A uthors:

Beszeda Istvan Unjversity of Debrecen, Dept. of Solid State Physics

H-40 I 0 Debrecen, Egyetem ter I .
Bigazzi, Giulio Instiruto de Geocronologiae Geocbimica lsoto pica CNR
I-56 127 Pisa, Via Cardinale Maffi 36.

Biro Katalin T. Hungarian National Museum

H-1370 Budapest, Pf. 364.
Dobosi Viola T. Hungarian National Museum
H-1370 Budapest, Pf. 364.
Elekes Zoltan Institute of N uclear Research of the HAS
(ATOMKI). P. 0. Box 51. H-4001. Debrecen, Hungary
Gratuze, Bernard Centre de Recherches Ernest Babelon, 30 me de La Ferollerie,
F-4507 1, Orleans Cedex 2

Oddone, Massimo: Diparcimemo di Chi mica Generale Universita di Pavia

1-27 100 Pavia, Viale Tammelli 12

Ringer Arpad: University of Miskolc, Dept. of Prehistory and Antiquhies

H-3515 Miskolc, Egyetemv~os
Rudoer Edina Z.: HAS Institute of Geographic Sciences
H -111 2 Budapest, Budaorsi 43-45.

ut

Silmegj Pal: University of Sz.eged, Depl. of Geology and Palaeontology

H-6722 Szcged, Egyetem u. 2.

Voros Istvan: Hungarian National Museum

H-1370 Budapest, Pf. 364.

CONTENTS

Dobosi, V. T. Archaeological investigations al Bodrogkeresztur-Henye

I. Topography . . . . . . . . . . . . . . . . . . . . .
2. Research History . . . . . . . . . . .
3. Archaeological finds . . . . . . .
3. 1. Tools . . . . . . . . .
3.2. Stalistical evaluation .
.
3.3. Technology . . . . . .
.
3.4. Comparison between traditional types and
3.5. Raw inaterial . . . . . . . . . . . . . . .
3.6. Sertlemem features . . . . . . .
.
4. Analogies . . . . . . . . . . . . . . . . . . . .
4. J. Contemporary sites in Hungary . . .
4.2. Contemporary sites outside Hungary .
5. Cultural affi.Jiation .
6. Summary . . . . . . . . . . . . . . . . . . . . .

. . . ..

5
5

13
16
.. . . .
. . . . . .
. . . .
their recent equivalents .
. . . . . . . . . . ....

Voros, I. : Bodrogkereszlur- Henye, Hunted mammals from the Upper Palaeolithic site
t . lnLroduction . . . . . . . . . . . . . . . . . . . .
2. Topographical distribution of lhe remains . . . .
3. Zoological and palueomological characterisation .
3.1.Equas.
3.2. Alces .
3.3. Cervus .
3.4. Bison .
3.5. Mammulhus .
3.6. Leo . . . . .
3.7. Lepus . . . .
4. Big game remains on the hunting campsite .
4. I. Quantitative distribution
4.2. Qualitative distribution .
5. Hunting-gathering strategies
6. Summaryl77

16
54
55

57

62
68
80
80

84
94
103
11 3
113
117
128

128
146

159
161
162

164
164
164

164
165

17l

Appendix:
Voros /. : Macro-mammals on Hungarian Upper Pleistocene sites

Ringer A.: Loess- and Palaeopecfostratigrnphical Data . . . . . .

Sumegi P. -Rudner E.-Beszeda /.: Stratigraphical and palaeoecological
investigation . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Bir6 1: K.-Bigau.i, G.-Oddone, M.: Instrumental analysis I.: The Carpathian

sources of raw material for obsidian tool-making. (Neutron activation and
fission truck analyses on the Bodrogkereszrur-Henye Upper Palaeolithic
artefacts) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Bir6 T. K.-Elekes Z.-Gratu4e, B.: Instrumentlll analysis II.: Ion beam analy es
of artefacts from the Bodrogkereszwr-.Henye lithic assemblage . . . . . .

186
213

Archaeological investi.gations at BodrogkeresztUr-Henye

Viola T. Dobosi

The financial ba.~is of the presem v.:ork \\-"as provided by the Foundation of Lhe Hungarian National Museum and the
National Science Foundation (OTKA Grant Nrs. 2794, T 020038). Publication was supported by Lhe Ministl) or
Cultural Heritage (Cultural Hetitage Directorate 1318-24-54/99.

1. Topography a nd the sunounding Palaeolitltic sites

Geographical endowments
The main fearures of the surface morpbology of the
Carpathian Basin were already established by the end
of the Tertiary period. The special interior development of the basin, surrounded by the Carpathians and
the interior volcanic arch was only modified by elevation and depression on local scale. The Basin is closed
from lhree sides. Its special micro-climate is strongly
influenced by its morphological position: more dry
than could be expected on the basis of geographical
co-ordinates. Also due to its basin morphology, ce.rtain
bio-geographical insulation can be observed in its development (SOMOGYI 1982, 77 .) The cca. 300 thousand square km area. beyond the above general statements. was probably dissected into several so far unknowo "ecological niches"during lhe whole time span
of the Pleistocene.
This relative isolation indicated by geomorpbologica l and climatic data seem to be proved in the light of
recent r~search at least for a restricted area and time
span (by the end oftbe Pleistocene SUMEGl 1996, 9)
At the same time, lb.is bio-geograpb.ical isolation cannot be traced in the Upper Palaeolithjc, though ii is
well known thaL the dependence of Palaeolithic communities "on natural conditions and resources'' were
sLill all important Quoting Hagen: " ...nature has been
the dominant partner and any examination of the relationship of man to nalure will inevrtably be a one-way
affair" (HAGEN 1972, 10) More exactly: the Upper
Palaeolithic population of the Northern third and the
central region of the Carpathian Basin were, beyond
doubt. participants or active agents of historical processes from the Intcrpleniglacial to the WOnn maximum (using recently suggested terminology, the MUP
= Middle Upper Palaeolithic period (MUSSI-ROEBROEKS 1996). Differences between the find material of Hungarian sites and those of the classical habitation zones of Palaeolithic cullures have been observed by most students of the period and tried to
solve the problem by terminological tricks. Differences in the archaeological material can be explained,
to our present knowledge, by the peripheral position

of the Carpathian Basin compared to the tribal area of

the great cultures. This could be further influenced by
relative isolation in the Basin, lhe possible direction of
contacts determined by geomorphological endowments etc. Accepting this theory, probably some retarded (?), surviving(?) cultures of the Transylvanian
basin, "outliving their time" can also be explained by
isolation detennined by geographical factors.
To the Easl of the Danube-bend, the Northern stripe
of Hungary is occupied by the Noithem Mid-Mountain Range. Its Easternmost momber is the Eperjes
(Presov)-Tokaj Mts.. known also in (erroneous) vernacular use as Zcmplen Mts. This name is mainly applied for the parts lying to the South of the Szalanc
(Slanec) palh. This mountain range belongs to the innerntost volcanic range of the young folded mountain
system of the Carpathians. This area forms a transition
between the High Carpathians and the Lowland
{Altbld) in several aspects: morphology. climate, biological environment. Due to the intersection of the
margin of the mountains with the lowlands, Alfold
intrudes deep into the river valleys (PlNCzES 1960,
1-2.). This paleogeographical fact bad a favourable
influence on the settlement strategy of Upper Palaeolithic communiLics, who were naturally most sensitive
to versatile ecological challenges. This an:a wos inhabiLed in all periods of the Upper Palaeolithic.
The Tokaj-Presov Mts. was formed as a result or
several volcanic eruptions in the Miocene I Sarmatian
period. Tt has different strike from that of other
members of the North-Hungarian Mid-Mountain
range, directed roughly North-South. Eruption centres, lying some 8-10 km fTom each other following
the main strike of the mountain in a zig-zag line
can be still recognised. Lava cones were typically
dislodged by modem age quarries. (GYARMATl
1972) (Fig.1.)
Sulphurous springs of the aren arc slight reminiscences of the Miocene volcanic activity. The postvolcanic activity following cbe intensive eruption phases
supplied the region - especially the southern border of
5

~ Rlollt

Hid rok'f8rolt

U mnolcnroll

[!]
f:;~
0
I

'

4
I

rKm

Fig. I 'l11c souLhem part of lhc Tok.aj-Presov Mts. with ra" m:ucrial outcrops
(Map by P. Gyarmnt1)

G Jzlrft

the mountains - with a ring ofsilicites. prepared from

1he soft pyroclascic rocks during the Late Pleistocene.
They served the prehistoric popuJation a vast quantity
of, mainly excellent rd~ material for tools. Silicic acid
leached by thermal springs from the o.ndesite (and rhyo litc) \\as scdimen1ed in volcanic cavities, clefts and
lakc:s in the form of opal, chalcedony or jasper at different phases of the volcru1ic activity. E.g., in limnic
quartzite. plant remains from the Sarmatian period or
Pannonian age wood trunk can be found, equally.
(PlNCZES 1960, 16.)
The body of the mountain is rising suddenly from
the flat surface of the Aillild. lts young, slightly
eroded fom1 recall, in spite of its relatively low altitude (5-700 m a.s.l.) real mountain forms.
The most ancient member of the mountain range is
lhe eponym Tokaj (or, Kopasz [Baldl) Mt. It is standing separately, to the south of the main body of the
mounrains: an irregular volcanic cone of 512 m altiLL1de, dissected by several dry valleys. Measured along
the Holocene terraces of the rivers, Bodrog and Takta,
its diameter 1s 5 km on the average. This cone is a decisive geological formation of lhe area. Tbe Kopasz
Ml. or Tokaj Mt. gave its name to the volcanic mountain range, tbe region as well as the wine..
It is separated frol!l the main body of the mountams with a 5-6 km wide stripe of ltiJls. These llills are
some 300 m lower than the Kopasz Mt, and at least
100 m lower than the margin of the mountain, consisting of smaller rhyolite cones covered by thick Late
Pleistocene loess. The highest point of the hills is Bodrogkercsztiir-Henye itself, at l98 m altitude.

Archaeological ropography
Investigating the topography of Hungarian Upper Palaeolithic sites, certain tendencies can be observed in
the spatial array of sites.
The earliest Upper Palaeolithic culture in Hungary,
Aurignacian wit11 split base bone points seems lo be a
short "cave episode" (perhaps serving as temporary
bunting camps of the open-air sites known to the
North and East of our borders; or, perhaps other ethnic
groups?). Caves are objective endowments. It is an
ideal habi1ation environment within the BOkk Mts.
though at other places of Central Europe the Aurignacian population did not live in them. It is of certain
importance that here again we imply a conscious
choice on behalf ()f Palaeolithic people the reason of
which is unknown to us. Both Aurignacian sites are
caves at relatively high altitude. though in BUkk Mts.
there are large, habitable caves at much lower allirude,
in 1he level of the sole as well.
Topographical investigation can only lead to some
general observations if the location of the sctl.lemenl
was arbitrarily selected by prehistoric people. TI1e
Hungarian open air Upper Palaeolithic siles belong

cultural!} to tbe Grovettian entity. spanning over the

Middle Upper Palaeolithic and the Late Upper PalaeolithJc periods.
Certain differences were observed in the senlemem
strategy of the two periods. As the number of sites
knovm is not too much. these difforences should be
rrealed as trends rather than laws. People of the MUP
/older blade industry/, i.e. Pavlovian complex preferred ro occupy multiple function strategic poinrs, on
e levations along lhe southern margin of the m.idmountain ranges, foothill slopes, terraces or hilltops
180-200 m altitude. on relatively pem1anenl settlements. Al t.hc same time, both cultural phyla of LUP
/Younger Blade Industry and Pebble Gravenian or
Sagvanan hunting communities. respectively) seemed
to occupy the whole territory of the Carpathian Basin.
Their "ad hoe" settlements I sites were found, comra.ry
to a long lasting geological "topos", i.e., there are no
Palaeolithic settlements on Alt<>ld), partly on loess
covered lee Age relict surfaces, 80-120 cm below the
present surface (J3.sz.5ag I Jazygian region}, partly in
embryoaal soil horizons of typical Jocss sequences.
several mcLers below I.he present (on the northern margin of Telecska-hills, Madaras). ln the given time period of the Upper Palaeolithic, however, t.he endowments of the actual surface could always decisively influence the selection of the habitation area.
The temporal and spatial scale of geomorphologicaJ studies is rather large-scale compared to the needs
of Palaeolithic research specialists. Still the IeconstrucLion of the original surface morphology of a given
settlement can be fairly successful.
The Henyc hill and its immediate environs is surrounded by, from North and South by the mountains,
from the East, river 13odrog, and from the West. the
flatlands of the srream Takca. Knowing the selection
criteria of the Giavettian entity, the site is ideal, about
190 m high above sea level, on a row of steep side
hills: a typical multiple function, strategical scene. II
is located at the meeting point of several ecological
niches:
- in the immediate vicinity (western margin) of the
area, a swampy flaUand served rich and varied lacustrine and terrestrial fauna (leading elemen1: Alccs),
without the danger of temporary iloods
- to the south, the Alfold oftered dry grassy steppe
(leading fauna! elements: Mammuthus. Equus)
- the Northern mountainous environment was less
uLilised for hunting, but thls area is a ric.h source of
variable stone tool raw materials (BIR6-P ALOSr
1983, BrR6-DOBOSI l 991 ). Ln the first place, we
find there one of the 'obsidian centres" supplying the
Carpathian Basin (probably even the areas beyond the
Carpathlans) with obsidian. This could be a decisive
facLor in the sy~tem of connections probnbly nJread)
in the Palaeolithic Period.

- last but not least, the stand-alone volcanic cone

(Ml. Kopasz at Tokaj) is an eye-catching. spectacular
geographical element in itself, a good point of orientation.
Densiry of finds (in situ or off-site) proved that
most of these attractions were valid in those days as
well, even supposing thar the present environment of
the site is more or less different from the conditions
during the lnterpleniglacial. Partly, the hills were necessarily lower because the great loess-producing periods were just to come. The Southern Zemplcn was
still an undislodgcd block. ln course of the study of the
hydrology and the surface fom1ation of the /\J1old
rcgiou. S. Somogyi came to the conclusion t.h::u U1e
immedinte surroundings of Bodrogkeresztilr-Henye
and its immediate surroundings were different from a
geomorphological point of view in the last third of the
Wurm period, same as the larger region, North-Eastern Hungary.
The hill and its environs was not an elevated shelter
in tbe marshes at tlte meeting point of the rivers Tisza
Md Bodrog, bw s imply the las1 member of the volcanic series bordering Llte right bank of the river Bodrog. Bodrog used to have a Longer course in the last
phase of thl! Tee Age. It reached the river Tisza at a
larger distance, at the margin of the Nyirseg and Hortobagy, the active bed of which was probably in the
line of Lhe streams Er and Bererty6 today. Tbe current
conditions were formed by Lhe end oflbe lee Age and
tbe Early Iolocene. with the elevation of the Ny!rseg
area and the depres!.io.n of the Bodrogkoz region. The
river Tisza was turning to North and "captured" several rivers running North-South, including Bodrog
(SOMOGYT 1982, 79.).
The Late Pleistocene endowmems of Lhe micro-region were, however. influenced only in small extent
by the different hydrogeologlca.I setting. Surface morphological fonns and dimensions remained constant in
the two periods (Fig.2.)
The vicinity of raw material sources, loess plateaux
ideal for settlement, tbe meeting point of several different biotopes (proved by fauna elements with different ecological demands), i.e .. ideal settlement conditions were present here, irrespective of the fact that the
Bodrog must have reached the Paleo Tl.Sza (6stisza)
not here but further to the South in the line of todays
Er-Beretty6 rivers.
PalaeoliU1ic finds were collected from all hill-tops
around the Henye hill in a circle of some hundred meters (Fip,. 3.). These hill-tops were all inaccessible for
excavations with the exception of the Henye itself. By
our last lield s urvey in 1988., new vineyards, old but

cultivated plantations and Lhe village waler plant slalion were in function on them.
This group of sites fit well into the chain of Upper
Palaeolithic sites \\ hich:
- follow lhe soulhem margin of the Northern MidMountain range, rich in raw materials
- occupied Lhe lowland~foothill surfaces i111r1.1ding
deep into river valleys from the Danube-bend till
Bcregovo and Korolevo, in other words lhe gate of the
Tisza entering the AlfOld (Great I lungarian Plain).
ll1is ~m::a had been studied relatively well from an
archaeological point of view. Due to the excellent
grapes propagating opt1ma.lly on the volcanic weathering products, highly developed viniculLure, intensive
working of the soil became permnnent, drawing early
attention to the archaeological sites. The environs of
the site have been subject to several (archaeological.
geological, geographical) studies as a collecling spot
"in the way" to any Nonh-East Hungarian st:udy ll'ips.
Therefore Lhc find material had to be collected from
unusually varied sources. 1 There were several field
surveys during the excavations and independent of
them. by professionals as well as private collectors.
Location of the latter collecting activity could not aly,ays be identified exactly. The find material as inventoried in the Herman Ott6 Musl!um of M1skok
(HOM) and the Hungarian Nutional Museum (I-INM),
respectively.
Significant quantiry of finds 1s known from the following collecting spots:
Kmicsos or Kavicsbanya diilo - from the h1lhop,
'1E and SE slopes of the hill (probably spread by
ploughjng): Pb 64/482-494, Pb 83/224-233.
Type distribution:
end-scraper on flake, l piece
double end-scraper on flake, l piece
lateral burin, I piece
retouched flake, 2 pieces
blade, blade-like flake, 8 pieces
Bake, worked fragmt!nt: 60 pieces (2 hydro-opallte.
3 radiolarite, I erratic flint 3 7 hydroquartr.ile and
17 obsidian)
core. core fragment 8 pieces (2 obsidian and 6 hydroquartzite)
Kastclytabla (mentioned as Csengos hill m lht.;
field survey repon by Lajos Toth engineer). dl!ep
pJougbing for vineyard plantalion probably disturbed
the cultural layer here: Pb 83/234-238
Type distribution:
2 end-scraper on Oakc made of obsidian
I blade (Szeletian folsitic porphyry )

A box of strnyfinds from the site, found in Lhc heritage of the Gabori-couplc got into the Hungarian National Mui;cum m

1998, which could not be taken into consideration because or Lhc closing of lbc manuscript
8

TOKAJ - EPERJES

Mt.
11 - 1 1 - 1 1 - 11 -

-11-11 - 11/I-JI-JI
11 1/ IJ-

- n-

-11 -

' ;:;.:_:~lfrfo:.'::::/Jj:'fo~;li#i#~tP _ i>,,

Q--

-v -- JI
-

- 11 -

JI -

. . . . . . . .... ,. _

- I. . .:l -1)

~ .. 1:1:..
,:::~ 1 ::". . .":..-:.
. ; .
:t"J I -

. 1 :....

II ~-..:::_::-: ~::: f ;J.\

;;.\:;o:-)J
11 _ ..:':./'1-:?~+=.-:-.:/F:>

'

J J - -1) -

- - IJ -

II -

11-11-J}-

JI'' "

:.

,,

:: ''

- JI - - }) l) J / --1'

}J -

JI -

JJ -

11 -

11 -

~
-

n-

IJ- J / - J l - -

-11-11

>1

JJ

p -- /)
,, -

,,_-_ ,, __
-

JJ -

- })

--

~ andesite

Sites:

t:::~~ ~.:} loess

-

2 .Tarcal Citrombanya

J)-

}/

,e'

- - surface f Inds
I

alluvium
hidroquartzit

oo obsidian
Fig. 2 The environs ofBodrogkeresztilr-Heny~

" ,,

,,

,.

JI

II

i'

fl I

,
"

ll
II

11

I I
' 1

11

'J ~..
I~

II

,,

11
ll

ll
h

. ~

,, \'. ll
I

Kavicsos

Ka

Kastelytabla

Ko

Kovesd

De Deak-hegy
Der Dereszla
Ci
?

Tarcal-Citrombanya
Sporadic

Fig. 3 Sites and collecting spots around Henye nill

10

20 pieces of flake, knapping debris ( 7 obsidian, 13

hydroquartzite)
Tcrezia-chapel- it is buill on the hill next to Henye,
belonging already to the administration of the village Tarcal. Some chipped stone objects are known
from here as well. The Palaeolithic tools: found
here (Pb 831222) were probably washed out to the
slope of the chapel-hill towards the vi llage by erosion or tumed out by deep ploughing for the vineyards. I 0 worked flake, (3 obsidian, 7 hydroquartzite)
Oereszla-hi ll - the northern part of lhe modern village Bodrog)<ereszllir completely fills the narrowing
valley of the river Bodrog, houses :itretching up,
pru1ly. lo the eastern flanks of the Dereszla hill. This is
an elevation with quite steep slopes, highest point at
162 m. a.s.l. The top of the hill was disturbed during
the construction of a pumping station of the local
water plants. Apart from finds from different other archaeological periods, Magdolna Hellebrandt found
here on mammoth bones and chipped stone artefacts
in the depth of 30-40 cm. (liELLEBRANDTSTMAN 1980, 88). The material of the 1982. field survey got into the Hungarian National Museum: Pb
831239-261 (Fig. 4. 1-5)
Type distribution:
angular burin (hydroquartzite and obsidian) 2
pieces
truncated blade I atypical end-scraper, I piece
burin-edge or worked fragment 4 pieces (2-2 obsidian and hydroquartz.ite )
blade 8 pieces ( Prut silex, hydroquartzite, obsidian)
flakes, knapping debris 28 pieces (hydroquamite,
erratic flint, radiolarite, "stone marrow'', hydroquartzite obsidian)
one borer and a small core-rabot from here are
probably not Palaeolithic.
Between Dercs.zla and Kavicsbanya diilo (location cannot be defined more precisely, field survey by
l stv:in Horvath during lhe 1963 excavations by Laszlo
Verres): Pb 64/495-498:
Type distribution:
2 blade and 8 mixed raw material fragments
Further collecting points are known from lhc foothill slopes along the Kopasz-Mt. To the North,
along the road lo Tokaj we find Bodrogkereszt.Ur Brickyards, where two artefacts were collected by
L3sz.16 Vertes from the sequence becween two buries fossil soil layers:
Pb 64/480: obliquely truncated blade, retouched on
both edges, from patinated hydroquartzite
Pb 64/481 : atypical core, egg-shape hydroquartzite
lump with flake negatives: seems somewhat older
than thi: finds of the surrounding bill-tops.

To lhe South of Henye, in the vicinicy of Tarcal,

two collecting spms yielded considerable amount of
material:
Tarcal-Kovesd Pb 87/23- 34. {Fig 4. 6-14)
Type distribution:
two flake scrnpcrs (obsidian), high end-scraper (erratic f1inl), burin-edge core fragment, conical core,
worked fragments (4 pieces). blade and. blade-like
flake (5 pieces). flake: 36 pieces (4 obsidian. 2 erratic flint and 30 hydroquartzite)
Tarcal-Oeak hill: Pb 87/21-22: a beautiful endscraper on blade made of hydroquartzite and 13 flakes
(12 hydroquartzite, l obsidian.)
There were two more unidentified collecting spots
at Tarcal, which could be any of the mentioned sites
and I or new hiU-tops as well:
Tarcal. half-way between the Roman Catholic
church and Hcnye hill, collected <luring the excavaLion ofVertes: 3 flakes and a wide, sbon end-scraper
on flake: Pb 64/499-50 I.
Tarcal, hill at the road bend, on the south.cm side towards Tarcal (above cemetery) 9 pieces of flakes were
collected made of mixed raw materials during the
1982. excavation season:
Inv.nr.Pb 83/223
The mounrain Kopasz itself is already above the
altitude level preferred by the Graveuian entity MUP
I (Pavlovian) population. Two objects marked as
Kopasz Mt.'' in the collection of the HNM could
originate from anywhere..
A beautiful, arched silex blade chipped off from a
core rim was donated to the HNM by I. frits, student
of pharmacology (DOBOSl 1975, 68), funher on, M .
Pecsi collected tlakes at a more closely unidentified
spot from Layered Loess over the uppermost fossil soil
horizon in 1964.
The southern foothill regions were not systematically surveyed as yet, and no information is available
on finds collected by others. The eastern slope of the
Kopasz, however, is panly natural and partly artificial
but altogether sloping very steep almosL vertical till
the Bodrag river. This wall is the famous Lebuj-bend,
locality for perlite and "marekanite", pearl-like glass
balls of a few mm. This side was forme-0 artificially partly, washed away by the Bodrog, partly disrurbed
by tbe road construction, with all possible P!cistocene
sediments desLroyed.
Sites on the Northern and Westem foothill slopes
belong aJready to the administrative custody of the village Tarcal.
On lhe northern slope, finds were coUected by
Gabor Gyombola on the Deal< hill and the Kovesdhill around the beginning of the 1980-ies. On these

11

10

6
8

11

12
fig. 4 Stray finds from the environs ofHenyc

cwo collecting spots, same as on Kastelytabla, Kavicsos and Oereszla, material similar to the Henye site
was collected, as much as we can say on the basis of
12

the scanty material. Quamicy of finds and extent of

area covered with finds cannot be compared to Henye.
A possible appealing imerpretation of the phenome-

noo can be that the Henye hill (I 98 m. a.s.I.) was surrounded by, on the lower hill-tops about I60 m altimde in a circle of cca. I km by satellite settlements.
Towards I.he North-we..'it, h:dfway between the
Ti.treat church and the Heo ye hill, in the loess wall
of a deep-cut road archaeological finds were found, an
end-scraper and a burin. This datum cannot be specified more closely.
To the Wesl, in the Tarcal - Brickyards. palaeonlological excavations by Denes Janossy resulted in
some obsidian- and silex artefacts (p.c. by D.
Janossy). Finds of uncertain stratigraphy were found
in the covering loess of the Railway station quarry,
some 200 m of il.
At Tarcal-Citrom banya (''Lemon quarry"), on
Fekete-hill, de1.ails of an Upper Palaeolitruc settlement
were excavated. The pyroxene-dacite quarry was
opened ar a "side hill" of the Kopasz Mt., about the altitude of the TTenye hill. The original surface was
about 215-220 m as.I. The original surface could nor
be reconstructed due to earth works preceding the
opening of the quarry. ln the covering loess, E.
Krolopp discovered archaeological finds in course of
sampling for malacological investigations, that were
later aul.benticated. Stratigraphical sequence: Pyroxene dacite is covered by 200-230 cm thick Late
WUnn loess, comprising two charred stripes in the
length of some 26 m in the depth of 140-180 and
80-1 00 cm, respectively. The lower stripe is archaeologically sterile, the upper charred stripe is a cultural
layer. [ts clay is slightly loamy, the time of its deposition can be placed to some Late Wann intersradials
(Laugeric>Lascaux?).

Fauna.
Vertebrates: Equus. Rangifer (JANOSSY 1975,
26.)
Mollusca: Pupil/a muscorum, Chondrula tr1dens
and Bradybaenufruticum, denoting relatively mjJd
climate (KROLOPP 1975, 28.) The slightly loamy
embryonal soil and lhe "mild'" malacofauna mutually corroborate the chronostratigraphical position
of the cultural layer.
A.nrhrakotomy Pi1111s cemhra (STlEBER 1975,
29.)
Archaeologicalfinds. (DOBOS! 1975, 9-25.)
35 too ls (end-scrapers, rabots, scrapers and ret-0ucbed blades)
25 blade
314 Oake
Dentalium and Vermetus '"trinket snails"
The numerous collecting points are material proofs
of the apparent attractions of the area in some periods
of the Palaeolithic, nornbly in the Pavlovian period.
One a fow square km, an unparaUelcd settlement density, habitation centre came forth. Were these settlements indicated by the surface finds of identical age or
not? And, jf they were strictly contemporary, were
there any rank order, functional, spatial or any olher
similarities or differences among them? These questions cannot be answered as yet.
According to the subjectjvc opinion of the author,
the hill-top settlements in the immediate vicinity of
the site Henye were contemporary satellite settlements
while on tlie slopes of the Kopasz ML, settlements of
younger age are also occurring (e.g. Tarcal-Cicrombanya)

2. R esearc h bi.stor y oftbe site Bodrogkeresztur-Heoye

The earliest record in the inventories of the Hungarian National Museum regarding Bodrogkereszcfu:Henye is lhe entry 33/1948. In the year 1920-2 l.
from the excavations of Jeno Hillebrand marked
"mesolithil'llID" . This excavation could be the campaign of Lajos Bella and Jeno Hillebrand in the Copper Age cemetery of Bodrogkereszrur (BELLA 1923,
7.). Probably, Hillebrand who had by that Lime considerable practice in Palaeolithic excavations found lhi.s
site in course of field survey. The Copper Age cemetery was excavated on the estates of count Ernest
Szechenyi-Wolkenstein in the autumn of 1920, spring
and autumn of 192 1. A. E. van Giffen from Groningen
took part on this excavation. K. Wolllik informed us
that during her study trip to the Nelherlands in 1987,
she came across 'Mesolitruc" finds from Bodrogkeresztlir in the university collection ofGroningen. Thus
we can justly suppose that 1-he note in the Palaeolithic

inventory book. "excavation by Jena Hillebrand" relates not the Henyc, bur the Copper A gc cemetery.. In
tile publicalion of the site, however. no traces of finds
from other (Palaeolitrhic or Mesolithic) period were
mentioned
This inventory entry (and the stone tools) draw tbc
attention ofLAszl6 Vcnes to Lhe site when he was revising the NE-Hungarian data in the Palaeolithic deposltory of the Hungarian National Museum.
In the s ummer of 1963, Llszl6 Verres entrusted
Lajos Toth, met.allurgical engineer and private collcclor from Miskolc to perform field survey on the area.
According to T6th's data, there were surface Palaeolithic finds on Henye, the slope of the hill witb the
TerCzia-chapel facmg the Henye bill as well as on the
Cscngos hill (unidentified c-0lb:ling spot). As the top
ofHenye hill was intended to be planted by vineyards,

13

the area was deep-ploughed again and the near-surface

culturaJ layer was threatened with destrucrion. Therefore Vertes planned authenticating excavations on the
area (HNM 1966).
Between 1 October - JS Octobe1 1963 -, Laszl6
Ve1tes performed excavations on the sit.e (HNM
1966). Vertes aimed at finding tbe original ca:tnp site
on the many times ploughed, disturbed area, a detail of
lhe settlement o.r at least the cultwal layer. He set out
trenches and sections in a roughly 100 m circle around
the geodetic spot height (concrete block), comprising
altogether excavated 165 m2 surface (Fig.5.).
His inve~tigations yielded partial results:
+ he could not find the central area of the sert.lement
or, judging from the results of his research, they musr
have been on the area most disturbed by erosion and

soil cultivation

* the archaeological

material collected from the

present surface and the settlement surface (cultmal
layer) was uniform, found in surprisingly large quantities, of exeellent quality both in technique as well as
.raw material
* the accompanying fauna is tlOt rich but adequate
for a bioscratigraphical classification of the site
charcoal co'Uld be collected for radiocarbon dating
* during the excavations, the site Kavic"'5banya,
later Kavicsos dulo could be identified (field survey
by L Horvath).
By the end of the excavations, Laszlo Vertes made
the following remark in the excavalion diary: lo the
north. north-west and south-west of his sections, towards the slope of the Henye hill covered during the
time of his excavations by lucerne 011 the area covered
by thicker loess there was still hope to find undisturbed cultural layers. The cam.psite must have settled
there not on the andesite but the loess thus settlement
features might be observable. He did not anticipate
much chances for further work because the area was
planned to be planted by vineyards, preventing
chances for research for years and decades. The hilltop must have been cultivated or disturbed already in
the time of Jena Hillebrand as he did not consider the
area suitable for fwther research. Under the "charm"
of the still fresh Cave Palaeolithic, he did not even
think of authenticating these surface finds. The forty

years of soil cultivation between Hillebrn.nd's collection till Ve1tes's excavations were obse-rvable in the
form of mix~d, dislllrbed stripes of humus in the se.ctions (VERTES 1966, 3.). The planned vineyard plantation; however, was hdd off.
Jn 1977, the author performed a field survey in the
region together with Agnes Salamon for the authentication of the site Tiszaladany-Nagyhomokos, Orgehat.
14

On the way, they visited the Henye she. It was found

thar not only the deep-ploughe-0 area remained a
waste-land but also the luceme fields chat were in use
in 1963 became wild and was not utilised for agriculture. Finds were collected on the surface.
fn "1981, a:rcbaeologist Magdolna I-:lellebrandl from
Miskolc perfonned rescue excavations ar Bodrogkereszror-Dcrcszla prior to the construction of Lhe village water p~t station and c01lected, among others,
Palaeolithic finds as well. Slie was informed that .the
idea of planting vine-yards to Henye was raised again
and soon the preparation of the soil would be started
( that is, a new deep-ploughing could be expected).
Thus the continuation of the excavations started by
Laszl6 Vertcs became topical.
This work was performed between 28 June and 22
J uly, 1982. Costs of archaeological excavations were
covered by J6zsef Fulop, chairman of lhe Central
Geological Bureau (Kozponti Foldtani Hivatal).
During the excavation, sondage was planted around
the flat plateau of the Henye hill, though U1e suongly
disturbed artifieial surfa.c e did not give any clues for
i:be reconstru1;1rion of the original surface. The extent
o f the site could only be.estimated because the surface
eccurrenc.e and coocentraiion of the finds had been
strongly influeneed by multi.pie <Lisrurbances. The
Vertes's "foe point1', the spot height could be found
broken on stiU identifiable place.
Excavation trenches were planted at three units:
(F;g. 5)

Bodrogkereszti1r-H.e11ye l emit
- to the north and east of the spot height, partly intersecting section A by Vertcs. This was necessary to
be able co eonnect the new excavation areas to
Ve1tes's sections. 8 exploring trenches were opened in
the length of 80 m . On finding the cu1tura11a.ycr in the
second rrench, A-B-C-D-E were opened to trace the
settlemen~ surface. During the 1982. campaign, altogether 42 m .. adjoining settlement surface could be
opene<lin cheA-B-C-D-E sections.

Bodrogkeresztur-Henye 11. unit

Abouc 80 ro the NW from the spot height 4 exploring trenches were opened, 20 m long each. Scanered

sertlement foatures, some in situ bones and stone tools

were observed as well as some surprisingly large
blocks of stone.

B odrogkeresztur-H enye lII. w1it

4 exploring trenches to the W of Venes's sections.
Mammoth tusk found in trenches I. and 2. might have
belonged to tbe same animal the mandible of which
was found by Verres in his section E.

~C

UNIT IV.

BODROGKERESZTUR-HENYE
EXCAVATIONS ( 1963,1982 )

o t~

//

~/ /
~ V~RTES 19&$
5-

~ 0080SI 1982

I-...

//

//

F fl
heart!\

~_,. I

~,~ ~
M

: ':

Hltlmnt

Mmmll thu
1 8 t 111nc hu

UNIT I.

.o point

-,t~vfwrr~ff+,,
M

2-

~- ;-;1 \'

~ B
1 ~--~

fl

1:
Ii

11

ij

:r
!1

.."u

..~

. , ,,,

.,~f : ~'~ . '~>,,'t,.

>'] I ' 111 , ,

'11

fl

''*I

41

/... : ',i.; ..,'!,~~'vt.,

~ ~
,,;f ~'
' , t ''

UNIT Ill.

__=--::J

., ,

~! : : :;';_,;,:,;.~,,;~:; ~
:'~ ~ ~. ~ ,~ ~I

:: ',
I''

:~ :
,.,
f'I

tit

Ll;

,, :

't

:, I,
I

l:
l!

N-

UNIT II.

Fig. 5 Comprehensive site map ofthc excavations by L:i.<r.ll6 Vertrs ( 1963) and Violn Dobosi (1981)
Ui

10m

Excavation s in 1963. and 1982. opened up at

Bodrogkere ztur-Henye altogether 425 ml s urface.

In course of the 1982. campaign, coUecting spots

from fonner survey were identified on the hilJ-tops

lying 20-30 m lower than the Henye site and coUected

further finds. The strayftnds could not give any data
for a possible interior chronology. All we can say thac
they do not differ, in ra\I\ materi31 or technique, from
the Henye material.

3. Archaeological finds (Photos 1.-4.)

All finds from Bodrogkereszn'.lr-Henye (including
several field surveys, collections and 1he two excavations) seem to belong to one culrural layer of one settlement. Therefore tools are presenccd cogelher. Comments by Vertcs on individual tool types are quoted.
3.1. Tools
( S-B number refers Lo type list nr. according to O.
de Soooeville-Bordes-J. Perrot)

End-scraper on blade (S-B J - 2 type)

75 pieces (Fig. 6.)
.. One of 1he most common implement types is 1he
end-scraper (21%). They are generally made of unretouched blades bza occasionally of rerouched blades
even on the end of Aurignac1an blades. There occur
oblique-edged shaped and short end-scrapers 100. The
''carinaled" rypes .are also made on blades".

(vERTES 1966, 10.)

Base fonn: Mainly regular, with triangular or trapeze cross section, on average, 30-40 mm long blade
which can be slightly arched. The detachment of the
blade is generally of Upper Palaeolithic character but
clactonian technique is also met. The bulb can be
thinned. Depending on the raw material (mainly in the
case of obsidian lumps) they can be slightly arched.
There are blade-like slices as well: one of the dorsal
planes of the triangular cross section blade is covered
by cortex.
Working edge: most typical is the arched working
edge at the distal end of the blade. H can occur also at
proximal e nd. There are some straight or concave
edge end-scraper on blade as well. A local speciality is
Lhe steep (high) scraper edge with projecting front
( 12 pieces), Aurignacian type elongated nosed endscraper (3 pieces). l 2 anefacts have asymmetrical
edge.
Others: side of the blades is more or less worked,
commonly with conlinuous or scattered line retouch.
in 4 cases with step-like (Aurignacian) retouch. The
other end of the blade, opposite to the working edge
can be straight or obliquely struck. Among the two
shouldered end-scraper on blade, one is fom1ed by
notches, the other dorsal "alternating" retouch.
16

Double end-scraper on blade (S-B 3)

8 pieces. (Fig 7. 2. 7)
Arched-arched or arched-straight working edge
combination made on regular, generally shortish
blade. Margins can be .retouched. Two objects deserve
special notice: on one too~ by Lhe combination of two
slightly arched and one deep concave scraping edge a
thick "L''-fonn, triple scraper was made; on the other,
alternately directed double concave working edge was
fonned on two edges of a thick prismatic flake .
Average length of the tools is between30-40 mm .

End-scraper with broken arch ( ''ogilvy" scraper) (S-B

4)
I piece
Regular scraper edge on retouched blade with uiangular cross section, lhe proximal edge struck off.
Hydroq uartzite
40mm long

E1ul-scraper on retouched blade (S-B 5)

This type was not separated because this type group
is only a variant of S-B 1-2 wiLh linle cypologicaJ
value. As it was mentioned at S-B 1-2, different working of the blade margins is frequent and is done ro different degrees of perfection
End-scraper on Aurignacian blade (S-B 6)
S pieces
(Fig. 8. 1. 2 Fig.JO. 2,3.5,)

The tools classified 10 this typological category

could be fit on the basis the q ua lity of the working
edge into the former end-scraper categories. Their
separate treatment is necessary because of the Aurignacian type step-like retouch we can find on these
tools (or, blades) is one of the diagnostic elements of

the archaic character of the industry within the Gravettian circle.

These scrapers are generally characterised by careful workmanship, neat. symmetrical finish. The working edge is arched or double, also containing fonns
with projecting front, steep (perpendicular) rerouch,
3nd also pieces where the working edge is placed on
the proximal end of the blade.
Average length between 40-60 mm.

End-scrapers

17

19

Serpentine disc, chopper, triangle

20

Fig. 6 End-scropers on blade. Scale I I

Fan-shape end-scraper (S-B 7)

3 pieces (Fig 8. 7, 8)
Made partly on already fan-shaped short. high
flakes. partly on Clakes made narrow at the proximal
edge by recouch or strike-off They are steep. slightly
arched edge scraping tools.

Typically, they are about 30 mm long. wide

flakes.
End-scraper on flake (S-B 8) 37 pieces
(Fig 9. 1, 2, 4. 5, Fig. JO. 6, 7. 8. Fig. 11 2 .5 6)

"FlnkP scrapers are alsofreq11ent 8111 011 the other

hand the "oul'ignacoid'' true carene forms and the
21

Fig 7 End-scra~r:. on blade. llnke and retouched blade. Scale t 1

.. ogivals" are missmg. There are altogether 3 nosescrapers in the finds bur even. these are not characteristic. Some end scrapers having a protrudmg tip on
their front above the scraping edge due to too steep
flaking resemble the types knownfirom Arka."
rVERTES 1966, JO)

22

Base fonn: typically. wide, large, rough flake .

There are core fragm~nts or regular cylindrical cores
split in two in Hs lengU1, slice pieces (made of pebble)
or corticated slice-like flake. core basis or prismatic

raw material

piec~s.

,
2

Fig. 8

End-~crapcrs on

retouched blade. fun-shaped end-scrapers Scale 11

Dei.achment of the nakc can be of Upper Palaeolith1c characr(}r, diCdrc and also clactonian as well.
There are obliquely directed 'dcjete" type flakes as
well Cross secLion is typically uneven. both in fonn
and thickness.

Working edge: directtld at variable parts of the bu.;)~

fonn. can be lobatc, U-fonu, arched. elongated and in
broken arch (ogival), straight (within the latter. perpendicular or oblique lo lhe longiludimtl axis of t11e
flake).

23

7b

L__~~~~~~~~~~~~~~~~~~~~~~~~~~~~J
Fig. 9 End-scraper on tlake. Scale I: l

The working edge can be placed at lhc proximol

edge of the flake, the tool can be hafted ("hafting" by
occasional or natural cleavage surface of lhe a flake) .
Recouch: t:radilional fan-shaped retouch, line.ar retouch, on the margins, occasionally bifacial and alternating retouch could be observed.

24

Variance of the dimension is great, this form also

comprise tools exceeding I00 mm.

Thumbnail or semicircular end-scraper

No significant difference from S-B 8, only the base
form of che tool is shorter and wider and the working

7
Fig I 0 End-s1 r.ipers on retouched blade. fla1<c. ~\:.lle I: I

edge is more strongly arched. It is not a characterisLi.c

type group. seems in most cases rather "'ad hoe" solution: the otherwise naturally arched cleavage edge was
worked by lobare rcrouch. lt is f.>l>!>Sible that Ulcy
licrved not scraper function. The Oakes are high. the

roiouch is su:ep or linear, there are several items with

steep I projecting front.
Carenoid end :;(roper (S-B J1-J 2)
6 pieces (Fig JI 2. 6. Fig. J6. 6, 7, Fig. 23. I J

25

2
1

Is

CJ

Fig. I I High end-scrapers, end..scr:ipers on Oll\<c. combined tools. Scale I: I

This type group is decisive in the cultural assign-

ment of the industry. The quality of finish is not as

high as in Lhc: Aurignacian industries but the carenoid
character of the tools is cvidenc. Among Lhe tools of
archaic character we can find two split longitudinally,

26

on one of them a double burin edge is found opposite

the scraper edge.
1t should be noted that on the two Hungarian Aurignacia.n cave sites, classified with certainty to this industry on the basis of split base bone points we can

1t. -

Y.II'..-ii,...;~ :S

'

- 1I

J! 1il' ~

2
Fig. 12 Rahois (planes). Scnle 1: I

11nd no carenoi<l type lith1c industry In this aspect, the

lithic industry of Bodrogkert:sLlur - Henye with this 6,
no! very Lyptcal carenoid end-scrapers 1s more "Aurig-

nacian-like" than the stone 1oois of 11e .:ii j,. ur!gnncinn sites in Hungary.

27

Fig. 13 Rabots (plane~). Scale l: I

Nosed end-scraper (S-B 13)

7 pieces
The Bodrogkeresz1l1r items ditrer from the classical
"grattolr museau" Lhat they are less carenoid-like: lhe
slightly nosed working edge is mainly formed on

28

flatl'mince" end-scraper on blade. Among lhe tools of

this group we rind an angular end-scraper: the nose is
in the right dislal comer of the tool.
Rabot (plane) (S-B 16)
19 pieces (Fig. I 1. Fig. 13. Fig. 14)

" ... "Rabors" are verv typical. They are also in most
msrances combined with hurm., doubled or ma)' even
have side-scraper edges. We find .wnilar 1mple111e111
the Arka industry. There are giant forms too which
are, however still smaller than those ofArka. One of

the characreristics oj tlte Bodrogkereszttir md11str; 1s

tl1e existence of the imermedial forms bet111een the
"robot wid tlte.~e carmate types ofhem'Y burins "
(Vl~TES 1966. 10.J

Fig. 14 Rabots (pl..ines). Scale I: I

29

Fig. I S End-scropcr-burin combin:uic:in. angular bunns. Sc:ile I : I

Base form: most typical, naturally, core: large,

conical, cylindrical or multiple cores but they can be
also found on rough nakes, pebble fragments and
core remnants, core bases cut in dHferent ways.
30

The working edge can be simple or double, within

the latter, parallel or allc.:mately worked.
The working edge. independent of its din.:ction to
lhe axis of the tool, can often be lobate or c:onca' e

6
Fig. I<> Borers. end-scrapers, comb111cd tools. Scale I: I

The degree of the cutting edge is mainly less than 90.

If the rabot edge was formed on the basis of rhe core,
lhc prepared core rim was lhinned from the dorsal site
b)

one blow.

Retouch: primary working is rarely rdined further

by linear or marginal retouch. The working edge, in
most cases is a lobate margin formed by chopp~r-re

touch. Prepared raw material blocks I flake cores can

31

be made by some poscerior working especially suitable for rough work.

Others: one of the tools was made of "stone marrow" with some defecl. By roughing out the base
form, a handy haft arose rendering the "raboc" a very
suitable tool provided its function and use really correspond to modern planes. (Fig. 12 JJ
Average length: 60-80 mm
Raw material: The correlation of raw material and
type is generally not significant, only the finish of
more fine - more rare materials is more careful. In
case of the rabots, however, the so-called stone marrow" occurring on geological source close to the site is
seemingly the most common raw material. 1t was an
ideal raw material for this large and not very minuteously worked tool type.
End-scraper-burin (S-B 17) 30 pieces
(Fig. I 5. 1- 3)

Base form: It can be a chisel/plane edge core (3

items), traditional blade (7 ilerns) and different nakes,
partly worked in the style of a scraper: 20 items
Working edge: The scraper ooge can be different,
depending on rhe base form: the plane is typically
thinned core base or crest with burin edge. The endscrapers on tlake are strongly arched, sometimes
asymmetrical or thumb-nail form, with some fan retouch or linear retouch. The end-scrapers on blade are
nJso traditional, generally retouched on the blade side.
The burin edge can be opposite the scraper edge or
combined with the latrer, in medial, flat. lateral, diedre
or angular variations. The burin edge can be made on
intact or truncated end of the blade.
Others: This is not a standardised type, there is
great variation within this tool group. Apart from
simple end-scraper-burin combinatio~ multiple combination is also occurring: at the comer of the scraper
edge as well as both sides of the proximal edge of lhe
blade, that is, a triple burin + end-scraper on flake
combination can also be found.
Average length: due to the mainly flake base fo.rm,
it is shon and wide: 30-40 mm.
Borer (S-B 23)
I5 pieces {Fig. 16. J}
This tool type is rarel)' represented in aU three sett-

lement waves of the Hungarian Upper Palaeolithic

period. The existing few forms are also different from
Lbe regular Western European borers or the types
frequently met in the Pavlovian. The base form is a
fine, slender blade or pan of it, tJ1e finish is careful
both at the mnrgin of tbe blade and t.he formation
of the borer tip. The tip is typically straight in the
longitudinal axis of the blade, worked with steep I al-

32

ternnting r~touch. There are some forms which are

slightly bent like a gabled chisel ("bec"-like), also
short lips, some can be formed by notches or using the elongated narural cleavages of the blade, making rhe naluraJly pointed distal tip sharper by small
linear retouch. Some lips were completed by burin
chip. It is not a characteristic group of tools but lhe
ones made were carefuUy made and of good raw
material.
Their size is variable depending on the raw material: on average, Lhey are 40-50 mm Jong.
The margin of Lhe blades is typically retouched,
blunted. Aurignacian type step-like retouch is also occurring.
Medial burin (S-8 2 7)
58 pieces, (Fig. I 7, Fig. 18, Fig. J9)
'Approximate(} 12% of rhe industry are burms.
Tlzese are rhe most variable in type have the best
workmanslup and among them the finest are those
made of blades: the terminal angular and side-edged
"diedre" burin.s. The rypes combined l-111h blunted
blades charac1eristic of the East Graveuian ore frequellf, as weJI as 1/ie double and multi-edged burins.
The angle burins made of large flakes - frequenr(v of
tabular flint - aie also characteristic of the indIJstry as well as the similar made /a/era/ b11rins.
Amo11g rhese lauer we find an especially well-worked
s11bground whose implements are probab~'I- rhe most
similar to the so-called "b1u-m carene " of r/Je French
Palaeolithic.
The core burms comprise 4.8% of t/Je industry.
They ha\e varied shapes, are well-worker implements
incl11ding.forms which present an archaic effect. Some
of them were also made in " carene form and cerroin
specimens are gigamoltths. "

rvERTES 1966. I 0.J

Dominant type within che most characteristic Looi

type group, i.e., burins.
Base form: variable - blade, blade-like flake, pebble slice, corefragment or -remnant. Base form seemingly has no imponance for lhe finished artefact.
Working edge: in the axis of the tool, perpendicular
or oblique to lhe planes, in case of high crest blades
sometimes on the crest. Could be produced by burin
stroke as well as lhe combination of retoueh and burin
stroke.
The margin of the blades or flakes can be more or
less retouched. Macrolilhic tools also appear in this
group, separated mainly on tht: basis of large dimensions. Some items which are closer to the average are
also classified here because of their character and hypot,ietical fWlction.

Fig. 17 Ilurins. combined tools. Scale I: I

Oblique b11rin (S-B 28)

5 pieces, (Frg. 10. 2, 7)
lts separation is not justified as quite ofien the
working edge of medial burins is also oblique compared to the ventral plane of the tool (i.e., in acute an-

glean not perpendicular). ln the separation, 1he former

classification of \lertes was followed, but this type has
no statistical value.
ln case of two tools classified to this group 1he burin-stroke was din::clt::d not from above towards !he
base bul. from the ventral plane towards the dorsal one
33

3
4

Fig. I 8 Medial and angular burins. Scale! I:I

34

The "' orking edge is the part between the oblique detachment and the dorsal plane of the tool.

Multiple b11ri11 (S-B 41)

49 pieces, (Fig 22. 3. 4 5. 7. 8)

Lateral bunn (S-B 19)

67 pieces, (Fig. 20. I . 3. Fig. 21, Fig. 22.)

Most \laried group, typical and nicely fini shed, outstanding from the rich set of burins. The separation of
the group applied by L. Vertes:
cwin form: made along two margins of the same
end of the bladt! or flake
double: made opposite or vis-a-vis
combined with slightly arched scraping edge
has no typological significance.

This is the most numerous group within the burins.

It is characteri:;c<l by variable base form and careful
finish . The margin can be rerouchcd, the end blown
ofI: The burin edge can be formed by one stoke or angular as well.
There are some lateral burins made on 1he broken
fragment of some other tools. with "ad hoe'' character.
Care1wid b11ri11 (S-B 32)
9 pieces, (Fig. 23. 4)
The artefacts classified here differ essentially from
the Aurignacian type. Busque" relates not the whole
tool but only the way of production for 1he burin edge.
The base fonn of these tools can equally be on planparallel silex sheet, core or core remnam. pebble fragmcnt. wide nake or slender blade.
Occurs also combi.ned with alternate scraper or
other burin forms.
The working edge can be closed by a notch, but
there is also a Brczillon's "burin busque sans encoche"
on corticated obsidian slice. This type gives an archaic
character to the industry but it is not a dominant feature.
Bunn 011 truncated blade (S-B 34, 35. 36, 3 7)

f'~w pieces are combined with truncation. More

frequently, multiple burins are formed on blades
wit.h muck-off end. Part of tbc multiple burins
were made on thick prismatic base form: they
were suitable for very robust suitable for use with
force.
Double pyramidal or prismatic fom1s: these bast!
forms are characteristic of the site, more typical than
mukiplc burins formed on flar blades or flakes. Some
tools are of alternate edge, thick, s uitable olso for
plane-like function. Tht:se tools were fonned by traditional (though rather large) burin strokes.
$ome multiple burins made on medial fragment of
wide blades resemble NoaUlcs-burins wit:h llie difforence that retouch at the end of the blades is substituted by one stroke.
Analysing dimensions, these tools can be divided
into two groups: traditional combined burins 30---+0
mm long, prismatic, alternating edge combined burins
wit.h plane function belong to the macro-tools with
average length of60-70 mm.

1-t pieces

Not very characteristic type group in low number.

Transversely truncated: 2 pieces
Obliquely truncated: 5 pieces.
Concavely truncated: 4 pieces,
Truncated in arch: 3 pieces,
Part of the rypes correspond to classical terminological criteria. Some of the tools with specially careful finish were made on more-or-less retouched
flakes.
Transversal burin (S-B 38. 39)
I 2 pieces

Made usually on flake. sometimes by dorsal burin

stroke al the proximal enc.I of the nake. Shaping can be
completed by linear retouch in rull width or parts from
the ventral sidc.
If the raw material was obsidian, am ore cage/decorticage flukes are frequent.

Core burin (S-B ./3)

34 ph:ces (Fig 14. Fig 25)
Due to the character of the raw material (hydrothermal silict:s of mainly banked occurrence), these
tools were typical ly made on geometric bas~ forms,
ra\\ material pieces in the shape of prismatic pyramids.
At Bodrogkernszrur, core burin formed on pebble
or geometrical base form with arched, wide medial
working edge were typical:
- on one side: of the burin, with long blade-shape,
arched parallel strokes
- on the other side. the burin edge was thinned with
some nat and wide scaled retouch
This rype of working resulted in a "carenoid" or
similar character.
Occasionally Lhis type occurred in combination,
most t.ypicaJly with scraper edge along the margin of
the nake.

35

Fig. 19 Medial and angular burins. Scale I: I

36

fig. 20 Lateral and core bunn.~. Scalt I: I

37

8
6

Fig. 2 l Angular. la1cral and multiple burins. St:ile I: I

38

Fig. 22 La.U!r:1l, angulnr. fl>ll hurins. Scale I 1

39

1
2
3

Fig.13 Cnreno1d und transversal burins. Sc1.1lc l: 1

40

Fig. 24 Core- and multiple burins. cale I ; I

Within the core burins, multiple burin-combination

occurs frequently with variable burin edges (lateral,
medial, angular, flat etc.).

Flat burm (S-B 44)

12 pieces

This tool Lype is not characteristic, fairly accidental.

The burin edge is mostly blunt (60-70). and the corn-

plete group has some atypical character. One of the

tools (Pb 83/615) deserve special meniioning: slightly
arched blade, covered on o ne side oft.he dorsal side by
cortex, on the distal end, flat burin edge, on I.be ventral
side,
combined with truncation on the dorsal side. A! the
base, on the back side oblique retouch: this piece
is nearest to a Kosticnki-knifc from the whole assemblage. The "relation" is only in type and
41

fl

ID

Fig. 25 Core and mulliple burin.c;. Scale I : I

42

places of finjsh. the type however is only reminding to a Kostienki-knife and not aclually identical
with that
B/ade poinrs (S-B 46-47)
26 pieces, (Fig. 26)

faactly half of this cool group ( J 3 pieces) consist of

"'pointes naturelles (Klingen mit naiUrlichen Spitze)'',
known also from Willlendorf. The pointed blades
were functional without an) further elaboration.
Among the retouched blade points t.hcre are only 6
ph.:ces which can be inserted into strict morphological
type categories (according to the S-B list), i.e., they
arc close to tbe typological criteria of the Chatelperron-knife.
Among the other retouched blade point we can
find tools fom1ed partly or totally "it'h marginal retouch, crested retouch, steep re.touch on the proximal
edge (hafting) as well. One of the tools deserve special
attention: here, the proximal end of the blade was
pointed. The retouch applied in the distal upper third
of the right side looks like a "comra-a-cran!".
Gra,eue points (S-B 48. 49)
14 pieces, (Fig. 27. J, 3, 4, 5. 6. 7)

Among the traditional blade poinlS. Nrs. Pb 83/594

resembling Vachons, and Pb 831710, resembling
Krems type points can be specially mentioned.
In case of some Gravette-points, initial hafiing can
be observed. Some of lhe bases were truncated by
steep retouch. In some cases, the cip of the blade point
was flattened by accidental crested retouch. Retouch is sometimes not blunted but simply marginal.
In other instances the slender blade with nalural edge
is only shaped to form and the retouch is not continuous.
Shouldered-hafted tools (S-B 5 7)
14

pieces

"There are many retouched blades and we find a

few Aurig11acian-rype ones too. The "pi'ece a cran''
jorm is frequent and 011 certain blades even 1he poim
in addizion to the area oftlie base is rerouched, but rhe
true "point 'a cran" is lacking. "
(VERTES 1966. 10.)

Typical a cran-forms, bolh ofEasrern and Western

European types arc missing.
Due to tbe almost identical function, tools retouched from two sides (hafted) were counted here.
At the end of the blnde (mainly on the distal, but
sometimes on the proximal end) a shallow shoulder or

hafling is formed on a short phase by steep rc:touch.

This tang is so short that it could hardl) correspond lo
chis I.ask-: the part observed as hafting could serve
other functions (e.g.. sreep truncation running at an
acute angle to the base of the tool?)
Shouldering is not typical technological feature of
the Bodrogkerl!szrur industry.
Backed blades (S-B 58. 59)

27 pieces
This type can be further divided into the foUowing
groups:
- blade blunted on one side
- blade biunted on one side and retouched (perhaps
with altcmaling retouch) on d1e od1er side
- blunted blade, transversally or more seldom,
obliquely rruncated on one or two end (Recbteckmesserchen). The retouch at the end of the blade can
also be on the dorsal side.
- blunting starts from the ridge of the blade with
triangular cross section
The average length of the complete blunted bladc:s
is about 30 mm. The transversally struck blades must
have had a sin1ilar function to Lruncated blades, the
natural cleavage surface of si lex can give equally useful working surface as blunting.

Truncated blades: (S-8 60-62)

26 pieces
transversally truncated (S-B 60) I0 pieces
obliquely truncated (S-B 6 J) 12 pieces
concavely truncaLed (S-B 62) 4 pieces
Part of the tools assigned to these groups were primary truncated blades, the other part formed on broken tool remnants or wide flakes obtained in course of
shaping a core, a straight, oblique or concave working
edge formed by linear retouch.
Truncation is often accompanied by the accidental
retouch of the margins. Among the truncated blades
we find several with high crest, thick, trapezoid section, prismatic blades, i.e., definitely "heavy" tools,
suitable for use by force.
Blades rewuched on one side (or relouched crest) (S-8
65)
126 pieces
"At ot1r sire blades having a so-called retouched
cresr are frequent. A part of these are .l'11rely not core
edges. "

WERTES 1966. 10.)

43

Fig. 26 Retouched blades nnd blade points. Scale I I

44

,,~
J

c.
It.

10

11

12

13

14

Fig. 2 7 Reluuched blades nnd blade poin is. Scnlc I: I

45

This tool type does not need much explanation.

The blades are on variable base forms with variable
finish.
The retouched crest blades assigned to this groups
are obviously not core margins. They are typical
blades with Lriangular cross section and blunted crest.
This group, together with retouched flakes are the real
..ad hoe" tools where shaping is minimal but probably
j ust adequate.
Blade refo11cl1ed 011 both sides (S-8 66)
24 pieces

Variable base lbnns and shaping, on different

places. Some of thi; blades can be struck off on one
end, obliquely or transversally. Tbis "terminus technicus" is used when the end of the blade is broken off by
one blow, i.e., not by retouch. This procedure is used
in large series and Lhe result is close to cnmcarion. Retouch can be marginal or slightly step-like. both from
ventral and dorsal sides.
Though inverse retouch ~as also found (marginal
or scaled retouch) in several tool iype. ventral working is not 1ypi1:al of the Bodrogkeresztur materia I

blades. can be fow1d. Tliis technique was used for

making the edge and a part of rhe .wrface ofan usually fine mou~reJoid point made of obsidian This 1mplemenr could be detennined as belongmg to the a
face plan group. it seems to J'il the best here."

(VERTES 1966 10.)

One of them is a wide decorlicating obsidian flake

(covcre.d on dorsal side with pebble cortex).
The other one is made of hydroquanzite, obliquely
stroke on the proximal edge and the left margin ~ ith
natural cleavage plane; on the right edge. corticated
surface and steep retouch.
Notched - lobateflake (S-B 74. 75)
18 pieces

Flakes are typically wide, high base form , core base

or core remnant also occurring, often wtth pebble cortex. They are bulky tools, Lhe notch can be formed by
one blow or retouch, sometimes Lhe margin of the
flakes are retouched or truncated.
Piece esquillee (S-B 76)

Blade w11h A11rignaciat1 retoud1 (S-B 67)

I 5 pieces, (Fig 27 9. !O. l 2)
The Aurigna1.ian relOuch is ty11...i,lly charn.ct~ristic
of the older Gra\ ellicn industri~~. thu!> also of Bod-

rogkeresztur. l bis t>-pe grour contain blade 1001:. with

Aurignacian retouch onl) sl culd they r .1n l1thcr
types, they were classified into their own calegory.
Aurignacian blades are typically large, bulky
blades, so-called ''heavy" types.
There are no exact data available, but according to
'the impression of the author both the Aurignacian industries and Lhe older Gravettian industries with
Aurignacian characcer are more large and bulky here
than the neighbouring Slovakian find assemblages. ll
is nor certain that Lhis difference in size relate the average of the industry but it is apparent in some key Looi
types.
The geographical proxim ity and the narrow temporal limits cannot illdicalc that these attesrable differences would mean temporal or spatial separation,
other genetics or differences on facies level. The exact
explanation is not known ycl: one possible answer can
be the accessibiliry or physical qualiues of the raw
material.
Mousterian point (S-B 69)
2 pieces (Fig. ~8. 1. 2)
''Also there a1-e 110 "Solutrean-type " Jeaf-poims b/lt
the surface retouch in a few instances, especially 011

46

9 pieces
Atypical tool type. flat , mainly square or similar
form flakes with c~nain amount of rl!touch on their
margm.
SfJe-scrapers (S-B 77)

37 pieces (Fig. 19. Fig 30)

" The side-scrapers and notched imp/emems a,e
frequenJ. There at'e three eS(/Uil, forms, 0 11e which reminds us of the typical East Gravettian concave-edge
burin. "
(VERTES 1966, 11.)

Important tool type, strengthening the sl ightly archaic character of the industry with significant number
of finds. The base form can be flake, core (Fig. 29 1)
cc>re fragment. pebble slice, or plary, flat raw material
piece of hydroqunrtzite.
The form of the working edge is basically arched.
17 pieces
Concave: 1 piece
Straight: 9 pieces
Multiple: I0 pieces
This later is pointed (the two working edges meet al
90 degrees), angular (at around 90 plain degrees).
double or multiple when the two working edges are independent of each other. On some of the tools, other
working independent of the scraping ~dge(s) is also

Fig. 28 Mo1.1s1croid poinl, pi.fee csquillee. Scale I : I

47

l,

Fig. 29 Relouched flakes, side-scrapers. Scale l :1

48

5
Fig. 30 Side-scrapers. Scale l :1

49

observable: scattered marginal retouch, scaled retouch, rruncation or notches). There are some scraper
combinations, e.g. with burin (Fig. 29. 2, 3).
Rac/ette (S-B 78)

l piece

Quadrangular, flal flake. with alternate retouch of

variable intensity along all four edges, straight and
concave working edges. Dimensions: 25 x 21 mm.
"Others" (S-B 92) were further divided:
R ewuchedflake (93)
60 pieces (Fig. 31)

Same as retouched blades, they are also ad hoe

tools fit to purpose. person and worldng style.
Hand-axe (94)
7 pieces

Cores can be grouped according ro form:

Cylindrical
23
33
Conical
Flat
17
Double
JS
Multiple, core base, core remnants, atypical raw material pieces:
68
Blades

551 pieces
Raw material:
Exotic
"Stone marrow"
Errat.ic silcx.
Mixed silex
Radiolarile
RydroquanziLe
Szeletian fclsitic porphyry
Obsidian

pieces
7
185

49
38
I I)

165
I

87

Flake, !01appi11g debris. blade fragment, blade-like

flake

Not the classical hand-axe forms, more leaf-like or

amygdaloid forms wiLh zig-zag edge line, pointed
bopping-tools. They are suitable, however, to perform
any task associated with hand-axes.

H ammerstones
21 pieces

Chopper - chopping-tool (95)

25 pieces (Fig. 32)

Egg-shape pebbled or pebble fragmen ts with concentrated traces of use on one or two end.

Classical Lower/Middle Palaeolithic types, made

on pebbles. pebble fragments, raw material plates;
chisel-edge artefacts fonned by zig-zag alternating retouch
On several tools we can find two genera.lions of
retouch. (BodrogkeresztUr finds often contain, apart
from the heavily patina1ed Qld shaping, fresh retouch)

Other objects

Pebble fragments

13 pieces

Geometrically split pebbles, slices, segments

Cores

1989 pieces, Made of variable raw materials

At those pans of the excavation where intact cultural layer \\as found, some other non-local ore pieces
(ferrous mica?), red hematile pigments were also col
lecteci
The broken, burnt bones found in ihe hearth burns
were in good state of preservation. Two of these bones
showed concentrated traces of use. l1bey were 7. and 8
cm long. respectively and had been in use probably as
retouchers.
Incised margin limestone disk (Verres s uten1s I moon
calendar)
lnv. nr.: Pb 64/408. (Fig. 33. T- 3}

156 pieces

"The cores are char acteristically Upper Palaeolithic wirh pointed, cyUndrical, double based and
flat forms which we1"e frequently transformed into
scrapers or burins. The boat-shaped core fragment
i.s frequem. it possibly had its own special function. "
(VERTES 1966. 12.)
50

" Among the finds was an object, carvedftom lime

stone. that was shaped like q half-moon or horseshoe,
ic measured 56 by 56 by J7 mm If the object were oriented, the top is "north '' left is "west " and right is
"east " the b~e of the ha/fm oon is "south". Near rhe
cenrer of the northern edge are f)vo near-vert1cal
caned lines. 6 to 7 mm long, that slightly converge ro
the north. Eastward and westward from the Imes. the
sha1p edges of the object are notches 011 the eas1ern

Fig. 3 I Worked fragmems $en le I: I

51

3
4

Fig. 32

52

Choppe~ -Chopping

tools. Scale I: I

----~ .....

--

. disc. Si:alc I. I
Fig. 33 Incised smne andserpemmc

53

side (lnt/ 12 on the western. all no1ches extend to the

reverse side. Parallel with and near the southern edge
is a carved /'ine 12 mm long.
1 chos.e to regwd rhe Bodrogkeresznir object as a
11ten1s symbol, although its lunar or solar shape was
noted. The western of the pair of converging lines
(northern edge) may be regarded as the symbol ofthe
new moon. The 12 notches to the west ofthis /me symbolize the crescent moon. The mmotched southern side
may be represenJ the 3 days ofthe.full moon.
The 11 no1ches on the east Side may represent the
days ofthe waning moon.
Finally. the eastern ofrile pair oflines may symbolize the vanished moon on rhe 28rh day (or 29th day, if
we allocate 4 days to the unnotched ~outhern edge:
the 28. 5 days ofthe lunar cycle may have led primitive
man ro reckon either 28 or 29 days).
This i111erpre1ation does not prevent me from mainlaining thar the object is a uterus symbol, but the moon
and uterus significances may be associated. "
(vERTES 1965, 855-856.)
(See on front cover).
Dimensions: 56x56x 17 mm
The description of the objeol is exact and detailed,
oo more completion is needed. The bright. double interpretation by Laszl6 Verrcs is one of possible many
interpretations.

3.2. Stati~ tica l enluation of the tools

Length dis1rih11tion

Llngth (mm)

6-10

16-20

29

21-25

80

26-30

108

54

- - --

157

41-45

109

5H>O

64

61-65

60

L_

66-70

-- -

38

II -

71-75
76-80

20

81- 85

86-90

91-95

--

II

_J
- ---

10

101- 105

111-115

116-120

Toti!

1124 p1ecc.s

Average:

45,11 nun

11

I'.!

>116

__ l

~..>

96-100

121- 125

---

.,~

167

36-40

Length-width racio ofthe tools:

The longitudinally split oval limestone pebble (Pb

rounded limestoae pebble with traces of use (as retoucher, Pb 64/236.) can be considered as forerunners
of the "moon calendar".
The function of the plan-parallel triangular sheet
of silex with patinated surface, re-shaped peaks (Pb.
6413 I6.) is nol known.
Dimensions:: 110 x 83 x 16 mm.

16

51-55
85
- - -- --- - - - - - - -

Flat. plan-parallel, slightly irregular, ovaloid disc

made of phyllite. Carefully polished on two sides, the
margin is profiled. Some small injuries along the distal
end and rhe margin. Traces of concentrated blows on
both sides, causing pits in rhe relatively soft material.
Dimensions: 92 x 81 x 88 mm .lnv. nr.: Pb 83/583

641258.) withow further shaping as well as the tlat.

31-35

Disc
(Fig. 33. ./- 5)

Retoucher (?)
Oval retoucher made of phyllite pebble, used on
both sides with po lished plane and margin.
Dimensions: 4 I x 3 I x 8 mm
The identical function of the rwo objects seems apparent though the former one is of much finer finish.
Inv.or.Pb 64/80

--

11-15

Pieces

ratio%

J, grovp

n group
m. group
IV. group
V. group
Vl. group
VU. group

Total

100--66
66-50
50-33
33-25
25-20
20-16
<16

pieces

455

41 ,2
18.5

204
263
11 2

23.9

10,2
3.7

42
17

10
1103

1,6

0,9

- - ---100

Bodrogkeresztt'.tr-Henye is a classical blade indusmarked by the base forms of the type list and the
number of the unworked regular blades. The nutncrous archaic tools made on nakes (e.g., side-scrapers)
and the worked I geometrically broken pebble tools
\\hich do not fit in a blade industry modify tbe ratio
numbers.
According co me calculations by Vertes. Lhe length
distribution of (a part ot) me blades at Bodrogkereszrur-Henye is as follows: (VERTES 1966, 12.)
ll),

"transported" raw materials show high technological

development skill. Significant differences between
"transported" and "local" raw material working and
utilisntion was demonstrated by Vertes in course of
the elaboration of the Arka-Herzsnret Upper Palaeolithic sertlement (VERTES 1964165).
3.3. Technology

Apart from generalities of an Upper Palaeolithic

tool-produdng technology, the following observations
< JO mm
0 blades
deserve special atteolioo:
Technological specialities d ue to the raw mate10-'.!0 mm
5 blades
rial a pplied:
20- JOmm
2 1 blades
"Stone marrow", a thick platy-banked (tabular) local hydrothermal raw material was specially important
30-40 mm
67 blades
for people of che Bodrogkeresztlir settlement Conse40-50mm
SI blades
quently, and in the first place, among the anefacts
I
made of this material, high crest, asymmetrical cross50-60 mm
44 blades
sectioa (often c lose to a right-angle triangle) were
60-70 mm
35 blades
,Ja:quent.
"Dli'etoffie natural creavage of "stone marrow", lbe
70-80 mm
18 blades
,___
cores made of originally plan-parallel raw material
80- 90 mm
6 blades
;,..Plates are prismatic, with parallelogram cross-section.
I
_ On tfie surtiice, sTl[x'lumps of3-4cm diameter can I
9G-100mm
4 blades
I
be
collected and wns probably available in those days.
-Their use resulted in the production of shonish bladeTo1;1I
251 blades _ _ .
like implements apart from Lhe classical blade tools.
partly covered by cortex, as slices and segments.
\
Vertes separated 251 blades from the J 963. year
-CalCUThtilrg-lhe ih?erage-length in a traditional way
excavation and performed statistical calculations on can be therefore misleading because this smaller sl.ze
them.
group of implements is connected more to raw mateln the last years of his scientific career. Vertes con- }rial lhan function. The banked hydrothennal silices
centrated on the quantitative analysis of morphologi- available in the immediate vic_tnity o~ si1e were
cal criteria for lithic technology, shedding new light on mainly used for macro-rools. The close analogies of
the development of their makers.
some i'Oagh~i.gantolit:lreS--cc>uld be found in the
He considered probability calculation a suitable material of Arka-Herzsaret, lying topographically not
cool for a mathematical expression of biological devel- very far but chronologically much younger. The
opment of prehistoric people. For the characterisation proximity of the abundant raw material sources lead to
of the stage of development he used a factor con- a "lavishing" use o f the local raw material, hydrosidered important by Vertes, i.e. Lhe statistical distri- quartzite.
bution of tool length data. The proximity of the disrriStriking platform
bution curve to ideal, Gauss-type distribution was con- ~ ~pmi fi"om OieeharacteriS'lic poiilf=IiKe striking 1
sidered as a representation of plan and finish, by the
latform Lhere are also di~dre and clacionia~ /
mind planning the instrument and the .more or less (110-120) and, sporadicaJly wor~~ki!'&._Plal.:l
trained hand to fulfil the concept (VERTES 1968, fonns present in Lhe material. fy'he flakes are often
4-5.)
ortentooooliquely, 1n'Oejcfe" fashion.
As a conclusion of his investigations, Vertes hyWorking edge
polhesed a general tendency that the standardisaWorking edge occurs cypicaJly on the distal side,
tion of the finds increase proportionally to the time more rarely on !he proximal end of the flake. Mulliple
elapsed as the assemblages grow younger. Al the same working edges can be parallel, convergent, and some
timl.!, adaptation to new raw materials is time-consum- rough loots with alternate edge (axe.adze function).
ing and in the initial phase of adaptation, learning how
Retouch
10 work a new material works against standardisation.
Most typical retouch type is direct, but inverse. alMaybc this is an explanation to differences in blade temating and bifacial retouch is nlso occurring The
length distribution, in spile of the fact that traditional mosl typical array of retouch is linear marginal. or

55

specially suited for the given Lool rype (abrupl and

semi-abrupt). Real scaled retouch was observed on
some blades and blade points. Apart from this we can
observe a type of "pseudo-scaled retouch": shallow,
flat retouch running up the steep distal plane of the
tool, resulting in a working edge resembling scaled retouch.
Quite often truncation is subsliLuted by one blow
perpendicular to the axis of the tool. The natural cleavage surface of the blown-off working edge is similar
in its quality to the working edge elaborated by steep.
blunted retouch. This type of shaping can be applied
together with other techniques. combination with
burin edge is especially frequent.
Burins
Burin on blown-ofT blade (s~ above) is characteristic; burin on truncated blade occurs much less.
The natural cleavage surface of obsidian is similar to
the slightly twiste.d burin chip negative. Part of rhe
typical burin edges can be the resu lt of unintentional
burin blows, but some are certainly intentional. The
w~ngnfsome bulky burins are so intensive tliar the
negatives of the aeracbed burin chips f5r a SJioulCJcr or
hafting. ln such cases, 1c cou1o tuive been jusL1fo: fili'fin
eag'?'i-rself that was in the hafting, and it is diffo.:ult lo
see the reason in such technical solutions (Pb 65/ 1269,
64/ l 86, 831273, 318.)
The surprisingly numerous "coarse''. large size
burios may indicare an environment rich in woods.
Part of these bulk-y burms are close to planes: in all
case, they could be used for softer material than bone
or antler. (This stacemom ~an be ~ixcially rel~\'311l for
combined burins over 80 mm with som1. other retouched working edges). Such large burins were made
of local cabuJar raw material, therefore Lhcir form is
close to geometrical (double pyramidal), resembling
most to Brezillon's "burin prizmatique" in technical
renns.
The working edge of lateral burins is often not perpendicular to the base but incorporatl.!S an angle
(around at least 70).
End-scrapers
Apart from traditional end-scrapers on blade, those
on flakes are also frequent. Typical of the site we can
find pieces with discoid base form, with slightly zigzag semicircular working edge, and end-scraper retouch (Pb 6-l/353, 289. 117, 2, 831286, 802, 277.)
They recall the form of chopping-toQI derivatives.
Base forms on blade or Oake have high crest and
steep ventral planes. Therefore sreep. almost rectangular scraper from is typical of this industry. Th~ dorsal
side of the tooJ is sloping steep towards the proximal
side.
High crest and prismatic base fonn favours the
pr~ence of carenoid end-scrapers which add up to the

56

Aurignacoid. or at least archaic character of the Bodrogkercszrur industry.

Also occurring, there are end-scrapers where che
shaping of Lhe working edge hardly exceeds an arched
truncated blade.
Characteristic type of the assemblage is the high
end-scraper, split longirudinally. Such tools resist use
better and probably do nor break easily. Splitting the
high fronted , carenoid end-scrapers was not accidental
l>ut intenLionaJ.
Planes (rabots), chisels
Similar to large, bulky careooid tools thru were,
probably, used for wood working, just like burins with
similar shaping (Fig 12- 14). Together with these, the
shaping of the margin of the cores can be hardly separated from the preparation of the core base. The intention could have been difforent, but the resulting raborlike core margin could be similar at I.he two working
stages.
In Hungarian technical Lermmolog)'J due toJhc oot
very forrunate translatiQP qf "burin". Pan of the Hunganan experts seem to be crouoled by the translation
of the term ..burin" (arveso),calling the tool t) pes
ciseau - tyteissel and burin - . Stiebel by the same term
"veso which is al leasL misunderstandable. They
should be diffcrentialed at least for the different function, therefore it is important 10 separate the cwo types.
Chisels are typically bifacially worked tools of 6-8 cm
size, the angle of thc working edge is smaller than
right angle (on Lhe average, about 70), the working
edge is in the plane of the blade/flake, having a
wedge- like function. The working edge of the burin is
formed perpendicular or at an angle to the plane of the
blade, with typically carving function. Wilh a little effort. flat cores. core remnants can be shaped to this
type by flattening the distal or proximal end. At the
same time, core preparation or use can result in :,imilar
forms. The two fi.inctions do not exclude each other.
Geometrical lameUne
Not mentioned in the type list, medial fragments of
bJades or par~ of them might have bad lhe same role as
geometrical retoucbed blades. (See also comments on
retouch; the perpendicular blow can have the same result as blunting retouch)
Cores

Apart from tradiuooaJ Upper Palaeolithic core

forms, the high number of prismatic - pyram idaJ cores
shou ld be emphasised as a consequence of Lhe spe~ial
qualities of the raw material, as well as the gcneraLiy
tab1.1lar raw material pieces, ill fitted for the production
of traditional blades . Diagonally split base fragments
of cylindrical - conical cores are frequent.
There are numerous flat obsidian core remnants
covered on one side by cortex with fine blade flake
scars.

