana, which make up almost all the
remaining biomass, have plB ratios simmilar to
H. canadensis; then we may calculate total com-
munity production as 6.4 gm·m- 2 yr- l , which is
comparable to the Pontoporeia spp. community in
the Baltic (Cederwall, 1977).
Ansell et al. (1978) recognised the impor-
tance of the patterns of elimination and produc-
tion occurring throughout the life history of an
organism. Elimination rates yield information
on patterns of mortality within the population,
while production rates give information on the
accumulation of biomass available to other tro-
phic levels. Two-thirds of the annual pro-
duction of Haustorius canadensis occurs during
the summer months; most of which is attributable
to the 1 year age class with minor contributions
by the 0 and 2 year classes. Klein et al.
(1975) found maximum elimination by Ampelisca
brevicornis to occur during October to December,
during which time approximately 60 - 70% is
consumed by predators.
In summary, the mean caloric content and
pin ratios for Haustorius canadensis fall within
the range expected for boreal amphipods with two
year life cycles. Caloric content follows a
seasonal pattern which is related to food avail-
ability rather than reproductive activity. The
cause a difference between the pin ratios cal-
culated based on caloric content as compared to
those based on ash-free weight. There is need
for further work on temporal patterns of produc-
tion and elimination rates as these may provide

information on the efficiencies of energy
transfer between trophic levels and therefore,
on the structuring of marine communities.
ACKNOWLEDGEMENTS: The presentation of this
paper was made possible in part by the Dept. of
Zoology, Graduate School, and Marine Program of
the University of New Hampshire; and by a grant
from the Symposium Secretariat, C.S.I.R. South
Africa to the first author.
5. LITERATURE CITED
Ansell, A.D., D.S. McLusky, A. Stirling,
and A. Trevallion. 1978. Production and energy
flow in the macrobenthos of two sandy beaches
in south west India. Proc. Roy. Soc. Edin. 76B,
269-296.
Cederwall, H. 1977. Annual macrofauna
production of a soft bottom in the northern
Baltic proper. in Keegan, B.F., P.O. Ceidigh,
and P.J .S. Boaden (eds.) Bio logy .2i. Benthic
Organisms. Permagon Press; Oxford. pp. 155-164.
Clarke, M.R.B. 1980. The reduced major
axis of a bivariate sample. Biometrika 67,
441-446 •
Crisp, D.J. 1971. Energy flow measure-
ments. in Holme, N.A., and A.D. McIntyre (eds.)
Methods for the Study of Marine Benthos.
Blackwell Scientific Pub.; Oxford. pp. 197-279.
Croker, R.A. 1967. Niche specificity of
Neohaustorius schmitzi and Haustorius sp.
(Crustacea: Amphipoda) in North Carolina. Ecol.
48, 971-975.
_____• 1977. Macro-infauna of northern New
England marine sand: Long term intertidal
community structure. in Coull, B.C. (ed.)
Ecology of Marine Benthos. Belle W. Baruch
Library in Marine Science #6, Univ. of S.
Carolina Press.
Cushman, R.M., H.H. Shugart, Jr., S.G.
Hildebrand, and J.W. Elwood. 1978. The effect
of growth curve and sampling regime on
instantaneous-growth, removal-summation, and
Hynes/Hamilton estimates of aquatic insect
production: A computer simulation. Limnol.
Oceanogr. 23, 184-189.
Dexter, D.M. 1969. Structure of an
intertidal sandy-beach community in North
Carolina. Chesapeake Sci. 10, 93-98.
Gillespie, D.M., and A.C. Benke. 1979.
Methods of calculating cohort production from
field data - some relationships. Limnol.
Oceanogr. 24, 171-176.
Hamilton, A.L. 1969. On estimating annual
production. Limnol. Oceanogr. 14, 771-782.
Harding, J.P. 1949. The use of probability
paper for the graphical analysis of polymodal
frequency distributions. J. mar. bioi. Ass.
U.K. 28, 141-152.
667
Hasselblad, V. 1966. Estimation of parame-
ters for a mixture of normal distributions.
Technometrics 8, 431-444.
Hastings, M.H. 1981. The life cycle and
productivity of an intertidal population of the
amphipod Ampelisca brevicornis. Est. Coast.
Shelf. Sci. 12, 665-677.
Klein, G., E. Rachor, and S.A. Gerlach.
1975. Dynamics and productivity of two popula-
tions of the benthic tube dwelling amphipod
Ampelisca brevicornis (Costa) in Helgoland
Bight. Ophelia 14, 139-159.
McLachlan, A. 1980. The definition of sandy
beaches in relation to exposure: A simple rating
system. S. Afr. J. Sci. 76, 137-138.
Percy, J.A. 1979. Seasonal changes in
organic composition and caloric content of an
Artic marine amphipod Onisimus (=Boeckosimus)
• affinis H.J. Hansen. J. expo mar. BioI. Ecol.
40, 183-192.
Ricker, W.E. 1973. Linear regressions in
fishery research. J. Fish. Res. Bd. Can. 30,
409-434.
Robertson, A.I. 1979. The relationship
between annual production:biomass ratios and
lifespans for marine macrobenthos. Oecologia
38, 193-202.
Sameoto, D.D. 1969. Comparative ecology,
life histories, and behavior of intertidal sand-
burrowing amphipods (Crustacea: Haustoriidae) at
Cape Cod. J. Fish. Res. Bd. Can. 26, 361-388.
Shelton, C.R., and P.B. Robertson. 1981.
Community structure of intertidal macrofauna on
two surf-exposed Texas sandy beaches. Bull. Mar.
Sci. 31, 833-842.
Strong, K.W., and G.R. Daborn. 1978.
Seasonal variation in ash and calorific content
of Idotea ba 1 t ica (Pallas) (Crustacea:Isopoda).
Can. J. Zool. 56,1917-1921.
Tyler, A.V. 1973. Caloric values of some
North Atlantic invertebrates. Mar. Bioi. 19,
258-261.
Velleman, P.F., and D.C. Hoaglin. 1981.
Applications. Basics. and Computing of Explora-
tory Data Analysis. Duxbury Press; Boston.
354pp.
Waters, T.F. 1979. Infulence of benthos
life history upon the estimation of secondary
production. J. Fish. Res. Bd. Can. 36, 1425-
1430.
_____• 1981. Seasonal patterns in produc-
tion and drift of Gammarus pseudolimnaeus in
Valley Creek, Minnesota. Ecol. 62, 1458-1466.
Wildish, D.J., and D. Peer. 1981. Methods
for estimating secondary production in marine
Amphipoda. Can. J. Fish. Aquat. Sci. 38, 1019-
1026.

BEHAVIOURAL AND PHYSIOLOGICAL RESPONSES OF A BURROWING BIVALVE TO STRESS
E.R. TRUEMAN (Zoology Department, The University, Manchester, M13 9PL, U.K.)
1. INTRODUCTION
S~obiQ~ plana (da Costa) is a tellinacean
bivalve found abundantly in muddy sand in
estuaries and on certain stable temperate zone
shores. Development of electronic recording
techniques (Trueman, Akberali, 1981) has made
possible studies of its behaviour in respect of
heart rate, valve movements and ventilatory
pumping both in the field and in the laboratory
under conditions simulating the natural
environment (Earll, 1975; Akberal i, 1978). In
many bivalves studied, e.g. Mya, MytitU6, O~tnea
(Thompson, Bayne, 1972; Walne, 1972), changes in
the level of pumping elicit little variation in
heart rate but in S~obiQ~ the heart rate
falls markedly with reduction of water flow.
Such increased sensitivity to pumping makes
S~obiQ~ a particularly suitable subject.
This article outlines the normal behaviour of
S~obiQ~ and how it may be affected by
stress caused by changing salinity, aerial
exposure, heavy metal or insecticide pollutants.
The consequential physiological events are
discussed together with the mechanisms of
detection of environmental stressors. Although
S. plana is found on stable beaches the tech-
niques used and general results are equally
applicable to many infaunal bivalves.
2. BEHAVIOURAL RESPONSES
The impedance technique (Trueman et al., 1973)
was used to record valve movements and the heart
rate of S~obic~ while a thermistor flowmeter
669
carefully placed in the exhalent water current
(Foster-Smith, 1976) allowed gill pumping to be
monitored. Simultaneous recordings of these
features on a multichannel pen recorder showed a
succession of activity cycles. Each cycle
shows a short period of pumping (8-60 min)
followed by a ventilatory pause of 10-45 min
duration. When pumping the clams exhibit a
high and clearly defined heart rate of up to
18 beats per min; bradycardia occurs during
each ventilatory pause, rates of 12 per min
being commonly observed (Earll, 1975; Akberal i,
1978). During pumping the valves may
frequently close and open rapidly, possibly to
eject pseudofaeces from the mantle cavity
(Akberali, 1978, Fig. 1).
When S~obicut~ is continually immersed in
seawater of normal composition the valves may be
adducted and held apparently completely closed
so that the clam is effectively isolated from
the external medium. This occurs at intervals,
which cannot be readily related to environmental
factors, and has been termed a quiescent period.
These periods occupy about 15% of the total
duration of experimental observations and vary
in duration, commonly being between 4 and 12 hrs.
Quiescence has been observed in other species
notably in the sublittoral bivalve ~ctica
~landica where it occurs apparently spontaneous-
ly (Taylor, 1976) and in littoral bivalves in
response to either aerial exposure or salinity
fluctuations (Akberali, 1978).
670
It was convenient to control quiescence in
S~ob~cU£~, by the appl ication of rapid changes
in the salinity of the bathing medium, to study
the effects of valve closure. A standard
salinity stress was applied by an immediate
reduction of salinity from normal to 20% sea-
water (30-6%0)' This results in valve closure
and inactivity while the tissues are effectively
isolated from the surrounding water. The water
was carefully siphoned in and out of a small
aquarium containing the clam. Control tests
were run using standard seawater throughout which
had no more than a momentary effect on the
animal's behaviour. The response of S~ob~cU£a-
~ whether to salinity change, pollutant stress
or aerial exposure was the same, namely siphonal
retraction and valve closure for up to 7 days
although in most experiments the valves gaped
for brief periods after 5 days. Th is allows
the clam to avoid pollutant or salinity stress
provided the application is of shorter duration
than 5 days. When the salinity of the medium is
reduced to 20% seawater the valves gape for
lengthening periods after 7 days and pumping
activity gradually increases in duration to more
than 50% of the time as the animal equilibrates
with the medium (Akberali, 1978). When tested
in mixtures of pollutants and seawater, for
example copper solutions, S~ob~cu~ retracts
the siphons and closes the valves at concentra-
tions of 0.01 ppm. This is followed by a rapid
fall in heart rate similar to that of My~Uh and
other bivalves (Manley, Davenport, 1979). After
2 or 3 hrs the clams commence to interact with
the medium but at higher concentrations, e.g.
0.5 ppm, the valves remain closed and the heart
rate low but replacement of the medium with sea-
water after 6 hrs leads to recovery within 10-15
mins. Should the pollutant continue to be in
the medium after 7 days valve gaping occurs and
is rapidly followed by death (Akberali, Black,
1980). This period is thus the maximum
duration of closure in Scnob~cU£~a and it is
likely that accumulation of anaerobic metabolites
is the critical factor. Similar responses have
been obtained with zinc ions (Akberali ct at.,
1981) and the insecticide Sevin (Akberali, Black,
personal communication).
3. PHYSIOLOGICAL RESPONSES
Valve closure during a natural period of
quiescence, salinity or pollutant stress results
in anaerobic respiration when, in common with
other bivalves, Scnob~cU£~ produces succinic
acid, alanine and other volatile fatty acids
(De Zwaan, Wijsman, 1976). During quiescence
the pH of the mantle cavity is gradually reduced
from pH 7.8 to 7.0 as acid metabolites accumulate.
In 2 hrs after valve closure the p02 in the
mantle cavity falls from 140 to about 55 mm Hg
but after 3 hrs a temporary increase of 20 mm Hg
occurs (Trueman, Akberali, 1981, Figs 3, 4).
Although the valves appear closed, this is
probably due to a slight opening of valve and
mantle margins allowing diffusion for it does not
occur when the valves are forcibly sealed. During
aerial exposure Scnob~cU£~a also shows a com-
parable increase in oxygen tension after 3 hrs
except when the valves are forcibly closed.
Oxygen tensions of less than 50 mm Hg have not
been observed in the mantle cavity of quiescent
animals and it seems likely that at this level
anaerobic respiration will have taken over.
During the same initial 2 hrs after valve closure
the pCo 2 increases from 2 to 10 mm Hg when the
valves are forcibly sealed, 5 mm Hg when closed
in response to an applied salinity stress. This
suggests that the lower levels are stabilised by
outward diffusion although the valves are
apparently closed. Recovery from quiescence
commonly involves the repeated rapid adduction
and relaxation of the valves and an overshoot in
the heart rate. This has been explained as
effecting repayment of an oxygen debt or the
flushing out of excretory products (Brand, Roberts,
1973). During recovery the pH of the mantle
cavity returns to 7.8 but shows a series of steps
each corresponding to valve adduction. It is

thought that one result of adduction is the rapid
flushing out of metabolites from the tissues
whilst between adductions normal ciliary pumping
removes these from the mantle cavity (Trueman,
Akberal i, 1981).
When seawater of low salinity is used to stress
S~obieul~a the concentration of all ions in
the blood and mantle cavity water, with the
exception of Ca 2+, fall to the level of the
external medium over about 14 days. Blood
calcium rises to about three times the
concentration of seawater (30 mM) within 5-7 days
when it drops towards that of the external medium
to stabilise at a new level after 18 days
(Akberal i et aT., 1977). Similar changes were
observed in the mantle cavity fluid. To
determine whether the increase in calcium ions
was a result of anaerobic metabolism clams were
placed in oxygen-free 100% seawater when they
exhibited only a slight rise in calcium levels
(Trueman, Akberali, 1981, Fig. 6). Only when
the valves close in response to 20% seawater do
the calcium levels increase markedly. Continuous
flow of normal, but oxygen-free, seawater through
the mantle cavity may explain the absence of a
significant rise in calcium ions for no anaerobic
metabolites would accumulate. Only when the
valves are closed would these accumulate and the
role of the increased calcium ions would appear
to be to buffer acid metabolites in the mantle
cavity and blood. Calcium is obtained from
the interior of the valves by dissolution
(Akberali, 1980) and the application of stress
for 7 days results in the mobilization of
tightly bound calcium from the shell, a decrease
in shell weight and weakening of the valves
(Akberali et al., 1983). Electron microscope
studies of the inner surface of the valves of
S~obieul~a from the mid shore, by the same
authors, have shown erosion of the surface after
aerial exposure for 6 hrs followed by a smooth
calcified surface after a similar period of
671
immersion. This suggests cyclical dissolution
and secretion of the valves caused by periods of
anaerobiosis consequent on tidal exposure.
4. DETECTION OF ENVIRONMENTAL VARIABLES
Little is known about sensory receptors in the
Bivalvia apart from the presence of pedal
statocysts, Frenkiel 's work (personal communica-
tion) on the role of the cruciform muscle organ
in mechanoreception and ultrastructural studies
on the ciliated receptors of the pallial tentacles
of P£aeopeete~ (Moir, 1977) and Lima (Owen,
McCrae, 1979). However the immediate response
of S~obieul~ to salinity and other changes by
siphonal retraction followed by valve closure
implies rapid detection on the extended inhalent
siphon or mantle lobes where they are in
immediate contact with the inhalent water current.
Se~obieul~a responds to falling salinities by
detecting changes in the concentration of ions
rather than to osmotic pressure changes or to
changes in the concentration of a single ion.
This animal is sensitive to Na+, Mg2+, and Ca 2+
(Akberali, Davenport, 1982). Ablation of the
visceral ganglion completely abolishes the closing
response of siphonal retraction to low salinities
(Akberali, Davenport, 1982) and no reaction is
obtained from the isolated siphon (Akberali et
al., 1981). In both experiments the sensory-
motor reflex arc will have been destroyed and
they contribute little to our understanding of
the location of the sensory receptors. Other
experiments have shown that siphonal retraction
continues to occur as a response to low
salinities when the cruciform muscle complex,
foot or gills are removed (Akberali, Davenport,
1982). These authors suggest that the sensory
receptors are located deeply on the mantle
tissues and that the vulnerability of
S~obicul~ to periodic siphon loss because of
predation is the reason for this. In MytiiUh,
however, the salinity receptors are on the
tentacles of the inhalent siphon allowinq low
672
salinity to be detected peripherally before
penetration of the mantle cavity (Davenport,
1981). A comparably efficient system is to be
expected in S~ab~Quiania and other Tellinacea.
Consideration of the receptors effecting shell
opening when normal water replaces polluted or
low salinity seawater suggests that they are
likely to be located on the marginal folds of the
mantle for although the values may appear tightly
closed the opening response occurs within 2-3 min.
(Akberali, Black, 1980). It is probable that
very slight valve movements occur periodically to
test the medium using receptors on the marginal
mantle folds rather than allowing the medium into
the mantle cavity. Such small valve movements
have not however been observed.
Further evidence regarding the distribution of
receptors in the pallial region is obtained from
scanning electron microscope studies. Hodgson
and Fielden (in preparation) have observed three
types of ciliated sensory receptors on the mantle
and siphons of Vanax ~~ and V. ~and{d~ and
have established that these are primary receptor
cells. These receptors occur both inside and
outside both siphons but most abundantly on the
Siphonal tentacles at densities in excess of
9 x 10 3 mm- 2 • Similar receptors occur in
SQnab~Qui~ but are not so significantly
concentrated on the tips of the siphons (Black,
personal communication). Attempts to study the
response of groups of ciliary receptors using
standard electrophysiological techniques have not
so far been successful in recording sensory
potentials in S~ab~Quiania. However, motor
potentials were satisfactorily observed but it
was not possible to determine which receptors
were concerned. There are two principal
functional possibilities for the ciliary
receptors, chemoreception or mechanoreception in
respect of suspended particulate material but it
is currently impossible to differentiate these
functions. A similar conclusion was reached by
Brown and Noble (1960) in respect of the
functioning of ciliary receptor cells in the
osphradium of Buttia. Nevertheless there are
certainly ample siphonal and mantle margin
sensory receptors in the bivalve species
mentioned to account for the behavioural
responses observed above.
5. DISCUSSION
The ability of S~ab~Q~a to close its valves
and to remain isolated from the environment yet
to continue to respond to external changes
suggests that this clam is particularly well
adapted to estuarine or littoral life. There
are two critical factors for this, first the
ability to hold the valves closed and to respire
anaerobically and secondly the rapid detection
of changes in the medium. These features occur
in a number of infaunal bivalves such as species
of Ven~ (A. Jenkins, personal communication),
An~Qa (Taylor, 1976) and Cen~zadenma. Little
is known of bivalves from high energy shores in
this respect although preliminary observations
by Dr A.N. Hodgson (personal communication)
suggest that V. ~enna respires anaerobically
when the valves are closed. This might seem
surprising in a species of a surf-washed zone
but the ability may be important for the
survival of clams stranded high on the beach.
From these studies it is clear that many bivalves
have the capability of anaerobic respiration and
due account should be taken of this in studies
of energy flow in molluscan populations. A
more difficult problem is the assessment of
what, if any, proportion of respiration in an
active clam is anaerobic. There are currently
few observations on this subject.
Simple experiments (A. Jenkins, A.R. Brand,
personal communication) during investigations of
the feeding behaviour of A~Znape~en suggest that
Ven~ ~~a may survive under anaerobic
conditions for a week or more while Tettina

6abula closes the valves for only a brief
period and does not appear to respire anaerobi-
cally. A~thop~~~w only consumes T. 6abula in
aquaria without sand for a shallow layer of sand
allows this species to escape. Tettiwa opens
its valves and is rapidly digested after
ingestion but when V~wUh is eaten it may survive
for a week or more in the stomach of the starfish
with the valves closed. This suggests two
alternative strategies by which bivalves respond
to stress, either rapid locomotion and escape by
species with streamlined thin shells or the
closure of thickened valves coupled with
anaerobic respiration and the ability to remain
quiescent until the stressor may disappear.
Further simple experiments with a wide range of
sessile and active infaunal bivalves are required
to elucidate this further.
6. ACKNOWLEDGEMENTS
I gratefully acknowledge the contributions made
to this article by my past and present research
students, in particular the active help of
Dr H.B. Akberali and the support of NERC grants.
7. REFERENCES
Akberali HB (1978) Behaviour of S~ob{cul~
plawa (da Costa) in water of various salinities,
J. expo mar. Biol. Ecol. 33, 237-249.
Akberali HB (1980) 45Calcium uptake and
dissolution in the shell of S~ob{cula~a plawa
(da Costa), J. expo mar. Biol. Ecol. 43, 1-9.
Akberali HB (1981) Effects of copper (Cu 2 +) on
an isolated tissue preparation from the bivalve,
S~ob{cul~a plawa (da Costa), J. expo mar.
Biol. Ecol. 52, 115-120.
Akberali HB and Black JE (1980) Behavioural
responses of the bivalve S~ob{cul~ plawa (da
Costa) subjected to short-term copper (Cu II)
concentrations, Mar. envir. Res. 4, 97-107.
Akberali HB, Brear K and Currey JD (1983)
Mechanical and morphological properties of the
shell of S~ob{cul~ plawa (da Costa) under
normal and stress conditions, J. Moll. Studies,
in press.
Akberali HB and Davenport J (1982) The detection
of salinity changes by the marine bivalve molluscs
S~ob{cul~a plawa (da Costa) and My~~ ~d~
L. , J. expo mar. Biol. Ecol. 58, 59-72.
Akberali HB, Marriott KRM and Trueman ER (1977)
Calcium utilization during anaerobiosis induced
by osmotic shock in a bivalve mollusc, Nature,
673
Lond. 256, 852-853.
Akberali HB, Wong TM and Trueman ER (1981)
Behavioural and siphonal tissue responses of
Sc~ob{cu!a~ plawa (Bivalvia) to zinc, Mar.
envir. Res. 5, 251-264.
Brand AR and Roberts D (1973) The cardiac
responses of the scallop P~~~w max{mu~ (L.) to
respiratory stress, J. expo mar. Biol. Ecol. 13,
29-43.
Brown AC and Noble RG (1960) Function of the
osphradium in B~ (Gastropoda), Nature,
Lond. 188, 1045.
Davenport J (1981) The opening response of
mussels (My~~ ~d~) exposed to rising sea
water concentrations, J. mar. Biol. Ass. U.K.
61, 667-678.
Earll R (1975) Temporal variation in the heart
activity of S~ob{cul~a plawa (da Costa) in
constant and tidal conditions, J. expo mar. Biol.
Ecol. 19, 257-274.
Foster-Smith RL (1976) Some mechanisms for the
control of pumping activity in bivalves, Mar.
Behav. Physiol. 4, 41-60.
Manley AR and Davenport J (1979) Behavioural
responses of some marine bivalves to heightened
sea water copper concentrations, Bull. Environm.
Contam. Toxicol. 22, 739-744.
Moir AJG (1977) Ultrastructural studies on the
ciliated receptors of the long tentacles of the
bivalve Uma h{a~ (Gmelin) with a note on the
long tentacles of the giant scallop Placop~ct~w
mag~!!a~cu~ (Gmelin), Cell Tissue Res. 184,
367-380.
Owen G and McCrae JM (1979) Sensory cell/gland
complexes associated with the pallial tentacles
of the bivalve L{ma ~a~ (Gmelin) with a note
on specialised cilia on the pallial curtains,
Phil. Trans. R. Soc. Lond. B. 287, 6-62.
Taylor AC (1976) Burrowing behaviour and
anaerobiosis in the bivalve A~~ca ~lawd{ca
({.), J. mar. Biol. Ass. U.K. 56,95-109.
-Thompson RJ and Bayne BL (1972) Active
metabolism associated with feeding in the mussel
My~Uh ~d~ L., J. expo mar. Biol. Ecol. 9,
111-124.
Trueman ER and Akberali HB (1981) Responses of
an estuarine bivalve S~ob{cul~ plawa
(Tellinacea) to stress, Malacologia, 21, 15-21.
Trueman ER, Blatchford JG, Jones HD and Lowe G
(1973) Recordings of heart rate and activity of
molluscs in their natural habitat, Malacologia
14, 377-383.
Walne PR (1972) The influence of current speed,
body size and water temperature on the filtration
rate of five species of bivalves, J. mar. Biol.
Ass. U.K. 52, 345-374.
Zwaan A De and Wijsman TCM (1976) Anaerobic
metabolism in Bivalvia (Mollusca). Characterist-
ics of anaerobic metabolism, Compo Biochem.
Physiol. 54B, 313-324.
 675

SOME ASPECTS OF THE ECOPHYSIOLOGY OF SCAEVOLA THUNBERGII, A SUBTROPICAL COASTAL DUNE PIONEER

N.W. PAMMENTER (Department of Biological Sciences, University of Natal, Durban, South Africa)

1. INTRODUCTION

Along the eastern coast of Southern Africa

Scaevola thunbergi i Eckl & Zeyh. is an

important primary coloniser of open beaches,

building dunes in much the same manner

as marram grass in temperate Europe.

The plant has erect stems up to 80 cm

above ground level wi th large (up to

120mm x 70mm) broadly elliptic to obovate,

sessi Ie leaves (Figs. and 2). Internode

length is variable (5 to 30 mm) and

possibly may depend upon the rate of

sand accret ion.

FIGURE 2. Stems of Scaevo I a thunberg i i
trapping wind-blown sand.

The plant shows sequential leaf senescence

with 15 to 25 leaves per stem (Fig. 3) and

leaves become more succulent with age (Fig.

4). The stems and leaves trap wind blown

sand and grow through the accumulated

sand, branching frequently. Adventitious

roots are produced from the stems as they

become covered. A densely covered fore-

dune, up to 10 m in height, will have an

interwoven network of below-ground stems

that may have originated from only two

or three plants.

