Documente Academic
Documente Profesional
Documente Cultură
Archaeometry 98
Proceedings of the 31st Symposium
Budapest, April 26 - May 3 1998
Volume I
Edited by
Erzsebet Jerem
Katalin T. Biro
ARCHEOMETRIE
Collectifs
ARCHAEOLINGUA
Archaeometry 98
Proceedings of the 3 lst Symposium
Budapest, April 26 May 3 1998
-
Volume I
Edited by
Erzsebet Jerem
Katalin T. Bir6
Edina Rudner
T5C
l Jo::i-1/-1
\
\
BAR International Series 1043 (I)
2002
\
Archaeometry 98
Proceedings of the 31st Symposium, Budapest, 1998
ThiS title Pbl
ll
{lcl. b
ArchaeaP:t Y
pablishers ss sponsored by
Gordon Bol.{ l.Jf }3 .
276 Banbu ry R.1:1
8 l'Itish Are haeolo
gical Reports
oxford i:l.q
E
oX2 7 D
En gland
www.archaeop
rs
s
l;allt
BARS1043
A r c haeolinPi.
O 'f.a \::
l1.tta.1 Europ
ean Series 1
Arcltaeometry 9
&: J.:.>r
Oceedz Hungarian National Museum
n s
g of the 31st Symposium, Budapest, April 26 May 3 1998. Volume J
t he indivir1
'-' \i
-
q_ l
qH.
....,t, o r s 2
002
ISBN 1841711<
ISBN 1841711< l 15
.
Enl
(
ti c::o.rnP1ete set of tw vol
h (t i
s Volum
o umes) {r
Pri.!lted m e)
C\
cl b
y The Ba .
n1
Word process:i.n singstoke Press
Im.re Szabo, Q . cl
E nglish text rYlq_ ;skt op ed t ng
\1
Kova, c I i and layout: Andras Kardos with contributions of
.
o
cover illustr a0
.
Ministry of Cultural Heritage
F. pulszky 188\3 :t\: <::::o
' .f:t-IJ PPer Age t ooI k .
"Th i t from Szeged-Sziller ca. 3rct rnillenium BC after
e Copp er A
ge in Hungary"
d. r
. .
lan Books
.
EDITORIAL PREFACE
we are honoured to present to the interested reader the Proceedings of the 31st International Symposium on
Archaeometiy, held in Budapest between April 26 and May 3, 1998.
ARCHAEOLINGUA Bringing the conference to Budapest was a major achievement. The possibility of inviting the meeting was raised first
during the Heidelberg meeting (1990). Although Hungarian archaeometiical research did not have at that time, and still
Central European Series does not have, a firm institutional background, the largely informal circle of scientists dealing with the broad field of
archaeometry was enthusiastic about the idea and ready to bring the meeting to Hungmy by 1994. Due to the significant
social and political changes of the early nineties, affecting also the scientific institutional network, this was not possible
at that time. In the course of the 1994 Ankara meeting and the 1996 Urbana/Champagne conference, however, we took
up the challenge and shouldered the task. The Hungarian National Museum agreed to undertake the role of main
organiser providing the infrastructure and institutional background for the conference. The Hungarian Academy of
Sciences provided the conference with a magnificent setting in the Ceremonial Room of the Headqumters of the
: ! hre
yea.rs ago, t e edi.tos of Archaeopress nd Archaeolingua first considered launching a
Academy. Several scientific institutions of the Academy - among them, in the first place, the Archaeological Institute
of the HAS - played an essential patt in organising the meeting. We also received impo1tant help from scientific
0 .J intly published series within BAR, under the
. title of Archaeolingua Central European Series institutions and associations working in different fields of archaeological science, which were mentioned in the
u
ea was o provide n w multidiscipl inary
forum for monographs and/or edited volumes
ope Abstracts volume published in 1998. A large Local Organising Committee was formed, partly to help the daily tasks of
to: chaeolog1sts and sc1ent1sts . _
involved in related studies. The series sees its beginning organisation and, even more impmtant, to disseminate information about the multidisciplinmy event among scientists,
.
present publicat . in the
ion . archaeologists, museum professionals and many other interested people. The number of registered pmticipants for the
Conference was 260, including 60 Hungarian members. As special arrangements were made for Hungarian students and
The Senes is
to have the fullest possible archaeological remit daily visitors, the number of persons who were able to benefit from the meeting can be estimated at twice the number of
.
period, g
ograph1ca ly 1t will focus on Central Europe.
This area is a very special entity both in
registered participants.
geo raph1cal
f and histor.ical terms, and is distinct in many
ways from northern and southern It is not by chance that we try to point out the benefit to Hungarian research of the Archaeometry Symposium. Our main
==
rn nd estern countries In Ce
.
tral Europe are integrated a huge range of influen
ces fro purpose in organising the conference was to promote related studies in Hungary, to convince scientists as well as
?
e a 1rect1ons which .
lead to a multicoloured and very specific cultura
l picture. The present-day archaeologists that it is worthwhile collaborating with each other, to show students a possible avenue of research and
geo ohtical
effe s Pos1t10
y
roc sses of 'New Europe' are also showing
these complicated and long-lasting also to demonstrate the possibilities and strength of an interdisciplinary approach to the management of museums and
.
caref I evaluation
in the heart of the contin nt lends both advan
tage and disadvantage and needs research centres. The efforts proved to be clearly successful, as reflected by papers presented to the conference (many
throughout the whole history of the past ten
related to one thousand years . All contributions of them written in the framework of international collaboration projects), the initiation of archaeometiy-orientated
or other aspects of the above subject will be workshops and seminars at different universities in Hungary, and last but not least, the high attendance of Hungarians at
welcome.
the ISA Symposia in Mexico and Amsterdam.
; ?
h eometry 98 is the inaugural title in the Archa
eolingua Central European Series This first title
its two large and impor
A chaeometr
!
tan volumes is based on papers presented
at the 31 t Internationa l
There are, however, still many problems as yet unsolved in Hungary, such as the lack of a proper institutional
background and the need for a standard periodical dedicated to questions of archaeometry. The minds of traditional
y Conference held 1n Budapest.
archaeologists/scientists, though ce1tainly softened by the gleaming perspectives, are changing slowly. It is clear that
the inclusion of costly analyses in the modest budgets of excavations without the hope of direct applicability cannot be
expected within such a short time. The building up of reference collections and the working out of methodology is a
time-consuming process. Neve1theless, we are confident that the 31st International Symposium in Archaeomet1y,
organised in Budapest, was a very important step in this direction.
Similarly, bringing the Conference to Budapest was a good oppmtunity for many Cential and Eastern European
countries to get involved in archaeometiical studies in general. The organisers made special effo1ts to support scientists
and students facing financial hardship with grants. The supporting institutions offered accommodation in guestrooms as
well as fmancial help which is greatly appreciated by the organisers. The Budapest meeting attracted, naturally enough,
people from Cential and Eastern Europe in large numbers as well as from countiies where archaeometry is tiaditionally
stiong. As we have observed, the venues of past and oncoming ISA meetings were represented by large number of
participants: Great Britain, as home of the idea, Greece, Gennany, Turkey and from the New World, the United States,
Canada and Mexico (Fig. 1, Fig. 2).
2 3
Participants by country (1 ), 31 ISA Budapest Main topics of 31 ISA, Budapest
6
17
[]Hungary
[] 3 field archaeology
ISlltaly
Ell 4a tech/prov metals
i;::J France
llD4b tech/prov ceramics/glass
CJ Greece
32 4c tech/prov stone/pigment/plaster
BTurkey
8 5 theme session experimental ..
D Austria
Ogeneral
!II Canada
IZIMexico
Others
20
Conference '98
h the proceedings of the Archaeometry
of papers, the editors decided to publis
Due to the great number
Participants by country (2), 31 ISA Budapest
ins the writte n versio ns of the prese ntations and
I conta
the structure of the meeting. Volume
in two volumes, following consists of the
5, and 6. Consequently Volume II
posters presented in sections 1, 2, 3,
the texts accompanying the
in section 4 a-c.
studies and material analyses presented
papers discussing provenance
Belgium
D Israel
first place,
onsuming technical difficulties. In the
the editors had to face several time-c
II Romania
ication.
related to the layout of the publ
r1China
5
4
TABLE OF CONTENTS
VOLUME I
1. Biomaterials
S.
BERD NIKOV, s L.) bones from excavations
analysis ofelk (Alces alce . .. .. . .. .. . .. . . .. . . . .. ... . 15
An osteological . .. .. .. . .. . .. ... .. .. ... .. .
.. .. .. ... .. .
.. .. . .
.. ... .. .. ...
.. .. . ..
a
.. .....
Latvi
..... ...
, Eastern
. .. ..
.....
Basin
.. .. ...
a
.. ..
Lake
.
na
. ... . .
..
in the Lub
.. . .. ..
....
S. & HERRMAN,N B.
SKOPF, B., HUMMEL, ...................................................
.......................... .............. 19
BURGER, J., GROS ......... ..............................................
lmiques in archa eome try ..... ........
DNA tec
& BARTOSIEWICZ, L.
KISS, ZS., CSAPO JR., J. ........................................................
......................... 23
CSAPO, J., CSAPO- 11th Dynasty Egypt......................
sis of bovin e soft tissu e from
Tue chemical analy
A.
A. M. S. & CHAMBERLAIN,
NIKIN, D. E., COPLEY, M. S., AHMED,
GERNAEY, A. M., MIN reported tuberculosis frequ ency sugg ests
mycolic acid prevalence and .. . . . . ... .. ... .. 27
A correlation between .. . .. .. .. . .. ... ... . .. . . .
.. . .. . . .. . . .. . . .
.. ... .. ... .....
..
.....
s
.....
lation
..... ...
popu
..
nt
.. ...
ancie
..
rty in
.. ....... ..
or pove
....
f
.
'
. . .. ...
'
..
arker
.. ..
biom
.. .. .
a
.. ..
DLETON, W. D.
G., SCHROTER, P. & MID er ................................................
.............................. 33
PRICE, T. D., GRUPE,
of huma n bone and tooth enamel: the European Bell Beak
Strontium isotope studies ,
O, G. 0. & ERICK REBORA H.
QUEVEDO LARA, T., CORONAD
QUEVEDO ROBLES, P., al bone s
ofmetals in hmnan and anim ...................................... .41
A preliminary determination ........................................................
locen e perio d in two regions of Mexico ........................
from Pleistocene-Ho
popu
.
c
......
spani
.. ..
Prehi
.. ...
can
. . ..
Mexi
...
in
..
A study ofmigration
..
A, G.
L., VALADEZ, R.,& ZARAZU ...................... 59
TEJEDA, S., MANZANILLA, n site ................................................
ns of estron tium and zinc in animal bones from the Teotihuaca
Paleodietary applicatio
, E. 0. & MARTINEZ, R. D.
VARGAS-SANDERS, R., DIAZ
Identification ofancie nt prote ins from a ceramic sculpture .......................................................
63
ess ................................................
ictlan tecuh tli at the Temp lo Mayor, Mexico - News and progr
ofM
2. Dating
try? . .. .. .
.. ...... .. ...
.. . .. ..
DAYTON, J. E. ................
................
........... 83
................
3. Field Archaeology
9
ERGENE
An eth\ \l}..i
an d th \)!I t 'tl
e l\) l:!l1 .
GARRlSG l)ll. o1og
e ical comparison: the Kerkenes archaeological survey WOLF, S.
:a: :
<\ (j
f erkenes city and the Keykavu
s castle ..........................................................................................................
.................113
.
Questions, answers and limitations: chemical, mineralogical and petrographic d1stmct10n
. . . .
RAi<:i ;: :
PAVLIS
. \'l:!\ <l \\i:t<:l
(
8{
g
:
ts near archaeological sites as a key to the past landscape evolut10n under human
d fr m the Kraichgau Hills in southern Germany .............................................
impact.
.
ofaeri l ..............................................131
s s , A., l i;: l\ I ' ?EORGILA, K., KOKKINOU, E., KARIMALI, E. & MANTZOURANI, E.
.,
GE
ca
1 gn at the Neolithic settlement ofKandou- Kouphovounos (C rus)
yp ......................................................................145
VOLUME II
The Isl S, Qi U, L., GKIOUROU, A., KARATHANASIS, CH., STEFOULI, M.,
q <\ l\ "KARIMALI, E., GIANNOULI, E., GAVALAS, G., ASVESTAS, G.,
SARRIS 'A \)f
4.a. Pr ovenance metals
A :' lOANNou' D & VRONTAKI E.
MAN \ V A "...'-lll.0rgos: micro-scale & macro-scale
.
'
. Al:<\" remote sensing ......................................................... ..
Ancient "\j '12
lDis, A., MERTIKAS, ST., GUY, M., VRONTAKI, E., ............................................
.149 .
ADRIAENS, A., EARL, B., OZBAL, H. & YENER, K. A.
SCURTU ' \. ct \) & KALP AXIS, TH. Tin Bronze Metallurgy in Transformation : Analytical Investigation ofCrucible Fragments
The geop t rirnoupolis, Lasithi): an archaeological site as a remote sensing laboratory............... from Tell al-Judaidah, Amuq (Turkey) Dating to circa 3000--2900 BC......................................................................................................273
.. l .....................................................157
SUMEGI, p i BALASSONE, G., BONI, M., DI MAIO, G., FARIELLO, M. & VILLA, I. M.
EllVlfO
at the
., i:l}:
i ,
tl\ :a1 s::::: ;
I
illa
E
oman town Porohssum m Dacia ................................................................
.
.
..................... l65
Analysis ofthe rings found in the Samnitic necropolis of Saticula (Sant'Agata Dei Goti, Benevento, Italy) .............................................279
BENVENUTI, M., STRILLOZZI, B., MASCARO, I., COSTAGLIOLA, P.,
ange and human adaptation in the Carpathian Basin
<l1:: if ROMUALDI, A. & TA.NELLI, G.
1Postg lacial transition.................................................................................. Tin-rich slags from a 6 th-4th century BC Etruscan settlement at Populonia (Tuscany, Italy) ....................................................................285
.....................................................................171
BEzUR, A.
What's beneath the surface? Metal working techniques in the Upper Mantaro Valley, Peru .................................................................... 291 .
Analysis ofsome Ancient and Medieval coins by neutron activation and X-ray fluorescence ...................................................................311
1: ;;,t
o
0 )J ing ofcopper and bronzes in sand moulds .......................................................
...............................................................185 CZAJLIK, Z. & G. SOLYMOS, K.
e
.\' ' 'flERI, S., & SANTORO, S. Analyses ofingots from Transdanubia and adjacent areas ........................................................................................................................ 317
GYULAI t he firing atmosphere in the
. "' ' F.
..
'O f production
Bwwstory. DILLMANN, P., FLUZIN, P. & CHEVALLIER, P.
Cs With ca1c1'te and chamotte mc1 us1ons ...............................................................
................................ .............................191 Application ofsynchrotron microdiffraction coupled with X ray fluorescence microprobe for the determination
IOVINO, l\' at 11.
ofthe iron making processes by identification ofinclusion in archaeological artefacts ..............................................................................327
Processiu
Szarvasgede (Hungary): a case 1cor b 10 ' rs1ty c onservat'1on .............................................
' d1ve ............. ............................. l97 ENIOSOVA, N.
MAR TON, l E"'
:'11!ilt h obsidian tools: the micro-wears ...................................................................
.................................................. ............203
Casting moulds from the Upper Dnieper Region (10--1 l th centuries AD) ..................................................................................................335
EREMIN, K., GRAHAM-CAMPBELL, J. & WILTHEW, P.
t' vl{_yArn' T Analysis ofcopper-alloy artefacts from pagan Norse graves in Scotland ...................................................................................................343
, , L, 'll
PAVLISH ._ % tect .
PA . . . .
..............................................................
. . . .
...........................
. ........ .................
...... . .... . ..
al< The Persian and Macedonian gold from Croesus to Alexander the Great.. .................................................................................................369
Flume e)(p :'.'\., P. .
ge e xpenmen ts ...........................................................................................................
!7':'.::t
' a ..................................................225 OOMORI, J. & TOROK, B.
PAVLISH, L. ""' ''l.BlNDIE NST , M. R. & SHE
PPARD, P. J. Technical examination ofthe Early Medieval ferreous metallurgical finds from Hungarian sites ..............................................................375
0 ZJ ond bone ..................................................................................................
...................................................231 INGO, G. M., DE CARO, T., ANGELINI, E., BULTRINI, G., CHIOZZINI, G. & MICHELETTO, E.
P
Copper plating on an iron artefact from the early medieval site ofPeveragno (Piemonte, north-western Italy) ......................................... 383
l.' brea kage from bow-driven arrowpoints in
VO=
Experimen"
two white tailed deer carcasses ...................................................
....................239 INGO, G. M., DE CARO, T., TUCCI, P., BULTRINI, G. & CHIOZZINI, G.
Early iron artefacts and slags from Tell-Afis (north-western Syria):
}.. SU.It s from the Sza'zha1ombatta
' Archaeolog1cal Park......................................................................... microchemical studies for locating the iron ore source ...............................................................................................................................389
.................................243
: J.
'M. & MAGGETTI, M. KASZTOVSZKY, ZS., REVAY, ZS., BELGYA, T., FAZEKAS, B.,
g ofclays using salt water ............................................................................
..................................................................249 OSTOR, J., MOLAR; G., VADAY, A. & FIGLER, A.
.
Prompt-gamma activation analysis ofRoman brooches ............................................................................................................................ 399 .
10 11
KOLTO, L.,K. \>
Analysis of h_i
i \
XRF analys is% \j"\
LINKE,R.,
ll
SCJ..\f :I.\1,1.\ % . llti.y
(Q: MACL
t L., v AN DER WESTHUIZEN,w. A. & THUYSEN,J. oos
\Ill }.1.' '. \ll ti. concentrat10
Non-destmct\ EAN, P.
. JACOBSON,
LIVERSAGE, D
lily ,isY bronzes ..........................................................................................................................................
.....................585
n finds with various method material for inter- and intra-laboratory calibration ............................
z l\1l\
s sann certified reference
4' ceramic- I: a new pottery
G. V. & BEUKENS, R. P.
69
\ )\ S
An industry ................... 405
PAVLISH,L. A., FARQUHAR, R. M., HANCOCK, R.
H. & ALRAM, M. KOCH,C. P.,
JNAA ofpottery from ll Lokeridede and Jarigole, Koobi Fora
" W1N1ER, ......587
OZBAL H AD Region,Kenya ....................................................................................
. , ., h_'I 1 /lh." or
'l'llcl<A. .
A.ustr ian Medieval .
MCCRAY,W. P. &
Minor metal!\ '\ sliver coins by energy dispersive X-ray fluorescence ..................................................
0
E.
...4 9 WARREN, A. K.
indications of
ll llp ,
l\l\ ll <\.Per alloy impurity patterns near the end ofthe Hungarian Bronze Age .............................................................4 Strangers in a strange land? - Renaissance glassmaking in Amsterdam ............................
......................................................................... 593
PROHASZKA,
&
,R.
MISONNE,B.,LADURON, D.,BRULET
Technical asp . ti l.\lltsBA.RL, B.
17
GEDIK, B. h ofLate Samian ware from Argonne,
t l\r Provenance study and technolo gical approac
REHREN, TII. s
asso 599
NW Gaul and Burgundy
ciat ed with Anatolian copper and bronze artifacts
llln of
.....
..........................................
The developU\ ll \\
MOMMSEN, H.,ANDRIKOU,E.,ARA
polymetallic ores ..........
..........................................
..........................................
...........................................................................................................................
..433
RYNDI NA, N. V Ill\('.\'l:N
\Ji.!;)!(
Ile
pottery
. Neutron activation analysis results ofBronze Age
a '. ................... 607
Eneolithic c st jars ofThebes .....................................................................................
lllirrors from Southern Italy ....................................... . . . . .... . . . . . ... .. ..... .... .. ...... ... ... ... ..... . 439 from Boeotia including ten Linear B inscribed stirrup
' \>,. l I<.. ITER, S. & KENYON,I.
,
ZS., 1' ''
. . . . ...... . . .... . . . . . . . .. . . .. ... .. .
&,.,
Ashuapm
.
at the
. ..
RE TSI S,
. ...
L.,
.... . . . ..... ..... . .
. .. .. . ... . . .. . . . .. . .
. . . .. .... . . .
.. . . . . . .. . . .... . . .. . ... . ..
llt,t l\ iv
l'.l;; ) 'li)t e analys s ofmedieval coins
The metallurgj ONGYOS SY, M. PALAGUTA, I.
: :
.....................627
SERTOK, K., S "' l ' gy dis
techniques ....................................................................................
from the reign Some results ofstudies on Cucuteni-Tripolye decoration
HANCOCK, R. G. V. & B'NAITY, J.
475
... .
1-Q.v
Of Copper Ore :!'\:\
. . . . . . . .. ..... .
TI,G.
Macroscopical,mineralogical and chemical characterization
"'"""'' b.
b s the Carpathian Basin ........................................................................... 481
&
Jura, Switzerland:
W e
ofpottery from an Early La Tene settlement at Alie (Noir Bois),
;li)ll :
a
. "-. "" e analysis as a tool for the identification
:; : s tion of 'low-' and 'high-tec h' methods ...............................................................................................
...........64 1
,r <::. S ites o
An example for the combina
'
W. T.
TULUN, T., UZGIL,E., GULE<;:, A., TEKIN, A.,BATUR, A.,
laye s ERDEM,A.,KAHYA,Y.,
A.,ALTUN, A.,ARLI, B.,
r on Qucun bronzes ...........................................................................
......................................................497
, G.,SAGLAME R, G., TANYEL I, G., TUTENG IL,
ARLI, H., <;:OBANOGLU, V. & EREL,C.
KOKSAL
DE LA FUENTE
A multidisciplinary study of ancient :iznik ceramics .. .. .. . . .
.... .. .. . . 655
.. ...... . .. . . . . . . . ... . .. ... . . .. . .. .. ... ... ... .. . . . . .
f,
. . .... .
Application o
. .. . . ... . . ... ..
in archaeolog ic {\
DEMIRCI, S., C \ Origin ofthe rosette-stamped storage jars from ancient Judah: ...667
P
..................................................................................................
Technologica l l \' '-l on preliminary results ofthe neutron activation analysis ............................
&
,.. ,
S, <:>lCalan :ixaly .
DURUZ,M. fvL\ \)
1s
. ' (provenance)
Mde
Cs:
penmental approach ...................................
An Early e i . \>t<\t ............................................................................................ 501
4.c. Provenance stones
G & B..
l'
EGGERT & \;.._ ' '\
'),._ II<. , E. N.,
Of s0111
TURKME NOGLU, A
.
OKE, H.
ft, e Medieval glazed pottery in Anatolia .........
M
x ATTANASIO,D., PLATANIA, R., ARMIENTO,D. & EMANUELE,
o
M L&. '\
.,. ."'t 'tty .
. . ... ..
M. C.
Q
EILAND
Infr ;ed :;rkshop from the 9th century at Reinach, Switzerland ................................................................. ..519 A New Electron Spin Resonanc e Marble Database : General features
... 673
c to selected cases ....................................................................................
GARCIA-HERAS \f- tl\i and preliminary application ofstatistical discriminant analysis
..............
H
Green - Medieval lead glass vessels la Heraclius a ,F. & SCORZELLI, R. B.
BELLOT-GURLET, L., DORIGHEL,0., POUPEAU,G.,KELLER
'
New developm
, :.,.. t >obQ.
525
by tot! "'""'\ '
'
................677
..................................................................................................
GASSNER, V. S
aeolog.cal ceramics
& 'tQ.e , 1 ....................................................................................................................................... and Ecuadorian sources using ICP-AES and ICP-MS ..............
BENEA, M.,MULLER, H. W. & SCHWAIGHOFER, B.
531
R. & TORNERO , J D
-
Archaeo metrical\
ANnEZ RUIZ,
)' '
a . . . ... . .685
.. ..... . . ......... .. .. .. .. . . . .. . . . .. .. .. . . . . .. .. . .
GEBHARD, R., GD''\-
na1ysis ofarchae
The single Roman marble quarry in Romania . . .. . . .. .. . . .
\\->' fluoresce
ological ceramics
. .
RIEDERER,J.,
nce (TXRF) ..........................................................................................
\"I. , .................................................54 1 D'AMICO, C.,BERNABO', M. B.,BIAGI, P., PEDROTTI, A. L.,
Miissbauerstudy te '. Archaeometrical analyses ofpolished stone tools ..............691
..................................................................................................
GRATUZE, B., Dz
l'lsation and provenance studies from the Neolithic to the Bronze Age in Northern Italy ..............
,C.,BORODI,G.,FIAT,T. & DARABAN,L.
!Ctn
on pottery found at Velia (Southern Italy) .......................................................
(_)..,,, :u_,ER,E. , HAUSLER, W .,
547
tZ K
A study ofHung DARABAN, L.,COCIU,L., ZNAMIROVSCID,V.,COSMA
& WAGNER, U.
........697
WAGR,
and semiprecious stones......................................................................
' ltic ott' Some physical control methods for the study of the precious
GIURCA,I.,SALAGEAN, M. PINTILIC A, A.
.. . ..... .... . . .. .. .. . . . . . . ... .. .... ... ..... ....... .. ... . ... ... . . . . .. . .. 705
a glass composition Obsidian provenance studies . .... .. . . .. . . .. .. .. . . .. .....
. .. . .. . .. . .
C) .
. . . . .. . ..
w
.... .. . . . .. ..
M.,
.. .
'
.
NEFF, H., RIEDERER,J. & WAGNER U
arion of Jnk
C I< ,
'
.
. HAUSL ER
I-- ,
"
::
ttery production: archaeometric perspectives ........... ....." ..................................... X-ray Fluorescence analysis ofobsidian tools . .. . .. .... . .
.. . .. .. . .. . .. . . .. ... . . .. . . . .. . . .. .... .. . . .. .. .. . . .. .. .. . .. .... . . . . . 709
.. .
'.
. . .. . . . . . .. . .
. ... .. . . .. . . . .. . .
. .. . .. .
'-
.. .
tlotte
ry from the Riet River Area, Northern Cape, South Africa ............................................................. Mineralogical and chemical characterisation ofRoman wall painting 715
.............. 581 . .. .. .. ... . . .. . ... . . . . . . . . . . .. .. . .. . .. . .. . . . . . .. . . .... ... . .. ..
from Medina Sidonia, Cadiz,Spain . . ..... . .. .. .
. .. . .. ... . . ...
.. . .. . . . . . . . . .. .. . . ... .
.. . . .. .... .. . . . .. . .... ... . ..
.. .. . . . . . ...... .
I. Biomaterials
bo
ae\V tttQllBte z. BELLA, S., PEREZ, M., RAMOS, J.,
t
Ga
o s <ltlcta als' souce areas and technological relationsh
MORATA, D. &
ips
CASTANEDA, V.
AN OSTEOLOGICAL ANALYSIS OF ELK (ALCES ALCES L.) BONES
l . . between minerals,
\l rehistonc non-fhn
\! t stone tools from the Atlantic Zone, Cadiz Province, SSW Spain FROM EXCAVAT IONS IN THE LUBANA LAKE BASIN, EASTERN LAT V IA
.............................................................. 72
'l'Qe
. . , N ASTICHENKO, 0., PSHEN
ICHNOV A. Y.,
3
<lC)y ."lllell
, CASEY, J., PAVLISH, L. A. & HAN
COCK, R. G. V. Faculty of Biology, University of Latvia, P.O.Box 66, Riga-JO, 1010, LATVIA
l neutron activaton analysis of siliceou
N e .
s mudstone from the Birimi Site, Ghana ....................................
011 lS ttqcj
., & ILLASOV A, L.
................................ 737
JUt,1 o\\the
A.eneolithic artefacts made from peculiar rock-types: finds
PrQ' l>.
The Lubdna marshy meadows are one of the most important Stone Age investigation regions in the East Baltic area. This
tQ foothills and in valleys ofthe Western Carpath
ians ....................................................................... ............. study deals with elk bones from the Abora, Zvejsalas, Zvidze and Malmuta River Estuary sites, dated from the Mesolithic and
J ............................ 743
Ofl:ll.l'LONG,D. G., HANCOCK, R. G. V. & PAVLISH L. A.
F.
Dt 0% sti.dies fLate Pala:o-Indian quartzite artifa
Neolithic periods. A total of I2,I20 animal bone, tooth and antler fragments were unearthed during excavations led by J. Loze
cts in the Great L akes Region
p t: enc a usrng destructive a nd non-destructive techniques .. ... .................
in I964-1990. Elk was the animal species that was most exploited by the human population. Its bones occupy the first place
Ge
SJ:! e Study ofIron Age amber from Bulgaria......................................................................................... .................................... ...
'
osteological analysis of this species. Some problems related to quantification are examined, such as the estimation of the
lo!?;ica rvANOV' N., AHLEY, G. .M., ?AL, R., GLASCOCK, M. D. & H. NEFF,
. .. 757 numbers of individuals of elk for each site. Two quantitative methods are used here: counting of the numbers of fragments and
ovellew. .
, ........................................................................
........ 761 with K. Paaver's analyses ( 1965) of previously excavated materials from Lubdna and with bone data from other Eastern Baltic
,
. Mesolithic and Neolithic sites. The results are additional to K. Paaver's data. The ecology of the area is discussed with respect
p A deteflalll tlon ofRoman marble sculptures from Pannon ia .............
............................ ...........................................................
l>t;'l'll-()l!?;. zA!<MANY, GY. & T. BIR.6, K. . 767 to the vegetational history and the adaptation of elk.
lh. )3:)(: chemical stues ofNeolithic stone tools from Tolna county, ungary
H ............................................. ................. ............. 777
KEYWORDS: ELK (ALCES ALCES L.), OSTEOLOGICAL ANALYSES,
, M., HO!lVATH, T. & KOVACS-PALFF Y, P.
-
A
ation ofthe stone llllplements
of a Bronze Age earthwork in Hungary ........................................
MESOLITHIC, NEOLITHIC, LUBANA LAKE BASIN.
.................................................. 783
e human population.
l>re t:Yl a stud'1es ofneol'thic stone tools from Baranya County, South Hungary.......... ......................................................... ..............
<>
1 1 .
of the most important Stone Age investigation regions in the Materials and methods
S<
. East Baltic. In 1960-1980, the Lubiina Lake excavation
S()lQeS1()e , erM. S. 797
}:l
l V ssus accuracy in the XRF analysis ofarchaeological obsidian: material has been studied by archaeozoologists, including Geographical characteristics of the sites
e 0n for archaeometry and archaeology .... ............................... ..................... K. Paaver (Tartu, Estonia) and V. Danylchenko (Moscow,
. ................. .............. ................................................ 805
i... Russia). K. Paaver's osteological study of the Stone Age fauna Lubana Lake is situated in the north-western and central
l', l),()<l :& STARNINI, E.
GY.
\! e?s
l>
r r
cal analysis ofpolished stone tools from some Neolithic sites of ungary ......
.
H ........................................................................ 811
(ITaaBep 1965) surveyed materials from the entire East Baltic
area, including Lubiina Lake Basin materials. In the Lake
East Latvia Lowland. The area is covered by wetlands. The
organic layer is commonly 1-4 m deep, in some places
St() . & T. BIRO, K.
v tic l:;: and classical stone quarries ('mines') in the
Carpathian Basin .......................................................................
Lubana material, elk takes the first place among all animals in reaching 6 m (Jlo3e 1979), a condition supporting bone
C: li; EELE, P., WEHLING, B., MOENS, L., DE REU, M.,
............. ............. 819 the Early Mesolithic. During the Late Mesolithic, Early and preservation. This study deals with elk bones from the Abora,
\! b:y e <l ct ?oNK. G., VON BOHLEN, A. & KLOCKENK
. AMPER, R.
Middle Neolithic, elk dominance gradually decreased, in Zvejsalas, Zvidze and Malmuta River Estuary sites, dated
()l t <l l tizat1n ofpigmen ts in Mercatellis manuscripts creasing again in the Late Neolithic. According to K. Paaver from the Mesolithic and Neolithic periods.
;::;::;::v;
,r
t croscopy .................................................................................... (ITaaBep 1965), the Early Holocene elk had larger body size
, ................. ............. 825
, . in comparison with the Middle Holocene elk. In the Middle Chronology
S(J Ce
: di s f prehistoric flint from the Gargano mines (Puglia, Italy): Holocene (from the Atlantic climatic period till the Sub
1'" c ensat10n ......... ....... ............. ........................................................................... Boreal period), elk body size gradually decreased, increasing Results of 14C-dating for all four sites are given in Table 1.
..................................................................... 82 9
.
\Vts t!?;Uu C., NIJAGUNAPPA, R. & PA VLISH, L. A. again only in the Late Holocene. Radiocarbon data were obtained from the St. Petersburg
:t;-l'e mound: provenance of unusual glassy materials ....... .... .................
. . .................................. ........................................ The present osteological material was collected in Technological Laboratory of the Institute of Archaeology,
............. 833
c() tli ' S., SARIN, P., OUSTER
HOUT, R., DE SENA, E. & WILLIAMS, W. 1964-1990 from all Stone Age archaeological excavations of Russian Academy of Sciences (JIE) and in the Geobio
gillents from Byzantine Cappadocia, part II ........... ......... ............ .....
... . . .............................................................................. Lubana Lake, including elk teeth, antler and bone fragments. chemical Laboratory of the Institute of Zoology and Botany,
....... . . 839
There are many known zoomorphic sculptures, which Archaeological Age, R.P. Age. fl.C. Arclwcological Cl.imatic period
represent elk, and hunting tools, made of elk bones and site period
antlers. The high percentage of elk in this material allows
complete osteological analyses of this species.
Zvidrc 7650+/-IOO - 57(Xl-2800 i
M ddle The end of Bnreal- the end
This type of study can help to determine changes in the
4750+/-60 Mesolithic- of Atlantic
population and morphology of Lubana Lake Basin elk during
Middle Neolilhic
the Stone Age in relation to climatic and vegetation
Zvejsalas 4905+/-70 2955 Early Ncolithic Atlantic
conditions, and to determine Stone Age hunting strategies in
Middle Neolithic
the East Baltic area.
The objectives of the study were: Malmma River -2500 Middle Neolithic The end of Atlamic
fragments and the minimum number of individuals; Aboral 3860+/-100 1890 Late Neolithic The second pml of the Suh
- to determine quantitative changes in the elk population Boreal
and morphology changes during Stone Age pe1iods;
- to compare our results with K. Paaver's work (1965) and Table 1. Chronology of Lub<ina Lake Basin
with other bone data in the literature; archaeological sites.
14
Proceedi11gs of the 31" Imernational Symposium 011 Archaeomet1y, Jerem, E. & T. Biro, K. (eds.), Archaeopress-Archaeolingua, BAR- Centi: Ew: Se1: I.
Berdnikov, S.
Estoni a11.
An osteological analysis of elk ( Alces alees L.) bones from excavations in the Lubana Lake Basin, Eastem Latvia
carbon cl Acac1e
('t at e lll y of Sciences (TA). There are layer and 46 (9.79%) for the Middle Neolith
them sf 40 radio- ic layer of during the Stone Age Conclusions
r the Zvidze site (ll 03e 1988), Zvejsalas site, 169 framents (30.95%) for M01phological changes of elk
occup t l
:\- 72 but only two of the Malmuta River
z t()ll. () and TA-180 l ) reflect the entire time period of Estuary, 823 fragments (12.91%) for the Abora
f h l site. Results
vejsa a (1'A.-
s e Zvidze site. There is only one date fo r using measurements of the first phalanx, it was possible to I.Minimal number of individuals of elk is 20 individuals
of MI e timation using various methods (the
largest value) in elk for the Zvidze site, 8 for Zvejsalas, 12 for the Malmuta River
(llo3e 1
. 988) 95) (JI03e 1988) and 3 for the Aborasl site are given m Table 2. estimate elk body size. The results indicated a decrease
, beginning of Atlantic period (Zvidze site) till the Estuary and 29 individuals for the Abora l site.
1972) ts
bu t on1y one (JIE-749) (Semyon The values depend on the method used, size from the
from K llsec1 h tsov et al. the Fieller &
ere. The dating Turner method yielding values significantly end of Sub-Boreal period ( Abora l ) (Table 3). 2. The elk population size gradually decreased from the
a
l> ave of the Malmuta site is taken lower then the
Pal The reasons for such morphological changes are not Middle Mesolithic till the end of the Middle Neolithic and
(the en e ,..clt s
A. Intic period), the Middle Neolithic layer l:iyer ml!lli)Ures
Caskel's
Quercus A n .
() f t nd T1lia pollen with Picea and Pinus additions
ttc (Jl03e 1988). At the Zvejsal Formulae
20
II 25 64
Greatest length Z\'ilJzc Atlantic I 7 I .8-89,3 79,86+!- l.46 s.7 6.85
References
period
Pi
e ?
(()<l c Pmus and Alnus occurred in the as site, Tilia, Ahoral Suh-Bore11I 18 7.l.5-81.l 77,52+/0.70 2,99 3.86
MN! ll
part of e 1 Atlantic I 8 58
Bridault, A., 1992, The status of elk during the Mesolithic,
88). The Aboral site was dated to the second GMT 19 12 29 60 K. Paaver's 32
(JI03e 1 Sllb-l3
Atlantic 72-90 79.16t/O,x7 4,9 6.19
Anthropozaologica 16, 151-160.
979). ' ()real period (Picea and Betula pollen) Fieller & Turner method
Casteel, R. W., 1977, A consideration of the behaviour of the
A to
y
tal 1
<Ima
ll 1 120 animal bone, antler and tooth fragments Table 2. Elk minimum number of individuals
Zvidze ec1 Greatest Zvidze 39 28,5-35.l 5,33
fauna! characterisation, OSSA 3/4, 129-151.
(
Atlarnk 31.68+/0.27 1.69
S57 6377. bones for the Abora l site, 3895 for Chaplin, R. E., 1971, The study of Animal Bones from
and 47 f f
0 from the Lubdna Lake Basin sites. hn::-.ad1h of
()
r()tll. r ZveJsalas [97 from the Early Neolithic layer proximal end
Archaeological Sites, Seminar Press, London and New
Malmu t a th e
The ltiv Middle Neolithic layer], and 546 for the Elk population quantitative changes during Abo1...il Suh-fiorl.'al 4] 22.S-35.5 :lf .41}+/.-0.32 2.07 6,59
York, 170 pp.
llte s er Estuary)
the Stone Age
Estuary er v r Daugnora, L. & Girininkas, A., 1996, Osteoarcheologia
a
l <ln<l z .10n of bone material is good (Malmuta River The elk percentage in the material was used
to indicate
K. Paaver's 31 3.91
sites) (Lithuanian Osteoarchaeology), Savastis,
A1lantk ]9.]5 32,,U+/-0.22 1,27
Lietuvoje
e Vt ze sites) or average (Zvejsalas and Aboral population fluctuations during the period from
Syste \Jst e d the end of V ilnius, 171 pp.
data
IJ1Jl1Jgy at the Faculty of Biology, University of Atlantic climatic period till the Sub-Boreal)
. Elk numbers animal bones from archaeological sites, Harvard
gradually decreased from the beginning to
Quantir the end of the University.
Atlantic period (from 13.48% in Zvidze Middl
Qtive e Mesolithic Table 3. Osteometrical analyses of elk first phalanx Dming, E., 1986, The fauna of Alvastra, OSSA 12, 210.
analyses of Middl Neolithic to the 13.40% and 9. 79% in
Some
the material Zvejsalas Early from the materials of Lubiina Lake Basin. Fieller, N. R. J. & Turner, A., 1982, Number estimations in
ll and Middle Neolithic). Elk numbers increa
such as tb. l'IJbl e s sed again during vertebrate samples, Journal of Archaeological Science 9,
e es lll . related to quantification were examined, the transitional period from the Atlantic climatic
each sit l'\l; tilllat period to the Comparison with K. Paaver's data 49-62.
the nutn () t o of the numbers of individuals of elk for Sub-Boreai (up to 30.95% at the Malmuta
River Estuary, Lougas, L., 1994, Subfossil Vertebrate Fauna at Asva Site,
quantitative. .
e rs methods were used: counting of Middle and Late Neolithic). During the Sub-Boreal
individu \Jf
f a climatic There were no significant differences between our Saaremaa, Reports of the Estonian Archaeological
<ll s
(Chapli n (lvf r gments and of the minimum numbers of period the elk population decreased again (to
12.91% in the measurements and K. Paaver's data (Table 3) and the trends 5, 71-93.
N l).
Society
Counting of the number of fragments Abora l Late Neolithic). Population quantitative
other s ecl 97 l)
changes are observed were generally similar. According to K. Paaver Semyontsov, A. A., et al., 1972, Radiocarbon dates of the
e s \\las used for estimating elk quantity among shown in Fig. 1:
The t e ... a ct f . . . . (1965), the number of elks increased in Sub-Boreal period, Institute of Archaeology 3, Radiocarbon 14, 336-367.
p e n or companson with other quantitative data. Middle Mesolithic - Middle Neolithic;
changes ta 1-2:
2-4: but our data suggested that elk numbers increased JI03e, M. A., 1979, Il03gHHR HeOJIHT H paHmn1 6poH3a
e
duti ng of elk was used to indicate population transitional period from the Middle Neolithic
estimati() to the Late significantly only during the transition period from the [Late Neolithic and Early
g t he Stone Age. Four methods for MNI
Jly6aHCKOR paBHHHhl,
ll. Neolithic; 4-5: Late Neolithic.
formula "-'ere Middle to Late Neolithic. This may be explained by the larger Bronze Age of the Lubana lowland], Pura, 3uHaTHe,
Used: MNI, GMT (Chaplin 1971),
(Fieller (f2a8
l'ut t ee1 1977) and Fieller & Turner's method
Casteel's The decrease of elk numbers during the Atlant
associated with declining food supply (Betul
ic period is scope of Paaver's material (entire East Baltic), compared to 211 c.
n er 19 a, Alnus , Picea, the Lubiina Lake Basin data, since the elk populations may JI03e, M.A., 1988, IloceJiemrn KaMemmro BeKa Jly6aHcKoli
82) Pin'.1s) due to climatic changes. During the have been different.
Osteo,11 Sub-Boreal pe1iod, IDI3IIBhI. Me30JIHT, paHHHR H cpegHHR HeOJIHT, [The
et,,. coruferous and successional (birch, alder)
lcql tree coverage Stone Age Habitation Sites of the Lubana lowland.
a nalyses of the materi increased at the expance of Quercus and Ulmus
Standa ct. b al , and wetland T he value of elk for the Stone Age human population Mesolithic, Early and Middle Neolithic], Pttra,
t expanded, creating good conditions for elk.
using the ()n e 3HHaTHe, 211 c.
A.. b . llleasurements were made for all fragments,
This sys t 35,00 , The high number of elk fragments (58.82%) in the Lubiina IlaaBep, K. JI., 1965, <l>opMupoBaHue Tepuo cpayHhI u
t e Q b. ttesch measurement system (Dries
ch 1976). iii
.....
allowing Lake Basin osteological material indicates that elk was the
.....
...... ......... ......... . . . . .. ..
.. .. . . . ............................ ...............................................
.... . ....... .......... . .... ...... ..
a8
.. .
(13.48%) l'll.a18 g numbers of elk bones and their proportions symbol for the Stone Age people (Bridault 1992) .
' 13 f \\lere obtained: 525 bones at the Zvidze site Fig. 1. Elk population quantitative
changes
r agrnents (13.40%) for
the Early Neolithic during the Stone Age.
16 17
DNA TECHNIQUES IN ARCHAEOMETRY - NEWS AND PROGRESS
Over the last 15 years, a series of molecular-genetic investigations has shown that, under favorable conditions, DNA can be
preserved for even thousands or millions of years. In the beginning, research focused on well-preserved soft tissues (e. g. Thomas
et al. 1989), but ancient DNA research soon shifted emphasis toward investigating skeletal components (e. g. Hage/berg et al.
1989; Hummel & Hernnann 1991; Hummel et al. 1996). Other archaeometric sources of organic origin or with organic
components were only sporadically or unsystematically checked for DNA content (e.g. Hardy & Raff 1997; Reese et al. 1996).
An archaeometric project at the Dept. of Historical Anthropology and Human Ecology, University of Gottingen, is now largely
devoted to detecting and analyzing DNA in cultural-historical and archaeological objects. The following remarks are intended
to show in examples the materials for which and the extent to which ancient DNA analysis can be used in archaeometric studies.
Three fields can be distinguished. An example will be given for each.
10 20 30 40
pl.'ime:t rR4.a
50 60 10 ao
sample sequencel
sample sequ
_ ence2
c primer rR4c-rc
Fig. 1. An aDNA-sequence from the archaeological bone was obtained twice from independent PCRs.
The comparison to reference sequences from GenBank shows that the sequences are identical
to the species Odocoileus viriginianus. Only at position 89 the anciemt sequences differ
from the GenBank sequence in one nucleotide (C-T).
DNA was obtained from bone powder by phenol!chlorophonn extraction. PCR primers that amplified a portion of 12S
ribosomal RNA gene were: rR4a: 5'AAC TGG GAT TAG ATA CCC CAC TA 3' and rR4c: 5'GAA GCA CCG CCA AGT CCT
3' Sequence analysis was carried out on AB! 310 Prism automated sequencer:
o,.____ ,.,,,. __ -'"''"I" lnornnt;ro,,nl ,<;,mmro<i11m nn Arrhnenmetrv. Jerem. E. & T. Biro, K. (eds.), Archaeopress-Archaeolingua, BAR- Centi: Ew: Se1: I.
Burge 1; J., Grosskopf, B., Hummel, S. & Hemnann, B. DNA techniques in archaeometry - News a11d progress
primers are labeled with fluorescent dye. ns in PCR tubes lead to mistakes in sequence P.. -0.. -0.. -0..
contanunatio . . E E c
Analysis was carried out on a 8% denaturing to the genera t10n of PCR artefacts. Fig. 5 <'d Cd <'<! ....
....
ana 1 s
ys1 or even . . . "'
-
VJ VJ VJ
s::
..
C
lly type
Our experiments show that it is possible to genetica
With re ..c':ATTCA.T GCA T ACGCA TCCT-3';
, The question is: how were paints mixed a couple of
st ictio crcA.
. A
GTGTTA-3') was cut thousand years ago ( Reese et al. 1996)? A superficial down to the systema tic level of an
an ancient sample
n en do
nucl e
a s es Dd ng that it is possible to
e! or Ndell. inspection of a Texas pictograph by SEM shows plant-like and individual. It is extreme ly fascinati
Assig microrganismic strnctures (Fig. 4). These superpositions material back to the animal or the plant
A. s i trace anthropogenic
\vl...<
t iell
s gna t
i on f t o
natio
n deposited on the surface at a later time can have crncial that delivered it. Questions of provenance can be answered
on
S. K., 1996, Ancient DNA from Texas Pictographs, 1989, DNA phylogeny of the extinct marsupial wolf,
Journal of Archaeological Science 23, 269-277. Nature 340, 465-467.
Fig
lvith . 3. Earl
a fra 111ode
gme nt as u rn scroll
/ nied t
Fig. 4. Example of a plant s11pe1position on a Texas
0 fit in the scroll. pictograph. Picture made by SEM (Jeol JSM-5400).
THE CHEMICAL ANALYSIS OF BOVINE SOFT TISSUE
FROM 11th DYNASTY EGYPT
Animal remains identified during the excavations of the hilltop Temple of Montuhotep Sankhkara (Thoth Hill, Thebes) included
the mummified remains of a young heifer's leg. It was recovered from the north-western foundation deposit and originated from
the right thoracic limb of an animal that was approximately 2 years of age. Pieces of dry tissue (including skin and tendons) were
removed from the pa/mar side of the metacmpal region. Chemical analyses of the 4000 years old sample revealed a remarkably
good preservation of proteins and high concentrations of sodium. The amounts of ammonia and free amino acids, indicators of
protein decomposition were low. Of the amino acids, sulphur containing cystine and remains of keratin were barely detectable.
Cystine is one of the least stable amino acids, and as such is known to have decomposed even in significantly younger samples
of wool. While soft tissue preservation in the dried cattle leg was most probably due to spontaneous mummification in the windy
and arid hilltop environment, the meat's pre-depositional treatment with salt cannot be excluded.
Proceedings of the 31" International Symposium on Archaeomet1y, Jerem, E. & T. Biro, K. (eds.), Archaeopress -Archaeolingua, BAR - Centr. Ew: Sei: 1.
CsaP ,
o, J., Csap
6-Kiss, Zs., Csap6 Ji:, J. & Bartosiewicz, L. The chemical analysis of bovine soft tissue from 11th Dynasty Egypt
cornposiuo . .
Baintner, K., 1976, Takarrna nyozastan (Animal nutrition), in
Calcium, mg/kg 6880.0 . however no such substance s could be identified
Hemes 1991), . .
Hom, A. (ed.), Allattenyesztes (Animal Husbandry) 1.,
Phosphoms, mg/kg 1330.0 fl U.
'
e excellent preservation of the. tissue..
reng th
Magnesium, mg/kg 430.0 s of total ad free ano acids, an attempt Budapest, Mezogazdasagi Kiad6, 295-414.
Aft er the analysi .
Potassium, mg/kg 1020.0 identify the D-ammo acids m the sample. The D- Baitosiewicz, L., 1984, Sexual dimorphism of long bone
was rnade to . . .
Sodium, mg/kg 11000.0 were identified both by 10n exchange growth in cattle,Acta Vet. Hung. 3213-4, 135-146.
anuno acids
.
Manganese, mg/kg 6.0 (Csap6 et al. 1989, 199) and by high Csap6 J. & Ms. Csap6, J., 1986, Ion exchange column
chrom atography
Copper, mg/kg 12.8 chromatography (Csapo et al. 1994). chromatography for the determination of keratin in meat
performance column
Zinc, mg/kg 40.4 meals,ActaAlimentaria 2, 137-150.
Iron, mg/kg 130.0 Free amino acid Csap6, J. & Henics, Z., 1991, Quantitative determination of
Amino acid samEle % bacterial protein from the diaminopimelic acid and D
Table 1. T he chemical composition of bovine soft tissue. Aspartic acid 37.936 8.92 alanine content of mmen liquor and intestines, Acta
Threonine 4.661 1.10 Agronomica Hungarica I-2, 159-173.
10 11.730 2.76
cm not contain ammonia, so that measurable amounts are usually Serine Csap6, J., Ms. Csap6, J. & Ms. Toth, L., 1986, Optimization
5.974 1.40
produced during protein hydrolysis in the laboratory. The
Glutamic acid of hydrolysis at determination of amino acid content in
31.261 7.35
decay that leads to the formation of glutamic and aspartic food and feed products, ActaAlimentaria 1, 3-21.
Glycine
87.888 20.66
acids, however, may result in some ammonia build -up even Csap6, J., T6th-P6sfai, I. & Csap6-Kiss, Zs., 1991, Separation
Alanine
Valine 35.516 8.26
during the most carefully conducted hydrolysis (Csap6 & 11.013 2.59 of D- and L-amino acids by ion exchange column
Methionine
Csap6 1986). The same holds true for fresh meat in general. 14.018 3.30 chromatography in the form of alanyl dipeptides, Amino
Isoleucine
Compared to fresh skin and hair, on the other hand, the Leucine 17.710 4.16 Acids l, 331-337.
amount of keratin identified (Csap6 & Csap6 1986) in the Tyrosine 27.619 6.49 Csap6, J., Ms. Csap6, J., Penke, B. & T6th-P6sfai, I., 1989,
sample may be considered negligible (Table 2; in spite of its Phenylalanine 26.345 6.19 Separation and determination of D- and L-amino acids
good preservation, no remains of hair could be detected on the Lysine 97.748 22.98 by ion exchange column chromatography in the form of
3.819 0.90 diastereomer dipeptides,Acta Alimentaria 4, 399-417.
surface of this find). That keratin was formerly present is only Histidine
12.502 2.94
shown by the very low weight ratio of cystine, which is too Arginine Csap6, J., Gombos, S., Henics, Z. & Ms. T6th, L., 1988,
Sum of total 425.380 100.00
small to make any reasonable quantitative estimate. Cystine is Modified method of diaminopimelic acid determination
one of the least stable amino acids, and it is likely that it in samples of biological origin, Acta Alimentaria 2,
completely decomposed (oxidised) during the four millen n ia Table 3. Free amino acid ratios by mg/JOO g sample and 159-167.
of deposition. For example, measurable quantities of such total free amino acid content in the bovine soft tissue. Csap6, J., Nemethy, S., Folestad, S., Tivesten, A., Martin, T. G.
Fi g. 2. Palmar view of the heifer sulphur-containing amino acids could not even be detected in & Csap6-Kiss, Zs., 1994, Age determination based on
'sfi .
.
samplmg area. oar sh owzng t he 2000 years old wool carpets (Csap6 et al. 1995b). Only D-histidine (93.79 mg/100 g sample, 16.7% of the amino acid racemization, A new possibility, Amino
The small amount of free amino acids (Table 3) also indicates total i. e. D+L histidine), D-phenylalanine (291.3 mg/100 g, Acids 7, 317-325.
Results a low degree of protein decomposition. The 78.92% gross amino 11.7%) and D-aspartic acid (438 mg/100 g, 7.4%) were found Csap6, J., Csap6-Kiss, Zs., Csordas, E., Folestad, S., Tivesten,
acid content corresponds to only 425 mg, most of these acids in the sample. The concentration of the other D-amino acids A., Martin, T. G. & Nemethy, S., 1995a, Rapid method
() 1'he good preservation of crude for the determination of diaminopimelic acid using ion
protein . being tied up in peptide bindings. Nevertheless, even this small lay at the limit of identification.
111Pl ete dehydration of the materials Ev is due
to the near quantity is indicative of initial decay. The sample was also tested exchange column chromatography, Analytical Letters
u was e so-called "soft"
bone dry, however, the org lli 28, 2049-2061.
}) anole . Conclusions
ecunen during dissection revealed
a so ptic study of the Protein total Collagen Keratin Csap6, J., Csap6-Kiss, Zs., Folestad, S., Orwar, 0., Tivesten,
() other remarkable feature is the ve ;h at gluey smell.
:-n Amino acid sample % In conclusion, it may be said that the artificial treatment A., Martin, T. G. & Nemethy, S., 1995b, Age estimation
g h c o ntribution of
"b() 1urn to the sample. The average Na c Aspartic acid 5.92 7.5 5.4 6.7 and subsequent mummification of this find preserved most of of old carpets based on cystine and cysteic acid content,
te t o
::i.IS cly is around 0.16% (Baintner 1976: 3
)
n f the animals' Threonine 2.12 2.7 2.1 7.0 the original amino acids in the soft tissue of the heifer's foot. Analitica ChimicaActa 300/1-3, 313-320.
1 naturally, uneven. Even cattle bloo bu t its distribution
Serine 3.06 3.9 2.9 10.4 No bacterial amino acids indicating decay could be identified. Csap6, J., Csap6-Kiss, Zs., Wagner, L., T::ilos, T., Martin, T.
d
() 0.37% sodium, w hich in no way W 0We ver, contains Glutamic acid 10.70 13.6 9.7 14.3 In addition to the small degree of decomposition, however, an G., Nemethy, S., Folestad, S. & Tivesten, A., 1997,
o 1
ex.p lain the high
""'" <liurn content of the sample presente . Proline 9.16 11.6 13.0 6.6 unusually high proportion of Na was also detected. This Hydrolysis of proteins performed at high temperatures
ain that d
this foot was salted or soak mth le I. It appears
and for short times with reduced racemization, in order
Glycine 15.67 19.9 22.5 5.4 suggests the application of some sort of salt, a substance in
di e
b Um-rich rind. The Na conten t of lUo d in s ome form of Alanine 8.19 10.4 8.2 4.2 great demand in the ancient world. It is not yet possible to tell, to determine the enantiomers of D- and L-amino acids,
de
0-25 times higher than that of u n rn-d a y bacon may Cystine 0.30 0.4 0.0 10.0 however, whether this treatment was part of normal food Analytica ChimicaActa 339, 99-107.
t e te
Q.i_IS\.:tthgte 1978: 90, 95) and the a ch
r a o
: beef (Paul & Valine 2.43 3.1 2.9 5.6 conservation (the salt content must have been very high by Gyulai, F., 1993, Environment andAgriculture in BronzeAge
cuss1on here is similarly rich in salt 0 gi cal find under Methionine 0.63 0.8 0.7 0.5 modem standards) or served specific ritual purposes. Hungary, Budapest, Archaeolingua, Series Minor 4.
Y A.nuno acids in the ancient sampl Isoleucine 1.98 2.5 1.5 3.7 The excellent preservation of organic materials is a joint Paul, A. A. & Southgate, A. T., 1978, McCance and
olysing the protein w ith 6M hy dr r i d ntified after
Widdowson ' s The composition of Foods, London, Her
l_ l oc Leucine 3.75 4.8 3.5 8.8 result of salting and the extremely arid climate. In Hungary, it
h.. C (Csap6 et al. 1997). The free anu nc a c id for 24 h at Tyrosine 0.90 1.1 1.2 has no yet been possible to detect sodium in prehistoric food Majesty's stationery Office.
.Q n o aC . 5.4
<:{'" Pl was determined without hydroly . id content of the Pheny!alanine V6ros, Gy. & Pudleiner, R., 1997, Preliminary Report of the
e sis a 2.50 3.2 2.2 4.0 remains, a fact more attributable to bleaching than to the
::tr. the s ample with 12.5% trichlor-ace _ fte r the treatment Lysine 3.29 4.2 3.9 3.3 actual lack of salt in the diet (Gyulai 1993: 43). Excavations at Thoth Hill, Thebes. The Temple of
tIC a
t ate the
c1Pl c id m order to
a protein present. Within th fr Histidine 0.56 0.7 0.7 1.1 Montuhotep Sankhkara (Season 1995-1996), Mittei
.
e a rne wo
surements, the quantity of lUI rk of these
a Uo . Arginine 6.25 7.9 7.6 9.5 Acknowledgements lungen des Deutschen Archiiologischen Instituts, Ab
tei n decomposition, was also ide a indic ator of
' nti d n Ammonia 1.51 1.8 teilung Kairo 53, 283-287.
1:'no a c ids during the amino acid an among the
aly s basic Sum of total 78.92 100.1 This research was carried out with the support of Grants
e ammoni a content of the samp le : . OTKA T 14916 and OTKA T 18179 of the Hungarian
is a
larl y to the material analysed ie rnarkably low.
here fr Table 2. Amino acid ratios by sample and protein content in National Science Foundation. The English text was revised by
esh h a
ir and skin do Dr. Alice M. Choyke.
24 25
A CORRELATION BETWEEN MYCOLIC ACID PREVALENCE AND
Fossil F uels and Environmental Geochemist1y, University of Newcastle, Newcastle upon Tyne, NEJ 7RU, UK,
1
2 Department of Chemistry, University of Newcastle, Newcastle upon Tyne, NEJ 7RU, UK,
JDepartment of Archaeology and Prehisto1y, Northgate House, West Street, University of Sheffield, Sheffield, SJ 4ET, UK
Mycolic acids, specific for Mycobacterium tuberculosis, have been found in archaeological human ribs from a 150-300 year
old infirmary cemete1y site. Sensitive detection of mycolic acids used an unequivocal combination of reverse and normal phase
high-pe1formance liquid chromatography (HPLC) coupled with fluorescence detection of anthrylmethyl ester derivatives. The
HPLC profiles of mycolic acids from ancient bones correlated almost precisely with those extracted from modern
M. tuberculosis, but differed from those derived from other mycobacteria. The number of ribs positive for mycolic acids
correlated with the documented tuberculosis prevalence and all soil samples were negative. Mycolic acids are biomarkers for
tuberculosis, and perhaps can be used to trace the palaeoepidemiology of tuberculosis, providing valuable information for
epidemiologists examining the resurgence of this disease. Biomarkers for tuberculosis frequency could also be "biomarkers"
for poverty in ancient populations.
0"0""""'""" ,1,0 I" Tntprnntinna/ Svmvosiwn 011 Archaeo111et1)', Jerem, E. & T. Biro, K. (eds.), Archaeopress -Archaeolingua, BAR -Ce1111: Ew: Sei: I.
Gernaey, A. M., Minniki
n, D. E., Copley, M. S.,
Ahmed, A. M. S. & Cha
mberlain, A.
ween mycol"ic acid prevalence and reported tuberculosis frequenc y suggests a "biomarker" for poverty
A correlation bet
Alpha hexane an ap
d Plied to a normal phase 5 m Merck clearly from those of other mycobacteria (Butler et al 991, i/ .
dissolvedlll . to
.
S 60 cartn ctge (150x4. 6 mm)for separation m Glickman et aI 1994' Ramos 1994) that share t is a-,
L1chfospher i-
methox y-, an
d ketomycolate fractions usmg methoxy- and ketomycolic acid pattern ( mmki 1987)
component -, .
a
futther reverse phase chromatography, usmg a 5 the mdivi ? .du ! -, methoxy- and ketomycolate classes (see
;
analysed bY . RP l8 cartridge (150x4.6 mm) wit .
L iCh rosph er h Fig. 3) provides incontrovertible evictence that the long
m Merck . .
m tet rahy drofuran (1:1). . .
chain mycolates ongma te from M tuberculosis (Mmmkin et
acetom
1e
al. 1993).
Data
Correlation with osteological examination
47) of the twenty-one ribs testd
R Five (SK31, 33, 40.' 41 and .
h own to contam co mponents corresponding to mycohc MycoIi ac1ds, characteristic of M. tuberculosis, have been
were s . e _
. ( F.ig. I) on iru "t"al
i reverse-phase HPLC (rpHPL C) , as co nf
iirmed m 5/ 2 I ribs under investigation (23.8%) and this
acids
. b SK33 (Fig. 2). Subjecting the total myco 1ate correlates with the recorded incidence of 27. Io or the
Alpha exemplified y
. s to normal-phase HPLC Newcastle Infirmary. The sample selection was random; those
fractio n from these fiive nb
. .
(np HPL C) showed the exp ected MTB profile of a- meth xy- mdividuals sampled were those that had been freshly reveled
?
an d ke o y colates, as seen .
m ig. 3, for SK33. Analysis of
F" on the days th at one 0f us (AMC) visited the site. Osteolog1cal
.
the m. dividual a-, methoxy and ketomycolates from these nbs examination followed site clearing; examination sh owed that
.
is given,
for SK33, m . F"ig. 4 and the other four n"bs gave one md"1v1"dual sampled for mycolate analysis (SK31) also had
. . h TB. All b samples
Fig. 1. Representatio s1mt1 ar patterns , but w1"th varying amounts of myco 1ate changes of the spine consistent w1t
_
n of the linkages of a-, . .
Methoxy- and Keto-m detected. The retention times f or the archaeological myco1ates tested for mycolates hd n bone lesions ( oberts t al. 1994).
supporting cell wall stru ycolates, to the mycob
ctures. The remainder acterial arabinogalac . ?
moiety does not necess
of the supporting ara
binogalactan is repres
tan of the
correlate almost prec1e1y with those from the modem MTB Osteological exammation showed on y 2/ 210 (0.95%)
arily have two a- and ented by R. Each pen . . . .
homologues, and the
single methoxy- and keto
mycolates. Each myc
tarabinose
(Fig. 4). The retention times of the a-' methoxy and individuals with skeletal changes consistent with tubercu!sis
methoxy- and ketomyc olic acid class consist
olates also include com s of several
ketomycolates from other mycobacteria (Minnikin 1987) can (D . J . Robertson, persona 1 commuruc ation) . Extrapol at1on
methyl branches, in add ponents with tra ns-cyclop
ropane rin gs and adj . .
be seen to be very different (Fig. 5) . All soils were negative
ition to the cis-cyclop
ropane rings which are acent from the numbers of bones positive for MTB mycolates
_
represented in the Fig
ure by triangles.
l
associated burial gro for mycolic acids. 88o/t0 of recorded literature values, but there is a margm o
und (AD 1753-1845
), provided an 1998) of a method dev error. It is interesting to note that the 23.8% prevalence f or
excellent opportunity eloped for clinical ma
to evaluate mycolic terial (Minnikin .
acids as reliable Interpretation mycolates, recorded here' is higher than ant1c1pa . ted for the
biomarkers for ancien et al. 1993). Samples
t tuberculosis. The Infi of bone (1.0 g), soil (1.0
rmary records, g), lyophilised .
1 opulation at that time (HardY 1994) ' but a hospital
for 1803 to 1845, sho M. tuberculosis Strain
tuberculosis (J. Nolan
wed that 27. I% of pat
, personal communic
ients died with 1984) (50 mg) and oth
C (Dobson et al. 198
er mycobacteria were
5; Minnikin et al.
heated at l00C
Is it really MTB? ==:; site could be expected to yield higher than average
ation), during a overnight in 8 ml Pyr disease frequencies.
time when tuberculo ex tubes, sealed by a
sis was thought to hav PTFE-lined screw
e been a major cap, with l ml 15% aqu Reverse-phase HPLC separates mycolic acids primaiily
cause of death in Bri eous tetrabutylammoni
tain (Hardy 1994). Bet .
ween November um hydroxide. according to their overall chain 1ength, with a lesser
1996 and March 199 After cooling, I ml wat Comparison with other biomarkers
7, skeletal remains from er was added and, afte .
210 articulated r centrifugation, contribution from the various mctional gro (Minnikin
individuals and at lea the supematants wer _
st 407 dis-articulated e mixed with I ml
individuals were 0.05% 9-chloro 1982) In the case of the mycolic acids from B , rpHPLC To date , the biomolecular analysis of archaeological uma
recovered (J. Nolan per methyl-anthracene in
sonal communication). dichloromethane and
;h
;J; .
the tubes mixed result in a characteristic "tight envlop " of peak (see F
.
remains f or TB has been restricted to the demonstration o
by rotation for I hou
r at room temperatu _
re. The lower which enable MTB specific mycolic acids to be stmgu DNA fragments from the insertion element IS6110, wit h some
dichloromethane layer
Methods was washed successive
ly with I ml
10% aqueous hydrochlo
ric acid and twice wit
h Im! water and
then removed, under
Sample collection and nitrogen flow, to giv
storage e crude mycolic
esters which were diss
olved in toluene (25
I) and the
Forty-two soil sample solutions adsorbed ont 1st
reverse 1st
reverse
s from 21 skeletons lab o toluene-pre-washed
elled SKI, 3-5, I ml BondElut MTB MTB
12, 21, 27, 31-33, 40- C-18 reverse-phase col
42, 4tH.8, 52, 102, 103 umns. The columns wer phase HPLC phase HPLC
, 107 and 108, e washed with
were collected. Two soil acetonitrile (4.0 ml),
samples were taken for followed by acetonitril
each individual; e:toluene (4: I)
one from the chest cav (4. 0 ml) and the
ity of the freshly exp washings discarded
osed, articulated . The purified
skeleton, prior to any anthrylmethyl esters
cleaning or lifting pro were eluted with ace
cedures and one tonitrile:toluene
from the area distal to (I:I) (3.0 ml) and eva
the skull. Soil samples porated to dryness.
were collected into
sterile polyethylene-ca
pped glass containers
and autoclaved
(121C, 15 min.) piio High peiformance liqu
r to storage at room tem id chromatography
perature. During
soil sampling, face ma of mycolic acids
sks and plastic gloves
were worn; plastic SK33 SK33
gloves alone were wo
rn for soil sample pro
cessing. Mid-shaft
rib samples were take Anthrylmethyl mycol
n from the 21 skeleto ates were analysed by
ns after lifting, HPLC using a
cleaning and osteologic Merck Hitachi L6200
al examination; plastic pump and a Gilson
gloves were worn 122 fluorimeter;
at all times during the excitation at 252 nm
se procedures and sub with a 435 nm interfer 5 5
sequent handling. ence filter. The 0 10 15 20 25 0 10 15 20 25
Bones were crushed usin flow rate was I ml
g a mortar and pestle min1, and the colum
to a fine powder; n temperature was
soil samples were air 37C. Hexane solutio
dried and powdered. ns of purified anthiy Time (minutes) Time (minutes)
lmethyl esters of
mycolic acid extracts
were applied to a rev
erse-phase 5 m Fig 2. Initial reverse-phase HPLC separatwns
Merck LiChrospher of Fig. 3. Normal phase HPLC separations
RP 18 cartridge (15
:
Mycolic acid extraction
and derivatisation 0x4.6 mm ), eluted .. .
with acetonitrile:tetra anth1ylmethyl esters of myco1.ic aCld s fironi MTB and SK31. of antluylmethyl esters of fractions collected a tial
hydrofuran (1:1). All
data were collected
The powdered bone using X-Chrom v. 2.0 Only the results from one rib sample are shown here and . '. n reverse phase HPLC (Fig. 2) from MTB an .
and soil samples wer 4k (Labsystems, Fis
e examined for ons pi .) software F. _ acid
mycolic acid, using data collection system e
zgs. 3 and 4, fior clarity. The total fraction of mycoltc Eluates corresponding to a-, Methoxy-
a modification (Gema . Fractions, correspon
ey-Child et al. ding to mycolic
acid derivatives wer
derivatives was collected for subsequ nt a al ys1s
b and Keto-mycolic acid derivatives were collected.
e collected, blown
down to dryness, normal-pIiase HPLC Conditions are given 111 Metho s. Conditions are given in Methods.
28
29
Gemaey, A. M., Minnikin, D. E., Copley,
M. S., Ahm ed, A. M. S. & Chamberlain A
een mycolic acid prevalence and reported tuberculosis frequency suggests a "biomarker" for poverty
A correlation betw
Salo et al. 1994; Nerlich et al. 285-306, American Society for Microbiology,
(Taylor et al. 1996;
C78 success
ificity of usmg this e1 ement has been Washington.
The spec
199?):
ucet -Populaire et al. 1996). Although evidence Bhatti, N., Law, M. R., Morris, J. K., Haliday, R. & Moore
Alpha C80 quesuoned
(Do . .
rese nce of the IS6110 msert10n sequence has been Gillon, J., 1995, Increasing incidence of tuberculosis in
for th e P . .
d l n virtually all modem isolates belongmg to the MTB England and Wales: a study of the likely causes, British
detec te
dam et al. 1994), a homology between a 181- Medical Journal 310, 967-969.
comp1eX (McA .
bp fragment
of DNA from mycobactena other than MTB as Brennan, P. J. & Nikaido, H., 1995, The envelope of
strated (McHugh et al. 1997). Usmg mycobacteria, Annual Reviews in Biochemistly 64, 29--63.
1s0 been demon
bridisation techniqu es, these authors obtained multiple Butler, W.R., Jost, K. C. & Kilburn, J. 0., 1991, Identification
;
b nding patte
rns from
isolates of Mycobacterium avium,
i, Mycobacteriun intracellulare,
of mycobacteria by high-performance liquid chromato
graphy, Journal of Clinical Microbiology 29, 2468-2472.
Mycobacteri um fortuit
acteria l mycolates are spec1es-spe c1fic (Butler et al. 1997, The decline of tuberculosis in England during
Mycob
Ramos 1994). victorian times, improved living conditions and selective
1991; Glickman et al. 1994;
pCR techni ques for ancient DNA detection require immunity, Abstract from P roceeding s of the International
or not this
amplification steps; it is still uncertain whether Congress on the Evolution and Palaeopathology of
procedure amplifies only the disease state, so that distinction Tuberculosis, Palfi, G., Dutour, 0. & Deak, J. (eds.),
bet ween TB, in the ancient individual, and mere infection by Tuberculosis Foundation, Szeged.
C83 Methoxy
MTB can not be made yet. Mycolic acid detection does not Dobson, G., Minnikin, D. E., Minnikin, S. M., Parlett, J. H.,
involve amplification steps, so the number of skeletons Goodfellow, M., Ridell, M. & Magnusson, M., 1985,
containing mycolic acids may reflect tuberculosis frequency Systematic analysis of complex mycobacterial lipids, in
in the original population. Chemical Methods in Bacterial Systematics,
Methoxy C85
Goodfellow, M. & Minnikin, D. E. (eds.), 237-265,
Mycolic Acids as a "biomarker" for poverty Academic Press, London.
Doucet-Populaire, F., Lalande, V., Carpentier, E., Bourgoin,
Tuberculosis is a disease of civilisation; more specifically, A., Dailloux, M., Bollet, C., Vachee, A., Muinaerd, D.,
it is a disease of overcrowding, so much so that it increases Texier-Maugein, J., Carbonelle, B. & Grosset, J., 1996,
proportionately to the degree of crowding (Wilson & Miles A blind study of the polymerase chain reaction for the
1946). Other factors involved in the increase in disease onset detection of Mycobacterium tuberculosis DNA, Tubercle
are malnutrition, stress and exhaustion (Mantagani et al. and Lung Disease 77, 358-362.
1995; Bhatti et al. 1995). Apart from an overwhelming Evans, C. C., 1994, Historic Background, in Clinical
colonisation by M. tuberculosis, disease onset is thought to be tuberculosis, Davies, P. D. 0. (ed.), 1-17, Chapman and
due to the recrudescence of the organism following a period Hall Medical, London.
of latency (Wayne & Hayes 1996). The organism would be Farer, L. S., Lowell, L. M. & Meador, M. P., 1979,
held in a dormant state until any or all of the above factors Extrapulmonary tuberculosis in the US, American
Keto resulted in a breakdown of the immune system, pennitting the Journal of Epidemiology 109, 205-217.
C86 initiation of lesions and a progressive tuberculosis (Wayne Gemaey, A. M., Minnikin, D. E., Copley, M. S., Middleton, J.
1994). Compromised immune status in modem populations C., Dixon, R. A. & Roberts, C. A., 1998a, Detection of
can be a result of infection by the immuno-suppressive mycolic acids confirms DNA evidence for tuberculosis
C85 MTB viruses, but no evidence exists to suggest that these viruses in medieval human skeletal remains, (submitted to
Keto .
threatened ancient populations, thus the single common Proceedings of the National Academy of Sciences USA).
denominator for the re-activation of M. tuberculosis leading to Gemaey, A. M., Minnikin, D. E., Copley, M. S., Ahmed, A. M.
M. marinum disease manifestation is poverty. S., Dixon, R. A., Roberts, C. A., Robertson, D. J., Nolan,
A biomarker that measures TB frequency in past population J. & Chamberlain, A., 1998b, Mycolic acids - biomarkers
will be a valuable adjunct in epidemiology and palaeo for ancient tuberculosis, Internet Arcaheology 5,
pathology. A "biomarker" that measures pove1ty in ancient http ://intarch.ac. uk/joumal/issues/gemaey-html.
populations will be a valuable adjunct to archaeology. The Glickman, S. E., Kilburn, J. 0., Butler, W. R. & Ramos, L. S.,
essential results in this presentation have been submitted for 1994, Rapid identification of mycolic acid patterns of
0 5 10 15 20 25
publication elsewhere (Gemaey et al. 1998b). mycobacteria by high performance liquid chromato
graphy using pattern recognition software and a
Time (minutes) Acknowledgements mycobacterial library, Joumal of Clinical Microbiology
0 5 10
32, 740-745.
15 20 25 D.E.M. acknowledges funding from The Wellcome Trust, Hardy, A., 1994, Death is the cure of all diseases: using the
for support in developing the mycolic acid analysis, and The
Time (minutes) general register of cause of death statistics for
Royal Society, for provision of a fluorescence detector for 1837-1920, Social History and Medicine 7, 472-492.
Fig. 5. Confimwtory
reverse-phase HPLC sepa HPLC.
of anth1ylmethyl esters
rations Mantagani, P., Jolley, D. M., Watson, J. M. & Rodrigues, L.
Fig. 4. Con of the individual ""
rv , Meth
firmatmy reverse-phase HPL C . oxy- C., 1995, Socioeconomic deprivation and notification
anthrylmethyl ester
separations of .
nd eto-mycolzc aczd . _
rated by normal phase HPLC F . Besra, G. S. & Chatterjee, D., 1994, Lipids and carbohydrates of McAdam, R. A., Giulhot, C. & Gicquel, B., 1994,
( ig. 3).
previously sep 1ate
d by npHPLC (data not
Conditions are given in Methods.
Condtt1: ons are given
sho n). Mycobacterium tuberculosis, in Tuberculosis: Patho Transposition in mycobacteria, in Tuberculosis: Patho
in Methods. genesis, P rotection and Control, Bloom, B. R. (ed.), genesis, P rotection and Control, Bloom, B. R. (ed.),
30
31
Gemaey, A. M., Minnikin, D. E., Copley, M. s. ' Ahmed' A
.
M S & Cham.berlam, A.
199-216, Ame iican Society for M icrobiology, Reichman, L. B.,1996, Multi-dmg resistance in
Washington. the world h
Presen t sltuat10n
.
. ,Chemotherapy 42,
2-9.
te
ST RONTIUM ISOTOPE STUDIES OF HUMAN BONE
McBugh,T. D.,Newport, I.E . & Gillespie,S. H.,1997, IS6110 Roberts,C.,Lucy,D. & Man
chester' K. 1994,Infl
homologs are present in multiple copies in mycobacteii a
l'es10ns of . nbs: an analysis of the
' amni at
0'Y AND TOOTH ENAMEL: THE EUROPEAN BELL BEAKER
other than tuberculosis-causing mycobacteiia,Journal of erry coll
. ecfIon,
PRICE, T. D.,1 GRUPE, G.,23 SCHROTER, P.2 & MIDDLETON, W. D.1
Amencan Journal of Anthrop
ology 95, 169-182
. Clinical Microbiology 35, 1769-17 71. Roberts,C. & Manchester,
K ., 1995, r1
Mi nnikin,
1 1ie Arehaeo
Che mical targets in cell envelopes of the
D. E.,1987,
gy 0!
lo
Disease, Sutton Publishing,
.
Stroud.
N rlic persistence,
U. & Hagedon,
Infectwn and Immunity 64,
Szei mies,
A., 2062-2069 various kinds of artifacts moved about from original sources 87Sr/86Sr ratios.
e h,A. G.,Haas,C. J., Z ink, Wayne, L. G., 1994, Dormancy of Mycobacteri
um tuberc losis to new locations. Often such evidence is interpreted to be Geologic units that are very old (>100 m.y.) and that had
B . G.,1997,Molecular evidence for tuberculosis in an an lat ncy of disease,Euro
? pean Journal of Clinic
indicative of the movement of individuals as well,but in fact high original Rb/Sr ratios will have high 87kSr/86Sr ratios
Egyptian mummy, Lanc et 350, 140 4. M1crobwlogy: Infectw . n and
al
Disease 13,908-914
Opravi l,M.,1997,Epidemio logical and clinica l aspec ts of .
W1lson' G' S. & Mil es, such an association is difficult to make. Artifacts can either be today. Such rocks include clay-rich units such as shale or
A. A. (eds.),1946,Tapley
!
rny cobacterial infectio n,nfction 25, 5 6 -59
.
. .