Typological-techno logical observatioos2

256 pieces of typical tools were analysed, selected
from both excavations and the surface collections. The
distribution of the selected ru:tefacts, according 10 Ira
dition:il archaeological typology is as follo-.vs:
I 08 end-scrapers
I 01 burins
11 end-scraper-burin combina1ions
11 rabots
17 blades
8 worked fragments
Raw material distribution:
203 pieces of mixed silices
53 pit:c~ of obsidian
The aim of our investigations was 'tO compare technical characteristics of recent cools and their hypothe
sed Palaeolithic equivalents. The fact chat there are
contradictions among the technical tenns used for describing some artefacts and their actual function has
been demonsrrated recently by se\leral traceological
studies. (Fig 34)
The basic process of tool-working and shaping is
cutting (slivering): from the object to be shaped,
small1.:r or bigger parts are trimmed off using different
lools.
The most important condition of the success of
work is the selection of working angle of the tool used
depending on tJte hardness of the piece to be worked.
The relation of the cutting edge angle of the tool and
the hardness of the material to be worked by this tool
is an axiom, which can serve as a suitable starting
point for detecting the rea.l function of the tools. Hardness of different materials is given by technical literature either as abrasive hardness (Mohs' scale) or resisrance to pressure (Brinnel's hardness). Quire often,
some individual experimental series are used as "etalon". For bone and antler, typically not their hardness
bul the bending strength is studied, because they are
considered not as lhe o bject of the working procedure
but as a tool or part of a composite tool.
The variety of indices fow1d characterise different
qualities of the material thus these values cannot be
converted to each other. On our graphs. on the hori
zontal axis Mohs" hardness considered. As Mohs
hardness is proportional to pressure resistance, Brinnel
values were exLrapolatcd.
The culling edge angle of archaeological types was
used to postulate the material worked by the tools. On
the basis of analogy between archaeological types and
recent tools, some modifications of the typology can
be suggesred.
We can observe the following on the graph of c utting edges:

2 pa.niaJ results published in DOBOSl- HOMOLA 1989,

I. The working edge of end-scraper inlays is suit.able

for making round sections and notches is best
suited for working on wood and bone. Most of
the cutting edges feU into the domain of hard
wood-antler hardness. (Fig. 34)
2. Working edge of different categories of burins are
more steep, the cutting angle bigger and the
working edge shoner. (Fig. 35)
Higher angle indicated harder mati:rial to be worked:
these edges could be adequate to work on fresh
bone.
3. Obsidian tools of both type groups started with
higher cutting edge angle: obsidian is more
rigid, fragile and sotf than silex, by the application of pressure the cutting edge gel injured
more easily.
4 . Working edge of tools made of obsidian were, by
both type groups, in a significantly more narrow
interval.
5. The incrementarion of the angle of working edge is
not evenly distributed. Disregarding casualties
and r.he possibilities of mistakes il seems that lhe
cutting edge was incremented, as a rrend, in
fi ve-degn:e steps.
3.4. Co mpar ison between traditiona l archaeological types and their 1ccent equivalents (Fig. 36)
The most typical Upper Palaeolithic type group~,
end-scrapers and burins were. due to their high ratio in
the lithic assemblage, probably Lhe rnosr typical working tools.
Using modem terminology, chisels
are cutting(slivering) tools wir.h one edge (Fig. 37). The edge
may have one or t\vo planes, lhe line of lhe edge
straight or arched line straight ones can be of right angle or oblique. The character of the working ability of
the tool, the cutting angle is determined by U1e user
how the chisel is held in the hand.
Chisels are grouped according to Uteir application:
Flat or ca1penter s chi.sel - this working tool correspond best, in archaeological typology, to "Meisselciseau-doloto". The working edge is in the dorsal-\lentral plane of the lool, opposed to most implements
called Chisels today, which are the descendants of is
archaeological ''Stichel-burin-rezec" and lheir derivatives: mortise chisel, plugging chisel, gouge or wedge
chisel.
Wedge chisels occupy a considerably important
place in the Palaeolithic tool kit as seeo in their occurrence ratio. Tts function is still the same as in the Palaeolithic period. The diITerence between a Pa laeolithic and a modem wedge chisel (inv. nr.: Pb 64/456)
is in lhe unworked interior triangular profile (Fig 37),

37-57.

57

:1

angle

110

!I

CUTTING

ANGLE

OF

END - SCRAPERS

~I
I

:1.
.,, i

100

:::,''

90

II

.,,

.7:1:
:::,+

80

.
I

..di++

.... ++

.r

-:zr+H++
.I

::++
/+
.

.,..

70

:: .....

r ...

+
+

60

'
'

..,
...,

/+

50

1+

/+t

40

s1 lex

+ obsl d lan

11:
I .I

.,+

30

1,

',.

20

,1

I I
10

I
I
ha rd ness / Mol'ls

Pin1m

Picea
Lari x

0
~

Acer

2
bone

Cll

0()

4
E

::>

Ill

Fig. 34 Cutting angle ofscrapers

58

...
Q

..,.

'
'+

angle

110

CUTTING

ANGLE OF

WEDGE . FLAT , PROFILE

.,/

CHISELS I BURINS

::;Ju+

100

..I+
,+
J!C+++
..:&: +++

90

: - /++

""'::

80

t+

"'1
,,+

70

4'++
I

. . ., ++
++

60

.;I+

'

50

ii+
..:t+

40

1.I++

-,
.,

sllex
obsid ian

I
I

-=rt

30

.,
I

20

I
10

I
I
hardness I Mohs

Pin us

Pieea

Lari)l
Acer

Fagus

3/

...
!!
c

:::

0
0

..,
~

qi

...

Cl.

Fig. 35 Cutting edge of ahlscls - burins

59

chisel

(A rchaeological)

Stl'tl.ight

cypes
endscr.lpcr

tlnt

silex

gnuftc

with
handle

gimlet

cndscraper
-butin

19

obsidian

other

cut

wedge:
'

spo~-

have

selling

50

stlex

ho we I drill saw

I -

obsidian

11

sife:r
>-

I
I

sllex

obsidian
TOTAL: 256

knife

angle

26
25

obsidian

robot

wilh

v..
shape

26

obsidian

under-

,,

3i/ex

sf/ex

blade

plane

reedplane

she -

obmlion
bunn

Experimental-fuoctionaJ type5

Traditional

21

2
2

40

51

93

I
16

10

Fig. 36 Archarologic.al types and hypothesed functions

as the metal working chisels a lso folio~ the Palaeo1.ithic fuU form.
Among the end-scrapers, the double end-scraper Pb
83/582 was analysed in (Fig. 37) . Its raw material is
Szcletian felsitic porphyry. The blade base form is of
trapezoid cross-si.:ction. ut the distal end containing a
75 cutting I slivering edgl! and a i.emicircular styleedge, at rhe proximal 1.:nd. a 82" cuuing slwering
edge and a straight I obliqul! style-edge was fom1ed.
This tool can be used in two ways:
- Taking io hand the margin of the blade, sliding
a long the smooth, flat ventral side. Working on wood
or bone, forming or innoothing a plane, the straight
working edge shou Id be used as the convex working
edge would be cootacled with the surface ro be shaped
on a very small surface and the effectiveness of the
work would be irrationally low. Using the traditional,
even end-scraper edge, a concave trough could be
deepened. The modem correspondents of this function
are called mortise chisels with different arcb radius.
- The other possibJe way of use is taking the tool at
the dorsal-ventral plaue and slide along the 75 and
82 degree scraping edge. Using il with fo rce this way
would hit the band against the substance to be worked,
and the p ressing force should be extended over the full
length of the blade to get it Imo lhe material to be
worked. This is not imaginable for a tool of this size,
therefore ii is suggested chat in this function it could
be used onJy hafted as a scraper inlay. Thus, similar to
most of the modem cutting I slivering cools, it was
functioning as a composite tool. Modern tools are
made, starting from the hardened working edge till the

60

tang used for fixing the handle are made of one material. while the. handle is made of a sofier, more elastic
material to resist more forces encountered during the
working process.
As this end-scraper on blade in the latter function.
most of Upper Palaeolithic - and. generally speaking,
Palaeolithic stone tools could nol be simple tools
(made of homogeneous material alone). judging from
size alone. Apru1 from the shaping of the working
edge, a firm grasp and optimal transfer of work.mg
force had to be ensured, occasionally by comforting
reoouch. The tools which were used, deduced from
small working edge angle, against very hard materials
could only function as stone inlays in composite tools .
Fixing the srone inlays into wood. bone or anrler
socket, apart from sticking probably U1e stretching
srress encountered as a counteraction to slivering was
also u tilised.
The analysis of robots (planes) proved that the
tools classified to this group according to traditional
terminology were probably used in such function.
They are tools held directly in the hand suitable for
working 011 a simple ( level) surface. Jes variations an::
the following:
- arched edge, narrow plane, the cutting angle is
45 (resulting in concave ribs on Lhe worked material)
- smoothing plane with cutting angle 45-48
Tn general, we can consider che tools with one level
surface, 45-50 working edge, fi t immediately in the
hand (without handle) as planes. The cutting angle of
the metal inlay at modern planes used by carpenters is

about 25-30.

~--

~
\

~
I

23mm--..I

~
I

'

:~: ) ',

I
I
I

I
I

I
I

I
I

:
I

~~

I
I
I

.._8mm
I

____;

wedged chisel made

of stone /full-form/

wedged chisel
made of steel

Fig. 37 Working edge nngles of end-scraper and modem chisel

Spoil shave blades can be used for the final shaping

of level and space curve surfaces. Their cutting angle
is perpendicular to the back panel. In our case, the
steeply/perpendicularly truncated blades or the
"blown-off" blades can be considered as spoil-shave
blades. This latter procedure sin1plified the rather
longish process of retouch by a well directed blow the result is an edge perpendicular to the ventral
plane. In some cases it should be considered that the
objecr.s simply called " blade fragments" could be real
cools.
Style-knifes are large hand tools used for several
functions. To fulfi l lhcst: tasks, variable working edges
are used: arched or lobate, nosed or concave. Their

general feature is a smooth base from where the c utting angles stan from . In the Palaeolithic choice,
mainly core rims, large retouched crested blades correspond to these criteria.
Borers do not belong to the typical tool kit of the
Hungarian Upper Palaeolithic period.
Among the non-retouched flakes, however, series
of objects can be found the unretoucbed lobatc edge of
which could be suitable for this function. The boring
tip is symmetrical, edges meeting a pointed Lip with a
concave edge. After this, a small "side-peak" prevented the tool from sticking in; in I.he correct position, splinters can be also detached from the hole by
the tool.

61

We are still at the beginning of a comparative study

of recent and Palaeolithic stone tool ftmctions and
technology. On the basis of our modest results we cannoL make any finite statemenrs in problems like the
atypical borers or, generally, the "no11-standard:ised"
tools. For example, the notched or pointed flakes
which were not retouched could function as '',ad hoe'
tools, using the natural cleavage qualities of the raw
material sparingly. Another possibility that these
flakes were intentionally left in a rough-out, halfr.eady stage.
The absence of saw blades in the Upper Palaeolithic industries of the interstadials whh rich arboreal
vegetation can only be explained by nor expecting the
typical Western European saw blades ro occur here.
flakes with zig-zag edge could be used as stone inlays
for sawing. Unlike the teeth of a typical saw blade
placed in one row, these flakes with zig-zag line,
resembling however in their cutting angles to modern
saw-blades were used in a different way. It was not the
saw moved to and fro' in the wood but the wood itself
was moved on the edge ofthe fixed saw. Cutting angle
data of modem saw-blades are 60 for pine-wood and
deciduous soft 1rees and for bard wood, 75. On the
flakes considered suitable for sawing on Lhe basis of
the array of"te.eth", the cutting angles were somewhat
bigger: 70-80.
Wedges are already tools for shaping without culting I slivering, i.e. splitting. They are important tools
for working wood, especially without tbe saw. Their
effectiveness increases by the incre-ase of the wedge
angle; at the same time, the difficulty of lhe work. the
force necessary for driving in the wedge is also
growing.
Typically, wedges were used as inlays fixed in a
socket. The ideal wedge angle for metaJ is 20-30, in
case of stone wedges it is necessarily higher. Hafting
is perforce necessary for the safety of the worker. AccordiDg to the injuries found on the butt end of the
s.tone wedges, however, this surface was not covered
by hafting lo ger maximal force of the blow.
Compared to the average lool kit of Gravettian/Pavl.ovian senleruents, the ratio of so-called
"heavy" tools is unusually high in th.e Bodrogker~itUr material. Tl:lese are really bulky, robust tools.
A reason for this could be, in tbe first place, needs of
the people: due to the rich arboreal vegetation in the
interpleniglacial period, the ratio of heavy wood-shaping tools was probably high.
The ecological condjtions of !he Pavlovian entity
were rather favourable (interst:adial periods) all over
Central Europe. ln the Bod:rog-l1isza-Takta region,
rich in waters, the natural endowments were even
more favourable. The forestal environment, rich in arboreal species and the high number of tools suitable
for tinlbet working indicates a varied scale of wooden

62

tools. Needs were also supported by the proximiry of

abundant raw materiaJ sources.
3.5. Raw material

The 1963 excavations by Laszlo Vertes yielded

about one third of the total material collected from the
si te which is deposited in d1e public collections mday.
Vertes e1,1a1uated the raw material distribution in the
followings:
'RAW MATERlAL:
Nineteen per cent of the implements are made of
obsidian.
Thirty-seven per cent ofthe implements are made of
so-called ""stone marrow'"', a white homogeneous
amorph silica-gel. This. material frequencly 1s found in
planparallel tablets 2-3 cm thick. The muhi-anguiar
buriJZS mid triangular amulets are nwde ofsuch materiaL

The raw material ofthe remaining 44% of the im

plements are mainly oflocal limnoquarzite while a few
of them are of s1ndstone (retoucheurs. hammerstones) and only a Jew are made from radiolarile or
jaspe1; i. e. from raw materials which r.hey probably
fo und in the form of pebbles at the not too disranr
northern river volley. 1\110 flakes or rather blades
made of ll speckled homstorie characteristic of the
Polish Swiec:ieclzow originate from farther ([Way. Tins
raw material is found at other Hungarian sixes. too.
and we can COUJ!Jt 011 the fact that. ir travelled greater
distances during the Upper Palaeolithic already as
commercial goods."
(VERTES 1966. 11.)

The finds collected from two excavation and surface surveys are unifonn in respect of raw material
utilisation, the variation in the ratios do not exceed
nonnal fluctuation due to different quantitjes collected.
In 1982, Erno Matyas, geologist of lhe Mad Oreamd Mineral Mining Co. investigated some raw material samples from the Bodrogkereszcur site (control
samples inventoried at Pb 83/924-935. in lbe HNM
Accordlng co his results, most of the loo! kit of the
Upp~r Palaeollthic site originated from the in1medjate
neigh.b,ourhood, the R3,lka-Mad-Tallya limnic basin of
Sarmatian (Upper Miocene) age. The geomorphological scheme of the formation of Sarmatian limbic
quartzite In the Ra.tka-Mad-Tallya triangle was also
reconstructed by Emo Matyas. (Vig. 38) The si licification process can be conncct.ed to several phases of
postvolcanic activity: sllici'lied rhyolile toff, pyrog~nic silt, geisirite and limnic quartzite (Fig. 39).
(NIA TYAS 196Q.). Siliceous bodies were easily prepared by local surface formi.Qg forces, mainly erosion

T he J 00 m thick lim nic sequence was accumJated in 1- 10 meter deep basins fed by hot spring
>S,0

2,0. s.o

0.0 .1,0

Ootn IYlter surfact

mmhv
medium

m1rthv fol'CSI

~'el!dltion

ra.u

extent of entamncnt

VCIV ROOd

bad (delta tvDe)

extent ofsortml!

bad

medium chani!cablc
thllJll!etbJ,
sandy tufiu:. fine clav

Vl!l'Vl!ood

OlMICIJeS!IDol\

sih~us

auantitv of colloide matenal

chanchtemtic fonnaaon

slov.

dom1~an1

11ua1unc. fine c!av

sandy ll.lfite connlomcrate

delllh ofwatu Cm)

fig. 38 Geomorphological schc.me oflhe fonnation oflimnic quartzite. rccollStruction by Em1' Ma tyas

8
9
10
11
12

13

14

Fig. 39 Different phases of the silici!icauon process, reconslfllction by Erno Mruy!s

1. hidroquamjte, 2. thick layers of limnoquartzite, 3. slnty limnoquartzite, 4, line clays, 5. siliceous s ilt, 6. lnolinite, 7. montmorillonite, 8. character of lhe sediment, 9. chemical. 10. pclitic, 11. tine sandy, 12. sandy, 13. gravel, 14. characteristic cations,
15. Si02, 16. kaolinite, ! 7. montmorillonite, 18. montmorlllonite, allcvarditc, 19. mixed structure minerals+ i.colites

during the Late Pleistocene. They can be co!Jectcd o n

the surface even today. (p.c. by Erno Maty~) . These

hydrothermal raw materials were the closest sources
of n1w materials; they arc not found on the site itself
but the sources are within sight.

The most special raw material used on the site is

"stone marrow". Neither contemporary sites at the
other side of the mountain (Megyaszo), nor younger
sertlemants in the north-western neighbourhood

(Arka) yielded comparable high ratio of Lhis raw mate63

or

rial; whereas tlle source is located in the middle

a
triangle of 20-30 km long sides.
In the late seventies, K. Bir6 surveyed the obsidian
industry of Hungarian Upper Palaeolithic sites. 1n
eoursc of this work she also studied the material excavated at Renye hill (Venes's excavation) as well as
material from field surveys around the site.
On the basis of the finds collected by Vertes, Bir6
recorded the followings:
''The first obsidian works}wp sites appear in the
immediate ~~urroundings of the Tokaj-Epe1jes
sources... at Bodrogkeresm11; exploiting the Hungarian (Carpathian ,fl) sources. Here, 1/wugh rJie complete tools were general'ly made of both ofihe known
obsidian genera. cores were made. almosl exclusively.
from local non-transparent Carpathian ll obsidian".
(BIRO 19&4, 17.)
K. Siman drew conclusions oo the function of se~
leroent on the basis of the distance of the raw material
used on the site from Lheir sourees and tbe character of
the bunted booty (SIMAN 1.988, 65.)

BodJ-ogkcrcszrur
pieces
'~

Szelerian felsitic porphyry

Hydroqua:rtzit.c
Silex
Radiolaritc
Obsidian
Opal and jasper
Stone marrow

o/o

Local

Mescolocal

Long-

distance

cores

91,70

2,59

2,7

blades

78,33

3.98

14.12

Oakes/waste
tools

77,36
62..88

1,12

8,18

4,25

15,78

ln Bodrogkereszt(Jr " ... most probably the population came relatively quickly through the Slovakian territories to s~.tle on the southern slopes of cite motmtains: the mesolocal raw material' had a smaller role,
while there is still a iarge tools.'' (SlMAN 1989,

93.J...
[h Europe the earliest Gravetlien dares are around
28000 years both at \Villendorf and Bodrogkeresztur.
In Bodrogkercsztur, " ...this Grave1tian population

uses much foreign raw material from- the norzh and

1iorth-easr. Cansequenily, this type of indusny must
halte t).iJ/1ed even earlier in more 11orthert1 rerritories.
Al Willendoif, at the same time. rhe indus1ry seems 10
be result ofrin i1111er development".

0,8

28,l
6,3

(SJMAN !990, 18.J

4.6
23.6

Sim an 's hypothesis on lhc origin of t;be indUsll)' do

.not exclude thal the fasL migration ~Gross Slovakia
could be of N-E direction as the ~ourus of almost all
of the mesolocal raw materials are located not only lo
the: North, bl to Lhe West of lhe s ite.
S iman's grouping of the sources corroborates the
.sceming~y obvious fact thaL distant raw materials were
in higher e$teem chan locals: a!J partial results of the
technological chain, made offar-fetched raw materials
were utilised ma.'Cimally. The obsei:vation that cores primary forms - were-111ade, in very high rati0, frnm
loqal material corroborates the same phenomenon llescribcd by Vertes in case of Arka, i.e., pan.of the tools
made of long-distance raw materials had been brought
to the settlement in ready-made form. The ratio of
mesolocal raw materials is strik:ingiy low. This can be
probably explained that we do not know all the raw
material sources of the Bodrogkereszt:Ur hunters; in
lhis case, chis category is merged into " locals" Another poiisible reaso,n can be that mesolocal geological
sources were only collected accidentally, in an ad-hoe
manner. The reason can be both - or a lot of another factors together.
,..\Daly.sing the raw material use ofth.e complete archaeological material in details we find that the_ratios
ar~modified only to some extent

74
54
276
13
415

,_ 35.L_

1170

100

l,L

The relative baJanee of loGal and distant raw materials, together with the fauna fit weU to local environment denote, according to Siman, a generalized settlement character for Bodrogkeresztiir (i.e., not exploitation sire or workshop~ SINIAN 1988, 65.)
Th~ contradiction between the two statements (i:e.,
if the Bodrog.keresztur site was a workshop or a settlement) is only ostensible. lt is clear that Bodrogkeresztl'.lr was not a quarry s ite, and the location of the
settlement, the accompanying fawJa indicates a general settlemenc with no specific function. By grouping facts and phenomena in a novel way, Siman found
further proofs for this. The el~boration of local obsidian as Bir6 supposes (BlRO 1984, 17) ~ is, at the
same time, just natural. The existence and extent of
local obsidian working cannot serve as a proof for obsidian workshops in industrial scale for trade and
transport, i.e., proofs ofa classical workshop site.
Examining pan of lhe raw material spectrum, e.g.,
the distribution or utilisation ratio of Northern Flint

64

Bodrog-

keresztilr

329

Total

varieties, the extent of different origin raw material is

g iven below, after Siman:
" Raw material distribution in percentage:

Raw material disuibution of Bodrogkeresztlir- Henyc archaeological material according to types (the extended lype list contains, apart from the S-B list, the complete artefact type list):
Types

1- 2
3
4

V'
8
9-10
11-12
13
16

17

.,,

.H5

5
65
9

..,..,
-~

44

4!6-47 ~"*"
~ --~.

57 rt.... .... ,,,_

5g..59b ... ~
60

61

J... ...... ~

c....

78
~ 1mnd-axe
choppar-

exotic

hydroquanzite

marrow

21

3
I
j

9
3
3

15

13
3

II
I

16

10

14

15

1
II

2
2

11

13

34
12
26
14
14
27
10
12

I
3

3
8

3
4

4
1

126

1
17

21

24

15

5
2

I
l
2
4
2

I
I

I
7

25

5
l

3
12
I

l
3
9
II
4

I
4
4

6
2

16

3
8

7
6
4

l
I
3

5
3
3
3
I

10

20
3

7
1

27

6
5

14
I

37

2
37

18

4
I

10

tzitc

I
1

I
I
2
I

Quar-

49

Szeletian f.p.

3
11

2
3

23
3

16
5
I
3

2
2
1

I
I

obsidian

s1one

radi<>larite

6
3

14

II

36

58

1
I

iChoooine-tool
'Ire

core base

_.. i.eeometrical oebble

- 11rrunmerstone

I
I

2
5
4

62

65
66 NA
67
69 f"VV\~r74-75 n~
76 l~ Pi 1~-11 .
77 ..\ U .I c.l,,

--

12

43

48-49

II
2

IS
58

/I
37
38-39 \
41

'-

10

7
19
30

..x

J ~6

75

hg..VI'

34

other

-(j)NA,

27
28
29
32

silex

5
3
37

</"'?

Erratic

8
I

pieces
total

Ull!de, blade-

157
5
13

49

39

19

5
7

21

561

50
3
2

168

14

187

91

fra11men1

65

Types

pieces
total

retouched

flake
fabrication

Erratic
silex.

other

radlolru:ite

60

exotic

hydroquartzite

stone
marrow

obsidian

Szeletlan
fp.

17

15

14

459

Quan.zite

I
1928

51

105

13

621

654

812

11 7
107

ll7
162

25

10
2.8

223
873

151

162

29

898

6ll3

32

2.24

279

5.4

38

1096

1049

805

12

32

62

7,8

1,5

l.1

305

29.3

22..4

03

0,9

16

debris

Tvoe (total)
Others (total)
Total - all

2n1
3589

artefacts
%

Apart from local materials, raw material varieties

obtained from considerable distance have special significance in "cross-culcuraJ connections". For indicating the distance of acquisition, the symbols used in lhe
Lithotheca are used:
"L = local; distribution of the rock not exceeding a
day :S walking distance
R = regional: raw material of 1'egional distribution.

cal, biological conditions here constitute abour 60% of

the Bodrpgkereszti1r lithic industry: bydroquartzite,
limnic quartzite, "stone marrow" .
The veins of the Mad limnic basin are located
w ithin a circle of I 0 krn from the site, to the North~

Flexible category changing by age and culture.

Roughly. this term den.ores raw materials used
by che people/cultural w 1it inhabiting the environs of Ihe source. The category can be de/.ineated very rediousiy as yet but has a veiy important
historical message.
LD = long distance goods: travelling over J11mdreds of
miles from the sowce area, a "cross-cultural "
item."

82%. Lumps of obsidian, from size. of a nal to that of

ah apple can be collected even today on the h.i lls

(BIRO-DOBOSl 199 1, 8.)

The term "regional" used in the Lithotheca approached the problem of the individual raw materials
not from the side of acquisition but that of supply, A
certain raw material can be of regional importance if
its distribution is in the order of the territory of a given
archaeological culture or community using this raw
material. lt is evident, that this category corresponds,
by and large, to Siman 's "mesolocaJ" group. Anyway,
this category is certainly between U1e two clear extremes, local and long-distance and separated within
the archaeological material on the basis of"feeling" or
" anticipation" rather than certainly. lo distance, this
category denotes the same as " medium-distance".

LocaJ raw materials

The study of the primary geological sources of ob
s idian and raw materials of hydrothem1al origin is
adequately studied, mainly due to modem industrial
utilisation of some hydrotheanal raw materials.
Along the south- margin of tb.e Tokaj-Eperjes
range, the postv.olcanic activity centres formed the
Mad Basin sediments. closely connected with the Upper Sarrnatian active voleanic processes. Variable hydrothermaJ silices fanned in different physical, chemi-

66

West, North and North-East (p.c. by Ern6 Matyas)

Obsidian can be considered also local: counting
them together, lhe ratio of local materials excet:d

around Mad. Carpathian 2 type obsidian, known

throughool lhe Upper Palaeolithic within the Carpathian Basin must have been distributed only from this
region.
Local raw material occurring in tabular, laminar or
nodular on the surface could be collected without special cxpenise and tools in unlimited quantity; these
sources were Che basis of more than form 80% of d1e
total Uthic industry. This fact iii itself could offer a
stable basis for the settlers and, at the same time, can
explain the lavishing use of1he raw macerial as well.
Further local raw materials which could be collected
practically from anywhere in the neighbourhood include basalt or silicified brecaia/conglomerates, silieified volcanic ash (Pb 83/695, 732, 478 ere.) as well as
!he quartz- and quartzite
Rock crystal should be specifically mentio11ed occurring fr-0m time to time in variable quantities on
.Hlunga.rian sites during the full length of the Gravettian entity. Rock crystal and hyalite can be collected,
in about a day's walking distance from tb.e s ite, in the
central regions of the Tokaj-Eperj,e s Mts. Tlle " rock
crystal" flake collected at Bodrogkeresztur, however,
could not be made of local material: hyalite occurring

in the region of Sima and Bask6 differs in quality

while rock crystal occurring around Tclkibanya differs
in size range and were not used for the production of

stone tools.
The .fluid inclusion study of Hungarian Upper Palaeolithic rock crystal artefacrs by I. Gatter excluded
tbe use of local-regional pure quartz for stone tools.
Most probably, rock crystal from "Alpine" paragenesi.s were used from epi- or mesometamorphic environ-

ment. The original collecting place for the rock crystal

could be among the numerous occurrences of the Eastern and Central Alps. These regions '"ere free of permanent ice sheet during the interstadials (DOBOSTGATTER 1996). ihis raw material is known to occur
on all senlements of the Older Blade Industry of the
Gravcttian!Pavlovian settlement wave (Bodrogke-res211ir, Megyasz6, l::lont-Parassa ITT) amd practically, it
is present in rhe whole Hungarian Upper Palaeolithic
/Gravettian entity. Most frequently, rock crystal was
found at the chain of Epipa1aeo1ithio settlements near
Pilismar6t Their occurrence indicates a systematical,
though not frequent and not necessarily direct contact
with the source areas.
Some pieces of so-called. "Tarpa silex" i.s also occurring among rhe Bodrogkereszrur finds, classified
by Pal Gyarmati as ..Oss2c:silh = baked" silicified rhyolite tuff(GYARMATI 1983, 17.). This light grey,
compact rock can be easily recognised on the basis of
characteristic resorbeated quartz crystal fragments.
Archaeological material belonging to several archaeofogicaJ periods was found at lhe outcrop. Tarpa hill
from the wall of a small mould disturbed by sand
quarry from a loessic sand (residual surface?). Most of
the finds were made of this material there (DOBOSJ
1983. 10-11). The exact geological source is unknown. The raw material might originate from the
Bodrogkereszrur rhyolite region rather than the small
byperstene andesitic ruined volcano at Tarpa; its occurrence at Bodrogkeresz.tllr is not surprising. The
small extent of its use can be explained by the abundant presence of other, better raw matecriaJs. ll seems
that this raw material can be non-local at Tarpa.
Though no typological or stratigraphical argumenrs
are available. the temporal limits of the utilisation of
this raw material might indicate that part of the Tarpa
material was of Upper Palaeolithic (Pavlovian?) age.
The appearance of the Tarpa raw material in the older Upper Palaeolithic find assemblages ofBodrogkeresztur (together with Slone marrow. which has its primary geological outcrop in all probability al the southern margin of the Tokaj-Presov Mts.) the contemporaneity, or at least, the common raw material source
utilisation of the two site groups can be supposed.
Long distance raw materials
Occurrence of source-specific raw materials falling
in the category of regional or long distance raw materials in consjderable ratio can be. to some extenl, proof
of certain control I ownership over the raw material

sources.
Tn case of l3odrogkeresztur, the known geological
sources point at the North and the North-East. Characr
teristic and unique items include. e.g.. felsitic porphyry from the Tatar-trough of the Eastern BUkk Mts.,
rhe characteristic Szelecian raw macerial known under
many names. The dotted c hert from Swieciech6w can

be located quite safely. The Eastern parts ofche Polish

Mid-Mountain range separated as Holy Cross Mts.
served as the primary geological source of this very
characteristic, small yellow dotted silex of good quality whicb has been found sporadically in the Hungarian
material since the MjddJe Palaeolithic. In the material
of Bodrogkeresztur, at leas1 5 tools were made of this
raw material (Pb 64/50, 65/1270, 66165, 83/624, 809).
The minimal distance between the settlement and
the geological sources, considering the route BodrogOndava-Dukla pass -Wisloka-Wisla valley is 300 km.
that is, far surpassing the daily movement, usual "action radius" of the community.
Another "long distance" raw material is the so
called Erratic I Baltic flint. The name itself indicates
the heterogeneity of the raw marerial, therefore its localisation is very .. fuzzy". This type of silex can be
callected in the terminal moraines, their primary
source could be practically anywhere within d1e exteoded "ice-catchment a rea". Their clo~sr generally
accepted secondary collecting spots arc ar the terminal moraines deposited in the Riss period in Silesia,
Special raw material collecting expeditions could have
been directed to these regions (KOZLOWSKI 1970,
8). Raw material classified as "Erratic" at Bodrogkereszrur were. in all means, long distance goods, t!ven
in the case when the terminal moraines could be found
more to the South than Silesia, as supposed by J.
KOZLOWSKI. We have no infom1atio11 on secondary
geologkal sources of Northern flint within 1he Carpathian Basin. Comparative Erratic flint samples of rhe
Lithotheca offer little help. The exact identification of
the quite often heavily patinated and carefully worked
artefacts is a task" for the future.
In Siman's paper, some tools are supposed LO belong to Baltic flint the raw material of which are not
known e ither from inside the Carpathian Basin or
from its envi:roos (Pb 64/228, 474, Pb 83/921, 279).
(SIMAN 1990, I 8.) The tools made of "Baltic flint" in
Sin1an's view were carefully worked, smaJI tools,
probably brought ready-made from a fonner station.
Some tools were made of the so-called "Prut silex",
i.e., a mainly graphite grey, silky lustre fine material
with brownish translucent edges. Bulk of the tools of
the Esztergom-Gyurgyalag Upper Palaeolithic site
were made of this material (VARGA 1991. 269.). The
utilisation of this material reached its heights, after the
Upper Palaeolithic period. in the Copper Age (Bodrogkeresztill culture), both in technology and quantities, appearing iii graves as high prestige grave goods.
The upper reach of rhe river Prut where U1is raw
material can be collected is about 400 km from our
site. The immediate connection of the two regions are
blocked by the Eastern Carpathian arch, reaching an
average height of2000 m altitude. The objects are the
following: Pb 64. l.23, Pb 64/292, Pb 83/60, 565, 572.

67

589, 616. " Prat silex" in archaeological practice is

often mixed with "Volhynian :flint" as well.
This exceptionally good Cretaceous fHnt variety
crops out some 300 km o the North of the upper reach
of lhe river Prut at the border of the famous Ukrain
"polesie" region, as a geographical {Eastern) continuation ofthe Poli$ Lublin rable. The exf')ens of this region separate 6 categories within the raw materials
used in the Upper Palaeolithic and Mesolithic of present-day Ukraine (p.c. by L. L. Zalizniak, Kjev, Arch.
institute). Of these, we can c<:>nsider the North-West
Ukrainian row material region with fine, compact.
dark grey - slightly yellowish cl1ert having a smooth
light colour cortex. Analytical data on the raw materials are missing as yet.
The other type of raw material possibly encountered at BodrogkeresztUr is also dark grey, easily patinated with bluish white patina reminds us most to
" doniecki krcmen". ft s hould be emphatically stressed
that the observed macroscopical similarity of raw materials do not allow us .to suppose any direct contacts.
There are "iotem1ediate" items considering the difficulties of raw material procurement between local
and long distance raw matedals, which could be obtained from the Carpathians or the Northern MidMountain range (Bnkk). We can caU them mediumdistance raw matedaJs, which t erm fits well to this
north-westemly exposed semicircular :zone with a radius of cca. l 00 km.
The primary geological sources of tlifferent Carpathian radiolarite-s and the Vihorlat: menilite (e.g. L
86/ 185, L 831696, L 8517) can also be reached within
this distance; they can be also collected on tbe.molass
are from river sediments in form of pebbles as supposed by Vertes for the jasper and radiolarlte tools
from Arka (VERTES 196~5. )
Some "appropriation" concerning the raw matetial
sources is expressed in the presence of locally specific
materials, as an exchange item. Such is the presence
of Szeletian fel.sitic porphyry from the Tatar-trough of
the Eastern part oftheBtikk Mts. (12 pieces). T.he distance from the source is 60 ~ as the crow flies .
The contacts of Bod.rogkeresztur-Henye indicated
by the raw materials point at longer or shorter distances but always to North, North-East, North-West.
As regards the topography of the site, this is not substantive information as t he s ite is situated at the Northern margin of the Alfuld {Gr.eat Hungarian Plains),
thus it is bordered by vast a reas void of raw materials
but rich in Living waters, wbicb shoul.d be crossed before reaching areas with s uitable raw materials.
Long distance raw material acquisition is not very
1i kely though the macroscopical features of the Henye
hill radial-a.rites resemble to some extent to radiolarites
of the Gerecse or Mecsek Mts. (p.c. by K. T. Bir6.)

68

More c lose Western conoections are best shown by

the presence of Szeletian felsitic porphyry from a distance of 60 km, crossing two m~jor rivers, Hernad and
Saj6. The raw material from the Tatar-trough in itself
is an " enigma". The geological outcrop is limited to a
very small area (today we know about the
Tatar-trough only). The raw mat~rial is of uneven
structure, platy, seems to us not the best: still it was
used in the Northern slTipe of the country, from the
Danube bend till Bodrogkereszt(1r from the Early
Gravettiai, tiJl the Epigravettian industries. (Its occurrence in the o lder industries of tbc Bolek. M ts. is reasonable)
3.6. Settlement features
The value of our conclusions o n the material of
Bodrogkeresztiir-He11ye is unfortunately lessened by
the unfortunate fact I.hat mosl of the finds were found
on the surface, .in near-surface position and I or disturbed layers.
Among tJ1e reasons we find antropoge neous interference (con!iiant cultivatio1l, deep-ploughing several
limes) as well as natural causes.
The .intensive surface forming effects of che climatical conditions before and after the lntcr_pleniglacial period can be demonstrated also in the
sections cut in the flanks of Henye hill. On the less
disturbed I eroded sides, below and over the signallike cultural layer (some bone morsels, flakes, or cbarct>al grains) a more sign.lficant bunch of loess was
scdimented and remained to these days than on the
hilltop. Collecting the sequence of the most distanl
sections, opened at a distance of 230 m in NW - SE
direction, 90 m in N-S direction we can conclude that
in the two phases of the Pleniglacial the ruling direction of the wind was different. .Before the existence of
the settlement .here (under the cultural layer), loess
was accumulated Jn larger quantities on the south-eastern side, while the covering loess is more thick on the
northern side.
&ealignment of the surface can also be caused or
in.:fluenced by the physical and chemical foatures of
the base rock. The basement here is of volcanic origin
"'rith great hydration capacity which is reacting faster
to climatic changes. Denudation and planation effected most the east~rn . southern and western footh ill
regions; however, most of the area was moved before
the young Wilrm loess covered the hillsides.
''... the orography formed up. co the end of the Tertia1y was subjected, during .the cold period of rhe
Pleistoaene to a suh~tantial upheaval. Periglacial
processes have not only transformed these surfaces,
bur they produced also a new kind oforography. "

(PlNCzES 1977, 30.)

During the two excavation seasons - l963 . and

1982., respectively - several secLions and trenches

- in the imact or only shallow ploughed sections

there was a yellow, homogeneous loess layer free from
concretions, its thickness varies between J0-20 cm.
- the cultural layer was on the border ofthis yellow

were opened on the top of Henye hill (Fig. 5.).

cypical loess and the underlying limy, whitish loess-

T he horizontal and vertical position of the finds

In opening the excavation sections, the primary fac-

like sediment. The boundary was not sharp and detitor was the concentration of finds. Not all of the sec- 11ire but gradual,
tions came up 10 expectations. This is especially ITue
- the cultural layer itself was a brownish embryonaJ
for Lhe sections cut in 1982, because the official and soil with humus and limy mycelia,
amateur collecLing activity spoiled much of the sur- 10-15 cm below the cultural layer a level satuface features. The seemingly disordered layout of sec- rated with loess concretions could be observed; under
tions and trenches therefore followed the surface mor- this. che andesite base rock followed.
.
phological feaiures and covered, approximately, the
The layer sequence. vertical array of finds were diforiginal (Palaeolithic) concentration of finds. For an ferent in the different exploring trenches and section~ .
easier orientation among the finds the introduction of According. to excavation experiences. not only ~he ~n
certain "Units" {see: settlement units, fig. 5.) became evenness of modern disturbances and culuvat1on
necessary.
.
could be reconstructed but the relaLive and absolute
The observations made during 1.he two excavations depth of the culmral layer allow some conclusions on
partly corroborated, partly completed each other.
the original surface, form and reliefofthe bill as well.
Dividing the hilltop according to quarters, the geodetic spot height was used. Sections and exploring
Observations concerning Lhe hill-top, 1982.:
1:renches were fixed relative to the spot height. Data on
- The Palaeolithic campsiLe was settled on th.e hillthe intra-site topography were collected from Lhe ex- top, at the supposed central parts, immediately over
cavation registries.
the andesite. Just in the periods preceding the formaObservations by L:iszl6 Vertes in 1963. were the tion of the settlement, considt:rable erosion must have
following:
been in operation: on the more protected slopes, origi- finds were concentrated in an irregular patch nal loess sedimenL can be found below the cultural
around the geodetic point in a circle of cca. 20 m dia- layer i.ncluding probably the sedjments washed off
meter,
from the slope.
- cultural layer was observed in the depth of 60-70
The No1th-East of the geodetic point, there were
cm,
two excavation uniLs, section -p from 1963 and
- just below the cuhural layer. at 70-80 cm the un- trenches 3. and 8. connecting unit I. from 1982.
derlying base rock, andesite was found.
.
.
ln section "F' (35-40 m to NE from the geodetic
- in the trenches, Lhere were 30--40 cm mixed. dis- point, the surface is 1.90 m deeper) in Lhe depth of
turbed soil contairiing sporadical fi nds in obviously 100--110 cm, burnt patches of cca. 10 cm Lhickncss
secondary position,
were found. No archaeological finds were found be- on the southern part of trench "A", immediately side Lhis hearth.
under the disturbed layer at 60 cm intact cultural layer
Base rock was fowid in the depth of 340 cm. Lo
was found. Ploughing did not reach that far,
trenches 3-8., the same phenomenon was encountered
- the material of the cultural layer (i.e.. brownish in the depth of 130-140 cm, together with scattered
Loess) did not seem a filling for pits or anificially co- settlement margin fearures, some flakes of stone and
loured area, much like if it were a fossil soil, also by bone.
its argilaceous touch, especially in sections H-F-D
Archaeological features were noL found jn comparable depth a l other places on the excavated hill-top. ll
( 1963).
.
- on the intact parts of trench "A" preserved m was. in fact, not possible to reach that deep (base rock
stripes, some silex and bones were found though not was appearing much higher, not deeper than I m).
Finds and the terracotta-like thick burnt hearth patches
manv
- there was a 60 cm deep ploughing on the are, cul- were " in siru", found in their original place and not
tural remains were intact only below this level,
washed in/down by erosion.
- probably, the central parts of the settlement were
destroyed by disturbances.
The possibllily of a second, deeper I old~r cultural
was considered and subsequenLly rejected belayer
loformation on che position of the cultural layer
cause:
from 1982.:
- in these NE sections, no traces of cultural layer
- there were worked Lools, flakes, fragments preabove
the hearth level was observed {no direct supersent in all profiles, in Lhe disturbed loess as well as the
position)
ploughed soil,

69

- no traces of the sediment typical of the main cultural layer was fow1d, characccristic of all sections
(with cmbryonal soil)
- quantity and quality of the archaeological finds
found here did not allow traditional archaeological
comparison for detecting possible differences within
the industry
A most plausible explanation of this phenomenon
can be that a deep cleft, (the side of which prevented
erosion) could be used here for some fire-related activity (consenJtion? drying?) The thickness of the incact loess covering the cultural layer here represent the
total time s pan of the formation of sediments.
To the North of the geodetic point we find unit I of
1982., the main excavation area.
The stratigraphy is thin and poor, the original surface strongly eroded and disturbed most by modern
cultivation also on these parts. On the western
parts of the sections, cultural layer appeared at the
depth of 30-35 cm and below thi", the base rock appeared at 70 cm. On the eastern part of the secuons,
base rock appeared in the depth of 40-80 cm below
the uneven surface
To the NorthWcst of the geodetic point. a little
furcher on, the modem surface is flattened a lirtJe,
forming a small plateau immectintely before the steep
slope of the hill-side. We bad good hopes to frnd undisturbed layers at this point.
In the four exploring trenches of unit ll. in
1982. A weak cultural layer wn found her~ m the
depth of 60-70 cm. lmmediately be ow the cultural
layer. 80 cm ht!low the !'Urfa c:. thi. .lSe rock \\ ..., ap-

pearing.

Characteristic features for the margin of the settlement appeared in the ditch facing the core of the sertlement. The centre of this settlement parch was not
found but obviously it is not the central (in position,
not in importance) settlement part reaching that tar.
The area between our sections here and the geodetic
poinL, at a distnnce of cca. 100 en was almost completely void of finds. Probably, this area was one of
the habitation surfaces among several within the settlement
To the West of the geodetic point, uojt ID of 1982.
connected the main sections by Vertes in 1963. ln the
excavation diary,. Us216 Vertes marked the possible
extension of the settlement surface towards these
parts. These sections were planted here to clear the
possjbility of a western continuation of settlement
traces in Vertes's sections.
The depth of the cultural layer here was uneven.
Between the two terminal trenches, in a N-S distance
of20 m the level of the cultural layer was elevated by
40 cm compared to the present surface: the depth of
lhe cultural layer was 9()-.100 cm in trench 2. while in
trench 4., only 40.

70

Vertes observed in 1963 that the original surface

must have been more steep 1owards the west than
today. Observations made in 1982. Prove Lbat it was
also stepper towards the south.
Completing these observations with experiences on
tile NE side we can suppose a constant erosion,
reworking of the sediments for the whole hill-top.
Levelling the uneven original surface was sta.rtcd
much before human interference and resulted in considerable differences in the thickness of lhe sediment.
There was, however, no possibility during the excavations to trace the position of the cultural layer by archaeological methods all over the hilltop.
The main sections of the 1963, Excavation were located to 'the South of poinl C. The profile of the initial
exploring trench by Venes tempted us with undistwbt:d settlement surface here. Sections E-J-G were consequently planted al this place. Some ~xploring
trenches of the 1982. Excavation were also connected
to the southern end of the first exploring trench.
On the basis of the first exploring trench Lasz.16
Vertes remarked in the excavation diary:
... the level of the cultural layer is getting deeper
more fast. discordantly compared to the present surfa ce. in squares :!-3 of se<:t.ion "E . some /l(,lkes were
evanfound at the depih of J m" (LASZL6 VERTES:
Excavation diary, llNM Archives)
At the southern end of trench '"A", andesite appeared at the depll1 of 2.5.
Two small sectiori.:. "Acre opened by Vertc~s to the
South-East of the geodetic point. Their relative positinn was only tentathely foxed: "C'. section was
marked on Vertes's comprehensive site plan but its
place was no1 found in 1982. Section "D" was opened
al a distance of 113 Ill from the geodetic point and 6 m
lower in absolute height, the position ofwhicb was noc
fixed and could not be found and mapped in 1982. According to the excavation diary, 70 cm deep in this
section a piece of charcoal and one blade was found
here in clear, slightly stained loess (embryonal soil?).
lstvan Vor& identified the fauna labelled stray finds"
hypothetically with animal bones coming from here.
He is supposing a fourth settlement patch on the basis
of the poor but authentic find material on the southeastern side of the hill. Archaeological argum6Jlts support only the extension of the senlement till this point.
The stratigraphy of the Henye hill on the side facing the Kopasz Mt was sim ilar lo the other profiles.
Considering together the observacions made on the
two excavations we can conclude the followings :
- The present surface is considerably transformed.
The position of the undisturbed cul rural layer indicate
that the original hill-side used to be much steeper in all

directions (stratigraphical relations can be only similar

those encountered on the hill-top at che south-east-

10

ern side).

- The average depth of the cultura l layer can be accepted as 70 cm; significant differences were o nly
found in che hearth level of the north-eastern clefL The
loess covering the seulemeot surface by lhe end of
sedimentation and the accumulation of the !oess was
probably much thicker (see rhe lhickness of the cover
sequences in the north-western sections). The lack of
these sediments today indicate a strong denudarion.
The: transformation of the surface. started in the Late
Pleistoceoe was finished by intensive modem cultivation and deep ploughing, destroying tbe glacial sedjments aJmost till the base rock.
- By lhe time of the Upper Palaeolithic habitation,
the loess formation of the preceding periods had already Jevelled the andesite surface. The different
depth the base rock from the current surface can be explained by this.
Hearths
On frle surface of the excavaced undisturbed settlement parts there were many, more or less scorched,
bwnt bone fragments found. Constructed hearths deepened into the soil or built lo any small degree were not found d uring the lwo excavation seasons.
The "hearths" were found on the margin of the hypochesed settlement surfaces (find concentrations). associated with no finds at all or a very scanty number of
fmds. They were found in section " F" of Vertes's
1963 excavalion and in trenches 3. and 8 in Unit l. of
1982. (Fig. 40). These '' hearths" were mainly irregular, more smutty Lban burnt patches with a minimaJ
quantity of charcoal in bad state of preservation.
The hearth place in trench 3.-8. was heavily burnt,
like cerracona in a depth oflO cm. lt was accompanied
by only very poor materiaJ, jusl enough to signal the
p resence of the cultural layer.
Verces .mentioned palmful dark patches on the intact settlement surfuce found in the loess. These ashy
patches were rare but marked the presence of the cultural layer even on the parts void of fi nds. Vivid red
spots occurred in all sections, however these were independent of human activities being the weathering
product of one type of the local rock.
Archaeo logical observations regarding the

fa una
Apart from the palaeontological evaluation of lhe
fauna, which was relatively rich and in adequate state
of preservation. we could observe that limb bones and
manmbles were fairly frequent within the material.
The occurrence frequency of these two regions of the
body depend on the way of utilisation of the booty,
The distribution of the bones is uneven on the settlement: probably, they were connected with the settlement patches I Units which could not be delineated
precisely because of I.he modem disturbances.
Vestiges of mammoth were found on the southern
pan of lhe settlement.

Though the section or surface colour of the settlemen1 surface did not suggest any artificial formation
of lhc surface, i.e.. pit, around the mammoth jawbones, the base level o f the mammoth mandible was a

definite walking level

The features around the mammoth mandible were
interpreted by Vcrtes in the followings:
"Mammoth Mandible. The mandible ofa mammoth
was fuund in squares 1, 2, 4 and S ofsutface G. It lay
in the yellow loessy material, undi~turbed, C1pprox1marelv 80-90 cm below the surface (Fig 41). U11der it
was an entirely sterile stratum 40 cm thick The mandible lay with the teeth downwards. forming an angle
of340 to the north-east. The mandible was damaged
near the roots ofteeth. The roots of the teeth were exposed, otherwise it is well-presened.
In the circle (with a ratlius of cca I meter which
can be drawn around the mandible we fo und seven implements 22 ttnretouched blades. 3 Jwmmer~stone
fragments made of sandstone pebbles and 30 flmt
flakes on the same level. Among these a few rested on
the bone. A part ofthe flint was (vi11g on a slam as if
rhe mandible liad been in a shallow pir.
The excavation squares around the mandible of the
mammoth were poor in bone finds. Generally speaking, few bones werefoimd in the westem secf1011 ofthe
trenches. The single mammoth mandible in the shallow pie surrounded by hanm2er-sto11es, implements
and flint waste was most likely to have been a sorr of
stool or anwl in a small workshop.
(VERTES 1966, 13.)

The state of preserva tion of the finds

On the basis of archaeological experiences, the embedding sediment of the cultural layer of Bodrogkereszttir-Henye resembled the classical loess of open
air sites.
The Henye-hill loess is more suitable for the fossilisation of the fauna than the weathered volcanic soil in
the interior valleys of the mountain. ln the latter, organic remains were co11sumed with the exception of
some tooth enamel plates (experiences by L. Venes at
Arka and K. Sim an at Hidasncmeti).
On the Bodrogkeresztur site, bones rcmaintld relatively completely. The difficulty is caused by the thick
limy coating on the lower surface of tbe bones which
could be detached only with part of the bone. This feature did not help much in recognising and preserving
worked bones.
The stone tools found "in s itu" within che cultural
layer and, to some extent, also the surface finds were
often covered with thick limy crust, sometimes 5 mm
thick. For the precipitation of such a bulky lime coaling, a lime-rich cover layer o f at least 70 cm thickness
as well as long eluviation period was needed.

71

-...J
N

UNIT

1: 20

trenches 3 and 8

- 110 - 120cm hearths

~,~

~
3trench

Sm

7m

6m

5m

4m

~ burned surface

_ ___.N

;~~:~:~ I charred

fig. 4() HeartS and bumt sur.fuces in the trenches 3. and It

ll!lEquus jaw
'

bone

STRATIGRAPHY of the Mammuthus- jaw l

DIIIIJ humus

rzzA

~ andesite tuff

ploughed

'"'""'"""' cultural layer

Fig. 41 SU-Jligraphica.I pos11ion of mammoth mandihle (after Vertcs L.)

-.I

loessy sediment

lI Mammuthus

-J

"""

...'

c:r
.,

'. ..~\

~c::::>

/I

..

.,

Pf

Ql(l

~:

~<\:;> ."""

a i?.o"'>

(\ll.,

<>\'I

()

.-

C>

0'

o~

E section
ff

'

...

Cb

()

burned

I Stone

\tn

ll
Ill

bone
tooth

N----Fig. 42 "E" !lection, Vertcs 1963

C:>

<=

'\)

/}

I
Q

o._
"9~ ~

~
II

~
~
~~ ,,)\

D~4

:-.

f.,

'~

"'i ''

,.

o.,,

<\'!

()

C:J,t:?~ ~ ~
~'o

c::>

.'
~

. ~
<'fj

\)
~

~
a ,~

~~

'
,

\0

(\

'

C>

O<v

1m
Verte~ 't

sectons

1963

~
~

~
""'

Q)
........

f-

.....
'
QJ

>.

=
~

Ill

<W

0,
IU

'-'

'

'lt

'

...
:l
"'
!:::

,...

::::::>

'<T

cil

u:

'
'\

..::::::70

c;:?

<W

'iii
0
..c

i..

'

=
...

00

:::>
QC)

......

t:).

(J

.;;t-

..c
0

<)tl.

cc..

't:I
Q

c:J

co.

'?:to
0

"".

>
75

I
0

~. \

()

\
~

..

"

.,. ..

IP

C>
1

(/)

c
0
;::;

~
()

~&

()

<I>

(/)

Cl
I

"~Q

I'

al

~ ~!;;

...;

t=
z

::>
0

....
~

()

a>

......
c
~

=i

i.;.

'.E
t"1

i..

a>

=
C>

76

'\)

/)

OQ

o~#

,:

l7

<;;,Q

z....._-

77

On the surface of obsidian tools, hydration crust

could be observed. As the thickness of this crnst depend on, apart from the time elapsed since its shaping,
the conditions of embedding to a large excent therefore
the thickness of the hydration layer in itself cannot
give i11fon11ation on the time of preparation of the tool.
(BIRO-POZSGAI 1982. l32)
On the raw ma1erials which were ready to patinate,
a thick layer of patina appeared on the whole surface
of lhe tools. The surface of Northern I Erraric flint varieties, which were typically fine grained, homogeneOllS~ good quality raw materials were typically covered
with bluish-white patina. The original colour of the
raw material .could be observed on fresh fractures
only.
Settlement units (Fig 5.)
Scattered, marginal settlement features with "in
siru" finds covered the whole surface of the hill in a
circle of250-300 m.
The density of finds can be considered average
within Hungarian Palaeolithic settlements. After the
quantity of swface finds, more tools were expected.
The relative scarcity compared to expectations can
also be explained that the " in silu" surface was found
in the sections deepened at the starting point of the
slope at the N ortbern, N 01tb-Westem part of the hill,
and probably these pai.ts were already marginal. The
scarcity of tools on the " in sicu" surfaces corroborate
lhe hypothesis that the central pruts of lhe settlement
was desrroyed over a large area which resulted in the
appearance of a great number of tools ,on the surface.
The centre of the campsite was probably settled immediately over the bedrock around the point "O" of the
1982. excavations. Due to erosion, the cultural layer
and its cover used to be very thin here which was disturbed by ploughing, leaving a lot of flakes and tools
on the surface. At the point where the hilltop turned
into slope the Pleistocene layers were getting more
thick, the original habicatiou surface was preserved prutly, immediately over the bedrock, partly - at lower
pruts - over more or less chick layer of loess. These
pru.ts were, however, only lhe peripheral parts of the
former settlement.
On the basis of the 1963. excavations Vertes supposed U1at the cenrral part of the settlement used to be
to !he East of trench "A", on a small plateau of the
Henye hiil (on the basis of the large number of finds
found in the disturbed layers of section ..H"). Judging
from tbe experiences of both excavations, this could
be only one oftbe settlement units. Probably, it used ro
comprise several comemporary settlement units.
On the basis of the surface concentration of finds
and tbe intact settlement surface details, four settlement units were reconstructed witb great probability.
The limits oflhe settlement patches are not sharp, the
fmds decrease gradually.

78

The four supposed settlement patches were the fol lowing:

I. The southern part ofVertes's (1963) trench "A''.
together wilh sections, E-J-G oflhe 1963. excavations,
in section" J", the small hearth and in section E", the
mammoth mandible and the surrounding workshop
assemblage (Fig. 42). The trenches opened after
Vertes's inruitions to the west of this area, the surface
of the sections of unit lll in 1982 (Fig. 43) belong ro
this settlement unit. It is remarkable that to the soulh
of the geodetic poim surfaee collected tools abounded
in stone marrow items.
2. Some 4~50 m from these parts and exactly to
the North of Lhe geodetic point, 8 exploring trenches
were opened in one stripe during the 1982 excavations. These sections were opened to unfold the cultural layer appearing in exploring trenches 2. and 5., of
these, s~ttlement u ni1 l. of 1982. was separated in secti<>ns "B-C-D- E '" (Fig. 44) .
3. Settlement Unit II. of the 1982. excavations was
found 80-100 m to the North-West of point 0. lr was
less intensive than the former one but concained a continuous surf.ace which was possible ro open completely .in trench 4. and further finds in adjacent
trenches marking the level of the settlement (Fig. 45).
The depth of settlement units I culmral layers I, 2,
3 was about the same from the present surface, i.e., the
plateau-like top of the Henye hill used to be, probably,
a plateau in U1e time of the habilation as well and the
factors forming the surface bad even effect on che
sediments.
4. Far from the central sections (around the geodetic spot height), in a distance of 1O~Ll3 rn on the
South-Eastern slope of the Henye hi1~ pieces of charcoal, blades and several animal bones were found in
the usual depth (70 cm). Their archaeological evaluation is not possible because of low quantity of finds.
I. VO'ros placed Unit 1V here because of the significant
faruna and the obviously authentic settlement features.
On the basis of Lhe distance of the settlement units
from each other, partly measured partly calculated and
their arrangement, the extension of the Bodrogkereszllir-Henye Upper Palaeolithic settlement reached or
slightly exceeded 200 rn diameter and incorporated
the plateau-like hi II top and a part of the South Eastern
slope.
Three heanh opened at unusually deep position at
rhe supposed north-eastern margin of the settlement
have to be mentioned speciaJ ly. They were found in
section 'F'' in lhe 1963 excavation by Verles 1963 and
in two marginal conjoining trenches (3. and 8.) at Unit
l. of 1982. (see ground plan). They contained no arcllaeological faurristical finds. Probably, these hearths
did not be long to the main culniral layel' of the: the
great difference in depth between the cultural layi:r of
Unit I and the level of the hearths cannot be explained

gi\lea lbe few meters' dhtance between them. The difTerence is bigger than between che culrural layer and
tht! CLtrrent surface. These buml patches can be even
the traces of natural lire as well.
l11e two hearths wer:e separated by a stripe void of
finds (Unit L trenches 5. and 7.) from the central areas
of the settlement. The level of the base rock (andesite)
he.re:: is about one meter deeper than in the ne ighbouring trenches. We can justly suppose that permanent
camp fire was placed here, sheltered by a natural clefl
or ridge (the soil was buml in a lhickness of I 0-15
cm).
These find concentrations, sertlemem units exceeded the extent of usual tent bases generally found
on temporary hunting-camps. The disturbances
spreading the finds on the surface effected only t11e
objects on the surface. ln case of the in situ" objects,
the horizontal array and the identification of at least
four contemporary settlemem units corroborate the
natural strategical importance resulting from the geographical position of Bodrogkeresztlir-Henye.
Though the specific settlement features that would
prove a permanent I w inter camp were missing and
there were no traces of anificial conscructions found.
even in the " in situ" Bodrogkeresz1ur-Henye was obviously more than a Lransitional camp-site used for
one hunting season.
Horizontal distribution of tools. b.alf products
a nd flakes
Bodrog_keresztiJr-Henye hill was more or less coverod with finds around the geodetic spot height. The
places where some concentration of finds were observed were o pened by authenticating excavations.
Thus approximarely three settlement patches were deLi neated.
The exact position of rhe surface finds col lec1ed
during several years cannot be identified any more.
On the s urface, agricultural cultivation of the soil reworked the fiads coming from the three senlement
units and spread them al 1over the plateau, covering up
the (possible) barren stripes. The significance of observations concerning the settlement structure was not
evident at the beginning of the excavations. The other
reason why lhese considerations are only of informative value is also rooted in heavy disturbances: the size

of the authentic, exca\ ated cultural layer surface at the

different sculemeni patches vary to a large extent.
The percentage distribution of different cypc groups
collected from the surface and the different excavation

units is given below:

Type groups

Surface

v cncs.1.26.l

Dobosi
wli1..1

Oobosi-

.uniUll
Tools

archaic cle-

17

15.8

I Dobosi
wliL.ll.

18,2

23,2

lmenlS
Blades

2.8

2.4

3A

1.7

14,4

11.9

18,7

9.1

Flakes

62

67

52,7

64,3

Cores

3,1!

2,9

7,0

1.7

Vertes 1963 and Dobosi 1982 Unit W: mainly

workshop character, low number of finished tooJs and
many flakes: il could be a refreshing, resharpening
place because the number of cores is also low
Unil 1: among all setllement patches. most of the archaic tools were found here. The relatively high ratio
of blades and cores indicate primary tool-producing
workshop activity
Unit 11: the highest ratio of finished tools was found
here, especially that of burins which is double the
value o f the two other units. The number of half -prodUcJS and cores is low.
T11e results are reliable in tbe basic tendencies:
there were only minor differences within the cype
spectra of surface collected finds and the three find
concentrations. T he close values within the settlement
patches of Bodrogkereszt:Ur-Henye indicate uniform
functions.
The evaluation of the nearest. analogy in all sense
(topography, chronology, serrlement features) Megyasz6-Szelesteto gave significantly different data: ratio
of tools is less than half bere (7%), ratio of blades
lower (5%) but the amount of flakes is much higher
(over 80%)
At Bodrogkeresztilr, the statistics of surface finds
did not show eminent values, surface collection can be
considered as a representative sample.

79

4. Analogies (FigA6)
and s ide-scrapers; these features were typ.ical of the

4.1. Contcmponry sites in Hungary

The (approximately) contemporary sites to Bodrogkeresztur were the following: Megyas26, Hidasnemcti. Saj6szcntperer-Margitkapu, Pl!spokhatvan,
Ront/Parassa 111. Nadap. Among these, only rhe mawrial of Megyasz6 and Hidasnemeri can be compared
sra1istically, due ro several reasons:
- lm:v number of finds: e.g. Nadap (le.ss than I 00
typical mols)
- asymmetrical rool kit due to special settlement
funclln n, !!.g. Pihpiikhacvan
- n1ateriaJ not publish~ yet, e.g.: s~j6szenipeler
Margitkapu or Hom
MegyaszO-Szelesteco (DOBOSl-SIMAN 1996)
The Sz.erencs hills rue co11nectecl to the south-western macgin of the Tokaj-Eperjes (Zemplen) Mts. as a

lbre-hill l-angc with ~ia.'<.imal q,ltilude 340 a.s.l. 011 the

southern part of the range, in the second l'ow of hills
facing U\e Al fold (Great Uungaria11 Plain) wc can find
Szelestet6. a plateau of230 m a.s.l. !n cour~e. several ex~avatkms <lc1ails of a lWo-layered settlem~mt

or

were opened on 180 square m surface. The locality

lies in the meeting poinc of several ecological niches
witJ1 m0s1 favourable natural endowments, in the immediate vicinity E>f tJ1e nonh-eastcm raw material
sources Th1.: sclllcmcnt sul'facc is of scattered ~harac
ter. l1 was j ust tl1e totally undislUrbcd lower cultural
layer of Megyasz6 which aonvinct!d us tJ1ar this is a
rypical feature of rhe interpleniglacial $1.."'ttlements or
ar least part of them. We cannot expccc intensive find
concentrations like- those found on the E pigravettian
or Sagvarian sites. Jt is therefore probable that the
"scattered. character'' of th.e Bodrogkeresztfu: industry
was not caused only b;y soil cultiv.atiou aud m@dern
disturbances. At Megyasz.6, the size and quality 0f the
assemblage, the complexity of the type List and rhe variety of raw material basis indioared a more lasring occupation. The site was a bona tide settlement, (i.e., nol
a hunting camp and I or workshop). The two cultural
layers were fo und separate.d from each olher by 60 cm
sterile loess. The cultural layers is located in sediments slightly turned to soil. The find material collected rrom the cu lturaJ layers and lhc s urface, respectively. show a mature, uni fom1 image. The lwo cultural layers used to belong to 1he sam e cu ltural phylum, bl)th being representatives of the Older Blade
/Pavlovian facies. Fihely worked end-scrapers, versatlle set of burins, the absence of geometrical microlilhs
and "'Aurignacoid" features i.n the shaping of the tools
were typical (many high-crested blades and O.akes. in
the baSt! fon11. consequently the finish is often steep.
step-like wi(h relatively high number of blade points

so

Hungarian Pavlovian industry.)

Fauna: Coelodonta. Eqi:ws, Bison.
C-14 date: deb. 5372: BP 27 070 680
Pfispokb atv~ n-Di6 , Oregszolo

(CSQNGMD1-BALOGH- DOBOSI 1995)

W0rksh0p at the w~tern margin of the souU1crn

parts of Cserhat Mts. faci ng the river Galga site
I) lamed on the hydioquartzite benches occurring along
the margin of the hi ll-side. Authenticated by field surveys and excavations. The rype List is asymmetrical
and deficient. It is evident that the site was a werksltop
planted on the exploitation and shaping of the rich
local hydroquartzite benches. Palaeonlological finds:
(Mamm.ulhus and Cervus)
Radiocarbon ehronof'ogical date:
deb 1901: BP 27 700 300
It is one of tbe closeat contemporary site to Bo<lrogkcteszl(Jr.

FJont-J>arassa UJ/Orgonas (DOBOS l-SIMAN in

press)
Four decades after the work ef M. Gabori, sites
around Hont were located on several field surveys. On
tine area of :cca. t 2-13 s~uarc km, more than 15 collecting spots were found. Among 1he authenticated
finds, we know so far only the site Orgenas which
could be classified with certainty to rhe Pavlovian culture, thus can be considered here as related_ to Bodrogkereszlitr.
lbe topographical position of the site can be cons .idered most traditional. The flood-plain. of the river
lpoly is separated from the loess terraces formed on
the north-eastern slopes of the Borzsony Mrs. by the
present higbwa.y_ The stratig raphy was observed till
L40 cm depth: humus - leached B soil level - typical
loess - carbona:tic loess - brown I inrersradial soil weathered py roxene andeslte..
Traces of Palaeolithic settlement level was observed in three horizons:
at 60.....70 cm, under the loess
at 80- 90 cm, within the carbonatic l.oess
at 13C>,-L40 cm, within the brown soil
The preliminary study of the artefacts ind icated that
the three settlement levels used to belong co U1e same
Upper Pa'laeolithic culture appearing on the site in
three di tTerent periods separated by a considerable
time from each other.
The age of lhe lower cultural layer was dated by radiocarbon measurements:
deb-5027 BP 27 350 610

~JJ~
-~""
.,

. ' /i
ll

.. 1
-~-')
- ~-:7~

..:t.

_'-.__

Fig. 46 Main upper Palaeolithic Sites in 1he Carpathian Basin

Bodrogkeresztur-Henyc I. Wlllendorf, 2. Aggsbuch, 3. Krems, 4. Langcnlois, 5, Grubgraben. 6 Dolni Vestonice. 7. Milovice. 8. StillfTied, 9. MomvMy nad Vahom (ukovska,
Podkovica), I 0. K.a5ov, 11. Cejkov. 12. Beregovo. 13. Korolevo. 14. Oas, 15. Mi1oc, 16. Molodova, 17. Kadar, 18. Nadap, 19. Mogyorosbnny:i, 20. Esztergom, 21. Puspt\khalvan.
22. Sajos.zentpeter, 23 Megyasz6. 24. Arka, 25. Hidasnemeli, 26. Sagv3.r
00

Sajoszentpeter-MargitkapudiitO
(RINGER 1994, 76-78.)
J\ 6 meter high. profile opened in a quarry pit in the
interior parts of the village, the opening of a small active side~valley of the river Saj6 was studied here by
A. Ringer. On the terrace-like shoulder of the valley at
155 m. a.s.1. a series of palaeo-soils was observed,
dissected by thin loess-like sediments. Layer 2
counted from upwards (at a depth of-80-120 cm) was
described as a cemozem-like dark grey fossil soil
which could be correlated with the Stilfried B soil horizon. Within ihis soil, an Aurignaco-Gravettian industry utilising mainJy local wooden opalite raw materials was found. Most characteristic types were high
crested .end-scrapers, blunted blades and Gravettepoints. The site is unpublished.

pieces of artefacts found, part ly during Lhe ex.c.avalion,

partly in course of field surveys there were only 7.7
pieces of tools in the strict sense (according to the type
list. of Sonnevi lle-Bordes); calculating percentage ratios therefore is not meaningful
lt is remarkable that about half of the tools (38
pieces) we.re blade points and blunted blades. The
function of the settlement ii; undoubtedly reflected in
the scanty material: hunting camp of a community
specialised on hunting Eqz.cus. The average dimension
of the tools is 42 mm.
Raw material of artefacts comprise 65% "Erratic
flinr", more exactly it is called erratic without knowing primary or secondary geological source. This ma
terial is bluish-white palmated, very fine texture, silky
grey and 35% li'ver-cmloured Gerecse rad iota.rite.
Fauna: 31 Equus germanicus, 4 Equus sp. II, 2 Asinns, 4 Rangifer and Bi.son

Pilismar6t (DOBOSJ J 996)

Hidasnemeti (S.IMAN 1989,)
Loess terraces of both side of the Danube-bend.
have been investigated for more Ulan sixty years.
l'bere are several, unique stray finds known from different local itieS that can be classified on typological
basjs lo the Older Blade industry ofthe Gravetrian entity. Sueh items include pieces collected from the loess
profile of the Basaharc Bt:ickyards and :field survey
and sondage on the Basaharc plateau The great .Pllismar6t series. (from Palrt till the Oregek di.Ho site) belong to tbe younger blade industry, the Epigravettian
cullural facies.
The small patches of settlement excavated in a distance of 800-1 OOO m of each other along the terraces
framing lhe allqvial plane of.the Danube were typical
representatives of srnaJI transitional hunters' campsites planted atong the route of reindeer migration.
The modest faunislical mateiial of Pilismar6tPalret
was classified, in accordance with this view as "kill
and depot site" by I. VOROS (see Chapter C., Environment and hunting strategies in I.his monograph).
Nadap (DOBOSI- IUNGBERTR- RINGER- VOROS
1988)

The site is located at the south-eastern entrance of

the valley transseccing the Velcnce Mts., at the margin
of the foot-hill s lope. facing the plains (later to be occupied by the shallow Velence-lake).
The site is weU stratified, lying within the loess sequence Dunaujvaros-Tapi6si.iiy, becween the Mende
Upper and Tapi6stlly H2 soil horizons. 'the quarry pit
planted for the cxtta0tion of sandy loess destroyed a
large parl of the settlement. The extent of the settlement must have been much larger than the surface
opened by authentic excavatio n~, judging from !he
quru1tiLy of lhe faunal remains alone. Among the 1100

82

The site was ex.cavatcd by K. Siman on the right

side (Western) terrace of the river Hemad. The industry shows typological similarity to. the Pavlovjan
(more exactly, cultuntl layer W9 of Willendorf).
Shouldered blade points lend a unique character to the
industry among the contemp.orary Hungarian sites.
Living in almost identical natu.rcil milieu to BodrogkereSztt'.lr and A.rka, it is different from both sites in typological composition as well as raw material basis.
Chronologically the sire was dated between Bodrogkeres.z tur and Arka The cultural assignatjon of K.
Simao was corroborated by receat studies. Accepting
the most reL"enl arrangement of the -Pavlovian culruce
by J. K Ko:z;towski, ~he "group with shouldered
blades" is Ute youngest facies of U1e Pavlovian(24.- 20
kyear BP, 1996). On a mere typological basis, Hidasnemeti is the only known representative of this cultural fades in Hungary so far.

lo lhe first publication of th.e site Bocirogkeresz.L'llr

Henye by Vertes, Ark.a licr7.Saret (vERTES
1964/65, 79-132.) was the most important - and al
most only - site comparable to Bodrogkereszrur (Fig.
47).
~ IGl\fl ~

IG

18

1
0

10

20

ao

jBlll

1Bd

GO

-'
110

ro

80

Pig. 47 Technological indices ofBodrogkcre~zrur and

Arka (VERTES 1966, f0)

In spite of geographical proximity and t he partly

identical raw material basis the two sices cannot be
in age

significant differences between the type list of the

two. sires: moreover, the type distribution percentages are surpris ing ly similar.
Among Lhe known Hungarian Upper Palaeolithic

in settlement function

sites, the find material or only a few can be used

connected

in raw material basis

in technological criteria (i.e., size, thickness of
Lool base forms and different techniques te5ulting,
partly, from dJfferent raw material basis)
Disregarding the different impression an the general image of the two iudustries, there are no

for slatislical calculations. These sites are: Me&ryasz6

and Hidasnemeti from the Pavlovian /Old Blade
industry, Sagvar and Mogyor6sbanya from the
Sagvariao I Pebble Gravettian, Esztergom and Arka
from the Younger Blade industry I Epigravettian fa-

cies.

Comparison of type lists :from Hungarian sit.es suitable for statistical evaluation

Types after

Bodroglre-

Sonneville-Bordes

reszttlr

Mcgyasz6

oemerf

Sagvar

Hidas-

Mogyor6sb~ya

Esztergom

-Ark.a

type name

l-16

end-scrapers

I&

26,8

38,S

10

0.4

23,.3

26,8

17-19

end-scraperburins

3,3

2.7

4.2

0,4

0,01

20-26

Borers

1,7

1.7

2,7

1,8

1,7

27-44

Burins

29

21

23.8

2(:),2

2,2

20,1

21

45-50

blade-points

0,6

10,4

6,1

1.5

51 - 57

Shouldered

1,7

1,4

0.1

0,7

2,3

1.4

blades

24

35,8

2,3

1,8

57,5

12,7

35,8

69-72

Poi.o'ts

0,2

0,2

0,05

0.5

0.2-

74-76

encoche, piece
esqui!ICc-

0,4

l,6

5.9

0,2

0.01

77-78

Side-se-rapers

4,2

7,4

4,3

3.0

1,7

1,8

7,$

79- 83

Geometrical
Microliths

2;7

84-91

Microliths

.12,I

10.l

0.4

3.1

92

Otl1e.rs

10.3

28,5

18,4

26,9

Cod~

58-68

points
Retouched

In the category " others", we can find tools missing

from the Sonneville-Bordes list, which are, however,
decisive for the image of the industry (archaic types,
hand-axes, pebble tools)
Esztergom:
DOBOSl-KOVECSES-VARGA
1991, 233-255.
Mogyor6sbanya: DOBOST 1992, 5-17.
Sagvar: CSONGRADl 1997, 17-44
In the comparison, sites of the Sagvarian I Pebble
Gravettian arc definitely detached from the type

composition of Blade (Older and Younger) indust'ries,

Le., Pavlovian and Epigravettian sites. This differentiation js present in all measurable criteria

(average length, ratio of lengthtwidth, i.e. laminarisation)

The picture of the large-enough Hungarian Upper
Palaeolithic sites, according the Otte's large categories, is the following: (1. and 2.: debitage and dechers
contracted, OTTE 1981, 72.)