The components of the physical environment

FIGURE 1. Stems of Scaevola thunbergi i of coastal sand dunes, with particular re-
showing leaf shape and arrangement.
ference to tempera te regions, have been
(Fully expanded leaves approximately
100mm x 60mm) • reviewed extensively by Chapman (1976)

and Ranwell (1972) . Conditions in sub-

tropical areas are similar, except for higher

676

Donnelly, Pammenter, 1983).

100
There has been considerable discussion con-
cerning the water status of coastal sand
E
..•
E 75 dunes (Chapman, 1976) and because of the

E high porosity of the soils there have been

:::. 50
suggestions that sand dunes can be consider-
ed to be edaphically dry habitats. However,
because of the coarse na ture of the so i Is
25
it is not until water content drops below
1.5% that soil water potential drops apprec-
iably. Thus almost al i the water present
10 15 20
Leaf number
in sand dune soi Is is potentially avai lable
FIGURE 3. Dry mass of individual leaves
to plants. Gravimetric water contents that
of S. thunbergii. Leaf is the smallest
(youngest} leaf. Mean values from plants have measur.ed at depths greater than 15cm
with 19-21 leaves, harvested in Apri I. Dry
range between and 15%, with the majori ty
masses are expressed as a % of the maximum
size for each plant. of values fall ing in the range 2 - 8%.

Sand dune soils are generally low in the

10 macronutrients, N, P and K but appear to
have adequate quantities of the micronutrients
(Chapman, 1976). Table 1 shows the range
of soi I nutrient contents have measured
on the Nata I coast compared with the ranges
reported in a wide variety of soi Is (Allen
et ai, 1974). Contents of Ca, Mg and P
of the sand dune soi Is are well within these
ranges. Levels of K are low and those of
10 15 20 N are extremely low. Despite the deposition
L.af numb.r
of salt spray, soil Na levels are not except-
FIGURE 4. Fresh mass/dry mass ratio of
ionally high; they never approach levels
individual leaves of S. thunbergi i.
associated with saline soils (Donnelly, Pam-
menter, 1983). this is probably as a result
tempera tures and, in some regions, lower
of the rapid leaching through the .porous
rainfall. High day-time summer temperatures,
soils (Chapman, 1976).
high I ight intensities (as a result of the
high albedo of bare sand) and high wind
Questions arise concerning physiological adap-
speeds result in high potential evapotrans-
tations to the environment possessed by the
piration and may give rise to plant water
plants of coastal sand dunes and concerning
stress. Associated with the high winds are
growth and reproductive patterns in relation
wind-blown sand and salt spray. Sa I t spray
to the habitat in which they naturally occur.
may be a causative agent in the vegetation
This report presents some prel iminary results
zonation frequently observed on coastal sand
of ecophysiological and autecological studies
dunes (Chapman, 1976; Barbour, 1978;
on S. thunberg i i attempt i ng to answer some
 677

TABLE 1

Ranges of nutrient contents of "normal" soi Is comp_'1red with

measured values for sand dune soi,ls (m mol 100 g air dry)

Na Ca p K N

"Normal" soi Is 0.1 - 1.0 0.2 - 5.0 0.2 - 2.0 0.01 0.3 0.1 - 1.0 8 - 35
Dune soi Is 0.1 - 0.6 0.1 - 0.4 0.01 - 0.2 0.04- 0.1 0.1 - 1.0

of these questions. was measured using a pressure chamber,

transpiration by short term mass loss of

2. WATER RELAT IONS
cut twigs, and leaf conductance using a
The water status of S. thunbergi i, together The well
diffusion res i stance porometer.
with air temperatures and saturation vapour
known phenomenon of water potential decl ine
deficits (SVD) over a 24 hour period in the
during the day with night-time recovery
field are shown in Fig. 5. (SVD is the
is apparen t. AI though leaf conductances
difference between saturation vapour concent-
at night were lower than during the day,
ration (or pressure) at ambient temperature
complete night-time stomatal closure did not
and the absolute vapour concentration (press-
occur. Transpiration declined to zero at
ure) . It is a direct measure of the dryness
night and responded early in the morning
of the air and is a more su i tab Ie measure
to light and increases in temperature and
than relative humidity). Water potential
SVD.

24

U
..
0
~ 22
E
:
Q
>
c 20
III

a
0-3

";
u-.
~

·
+0
: .2
E
'.E :
...~ ~ 0·3 -'-2 !
~
~
•
u• .
~
c

·
0
! ',1 i 0'2 ·1-4 i
0 8 a;
!'0.
..·• ,,'
'i ~
:!
... '-1 -'-6

12 It al 14
Tim. 01 d.~

FIGURE 5. (a) Environmental parameters and (b) plant water status of S. thunbergi i during
a 24 hr period in October. (a) 0 - - 0 air temperature, . - . SVD
(b) . - - . transpiration, 0 - - 0 I eaf conductance, . - - . water potential.
 678

Transpiration rates can be expressed as transpiration from before midday. They

per unit leaf area, per unit leaf fresh mass attributed this to decrease in stomatal aper-

or per unit leaf dry mass. Only if the fresh ture associated with leaf water deficits.

mass/dry mass ratio and specific leaf area The transpiration rates reported by Willis,

are known can these three be interconverted. Jefferi es ( 1963) are considerably higher

The maximum rate recorded on this occasion than those reported here (maximum values
(0.26 mg (g dry mass)-l s -l is equivalent to 15-20 as opposed to 5-7 mg (g fresh mass)-l min -1).

1.9 9 dm- 2 hr- 1 and 1.7 (g fresh mass)-l min- 1 These authors do not give fresh mass/dry

These measurements were taken on a clear mass or mass/area ratios, but from their

day in spring and transpiration rates were data it is possible to estimate fresh mass/

not particularly high. Measurements taken dry mass ratios that are simi lar to those

over shorter time periods in summer (11.00 for S. thunbergii. Thus on a dry mass

hrs to 16.00 hrs) when temperature and SVD basis and probably on a leaf area basis

were higher (average midday values of 32° as well (specific leaf areas are unlikely

and 11.1 9 m -3respectively) yielded average to be vastly different) the max imum trans-

midday transpiration rates of 4.8 mg piration rates measured by Willis, Jefferies

(g fresh mass)-l min- 1 (approximately (1963) are two to three times higher than
-2 -1
5.4 g dm hr ), nearl y three times higher those measured for S. thunbergii.

than the spring values shown in Fig. 5.

From a know ledge of transpiration ra tes
On one exceptional occasion with an air temp-
-3 and plant densities it is possible to estimate
erature of 36° and SVD of 26 g m , a trans-
a water budget for a fore-dune colonised
piration rate of 7.3 mg (g fresh mass)-l min -1
by S. thunbergi i, but it is emphasized that
was recorded. Most reported transpiration
this is a very preliminary estimate. Salisbury
rates for a wide variety of species over
(1952) calculated that an individual plant
a wide range of conditions f a II within the
of Trifol ium arvense growing on beach soi I
range 0.5 to 2.5 g dm- 2 hr -1 (Leopold,
shou Id ut i I ize all the soil water available
I<riedeman, 1973), although an exceptional
to it in four days and invoked the process
value of 13.2 g dm- 2 hr- 1 has been reported
of dew formation within the soil as a water
from the Southern Sahara (Stocker, 1976) .
source for the plant. Measured transpiration
Thus the summer rates of transpiration of
rates of S. thunbergii range from 0 to
S. thunbergii can be considered to be fairly
mg (g dry mass) -1 s -1 The high rates
high. However, on only one occasion was
are confined to a few hours in the middle
midday transpiration rate lower than morning
of the day at the height of summer. Measured
or evening rates. Thus, despite the high
values are frequently lower than 0.3 mg
transpiration rates, midday stomatal closure
(g dry mass) -1 s -1 and thus an average
as a mechan i sm to reduce water loss was
of 0.3 mg (g dry mass)-l s-l for 12 hours for
on I y an infrequent occurrence. It would
each day of the year is a generous est ima te.
appear tha t the water supp I y to S. thunberg i i
From the dry mass of leaves per stem (9.3g)
is generally adequate to meet the environmental
and maximum stem density (20 m- 2 ) the
demand.
daily consumption of water can be estimated

Willis, Jefferies (1963) investigated the water as 2.4 kg m -2 of soi I surface. Excavations

relations of sand dune plants from North to 2 m below soi I surface have shown

Devon and reported marked reductions in apparen t I y healthy adventitious roots at

 679

this depth. Assuming a soil bulk density

of 1.3 and a soi I moisture content of 5%

100
a column of soil m2 in cross section and
•
2 m deep wi II contain 130 kg of water.
•
Assuming that all the water is avai lable

to the plants tnis water is sufficient to last

130/2.4 54 days. The available water

is over-estimated by assuming that all the

soil water is available to the plants and

25
under-estimated by using the maximum stem

density and by assuming that rooting is

I imited to 2 m (the greatest depth excavated);

10 20 30 0 40
Leaf tempera ture ( C)
plants that have grown with a dune that

is 10 m in height are I ikely to have adven-

FIGURE 6. Net photosynthesis (P N ) of S.
titious roots at depths considerably greater thunbergi i as a function of leaf temperature.
The different symbols represent different
than 2 m. The estimates of soi I water supply
plants. Values are expressed as a % of
will be less than 54 days in summer when the maximum achieved by each plant.

transpiration rates are high, but this is

Figure 6 shows the temperature response
the rainy season. Estimates of supply wi II
of net photosynthesis of fu II y expanded
be greater than 54 days in winter, which
leaves of S. thunbergi i. There is a fairly
is the season of limited rain.
wide optimum centred around 26-28°. At

-2 -1 tempera tures that are rarely recorded

A transpiration rate of 2.4 kg m soi I day
during the day on the Natal coast, photosyn-
for 365 days is equivale'nt to 876 mm water,
thesis is sti II about 40% of maximum. Temp-
which is somewhat less than the average
eratures as high as 40° occur only infrequent-
annual rainfall for the Natal coast (Port
Iy but net photosynthesis is nearly 60%
Shepstone 1114 mm, Durban 1011 mm, St.
of maximum at this temperature. Photosynthe-
Lucia 1128 mm). AI though nothing is known
sis of S. thunbergi i appears to be well
about loss of water from the system by deep
adapted to the temperatures that may be
percolation, or the effect of any water table,
exper i enced by the plant throughout the
it does not seem to be necessary to invoke
year. The I ight response curve is shown
dew as a water source for the plants.
in Fig. 7. Saturation occurred at 800-1200fJE
-2 -1
3. PHOTOSYNTHESIS m s or about half full sunlight. Fig.

Isolated plants occurring naturally on the 8 shows the effect of leaf age (position)

beach were transferred to pots and used on net photosynthesis. Photosynthetic rates

in photosynthesis studies when they had declined quite markedly with the old leaves

become re-establ ished. The physiology of showing only 10% of the rates of young

and H 20 vapour exchange was studied expanding leaves. However, the old leaves

on single, attached leaves using standard sti II make a positive contribution to the

infra-red gas analysis techniques. All leaves C economy of the plant and probably continue

that were used for gas exchange measurements to do so until shortly before abscission.

had been produced in the field before the

It is known that the stomata of some plants
plants were potted. respond directly to ambient humidity (Hall
680

declined only marginally, suggesting that

the residual (mesophyll plus carboxylation)

10.
'. resistance to CO 2 uptake is somewhat greater

than the stoma ta I resistance. Ra tes of

75
transpiration measured (0.2 to 3.4 g dm- 2

hr- 1 ) were similar to values obtained in

field studies. The field studies indicated

that stomatal closure in response to water

demand or water stress was infrequent.

However, SVD values of greater than 15

-3
g m were only occasionally encountered.

Additionally, Lud low (1976) has pointed

400 800 1200
PAR (pE m- 2 s-,) out that plants maintained in the field

or under artificial conditions may well show

FIGURE 7. The response of net photosynthesis
of 5. thunbergii to photosynthetically active different responses to water demand.
radiation (PAR). The different symbols
represent different plants.
pressed as a
Values are ex-
% of the max imum achieved The transpiration ratio (g H2 ° transpired/
by each plant. g CO 2 fixed) is a measure of the efficiency

with which a plant uses water. The value

wi II obviously be very dependent upon environ-

1.0 • mental conditions, but there are noticeable

differences between plants wi th d i fferen t

photosyn thet i c pathways. Transpiration

ratios for C3 plants are generally within

the range 450 - 600, C4 plants 250-350 and

CAM plants, when they are fixing CO 2 at

night, 25-150 (Szarek, Ting, 1975). The

25
value of this ratio for 5. thunbergii under
•
the conditions employed in this investigation
-2 -1)
I. 15 20 (excluding light levels below 100fJE m shad
leaf number
+
a mean of 133 - 63. Water use efficiency
FIGURE 8. The influence of leaf age on
varied with leaf age, the young expanding
net photosynthesis of 5. thunbergii. Leaf
1 is the youngest I eaf. Va I ues expressed leaves being two to three times as efficient
as a percentage of the youngest leaf tested.
as the old, pre-senescent leaves.
The different symbols represent different
plants.
Rates of net photosynthesis under optimal
et a I, 1976) • The response of both net
conditions were about 20 mg dm- 2 hr- 1 and
photosynthes i s and transpiration of 5. thun-
the CO 2 compensation point was found to
berg i i to 5VD is shown in Fig. 9. At SVD
be 55 vpm. These values, together with
values above 15 g m -3 transpiration rates
the saturation of photosynthesis at about
remained constant, indicating partial stomatal
half full sunlight and the lack of Kranz
closure. (The significance of the sigmoid
anatomy suggests that 5. thunbergi i has
shape to the curve, rather than a linear
the C3 pathway of photosynthesis (Black,
increase in transpiration up to 15 g m- 3
1973) . Water use efficiency appears to
is unknown.) Rates of net photosynthesis
 681

.
-
~IOO
K
~ •
• •
e • •
•
a·

b
100 •
•
E 75
"
.
E
'w
e
'050
!!

10 25
SvD (gm"'1

FIGURE 9. (a) Net photosynthesis and (b) transpiration (E) of S. thunbergii in response to
saturation vapour deficit (SVD). The different symbols represent different plants, the same
symbol being used for both photosynthesis and transpiration. Values are expressed as a %
of the maximum of each plant.
-1
be very high, higher than most plants and 1.4 mmole g respectively) (Drennan,

an attribute admirably suited to rapidly Pammenter, 1982) • This suggests that S.

groiNing plants in areas of high evaporative thunbergi i does not possess an efficient ex-

demand. elusion mechanism at the roots. (Expressed

on a fresh mass basis the comparison between

4. INTERNAL NUTRIENT CYCLING
marina is not as marked;
S. thunbergi i and A.
S. thunbergi i shows sequential leaf senescence
0.25 and 0.42 mmole g-l respectively). The
and soi I contents of the important macro-
increase in succulence with leaf age (Fig.
nutrients are low (Table 1). Thus some
4) might be related to this continued Na
form of mobilization of inorganic nutrients
accumulation. Leaf P contents initially de-
from old to younger actively growing tissue
clined, indicating a more rapid leaf growth
may be expected. Figs. 10 and 11 shows than P uptake, and thereafter remained
the leaf concentration, on a dry mass basis, reasonably constant with no marked
of Ca, Mg, Na and P for leaves of different re-mobil ization from the older tissues.
ages. Na and Mg were accumulated throughout

the life of the leaf and Ca was accumulated Leaf contents of K and N are shown in Fig.

unti I short I y before fu II leaf expansion. 12. For both nutrients there was an initial

Despite the low salinity of the soils, leaf decline with leaf age, indicating more rapid

Na contents of S. thunbergi i reached very leaf growth than nutrient import. However,

high values, higher than the Na content the decl ine continued after the leaves had

of the mature leaves of the mangrove Avicennia reached full expansion, suggesting a marked

marina, expressed on a dry mass basis (2.5 export from the 0 I der tissues. Plots of total
682

~0·3 0'3
'..
'0
E
E
:0·2 0·2
c:
!c:

-.
0
u

~ 0'1 0'1

o 15 20 10 15 20
Leaf number

FIGURE 10. The contents of Ca and Mg, expressed on a dry mass basis, of individual leaves
of S. thunbergii of different ages. Leaf 1 is the youngest leaf.

'..
Na
'02-0 0·1
E
!
c:
!
Ii
..
u 1.0

...•
0·05

o 10 15 20 10 15 20
Leaf number

FIGURE 11. The contents of Na and P, expressed on a dry mass basis, of individual leaves
of S. thunbergii.

100 2'0

75 '", 1.5 '",

'0
E E
~
E .§
'j(
250 E
!c:
'0

..
0
u
~

~Z5 0·5

• Lea' number

FIGURE 12. Contents of K and N expressed as per uni~ldry mass and total per leaf 0":1 individual
leaves of S. thunbergii. . - - . mmoles (g dry mass) . - - . total, mmoles leaf
 683

K and N contents per leaf against leaf number duri ng summer but leaves were produced

(Fig. 12) show that this export starts before throughout winter at about 60% of the sumrn,er

the leaves have reached fu II expansion. rate. The low rate recorded in October coin-

The decline in net photosynthesis with leaf cides with peak flower production, but this

age (Fig. 8) may be related to the decline may not be significant; fruit growth continues

inN con ten t. until the end of January, the period of maxi-

mum leaf production. It is interesting that

The older leaves were slightly smaller than
leaf longevity (mean 167 days) varies with
the maximally expanded leaves (Fig. 3) •
the date of initiation leaves produced in
It is not known whether this represents a
winter have a I ifespan of over 50% longer
mobil ization of C reserves before abscission,
than those produced in summer (Fig, 14).
or whether it is a seasona I phenomenon

the older leaves were originally produced

at a different time of year to the maximally

expanded leaves and may not have grown

to the same final size.

5. GROWTH STUDIES

In the results presented here leaf position

.
(number) has been taken as a measure of ~ 100

leaf age. To be able to make estimates

j
of productivities and nutrient budgets, rates
50
of leaf production, leaf growth and fruit

production as well as ages of leaves will

have to be known. Pi lot studies of tagged N 0 M A M A o N 0

Time of year
plants have been undertaken. Fig. 13 shows
FIGURE 14. Lifespan of individual leaves of
the rate of leaf production throughout the
S. thunbergii in relation to date of appearance
year. Maximum leaf production occurred of the leaf. Mean of five plants.

. From

I iminary
rates of

estimate
leaf production

of primary
a very

productivity
pre-

can be made. An average of 41 leaves,

-, ., and hence stem containing 41 nodes, are

~
+
produced per stem per annum. Assuming
-2
15

H-
an average density of stems m on a
.,
+ + well vegetated dune yields a production figure

+
- of vegetative parts of 0.8 to 1.0 kg m- 2yr- 1

depending upon the extent of fruit production

.,
(equivalent to 15 to 18.8 MJm- 2 yr- 1 assuming
-1
a conversion of 18.8 k J g dry matter (Odum,

1 1971)). This compares with values of 0.86

JFMAMJJASONOJfMAM
Time 01 year
+ 0.21 kg m- 2 yr -1 for tropical savannas

and 1.14::: 0.35 kg m -2 yr- 1 for tropical grass-

FIGURE 13. Rates of leaf production by
S. thunbergii in the field throughout the I ands (Murphy, 1975). Below-ground product-
year. Mean of five plants.
684

ivity is likely to be low as most of the below- growth form and response to accreting sand
ground biomass is stems that were produced on the other. Studies are underway to eluci-
above ground and have subsequently become date this latter point.
buried as sand accumulated around the plant.
ACKNOWLEDGEMENTS
6. CONCLUSION
The author wishes to thank the Council for
s. thunbergi i is a tropical coastal dune
Scientific and Industrial Research for financial
pioneer that grows well in accreting sand.
assistance, Professor C.F. Cresswell of the
I t shows moderate to high transpiration rates
Department of Botany, University of the Wit-
but seldom seems to suffer severe water stress.
watersrand, in whose laboratories the photo-
The water supply from rainfall generally
synthesis studies were performed during a
appears to be adequate to meet the atmos-
sabbatical leave, and Mr. I. F. Garland of
pheric demand. The photosynthetic character-
the farm Twinstreams, at Mtunzini, Natal,
istics are predominantly those of a C 3 plant,
on whose property much of the field work
but it appears to have a very high water
was done.
use efficiency. (However, if water use effic-
REFERENCES
iency is calculated in terms of water trans-
A lien SE Grimshaw M Parkinson JA and
pired per unit dry matter produced (rather
Quarmby C (1974) Chemical analysis of ecolog-
than g CO 2 fixed) the values obtained from ical materials, Blackwell Oxford.
Barbour MG (1978) Salt spray as a micro-
the estimates of annual transpiration and
environmental factor in the distribution of
productivity range between 650 and 820, beach plants at Point Reyes California, Oeco-
logia (Berl.) 32,213-224.
depending upon the extent of fruit production.
Black, CC (1973) Photosynthetic carbon
This is within the range of C 3 plants (Black, fixation in relation to net CO2 uptake, Ann.
1973) • The reason for the discrepancy between Rev. Plant Physiol. 24 253-286.
Chapman VJ (1976) Coastal vegetation,
the two methods of estimating water use Pergamon Press Oxford 2nd Ed.
efficiency is as yet unknown.) S. thunbergii Donnelly FA and Pammenter NW (1983) Vege-
tation zonation on a Natal coastal sand dune
shows a conservative mineral nutrition, re- system in relation to salt spray and soil sa-
linity, S. Afr. J. Bot. 2,45-51.
cycl ing those nutrients that are present in
Drennan PM and Pammenter NW (1982) Phy-
the soi I in low concentrations. It appears siology of salt secretion in the mangrove
to be very efficient with respect to N acqui- Avicennia marina (Forsk.) Vierh, New Phyto-
logist 91, 597-606.
sition, attaining moderate leaf N contents Ha II, AE Schu I ze E-D and Lange OL (1976)
from soi Is that are almost devoid of N. Current perspectives of steady-state stomatal
response to environment. I n Lange OL Kap-
Growth occurs throughout the year, the winter pen L and Schulze E-D eds. Water and plant
rate being about 60% of the summer rate I ife. Problems and modern approaches,
Springer Verlag, Berlin, pp. 169-188.
and prel iminary estimates of productivity Leopold AC and Kriedeman PE (1975) Plant
class it with tropical savannas or grasslands. growth and development, McGraw Hill, New
York, 2nd Ed.
Apart from the high water use efficiency Ludlow MM (1976) Ecophysiology of C4
and N acquisition ability S. thunbergii does grasses, I n Lange OL Kappen L and Schulze
E-D eds. Water and plant I ife. Problems
not appear to possess any particular physiol- and modern approaches, Springer Verlag,
Berlin, pp. 364-386.
ogical characteristics that adapt it to its
Murphy PG (1975) Net primary productivity
habitat. I ts success wou I d appear to lie in tropical terrestrial ecosystems. In Lieth
in physical characteristics such as resistance H and Whittaker RH eds. Primary productivity
of the biosphere, Springer Verlag, New York,
to high wind speeds, salt spray and sand pp. 217-231.
Odum EP (1971) Fundamentals of ecology.
abrasion on the one hand, and its particular
 685

WB Saunders, Phi ladelphia, 3rd Ed.

Ranwell D (1972) Ecology of salt marshes
and sand dunes, Chapman and Hall, London.
Sal isbury EJ (1952) Downs and dunes,
G Bell. Cited from Chapman (1976).
Stocker 0 (1976) The wa ter-photosyn thesi s
syndrome and the geographical plant distri-
bution in the Sahara deserts, In Lange OL
Kappen L and Schulze E-D eds. Water and
plant life. Problems and modern approaches,
Springer Verlag, Berlin, pp. 506-521.
Szarek SR and Ting IP (1975) Photosyn-
thetic efficiency of CAM plants in relation
to C 3 and C 4 plants, In Marcell E ed.
Environmental and biological control of photo-
synthesis, W Junk, the Hague, pp. 288-297.
Willis AJ and Jefferies RL (1963) Invest-
igations on the water relations of sand dune
plants under natural conditions. In Rutter
AJ and Whitehead FH eds. The water relations
of plants, British Ecological Society Symposium,
l'lo. 3, Blackwell, Oxford, pp. 168-189.