Pnnc1ples of Bacteriology
W
and ilson's earned or exchanged by those who made them. Given the fact igneous rocks that have high potassium and silica contents
Ra . and Immunity Vol II 1291 such as granite (Faure 1986). In contrast,rocks that are
rno s,L . S.,1994,Characte nsat10 n of m ycobact enal spec ies Arnold,London.
' ' that artifacts are indirect,or proxy,evidence of movement in
by high performance liquid ch ro matography and pattern the past,
a better approach is to investigate prehistoric humans geologically young (<l-10 m.y.) and that have low Rb/Sr
recognition,Journal directly for indications of residential change. ratios, such as late Cenozoic volcanics, generally have
of Chmmatographic Science 32,
219-228. A number of new archaeometric techniques are in 87Sr/86Sr ratios less than 0.706 (e.g.,Rogers & Hawkesworth
development that will allow us to pursue this question. The 1989). Rocks that had very low initial Rb/Sr ratios,such as
general categories of such techiques involve ancient DNA and basalt,can have 87Sr/86Sr ratios less than 0.704. These
isotopic bone chemistry. Several different isotopes of lead, variations may seem small,but they are exceptionally large
oxygen, and strontium are being considered as possible from a geological standpoint,and far in excess of analytical
indicators of past movement (Carlson 1996; Ericson 1985, error (0.00001 for 87Sr/86Sr) on thermal ionization mass
1989; Gulson et al. 1997; Price et al. 1994a,1994b; Stuart spectrometers.
Williams et al. 1995). Of these,strontium seems to show the Strontium in bedrock moves into soil and ground water and
most potential at present. In the following paragraphs we from there into the food chain. In humans and other animals,
outline the basic principles of strontium isotope analysis for the vast majority of strontium is deposited in the skeleton.
the investigation of human migration and provide a case study The mineral matrix of bones and teeth consists mainly of
from the late Neolithic Bell Beaker period of Central and calcium phosphate hydroxyapatite [Caw(P04MOH)2]. Sub
Western Europe. Results from the study of human skeletal stitution of strontium for calcium in this apatite produces
m aterial from cemeteries in Bavaria are presented and suggest strontium concentrations on the order of 102-103 ppm in hard
that both human mobility in the Bell Beaker period and the tissue. Although local levels of elemental strontium in plant
potential of strontium isotope analysis are quiet high. and animal tissue vary due to many factors,the isotopic
composition of strontium is not changed (fractionated) by
Strontium isotope analysis biological processes because of the very small relative mass
differences among the strontium isotopes. The strontium
The basic principles for the strontium isotope analysis of isotope composition of human bones,therefore,match the
human skeletal remains are straightforward and borrowed diets of the individuals,which in tum reflect the strontium
largely from geology. Both elemental strontium content and isotope composition of the local geology.
its isotopic ratios in rock,groundwater,soil,plants,and In human bone,the strontium isotope ratio serves as a
animals vary depending on local rock types (Dasch 1969; signature of the geology of the area where an individual died.
Hur st & Davis 1981; Graustein 1989),primarily on the age Bone undergoes complete replacement or turnover of its
and composition of the material (Faure & Powell 1972; Faure inorganic phase during life (e.g., Parfitt 1983), so that
1986). The stable isotope 87Sr is formed over time by the measurements of bone strontium reflect the last years of the
radioactive decay of rubidium ( 87Rb,t l /2 - 4.7 x 1010 years) life of the individual. The actual number of years is related to
and comprises approximately 7 .04% of total strontium (Faure the type of bone under consideration and its turnover rate. The
& Powell 1972). The other isotopes of strontium are non- enamel in teeth, on the other hand,forms during gestation and
32
Proceedings of the 31" International Symposium on Archaeometry, Jerem, E. & T. Biro, K. (eds.), Archaeopress -Archaeolingua, BAR - Centi: Ew: Sei: 1.
er
Enam el: The European Bell Beak
of Human Bone and Tooth
Price, T. D., Grupe, G., Schrote1; P. & Middleton, W D. Isotope 5tud"es
i
Strontium
infancy and undergoes relatively little change dming an have been found in the graves are frequently robu
t
individual's lifetime. Enamel has ve1y few internal organic with a distinctive "short-headed" skull (e.g., Gerhar Ill
t
re a197
sttuctures and is thus considered inert tissue which does not 1978). Settlements from this period, however, we
hn
recrystallize or remodel after fotmation (Steele & Bramblett unknown until recent years.
1988). Differences in strontium isotope ratios between the Bell Beaker materials are distributed irregular!
bone and tooth enamel of the same individual thus provide an Denmark to Sicily and from Ireland to eastern
1 Uro
indication of mobility and residence change (Eticson 1985, Compt atton and anal ys1s of rad"wearbon dates fro
.
1989). period suggests an origin in the Rhine delta region :0
.
Post-depositional contamination of bone and tooth before 2500 BC m a Corded Ware context (Lanting & v ltl
n
(diagenesis) is not a significant problem in strontium isotope Waals 1976). This early expansive Bell Beaker ph :.
studies. Bone is more susceptible to diagenesis than the assumed to have split into three "regional groups " e 11
. . the
denser dental enamel (Molleson 1988; Vemois et al. 1988). southern Bell Beaker with finds from Spain, P ortu
g
Cleaning techniques have been developed which remove southern France and Italy; the western Bell Beaker with ali
sues
.
m central and northem Franee, Great B" ntam and Irelan h Pommelsbrunn
1992; Sillen 1989). Moreover, Sealy (1989) tested the acid Benelux countries, the Rhine region, and the north Ge e
wash procedure of Sillen on the isotopic compositions of lowlands ; and finally the eastern group in Hungary, the Cz
87sr86 =
strontium in bone and was successful in recovering the Republic and Slovakia, Austria, and southern Bavaria.
some regions such as the British Isles, the disttibution of Be
non-uniform
known, biological values from bone which was initially
contaminated with diagenetic strontium of a different isotopic Beaker mateiials is almost continuous, while in others the
ratio. Other researchers (e.g., Staudigel et al. 1985; Schmitz et remains are very sparse. The end of the Bell Beaker is varia&l
e
al. 1991; Koch et al. 1992) using such acid-cleaning proced in these areas, depending on the date of appearance of Bronze
ures have had similar success obtaining Sr87186 measurements Age materials. The Bell Beaker period lasts longer in Brit ain
Price et al. 1994a, 1994b; Sealy 1989; Sealy et al. 1991, 1995) structures and the emergence of more mobile groups (Sherratt
has demonstrated the potential of strontium isotope ratios for 1994). This interpretation has remained a standard view of
the study of questions concerning migration. Sealy's (1989) Bell Beaker. Others, however, including Harrison (1980: 164)
study of inland versus marine adaptations in South Africa, for and Engelhardt (1991), have suggested that increased social
example, distinguished groups of people on the basis of ranking, not population movement, was responsible for the
terrestrial, marine, and mixed diets and thus could infer adoption and spread of Bell Beaker materials as symbols of
patterns of movement for the different groups. Price et al. wealth and status. The questionof the importance of mobility
(1994a; Ezzo et al. 1997) examined human bone and tooth in the Bell Beaker period is unresolved. It has not been
possible to determine if these distinctive materials were 0 100 km
samples from the sites of Grasshopper Pueblo in north-central
Arizona where ceramic and architectural evidence suggested brought by their owners or imported.
isotope ratios
that several human groups had settled the site sometime in the major foations nd stimate for strontium
Fig. 1. A schematic map of the geology of Bavaria, showing also show n.
latter part of the 13th century. The results of the strontium Bell Beaker Folk in Bavaria in Bavaria used 112 this study is
in rock. The location of Bell Beaker cemeteries
isotope ratio analysis provided good evidence for the presence
Bavaria was the focus of this study because of the large er of Bavaria is comprised
of both migrants and locals at Grasshopper Pueblo and to Foreland at the southern bord
brunn, Straubing-Oberau, and Ttickelhausen (Fig. 1). These
ted chalk sediments: The loss
distinguish different groups of migrants at the site. number of excavated buiials from graves and cemeteries of primarily of glacially redeposi
burial sites occur primarily along the Danube River, but there isotope ratios rangmg
tium
the Bell Beaker period in this area (e.g., Gerhardt 1953). More and marine carbonates have stron
The Bell Beaker period
are also examples north and south away from the river. These
e valu es were confirmd y
than 100 Bell Beaker sites have been excavated, with the between 0.708 and 0.710. Thes
sites have been excavated over the last 100 years with several Beak er cemetenes m
Bell
majority found between the Danube and the Alps. The analysis of four soil samples from
the result of rescue from recent construction (e.g., Kociumaka betw een 0.708965 and
pe ratio
The Bell Beaker is one of more intriguing, and less known archaeological and anthropological evidence suggests that the this area with a strontium isoto
& Diettich 1992; Weinig 1992). This matetial is housed in the essen tially the sam as
9377 ,
periods, in European prehistory. The term Bell Beaker is used closest connections of the Bell Beaker materials come from 0.709892 with a mean of 0.70
State Anthropological Collection, Munich, Germany. tium isoto pe ratios,
the northeast, likely in the Czech Republic, Hungary, or s of stron
for a type of pottery and a group of people, as well as a pe1iod . seawater (Faure 1986). In term
Bavaria can be divided into several zones geologically, tical ly.
analy
of time (Ltining 1994; Sangmeister 1972; Shennan 1986). Austiia, and spread southwest into Bavaria along the major these are easily distinguishable
divided between n01th and south by the Danube River
The Bell Beaker period appears at the end of the Neolithic rivers which provided water, food, and transport (Gerhardt
(Fig. 1). Zones to the north of the liver are of m i xed, non n
and dates from approximately 2500-1900 BC cal. The period 1953). The Danube must have been a primary route of Results and interpretatio
movement and settlement in the past as well as today.
uniform geological formations. The region nort eas of t e
is named after a distinctively shaped ceramic vessel,
river is charactetized by granitic sediments with 1sotop1c
Human skeletal materials have been found either in small each of the Bell Beaker sites
probably a drinking cup. Bell Beaker pottery is found most
ratios greater than 0.710, ranging to 0.750 and higher (Sollner The number of samples from
cemeteiies (10-30 individuals) such as Augsburg, Irlbach, Tabl e 1. Enamel from the first
commonly in graves that also contain other distinctive in Bavaria is presented in
unpublished data). Sediments south of the Danube are largely cal bone from the femur
corti
materials including jet and amber ornaments, some of the Ktinzing-Bruck, Landau SO, Osterhofen, and Weicheting, or peimanent molar and compact
glacial in otigin, including loess deposits in a wide band alng where both tissues were
first gold and bronze objects in Europe, and archery
as single or very small groups of burials (no more than five was collected from skeletons
the 1iver. The area between the river loess and the Alpme
equipment (Sherratt 1994). The Bell Beaker individuals that individuals), including Altdorf, Landau, Manching, Pommels
35
14
Price, T. D., Grupe, G., Schrote1; P. & Middleton, W. D. Tooth Enamel: The European Bell Beaker
Strontium Isotope Studies of Human Bone and
Cemetery Samples Migrants@o.001 %Migrants i Samples were then solubilized overnight in 3 ml cone BNo ()
.
50.0
i ()
l 60oc . 3
l
Altdorf 1
(Madison) or by wet ashing under pressure for 6 h at
I ml cone. HN03 (Munich). After evaporation of the
14 2 Tooth Bone
acid IQ
Augsburg 14.3
I
irlbach 12 2 16.7
80C , the remnants were solubilized again in 3 ml at 400 Augsburg ,::;
0.0
HCl
and ... -2
passed through a cat10n
Kunzing-Bruck 6 0 Erlbach
. exchange column ( 200-400 rn h
es'
()
Landau 9 4 44.4
HCl as mobile phase) to separate Rb from Sr. '.._;
Manching 3 1 33-3
8
The results of the analyses of these samples are pre
Osterhofen 3 37.5 . s ented
m F'1g. 2 as a graph of elemental strontium content
.
1
. . vers
0
Pommelsbrunn o.o ""16
strontmm Isotope rat10 value. Values are plotted
Straubing-Oberau s 2 40.0 againSI
e1ementa1 strontmm
. . part to spread out the distribut
m Sr ppm
TUcklehausen l 0 0.0 ion
7 2 inspe tion. In addi io , stronti m content in human
Weichering 28.6 ? sk el et
:
Total 69 17 24.6 matenal may co tam mformat10n on the composition
of diet
200 10
.
and on status (Pnce et al. 1985; Sillen & Kavanagh 198
2).
Table 1. Samples size and immigrants at individual Bell The basic expectation of the study was that differe . . :
. . nc es in
Beaker sites. Cut-off value of 0.001 is used strontmm Isotope ratios between bone and tooth of the ... 16
sam t ...
to identify immigrants. individual would indicate residential change. There ... ...
.9
...... ...
differences between bone and tooth in the data but ther
ar:
e is
preserved. Paired samples of bone and tooth enamel were also a good bit of variation in the strontium isotope val
ues.
obtained from a total of 62 Bell Beaker individuals and Because of the variation present among the samples, it is
not
prepared fo r anal y sis. I n addition, there are 12 individuals, always obvious how to distinguish migrants. Samples T
--- -----
with
largely children, for whom onl y enamel was available and 1 very high isotope ratio values clearly indicate residential 0.710 0. 712 0.714 0.720
shift, 0.708
sample of bone on ly, a total of 7 5 individuals. Approximately but what about the large number of intermediate val
ues?
half the samples w ere analy zed in Madison and half in Several approaches could be used to distinguish immigrants
in Sr 87/86
Munich. Se ve n samples w ere analy zed in both laboratories to these data (cf. Ezzo et al. 1997), but any cut off value
to
ensure compa rability of the results. A preliminary report on distinguish migrants will be regionally specific and somewhat Fig. 3. Scatterplot of elemental strontium content versus strontium isotope ratio for samples
from th sites of Augsburg
this work appeared in Price et al. (1994a) and a more detailed arbitrary. In this paper we discuss two methods for and Ir l bach, showing the distinct e
diff rence in bone isotope levels between the two sites.
study in Grupe et al. (1997). T he complete dataset for this determining the cutoff point: (1) a value based on geological
study is report ed in Grupe et al. (1997). differences in the area, and (2) a value based on bone measurement units, depending on precision. Migration from a mean+ 2 s.d. value is 0.7103. This bone value is used as a
Bone and enamel samples w ere washed and the smfaces strontium isotope ratios as an indicator of indigenous values. granitic area (with values greater than 0.710) into a region cutoff point in the enamel 87Sr/86Sr data to distinguish
removed me chan icall y (in Madison) or by ultrasonic etching (I) Although the geochemical differences in 87Sr/86Sr ratios dominated by carbonate-rich soils (with values averaging immigrants. This value substantially exceeds the maximum
(5 ruins. in 99% HCOOH, Munich), ashed for 12 h at 500C for the geology of Bavaria appear to be small, they are in fact slightly more than 0.709) would be readily visible in the 87Sr/86Sr value for soil samples from the project area (0.70992)
to remove the organic fractions, and finally homogenized. To highly significant. Modem mass spectrometers have a 87Sr/86Sr ratios in bone and enamel. A conservative cut-off and is a very conservative estimate of immigration. Using this
about 100 mg homogenized sample was added a spike of Sr measurement error between 0.00003 and 0.00001 for value of 0.001 between the bone and enamel measurements criteria, 13 enamel samples lie beyond the cut-off value of
(strontium carbonate standard SRM-987, National Bureau of strontium isotope ratios. Thus, a difference of 0.001 (0.710- should pennit probable immigrants to be identified. As an 0.7103. These 13 individuals are 17.6% of the total number of
Standards, Washington, D.C) with a ratio of Sr: spike of 1:50. 0.709) is the equivalent of between 33.3 and 100 example, Fig. 3 shows the data from the sites of Augsburg and enamel samples (74) that were analyzed.
lrlbach. Immigrants are readily detected and the geological Thus, our estimates for the number of immigrants in the Bell
20001 boundary between granitic and carbonate soils, marked by the Beaker burials from southern Bavaria range between
Danube River, is distinctive. Augsburg is located at some 17.5-23.8%. These are very conservative estimates for several
Bone
distance from this boundary, approximately 220 km along the reasons, including ( 1) the geological zones of Bavaria are quite
river valleys. lrlbach is situated very close to the Danube, on large and migration within the same geological region will not
1600 .A. Tooth the southern bank where the soil is dominated by loess. While be discernible with strontium isotopes, (2) the numerical cut
the isotope ratios in the Irlbach samples are generally closer off values are cautious as discussed above, and (3) the turnover
to 0.710, reflecting the mixing of sediments in the river valley, rate in cortical bone is very slow, as discussed below.
E the individual in grave 16 is clearly detectable as an Given that this estimate is conservative, does 17.5-23.8%
0...
1200
II immigrant by the high strontium isotope ratio of the tooth represent a high or low rate of migration? First of all it is clear
0... enamel.
that not all individuals are changing residence; there are no
I...
Vi A Using a cut-off point of 0.001, 14 of the total of 62 tooth indications of mass migration or demic diffusion. Clearly
bone pairs from there are several generations of the population represented in
Boo-" . ... the Bell Beaker period indicate immigrants.
In addition, a Bell the burials and some of these have not moved from outside the
Beaker child from the site of Straubing has
fl.\ ia A
a very high
Sr isotope ratio in tooth enamel of 0.716210
(no compact femoral
bone was available). This child must
local geological regime in their lifetime. Secondly, a ,,normal"
rate of ffiigration in a model population can be estimated from
..
400 have moved a
substantial distance in the early years of its life life table data. Herrmann et al. (1990) use a migration rate
JJ!a A Thus, a total of .
A 15 of the 63 Bell Beaker individuals, almost based on histmical data of 10% for replacement in a model
.... A A 25
(23.8%), moved residence to new geological regions population. If this estimate of 10% is appropriate for
A dunng their life prehistoric Europe, then 17.5-23.8% mobility among the Bell
times.
0 2) An other cutoff value was also used, based on isotope Beaker people was very high indeed.
ratio value
0.709 0.711 0.713 s in bone, which provide a conservative estimate for The strontium isotope data provide details on other aspects
0.715 0.717 long-term
residents of the area. For this cutoff value we used of migration in the Bell Beaker period. Examination of the
81Sr/B6Sr he mean value +2 standard deviations for bone strontium chronology of the individual sites indicates a tendency for
is otope Bell
.
atio m the
sample. Thus, >97 .5% of the bone ales h hig er migration rates in the earlier part ofthe Beaker
Fig. 2. Scatterplot of elemental strontium content versus strontium isotope ratios in bone
fall Within
this cut-off point. The mean and standard devrnt10n period (Table 2). The sites from this initial period are smaller
or the b one
and tooth enamel for Bell Beaker graves in Bavaria. 87Sr/86Sr values are 0.7090410.000616; the b o ne and measured rates of migration vary from 33% to 50%.
Price, T. D., Grupe, G., Schrote1; P & Middleton, W D.
Strontium Isotope Studies of Human Bone and Tooth Enamel: T he European Bell Beaker
The larger cemeteiies of Irlba ch, Augsburg , a nd Weicheri ng migrants came from a number of differe
nt areas. We Ag at oeath 45 Dasch, E. J. , 1969, Strontium isotopes in weathering profiles,
are dated to the younger phase of the Bell Beaker pe1iod and roughly estimate the minimal distances travers ca
ed by mi n + + + ++ + + ++
deep-sea sediments and sedimentary rocks, Geochimica
. . . -... +... +++ +
contain more buiials, 20 -30 inhum a tions, perhaps an md iv1duaI s. Among the sites we investigated gran t 100
the ce 0 et CosmochimicaActa 33, 1521-1522.
' m ete
indication of a more sedentary popula tion; migration rates a re Augsburg was at a maximum distance ry at oO
from the gra Engelhardt, B. , 1991, Beitrage zur Kenntnis der
ru tic .Q 80 0
lower, 14. 3% to 33%. depos1ts northeast of the Danube (Fig.
.
tted. Beakers and Faure, G., 1986, Principles of Isotope Geology, New York:
using the 0.001 cutoff value. T hu s a h igher prop01tion of 45. Variable bone turnover rates in the mid-shaft femur and m
daggers do not provide direct evidence
of human mobility. the anterior iliac crest are seen in the curves of re-calibration. John W iley.
females appear to be migr a nts. T his higher proportion might That information can come from huma
n skeletal material On the basis of such models, it will be possible to estimate the Faure, G. , and Powell. T. , 1972, Strontium Isotope Geology,
be explained by the practice of female exogamy, the and analytical techniques such as ancien
t DNA analysis and age of movement by comparing strontium isotope ratios in New York: Spiinger-Verlag.
movement of at marria ge fr om a n ancestral
females isotope ratios in human bone.
different bones and teeth. Frost, H. M., 1969, Tetracycline-based histological analysis of
residence to the community of the male. At this point, This study will be expanded along severa
l lines in f uture Strontium isotope analysis of human bone and teeth is a bone remodeling, Calcified Tissue Research 3, 211-237.
however, the evidence is only suggestive and additional research, attempting to locate the specifi
c areas from which very useful technique for the investigation of prehistoiic Gerhardt, K. , 1953, Die Glockenbecherleute in Mitte/- und
research is needed. these migrants came and examining
the life history of population movement and residential change. The range of Westdeutschland, Stuttgart.
migrating individuals. First, we will contin
ue the analysis of applications is wide - documentation of foreign rulers, Gerhardt, K. , 1976, Anthropotypologie der Glockenbecher
Sex #Burials skeletal material and sediments from Bell
Beaker burials. questions of animal pastoralism and transhumance, evidence leute in ihren Ausschwarmelandschaften, Glocken
Male 38 The strontium isotope ratio values of the
migrants in our of conquest, marital residence changes, and many others. This bechersymposion Oberried 1974, J. Lanting & J. D. van
Female 24 Bavarian study suggest that the areas from
which individ uals list continues to grow and we anticipate a variety of reports der Waals (eds.), pp. 174--164, Bussum-Haarlem.
Unknown 13 moved were to the east and north. Probable
source areas for utilizing this new and important method in the coming years. Graustein, W. C. , 1989, 86Sr/86Sr ratios measure the sources
the Bavarian migrants lie in Austiia, Hunga
ry, and the Cz ech and flow of strontium in terrestiial ecosystems, in Stable
Table 3. Sex of Bell Beaker burials. Republic and we are currently seeking these
samples. Acknowledgements Isotopes in Ecological Research, Rundel, P. W. ,
Second, we will examine individual vaiiat Ehleiinger, J. R. and Nagy, K. A. (eds.), pp. 491-512,
ion in strontiu m
Two cases of migration are observed in su b-adults. isotope ratios. Most of this variation is in fact
associated with There are a number of individuals and institutions to be New York: Springer-Verlag.
A female in Grave 9 at Augsbu rg died at approximately 15 diet and migration and the number of years
an individual has acknowledged in a project such as this. Funding for this Grupe, G., Price, T. D. , Schorter, P. , Sollner, F., Johnson, C. &
years of age. Her 87Sr/86Sr e namel ratio of 0.711687 indicates spent in the new place of residence. However,
other sources of research was supplied in large part by the Deutsche Beard, B., 1997, Mobility of Bell Beaker people revealed
an oiigin north of the Danube, while her 87 Sr/86Sr bone ratio of variation may come from intrinsic local variab
ility, multiple Forschungsgemeinschaft. Additional funding came from the by stable strontium isotope ratios of teeth and bones,
0.708184, matches the local levels at Augsbu rg. This juvenile changes in residence, changes in diet, age,
or the type of U.S. National Science Foundation (BNS-870273 l and BNS- A study of southern Bavarian skeletal remains, Applied
female must have made the journey duri ng infancy for the skeletal tissue measured. We are particularly
interested in 9111680). A number of colleagues have contiibuted to this Geochemistry 12, 517-525.
change to appear in tooth enamel. Grave 1 from Straubing was variation within the skeleton during an individ
uals' lifetime research including Brian Beard, James Burton, Bernd Gulson, B. L., Jameson, C. W. & Gillings, B. R., 1997, Stable
a small child with an isotope sign ature in th e tooth enamel for information about the life history of that
individual. Bones Herrmann, Clark Johnson, Jan Lanting, Dr. A.M. Parfitt, Steve lead isotopes in teeth as indicators of past domicile -
(0. 71621) comparable to granite. N o bone sample was and teeth remodel at different rates. Denta
l enamel from Shennan, and F. Sollner. Sample preparation was done in a potential new tool in forensic science? Journal of
available from this individ u al, bu t local soil samples provided different teeth develops at different ages
(Hillson 1986). Madison by KathieEvans and Bill Middleton and in Munich Forensic Science 42, 787-791.
a value of 0. 709656, indicating tha t the child was not born at Published data for tum-over rates (e.g., Frost
1969; Jowsey & by Dirk Weickmann. Hamson, R. , 1980, T he Beaker Folk: copper age archaeology
the site. These two cases of migration duri ng childhood, Gordon 1971; Parfitt 1983; Simmons
& Grypans 1989 ) in Western Europe, London: Thames & Hudson.
combined with the higher mobility of females, strongly indicate that enamel is stable, femur turnov
er is on the order References Herrmann, B. , Grupe, G., Hummel, S., Piepenbiink, H. &
suggest that people did not travel a lone (as traders or of 20 years, while the anterior iliac
crest remodels in Schutkowski, H. , 1990, Prii.historische Anthropologie.
craftsmen), but rather in sma ll groups, pr esumably famili es. approximately two years.
Carlson, A. K., 1996, Lead Isotope Analysis of Human Bone Leitfaden der Feld- und Labormethoden, Spiinger
As noted above, cemeteries were smaller duri ng the early Bell Sealy et al. (1995) and Cox & Sealy (1997)
have suggested for Addressing Cultural Affinity: a Case Study from Verlag, Berlin.
Beaker phase, perhaps suggesting smaller, co-r esidential units that different skeletal components should
provide information Rocky Mountain House, Alberta, Journal of Arclwe Hillson, S. , 1986, Teeth, Cambridge: Cambiidge University
at that time. Nevertheless, only half or less of the interred on changes in diet and residence during life.
Thus the human ological Science 23, 557-567. Press.
individuals were migrants, indicating that the entire group in skeleton preserves not only an infancy/deat
. h contrast in Childe, V. G. , 1950, Prehistoric Migrations in Europe, Oslo: Hurst, R. W. & Davis, T. E. , 1981, Strontium isotopes as
the cemetery had not tra veled as a un it , arguing against the sis 1s s .
r 6 r rat10s, but essentially records the isotop lnstitutet for Sammenlignende Kulturforskning.
ic life history tracers of airborne fly ash from coal-fired plants,
amval of Bell Beaker folk e n masse as warriors or invaders. of the individual. Based on differences within
an indiv idual Childe, V. G., 1957, The Dawn of European Civilization, Environmental Geology 3, 363-397
It is possible to examine dist a nces a nd direction of skeleton, it is possible to predict expec
ted strontium isotope London: Routledge & Paul. Jowsey, J. & Gordon, G., 1971, Bone turnover and osteporosis,
migration as well. The overa ll direction of migration for the ratios each year after a change in residence. Cox, G. & Sealy, J ., 1997, Investigating identity and life
A model based on in The Biochemistry and Physiology of Bone. Vol. III:
Bell Beaker people, base d on the strontium isotope data different turnover rates in tooth enamel
of the first molar, mid histories: isotopic analysis and historical documentation Development and Growth, SecondEdition, G. H. Bourne
appears to be from northea st to southwest, i. e., from granitic shaft femur, and anterior iliac crest is
presented in Fig. 4. of slave skeletons found on the Cape Town foreshore, (ed.), p. 201-238, New York: Academic Press.
to chalk and loess sediments. The varia tion seen in the In this hypothetical example, an individual
moved at the a ge South Africa, Intenationa/ Journal of HistoricalArchae Kociumaka, C. & Dietrich, H., 1992, Ein Grabeifeld der
strontium isotope ratios for tooth enamel suggests that of five from an area of a normalized
value for strontium ology 1, 207-224. Glockenbecherkultur vom Spmtgeltinde der Universitat
38
39
Price, T. D., Grupe, G., Schroter, P. & Middleton, w D.
Press.
Raton, Florida: CRC Press. for each metal. The results on the content of Ca, Ba, Pb, Cd, Zn and Mg on every sample does not show any insight about their
Sillen, A. & Kavana h, M.,
1982, Strontium and pale
Price, T. D., Schoeninger, M. J. & Annelagos, G. J., 1985 odietary feeding habits for the high levels of contamination in the second human sample. It is required a palinologic study in order to
research: a review, Yea
Bone chemistry and past behavior: an overview, Jouma i 25, 67-90.
rbook of Physical Anthro
pology shed more light about it. Nevertheless the studies gave some interesting results, especially on sample SFIOOJ -H, as it is the
of Human Evolution 14, 419-448.
Price, T. D., Blitz, J. Burton, J. H. & Ezzo, J., 1992,
Simmons,
J. & Grynaps, M. D., 1989, Mechanisms of
contamination by vulcanism which can modify the studies about migration routes in the population of the Americas.
bone
. . . formation in vivo, in Bon
m Prehistoric Bone: Problems and Solutions, KEYWORDS: METAL-TR ACING, PALAEOZOOLOGY, PLEISTOCENE, HOLOCENE, MEXICO.
Diagenes1s e, Vol. I: The Osteoblast
and
Osteocyte, B.K. Hall (ed.
), pp. 193-202, Caldwe
Journal of Archaeological Science 19 513-529. ll: The
Telfo rd Press.
Price, :
D., Johnson, C. M., Ezzo, J. A , Burton, J. H. &
Introduction natural input with food, water and air, which depends pre
P. & Zindler, A., 198
Stau gel, H., Doyle,
Encon, . J. E., 1994a, Residential mobility in the 5, Sr and Nd dominantly on the local natural geochemical conditions, and
isotopesystematics in fish
prehi tonc southwest: a preliminary study using . teeth Ea1tl and Plane
,
tmy This paper forms pmt of a preliminary study focused on the the synthetic materials in the immediate or more distant parts
. S c1ence Letters 76, 45-56.
strontmm isotope analysis, Journal of Archaeological detennination and comparison of metals on two human of men and animal 's environment.
Steele, D. G. & Brambl
ett' C A 1988, 1rhe , Anat omy and
Science 21, 315-330. samples and four different mammal species collected during a More information is provided by a comparison of temporal
f!
40
Pro ceedings of the 31" lntemational Symposium on Archaeo111et1)'. Jerem, E. & T. Biro, K. (eds.), Archaeopress-Archaeolingua, BAR- Centr. Ew: Se1: 1.
of Mexico
per iod in two regions
Pleistocene-Holoc ene
and animal bones from
na tion metals in human
0
det ernu
g1ugtgl
Ba1ug/g) Ca(uglg) 'M
--- -- - -- _J Estim. Age Placd(uglg) Pb(ug/g) \Zn1ug/g)
;
i:!lI type -- - -- - --
1 4
-
31 i 681.
-
1
-
!
0 .3
I
0 156.94
I
- --- -----------
20,000 vears Oaxaca I 0
human {female) SFl-H001
- ---
38.811327-
----------
,1')
-KRAUTAC KER
CA L IN VE ST IG ATIONS AT SOPRON
TIIRACOLOGI
AN (NW-HUNGARY)
2
JEREM, E.
RUDNER, E. Z.1 &
Hungary,
rsi u. 43-45., Budapest,
cal Res ear ch Ins titute of HAS, Budao
1 Geogra phi ngary
Uri u. 49., Budapest, Hu
cha eol ogi cal Ins titu te of the HAS, 1014,
2Ar
- SETTLEMENT
CEMETERY
100 km
' (
,./
I
0 som
ted area.
Fig. 1. Sopron-Krautacker excava
Centi: Ew: Sei: 1.
Archaeolingua, BAR -
E. & T. Bir6, K. (eds.), Archaeopress -
Jerem,
um on Archaeometry,
Proce edings of the 31" Imernational Symposi
Rudne1; E. Z. & Jerem, E.
Anthracologica/ investigations at Sopron-Krautacker (NW-Hungary)
Nevertheless, H
anthracolog1c aJ
. u n gary
was among
an al s s
the first places
where
Deininger (1890 1
Methods
, 8 1 on
was made
t Bronze Age
thanks to the
work of
Accordingly' charcoal pieces
remains vaiies betwee 0
n 0
th e syste The size of our charcoal
matic sam .
20.0 cm, however we only examined the
span of the si . plmg fo r the
338-304
whole life
te cont
med, size5'
or Urngrave 23.
fraction in the
0.2-20 .0 cm (Schoch 1986) deb-5841 -25.15 217540
eside various
212-172
charcoal sample
.
s i nt n biomaterials
ed to ide
ntify the natu
vegetation grow ral arborea l tigations (Stieber 9 362-262
in g W ithi Reflected-light-microscopical inves
1
n a fe w hun
Kiln 365. -25.65 222030
dreds of meters .
deb-5833
the settlement
98 ), and eingrnber 1978, 9 259-194
(Ve rn et around and vanous tree-anatomy atlases (Schw
application of th
is veget
also the mult
ivariate . 1 90
were used m
ning
a sean j 378-337
ation m arc tree species identification and
funerary practi hitecture, ma .
Y hce (Figueiral s (Rudner et al. 19 ? deb-5809 House 100. -25.02 226045
ces and e nufacturi ng '
1988; Smart & H 306-210
eI ectron ffilcroscope for taking photo
9 )
veryda I'
1996 Boyd
1988). As a
.
the
offman We have exaffilned
711-709
three fracture faces of the remain
conclusions con
result, we ca draw . s ' 1
.e
. .
512-441
.
House I. -25.07 241050
cernmg
. tat'IOnaI-c1 im deb-5795
transversal, long1tudmal radia
1994. F'i
vege l and tangential
439-400
(B adal et al. atological pla nes
gueiral 199 events
sometimes we attaine h
)
Fractures were made by hand or
: 5 dte
surface by cuttmg the fragments 726-677
.influence . on the na . . as well as human .
House 361.
appropnate
-24.80 243050
deb-5793
. u1 al son-ou
e re a1rea 542-407
rnvest1g at10ns w ndmgs. Althou with a ra
zor.
1
gh botanical .
al. 984- 85; Fac
blade. No cheffilcal
radiocar bo
n
dy came treatment was applied so that
&
d out at this
726-677
site (Jerem et
' Jerem 198 ,
sa r
5 Jerem forthcoming),
datmg wouI d be possible as well.
House 138.
.
they pro vide info
rm ation . All microscopic investigations were deb-5803 -24.96 243050 542-407
mamly on canied out at
on the b the cultural the
727-677
. as es of vegetation Department of Plant Anatomy, Eotvo
. . archaeo bo
rnvest1g at10n s. T
.
House I.
s Lorand Univers
deb-5843 -24.40 2435 45
tamc
'
aI and carpol
542-410
h e re1,.0re ogical ity,
' our prese Budapest, Hungary.
Hungary as suc h nt results are
7.:SU-659
a I arge unique in The preparation of the charcoal for
area w1t h a radiocarbon dating w
n ana1 y House 308. -25.57 243550 560-557
great amount of
deb-5789
has not .yet b ee
(1982). Measurements
. material made according to Csongor et al.
549-408
sed to ex
composI110n an d
u s es 0f
PI ore the deve
lopm ent, done according to Hertelendi et al.'
s method (1987, 1989) in
we :
Age; it is also u ni 740-630
arboreal
que Ill Ce
vegetation from
House 301.
. the
deb-5801 -25.68 245050
the Iron
574-416
ntral E urope c Nclear esearch Institute of the Hungaiian
su bsequent time 10r the previous Academy of
p eri0 d s
as w ell. and Sciences m Debrecen, Hungary. En-
746-601
ors were taken into acco
unt
in congrnence with Hertelendi (1990
). Calibrations follow the deb-5817 Pit 192. -24.07 246545
Archaeolog
method f Stuiver & Reimer published 598-426
in 1993. All the samples
ical c
ontext and ?
sampling
House I. -25.63
were umformly taken from Quercus
Due to the e
xcellent
robur!petraea species.
deb-5819 246550 749-424
topographi. In some cases, derivatographical analy
ses are being carried
finds chrono log i c position of
c a l l y ra the site ' the out by Gy. Szo6r at the Depa
House 270.
nge from
deb-5799 - 24.52 248560 766-418
M'ddl
1 the COpper Ag rtment of Mineralogy and
e Ages ' a
nd trace e to the late Geoloy, Kossuth Lajos University
s of vari ou . , Debrecen, Hungary,
be demonstr s cuI turaJ mflu
od (1971). The aim of these
. ate d. T he ences could according to M. Foldvari's meth
golden er
the fiirst deb--5786 Urngrave 167. -25.31 291540 1123-1021
from the si xth to a 0f this settlem
ent dates analyses is to specify from the
c . century B burning temperature if the
iound m the cem C on the basis
erem 1981,
of items
1984-85) an d the
etery (J charcoal samples came from a heart
1987; J erem et al. h, a burnt or a buried
deb-5821 Urngrave 75. -24.77 296550 1242-1069
same is .
house, or just from buried vegetation.
data (Table 1, F
ig. 2).
con firmed
by some radio
carbon
Charcoal sam
deb-5810 Urngrave 167. -25.47 297555 1270-1075
ples Radiocarbon evidence
re collected
other structures from the graves
of th
rnfiel cu and
1397-1246
Hallstatt an d th e lture, from the
Urngrave 159. -24.78
end of the
deb-5811 307050
La en It was the first time that an exten
1425-1137
e periods. sive 14C analysis was
of bonfires, Wo They containe
nearby vegetati o especially on Iron Age
oden g . d remains made on Late Bronze Age and
rave-posts
2853-2796
or JUst remain
coevaI and La
n . F urt s of the
Pit 32.
material in Hungary. This first
deb-5836 -24.71 413045
her sam series of data represents
2733-2596
ples came fro
Tene s m obie J cts of almost the whole life span of the site
ettlemen t
46
47
R11d11e1; E. Z. & Jerem, E. Anthracological investigations at Sopron-Krautacker (NW-Hungary)
Sample name
House I. II 1 deb-5795
House 361.
II Ml deb-5793
House I. II Ml deb-5843
House 308.
,. deb-5799
House I. deb-5919
Urngrave 159.
im::IO deb-5811
Pits and the nearly entirely analysed House I., from where Rosaceae. Nevertheless our work has provided more in
3. Fraxinus excelsior/omus 4. Carpinus betulus
45 samples were taken from various parts of th e house, formation on local arboriflora.
contained a more diverse arboreal flora with Quercus In some cases, sieving of section material took place, and
robur!petraea, Fagus sylvatica, Crataegus, Pyrns, Tilia, more than 2 g charcoal/I 00 g sediment ratio was found. This
Ulmus, Sorbus, Acer, Salix-Populus an d pines. is relatively high in comparison with the data of Carcaillet &
No significant differences could be observed yet in the Thinon (1996), thus suggesting a nearby forest or/and a major
subsequent age s. fire phase.
Number of Number of
Tree species structures samples with the
where this type occurrence of the
of tree tree sp.
occurred
48 49
)
Sopron-Krautacker (NW-Hungary
Anthracological investigations at
Rudner, E. Z. & Jerem, E.
although
y (Jerem et al. 1984-85),
were pollen analysis and carpolog
struction. Mature trees ence of a more open env
iron ment
luable for con the latter data provided evid
ha r d - wOod' va ny cas es frag me nts with viti es. But
it i s a . We found
in ma nsity in agricultural acti
urp ose as a result of growing inte
.
b'lity seems
1 198 8). These natural milieu of arch
avai la & Ho ffm an could grow
l 1996; Smart eristic of loess areas,
o o d (Figueira .
young branches. maple-oak forest, charact
fue 1 w
st cases, rem ams of
in the two latter forests.
These types
eS Cont
ained in mo able here and beech could mix
1a c have profit
e p know n to
.
fi r
tica (Plate 1.2)
1s closer environment.
Fi us sylva rco al. Ac cor din gly, might have occupied the e use,
cha
to provide goo d te 1.1) had the widest scal
g g qualities and industrial kilns and furnaces (Jerem
urnin Quercus roburlpetraea (Pla
e1ti es. It was
l in quantity and wood-prop
used as fue .. probably because of its
1t was 8; ilhs et al. 1998).' .
e m et al. 199 "':' also the main fuel for hear
ths.
1984; Jer
.
, for
around bmldmgs and pits d as fuel in industrial
kilns and
were found m and Fagus sylvatica was use
Soft woods Crata egu s and Pyru s, al. 1998).
Sorbus, m et al. 1998; W illis et
alix-Populus, furnaces (Jerem 1984; Jere
ample S veg eta tion (no n
s and pits,
: the bur ied
the remains of in and around building
obably as Soft woods were found
l, Bo yd 198 8) or they were used for probably represent the rem
ains of
charcoa used for basketry and they
artefactual
nks to their flexibility.
ueira! 1996), tha the bmied vegetation.
basketry (Fig ur/petraea (Plate 1 .1) to our
y rites, Quercus rob to nowadays, according
During funerar The climate was similar
spe cies are thought to Hun gary
sp. were utilized
. These zonal vegetational map of
a d Sorbus
n results and the present day
n (Boyd 1988), so eo
k trad itio plex pala
ropean mystical fol gested by a series of com
have a long Eu e. (Fig. 3), which was sug
mal aco
purpos (pollen, microvertebrate
,
e been selected on
they might hav ecological investigations
cal
underline the archaeologi site lies in
The fire dam age to the rem ains l (Jerem forthcoming). The
gh itlogical analyses) as wel
wo uld be
ng house-fires. Alt hou
etat ion as wel l as climatic in
obsevvations concerni a border zone regardin
also pro vid e
g veg
s, the y can
the causes of fire d nowadays, too.
interesting to specify evefluences, as can be observe
r, we still
, how out
climatological informa
tion. To achieve this
tive analyses are being carried
Individual and compara n
tal reco
need further rese arch. ing a complex environmen
with the purpose of mak faun a
the
whole of the flora and
struction including the d
Conclusion ral setting and the dom
esti cate
comprising both the natu
clim atic fluc tuat ions and shed
reve al
utacker species. They may also
settlement at Sopron-Kra caused by human activity
.
Charcoal remains from the ral ligh t on phe nom ena
near by cemetery prov ide evidence on the natu es are uniq ue in Hun gary as well as
and the These investigation seri
2. Alder grove few hun dred s of met ers e number of
within a Europe owing to the larg
arboreal vegetation growing in the whole of Central
et & Thie baul t 198 7), whic h relatively
around the settlement (Vern of the analyses and the
r-ba nk alde r samples, the complexity
types such as rive ly accurate
consisted of edaphic vegetation n. They promise a fair
es (Pla te 2.3) . large chronological spa
ash- elm grov ng a home
groves (Plate 2.2) followed by oak- the natural sphere providi
st (Pla te 2.1) coul d picture of the region and
fore
As a zonal community, hornbeam-oak
for an Iron Age population.
it was also suggested by
3. Oak-ash-elm grove have existed in the mentioned ages, as
forests;
1. Zone of montane beech
Hungary (Borhidi 1961) -
Fig. 3. Climazanality map of Zon e of oak forests;
eam-oak forests; 4.
2. Zone of sub montane beec
h forests; 3. Zone of h ornb garian Plain,
arid zones of the Gre at Hun
5. Zone of forest-steppe; 6. 1soxere
s showing the indentically
Plate 2. Present day pictures of th e reconstructed vegetational types on the bases of semiaridity-index
values.
leisztocen vegetacio
Stieber, J., 1967, A magymorszagi fels6p
Age to Roman, at the archaeological site of Cast Ko
. "lto", L. &. Bm tosiewicz, L. (eds.),
enyek tiikreben,
ch in Hungmy II, 85-
Acknowledgements ro m t6rtenete az anthrakotomiai eredm
trade 95.
Penices, N.W. P01tugal, Vegetation History met1.i.c I resear ocene in Hg
'
.
Archaeo (Vegetation history of the Upper Pleist
C & Hem z, c . ,
a a .
Bartha, D., 1992-1993, A magyarorszagi dendroflora tagjai Heinz, C., 1991, Upper Pleistocene and Holocene vegetation B. E. ( ed.),
ey & Son
Bennett, K. D. & Toth, A.,
s. 117-12.
nak florisztikai, conologiai, okologiai es termeszetvedel in the south of France and Andorra. Adaptations and first Palaeohydrology, Wil
W illis, K. J., Siimegi, P., Braun, M.,
' Microscopi
F H c Wood Anatomy,
mi mutatoi, Erdeszeti es Faipari Tudonuinyos Kozleme ruptures: New charcoal analysis data, Review of Schwemgru b er, 1978
in Hungary: who, how
1998, Prehistoric land degradation
.,
nyek 38-39, 13-32. Palaeobotany and Palynology 69, 299-324. Zurcher AG, Zug.
and why? Antiquity 72, 101-113.
Schweingruber, F. H.,
ds,
Borhidi, A., 1961, Klimadiagramme und Klimazonale Karte Heinz, C., Badal Garcia, E., Figueira!, I., Grau Almero, E., 1990, Anatomy of European Woo
Jakucs, P., Ka-pat hi, I.,
Zolyomi, B., Barath, Z., Fekete, G., .
Thiebault, S. & Vernet, J.-L., 1988, Identification des
Karpathi, v., Kovacs, M. & Mathe,
Bern-Stuttgart. ?
I., 19 7 Emre1 ung
..
Ungams, Annal. Univ. Sci. L. EO!vos, Budapest. se1: biol. .
Smart, T. L. & Hoffm
an, E. S., 1988, Envrronmental
.
Feuillat, F., Dupouey, J.-L., Sciama, D. & Keller, R., 1997, A Osthallstattkreis, Mitt. der bsterr. Arbeitsgemeinschaft
new attempt at discrimination between Quercus petraea fiir Ur- und Friihgeschichte 37, 91-101.
and Quercus robur based on wood anatomy, Can. J. F01: Jerem, E., forthcoming, Anthropogenic Effects on P re
Res. 27, 343-351. historic Environments in Hungary, Archaeolingua 11.
Figueira!, I., 1992, Methodes en anthracologie: etude de sites Budapest.
du Bronze final et de l'iige du Fer du nord-ouest du Jerem, E., Facsar, G., Kordos, L., Krolopp, E. & Voros, I.,
Portugal, Bulletin de la Societe Botanique de France 1984-85, A Sopron-Krautackeren feltart vask01i telep
e
139, 191-204. regeszeti es kornyezetrekonstrukcios vizsgalata, (Th
of th e
Figueira!, I., 1993, Charcoal analysis and vegetational archaeological and environmental investigation
cker ),
evolution of North-West Portugal, Oxford Journal of Iron Age settlement discovered at Sopron-Krauta
Archaeology 12, 209-222. ArchErt 111, 141-169; 112, 3-24.
Jerem, E., Balla, M. & Balazs, L., 1998, Early Celtic stamp
Figueira!, I., 1995, Evidence from charcoal analysis for
ed
environmental change dming the interval Late Bronze pottery in the eastern Alpine Area: workshop activi ty and
53
TUDY OF MIGRATION IN MEXICAN PREHISPANIC P OPULATIONS
A S
BY USING DNA ANALYSIS
The movements of Mesoamerican Mexican populations through Mexican territory have been motivated by various natural
and social causes. The results of archaeological research and physical anthropology data show similar cultural characteristics
between Teotihuacan and Tula populations which are represented clearly by Coyotlatelco and some Mazapa ceramics,
offerings, burial patterns and cranial deformations (Epiclassic occupation). These facts suggest that both populations could
belong to the same genetic group in this Epiclassic occupation. In this work the genetic frequencies of these populations were
compared to see whether they are in agreement with the historical sources as well as archaeological and physical
anthmpological studies. Comparative analysis of allele frequencies of these populations vaiy in time and space.
.
... "'" F & T. Biro, K. (eds.), Archaeopress -Archaeolingua, BAR- Centi: Ew: Se1: 1.
A study of migration in Mexican prehispanic populations by DNA analysis
Sa/azw; C. Z., Arrellfn, R. & Vargas-Sanders, R.
CJ) Q) Q)
chain reaction (PCR) (Piiiibo et al. 1989; Akane et al. 1993; Genetic results reveal that Teopancazco and La Vi . -0 co
.....
Q) -
-0 -ro Q) en
c:
.....
enu Q) CJ) .c 0 co c:
& Salvo 1990), composed of Maillard products (Piiiibo
c: - - co -0 <(
Rogan bmials, wh ich represent the Classic period at T eot -0 Q)
:2
-
ih co c: .....
co
<( c: - c: ::J 0
-
.....
-
(.)
CJ) ::J
1989) or humic acids (Hagelberg & Clegg 1991), fulvic acid have two identical genes, while the other three are diftluacaii
.....
Q) .0 Q) Q)
t- CJ) ;:l c: 0 (.)
<( - Q) - -0 c:
erellt. -0 w
.....
a. CJ) - Q)
(.) CJ) '
(Tuross 1994) or the Heme group (Goodyear et al. 1994). But if we compare allele frequencies from the Cla j 0 a. Q)
ss CJ)
-ro :s: -
-0 Q) - -0 - CJ) .....
co Q) co
>- -0
Epiclassic from Teotihuacan burials there is no coincid
e Q)
.....
Q)
DNA extracted from prehispanic bone was cleaned using e <..?
-ro Q) - ::J - c: Q) ::J ::J ::J
::J
::J - -0 co Q)
CJ) co CJ) c: -0 -0
er
.0
various methods of agarose gel electrophoresis. Contemporary frequencies. However, similruities arise again in lJt 0
_..J
-0 co
>
(.) -0 (.) c:
-0
>
E
(.) (.)
-0 Q)
ro co
.0 .0
co
-
0 c: 0 ro CJ) Cl) c: '+- -
DNA was used in all cases as a control to test that the Postclassic in some genes such as GYPA and BB;;; 0 -0
- ::J Cl) -0 -0 Cl) co 0
::J c: 0
0 CJ) c: c: CJ) ro ..... c:
0.. Q)
Nevertheless, the Coyotlatelco and Mazapa phases wh c:
purification methodology did not interfere with amplification ic h 0 a. ro Q)
E
::J (.) -0
=
ro 0
0::: a.
c: Q) 0 0 Q) c: ::J -
and also as a negative control to show that there was no very close phases in time show more differences that tho
IJt c: Q)
E
a. x 0 c: ro
.....
E
0
-
.....
t- Cl) co a. .o
Q)
.....
z co (.) c: CJ) co c: co E
contamination during the extraction and purification methods. Classic and Epiclassic periods. E
CJ) Q) c: c: -
0 ro
co -
'+- .....
<(
.....
c: (.) 0 c: 0
E 2
.....
'+-
DNA was purified using the following two methods: (1) On the other hand, allele frequencies in Tula populatio 0 c: Q) .0 0
-
.....
c:
'+-
_..J Q) . CJ) Q) ..... ' Q)
ns <( - CJ) Q) >-
-0 co 0 Q) ..... co -0
- - Q) 3:
DNA recovery from NuSieve Agarose (Perkin Elmer Co). are different except in LDLR. ro
-
- - - - ro
()
.....
ro Q) c:
-
(.) 0
E
c: a..::: Q)
c: ro c: CJ) c:
E
::J ro
After performing electrophoresis on the aDNA in for If we accept the theory that people
coming fro o. c: c: ::J Q) a.
E
(/) 0 (.) Q)
Q) -ro c:
2
.....
c-
.c -
was incubated for three minutes at 45C. An equal volume of populations we would expect slight differences between them. >- <') (.) CJ) - - Cl)
ro
>- CJ) Cl) -
- x c:
ro
>- CJ) - 0 - Cl
.....
ro ro Cl> Q) ro
phenol previously heated at 50C was added and the sample Neve11heless, results indicate an important change as can be
.....
- ro co Cl> Q) Q) ..... c:
(.) - -0 co Q) -
(.) -
.....
Q) CJ) Q) ..... c: Q) -
.c ..... -0
'+- .....
0
0 - c: (.) Q) ......
.c a. ro -0 c:
::J
CJ)
.....
Q)
.....
0
stirred until homogenised, and centrifuged for 10 minutes. observed in Table 2. Although the other theory is focused on ro -0 Q) .0
-
c: .0
.....
-
.....
Q) c: (.) .0 a. ro ro u)
a.
Cl) - - c: ro
-
ro
.....
Q) 0 ro 0 CJ) ::J
(..)
.....
Q)
Two successive phenol-chloroform and chloroform-isoamyl a northern origin for the early settlement at Tula, at this - Q)
(.) 0 ro
Cl) ro .c a. .c a. (.) -
c:
u er -o
a. c: 0
<( c: ..... Q) - 0 CJ) Q)
alcohol extractions were performed and precipitated with moment we do not have samples from that geographic region u) ro ro
E
0 ro Q) -0 Q) (.) -0 CJ) CJ)
t- Q) ro c:
(.) ro CJ) co
(.) .c Cl) .c c: c: Q) Q) -0
ethanol (Salazar 1995). (2) Electrolution into troughs. The to comprue these populations. <( .c
(.) c: co
c: -
.....
ro ::J ro
Q) ro ::J
.....
c: E (.)
E
0 -
Q) ::J CJ) -0 0...
E
.c CJ) - c: .c -
(.) Cl) co c: 0 E .o c:
method was employed as reported by Vargas-Sanders et al. One may also compare the genetic traits in two succesive _..J c:
Q) -ro
- Q) co
CJ) -0
Q)
Q) <.D 0
-
ro .._
E
ct
-
CJ) ..... CJ) Q)
<( .c .c .....
O'l
Mazapa -
:J c:
'+-
::J -0 - 0 c:
(1996). Once the aDNA was visualised on Agarose gel stained populations within in Tula: Coyotlatelco and Q) 0 -0
u
'+-
- -
() -0 3:Q._ 0 Q) ro c:
O'l c:
-ro
.....
Q)
E
Q) - a. 0 ro ro
3:
.....
E E
:::i
..-
with ethidium bromide and the UV light, a trough was cut (Table 2). In this case the LDLR gene showed the same ro
.....
w
-a
ro 0 > - .....
-
(.)
Q.)
-ro (.) co
0 > 0 0 ro
::J
E
0 Cl)
lx buffer:glicerol (1:1). Every 2 or 3 minutes eluated fluid In general, we may say that the genetic frequency of the I 0 ro
(.) ..><: 0
-
(.) Cl) w - Q) (.) 0 -
Q)
-ro - 0 -0
(.) w -0 co
.....
a.i E
(..) 0 Cl>- -;;:; Cl) .c ro Q)
Cl)
ro
:
..... .....
.c c: c: CJ) 0
ui
::J 0 a. c: ::J
from the trough was recovered and then refilled with fresh same cultural group in the two sites are different. The genetic 0:::
.3 ui
::J
.....
..><:
0 ::J .c - ::J > ro co .0 .c
c:
.....
Cl)
<(
0
::J .c 0 ro
0 0 Q)
-
ro Q)
solution until all the pigment-free DNA was obtained. frequencies of the Teotihuacan and Tula populations change 0 - Q) - a - (.) >-
.....
c:
0 -0 - -
Cl .0 Cl)
-
-0 - N -0 Cl)
2
.0 0 c: Q) ro - (.)
2
..... - -
.....
E
-.c Q)
> - Q) ro ro
2
.c Q) Q) Cl) c:: ::J
Pigment free-DNA precipitated by ethanol, the pellet was with time. CJ)-
-
-0 .c > - - .q- O>f- co - x
.....
- Cl> ro (.) Q) > 0
x O'l
- :2
ro - Q) ::J
E
.I-
c: 0 0 o::: u I
:s:
(.) Q) (.) Q)
resuspended in sterilised distilled water and stored at -20C. T hese results suggest that it is possible to have more Q) ::J - (.) w Q) O'l
0
Q)
co c: 0 a. -0 -0 .c
c: ...... z
.... .
z 0 .0 ro ..... co
.....
::J ro
by Hagelberg et al. (1991). (PCR) analysis, J. Forensic Sci. 38, 691-701. - 0 ..... ::J co
(..)
E E ::J
.....
c: (.) ::J
& Mason, I. J., 1994,
ro
Forty cycles of polymerase chain reactions were petformed Goodyear, P. D., MacLaughlin-Black, S. <..? Q) N
>
er
CJ) -0
er
CJ)
> co -0
c:
as follows: denaturation at 95C for 60 s, annealing at 57C A Reliable Method for the Removal of Co-Purifying O
w co
(.)
c:
Cl) c:
ro
>- ro
>-
ro ro ro
I-
.-1
_..J _J ro
- c:
for 1 min and extension at 72C for 1 min. PCR Inhibitors from Ancient DNA, Biotechniques 16(2), 0- 0 Q) c: Q)
w (/) - a.
.c
.....
.c .....
0 0 (.) co
Q) (.) co
Amplification products were analysed by gel electro 232-235. <( Q) 0 > c c:
.c
0 c: .
I
& ro
I- Q) (..) (..)
phoresis. Positive amplifications were hybridised as indicated Hagelberg, E. Clegg, J. B., 1991, Isolation and -ro -
- -
(.) .....
> (..) (ii -
:2 :2
,_
c: Q) ro
The results display a similar archaeological context between DNA Analysis, Nature 352, 427-429.
.....
Q) co Q._
""
.....
& Robert, H. C., in press, La cultura
.0
in morphological and metrical traits and cultural body Mastache, A. G., 0
ro 0::: a. _..J Cl) >-
-o
"Q'.O
Cl'.'. = >- 0 >-
Q) .0
modifications such as cranial deformation with two modalities Coyotlatelco en el area de Tula, in Las industrias lfticas Q._
:;::::; .c
c:
..... -0
0.. (.)
0... Q)
I-
de Q) -0 -0 - Q)
found in Coyotlatelco and Mazapa phases (Table 1). Coyotlatelco en el area de Tula, Instituto Nacional
>- Q)
2 (.)
Q)
.....
The allele frequencies for Low Density Lipoprotein Antropologia e Historia, Colecci6n Cientffica. ro
(.) (.) ..... 0
Q) Q)
_J 0
Piiabo, S., 1989, Ancient DNA: Extraction, Characterisation,
.....
57
56
Arrellfn, R. & Vargas-Sanders, R.
Salazai; C. Z.,
ESTRONTIUM
HBGG D7S8 GC EOD IE TA RY APPLICATIONS OF
GYPA =
pAL N SITE
Period or phase
LDLR
- IM AL BONES FRO M THE TEOTIHUACA
A=O
-
A=0.5 A 0 A= 1 A=0.5
ND ZJNC IN AN
G.1
ADEZ, R . .' & ZARAZUA,
Teotihuacan Classic A
B= 1.0 B=0 B=0
MANZANILLA, L..' VAL
B= 0.5
(Teopancazco) B=O '
JEDA, s.,
TE
C= 0.5
luca,
Nuclem; Can: Mexico-To
A=0.5 A= 1.0 A=0 iones Nucleares, Centro
Nacional de !nvestigac
A=0.5
A=0.5 1 Instituto
Teotihuacan Classic (La nes Antropol6gicas,
B= 0.5 B=0.5 B=0 B= 1.0 2Instituto de Investigacio
xico, D.F.
Ventilla) B=0.5
Me xico , Ciu dad Universitaria, 04510 Me
Nacional Aut6noma de
Universidad
C= 0
A= 1.0 J
levels ca
Teotihuacan Mazapa d ca lcLUm
m zinc an n valu es
ELEMENTAL DISTRIBUTION
La Ventilla human bone
TEOTIHUACAN SITE
40 r-....--
35 .
.
,.O
,. 0 Ca/ -1,5
.. a
'C.
- h.-. Si
- 30 .
p. ... ...
a
.. - a --<>-P . I -2
RABrTWEA
25 -e--AI ST
;::: ' RABBIT
/:r. -' - cG -2,5 CACOMIXL
r- sBrr.
20
OOG
'
HUMAN HUMAN SKUNK.
f-
z ,,
- - / (.)
HUMAN HUMAN
{:>--
15 ,\ -
-- --
/ ...
' -3
z ' /
HUMA.N
0 / "' CANNIS
I
u 10 \ /
'i
'
\ /
/
(.!)
5 .
/ 0 -3,5
..J
4,45 30,1 80,1 180,1 1801 1951,1 1960 -4
DISTANCE (microns)
-4,5
Fig. 1. Elemental distribution of Si, Ca, P and Al in a section of human bone from La Ventilla.
SPECIES
- rescence. The Si(Li) detector used for collecting Results and discussion
by X ray fluo . Fig. 2. Log Sr/Ca values for species from Teotihuacan site.
fonnat10n from the sample has an active area of
the spectr al i n
80 rnm2, an FW
HM of 160 eV at 5.9 KeV and a bias voltage The E l 9PER, ZENT7 and E93COM samples show high
v. Samples were exposed to primary radiation from the concentratJ.ons of rron that means soil particles inclusion.
Of 900 .
238P u sources. An acquisition time of l000 s (live (Table 1). The data of potasmm concentration are simil arm TEOTIHUACAN SITE
l 09Cd and
d to fluoresce the samples and standardize. Data g de, w1c
h means a few differences in ionic exchange and
time) wa s use
ozroYAH OZENT7 HUMAN 1223.8 3.5 763.9 9.4 Fig. 3. Log Zn/Ca values for species from the Teotihuacan site.
ozrovAH OZENT8 HUMAN 652.9 5.1 692.9 4.8
ozrovAH OZENT14 HUMAN 778.7 16.6 646.2 12.5
Acknowledgements
VENTILLA E08VE HUMAN 244.4 3.5 1433.4 3.5 Conclusion
VENTILLA E23VE HUMAN 140.7 18.9 1549.3 15.0 acknow ledge support from
The authors gratefully
Evaluation of the data from Teotihuacan animal and human
VENTILLA E27VE HUMAN 180.3 15.6 1643.3 6.9 Materials Depart ment of Nuclear Resear ch Nation al
bones of barium, strontium, zinc and calcium in a variety of
Carapia
V ENTILLA E31VE HUMAN 109.4 4 1783.3 3.5 species from different tropic levels, including herbivores, Institut e. We also thank J. Francis co Cruz, Leticia
analysing
omnivores and carnivores, as well as humans reveals that and Thelma Falcon for help with prepari ng and
CAVES ACVE1 RABBIT 629.5 4.2 1307.7 13.9
barium and strontium values show slight differences in mean samples.
CAVES ACPAP1 RABBIT 982.6 3.5 618.2 7.1
values for species. Thus herbivores have closer Sr/Ca ratios
-
CAVES ACVN4 RABBIT 613.4 3.5 569.9 10.8 than the ratios for carnivores. References
CAVES ACVE1 RABBIT 693.1 3.6 602.8 3.5 The examination of thirty archaeological samples from
Teotihuacan site confirms the closed model for this Burton, J. H. & Price T. D., 1989, The Ratio of Barium to
CAVES E9ZO SKUNK 490.0 3.5 n.d. ption of
environment. Strotiu m as a Paleod itary Indicat or of Consum
CAVES E93COM WEAST 1400.0 70.0 4500.0 675.0 17,
In the Sr and Zn data, there are significant differences Mmine Resources, Journal of Archaelogical Science
CAVES E77TLA OPPOSUM 180.0 3.5 n.d. according to species, and the human values are clearly more 547-557.
CAVES E185CAC CACOMIXL 450.0 23.0 4500.0 675.0 similar to carnivore values that to herbivores values, Bmton, J. H. & Price, T. D., 1990, Archaeometry, Birkhauser
- consistent with the variety of food resources of the diet. Verlag Basel, 787-785.
CAVES E2CAN CANNIS 177.6 4.6 1012.4 9.2
Burton, J. H. & Wright, L. E., 1995, Nonlinearity in the
Although trace element studies focused primarily on
CAVES E119PER DOG 2621.0 33.4 789.3 78.2
- trontium and barium, zinc is also a valuable paleodietary Relationship Between Bone Sr/Ca and Diet, American
-
CAVES E862BCA CANNIS 320.0 4.0 n.d. Indicator. The use of bmium, strontium and zinc permits not Journal of Physical Antropology 93, 273-281.
CAVES E85CAN CANNIS 320.0 4.0 n.d. only the Ezzo, A. J., 1994, Zinc as a paleodietary indicator: An issue of
- identification of the trophic position, but also stretch
differences for species in semimid environments like theoretical validity in bone chemistry analysis, American
61
60
,
Tejeda, S., Manzanilla
' L. ' Valadez R & Za razua, G.
T PROTEINS
IC A TION OF A NCIEN
IDEN TIF NTECUHTLI
ESPECIES LOG Zn/Ca LOG
Ba/Ca
S CUL P TU RE OF MICTLA
IC
H UMAN
ROM A CERAM MEXICO
PLO MAYOR,
-3.57015361 -2.44521488
F AT THE TEM
HUMAN 2- .91584917 -3.38421398 -2.58258164 D.'
MARTiNE Z, R.
R.,1 OR TIZ DiAZ, E.1 &
HUMA N -2.98517409 -3.26126287 -2.76066052 DERS,
S-SAN
ARGA co C. 04510, D.F.
MEXICO,
HUMAN -2.96165394 -2.8760585 -2.40395804 V ma de Mbcico, Mbci ;
Nacional AWOno x1Ca,
ntrop ol6 gic
' Univ.,,idad ndad Nacwnal Autmwma de Me
HUMAN 3- .08801512 n.d. gadane> A
Fac ultod de Med<Cma, Um ver
. to de 1,,.,.,ti a Expenmental, .
- ento de Medicin C.P. 04510.D.F.,
MEXICO
HUMAN -3.05833551 n.d 1,,,nwzp a
epart '" Me xico
,
n.d
,,tanding t/1
approach to unde
n.d gical mata ia1' off"' a new
nd on diffmnt m
chae olo hemical and
nt pmtein' fou re,ul1' of the bioc
cie will pre ,.nt the
RABBI n.d . c
ifi ation
of an
cted i' haemoglo
bin. Thi' papa
in Mexico C ity.
This wamic
fheiden the protein' that can dete ulp ture fou nd at the Templo Ma yo' re,.nted
RABBI n.d f out on a ceramic " w ith blood, o' rep
at we carded monie>, wa' covered
RABBIT
a
p O o<
o
c l an
o l y'i' th
ath, Mictlantecu htli
who, in ritu al cere
tein' on the figu re,
but to al'o
1mman a God of De the prmnce of pro
n.d
:;
e
at, t /1 M .x ic
objective of our. "
"arch i' not onl
y to deten
.
n ine
is human or not.
RABBIT 2- .42597903 n.d figuMe r argliabechi codex The . niq ues, whether the blood
che nu cal tech
the immuno
SKUNK -2.7679898 -2.88401611 -3.38916608 ify' by JEF and
1.d ent1 CUHTLI, BLOOD,
URE , MICTLANTE
WEAST 2- .49840498 -3.2155998 O RD S: CER AMIC, SCULPT
KEYW IEF, IMMUN OCH
EMI CAL .
HAEMOGLOBIN,
OPPOSUM -2.22051 raphic, biochemic
al,
hemastix, crystallog
periments including
CA COMIXL -3.62324929 h et al. 197l a, 197l b;
of
and the reproducibility
become standard,
blood proteins has
klin 1988; Smith &
Wils on
(Gurfinkel & Fran
.
Glascock, M. D.: Sand ord, M. K., Gerritsen, Sanders, W., Parsons, J. & Santley, R " 1979, The Basm of
. the methods is low
1992; Kooyman
et al. 1992; Man
ning 1994 ; Down
s &
of
J F. . n, soil conditio ns
Farnum El m ent s m Ancient Human Bone and Mexico ' Stud'ies in Archeology, Academic Press, Fiedel 1996 ). Ofte
.
,
S ., 1 99
5, Trac e Lowenstein 1995; idity ,
k le ta r - &
on s e t . Anz. 4, 361 3 70. Q rition dev eloping (Gurfinkel
ean ar ea, An ro
Sutton, M 1994
' I ndi rect Evidence in Paleonut exts have been
Mediterran
.11 L . , 199 ,
G
4 eografta Sagrada Inframundo en
e The vzet and Health of P rehistoric Americans
Studi s
archaeological cont
yman et al. 1992;
Cattaneo et. al. 19 93;
22, 99_'194.
Franklin 1988, Koo it
, . 199 6 ). Obv iously,
M anzaru a , 11. Instituto de In - et. al.
11tropo/0 1 s N.
a al. 1995 ; Tuross
Teot h i ua can, A Varg as-Sanders et dep ositi on
lo as
t opo i , Universidad Nacional Valdes, F. M. & Jaramillo' A ' M., 1994, Sistema para el laboratory the
reproduce in the
ve stigaciones An - . is impossible to
ns of preservation
with
exico, 53 6 5. andlisis por fl uorescencza de rayos x Manual de!
e Me ico, M
x fy the conditio
ntro de Estudios al Desarrolio Nuclear, La
ono m a d . process and speci ess ary to
::Ca, 76p.
Aut11 o ez. C. & Freter, A. C.' 1996, Oat 1ng ' s, it is nec
U to these problem
L. L p . . precision. Added s from
M anzaru a, , difficult to recover
protein remain
vauons m Qu Tunn els behind the H
t Exca consider that it is d
Resul s h e S un At T eot lhuacan, Ancient Meso- Valadez, A. R., 1992' Imp acto de! Recurso Faunfstico en la ic tools, espe ciall y those manufacture
t
i of . cert ain kinds of lith
P yra m d
sile x.
Sociedad Tceot111uacana' UNAM Facultad de Ci encias obsidian and
. a 7 245-266. glasses such as
. out of volcanic rtunity to obtain
amenc oppo
l., !991, Reconstruction .of Anasazi
.
mic sculpture fr o
esa n . . samples of a cera s. In
Black M
ain
d iet, N utnents and 00d m Paleonutrition, Academic cient protein rem
tried to recover an
excavations, we the
'
63-70. u ynh. V. H., 1996, App lication of EDXRF P ress, pp. 5, 6, 4 4. two advantages:
firs t, the porosity
of
T H &H .
Nguyen, . nni at10 l ead and O ther Trace Elements in
this case we have
refe rred to as havi
ng gone
n n of mate rial and, seco
nd, this figure is
t o the dete . l the Magli abechi
Codex
fluids of Ind ustna W orkers in Vietnam, X- shower of blood in
the BJOdy 4 through a ritual
- 1 .
)' 25, 3
ctlvmeti (Lopez 1995).
ray Spe
context
Archaeological
Mictlantecuhtli: n
of preservatio
and conditions
the foundation of
Mexicas begins with
The history of the ntown
present-day dow
1325 on the site of
Tenochtitlan in AD red an area
, the city cove
1). Built on a lake
Mexico City (Fig.
Fig. 1.
62
Sanders, R. V., Ortiz Dfaz, E. & Martinez, R. D. the Temp/a Mayo1; Mexico
ion 0r ancient proteins from a ceramic sculpture of Mictlanternhtli at
Iden.11;!;cat
1
Fig. 4.
Sample gr/ml Total gr. group in the presence of HP2, the evidence being the
Fig. 2.
oxidation of a colourless diaminobenzide to a coloured
PBS
Back+PBS (1) 7,000 350 diaminobenzidine. The results of human haemoglobin from
of four quare mile. In one of the most impressive buildings
6,600 330 archaeological samples after IEF 3-9. In this case, only three
of Mxico-Tenochtitlan - La Casa de las Aguilas two Back+PBS (2)
Fig. 3. 1,600 80 bands were stained, corresponding to the different
_
64 65
Sanders, R. V., Ortiz Diaz, E. &
Martinez, R. D.
2. Dating
0 n .
reagent, J. Biol. Chem. 193
, 265-2?5. Phenot
Loy, T. H., 1983, Prehist
oric blood residues <let BLACKW ELL, B. A. B.,1 OLSEN, J. W.,2 DEREVIANKO, A. P.,3
. . ect'ion on
surfaces and identific ation of species of ori . . toot TSEVEENDORJ, D.,4 SKINNER, A. F. R.1 & DWYER, M., 5
g ,
200, 269-271. m Science
Loy, '!" 1 Dept of Chemist!y, Williams College, Williamstown, MA, 01267, USA,
H. & Wood, A. R., 1989,
Blood residues
aayonii Tepesi, Turkey, J.
Field Archaeol. 16
an 1 .
45
1s at bonnie.a.b.blackwell@williams.edu, anne.1:skinner@williams.edu,
Loy, T. H. & Hardy, B. 60 2Department of Anthropology, University of Arizana, Tucson, AZ, 85721-0030, USA, olsenj@u.arizona.edu
L., 1992, Blood resid
e an a1 : 3/nstitute of Archaeology and Ethnography, Russian Academy of Sciences, Siberian Branch,
Ys1s
90,000-year-old stone tool
s from Tabun Cav
e, Israel Novosibirsk, Siberia, 630090, Russiavolkova@archaeology.nsc.ru,
Antiquity 66, 24--35.
Manning, AP., 1994, A caut 4Archaeological Sectm; Institute of History, Mongolian Academy of Sciences,
ionary note on the use of
hematix Ulaanbaatm; Mongolia, ganbold@magicnet.mn,
detection and confinn . and
dot blot assays for the
. ation 5Townsend Harris Highschool, Flushing, NY, 11367, USA
archaeo1ogica1 blood residue
s, J Archaeol. Sci. 21,
of
McDonnel, A & Staehe 159 -1
lin, A, 1981, Detection
chrome f, a c-Class Cytoch
rome, with Diaminobe
of C
Y
. .
m po1yacry1amide gels, Ana nz1.din. e T sagaan Agui Cave lies south of the Gobi Altai massif. Sedimentmy fill includes intercalated loess, sand, and clay units, with
l. Biochem. 117, 40-4
4.
Newma , M. E. & Julig, P.,
? 1989, Protein residues some soil horizons, as well as ironstone concretions, gypsum, eboulis, alluvial, terra rosa, and gravel beds in deeper layers. In
on litltic
artifacts from stratified bore Test Pit 2, Stratum 3 contained Levallois-like points, whereas Strata 4 and 5 had flakes and burin-like tools, among other
al forest site' Cana
d"zan J. artefacts. Deeper layers yielded retouched flakes, burins, notched tools, and scrapers. ESR was used to date seven subsamples
Archaeol. 13, 119-132.
Newman, M. E., Yole I, R. M., Ceri, H., & Sutton, fmm one Equus and one cervid tooth, both from Stratum 4, associated with transitional Middle-Upper Paleolithic materials.
Fig. 5. M. Q., ESR dating uses the radiation-sensitive signal found in well c1y stallized fossil tooth enamel, but not in modern teeth to date
1993, Immunolog1cal pro
tein residue analysis of
Ii thic archaeologic .
al materials, J Archaeo
non- fossil teeth. The equivalent radiation dose needed to produce the observed ESR signal is the integral with respect to time of the
References l. SCl.. 20' natural, environmental dose rate experienced by the tooth after its deposition. Since the age depends on the uranium (U) uptake
93-100.
Remingtn, s_. J., 1994, history assumed, three limiting cases are calculated assuming early U uptake (EU), continuous (linear) uptake (LU), and recent
Ascenzi, A., Bnmori, M., Citro, G. & Zito, Identifying species of orig
R., 1985, in from U uptake (RU), respectively the minimum, median, and maximum ages. ESR isochron dating requires numerous subsamples
prehistonc blood residues,
lmmumulogical detection of haem Science 266, 298-299.
oglobin in bones of Sensanba gh, G . Wilson, from large mammal teeth, but eliminates in situ dose measurements or sediment sampling, because the tooth acts as its own
ancient Roman times and of Iron and A. C. & Kirk, P. L., 197l
Eneolithic Ages, . a, dosimetn: The teeth yielded a mean age 33.2 3.8 ka (EU), 36.2 4.4 ka (LU), and 39.2 5.1 ka (RU). 230Th/34U dentine dates
Prot m stability in pres
erved biological remains
Pmc. Natl. Acad. Sci. USA 82, 7170
-7172. . ]. might clarify whether U leaching has occurred. Given that little difef rence exists between the various model ages, Stratum 4
Survival f biologically acti
Cattaneo, C., Glsthorpe, K., Phillips, R. & Soka ve proteins in a 8-year-old
l, R. J., 1993, sample dned blood, Int. J. probably dates to 33.2 3.8 ka, assuming EU, but may be as old as 39.2 5.1 ka. Stratum 4 was deposited during Isotope Stage
Blood residues on stone tools: indo Biochem. 2, 545-557.
or and outdoor Sensanba gh, G. , Wil 3, a time when the Gobi experienced less arid conditions.
experiments, World A rcha eology son, A C. & Kirk, P. L., 1971
25, 29-73. . b,
Protm s ab1lity in preserve
Do wns, E. E. & Lowenstein, J. M., 1995, Ident d biological remains II.
KEYWORDS: ELECTRON SPIN RESONANCE (ESR) DAT ING, MAMMAL TEETH, ESR ISOCHRON DAT ING,
ification of Modificat10 and aggregat
archaeological blood proteins: ion of proteins in an 8 year-old
A cautionary note sample of dried blood, Int. TSAGAAN AGUI CAV E, MONGOLIA, MIDDLE-UPPER PALEOLITHIC, LE VALLOIS-LIKE PALEOLITHIC,
J. Archaeol. Sci. 22, 11-16. '
. J. Biochem. 2, 558-568.
Smith, T. R. & Wilson, M. LAT E PLEISTOCENE, GOBI DESERT.
Fiedel, S. _J., 1996, lood from stones? Some T., 1992, Blo od residues on ancient
methodological tool surface: A cautionary
and mterpretatJve pro blems in blood note, J. Archaeol. Sci. 19'
residue analysis 237-241. Introduction T he Tsagaan Agui Cave contains a narrow, inclining
J. Archaeo/. Sci. 23, 139-147. '
Tuross, N. & Dillehay, T., 1995 entryway, a lower grotto, a rotunda-like main chamber, and
Gurfinkel, D. M. & Franklin, U. M., 1988, A , Mechanism of preservation at
study of the Electron spin resonance (ESR) date archaeologically at least two smaller chambers behind the main rotunda.
!"fonte Verde and one use of
feasibility of detec ting blood resid biomolecules in archaeologica l related mammal teeth between - 30 ka and 5 Ma in age and
ue on artifacts' mterpretation, J. Field Arch T hese innermost chambers were not systematically in
J. Archaeo/. Sci. 15, 83-97. aeol. 22, 97-1 IO. heated flint. ESR dating can provide a real advantage over
Tuross, N., Barnes, Y. & Pott vestigated during the first maj o r excavation by the Joint
Herr, J. C.,_ Benajamin, D. C. & Woodward s, R., 1996, Protein identification other methods, because the teeth and flint represent artefacts
, M. P., 1989, of blood residues on Mongolian-Russian-American Archaeological Expedition
Detection of human origin of blood experimental stone to ols, with archaeological significance, unlike some methods which
on tissue, in Allen, J Archaeol. Sci. 23, 289-296 (JMRAAE) in 1995, but constituted one important focus
R. 0. (ed.), A rcha eology Chemistry, Wash . date only stratigraphically associated sediment (Blackwell
ington, DC: Vargas-Sanders, R. & Orti & during the 1996 and 1997 T sagaan Agui excavations
American Chemical Society, 39 4-40 z, E., 1995, Advances and Schwarcz 1993a). ESR dates have compared well with 14C
6. pe specti es in t e analysis (Derevianko et al. 1996, 1998).