83

Bodrog- I Mcgyas71>
kt:reszrur pavlovian
pavlovian

Debi Lage.
dec.he.ts
de fabrication

Outils "a
oostcriori''

Outils
faconnes

74%

3
23

'
I
I

'

Mogyor6sbanya
Sligva-

Esztergom
epigrav~t-

rian

tian

84%

87%

6.0%

35

The statistical investigations onJy support the facts

visible to the naked eye. The pereentage rati0 of Otte's
1-11 categories is highest at Mogyor6sbaaya. fo this
sense. the fragmentary character of Sagvarian is reflected which served as a primary faccor in the separation of the industry. The high ratio of deb.itage can be a
signal of abundant raw materials but also it can show
that pebbles were Jess suited for the production of Up,
per Palaeolithic types. Just the opposite is valid for
Esztergom where excellent quality raw material was
used with great efficiency to produce perfe<:t tools.
Blades and flakes belong to the same category of artefacts, i.e. blanks which served mainly as the base form
oftoal types. Turning back to Mogyor6sbanya: the unusually high ratio of pebbles, pebble fragments and
flakes compared to blades mean at the same time the
relative scarcity <;if traditional Upper Palaeolithic tqol
types.
4.2. Contemporary sites outside Hungary
Before comparing Bodrogkeresztt'.ir-Henye to contemporary sites available in published form, let us
comment on the difficulties of comparison by typological and other methods.
The scheme suggested by F. Bordes' (applied for
the Upper Palaeolithic by D. de Sonneville-Bordes
and J. Perrot) typology working adequately for several
decades for the comparison of sometimes very different materials was found already inadequate, as stated
by M. Otte: "Le temps ou les publications d'ensem~
bles lithiques s'arTetaient aux diagram.mes cumulatifs
est aujourd'hui depasse." (OTTE 1989, 10)
The application of the French type cumulative
curves for Hungarian Upper Palaeolithic materials
was always tedious. First of all, because Hungarian
tool types were very different from those tbe cat.ego"
ries used to be set upon; second, priorities in setting
the technical terms denoting the types were not always
clear.
l nfonnation condensed in the typological curves,
however, cannol be substituted by the rather minute
and complacent figures by modem computer-assisted
elaboration. Cumulative curves represented the mate-

84

rial in totality. gave a prompt impression on dominating tendencies within the industry and the ratio of
majn types. The need for reflecting certain details of
technological niceties were not always fulfilled, even
by the indices, but this does not necessarily mean that
we have to disregard the method totally.
In che well illustrated comprehensive type book of
P-Y. Demars and P. Laurent, the authors aim at presenting the c,0mplete European type spectrwn. In rca~
ity, they were only collecting the Upper Palae.olilhic
t}'pes from Centta.1 and Western .E urope (D.EMARSLAURENT 1989).
Apart from rhe cca. 60 types presented in the gloss.ary, besides 27 general Upper Palaeolithic type there
28 tools typical of a narrow range of sites, almost sitespecific, the general significance of which is negligible in a European Upper Palaeolithic typespectrum.
Knowing lhe Hungarian Gravettian find material
the author is convinced that a great typological overv icw valid for Central and Eastern Europe wi II not be
based on the Hungarian material directly, which is neither excessively rich nor very characteristic. Though
the triple division of the Gravettiau can be furnished
wi th typological content, the almost 40 sites of the
three settlement waves yielded hardly enough material
for such a synthesis: less than half of the sites can be
evaluated statistically. Simply, time has not come yet
for the elaboration of a generally valid typological
system illustrating both similarjties and differences, if
such a system can be achieved at all without falling
captive to traps
- la terminologie est confuse. utilisam des qualificatife d 'origine fonctionnelle. technique, morphologique. roponymique, statigraphique:
- ii ~iste rme incoherence. aplusieurs niveaux dans
la classification. des caMgories, l 'imerieur ile cercains t.ypes notammenc au niveau des fossiles direc-

teurs:

- la classification des type:;, iaisse. place a la subjectivite. notamment clans les problemes de dimer'l-

sions:
- elle ne r.end pas compte de touie la complexit, de
la for me de la piece.

In the present comparjson. sometimes hardly compatible syscems used by several authors have to be
used. As a compromise of formal vs. functional typology, the most realistic platform for comparison is ana~
lysing type gr,aups: that is a necessary favour for func1ional typology, onc.e the stricL prescriptions of a formal typology cannot be accomplished.
The aew system saggested by P-Y. Demars (DEMARS 1990) could not be adopte.d for the study of Lhe
Bodrogkereszu:ir material as yet.
Emphasis in cpmparison was given to closely related sites, however, s~1ne si~es which were neither
con,temporary nor spatially connected were also com-

pared to Bodrogkercsznir if data of their publication

made it possible.
Finds of the younger seLllement horizon were necessarily different. The exact definition, expression of

these differences, however, contributed to finding the

exact character of the Bod.rogkereszrur industry.
Sites, siu: groups were studied according to directions of possible contact. These suppos ed directions of
contact were considered at different degrees of probability due Lo geographical endowme11ts.
Orth-West

In the comprehensive study of J. Barta, several sites

were enumerated without typolog.ical details as contemporary sites with Bod.rogkeresztur, mainly from
the Vab valley, on both sides of the river:
Vlckovce, Ncmfova, Nove Meslo. Further sites
enumerated include Svodin, Sladeckovce and some
cave sites of the White Carpalhes.
The s ite complex Moravany along the river Vag
has to be mentioned specially. (Palaeolithic sires of rhe
eastern, left s ide of the asymmetrical Vag-vaJley /the
foothill terraces of Povai.sky 1novec were published
by HROMADA- KOZt.OWSKl 1995, 17) The older
tool assemblages of the typologically-chronologicalJy
different sites show a lot of similarities to Bodrogkeresztilr, though their chronology is younger (BARTA
I 965, Taf. 67). One of the base camps. Moravany nad
Vahom - Podkovica, with Kostenki knife, served as a
key s ite for the delineatiol'l of the origin and distribution of the Gravcnian culture in lhe theory of M. Otte
(01TE 1990, Fig. 6.)
Dolni V~tonice is a key site for the Central European Gravettian culcural complex. The material of this
c lassical site was published according to Lhe Sonneville-Bordes-Perrot system (KLtMA 1963), thus it
can be directly compared to Bodrogkereszrur:
Comparison of type lists from Dolni Vestonice and
Bod.rogkeresztlir according to the SonneviUe-BordesPerrot type list

4,6

2,2

0,05

I.I

10

0,31

I. I

II

o.os

0,7

12

0.2

0.7

13

0,9

14

O.J

16

0,1

2.3

17

2.2

3,7

18

0,05

19

0,3 1

20

0,05

23

1,25

1,8

24

0,10

25

26

0.15

27

3.09

7, 1

28

2,0

0.6

29

1,56

8,0

30

5,87

31

1.61

32

0,85

1,1

33

0,05

34

1,94

0,2

35

l.61

0,6

36

1,68

0,5

37

0,2

0,4

38

0,31

1,5

>----

Dolni V!stonicc

Bodrogkereszt\Jr

39

0,31

1,5

40

0,87

6.55

9,2

41

I.I

6.0

1,77

9,2

42

0,35

0 ,52

1,0

43

2.35

4,2

0,42

0,1

44

1.62

1,3

0,2 0

45

1,31

0,8

46

1,40

3,2

0.3

47

1.89

3,'.?.

fype

I--

85

48

,_____

.-

- -

2,40

1.7

88

0.25

49

1.05

1.7

89

1,10

5-0

0.25

54

0,05

55

0,05

56

0.05

57

0.05

1,7

58

0,41

3,3

59

0.35

3,3

60

0,88

1.2

61

1.57

1.5

62

0,93

0,5

63

0,47

65

3,78

15,5

66

3,72

3,0

67

0,10

1.8

68

0,05

69

0.25

0.2

74

1.15

2,2

75

0,52

76

1.51

I.I

77

1,83

4,6

78

0.1

84

0.1

85

21.93

86

87

5,87

0
0

Comparing the two type lists, they agree well in lhe

lack of certain tool types (negative identity) which can
be considered culrure-speciiic: general lack of borers
or low number of them, tofal lack of Western. type
blade points and classical leaf points
In most types, the differences are not significant
(BodrogkeresztUr is less complex). 1n case of differences, BodrogkeresztUr seems more archaic (higher
ratio of side-scrapers); its more ro ugh, bulky character
is manifested (type 65. in favour of Bodrogkeresztur,
type for 85~89. Dolni V~stoni ce)
Dufour-lamellae considered "fossile direc toire" at
some other sites are missing from both sites .
The typological system used by J. Svoboda, for the
analysis of Dolni Vestonice ll - Weste rn Slope
(SVOBODA 1991) could be adopted well:
His types were grouped in 5 units, similar to the
taciturnity of Demars's column graphs (DEMARSLAURENT 1989, Fig. 62.), thus investigaling the distribution ratios of the tool kit This condensation of infonnation, however, has its draw-backs:
finer details of the induslry are lost - just L11e ones
which elucidat~ its chronologicaJ position, roots and
contacts (such characteristics for Bodrogkereszrur
would be- the archaic, A.urignacoid character, the trend
for macrolithisation due Lo abundance of ra\'V material
or other reasons, e.g., forest environment)
high ratio of "other tools" covers up the contraction of heterogeneous tool groups. ln case of Bodrogkeresznir, 40 % of the total material wouJd be disregarded this way, and it seems just too much infonnation wasted
backed implements and microlithic tools overlap
considerably.
Comparison between Dolnf V~stonice and Bodrogkeresztur according to Svododa's categories is
presented below:

Comparison of type groups from Dolni V~ston.ice and Bodrogkcreszrur, Svoboda's categories:
Dolnl Vestonicc I
Type group

(KLIMA 1963)

Dolni Vestonice 11
(SVOBODA 1991)

Bodrogkeres:.itur

pieces

pieces

pic:ces

283

14,8

10

200

26,1

!Burins

536

28.1

102

31

213

27,8

J13ackcd implementS

562

29,5

112

34

54

~crolitbs
ther tools

147

7,7

20

378

19,8

63

19

300

39, J

Total

1906

100%

329

100%

767

End-scrapers

- -- -

86

100%

As it is emphasised by J. Svoboda, the chronological position of Bodrogkeresznir-Henye is unambiguous:

"Posirion qf Pavlovian at Pavlov (27 000- 25 OOO

Henye (about 28 OOO B . P.). In this eastern region, the

lower Graveai.an horizons still may appear in cher nozem soil (the Mende soil). In South Poland and irr
Austria. the sites ofSpadzist.a C2-iayer IV and the Wil-

B. P) and Predmosti (26 870250 B. P.) is chrono-

len.do1J sequence. beginning wirh layer 5, fall in this

log1cally comparable to the mean dahng ofD V 1 and

II. while a1 Stranslui skala Ila we still found /a1e
Au.rignac1an m the corresponding srra1igraphic level.
The Graveuian settlement at Milovce seems to be
slighr(v more recem (25 500-22 OOO B. P.). Generally.
the Pavlovian may be placed into longer Lime-span berween 29 000-20 OOO B. P.
Earlier phasis of the Moravian Pav/avian is concempor01y wirh several Gravettian sites in the Carpathia11 Basin: Nemfowi (28 570 ::!: 1345 B. P.), S/aninova Cave (27 950 270 B. P.) and Bodrogkereszhir-

same period".

(SVOBODA 1991, 14.)

This chronological position is corroborated by the
typologi caJ corn position of Bodrogkeresztt'.l.r-Henye.
Different sires of the Pavlovian culture were compared several times and by several authors. One of
these was compiled by K. Valoch, using the classical
indices (VALOCH 1986/87, 60.). The values of Bodrogkeresztilr have been supplied accordingly:

Dolnl VCslonice 3
Habitmion

Partie
lnferieure

runasd' os

215

IG

- -

Pavlov n 4

BodrogkeresztUr

ll

77

133 8

416

3 88

812

26.SJ

t.S,19

12,71

12,01

25,00

18,6

IB

35.55

3'2.47

29,07

37,98

45,62

35,9

IR

6,05

6,49

0,90

1,20

1,55

5,8

ID

15,81

12.9.9

37,67

19,95

12,11

11 ,9

lpf

0.30

(0.24)

(0,46)

(0,'.!4)

appointcs

1,40

10.38

4,33

4,57

2,06

J.2

IGA

2 ,33

0,15

0,26

1.6

IBA

0,47

0.37

0,72

1,2

199

73

1259

385

341

634

ComposiLes

8,04

8,22

6;17

8,05

13,78

2 1,8

Couteau
Kostcnki
Lames

In the essay of J. Koz.lowsk.i (1984), material ofirrationaUy large geographical distances are ccompared
to each other (Doln! V~tonice, Pavlov, Ostrava-Petrkovice, Corbiac; KOZLOWSKJ 1984, Table Ill.) for
expressing essences of the cultural phyla I chronological levels.
Comparing statistical values presented here with
their equivalents from Bodrogkercsztu.r, evidently,
co mparable values were found only on the Moravian
s.ites.
Whatever was said for Doln! V!stonice is valid for
Pavlov and Ostrava-Petrkovice as well, that is:

in the first place, they share negative evidence,

i.e., the complete lack of certain type group s
the presence of rabols and carenold end-scrapers
lend Bodrogkereszrur an archaic character

The recently excavated South-Moravian settlements from the Pavlovian period (Milovice) represent
a younger phase of the culture. The radiometric dates
are around 25 kyear, whloh means a younger phase of
the lnterpleniglacial. From the arcbaeologjcal point of
view, the industry seems fairly archaic, with Aurignacoid character (OLlVA l 989, 112.).
87

The arch of Upper Palaeolithic settlements with

mammoth hms. ranging now from Southern Moravia
till the Central Russian plain in an arch, th~ stations of
which rnay represent milestonesof a possible connection from the Don-bend via th:e settlement along the
Dniepr (Mez:in) and Soul.hem Poland (Krak6w),
through the Moravian hills tiJI Lower Auscria
(Langenlois), with important ''evasions'" to the North
like Zaraisk.
The Hungarian equivalent of the yaunger Palvovian horizon, represented in Moravia by Milovice can
be Nadap. Its geocbrooological position allow this
supposition. Over 60% of the 58 tools of the heavily
disturbed senlement were made of erratic flint (supposed Northern, at least Silesian connections!). The
fauna is dominated by Equ11s germanicus remagencmsis (DOBOSl et al. 1988). Traces of mammoth huts,
even mammoth bones were not found at Nadap.
West
The Upper Palaeolithic of the territories adjacent to
Hungary from the West
- belong to the base tribal area of the given cultural
entity
- fortunately for us, bas been investigated well and
for a Jong Lime
- has been complacently published, mainly in de-

rails
Within a circle of not more than 20 km, classical
sites line up in the vicinity of each other from Wachau till the Kamp valley like Aggsbach, Willendorf, Krems and sites of the Kamp-valley:Langenlois,
Kamegg and Grubgraben.
The area must have been densely populated in
some periods of the Upper Palaeolithic (or, studied
most intensively in our age). The ropography of the
area is similar to the Danube-bend area in Hungary:

"Bei Melk be1rirt Donau, die bisher in einem breiten. alnerciiiren Tale fliesst, ein enges. romamisches
Durchbruchstal, dos in das moldan6lbische Massh1 tie[
mu/ mil steil abfallenden Wanden eingenagc ist, die
sogenannte Wachau. Bei Krems verliisst die Dona11
das a/Je Gebirge, um das g1osse. bis m 15 km breite
Becken des T11/l11er Fe/des zu dwchsrromen. das in
Norden von einem von~1iegend aus rereti.i:i1en Ablagerzmgen bestehe11den liiigelgeliinde begrenzt wird,
dessen mehr oder minder sreiler Rand '' Wagram"
genannt wird. Das 'Hangende des Wagrams bildet allenJhalben der loss. "
(BRANDTNER 1954-55, 7-8.)

ither immediately on the terraces of the Danube,

or on the plateaux over the adjacent wider stream valleys, the utilisation of natural endowments resulting
from the meeting point of several biotopes created
similar conditions for habitation to those of the
Danube-bend region.
Settlements here are aJso ranged into rwo chronological horizons:
the older horizon is represented by Willendorf 11/5
and the associated sites like Aggsbach
the younger horizon is represented by Grubgraben
and other Epipalaeolithic sites in the Kamp-valley
Recently, Haesaerts compared sedimentologicaJ exper1encos in the recent Willendorf stratigraphy with C
14 dates (HAESAERTS 1990, 212.)
his results, partly, contradict Hungarian observations: Upper Palaeolithic settlement waves and interstadial phases did not overlap, moreover, the cultural development is more continuous than discontinuous
By finding the Pavlov soi l horizon, he draw attention to an importanl fossil soil Level which was not
found yet in Hungarian loess stratigraphy. (Between
Mende Upper and Tapi6stlly horizons). Accepting the
first C- l 4 date for BodrogkeresztUr literafly, the se11 lement fits in the uppennosL layer of StiUfried B soil horizon; lhe settlements found c losest as archaeological
analogie.5 are, however, younger.
Stillfrieo in itself is a key site. Its topographical position is optimal: the cca. 400 m a.s.I. hilly region of
the Morava/March vaJley is slowly opening towards
the plain extending rill the Danube, the Moravian
Field. Tt is an ideal setting for a Palaeolithic settlement.
Stillfried is a key site for Hungarian geomorp hol.ogicaJ research as well, because the key section of
its loess profile served as an type section for Hungarian soil- and loess-formation periodisation as well.
Tile humic level of the Bodrogkercszrur-Henye cultural layer could be paralleled Lo Stillfried B soil horizon. In respect of the site itself, it could serve as a station of tbe route leading from Bodrogkereszuir-Henye
towards the base areas of lhe Pavlovian culture, Willendorf and Pavlov.
The r~sull of Ute studies over 25 years is the publication ofWestwaU's n.e w excavations.
The archaeological finds comprise; 4% endscraper, 2,8% burin, 14% Graverre and micrograverte,
26,9% Gravecte blade, i.e., backed blade3, 16,4%
backed blade, 33, 7% others: altogether 249 objects.
The cultural affiliation of Stillfried was described
in the followings:

3 I fore we hove to note that is i.s difficult to decide on a proximal fTagment if it used ro belong to a Gravelle point or a

Gravette blade, thus this value is tentative.

88

" Die geochronologisahe Stellung unseres Paliiolithikums isr daher eindeutig: kno.pp Uber dem eigenJlichen Srillfried B 11nd in der begirrnenden '</etzen
Kalrzeit " (Wiirm III). Der in tmserem Profll auftretende gleyarti,ge Horizonr enrspricht der "Gleyfleckenzone ''. die ouch ander!.wo als iiber Stilljiied B
/iegend beschriehen Ktil'd (FINK 1979. 79.). VielleicJ11
li:l.sst e1 sich mic einer en.tsprechenden " vermehnmg
des Wasserha.ushaf1s '', und !assen sich. die .doch leich1
umerschiedlichen Schneckenfaunen in den Schichten
1-5 mil den moglichen Klimaoszilla1ionen im Stillfiied B erkltiren. ''
(FELGENHAUER 1980, 26.)
The author did not see enough reason for adopting
the Moravian name "Pavlovian" for the Lower Austria
sites, gave priority to the name "Aggsbachian" introduced after J. Bayer in 195 I. The eponym site is in
Wachau, dated fo the middle of the Wilrm lJ period
(accord mg to FELGENHAUER J 95 l , 261 .)
"Wir glauben, so Aggsbach frmerhalb unserer erz...
geren Heimar in die f riihe Mitte des Wurm 11-Stadlales
stellen zu dllrfen .zmd somit alter als Miesslingtal rmd
}'iinger als Wil/endorf JJ/5 n1 bezeichJ1en. Jn dieser
zeitlichen Stellung bildet es mit eine der Stationen der
Nfoder6srerreichisch-Miihrisch.-Slova'kischen Gruppe
des osilichen Gravettien.
De11 Terminus "Aggsbachien " gfauben wir ablelmen zu diirjen ~ sowohl im Sirme einer eigenen Kultur a'/s ouch im Sinne eit1es zeirlicli oder raumlich

seJbstemiJig gewordenen r_..,.pus. "

The main cultural layer of tb.e eponym site of the
Aggsbachian was dated around 25 kyear by C-14
method, parallel to the younger phase of the
Pavlovian, i.e., the upper levels of WiUendorf II
thougb typologically and technologically it is different
to some extent (with 50%, ''tlecliette'' and archaic

forms).
The lower I Northern end of the Wachau is lhe region of .Krems where the Danube leaves the midmo\:lntain environment and lbe widening Danube-valley is transsected by, from the North, the Wa:gram
bills, from the south, the flattening slopes of the Vienna forest..
Krems itself is known for its numerous Gravettian
sites. Recent excavatiG>ns identi fi,ed several more sites.
The publications available for lhe author were
partly reappraisals of old excavations: Lhe modest

fo course of her Central Buropean activity, A. Mon-

cet-White encountered aJso the difficulties of working

on poor assemblages - both in absolute quantity and
quality ( Bosnia: Kadar and Austria: little part of Grubgraben)-more exactly, wbere the find material could
be accessed only in fragments or details. She was
trying to fit Grubgraben material into dle range of
sites using the SonnevilJe-Bordes type list. Her
grouping offered a more realistic approach though
assemblages under I 00 pieces should, as a mle, not be
treated in the statistical evaluation (MONTETWIIlTE 1990, 155)
Grubgraben belongs to the younger phase of the
Gravenian culture~ On the basis of the few tools published in drawings (MONTET-WHlTE 1990, Fjg.
lX. l-12.), the material is analogous with the material
of the m iddJe settlement waves of the Grav.ettian culture in Hungary.
The real significance of rhe Grubgraben material
excavated and published by Montet-White lies in the
unusually carefully documented settlement features,.
as Lowe r Austria. is known to be very rich in archaeo logical finds from this period. Same as Gnib.graben is
much more than I.be publ.isbed little details as I had the
chance to see, thanks to the kindness of Prof. F.
Brandtner.
The Early G.ravettiantooUcit of Langenlois, Kargl's.
Brickyards was found accompanied by a hut base (?)
formed of mammoth bones and teeth and a tool producing workshop.
The site would deserve more attention, as far as the
dwelling hut is concerned, but unfortunately l have no
detailed infurmation on the find material (HETNRICH
1974-1975, 16.).
Langenlois is so far the most westward point in the
slightly arched, very long, east-west directed stripe of
settlements extending over I 500-1800 km, characteri~d by the mammoth hut and, partly, .t he Kostenkipoints.
South-West

The morpbGlogy oftl1e southern parts oftb.e Carpa~

tnian Basin is changing decisively only south of the
Sava valley, by the rising northern fore-hills oftbe Dinarian Alps. The avruJable scanty data may reflect low
number of inhabitants in tbe Upper Palaeolithic and/or
low efficiency of archaeological studies on the peri0d.
(DOBOS! 1967. l84-1'93.)

Grravettian material of Krems-Hundssteig, rich in

The best known site along the Southern bank of the

m icrograv.ettes (bardly SO tools :that were closest to

Wit.lendorf IJ/5 according to the publication (RAHN
1972, 88.)

river Sava is Kadar. (MONTET-WHITE et al. 1986),

which has, at the same time, comparable lithlc industry:

89

Comparison of Kadar and Bodrogkcreszrur-Type

groups according to the SonueviUe-Bordes-Perrot

I
I
End-scrapers

Kadar%

Bodrog-

kercsztlir %

26

2.7

3,2

28,9

Backed cools

18. l

2.2

Within I.he fonner,

microliths!
Retouched blade

23%

End-scraper
combina1ions
Borer

-Burin

I
I

-Side-scraper group
Lamelle group

20,3

19

6,8

14,3

:?8,6

3,3

As it is apparent, Kadar is a typical young Epigravettian industry, with very high Iatio of backed
blades and microlithic tools, specialised on shouldered
poinlS. There are very few common elements with
Bodrogkcreszrur: onJy the indifferent retouched blade
ratio seems similar (MONTET-WHITE er al. 1986,
59.)
The archaeological material of Ll1Scic (Aurignacian), a conlemporary site to Bodrogkereszt\1r towards the south, sourh-west, cca. 500 km far from
BodrogkeresztUr, did not reach the lower limit of statistically valid quantities (MONTET-WHITE et al.

1986, 83.)

800 km from the cenJ.raJ area of events in a cave of the

Karlukovo karst.
Branching off the Gravettians to SE Balkans in
the early phase can only be raised yet, without connecting evidence, as a " tentative" hypothesis.
treating the contradictions between theories and finds, the
facts should receive priority: in our case, che TemnataWilleodorf Oraveltian analogy.
On the occasion of Tcmnaca cave, scanered finds
from the Epigravettian of the Souihem pan of the
Alft>ld (Bacska region) are mentioned by the authors
(DROBNIEWlCZ-GINTER- KOZLOWSKJ
1992.
420.), which cannot be connected to the Bulgarian
site. The real analogy lies some 250 km more to Lhe
north-east, which is not so much more (250 km) further on, at Bodrogkereszrur.
One of the raw mater types described from Temnara cave was classified as Northern Hungarian limnic
quartzite by the authors (DROBNIEWTCZ-GlNTER
1992, 420.). lf Lbis opinion can be proved, it will also
&erve as a corroboration of anocher observation made
on long distance spread ing of raw materials at such an
early date. Namely, the bulk of the Esztergom- Gyurgyalag Upper Palaeolithic s ite was made of Prut s ilex,
rested by neutron activation analysis as well (VARGA
199 J, 269.). Some preferred raw materials could be
traded/carried much fwther than in the "the original

rn

cradle territories of this taxonomic unit" (DROBNIE~\IJCZ-GlNTER-KOZLO WSKJ 1992, 420.). Should
the raw material Nr. J 7 from Temnata 11eally originate
from North-Eastern Hungary, this fact would also indirectly corroborate the connection between this site
and tbe contemporary BodrogkeresztUr, lying in the
centre of North-East Hungarian limnic quarLZite
sources.

South-East
East
Because of opinions expressed on the Gravettian
culture and references lo Hungarian material, it is
necessary to deal with the material of the TernnataCave.
Apart from the general theorems on the origin, distribution directions and rhythm of the Gravettian culture, there are three reasons for discussing Lhis site in
the monograph on the Bod:rogkeresztU:r-Henye site:
It belongs to the earliest Gravettian settlements, so
it is a contemporary site. its analogies in the Central
Danube-Basin and to the east of the Carpathian Arch
agree with the list presented for Bodrogkereszt:Ur. The
site can be hardly fit to the chain of events hypothesed
by DROBNIEWICZ-GTNTER- KO:Z!.OWSKJ 1992,
412, i.e., formation of the Gravellian culture in the
Cenrral Danube-Basin, spreading towards the West
and after getting stronger there, invasion towards the
East and flourishing between the Don and Dniepr
rivers. Temnata lies, even as the crow flies, minimum

90

Examining the Eastern connectio ns of Bodrogkeresznir-Henye we have to consider territories with different level of cognisance and/or documentation . Thus
the potential contact direction, content and chronology
will be presented hete in geographkal order.
Transcarpathian Ukraine

The importance of the marginal"-regions is unde rlined by the se~lement pattern of the north-eastern
limits of the A lfOld (Great Hungarian Plain) in the Upper Palaeolilhic period.
Either because cultures were "jammed" before the
arch of the Carpathes or j usl as a consequence of using
well-known advantages of the meeting point of several ecological niches, on the small area belonging to
Ukraine a range of varied Upper Palaeolithic industries were found.

Comparing Bodrogkeresztlu to Lhe Korolevo Upper

Palaeolithic industry is not realislic, because
the number of true tool types is very low within
the assemblage

Mts. (Nagyhegy), which yielded, however, a low

number of finds nor suitable for a statistical evaJualion
(TKACHEJ\"KO 1989, 222.).

over 90% of all artefacts were made of andesite

The Kostenki-Borsevo circle is already falling o utside this essay. In the archaeological elaboration, the
traditional method of hunting analogies by each type
was not followed. There is, however, one exception:
Laszl6 Vertes found an exact analogy of the obsidian mousteroid point from Bodrogkeresztlir (Fig. 28.
I) at Kosten.ki 15 (Gorodcovskaia) site (ROGACITEV
1982. Fig 58.12.).
This special industry is to some extent different
from the general image of the Kostenki-type industries. Al the same time, it is an eponym site for a facies, Gorodstovian within the Kostenki circle occurring at several points.
The distribution of the eca. 370 tools is quite specific, with 25% end-scraper, 30% piece esquillee (!)
and Chatelperron type blade points. According to
chronological position, t.11is facies belongs to the
middle horizon of the culture, around 21 Kyear. (ROGA CHEV-SrNITZYN 1982, 162- 171.)
The tools of Kosrenki 15 are known to us only in
diminutive drawings. Analogies of the fonns were
found in great number without implying immediate
genealogical connections.

the chronological position of both Upper Palaeolithic cu.ltural layers rendered any statistical comparison unreal
(On the basis of the drawings, Korolevo TI repre>sents partly a Babonyian type Middle Pa.laeolithic,
partly transitional forms between Middle and Upper
Palaeolithic forms like ' Eger-culture". From the artefact inventory of Korolevo I some forms resemble
Bodrogkercsztur, as much as one can judge from the
drawings. (OLADlLIN-DEMIDENKO 1989 Fig. 1920. USLK 1989, 179.)
The stratigraphical position of lhe sites was found

".. . nor fitting into the generally accepted chrono-

logy..." by the authors (GLADTLIN- DEMIDENKO

1989, l 77.) They considered the Carpatho-Balkan region as a centre of the multi-rooted development leading from the Middle Palaeolithic to the Upper Palaeolithic c ultures.
The topographical position of Korolevo practically
invited a Palaeolithic habitation. The river Tisza enters
the plains between Huszt and Nagysz61os, in other
words, it means a possible route for crossing the Carpathes along the We-stern Slopes of the Carpathian
forests, the Avas hills and Nagysrolos with 800 m
average altitude a.s.I., leading till the Tartarian pass.
Anyway, ilS stratigraphical position under the
Brorup fossil soil needs further arguments (provided
that Brorup/Anunersfoort is identical with Emiliani
Sfc, i.e., Lhe Mende base soil also in a Transcarpathian
sense as stated by both Pecsi and Ringer (PECST 1992,
Fig. 10., RINGER 1988, 80-8 l.) ln this case, the
Korolevo I soil c-0rrespond the "acme" of Hungarian
Middle Palaeolithic.
Reviewing the sites of the " Beregovo group" we

can observe the following:

Their raw material utilisation. type distribution and
tool producing technology and traditions are common.
The most numerous assemblage is Beregovo l with
114 tools, the type distribution of wbich seems similar
to Bodrogkeresz:tUr at least in ratios and its well documented stratigraphical position is also corresponding
to that of Bodrogkereszt:Ur (as Tkachenko wrote: .. . in
rhe lower part ofpale-yellow loam 4 above the second
paleosol Paudorf.. ," TKACHENKO 1989, 2 14-216.)
The distance between the two sites is not more than
80-90 km. Most probably, they both belong to the
same cultural phylum. This may well be true for the
other localities belonging to the Beregovo-group,
ranging along the southern margin of the Beregovo

Romania
Contemporary s ites of the Romanian Upper Palaeolithic should be further divided into two geographical
groups: inside nnd outside the Carpathian arch. lo respect of BodrogkeresztUr, basically the former region
should be considered: marginal regions within lhe
Carpathian Basin, Upper Palaeolithic sites from nonhwestem parts of Romania. This area is heavily dissected by river valleys, opening lowards the Wes t to
the Szatmar plains. In the Upper Palaeolithic cultures
of the region all phyla are present that were found on
the foothill slope regions of the neighbouring countries, which are politica.lly different now bul the geomorphological, topographical conditions are absolutely identical with those of this region.
The Gravettian industries of the subsequent intcrstadiaJ settlement waves can be obviously connected
to the material of the settlements around Beregovo.
Though the available publications cannot be used
for statistical evaluation, the sites ofBitiri's Gravettien
f. phase ( "p/11s a'1Ciem "'), Calinesti and TuruJung were
obviously conlemporary with Bodrogkere&ztur as
"Post-Aurignacien assemblages (BlTIRl 1969, 530)
The early phase of the Gravetlian cu.lture in the
Avas (Oas) region - according 10 the terminology used
by Carciumaru, O haba interstadial- was found different from the Sagvarian by Chirica as wdl (CHIRJCA
91

1991, 12- 14.). Most probably, these sites belong to

rhe- circle of Bodrogkereszrur. This observation seems
to be supported by the map published by Chirica
(CHIRICA 1986, Fig. l .), on which the Transylvanian Basin is strikingly empty. Sites are concentrated mainly a long Lbe Eas1em slope of the Carpathes
along rht.! rivers Sirec and Prut as well as the valleys
opening to these river basins, running down from the
high mountains.
Topographically, m.alnly the Gravettian settlements
in the Sz.amos (Somes) valley in the Avas (Oas)Iegion
should be considered as potential connections for
Bodrogkereszrur I (CHlRlCA 1986, sites 61-67.)
The more large, populous and more important Upper Palaeolithic centres were found outside tJ1e Carpathian arch, in Lhe middle part of the Prut valley. The
"chief site" is Mitoc-Malu Galben. Apart from the
completely identical chronological position ( HONEA
1994 , 144 ), some further derails offer further important analogies: amulet pendant with incised margin
and carved surface (CARCTUMARU-CHIRICA
19&7, 68.; GRTGORIEVA-ANIKOVITCH 1991, Fig.
9.)
The importance of the incised margin discs is envisaged by the author less in their supposed or real
symbolical meaning bur rather the connecting role of
this object 1ype, allowing an asymmetrical connection
between Bodrogkeresztur and Moldova.
Asymmetry is understood in distance: the distance
between Bodrogkeresztfu and Mitoc extend to some
4/Sili of the distance between Bodrogkcreszrur and
Moldova. The importance of incised margin discs is
seen by the author not in their hypolhesed or real
symbolical meaning but their significance as direct
proof of common roots. The identity of shape and
ornamentation on objects not influenced by functional
requirements of everyday work can be a strong
argument on common cultural roots and close relation.
Mitoc-Malul-Galben belongs accordfog to the
chronological system by Chirica lo the second period
of Gravettian industries in Romania, dated between
28n9 Kyear-24 Kyear. This period is represented by
eminent sites Like the great Austrian or Moravian settlem.enls. Bodtogkeresztur with slightly older C-14
date belongs to this horizon, but in the opinion of
Chirica the absolULe chronological date of the Hungarian site " ... contains a very great range of wzcerlainly..." (CHIR.TCA 1989, 150.)
Ripiceni Izvor is the other multi-layered Palaeolithic site along the central parts of the Prm basin with
significant Gravettian settlement (28 thousand artefacts. two Cllitural layers with four levels). This site
contains an important industry of the Molodova-Mitoc
c ircle. showing continuous development during the
Upper Palaeolithic period.

92

The area between the Podolian plateau and the Carpathcs known as "Ukrainian Switzerland", dissected
densely by rivers and with ample excellent Cretaceous
raw material deposits offered ideal environment for
the interior development of Palaeolithic cultures condensed before the Carpathian arch.
In four levels of the Gravettian period starting with
the Tursac interstadial period Lhe litbic industry is
fairly homogeneous (the ratio of end-scrapers increase
linearly towards rhe younger levels from 26 to 38%
while the ratio of bur.ins decrease to some extenl from
16 to 15%). The specific character of che industry is
due to bifacial working and the low ratio of the
eponym type (Gravette-points, PAUNESCU 1993,
153-171.).
Molodova
Among the eastern connections the localities ofchis
region should be also mentioned briefly.
Bodrogkereszttir lies about midway berween Molodova-Corpaci and Willendorf. The disiance between
Bodrogkereszllir and WiUendorf towards the Wesl is
roughly 500 km as the crow flies, the environs of
Molodova (Cen1ral Dniestr region, CHIR.lCABORZlAC-CHETRARU 1996)) is roughly the same
to the East.
The conneocion, however, is far from symmetrical:
while towards the Wacbau. the route leads along the
Danube valley amidst mountains weU below I OOO m,
towards the East the Carpathfan Basin is closed by the
range of the Eastern Carpathes surpassing 2000 m.
which was not the typical living area for the Gravettian people even in the possible ice~free periods.
The gates of the East.cm Carpathes, the passes
(Verecke, Tatar-pass), are narrow passages at irn elevation of 900 m. The circumstances were not favourable
for the diffusion of cultural effects (DJTNDJANKOZLOWSKT-OITE 1999, 30). Lnformation based
on condensed large scale maps however can be contradicted by personal observation. The Alps which had
similar or even more rough orography were not. impassable during the Upper Palaeolithic period. Regular. stable connections, however, were not established.
The possibility of Eastern direct contacts could
only be imagined for those who, as F. Bordes put it,
.. were eager to know what was over the mouncains".
In Western - Nonh-Westem direction (Moravian basin, Danube--valley) ll1ere were no obstacles to lasting
contacts.
Towards Lhe south-east.. the climate of the Transylvanian Basin is much colder than Lhe average temperature of the Carpathian Basin even today. Apa1t from
the unfavourable conditions, large distances and high
altitude separated them from the setilements of the

Moldavian culture flourishing along the Eastern

slopes of the Carpathes.
North

AcGording to personal impr<!ssions of the author,

the Cejkov industry is smaller, more, gracile than Bodrogketesztilr.
Comparison with th.e site Krak6w-Spadzist.a hits

against double difficulties:

A long the Southern, Hungarian margin of the
Tokaj-Eperjes Mts. Lhere were several Upper Palaeolithic settlements - among them, Bodrogkereszt:Ur. On
the eastern margin of the northern parts of the mountains there were two large settlements, adjoining the
Bod.tog valley: Kasov and Cejkov.
The Kasov industry can be assigned to Epigravettian, one geochronological phase younger than BodrogkereszrUr, settlement of lhe medium wave of fue
Giravettian population. It is a Large and rich settlement
with more than 4000 tools/artefacts.
"Le nivea/I inferieur presenie wi habitat assez
re.streint (12 x 8 m) caracteris, par une industrie a.vec
des traits gravettiens plus prononces (aux d 'outils <!i
bord abanu J0,5%) et executee en grande partie a.vec
d11 silex ''nordique" provenant d11 Sud de la Pologne
(54,4%). L 'importaticfl systemaJique du silex "nordique " a l "inierieur du bassin carpathique et du

Danube moyen est un phenomene caracteristique... "

le niveau supiriew~ probablemenl postp/eniglaciatre, d'apres Jes donnees sedimentologiques,
represenre une industrie avec foible raux. d'etements a
bord abatt11, sans formes diagnostiques gravettiennes,
mais /argemem domineepar les gratioirs et les burins.
011 y remarque certoins e/eme.rics ..aurignaciides "
co1i1me des grattoirs a fro/II eleve, des bu.rins
carenoides: cest 11n phenomene plus large dans certa'iines ind11stries post-p/eniglaciaires de /'Europe
Centrale."
(BANESZ et aJ 1992, 19.)
T he geographically closest analogy to Bodrogkeresztlir was the site Cejkov in Slovakia, lying in a
distance of cca. 50 km from BodrogkeresztUr, in the
other (Slovakian) part of the Tokaj-Eperjes obsidianregion.
The older level among the two cultural layers found
here probably belonged to the Denekamp-period
(BANESZ 1962, 758.; BARTA- BANESZ 1981, Fig.
I.) The main cultural layer of the site is the younger
Gravettian horizon. On the basis of chronological
data, the Cejkov interstadial probably corresponded to
the older embryonal soil boriwn of the Sagvar period.
(BANESZ 1996, 7)
The older level, which would be more important in
respect of BodrogkeresztU:r, has not been published in
details as yet.

L. BodrogkeresztW--Henye did not belong to the

Ko.stenkian culture like the statistically interpretable
Site B, with mammoth bone hut
2. The other side of the same difficulty: the probably closely related Pavlovian industry (Sit~ C2 Level
IV.) with its 24 pieces of tools were not adequate for
comparison.
To demonstrate the essential difference between
BodrogkeresztUr and the Kostenkian industry from
Krak6w-Spadzista, the statisrical comparison of Lhe
two sites were performed and presented below:

Krak6w-Spadzisra,
B

Types

Dodrogkcrcsznir

.Pieces

Pieces

end-scrapers

2,97

l57

24

Burins

57

24.25

173

27

cembined tools

0-,42

16

2.5

Truncations

2,97

13

2,0

27

11.~8

24

10,21

90

14,0

0,42

18

2,8

Backed imp.

36

15,3 1

32

5.0

sh ouldered

63

26,8

1,4

retouched

12

S, I

60

9,3

57

8,8

20

3,1

Kostenki

kniv.es
ret. blades

Borers

flakes
side/scrapers

Notched

(KOZLOWSKI-SOBCZYK.1987, 66/68.)

Disproportionate features arising from tlte comparison of the type list

dominance of cultufe specific types in Krak6w
(shouldered blades, Kostenki-kni.fe)
higher number of archaic types (side-scrapers,
notched. tools) at Bodrogkereszttir justify the chron.0logical pos.ition of the Kostenkian type industries and
the Bodrogkeresztlir finds.

93

S. Cultural affiliation

In the first publication on Bodrogkereszrur-Henye,

Laszl6 Vertes wrote die followings:
"ANALOGJES OF THE INDUSTRY OF MT.
HENYE. The mdus11y is evidently a manifestation of
the East Gravetlian group and absolure chronological
dare places it lo the older ones. Typologically this is
shown by tl1e frequemly found A urignacian retouch on
rhe implements. while the A 1trignaaia11-type endscrapers are missing; by the relatively low percentage
ofbacked pygmy blades. by ihe uniformly large size of
the implements, i. e. the lack of microlithization which occurred already in Arka - it creates a striking
comrast 'fith the gigantholiths there. Eve1y old East
Gravettian material contains analogies to some ofthe
types known from Mt. Henye. rvERTES 1966, 13.)
The seventeen thoiLrand year old material from
Arko and Sagvar. the 10000 year old material from
Szeksza'rd and that from Szob being also yotmg
(vo1mger tlum the maxin11m1 of the W 3) contain certain elements which could be defined as aurignacoid i. e., thick. short or almost circular end-scrapers
which are fairly close 10 the keeled scrapers...
... This means Iha! the awignacoid" feat11resfo1md
in ihe indL1Stries of W3 or even sooner do not necessal'i(l' relate them to erhnic units wilh temporal and
spatial contiguity. i e.. to relate them directly to the
i11/ieritance of human groups living I 0-15000 years
earlier. These types may be regarded as the result of
new invention co1111ected with increasing mic1olithizatio11. In this connotation the two types of Gravettian
"groups" found in Hunga1y might represe11t two temporal horizons ofan idemical cu/111ral phylum.

(VERTES J966, 14.)

The large amount of scientific knowledge accumula1ed during more than three decades passed since the
first publicacion re-arranged our informarion on the
Upper Palaeolithic of both Hungarian and other
Central European sites. Its seems that the cultural
development we are able to follow by archaeologicaJ
methods is gerriog more and more precise: valid for
shorter time span and smaller regional unitS. In 1966,
the more or less generally accepted view stated that
the population of a uniform culture was present during
some 20 thousand years from the UraJ Mts. till the
Rhine, i.e., the Gravettian culture. The gradually accumulating wealth of knowledge made possible the
separation of several chronological-areal groups and
tech.nocomplexes in recent years. This fact, however,
was not so fortunate for the nom enclature.
Let us briefly survey the most important essays for
dissecting-grouping the Hungarian Upper Palaeolithic
period:

94

KOZLOWSKI-KOZLOWSKI 1976, 1983 ,1996

The whole "Pavlovian" tem1inology was introduced to public awareness of the profession in the sixties, as summarised by the important paper on the history of research by K. Valoch (VALOCH 1996, 205206). This terminology was raken over by J. K.
Kozfowski and used in the ir review paper written
together with S. KOZLOWSKI, as well as in his own
more recent articles.
On the map published by J. K. Kozlowski in 1983
the core of the sJ. Gravetrian culture was transposed
to the PavLovian base areas. The cultural emanation (in
respect of certain characteristic types) was staited
from this core to different direc tions in the different
periods.
The base fonn of the Kostenki knife also appeared
first in the "Industries de tradition Pavlovianne avec
poims cran du type oriental MorCiVany Podkovica"
and reached from here partly to the West (Wi llendorl)
and U1e east (towards Cejkov-Kasov, later Molodova).
Cejkov and Kasov represent the Pavloviao sites in
Eastern Slovakia. from here, culture radiated immediately towards north-east (Kostenk.i), or to the north
towards the South-Polish (Krakow-Spadzista, Zwierzynec, Witkowice) sites - and maybe from here direct
to the E~ in direction of Kostenki. ln the meantime,
the route ofthe Molodovian culture proceeding slowly
towards the North was crossed (Lipa). (KOZLOWSKl
1983, Fig. 19.)
Jn respect of nomenclature (i.e., " Pavlovian''),
again we can raise I.he question - can it be valid over
such a large area?
Though from reasons to be specified below the
author herseJf is using the Pavlovian terminology,
there are some intriguing points here:
Condensing such wide temporal range and such
variable tool kits under the name of one culture reduces the information simply to I.he name of a lime
span, more exactly, a chronological phase of a certain
area. There are also arguments for keepi'n g the traditional term "Gravellian" for the specific industries of
several small communities as the name indicates
something common (basically, though in a secondary
manner. presence of the tool cype named after the
original site.) ralher than preferring 1he term Pavlovian,
wb..ich has.more regional connotations.
His views were modified to some extent in the review papeqlublished in 1996 (in which the Hungarian
Upper Palaeolithic is still irrationally under-represented): The Gravettian of the Danube-basin was divided into U1e following chronological levels: Early
Pavlovian (30-27), developed Pavlovian 27-24) and
lhe horizon of shouldered points (24-20) (KOZLOWSKT 1996). On the basis of this grouping, the

low representation of Hungarian sites can be reasonable, because this division is not valid for the Hungarian sites at that moment In the most recent comprehensive work, he mentioned only five Upper Palaeolithic sites from Hungary (DJINDJAN-KOZLOWS-

Kl-OTIE 1999. 383).

OTIE 1980-81
M . Otte presenting the preliminary results of his
Gravenian monograph mentioned only the Epigravetlian sites from Hungary (Sagvar, Airka, Pilismar6t.
Szeged)therefore he wrote:
D 'une .facon generale lo plupart des sites gravei.ciens eta/et)( contemporain<1 soil de la fin de l 'imers/ade de Still.fried B (= Arcy-Kes~eli). soit du scade
rigo11re11x s11ivallt (portant ici le no fil); certains vonc
j11sq11 "a /'oscillation remperes correspondant a
Laugerie-LaSCQIL'C. Les ensembles epigravettien.s hongrois semblem plus rece1'1/S encore.
D 'apres les dates radiocarbones disponibles, la
periode d 'occ11paJion gravel/ iene s 'etend de 2 7000 a
20000 BC (sauf pour les sites ho11groi) el I 'on observe les occupations les plus ancie1uies en BasseA 11triche (Wi//eJ1d01f il/5) et en Podolie (Molodova
VIX IX).

(CITE 1980, 175.),

and classified the Hungarian_ sites summarily into the
last g roup (Nr. 8) to the Epigravettian industries
(OTIE L980. 189.)
The large-scale comparison by M. Otte considered
the tool kit of 21 Giavettian sites. (Mainz, Geissenklbsterle, Brillenht>hle, Mauem, Bilzingsleben,
Lubna, Revnicc, Willendorf Il/5-9, Aggsbach, Dolnf
Vestonice 11, Pavlov 11, Pecrkovice, Manmtova, Wojcice, Krakow-Spadzista, Cejkov, Molodova V/Vll)
Among the sites studied, the number of artefacts were
typically well under the range of BodrogkeresmlirHenye: exceptions to rhat were only Willendorf U/9,
Aggsbacb and Molodova VNm complex where the
number of tools studied was adequarely large. (OTTE

1981 ,86.)
The type groups ("Types de vestiges. OTIE 1981,
73) specified by Otte were defined for Bodrogkeresztlir and the values were fit in the comprehensive
table connecting several sites of the entity.

Otte 1) pes de

vestiges

.BodrogkcrcszturHeO\c
Megyas1.6

Willendorf IU6

! Pavlov II

1-1
debiiage
dcche!S
doutils

posteriori

Outils
retouch6

73.8%

2,6%

23.5%

84%

9%

7%

71%

1.6%

27,45%

62%

2,2%

39%

KrakowSoa<Uista
Dolni V~tonice

Aggsbach

_u

-'- -

594'/o

10%

31 %

53%

5%

41%

50%

1,3%

49%

The ratio of the groups separated by Ottecan be influenced by the extent of collection, the ratio of excavated surface compared to the total area of the settlement and the function of the of the settlement Still,
the observable tendencies seem to indicate an increase
in 1001 production efficiency.
After t)le analysis of the tool-kits, the foLiowing observations could be made
there is no connecLion with rhe Central Euroean
sites (Otte's Nrs. 1-5, comprising Mainz, Geissenk6slerle, Brilleohohle, Mauern, Bilziogsleben)
sites 6- 11 (Lubna, Revnice, Willendorf IT/5, 6, 7,
8) and the Bodrogkeresztt'.tr finds resemble most in the
ratio of worked blades (Otte's type groups 4, 5. 12, 17
and composites)
relation with Pavlov n is observed in the ratio of
burins, percoirs and bees
the ratio of end-scrapers in Mamurova cave and
BodrogkeresztUr was equally around 20-21 %
at Molodova VN ll, the ratio of lamcs appointes
and encoche-tools were identical with Bodrogkeresztlir

the individual characteristics of Bodrogkeresztl'.Jr were

mainly the negative features: Otte's type groups 6-9
{microliths, shouldered points) as well as types l4-15
(knives) were totally missing. The only " positive divergence" from tbe sites analysed by Otter were the
high ratio of sjde-scrapers (4,6%) which could be explained by chronological arguments {archaic character)' but also functional reasons (OITE 1981, 87 .)

111e basic grouping was aJso kept in 1996 when

connecting the Upper Palaeolithic sites of lhe PrutDniestr region to the Willendorf-Kostenl..-i circle
(OD"E et al. 1996, 213-226)

In 1989, M. GABORJ argued for a chronological

division of the Hw1garian Upper Palacolilhic/Gravertian period WiLhin the three "Zeitliori.zonte" (between
30-27 18- 16 and 13-23 kyesr, respectively) he did
not se~arate cultural phyla (GABORI 1989)
G. GRJGORIEV AND O. SOFFER 1993 inrroduced a new category for marking a large group of
sites: Willendorf-Pavloll-A dveevo-Kostenki Archaeological Entity (GRJGORJEV 1993, 52.. SOFFER
1993. 45.). This unit was rather widely delineated in
geographical and chronological sense and stand for a
former phase of the so-called Eastern Gravettiao Lecbnocomplex. Though lraditional archaeological criteria

95

for belonging to this "entity" were defined raU1er magnanimously, it is true that al Bodrogkereszrur more of
these features are missing than presenL Leading finds
of I.he entity like Kostenki knife and point or tleche1te
were not found at Bodrogkeresztur.
Considering the missing and existing features it is
suggested here that B-0drogkeresztlir still belongs to
this group.
The evident contact dfrection was to the north-west
(immediate connection to the western units of tbe entity along the Danube-val1ey). Direct contact with the
sites to the east of the arch of tbe Carpathes was hardly
probable.
Arguments for .integrating Bodrogkereszwr Lo the
above entity are the following:
chronological position of the site
presence of mammod1 as prey (even in low quantities: SOFFER 1993, 43.)
the topographic.al position of the site
typological similarities with the relevant sites:
few by types but more in general image
technological-formal criteria which are very difficult to turn to written text, more like general impressions on the Bodrogkeresznir industry

....

M. MUSSI AND W. ROEBROEKS 1996

The study period is divided into three major parts
by the authors, according to the tenrunology of the
1995 Pavlov conference/conveniion:
_ EUP I Aurigoacian culture: both chronologically
archaeologically falling outside our present inquiries. The C- 14 dates however do not exclude the
possibility that Aurigoacian n (Olschevian), identified
by Laszl6 Vertes in lhe upper culcural layer of the
{still6sk0 cave lived here till Lhe tum of the 301h Millennium bp. (VBRTES - de VRIES 1959: 195). The
last d1ird of the Worm period, which can be divided in
lhe Carpathian Basin into three phases, served as the
chronological framework of the s.I. Gravettian period.
On tbe basis of chronological and

and

DOBOS! L996a 126

Technical I lypologictd features this large cultural
complex was divided inlo Lhr~e phyla. all of them corresponding a cultural facies of this complex_ h means
that among roughly identical ecological conditions,
similar way of life and similar "leading finds" (at
least, as regards the gravene-points) site groups for
three different stone-working traditions., technologies
could be separated. The triple division of the Hungarian Upper Palaeolithic period partly follows the drvision suggested by the Pavlovian convention:
- Older Blade lndustries, suggested terminology:
Pavlovian = Gravettian
- S~gvarian /Pebble Graveuian is intercalated (temporally) these two, pa11ly contemporary with the

96

younger phylum, characterised by preferential use or

pebble material.
- Younger Blade lndusrries, suggested terminology: Epigravertian = Epigraveuian
Summarising the efforts for the division of the
Gravettiao entity in Hungary: the best possible solution for a generally valid bul easily adaptable system,
liberal but adequately precise system in tjm c and
space is a mixture of Soffer's entity-t and the division
of the Pavlov convention. The range of sites aU of the
demonstrating some individual features which were in
loose contact with each other, the "big mosaic" can be
easily covered by the tem Gravcttian entity, separated
geochronologicaJly as MUP and LUP, comprising a
range of local cultures.
The considerable theoretical and pracrical differences apparent in the determination of traditional ar-

chaeological cultures, the dirferences in attribures and

I or certain group of sites these cultures cannot be deli neared convinci:ngly enough. It is also doubtful if
they will be ever more evident because-the increase of
the authentically excavated sires increase, or can increase the scale of individual characterisrics just as
much as the possible common features necessary for
the determination of the cultural units. It is also doubtful Lhar once it is so uncertain, is it really necessary to
force sires into any chronological-geographical framework or rather we should describe individual features
within the loose frames of the 'entity".
The real contents of the name Willendorf-Kostenki
culture can be questioned on lhe basis of the large distance betweeL1 the two eponym sites (1800 km).
The route of the Aurignacian migration was even
longer than this distance; but this route was not a
range of contemporary sites but a well dated move
ment from place to place. Sucb a vast area could oot
be covered by one culture already in the Middle Palaeolithic. To the West of the Wachau (WiUendorl), the
same distance would be already deep in the Atlantic
ocean. The two eponym sites are separated from each
other in the given period by a rich and versatile culture
area (Upper Palaeolithic cultures on the European part
of Russia, Ukraine, Moldova, ROGACSEV- ANYlE<OVICS 1984, Fig. 72.) le is hardly possible that the
same population with full contents and culture were
present at lhe far ends. Therefore the name "entity" is
more justified than culture which has too strong implications. Population of the contemporary (MUP) Hungarian settlemenLS could be tbe "filiaJs" of the base
areas in Central Europe for whom the almost idemical
natural endowments offered similar conditions for
living. Ln a relatively short period, missiles,.
from lhe core areas invaded North-Eastern Hungary.
The closest Eastern centre, along the river Dniestr

seemingly did not have much influence on the events

happening at the interior parts of the basin. The population surplus condensed in the Eastern forelanus of
the Carpathes - whether they had immediate contact
with the more distant Eastern cultures or not - could
advance towards the West by avoiding the high mountains towards the Nonh (or, possibly, t11e South).
To answer the question whether the Upper Palaeolithic culrures appearing in the interior pans of the
Carpathia11 Basin were of Eastern origin, independent
of nomenclatorical considerations (I.e., to distinguish
them from Western- and Central European cultures),
the answer cnn only be given in two separate phases.
Eastern. in a sense that the appearance of the Upper
Palaeolithic cultures and Homo sapiens in this region
can be explained from the East-South-Easl Apart
from the general Eastern origin, the population of the
Carpathian Basin during the whole period of the UP,
more exactly, the int()rstadial periods took place probably from north-western direction, the core area at
Wachau-Pavlov region. The mutual and constant connection of the original population with the furtherlying eastern cultures took place, most probably, Lo the
North of tbe Carpathian arch along the corridor be~reen the ridges of the high mountain range and the
permanent ice sheet.
Another problem of nomenclature. the question of
the so-called Sagv!rian culture should also be raised
here. Currently it seems probable that the LUP pebbleGravettian (Sagv!rian) was an independent facies of
the above la rge cultural entity; however, it is a fairly
enigmatic find assemblage. The latent development of
Middle Palaeolithic pebble-working traditions, even
their preservation over such a long period in spite of
the complete anthropological and culrural basis needs
further proofs. The orig in a11'd movement direction of
the population cannot be described yeL It is not connected to Bodrogkeresztur anyway. Let us state here
again: the cultural name Sagvarian is still used in
some Weste rn European technical I iterature erroneously: it has no relevance for late backed blade industries in the Carpathian Basin (KOZLOWSKI 1979,
42.). The second (medium) phase of the chronologically and typologically well separable three settlement
borizons connected to tbe Gravettian entity is caUed
Sagwir period. The eponym site of this period is the
Sagv!r Upper Palaeolithic settlement. Lts contemporaneity with the Lascaux interstadial has been proved by
V. Gabori-Csank {GABORI-CSANK 1978. I 1.). By
the refinement of archaco-stratigraphy the duration of
this period was extended over lhe embryonal soil horizons containing the two cultural layers of the Sagvar
settlement (in Hungarian loess profiles: hi aud h2) as
welJ as the sedimentatio n of the I m thick loess layer
1n between them, corresponding to the Laugcrie-Lascaux interstadials in western terminology and the in-

termittent Jiule micro-stadial phase (DOBOS lVOR6S 1987, 58). T he older settlements of this geochronological phase can be connected to Sagvar site.
That is, the so-called "pebble gravettian" (Sagvar.

Madaras, Mogyor6sbanya sites) constitme the Sagvarian culture in an archaeological sense. The unity of
the find material serve adequate reason for separating
this small group within the large Gravettian entity.
SagvAr is one of the oldest known and certainly the
most quoted Hungarian Upper Palaeolithic site (Habent sua faro situum: it may take several decades before one site becomes universally known among professionals). Inhabitants of the site were specialised on
hunting horses (37 individuals) and rein.deer ( 126 individuals; VOROS I 982). Its two cultural layers, poleposc dwelling hut, pierced antler ("commanding
stick") and the rich litluc inventory really disLinguisb
Si\gvar among the average of Hungarian Upper Palaeolithic sites (GABORl 1959). Tt was among lhe first
Hungarian UppeT Palaeolithic localities dated by radiocarbon method. The chronological - culturaJ identification of the site was nol made easier by that in
those day as the only comparable (Palaeolithic) C- l4
date was known from the tst.al16sko cave and the
number of known Hungarian Upper PaJaeolithic sites
was only a fraction of those known today. Thus comparison was only possible to far-western. French data
(OABORl-CSANK 1960. 128). For several dcl!ades,
Sagvar was "the" open air Upper Palaeolithic site in
Hungary. The generally spread - and mainly erroneously used - technical term "Sagvarien" can be attributed to its universal notoriety and its special character, recognised already by M. Gabori. M. Gabori and
later V. Gcibori Csank separated the fmds of the southern region in Hungary and applied the term "Sagvar
group" for it. The sites of the Danube bend region
and the north-eastern parts of the country, were named
as the "'Pilismar6t group" (GABORT 1968 186,
GAsORI-CSANK 1984) ). Following this, in the
foreign archaeological literature the '"Sagvarian
group" became a synonym for Upper Palaeolithic cultures in the central pans of the Carpathian Basin denoting summarily and exclusively all Upper Palaeothe Auriglithic features here (with Lhe exception
nacian).
The period " framed" by the two cultural layers of
the Sagvar settlement immediately following the small
interstadial after WUnn 3 maximum till the end of the
Lascau.x period seems in our interpre1ation rather a
geochronologically well defined period (DOBOSIVOROS 1987, 58) which can be associated with (at
least) two different cultures following different toolworking traditions.
the cultural unit formerly called "pebble gravetrian": the name Sagvarian is suggested for this, and
only this unit.

or

97

 LUP I Epigravettian culture suggested by the

Pavlov convention mentioned in Hungarian technical
literature as 1Graverrian with younger blade industry".

The framework of space and time where the

archaeological industry of Bodrogkereszttir-f{enye
could be fit is adequarely circumscribed geegraphically as well as chroAologically. Tiie difficulty of comparing sites and features is hidden in the approach of
specialists consideri11g different typological-technologicaJ or other features they consider important for
defining a cultural phenomenon.
The methods comparing these different features
with variable success do 001 always allow us to quan-rify the identity or difference of sites. The very general
and elusive, but existing identity within the Gravertian
culture or which can serve for the separation of gro1..1ps
within the large entity is almost impossible to express
by mathematical terms, same as the dO"finition of individual features of one- site, Le., J1ow much it can be fit
wilhin the ftamewerk of a given cultural unit.
Notions on the "quality" of Hungarian Upper Pa~
laeolithic settlements is ambivaJent There are some
settlements where the stone Lools are of comparable
quantity and quality to orher Central European.settlements.
The permanent base camps, however, were missing. The Hungarian settlements (Bodrogkeres:ztUr,
Honl, Megyasz6) could be large enough with ample,
typical, good quality tools: but the s.ettlement features,
constructed 0bjects, exotic and artistic material which
were ass0ciared with them were missing.
The bone industry of open-air Upper PaJaeolithio
sites in Hungary was strikingly poor: this could be explained by unfavourable fossilisation conditions supported by the condition of the fauna found on the settlements.
The connections of Bodrogkeresztlir in. this comparison point at che older Pavloviau sites of Central
Europe.
lt is worth ro deal with some paper,s, by M. Otte at
some length because in his sUll1.111aries (OTTE 1981)
the Hungarian sites were also considered to some extent. Bodrogkeresztur was connected by Otte on theoretical basis, possibly its chronological position - to
the Willendorf ll/5-6 settlement level. His Gravettian
chronology (I 986) was based on the re-considered
layer sequence af Willeodorf.
The main artraction in a chronologicai I typological
grouping is the comparison of chronological I typological levels, "facies" rather than "cultures". The
"cultures" of the Gra,vettian circle; probably having a
common root cover probably smali lecaJ groups beeause the ecologieal, geomorphoiogical background
necessary for the introduction of the slightly. mystical
ccchnical term "culture" were identical for all. In other

98

words, in vain we try to separate on the basis 0f technical-typological features small find assemblages
naming them "cultures" when the settlement features,
ll1e ecological (faunistical, botanical etc.) back.ground
is identical over the area of half continent Jn the recent work of Otte, rhe chronological levels were filled
with typological contems.
Bodrogkereszlur preserved its place among the earlies Gravettian sites, the suatotype (leading site) of
'?l~hlch was: Willendorf II 516 (OITE 199 L, 50). This
place was maintained only on the basis of chronological position, based on Vertes's publication.
The analysis 'Of the complete archaeological materlal- the general image of the industry, ilS typo logicaltechnologi'cal features - supported the validity of this
classification. At the same time, in Bodrogkereszt6r as
weU as all the Hungarian MUP sites, the leading finds
of this "staae" were either completely missing (Kostenki-knife) or present in negligible amount (microliths, fl.echette).
Tbe Hungarian sites seem to be mmre strongly related on typological grounds to Stade Ll, the bladep.oint facies.
Considering the connection establishes between the
chronological pos.ition and the fossile direcLOire define.d by Otte valid for Hungarian MUP sites, the Hungarian sites were shifted ty.pologica1Jy compared to the
core are1:15.
Partly, at Bodrogkeresztur the early date is accompanied by young types; partly, Lhe- ratio of side-scrapers and some retouch-type srrengtbened tbe archaic
character of the industry.
The sites contracted in Otte's group 4. were in fact
settlements of contemporary Grav.ettian population reLate.d to each other. The Hungarian .relations of Otrc's
group 8. were certainLy slubbe.red (OTTE 1981 , 134135.); our current state of.knowledge allow the delineauon of a much more detai.led picture on the "rest".
TI1e summarily contracted "other Hungarian fimds"
where Sagvar and Arka were contracted can be dividea into several chronological and cultural levels
(DOBOSI 1998, 131-132)
11he theory Qf a polycentrica1 origin of the most varied and colourful industry of the Upper -Palaeolithic
called in short "G.ravettian"' characterised by ubacke(l
blades" was delineated by the mid-seventies (KOZL0.WSK11979. fig. 13.). The individual, variable Loot
kit of conlernporary sites support this theory. Though
the only real common element is the backed blade,
they appeared. fairly early, during the Arey intcrstadial
m the basis of loqaJ roots in SW France, S-Ttaly or
Central Europe. They were considered descendants of
an early Upper Palaeolithic industry, -which is independent of the Aurignacian (KOZLOWSKI 1979),
The formation area 0f the Pavlovian (MUP), as we
Jmow today, was in Western Slovakia, Southern

.Moravia. Cultural effects. sometimes groups of people

radiated from here towards lhe Rhine and the river
Don. The latter direction. though chronologicaJ data
seem to support the direction of lhe migration I invasion clearly, could nol be a real migration taking into
consideration the vast distances, the shortness of time
and the richness of cultures unfolding on lhe Eastern
territories. This period is fit inlo such wide spatial and
temporal scales lhat it could be hardly summarised
within one culture.
The most recent Upper Palaeolithic summary already divided this period into two parts. In the nortbwestem corner of the great Central European base area
(gravettien ancien d'Europe centrale) appeared a
"Kostenki-episode", limited in time and space with hypothetical connection to the Kostenkian base areas
(DJJNDJLAN- KOZLOWSKl-OlTE 1999, 206-207).
There must have been a certain base population the
culture of which was coloured by the innovations
overtaken or just the other way, which influenced the
local appearances of the strong cultural expansion by
their own traditions.
11ie Carpathian -Sasin is located to rhc south of the
base area and lhe main expansion routes of lhe culture
(mammoth-hut route): the most Likely direction of lhe
immediate or indirect spreading of culrural effects, the
ice-free corridor Lo the south of the permanent ice
sbeel anq to the north of the arch of the Carpathcs.
open till the liver Don and even further on with no
high mountains blocking the way.
Connection with the neighbouring Moravian 1erricories could be realised through the north-western gate
in the Danube valley, connecting Bodrogkcrcszt\ir
with the base areas of the culture. The same route
might serve for the inflow of raw materials of Silesian
(?)origin. This route along the Danube lead along the
Danube-bend towards the foothill regions rich in raw
materials. The foothill region of mountains in the central parts of lhe Carpathian Basin couJd not hold up the
communication between areas. The river valleys runAing from North to South, which were often menlioned could o nly serve as raw material acquisition
routes or traffic with lhe small brotherly communities
(Slovakian sites): these rivers run inside the Carpathian watershed, and the main obslncle of contacts outside the Carpacbes was j ust the presence of the high
mountains.
l'he morphological endowments of the Central
European orography, the effccr of the Carpathian arch
as a geographic factor Oil the direction and pace of the
spreading of different cultures was studied by F.
Djiodjan in an inspiring lecture (DJINDJAN 1992.).
This effect was immediate and strong: the c losed
range of the Eastern Carpathcs set a limit to immediate
eastern connections of BodtogkeresztUr which was

hard to surpass among interstadial circumstances and

practically impassable among stadia! concLitions.
To che east oflhc Carpathian arch, taking advantage
of the favourable conditions of the so-called Ukrainian
Switzerland. Lhe population was practically condensed
and lived constanlly further on, unfolding the Molodova culture in its full richness. In lhe development of
these events, Bodrogkereszrur bad hardly any immediate role.
The possible connections berween Molodova and
Bodrogkercszuir were observed first by Grigorjc\ a
and Anikovich (GRJGORIEYA- ANlK.OVTCH 1990.
89.) Apan from the stone tools, the most striking common eleinenc is a carved rim biconvex discoid object.
similar to Lhe ''moon calendar'' of BodrogkercsztW-.
The presence of this special o bject at both sires.
however is not adequate for proving direct contaclS.
In spite of fonnal analogies, the sites of the middle
phase of the river Prut can be connected more to the
geographically neighbouring regions than to Bodrogkeresz.Wr.
The scenery of migrations in the Stillfried B Paudorf- Maisieres interpleniglacial period ca11 be lt)cated with great probability in the northern ice-free
stripe. Probably this movemenc took place in both directions, though the absolute chronological data
mainly point at movemems from west to ~ast. This
statement is to some extent contradictory lo accumulated evidence on the direction of spreading of faunal
waves. The movemems of different directions took
place not necessarily at the same time horizon: lb.e
migration of surplus population from the Gravettian
base areas towards the West could take a very short
tin1e within the period of the fauna movemenL
Es1imating density of population on lhe basis of
the existing archaeological evidence il seems that during some thousands of years lhc cultures rooted in lhe
Central European base area expanded into flourishing
at the East Side of the Carpathian arch.
The Upper Palaeolithic sites within the interior
river valleys of the Carpathian Bas in opening towards
the plain on foothill-, terrace or plateau geomorphological position represent two chronological phases.
According to our present knowledge. Middle Upper
PalaeoHthic (MUP) s.ites with variable local culruraJ
name!> (on the interior parts of the Carpathian Basin,
Pavlovian or Gravettian Older Blade Industry) delineated the route of outflow from the base area towards
the East (DOBOS! 1998, Fig. 2) Knowing lhe geomorphological endowments of the area, other directions of movement would be hard lo imagine within
the Carpathian Basin. This west-east movement in the
northern stripe of the basin could be almost completely independent of the larger scale and more lasting connection between the large Central European
Upper Palaeol!Lhic cultural units, the route of which

99

lead ro the North of the Carpathian arch along the icefrec Souch Polish corridor. The contradiction hidden in
the dataset on the possible directions of migration and
its chrnnology can hardly be solved on the basis of
finds within the Carpathian Basin. Fauna! waves proceeding from east to west and theoretically closely related .movement of "human waves" are opposed to absolute chronological data which show a trend for getting younger from west to east. Theoretically, the eastwest direction is supported by a more general and
more significant trend in human evolution lasting
much longer than the events tre.ated in this monograph, the east-west advance of Homo sapiens.
The stirring history of ilie Central- and Eastern
Eui:opeanPalaeolithic obviously cannot be stringed on
one chain of events, whatever direcrion, chronology or
ropography. One possible sequence is reconstructed
below whicb do not contradict to data known so far:
The earlfost Upper Palaeolithic with Homo sapiens and the archaeological industry called with inverse logic Aurignacian after its westernmost and
youngest form of appearance arrived along the route
Near East-Balkans, from the south-east into Eastern
Europe.
This population proceeded panly, conserving its
archaeological industry more or less intact, towards
the west; partly, had some influence on. differenl indusrries with Middle Palaeolitbic stone working trad.i
tions
Arriving from the Balkans, to Lhe east of th.e
ridges of the Eastern Carpathes this population could
reach to the North along the river valleys running from
North to Souch. Local Middle Palaeolithic cultures
flourishing there developed here undisturbed for a
long time. FoUowing the impetus o.f Upper Palaeolithic innovations whiob are very difficult to point at
but undoubtedly existing, strong and vigorous Upper
Palaeolithic centres were formed l:letween the rivers
Prut and Dniestr.
The same process possibly took place on the area
excending from the Wachau till the Moravian Basin
which could also serve as an ideal scenery for development.
The U pper Palaeolithlc centre formed here possibly entered into constanl and strong interaetion with
the territories to the North of the Carpathes in Soulbem Poland where Upper Palaeolithic cultures which
had infiltrated formerly from the Balkans, spreading
towards the East were met
The Northern stripe of the Carpathian Basin was
probably populated on tbe margin of the great movements as a precipitation of the surplus population.
Even without knowing the exa.c t origin of the utilised long distane<e raw matedals called surnroarisingly
" erratic flint" Lt js certain that the population of Bod-

100

mgkeres2tilr-Henye used to have some northern connection. The immediate route,,direction of these cont.'lcts, however, were not possible to reconstruct as yet.
Lt was already grnuped by M. Gab?ri among sites with
apparent northern connections (GABORI 1968, 1969).
The base area of the Pavlovian culture (North-West)
and possible directions of raw material pmcurement (
North-) confine well the acrion radius characteristic of
population of. tbe Pavlovian I Older Blade industry.
The g reat aUuv.ial plain adjacent from South to the
stripe of settlements was either not suitable for human
settlemems or contacts have not been documented
from this area as yet.
In the period around the tast co.Id maximum of the
Wilnn period, a new technocompJex emerged in the
interior parts of rhe Carpathian Basin, the Sagvarian
culwre.11 is slillnot known if the embryonal soil horizons of the Sagvarian cultural lay,ers were the products
of smaU interstadials prior or after the Wilrm cold
m aximum. Tbeir absolute chrenological position
(iaround 19 kyear BP) seems to. indicaie that the settlement waves took place after the coJdmaximum; however the 11verage temperature values calculated on the
basis of palaeontological, evidence al lo.w I.be hypothesis that on the inferior parts of lhc Carpathian Basin
tlle cold peak was late compared to general history of
climatic events in Europe. (See: Voros in DOBOSIVOROS 1987 .) So far, only four s.ite.s belonging lo
S'Cigvarian culture are known, the archaeological material is fairly characteristic. On ~be basis of the lrain of
thoughts presented above, it can be rooted in a pebblew.orking Middle Pala.eolithic tradition. The time and
place of the formation of this culture is hardly known.
lt must have been a vigorous and strong culture as
reflected by the multi-layered (Sagvar) and extended
(Mogyor6sbanya) settlement!). The settlements of the
LUP Eplgravetlian cuJture seemed to belong, according to oi1.r present knowledge, dispersed small commurLities all ove.t the Carpathian Basin, which were to
some extent.separated from each other within the great
Gravettiati entity. The pres~rved Gravettia.g way of Ufe
and tool working technology I typology inight bave
been the res1,1lts of optimal adaptation rather than adherence to tradition. Their archaeological material is
not very significant, boch in quantity and quality.
Archaeologically. the system of contact among
Aung!!firu.i sites was fairly '"inbred".
Surveying referenc.es, analogies in archaeological
publications, the geographically set narrow framework
r.emained stable. There is little infom1ation 1-Udden in
~isting analogies of in.dividual Loo ls {not types) further
on to the east or west. lt is difficult to decide if analogies in style were only formal convergence or real genetical affiliation. All the more as these individual
analogies relate, much to our regret, not the " leading
fossils" (e.g., Kostenki-knife) or decisive settlemem

features (e.g., mammoth hut), not to speak of lhe art

objects. There were only few elements found among
the generally accepted criteria for e~i.imating duration
of settlement or populaLion size. Most of the Hungarian sites wete poor in finds and settlement features
with ad-hoe types and scarcity of features. This can be
explained by the lack of the base camps known from
the neighbouring areas (east, west, Dorth). Conclusions can be buih upon, however, existing data only:
the immediate Easte.m connection of the Hungarian
Gravettian culture- cannot be accepted and it seems
better co reside from classifying sites without adequate
data inLO traditional cultural schemes.
The base area of the MUP I Gravettian entity
shows a rich and most varied cultural image from the
Lower Austrian loess regions till the Moravian basin.
which can be classified into several cultural horizons.
This area has been studied for over a century. The relevant technical literature is rich and accessible. Let us
mention here only two recent summaries, a substantial
catalogue edited by, and partly written by J. Svoboda
(SVOBODA 1994) and a short but concise site catalogue from Austria (NEUGEBAUER-MARESCH

L995).
SLarting from the Willendorf-Pavlov arch, we can
survey lhe sites of the period in the Carpathian Basin.

ln Western Slovakia, close co the Pavlovian core area

the MUP-LUP period was begrudgingly rich and varied. The valley of the river Vah (Vag) running by and
large from north ro south is transsecting variable soil
types. Following its high-mountain origins, the middle
phase ofthe river runs across wide loessy hills. This
phase must have had a great ecological potential in the
Late Pleistocene: a great variety of Upper PaJaeolithic
industries and typical, rich settlements are known from
these regions. To mention some oftbem: Tren~ianske
Bohuslavlce with regular, arched edge blade points,
(BARTA 1988) Moravany-Podkovica, closest to the
Pavlovian core areas, classicaJ Kostenki points characterise MUP (VALOOH 1995, 604), a11d in the close vicinity, wi1hin a few kilometres, the leaf-point industry
of Moravany-Dlha and a range of LUP Epigravettian
settlements were excavated (Moravany-Zakovska,
HROMA DA 1995). At the entrance of the next valley
in Pova2.ski lnovec, on the plateau over the river Nitra
we can find another classical site, Nitra-Cerman, also
with Kostenki-points (VALOCH 1995, 604).
From the range of settlements along both sides of
the river I poly, tbose on the left side are better known
(GABORT 1956). The concentration of finds here
embrace within narrow spatial dimensions the complete periods of MUP and LUP. Towards the east,
within the Kosice Basin in the valley of the river
Homad (Hemad)large and special sites of ~e LUP
period are known: Kasov and Cejkov (BANESZ

1996).

The Upper Palaeoltthic tools of the Korolevo multilayered Palaeolithic complex were classified by the
excavators of the site as "ne;v. well perceptible and
expressive centre of early Upper Palaeolithic indus-

tries" (GLADILTN-DEMIDENKO 1989, 177). The

topographical position of Korolevo was stratigraphi~
cally favourable, corresponding to Older Gravettian
traditions. Probably this favourable position accounts
for its "versatile" characLer; in other words. the variable environment of Lhe si:te was most attractive as
landscape, Loo. It is true, however, that the site was
wedged in the north-eastern comer of the Alfflld. Consequently, the direction of possible contacts and the
flow of innovation - infonnation - raw material was
limited. So far we do not know about similar old Up-.
per Palaeolithjc sites in North-East Hungary which
could be connected with this cultural circle; moreover,
we have no authentic contemporary site excavated o n
the regions geographically connected to ~he Transcarpathian area.
It seemed evident that the Korolevo Upper Palaeolithic industry with its unique chronological position,
raw material use and typology should be compared lo
the Upper Palaeolithic settlements lying around Beregovo, known for s:everal decades, some 20 km of the
site Korolevo. 1n the opinion of V. l. Tkacbenko who
evaluated the sjtes around Beregovo, Korolcvo and
the settlements around Beregovo represent two independent Upper Palaeolithic groups in respect of
chronological position and typological spectra
(TKACHENKO 1989, 222). Condensing all sires
around Beregovo on mere topographical grounds
would not be necessarily jus:tified. On t.he basis ofthei'r
s:tratigraphical position, they can form a link in the
chain of MUP system of cultural connections. Ln respect of geographical position, these sites form the
easternmost elements of the large settlement wave running along lhe Carpathian Basin (DOBOS1 1998. 130)
Geographically, the Gravettian sites of the Oas region in Romania belong to this horizon (Boinesti, Somos, Calinesti). According to the authors BlTIRJ-SOCOLAN, these sites were dated to Wiinn 3. Jn their
opinion, though they had a parallel development ro
similar age Eastern Gravettian groups, still they
fortned a special group of sites (facies paniculier), the
analogies of \Vhich could be found on Hungarian and
Czechoslovakian sites (BJTIRI-SOCOLAN 1966,
24). In a more recent publication, the chronological
separatien of the sites could also be accomplished: Turulung and Calinesti represent an older facies of che
Gravettian culture while Boinesti and Somos I-TI
belong to a younger horizon (BJTIRl l969, 53). Lt
would be very important to define these "particularities" exactly as this area used to be a passage towards
the Transylvanian Basin, which is very deficienlly
known.