ENERGETIC VALUES IN INTERSTITIAL ISOPODS AND AMPHIPODS FROM SANDY BEACHES AS A FUNCTION OF BODY SIZE
AND SEASON (WESTERN MEDITERRANEAN) .
N.COINEAU (Maitre de Recherche, C.N.R.S., Laboratoire Arago, 66650 Banyuls-sur-Mer - France).
1. INTRODUCTION
Modern ecological literature includes a large
number of studies dealing with energy flow,
aiming at a better understanding of the func-
tioning of ecosystems, and of the interactions
between different ecosystems. In this field,
only the South African sandy beaches (Dye, 1981
Griffiths, 1981 ; Koop et al., 1982 ; MacLachlan
et al., 1981) and the Indian beaches (Wafar et al.
1980) have been taken into account. Some isolated
data exist in Europe (Lasserre, 1980 ; Lasserre
and Renaud-Mornant, 1971). As reproduction cycles
and population dynamics of interstitial isopods
and amphipods of Mediterranean sandy beaches are
well known, such studies can be carried out. In
order to establish the energetic budget of these
crustaceans, and within the framework of a study
of energy flow in these animals, calorimetic
mesures provide an element contributing to the
definition of the potential energy fraction that
they represent for conceivable predators, and of
their nutritive value. The present study deals
with three interstitial crustaceans : the isopods
AngeZiera phreatieoZa and Mieroeharon marinus,
and the amphipod BogidieZZa chappuisi ; all live
in the sandy beaches of the Roussillon coast
(France, Western Mediterranea).
2. MATERIAL AND METHODS
The crustaceans were caught on the sandy beach
at Argeles-sur-Mer (France) by the Karaman-
Chappuis' method at weekly or bi-weekly intervals,
according to weather conditions. Animals were
brought alive to the laboratory in the intersti-
687
tial water (brackish water), and were sorted,
briefly rinsed in distilled water, grouped into
genera and dried to constant weight at 70 D C in a
vacuum oven. Samples were taken at regular inter-
vals from February to October 1980, and from
January to September 1982.
Weight was determined to the nearest 0,01 mg on
a Cahn electrobalance. The monthly or seasonal
individual mean dry weight was obtained by weig-
hing a smal sub-ample comprising young, medium-
sized and adult individuals of each species. This
sub-sample is proportional to the total number of
every size class for one calorimetric experiment.
As a matter of fact, the dry weight of intersti-
tial crustaceans was very lowe juveniles : from
1.25 to 4.82 pg ; adults: 4.83 - 9.05 Vg). So
it was necessary to group batches of more than
400 isopods to get the required minimum weight
for a combustion (1 mg) ; several samples, and
occasionally even the samples of two or three
successive months, were therefore grouped (in
particular in Winter).
Caloric contents were determined by complete
combustion with a Phillipson microbomb calorime-
ter. The recorder was previously calibrated with
benzoic acid after 15-20 tests. Ash remaining in
the microbomb after firing was weighed to obtain
the percentage of mineral content.
3. RESULTS
3.1. Seasonal variations of dry-weight
3.1.1. AngeZiera phreatieoZa (Isopoda). The mean
values for individual dry-weight Cpg) varied from
4.2 to 6.8 ¥g, with the extremes 4.1. in December
688

and 9.0 ~g in July. Means increased from October

~9 cal/mg
to May, with a high rate of increase up to June
,,.,,
7
and then regularely decrease. Every increase coin-
cided with the beginning of ovogenesis (autumn),
"
, , ,,
,
the progressive maturation of ovocytes at the end
6
, ,,
~'
I .--'~ ... -~
, , '\

., .
of winter and, in spring, egg-laying and growth , / '.
,, \, \

.'
5 7
of the embryo during the embryonic development
(June-July) ; the female keeps the embryo inside
,
... --- ~ ..
\
'.'\
.-'1:f-
-_ ..
4
the marsupium for 40-45 days (Coineau, 1969, 1981) \.b 6

(figure 1 ,a).
3.1.2. Microcharon marinus (Isopoda). Means varied 3 5
between 3.9 and 5.9 rg,extremes were 2.3 and 6.0
pg. The seasonal fluctuations closely followed 2 4
those of Angeliera, with a less pronounced decrease
from July to September. The number of adults of
3
the medium-sized class and ovigerous females
remained high during this period (Coineau, 1969).
The increase of these means occurred also during
J FMAMJ J ASOND
ovegenesis, egg-laying and embryonic development
in the marsupial-chamber (60 days) (figure l,b). FIGURE 1. Seasonal variations of the mean indi-

3.1.3. Bogidiella (Amphipoda). Means of individual vidual dry weight (pg) in the isopods : a)

dry weight variation were opposite to that of the Angeliera phreaticola, b)Microcharon marinus.
isopods : regular decrease from Januray to Seasonal evolution of the caloric content (cal/

August-September. We note on the other hand a mg ash-free dry weight) c)A.phreaticola, d)

quick increase in October and November 1981, M.marinus. 1981 data.
corresponding to the presence of ovigerous females
of large size and females ready for egg-laying of the three genera in the samples ; the weight

with three enormous ovocytes. This increase was of the pellets varied between 0.92 and 3.84 mg

less marked in 1982. The low individual dry weight in 1981.

in August and September was due to the large 3.2.1. Isopoda.The results for 1981, expressed

number of juveniles' of stages I and II as compared in calories/mg, ranged between 4.80 and 7.02

to the number of the adults (92 % juveniles in cal/mg ash-free dry weight for Angeliera (mean

1982 ; 64 % in 1981) (figure 2,e).
The variations of individual weight were less pro-
nounced than those of isopods. On the one hand
this was because of the sex-ratio favouring
the males (males are lighter and smaller than
females), while on the otherhand it was because
5.45 in 1981 ; 5.55 in 1980), and between 2.73
and 4.60 cal/mg ash-free dry weight (mean 3,93
in 1981) for Microcharon. The percentage of ashes
after burning at 1 200°C varied between 3.6 and
18.3 % of dry weight (mean 10.3 %) for Angeliera,
and between 7.5 and 30.8 % of the dry weight

the size of ovocytes is relatively smaller (mean 19.3 %) for Microcharon. The values of ash

than in isopods. content, determined on separate pellets weighed

3.2. Seasonal change of caloric content
Batches of 280 to 800 individuals have been
grouped, according to the relative abundance
before and after incineration for three hours
at 600°C in a muffle furnace attained between
5.0 and 21.6 % for Angeliera (three tests) and
 689

between 6.4 and 28.2 % for Microcharon (two mea- f9 cal/m9

sures)(figure 1, c and d). 15

Comparable results were obtained in 1980.

Seasonal variations of the caloric content follo- 14
e-___ -- ..:....,. . . . ...........
""ee
... ...
'.'
wed those of the dry weight with an obvious paral- -'~'

lelism for the two isopods (figure 1 ) : slow ... ~'"

13 7
and regular increase from October to June, more
rapid from the end of May to the end of July,
followed by decrease till October, which is more
accentuated for Microcharon. The four curves reach
their maximum at mid-July, when the greatest num-
ber of ovigerous females were present in the two
genera (Coineau, 1969, 1981). The slower decrease
of September-October in Angeliera can be explained
by the fact that a certain percentage of ovigerous
females persisted till mid-October, a phenomenon
not occurring in Microcharon, in 1981. An inter- M A M J J A SON 0 J
esting value is the caloric content in a batch
FIGURE 2. e) Seasonal fluctuations of the mean
of ovigerous females of Angeliera of May-June and
individual dry weight in the amphipod Bogidiella
July 1980 : 6.88 cal/mg ash-free dry weight (ash
chappuisi. f) seasonal variations of the caloric
content 5.35 % /dry weight).
content in B.chappuisi. e,., 1981 data; •
Caloric and weight variations appear closely
1982 data.
related to the reproductive cycle.
The ash-percentage/dry weight, high in winter,
(Coineau, unpublished). Nevertheless, we should
increased in spring and summer, increased again in
note that the value of November 1981 seems abnor-
autumn, and therefore varied in a way opposite to
mally high (it was impossible to make another
the calorific variations. The same is true for
calorimetric test at the same period in 1982)
Microcharon, although less distinct.
that of December-January-February 1982 seems
3.2.2. Amphipoda. The number of amphipods in
more compatible with size histograms established
the samples was often too small for regular
for a study of populations dynamics in Bogidiella.
measures; data for 1981 and 1982 were therefore
The lowest point of the calorific curve agrees
added.
with the presence of high percentage of juveni-
The caloric content oscillated between 3.52 and
les (figure 2 f.).
6.92 cal/mg ash-free dry weight (mean: 4.77).
The ash content percentage of dry weight followed
The ash content ranged between 8.4 and 17.3 % of
the values of weight and calorific content, in
dry weight (mean: 13.45). Figure 2 allows
contrast to the pattern in isopods.
us to evaluate the parallel seasonal change of
the curves of weight and calorific values. There
4. DISCUSSION
was a decrease from April to September-October
Owing to the small size of interstitial crustaceans,
followed by an increase in autumn and the maxi-
the sampling method of Karaman-Chappuis was used
mum occurring in November-December in 1982 ; there
to obtain a larger number of individuals. This
is a peak in April-May 1981. The increase is rela-
number is still insufficient for calorimetric
ted to a great number of ovigerous females
tests for each or for single size classes.
690
Quantitative samples aiming at a productivity
evaluation will provide data about the energy
potential per square metre of sandy beach.
The small size of these crustaceans accounts for
the slowliness of all operations and tests.
Weight data reflect good trophic conditions of
the beach milieu, particularly in spring. They
also reflect the reproductive behaviour ; we can
see a slow increase in spring during the slow
maturation of the ovocytes ; the acceleration
takes place during the last phases of the ovo-
cytes growth with storage of nutrients, and then
during the egg laying within the marsupial chamber
of female. The calorific value fluctuations also
appear to be related to the reproductive season for
all three crustaceans, the maxima coinciding in
two cases. We should also notice here the adapta-
tion to the mesopsammic environment which is
demonstrated by the reduction of the number of
eggs, and by their large size due to the large
volume of yolk reserves. This reduction in number
is most pronounced in Angeliera, which lays only
one enormous egg per spawn ; it is in this form
that we find the highest calorific value (7.02)
in June-July, when the greatest number of repro-
ducting females can be found (ready for laying,
or with an embryo in the marsupium) (Coineau loco
cit.). The caloric data of ovigerous females of
Angeliera agree with this idea. Furthermore, the
high caloric content is due to storage of energy
sources in the form of lipids ; we can observe this
on histological sections where large lipid droplets
can be seen in spring and summer. The part played
by the males in the calorific increase is noticea-
ble because of their sexual maturation during the
same period. Indeed, Angeliera males copulate with
juveniles I and II (Coineau and Renaud-Mornant,
1977), which explains their part in the increase
in June-July and the quick decrease in August-
September by energy loss through gametes and res-
piratory cost. On the other hand, higher percenta-
ges of juveniles in August-September-October result
in decrease of the caloric content, which is oppo-
site to the general rule. Strong and Daborn (1978)
stated that juveniles invest lipids which are
converted to lower energetic proteins and carbo-
hydrates used in growth. This can explain this
decrease of the caloric content, beyond the disap-
pearance of ovigerous females and the death of
many adults of large size. The Strong and Daborn
hypothesis may also explain the importance of
ash percentage of dry weight during the same
season.
The ash content of interstitial crustaceans is
relatively low as compared to that in other
isopods (other isopods : 45-72 % dry weight)
(Shapunov, 1969, Cummins and Wuycheck, after
Strong and Daborn, 1978). It shows smaller varia-
tions ; the proportion of C0 3 Ca, not determined
here, is probably low, or even negligible (Percy,
1979), because interstitial crustaceans have a
very thin integument which is little calcified.
With a low percentage of ash, it is possible that
a better correlation exists between caloric content
variations and the reproductive cycles (Wiegert,
1965) .
In contrast to the low ash content, calorific
values of the three studied crustaceans appear
normal for Microcharon, but high for Angeliera
and Bogidiella when compared with other benthic
isopods ans amphipods. Generally, the mean value
is lower than 5 cal/mg ash-free dry weight for
benthic invertebrates (Griffiths 1977, Wissing
et al., 1973) although the calorific content is
higher for pelagic forms. Few isopods have been
studied recently. The values range from 2.7
(Idotea baltica, Strong and Daborn, 1979) to 5.3
cal/mg a.f.d.w. (Ligia dilata, Koop and Field,
1981).Among the amphipods, the mean caloric
content lies between 2.74 (Leptocheirus pinguis,
Tyller, 1973) and 6.91 (Pontoporeia affinis,
Green, 1971) ; the general mean values vary
around 3.5-4.4 cal/mg ash-free dry weight (Wissing
et al. 1973 , Boates and Smith, 1979, Brawn et
al., 1968, Baitchorovand Sementichenko, 1977,
Wissing and Hasler 1971, Perry 1979, Mathias,

1966, (unpublished data, in Cummins and Wuycheck,
1971), Hargrave, 1969, Le Bouteiller-Bougnoles,
1981). The higher values are found in Pontoporeia
affinis : 4.4 - 6.9 and Parathemisto : 5.8. With
a mean of 4.77, Bogidiella represents a relative-
ly high energy content. Little work has been
published about interstiti,al invertebrates
(McLachlan et al., 1981).
Working on the Slobodkin and Richman (1961) hypo-
thesis, Griffiths (1977) showed that benthic
crustaceans of low value live in a stable environ-
ment, and that crustaceans of high calorific con-
tent live in an unstable environment. Littoral
interstitial crustaceans live in parts of sandy
beach near estuaries and are under the influence
of wave action : they are therefor placed in the
second category. High energy content and strong
variations can be related to storage of lipids.
These high energetic values suggest that the
part of interstitial crustaceans in the energy
flow is far from being negligible. We also know
that as element of interstitial fauna, they are
considered the decomposers of the surf zone-beach
ecosystem (McLachlan et al., loco cit.).
Acknowledgments : The author wish to thank Mrs.
S.Razouls, M.J. Natero and N. Batailler for advice
on the operation of the microbomb calorimeter,
and F. Duran who collected the material.
REFERENCES
Baitchrov VM, Sementichenko VP (1977) Size
weight characterictics and caloric content of
Pontoporeia affinis Lindstr (Crustacea, amphipoda)
from different habitats, Gidrobiol. Zh. 13 (2),
70-73.
Boates JS, Smith PC (1979) Length-weight
relationships, energy content and the effect of
predation on Corophiwn volutator (l?allas) (Crusta-
cea, amphipoda), Proc. N.S. Inst. Sci. 29, 489-
499.
Brawn VM et al. (1968) Calorific content of
the standing crop of benthic and epibenthic
invertebrates of St. Margarets Bay, Nova. Scotia,
J. Fish. Res. Bd Can. 25, 1803-1811.
Coineau N (1969) Les Isopodes interstitiels.
Documents sur leur ecologie et leur biologie,
These Doct. Etat, Paris, 229 p.
Coineau N (1981) Aspects de la. d:ynamique des
691
populations chez un Isopode interstitiel, Bijdr.
Dierk. 51 (1), 20-30.
Coineau N, Renaud-Mornant 3 (1977) Etude
anatomique des Isopodes interstitiels. II. Mor-
phologie fonctionnelle et evolution saisonniere
de l'appareil genital femelle d'Angeliera phrea-
ticola Chappuis et Delamare Deboutteville 1952,
Archs Zool. expo gen. 118 (3), 349-366.
Cummins KW, Wuycheck JC (1971) Caloric aqui-
valents for investigations in ecological energe-
tics, Mitt. into Werein. Limnol. 18 (IS), 1-158.
Dye AH (1979) The measurement of biological
oxygen demand in sandy beaches, S. Afr. Jl Zool.
14, 55-60.
Dye AH (1981) A study of benthic oxygen con-
sumption on exposed sandy beaches, Estuar. Coast.
Shelf Sci. 13 (6), 671-680.
Dye AH et al. (1981) The ecology of sandy
beaches in Natal, S. Afr. Jl Zool. 16 (4), 200-
209.
Green RH (I971) Lipid and caloric contents
of the relict Amphipod Pontoporeia affinis in
Cayuga Lake, New-York, J. Fish. Bd Can. 28,
776-777 .
Griffiths D (1977) Calorific variation in
Crustacea and other animals, J. Anim. Ecol.
46, 593-605.
Griffiths CL (1981) Grazers and filter feeders
in the Nearshore Zone of the southern Benguela
System, Trans. Roy. Soc. South. Afr. 44 (3),
335-340.
Hargrave (1970) The effect of a deposit feeder
amphipod on the metabolism of benthic microflora,
Limnol. Oceanog. IS (1), 21-30.
Johnson WS (1976) Populations energetics of
the intertidal Isopod Cirolana hardfordi, Mar.
BioI. 36, 351-357.
Koop K, Field JG (1981) Energy transformation
by the supralittoral isopod Ligia dilata Brandt,
J. expo Mar. BioI. Ecol. 53, 221-233.
Koop K et al. (1982) Biodegradation and carbon
flow based on kelp (Ecklonia maxima) debris in
a sandy beach microsom, Mar. Ecol. Prog. ser.
7 (3), 315-326.
Lasserre P (1980) Energetic role of meiofauna
and epifaunal deposit-feeders in increasing level
of microbial activity in estuarine ecosystems,
at the water-sediment interface. Colloques intern.
C.N.R.S. nO 239 : "Biogeochimie de la matiere
organique a l'interface eau-sediment marin"
309-318.
Lasserre P, Renaud-Mornant J (1971) Consomma-
tion d'Oxygene chez un Crustace meiobenthique
interstitiel de la sous-classe des Mystacocarides,
C.R. hebd. seanc. Acad. Sci., Paris 272, 1011-
1014.
Le Bouteiller-Bougnol C (1981) Biologie et
reproduction d'une population d'Amphipodes de
Banyuls-sur-Mer : Siphonocetes dellavallei.
These Dr 3e Cycle, Univ. P. et M. Curie, Oceanog.
Bio 1., 1 IS p.
Mathias J (197I) Energy flow and secondary
production of the Amphipods Hyalella azteca
692

and CPangonyx richmondi in Marion Lakes, British

Columbia, J. Fish. Res. Bd Can. 28, 711-726
McLachlan A et al. (1981) Sand Beach Energe-
tics : an ecosystem of high Energy Interface,
Estuar. Coast. Shelf Sci. 13 (1), 11-25.
McLachlan A et al. (1981) The ecology of sandy
beaches in Southern Africa, S. Afr. Jl Zool.
I 6 ( 4), 21 9-23 I.
Newell RC (198[0) The maintenance of energy
balance in marine invertebrates exposed to changes
in environment temperature. An. Environm. Fitness,
pp.561-582, R.Gilles Ed.,. Pergamon Press, Oxford.
Paine RT (1971) The measurement and application
of the calorie to ecological problems, Ann. Rev.
Ecol. Syst. 2, 145-164.
Percy JA (1979) Seasonal changes in organic
composition and caloric content of an arctic marine
Amphipod Onisimus (= Boeckosimus) affinis H.J.
Hansen, J. expo mar. BioI. Ecol. 40 (2), 183-192
Shapunov VM (1969) Determination of the calori-
fic value of dry organic matter mixed with benzoic
acid, Hydrobiol. J. 5, 92-97.
Schlindler DW et al. (1971) Seasonal calorific
values of freshwater zooplankton as determined
with Phillipson bomb calorimeter modified for
small samples, J. Fish. Res. Bd Can. 28 (4), 559-
564.
Shafir A, Field JG (1980) The importance of
small carnivorous isopod in energy transfer, Mar.
Ecol., Prog. Ser. 3, 203-215.
Slobodkin LB, Richman S (1961) Calories/gm
in species of animals, Nature, Lond. 194 (4785),
299.
Strong KW, Daborn GR (1978) Seasonal variation
in ash and calorific content of Idotea baZtica
(Pallas)(Crustacea : Isopoda), Can. J. Zool. 56,
1917-1921.
Strong KW, Daborn GR (1979) Growth and energy
utilisation of the intertidal Idotea baZtica
(Pallas)(Crustacea : Isopoda), J. expo mar. BioI.
Ecol. 41, 101-123.
Wafar MVM et al. (1980) Physical-chemical
features and energy input in three sandy beaches
of West-Coast of India, Ind. Jl Mar. Sci. 9,
25-30.
Tyler AV (1973) Caloric values of some North
Atlantic invertebrates, Mar. BioI. 19, 258-261.
Wiegert RG (1965) Intraspecific variation
in cal/g of meadow spittle Bug (Philaneus spuma-
rius), BioScience 15, 543-545.
Wissing TE, Hasler AD (1968) Calorific values
of some Invertebrates in Lake Mendota Wisconsin,
J. Fish. Res. Bd Can. 25 (1), 2515-2518.
Wissing TE, Hasler AD (1971) Intraspecific
change in caloric content of some freshwater
invertebrates, Ecology 52 (2), 371-373.

A SYNOPSIS OF COMMUNITY RESPIRATION STUDIES ON EXPOSED SANDY BEACHES
A.H. DYE (Department of Zoology, University of Transkei, Umtata)
lNTRODUCTION
The study of community metabolism has a long
history dating back to the late nineteenth
century When Hoppe-Seyler in Hutchinson (1941)
suggested that oxygen consumption in deep lake
water was a measure of biological activity.
The early literature has been reviewed by
Pamatmat (1977); of the 80 or so references
reviewed, only one, (Pamatmat 1968), concerns
benthic community metabolism of intertidal sands,
in this case a very sheltered area. The first
in situ measurements of biological oxygen
demand on exposed beaches, (Dye 1979), were
used as a comparison for a laboratory method
Which was being developed. One of the major
reasons for the lack of studies of this type,
is the exposed nature of the environment.
Equipment left in the intertidal zone receives
considerable pounding from waves and may be
damaged or lost entirely. Considerations such
as these are reflected in the fact that all the
studies on sandy beach community metabolism have
been conducted in the laboratory where
conditions are more easily controlled.
Laboratory studies have made use of two basic
approaches; either bringing intact sand cores
to the laboratory and measuring their oxygen
consumption (Dye 1979, 1980, 1981), or
constructing some semi-permanent system
designed to simulate the natural environment
(Johnson 1970; McIntyre et al 1970; Pugh 1975,
1976; Boucher and Chamroux 1976; Chamroux et
al 1977; Munro et al 1978; Wormald and Stirling
693
1979; McLachlan et al 1981).
METHODOLOGY
Most in situ studies of benthic metabolism have
been in sheltered sediments and have involved
the use of bell-jar techniques in Which a known
area of substratum is enclosed by a container of
known volume. Decreases in oxygen in the
container are then related to the metabolism of
the system. There are practical problems in
applying the bell-jar technique to sandy beaches.
Apart from the potential wave damage mentioned
earlier, the high drainage rates of water in
sandy beaches poses special problems. Methods
relying on the measurement of oxygen changes in
water overlying cores or areas of substratum
restrict the flow of oxygen to that supplied by
diffusion only. This is clearly inadequate in
a well-drained system Where oxygen penetrates
deeply due to flushing rather than diffusion
from the surface (McLachlan et al 1979). Bell-
jar methods used under these conditions will
thus under estimate the true metabolic activity.
In a study of a South African sandy beach Dye
(1.979) compared in situ measurements with those
made on a series of cores taken from the surface
to the permanent water table. Oxygen con-
sumption was measured on interstitial water
drawn from the core after a suitable incubation
period. The field measurements gave an oxygen
consumption estimate only one-sixth of that
calculated from the integrated core profiles.
694
Another problem associated with the use of
surface enclosures is that of defining the
volume of substratum under study. Since the
bell-jar is "bottomless" it is not possible to
enclose a known volume of sediment nor to pre-
vent leakage from surrounding areas. In
addition, the effects of tides, currents and
desiccation will be damped. It seems that the
most viable method of obtaining estimates of
oxygen consumption from a number of points in a
system is that of measuring oxygen changes in
the interstitial water of intact cores over short
incubation periods.
For longer-term studies, which involve not only
oxygen consumption measurements but also measure-
ments of mineralization rates as well as
monitoring changes in populations, the con-
struction of systems such as sand columns offers
the best solution. A considerable degree of
back-up is required if such systems are to
simulate adequately the natural environment.
One of the most important considerations is the
flow of water through the system. This should
be comparable to flow rates found under natural
conditions and should be of a "tidal" nature.
McLachlan et al (1981) found that an increased
flow rate in their sand columns resulted in
abnormally high rates of oxygen consumption and
they derived a direct linear relationship
between flow rate and oxygen consumption. An
alternative used by Pugh (1975, 1976), in which
a wedge of sand is exposed to continuous
inundation and wave action, appears to have come
closest to simulating natural conditions
although necessitating considerable technical
back-up.
RESULTS FROM INTACT CORES
Since it is possible to exclude the macrofaunal
organisms from sediment cores, either when they
are taken or by disregarding results from those
found to contain large forms, this method
measures what is more correctly termed inter-
stitial metabolism. Measurements of the oxygen
consumption of any macrofaunal organisms in the
sand should also be made to obtain an estimate
of metabolism for the system as a whole. As
mentioned the only published work in which use
is made of intact cores is a series of papers by
Dye (1979, 1980, 1981). In these studies,
which covered an 18-month period, the beach was
treated as a three-dimensional system with the
depth limit being the permanent water table.
Comparisons were made between a sheltered beach
and an exposed beach. Intact sediment cores
were taken horizontally from the surface to the
water table at intervals of 20cm at the high-,
mid- and low-tide levels. Parallel studies
were made of the oxygen consumption of macro-
faunal organisms present. Estimates of the
contribution of protozoa and bacteria were made
on the basis of direct counts. Table 1
summarizes the information and compares it with
that obtained by Smith (1973) and Smith et al
(1972), these being the most comparable,
although they were sublittoral. Although these
results are comparable it should be remembered
that in view of the comments made earlier con-
cerning bell-jar measurements, the sublittoral
estimates may be somewhat low. Considerable
variation between areas also results from the
relative size of the various populations in the
community. A large macrofaunal biomass can
increase community metabolism considerably and
variations in sediment particle size and organic
content can affect microorganism populations and
so also community respiration. The effects of
periodic inundation and desiccation on the
respiration of beach sediments has received
little attention. Tidal effects have been
studied on cores from South African beaches by
Dye (1980) who found a positive correlation
between oxygen consumption and water saturation
above 30%. The greatest effects were found
near the surface of the sand at the upper tidal
 695

levels (Figure 1). Maximum oxygen consumption consumption in relation to area, giving the
coincided with periods of inundation and this is depth of the sediment and the temperature regime
important When hourly rates are extrapolated to as additional information.
longer periods.

TABLE 1
Comparison of community respiration studies from two South African beaches (core measurements) with
those from similar (sublittoral) sediments in Bermuda and Georgia (bell-jars).

Total 02 Macrofaunal Bacterial Meiofaunal

Area Temperature uptake uptake uptake uptake Reference
m.e/rrf' /h m.e/m2 /h m.e/m2 /h m.e/m2 /h

Sheltered beach 20°C 17,6 0,7 10,6 6,3 Dye (1981)

Exposed beach 20°C 13,5 8,0 3,0 2,5 Dye (1981)
Sublittorala 23°C 25,8 0,6 7,8 17,4 Smith (1973)
b 15-20oC 81,4+ 6,2 26,1 Smith et al (1972)
11 49,1

+4% silt/clay; a - Bermuda; b - Georgia.