Hyland, D. C., Tersak, J.M., Adovasio, J.M. & Siege of proteins from projectile dates for samples ranging up to 20-25
l, M. I., pom ts: b10chemical stud ka (Grun 1989), and
1990, Identification of the speci es ies of haemoglobins, paper with thermoluminescence (TL) and 230Th/234U
of origin of residual prese ted at the 1st Sym dates in the
blood on lithic material, American
posium of Archaeological 30-200 ka
Antiquity 55 ' Che 21stry, Division of Hist range (Blackwell et al. 1994; Blackwell 1995).
104-112. '! ory of Chemistry, 209th ACS
National Meeting, Anaheim
Kooymn, ., Newman, M. E. & Ceri, H., 1992 , U.S.A.
, Verifying Vargas-Sand rs' R., 01ti
_
reliability of blood residue analysis z E. & Martinez, R. D., 1995,
on archaeological Dete . The Tsagaan Agui Cave deposits
tools, J. Archaeol. Sci. 19, 265 -270. nat10n of ancient hemoglobin by biochemical
and 1
Lopez, L., 1995, Guena y mue1te unol ogical analysis, paper presented at Xllf In the eastern Gobi Altai range, in Bayan Hongor
en Tenochtitlan International Congress of aimag,
. Prehistoric and Protohistoric
Descubrimientos en el recinto de Mongo lia (N 4442'32.6", E
las Aguilas, Sciences, Forli, Italia. 10110'08.8") lies T sagaan Agui
ArqueologfaMexicana 12, 75-77. C ave (Fig
. I). The cave occurs in lightly metamorphosed,
Lopez, L. & Mercado, V., 1996, Dos esculturas
Precambrian
dolomitic limestones, replete with karst
de
Miclantecuhtli encontradas en el lan dforms
recinto sagrado de (Fig. 2; Derevianko et al. 1996). Today, a few semi
Mexico-Tenochtitlan, Est. Cult. Nahuatl n omadic
26, 41-6 8. herdsmen use the Gobi Desert's northern borders to Fig. I. The Tsagaan Agui Cave archaeological site,
gra ze their flocks. At
times in the past, however, the area has Mongolia. Tsagaan Agui Cave occurs in Bayan Hongo1;
support
ed much lusher vegetation and a more diverse fauna southwestern Mongolia, in the edge of the Trans-Gobi Altai
(Derevianko et al. 1996, l 998).
near the Gobi Desert.
66
Proceedillf?S of the 31" International Svmvosium Oil Archaeometrv. Jerem. E. & T. Biro. K. feds. I. Arrhaennl"l'S<-ArrhnPnli11<>11n RAR - rPntr l'"11r SPr I
,, ..
Blac"1vell, B. A. B., Olsen, J. W., Derevianko, A. P., Tseveendori, D., Sk11111e1, A F. R & Dwyei; M.
. Mongolia
Dating the Paleolithic Site at Tsagaa n Agui,
ESR (Electron Spin Resonance)
24 25 26 27
21 22 23
19 20
d
In 1995 and 1 997' the cave's lower urotto y .
. ,,, . ield
er th a ne an
extens1ve stone tool collection typologically earli
. .0tn -300
os1ted fr
elsewhere within the Tsagaan Agui complex,
.
as the I . AA-23158 33.84 0.64 ka.
.300
. . 11ItiaJ
sed1mento1og1cal analyses hinted.
2, the upper beds are mainly clasfI c se
In Test Pit
dim
with secondary mmerals, such as gypsum and carbo ent
nat es, a n
.
-400
. ct
.
limestone eboulis, while the lower beds contam
if Iner cl
Ienses mtercaIated with coarse sand layers (Table
l F' g ay
.
n :lyses
3;
a
Derevianko et al. 1 996). Palynological and fauna!
,,....
e
have not yet been completed for these layers, n .
"'
o1 h a s the
detailed sedimentological analysis.
-500 "'
_,
4l
m the Tsagaan Agm excavations. Although all artef g .J::
acts were
1y classified in the fi eld, only a small fractio ,&>
; .. 0
prelimman
n h ve J:l
"'
been th oroughly analyzed. Preliminary data from
the l "V Burrow -600
.
-600
excavat ons at Ts gaan Agui reinforce several
gene ::i 11C Date
TL Date
conclus10ns draw m analyzing the archaeological
mate rials
+ ESR Date
from the two prev10us field seasons:
1. Stone
-700
raw material, mostly jasper and o
. ther
t Preferred
-700
cry tocrystalline quartz minerals, appears exclusively
m LJ . Io 1'
Upper Plioconc/ Lower
havmg been obtamed within just a few hundred metres
Plcmoconc pcbblr of e
cave entrance. m
-
tcmcc deposits
-
0 1
t
2. A stratified cultural sequence representing the
Lo''r Prc::i111hn;111
dolrimiu:: . . late
llTHCSIOl\C :
prehi tonc early Bronze Age through Middle Paleolithic h
as
at the T sagaan Agui excavation in the main chambe1:
Fig. 3. Stratigraphic profile for the northeast wall
Upper Prcca111hrun mwmorphic
been 1dent1fied.
Quatcrnm scrl 1mcnt from translocated from their positions to this stratigraphic
profile. Where
D 3. In the deepest strata, recovered tools consist mostly of The locations for the dating samples have all been
sandstones and ,olcanics sa1ra :md lo" s.11ra tl!rraccs
. . in relativ e depth between the
been adjusted to compensate for dif e
f rences
fl ake s rapers, compnsmg only approximately 4% in the lithic necessary, their positions relative to the datum have
units is correct . QT40 and
1Uffs
v
G:Zl Upper Jurass1c/Lo\\cr (rel"'
bas.11ts and
"'
0Fault collect10n from these horizons. sampling location and this profile, in order that their
positions relative to various stratigr aphic
14C
4.
4 in Square A24, while all the
from the lower part of Stratum
Flakes were derived from both prepared platform QT41, plus associated sediment samples were collected
Upper CrctJccous pebbles. r7.7l Siliceous \C1ns ng layers, and all the TL sample s from lower in the sequen ce.
"Levallois" (sensu Okladnikov 1986; Alekseev 1990,. dates so far obtained come from higher in Stratum 4 or overlyi
Derev1'anko l 998) and polyhedral cores. Primary reduction
0 0
boulders. and sands1onc
portion
was deposited from at 33 1 ka, althoug h the lower
The six AMS dates and two ESR dates confirm that Stratum 4
. ve1y much older (modified from Derevianko et al. 1998).
Fig. 2. T he geo l ogic setti11gforTsagaanAgui. occurred ots1de the cave, principally at the raw material source. could possibly be as old as 38 ka. Deepe r units may be
.
Lightly metamorphosed Precambrian dolomitic limestones The lithic r duction waste products litter the limestone
. . ed below.
in addition to the ESR determinations describ
contain many karst features, i11c/11di11gTsagaanAgui Cave massif contaimng Tsagaan Agui Cave. S urrounding many found. The plane was made on a large spall, the edge trimmed
. 3 and the upper part of 4 average d 33.00
The dates for Strata
(modified from Derevia11ko et al., 1996). Jasper cobbles and boulders that outcrop just above the cave by removing several large flakes. Made on a massive flake,
1 .2 1 ky BP. These are associa ted with a typolog ically
e trance, l ge primary flakes and smaller chips indicate in
. the combination tool combines a sidescraper with a spur. The
-Upper Paleoli thic flake and core tool
transitional Middle
In 1988 and 1989, joint Soviet-Mongolian expeditions sztu educt10n. Detailed contour and scatter density mapping spurs found in this site seem unique to the Mongolian
cultura l
excavated a trench 1 6 m long and 2-6 m wide s panning the drip
reinfor ces the interpr etation that
of this workshop was completed in 1 996. Paleolithic, mainly because they are produced in several assemblage, which
below Stratum 4 are substan tially earlier,
of the cave'sinclined entryway In Strata 4 and 5 in Test Pit 2, 46 Paleolithic artefacts were d ifferent ways. Essentially, the collection resembles a materials occurring
line along the south
p rofile in that
thic in
margin
perhaps early Upper Pleistocene/Middle Paleoli
(Derevianko & Petrin 1995). In 1995, the north fou d, icluding two core-like artefacts, several spalls, a Levallois-like transitional Middle-Upper Paleolithic
.
and 12
trench was extended 50 cm to the north and 2.0 m east into the
for the Strata 5, 11,
0ne
affinity. Preliminary TL dates
partial dihedral blade 29 flakes, and 12 d e'b1.tage pieces. assemblage (Olsen et al. 1 996; Derevianko et al. 1 996, 1 998).
might
sample recovered in the Tsagaan Agui excavations. A wide Since the early 14C dates indicated that some layers
was expanded to the east and west in order to determine the The smaller flakes have damaged striking platforms, while the
or one
limit, ESR dates were attemp ted f
range of mammalian and avian species has been identified exceed the 14C dating
maximum depth for the deposits bearing cultural artefacts and lger ones show defined residual striking platforms. A to ol
o und in
. . sp. molar and one cervid cheek tooth, both f
to resolve the degree to which post-occupational ro offall has kit, cons1stmg of a plane, two burin-like artefacts a notch thus far, many with paleoecological implications, such as Equus
'
2 in Square A24, within the lowest 5- 1 0 cm of
affected the underlying sediment. In 1997, JMRAAE linked the made on an oval flake, and a combination tool, was also biso n, gazelle, horse, and ostrich. In 1997, additional organic Test Pit
Stratum 4 (Fig. 3; Table 3).
original Soviet-Mongolian soundings made in 19 88-89 with samples were collected from Ochotona (pika) and Alticola
our own excavations from 1995-96 to yield a continu o us
Stratum Age Sediment Type (vole) nests near the cave, since the 1 996 samples yielded
mostly gray sandy silt 'With gypsum crystals, ESR dating
_
l Holocene
longitudinal profile through the cave's main chamber down to ages [:S 1 .30 0.14 ky BP (GX-22673)]. JMRAAE staff are
carbonate crystals, limestone eboulis
the bedrock floor. Bedrock and large blocks of dolomitic debris currently analyzing these and other rodent middens from
2 extensively
occur at depths below 4 m beneath the present surface i n the
Holocene? mostly gray sandy silt with gypsum, excavations at Chikhen Agui Cave to assess paleoecological The theory underlying ESR dating has been
carbonate crystals, limestone eboulis re (see
cave interior. The open chimney in the main rotunda roof and
ical and archae ologica l literatu
Variation. described in the geolog
3 dark gray to grayish yellow sandy silts with fossil tooth
sporadically active stre an1s passing through the cave complex
Late Pleistocene recent review in Blackw ell 1 995). Since the
-8
In 1996 and 1 997, Tsagaan Agui 's principal inner chamber
angular limestone gravels and gypsum crystals
enamel acts as a natural dosimeter recording its total
Was excavated. Artefacts there suggest that that late Neolithic
9
itself have caused e r osional events, which have profoundly total
the cave. Stream activity seems most prevalent in units at and . stone slab there may have served as an altar.
where the dose
while exactly when the chimney o pened is Six AMS 14C dates have been completed for the cultural pennits one to date a tooth. For tooth enamel,
below Stratum 3, n pertains:
rate, DS(t), varies with time, more comple x equatio
currently difficult to assess. Table 1. Stratigraphy in Test Pit 2. sequence in Tsagaan Agui's main chamber (Table 2; Fig. 3),
68 69
Blackwell, B. A. B., Olsen' J. W., Derevianko, A. P., Tseveendori' D "ki'
., ..,, n11e1; A . F. R.
& Dwye1; M.
ESR (Electron Spin Resonance) Dating the Paleolithic Site at Tsagaan Agui, Mongolia
calculated
based on the radioactive element concentrations in ery recently. L RElX ESR spectrometer with a microwave frequency (EErnl (EErni (EErnl (EErn)
JEO
el and associated sediment, bone, and dentine. power, under a 100 kHz field
A. QT40, 96TA1, Cervid
th e enam EU a es represent the
minimum possible . of 9.445 GHz at 2 mW QT40enl 0.21 13.63 5.83
Th e calculated ESR age strongly depends on the estimate d from a given accum ESR age denv e<J of 5 mT/T using a 0.3 second time constant.
ulate d dose, while RU modu lation 0.02 0.02 0.02
dose rate, Dex1(t). Although Dex.Ct) is best m easured by ages are usually B.
external maximum. Alt th Spectra were scanned over 50 mT centered at 360 mT with a
QT41,96TA3,Equus:
hough a coupl ed e
TL or c dosimetry at the oiiginal sampling sites (Blackwell et
. ESR-230 Th/234 QT4len l 0 II 3.51 4.67
calculat10n can give
the exa
ct up take param_eters
U ag e 4.0 minute sweep time. The receiver gain vaiied from 10 to QT41en2 0.08 3.15 3.15
(e.g., Grun&
al. 1992, 1994), that was not possible for this initial study. McDe1mott 1994) _ 400 d epending on signal intensity. QT41en3 0.08 3.35 5.83
' these teeth contam so little u (see b
Since the teeth were collected before the excav ators d ecid ed elow) 0.07 2.63 2.63
After powdeiing, associated sediment samples, and each
QT4len4
that no succes sful 2i0Th/234U
has y e t been com QT4len5 0.12 5.03 3.35
pleted.
to try ESR dating, neutron activation analysis (NAA) was dentine and enamel subsample were analyzed by NAA for U QT41en6 0.02 5.83 3.66
us ed to measure the radioactivity in the asso ciated sediment with delayed neutron counting (DNC) for 60 seconds, Mean 0.08 3.91 3.88
The analytical
for this expeiiment. Dexi(t) was then calculated from
method 0.03 1.23 1.17
following a 60 second irradiation and a 10 second delay. For
Typical errors' 0.02 0.02 0.02 0.02
radi oactive dosimetry (e.g., Nambi & Aitken 1986). This For tooth ename
l the ESR datmg . K analyses, s amples were irradiated for 60 seconds, and after
NAA detectio11 0.01 0.01 0.01 0.01
protocol is well
m e thod requires that sediment water content be known, since established (Blackw
l
e l 198 9) The teeth 24 hours' delay, counted for 20 minutes using a y counter, limits1 0.02 0.02 0.02 0.02
water weakens all radiation dose rates, particularly the _ were p hotographe
and sketched in six ff .
di erent orientatJO d while for Th, samples were irradiated for 60 minutes and l
Typical NAA detection limits and errors depend on sample mass and tissue type.
ns m order to record the
analyzed for 20 minutes on a y counter after a week's delay.
SamEle Numbers Sampling Location The sediment water content was measured by weighing the Table 4. U Concentrations is T sagan-Agui Teeth.
1
Collection
Type Quadrat Stratum Horiz.on AMS wet sediment, then heating at moderate temperature for
>ODatum < Surface2
14CLab within Stratum 14CA several weeks in order to diive off the water, and reweighing The sediment surrounding QT40 and QT41 generated the
(cm) (cm)
TA97-6 Wood A20 3 (kyBP) the samples. external dose rates, Dex1(t) =0.957 0.121 and De,,(t) =0.963
334 40
AA-26586 charcoal 0.931 0.122 mGray/y respectively (Table 5), while the mean Dex.Ct)
TA97-15 0.065 ESR age calculation for Stratum 4 based on nine samples was 1.039 0.159
Wood A20 2 lowest
AA-26586 430 33.78 mGray/y, which is not significantly different from that
charcoal gravel layer just above 3
0.59 By fitting the data to a saturating exponential curve and immediately surrounding QT41. For layers 1 through 4, the
TA97-17 wood A120 3 surface
436 a ssuming a 1//2 weighting, the Vfit program calculated the AS mean Dex.Ct) measurements differ little from layer to layer,
AA-26586 charcoal 33.50
values and their associated errors. The ages and their suggesting very consistent sources for the cave sediment.
0.60
TA97-16 Wood A'23 3 top associated errors were calculated assuming the ku-value for Therefore, although only one sediment sample was available
3 55 80
AA-26586 charcoal 33.84 enamel to be 0.15 0.02, its density, 2.95 0.05 g/cm3, and associated with QT41, rather than the usual 6-10 to estimate
TA97-18 wood A'20 4 surface
0.64 no Rn loss. The initial activity ratio, (2 34U/238U)0, was assumed an average sedimentary De,,(t) the calculated value does seem
AA-26586 probably derived from 3
390 30.94 to be 1.2 0.2 in the cave system. Dose rate calculations to provide a good estimate for the whole stratum.
charcoal
0.48 assumed some attenuation due to water in all phases, The associated sediment for QT40 and QT41 showed no
TA97-12 Wood A26 4 near surface 27 4 69 including the enamel, and some attenuation of p dose due to significant change in water content when diied, probably
AA-26586 charcoal 32.96
density and sample thickness. Water in sediment was assumed because the sample bags in which the samples were stored
0.67
-
1 Relative locations
shown on Figure 3 to average 10 10 wt% over the time the tooth was in the after collection could leak air or moisture. More sediment
2 Present ground surface ground, while the cosmic dose rate was assumed to be 0.0 samples, collected and properly bagged on site in August,
Table 2. AMS uc Dates.for the Main Chamber' Tsagan-A g 1.. mGray/y, because the teeth occmTed more than 10 m below 1997, were tested now for water content. These averaged
Lil, Mongo 1a.
the ground surface (Nambi & Aitken 1986). < 1-2 wt% for all samples. Although the region is quite dry
The programs calculate standard errors using standaid today, in the past, the cave has seen intermittent stream flow,
ESR SamEle Numbers Location1
Tooth radiation physics. The largest error found in ESR enamel age as suggested by the sedimentology (Olsen et al. 1996;
Analysis Ca talogue Accession Layer Square Depth Species calculations usually deiives from those associated with the
Type Derevianko et al. 1996, 1998). Strata 3 and 4 do appear to
(cm) accumulated dose, Ar., and external dose rate, D<,,(t), while have expe1ienced some stream flow, although this must be
Qr40 96TA1 SS44 Bottomof4 A24 370 Cervid cheek tooth some 30 other factors produce fairly insignificant errors. To confirmed by detailed sedimentological analysis. Moreover,
Qr 41 96TA3 SS86 Bottomof 4 A24 370 Equus sp. molar test the robustness in the calculated ages, the effects from duiing the late Quaternary, at least one laige playa lake
Relative locations shown on Figure 3 Variability in Rn loss, sedimentary water concentrations, and formed in this region, suggesting moisture may have been
Table 3. ESR Enamel Samples ftvm Tsagaan Agu1 somewhat higher in the cave sediment duiing such high lake
Dex1(t) were all analyzed, using the data from typical sample,
, Mongo 1.
. w.
QT4len2 (see below). stands. The sedimentary character and prolonged dry season
70
71
'''
60
40 ,.,, 1
_,,C.,
on""
, ,ce,,
n,,_,trati_n.s.......
, External Das
e Accumulated ESR Ages
'DI K
--_
9ITA22 2b
o.117 . 4.5
1.34 5.05 2.95 0.01 I.003 106t- L20 3.6 4.1
01 : 2b 170 5.53 2.86 0.00 1.042
<( QT41en2 1.84
1:108 36.7
; 4.33 2.95 0.01 L020 LOS4 35.78 33.1 35.1
34 as reported 9ITA2S .57 5.33 2.75 0.01 l.198 1.274
QT4len2
grrA26 3 1.02 3.9 4.3 4. 7
,/ 2b 1 64 4.88 2.70 0.00 0.972
9ITA27 . 1.033 35
2b O.Ol
9ITA28 2 b __J,::61:--:5-- .
3:
:-.. 2.:. 9 7 __:
0,0 1-- L048 .--14--l!._ 42. 56 41.2 42. 8 43. 7
4 -- : c::
'.___ QT41en3
Mean 2b 1.95 5.08 2.86 0.01 L047 1.63 5.1 5.5 5.7 *RU
1.rn ""
32 Typical 0.74 0.39 0.10 0.01 0.151 0.107 OT41en2 LU
error 9TfA 6 1.71 36.61 33.9 35.7 36.9 os reported
9ITA7 2:. 10
6.29 2.26
7. 58::
0.949 l. QT41en4 EU
':'- ----C:.: --' :
---:= ___;2::.!.18 _:::_
L 03
J __jl o l.46 4.1 4.5 4.9 28+--------
:oii-
_ __
1.91 20 30 40
__
4.5 5.0
__
1.38 4.0
___
0. ESR ages average 33.8 4.0 ka (EU), 36.14.3 ka (LU), and assuming linear U uptake model
LU 5. 2. wt%
radon loss in all tooth phases, Rn,...,
0. %
=
3.20 0.70 mm
dentine thickness, ..., Calculated using external dose rate derived from associatE thickness removed from inner enamel, i/Jin= 0075 0.025 m
!JI 0.02 ppm significantly from one another at the 95% confidence level,
sedimentary U concentration. U..i sediment 91I'A4 (Table
, 4). thickness removed from outer enamel, 0,,., 0.025 m
3.83 0.14 ppm 0.044
sedimentary Th concentration, Th..i however, the U uptake model assumption is not critical for
sedimentarv U concentration, K..i
3.04 0.12 ppm D.,.1jt) = 0. 963 0.122 mGray/y dentine thickness, i/idcn 3 . 20 0.70 mm
these samples. In several dry environments, EU ages hav e sedimentary u concentration, u.ed 1J I ppm
=
proven to be the best uptake model (e.g., McDermott et al. Without Qr4len3. 0.14
he obi even during wett er 1 993). Therefore, the sedimentary U concentration, K.ed 3.04 0.12 ppm
that would have occ ed in t G urr EUrepresents aage probably
cosmic dose rate, D00,(1) 0.000 0.000 mGray/y
nt saturation. Since the reasonable median age for this tooth. For the two teeth, the Table 6. ESR Ages for Tsagaan Agui, Mongolia
periods, how ever, arg u e agai consta
nst
centra tio n is likely to have been combined mean age is then 33.2 3.8 ka, assuming EU.
time-averaged wate r con
it w as assumed to be 10 Because water in the sediment slows the incident radiation, require that sediment was completely saturated for at least actual external dose rate were 0.240 mGray/y higher, then the
higher than the 2 wt % seen today,
ns. a change in the water concentration over time significantly 70--80% of the time, or - 30 of the last 50 ky, a scenario calculated ages would decrease by 6-7 ky. More sediment
10 wt% for the age calcula tio
ced only one subs mple and a changes the external dose rate, D.x,(t), and, therefore, the that seems excessive for the nmthem Gobi. T he evidence samples around each tooth will be analyzed to ensure the
Since QT 40 was small, it produ
34U analysis to assess the U calculated age, t. To test the effect of the sedimentary water currently available suggests that the maximum time-averaged accuracy in the ages. Coincidentally, however, the sediment
could not be submitte d for 2fh/2
chron age be calcu lated. The assumption, ages were recalculated using different water Water concentration did not likely exceed 2 0 wt%. samples closest to QT40 and QT41 yield the lowest D.Jt)
uptake model, nor a n iso
ges fo r QT40enl and QT4 l concentrations. If the cave sediment had been much wetter, o n T he external dose rate calculation strongly influences the values seen in Stratum 4 (Table 5), but the other sediment
similari ty in the c a lculat ed a
calculated ages. T he inhomogeneity in the sediment close to samples were from more than 1-2 m away from the locations
QT4 0 ages are reliable. The QT4 0 average, since 50 ka than that seen today, the calculated ages
(Table 6a) suggests that the the teeth
36.9 4.2 k a (LU) would be significantly older than those reported herein (Table 5) is sufficient that the actual external dose where QT40 and QT41 were found. Measurements for
ages are respectively 30.0 2.9 k a (EU),
rte, D.,,(t), Stratum 5, which does contribute some dose to QT40 and
, errors of 1 0-14%. (Fig. 5). If the sedimentary water concentration had average d that the teeth experienced may have been
and 46.6 6.3 ka (RU) it w h
Sign ificantly different than that calculated herein, causing QT41, have not been completed. T he TL ages for Stratum 5
es, the internal dose rates, 30 wt%, instead of the 10 wt% assumed here, the EU ages
For QT41, the accu mulated dos
consist ent for all subsamples would increase by 2 7%, the LU, by 30%, and the RU, by 3zo/o. p ot e ntial inaccuracies in the ages (Fig. 6). If Dexi(t) were hint that it is much older, and may have been deposited under
the EU, LU, and RU a ges ar e
actually 0.240 mGray/y lower than that calculated here, the very different conditions. Until sediment from Stratum 5 can
en3 s eems rather different, wh en Despite the high lake levels intermittently in the past, however,
(Tables 4, 5). Although QT4 l calculate
nsid ered, it does not d iffer to maintain 30 wt% sedimentary water concentration would d ages would increase by 9----1 2 ky. If, however, the be analyzed or in situ dosimetry can test the modem D.,,(t),
its associated err or is co
72 73
.
Blackwell' B. A B. , Olsen, J W. , De11!vwnko, at TsagaanAgui, Mongolia
A. P., Tseveendori, D., Skinne1; A. F. R. & Dwyei; M. Spin Resonance) Dating the Paleolithic Site
ESR (Elec tron
EU
3001
A. Conclusions
I
---- -
Sediment
j
1
D.,,,.1.,y(t) d
D int.eu( Din1 .reu(t}1
Due to low U in the enamel, few signi
ficant . saniple (mGrav/ ) (mGray/ ) (mGra i )
250 .
QT40<liffe ren /\ QT40 96'TA1. Cervid:
exist between the model ages for QT41
ages 0.125 0.004 0.005
smn1 ar to those for QT41. Therefore, the , 0.257 0.017
average E
0.037
o. 957 0.001
ag
Qr 40enl 0.003
4
0.017 0.004 0.008
33.2 3.8 ka, provides a reasonable age estim 0.121 0.006
0
200 ate or Stra
_::.
"-"
4. Small changes in the time-averaged sedime
ntary . J3 4 , 96TA3, Equus: 0.004 0.004
content would not cause significant diffierences . 1Uo1s QT 1 0.020 0.055 0.009 0.027
O. 963
ag , if
m the qr 41enl 0.025 0.004 0.004
0.007
overesti e '
(j) 150 however, the external dose rate was under- or 0.963 0.015 0.052
0.004
(Jl ma1e4 QT4ten2
0.016 0.027 0.007 0.0 1 3 0.004
RU
<( here, the ages could change significantly. 0. 963
* QT41en3 0.020 0.004 0.004
0.013 0.043 0.006
LU Further ESR and associated analyses may
. . c1anfy
. sotne QT41en4
0.963
0.010 0.048 0.004 0.005
0.099
EU remaimng questions. If obtainable ' 230Th /234U 0. 963 0.021
100 0.005
analyses QT4ten5 0.087 0.002 0.042 0.000
show w hether U leaching has occurred from the e lllay 0.963 0.004
. namel QT41en6 0.007 0.029 0.003 0.004
OT41en2
sediment analyses should help to establish the pre . '1oic 0.963 0.014 0.061
QT41 mean
0.002
4
reported cise ext 0.002 0.002 0.002
dose rates for Stratum 5, which does influence teet fr e
Typical
0.004 0.003
h o lllal
50 0.122
m the
error
f-t--1 lower Stratum 4. We have now collected severa -
EU assuming early U uptake model
1 more t "lll L Abbreviations:
specificall
. y for ESR dating from Strata 3, 4 and so e LU assuming linear U uptake model
me Iayers
0-;t-r-r-'T..,-,-,-fr-r-T-r-r-r-,-
-, lower m the sequence. RU assuming recent U uptake model
External dose rate, y component
0 400
Daxt.r (t)
800 1200 1 500 2000
Compared to the 14C and TL ages, the ESR ages
Dext(t), External Dose Rate (Groy/y) . .
consistent with the stratigraphy (Fig. 3). Interest '
are tota)J
Y Dmt,en(t) Internal dose rate from enamel
mg1Y, how.
4
ever, the 1 C and ESR ages combined with the TL Dim.,dcn(t) Internal dose rate from dentine
ag es do mGray/y
cosmic dose rate, D"""(t)
Fig. 6. The effects from variations in the external dose rate 0.000 0.000
suggest a great hi atus m deposition between Stratum 1 Calculated assuming
' 4 and
sediment density, p .,d
Dex(t), on the ages for QT4I en2 . As the external dose rate 2. 66 o. 05 g!cm3
f cave
Stratum 5, perhaps indicating a prolonged period
o wt%
ru g/cm3
external dose rates . Smee the sediment associated with .
chi mne above the main gallery, which would have allow 3 Calculated assuming dentine density, Pilm 2.75 0.05
QT40 w s imilar to that for QT4I, effects on QT40 would . ed
water in dentine, Wden 5. + 2. wt%
wt%
much higher sediment volumes to enter the cave.
be s1m1lar. A small change in DeJt) will cause the tratum 4 rpresents
layer deposited in the Middle Late 23 Calculated assuming Rn loss, Rn1edh 0. 0.
wl/O
3.8, but possibly as early as 39.2 5. l ka, during
mGra /y, a 25% increase, decreases the ages by I 9-2I%, water in enamel, W.., 2. 2.
mm
?' Oxygen sotope Stage 3 and continuing until 33.0 0.7 ka
wzle decreasing DeJt) to 720 mGray!y, a 25% drop, enamel, <Poo 0. 70 0.10
(cf. Martms n et al. 1987). At the time, glaciers would !\ave 2,3 Typical thickness ranges:
. 1.10 0.10 mm
mcreases the ages by 28-32%, depending on the uptake been retreatmg from the continents in the northern hemi
0.06 0.02 mm
model . The data plotted here were calculated assuming: shere.' cusing . dramatic climate changes in this are a , as inner enamel surface removed, fmner
0.02 nun
outer enamel surface removed, -.,,
0. 06
climatic Circulat10n systms realigned in response to rising
. dentine, tPdoo 3.20 0.70 mm
initial U activity ratio, (234Uf238U).,,...h
sea lvels and slight mcreases in global temperatures.
0.01 0.02 mm
1.20 0 20 cementum, <Poem
a
particle efficiency k 0.15
Certainly, vegetation and fauna would also have been
ual subsam pl e thickn esses, U concen trations in various tissues
Typical errors usually depend on the individ
0.02
enamel u concentration: u; 0.08 0.02 ppm
changing in re ponse to the changing climate, requiring that
4
dentine U concentration, Uden 3.15 0.02 ppm the hman res dents also make significant changes in their (Table 2) and their associated errors.
enamel water concentration, w
2. 2. wt% behav10ur. Dunng times in the late Quaternary, this area was
dentine water concentration, w 5.
mu:h .more humid than today, and might be more humid Table 7. ESR Dose Rates for Samples from Tsaga
an Agui, Mongolia.
J
2. wt
enamel density, Pen 2.95 'e)cm
3
0.05
dentine density, p"'
g/ cm agam m the future.
& Debenath, A.,
Blackwell, B. A., P orat, N., Schwarcz, H. P.
References
. 2.75 0.05
sedimentary water concentration, W...i 10. 10. wt%
tooth ename l: Comp arison with
1992a, ESR dating of
Acknowledgements
=
'E/cm.3
sediment density, p,.,.. 2.66 0.05
230 /234 dates at La Chaise -de-Vo uthon
enamel thickness, .. l.050 Alekseev, V. P. (ed.), 1990, Paleo/it i Neolit Mongolskogo Th U speleothem
. 0.100 m
thickness removed from inner enamel (Charente), France, Quatern ary Scienc e Review s 11,
0.075 0.025 m
' "wz We thnk M. Divjak, B. Lau, D. Conway, J.l.B. Blickstein, Altaiya, Nauka, Novosibirsk, 644 pp.
thickness removed from outer enamel
C. .Moms, H. Chan, G.Q.P. Xiang, and A. Dawoodi for
ourer 0.044 0.025 m 231-244.
Blackwell, B. A., 1989, ESR Dating of Tooth Enamel,
Blackwell, B. A., P orat, N. & Schwarcz,
dentine thickness, <1<n mm H. P., 1994, The
and
. 3.20 0.70
sedimentary U concentration V. assistance in preparing and analyzing the teeth. J. Johnson McMaster University, Dept of Geology, Technical Memo
1.31 0.02 ppm ologic al sites in the Great
sentary Th concentration, 77i:: 3.83 0.14 ppm J. 'Yelsh, McMaster University, performed the NAA analyses, 89.2, Hamilton, 234 pp. potential for ESR dating archae
74 75
Blackwell, B. A. B., Olsen, .
J. W.., Derevianko, A. P., Tseveendori D. Skin
ner A F R & Dwye1;
M.
' '
& Tseveendorj, D. (eds.), HAT FOR AND WHAT LIMITS CAN AMINO ACIDS
HAT WAY, W
;
etric U-seiie llb
date s
f or I ,
1998, Archaeologzcal Studies Car Neanderthal/early mod
em hominid Sit s
ried Out by ti1e JOtn
r . f 252-255.
es, N
ature M N CID RACEMISATION BE USED IN ARCHAEOMETRY?
A I OA
Russian-Mongolian-America
in 1996
.
n Expedition
Nambi, K. S. V & Aitk
en, M. J., 198
Izdatelstvo, Russian Academ
f 6' An CSAPO, J.,1 NYBERG, J.,2 MALMGREN, B.,2
Archaeology and Ethnogra
y of Sciences, Institute o
conversion factors for TL . nu al
phy' Novosibirsk ' 406 pp. and ESR datm
. 28, 202-205.
g, Arch
ae0 CSAPO-KISS, ZS.1 & CSAPO JR., J.3
Derevrnnko, A. P. & Petrin' V
T. 1995 , 1ssledovam.ya
Peshhernogo Kompleksa Tsag , P., 1986,
Okladnikov,
A.
an-Agui na Yuzhnom Fas Paleo/it Mongolu.. 1Facult of Animal Science, Pannon Agricultural University, Guba S. u. 40., Kaposvdr, 7401 Hungmy,
Novosibirsk, 433 pp. , Na y
2Geological Centre, University of Gateborg, Goteborg, 41298 Sweden,
Gob1skogo Altai a v Mongoli
i, Izdatelstvo, Russian
Academy of Sciences, Insti
Olsen, J. .,D erevianko, A. P. & Tse
6., Pees, 7624 Hungary
S., 6 .r Natu ral Sciences,
tute of Archaeology and . veendorj, D N atural Geography Department, Janus Pannonius University, lfjusag u.
Ethnography, Novosibirsk, 132
Invest1gat10ns in Tsagaa . 7 I faculty 01
n Agui Cave, m J
pp. . Archaeo I
. og;
, R., 1989, The ESR dating
..
Grun Studzes Carried Out by
the Joint R uss1
method, Quat. emary .. an-Mo
Amencan Expedz.t1on . o.
8 han.
.
1nternatwnal 1, 1-69. in Mongolia in 1995 :
Der VI. v wing prev ious attempts to use the extent of amino acid racemization (AAR)
for the dete1mination of the age of
8: McDermott, F., 1994, Open system modelling for A. P , lsen, J. W. & Tse alJko, A er re ie
Griin, R.
.
: veendorj, D. S.
eds.)
,
C t ogical samples contai
ning proteins, we present our own approach. Before starting amino acid racemization research, a
chaeolo 74-89,
U-snes and ESR datm
Publishing House of
the Institute of Ar arc aeol
g of teeth, Quaternary Scie
gy and rated for protein hydrolysis with ve1y low racemization of amino acids. The hydrolysis was pe1formed at
nce ethod was elabo
13, 121-125.
Ethnography, Novosibirsk
to the original (l 10C, 24 h)
m perature (J70C) for a short time (30 min) which caused very low racemization compared
Reviews
Maitinson, D. G., Pisias, N. G.,
Skinner, A. F. R., Blackw
.
ell B A B ., Chast
ne:v,,
hig
Hays, J. D., Imbrie, J., Moo een, D. E.,
re, J. M. & Mm, S. S., Shao, equ ently the D- and L-amino acid content of fossil bone samples of known age (radiocarbon method) was detetmined
'
: HPC af er precolumn
T.
dating and the . mproveme .
nts t derivatization. Based on the half-lives of racemization obtained and plotting the DIL ratio as a function
datmg tooth enamel by
orbital theory of the Ice Age ESR, Applied R . .
s: Development of a high ad1at1on & III acids, calibration curves were obtained which can be used for the age detemiination of fossil bone
:;
Isotopes 52. time for
v arious amino
0 to 300,000-year chronost
.
resolut10n
ratigraphy' les
amp in the range of 2000- 500,000 years. Another method is presented for the detemiination of age of textiles in the range of
Quaternary Research 27, 1-29.
J(){)-1800 years. This method is based on the dete1mination by amino acid analyser of the age-dependent alteration of amino acid
Cystine, methionine and tyrosine content decreased, while cysteic acid content increased with age.
com position of proteins.
Predicti on equatio ns were developed as linear regressions of age of wool based on cysteic acid, cystine and tyrosine content.
Introduction Recently Marshall (1990) established that the bones are not
reliable materials for AAR testing, particularly if they come
The subject of this paper is a survey on the possibilities of from a warm environment. The statement was based on
the determination of the age of archaeological samples differences observed between the age of the California bones
containing proteins based on transformations of the amino determined by 14C accelerator mass spectrometry (5000-6000
acids therein. The first possibility is to base the age years) and by AAR (50,000-60,000 years). Milford Wolpoff,
determination on the extent of racernization of the L-amino paleoanthropologist, expressed the opinion (cited by Marshall
acids which are the constituents of native proteins. Amino 1990) that many people currently regard AAR as "some kind
acid contents in fossil shell, bone and tooth samples from of joke".
early ages were reported first by Abelson in 1954. In 1967, Since various changes in temperature during the past and
Hare and Abelson reported that D-amino acids in fossils other conditions influencing dead biological organisms are
resulted from conversion of L-amino acids of protein. It was not well known, the reaction temperature of racemization can
found that the older the fossil the higher the D/L ratio and, only be estimated and not accurately determined. This is the
after a certain age, amino acids occurred in racemic form. The reason that in this study at first contents of D- and L- amino
ratio of D-allo isoleucine and L-isoleucine content in a acids and their ratio were determined in samples of known age
fossilised shell sample was found to be 0.32 and the fossil was (as determined by the radiocarbon method). These data were
estimated to be 70,000 years old, as reported by Hare and then compared with data obtained from the analysis of amino
Mitterer, in 1968. This is considered as the first application of acids in samples of unknown age. To make the comparison
am ino acid racemization (AAR) - or rather epimerization - in more accurate, the antecedents of samples of known age when
geochronology. analysed were the same as or similar to those of unknown age.
Subsequently, racemization of amino acids was used for All of the samples analysed in this study originated from the
age determination same Hungarian environment. The samples were excavated
of various materials containing protein
CWehmi!ler & Hare 1971; Williams & Smith 1977; Miller & from a depth of approximately 1.5-2.0 metres (the depth is
Hare 1980; Bada 1985; Taylor 1987; Csap6 et al. 1988, very important because of the temperature; the yearly average
1990). Isoleucine was given special attention because L 11C). The pH of the soil at this
. temperature at this depth is
is oleucine - depth in most cases was very near to neutral and never
containing two chiral centres - can be easily
separated
from the diastereomeric D-allo isoleucine by an exceeded 7.0 (the pH ranged between 5. 5 and 6.8). Therefore,
amino acid analyser. For the moment Miller et al. (1997) used approx. 150 fossil bone samples previously analysed by the
the amin
o acid racemization to determine the integrated radiocarbon method were collected from various Hungarian
temp erat
ure exposure during a long period of time. museums, and their D- and L- amino acid contents were
However some errors of age determination based on AAR determined. The D/L ratio was plotted against time, which
by Williams and Smith in 1977. Temperature,
Were reported produced a calibration curve. This curve can be used for age
PH, soil composition and various contaminants should also be estimation of samples of unknown age after their D- and L
c o nsider
ed when estimating the age of fossil bone samples. amino acid contents have been determined. The D/L ratio for
76
Proceedings
of the 31" Jnternatio11al Symposium on Archaeomet1y, Jerem, E. & T. Biro, K. (eds.), Archaeopress-Archaeolingua, BAR- Centi: Ew: Se1: I.
,.,..,. in archaeometry?
o acid racemisation. be used
s can amino acids and amin
hat fior and w hat limit
Csap6, J., Nyberg, J., Malmgren, B., Csap6-Kiss, Zs. & Csap6, J. Jr. w
Jn wIi
ay
at w '
amino acids
m ples The D/L ratios for various
2-3 amino acids was detennined for each sample and the hydrolysis and racemization have been carefully in
vesri Age of sa
ga ed by Val
mean value of ages estimated from calibration curves was It was concluded from the data of our experimen determin Glu Ala Ile
ts His Phe Asp
considered the true age of the fossil sample. rae of the racemiztion in the course of the hydro t cone cted
. the 14c
. ear)
In the course of these studies it was found that in the case acid catalyzd prtem hydr lys1 decreases in the fo oo method (y
-
U
of samples containing proteins with low D/L amino acid ratio ?
order: aspai:1c ac1 , glutamic acid, threonine, pheny
(fossils under the age of about 2-3000 years) it is not possible
_
alane, valine, _ _
stidine.
he rate of the racerni
la! a
ltine,
2200
0.138
zation 0.162 0.101
_
to achieve reliable data for their age. Another possibility for peptide bound amino acids m 6M hydrochloric acid b of 2800
etw 0.181 0.109
the age detennination is to base the measurements on the 110 and l 70C is 4-7 times higher than that of the fr
ee 3110
0.199 0.128
changes in the amino acid composition of the aged samples.
The basis for using the amino acids for age detennination is
acids. The racemization at high temperatures a
nd
reaction times is by 20-55% lower than at l 10C for ott
3240
4630
0.253 0.179 0.109
archaeometry - to determine the age of wool carpets and presented; none of the ratios lower than 0.1 or higher than 0.7 72400
radiocarbon method.
textiles (Csap6 et al. 1995). presented, in these cases, the accuracy of age determination
Table 1. DIL rati os for vari ous amm
.
o a
.
1 ; .
ds onc rn '. ng ages if fossil
samples determined by the
ino acid to reach 0.333,
In order to be able to measure the changes of the amino acid was doubtful. The half life of amino acid racemization, i.e. the _ at n .e. t e time re qu1- 1 d for the ratio of D- to L-am
composition and/or the enantiomeric ratio of the individual time required for the ratio of D- to L-amino acid to reach T he ha lf life of amino acid race nu z w , t samples during burying was
approximately 1 1 oc,
the data of Tab le 1 . T he temperature oif . J1e
amino acids in very small archaeological samples the use of 0.333, was calculated by interpolation or extrapolation from was calculat ed from
was between 5.5 an. d 6.8.
the surroundings of the bone
while the pH of the soil in
the highly selective and sensitive high-performance liquid the data of Table l , and is presented in Table 2. The tempera
As
n curves is presented finally.
chromatographic method (HPLC) is inevitable. Pre-column ture of the samples during burying was approximately 11C, The applicability of calibratio
Amino acid Half-life (year) of whic h was unkn own
ple the age
derivatization of a-amino acids using o-phthalic aldehyde while the pH of the soil in the surroundings of the bone was an example one bone sam
he
(OPA) and various thiols as the reagents leads to fluorescent between 5.5 and 6.8. ina edfrom the end of the Neolithic Age or from
and orig
Histidine 5500 g to the arch aeol og sts
accordin
formed derivatives, while another reagent, 9-fluorenyl-methyl From the data of Table 1 it is evident, that, D-His, D-Pl\e, beginning of the Copper Age
Phenylalanine 8500 acids and the followmg
chloroformate (FMOC-Cl) transforms *-amino and imino D-Asp, D-Glu and D-Ala contents can be used for the age yse d -"or L and D-am ino
was ana1 i -
8600
Tyrosine own sample was excavated from
acids to fluorometrically active derivatives. Both types of determination of samples which are 2-12,000, 3-20,000, results were obtained (the unkn
13500 ):
5-35,000, 8-70,000 and 10-80,000 years old, respectively.
Aspartic acid the pH of the soil was 6.50
derivatives have good chromatographic properties enabling
16500 a depth of 1.85 metres and
Serine
several amino acids to be separated and measured within one Age of samples older than 30,000 and 50,000 years can be
Threonine 17000 mg, D/LHis = 0.428
chromatographic run at very low concentration levels. detennined on the basis of Ile and Val content, respectively. L-His: 0.0697 mg, D-His: 0.0289
Glutamic acid 28500 n curve: 7100 years;
If the aim is the enantiomeric separation of amino acids, Data were corrected (reduced) with the D-amino acid content Age calculated from calibratio
Alanine 32000
chiral derivatization agent should be used which transforms of a fresh pig bone to eliminate the errors of analysis. SEM = 337.
Isoleucine 110000 mg, D/LPhe =0.254.
the amino acid enantiomers to pairs of diastereomeric de Concentrations of the D-form for the other amino acids were L-Phe: 0.0543 mg, D-Phe: 0.0138
Leucine 140000 n curve: 6950 years;
rivatives separable on achiral HPLC columns. The chiral 1-(9- negligible. However, all analyses were corrected for the small Age calculated from calibratio
Valine 180000
fluorenyl)-ethyl chloroformate (FLEC) reagent (Einarsson et concentrations present in fresh pig bone. SEM = 191.
mg, D/LAsp =0.182.
al. 1987a,b) and also the OPAfthiol reagent are suitable for Studying the calibration curves, it can be concluded that, in L-Asp: 0.1346 mg, D-Asp: 0.0245
the case of the D/L ratio being lower than 0.1, the D-amin
Table 2. Half-lives of racemization and ration curve: 6900 years;
this purpose if a chiral thiol is used in the latter case. Both Age calculated from calib
It epimerization of various amino acids found
types of reagents were successfully used in this study. If the acid content is too low and age detennination is uncertain. SEM = 465.
e in Hungarian fossil bone samples.
aim is only the separation of the isoleucine and D-allo is obvious that the calibration curves can be used for ag
of the
detennination most satisfactorily in the linear range (betw een sample is the mean value
isoleucine, no precolumn derivatization is necessary since
years The estimated age of the
0.1 and 0.5 where D-amino acids are present in w
eJI Sample 1 estimates vmy from 562 to 600 years. This mea n valu e has a
they are diastereoisomers, and the separation and quantitation s above estimates, or 6980 _
le and average 580 years. All four sets of confi nc limit
d
s and the 95% confi denc e mte rval
can be made by amino acid analyser without chiral selectors. detectable amounts). The optimum D/L ratio for each s amp standard enor Of 202 Year
e overlap, indicating that the estimates are not s1 ni.fic
antly
The prerequisite of a successful chromatographic analysis can be found by analysing the amino acids best suited for a g een would be 6554 to 7406 years. .
1 ts betw r
detennination. E.g., for fossil bone samples of 11,200 years
of different. We can be 95% certain that sample od includes the analytical erro
is the complete hydrolysis of the protein content of Of course the described meth
, 550 and 610 years of age. the effe ts on AAR of
age the D/L ratios for His, Phe, Asp and Ala are 0.682, 0. 473 method, but
archaeological samples. All hydrolyses were carried out using
344 years of age estimation by the 14C
6M hydrochloric acid. If the aim is the determination of the 0.271 and 0.112, respectively. In this case the D/L ratios
of Estim ates for sample 2 range from 297 to
re, pH and the com position of sol have been
ap, and we temperatu _
overl ld be
for 2 to 4 amino acids shou
of an d average 326 years. All confidence limits .
enantiomeric ratio, it is very impor tant to keep the extent of Phe and Asp are recommended for determining the age and 364 eIiminated. The D/L ratio
ed can be 95% certain that sample 2 is between 288 mea n valu e of estim ated
racemization at the lowest possible level. The effect of the samples; however, the D/L ratios of His and Ala can be us determined for each sample,
and the
yea rs of age.
reaction times and temperatures on the extent of the to confirm the estimate based on the ratios of Phe and As p.
79
78
archaeome try?
csapo,'J" Nyberg, J., Malmgren, B., Csap6-Kiss, Zs. & Csap6, J. li:
o acid racemisation be used in
limits can amino acids and amin
Iat wa) ' what for and what
'
In w1
ct con
between the data from the calibration curve and were approximately 10, 20 and 30 times as gre d w e ca
ce at at 50() .
differen
.
. 1589-1599.
14C d a tmg was generally negligible. However, our and 1700 years, respectively. Cystine conte I an
364 y e and decr easmg cystme and tyro sme
, Racemization of amino
Hare, P. E. & Abelson, P. H., 1967
m . cysteic acid
those fro nt d e e
reaA
ts
nt co 10% m 1600-1700 years The differences are conte n an accurate basis for
to provide
ere
due to diff di even o be exercised in 66, 526-536.
betl sh We or cloth. Howev
o ther calibration curves can be formulated for n er, caution sho uld
when expressed as a ratio of cysteic ac1 , Laboratory simulation of
Hare,P. E. & Mitterer, R. M., 1968
re sul ts,
on these cystine.
ment o n the basis of the methods described here. 1600-1700-year-old samples had a ratio more th wo l c
ion equations pre
sen ted in this paper.
ls, Carnegie Inst.
amino acid diagenesis in fossi
on
e predict
ir an 100 .
each env
antage of this method compared to the other as great as that of modern carpets. u sin t
working with wo ol fab ric of a different
t laboratory
at adv .
The gre
A d i11e ren either to Washington Yearb. 67, 205-212.
low size of sample required. 2-1o mg nalyse some sam
ples of kno wn age
sed w11h a ion of 1ibonuclease with
s the ve:r_ Both methionine and tyrosine contents deerea a Hirs, C. H. W., 1956, The o xidat
_i ld
methods g
n. . ou
0.1-1.0 mg protein, is sufficient for on n sh
ility of these equations
or to develop a set
performic acid, J. Biol. Chem. 219,
cont ammg In the 1750-year-old wool fibre and one of the 1 611-621.
e applicab .
siz e 600_Year.01
sam ple
t en samp1es.
th
AAR.
f
.ons appropnate forh
based on The radiocarbon age venY
wool fibres we were unable to detect the presence . dating: Great expectations,
pU
nin ation
. Marshall, E., 1990, Racemization
en
age det . of e1ther .
on met ho d can be used only for those samples tyrosme or methionine . The methiomne d des cnb ed her e we can only estimate the age
detennin
ati content of m etho Science 248, 799.
samples decreased to 85% of the original leve1 .
B th e
the can demonstrate with 100
% confidence
which are
than 80,000 years. By means of AAR 1945, Analytical chemistry
: pet, but we Martin, A. J.P. & Synge, R. L. M.,
m ore recent
. m Hl0-140 of e car een two car pet s, and we can
the me ans of Ile, Leu and Val) the age of the years, to 55% m 400-500 years and to a1 o 100 years betw & Edsall, J. T. (eds.),
of the proteins, in Anson, M. L.
y by
(especiall m st 1 'f th ere is inals if the
em onstrat
arpet is older than
100 years.
Advances in Protein Chemistry
mple tely ori gin al c New York, 2, 1-83.
Age est! mation of ol carpets based on decomposed at 1600-1700 years.
Miller, G. H. & Hare,P. E., 1980
, Amino acid geochronology:
cystrne and cyste1c acid content Linear regression equations, correlations and st dard Acknowledgements ix and potential of
. an Integ1ity of the carbonate matr
.
errors of estimate for the regressions of age on ammo a 'd Hoering, T. C. & King,
Ct
Fun d for Scientific molluscan fossils, in Hare, P .E.,
Jn accordan
ce with the statements in the previous chapter contents were calculated. Based on the correlatio
' ns and errors The financial
support of the National amino acids, J. Wiley &
K. Jr. (eds.), Biogeochemistly of
the age o
f ol textiles (100-1800 years) is not high enough t
d
_
from the
H arian National Museum and the Hungari n
ung : es 1mators, c rrelatron coefficients were -0.980 for cysteic
Bada, J. L., 1985 , Ami no acid race mization dating of fossil
recording equipment, in Colowick
, S. P. & Kaplan, N.
pplied Arts. Samples from contemporary wool acid and cystm , -0.990 for cysteic acid and tyrosine, and 13, 241-251.
bones,Ann Rev Earth Planet Sci
0., Methods in enzimology 6, 819-
Museum f 831.
g1 ate from th Domus furniture and furnishing 1988, Archaeological age
+0.970 for cystme and tyrosine. As a result of these high Csap6 J., P ap I. & Kolt6, L., mato grap hic deter mination of amino
carpets on . les containing protein Moore, S., 1963 , Chro
he city of Ka_posvar. New wool of Hungarian Merino correatrons bet veen inependent variables, very little could determination of fossil bone samp c recording equipment,
store in t _ :-
n and epimerization, acids by the use of automati
obtamed d!fe ctly from our experimental farm. be gru ned by usmg multiple regression. However, the average based on amino acid racemizatio
sheep was . J. Biol. Chem. 238, 235-245.
to the age determination based on amino acid Anthropologia Hungarica l , 67-86. , F., 1955, Determination of
Osono, K., Mukai, I. & Tominaga
Jn contrast of estimates based on cysteic acid, cystine and tyrosine was
Csap6, J., Csap6-Kiss, Zs., Kolt6, L. &
. P app, I., 1990, Age
on, in t is case it is not important to minimise the found to have a standrud error of approximately the same size and peptides, Nagasaki
racenu zati based on the ratio cystine and cysteine in proteins
ation durmg the hydrolysis. It is, however, most as that based on cysteic acid alone. Use of the average of determination of fossil bone samples
racenuz _ 90, Iggakai Zassi 39, 156.
of amino acid racemization, Arch
aeometry
t to void_ t h e oxidation of the amino acids. In stIm ates allows compensation for errors of analysis for ics in archaeology and
imp ortan _ . Taylor, R. E., 1987, Dati ng techn
ce with this the hydrolysis with 6M hydrochloric mdiv1dual amino acids. Birkhauser Verlag Basel, 627-635. . 59, 241-251.
Csap6, J., Csap6-Kiss, Zs., Folestad, S., Orwar, 0.,
accordan Tivesten, paleoanthropology,Anal. Chem
carried out at l l0C with the complete exclusion 1971, Racemization of amino
Vehmiller, J. F. & Hare, P. E.,
P redicted age, standard errors and 95% confidence limits
acid was
A., Martin, T. G., & Nemethy, S., 1995 , Age estim ation
pheric oxygen. Careful optimisation of the reaction for estimates of age of samples 1 and 2 are shown in Table 3. ce 173, 907-914.
of atmos cyste ic acid conte nt, acids in marine sediments, Scien
has shown that the usual hydrolysis of 24 hours of old carpets based on cystine and G., 1977, A critical evaluation
Williams, K. M. & Smith, G.
Sample 1 estimates vary from 562 to 600 years and average
conditions 20.
icient and a reaction time of 48 hours is necessar Analitica ChimicaActa 300(1-3), 313-3 acid racemization to
is not suff 8 yars. All four sets of confidence limits overlap, of the application of amino
e complete hydrolysis of the proteins in th md1catmg that the stimates are not significantly different. We Einarsson, S., Josefsson, B., Moller, P. & Sanchez, D., 1987, metry, Orig. Life. 8,
t o achiev l amines geochronology and geothermo
can be 95% certam that sample 1 is between 550 and 610 Separation of amino acid enantiomers and chira
textiles.
derivatization with ( +)-1-(9- 91-144.
using precolumn rmination of methionine in
Am ino Acid ed-ph ase liquid Yoritaka, T. & Ono, T., 1954, Dete
Sample AA content Estimated age Standard error 95% C.L. fluorenyl)ethyl chloroformate and revers saki Iggakai 'Zassi 29,
proteins and peptides, Naga
g/100 g year L.L. U.L. chromatography,Anal. Chem. 59, 1191-1199.
400-405.
Table 3. Estimates and confidence limits for ages of two cmpets of unknown age.
81
80
THE PROBLEM OF CHRONOLOGY
DAYTON, J. E.
clo The Institute of Archaeology, 78 Dean Street, Soho, London WJV GBE, UK
for large volcanic eruptions in ice cores and from tree rings give us an opportunity for a fixed chronology for the
Two dates
e Age of the Aegean and Egypt. The archaeological fallacies of the present system will be explained, and a solution
Late Bronz
fior the chaotic and destructive end of the Bronze Age itself.
proposed
KEYWORDS: DENDRO-CHRONOLOGY, ICE CORES, EGY PTIAN KING LISTS, VOLCANOES, MASS
EXTINCTIONS, SEA PEOPLES, SANTORINI, ASH FALL-OUT, TSUNAMIS, RAMSES III, THE EXODUS.
archaeologist and a geologist, has who built the three great pyramids. Herodotus' description of
Th e writer, a Middle East
Archa eometry Conferences almost since their the walls of Babylon and the Tower of Babel were found to be
attended
Oxford, and this paper is an attempt to bridge the conect by the German excavators in 1908 so why should he
begin ning at
gap between archaeology and science. The be wrong about Egypt where he also described the Labyrinth.
widening
archaeolo gists, now specialise and ignore the scientists. The 5. Astrological Venus tablets, of little value, but a solar
diffusion of technology is out of fashion. The scientist is not eclipse in Babylon on 15th June 763 BC is a fixed point from
aware of the problems, such as provenance, which the which we have floating Assyrian King Lists.
archaeologists now avoid. Sir Alan Gardiner (1966), the great Egyptian scholar has
It is not realised by scientists, nor by many archaeologists, summed up Manetho perfectly:
that firm dates do not exist for archaeology and h istory until "In the earlier days of Egyptology, its adepts accepted the
about the time of the First Olympiad, assumed to have taken evidence of Manetho with child-like credulity ... and today
place in 776 BC. Even our modem dates are wrong. For there are still scholars not innocent of this enoneous method
example Herod the Great died in 4 BC when C hrist was born. of calculation. The student has to be on his guard against
Therefore we will be celebrating the Millennium 4 years late popular traditions that are offered as history, and assertions
in 2000 AD. The Christian era was only organised by a monk, containing a kernel of truth but resented in a distorted form."
Dionysius Exiguus who lived in 496-540 AD. Later in Serious chronology only began with Sir Flinders Petric just
165 AD Bishop Ussher of Armagh stated that the world over 100 years ago at Abydos in Egypt. He excavated with
was created in 4004 BC which many still believe to be true to care, as distinct from looting pretty things for museums, and
this day. invented 'sequence dating' by which at a multiple series of
Like science, archaeology is a mere 200 years old. levels, the deepest was the oldest. This method did not allow
Napoleon's team of scientists found the Rosetta Stone in for earlier excavated 'fall-out', when sites were levelled for
1798. In t h e 1820's Thomsen sorted King F rederick's rebuilding, or for intrusive pits and disturbances by ancient
'cabinet of curiosities' into the 'Stone', 'Bronze' and 'Iron' tomb robbers. Pettie reduced his original date for the Ist
Ages that we now have. Dynasty of Egypt from 5546 BC to 4145 BC. Breasted
A mere 50 years ago, only 4 aids were available for dating reduced it further to 3400, and now the 'official' date has
the prehistoric past: reached 3114 BC precisely. One can see that a 'fixed' date for
Egypt is not yet a reality.
1. The Bible, with its genealogical tables, its description of "The chronology of Syria, Palestine, Crete and the Aegean
known places, and the important description of the Exodus is all based on tenuous relations with Egypt." Renfrew (1973).
under a Pharaoh Ramses. "The whole chronological problem could be clarified by a
2. Five Egyptian King Lists, all of fragmentary form giving single positive synchronism which all scholars could accept.
the na mes of some kings and the lengths of their reigns. All But such does not exist." Ward (1971).
these lists are incomplete and floating in time.
3. A semi-mythical list of all and every possible ruler of the Petrie' s Ongoing Great Mistake
"Thirty One Dynasties" of Egypt. Compiled by Manetho, an
Egyptian priest for the Macedonian ruler Ptelomy c. 300 BC. W hen Petrie at the end of the last century was excavating at
Some of the Dynasties certainly did Abydos in Egypt he was in fact digging in rubbish pits. He
not exist. Egypt consisted
of t wo king
doms "Upper" and "Lower", and the rulers may states:
not have actually
ruled both, although claiming to. Manetho's "Green glazed baboons were found at about level 70 of the
dynasties add
up to 5608 years precisely from the first ruler Ist Dynasty, and more were found at level 28." He continues
Menes or Minos (Egyptian like Arabic and Hebrew has no "We have already noted the difficulty about these things being
vowels.)
found at so high a level .... but it is clear that the objects are all
4. The much maligned ancient Greeks, Homer and of the Ist Dynasty." There is also "burnished black pottery
Herodotus.
Herodotus visited Egypt and Babylon c. 450 BC from Crete". This was in fact Tel el Yehudiyeh ware, now
and was
told that Min was the first to rule the two kingdoms, dated to c. 1600 BC.
and also
the names of the builders of the pyramids at Giza. He What alerted the writer to this c hronological enor (Dayton
al
b so cu nously put Sesostris a great ruler of the 12th Dynasty
1978) was "At the beginning of the Ist Dynasty we meet the
efore Che
ops, Chephren, and Mycerinus of the 4th Dynasty art of glazing fully developed, not only monochrome but for
Proceeding
s of the 31" lntemational Symposium 011 Archaeomet1)', Jerem, E. & T. Biro, K. (eds.), Archaeopress -Archaeolingua, BAR-Centi: Ew: Se1: 1.
Dayton, J. E. The p roblem of chronology
longer holds Citi carbo 1984) volume of Krakatoa in 1883 the only comparison of a Plinian
Crescent and spread to the barbarian West no oaks (Pilcher et al. and matched
amazing statement: "Las t ye a . ears on Irish
i m any pieces of pottery closely "W'11h t h e calibration of carbon 14 dates they
are unteWillet b
tast 6 Cologne, Stuttgar t, and Gottingen. explosion in an ocean that we have.
nabI
resembling Aegean ware c errnans at
by the
.
ame t o Ii ght m the tom b of as t he temples m Malta predate the pyram
ids and S 1984). Kuniholm , Kromer and What is impo1tant for archaeology is the existence of a
Mersekha and in tha t o f D en
of the I st D ynasty" o nehen r & Delorme
pha e III is earlier han Mycenae." (Renfrew ; g IT P.USCh ne a Minoan/Cycladic town buried under 40 metres of white
Identi 1973 . \..,.. . produced a long sequenc e for Anatoli
'The body of the ware IS . have also
cal with t hat oflater Aegean Libby checked his first dates against carbo
M na 1996). Whereas the Irish, volcanic ash (ignimbrite, glassy silica shards) with diagnostic
or Mycenaean pottery, th .. nised b
500 BC (Kuniholm et al.
.
e
Pattems are conm1on on pompen, and against wood from a
boat of the . "Ulll from are not tied to any pottery which has been little published. This city of Akrotiri
Mycenaean pottery, and llldee . tune and American sequences
d no patterned Egyptian pottery CaIigu1 a, and found that they m atched perfec Of a errnan, .
also contained magnificent frescoes depicting sophisticated
. a date from umbers of the Kas
is known until the XVIIIt h tly H is
' dates for date' Kuniholm has
Ynasty. This pottery of a fabric the Near East did not fit Petrie's chronology bistoncal . .
,,
and of forms ent'Ire1 Y iorei d
an so they . k of 1305 BC which
contamed Mycenaean pottery. shipping and the siege of a multi-storied town such as Troy.
g n to E . ...
w"'0 shiPWree . .
Christos Doumas (1989:39) wonders if the warriors in the
also has doubts regarding the 'offi
gypt, and f European wiong. I t never occurred to anyone that the cml' E gypttan
character, is now absolutely d at Egypl a
with large wooden beams for the roofs, and with 14 rooms on
chronolog Y 0 f
the Near East
the last place which these materials reached
Hyks s, c. 1600 BC. Anachronisms such
at the ti :: ;:
0
Y asty to which Petne depopulated twice, before and after the Trojan war. If the
it. But it was even younger. atomic or volcanic explosion eventually settles on the snow of
(That the dates for Knossos, Pylos, Mycenae, Santorini & the polar regions. This 'dirty' snow which can contain up to Thera frescoe really is of the siege of Troy, then the eruption
An XVIIIth Dynasty scarab ' .
a faie. nce foundat10n plaque of
Thothmes III c. 1450 ' green f .
Ramses II are the same, whether calibrated or not is very 56 different elements, strontium 90, sulphuric and nitric acids, was after that event, quite late in the Bronze Age as is agreed,
aience bab significant, marking as they do, the end of the Bronz Age in
and a cartouche of Necho of 6
oons as at Ab ydos,
0 B C were found in the same
hydrofluorides, dust from volcanoes, lead from petrol, oxygen thus pointing to the 1159 BC date for the Thera eruption. It is
the ear East.) (The Knossos date was on barley grains.) 16 and 18, etc. does not melt in the very low temperatures, and now also agreed by vulcanologists that there was no great
arry Snuth Wnt
deposit As Professo r H
es Adams (197 4 ) . no It is agreed that Carbon 14 dating is not a precise tool, as
m period of time (a few months at the most) between the
object from the Main Dep Sit . is covered in succeeding years by more snow. The layers in
c an be safely attributed to an the cosmic ray neutrons are produced by the sun which is not
early date .' At Abydos, di scu the sliced ice core can thus be counted individually and warning rumbles and earthquakes on Thera, which enabled
ssed ab .
ove, a senes of New a stable star. Over very short periods, e.g. 460-390 BC, and visually with back lighting back to about 8000 years with the people to escape with their valuables, and the final great
Kingdom temples c. 1550-13 20
r est on the levelled remains 1880-1960 AD, there have been major anomalies (Pearson et great accuracy and to 160,000 years by calculation. explosion, and so no gap of some fifty years at Knossos.
of the Old Kingdom buildings,
. hence the mix up al. 1986; Budyko 1969). The GISP2 core gave a superb record over the last 7000 In a deep sea core no. 58 at a distance 110 km from Thera
Based on Petne 's dates the
wh 1 e of .
. 0 Egypttan chronology Eight samples of grain from the same storage jar at years. Less than 1 m of core was lost, and large scale bedrock the volcanic ash was 78 cm thick. The tephra in Core 50 at
is wrong but no archaeo1og1st dar . .
e adnut it. Santorini gave dates of 2037, 1850, 1420, 1350, 1394, 1300, folding did not exist (Zielinski et al. 1994). The dust profile 200 km distance towards Rhodes, where Trianda is destroyed,
The above exposition, I . .
hope, ex
plain s the basis of 1110, 960, and 900 BC. Six threads from the Shroud of Turin w as v ery accurate and known and unknown volcanic was even thicker at 212 cm (Ninkovich & Heezen 1965). This
archaeological chronology to this
day. produced dates of 1260, 1265, 1295, 1305, 1345, an d 13 82 er uptions, together with the dust of the Siberian meteor of shows that the wind was from the south-west at the time of the
AD, but they were good enough to show that The Shroud did 1908, clearly visible ( Anklin et al. 1993). eruption, perhaps in early Spring as some of the storage pots
Carbon 14
not date to 33 AD when Christ was crucified (Aitken 1988). The volcanic Plinian explosion must be strong enough to contained beans which had started to sprout. The coasts of
.
Fortunately we now have two techniques which pro vide us eject the dust
through the tropopause at a height of about 20 Anatolia would have suffered greatly, so putting the peoples
This was the first scientific tec . .
km. Some eruptions have reached 40 km. El Chichon had two
.
of that area in motion. Knossos and Mallia are only 120 km
hni que to stnke with accurate dating within an error of 5 years, they are
and it was a revolutio n. C0Iin R
1968 to recognise itsimpo1tanc
archaeology,
en frew was one ofthefirst m
.
nngs and the Ice Cores from the Polar Regions.
tree
asts to 4 and
.
38 km. Tambora also two to 33 and 44 km. away so the damage to Crete can be imagined. But we must
e but I. ts .imp cations
li h ave esuvms m 79
AD is calculated to have reached 33 km. In the not forget the other islands in the Aegean and the mainland of
still not been realised by archaeo ' .
It has shown that 'barbaric'
logists Tree rings or dendro-chronology ; ce cores,
together with the tree rings, we therefore have a t Greece and Turkey only 200 and 300 km away.
are as of western Europe had ast, a dat
ing tool accurate to a single year. At Trianda, at the north of the island of Rhodes, recent
technology and a civili sa t'ion of .
. sorts earlier than those of the In an endeavour to correct the original Carbon 14 dates that excavations have revealed layers of Tephra from Santorini
Middle East and the Fert11e C
ha d so upset the Egyptologists, the tree rings of the long Jived
.
from 30 to 50 cm thick (Marketou 1988). Tephra up to 39 cm
rescen t. Thi s, as the wnter Santor
demonstrated in 1971 and 1978,
ini
bnstle-cone pine which grows in the White Mountai ns of
. .
. Was peitect
-" I y 1og1caI , as the thick has been found on Kos, 95 km to the N.W. Both Kos and
mineral deposits of copper, tin
to make
bronze
.
with copper, Nevada at a height of some 10,000 ft. (3,500 m .) w e Santonru,
also known as Thera, is the key to the chronology
Rhodes are only a few km from the Turkish coast, and Lycia
and silver were all m the We
st and outside .
the Fert11e analysed for their Carbon 14 age and studied for variations in f the M
iddle East, Egypt, and the Aegean, and its date will is from where the Lukka a contingent of the 'Sea Peoples'
cosmic ray activity. The rings also show climate ch ange and
Crescent. Tin bronzes were not .
discovered na!ly corr
. by Mesopotanuan ect the cluonological nonsense of the last 100 years. came from. At Xania on Crete 15-20 cm of ash has been
pro spectors in Bohemia as Chil
de (19 9) dramatic climatic events, such as ve1-y narrnw rin gs caused y
b The isl
had postulated but and, some 110 km north of Crete is the remains of found. A deposit of only 10 cm of tephra can put fields out of
had a rrived in the Near East with . . an enor
mvadmg peoples. The great frost damage brought about by m ajor volca nic ev ents mous volcanic eruption which left a huge flooded action for ten years.
84 85
Dayton, J. E. The problem of chronology
There appear to be only three major volcanic events in the silver, antimony, tin, cobalt, tho1ium, bism uth
ice cores and tree rings in the two millennia BC. They are at: telurium. The damage to Alaska and northern c' SU Phur
1
BLACK _SE
m the period 1 9 19_ 18
. anact a.
2345 BC in Irish oaks, with Icelandic tephra in peat bogs of known. Z1es calculated that .
2310 BC dated by C l 4 1,250,000 tons of hyochloric acid and 200, o2 sollle
. .
hydrofluonc acid were discharged into the atmosp
00 o
ns Of
!
1628 BC in ltish, German, and English oaks, in Anatolia,
here.
and in Bristle. Cone pines.
1150 BC which is very clear in Irish oaks, starting at 1159 2. Tsunamis.
BC and the climatic effect lasts for 18 years until trees begin These are great waves caused by volcanic exp!
o s1ons
and
to recover. 18 years also show up in Fennoscandia (Briffa by earthquakes, often out at sea. As the waves
. approach -
1994). The above information is in Baillie (1995) and the contmental coasts and the waters become shal "JC
lo w so
same dates show up in the Ice Cores (Hammer et al. 1980). .
he1ght of the tsunami increases often up to the
35
But as a cave at Zielinski G. A. et al (1994) list 3 large Krakatoa. Sometimes the anival of the wave is
prec = at
eruptions at 1459, 1457, and 1454 (VEI 164). Also 2 at 1192 the sea receding exposing the seafloor. At Puyehue
1960 people on the shore noted that the tide had
in Chile y
s dd Ill
(VE l 110) and 1190. (VEI is the explosive index).
The 1623 BC date at which archaeologists had confidently
goe out. Fi een minutes later the first of 4 wav
es 6
nly
jumped has now been proved by analysis (Zielinski & amved, causmg much destruction. Twenty-two hour
ave reached Japan, 17,000 km away with great
s later
Germani 1998) not to be from Santorini. We therefore have a force and
k illed 180 people. In Lisbon in 1755 the same thing
choice of other fixed dates for a ve1y big volcanic event. The ocCUITed.
whi1e m Japan in 1703 100,000 people were killed
.
effects of which lasted 18 years. The writer will argue for the by sueh a
acceptance of 1159 BC on the archaeological evidence of the
. .
wave. Santonru would have produced such a series
t
. . of waves Cf)
incursions of the 'Sea Peoples'. of smu1ar magrutude causing devastation on all the
sho re s f Q)
It is agreed that the Bronze Age throughout the Near East the eastern Mediterranean. The waves would have dest ro 0.,-,
0
Q)
ends in a phase of semi mass extinction. All the archaeolo the Minoan/ Mycenaean fleets, and caused enormous
coa taI
gical sites from Pylos in w estern Greece along the shores of d a mage. a..
Turkey to Syria & Palestine are destroyed by fire. Even Egypt The retreat of the tsunami obviously accounts for
t he al
and Mesopotamia decline into darkness. This extinction 'miraculous' parting of the waters that we hear of in 'Exodu
s'.
coincides with a mass movement of peoples by sea and by Someone had seen the effect of the tsunami of Santorini on
land seeking food and shelter in Egypt. They come from the coast of Egypt. Krakatoa caused a succession of waves of
Libya, by sea from the islands, and by land with their families two types, long with periods of about an hour, and short
and animals from Greece, Asia Minor, Syria and Palestine. It irregular waves. They were felt 9,000 miles away in Chile
is clear that this mass movement is caused not by Dorians and where ships broke their anchor chains. Two hundred and
other invaders but by the mass destruction to agricultural land ninety towns were destroyed in Indonesia by Krakatoa. The
caused by the Santorini eruption. destruction in the Aegean must have been similar. Meswesti
The evidence from other kno w n volcanic eruptions is as .......__
follows: The evidence for 1159 or J J 92 BC 0 100 200 300Km
1. Santorini was 5 to 6 times greater than Krakatoa, our I. The Bronze Age ends with universal destruction.
of Turkey are within the 400 km
only comparison for the walls of a Plinian volcano collapsing 2. There is a great migration of peoples to Egypt fro m all Map 1. The eastern Mediterranean. Mainland Greece, all the islands, and the Greek settled shores
destruction circle. From the destruction evidence it appears that Cyprus and Libya were effected within the 500 km circle, and on
into the sea, with the consequent inrush of sea water and the r oun d the eastern Mediterranean. These people are not
production of glassy ignimbrites and hot poisonous gases. them.
raiders, but bring their women, children and cattle with the evidence of historic eruptions, the dust would have reached as far as the Delat of Egypt. Tree rings in Ireland, Europe, and
3. The map 1 shows the area affected by poisonous ash Nevada show an J 8 years period of little growth after 1159 BC. With grazing destroyed for such a long period mankind could not
Thera had a high sulphur content. The ash/ignimbrites are
fluidized when the sea water breaks in. At Mt. Pelee in 1902 from Santorini. The Libyans (Meshwesh) invade Egypt from have surviv ed in the area. A few survivors migrated and the area was wasteland that we find in the early decades of the Iron Age.
the ignimbrite flow covered the 4 miles to St. Pierre in 3 the west. The 'Sea Peoples' from Syria and Palestine. Pale stine at the greatest distance from Thera recovering fairly rapidly under the Philistines and Tjeker.
minutes. The heat of the flow was 700C and 30,000 people 4. Egypt with the Nile's annual flood washing away
were killed in one minute, their lungs burnt out and their body poisonous ash would have had food. Other areas would have isotope dating has shown that the sulphide copper ores of 11. The Bible. The account of the Exodus clearly relates to
fluids evaporated leaving dessicated corpses. One man in the required some 15 years of rain to clean the land. Cyprus were not exploited before the conventional date of a great volcanic event. The Israelite slaves are building a city
town jail survived. 5. The evidence from Medinet Habu in Egypt depicts 1400 BC (Dayton 1984). Again a date of 1627 is ruled out. or palace for a Ramses, Pithom in the Delta (Exodus 12:37):
At Laki in 1783 and Hecla in 1970 in Iceland, thousands of Ramses III (conventionally ruling from 1198-1166) defe ating Plague and a murrain that kills cattle is a n01mal after The great eruption of Santorini and the confusion and
animals died when the volcanoes produced fluorine rich grass. the invaders. This date is near to the 1159 BC event. effect of volcanic events. The great eruption and dust veil of darkness (Exodus 10:21) gave them the opportunity to escape.
It has been calculated that if dry grass contains more than 250 6. Keftiu (Cretans) still appear as t1ibute bearers at the time 536 AD brought plague and famine to the Byzantine world of We also read that the wheat and the rice were not grown up,
ppm fluorine, animals grazing on it will die within 2 or 3 days. indicating that it was springtime in Egypt. The fish in the river
of Ramses II (conventionally 1304-1237) so Crete was not Justinian, and spread to Gaul and Britain by 542 AD (Baillie
In 1815 Tambora ejected 150 cu. km of ash over a distance once before the Trojan War, and immediately after it. "T
he cattle in
Egypt die. thunder were also caused by its atmospheric effect. The Lord
of 1300 km and 90,000 people died of starvation when their Cretans reward was to return from Troy to find plague and 9 . The Israelites were building a palace for a Pharaoh put a column of fire to guide them (n01thwards) by night and
crops were either poisoned or destroyed. Its dust cloud created fami n e among themselves and their flocks. Crete was the
n Ramses before the Exodus. This rules out the 1627 date, a column of smoke by day (Exodus 13:21). This must have
"the year without a summer" in 1816 when crops thousands of depopulated for the second time" (Herodotus 7, 1 70-1) Th
e Which is when the Hyksos were ruling in Egypt. been the eruption of Santorini which in spite of the curvature
g the 10. Prevailing winds. In summer the prevailing wind in the of the earth, must have been very great and visible from
miles away in Europe and America failed to ripen. durin
fall of Troy is conventionally dated to 1184 BC i.e.
The great Katmai eruption in Alaska in 1912 released so reign of Ramses III.
Aegean is
the Meltemi from the north, but from the end of Egypt. Then when they had crossed the 'Sea of Reeds' the
much hydrochloric and hydrofluoric acid that washing hanging 8. The King of Cyprus writes to Pharaoh "do not take i
t to September
to the end of M ay, the Aegean has variable winds Lord put the column behind them. (In other words they had
on the line in Chicago, 3000 miles away was damaged. Katmai heart that there is so little copper, since plague has kill e d all from the
south-west which would have carried the ash across reached the shore of the Mediterranean, where they met the
ad the sea to
also gave off traces of lead, zinc, molybdenum, copper, arsenic, the people, and there is nobody left to smelt the copper." Le the shores of Turkey (Map 1 ). tsunamis of Santorini and turned south into Sinai, and in three
86 87
Dayton, J. E. The problem of chronology
days they came to the Bitter Lakes (Exodus 15:23). All this is known in Egyptian records as The Third Inter Hammer, C. U., 1987, The Minoan eruption of Sant01ini dated
mediat
perlectly reasonable. We also hear that they could not flee e. P lit References to 1645 BC, Nature 328, 232-235.
recov e o
hundred years 1s too long for a civilisation to
r e
through the land of the Philistines, who did not settle in from the Santonm d'Isaster. Israel and the Phil' ' ven
Hughes, M. K., 1988, Ice layer dating the eruption of
1s1Jnes
Palestine much before 1150 BC and were part of the Sea . Pro 8Per Santorini, Nature 335, 211-212.
perh aps m only two or three generations, and so ., Ancient Hierakonpolis, Warminster, England.
the St B 1974,
People that Ramses III fought against. The mention of the that Solomon started building The Temple in 1 atelllen1
AdaJll5 .. .r A 1988, The Thera Eruption, resume of dating,
Jouzel, J., 1993, Extending the Vostok Ice Core record to
erusal
Philisti nes here is a later anachronism. years after the Exodus cannot be correct nor th eIll, 480 Aitken, iv.. ., 3700 m, Nature 364, 203-207.
etry 30, 165-182.