IOI

At the soulh-eastern corner of Lhe A l:fOl<i. the

Danube-Tisza.-Maros-Temes al luvial plains. are escorted by a range of hiJls in a wide stripe. From th.e
plains upwards, following a "step" of some L400 ave-

rage altitude (Almas, Szorenyi -and Ruszka Mts.) we

hit against the 2500 m high range oflhe Southern Carpathes dissected by .steep river vaUeys. The historical
name for the plains and the foothill region is Banat
(Banat). llS Upper PalaeoJ,itb.ic history is fairly well
known. The chronological position of the rich Aurignacian sites found here seem to cootradict topography.
The only pass_able way of the Aurignacian migration
ffTom South-East (Bulgaria) must have lead along the
Baoat forelands. Compared to lbe more northemly lying Aurignacian sites in Hungary the Bil.nBt sites were
dated at a surprisingly young age (tardiglaciere). 1n the
apinion of Mogosanu: " . .l'aurignacien du .Banat
represente 1111e e1ape tardive ..." (MOGOSANU 1978,
147). On the Late Pleistocene terraces of the rivers
Bega and Ternes, sites around the altitude 250 m a.s.L
~On the terrace of the Temes, 1"11lcova. on the terrace
of the Bega, Romanesti-Dumbravita I.VI. levels)
agreed ohreaologieally with the younger blade induslry facies in Hungary (Lascaux - Romanesci iotcrsLad1al); 1heir find material was poar (MOGOSAWU
1978, fig; 37.). The "quartzite Palaeolithic" (paJeolilic
-p,iartitic) industries living parallel to classical UP industries in the Ba.oat represent a special instance. This
it1dustry was derived by Romanian experts from the
local Mousterian (MOGOSANU 1978, 14.8). Thaugh
so far the authar had no possibility to see lbe material.
personally, the idea can be raised that this type of industry might be connected with the Pebble Gravettian
of the interior parts of the Carpathian Basin, i.e.,
Sagvacian containing pebble tools made according to
Middle Palaeolithic pebble-working traditions. Geoc hronologicaJly, these sites were dated around the
Hetculaae I. interstadial .
At Szeged-Othalom, prolniding the alluvial deposits of the river Tisza with its 90 meter alritude yielded
archaeological material from several periods. ln 1935,
4,5 m deep from the surface remains of mammoth and
Equus were found in tw.o well separated levels which
were remains ef a small Upper Palaeolithic hunters'
camp (BANN BR 1935). Charcoal found here was dete1mined by P. Greguss as Abies alba. llhe arcllaeOlogical material comprised 27 worked objects. Apart
from some blades, it contained a core fragment, nakes
and knapping debr.is. The raw material utiUsed included bydroquartzite, jasper, Prul (?) silex. The archaeological material, as much as we can judge on the
basis of the scarce find material can be fitted well to

Younger Blade industries, i.e., the LUP I Epigravettian

settlement wave according to the Pavlovian convention. This chronological assignment is, to some extent.
contradicted by the prusence of mammoth in the fauna
l02

the stock of whicb was already Iittle in the Carpathlan

Basi h by the end of the Ice Age. The same related to
the s.mall industry of Dunafoldvar, which was even
mor.e poor than Othalom .
The l0ess hills at the southern parts of the DanubeTisza interfluv.ial region are known as the Telecska
hills. The authentic Palaeolithic site of the region is
Madaras - Teglaveto (Brickyards). and possibly some
stray fi nds can a lso be coilllected to this circle (?) to
tlte south of Bacska region. ln spite of troubled find
circumstances (the site used to be an active quany for
brickyard), Madaras is outstanding among the Hungariian sites with its rich and well interpreted scientific
evidence. Madaras-Teglaveto used to serve for a transillonaJ campsite for a smal I group of people hunting
Equus and coUecring .agopus eggs (DOBOS! 1989
63-64). Tbe settlement used to be relatively large but
of low intensity with five hearths coma.in.ing charcoal
of Betula and Pinus. The arcnaeological material can
be c lassified to the so-caJled. ..Pebble Gravettian" or
S~arian facies.. The geological sources of the raw
material, of the small. often corticated tools were in the
Bakony and Mecsek Mts., probably also Northern Erratic flint: according to the categories. used by the Lithatheca, regional and lon.g distance materials (BlRO
1989~ BlR~DOBOSJ 199'1).
ln a strict geograpwcal sense. the territory of
Northern Bosnia does not belong to the Carpathian
Basin. the border-line being Lhe river Sava. Kadar is
on the ri,gbt side of the river but it deserves to be men~
tioned on accoWJ.t af its relation to some Hungarian
sites ( MONTET-WAlTE et al . 1986). Its chtonology,
part of .teal types, Lhe preference of pebple material
and the length/width ratio of fools connect Lhe site to
Sagvar. The pn;senee of the exclusive cool type group,
sb.ouldered blades however separated them fi:om each
\lilier. Montet- White described the material of this sile
as an Epi'g ravettian industry fom1ing a transition between ~1e developecl Gravc;trian of Willeadorf ITJ9 and
Lhe old ~pigravettian of the Adr.iaLic region (MONTET-Wl{TTE el al 1986, 65).
The hunters' campsire Nadap was situated at the
entrance of a valley ~anssecting the Velence Mts. Several meters of loess cover was accumulated here over
slope sedltuems dissected by grit Lines. The good stratigraphical position offered reliable chronological
frames for the heavily disturbed and partly destroyed
settletrtent About the middle of the period between
the Paudorf I Maisieres and the Laugerie interstadials
(classical period of MUP), a group of people hw1ting
F,quidae was settled here for a short time. Complementary game hunted were Rangifer and Bison. The
typical archaeological material (classical Gravettepoints, slightly high end-scrapers, thick burins) was
made of two types of raw material, liver-brown silky.
lustre radfolarite and bluish white patinated flint, cam-

Basabarc loess profile, serving as pedological key

section. The early phase of LUP is represented by
Mogyor6sbanya, assigned to the Scigvanan culture
whife the range of Epigravettian settlements is si-

prising altogether 90% of all artefacts (DOBOS! et al.

1988), offering a special attraccive character for the industry.
The Danube-bend region seemed to possess in all
phases of human history the geographical energy to
attract a dense network of population. During the
MUP and LUP periods of the Upper Palaeolithic, a
string of sites were settled here on a cca. 20 km long
phase along both sides of the river. MUP is represented only sporadicaJJy among the soil layers of the

tuated between Pilismar6t and Dt>ml>s. All three MUP

sites were just sporadical, the Pebble-Gravettian
Mogyor6sbanya in itself, the Younger Blade or Epigravettian industries occurred in a chain of some
10 kms around Pilisrnar6L

6. Summary

Bodrogkereszrur-Henye is a 195 m hill in NorthWestern Hungary at the margin of the Tokaj-perjes

(Presov) mountain range. The Upper Palaeolithic sjte
on top of the hill has been known for several decades.
The find material accumulated mainly in the Hungar-

ian National Museum4 originated from two excavation

campaigns and several official and private field surveys. Deep ploughing and planting of vineyards disturbed the sediments at some places till the base rock,
including the Upper Palaeolithic cultural layer.

HI LLTOP

i::s 90m
UNIT 1/8

I
I

..'.._ I

ea-..!. I
..!....

Ca-;- .L 7

.!.... 1

I
I

I
I

-1-1 - I -

-;-..!.
Ca ~ I~-;-:_ I_!_
-1-1-1-1-t

..!_7 _!_1.:.. -;- ~-;- ..!...1

1-1-1-1-1 1 -1 1 1 -1
1 -1-1-1 1 1-1-1-1-I
1-1-1-1- 1 1 -1-1-1-1
1 -1-1-1-1-1-1-1-1-1

2-7 ~I...!.... 7 ...:_I__!_I

I

I l

J
I

I
I

I t

I I

1x1

1X l

-1-.-1-1 1
1-1-1-1-1-1-1- 1 1-1
1 -1-1-1- 1-

1- I

Ca

1 1

1 - 1 - 1 - 1- 1 - 1 - 1 - 1 - 1- 1
1- 1 - 1 - 1 - l -

I
l

'Ca

..!.7.!...7J...1..!.. X I

I ~7..!.

I 1- I
I -1 I I
I
I -1- I
-1-1-1-1-1-1-1-1-1

.L7..!._7.!._
J..!....t-1-J
1

'x:
I
I

!.1..!....1 ca-!. 1...!... 1

1 -

I
I

I
I
I

UNIT 11111

VERTES ,.E"

I :
I -

t -

1 -1-

1
I
I ....!._

I 1
I -

-----I_!-t

- - -------
-

I -

1-

1 -

1 -1-1--1
1- 1 - 1 - 1 1-

- 1 - 1 - 1- 1-1

~1...!...1..!..-;-...!....1~1

1 1-1-1-1-1-1-1-1-1
1 1- 1 -1-1-

-----
--- - -
1-1- - - -

IT

IT

I -

I
1

xxxxxxx
..!..
..!..
..J..
.J...

...!..

.!...l.!...l.!...T.2.1...!....T
..!.. Ca I...!.. I 1 1 ..!..I
I I I -GaI -1-1 1-1
..!....1 _!_-Cl!_ T ..'._- I I
, 1 1 1 -Ca1 .!...-;-!..1 .!.... T .!... T .!...
-- 1 -1-1- 1
1~-1

1 ~ :_ T .!.....
I

-1-

ttJ--1!.I
'-'':..-.--.....

Fig. 48 Hmye hilltop stratigraphy N-S direction

4 A small part of the material is in Hennan Ott6 Museum, Miskolc

103

......
0

.b.

NW

HILLTOP

SE
., 230m

UN IT II / 1

YEATES .,D"

UN IT 117

Wll Will 11114

recent su rtace

recent soil

[ ~ ~- _: ] loess
-

I
I

'

ea'I

Ca

Ca,

:,

I
I

!_

1-1 - 1 - 1 - 1 -1-1 1-1- 1

._,_,_,_, _
_,_,_,_,_,

xx 'x x' x x x x x

1-1- 1-1-1-

-1-1-1-1-1

~~I t
I - ~
IV-1
~-m-

~
1
1+.... +
1

- - ...

..

...

'

'

I-!.. ~/\AIX
I

-I

'\

-I

!_

-1--l-I

1 -1

I - I..!....-'"- I _ - 1 -

-1-1

I -

1x -

-.~
1

-I

_,- I

.6

-I

I -I - I
_,._

11...W'l I - I I - I
2...7 ..!.. _ ,
,c+r
1-1 I -
I -r-"1
-X -7"'
1-- 1 --;--~-
1
1
' .L I , .:_ -;- .:_X - I 2-, _'(:)I
- I- I
+
_,_!_r;:"t-'8
-

---1-1-1-J
1 -1-1-1-1-1-1-1-1-1
1 - l - 1 - 1 - 1_,_,_, -1-1
1 - 1 - 1 - 1 - J-1-1-1-1-1
1-1- 1 -1-11 -1-1-1-1
1 -1-1 -1' I I I I I I I I I
I I I
I I
I
I I I I
I
'
I I I
I I I I I

xxxxxxxxx
11
I

ea

c::J
xxxx

embryonic soil
CaCOn

andesi te
cultural layer.
finds bumed

1' 1' ' I , 11

t It' 11, 1 t' 1': I 1 1'1

j~I
~tt..:..JT + .:
+
t

1' ''I'

l1 ~_ L]

+ + + '

1 I 1
I I, 1 I 1, I 11
I t 1 1 11 I
11'1'1 1, I I
I I I I I, 1,
I I I 11 I I I .

I II I I I I 1 I

I1I
It 1
I1I
'1 I
I 11

II I

I I

I 1l

'I I

1
11
1

1I

11

j II

I1

I I

1I

'

I I

I I

I I
1
I

I:
I

1I

I 1I
I
1I 1 I I I
I t 1I
I
I I 1
If I
11

I:

I 1

I I: I I 11

r , 11 1 r 11
1
1 1, I I 11
I I I 11 I 1 II

I I I 1 1 I I 1 I I
I I: 1

I I 11

I I
I

I:
I

I I
I

I .I

11
1

1
I

I I
I I

I I
11

I I

11

I 1

I I

Fig. 49 Hcnye hillldp stratigraphy NW-s.E direction

dark yellow loess

I

'1'

I 11
1, I
I I I I
I 1 11:

On the basis of the stratigraphy obtained during excavation experiences

the Wurm period reliefof the hill must have been
more uneven with higher relief energy

tributaries). rmmediate eastem-soulh-easrem connection with the regions outside the Carpathian basin are
hard to postulate.
Leading finds of the culture like Kostenki knives,

the Late Pleistocene loess cover of variable thick-

permanent above ground constructions/mammolh huts

ness smoothed away these uneven surfaces

the subsequent intensive climatic events re-arranged the acrual surface to variable extent (Fig. 48-49).

were missing from Bodrogk.ercszrur and all the other

contemporary Hungarian settlements. Art was represented in the form of incised rim polished discs. WeUk:nown and spectacular "fossile directoire" linds were
always empbatically considered in the cultural classification of sites. They are, in fact, adequate but not indispensable conditions for assigning siLes into cultures.
Supposing that the "people with mammoth hucs" were
a separate, somewhat younger episode of the Gravetlian entity (DJINDJAN- KOZLOWSKI-OTTE 1999-.
207), the Hungarian sites could well be a more general,
less typical and poorer represenratjves of the same culture. The geographical and chronological framework
covered by Lhe mammoth-huts could be ro large for being filled out by one culture in the tiaditional arcbaeological sense. Tbe possibility of other explanations i.e., the construction of mammoth huts were specific
not for the culture but the climate s hould also be con
sidcred The hypothesis can be risked that the distribu_tion stripe of mammoth huts coincided with the grassy
tundra vegetation zone witb no arboreal vegetation
(JA.RAINE 1997, 9). The Carpathian Basrn was situa-

The archaeological material is uniform. It is in good

accordance with Hungarian and Central-Eastern European analogies, che radiocarbon results and the fauna.
On the basis of areal extent and the quantity of the
archaeological and palaeontological material, the Bodrogkereszrur-Henye site was outstandfog among the
Hungarian Upper Palaeolithic sites. The finds allow us
to postulate long lasting permanent and I or repeated
st.ay of the same cultural entity. The criteria on the basis of which permanent settlements were separated in
the Moravian Basin or along the river Don had been
missing in the complete Hungarian Palaeolithic evidence; still, Bodrogkeresztiir-Renye was undoubtedly
one of the most varied and complete Upper Palaeolithic sites in Hungary. The centre of habitation could
be observed at several point of the plateau. Topographical "slip" may indicate chronological differences
but the unity of the archaeological evidence clearly indicate that the complete find assemblage used to be.long to the same archaeological culture.
Bodrogkereszttir-Henye was an important s ite of
the Older Upper Palaeolithic I Gravettian population
wave which could be followed from Willendorf and
the Pavlovian base areas till Molodova
[n che large sense, it was a member of the Willendorf-Pavlov-Avdeevo-Kostenki Entity. or, from anoLher point of view, into the frames of the (Central European) Middle Upper PalaeoLithic (MUP). The identity
is understood more in the similarity of the ecological
relations of the interpleniglacial or younger open-air
settlements than specific (typological or other) analogies. For this, the presented framework is far too large
in chronological as well as geographical sense.
Immediate typological connections were found
within the interior margins of the Carpathian Basin as
well as southern parts of Moravia. Bodrogkereszrur
used to exist to the east of the core area of the interpleniglacial backed blade industries, to the south
of the hypothetical main spreading route of this culture.
The most likely route for the population of the interior areas of the Carpathian Basin is considered to be
across the Danube valley (with adjacent hills and midmountain regions). Other routes might have also
played minor roles, running towards the interior parts
of the basin from the main communication road or the
watershed (in the ftrst place, the river Hernad and its

ted to the souch of d1is stripe, with its specific, more

protected climate already in the steppe zone.
As we are getting closer in time, the amount of
info rmation is getting objectively more numerous. Our
perception, however, multipLies it subjectively. It
seems that the resultant tendencies do not favour integration in the Upp~r Palaeolithic. Tn the period of sensible separorion we can count on a big mosaic of local
communities built on each other replacing the traditional large cultural units of the Lower or Middle Palaeolithic period. The given territories. chronological
frames could overlap to some extt:nt.
The summary or part of features which could be investigated with archaeological methods included:
selection of the place of settlement (close to rivers, on multi-purpose strategical points
large system of connections (e.g., procurement of
" long distance" raw materials)
distribution of certain obje~t carrying probably
the same conceptual contents (polished serpentinite
discs)
type spectrum and analogies of types.
- These features set a common framework to the
s ites of this period. T he above reconsLrUcLion of the
Older Blade lndustry I Pavlovian/MUP sites (position
of the site Bodrogkeresztlir and its circle) within the
Hungarian Upper Palaeolithic could be realised accordingly, on the basis of our present state of knowledge.
105

Radiocarbon dates from Bodrogk eresztlir- Henye:

GXO 195 (charcoal, cultural laye-r}
deb 2S55 (charcoaJ, sediment
with arch. finds)
deb.3381 (booe, culturalJayer)
Hv 12986 (bone, cultural layer)

28 700 3000

According to the archaeological, palaeontological,

palaeobotanical and pecfological centributions, the

26318 365
18 575 208
10 630 270

26-28 ka BP (non-calibrated) chronologicaJ positi on is nccepfable.

Upper Palaeolithic (H1111gary) Synoptical table

J..-y~r

Period SITES
/having C 14 d~res

Soillloess

Traditional

July mean

sequence

inters tadfals

teruperatnrc

lstall6sko VI,
Janko:vicb 5
13

Arkafupper,
upper layer

15

Pilisszanl611

16

o.c

Hungary

BP

AllcrBd
15.7
14.3

P~kO

e
s
s

C limatoraunal PHASE';S
named after
main sites

B
A
l

13,6

1"2,1

upper 5/a

CULTURES

E
G

p
J

G
R
A

Pilismar6t. Budapcsl-

Csillnghegy

17
l&

A.

Esztergom.

Stigwir upper a tlt11ral

layer
Jrisefelslkemgyorgy 1-11
Arkal /ov,rer cultural layer

A
T
A

Madaras

Mogyorosbdnya

h2 sail

Stigvr:ir lawer cu/t11ra/

Duna11jwiros

layer

La.<;eaux
L
l

e
s
s

s
19

"'

T6pio1iily
soil
I

14,3

16,2_

Laugerie

s
s

z
A

A
G

v v
A

E.
T
T

R
E

J
A

N
T

20

22

Remete Lower,
Pilisszant6 1 middle laye:a'S
Pcsko middle layers

23

eilissziln.t6 I lower layers

21

L
0
E

6
Tursac?

Bodrogkcrei.'Zttl r,
Mcgyaszo
Pilsp!ikhatvnn,
Ront/ Paras11a

17

p
A

N
N

Hidasncmeti
fst3116sko JV-V; Nndap
28

l5,6

s
s

Me.11dtt

Denekai:np?

16,3

17,4

M
E
D

Vpper
soil J

v
0

v
I

Saj6szenfp~ter

3 1- 32

17,4

lstillosko upper layers

lSTALL6SK6

AURIGNACIEN II

'v.h.h olschewapoint
> 35

Istall6sk0 lower layer

Mend!!

Upper
soil 2

Hengelo'?

AURIGNACTEN I
W)Lh Sp(itred

base point

106

REFERENCES

BARTA 1965
BARTA-BANESZ 1981

BA.Nusz 1962
BANESZ 1989
BAll.IESZ 1996
BANESZ et al. 1992.

BELLA 1923

BiRO 1984

J. Bai1a: Slovensko v starsej a strednej dobe kameoncj. Bralislava, 1965.

J. Barta-L. Banesz: Tne Palaeolithic and Mesolithic. Archaeologic.al Research in Slovakia. Xth lntemational Congress of Prehistoric and Protohistoric
Sciences. Nitra, 1981. 11-30.
L. Banesz: Vyskum pelolitickej stanice Cejkov i v roku 1961. Arch. Rozhl.
XIV/ 1962.6.
L. Banesz: Problems of the upper palaeolithic in the north-western part of the
Carpathian Basin. Anthropologie XXVIlJ2- 3.
L.Banesz: Predmety umeleckcho prejavu z paJeolitickej stanice pri Cejkov a
Kasove. Slovenska Archeologia 44. 7-24
L. Banesz-J. Hromada-R. Desbrosse--1. Margcrand-J. Kozfowski-lC
Sobcz.yk-M. Pawlikowski: Le site de plein air du paleolithique superieur de:
~ov 1 en Slovaquie orientale. Erode preliminaire d ' une structure spaliale
des outillages epigravettiens en obsidicnne. Slovenska Archeol6gia XL- L

5- 28.
BelJa L.: A bodro_gkeresztUri aeneolitbkori temetl>. Az orszagos Magyar
Regesuti TArsulat Evkonyve I. evfolyam, 1920-22. Budapest, 1923. 7- 18.
K. T. Bfr6: Distribution of obsidian from the Carpathian sources on Central
European palaeolithic and mesolithic sites. Acta Archaeologica Carpathica.

xxm.

BJR6-POZSGAI 1982

Bir6 K. T.-Pozsgai I.: Obszidian hidraci6s kergenck vizsgalata kormeghatarozas ceJjab6l Archaeologiai Ertesito 109. 124-132.
BlRO- PALOSI 1983
K. T. Bir6--M. Palosi: Sources oflithic raw materials for chipped artefacts in
Hungary. M. All. FOldtani lntezet evi jelentese 1983.407-435.
BfRO- DOBOSt 1991
K. T. BirO-V. T. Dobosi: Lithotheca, Comparative Raw Material Colleccion of
the Hungarian National Museum. Budapest, 1991.
BJTIRl 1969
M. Bitiri: Gravetianul in Nord-VescuJ Romanici. SClV 20. 1969. 515- 531 .
BRANDTNER 1954-1955
F. Brandtner: Kamcgg. eine Freiland-station des spliteren Palaolithikums in
Ni~denisterreich. Miueilungen der Prtihisrorischen Kommission der Osterreichiscben Akademie derWissenscbaften VU. 1954- 1955.
CARCTUMARU-CHTRICA
M. Carciumaru-V. Chirica: Decouvertes d'art Paleolithiquc sur le Territoire
1987
de la Roumanie. Bibliotheca Archaeologica lassiensis CL La gene.se et !' evolution des cultures paleolithiques sur le Lerritoirc de la Rournnnie.University
"Al.LCuza" Tasi.
V. Chirica: La cbronologie relative el absolutedes habitats aurignaciens er
CHIRICA 1986
gravettiens de la Roumanie. The pleistocene Perspective 1. LondonSouthampton, 198.6.
V. Chirica: The gravettian in the east of the Romanian Carpathians. "Al. I.
Cf-DRICA 1989
Cuza" Universicy of Jassy.
V.Chirica: Le gravettien en Roumanie. Le paleoUthique le neolithique de la
Cl-ITR.ICA 1991
Roumanie eo contexte europeen. BAI 4. Iasl, 1991. 7-24.
V. Cbirica-L Borziac-N. Chetram: Gisemeots du paleolithique superieur
CHIRICA-BORZIACancien entre le Dniestr et la Tissa. BAJ V.
CHETRAU 1996
Cscmgradine Balogh E.: A sagvari kesoi felsopaleo lit lelohely tipol6giaj es
CSONGRADI 1997
statisZlikai vizsgilata. Folia Arch. 46.17-44.
CSONGRADI- DOBOSI 1995 E. Csongradi-Balogh- V. T. Dobosi: Palaeolithic Settlement traces near
PUspokbatvan. rolia Arch. 44. 37- 59
P.-Y. Demars: Proposition pour une nouvelle liste typologique des oulillages
DEMARS 1990
lithiques du paJeolithique supericur. Paleo 2. Decembre 1990. 191-202.
P.-Y. Demars-P. Laurent: Types d"outils lithiques du paleolithique s uperieur
DEMARS-LAURENT 1989
en Europe. Cahiers du Quatemaire No 14. 88-381-790.
F. Djindjan: Vais de Communications entre Europe Centrale et Europe OcciDJINOJAN 1992
dentale: Occupations d.altitude et limitation climatiques. Adaptations au
107

Milieu montagnard au Paleolithique et au Mesolithique. Tremo-Viote 5-11.

Octobre 1992.
DJINDTIAN-KOZLOWSKI- F. Djindjan- J. Kozfowski-M. Otte: Le paleolitb:ique superieur en Europe.
OTTE 1999
Aanand Collin 1999.

DOBOST 1967

T. Dobosi V.: Uj felsopaJeolit telep az Alfoldon. Arcbaeologiai Ertesfto 94.

DOBOSJ 19 75
DOBOSI 1975 a

T. Dobosi V.: Adatok a BodrogvOlgy oskonihoz. Folia Arch. 25. 9-2~.

T. Dobosi V.: MagyarQrszag os- es kC>zepsokori lelobcly-katasztere. Archaeologiai E~itO I02. 64-76.
T. Dobosi V.: Asatas TarpaMarki-tauyan. Communicationes ArchHung.
L983.
T. Dobosi V.: Madaras-l"eglaveto fels8paleoli1 telep regeszeti feldolgozasa.
Cumania 11. 45-58.
V. T. Dobosi: Discontinuity in the Up.per Paleolithic of Hungary. In: Les bassins du Rhin et du Danube au Pahfolithique superiur: environement' habitat el
systemes d'eebange. ERAUL} 43, Liege, 1991. 17-25.
V. T. Dobosi: A new Upper Palaeolithic Site at Mogyor6sbanya Communications ArchHung 1992. 5~17.
V. T. Dobosi: Upper Palaeolithic in the Danube Bend. Palaeolithic in the
Middle Danube Region. Bmo, 1996. 25-37.
Y. T. Dobosi: Eastern or Central? Gravettianconnections/Hungary. Vostochnyj
gravett. Mos~kva 1998. 125-134.
V. T. Dobosi-B. Jungbert-A. Ringer-I. Voros: Palaeolithic settlement in
Nadap. Folia Arch 39. 14-40.
V. T. Dobosi-L Gacrer: Palaeolithic tools made of rock crystal and their preliminary fluid inclusion mvesti.gationi Folia Arch. 45.31-50.
T. Dobosi V~Homola l~ tipol6giai-technikai megfigyelesek paltintott
koeszkoz0kon. Folia Aroh.. 40. Budapest, 19.89.
V. T. Dobosi-E. K.Ovecses-Varga: Upper Palaeolithic Sile at Esztergom-

184-193.

DOBOSl 1983

DOBOSI 1989
DOBOSI 1991
DOBOSI L992
DOBOSJ 1996
DOBOSI I 996a
DOBOSJ et aJ I 988
DOBOSl-GATfER 1996
DOBOSJ- HOMOLA 1989
DOBOSl-KOVECSES-

VARGA 1991
DOBOSI-SlMAN 1996

DOBOSl-VOROS 1987
DROBNJEWICZ-GINTERKOZLOWSKI 19.92

FELGENHAUER. 1951
FELGENHAUER 1980
GABORI 1968
GABORI 1969

Gymgyalag. Acta.ArchHung ASH 43. 233-255.

V. T. Dobosi-K. SinlfilL: New Upper Palaeolithic sile at Megyasz6Szelesteto. Communicationes ArcllHung 1996. 5-2V. T. Dobosi-1 Voros: 1'he Pilisstlnt6 L Rock-shelter. Revision. Folia Arcb
38. 1987. 7-64.
B. Drobniewicz- B. Ginter-Ji TC Kozlowski: The Gravettian sequence.
Temnata Cave. Excavations in Karlukovo Karsl Area, Bulgaria. Jagellonian
University Press. Cracow. 1992.
F. Felgenhauer: Aggsbach Ein Fundplatz des spaten PalaLithlku~s in Niederosterreich. Mitteilungen der Prabistorischcn Kommission der Osterreich.ischen Akadem ie der Wissenschaften. V. Band, N r. 6. 1951, 160-266.
F. Felgenhauer: Bin jungpalliolithisches Steinschlageratelier aas SLillfrit!d an
der March, Niederosterreicn z1.1r Herstellungstecbnik von Mik:rogravettespitzen. Forscbungenin Stillfrjed .Bd. 4, 1980, S 7-40, Taf, 1-11.
M. Gabori: Archaeology and prepistory of U1e Hungarian "Loess-Palaeolithic". La Prehistoire 1968, 183-189.
M. Gabori: R"egionale Verbreitng pal!iolithischer Kulruren Ungams. Acea
ArchRung2l. L5~165.

GAilORI 1989

M. Gabori: Die letztc Phase des Pat~olith ikums in Ungam. Quartfu' 39/40,

GABORl-CSANK 1978

V. Gabori-Csank: Une escillation climatique a la fin du Wilrm en Hongrie.

Acta ArcbHung ASff 30. 3- 11.
G6.bori-Csank, V.: A felsopaleolitikum nyoma Budapesten. Budapest
R egisegei 25.7-14.
V. N. Gladilin-Yu. E. Demideriko: Upper Palaeolithic stone tool complexes
from Korolevo. Anlhro1>ologie XXVU/2-3 . 143-178.

\3.1-14.0.
GABORI-CSANK 1984
GLADTLIN-DEMTDENKO

1989

108

GRIGORIEV 1993

G. P. Grigoriev: The Kostenk.i-Avdeevo Archaeological Culture and the Willendorf-Pavlov-Kosrenki-Avdeevo cultural unity. In: From Kostenki to Clovis.
New York, 1993. 51-65.
GRJGORJEVA-ANLKOVITCH G. V. Grigorieva-M. V. Anikovitch: Au subjet des Hens culturels entre cerrains tndusrries du paleolithique superieur d"Hongrie et de !'Europe d'est. ln:
1990
Le paleolithique et le neolithique de la Roumanie en contexte europeen. Bibliotheca Archaeologica Jassiensis IV. lasi, 1990. 72-89.
GYARMATI 1972
P. Gyarmati: Structural-volcan-tectonic sketch of the Tokaj Moumains. 1972.
(manuscript)
GYARMATl 1983
Gyarruari P.: A Tarpa-Ma.rki tanyan el5.keril.lt pattintott k6cszkozok nyersanyaganak vizsgaJata. Commmuniatiooes ArchHung 1983. l 7.
HAESAERTS 1990
P. Haesaerts: Nouvclles recherches au gisement de Willendorf (Basse
Autriche). Bull Inst. Royal Sciences nat. de Belgique t. 60. (Sciences de la
terre) 203-218.
HAGEN 1972
A. Hagen: Man and Nature. Reflections on Culture and Ecology. Norwegfan
Archaeological Review. Vol.5.No.I. 1- 22.
HAHN 1972
J. Hahn: Eine Gravettien-lndustrie von Krems- Hundssteig (Niederosterreich).
HOMO 72. Heft 112, 81-89.
W. J-leiorich: Palaeolithforschang in Ostecreich - ein RUckblick aus die lc tzten
HBJNRlCH 1974-1975
25 Jahre. MitteiJungen der osrerreichiscbe Arbeitsgemeinschaft fi.ir Ur- und
Frilhgeschlchte 25. 1-40.
HELLEBRANDT-SIMAN
Hellebrandt M. -Siman K.: A Herman Ott6 Mtizeum 1979. evi asatasai es
leletmentesei. Herman Otto MUz:eum Evkonyve 19. 87-97.
1980
HONEA 1994
K. Honea: Tranzitii culturaJe in paleoliticul superior timpurui si cronostratigrafia de la Mitoc-Malu Galben. Arheologia Moldovei 17. 117-146.
H ROMADA-KOZLOWSKI. J. Hromada-J. Kozlowski: Complex of Upper Palaeolithic sites near Mora(eL al.) 1995
vany. Western Slovalda.Krakow, 1995.
Jaoossy D.: Tarcal-Citrombanya 6sk5kori lelohely gerinces faunaja. Folia
JANOSSY 1975
Arch .25. 1975.26.
Jaraine Koml6di M.: A legut6bbi azaz holocen beerdosOdes n6rarorteneteroL
JARACNE 1997
Botanilcai Kijz.lemenyek 84. 3-15.
B. Klima: DolnJ V~tonice. Vyzkum mboi~te lovcu mamutU v letech
KLTMA 1963
1947- 1952. Praha 1963.
J. Kozlowski: The origin oflithic raw materials used in the Palaeolithic of the
KOZLOWSKI 1970
Carpathian countries. ffird Symposium of Carpathian Archaeology, 197.
3- 24.
J. KoZ:lowski: Le paleolithlque superieur en Pologne. Anthropologie(Paris) 87,
KOZLOWSKI 1983
49-82.
J. Kozlowski: S igni:ficance of the inter-site and intra- site variability in the upKOZLOWSKI 1984
per paleol.ithic. 317-348. Francisco Jorda obhua scripta praehlstorica Salmantlcae 1984.
J. Kozlowski: The danubian graveUian as seen from the northern perspective.
KOZLOWSKl 1996
lo: PaJeolithic in the Middle- Danube Region. Ed: Svoboda, J. Bmo, 1996.
11-22.
KOZLOWSKl-KOZl.OWSKT J. Koz!owski-S. Kozlowski: Upper palaeolithic and Mesolithic in Europe.
Taxonomy and Palaeohfatory. Polska Akademia Nauk-Odzial W Krakowie
1979
PraceKomisji Archeologicznej. Nr. 18. 1979.
J. Kozfowski-K. Sobczyk: The Upper Palaeolilhic Site Krak6w-Spadzista
KOZLOWSKl-SOBCZVK
Street C 2. Warszawa-Krak6w, 1987. Universitas Iag,ellonica Acta Sci. Liu.
1987
847.
Krolopp E.: Tarcal-Citrombanya 6sk&kori lelohely mollusca-faunaja. Folia
KROLOPP 1975
Arch 25, 27-28.MNM 1965 MNM Adattar V. 94 ( 1965).
Matyas E.: A ratkai fels6szarmata 6desvfzi medence fOldcani es teleptani
MATYAS 1966
viszonyai. Foldtani Koz!Ony 96. 1966. 27-41
F. Mogosanu: PaJeoliticul din Banat. Bucuresti 1978.
MOGOSANU 1978

109

A. Montet-White (Ed.): The epigravetlian site of Grubgrabeo Lower Austria:

the l 986 and 1997 excavations. ERA UL 40, Liege 1990.
MONTET-WIITTEA. Montet- White-H. Laville-A. M. Lezine: Le paleolithique en Bosnie
LAV ILLE-LEZINE 1986
du Nord, Chronologie, Environment, et Prehistoire. L' Anthrop.ologie 90,
1986.229-288.
MUSSL-ROEBROEKS 1996 M. Mussi-W. Roebroeks: The Big Mosaic. Current Anthropology 37
697-699.
NEUGEBAUER-MARESCH
en. Neugebauer-Maresch. Altesteimzeit im Osten Osterreichs. Wissenschaftliche Scbriftenreihe Niederosterreich. Wien.
OLIVA 1989
M. Oliva: Vyz.kum mladopaleoliticke stanice u Milovic, Ok.r. Breclav
vysledky tri uplynulych sez6n. (The Excavation of the Milovice Upper Palaeolithic Site Three- years Results.) Soucasny stav perspektivy vjzkumu
kvarteru v CSSR.. (The Present State and Perspect.ives of Quaternary Research
in Czechoslovakia. Bmo 1989. I 02-1J6.
OTTE 1980
M, Otte: .Les groupes Gravettiens en Europe Centrale. In: Colloque International L' Aurignacien et le Gravetticn (Perigordien) dans leur cadre ecologique. Nitta, 1980.
OITE 1981
M. One: Le Gravenien en Europe Centrale. Disstert.aLion Archaeologicae Gandenses De Tempel Brugge, 198 J.
OTTE 1990
M. Otte: Revision de la sequennce du Paleolithique Superieur de Willendorf
(Autriche). Bulletin de L' lnstirut Royal des Sciences Naturelles de Belgique
60. 1990.
OTTE-KEELEY 1990
M. Otte--L. H. Keeley: The Impact of Regionalism on Palaeolithic Studies.
Current Anthropology Volume 31 . Number 5. December 1990.
OTTE ec al. 1996
M. Otte-P. Noiret- V. Cbirica-1. Borziak: Rythme evoluLif du .graveltien
0riental. UlSPP-CoUoquium XILF<.uLi 1 '1996. 213- 226.
PAUNESCU 1993
A. Paunescu: R.ipiceni-lzvor Paleolilhic si mezolithic. Biblioteca de arheologie Lll. Bucuresti, 1993.
PECSJ 1992
M. Pecsi: Locss of lhe last gl.acilion. Atlas of Palaeoclimates and palaocnvironments of the Northem Hemisphere. Budapest-Stuttgart, 1992.
PlNCzES 1960
Pinc7.is Z.: A Zemplen hegyseg deli resuoek termeszcti fOldrajza. Manuscript Budapest, 1960.
Z. Pinczes: Periglacial planation surface and sediments in the Hungarian
PINCzES 1977
Mountains. Fl:lldrajzi KOzlomenyek2S/1977. 45-55.
RINGER 1988
A. Ringer: Possible correlations between loess and cave deposit srrarigraphies
for the upper PleisLocene in Hungary. Tn: Paleogeography of Carpathian Regions. Budapest, 1988. 65-5.
RINGER 1993
Ringer A.: Eszak.kelet-magyarorszagi geomorfol6giai szintek es regeszeti
adalaik.. Miskolc 1993. Manuscript
ROGACHEV-ANfKOVlCH A. N. Rogachev- M. V Anikovich: Pozdnyij paleolit Russkoj ravnyinO i
1984
Krima. in: Paleolit SzSzSzR. Boriszkovszkij, I.. J. (Eel). Moszkva, 1984.
162-27l.
ROGACHEV-SINITZYN J 982A. N. Rogacbev-A. A Sinitzyn: Kostenki 15 (Gorodtzovskai.a stoianka) ln:
Palaeolithic of the Kostenki-Borshevo area on the river Don 1879- 1979. N. D.
Praslov - A. N. Rogachev,(Eds.) Leningrad, I 1982. 162-170.
SLMAN 1988
Si.man K.: Teteptilesi fonn::ik Borsod-Abauj- Zemplen megye tertileten a paleolithikum idejen. Hem1an Ort6 MU.ZCurn Evkl:lnyve XXV-:XXY.l. 55-67.
SlMAN 1989
K. Simao: Northern flint in the Hungarian Palaeolithic. In: "Northern" (Erratic
and Jurassic) flint of south Polish origin in the Upper Palaeolithic of Central
Europe. Krakow, 1989. 87-94.
SlMAN l 989a
K. Simao: Hidasnemeti - Upper Palaeolithic Site in the Hernad-valley (Northeast Hungary). Act.a Archaeologica Carpatbica 28/1 989. 5-24.
SIMAN 1990
K. Sim3.n.: Populatian fluctuation in the Carpathian Basin from 50 ro 15 thousand years BP. Acta AcchHung ASH42. 13-9.
MONTET-WHITE 1990

110

SOFFER 1993

SOMOGYI 1982

0. Soffer: Upper Palaeolithic Adaptations in Central and Eastern Europe and

Man - Mammoth Interactions. In: from Kostenki to Clovis. Upper Palaeolithic-Paleo-Indian Adaptations. New York, l 993 . 31-49.
Somogyi S.: A K.arpat-medence termeszeti viszonyainak valtozasai a bonfog-

lahis elott. Bevezetes a magyar ost0rtene1 kutatasainak forrasaiba. lV. BudaSTIEBER 1975
SUMEGl 1996
SVOBODA 1991
SVOBODA 1994

pest. 1982. 76-81.

Stfober J.: TarcaJ-Citrombanya oskokori le16hely faszeneinek anthrakot6miai
vizsgalata. Folia Arch 25. 29.
Silmegi P.: Az EK-mngyarorszagi lOszteri.iletek osszehasonlit6 osklim yezeti
rekonstrukci6ja es retegr.ani ertekelcse. Kand.dissz.tezisek 1996.
J. Svoboda (Ed.) Dolni Y~stonioe JI. Western SlopeERAUL 54. Liege, 1991 .
J. Svoboda (a kol). PaJeolit Moravy a SJez.ka. The Dolni V~tonice studies. L

Bmo, L994.

SVOBODA 1996
T KACHENK.O 1989
USIK 1989

VALO CH 1986-1987
VALOCH 1995
VALOCH 1996
VARGA 199 1
V ERTES 1964-1965

VERTES 1965
VERTES 1966
vERTES 1968

J. Svoboda: The pavlovian: typology and behavior - ln: Paleolithic in the

Middle Danube area. Svoboda, J. (Ed.) Bmo, 1996. 283- 301.
V. L Tkachenko: T'he Beregovo group of Upper Palaeolithic Sites in Transcnrpath.ia. Anthropologie 27. 2 13-222.
V. l. Usik: Korolevo- Transition from Lower to Upper Palaeolithic according
to reconstruction data. Anthropologie XXVII/2- 3. 179-2 12.
K . Valoch: Les questions du Pavlovien. Antiquites Nationales No. 18/ 19.
55-62.
K. VaJocb: Territoires d'implantation., contacts et diffusion des societes du
paleolitbique superiem dans I' ancieone Tchecoslovaquie. L' AnLbropologie 99.
593-608.
K. Yaloch: L' origin du gravettien de l' Europe Centrale. UlS PP Colloquiwn
XU. Forti, 1996.
I. Varga: Mineralogical analysis of the lithic material from the palaeolithic site
of Esztergom-Oyurgyalag. Ac ta Archl-Iung 43. I991. 267- 269.
L. Vertes: Da JungpaHiolithikum von Arka in Nord- Ungaro. Quartar 15116.
79-132.
L. Yertes: "Lunar Calendar'' from the Hungarian Upper Paleolitbic. Science
149. 1965. 855-56.
L. Vertes: The Upper Palaeolithic Site on Mt Henye at Bodrogkeresznir. Act.a
ArchHung ASH 18. 3- 14.
L. Vertes: Rates evolution in Palaeolithic technology. Acta ArchHung ASH
20. 3-19.

111

''L

'I

Hunted mammals from the Gravettian campsite Bodrogkereszrur-Henye (Ne Hungary)

Istv:in VOROS

L Introduction
The Henye hill is situated between the so-called
H cgyalja region lying along the Bodrog river at the SE
margin of the Zemplen Mt.s. at 20~300 m altitude
and the Tokaj (Kopasz) Mt, 5 I 5 m a.s.l.near Bodrogkeresz tur village. ln 1963, Lasz l6 VERTES excavated
the Upper Palaeolithic settlement lying on the flat cop
of Henye hill. Tn 1982, Viola Dobosi continued work
by sondage. In the two excavation seasons, altogether
423 m2 was unearthed :
165 m2
1963, YERTES
1982, DOBOS!
258 m2
resulting in altogether 3589 pieces of stone, 1803 animal bones and 154 charcoaJ pieces
Animal bone remains of the 1963. excavation season were transported into the Palaeovertebrate Collecci on of the Hungarian Geological lostitute (Since
1994, the collection belongs to Lhe Geological Museum within the Hungarian Geological lnstitute. Inventory numbers: Y.10974, 10976-77, 10979-l 0990,

There were remains of six big game species found

at the Renye-hill settlement. Among them were five
herbivores: Equus sp.. Alces alces {L ). Cervus
elaphus L.. Bi.son priscus /Bojanus/, Mammuthus
primige11ius /Blumenbach/; a large carnivore, Leo
spefaeus /Goldfuss/ and a medium fur game: Lepus sp.
/europaeus Pallas/ (Table 2.)
The traces of the Upper Palaeolithic settlement on
the SE plateau of the Henye hill could be observed on
a surface of 250 x 75- 80 m area. Excavations were
perfonned on fow areas (Unit 1- JV) by archaeological
sections (sec.) and trenches (tr.) (Fig. 1 .), which were
further divided by a grid system into l m 2 ( = - ) surfaces.
Occurrence of animal bones o n the excavated surfaces is a.symmetrical: sometimes they are concentrated, at other places, missing. Animal bones occurred on 36,4% of the total excavated surface ( 154
m2 of 423 ml): (Fig. I .)

l 0992.)
This part of rhe assemblage was investigated. by
Mlkl6s KRETZOT (KRETZOl l 964a.); the "extract"
of this work, i.e., tJie faunaJ list was published by
Laszl6 vERTES in his article on Lhe site (VERTES
1966. 7):

Table 1.

1.Il. 63. A trench., E, J , G. H, C, D. sections

Equus

94ps

9~MNl

A Ices

59

Ccrvus

B lSOO
,__

Mammuthus

Leo

L~pus

168

26

TOTAL

+in KRETZOI 1964, min. I0, but more

Animal bone remains of the 1982. Excavation were

transported into the Quaternary Section of the Archacozoological collection of the Hungarian National
Museum.

TotaL

Animal
bone
remains
found
(pieces)

area

Animal
remains

excavated
(m2)

found
(m2)

Unit I. ( 1963, 1982)

151

57

1083

Unit 11. ( 1982)

80

28

2 12

Unit ill. (1 963, 1982)

184

61

474

34

423

154

1803

Unit IV. ( 1963)

TOTAL

On the Henye-hiU, lhe surface of the andesice base

rock is uneven and is sloping more than the current
s urface of the sedimenl deposited on it.
Observations made duri11g excavation prove tha t
the cultural layer of the upper Palaeolithic settlement,
the contemporary walking level followed the surface
of lhe base rock, i.e., it is running paraUel to the
andesite surface. Consequently, the thickness of the
sediment overlying the cultural layer is a lternating.
getting thin on the top of the hill and more thick towards the slope. Due to viniculture practised on the
hill, the covering layers were mixed several times at
some places till the depth of 8~ I 00 cm. Conse-

113

'

BODROGKERESZTUR - HENYE
EXCAVATIONS ( 1963,1982)

UN IT IV Jm c

\
\

'

""

/ /

//

sel1 fement

aectlons

--- - - -.....

...........

------

h==

..........

'

'
:1

,,

'I

1'

,l

ti

42

Li);,.'''''
,Jr
'li-";f/,::1,,#.r !
~' 1l'1 1

u) t

(z

1
1

" ~~~',/I'.,~~
,,
11 . ,,.$,~'{s,~,
,,,. ':1,"1 , '
/

"

,,:;"':1,: 1~ 1.,..: .
~

"

4c___

~,,,:,~",:r,,,
1

I I

UNIT II.

~J B

..~"::::
,-:.

.,.:;, I
,:,;'.:
,, I
.Jl1,
i,

1
,;_:;' 1

,;::

,,

~~

...,, ,.,,,,,,h.,"'1'
...~ . wM
QM .,,. ~2 .:."I
..

.1 4'

... ;

"/"

""

.........
I

_,...,...

:~'.:.

lt I'

I I
/ I4
/ ,.,/
I

bOrder of I he Unlls

'i
'
'I

. :!.~

fauna

-- -

/ /
/
/ , :.,!,,; '"''',
/
~. ~

\
/

A-H

,,,,,,
/

6 \

- -

~')~/
........

--

trenches

.........

F '- ~'

1- 8

'

......

D~

8L~3

'\.

UNIT Ill.
.........

_1

N - -- - ;

Fig.. I Bodrogkeresztur-Heoye. Excavation areas of the Upper Palaeolithic site. Unirs 1.-Il.-Ul.-rv.

10 ..

quently, animal bones were also occurring in original

posjtion in the cultural layer, "in situ" as well as the
present surface.
Distribution of animal bones according to excavation units was the following:

trs. Excavated surface 184 m2, wilh 4 74 pieces of animal bones found on 61 m2 of the area (Table 2).

No animal bones were found at 1.63. B, F. sec. I.82. I.

4, and 5 tr.

Samples for C- l 4 analysis:

Equus bone - III. 82. 3 tr. W part, 7 0

Unil I. (Fig. 2.)

The Northern part of the cencre of the Palaeolithic
settlement: l.63. B, F sec., A tr. 1-15 m. Excavated
surface 151 m2, with 1083 pieces of animal bones
found on 57 m2 of the area (Table 2.).
No animal bones were found at 1.63. B, F. sec. l.82.
1,4, and S tr.
Samples for C-14 analysis:
I. loessy charcoal (300 g) - I, 63. F. sec. I 00-110-130
cm (from bum layer with charcoal, vERTES HNM
Archives V. 94.1965.)
2. Equus bones - I. 82. B. sec. 4 0 , f. 82. 7 tr.
Also from the profile ofl.63. F sec., samples wereoollec1ed for sedimeotological analysis (VERTES HNM
Archives V. 94.J 965.) Results by D. Moldvay were
published by L. VERTES (VERTES 1966. 7).

Fix depth data for the animal bones are the following:
111.63.H. sec. "disturbed layer"
-80, - 90 cm
E sec., E part
-{)0, - 80 cm
E sec. 2- 3 0
-100 cm
G sec. 2, 5 0
-80 cm(Mammuthus
mandible)
-80 cm
III.82.l . tr. 5 0 ..disturbed humus'',
(Mammuthus molar)
2. tr. 6 0

-90, -l 00 cm
(Mammuthus tusk fragm.)

3. er. 9 0

-80, -90 cm

4. tr. 3-4 0 "disturbed humus"

-80 cm

Uniz IV. (Fig. 1.)

SE edge of the centre of the Palaeolithic settlement:

fix depth data for the animal bones are the following:
1.82.C. sec. "under humus"
-80 cm
3. tr., Spart
3. tr.
6. tr. '' in deep-ploughed soil"

- 130 cm
-100,-llOcm
-60, -80 cm

Uni1 If. (Fig.3.)

NW pan of the Palaeolithic settlement: l.82. 1-4
trs. Excavated surface 80 m2. with 212 pieces of animal bones found on 28 m2 of the area (!'able 2).
Fix depth data for the animal bones are the following:
I.81. tr.
-60, -70 cm
4. tr.
-60, -70 cm
2 . tr.
-50, -70 cm
3 tr. W part
-{)5, - 70 cm

Unic m. (Fig. 4.)

Southern part of the centre of the eataeolilhic settlement: lll.63. A, tr. E, J, G, and ~sec., m. 82. 1-4

IV.63. C, D sec. Excavated surface 8 m2, with 34

pieces of animal bones found on all 8 m2 of the area.
No other archaeological material was found here. The
material inventorised as "stray fi.nds" in the Palaeovertebrate Collection of the Hungarian Geological lnsLitute as well as some unmarked pieces from the sit!!
correspond to the quanti~ and description and excavation observations by VERTES, which said for the D
sec: many horse, mammoth, Megaloceros . . . (Old
settlement) ... (HNM Archives V. 94. 1965.). Among
the provenance of palaeontological finds. the A tr. and
a1J sec.-s are found with the exception of D. and C.
sec-s. Probably, we can identify D. sec. With "stray
finds" and unmarked finds with C. sec.
Fix depth data for the animal bones are the following:
- 20 cm
TV.63.C. sec. "disturbed layer''
(Mammurhus molar)
D sec., "cultural layer in
-70 cm
original position"

11 5

'

Table 2. Bodrogkeresztlir- H enye. Distribution of animal remains on the hunting station (N umber of specimens)

Sec:lion

I.

Unit

J982.
D.

c.

Ii.

B.

A.
2.

Trench

Sum

10

10

12

14

JO

61

JS4

24

IS

28
46

39

11

24

52

142

JO

483

85

16

12

11

40

16

J2

10
40

14

97

224

13

-~
- -

20

42

10

21

767

A Ices

17

43

12

12

Mammulhus

12

Dison

21

245

28

47

Bone frag."

22

398

73

Total

32

643

IOI

lJnrB

57

+ OufQ from 1hc c.\CllV!lhon di:iry

284

16

153

26

200

15

42

+
+

4.

f-

~ ~

384

14

18

13

- 699

1083

20

36

26

45

71

20

81

97

1-4.

Sm.

8
35

Lepus

A.

Tolol

s
26

Leo

o.

12
189

Sum

J.

1.

3.

E.

1963.

7.

6.

2.

H.

I V.

'82
_._

8.

Equus

Ill.

l96J.

l.

Surface m-

Ccrvus

n.
1982.

. 3

53

40

73

272

90

202

10

1036

16)

474

32

1803

12

67

27

87
125

57

55

14

212

12

124

82

53

40

U I. 82 .
E8

Fig 2

N - -- - - '
7

E4

E7

14

E 1

E 2
6
!01

A9
,,,

L
All

E 13 E2
3
l

Ell

E 12 E 15

~2 ~n 5 II 5

EJf> E 41 E 2

~-

..

L'

"

~ 12 ~ 4

:t I! 12 E 9 E II

Ill 1

E 2

,s

A 17
"" 3

E2

\~

f 3
E 3

Sectlons

trenches

36 78

"4
~ 2

..

I 7

E I

:t

!
[

IE 3

'

~-"

EI
1

c
A. B.C.D E

B
E 11

..

,.._
2.

E Equus . A Alces . M Mammuthus .

X burned bones

Fig. 2 Bodrogkereszni:r-Henye, Unit l. 82. Bone distribution among the species

2. Descriptio.n of t he big animal remains- topograpbical distribution

2. I. U11it 1. (Fig. 2.)
T. 82. A. sec.
I. 0

2. 0

3. 0
4.0
6. 0

UnIB (unidentified I p. (piece)

bone)
1 - os sesamoideum dist. fr.
J!quus
9
- s. rad. prox. med. fr (4 ps), 3 carpus fr., me. diaph.
Alces
fr. (2 ps)
18 ps.
UnIB
3 -2 scapula fr., me. diaph.
A1ces
3 ps.
UnIB
4 ps. (2 burnt)
UnIB
3 - rad. diapb. fr. (2 ps), os capitatum (CJ)
Equus

C. 82. 2. tr. (parts between A sec. l -3 D and B. sec. 7-9 D)

1. D
Equus
4 - tib. diaph. fr. (4 ps)
2. 0
Equus
3 - tib. diaph. fr. (3 ps)
Uurn
2 ps.
Equus
1 - M sup fr.,
3. 0

UnJB

5 ps.

117

L 82. B. sec
2.0
3. 0

Equus

4.0
5. 0
6. 0

Eq11us

7. 0

Equus

8. 0

Equus

9. 0

Equus

Equus
Alces

UnlB
Equus
A lees

UnlB

Alces
UnlB

2 - ti b. diaph. fr. (2 ps)

11 - 3 M sup fr. ,brain skulJ fr. , s .tib. diaph. fr. (7 ps)
3- stenebra fr., s. rot. diaph., mp. diaph . fr.
l9 ps .
3 - d. tib. diaph., d. astragalus fr., d. calc. fr.
3 - d .[2-3, tib. diaph. fr.
7 - 2 hum. diaph. fr.. s . hum. distr. fr., d. me. dist fr.,
d. ml. 2 trocbleae. mp. diaph.. fr.
7- hum. diaph. fr., d. ('?) calc. fr. , os cuneifonne Lat (TJ),
os cuboldeum ((T.:i-s), d. mt prox., 2 mt. diapb. fr.
S - d. p4_M1-3, d. astragalus
3 ps.
4 - d. rad.-ulna prox., rad. prox fr. Ouv.), d. os ph. I. post.
2 - s . tib dist . fr. (2 ps)
3 ps.

T. 82. C. sec
l. 0
2. 0

Eg,uus
Eguus

12 - s. lib. d ist. diaph. Uuv., 12 ps)

25 - s. M112, d. ~ . diap. (J ps), os capitatm (CJ),
d. pelvis (6 ps, burnt}, fem. diapb. (4 ps), tib. diaph.
(3 ps), d. tib. dis~ . mt. dlaph. fr., mp214 prox. fr., os sesa-

molqeul11 prox., OS ph. il. prox. fr., OS ph. II. dlst.

Alces

(burnt), OS ph. ITT. fr.

4 - s . hum. dlisL. diaph. Quv.), 2 me. diaph. fr. , os ph.
prox . fr.

UnIB

14 ps.

3. 0

Eq1ms

8 - pelvis fr. Guv., 4 ps), d. rib. prox.-<list. diapb. (4 ps)

UnlB
Equus

3 ps.

4. 0

5. 0

A Ices
Un TB
Equus

6.0

Alees

9 - s M3, s. pel vis fr., d. fem . prox. (6 ps), os

naviculare (Tc)
I - mp. prox. cliaph.fr.
6 ps.
41 - 3 M sup. fr., s M112, s P314 fr., d PJ/4, d. M 112 fr.,
corpus mandibulae fa, hum. ctiaph. fr. (4 ps), s. rad. pro x.,
3 rad.prox. med. -fr., mc.prox. fr. , 2 mc.diaph. ., s. me.
disl., mp214prox. fr.. 2 s. caput fem., 2 fem . prox. fr.,
2 fem. dlst.fr., 2 tib. prox. fr., tib.dlaph .. fr. (8 ps), s.-d.
ti.p. dist., S: calc., d. OS ph. Il. post.
7 - 2 tib.diaph.fr., d. mt. prox. fr., (4 ps), d. mt. diaph.

Unffi

43 ps.

Equus

12 - d p4, d. ram. mandibulae fr., d PJ;4 fr., s. M2,1mm.

prox. diaph. fr., d. _pelvis (burnt). d. tib. diapJ1. fr. (5 ps).
OS p!h. Il. prOX (burnt)
2 - me. dlst dlap.h. fr. (2 ps).

A Ices
7. 0

8. 0

Unffi
Equus
UnlB

Equus
A lees
UnlB

9. 0

Equus

16 ps.
2 - d . .hwn. diapb. fr. (2 ps).
5 ps (burnt).
2 - d. rnt. disL, OS ph. ill. fr.
12 - rad. diaph. fr. (4 ps), fem. diaph. fr. (3 ps), s .-d.
rib. dist., mt. prox. fr. (3 ps).
6 ps.
l5 - d p2 fr., s . P2, s. P314 2 M inf. fr., corpus mandibulae
oral fr., d. rad. proxfr., rad.diapb .. fr.. (2 ps), fem . diaph.
fr. (3 ps), d. os phl. dist. ant. (?), d. os ph.11. fr. post.,

s. os ph n. anr.

118

r.

A Ices

Un TB
10. 0

Equus

5 - 2 me. prox fr., d. mc.dist.. fr., tib. dist. fr., os

malleolare fr.
I 0 ps (2 burnt).
30 - s. brain skull fr., s. 11-2, d. f3, scapula fr. ( 11 ps},

rad. prox. fr., s. os pisiforme (Ca), pelvis fr. (juv., 3 ps),

ll.O
12. D

Alce.s
UnIB
Equus
lllc es
UnIB

Eguus
Mammuthus

s. fem . prox. fr. , tib, prox.fr. (burnt), tib. diaph. fr. , s.

calc., 2 mp2/4 fr., s. o:s ph. I. ant. - post.. d. os pb I. ant.,
os ph. L prox fr.
3 - d. mt. prox. diaph., d. mt. dist. fr. (2 ps).
6 ps.
2 - OS ph. l. post. lal fr., OS pb. l. distr. fr.
4- d. tib.diaph .. fr. (4 ps).
5 ps.
2 - s P'\ s M 3
5 - corpus mandibulae fr (5 ps)

I. 82. C. sec. "strayfinds"

Equus

A Ices
Bison
MammuJhus

UnIB

29 - I inf. fr., 4 M inf. fr. , corpus mandibulae fr. , s. ram.

mandibulae (burnt), hum. distr.fr., rad. prox. med. fr., 3
pelvis fr., fem.diapb. . fr., (4 ps), 3 tib. prox. diaph. fr., tib.
diaph. fr. (3 ps), 3 mp. dis t.fr., mp214 prox. fr., 2 os sesamoideum prox., os ph. II. prox fr. (burnt)
5 - rad. prox. fr., fem.diaph. (2 ps), 2 os pb. I. dist. fr.
I - s.tib.dist.
7 - rusk aboral temllnal piece (2 ps), praemaxilla fr.
(5 ps)
284 ps (154 bone grit, l30 "bone morsels") 25 burnt,

L 82. E. sec.

2. 0
4. 0

Equus
Equus

Un.IB
5. 0

7.0

Equus
UnIB
Equus

UnIB
10. 0

Equus
UnIB

I - d. OS ph. I. ant.
l - mp.diaph. fr.
3 ps.
2 - scapula fr., bwn.diaph. fr.
8 ps.
13 - d. rt-2 Guv.), d. p 3/4, s. P3/4, fem .diaph. (8 ps). OS pb
II. prox.fr.
J ps (burnt).
2 - mp.diapb. fr. (2 ps).
2 ps .

L 82. E. sec. " during excavation"

Equus
A lees
Bison
UnIB

7 - hum. dist. diaph., astragaJus fr., 2 mp214 fr., 2 os

pb. l. diapb. fr., OS ph. 11. dist. fr.
I P sup. (germ).

I d Mt-257 ps (19 burnt).

I. 82. D. sec.

2.0
4. 0
7. 0

8.D

UnIB
A lees
Equus
Mammuthus
Alce.s

On TB

2 ps.
8 3 costa fr. , s. tib. dist (2 ps), tib. diaph. (3 ps).
3 - s. pelvis fr. (juv. 2 ps), rib. diaph. (burnt).
I - tooth lamella fr.
9 - 2 vert. lumba. fr., 3 costa fr., tib. diaph. (3 ps), me.
prox. diaph.
20 ps.

Il9

l. 82. 3. tr. - 100-110 Cni

Equus

7- s. JI Guv.), s. dp3 + p3, s. p2 Guv.) s. Ml-2, M3

(genn).

Equus

I - costa fr.

1. 82. 3. tr. S. part. 130 cm

I. 82. 6. tr. "in deep-pl.oughed soil" -60, -80 cm
Equus

Un TB

7 - s. tib. dist. diaph. (2 ps), s. as1ragalus - calc., s. os

cubeideum (T4-S), mp.diaph. fr. (2 ps)
Gps.

I. 82. 7. lr.
UnIB

4 - vert. cervicalis fr., 3 costa fr.

14 ps.

" burnt bones"

(data from the excavation register)

Equus

r. 82. 8. er.
2.2. U11it //. (Fig. 3J

Description of the trenches from N to D:

11. 82. I . tr. I0-20 m

16. 060-70 cm

Equus
Alees
Equus

I - d P2 fr.
7 - corpus mandibulae fr., mp. diapb.fr. (6 ps).

3 - s. me. disl. fr. (3 ps).

Alces

9 - costa fr. (burnt), d. tib. dist. (8 ps).

Equus

7 - d. Ml -2, 2 costa fr., d. rad.dist. fr., 2 rad. diaph. fr.,

os pb. T. dist. posL
l - tib. diaph.fr.
2 - cib. diapb.fr., me. diaph. fr
8 - d. rad.-ulna prox., rad. diaph. fr. (6 ps).
3 ps.
I - s. pelvis ft.
4 - magnum (C2+J), os uncinatum (C4+s), os scaphoideum (Cr), os triquetrum (Cu)

II.82.4 tr.
L. D

5.0

Equus

6.D
7.0

AIces
Equus
Un!B

8.0

Equus
Alces

14. 0

UnTB
Un TB
Alces
UnTB
UnTB

16. D
20. 0

Alces
Eq1ms

L2. 0

13.0

12 ps.
9 ps.
5 -costa fr. (3 ps)., 2 mp. diaph.fr.
8 ps.
3 ps,
I - s. mc.prox.
7 - d. caput fem., lib. dist. fr. (5 ps), os ph. 11. prox.

ant.
UnIB

10 ps.

Equus
Equus

I - d. os ph. TI . ant.
2 - M sup. fr., fem. dist. lat. fr.

Alc~s

8 - P sup., d. tib.dist., d. ascragalus, d. calc., d. os

et.fr., d. os cuneiforme interm. et lat. (T2+3). d. mi. prox.,
mt diaph. fr.
49 ps (20 burnt)

H.82.2. tr.
i. D
8. 060-70 cm

UnIB

120

u . 11.a2.

Fig 3 .

N4

I! 7

E 1

E 1

f2

A 2

E 8

E t

A4

E
A 8

to

re

A S

A5
8

E1

E3
A7

At
C1

E 3

A t

" 9

E "'7
20

10

1. 2 .3.4 . trenches

E Equus
A

Alces
Cervus

1--i
lm

Fi.g. 3 Bodrogkcrcszmr-Henye, Unit a. 82. Bone distribution among the species

121

11. D

Un.IB

9 - d. 11, d. os zygomaticum fr., carpale fr. , os capitatum

(C3), os lunatum (C1), os triquetrum (Cu}, s. me. clist.,
mt. dist. Jat. trochtea (burnt), s. os. ph. l. prox. lat.
fr. ant.
3 - s. corpus mandibulae fr. +M1-3, s. mt. prox. fr. ,
os lunatum. (Ct)
8 ps.
2 -capur hum. fr., tib. d.iaph. fr.
6 ps.
1 - tib.diapb. fr. (L 210 mm)
8 ps.

Eqrms
UnIB

2 - M sup. fr., d. hum.dist.diaph. fr.

2 ps.

Equus

Alces

14. D
18. D

UnIB
Equus
UnlB
Equus

IJ.82 . 3: tr. E part (1-10 rn)

lI.82. 3. tr. W part (I 0-20 m) -65, -70 cm

Equus
A Ices

Cervus
UnIB

I - s. corpus.rnandibulae + dp2-4, M 1.2.

1 - tib. di.St. fr.
1 s. tib.dist.
7 ps.

2.3. Unit ill. (Fig. 4.)

Ill.63. A tr. 26-30 m

Equus
Bi.ton

Mamm.utlws

2 -s. M3., d. rib. d.ist,

2 - mp. dist. trochlea, os ph. TI.
4 - too~ lameUa fr., diaph. fr. (3 ps, tib. ?).

Ill.63. A tr. 36-39 m. (area adjoining lll.63. E. sec. 28-29-30 D , Fig. 5.)
3 - s. fa, os scaphoideum (Cr) fr. , s. me. disl
36.D
Equus
3 - s. Mll2, Iiinf. Uuv.), d. tib.dist.epiph. fr. Uuv.).
AIces
37. D
12 ps. (cfata from excavacion register)
.
UnIB
10 -.s. praemaxilla + r t- 3, d. 11, d. corpus mandibulae fr.
38. D
Equus

+ M112, costa fr., rad. prox. fr. (2 ps), d. fem. dist. epiph .
Alces

UnIB

fr.. patella fr., tto. diaph. fr. (2 ps).

3 - cl corpus mandibulae (.inf. 8-1 0 months old). s. corpus
mandibulae fr. (ad.), s. ram. mandibulae fr.
2 ps (diaph. fr.).

lU.63 . A tr. 39-42 m . (area ac1Joining lll.63. J. se.c. 28-29-30 D (Fig. 5.)

40. D
41. D
42. D
Of these remaining:

48 bone ps. (data from

30 bone ps.
2 bone ps.

Eguus
Alces

Mam mu.thus
43.D

UnJB

lll.63. A tr. "stray finds" (probably 39-42 m.)

Equus

122

~cavation register)

l.0 - d. Jl, s. 11-3, d. p 2, 1 s. -2 d. M ll2, d. corpus

manc!ibulae :fr. + P 3t4. os ph.1. ant. fr.
8 - d. max. ;fr. + dp2-4, d. max. fr. Ml, (M2 gean),
d. M'n, s.-d. mandibula symph. fr., d. corpus mandibulae
+ P4 - M 1-3, s. M3, d. me. prox..
5 - ~surface stray fmd" praemaxilla fr.; 2 costa a-t. 190,
360 mn;1 ~ M sup. aboral fr. tooth lamella fr.
I sp (data from excavation register)
14 - I sup. fr. 2 d. p3/4, M sup. fr., s. MJ, d. M3 fr.. d. Pi,
d. Pat<1, d. M inf. fr., d. M1/2; 2 s. - 2 d. M3

U.111.63. 82.
N ______,

Fig. 4

..........
Im

20

10

10

20

Fig. 4 Bodtogk:eresroir-Henye, Unit Ill. 63., 82. Animal bone distribution

u. Ill.
H
38

....

"
6

12

...

11- - - . .
~

$:>.c:>

......

Ag . 5

sec

31
~

'r::> ~

63 .

,,.

f1~

<=::>

~~ ~,

E sec
E

13

12

Bison

Equus

A lees

M Mammuthus

15

<:>.

~60m

...30

J . sec

e 4
A 10
B

c::>

I?

'J

A trench

e 1)e 1

41

Fig. 5 Bodrogkeresznir-Henye, Unit m. 63. "A" tr. Bone distribution among lhc species.
C-0mple1ed a fter Vertes I 966, Fig. 3.

123

Alces
Bison

2 -. rad. dist, d. tib. dist.

1 - s. astragalus

Equus
Alces

6 - s. p3/4, M sup. fr., d. Ml/2, d. M3, d. Mui, d. me. dist.

6 cast right side antler beam, 2 P sup., d. rad. dist.
epiph., s. uncinatum (C4+5), d. os lunatum (Ci).

111.63. H sec., "disturbed layer"

111.63. E sec., (Fig. 6.)

Data from excavation
3. 0 - LOO cm
22. 0
25. 0
26. 0
27. D
28. D
29. D
30. D

register.:
1 bone ps.
10 bone ps.
14 bone ps.
2 bone ps.
1 bone ps.
10 bone ps.
8 bone ps.
9 bone ps.

We have no further infonnation on the large, cca. 95-100 cm long object published from E sec. 30. 0
(vERTES 1966. Fig.2.) It did not appear in the bone material investigated (KRETZOJ 1964 a). In the excavation register, there were drawings of8 animal bones and one tooch beside this object (vERTES HNM Archives

v. 94. 1965)

111.63. E sec.
2. 0 I 00 cm "cultural layer"
Equus
A lees
8. 0
Alces

13-15. 0

Equus

24. 0

Mammuthus
Alces
Bison

Alees

2 d. corpus mandibulae fr., + M3, d. hum. clist. med. fr.,

3 -s.-d. magnum (C2+3), s .unicatum (C4+5)
3 - corpus mandibulae fr., s. ang. mandibulae fr. , s. scap.
dist fr.
5 - 2 d. p314 Guv., ad.), d. M tn ), s. hum. prox. fr. Guv.),
d. lib. clisL ft.
2 - s. ram. mandibulae fr. , scap. fr.
1- d. m2
I - s. corpus mandibulae + P-M.
l - s. me. disl.

111.63. E sec. "disturbed layer"

Equus

Alce..s
Cervus
Bison

4 - s. Mll2, d. M 112, d. hum. dist. med. fr., d. os ph.

U. posL (?)
3. s. OS unicatum (C4+ 5), s. OS lunatum (Ci). s. OS
triquetrum (Cu).
I - end of antler tine (L. 57 mm)
I - s. astragalus

lll.63. J sec. (Fig. 6)

According to excavation register, lhere were ''one-two, more or less and
sometimes abundant" remains of animal bones in the foJlowing 0-s:
4-5 ., 8-9., 10., 13., 17.. 19-20 .. 24.
I. 0
Equ11s
2 - s. Ml, d. astragalus fr.
26-29. 0
Alces
5 - s dp4 fr., d. P<t, 2 s. - I d. corpus mandibulae fr.
2 - d. hum.disc. med. fr., d. tib.dist. Guv.).
28. 0
Equus
13 - d. Mll2, s. Ml, M sup. fr., s. di21J, dp2-3, s. corpus
Alce.<>
mandibulae + dp2-4, d. corpus mandibulae fr. + M112, d.
rad. prox., d. olecranon fr., 2 mt. diaph. fr., os ph. Il.
prox. fr. post.
124

u. Ill. 63 .

N4
i------

36m

H -----1

38
9

6
1

12

10

30
x

15

19

14

11

10

13

21

10

13

x
17

15

81

11

2~

19

x
2

10

22

15

2
32

12

48

34

14

2
15

10

45

43

x
1

10

12

x
6

x
3

31
14

t--- - -

D
1m

26

19

x
5

44

26

31

36

39

10

23

E - -- +-- - - J

-sllex

28 29 30
25 21S 27
22 23 2"'

-bone

18
18

~o

21

18

18
10
7

17
14
11
B

5
2

..

15

12
9

e
3

quad rats
Fig. 6 Bodrogkcrcszuir-l:lenye, Unit Ill . 63. E-J-G sec. Animal bone and silex disllibution

125

27-30. 0

Eguus
A Ices

30. D

Ce.nJuS
Bison
AIces

3 - d.ra.d.dist fr. Guv.), patella fr., s. rib. dist fr.

18 - d. max. fr.+ p3+4, s. dp2 -4, d. dp2-3, s. p 3, 3 d. M 112,
d. corpus manctibulae + P+M, d. dp2-4, s. M112, tib. dist.
fr., os ph. I. m.ecl post., os sesamoideum prox.
1 - d. tib. dist.
.
2 - d. os cen.trotarsale med. fr. , mp. dist. trochlea fr.
I -d. corpus 111.andibulae + P+M

IIl.63. J sec. ("disturbed layer")

Equus
A Ices

IIl.63. G sec. (Fig 6.)

2. D
Mammutlms
S. 0
JJammuthus
19. -22. -25.-28. 0
Equus
22. D
Alces
21.-24.-27.-30. 0
Equus
Leo

3 - s. ram. mandibulae fr., M . sup. fr., d. tib. dist. fr.

3 - s. corpus mandibulae + dpi-4M1-2 ( 14-16 months
old), d. ram mandibulae fr., d. astragalus

2 - fem. diaph. fr. (2 ps).

3 - pair of mandibulae (Fig. 8.1.) fem. ctiaph. fr. (2 ps)

8 - s. p.3/4, d. dp114, d. P3/4, s.-d. M112, s.
hum.dist. fr., hum. dist fr. , s. os naviculare (Tc).
J - d. corpus mandibulae + P+M.

9 - s.-d. I 1, s. I 1-3, d. 13, s. mandibula oral fr., s. tib.dist.

fr.1 t:ib. diaph. fr.
I - d. mt3 .

lll.63 . D sec. "during excavation" - 50, -60 cm (Probably the area between the squares 28. - 19. - 21.-30. D ,
the.Eastern part of the section )
7 - r ight side brain skull fr., s. p314 fr., d. M112, M3. fr.,
Equus
s.-d. rad. prox. fr., m p. dist.fr.
8- brainskuU fr. (5 ps), l 1/2(gem1), P3/4, d. me.
A lees
prox.fr.
ill. 82. I. tr (Fig. 7.)

5. 0

Equus
Mammuthus

6 - oral fr. ofmandi.buJa, s. astragal.us fr., s. calc. fr.,

mc.prox. fr., mt. dist. fr.. mp2/4 prox. fr.
3 - s.-d. mandibula sympli. fr., M inf. fr.

ill. 82. 2. tr

6. 0

Equus
Mamm~tthw

3 - d. tib. disl. fr. (3 ps).

2- twlk fragment (L: 1800 mm, greatest diam .:
110 mm, (Fig. 8.2.), s. fern. diaph. fr.

HI. 82. 3. tr
4. 0

Equus

A.lees
DnIB

7. 0

Equus

9. 0

Alces
Equus
A lees

Leo
UnIB

126

7 - c. fem. fr., fem . diaph. (2 ps), tib. diaph. ( 4 ps).

3 - costa fr. (3 ps).
20 ps.
13 - 6 cost-a fr s.scapula fr. (3 ps),., d. hum dist. (2 ps),
d. rad. prox. fr., tlb.diaph.
3 - costa fr. (2 ps), os ph.1. di.St. fr.
1-sP:i

2 - s.-d. corpus mandi bulae

1 - OS ph. 1. s. I. post.
10 ps.

u. Ill. 82.
N--i

E 6
loll 3

10

E 22
A

so

e. a
"' 2

20

10

E Eqws

Alces

C Cervus

Leo

,,,,

M Mammuthus

Fig. 7 Bodrogkcresztilr-Henye, Uni1 DJ. 82. 1-4. Bone distribution among the species

ru. s2. 4. tr

3.-4. D

Equus

22 - s. MJ fr., costa fr. (8 ps), c. hum:fr. (2 ps), d.

Cerv11s

hum.dist, d. pelvis fr., caput. fem. fr., fem. diapb. (3 ps),

tib. diaph. (5 ps).
6 - costa fr. (4 ps), scapula fr. (2 ps).
I - antler tine

UnJB

60 ps.

A Ices

127

N___.i
G4

GS

G1

G2

2.

1.
Fig. 8 Bodrogkereszttir-Henye. Unit Ill. L 63. G sec. anvil made of mammoth mandible (After Vertes 1966 Fig. 5.,
completed I corrected). 2. 82. 2. tr. marnmot.h tusk.
l.4. Unil IV. (Fig. / .)

IV. 63. C. sec.

Equus
Mam mu.thus

I - astragalus fr.,
I - NJ sup. fr.

Equus

I 0 - s. p2 Uuv.), 3 s. p314, 2 d. PJ/4, 2 d. M 112, d. MJ fr.

rad. cUaph. fr.
2 - 2 M inf. fr.
l - s. me. dist. fr.
8 - M fr., long bone fr. (7 ps)
l - ulna fr. (data from excavation register and KRETZOJ
l 964a, lost).
10 ps.

IV. 63. "stray finds" (=D. sec.)

A lees
Cer.tus

Mamm11th11s
Lepus

UnJB

3. Zoological a nd palaeontological characterisation of tJ1e animal species

3.1.Equus

Equ11s remagensis SKORKOWSKJ 1933.. Eq11us germanicus '' (NEHRING 1884)

483 remains - 50 individuals

3.1.1. A natomlcitl oftb e E quus remains

He.od region

At the Palaeolithic campsite Bodrogkeresztl1rHcnye, 483 pieces of Equus bones could be identified. The anatomical division of these are listed

3 brain skull fragments (basioccipitale).

2 facial skull fragments (praemaxilla fr., os zygo-

below:

maticum ),

128

15 isolated upper inc.isives

7
4

II

12

i3

2 sin. I sin. -

3 dext.;

21 isolated ~pper premolars

5
P
2 sin. p3
I
l sin.,
p 3/4 6 sin. II
p4
3
l sin. 30 isolated upper molars
4
M
2 sin. M2 l sin. 2
Mt/2 2 sin. s
M3 4 sin. 7
12 mandibulae
4 corpus mandb. I sin. 2 mandibula lower
edge, fr.
3 mandibula oral fr.
3 ram mandib. fr. 2 sin. 8 isolated lower incisives
3
2 sin. 11
I
l sin. h
3
2 sin. 1J
13 isoJated lower premolars
2 sin. 3
P2
9
P314 2 sin. 28 isolated lower molars

11
l
8

5 dext.
3 dext

3 dext.

11

113

fr.

3
I dp sin.

5 dext.
2 dext.
2
l
3
3

dext.
dexL
dext.
dexl;

Hind limbs

12 Mfr.

22 pelvic
(pars acetabuli, corpus o .
ilii, ala o. ilii) 4 sin. -

l lfr.