RESULTS FR<M EXPERIMENTAL SYSTEMS
Studies with experimental sand systems are of
fairly recent origin. They have been used for
a number of purposes, not all of Which were
aimed at measuring community respiration or
other indices of metabolism. Johnson (1970)
used a sand column to study the effects of oil
on beaches. The major effect seems to be
mechanical rather than chemical, particularly if
aged oil is washed ashore. The interstices
become clogged and this reduces water and oxygen
supply.
Community metabolism has been studied by
monitoring oxygen consumption and rates of
mineralization of organic matter. The info v-
mati on available is scattered and, unfortunately,
there has been little consistency in data
presentation, as illustrated in Table 2.
As with the core and field studies considerable
variation in oxygen consumption data from ex-
perimental sand systems appears to be the norm.
To What degree this is an artifact of the
methods used or a real variation is difficult
to determine. Much of the data should be
treated with caution because of the large
variation in flow rate of water used, usually
with little indication of the real environmental
conditions. Flow rates range from 25_60.em- 2d- 1
to simulate a IIshelteredll beach (Wormald,
Stirling, 1979), 100.e m- 2 d- 1 for anllexposed ll
Scottish beach (McIntyre et al 1970) to
550.e m-2 d-1 in a 11 large " sand column and
-2 -1
330-1000.e m d in a "small" sand column to
simulate an exposed Indian beach (Munro et al
-2 -1
1978). A flow rate of 300.e m d was used to
simulate an exposed South African beach by
McLachlan et al (1981).

Comparisons between studies are severely limited Although there is more consistency in the e~

by the lack of standardization in the units used pression of mineralization rates, there is still
as indices of community metabolism. I t is considerable variation, stemming largely from
desirable, for example, to express oxygen the methods used. McLachlan et al (1981)
696

Inundation

o~~----~----~----r----'-----'----~----~------~
IT HT L
LT HT
FIGURE 1 Fluctuations in interstitial oxygen damand as a function of tidal inundation
on an exposed sandy beach. S - surface; 20cm; 40cm; 60cm are depths below the sand
surface.
TABLE 2
Summary of data on community metabolism measured in experimental systems.

Index of community
System Temperature metabolism Reference

sand wedge 5°C 0,44 mg 021.e/h Pugh (1976)

II II II 80% mineralization II

sand column 20°C 100% II Boucher and Chamroux (1976)

II II 28°C? 0,5ug O/g/h Munro et al (1978)
II II II 70% mineralization II

sand column 18-25°C 43-100% N - II Wonnald and Stirling (1979)

II II II 60-99% p - II II

sand column 18-20 oC 39,2 ml 0/m2/h McLachlan et al 1981

" " " 18-45% N- mineralization

"

measured the rate of mineralization of naturally have much relevance to the natural situation.

occurring organic matter as well as that of McLachlan et al (1981) calculated that in ex-

added amino acids. The addition of amino acids posed, well-drained beach sands one-third of the
stimulated microbial activity and increased the organic nitrogen should be mineralized in the

removal of organic -N; NH4 -N; N02 -N and upper 50cm of the sediment. The finding by
increased the generation of N03 -N. Since Boucher and Chamroux (1976) of 100% mineraliza-
amino acids are more readily utilized than tion in a 16cm column is almost certainly an
naturally occurring organic matter, it is doubt- overestimate resulting from their use of amino
ful whether results based on amino acid addition acids as organic substrates.

It is unfortunate that data on the two most im-
portant parameters used to estimate community
metabolism, namely oxygen consumption and
mineralization rates, suffer from large methodo-
logical variations in addition to any natural
variation that may be present. This greatly
reduces the usefulness of the data collected so
far. Although these studies have been of great
importance in the development of simulation
'systems, a system has yet to be developed which
truly simulates beach conditions, particularly
with respect to water-flow,characteristics.
Data are available on water flow in inter-
tidal sands (Steele et al 1970; Munro et al
1978; Riedl and Machan 1972; McLachlan 1979)
so it should be possible to construct a system
with the necessary characteristics.
To summarize it is clear that the experimental
approaches used to study metabolism in sub-
littoral and sheltered areas cannot be applied
directly to sandy beaches. Two viable
alternatives exist. One is the use of intact
cores enabling integrated profiles of metabolism
to be produced at a relatively large nuniber of
sites. This is particularly useful for large-
scale comparative studies and monitoring
purposes, although parallel studies on macro-
faunal metabolism would have to be carried out.
The other alternative is the experimental sand
system. Many of these systems have not
adequately simulated the natural conditions,
particularly with respect to water flow
characteristics and wave action. Until these
aspects have been improved extrapolations from
laboratory to field should be made with caution.
Finally, standardization of methods and
presentation of results is a prerequisite if
meaningful progress is to be made in the study
of community metabolism on sandy beaches.
697
REFERENCES
Boucher G and Chamroux S (1976) Bacteria and
neiofauna in an experimental sand ecosystem I.
MateriaLs and preliminary results. J. expo mar.
BioI. Ecol. 24, 229-237.
Chamroux S, Boucher G and Bodin P (1977) Etude
experimental d 'un ecosysteme sableux. II.
Evolution des populations de bacteries et de
meifaune. Helgoland. wiss. Meeresunters. 30,
163-177
Dye AH (1979) Measurement of biological oxygen
demand in sandy beaches. S. Afr. J. Zool. 14,
55-60. '
Dye AH (1980) Tidal fluctuations in biological
oxygen demand in exposed sandy beaches. Estuar.
cstl. Mar. Sci~ 11, 1-8.
Dye AH (1981) A study of benthic oxygen con-
sumption on exposed sandy beaches. Estuar.
cstl. Shelf Sci. 13, 671-680.
Hutchinson GE (1941) Limnological studies in
Connecticut: IV. Mechanism of intermediate
metabolism in stratified lakes. Ecol. Monogr.
11, 21-60.
Johnson R (1970) The decomposition of crude
oil residues in sand columns. J. Mar. BioI.
Ass. UK 50, 925-937.
McIntyre AD, Munro ALS and Steele JH (1970)
Energy flow,in a sand ecosystem In: Marine
Food Chains, J.H. Steele, Editor, Oliver and
Boyd, Edinburgh, pp 19-31.
McLachlan A (1979) Volumes of sea water
filtered through eastern Cape sandy beaches.
S. Afr. J. Sci. 75, 75-79.
McLachlan A, Dye AH and van der Ryst P. (1979)
Vertical gradients in the fauna and oxidation of
two exposed sandy beaches. S. Afr. J. Zool. 14,
43-47.
McLachlan A, Dye AH and Harty B. (1981)
Simulation of the interstitial system of exposed
sandy beaches. Estuar. cstl. Shelf Sci. 12,
267-278.
Munro ALS, Wells JBJ and McIntyre AD (1978)
Energy flow in the flora and meiofauna of sandy
beaches. Proc. Roy. Soc. Edin. 76, 297-315.
Pamatmat MM (1968) Ecology and metabolism of a
benthic community on an intertidal sand flat.
Int. Revue ges. Hydrobiol. 53, 211-298.
Pamatmat MM (1977) Benthic comnnmity metabolism:
A review and assessment of present status and
outlook In: Ecology of marine benthos, B.C.
Coull, Editor, Belle W. Baruch Library in Marine
Science No.6, University of South Carolina
~ress, Columbia, South Carolina, pp 89-111.
Pugh KB (1975) A model beach system. J. expo
mar. BioI. Ecol. 18, 197-213.
Pugh KB (1976) An annual cycle, at constant
temperature, in a model sandy beach I. Nutrient
chemistry. J. expo mar. BioI. Ecol. 22,
179-192:.
Riedl RT and Machan R (1972) Hydrodynamic
patterns in lotic intertidal sands and their
bioclimatological implications. Mar. BioI.
13, 179-209.
698

Smith KL (1973) Respiration of a sublittoral

community. Ecology 54, 1965-1975.
Smith KL, Burns KA and Teal JM (1972) In situ
respiration of benthic communities in Castle
Harbour, Bermuda. Mar. BioI. 12, 196-199.
Steele JH, Munro ALS and Giese GS (1970)
Environmental factors controlling the epipsammic
flora on beach and sublittoral sands. J. mar.
bioI. Ass. UK. 50, 907-918.
Wormald AP and Stirling HP (1979) A preliminary
investigation of nutrient enrichment in ex-
perimental sand columns and its effect on
tropical intertidal bacteria and meiofauna.
Estuar. cstl. mar. Sci. 8, 441-453.

SANDY BEACH ECOPHYSIOLOGY - WORKSHOP REPORT
A.C. BRmJN and R. BALLY (Department of Zoology, University of Cape Town, Rondebosch 7700, South Africa)
The ecophysiology workshop held on 21.1.83 was
designed to concentrate primarily on gaps in our
knowledge and possible errors in our concepts,
although in the event a broad range of topics was
discussed.
There was general agreement that the most serious
omission in the work so far undertaken is the
lack of any solid appreciation of the ecophysio-
logy of animals inhabiting the surf zone and in
particular of the surf zone zooplankton,
including mysids and shrimps. Such knowledge is
critical in view of concepts now emerging of the
sandy-beach ecosystem as an entity which includes
not only the intertidal sands but also the water
and substratum of the immediate sublittoral.
Understanding is especially needed in the area of
surf zone dynamics and their effects on the
distribution, behaviour and physiology of the
surf zone biota. Some data on mysids and shrimps
were presented at the conference and there are a
few papers on the behaviour of surf zone fishes,
but these contributions, while valuable, serve to
highlight our ignorance of the ecophysiology of
this part of the system.
Some discussion ensued on the techniques that
might be adopted in the experimental study of
surf zone zooplankton. It was felt that keeping
such animals in the laboratory would be likely to
give unrealistic results unless great care was
taken to simulate conditions prevailing in the
surf zone. The best laboratory is undoubtedly
the field itself but while large animals such as
699
Vo~ax may be studied experimentally in the field,
the study of zooplanktonic species presents
difficulties which might well force the
experimenter into the laboratory. Dynamic tanks
and flow-through systems are in use in some
laboratories which reduce the problems of relat-
ing laboratory results to field conditions and
may also overcome difficulties associated with
feeding. Field studies should, however, be under-
taken wherever possible and the aim of
laboratory studies must always be to relate the
results back to field conditions. There was
considerable support for the suggestion (Brown,
this volume) that we urgently need multivariate
studies on the tolerances of organisms and that
these must include a time component. Such studies
are needed not only on the surf zone zooplankton
but on organisms filling all niches within the
sandy beach ecosystem.
Surf circulation currents are probably used for
movement by the zooplankton and there are likely
to be complex interactions requiring inter-
disciplinary studies. However, if we want to
model animals' responses to water movements,
frequency and amplitude models would be required
so that a great deal of sophisticated apparatus
would be involved. One would probably not
achieve the ideal system but have to compromise
and scale down the whole operation. There was
agreement that even a limited project along these
lines would be most valuable.
The ecophysiology of sandy beach plants, including
700

dune plants, intertidal single cells and the surf
phytoplankton, was identified as another serious
gap in our knowledge. There was considerable
support for suggestions that such work should be
carried out, although a motion that a committee
be set up to deal with the problem was defeated
on the grounds that the workshop was not
competent to pass such a resolution. Individual
scientists should, however, be encouraged to work
in this field.
Little is known of the ecophysiology of the micro-
and meiofauna and much of the work that has been
undertaken is of the "black box" variety, the
studies being of communities rather than of
individual species. Such studies may be both
valuable and time-saving but cannot replace
investigations of known species. Most speakers
agreed that the time had come to undertake true
ecophysiological studies on individual species of
micro- and meiofauna, although a mixture of both
approaches might pay dividends in the long run.
It was also pointed out that laboratory
experiments with meiofaunal species may well give
more realistic results than those conducted on
the macrofauna.
building up populations. A strong plea was made
that ecophysiological studies be attempted on
larval forms as these are frequently more
important to distribution than are the physio-
logical responses of the adults.
We also need to start considering evolutionary
principles in sandy beach biota and investigating
the effect of genetic variations between popula-
tions on their ecology. Evolutionary
considerations not only constitute a gap in the
literature but were not even mentioned by any
speaker at the conference. Sandy beach
ecophysiologists appear to be trapped by the
species concept despite the fact that numerous
distinct populations may be identified within
single species on the results of such techniques
as gel electrophoresis. Distinct populations
may also be identified on behavioural grounds,
although here only some of the differences may
be genetic as sandy-beach animals are among the
most adaptable forms, changing their behaviour
to suit prevailing conditions. Experiments on
gene flow between differently located and
behaving populations are indicated.

Sandy beach animals are ideal tools to test, and

perhaps develop further, the rand k selection
theories, yet little has been done in this
direction although it is known that in some
psammolittoral species, the number, size and
packaging of eggs vary according to circumstances.
Beaches differing in exposure would make ideal
workshops for the study of reproductive
strategies. There was a subsequent discussion on
reproductive rates and attention drawn to the
fact that the yolk content of VotiLia eggs
differs with the intertidal position assumed. As
far as distribution and recruitment are concerned,
we need to know whether there are mechanisms for
retaining the planktonic larvae of psammophiles
in the surf zone and whether this is a factor in
 703

MANAGEMENT OF SANDY COASTLINES - REPORT ON REVIEH AND HORKSHOP

A.E.F. HEYDORN (National Research Institute for Oceanology, Stellenbosch,

South Africa)

1. SUMMARY OF REVIEH PAPER BY J R CLARK IDENTIFIABLE IN 'i'HE LCOSYSTEMS TO BE

Prof Clark made it clear that he would MANAGED. It is therefore of great
deal with management and conservation in importance to yet managerial systems
the widest sense, including adminis- to recognize themselves as Deing
trative and legal considerations. On responsible for HHOLE LCOSYSTEMS and
this premise he emphasized that: to yuard against utilization or
manipulation of COMPONENTS of
Beach management can only be effective ecosystems (i.e. dunes, beaches or
when seen in the overall context of floodplains) if the ecosystem as a
Coastal Zone Management. Management whole can be harmed in the process.
procedures have to be attuned to
natural processes if they are to De For these reasons it is equally
effective. This means that they must important that scientists should
be seen in relation to coastal recognise that they are responsible
environments from watersheds in for the transmission of information
catchments to water in the sea below about the processes and functioning of
the breaker influence. ecosystems in understandable form to
environmental management agencies.
The key to successful management is They must therefore ensure that their
simplicity in spite of the complexity research, even if it is of fundamental
of governmental systems. There can be nature, must eventually be relevant to
10 to 20 authorities involved, the need for scientifically Dased
belonging to Cen tral, provincial, environmental management.
District or Local levels of
Government. This often leads to In spite of the need for a HOLISTIC
boundary disputes where does the ECOSYSTEM APPROACH as advocated above,
jurisdiction of each Authority end? practical exigencies dictate the need
for the utilization of ecosystem
In spite of this, all interactive components. To achieve this without
processes of an administrative and causing unnecessary damage, overall
ecological nature must be considered. environmental management plans need to
THUS ADMINISTRACTIVE PROCESSES SHOULD be drawn up in which the determination
BE STRUCTURED ACCORDING TO THE of boundaries is very important.
COMPONENTS AND PROCESSES WHICH ARE However, administrators and ecologists
704

frequently attach different values to that of the coastal zone (of which
boundaries. While a river might be a beaches form an important component),
useful boundary to an administrator, needs to L>e structured to be effective.
the ecologist might insist that In general processes of management must
watersheds should be used so that be based on:
effective management of catchments can - assessments of the resources involved;
be achieved. Ecologists should - the setting of goals and objectives;
therefore strive to separate coastal - the implementation of management
ecosystems into components which can decisions to meet these goals.
be used by administrators. This has
been successfully done in the U.S.A.
but the alignments of boundaries must applied specifically to beaches
be coordinated with due regard for the The beach and its characteristics:
relationship between natural and In any assessment of the resources
administrative processes. represented by or associated with a
beach, it must first be decided whether
Research should be global in concept. the beach should be regarded as:
Thus, although there is great
variation in coastal environments a limited ecosystem, or
certain principles are universal in a component of the ':Jreater coastal
concept and should be applicable in zone.
coastal environments around the world.
Thus it must be decided what the spatial
On the basis of the above, the major 1 imi ts are, e. g. the reg ion between the
components necessary for effective watershed and as deep as sediments are
Coastal Zone Management were outline as: capable of being moved onto beaches from
1. planning the seabed; OR the beach between two
2. Co-ordination headlands; OR the portion of beach
3. Control (permits) affected by artificial structures or an
4. Review (plans, progress reports, estuary or ri ver mouth; OR a transect
etc) of a beach in which physical, chemical
5. Envi ronmental assessment (prog rams, and biological interactions with
projects, etc) adjacent parts of the beach, with dunes
6. Consultation or with the sea can be determined.
7. Guideline criteria (buffers,
mitigation, etc) S i mila r 1 y , ..;;t...;:e",m",p~o::..::..ra=l_--,=l:.;:i:..;cmc.;l=-'t=.=,s must be
8. Designations (critical areas, decided upon, e.g. tidal, diurnal,
reserves, etc) seasonal, episodic, etc.
9. Technical assistance (e.g. to local
authorities) • Finally, the state of the beach must be
assessed" i.e. is it stable, is it
2 DISCUSSION FRAMEWORK PRESENTED BY eroding, is it ecologically viable, or
THE CHAIRMAN is it polluted? Obviously the state of
Any process of management, including the beach will also be affected by its
 705

present and predicted utilization. Altermatively, should a beach be managed

for a specific goal such as:
Assessment of the state of knowledge of
processes. - recreation (swimmin~, angling,
surfing, sunbathing, etc);
The scientific expertise available - ecomonic exploitation (mining of heavy
including its predictive capability. minerals, diamonds, sand reclamation,
harvesting of organisms, kelp
Knowledge of the legal and adminis- collection etc);
trative framework within which the beach - a component of the coastal zone (to
must be managed. protect developments in uune areas,
roads, railway lines, etc).
Financial resources available for
management. Finally, is it feasible to allow
modification of beaches, e.g. through
Resources available for implementation harbour development and the associated
of the management policy decided upon. construction of groynes, revetments,
etc.? If so, will accretion and
2.2 The setting of goals and objectives erosional problems result which might
of management (including an eval- require expensive maintenance or
uation of options). sediment replenishment procedures?
An assessment must be made of all
potential threats to the Leach in 2.3 'l'he implementation of management
question and this will determine the goals
urgency with which action must be taken. Having assessed the beach resources and
having set management goals and
Priorities for preservation must be objectives, the next step is to decide
determined, e.g. should its present on the requirements for implementation
status ~ be maintained in terms of: of the management goals. Relevant here
are:
- its intangible values such as
scientific, educational or - engineering ~uidelines;

recreational ones or - the legal, administrative and

should it be regarded as a core or a enforcement infrastructure which is
buffer area, e.g. to protect dune particularly important when the
environments (particularly where viability of beaches as natural
development has taken place on dunes), systems is to be maintained;
to contritube to the equilibrium of - the creation of facilities for
estuary mouths, in relation to harbour effective information transfer without
works, etc; or which the all-important public
- should the beach be maintained as a awareness of the need [or implemen-
specific habitat, ecosystem or tation of management procedures cannot
reference area. be promoted.
706

From the considerations of 2.1, 2.2 and readily used for management purposes in
2.3 above, four key questions arise: terms of:
- format;
Question 1: - understandability;
What administrative structure is quantification, and the creation of
considered ideal for beach management'? adequate public awareness,?
Inherent in this question is whether
beaches can really be managed in 3. GBNERAL DISCUSSION
isolation or only as components of While it had been hoped that the
coastal ecosystems'? ~,orkshop would provide answers to the
four key questions posed by the Chairman
Question 2: under "Discussion Framework", it soon
Do we know enough about our beaches to became clear that the~topic of MANAGBMBNT
be able to manage them effectively: is so wide that it requires a symposium
- for conservation, on its own.
- over predictable time horizons'?
A closely related and basic question ~uestion 1: What administrative
arises from this: do we really need deep- structure is considered ideal for beach
going biological knowledge for purposes management, and, can they be managed in
of beach management, e.g. for the deter- isolation'?
mination of the inshore angling
potential'? The real question should be: "HOW does
scientific input get through to
Question 3: administrators'?"

If manipulation in the coastal The need for cost-effective decisions

environment is necessary, do we have the and procedures in coastal zone
predictive ability on the basis of management is largely ignored although
current research to assist planners and much relevant information is available.
engineers with answers to questions such This points to the need for land-use
as: plans based on the recogni tion of
- what will artificial sand replenish- ecological processes. 'I'his also
ment requirements be'? emphasizes firstly the need for
what ecological effects .will temporary university education in this field and
disruption have both in the long and secondly the r61e which ecologically
the short term, e.g. of diamond mining orientated engineers have to play.
in the sub-tidal'?
will, for example, revetment construc- There is a dire need tor competent
tion, have permanent detrimental resource managers, but a glut of
ecological effects'? administrators and politicians exists.
Transmission of available information on
Question 4: coastal zone management in under-
Is the type of research discussed at the standable form to ALL authorities in
symposium producing results which can be coastal zone management is of vital
 707

importance. matter.

The problems of multiplicity of control In many cases there are conflicting

in the coastal zone were reemphasised. interests which require equally con-
flicting managerial techniques.
In
QU est ion 2: :::::D..::o=----.:.:w..::e=----.:.:k.!;n:.::o~w"__....:e~n.!;o=u...
g'_"h'___..:::a:..::b'_'o:..u=t_o=u.::..r these cases available knowledge is often
beaches to be able to manage them ignored or even twisted to suit the
effectively for conservation and over desired form af utilization of the
adequate time horizons? environment. Because of this very
typical situation, effective management
Because these two questions entail the is only possible through strict
need for knowledge of physical, chemical recognition of the requirements of
and biological interactions within natural processes and the need for their
beaches and with the surrounding maintenance in both ecological and
terrestrial, marine and estuarine economic terms.
environments, and because time horizons
encompass natural changes, episodic very often scientists are reluctant to
events and short-term human manip- give advice lJecause they feel that more
ulations, it was decided to combine research needs to be done. Because of
discussion of these factors with that of the rapid rate at which degradation is
Question 3, below. taking its toll of the environment and
its resources, they should rather be
Question 3: Is deep-going biological prepared to use available information
knowledge really necessary for purposes and on this basis advise NO'IJ to the lJest
of beach management? of their ability of WHAT NLEDS ~O BE
MANAGED AND HOW THIS SHOULD BE DONE.
Deep-going biological knowledge is
essential for all forms of beach There are two types of management:
management but not
available. always (a) deferred management where decisions
Where knowledge is available its may have to lJe deferred lJecause
usefulness depends on the type of stress more data needs to lJe gathered;
imposed on the environment by human (b) adaptive management where existing
action, e.g. the consequences of impacts knowledge is used but it is
due to interference with dunefields, or accepted that if mistakes are Idade,
the effects of breakwaters on beaches they will be corrected as more
are predictable, but the available experience is gained.
information is often ignored. On the
other hand, the effects of toxic waters, Question 4: Is research producing usalJle
cooling waters, or multiple discharges results for management purposes?
into the sea or out of broken effluent At this stage time ran aut for the
pipelines, are usually unpredictable and workshop. In the final stages of the
more imformation is obviously required. general discussion and also in
Whether such information will be used post-workshop discussions, various
when it does become available is another participants emphasized the need for an
708

understandable format for the scientific results are and will

presentation of research results and increasingly find practical appli-
warned against a tendency for many cation. It is a pity that this state of
scientists to criticize managerial affairs Vias not achieved 25 years ago
procedures. It is essential tha t the because much damage to the coast could
scientist does not adopt a lofty have been avoided. Hi th the increasing
attitude but rather that he convinces pressure Lrought about by growth in the
those who carry responsibility for the human population throughout the Vlorld,
implementation of coastal zone the need for a scientific basis to
management procedures that he has an coastal zone management will become more
essential contribution to make. In this and more vital in the years to come.
context the recognition of critical
processes is of particular importance,
e.g. the prediction of vulnerability of
certain areas which in the U.S.A. has
led to the protection of land behind
barrier dunes through the withholding of
Federal support for roads, sewage works
or disaster assistance AND of insurance
or tax relief in such areas. About 600
miles of the Atlantic coast of the
U.S.A. is involved and the protection of
critical areas here can be attributed
directly to the fact that scientists, on
the basis of the research results they
produced, could convince the Government
tha t i t Vlould cost more to develop such
areas than not to develop them.
Similarly the predictive ability of
scientists has led to the acceptance of
the concept of buffer zones in the
U.S.A. For example on an eroding
coastline it is necessary to know where
the coast will be 50 years from now and
no building is permitted in front of
this line.

4. CONCLUSION
It can be accepted therefore that
although a great deal needs to be done,
research is producing usable results in
many countries and subject to adequate
communication between scientists,
engineers, administrators and managers,

FACTORS INFLUENCING BEACH EROSION AND ACCRETION:
E.C.F. BIRD (Department of Geography, University of Melbourne)
Introduction
Sandy beaches are dynamic environments. Even
where they are neither gaining nor losing sedi-
ment they are subject to short-term changes in
response to wave regimes and weather conditions:
their transverse profiles show alternations as
beach-face sands are re-cycled and decline as
the component sand grains are reduced in calibre
by attrition and weathering. On some beaches
these trends are masked by accretion as new
sandy sediment arrives: if it is retained on the
beach, the coastline advances. On others there
is a loss of sediment, marked by diminishing
beach volumes and coastline retreat, with re-
incorporation of backshore beach and dune sands
on sectors that previously prograded.
In recent years, geomorphologists have given
much attention to changes that have taken place
on sandy coastlines around the world. In 1972
the International Geographical Union established
a Working Group to investigate such changes by
comparing historical maps, charts and air photo-
graphs with the present coastline configuration.
This initiated a project that has continued as
part of the programme of the I.G.U. Commission
on the Coastal Environment, which succeeded the
Working Group in 1976. As chairman of these
groups and convenor of the project on sandy
coastline changes the present author has been
responsible for co-ordinating information
supplied by 116 coastal scientists from 66
coastal countries (e.g. Koike 1977, McLean 1978,
Thorn 1968), and results have been presented in a
number of reports (e.g. Bird 1976, 1981).
709
A GLOBAL REVIEW
The major finding is that there has been an
almost world-wide predominance of erosion on
sandy coastlines, and that this has been the
case for periods ranging from a few years to
several centuries. Sectors of recent and con-
tinuing progradation of sandy coastlines are
restricted in extent. Discussions have indi-
cated that there is no single or simple explan-
ation for this predominance of erosion: it is
the outcome of a number of factors, the relative
importance of which varies from sector to sector
around the world's coastlines.
In the last few years the term coastline
(indicating the margin of the land at mean
highest tides) has been preferred to shoreline
(which can be taken as the edge of the sea at
any specified inter-tidal level). Changes have
certainly taken place on various shorelines thus
defined, but it is the advance or retreat of the
land margin at the limit of normal tidal sub-
mergence that attracts most attention, being a
measure of actual gains or losses of terrain.
The author's interest in this problem dates
from work on the Gippsland coastline in south-
eastern Australia, starting in 1957 (Fig. 1).
This coastline consists of a long sandy barrier,
enclosing a series of coastal lagoons (the Gipps-
land Lakes), and bordered on its seaward margin
by the Ninety Mile Beach. Although this barrier
consists of parallel dune ridges, with success-
ively deeper soil profiles and related sequences
in vegetation communities landward from the
Ninety Mile Beach - evidence of successive
addition of such ridges on a prograding sandy
710

OUTER BARRIER

m Prograded area

'-_--=5...
00_ _ _'.:.000 metres

~3:;y~"'---RECURVED SPIT .~ ERODING SANDY COASTLINE

~----~IO~~~2~0----~3~0~~4~0----~~Km.