An Egyptian papyrns "The Prophecies of Neferti" . . ' at they
8llen1
A. chaeom Johnsen, S. J., 1992, New dating from the GRIP core (revises
430 ears m the wilderness (Exodus 12:41). Greenland Ice Core Project, Nature 364,
f\l""'n, .r 1993, The
in.,
trli Ii ..
(St. Peterburg I I 16b, recto) states 'the sun is veiled and It IS also clear that Egypt with its crops and cattl Dye 3), Nature 359, 311-313.
shines not in the sight of man. The river i s dry, even the river
e de
stroyed 20 3- 207. Kuniholm, P., 1996, Long tree ring chronologies for the East
.. .r G L. , 1990, Irish tree rings and an
also suffered from the eruption, which was mu c 'llie, in. event in 1628 BC,
. . h greate
r than .
of Egypt. (This is because the emption took place in April or Bai Ill Mediterranean, 29th Symposium of Archaeometry,
has been Imagmed, or experienced in historical
fImes. D. A. (ed.), Thera and the Aegean World
The in Hardy, TUBITAK, Ankara, 401-409.
May when the Nile is at its lowest.) 'The river is blood. Oh length and vacuum of this 400 year period 166.
hasforJ vol. 3, p. 160-
tha the earth would cease from noise. Trees are destroyed, no
. .
womed archaeologists. To raise the date of Sant
onru to 1 ng
. o
L., 1994, Dendrochronology & the nature of the
Kuniholm, P., Kromer, B. & Manning S., 1996, Anatolian
M. G.
627 Baillie, &
frmt or herbs are found. The land is without light.' (cf. Exodus Tree Rings, Volcanic Eruptions, the Absolute
BC would make this period even longer. The Holocene 4, 212-217.
AD 536 dust veil event.
10:21). This is exactly describing what it must have been like Chronology of the Eastern Mediterranean 2220-718 BC,
The 1159 date fits the Egyptian dates for Ram chronology and the chrono
ses III b M . G. L., 1995, Dendro
during the eruption (Vitaliano 1973; Gardiner 1914; Pritchard UI Baillie, Nature 381, 780-782.
means that the date for the destruction of Knossos
and the nd Bronze Age, in Waddell, J. and Shee
e logy of the Irish
1950). Krakatoa caused darkness in Bandung 140 km away . . .. . Lamb, H. H., 1970, Volcanic Dust, P hil. Trans. of the Royal
of Mmoan CIVIlisat10n, which is now contemp Ireland in the Bronze Age, p. 30-37,
orary With Twohig, E. (eds.),
for three days. Mycnaean, must be reduced accordingly. Society London A266, 425-533.
Dublin.
It IS to be hoped that analysis of the volcanic dust . Leuschner, H. H., & Delorme, A., 1984, Verlangerung der
G. L. & Munro, M. A. R., 1988, Irish tree rings
. ,
Conclusions . m the1ce Baillie, M. Gottingen Eichenjahningchronologien ftir Nord und
cores nught fingerprint one of the two volcanic veils, Nature 332, 344-3 46.
da tes10 Santorini and volcanic dust
Santonm .. which we know German oak tree chronology
Stiddeutschland bis zum Jahr 4008 v.Chr. Forstarchiv
was Iich in sulphur. 1981, A 2350 year south
Becker, B.,
The archaeological picture of the Late Bronze Age reveals 1159 also gives us a fixed date for The Exodus w l, 64. 55, 1-4.
hich can in Fundberichte aus Baden-Wiirttemberg 545-5
much confusion. It is clear that it ends in massive destruction be seen to be based on fact rather than fable. of the Hyksos. British Moore, J. C., 1992, The chemical basis for the electrical
M., 1996, Avaris: the capital
Bietak,
and burning that stretches from Pylos, in the far west of Greece stratigraphy of ice, J. of Geoph. Res. 97, 19803-19812.
Museum Press, London.
to Mycenae, Tiryns, Midea, Gia, Lefkandi on Euboea, to Kea Ninkovich, D., & Heezen, B. C., 1965, Santorini Tephra in
Addenda Bond, A. & Sparks, R. S. J., 1976,
The Minoan Eruption of
and other Mycenaean sites coastwise along the shores of Submarine Geology & Geophysics, Colston Papers 17,
Santorini. J. Geo!. Soc. of London 132, 1-16.
'.'1edite aean to Turkey, Miletus and Tarsus. The end of Troy
rr , Since giving this paper in April 1998, G. Zielinski and
M. Briffa, K. R., 1994, Holocene climate change:
evidence from tree 413-453, Bristol.
conference. 913-914.
Gard'mer Sir, H., 1966, The Sceptre of Egypt. Stothers, R. B.& Rampino, M. R., 1983, Volcanic eruptions
.
Gardiner Sn, H., 1914, Journal of Egyptian Archaeology, in the Mediterranean before 630 AD, J. of Geoph. Res.
Sheet, Nature 270, 482-486. eruptions, with special reference to tephra and gases, in
Fig. 1. Part of the relief from Ramses Ill's palace at Medmiet Hab'.1, near Ham m er, C. U., 1980, Past volcanism in the Greenland Ice Sheets, P. D. & Grayson, D. K. (eds.), Volcanic Activity &
Luxor.-
,
T,ms
, sIwws the 'Sea r . i thetr
ne a p/es coming by land wit!
women, cu / 'Idren and ox-carts Sheet, Nature 288, 230-235. Human Ecology, p. 125-159, Academic Press, London.
'
. ,
The men with the 'feather head dresses are Peleset (Philistines) ,"'h . ' Hammer, C. U., 1980, Acidity in polar ice cores in relation to V italiano, D. B., 1973, Legends of the Earth, Bloomington,
.
, e man wit
. . . . .
men : enmy pe1 haps from Sardmw. The men with wigs are Egyptians takti' g ' a Pl,1.1.ist me
pnsoner: absolute dating, Journal of Glaciology 25, 349-372. Indiana.
88 89
Dayton, J. E.
m
Zielinski, G. A. & Germani, M. S., 1998
Zeigler, P., 1970, The Black Death Pelican Books . , New I
ce
, London. evidence challenges the 1620's BC age fo
r the S an core 1 Laboratoire de Perrophysique, Departement de Mineralogie, Universite de Geneve,
tori":
(Mi noan) Emption, Journal of Archaeological 13, rue des Maraichers, CH 1211 Geneve 4, Switzerland,
sc1ence25,""
2Departement des Recherches Archeologiques Subaquatiques et Sous-Marines, DR ASSM Annecy,
58bis rue des Marquisats, F-74000 Annecy, France
data sets are now available for various regions of the globe, howeve1; the scatter in the directions is larger
Arch aeomagnetic
would expect from a consideration of the stated errors in direction and age. This is despite laboratory tests designed
tha n one
poor quality magnetic data. Various sources of error have been investigated over the years, such as magnetic
to elimin ate
a kiln wall fall-out, and magnetic anisotropy. Little attention has been given to possible movement due to subsidence
refr ction,
overburden. This is because post-abandon defomwtion of hearths and ovens is difficult to assess as these
or pressure from
at high temperature in antiquity. The recent discove1y at Laprade in the
structures could have been deformed whilst functioning
Fre nch RhOne valley of a series of flat hearths dating from the Late Bronze Age /lb has enabled a defomwtion test to be carried
out by AFAN along
out. This archaeological site, between Orange and Montelimm; was one of many rescue excavations carried
the future Valence to Marseille TGV line. One of these hearths was markedly dome-shaped and gives the poorest
archaeomagnetic result of the hearths analysed. However, if the individual specimens from the dome-shaped hearth are tilted
back into a horizontal position the archaeomagnetic result is greatly improved, and in better agreement with the other hearths.
This study shows the importance of applying, where possible, a tectonic correction to the archaeomagnetic analysis of defomied
structures.
;
;
Plan Section
;
; Baked clay (floor),
;
yo22uso i
;
;
/
/
/
successive stages (section)
; /
/
/
/ [:-:-:::-:::-=-:-j f:-:-:::-:-:-:-j Partially baked clay
; ..... /
/ /
;
;
/
/
/
/
/
/
l'..-..'.:)l9J21 Gravel
/ /
Silt with carbonate nodules
mi /
/
/
/
I II I
I!!! / /
Cl!/ / /
/
Clayey silt
:;:;i
.
/
/
l:tMtttJ Hlf:Htfl
(!).,
/
/
i-:1
;
; F1012
/
/
Silt, traces of firing (section)
y: 221.1
;
+IF1019 E
;
'8till 4
I
0
i 3 I{)
LAPRADE
; 0
;
;
;
;
; IJ5 0
0 0,50m
n of archaeomagnetic samples.
Fig. 2. Plan and section of deformed hearth 2014 with locatio
bull's eye
0
0
Results / spirit level
M
le
The archaeomagnetic results from the three hearths are
shown in Table 1. All three hearths give acceptable
/ /
'' -...____ ... - '- \
(2014) -
/\ _,,--...../ / \
-
Fig. 1. Geographical location of the archaeological site of Laprade. archaeomagnetic results with the dome shaped hearth
b.
showing a greater dispersion of the directions of the individual
ho1izontal, was markedly deformed into a dome shape mrnw corresponds to the strike of a geological sample. The specimens than is the case for the other two hearths. The mean
. .
(Fig. 2). The discovery of the disturbed hearth along with the inclination of the smface (dip) was measured using an directions of all three hearths are similm with 2014 having a
. magnetic North
others from the same period enabled a field test to be camed electronic inclinometer (Anglestar, 0.1). slightly lower inclination than the other two.
. .1ar to th e fold test in palaeomagnetism.
out smu A portable magnetic susceptibility meter (Bartington MS 2, Considering the longer reversed store period the viscosity
/
probe D) was used to select those parts of the hearths which coe fficients (Table 1) are reasonable and the remanent '-
Experimental study / '
,,,--...../ / \
Sampling had been the most heated. This is due to the large incr ease in magnetisation of the hearths can be considered to be stable.
c.
_
susceptibility of the clay with maximum heating tempe rature. Co ncerning the possibility of a magnetic refraction in the
The b aked clay surfac of the hearths was sampled in the thin sheet of the hearth it would seem that the intensity of
. Elevated temperatures can sometimes lead to a slight lowering
field usmg the g ued-disk technique (Clark et al. 1988). of the susceptibility due to vitrification but this can be eas ily ?1 ag netisation is not sufficiently high for such an effect to be
Fig. 3. Archaeomagnetic sampling method;
A polycarbonate disk 25 mm in diameter and 5 mm thick was recognised due to the greater hardness of the surface of t
he impo rtant. Calculations of Coe (1979) show that for a thin Glued disc technique.
stuck on the surface of the baked clay using a few drops of vitrified clay.
sheet with magnetic properties similar to the baked clay of
the thicker and
instant glu e (cyanacrylate). The plastic disc enables an Laprade one could expect a shallowing of the inclination of subsidence of the poorly fired edges around
.
0nent auon arrow to be easily drawn on the . around one degree. better fired centre.
specimen an d Laboratory analyses to those of
serve s as a ps
1p ort for fragile samples (Fig. 3). Because of the anomalous shape of 2014 it was decided to The mean direction of remanence moves closer
be disting uished at the 95%
A geologist s compass (Brunton) was used to draw an on apply a bed
ding correction to the individual specimens by the other two hearths and cannot
The specimens of baked clay were trimmed, using n
arrow m th e d"Irec!Ion of magnetic north on rotating each of them about their strike so that their upper confidence level.
those samp1es magnetic tools, in the form of cylinders (diameter 22, le n gth
wh ich
.
. were honzontal. The possib
le influence of a strong1y 18 mm) to fit inside the measuring space of the magnetome r,te Utface ?
ecame horizonta l. This is equivalen t to the
magn euc h earth on the compass needle was reduced by usmg
. Ypothesis that in antiquity the hearth was a horizontal sheet
Conclusions
. a Mo/spin fluxgate spinner.
a 10 cm p1 ast1c di stance ed of baked
. .
piece between the compass and the The magnetic stability of the samples was investig at clay. The dispersion is then much reduced with
are alpha 95 Application of a tilt correction to the deformed hearth of
heru1h. The. direc!Ion of the orientati"on anow . was then using a viscosity test (Thellier 1981) where the sa m ple s % dropping from 3.4 to 1.4 and the precision
chec ked usm a sun compass, thus eliminating any influence n .a ara Laprade not only improves the precision of the archaeo
re-measured, after having been stored for one month i meter increasing from 90 to 490. Such a dramatic
of the m agne hc haith on the orientation procedure. n eUC imp rovement strongly suggests that this hearth was deformed magnetic result but also brings the mean direction obtained
reversed position with respect to the laboratory Geomag
In the case of lilted samples a hmizontal arrow was traced w h at fter the last firing and that magnetic instability is not into better agreement with the data from the two other hearths
e
field. The samples from Laprade were kept for a so m
. , import from the same site. The unique chance of being able to carry
on the disc s upper surface using a small spirit level, this longer period (9 months) in a reversed position. ant. The deformation was most probably due to
92 93
Hedley , I. G. & Billaud, Y.
b N
TOWARDS A RADIOCARBON CHRONOLOGY
OF THE ROMANIAN PREHISTORY
LASZLO, A.
Fae. Jstorie, Universitatea "Al. I. Cuza" Bd. Colou No. 11, 6600 /a:ji, 9, C.P. 1802, Romania
* All specimens horizontal. Available data 49-50: Palinca I996: 244-246; Fischer 1996.
N : number of specimens. 5I-52: Laszlo 1993: 28, 32, Tab.I,Fig.I and 3, no.4, 7.
# NRM, (natural remanent magner At the present time there exist 269 radiocarbon datings
ization), average value stan
94 Proceed
ings of the 31" /11tematio11al Symposium on Archaeometry, Jerem, E. & T. Biro. K. (eds.), Archaeopress - Archaeolingua. BAR - Centi: Eur. Se1: 1.
gy of the Romanian Prehistory
Towards a radiocarbon chronolo
Ldsz/6, A.
---------- - ;;E'"A(;"EB
r l.ABORATORY MATERIAL'
j__
' 1 P
JPERIODS tv_1B_ E_ R_O_F__- 1_4_D_A _T__
C_ A _ ENTIRE RANGE OF DATA(!
-
1 S\U.-- z:lil:TURE
. NO ANALYSE
-
l _N _U_
t\
--
1---
\ .
:
J - \No: ( l_<:;L__
i h'OT AL
J
-
I
: __ _ _____
I
P hases
Cultures
Historic-geographical regions
DATA 1>------_J
. ; \ . . .. .GrN - 0 5 .
+ ----
I
- ---
D JG B T BP CAL BC
I
l 2
.:: ... ..
,,
.
'B
-.--. , 54000 +
\ Mouste.nan
M W 0 15 5 .
BB,CIW
!
_
- 11230
L--
\
\
f--;--\:Ri p j;;- - ;;i;;; niv.lll Mo
l Bor o:;;tcni
l 1 ustrian GrN 1-.17M
L 1 ! 3 4 5 : 6 7 8 9 10 11
46400
i I 28
IT C +4700 +170{)
1
r;:.-air; H ! I 11 1 14 !>54000146400 -29700
-2900 CiW. A
i
I
.
! """" rRhmo v
--
"GLIT\ "9700
i i l
-- -
\ n1v.\lb
fUPPER PALAEOLITHIC
-
l Aurignacian
! Gravettian I
23 I 32730 220- 241001300 GrA- 1357
GrN - 10529
32730220
24100 1300
\C
Ci\V. A..BB
s1 4 57
l 28910 480 - 16150 35
i EPIPA LAEOLlTHICi - . ...{.. ... . . . - .. .... .. .
; ME SOL IT HI C GrN _ 12636
c;;N:_1_(;52g----t--16lsiH 350
i 289LO:':_< .. . L.c'. _..............................
! Tardigravetlian 12600 120- 10125 200 + \ C!W. f\
i Schela CladoV"_'_ i -. .
I CIW. A
_-1-
__,_ ___ 1
, 1-+-- i l_l_ +-_ _ 8_0_4_0_ 60_-_7__
_ 1 _ 19_5_
_ 1o o_ 6 2_
_ -t---_1_ 0-_ _5_97_ tl_ -I .
...... j. ............
_,___ _
_. , __
- \ L2600 120
ITardenoisian
_______
!
l NEOLITHIC.
7850 215 7060- 6450 i I
... .. ..
. ; .. .. I
j C.:'\V, A. .BB
....-.
, AENEOLITHIC
l\ii2S .i 200
! Starcevo-Cris lll - TV I
i
i j
4 6 6665 45 - 5860 60 5630- 5494
804().;, 160 ! ("i\V
i i -4835-4715
um
- 4896- 4621
esti j I j
1
6280 JOO 5330- 5148 i
! Linear Pottery 1 !
5319- 5070 1 1 2 ! 6245 100- 6170JOO
t l 1
52
-------+---l---+'1 -+--i---+---+--.;---+----------1-- 9_::_:2:.?J
! Boian IV (Span\ov), I
! Transition I
'1_'1_-_9''.'"_lfllt<: t---_:_-
:
5060- 4780
-
_J _E_+-_5_ 98_ 5 12
_ _0_-_ 5_7_0_5_
__-+-__471
_ 80 _ _ 3_ -_ 4_4_6_6-ci
I G u meln i Ja A2
___ __ +- _ _
I I :I 14 I. I I 15 5715 70-5360 70 4713- 4481
' I l - 4336 - 4046
! P etr eti A 3 i 5900 100- 5710 100 4933- 4720
- 4720- 4460
i S;kuJa 11 TJI 6 6 5710 45 - 5260 60 4668- 4504
- 4226-3998
_ 1_11_ _____;__-+ _ 2 --'-- 2
-+ -+---"---l- -+-- -+- _ 0
_ _ 5_88 _ _ 7_____
_
_ _ngi a 89
_-+-_4__ 4 _-__47_1 _8 _
j Precucuteni II - 111
;: a
_ _ma o_
5820 50 - 5530 85 4780-4619
H
4 4
t
- 4465 - 4339
i Cu cu e ni A 1-4 25 9 34 5690 50-520.5 I 00 4662-4469 !
i -4222 - 3827 i
I Cucut e ni A-BI 4214-4002 ......i... ........................................
I i
5246 24 - 4996 26 !
\ ic
[-cucuteni B 1-2
-3902 -3715
4 4 5162 37-4890 29 3999 - 3970 ! l CIW
- 3772-3646 !
_ _ _r _1 ----l ---
Coofen i I I 1
, Horodistea I - II
__ -
! ' -'-- -l-- -l
! !
i 3 _ , ___
-
9
__
12
-+- ---
4 570
1 - - - _
60- 4030 75
J
- _ _ _ _ _ _ _ _ -- - f--
- ;_ ;
_ _-;0
-
_ -j
1' j
i 4495 18 -4235 30 3331 - 3101 !
---------+---1-,--!----t-- i l
i +--+---t----+---1------------+-
'
- :::.
2908 _:__ _ i
.::_.:__:_;
.:..: -2783
I Ochre Graves I 4 ' T
i
4280 65 - 3400 150 3011 - 2785 !
I i
! EARLYBRONZE A GE
- 1900- 1520!
! Livezile Group
'MIDDLE BRONZE AGE
. I i
!
4109 44 2700 - 25&0 ,
l 42 li Co1o feni 111 UZ-2 870!ETH-'J278 4010 75
iB
I'
Po ana Ampoiuhu
i
IWietemberg III (C) I
4495 18
5 2 4 _:_
__:_:_.::_
HD-14785
-----C-__:_:_.:.=._
l
!t -+---t--+--1--t---+- -i- ------
\ Horndi$te
_
. 4J Ii Honx\i_::"_'I_.
rLATEBRONZE_ A_G_-E ----i
--
1685-1
1 HD- 1489g . \ 4215 30 .... \
Horodistea .
Ti;\I,<
\ 4 ! Horodistca . !
3270 30- 3200 30
rN 19 !C
! \Te 1/Ve rbicioara lV-V)
-1280 65
, Fundeni - Govora 4 4 1
1610- 1460
:;; ;;;;;;
!H;-
. ... .. \ . .
_
. ..
G
95
: 1o - :-J400c' 150 C
--1
I,
.
' - Bl n 25 \
i'
- --
-1510-1425 .'
i
i
--. .?... .L, --'.:._ ..:.'..-:--::
... <. . 1 i_- \vi--;-ili" (c )
. Bl -4622 --1\ ? (: 1 68 5 -t 5'24BC11.
\ 48 l Sih1 oara . -- - .... .
n
. mttent
+ -+--+---+-+--+--+--- -------l- ---__J
_j__-+-- \ I Viilor"
l CiW
--+-- --
I
B
- _______
.
3270 30
o
\ 49 .Popesii
n 6101
u 1 1i
Noual-1 1** 4
_
_j_ \\
f d Ve:rb
e - lcion -
3260JOO- 3100 + 55
Ci w
1 ra rlV-V
a :l.) ,- - ---i--32()(;-;:J"(l-Ci\/:I
1 680- 1430
:, C/ W
--
- - -;;-
I _
105:--
cTi/
-;;;;-;
soh
-1434-1313 l
:
L ---- ---
--- -
-- -
j; -Fundem - Govora B-6 3200 + 30
.
\ 50 Popeti ra IV- \'') .\---- -- - ..l --- . ...... .... .. . .
-
*Abbreviations: M =Moldavia; D =Dobroudja; W =Wallachia; 0 =Oltenia; JG = Iron Gate; B =Banat; T =Transylvania
Bin - 1086
\tTei!Verbic1oa
1
N ,ua l
GrN .
SI \. Mahala. 111\ . U a
--
_,___.
31\l(U 55
_ ..-.:'..'. _
j -_.::.;_ _
:..:. ..: :
_ _-'---
near Cerniiufi (Cernovcy, Northern Bukovina, present-day Ukraine). We do not have the pemiission to mention here the four -____j_ ':'._:
ol3.5
l NOlm Ll -
-l_J..:. .:--:::...i._1_';_.-..-1?. .. . ood.
14C datings obtained in the Groningen La.boratory for the Noua-site of Crasnaleuca, Botoani county, North-Eastern - e; C =charcoal; W =w
ONS: A =ashes; B =un
burnt bone, BB - burnt bon
*ABREVIATI
Romania. T hese datings will be published by Mrs. Lidia Dasciilu.
esented in Table 1.
radiocarbon dati.ngs repr
Table 2. Appendix to the
Table 1. The distribution of the available radiocarbon data by ages and regions.
97
96
wsz/6, A.
the Mitoc-"Malu Galben" upper palaeolithic site Damblon et OF CORAL BA SED ON A MINO ACID COMPOSITION
E ESTIMATION
Fischer, C., 1996, Probele 14C din valul de
e
b
al. 1996). Taking into account these criticisms, the values in de la Popeti (jud. Giurgiu), SCJVA 47(3), rollZu! !G
2
5 NYBERG, J.,1 MALMGREN, B.,1
the last column of Table 1 might suffer slight modifications. Gimbutas, M., 1991, The civilisation of the Go
dess 2
CSAP O, J., CSAP O -KISS, ZS. & CSAP 6 JR.,
of Old Europe, San Francisco. . h ewo 2 J.3
Conclusions Honea, K., 1994, Tranzitii culturale In pale
oliticu1
superj
timpmiu i cronostratigrafia de la Mitoc -M
alu G alb 'Geological Centre, University of GOteborg, G0teborg, 41298 Sweden,
The available . . . enG Hungary,
radiocarbon data do not cover, B otoam). Arheologia Moldove1 17, 117-146. Ud. 2fjaculty of Animal Science, Pannon Agricultural University, Guba S. u. 40., Kaposv{u; 7401
geographically and temporally, all archaeological pe1iods, Honea, K., 1995, Ambiguities of Aurigna Departmen t, Janus Pannonius University, Jjj u sdg u. 6., Pe es, 7624 Hungary
cia n - G ., ra 1 Sciences Natural Geography
cultures and phases; we need new datings to be able to outline raveltj t 0F1Yatu .
J Facul y 'J
'
oie
. . .,. de la . . . . .
Bronze Age in the Carpathian-Danubian area, generally ve ry s to other ammo acids, or d ecomposltw n. Th e
c1v1
. 'Iisation Noua-Sabatmovka-Coslogen i, . . can be explained by the ox1dat1on or conversion
Culture et am1no acids ' which
speaking, became unavoidable. But, in this respect, one . contammg . .
in the concentrat ion of Glu, Cly, Ala, Val,
au Bas Danube X, 21- 41.
Czvz11satzon slight in the case of Asp, Pro and Arg, and there were no changes
decrease was very . .
mcrease
cannot elude the "delicate" aspects of the contradictions Laszlo, A., 1997, Datarea prin radiocarbon in arh His in the course of 250 years. These AAs are not sensitive to environmental influences. There was a minor
existing between the historical and the radiocarbon
eo/ogie
. Phe and ion of other
(Biblioteca Muzeului Nafional II), Bucureti. vatt ' n 0F Ile and Leu during this time' which was surprising, but may be explained by the decomposit
in the coneent " o 'J .
chronology both of the East Mediterranean-Aegean World Mantu, M., 1995, Clteva consideratii privind cron to relatively higher amounts of these two AAs m
.
the protem .
Plo
tzng
.
213-235.
I
References Monah, D., 1978, Datarea prin C l 4 a etapei Cucuteni KEYWORDS: AGE DETERM INATION OF CORALS, AMINO ACIDS, D-AMINO ACIDS,
A2
RACEMIZATION OF AMINO ACIDS.
I
tiJtl
e These
of t he ot e .
.
. h r anuno actds, while the Ile
and Leu did not change
.
Gly
Ala
5.28
5.17 0.284 -0.006 0.0019 0.0392 -0.85 0.0329
for those samples where the age was between g for d the relativ e conce ntrat10n of these two
c ntr
100 and 25 at o n , an
five minutes through the hydrolysing agent by glass capillary 0 Years. 0.068 -0.002 0.0005 0.0940 -0.91 0.0110
!II con e
1.41
i
Cys
.
. In contradistinction to these results, the co
ncentratio . ds increased. . -0.003 0.0003 0.0556 -0.98 0.0005
After bubbling with nitrogen, the Pyrex tubes . . .
ammorua mcrease d over time. Table l gives . n aJlllno act It can be Val 3.44 0.040
were f
inform at O tlte sho ws th e parameters of linear regress10n. .
immediately closed, and put into the heating oven at ion Table 3 Met 1.95 0.078 -0.004 0.0005 0.1084 -0.97 0.0011
l 10C the quantities of amino acids in the coral, but negati ve correl at10n between
do es no t .0n is a very close
for 24 h. After hydrolysis, the tubes were cooled at t hat there 3.68 0.050 -0.001 0.0003 0.0690 -0.93 0.0080
room any information on the quality of the coral gtve stated al and the amino acids presen t. The only Ile
temperature and the pH of the samples was set to 2.2
protein . o fthe cor .
0.090 -0.002 0.0006 0.1244 -0.14 0.7937
with 4M In Table 2 the amino acid composition of the the age . where r is -0.14. For the other ammo Leu 5.11
protein in n ts 1eucm , e
NaOH. During neutralisation, the temperature was kept excepuo Tyr 4.28 0.271 -0.010 0.0017 0.0373 -0.94 0.0047
below (g ano acid/100 g pr tein) an be seen. This (Arg). There was a
30C by means of a sodium chloride-ice mixture
. After
? _
quan1ty of the respectJ e anuno acids resen
table show : .
acids v ts b
etween -0 73 (Glu) and -0.98
. .
itive correla tion betwe en the age of coral and Phe 4.66 0.153 -0.007 0.0010 0.2112 -0.96 0.0022
dilution with citrate buffer (pH=2.2), the hydrolysates : ? t in 100 g
Cora) ry close pos Lys 5.54 0.723 -0.013 0.0048 0.9957 -0.80 0.0580
were
filtered and applied to the automatic amino acid analyser
.
protem. The concentration of those anuno acids
sensitive to the circumstances of the environm
that are v
e
:UU onia (r = +
0.96).
His 2.37 0.140 -0.003 0.0009 0.1930 -0.88 0.0260
The amino acid contents of the hydrolysates
were The highest decreases were observed in the case
ent de cr ea J References Arg 4.96 0.302 -0.008 0.0020 0.4156 -0.90 0.0139
of Thr, Tyr 0.0060 1.2280 +0.96 0.0021
determined by using an Aminochrom-11 and LKB 4101
type and Ser. The decrease in the case of Met and NH3 13.69 0.890 +0.041
amino acid analysers. Otherwise the analyses were carried Cys can be
out explained by oxidation of the two sulphur-con . os J. E. & Knutson, D. W, 1974,
Buddemeter, R W Marag '
Table 3. Parameters of linear regression (y=A + Bx).
taining amino
as described by Csap6 et al. ( 1986).
of reef coral exoskeletons: rates
.,
Glu 14.45 13.00 9.48 10.62 10.87 10.36 Bull. Ma1: Sci. 55, 151-175.
Asp 27.05 27.23 23.20 24.33 23.11 25.12 some Paleozoic and modem corals, Mem. Bur. Rec
Winter, A., Goenaga, C. & Maul, G. A., 1991,
Pro 3.94 3.10 Carbn and
2.72 2.54 2.29 2.27 Thr 5.15 3.04 Geo!. Min. 89, 497-506.
4.36 3.84 3.75 3.74
Gly 5.69 4.61 4.82 4.62 4.46 4.33 Ser 5.60 4.27 4.05 3.39 3.03 2.89 Knutson, D. W., Buddemeier, R. W. & Smith, S., 1972, Co al ; oxygen isotope series from an eighteen-year Canbbe
an
Ala 5.64 4.71 4.21 reef coral. J. Geophys. Res. 96, 16673-16678.
3.94 4.00 4.02 Glu 10.71 11.02 9.75 10.42 10.62 10.07 chronometers: seasonal growth bands in reef cora s.
Cys 1.53 1.22 1.11 1.10 1.00 0.96 Pro 2.92 2.63 2.80 2.49 2.24 2.21 Science 177, 270-272. .
Stearn, c. w., Scoffin, T. p & Martindale, W., 1977, Calcrnm
Val 3.48 3.33 3.10 2.96 2.88 2.79 Gly 4.22 3.91 4.96 4.53 4.36 4.21
Met 1.98 1.81 1.44 1.16 1.04 0.98 Ala 4.18 3.99 4.33 3.86 3.91 3.91 carbonate budget of a fringing reef on the west coast of
Ile 3.72 3.53 3.58 3.44 3.28 3.33 Cys 1.13 1.03 1.14 1.08 0.98 0.93
Leu 5.09 5.22 5.01 5.12 4.92 5.19 Val 2.58 2.82 3.19 2.90 2.81 2.71
Tyr 4.41 4.10 2.74 2.57 2.14 2.07 Met 1.47 1.53 1.48 1.14 1.02 0.95
Phe 4.63 4.54 3.87 3.41 3.09 3.13 Ile 2.76 2.99 3.68 3.37 3.21 3. 24
Lys 6.46 5.09 2.93 2.93 3.15 3.23 Leu 3.77 4.42 5.15 5.02 4.81 5.05
His 2.51 2.00 1.88 2.10 1.75 1.45 Tyr 3.27 3.47 2.82 2.52 2.09 2. 01
Arg 5.48 4.33 3.60 3.60 3.38 3.14 Phe 3.43 3.85 3.98 3.34 3.02 3.04
Ammonia 14.88 15.10 16.48 19.76 23.49 23.68 Lys 4.79 4.31 3.01 2.87 3.08 3.14
1.41
Total (without ammonia)
His 1.86 1.69 1.93 2.06 1.71
120.02 102.89 80.72 82.19 78.82 79.19 Arg 3.05
4.06 3.67 3.70 3.53 3.30
Total (with ammonia)
Ammonia 11.05 12.81 16.99 19.40 22.96 23.02
134.90 118.00 97.20 101.95 102.31 102.87 Total 100.00 100.00 100.00 100.00 100.00 10 0.00
100 101
3. Field Archaeology
OF KITROS
THE AR CH AE OLOGICAL SITE
LED STUDY O F
T NETIC
p E AI COMBINED MAG
!-. (N. GREECE) BY
ETRY METHODS
AND SPECTROM
S, Y. 1
NDOPOULOU , D.1 & SANAK!
SARRIS, A.', KO
AJDONA, E. ' ,
ece.
06 Thessaloniki, Gre
y of Thessaloniki, 540
1 Geophysical
Laboratory, Universit te, Gre ece .
H., Rethymnon, Cre
nean Studies, F.0.R.T
2/nstitute of Mediterra ", Athens, Greeece.
ter "Dem okr itos
3Nuclear Research Cen
' "''". . -
'
..
A
, ' '. ' ' ,..,
;
.IQ!)
i_
DARK GRAY
SAND
.-. -!Sil
:::::;,.
--c...
-
Q ;
r
-!%
. IJROWNSANDY "::..
... LAYER
I
.... , .]jfl
.i.
-HHi
.)ijij
!
REDDISH
L---'---1:}
CLAY
..\:
)i; "
SuHfplibility
Xfd '?!
Hi
Ont] TOrSOILWrm
CERAMICS
-411
,.,Jj A
.0.5m --
\
"' ,'
H I SANDY-CLAI'
. ..
ltn
UROW:"COLORE.D -ti I
i l
'
{
-1
' i
'
1.JGHT"\WN
;"'
0
! UORIZON -120 '
' I
11 SANDY-CLAY
. r-
BROWNCOLOREll
JI r
' . \!}(J
I
.
. It u / /.
-11)
. .. w II
t' /
-. _t
JOI I
REUOlSH l
.?.:{i
-200
IA / 10 c
CLA\'
'I I \ S6.
-io
!!O
1 --
)0 6(1 l.t'. Dnrr;nt Contra!
Suufptibility
II III .
IV V VI VII VIII VI. Fig. 2 Distribution of the magn
etic susceptibility
I
I. tren h
SJ and S2
S5 Tl measurements versus epth for
: ;
.
: :: ;
I r ne Contrast
I
r w
I
100 I-
e i
i : : :1 :,: ;; :v: : ae lo i al trenches.
'.;.!;:
.,,. . - s
20 I-
Xfd % perc entag e contn'bution
susceptibility, could not have been created in abundance. .
by the shap e and the
samples (Table 1) was estimat ed
Several samples were taken from kilns, in a radial direction, in symmetry of the Mossba u er spectrum.
Spectrometry analysis
order to define the variation in magnetic susceptibility from the
I I
40 liO 80 i 00
I
20
I
center towards the outer p ait of the kiln. For kilns F l and F2
Distance (rm)
In an effort to investigate the chemical composition of the
(Fig. 4), the values of magnetic susceptibility exhibited a samples, we peiformed ESR measurements in the Nuclear
significant reduction in a direction away from the center of the Fig. 3 Magnetic suscept1b1l
. . ity 9%
.
values versus distance inside
Sl-12
Research C enter "Demokritos" were performed. The
kilns. After 1.5-2 m the values increased again, reaching the .
1n tiie sa me diag ram th
.
e (Cross-section /gray horiwn)
7%
analysed samples (10 in total) were collected fro m the and outside the archaeologzcal szte.
F3-l
104 105
Aidona, E., Sarris, A., Kond
opoulou, D. & Sanak
is, Y.
study of the archaeological site of Kitros (N Greece) by combined magnetic and spectromefly methods
Jm A d etailed
lm MAGNETIC FIELD
T=IOK
so T=IOK
T=IOK
&
Tl-9
E
60
Im l.m
.
.
\
2ra :;;
40
\
\
"""
20
\
'"' ,,,.
t20c111
0
0 lO 100 ]j(I zoo 250 JOU
Distan cc(cm) 351)
t' S2-3
J \ ('\
- \
I
JOO ;:..
<'.
''.: '''"":
\. \
i
I
:t
High FrcqucnC'y
\ !\\ \ \
\\ )i
) Tl-9 \
Jm lw S4-4
\
f,
80 827 !
\i I
\
'"
\\ /
I
' ""
\'J \
... I i ' I
60 v
\;
u
.
...
'-
- ---!
IOI 40.'G }WJ S000
v,
I
:;;
40 ,.,. '
'
1 (1);1 ,.,., -
of the arc
v.
107
IS AND E DGE
PR IN CI PA L CO MPONENTS ANALYS
OF
PLICATION ES FOR THE DE TEC
TION
AP M EN T TE CH NI QU
ENH ANCE ANATOLI A
SI TES IN CENTR AL
OF PREHIS TORIC
ER, A. M.2,
ERDOGAR, N. J.1, OZ
, M. E. 1,
YILDIRIM, H. 1, OZEL
, U.3
GULCUR, S.3 & ESIN
Kocaeli, Turkey
PK 21, 4I470 Gebze,
ent , TU BY TAK Ma rmara Research Center,
artm 3I Ankara, Turkey
l S pace Technologies Dep hnical University, 065
ometry, Middle East Tec
2 Deartment of Archae 59 nbul, Turkey
, Beyazit, 344 Ista
ory, Istanbul University
Department of Prehist
o
en" ;n Centml AnatoU
o.-chaeologkal >ettlem
app Ued ;n ;n, ,,U gmfon of wh;,,odc eme nt tec hn; qu< >
Remo" ,,.
,;ng techn;q
ue' wm
wn e u>ed to ''",;n;ze the
""dy a<ea. Edge enhanc
matk mte lU" ;mo g" nor th-e ast, nor th
ond SP<fT Panchro 'a''' ;n
;magn. The PwwUt op<
Landsat TM andent road' and "tt
lemen" on the mtelWe
(PCA) was used to extr
act
h;g hUg ht the pon ent s Ana lysi s
ed to . Finally, Principal Com
were appU ctio ns wer e app lied app lica tion .
th-west dire e the best results for this
th-east and sou components which giv
west, sou ima ge and to find out the
infonna tion from the
significant
S.
Y, SATELLITE IMAGE
SING, ARCHAEOLOG
KEYWORDS : REMOTE SEN
eigenvectors
A are composed of the
The rows of the matrix
Introduction of the covaiiance matrix
Kx between the plan es. The matrix A
mat rix Kx such
tion of the covariance
performs a diagonaliza
archaeologists to find ima ger y,
The study area
was surveyed by the trix of the transformed
that the covariance ma
lem ents; such as flat
archeological sett
out the types of
anc ien t roads. The exact
uli as well as Ky=AKxAT
settlement, mound, tum
Magellan GPS
were measured by the
positions of these places
historical places nvalues of Kx
tem) to locate the pre se elements are the eige
(Global Positioning Sys is a diagonal matrix who
I on the Landsat TM and
SPOT Panchr om atic satellite images.
arranged in descendin
g value, (Pratt 1978).
Prin cipal Com
rected into bands of
were geometrically cor
I
sev en
These satellite images A) was applied to
arch aeological ponents Analysis (PC
red and fuse d. The
UTM coordinates registe Landsat TM satellite ima
ge.
ge. power of
on this fina l ima ( Table 1) indicates the
I
settlements were marked The correlation ma trix
aeological Landsat
ate the typi cal characteristics of the arch erentiating the ban ds of the
To investig
r eart h surface the PC A analysis in diff dsa t TM
from othe ds of Lan
settlements which distinguish
them sec ond and third ban
wer e app lied to TM ima ge. The first, m their
tech niqu es ecte d fro
characteristics, image enhanceme
nt hly correlated as exp
satellite ima ges were fou nd to be hig spe ctru m.
tion from agn etic
extract archaeological informa ns in the electrom
close frequency locatio
(Erdogar 1997).
Edge enhancement
lied in order to bring
Edge enhancement technique was app
not easily visible
out edges, lineaments and lines existing but
. Prew itt mas k is a compass
by naked eye on satellite imagery
edg es in all directions
gradient mask searching for all
(Gonzales & Woods 1993).
on the SPOT
In this study, Prewitt operators were applied
in north-east
Pa chromatic imagery (Fig. 1) of the study area
. 2c) and south
(Fig. 2a), north-west (Fig. 2b), south-east (Fig
the masks
west (Fig. 2d) directions. In all these four images,
applied,
showed the roads and rivers. After these masks were
'.11'chaeological features and ancient roads were noticed in an
improved fashion (Erdogar 1997).
in the south-east
I
direction.
in the south-west direction. of Principal Components Analysis.
of Principal Components Analysis.
I
of Landsat TM, the sixth one differs from the others. The
seventh component of the PCA (Fig. 3g), contains no
information, as the satellite image confirms.
110
111
Erdogm; N. J . .. .. .
., Ozer, A. M., Yild111111, H., Oze/, M. E, G11/p1rS. & Esm, U.
Conclusion
References
AN E THNOARCHAEOLOGICAL COMPARISON:
Edge enhancement techniques were 1oun c
. d usefu1 m Erdogar, N. J., 1997, Applica E RKENES ARCHAEOLOGICAL SURVEY AND THE LEGEND
tio n
TIIE K
Th. er aims to present the similarities between some of the findings of the Kerkenes Archaeological Survey, namely those
1ap
t :j ahmurath Ethnoarchaeological Fieldwork and the legends that still exist in the Sahmuratb village located on the
Jrom s/on es of the Mt. Kerkenes in Yozgat, Turkey. The oral history recorded during the ethnoarchaeological study in
r
eastern .
h uratl corresponds to events that took place between the Medes and the Lydians (Herodotus), the Medes and the
l
ah:em enids (Herodotus), the "Kay Kawus legend" mentioned in Babylonian tablets (Lewy 1949), the Persian sources (Taheri
Arabs and the Byzantians (Bakai 1769). The various elements of beliefs mentioned in the legends still can
J;BO) andi thetheAmevi
be found n beliefs and behavior of the modern day Sahmuratlz villagers.
Geoffrey and Frarn;:oise Summers of METU at Kerkenes Summers 1994). There is "a minimum of five ancient
Mou ntain top Iron Age City (Summers & Summers 1994, gateways" and a stone rampart envelops "the outer face of
1996, 1997), an ethnoarchaeological investigation was the wall" (Summers & Summers 1994: 13). Inside, the
condu cted in the Sahmurath village (Ergenekon 1996, 1997, average height of the wall is 2 m and the construction reveals
1998). The village is located on the eastern slopes of the that the walls were "designed to carry a massive mud brick
mount immediately below the ancient city walls. The area superstructure" which was subsequently never built
within the city walls is 53.3 km2 and is submerged today into (Summers & Summers 1994: 13). The area within the city
silence as the official pasture of $ahmurath (Ergenekon walls is totally utilized by buildings, roads, etc. There is a
1996a: 10). The Sahmurath farmers (manipulate differently) highly developed water collection and distribution system
an environment somewhat similar to that of the Iron Age City "with artificial channels and reservoirs" (Summers &
of Kerkenes. Childe, the pioneer of socio-archaeology, Summers 1994: 13). Summers & Summers have identified
suggests that "the study of living human societies as two areas, one at the north (1290 m) and the other at the
functioning organisms" is revealing for archaeology (Gjessing south (14 72 m), which distinguish themselves as important
1975: 324). Therefore, "ethnoarchaeological research locations enclosing public places, a palace and a wide street
investigates aspects of contemporary sociocultural behavior (1994: 14). The construction of the city was not totally
with an archaeological perspective" integrating "ethnographic finished when it was abandoned and the city was only about
and ethnohistoric data with archaeological data" (Kramer to survive "perhaps less than a generation" (Summers &
19 79: 1, 3; Mcintosh 1986: 148). Summers 1994: 14), before it was destroy ed by fire. The
pottery found at the site dates perhaps back as far as the 6th
Methods century BC (Summers & Summers 1994: 14). Since no
buildings, roads and enclosures seem to overlap, the greatest
Field-work techniques borrowed from social anthropology, part of the city was not reoccupied, except for two much
uch as participant observation, in-depth interviewing and smaller locations, one (at the peak of Keykavus) which was
mv estigation of
documents and maps are used in this research. inhabited at the Hellenistic period, and the other at the
Ther efore some
techniques of the ethnoarchaeological southwestern end of the city occupied until the early
method are used
as recording oral history, legends, ly rics and Byzantine times (Summers & Summers 1994: 14). The other
plac e names.
sites of the settlement in the vicinity of the city walls span
from "the Early Bronze Age (ca. 3000 BC) to the medieval"
The archaeological survey (Summers & Summers 1994: 15). The Gozbaba burial
mound in the north, the highest point on the mountain
Th e ar
chaeological investigations carried out since 1 993 (1533 m), is "about two kilometers from the city wall" with
Mount Kerkenes, (also
identified as Pteria) (Summers a "late Roman or Byzantine watch-tower" (Summers &
I( 9 7: 81-94)1 have revealed the following: the city of Summers 1994: 14). The archaeologists Summers &
er
t kenes was founded on a grano-diorite rock mass with Summers suggests that this tower may be one of those in the
(: peaks. The
Keykavus Castle (1454 m) and Kirernitlik region that gave "warning of Arab raids corning across the
7 4 m) are located to the east and south of this ancient Capadoccian plain from the south" (1994: 14) .
I 42km
southeast ofYozgat in Turkey.
112 Proce e
dings of the 31" lntemational Symposium on Archaeomefl)', Jerem, E. & T. Biro, K. (eds.), Archaeopress-Archaeoli11gua, BAR-Centi: Ew: Sei: 1.
of Kerkenes city and the Keykavus castle
Ergenekon, B. comparison: the Kerkenes
a rchaeological survey and the legend
I1aeological
Ane thnoarc
. .
Siavu sh to one of his daugh
Yozgat. It 1s a village of 335 inhabitants (1997) who live in 76 beIief m
Sahmurath that remind us of the a n d ideas ed ov s was taking ove r Pten a or Efrasiab married
leg ends piourn app ened when Cym gave birth to Shah
households. The bulk of the population is between 20 and 55 ofI<.e th is h l him murdered. His pregn ant wife
(Kay Kawus). yka wheth r e evi raids on Turkey and Ista nbu
year of age. The e are low birth and death rates, but Arabian Am . .
m Keyhiisrev (Cyrus). To reven ge his son Siavush, Shah
g mg to Hay rett
. d111..- of 717-740. But accord . y afterwards. He
contmuous ellllgrat10n to industrial centers in Turkey and th eYears Keykavus attacked Efras iab but died shortl
The legends of Kerkenes cit IJelWeen , ger, Shah Mur at (who appears agam on
Y anct earn e d villa by his grand son Shah Keyh usrev, who
Europe. An elected headman (mukhtar) and the board of Keykavus Castle the Saer' a
l
a! Gazi6 to conquer the was then succeeded
8th c.) asked Batt on his mother's
elders ovem the village. All of the villagers are farmers by the scene at the by then also had a well of golden
inherited this fight to reven ge his grand father
kavus, which ab fled to the land of the
profess10n. Agriculture is almost totally mechanized We learn about Mt. Kerkenes earliest castle of Key then held by Cluistians side (Taheri 1980: 512). Efrasi
from H.itt. could catch. It was a pool (Tahe ri 1980:
(Ergenekon 1998b). Products of dry ite tab)... n o o ne the Romans (Anatolia) and was killed in
agriculture such as (Gurney 1995). The Hittite King who t . of the Moslem "martyrs" and
' rave1 led ... IJeeS s). The graves ng about the existing
wheat, rye, barley, vetch, lentils and chick-peas are cultivated chanot in a War /Q.yzanUan Keykavus fort1 fi
1cat1 0ns are 511-518). 8 Taheri mentions nothi
and harvested. Viney rds that once yielded high quality f " .
t (Ergenekon l 997 a). It mu st be
Mt. Kerkenes and then climbed up Mobht Daha evidence this fac and the religious
grapes have een left idle after higher profits were optained is Kerkenes -, sanctified it and went to
-
Which beli. eved to ak meets its When we turn our attention to the story
visit ors wit h
. Ankuw a e that the Gozbaba pe understand that
menuon ed her
s we
from mechamze agnculture and remissions from expatriate 1995; Ergenekon 1998a: 2). He may have"
(Gurn ey ng flights arou beliefs derived from
nd them not Babythelonian tablet
II bees that make warni
. Mount Kerkenes Great Road" runnmg to the east and "The us of the battl
C ds rem ind known in
offers a rich variety of bushes and herbs. The water cisterns 0 f i Road" in hese legen (Samas), the Moon (Sin) and Istar (Venu s) also
. .
the souteas by th
@
agers (Ergenekon l 9
) first between
the Medes and Lydians, then
the Lydians and the
ari sect of Turki ye.10 Nabunaid
the ancient city are now ools of mud fostering frogs and The histoncal ongms of this village are rooted . Christianity9 and the Hayd
this first battle (which
hree folk . According to Herodotus, se he hailed
leehes. These places which have international fame are
I \. Achaemenids (Kay Kawus) was killed in a civil war becau
legends told by the villagers, one about the eclipse on May 28th,
villag , he o five years) ended with a solar his country
beheved to cure skin d iseases such as scabbies, hemorrhms ther went on for higher his uinity than
d by Thales
the only god, Mard uk, of
'd one about Kerkenes city and the third one ab already calculated and announce
. . . .
3
out the Ke k a
Y vus 585 BC, an event Iraq, claimed that he ascen ded to the heave ns on a throne
and 1 ferulit. It 1s believed that one has only to bathe in them C astle. The legends recorded in Y e, Arie nes, the
.;;: ahmurat1I bet ween in Miletus. As a sign of peac
to enJOY therr regenerative benefits (Ergenekon 1996a: 10).
1995 10 his citizens drawn by four eagles, which
King
shock ed the Mard ukian s who
ian
and 1997 (Ergenekon 1996b, 1997b, 1998a Lydian King Alyattes and the Med
Horses and donkeys have been totally replaced by tractors,
) were also daughter of the thought that the heave ns
dedonly belon ged to their god, built
recorded here by Bittel (1960-61) in 1908 . A ' ibly settled at the newl y foun
. ccordmg to oral Astyages, mairied and poss by archa eolog ical excav ations along
cars and minibuses. Animal husbandry has decreased observatories11 supported
een Arien es'
history, the village is said to have been founded d battle took place betw
. 500 ('Ii t city of Pteria. The secon Babylon in Iraq to
bly, leaving only three families who keep flocks for
cons1dera the Tigris, carried the imperial palace from
the
. h_r
2500) years ago by t ee bands of the Turkish
Ogu z ri '
: brother King Croesus
of Lydia and Cyrus the Grea t7 of
1949) , learne d how to interpret the
commercial purposes. A few families have bee-hives in their Balh in east Iran (Lewy
who settled at an ancient site founded by an d . .
Persian Acha emen id Dyna sty. When the Medes Dynasty and the services of
stars and did not need
vin yards. "The vill gers fllow a folk calendar adjusted to called Sah Murat (Dogan 1995: 257; Ergeneko 1 future according to the
by their cousins the
. . 9'. their Empire were overtluown claim ed that the one and only universal
their ecosystem, which dlVldes the year into four seasons ", A combmat1on of all of these stories is told by v1 the Mardukian clergy,
of Cyrus the Great in 547
.. agers as 11 Achaemenids under the leadership he was Sin's proph et (Lewy 1949:
months are referred to in numbers" and "the year starts in II
. .
fo ows.. .. The Greeks lived m Kayseri, the Shah at Kevk God was Sin and that
the prophetesses of the
BC, ng Croesus of Lydia consulted the runic writin g in his dream,
September, when" villagers start to prepare their fields for the 33-56) and that he had learnt
e would fall apait
They traded with each other. Camel caravans went throug
": Oracle at Delphi. They predicted that an empir it before (Lewy 1949: 68). He also
next ye' (Ergenekon l 998b). Traditional houses built of though he had not known
(K1z1hrmak).
.
road by the mountain.4 In those days, Turkey and Iran were if King Croesus crossed the river Halys Mard uk asked Nabu naid in his
grano-d10nte rocks from the mountain have been rep lace d said that one night the god
crossed Halys
.
one and the same country. When the two brothers o{ Iran Interpreting this prophecy in his favor, Croesus city Harra n.12 Nabu naid told
wit
'hb' nck houses and tile roofs with new floor p Ians. dream to conquer Sin's holy
ed and the
. .
fought each other, one of them came and settled with his but was chased back by Cyms. Sardes was occupi s Empi re. Mard uk then
Marduk that the city was in the Mede
TradiIlona 1 weavmg and dress have been abandoned. All of way back,
desc ndence at Kerkenes as a king. He assigned his two sons
Kingdo m of Lydia disappeared the same yeai. On his er the whole empir e one
conqu
the vil agers are related to each other through endogamy to bmld walls around the c1ty. Th ey started as young men at according to Herodotus Cyrus warred with Queen
predicted that Nabunaid
Tomris
would
of the
the Great .
was realiz ed by Cyrus
prefng to m either their parallel or cross-cousins. day (Lewy 1949: 82). This
the same point and worked their way in opposite directions. Sakas east of the Caspian Sea and was killed there. noma dic Chald eo
was introd uced by
Fanulies are patnlocal and patrilineal. They belong to the en they met again The Sin (moon) Cult
both had become old men with white The name Keykavus is intriguing because it is the title inated the desert,
Sunnite sect of Isla m and hold many pre-Islamic beliefs, such accordi ng to Aram eans to Anatolia. While the moon illum
h Ir and beards. Meanwhile, the Kerkenes King's brother, the attributed to the last Babylonian King Nabunaid
night, Merc ury, Venus , Jupiter, Saturn
as the star, ancestor, wood and water cults. Food, pastries and King of Iran, had a son called Murat. When he grew up, his some tablets and cylindrical seals (Lewy 1949: 107). He is people who travelled at
them their direct ions. The sun, on the
sweets m ade of grain, dairy products, eggs, vegetables and father told him to go to Kerkenes and kill his reigning uncle known for his polytheistic beliefs about the Sun (Samas), the and the stars showed
domi nant divini ty for agricultural
grapes offer a variety of dishes and drinks. Poultiy and meat there. The son entered the city disguised as a hunter. But when Moon (Sin), Venus (lstar) and other heavenly objects as other hand, became the
66). Kay Kawu s, who was a
are consurned only on important occasions such as religious Murat was caught he confessed to his uncle that he was sent opposed to the monotheism of his native Mardukians of Neo societies (Lewy 1949:
. . ramea n origin , was some times
and social celebrat10ns like engagements, weddings, maIe tere to kill him. Instead of punishing him, the king disa rm ed Babylonia (Iraq). The tales of these events are recorded as Babylonian of Assyrian-A
. .. . "!star 's Husba nd" to show that he was
Clfcumc1s1 ons, farewell and homecoming parties before and him the Iranian legends by Persian authors (Lewy 1949: 29, 108-109). called "!star's Son", or
. . nd married his cousin to his own daughter. After n (Lewy 1949: 77, 79).
after Ilgnm ge to ecca. Politically the villagers adhere to weddmg, they settled in this city (village), hence the nam e Accor ding to Ebu-Cafer of Taberistan (Taheri 1980) "Kay" is a semi-god by way of famil ial relatio
goddess lstar honored
the nght wmg Nationalist Action Party and the recent Sahmurath.5 a title meaning "The Good" in Persian and is given as a title In another version of the legend, the
that she would carry him like a tiara on
reIi. g10us We1fare Party, the social democrats being m the However, according to another version the king and his only to Shahs (Taheri 1980: 443). Taheri states that Shah Kay Kawus by saying
. . . Like Mose s at Mt. Sinai (Sin's
nunont . There 1s a 100% rate of schooling among the new people were killed by warriors who came out of the bags of Kykavus was Shah Kay Qubad's son. Shah Kavus had many her head (Lewy 1949: 93).
at the moun tain tops, Kay Kawus
generat10n, but the few that graduate from universities seek the camel caravan that entered the city (perhaps through the wives, one of whom was the daughter of Efrasiab, the King of mountain) and Solomon
. place s were closer to heave nly gods.
pro1ess10na
c 1 JObs in cities. A local dialect of the Turkish Contrary Road and the Kayseri Gate in the southeast). The Turk istan. Keykavus sent his son, Prince Siavush, who was also believed that high
.
invaluable help. Fro m the archaeological survey, we k now that Kerkenes contains the remains of ancient water-reservoirs
,
' According to Babylonian and Persian sources Keykavus is a title attn'buted to the 1ast Babylonian son
King Nabunaid and his successor, Cyrus the Great's The Father, the Son and the Holy Ghost.
10 considering the last as God.
Cambyses (Lewy 1949; Taheri 1980).
. The Hay dari Arabs of Hatay believe the Prophet Solomon, the Prophet Mohammed and the Prophet Ali,
' 11 stairs around it outside.
Mt . Kerkenes was both on the King's Highway and the Silk Road The Sin temple he built in Hanan is called Ehulhul (Lewy 1949), which is a conical ziggura t with
5 .
As mentioned above potential maniage panners for Sahm urati I youngsters are parallel or
cross cousins (Ergenekon l 998a: 3).
12
Hanan by Urla at southeast Turkey.
114 115
and the Keykavus castle
the legend of Kerkenes city
archaeological survey and
Ergenekon, B
. .
rcIiae .
arison . the Kerk enes
.
ological comp .
j\11ethnoa
. t: How w e
cI nt osh J
. 1986 T he Practical Archaeo
log1s
Koy M
the local mosque and its minaret at $ahmurath today. through periods of fainting (epilepsy) and t 996a, Arkeo
metri: Kerke nes ve Bir
; ' ;
hat w know about the
Past, The Paul Press
B" 1 k o
Kerkenes, which was considered as a holy mountain by the consciousness, going through short periods of
hus lo
st ergeneko'. . .n Sila Y 1l l , Say1 2, Dogus Ma :
tb as1 Yoz gt'
.
;.
lt d 146-173.
. . me n t ia tt
de J(iiJturu,i es Dagi: Bir Dag Efsanes1
mn
1996b, Kerken ac-i Nebeviyeye
Hittites, must have been a suitable place to worship celestial moved his seat from Mecca to Medina, he chos hi h
e g Q
ergenekon ' B., nbu1, 136- 144.
B
Sm'd N ursi, 1991 Otuzbirinci Soz: Mir .
and towers to call his followers to pray to All h Pia
I sta
" '
Ka
D ai'rdir, Risale-i Nur Kiilh
. ,
pe1nde,
bodies and honoured as holy during the time of the Medes rg1 s1 yatm dan
Atlas De
.
a es Sahrn urat h
Empire under Kay Kawus. Cyrus the Great, who was Kay ascended into the sky after his death to join All ah ( '. nd he ergenekon B., l
.
997a Arkeometri:
Kerken
593-626.
1? , The Mountain Top c
Sai N11f1i Say1 3, .
ogus M atbaa 1 Yozgat' .
ity
mers, G. & Summers, F., 1994
Kawus' political opponent and successor, believed in the one 1991; Smw; 1997). '.
.. . sii, in Sila
'
Y 1J l , . .. .
. ..
o
J( p
ru
7b, Ekoloj ik Deg1sm e : Geling ullu BaraJl Sum
Cap adoc ia, Ark eolo )I ve
K k e Dag (Yozgat) in
universal god, Marduk. His son and successor, Cambyses The Sin temples were symbolized by a cresc
e nt T ergenekon I B 199 l,
, Y 1l ll'.
Say1 5,
(Kay Qabus), burned down the moon temples when he came ziggurats were conical in shape and had stairs
o uts de
hese
l
that v e Es n g Kasa bas i,
8a Keykavus ve
S1la Der gisi
Kerkenes Efsanelen, m
:
na 1 , Say1 62/63, 3-20.
&
Test Trenches
to power (Lewy 1949: 94), but belief in them lingered on. went all the way to the top. Today, the domes of t
ergene ., 199 , G . S ummers, F ., 1996 , Survey and
y,
hem ko n B at. Summers,
tba as1 ozg on the 1996
Today, the Sahmurath villagers still remember the game of and the narets in Sahmurath and the rest of the v:ues S'la y'11 2,
Say1 6
'
Do gu Ma
.
m Sahmurt1 i Vll 1 age v
at n.er
.
kenes Dag-. A Prelimina1)1
Rep ort
rcha eolo gy ara.
ges I b, Eth noa Archaeol ogy at Ank
jumping over the fire called "Sin-sin" their fore-fathers played surroundmg Mt. Kerkenes have crescents on their
ro ofs ergenekon, .
B.,1998
in Turk ey, in Pmceedmgs of the
Season, British Institute of
Summers, G. & Summer '
ns
.
atio
(Ergenekon 1996a: 14). minarets are tall, cylindrical towers with winding stairs:
. . in side 1'.he by Kerkenes
E xcav
?' ympo s1um, University of
s F., 1997 , T he Kerk enes
B
Dag
. .
ntls h
S 1997
.
th e British y Eas . . .
. s.m g, G " 1975, Socio-arche
ology, Cu11ent Ant hrop olog at1,
Taberi's history, Shah Cyrus chasing King Croesus west of observed in Turkey (Tulunay 1993). A red scarf is put disi: Peycambenm1z111 Hay
around GJes Suruc, S., 1997, Kainatin Efen
Halys back to Sardes inHerodotus' history and the versions of the neck of the bride and a green one around the groom 16,323-3 43. Vol. J,11, Yeni Asya, Istanbul. . .
.
at . .
s1, Cllt I,
E . C. M. B. c., 1980, Tarih-i
these mythes told in $ahmurath. According to Taheri (1980), weddings. Here the red symbolizes both the female and <l o tus I.76. . . Taben Tercume
the Hero llca twn s of Ta beri
1979 Ethnoarchaeoloqy:
lmp '
Shah Keyhusrev, who conquered Turkistan (530-532 BC), color of the sun, while the green symbolizes the male and Kramer, C., (ed . ), Konya.
, Co1umbia Univ. Press ,
the '
t. of
ve Gilne, Serm'nar, Dep
Archa eo logy
built a city called "Keykred" (the location of which is belief in Islam in Sahrnurath. Today, the Persia n word Ethnocraphy for
Tulunay, Y., 1993, Evre n
ara.
unknown). He asked the Prophet Solomon to help him and "Sitare",meaning both star and Venus,is used as a girl's name NY . .
Kaus Archaeomet1y, METU, Ank
H 1949 The Babylon
ian Background of the Kay
Solomon sent him his giants. They were ordered to build walls and "sin, sun, san, su" as syllables in girls' names s e!\ as
u Lewy, ., '
II, 28-109.
of 7 parasangs in length and 4 f loors in height.However,Allah "Aysun"13 relating to the moon in $ahrnurath and other places. Legend, Orientalia S V
sent an earthquake and the city was destroyed (Taheri 1980). The $ahmurath villagers use the expression "I will make you
This version and Summers & Summers' archaeological survey a crown on my head" when they want to show their gratitude
show similarities. The very name of the Castle Keykavus leads and great respect to somebody, which reminds us of Istar's
us to seek a connection between this Median City of Pteria and expression to Kay Kawus: "I will carry you like a tiara on my
Babylonia. As mentioned above, the name Keykavus is head". This term is widely used for mothers in $ahmurath and
provoking because it is a title given to both the polytheist elsewhere in Turkey.
Babylonian king Nabunaid, as well as his political and F rom the Keykavus Castle the Sahmurath villagers can see
religious opponent, the monotheist Mardukian Cyrus's son, as far as Mt. Erciyes in the south and Hattusa in the north. It
Cambyses. The former is, however, spelt Kay Kawus, while must have thus served as a proper watch tower for the Pterians
the latter Kay Qabus (Lewy 1949; Taheri 1980). both for reasons of defense and visual communication. The
The beliefsNabunaid and his opponents represented and the palace remains identified by Summers & Summers (1996,
symbols they used appear as national and religious symbols in 1997) join a wide ceremonial street in the east-west axis. Mt.
Sahmurath today. Briefly, Shah Keykavus believed the trinity Kerkenes no doubt was an appropriate place to watch the
of Sin,Samas and Istar and other celestial bodies to be holy,he movements of celestial objects.
rebelled against the Iraqian god Marduk's religion (Lewy It is still too early to say anything about cultural continuity
1949: 31), he claimed to have ascended to the heavens seated between the Pterians and the $ahmurath villagers, because
on a golden throne, he built a tower like temple-observatories (although we know a lot about Sahrnurath) the archaeological
to watch the holy stars, he went through a period of dementia survey so far does not lend much evidence about all aspects of
in Yemen (Lewy 1949: 32), he claimed that god, Sin was the their culture. However, it can be concluded that the religio u s
"only universal god" and he was Sin's closest associate on beliefs and some symbols o f the $ahmurath villagers tod ay
earth (Lewy 1949: 33-56),although he didn't know how to use have certain counterparts in the beliefs which are known to
the stylus on a tablet, he claimed to have learnt to use it in his have existed in the 6th century BC in the Media n the and
to ry
dream (Lewy 1949: 68).He moved his capital from Babylonia Persian Empires. Today the Sahmurath villagers' oral his
la nd
in Iraq to Balh in east Iran (Lewy 1949: 31). testifies not only to their ownership of Pteria as grazin g
u ally
The Prophet Mohammed, who lived in Arabia a thousand but also that the villagers identify themselves intelle ct
years later than the Pterians (570-632),is highly respected and with the ancient inhabitants of Mt. Kerkenes.
followed by Sahmurath villagers today. The Prophet
Mohammed has similar properties to that of Kay Kawus, in References
spite of the fact that Kay Kawus was a polytheist, while
Mohammed was a monotheist. Mohammed communicated Bakai, K., 1769, Seyid Batta! Gazi Destam.
ag
with the one true god "Allah" in the Hira Cave of Mountain Bittel, K., 1960/61, Legenden vom Kerkanes-D
Nur. The revelation took place in his dream; he was illiterate (Kappadokien), Oriens 23124, 29-34.
orgu0
until Allah sent the holy script of the Koran to him and Dogan, D., 1995, Sahmurath Koyil, in Sorgun, S
ordered him to read, which then he could. Mohammed went Kaymakamhg1 Killtilr Yaymlan.
13 The syllable "ay" means "moon", the syllable "sun" means "dedicated to". Other examples of such names are Aysan, Aysu, Aymelek, Aylan, Ay dan an AYsel
d
117
116
H EOLOGICAL CORING, USING CRYOPROBE TECHNIQUES
AR C A
GE o AN EXAMPLE FROM THE AMERICAN SOUTHEAST
I v 2 3
GARRISON, E. G., SERMAN, N. & SCHNEIDER, K. A.
Franklin College of Arts and Sciences, Department of Geology, The University of Georgia, USA,
1
S' location and characterization in wetland areas or saturated soils has benefitted from the combined use of ground
) and cryogenic soil probes. This synthetic protocol has been developed and successfully applied in the
'ating radar (CPR
penerican South over the past five years. It has been applied to both prehistoric and historic archaeological sites at working
mhs of 2 meter s or more. The CPR systems are standard models familiar to many workers while the cryoprobe as a variant
e
h types developed in Germany and Switzerland in the early 1990s.
of Thee single greatest difef rence between the U.S. and European version is the complete reliance on nitrogen as the coolant in
due to the elevated ambient air and soil temperatur es. Rapid recovery of intact, relatively uncompressed,
the Am erican South
n
sedime t colu mns has been obtained for in-field comparison with CPR data of adjacent archaeological deposits.
Th is pro tocol has increased confidence in the inte1pretation of the radar data enhancing its role in the overall
ch ara cterizatio n of the archaeological site as whole.
KEYWORDS: GROUND PENETRATING RADAR (GPR), CRYOPROBE, SEDIMENT CORING, SOIL FREEZING,
ETOWAH, SCULL SHOALS, GEORGIA, USA.
Introduction Cryoprobes
To characterize archaeologically interesting alluvial deposits "Soil freezing" Cryoprobe technology used in archae
we have used geological coring to obtain soil/sediment samples ological studies has a recent history. Swiss and German
from archaeological sites. A recently developed coring investigators have, in the late 80's and early 90's, deployed a
technique based on cryogenic technology was used to extract variety of coring techniques that freeze saturated soils
nearly 2 m long sediment cores. The "cryoprobe" technique (Capitani 1993; Grebothe et al. 1990; Hochuli 1994; Lassau &
was first used in Europe in the late 80's and early 90's (Hochuli Riethmann 1988) (Fig. 1).
1994). It uses a closed 1" (2.54 cm) stainless steel pipe through The technique is patently straightforward. A hollow pipe,
which a cryogenic fluid is circulated. This pipe, after insertion closed on the insertion end and open at the top, is driven into
in the sediment, freezes a continuous coating along its length the soil by manual or mechanical means (Fig. 1). Into this
which, when extracted, yields an undistorted record of the pipe a cold substance - dry ice, liquid, or gas such as C02 or
lithostratigraphy. This sediment column exposes the depos nitrogen - is introduced in order to freeze the surrounding soil
ition al record for use in a variety of subsequent analyses - to the pipe.
pedological and geochemical. At archaeological sites in The reasons for using this approach for stratigraphic
Georgia the first American prototype was successfully sediment/soil sampling include (a) recovery of relatively non
developed by the University of Georgia Geology Department. compressed samples in contrast to those recovered by
CJ
CJ
C:]
D
0
l
0
a b c
Fig. 1. Insertion, application of freezing media, extraction of cryoprobe (from Hochuli 1994).
Proceedi
ngs of the 31" lntemational Symposium on Archaeometry, Jerem, E. & T. Biro, K. (eds.), Archaeopress-Archaeolingua, BAR- Centi: Ew: Sei: 1.
h
Garrison, E. G., Se1111an, N. & Sc/111eide1; K. A
. Geo archaeological Coring, Using Cryoprobe Techniques - An Example From the American Southeast
tr eated
as a reverse of a heat-loss problem in thennodyn allllcs
. .
Necessarily, the freezing time
is muc h sI o wer th a
is roughly prop or .
t1onal
In clay8
n sands
.
/'
.,. were deforested and used to produce cotton for the textile mill
J w & Jaeger 1986). In this case the process 1s cooli ng over the abso lute coolant temperatu to one along with subsistence crops for the workers, managers, and
( Cars a re or
,.H,.o unding medium rather than heating it. The fommla Freezin g Time - 1/ absolute
coolant temper other village inhabitants (Hunt 1980; Stovall 1 984) .
the Su . . ature
is as follows: The freezm g l ime is not linea
for this r follow g, . We extracted two cores of nearly 1 .6 meters each inside the
m instead
h y erb o 11c ath such as seen in mill and along the natural levee on the Oconee River (Fig. 5).
two coolants of _ 0 a
20
40 C. Th e llme required for the a nd These sediments were used in geochemical and plant
-40C cool 1
-20 C c oola n t. Lower temperat mqmili
macrofossil studies. Macro-pedological study of the cores
. ure coolants cut f at of
reez1ng ti
.
0
is the "heat" lost, in our case "cold" gained; Tcr the mcrea se the strength of the
frozen soil and . Ille, delineated the anthropogenic and natural deposition in the
h ere T . Increase
w perature of the "country rock" e. g . the soil and TP, the . g radrns. One observes that mill's sediments at a depth of up to (1.6 m) as determined by
the freez
freezm the
tem mg lt.me .
18
.
rature of the probe, all in degre es Celsius. At the
proport10nal to the square of the cryoprobe borings. The pedologic profiles delineated the silt,
freezing radius as
ternpe
probe
0 . . Well ( g>).
::::20; Tpr- 197 t h en T ==-177. Usmg an
exponentia 1 curve An exact calculat10 n of freezing time t clay, and sand percentages are shown in Fig. 6.
I'f Ter ' f 0IIowi. ng
Joss, as Carslaw and Jaeger su s Khakimov, is:
f0 he at
gge t, we can expect a
t'. \or of 2 "dro " p in the soil temperature -88C , and Con cl usions
c
at 2 cm to
.
c m th e temperature 1s onI y -30C. Factors influencmg . the Ir== t'
at 4
C, Yt ro2f'A.1
fl w of cold include water conten t - the more water increasingly
ake or o hea t conductivity of the frozen soil From the mid-1990's the authors have
where: "-1
: : reater t e latent heat and slower heat loss, slower frezing incorporated technological advancements such as the
=
g
=_in meters
SroJt
.
r, it 1s the water content that detemunes the sohd1ty of cryogenic soil probe into a geoarchaeological methodology
r
w
Ho eve c, =specific heat of the froze
n soil in Kcal/kg:C
sample after it freezes to ice. designed to characterize archaeological sites such as Scull
the core . Yi = unit mass of the frozen soil in
ognized soil expert and pioneer in the science of so ti
3
Shoals and Etowah.
The rec
Kglm
Fig. 2. Map of the Etowah site (from Thomas 1894).
external radius of the freezing pipe
.
w as the Soviet academician, Kh. R. Khakimov. His
f eez ing
ro =
(m)
rt b ook entitled "Artificial Freezing of Soils, Theory and t' determined from tables, an empirical
rctice" (1957) laid the groundw?rk
=
th ose encountered in the construction of the Moscow powered textile mill (Fig. 3). Typically,
'
f sediments. Silty soi ls freeze outs1de-m" , e.g. external At Etowah the radar has prove to be the most eff ectiv e
soils o
al as the free H20 is squeezed radially outward. geophysical tool for examining the interior structures of the
"
to intern
oil freez es top-down with an upward migration f free 13th century mound . Evidence suggests at least two
Sa ndy s
810. Even in unsaturated sandy soils water vapor nugrates constructio n phases and a possible interior feature. Co ng has
ri
freezing boundary. produced sedimen t profiles to -5 meters withi n the m ound.
oa rd th e .
1 The freezing boundary is termed the "freezing radius", z. the
These profiles have been used
to (a) characterize
aboring the audience with the differential equations sediments used in construction and (b) calibrate the
With out bel
balance, it is assumed that there is a linear relationship geophysica l data (F ig. 4).
he at .
and the distance, R, from the freezing pipe to a At Scull Shoals both magnetic prospection and radar h ave
c
between it
the soil, where T is the same as b efore freezing so: produced useful data. The geophysical
data indicates that
the
point, in
present ground surface represents a ui vale nt
level in the mill eq
R ==a?f' to the first-st01y of the 19th century structure. What th e GR
data show is a signi ficant build-up lluvial
of overbank a
hi
a is a constant for a specific soil type. Khakimov wJt n the mill due to increased flood frequency
at
where
d epos1ts
120 121
Garrison, E. G., Se1111m1, N. & Sclmeidei; K. A.
'toV' to be
Inc. fre docu men ting past
mg a wduable arch w ing fart her
gical in,.'1igotion,, thu' allo
Grebothe ' D., Lassau, G., Ruckstuhl y in the context of geoorchaeolo
& s eifert ' M 'truc t the collu,ial c hronolog
3.8 p' to '"'" "
199() #ng hel
und er hum an and/or climate
impact.
.
Thay ngen SH m
vve1er: Trock eneissondiernn 19 9,
Jahrbuch of landscape chan ges
.,
a
elucid tion
-
un Fru- hgeschi
I
chte
Dark Brown Silt-Clay de1 . Schwe 1zenschen Gesellschaftfiir U 1 G. SOIL
NCE, OSL, OPTICAL DATIN
73, 167-175. OLO GY, OPT ICA LLY STIMULATED LUMINESCE
ARCHAE
l(EYWORDS' GEO IMENTS, IRON AGE, GER
MANY .
Hochuli, S., 1994, Unter den "Bahn 2000" . EROSION, COLLUVIAL SED
, Gefrie rkem
5.0
ch we1. z 17 ' 25 3
Bohnmg in Kanton Zug' Archaolog1'e de1. S t luminescence signal keep
Hunt, C. C., 1980, Historical Su1vey Rep0
radiation damage and a laten
Fig. 5. Clyoprobe 1 .t, S cull Sho- als
0.
Introduction grain s are exposed to daylight as
Fig. 4. Sedim ent descri ption . growing with time, until the
core at extraction, C omm uruty, Greene County, Georgia' USDA Forest . 2). After being transported
Atlanta.
(Fig.
of Mound B cryoprobe c ore.
a consequence of soil erosion
length is 160 cm.
. Service,
ered
s of Middle Europe have suff ral grains are buried by
The old settled loess area
. .
mine
and redeposited downhill the
Khakimov' Akh R" 1957 , Artificw l Freez ing of so11s. Theory e the onse t of agricult ure in
t luminescence signal
considerable changes sinc other colluvial material so that
the laten
Additionally' the prot o t ype c1y ogeruc conng teehniques
.
s of the u s SR
. . pletely
ted in partially or com
p
and Practice' Academy of Sc1ence
Neolithic time, documen are samp led, e.g., in the course
starts growing again, until they
co pled wit geophysical, geochemical, macropedological,
.
s exposed to soil erosion by
p
; _
rrnafrot Institue in V.A. Obrnche, Translation
b ; isrei truncated soils in relief position of an archaeological digging
cam paig n. In the luminescence
. soil material detached and sample is measured
a? plant Illlcrofossil studies provide a robust geoarcheolo- ogiam m Scienti fic Translations ' (1966) water. On the other hand, ence of the
laboratory the natural luminesc
Lassau, G. & Riethmann, p" 198 8, Trockene1ssondi
.
. onmental studies. At Scull : esse s was redeposited, at least irradiated sample
gical methodology for pa1eoenvu erun g, ein transported by slope wash proc addi tiona lly,
as well as the luminescence of
.
. .
topo graphical depressions.
:;
in
Shoals ' Georgia ' this convergence of modem technologies has Prospe uonverfahr en im Seeuferbereich, Jahrbuch der partially, on the foot slopes and tive dose proto col).1 From the
. . . _ material from else portions (multi-aliquot addi
lead to ms1ghts into t he dynaffi!CS of paleoenvrronmental of lost soil added dose the
nshcen Gesellschaft fiir Ur-und Friihgeschicte 71, As corresponding remnants ence grow th with
.
sent an archive with a correlation of the luminesc
changes m a relatively restricted locality. At Etowa h, the where these colluvial depo sits repre also determined the
- Hav ing
paleodose (P) is extrapolated.
empasis has been on the characterization of prehistoric
. ating phases of
M oshkin, Y I., 1933, The Problem of the Formation of Ice memory of the landscape histo ry indic year - the time
.
ral dose in situ per
. 1 organ-
dose rate (D) - i.e. the natu
architecture and its role m Nauve Amen. can socta tals lity in the catch ment area (Fig. 1).
m Frozen Soils, Preventing the Swelling of geomorphic activity or stabi ding P by D.