42 femur
8 prox. epiph. 3 sin. 30 diaph.
4 dist. epipb.
2 patella fr.
96 tibia
3 prox. epiph.
81 diaph.
I sin. 12 disl. epipb. 3 sin. -

I dex'f.

l dex:t.
I dext.
I dexl;
I dext

7 dexl; dp314 I sin

M112

2 sin. -

9 dext.

Mi
MJ

I sin.
5 sin. -

3 dext.;

8 Mfr.

Trunk region
I vert. cervicalis fr.,

21

C0Sl8

fr.

8 dex.t.

I dext

9 dext
9 dext.

l8 tarsale
7 astragalus 2 sin. 3 dext.
6 calcaneus 4sin. 2 dcxt.
2 os naviculare Tc, os cunc ifom1e lat. TJ, 2 os
cuboideum T4+5
8 metatarsale
2 prox. epiph.
I dext.
3 diaph.
3 dist. epiph.
I dext
17 metapodial / mc/mt/ diaph.fr.

Phalanges

Fore limbs
15 scapula fr.
24 humeri
4 prox. epiph. I sin. Guv.}
12 diaph.
8 dist epiph. 2sin. 33 radiis
15 prox. epiph. 4 sin. -

16 diaph.
2 dist. epiph.
2 dext.
2 ulna fr. (olecranon)
2 dext.
8 carpals 3 os capitatum CJ, os triquetrum Cu, os
pisiforme Ca. os lunatum Ci, os scaphoideum Cr,
carpals fr.
l 0 metacarpals
I prox. epipb.
2 diaph.
7 dist. epiph. s sin.
2 dext.

14 OS ph.l.
4 dext.

2 dei...i.
3 dext.

12 OS phJl.

5 ant - 4 posL; (7 complete,

2 prox., 3 diaph.. 2 dist. pieces)
3 ant. -2 post.; (5 comi>lete,
5 prox., 2 dist pieces)

2 OS ph.ill.-fr.
4 os sesamoideum 3 prox. I dist

Anacomical division of Equus remains within che camp site are presented on Table 3.

3.1.2. T~tb
For the exact classification of the Henye hill horses,
their relationship, chronology and species evolutionary level they should be compared to Equus finds from
other well known and documented Upper Pleistocene
sites. For the comparison, primary and characteristic

dimensions of fossil material, teeth and posicrall'ial

bones can be used.
Skulls were cracked and destroyed. with only 66
pieces of isolated upper teeth remaining.

129

Table 3. Bodrogkeresztur-Henye. Anatomical distribution of the Equus remains on lhe hunting station
(Number of specimens)
!.

Unit

D.

Section

A.

r~neh

P sup.

M sup.

mMdb.

r mr

,I'.inf.
_

M inf.

10

dt

rad px

dph

-I
2

dph

IS

dph

17

dt

dpb

dt

ph I

rt.

Lll

7
2

5e5Blll

189

12

I!

IS

- 162

->-

22

JU

,_

5 I

Kl

12

J
I

2
I

mcimtdph

mtpx

llus:1I

r
r

2
II

IS

--r

calc.

21

2
2
2

28

I
4

astg.

I I

11

13

39

3
I

di

130

patella

~ll

21

30
12

6
2

I
7

18

Total

r2

!Cm p-c

dph

dt

.l.

Clll'paJ
me px

3.

:!

Totnl

ulnn

'-'--

I.

TV.

c. D.

I
I

dl

pelvis

,\.

I
I
I

dph

2.

costn
II

&a1pula

I.

I l

vcrtcb.

hump)(

~p

6.

7.

G.

J.

focitll skull
3
3

F_

H.

2.

bann skull

Ill.

11.

c.. r=. n.

17
I~

12

I
26

35

2-1

I IS

39

II

10

24

21

10

483

Among the 23 incisors (15 sup. - 8 inf.) there were

no deciduous ones. 3 of the incisors are not worn yet.

Dimensions of jav. and ad. upper incisor crowns are

presented below in mm:

JI

I Width

Sag.L

12

13

16.5

19

19,5

20

17

18, l

21

21

17

18

19.S

20

ll

12,2

12

10

11,1

10

12

10

13

12

13

juv.

juv.

juv.

Canine (C) is absent from the c.arnpsite 0f Henye hill.

The Henye-hill horses can be characterised well on the basis of29 upper and 4 l lower cbeek teeth and teeth
fragmems (Table 4-5).
Height ofthe teeth
Dimen5ions of cheek teeth, e.g. height depend on
the species-evolutionary level and the age of llie indi-

G. Ne bis (NOBTS 1971. 7.) assigned tbe height of

tee-th into three abrasion levels, taking into consideration abrasion of fhe Leeth with the advance of age:

vidual.

Height of the teeth (mm)

Age

Abrasion levels

A.l.

From after the appearance of

the pennanent te-eili till 6th
years or age

All.

7-16 years

A.IO.

Afier I 7. years

Upper

Lower

75 <

60 <

75-55

60-4-0

<55

<40

Height of the tootb was measured from the bifurcation of the roors to the upper grinding surface (or lOp) of the
looth .
Upper teeth from Henye hill :
NOBIS categories
A. II.

A. Ill.

M112

M3

A.I.

p2

p3f4

Upper teeth height (mm)

~UV.

Ml/

85

96

?3'
98

9(1

IJO

Ml

8&

86

8.S

gs

RS

75
75

8J

64

75

80

mru.

ad.

subad.

P2

SlOll

60

54

58

60

45

65
65

76

75

:>S

66

66

Average height of lhe upper leeLh in A. I. was 83,8] (n = 16), 98-75 mm; ii\ A.TI. 63,00 (a= 8) 66-58 mm.
Average height of upper teeth (in mm):

pllJ (Ll = 9) 72,89 96-58

Ml 1l(n-= 10) 82,30 98-65

131

Table 4. Bodrogkeresztur-Henye, Dimensions of the upper cheek teeU1 of Equus (measured al the occusal
su1face; in mm)

,I-

p3/4

t.

JR,5

39.0

l,

26.5

26..5

30.D

--~----+-

15,0

4S,O

- s.
1.45

l,H

maJ

I.

3.

35,0

26,0

28,0

26,0
76.0
14.0

1s.s

75.0

32,0
96,0
13,0

4.

S8,0
l S,5

34.IJ

34,0t

83.0
17,0

60,0
14,0

85,0
11.5

15,0

mat

35,0

2.

64.0
14,0

32.5

____,i -_2_9._S _ f - 28.0_I

- - --+- - - - + -- 4_s_.4_o-l-__
4~
....:.1_ _ _
so-'-.o--1-_.
_~--1--s~
2 ~--1---3;.;..
3';.;..s_+--44~ _
9,0
9.3
9,7
9,6
10,0
9,2
1.00
0,96
0,98
1,03
1.15
1,17

6.- - + -1-0.-2-+---1-0-,4

l___2:_

30,0
31,0
31.5
_,__31_.o___
~ o_.o
___J o_.s_

54.0
14,SO

7 5,0

4.

I"

30,0

---+--J_o_.o_

5.

65.0
13,5
48.2
7.9
o.98

28,0

:!8,S

31,0

32,0

32,0

i- -2_9.;_.o_+-_ 2_1_.s_ -+-_Jo.;_.o_ -+-_ 21-'-.o--+-_

65,0
85,0
86,0
8Q,0
IS,O
1-5,5
14.8
12,0
S3.S
54_,J_ t - _4_7._7__,'--_
37....:.,:i__
1
8.S?
8,2
9.3
8.6
o.96
1.00
1.04
1.1s

32,0

31,0

2s- '-.o--1- 2~
98,0
88,0
16.0
50,0
8,9
9,0
1,14
1.13

2~

90,0
15,0

37,1
SJ.8
45,4
6.
I
11,2
6,7
8,6
1.
I 1,09
1.00
1,21
::::=====~===========---===:;::....-=============~-----L- - - ' ----''-----L- - - - " ----'---'--..;.._-'
MJ

28,0
l4,0
66,0

I.

2.
3.

28,0

:?ft,()

30.0

32.5

15,0

26,0

27.1

75,0

ss.o

TI.O
85.0

66.0

II.I
3-l.2

'--_ .i_._-+-_ 16,0 --1- - --t--l_S,O

_ __ l_J,_0--1
I s.
s1,2
53,s
I 43,4

6.

6.7
1, 17

7.

7,0

7.3

1.12

6,9

8,8

1,30

1,20

m111

1. length., 2. breadth, 3. height.. 4. lengtl1 of Pc., 5. Index of Pc., 6. occlusal surface measure (cm2), 7. proponion length to
breadth += emhryyo (not yet e rupting; the tooth crown was still sunk down io the alveolus

Lower teeth from Henye hill:

A. L

NOBIS cateanries
A 11.
2

A. ID.

I
-

1'1

Pvt

M111

~J

Lower teeth height (mm)

juv

Subad

ad.

84

P31-4

Mia

scn

mtiL

SS

P2

9S

90

M3

so

60

80

75

78

Sl

70

60

58

56

Average height of the lower teeth in A.1. was 81,72 (n = 7), 95-70 mm~ in A. ll., 56.28 (n = 7) 60-50 mm.
Average height of lower teeth (in mm):

M112(n = 4)85,00 95-75

MJ (n = 2) 74,00 78-70

132

35

28

54

45
211

Table 5. Bodrogkeresztur- Henye., Dimensions of the lower cheek teeth of Equus (measured at the occlusal
surface: in mm)
P2
I

33,4

.14,0

18.2
18,0

2.

3.
4

dp314

P314

36.6~

38.0

33,0

3S.O

29,0

15,0

17..S

17,S

16.5

15,S

19,o

so.o

58,0

6,07

35.0
5,10

6.38

6.67

scn

m111

5,25

S,-15

S.51

18.I
84,0

l.

3.
4.

33,0
17,0

35,0+
18,0

50,0

40,0

5,61

6,30

35.0
18.0
60,0
6,30

26,S

2
3

21,-1

20,0

S.89

6,72

60,0
6,SO

27,0

27,0

27,0

27,8

28,0

18,0

16,5

17,9

16.0

4,86

4.48

4,97

4.4~

33,0

33,S
15,(}

17.0

S,04

~.60

28,S

33.0

34,0

35,0

33,0

16,0

18,S

IS,O

16,1

13.5

95,0

90.0

80.0

4,56

6.10

6,12

s.64

78,0
4,46

IS,5

17,l

M1

M112

28,0

17.0
52,0
4,76

28,1
17.S

.-

19.0

28,1
17,5
7:5,0

5.34

34.0

2.
3.

15.0

15,0

4.

5.10

60,0
4.93

M3
36.S
13,S

32,S

--

19.0

22.0

32,0

1~.o

M112

31.0

5.55

S,43

P31~

30.0
18,S

30,0

14.0
47,0

16,5

58,0
5.45

S.03

mat

36,S
15,0

70,0

56.0
5.43

1. iegntb, 2. breadth, 3. height, 4. occlusal swface measure (cm2), += germ (not yet e rup1ing; the tool.b crown was s tlll !\Unk
down in the alveolus)

Cheek teeth belonging to abrasion stage A.I. are

characterised by, beyond question, great hypselodonty.
Th.is feature corresponds well to large vaulted facial
skull, high maxilla (e.g., horse skufl from Pilismar6tBasabarc, VOROS l990a). The corpus mandibulae is

The differences in the height size of teeth is also influenced by the status of the tooth - i.e., milk teeth or
permanent teeth. The row of leeth of three juv./subad.
individuals show the differences weU (height of teielh
io mm):

P2

PJ

P4

Ml

M2

so 1

40 1

601

90

80

90
86

90

85

MJ

tnf.
sup.

75

genus permanent teeth under the deciduous teeth

unworn (withoul abras ion surface)

wide and high. The fast abrasion of teeth in stage A.ll.

shows that the resistiv ity of enamel to contemporary
vegetal nutrientS was low. In the case of this Eq11us
form, the inadequate hardness of the enamel is "compensated" by, for the teeth ex.posed to fast abrasion, by
hypselodonty.

88

Age

(years)

CC:i. 2
years
752

2- 2,5
3-3.5

Uni1
1L&3. 3. tr.
J.82. J . tr.
1.82. B sec. 8. D

Length and breadth ofteeth

The length and breadth size of the crown - tooth
column - detennine, apart from the absolute dim ensions of the teeth, the fonnation of the chewing surface, the "useful area" of the tooth as well. W.
Reichenau ( 191 5, I 05-) draw anention to the phe133

nomenon I.bat the high loorh crown of horses is getting

gradually narrow rowards the neck of the root Consequehtly, the length - width data of the chewing surface gradually decrease by the advance of age. The
" macrodontous" dentition becomes ..microdontous".
The width data of the crown, especially on the lingual
side, is also influenced by the thickness of tooth
cement. 60- 70 mm length decrease (tooth abrasion)
can result in decreasing lhe c hewing surface by
5-6mm2.

MI

between

them. The decrease of breadth of P2

and M3 tooth column is more significant at the upper

reeth than che lower ones.
Tooth finds of Equus at Henye are dominated by
abrasion stage I. and n. Among the teeth with known
heig ht, only 2 sup. and 3 inf . were found belonging to
abrasion stage UI. Jt is characteristic of the construction of the dental columns that dimensions of different abrasion stage Leeth can be similar, or s lightly different from each other (Tables 4.-5.). Similar phe-

Table 6. Length and breadth measurement data measured at rhe occlusal (or grinding) surface. Statistics on the
cheek teeth ofEquus found at Henye (in mm)
r

mean

min-max.

II

m~

38,5-39,0

26,5

min-ma.x.

mean

mm-rrtax.

30,43

29,0-32,0

1.i.45

l l.S-17.0

10

27,9 1

26,0-30.0

14,16

l2,0- 16.0

28.0-32,5

25,02

24,0-27 .0

13.78

11 , 1- 16,0

33,4-3&,0

17.05

15,0-18.2

33.0-35.0

:?

30,95

29,5-35,0

10

19,o6

17,0-22,0

29.08

36,5-35,0

16

17,28

15,5-19,0

34.42

33,0-36,5

14.68

13,5- 16.5

p2

p 3'4

12

31,95

29.5-35,0

Ml/2

10

30.25

26,0-33,0

M3

29,3

P2

35.-17

dp314

P314

M112

13

MJ

15.0-16,5

The length - breadth and height values of the teeth

are presented in Table 4-5.
lt seems that in the dental set of the Upper Pleistocene horses the lenglb - breadth dimensions of the
lower P2 and M3 are m o re unified on individual and
phyla- chronoevolutionary level than that of P3/4 and

nomenon could be observed in case of the horses

found at Sagvar, where an teeth belonged to A.ID .. but
apart from the normal "old age microdonts", there
were also mesodoat size teeth fow1d as weU (Tables
7.-8.). On species evolutionary leve l, " prismatic"
tooth with parallel walls and ..truncated pyramidal"

Table 7. Comparison of measurements of Equ11s upper teeth from Bodrog keresz.tllr and other UP sites (in mm)
'

lt:nglh

TeethlSiJc
D

'(

38,75

Proiocon length

bmidlh

minmax

:38.S-39,0
41,0
40.0
I
45.0
34.0-41.S

26,S

34.0-38,0
.36,0 41 ,5

min-ma.~

pl

Bodrogkertnttl r
Pillsmar61-Bh I

~velvlML
~

LL

1
2

40.00

~adapt>

37,85

tS;igvtlr7

36,S

Cllmbc Grcnal 8
Solu1re8

25

39.08
36.82

Willi:ndorf IL9
Ttldc 10

Sandalja nII E.-F

vH
Tcmn111n8 V.

134

39.t8

39,40

39,06

35)!- 38,7
34,5-36,S
365-41,2
39,l
38,HO,J

I
2
I
4

33

8
2
II

!S.38

24,55
26.48
24,86

27.67

26,00

25.74

28,0
27,11--"J
36,0
23,0-28,S
20.0-26.2
M.~29,0

23,2-'..6,0
25,G-27,0
25,8-29,8
26.8
22,5- 28.1

"

min-max

Table 7. cont. (2)

length

'll!eth/Site

!2
5

min-max

Protocon length
x

J0,4-J

29,0-32.0

14.45

11.5- 17,0

30,5-33,0
26,0- 33,0

5
5

13,90

12,4-18,5

31,50
29,00

16,~0

29,25

27 10- 31,0

14,23

14.5-17.5
13,0-15,0

bread lb

mln-max

31 ,95

29,6-35,0

33,00

min-max

~14

8<>drogkere11:tlir
PilJsmer6tBh 1

Pillsszilnr.6 IL2
Kiskuvclyl1JL

21.90

32,0- 33.5
)0,0-34,0

29,.U

27,0-32,0

LL

32,0-33,0

23,5-H,O

4
2

30,22

25, 1- 33,S

29,12

25,5-31 .5

15,00

9.5-18,0

3DJ 8

25.0-32,6

27.07

25,S- 29..0

14.00

12.5 16,5

Nedlip6

29,06

27,0- 33.S

26,67

26,0-18,0

3
4

14,12

!3.S~ l-45

Sl!gvar7

LI

29,01

25,6-31.U

II

27,ll

25,8-19,U

14

13,12

11.3-15,0

Cumhe: Grcnn!8

73

30.28

27,0-33.S

67

28,33

26,0-aR,0

lJ,79

11.!HR,O

28,60

26,5- 3 1.0

19

27.40

23,2-3{).0

20

13,54

12,0- 15,7

WilhmdorfTL9
Tilde1ff
Sandalja 11 11 Ef

20
2

29,3-29,5

37

30,80

26,$-35.3

18

211.80

24,1-33,4

J6

14,69

11.5-17.2

2 9,10

27.60

CHI

22

31.00

IS

28,00

-icrnn:u.,Rv

29.48

27.7-31.4

29.26

2S,0- 31.8

21.7-3 l.4

Bache Kiro 11

IS

27.10

24.5-28,8

13

28,30

25.9-3'1,0

Mndaras4
Plliw.llntOL5

Solutrc8

13,0-15,0

29,S-31.,0

11.4-~

243-28.8

Table 7. cont. (3)

't'!:<'thJSitc

length

Prot~on

br.cadlh

li!r!glh

mln-max

min-max

mlnma.\

Bndrogkr.l'C5Ztllr

10

J0.25

26,0-33,0

10

27,91

26,0-30,0

!)

Pilismonit-Bh 1
Pilisszilnro IJ.2

6
2

29,08

28,5- 30,0

30,21

2?,S-31,0

14,26
l.S,20

13,0- 18.Qo

23,2-31,0

Ktskevt!tr UL

26,17
24;l)4

2S,()-27,2

M112

Madaras

Piliss?iinlo

1.'
u.

[ i-:adap6
lsrigvar7
Combe Greoal8
Solwnl8

2
2
6
15
71
23

28,0

12,0~15,E>

13, !2

I l.J-15,0

26,93

24,5-29,S

72

14, 13

ll,0-17,0

26,15

24,0-29,5

22

14.39

13.0 16,9

23,1-'..9,2

16

14,60

12.i-16,9

14,78

13,1-15,9

60

24.5-29,0

20

28,40

.27,70
23,40

27,0-28,0

25,C>-26,5

26,4-34, l

16

26,90

26,50

13

26.50

27,51

25,C)-30,0

25,50

24,5- 27,0

ZJ,S-32,5
21.8-24,5

13,a-1s,o

2
14

23,0-30,0

16

25,8-29,0

25,99

G-ll

l-0,7- l5,5i

13.21

26,82

27.10

10,0-15,Si

l5

12

12,0;._16,Cl

13.5

28,C>-3 l,S
25,S.-28,S

27.20
27,12

SandaUs 11 I I E-r

I
2

29,4

24,4-25,0
2..'.S-29,5
24,4-27,0

16

21,C>-29,0

22,8-26,1
28,S.-29,0

Wllhmdorfll.9

26.5-30,0
24,1-28,1

27,36
25,22

111de 10

Tdino!!.Ul8 V.
.Bad10 Klru 12

2.8,12

12,0-16,0

135

Table 7. cont. (4)

length

Tepth/Sitc

lw
IBodrog);,Cl"C$Zllir
PiUsmeriit-Bh1

IKlskeve1Y3 Ml.
Nii!dap6

[sagvr7

IComb.e GrcOAl8
Solutr68

IWlllendorf lt.9
ln1<kro
SandnUa U11 E-1'
G-H

Temnata&v.

s
3
2

)(

mlnnla:<

'25,02

23,0...27,0

13,78

ll,l-16,0

24,0-28,0

IS,30

14.5-16,0

22,0-23.5

2
l

24.0-24.5

15,00

23,82

18,0-29,0

55

14,59

11.5-19,0

22.84

~23,0

14,80

14.4-15.5

2J.0-23.9

14,62

12.8- 16,2

23,0-24,6

18,0 32.5
23.5-31.5

23,0-29,0

25,0

25,0-26,0

22.67

2110...28,.0

23,0-32,.S
26,0-28,3
30,0-30.5
25,8-29,0

49
3

46

26,69

4
2
6
3

27.07

j'lliO!flal<

29.30
29,70
25.67

27,58
28.20
27.30

26,0-30,0

length

rqin-ma.~

3
2

Prot~n

bread1h
n

)(

15,0-L6,0
12.0

24,.5
7

21,84

3
5
2

22.20
22.10

l
14,5- 15. 7

1. VOR6S L990a. Tahl. l.; 2. VOROS L986b. 37; 3, VOROS 1994. Tab. J .; 4. VOROS 1989. Tab. 2.; 5. VOROS 1987c.43.;
6. VOROS 1988b. Tabl. 2.; 7. VOROS 1982. Tabl. !.; 8. DELPECH-GUADELU 1992. Tabl. 28, 31-35 .; 9. THENTUS.
1959. L60.: 10. IUEDEL 1980. Tab!. I.; 11. f.ORSTGN 1990. Tab!. 2.; 12. FORSTEN 1982. Tabl. I.

Table 8. Comparison of measurements of Equus lower teeth :from Bodrngkeresztllr and other UP sites (in. mm)
~

breadth

lc11gth

Teeilli'Site
n

min-max.

min-max

35.47
38,SO

33.4-38,(1

4
I

35,00

5
II

17,05
18,05
16,20
17,02

15,0- 18,2

38,5

'P2
Bodrog1'crozmr
Ki$keve!y C. LL2

' Nadap4

5
11
I

34,09

31,0-38.0
30,..C.....36,0

34,00

34,0

49

34.1 7

32,87

29,G-38,0
29,0-.35,6

SI

16

34.:55
35,S

30,3-37,9

SruJdaUn II C. E-F8
G-H

6
4
6

Tem1w1a C.6

Sa&vru-5
Pilf=MIJS II. Rsh. 1
C1;1mbc Gn:uat 6
Sol11tre6
Tilde Ci: Ti7

17
2

33,3-35.S

5
6

35,6

34,64

18,5
15,0-IX.O

16.0
15,39

15.0-1 M
16,0
13,0-18.0

15.12
16,76

135- 18,0
16.4-1$.4

IS,00
1:5,+

16,S

16,0- 111,()

Table 8. cont. (2)

breadtl1

leQglh

T.;cth/Sitc

mu1-mnx

10

19,()

17,0-22,0

5
4

18,04
20,75

20,5-21,0

28,(}-32,0

17,04

16,0-19.0

27..2- 30,0

16,5-19,0

28,6!i

25,0-32,0

18

17,56
18,70

uo

29.97

25,5'-34,0

112

17,54

15.0 21,0

32
24

2.&,65

26,0--30,7

32

16,55

14.J-t8,3

31,.50

26,ll-3.S,2

24

18,2

15,5-21,2

~.10

JI

16,6

6
4

29,IO

15,IJ

19,92

28,S:-J 1,.S

16,02

10

28,80

25,()..12,7

lO

16,20

11

Jt

mirt-mM

9
5

3 1,95

29,0-35,0

Pili!;5zlint6 rt. Rsh. 1

Ki.skcY<!ly v. u .2

J0,80

28T0-32.,0
30,0-33,0

Mllilarrui ML- LL3

29,50

N$p~

il.8,84

Soigv;\r5

7
19

Com1>eo~16
Soh.1tn:6

P314

Bodrogknrcnttir

nld~ C T[1
Snndalja n c. E-P

G-11
"Jemnalll c.6

Bacho Kiro C.~

136

Jl,05

11i,S- 2 1,0

17,0-21,0

13.2-1715
14,5-17,7

Table 8, cont. (3)

lcnglh

reetb..ISire

bre11dlh

11

l(

min. max

Bvdrol!-k~rtntli.r

13

29,08

l 7,18

16,S- 35,0
26,7-21,0

16

Kisll.ll\itly C. UL2

Mv,
LL

2
7

28,5-30,0

28,57

25,5-34,0

Pnis.miniO a. Rsh. I

28, 10

Sngv~

11

25,46

27,3- 29,0
23,6---27,0

6
3
10

Marlnras U.3

Comb"' OrenaJ6

124

Snl111r~6

27

11l~CT1 7

21
3

Sandalj11 IJ C. -p8

27,78
27, 10
31,70

27,3

27,40

IBuchoi Kiro c.9

II

25,5

G-11

15,68

IJ,()-11,5

18.40
16,80

18,0-20.5
15,0-19,J
13,0-IS,S
13,0-18,0
12,7 16,0

122

15.67

26

14,54
16,30

21

u..1

l Tc:mnata c.6

1115-19.6
18.0-20.5

23,0-25,0
1 1,5-Jl,O
24,6--29.S
26,6-36,2

l5,.5- 19,0

2.

NMap4

min-1na.x

14.5
14.7

16,11

13

15.5

14.8- 18,2

25,3-29,0
24.0-28,6

- 1~
mi" rnax

min-m"x

14,68
15.0

1.3,5-16.S

15,0-~

13,8- 16,8

Table 8. cont. (4)

length

Teetlv'Site

M3
Bodrogkcrumir
Ki.skcvcly C. UL2
ll
MadanisML- LL3

breadth

34.42

33,0-M,S

34,o

)4,0

Jr,i>-3&.o

IS.O
1'7.0- 18,0

34,0-36,0

Nadap

34.0

34,0

t4.0

Sagvar5

14

32.15

29.0-34,0

12

PUissT.ilnl6 ll. ~h. I

30,0

30,0

14.15
13,0

30.0-3&.0
285-'.li;.5

48

1.3.93

13.71

l ;?,5-15,5

29.'2-l3.5

14,l
13.2

13,6-14,4

11

Comb~ Gn:rut16

4.8

33.48

Solutre6

')

TIJ~CT17
S11ndttljtt JI C. EFS

31.9'2
3 1,70

33,9

31,'.l

34,53

G-.t.!
TemruUaC.6

3
J2.S-36,0

13,0- 13,6

12,2

t4,87

_ _1_4,0

U ,0-16,0
13,()
I

12..0-16.P

13,3-15,6

1. VOROS 19S6b. 37; 2. VOROS 1994. 2-3.; 3. VOR6S 1989. Tab!. 2.; 4. VOROS L988b. Tabl. 3.; 5. VOROS 1982. Tab!.
2.; 6. DELPcCH-GUADELLI 1992. Tabl. 36-42.; 7. RJEDEL 19110. 32- 33. Tabt l.; 8. FORSTEN 1990. Tab!. 2.; 9. FORSTEN 1982. Tabl. l.
tooth with narrowing wall towards the t0oth neck can.
occur as individual variation. The facial skull (maxilla) of the horses w ith " prismatic" teeth and the cor
pus of the mandible, however, had to be more robust.
The deposit ion of layers of tooth cement is less significant in case of the upper teeth while it is considerable on the labial side of the lower teeth.
The dentition, tooth size of the Heqye hill Equus
finds correspond to the large mesodont horses of the
Upper Pleistocene W 1-2 pe riod (Tables 7.-8.). One
item, a sup. Ml/2 (A .1.) differs from the rest of the

Henye hill Equus fends in its small dimensions (26 x

26 Pch. 14 mm) (Table 4.).

Occlusal (or grinding) swface in square centimetres

The above dental parameters, i.e. the product of
length and breadth can give the " useful" chewing area
of the tooth. This is rhe surface where the animal,
during nutrition. grinds vegetal food. Dimensions of
the occlusal (or gcioding) surface on the cheek teeth
are given below in square ems for the Equus finds at
Henye:

IJ7

mean

min-max.

Upper teeth

p2

10,2-10,4

p :i/4

9,72

M112

p 314_ Mli'2

M3

10

17

9.0-1 ' ~2

8.43

6.7- 93

8,96

6,7- IJ,2

7.34

6.7-8.8

mean

min.-mlllL

A-I-Ill.

p l.l\'(3

17

8. 96

6,7- 11 ,2

A.LII.

p Y4

9,&5

9.J- 11 ,2

A.ID.

pV4

9,0

;U-W.

pll4

A.L
A.1.-11.

M l/1

Ml/2
Ml/2

9.0
9.'12
6,7
li,63

7.9-9,3

10

8,d3

6,7-9.3

p J.14

8,32

7.9-ll.7

1Jl4

i;l

7,2

Mill

7,0

1,0

pl.M2

8
4
4

7.!12
S.S2

'6,7-8.7

7,32

6. 7- 7,9

pJ.M2

7 ,58

6,!>-8,4

p311

7,48

Mm

7.65

-6,9- 8.4

IP3M2

26

7,43

6.<Hl.6

1 1'3'~

II
IS

8.24

6.6-9 .6

6,S3

6.0-7.6

Bodroekutnulr-Hmiyr

A.1. -TI.
J'1lillszdnr6 1. R!h. 5
A.1.-11
A.Ill

Lowc:r icclh

.Pi

P 314

M112

,_PJ14-M l/1

M3

5.l-6.6

5.98

5.4-6,7

p3/4
Mtll

A.1.-IL

5.o?

4,4-6.1

22

5.44

4,4-6,7

5,08

Table 9. Comparison of occlusal surface (in cm 2) of

upper ceeth of Equus from similar chronolOgical l'eriod sites in Hungary
Pilism11r6t-Bruroharc 1
Al.

n
pl-M?

II

pl/4

5
6

M1 12

M"lldaras ML-LL2
A.1.-ll.

9,45
10.30

8,75

9.30

7,3-10.4

9,91

9.S-10,4

Mill
p3.M2

pJ/4

2
2

!P,ilimiinl6 11. R~h.~

!

A.L-IL

M'a
pl.Mz.

)>JJ.I

Ml f2
A .UL

138

Mrn

l
I

A.Ill.

Mlll

[n the Hungarian Upper Pleistocene Equus populations, lhe trend of changes in the grinding surface is
the following (data in,cm2):
Upper teeth - -p3.M2
A..1-11.

S1Le

mean

Pilil>lnn1'6t-Bnsahnre

9 .45

II

Madaras Lower Levcl-Mtddlc Lew i

930

Pil!SSTAnl6 Il. Roclcsheltcr

Ki~kevely Cot\'\:, Lower Lovd
:Bodros;J<cres:ciur-Hcnye
P!J.lssz:3nt6 I. Rc>ck.,ht:lll.'I.,.

9.28

Ki~kevely

A.Ill.

7,3

Cave, Urmcr L~vel

17

8,32

7,92

7.58

7.43

26

flll."tdlltllS Lower l.evcl

6.15

l'iliSSZllnt6 I. R'kshchcr
Pili5cl1116 II. Rocblwlter

7.1 2

6.8

s,s

9.2'1

8 ,1 -1().9

Lower teeth - P3-M2

9,2-1 0,.9
8. 1-8,.9

A:.I-n:

9,3
6,1

8,96

Sl\gvdr

5,50

9,2S

9;1.7

Nu.clap

6,80

9,2.8

I
I

7.0~8.4

I. VOROS l990a Table I.; 2. VOROS 1989 Table 2.; J. VOROS

L994 Table J .; 4. VOROS J986b 37.; 5. VOROS 1987c 43.; 6.
VOR6S 1988b Table 2.; 7. VOROS l982. Table 3.

8.4-'9.')

lillkevcly c. LL3
A.1.-11.

--

&,4-9,9

3
l
1
l

7,30

S~ar7

9.9-11,0

pJ!~

P3i4

A.l.lL

min"max.

P3-M2
M 112

A.rt!.

mean

R. Hl,7

Nn'1ap6

4.4-5.4

ln case of the "pyramidal" teeth, the great variation

in the grinding surface can be attributed to different
abrasion stages. ln case of these teeth. by A. ill.,
grinding surface can decrease almost to half of the
original. ln cechnical literature, the mean value of the
sum of lower P3 and M2 teeth is also used for the
characterisation of the Equus population of a region or
a site. As there is significant difference wirhin the
same Abrasion stage between the praemolars and che
molars as well as Ll1e grinding surfaces in different
abrasion stages, the Equus tooth find of Hungari'an
sites were treated separately (Tables 9.-10.)

6.7

Kb kcvClyC. UL3

5,13

13

9.0

6.10

+onlyP3f4

Sltc

mean

KiskeYc!ly Ct!ve. Lower level

6,17

Bodroglrercmur-Htllyc

5,44

8,1-11,l

Pili.swlllt6 ll. Roclc:shclU."I'

5,43

8,1-11,l

Milchuas-Middle Lcvct'

S.Ct4

12
It
5

9,3

Nt1dnp

4,98

II

6, 1

KiSkiwely Cave, Upper Level

4.96

Lower teeth - P3-M1

A.Ill

!Site
ISdgvi\r

fll(!aD

4,53

28

INadup

'.l,66

3,56

IM0dnras Lower Level

achenheimensis is somewhat higher: sup. 9,75 Limil

8,6- 10,.9, inf. 5.7 Jimit 4,6-6,3; I.hat of EquU remagensi6 sup. cca. 8,0 and Equus solutreensis sup.
cca. 8.05 inf. cca. 5,3 are a little lower than those of
the horses ofHenye (RIEDEL J980. 36, NOBIS l 971.
39, 41 , MUSIL 1968 a.).

l11e dimension of the grinding smface Henye hill

horses, both for the upper and lower reeth, is typically
Iarge medium -medium value: for the upper p3.M2,
the average is 8,96, while that of the lower P1-M2 is
5,44. The dimens:ions and average of the grinding s urface P3 - M2 ofthe H enyehillhorses show greatsimilarity to quus teeth at Tilde near Triest sup. 8,48
Limn 6;99-11.22; inf. 5, 13 Umit 4,48-7,0 (RIEDEL
198@. 36) The grinding surface average of Equus

Protocon.us /engch (in mm)

n
9
9
4

p3f4

Mlll

Ml

mewi
14,.50

min-max.
11..5-17,0

14,25

12,0-16.0

13.78

11.1- 16,0

The mean value of p3JM2 Pc length average of the

EquusfmdsatHenyehill is 14,35 mm .

Table l 0. Comparison of ocdusal surface ( in cm2) of lower Leeth of Equus from similar chronological perios
siles in Hungary
K.WcvC!y c. u 3

m~n

mir\-ma.it.

P3M3

r~'4

6,17
6,42

6,3~7

M112

A.L.llJ.

l'1.M2

4,4-6,7

P114

22
9

5,44

ALIL

5,98

5,4-6,7

AJlfl.

Mrn
M1tz

l3

5,07

4,~. I

4.48

4,4
5,0--5.9

AJ.-II.

S.l~.7

S.l~.2

Bqdrngl;~llir-Bcnye

J\ .IIL
Pifusztlnt6 ILRsb.4

5,43

P114

.5

5,59

5.2-S,9

M1h

5,15

5;0-5.2

Pl}I

5,04

4,-t- 6,0

?114

4,0

M1n_

P3M2

11

4,98

Py 4

5.14

4,6 5.9

M112

7
4

4,68

4,2 S,2

A.lIJ.

M112

A.1.- Il.

M112

AJIJ.

PrM2

28

4,53

1'3/4

1.8

4,67

4,7-6,7

M1n

10

4.U

3.S-.S,l

A.lIL

P;rM2

M11dw-as ML-LL1
A.t.JL
AJIL

4.0

---

2,9-3,8

Nadup6
Al.11.

K ik~,~I~ C. UL3

--

4,3- 5.9

3,3-3,9

4,7-5,2

S(lgv6.r7
3,5--6,7

2 . VOROS 1989. Table2.; 3. VOROS 1994. Table3.; 4. VOROS 1986b. 37.; 6. VOR6S 1988b. Table 2.; 7. VOROS 1982.
Table 3.

139

Among the investigated Hungarian Upper Pleisto

cene populations, the average of pJM2 Pc length values are given below, in decreasing order (mm):

A 1.-11
l'flismar6t- Basabarc
Dodrogkeresztiir- Henye

Prllssi.io.to I. Rocl;shdLcr
A L-IL

IDCllll

min-max.

Madaras LL-.'vflJ
l'iliss:ainio U. l(()(;ki;helrer

16.50

"6

l 14,S-17,5

16.)4

15,0-18,U

l'ilismar6t-Ba~eltarc

14,62

11

12.4-ll!,5

Oodrogkcrcszhlr-Rtnyo;

1-1.35

18

ll,5-17,0

Xiskcvdly CaYe, UL

13,86

10,()-.J6,0

Nadap

13.72

.8

12,l-!5.0

Pillsszanl6 L Rock.sh~ltcr

12, 17

11.0-13.0

Am

mcail

min-roa.'I.

sag~nT

13,52

25
I
I

11,3 .. 17,0

9.5-10.7

Pillss-r.fin10 I. Rockshelter
PUissi.An16 IJ. Ro~kshdt1.-r
Mlldams LL
,__

10. 10

16.5
12.0

The higher valties of Pc length (16-18 mm) occur

dominantly at the p 314 teelh. Pc dimensions for this
tooth do not change much WiLh age: even in stage
A.m.. they remain about the same: see Sagvar or
Pi I isszant6 I. Rockshelter.
The Pc length dimensions of the Henye borses
again show the greatest similarity to the EqLw~. borses
at Tilde Cave Ti: mean 14, 7, fur .n = 52> limit
U,5-17,2 (RIEDEL 19.80. 42. Tabl.1., 6,.).
The Equus Pc length of S.andalja U.Cave. E-F.
L.evel waS found to be lh.e folio.wing: mean 13,6, n 14, G-H. Level mean I 4,2. n - 33 (FORSTEN 1990.
Tabl. 1.).
p3 - M2 Pc length of Eql.Jus achenheimensis was
between 13,9-17,8 mm (NOBCS 197 1. 36., 41).

Protocom1s index (Pc length I length of the grinding

surface)
mcar1

min-max.

Pi-Jjsszrun6 r. RtickshellCT

PL~ U. Rcx:~shelter

M:adnra.s LL

/\ t-Jl.

33.8-52,3

47.7!

31,5-54.3

Pi llss1An16 IT

Ml

47,07

43.2-57,2

Pili,manl6 U. Rockshcllci
Klskc,ely Cnvc:, UL

Na.cfup

140

48,4-53.8

50,16

-10,IHIO.l

48,85

44.4-53,7

mh1-111ax.

J 4 l,3--60,(1

I
I

41,0

66,0
)1,7

] 35.1-46.9

Values of the Pc-lndex. correlate qllite well with P.c

length values (e.g., Madaras 59,4; Sagvar 60,0;
Kiskevely Cave UL 60,3; Pilismar6t-Basaharc 63, l ;
Pllisszant6 1. Rockshelter UL 66.0.)
By simil.ar Pc length, the higher Pc-Index values
appear at the shorter teeth.
The Pc-lndex value of lhe Hen ye hill horses, 46,47
belong to the medium values, low values are found at
the teed1 with shorl Pc, high values are characteristic
oftbe teeth wlth short crown.
The dentition of Upper Pleistocene horses studied
so far in HWlgary cannot be considered uniform. because of the variation in dimensi.onal parameters. In
case of the horses with sjmilar geological and individual age (same Abrasion s1.age), di.inensions of Pm
and M as well as !.he grinding surface size are basically s imilar, but tooth find occur in both higher and
lower domains. The general trend of changes .ih tooth
dimensions and mean values show a decreas.ing trend
by lhe advance of geologica.I age (FORSTEN J 988.,
1991.)
The dentition of the Late Upper Pleistocene horses
in Hungary can be classified, on the basis of si.ze categories, as fo llows:

45,23

Mud~LL-ML

33,8-54,3

LL; Lower Level, ML: Middle Ll!vel, UP: Upper Level

min-max.
44,8- 59,4

35,l-JI0,0

25

mean

p3l~

mean
53,12
51,40

46,-1,7

37,90

50,56

M'n

A L-11.

min,mnx.
37,0 63,1

A TTL

Pilismur6t- Bas;llwc
Mndiir.is Ul-LL

The p3 - M2 Pc Pc-Index average for the Henye hill

horses was 46,47. large meditim size. The P.c-lndex
average of p3 - M2 among the investigated Hungarian
Upper Pleistocene populations are given below, in decreasing order (mm):

11
18

S4vlir

Bt1drogkcren11lr-Benye
n

mean
47,67

Rookshelter LI.
Kiskcvc,\Jy CA_..c, Lt.

Protoetlnus
M

Ttttb
L
L
L
I.

L
L

'L

Sh

,..,,

l.

Kiskevely Ctwo, UL
l'Hiss:cln!O I. Rooksbchcr LL

L
L

Sh

N'ildnp

$h

SmJ M
Sm

Sh

L
L

Sh

AIU.

S:i&var
Pilissz8n16 I. 'R.ockshelter UL
l'llissz5.n16 II. Rookshelter UL
M.adllmSLL

Sm
Sm

L
l

Sh

L: Large/long, M: Medium, Sm : Small, Sh: Short

UL : Upper Level, LL: L-0wer Level

1:

Investigating fhe extreme values of the tooth dimension range we find that by relatively constant
higher val.ues, the variation of the lower values is very
large (Tables 7-8).

"Macrodontous" teeth, dominant in the MiddJe Upper Pleistocene, can be found still, though rai:ely (e.g.,
Madaras, Sagvar). The teeth of the Latter period are
more hypselodo"' the walls ofthe prismatic crown are
parallel. Their size is hardly decreasing, even by
abrasion stage A. ill. Such fonns are Eqirus from
Bodrogkereszn1r-Henye (Table 4-5.) and those of
sediments under layer C-6 in Kulna Cave. These are
medium size horses, dentition assigned to the
form E. scythicus (MUSIL 1990, Tab!. 2., 7 ., 14., 33.,
41 . 47).
One ofits last occurrences is known from the Dryas
IT. period, from layer VIII. of Kniegrotte (MUSIL
J 974. Tabl. 2-23).
The evaluation and interpretation of the lower
values of tooth djmension is more difficult. "Microdontous,, teeth found on the sites can originate from
real "small" teeth as well as 1senilis., (A.III.) horses.
To decide this question is no problem in an assemblage o.f optimal distribution. Tf the " microdontous"
teeth came forth only from A.m. scage, it is not knowable if they originated from small teeth or justrema.ins
of "pyramidal" teeth. In case when the same layer
yi.elded ''macrodontous" A.UI. teeth as well, the difference is evident the fonn Equus TT. "small horse"
and E. /A./ cf. hydruntinus was' separated on this basis
at several Hungarian Upper Palaeolithic sites (Sagvar,
Nadap, Madaras, Pilisszam6 1.-11. Rockshelters). Ler
us note that these "microdontous" teeth appear quite
often with totally abraded grinding surface, worn deep
till rhe roots (Madaras: VOROS 1989. Fig. 2. 2- 7.,
Pilisszant6 n. Rockshelter: VOROS 19&6. b. Fig. 4.4.,
Nadap: VOROS 19&8. b. 34.). They show that their
e namel was less hard and resistant to contemporary
vegetal nutrients.
Horses with senile" m.icrod0ntous teeth appear in
Hungary at Madaras Lower Layer and the Medium
Layer of the Pilisszant6 I. Rockshelter during the
WII/DJ interstadial (VOROS 1989. 1987 c.).
On some sites, though in low number, there w,ere
small and small medium size teeth found as well. One
example was found at Henye hill; one dexl. M 112 (82.
11. 4. lr. l . D, Table 4.)
The pre.sence of real microdontous small horses
during the Late Upper Pleistocene is marked, apart
from the asi11ien, by the Eq1ms tooLb find of the North
Bulgarian Bacho Kiro Cave (W TI/ill interstadial,
PORSTEN 1982. Tab!. l .) as well as the afore mentioned Kniegrotte Cave VI ll. from the Dryas ll. period, where microdontous leelh were found al.so in
Abrasion stage 1. (MUSIL 1974. 2- 23).

All BodrogkeresztUr-Henye teeth belonged to the

large medium category wrth the exception of the one
small medium piece. Tooth remains from small size
Equus aod Asinus were missing from this Upper Pa-

laeolithic campsile.
3.1.3. Postc.raniaJ skeleton elements
72,8% (352 pieces) of the Equus finds from the
Henye site were postcranial bone remains (Table 3 .).
this is the highest number on the Upper PaJaeolithic
sites in Hungary, but due to its fragmentary character,
only I I ,0% could be measures.
The osteometricaJ data of the bones, measw-e<l according to the methods published by J. U. DUERST
{l926.) and A. van den DRIESCH (1976.) are presented on Table 11.

Table JJ. Bodrogkeresztitr-Henye. Measurement of

postcranial bones of Equus (in mm)
I.

2.

95
-

-1.

5.

3.

6.

7.

56

so

35

48

R.'ldius

-- -

61

30

42

28

.m

24,5

40)C

st.~

Metac:irpus

58

4()

56

26,5

41

56

.J-0

')0

!lO

56
SS

58

'tibia

8:S
80

51

SO

60

54

53

28

40

97
91
90

38

54

21

30

64
-

41
41

55

41

22

23

50

39

21

24

Os phalengls I.

posL

53 Guv.)

S2
S4
S2

Mu1atarsus

llnl..

92

91

40

52

43

16

46,5

85

28

28

22
18

2.7

01 phnllmgls U.

nnt.

47

65

56

61

35

26

30

46

64

54

S'.1

35

Z5

30

35

Ml

pCXL

I.

'.I.

48

61

51

55

~5

47

"

57.5
50

63.S

;\srragalus
I

l.

8.

9.

68

60

128

63
54

70

5.

6.

34..5

25

3<1

ll.5

38

25

Measurements w.ere made on br~dlb and diameter

of prox. andlor dist epiph. of Jong bones a.nd length breadth dimens ions of os Pb. I. ll. Long bones belong
typically Lo Lhe large medium category wilh 5 exceptions.

7.

29

These exceptions are tbe smallest within the assemblage. They are the following:
s. me. dist ( I. 82. C. sec. 5. D).
2 s. calc. (T. 82. C. sec. 5. O; 1. 82. 6. er.
os ph. I. s. post. (1. 82. B. sec. 9. 0 )
os ph. 11. d. post. (T. 82. C. sec. 5. 0 )

t 63

Calcaoac.

- -

60

1. length, 2. prox. cpiph. breadlh, 3. diaph. min. breadth, 4.

dist. epiph. breadth. 5. prox. piph. diameter, 6. cliapb. min.
diameter, 7. disl. epiph. diarnerer, 8. breadth, 9. heigh

Their dimensions belong to the small medium size

category (fable 11).
Comparing the Equus bone dimensions on some

Upper Pleistocene s ites we can observe the follow ing

(Table 12):
Table 12. Comparison of measurement ()fpostcranial bones of Eqrws (in mm) from other Upper Palaeolithic

Sites
RndiuJ pnu. br.:.adth
Bodrugkcxcv :lurl lcnye
1
K~evi!l) C 1

mean

95,0

mln max.
95,0
8.t,0-85.0

dlnmtter

Pilillz:int6 II. lbh.3

Achathctm6

l r.1dc c

r,

1-1

87.8

92.8

IKulnaC 1
Bolltngl\~1iir-l lcnyc

~LLl

l'ilde C

57,11

6
I

S4,3

56,0-01,0
SJ.0-56,0

ss.o

ss.o

51.0

48,0-SS,O
49.0-{)0,0
52,4-58,9

r 1s

KillruiC.7
rndc C. r 11
1Mctalsl'iius pros. breadth

27

55.l
55,3

10

Tlhill
,
_dlt. bn:ndlh

Bodrogkcremllr-Henyc
Mrulnms LL2
PilL'Winto L Rsb.1
Kiskevely c 1
Achmhc1m6

2
l

--

86.2

77.0

24

76.7
88. 1

84,1

70.0-80.0
82.0-96.0
73,0-90,0
80,0 92,0

60.0

56.0

44

r~c.r,s

15

lsUUlu~ko
,___.

c. uc1.

dlsL breadth
-MellltnruJ
Bodrogkercttturl l""Yc-

Kiikevcly c. 1

~nhaim6
ITllW! c:. 118

142

80,<>-90,0
79,0 82,0

77,0

~-i
Pllimu1r61-Dl6s~

47,0

49,0

minmn.
-18,0- 36.0
-16,0-48.0
I

47.0

47,4-53.2

40,3 4 7.7 -

40,6
40,2

40,0-42,0

6
2

37,5

36,0-39,0

LI

40.1

38,2-12,S

S3,0

50,(l- 56,0

37,~2.5

diameter

Kiskcvcly c.1
Ach.cnbcim6

Bodro~c~7tlirl ltt11) c

SI A-

diameter

Kllt1:\'ely c 1
Pilw7,Sn16 l1. !Uh.l
Ach<:nhe1m 6

mean

3
'!

80.0-95,0
88,5-97,0
SU. 89,4

Me1ncnrp11a dllit. brc11dth

52.0-54.n

4 7,S

~ 7.5

48,0

46.0-SO,O

Sl.2

52.S-54,0

60,0
SJ,0- 58.0

58.1

46,0-68,0

55,7

52,5-62.J

SJ.2-62.3
I

49.0
51i.O

49.0
S8,0

6
41

53.0
56.2
S6.2

51.0
55,0-58,0

12

56,5

54,1-58.9

djan1ctcr
I

I
6

40.0
42.6

II

41.0

40,0

40,n-t4,0

46.~.o

31!,X-..Jl.2

Table 11. cont.

o~

pb. l..ant+poil length

BollrogkcrcSZ!llr-t!eh:)'i:
Kis~e\~J) C. UL 1

li
Pill~znnt6 Ii R511.J
! PiJ\s.'*116 I. JQ;IJ.~
!1.chcnhcJm6
K1JhU C.7
T1l~c C. T 18

mean

min-mu.

91,0

85,0-97,0

11

86,S

78.()'.92.,0

91.0

87,0-100

8!1,0

80,0

8*,0
80.0

39

93.8

88,0-Hll

24

90, 1

86.0..95,0

Bodrogkeres:z:nir-Nc.nyc

64,0

64,0

Ki~keve.ly C LiL1

17
7

60,6

53.0-64,0

65,4

62.0-67,0

62,U

62,0

60,~

60JJ

Achcnhcim6

37

63,6

58.0-72.5

Kulua (' 7
litdc c. T18
di11ph. brc11d1l1

21

dhuq~cr

I
I

91.2- 9.l,6
I

LL

Pltlss1.Mt6 U. l{Sh. 3
'Pillss1.a1116 I. Rsh.~

Oodrop,kercmiir-UJ:nyc

min-maL

prox. breadth

IRlskev<ly C. UL I

m e,:u1

55,2-63,9
61,7

55.0--69,0

39.2

36,0-41.0

17

38.4
42,4

40,1)-45.0

P ilissz.alllD IL Rsh. 1
l'ilis.mim6 L Rsh.4

36,0-41.0

39.5

39.5
35,0
37..S-47.0
3'6,5--44 ,9

AcnQJ\hc!m0

39

35.0
41.9

llldt<C. T 18

2.1

40.2

r,

di~ I. breadth

._Bodrogktte!ZLUr-Hcnyc

Kilik~vcly

c. Ul I
LL

ll'ilismim.ii ll Rsn.3
l'ills!>Z<ln1.ii I Rsh.~

51.5

46.5-55.0

50,6

45,5-Sl,O

54,4

51.0-57,0

49.5
48,0

49.S
48.0
41.1-50,9

48.6

45.0-51.9

1
2

Kitlrui C.i

l111dc c. T 1a

2.~

I
I
I

<

~h. lL 11n1+pclJI length

4

I
I
I
I

47.0

46.0-4R,()

44,4

42,ll-47,0

48,3

46,(}- 5),()

45,0

45,0

~10 1 Rsh.~

41 ,0

CQmbe-Or~al9

Solui.r\!0

25

drn gkerci;Wil'-Rcnyc

l<is~vc!ly C Ul 1
LL
Pil~.dUlrull R.\!h:3

Acho.nhcim

]"ildeC:..: !/
Rulna C 7
lcmr;;;;a<.:.9

8
7

'

41 .0

35.5-42.6

l4

38.6
41.0

15

40.9

'37.S-44.4

5 0.0-55.9

--

- --,

38.0-475

43.9

43,9

61.7

55.0-65.0

56.8

53,0-60,0

62,2
5S,u

59,0-62.0

u.
' Pilissl:Jin16 n. R.sh.3

39,0-39.()

prox. brMdch
' Bodrogke(eswir-J [cnyc
IKrskcvcly C. Ul. 1

I
I

S.R,O

'

143

Table 12. cont.

pro:i:. brctldlh

mean

mln-mu.

11 il~zan10 1. Rsh."

57,0

S?,O

1
66

56..3

56.5-58.0
51.0-62,0

I~

58.9

54.0-66.5

(J

57.6

S0.0-64,7
60,8 {>4,7

C'ombt-Cirenalq

SohJlrc9
,____

Ac~hcim
Taldc c T 18
KWna<.:.7

Tcmruuac 9

63,1

63.1

51.5

45.0-56,0
43,0 Sl,0
.50,1>-SS.O
47,.S

diarnctcr

mln-mu.

mean

---

1Jlapb. breadth

Bodrug1'crcsnur-llcnyc
'K1s11.eCl)I c. UL 1
LL

8
1

52,1

Pillsr.7.An!O II Rsh.3

>11,5

l'ilisuant6 LRsll.4

-IS,0

2
70
14

46.7
J9,2

.l-1,4-52,S

48.2

411.2

Comoo-Cim1alq
Soh1trc!9

l1kl<:C T1 8
TernnllL, C.9

411.4

I
I
I

45.0
~5-47,0

43,0- 54,0

di.fL breadth

Bodrogkae!>Ztfu'-Heny.:

56.A

S?,S 61,0

Ki.skcH!ly C. UI)
LL

53,3
55,3

47,0-56,j
54,0-59.0

Pill!>SzAnl6 a. Rsh.3

I
I
2
67
13

l'iliSS7.iint6 I. Rsh.
.____

Cornhe-Grenal9

Solutr.!9

~ldcC. T 13
, lcmnata C.q

51,0

51,0

50,0

so,o

I
I

49.S-Sl.O
50,3

51.6
$4,2

47,0-S4.2
45,b-SS,.S
54,2

I
I

t. VOROS 1994. Table 4.; 2. V6ROS 1989. Table 2.; 3. VOROS 1986b. Table 2.; 4. VOROS 1987c. Table.; 5. V6ROS
198lc. 22.; 6. NOBIS 1971. Tab. u m., LXTX- LXJX.; 7. MUSIL 1990. 25-26.; 8. RIEDEL 1980. Table ll.; 9.
DELPECH-GUADELLI 1992. Table 48.

Dimensions of radius, metacarpus, tibia dist.

e pipb., metatarsus prox . and dist. epiph. as welJ as os
.Qh_Lfrom Henyc horses agree most with those of
K iskevely Cave LL (VOROS 1994.) and medlum size
Equus from Kulna Cave (MUSU. 1990.), as well as
the Achenheim form (NOBlS 197 1.) and the big
horses of Tilde Cave T1 (RIEDEL 1980.) Length of
the os ph. Il. agrees with those of the large horses in
Kiskeve ly Cave LL and Achcnheim horses. The same
bone is longer in case of the medfom large horses in
Kulna Cave and shorter on all other site'S (Table J2.)
The postcranial bone dimensions of the Kniegrottc
Cave VllJ by the end of the Late Pleist0cene with micro- and macrodontous teeth are smaller than the
llcnye finds (MUSIL 1974. Table 24).
lt is an interesting phenomeno n that whiJe I.he dimensions of os ph. T. correlate well with the long
bones, that is they change accordingly, that of os. ph.
11. a llow more differences: small and medium size
144

forms bul also phalangees with wide prox. epiphysis

coul.d equalJy appear.
Withers he ight
There were no complete long bones preserved at
th e Reny e. We can estimate the lengm of the long
bones adequately from the width of the prox. and dist.
epiphyses. Thus, typical long bone data can be reconsr:ructed accordingly :
cca. 340- 350 mm
radius
cca. 240-250 mm
metacarpus
cca. 350-370 mm
tibia
cca. 280-290 mm
metatarsus
The withers height based on I.he calculations of lhe
lo.ng bo ne values, elaborated on domestic horse materia I by Vitt (1952.) are gi ven bdow in cm. This
method is applied here for lhe sake of unifonn comparison (in om).

wnrm m
fu;cd-Oth.tlom

Mc

Mi

141.3

Il'ili&s7.dnt6 IT. R.ochhchcr

IJS,5-142,4

ISoluln!1

1285-1 45,J

127,4- 145.0

138,6-144.0

137,S-140,2

W. llilIHrT
I !Gske\'~ly Ca"c UL
W1llcndorf I.2 (eel.I
WUnn IJ/1lJ U
~ogkcre$Zllir-Jienyc.
1

1311.6
(=i) J47,0-15l,0

146,0-152,0

1'10.2--154.6

141,)-157.3

1+4.5-150,9

146.1 IS0.9

Rcmogen
Tiide Cave 1 T1
Achcnheim UL 1

149.3-160.0

IS0,0-1.SB,4

Ache11hc1m LL

148.0 155.2

IS4.6-16S.S

148,8

147,2-1 SS,8

151.4 153.5

l '3.4-1 54.5

WOnn 1-11
~skcvcly Cave I.I.
>ovnca1

( 1. RIEDEL 1980. rig. I.. 2.. 2 THENIUS l<.>59. 160., 3. RAKOVEC 1965. Tabl 27-28.)

The average height of the Henye Equus is cca.

146,0 cm. ll is similar to that of the Tilde Cave T1
horses 146,S cm, (E. remage11sis) and can be considered intermediate between the larger form of E.
achenheimensis and the small E. '"germonicus " - E
solutf'eensis.

3.1.4. Character and definition of the Henyc

horses
The Eq1.111s remains .from the I lcnye hill can be class ified on che basis of clifferent features and dimensions
into two groups: absolutely dominant large meclium
s ize horses (dentition, skeletal bones) and sporadically
occurring small medium size horses {dentition, skeletal bones).
The dental dimensions of the large medium size
horses are between the forms E. acl1enheime11sis and E.
remage11sis. Width dimensions of the postcranial bones
suitable for analysis as well as os Ph.1. agrees well with
the form E. mosbachensis - E. achen heimensis and
larger than . "gennanicus - E. solutree11sis.
The large E. achenheimensis described from the
lower loess layers of Achenheim (20 c./d-20a, NOBIS
197 1.) probably appeared already in the Riss period.
Jn Hungary, the occurrence of E. achenheimensis is
known from the Hungaria Hill Cave at Dorog
(JANOSSY-VOROS 1987.) The reason why the
Equus of Dorog was only marked "cf." here is that in
the chronologically closed material, the dimensions of
large Equida bones can be differem. in a mosaic-like
pattern from the typical E . achenheimensis.
The dimensions of the Dorog Equ11s hum. disL
agree with those of E. mosbachensis (PRAT 1968.,
-oars 1971. 1981.), the me-me length dimensions
with those of the short form of E. mosbachensis;

however, the epiphyses and the diaphyses are wider.

Al the E ache11he1mensis, mc-s are shom:r and more
narrow while at E. steinheimensis, mc-s are smaJler
(PRAT 1968., NOB IS 1971.). Among the mts, longer
pieces occur as well. Dimensions of lhe lib. dist. agree
with the smaUer form of the E. mosbachensis, and the
dimensions of E. achenlzeimensis - E. steinheimensis
(PRAT L968., NOBIS 1971. 1981.).
The large Equid from Dorog has large mesodont
dentition, longer and wider bones that the CentJ"alEuropean E. achenheimensis a typical steppe fonn.
lts latest occurrence is known from Pilismar6t- Basaharc brickyards (s.L W 1-Il, VOROS I 990a.).
The medium large E. remagensis. (SKORKOWSKJ
l933) forms in a way a transition between the laige
form of E. achenheime11sis and U1c small forms of E.
''germanicus" - E. solutreenszs.
1n the Late Lower and Middle WUrm, the names E.
piveteaw (PRAT 1968.), the E. remagensis. (remagenensis) (SKORKOWSKI 1933.) and E. scythic11s
(RADULESCO-SAMSON 1962.), E. latipes (GROMOVA 1949.) are used partly for wogeographicaJ dis1.inction and, partly, as chronospecies names.
The large medium size Equus finds from Bodrogkereszti1r-Henye belong geologically as well as phylogeoetically to this large circle. Conventionally it can
be identified with the Central-European chronospecies
E. remagensis. A. Riedel identified the horses from
Tilde Cave T1 with the Upper Wiirm E. '"germanicus '"
or rather E. remagensis chronospecies, noting Lhal
they comprise some more archaic (Riss?) forms as
well (RJEDEL 1980. 73).
The Equid material of the Kiskevely Cave Lower
Layer can be also classified here, which is, on the
basis of its osteology, a forest livins. form. The large
p2 (dimensions: 45 x 36 mm, VC>ROS l 994. 31. Fig.
6. 1., Tab. 3.) found together with remains of A. hydruntinus in the lower level of the Kiskevely Cave belongs Lo E. mosbachiensis (REICHENAU 1901.).
Knowing the layer sequence and excavation methods
applied in the Kiskevely Cave (DOBOS1-V0RO
1994.) this view can be supported.
The collective name for the small bulky Equid
dominating in Europe in the Upper Wiirm, mainly in
its second half is E. germanicus chronospecies: E.
gallic11s (PRAT 1968.). E. "germamcus'' (NEHRING
1884.), E. soJutreensis (NOBIS J 971.), E. ferus
(BODDAERT 1785,. on the basis of the interpretation
of G. NOBIS 1971. SI). Apart from the " local" names
used for l.be small Equidae on the given geographical
region, any other species I subspecies assignation (e.g.
E. ferus ssp., E. caballus ssp., E. przewalski ssp.) seem
mo re of a taxonomical-phylosophicaJ category than
ostcological, zoogeograph.ical facts.
Conventionally, the small size horses have an me
shorter than 225 mm. ml 270 mm and os ph.I. shorter
145

than 85 mm. An early appearance of E. "germanicus"

is represented by the Equ:us found at Sveduv Stul in W
IT- ill (identified as E. cf. gmelini MUSTL 1962.)
The small medium size Equus finds from Bodrogkeresztur-Henye hill can be identified with lhe E.
"germa.nicus " chronospecies.
At the Bodrogkereszti1r campshe, the joint occurrence of two horse species, E. remagensis and E. "germanicus " cannot be excluded. lt cannot be disregaroed, however that one of the horse bone sample
submitted for C-14 dating from 1.82. B sec. yielded an
absolure date of 19-J 8 OOO BP! ls it a casuality, or the
si.te might have had a younger phase as well. Archaeological analysis of the existing find material seem to
exlude the possibillty.
Those Upper Pleistocene sites where the two caballoid horses {large - small, large - medium. medium smal l) occurred together were collected last rime by A.
Forsten (FORSTEN L988. 169-17 I). The presence of
lhe small Equid can be defined weU where teeth and
long bones were found together.
It seems Lhat a general trend of size decrease can be
observed in Ew-ope among lhe Upper Pleistoc.ene
cabalJojd h.orses (FORSTEN 19.88. L7 L-173., 1991.)
but tliis trend is manifested in the Equus material of
some sites in a mosaic-like pattern_The joint effect of
climace and nutrients, the deterioration of the habi1al
for Equu:s and ir.:s direct and indirect effect on dentition
and body size cannot be felt directly on Hungarian Palaeolithic settlements of lhe Upper Pleistocene. It can
be demonstrated, however, that at the beginning of
Wilrm m (Madaras Lowe1 Layer) around the coJd
maximum of the Wilrm period the Equus teeth, especially those of smaller size became abraded totally
very fast Such ''worn out" teeth were on.ly met for the
Coelodonta. S11s and the Crocotta.
ln the second half of the Upper Pleistocene (WUrm
ll.-IJJ.) ll1e history of events 'for climate and sedimentation was faster and more varied than that of the
preceding period. Such fast changes could not be
foll.owed by a "indigenous " horse population neither
in pheno-, nor in genotype. It cannot be neglected,
however that in these times large mammals, thus
horses among other reacted with long migrations,
drifted by " fauna waves" into other regions. This
v.iew is supported by the study of A. Forston1 on
the basjs of numerous find materi.al collected from
large area: apart from horses with long metapodia
and phalanxes, horses of short me appear In W U
(time unit 6.), while with short os _ph. I., they appear in W. Ill (lime unit 7.) (FORSTEN 1991. 416.,
Fig. 2.-3 .). The same pllenomenon can be observed im case of the Hungarian Late Upper-Pleistocene Equidae.
Tl cannot be decided that the large size Eqw1s form
at a given place was a local "archaic" or migrated ele146

mem. Also, we have no details as yer on the chronomigration of Lhe small size Eqmts forms.
rrhese problems can he forwarded only by the regioaaJ smdy. of Equus materials which are exactly
dated and separated well in zoogeograpbical and
chronological aspects.
In my opinion, the Late Upper Plei.stocene decrease
of dimension of Equus can be explained not only by
cold and arid climate and deterioration of the envlronmen~ but also by the appearance of small Equus forms
in Europe by migration.
3.2. Akes alces (L)
224 remains - 34 individuals

'.fhe .relatively abundant occurrence of Alces finds

observed already in the first excavation
(KRETZOI t 964 a.). The determination of the A lees
remains was made easier by the nigh number o.ftoolh
finds, especially lower row of teeth (rnandibulae). The
build-up of lhe crown., the position of the emo- and
hypoconids oflhe praemol.ars, the diagonal crest of the
proto-entoc,onid of P4 proved without doubt the presence of the Alces genus. Dimensions of teeth and
bones correspond to, at t.he sam.e time, the speciJic
data of the species Alces alces.
was

3.2 .1. Anatomical division of the A k es remains

At the Henyc Upper Palaeolithic campsite, 224
pieces of Alces bones were determined. The anatomical distribution of these is given below:

Head-region
I cast anLler beam piece
5 brain skuU fragments (refillable pieces)
3 facial skull fragments E3 dexl maxillae, 2 juv.
ad.)
19 isolated upper teeth
1f P (of these, 6 dp)
8 M (M tl21 sin- 5 dext., 1 M' sin., l Mfr.)
14 isol~ted lower teeth
31 (t di, 2 Ijuv.)
TP (I P<1 dext, 2 dp:?, 2 dp3, 2 dp4)
4 M ( L M112sin., I M3 sin., 2 M inf. fr.)
24 mandibulae (14 sin. - J0 dext.)
l 6 corpus mandib. 8 sin. - 8 de>rt. ( L inf., 2 juv.,
13 ad).
L mandb. lower edge fr. sin.
4 mandb. oral fr., 3 sin. - 1 dext.
3 ram. mandb. 2 sin. - l dext.
Trwtk-1-egion
2 vert. lumbatis fr.
I 9 C-OSta fr.
l stenebra fr.

Fore limbs
6 scapula.fr.,
4 humerus
3 diaph.,

33 tibia
13 diapb.,
20 disc. epiph. 6 sin. - 12 dext.
I as maJleolare

l dist. trochlea sin.

5 tarsals

12 radius
6 prox. epiph. 2 s in.,
4 diaph.
2 di.st. epiph. 2 sin.
I ulna (olecranon) fr. sin.
16 carpale
3 os magnum Ci+3, 4 os hamarum + uncinatum
C4+5 l os scaphoideum Cr. 3 os lunatum Ci, 2 os triquetrum Cu, 3 carpale fr.
15 metncarpaJ
5 prox. epiph. 1 sin. - I dext.,
8 diaph.,
2 dist. epiph. I sin - I dext.

llind limbs
5 femur diaph.,

2 astragalus dext.
I caJcaneus dext
I os centrotarsaJe dext , I os unc iforme int. - lat.,
T 2+3 dext.,
19 metatarsal 4 sin. - 8 dext
10 prox. epiph. I sin. - 5 dext.,
5 diaph., l sin. - I dext.,
4 dist. epipb. 2 sin. - 2 dexl.
l 2 metapodia (me/ml) diaph. fr.

Phalanges
5 OS ph. L
I OS ph. U
I os sesamoideum prox.
The anatomical distribution of Alces remains is presented on Table 13.

Table 13. Bodrogkeresztlir-Henye. Anatomical distribution of the Alces remains on the hunting station
(Number of specimens)
Unh
Seel.ion

c.

D.

Trench
antler
brwnskull

I.
E.

m.

lL

B.

A.

E.

H.

I.

4.

2.

M sup.
mMdb.

2
6

6
6
9
7

I'm(
,_

M lnf.

I inf

vurtcb

COSllt

stcncl>r.1

scopula

II

8
24

I
4

6
3

.--

I
4

rad P'
dph

6
4

dl

ul!lo
cw,i:ll

::!.
I

me px

maltco

16

:s
13

::w

-3s
2

'

s
6

3
2
19

di

astg.

.s

l1w11dph

.s

Tout!

),

D.

4.

r xup.

dl
rem dph
tib dph
di

G.

/\,

fUcinl skull

dpb

IV.

J.

147

Table J3. con.

Unit
Section

I.
E.

c.

D.

m.

II.
B.

A.

Tr<!nc:b

I
I

H.
I.

4.

tarslll

mtpx

dpli

di

mcJml dph

ph I.

3.

2.

calc.

E.

IV.

G.

J.

D.

A.

Tollll

3.

4.

10

.s

12

IJ.

I
I

sesnm.

To!J!I

I
I

3.2.2. Teeth, bones

For Lhe character.isation of ;,he Henye A.Ices, dimension of teeth (Table 14.) and some postc~anial
bones (Table l 5J can be used. The Alces remains,
similar to other animal species found on tb:e s~te are

very fragmented . Bones exfoliated already in the

soil due to physical effects and the wall of lon,g bones

we.re disassembled Lo srmall plates during excavations.

Table 14. Bodnt>gkeresztur-Henye. Dimensiens of the upper and lower cheek teeLh of A lees (measured atz crown,
in mm)

Upper teeth

l.
2.

dp!

dpl

19

l2

15

20

pl

dp4
22,0

21,5

16

23,0

21,5

35

24

14

pl/4

P4

14

14

14

28

32

Min.
24

27

25

}J

28

29

28
28

28

33

30

30

31,0

30

34

31.S
juv.

Lower teeth
P2
1.

P4

P3

dp

14,0

2.
9,5

dp

14,S

10,0

1.

:?.

18,0

IS.O

18,0

13,Q

28.0

17,5

14.0

31,0

18,0

30,0

1 1,0

148

28,U

20,0

16,S

27,0

U,O

21,0

25.0

21,0

30,0

17,0

Mz

-2 I.

--.1.
I.

2.

40,0

23,5

21 ,)

26,0

21.S

21l,O

21.5

28,0
21l,O

2J,O
22;0

38,0

22,0

20,5

28.0

11.S

2J,0

W,5

22.0

17.0
15,0

24,0

22,0

25.0

18,0

26,0

IS.5
19.S

24.0
2.6,0

24.0

1 1,0
21,0
21.5

27.0

22,0

40,0

23,.S

26,11

18,0

;!8,0

20,0

40,Q '

io,o

21.0
20.S
22,0

27.0

21,0

l ll,0

22,0

2ll,O

2 1,5
40.0

23.0

39,S

11,0

22,0
22.0

j!j,)

13,S

24,0

16,0

26,0

19.0

14,0

24.0
23.0
2S,O

18,0
17,0

26,0
26.0
29.0

24,0

27.0

l.l,-0

2.

28,0

13.0
14,S

18,0

27-0

19.0

19.0

M1
2.

19,0

dp

18,0

30,0

22,0

26,0

23.5

28,0

22,0

Table con.
L ower teethrow length (measured at alveol, in mm)

dp

The only antler piece found is a right side cast

beam: the diameter of the rose is 230 mm, the diam eter of the stem is 175 mm.
Skull fragment is known only !Tom Unit Jll. 63. 3

1'-M

l'

M1

2.

I.

sees., one parietaJe (broken to S pieces) and fragments

91

38

of 3 maxilla dext. (Table f 3.) The length of milkpraemolars of the AIces calves were:
upper 6o-65 mm,
lower 50- 6H7 mm.

95

40

50

66
67

dp

166

67

68

38

71

173
172

72
73
77

77

97

40

99

40

l . length, 2. breadth

The dimensions of Late Upper Pleistocene Alces

teeth and rows of teeth are fairly uniform (Table 16.
Table 17). The differences in the extreme values of the
tooth crown size Limit is caused by individual age and
sexual dimorphism, typical of the A rtiodactyls.
The few measurable bones of Alces from Henye
correspond to both Upper Pleistocene (THENTUS
1959. L53 ., JANOSSY 1964. 144; as well as H6r-valley and IstaUosko Cave specimens) and recent Eastern
Ew-opean elk (TOPAL-V6 R6S 1984., GROMOVA
1960. 99.. I 11).

Table 15. Bodrogkeresztur-Henye. Measurement of postcranial bones of A lees (measured at crown, in mm)
I.

2.

3.

Radius

Mdl!CillpUS

llumcrus

79
-

libia

Meuuarsus

>--

44

38

l.
78

8.

9.

Asctaialus

52

48

82

54

43

CalcrutC\1$

168

54.

Msgnum

40

35

25

40

J2

27

47

32
2&

26,S

)J

4J
34

35

36

40

60
54

40

43

23,.S

58

45

27

67
77
33,5

1.

14
30

48

82

74

Os phll

45

-IS

35

54

79

rnquelrum

77,S

~m

6.

Os pbJ.

~iwltUm

"

s.

92

42

56
58
60
41

48
26

149

Table 16. Comparison witb other teeth dimensions ofAlces from the same chrono'Jogical period (in mm)
length

breadth

mun

bliDW:U.

mean

rnln-m u.

23,8
23;0

23.0-24.0
23,Q

26.4
26,5

21,S-32,0

23,6

23, 1-23.S
23.2-23,7

1
2
4

'1.7,?
28,9
27,0

24,0.--3 1,0

27,2-30,0

p314

Bodrogkemznir- Benyc

Kiskevely c. LL
Bellllov c.1 sp.

BabjuC. 2j.
M l/2

IBodroi1tcrettbir- B.cnyc.
BabjaC. 2j_

3
1

Vet'l!itlctl C. IL:l

21.0

29,8

26,S
22,0-26,5
29.S- 30,0

3
1

30,8
29,0
17,0

28.0-34,0
27.9-31.8
27,0

13.8
14,0

13,0-15.0
14,0

p!

Bod rog.ktr~Jzflir-Heuyc

18,7

18,0-f9,5

Nagymaros~

1
2
I

19,0

19,0

6
I

L9;0-t 9,2

13',8

18,8

13,4

l3,4

8odrogkcrcsztiirHenyc

23,2

22,0-25,0

17,4

l.S:,0-20.$

Nngymaros4
BeUllOV c. 1sp.
Bttbia C. 2 j .

22,S

22,~

24,5

:ZJ,9-25,0

I
3

18,0
16,S-16,8

23,4

2),4

18.0
16,7
17,2

Bctalo11 c. 1sp.

Bllbja C.2 J.

IJ,6-J6,8

1'3

17.2

p~

Bodrogkues:Uur-H1111yc

II

26,8

19,9

lli.5-12.0

29,0
30,5

24,0--30,0
29,0
30.5

10

Nagymar9S4
l'itismar6t-Od.4
Bceialov C. 1 sp.

II

19,8

27,9

2 7,3- 28,5

5
2

19.8
21.0
18,8

B~bjn C.2 j .

I
5
2

2:7,9-28,6

21,0

17,.S:-20.4

20, 1-lJ.8

Mt
Bodrogkercsztul" B e:rrye

25,6
21!,6
30,0
28,0

N agyl11lllos4

9
I

Pilismarot-<".>d.4

Pilisw1l6 n. Rsh.s

BnbjaC.2 j .

Vetemica C. h.3

25.2

25,2

1
2
t

9
l

25,6

13,0--28,0

18,6

28.6

30,0

211,0

30,0
28,0

'2.

25,'2

24,8-28,3
25,2

10,l

20, l

ll

27,8

26,!}-30,0

1l

'.?l,9

20.0.-23,5

1
I

31.0

31,D

23 ,0

23.0

211,0

28.0

20.0

20,0

2&,6--29,2

2
7
1
I

M.t
Bodrog.kercsztlir-Rcnye
Nagymaros4
Pllismarot..Od.4
'Piliss:zanto u. Rsh.5
BabjnC.2 j.
Vei...mlc11C. h. 3
Mz
Bodro.g.k ert'iitlir-Benyt
Nagymnros~
Kisll~vay

C. LL

Babja c.: j .

24,8 28,3

2
1

?3,0-28,0
28,6
30,0
28,0

9
1
1

20.6
22.0
22.0
19.0

tS.0-21,5
22.0
22,0

19.D
20,6- 21,7

20.1

20.1
18,0-21 ,5

20,6
22,0
22,0

19,0

u ,o
22,0
19,0

20,6-21,7

22.1- 22,5

Ml
Bodrogkcrcsztiir-flcnyc

39.3

l&,G-40,0

NB&)'maiv~4
Kiske,tly C. LL

41,2

Babjn C.1 j.

43.0
40,7

41.2
41,0
40,7

150

21.8
23.0
23,0
22,7

20.0-2.l,)

23,0
23,0
22,7

Poroszl6, Tisza river bed (JANOSSY-VOROS 1979.

Nr. 486)

Table 17.
Lowt:r leethrow length
Bodrogke,..,,'llt1r-Rcnyc

P- M

I 70.4

166,()-1 1),0

rm
Bodrogkuattlir-llenyc

72,2

Na.gymlll'OS

6 8.0

Bud rogkcre,,ztu r-1 lenye

93,5

N~mnros~
Bahja C.2.j.

1
1

98)1

9~

99.0

99,0

Szeleta faonal phase 11 sites Wtlrm 1.