FIGURE 1 - The Ninety Mile Beach in south-eastern Australia is a

retreating sandy coastline, except in the prograded
sector on either side of the artificial stone jetties
that protrude at Lakes Entrance.

shoreline during Holocene times - the seaward
fringe has been truncated by more recent erosion.
The beach is thus backed by crumbling cliffs of
sand where storm waves have attacked the dunes.
It was at first thought that this was a temporary
condition; that phases of erosion were necessary
to separate the successively-formed dune ridges,
and that in due course a new sand ridge would
begin to form along the length of the Ninety Mile
Beach. However, it was found from historical
evidence, notably measurements from maps and
charts made in the mid-nineteenth century, that
the Ninety Mile Beach had been a receding coast-
line for over a century, the only exception
being sectors bordering the stone jetties that
were built alongside an artificial entrance to
the Gippsland Lakes, opened in 1889. Here sand
accretion had taken place in front of a formerly
cliffed dune coastline, with development of young
parallel dune ridges on the prograded forelands.
Beyond the margins of this prograded sector, the
Ninety Mile Beach shows evidence of up to 150
metres of coastline recession between 1870 and
1970, and although there are brief phases of new
dune formation at the back of the beach during
fine weather, this trend of coastline retreat
has been maintained. Some of the eroded sand is
blown landward in backshore dunes; some has
drifted alongshore to be incorporated in the pro-
graded foreland at Lakes Entrance; and some has
been withdrawn to the adjacent sea floor (Bird
1980) .
Similar features were observed elsewhere in
south-eastern Australia, where beach ridge and
barrier formations produced by sand deposition
and coastline progradation in Holocene times are
now generally truncated along their seaward
margins. Continuing progradation, with addition
of further beach or dune ridges, is confined to
a few sectors such as Lakes Entrance. Elsewhere,
the seaward fringe has been cut back, and in
places dunes have been mobilised across barriers
that have thus become transgressive formations
711
behind a retreating coastline. Fig. 2 shows the
typical sequence. Subsequent research has ex-
tended this conclusion to many parts of the
world's sandy coastlines, including comparable
sectors in south-eastern South Africa (Heydorn
and Tinley 1980) and North and South America
(Tanner and Stapor 1972), and there has been much
discussion of the various possible causal factors.
A tendency to adopt a single, simple explanation
for beach erosion or accretion on the local scale
has given way to a realisation that there have
been multiple causes of such changes.
Beach erosion and accretion
Sandy sediment delivered to a coastline from
offshore or alongshore is retained on sectors
where waves and currents operate constructively,
a condition which depends partly on the form of
the waves and the strength and direction of near-
shore currents, and partly on the existing con-
figuration of the coast, in particular its trans-
verse profile extending seawards. Some coastlines,
notably those where the prevailing waves break
gradually, sending forth a strong swash across
shallow nearshore waters, may be considered
receptive to sand deposition, and are typically
prograding sectors. Progradation will not occur,
however, if the deposited sand is quickly moved
away alongshore by obliquely-arriving waves and
longshore currents, or if strong onshore winds
winnow it from the beach face and carry it land-
ward to the backshore zone and beyond, or if it
is washed over the beach crest and deposited in-
land.
On the other hand, there are many coastlines
where beaches are diminishing or absent, and yet
sand deposits are abundant on the adjacent sea
floor. Examples of this have been recognised on
the coasts of Victoria and northern Tasmania
(Bird 1980, 1982). Such coastlines are rejective
in the sense that the prevailing wave and current
processes have not been capable of delivering
sea floor sand to the present coastline. The
712

Continuing progradation
..

...
Holocene 8.dvance, then recess:on
--~)
-

FIGURE 2 - Sections showing the morphology and associated

vegetation on a prograding sandy coastline (top);
the onset of erosion (middle); and the development
of a transgressive barrier (bottom).

two situations are shown in Fig. 3.
It is interesting to speculate on the environ-
mental changes that could lead to a receptive
coastline becoming rejective (onset of erosion on
a previously prograded beach sector) and vice-
versa (onset of accretion on a sector previously
beach-less or subject to beach erosion). Such a
speculation can be informed by consideration of
the 10 factors that have been recognised in the
course of the I.G.V. Commission's project on
sandy coastline changes as influencing the erosion
or accretion of beaches.
The ten factors (each of which was exemplified
and illustrated in the oral presentation) are as
follows:
1. Variations in fluvial sediment yield.
Beaches may still be prograding at and near
the mouths of rivers delivering sandy sediment to
the coast, especially where fluvial sediment
yields have remained high, or have increased as
the outcome of natural processes (vulcanicity,
glacier melting, earthquakes, or occasional down-
pours in the river catchment) or of man's activ-
ities (accelerated soil erosion in the catchment
due to deforestation, overgrazing, cultivation,
or the effects of mining). Short-term prograd-
ation may be a consequence of unusual weather
conditions, such as torrential downpours result-
ing in massive fluvial sediment yields to the
coast. A reduction in fluvial sediment yield
(e.g. by dam construction and the subsequent en-
trapment of sediment in reservoirs) is usually
followed by reduced progradation or the onset of
erosion on beaches maintained by fluvial nourish-
ment. On the local scale, changes in the position
of river mouths can result in the onset of erosion
on previously nourished sectors and accretion on
sectors to which the outflow has been diverted.
2. Variations in sediment yield from cliffs and
rocky shores.
Some beaches are supplied with sandy sediment
eroded from arenaceous outcrops in adjacent cliffs
713
and rocky shores. If this supply is reduced or
cut off (e.g. by the building of sea walls to
halt cliff recession, or as the result of a
natural decline in cliff erosion) these beaches
are likely to be depleted by erosion. Beaches
receiving sediment from bordering coral reefs
may be augmented if coral growth is damaged or
destroyed (e.g. by pollution, or the effects of
seastar predator infestation), or eroded if the
rate of production of coralline sediment di-
minishes.
3. Variations in aeolian sand supply.
Prevailing winds are generally onshore in
coastal regions, but there are some sectors where
dunes blown into the sea by winds from the hinter-
land are nourishing prograding beaches. If the
dune sand supply runs out, or if the dunes become
stabilised (naturally or artificially), beaches
nourished in this way are likely to be depleted.
4. Variations in sand supply from the sea floor.
Substantial quantities of sand have been
carried shoreward from the sea floor to build
beaches, especially on oceanic coasts behind
shelf areas generating sediments from benthic
organisms (algae, foraminifera, bryozoa) or where
the coastal waters are rich in shelly fauna.
Quartzose sands deposited by rivers on the emerged
sea floor during Pleistocene low sea level phases
have also been swept shoreward during and after
subsequent sea transgressions. On some sectors
this still continues, with shoal sand moving on-
shore, but many coastlines formerly nourished in
this way now show signs of a deficit, there being
a decline in supply from the sea floor. This
seems to be part of the explanation for the onset
of erosion on the Ninety Nile Beach in south-
eastern Australia, where progradation has con-
tinued only in sectors adjacent to artificial
breakwaters (Fig. 1). Sand lost from the beach
has been partly removed to the sea floor or
carried away along the coast, and partly driven
onshore in dunes spilling landward (Fig. 2).
714

FIGURE 3 - Coastlines may be rejective of beach sand accretion,

even though there are sand supplies on the adjacent sea
floor (A). Others have been receptive, with progradation
(a,b,c,d) of beaches (B).

5. Losses of sand onshore, offshore and alongshore.
Sand winnowed from a beach to backshore dunes,
or washed inland by surges, or withdrawn to the
sea floor, or drifted away along the coastline is
lost from the beach system, and if it is not re-
placed by new sand deposits, erosion results.
Where sand moves alongshore, losses from one sector
may be compensated by progradation down-drift,
thereby reorientating the coastline. Many spits
and cuspate forelands show such a balance of loss
and gain as their evolution proceeds.
6. Interception of longshore drift.
Progradation of beaches occurs where sand
accumulates alongside a headland or coastal land-
slide, or in the zone of wave convergence behind
a nearshore island, offshore breakwater or wrecked
ship, or adjacent to a tidal inlet with strong
transverse ebb-and-flow currents, or bordering a
breakwater that projects from the coastline. In
each case the sector down-drift is deprived of
sand supply, and may be eroded. Accretion on the
up-drift side of breakwaters is commonly balanced
by erosion down-drift.
7. Losses and gains due to other human activities.
Apart from the effects of breakwaters, beaches
may be modified by such activities as sand ex-
traction, which is a common cause of erosion, as
is dredging in the nearshore zone, which deepens
the water and allows larger waves to attack the
shore. Some beaches have been augmented by the
deposition of quarry waste on the coast, and in
recent decades coastal engineers have re-nourished
eroding beaches artificially by dumping sand
brought in from the sea floor, alongshore from
an accreting sector, or from inland quarries.
Such beaches are a more acceptable and useful
means of coast protection than sea walls, boulders
or tetrapods.
8. Effects of weathering.
The volume of material on a sandy beach may be
715
reduced by attrition, when the component sand
grains are worn by repeated agitation and grind-
ing by wave action, or by solution, when soluble
constituents are leached out. As beaches become
finer in texture, sand is more readily winnowed
from them by wind, waves or currents; and as
their volume declines, the coastline begins to
retreat. On the other hand, cementation of beach
sands by carbonate precipitation, widespread on
tropical coasts, produces beach rock which may
impede subsequent erosion.
9. Changes of sea level relative to the land.
Coastal emergence, due to land uplift or a
fall in sea level, leads to an advance of the
coastline, and also stimulates shoreward drifting
of sea floor sand deposits. Submergence results
in coastline recession and the loss of beach sand
to the sea floor [the "Bruun Rule"] (Fig. 4). It
should be noted that these effects can be offset
by sediment transfer. Thus an emerging coast will
show coastline recession if there is accompanying
removal of coastal sediment by wave scour, and a
submerging coast can prograde if the effects of
submergence are offset by rapid accretion of
sediment (e.g. from fluvial sources or by long-
shore drifting). Within recent decades an over-
all rise in mean sea level has been recorded on
tide gauges in many parts of the world, and coast-
al submergence is likely to become extensive if
global warming and consequent reduction in
glaciers and ice sheets augments existing ocean
water volumes.
10. Climatic variations.
Advance or recession of sandy coastlines may
be correlated with climatic changes that increase
(e.g. of amount and intensity of rainfall and
consequent runoff) or decrease fluvial sediment
yields where the beaches are nourished by river
sands. As beaches are eroded in stormy weather
and build up in calmer weather a change to a
stormier climate will result in increased erosion
716

FIGURE 4 - According to the Bruun Rule a sandy coastline with an

equilibrium profile will retreat in response to a rise
of sea level relative to the land until the profile has
been restored by sea floor accretion.

Coastline receding
O·2m/yr.

Sea

FIGURE 5 - Analysis of sand losses from a retreating barrier beach on

Rhode Island, USA.

and the onset of calmer conditions can be marked
by beach progradation. Coastline advance or re-
treat related to such a change is likely to be
short-lived with adjustment to the stormier or
calmer conditions: continuing erosion requires
increasing storminess.
Conclusion
Each of these factors has contributed to beach
erosion and accretion. It is evident that no
single factor will account for the modern pre-
valence of erosion on sandy coastlines: it is not
simply the outcome of man's interference, or a
sea level rise, or an increase in storminess in
coastal waters, or a decline in sea floor sedi-
ment supply, but the result of combinations of
these factors, the dominant process having varied
from sector to sector. There is no evident
correlation with any known cyclic pattern of
coastal climate or tide regime: the world's sandy
coastlines are generally in a state of dis-
equilibrium.
It is thus necessary on each sector of sandy
coastline to consider all of the factors, and to
decide which have been dominant in producing
beach erosion and accretion. A good example of
quantitative analysis has been given by Fisher
(1980), who on the sandy Rhode Island coast docu-
mented an average retreat of 0.2 metres per year
between 1939 and 1975, and calculated that 15% of
this was due directly to coastal submergence, 26%
losses by overwash processes carrying beach sand
onshore, 35% by losses into tidal inlets, and 24%
by withdrawal to the sea floor (Fig. 5). Such
attempts to analyse the advance or retreat of
sandy coastlines in terms of volumetric budgeting
will characterise the next phase of research into
beach erosion and accretion.
References
Bird ECF 1976 Shoreline changes during the past
century, University of Melbourne.
Bird ECF 1980 Historical changes on sandy shore-
717
lines in Victoria, Proc. Roy. Soc. Victoria, 91:
pp. 17-32.
Bird ECF 1981 Recent changes on the world's
sandy shorelines, in E.C.F. Bird, K. Koike (eds.)
Coastal Dynamics and Scientific Sites, pp. 5-30.
Bird ECF, Paskoff R 1979 Relationships between
vertical changes of land and sea level and the
advance and retreat of coastlines, in K. Suguio
et al. (eds.) Coastal Evolution in the Quaternary,
pp. 29-40.
Bloom A 1977 Atlas of sea level curves. IGCP
Sea-level project.
Davies JL 1957 The importance of cut and fill
in the development of sand beach ridges, Austr.
J. Sci., 20: lOS-Ill.
Fisher JJ 1980 Shoreline erosion, Rhode Island
and North Carolina coasts, in M.L. Schwartz, J.J.
Fisher (eds.) Proceedings of the Per Bruun Sym-
posium, pp. 32-54.
Heydorn AEF, Tinley KL 1980 Synopsis of the
Cape Coast, Estuaries of the Cape, Part 1,
C.S.I.R., Research Report 380.
Koike K 1977 The recent change of sandy shore-
lines in Japan, Komazawa Geography, 13: pp. 1-16.
McLean RF 1978 Recent coastal progradation in
New Zealand, in J.L. Davies, M.A.J. Williams
(eds.) Landform Evolution in Australasia, pp.
168-196.
Tanner WF, Stapor F 1972 Accelerating crisis
in beach erosion, International Geography (I.G.C.
Montreal), 2: pp. 1020-1021.
Thom BG 1968 Coastal erosion in eastern Aust-
ralia, Austr. Geogr. Studies, 6: pp. 171-173.
 719

ACCIDENTAL FORMATION AND SUBSEQUENT DISAPPEARANCE OF A CONTAMINATED BEACH:

A Case History of Environmental Management

D.V. ELLIS (Biology Department, University of Victoria, Victoria, B.C., Canada V8W 2Y2)
and J.D. POPHAM (Seakem Oceanography Ltd., Sidney, B.C., Canada V8L 3S1)

INTRODUCTION 1981) . This paper provides additional

People produce wastes. Where they live in information about the shoreline recovery

industrialised urbanised societies they process.

produce large amounts of concentrated wastes The data reviewed here comes from two
with the potential to have many different initially separate investigations by the
impacts on the surrounding ecosystem. In spite authors, which were brought together in 1980.
of legislation such as the U.S. NEPA (National D.V. Ellis started ecosystem investigations in
Environmental Policy Act) and regulations in 1972 at the Jordan River Mine (see Fig. 1),
many countries which require treatment and which discharged the tailings from its copper
beneficial recycling of deleterious concentrate mill sub-tidally from 1972-74, and
components, in practice there remains a still to the beach during earlier operations (Ellis,
frequently used option to discharge treated or 1975). A series of 15 data and review reports
untreated wastes in liquid phase to the from the monitoring was filed with the B.C.
surrounding ecosystems. If the environmental Pollution Control Branch, Victoria.
impact of such discharges is to be minimised, Multi-element contaminant investigations were
considerable objective information is needed independently initiated in 1977 by a research
by decision-making regulatory agencies. The team, of which J.D. Popham was a member, at
information should include results from other the Anaconda Britannia Mine (see Fig. 1). The
similar cases, whether rated as successful or mine discharged its copper concentrate mill
not. This paper is a contribution to knowledge tailings to a small creek draining across a
about the discharge of liquid particle-bearing pocket beach to a silled fjord (Stump et al.,
wastes (slurries) at high energy shorelines. 1979; Popham, D'Auria, 1983).
The impacts of continuous discharges of
mine and other wastes at low energy fjords 2. JORDAN RIVER MINE
have been intensively investigated in recent The sub-tidal discharge of copper mine
years (Ellis, 1982; Freeland et al., 1980). tai lings was monitored from 1972-74, and
Considerable information is also available subsequent recovery from the discharge was
about both high and low energy shorelines monitored until 1980 (Ellis, 1980). A set of
impacted by and recovering from accidental effluent controls was placed on the discharge
one-shot discharges such as oil spills (see Table 1). The controls were
(Stevenson, 1978). Recovery from continuous site-specifically modified from provincial
discharges at low energy shorelines also objectives (Department of Lands, Forests and
occurs following implementation of pollution Water Resources, 1974). The discharge point
controls (Cross, Ellis, 1981; Ellis et al., was speCified at 12.5 m depth, implying 1.2 km
720

VANCOUVER CANADA
U.S.A.

ISLAND

FIGURE 1. Location map.

offshore. A monitoring programme was required, precluded accurate and precise quantitative
but could be implemented only two months prior population estimates, restricted the number of
to the start of discharge. Of the parameters species available for toxin testing and
monitored only a few were found to be locally constrained near-shore water and sediment
satisfactory and they were implemented too sampling. Tests which were finally carried
late for adequate baseline information. through the discharge and recovery period to
Difficulties arose due to extreme variability 1978 are listed in Table 1.
in the biological populations, combined with After start-up the tailings outfall
dangerous surf conditions and currents. These suffered two breaks (see Addendum) within four
 721

TABLE 1. Discharge constraints for Jordan River Mine (B.C. Permit 427-P) (ppm mg/L) •

Rate: 1,000,000 IGPD (= 0.05 m3 sec- 1 )

Suspended solids: 250,000 ppm
Total solids: 250,000 ppm
pH range: 8.5 - 10.0
Cyanide: 0.1 ppm (total)
Zinc: 0.1 ppm (dissolved)
Copper: 0.05 ppm (dissolved)
Sum of total Cu, Pb, Zn, Cr, Cd, Ag: 1.0 ppm (dissolved)
Outfall location: 12.5 m depth, 1.2 km offshore
Emergency spill pond: location specified near mill
Monitoring parameters (those applied consistently during and after discharge)
Effluent rate (during discharge only)
Effluent composition (during discharge only)
Beach sediment copper
Tissue copper (Protothaca staminea)
Macrobenthos: presence/absence
Other parameters during discharge
Sea water composition and turbidity
Sediment mechanical analysis, and organic content

months (plume dispersal patterns are shown in Discharge ceased in December 1974, as a
Fig. 2 and Ellis, 1982a, p. 245, Fig. 3) on result of a slow two-year regulatory due
the low gradient offshore slope, following process eventually requiring replacement of
which the discharge (only 100 m off MTL) built the original offshore terminus, combined with
a well sorted fine sand beach of tailings poor economic conditions. From 1974-78 copper
(median particle size 0.15-0.20 mm). This in both sediments and clams was reduced
contained high levels of copper (Ellis, 1980) (Ellis, 1980) closer to reference levels. A
-1
at 500-1300 Ilg.g relative to a reference statistically significant increase (P <
-1 -1
(control) site 20 km to the east ( 20 llg·g ). 0.01) from 16 to 75 Ilg.g by 1978 in sediment
The one beach inhabitant (the littleneck clam copper at the reference site suggested that
Protothaca staminea) which was continuously the reduction in copper levels in the
present in a nearby pocket of poorly sorted discharge area was due to physical
gravel and mud and could be monitored also displacement of the beach-forming tailings,
-1
showed high levels of copper (20-30 Ilg.g dry i.e. erosion and long-shore drift. By 1978 the
wt) almost double that of reference specimens physical displacement of the tailings beach
(Ellis, 1980). The actively mobile beach and and sub-t idal recovery was apparent (see Fig.
surf zone very close to the outfall break were 3) , and the topography had returned to the
bare of the diverse plant and animal configuration prior to discharge.
populations occurring elsewhere in the area In summary, the poorly sorted beach in
but organisms did inhabit light deposits of the area had been physically, chemically, and
tailings. The accidents precluded some of the b io log ica lly modified by the accidentally
planned monitoring. produced near-shore tailings discharge. A well
722

FIGURE 3. The recover ed beach at Jordan River

in 1978 four years after discharge ceased. The
ou tfall channel remains unpopulated but algal
beds have grown back over the ar ea that was
depopulated in 197 3-74.

3. COMPARATIVE INVESTIGATIONS AT BRITANNIA

BEACH
FIGURE 2. Surfacing and dispersal of mine Multi-element sediment and tissue a nalyses by
tailing slimes at Jordan River in 1972. The
X-ray en e rgy spectroscopy at pocket beaches
dredged and backfilled channel for the outfall
and algal beds are shown as a light lin e from near th e Britannia copper mine in 1977
the beach and dark patches re sp ectively (B.C.
demonstrated persistent high levels of cop p er
Govt . photo).
and some other elements (Stump et a1., 1979).
sorted beach formed, bu t in . the most dynamic In 1980 comparative investigations were

areas was too active to develop a normal plant undertaken in th e Jordan Ri ve r and Britannia
and animal community, although an array of areas to explore the extent and lev e l of
species could exist in areas not heavily tissue metal di ffe ren ces for comparison of
covered by tailings. The one sp ec ies which recovery rat es . Sampling strategy was to test
cou ld be rep ea tedly monitored demons t ra ted a one species eac h of a marine plant, a deposit
suscept ibi 1 ity to copper bioaccumulation feeder, a beach suspension fe e der and a rock
during the discharg e period, fo llow e d by face suspension fe ed er in order to compare
flushing of the contaminant over the next six different trophic niche metal levels at
years. progressive distances from th e tailings
source. In practice, the complete design could
 723

-1
TABLE 2. Comparative metal concentrations (llg.g ) from specimens and sediments collected 22/3/80
and 31/5/80 (whole specimens unless viscera or gills indicated).

Copper Zinc
Viscera Gill Viscera Gill

Mytilus edulis
Britannia Beach 645 219 193 607
Furry Creek 172 67 95 1156
Little Cates Park 9 6 82 385
Jordan River 21 9 108 179

Protothaca staminea
Furry Creek 119 12 100 90
Little Cates Park 36 7 97 76
Jordan River 9 67

Macoma baltica
Furry Creek 93 160 448 91
Little Cates Park 25 10 885 91

Fucus sp.
Britannia Beach 111 6
Furry Creek 30 96
Little Cates Park 4 44

Dlva sp.
Britannia Beach 345 218
Furry Creek 22 66
Little Cates Park 7 22

Laminaria sp.
Furry Creek 44 70
Little Cates Park 17 43

Sediments ( < 0.149 mm)

Britannia Beach 2700 829
Furry Creek 522 425
Little Cates Park 56 239

not be implemented due to lack of species at and known bioaccumulator M~ti lus edulis

accessible sampling sites, but some data on (Phillips, 1980) were higher at Britannia

each trophic niche was obtained. Analytical Beach than at Little Cates Park (reference

methods followed those reported previously site) and Jordan River, and intermediate at

(Stump et al., 1979; Popham, D'Auria, 1982). Furry Creek (6 km from Britannia Beach) . A

Insufficient specimens in some locations similar trend occurred for the beach

precluded replication for statistical suspension feeder Prothaca staminea and the
analysis. Table 2 documents results for those other species (Britannia series only) but at
two metals showing order of magnitude generally lower levels. The expected higher
differences between sites. levels in viscera than in gills showed in most
Copper levels in the suspension feeder cases.
724

The multi-element analytical system contaminated sites and species of different

demonstrated that simultaneously with copper trophic status, and the rates of recovery from
differences, zinc (which was not part of the contamination. In this case the Jordan River
mine's extraction objectives) was also site was virtually clear of contamination six
elevated in some spec ies close to the years after discharge ceased, whereas the
Britannia Beach source, but not in others, Britannia area remained substantially
e.g. Mytilus and Ulva but not Fucus and contaminated with a depauperate fauna at the
probably not Protothaca, Macoma and Laminaria. discharge site. The differences appear to lie
There was also a progressive decline of copper in several causes:
in sediments < 0.149 mm fraction) and a 1. The Jordan River area has a high
similar trend in zinc. energy shoreline compared to Britannia
Jordan River specimens of Mytilus and Beach.
Protothaca had lower levels of copper than at 2. The Jordan River discharge went
Britannia Beach, comparable to reference directly to the sea intermittently
station levels and as known elsewhere (Stump over twenty-four years, and was under
et a1., 1979; Popham, D'Auria, 1982). Similar effluent and outfall design controls
differences appeared for zinc. in the last two-year period of
In a number of species at both areas, operation.
there were unexpected tissue elevations shown 3. The Anaconda Britannia mine discharged
by the simultaneous multi-element analyses. over seventy-five years up to 0.2 m3
These included differences between the two sec- 1 of acid waste to a coastal
sites in nickel, iron, arsenic, selenium, lead stream from which tailings and
and the non-metal bromine. Species differences effluent entered the sea at the beach
in metal levels at Jordan River appeared face.
between Mytilus edulis and Protothaca staminea 4. Effluent composition differences arose
as was to be expected in some cases, e. g. from two ore-bodies and mi 11 ing
copper, zinc, manganese and iron, but did not processes (comparable tailings and ore
appear in arsenic, selenium, bromine, lead and analyses were not available to us).
strontium. At Jordan River Mytilus edulis
tissue levels declined over 800 m from the 4. DISCUSSION
outfall in manganese and iron (21.64 to 5.36 The Jordan River Mine monitoring programme was
-1
and 1914.87 to 786.8 ~g.g respectively). designed and initiated in 1972 at a time when
We recognise that lac.k of statistical little thought had been given by regulatory
analyses require that these differences must and industrial scientists (particularly those
be taken as trends requiring proper serving small, marginally economic industrial
investigation, even though unpublished plants) to the pattern of impacts which occur
estimates of sample variability support the at various kinds of point source discharges,
trends as real phenomena. The data are and the nature of monitoring tests on which
reported here to illustrate the value of the appraisal of environmental impact and quality
simultaneous multi-element analytical control of the discharge stream can be based.
procedures in comparative studies of complex These impact patterns are now more apparent.
waste discharges, the nature of They include smothering, suffocation,
bioaccumulation differences between contamination, poisoning, pathogen
 725

transmission and turbidity/discolouration occurs, as with sanitary wastes) be rectified.