Rru way Subgiade, Sb. NIT. Puti NKPA
t since burial is calc ulated by devi
Optical dating
BACKSWAMP SEDIMENT
100
and BURIED SOIL
e 1so
d:
Optically stimulated luminescence dating (also calle
optical dating) offers the oppo1tunity to determine the time
ir .. .. ....
<>.
200 elapsed since a mineral grain of a sediment was last exposed
cool.. climatic History
WEATHERED GNEISS t
to daylight and has been buried since. Luminescence is the
SAPROLITE Wet ... .. ...... . .
measurable expression of a radiation damage, which non
Time
250
conducting materials, like quarz or feldspar accumulate, ...
I
according to their own natural radioactivity and their
Fig. 1. Colluvial sedim ents near
300 archaeological sites:
urroundings, mainly the covering and underlying sediment
scape chan ges
ting past land
an arch ive documen
ay e rs. As long as the grains are sheltered from light or are not
in the catch ment .
. I 350
strongly
heated (as e.g. in the case of a prehi
stmic fire place)
Sand %Clay
erative, multiple-aliquot, single
Reviews on the different me tho ds to detem
escence dating (additiv, regen
odose (or equivalent dose) in lumin
( 1998).
une the palae
( 1998)
Fig. 6. Sedimentfi'actions, SrnllShoals core #l.
aliquot) are given e.g. in Aitken and Wagner
latent p . . rt herwe
a1h e v i 1 p file with deposited either in the Roman Period or the Iron
Age La Tene
collu a
honzo n, Ih us b econung a clay enriched B i- in the uppe r part. The
age = . rect carbonife rous material
luminescence 0 .
honzo n corresponds to the unweathered
I-honzo
n. The
bo om
;
an ilty
ical cons isten cy f or the lower 4 Period.
ii b) . . ca1carou ows chr
onolog
c
DI . . 9,_; s01l eros10n progrades from top to bottom
s IGes s
, c1a11n g s h
0 . a comp et
!
a) of correlatmg colluvial sediments would . . eP
c
.. . . ex hib11 a
ve , OSL'
Iayenng m a reverse order of the described .
u m-situ sequ l1i
: ence ample preparation:
r Y: 1 -10 Gy/min
' 90S /90
some o f t h e ceI Jars - once dug some 2 m d .
'
An insufficiently bleached signal however may cause age to ographically deep pos1t10ns. Soil erosion stripped
off all instrument: Rise-Reader DA12
overestimation.2
so an me of the loess material from the hilltop,
the upper
preheating: 2 min/220C
The proper signal reduction of slope wash sediments was n n e slop e and even some of the lower slope, again
near infrared: TEMT484-diodes (880 !:!. 80
run)
. '
md1catmg re !ief lowering of at least l m.
checked b y applying OSL-dating to 22 indepe ndently -
stimulation:
As evealed by drilling cores and excavator ditches the foot
radiocarbon or archaeologically - dated, colluvial samples. detection: photomultiplier EMI9235Q
slope is overed by 4-5 m deep colluvial deposits. This means
The ages showed good agreement within the erro r limits of shine down: ?:. 60 sec
that dunng the Holocen e the relief was equalized by soil
the techniques, with transportation distances of only 50 . BG39+2xB G3+ GG400, KRBETSCHEK et al. (1996)
er s1on and accumulation by at least 6-7 m
violet-blue:
metres proving sufficient for succesful OSL-dating (Lan g ? as compared
with a total of 25 m height difference between hilltop and 390-450 nm 3 mm each
_
1996). Therefore optical dating was applied for geo KRBETSCHEK et al. ( 1996)
BG39 + GG530,
valley floor. yellow:
archaeological investigation to retreive the memory of
The Jd st colIuvium, found in the deepest position on top 520-620 nm 3 mm each
colluvial sediments in the test area Bauerbach. ? . DULLER (1994)
of the n-s1u-s01I remnants, is of a light colour, decalcified analysis: Analyse 5 .22
As the Bauerbach loess derivates are rich in silt and GRON (1993)
and qmte silty. Further pedological studies are necess ary to Simplex 8/93
feldspar content we used the standard Heidelbe rg multiple AITKEN &XIE ( 1 992)
co nfim1 the field hypothesis that this colluvium, which has special trea tmen t : '
'late l ight subtraction
aliquot additive-dose protocol for polymineral-fine-grain
analyses (Table 1; see also Lang et al. I 996), for which
not been OSL-dated yet, is borne from the AI-horizon of the fading tests: uniITadiated + P-irradiated aliquots:
sample preparation (further developed from Zimmermann luvisol.
storage: room temperature
In the field a total of 8 different colluvial layers can be
1971, in Aitken 1985) is carried out under strongly subdued
distinguished, the lower ones being decalcified and - apart measuring: ?:. 3 months after preheating/OSL-measuring
yellow-green light, and luminescence is stimulated in the near
from the above mentioned oldest one rich in clay conten t.
infrared around 880 nm . Shine down curves of at least 60 s are _
before measuring for 2 min at 220C and fading tests are rooting trees once grew into the underlying not yet OSL P-counting: Ris0 GM-25-5
s
carried out after at least 3 months of storage after preheat. dated Al-borne colluvium . Within the dark layer two nail
were found, which might be of Roman age (pers. com. H.
_I-spectrometry:
These however, have n o t yet been practiced for the samples
Eberspacher 1997). lab o rato ry (low level): Ge-detector
presented here. -
Signs of bioturbation mark the upslope transition, wh ere field: portable NaJ-detector
-
the colluvium no longer rests on Et-remnants but on bare NAMBI &A ITKEN (1986)
Test area he !ose-rate-calculation : dose conversion factors
loess. S mall burrowing animals mixed loess material into t
overlying colluvium and vice versa. a: a-counting, y-spectrometry
The study area Bauerbach is located in the Kraich gau Hills -
sa
in southwestern Germany, some 50 km south of the city of Within the uppennost metre of the colluvial sediment P: -counting, y-spectrometry
,...._
layer of anthropogenically dumped, extranous m ate rial R IESE R ( 1991)
Heidelberg (Fig. 3). Situated on the eastern border of the y: y-spect ro m et ry
,__
Rhine Graben, the Mesozoic geological base ment was
intem1pts the n01mal course of slope wash deposits. PRESCOTJ' &H UTTON (1994)
cosmic:
-
Exp eriences of dating insufficiently bleache d sed ime nts with elabo a e OSL metho s are
rt d reported e.g. by Fuller et al. ( 1994), Lamothe et al. (1994) and
ly bleached fine grained sed iments. Table 1. Multiple-aliquot additive-dose protocol fo r the analysis of polymineral fine-grain
fractions - brief overview.
Singhvi & L ng (in print). S o far n o reliable pr oto ol e
a c xistsforrou1 n
i de a1in gofinsuflicient
Kadereit, A., Lang A & m
, rragne1; G. A.
i
Dresden\
_j')
v
',.
ch.
the test site Bauerba
Fig. 5. View from the SE of 4-5 m thick coll uvial fillin
g.
hill slope cont ains a
to the south of the d nw the hill top.
The drainage line Age ame, ;, >iwnte
eal> 1.5 m >OU ,,.,,,;an >ina Iron
nw
a,aaan >ite, w hich
The a<thaeologicol exc since the preroman
due to soil erosion
coll uvial deposits is
Middle Ages (here:
Results espe cially since the
Iron Age period and consistent with the
%)). These results are
in pro gress 2.8 m of 4.3 m ( 6 5 thern Kraichgau,
erbach is still sou
the test site Bau ( 1 989: 171) in the
As the w ork at 8a sh ows observation of Eitel of colluvial
marized. Fig. met res
results can be sum ce of several
only preliminary an loess who found eviden the Middle
young Wii rmi since
situation with a few hundred years
the pre-holocene ef. Yet the sedimentation within sol- Bt with com
paleo- reli luvi
ing the underlying Middle Ages the
covering and equaliz ntry' is only Ages. Also in the up by soil erosion
'loess hill cou strength was used
ical term periglacial p arably high shear
geom orpholog w point, one mig ht as sion ev ents bec ame
archaeologist's vie high magnitude ero
half truth. From a geo colluvial processes, so that rain storm eve nts.
al and ding
ocene 'soil erosion the case of outstan
.! well speak of a Hol proo vably even more likely in dsc ape and reli ef
be en ant lan
hough the area has dence of signific
Fig.3. Th e loc a landscape' (Fig. 8b). Alt port ion of the D espite the evi
tion of the test . quite a considerable
settled since the Neolithic,
site Bauerba ch zn .
. the loess covered Kraichg au Hills .
zn SW-Germany.
At the top of the
more clayey, subsoil-d . . . .
enved colluvium
samples HDS-49
7 and HD
S-496, taken from
epth >
0.08 ka ( 1140 AD 80
i';;' ) md>0otmg the High Medicro l
50m
270-277 cm
and 25 2 p0<iod. Aoco<ding to kn
Bor Bork (1987) and Bork (1989)
o
'q"'J.''' I0.86 -
57 cm b low
i '
.12 ka (1140 AD
grnund sur ace,
_of
give this is a period well own for high magnitude rainfall and
-%\%%\ -HMM-
---
I
I
0
0 0 \
Stratigraphically inconsiStent ages appear in core metre 1
"- J
0 00
----
,
-
.---
I I
and 2. For the uppermost metre this is due to the fact that with
00 I
_
-
\
A1 extranous and n 0t properly bleac hed matenal .
was sampled.
.
0 -
--
-
HDS-492. 0 .- -
-
-
-
-
--
;
Age overestimations ho ever, are also noticed for samp les I- -
-
-
Slit ditch
9
-
.
cm below ground 1 ve1. t s not un kely that high magnitude Upper Iron Age
Bv h
soil erosional even s as t ey occured in the High Mediev al Upper Neolithic period (?)
Pit house, Latene
Period(cf above) caused the depos 1t1on of not prop erl y
C c/C v Pit of Michelsberg
Culture
. .
n
ne period house
bleached colluvial matenal m core metre 2 .
Eaves ditch of Late
As none of the up er flou sam les fails to produce an
:
Lower Iron Age
Grave of Latene peri
od
IRSL-shine plateau on y strat1graph cal inconsistency wit hin att period
1
2m
c the profile indicates that the top sed ments were obviously n o t
o Cellar or pit of Hallst
127
Kadereit, A., Lang, A.
& Wagnei; G. A.
Colluvial sediments near a rchaeological sites as a key to the past landscape evolution under human impact
Core Herrenbuckel llb
PRE-HOLOCENE
. 1 996b, Jungsteinzeit - mittlerer Abschnitt (Mittel- Rieser, U., 1991, Low-level Gamma-Spektrometrie zum
B
periglacial loess hill country
JJeideollthlkum), in Behrends, H. (ed.), Faustkeil - Urne - Zwecke der Dosisleistungsbestimmung bei der Lumines
Depth ne
Sample IRSL-Age wert. A
rchaologie in der Region Karlsruhe, Badenia zenz-Datierung, Diplomath.esis, Faculty of Physics and
Sch
[ml Number [B.C./A.D.J [years] K arlsruhe, 55-62. .
Astronomy, University of Heidelberg, 67 pp.
solifluctio
n laYer Verlag, Singh.vi, A. K . & Lang, A., in print, Improvements in optical
M. R., Rieser, U. & Stolz, W., 1996, Optical
1
I J(rbetschek,
I ..-. a
dating: Some lunescene properties of natural feldspars, dating of partially bleached sediments: The differential
-- .J' HDS-491
- . '
silfy _
lementi di Geografia Fisica e Dinamica Geology and Archaeology, Springer Verlag, Berlin,
urope, Supp
calcarous I J
HDS-494 Sup plemento III, 241. Heidelberg, New York, 467 p.
580 A.O.
,
-
aria
I I 170 HOLOCENE
Quatern
I 'j HDS-495 820 A.O. LaJlg, A., Lindauer, S., Kuhn, R. & Wagner, G. A., 1996, Wiggenhom, H., 1995, lnfrarot stimulierte Lumineszenz von
190
colluvium soil erosional Por cedures used for optically and infrared stimulated lumi Feldspat - Grundlagen und Anwendung als Datierungs
silfy/clayey of sediments in Heidelberg, Ancient TL 1 4, methode, PhD-Thesis, Faculty of Physics and Astronomy,
calcarous '
and colluvial landscape ne scence dating
University of Heidelberg, 127 pp.
HDS-496 1.140 A.O. 7-11.
80
HDS-497 Lamothe, M., Balescu, S. & Auclair, M., 1994, Natural IRSL Zoller, L., 1994, Wiirm- und Rij3lO.fJ-Stratigraphie und
1.140 A.O. 120
b Middle Ages intensities and apparent luminescence ages of single Thennolumineszenz-Datierungen in Siiddeutschland und
- colluvium
colluvium
feldspar grains extracted from partially bleached sediments, angrenzenden Gebieten, Habilitation thesis, University of
... Iron Age I Rom
clayey an time Radiation Measurement 23, 555-561. Heidelberg, 174 p.
HDS-498 - soi/relict
50 A.O. 300 Narnbi, K . S. V. & Aitken, M. J., 1986, Annual dose
decalcified HDS-499
330 B.C. 310 conversion factors for TL and ESR dating, Archaeometry
2 8, 202-205.
Fig. 8. re-holocene periglacial loess hill countr
Prescott, J. R. & Hutton, J. T., 1994, Cosmic ray contributions
soil relict developz towards a holocene soil erosional andy
to dose rates for luminescence and ESR dating: Large
deposzt1onal landscape:
preliminmy scheme.
depths and long long-term time vaiiations, Radiation
Jungstemzeit - jiingere
'
. . r Abschnitt
(Jungneohthikum), in Beh
rends, H. (ed.), Faustkeil-
Urne
- Schwert, Archaologi
e in der Region Kar
lsruhe
Karlsruhe, 62-72.
Fig. 7. Co//1111ial c/110110/ogy of the drilling
'
quired to recog ze nu tmg 1actors for the applica cksed1mente. Ein Beitrag zur Landschaftsgeschichte
_ tion of OD
Sudwestdeutschlands,
and to accomplish data mterpretation. Stuttgarter Geographische Studie ll
111, 205 pp.
Fuller, I. C., Wintle, A.
References . G. & Duller, G. A. T., 1994, Tes of
Ptrnl bleach methodology
as applied to the in fra-red
stim ulated luminescence
Aitken' M. J., 1985, The11110/1 1111i11esce11ce Dating . of an alluvial sediment fro!11
.
Archaeological Sc1e11ce, Acadentic Press, 359.
, Stud' 1es ll1 anu be, Qua ternmy Scie nce Reviews 13, 539-543.
Aitken, M. J., 1998, An i11tmd11c
Grun, 1993, Simplex, Version 1993, Computer progr am
, 0x1or for
tio11 to optical dating
. . c d
lummescence data ana
University Press, m pnnt. lysis.
Heide, B., 1996a,
Aitken, M. J. & Xie, J., 1992, Optical dating using infra-re Jungsteinzeit - alterer Absch nitt
d (Altn eolit hiku m),
diodes: young samp les, Quatemmy Science in Beh rends, H. (ed.), Faustk eil- Urne
Reviews 1 1
' Schwert, Archiiologie
1 47-152. in der Region Karlsruhe, ad ni a B e
Verlag, Karlsruhe, 34-
54.
128
129
PRELIMINARY GEOARCHAEOLOGICAL REPORT
TEL KEDWA, SINAI, EGYPT
1 rchaeomet1y Laboratmy, Department of Physics, University of Toronto, Toronto, Ontario, M5S JA7 Canada
A
2Near and Middle Eastern Civilizations, University of Toronto, Toronto, Ontario, Canada
gical survey was conducted at Tel Kedwa in the North Sinai in conjunction with archaeological excavations
A geo a rchaeolo
uipose of ascertaining the site's length of occupancy through the determination of the depth of cultural material. T his
'th the p
WI accomplished by coring the site. Also, a topographic map was generated and the relative elevation with respect to sea level
was
was deter min
ed.
J{EVWORDS: GEOARCHAEOLOGICAL SURVEY, WATER TABLE DEPTH, TEL KEDWA, TEL EL HER (MIGDOL),
TEL ABU SEIFI, TEL FARAMA (ANCIENT PELUSIUM), TEL EL-MASHKUTA, QANTARA E AST, SINAI,
SUEZ CANAL, NECHO II, AUGER HOLES, TOPOGRAPHIC SURVEY, SAITE.
Proceedings
of the 31" lntematio11al Symposium 011 Archaeo111et1)>, Jerem, E. & T. Biro, K. (eds.), Archaeopress-Archaeolingua, BAR- Ce1111: Ew: Sa 1.
.
sl/1at, Egypt
gical report: Tel Kedwa,
P reliminary geoarchaeolo
Pavlish, L. A. & Redford, D. B.
NORTH
SINAI
HIL1CN'OUS
.'!l\!
\
\,
:::
-
'
.
-
Eastern
....
Western
Desert Desert
.',
- ---------------- ... __
; :: :
lt to that level and
was burnt and partially destroyed for the first time the site to Tel Kedwa during the 1997 field season. This 1 ) that they were bui
ls, bu a
of heavily decayed wal
:;
(600 BC I 568 BC). Based on parallels with Saile fort Tel el survey had two major objectives: 1.) asceitain the dept!\ of
r present the remains
ed stru ctu re; or, . h
an unfinish .
first few courses of own pend
Mashkuta located to the southwest near present day Ishmalia, cultural remains on the site; and, 2.) measure relative
systematically at som
eu
bricks were removed
the site ran d
which shows three distinct destruction levels at 600 BC, elevations on the site and in the surrounding environs to l
cultural materials on
568 BC and 525 BC, a likely candidate for the destruction calibrate the depth measurements. after construction. The
3 metres. ig.
ent to aproximately
would be 568 BC. The fort was rebuilt and appears to have To accomplish these objectives a grid was setup using the depth from being abs
nt section (OE/ON).
survived for almost a half a century before being again inner comer of the northwest fortification wall. This location the the site. shows an typical sedime
re sec tion through
destroyed sometime late in the third quarter of the 6th Century (60 metres East, 103.45 metres North) was the benchmark for Rig. 2. T hree-hundred met
BC leaving a thick stratum of ash and reddened brick. Again a the survey. The elevation attached to this point was 2 metres
rebuilding was planned almost immediately, incorporating above sea level which is very close to the true elevation for
walls 13 metres thick with foundations 1.20 metres deep on a this point on the site. This observation was confirmed with off
sand pad in excess of 5 metres deep with a surrounding "moat" site measurements showing that the transcendental beach to
canal approximately 10 metres wide and several metres deep. the south of the site is at an elevation of approxi mately I
The moat served both defensive and engineering purposes as it metre above sea level; and, that the transient lake bottom
in
helped to stabilize the mudbrick foundation walls which have sediments are 10 to 20 cm below that of the beach
Ke dw a
been built to ground level, but which appear never to have been elevation. This lake bottom effectively separates Tel
sou th ea st
completed. No stratum of deposition is associated with the from its rather better known neighbour to the
walls, the present ground surface running directly up to the (2.5 kilometers) Tel el Her, or Migdol. .
t
face, and the foundation trench descending from the present The major axes of the site grid are two lines running at ngh
O E/O N to
surface. There seems no doubt that this last fortress had been angles to one another. The East-West line went from
th
300E/ON crossing the inner wall line at 60E/ ON on
planned, but never finished and put to use. Events must have
sid e 0
caught up with construction, rendering the plans useless. But western side of the site and at 249E/ON on the eastern
N /6 0E to
what event, clearly dated to the third quarter of the 6th century the site. The North-South line went from 120
could have overwhelmed a state administration able to 103.45N/60E.
Sixty-three 10 . cm diameter auger holes were dug with th
conceive of and implement such a massive foitification? There
objective of ascertaining water table depth, wall depths,
s
is only one: the catastrophic defeat in the northern Sinai in
de pth s .10
525 BC of Psammetichus III, last king of the 26th Dynasty, at foundation platform depths, moat locations and
re m
addition to the spatially varying depth of cultural
the hands of the Persian army of Cambyses. The collapse of
e
the Egyptian state opened the door to foreign occupation, and (Fig. 1 ). Thirty-nine of the auger holes were dug alo ng th
mete East-West gri line .at ON providing a reasonable c o
: ; l material at Tel Kedwa.
for 120 years Egypt became a province of the Persian Empire. the depth of the cultura
a Fig. 3. Map showing
The eastern Delta and north-west Sinai ceased to be a frontier sect10n through the site (Fig. 2). The average borehole w
133
132
Pavlish, L. A. & Redford, D. B.
Two-hundred and twenty-five elev
ation measurements
ere taken usm _ g
a standard military tran
sit (400 degrees - -
circle). In adelit10n to
the measurements take .
Unconsolid
at ed INVES TIGATION OF ARCHAEOLOGICAL SI TES
n for the d Sediments
Shercl Layer 2
Sand Lense
to present some case studies of our non-destructive researches in Hungary in the framework of M3 highway
in (5Y7/1) This article seeks
grey-green cla an d Frenc h-Hungarian cooperation. To develop the aerial archaeological, geophysical, etc. technologies for pmspecting is
y roject
(10YR7/l)+San
d ow ava ilab le in
aborat
Hungary.
ion of
However;
many
the integration of measurements and archaeological data needs a highly-developed GIS-system
specialists. T he next step is to analyse the superposition of measurements and the archaeologically
and the coll
data (e.g. pits, post holes, etc.) if we want to understand the spatial layout of sites and features. As a result, we
obtained physical
Sherd Layer 3
can of e
f r a sequence of data collection which could be quicker and more efficient than non-integrated, single-stage methods.
E.g. Gyula Novak.i's research at Velem-Szent Vid-hegy and in the Sopron area: Szekely et al. (1992), Szekely & Puszta (1996), Kolto & Bartosiewicz (1998)
271-2 74, with further bibliography.
With the exception of a few experiments: Cucarzi (1992), Jerem et al. (1992).
Braasch (1994).
Goguey & Szabo (1995).
,
F r the a
6 o rchaeological control of the results of aerial photography of Neolithic rondels, cp. Zalai-Galli ( 1990).
Becker (1996) is a notable exception. For methodology and background research, cp. ibid., 265-283, Stanjek & Fassbinder (1996), Fassbinder & Stanjek
F,. 7 Cl 996) and, more recently, Powlesland et al. (1997).
.
ig. 5. Map showin Czajlik
g elevation above sea leve & Holl (1996), Czajlik et al. (1997).
l at Tel Kedwa.
Raczky et al. (1998).
134
Proce edi11gs of the 31" !1Zternario11al Symposium on Archaeometry, Jere111, E. & T. Biro. K. (eds.). BAR-Archaeoli11gua, Cemral Europea11 Series I.
sites
vation at archaeological
.. , eys and archaeological exca
. ' p., Piiszt a, S., Cza1lrk Z Holl B &
Racz/.y
.,,
mart on, A. of aeri al phot ography, magnetometer surv
on
'
applicati
.,
,
rate
/nteg
of a few seconds. The data were .
hen linearly filtered: and the results of data collectio
.
geological and archaeologica n. Wi sha
I anomalies were d'istmguished ll the n
results of our archaeological d
techniques than from trad- a depth of 20-50 cm (Fig. 4). These . struction
:Uenof,1 thepe.French
ditch es at othe rs, potte
For a descnpt10n of the s Neolithic rondel of two containing, among
io T
Raczky et al. (1 998).
as well as of the sunk en houses and refuse pits
his excavation was a re enabled the identification of this rondel,
11 an
-Hung an coope rat10n and was led by ever ,
Goguey & Szabo (1995) 67. V:aJk Cserrneny1 (Szent Istvan
, Kiraly
, Museum, Szekesfehervar). (1998). (1992), Raczky et al. (199
4).
(1997). Cp. also Raczky
" ky et al. ( l 997b), Fodor
Cp. Raczky et al. (1997a), and esp. Racz
136
137
cal sites
excavation at archaeologi
eys and archaeological
aphy, magnetometer surv
A.
n of aeri al pho togr
Raczky, P., Puszta, S., Czajlik, Z., Holl, B. & Marton, ated applicatio
/11fegr
1200
1100
1000
.
intensive anomaly of the innennost ditch
- Which,
strength of the archaeological evidence was the . on ! :mo
e li e
. ar st '
caused by the burnt daub fragments m Its fi1ll
Its fonn
rectangular with rounded comers, rather than .
. C trc u l
mam s1.des are at iight angles (90o) to each oth
er. This
....
i
through the third, fourth and fifth ditch . ere
cuttmg
. ..
..;
_,
m ' N l>
" .J "'
m 40 Iii m m "" I\) ii> c
i m c
n of (JI a; tit th
w
m
0 I\) A Ill "' I\) A
0
Ill
0
"'
0 0 0 0 0 0 0 0
l> Ill 0 0 0
0
the entrances could be traced within the ronde1 The cent
1\1 0 0 0
0 0
0
. ra)
Fig. 7. Magnetometer surv
.
zone of the tell was also magnetically i'ntens1ve complex.
and tlle 1ater ey of the Csffszhalom site
.
excavat10ns revealed that these anomalies were cause d m ai. n1y
st oriented
also revealed an east-we
bY house and hearth remains. the possible The magnetometer map
knowledge, of and turning
to our present the one-time river bed
The smaller, non-continuous anomalies v1's1b1Y ceased in According feature starting from
, buildings, wells and mad e out on
tures - such as graves r section could also be
the nort hm part of the horizontal settlement, correspondin archaeological fea towards the tell; the latte
se positive anomalies
). This was a double
the aeri al photo (Fig s. 6-7
the
e d.
h from
as the aenal photographs showed - to the one-time riverbg pits - only the
latter or the latter two cau ditc
ber of pottery sherds in neto met er map ,
ng to the high num according to the mag
These could be correlated with the boundary of the sett1ement of this type owi fitting of the Sarmatian period whi ch, ed
firmed by the GIS turn
their fill. This is con
then
Fig. 5. Aerial prospecting oif the Csffszllalom s1't e comp1ex' -time watercourse and
as estabIished by the field survey and thus these anomali -continuous ran parallel to the one
non aerial photo
taken by 0. Braasch (1993). can also be associated with settlement features (Fig. 8 information layers. Sin
ce in the wes t thes
at the Cs6szhalom-du1
e
6 site can be southwest - and thus the enigmatic feature on the
anomalies - which could now be interpreted.
pits bes ide the houses - the tell, the
primarily linked to the Lat
e Neolithic uded the central area of
as defined The excavated areas incl
the bounda ry of the site pho tos and the
could be noted well beyond revealed by the aerial
site is larg er than rondel of five ditches the M3
indicate that the line of
by field surveys, it may as well as the planned
magnetometer surveys,
previously thought (Fig. 9).
-
-::----------- -- ,, ha1om site, taken by O. Braasch 0993).
-- of .the csosz
(1989-1997).
Fig. 6. Aerial prospecting
139
138
Racz).y, P., Puszra, S.,
Cza':/'lik Z"
Holl, B. & Marton,
A.
'
cavation.
\\'I't h0ut ex
tric and methodological
aluation of the archaeome
The ev data integrated into the GIS system
. pJicau ons of the . .
ove can be hoped from a detailed analysis of the
jJJ1 'b d ab
e;; d al feature
s of ruchaeologial sites and from the
.
detailed geomorpholog1c and pedolog1cal data.
: nection of
Summary
1
2 E.g. Becker (1996) 77-81, 155-158, 293-295.
140
141
n at archaeological sites
archaeological excavatio
magnetometer surveys and
Racz/..y, P., Puszta, S., Czajlik Z Holl B & 1narton,
,,, A, . licat ion of aerial photography,
ed app
teg1at
/n
' ., ,
Bokony1, S. (ed.), Cultural and landscape changes m rche l gie en France et en Hongrie - Ugi fenykipezes
South-East Hungary I. Reports on the Gyomaendrffd
es regeszet Frrn .... czaorszag
, ban , es Magymvrszagon,
,
:
.
v1ect, Archaeolingua I, Budapest, 13-40.
.
Institut Franrai
v
s - consetl Regional de Bourgogne - .
CzaJ , Z., Holl, B., 1996, Archaeological Chronology and Eotvos Lorand Tudomanyegyetem ' Budapest 127 p.
.. .. ,
E xcavat10n Data Bases, in The F uture of Our Past Jerem, E., Kiss, Zs., Pattantyus-Abraham' M. & "arga ' A. '
'
v.
.
'93 '95, Hungan National Museum, 144-150. 1992 ' T he combmed use of archaeometric m ethods
. . -
CzaJlik, Z., Marton, A. & Holl ' B., 1997, A z M3-as autopalya,
, .
preceding the excavation of archaeological sites' in
regeszet1
, , leletmenteseinek " S. (ed.), Cultural and landscape changes in
"konyi,
terinformatikai eldo lgozasa
, - B0
The GIS processmg of the rescue excavations associated Souh-East Hungary I Reports on the Gyomaendr6d
. .
w1th the M3 motorway in Hajdu-Bihar County' in
Pro1ect, Archaeolingua 1. Budapest 61-98 '
al
'
R aczky, P., Kovacs,, T. & Anders, A (eds ), Vtak a
. . , 0 o, L. & Bartosiewicz, L. (eds.), 1998 Archaeometric
K"lt"
multba - Paths into the Past A z M3-as autopalya
, . , Research in Hungary//., Budapest-K posvar-Veszp rem
regeszet1
, leletmentesei - Rescue Excavat'ions on the M3 322 p.
.. ..
M otorw y, Eotvos Lorand Tudomanyegyetem - Magyar
Powlesland, D., Lyall, J., & Donoghue, D., 1997, Enhancing
nd
emzet1 Muzeum, Budapest, 153-155. the recrd through remote sensing: The application a
. . st ng
Fassbmder, J. & Stanjek ' H., 1996, Magnet1sche Boden- .
mtegrat10n of mult'I-sensor, non-mvas1ve remote sen
. .
baktenen und deren Auswrrkung auf die Prospektion . d
techmques for the enhancement of the Sites an
.
?
archii log1 cher Denkmaler, in Becker, H., (Hrsg.),
.
Monuments Record, Heslerton Parish Project,
N .
. tt p://
Archaolog1sche Prospektion. Luftbildarch" aolog1e und (h
. _Yorshire, Egland, Internet Archaeology 2,
Geophys1'k A ibettshefte des Bayerische11 Landesamtes
' .
mtarch.ac.uk/joumal/issue2/pld_index.html).
143
142
O HYSICAL CA MPAIGN AT THE NEOLITHIC SET TLEMENT OF
A GE P
KANDOU - KOUPHOVOUNO S (CYPRUS)
1[,aboratoiy of Geophysical, Satellite Remote Sensing & Archaeoenvironment, Institute of Mediterranean Studies,
Research & Technology, Hellas, P.O. Box. 119, Rethymnon 74100, Crete, Greece, asaris@ret.forthnet.gr
p; mdation of
lenn G. Black Laboratmy of Archaeology, University of Indiana, Bloomington, IN 47405, sball@ucs.indiana.edu
3Laboratory of Applied Geophysics, Dept. of Mineral Resources Engineering,
Technical University of Crete, Chania, Crete, Greece
4Foundation of Hellenic World, Akademias 15, Athens, Greece
5Department of Archaeology, University of Athens, Zografou, Athens, Greece
The g eophysical mapping of the Neolithic Settlement of Kandou-Kouphovounos, in Cyprus, was conducted as part of the
continuing archaeological research program in the specific area. The goals of the project included the mapping of the shallow
depth architectural remains of the site in order to assess the limits of the settlement and plan the future excavations of the site.
t
Mag netic surveying was successfully applied, together with soil resistivi y prospecting. Both techniques resulted to a number of
fea tu res that have been correlated to the existence of architectural features. The electromagnetic survey (EM38) encountered
various problems related to the balancing of the instrument due to the extreme high temperatures, the high resistivity of the
shallow bedrock and the lack of contact with the suiface due to the dense vegetation islets and bedrock outcrops that were
scattered in the area. The magnetic susceptibility maps, covering an area with existing architectural relics, resulted to a detail
mapping of their inner details, in close correlation to the other techniques. Subsequent excavations in the following year
revealed a number of architectural features which were located by the use of geophysical prospection techniques. The
geophysical maps indicate that the site extends in the north, south and west directions with no evidence of a surrounding
defensive structure. F urther geophysical work is planned to map the relics on the slope of the hill. The final product will consist
of the superposition of the excavated regions on the geophysical maps to provide a tool for the management and preservation
of the site and its environmental settings.
Proceedings of the 31" Imematio11al Symposium 011 Archaeometry, Jerem, E. & T. Biro, K. (eds.), Archaeopress-Archaeolingua, BAR- Ce1ltl: Ew: Se1: 1.
Sarris' A ., Ball' s., Georg1la,
K., Kokkinou, E., Karimali, E. & Mantzoura A Geophysical Campaign at the Neolithic Settlement of Kandou - Kouphovowws (Cyprus)
ni, E.
1so
240 ' 0
.
220 Ill 220
200
Giophy1'1Gnd1
200' SoilR;J11nc1>511rn:_r(rnklP1ooc)
180
'
180
160
140
120
100
60
.... X.!J.Vll )XJJVlll xx1x...
40
,xxxv
20
146 147
Sarris, A., Ball, S., Georgi/a, K., Kokkinou E Karimali' E & Man
'
t.zourmu,. E.
probably related to variations of the bedrock's anaglyph, registration), intensive soil corrosi
_ on Phen o THE ISLAND OF AMORGOS:
rn ena
whi ch could be considered as constituting a natural defensive s1gnaI d1. stmc t10n) and unfavorable pr . O ac "1.
eserva11o n
NSING
MICRO-SCALE & MACRO-SCALE REMOTE SE
border for the settlement to the south. Evidence of intensive condit
Kantou-Koupho v o u n o s, the recog
nit" ions
. 10n of th
soil corrosion to the south enforces the supposition that the e geop .
signatures (depen d ing on the metho
d used) rne hy8j,.. 1
UROU, A.,3 KAR ATHANASIS, CH.,3
. t a nu
ARRIS, A.,1 MARAGOU, L.,2 GKIO
settlement was mainly expanded in the northern pru1 of the d1fficu 1Iles d u e to the rough geo mo rnber"<IIf
rph o 1 ogy, th o
hill. thick vegetat1.0n, the outcrops of the
e fre S GIANNOULI, E.,7
LI, M.,4 MAVROIDIS, TH.,5 KAR IMALI, E.,6
dr qu ency 0f
sTEfOU
. . be ock
In ntrst, the north section of the hill appears to be more su ffic1ent thickness of soil, and the
the 1a
ck of
GAVALAS, G.,2 AS VESTAS, G.,3 GEORGILA, K.,3
'
size a n d
n ature
pronusmg m relation to the existence of architectural remains. potential taigets ( some of which were of the
even engr av
1.
A specific region was surveyed by all methods, in order to be bedrock) (Sanis 1998). The parallel prosp
e d I.nto
the IOANNOU, D.3 & VRONTAKI, E.3
ec ion of
used as a guide in the interpretation process. The internal . th e s.u
diffi e by
. .erent .techniqu es and intensive filtenng of th
e data nt, Institute of Mediterranean Studies, Foundation
details of architectural remains are evi d ent in all data sets. cnt1cal m th e enhancement of th
e 1,,eatu r e 8
were
1 atory of Geophysical, Satellite Remote Sensing & Archaeoenvironme
.
S?il resistivity was especially successful in outlining mterpretation of th e m.
an d t
he [JI b or gy, Hellas, P.O. Box. 119, Rethymn on 74100, Crete, Greece, asaris@ret.forthnet.gr,
of Research & Technolo
differences that were caused by the existence of wall 2Dept. of Archaeology, Universit y of Ioannina , Ioannina 45110, Greece,
structures and/or corridors. Similar results were produced by ring, Technica l Universi ty of Crete, Chania, Crete, Greece,
References 1 ophysical Laboratory, Dept. of Mineral Resources Enginee
the electromagnetic survey (Fig. 5), c ontraiy to the cor Ge Mesogei on 70, Athens 115 27, Greece,
4Institute of Geological & Mineral Explorat ion,
responding ones in the sou them portion of the site. s Engineering, Metsoveio Technica l Universit y of Athens, Zografou, Athens, Greece,
Mantzourani, E., 1994, Ekthesi Apostelesmaton t' An 5Dept. of Mineral Resource
18 askafi8
st1 thes1 K a n d o u-Ko uph o vounos , Epi.stim 6Foundation of Hellenic World, Akademi as 15, Athens, Greece,
oruki EpetiJi
.
Sams, A., 1998, Geophysical Issues in Arc
haeological
The emp oyment of geophysical prosp ection techniques in .
Research: Paradigms, Uncertainties & I c . s
nterencie
the srveymg of Neolithic sites needs special attention, Int ing and macro-scale satellite remote
. m on Remote Sensin
Sympos1u has become the target of micro-scale geophysical prospect
g in Archaeology, B '
oston, The island of Amorgos
applying a wide spectrum of remote
especially when dealing with a dry enviro nment (weak signal USA, 16-19 Aplil 1998. of the project was the formulation of a general strategy for
sensing investigations. The goal framewo rk of archaeological research.
es and integrating the corresponding results to the general
sensing techniqu site and a hilly site dated
le only by foot, a coastal Roman
conducted in two remote areas, accessib
The geophysical survey was r sites for mapping
e surveys were carried out at the particula
from the Early Cycladic (EC) period. Magnetic and soil resistanc
the same grids. The correlati on of the geophys ical maps and
suiface sur vey was applied on
the subs urface relics. A systematic g was also
results about the boundari es of the EC site. Geophys ical surveyin
the swf ace sherd concentration has drawn specific
are either covered complete ly by alluvium deposits, or are
site, most of which
able to detect architectural remains of the Roman
photogra phs were used for creating the aerial mosaic of the
Aerial
submerged within the sea due to the past seismic activity.
and both layers were superimp osed on the DEM of the
aerial images
wider region. The geophysical grids were registered to the
satellite mosaic of the whole island. Processi ng of the diff erent
region. Two Landsat TM images were combined to produce the
image was also superimp osed on the DEM of the whole
bands was used for creating a number of thematic maps. The satellite
added for producing a GIS. Supervised classification
island and other layers such as the digitized geological maps were
the island. The island of Amorgos has been used as a pilot
techniques have been used for modelling the settlement patterns of
study for the application of large and small scale investigations of the archaeological sites of Greece, proposing a model of
ar chaeological site assessment and the creation of an electroni c archaeolo gical and monument record through the use of
In the current presentation we will deal with the application Aerial and satellite imagery was employed to capture the
of micro-scale and macro-scale investigations in the island of geomorphologic characteristics of the island and provide the
Amorgos, in the Aegean. A nationally funded campaign was geographic context for the accurate registration of the known
Carlied out in the island of Amorgos, in the Aegean, as a pilot archaeological sites. The digitization of topographic maps
project for the application of different remote sensing provided useful information on the altitude, slope, and aspect
tech niques and the integration of them in a Geographic of the archaeological sites and became the base for the
Information System that could be used as a tool for the production of 2-D & 3-D thematic maps of archaeological
man agement of the archaeological sites (Sarris 1994). interest.
The micro-scale module of the project included the systematic The digitized archaeological and geological data were
surface survey correlated to the satellite images in an effort to recognize the
and geophysical prospection in two remote
eas located in the north of the island, SW of the capital of spectral signatures of the archaeological sites. Supervised
atapola. The targets of the geophysical campaign were the classification techniques have been applied to identify areas
coastal site of specific archaeological interest. A number of thematic
of Kato Akrotiri, consisting of Roman/Byzantine
: chitectural remains, and the hilly settlement of Madres of maps were produced, regarding the archaeological sites of
1'oussos dated to the Early Cycladic period (Maragou 1990). different era and "risk" areas of similar interest.
he magnetic and soil resistivity surveys were conducted in
er to locate and map remnants of architectural remains. Methodology and results
e geophysical data were correlated to the results of a
system investigate the The site of Kato Akrotili was surveyed by magnetic and
atic surface survey, in order to
habi_ tatio n soil resistance (Twin probe) techniques with a lm sampling
trends within the sites.
Gk;,,ro,,, A . Kw"''
"'";,, Ch.. S"fa
Cav a/ as, G., . "U, M., M"""'d", a 1
Asvestas,G., G
eorgi /a, K., Ioannou,
Th, Kanm 4 E. . c"'""""
inervl (Fig. 1) The pr D. & Vront aki, E. /', ., The Island of Amorgos: Micro-scale & Macro-scale Remote Sensing
x .
gnds mtluenced the resis1Jty of the c o a st to the geophysical c
v11y ommon coordinate syste
sea. S?aded retie maps an readings in the direction of the pro m was helpful
for the s
d di rec tional filtering emph cessing of the data. imult .
the eXJstence of linear featu asi zed
aneo
lls Magnetic grad'zent sun1ey at Mandres tou Roussou -Amorgos 1996
res, Which co Geophysical survey
ing was able to
the existence of coastal archi uld be correlate
d to de te c t archit tu
tec remains of the R oma
of the residuals nd downw tural remains. The calculation com n site, most of w hi ecrJa
. c h are ei
ar . ther c
vertical magnetic gr d con tinuation re pletel y by allu vm m epos ts ' or are ovel'ed
si a 30 m step inte d by
enhancement te.c hni ues, ca ve processing, including edge
10 nT
rval digitization of 0
the 1 :5000 z
lculation topographic maps (Fig. scale 1 5 nT
4). Although the superp
-
Soil Resistance
sur vey at Kato Akrotiri
1 L:=====-=-.: -Amorgos 1996
I
__
:- :- .
:-- 'J Roussou.
Fig. 2. Magnetic Gradient Survey at Mand res 0.r
__
__
- --
--
I
70
I
I
:I
so
.. 079 149
N
0 .. 078 549
R 40
T
H
. _, 077 9'9
(m)
30
' . .. . . .... .. 077 30
.
ii
. .
. ,....
Bi 20
' 076 7'9
I
10
' 076 149' ... 076 149
I
\
I ' 075 549 ... 075 549
\
. .. ... \
10 20 30 40
..
-.;
" 50 60 70 80 .. 07' 949
.. EAST (m)
.
.Mat:a.ra
Fig. 1. Soil
Resistance Survey
at Kato Akrot iri. .
F.1g . 3 . Mosaic of aena
. l image!)' of the wider area of interest.
15 0
151
sing
. & Macro-scale Remote Sen
.FAmorgos. Mi"cro-scale
The Is/and OJ
Sarris, A., Maragou, L., Gkiourou, A., Karathanasis, Ch., Stefouli, M., Mavroidis, Th., Karimali, E., Giannouli, E.,
Cava/as, G., Asvestas,G., Georgi/a, K., loannou, D. & Vrontaki, E.
Fig. 4. 3-D representation of the aerial imagery of Kato Akrotiri (view from the sea).
1, 2, 3 ).
l La ndsat imagery (Bands
ime nsio nal rep rese ntation of the real co our
Fig. 6. 3-D
4 O
l:J 14
7
1 '
-<
4 073 147
059147 Motv-t-.-----;-+--;.---;-;-;-----;-+---;-...;---;-+-..---;--;--
li.53, 1151 IM>tu:a lilil 4117 lili1 457 H4 451 li57 451 510 4117 573 457 1115 4117 li7!1 451 iU 07
satellite imagery, as well as the archaeological data, to the preference towards the north and south directions. Finally, it
Transverse Mercator System was based on the rectification of can be observed that in the Classical-Roman era, most of the
the images using the affine transformation. True color and already discovered settlements are built in the northern side
of
nd has a
pseudo-color maps were produced by the combination of the island. Given that the northern side of the isla
ve
different spectral bands. Both aerial and Landsat imagery much smoother slope than the southern side of it, the abo
were superimposed on the DEM providing a 3D represen observation suggests that the access to the sea was of prim ar}'
of
tation of the island (Fig. 6). importance in the historic times. Furthermore, a network
Statistical analysis proved that in all times settlement was towers has been created in the last historical times, possibly to
confined in regions of lower elevations. About 70% of all sites serve as a system of communication, control and eas y
are found in elevations of less than 240 m above the sea level, warning of the inhabitants.
as
indicating a continuous trend to coastal and plain areas. In In order to test the above hypothesis, viewshed analysis w
fact, most of the recent villages of the island are built in the performed in both the sites of historic times and the ww
ers.
_
The viewshed process calculates the observation bo un d a ne s
proximity to the coast or valleys. Furthermore, most of the
d ata
lower elevation hinterland has been exploited for cultivation from a specific point of reference, based on the DEM f
purposes based on a ten-ace land use system. A smooth slope Viewshed parameters considered a 6.3-10.8 km distanc 01
of less than 30% is also observed in a similar proportion of sight, a 360 horizontal sweep angle and a +/- 80 ve
rUC
analysis gical sites of the island
of Amorgos.
sites. Similarly, the orientation of the sites shows an obvious sweep angle (Fig. 7). The results of the viewshed s for selected archaeolo
Fig. 7. Viewshed analysi
153
152
. sing
.
: M icro-scale & Macro-scale Remote Sen
The Island of Amorgos
Sarris, A., Maragou, L., Gkiourou, A., Karathanasis, Ch., Stefouli, M., Mavroidis, Th., Karimali, E., Gian no
uli, E.,
Gavalas, G., Asvestas,G., Georgi/a, K., loannou, D. & Vrontaki, E.
were patticularly successful for most of the island with the There was a clear distinction between the spectral s
i gnatures
exception of the NE side of it, which is also the least explored. sites - no sites and between the Mycenean settlem en Of
ts and
For example, the central part of the island, including the gulf rest of the sites. In the final model, the confusio e
n
of Katapola, was mainly controlled by the sites of Minoa, exhibited an accuracy of 69-100% (Fig. 9). The app
lica
:
atrbt
Arkesini and the port of Kato Akrotiri. This observation, of
together with the results of the geophysical survey,
the classification model to the satellite mosaic of
island produced similar results. A fmther refinement
the
;0
Ole
strengthens the role of the port of Kato Akrotiri. In another th
model and veiification of the preliminary results is exp
ected e
example, one could have found difficult to explain the reason be carried out by the addition of the unpubli shed sites
of existence of three neighboring towers in the SW part of the archaeological database.
to
island. Viewshed analysis suggested that the particular
network not only had control of the nearby valley, but it could Con cl us ions
also establish communication with the northern part of the
island as well. Indications of architectural remnants have been pro
vided
The superposition of the archaeological database on the by geophysical techniques. Natural and anthrop o
geni c
digitized geological map proved that there is no preference of changes of the landform impose specific limitations in
the
the settlements with respect to the geological regions of the survey of archaeological sites and the interpretation
f o
island. On the contrary, most of the sites are observed at the geophysical anomalies. The high-resolution magnetic survey
boundaries of geological regions, suggesting that there was a conducted at the E.C. site, in an effort t o locate its boun daries
tendency to have access to various natural resources (Fig. 8). and recognise residues of past habitation (shelters, to mbs,
A similat conclusion can be drawn by the watershed analysis. etc.), had limited success. The nature of the targets on one
In the particular process, watershed areas (namely hydro hand, consisting mainly of rock-cut shelters and other similar
logical drainage regions) and flow paths were calculated based features, and the different levels of the terrain on the other
2, 3, 4, & 7.
on the elevation values. Drainage patterns were also clearly est to spectral bands 1,
hand, imposed specific obstacles on the survey and the .
as OJ
,f'p rob able archaeological i nter
represented by the combination of Principal Components 4, 3 interpretation process. Still, the methodology used was the F.ig. 9 Classiti
'
" cat ion model rela ting are
.
es
h e spet ral signatur
& 2. Most of the settlements of the island have been found at hou h, sup erv ised clas sification of t .
only way of detecting the specific remnants, compared to . .
ncuon of the arch aeo log ical info rma tion Alt
has been relatively sati
fac to i
: m t:
h
the boundaries of the drainage polygons. This is also in other prospection techniques, even to the traditional surfac e system, and the rest i a1 !
of arch eolo gica l sites
data The spe ctra l sign atu re of arch aeo log . .
t h e kn o w n arch aeological sites, th
agreement to the general observation that most of the survey. Similar restrictions due to the bad preservation of the tothe publis hed
.
rec om zed on1 w he _Y d ling with correct registration o overestimated
sites can be con fide ntly maps h av e probably
settlements extend on the top of the hills of the mainland. monuments and the changes of the landform were also .
: : har act er cor resp onding thematic . .
env iron me ta , the accelerated
sufficiently large are as
wit h sim ilar l ogical mtere st. Still
Prediction modeling was also perlormed based on supervised encountered in the case of the Classical-Roman port of Kato
par ame ters on the the are as of arc hae o d f or an
. . men tal the nee
isucs. The variability of
the environ . ural pre ssu res indicate
classification of the Landsat spectral signatures. Modeling was Akrotiri. of arc hae olo gical environmental and cult an d use d
ility ned
ited availab can be con tinu ous ly refi
.
limited to the lower part of the Landsat imagery, which was Changes in land use strategies and soil erosion processes one hand and th e lim . s on arch aeo1ogic aI GIS that
rest rict ion (F 10).
011 141
D
a 11
85
EASTING
.
. ,f' known & published sites).
(mapping OJ
haeological map ,f' the island of Amorgos
OJ
Fig. 8. Correlation of the digitized geological map with the archaeological inventory. F.1g. JO . Electronic arc
155
154
Sarris, A., Maragou'
L" Gk'1ou1vu, A., Karatha ' '
Gava las, G" Asvesta
nasis' Ch" SteJ'F
0 &. d"
. 11/1', M., M.avro1 1s' " ' " Kan. ma /'
111 .
s, G., Georoi/a .1, E " Gwn
noufi' E
, K., loa1111oi1, D. .,
vrontak '1.' E
o
R eferences
!l .
arvard Uruver . , Bos Past", 1raborato1y of Geophysical, Satellite Remote Sensing & Archaeo-environment, Institute of Mediterranean Studies,
sity ton, USA., 14- 16 Octobe
1:atans, A, 1965, The geo r. un dation of Research & Technology, Hellas, P.O. Box. 119, Rethymnon 74100, Crete, Greece, asaris@ret.forthnet.g1;
tectonic position of Amorg Fo
os island ' z m of Mineral Resources Engineering, Technical University of Crete, Chania, Crete, Greece, vafidis@mred.tuc.g1;
Bull. Geol. Soc. Greece
6, 239-247. oepart ent
328 Montcalm, Ramonville, 31520 France,
4Geophysical Laboratory, University of Thessaloniki, PO. Box. 352-1, Thessaloniki, Greece
For the past 4 years, a coalition of European researchers has been studying the archaeological site of ltanos in North-East
rete, Greece, from various archaeological, environmental and geophysical aspects. The project, supervised by the Institute of
C
Mediterranean Studies in Crete and the French School of Archaeology in Athens, has integrated a number of surveying
q accessing the archaeological and environmental parameters of the wider archaeological region. The purpose of
tec hni ues for
the geo phy sical project has been to map the buried archaeological relics, including remnants of buildings, streets and walls in
the ex tend ed area of the archaeological site. Furthermore, the site has been used for testing a number of high resolution
co nve ntional and modern survey techniques. Seismic refraction and reflection techniques were used for detecting the ancient
port of Jtanos and mapping the bedrock of the area, covered by alluvium deposits. The ancient port was also the target of CPR
and electrical tomography prospecting, which encountered serious problems due to the shallow depth of the water table.
T he geophysical, topography and archaeological survey data have been superimposed on to an aerial mosaic of the region.
The overall assessment of archaeological sites has been extended through the use of Landsat and SPOT satellite imagery (the
last of which has been re-sampled to 5m accuracy, through pixel-mixing techniques). Sub-centimetre accuracy with CPS
re ceivers has been used specifically for the above image registration as well as for precisely locating the archaeological,
environmental and geophysical features. The archaeological site of ltanos has been evolved to be a natural Remote Sensing lab,
where dife
f rent scientific approaches have been applied to effectively extract the archaeological information of the site.
6). The
h of the two acropolis (Fig.
n techniques was identified in the region sout
by different prospectio the theoretical assu mpt ions , which
.
g of the site sical model is in agreement to
s e
in orde
r to examine the
behavior of the geophy
tati on proposed the specific loca
tion as a better cand idat e for the
nce level of the interpre ic
d1d1 and in
crease the confide
ieve d coastal installations of
the site due to its geog raph
sical images was ach
51gna 5 Enhancement of geophy
s and properties.
proces volution of gradient,
Laplacian, high-pas
ion survey was implemente
d by
the con The ground-based prospect
tbfO:
ball .P
ass ope
rators (Fig. 2). But terwmth and Gaussian
domain have been proven
effective geodetic surveying using an
Ashteck total station and
the
a
kine
mob
mat
ile
ic
the frequency e was applied for
n g in
related to the unit. "Stop & Go" techniqu
fjlt en sho1t-range anomalies
.
a num ber of
ing the on of Itanos, taking
utl in and other surveying of the wider regi
.
ill te ctural remains of
the site. Linear features
ugh the application of dire
ctional high-accuracy read ings of grou nd control points (GCP) to
and sate
be
llite
arc ' ls were emphasized thro correction of aerial
: tal
te rin g (1 st and 2nd
horizontal derivatives).
evident in the electroma
The rem
gnetic
ova l of
dat a,
used for the geometric
images. A number of expe riments of static surveyin
accuracy of the syst em
g were
and
lo ca l trend, especially 3). conducted for testing the
tbe ies (Fig.
essary to enh ance the residual anomal grids and profiles to the gene ral
was ne c . collect field data for registering the geophysical
was used to e obta ined
g I Swift system site. Measurements wer
The Stin ys, topographic layout of the
nner and dipole-dipole arra to the Greek Geodesy Syst
em
imaging. Both We in UTM and then tran sfor med
resistivity sure
1 m, were employed to mea measurements were used
for the
ctrode spacing 0.5- of Axes ('87). The above
wi th ele ). The 2-D
at different depths (Fig. 4
Seismic Survey (Seis-gun) l ima gery .
t resistivity values atic and multispectra
appa ren ugh registration of SPOT panchrom
arent resistivity data, thro
of the measured app
jnversion and a non-linear least
ce forward modelling Final remarks
finite-differen
into true resistivity was
optimization technique,
squares n and interpretation
accurate depth determinatio ns of
following the same directio
necessary for ography was Site recovery will continue
Barker 1 996). Electrical tom sical prospection wor k will be
of th e data (Loke & avations verified the surveying. Fmther geophy
area of grid H, where exc use of conventional and non
applied to the tified by other carried out through the
ber of wall structures iden ing the geophysical sign atur e of
existence of a num conventional techniques, refin
th of the two
es, and in the open area sou ng the interpretation proc
ess.
geophysical techniqu subsurface targets and imp rovi
nd penetrating
a pulse EKKO 1 000 grou only be required to inve stiga te
acropolis. In addition, In this way, excavations will
. 5). Pre
in the same regions (Fig rest, providing a feed back to the
radar was employed regions of significant inte
saturation
signals included a signal n. The goal is to study in dept
h the
proce ssing of the radar process of data manipul atio
ity of the
level removal. Continu methodologies used and
take the
correction and a DC dynamic potential of the
by the application of a gain t of the
reflecting signals was assessed appropriate actions tow
ards the improvemen
ortio nal to the signal strength. valu able
control filter, inversely prop thereby supplying a
techniques and sensors,
appl ied along the profile of the . The
A horizontal average filter was e prod uctive application of them
flat infrastructure for a mor
manner for emphasizing port's basin will be inte grat ed by
Global Positioning System (Base Station) Electromagnetic Survey data in a trace to trace 3-dimensional model of the
ctors in the grou nd (Bev an & be fed to
lying/slow1y dipping refle the resulting model will
e further seismic surveys and
. The purpose of the abov eate the
Kenyon 1 975; Goodman 1 996) ounding region to recr
the the DEM of the surr
in a clear picture of stion.
experiments was twofold: first to obta of the particular area in que
ains and second to environmental settings
signal expected by architect ural rem ronm ental
, geological and envi
g and seismic survey SPOT Panchromatic imagery
correlate the data with the results of corin regi on will
al data of the wider
parameters, and archaeologic
Info rma tion
(Camerlynck et al. 1 994). ractive Geographic
be synthesized to an inte
to map the top of the ent
Seismic surveying was performed for a more rational man agem
of the overburden, System, which could be used
basement and calculate the thickness regi on. This
ental resources of the
reflection and of the cultural and environm
which consists of alluvium deposits. Both of grou nd and satellite prospection,
GPS
and hammering type of integration e
refraction methods were employed. Seis-gun will continue in an even mor
ic waves that and GIS technologies
were responsible for the generation of seism llanos as a natural rem ote sens ing
The seismic systematic way nominating
were recorded by 48 units of 1 4 Hz geophones. nts wou ld allow us to expand and
modify
orth and lab where experime
trave rses were set up in approximately south-to-n the future.
es were the research strategies for
east-to-west directions, whereas parallel profil
hone
separated a distance of about 1 5-25 m and geop
at
References
separation was 2 m for all profiles. Shots were performed
of vary ing
e middle of the arrays, at both edges and at offset 1 975, Ground Penetrating
Radar for
eve, depe ndin g on the Bevan, B. & Kenyon, J.,
distance when this was poss ible to achi ca News Letter 2( 1 ), 2-7.
all Historical Archaeology, Mas
restrictions of the landform and vegetation coverage. In M. & Panissod, C., 1 994,
the Camerlynck, C., Dabas,
s ections, processing of the data resulted in three layers of and four Elec trom agn etic
Comparison Between GPR
s bsurface. In addition to seismic refraction techniques, a
Methods for Stone Fea
tures Characteriz atio n: An
ed out
high resolution seismic reflection survey was carri Prospection l, 5- 1 7.
Hz Example, Archaeological
along one of the profiles, with the use of a seis-gun, 00 1
Pen etrating Radar Simulation
in
Goodman, D., 1 994 , Gro und
rec eivers and a 0.5 m geophone separation. The optim um 224 -23 2.
y, Geophysics 59,
) was 7 m. Engineering and Archaeolog
wind ow (offset between source-first geophone R. D., 1 996, Rapid Lea st-S qua res
Loke, M. H. & Barker,
Global Positioning System (Mobile Unit) Vertical iHagnetic Gradient Survey
J Pro cessing of the seismic reflection data included geometry
co rrections, trace muting, AGC, stacking, deconvolution, and
Inversion of App aren
Quasi-Newton Method,
t Resistivity Pseudosectio
Geophysical Pro spec
ns by a
ting 44,
. ed the
nugration. Processing of the refraction data has offer
1 3 1 - 1 52,
fir st 3-dimensional model of the ancient port which has been
Fig. 1. Details of the Geophysical survey in ancient Itanos, Erimoupolis (E Cr,ete, Greece.
)
1 59
1 58
Sarris, A., Vafeidis, A., Mertikas, St., Guy, M., V rontaki' E
., man
u akou M &
'
valp
" A ncient Jtanos (Erimoupolis, Lasithi): an archaeological site as a remote sensing laboratory
axis
, Th.
14
12
10
_g
8
;r.:
E-<
""
0
:z
4
2 4 6 8 10 12 14 16 18 20
EAST (m i
12
10
_g
8
::i::
E-<
""
0
z
2 6 10 12 14 16 18 20 22 24 26 28 30 32 34
EAST (m)
_g
8
::i::
E-<
6
0
z
4
2 4 8 10 12 14 16 18 20 22 24 26 28 30 32
EAST (m )
Fig. 2. Top: Bandp ass filtering of magnetic data. Bottom: A: Grid H. Upward con tinuatio'.1 ofmagnetic Fig. 3. Vertical magnetic gradient was successful in mapping the shallow depth archaeological relics of the site.
abo ve the suiface). B: Grid H. Butte1worth filtering of magnetic data (cut-off: 0.125) C: Gnd H. un;;:rt
(8 m T he application of directional filters and the calculation of the residuals emphasized
of magnetic dat a (cut-off: 0.250) D: Grid H. Butte1worth filtering of magnetic data (cut-off: O. ' a number of features probably correlated to remnants of wall structures.
160
161
Ancient ltanos (Erimoupolis, Lasithi): an archaeological site as a remote sensing laborato1y
0 .9
1 .7
2 .2
2.1
3 ,3
3,9
4,6
i!JH.J; Ila
64. 0 .. - -
128 25 6 5!2- -
Unit Electrod
e Spac
ing -
_ Le
M
5eud0 1 .0
J)eJ'th
9 ,5
1 .0
1 .5
2 .0
2.6
3.1
3.6
4.1
4,6 cl A J'arent Resistivit
p y Pseudosecti on
re
easu
5eudo 1.0
l
Del'tl
9,5
1.0
1.5
2.0
2.6
3.1
3,6
4,1
4,6 "AJ'J'arent Resisti
vity Pseudosection
ate
a1c:1.1l
iteration 9 RHS e.r.ro:r = 3. 3 X
5.0 9.9 1.3.9
1.0
DeJ'tll
0.z
9,8
1.4
2.z
- le
D.pole Dipo (TOP) and Wenner (BOTTOM) tomography (a=I m), para
Fig. 4. i llel to an excavation trench.
Fig. 5. G
.r gn'd H
round penetrating radar cross-section in the area OJ - P1V1
along the direction of the trench (W to E).
.+:fe
162
163
Sarris, A., Vafeidis, A., Mertikas, St., G11J\ M., Vivmaki, E., Manakou, M. & Kalpaxis, Th.
SCURTU,E. F.
-20m
-nm
-24m
-26m
The paper describes the results of a geophysical (magnetic and geoelectric) siavey on an area of more than 2 ha where no
-28m ace of ancient Roman buildings of the town (that had about 20,000 inhabitants in the 3rd century AD) is visible at the surface.
-.Hlm
-32m
-34m
he main result of the judicious use of geophysics (especially a high accuracy of field measurements and an appropriate
the data) is a veJ)' precise plan of the area (Fig. 5), that s hows streets, buildings and even room entrances.
filte ing of
-]6m r
-3Sm
40m
KEYWORDS: GEOPHY SICS, ARCHAEOLOGY, DACIA, ROMAN TOWN, FORT, BUILDINGS, ROSETTE.
Classical archaeological methods need a huge expenditure La Tene epoch due to some Dacian small forts built up during
o f human energy. More than that, as someone said, the 2nd-l st centuries BC around a large one on the Magura
-14m (magnetic and electrical) methods on a smface of about 2.2 ha area. It was a big production and commercial transit centre of
-17m in the Salaj county, Roumania, right to the south of the Roman about 20,000 inhabitants (Forni 1976). The civil settlement
-18m fort built up on the Pomat hill (Fig. 1). already existed in 106 AD, when a Roman veteran was
established there. After its permanent growth, the town was
-12 m
declared municipium at the end of the 2nd century.
-6m
An other outstanding moment in the history of the town
-30m
Proce
164 edings of
the 31" J11tematio11a/ Symposium on Archaeomell)', lerem, E. & T. Bir6, K. (eds.), Archaeopress -Archaeoli11gua. BAR- Centi: Ew: Sfl: I.
Scurtu, E. F.
The geophysical image of a part of the ancient Roman town Porolisswn in Dacia
/
.......
/ '
/ ' '
/ ,,.,. --- ', \ The petrophysical background of the geophysical survey at
I ,,..
---...
I I '/' I
I /
I I I -" . --/ I Geophysical results
I I I "' /
I t ,,_,.,. II
I I
I, I"'--:"/ . I
I \ ' -- . t\ I The magnetic survey started with measurements on an area
I '---
'\ _,/,:
.. f "t\:)\:) '///}; l)fJI \.:,\'
,-!! l/i / '\\)
S
JI 111U I
\
........ _,/
/
/
/
/y
of 400x250 m, on a 20x5 m giid, with the aim of separating
areas of shallow magnetic sources from those of deep
magnetic sources.
" I
\,
.,-- ,
/
t!j
1\
I I\.
'],
/ ", I.,
.
/!/
. I
'.... -- ,,,,' The next step was to make a pilot survey in a very dense
"
grid (lx0.5 m) on an area about 100 m long and 20 m wide.
I\ I\
I . .____.- /
_,,./'/ <::;
. , . !;
\\ -.--"" // I
After data processing, filtering and graphical presentation, the
I
\ ' ........, .//' I / underground remains of the walls are quite visible, as we
/,,,
'--- =.:"-,,,.. : already concluded from the mathematical modelling made
\
,,.,,
D E F G H J K
A B c
l
8
F. SCURTU
POROLlSSUM, 1996
I
I 7
'it I
m I
Q
21" 22 2,.
6
Fig. 2. Dacia in the first half of the 3rd century AD.
Volcni rocks (quartz-an dhes
ites and quartz-po h ritic
. Th magnetic method makes use of the property of some 5
nucrodontes), epresented
Sarmatlan age, pierce the sedim
by subvolcanic bo e
.
of matenals especially those contrumng iron or iron oxy des
entary rocks m some areas
166
167
/issum in Dacia
t of the ancient Roman town Poro
Scurtu, E. F.
The geophysical image of a par
objective
ological methods, this
Using classical archae
: ::;eo:o; but te map is _still rather confused. 6x6 m each (E4), just NW of the rounded
Part of the file (AB
in Fig. 1) of
vertical electrical sou
ults "".ere qua ntit ativ ely
ndings,
inter would take probably
dozens of years, wit
hout
large
tak
sur
ing into
face by
t, whose res truction on a very
Oflent
roac1,
account the land des
are c ear on this map because the image 1s clear enough to show even the
west-eas
;: ;::
s
:n
: : s
e squares 13 and 14 and
. .
Some trends point to a kind of wall ms1de t
its signature is different to the external o
he big
e
rntran
o otn ed 7s
n
a ve th
e depths
and the
under
of the lim
lay ing sandstones and
conglomerates
s rocks in the
uncovering the bm ied
gical excavations on lim
constructions. Of cou
ited areas are necessary
, but the y must
oriented NW E or -SW, parallel to the fort walls. which appears very often on all kind of monum gn eo physical resear for its aid to present the
. ents found. Th e g
, allowed us to pre sent evidence
io11owmg ones
The most evident features are the c the poro 1.issum area, it seems to be the town embl Ill on the Pomat hill
rt
1 . Two parallel walls, at about 7 m distance of e ch other,
. .
case this building including a representation of
e m. In I tha
flm an fo
number of ancien
t constructio ns, absolutely
References
the to a large ent ed in the
0 arly repres
. wn face but very cle
markig ?ut a Roman road climbing from SE (F2' E3), mblem may be
one of main importance fo r the . vis ible at the sur
town. 1lte rk. gli amfiteatrei in
the geophysical wo dagine demografica e
changmg its direction after about 70 m from the south border
. . .
image of the central pedestal of the rosette may be .
mte rpret ed age due to
spectacul ar res ult of this research is the
map Forni, G., 1976, L'in
-lSl.
of the map (E3'E4), contmumg about 80 m to NE and gettmg as a s1ope (stairs?) descending to the un derg The m ost tte r of Dacia, Apulum 13, 141
round c measurements. As
a ma arheologic daco
olis sum, un complex
out of our map (H6). The route is divided longitudina11y mto . m the magneti Gudea, N., 1989, Por
deduced fro
construction described above.
.
two equal parts (as It is v1S1ble especially in F2, but t races are 4. A construction of circular section a d small (about
2 m) fac t, it is al complex at
. . .
, . 17 centuries ago. Romanian archaeologic
also V1S1ble m F4 and GS). As the f:ort s gate porta principalis diameter, visible inside the road arch (F , seems to be o mans more than (Porolissum, a Daco-
also a left b y R on a limited
. rical me asu rements, made only of the Roman empire
), Acta Musei
dextra is also d1"v1"ded mto two parts' the westward one statue pedestal, or may be a fountain. The g eoelect inf ormation on the northern border
resting
lack of time, adds inte Porolissensis 13, 1-1
178.
reserved for the carts and the eastw ne fo r the pedst an,
.
S. Eastwards of the route a lot of bm1dmgs (some o em fth area due to the
a. ei (Atlas for the
structure of the are Pascu, S., 1983, Atlas
pentru istoria Romani
it seems that this route enters the by porta przncipahs "".1th rounded comers generally parallel to the route change of the geological of
. _ were obt ain ed by a judicious use Bucarest.
dextra. The archaeo1og1ca1 trench dug in F2 (the two parallel direct10n) and streets are visible. These results ing the history of Roumania),
stage (inc lud of resistivity
. ting with the project Computer calculation
alignements oriented NE-SW reflect the heaps of soil taken 6. Westwards of the road some othr build.ings are visible geop hysics, star Scurtu, E. F., 1971,
sic al signature of the ductor body,
ing of the geo phy a buried spherical con
out from the trench) have confirmed the geophysical image in (A7' A8, B7' B6' BS' CS)' but their walls direct1ons are mathematical model pseudo-sections of
h accuracy of the
. continuing with the hig g 20(3), 60S-62S.
the area. vable and generally different from NW-SE or NE-SW expected targets) and Geophysical Prospectin
processing of the , E., 1986,
.
2.
large building of 70x40 m (columns D-G ' rows 4-7), whic are characteristic fo r the east zone, much more field measurements,
with an appropriate
proper interpretation
of the final Stefan, A., Rusu, A.,
Bratosin, I. & Colios
in the link zone
field data and with the the alpine magmatites
mcludmg a smaller one' of 32x15 m, and three others of orgaruzed. Petrological study of
-Gutai-Tibles
geophysical data. Mountains and the Oas
between the Apuseni
absolutely non-destructive, uresti 70-71,
The geophysical research, Institutul Geologic Buc
volcanic chain, DdSS
st be exte nded to a larger area,
quick a n d reasonably cheap, mu town 242-262.
8
map of the entire ancient Roman
in order to draw the
ient Rom an
capital of the anc
Porolissum, probably the
province Dacia Porolissensis.
7
6
(.__,
2
:.../
... I"' I/
v i'
I\
5
v
4
' '
["...lu- 7 \ v" I/ .
4 6
'
I ' I/
\.
3
8
I 1'. i/ I
10 .....
2
12
\ IJ i
14
340 350 360
300 310 320 330
X (metri)
260 no 280 290
230 240 250
Ii (rnetri) J9Q 200 210 220
150 !GO 170 180
110 120 !JO 140
RO 90 100
. 1.
ng the AB profile in Fig
ermeable basement alo
Fig. 6. Relief of the imp
Fig. 5. Filtered map of the surveyed area.
169
168
ENVIRONMEN TAL CH ANGE AND HUMAN ADAP TATION
IN THE C ARPATHIAN BASIN AT THE LATE
GLACIAL/POS T GLACIAL T RANSI TION
'Department of Geology and Palaeonthology, University of Szeged, 6701 Szeged, P.O.B. 658, Hungary,
2Damjanich Museum, 5001 Szalnok, P.O.B. 128, Hungmy,
3Nuclear Research Institute of HAS, 4001 Debrecen, P.O.B. 51, Hungmy
ng climatic, bedrock and soil conditions from the viewpoint of early agriculture these conditions seem to form a
Consideri
. with in the Carpathian Basin which detennined the chances of the northern distribution of Early Neolithic Koros-Starcevo
1' it
u e o Balkan origin. We called this line, which during the Early Neolithic limited the northem distribution of Balkan type
r r f
olithization in
the Carpathian Basin, Central European-Balkan Agroecological Barrier (CEB AEB).
Different palaeoecological data indicate that a strong global reminiscent of those developed in the Near East. The Early
climatic change occurred at the late glacial/postglacial Neolithic cultures of the Cai-pathian Basin aie characterised
iransition, between approximately 12,000-9,000 years BP. by dense occupation along the river and creek system (Kosse
Climate change at the late glacial/postglacial transition resulted 1979). These alluvial environments were attractive to early
in a dra matic change in the ecosystem because the rapid and farmers because crops were cultivated by intensive
strong change in climate resulted in some rapid changes in the horticulture on hydromo1-phic soil (Sherratt 1980). Sheep,
environmental processes, such as landscape evolution, soil goat and emmer were their economic staples but these faunal
formation, plant succession and faunal migration (Roberts and floral elements lived on the borderline of their natural
1989). Thus, during a short transitional period from glacial to distribution area so they could survive only with human
interglacial conditions a new environmental situation de protection (Bokonyi 1974).
veloped, and a new geological epoch started that is called the To the north of this zone two groups of the Linear Pottery
Holocene. This new environment transformed the habitats of Culture (LPC) formed (Kalicz & Makkay 1977; Kalicz 1983;
the human populations and their physical and biological Makkay 1982) from 5500 cal BC (Hertelendi et al. 1996). The
environment, thus the potentials of the human communities' economy of LPC was based on cereals but animal husbandry
economy changed again. As a result of these changes some switched from sheep and goat to the reliance on cattle and pig
cultural modification began, and Mesolithic followed by whose wild progenitors were available for local domestication
Neolithic cultures developed and spread out in many areas of in Southeast Europe (Bokonyi 1974). The Neolithic agro
the world. The archaeological data show that one of the most ecological system had changed, too, when the first and local
important and earliest agricultural centres formed in the Near groups of the Linear Pottery Culture developed in the
East approximately 10,000 years ago and moved up to the Carpathian Basin. These communities started using the
Balkans finally reaching Northern Europe approximately 1979; Sherratt 1981,
loessic soil-vegetation ecosystem (Kosse
between 5000-4000 BP (Ammerman & Cavalli-Sforza 1971; 1982). One of the first LPC developed in the western part of
7.ohary & Hopf 1988). the Carpathian Basin. This branch of the LPC subsequently
Reg ional effects of this global climate change can also be spread fast through the central, western and northern parts of
detected in the different palaeoecological sites of the Europe (Ammerman & Cavalli-Sforza 1971).
Carpathian Basin (Willis et al. 1995, 1997; Siimegi 1996; But why did the Balkan type of neolitization process stop
Kordos 1977). This region (including Hungary) is an import in the central part of the Carpathian Basin? Why did animal
ant intermediate zone between the Balkan Peninsula and the husbandry, land use and culture change in this region? Were
estem, eastern and northern parts of Europe. Thus it plays an there any settled Mesolithic communities ready to accept
unportant role
in understanding the interactions of migration, farming as a way of life or was there a barrier? Archaeological
ettlem ent process, culture spread and environmental changes and palaeoecological data suggest a number of possible
Ill Europe. explanations. Our investigation aims to find some answers to
Archaeological evidences suggest that the first farming these questions.
50
01l1Inunities developed in the Carpathian Basin from ca.
0 cal BC (Hertelendi et al. 1996). The earliest agricultural Methods
unities belong to the Koros (including Transylvanian
Oros) and Starcevo cultures, long recognised as a part of the We collected samples from 25 Late Pleistocene/Holocene
B al a n Ea
1 k rly Neolithic complex (Kutzian 1947; Kalicz 1980, sediment sequences of different palaeoecological and geo
983, 1990; Kalicz et al. 1998; Raczky 1988, 1989; W hittle aichaeological sites for sedimentological, geochemical, pollen
19 6). These two groups developed parallelly in the south-
9 and quaitermalacological analyses, wood anatomy, and radio
easte rn,
eastern and southwestern parts of the Carpathian cai'bon dating (Ke1tesz et al. 1994; Nyilas & Siimegi 1992;
Proceedi
ngs of the 31" Jmernational Symposium on Archaeometry, Jerem, E. & T. Biro, K. (eds.). Archaeopress -Archaeolingua, BAR - Cellf1: Eiu: Se1: 1.
ition
: at the late glacial/postglacial trans
1uan. basm
nge and human ada ptation in the Cmpat .
Siimegi, P., Kertesz, R. & Hertele11di, E. . ental cha
Enwrom
S i.imegi 1996; Willis 1997; Willis et al. 1995, 1998). Chron parkland in the central part of
the Carpa '
ology was obtained from 118 radiocarbon age detenninations. Koml6i 1987; Willis 1997; Willi
h
Basin (
\ ..
/-:.,)'
s et al. J
,., \: !
9
7 l
The radiocarbon dates were calibrated using the CALIB 3.0 The climate and vegetation chan
: 199 ,
ge could '
program (Stuiver & Reimer 1993) and then conve1ted to cal BC
h ave c
ch ange m soi 1 format10n processes au
.. and th e fi
ge in
naJ sta .
to enable comparison with archaeological data. According to trans1t10n process was the transfoi
mation of p
oiJ
od zoJ s
these radiocarbon-dated palaeoecological records we re brown earth (Willis et al. 1997). The
molluscan
. record .
constructed the vegetation, faunal, soil and environmental early post glaci al also indicate an s tn
inter mi
changes at the late glacial/postglacial transition and the early ecologically different faunas. There x re Of
. are characte .
YI
. nstt cau
Holocene environment with the human impact. glacial, cold-re sistant species and
ear]y p
For the archaeological interpretation, a review was made of thermophilous species. These highly ostglacf
. mi xed co
mmuni ti
published archaeological sites covering the time period from which southe astern European and es
Bore o-AJp .
t n monu
t h e Mesolithic to the Middle Neolithic within the Carpathian species lived together, h ave no mode
m analog
Basin (Barta 1980; Dumitrescu & Vulpe
..
among regio
u
1988; Kalicz 1990; compos1t10n of mollusc faunas varied
ns and I
&
.
Kalicz Makkay 1977; 1996; Makkay 1982;
Kertesz was very Slilll 1 ar to the reg10nal mosai
c vegetatio
n Patterns
.
Neolithic agro-ecology (climate, vegetation and soil type), and 1997; '."illis et al. 1995, 1997) and malac olo
gical a
(Siimeg1 1996; Krolopp & Si.imegi 1995), continu
model the interfaces between the environment, society and
e d on
economy during the Early Holocene (Siimegi & Ke1tesz 1988). different regions of the Carpathian Basin. These
postglaci : ;
regional environme ntal differences are very simila
,
r to the
ppe-forest climatic zone,
. Hungary. CF - Temperate cl.1mat'ic zone, BS= Ste . e,.rr.
. atic JJec t
Results recent deviations in climate (Fig. 1), vegetation (Fig.
.. tic years 2. West European clim
2) and
F.ig. 1 Koppen's climatic
.,
ranean cz1ma
region s zn
JFe
soil type (Fig. 3), which characterize the Carpat hian
Boreal climatic zone, 1.
Basin of sub med iter
DF
,
enc
today (Siimegi & Kertesz 1998; Kertesz & Siimegi l999).
The equ
. JJ' et'. 4. Submontan climatic effect.
y c l'imattc e,Fre
The radiocarbon dated molluscan (Willis 1997; Siimegi :
3. Submed1ter . ranean
1996; Krolopp & Si.imegi 1995) and pollen data (Willis et al.
=
climate became progressively warmer and wetter in the Early Holocene. The pollen and charcoal records of several
Carpathian Basin. This increase in both precipitation and Hungarian sites show correspondences between some small
temperature have enabled transition from forest steppe to peaks of hazel pollen and microcharcoal (Willis 1997; Willis
broadleaved coniferous forest and ma ximum expansion of et al. 1995, 19 97, 1998), which suggests that p eople may
coniferous forests. Within the boreal forest there were also have brought a bout the vegetation change before 6500 cal
l
pockets of deciduous trees. Composition of coniferous forests BC. However, there is no archaeological evidence for
predominantly was composed of Picea and Pi1111s in the Neolithic occupation of the Carpathian Basin before this time
eastern, P inus, P icea and Larix in the northeastern and Pinus ....