BildOspest Cave, upper (6) greenish grey layer aad

67,0-77,0
6&,0

1\1

91,0-99,0

I. RAKOVEC 1956. :?.; 2. POHAR 1985. T11bL 11-12.; J. MALEZ

1963. 126.; 4. MOTrL 1942b. 54.; 5. VCRCS 1986b. 37.

3.2.3. Occurrence of Akes alces in the Upper Pleistocene of the Carpathian Basin and its

environs
According to general opinion, Alces alces sho\vs in
" ice-free" Europe chronologically and regionally discontinuous occurrence patters during the Upper Pleistocene (WUrm) period.
Its widest distribution can be placed lo the last interglacial (Eem) and the fo llowing Prae- and Early
Worm periods. ln tl1e Middle and Late Upper Pleistocene (W IJ- l ll period) which is in the focus of interest
due to the A.foes finds at Henye , the occurrence of Alces alces was restricted to certain periods or regions.
Hu11gary 42 localities
ln Hungary, remains of Aices are known from 46
Pleistocene localities. It appearance can be connected
to two fauna! stages.
The first is the Middle Pleislocene Biharian faunal
stage, of which the cartiest occuf'!'eoce of Alces is
known:
Lower Biharian substage, Osztramos 2. Alces
group (JANOSSY-KORDOS 1977. 43.)
Upper Biharian substage - Tarpa hill Alces sp.
(together witb Trogonrherium, JANOSSY- VOROS
1979. Nr. 468), Solymar-OrdOg Cave A/7a, Alces
latiftons. Alces brevirostris (KRETZOI 1946 a..
VOROS 1985., 1988 a.), Gyorujfalu Alces latifro11s
(JANOSSY-KROLOPP 1994.).
The secand is the Upper Pleistocene Utrechtian
fauna! stage in which the a.umber of sites with A/ces
alces is tenfold: 42 localities. With the exception of the
last third of the fauna! stage (Wtirm ill), AIces is present though with variable intensity.

lower (dark grey) c ulture layer (5-3.) (KADIC 1934.

64, MOTIL 1941 . 15),
Herman Otto Cave, 2. layer (KADIC 1916. I 0, EHlK
1916.25)
Szelim Cave, 3. "C" layer (VERTES 1965.345)
Fegyvemek, Ti.sza bed (JANOSSY- VOROS 1979.
Nr. 497)
KisJ...'Unfelegyhaza - Brickyard (fauna: Arvicola terrestris. Panthera spelaea, large Eq11us "abeli'', CervJtS
elaphus, Mega/oceros giganteus, S116 scrofa (BENDA
1929. 268-270 JANOSSY- VOROS 1979. Nr.
449-450)
Raba bed, Ikreny (JANOSSY- VOROS 1979. Nr. 6)
Sagvar - "sand with concretions" under the loess
(GAAL I933a. VOROS 1982.J
Szolnok-Tisza bed (JANOSSY-VORQS 1979. Nr.
501)
Tiszakecske-Tisza bed (JANOSSY-V6R6S 1979.
Nr. 521)
Tiszasilly-Tiszai bed (JANOSSY-VOROS 1979. Nr.
491)
Vezseny-Tisza bed (JANOSSY- VOROS 1979. Nr.

515)
1st.all6sk5 faunal phase 27 sites, Middle Wilnn, W ll,
W 11- IU.

Ista116sko Cave, lll. (JANOSSY 1955 160, VOROS

1984.).

Jankovich Cave, upper yeUow layer, rich in microfauna (Hillebrand 19 19. 9, fig. 3.)
Kiskevely Cave 3. layer (MOTTL, 1941 17, VOROS
1994.)

Pesko Cave, grey layer, "sample nr. 12." (HTR 1990.,

Table 1.):

Pilisszaot6 D. Rock Shelter l'vll..., 7. layer (VOROS

1986 b.)

Solymar-Quarry, cleft filling (KUBACSKA 1927.

VOROS 1988 a. Tabl. 4.)
Szelim Cave, B2 layer (GAAL 1935., MOTIL 1941.
15)
Tark() Rock Shelter, Block II upper "2" {JANOSSY

1976. 15., TabL. 11/C.)

Bodrogkeresztur - Benye

Dunaszekcso, loess (JANOSSY-VOROS 1979. Nr.

Varb6 fauna! phase 1 site, Prae-Wa.rm
Lambrecht Cave, Layer lV-V (JANOSSY J964. 148)

298)

Subalyuk fauna! phase 3 sites, Lower WUrm

Kiskevely Cave, Layer 4. (VOROS 1994.)
Lambrecht Cave, m. (JANOSSY 1964. 148)

1979. Nr. 342)

Filzesabony 1979. Nr. 343)

Filzesabony - gravel quarry (JANOSSY-VOROS

sand quarry (JANOSSY-VOROS

151

Fokoru, Tisza bed, (JANOSSY-VOROS 1979. Nr.

494)
Kotelek, Tisza bed (JANOSSY- VOROS 1979. Nr.
493)
Lovas - grit quarry, Palae.olJthic paint m.ine (DOBOSI-VOROS 1979.)
Madaras-Brickyard, " Lower" loess under the Upper
Palaeolithic site (VOROS 1989.)
Mlilyi - gravel quarry (JANOSSY-VOROS 1979,
Nr. 374)
Nagybatony (JANOSSY-VOROS 1979. Nr. 304)
Nagymaros-Ujvolgy, Fischer-T6bifls's " soil quarry"
(MOITL 1942b. 47, - VOROS 1998.)
Nagymaros - beside narrow-gauge railway (MOTIL
1942b. 48, - VOROS 1998.)
.
Pilismar6t-Oregek du16 (MOTTL 1942b. 53-54)
:eotgar, Tisza bed (JANOSSY- V6ROS J979. Nr.

482)
Romhany,

in cavity of Triassic limestone

(JANOSSY-VOROS 1979. Nr. 218)
Szolnok-Samyak, Tisza bed (JANOSSY-VOROS
1979. Nr. 503)
Szolnok..Sokoru, Tlsza bed, (JANOSSY-VOROS
1979. Nr. 508)
Tiszalok-RAzom puszta (JANOSSY-VOROS 1979.
Nr. 478)
T6szeg cliff, 'fi:sza bed (JANOSSY-VOROS 1979.
Nr. 512)

Apa (2), Baia de Fier (3a, SulrCarpathian), BistriJa

(7), Borsec (9), Bratei ( 12), Cimpia Romana (16,
Mcmntenia), C uina Tmcului (Danube valley, E:pipa-

laeolithic, 20a)
Lespezi (30a), MaJini (32), Meziad (34), Nandru

(36a), Noul Sasesc (37). Onca Sagatag mine

(CZIER 1988.), Orme~ (40), Ripa (44), Ripiceni-lzvor (Prat valley, CHIRlCA 1987. 29-30), Scari~oara
(46).
Alees alces occurred in Roumania during the Middle Wurm (W 11-li/lll), the Upper Wiirm (W ITT), lhe
end of the foe Age and also in Lhe H olocene (MACA ROVTCI 1962. 68).

Y11goslai1ia 3 localities
Aradac (Voivodina) is known from old teebnical
literature as a site withAlces (Koch 1900. SS 1). lt was
dated to s.1. Upper Pleistocene.
Vucedol, Danube valley in loess from W ill
(?)(MALEZ 1972., MALEZ 1986. l 07).
One of the southernmost exposed sites with Alces
alces is known from the Palaeolithic settlement of
Crvena Stijena (Crna Gora, Mousterian-Tardlgravet,tian~ MALEZ 1972. 176, MALEZ 1986. 107, J URIC-BASLER 1979. 380. Nr. 116.).

Croalia 12 localities
Croatian sites with Alces are well known from the
work of M MALEZ (MALEZ 1972., 1979., l986.)

S /ovalda 3 localities

TheAlces remains were found both i.n Cave sediments

Apart from data in older technical literature from

Presov (Torysa river bed) and Tovarne (Ondava river
bed) (KOCH 1900. 551 ). J. Barta.mentioned fossilAbces finds around Tren~in (B ARTA 1964. 17). Their
age is s.L Upper Pleistocen~.

and open air sites.

Localities with Alces include: Kanegr~ Krapioa C.
(Mouscetian, W 1- 11., MALEZ l9.?9. 236. Nr. 42.),
Peeina in Bri.na near Dmis (with A. hydruntinus. in
Aurignacian-Gra.vettcian contex.i, MALEZ 1979. 247.
Nr. 78.), Pisana Stina in Opor Mountain (Dalma:tia),
Romualdo Pecina layer "c", W. lIL (MALEZ 1979.
2)2. Nr. 58.), Samobor (Mousterian. W l ., MALEZ
1979. 253. Nr. 45 .), Sandalja II. Cave (Aurignaoian-

Roumania J 7 localities
In Roumania, there are 17 Upper Pleistocene sites
known with Alces (CZlER-JURCSAK 19&7., CHIRICA 1987., CZlER 1988.). lo fayer ID. oftbeMousterian site Ohaba Ponor, the "Alces machlis Og. f. diluvialis" lt anddi1 (GAAL L943 a. 21-22., Tfil. Il. 11)
seem to be, on the basis of tbe photos published, not
Alces but i:emains of a large Cervus (incisors published as Saiga are probably Rupiaapra, ibid.). Thus
the site should be cancelled from the list of Alces sites
(CZJER-JURCSAK 1987. Nr. 39). ln the following,
Romanian sites are numbered according to the above
list.
With the exception of the Moilntenian Late Upper
Pleistocene (16) and the Sub-Carpathian Moustereauri.gnacian (3a) as wel.L as the Aurignaco-Gravettian
site of Ripiceni-lzvor in the Prut valley, tbe bulk of. the
occurrences of Alces were registered fro01 Transylvania. Most of the known occurrences are valleys or

close to valleys:
152

T:ardigravettian. W 11-ill. ). Veternica Cave layer ..b.,

~ Ii'U., MALEZ 1963. 126-127. revised by MJRACLB-BRAJKOVIC 1992.), Vlndija Cave (Aurignacian-Gravettian ('?Mousterian, ? Meso.lithic )' Vcli.ka

Pecina, Gora.nee, Ravna Gora (Mouslerian-Gravettian,

W. 1-III., MALEZ 1979. 263. Nr. 35.), Vinkovci
(KOCH 1900. 55 1.), Zarilac in Basin near S lavonska
Poiega.

Slovenia 7 localities
The chronology of Slovenian AJces sites is relatively well known (RAKOVEC 1956,, 1975., BRODAR-OSOLE 1979., POHAR 1983 ., l997.).
Localities with Alaes include: S.I. Wurm: Tomaiek

Cave near Sezana (RAKOVEC 1956.),

Lowe r WUrm - W 1.-W lln.: Betalov Spodmol,
layers 3-4. (BRODAR-OSO~ 1979. 136. Nr. 16.).

Matjaieve kamre layer 3., (BRODAR-OSOLE

1979. 143. Nr. I I.).
Middle WUrm: W TT-TI/III.: Parska Golobina, layers 3-4. (BRODAR-OSOLE 1979. 147. Nr. 27

Lower A ustria 3 localities

E. Thenius remarked the followings concerning the
elk finds of Willendorf V (Wilrm O/ITT ?): Alees was
occurring only rarely in young Pleistocene sediments

Cravettian),

and enumerated three sites (Rixdorf, Klinge bei

Late Upper Wurm - end of W 111 - Bolling incers1adial: Babja j ama Cave layer 5 (Epigravellian, POHAR l 985.), Lukenjska Jama Cave layer 4. ( Epigravenian, POHAR 1983.), Matjafeve kamre layer 2.
(Epigravetlian, RAKOVEC 197 5. Tab!. I .), Zupanov
Spodmol layer 2. (Epigravettian, RAKOVEC L975.
Tabl. I.).
On the NW part of the Balkan peninsula - the foothill region of the Eastern Alps - A lees occurred since
the beginning of the Wilrrn period. The migration
of Upper Pleistocene Alces to Slovenia and Croatia
came from tJ1e north-west, from the direction of the
Eastern Alps (RAKOVEC 1956., MALEZ 1972. 177,
MALEZ 1986. 107).
Their further distribution towards the South is
dated for the W Ill. stadial.
The age of the Alces remains from the Sava valley
is uncertain; their dating 10 the Holocene period, exclusively, probably has co be revised (RAK.OVEC
1956., MALEZ 1972.).
A/ces remains described from "heavily, mixed"
mammalian fauna, dated to the interstadial periods
"Bolling and AllerOd". following che W Ill stadial
maximum as well as the "Oldest, Older and Younger
Dryas'', with no Eq1111s ar all (POHAR 1997.)

Cottbus, Dilmten ). From other Layers ofWurm, il was

not know11 but a large Cervus form was found "insread" (THENIUS 1959. 82., 151- 153., 167.)

Nortliem Italy I 0 localities

In the Upper PleistoceneA/ces occurred on the concinental zoogeographical region - from Alps to Bologna: in W Liguria (I site), N Lombardia (I site),
Veneto (6 sites) and friuli Venezia Giulia (2 sites)
(MINLERJ et al. 1995. Tab. 1.).
Sites (Nr. ta.ken from MINIERJ et al. 1995.):
Grotta di Grimaldi (5.); Alluvioni Pavesi (15.);
Grotte di S. Bernandino (20); Riparo Tagliente (23.);
Grottina Azura (locality ofGroua di Paina, Epigravetlian, (BARTALOMEI et al 1988. Tab. 1. In: POHAR
1997. 156.); Grotte del Broion (25.); Grom1 di Irene
(26.); Ponte di Veja (3 L); Cave di Ca'Negra (40.),
Tilde C. (W IT.-UmI. near Gabrovizza - earlier
names: La grotta del'Alce, Losova jarna, RIEDEL
1980.).
Apart from the s.I. Middle Wurm Grotta di Grimaldi (W lfil.) and Alluvioni Pavesi as well as Tilde
Cave near Tricst dated to the second part of the
Worm Glaciation, the other seven sites can be
dated to the Late Pleistocenc I Holocene transitional
period (very end of the Wilrm g laciation (MI~
N IERl et al. 1995., MASSETJ et al. 1995 .. RUFFO
1995. Fig. 10.).

Czech Republic 9 localiries

lo Morava region. Alces remains were published
from three localities: Klilna Cave, layers LI .a.-7 RJW
- W U/IIl and layer 5 (end of W III.), Magdalenian
(MUSIL 1969. 10- 17.), Pfodmosti W IL-W TI/In.
(ABSOLON-KLIMA 1977. 73.) and Sipka (Stramberk), end of W UL, Magdalenian (MUSIL 1958.

25.).
Four other sites were mentioned in connection with
the fossil Alces finds of the Northern Bohemian
Zechovice 11. srone quarry and Novy druh u Volyne:
Suchoma.st, Sudslavic, Trmic and Josefova (ZELlZ
KO 1923. 250-25 L, 264-265.).
Poland 21 localities
Ln the catalogue of K. Kowalski on Pleistocene
mammals in Poland, pubJished in 1959, 59 localities
with A/ces finds were enumerated. The author used
mainly old reference dala and remarked that most of
the Alces finds were Holocene and a closer dating of
the Pleistocene finds was impossible (KOWALSKI
1959. 39., 237.) . On the basis of other UpperPleistocene species found on the sites with Alces 21 locaJilies
proved ro be Pleistocene wilh great certain[)
(KOWALSKI 1959. Cat. I 71- 176.). In lhe followings,
the number after the sites is Kowalski' s catalogue
number.
Three localilies from Northern Poland: Elbi~
(18.), Grupa (25.) and Miniccta (21.) from the Early
Upper Pleistocene. On localities 18. and 25., A lees
was found together with Megaloceros.
From the southern margin of Soutb-Poland, 18 localities wi1h.A/ces are known:
From the Sudet Mountains in Upper and Lower
Silesia. 7 sites: Brzeg Dolny (32.), Chorz6w (40.),
Dzierzno (39.). Masl6w (34.), Radochow {38.), Strzegom (37.), Witk6w (1.).
From the region ofKrak6w-Wielun Upland. 11 sites:
Bf(blo, Bf(blowska Cave (43.), Czul6w, Matkq_
Boskl\ Cave (51 .), Potoczkiem Cave (52.), Murck
Cave (53.), Maszyce, Maszycka Cave (46.), Mnik6w,
Milasz6wce Cave (49.), Kochanka Cave (50.),
Piekary, Galoska Cave (48.), S!\,s p6w, Koziarnia Cave
(42.), Wienchowic, Wierzchowska Cave (44.), Mamutowa Cave (45.).

153

The site Maszyce Cave (46.) is an Upper Palaeelitbic s ite from the Magdalcnian period. AU the otliers
rn.
probably come from l:he s. I. w n.-If/Ill., late
period.

Localities wiJh Alces from the territories .to tlie east

ofthe Carpathes 12 localities
Alces occurrences of Late Upper Pleistooene age

are .known from the river valleys in the Western

pan ef l:he Voll1ynian-Podolian table (GROMOV
1948., PTDOPLICHKO l956., LUCIUS 1970.
CZlER-JURCSAK 1987.,DAVID 1980..).
'
localilies are the fo llowing:

West-Ukrai11e 3 localities:
Sirec (PlDOPLICHKO l 956. 87.), Temopol, Bug
valley (PIDOPLICH KO 1956. 131.), Gura Putilei
(CZIER-JURCSAK 1987. 620., Nr. 27).

Prut-valley: 4 localities
Cemovici (Ukraine, with Megalocel'os, PIDOPLICHKO 1956. 169.), Brinzeny Cave I. (Moldavia,
with Megaloceros, DAVID 1980. 23. Tab. 8 .) Stari
Duruitory C.Il.(Moldavia wilb Megaloce1os, DAVID
l 980.23 .Tab.-3.), Ripiceni-lzvor (Romania, see there,
CHIRICA 1987.).

Dnie.srr-valley: 3 localities

Molodova V (Ukraine, PTDOPLTCRKO 1956.

168., layer 7- end ofSolutrean, layer /.-l. end ofLate
Wtirm, CHERNlSH 1961. 72-73 ., 7. W IIYIII. 6. beginnir!g of W Ill, 5-2. second half of W rn. LUCIUS
1970. 76-81 ., 6-1 , last Valdai (W) interstad.ial, beginning of the Holocene, BORZJYAK 1993. 68. Tab.I.),
Kom1an IV (Ukraine, CZIER-JURCSAK l 987.
619., Nr. 18, ~3 l'ast Valdai (W) interstad.ial - eegi:nning of tbe Holocene, BO.RZIYAK 1993. 68. Tab. I.).
Ras.kov Vil (Moldavia, Epigravettian, end of W fil
DAVID 1980. Tab. 10).
Central Russian Plain I /.ocality
Among tbe 29 Upper Palaeolithic settlements, 0.
SOFFER ( t 985. tab. 2.1.2.) mentioned no Alces. From
Goney in d1e Udai valley (SOFFER L985.Nr. 7.) V. f.
GROMOV published A/ces remains (GROMOV
1948. 92.). In the publication of E. LUCIUS, (1970.
90.),Alces was also missing.

Dniepr valley 1 locality

A:lce.s is mentioned fr'om the sit~ Kremenchug
(Ukraine, with Megaloceros, PIDOPLlCHKO 1956.
l08).
Alces occurred in the W I. period, the WUrm interstadial.s and the end of the Wi.im1 m I Holocene transition.
l54

B ulgaria 2 localities
Fossil occurrences of Alces are known from Northern
Bulgaria, o.t the following sites:
Bacho K.iro Cave 8., cf. Alces sp. (K.UBIAK.NADACHOWSKI l982. 63., Tab. l ., Gra:vettian,
Middle Wlim:t, GINIBR-.KOZLOWSlU J982.)
Temnata Cave TD-I. 3d and TD-V. 3a-3d-3g; Alc.es
alces and Alces seu Megaloceros was described,
respectively (DELPECH-GUADELLl 1992. 154155., 158- 159.., Tab. 1- 2.). On the basis ofthe published dimension of os ph.. ill. from TD-V 3d
(DELPECH-GUADELU 1992. 155. ), it can hardly
be Iemains of Aices but muoh smaller game. Layers
3d-3g were of Middle Wtinn age, Gravettian, layer 3a
of Late Upper Wfirm age, Epigravettian (GlNTER-KOZLOWSKl 1982.).
Ji,fegalocerns and E. hydrumiius were- found at
both siles and Rangifer was missing.

3.2.4. Late Pleis tocene cbr(jnological distribution of

A Ices alces

The occurrence of elk n1med out to be discontinuous after the Riss/Wilrm (Bern) .interglacial - Early
W U:rm l. period, both in space and time. T0 answer the
question, where ms the areal of.elk in Europe during
the Middle and Upper WW:m, the study of kno,\vn sites
is necessary.
Altogether 132 so-c.alled Upper Pleistocene localities were collected from the Carpathian Basio and the
adjacent areas where elk used to live (s.I. W. l. W.
1.-TL, late W Ill; 'fuble 18., Fig. 9.) Areas to the NortbWest and North of the Alps were not included in the
slllr.Vey. This area could be tbe western stripe of the
Pleistocene I Holocene dispersion of e lk. Upper Wttrn1
data were nol obtained. W. v. Koenigswald published
elk from 9 sites from lh.e Northern Upper Rhine region, in falllia ftom the la.s t interglacial. In the fauna
list, ALees sp. in Tab. 1., Alces alce.s was mentioned as
a climatically indifferent species sirnilar to a1uoch (?)
(KOENIGSWALD 1991.). In the Lale Glacial fauna
of Gamssu.lzenhohle (0berOsterreich), elated cca.
14-10 000 BP., elk can be possibly considered as an
expansional fauna element (FRANK et al. 1995.,

Tab.l.}
Distribulion and frequency of Late Pleistocene elk

sires (Fig. 9.) demonstrate clearly that Alces used Lo

occur, dotnlnanlly, within tbe area berdered by the
Western Alps - Southern France -Eastern Alps- Ore
Mts. - Sudetes - Car,patbes; to the south~ in the norlhe.rn f.oatl1ill region area of the Po valley, further on,
outside the Sudetes and the Carpatheson the Northern,
Northl3astem marginal zone. Occurrences in Southern Romartia and Northern Bulgaria can be considered

marginal.

Table J8. Distribution ofA lees in the Wilrm and Holocene period (site number)
Period

Poland

18

Moldavia West Ukraine

II

Bolli:mla
Momvin
Slo\al:ia

WI.
~

Au.<trrn

Hungary

38

RoumiUlia

17
2

Bulgnrio

llolocene

w 11-0/ITJ- m
2

3
3

I
1

PldsU>Q:nc

St

l:nd ofW lll

Prachi.~

Hist

SS

35

49

2
I

II

27
4

2
l

38

Yugosl~vin

CrnatlB

12

Slovenia

No.rtbcm hnl)

10

JI

For references, see chapter 3.2.3.

1. 2. 3 .... localities
!l\Slll.'m recent dlstr.

Fig. 9. Topographical distribution of Alces alces during the Upper Pleistocene in Europa

l55

To answer the question: to what period(s) the elk

sites can be dated. more exactly, when did elk tum up
at a given region is more difficuH.
Mosl of the sites have no exact geoJogical, archaeological or radiometrical dating. [n most cases, "only.,
the faurustical dating is available, resulting from the
joint occurrence of chronologicaUy important species.
lndividual finds make the correlation of the faunas
more difficult, which is at the same time a self-evident
vicious circle. Which is saying: species X belong to
fauna Y because species X used to occur1 so far, in Y
fauna. This must be true, knowing the speci.e s evolutionary trends of a given region (e.g., in the Carpathian
Basin: KRETZOI 1969 a., JANOSSY 1979.) In case
of a chronologically discontinuous species, tbe exact
faWJistical-chronological assignment, i.e., which
chronophase of a given occurrence the actual remains
can be assigned lo cannoi be performed without other
data like faunal assemblage, archaeological cul.t ore,
geological, radiometrical dating o.r species evol.utionary data.
For the chronological assignation of a part of the
Upper Pleistocene elk sites, the author made use of,
apart from the basic publicationsl the work of A.
Paunescu (1984. Fig. 1.) and Paleolit SSSR (1984).
Taking into cnnsideration only the mentioned archaeo logical cultures, disregarding now localJy different
terminology cannot given "faunaJ assemblage" of a
region ma concrete period of time, and their regional
and chronological changes. Taking chronological level
as priority, regionalJy it can comprise several archaeological cultures. Taking culture (eulturaJ complex) as
a priority, they can comprise a long period including
several otimatical phase-S. The s.i milar character of
econom ica]-cuJtural heritage (e.g., way of life, use of
rooJ types) can be spread .in time. diffuse in space. For
example in Europe there is a difference of about 4000
years between che Mesolithic cultures of Southern and
Northern Europe.
The known Upper Pleisfocene elk sites can be separated into three periods:

1. W 1.- W UIL
2. W n .-Wil/IlL

Mousterian
Aurignacian-Gravettian
3. Second half of Will., Magdalenian, Bpigravettian,
transilion towards the
Epipalaeolithic, TardigravetHolocene
tian etc.
Period 3. comprise also U1e sites classified .by archaeologisls to Late palaeolithic on the basis of archaeologica I heritage; on the basis of C-14 dates, however, they are o bviously younger than I0000 years
BC, that is, belonging to the Holocene period. The re1 see at the list of Localities

156

vision of these finds, however, al.ready fall outside the

present scope.
"
In sp.ite of the scanty chronological data, the known
distribution of elk can be used co reconstruct the Upper Pleistocene areal of Alces.
Upper Pleistocene occurrence of Alces (Fig. 9.) was
dominant in two areas:

I st
Carpathian Basin
T11tnsylvania
NE-Balkans

2nd

South Po.land
West Ukraine, N ortb
Moldavia
(Romania. N Moldavia)

72 localities
3 7 localities
L3 localities
22 localities
30 localities
L8 localities
l 2 localities

54,6%
28,8%
9,8%
16.8%
22.7%
13,6%
9,1%

The greatest frequency of Alces was encountered in

the Carpathian Basin and the NW-Balkans.
Calculations can be made without the 13 Upper
Pl.e istocene finds from the Tisza bed; in this case, the
ratio of cave l open air sites will be 9 I J 5. We may
aL">O add the 13 localities .Uom the Sava bed, which
were described by the authors (RAKOVEC 195.6,
MALEZ 1972.) as Holocene but in the opinion orthe
author, they are Pleist0cene finds: in this case, the
number ofNW Balkan s ites will be 35. The l3 Transylvanian (Romanian) Alces sites, mainly from river
vall.eys, completethe pictur e nicely.
Localities wfth Alces in Silesia, outside the mountain range of the Sudetes and the Carpathes are less
numero.u s (7 sites), more on tbe Krak6w-Wieh'.m Upland (1'1 sites). ln Western Ukraine and the upper
reach of the Prut valley 7, in the I>niestr valley 3, in
the Dniepr valley and the Central Russian Plait1 ll
sites with Alces are known.
The total number of localities with A Ices from the
other regions - Northern ILaly, Lower Austria, Bohemia, Slovakia, S Romania, N Bulgaria (Fig, 9). is altogether 30-22,1% of all localities. Numerically, lhey
agree well wjtb the Joc.aJJ,ties of the N-NW marginal
region of the Sudetes and tQ.e Carpathes.
It is remarkable how much the areal of Alces got
pimched within the area bordered by the Central European Eastern Alps and the Carpathes, its southern border formed by the Po valley, and the stripe of the margimal (outer) zone of the Sudetes and Carpathes.
It seems that the Carpathian range means no physica~ obstacle fQr Alces; neither today, nor in the Upper
Pleistocene. Along the low-lying passes, valleys, elk
could c ross the mountain range without problem .

Moravia, Northern Bohemia and Slovakia form a sort

of faunistical transitional zone berween the Carpathian Basin and the South-Polish region. It could not
be ..by chance" lhat in the Late Upper Pleistocene, elk

Western Balticum (Denmark) - J5 localities

U1e Glacial "Aller&i- 3 sites

~AARIS.SORENSEN 1993. Tab.

turned to be a Hmountainous forestal species" not only

Mesolithic 12 3i1.CS

(lill lhe end of the Allantic period)

(JARMAN 1972. Apptndix 1.)

i.n Central Europe, but also in the distant Caucasus

{BARYSHNJKOV 1982.).
According to our present state of knowledge, the
soUlhemmost known occurrence of Alces ill Europe
during the Late Upper Pleistocene was found in the
Crvena Stijcna Cave, in Cma Gora at 43 latitude, in
the Prut-Dniestr valley, at 48 latitude wh.ile in the
Don valley, it is over 51 latitude.
ln Hungary, Alees appeared in the Varb6 and Subalyuk fauna! phase of the Utrechtian fauna stage, became common in the Szeleta fauna! phase (1 1 localities) and frequent by the lsta116sko faunal phase,
when the number of known occurrences exceeds the
double of the former period (27 sites). By the rime of
the WUnn 111 maximum (beginning of Pilisszant6 fauna! phase) Alce,i, disappeared from the Hungarian
fauna.
3.2.S. Holocene occurrences of A lees (Fig. JO)

p.

0., FISCHF.R 1993. 56.)

Gel'mmry - 2 localities

I(JARMAN 1972. Appendix I .)

IMe..<olithic 21itts
Poland - 93 localities

"Holocene"38 sites, (KOWALSKI I 959. Catalogue

171-176.):

Mesolithic I, Neolithic 6, Bronze 1, Iron-Roman

12, Medieval 35 sites (WYROST 1994. Tab. 5.
119-120., 149-150.).
Western part of Switzerland - 73 localities
(CHAIX-DESSE 1981 ., Annexe 1.2.1., HESCHELER-KUHN 1949. 264., 282., 507., 516.):

Mesolithic 3, Neolithic 23, Neol-Jron 10, Bronze 3,

Iron-Roman 6, "Holocene" 18, unknown I0 sites.
France - 8 localilies

(CHAIX-DESSE 1981., Annexe 1.2.1.,):

Mesolithic 4, Neolithic 4.

In the transitional period between the Late Pleistoceoe I Holocene, Alces accumulated along the Western
margin of the Alps and advanced along the Eastern
side of the Carpathes towards the North, North-East.
Elk reached the northern limit of its recent occurrence
during the Preboreal - Atlantic phase, during some
2-3000 years (Fig. JO).

Morava - 1 locality
"prel1istoric I Iron Age" + mixed of Pleistocene
species,( MUSIL 1969. 10- 11.).

Northern part of Sweden - 64 localities (IREGREN

(SZALAY 1916. 6.).

1983. Tab.4.1.12.).
Mesolithic 2, Mesol-Neolithic 6, Neolithic l I,
Mesol-Neol-Bronze-lron I0, Neol-Bronze-lron 26,
Bronze-Iron 9 sites.
Eastern Balricum - 57 localities (PAAVER I 965. Tab.
67).

lEarly Holocene

2 sites
12 sites
43 sites (Tab. 67).

IMiddle Holocene
1Late

Holocene

(In th~ Fig. 41 different data wcn: given by PAAVER 1965,)

12 ~lies
13 sites

Earl)' Holocene

7000-3000 BC

Mlodle HlllQUllC

2000 BC

Late Holbccne

second
half
of
Millennium BC
Millennium AO
tlnd Millenn1um AO

lst 20 silc:s
lst

lOsitH

Slovakia - 1 locality

Jablun.ka

peat (Orava county), "prehistoric",

Hungary - 2 localities
"prehistoric~' (Belavari berek =grove, near Heresznye, Somogy county, (SARK6Z'!" 1~16. 247- 248).
Copper Age (fis-zaluc-Sarkad, VOROS 1987 a.).
Slovenia - 4 localities
Mesolithic 2, Bronze 2 (RAKOVEC 1956., 1975.
Tab. J., MALEZ 1972., POHAR 1983., BRODAROSOLE 1979. 136.).

Remains of AJces from the Sava valley, flowing

across several countries were found at 12 sites,
considered Holocene (RAKOVEC 1956., MALEZ

1972.).

Romania - 38 localitie6
"Postglaclaa" 9, Neolithic 4, Bronze 5, "Holocene''
LO, unknown I 0 sites (CZIER-JURCSAK 1987.
618-622).

157

Penod

All

Prb

PALANK

Fiaunal ph.

Cultures

KOR CS

10 00

8000

5000

6000

3000
2

NSweden
Baltic regmn

BOKK

Al.FOLD

2000
1

45

1000
1000AD

12

43

12

Poland

35

- - 12

Denmark
France

SA

Br

Neal

Mes

BP

Be

SB

At

2J

W SWitzertar:id

Germany

Morava
Slovakia
Hungary

Slovenia

Romania
Moldavia+

WUkra1ne

2
4

16

26

&1'mlR

recent distr.

1.2.3 ... l ocall11es

Fig. 10. Topographical and chronological dis1ribution of Alees alces during the Holocene in Europe

158

Western pans of Ukraine a11d Moldavia - 86 localities

Mesolithic 3, Neolithic 16, Bronze- 1ron 4, Iron 22,
Roman 6, Medieval 7, " Holocene" 28 sites.
(PTDOPLICHKO 1956., CALKIN I 966., TRING-

slowly cowards the North, Nortb-Easl from the region

of the E Alps and the Carpathes; the areal of Alees in
the Caucasus remained isolated.
ln the boreal I subboreal zones of our days, Alces

HAM 1969., LUCTUS 1969-70., TELEOIN 1985.

can be considered as an allochronous element of the

Tab 3. 135.)
At the beginning of the Mesolithic period, a fast
migration can be observed from the Late Glacial refugia of the mountainous parts of the Carpal.hes and
the Ca.rpal.hian Basin towards the North-Easl and the
East, getting gradually slower (W part of Switzerland,
France, Denmark, Poland, EasLem Balticum, W
Ukraine and Moldavia).
ln the Neolithic and Aeoeolithic period, Alces was
still frequent in Wl!stero parts of Sv."itzerland and
Western Ukraine.
During t11e Bronze Age generally described as cool
and humid, the Central European occurrence of Alces
decreased to minimal. Later on. during the Iron Age I
Celtic /Roman period they were appearing again on
the territory of Switzerland~ Morava., Poland and WUkraine. In Western Ukraine, during the 8-2 century
BC, Al<Jes was found on 22 sites from the Skythian
period. The occurrence frequency of elk. on the Central Russian Plain changed in a very interesting way:
from NE to SW, their number was decreasing. While
at the forested NE parts and the Orlov - Kursk forestal
steppe elk formed 19- 26% of the fauna, while Cervus
was either missing or occurring in very low quantities
( 1.5%). Al I.he same time, in I.he region of the Dnieprlngulce and the lower reach of the Southern Bug, the
ratio of Alces was only 1.5% while they were oompletely missing from Moldova. Ratio of Cervus at
the latter territories reached 26,5%, in Moldova,
even 51% (CALKTN 1966., 59., 86-87., Fig. 35.,
Tab. 77).
In the historical periods, A Ices was known in Large
numbers during the 9-lOth cenrury AD in the Eastern
Baltic countries, in Poland and to some lesser quantities in W-Ukraine and Moldova (PIDOPLlCHKO
1956, CALKIN 1962., PAAVER 1965., WYROST
1994.)
Summarising the chronodinamical changes in the
areal of Aloes we can conclude the followings:
During Lhe Upper WUrrn period, A!ces in Europe
was a typical submountainous animal living in the
Eastern Alps - Circum-Carpathian zone. lts occurrence was also noted to the North of the Pyceneans
(Tarn, Garonne-valley, DESBROSSE-PRAT 1974.
151} and to the South of the Lower Danube (N Bulgaria, TCUBIAK-NADACHOWSKJ 1982., DELPECH- G UADELLI 1992).
ln Lhe Upper Pleistoccme, the areal ofA!ces was disjunct. The closest neighbour Lo the East from the Central European areal was located in the Caucasus. After
I.he withdrawal of the polar ice cover, Alces migrated

fauna.
The present areal of elk. generally characterised by
great vagility (mobiLity) can be diviO'ed into several
parts. In the swnmer period, Lhey reach the forestal
tundra, otherwise they pre fer the deciduous forests in
and around lhe watercourses. They are migrating seasonally to different habitats according to the avajlability of nutrients.
Compared to its recent seemingly continuous areal
we can observe intensive long distance migraLion of
AJces witll long tem1 SW, S expansion throughout
prehistory, historical ages and modern times (HEPTNER et al. 1966. 283, BRJEDERMANN 1982. 99.,
TOPAL- VCROS l984 with further references).
Taking into consideration data from I.he past 200 years
the observer gets the feeling that Ille currenL bon~aJ
areal of A lees became conserved as a result o f natural
(fuunistical) and artificial (antropogeneous) factors to gether; fi'om time to time, elk tries to "break through"
these frames. The reason for these areal expansions
are not really known: it can be a result of environmental, climaHc and I or population dynamical motives. Anyway, the environmental factors of them ilder
climatic zones do not mean for Alces a pessimal habitat, with the exception of human interference.
lo Europe Alces, having a more universal distribution in the Upper Pleistocene R-W, Early W periods,
can be found ()nly in the refugia along the mountain
ranges during the w n-m period (submountainous
range, basins, Pyrenees Western Alps, Eastern AlpsCarpathes, further Caucasus. To answer the question,
where this species used to stay in the "elkless" periods. further fauniStical studies, mainly site chronology are necessary. Was it ''no elk" or " no bunter'' on
the area?

3.3. C ervus elaphus maraJ (Gray 1850)

5 remains - 3 individuals

...""'

..

There were relarively few remains of Cervus found

o n the Upper Palaeolithic camp.site Bodrogkereszc:UrHc nye , altogether 5 pieces. The anatomical djscribution of these are given below:
ill.63. E Sec. 2. D
antler tine piece (L. 57 mm)
ID.82. 4 tr. 3--4. 0
antler tine piece (L. 85 mm)
ill.63. D Sec.
s. me dist. fr. Guv.)
"stray find"
ILL.63. J Sec.
d. tibia dist. piece

27-30. D
s. tibia dist. piece

ll.82. 3 tr. W

159

The two distal tibiae were remains of stag.

Table 19. Dimensions of Cervus tibia disL (in mm)

Bodrogkcn:ttttir

Klskcvely Cave 3-4

SvCduv s1Ul 1

disL breadth
66,5
70,0

disL diameter

61

S2

63.6-67
52,)-56.4

Vctcmitla C2

58

Aszi>J (Late NroL)3

- male nl4

S7,0-67,0

- female n-81

49,5-57,0

Ccrvus llUll1!J (Neol Ocorgin~

RJ:ccnt {Hungary)>
male n-12

- fc:mnlcn-21

49.5-54,0
42,5-49,0
~

I. MUSIL 1962. 199. 2. MALEZ 1963. 121., 3. VOROS 1975.

19&3 a. Tabl. I., 4. BURCHAK- ABRAMOVICH 1972. Tabl.
42.

Red deer was the third most common species

in the herbivore fauna of Upper Pleistocene sites in
Hungary, foll0wing bison ahd horse$ (See Appendix,

'Fable H.).
Among the finds of Hungarian Upper Pleistocene
red deer, a larger form is constantly found tbougb in
smaller quantities. Tbe differences, surpassing (?)
sexual differences were observed by palaeontologists
fairly early and marked in the fauna list under various
names: e.g., Cervus ssp., Cervus cf elaphus, Cenus
elaphusf major. Cen111s cf maral, Cervus canadensis
asiatic:us Lydekker, Cervus marol Ogilby e lc.
The large form of Cervus was found on 25% of
Hungarian sites (22 locaHties), ..dominating" in
roughly equal quantities (6 and 9 localities, respectively) in the Szeleta and IstAJ16sko fawial phases. Red
deer was still present in the first phase of tbe
Pilissz.Ant6 faunal phase, later - by the WUrlll III cold
maximum - displaced by Rangife1:
In course of the cbronological body size analysis of
the Hungarian subfossil red deer it was obsecved that
apart from a general dec rease of size, differences
could be observed also in the size differences due to
sexual d.imerphism. On a given place. e.g., the Carpathian Basin~ rhe difference between tbe two sexes in
lhe arid continental period was larger than during the
humid Atlantic period (VOROS 1975.) On a larger
scale, e.g., Europe, proceeding from the Southern
zones towards the North the sexual differences decrease.
Knowing this as well as the 8-10% difference in
size between the male and femaleBisons should put us

160

on the guard. The possibility of having a large red

deer form in the Upper Pleistocene however is corroborated by finding among the small deer male
(small) and female (smalJest) bones as well. To put it
in different ways, the occunence of large size bones
on some sites would indicate the hunting on the biggest srags exclusively.
The fauna! history of subfossil large mammals in
the Carpathian Basin sh.ows well that the deer form
moral appeared in Hungary together with the Eastern
European wild boar, the Caucasian Bison, Capra
ae~arlfs (this latter. however, needs further proofs:
VOROS 1996. 46-47), Easlem European wild
horse "hemionus" (= "Asinus") and the Persian Liou
in a so-called second subfossi l fauna wave (Late
Neolithic - Copper Age: VOROS 1981 a, 1983 b,
1986 a, 1987 b.) The more drasticaJ changes in climate and environment during the Upper Pleistocene
could reasonably even more drive in the "maraf' deer
into Central Europe, more specilicaUy. the Carpathian
Basin.
The red deer specimens at BodrogkeresztUr belong
also to this "fonna major" and could be identified with
Cervus elaphus maraJ (Ogilby)
The dimensions of the distal tibia (Fable 19) show
the large dimensions of this Cervus fonn very well. It
is interesting to note that the red deer from Vetemica
C (Croatia, MALEZ 1963. 121.) was much smaller. It
was possible to observe much smaller red deer during
the subfossil Late Neolithic period on the South.em
parts of Hungary than the Lowlands, especialJy the
Northern parts of the G reat Hungarian Plain (AlfOld)
(VOROS 1975, 1983a.) Today we can already render
probable that tbe red deer form living in the illyricum
was smaller during the Upper Pleiscocene as welJ as
the Holocene chan U1e fore type in Ceotra1 Europe.
The large form of Cervus, appearing latest in Lhe arid
continental climatic phase of Lbe Late Neolithic
(4000-3500 BC) in tbe Carpathian Basin, did not proceed towards tbe South to NW Balkans and I or N
Italy.
The forest red deer, maral deer could be found in
the Upper Pleistocene universally, though in uneven
frequency; with the exception of the Eastern European steppe region where their presence was confined
to some river valleys e.g., Dniestr valley - Molodova
V. ~CHERNISH 196L}, Don vailley- Kostenki T., VI.,
XI V., XV., XVI., (LUCCUS 1969-70. 41., 49., 57.,
59.).
The red deer considerable withdrawal, decrease in
Central Europe took place in the cime of tbe WUrm ill
cold maximum, when this fonn was driven back to
submountainous - submediterranean regions towards
Lhe South.

3.4. Bison priscus ssp

Bison represented the most frequent element of the

Upper Pleistocene herbivorous fauna in Hungary. Its
dominance maximum was reached during the Szeleta
and the IstAll6sko fauna! phases (Appendix, Table 11

I 0 remains - 7 individuals

There were altogether I0 pieces of Bison remains

o n the Upper Palaeotithic campsite BodrogkereszturHenye hill. The anatomical distribution of these are
given below:
d. MJ/2 (I. 37, br. 19, b. 50 mm) 1.82. E Sec. "during
c leaning"

hum.dist. fr.
s. mcdist

lII .63. D sec. 19-22-25-28. 0

Ill.63. E sec., 24. 0
I. 82. C sec."surface"
111.63. E sec., 2. 0

s. lib. dist

s. astragalus
s. astragalus

lll.63. A tr. 38-42. 0

d. centrotarsale

111.63. J sec. 27-30. 0

111.63. J sec. 27-30. 0
ll.63. A tr. 2,6-30. 0

metapodium dist. fr.

metapodium dfat. fr.
OS

ph. 11.

ill.63. A tr. 26-30. 0

Table 20. Comparison of Bison bone dimensions of

Bodrogkeresztur and other UP sites (in mm)
I

I.
Tibia <fist.

Rodmgke rtszttir

,_Willcndori 1. I
Willendort v. 1
,__

4.

s.

90

98.S

91

SS

90

3.

6.

7.

60

31

44

I Mcllli:llI'J)US d.1SL
Bodrogker~n.w~

I.

l/\slrtlgal~
Bodrogkcrentur

I Ki~kevely Cavc3-l
I
I

3.

2.

95

69

58

97

68,5

SS
52

93

60

W1llend-Ofl' I 1
Bnbja jam112

92,5-94.0

61.0- 62,1

92,S

64,2

SS

Moravska :c.2

9S.2

6).8

56

3.

4.

5.

6.

1.

28
28,S

l8.5

35
33,S
36

38

Os phalangls II.

so

37

l.8

29

Kiskcvc!ly Cave 3-1

51

4()

JL
l!>

31

38

37
Sl
32
S7
42
- 34
I. THENIUS 1959. 158., 2. POHAR 1985.

4J
45

Wi lh:mlorf I. I

Without the horn core, the detennination of the

subspecies cannot be realised directly. The Bison bone
dimensions of the Bodrogkereszti'.1r agree best with
the dimensions measured at the Amvrosievka kill s ite
(983 individuals, 78 209 bone pieces identified,
PidopJichko 1956. 113., 20--21 OOO BP) which was
c lassified as short-horn Bison prisc11s MediaJor
Hilzheimer fonn (KROTOVA-BELAN 1993.
13 7-138.),
According to our present knowledge, Bison priscus
was known to occur still in the Baj6t fauna! phase.
During the Holocene, another fonn of Bison is encountered, Bison bonasus, more precisely B . bor1asus
hungarorum KRETZOI
(= Caucasian form,
KRETZOI 1946 b, VOROS 1987 b.). Finds are known
from the Neolithic and Celtic periods as well as the
early medieval times.
Recent archaeological excavations on the Greal
Hungarian Plain yielded very large size Bison mc-s,
from Mesolithic (Jasztelek I.) and MiddJe Neolithic
(Sza.rvas I.) sites, located along river valley terraces,
tbe size of wh.icb exceeded those of known fossil
Bison in Hungary. (Table 21). The presence of large
B ison in the Holocene can be explained in Lwo ways:
e ither they can be considered "surviving" B. priscus at
the beginning of the Holocene (or bonasus males?), or
they were simply of Pleistocene age . There were also
other instances of this: e.g. al Tiszaluc-Sarkad 1989.,
431 Sec. W. 1-2 spade levels, fragments of mammoth
pelvis (pans of acetabulum) were found in a Copper
Age pit, as a collected, " rarity" together with bo~ of
domesticated a.aimals. Currently, physical methods are
applied to decide the problem.
Table 21. Comparison of Bison bone climensions of
Bodrogkereszmr and other Pleist0cene and Holocene
sites (in mm)

2.

Dodro:ktres7.llir

with 35.35 sites.).

I.
Uodrogk~resztur

BVavli!gyi Rsh.1
lstall6sk6 Cave
Bronz Cave (Pllis
mts)
Budapest. Cor\in

sq.

The measurable bones of Bison at Bodrogkereszrur

(6 pieces) used to belong to male individuals, the
humerus, Lhe two merapodium fragments and the os
ph. U. arc smaller.

2.

3.

4.

s.

6.

7.

l'lcislocene

l<re.uzberg2

Amw<!Slc:vka1

(m11le)

55

- 90.a::

90

.n
-

44

90

235
230

75

42,S

75

4S 28

39

2l6

77

45

79

4K

28

42

80

56

-18

221

83

51

85

10

214,S-

75-93 45,4-58 83,5--98

2.46

(fi<mBle)

113-246 67,531,5

39,.5-St

70,~l

161

Table 21 . con.
3.

4.

s.

6.

5$

5(\

- -

8?

57

9 1,5

53

35

2M

73

47

68

29

Killmfuin!1i poL-hole
Cclffe

220

67

37,5

64

39

24

Szabolos6 Medieval

230

85

53

80

50

30

204216

66-73

37-U

6H9

20923 1

73-86

T45-52

6?~0

I.

l.

262

82

252

Pr:ehil;torit.

Uolocim.c

Jaszmlek 1

? Mesolithic
4

S7arvns 1.

'I Neolithic

l,jubUansllo b.5

7.

48

~
8

46

~ctcnl

Biulo\\ iel.it7
(fomah:)
(male)

I MOTIL 1938a 34-., 2. PHLEPS 1907, 40., 3. KROTOVA-EELAN

1993. Tnbl. 3 .. 4. B0K0NYJ 1987., 5. RAKOVE~ l953. 273-274., 6.
vOitos 1990bTabl. 4.. 7. EMPEL-ROSKosz 1963., Tabl. 13.

ln the Upper Pleistocene, Bisor1 prisous used to be a

typical steppe species of the open landscape areas.

Remains of mammoth. have been. registered so far

from 41 O sites in Hungary. On 71% of the 552 "strayfinds"/ open-air sites (392 localities) and 24% of the
75 Cave/Rocksbelter sites (l 8 localities). It is even
more striking t hat on Palaeolithic sites - bot.h Cave
sites and open-air settlements - the number of hunted
mammotb is acwaHy very low. Considering provisionaUy all the 18 Cave I Rocksbelter mammoth remaios
as products of human hunting, we can only add S
open-air settlement.$ wilh mammoth in the fauna
(Tata, Erd, Bodrogkeresztfu,Verseg and Cs.illaghegy).
The tusk lip of the Csi llaghegy baby-mammoth
was formerly published as reindeer antler
(GABORT-CSANK 1980. 238) and later corre.cted
(GABORJ-CSANK 1984. 7., L986A.). lts connectioll with the cultural .layer is not clear.
On the Lerri.tory of Northern Hungary where Bodrogkereszn1r is also located there were 146 lo.calities
with 513 pieces of mammoth booes found, presenced
in .a former survey (VOROS 1-980. 33.).
Their anatomical Cl.istribution is the following:
8 picees

sk\111 fr.

3.5. Ma111mutbus primigenius (Blumenbacb 17:99)

31

J:ruU1dible.fr.
molar, molar fT.

42 remains -8 individuals

tusk f:r..
vcn:brae

.'203
_
46
41

The topographical division of the mammoth remains

is ~be folJowiog:

cibs

40

1.82. D. &.~. 7 0
, 1.82. C. Sec. 12 D

Lamcll nc
ro111us;mundibulau ft. (5 pi~~)

11

I "sll'lly finds"

l tuSk aboral fr., pracmaxnfa fr. (5 pic=s)

II

sempula fr.
humerus fr.
radius fr.
ulna fr:

p,ihis fr.

Femur diaph. fr. (l.pic=s)

~ll' of mnndibl~ (Fig. 8.1. 20-25 years old),

filmur IT.

15

ffi,63. E. Sec. 13-lS. D. d onM2 (4-5 y~.otd)

111.C.J. G. s~e. 2. 0
ill.63. G. S.ec.. 5. 0

11ur diapltysis fr: (2 pieces)

1.2

I
13

m.n L 1r. s. o

M inf fr., Sd, symphysis mand. r~.

lib'Ju Er.

tusk. (slighl.I)' curved. ju"cnlle Fig. 8.:t 1n sicu

dimension.-;; I. 1800 mm, base diam. no mm).
femur dlBph. fr. (I. 26S mm, w. 70 mm)

fibula ft

m.&2. 2. u-. 6. o

cttqo\l.S I tnrsus

32

lli.63. A tr. 26-30. 0

M fr., long b.one diaph. fr. (3 pies)

mciWO<lium

l costa (I. 190 mm. 360 nun)

pracmaitilla fr., 2 M sup. fr:

111.63. A u.36-40. 0

m phi!laiigis

ffi.63. A er. 39-42. D

IV.63. C. Seo.
IV.63. 0. Sec,

M fr. long boac diaph. fr. (7 pk~J

tUsk fr.

prniimaxl11n fr.

mol.Sup. fr.

mandibulll fr.

mol. fr.

fr.

femur diuph. fr.

long bone dlapb. fr.

5
10

41 piccc!r

J.62

----

30

513 pieces

M sup. fr.

The ana(omical distribution of Mammoth i:en;i.ains:

COS!a

3
10

pE11ella

long bone rr.~

Among the of a1J mammoth remains (203 pjeces)

29% are most resistant teeth and its fragments.
Lo Northern Hungary, at 137 localities rhe mammoth died of natural causes and the rernaihs stayed on
the primary spot or were cransported to secondary
locations. There were 6 articulated skeletons or larger
parts of the skeleton found (Ape, Hont, Hatvan, Miskolc "eogine-.house settlement", Samsonhaza and
Zsujta.) At Jo_bbagyi in the valley of riv.er Zagyva,
there were remains of mammoth breccia fou nd at an
area of "25 x 50-60 feet in half meter depth, with
weathered bones and molars" (HALAV ATS 1899.)

found during the construction of the railway by the

end of the 19th century. The mammoth remains were
transported co the Hungarian Royal Geological fnstilutc; the teeth belonged to juvenile, prae-adultus and
adultus animals. The taphonomy of the whole serting
recall a typical flood catastrophe, similar to the conditions e:ncounlered at Visonta K-IJ open-cast mining
she where remains of Arch. meridionalis were found
in 1983 (FODOR-V6ROS 1990.).
At 8 localities - in 6 Caves {Aggtelek-Baradla,
Berva, Di6sgy3r-Tapotca, Lambrecht, Subalyuk, Szeleta Caves) and at 2 open-air sites (Bodrogkereszrur,
Marraderecskc) the mammoth was obviously present
as hunted pray.
On the 146 Northern Hungarian sites a ltogether
some 180-200 mammoth indivjduals were found. Of
these, we have information of the age group of 80 individuals given below:
neonatus 2, infantilis 4, juvenilis l 7, post-juvenilis 7,
prac-adu1tus 5, adultus 25, senior adultus 20.

lt is interesting to compare the mammoth finds

from Northern Hungary 10 the South-Polish mateiial
(KUBIAK 1965.) as well as the evidence found on the
Pl'edmosti mammoth-hunter's campsite (MUSIL 1968
b. in GUENTHER 1977.):
l lkg10
~em Hungary (individual%)

lllhlml l'Oltlnd (pieces %)

Ptcdlnost (picces %)

)'Dung

adult

liellior

37,5

37,.5

25

2S

2S
18,S

18,~

63

so

It is surprising that in the more or less random sample from Northern Hungary only halfofall individuals
reached sen. aduJtus I maturus age than in Southern
Poland. What could be the reason for this widely different mortality? [t is not possible to answer this question as yet without exact chronological and sedlmentological anaJyses. The following observation may
serve as a contribution to the problem: young individuals occurring dominantly lo ri.ver valleys
(JANOSSY- VOR6S 1979. Abb.2.) at the bottom of
loess bunches over fossil soil horizons. The young animals were probably victims of unfavourable natural
conditions.
At the Pfodmost mammoth hunters' campsite \lle
large number of young individuals reflect, similar to
hunted mammoth fauna in Hungary, specialisation on
mammoth calves and young animals. The same can be
observed at Tata and Bodrogkeresztilr as well.

Table 22. Dimensions of mammoth teeth (in mm)

MandibleM1

s.
tooth I.

t91}t

M sup. fr.

Lm

d.mM?

d.

~3

185+

31

toolh br.

7J

81

30

tooth II.

110

105

150

43

gs I.

90

115

gsln.

16+

II
14+

Ith

10
6

S,6

cth

1.~

J.5

1,3

1.4

~I.

lnd.:x

4+

II

+fragmented, incomplete piece

gs 1.:lengtb of the grinding surface, gs: In. lamelta

number of the grinding surface, In. lamella number,
ltb:lamella thickness. eth: enamel thickness, lndex:
number of lamellae in 100 mm
On the basis of the thickness of the tooth larnclla
(5.6-7 mm) and thac of the enamel ( L,3-1,5 mm) this
form can be assigned to the so-called "jung prirnige11ius" type, i.e., the evolved A{ammlllhltS primigenius form (KRETZOI et al. 1982. Tab!. 2.)
Dimensions of the mandible: maximal breadth 490,
length 440 mm, The height of corpus mandibulae is
154, its width is 123 mm. The interior width of tbe
symphysis is 80, its height is 88 mm. The distance between the lingual margin of the two M2-S is 90 mm.
In the Carpathian Basin the ancestral form of I.he
taxon Mammuthus prinugenius probably appeared already in the upper substage of the Biharian faunal
wave. At the Castle Hill of Buda, in the localities 5.
and 9. OrszagMz str., remains ofM. primigenius were
found in limy silt ''in loess-like porous calcareous clay
layer'' containing anLiquus (!) (KADCC 1939.,
VOR6S 1991 a, 1983 c. Tab!. 1.).
During the Upper Pleistocene, M. primigenius is
still a typical element ln tbe !stall.6sk5 phase of Lhe
Utrecht faunal wave (found at 4 Cave- and 36 open-air
sites, see Appendix Tabl. ITT.). Subsequently, it is disappearing from the faw1a.
On the Palaeolithic settlements younger than
~8-20000 BP the occurrence of mammoth seems to be
related not with the hunting activity but collecting
(VOROS 1991b., 1998.). The smooth swface of large
mammoth bones, mandil5l'e"S wereused on livin1ffloor,
or carved, split for anefacts or parts of artefacts. AL
Szob-lpolypart, femur diaphysis (GABOR! 1969. 6.,
Tai. I. 1-2), at Esztergom-Gyurgyalag, tibia (VOROS
1991 b. Fig. I.) was used for local activities.

163

3.6. Leo spelaeus G oldfuss 18 10

3.7. Lepus sp .

2 remains 2 individuals
Among the two large predators, Leo and Crocorta,
only Leo 's two bone remains of the feet were found:

The onJy rabbit bone of Lhe Upper Palaeolithic settlement was an ulna fragmen t (IV. 63. D. Sec.), lost
(KRETZOl 1964 a).

d.rnL ill

4. Big game remains on the bunting campsi.tc

S. OS

IU.63. G. sec. 21-24-27-30 0

Ill. 82. 3 tr. 9 0

ph. J.

4J. Q11antilative tlistribution

Bone dimensions (in mm):

m L Ul

_?Sph. I

I.

2.

3.

15

14.5

Sl

19.5

n.o

5.

6.

I -

l6

II

16

L-

7.

Bolh bones are small size.The medial length of

phJ. is 44 mm.
Cave lion and Cave hyena were constant representatives of the Hungarian Upper Pleistocene large
mammal fauna. They were found on one third of the
studied 180 sites (60 localities); at 27 localities.
together (45%), on 22 localities (36,6%) only hyena,
at 11 localities ( 18,4%) only lions were found (Append.ix Table 1). Both animals were typical large predators
of clearings near scrubs or open spaces. Hyenas were
more frequent (49 localities); their separate occurrences were double the amount of lions. lt is all the
more surprising lhat on open air sites, lions were more
frequent: 16% ( 6 looal i1ies) contained lion and on ly
8% (4 sites) containe--0 hyena. Jt seems hyena deserved
more its attribute name .. spelaeus".
The dominance of the two large ptedators was
reached by the Szeleta and tstall6sko fauna! phases.
however, they were still occurring in Caves and rock
shelters in the first phase of the Pilisszant6 faunal
phase. Due to climate deterioration, getting nearer to
W ill cold maximum, lion disappeared earlier from
the fauna than the hyena. Hyena lasted longer and its
site frequency was about double (7 localities) than that
of lion (4 localities). T hey were occurring together
only in the middle layer complex of PiLisszaot6 I.
Rock.shelter.

The estimated extent of the Palaeolithic settlement

on the basis of the surface distribution of finds on the
Henye was over 18-20 OOO m2. Some 2-2,5% of the
total area. was excavated, 423 m2. The collected 1803
pieces of animal bones were found on lt!ss lhan halfof
the excavated surface, 154 m2 (36.4 %) where the
average density of bone finds was 12 (" 11,7") pieces/
m2. Calculated for the total excavated area, 423 m2,
tlie bone find density was 4 ("4,26") pieces/ m2, one
third of the above value.
60% of the 1803 bone pieces were found in Unit I
( I 083 pieces); bone find density here was 19 pieces/
m2l

26,4% in Unit ill (474 pieces), bone find density 8

('7 ,8") pieces/ m2;
11,8% in Unit ll (212 pieces). bone find density 7
(" 7,5") pieces/ m2:
1,8% in Unit rv (34 p.ieces). bone find dens ity 4
("4,25") pieces/ m2.
Number of unidentified bone (UnlB) fragments
was J036 (57,5 % of total bones found), mainly frag -

ments of bone cortices from split long-bone diaphyses

and fragments of " bone plates". Their s ize varied between 5 and 50 mm. Their physicaJ frittering was forwarded, apart from intensive soil cultivation, extreme
climatic conditions as well. 67,5% (699 pieces) of unidentified bone fragments were found in Unit l; 19,6%
(202 pieces) in Unit Ill; 12,5% ( 125 pieces) in Unit U
and 0,9% (10 pieces) in Unit fV.
Among the 1803 pieces of animal bones, 767
pieces (42,5%) could be determined anatomically and,
consequently, taxonomically as well: half of them -

Table 23. Bodrogke~szttir-Heuye. rnsoibution of the animal species

1-57 ni1

Unit

rt-28 m1

IU-61 ml

b~ un~ts (Number of specimens, %)

I-IV L54 m 1
IV-8 ru1

pieus

"I.

pieces

~.

pleas

o/.

pir<Gf:S

pieces

1.

284

ss.o

>16

9.5

1~2

29.S

11

2,:?

100,0

-Alccs

8S

38,0

>16

17,8

97

4) .3

0,9

48]
124

Dison

Equus

1)

M~mmuthus

CcrvuJ

J
2

10

5
1

TowJ idtntlncd

384

so.o

87

11.J

unidentified (UnlB)

699

61.S

125

1083

60,0

212

164

Leo
Lcpus

TotnJ

100,0

42

3S.S

24

3,2

767

100,0

12,0

:?72
2.02

19.6

10

0,9

1036

100,0

11,8

47J

26,4

34

l,R

1803

IOO,O

50%, 384 pieces - found in Unit L, 35,5% (272

pieces) in U nit III; 11, 3% (87 pieces) in Unit fl and
3,2% (24 pieces) in U nit rv. (Table 23.).

Bone accumulation is highest in unit I.; less than

haJf in Unit TI and Unit II and in Unit JV., there is an
order of difference in bone density.

lnvcs rigating bone distribution in the different units, we find the follqwings:

juni1
-Detennined
UnlR
ITuui1

a.

pieces

384

35.5

Bi

699

64..5
100.0

125

IO!tJ

Ill
piea:s

%
41,0
59,0
100,0

pieces

202

IV.
%

pieces

272

57,4

202

42,6

474

100,0

10
34

rv

I
I

pieces

o/o

70

767

42,5

30
100,0

1036

57.5

101t3

100,0

o/a

UnlB: Unidentified bone

4.2. Qualitative distribution

4.2.1 Distriblltion ofthe remains by species
92,2% of the decisive bone remains (707 pieces)
used Lo belong to two species while the rest 7,8% (60
pieces) belong Lo five species (fable 24.).
The two most imported hunted big game for the
Bodrogkeresznir-Henye Upper Palaeolithic hunters
were horse and elk. The fTequency of horse remains

Among the 224 pieces of elk remains 43 ,3% (9 7

pieces) were found in Unit III.; 38% (85 pieces) in
Unit l.; 17,8% (40 pfoces) in Unjt 11 and 0,9% (2
pieces) in U nit lV (Table 23.).
lt can be concluded from the data that horse dominated in Unit I - the occurrence of elk here was 20%
lower. At the same time, elk dominated in Unit m
where ratio of horses was l 4% lower.
Among the two ma.in big game, the distribution of
elk is more even on the settlement: difference berween

Table 24. Bodrogkeresznir- Henye. Distribution of the animal species

unit, minimum number of individuals
I

Unit

p ieces

JO

46
40

Equus

28-1

74,0

Ale~

85

22,1

Mammlllhus
Bison

13

3,-1
0,5

I
2

>---

11

MNl

pieces

Cmiu:;
,__

mthe units (Number of specimens, % in

IV

"'

MNI

pieces

MN!

pieces

52.8

10

142

52,4

46,0

97

ls.it

Ill
19

20

?,4

3.0

I.I

0.3

272

100

53

1,2

I- IV

II

483

63.0

so

224

29,2

34

42

s.s

10

10

1,3

0.6

0.2

384

unidcnllilcd {UnlB)
Total

699

125

202

10

!036

1083

2l2

474

34

1803

32

87

100

16

were 63%, those of elk 29,2%. The frequency of the

accessory game was strikingly low, mammoth 5,5%,
bison l ,4%, red deer 0,6%, cave lion 0,2% and rabbit
0.1% (Table 24.).
Bone remains belonging to different species were
unevenly d.isLiibuted on the settlemenL There were
important quantitative differences observed (Table
23.). Remains of the two main game were found in all
four units but their accumulation frequency was different with.in the settlement.
58,8% of all horse remains (284 pieces) were found
in Unit!.; 29,5% (142 pieces) in Unit Ill; 9,5% (46
pieces in Unit 11 and 2,2% ( 11 pieces) in Unit IV.

pieces

Leo
lepus
Total identified

100

MNl
2

24

100

M.Nl

II.I

767

100

108

Limit I and ill is only 5% while the same value was almost 30% for the horse. ln lhe modest bone material
of Unit IL, horse remains slightly surpassed tllat of elk
while its frequency was slightly less than half of those
(fable 23.).
Among the accessory game, mammoth and red
deer occurred in 3-3 units (mammoth: Unit 1.-lll.-TV.;
red deer: Unit Il.-m.-IY.); bison occurred in two units
(Unit L- ID.) while cave lion (Unit ID.) and rabbit
(Unit IV.) in one.
Examining the units of the settlement separately for
the distribution of bones (Table 24.), the frequency of
the two main big games differs according to tbe quan-

165

tity of aCC(!SSOry games: 98,8% for Unit TI .. 96, I% for

UniL I., 88,2% for Unit UL
Horse dominated all units: 74,0 - 52,8 - 52,8% in
Units 1- 11-ffi; followed by elk 22. I - 46,0 - 35,8% in
Units 1.-n.-ru. Percentage was not caJculated for Unit
IV due to low number of occurrences. Most bone remains here (I I pieces) belonged to horses, followed
by mammoth ( 9 pieces).
Among the accessory games, 4 species were found
in Un.ii ill., 3 species in Unit TV., 2 species in Unit I.
and one in Unit II (fable 23.).
4.2.2 Distribution ofbone remains according to
anatomical regions

The anatomical distribution of animal remains

found at Bodrogkeresztur-Heaye hill is heterogeneous. It means that the single bones. bone fragments are
present in different number and ratio: some parts are
missing.
The method for classifying animal bone remains
found on archaeological sites - i.e., antropogeoeously
selected - into anatomical regions was elaborated by
Mikl6s Kretzoi on the animal bone evidence of the
Middle Palaeolithic site Erd (KRETZOI 1968 b. 230.).
Tbe animal was grouped, according to anatomical
build-up, muscles and meat cover into five body regions which were all different in respect of use or utilisation:
A. head region-skull and mandibles, teeU1, lingual
bone and perhaps the atlas
B. trunk region-vertebral column (cervical, dorsal,
lumbar and sacral vertebrae), ribs and sLemum
C. meaty limb region (limbs and shoulder region):
scapula, humerus, radius-ulna prox. parts, pelvis,
patella, femur, tibia-fibula prox. pieces
D. dry limb region-radius-ulna dist. pieces, carpus,
metacarpus, tibia-fibula dist. parts., tarsus, metatarsus
E. terminal bones-phalanges, caudal vertebrae, penis
bone.
The ratio of bones belonging to the ana1omical
body-regions. the dominance or absence of certain
regions, separation of animals represented by all body
regions from those which were represented deficiently can give information on the utilisation of the
animals, in our case, hunting specialisation and
methods.
Such primary data can be given fo.r the activities
performed on certain parts of the settlement by units

166

and within them , by squares, analysing distribution of

body-regions by excavation units.
Horse

The distribution of horse remains was more even

than that of elk. Dominant occurrences of the individual body regions by Units are given below:
bead region in U.1. 82.C. B. sec., in U. m. 63 A tr., G
sec.
meaty limb region in UJ.82. C. sec., U.U.82. 4-2. tr.,
U . ill.82. 3-4 tr.. Less in U. m. 63 . E-J-G. sec and A tr.
dry Limb region in U.1. 82. C-E-0. sec., U. m. 6J . J.
sec., U.IJl .82. I tr.
phalanges U.1.82. C-E. sec
From the trunk region, only i11 U.L 82., 3., 7.cr,
U.11. &2. 4 tr., U.III. 63. A. and 82 3-4 tr. yielded some
pieces.
The anatomical units, body regions of horse remains found on each square meter were given in Table
25. Summarising data, i.e., where and in which combination lhe horse remains were found the followings
could be concluded:
on the settlement, bone rl!mains represenring "horse
meat" were:
scap-hwn-rad and pelv-fem-tib, found in U.T. 82. D-CE sec. and U. UI. 82. 3-4 tr.,
hum-rad and pelv-fem-tib in U.ll. 82. 4-2 tr.,
hum-rad and lib in U.I. 82. B sl!c. and U. 111. 63. G.
see.
rad and ferrHib in U.Ill. 63 A tr.
bum in U. rn. 63. E-J. sec.,
rad in U. TI. 82. A sec. and U. rv 63 D. sec.
-The shoulder and (scap-bum-rad) and ham (pelvfem-tib) body-pelts caning from the trunk region was
seldom (in 9 cases) connecced to the lower parts of Lhe
leg, carpus/t.arsus-metapodium and sometimes phalanges. ln most cases, the limb was truncated at the
wi:ist/ankle joint: in case of U.1.82..E-B. sec.,
U.TT.82.2-3-4 tr. UJil.63. E-J-G. sec. and U.lll.82. 3-4
tr. 1he carpus/metacarpus bones, in case of U.I.82.D.
sec. 2. tr. U.ll 82. 4. tr_, U.lil.63. A tr. and E sec., as
well as u.rn. 82. 3-4 er., the tarsus/metatarsus bones
were missing.
At four places, we could find only the dry lin1b
bones, these were: metacarpus in U. ll. 82. I tr. and U.
m. 63. TT. sec and tarsus/metatarsus/metapodium in
u.m. 63. C sec and 82. 1. tr.
Os ph. I. was al~o found singly at two places,
U.1.82.E. sec. 2 D, u.m. 63. A tr. 40-42 D , other
proximal bones of the feel were missing. (Table 25.
Fig J!). {Table 25,)

Table 25. BodrogkeresztUr-Henye. Topographical distribution of Equus remains

ILocation

Kead

Tl'Wlk

Fore leg

Termina l bo11et

llind le&

ofreec

Uuir L '82
sec. 70
-D.
C
ID

pclv

.Re.

C" "c 2D

tooth

rad

sec. 30

51.~.

SClC.

amlcr

-C sec. 9-120
C sec 7-10-110

skull

E. sec. 20
E. sec. S-4-7-lOO
B. S<IC. 20
B. sec. 30

bl>

fem

tib

pelv

4-50
6-80

lib

tooth

hum

rad

pelv

(cm

lib

clrni

ph.11.

mdb

tooth

bwn

md

pelv

fem

Lib

mt

pti.1-D.

mdb

tooth

mp

ph.1-111

scp

fem
fem

lib

tooth

rad
rad

pdv

mdb

hum
hum

bl>

t/mp

pb L

IO()th

scp

hum

t/mp

ph.ll'TI.

me

pdv

fem
lib

skull

tooth

lib

tooth

lib

max

hlllll

t/m1

md/uln.a

ph.I.

lib

sesam.

A. stc. 20
,.\,sec 6 0

tr
b'

er.

rad
tooth

rill

I
r ib

tib

tlmp

vcrt

Unit Cl. '82

...,___

1-

I u 16CI
4-2.

ph.l

l!. ir.1-30

pbJL-m.

ml

mdb

--

B.scc. 4-50
,.__
B. sec. 7-80
B. s.ec.90

pclv

me

u I- ID

rad

rib

ph.l

lib

<I. tr. 50

t-

.I, tr. 70

rad/ulna

4-2 11 K-8CI

2.U. II 0
....._

pelv

tooth

slmll

fem
mt

hum

2. tr. 14.0

tib
(cm

4. u:. 20.0
3. tr. Ii
l . b.

ph.l.

tib

2. It. 18.0
....._

I -

d mc

tooth

phJI.

lib

hu.m

tooth
mdb

Unit Ill. '63

II U' 37.D

tooth

fem

c/mc

ph.I.

tooth

l.D
-lJ. sec.
sec. 27-28-30.D

lwm

mdb

I:. IC~. 13-15.0

hum

lib
lib

l.i. 1r.J-'1.D

-~

tooth
mdb

tooth

hum

mdb

moth

hum

rad

tlb

t/mp

skull
t/mr

mdb

tlb

2.. tr 6.0

~o0

lib

tooth

5.0

md

rib

max/mdb

A. lr. 40-42. 0
E. ~cc. 2.0

[m.s2

bl>

tooth

~~9.0

Ci. sec.. 19-2&.D

G._,_ 21-30.D

me

tooth

H. see.

IA U'. 26-30.D

rib

scp

hum

tib

rtld

IDOlh

IOOtb

rib

bum

pclv

fcm

lib

167

Table 25. cont.

Read

Local ion

Tru.nk

Fore l~g

rcnn in al bones
of feel

ffiudlcg

llnit IV. 63'

C. sec.

1ooth 1

D. ~t...

rad

Fig. 11 . Boclrogkereszu1r-Henye. Anatomical distribmion of horse bones

Elk

The distribution of eJk bones on the settlement was

more differentiated than those of horse. Remains of
the individual body regions were more separated
topographically from each other (Table 26),
For example, 90% of finds belonging to the head

region were fol.md in sections U.ITI.63-82. In U.I.82.,

t.here was only one tootb more (E.sec.).
82% of meaty limb bones were concentrated in
U.f.82. D-C., B-A sec.; 2-2 more pieces were also
found in U.11163.E and J sec.as well as 82 4. tr., l
piece came forth from U.U.8.2. 4. tr.
Body-pelts cutting from the lrunk regioQ, shoulder
(scap-l1um-rad) and barn (fem-tib were generally connt~cted to lower pan of rhe leg (carpus/tarsus-rnetapo168

dium, 11 instances). Basic differencec compared to

horse remains is the common presence of dry-limb
bones occurring in themselves at 6 places: Unit Jl82.,
t-2-3 lr. and u.m. 63. H-G. sec., A tr.

At six p.laces, carpus/metacarpus was occurring in

itself -U:I.82. D. sec., U. n. 82. 4-2 tr., u.m. 63. E-G.
sec.; at four instances, tarsus I metatarsus bone was
found alone: U.L 82. B. sec., U.n. 82. 2 tr., U. ID. 63.
J. sec., without higher proximal leg bones (Fig. 12).
Finger-bone was .found at one place in the company
of a .rib bone (U .III.82. 3 tr. 7. ?).
h can be scared that on the Bodrog,kersztfu settlement there were less shoulder-ham parts and more
"dry limb" bones for elk than for horses. (Fable JI).

teeth
19sup
14 int

2
16
5

---- - 5- -- Fig. 12. Bodrogkcrcsztlir-Henye. Anatomical dis1ribution of ellt. bones

Table 26. Bodrogkeresztur-Henye. Topographical distribution of Alc:es remains

Location

~ f. '82

liuid leg

rih

me

hwn

S<.'C. 4-.>-60

'

me

rad

me

fem
fem

t'. scc.10110

!!.sec.

scenebra

hum

5Cap

me
rad

160

m1

lib

ml

tlb

ml

lib

ml

lib

mp

ph.I

1-

rib

mdb

4 lr. 60

4-:?. lr. K-80

1001.h

1. lr 11.0

c
c

mdb
I

l. tr 13.D
:? u 16.D

- -

lib

clmc

~ni1Il. 1 8l

3 uW

IOOlh

B. sec.. 34-90
A. s.:c.1-3.0
lr.

pb.I

me
rad

Terminal bone:s
offetI

cih
Lib

verl

rib

(sec. K-90

IL

I tib

cc.{0
ec. 80
C <ec. 20

l"

Fure leg

T1'11llk

Head

tib

mp

tih

Lint

ml

rib

mp
me

Lib

-,
-i
169

Table 26. cont.

Lclaulun

Unir Ill. '63

II. sec

IIcad

-l

tmLllU

Air F D
\ 1r 39.D

,_.'\ tr

~db

~cc.

10- 150
E. ie\:. 24.0
j $~(.

J SC.:.
} Sl:C,

mdb

4. tr 3-4 0
] . IT 4.0
3 IT. 7 0
3. tr. 9.0

rv. '63

- ,...,...._
I

... skull

'

,_

ml

- ,_

'

me

I
scap

-1

IOOlh

>C.'innl .

'

rib
mdh

ph.TJ.

n'b

,__

ph.I

lih

rib

--r

ioolh

lib

Mammoth
Remains of mammorh were typicaUy found in
" patches" al aJI three excavated unilS wic:hin the sertlemenl with the exception of U.U.82. Thus, a mandible
fragment, LUSk and pracmaxilla fragment found in
U.l.82. C.. sec., looth, praemaxilla fragment, rib and
long bone fragments in U.111. 63. A tr.., pair of mandible~ and 2 femur diapb. fragments in G. sec., mandible and tooth fragment in U.111.82. l tr., tusk and femur diaph. in 2 tr., tooth in U.IV. 63 C. sec. and tooth
and long.bone fragment in D. sec.
Real ..meary limb bones" of mammoth could possibly be the femur and femur and/or tibia long bone
diaphysis fragments. The two ribs belonging to the
ITunk region could also represent meaty parts. Still the
general impression on rhe rype and distribution of lhe
bones can be that the mammoth bones, especially the
parts of the head region were not taken to the settlement as a pan of th~ diet.
ll could be proved for U.111.63.G. Sec and supposed
for U.J.82.C. Sec and U.III .82. 1. tr. that the mammoth
mandible served as an "anvil" (flal, smooth bone
plane) used for the production of the stone
tools.(VERTES 1966. 13) Similar role can be attributed to the femur ("'long bone") fragments al
U.ill .63. G. sec., 82. 2. rr., U.IV. 63. D. sec.

170

me
c:

..