(Ellis, 1982b). These impact patterns can be There is, for example, some evidence (Hay,
used to des ign a rational programme of 1982) that a particulate waste stream (a
site-specific environmental impact assessment slurry) such as mine ta il ings can flow
and monitoring provided adequate time (minimum underwater either as a density current
one year) is allowed prior to the design of settling to the seabed from the outfall
waste disposal procedures and start-up. terminus, and running dynamically between
Real-time bioassays with automated feedback to depositing and eroding levees, or
process controls are now under development alternatively by apron flow over a wide area.
(Cairns, Gruber, 1980) and are an obvious Appropriate outfall design can take advantage
objective in improving environmental quality of either form of dispersal according to the
control. Similarly multi-element environmental topography of the receiving area. It is by no
and biological (tissue) analysis is a major means clear which discharge option would be
step in reducing environmental quality control most desirable in the sub-tidal environment
time-lag from sequences of analyses of off sandy beaches. A site-specific decision
effluents and receiving ecosystems dependent may have to be reached depending on such
on a priori guesses about toxins of concern. factors as:
Although the controls were not effective 1. Steep submarine topography permitting
in preventing contamination at Jordan River, density current flow to the bottom of
circumstances showed the most serious flaw. It nearby deep channels, or gentle
was the inability to ensure that the offshore gradients causing formation of perched
discharge point was maintained by effective aprons of deposits.
engineering and by quick administrative 2. Exposure of the coast to rare event
responses to accidents, i.e. outfall breaks. windstorms and tsunamis which could
In contrast there is some evidence that the damage an outfall or resuspend a
province-wide effluent composition objectives perched apron of deposits with
(Department of Lands, Forests and Water contaminated interstitial water.
Resources, 1974) which were site-specifically 3. A populated coast with many residents
modified for Jordan River have been effective dependent for subsistence on local
over a ten-year period in preventing bioaccumulating fish and she llfish
widespread bioaccumulation sub-tidally species, and with a limited range of
(Pelletier, 1982). Br~tannia Beach in contrast upper trophic level species thereby
reflects the earlier virtually total lack of biomagnifying contaminants (as was the
pollution controls (1899-1974). case in the Minamata disaster
It is important that, if circumstances (Forstner, Wittmann, 1979; Young, in
should require a sand beach coast to be used press».
for receiving waste discharges, there be In summary, if site-specific
effective regulations, adequate engineering circumstances indicate that waste discharge to
design, construction and monitoring and the sea across a sandy beach may be needed as
particularly, efficient enforcement and speedy a disposal option for an industrial plant or
constraints. Even more important is that the city, there are possibilities for mitigating
present lack of information about offshore impacts previously found on other beaches.
waste dispersal (and assimilation if it
726

ACKNOWLEDGEMENTS Pelletier CA (1982) Environmental data

handling and long-term trend monitoring at
The authors acknowledge with gratitude funds
Island Copper Mine. In Ellis DV, ed. Marine
provided by the B.C. Ministry of Environment, tailings disposal, pp. 197-237. Ann Arbor,
Michigan, Ann Arbor Science.
the University of Victoria Faculty Research
Phillips DJH (1980) Quantitative aquatic
Fund, the Natural Sciences and Engineering biological indicators. London, Applied Science
Publishers Ltd.
Research Council Strategic Grants Programme to
Popham JD and D'Auria JM (1982) The effect
the Environmental Toxicology Group, University of body size, season and trace element leve ls
in Mytilus edulis, Arch. Environ. Contam.
of Victoria and Dr. J. D'Auria, Department of
Toxicol. 11, 273-282.
Chemistry, Simon Fraser University. Popham JD and D'Auria JM (1983) Combined
effect of body size, season, and location on
trace element levels in mussels (Mytilus
REFERENCES edulis), Arch. Environ. Contam. Toxicol. 12,
~
Cairns J and Gruber D (1980) A comparison of
Stevenson JC (1978) Recovery potential of
methods and instrumentation of biological
oiled marine northern environments, J. Fish.
early warning systems, Water Resources
Res. Bd. Canada 35(5), 499-796.
Bulletin 16(2), 261-266.
Stump IG, Kearney J, D'Auria JM and Popham
Cross SF and Ellis DV (1981) Environmental
JD (1979) Monitoring trace elements in the
recovery in a marine ecosystem impacted by a
mussel, Mytilus edulis, using X-ray energy
sulfite process pulp mill, J. Wat. Poll. Cont.
spectroscopy, Mar. Pollut. Bull. 10, 270-274.
Fed. 53(8), 1339-1346.
Young DR (in press) Methods of evaluating
Department of Lands, Forests and Water
pollutant biomagnification in marine
Resources, Water Resources Service, Victoria,
ecosystems. In Proceedings of the workshop on
B.C. (1974) Report on pollution control
meaningful measures of marine pollution
objectives for the mining, mine-milling, and
effects, Pensacola, Florida, 1982.
smelting industries of British Columbia.
Victoria, Queen's Printer.
Ellis DV (1975) Pollution controls on mine
ADDENDUM
discharges to the sea. In International
conference on heavy metals in the environment, The verbal presentation of this paper was
Oct. 27-31, 1975, pp. 677-686. Toronto,
based on a concept apparently well-known to
Ontario, Canada.
Ellis DV (1980) Environmental consequences symposium participants, that of Murphy's Law.
of breaks and interrupted construction at
"If something can go wrong, it wilL" An
marine outfalls in British Columbia. In
Coastal discharges, pp. 123-126. London, Environmental Engineering Corollary was
Institution of Civil Engineers.
formulated. "An environmentally engineered
Ellis DV, Gee P and Cross SF (1981) Recovery
from zinc contamination in a stock of Pacific development is only as good as the engineers
oysters, Crassostrea gigas (Thunberg), Water
who design it, the contractors who build it,
Poll. Res. J. of Canada New Series 15(4),
303-310. the scientists who monitor it, and the
Ellis DV, ed. (1982a) Marine tailings
regulators who police it."
disposal. Ann Arbor, Michigan, Ann Arbor
Sc ience.
Ellis DV (1982b) Ocean disposal in British
Columbia. In Oceans 82 conference record,
Sept. 20-22,1982, pp. 1101-1106. Washington,
D.C.
Forstner U and Wittmann GTW (1979) Metal
pollution in the aquatic environment. Berlin,
Springer-Verlag.
Freeland HJ, Farmer DM and Levings CD (1980)
Fjord oceanography. New York, Plenum.
Hay AE (1982) The effects of submarine
channels on mine tailing disposal in Rupert
Inlet, B.C. In Ellis DV, ed. Marine tailings
disposal, pp. 139-181. Ann Arbor, Michigan,
Ann Arbor Science.
 727

MONITORING BEACH AND DUNE ADVANCEMENT AND VEGETATION CHANGES 1937 - 1977 AT THE FARM TWINSTREAMS,
MTUNZINI, NATAL, SOUTH AFRICA

WEISSER, P.J. and A.P. BACKER (Botanical Research Institute, Private Bag X101, Pretoria, 0001,
South Africa)

1 . INTRODUCTION
Most of the South African coastline is static or

~
receding. Therefore, the coastal stretch between
the Tugela River Mouth and the Mlalazi Estuary Mouth
is of special interest, as coastal prograding is
area
taking place. Weisser et al. (1982) studied the
250 m
dune advancement using air-photo interpretation at I I
Siaya Lagoon
the'MlalaziNature Reserve (1937 - 1977), finding a
coastal progession of 95 m (2,4 m per year). During
1982 maps of the Siaya Lagoon and parts of Twin=
streams Farm were produced by Mrs D Selby and Mrs
JE Perry of the Coastal Engineering and Hydraulics

'Division of the National Research Institute for
Oceanology, (NRIO), Stellenbosch. The maps were
based on air-photo interpretation and open up the
possibility of doing additional measurements on
coastline changes.
The coastline studied is part of the Farm Twin=
streams and stretches from the Siaya Lagoon Mouth
{latitude 28° 58' South, longitude 30° 45' 45"
EasB to 1,5 km South (Fig. 1). From Mtunzini to
the Tugela Estuary, the coastal plain is absent and
a series of Holocene beach ridges forms a prograding
shore at the margins of subdued hills (Fig. 2).
Offshore, a comparatively shallow continental shelf
terminates in a pronounced slope change at depths
of between 90 and 135 m From Cape St. Lucia the
shelf widens southward, reaching a maximum width of
45 km southeast of the Tugela Estuary before
narrowing again to the south (Orme, 1973).
---,--
1937
.-/ 1'151-
1977
../ i;;dian Ocean N
FIGURE 1. Map of part of the study area
showing the advancement of the position of the
1937, 1957, 1965 and 1977 limit between beach
and foredunes. Based on maps by JE Perry & D Selby
(NRIO, 1982).
The climate is humid and warm to hot with a high
year-round rainfall (Schulze, 1965), the mean
annual temperature at the Cape St. Lucia Station
being 21,5° C. The mean annual rainfall on the
nearby Storkwood farm is 1 215 mm (NRIO, 1982).
The following introductory paragraphs on factors
affecting dune formation, such as surface winds
and shore zone dynamics are based on Orme (1973).
728

The prevailing surface winds along the Natal coast

are northeasterly and southwesterly. These winds
flow roughly parallel with the coastline and are
almost equally divided in frequency and velocity.
They are spread fairly everly over the 12 months
of the year. Wind velocities are subject to strong
diurnal variations, daytime winds being signifi=
cantly stronger than wind at night.

Shorezone dynamics are conditioned by prevailing

northeasterly and southwesterly winds, by complex
nearshore and coastal circulation systems, and by
seasonally high sediment discharges from neigh=
bouring watersheds. Nearshore circulation patterns,
consisting mainly of asymmetric cells, generate
longshore currents exceeding 1 m/sec Inner and
outer breaker zones reflect offshore bars. Beyond
the nearshore circulation, reversing coastal
currents related to the Agulhas Current System
flow parallel to the shore with surface velocities
of 0,2 to 0,8 m/sec, reduced by a factor of 0,4
near the seabed. Both coastal current and near= FIGURE 2. Aerial photograph of the approximately
450 m broad area showing the dune ridges and
shore circulation systems favour a northeastward
their vegetation ranging from dune pioneer
movement of materials along the southern Zululand (on left) to the dune forest reaching 18 m
(October, 1982).
coast, with strong onshore components added.
During the winter dry season, strong littoral drift
information; (3) to revise previous data on
creates barrier beaches across estuaries. During
vegetation-succession chronology.
the summer rainy season, these barriers are
breached and vast sediment plumes extend seaward
2. METHODS AND METHODOLOCIAL CONSIDERATIONS
and alongshore. Orme estimates that the Tugela
Thirty systematic sampling lines per=
brings 10,5 x 10 6 tons of sediment to the coast
pendicular to the coast each 1 cm apart were
annually. The materials contributing to dune
drawn on the maps published by Selby and Perry
growth come primarily from the high sediment loads
(NRIO, 1982) and changes in the position of
brought down seasonally by Zululand rivers. Cliff
the beach-dune limit measured for the year 1937,
erosion is a minor contributor of materials.
1957, 1965 and 1977. Data for years 1953 and
1969 were omitted because of the very short
Additional information on sediment loads of the
periods between the flight.
Tugela River and sediment dynamic of the coastline
can be obtained from J. Nicholson (unpublished).
The potential and limitations of air-photo-
interpretation in vegetation studies have been
Objectives of this study were (1) to monitor and
discussed by Edwards (1972) and, in relation to
quantify beach and foredune advancement; (2) to
dune vegetation in Zululand, by Weisser (1979).
describe the dune vegetation as background
 729

Concerning this study the following points should

be borne in mind: the photographs are at different
scales; they were not taken at the same time of
the year, which impairs their comparability; and
the differing resolution of the air photos, the
1937 photos being the weakest.

As pointed out by Weisser et al. (1982), the inter=

pretation of the results is complicated because
dune advancement is not a linear process. It
occurs in pulses, each pulse corresponding to the
detection of an additional dune ridge on the air
photo. There is a time lag between the establish=
ment of Scaevola thunbergii seedlings, the
accumulation of sand and the appearance of the new
ridge on the air photo. These dunes are at first
isolated and later coalesce to a ridge parallel to
the coast (Figs 2 and 3).
3. RESULTS
3.1. Beach and foredune prograding
Figure 4 summarized the results. Taking the
situation of 1977 as baseline, the 1965 beach/fore=
dune limit was situated approximately 68,8 m
landwards; the 1957 limit 97,1 m ; the 1937 limit
121,5 m (Fig.1). This 121,5 m represents the
seaward advancement during 1937 - 1977 of the
beach/foredune interphase and the rate over this
forty year period was 3,04 m per year. This value
is similar to that reported by Nicholson (un=
published) . The advancement rate shows an in=
crease from 1,25 m per year for the period
1937 - 1957 to 3,56 m per year (1957 - 1965) to
5,67 m per year (1965 1977). Figures 1 and
4 may imply a progressive advancement of the
foredunes. This could be misrepresentation,
because storms and equinoctail tides could
temporarily reverse the trend and destroy the
foredune partly or completely. Therefore, some of
the ridges found today may not be those that were
established first.
FIGURE 3. Foredunes at Twinstreams, Mtunzini
(June, 1978). The new Scaevola thunbergii dune
started in June 1973
The dune advancement at Twinstreams is occurring
through the formation of additional phytogenic
dune ridges (Fig. 3). Seedlings of Scaevola
thunbergii, a common dune pioneer, establish
themselves on the beach seawards of the fore=
dune. They act as sand traps, as sand accumu=
lates around the newly established plants.
Upward growth of the plant prevents their
burial and causes the formation of the ridges.
3.2. Dune vegetation
This description of the vegetation at the study
area is based on information published by Moll
(1972) and unpublished data gathered by J
Walker and Weisser in 1979 and is given as
background information.
About sixteen vegetated dune ridges parallel to
the shoreline covered with vegetation from
recent pioneer stages to more mature dune
forest were present in 1977 at Twinstreams
(Fig. 2). There is a successional gradient,
730
seral stages and zones are distinguished for
descriptive convenience.
3.2.1. Dune Pioneer Zone. It is about 50 m
to 70 m wide and comprises the dunes mainly
colonized by Scaevola thunbergii. Other species
that can be found are Arctotheca populifolia,
Ipomoea pes-caprae,Gazania rigens var. uniflora
and Tephrosia purpurea subsp. canescens. The
community structure is simple, the field-
layer usually 0,5 - 1,2 m high. A total cover
of five per cent was estimated at the sampled
site (Fig. 3).
3.2.2. O~en Dune Scrub. This community be=
gins with the establishment of the shrubs
Passerina rigida and Helichrysum ericaefolium
and the grass Stipagrostis zeyheri. Protected
from onshore winds by seaward ridges, the sand
movement in this zone is slight. The plant
cover and species diversity increases land=
wards. Species present in this zone are
Chrysanthemoides monilifera, Carpobrotus
dimidiatus, Chironia baccifera, Gloriosa
superba, Helichrysum kraussii and Cyperus sp.
Seedlings and young plants of the Closed Dune
Scrub such as Eugenia capensis and Mimusops
caffra occur (Fig. 2) and become progressively
dominant when advancing landwards.
This community has two layers. In the
sampled site the field layer was 0 - 0,8 m and
the scrub layer mainly 0 - 2,5 m. Average
cover values were 10 - 60 per cent for the
field layer and 5 - 50 per cent for the scrub
layer.
3.2.3 Closed Dune Scrub. There is a gradual
transition from Open to Closed Scrub (Fig. 2).
Common species in the Closed Dune Scrub are
Eugenia capensis, Rhus nebulosa, Brachylaena
discolor and Colpoon compressum. Other canopy
species forming the thicket are Mimusops caffra,
Apodytes dimidiata, Canthium obovatum, Tricalysia
sonderana, May tenus nemorosa and Rhoicissus
digitata. Many of these species also occur in
the Dune Forest. In the field layer Phymathodes
scolopendria is a conspicuous and often domi=
nant fern.
Structurally this community is a thicket with a
dense, windpruned canopy caused by salt laden
onshore winds. Total cover values ranged be=
tween 70 and 90 per cent and the height varied
from one to four metres.
3.2.4. Dune Forest. At Mtunzini the Dune
Thicket develops into Dune Forest in the dune
slacks, the trees become taller, the stem
diameter broader, the lianes and epiphytes
more abundant and the community multilayered.
Abundant litter is found on the forest floor
and soil horizons are well defined.
Common canopy trees are Mimusops caffra,
Apodytes dimidiata, Sideroxylon inerme, Euclea
natalensis, Dovyalis longispina, Allophyllus
natalensis, Olea woodiana, Canthium obovatum,
Scolopia zeyheri and Vepris lanceolata. In
the understory Peddiea africana, Monanthotaxis
caffra, Kraussia floribunda, Carissa bispinosa,
Psychotria capensis and in the field layer
Phymathodes scolopendria, Scadoxus puniceus,
S. natalensis, Crocosmia aurea, Oplismenus
hirtellus, Cyrtorchis arcuata and Dietes sp.
were noted. The dominant field layer species
Phymathodes scolopendria is replaced in the
older parts of the forest by the densely grow=
ing Isoglossa woodii. Climbers included
Rhoicissus rhomboidea, R. tomentosa, Asparagus
falcatus, Dalbergia armata, Dioscorea spp.,
Acacia kraussiana and Pupalia atropurpurea.
Epiphytes such as the fern Microsorium punctatum
are present.
The structure of the Dune Forest is defined by

canopy trees ranging from 5 to 18 m with an
estimated cover of 40 - 90 per cent. The under=
story is poorly defined, heights between 3 - 7 m
and cover values between 3 - 50 per cent were
estimated. Frequently the shrub Isoglossa
woodii occupies the following lower layer form=
ing a dense vegetation of 0 - 2,5 m high and
with a cover of 60 - 95 per cent.
The vegetation structure and floristic composi=
tion found at Twinstreams is similar to that
reported by other authors for Natal dune vege=
tat ion (e.g. Edwards 1967, Venter 1972 and Ward
1980). The main difference is the "consertina
effect" that the prograding of the coast causes
at Mtunzini. Young, medium and older dunes
spatially well separated and thus presenting
the vegetation in a "model situation" ideal for
the study of vegetation succession on dunes.
This was already recognised by Moll (1972).
Sugar cane plantations replace the forest on
the old dunes with red soils westwards of
Mtunzini (Fig. 2). Venter (unpublished) found
south of Richards Bay the Mimusops caffra Dune
Forest being followed in the succession by a
Celtis africana Community. Weisser (unpublished)
found the same north of the Mlalazi Mouth. This
suggests an incomplete successional sequence
at Mtunzini.
3.3. Vegetation-succession chronology
Weisser et al. (1982) published a preliminary
vegetation-succession chronology by dating the
dunes assuming a yearly dune advancement rate
of 2,4 m and applying this to the tranSect
published by Moll (1972). The new rates of
dune advancement applied to the vegetation in=
formation from orthophoto map 2831 DC 25 Ekuhleni
(1977) and air photo 160, Job 240 (1976) allow
a refinement of that vegetation-succession
chronology.
731
120
100
80
60
§
~40
OJ
E
OJ
,.~ 20
>
"0
«
1930 60 70 80
Year
FIGURE 4. Foredune aavancement at Twinstreams
Farm. Total distance of the dune advancement
1937-1977 was 121,5 m and the average rate 3,04 m
per year.
If one takes the spatial position of the seral
stages as an indication of their age by apply=
ing and extrapolating from the Figure 4 graph
onto a vegetation profile, then the following
vegetation-succession chronology (Table 1) is
obtained:
(1) it takes about 5 years for a dune ridge to
be formed under present conditions (Fig. 3);
(2) it will take about 13 years for the Open
Dune Scrub Community to commence colonization;
(3) it will take about 71 years before the
dune 1S covered by a Closed Dune Scrub;
(4) if protection against seawinds and salt=
spray is given by the seaward ridges and their
vegetation, a Dune Forest could develop after
about 119 years, beginning in the dune slacks
and later spreading from there.
As pointed out by Weisser et al. (1982) the
space/time relationships obtained here must be
regarded as a gross approximation and should
be followed up by studies with permanent plots.
C J Ward (pers. comm.) and MacDonald and
Pammenter (unpublished) have laid out permanent
vegetation plots and transects in the Twinstreams
area and Mlalazi Nature Reserve which will allow
future refinement of this interim dune vegeta=
tion-succession chronology.
732

TABLE 1. Spatial range of dune vegetation and its inferred age resulting from applying graph on
Figure 4 onto a profile (1977) through the dunes at Twinstreams

Distance from landward Time taken for the Approx. spatial Approximate
beach limit to begin= community to begin range of duration of
ning of sere to establish Communities seral stage

m Years m Years
Pioneers 0 0 0-70 ~13
Open Dune Scrub 70 13 70-160 ~58
Closed Dune Scrub 160 71 160-220 ~48
Dune Forest 220 119 220-450

4. ACKNOWLEDGEMENTS Engineering and Hydraulics Division, National

We would like to thank Drs J C Scheepers, D Research Institute for Oceanology, Council for
Edwards, Prof. P Hall, Messrs W de Waal and Scientific and Industrial Research, NRIO
P Frost, Mesdames S Frost and J N Weisser and Memorandum 8134.
Miss L Smith for their valuable comments; Mr Orme AR (1973) Barrier and Lagoon Systems
C Buthelezi for his help during field work; along the Zululand Coast, South Africa. In
Mesdames D Selby and J E Perry for permission Coates DR, ed. Coastal Geomorphology, pp. 181-
to use their maps and Mrs M v d Merwe for 217, Binghamton, State University of New York.
typing the manuscript; Mr R E Crofts (office Schulze BR (1965) General survey. Climate
of the Surveyor General), Prof. D A Scogings of South African Series Part 8, Weather Bureau
and Mr A Bikaroo (Survey Department, University Publications, Pretoria, Dept. of Transport.
of Natal) for their collaboration in obtaining Ward CJ (1980) The plant ecology of the
aerial photographs; Mrs A Romanowski for Isipingo Beach Area, Natal, South Africa, Mem.
photographic work; and the Natal Parks, Game bot. Surv. S. Afr. No. 45.
and Fish Preservation Board and Mr I F Garland Weisser PJ, Garland IF and Drews BK (1982)
for their support during field work. Dune advancement 1937-1977 at the Mlalazi
Nature Reserve Mtunzini, Natal, South Africa
5. REFERENCES and a preliminary vegetation-succession chrono=
Edwards D (1967) A plant ecological survey of logy, Bothalia 14, 127-130.
the Tugela River Basin, Mem. bot. Surv. S. Afr. Weisser PJ (1979) Suitability of air-photo
No. 36. interpretation for monitoring coastal dune
Edwards D (1972) Remote sensing in the evalu= vegetation of the Zululand Dunes, South Africa.
ation of the natural vegetation resources of In: The use of ecological variables in environ=
South Africa, Proc. 5th Symp. Remote Sensing, mental monitoring, The National Swedish
Pretoria, CSIR, May 1972, pp. 99-102. Environment Protection Board, Report PM 1151,
Moll EJ (1972) A preliminary account of the pp. 62-72.
dune communities at Pennington Park, Mtunzini,
Natal, Bothalia 10, 615-626.
NRIO (1982) Hydrological hydraulic study of
Natal Estuaries. The Siaya Lagoon, Coastal
 733

MANAGEMENT FOR SURVIVAL - A REVIEW OF THE PLANT ECOLOGY AND PROTECTION OF THE 'MACHAIR' BEACHES OF
NORTH-WEST SCOTLAND

R.E. RANDALL (Girton College, Cambridge, U.K.)