(Hertelendi et al. 1996; Chapman 1994; Whittle 1996).
in the c e ntral parts of the Carpathian Basin (Willis 1997). The Probably, these palaeoecological and archaeological data
combination of the acid litter of coniferous trees, and the cool indirectly show that the h-f-g peoples of the Mesolithic , ...
late glacial climate could have ensured that a podzol soil period used fire for the alteration of the vegetation in this
develope d in some places of this region (Willis et al. 1997) region about 7000 cal BC. These results are consistent with
where the bedrock was acid. At the same time malacological both the new and old archaeological data (Gabori 1956; Barta
records indicate that there could have been areas where the 1980; Kertesz et al. 1994; Kertesz 1996) as far as these data
vegetation cover showed a mosaic pattern, although the have indicate the development of Mesolithic activities within
dominance of open habitat preferring mollusc declined. The the Carpathian Basin at that time. Some e xcavated
ed
climatic and vegetation change resulted in the extinction of Mezolithic sites and finds are located close t o the ana lys
palaeo-ecological sequences (e.g. t h e Cuimesti II lith ic
the cryophilous mollusc in the central part of the Carpathian Meso
or th e
site can be found c. 2 km SW of Batorliget marsh
1
to higher sea level altitudinal regions of the Carpathians. vicinity of the Sarret core-point). Thus ther e are som '3
1
an
Composition of the Late Pleistocene mollusc faunas suggest palaeoecological evidences for Mesolithic burning 5rz:I
a S
vegetation modification in the Carpathian Basin at 7
.
rutle
Carpathian Basin. BC. These data suggest that the Mesolithic cou 7
li fe, an
adapted to the transition of Neolithic agricultur e
8-
After the development of taiga environment the late
glacial/postglacial transition occurred between approxi they were in the substitution phase (Zve!ebil & Ro wley ,
. '
step-fo
orth .
. /us0Querc1
.
,_,;''
.,
j \
.. ... .
.
\. L
.'J l
1
\::
.; .
:
1
10
a
11 oO
2
Ill
cE'.2
.. ; . .j
L
IV
"" d
.. -
. .
. --: . ......,...
'1 ,....
Fig. 3. Distrib
ution of soil types in the
1. Distribution of the Carpathian Basin.
soil types which develo
ped on loess bedrock; Fig. 4. Central European and Balkan A groecological Barrier and the distribution
on san 2. Distribution of the
dy bed rock; 3. Distrib soil type which develop .
ution of the soil type wh ed .
of Mesolithic sites, green corn do rs and. the first tmces of human impact on the vegetation.
ich developed on com
pact bedrock . .
(volcanic, metam01phic I. Strong human impacts (burning, soil erosion) between 6 -6500 l BC 11 Strong human impacts (burnmg, soi1 erosion )
or limestone rock). '
r
,
the Early Neolithic culture of Balkan origin. We called this line, This was supported by OTKA
research Oxford. Stuiver, M. & R eimer, P. J.,
1993, Extende
which during the Early Neolithic period limited the nmthem F-23129), FKFP (No. 502 and 2006 (F-Ol8Q5g Reports, & umegi, P., 1995, Palaeoecological recon- ram,
S I4C age calibration prog
f{!OloPP
) and For CAL IB 3.0
expansion of Balkan type neolithiz a tion in the Carpathian Prehistonc Archaeology of the Jaszs
. .
Studjes
ag Foun d . on the
,
of the Late Pleis
tocene, Based on Loess
revised
at ion. str11 ctl
on Rad ioca rbon 35, 215-230.
Basin, Central European-Balkan Agroe cological Banier (CEB
. .
K 1acofauna
Geolournal 26, 213 -22
2.
. ptfragen des Neolitikums
er
Trogmayer, 0 ., 1968 , Die Hau
in Hungarv
1n3
J,
AEB: Siimegi & Kertesz 1998).
t10nes
References
, eum E v-
1947 , The Koros
Culture, D1ss erta
Mora Fere nc Muz
an, I., Ungarischen Siidtiefebene,
l{utzl ae II, 23.
pannonic ' t1e
I kanyve, 11-19. . . .
Conclusion Amm erman,A. J. & Cavalli-Sforza L L.,197 M New results in the rese
arch 01
in the Neol1t/11c, Cambnd
ge
' 1, e J., 1982, Whittle, A ., 1996, Europe
tAakkHay,nga.nan Neolithic Akademiai
asur i apes t.
in Europ e, ng the
rate of spread of early fanning . Kiado, Bud
. University Press, Cambridge.
Man 6, t Ukraine, Kiev. cm
'
CEB AEB, determining the northern distribution of the 647-688. 11 L. G ., 199 1, Mes olith ic in Wes
act of Early Agriculture upon
the
tJa tskev o1 , Willis, K. J., 1997, The Imp
&
,
K0ros-Stareevo culture groups with Balkan contacts in the Ba1ta,J., 1980, Das Paliiolithikum und M . 1
Rus a Hungarian Landscape, in Cha
pman J. Dol ukha nov, P.
esolithikum Molluc fauna f
Carpathian Basin, existed only in the earliest Neolithic. The Sumegi, P., 1992, The Euro pe
Nyd. as,
S/ovenskd Archeol6gica 28, 119-13 , Central and East ern
6. . and m (eds .), Landscapes in Flux
expansion of populations having Mediterranean economic Bokonyi, S., 1974, History of domestic , gy at the end of the Pleistocene (Wurm 3)
mammals in Ce I Hort ob a in Antiquit y, Oxb ow Boo ks, Oxford.
and cultural roots became less intensive in the marginal zone nt a ene, P roce edin g of 10th
Internat iona l Mala-
& Toth,A., 1995: The Late
the Holoc
and Eastern Europe, Akademiai Kiad Brau n, M.
o, Budapes egi, P.,
.
. t. Willis, K. J., Siim ,
.
of the area where climatic and cultural influences of Balkan Chapman, J., 1994, The ong . ms of
farming in sou h Congress, 48 1-4 86. . History of Batorliget, N.E
t East o ological .
der Slowake1, Quaternary Environmental
Europe, P re'I11sto1re Europe,enne 6, 133-1
. . ranu k m
origin flourished, that is, during their northward movement , 1980 , Alte re Line arke y, Palaeoec olog y, Pala eo
56. pavuk, J., Hungary, Palaeoclimatolog
within the Carpathian Basin, and even stopped when they Dumitrescu, V. & Vulpe, A ., 1988 ol6gica 28, 7- 90 . . .
' Dacia 1nazu Slovenska Arche , . es , 7.
de
te geograph y 118, 25-4
,
kronl gza1
reached CEB AEB. W hat did it mean from the viewpoint of . 8, A Tisza-videk ultural1s egi, P. & Toth, A., 199 7, Does
1 is, K. J ., Braun ' M. ' Siim
dromihete, Bucuresti.
Raczky, .
P., 198
-versa'].
W'll
those autochthonous Mesolithic communities which lived to Gabori,M.,
kapcsolata1 a B
a neolltzkum,
1956, Mesolitsche Funde a lkdnnal es az Egeikummal n chan ge or vice
von soil change cause vegetatio
g
, Szo dli et
the north of CEB AEB? Archaeol6giai Ertesftff 83, 177-182. Szolnok. Hungary, Ecology 78,
rezkor idffszaktiban, A temporal perspective from
,
.y a d
1. Human groups, possessing inn ov ations from the Early Hertele di, E., Kalicz, N., Raczky, P., Hor vath, logi cal F ramework of t h.e Earl
F., Vere s, E., Raez P., 1989, Chr
ono
740-750.
n, M., Bennett, K. D . & o
, , 10n, Vana
I. & Bartosiewicz' L ., 199 Tisz a Reg T'th,
Willis, K. J., Siimegi, P., Brau
Neolithic food producing economy had came close to Svmgor, E., Futo, ic in the
6, Re.. ddle Neolith
them, thus creating possibility for acculturation. evolution of the Neolithic in eastern Hungar 2, 233 -251. Hun gary : who,
A., 1998, Prehistoric land degradat
y based on Archaeologica Hu nga rica . ion in
. ory,
e. An Env ironmental Hist
Roberts, N. 1989, T
2. Because of the depletion of po ss ibili ties to practise c alibrated rad10carbon dates, Radiocarbon he Hol ocen 101-113.
37, 239-241 how and why ? Antiquity 72, .
Jara i Koml od i, M., 1987, Postglacial climate and
,
al '
History in Hungary, in Kordos, L. & Pecsi, and seasonality m early cere West
M. (eds.),
groups were unable to occupy the lands north of the er soil of cultivated plants m
Sherratt,A ., 1980 , Wat World: the origin and spread
barrier. Thus, Mesolithic communities gained time to ology 11, 3 13- 330 . Press, Oxfo rd.
'
Holocene environment in Hungwy, 37-47, Geogra cultivation, Wor ld Archae Valley, Calderon
accept technical innovations wit hou t being absorbed Institute, Budapest.
phic al
Sherratt, A., 1981, Plou gh and
.
pastoralism: aspects of the
Asia, Europe, and the Nile
Zvelebile, M. & Rowley-
Con wy, P., 1 :
86, Fora gers an
either culturally or economically or demographically Kalicz,N., 1980, Neuere Forschungen ilber die Entstehu secondary products revolutio
n, in Hodder, I., Hanunond, ._)
pe, in Zveleb1l, M: (ed , Hun
te1 s
ng farmers in Atla ntic Euro
erns of the Past, 261 -30
5,
N. & Isaac, G. (eds .), Patt
into communities of Balkan origin . CEB AEB seems to des Neolithikums in Ungarn, in Kozlowsky, J. K. & bridge Um vers 1ty Pres s,
. in Transition, 67- 93, Cam
Machnik, J. (eds.), Problemes de la neolithisation dans Cambndge.
play a fundamental role in the formation of the Neolithic Cambridge University Pres s, Cambridge.
of Neolithic and Copper
Sherratt,A ., 198 2 , The
culture with a completely new, different character, certaines de l'EU1vpe, 97-122. deve lopm ent
Hungarian Plai. n. part 1 : Th
e
adapting to local conditions, n01th of this banier. Kalicz, N., 1983, Die Koros-Starcevo-Kulturen und ihre Age settlement in the Great
l,
3. Those Mesolithic communities whic h lived to the south Beziehungen zur Linearbandkeramik, Nachrichten aus regional setting, Oxford Journ
al of Arch aeol ogy
of CEB AEB had absorbed both culturally and Niedersachsens Urgeschichte 52, 91-130. 287-3 16. .
the neolithization process of eoecolog1ca l R econ-
. p. 1996 , Comparative Pala .
.
demographically into Kalicz,N., 1990, Friihneolithische Siedlungsfunde aus S.umeg1,
at10n of NE Hungari
Mediterranean type, except those places where the Siidwestungarn, Akademiai K iado, Budapest. struction and Stratigraphical Valu
& Makkay, J.,1977, Die Linienbandkeramik in der Manuscript, Budapest (m
possibility for isolation was given for sometime (e.g. in Kalicz, N. Loess Region, PhD Thesis,
the Iron Gate area). Grossen Ungarischen Tiefebene, Akademiai Kiad6, Hungarian).
4. CEB AEB existed only in the Early Neolithic since Budapest.
during the period after the formation of the Linear Kalicz, N., M. V irag, Zs. & T. Biro K., 1998, The Northern
Pottery Complex the neolithization process to the north Periphery of the Early Neolithic Starcevo-Culture in
of CEB AEB produced cultural groups of not Balkan South-Western Hungary,A Case Study of an Excavation
type but of local autochthonous character. Their plant at Lake Balaton, Documenta Praehistorica 25,
and animal stock, production experiences were adapted 151-187.
to local conditions therefore from ecological point of Kertesz, R., 1996, The Mesolithic in the Great Hungarian
view this agroecological limit did not mean a banier for Plain: A survey of the evidence, in Talas, L. (ed.), At the
them. fringes of three worlds, 5-34. Szolnok, Damjan ich
5. Considering the importance of not only climatic, Museum Press.
bedrock and soil conditions in the Carpathian Basin Kertesz, R.,Siimegi,P., Kozak, M., Braun, M.,Felegyhazi,E.
during the process of neol ithization but also & Hertelendi, E., 1994, Archeological and Paleo
hydrographic conditions, it s eems that within the ecological study of an Early Holocene settlement in the
Carpathian Basin neolithization process took place along Jaszsag Area, Acta Geographica, Geologica et
the rivers, the so-called "green corridors". We consider Meteoro/ogica Debrecina 32, 5-49.
green corridors to the south of CEB AEB as Neolithic Kertesz, R. & Siimegi, P., 1999, Theories, critiques and a
infiltration zone, while those to t he north of this banier model: why did the expansion of the Koros-Starc evo
as Mesolithic infiltration zone. Archaeological data and culture stop in the centre of the Carpathian Basin ,
palaeoecological records suggest that the Neolithic Tisicum 11, 9-22. (in Hungarian).
acculturation process of the Carpathian Basin took place Kordos, L., 1977, Changes in the Holocene clim ate of
in these infiltration zones between approximately Hungary reflected by the "vole-thermometer" method.
6500-5500 cal BC. FO/drajzi Kazlemenyek 25, 222-228.
177
176
5. Experimental Archaeology
sCK, c. w.,1 STOUT, E. C.,1 LEE, K.,1 TODTENHAUPT, D.2 & KURZWEIL, A.2
IAmber Research Laboratory, Department of Chemistry, Vassar College, Poughkeepsie, New York 12604, USA,
2Arbeitsgruppe Teerschwele, Museumsdoif Diippel, 14163 Berlin, Germany
Among the pre- and protohistoric methods of manufacturing pine tars in Central Europe are the Meile1; in which an earth
stack of wood is partly burned and partly pyrolyzed, and the double-pot method, in which wood chips in a perforated
covered
vessel are pyrolyzed by external heat while tar is collected in a second vessel below. Experimental tars made by both
ce ramie
been analyzed by NMR spectrometry and their individual constituents have been determined quantitatively by
e thod s have
c-MS in o rder to correlate the composition with the manufacturing method. Both techniques reveal differences in composition
that distinguish between the two manufacturing methods and that indicate that the double-pot method involves higher
es than the Meiler method. Allowing for depositional degradation of archaeological tars, the data can be applied to
temperatur
the ide ntific ation of prehistoric manufacturing methods.
Introduction with a tight-fitting lid. A fire is lit in the space surrounding the
upper pot, so that the charge is heated externally and the tar
The composition of archaeological tars and pitches reflects drips into the lower pot. T his small-scale production method
the raw materials from which they were made and the has been more fully described elsewhere (Kurzweil &
tech nology by which they were made, a relationship that may Todtenhaupt 1990, 1991). For the sake of consistency, we will
be obscured by compositional changes suffered during their use the German term 'Doppeltopfteer' for tar made by this
long deposition in a terrestrial or marine environment double-pot method (Fig. 2). T he Doppeltopfteer used in this
(Reunanen et al. 1990). We have shown earlier that tars made study was made at the Museumsdorf Dtippel.
from resinous wood and from bled resin can be readily
distinguished by the presence of resin acid methyl esters in Analysis by 13C Nuclear Magnetic
the former, and that the manufacturing temperature can be Resonance spectrometry
estimated both from the degree of decarboxylation and from
the concentrations of individual resin acids and their We have shown elsewhere that quantitative information
degradation products (Beck et al. 1998). We now report an about the degree of decarboxylation and aromatization of pine
experimental study in which two pine wood tars, made by two wood tars can be obtained conveniently by a single CMR
different methods, are compared by means of nuclear spectrum, provided that a sufficiently large relaxation time
magnetic resonance (NMR) spectrometry and by principal (20 seconds) between pulses is programmed (Beck et al.
component analysis using gas chromatography - mass 1998). CMR data of the two tars, run on a Bruker Model
spectrometry (GC-MS). AC300 NMR spectrometer equipped with a Bruker Q and P
probe, are given in Table 1. Comparing the Meilerteer to the
Central European pine wood Doppeltopfteer, there is no significant change in the degree of
tar technology decarboxylation, but a doubling of the resin esters at the
expense of the resin acids. In addition, the percentage of
Early tars were made in prehistoric times by two distinct unsaturated, i.e. olefinic and aromatic carbon atoms has
met hods. In the older, pre-ceramic method, a pile of wood is substantially increased. Both of these changes indicate a
covered with eai.th and ignited. By controlling the access of air, higher operating temperature for the Doppeltopfteer.
some of the wood can be made to bum and provide heat, while
the remainder is charred, producing both charcoal and tar. As
a means of
producing chai.-coal, this technology was essential Carbon Meiler Doppeltopf
to metallurgy, but it was practiced long before the metal ages
for the manufacture of tar and pitch, which were needed for
COOH 2.14% 1.88%
ha fting, caulking and lubrication (Todtenhaupt & Kurzweil
991, 1993). Since the English term for this system is COO Me 0.32% 0.63%
charcoal pile' which wrongly suggests only one purpose, and
the wrong one at that, we will use
the German 'Meile1teer' for sp2(unsaturated) 31. 15 % 38.42 %
tar m ade
by this method (Fig. 1). T he Meilerteer used in this
Work was made at the Museumsdorf Dilppel.
sp3(saturated) 65.49 % 57.21 %
In the second method a ceramic pot is buried at the bottom
of a con ical pit, a second pot with a perforated bottom is
Placed on top of it, charged with wood chips, and covered Table 1. Carbon NMR data.
Proceedi
ngs of the 31" International Symposium 011 Archaeometry, le rem, E. & T. Biro, K. (eds.), Archaeopress -Archaeolingua, BAR - Cenfl: Eu1: Sei: I.
Beck, C. W., Stout, E. C., Lee, K., Todte11/w11pt, D. & Kurzwei/, A.
Ancient pine tar technology: corellations between manufacturing methods and composition
a w
th t rs
e re
e Isome
;ed to
;
19 'b n identified in the pyrolysate of pure cellulos
e Robinson et al. 1987; Reunanen et al. 1990; Beck & Borromeo
the theodynamically more stable a-pinene
az om e
th an
Mod
e.
el
The
5995
GCIMS
equipped
instrument used was a Hewlett
with a Model 9153C work respectively. The total monote rpene content
and li
decreases al
nene, so e e
torova
1997). They persist in archaeo logical tars: Lange 1990), in Etruscan wine flavored with pine resin (Hostetter et
pack"'
d . most found 4-methylguiacol and an unidentified ethyl
al. 1993), in deep-sea sediments (Simoneit 1977), in atmos
d a 1 5 m x 0.25 mm capillary column w ith a two-thirds from 7.1% to 2.6%. The appearance (P;3)
an of the o pheric aerosols (Simoneit & Mazurek 1982), and in amber
station terpene degradation product l-ethyl-4-meth
m no. (l acol in the tar of a Medieval shipwreck. The Doppel
al'Y phase of RSL-150 poly(dimethylsiloxane). Injection
station Y1eyelohexan (Botta et al. 1982; Mills et al. 1984/85; Czechowski et al.
re was 250C, mass analyzer 180C, source suggests the first step of the pyrolytic destr uc gu eer contains five guaiacols, at a total concentration of
temperatu tion of t e toP 1996). However, fichtelite, which requires reducing conditions
po rt monoterpenes that would be expected at higher
he
n1. This three-f old increase
is in complete accord with
l 72C. The initial oven temperature of 50C was
l 9 -10.
ratur e . t em peratures
tem pe .
.
Surpnsmg, however, 1s the presence of two terpin 1 uthenberg et al. (1996), who have shown that the for its formation, has so far been found only in fossil conifer
d at th e rate of 5C/minute to 250 C and then held ' ults of R
rampe . sm b .
sis produc ts increases with the wood (Bromeis 1841), in Tasmanian retinite (Douglas &
mentat10n was by electron impact at 70 eV. the Mei1e1teer and, at insigni ficantly lower conce nt ncentrations
of lignin pyroly
consta
nt. F rag
the Doppeltopfteer, because these te1tiary alcohols ar
rati ons ' in : rature. A s all amoun t (0.3%) of an add Grantham 1974), and in peat pyrolysates (Kumari 1990). All
icatio n of the components relied on the 6th edition of the . e so p rone operating tempe .
Identif m the Doppelto pfteer, tentatively these earlier studies have been useful in establishing the
l databas of McLafferty & Stauffer (1997), on to dehydration that they do not yield molecular 10ns . 'tlonal phe
nolic compound
pectr m therr
mass s .
mass spectra. Apart from thi s unexplained anomaly th as l,2-dimethoxy-4-(1-propenyl)-benzene, is not
structures of these hydrocarbons. Since there is no agreement
the lit erature 1t
d, a d on our own collection of mass e amounts :dentified
data in and structures of surviving monoterpenes clearly ind
on the order of elution of the two compounds, the distinction
wn in Fig. 4.
-pinene
0.4%
'
1-ethyl-4-methyl-cyclohexane
2.3% 0.7%
arene
6A. limonene
0.8% 0.3%
Xi erpinolene II
II
Q, 04% 0.3%
4-terpineol
0.4%
180
hod s and composition
ween man ufa ctur ing met
nology: corellations bet
Ancient pine t ar tech
Beck' C W" St out' E C" Lee, K., Todtenhaupt, D. & Kurzweil, A.
ts
Acknowledgemen
between the epimerc .
i 18- and 19-norab1eta-8,l
Summary
l , l3-trienes is nothe pathway by which both palustrate an d ne ledge financial
. ?abieta te ar s gratefully acknow
based on Simoneit's (1977) bserval!on isomenzd to abietate (Takeda et al. 1968) e nces between The American author
that the ratio of the
re are distinct differe National Science
ch th e n,
' h data show that the by the United States
ions m/z 256 and m/z =
21 is .
lower for the 18-nor-isomer. gether with the abietate already present, : to.
show that supp01t of this work
be o f. e NMR spectra alone ogy) and of the
=
.
s th e s
R-96-00254, Anthropol
'
As shown in Fig. 5' the SIX Id ent1fied . o urce b y the two methods. Foundation (Grant SB
hydrocarbons and three of the substantially increased amounts of deh ma de ile1teer by a College. This
. droabie t tars uished fro m the Me oratory (ARL) by Vas sar
more that could be only c teg onc ally . Y at p fte er can be disting Amber Research Lab
(from 19.3 to 28.3 %). As we have sho wn earlier (B lto their
classified (one of them eek et e e n of resin acids to
may actually be an este1) together compnse . al p pp
o in cre ase in the conversio rep ort is ARL Pub lica
tion No. 137.
9.6% of the 1999), methyl abietate is the only resi'n aci d that su l d rat ion.
rvive s a iwo fo increase in unsatu
Doppeltopfteer . Only three of t he . t esters and by a nearly 25%
(mcluding the suspected hi gh temperatures. The appearance of the deg h
t yl of the chane that
s
ester) are found in the Me1']erte er, with a
rad ation more detailed pictre References
total concentral!on of
products 17-nordehydroabi etate and abiet ' a-8 11 data give a ile1teer.
r relative to the Me
13,15.
-MS
.
2.0%. The concentration of the uIl!mate degradation . . ze the Doppeltopftee
tetraenoate m the Doppeltopfteer is consonant with a . teri t and the
ff 9.4% 40%
1999, Central Europ
Biomolecules 2, 281
ean pine tar
-293.
tec hno
:""
. .
gel is, F., Nicoletti, R. & Tricar
mcreasmg temperature. Botta, M., De An s of amber,
alysis in the gen esi
methyl palustrate The role of acid cat
f Resin acids (Figs. 6a and 6b) P hytochemistry 21,
381-384.
n Liebigs 37,
1, Ub er den Fichtelit, Annale
1.0% Bro me is, C., 184
The identifci ation of 15 resm .
acids,
28.4%
304-306.
many of which are hanbinski, M. ,
oneit, B. R. T., Sac
closely related isomers o r even stereo
m ers, rests on their
methyl levopimarate Czechowski, F., Sim chemical
iec, S., 199 6, Phy sico
reIat1ve retention times as well as t .
heir m ass spectra. Both are
methyl dehydro-abietate Chojcan, J. & Wolow osits in
s fro m dep
.
. zation of am ber
based on the literature cited above . .
m the secl!on of
1.8% structural characteri
0.8% -83 4 .
. em. 11, 811
deterpenoid hydrocarbons o ublished & Poland, Appl. Geoch
work by Zinke! et al. gerprint gas
I, Gra nth am, P. J., 1 974, Fin
(1971) and by Anderson mans (1991), and on our own Do uglas, A. G. &
ive bitumens,
e nat
u-
methyl isopimara-8, 15-dien-18-oate
analysis of som
mass spectral files. chromatography in the organic
, anc es on
substances Adv
There are dramatic changes in the .
kinds and amounts of asphalts and related
. 10.5% 7.7% Paris, 261-276.
diterpene resin acids in the t tars. The total decreases from Geochemistry 1973,
ier, D., 199la,
I., Bremer, J. & Me
.
85.6% in the Meilerteer to 7 Faix, 0., Fortmann,
% m the Doppeltopfteer. That
9.
temperatures can be derived from the changes Werkstoff 49, 213-21
in individual D., 199lb,
Bremer, J. & Me ier,
resin acids. The ring cleav g pro ducts
' Faix, 0., For tma nn, I.,
, which increase both collection of
pro ducts of wood. A
: methyl sandaracopimarate
in number and amount in t Thermal degradation
opp ltopfteer, are known to be polysaccharide
) mass spectra of
de1ived from levopimarates y bas1c yrolysis electron-impact (El
(Takeda et al. 299-304.
lz Ro h-Werkstoff 49,
\ p 0.5%
1969). The intem1ediate role o f levop1m ' 0.3% 3 oxy-abietic acid 4.0% derived products, Ho
3, A Bronze
anc ac1d is supported
--- - -------------------------------
182
Beck, c. II', Stout, E. C., Lee, K., Todtenhaupt, D. & Kurz.weil, A.
re pr ae
c0 l Geology 1 5 ,
,
6 3-69. posium o
coo n d
at AND BRONZES IN SAND MOULDS
J{., Czepel, H. & Gratz!, J., 1965, Gaschromato Urana, IL. Archaeornet
. ry
.11atzl, 111 sche Studi. en zur Pyrolyse des Lignins, Holz Roh S1mone1t, B. R. T., 1977, Diterpenoi ,
raP . d com pou ECCLESTON, M. & OTTAWAY, B. S.
.
em e rekonstruierte mittelalterl ich M et ode Simoneit, B. R. T. & Mazurek,
M. A., 1982 o
ga c m
463-47
6. University of S heffield, Northgate House, West Street, Sheffield SJ 4ET, UK.
. f"lolzte ergewmnung, Experimente/le Arclwolog1 e 1111 the troposhphere - II. Natural atter of
. . . backgrou
' nd
of bio' .
: ,seu ms doif Diippel, Oldenburg, Staatliches Museum lipid matter m aerosols over
. .
the rural Wi
estern u .
gelllc
"!' rkunde und Vorgeschichte in Oldenburg. S !ates, A tmosp h enc Envuvn lllted
Natu
fll ment 16, 2139
_ 2159 r[he a bsence of mo ulds in the archaeolo gical record of the earliest metal using periods in Euope has. led to the suggestion
vie1'
.1 A &Todtenhaupt, D ., 1991, Technologie der
Truced , H. 1 chulle , VY_ H. & Lawrence, .
. R. V., 19
6 tmOulds
made of sand were used. This paper describes some of the experimental work earned out m castmg copper and
,
.
J{Llrz )ztee rg ewmnung, Acta Praeh is t. A rclweol. 23, 63-9!. t enna 1Somenzat1on of abietic
acid J 0rg
8 The .
. Ch ern. 3 es using sand moulds with varying amounts of clay. The resultant coo/mg curves of the metal as well as of the mauld'mg
'
s of levopimari
Rosebr k
.
c acid sa
00 '
KEYWORDS: EXPERIMENTAL, CASTING, COPPER, BRONZE, SAND MOULDS.
eity, F. W. & Stauffer, D. B., 1997, The \Vi/ey!NBS J. 01g. Chem. 34, 1459-1460. lt s,
McL' i stiy ofMass Spectral Data, 6th edn. Wiley, New York . Todtenhaupt, D. & Kurz weil A.
1991 HoI ztee
cter Introduction China clay was added. It was important to ensure that the
. 1\egs., rm
'
,
White, R. & ?ough , L. J., 1984/85, The chemical Tnbolog1e, Acta Praehist. Archaeol.
. .
23, 47_53. moisture content of the sand/clay mixture was constant for
r.'11ll5 tion of Baltic amber, Chem. Geo/. 47, 15-3 9. Todtenhaupt, D. & Kurz weil, A 1993 .
JJ1po si Wood tar m tn. 'A brief survey of the archaeological literature reveals that each cast. Similarly, the density of the compacted sand had to
c b ology
.o va , I., 1997, Chemicalli y li nkin g past and present: Tribal. Schmierungstech. 40, 108-110.
'
st 10 the r e is a dearth of casting moulds amongst the archaeolo be uniform for each mould. This was achieved by testing the
Zink D . F., Za, L.' C. & Wesolows
'
. . ki, M. F., 197 gical material from the Chalcolithic and earliest Early compacted sand with a Ridsdale-Dietert ("B" Scale) Green
co . 1
issertauon, Ur uversi ty of Amsterdam. 1te1pene Resm Acids. A Compilation
ofln'rared, .., z
'J' Bron e Age from areas such as the Levant, the Aegean and Hardness Tester to maintain a reading of fifty in every mould.
'
p .
n N., Evershed, R. P., Hi ggs W J, NucIear MMagnet1c R esonance, Ultra
mass,
J ian, K. &
nsO ' ' . . em viol et Spectra
and Egypt. This was also found to be the case for most of These readings are empirical and bear no direct relationship to
!{obi B [inton, G., 1987, Proof of a Pine Wood origin for Pitch Gas Chromatograpluc .
Retention Data (of
the Methyl Europe. In all these areas metallurgical activities are any definite pressure per unit area, though they are valuable
01 Tud or (Mary Rose) and Et ruscan Shipwrecks: Esters), U.S. Dep artment of Agricult
ure, Forest Ser i
v ce, evidenced only by the finished metal artefacts themselves for comparative purposes.
ppli ation of Analy tical Organic Chemistry to Forest Products Laborator y, Madison,
c W I. and occasionally by some ores, slags and fragments of Temperature measurements were taken using a five channel
logy, Analyst 112, 637-644.
j\rchaeo . crucibles. There are moulds of stone, clay and metal, but electronic thermometer. The thermocouples were placed at 2,
n. M., Ekman, R. & Heino nen, M., 1990, Alterauon
they are mostly
u1at1e . . associated to material from subsequent 5, 10 and 20 mm from the metal/mould interface. Readings
!{et of pine tar m a manne environm ent, Ho liforscl11111g 44, from the three dimensional centre of the metal itself were
peciods. This negative evidence in the earliest metal-using
211-218 . periods has led to the hypothesis that sand moulds preceded taken using a "Pyrotenax" thermocouple. The first tempera
other casting media during the formative metalworking ture recorded was, for safety reasons, taken at some time
periods (Ottaway & Seibel 1998). between 10 and 20 seconds after the metal was poured.
The aim of the present study was to conduct a small series All following temperatures were taken at 30 seconds intervals
of casting experiments, using moulding material easily for one hour. Each experiment was duplicated.
available in prehistory, namely sand and clay, to observe the The metal, contained in a new plumbago crucible, was
cooling behaviour of metal and moulding material and to see melted in a modem gas furnace at temperatures reaching
what effect, if any, increasing ratios of clay in sand had on the 1 300C. In all cases a 1500 g charge of metal was used and
final bronze product. for the bronzes the tin was added immediately prior to casting.
Altogether six casts were made: two 10% tin bronzes were
Experimental procedure cast in sand with 20% clay, two 10% tin bronzes were cast in
sand with 30% clay and two pure copper objects were cast in
The experiments aimed to investigate the effects differing sand naturally containing 1 2% clay. To avoid contamination
compositions of sand have on the cooling rate and the pure copper was cast first.
micr ostructure of copper and bronze. This was achieved by Samples of the metal were cut from each cast to analy se
varying the level of clay present in the sand, from the natural the metal by inductively coupled plasma spectroscopy (ICP)
12% by weight contained in sand to 20% and 30% of clay. and in order to study the effect different cooling rates had on
For ease of comparison the methodological approach the microstructure of the cast metal. For the latter, all
a dopted was based on that of Ottaway & Seibel (1998). The samples were mounted, ground and polished and then etched
pattern for the cast was made of pine and measured with alcoholic ferric chloride in preparation for metallo
l.5x4x10 cm. These dimensions were similar to those of the graphic analysis. Vicker's micro-hardness tests were also
fa
l nged axe found with the Austro-Italian 'Man in the Ice' carried out.
(Spindler 1993 ), who has been dated to the second half of the
fourth millennium cal BC, i.e. to the late Neolithic of that General background to casting in sand
region. Modem foundry two-piece moulding b oxes with
in ternal dimensions of 15.2xl5.2xl0 cm were used in our M ould
experiment to contain the sand. These are the small moulding
boxes found by Ottaway and Seibel to be best suited for the Moulds for casting consist of a particulm refractory
si ze of cast. material, such as sand, that is bonded together to hold the
The natural "Mansfield" sand used for the moulds had a intended shape during pouring. The refractory material should
grain size of 119 and contained a nominal 1 2% clay. To be thermally stable, chemically umeactive and be free of
in crease this percentage of clay, dried and crushed potters volatiles that produce gas upon heating. The porosity of a
184
Proceedings of the 31" Jntemarional Symposium on Archaeomell)>, Jerem, . & T. Biro, K. (eds.), Archaeopress -Archaeo/ingua, BAR- Cenll: Eiu: Set: I.
Eccleston, M. & Ottaway, B. S. Experimental casting of copper and bronzes in sand moulds
mould is impor tant as it controls its permeability. A mould have relatively poor thermal conductivit of copper at 1083C, while the 3. Metallography
y the te I w the melting point
stbeO
with good penneability allows gases produced during pouring distn"but10n, ' mperat 1y. The
and thus the expansion, is not u ure JU ifies. The curve then falls off steep
nifonn metal solid bronzes show large
to escape through the mould (Piwonka 1988). If a mould is ure with a complex phase The micrographs of the as-cast
p on the other hand, an alloy
too porous, metal may penetrate the sand and cause flashing. Heat transfer and transformation zones bronz es over a range of tempera tures and this 1s columnar grains and long dendritic structures, consistent with
di agra m. Solidify
On the other hand, if permeability is too low, the ability for ature. slowly cooled tin bronzes. Those cast in moulds containing
t d by the steady fall off in temper
the gas to escape is decreased. Heat is transfen-ed from the molten meta reflec e 1 ernce m th e cooli ng curves 20% clay have more eutectic phases than the bronzes cast in
l to the .
adJacen It w as
expected to se some d"ff . .
moulding material. O'Meara et al (1988) t differen t rat10s of sand moulds containing 30% clay. Microscopic examination of the
state that h th casts canied out m moulds with
Sand and clay moulds t e ofd e pure copper samples showed that the metal formed was a
amount of heat transfen-ed 1s a funct10 n of ever as can be seen in Fig. 1, the differences are
.
.
-W--20%clay
&sand
at5mm
__.,_30%clay
-Copper &sand
in at5mm
sand
-20%clay
_.,_.Bronze &sand
in'.20% at10mm
clay&
sand
-+--30%clay
&sand
--it-Bronze
at10mm
in30%
clay&
sand
-20%clay
&sand
at20mm
o .!:.....,.....,.-.--..-.-..--.-r-r--.-r-r---r-ir-r..,--r-.-.-.--.--r.-r--i -30%clay
\:)"'
0.0 -t=ir-r-.-....-.-,.--r-"T-r-T-,.-ir-r--r-.-...-r-T-,.--r-.,-.--.-.---r--.-...----___;
&sand
at20mm
Time (minutes)
Time (minutes)
for
Fig. 2. Cooling curves of moulding material at 2 mm, 5 mm, 10 nun and 20 mm distance from the metal/mould interface
Fig. 1. Cooling cw11es of copper in sand moulds and bronze in sand/clay moulds. bronze cast in moulds with 20 and 30% clay.
186 187
Experimental casting of copper and bronzes in sand moulds
Ecc/esto11, M. & Ottaway, B. S.
study Acknowledgements
periods, outlined in an earlier pilot
Pure Copper Pure Copper 10% Tin Bronze 10 % Tin B . B
E A
Mansfield Sand
ronze
.ot!iCSt & eibel 1998).
Mansfield Sand S y for help in the
20% Clay & S and 30% Clay "" a wa Y nce for the The authors would like to thank Alan Harve
& Sand (ort su gg ested that this could be taken as evide the Department of
Cu Phase Cu/CuO Eutectic It wfas petiod s, which , after use, would foundry and Dr. John Whiteman, both of
ulds in these
0 sand mo Engineering Materials of the University
of Sheffield, for
e rate to ' dust in the wind ' .
As Cast 51.7 3.4 66.0 3.9 u a11Y disinteg due to Dr. David
106.4 7.9 95.910.3 cted to test if advice and support. Further thanks are
v mall series of expe1iments was condu tment of Archae
[HV2s] s
in sand mould s would affect the Dungworth, Teaching Fellow in the Depar
lHV2s] component s
fHV100] sing clay ology and Prehistory, for help with the practic
al work of this
norea
fHV100] lex behav iour
n t bronzes. The result s reflec t the comp
50% Cold Worked 112.04.3 121.6 7.1 1
199.813.5 180.92 2.7 u lt a sand/c lay mould ing mater ial project.
n cooling and of
rnet al upo
[HV2s] lHV2s] fHV100] fHV100] Po
n heating. It was obse rve . that bron ze cools m
gina lly
those Bibliography
in s and moulds contammg 30% clay than . m
slo wer needs to be earne d out to
more work
ning 20% clay and -Heinemann, Oxford.
3. Hardness values for copper and b.mnzes cast zn contai Campbell, J., 1991, Castings, Butte1w01th
.
Fig.
to elucid ate the reaso ns for this
these results and
sand and sand/clay moulds.
ed axes of the north
confirJll and metal, Kienlin, T. & Ottaway, B. S., 1998, Flang
ur. Compared to moulds of clay, stone ilities of use
complex ehaviour. The bronze cools marginally slower in the to rather brittle metal ' by the pre beh avio and of sand with alpine region: an assessment of the possib
sence of copp sand with natura lly occur ring clay
moulds wtth 30% clay than in those with 20% clay. At 2 and .
eutectic phases, respectively.
er oxides
or moulds of cooling wear analysis on metal artefacts, in Morda
nt, C., Pemot,
of clay are the slowe st
ficially increased levels
5 mm away from the metal' the moulds with 3001
'lo h ave a art i M. & Rychn er, V. (eds.), L'Atelier du bronzi
er en Europe
. . The results of the hardness tests seem to show
higher te perature, whereas at 10 mm from the metal the that the me
tal moulds. II. Colloque International 'Bronze '96',
Neuchatel et
m the as-cast state produced here is tic
harder an e leads to an increased presence of eutec
moulds with 20% clay have the higher temperature. 20 mm stan dard The slow cooling Dijon, 271-286.
hardness values published by Scott an increa se in the coppe r oxide
(1991 8 e s and to
This w as als phases in the bronz The properties
for bronze, in
al and preven t disasters simulation of historical casting processes
time they could act as a bindin g materi
smiths , being clearly very International Symposium held in Innsbruck.
-Copper
when the moulds crack. Prehis toric
smann, Munich.
the numer ous pure coppe r and Spindler, K., 1993, Der Mann im Eis, Bertel
experienced in casting, as
In
structure of ancient
show, would Scott, D., 1991, Metallography and Micro
sand
bronze artefacts in the archae ologic al record
and historic metals, The Ghetty Conse
rvation Institute,
mould
could be
have welcomed, and used, any additional benefit that
........-sronze
Los Angeles.
in20% gained from moulding material.
clay&
800.0
sand
iii'
:I
'iii
4i -W-Bronze
in 30%
CIJ
...
clay&
:I
600.0 sand
'IV
...
CIJ
Q. -+-Clay
E
CIJ
I-
400.0
-&-Metal
0.0 +-;;-...,-r---
- -
:--....-
-: -- -.---- ' --
-
2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0
Time (minutes)
188 189
THE IMPORTANCE OF THE FIRING ATMOSPHERE
JN THE PRODUCTION OF COARSE CERAMICS WITH CALCITE
AND CHAMOTTE INCLUSIONS1
1 CNR, Research Institute for Ceramics Technology, Via Granarolo 64, 48018 Faenza, Italy,
zArchaeological Department, University of Bologna, Piazza S. Giovanni in Monte 2, 40124 Bologna, Italy
The firing of coarse ceramics, which constitute about 70% of the ceramic finds in Roman archaeological excavations in
1s to have een mainly realied in pit or pile-kiln. In the present study, some experiients hav been carried
North ern Italy, see1 . .
111, in w hich a series of specunens were fired m a pit-kiln. The specimens were composed of a body with mcluswns of ground
a /oit e or cham otte. The
characteristics
aim
of
was
the
to verify
different
the behaviour
mixtures. To
of the
evaluate
material
the role of
during
the
firing
firing
as well
atmosphere,
as
the
the microstntctural
samples were also
and
fired
mech ani cal
in an electrical laboratmy kiln, with an oxidizing atmosphere. The firing atmosphere does not influence the characteristics of
samples containing inclusions of chamotte, while different results are obtained with calcite inclusions. In particulm; pit-firing
allows temperatures to rise as high as 800C without promoting calcite decomposition and subsequent decohesion of the
artefacts as in oxidising environment.
I .
This research project was carried out under the auspices and with the financial support of the "Cultural Heritage" Finalized Project of the CNR, Italy.
Proceedings of the 31" /11ternatio11al Symposium on Archaeometry, Jerem, E. & T. Biro, K. (eds.). Archaeopress-Archaeolingua, BAR- Cenfl: Ew: So: I.
otte inclusions
e ceramics with calcite and cham
sphere in the production of coars
Fabbri, B., Gualtieri, S. & Santoro, S. The importance offiring atmo
phases S . fi . It In
vol
.
came d o ome pit- mng experiments were also
the preparation of clay bodies containing chamotte ve s 750 and 850C) with a
ut in w hi . . . 1 ' Carb onate from test pieces of that different maximum temperatures (650,
sh ape an h a senes of test pieces of predetennined and s1.Iicate sand mclus1ons, as well as of C ay . e inclusions were prepared 20 hours was established,
d siz e, ma COntaining charnott slow cycle.A total firing time of
. . . . all the bod es, fired n a pit.
cha m Otte . de from nux tures contammg calcite or . 110ns for the purposes of comparison and ide
me no Jll mate1il use as the base for about 8 hours, a maximum
mcl u . ntifymg sa e . calatrons of with a rising-temperature phase of
. sr ons, were fire . the s1ons were obtained from mter , and a falling-temperature
behav our d The arm was to observe the positive and negative effects of adding the temper. 'fhe calcite mclu temperature phase lasting 1 hour
1
characteri
of the m
stics of
te
?
.
als
.
dunng firing, as well as the The chamotte was prepared by grinding the bod . sp atic c
alcite in the limestone
.
formations of the Romagna-
phase lasting about 11 hours.
the rrushed products. 1 Y Withou t
. penrunes.
C learly' this temper, fired under different conditions a so m
ord er to arches A to a maximum
was n0 t stnctly .
ascertain the effect, if any' of this techno1ogr. cal v rials were then ground down
archaeolo speaking a n "experimental All the mate Raw material and product chara
cteri sation methods
d ata be hase
and not
mg sought was m fact of a general
character bo dY m different amounts (up to 30%) and gram . . & Bolena) in
a ratio of approximately 9: 1.
ranging from 2000 mm to 63 mm.
quantity
. Above all, ined were then mixed with the
. .
the type of fuel rs an extremely The bodies thus obta rials the mineralogical
.
coarse m . e was carried out
clus10 ns, with the objective of ascertaining 1 and poros1metric analyses.
standpoint through minera1og1ca Firing in a semi-reducing atmospher is given by: R= 3PD/2LS2,
m in diameter and about bending strength (R) in kg/cm2
FIRING
SEMIREDUCING ATMOSPHERE (PIT-FIRING) not m ore than an hour.We next proceeded to perform the final
operations on the pit, spreading alternating layers of earth and
osition of the clay used
Chemical and mineralogical comp
OXIDISING ATMOSPHERE (650C - 750C - 850oC) Wet straw on top of the embers to help maintain the heat
as base body is reported in Table
1. It turns out to be an iron
beneath the covering for a long time.The wet straw, burning
carbonate minerals.
rich illitic-kaolinitic clay without
_I I
slowly, creates small channels that communicate with the
I I
distr ibuti on curve s obtained for the two
outside, permitting release of the smoke during the final stage The grain size
tly different.For calcite the
- l-E-S_J
types of inclusions are only sligh
0
=R=E=DB 1D
of combustion and the start of the cooling phase of the
C C
HA;: ;; R AC T;,
E ;,
R l
SA TI=
ON:-=
O F FI
approximately, whereas the
-b-_!_ .
mean diameter (<p50) is 450 mm
( shr i nk age, endmg
c eram ic material.This phase generally lasts about 48 hours,
otte.
value of <p50 is 540 mm for cham
strength crystalline p has es, porosimetry, thermal shock resistance)
,
after w hich the "vault" is opened and the pieces are taken out.
,
:- - - tory mixer.
as a rotary powder container for small labora
Jar: cylindrical porcelain container used
Fig. 1. D escription of the comprehensive project.
193
192
Fabbri, B., Gualtieri, S. & Santoro, S. calcite and chamotte inclusions
The importance offiring atmosphere in the production of coarse ceramics with
+
0.25 K-feldspar the percentage of calcite and cham otte
+++++
17.31 Na20 ilar
Ah0J in the bo . 811n to
art ,. dies.
tr
K10 1.97 Plagioclase + The bending strength of the pit-fir
ed alternative to firing in an oxidising atmosphere, with an
Ti02 0.66 . . e1acts w hi
( wit h calcite. mclusions) or at least co as equally high level of productive complexity.
++ mparable g her ineralogy of bodies with calcite inclusions.
Fe20J 5.01 s 0.02 Illite
1us1on
me . that of any of those
s) with (chamotte Table 3. M
fired m the e
+
.. lectc .
0.05 P20s 0.08 Kaolinite In these last cond111ons, the bending strength gical requirements and function References
MnO f the b o kiln, f financial-technolo
terJllS 0
0.74 P.F. 5.75 Chlorite trace
highest with firing at 850oc, going from
4 50 k dies Was
g/c m2 (f+ .
ali of
the finished product.
MgO
finng at 6 5O and 750C) to approxima
in an abu .
sam e unifo rm red-orange colour, which was slightly The mineralogical compositions of the body ably utilise pit kilns. For example, the finds m metrica, Atti l a Giornata di Archeometria della Ceramica
on the with chamott also more profit
temperature. were practically the same before and after e "II contributo delle analisi archeometriche allo studio
er at the 850C ian and Northern Italy (Mazz occhin &
light . . .
firin g, except fo
r both central Tyrrhen
With pit fuing, on the other hand, there was an alternation th e destruct10n of kaolimte and c hlorite as a of Roma earthenwar e jars(2n
BC- A? delle ceramiche grezze e comuni", Bologna, 28/2/1997, a
result of finn . g. Ag ostini 1997)
of oxidising and reducing conditions, depending on the firing The composition of the calcite body on the other atmosph ere, contammg calcite cura di S. Santoro Bianchi e B. Fabbri, 136-142.
. . ' han d 'Varied Isl c.) fired in a reducmg
stage and the position of each artefact in the pit with respect to greatly dependmg on the nng c n itions(Table
? 3). In this
the fire and the wind. These heterogeneous conditions case, there were substantial vanallons in the
quan tity f
produced pieces that were themselves not uniform, as well as calcit, which decreased sharply between 650C
and disappeared completely at 850C. At this tempe
and 7500 1
each other. Colours ranged from pink
being different from rature we
to shad es of grey, and to black caused by carbon particles observed the presence of calcium oxide, traces of
hues . which ere
.
site d on the surface or infiltrated to a depth of 1-2 mm. In also ,.iound m pit-fired artefacts. However, pit firing
tended to
depo
in cros s-sec tion m o s t pieces exhibit a grey-black surface preserve much of the calcite.
fact ,
sur roun ding a centre that is generally lighter in colour. The presence of calcium oxide is likely the cause
layer of the
Tabl e 2 sho ws the results o f the physical-mechanical lack of cohesion of the specimens fired at 850C, because
it
tends to rehydrate and then to yield calcium carbonate, so
characterisation tests; for each of the pit fired bodies. Test
p romoting a bloating of the particles. The same phenomenon
pieces were selected so as to represent the entire range o f
above. should occur after firing at 750C, but no calcium oxide can
colours described
Ther e does not appear to be a substantial difference between be observed by XRD because it is not in a crystalline fonn.
obtained for the same body at different
total sh rinkage values
firing tem per ature s in the electric kiln, nor between the values Conclusions
ed for pit and electric-kiln firing. On the contrary, there
obtain
me difference between calcite-containing and chamotte We can conclude that, for the ancient potter, firing in a
is so
reducing atmosphere was the best method in terms of
containing bodies: the shrinkage is lower when chamotte
5.3% on average). managing the firing process and functionality of the
inclusions are present(4.7% against finished
As regards water absorption, it was impossible to measure product.
containing calcite inclusions and fired in an If the clay used had an abundant calcitic component, or if
the sample bars
This was due to the total calcite inclusions were added for other reasons(reduction of
electric kiln at 750C and 850C.
'disintegration' of these test pieces when immersed in boiling deformation during drying, reduction of sensitivity to thennal
slowly(over a few months), when exposed to shocks) firing in a reducing atmosphere provided greater
water and , more
air. This means that, if a calcite-containing body is available, assurance of success because, unlike firing in an oxidising
oxid ant firing is dangerous when temperatures are not well atmosphere, it did not impose any special temperature control
too much. requirements. If the inclusion added was chamot te, the best
controlled and increase
There were significant differences in water absorption for results in terms of functionality of the finished produ ct we re
bodies: calcite-contained bodies display obtained with firing in a reducing atmosphere, or with firing
the two types of
in an oxidising atmosphere at 850C(a considerably high
Parameter Pit-firing Oxidant firing temperature, more difficult to control and more costly to
Body
650 c 150c s50c attain). Overall, firing in a reducing atmosphere theref re
Shrinkage (%) 5.3 5.4 5.2 5.3 offered a better guarantee of good results, presumably with
(
vu+cA Water absorption %) 13.7 13.7 (*) (*) considerable energy savings in the case of firing artefac ts
' 79
Bending strength (kg/cm ) 51 48 67 containing inclusions.
4.9 4.5 r
From the archaeological point of view, it follows that fo
Shrinkage(%) 4.7 4.5
vu+cH Water absorption(%) 17.6 16.4 17.6 17.3
y th ose
' the "weaker" productive structures, that is to sa .
Bending strength (kg/cm ) 59 41 43 65
(*)N ot evaluable.
organised for production at the lowest possible co st, for the
lC
own consumption or for local sale having a limited econoJJl
the mo st
Table 2. Total Shrinkage, Porosity (water absorption) value, filing in a reducing atmosphere was always
i. n
and Bending Strength. advantageous choice, offering better performan ce both
194 195
BIOHISTORY F ARM AT SZARVASGEDE (HUN G ARY):
A C ASE FOR BIODIVERSITY CONSERVATION
GYULAI, F.
to pollen analysis, archaeobotany ( concemed with the evaluation of seed and fruit remains) concentrates chiefly
In c ontrast
that were deposited as a result of human activity. Plant cultivation in the Carpathian Basin started 8000 years ago.
on fin ds
r in tention is to reconstruct in detail, this cultural heritage. Our aim is to establish a bio-history reserve where we can collect
(Ju
and produce, in a pure biological manne1; the cereals, fruits and grapes species of the Hungarian Medieval periods in the
Carpathian Basin, saving them from extinction. These species are the result of many centuries of adaptation to the specific
climate of the Carpathian Basin. The production of these ancient species together with the supply of additional information on
them would mean the sun1ival of this cultural heritage. One way to know more about the value of ancient cereal is chemical
an alysis. The analysis of cereal and food remains poses a great challenge. This work requires simultaneous familiarity with
bot an y and chemistry. Nevertheless, results have contributed significant information on the food consumption and subsistence
culture of prehistoric people. We intend to reconstruct this ancient plant cultivation using the latest scientific data. In our
Biohistory Farm of Szarvasgede (Hungary) beyond saving and cultivating cereals that are threatened with extinction we would
also c onduct experimental archaeology. This is a multidisciplinary research, the synthesis of cultural anthropology, archaeo
and ethnobotany. These all together would help preserve tradition se111e the environment and education.
Description of the project Our project started in 1995. We plan to establish a 2 hectares
model habitat on Szarvasgede (N6grad County) 70 km from
Our aim is to establish a biohistory farm where we collect Budapest (Fig. 1). The "Biohistory Farm" is bordered by the
and produce old cereals and fruit species saving them from Aranyos creek. The lot is located in the village. The climate is
extinction. These species are the result of many centuries of favorable; the distribution of annual precipitation is good. The
adaptation to the specific climate of the Carpathian Basin. land is rich in nutrients. This is a virgin territory. The
Proceedin
gs of the 31" lntemational Symposium on Archaeometry, lerem, E. & T Biro, K. (eds.), Archaeopress -Archaeolingua, BAR - Centi: Ew: Sa 1.
Gyulai, F. Biohistory farm 011 Szarvasgede (Hungmy): a case for biodiversity conservation
countryside is free from industrial activity. It is an archaeotype knowledge, economic strategies, subsistence olo gical excavations and living plants from the model
. pa tterns
and arch ea
of symbiosis between man and nature where the natural and environmental changes.
habit t.
a
cultural landscape (agriculture, settlement) composes a holistic In the Carpathian-basin plant domestication beg m
A ong
cereals we wish to produce chaffed wheat ( einkorn,
. . a n ca 800Q
unit. This would be a model garden and exhibition. The results years ago, fnn t growmg ca. y ears ago (F
2000 ig. ). Jll!ller, spelt) as most impmtant. We collected the grains from
of our research could be used in the training of professionals mtent10n 1s to reconstruet, m
detai1, thi s cult ural Our 2 rational archaeological farms and from genebanks
. henta
ge w,
e
eadY opep1ivate
and in education. suppose, based on archaeological and botanical studi individuals in Hungary and from abroad. We
and fr
om
. es, that
a t
Similar to projects designed to save ancient domesticated the time of the Iandtaking Hungana ns, they had arri d c e in the Biohistory Faim since five yeais the oldest
Ved to the
u
. o
animals we shall care for and maintain ancient plants (region Carpathian-basin . carrying wit'h th em the knowle
he t: the
a einkorn on half of hectare and we make experiment
. dge o f cere
al'
specific cereals and frnit) threatened with extinction. We grape and fnnt . cu1
11vat1on. Documents fro
m the e 10 prepa
re it for food (Fig. 3).
. ,, . . arly
intend to establish this habitat in harmony with agriculture Medieval Arpa? penod arely mention fruit g rowing. Einkom (Triticum monococcum), the bread wheat of
and the environment. We wish to join the project, entitled Jn the medieval penod the hungarian wheat an d ancie nt
cultures, could be the choice of present day reform
. fru'u
"Environment and ecologically friendly animal husbandry products were famous and m demand on the mark diets (he alth food diets) due to its high raw protein content
ets of
and agriculture for private and state owned farms on nature Europe. The new flavour and good character of these "re . and rich microelement constitution. Its favourable qualities,
gion-
preserves and their protective areas" under the aegis of the specific" products made them beneficial for hum hardyness, weedsurpassing ability, great resistance against
. an
Nature Preservation Office of the Ministry of Environment consumption. T e archaeolog1cal recrd however conta
pests make einkorn a prime candidate
for chemical free
ins
and Land Development (Tardy 1994). Our project connects d o mesticated fnnt seeds and plant remains. The floral remain cu ltivation.
s
with the Biodiversity Protection program. of medieval sites help us t learn about the species ot the We wish to produce in a pure biological manner frnits
Our main tool is archaeobotany. In contrast to pollen p eiiod. The une arthed botamcal matter is the exact mirror of (pples, pears, peaches, apricots, plums, cherries, sour
analysis, archaeobotany (concerned with the evaluation of seed past production and plant knowledge. The archaeological data cherries, walnuts) and grapes. They are all famous ancient
and fiuit remains) concentrates chiefly on finds that were can be used to cross-reference the existing historical and region-specific species that are pushed onto the periphery in
deposited as a result of human activity. This discipline deals iconograph y sources. Archaeobotanical interpretation is a the Carpathian region for different reasons. Our team with the
with the identification of plant remains and associated major element in the study of historical ecology. The study of h lp of the University of Horticulture and Food Industry,
e
artefacts, historical sources (iconography and written data) and plant remains can help us understand local subsistence Budapest, and the Pannon Agricultural University, Keszthely
those areas of the social sciences that concerns agriculture. Its p atterns, plant use, past environments and their utiliz ation. collec ted the old frnits and grapes species from rare refugee,
focus is the relationship between man and environment and The ancient cereals, fruit and grape species collected on the from their own native growing sites in the Carpathian-basin
human economic activities in the past. model lots provide the means for identification of other where they could still be found. We planted near by 200 piece
The major area of archaeobotany is the study of plant medieval plant remains. With the help of the Prism-Imageplus of trees and 100 piece of grapes in 1997 /1998 on Biohistory
domestication, cultivation and the development of agriculture. software and computer system we can conduct morphological Harm, Szarvasgede (Fig. 4).
Ecofact analysis can shed light on past human biological comparisons between plant remains (seeds and crops) from In our Biohistory Farm of Szarvasgede beyond saving and Fig. 3. The einkornfield at the Biohistory Farm.
cultivating cereals that are threatened with extinction we
NeoUthic Roman Period Ugratlon Uddl< Ages
Copper Age Bronze Age Iron Ag<
would also conduct experimental archaeology research. In Bocz (1988) analized the macro- and microelement
middle late Barbaricum Period
early middle late Pannonia early la
u u u u u u u u 0 Q Q Q this case we built in 1997 an early medieval ( Arpad Age) composition of ancient and of 18th century steppe plough
= = = = = = .. " "' <
8 8 8
.. pithouse founded on the motorway excavation: excavated by wheat crops. He compared the results with data gained from
"' " "' 8
...
"
'" " " '
Zoltan Beneze (Budapest Historical Museum), designed by recent plough-wheat analysis. He notes that with the advance of
illibor Sabjan (Enthnographic Museum, Szentendre) and improvements the nitrogen/phosphorns, nitrogen/potassium,
Miklos Takacs (Archaeological Institute of the Hungarian and nitrogen/magnesium ratio decreased in the grain while the
Academy of Sciences) (Fig. 5). nitrogen/calcium, nitrogen/manganese, nitrogen/zink and
nitrogen/copper ratio increased.
Nutritional value of
to------ --------------- The food remains from Hungary
the ancient cereals
Kiirber-Grohne (1989) pe1formed comparative growing We intend to reconstruct this ancient plant cultivation and
experiments with einkorn, emmer wheat, spelt, durnm wheat, food consumption using the latest scientific data. Through
plough wheat, double row bare barley, rye and oats in chemical analysis we will learn more about the value of
Stuttgart/Hohenheim and in Schwabis Alb. Their results on ancient cereals. The analysis of cereal and food remains poses
crop inner content values (ppth. weight, raw protein, a great challenge. This work requires simultaneous familiarity
Carbohydrates, raw fat, ash, raw fibre and dry matter) are with botany and chemistry.
Useful for quality comparisons. According to their test results An increasing number of archaeological excavations have
the average raw protein content of ancient chaffed cereals been accompanied by scientific analyses during the past few
(einkom, emmer wheat, spelt) is 50% higher than that of decades. As a result, prehistoric food remains have also been
ploug h-wheat. The raw protein content of plough-wheat is recovered. Remains of carbonized bread were found
- -
about the same as in barley, rye and oats. This is especially following the water-sieving of samples gathered on the floor
chara cteristic level of burnt Ottomany (Middle Bronze Age) culture houses
-
with einkorn and emmer wheat where one fifth
---- 20%) of dry matter in the crop is raw protein. Fatty content at the tell settlement of Turkeve-Terehalom (led by J. Tarnoki
18 about
---- - 30% higher in chaffed wheat. The highest is in oats. in 1990). The porosity of charred pieces, that is the size and
l'he c arbohydrate content is the highest in baie barley and rye. shape of holes, shows the typical strncture of leavened bread
It is important to note that raw protein and carbohydrate (Gyulai unpublished data).
content During study of the inside of another sherd from the
in the crop depends on the nitrogen content of the soil
(nitrogen Copper Age site of Zalaszentbalazs-Sz616 hegyi mez6 in
fertilizers increase it), on the climate (warm, dry
We a ther increase the value) and on the morphology of the crop Hungary (led by E. Banffy in 1992) using a stereo
few, common
(the raw
protein content in flat seeds is higher than that in microscope, a small brownish-blackish spot of typical burnt
roun d
Fig. 2. Characteristic finds of plants from Hungmy. seeds) (Brouwer 1972). food remains was discovered (Gyulai 1995). The shape, color
198 199
Gyulai, F. Biohist01y Jann on Szarvasgede (Hungmy): a case for biodiversity conservation
i
;;
::f 1
i :
Fig. 5. Early medieval pithouse reconstruction in Szarvasgede.
and thickness of this spot, on the other hand, is very similar to as a result of burning, their elongated shape may still be
analogous remains found at lake dwellings in Switzerland and recognized. Seed germs fell out of these seeds which may be
Southern Germany and to what was identified as simple flour indicative of cleaning since germs are often lost during the
soup (Wahren 1985). This may be an example of "seasoning" course of threshing and cleaning. Carbonized lumps of
of ceramic dishes to be used for cooking by sealing pores of ground meal were also found probably made from millet
these vessels with this sticky kind of fine flour soup. because they included clumps of two to ten millet seeds
Following this stage, the dish would not be washed in order to cemented together. It may be reasonably assumed that these
make sure that the substance cemented into the pores made it finds were deposited in the pit within a short period of time,
waterproof (Fig. 6). if not simultaneously. It is also possible that they were already
During the course of excavations at Keszthely-Fenekpuszta burnt during the cooking process or subsequently, together
in 1980 a 60 cm thick, ashy layer was found when a pit dated with other artifacts (Gyulai 1996).
to the Celtic Period was excavated (led by I. Erdelyi in 1980). When the residue was sorted under a stereo light
The archaeozoologist Istvan Takacs (Agricultural Museum, microscope, numerous animal remains were found as well.
Budapest) collected significant amounts of fish remains here Within this assemblage, the archaeozoologist I. Takacs
as well as some 1.5 liter's of black, ashy sediment. L. identified a number of bones from fish and small birds, which
Bartosiewicz archaeozoologist (Eotvos L6rand University, showed signs of roasting.
Budapest) carried out the identification of animal bones.
Ma croscopic as well as microscopic analyses took place in the
Archaeological Institute of the Hungarian Academy of
Sciences in addition to analytical studies (macro- and trace
ele ment, amino- and fatty acid analyses) at the Central
Laboratory of the Faculty of Animal Science at the University
of Agricu
ltural Sciences in Kaposvar (by J. Csap6). These
rnet hods all lead to the conclusion that the grayish-black ashy
layer, rich in
fish bones, were the remains of a "fish supper"
(Gyulai unpublished data).
Carbonized food remains recovered from a Tumulus (late
Bro nze
Age) culture refuse pit at the site of Balatonmagyar6d
idvegpu
szta (led by L. Horvath in 1987) are worthy of
discussio
n as well. A significant portion of archaeobotanical
finds o rigi
nated from legumes. All of these seeds represent
Vege tab
les popular during the Bronze Age including peas
(Pisum
sativum), ervil (Vicia ervilia) and chickling vetch
(lathyrus
sativus). Husked seeds from true millet (Panicum
Il1
l1a
[" c
e um)
are carbonized. Although these pieces puffed up Fig. 6. P rehistoric food remains on ceramic sherd.
200 201
Gyulai, F.
RS
p&OCESSING FISH WITH OBSIDIAN TOOLS: THE MICRO-WEA
The material obtained for study also contained a significant
References
number of carbonized remains ongmating from a
homogeneous but spongy material. They seem to be Bocz, E., 1 998, Buza:
osi fajtaval-telj'es . IOVINO, M. R.1
fragments of a piece of larger volume. None of the fractured . nun ose ge t ru,._
N
wit. h o 1d vanetas-complete
quality) U 1 ,, n
surfaces displayed holes, bubbles or cracks: the small, round agyar
gazdasag 53, 14. '
Me14 Viale Tunisi, 8, 96100 Siracusa, Italy, e-mail: miarosa@tin.it
pores were evenly distributed. These fine holes cannot be the Brouwer, W., 1972, Handbuc
h des speziel/enP
'jlan zenb
result of fermentation caused by Lactobacillus lactis. They ( ",
vve1zen, Roggen, Ge ste, Hafe aue s /,
r r Mais) B Ii
are rather indicative of the baking of an evenly kneaded, fine ayuiat, F., ' . er n
J)unng1
. the Mesolithic and Neolithic
periods, fishing was important. In an archaeological context, it is difficult to understand
,.
r mat
. .
These data, in addition to the instrumental analyses (the m the Ha/10t Basin, South
-West Hung ary
An t H istory ftsh Ju nctwnal. studies have addressed the micro-wear jirom processing fish with obsidian tools. In two well preserved
tae us 22
identification of Phytolits, macro and microelements, amino 145- 1 57 . '
, fsW . h.
1e o ltt ic
-Neolithic fishing sites (Vzzo's Cave, Trapani, Sicily and Franchthi's Cave, Argolide, Greece) we have conducted
and fatty acids) all indicated that remains of Bronze Age : "' ' s . Sparus aurata, Dtp1odus sargus, D entex dentex,
kesQb nz
strawberry shortcake were found. The dough was made of IJJerent k"in d oif
fonnatwn from tI1e d'f'F.
:
1
Consulting Engineers and Archaeologists-ARCHAEOME TRY R ESEARCH L ABORATORY- VIA G. MINGAZZINI, 7 - 00161 ROMA-ITALY.
1
This fish has very small scales in almost all body, only close to the corsaletto has bigger scales. Also these scales are ctenoid, but with very small ctenii.
:
margin of the scale (Casteel 1976). Even if all our scales are The cutting of fish generally produ ces a cle
ar
ctenoid they show differences in their shape and in the ctenii's alteration with noticable features. The edge is o ce
ften g tly
shape. Dentex dent ex shows a good development of cteni i, rounded, but more rounded edges were also o
bserv ed
like a comb, while the other scales have very small ctneii, Sometimes one can observe on the edge a slightly
s 0
0th
almost imperceptible (Fig. 2). polish never invasive. The striations are mostly
ough.
Another difference w e underline here: Dip/odus sargus, bottomed and sleeks, but also intermittent str iation
. could
Sparus aurata and Sc01phae11a scmfa have less resistant occur. They show a parallel and diagonal orientatio
n. S!ight
scales. Ephinephelus, Mugi/ cephalus and Dentex dentex h ave bands of attrition are visible near the edge but mostly o
f them
more resistant scales. are away from the edge (Fig. 3).
The cutting of Thunnus produces a different wear pat
te
?
c:J
MUGIL CEPHALUS The edge is rounded, a dull smo th (little corrugated)
DENTEX DENTEX pol l
occurs. The transformed surface is covered with a lot of
short
parallel intermittent and rough-bottomed striatio ns. Sleeks
are
scarcely attested. Some narrow bands of attrition are vis
a
ible
EPHINEPHELUS
SPARUS AURATA
away the edge, but in some cases patches of attrition are als
o
visible on the rounded edge. The wear is very invasi
ve
(1000 mm), and it looks like a band along the edge (Fig. 4).
DIPLODUS SARGUS SCORPAENA SCROFA
Scaling of fish is more difficult to detect. It is almost
. on
cati 20x 4x.
impossible to observe any diagnostic surface modification l o b s1'd'tan tool . Magnifi
s fish with an experimenta
Fig. 4. Cutting Thunnu
Fig. 2. Schematic representation of the difef rent typo logy scaling only a few fish or scaling fish which are not fresh .
of the scale's shape of the fishes object Scaling fish does not always develop a diagnostic micro wear
of our experimental activity. pattern. The used edges, usually, shift from slight to rounded.
The scaling of some fishes like Spaurus aurata and
Data Scorphaena scrofa did not give any peculiar wear. The tools
show only few short, narrow, deep and shallow rough
We attempted to observe the eventual divers characters in bottomed striations. The same pattern was observed scaling
the micro-wear traces formation result from scaling and 3 kg of Diplodus sargus.
butchering the above fishes. Slight edge rounding, few sleeks and rough bottomed
We employed different quantity of fish, beginning with few striations, narrow and wide, mostly deep, with perpendicular
hectograms up to six kilos. orientation were found on the experimental tool used to
The experimental tools were mostly flakes with a right edge scaling tree big Ephinephelus fishes. Some band of attrition
produced with Lipari raw material. we can observe (Fig. 5).
Twenty different edges were used to scale fishes for a total The scaling of Mugil cephalus gives a slight edge rounding,
account of 5 hours and 35 minutes. The minimum single use with few grouped sleeks and some little rough bottomed. The
time was about 4 minutes. The maximum was about 41 striations are mostly shallow and deep. Little luminescent
minutes. To cut fishes w e used eight edges for a total time of attrition we can detect on the used edge.
1 hour. The minimum single use time was about 4 minutes The scaling of Dentex dentex shows a rounded edge, some
The maximum 20 minutes. smooth dull polish, sleeks, rough bottomed and stepped, long
On experimental obsidian tools the processing of fish is and short striations. Severe attrition could be present on the
. gnification JOx 4x.
mostly visible. edge (Fig. 6). Ephinephelus fishes. Ma
. e resulting from scaling
F1g. 5. Micro-wear trac
.
gnification JOx 4x.
. D entex dentex fishes. Ma
e resulting from sea /'mg
Fig. 3. Cutting Dentex dentex .fish with an e).per im ent al obsidian tool. Magnification 20x 4x. Fig. 6. Micro-wear trac
205
204
Iovino, M. R.
fish processed.
Acknowledgements Th e main topics of this paper are: A - the excavation evidence of weaving in authentic archaeological material in
f{ung ary, B -
the constntction evidence of a loom based on these archaeological data, the weaving experiment.
on the difference between the meat and the bones structure of Today we know only a few, unpublished, textile remains from
europei, Mursia, Milano.
the Neolithic and Copper Age periods and they are known
thunnus compared to the other fishes. Moss, E., 1983, The functional analysis of flint implements,
l
British Archaeological Report, Intemationa Series
Both Semenov (1964: 10 7) and Moss (1983: 104, 105) 177, from personal remarks of the archaeologists (Table 1).1
observable rounding of the edges and striation. In most cases patterns. Pots with a painted textile pattern had been
American Elsevier, New York, 120-131.
Semenov, S. A., 1964, Prehistoric Technology,
these microwears do not show the particular species of Cory, Adams decorated by a stripe of fabric dipped into paint. They are
worked fish. We can observe a good pattern scaling dentex. and Mackay, London. known from Neolithic and Eneolithic sites.2 We can say the
same about the Bronze Age pottery. 3
We have one curiosity as well from the middle Bronze Age:
some symbolic figural representations of a carpet or a loom
(Figs. 1, 2 and 3).4
Weaving material can be demonstrated by all of these
sources in the Iron Age as well (7th-4th century BC). Textile
imprints can be seen on the following published iron objects:
the certosa fib ula from a cremation grave of a 1-5 year old girl
(Szentl6rinc), the iron ring from a tumulus grave Kismez6.5
Textile imprints can be found on a pot from the fortified
settle ment of Szazhalombatta and on a ring shaped loom
We ight from stray finds of Velem-Szent Vid.6 We should also
mention the well-known figural representation of the weaving F ig. 2. Site unknown, Szalnok, Damjanich Museum,
process on the um from Sopron Burgstall.7 figural representation of a carpet on the loom.
I
2 H orvath & Marton (1998), ann. 6. 8.
Asz 6d: Kalicz (1985, 1998); T. Biro (1992: figs. 18.2, 21, 22). See in Horvath & Mruton (1998: 258, figs. 4-6), Asz6d, Se: in Kalicz (1998: 80, abb 25-6;
3 82, abb 27-9; 122, abb 50; 123, abb 51). Textile or basket imprint: Zalaszentbalazs-Sz616hegyimez6, Early Copper Age, in Banffy (1995: 20, abb 6-3).
Ember (1952), see also in Horvath & Marton (1998: 260, figs. 9-10).
,
Kalicz-Schreiber (1984), see also in Horvath & Marton (1998: 259, fig. 7).
m (1968: 169, t. XXXVIII, 1-2), Vadasz (1997: 29, abb 2-10).
, Jere
,
Szazhalombana, unpublished, unrestorated material Str. Unit 23/1994 from E. Marton's excavation. Velem St. Yid: Hrivnak & Polka (1998) dissertation of
1 Berzse nyi D. Highschool, Szombathely, Dept. of History (1998) (in manuscript, with pemlission of the authors).
. JOX 4x.
n cat n
8. Residue of a scale of Dentex dentex observed on the suiface of an experimental obsidian tool. Mag ifi w
Gallus (1934).
F ig.
206
Proceed
ings of the 31" Jntemational Symposium on Archaeometly, Jerem, E. & T. Bir6, K. (eds.), Archaeopress -Archaeolingua. BAR - Centi: Eiu: Sei: 1.
HoJ111ith, T. & Marton, E.
The warp-weighted loom in the Carpathian Basin (Hungmy) - Experiment in the Archaeological Park, Szdzhalombatta
AGE SITE CULTURE FIND No. SHAPE of the weight LOOM and
SITE CULTURE FIND No. SHAPE of the weight
L OOM and i\GE WEAVING
Neolith Tiszajen6
Kiiriis weights ? pyramid WEAVING
Bajc-Retz trawl remains ?
Neolith Gyomaendr6d
Kiiriis textile imprint warp-wei
g pe r Gy6r-Szabadretdomb
5 hted cop pot with basket basket weave
szen tbalazs Late Lengyel
(line) ;;,rp _weighted er Zala
copp weave imprint
basket weave s drrecti
on
of twisti Tizaujfalu- Baden piece of a fiber
(Salix sp.) ng copp er
Neolith Devavanya (line?) grave "s" direction of
Protovinc a weights tabby vmhegy
5 ring, 'tomato' -shaped twisting
Neolith Devavanya- wa rp-weig
Szakfilhat? two rolls of weights hte d
Rehelyi-Gat H6dmez6vasarhely- Perjamos? small bone tablet stray find
Tisza? between two post holes 21 Eaf lY
conic with 6 holes tablet weaving?
in a house warp weig
hte d ze Szaklilhat
Bron 3x3 ems
Neolith Magyaratact
Emly Linear Pottery pot with matting carpets 2
Hatvan pots with textile imprints ?
warp weigh T6szeg-
imprint ted Bronze
"z" directi on Laposhalom
of twisting Szigetszentmik16s Nagyrev, Vatya pots with textile imprints ?
Bronze warp-weighted
Neolith Szeghalom-
Szakfilhat? supposed loom ? ?
twill
Bronze P
lrnzdvar
.. ..
, Vatya weights ?
Kovacsdomb cilcske-Vorosgyrr Nagyrev,Vatya weights 3 pyramid warp-weighted
Tiszai? Bronze B
Neolith Tcirekpuszta-Kavicsbanya Szazhalombatta Nagyrev, Vatya weights 18 pyramid, sphere warp-weighted
Linear Pottery pots with textile imprints 2 Bronze
Neolith Mesztegny6 ? unafiildvar Nagyrev, Vatya weights ? pyramid warp-weighted
Linear Pottery pot with textile imprint 1 Bronze D
Neolith Pari-Altacker naujvaros- Nagyrev pot with loom- a supposed
Transdanubian textile imprint on the 15 Bronze Du
warp weight d Kosziderpadlas representation loom with carpet
Linear Pottery base of pot e 2 a supposed
Nagyrev pots with loom-
"z" direction of Bronze Nagyrev-Zsid6halom
twisting representation loom with carpet
Neolith Bazsi Vatya I II. weights ? conical warp weighted
Late Transdanubian textile imprint on the tabby, twill Bronze Alpar-
warp weighted Vardomb Koszider
Linear Pottery base of pot
Neolith Battonya- Vatya weights ? pyramid warp-weighted
Early weights from a h ouse,
tabby Bronze Kakucs-Balladomb
? Gava weights 2 ring stray finds
Parazstanya Szakfilhat warp weighted Late Baks-
seclion V.
Neolith Battonya- Bron ze Temet6part
Szakfilhat-Tisza piece of a thread, ?
Parazstanya ? Janoshaza- Urnfield textile debris from ? ?
house No. 6. Late
Neolith H6dmez6vasarhely Bronze Orszaguti du16 grave No. 1983/2
Tisza weights with textile
warp-weighted Nemetbanya Urnfield weights, 3 ring, pyramid warp-weighted,
debris Late
Neolith H6dmez6vasarhely- small bone tablet 7 tablet weaving?
Tisza textile debris from a Bronze-
Gorzsa ? Early with holes, 9x4 ems 1
girl's grave
Neolith Szegvar-Tuzkiives Tisza Iron
weights
Neolith Ocsiid-Kovashalom warp-weighted LB-EI Kese116hegy-Ziiriighegy Urnfield weights ? ring, pyramid warp-weighted
Tisza weights star, trapezoid warp-weighted Szazhalombatta- Late Urnfield, weights, a supposed 9 ring, pyramid, warp-weighted,
Neolith Gyomaendr6d-Endr6d Tisza Late
textile imprint on a
warp weighted Bronze- Sanchegy Hallstatt C-D loom, sphere width: 200 cm
grinding stone
"z" direction of 0 Early pot with basket weave
(vegetable fiber-line?)
twisting lron imprint
Neolith H6dmez6vasarhely- tabby LB-EI G6r-Kapolnahalom Urnfield, Hallstatt pits with weights ring, decorated warp-weighted
Tisza matting imprint 4
Kiikenydomb "z" direction No.E-F/5 "c" pit 8 pyramid
(Typha latifo/ia)
matting weave - No.I-6 "c" pit l
Neolith Battonya- twill 9
Tisza pain1ed textile imprint 3
Giidriisiik "z" direction of LB-EI Velem-Szent Vid Urnfield, Hallstatt weights 5 ring, pyramid, warp-weighted
on pot
twisting 6 sphere, donut
(yellow)
tabby lron Vaszar Hallstatt C textile debris o n iron ? tabby
Neolith Se ring of a bridle
Early Lengyel bowls with textile 2
tabby Iron Kompolt Pre-Scythian debris of cord from a 8 "s" direction
imprints
Neolith Asz6d- Hallstatt B3 well ? of twinning
Lengyel weight and grinding
Papi fiildek pyramid warp-weighted (bast fiber)
stone with textile
tabby Iron Saghegy - Hallstatt C textile debris on iron tabby
imprints 1
Neolith Asz6d- Kismez6 ring of a bridle
Lengyel painted textile imprint 2
Papi fiildek ?inkle weaving or Iron Fehervarcsurg6 Hallstatt C-D weights 2 half-ring warp weighted
on a vase
twill Iron Sopron-Burgstall (Varhegy) Hallstatt C-D weights, ? pyramid, ..? warp-weighted
(yellow)
um with loom
Neolith Asz6d- representation
Lengyel rolls of weights 32
Papi fiildek truncated pyramid warp weighte d Iron Sopron-Krautacker La Tene A "the weaving house" ? weights warp weighted
Neolith Kajlirpec- Iron Szentl6rinc La TeneA fibula with textile debris tabby
Lengyel matting imprint on a
Pokolfadomb matting weave Iron Szentes- Scythian textile debris from the ?tabby
base of pot
Neolith Zeng6varkony Vekerzug grave No. 53.