-~--;;lib

tadlulna

mdb

Unh Ill '82

Unit

roolh

~.o

0 s.D

>---

toath

mdb
mftl\

J sec.
G !lee
,__

rndb

mdb

rad

scap

--

rudb

21-30 0
30 0

~np

mdb

28.0
26-29.0

rad

Tcmlinul bunt~
of feet

lllud leg

--

lm"-\'.tndb

-J. ~c.

1uolh

-1()-12.0

I ...:c 2.0
I;.. Re. 8 0

h)()lh

_J

ore ltg

Trun.k

ph.I

Bison

Remains of bison were found only in two units: at

U.l.82. C. and E sec., as well as U.111.63. A tr. and -JG.sec. Apart from one tooth (UJ.82. C. sec.) and one
os ph. ll. (U.llI. 63. A tr.) all the bison bones found belong to the so-called dry limb region. No bones belonging to trunk- or meary limb region were found in the
excavated area. (Table 30) The meal of the bison was
consumed by the Palaeolithic hunters but on Lhe settlements it is more typical to find dry-limb bones and
phalanges while crun.k and meaty limb bones were accumulated mainly at the narural death places (pothoJes., clefts, waterfront ere. thanatocoenoses).
Red deer
Remains of red deer were found at 5 different
places oo the settlement. One aatler tine each at U.11 l.
63 .E. sec. 2. D and at 82. 4. tr. 3-4 O; fragments of

one libia each al U.llJ. 63. J. sec. 27-30 CJ and U.U.

82.3. tr., the fragment of a metacarpus at U. TV. 63 . D.
see. Apart from the small antler tine fragments, the
three bones belong to the dry-limb region (Table 30)
C:avc lion

Both lion bones were typical for parts remaining in

the flayed hide /mt/ of the dry limb region and a fin-

ger-bone (os ph. 1.) of the terminal bones (Iable 31).

The split mt was found in Uni1 lfl.63.G. sec. 21- 24,
27-30 0, the OS ph.I. was in Unit m. 823 LT. 9 D .

bones of possible "pilferers" of the nearby (?) butchering site (hyena, wolf, fox etc.) found. Collecting
micro-mammals is a question of excavarion technique. At Bodrogkeresznir, collecuon of micro-facies

Rabbit

was not practised; the presence of rodents on the

The only and lost rabbit bone (ulna fragment) belonged to lhe meaty limb region. Rabbits were typi-

site in either its active or inactive phase was documented on the surface of the bones (KRETZOI
1969 b.).

cally exploilcd as source of meat and ll.ide.

Table 27. Bodrogkereszh'.lr- Hcnye. Individual number of animals in the excavarion units
Onll

n. si.

L '8l.

0.

r.

:.

B.

Fquu5

:i

,\Jus

g..,~1100

Ill. '82..

A.

rnmcl1

H.

2.

2
l

Ii.

1.

.i.

3.

A,

tbus

Bison

l I

I--- -

c.

Mmnmu

c~r\'u~

J.

3.

Total

I C.
I.

1.

I I I

3.

13

:iU

34

]()

1-

l
I 2

.i.

13

'9

II

Su1n

D.

Leo
Lcp11!.

1\. "63.

Ill. '82.

E.

:?

I
108

Table 28. Bodrogkeresznir- Henye. Individual number, age distribution and amount of usable meet of bunted

animals
Age jlroup

~~ullUll

jU\'Clll!iS

MNI

MNI

nl~UI

so

8000

3-l

401200
241130

10

9 500

Bi&on

71200

1400

Co:rvus

-l

3/11)()

300

L eo

'.!

:?160

120

Eqo~

-Akes
-Mammulhus
-

Lcpus
Total

_,_

I'.?

I
108

MNI

meat
80

4050

1:so
S,8-0

450()

l 100

-100
100

1160

60

18372

'l IS

13SO

13SO

IOlal amount ofu.~<'

nlfuntilis

MNl
2/Sn

meat
100

100
10462 kg

m.-:n.

s. Hunting-gathering strategies at Bodrogkercsztur

640

5.1. Species used as food

UP campsite

Al I the 1803 pieces of bones found in the excavarcd

area of Bodrogkcreszttl.r-Henye hiU could be classified
as cultural or archaeozoological finds. Animal bones
of all the seven species come from animals killed in
hunting. These were horse, elk, mammoth, bison. red
dee1: cave lion and rabbit.
Casr antler of elk, red deer tines and mammoth
tusks were also coUected.
Local small mammals were not found among the
species of the open air site, neither were the

The minimal number of individuals (MNT) determined on the basis of the hunted bone remains found
at Bodrogkereszuir-Henye hill is I 08. The areal disrribution of the individuals belonging to cerlaio species more precisely, pans of their body - is demonstrated
on (Table 27). Most species 6 and maximal number of
individuals (44) was observed in Unit UI. 63. Five
species were present represented by one individual
each in Unit IV. 63. D. sec. Apart from 29 indh iduals
of the two main big game (horse and elk) there were I
mammoth and 2 bisons in Unit r. 82. In Unit U. 82.,
171

altogether 14 individuals were found with only I red

deer.
Mosl of the hunted mammals were killed in ad11lt11s
(92.5%, 89 individuals) or .wbadtJlllls (9 individuals)
age. The ratio of developed individuals was 80% for
both main big games, tbat is. 40 horses and 27 elks.
(Table 28)
9 indiv.iduals represented subadullus age, all of them
horses:
Cea

mammoth bones were obtained with and for meat, we

could add 400 kg more.
None of the hunted mammals were taken totaJly to
lhe site as demonstrated by the anaromicaJ distribution
of bones, only some parts of their body.

I indhid11al

belwccn 2.'.i-3 Y<!ll:->

J )'ea!X

I individual

6 iru!Mdual~
I irum\dWll

be1wee1,1 3,()-3,S Y=

between 3,5--4,0 years

On the basis of the individual number ( I08) hunted,

calibrate.cl by age group data (Fable 28), the theoretical
useful meat q~tantity could be 20462 kg, with considerable amount of fat.
On the basis of the skeletal parrs (fable 29) found
on the four excavatiqn surfaces, less than half of the
potential mea.L quantity (6 902 kg, 34%) was present
Of this total weight, horse-meat couJd amount to 4750
kg, elk. 1950 kg, bison 200 kg, rabbir 2 kg. I.rt case the

Table 29. Bodrogkereszair- Henye. Anatomical distribution of hunted auima.ls (pieces)

Alets

Equllll

Bone

antler

Mamm11 1l1~

Diull
m.uu.liblc:

fl,

12

2.4

\t)lllh

ll :i

33

11

2
19

6
.;

\'crtebra

--

2l

C0$111

sten.:bm
SCllpl\ln

IS

h11mcrus

24

r<1di,us

33

I?

llln.a

CTil')lllS

me1acarp11,~

lO

-----pelvis
femur
pau:lla

llbill

')6

I.at.SOS

18
8
17

mellltlltiUS
mctapodlum
ph. l.

Ctf'\'U~

L~pus

-- -

-5

ph. lll.

'.!

sosamnidrum

hone fr.

r
I

--

3
l

19

12

--

,,

- --

l
10

483

33

14

12

16
IS

42

ph. u.

To1nl

Bl!oo

224

.i2

10

On the basis of the dentition status, tbe abrasion of

the chewing surface and the ossification of the eplphy-

In case of the two main big game, horse and elk,

both the anatomical (Table 29) and the body region

ses, the distribution of the young individuals can. be

giv.en below:

distribution show e.ert.ain simUariries and differences

Rcddeu

7 indlv:iauals: 1 of them. 8~10 months olli, S individuals

14-16 m<>1~ths old
l irulivklunl: eta.1 Y= old
I individual: CGB. 2 yean old

MammQlh

I lndividu.il: +5 years old

Elk

Horse

172

(Table 30. /l'ig. 15).

For both species, the contribuLion of the bead region - especially teeth - was relatively higp: ho1rse
27,3%, elk 29,5%. At the same time, presence of tire
trunk. region (horse 4,6%, elk 9,8%) and ph.alanges
(horse 6,6%, elk 3,1%) was negligible. The basic difference between the remains of the two main game

Table JO. Bodrogkeresztt'.rr-Henyc. Distribution of hunted animnals according to bodyTegions (pieces,%)

h<ad

sp~'jl l~

ll

pieces

pietes

pims

picm

Equwi

483

132

27.l

d,6

--

Akes

:24

66

29.5

222
J9

Mwnmwlms

42

25

B illon

Hl

22
22
2
-

Ct1~w

-Lepus

-Leo

9.8

''Ml
46.0

pic:o:~

17,4

90

7S

pieces

3~

6.6

40,!

3.1

I
I

-1
A
B

30

20

D
E

Eq Ac

Eq Ac

Eq Ac

--

Eq Ac

'h
15.S

11n~m

15

di) limb
D

n(Sb) hmb

fr\llllc

head
t runk
meaty limb
dry limb

terminal

Eq A

Eq Equus
Ac Alces
Fig. 13. Bodro~res7.tllr-Henyc. Distribution oflirirse and elk remains according to body regions(%)

was observed in the ratio of meaty Limb region versus

dry limb region (Table 30): while 46% of horse remains was belonging to the meaty limb region, only
17,4% of elk bones came from the same body region.
To present the same phenomenon from the other side,
while the dry-limb region constitured only 15,5% of
total horse bones, the same ratio was 40,2 for elk.
fl means that the meacy - dry regions of !be two
main game showed adverse tendencies and, consequently, utilisation (Fig 11- 12).
What primary conclusions can be drawn from the
above data?

The two large herbivores were of different milieu

and habit; these animals could not be bunted in the immediate vicinity of the seltlement. This is funher corroborated by the apparent Lack of trunk bones. The total number of trunk bones found comprise 3 vertebrae
and 40 rib fragments and one stenebra (Table 29,30);
these bones altogether would not be enough for the recons1ruction of the vertebral column of one ru1imal not LO speak 84 indlvjduals (50 horses and 34 elks).
Both big game had been butchered at the hunting
site. The bide was stripped off there, 100. It is interesting to note that for the artyJodactilous elk, 7 pha-

or

173

UniL Ill. 63. A tr., E-J-G sec.,

UniL TIT. 82. 3 and 4 tr., on the surface of each
squares
elk meat- Unit 1. 82. C-B-A sec.
Unit lll. 63. E. sec.,
Unit 82. 4. u. on the surface of each squares

langes, for the perissodactylous horse, 32 phalanges

were found on the settlement. The feel and the phalanges were cu1 off from the fore and hind legs. Still,
on the relative!) less numerous horse feet more Of
phalanges were found than on the elk feet.
Jn thc meaty linib remains of horses (222 pieces)
there were more than two tin1es more hind legs ( 150
pieces) than fore legs (72 pieces). ln case of the elk retn?iru;, the difference in the same region is negligible:
bones belonging to 21 fore legs and 18 hind legs.
On the basis or the two main big games hunted remain distribution (Table 30, 3 !) the meal can be stated
thar the mealy limb region as well as the head region

rabbit meat- Unit IV. 63. D. sec.

Reooains of feet detached :from the limb (c/t, mp,
pb) were observed independently at 6 places in case of
the horse, a't I 1 places for elk, at 4 Rlaces for bison, at
rwo places for cave lion and one case for red deer (see

Table 31. Bodrogkereszhir-Henye. Body regions of htmted animals according lo body regions (individual cases)
PW'U or 1h~ llc>dy

Fo~limb

TnlrtJ(

llc:id

Bone

Sk.

mdb

Vertirib

Sp

EqL[\lS

28

11
13

7
8

Akes

Mamm111hus
Bll

9
6

L4

Mc

16

rv
9
-

Mlimp

J>h

10

.25
13

12

14

10

3
l
I

F~rs

lllcu! limb

c:
4

C<.'T'VWI

Leo
Lcpu:s

H
15

I
I

Sk.-skull; mdb-mandible; Vert-vertebra; Sp-scapula; H-hunierus; R-radius; U-ulna; c-carpus; Mc-melacarpus; Pv-p.eJvis; Ffemur; T-tibia~ Harsus; Mt-memtarsus; mp-metapodlum, Ph-phala.ngis.

of the horse was primarily accumulated on the settlement. That is, results and product of horse-hunting
was lransponed to the excava1e.d parts of the .settlement and was consumed there.
Ln case of elk, the situation was differem: bones of
tbe dry lim_b region and the head region dominate.cl the
assemblage. Apart from "occaslonal" consumption of
elk meat, signs for a meat depot or meaty bones were
missing from the excavated parts of the settlement. lt
seems as jf the place for elk-meat consumption were
different .from the parts ex.ea vated. Was the BodrogkeresztUr settlement only a temporary stage in elk hunting? The frequency of the head region (29, 5%) seem
to contradict this.(Tabte 30)
The archaeoioological evaluation of the material is
made more difficult by tl1e scanty evidence on horizontal and vertical periodicity of the site. We have no
information on important circumstances: were the four
units utilised at tbe same rime or separately, fot bow

chapter 4 .2.2. Table 25. 26). Tt is probable that these

terminal dry bones were transported to the settlement
in skinned hide, and were cut off fro m the end of foot
at their spor of occurreoce.

long and which pe1iods1

pattern.

On the basis of the limb bones representing "food

residuals" on the site, local stripping and I or conswnption of meatcouJd be localised at the follow.ing
micro-regions within the Bodrogkeresztur site:

horse meat - Unit 1. 82. C. sec. (dominant) and adja

cent B and E Sees.
Unit n. 82. 2-4., Easrero halfof 3 tr.
174

5.2. Envirot1me11t- lluuling strategies

Tbe slubbered Upper P'leistocene vegecation history
of the Carpathian Basin .bas recently started to formulate chron.ologically as weU as regionally as a result
of the activity of P. SUmegi and his team (KERTESZ-SUMEGI 1999. with further references).
The faunistical examination of large mammals,
similar to ilie study of birds seem to exclude completely the previous old static view on the exclusive
dominance of "tundra'' and "steppe" vegetation zones
ia Hungary. Ve.getation zones determined by surface
morphology and river valleys were mix~d in patches,
the individual compo11ents appeared in a mosaic-like
Large he~bivores - ilil our case, hunted pig game were living in the environmental boxes suitable for
tbeir requiremehts. The natural caducicy of the animals
moving along river valleys or to drinking places was
seemingly most different, depending on the size of the
animal and taxonal characteristics.
for example: in the current bed of the river Tisza,
remains of horse were found only at two places (Pe-

roszlo, Szolnok) opposed to mammoth (47), wooUy

rhinoceros ( 19), bison ( 19) and Cervidae (Megaloceros 19, elk 13, red deer 8, JANOSSY-VOROS
1979.).

The more variable was the zonality of d1e vegetation, the more mobile the large herbivores could be. so
much the more species with different milieu could be
found in a given region. By the srudy of the fauna! remains, mixed species assemblages of widely different
areal could be reconstructed.
The landscape structure of tlie Carpathian Basin
was much more variable than the Western European
territories or the Central Russian Plain in the Late
Upper Pleistocene. Consequently, it is essentially different from a faunisticaJ point of view as well.
All ll1e faunal remains of big game at the Upper Palaeolithic open air site Bodrogkereszrur-Henye, with
the possible exception of elk and red deer antler pieces
came from animals killed in huncing.
The species composition of the hunted animals
was poor, typicaUy reflecting antropogeneous selection.
The Upper Palaeolithic hunters were specialised on
the kill of two large herbivores, horse and elk. 92,2%
of the animal bones which could be identified (707
pieces) and 77.8% of the hunted individual game (84
individuals) used to belong to these two big games.
The rest of the bones (7.8o/o - pieces) were distributed
among 5 species - mammoth, bison, red deer, cave
lion and rabbit, altogether 24 individuals (Table 24.).
The hunting strategy of the Upper Palaeolithic hunters
was generaUy based on four main big games (horse,
bison. red deer and reindeer; GAMBLE 1986. 106).
At Bodrogkeresztur, of all hWJted herd animals horse
was absolurely dominating (63% of all bone pieces,
46,3% of all individuals) and the occurrence of bison
and red deer was low (1,9% of pieces, 10 of individuals, Table 18 ). Reindeer was missing from the fauna
due to faunistical reasons.
Elk served as an alternative resource for Palaeo1ithic hunters (GAMBLE 1986. 106. , 109.). At Bodrogkereszliir Jt was the second mos! frequently occurring bunted game (29,2% of all pieces, 31,5% of all
individuals).
This is not surprising as the faunal dominance of
elk was reached during the lstA116s:ko fauna! phase
(64,3%, App. Table VII.)
The mobility of big game and their tendency to
form herds is different Apan from the resident red
deer, the other big games - mammoth. bison, horse
and elk - are migrawry species. The aggregation potential at set seasons of t11e year is high at the bison,
medium for mammoth, horse and red deer and /ow for
the elk (GAMBLE 1986. I 05 .. Tabl. 3. 10). ll 1s inceresting to note that Gamble classified elk to residentaJ
species.

The living space of Upper Pleistocene large mammals, their characteristic milieu was not necessarily
identical to those of the species living today. We can
say, conventionafly, that horse was a steppe element,
bison steppe or forest-steppe elemt:m, red deer characteristic of the forest while mammoth seemed to tolerate a wide range of environments. Elk generally indicated a marshland environment but it is also known
that in summer elk rather avoided marshy environment because of tbe insects (STURDY-WEBLEY
1988. 263.) Knowing the elk diStribution data in the
Upper Pleistocene-Early Holocene (Tables 18. Fig.
9- 10) it is difficult to recognise today's peripheraJ
taiga belt. It is known, however, that elk would break
out towards the south from time to time (TOPALVOROS 1984. with further references).
The basic environmental motive for the movement,
seasonal migration of large herbivores is a lways winter
food; obstacle being thickness of snow Layers and the
eJnent of flood covered areas during spring-time.
The Upper PaJaeolithic bunters at Bodrogkeresztur
would chase hunted around the Zemplen-Hegyalja
hills, close to the local floodplains of the rivers
Hemad-Takta-Bodrog-Tisza and in the gallery forescs.
Single or herding animals were probably captured by
col'lective hunling by a pack. The actual hunting
weapons could be identified by functional analysis C>f
the stone implements. Former "bone'' (antler, mammoth-tusk) points of the Aurignacian hunters were
missing from Bodrogkereszltir. Bow and arrow are
"long-distance" kill weapons penetrating deeply into
the body, injuring I stopping the animal, and can be
lethnl in fortunate cases (FRAYER 1981 ). Probably,
tilting spear musl have been in use as well, by the help
of which the injured animal could be killed at close
.range.
The consumption of horse meat, its storage was
dominant in U. l. 82 C sec. and its vicinity, sec. B-E.
(Fig. 2.). Further meaty parts were found at UJI.82. 42-3 tr. (Fig. 3.) and U.ITI.63. A tr. E-J-G sec (Figs.
5- 6)
The "occasional consumption" of elk meat took
place, inrerestingJy, at the same place where the depot
of horse meat was found, i.e., U.l. 82 C-B-/\ sec. as
well as U. m. 63 E sec.
Meaty bones of bison and red deer were not foun d
on the excavated parts of the sertlement (Tables
29-30). 80% of both horses and elk hunted were adu 1tus individuals indicating developed bunting technique and organised hunting strategies. On the Upper
Palaeolithic sites Sagvar and Nadap, head region (in
the first place, teeth) dominated among horse bones,
followed by the dry limb parts (Table 32). At the same
time, the material of the reindeer hunrers campsite
sho' different character: Pilismar61-Palret was a killand depot site, Sagvar a transient campsite, while
175

Table 32. Distribution of body regions of Equus on Hungarian Upper Palaeolilhic campsites (pieces, %)

li~d

-Trunli

1---'-m~llY lintb

l>ry limb

Nada:p1

S6gvu1

R.:glon.ISitr

'

Pi.:ces

PI~~

t8S
I

I:

7:l ,&1

SS

SS,56

132

0,.11

6
42

--

Phelan~

8odrogk<:re$%1i1r

Piece.~

!Cl

2~

%
27.3

i .88

?2

4.(i

2,4.5

11

7,1 8

212

-l;l,0

17.22

.;,o

75

I 5.S

41
7

100,0

15.3

31
483

100.0

26,l!O

4,38

100,0

6.6

l. Voros I !'J&2., 2. Voros 1988b.

Table 33. Djs1.ribution of bedy regions ofRangi/u and A.ices on Hungarian Upper Palaeolithic campsites
(pieces,%)

Regl>tJ' :...

I
I
I

H~a.t

t --

----Trunk
~~linlh
Ory limb
Phalangi;s

rI
I

R:mg\fCU'

sngvw 1
l'i~ccs:
10~

48
.S6
2Jl
5-lt
l.!lll!

--

flllL~i.t.)1.filn!tJ
~.

Pieces

11 .:\'}

J '1

120

12.06
;0,10

116

29,7S

396

IS.61

:ill

49

1'4',87
I ?,St.

~31

J.98
100,ll

408

1\10.11

1-410

).3:?

-.------

I
I

PillS$UnJI> t. ll'lh 4
I
%
l'ieci:;

3,76

UC)

'"

'U3

A Ices
1Wrogl-<1rcS'll~r

''

Pieces

'!r~

66
22

29.S

--so
I

15,Rl

28,08

I
l

23,62

"JR,94

3.1

lOl.i.O

224

loo,u

3.54

39

9,11
17.4

911

40,2

I. Voros 1982. 3. Voros l983d ... 4. Vt\~os 187c.

Pilisszant6 L. rock-shelter v.as a rciutker meal depot

(Table 33.).
The mammoth femm (and fo::lllW' u1 tibia) belong to
lhe category of meaty bone. lwr ;- not, ef) likely
U1at the whole cut-off limb could have been trans-

ported to Lhe settlement The r.neat of the "mega herb.ivore" mammoth was probably cut off from the
bones. The large bones or parts of their diaphysis were
probably colleoted from the kiU sile or other places.
PaJaeolltliic hunters huoted not only rnammotb calv~
bm also full-grown males as well: this was proved by
Lhe mammoth skeletons from Dunat<>ldvar and Mitradcrecske, where stone tools (weapons?) were fow1d
associated with the bones (CSALOOQVITS 1936.,
VERTES 1954., VOROS 1975., BTRO 1984.) Dunaf6ldvar was a kill-s:jte when.: rwo juvllni.te 1nammoths
and a bison were captured from fhe animals wbich

used to go J:here for drinki11g. The prey was butchere.d

on spot and parts canied away. For example, only lhe
vertebral column of the bison were le.ft behind. At
Mau:adl!Tecske, a die.cl (?) mammoth was hunced 00
the site, proved by the existence of the complete skeleton. The head, leaning on the tusks, was foun,d in verlical position at the Jefl si~e of t11e body. The carcass
was net disassemb)ed. A large stone tool was found
6-7 rn far. from the head of ilie animal in the same
level.

176

MammoU1 l\JSk was used for the production of thin

spear-heads, "Lusk stic~" - this \VBS special ly characLi!rislic of Lhe Aurignacfan culturl!. AL the si.tc Verseg,
from habitation surface Nr. l. (C section 4 block, - 120
crn) an arched triangular "spoon was found, made of
mammoth tusk (VOROS I 99lc. 88. Figs. 4-5). Similar but bafted specimen, pierced a,nd ornamented with
rows ef dots described as "amulet was published
from :arinzeoi cave, Moldavia (DAVID 1980. 47-48

Fig. 15.).

We can calculate the hunting season on Lhc transit

campsite Bodrogkeresznir, based on the age of some
young animals:
an1ong the elk calves. 2 individuals were .hunted in
spring (U.lli.63. A. Lr., lf.ID.82. 2 Lr.). 5 individuals
were caprures by the end of Slimmer-beginning ef
autumn (UJlI.63. J-G. &cc., U.T.82. E sec.);
arnong the young horses, 6 individuals were taken by
the end of summer-beginning of autumn (l,1.1.82. E-CD. sec., UJil. 63. E-J.1-2), one individual was killed
in spring (U.Ill.63. G,sec.) and cwo in winter (U.L82.
3. Tr., U.Il. 82. 3 tr.)
The total amount of useful meat which could be,
theoretically utilised was 20 462 kg (Table 28.), of
which 34%~ 6 902 kg meat was actually documented
by the bones found on the excavated parts ef the settlemenL This was a considerable amount of food in itself.

5.3. Clrro110-bios1ratigraplty

The lithic material of !.he BodrogkeresztUr campsite

used co belong to the Upper Palaeolithic Early Gravettfan I Pavlovian emity, , whicb can be dated chronologically to the second half of the lstaU6sk6 fauna!
phase (W H/111., Stilfried B interstadial). The 7 hunted
mammaJs found on the site may occur any time with
the exception of the end of the Late Pleistocenc. The
terminus ante quam for the site is the Szeletian phase
(with Megaloceros) and the post quam is the
Pilisszant6 fauna! phase without Alt:es and Mamn111th11s. The large Equ11s remage11sis, known so far
from Lhe Late Lower-Middle Wtinn was an ..archaic''
element in the fauna, but Lhe typical small-medium
size horse of the Upper Worm, E. "germanicus " was
also present. The large Cervus elaplms maral, Bison
priscus ssp. with mediator bone dimensions and Mamn111thus were, in this case, indifferent chronospecies.
Their presence, however, was important for the eco1ogicaJ provenience of the fauna. Elk, red deer and bison were forestal species; among the horses, the large
form was probably of foreslal area and the smaJl-medium size form typical of rhe steppe region.
The Kopasz Mt ofTokaj is a member oflhe Zemplen-Hegyalja, protruding towards the South like a
peninsula. It was and has always been in the junction
of human and animal routes. On the basis of the reconsrruclion of the environment of the Kopasz Mt. and its
environs (KERTESZ-SOMEGI 1999. 72.-) we know
rhat the Northern side of the mountain facing Lhe
Henye hill was covered in humid temperate periods by
forest composed dominantly of spruce (Picea). On the
north-cas1em side of the Kopasz Mt. which had speciaJ surface morphology,..a mixed forest refugium was
formed (KERTESZ-SUMEGI 1999.) by the Sagvaria.n chronological stage (DOBOSJ-V6ROS 1987.
Tabl. 8.).

Ar the feet of the Kopasz Mt.. faunal elements from

extreme arcals could be found together: rhis was
documented for example at rhe site Tokaj-Csurg6kUt
4a (20 OOO BP) where extremely steppean elements,
Al/actaga and Sicfrta species where found (KORDOS-RINGER 1991. 527.). Allactaga ">'as not
known before from the Upper Pleistocene. Si.milar
a lien element of the fauna immigration previously was
Alpine marmot known from north-eastern confines of
Debrecen (KRETZOl 1954b.), the Pilismar6t loess
sequences (JANOSSY- V6R6S 1979. Nr. 164.) as
weU as the lower, light brown layers of the Bervavolgy rock shelter in the BCikk Mts. (MOTIL 1938
a. 32-33. Fig. 10).
Tbc immediate surroundings of BodrogkereszturHenye yielded the following macro-mammal localities
from the WUrm period:
- BodrogkeresztUr - Mammwhus primigenius

(JANOSSY-VOROS 1979, Nr. 428.)

- BodrogkeresztUr-Dereszla, Site 3., -30 -40 cm
deep Mammuthus primigenius (HELLEBRANDT
1980, 88, with archaeological finds)
- Bodrogkeresztirr - Lebuj tavern, from loess

.Mammuthus primigenius (KOCI-I 1900.)

- Tokaj - Kopasz Mt. Cervus elaplzus fossilis
(KOCH 1900.)
- Tokaj - Patk6 quarry Ursus arctos (JANOS-

SY-VOROS 1979, Nr. 424.)

- Tokaj - 8 odrog interfluviaJ region, from loess
Mamm11th11s primigenius, Coelodonta cmtiquitafis
(VOROS 1980, JANOSSY- VOR6S 1979. Nr. 422.)
1

- Tarcal - Brickyards, from loess Mammlflhus

primigeni11s (VOROS 1980.)

- Tarcal - Citrombanya. from locss Mamm11tlws
pri'!.lig~nius, Ursus. spelaeus. f!.:an8ifer tarand11s
(VOROS 1980, JANOSSY- VOROS 1979, Nr.
421.); on the Palaeolithic settlement, Eq1111s sp.,
Rangifer rarandus (JA.NOSS Y 1975. 26).

6. Summary
l. From the 423 m 2 excavated surface on Lhe site
Bodrogkereszrur-Henye, 1803 pieces of animal remains were collected. Species found included 6 big
games: Eq1111s, Alces, Cerv11s. Bison, Mamm111/111s and
Leo as well as lep11s.
2. Equus was most frequent on the site (63%): the
specimens found belonged to, with the exception of
0ne small medium tooth and 5 postcranial bones to the
large medium size range. Withers height estimated
was 146 152 cm. The large medium fonn horse could
be identified with Central European E. remagensis, lhe
small medium form with lhe chronospecies E. "germanicus" The decreasl! o f s ize experienced in Late
Upper Plcistocene Equidae could be attributed, apart

from rhe effect of the deterioration of environmental

conditions (cold, arid climate) to the appearance of
migrating small Equus forms in Europe.
3. A/ces (occurrence frequency 29,2%). Tooth and
bone dimensions of the Bodrogkereszror A.Ices agree
well with Late Pleistocene and recent Eastern European Alces dimensions. ln the Upper Wilrm period,
11/ces is lypical submountainous species in the Eastern
Alpean - Circumcarpathian region in Europe. h was
al:so registered North of the Pyreneans and south of the
lower reach of the river Danube. Tbe areal of lhe
Central European A lees was shi fled following the
slow withdrawal of the ice sheet to the North. Nor1.hEast. l cs current boreal discribution was fom1ed as a

177

result of climatic as we.II as anrropogeoeous factors;

they may migra1e to the south to considerable

distances from their permanent areal.
4. Cervus (occurrence frequency 0,6%) belong to
"forrna maj01" (Cervus s. maral). ln Central Europe,
the red deer was driven back before the cold maximum of WUrm ill towards the south to the submountainous-submcditerranean region.
5. .Bison (oecurrence frequency 1,4%). The postcranial bone dimensions agree best with the short-horn
Eastern European Bison prisc11s med.i ator form.
6. Mammuthus (occurrence.frequency 5.5%): compared to the frequency offaunistical fmds, the ow1'1ber
of mammoth remains found on Upper Palaeolithic
sites as hunted is surprisingly low. On the Upper
Palaeolithic settlements younger than 18-20 ka BP.,
mammoth bones were probably transported not as a
result of hunting but much rather, collecting of bones.
7. The rop-predaror Leo and the Lepus were- rare
elemenis in che fauna.
8. At Bodrogkeresztllr, the frequency of animal remains was unevenly distributed among the settlement
units. The two main big game were horse and elk;
frequency of accessory game was low. Horse remains
dominated in Unit I., elk remains dominated in Unit

ill.
9. The anatomicaJ distribu1ion ofthe bones prove-cl
that none of the animals bad been taken to fue settle-

178

ment in whole. The tolal amount of useable meat calculated on the basis of individual number of hunted
animals (108) was 20462 kg, the amountof meat associated with the bones actually found on the settlement
was 6902 kg. The accumulation of the meaty vs. dry
limbs of the two main big game show complementary
tendencies: horse remains were dominared by meaty
limb booes, elk remains were dominated by dry limb
bones. Bison and red deer were mainly occurring as
dry limb bones. Remains of mammoth were probably
not for the purpose of eating. the remains of the cave
lion found were typical for terminal bones left in the
stripped skin.
10. The Bodrogkereszrur hunters used to go after
their prey along the southern margin of the TokajEperjes Mts., the gallery forests of the floodplain of
the rivers .Bodrog and Tisza. Herbivore large mammals used to live in these ecological niches. Elk, red
deer, bison and the large horse could be of foresral
area, the small medium horse from che steppean region. With the exception of the resident red deer, the
other animal species were migratory types. On the basis oftheJnortality of the hunted animals, the hunting
season could be hypothesed for .spring, late spring-

autumn an<! winter.

11. The f:allllaoftheBodrogkereszt(rr-Renye Upper
Palaeolithic settlement could be dated to the second
half of the Ist:aU6sk6 fa u nal pbase.

REFERENCES

ABSOLON- KLIMA 1977

K. Absolon-B. Klima: Pl'edmosti ein Mammutjagcrplatz in Mahren . Fontes

Arch. Moraviae 8. 1977. l>raha 215. Bartolomei et al. 19&8 G.

AAR1S-SORENSEN 1993
BARTOLOMEl et al. 1988

BARYSHNIKOV 1982

Aaris, K.-Sorensen,: The Fauna. In.: S. Hvass- B . Storgaard: Digging into the
Past. 25 Years of Archaeology in Denmark. Copenhagen 1993. 28-30.
G. Bartolomei, A. Brogli, L. Caltani, M. Cremaschi, M . Lanzinger, and P.
Le-onardi: Nuovc ricerche net dcposiro ;pleistocenico della Grona di Paina sui
Colli Berici (Vicenza). Arti Dell 1nstit1.11to Venero di Scienze, Lettere ed Arti

146 (1987- 88) 1988. I ll- 160.

G. F. Baryshnikov: Late Pleistocene Mammals of the Greater Caucasus. In.:

lntetnat. Union for Quaternary Research. Xl. Congress. Moscow 1982. Abstracts vol. n. 403.
BARTA 1965
Batra J. Slovensko v stadej a strednej dobe kamennej. Die Slowakei in der
tillere11 und mlttleren Stei:nzeit Bratislava 1965. 232.
BENDA 1929
Benda, L. : A magyar Aill>ld ostOrtenete. II. Acta Sabariensia 3. Kaposv.3.r
1929. 352,
BIR61984
Bir6, K .: Oskori lele.tek aMatra hegysegb61. Agria 20. Eger 1984. 5- 11.
1. A. Borziyak: Subsistence Pract ices -0f Late Paleolithic Groups along tbe
BORZTYAK 1993
Dnestr River and Its Tributaries. In. 0. Soffer-N. D. Praslov: From Kostenk.i
to Clovis. Upper Pal.eolithic-Paleo-Jndian Adaptations. New York-London.
1993. 67- 84.
Bokonyi S.: Szarvas-1. lelohely, egy keso t:1jkokori telepreszlet allatmaradvaBOKONYl 1987
nyainak a.rchaeozool6giai .v izsgalata. Archaeozoological srudy of the animal
remains found in the late neolithic settlemenL at Szarvas-s.ite No. l. MMM
K6z.J. 1986-81. Budapest 1987. 89-<103 .
L. Briedermann: Das Elchwild Alces aloes (L.). In.: H.Stubbe ed.: Buch der
BRIBDRRMANN 1982
Hage. l. H.aarwild. Berlin 19&1.. 99-109.
M. Brodar-F. Osole.: PaleoHlske i mezolitske regije i kulture u SJoveniji. In.:
BRODAR---OSOLE 1979
Pra.ist0rija Jugoslavenskih Zemalja. 1. Paleolitsko i Mezolitsko Doba. Sarajevo 1979 159- 194..
BURCHAK-ABRAMOVICH Burchak-N. I. Abramovich: Fauna pesery SagvardsiJe v Zapadnoj G.ruzii.
Tbilisi 1972. 99.
1972
V. J. Calkin: Zivotnovodstvo i ohota plemen vostochnoevropeiskoi lesostepi v
CALKIN 1966
rannem zeleznom veke. MIA. Moskva 1966 107.
L. Chalx-J. Desse: Contribution a la connaissance de l'elan (A Ices alces L.)
CHAIX-DESSE 1981
postglaciaire du Jura et du :Plateau suisse. Quai:tar 31-32. [981. Boon
139-190.
0. P, Chernish: Paleolitichna stojanka Molodove V. ANU RSR Klev 1961
CHERNISH 1961

L72.

Cl-IlRlCA 1987
CZIBR 1988

CZI ER-JURCS~K 1987

CSALOGOVITS 1936

V. Chirica: La genese et 1 evolution des cultures du paleoHthique superieur

clans la zone du Prut moyen d apres les recberches recentes. Bibliotbeca Archaeologicalassiensis n. [~i 198/ 25--40.
Z. Czier: Asupra unui de A Ices alces L. de la Ocna $ugatag, Judetul Maramures. Crisia LS Oradea 1'988 697-700.
Z. CzieE-l1. JurcsAk: Un craniu (le elan (Aloes alces L) descoperit in Avenu.1
lui Avnlmut din Cheile Somesulul Cald. Crisia 17. Oradea l 987 61 1-625.
Csalogovits J.: Oskori leletek Duna:tl>Idvaron. Tolna Varmegye Multjab61 2.

1936. 7-14.
A. 1. David: Teriofauna pleistocena Moldavii. Kishinev 1982 . 186.
DAVID 1980
DELPECH-GUADELLl 1992 F. Delpech-J-L Guadelli Les grandt;S Mam.rniferes gravettiens et Aurignaciens de la gror.te de Temnata. Tn.: K. J. Kozlowski et al.: Temnata Cave. Excavations in Karll.l.kovo Karsl Area Bulgaria Vol. I, Part l. Krakow 1992.
141-216.

179

R. Desbrosse-F. Prat: C Elah magdalenien de Poerre-Chatel (Ain). Quartar

251974Bonn 144-157.
DRlESCH 1976
A. Driesc]i von den: A guide to the measurement of animal bones from arcbaeol'ogical sites. Peabody Mus. 'Bull. I. Cambridge 1976.
DOBOSI-VOROS 1979
V. Dobosi:--1. Voros: Data to an evaluation of the finds assamblage of the Palaeolithic Paint Mine at Lovas. FolAroh. 30. 1979 7-26.
DUERST 1926
J. U. Duerst: Vt:rgleicheade Umersuchungsmethoden am Skelert bei saugem.
ln.: E. Abderhalden: Handbuch der biologischen Arbeitsmethoden. Abt. VlL
Heft 2. 1926. l 25-530.
BMPEL-ROSKOSZ 1963
W. Bmpel-1'. Roskosz; Das Skelett der Gliedrnassen des Wiesents, Bison
bonasus (Linnaeus, J 758). Kosciec koncz.yn zubra, Bison bonasus (Linnaeus,
1785). Acta Theriologica 7, 13. 1963. Bialowieza. 259-320.
BHJK 1916
Ehik. Gy.: A Hennan Ott.6-barlang asa.tasainak faunisztikai eredmenyei. Barlangkut. 4. Budapest 1916, 24-29.
FISCHER 1993
A. Fischer: Tbe Late PalaeoliU1ic. Tn.: S. Hvass-B . Storgaard : Digging into
the.Past. 25 Y.ear.s of Archaeology inDellillark.. Copenhagen 1993. 51 - 57.
FODOR-YOR6S 1990
Fodor, L.-V~ros, L: Pleisztocen 6sem!J)smaradvanyok Visont.ar61. Diluvialit
Ursiiugetier-Reste aus Visonta. Agtia 24. "Eger 1990. 5-20.
FORSTEN 1982
A. Forsten: Equidae. In.: J. K. Koztowski ed.: Excavation in the Bacho Kiro
Cave (Bulgaria) Ffnal Reort. Warszawa 1982 56-64.
FORSTEN 1988.
A. Forsten: The small cabal laid hor.:se of the upper Pleistocene and Holocene.
P. J. Anirn. BJeed. Genet 105. 1988. Hamburg-Berlin. 161- 176.
FORSTEN 1990
A. Forsten: Dental size trend. in an equid sample from the Sandalja ll cave of
northwestern Yugoslavia. Palaont. Z. 64. 1-2. Sturtgard 1990 153-J 60.
FORSTEN 1991
A. Forsten: Size decrease in Pleislocene-Hohocene true or caballoid horses of
.Europe. Mammalia 55,3. 1991. 407-4l9.
FRANK et al. 1995
Cb. Frank~ . K. Kunst--J. Mlikovsky-D. Nage.1-G. Rabeder - K. Rauscher
G. Reiner: Lisle der fossiJlen Faunen der GamssulzenMhle im Toten Gebirge
(00). Mitt. Komm. Quartll.rforsch. 9. Wien 1995 51-52.
F:RAYER 198 l
D. W. Frayer: Body size, weapon use. and natural selection in the Europe-an
Upper Paleolithic and Mesoli.Lhic. Am, Antluopologist 83, I. 1981 57-73.
GAMBLE 1986
C. Gamble: The palaeolithic senlement of Europa. Cambridge 1986 471 .
Grui~ l.: A javorszarvas legregibb csontmaradvanya Magyarorszagon. P6t!uz.
GAAL 1933a
A Term. Tuel Kozl. 65. Kotetehez 1933. 1.89. Faz. 19-22.
GAAL 1935
Gaal~ I.: A bfmhidai Szelim-barlang asat!sa. P6tf'Uzet a Term. l1ttd. Kozl. 67.
Kotete'hez l935. FUz. 49-63.
Ga!J, 1.: Ujabb ember- es eml6scsontleletek Erdely mouster1jeb01. Neuere
GAAL 1943a
Menschen- und Saugetietknochen aus dem MousteITien SiebenbOrgens. Kozl.
az Erdelyi Nemieti Muz. Erem- es R6gis6gtarab613,1. 1943. Kolozsvar 1-46.
M. G.Abori: Palaolitbische Schnecken-Depots von Szob. Acta AJ:chHung 21 .
GABO.Rl 1969
1969 3-11.
Gaborine-Csank, V.: Az 6sember Magyarorszagon. Budapest 1980 265.
GA.BORJNE-CSANK 1980
Gabori-Csank V.: A felso paleolitikum nyema .Budapesten. Spuren des oberen
GABORI-CSANK 1984
Pal!:iollthikwn in Budapest. Bud.Reg, 2S. 7-14.
Gabori:-V. Csank: Spuren des Jungpalaolithikums in Budapest. ActaArchRung
GABORl-CSANK 1986
38 .1986 3-12.
GINTER-KOZLOWSKI 1982 B. Gi.nter-J. K. Koztows_ki; Conclusions. Tn.: J. K. Kozlowski ed.: Excava-"
t:ion in the Badlo Kiro Cave (Bulgaria) Final Report. Warszawa 1982

DESBROSSE-PRAT 1974

169-172.
GINTER-KOZLOWSKI 1992 B. Gin1er-J. K. Kozlowski: The archaeological sequence. 1.n.: J. K.
K-0ztowsk.i et al. ed.: li'emnata Cav~ Excavations in Karlukovo Karst Area
Bulgaria Vol. L. Part 1. Ktak6w 1992 289-293.
V. l. Gromov: Paleontolegiches.koe i arl\hcologicheskoe obosnovanie stTaLiGROMOV 1948
grafii kontinentalriib otJozeoij chetverticlmogo perioda ha territorii SSSR.
AN.SSSR. Tr.Tnst Geol. N.64., Geol. Sef. 17. Moskva 1948 521.

180

GROMOVA 1960

GUENTER 1977

V. Gromova: OpredeliteJ mlekopitajushih SSSR po kostjam skeleta. V. 2.

Opredelitel po kropnym kosljarn zapljushy. AN. SSSR. Tr. J<om. Izucb.
Chetven. Perioda 14. Moskva l960 116.
Guencer, E. W.: Die Backenzfilme der Elephanten von Taubach bei Weimar.

HALAVATS 1899

Quartarpalaeontologie.2. 265-304.
Halavats, Gy.: A jobbagy:i-i (N6grad m.) Marnmuth-leletet. FoldL Kazi. 29.

HELLEBRANDT t 980
l:IEPTNER et al. 1966

1-!ESCHELER-KDHN 1949
HILLEBRAND 1919
IDR 1990
lREGREN 1983

JARMAN 1972

JANOSSY1955
JANOSSY 196.4

JANOSSY 1975
JANOSSY 1976
JANOSSY 1979
JANOSSY-KORDOS 1977

JANOSSY-KROLOPP 1994

JANOSSY-VOROS 1979

JURlC-BASLER 1979

KADlC 1916
KADLC 1934

KADTC 1939
KADTC-MOTIT.. 1938

1899 39-41.
Hellebrandr, M.: Bodrogkeresztfu (szerecsi J.). ln,: HeUebrandt, M.-Siman,
K. Regeszeti kutatasok BAZ megyeben 1979. HOME 19. Miskolc 1980 88.
V. G. Heptner-N. P. N aumov:
Stiogetiere der Sowjetunion. Band 1.
Paarhufer und Unpaarhufer. Jena l 966.
K. Hescheler-E. Kuhn; Die TierwelL Jn. : Tschumi, 0. ed.: Urgcschichte der
Schweiz. Bd. l . Frauenfold 1'949 12 1-368.
tlillerband, J.: Az 1917. evben vi!gzett asatasaim eredmenyei. BarlangkuL. 7.
Budapest 1919 6--13 .
Hir, J.: Retegazooosit6
a Pesko-barlangban. Layer identifying excavation in the Pesk8 Cave. Karszt es BarJang 1990. 2. Budapesl 97- 100.
E.J'regren: Cataloge of Bone Remains in in Prehistoric Sites in Northern Sweden . In.: J. Ekman- E. Tregren: Early Norrland 8 . Archaeo-Zoologiea! Investigations in Northern Sweden. 1.983. Scock.holm. 43-95.
M. R. Jarman: European deer economies and the advem of lhe Neolithic. In.:
Higgs, E.S. ed.: Papers in Economic Prehistory T. Studies by Members and
Associates of the British Academy Major Research Project in Early History of
Agriculture. Cambridge 1972125-147.
D. Janossy: Die Vogel-und Saugetierreste der splirpleistozlinen Schiccen d er
Hohle von 1staJl6sk6. ActaArcbH:ung 5, 3-4. J955 L49-J 81.
D. Janossy: Letztinter-glaz.iale Vertebratenfauna aus der Kalm.an-Lambrecht
Hlihle (Bilkk-Gebirge, Nordost Uogam). 0. Actn ZooLJluog. l0. 1964
139-197.
Janossy. D.: Tarcal-Citrombanya Oskokorj lel6hely gerinces faunaja. Fol.Arch.
25. 1975 26.
D. Janossy: Die Felsniscbe Tark5. Besc:hreibung der Fund.stelle und der Vertebratenfauna. Karszt- es Barlangkut 8. 1976 ~106.
Janossy, D. A magyarorszagi pleiisztocen tagolasa gcrinces faunak alapjan.
Budapest 1979. 207.
Janossy, D.-Kordos, L.: Az Osztramos gerinces lelohelyeinek faunisztikai es
karsztmorfo.16giai Attekintese. The futmistical and karst-morphologicaJ review
of paleontological localities for ''e.rtcbrates at Oszt:ram.os /Northern Hungary/ .
Fragm. Min. et Pal.8. 1977 39-67
Janossy, 0 .- Krolopp, E.: Als<hpleisztocen Mollusca- es gerinces fauna a
gyorujfalui kavicsbaoyab6l. Lower Pleistocene mollusc and vertebrate fauna
from the gravel pit of Gyoruj:fitlu INW Hungary/. F6ldl.Kozl. 124, 4. 1994
403-440.
D . Jaaossy-I. Vor5s: Grossauger - S.t reufunde aus dem Pleistoti.ln Ungars.
Frngm. Min. etPaJ. 9. 197.9 21-60.
M. Juri~-D. Basler: Prirodni o kviri i rtid na istrazivanju paJeolitskog i mezolitskog doba u Cmoj Gori. In.: Ppraistorija Jugostavenskih Zemalja. I. Paleo litsko i MezoJitsko Doba. Sarajevo 1979 379-382.
Kattie, 0.: A Herman Ott6-barlang H,;iinor k6zseg hataraban. BarlangkuL. 4.
Budapest I 916 6--17.
Kadic, 0.: A jegkor cmbere Magyarorszagon. Der Mensch zur Eiszeil in Ungarn. MKFT. Evk. 30.1. 1934. 148.
Kadic, O.: A budavari barlangpincek t'l51dtani viszonyai. A Szt l.stvan
Akad.Menny.-Term.Tud. Oszt Ert. 3, 4. 1939. L-20.
Kadic, 0.-Mottl, M.: Felsotiirkany videkenek barlangjai. Barlangkut. 16. I.
1938 g,....,39, Die Hohlen der Umgebung von Felsotarkany.

me

asatas

181

KERTESZ-SUMEGI l<J99

Kertesz, K-Slimegi, r.: Az Eszalci-kozephegyseg negyedidl:iszak veg1

os:Wrtenete. Emder
kOruye.z.et kapcs0laca 30,00 es 5.000 BP evek kozt>tt a
szubkarpati regi6ban. N6grcid MM Hvk. 23. 1999. 66- 93.
Kardos. L.-Ringer, A.: A magyarorszAgj felso-pleiszlocen Arvicolldae-sztrarignifiajanak klimato - 6s archeosmatigraiiai korelaci6ja. MAFI. Evi Jet.
19189. Ev.roJ (1991) Budapest 523-534.
Koch, A.: A magyar korona orsz,agai kovUlt gerinczesallat rnaradvanyai
rendszeres atnezete. Magy. Orv. Es Term . Vi.zsg. 30. Vandorgy. Munk.
Sz.abadka.1900. Budapest 526-560.
W. v. Koenigswald: Exoten .in der Giossauger-Fauna des letzten Cnterglaziais
von Mitteleuropa. Eiszeitalter u. Gegenwa:rt 41 Hannover 199 l 70-84.
K. Kowalski: Katalog ssak6w plejsmcenu Polski. A cataJoque of the pleistocene mammals of Poland. PAN. lnst..ZOo.Oddzial w Kiakowie. 1959.
Warszawa-Wroclaw. 267..
M. !<.retzoi: Die Wirbeltierfauna aus dem alteren Pleistozan van Solymar
(Manuscript). 1946.
M. Kretzoi: On Bison bonasus hungarorum n. ssp. Ann.His t.-nat. Mus. Nat.
Hung. 39-. 5- 6. Budapest 1946 105- 107.
Kretzoi, M.: A Bodrogkeresrnir Henye hegyi kefoti gravetti telep faunaja. ln.:
Vertes, L.: Bodrogkeres:ztur Heny.e hegy kefeti gravetti telep asatasa. Jelentes.
MNMAt. V. 94. 1965.
M. K.relzoi: Marmota-maradvanyok - Debrecenbol Mannot - remains from
Debrecen. Foldt.KlizL 84. 1-2. 1954 75-77.
M. Kretzoi: Etude Paleoncologique. ln.: Gabori-V. Cs<ink: La station du
paleolithique moyen d' Erd-Hongrie. Mon. Hist Budapestinensia 3. 1968. Budapest 59-lOJ.:
M. Kretzoi: A magyarorszagi qnarter es pliocen szarazfi>ldi biosztratigraiiajanak vazlata. Sketch of the Late Cenozoic (Pliocene and Quaternary) terrestrial stratigraphy of Hungary. Foldr. Kozl. 17 (93) 3 1969 179-204.
M. K.JTctzoi: Ergaozende Angaben zar Kenntnis der jungpal.aolothischen ReststeJle Bodrogkeresztur-Henye (NO-Ungam). Manuscript 1969.
M. Kretzoj-P. Manon-M. Pecsi-R Schweitzel'-I. Vikos~ Pliocene-pleistocene piedmont correlative sediments in Hungary (Based on lithological,
geomorphological, paleontological and paleomagnetic analyses of the exposures in the open-cast mine at Gytsngy0svisonta). Quaternary Studies in Hungary 1982. Budapest 43-73.
Kubacska, A.: A solymari sziklailreg pleisztocen csontlelele. Barlangvilag I
1926 1-4. ( 1927) J 7-26.
H. Kubiak: Slcmie kopalne Polski poludniowej . The Fossil 'Elephancs of South
Poland. Fol. Quatemaria l 9.J965.Krak6w. 1-43:
H. Kubiak-A. Nadachowski: Artiodactyla. In.: J., K. Kozlowski ed.: Excavation in the Bacho .Kiro Cave /B tilgariaJ FinaJ Report. Warsza.wa 19.82 61-66.
A, A. Kiotova-N, G . Belan: Amvresievka.. A Unique Upper Paleollilii.c Site
in Eastern Europe. In.: 0. Soff.er-N. D . Praslov ed.. From .K:ostenki to C lovis.
Upper PaleoJithic-PaJeo-Indian Adaptations. New York-London 1993
125-142.
E. Lucius: Das Problem der Chronologie jungpalaolithischer Stalionen im
Bereiche der europllischen UdSSR. Mitt.Prahist.Komm.O.st.Akad.Wiss.
13-14. Wien 1969-1970 BO.
N. Macarovici: Contribution a la connaissance de Alces palmatus Ham . Smith
dans le Quaterna.ire de la Roumanie. Eisuitalter u. Gegenwart l 2 OhringenJWilrtt. 1962 66-72.
M. Malez: K varra.ma fauna pecine Vetemice u Medvednicl. Die Quart.are
Fauna der Hohle Vetemica (Medv:eclnica- Kroatien). Palaeoot. Jug. 5. Zagreb.
1963 193.

es

KORDOS-R:INGER 199 J

KOCH 1900

KOENJGSWALD 1991
KOWALSKI 1959

KRETZOI I 946a
KRETZOI l 946b
KRETZOI 1964a

KRETZOI 1954b
KRETZOI l 968a

KRETZOI I 969a

KRETZOI l 969b
K:RETZOI et al. 1982

KtJBACSKA l 927
KUBIAK 1965
KUBJAK-NADACHOWSK.I

1982
KROTOVA- BBLAN 1993

LUCfUS 1910

MACAROVICI 1962

MALEZ 1963

182

MALEZ 1972
MALEZ 1979

M. Malez: Rasprostranjenost hladnodobnih zivotinja u gornjem pleistocenu

jugoistocne Evrope. Uber die Verbreitung kaltzei1licher Tiere im jOngeren
Pleistozan SOdost-Europas. Ljetopis JAZU 76 1972. Zagreb 133-J 80.
M. Malez: Paleolitsko i mezoUtsko doba u Hrvatskoj. In.: Praistorija Jugoslavenskih Zemalja. L Paleolitsko i Mezolitsko Doha. Sarajevo 1979
197-276.

MALEZ 1986

M. Malez: Die quartll.ren Vertebraten-Faunen in der SFR Jugoslawicn.

Quarrarpa1~ontolog.ie 6 Berlin 1986 I 0 l.-117.
MASSETI et al 1995
M. Masseti- P. Mazza-M. Rustioni-B.Sala: Large-sized italian Ungulates
at the Late Pleistocene-Holocene transition: An Overvil!w. In.: Archcozoologia l Convegoo NazionaJe. Padusa Quademi I . 1995 Ro' igo. 89-96.
MINIERI et aJ 1995
M. R. Minieri-C. Petronio--R. Sardella-M. Scarano: Le faune a mammi
feri del Pleistocene superiore dell ltalia Peninsulare. lo.: Archeozoolog1a. 1
Convegno Naz:ionale. Padua Quaderni I . 1995 Rovigo 75-78.
MIRACLE- BRAJKOVIC 1992 P. Miracle-D. Brajkovi~: Revision of the Ungulate Fauna and Upper PlelsLO
ceme Stratigraphy of the Veternfoa Cave /Zagreb, Croatia/. Geol. Croatica 45 ,
Zageb 1992 1-14.
MOTIL 1938a
Mottl, M: A Bervavl:\lgyi sziklaOrcg. 6slenytani eredmenyek. Tn.: Kadic, O .
Mottl, M.: Fels6tarkany videkcnek barlangjai. Barlangkutatas 16. 1. 1938
MOTIL 1941
MOTIL 1942b

MUSIL 1958

MUSIL 1962
MUSIL 1968a
MUSIL 1968b
MUSIL 1969

MUSIL 1974
MUSIL 1990
NOBIS 1971

NOBIS 1981

PAAVER 1965

eALEOLIT SSSR 1984

PAUNESCU 1984

25-34.
Mottl, M .: Az interglacialisok es lnterstadifilisok a magyarorszagi emlosfauoa
tilkr.Cben. Ftlggelck a MKFI 1941 .Evi jelentesenek. 194 I 5-42.
M. Mont: Adatok a hazai 6- es Lijpleisztocen foly6terraszok cmlosfaunajAhoz
Beitrage zur Sliugetierfauna der ungarischen Alt- und Jungpleistoz.fulen Flussterrasseo. MKFI. Evk. 36. 2. 1942 1-70.
R. Musil: Fauna moravskych magdaJ~nskych stanic. Anthropozikum 7 (1958)

7-26.
R Musil: Die Hl\hle "Sv&Juv stul'' - dn typise.her .HOhlenhyanenhorst AnLropos, 13 1962. Bmo 97-260.
R. Musil: Die Equiden aus dem Travertin von Ehringsdorf Abh. Zentr. Geol.
Inst Palaont. Abh. 23. Berlin 1968 265-335 .
R. Musil : Die Mammutmolaren von Predmost (CSSR). Palaont Abh., A. 3 .1 .
1968. Berlin 1-192.
R. Musil: Die Encwicklung dcr liergeselJschaft im Uufe der Sedimentation in
der Kulna-Ht>hle. ln.: K. Valoch et al.: Die Erforschung der Kulna-Hohle bei
Sloup im Mahrischen Karst (Tschechoslowakei). Quartllr 20 1968 Bonn 8-20.
R. Musil: TiergeseUscbaft der Kniegrotte. In .. R. Feustel ed.: Die Ki1iegrotte.
Einc Magdalenien-Station in Thilringen. Veroffeotjjchungen des Mus. fUr Urund Frtlhgeschichtc Thilringens S. 1974 Weimar 30-95.
R. Musil: Pferdefunde /Equus raubachenis und Equus scythicus I aus der
Kfllna-Hl:lhle in MIDu'cn. Weimarer Monogr. zur Ur- u. Frilhgeschicb1e 26.
1990 Weimar 84.
G. Nobis: Zum Wildpfred zum t-lauspfred. Studien zur Phylogenie pleistotilner Equiden Eurasiens und das Domestikationsproblem unsercr Hauspferd. Fundameota B.6. 1971 K6ln-Wieo 96.
G. Nobis: Equss mosbachensis v. Reichenau aus Ablagerungen des cromerzeillicben Mains von Randersackcr bci Wiirzburg. Quartlkparaontologie 4.
1981 Berlin 93-104.
K. Paaver.: Die Entscehung der Saugetierfauna und die Variabilirlit der
Siiugetier der Os1baltikums im Holo'l1ln. AN. EEER. 1965 Tartu. 494 (in Russian, with Gem1an summary).
Paleolit SSSR. Ed.. P.1. Boriskovskij Arkheologija SSSR 1984 Moskva.
A. Paunescu: Cronologia paleoliticuJui ~1 mezoliricului din Romania in contexlul paleoliticului Central-Est si S11d European. SCIVA 35. 3. 1984 Bucur~ti

235-265.

I83

PHLEPS l907

PfDOPLICJ-IKO l956

0. Phleps: Ueber das Skel,e tt eines we~blichenBison priscus Boj. sowie andere
Bison und Bosreste aus dem DiJuvium Siebenbti:rgens. Verhandl. Mitt. Siebenb. Vereins.f. Naturw. 56. Hermannstadt ( 190.6) 1907 1-44.
I. G. Pidoplichko: .Materiali do vivchenya minulih fauna URSR 2. AN. URSR.

POHAR 1997

1956 Kiev 234.

V. Pohar: Poznoglacialna :favna iz Lukenjske jame. La faune tardiglaciaire de
lagrotte de Lukenjskajama Geologija 26, '1983. Ljubljana 7 1- 107.
V. Pohar: Kvartarni sesalci iz Babje jarne pri Dobu. Les mammiferes quaternaires de la grotte de Babja jama (Dob). Razprave 4. Razreda SAZU 26.
Ljubljana 1985 97-130.
V. Pohar: Late Glacial mammalmacrnfauna in Sloeiiia Quartli.r 47/48 1997.

PRAT 1968

F. Psat: Recherches sur les equldes pleistocenes en France. Th6se Sci. nat.

POHAR 1983
POHAR 1985

14~158.

RAK.OVEC 1953

RAKOVEC 1956

R.Al{OVEC 1965

. RAKOVEC 1975

REICHENAU l9t5

RIEDEL 1980
RUFFO 1995

sARKO!lY 19 I6
SKORKOWSK1 1938

SOFFER l985
STURDY-WEBLEY 1988

SZALAY 1916
TELEOIN 1985
THENTUS 1959

TOP AL-VOROS 1984

TRINGHAM 1969

184

Bordeaux. 226. 1968. Bordeaux 662.

I. Rakovec: Bizon i2 most1scarske dobe na ..dubljanskem barju. On.Bison of
Lhe Lacusstrine Age from me Ljubljansko barje in Slovenia (NW Yugoslavia).
Ar. h. vestnik. 3. 1953. Ljubljana 257-287.
1. Rakovec: 0 ostankih evropskega Losa v Jugoslaviji. The remnants of the Elk
(Alces alces L.) in Yugoslavia. Geol. Anali Balkan. poluostrVa 24 Beograd
1956 l-14.
I. Rakovec: Pleistocenska sesalska favna iz R isovace pri A.randjelovcu. Pleistooene mammalian fauna fft'om Risovaoa near Arandjelovac (Serbia). Razpr. 4.
razr. 8. Ljubljana l965 225--311.
l. Rakovec: Razvoj k.v.artame sesalske fa.vne Slovenije. Uber Quartiire
Saugetierfaune.n Sloweniens (NW Ji.tgeslawien). Arb. Vestnik24 (1973) 1975.
Ljubljana 225-270.
W. Reincbenau: Beitrage .zru nfilleren.Kenntis fossHer Pferde aus deutschem
Pleisrozan, i:nsbesondere Uber di.e Entwicklung und die Abkaustadien des Gebisses vom Hochterrassenpferd (Equus. mosbacbeosis v.R.). Abh. bess. Geol.
Landesaostalt 7. l l915.. Darms.tadl 155.
A. Riedel. l l cavallo delta grotta Tilde. The horse of the Tilde Cave. Atti
Mus.Clv. St. Nat., 32. (1) 2. 1980. Trieste 19-110.
M. Ruffo: Origine degli lnsiemi faunistici nei vari livelli deposizionali della
G~otta delle Capre ~ S . Felice CiI:ceo (LT)-1993. In.: Archeozool0gia. 1 ConvegnoNazionale. Padusa Quaderni I. l 995. Rovigo I 09-123.
Sarkozy, Gy.:Vadaszlap 1916.jul 25. 247-248.
E. Skorlcowski: Badan.ia nad systematyka konia. Studien Ober die System.atik
des _Pferdes. Prace Rolniczo-Lesne 32. 1938. Krakow 125.
0. Soffer: The Upper Paleolithic of the Central Russian Plain. Orlando 1985
539.
D. A. Sturdy- D . P. Webley: Palaeolithic geography: or where are the deer?
World Arch. 19.3. 1988 262-280.
Szalay, B.: Javorszarvasunk. A magyamrsz8gi Alces palmatus tOrtenete.
KUlonlenyomat a Vad3.szlap 1916. evi 37: evf.-b6l 1-J l.
D. Ya.Telegin: Pamjamiki epokhi mezoLita na territorii ~koj SSR. ANU
SSR. 1985 Kiev 183.
E. Thenius: l)ie jungpleistozfule Wirbeltierfauna von Willendorf i. d. Wachau,
N. O. Tn.: Felgenhauer-, T r. ed.: Willendorf in der Wachau. Mitt Pr1iliist
Komm., 8-9. l. 1956-1959. Wien 128-170.
Gy. Topal-]. Voros: Notes on history and recent records of elk (A lees alces
/L./) in Hungary. Vert. Rung. 22. 1984 83-94.
R. Tringham: Animal domestication in the Neolithic cultures of the
sooth-west part of European USSR. In.: P. J. Ucko-G. W. Dimbleby ed.: The
domestication and exploitation of plains and animals. London
1969 381-392.

VERTES 1954
VERTES 1965
VERTES 1966
VJIT 1952
VOROS 1974

VOROS 1975

VOROS 1980
VOROS 1981a
V6R6S 1981c
VOR6S 1982
VOROS 1983a

Vertes. L: Nehany uj 6skori leiohelyiinkrol. Fol.Arch. 6. 1954 9-21.

Vertes, L.: Az oskokor es az atmeneli kokor t!mlekei Magyarorszagon. A Magyar Regeszet Kezikonyve 1. Budapest 1965. 385.
L. Vertcs: The Upper Palaeolithic Site on Mt Henye at Bodrogkeresztur. Acta
Arch Hung 18 1966 3-14.
0. V. Vitt: Losadi pazyrykskih kurgano\. Sov.Arh. 16 1952 163-205.
Voros, I.: A Karpat-medem:e Elephantidae-Leletanyaganak artekintese. Egyetemi Szakdolgozat. Debreccn 1974. Kez irat
Voros, I.: A magyarorszagi subfossiJis szarvaspopulaci6k archt!ozool6giai
vizsgalata. Doktori Disszertaci6. Debrecen 1975. Kezirat.
Voros, I.: Magyarorszag fosszilis Elephantidii. L Eszak-Magyarorszag Ell.!phanudae lelete1. FoLHisL-natMus.Matr. 6. 1980 13-49.
I. Voros: Wild Eqwds from the Early Holocene in the Carpathian Basin.
Fol.Arch. 32 198 1 37-68.
Voros, I.: A Pilismar6t-Di6s fe!sopaleolit Lelep allatcsontleletei Comm Arch
Hung. 1981 22-24.
I. Voros: Faunal remains from the Gravettian Reindeer hunter"s campsitc al
Sagvar. Fol. Arch. 33. 1982 43-71.
I. Voros: Archaezoological investig.ations ofsubfossil red deer populations in
Hungary. 111.: M. ~ubasiowcz. ed.: Arcnaeozoology. r. Szczecin (l979) 1983
637-642.

V6ROS 1983b
V6ROS 1983c

V6R0S 1984
V6R6S 1985

VOROS 1986a
V6ROS 1986b
V6ROS 1987a

V6ROS l 987b
VOROS 1987c

VOROS 1988a
VOROS 1988b

VOROS 1989
VOROS 1990a

T. Voros: Lion remains from the Late Neolithic and Copper Age of Carpathian
Basin. Fol. Arch H. 1983 33-50.
I. Voros: Elephantiden-Reste aus dem Karpathenbeckeo. Fragm. Min. et Palaeont. 11. 1983 61-84.
I. Voros: Hunted mammals from the Aurignacian cave bear hunters' site in the
IstAJJ6sko Cave. fol.Arch 35. 1984 7- 3 1.
I. \'or6s: Alces brevirostris Kretzoi from the Ordoglyuk Cave at Solymar
(llungary) Fragm. Mio. et Palaeonc. 12. 1985 59--66.
Voros, I.: A szerencs-takta.fBldvari kesoncolil telep t.\llatcsontleletei. Nal. Borsod. I. 1986. Miskolc 98- 124.
I. V6ros: Animal remains ofU1e Pilisszim6 Rock-Sheller II. Ln.: V. T. Dobosi1. Voros: Chronological revision of the Pilisszant6 - Rock-Shelter II. Fol.
Arch. 37 1986 32-42.
l. Voros: A Tiszaluc-sarkadi rezkori celeptlJes allarcsontleletei (El5zctcs jelentes). Animal remains from the Copper Age Settlement at Tiszaluc-Sarkad
(Prnliminary Report). Fol.Arch. 38 1987 121-127.
I. Vor6s: Large mammalian faunal changes during the Late Upper Pleistocene
and Early Holocene Tunes in the Carpathian Basin. In.: M. Pccsi ed.: Pleiscocene environment in Hungary. l 987. Budapest 81-l 0 l.
I. Voros: The mammal fauna of the Pilisszanl6 1 Rock Shelter. In.: V. T.
Dobosi-1. V6r0s: The Pilisszant6 l. Rock-Shelter revision. Fol.Arch 38. 1987
31 - 50.
I. Voros: Middle Pleistocene mammalian fauna from the Solymar Cave.
Fol.Arch. 39. 1988 41-58.
I. Voros: Animal bone remains from the Nadap Upper Palaeolithic SettJemem.
Jn.: V. T. Dobosi-R Jungbert-A. Ringer-r. Voros: Palaeolithic Settlement in
Nadap. Fol.Arch. 39. 1988 32-39.
Vorfis, I.: Madaras-Teglaveto felsopleisztocen emlos maradvanyai Cumania
11 Kecskemer 1989. 29-43.
L Vfiros: Fds5 plcisztocen Equus a.chenheimensis Nobis 1971 koponya Pihsmm6t-Basaharcr61. Equus achenheimensis Nobis 1971 Schadel aus dem
oberen Pleistozan von Plllsmar6t-Basaharc. Duoai Regeszeti Kozl. 3. ( 1989)
1990 5--10.

185

VOROS 1990b

I. V0ros: Szabolcs isptin.~agi szekhely A.r_ad-kori allatcsontleletei. ArpMAge

animal bones from SzaboJes. JAM Evk. 27-29 ( 1984-19&6) 1990. Nyfregyhaza t65'-188.

VOROS J991a

Voi:os, 1.: MagyarorszAgi Elephantidae-k biosztratigrafiaja. 1991. Kezirat

VOROS 1991b

1. Voros: Large iruunmaJ remaiJls from the Upper Paleolithic site at Esztergom-Gyurgyalag. Acta. Arc.h. Hung. 43. 1991 261-264.
I. Voros: A versegi paleolit telep emlosfaunaja. Die Saugetierfauna der Paleo1ithsiedlung von Verseg. Stud.Com. 21.1991. Szentendre 85-96.
1. Vliros~ Bird and Mammal faunal revision of Kiskevely Cave. la.: V. T.
Dobos.i-I. Vor0s: Material and chmnological revision of the Kiskevely Cave.
FoLAtch. 43. 1994 21 - 36.
Voros, t. : A balaeai neolttikus telep a:Uatcsont-maradvanyai. Com.Arch Hung.
19% 45-51.
Vorlls, I.: A Duna-kanyar bal partjflnak pleisztocen nagyemlos Jeletei. Kezirat.
P. Wyrost: Dawna fauna Polski w swi-etle badan kostnych materia16w
archeologicznych rozmieszczenie w czasie i przestrzeni The fauna of ancient
Poland in the Light of archaeological bone materials research. Rocz.ar. Pozn.
259 1994 Poznan 75- 176.
J. V. Zelizko: Dilmrialni faunn od VoJyue. Zprava o vyzkumu za rok
1918-1919. Fauna pleistocene de Volyne. Sboro. Stat. geol. u. Ceskoslovensko 1923. n.2. Praha 237-275 .

VOROS 199Jc

VOROS 1994
VOROS 1996

VOROS 1998
WYROST 1994

ZELrzKO 1923

APPENDIX
M acro-mammal remains on Hungarian Upper Pleistocene sites

Estvan VOROS
Hungarian vertebrate palae0nto.logical. research
J:>Ublished 100 years ago the large-scaJe catalogue site
List known as the ''Koch-catalogue". Its exact reference is tbe fCi> llowing: Koch, A.: A Magyar korona
orszagai kliVUlt gerinczes aHal maradvanyainak
rendszeres atne.zete" (A systematic lisc of the vertebrate fossil remains of the land under the auspices of
the Hungar.ian Crown} (Magy. Orv. es Term.-Vmg.
Szabadkai Vandorgyules Munk., 30. Bp. 1900. 526560). [n this work- without references to technical literature due to lack of space - 73 species from the
Quaternary period were enumerated. Apart from localities with cave bear, hyena and lion, sites wltli 20
herbivorous macro-mammals were listed.
The publication of the next, dominantJy macromammal lists can be attributed to the renewal of
Hungarian spelaeol0gica1 research from the tum of
the twenties, thirties of the 20th centucy, which
became mcreasingly more intensive. At first, only th~
"species" of the individual cave layers (KADJC
1934.), later "faunas" of layers (MOTrL 1938b,
1940.), later Hfaunas" of the established Palaeoli.thic
186

cuJtutes were published (Solutrean: MOTTL 1938b.,

Aurignacian; MOTIL 1942a.). Later on, the faunistical result& of the open-air sites along the Danubebend (MOTTL 1942b.) and that of the BUkk caves became know (MOl1TL 1944, 1945.), though with some
delay.
The first virtually complete list of the new Mammalian faunas extending over the whole Pleistocenc,
following chronological order - but unfortunately
without references - was published by M.Mottl, under
the title "Az interglacialisok es interstadia.Jisok a
magyaro~zagi emlosfauna ttikreben1' [lnterstadials
and interglacials as reflected by the Hungarian M~
malian fattna] (Foldt.Jnt. Evk. 352., 1941 75-112).
In Lhe m<mograph uA Magyar Regeszet Keziki.S1~ve J." [Handbook of Hungarian Archaeology I]
(VERTES 1965.) L. Vertes published fauna lists in
chapters 26. "13ar:langi lelohelyeink... " [Cave sites in
Hungary] and 27. "Ktllszf.n.i (losz es terasz) telepeink
termeszettudomanyos es regeszetti adat.ai "[ArcbaeoJogical and scientific data on Hungarian open-air
(loess and terrace) sites]. Fonner fauna lists were

quoted, sometimes inaccurately. cornpiered at some

placc::s by dala from D. Jlinossy and M. K retz-0i - again
without references.
The publication of the comprehensive monograph

by D. Janossy (JANOSSY 1979.) became necessary

due ro the correlation efforts based on the faunistical,
bio- and sediment stratigraphical researcb. of the 60ies and lhc 70-ies. Its title is: "A magyarorszagi
pleisztocen tagolnsa a gerinces fauna!< alapjan" fThe
division of the Hungarian Pleistocene on the basis of
lhe vertebrate faunas]. Late Wilrm faunas and fauna
lists of Palaeolithic siles were omitted from the volume due to conceptual considerations (referring to Lbe
monograph by Venes (1965.), (JANOSSY 1979. 8)
The fauna lists of further 533 Hungarian localities
where less than I 0 species were found, the so-called
sporadic sites were also omined from the volume
(JA.NOSSY-V0ROS 1979.).
There 1s still no comprehensive catalogue on the
Upper Pleistocene fauna or its localities. This can be
the reason for some misinterpretations. like the mistaken faunistical. biostratigraphical assignment of
some fauna! assembla.~es .as turned out later: see, e.g.,
Lovas (DOBOSJ-VOROS 1979.) or Jankovich cave
(GABORI-CSANK 1993.).
Dealing with macro-mammals, lhe author cannot
aim at completing this gap fully. At the same time it
became indispensable Lo compile al leai t a basic site
catalogue of macro-mammals. These species were, at
the same time, the most important big games for prehistoric man. Species names published in technical
literature - knowing the different reliability and
somce value - had to be taken as published. These
data cannot be altered Lill the actual revision of the
animal remains is accomplis.hed. Where revision has
been already completed, results of these were also
used. Species of Bovidae (Bos and/or Bi.son) were
treated together, which is an acceptable compromise
within the aims of the present work. In the future, the
existing evidence of Bovid. Ccrvid species have to be
examined again. It occurs quite often that assignacion
of species ro Cerv1ts elaplrus, C. "maral" and Megaloceros is not correct. Taking this into consideration as
weU, the author believes that further corrections may
modify the image on Upper Pleistocene herbivorous
fauna in the Carpathian Basin, but the main trends will
not change essentially.
Among the 15 macro-mammal species analysed, 2
arc top predacors (carnivores) and 13 herbivores:

Leo spelaeus (Goldfuss)

Crocotta spelaea (Goldfuss)
SusscrofaL
Capreolus capreo!tts (L.)
Cervus elaphus. l.
Cervus "mara/" Ogilby

Alces alces (L}

Dama dama(l)
Megaloceros gi,ganteus (BlumenbC1ch)
Rangifer tara1u:111s(L)
Bovidae (Bos/Bison)
Asinus hydrr1nti1111s (Regalia)

Equusssp
Coelodonta a11t1q11itans (B/umenbach
Mamm111/111s prim1ge11i11s (Blumenbadi).

The scudied Upper Pleistocene, Wtirm faunas and

localities were assigned to the following fauna!

phases:

SOITO
VARBO
SUBALYUK

- uncertain, belonging to s.L R-W:

- prae-Wfrrm;
- Lower WUnn;

SZELETA

- Middle WUrm;

ISTALLOSKO
PILISSZANTO

- Upper Wilrm as well as

-before the worm rn cold max
-following the m maximum.

BAJOT,

Among lhe fauna! phases of the Upper Pleistocene

in Hungary, Tokod phase has to be annulled because
its type faunas are older species assemblages
(JANOSSY-VOROS 1985. 1987. GASPARIK 1993,
1998.). The chronological assignment of the open-air

sices was performed on the basis of the so-called type

species.
Sites

I. SOTI6 FAUl'AL PHASE

I. Csakva r cave, Vertes Mts.
Side branch - upper part of yellow layer (mixed):
Dama. Also occurring: Crocotta, Sus, Capreolus,
Cervus, Bison sp. Equus, Coelodonia (KRETZOI
1954 a. 43-44).

2. Simo - Diosvolgy quar11'

In the cleft: Leo, Crocorto. Sus, Capreollis, Cen111s,
Bovidae, Eq11us (KORMOS L925. 165). Later on,
Janossy identified the site with clefts Si.\ttlS I., 2 and 4
(JANOSSY 1979. 118. Fig. 29).
Siltto 6/3. reddis11 brown layer: Dama (JANOSSY
1979. 118)
StittO 8. Crocotta (JANOSSY 1979. 118)
S!itto 9. Ce.rvus, Bos seu Bison, Equus (JANOSSY
1979. 120)

3. Subalyuk cave, Bll.kk Mts.

Lower layer complex:
l. high red layer: Leo. Crocofto, Cervus, Bos.
Equus, Coelodonta.
2. high red layer with breccia: Leo. Cervus, Equu.s,
Coelodonta.

187

3. yellowish red layer: Crocotta, Cervus. Bos,

Equus, Coelodonta.
5. dark brown layer: Crocotta, Sus, Cervus, Bas,
Equus. (MOTIL 1938c. 215, 221-225).

11 . Solymar-Ordoglyuk cave, Budai Mts.

Denever (Bat) room A/5. yellowish brown layer:
(ffyscrix), Leo, Crocorta, Capreolus, Cervu~, Eq1111s,
Coelod01ua (VOROS 1988a. 44 Table 4).

U. VARBO FAUN AL PHASE

12. Siitto-Calcareous tuff,

(Hystrix), Sus, Cervus, Megaloceros.

/l./ Csakvar cave, Verte Mts.