Low, flat windswept sandy plains usually rich in the hills - 'blackland' (Fig. 2). Machair sand
shell fragments are widespread in the Outer
Hebrides, Tiree and ColI and also occur on other
islands of the Inner Hebrides and the Northern
WEST EAST
Isles as well as the Scottich mainland west-coast stable
grass
machair
marshes
black land

mobile cultivated machalr moor·

(Fig. 1). Known regionally as 'machair', these dunes machai r loch land

Sea

FIGURE 2. A cross-section showing geographical

relationships of machair and associated habitats.

consists of a mixture of calcareous and siliceous

components in variable proportions. Particularly
on the mainland the CaC0 3 content may drop below
20% whereas on some of the Outer Hebridean sites
it may be over 80%. Typically there is a decline
in CaC0 3 content from the shore inland as organic
matter increases in the substrate. The silica
fraction, along with the shingle ridges which
often occur at the back of beaches is the reworked
remnant of glacial and fluvioglacial sediments
dropped on the shallow offshore platform around
the islands. The calcareous fraction comprises
comminuted shell-fragments of marine animals
FIGURE 1. The main machair coasts of Scotland.
which existed or still live in the shallow waters.
Maximum shell-production is thought to have
beach systems are best developed along the west
occurred in earlier Holocene times with lower
coast of the Uists. In classic sites the machair
sea-levels and the major period of sand deposition
has a line of dunes on its seaward side which
occurred between 5.700 BP and 2,000 BP when there
vary in age and stability; its inland boundary is
was an excess supply of material in the offshore
often demarcated by a line of brackish lochs and
zone. Currently there is only a small amount of
impeded drainage beyond which occurs acid peat of
734
sand available for further beach growth except at
a few sites where an abundant shelly fauna still
survives (Barra and S. Uist).
The sandy shores on the seaward edge of the machair
sequence may be long and broken only by low, rocky
outcrops of Lewisian gneiss, as in the Uist~, or
may be confined to small bays between high
headlands, as at Opinan south of Gairloch. At its
upper limit the shore is usually replaced by a
shingle/cobble ridge on the back of which dunes
may develop either as a single crest ridge or
covering a wider area. These dunes reach a
maximum of 20m OD. In a few localities dunes are
absent. The dune systems, where present, are
usually severely eroded by wind. Corridor
blowouts cut through some of the coastal dunes and
result in redepositional hummocks further inland.
At other sites cauldron-blowouts erode down to the
shingle/rock platform or to water-table giving
sites for dune hollow or dune slack vegetation.
Blowouts may well occur through extreme natural
events but have been exacerbated by disturbance
by man and the 19th and 20th century introduction
of rabbits to the region. Locally organized sand
control is virtually nonexistent.
Behind the dunes the machair plain occurs. This
area of low relief, usually around Sm OD, is
normally flat or sloping gently inland, and is
geologically older than the dunes. The present
machair plain is thought to have been created by
erosion and reworking of terrestrial sand since
the phase of accretion from a seaward source
ended around 2,000 years ago. The topography is
characterized by a flat plain of deflation which
is often complicated by redeposition of sand
hillocks from dune or machair plain blowouts.
The level of the machair plain is determined by
the underlying water-table and this level lies
somewhere between its minimum summer and winter
positions. In the long formation of the machair
plain it is thought that an original, high-level
surface has been eroded by wind action and that
eventually the deflation plain will extend from
the sea to the blackland.
The peripheral lochs and lochans are sited at the
juxtaposition of base-rich and base-poor habitats.
Because of their relatively recent origin (i.e.
around 7,000 years) and strong marine influence
they are eutrophic, variably brackish and
virtually unique in western Europe. They contain
a rich invertebrate fauna and dense banks of
submerged and emergent macrophytes which contrast
markedly with the acid peat land lochs inland of
the beach ecosystems.
Although most machairs lie between latitudes
S7°N - 60 o N, the oceanic location results in
mild winters but low accumulated temperatures
in the short growing season. Sunshine hours are
low but the annual precipitation is around
1S00mm - hence potential evapotranspiration
deficits are uncommon. The sandy soils and
strong winds make drought a frequent summer
problem, yet the higher winter rainfall results
in considerable flooding from October to March.
The summer water-table may be over 3m below the
surface of the machair plain. Wind is the most
notable feature of the climate: 44% of days have
records of gale force gusts mainly from the south.
Storm winds may occur from any direction and at
any time of year and sand from blowouts at the
rear of the machair may well be reworked down to
the foreshore.
 735

plain is dominated by Festuca rubra with typical

frequency values of over 80%, and legumes (Lotus
ssp. and Trifolium spp.) as well as many other
15 • over 50 km/h. § over 20·5 km /h vascular and non-vascular species - up to 35
I!lIIll over 40 kmJh o total ·wlnds
species per metre square being common (Fig. 5).
~ over 33 kmJh
In some areas endemic sUb-species of orchids are
frequent. These characteristic plant communities
are kept in balance and held from succession by
grazing, without which many areas would become
Calluna or Molinea heath.

40- 170- 200- 230- 260-

130 160 190 220 250 280
N NE E SE S SW W
Degrees True
FIGURE 4. Typical dune vegetation, Monach Isles.
FIGURE 3. Percentage frequency of winds of
different speeds from each direction calculated
over four years. Benbecula.

Thus machair is a naturally dynamic and fragile

coastal ecosystem but is normally well-covered
with vegetation. The insularity of the area does
not curtail the diversity of machair plant
species - the 200ha of the Monach Islands contain
220 flowering plants. Floristic richness
distinguishes machair plant assemblages from the
more siliceous coastal systems elsewhere in
Britain. Dune areas are characterized by the
presence of Arnmophila arenaria and exposed soil
surface. Compared with dune systems elsewhere
FIGURE 5. Herb-rich machair sward, Monach Isles.
Agropyron junceiforme plays a limited role and
non-vascular plants are scarce with the exception
Indeed grazing and occasional ploughing have been
of Rhytidiadelphus squarrosus and Peltigera
so integral a part of the development of the
canina (Fig. 4). The grassland of the machair
machair that, without it, machair would not exist
736
in its present form. The machair has been
populated for over 3,00 years and during this
period it has usually been farmed in such a way
that little loss has been caused either to its
fertility or its ecological interest. The economy
of the crofters of the machair has traditionally
been based on seasonal cattle and sheep grazing
and some potato, oats and rye cultivation. In the
past manuring and the spreading of kelp
(Laminaria spp) over the machair were common
agricultural practices. Chemical fertilizers do
not add to the organic matter of the friable soil
and increase its susceptibility at erosion. More
frequent ploughing, excessive traffic, overgrazing,
or large rabbit populations all result in
destruction of vegetation and severe wind erosion
(Fig. 6).
FIGURE 6. The effect of excessive caravan traffic,
Clachtoll.
Contempory documentary evidence on the condition
of machair and its use by man is rather limited
prior to the nineteenth century. However, the
Old Statistical Accounts of the 1790's and the New
Statistical Accounts of the 1840's, which provided
descriptions of Scottish parishes contain frequent
mention of sandblow in the areas of machair. This
new phenomenon of the late eighteenth century
onwards occurred at a time of socio-economic
change in the Highlands and Islands of Scotland.
With growing populations and re-organization of
land tenure, great pressure was put on the machair
landscape. Concurrently the growth of a kelp
industry (for its chemical constituents) led to a
decline in the use of seaweed for mulch and
fertilizer for the machair and increased pressure
of foot and cart traffic across the machair plain
to the shore. Agriculture changed from occasional
crop years and long periods of fallow to frequent
cropping, short fallows and low yields. As crops
declined so Ammophila was used for thatch instead
of straw and this reduced the stability of the
dune systems. In some places dunes were
'reclaimed' for cultivation. In other areas of
machair the Highland 'clearances' of population of
the mid-nineteenth century and the conversion of
areas to sheep farms led either to a reduction in
grazing pressure and redevelopment of stable dunes,
or to drainage and therefore increased suscept-
ibility to summer erosion. The outcome of these
changes was extensive planting of some dunes and
machairs with Ammophila to prevent sandblow and/or
regrading and turfing of some erosional faces so
that stabilization eventually returned at a lower
level. Hence many of the current machair surfaces
are only a century or so old.
The present period is seeing another marked change
in the land-use and control of machair which may
have much more severe consequences than that of
the nineteenth century. There are both changes in
the agricultural practices on the machair, and
the system of land holding is also being altered.
Many areas that were previously held as common
grazing are now being split up and rented to
individual tenants. Concurrently the whole legal
basis of crofting tenure has been reviewed over
the late 1970's. The machair provides a relatively
dry and sheltered environment for stock over the
winter in contrast to the rough hill grazings.
In summer it is used for the production of winter
feed for stock, primarily small oats (Avena
strigosa) and an indigenous rye (Secale cereale),

and also for potatoes. With low soil fertility
and proneness to drought, there is usually a one
or two year agricultural use followed by two or
three years fallow, when the area becomes
repopulated with the natural legume-rich Festuca
sward (Fig. 7).
FIGURE 7. Modern agricultural use of machair beach
at Borve, Benbecula.
This form of agriculture results in low yields,
especially in dry years when crop failures and
sandblow occurs. Modern forms of grain will not
tolerate the high pH or manganese deficiencies in
the soil and factors such as small field size and
large-awned grain prevent the use of modern
machinery. The excesses of calcium result in
deficiency diseases in cattle and sheep due to
unavailability of copper and cobalt. The only
viable alternative form of agriculture for the
machair is perhaps increased grass production for
silage. This is being advocated in some quarters
using sown grass crop drilled below the surface
immediately after ploughing and well fertilized,
or by drilling into uncultivated land previously
treated with complete herbicide. Such methods are
only viable if common grazing ceases and fencing
enables establishment and management.
Sown fodder-grass, fertilizers and herbicides will
obviously eliminate the colourful herb-rich
737
successional communities that occur during periods
of fallow and will also seriously affect the large
snail and invertebrate fauna. Also because of the
denseness of the bird-breeding population (e.g.
corncrake, phalarope, ringed-plover, oystercatcher,
dunlin, black-tailed godwit and ruff) and winter
migrants (especially barnacle geese) this coastal
system has been designated of global significance
by the World Conservation Strategy. These, birds
too, would not survive the change to a silage-
based agricultural economy. Until now such a
marked agricultural change has been hampered
above all by lack of money but as of 1/9/82 the
Western Isles Integrated Development Programme
administered by the Department of Agriculture and
Fisheries for Scotland has made available £200,000
for amalgamation of holdings and rearrangement of
common land as well as drainage, herbicides and
seed. In all £20 million will be made available
to agriculture over the next five years. At
present there are 35 Sites of Special Scientific
Interest in the machair which would not directly
be affected by agricultural change but it is vital
that some mechanism is provided to ensure that
measures taken within the IDP are compatible with
the total machair beach ecosystem. So far little
funding has been made available for scientific
monitoring of the possible effects of such
activities as land-drainage, chemical fertilization
or intensification of agricultural use. Because
this area is one of Europes 'less-favoured'
socio-economic regions, the investment is being
made mainly agriculturally rather than via tourism
or craft-industry which could be more easily
controlled and integrated into the wildlife
heritage which the machair beaches possess.
Agricultural practices on the 'blackland' adjacent
to machair sites can also affect the coastal
ecosystem. During the 1960's around 6,000ha were
improved by application of calcareous beach sand
at 20 tonnes per hectare. If this activity is
resumed under the IDP careful planning will be
required to conserve the more important scientific
738

areas, especially as there is little contempor-

aneous replacement of shell £ragments.

The geographical isolation of most machair be,aches

makes the exploitation of local sand or shingle
for construction work essential, particularly as
there is infrequently an alternative source of
fluvio-glacial sands and gravels. Nevertheless
this can seriously endanger coastal stability and
over recent years many sites have been made the
subject of Coast Protection Orders. Crofting
communities do have a legal right to extract sand
and shingle within their townships and there is
still need to protect the more important beach
systems especially in the Outer Isles of Lewis
FIGURE 8. Regrowth in non-grazed exclosure on
and Benbecula.
Opinan machair.

Recreational pressures on machair beaches are

Short-term results show a marked increase in
relatively low but increasing as road and ferry
flowering of the machair, especially in rabbit
access improves. A recent survey showed 80% of
exclosures but also in areas of domesticate
campers and caravanners in north-west Scotland
control, an increase in winter accretion of sand
sited on or adjacent to a beach. At the same time
within the higher biomass and a healing of blowout
rabbit numbers have begun to build up again after
ruts caused by vehicle wheels. At Achmelvich and
the ravages of myxamatosis. Thus the scientific
several other sites (e.g. Clachtoll, Mellon
interest of several fragile areas could easily be
Udrigle) reseeding of blowouts with Lolium in a
destroyed through overuse, as erosion increases.
peat mulch or planting of Ammophila and Elymus
At the same time the recreation value of the sites
has enabled beaches to regain their attraction.
is destroyed. The Scottish Countryside Commission
This, coupled with the provision of car parks and
has set up case-studies at two mainland machair
small-scale camping sites funded under the
sites at Opinan and Achmelvich to examine grazing
Countryside (Scotland) Act, 1967, has enabled
and recreational pressure. At Opinan a 30ha beach
several areas to sustain much greater numbers of
has been excluded from grazing pressures and
visitors without harm and so help to promote local
vehicular access by a perimeter fence. Personal
prosperity.
access is directed via stiles and gates and an
agreed rota of limited grazing has been implemented
However care must be taken in such development
(albeit with difficulty). Control plots and
for many beaches lose both tourist and scientific
rabbit exclosures have also been erected (Fig. 8).
attraction if their wildness and remoteness are
lost by overuse. Organized, large-scale
development must be limited to the ideal sites.
Crofting tenancies and an ageing population will
limit development at the majority of beaches in
the short term but farsighted planning is needed
to decide which locations might be carefully
 739

developed for tourism and associated craft
industries and which preserved undeveloped.
Total machair planning must occur before damage
becomes irreparable.
A final potential pressure on the machair
ecosystem is that of pretroleum production in the
Mesozoic and Tertiary sediments of the continental
shelf off north-west Scotland. If recoverable
resources are discovered, pipeline landfalls need
flat approaches and processing plants need flat
land. Machair complexes would be ideal, yet highly
susceptible to any increases in human activity.
into the system have resulted in a non-equilibrium
situation and severe erosion. Where this has
happened on a local scale the system has
readjusted when the activity has ceased (Fig 9).
It is vital that the potentially more serious
perturbations of the equilibrium conditions at the
present time do not permanently deflect the
return to an equilibrium system.

Area above high water Area above high water Eroded face on dune
Unvegetated face
mark with large sand supply mark with limited sand supply or sand hill

Developmental phase

Initiating

Phase

Developing
general landwards
movements of sand
~_ ----Sand hi II CD u';:;-_

~
_ _ _ _ _ _ _ __

local
Mac'hair
salt marsh
---
- ~
t
Phase

sand/I
erosion
_---- Dune slack (~assland) - - - - r- ___ \ ~

\
/"
.'.
Culimating deposition of

from coastal area ~

~ complete remo~:~a::O:and and
Phase landwards movement of dune front
_ _ _ sea
- - - - - - Machair grassland (Grassland) - - - - - - --- Machair (Grassland) - - - -...-. invasion

Developmental Sequence Primary Sequence Composite Sequence Secondary Sequence

FIGURE 9. A hypothetical sequence of machair development. Developmental path ---->. Extent of

vegetation unit < - - - - >.

The machair ecosystem is a very delicate land form
which is in the medium term in equilibrium with
contemporary biological and physical processes.
Any short-term changes where man has failed to
adapt his agricultural or other activities to fit
REFERENCES
Caulfield, C. (1982) Friends the Hebrides can do
without. New Sci., 24th June, 862-863.
Dickinson, G. & Randall, R.E. (1979) An
interpretation of machair vegetation. Proc. R.
Soc. Edinb. B, 77, 267-278.
740

Dunn, E.E. (1980) Cropping the machair in: Sand-

dune machair 3, ed. D.S. Ranwell, 6-7,
Cambridge lnst. Terr. Ecol.
Mather, A.S. (1980) Man and the machair in the
nineteenth century. in Sand-dune machair 3,
ed. D.S. Ranwell, 12-13, Cambridge lnst. Terr.
Ecol.
Ritchie, W. (1979) Machair development and
chronology of the Uists and adjacent islands.
Proc. R. Soc. Edihb. B, 77, 107-122.
Randall, R.E. (1976) Machair zonation of the
Monach Isles NNR, Outer Hebrides. Trans. Proc.
Bot. Soc. Edinb. 42, 441-462.
Randall, R.E. & Band, W.T. (1981) Human impacts on
the beaches of north-west Scotland. Ecos. 2,
22-26.
 741

PART SIX

SOME ABSTRACTS OF PAPERS

NOT PUBLISHED IN FULL
 743

MUD ACCUMULATION OF A MICROTIDAL OPEN region. They are exposed in a broad

OCEAN BEACH coastal plain where a giant lagoonal
system is placed, being isolated from
L R MARTINS, J A VILLWOCK and I R the Atlantic Ocean through the growth of
MARTINS, (Universidade Federal do Rio a multiple complex barrier. 'l'he sandy
Grande do SuI, Instituto De Geociencias, deposits which constitutes the present
Porto Alegre, Brazil). coast line belong to the last barrier
that was built after the Holocene
The cyclic deposition of mud layers over transgression.
the fine and clear quartzose sands of
Cassino Beach in southern Brazil, have A transverse profile trough Cassino
been causing damage to the recreational Beach shows that the coast zone above
activities there. The study area, part the higher storm level is constituted by
of the Rio Grande do SuI Coastal dune fields and/or sand flats. The back-
Province, is composed of two major shore is 200m wide and has little dunes
geological elements, The Basement and whix:h are constantly reworked by the
the Pelotas Basin. The former is prevalent northeastern winds that
constituted by Precambrian Cristaline removed sands from the beach to the
Complex and the paleozoic and Mesozoic adjacent dune field and sand flat. The
sedimentary and volcanic sequences of fore-shore is narrower, 50m on average,
Parana Basin. The latter, a marginal as a consequence of a tidal amplitude
basin, was originated by faulting less than O,5m. There are beach cusps
related to the early stages of the there. 'l'he off-shore zone has a gentle
opening of the South Atlantic Ocean, slope and under the breaker zone there
during the Cretaceous. are sand bars.

Since that time the Pelotas Basin was The inner continental shelf of this area
filled by terrigenous clastic sediments is covered by relict sandy sediments and
deposited in continental, transitional near the Patos Lagoon outlet there are
and marine environments, which were sil ty and clayey sands which are l:Juild-
moved in time and space by successive ing a shoal fringe just in front of the
transgressions and regressions con- jetties that were constructed to protect
trolled earlier by the balance of navigation activities of the Rio Grande
subsidence and sedimentation rates and Harbor. These mud deposi ts are the
later, by the glacioeustatic oscillations unique recent sedimentary contribution
which took place during the Cenozoic Era. from continent to the shelf. The
suspended material that travels along
The upper part of this sequence is the Patos Lagoon by the Action of wind,
composed of predominantly sandy deposits waves and currents, is flocculated under
that enclose a series of discontinuous the influence of higher saline waters.
lito-stratigraphic units, variable in It concentrates near the bottom forming
age from Tertiary to Recent, as a result a fluid muddy layer along the deeper
of the migration of several sedimentary channels from where it is spreaded over
environments over the same the continental shelf by the ebb
744

currents. Mud accumulation over the back-shore were recovered by the wind
shore face is a periodical event which blown sands. In fact mud layers may be
is related to storm surges that beat the seen in trenches opened into the
coastal zone. When storm waves begin to back-shore areas of Cassino Beach.
produce a tangential pressure over the
muddy bottom of the shoals, the fine and Textural and mineralogical studies were
cohesive material begins to move towards made of this mUd. It was a clayey sandy
the beach zone. These waves operate as silt, with smectite, interstratified
a "bulldozer" that piles the muddy illite-smectite and kaolinite, the same
suspension up to the back-shore by an characteristics showed by the finer
unusual kind of mass movement. A sample sediments of the Patos Lagoon bottom and
of this material, collected at the the muddy sediments sampled at the
moment it reached the beach, contained shoals in front of the Patos Lagoon
213 g.l-l of solids. outlet, inner part of Rio Grande do SuI
continental shelf.
As soon as this black material is spread
over the sandy substrate of the beach it
looses part of its water content,
increasing its density and viscosity
which precludes the return of mud to the BEACH AND RIVER-MOUTH PROCESSES, NATAL
sea. Swash and backwash movements COAST, SOUTH AFRICA
produce a vertical accretion and
partially rewor k the recently deposited I L VAN HLERDEN, ( Coastal Studies
material, developing features similar to Institute, Louisiana State University,
rill marks. The black color of the Baton Rouge, Louisiana, 70803).
original suspension at the moment of air
exposure changes to brown by oxidation. A sedimentary process-response research
program was undertaken along the Natal
In 1978, a wedge like deposit (3,000m coast from April to august, 1976.
long, 100m wide and 1 m of maximum Specific site locations were Umgeni
thickness), was created in less than 24 River mouth, Umkomaas River mouth, and
hours, killing all the shore animal Amahlongwana Beach.
community and causing damage to tourist
activities. Fortunately this kind of Data showed that there was a close
mud accumulation has an ephemeral life. relationship among the northward
As it dries mud-cracks begins to develop migration of low-pressure cells, wind
producing a general fragmentation of the direction, swell approach angle, and
deposit. At the fore-shore wave erode resulting longshore drift. Longshore
and rework mud blocks which are readily drift was generally directed to the
transformed into clay balls that are north, although under onshore swell
scattered over the shore face and conditions the drift current may have
transported offshore. This mud remains had a southerly orientation. Field work
only where rapid burial occurs, e.g. was conducted after a period of major
depGsits which accumulated at the floods. 'l'he resultant large amount of
 745

sediment dumped in the nearshore greatly BEACH BARRIER SEDIMENTATION, CHANDELEUR

influenced coastal processes for a few ISLANDS, LOUISIANA
months thereafter. Generally, river
mouths are characterized by a sandy I L VAN HEERDEN, S PENLAND and R BOYD,
southward prograding spit and rocky (Coastal Studies Institute, Louisiana
southern bank. However, northward- State University, Baton Rouge, LA,
directed spit growth occurred at both 70803) .
the Umgeni and Umkomaas river mouths
during the early part of the study. Sea level transgression results from
This abnormal situation was related to subsidence following major distributary
the abundance of nearshore sediment as abandonment in the Mississippi Delta.
well as direction of swell approach. Subsidence here is mainly due to
compaction and dewatering of silt- and
The nature of the beach face, beach clay-sized sediments. Rates of
width, and size and organization of subsidence in abandoned delta complexes
beach cusps at Amahlongwana Beach were range as high as 1.3 meters/year. During
found to be predominantly swell influ- the resulting sea level transgression,
enced, wi th some tidal response. Cusp marine processes dominate sediment
development occurred under the influence dispersal and the generation of trans-
of onshore swells. The origin and gressive sedimentary facies.
spacing of these depositional features
appeared to be related to the position The combination of deltaic sediment
of nearshore rock outcrops and "swash source, subsidence, and marine re-
distance." Erosional cuspate features working on the hurricane-dominated
(giant cusps) and cuspate horns are microtidal Gulf Coast results in a
related to northward-directed rip current characteristic distribution of coastal
activity under the influence of southerly sedimentary environments. Wave
swells. Giant cusps had spacings greater reworking of deltaic sand bodies
than 500m. produces elongate barrier islands up to
40 km in length. These sandy coastal
Overall, beach and river-mouth geomorphic barrier tacies exist as flanking
conditions were dominated by the barriers, barrier island arcs, or inner
relative abundance of nearshore shelf shoals, depending on their age.
sediment. However, time-specific beach Seaward from coastal barriers are
and/or river-mouth process-response located offshore sand sheets marking the
features were controlled by major landward migration path taken by the
meteorological forcing functions. transgressive barriers. Hithin these
1-5 m sheet sands are buried tidal inlet
sand bodies up to 10 m thick and
characterized by migrating spit
sequences. The landward margin of
coastal barriers is composed of
horizontal, thinly bedded washover
deposits overlying fine-grained, shelly
746

lagoonal sediments. twice a week, i. e. 100 samples a year.

Each one included
The major marine processes determining organic carbon (dry burnt with LECO
sediment dispersal in abandoned Analyser)
Mississippi deltas are waves, tides, and - organic nitrogen (Kjedhall analysis)
wind-driven currents. Waves are - total bacterial biomass
responsible for eroding deltaic (epifluorescence direct counting)
sediments and for subsequent longshore living bacterial biomass (Zobell MPN
transport of sand-sized particles. counting)
Tidal currents are responsible for - heterotrophic potential (incorporation
generating tidal inlet sand bodies and of labelled ~lucose and Glutamic acid)
for the transport of fine-grained - Meiofauna (hand cores and counting of
sediments to protected lagoonal Nematoda, Harpacticoida, Copepoda .•. )
environments. Wind-driven circulation
accounts for seaward transport to The bacteria biomass data, from the
offshore sheet sands and landward living counts fits well with the total
transport to washover deposits during biomass (direct counting) and the
storms. variations agree with those of the
heterotrophic potential. We observe
dUring the year, six periods when the
number of bacteria is high. The highest
values are noticed dUring the cold
BACTERIA-MEIOFAUNA RELATIONSHIPS IN A months. They are close or greater than
SUBANTARTIC SAND BEACH (KERGUELEN) 10 8 bact.g- l sediment. The values
are lower in May and January (10 6 bact
F DE BOVEE, G CAHET, D DELILLE and J g-l) •
SOYER, (Laboratoire Arago, 66650
Banyuls-Sur-Mer, France). Meiofauna counts iluctuated widely (45 -
12500 .10 cm- 2 ). He observed seven
The previous studies made on the important increases of the fauna during
continental shelf, at a regional scale, the annual cycle. The values are medium
during oceanographical researches do not or low in summer (1000-45 .10 cm- 2 ).
exhibit any clear relations between the This observation can be correlated,
trophic supplies and the abundances of during this season, with a low
the meiofauna. The study of annual respiratory activity of the sediments,
cycles (organic matter, bacteria, the C.O.D. and the photosynthesis are
meiofauna) in the sublittoral area, important.
suggest that the different components
have different dynamics. An easily For each period the meiofauna and the
accessible beach was selected to study bacteria increase in the same time, but,
this. the microflora decrease when the
meiofauna reaches its highest numbers.
We studied a station on a clean fine This observation suggests that predation
sandy beach. The sample periodicity is can be an important fact.
 747

The annual production of the meiofauna FEATURES OF SOME SANDY ESTUARY SYSTEMS
and especially Nematoda is equal to or
higher than found in other littoral or J R GRINDLEY and Y VON SHIRNGING,
subtidal biotopes. (University of Cape Town, Rondebosch
7700, South Africa).