Lengyel matting imprint on a 2 matting weave - Iron <:;sanytelek- Scythian textile debris from the ?tabby
base of pot
?twill Ujhalast6 grave No 16. 44. grave
(Typha latifolia) Iron ?tabby
Neolith Bicske Sandorfalva- Scythian textile debris from the
Sopot-B icske weights?
Eperjes grave No. 1 38.
objects with 2 10 star ?lucet or threa d Iron Tiszavas vliri La Tene C a ring shaped weight on ring ?warp weighted
or more arms
Neolith Nemesvlimos-Balaca winders the woman's basin from
Sop ot-Bicske weight with animal
conical - zoomorphic, warp-weighte d No. 3 1 . inhumation grave
head
thread win ders? Iron
Velem-Szent Vid La Tene D? weights ? donut warp-weighted
star shaped objects 3 star 2
with 4, 6, 8 arms Iron
Neolith ecsehely- Bodroghalom La Tene C-D textile debris on an iron 1 ?tabby
Sopot textile imprint on a
Ujmajor tabby knife
base of pot
"s" directio n of No. 1977/10 grave
(hair or kemp?) Iron
twisting Kosd La Tene C-D textile debris from the 5 ?tabby
graves No. 46. 54.
'In situ'found loom weights and their problem reconstruct the ancient weaving technology, we had to buil
d
up a ve1tical loom, making clay loom weights and to construct
Fig. 4. In situ found loomweights of G6r-Kapolnadomb.
The idea - All the archaeologist can suppose, as Barbers a warp with weights.
says: 'The most direct proof that a set of weights belonged to
a loom comes when the loom happened to be set up for use at 4. Rapport
the time it was destroyed, so that the weights are found lying The weaving experiment
) technic
in distinctive rows, having dropped a few inches to the ground - a constructed vertical, We wove with plain weave (and basket weave
when the warp threads from which they had hanged were warp weighted loom11 (Fig. 5). .
the above
destroyed by fire or decay. Such was the case at Troy... : This is the most favourite technology m
between two post-holes and the wall lay three or four rows of imprin ts for a long period . The appea rance
1. The loom construction mentioned textile
clay loom weights just had fallen (Blegen 1963: 72)'. logy is an impor tant issue here. We are sure
of the twill-techno
Basin will
The main question of our experiment was: how is it We used several data based on the literature, ethnographic that the archaeological material of the Carpahian
possible? analogies, figural representations and other archaeological have a surprise in store.
reconstructions.12
'In situ' evidences Conclusions
2. Loom weights
r to
After collecting and analysing many published or non The main task of our experiment was to give an answe
published archaeological evidences of the so called 'looms' The colleagues of Matrica Museum made copies of the clay Barber's idea:
tme,
in Hungary, we realised that from the earliest Neolithic data weights found in the middle Bronze Age tell-site (Excavation The warps with weights were standing for a long
m a
(Tiszajen6-Szarazerpart, 14C date: 7000-6000 BC)9 until the of I. P oroszlai). They made two types of weights, (pyramid during the weaving procedure in a pit, in a house or
Iron Age 'weaving pit' (G6r-Kapolnadomb 14C data: sanctuary. Weaving could last for some weeks or for ome
and sphere shaped), cca 50 pieces. It weights about 100
ts
269040 BP, 892-872 BC and 288040 BP, 1138-992 BC) gr/piece. months. If the textile was ready, the warp and the weigh
(Fig. 4)10 we find the same authentic circumstances in the became useless and they were cut down. They have hange d no
excavations: a lot of weights in rows and two timber holes. 3. About the material of the threads more. They were used as real weights, collected in a pit or left
There are differences in forms of the weights and in the at the same place, under the loom. This is why we doubt tht
estimated distance between two post holes. The number of We used linen threads for the warp and woolen threads for the position of the fallen weights could help us m
weights could be indifferent. In the published prehistoric the weft. Our choice was not based on the botanical or reconstructing the loom. If you make the warp, you will see
material of the Carpathian Basin we could not find wooden zoological investigation of the tell-site, because the bon e that knotting the weights is simple, and it is not a hard work.
remains or other organic matter, as an excavation evidence, material has not been analysed yet, the botanical materia l is On the excavations the problem comes from the missing parts
that may be brought into connection with the construction of only partly analysed.13 We followed the weavers' pra cti cal of the loom structure: the heddles, the shed bars, the measure
a weaving loom. The only evidence is the existence of the experiences, but we do hope, we will find the authentic of shed opening. There are no excavation evidences for these
weights and the supposed existence of two vertical beams of answer in the botanical and zoological investigations in th e and we have not found any of them in our material. There is
a loom. During the technical examinations, in order to near future. only rare chance that we could register the 'ideal' situation in
an excavation, that is, when the house caught fire and the
Fig. 5. Tabby imprint of Szentlffrinc.
8 Barber (1990). people had to leave their house quickly, leaving a halfmade
9 Selmeczy (1969). carpet or the hanging warp on the loom.
for Istvan Ecsedy, Andras Figler, Ferenc Horvath, Gabor Ilon,
'0 14C data: lion (1996: 184, ann 34). The material of the weaving pit I-6 section C will be published by E. Marton with Gabor lion's kind pemussion.
,
rop ol Szi e1
g .
"
ee
,
6, * pottenes and obje
n
Archaeomet1y Laboratmy, Isotrace, University of Toronto, 60 St George st., M5S 1A 7 Toronto, Canada
* U n e 1. Fuz 2
K ., 1999 ' Marton E 199
,
Almassy, 8 7 2 . ,...
;:
7 1 9 9
'
cts mad
'
JAM XXIX-XL, . J<lan g Velem -Szent Y
clay i
T he bow
Gyu1a1,
eJi nu n Szazhalombatta, in
F., 1996 ' Pr 'te of Oravetz, H., 1997,* Late
a ]993-1 96, Poroszlai, I.
Te J l- s I . . Ki:iri:is (Protovinca) settle
mbatt
of the 9 me t and
Investigation bunal at Devavany a-Barcei kisha
a lo
a.
lom, CommArc
Four expe1iments (Ballistics, Distance, Accuracy and The bow used was a standard twenty-five pound draw
ombatt
. s- a z h hnung,
Excavation s at 5-25.
l hal Patek, E., 199 3, * We stu l am Hunt Sim ulation) were carried out using projectile points fibreglass one with a multi-strand linen drawstring, loaned for
2 6, S z az hted loom, Studies in the l g ill der Halls tattzeit' VCH A
(ed.) ' l - 6 ,, wmp-weig
, TM im plement, Norway. Humaruora. Quellen und Forsch. Zur
manufactured to an archaeological standard (Epipalaeolithic the study by the University of Toronto Scarborough College
logy 01
., 1 96 4 c ie n t .
Hoffman n, M .ra n a n . priihist. Und p projectile p oint type: Nachorine; cav e site in Anti-Lebanon Sports and Recreation Depaitment. The arrows were purchased
H.uo Te ch n o prehistonc Spinning and vinzi ro. . .
t ryand . *, . alrom. Arch. 7, Weinheim. .
& J.,
18 mountains). Manufacture of projectile pomts was accom from Hercules Sporting Goods, Toronto, Canada. They were
on, E.,
Horva'th, T. e c arp <
B asm (Hungary), in
..
M art an
ung a1 II. Kolto, L. &
i Reg
s fr om th
at h eny e, 1996 ,* Fundorte der Sopot-Kultur
,,
y auf dem plished by first truncating flint blades provided by the standard 74 cm, 20.0 gms (average weight), three feather
Weaving Object
. se a1 .c h e s i n H. Balatonhochland, Com111ArchH1111g, 23-41.
archa eolog y laboratory of Scarborough College (courtesy of fletched, metal pointed arrows with plastic nocking tips and a
l Re
R. B. Schroeder), and then forming the edges and notches
Archaeometnca 49-26
9. Rezi-Kat6, G. 1999,* Kaj<irpec-Pokolfadom
b, poster in centre of balance at their midpoint (approx. 37.5 cm from base).
Bartosiew icz , L . (eds.), T/le T y p o l ogy ?f spin le and
Lengyel'99, Veszprem 11-13, October, abstract
M., 99
8
"'' , p., & Polka,
, book. with a 0.32 cm diameter copper punch. The manufactured
. D1ssertat10n of
ent Vid,
.r
Selmeczy, L., 1969, Das Wohnhaus der Ki:iri:is
s OJ . em - Sz
Hnvni:U\. Gruppe von points were given numbers, and those numbers were placed The hafting
Vel Dep .
t. of History
athely,
weaving weig ht s zo m b Tiszajen6, MFME 196912, 7-23.
seve ral areas of the projectile points so that they could be
school, T.
.
Early Axe
on
D. H i g h Biro, K., 1992, Data Oil the Technolo y of
Berzseny1 g identified in the case of breakage. The ten specimens were The attachment of the manufactured points to the arrow
rip t). l! h a ndwerk der U men-
ta Production (Adatok a korai baltakeszftes techno/O d
vor1 aufiger
(in manusc
as, Ungam). "
Me
. gi
jahoz), Acta Muse i Papensis.
zum each measured for a number of potentially significant shafts required that the metal hunting points and the wood
11on, G . 1996 , , (K onut at V
Be1t..rage .
Gor , E & L'ippert, A.
attributes: length, width above notch, thickness at the base, within them (total weight 1.5 gms) be removed. The shaft end
ttku1t 11 '
felderkultur -
r Jerem . The late neolithic in the Tisza region, 1987, * catalog,
thickness at the midpoint, thickness at point, truncation was then shaved from 9 mm diameter to approximately 7 mm
.e o sthalls ta s
.
.
ngu a : B
Bericht, in Di t Szolnok-Budapest.
ud a p e . .
s V.
. angle, truncation length, point angle, left and right notch diameter for a length from the edge varying from 1 cm to
tbro n z
(Hrsg .), A rch ae o!i ez e it h c he Huge lgra ber in Vadasz, E., 1997, Das Hilgelgrab von Kismezo bei dem
'. l ,
dista nce, diameter, depth and, finally, weight. As a rough 3.5 cm. This was done to test for potential shaft shattering.
c aArcl1Hill g
2 ' * Spa
Jankovich, K.,
3 -81
,, .
1 99 g, Zalai Miizeum 8, 27-39.
t
Sagber
V. Szabo, G., 1996,* Forschungsprobleme der Csorva-Gr uppe
me asure of comparability for the projectile points, the A split was made in the shaft with an Exacto knife and a notch
gend, A meter y
o f Szentlonnc, Acta
der Bakony-Ge e ce standard's weight, and the mean and standard deviation of was then cut to incorporate the stone point. Shallow grooves
n Ag
T he Ir o
Jere m, E ., 1 968 ' und der Gava-kultur aufgrund einiger Fundverbande aus
komitat Csongr ad , MFME StudArch ll, 17-18.
. the manufactured points, were calculated and compared. were then carved around the circumference of the shafts in
8. a n Textiles u ntil AD
The standard weighed 0.76 gms; the mean (X) of the manu
S , 1 96 r o p e dem
Arc h . HA / Eu
V. G., 1997, * Sie Sahen die Tore von llioll. F1mde aus
Norfl alignment with the notches in the stone points. The points were
L. B. , 1991, Szabo,
forgensen, , factured points was 0. 78 gms with a standard deviation of then bound into place along the notches with Strikemaster
enmark
1000, Aa , D A szod.
rh uis o n , der Bronzez eit von mittleren T heiss gebiet, Gyulai Katal6
'd
lu A sl_o
J
mm diameter,
Kuns1
o.18 gms. The mean weight for complete projectile points
n
braided nylon casting line (0.5 25 lb (11.3 kg)
ffkon'fia
' K b emalte Keramik aus
.
985 u d gus ok
.. r Jiche
K alicz, N., 3.
Afcha e olin g ua Ser. Min.
from the Nachorine archaeological sample from which the test) purchased from the Canadian Tire Corporation. The points
K alicz, N., 1 9 98
, Figu
ng s
um Westu
. arn . standard was taken is 0.62 gms with a standard deviation of were then dipped to the binding on the notches in Esso Imperial
.
dem Neolithik o.25 gms. Parowax to hold them in place. Following this procedure the
The points were made from a dark brown Lebanese chert entire shaved shaft area is tightly bound with the nylon line. The
With the exception of No. 6 (Fossil Hill Che1t, Ontario, total binding was then placed in the hot wax for 5 minutes to
Canad a), so uniformity in results might be more likely permit any trapped air in the hafting area to be replaced by wax.
durin g experimentation. The experiments were designed to Then, after removal and cooling, the hafted points and their
determine what the performance characteristics of bindings were dipped in the hot wax quickly and cooled for a
re co nstructed
projectile points might be, and from these half dozen times, which permitted a elem layer of wax to be
character
istics, what the probable modes of hunting might built up on the entire hafting area. The wax was then trimmed
h ave been
:scertain the bow and arrow or atlatl configuration may have
within a highland cave setting. It is not possible to away from the point area. Clear wax was chosen because any
stresses in the hafting aiea would frost the wax and be readily
een tha t
propelled points in antiquity. The vagaries of observable. With a bow and the prepaied arrows the four
arch aeo
logical time will have destroyed most of the clues to experiments above were carried out.
reco nstr
ucting this aspect of ancient behaviour. But, for the
Test 1: Initial velocity estimates
Purposes of
this study, this lack of information is not deemed
he vital. The experiment is that of a 'best-of-all-worlds' or
Ideal typ
e" bow and arrow shaft, one that reproduces an
Idealized
. The most important quality of the bow-anow-stone point
shooting situation ('best-of-all-worlds' from the system that can be ascertained expe1imentally is the initial
illod ern
tech nologist's point of view), so that the stone point
:
rfo rrnance characteristics may be more easily isolated and
velocity that the bow can impart to the arrow shaft and its point
ogue.
ble I catal I entifie
when drawn. Thus, each numbered aiTow was maiked at 30 cm,
s h ed finds to 1'a
References of pubh.
d. 45 cm, 6 0 cm and 66 cm from the nocking tip notch. These
*
ings of the 31" /11tematio11al Symposium 011 Archaeomet1). Jerem. E. & T. Biro, K. (eds.). Archaeopress -Archaeolingua. BAR - Ce1111: Ew: Se1: 1.
212 Proceed
Pavlish, L. A.
Experiments with a bow and arrow
mis
_
th e weight.
slightly lower velocity values than those
calculated from the ap _Y n with a standard metal-tipped arrow. The experi Market in downtown Toronto, Canada. The cut of meat
com p a riso
photography. ment co nsisted of shooting arrows from a kneeling position at weighed 9 lbs 14 oz (4 .5 kg) and was composed of layers of
from v + dg where v (velocity) the sque meat, fat and bone. The average thickness was 8 cm.
ro ot of th prescrib ed distances (2 , 4 , 6, 8 and 10 metres) into a
Test 2: Distance exp eriment
=
target with a 7-cm diameter bull's eye centred The length was 29 cm, tapering to 17 cm, and the width
0 lties styrofo am
in the field of approximately 3 0 m/sec) T .
his ve i o ci
ty is I m etre above the floor. The experiment was conducted in two averaged 21 cm with a maximum width of 24 cm.
e the distance
The purpose of this experiment was to determin equivalent to a draw on the experimental bow
. . of b e tw ee 45 parts. Part One consisted of shooting, with a 60-cm draw, a The experiment was conducted by hanging the piece of
to a standard n
ling system would impart and 60 cm which would create impact forces
that a modem propel quite adeq stan dard an-ow from the five positions and then shooting the meat backed with newspaper on the styrofoam accuracy test
compare these results 1
ior killi ng aru ma s, 1'f a degree of accurac y cou
uate
arrow and a stone tipped an-ow; and, to "
ld be attaine sto ne tip an-ows as a group without individual discrimination. target and shooting a standard arrow and stone point tipped
. 1
an atlatl.
. d.
unweighted atlatls by Peets (1960: 108-11 0 ) and Howard Pavlish and Thomas 74.06 meters 26.94 meters co nsisten
cy and within 100 metres with random luck. In low
80 y=8.5x-22
and are less than those notd by game density situations, or in areas where the game would 4-9
Y=l5.75x-32.3
(Howard 1974: 102- 104 ) 60
ed by Pavlish & shots into the light breeze
58
and observ
)
Brown (194 0 (Howard loc. cit. ) 1 he spread out over a wide ten-itory, the most useful stone
40
r (197 5). The velocit y attained by JlOint arrow shot would be one made at close range, where Y=l.5x+32
t.
Thomas (1974 ) and Butle
the dart was approximately 24 metres per
second (calculated Table 2. The distance test for anvw and atlatl dar acc uracy
could be combined with killing power. This is 20
readily
apparent when one realizes that with this method of 0
the stn ng at the a ppropn ate Place
to the
was anamed by placing the bow in a fixed post. t.10n and drawmg hunting
The measurement of the poundage pull of the bow there would probably be no need for standardization
1
of ar
0 2 4 6 8 10
required distances with a calibrated scale s pring. al eolithic ro w shafts beyond general parameters in a manner
carried out with a replica of a 20,000 year old mammoth tusk boomeran g recovered from an Upper
Recently, distance and velocity experiments were p
a unda
ge analo
gous to the stone points. Resultant from this, one Fi g . 1 a and 1 b. Accuracy results.
site in Poland (SOO gm). The results of this experimen
t suggest that the range and velocity of a boome rang are comparable with those of a bow with
21.4 gs maY !n
replica travelled up to 66 meters and reached a verage veloc1ues
of 1ess th an 25 pounds, and an atlatl The boomerang
of meters/second. Boome .
th e y e 1ess d' '
a gle will lessen the probability of g an1e observauon and flight; and, 2) l'he acquisition of road-kills from the Mi istry of Resources pemlitted carefully preplanned experiments which evaluated the performance qualities of these
. . . ed 111
confer two advantages on the user: I.) the lo w approach n
1sadvantag
(Bahn 1995. 562).
214 215
Pavlish, L. A.
Experiments with a bow and arrow
01 s'
b reak 101 . bec arne mde pendent of the haft. One
. to
. the pro
point did stone point delivery systems is not With biem f
oi nt . o on accuracy and distance of the atlatl, Carried out at
here the P bone th at
was no t nea r the . . o ut
This approach d1d he1 p to elmun ate som
meat surface " Its ad an
Plane w
ok e n
s ide th e hafts and
e of the
v tages
source f aJul, N_
P. G., 1995, Flight into Pre-history, ature 37 , 562. Bronte Creek Provincial Park.
. point br 1
n in the point, while that may have been B Peets, 0. H., 1960, Expetiments in the use ofAtlatl Weights,
broke
variation p res e n t if a reco so
J., 1940 ' Proiectile
.
othet n strucf J points, American Ant1qwty 5,
nt s w ere pronu
ded substantially into srown ,
.
pot o the mea t been attemp ted usmg some ethnograp Ion had
Two . bone, a \s . hic mod
el In addi . 20 9-213.
American Antiquity 26, 108-110.
be d d ed in . this approach to expenmentatio n did hon, J., 1606-1625 [1907], Narratives of Virginia, Tyler, L.
Sm ith,
sutIer, W. B., 1975 ' The
ac compli. . atlatl: the physics of function and
im e (Table 3). re a t a d vantage m hun ting with hafted
.
e, w hi c h was to sh Its . G. (ed.), Scriber's, New York.
objecuv atte mp to
be a. g
abov t isola te
ht e , b rea k and . som pedonnanc e, Plains Anthropol ogist 20, 105-110.
Th ete
mig bon leav e a large portion performance cha actenst1 r . c of the replic e of
htt a c ti. ssue. Suc Native American Bows, Bagley, M.(ed.),
the
ated s tone .
ouId Jiam ilton, T., 1972,
.
. t ha t w le h a breakage pattern the stone-tippe
pomts .
T
po in t s u s . velocities of d a . he
he rn cri pple an arum
now
G.
s for vano us Schumway Publishers.
into t . bow dra York Penna,
ing al. Bow strength and .
slightly
ru d to be less w
quicklY
were found th s
prot ' ut c ompar
a n b
. able to th
c ei t ainl Y pr o bably do not dictate this kind of with a modem stan . ose
wouId obtained dar '
. a \one generaiiza .
d meta1 -ti
"' vel oc1t Y s ueh .
uon s are premature
pped a
rrow.
Maximum distances were found to b e a .
arro"' t tedlY
size, . . p pro x mately t
d 1p le but the se observations do 1
br e ak a ge. A he s an same. Accuracy vaned with ston e tipped a r he
o n s1 d eIs t hi c can
h serv e . rows, but on
ec as testable model th e
when n whole tended to drop off with g r eater rapidi
w
u ns
. id ques o \e , u n Iess a pomt-breakage ty th an t
.
p otential 1 to of a strategi c behav1our usmg such anow points In pn. .
' ld se e n nc1p1e,
ou
point w
.
. the
ak age en counte red m the experiment technological subsystem, the understanding of whic h
mo deI .al bre< ' .
f patterns nugh t be associated with
may
ene1
m atelY
f a p proJU
o
Ih:!ill'. Results
[Jii. 30 penetrates several cm into meat, and when hitting bone bounced ou t of the meat.
r
Stan da d 45 penetrates into the mea t t hrough entire thickness most of the ti me when not hitting
Stan dard bone
60-66 penetrates meat comp letel , and penetrates the bone 2-4 cm
a y
St and rd 30 penetrates point depth into meat
6
4 45 point breaks in bone
ti
11
30 just sticks in the bone and sna ps point
I
5 30-45-60 penetrates meat to varying degrees
3
2 66 breaks in bone
ti
11 30-45 penetrates meat
5
3 60 breaks at base in bone
5
3 66
7 breaks at base
. ii s Naillltives of\rllgi ia (1606-1625) noted t hat : "... Forty yards will they shoot level, or very neare the mark, and 120 is t heir best at
n
sniith 10 1 e tO the local Indian
s.
' JoIio " in refer en>
randoOl
216 217
THE HEATING ISSUE IN ANCIENT TECHNOLOGIES
PAVLISH, L. A.
Archaeometry Laboratory, Isotrace, University of Toronto, 60 St George st., M5S IA7 Toronto, Canada
whether or not prehistoric peoples employed heat treatment of chert and flint as an integral part of their lithic
'[he issue of
roductio n technology is a matter that has long been debated. A heating model is presented along with underlying
/Iopso phical principles upon which the model is constructed.
1
T he methodology is developed and discussed. T he overall
P 1 ctusion reached through this work is that the heating issue will always have to be addressed on a case-by-case basis, and
:
aoncep tual
tools employed to make the decisions will be phenomenological in their make-up.
J(EYWORDS: P URP OSEFUL HEATING: TOTALLY HEATED MATERIALS (BAKED OR BURNT); PARTIALLY
General perspective on heating formed into a sharp point. In other words, knowledge of both
degrees (partial-to-total) of heating and intensities (roasted
'The application of heat to raw materials to make them more to-bumt) of heating as well as knowledge of organic and
useful probably began with the first control of fire by humans. inorganic mate1ial properties was available to the ancient
Advances in the understanding of py rotechnology eventually manufacturer. In the case of stone, experimentation indicates
led to the firing of ceramics, the smelting of metal and the that some of the same principles of heating apply as those
potential for large-scale material production and consumption. which would apply to wood or bone. In all media, there would
What is the evidence from pre-ceramic periods for the necessarily have been many more options and choices
expl0ration and development of a heating technology asso available to the ancient technologist than merely heating
ciated with the manufacture of stone tools? For the most part, indiscriminately; and, undoubtedly, this situation would result
the study of pre-ceramic, heat-altered materials has been in a broad range of intentional heating patterns at times
primarily ad hoc, with the limited objective of determining only overlapping with the range of inadvertently heated examples.
whether or not a particular specimen has been altered by heat in The fundamental question arising from this theoretical
antiquity. Review of a wider, and largely incoherent body of observation is:
literature on the subject suggests that much more could be How can purposeful heating and potential for variability in
learned about the prehistoric use of heat to alter materials used that heating be detected in the archaeological record?
for making tools. Consequently, an instrumental approach to
this subject, coupled with experimental observations on Recovering evidence
reproduced, archaeological-like specimens, may point the way
toward a better and more systematic understanding of some of In order to recover successfully evidence of partial heating
the prehistoric artisan's uses of heat. from the archaeological record, one must first establish that
In some cases, evidence of heat alteration can be detected; evidence of this kind is archaeologically recoverable. A lack
and, under certain conditions, it can be shown that such of knowledge of the differentiation of degrees of heating has
alteration probably was intentional. Analyses (e.g., TL) can existed because researchers have rarely looked for it
show also that the study of prehistoric heating technology is a sy stematically. Until recently, the problem of distinguishing
far more extensive and complex undertaking than had accidental or incidental heating of any kind from purposeful
previou sly been supposed. In addition to providing an heat treatment2 was given little consideration (Klein 1973;
app reciation of the potential role of temperature in ancient Rick & Chappell 1983; Morrow 1981). This matter of
manufacturing sequences, these experiments establish that purposeful vs non-purposeful heating can be rectified by
certain prehistoric technologists not only used temperature employing thermoluminescence ( TL) as the determinant of
ht, at times, used it with great skill. For example, when one heat treatment. Materials heated in antiquity will have
Wished to make a wooden spear point, it was possible to decidedly different TL outputs in the form of light than will
Partially bum the end of the wooden shaft, softening the their unheated, geologically-similar counterparts. In addition
material, making it easy to remove the outer section. Then the to determining the potentially differentiable TL outputs, it is
remai ning unburnt core material (Adam 1951; Movius 1949, necessary to analyze a large sample size (N>100) and, if
!950),1 possibly partially heated, and perhaps hardened - possible, confine the sample to a specific stage in a lithic
depending
on the species of wood chosen - could easily be reduction sequence. The rationale for the large sample size is
I
'I'he
8-foot long yew wood spear found in association with Elephus antiquus and Levallois flakes at Lehringen, Germany appears to have been fire-hardened
and used in the demise of the naturally-bog-trapped elephant (perhaps fire-driven). The site is dated between 80,000 and 125,000 years ago (Adam 19 51;
Movius 1949,
i 1950).
'Purposeful
heating' of stone is a term used here to suggest that there was direct, conscious human intervention through the application of heat to stone or
Ot her m
aterial for the purpose of facilitating some objective, be it technological with respect to manufacture, cultural oraesthetic. 'Non-purposeful' heating
by human agency may be 'accidental' (e.g., retouch flakes in a fire), or 'incidental'
(point heated during hafting with a mastic like bitumen (cf., Boeda et al.,
l996,
336-338).
Proceedi
ngs of the 31" Jntematio11al Symposium 011 Archaeomet1)'. Jerem, E. & T. Biro, K. (eds.), Archaeopress -Archaeolingua, BAR - Centi: Ew: Ser. 1.
Pavlish, L. A. The heating issue in ancient technologies
1 ignal tha n
s
b ad e.
.
IO
reu
\ al heating divisio
ns are
.
discus sed below
expenm
.
.
.
entat1o n, and
Accidental or incidental total heating (burnt or baked), or
partial heating (roasted) of archaeological material without
1
core. Pa11ial heating would be a logica of observ ations f rom
materials' disposition. Isolating a
a n1cu ar stage m_ l exp an
at
the th the support the direct conscious application of manufacturing directives
. .
.
a lithic kinds of resu lts. w in mind that the categor ies are part of a
e ana 1ytica 1 potential
ion for these various
' '
e cteveloped
class1ficat10n system for heating b
1 This the obvious archaeological context. The phenomenon could be created by
.
Heated versus Non-heated followed by the sub categ standing
c ass1fi1cat1on system must addre s s the issue - ones of temp erature control, and, notwith
s associated with an infinite variety of situations (e.g., a hearth dug into older
. only work within
both the physical state of the materia 1 difficulties, an ancient technologist could
(e.g., no heat; some Classification of h eating practices question may become deposit with lithics), but proper analyses would rarely
heat; total heat) and its disposition tho se p hysical constraints. This
. . (e.g., unheated; non- nding of the effects provide a suite of uniformly heated materials that could be
purposeful heatmg, purposef ul heatin somewhat easier to address as our understa
g) The TL method can Physical state of the material s, thus enabling mistaken for a technology. An item might be heated for no
then be appIied to a large sample of
A
however, that is 'partially heated' (the t
. erm roasted' will be such an activity might have been counter-productive, because
used mterchangeably below) would in som .
. . e cases provide the Fig. 1. Schematically o utlines a theoretical classification of heating would increase the possibility of point fracture (Pavlish
ancient stone worker with advantages .
The heated extenor for
' heating that may be present in a prehistoric technology of 1985, 1987). The heating of those projectile tips was probably
example, of a blank or prefonn could .
m ore eas11y be removed incidental to some other activity. That activity may have been
. ' bone, wood or stone. While the model addresses the issues of
and the unheated matenal that rem aine
d wouId have all the
partial or total freezing, they will not be discussed below. A N"-A the cooking of food with the projectile point being part of the
of the natural stone, providi ng
qualities .
. , maximum durability
The total spectrum of heating from Unheated to heat skewer, or perhaps the application of a heated poison to the tip
This advantage afforded by roasting' w
ouId be substanti. al m . Natural
(this latter possibility can also be purposeful if the hunter's
artial Total Human
destroyed (Decrepitation) would have been available.
A<cidi. Dci""''
.
the manufacture and use of b1 faces In a .
. dd' 1t10n ' usm g TL on Undoubtedly, an y of these material dispositions may have intent is to leave the point tip in the animal) (Judge 1973).
large samples m the archaeological re .
. cord provides a means
a fforded advantages in a particular situation. Non Accidental or incidental heating may be a factor in pre
of recogruzmg the roastmg of stone Su
ch resu1ts can provide st manufacture (e.g., mining), or post-manufacture (e.g., midden
Purposeful Heating ma y have been accidental (e.g., fore
L
.
,.0
. e of the curious (e.g., cookin g) and may display degree s of
accounts of stone heatmg that pervade .
fire) or incidental
the North Amencan heating (partial--> total). Pwposeful heating may take the technology (e.g., soil material burnt because of chance
ethnographic literature (Hester 19?2) U
. nfortunately' partial
form of partial (Roasted: removal of altered su1face leav
ing ortlI Torn Tool'""'' Total proximity to a fire).5 When a number of accidentally or
heatmg can only be detected from .
TL expenments on the core unaltered); 01; total (Baked: totally heated, Burn :
t incidentally heated materials are analyzed in a systematic
archaeological materials fo r wh i .
1
a
ch the manu factunng totally heated with swface damage, or Decrepitated: total
l y
Fig. 2. shows the classification system manner, non-purposefully heated materials will not show
teehno ogy is known. A m acro-blade
core an d blade-flake
heated with damage in the Jann of cubic spallin g, po
tlid consistent pattern in their spatial distribution. For example, in
fractures, cracks, dehydration etcetera). the behavioral (disposition ) component of heating an activity area used for tool production, the presence of a
from the material's physical state. scatter of waste flakes on the old land surface would be
3 Electron Spin Resonance (ESR) is noth er an 1 a
a y i. cal method that
! shows promise, but aspects of the signal prod uction are still imperfectly understood (Rink.
personal conununication). l the intensity of the fire jumps from place to
Fore st fire damage, while intense at the surface in confined places, is often inconsistent because the fire and
o e 1 analogous w . th
This process of 'partially h eating' stone is lo
s Y technique
ing Desert conditions may heat all surface materials as well
i the role of fire and heat in the yew wood spear point hardening and sharpen p lace; and, rarely does the fire influence mate1ials below the surface because of ground insulation.
of manufacture.
as those buried at small depth. These temperatures would rarely exceed 50C.
220 221
Pavlish, L. A. The heating issue in ancient technologies
ed when the
expected. If a contemporary firepit was located in the centre g of the spear point might partial or total heating would have been employ
Totally heated materials heated (burned), the heatin
been dere worked in its natural fonn.
that the lithic material could not be easily
of that activity area, then those flakes present in the pit dming
bi!S consi d to be incidental if one assumes heating was a
From our modem-day perspec tive,
u e, and those redistributed across the old land surface by ed to particular woods
Totally heated materials are those w
hich have b be dening properties of heat when appli enhanced an
either human or natural agency subsequent to heating, might ee n bak n. If one considers that technological factor that would have greatly
to some optimal temperature with some ed bilf ot known to the ancient aitisa
old land surface, they may or may not find that it was heated. were s1tuat10ns m antiquity where any heating of a stone tool blank or core
.
disa al of materials,
. ntage th ut Butler 1964), facilitated greater ease of remov
It would be unwise to assume that the remaining surface ffil ght be accrned from the heating and . at g qualities alluded to above, witho
weakerung surficial flakin s in the compl etion of a high
scatter was also heated. Fmthermore, it would not be logical
. 0f the al, purpo seful thus promoting increased succes
matenal was offset by some other set of adv the internal materials, is a technologic
. antageous f
actors, h anging quality final product.
to ascnbe a human intent to the finding. If, however, a large be they technological (e.g. poor heat-manage of stone , it is unlike ly that the non-purposeful
' ment fac1Ii act. In the case
lie s), accidental
sample of the flakes is examined, and shows a percentage to heating of the inner areas would be either
.
archaeologists can assume that the source of the heated flakes example of heating might afford certam .
advantages. to test its validit y. To aid
was the hearth. Other flakes in the vicinity were never heated. . archaeological record, it is necessary
Increasmg damaging power through the mt ro . Hypotheses
d
00 a tical criteria have been
Should the analysis of a large number of flakes show that all
. .
. . in the testing programme, sets of theore
: 1ct1m of a number of small cutting smfaces ffil ght n l
I crease cation of stone into genera
a model of stages in various developed that permit the classifi
o them are heated, irrespective of whether they are in the fire mtema1 l l lJUry
and foreshorten mobility and life The previous section outlines g,
through Non-P urpose ful Heatin
. categories of Purposeful Heating,
It or some distance from it, then purposeful heating may be
echamcal damage, and, within a rather short period o f li. me with ancient technological heat
phenomena associated given to the
mferred. Why? The archaeologists can assume that, because be used to test for the presence and Non-Heated. More specifically, the terms
(i.e., days), inf ction - a dum-dum projectile of stone. management. This model can
are 'burnt' , 'baked' and
almost all flakes have been heated, there is a high probability . g using TL analyses on a large physical state of these criteria
Other expenments indicate that unheated tanged
pomts or absence of purposeful heatin eful and non-pu rposeful
from a single stage in a lithic 'roasted' as they apply to purpos
that heating must have taken place before the flakes were shot'mto b one under muscle, broke at their bases 1 .
eavmg a number of selected samples the presen ce
heating. The primary objective is to demon strate
generated. Therefore, the heating was probably part o f the sturdy cuttmg edge lodged in the bone (Pavlish & S tical classification of heating
technology. Basically, the theore
a ce. In additio n to
. . lithic tool biface-production sequen
bs rvations on heated materials from a spe cific stage in a
p mt that is relatively
' Stewait Struever used the tenns 'baked' and 'roasted' in reference . .
. are different the
Journal of Archaeology 4, 221-254.
(Struever 1973).
one can be certain that, in a number of instances, purposeful
222 223
Pavlish, L. A.
Boeda, E., Connan, J., Dessort, D., Muhesen, S., Mercier, N., Pavlish, L. A., 1987, To heat or not
FROZEN BONE BREAKAGE EXPERIMENTS
. to heat,
Valla as, H. & Tiserat, N., 1996, Bitumen as a hafting
_ question! 52nd Annual Meeting
of the S .
that is
the
. o ciet ,, PAVLISH, L. A., 1 BE VER, W. F.2 & SHEPPARD, P. J.3
matenal on Middle Palaeolithic arefacts, Nature 380' Amencan Archaeology, Toront o C y tor
' anada 6-l
336-338. Pav Ii sh, L. A., & Sheppard, P. J., 19S3 T ' 0 May,
. . ' hennolu .
1Archaeomet1y Laboratory, Isotrace, University of Toronto, 60 St George st., M5S 1A7 Toronto, Canada,
Crab ree, - E. & Butler, B. R., 1964, Notes on expeiiments deterrnmat1on of Pa!aeoindian heat t
reatment .
llUnescen t
. In Ontan0 2Ministl)' of Finance, Govemment of Ontario, Canada,
m flint knapping: heat treatment of silica materials ' Canada, American Antiquity 48(4) 793 '
' -799
3Anthropology Department, University of Auckland, Auckland, New Zealand
Tebiwa 7(1), 1-6. Pav lish, L. A. & Savage H 2001 Bone
. . . ' ' ' and stone
bre ak ag e
Hester, T. R., 1972, Ethnographic evidence for the thermal m white-tailed deer, in this volume.
alteration of siliceous stone, Tebiwa 15(2), 63-65. Purdy, B. A., 1975, Fractures for the
. archaeolog .
Judge, J. W., 1973, Palaeoindian Occupation of the Central 1
S wanson, E ., L ztuc 'T'
1echnology. M ak"
ist in Qualitative experiments are carried out on frozen animal femurs in an attempt to establish whether or not bone fracturing
_ mg and U .
. sm g ties vary with the degree of freezing. Fracturing of samples cooled to 1 96 C, -40 C, -25 C, 0 C and room temperature
Rw Grande Valley in New Mexico, University of New Stone Tools, Aldme Press, Chicago, 133_ r op er
-
141. f20 oc) is accomplished using a simple two-support beam configuration and a hammer. All sample specimens are broken by
Mexico Press, Albuquerque. Purdy, B. A., 1982, Pyrotechnology: Prehistoric . .
application
with the exception of the -196 degree samples which are easily broken by hand. Although all bone samples fracture in
hammer
Klein, J. J., 1973, T hermal alteration: an aspect of lithic to
che1t methods in North America ' in T' e E
11 volution if
. the roximately the same way, the ease of fracturing increases with decrease in temperature. T he -40 C samples are found to be
app
ashi ng ton
Melcher, C. L. & Zimmerman, D. W., 1977, TL determination D.C., 31-41. exhibit two fracture types: one which cuts across bone grain structure and one that follows bone grain structure boundaries.
The unifomi fracturing qualities are postulated to be the result of bone architecture which remains basically unaltered by
of heat treatment of che1t artifacts, Science 197 ' Rick, J. W. & Chappell, S., 1983, Thermal alterat10
" n of s.11.ca
I cooling. It is suggested that the freezing of water in the bone is responsible for the increased ease of bone fracture.
1359-1362. matena 1 s m technological and functio
nal pe rspe ctiv
. e,
Morrow, C. A., 1981, Thermal alteration testing of Ft. Payne Lithic Technology 12(3), 69-SO.
chert, Appendix I in Johnson, J. K., Lithic Procurement
KEYWORDS: FROZEN BONE, FRACTURE MECHANICS, PURPOSEFUL FRACTURING, PREHISTORIC
Rowlett, R. D., Mandeville, M. D. & Zeller' E J , 197
4 The TECHNOLOGY, BONE TOOL TECHNOLOGY.
mterpretation and dating of humanly work '
.
Mechanical properties
prod uced by grinding
dry ice and p ac
he 40C
king the
samp le
.
s w er
In
e
\ er. Of the
water preent, some 10-15 ercent of the total
.
;ial m e is in the spaces m the bone occupied by canahculus,
altered by temperature change, while resistance to advancing
fracture fronts diminishes with lower temperatures.
as ex -25C temperature was bon e in
pressed m obtained b . it Th "u
make-up of bone, designe the architectural Y P lacing the e canals, and osteocytes; 85-90 percent is m the One of the interesting implications resulting from this
d ice and momt bone i versian
temperature . n dry .
stress and strain in mobile
s, ave a
cope with normal
placed in a very small
hole that was dri
1 h a th
. e nno m
series of observations and their explanation is that any
the fract re propertie of th
at b
direct inf luence on
end of the bone. This m to e
extre m
e ter
stances,rying
and in the hydration shells of the crystals of bone measuring device which applies very slow stress-strain rates
transfemng compress1on aJ
. . forces
Smee bones
alon g the.
are capable of
was thought not to ser
hole was plac e
iously affect the
d i a po
sition
W hi ch
e
ne r al.
: e lagen andat 100C
D removes the water of constitution will not generate fractures of a type that will be paiticularly
fth col
m a ve1y efficient manne IJ" longitudinal axes fracturing . of other organic matter, but not the water meaningful to the experimentalist attempting to ascertain the
might expe
r, one the sa ple (Brooks et qu ality
al. 1970). The
would be more difficult
conversely, much easier at ri
in the
n o nnal loa
ct that fracturing
d direction' and
placed m a controlled
freezer, as w
oc samp
ere the free
les W e r
of
e
:fthe bone little
rystal. The water of compact bone is so firmly
or none can be removed from the finely
ease with which early humans in the New World may have
fractured frozen bone.
. ght an samples. ze- drie 1>ound that
bone. In fact, it can be ant gJes to the lon d
.
g axis of the w ed cortical bone by enmmous centrifugal forces - 5. Samples of beef cattle femur bone not used in the
icip ate
d that fr Po der
vary throughout the bon e actunng qualities
Experimental re
held water from the original experiments remained in the freezer for a period of
design (Bonfield et al. 1977
as r
flection of its
will
sults
forc es sufficient to strip all mechanically
an d observatio
architectural crystals..." (McLean & Urist 1968, with reference to nine months and expe1ienced a degree of freezer drying.
Gorde 1968; _Yamad a l97o '. u ns
) F n
ey 19 70; These bones were found to be substantially more difficult to
Evans 1957; Ro binson & Elliott 1957).
fractunng expenment, it se om the standp fracture than fresh frozen or unfrozen bones. An explanation
em 1o oint of a Twenty-six bones wer Obviously, with water distributed in small spaces throughout
format which would most li
; ke! a
gical to
PproX!mate.
cho ose a fracturing
catt le femurs selected from
e fractured. The t
those bones col
wenty-four
be ef the bone, it is logical, as was originally considered, that it be the for this difference is that water loss from the bone increased
have been used by a peo J that which might . lected or t basis for assuming the fracturing qualities in various degrees of the bone porosity. Thus, the propagation of fracture fronts in
. p e empJ exp enm ent were of similar gen he
Therefore, 1t was decide 0Ymg a bone tec eral morphoIogy
d that
hnology. of the experiment, whi . The result frozen bone might be different. This idea is reinforced by the bone would become more difficult (Gorden 1968). Air
ch was repeated s
perpendicular to the no nnaJ 1 fracturing . on three separ
of whole bone occas10ns, are given in ate studies that demonstrate that crack propagation is retarded by dried bones of animal and people have been found to be
oad-be
.
appropnate and that attem
p ts to l
anng Iongitud
f a ke the
. .
inal axis was .
n Mach.
testmg dev1ce (e.g., Instro d that an existing (Bonnichsen 1979). need not diminish the argument because air-dried frozen bone
C and hat temperature is
the measuring of torsio i e) would not the most important factor in this
n al ' te Though not directly applicable to the frozen bone fracture fracture ease may, in contradistinction to the unfrozen air
f
be used for
(as compared to pressure for
unfrozen mterface situa
stresses and stra ms m sm n sionaI and ion
all bo compressional example). Assuming that problem, an interesting ancillary observation is that if a small dried bone fracture ease, be postulated to be somewhat less.
idea of using a free-fall h- u uer
e
n-.- samples. Further
more, the analogous interface situa
bone -water has a some hat amount of water remains unfrozen until approximately -40C 6. A theoretical argument can be made for the liquid
because the hammer wo (1 mpact h tion, one might expect that a similar
uld h av amm er) was rejected then ion exchange can take place. Thus, the implication arises constituents of bone being a key element in the qualitative
phenomenon will be b one.
e h a d to present m In bone, however,
height in excess of that . that the fossilization process by mineral replacement can
. . . fr av be dropped from a change in fracture ease noted in the experiment. The funda
1
ai able . bec ause of its vacuolar nature,
on converuen one might also expect pressure
order to lllltiate actures in t devices in continue even in permafrost conditions. Such an observation mental success of bone as a weight-bearing device comes not
. . the bo to pay a mor e important
to des1gn a dev1ce espe ne It was role in determining the final
cial ! deemed premature free has implications for the distribution of minerals through from its strength or stiffness, but from its toughness. To
Y for this zmg temperature In a
expenm entat10n, ddi tlon, the water present in the
smce the re wo rk prior to any Partially or totally fossilized bone (Farquhar et al. 1978). appreciate this point, one need only consider ceramics, for
. sults f bone ":'ould be present in
rom such acti a solu te concentration of some kind
provide some useful guidejj vity might 4. The fracture surfaces in the bone were found to be of two example, which may have great strength and stiffness, but,
n es f r and this factor would also
influence final interface freezing
their hardware design. I
n refl ? future experiments and temp kinds with some having cleavage planes that cut across the nevertheless, have little toughness, breaking easily if dropped,
ecti o n o e atures. It is at least
standard two-suppo rt beam f these decisions reasonable to posit that the bone's grain and others that follow the grain boundaries of the
cOn fig ' a fractunng qualities of the due to brittleness. Bone, on the other hand, is a very tough
was set up for the bone urati o n using bone might be altered with the bone closely. These fracture types have implications for the
fractu . cement blocks removal of these thin (onl material because it is a composite. If one considers bone's
n n g. Th _ y several molecules thick) layers of
accomplished using a ha e actual fracturing interpretation of the experimental results. These two types of index of toughness (work of fracture: kg cm/cm squared
nd - he1 was wat er simply becaus e th .
d t wo eu presence would provide a
Two lots of bone were a kilogramme hammer . fracture according to Piekarski (1970) are dependent upon divided by ultimate tensile strength: kg/cm squared), one
ssem bJ dam pem ng effect to fracture prop
ed 01. the . _ agation. Admittedly, it is
was repeated on thiee se . expenment, which strain rate. He states that: discovers it to be comparable to that of teak wood
paIa te not readily apparent how
pre-frozen to 0 Cin a fre 0 oc
cas1ons. The this bone-water interface proble m
ezer . first Jot was can be co clusively proven. "...a crack propagates at slow rates by following weak (approximately 6.00), about twice that of key steel (2.81) and
roVided ? Some insight into the bone-water
of the Fauna! Oste o-Arc
Toronto; the second' c
fhaeo
0gy Lab
by Dr. Howard Sav
oratory, University
age prbl m m general, how
ever , is provided by McLean and
interfaces of Haversian lamellae or through interstitial bone,
Which seems to be weaker than Haversian systems. At high
over twenty times the toughness of firebrick (Tattersall &
Tappin 1966). This toughness comes from the composition of
oJJecte of Unst m their text Bone:
d
experiment, was kept at r
The experiment design
o om t
em peratu
1mmediately befo
re.
re the
The ater content of bone varies with
the species of the
strain rates, fracturing propagates indiscriminately through all
micro-constituents." (Piekarski 1970).
bone that has approximately a 40 percent mineral, 40 percent
collagen and 20 per cent liquid and colloid content in its
ca li ed amma!, with age, with
van ous freezmg temper fo r the brea the nutr itional state of the individual,
atures . kage of bone at and With the nature of the Piekarski observes that at slow strain rates fibres pull out structure (Piekarski 1970). The bone structure consists of
m a ddihon . bon e tissue under observation. As
to normal room _ from between interstitial bone arid approximate the grainy amorphous mineralization, and, more importantly, crystalline
a representative standard
for the compact bone of an adult surfaces observed in the experiment. hydroxyapatite fibres (Ca10(P04MOH)i) which are imbedded
226
227
Pavlish, l. A., Beve1; W. F. & Shepp
ard, P. J.
Frozen bone breakage experiments
in the less stiff crystalline organic .
collagen matrix This matnx
contns Ii This argument would hold
qmd s and colloids which occupy true even if Paper No. 89, National Museum of Man, Mercury
.
canaliculi, and lacunae
Haversian canals, and at -40oC it is not altogeth
er obvious th
sf rem
a ins c
on stan
erved but would also become totally saturated with water.
. " .All of these spaces are
case. !he hydration shell of uch Wou Pe bone were subsequently frozen, it might acquire some Series, Ottawa, 297.
interconnected and the flow of the apatite fib :: ld be t
liquids under stress can absorb
at a given sub-zero tempera may h Jfjher different fracturing qualities caused by, among other Brooks, D. B., Burstein, A. H. & Frankel, V. H., 1970, The
large amounts of energy increasi ture as sugges Well free
ng toughness..." (Piekarski ted ab ov ze ra. gs, m o-cracks due to water expansion. An expe1iment
icr biomechanics of torsional fractures, Journal of Bone and
less strong fibre matrix, sf'. e creat
1970). Conver ely, one may dedu
energy absorbing potential of
ce that an alleviation of the
liquids and colloids in bon
the matrix sm' is unlikely to
In addition, the .
remain stab! e .
yie ld streng
With the fr
ing a
th
. of
e
ling with
im
this problem should be relatively easy to set up.
ents to establish whether an environmentally
Joint Surgery, American Volume 52-A, April, 507-514.
Canbly, T., 1979, The search for the first Ame1icans, National
the inte r- and extra- cellular water eezing xp er
w? ud tke place in frozen bone and may de of
. The act of freezing, while no thermal gradient in bone (e.g., -40C at the surface to Geographic 156(3), 330-363.
enu?ating these constituents
from the bone, removes their
In other words, sc may well
be diminished
t sc
c ease
to Sill"
.
. duced Currey, J. D., 1968, The effect of protection on the impact
sc" =sf' Vf1 + sm"(l-Vf ) Where C in e inner regions or vice versa) might create stresses
th
pnc1pal toughening advantag (3)
e and replaces it with the improve the fractming quality of the bone should be strength of rabbit's bones, Acta Anatomy 71, 87.
. I
(2) however, suggest that the general spiral fracturing of bone Milne, J. S., 1907, Surgical Instruments in Greek and Roman
were for a time in a cool water envir
onment, it would be well remains unaltered with cooling due to bone architecture, and Times, Clarendon Press, Oxford, 178.
Number of Temperature at Breakage that bone fractures and flakes more easily at lower Okamoto, T., 1955, Mechanical significance of components
Fracturing Observations
Bones Sampled Degrees Celsius temperatures. The water in the bone is postulated to be the of bone tissue, Journal of the Kyoto Prefectural
Relative scale
of Ease
important contributor to these qualitative changes though University of Medicine 58, 1004-1006, in Yamada, H.
(2) -196C bone's composite nature is probably also important. Future 1970, Strength of Biological Materials (F.G. Evans, ed.),
Broke very easily Broke \\lith spiral fracture
Liquid Nitrogen experiments will no doubt generate yet more questions for Williams and Wilkens Co., New York, 297.
(6)
Hand broken
-40C answering; but they will, it is hoped, answer some that have Physical Sciences Study Committee, 1965, The Physics of
All samples Spiral fracturing, flakes
(dry ice) already been posed. In conclusion, one must leave open the Lilliput, in Physics, Copp Clarke Publishing Co., 48-51.
fractured
easily detached easily and have
final deposition of the question of a prehistoric frozen bone This text makes references to G. L. Galileo, 1638,
sharp edges; flakes and scars are
technology, but one can suggest that the findings make the Discorsi e Dimonstrazioni Matematiche, Northwestern
similar to conchoidal fractures
associated with cryptocrystalline possibility a more feasible one. University Press, translated by H. Crew & A. de Salvio.
materials; t\\lo types of cleavages are Piekarski, K., 1970, Fracture of bone, Journal of Applied
(4) -25C
noted. References Physics 41, 215-223.
Two blows for one
(dry ice) Substantial amount of shatter with Robinson, R. & Elliott, S., 1957, The water content of bone,
and one for the numerous splinters, some flakin
other three
g. Anderson, D. M. & Tice, A. R., 1973, The unfrozen inte1facial Journal of Bone and Joint Surgery 38-A, 167-188.
(6) 0C water contents in frozen soils from surface area measure Tattersall, H. & Tappin, G., 1966, The work of fracture and its
All broke with one The bone produced spiral fractu
(freezer) res
very forceful blow ments, Ecological Studies 4, Springer, New York N.Y., measurement in metals, ceramics and other materials,
but would not flake. The blow
Journal of Material Science IO, 296-301.
s
required were more forceful than those 107-124.
Aoji, 0., 1959, Metrical studies on the lamellar structure on Uehira, T., 1960, On the relation between the chemical
(5) +20c
required for the lower temperature samples.
(room temperature)
All broke with one The fractures are similar to those human longbones, Kyoto Prefectural University of components and the strength of the compact bone,
of
very forceful blow froze n bone , but they break in
a Medicine 65, 941-965, in Yamada, H., 1970, Strength of Journal of the Kyoto Prefectural University of Medicine
noticeably less dynamic manner Biological Materials (F. G. Evans, ed.), Williams and 68, 923-940, in Yamada, H., 1970, Strength of Biological
and
(3) 0C
have more resilience.
Wilkens Co., New York, 297. Materials (F. G. Evans, ed.), Williams and Wilkens Co.,
T\\lo ve1y forceful These bone were very difficult
(freeze-dried) to break Bick, E. M., 1933, History and source book of Orthopaedic New York, 297.
(9 months in freezer)
blo\\ls for t\\IO of moreso than bone at room temp
erature. Surge1y. The Hospital for Joint Diseases, New York, 254. Wolff, J.,1892, Das Gesetz der Transfo1mation der Knochen,
them and one for
the third. Bonfield, W. & Grympus, M. D., 1977, Anistrophy of the Quarto, Berlin.
Young's modulus of bone, Nature 270, 453-454. Yamada, H., 1970, Strength of Biological Materials (F. G.
Bonnichsen, R., 1979, Pleistocene bone technology in the Evans, ed.), Williams and Wilkens Co., New York, 297.
Table 1.
Beiingian Refugium, Archaeological Survey of Canada
228
229
FLUME EXPERIMENTS WITH STONE AND BONE
1Archaeometry Laboratmy, 1sotrace, University of Toronto, 60 St George st., M5S 1A7 Toronto, Canada,
2Department of Anthropology, Univeristy of Toronto, Toronto, Canada,
3Anthropology Department, University of Auckland, Auckland, New Zealand
A recirculating laboratory flume was used to study the sorting and relative transport of stone and bone materials. Fifty-one
flume runs were conducted with Froude Numbers between 0.06 and 1.71. Bone and stone movement was plotted as a function
of their rate of movement in centimetres per second (cm/sec) with respect to their individual mobility numbers (U/OJ).
Low, transition and upper energy fl ow regimes show a clear separation in the rate of movement of bone and stone.
Settling velocity and drag coefficients are calculated for the material. T hese results have implications for understanding
distributions of artifactual material on some archaeological sites, and may provide a key to interpreting the potential for
movement on others. Results are applied to field problems.
Proceedings of the 31" lntemational Symposium 011 Archaeometry, Jerem, E. & T. Biro, K. (eds.), Archaeopress -Archaeolingua, BAR - Centi: Ew: Se1: I.
Pavlish, L A., Kleindienst,
. M. R. & Sheppard, P. J.
Flume experiments with stone and bone
transition velocities tially two-dimensional, and do not make allowances for the
12 28 cm fsec and 0.018 cm/sec; the
o.OOng between 0.02 cm/sec and o.44 cm/sec; and the upper fact that stone or bone in a natural environment might fan out
11 f811gJ regime velocities of 4-38 cm/sec (plane bed), from a point of origin in a wide arc, which theoretically could
D0o cm/sec (standing waves) and 20 - >100 cm/sec approach 180 degrees. The flume, however, does provide a
- 10 Z3 'du n es). Bone velocities were as slow as 0.0001 cm/sec good approximation of net downstream movements, and,
(an
E 9 for
e lower flow .regime, but were comparable with stone in thus, it simulates results that are a measure of the potential for
(..) er flow regimes. differential dispersion in a stream environment.
th upp
en 8
he data from the upper flow regime indicate that It is difficult to quantify the total movements of individual
- differential
transport velocitis of particle are trnslated ito particles. This problem is particularly acute in the lower flow
"O sort ing of the matenal by respective settlmg velocity. regime where bedform movement has a strong influence on
e
{f cti ve
e 7 the potential for particle movement and the direction of that
this regime dispersal is quite rapid, increasing with the
_g 6 movement. With the movement of ripples and ripples/dunes
hang e of bedform from plane bed through standing wave to
:; along the stream bed through an area with prop01tionately
e 5
tidune. Sorting takes place in the lower dune/ripple flow
larger stone and bone particles, these particles are alternately
"'O
egime, but at a greatly diminished rate. The lower energy
r
regime runs plotted in Figs. 3 and 4 used 99.9% of the run
..c 4
- time, but the actual rate of transport was on the average only
STANDING WAVE/ANTIDUNE
/20,000th that of the upper regime runs. This observation is
T 'IRANSffiON: PLANE BED/STANDING WAVE consistent with observed natural geological processes.
2
0 PLANE (FLAT BED) Thus, relatively instantaneous transport velocities translate STANDING WAVES AND ANTIDUNES
0 TRANSffiON: RIPPLE/PLANE BED
.A RIPPLE/DUNE
over a period of time into sorted materials. An important cave
I e NO MOVEMENT at that needs to be noted is that the flume studies are essen-
. .. .
100
.
. . 1
.
.
... . ... I
Q..__22 _J.-J..
..., 5-l--L._l.OI ---:-i_--L
_ L-.L...l___j
.L...L
I
UPPER
PLANE SEO AND STANDING WNES
50 IOO
U (mean velocity: cm /sec) FLOW . ...
REGIME ...
..
.
understanding and evaluating archaeological distrib . are not spheres: i) velocity can I
utions that
~
column do not have to follow v
a. FLUID DRAG( u,) a veitical path (Middleton & o Rl..tl" 4 TRANSITTON $ v
V RUN '11<47
'
The 0bectiv e m runrung the flume in all flow regimes was
b.
. FLOW
smtm g, entrame m nt potential and deposition 0.01
. ...
..,,. ..
of stone and T ., T o" REGIME
bone matenal within each
. .. ...
flow regime, and between the m. o;f1p" Io
.. ..,
.
.
plot of the average rate of transport of stone flakes .
Fig. 3 is ., .
as a funt10n of their respectiv ., '"' 'f'
+&lit--
232
233
Pavlis h, L. A., Kleindienst, M. R. & Sheppard , P. J.
Flume experiments with stone and bone
num ber (U.lw) resulted in a 100 l ustrnent movement as one would expe
and bone particles % incr e m obility
1961; Isaac 1966). When cultural material is exposed to the
the dista c n of te
with the pecentae
of exposure decreasing as stream velocity
npples/dunes develop deeper troughs and
traveled by a particle. The formula
permits one to alculate the average
(U.l o 25)/0.3
8
n e
se large
particles have values .near. the origi
ionship betw een
.
mobility streamflow, it will tend to be concentrated by fluvia
l
increases: 1.e., relative dis ce :::: d!D
imensionless graph where the relat select ive sortin g as a result of the remov al of
litudes. When exposed on the surface, the particle expect th particle to travel with respe one Wo d is non-linear. In these experi washi ng and
long er amp uld er and relative distance ral mater ial is
ct to the gher
m num b tends towards the host sedim ents by erosi on. As the cultu
rnust be subj et d to a current that exceeds the velocity numbers m the sample. This formu1 o bility rs that the slope [(U,/ro)/()] be dictated by
_ _ . a may be useful .
With platy me nts, it appea s, beca use of the expo sed, its own poten tial for move ment will
1mtiat e movement: the latter is termed the part!c1es t?at have settling velocities becomes smaller. In other word
required to between 7 and 25 0 as U,/(J) its own U./w. Because a parti cle's settli
ng velocity depends
veloity or critical velocity (Uc). The . Cm/sec. e large r mate rials, once
competent threshold The impl ication of the analysis is that high velocities that are required to mov actual potential for
n average, do so as on its size, density and shape, the
b
velocity can vary substantially for non little ably will cont inue to
competent threshold movement can be expected when
mo I ty numb a particle staits to move, it prob strea m location and
such as silts, sands and larger stone er s are and not trapped movement will be dicta ted by its
cohesive sediments under 0.25. A wide range of flow cond exposed at the smface of the bed
itI"ons, however
, do es Jo ng as it is that are present consequent U.
alread y noted, threshold movement has been Penm.t some movement at lower mob" stream
pa rti cles. As
lit number . by J ower energy micro-regimes in the
F hs. The
exp res sed in terms of stream velocity, bed shear stress and exmple, experiments using large stone
I m icles ra
ngi
ng
or
in bedform depressions such as dune or
antid une troug
stone Bone movement
stre am pow er. Few data are available, however, on the
required for the movement of bone and
weght from 0.5 gm to 365 gm (core) had
:o
.
b Iity num? rs th
in
at experiments show
that trapping efficacy of
ll
large
stone
r
s were
esh old condit ions vaned from 0.36 to less than 0. 05 -vis the smaller one. Sma fair measure of the
th r
.
. t veloc1t1es
of particles exists vis-a of a lai ge Mobility number still provided a
stone materials in silty, sandy and gravel bed streams. The 30-40 cm/sec and U of approximately by the eddy system m the lee
5 cm/sec ' mate n.als sometimes trapped ents. The distance
s), and movement potential of the bone elem
threshold depnds in a complex manner on both the properties with mobility numbers greater than 0 20 t in sedimentologic al te1m
stone (a shadow wake effec sets were averaged for
.
. .
riments, on average, it
maten.als approximately 10 gm to 365 gm, .
bed. Larger
Flakes that were less than element group. In these flume expe
particle sizes from 20 gm to 365 gm. er mobility numbers
did not move appears that bone elements with great
Th sand grains, aveaging 0.3 mm in diameter, that were ::: .
bly m . he downstream direction when exposed
. to 2% by weight of the trapping piece
could consistently be
will travel farther proportionally in a
quantitative relationship
us d m the flume expenments
began to move at velocities of regime (U=86.94 cm/sec).
e condit10ns for 45 minutes (U=99. 7 cmlsec
, removed from a standing wave flow
m 17.5 cm/se c to 21 cm/sec. The smaller stone particles U.=7.9). However, they d"d .
fro I undergo mmor adjustment to the
o w ed in ipient ovem nt at 25 cm/sec. This type of higher energy flow condit1ons
sh by settling into the sand This -
ent 1s asso ciated with the particle realigning itself so was a result of sand bemg
d ire ction of city for the smaller As a consequence of thi s type of movement,
some pieces
particl es with lengths measuring less than 35-40 mo:ed upstream as much as 20 cm during the 45 minute
sto ne mm
1-
achi eved at app roximately 25.5 cm/sec.' As the antidune run Block Y pieces
w as tended to settle almost vertically 0.8
rea mflow velocity was increased, the distance that an w hile platy pieces tended to move slightly upstream or
st SCAPULAE
might move when exposed also increased.
individual particle downstream (Allen 19S2).
On th e average, the maximum distance of travel per exposure Fig. 5 als shows the plot of U,/ro versus for the larger- 0.7
.
one dune wavel ength from dune trough to dune trough. s1zed matenal (symbo1 'O'). Mob1lity .
was . numbers of less than CART/LAG
A flake could cover any lesser distance. While this movement
was primarily in the dO\:nstrea direction over 179 degrees
0::
E2 0.6
LONG BONES
(LARGE)
RIBS (SMALL)
of arc, there were occasional displacements in the upstream
0.9
direction onto the face or peak of an oncoming (downstream ..J
<::(
METACARPALS
a
moving) dune. The ctual path takn by an individual particle 0.8 * t- 0.5
RIBS (LARGE) METATAfiS..LS
often apeared erra1c and unpredictable, with each exposure ::> z
. 0.7
ence on the downstream direction or w
havmg little or no mflu
..
dp/0(10+)
direction and distance implies that sorting of the materials in Cl. MANOIBLEe
::?: 0.5
CARTILAGE
the flume will take place very slowly, and need not be clearl :!I w FMORA
O)
dist nce that the md1v1ual particle moved divided by the _i.; "1 __,__o---'-
1
. -=0 7-
. I -1-'--1...
-' 0.2 0.3--'--L-'--
.L -L-L..L.
0.4 0.5 0.6 0.7__,--'---
0.8__._
-'- 0.9 _,__,
___._ _ l.O
maximum distance atamed by the particle travelling the QL--. 0.6 .L--1-..-1---Li....-.L.-...\--1-_a...--t.j,,._...l.-o.i-. u ...i-_..--'"---'--'-_,___..'---*
234
235
Pavlish, L. A " Klei
nd'1enst, M. R. & Shepp
ard, p J. Flume experiments with stone and bone
increases th
e p0tenti.al for results are based on average values, and must be
app roximation, however, as bone behaves rather differe n t} Th ese
once the bone is entirely m ovem that obtain for such values. in Geomorphology, John Wiley & Sons, New York, 360.
than stone in various flow regimes because of shap e a n .
mstantaneously realign
. par
exposed' b
a 11 e 1
.
with the stre
.
ecause It will
aimost
ent .
viewe
d subject to the caveats
vertheless, they may h e1p to present a meanm
gfu1 picture Middleton, G. V. & Southard, J. B., 1977, Bed Configurations,
density factors. which often immediately am, an a N Mechanics of Sediment Movement, Society of Economic
At U as low as 10 cm/sec some small bones and c artila g e .
At high velocities ' the larg
precedes m
ovement
ction
; the m echa
nics of movement and sorting of stone and bone
bee, are, and will be taking place as long as water
Palaeontologists and Mineralogists Short Course No. 3,
re
.
alig ned t o present the least smface area t o streamflo
w
.
for trappmg small bones
e b
a phe nome
o nes have the sam
e pote nti
al
at have
Binghampton, N. Y., 7.1-7.42.
(incipient movement). At velocities between 13 and non that was als an d sediment mteract.
'
20 with
large stone particles Bee
ause of therr
o note d Pavlish, L. A., Jopling, A. V. & Zhang, Z., 1987, Fluvial
crnfsec the remaining small bones showed in cipient Ji near nature
Thus, while fewer bones were exposed at any one time, th ese .
p oportlon that can be moved in streainflow BAR, International Series No. 352, Oxford. 86-107.
dunes generated larger troughs and thus provided a greater
through substantial archaeology: Earth Science and the Past, Duckworth,
. distance approaches 100% ! Therefore, it follow
s that sorting London, 408. Schick, K., 1992, Geoarchaeological analysis of an Acheolian
potentia1 for total bone exposure and consequent movement
of 0 atenal by size, with an almost complete to Site at Kalambo Falls, Geoarchaeology 7(1), Zambia, 1-26.
At velocities greater than 45 cm/sec, the skull would mov : removal of one Isaac, G., 1966, New Evidence from Olorgesaillie relating
fraction, wo ld be an unlikely outcome of exposu Shackley, M. L., 1975, Archaeological Sediments: a survey of
downstream at velocities that approached that of th . re to high the character of the Acheulian occupation of the site, in
analytical methods, Halsted, New York, 159.
streamf low. This element separates out so quickly that on : water conditions. In comparison with the high water
clunng 1ow water conditions, the paiticles <30 mm
conditIo ' ns,
have
Cuscoy, L. D. (eel), Actas del V Congreso Panafricano de
y ed Estudio de/ Cuaternario II, Sheppard, P. J. & Kleindienst, M. R., 1996, Technological
would never expect it to be present if sorting by water were a P rehistoria
mob 1'Jity numbers >0.25 (U./w) and >l (U/ro), and therefore change in the Earlier and Middle Stone Age of Kalambo
Publicaciones del Museo Arqueologico Santa Cruz de
f actor in a bone assemblage distribution.
are
s bject to movement ( l/1560th that of stream), while paiticles
Falls (Zambia), The African Archaeological Review 13,
Small bone elements moved downstream relatively rapid} Tenerife 6, 135-145.
sized 35-40 mm and above are immobile. Lag formation by
Jopling, A. V., 1965, Hydraulic factors controlling the shape
171-196.
with respect to other whole bone elements. They did follo
deflation may still take place.
of laminae in laborat01y deltas, Journal of Sedimentary Simons, D. B., Richardson, E. V. & Albertson, M. L., 1959,
normal movement patterns, but because of their mobilit Consequently, one can
conclude that for an almost Flume studies using medium sand (0.45 mm), U.S.
potential, they were often carried substantial distan ce
Petrology, 784-786.
complete removal of a certain size range of particles from an Geological Survey Water Supply Paper 1498-A, 17.
Kleindienst, M. R., 1961, Variability within the late Acheulian
(20-30 cm) with each exposure. Both the pelves and the assemblage to occur, that size must be mobile2 during average Stein, J. K. & Farrand, W. R. (eds.), 1985, Archaeological
?
scapulae moved su stantial distances with respect to other stream flow conditions, while larger sizes are not mobile.
assemblages in Eastern Africa, South African
Sediments in Context, Peopling of Americas Edited
Archaeological Bulletin 16(62), 35-52.
bone elements. Their plate-like morphology presents a large
Also, the percentage of particles that are removed above the Volume Series, vol. 1, Centre for the Study of Early Man,
Kraft, J. C., Kayan, I., & Erol, 0., 1980, Geomorphic
surface area to the stream currents in most attitudes. size that is totally removed is therefore a function of exposure Institute for Quaternaiy Studies, University of Maine at
reconstructions in the environs of Troy, Science 209,
Therefore, one would expect these bones rarely to be present
to above-average water conditions. Thus, the comparative Orono, Orono, Maine, 147.
776-782.
in well sorted taphonomic assemblages, but they might be
absence of a size fraction in an assemblage may serve as a Lane, E.W. & Koelzer, V. A., 1939, The relation of suspended
fo und in downstream depositional contexts. The vertebrae
. gauge for measuring average stream conditions in an area
to bed material in rivers, Transactions, American
appear the next most likely to be absent from a stream-sorted over long periods of time.
Geophysical Union 20, IV, 637-641.
collection, along with an assortment of caitilage eleme nts.
Carpals, tarsals, metacarpals, metatarsals and phalanges all
Summary
seem to move at approximately the same rate with respect to
other elements. The ribs, as noted (Fig. 6), are immobile even
This flume study attempts to bridge the gap that often exists
when 90% exposed, because of their morphology, which in
between the results gathered from research carried out in the
tum explains their immunity to relative movement a
laborat01y and the subsequent application of those results to
expresed through their mobility number. The mandibles an interpretation and solution of field problems. The detailed
humen appeai to have the same movement potential in spite
flume experiments permit some general assumptions about
of their differing mobility numbers. The femora, tibiae
stone and bone particle movement to be quantified through
fibulae, radii and ulnae all appear to have a similar potentia l the use of mobility numbers. A mobility number was derived
for movement. The movement of these longbones
. IS from a measured settling velocity of a particle and the stream
' W hen exposed totally they align themselve
charactent1c. s shear velocity or mean stream velocity. These values were
par allel with the current, but as they drop over a dune fac
' plotted against the relative distances travelled by each
they will align themselves parallel with the axis of the du : respective particle, and trends of movement potential versus
u ough which is approximately at right angles to streamflo : mobility number were quantified.
236 237
ONE AND STONE BREAKAGE FROM BOW-DRIVEN ARROWPOINTS IN
B
T WO WHITE TAILED DEER CARCASSES
'Archaeometry Laboratory, Isotrace, University of Toronto, 60 St George st., M5S IA7 Toronto, Canada,
2Department of Anthropology, University of Toronto, Toronto, Canada
Stone projectile points hafted on modem arrow shafts were shot into the left side of two white-tailed deer carcasses. T he bow
used in the experiment was calibrated using both a ballistic pendulum and high-speed photography. A 66 cm draw on the bow
was used for the experimental shots, and it produced a range of velocities of between 35 meters and 42 meters per second. T he
bone damage done to the scapulae and the ribs was recorded. Bone and stone broke with forces of between 300 and I 000 lbs
[J.3 x Id nt and 4.4 x J(J3 nt]. T he bones were cleaned and the damage was photographically catalogued. T he observed damage
to the white-tailed deer bones may provide a useful frame of reference for the interpretation of similar damage on bone recov
ered from an archaeological context and insights into ancient hunting technologies.
The experiment reported below represents a small but sig Two white-tailed deer carcasses were obtained from the
nificant part of a project which had as its goal the reconstruc Ministry of Natural Resources of Ontario and stored in the
tion of aspects of ancient technology. The objective herein was freezers of the Department of Zoology in the Ramsay W right
to quantify the velocity of stone point-tipped arrows, the forces Building, University of Toronto until used in the experiment.
of impact such arrows have when hitting bone, and the conse Archery space was provided in the IsoTrace Laboratory by
quent bone damage. The forces generated are capable of A. E. Litherland, Department of Physics, for the experiment.
destroying both bone and stone, as well as the occasional A small crane was used to suspend each deer, after thawing,
arrowshaft. The resultant damage to deer bones and stone by the forelegs. Prior to the experiment, it was observed that
projectile points was recorded. The observed damage provides the animals (FA-343-18 and FA-343-21) had multiple frac
a frame of reference for the interpretation of similar damage tures of the extremities, ribs and mandibles, primarily to the
observed on archaeological samples. This paper provides an right sides of the deer. The arrows were shot horizontally from
outline of the experimental design, the results obtained from the a measured distance from the targets, and all pertinent data
perspective of the damaged bone, and observations derived were recorded. The bones were then cleaned and the damaged
from those results. Bone is a composite material which is areas identified and described.
extremely durable having an ultimate tensile strength of
approximately 1400 lbs/in2 (100 kg/cm2). Its tensile strength is Results
one-half that of brass, twice that of teak wood, and 40 times that
of cast iron. Bone is very capable of absorbing energy from the A total of 122 shots were made at the two deer carcasses,
impact of stone projectiles and is tough enough in many using 62 different projectile points at an average distance of
instances to break or shatter the stone. Bone injuries resultant two metres. Thirty-eight percent of the shots resulted in point
from stone and metal implements used for medical purposes breakage (46). Bow draws of 30 and 45 cm did not generate
(e.g., trephination, Ortner & Putschar 1981) or weaponry (Noe forces sufficient to penetrate the animal skin, while 60 and 66
Nygaard 1974; Knowles 1983; Turner 1983) have been report ems draws did penetrate to the bone or into the chest cavity.
ed from time to time in the archaeological record. In addition, Breakage of projectile points was primarily due to contact
studies of ancient butchering practices (Guilday et al. 1962; with bone (43), but such contact did not always lead to break
Lyman 1978) and a number of replicative experiments with age as 64 shots (52%) were influenced by bone. Fibrous liga
respect to butchering techniques and their effect on bone have ments and tendons often adjacent to the bone are believed to
been carried out by a number of researchers (e.g., Stanford et arrest the arrow's movement without leaving a mark on the
al. 1981). However, there appears to be no known literature bone. Therefore, only 67% of the shots striking bone resulted
which attempts to replicate and quantify bone injury due to in the breakage of the projectile points (cf., Table 1). Bone and
stone projectiles. Thus, this study may be useful for establish stone breakage occurred with calculated forces of between
ing the ground work for fu1ther research of this kind. approximately 300 and 1000 lbs (1.3 x 103 nt - 4.4 x 103 nt)
being generated by the impact of the arrows. Table 2 gives the
Experimental design depth of penetration and associated work and rate of work
(power) done for producing the bone injuries. Bone injuries
Bow and arrow due to the impact of the arrow-points on the bones themselves
were evidenced by finding the projectile points imbedded in
The experiment was carried out using a standard 25 pound bone, indentations and scratches on the bone surface, notches
bow with modem arrow shafts hafted with stone projectile points on borders or iidges of the bones; or bevelling of the fracture
manufactured to an archaeological standard (cf., Pavlish 2001). planes inward from the direction of flight of the arrow-point.
Proceedings of the 31" lntemational Symposium on Archaeo111et1y, le rem, E. & T. Biro, K. (eds.), Archaeopress -Arclweolingua, BAR - Centi: 111: Se1: 1.
Pavlish, L. A. & Savage, H. Bone and stone breakage from bow-driven arrowpoints in two white tailed deer carcasses
left to g mass and tissue resistance between the animals (about two
. . . n ht
Table 1. The number and percentage of the 122 bow shots Two 1mbedded pomts were found m the left scap the doe's impressions or scratches that were independent of related
. . ula, and tw and a quarter times as many shots passed through
that occurred in the four non-exclusive categories: . . o
pomts m the nght nbs. Some aJTows were observ through that of the buck). (Table 4 shows that injmies. Eighteen anows caused chipping while six totally
. . ed to bounc
e body cavity as
1) stone remaining in bone; 2) bone breakage due to stone; .
back, after st nking the carcass, especially after
striking It Ill 7.5 times as many chips and notches were found on the doe's perforated the bone (e.g., scapula). Two shots shattered ribs
3) stone breakage; and, 4) bone influenced shots. . . . .
Table 3. Relative bone damage Table 4. Bone injury from arrow point damage
Table 2. Depth of penetration, time, acceleration, force, work and power for selected shots. as a function of increasing force. as a function of location.
240 24 1
Pavlish, L. A. & Savage, H.
Observations
References
EXPERIMENTAL RE SULTS
General
Baker, J. & Brothwell, D., 1 980, Animal Diseases in
Arc hae
FROM THE szAZHALOMBATTA ARCHA E OLOGI C AL PARK
1 . Based on expeii . ology (Chapter 7, Traumatic Injury), Academic
mentation, bone Press,
. damage due to hunting London. POROSZLAI, I.
proJect'l
1 es will .
be relat . .
ively rare m the archaeol
context. Archaeol
ogicaIIY related proce
ogical Guilday, J. E., Parmalee, P. W. & Tanner, D. P., 1 962, Abo
ri _
.
rarefy sue h disco ..
veries [bone InJUn .
sses will further
es from natural causes in
inal butchering techniques at the Eschelman Site (3
6 a l Matrica Miizeum, Szdzhalombatta, Gesztenyes u. 1-3., 2440, Hungmy
Bone damage pathological conditions in human skeletal remains, and firing took place together with the use of prehistoric tools. In 1997, our work enlarged into environmental reconstuction. Our
Smithsonian Contribution to Anthropology 28, 479 pp. aim is to present the natural and cultural landscape of the Middle Bronze Age and Iron Age. Our present-day climate is very similar
The types of da (trephination pp. 95-1 00). to the subatlantic climate of both prehistoric periods. This means that oak forests advanced at the expense of beech. 117e replanted
ma ge w hi ch are
observed on deer bone as a Pavlish, L. A., 2001 , Experimental Archaeology: Bow, Arrow vegetation will thus be representative of the Bronze and Iron Age landscape. The park is going to be a centre for experimental
result of these exp .
enments are as
. . follows: and Projectile Points, Paper given at the 31st International archaeology and education. The goal is not only to create a place for tourists but also an area where eve1ything is authentic and
l . The m dentmg of b . .
one surfaces, with or without retention of
.
the prJectJ1e poi . Symposium on Archaeometry, Budapest, Hungary, in this constructed on the basis of the archaeological evidence. This will be a place where we will be able to present prehistoric lifestyles,
nt tips.
2. Notching of thin volume. handicrafts and agriculture. Archaeologists will be able to carry out various experiments as well as to test their ideas and
. bone edges, e.g., .
nbs Stanford, D., Bonnichsen, R. & Morlan, R., 1 981, The Gins hypotheses derived from the results of different excavations and empirical obse1vations.
3. Creahon of frac .
tures with bevelled
242 Proceedings of the 31" Internatio11al Symposium on Archaeomet1)>, Jerem, E. & T. Biro, K. (eds.), Archaeopress-Archaeolingua, BAR- Ce11t1: Ew: Se1: 1.
Poroszlai, I. Experimental results from the Szdzhalombatta Archaeological Park
11 ilut1,, dividing wall (Fig. 5). The mud walls were ve1y thin, 5-10 cm.
On the excavation, there was no evidence for the entrance
ii&; Fig. 1. Axonometric view of reconstructed tumulus No. 115. but according to the orientation of the house we consider it
v.o ' e.--, may have been on the SW or SE side.
f$4
Prehistoric life is also introduced to the public within the 6 Two types of reconstruction could be supposed:
hectare area. Next to the reconstructed tumulus an area was - a gabled roof house
carefully chosen and enclosed by a wattle fence. In two units - a hipped roof house.
(Bronze and Iron Age) houses are being built (Fig. 4). We are In both case the basic problem was that the thin walls could
not going to build "a village" but using original ground plans hardly bear both their weight and the weight of the roof.
from excavations, constructions of houses, pits, outbuildings Consequently the post holes found outside the wall must have
Map 1. The Iron Age (Hallstatt culture) tumuli together with the Iron Age and Bronze Age fortified settlements, Szazhalombatta.
tumulus can be visited from the W-E direction to demonstrate tumulus you are on a steel bridge, under and around it the stone
that it is an alien process. The architects considered it very packing can be seen well. A 20 cm thick black painted feno
important to use the most modem material (concrete and steel) concrete shell is above the grave chamber and original grey is
to show the contrast with the original material (wood and above the conidor indicating the difference (Fig. 3).
stone). The tumulus inside is divided by a corridor, parallel to The spectacular presentation has a definite scenical order:
the original funnel shaped one (Fig. 2). The approaching first a 10-minute-long multimedia show can be enjoyed about
corridor has a sloping ramp with concrete wall to give the the site, life and bmial customs of the people living here, then
feeling of going downwards when going upwards. Entering the tumulus No. 115 is presented with sound- and light effects. Fig. 2. Inside reconstruction plan of tumulus No. 115.
244 245
aeological Park
the Szdzhalombatta Arch
Experimental results from
Poroszlai, I.
0
- - - - - - -. .-. - - - - - - - - - - - ... - - - - -- - - - - - - - - - - - - - -- -
.
0
house type was chosen. There are two mai n posts mside, the 00
lengthwise The roofing 0f th
purlin does not run to the end e
!.
. . .
tion both edge s of the house
is 180 cm. With this reconstruc
.
porary houses were about
are symmetncal. As the contem .
8-12 m, a 11 m long and 6,5 m wide house was bm1t here
(Fig. 6) Oak was used for the main posts and a.sh for other
0
'
-
. ()o
- - -
-
covered with reeds.
I - - - - -
- -
- - - - -- - - - - - - 0
. 0000 0
'. -
l
/
j
/
of house 4.
Fig. 6. Rec ontruction plan
to
-long experiences
we need 8-10 year
s were Our opinion is that
overed - e.g. tree crop.
ctions were disc count a real average
more of their fun post- holes planted again, some
of
sickles; to dig of the har vest is
or better cur ved The larger amou nt
barked with axes dles, anot her the seeds from the
ear.
ght han ing after shelling
used with long strai it is used for bak
shafthole axes were onm enta l reconstruction
chisel. The main result of
1997 is an envir
axe was used as a istoric umn 1997 and Spri
ng
and growing preh are territory. In Aut
is an area set out for sowing plan for the 6 hect nze
There and ta - istic of the Bro
bushes - character
ccon spel
seeds - Triticum dico 1998 several trees and
seeds. Both types of Age s. It was le species, oak spe
cies,
and Iron were planted: map
during the Bronze- and Iron Ages -
were widespread arm, who sent
Ancient F el, haw, sloe etc.
Ji ctor of the Butser ash.walnut, elm, com
Peter Rey nolds, dire crop ural landscape of
!'
ons how to carr y out the " sent the natur al and cult
them and gave the
instructi Our aim is to pre
m eters in 16 Age. Our present-day
ent. Within 400
square Age and Early Iron
fl growing" experim the Middle Bronze both
are sow n. The lantic climate of
350 gramm seeds/plot similar to the subat
exper imental plots, climate is very at the
ut 5 cm deep;
the forests advanced
into drills abo It means that oak
seeds are sown in
row s
man ured , prehistmic periods.
ome plot s are
rows is 30 cm. S expense of beech.
distance between the sure d ever y tw o ental reconstruction
the
growing of the whe
at is mea of the environm
som e are not. The spelta In the first phase oodl and .
the dicoccon and and create the w
height of both the original grass
weeks. The avarage goal is to replant grass
est. The h arves t is ies of the original
135 cm before the harv of the total spec
was between 55 and The development On
bronze or ston e-bl aded bush es takes 4-5 years.
with the copies of the growing of the
made by hand or takes 6-7 yea rs, fruit
and eastern border,
r the ear. st, on the northern
sickles. The stem
is cut unde
from 5,6 the edge of the fore
had some 126 kg
first year, in July 1996, we trees will be grow
n. es,
In the In one plot ses with vegetabl
means a more than
20 fold har vest. ens around the hou
kg seeds which There are small gard e wer e
entl y with in 1 cies planted her
we had 0,32 kg,
consequ lentil, pea - all spe
on 1 square meter e.g.: horse-bean,
hectare it means
3.2 tonna.
alombatta. 247
Fig. 4. P lan of the Archaeological Park in Szazh
246
Poroszlai, I.
"If he wants white pottery, as is more frequent, he uses sea-wate1: If he prefers red or bronze coloured potte1y he makes use
of well- or cistern stored-wate1:" This exce1pt from the book on the potters of Djerba by Combes & Louis (I967) makes reference
to the central theme of von der Crone (I994). By adding salt to a redfiring calcite-rich clay its su1face turns white during firing.
This bleaching phenomenon has been studied in von der Crone (I994) by using simplified "synthetic mixtures" consisting of at
lea st two up to five of the compounds illite, quartz, calcite, dolomite and hematite, considering parameters such as
mineralogical and chemical composition, quartz and calcite granulometry, the firing conditions (oxidizing atmosphere) as well
as the nature of the additives - mainly NaCl along with other salts. These results are then compared to eight natural clays taken
fmm different localities. In using XRF spectroscopy, powder diffraction, Gandolfi, electron microscopy, neutron activation and
Mossbauer spectroscopy it is shown that the bleaching obtained with complex synthetic mixtures is comparable to that of
natural clays and no supplementary firing phases occur with the natural clays. In this article some of the results from the
analysis of these natural clays are presented. The best bleaching is achieved with added halite at temperatures of 900-950C.
The drying is of great importance and it is during this process that the water soluble halite is enriched on the surface of the pot.
. i.t;; Reconstruction of the Iron Afterfiring, the suiface is composed mainly of quartz, plagioclase and pyroxene with no hematite. Depending on the Ca content,
2 ' .
. f} gehlenite may be present. The absence of hematite is due to the newly formed Ca-silicates. Howeve1; the red-coloured core of
the same sample contains considerable amounts of hematite with less pyroxene and plagioclase. The chemical composition of
the clay is the decisive factor influencing good bleaching. Clay rich in CaO permits the formation of Ca-silicates, such as
pyroxene and gehlenite which combine with the iron in their structure. Also, the purity of the white colour depends directly upon
the iron content - the more iron present, the more yellow the colour becomes. Halite accelerates mineral decomposition and
subsequent phase fonnation. Actually, gehlenite and pyroxene appear at temperatures as low as 700C. Furthermore, there is
no free CaO observed during the whole firing process. Another direct effect of halite addition, supplementary to the bleaching
effect, is the loss of K, Na and Rb in the form of chlorides.
Introduction as a mineralizer in the clay, also attack the walls of the kiln and
can even destroy them. The bleaching in connection with salt
Fig. 7 Environmental reco The most remarkable thing beside the big choice of potter or sea water has been studied by Dufumier (1982), Bearat
nstruction - P hase I.
wares in Tunisia are their colours. The eye is blinded in the (1990) and Beaiat et al. (1989). Dufumier (1982) concluded,
found in the layers of the nearby Bronze Age tell sunlight by a brilliant white. Upon looking closer it can be that the bleaching occurs only in calcite-rich clays, that it is
site or Iron Holport, A.,
Age settlement site and analysed by archaeobotanist
1993, Kora vaskori halomsfros temet6 Erd
s. seen that almost every piece has areas where the white colour only the combination of calcite and halite/sea water which lead
Szazhalombatta hataniban, in Poroszlai, I.
This reconstruction gives an impression that the (ed.), 4000 ev merges into red. Also, the inside of narrow and deep jars turns to the bleaching, that the iron content remains constant, that
Archae a 100 halom vdmsaban, 23-34.
ological Park is the small-scale reproduction of the red close to the bottom. With the aid of ceramic bodies it can the Cl-ion is the responsible factor and not the Na-ion, and that
prehistoric Holport, A., 1996, Architectural remain
landscape (Fig. 7). s in the tumulus be seen that it is just a thin layer that is actually white. The the addition of salt to clay results in a lost of K and Na during
cemetery of Szazhalombatta, in Poroszlai,
The park is going to a centre of experimenta
be I. (ed.), ceramic bodies look like a sandwich, with a white rim (on the firing. He concluded furthermore, that the glass phase
l Excaations at Szdzhalombatta I989-I995, 34-42
archaeo logy and education, a place where every . both sides) and a red core. Although the white rim looks like plays an important role. Von der Crone (1994) showed how
thing is Holp01t, A., 1999, Theoretical and practic
.
au hentJ.c and constructed on the basis of archae
ological
al problems of a slip, it is not. The white outer side and the red core are made ever, that the number of crystalline phases is higher in the
reconstruction in the case of an Iron Age
v1dence. It can be used both by teachers to teach tumulus, in of the same single step clay and were not treated differently. white rim compaied to the red core.
prehistoric Jerem, E. & Poroszlai, I. (ed.), Archaeology
lifest le, handicraft and agriculture, and by of the Bronze The only difference to complete red pottery is that the potters Based on the following observations of Fabbri & Fiori
practJ.ce what they learn from books. The park is
students to nd Iron Age, 303-309, Archaeolingua 9, Budapest. use sea water instead of rainwater or add table salt to the clay. (1986) and Schmidt (1972) it was concluded that formation of
expected to Morgos, A., 1999, The conservation plan
be used by archaeologists to carry out various expe of the Iron Age timber It can be assumed, that the potters know that the dissolved HCl is related to the reaction of halite with the clay minerals
riments as structure of the grave chamber at Szazha
well as to est their ideas and hypotheses derived
from the Jerem, E. & Poroszlai, I. (ed.), Archaeology
lombatta. in salts in the sea water are responsible for the bleaching on the at 500C, complete decomposition of calcite at temperatures
results of different excavations and empirical observa e
of the Bronz surface, but they surely have not thought about the chemical between 500C and 700C and absence of CaO dming the
tions
ince Autumn of 1996 each year our museum
organi es
and Iron Age, 335-343, Archaeolingua 9, Budap
.
Poroszlai, I., 1992, Szazhalombatta-FOldvar
est. and mineralogical processes which lead to this bleaching. whole firing process, Bearat (1990) and Bearat et al. (1989),
environmental and experimental archaeological , in Bona, I. (ed.), This trick and the amounts of salt used are based on empirical whose research is a continuation of the work of Dufumier
courses for Bronzezeit in Ungarn. Forschungen in Tell-Si
students. The paper are published. References can edlungen knowledge and tradition. (1982), assume that the alkaline-metals volatilize in the form
be found on
an Donau und Theiss, 153-155.
the Web site of the "Matiica" Museum and Archaeolo The use of salt is not only known from Tunisia but from the of chl01ides. They explain the effect as an accelerator of
gical Park Poroszlai, I., 1996, Excavations in the Bronz
Fmther information by e-mail: matlica.muzeum@m e Age eaithwork in whole of the North African area to the Orient (Persia). Brooks reactions of halite with the possible formation of CaCl2 by the
atavnet.hu.
Szazhalombatta between 1989 and 1993,
in Poroszlai, I. et al.
(1974) and Matdon (1971) write about Antique (white) reaction of HCl and calcite. The following article focuses on
(ed.), Excavations at Szazhalombatta I989-
References I995 5-15. ceramic bodies from Israel and Mesopotamia. Arnold (1971) the mineralogical reactions taking place dming firing.
Poroszlai, I., 1997, Ein archaologischer arkP in mentions potters in Yucatan who use salty clay and Rye (1976)
Szazhalombatta, Ungam, Das Altertum
Holport, A., 1985, Asatasok Szazhalombattan 1978 -1982, 43 ' 59-68. observed potters in Pakistan who added ground salt to the clay. Composition of three natural clays
Viragh, D., 1980, Varsanyi Janos, a
el6zetes jelentes, Studia Comitatensia 17. magyar regeszet els6 Salts as an additive in ceramics are not only used by aitisans
geodetaja, Geodezia es kartogrdfia 1.
but also in industry (Heimann 1989; Laird & Worcester 1956; Out of eight clays (from different localities) investigated in
Palomo et al. 1985; Schmidt 1972, 1978). Attention should be von der Crone (1994) three characteristic ones, AG, AD and
paid to the use of salts because the gases, which act positively TM 1, have been picked out for this article (Table 1). AG is
248
Proceedings of the 31" Intemational Symposium on Archaeo111et1)', Jerem, E. & T. Biro, K. (eds.), Archaeopress-Archaeolingua, BAR- Centi: Ew: Se1: I.
Experimental firing of clays using salt water
von der Crone, M. J. & Maggetti, M.
Wo Sp Ana 11
Ge Or He
- Qz An Px
AG v v v
AD TM 1 AG AD - xx
TM 1
Illite 60 65 28 32 40 Si02 74.50 55.96 -
-A. G 1 xxxxx
x
v v
60. 26 xxxxx
AG 2 v
Chlorite 0 x
0 6 Al20 3 17.02 14.16 xxxx x
12.70 AG 3 v
v v x
Kaoli nite 35-40 60- 65 6 Fe203* 2.00 xxxx xx
5.93
4.69 AG 4 v x
x x x
xxxx xxx
Montmorillonite 0 5-10 48 Cao 0.24 17.92 AG 5 v x
14.27 xx xxx x
xxx xx
AG 6
MgO 1.29 2.30 xxx x
4.61 xxxx x
AG 7
Quartz 40 20 28 K20+ Na20 4.20 2.64 5.6 4
. T he relative quantities of the dif.fe
ren phases
2 - AG 7, all fired at 95ooc.
Carbonates 0 25 29 C0 2 0.27 7.63 12.48 of AG and the mixt ures AG
Table 3. XRD-analyse s
- t-<., An Anorthite, Px Pyroxene, Ge = Gehlemte,
. . .v = l/2x . Qz - Quar .
=
Feldspar <5 < 10 5 LOI 4.30 10.60 the Iie1ghts .r
OJ the pe aks .
=
- atite, Wo m
0r- Orthoclase ' He - Hem
=
Table 1. Mineralogical and chemical composition1 of the three clays (AG, AD, TM 1). association present in the experimental
theoretical association
Only selected major elements listed. *Fe101 exper. specimen (x-ray diffraction)
Fig. 2
specim. Fig. 1
from the clay quarry Aga, 1 km to the east of Kleinaga Clay AG Qz + Sp+ [11] + [Or] + [Ana?]
Qz/Mu
AG 1 Qz+Mu
(Thi.iringen, Germany). It is a tertiary clay of the "Luckenauer Qz + Sp+ [Or] + [Ana?]
Qz/Mu
Tonhorizont" with a lot of silt and fine sand. AG is very rich Diffraction analyses of the ceramic body of the unaltered AG 2 Qz+ Mu
Qz+An (+He)+ [Or]
Qz/Mu+Di
in qumtz and contains no carbonates. The clay minerals are clay AG show quartz, orthoclase, spine!, anatase(?) and illite AG 3 Qz+ Mu +An
Qz+An (+Px+ Ge+He)+ [Or]
illite 2/3 and kaolinite 1/3. The iron content is low. The firing (relictic phase) (Table 3). The addition of halite (AG 2) leads Qz/Mu+An+ Di
AG 4 Qz+An e)+ [Or]
colour (without any additives at 950C) of AG is light orange. to a change of colour at the surface from orange to deep red. Qz+An+ Px+ Ge+ Wo (+H
An+ Di+ Qz
AG 5 An+ Di/Wo + Qz
AD is from Tunisia from the hills beyond Guellala (isle of The core is more lightly coloured. Hematite, which is Qz+ Ge+An+ Px +Wo (+He)+ [Or]
An+ Ge/La+ Di
Djerba) and displays the typical salmon colour of these clays. considered to be responsible for this change, appears only AG 6 An+ Di/Wo +(Ge) Qz + Px +An+ [Or]
Di
AD contains about 25% carbonates. The clay minerals are when calcite is added (AG 3), whereas the peaks of spine!, AG 7 An+ Di/Wo +(Ge)
kaolinite, illite and montmorillonite. The iron content is anatase and illite disappear. More calcite leads to the
medium. The firing colour (without any additives at 950C) is formation of plagioclase (AG 4) and then pyroxene and ons of AG 1-7.
ed and theoretical phase associati
red. TM 1 is a mixture of clays, dolomitic marls and sand gehlenite (AG 5). The colour in the core remains orange-red Table 4. Experimentally detemiin
. 2 the different mixtures of
stones from the upper "Siisswassermolasse" from the Grube but the surface turns white - at first partially, then the whole pose. Wl thin the system of Fig. .
erature of 95ooc is not high e
Mettlen (Thurgau, Switzerland) and a rnicaceous sand from phase, it is assumed that the temp AG 3 and AG 4 contam som
of the experimental specimen is covered (AG 5). AG describe the same path.
Weinfelden (Switzerland) and is used by ZZ Ziegeleien Plagioclase, pyroxene and gehlenite are present in the core enough. . .
. metastable gehlenite. . .
Ziirich for making tiles. The content of clay minerals is about and in the white rim. Hematite is found only in the core, The addition of calcite to AG
makes its compositlon n_io:e
The diffe renc e betw een AG
6 and AG 7 is clea r y. vlSlble
35%, mainly composed of montmorillonite and illite. It from the j oining line of quartz-m
ullite to the aO-pole :'ithin ralo gica l com posit10n. G
nepheline only in the white rim. In AG 6 the rim becomes to the naked eye and in the mine
mm) but turns a little yellow because of the the syst em Of Fig 1 From AG 3 on there is form
at10n of
whe reas in AG 6 ge leite,
contains about 30% of carbonates (calcite and dolomite). The thicker (about 1 . .
7 cont ains only pyro xene ,
5
maximum m G 5. From AG
oth ssociat10ns
iron content is medium. The firing colour (without any anorthite. Anorthite reaches a pres ent.
higher Fe content. Now gehlenite is the dominant phase .
thite anor thite and pyro xene are
in the ubtnangle anor
additives at 950C) is red. to AG 6 (the mixture is then um situat10n (Fig . 2).
besides quartz; pyroxene increases and plagioclase decreases roug hly represent the equi libri
gehl enite ) anorthite and quartz decom-
compared to AG 5. The yellow from AG 6 can be turned back diop side/wollastonite-
Results of the natural clays to white by replacing calcite with dolomite (AG 7). In this
after firing case more pyroxene is formed, gehlenite and wollastonite are e AG
no longer present and plagioclase has strongly decreased.
e AD
All three clays were investigated with and without halite Table 4 shows the experimentally determined and the
(Table 2). AG, which has a very simple mineralogical theoretical phase associations of AG 1-7. o TMl
composition (no carbonates), was studied by admixing According to the chemical composition of AG 1 and AG 2,
calcite, dolomite and hematite in addition to halite. If no other spine! does not represent an equilibrium phase (Fig. 1). Its
indications are given, the samples were fired at 950C with a formation can be explained by local illite-illite points of
1 hour holding time. contact. For mullite, which would represent an equilibrium
Forsterite
f rent admixtures to this clay and su1face colour after firing (Munsell soil colour charts).
Table 2. Shows the dife Fig. 2. Mixtures AG 1-AG 7, AD and TM 1 in tl e hase
Fig. 1. Mixtures AG 1-AG 7, AD and TM 1 in th phase
.
diagram after Osborn & Muan (1960). Abb e 1at10ns:
1 The values of the oxides (Si02 - K,O in wt%) are presented without the LOI. Results of the analysis of the clay minerals in relative quantities (wt%), quartz,
diagram after Osborn et al. ( 1954). Abb1e ia wns: see Table 3, La = Larnite, Ra Rankuute.
:
=
251
250
Experimental firing of clays using salt water
von der Crone, M. J. & Maggetti, M.
ses
occurs first at temperatures around 750-800C and is best 1.1 and TM 1.3 is shown in Fig. 4. P m T TM l, AG 7 fits best with
. yroxene
cr CaM gosFeo,;Alo.sSi 1.sO6 The distribution of . .
around 900-950C. Diffraction analyses were made of the drastically from TM l to TM 1.1. The diffe m ea ses sized by
. . . rence betwee fassait (ASTM 25-1217) or a pure diops de synthe .
specimens at 650C, 750C and 950C (see Table 5). At n T M
I. I to TM 1.3 1s low, that is, hahte reaches its a low iron content which
Kreimeyer. Both pyroxenes indicate
max1ma1 eff
650C calcite is mostly decomposed. Firing phases are mainly ect
at around 1 wt%. (compared to the other
is effectively the case in AG 7
presented by anorthite, pyroxene and subordinated wollas
pyroxene bearing samples).
tonite and gehlenite, where gehlenite only occurs in the core. Correlai?n between chemica
The same is true for hematite. compos1t1on and firing colourl Conclusions
The projection of AD falls within the system shown in
Fig. l in the subtriangle anorthite-quartz-diopside/wollas In the Figs. 5 and 6 the opticl observations
(colour) are These investigations have shown that bleaching is possible
. . .
tonite and in the system shown in Fig. 2 in the subtriangle con-elated to the chenucal composit10n. It is valid
for all eight with all types of clays by adding halite (provided that thy
. .
anorthite-diopside-gehlenite/larnite, both times close to the clays and the "synth'
etic nuxtures"from von der Cron
e (1994) contain carbonates). A content of about 1 wt% f hal 1te is
joining line of anorthite-pyroxene. That means that the .
The second diagram in particular shows that the pure most favourable. It is already sufficient to obtam mmeral
observed association at 950C fits the theoretical association. the white colour depends directly upon the iron content.
ness o f
associations close to equilibrium situation at 95ooc, where s
I.I
3. Experimental specimens of TMI, TM
Fig. a non equilibrium situation (Magetti 198 6) is normal at this
Qz Cc II/Mont An Px Ge Wo 95oo c. Beca use of the thick ness . .
Or He and TM 1.3, fired at temperature. The observation on natural clays contammg salt
the sample the outer zone of TM I is oxidi ze more
xxxxx of or halite can be summarized as follows:
650C xx xx v xx x v
than the core and contains more hema
tite. By addmg 1 to/o
.
750C xxxxx v white and contams
een rim and core becon_ies
x x x outer zone turns
v of halite (TM I.I) the
- The difference in colour betw
x
_ e (TM I.3, 3 wto/o)
halll
to processe occumng
addin g more
no more hematite. By is due
visible with firing, but it
950C xxxx xxx xxx v v xx
this zone becomes thicke1:
mula tion of salts m the outer
during the drying stage (accu
zone).
g at 650C 1 a dryg
Best results are achieved by dryin
f rent phases are
Table 5. XRD-analyses of AG (+ 1 wto/oNaCl) at different temperatures. The relative quantities of the dife
small amount of a com-
.
edral coordination wheeas
the unfired clay AD, the core and the rim of AD (fired at atom s coexist in symetric tetrah n of the white nm consists
ination. It is not possible - The mineralogical compositio
All experimental specimens of TM I show zonation after 950C with 1 wt% halite) were investigated by Mossbauer the others display octra hedra l coord
firing (Fig. 3). In TM l (without halite) the zone close to the spectroscopy. Fig. 7 contains the different spectrums.
Pyroxene Gehlen ite
surface (3-5 mm) is red, the core light brown. The boundary Plagioclase
AD contains 5.2% Fe,0,. Of these 5.7% is hematite.
between core and rim is sharp. Despite this optical difference, Hematite is already indicated in the room temperature 501-------i
15------
25-----i
mineralogical (XRD in Table 6) and chemical analyses (XRF) spectrum (RT spectrum) with a weak six line spectrum (Fig. TM TM
TM
of core and rim are identical. 7a). The abundant ferrihydrite is indicated in the RT-spectrum
'$
Three zones are visible in the cross section of TM 1.1 and in the Fe3+ component with large quadrupol splitting and is 20 40
TM 1.3. From the surface to the inside a yellowish zone well visible in the 4.2 K spectrum (Fig. 7b) as a six line
10
follows a light red zone and then the greenish grey core. The spectrum with a smaller magnetic field.
transition area of the outer zones is diffuse. In TM 1.1 the The ferrihydrite is therefore present in form of super
light red zone is much thicker than the yellowish, while in TM paramagnetic particles with a size < 100 A. The Fe3+ doublet
1.3 it is simply inverted. The change from these outer zones to exists in the RT-spectrum as an overlying of several
the core is sharp as in TM 1. The zoning of these experimental components, while in the 4.2 K spectrum it is mainly due to the
specimens can be explained by their thickness (which is more clay minerals (based on the small quadrupol splitting
than twice that of the others) and the influence of the gas characteristic for clay minerals). The ferrihydrite contributes
phase during the firing. At the surface the oxidizing about 35% of the whole iron, the clay minerals at least 55 %.
atmosphere predominates and leads to the formation of Furthermore, there is a low amount of a component with Fe2+.
hematite, while the different atmosphere (guided by the gases The existence of hematite (12.9%) in the core of the fired
1.3 I. I 1.3
of the reacting phases) in the core does not allow the I.I
specimen (Fig. 7c) is remarkable, while it is missing in the I I.I 1.3
formation of h ematite. The Al203 content of TM 1 is 12. 6%. RT-spectrum as well as in the 4.2 K spectmm of the rim I.3, fired at 950C.
em't.e m T'M I, T'M I I and TM
.FP lagwclase, pyroxene and gehi
This is important because the coexistent field of anorthite Fig. 4. Quantitative analysis OJ
(Fig. 7d and e). Whereas all the iron in the core exists as
Qz An h & ili
600C v x v v
800C 900C 950C 1000c l050C x v v
xxxxx x x
TMl v
TM 1 6,25YR 6/6 x v
5YR 5,5/6 5 YR 6/6 5 YR 6,5/6 5YR 6,5/6 7,5YR 7/5 xxx xxx v v
TM 1.1 xxxxx
TM 1.1 7,5YR 6,5/5 v
5YR 7/6 lOYR 8/4 2,5 y 8/3,5 xxxxx xxx xxxx
TM 1.3
TM 1.3 7,5YR 6,5/5 5YR 7/6 2,5 y 8/3 2,5 y 8/4
ive quantities of the different
. /i I and 3 wtm 'J halite' fired at 950C. The relat
70 0.F .
'J TM I'. TM I wit
Table 7. XRD-anal yses 0F viations as m Lable 3'
r
. F t11e peaks v - l/2x Same abbre
heights OJ
Table 6. Firing temperatures of TM 1 and colour of the fired experimental specimens (suiface).
=Penclase, L1 =L une.
La=Lamite, Le=Leucite, Per
TM I.I= TM 1 + lo/oNaCl, TM 1.3 =TM I+ 3%NaCl.
253
252
van der Crone, M. J. & Ma
ggetti, M.
Experimental firing of clays using salt water
Fe2o3
0 HL Proceedings of the International Clay Conference
mainly of pyroxene, plagioclase and quartz (subordinated
0 AG Denver, not paged (first proofs).
0 BL nepheline, wollastonite, orthoclase).
90 O GI hematite - is bound in Heimann, R. B., 1989, Assessing the technology of ancient
0 AG _ The iron - in the rim not present as
6 NE 20 is white, although it contains Fe. pottery: the use of ceramic phase diagrams, Archeo
80 0 GI py roxene pyroxene
80 + PN, AD, DJ, 6 NE It is possible to predict bleaching if the chemical or materials 3, 2, 123-148.
_
TM I, TM2 0 /';
Kreimeyer, R., 1985, Die Abhangigkeit der Brennfarbe von
red 40
+ PN, AD, mineralogical composition of the clay is known.
D J,
10 <:no 60 TM I, TM Depending on the iron content whether the bleaching may Ziegeln von der Zusammensetzung der Ausgangs
0 70 red 0 2 _
ei/
+/';
result in a white or yellow colour. An iron content of more substanz, 5th Meeting of the European Clay Groups,
0
60 o than 5-6 wt% Fe,01 leads to yellow. Konka, J. (ed.), Charles University, 515-520.
I 60 40
white-red/ yellow-red _ The optimal firing temperature for good bleaching is Laird, R. T. & Worcester, M., 1956, The inhibiting of lime
I between 850-950C. blowing, Transactions of the British Ceramic Society 55,
50 yellow 545-563.
20
References Maggetti, M., 1986, Majolika aus Mexico - ein archaomet
0 white risches Fallbeispiel, Fortsc/11: Mina 64, 1, 87-103.
CaO+MgO 20 40 Arnold, D. E., 1971, Ethnomineralogy of Ticul, Yucatan Matdon, F. R., 1971, A study of temperatures used in firing
Cao 60 80
Mg O potters: Etics and Emics, American Antiquity 36(1), ancient Mesopotamian pottery, in Science and Archae
20-40. ology, Brill, R.H. (ed.), 65-79, MIT Press, Cambridge,
Fig. 5. D agram (Si02+Al201)-(CaO+MgO)-Fe201. The Bearat, H., 1990, Etude de quelques alterations physico Mass.
Fig. 6. Diagram Fe20rCa
boundmy !mes are based on the mixtures with eight natural . O-MgO. A bleach'
mg to pure chimiques des ceramiques archeologiques, These de Noller, R., 1983, Untersuchungen zur Farbigkeit oxidierend
wh 'lte is only possible with
clays and some "synthetic mixtures" a low iron conte gebrannter Tonkeramik, Dissertation der Freien
nt. doctoral, Univ. de Caen, 162 p.
(see von der Crone 1994).
Bearat,H., Dufournier, D., Nguyen, N. & Raveau, B., 1989, Universitat Berlin, 165 p.
Influence de NaCl sur la couleur et la composition Osborn, E. F., Devries, R. C., Gee, K. H. & Kraner, H. M.,
a)
chimique des pates ceramiques calcaires au cours de leur 1954, Trans. AIME 200, 38-39.
cuisson, Revue d'Archeometrie 13, 43-53. Osborn, E. F. & Muan, A., 1960, Phase equilibrium diagrams
Brooks, D., Bieber, A. M., Jr., Harbottle, G. & Sayre, E. V., of oxide systems, plate l, American Ceramic Soc. and
1974, Biblical Studies through Activation Analysis of Edward Orton, Jr., Ceramic Foundation.
Ancient Pottery, in Archaeological Chemistry, Beck, C. Palomo, A., Vazquez, T., Blanco-Varela, M. T. & Puertas, F.,
W. (ed.), Advances in Chemistry, series 138, American 1985, The mineralizer effect of fluorspar in relation to
Chemical Society, 48-80. the form of addition to an industrial raw mix, Tonind.
AD Combes, J. L. & Louis, A., 1967, Les potiers de Djerba, Ztg. 109, 10, 752-755.
unfired
96 RT Publication du Centre des Arts et Traditions Populaires Rye, 0. S., 1976, Keeping your temper under control, Arch. &
d)
e)
10
10
I
I
i
I
I
I
93 AD Rim
RT
95 AD Rim
4.2 K
-6 0
+6 -10 -- ---1
0 +10
Velocity (mm/s)
254
255
CHEMICAL AND PETROGRAPHIC DISTINCTION BETWEEN
THREE MEDIEVAL BRICK PRODUCTIONS IN SWITZERLAND
WOLF, S.
The bricks from St. Urban, a Cistercian monastery in the Canton of Lucerne became subject of archaeological and
archaeometric inqui1y because of their unusual size (45x30x25 cm) and their extraordinmy decoration. From decoration
themes it was assumed that apart from the manufactory at St. Urban two other production sites at Fraubrunnen and Frienisberg
(Canton of Bern) existed. Chemical and petrographic analysis of bricks from the three sites has been carried out to address the
question of whether the bricks found in Fraubrunnen and Frienisberg were made in St. Urban, or if these monasteries had their
1 tion
own production sites. The combination of the results of chemical and quantitative image analyses and arts-historic inf0111a
on the decoration indicates the existence of three individual manufactories.
FRANCE
SWITZERLAND
North
t,Okm
ITALY
0
1
-- 1
Fig. I. Map showing the geographical and geological locations of the three Cistercian monasteries discussed in this study.
Fr: Fraubrunnen, F: Frienisberg, U: St. Urban. Grey area indicates distribution
of the Lower Sweet Water Molasse in Switzerland.
Proceedings of the 31" lntemationa/ Symposium on Archaeometl)', Jerem, E. & T. Biro, K. (eds.). Archaeopress-Archaeolingua, BAR-Centi: Ew: Se1: I.
Chemical and petrographic distinction between three Medieval brick productions in Switzerland
Wolf, S.
Qtz
Analytical methods Geological situation
1400
Sampling
.
Th geograp cal nd geological situation of the
studied 1200
.
sites is shown m Fig. 1. The three sites are situ
ated n
A total of 86 biick samples from St. Urban (n=51), sedimentary deposits (Lower Sweet Water Molasse)
of t .-21000
Fraubnmnen (n=20) and Fiienisberg (n=l5) were selected for same geological unit (Letsch 1907). The geologica
chemical, mineralogical and petrographic analyses. The major comprises red clays, silts and sandstones (Keller et al.
l uni Ul
c
Q)
Qtz
199Q c 800
-
ity of the bricks are part of Swiss Museum collections in Basel, Muller et al. 1984; Peters 1949). The sandstones
ar Q)
Bern, Lucerne and Zurich. Samples of 15 to 20 gm were needed composed of quartz sands and granite rock fragments;
>
the ,p
ro 600
for full chemical and mineralogical analysis and thin section. clays consist mainly of illite, muscovite, chlorite and an iron
0:::
hydroxide (Wolf in prep.). We assume that the red illitic clay 400
s
Microscopy and quantitative image analysis and the sandstones represent the raw material for the bricks.
200
A great number of thin sections (n=86) of the three sites has Results and discussion
been studied in detail under a polarising microscope. A careful of petrographic, mineralogical
selection of 13 thin sections, which are representative for each and chemical characterisation . -....,_ -
-. .... ... ...._.
.-... __
-
.. 20 25 30
-...
-
- -- .._,
-
....:... .:._...
- .
.
2 '.- . , -- , _ 111
-!'
five areas (6.88 mm ) were analysed per thin section. The section (refer to Maggetti 1982) we can distinguish between I - ,, .. .
..
.. Fig. 3. X-ray powder diffraction pattern of three brick
images were processed using digitised analysing software a fine grained matrix (grain diameter <0.015 mm) and non
:t .-
. -.- ..
-
... -
.
samples. The diffraction patterns reveal similar phase
-
. -
.. ... "
(IMAGENIA, Biocom, Lyon). plastic inclusions (>0.015 mm). The bricks show a red - ... 116:
-
of Fribourg, Switzerland). A step angle of 0.02 29 and a Small muscovite particles (0.015-0.1 mm) are hardly
. . .. . .... ....
.... .
... _,
...
-
i -'!'
, - -
. - ..
-
-- trend, whereas Ti0/Al203 (Fig. 4b), Cr+
Ni/MgO (Fig. c)
measuring time of 1 sec./step was chosen for analysis. observed. We determined the abundance of grains, i.e. the - ...., positive con-elauon
and Na,O + KzOI Fe203 (Fig. 4f) show
, 111'
.
-
. .... .... -
and most of the
. - --
ratio of the surface occupied by the grains to the total
trends. No con-elation has been found for CaO
..
surface, and the grain size distribution. Following Delesse's er, the con-elatin
trace elements (Rb, Sr V, Y, Zn, Zr). Howev
Chemical analysis ..
'
,1..
of non-plastic
.
Principle the brick's temper proportion ( vol%) equals the
can be ascribed to varying proportions
.. . I
Chemical compositions were determined by analysing ratio between the cumulative surface of the grains and the _.....,., - .. . "' - .
.. ..
inclusions as already confirmed by quanti
tative thin section
major (Si, Ti, Al, Fe, Mn, Mg, Ca, Na, K, P) and trace element total surface of the analysed area (De Hoff & Rhines 1972).
-
. .
.
.. .....
analyses. Furthermore, referring to the ge . logical si ation,
(Ba, Cr, Cu, Ni, Pb, Rb, Sr, V, Y, Zn, Zr) concentrations using Digital images (Fig. 2a-2c) of parts of typical thin section
I '
high roportion f
high silica concentrations are related to
,
,.
-
-
X-ray fluorescence (Philips PW 2400, Rh-cathode) at the (area=6.88 mm2) of each site visualise the petrographic
.., .. -
"
,,
silica-rich minerals like quartz and feldsp
ars. High values m
Institute of Mineralogy, University of Fribourg, Switzerland. characteristics of each sample group. Referring to
...
..
...
.,
aluminium, magnesium, iron, potassium,
.
chromium ad
The calibration included 39 standard compositions for major quantitative analysis (Table 1) for a total of 13 representative - -
nickel, which represent the composition of
the initially plastic
._.,
t1 .
.
elements, and 20 for trace elements. For sample preparation non-plastic trae
thin sections the three sites show remarkable variations of
matrix (illite, muscovite, chlorite, biotite and
refer to the PhD thesis of Zanco (1998). proportions of non-plastic inclusions and matrix: St. Urban tion f non-pls uc
.. ':" .. .. . minerals), are related to a low propor . .
ted m a high silica
samples have a high (about 24 vol%) content of non-plastic
inclusions. In this respect the sites are separa
inclusions. The bricks from Fraubnmnen have a
- '
aluminium -
low aluminium (St. Urban) group and a high
Data treatment ..
- as lso
.. low silica (Fraubrunnen) group. Frienisberg
!.
.
.. ..
Chemical data were analysed using statistical programs whereas Frienisberg samples with about 10 vol% lie tative results - an interm ediate compo sit10n.
quanti
with
available under the statistical package SY STAT (version 5.2). between Fraubrunnen and St. Urban.
. . " ... ... , ... shown
Rb/K20 (Fig. 4d
By looking at the scatter plots CaO/MgO and . .
_
The raw data were recalculated by x-m/cr, where x is the given Mineralogical analysis: X-ray diffraction analyses ed as outliers, Le.
and 4e) three samples can be identifi
percentage of a constituent, cr is the average and s is the showed that the bricks have the same phase composition at (112) and the
ungrouped samples: the Fraubrunnen
standard deviation of the group (Picon 1984). This treatment the three sites (Fig. 3). We identified quartz, plagioclase,
rnrn 3 by their ele:a ed
Frienisberg (139) samples are characterised .
was necessary to account for concentration differences between potassium feldspar, hematite and spinel. proport10ns of illit
Fig. 2. Digital images of a representative thin section K,O and Rb concentrations. Different
the chemical constituents. Cluster analysis uses the average Chemical analysis: Table 2 contains the analyses (average , Fez03) can explam
no. 7: St. Urban (a); no. 105: Fraubrunnen (b); mscovite (K,0, Rb) and chlorite (MgO
linkage of the mean Euclidean distances matrix which is and standard deviation) of the bricks from Fraubrunnen, one in sample
no. 116: Frienisberg (c). their "outlier'; position. The presence of limest .
defined by the measured concentration values of 18 (Si, Ti, Al, Si02 and
Frienisberg and St. Urban. All the samples are aluminium 142 (Frienisberg) can explain its lower
Fe, Mn, Mg, Ca, Na, K, P, Ba, Cr, Cu, Ni, Rb, Sr, V, Y, Zn, Zr) silicate iich bricks characteiised by a silica (Si02) content Al,03 values (Fig. 4a) and CaO concentratio
n that is
of
elements. The probability of right assignment was tested by ranging from 60 to 75 wt%, aluminium (Al203) concentrations no. of thm images per total counts vol% dobble the values of the rest of Frieni sberg samples
259
productions in Switzerland
ction between three Medieval brick
Chemical and petrographic distin
Wolf. S.
5
142 0 112
Fraubrunnen St. Urban
. Frienisberg St. Urban
Conclusions thirdly, the three sample groups foim linear trends m 51 outliers 0139
. scatter 15 20
d iagrams. The assumption that the biicks of the thre . n ()
() average
6)
e sites average () average
A couple of observations and analyses (Wolf in prep.) were produced from the same raw materia1s renders
. . the
suggest that the source of raw material of the three sites was di stmction between the possible production sites di 63.85 2.15 70.29 2.10
fficult. Si02 64.46 1.47
extracted from the same lithological units: firstly, the three Moreover, we have no indication (kiln or waste pie .
0.79 0.03 0.64 0.06 N' o
. . ces of Ti02 0.73 0.03 - Frienisberg
monasteries are all located on Lower Sweet Water Molasse nc
b k ) f
. .
s or a bnck production site at Fraubrunnen
or AlPJ 17.70 0.69 18.13 0.84 15.81 l.03
deposits C<
eller et al. 1990; Muller et al. 1984), secondly, the F nernsberg. Nevertheless, quantitative analysis and che .
. . ffilcal FeP3
' 6.40 0.30 6.84 0.49 4.98 0.60 0
.
composition of the bricks evinced variations in matrix- 0.08 0.02
nuneralogical composition of the bricks is the same and 0.15 0.01
MnO 0.13 0.01 u
3.02 0.36 l.99 0.28
MgO 2.89 0.18
80 Friertisberg (n=15) CaO l.78 l.24 1.39 0.67 1.13 0.50 -5-
l.76 0.19
a) 6 Fraubrunnen (n=20) b) NaP l.66 0.14 l.78 0.12
3.20 0.28
+ St Urban (n=51) 112 4.10 0.24 3.90 0.23
75. 0.9 KP
0.15 0.02 0.10 0.02
PPs 0.14 0.06
:j:\., 100 100
---l
-70
lat* -
0.8
A sum 100
- 1 0+--+-....
...
0
.... +
0
0.7 + + Ba 449 33 420 17 431 19 -10 -5 0 5 10 15
65
- - +ittt
11 153 14 118 12
FACTOR(l)
illlt * Cr 150
C'I
A 0).6 * -t
142
Cu 39 3 50 14 28 5
O
..... ro. 142 A ..... \f+ Ni 78 5 83 11 54 7
112 ++ 163 11 Fig. 5. Results of discriminant analysis using Si, Ti, Al, Fe,
CJ) 0.5 167 10 178 9
Rb Mn, Mg, Ca, Na, K, P, Cl; Cu, Ni, Rb, Sr, Y, V, Zn, Z1:
93 11 103 31
55. Sr 95 20
0.4 34 2 30 4 Fraubrunnen (n=19), Frienisberg (n=13), St.Urban (n=51)
Pb 69 52
134 8 110 15 sites and outliers (n=3) are clearly distinguished.
v 129 12
so 3
2 40 4 31
10 15 20 25 10 25 y 34
12 osters St.
Zn 113 7 122 10 95
Maurer, C., 1998, Die Backsteine des Zisterzienserkl
Al203 (wt%) 17 177 19 ag am "Greifswalder
Urban im Kanton Luzem, Vortr
Zr 156 5 168
fswa ld, Germany, 16.-
Backsteinkolloquium 1998, Grei
300 Fe20/ total iron as Fe3+ 19. 4. 1998.
c) 139 6 d)
=
79-88.
A
- s from St. Urban: analytical
-
139
. 139
1972, Microscopie Wolf, S., in preparation, T he brick
'2J.
.... 6
A Hoff, R. T. & Rhines, F. N.,
Cistercian bricks in
and technical investigation on
De
-
s
-
- + fln' ff "" quantitative, Masson editions, Paris
, 147-178.
Switzerland, PhD thesis, Univ ersity of Fribourg, CH-
St. Urban, Bericht der
-
Keller, B., Blasi, H.-R., Platt, N . Western Switzerland,
Terra Sigillata Imitations from
-
F e203 tot (wt%) Geologie der Schweiz, Geote chnische Serie 4, 159-160.
Fig. 4. Scatter plots of the three discussed groups (n= 86): 4a) SiO/ 1 + N'111'M
IA[2o3 4 c) c.
- Al2O3' 4b) TiO:f1'1 . g 0,
4d) CaO/Mg O' 4 e) Rb!'K2o 4f) Na20 + K20/Fe203 tot) (for discussion see text) (group symbols as in 4a).
'
261
260
General Archaeometry
The paper presents the applicability and effectiveness of structure analysis in archaeometry. As a consequence of structure
analysis of the Holy Crown of Hungmy, it is demonstrated that the Crown is a uniformly designed and produced unity as a whole.
Plenty of hypotheses exists about the origin of ancient and from these the measures (width, heigth, diameter etc.)
remnants. In some cases analysing the structure helps us to in mm (see Table 2).
choose the correct version among them.
Measured Angles Coordinates Remarks
spot
Application of structure analysis (degree,minute,second) (mm)
Hor. Vert. Hor. Ven.
exactness, possibility to keep the method under strict A, 27"44,2' 9021,1 90,4 -4,l Y.
A., 3158,9' 9018,0' +116,4 -1,6 Y.
control,
H" 29"18,0' 8919,9' -13,8 +44,l Y,
- no starting suppositions will be needed, H,, 30"24,0' 89"19,9' +38,0 +44,l Y,
- all the characteristics can be measured without any direct v, 90"19,4' -2,7 Y.
contact with the (generally) easily destroyable object using v, 9123,6' -53,l Y,
s, 2907,0' 88"56,5' 23,2 +64,5 Y .CO
remote sensing technics (e.g. geodetical methods, stereo
s, 30"37,7' 88"52,0' +50,4 +68,2 Y .CO
photography, laboratory version of aerial and satellite -----..------------- ----
--
Proceedings of the 31" lntemationa/ Symposium on Archaeomet1)>, Jerem, E. & T. Biro, K. (eds.), Archaeopress-Archaeolingua, BAR- Centi: Eiu: Se1: I.
Ferencz, Cs., Ferencz-Arkos, I. & Fehe1; A. Application of structure analysis in archaeomeny
measure of bands without omamentics 50,4 $ 51,5 1,98..$ 2,03 Full height 127,0 5 l27,0-lZ7 .5 -0
Accuracy of scattering of these measures: 0.5 - 3.5 the cross, from base
%. Lenght of the cross
Side lenght of the upper Pantokrator
Consequence: The Crown is a precise handmade object. from the base- -56.S < 60,0-(2,22 < 2,36) -- meaningless
plaque [15, etc.] -50 -l,97
Distribution of the base-measure: -rim (not integer)
Frontal Pantokrator plaque
Lower part (corona graeca): cca. 48.7 - 52 mm. height:$ 48,7 $1,92
(inclined cca. 16' backward) Front-back diameter 215,9 8,5 215,9-216,8 -0
Upper part (corona latina): cca. 49 - 51.8 mm. together with width:::; 49,7 $1,96
[The "ve1tical" height of the cross: 52.3 nun.) mounting
The base-measure is common with very high accuracy. This - the bead-Wire fomting a rim Table 4. Analysis of the enframing sizes of the Holy Crown.
covers the band sufficiently ;::52 ;::2,05
means that there must have been one "etalon" for cutting these
accurately ax:,;:,31t11:26
accessories (ring, bands, Christ-plaques etc.) for both parts -49 1 93
of Apostole-p!aques on the
-
Height ,
circumterence
(Table 3). bands (characteristic, [15, etc.D
Distance of the small golden
B) Main measures of the Crown loops on the ring, holding -;2 > 50 -2,05 > 1,97
the strings of small pearls
Fig. la. Measured points on the Crown. - Frontal view. (average distance, calculable
Let the measures be presented in inches, or let the width of
from the circumference too)
Christ's throne be used as a unit. This latter one is cca. 25.4
mm - 1 inch on both Christ-plaques. It is to be noted, that all Mean value cca. 50,5 cca. 1,99 "' ----
+.-- ----
the measures of the frontal view are integer without exception Maxima.I deviation from the mean: t3,5%
(Table 4,Fig. 2) !
The numbers representing the measure-data in inch - but Table 3. Measures near 2 inch on the Holy Crown
solely on the complete Crown, and not on the upper and lower ( 1 inch=2,54 cm).
parts separately - produce a complete golden-sectional se
quence: 3
1,2,3,5,8.
e.5
The circumference data of the elliptical ring and the bands
confirm the above discussion (Figs. 3, 4). Fig. 3. T he base line of the Crown is elliptical, its perimeter
is the closing value of the dimension sequence and the ratio
C) Control of the method sequence at the same time: the perimeter is given in inch,
the numbers of ornaments along it are: 8 ornaments
Let the ratios of the mantle be examined. (plaques+gems) and the 13-13 loops fixing
The measures on the mantle are unstable, but the ratios are the strings of little pearls.
stable reasonably. Let the main divider circle of the drawings <')
.I \. '
(today inegular because of damages)
Originally shaped Width of bands Without
"
m
omamentics as seen today (calculated) cca. 50,4 $ 51,5 min "' E
Lenght of the cross from base to top cca. 56,5 !Dll1 lii
E
Total lengh t of the cross cca. 58,5 !lllll
Projeeted lenght of the cross along K
La 1 + 2 + V,. ,.. 2 + V,, -t 1 s 6 ... 2V, L > ; -= 6 + 0, 28; L < 8
the vertical axis
from the base 52,3 mm
2v,. > o.28 data: v. co ... -inch
from the contourline of the
upper central plaque 55,3 - 55,5 lll!ll 8 Fig. 4. T he closing of the contour dimensions on the bands
(frontal view). Simultaneously the necessity ofV
Fig. 1 c. Measured points on the Crown. - Rear-side. Table 2. Some measures of the Holy Crown. Fig. 2. Frontal sketch of the Holy Crown. and the range of its values will be obtained.
264 265
Ferencz, Cs., Ferencz-Arkos, I. & Fehil; A.
Now the ratios can be seen on the next figure: These ratios Conclusions EXPERIMENTAL DATA ON THE TRADE OF RAW M ATERI ALS IN
are very near to a golden sectional graphics, but are not
EUROPE DURING THE LATE BRONZE AGE
complete. Now a hypothesis presents itself. Namely, in case Structural analysis may be an effective
the mantle was originally planned according the rule of the tool in
archeometrical investigations.
SZABO, G.1 & SZONOKY, M.2
golden section, and later a part of the graphics was cut. So one In the presented example this means the
followings:
may calculate the original shape. This can be seen as a dotted - The Holy Crown of Hungary is a uniformly
designed an
d
1 Koszta J6zsef Museum, Szechenyi liget 1, 6600 Szentes, Hungary,
line in Fig. 5. produced unity as a whole.
2Dept. of Geology and Paleontology, J6zsef Attila University, Egyetem u. 2-6, 6722 Szeged, Hunga1y
This result is acceptable and it is in an excellent - The Holy Crown was made according to a
very strict tu 1
agreement to the earlier reconstructions (Balint 1977; e
system me
. Iuctmg the golden section's iules as
_
well.
Mihalyffy 1928). The method of structural analysis works - It was planned in the measure: the ancient
used inch. The technological, manufacturing processes, which very often recquired special knowledge, abilities and technical methods
well in this case too.
and tools as well as the utilization of special raw materials pointed to not only regional relationships, but to such connections
References (only as samples) as well, with the help of which we were able to make historical conclusions. Our analysis of the bronze finds and evaluation of
=1,8
2 field-work observational data have been widely supported by a series of archaeological experiments carried out in Regoly, then
Beothy, M., Feher, A., Ferenczne Arkos, I. &
Ferencz, Cs., later at Bolcske and in the Archeological Site of Szazhalombatta besides the number of scientific analysis carried out in various
1981, Egy regi kor kozmol6giajanak emleke: A laboratories. It can be clearly seen that the acquisition of the rocks used for bronze manufacturing can be connected to three,
magyar
Korona (A Record of an Ancient Cosmo
logy: The clearly distinguishable supply areas. With the help of the results of the experiments carried out so far we managed to draw the
I
Hungarian Crown), (in Hungarian), Fizikai
I Szemle 31 , borders of two regions, which were responsible for the Late Bronze Age metal raw material supply ofTransdanubia. The bronze
\ 473-482. objects northern in their origin and the sandstones with special qualities convenient for casting, which were found in the
\
\ Balint, s., 1977, Unnepi kalendarium
\ (Celebrational Carpathian Basin reinforce the assumptions of the existence of a trade route pointing beyond the Alps. This also poses the
\ Calender), (in Hungarian),
\ Szent Istvan Tarsulat, question of international trade and the importance of a Europe-size division of labou1; greater than that of the Carpathian Basin
'
\ Budapest.
' size besides the question of the bronze supplying raw materials. Since in this case we can definitely talk about not only an
'
'
'
Mihalyffy, E.,1928, Pannonhalma, (in Hungarian), Magyar international trade meeting the demands of some people in terms of consumer's goods and luxurious commodities but a well
...
Mtiveszet 411 .
' ........ organized international trade of special industrial raw materials at a European level, as well.
L Ferencz, Cs., Feher, A., Ferencz-Arkos, I., Henne!,
..-.. ___ ____ ,,,..
S. &
Beothy, M., 1994, Structure analysis and other aspects
in KEYWORDS: LATE BRONZE AGE, EXPERIMENTAL-ARCHAEOLOGY,
Fig. 5. The ratios of the drawings on the mantle. the investigation of the Holy Crown of Hungary and the
ARCHAEOMETALLURGY, PETROLOGICAL ANALY SIS.
coronation regalia, in Sacra Corona Hungariae, Bakay,
K. (ed.), 189-297, Varosi Muzeum, K6szeg. The new bronze finds and foundries which have come to only a couple of copper-ore occurrences in Hungary (Ruda
light recently in Tolna county along the southern parts of the hegy, Lah6cza-hegy, Bajpatak, Velem, Bozsok, Rohonc,
river Danube pose a number of technical and technological Tokaj),5 a number of foundries and metallurgical centers
questions.1 The microprocesses observable in the bronze finds might have existed in this area,6 which manufactured metals
drew our attention to the phenomenons and processes related in different qualities due to the composition of the ore. (The
to the manufacturing and utilization of these objects.2 The fact that the component elements of the ore change in parallel
technological, manufacturing processes, which very often with the mining of the ore-deposits makes this picture more
required special knowledge, abilities and technical methods colorful and sophisticated.)7 According to Szegedy E., on the
and tools as well as the utilization of special raw materials basis of the metallurgical, technological analysis of the Late
point not only to a system of regional relationship, but also to Bronze Age objects, it is possible to define the once existing
such connections, with the help of which we are also able to metallurgical centers and their trade relations as well.8
make historical conclusions.3 Our analysis of the bronze finds Recently, however, Czajlik Z., on the basis of his personal
and evaluation of field-work observational data have been observations and experiences at various sites has claimed that
widely supported by a series of archaeological experiments there are no real existing data indicating or referring to Late
carried out in Regoly, then later at Bolcske and at the Bronze Age metallurgy at localities within our national
Archeological Site of Szazhalombatta as well as a number of borders.9 For example from what we may call the most well
scientific analyses carried out in various laboratories.4 In the known metalwork center in literature; the Velem locality
following parts I would like to give additional inf01mation on despite the numerous field-experiments that had been carried
the trade of the two basic raw materials indispensable for Late out there, no evidence - finds or traces - of metallurgy came
Bronze Age metallurgy: the bronze used for melting and the to light. Though in most cases in the area near the ore
sandstones used for making moulds. occurrences the waste rock burrows are still clearly visible on
A number of Hungarian scientists agrees with the the surface even hundreds of years later as well. 10 There seems
traditional view that though we may suppose the existence of to be a contradiction here, the reason for which might be that
266 Proceedings ofrhe 31" lllfemario11a/ Symposium on Archaeo111et1)'. Jerem, E. & T. Biro, K. (eds.), Archaeopress-Archaeoli11gua, BAR-Ce1111: Ew: Sei: 1.
Szabo, G. & Sz6n6ky, M. Experimental data to the trade of raw materials in Europe during the Late Bronze Age
the earlier records in literature do not make any difference order to gain better drying results a more loose-strnctured characteristic strnctural elements of the Late Bronze Age special technical and technological knowledge, which are
between the ore-processing metallurgical centers, and the maybe organic material thinned mould cast was prepared. 1 metallurgy. If we examine the occurrences and locations of basically characteristic and occur in large amounts in the
metalwork centers doing the manufacturing of finished or Even hundreds of wax moulds could have been prepared, and objects similar to the ring-shaped finds circular in cross Trasdanubian regions, in the Tiszaszentimre find set together
semi-finished bronze raw material pieces. This is also true for by embedding them next to each other they could also have section, and having ornamentations made up of spirally with the rhomb cross section bronze rings generally
the other major metalwork centers in the Carpathian Basin: in been casted at the same time. The b1ittle fracture surfaces running, circular sets of lines as well as their roles or widespread in the Carpathian basin indicate the movement of
their neighborhood the evidence of ore-processing, the observable in most cases on one edge of the objects prove that designations in the Late Bronze Age it can be clearly seen, that this unique object as some kind of currency. 24 Its role of
burrows are also missing. 1 1 the casting taps had been broken from the casted strings or in Transdanubia the ring-shaped raw material finds are equally simply being a raw material however, would not explain such
Consequently in Hungary due to the geographical location lines while warm, then the bronze line still straight was occurring in the major metallurgical centers (Velem, Lengyel, distant occurrences of this really peculiar, thus well-traceable
of the ore-mines and the nature of the work, the metallurgical reheated and heat-treated so that it could be formed to gain the Regoly etc.) as well as at minor human settlements, such as ingot like areas along the tiver Tisza. Namely in this area the
centers connected to the ore-occurrences are separated desired shape. This way with the help of a simple Kisdorog or Szakaly. Besides the almost industrial scale nearby metalwork centres - situated either to the north or the
geographically as a matter of course from the metalwork technological process one could gain numerous raw material production going on at the major metalwork centers, the ingots east - were basically responsible for raw material suppo1t. All
centers doing the further manufacturing of bronze on an pieces, which could have been widely used and easily reaching the tiniest villages could have served as raw material these factors point to the fact that the pieces Transdanubian in
industrial scale on different levels, similarly to the structures transported either by having them striped onto a rod or a supplies for the local blacksmiths. The massive spread of their origin and appearing in distant places also point to the
of human settlements. Thus the extraordinarily rich, Late leather stripe. bronze tools made the local manufacturing of simpler objects movement of the rings derived from their function as some
Bronze Age metalwork developing in the Carpathian Basin As I have already mentioned this is only one of the possible such as bracelets, ankle- and foot-1ings, needles and fibulas,20 kind of a currency, the well-developed Late Bronze Age trade
could have gained its supplies of raw materials only from forms of the raw material pieces, since all the major which could be produced from single bronze strings or lines, and circulation of money and the well-organized support of
mines in the margin areas due to the geographical location of metalwork centers working on an industtial scale could have possible and necessary besides the production and mending of bronze raw materials.
the ore-occurrences. This is not a unique phenomenon in developed besides their peculiar products their own scale of simpler bronze mouldings or casts. The manufacturing of the In the permanent foundties there was some kind of
Europe since with the exception of the recycled objects, the raw material forms, either in the form of tools, jewelry or bronze raw material itself however - even because of the lack preference for the sandstone moulds, which could have been
flourishing northern Bronze Age metalwork utilizes only weapons as well. We have knowledge of sickle-shaped ingots of ores - would have been far beyond of what the minor repeatedly utilized. In the two foundries excavated in 1995 at
imported raw materials as well. 12 Besides the simplest loaf, - which are otherwise inconvenient for utilization - such as settlements could have achieved. But with the help of the the Bolcske-Sziget location the sandstones used for the
tongue and bar-shaped raw material pieces, in the domestic the ones mentioned by Hamper J. from Kalnyikfalva17 as well ingots, which reached them through the means of trade - and production of moulds display a great variety. Besides the
find sets one can find pieces of particular forms and as an octagonal rod, such as the one known from Regoly were quite valuable thanks to their material and this way could heavily compacted materials thinned with sieved quartz sand,
designations. 13 Recently our attention was caught by a ring Veravar. 18 On the latter one even the steps of cutting are have served as some kind of money substitutes - the minor which are almost impossible to tell apart from the real
shaped, multifunctional piece, unique in its ornamentation clearly visible, during which from the solid rod minor pieces settlements could also join a larger scale net of division of sandstones, there are real sandstones occurring in large
and appearance. 14 were cut off for purposes of further manufacturing. The traces labor, that of a Carpathian Basin size. numbers as well. This time I would like to give a description
Among the ring-shaped finds coming from locations at of cutting and the fracture smface refers to the following: first The ornaments of the ring-shaped finds, which recquired of only the most important objects, which were also examined
Szakaly-Ftityiilos and Kisdorog-Hegyiszant6k we managed to the heated rod had been cut in circularly by a chipper or - as special technological knowledge did not intend to represent the in thin-sections.
find numerous pieces circular in cross-section and having a we also checked it out in practice - by a sacketed-axe, then a real or possibly barter value, or the weight of the object. The
spirally running, circular ornamentation, made up of 5-9, piece of the desired size was broken off. Bronze heated above characteristic ornamentation and the technological knowledge Mould for the casting of needles
almost parallel, gently contorniated lines on their surfaces. 800C can be untempered to such a degree that the cutting related to it were convenient for telling apart the products of with a head similar
The ornamentation is missing near the edges of the iing surface becomes totally smooth, such as in case of the casted the different metallurgical centers. The bronze rings appearing to that of a poppy-head
shaped piece; the contorniations or inscriptions occurring pie of the Tiszaszentimre treasure finds, from which a pie in the Tiszaszentimre find set together with their analogies,
very often faintly on the outer side are even more difficult to slice was cut out in all likelihood by using a similar method.19 which primarily can be considered as ingots ve1y typical in The find is intact only at the upper side, where tt is
notice on the inner side of the ring-shaped piece. On one edge The edge of the tool - the well-hammered, previously their appearance, and the other pieces different in their shapes completely straight lined. The left hand side is not totally
of the pieces a jagged surface of fracture with faint edges, thoroughly heat-treated sacketed-axe - used during the but similar in their roles or designations - such as the open complete, thus it is impossible to tell accurately whether there
while on the other edge very often a rounded ending can be experiments did not go blunt relevantly even after several ended, thick casted rings, which very often have cast seams could have been more needles in the mould besides the five
observed. Similar finds are very frequent in the Trans cuttings. After cutting in the rod we relied on the fact that the and remnants of casting taps on them (Pacin,2 1 Retkozberencs22 ones presently visible - though it is very likely since there is
danubian (Dunantul) treasure find sets, but they can also be bronze turns extremely brittle at approximately 600C, thus etc.) - could be convenient objects for revealing the market a gradual decrease in the size of the needles toward the right
observed in the Great Hungarian Plain (Alfold) - in the by hammering the hot bronze ingot we managed to break off areas of once existing ancient bronze raw material trade. hand side. The right side of the mould is ragged, but the
vicinity of the major metalwork centers as well, just like in pieces of the size indicated by the in-cuts . At other times The ring-shaped finds or rings thanks to their raw material cambered connection of the outer needle's edge to the casting
settlements made up of only a couple of homes. On the when we were cutting larger bronze ingots for melting and content represented a great value in the given age. As the gutter point to the fact that there were no more forms
ornamented ring-shaped finds one can often observe traces casting, we did not cut them in advance, but we cut them up thoroughly collected, minor bronze pieces appearing in the inscribed into the mould on this side. On the lower part only
and signs of hammering, shaping and cutting indicating into shapeless, smaller pieces by cold cracking. The treasure find prove, the pieces thanks to the extraordinary incomplete remnants of the casting gutter can be seen. In the
further processing of the material. 15 characteristic, coarse, "greasy" fracture surface is a reliable importance of the bronze raw material had extremely high mould the sides of the needles are rather rectangular - when
As the phenomenons observable on the surface of the indicator of this cold cracking method even hundreds of years values, thus as some kinds of indicators of value they could moulded they could have a diamond-shape, which indicates
objetcs and the archaeological erxperiments proved the later. have served as means for making the barter and normal trade that they must have been hammered subsequently to gain a
manufacturing of the above mentioned ring-shaped pieces Disregarding certain objects, on the basis of what have been easier. The ring-shaped raw materials, due to their material, round shape and thus stretching the otherwise small, 6.7-7
might have consisted of the following steps: first a wax mould said earlier, the peculiar ingot forms, which can be dated to a shape, widespread occurrence and weight changing according centimeters long needles.
covered with more liquid clay layer was used, then later in wider time period within the HA A era outline some to given units23 were especially convenient for functioning as The height of the mould is 8.8 centimeters, the width is
some kind of a good having universal equivalent, thus 8.4 centimeters and the thickness is 3.6 centimeters. It is
fulfilling the role of the ancient currency. The bronze rings or made of reddish -greyish-black in the highly burnt areas -
" According to Pasztor G. a mine and a smeltery must have been working in Velem (Pasztor et al. 1990: 10). His views are generally accepted in the ring-shaped objects besides their primary role of being a raw friable, tabularly compacted ceramics with high quattz
archeological literature as well (Miske 1929), though researches done by a magnetometer so far has not been able to detect anything about the presence of
material, wonderfully accomplished this new function as well. content. The locally derived (Danubian) fine-grained sand
furnaces in the locality (Vastagh & Kiszely 1962: 232).
The appearance of the two pieces - having a spiral running used for thinning the clay is clearly recognizable in the
12 Bronsted, on the basis of the rod-shape raw material, supposes that the people of the northern Bronze Age acquired the metals in this fom1 - as it is proved
by the tight relations even from the area of the Carpathian Basin as well - thanks to the local absence of raw materials (Bronsted 1962: 99-100). circular ornamentation made up of sets of lines and recquiring material of the mould.
13 Hampel 1886-96. Further details can be found in Mozsolics 1984; Fekete 1985; Maraz 1989.
1' Szabo 1996.
15 ibidem 20
Hampel 1886-96: CLXXII.T.l
6
1 Zsak 1960. 21 B. Hellebrandt 1989: 99.
17 Hampel 1886-96: CCV.T. 1-4. 22 Mozsolics 1985: 194, T.1-11.
1' Szabo 1993: 188, 11.T. 5, 7; K6szegi 1993: 225-235. 23 Mozsolics 1963: 68-69.
1' Hampel 1886-96: CLXXII.T.l. " Szabo 1995.
268 269
Szabo, G. & Sz6n6ky, M.
Experimental data to the trade of raw materials in Europe during the Late Bronze Age
A piece of a mould for casting a larger deposited later during the history of the Earth, in the heavily compacted structure of one object examined in thin am Velem-St.Veit-Berg,Westungain, ARCHOST 612,
sacketed-axe with ornamentation Miocene or Pliocene eras. Rocks of this age are very section. The piroxene andesite cannelura rock coming from 30-35.
of circular striations on the sides common in the Carpathian Basin but they also occur in many the Bolcske foundry has a rock material, which is originated Damay, K., 1910, Az uzsavOlgyi bronzkincs, ArchErt 30,
places all over Europe, thus the origin of this piece can not in all likelihood from the Carpathian Basin. However it is not 426-431.
The side of the axe, which is aproximately 3.2x4 centimeters be determined. possible yet to tell the exact place of origin, for this purpose Eibner, C., 1982, Kupfererzbergbau in bsterreichs Alpen, in
in diameter lies 2.7 centimeters far from the upper side of the The stone axe has a very unique stone material coming further comparing analyses are needed. Siidosteuropa zwischen 1600 und 1000 v. Chr., Band 1.,
mould. This distance determines the maximal possible length of from the Alps: garnet gneiss, or retrograde eclogite. The Hensel B. (ed.), Berlin, 399-408.
the casting gutter and the clinker pit. On the form under the material of the polished stone also from this locality is a Summary Eibner, C., 1993, Zurn Beginn expe1imentalarchaologischer
sides a minor handle can be observed. The width of the heavily typically Alpine-type rock which also occurs in Burgenland: Untersuchungen einer bronzezeitlichen Kupferver
damaged mould, which was originally chiseled to a prismatic micaceous quarzite. It can be clearly seen from what has been said earlier, that hiittungsanlage in Muhlbach, Salzburg, Archeologica
shape is 7.8, its thickness is 5.4 while its incomplete length is
the acquisition of the rocks used for bronze manufacturing Austriaca
77, 31-38.
12.2 centimeters. It is made of red sandstone. A piece of a larger casting mould, can be connected to three, clearly distinguishable supply Fekete, M.,
1985, Adatok a koravaskori otvosok es
The red, fine-grained sandstone is one of the most widely which was chiseled and polished areas. The first region is more or less equal with the area keresked6k tevekenysegehez, ArchErt 112, 68-91.
used types of stone - in all likelihood due to its frequent smooth and straight on one side lasting to the borders of the settlements; this region served as Hampel, J., 1886-96, A bronzkor Magyarhonban I-Ill,
domestic occurrences. Among the sandstone moulds the red,
a supply of quaitz sand used for the production of ceramic Budapest.
possibly Upper-Permian or Lower Triassic, fine-grained The inscribed outlines of two objects can be observed moulds and the tamped sand forms. The eastern margin of the Horst, F., 1986, Die Jungbronzezeitliche Kannelurastein des
sandstones are the most frequent. Rocks similar to the on the smooth side. Unfortunately they are to faint to Alps and the mid-mountains of Hungary comprise the outer mitteleuropischen Raums- Werkzeuge fiir die Bronze
material of the examined samples can be found in the western determine the objects themselves. The other side of the border of the second region. This aiea more or less complies verarbeiteung? Helvetia Archeologica17, Heft 67.
parts of the Mecsek Mountains and the Uplands near Lake mould is loosely adjusted, slightly semi-rounded. The with the Transdanubian region of the "umfield" culture and Indreko, R.,1956, Steingerete mit Rille, Stockholm.
Balaton in Hungary. But this type of rock is also well-known object must have been 16 centimeters long, 11 centimeters whose connections can be traced until Tokaj, the line of the J6sa, A., 1910, Halstatti vagy "nyfri" kultura, MKE 4,
in Northern and Western Europe. wide and 6 centimeters thick originally; unfortunately it is river Tisza, but this does not continue towaids Transylvania. 109-117.
impossible to give the exact measures of the piece, which The third supply area points beyond the Alps, streching as fai Kemenczei, T., 1983, A Tatabanya-banhidai bronzlelet,
The mould of a sacketed-axe suffered severe damages because of the plough. It is made as the quarries of Northern and Western Europe. ArchErt 110, 61-67.
of the typical fine-grained Cretaceous sandstone occurring W ith the help of the results of the experiments carried out Kemenczei, T., 1984, Die Spiitbronzezeit Nordostungarns,
On the side of the hose traces of 2, while on the hose itself only at territories far from the Carpathian Basin along the so fai we managed to draw the borders of two regions, which Budapest.
traces of 3 striations can be seen. The body of the axe itself coasts of the Baltic and North Sea, and which has dark were responsible for the Late Bronze Age metal raw material Koszegi, F., 1988, A Dunantul tortenete a keso bronzkorban,
under the horizontally running striations is smooth and gray color when burnt. The sandstone having a grayish supply of Transdanubia. (The local one is naturally out of the Budapest.
undecorated. The length of the molded axe is 7.8 centimeters, color when burnt has been separated during the analysis of question, since there are no copper-ore occurences in Koszegi, F., 1993, A Regoly-Veravar keso bronzkori
the width of the hose (outer) is2.8 centimeters. The width of the objects coming from the Bolcske locality and it is Transdanubia.) The first one is the Eastern-Alpian region, leletegyiittes tipol6giai es idorendi kerdesei, Wosinsky
the axe at the lower part of the hose is 2.1 centimeters, the considered to be an alien material in the Carpathian Basin, which participated in the bronze supply of Transdanubia by M6r Muzeum Evkonyve 18, 225-235.
width of the edge is 4.4 centimeters. The edge and the hose is which was imported from remote places. The moulds of a means of its casting pies, having distinctively low alloy Lazar, J., 1941, A saghegyi I. es II. szamu bronzleletek
separated by one horizontal striation. The length of the casting number of objects coming from the locality were made of component content and having been transported in large ismertetese, Dunantiili Szemle Konyvei, 185.
tap at the upper part of the axe is 1.2 centimeters. The mould this stone, and we also found untouched stones of this type amounts. We must seek the other, more remote region in all Lazar, J., 1943, A saghegyi 6skori telep bronzmuvessege -
itself is intact, originally it must have been used for casting without any carvings or inscriptions. The raw stone block likelihood in the Erzgebirge from where the peculiar ingots Die Bronzeindustrie der urzeitlichen Siedlung am
some other object as it is indicated by the uneven rectangular refers to the fact - just like the shape of the objects used with higher tin content, and very often having a ring-shape - Sagberg, DuSz,280-287.
shape of the mould, and it should have been chiseled in the Carpathian Basin in the ready made, major moulds the production of which recquired special knowledge - were Majnaiic-Pandzic, N., 1992, Ljevaonica broncanih predmeta
subsequently to gain the mould of the axe. Its length varies - that the sandstone had been transported here in forms of transported to Hungary. The bronze objects northern in their na Kalniku kod Krizevaca, Opusc. Archaeol. 16, 57-73.
between 10.5 centimeters and 9.5 centimeters, its width is 5.4, larger blocks and then it was manufactured later on origin and the sandstones with special qualities convenient for Maraz, B., 1989, Az 6skori bronzmuvesseg viragkora a Del
its thickness is 3 .1 centimeters. Its material is Miocene fossil locally. casting, which were found in the Carpathian Basin also Dunantulon, Baranya 198911-2, 7-25.
bearing calcitic tuff, which in all likelihood comes from the The so-called cannelura rock which has come to light for reinforce the assumptions of the existence of a trade route Miske, K., 1904, Die Bedeutung Velem St. Veits als
Tokaj Mountains. the first time from the previously mentioned locality in pointing beyond the Alps. This also poses the question of prahistorische Gusstatte mit Beriicksichtigung der
Hungary is pointing towards connections with the n01thern international trade and the importance of a Europe-size Antimon-Bronzefrage, AJA 2, 124-138.
A mould for casting needles, closed by areas. The object, which might seem to be a simple, rounded Miske, K., 1907, A Velem Szt. Vidi ostelep, Wien.
division of labour, greater than that of the Carpathian Basin
fracture surfaces on each side grinding stone for the first sight has slightly concave lower Miske, K., 1908, Die priihistorische Ansiedlung Velem St. Vid,
size besides the question of the bronze supplying raw
and upper parts, and on the side there is an indentation materials. Since in this case we can definitely talk about not Wien.
The
4 mm wide, 3 mm deep "V " -shaped contorniations running around, the cannelura after which the object was only an international trade meeting the demands of some Miske, K., 1913, Bronzkori typologia II., MKE 7, 8-24.
running parallely 0.9 centimeters far from each other on the named. This object is very often described in the literature as people in terms of consumer's goods and luxurious Miske, K., 1928, A t6rtenelem el6tti id6k fernkohaszata, TtK,
casting mould served for casting the sides of the needles. The
an ore-stamping hammer,25 this assumption is not true commodities but a well-organized international trade of 476-480.
minor traces of rubbing running lengthwise parallelly with however, because there are no ore-occurences or 28
special industrial raw materials at a European level as well. Miske, K., 1929, Bergbau, verhiittung und Metalbearbeitungs
each other on the side of the contorniations indicate that they metallurgical centres in the nearby areas where this object werkzeuge aus Velem - St. Veit (Westungam), WPZ, 81-94.
must have been rubbed or chiseled from the material of the appears - such as in the case of the Bolcske locality. Its References Mozsolics, A., 1963, Der Bronzefund von 6palyi, AAH 15,
mould. The color of the surface of the light-gray, fine-grained general area of occurrence is the region of the northern 65-82.
sandstone grew reddish in the middle. The heat effect must Bronze Age26 and Lake Zurich, furthermore we have B. Hellebrandt, M., 1989, A pacini IV. bronzlelet, ComArcHung, Mozsolics, A., 1973, Bronze und Goldfunde des Karpaten
have been stronger on the inner side since the discolored layer knowledge of 1-1 pieces from Bolcske, Hungary and Kalnik, 97-117. beckens, Budapest.
is thicker there. The measures of the piece: width 7 Croatia.27 The number of molded products are very Bronsted, J., 1962, Bronzezeit in Danemark, in Nordische Mozsolics, A., 1984, Ein Beitrag zum metallhandwerk der
centimeters, height 5.4 centimeters, thickness 2.6 centimeters. significant in the area of occurrence of the cannelura rocks, Vorzeit II., Neumiinster. ungaiishen Bronzezeit, BRGK 65, 20-72.
The mould is made of a looser rock; a light-colored,
thus we might suppose that this special tool was used in the Czajlik, Z., 1993, Exploration geolarcheologique du Mont Mozsolics, A., 1985, Bronzefunde aus Ungarn, Budapest.
yellow, thin-bedded sandstone, which must have been
foundries - for tamping sand forms as it is indicated by the Szent Yid, Acta Arch Hung
45, 317-347. Otto, H., 1973, Uber die Auswertung der Metallanalysen in
Czajlik, Z., Molnar, F. & Solymos, G. K., 1995, Angaben zu Band l und 2 der Studien zu den Anfangen der
" Indreko 1956: 9-11. der Spatbronzezeitlichen Metallrohmaterialversorgung Metallurgie, APA 4, 1-9.
26 lb.
Fig. 12.
27 Horst 1986; Majnaric-Padnzic 1992: 57-63; Yrdoljak 1992: 75-87.
28 Tylecote 1987: 18-22.
270 271
Szabo, G. & Sz6n6ky, M.
272