On the west coast of southern Africa

there are several sandy estuaries.
COMMUNITY STRUCTURE OF ICHTHYOPLANKTON Elsewhere most estuaries are character-
OFF SANDY BEACHES IN ALGOA BAY, SOUTH ized by fLiuddy substrata but the south-
AFRICA western region of Africa contains
exceptional volumes of sand and some of
L E BECKLEY, (Zoology Department, the estuaries are predominantly sandy.
University of Port Elizabeth, P.O. Box These estuaries have little rooted
1600, Port Elizabeth, 6000, South aquatic vegetation and no intertidal
Africa) . salt marshes. Despite the absence of
productive salt marsh vegetation some of
The ichthyoplankton community occurring these estuaries support a substantial
off sandy beaches in Algoa Bay, South biomass of fauna. Bigh faunal biomasses
Africa, is described from results of a require rich supplies of organic
two year survey. Quantitative samples detritus but endogenous sources appear
were collected off Sundays River Beach, to lJe restricted. However, the
Coega Beach and Kings Beach using a l,5m sUbstantial volumes of kelp (Ecklonia
diameter plankton net with a mesh size maxima) washed into these estuaries f.lay
of 500 urn. Larvae of numerous fish be important. The rale of kelp detri tus
species were found to occur in the in sub-tidal and littoral marine systems
plankton and the pelagic species has been well described so it is
Engraulis capensis, Etrumeus teres and possible that kelp might be a major
Sardinop ocellata were often numerically source of detritus in some of these west
dominant. various species of Gobiidae, coast estuaries.
Blenniidae, Sparidae, Sciaenidae and
Soleidae were also regularly encountered However, there are other possilJle
in the plankton. During the summer sources of detritus. Several of the
months ichthyoplankton was more abundant west coast estuaries have reed swamps in
than in winter and densities of up to the lower reaches of the rivers so
four fish larvae m- 3 were recorded. detritus derived from them might lJe
contributed to the estuaries. The main
reeds are Phragmites australis, Scirpus
littoralis, S. maritima and
capensis. Phytoplankton in the estuary
or washed in from the sea might also be
significant.

To obtain some understanding of the

748

functioning of these estuarine ecosystems Klaasjagers estuary may be largely

it is important to investigate the major dependant on allochthonous sources for
source of the organic detritus on which the detritus on which the ecosystem
they depend. Determination of the depends. 'l'he extent to which these
ratios of different carbon isotopes in observations apply to other sandy west
the detritus can provide information on coast estuaries has not yet been
this. This has been attempted in the determined.
Klaasjagers estuary in the Cape of Good
Hope Nature Reserve. This might /Je
regarded as a typical sandy west coast
estuary so it could provide information
applicable to other comparable systems. VARIATION IN SANDERLING FLOCK SIZE AND
STRUCTURE ON A SOUTHWEST CAPE SANDY BEACH
Four different detritus samples from the
KlaasJagers estuary have s13 C values of A CROWE, (Fitzpatrick Institute,
-17.44, -17.69, -17.95 and University of Cap~ Town, South Africa).
-19.40%0. The latter value is for
a sample including faunal fragments. Sanderling flock size data collected on
a southwest Cape sandy beach during
The relevant S13 C values of possible mid-summer I,lonths suggest that there is
primary producers are:- no optimal flock size for this species.
Phragmites australis -24.6 ° /00 Rather, a frequency plot of sanderling
~ capensis -26.4% 0 flock size data shows a clear bimodal
Scirpus maritima -26.82% 0 distribution, favouring very large and
Ecklonia maxima -12.0 to -15.07% 0 very small flock sizes. Also, contrary
porphyra capensis -14.9% 0 to optimal foraging theory, there is no
Gelidium pristoides -14.5°/ 00 obvious relationship between flock size
Ulva lactuca -13.8% 0 and invertebrate prey density. However,
if large flocks are analyzed separately,
The relatively high detritus values there appears to be an inverse relation
eliminate most of the low (i.e. more between flock size and prey density. In
negative) sources such as the various other words, larger flocks frequent
species of reeds. Kelp samples do areas with low prey density. This result
however, have high S13 C values and suggests that large flocks comprise
might be a source of this detritus. individuals searching for food, and that
Other algae frequently washed up with these flocks fragment into snlaller units
the kelp have similar values and might once patches of highly dense prey are
also be contributprs. Inshore phyto- discovered. Moreover, ~anderling flock
plankton !" ,'om the Benguela current, for cohesion varies with microhabitat.
which Prof J Field h~s determined a Birds foraging in wetter areas tend to
value of -17.9%°' is also a likely cluster, while those which forage in
source. drier areas tend to do so in loose
groups. It is suggested that flock
These findings suggest that the feeding facilitates the discovery bf
 749

patchily distributed subterranean prey They thus utilize the surf zones as
(Le. in wet sand), but is of little nursery areas. Both ovaries are
value when the primary prey dwells on functional and the litter size increases
the sand surface (i.e. in dry sand). with the total length of the female.
Catch per unit effort results from
angling data and tagging data showed an
offshore migration of adults during the
winter months.
THE IMPORTANCE OF NON-TELEOST FISHES
(ELASMOBRANCHS) IN THE SURF 60NE WITH
SPECI~L REFERENCE TO Rhinobatos annulatus

G J ROSSOUW, (Zoology Department, ARTIFICIAL SANDY BEACHES AND

University of Port Elizabeth, Box 1600, ENVIRONMENTAL IMPAC'l'S DUE '1.'0 DUMPING OF
Port Elizabeth, 6000, South Africa). COPPER MINE 'l'AILINGS AT CHANARAL AREA,
CHILE
The most important predator on benthos
in the surf zones off sandy beaches in J C CASTILLA, (Laboratorio de Zoologia,
Algoa Bay is the lesser sand shark, Departamento de Biologia, Pontificia
Rhinobatos annulatus. Other elasmobranch Universidad Catolica de Chile, Castilla
species of importance are the rays 114 - D, Santiago, Chile).
Dasyatis pastinacus, pteromylaeus
bovinus, Myliobatis aquila and the Untreated copper mine tailin~s have been
sharks Carcharhinus obscurus, dumped directly on to the shore at
Carcharhinus brachyura and Odontaspis Chanaral area, Chile, since 1938. Two
taurus. Sampling was done by organized dumping sites have recei ved mining waste
angling parties and beach seine netting materials. Chanaral Bay (29 0 02'
over a period of two years. Nutrition, S.Lat.; 710 30' H.Long) accumulated
growth ageing and reproduction were ca. 150 x 10 6 metric tons of
studied in R. annulatus. The major prey sediments between 1938 1974. 'l'he
species taken were the mysids, dumping bite was moved up in 1975 to
Gastrosaccus psammodytes, Mesopodopsis Caleta palito, about 8 km north of
slabberi and the sand mussel, Donax Chanaral Bay. Since then the actual
serra. Ring counts on the vertebral dumping site has received about 70 x
centra were used as material for 10 6 metric tons of sediments and ca.
age-determination and growth studies. 2.5 - 3 times as much fresh water (used
One ring is formed once a year in both for the treatment of raw mining mater-
sexes. Sexual maturity of this species ials for copper and molybdenum
is at the age of three years for males extraction). Furthermore, unknown
and four years for females and both quantities of chemicals used in the
sexes can reach an age of eight years. processing of both minerals have reached
R. annulatus reproduces once a year the dumping sites and surrounding
during the months November to February coastal areas.
when they move into shallow waters.
750

As a result of these activities some of samples were taken through a 0100m at

drastic geomorphological coastal changes three hour intervals over 24 hours. ~he

can be observed along the nearly 16 km data show that cells of A. birostratus
of shore actually affected by the were present in higher concentrations at
tailings both at the old and new dumping the surface than in the rest of the
sites. At Chanaral Bay an expansion of water column during the day. 1'hey
the original sandy beach of several disappeared almost entirely from the
hundred meters and a modification in surface and decreased in the rest of the
coastal contours due to the accum- shallow water column at night. Cells
mulation of tailing sediments are present in the surface layer of water
reported. The same phenomena are taking during the day 0ecome concentrated over
place at Caleta Palito and now arti- the rip currents of the surf system to
ficial sandy beaches are part of the form distinct, dark brown patches of
scenery. phytoplankton.

Sand beach macrofauna was monitored in preliminary estimates of primary pro-

1975 1976 and 1982 at four polluted ductivity per cell were Iilade and found
and unpolluted sites along 25 km in the to 0e higher in water samples with lower
Chanaral area. Species composition and concentrations of cells. productivity
zonation patterns are reported. Main per cell therefore, was higher outside
environmental impacts in the sandy than inside the 01ooms. These results
shore ecosystems of Chanaral area are are, however subject to further
discussed. experimentation. Bnvironmental data
such as wind direction and intensity
have 0een recorded for several extended
periods together with information on
bloom intensity. A positive linear
regression was found between 0100m
PRELIMINARY INVESTIGATION OF SURF ZONE counts and the average of the three
PHYTOPLANKTON BLOOMS OCCURRING OFF THE previous days wind conditions (wind
SUNDAYS RIVER BEACH IN ALGOA BAY intensity and direction). In this case
too, more data needs to be collected
D S SLOFF, (Department of Botany, before a final conclusion can be drawn.
University of Port Elizabeth, PO Box
1600, Port Elizabeth 6000, South Africa).

Surf zone phytoplankton blooms dominated

by the diatom Anaulus birostratus Grunow
have been recorded along the south coast
of South Africa. A series of samples
were taken through a bloom and the areas
adjacent to it at four depths during the
day (12hOO) and at two depths at night
(21hOO) . On a second occasion a ser ies
 751

FACTORS INFLUENCING THE DISCONTINUOUS this seasonal return migration the

DISTRIBUTION OF THE LUGHORM - Arenicola above-mentioned ecological factors
marina (L.) - ON THE BEACH OF THE NORTH continue to regulate the exact location
SEA (BELGIUM) of the patch where lugworms occur.

S CLAUS and A F DE BONT, (Kath. Uni v.

Leuven, Belgium).

Although the beach of the North Sea

along the Belgian coast is entirely ECOPHYSIOLOGYCAL ASPECTS OF THE GENUS
sandy, the distribution of lugworms is Donax 1. ENVIRONMENTAL FACTORS
discontinuous. The study of an area CORRELATED WITH ABSOLUTE AND RELATIVE
with isolated groups of lugworms shows a DENSITIES
relation between the occurence of worms
and several locally variable L NEUBERGER and G DE MAHIEU, (INTECMAR
characteristics of the ecotope. The Univ. Simon Bolivar, Apartado Postal
size of the sand grains, the water 80659, Caracas 108, Venezuela).
content of the sediment at low tide, the
hardness of the soil and its tixotrophic Donax denticulatus L. and Donax striatus
properties, and also, the presence of L. (sympatric species) were studied in
food-particles are the most important the central western Venezuela coast
factors. All these characteristics, (Boca de Aroa and 'l'ucacas) between July
except, to a certain extent, the last 1980 and August 1981. 'l'ucacas' beach
one, are interrelated. On the other slope and mean ,"rain size of the
hand, the most implicated properties of sediment were smaller than in Boca de
the lugworms, in relation to this patchy Aroa. The environmental factors studied
distribution, are: were temperature, salinity, organic
a) capability of the lugworms to breath carbon in the sediment and particulate
dissolved oxygen in water and gaseous organic carbon in the water. Salinity
oxygen as well, b) possibility to fluctuated throughout the year in
survive, for a certain time, in Tucacas between 32.2 0 / 00 (NOV.1980
completely deoxygenated water, and, c) and June 1981) and 37.2 0 / 00 (JUlY and
ability to dig a burrow easier in finer Sept. 1980) reaching the lowest value of
sand with a high water content than in 5.8 0 / in May 1981, which
00
dryer and coarser sand. corresponds with the rainy season. In
Boca de Aroa, it fluctuated between
A slight shift of shape and outline of 23.2 0 / 00 (June 1981) and
36.8 0 / 00
the occupied patch and of the density of (July 1980) reaching the lowest value of
worms in it was noticed between December 16.9 0 / 00 in April 1981 (rainy
and March: higher on the beach in season) . The sediment temperature was
December; nearer to the low-tide line measured throughout the year and the
in March. This is in accordance with minimum and maximum registered in
the known migration model of lugworms on Tucacas were higher (30.5 and 36.0 0 C)
the northeast coast of England. During than at Boca de Aroa (30.0 and
752

Tucacas also presents a ECOPHYSIOLOGYCAL ASPECTS OF 'I'HE GENUS

higher value of organic carbon in the Donax II. FILTRATION RATE IN Donax
sediment, between 0.167 and 0.246 %H&B denticulatus L.
(May and April,1981); at Boca de Aroa
the values varied between 0.083 and L NEUBERGER and G DE HAHIEU, (INTECHAR
0.113 %H&B (May and March, 1981). Boca Univ. Simon Bolivar, Apartado Postal
de Aroa present a particulated organic 80659, Caracas 108, Venezuela).
carbon in the water between 1.816 mgC/l
(Nov 1980) and 5.386 mgC/l (March 1981) The effect of salinity and temperature
except 15.156 mgC/l (August 1981), and on the filtration rate of D. denticulatus
Tucacas between 1.690 mgC/l (March 1981) collected in Boca de Aroa (central
and 4.220 mgC/l (May 1981) except 8.921 western Venezuelan coast) were determin-
mgC/l (June 1981). ated using an indirect method (Coughlan,
1969) employing the alga Thalassiosira
D. denticulatus is always found at both psuedonana. The filtration rate was
localities and in densities greater than measured in a total period of four hours
D. striatus. The latter is only found with determinations made each hour.
at Tucacas with the exception of a few
specimens that were found in Boca de The salinity and temperature modify the
Aroa. filtration rate, with the optimum at
25 0 / 00 at a constant temperature of
The absolute densities of D. 26°C. Increasing the temperature at a
denticulatus in Boca de Aroa give a constant salinity of 28 0 / , the
00
positive correlation (0.01%) with optimum was between 27 0 C and 29°C
salinity~ temperature, organic carbon in during the four hours of measurements.
the sediment and particulate organic
carbon in the water. In Tucacas, the re It was possible to establish
is a positive correlation (0.05%) with relationships between the optimum
all these factors except the temperature. filtration rate at different salinities
D. striatus in Tucacas gives only a and temperatures and with the
positive correlation (0.05%) with the distribution of Q. denticulatus near the
salinity. river. Nocturnal migrations were
observed at regions in the beach where
sediment temperatures were approximately
27 0 C and 29 0 C which coincides with
their optimal rate of filtration.
 753

THE GAMTOOS AN EXAMPLE OF BEACH/ fragile, being vulnerable to trampling,

ESTUARY INTERACTION damage caused by off-road vehicles and
other forms of human interference. If,
T J E HEINECKEN and A E F HEYDORN, therfore the equilibrium of the estuary
(Estuarine and Coastal Research Unit, and coastal dunefields is disrupted
National Research Institute for through undue disturbance of the natural
Oceanology, POBox 320, Stellenbosch, vegetation, the ecological (and therefore
7600, South Africa). economic) viability of the system could
be placed in jeopardy.
The Gamtoos River Estuary, which is
located on the Southern Cape Coast 50 km Careful planning tor the future use of
west of Port Elizabeth, forms part of the estuary and coastline is essential.
one of the major Cape river systems.
The river drains extensive parts of the
southern Karoo as well as four mountain
ranges and has a total catchment of
34 2
438 km • It is subject to extreme THE IMPLICATIONS OF RESOURCE
flooding and carries a very high silt PARTITIONING FOR 'I'HI:: STRUCTURE OF A
load which is deposited mainly in the SAND-BEACH MEIOFAUNA COMMUNITY
estuarine area, forming extensive
mudflats and saltmarshes. R /Ii HAR,IICK, ( Institute for ~'larine

Environmental Research, Prospect Place,

The estuary mouth is situated in a The Hoe, Plymouth, U.K.).
dynamic environment of bare transverse
coastal dunes and is highly susceptible It is now well established that the
to the combined effects of the nature of the food supply to sandy
predominant wind, wave and longshore beaches influences the uistribution of
current regimes. Under the influence of macrofauna species, the trophic
these forces the mouth of the Gamtoos organization of the macrofaunal
River tends to move eastwards but at community and t~e pattern of energy flow
times of flood it breaks through the through this component of the beach
transverse dunefields towards the west. fauna (Griffiths et al., this volume).
This is to be illustrated by a series of By contrast, the "black box" treatment
aerial photographs taken before, during that meiofauna usually suffer fails to
and after flooding. recognise the varied trophic components
(bacterial-carnivores etc.), with the
The estuary with its good fishing in the result that physico-chemical parameters
river and along the coastline, its such as grain size, oxygen concentration
plentiful supply of bait and numerous and salinity are almost invariably
other attractive environmental features, invoked to explain species distr ibution
combine to make it an ideal area for patterns and community structure. This
recreational development. However, it paper sets out to show that food supply
has to be borne in mind that the is also a most important factor in deter-
saltmarsh and dune vegetation is very mining many properties of community
754

structure and distribution in sandy Fine scale vertical distribution of

beach meiofauna, and that without a meiofauna in the sand-flat is also shown
sound knowledge of the trophic to be controlled by trophic special-
organisation of the community it is isations of the component species.
usually impossible to interpret the Distinct spatial separation of the
observed patterns. harpacticoid copeopd species on a
millimeter scale is indicative of
partitioning total energy flow between competitive interactions among yeneral-
the trophic components of the meiofauna ist feeders with broadly similar food
of a sand-flat at Exmouth (Devon, U.K.) requirements. Nemtodes, on the other
shows an overwhelming dominance of the hand, IJave a much less well defined
predator/omnivore group, a significant separation of species with depth: the
contribution by epigrowth (mainly range in form and size of their feeding
diatom) feeders but the virtual absence structures permits a high degree of
of a deposit feeding element. trophic specialization and the avoid-
Comparable data for an intertidal ance of interspecific competition. 'l'WO

mud-flat show that diatom feeders species of nematodes which have feeding
dominate, with predators of least structures almost identical in form and
importance. These distributions clearly size do, however, have a distinct depth
reflect the varying proportions of separation from one another.
different types of food resource
available in the two types of sediment.
In both communities resource
partitioning appears to occur only on
the basis of prey size in the predators/
omnivores, with a consequent restriction DUNE: A GEOBOTANICAL MODEL OF FOREDUNE
in the number of co-existing species. DEVELOPMEN'I'
In other trophic groups, especially the
assemblage of species feeding on D J DISRAELI, (Department of Geography,
unicellular plants, resource University of California, Santa Barbara,
partitioning is much more complex, California 9310G).
allowing a much larger number of species
to co-exist. Species importance-rank The purpose of this investigation is to
curves are strongly concave in the illucidate the rOle of Ammophila
sand-flat which is dominated by a few breviligulata in the tormation of fore-
predators/omnivores, but much less so in dunes in the Northeast United states.
the mud-flat where non-predatory species Earlier investigations by the author
apportion resources much more evenly. have sIJown the ability to model the
Thus, increased dominance by a few dynamic interaction between plant growth
species in the sand-flat is a natural and sand transport. Six simulations
consequence of the trophic organization were run with varying conditions of wind
of the community and the different scope speed and drift-line establishment tor a
for diversification with broad trophic period of 14 years. The results of the
groups. simUlations indiciate that one study
 755

site is in a juvenile phase, while the SOME ASPECTS OF DRIFTSAND RECLAMATION IN

other is more mature. The reason for THE REPUBLIC OF SOUTH AFRICA
this comes from the fact that the
equilibrium shape (steep windward face P REYNEKE and M BURNS, (East Cape Forest
and shallow leeward side) is similar to Region, Forestry Department, Department
the second site, even with starting of Environmental Affairs and Fisheries,
conditions from the first site. It was South Africa).
also concluded that the juvenile form
was most affected by the wind power, The policy towards driftsand management
whi Ie the older form was not. Only the in South Africa has changed a lot during
dune crest height of the older shape was recent years. Indiscriminate reclam-
affected by the wind power. ation of <.lriftsand areas is no longer
carried out and the tendency is to
The model has been validated for bi- re-establish indigenous species in
weekly increments by comparing the preference to exotics.
predicted measurements to the second
site, which was not included in the The reclamation operation is divided
original calculations and from long term into two main activities referred to as
measurements of Ammophila foredunes. temporary and permanent stabilization.
There are only two observable problems
with the model: (1) an annual over- Temporary stabilization involves the
prediction of 15 cm of sand deposition halting of sand movement by packing the
over 25% of the dune and (2) a slower open sand area with branches, reeds or
than observed windward migration of the baled ~rass. ~his is followed by the
dune front. The former is explained by establishment of a permanent vegetation
the lack of precipitation measurements, cover throu~h in situ sowing of suitable
a disregard for leaf flexibility, and no indigenous seed and planting a limited
means for calculating erosion. The number of seedlings raised in the
latter is explained by a lack of drift- nursery.
line measurements. DUNE can be
succesfully applied to any Northeast An attempt is made to initiate and speed
foredune that faces the onshore winds up secondary succession towards the
and with a grain size of .75 mm. climax communities occuring naturally in
the area by following the stages which
presumably occur. Grasses and shrubs
are established initially and the woody
elements are introduced at a later stage.

A considerable effort is being made to

eradicate exotics have become
established on driftsand areas. Species
such as Casuarina equisetifolia and
various eucalypts do not pose a big
problem. Acacia cyclops and A.
756

cyanophylla, hO\lever, are very difficult ranean. Rainfall occurs only in winter
to eradicate but methods have been (November-April) and the summer is hot.
developed to replace them successfully The average annual rainfall is 600 mm.
with indigenous species. The average air humidity is 67%. The
mean daily temperature is 19 0 C, \lith a
igdression of 7 o C. The extreme
maximum temperature, determined for 65
years, was 22.7 o C in January and
THE SANDY HAIFA ACRE PLAIN AS A TYPICAL 38.4 o C in Hay. The tJrevailing \lind is
PHYTO-ECOSYSTEM OF ISRAEL'S COAST south-west and west with an average
velocity of 1.1 2.5 Beaufort. Five
D E TSURIELL, (Israel). regions may De distinguished (from \lest
to east):
The Haifa Acre Plain is an area of 1) The sea-shore exempt of any
intensive industrial and urban vegetation, Decause of high salinity and
development, necessitating works of frequent sea inundation.
maintenance for stilling the shifting 2) A narrow belt characterized by two
sand-dunes which cover the area, as well dominant plant species: Sal sola kali and
as to protect the coast against sea lpomaea stolonifera accompanied by the
invasion and against new sand inundation following species: Cakile maritima,
originating most probably from the Nile Euphorbia paralias, Sporobolus arenarius
Delta and transported by the currents of and Pancratium maritimum. All these
the Medi terannean to the North Coast of species are resistant to salt spray,
Israel. most of them are succulent. Some of
them may penetrate into the posterior
The sub-surface geological structure of belts.
the plain shows a trough in which 3) The third belt comprises tvlO dominant
formations of clays and silts species: Sporobulus arenarius and Lotus
alternating with sandy marine formations creticus, accompanied by Agropyrum
have been deposited. Four marine junceum, Diotis candidissima, Oenothers
formations may be distinguished and may drummondi, Silene succulenta, Senecio
be attributed to sea transgression, joppensis, Concolculus secundus and
whereas the unconsolidated upper marine Crucianella maritima.
layer is attributed to the Flandrian 4) The fourth belt comprises Ammophila
transgression. The sedimentary data, arenaria and Cyperus conglomeratus. The
confirmed by pollen analyses, indicate first one covers most of this area.
that the climate has not undergone They are accompanied DY Tamarix gallica,
marked variations during the Pleistocene Ipomea stonolifera, Atractylis flava,
age and that it differed little from the Juncus acutus, Scirpus holoschoenus and
present. The mobile actual dunes occur Saccharum agyptiaceum.
without interruptions along the whole 5) 'L'he last belt is larger than any of
Coast between Haifa and Acre, the width the others, 200 - 300 m wide and mainly
not exceeding 2 km, the length being consisting of active naked dunes and
appro 15 km. The climate is mediter- sparsely covered by vegetation. In the
 757

rear of the belt, there is a rich species planted in each of these three
vegetation cover consisting of Artemisia locations. Most of them have to Le
monosperma, cyperus mucronatus, Retama chosen among the natural plant cover
roetam, polygonum equisetiforme, occurring in the ecosystem, the
Chrysanthemum viscosum, Echium preferred species are those which are
augustofolium, Neurada procumbens and most resistant to salt spray, bigh wind
Medicoga marina. All five belts appear velocities and to temporary sea
approximately in a parallel order. inundation. All of them must be able to
accumulate sand.
Two main problems exist:
I) The area being a valley fronting a The main species suggested for this
menacing sea which ascends slowly, may objective are: (a) for tbe building of
be inundated by the sea water and the foredune: Amrnopbila arenaria,
partially or totally eroded by it. Agropyrum junceum, Lotus creticus and
2) The existing sand-dunes inside the Saccharum biflorum. (b) for the area
area, most of them still being active, behind the foredune: Artemisia
disturb industrial development and monosperma, Retama roetam, Calligonum
traffic. This problem is aggravated by comosum (introduced from the Negev)
the fact that considerable amounts of Oenotbera drummondi
new sand masses are ejected from the sea,
increasing the activity of the shifting Fences are being used in order to
sand dunes. accelerate tbe building of the foredune
and to obtain any required height of it,
Preliminary protection works on according to the data collected by the
relatively large scale under the Corps of Lngineers (Army of the United
specific climatic and geo-morphological States of America).
conditions of this ecosystem, have
shown that the building of artificial
"sand-dunes" made by plant cover,
parallel to the sea coasts, seems to be
most efficient means not only for main-
taining the shore against sea
penetration, but also to decrease the
shifting of the sand particles inland.

The techniques of building such

foredunes are based upon the French
model used in the landes (southwest of
France), a protecting dam whose windward
side is of a moderate slope, with a
large plateau in the middle and a steep
slope on the leeward side. 1'he
different degrees of slope may be
obtained by the right CJ;hoice of plant