Side branch - upper part of yellow Layer (mixed)
Leading fossil: (Hystrix), Megaloceros, Asi11us.
Other taxa: Crocotta, Sus, Cervus, Capreolus, Bison sp., Equ us, Coelodonta (KRETZOJ 1954a. 4344).

(KORMOS 1925. 162-163).

4. Di6sgy6r-Tupolca cave, Bilkk Mts., Castle bill

Brown I Dark brown layer, marked "b": (Hysrrix
p.c. by M. KRETZOI), Crocotta, Sus, Cervus, Megaloceros, Bos, Bi.son, Equus, Coelodoma, MammuJhus
(SAAD-GAAL 1935. 6~2).

Bovidae

13. Ta rkc'.S rockshelter, Bukk Mts.

Block Illa, lower brownish Layer marked "5":
(Hyscri:r), Leo,
Block VI. Terra rossa, layer "8": (Hyslrix),
Cervus, Rangifer, Bison (J ANOSSY 1976. TabI.
11/B-C)

fJl. SUBALYUK FAUNAL PHASE

14J Di6sgy6r-Tapolca cave, Biikk Mts., Castle Hill

5. Kiskevely cave. Pilis Mts.

Yellow layer nr. 5. " with hyena": (HystrLt), Crocotta (VOROS 1994. Fig. 8. Table L.).
6. Kiskobat pot-hole, BUkk Mts.
Reddish brown layer: (Hystrix), Cerms, Rangifer,
Bison, Equus, Coelodonta (JA.NOSSY 1979. 129).
7. Lambrecht cave, BUkk Mts.
Layer lV: dark grey: (HystrLY), Crocotca, Sus,
Capreolus, Cenus f. major. Alces. Megaloceros.
Rangifer. Bison, Asinus, Equus, Coelodonta, Mammuthus

l/4. brownish grey layer: Sus, Bos seuBison.

I/3. yellowish-reddish brown layer: Megaloceros.
Ra11gifer, Bison. Equus, CoeJodonta, Mammuthus
TT/5. light grey layer: Crocoita, Bos seu Bison. Asim1s. Equus. Mammuthus

ll/4. reddish grey Layer: Megalnceros

U/3. red.dish brown Layer: Leo, Crococta, Cervus,
Megaloceros, Rangifer, Bos, Bison, Coelodonta,
Mammuthus (HELLEBRANDT-KORDOS-TOTH
1976. 31 ). In the grey-dark grey layer "c" of the old
excavation there were no Crocoaa and Mammurhus
(SAA.n-GAA.L 1934. 18), it is erroneously published ia HELLEBRANDT et al. 1976. 31 ).

Layer complex V: from black to reddish brown :

(Hysrrix), Leo, Crocotta, Sus, Capreolus, Cervus f.

majo1; Ale~~ Megalooeros, Rangifer, Bison. A.sinus.
Eq1111s, Coe/odonta, Mamnmrhus
Layer V/a (=lower part of Layer V.): (Hystrix), Bison (JANOSSY 1963. 294, 298).

IS.I Kiskevely cave, Pilis Mts.

Layer 4. dark brown /reddish brown , Leo, Crocotla, Sus, Cervus, Alces, Megaloceros. Bos, Bison,
Asim1s, Equus Coelodonta, Mamm11rhus, (MOTTL
1938b. 40-4l, 1941. 13,)Bos=Bison, Megaceros =AJ-

ces (VOROS 1994, Tabl. l . Fig. 8).

8. Porlyok cave, Aggtelek Karst area
"Flat room" - lower lst reddish brown layer
(JANOSSY 1979. 121 = layer 4. In JANOSSY et al.
1973. Fig. 5.): Capreolus, Cervus, Equus (JANOSSY
et al. 1973. 34).
9. Poroslyuk, Bilkk Mts.

Red/brown -reddish brown layer: (Hystrix), Su.s,

Capreolus, Cervus, Bison, Equus (fANOSSY 1979.
128).
10. Porgolbegy (Szarazgerence) cave, Bakony Mts.
V. " terra rossa" layer: Cervus (VARROK 1955.

494).

188

In her earlier publicarions M.Mottl did not mention

Hysrrix from this layer (l 938b. 40-41.). Later this
species has occured in the faunaJ list (MOTIL 1941.
13) probably from the 5th layer.

n.1 Lambrecht cave, Btlkk Mts.

UL yellow layer, Capreolus, Alces, Megaloceros,

Equus (JANOSSY 1964. 145, 148).

14. Lengyel cave, Gerccse Mts.
Leo, Capreolus, Cervus. Megaloceros, cf Bison,
Asinus. Equus. Coelodonta, Mamm1dhus (JANOSSY
1979. 135).

/ LO.I P!frgll lhegy (Sz.irazgercnce) cave, Bakolly

CV: SZELETA FAUNAL PHASE

Mts.
lV. yellow layer: Megaloceros, Rangifer, Bison,
Mammurhus.

Ill. grey layer: Sus. Capreolus. Cervus, Cervus c[

maral, Rangifer.
rr. reddish brown layer: CapreoJus, Cervus sp.,

18. Arnocko cave, Bukk Mts.

2. light brown layer: Leo, Ce1vus, Megaloceros, B;son, Equus (KADIC- MOTTL 1938. 5 l ), Cervus
e/aphus f. major (MOTIL 1940. 1911 ).

Meguloceros, Rangifer, "OV1bos (=smaJI Bison, acc:ord.ing to I. VOROS), Asinus, Equus, Coelodon.ta
(VAR.ROK 1955. 493-494).

19. Balla cave, Biikk Mts.

Rear part of cave, greyish-greenish layer: Leo ?,
Su.tt, Rangifer,Bos sp. (HILLEBRAND 1912. 765)
(Lower) greeni.sh grey layer: Leo, Crocolla.

13.! S uhalyuk cave, Bukk Mt'>.

Upper layer complex:

Capreo/11.'>, Megaloceros, Rangifer, Bison, Eq11us

7. greenish grey layer: Crocolla

8. yellowish brown layer: Crocotta, Cervus sp.
"mural . Bos. Coelodunta, Mammuchus
9. greenish yellow layer: Leo. CrocotTa, Cenrus,
Cervus sp. "maral ", Rangife7; Bos, Equus. Coe/odo1110, Mammuthus.
10-t 12 dark grey layer: Leo, Crocotta, Cervus.
Cer1us sp. "mara/", Bos, Equus, Coelodonta.
I I+ 14 Ltghl brown layer: Leo, Crocoua, Ccrvus.
Cervu.\' sp "moral", Mega/oceros, Rangife1: Bos.
Bison, Eqtms. Mammuthus (MOTIL 1938c. 2 112 16).
15. Erd - Parkvaros, Middle Palaeolithic site
Lower and upper layer complex: Leo, Crocolta,
Cervus, Mcga/oceros, Rcmgifer, "petit B ovid" (? Ovibo.\, small Bison according to 1. VOROS.), Bison. Asi11us, Equus, Coelodonta.
Also occurring in the lower hlyer complex: Mammuthus, in the upper layer complex: Sus and Rangifer
too (KRETZOl 1968 a .. 77-90. Tables on pp. t;i2--63).

t 6. Poroszlo, Tis:za rive-r bed

Alces. Megaloceros. Bison. Asinus, Equus, Coelodonta, Mamm111h11s (JANOSSY-VOROS 1979. Nr.

486).
17. Tnta, Porbanyo quarry, Middle Palaeolithic
site
KORMOS's cx.cavation ( 1909)
ln loess: Leo, Bison, Mammurhus (KORMOS L912.
16, 20).
"ln rock clefts": Crocoua, Sus, Cervus, Megaloceros, Equus. Coelodonto (KORMOS 1912.
23- 25).
From the travertina: Crocotla. Megaloceros. Asinus, Equus, Coelodonta, Mammulhus, (KORMOS
1912, 22-23., KRETZ011964b. 116., 119.).
Vertess excavation (1958-59), Sus, Cervus, ?
Megaloceros, Bos seu Bison, Asim1s, Equus, Coelodoma, Mwmmahus (KRETZOl l964b. 121- 122).

(MOTIL 1938b. 37, 1941. 14).

20. Ballavotgy rock shelter, Biikk Mts.
Yellowish brown layer: Crowlfa, Cervus, Megaloceros, Rangife1-, Bison. Cen.ius js missing at MOITL
(L941. 18), completed by M. KRETZOl in VERTES
1965. 283.
21. Berva cave, Biikk Mts.
2. light brown layer: Crocotta, Cervus sp. (elaplms
major, moral ?), Megalocero.~, Bison, Equus, MammuJht1s (KADlC-MOTTL 1938. 20, MOITL 1940.
1905, 1941 . I 3). Cern1S os ph. 11. is correctly os pb. I.
(KADlC-MOTTL 1938. 19- 20).

22. lJenavolgy rock shelter, Biikk Mts.

1. light brown layer: Crocorta, Sus, Cupreolus,
Cervus. Megaloceros, Rangifer, Bos, Bison (MOTTL
1936. 151, MOTTL l938a. 34, later only Bison
MOTIL 1940. 1910. 1941. 13). Equus, Coelo-

donta.
23. BildlSspest cave" Biikk Mts.
6. greenish grey layer: Cervus, Alces, Megaloceros.

Rangifer. Bison, Equus, Coelodonta, Mamm111Jws

(l<AOIC 1934. 64).
S. upper (dark grey) culture layer: Crocotta, ''large"
Cervus, Alces, MegaloceJ'Os, Rangifer, Bos seu Bison,
Bison. Equus, Coelodonta (KRETZOl 1-927, KADIC
1934. 64, MOTTL 1938b. 49)
3. lower (dark grey) culrure layer: Uw. Crocotta,
Cenus, Alces, Megaloceros, Rangifer, Bos scu Bison,
(K.RETZOl 1927, later Bos KADlC 1934. 64 and Bos,
Bison M01TL 1938b. 49, 1941. LS}, Equus, Coelodonta. Cervus is only mentioned by KRETZ01
( 1927), missing from the publications of Kadic and
Monl.

I JJ Csakvar cave, Vertes Mt .

Main branch - light brown layer: Leo, Crocotta,
Cer\llls (pierced "pearl tooth"', canine), Megaloceros.
Rangifer, Bison, Equus, (KADIC- KRETZO! 1927.
4-5).

189

Light grey layer: Leo, Croco/fa, Cervus sp, Me galoceros, Rangifer, Bison, Eqrms, Coelodonla
(MOTTL 1938b. 42. 1941. 14, KRETZOJ 1954a.
42-43).

28. Megyefa rock s helter, Mecsck Mts.

Leo, Crocolta, Cenl/lS elaphus sp.. Megaloceros,
Bison, Equus. Coelodonta, Mammuthus (K.RETZOl
1942a. 364).

24. Gorombolytapolca cave, Bilkk Mts., Szentkereszt Mt.

Niche ill., 3. brown-brownish yellow layer: Crocoua, Cervus, Megaloceros, Rangifer (VERTESKRETZOl- BERTALAN 1.965. 82).

29. Mexik6v61gy cave, Bilkk Mts.

4. brown layer: Crocolla, Megaloceros, (MOTTL
1938b. 41).
Leo, Cerv11s elaphus sp., Megaloceros (.KRETZOI
l 942b. 268).

25. Herman Otto cave, Bilkk Mts.

30. Miskolc-Tapolca rock shelter, BukkMts.

Mam branch - entrance hall

5. yellow layer: Megalaceros.

Megaloceros, Bison, Equus, Coelodonra, Mammuthus (JANOSSY-VOROS 1979. Nr. 379).

3. yellow / limestone breccious layer: AIces.

2. yellow layer with limestone grit: Crocotra~
Cervus (2 pieces of pierced "pearl tooth.. canines), Alees, Megaloceros, Bos sp. ? (KADJC 19l6. 10-1 1,

31. Remete Upper cave, Buda Mts.

l. Small hall
4. yellow layer: Leo. Cervus, Megaloceros sp..
Rangifer, ''Ovibos" (= small Bison VL). Bos seu Bison. Equus, Coelodonla, M.ammuthus.

Fig. 9, EHIK 1916. 25, MOTIL 1938b. 41-42, 1941.

11) Bos (MOTTL 1942 a. 101).
26. Jankovich cave, Gcrecse Mts.
Side branch - Dome room. reddish-yellow Jayer:
leo, Crocollu, Rangifer. EquUJ', Coelodunta (lfilLEBRAND 1915. 133. MOTTL 1938b. 45, 1941.
14), Without layer assignation Megaloceros, Bos seu
- Bison - data by M. KRETZOT (VERTES 1.965.
305).
27. Kccskesgalya cave, Bukk Mts.
Main hall - 3. Light brown lay(lr (crwncoui;Jy pub~
lished as dark brown in VERTES 1965. 3 LO): Crocolla, Megaloceros, Rangifer, Bos '!, Bison, Equus.
Coelodonta (KADIC 1940. 219, MOTf L 1940. 1903,
1945. 1526).
4. dark grey layer (erroneously published as greenish grey by K.ADlC 1940. Ill. map, adopted by
vERTES 1965. Fig. 18.): Leo, Crocotfa. Cervus,
Mega!oceros, Rangif~r. Bison, Eqmts (KADIC 1940.
220-22 1, MOTIL 1940. 1902- 1903). Jn the publication of K.adic, Cervus canadensis asiaticus or Cervus
maral was missing from the fauna list though, it
was included in the description of species (KADIC
l 940. 219-22 1).
ln the contracted tauna list of 4. dark grey+ 3. light
brown layers contains Cervus elaplws f. major
(MOTIL 1941. 10) as well.
Rear pan of the cave (results of the 1937 excavation):
3. light brown layer: Oocotta, Megaloceros, Rangrfer, Bison, Equus, Coe/odonta.
4. dark grey layer: Leo, Crocofla, Cervus (maral ?),
Megaloceros. Rangifer, Bison, Equus, Coelodonra
(MOTIL 1945. 1526-1527).

190

11. Large hall

4, yellowish brown layer, upper level: Leo. Crocotta, Cervus, Mego/oceros sp., Rangife1; "Ovibos " (=
small Bison VI.), Bos scu Bison, Equu.\, Coelodonla.
Mam1T11Jthus.
- lower level: Leo, CrocotttJ, Cervus, Megaloceros
sp., {data by M. KRETZO! in GABORl-CSANK
1983, 1993. 61-63).
32. zeleta cave, Bii.kk Mts.
61 (upper) light grey 1 red culture layer: Leo, Crocolfa, Megaloceros, Rangifer. Equus, (KADlC 1934.
35, MOTTL l938b. 45, 1941. 14).
4. dark grey layer: Crocotta, (KADIC 1934. 35).
3. li&ht brown layer: Leo, Crocotta, Megaloceros,
(KADIC 1934. 35).
2 (lower) brown /dark brown layer (erroneously
published as light brown by MOTTL 1941. 14, 1945.
1521): Leo, Crocotta, Cervus (elaphus), Megnloceros1
Rangifer, Bison, Coelodonta, Mammulhus (KADIC
1934. Fig. 4. 34, MOTIL l 938b. 36, 1941. 14, 1945.
1521).
33. Szelim cave, Gerecse Mts.
3. brown/dark brown "C" layer (with hyena): Leo.
Crocotla. Cervus canadensis as1aticus. Alces. Megaloceros, Rangifer, Equus, Coelodonta, Mammu thus (MOITL 1938b. 43, 1941. 13, GAAL 1943b.
435).
1. yellow (GAAL l 935.). light yellow (GAAL
194l. fig. 7), brownish yellow (GAAL 1943b., 434)
"E" layer: Leo, Crocotta. Megaloccros, Rangife r, Bison, Mammuthus (data by D . JANOSSY in VERTES
1965. 344-346).

3.t. Ba latonsza badi - Safranykert

1n cbe bonon1 of a high locssy-sand wall, I 08 m.
n.s.I. in loose sand: Megaloceros, Bison, Equus,
Coelodonta, Mammuthus (KADIC 1911. 9).

47. Kiraly kut spring, cleft

Me;alocero.r. Equu.s, Mammuthus (J...\.NOSSYV0 R0S 1979. Nr. 367).

48. Kiskunfelegybli.ta - Bnckynrd

35. Bares

Megaloceros, Co(!/odonta (JANOSSY-VOROS

1979. Nr. 69)
36. Csepcl - gravel quarry, Da nube "alley
Megaloceros, Mammut/111s (JANOSSY-V0R6S
1979. Nr. 2 L2).
37. Csongrad environs, Tisza river bed

Cervus, Megaloceros. Bos, Coelodonta, Mamm111h11s (JANOSSY-VOROS L979. Nr. 527).

" under the level of forest soil": Loo. Sus. Cervus.

A/ces, Megaloc:eros, Equus, (BENDA 1929 268270., JA1\0SSY-VGR0S 1979. Nr. 449-450).
49. Nagymaros, loess

Megaloceros, Eq1111s, Mammuthus (data by M.

KRETZOT in vERTES 1965. 357.).
50. N6grndver6ce - Bnckyard, loess
Megaloceros. Eq1111s. Mammuthus (MOTTL 1942b.
48-49).
s1. Paszt6

38. betweeo Csoograd and Szegvnr. Tisza river bed

Sus, Cervus, Megaloceros, Bos, (JANOSSYVOROS 1979 Nr. 529).

Nr 311).

39. Danube ri"er bed, without closer locality

Megaloc:eros. Bos seu Bison. J.fammutlzus
(J.ANOSSY- VOROS 1979. Nr. 265).

52. P6kaszepetk, forest sa nd quarry

Megaloceros (JANOSSY-V6ROS
62.).

40. Endr()d
Mega/ocero.~, Mammuthus (JANOSSY-V6ROS
I 979. Nr. 462).

53. Pilismarot-Basaharc, B rickyard

41 . Eger

Megaloceros, Bison (JANOSSY- VOROS l979.

Nr. 361).

Cervus, Megaloceros (JANOSSY-VOROS 1979.

1979. Nr.

Upper loess complex: Equus.

In loess becween lwo layers of loam: Megaloceros, Bison, Equus, Mammuthus.
Lower lhick loam layer: Cervidae. ind_ Bison, Coelodonta, Mammt11h11s (MOTIL l 942b. 52. V6ROS
l 990a. 8. fig. I.).

42. Ercsi

Megaloceros. Coelodonra, Mammuth11s (JANOSSY- VOROS 1979. Nr. 246).

43. Fegyvernek, Tis:ia rin r bed
Cervus, Alces, 1\,fegaloceros, Bison. Mammuthus
(JANOSSY~V6R0S 1979. Nr. 497).

54. Raba river bed, Gycir environs

Alces, Megaloceros, Ra11gife1; Bison. Coelodonta,
Mommuthus (JANOSSY-VOROS 1979. Nr. 6).
55. Saj6kazinc,

Megaloceros, Coelodonra. Mammuth11s (JANOSSY- VOROS 1979. Nr. 386).

44. Gyoma

Megaloceros, Mammuthm (JA.NOSSY-VOROS

1979. Nr. 461).
45. Hatvan - Brickyard, gravel layer

Megoloceros. Bison, Equus, Coelodonta, Mam m11thU1 (JANOSSY- VOROS 1979. Nr. 324).
46. J:is2bcreny - sand quarry beside lime deposit
Megaloceros (JANOSSY-VCROS 1979. Nr.
440).

56. S:igvar -in "sand with concrelions", under loess

Alces (GAAL J 933a., VOROS 1982. 45).

57. Szeged environs, Tisza river bed

Megaloceros, Bos, Bison. Coelodo111a, Mammuthus
(JANOSSY-V0R OS 1979. Nr. 531).

58. Szob - Brickyard, loess

Megaloceros, Equus (V6ROS l 998).

191

59. Szolnok - Tiszn river bed

V. I STALLOSK6 FAUNAL PHASE

Cervu.~.

Alces, Megalocervs, Ra11gifer, Bos, Bison,

Equus. Coelodon1a. Mammu1h11s (JANOSSY- V6-

72. Aggtelek - Baradla cave, Aggtelek Karst Mts.

ROS 1979. Nr. 50 1).

60. zolnok - Pusztavarsany, Tisza river bed

Megaloceros. Bi.von, Coelodoma, Mammwhus
(JANOSSY-VOROS 1979. Nr. 492).
61. Szolnok-Zagyva fork

Main branch and Dencver (Bat)-branch, yellow-red

layer: (Ursus spelaeus. Vu/pes v11lpes),

Sus scrofa, Ce1vus elaphtL'i, Ran~fer rarandus, Equus

sp., Coelodoltla a11tiquitatis, .MLVnmutJms primigenius
(VERTES 1954. 18, VOROS 1974., 1980. 17.
JANOSSY-VOROS 1979. Nr. 406).

Megaloceros, Mammullws (JANOSSY-VOROS

L979. Nr. 502).

124J Gorombolytapoka rock s helter Biikk Mts.,

Cervus, Megaloceros. Bison, Coe/odonta, Mummwhus (JANOSSY-V6R6S 1979. Nr. 485).

Szenlkereszt Mt.
Yellowish brown layer: Oocotta, Cervus, Rangifer
(HlLLEBRAND l935. 24, MOTIL 1941. 12). Bos
seu Bison, without layer assignment. data by D.
JANOSSY in VERTES 1965. 294).
~iche ill. yellow layer: 3 pieces of pierced Cen,us
teetlh (incisivus) (VBRTES- KRETZOI-BERTALAN 1965. 8 l-82. Taf. I. 1- 3).

64. Tiszukccskc - Tisza river bed

Alces, Megaloceros, Bison (JANOSSY-VQROS
1979. Nr. 521).

73. I Laromkuti cave, Bilkk Mts.

First room - greenish layer: Crocotta, Capreolus,
Cenrus(KADIC 1914. 189, MOTTL 1941.20).

65. Tiszasiily - TiS7Jt river bed

74.lstalloskll cave, Bukk Mts.

V. yellow layer: Leo. Crocotla (VOROS 1984.
TabJ.1.).
IV. yellowish brown layer " in situ" large hearth:
Leo. Crocolla, Sus, Cervus f. major , Rangifer, Bison,
&Jrws (Mammut/ms in the form of a tusk artefact),
(MOTIL 1942a. 91 , l 944. 44-45, JANOSSY 1955.
159-163, VOR6S 1984. Tabl I.).
III. (dark brown. reddish brown, hghl brown. greyish brown, grey) layer complex, three different cultural layers: Leo, Crocotta, Cenms f major, Alces.
Rangifer, Bison. Equus (Mammutlms in the form
of a tusk artefact), (VERTES 1955. 119, 129-130.
IANOSSY 1955. 159-163).
II.dark brown, "lower Layer with microfauna"
I. layer complex(light brown, dark brown layers)
with two culturaJ layers: Leo, Crocolfa, Cen~us f. major, Rangife1; Bison (Mammuthus in the form of a rusk
artefact) (VERTES l955, I L9, 129- 130, J ANOSSY
1955. 159- 163).

62. Tiszad.ada - T isza river bed

Megaloceros, Bison, Mammuthus (JANOSSYVOROS 1979. Nr. 479).

63. Tis1.afiired - Tisza r iver bed

Cervus. Alces. Megaloceros, Bison, Mammuthus

(JANOSSY-VOROS 1979. Nr. 491).

66. Tiszaszc>Hls - Tisza ri ver bed

Megaloceros, Bison, Coeiodott1a, Mammuthus
(JANOSSY- VOROS 1979. Nr. 487).
67. Tiszaug - Tisza river bed
Mega/oceros, Bos, Bison. Mammt11hus (JANOSSY- V6R6S 1979. Nr. 524).
68. Tiszaug - Kecskemeti sz1krn, Tisza rh er bed
Cen11s, Megaloceros. B1w11, Mammu1l111s (JANOSSY- VORCS 1979. Nr. 525).
69. Vac-Csipkes - Upper Palaeolith1c site. underlying
loess
Megaloceros, Coelodonta (VOROS 1998.).

70. Vezseny - TiS"La river bed

A/ces, Mega/ocePO.'i, Mammuthus (JANOSSYVOROS l 979. Nr. 515).
71. ZaJaegerszeg-11. Brickyard, 10-15 m deep

Megaloceros , Bison, Equus, Coelodonta, Mammurhus (VERTES 1954. 17, JANOSSY-V0ROS

1979. Nr. 56).

126.f Jankovicb cave, Gerecse Mts.

"Uppermost"

yellow

layer:

(HILLEBRAND 1919. 9, fig. 3.).

192

Rangifer, Eqiars,

Coelodonta (KORMOS-LAMBRECHT 1914. 78-79,

MOTIL 1941. 16).
T.n the front part of the cave, "Magdalenian age", in
"microfnuna-rich" part of yellow ? layer, polished
bull-head bone awl made of Alces metapodium

/27J Kecskesgnlya cave, BUkk Mts.

Hatl. 2 yellow layer: Bison. Equus (K.ADlC 1940.
219, MOTIL 1940. 1902).

81. Pmsszant6 II. rock shelter Pills Mts.

7. light grey layer (in the m iddle): Leo, Cmcoua,
Alces, Bos, Bison, Equus, Coelodonto (VCRCS
l 986b. 35, Tab!. I).

ISJ l<iskevcly cave Pilis mts

3 yellow layer, Leo, Ce1vus. Rangifer. Bison.
Equus, Coeladonta (MOTIL 1941. 17).
4/3. layer: Cervus. Alces, Bison, Eq11us. Coelodonta (VCROS 1994. Fig. 8.Tabl. 2.).
75. Uikviilgyi cnve, Bilkk Mts.
ln main hall: leo, Crocofla, Cervus, Bison
(MOTTL 1938b. 38., 1940 1913. 1941. 12).
76. Orfii, pring cave. Mecsek Mts.
Cervus, Equus, Coelodonta, Mommulhus (vERTES 1954. 21).
77. Perpnc cave, Bukk Mts.
Light br~wn layer: Crocolla, Rangifer, Bos seu Bison (KADIC 1940. 2 l 0).
78. Pesko cave, Buk.k Mts.
3. red brick red layer. rich in miorofauna: Leo,
Cmcolfa, Capreolus. Cervus ( major, Rangifer. Bison, Eq1111s, Coelodonta (EHtK 1914. 194, KADIC
1935. 51. MOTIL 1940. 1915, 1941. 16, 1944.
13-16).
4. greenish grey layer (erroneously mentioned as
greenish brown by MOTIL 1941. 11 ): Leo, Crocotla,
CeJws f. major. Rangifer (KADlC L935. 51, MOTIL
1941. 11, I 942a 95)
5. dark brown layer: Crocotta, Sus, Capreolus.
Cervus f. major, Rangifer. Bison. Equm (4-5. layer
l\IOITL 1940. 1918, 1944. 9- 22). Mentioned large
deer and small Bison (MOTIL 1944. l l.).
J. Hir performed an authentication excavation of
the so-called "V6rtcs's section" (HlR 1990., Table
I.):
4. grey layer:
in samples 7, 9, 11 , 13: Cervidae
in sample
12: Alces
5. brown I dark brown layer:
in sample
16: Cervidae
in sample
18: Crocotta.
79. Petenyi cave, Biikk Mts.
Lower brown layer: Crocoua (VERTES 1956. 4).
80. Pilisszanto I. rock shelter Pills .Mts.

Lower layer complex! D6 brownish grey

D7

greyish brown layer: Leo. Crocoua, Cervus "'maraf',

Rangifer, Bison, Equus, Coelodonta, Mammuthus
(K0&\10S 1915. 328-329, VORCS 1987c. Tabl.
1- 2) Pierced deer "pearl tooth" (caninus),( KORMOS
1915. Table
8.).

82. Remete Lower ca"e, Budai Mts.

12. rusty brown layer "c": Cervus (JANOSSY
1953. 420, 1979. 152).
1n the 11. yellow loessy layer "b" the Coelodoma
find belongs to this pbase in age as well.
83. Solym:ir - Quarry. m cleft, Budai Mts.
Red-brown layer: Leo. Crocorta, Capreolus, Cer111LS. Alces, Equus, Coelodonta (KUBACSKA 1927.
63-64, VOROS I 988a. Tab I 4 BI I).

133.! SzeUm cave, Gerecse Mts.

4-5. (upper) yellow 'B" layer:
5. B1 layer: Rangifer, Coelodonla (GAAL 1935
55).
4-5. 81+2 layer: Crocoua. Cen'IJS, Cervus can.
asiat., Cervus moral, Aices, Rangifer. Eq1111s, Coelodonto, Mammuthus (Data by, JANOSSY in vERTES
1965. 345).
4. 8 2 layer: Crocolla, Rangijer Alces. Equus,
Coelodonta, Mammuthus (GAAL 1935. 54-55,
MOTTL 1938b. 45, 1941. 15).

The marking of layer B2 as Upper or Lower is erroneous (vERTES 1965. 345), it should be correctly
marked as Layer B upper= B 1; Layer B lower= B2, respectively.
/13J Tarkli rock shelter, Bilkk Mts.

Block II. upper "2" brown layer: Capreo/us, Cervus

cf e/aphus, A/ces, Equus.

Block ll. lower "3" brown layer: Capreolus,
Cervus, Rangifer, Bison.
Block Clla, upper "4" yellowish layer: Leo, Cervus
cfi e/aphus, Rangifer, Bison.
Block ura--b+V. (= KA.DIC 1944. 79), "6" yeJlowish grey layer: Leo, Crocolta, Cervus, Bison.
Block VII. "9" layer: Leo, Cervus, Bison
(JANOSSY 1976. 11- 17, Tabl. 11/B-C. Cervus of
Block VII. is missing from Tab!. Il/C).

84. Abaujszint6, gravel

Rangifer, Equus, Coelodonta, Mammuth11s (IANOSSY- VORCS 1979. Nr. 419).

85. Bodrogkercszl11r - Henye hill, Upper Palaeolithic settlement

Leo, Cervus "'moral"', A Ices, Bison, q1111s, Mam-

muthus.

xxm.

193

86. Badacsony, in sand

Bison, Equ11s. Coelodonla, 1\llammuthus (JUGOVJCS

KRET701-CSANK 1954 92}

87. Dunafoldvnr - Coboljaras, Mammoth skeletons,

Upper Palaeolithic butchering ~ite
Bison, Mamm111hus (CS.ALOGOVTTS 1936., VO-

ROS 1981 b.).

88. Dunaszekcso, loess
Alces (JANOSSY-VOROS 1979. Nr. 261).

89. Esztergom environs

Rangifer. BiM n. Coe/odonla. Mammuthus

101. Lovas, dolomite quarry, Upper Palaeolithic

paint mine
Sus, Cervus, Alces. Rongifer, Equus (DOBOSlVOROS 1979.).
102. Madaras-Brickyard, '"Lower" loess under the
Upper Palaeolithic site
Alces, Bison, Equus, Mommuthus (VOROS 1989
Tobie 1).

103. M:\lyi - gra\'cl quarry,

Alces (JANOSSY-VOROS 1979. Nr. 374).

CJA-

NOSSY-VORCS 1979. Nr. l61).

90. Ftizcsnbony - gravel quarry
Alces. Bison, Eqzrns, (JANOSSY-VOROS 1979.
Nr. 3421

104. Matraballa - sand quarry

Bison, Coelodo11ta, (JANOSSY-VOROS 1979.
Nr. 376).
105. Merk - Kruszna cbanncl

Crocotra, Cervu.<'. "P

Bison,

91. FilLcsabon) sand quarry.

,J/c, tJA V>~S'1 VOROS I '74 l\r .,.+3).

mwhus (HALAV AT 1898 ).

92. Fokoru, Tisz~ river b<'d

Alces, Mammutlm.v (JANUS!::IY-V6RCS 1979.
Nr. 494).

106. Miskolc
Crocotta, Bison, Equus
1979. Nr. 389).

93. Galgah~h i..t

Bos scu B1so11, Eq 11 ~ Coeludonta, Jlummuthus
(JANOSSY-V0ROS 1979. Wr. :!QR).

107. Miskolc-Quar~1

94. Hatvan

Cer111s. Mammuthus (JANOSSY-VORCS 1979.

Nr.321}.
95. J aszkaraj eno

Bison, Mammuthus (STANCZlK 1975 JANOSSY- VCROS 1979. Nr. 514).

96. Kaposvar, loess

Cervus, Coelodonta, Mammut/111,'i (JANOSSY-

Coelodo11ta. Mam-

(JANOSSY- VOR6S

Bison, Mammuthus (JANOSSY-VOROS 1979

Nr. 39 1).

l 08. Monor, Mummoth skeleton

Cervw;, Equu.\'. Mammuthus (GAAL 1928.).
109. Nagybatony
Alces, Bison. Equus, Mammuthus (JANOSSYVOROS 1979. Nr 304).

llO. Nagykoril, Tisza river bed

Bison. Mammuthus (JANOSSY-VOROS 1979.
Nr. 496).

VOROS 1979. Nr. 11 2).

97. Kaposvar - Pecsi road., m depth of 15 m
Rangifer, Coelodonw (JANOSSY-VOROS 1979.

Nr. 11 l).

111. Nagymaros - Ujvolgy, Fischer Tobias's "sotl

quarry''
Cervus f. mo1or, Alces, R'!_ngifer, Bison, Coelodo111a, (MOTTL L942b. 47., VOROS 1998.).

Cervus, Mammu1hus (JANOSSY-VORCS 1979.

Nr. l 10).

112. Nagymaros - beside narrow-gauge ("small") railway-station.

Alces (MOTTL 1942b. 47, - VOROS 1998.).

99. Kotelck, Tlsza river bed

Cervus, Alce.\. Bison, Ma111muth11s (lANOSSYVOROS 1979. Nr. 493).

U J. Padragkut - Sand quarry,

Rangifer, Eq11us Mammmhu.s (JANOSSY-V6R0S 1979. Nr. 27).

I 00. Labatlan
Leo, (Ursus), Coe/odonta, Mammutl111s (JANOSSY- VOROS 1979. Nr. 154).

U4. Ncsbnnyatclep - Kaposztas valley, Mammolb

skeleton
Cervu.<>, Mammutlrus (RlHMER 1935. 28).

98. Kolesd

194

US. Pillsmarot, loess

(Marmota), Cervus, Rangifer, Equus, Mammuthus
(JANOSSY- VOROS 1979. Nr. 164).

129. T6szcg cliff, Tisza r iver bed

Alces, Mammutlius (JANOSSY-VOR6S 1979.
Nr. 512).

116. Pilismar6t-Oregek dulo, Upper Palaeolithic

site
Alces, Rangifer, Equus, Mammuthus (MOTTL

130. VAc-DCM, loess

1942b. 53-54).

131. Verseg - Kcrtekalja, Upper Palaeolithic settlement

Rangifer. Bi.~on, Equus, Mammuthus (VOROS
199lc. 88-89).

U 7. Polgar, Tisza river bed

Alces, Mummuchus (JANOSSY-VORQS 1979.
Nr. 482).

Ll8. Rakoc.tifalva, Tisza river bed

Bison, Mammuthus (JANOSSY- VOROS 1979.
Nr. 510).
119. Rombany. in cavity of Triassic limestone
Cervus, Alces, (JANOSSY- VOROS 1979. Nr.

Bison. Equus, Mammutlws (VOROS 1998.).

132. Visegrad - Hamzsabeg

Cervus, Mammuthus (JANOSSY-VOROS 1979.
Nr. 170).
133. Zebegeny
Cervus, Rangifer, E{!mlS, _Co~lodonta, (KOCH
1900. 548. 550, JANOSSY-VOROS 1979. Nr. 175).

278).

1979. Nr. 338).

134. Zebegeny - K alvaria hill, loess

(without exact layer assignation): Cervus cun.
asiat., Blvon, Eqt111s, MammrJfhus (GAAL I933b.
130. VOROS 1998).

12 l. Szajol, Tisza river bed

Bison. Mammuthus (JANOSSY-VOROS 1979.

135. Zirc-Cuha stream, loess

Bison, Mammuth&1s (JANOSSY-VOROS 1979.

Nr. 500).

Nr. 29).

Jl20. Sirok
Bi.son, Eq1111s, Coelodonta (JANOSSY-VLlROS

122. Szende hely

Leo, Cerms, Equus, Coeiodonra, (JANOSSYVOROS 1979. Nr. 282).
I 23. Szob - beyond Kalvaria, upper sandy part of
gravel
Cerv11s can, asiat., Rangifer. Mammuthus (KEZ
L934. 14).

124. SzOdliget
Bison, M.ammuthus tJANOSSY-VOROS 1979.
Nr. 195).
125. Szolnok - SArnyak, Tisza r iver bed
Alces, Mammuthus (JANOSSY-VOROS 1979.
Nr. 503).
126. Szolnok - Sokoru, T isza river bed
Aices (JANOSSY-VQROS 1979. Nr. 508).

127. Tiszal6k -Rbom puszta

Alces (JANOSSY-VQROS 1979. Nr. 478).
128. TiszaH>k - Dam construction, Tisza river bed
Bison, Coelodoma. Mamm11thm (VERTES 1954.
L9).

VJ. PILISSZANTO- VIL BAJ6T FAUNAL

PHASES
136. Also Sz.inva (Cave over the "lower spring" of
Szinva stream) Biikk Mts.
Cervus (JANOSSY-VOROS 1979. Nr. 366).
137. B ajts cave, Gerecse Mts.
Yellow, "b" layer: Cervus can. asiat.
Lighter yellow, "c" layer: Leo, Rangifer (GAAL
1929. 4-6).
/19./ BaUa cave, Biikk. Mts.
Front part, (upper) yellow layer: Capreolus,
Rangifer, Bo\' sp. (=Bison) (HlLLEBRAJ'1D 19l2.
780).
/22J Bervavolgy rock s helter, Btikk Mt.

2. yellow - hght yellow layer: Crocotta, Capreolus,

Cervus {f. major), Equus (MOTIL 1'938a. 26,
MOTIL 1940. 1909).
U8. Bivak cave, Pills Mts.
Pleistocene upper 1. (yeUow) layer: Sus, Cervus,

Rangifer.

195

3. (yellowish grey) layer: Croco11a, Rang{fer, Bison

sp., Equus
Ple1sroceoe lower 4. (grey) layer: Sus, Cervus,
(large form. fonnerly "Megaloceros", JANOSSY
1979. 144) Rangifcr (KRETZOl--VARROK 1957.
22-23).

and Crocotta are older fauna elem eurs occurring in 11

secondary position.
Middle layer complex:
03 greenish grey + D4 greyish yellow + Ds rusry
red layer: Leo, Sus, Cervus moral, Rangifer, Bi.son.
Equus (KORMOS 19JS. 326- 327. VOROS 1987c.

Tabl. l-2).
/23J Biid()spest cave, Biikk Mts.

7. reddish brown k1ycr: Cervus (formerly "l11Jegalocero.f' KRETZOI 1927, KADIC 1934. 64), Rang/fer ,
Bison. Equ11.\ (KRETZOI 1927. KADlC 1934. 64)
n 4J Istfi116sko cave, Bukk Mts.
Vl. light yellow layer marked as "upper microfauna-1ich layer": Rcmgifer (V6ROS 1984 . Tab.I. I.).

126J J ankovich cave, G erecse Mts.

ll. Block 4. (between 9~11 0 cm) Rangifer.
(BACSKAY- KORDOS 1984. Table 1.).

139. Kiralykut pot-hole, Bilkk Mts.

Crvco11a, (SEBOS 1934. 14).

ISJ Kiskevely ca ve, Pi.lis M ts.

2 ycl lov.ish g1e~ layer. Rc111gifo?r, Equus (IITLLEBRAND 1913 2(}..21, VOROS 1994 ).

140. Koros cave, Bukk Mts.

Dome hall, grey luyer: leu (KAD!C 1944. 76).
141. L illafiired rock shelter, Bfi k k Mts.
Yellow layer: Sus, Cenw (KORMOS 191-+a. 202).
142. 6huta (Btlkk.szent laszJo) - Cseogos teber,
Bfikk Mts.
Rangifer, Bison (VERTES 1954. 2J ).
143. Orfii - Sark.Snykut, ivtecsek Mts.
Crocoua, Bison, (KOCH 1.900, 542, 552).
144. Olyvesko rock shelter, Bilkk Mts.
2. light brown layer (eotrance): Rangifer, Equus.
3. greenish grey layer: Leo, Rangif er (MOTTL
1944. 63, 66).
/78./ Pesko cave, BUkk Mts.
2. light yellow I yellow layer: Rangifer, Bison,
Equus, Ovibos (MOTIL 1944. 19).
/80./ Pilisszanto I. rock shelter , Pi Iis M rs.
Upper layer complex:
D1 light yellow + D2 dark.er yellow layer: (Leo,
Crocotta), Cervus canadensis asia1., Rangifer, Bison.
Equus (KORMOS 19 15. 324-325, VOROS l987c.
Tabl. 1- 2). According to U1e opinion of the author, Leo

196

145. R cjtck I. rock sh elter, Blikk M ts.

TriaJ pit ..3" Crocotta sp.
Block UJ. I 4~160 cm. "9": Sus (? domes1ica,
JA.NOSSY 1962. 57), Cervus, Rangifet.
180-200 cm "ll'': Bison (JANOSSY-KORDOS
1976. Tabl. II.).

/82./ Remete Lower ca' e, Budai Mts.

11. yeJJow lb/ layer: Cervus, Bison (JANOSSY
1953. 420, 1979. 152).

146. Rcmetehcgy rork shelter, BudJti Mts.

Upper yellow layer ("with rodents"): Bos seu Bison.
Lower reddjsb brown layer: Crocolla ?, Rwzgifer,
Equus (KORMOS L914b. 357. 36~361).
/32 J Szeleta caH, Bukk l\1ts.
Entrnnce, light yellow layer: Rangifer, Eq1111s
(KADIC JQ34. 35).

147. Vaskapu cave, Bilkk Mis.

2. hgbr bro" n layer: Rangifer, Bison,
3. greenish grey layer: Rangifer (KAD!C-MOTIL
1938. 6~ 1).
148. Arka - Hcrzsarct, Upper Palaeolithic settlement
R.angifer, Bison, Equus (VOROS 1987b. 89-90,
Tabl 1.).
149. Esztergo m-Gyurgya lag, Upper Palaeolithic
settlem ent
Rangifer, Eqm1.S (anefact made of Mammuthus
bone) (VOROS l99 l b.).
150. Nadap - Stone quarry, Upp er Palaeolithic settlem ent
Rangifer, Bison, Equus (VOROS 1988b. 34-35,

Tab1. 1.).
151. Pilismar6t - Dios, Upper Palaeolitbi.c settlement
Rangifer, Equus (VOROS 1981 c. 22).
152. Pilismarot - Pilret, Upper Palaeolithic settlemen t
Sus, Rangifer, Bison, (VOROS 1983d. 299).

153. Tarcal - Citrombaoya, Upper Palaeolithic settleme nt

Rangifer, Equus (JANOSSY 1975. 26).

156. Vac-Csipkes,

Upper

P alaeolithic

settle-

ment

Rungifer, Equus (VOROS 1998.).

154. Sagvar, Upper Pal:aeolithic settlement

Rangifer, Equus (VOROS 1982.).
155. Szeged - Othnlom, Upper Palaeolithic settlement
Rangifer, Bison, Equus (VOROS 1987b. 89, Tabl.
I }.

On the 157 s ites, altogether 24 l layers / levels were

separated, assigned currently into 180 chronological
''unitS'. Among these 180 units there are 2 clefts, 75
caves and I 03 open air sites.

The biostratigraphical division of site types: (Tables

1-111)

~""'"'~
no

3
11

ARBO
'\llBALYlJI-:
S7!:LETA
ISTALL6SK0
P1116Sz.ANT6.-

32

BAJ T

I. cleft, 2:

C.avc site!>

Slte.)IOlld

cnkar.:ou~

Jl'r--1

Open 81r Sites

'--11~

SS

17

J8

70

18

5:?

22

10

157. Zebegeny-KilvAria bill, Erzsebet s1.r. 25 (currently, Szonyi I. str. 23), Upper Palaeolithic settlement
fn the bottom of the 14 m high loess wall, from I.he
culrwal layer: Rangifer, Ovibos, Eq1111s (GAAL
1933b, VOROS 1998.).

Frequency o(
~'84if 1es on 1hc:

Species

Pcrcenlllj!C fo

Nwnbi:t of sries

StlCS

1 0 Upper

PlelSlllCellC mes

Bov1cl4e

103

51,.3

2
3

Equus

92

51.2

Ccrvu~

88

Mnmmu1hus

K4

'16,7

5-6

Rnngdu Megalll-

66

Ct!fOS

60

36,7
36,7

63

JS.Cl
27,3

, Coelodonta

Cm:otlll

IJ

A lets

10.

Leo

49
42
38

48 9

23.~

21.2

11.

Sus

26

12

Caprcolu.s

18

14.S
10,0

13

A51Rllll

10

S.6

14-15.

Ovlbos/ D1111111

'

tuff

1.2

Tahle I. Carnivore and herbivore macro-mammals on Upper Pleistocenc localities (Tables 1-Tfl)

SOTTO FAUNAL PI IASE 3 sites

11

Csilk1-Ar c

Siltit'l(clvftl

jl

$ubal}U~

L
I.

c L

Sus

Cp

Cv

Br

Sus

Cl!

Cv

BsB1

Sus

Bs

CoeI

Cv

C:oel
Mumm

Mamm 1

VARBO FAUNAL PI LASE 11 sites

II I

Cslilm\r c

D16!gyor-Tapolc:i ~

5
6
7

K1sktvel)

S.

K1sk61Ui1
Lambrecht c IV- V

Hx
H'<

Hx
Hx
Hx

Su~

c
c

Sus

C11

f---

Sus

Cp

Cv

Mg

Cv

Mg

Cv
Cv

Mg

Bi
BsBt

Rg

Bt

Rg

Bt

As

As

Coe!

Coel

- --

Cocl

Mwnm

Coe.I

Mumm

197

VARBO FAUNAL PHASE l l siles (con.)

I

Porlyu\..c I

9.

Poroslyul. c

10
II

Pl)rgOlheg\ c V

ll

J3

llx

:;us

Cp

C1t

Co

Cv

H.'C

Suuocak 111ff
rarku ll~ ll la. VI.

Hx

I c

c~

Cp

llx

Cv

Coe I

Bov

Mg

c,

l.

l)j

c,

Solymarc. AJS

Rg

-~

131

Rg

SUBALYUK FAUNAL PHASE 9 sites

IS

016sgy6r-Tapaka c
c 1-11
K1skcvtly c 4

17 1

LlllT1brcoh1c 111

1-t

L~nrod c
l'Orgnlhcg) c IV-tn
Suhillyu1'; c U

14.f

110
/3./

IS
16
17

LrdHI
Porosz:lo

/17.I

Tata travemoo

I.

Sus

Cv

Mg

Sus

L
Sus

Cv

Cp
Cp
Cp

A
A

BsB1

Al;

Caci

81

/Is

Coel

Mn mm

Me.mm

Mg

Cv

Mg

81

As

CoeI

Mamm

Cv/m

Mg

Rg

Sr

Coel

Mmnm

Cv/rn

Mg

l<g

F.

C<M:I

fl;!amm

Mg
Mg

Rg

BsB1
Bi

As
As
As

Coe I

Mnmm

131

l\s

Coel

6sl31

As
As

Cod

F.

Cocl

Mlllllm
Mamm
Mn mm

,_

Sus

c,

c
c

Sus

C\

rain Joess

Rg.

Mg
Mg

SZELETA FAUNAL PHA SE 17 caves..- 38 open air site= 55 sites

18.

AmMlicl

19

Balla c

I.

20

Ballav()l~

21
22

Bcrva c

2J

BOd<lspeSI c 6-3

II./

Csakvllrc
Gllromb<llyurpolca c

24

c
c
c

Rs

BcrvavOlg)' Rs.
L

r1113

25

Herman 0 c 5-2

26

J.1111..ovich c

27
28
29

Ke,~kesga.lya c

3-4

l.

Mg
Mg

131

Rg

81

Cv
Cm

Mg

Rg

Br
Br

Cv

Mg

Rg

I:

,.__ Mamm
-

Mg

Rg

R1
BsB1
11

Cm

Sus

Sus

Cp

Cp

Cwm

c
c
c
c
c
c
c

Cv

Mg

Rg

c. .

Mg

B~B1

BsB1

Cv

Mg

E
I:

Cv

32.

Szele111c6-2

33.

c
c

Cv/m

3.5

Szehm cC-E
Bsl:uonnabadl
Bares

36.

c~pd-srnvcl-q_UllTT)'

Cv

Mamm

Coe I
Coe I
Cod

Cm
A

M&

Rg

Mg

Rg

Mg

Rg.

Mg

Bi
81

81
81
Bi

E;

CoeI

Mamm

Cocl

Mnmm

I:.

Coe I
Coe I

!:.

Coe I

Mamm
Ma.mm
Ma.mm

Cv

tv!g

CV

Mg

Bs
Bs81

Dnnubc nvcr bed

Mg

4-0.
41
42

endrod

Mg

Eger

Mg

.n.

Erc.si
Fegyvcrncl Tb

44-.

Gyoma

Mg

45.

lflUVllJl

Mg

C1:>el

Bs

Mg

Mnnun
Mumm
Mamm
Ma mm

81

Mg

Coe!

Mg

Cv

Mamm

l!
E

Mg
i

Sus

--

Mg

39

L98

81

Mg

Cocl
Coel
Cocl

Bs

Rg

Csongrad-Sn~gvlir

Mg

Rg

JI.

38

M.&

Cse1ngrtitl c 1'inn b

E:

Mg

Mc.'Cllco' 61gy c
M1skolc-Tapolco
Remcte Upper e 4

-37.

Cm

34.

,_

Cv

Mcgyefo IU

30.

Mg

Coel

Mamm
Mamm

Cocl

Mamm

BI
D1

SZELETA FAUNAL PHASE 17 caves + 38 open atr site= 55 sites (con.)

41>

Mg

47

J11SZbcreny
K11ilyk111

~8

Kl.skunfeleID'hli711

Cv

Sus

Mg

e:

Mg

Mllmm

ol9

l'fag,ymaros

Mg

I:.

Mimm

50

N6grild~cmcc

MITTllm

'ii

Puszl6

Mg
Mg

S;?

P6ka.w:pc!l::

S3

1'11Wl14r6t-Ra511harc

5"

R6ba. bed

"5

Saj6ktlzi nc

S6

38

ISigvar
S7.cgebTb
Swb

~9

1<;,,_0JnC1k Th

57

J6U
61

, 1>2
63

c,

Mg
Rg

Mumm

B1

Coe!
Coe!

Mnmm

Coe!

Mumm

Coe!

Mamm

Coe!

Mamm

Mnmm

A
I

BsB1

Mg

r.

Mg
C'

67
68
6!1

Vlic-AI~. Csfplc~s

11)

Vrn<.-n}

Bs81
flt

c~

Mamm

Mg

Bi

Mamm

Mamm

Mg

Bi

Mg

BI

Mg
Mg

Bt
81

Mg

BsBi

Memm

Mg

BI

Mmnm

c,

Cod

;\

Zalacgerszed n Bt

Ml!mn1

CoeI

Mmnm

Coel

Mg

l'b

Cv

T1szasuly Tb
Ti$zas:zoli Tb
Tl$T.lug1'b
111iz11ug Tb

Rg

Mg

f1S?affirtd Tb

!65

Mg

Mg

S7.olnol..-Pv Tb
ISzolnok-Zlgyv11 bed
T1S?Jldndu Tb

1'15lllkccske 10

[ 11

Mg
Mg

64

66

BI

Mg

Mamrn

Mg

B1

Mg

F.

Cotl

Mllmm

I:.

CoeI

F
F.

Cool

lSTALL6SKO FAUNAL PHASE 1S caves t- 52 open air sites = 70 sites

n IA!!lilc!_ck__

- - - -

73

Haroml.il~

7J

tsUJl!i>sko c V-1

c
Cp
c
Sus ,__ --

R.s
L

126' Jrinkovu:n c
'27

Kecskc.qgaJya c

r,i-s

K1skcvd> c 4/3
LO~vOll!.)'1 c

76

Ortll

Peqitle

78.

Pcsk6 c 3 S

I-- -

19

RO

I-

__
n Rs

82

Rl!metc Lower c

- -

R5.

Bodro1:ktrw.11ir

86
117

l3adDC!iOR)'
Dunal.!ldar

011nasukc.W
~-_ Esllcrgo1_n_ _

Cm

r---

c
c
c
c

Cv
Cv

Sus

Cp

Cm

-BsR1 -

A
,_

Rg

R~
81

I::

Rg

.B1
Bi

BsB1

Rg
Rg

Bi

Cm

R&

I:.

B1

E.

BsB1

c
c

-L

Cp

1-

Cp

.__ - ,_____

_...___

Coel

c.

Cvlm

Rg

"

Cv/m

Rg

Cm

- - - Rg

Bi
B1

lA
L_l~

--r
Rg
81

Coe)
Cod

c---

CoeI
CoeI
Coe I

Cocl

Mamm

Cocl

MAmm

F.

I J\'btmm
Coe I Mamm

I:.

Bo

81

\ltamm

,__

B1

Cocl

C1

SolymArqLlll.fl)'
Szelhn c 11
Tark6 R.< lHIJ V., VIL
Abau1winl0
_ _ _

Cv
Cv

Cv

Peten'tl c

Plh~S2.0n10

84
,_.

,___
Plhsi1.m10 I. Rs

81

83
r.13
D

-- -

Rg
Rg

Cv

Sus

'.24.. Gi'mimbOly !aJJOlca c

-, -iE

>-

Mumm

Coe!

MAmm

199

ISTALL6SKO FAVNAL PHASE 18 caves -r 52 open air sites= 70 sites (con.)

Ftlu:sabony gm\ quarry

II

9l
Fiizcsalxmy 5and quarry
;..92 Fokorulb

90

93

97

Gnlgahc\'I.'
Hauvun
JdszkaraJCnO
Kapusvnr
Kaposvnr Proad

98

Kole$d

99

KotclakTb

94

95
96.

c,

MBdlltllS B).

103

MAlyi-snwcl qW1rr)

104
105

Mitraba.Jla

108

Munor

109

Nag~~lOnv

110

Nagyk<lru Tb

,Ill

Nagymum~-UJvOIID

112

Nl!gymntOS-S1$1

llJ
114
l lS.

Padragkut
1c'3bAnya1eler1
Pllasmarot (+Marmo1n1
l'll1.smtlf61..()d
PolgAr

116
~

117
118

Man11n

'
Rg

I
Cv

Coe I
Mamm

Bi

Mwnm

Cod

C\

Sill

Mamm

Bi

/\

B1

c,

c
c

CueI
Cool

Bi
B1

Cv
A

81

Cm

I:.

MD.mm
Ma.mm

Mamm

B1

RG

81

Cocl

Rg

Rg

F.

MBmm

Rg

r:

M11JT1111

I Mnmm

Mumm

C\

Cv
A

M111nm

Mamm

B1

Cv

Rombany
S1rok

Bi

121

81

122

SZDJOI Tb
S.."l!lldehely

l'.!3

Srob-Klilvlirm

12~

126

Srodhge1
Szolnok-Sny Tb
Szolnok-Sk. Tb

127

Tis:mlOkR:ii.ofll

128
129

T1SZ11IOl.:-Dam Tb
TOs7.cticlilTTb

'

Cv
Cm

Ii

Coel

Coe!

Mttmm

Rg

Mllmtn

Mamm
Mamm

01
I

81

Coel

F
E

81

IJO

Vlil:-OCM
Vel'5C8

132

Visegrdtl-11

1.33
134

ZcbcgCn)'

c,

Zcbcgeny-IUlvlirla

Cm

135

Zire-CUhll Slmn

R.g

BJ

Mamm
tvfamm

Manun

Cv
E

R.g

Mamm

Bt

Cod

Bi

Mamt\1

PO..ISSZANT0-13AJ6T FAUNAL PHASES 22 caves;- 10 open air siLes = 32 sites

Cv

136, Als<>szmv;i t

Basu; c

Cm

137
119 I BnlJnc
f22 I BervavOljiyi Rs2

Cp
Cp

c
I

Sus

R.g

R&

Di

Rg

81

r
I

Cm

Cm

MAmm '
Mwnm

131

200

mm
-Ma--

Rak6czlfalva Tb

B1vt1k c I, l-4

Mn.mm

I:

BI

119

138

Mamm

Rg

.'vl:unm

Mamm
Mamm
M:imm

81

120

123

Coe I

Cv

MllrJ.:K=~b

106. Miskolc
107 Mifiolc-qull1T)

Cv

Lo\185

10:?

I
I

Bs01

100 Li\bal11111
.__
101

81

PlLISSZANTO-BAJOT FAUNAL PHASES 22 caves + 10 open air sites= 32 sites (con.)

!F.?3

'BUdb;pc>I c 7

nJ

bmll09.il ~ VI

11'26

15

Jankov1ch c II
K1mlykill c
K1skcvcly c ~

140

Korosc

141

L1 llafurcd Rs
6hu1a
Orfu-Sk
OJyvesl.o c 2. l

139

142
143

1-14

c,

Rg

Rg
Rg I
Cv

Sus

Rg

Rg

11..i

.c1

Remele LO\\.Cf c I I

146

Rcmcu:hecgy Rs
S.ll!lctu
Vo<npu ~ 2- 3.

Arb

L~ncrgum-0y ttrgy

ISO

1''nd.ap
Pillsmor01-Dios

Rg
Rg
Rg

Pillsmuro1-Pliln:I
T~rccil

Rg

j 1S4

1
E.
E

01
131

Rg

BI

Rg
-

Sus

S6gvi\1
ISS Suged-Othalom
I t56 I \'tic-AM Cslp~cJ

Rs

Bf

Rg

Rg

Ilg
I

Z.:bogi'n)-K~hilr1a

E
E

B
BsBt

Rg

lSZ

01
81
01

Rg
Rg

153

~7

Cm
Cv
Cv

149

Sus
Sus

Rg
l

l-=

81
81

l.

.'8.l

15 l

E
I

Pesl.6c 2
Pllma1110 I Rs UL
l 4 S Rcjtckl Rs1

148
,...._.

,7g

.....147

1so.

.J2

I Bi I

Rg

Bi

Rg

Rg

LJ

Abbf'l!\"IDliOllS

Hx I lystm.. I Leo, ( Crocoua. Sus. Cp Cnprcolus, Cv Cervu.~ claphus, Cm Ccm'Us "marul". A Akes, D Dama. Mg Megaloccros. Rg
Rnng.uer, 61 Rison, As Bo.:1, Bov Ilovidae, As Asinus, t! Equus, Coe.I Coclodonta. Mnmm Mumniutbus, 0 Ov1bos. Tb-T1.ru1 bed, Zb-Zagyvu
bed, By-Bricl--yard. c cav~, Rs rock-shelter, lJ Upper, L Low.:r

Table 11. Species frequency of the Upper Pleistocene faunal phases (180 sites with carnivore and herbivore
macro-mammals - number os sites)
SiltS
Fnum1J ph:lsc

S rcc 1 c:s

tolOI

Su6

Dm

Cp

Cc
3

Cm

1n1ly H d
Mg

VARB()

11

s
s

3
6

StmALYUK

'.!

4
8

S4

szr:.tTA

SS

!:!

IS

3
2

19

11

lSTALLOSKO

70

II

IJ

2J

27

0>\JOT

31

Tot11I

180

JS

49

26

18

66

ll

42 I 66

Pll.l~SZANTO-

Rg

As

SUTIO

Cl
2
5

Mm

8
35

II

'.!4

J.4

21

9
23
31

35

~4

40

23

19

IS

92

103

63

84

3
4

2
8

13

66

10

Abbnmallons:

I Leo, C Crocona, S Sus, Dm Uama., Cp C.apret'llus, Cc Ccrvus dnphm. Cm Ccn-us "marn.I", A Alces, Mg Megaloceros, Rg Raogifer. As
Asmus, B Dov1dae (Bos, Bison), !-, Equus, 0 Ov1bos, Cl Coclodonta. Mm Mnmmulhus

201

Table /II Species frequency of the Upper Pleistocene faunaJ phases according lo site types (180 sites with

carnivore and herbivore macro-mammals- number os sites)

S11es

Faunal phMC

Species

tolal
L

Sus

Ce

C11\<e site
VARBO

::!

3
J

Dm
2

Cp

SUTT6

II

3
9

Cave sue

II

'i

SlJBAL't'UK

9
6
J

6
4

CB\ C 1itc
Open air site

SZEl..ElA

SS

2
12

Mg

Rg

3
3

5
3

15

II

54

13

2
2

:!

19
10

17

II
I

15

ISTALLOSKO

70

II

ai.., ~1tr:

--

18

14
12

Optm air ~he

'i2

Pllll)Sl.ANT6BAJOT

l?

Cli\'t StlC

22

Open air sue

TOTAL

10

I
49

23

10

9
5

13

IA

66

Cl
2

7
7
8

s
s

2
2

8
5

9
6
3

23

Mm

JS

24

34

17

II

13

IS

IO

8 I 37
27

2
21

10
)I

:!O

I~

27
40

10

12

35
10

24

8
19

10

"

19

25

14

36

25

19

15

IS

10

II

10

92

103

6J

84

26

J
J
7

3
2

Open oir site

l8

I
I

1
8

Ca'c site

180

Cm

ana l ysed

42

66

66

10

Abbruvratl()m.
L Leo, C Crocorta, S Su.s Dm Oumu,. Cp Caprc:olus, Ce Cervus clapbus, Cm Ccrvus kmaral'', A Alce.>. Mg Mcgaloc!!l'os, Rg Rnng1fcr, A:>
Asmus, B Bovidac (Bos, Bison). E Equus, 0 Ovibos, Cl Coclodonia, Mm Mammuthus.

A SHORT CHARACTERlSATION OF THE FAUNAL Pl;iASES (Table IV, Fig. I.)

Most characteristic of the VARB6 fauna! phase is
the presence of Hystrix; this IS the main feature distinguishing it from the SVBALYUK level. Cervus dominates this faWlal phase (17,%); Equus ( 11 ,8%) and
Bovid ( 11 ,9%) are significant. Capreolus, Sus and
Coelodonta are present. Also appearing: Rm1gifer,
Mammuthus , Megaloceros, Alces and Asinus (Piovided there is no mixture in the layers and finds of
Csikvar /lJ and Lambrecht n .I caves.
SUBALYUK faunal phase, the species Cervus,
Equus as well as Megaloceros, Asi1111s, Bovid, Coelodonta and Mammutlws are represented evenly (between 9.5-10,8% each). Also present: Sus, Capreolus,

AIces and Rangifer.

SZELETA fauna! phase is dominated by Megaloceros (2 1,4%) with less Bovid (l3,9%) and Mammuthus (13,5%). Cervus (J0,0%), Equus (9, 1%) and
Coelodo11ta (9,5) are significant. About half this
quantity is present of Rangifer (5,2%) and Alces
(4,3%). Sus and Capreolus are present, Asinus miss-

ing.
!STALL6SK6 fauna! phase is dominated by and
Mammuthus ( 16,5%), followed by Bovid ( 14,5%),

202

Cerv11s (13,2%) and Equus (12,2%). Alces (11.2%)

Coelodonta (9,8) and Rangifer (8,6%) are significant. Sus and Capreolus arc present, Megaloceros

missing.
PILISSZANT6-BAJOT fauna! phase is dominated
by Rangifer (28,4%), Equus (21 ,6%), and less Bovid
( 17,0%). Ovibos appears. Missing from Lhe fauna
spectrum are: Alces, Coelodonta, Mammuthus.
In the Upper Pleistocene Wiirm period, 7 species
are occurring in all faunal phases (Sus, Lea, Crocotta,
C(Jpreolus, Cervus, Equus, Bovld). 2-2 species are
present in.five (Coelodonta, Rangife r) and/our (Mammulhus, Alces) fauna/ phases. Megaloceros is present
io three, Asinus in t\vo, Dama and Ovibos in one fau-

nal phase only.

The frequency of occurrence of the 15 species analysed in the six fauna! phases of the Upper Pleistocene
is demonstrated on Table TY., (Fig. I} .
The most striking features are the occurrences of
Dama, Asinus and Ovibos connected to special periods
as well as the chronological changes in the species
dominance of Cervidae. Cave lion and hyena were
present constantly. Ratio of mammoth increased con-

stantly till tJH:: Pilisszant6 fauna! phase (3,4-16,5%);

woolly rhinoceros remained fairly stable (8,75- 9,8%).
Apart from U1e two giwit herbivores, mammolh and
woolly rhinoceros. the two rop-predators, cave lion
and hyena, Uie frequency distribuuon of the remaining
11 species shows more differcnr1a1ed changes (Table
f!; Fig 2.) Cen111s, Bovtd and Equus were present in
nearly equal ( 15%) percentage chroughour the Upper
Pleistocene. Their slJght increase (with 5-8%) took
place in different periods: Cervus in Varb6 and
lstall6sk5. Bovid earlier in Szcleta and Equus la1er,
only in Lbc lstall6sko fauna! phase reached frequency
vaJues over 20%.
The rnre SllS can be rrnccd throughout the Upper
Pleistocene. lLS about l 0% frequency in the Varb6 Suba!yuk faunaJ phases decreased to one quarter by
the Sze!eta - lstilll6sko phases (cca. 2.5%), raising
again, surprisingly, by Lhe Pilisszant6 phase (6,5%).
Asinus is present only in the Varb6 (4,6%) and
Subalyuk (14.2%) founal phases. It is interesting that
according to lhe shape of the graphs (Fig. 1- 2),
Asinus advanced in the fauna nor "on expense of' s.I.
steppea11 elements Equus and I or Bovidae but
Megalocerv.~ and forcstal Cervus. Ovibos appeared
only in the fauna by Lhe end of the Late Upper Pleistocene.
Among the herbivore large mammals, seemingly
Cervidae are most sensitive for environrnentaJ
changes (Ti.1ble VI. Fig. 3.). Their frequency was
changing, or they left the area of the Carpathian Basin.
Dama occurred. according to our present state of
knowledge. only in the Slitto fauna! phase.
l11e rare roe deer (Capreolus) can be traced
throughout the whole Upper Pleistocene: its occurrence is highest at the beginning {Varb6, 2 l %) than,
decreasing gradually, reaching lowest value in the
middle phase (Szcleta, 2,0%) Similar to Sus. its frequency is raising again by che two upper phases
(lst3116sko. 4.7%, Pilisszint.O, 5,6%).
The abundant Cen1us is constnntly present. It
shows relative decrease (by 10-l 3%) only in the
Szelcta (23,8%) and Pilisszam6-BaJ6t (25%) phases.

On the graph, (Fig. 4., Table VT) the fonns C.

e/aphus (63) and C. "marar (6b) forms are represented separately. The occurrence of the laner variety.
if the dctennination of the species is adequate. 1s rais-

mg throughout the Upper Pleistoccne.

Alces appeared in lhe Varb6 phase (4,3%),
represented by I 0% in the Subalyuk and Szcleta
phases, reaching triple value in the Tst3116sko phase
(32,2%).
MeKaloceros appeared also in the Varb6 phase
(17,4%), than increasing intensively tripled by the
SzeJeta faunaJ phase (51 ,4% ). Occurrence of A lees is
low till Lhe maximum occurrence of Megaloceros m
the Sz:eleta phase. By the disappearance of Megaloceros, Alcru reached its maximum frequency (32,2%,
lstitll6sko phase). H 1s interesting to note Lhat among
lhe species studied, the ones which reached an
outstanding species dominance value in a given
faunal phase are missing from the fauna of I.he next
phase.
Rangifer is constantly present, significant from the
beginning of the Upper Pleistoccne (Varb6 13, 0%,
SubaJyuk 14,8%. Szcleta 12,4%). Its quantity doubled
in the lstall6sko phase (25,0%). Dy the end oft11c Late
Upper Pleistocene (Pilisszant6-Baj6L phase. 69.8%)
reached absolute dominance w1thm the Cervidae and
also between the Quaternary macro-mammals in general (Table l V. I~) .
The chronological distribution of Upper Ple1srocene species is demonstrated on Table 1'11. and Fig 4 ..
respectively. This figure can present well the characteristic specfos in the given fauna! phases as well as
their changes in dominance.
Summansing macro-mammalian species of the
Upper Pleistocene in Large scale founal phases is
only a first step in faunistical analysis. Differences
among the faunaJ phases s.how Lhe main directions
and lendcncies of global climatic changes. At the
same rime, they cover up chronological and regional differences within the fauna! phase. The
study of these factors is lhe task of further research.

203

100

"'

Ovibos

Ma mm.

80

60

20

Leo

Sb

P- B

Sz

Fig. !. S: Su'tto. V: Varb6, SB: Subalyuk. Sz: Szeleta, l: IstAllosM. P-B: PilisszanL6-llaj6t pbaunal phases
Mamm.: Mamrnuthus, Coel.: CoeJodonta. As.: Asinus, Rrulg.: Rangifer. Meg.: Megaloccrns. Ale.: Alces, Cp.: Capreolus, D.:
Dama, Ccrv: Cen~us

T(lb/e 1V. Species frequency of the ,LJpper Pleistoceoe faun al phases in percentages (180 sites with carnivore and
herb'1vore cnacro-mcpnmals l :>~ specm.>s
. )

Fauna! phase

Species

nnalysed

Su.~

Om

Cp

n.o

l3.0

2
R,7S

8.7:>

8,7$

S'.S

s.s

10
17,0

I
1,8

9
10,8

J,6

9,S

4,0

9,S

13
S,2

23
9,1

13,0

13.0

3
L3.0

CcJm

A.

Mg

Rg

As

Cl

Mm

SOTTO
Sites
%

2
&,75

VARBO
Slle$

s.o

s.s

6
7,0

6.0

6.0

J.6

12

IS

25

II

54

4,7

6.0

1,6

o;&

10,0

4,3

21,4

l2
13.2

27

21

31

JS

24

40

11.2

8,6

12,7

14,S

'>.R

16,S

19
21 ,6

15

17,0

2,3

4
6,8

3
S,0

:?
'.l,4

11,t

11 .8

8,5

3,4

10,8

9,5

8
9..5

8
9.5

35

24

l ~.9

9,5

34
13.5

SUBALYUK
Siu:s
%

SZ~LETA

S11es
%
ISTAL.LOSl<.0

14

5,8

1,6

1,6

25

4,S

8,0

5
S,7

l.3

10,2

28,4

Sni:s

II

4,5

P!LJsszMn'O BAl6T
Sites
Ofo

Abbreviations.
1- Leo, C Crocon.a, S Sus, Dm Damu, Cp Capreolus, Cc/m Ccrvus claphus / "maral", A Alce;s, Mg Mcgaloccros, Rg Ranglicr, As A3mus, B
Bovidae (Bos, Bison), E Equus, 0 Ovibos, Cl CoeJodonta. Mm Mnmmulhus

204

100

Ovibos

Bovidae
80

Equus
60

20

Sus.
0

Sb

Sz

P-B

Fig. 2. S S!ltto. VVarb6. SB Subnlyuk. S1 Szelcta, Ifstflll6sko.. PB: Pilis.w\nr6-13aj6t pbounal phases
Mamm Mammulhus. Coe!.. Coclodonta. As.: Asinu:.. Rang: Rangifcr, Meg.. Megaloccros. Ale.: Akes. Cp.. Capreolus.

O. Dnma. Ccry: Ccrvus:

Table V. Species frequency ofthe Upper Pleistocne fauna! phases in percentages ( 180 sites with herbivore
macro-mammals, LL species)
Spec:its

FIWlal phase

Sus

Dm

Cp

Cc.m

Rg

Mg

~nalysed

18.7

~.

2
l:?,6

2
12.1>

18.7

3
18.7

J
18 7
I

2.3

9,0

6,8

4,6

lo!,2

7.3

1-1,2

14,2

11,.3

11.3

11.0

S,4

16.1

3
S,4

25

II

54

13

23

2.4

I,:?

IS.O

6,S

32.4

7,8

llJI

4
2,6

"

32

2,6

W,8

27
11,S

6.5

2.6

11,6

10

' 22.7

7
I

16.0

7
16,0

..

I
-

8
14,:?

SZELETA
Srw..

0
I

SUDALYL'K
Sm:s

S1tcs

Sues

VARRO

c.

/\$

SUTTO

35
20,9

1ST..\u.OSK0
Sites

...

21

31

35

13,6

20,2

217

2.S

19

IS

32,S

24.6

19.S

PILISSZANT6-BAJOT
Sties

...

2
2,6

II bbroVlat/qn.s,

Dm Dama, Cp Capreolus, Cc/m Ccrvus claphus / "meral", ,\ Alces, Mg Megnloccros, Rg Rang!fer, A~ Asinus, B B<1,tdae (Bos, Bison).~
Fquu;,. 0 Ovibos

205

100
~

Rang.
80

Meg .
60
\

'\
\
\
\
\

40

\
\

\
\

\
\

20

Cerv.e

\
\

Cp.

\
\

Sb

P-8

Sz

Ftg.J S: S0110, V Varbo. SB Subalyuk, Si: Szeleta, I: ls1flll6sk8, r-R: P11isszAn16-Baj6t phaunal phasl!S
Mwnm.: Mammulhus. Coel.: Coelodontn. As.: Asinus, Rang.. Rangifcr, Meg. Megaloccros, Ale.: Alces, Cp.: Capreolw;.
D.: Dama, Cerv: Cervus

Table VI. Cervidae species frequency of the Upper Pleistocene fauna I phases in percentages ( 180 sites)
Spee les

Faunnl phnse

Dm

Cp

Cc

11n11lyscd
Cm

Mg

R&

SUTTO
Sftus

1.

28,6

'.!S,6

42,8

5
21.8

39.2

4.J

4,3

17,-1

13,0

Sites

..

%
SZGLETA

II.I

!6,0

7,4

3
II ,I

29,6

14,S

19
18,0

6
S,8

10,4

54
51,4

11.-1

9
10,7

32,2

VAR.BO
Siles

,.

SUliALYUK

Sitts

~
lSTALl.0SK0
Sue)

2,0

PILISszA NT0-BAJ0T
Sites
"lo

23

4,7

27,4

5
13,8

II

27

25

11.2

69,4

Dm Dama. Cp C11preol11s. Ce Cervus elaphus, Cm Cervus "moral''. /\ Alces, Mg Megaloceros, Rg Rangifer

206

21
25,0

Abbre11101/ons:

'

13

I
5,6

P-B

Sz

s
s

Cr

Cp CV

Rg Cl

Mm Al

Mg As

Fig../. S: Sotto. V: Varb6. SB: Subalyuk. Sz: Szeleta. J: lstall6sk6,. P-B: Pilisszant6-Baj6t phaunal phases
S Sus, L Leo. C Crocoua. Cp Capreolus, Celm Cr:rvus claphus / 'maror, E Equus, B Bovidnc (Bos, Bison). Rg Rangiti.:r,
Cl Coelodonta, lV1m Mammuthus. A Alccs. Mg Mcgaloceros. As Asinus. Dm Dama. 0 Ovibos

Tabll! VJ] Chronological division of the Upper Pleis tocene species in percentages (180 sites. 15 species)
Fnunnl phase

Specie~

Sus

....._

sun'O

c .__ Cp
,_

CeJm

f--W

analysed
Cl

Rs

-~~ Mg j

~~

As _

.__

I -

S11e.s

VARBO
Suc-s

...

11 6

'3

3
6.2

:?

II I

34

3.1

I -

2.9

- ,__

2
32

,_

- -~

LO

.s

19.2

79

10-2

27.8

11 3

1.6

68

46

8.0

24

23

R
95

100

,61
_ 10 0

SU13AJ.YUK
Sites

.I

19.2

IS.8

102

166

10 3

98

78

60

128

23

35

I -

SZELrTA

S11cs
.____

12

IS 4

JI S

~--

---

ISTALLOSKO
S11es

>---- -

154

PJLISSZANr6-BAJQT

Sitts

II

IS

25

306

111

28 4

14

29.0 ........28..S

r- -

2S.O 34.0
.-----,____

32

JI

22.3

36 3

33 8

JS
)~

-- - -

--

13

24

11

54

!9_8

380

40..S

26:

81 8
~

21

24-

4-0

JU

38.0

47 6

-4

19

JS

2S

19:?

10 5

14 l

111

10 3

20 6

14 s

37.8

7! ,___
121
800
-

-- .....__

27

64 l

.,

_:__

100

TOlAL

f---

Sito

'?6

311

J9

Ill

88

92

1.

100

100

100

100

100

100

103

66

63

84

42

100

100

100

100

IOO

66
1011

10

'?

100

100

100

Abbrt:vlntmris.
S Sus. I Leo, C Crocoua, Cp Capnx;ilus. Ccim CCJ"\'U.~ elaphu.s I "marnl.., E Equus, B Bovidac (l'los, Bison), Rg Ransffcr. Cl Coelodunta,
Mm Mammuthu~, A Alces. Mg Megaloccro.s, As As_inus. Dm Dama, 0 OY1bo~

207

REFERENCES
BACSK.AY- KORDOS 1984

BEN DA 1929
CSALOGOV l rs J 936
DOBOSI-VOR6S 1979

EHIK 1916

GAAL 1928
GAAL 1929

p-Ar\L 1933b
GAAL 1935

E. Bncskay L. Kordos: F ontosabb sz6rvanyleletek a MA.FI gerincesgyujtemfoycbeu 9. Kozlemenyi, Major finds scattered fossils in the Palaeovenebrate Collection of the Hungarian GeologicaJ Institute /Communication
No.9/. MAFI Evi Jelentese 1982. evr61 / 1984/ 355- 361.
Benda, L.: A magyar AlfOld ostOrtenete. l1. Acta Sabarieruia 3. Kaposvar
1929. 352.
Csalogovils. J. : Oskori leletek DunafOldvaron. Tolna Vannegye Muhjab61 2.
1936. 7- 14.
V. Dobosi- L Voros: Data to an evaluation of tbe finds assamblage of the Palaeolithic Pain! Mine at Lovas. Fol.Arch. 30. 1979 7-26.
Eruk, Gy.: A Herman Otto-barJang asatasainak. faunisztikai eredmenye1. .Barlangkut. 4. Budapest 191624-29.
Gaal, 1. : A legtijabb monori mammut-lelet. P6tfiiz.et a Term tud. Kozl. 60.
Kotetehez 1928. Fiiz. 182.
I. Gaal: Diluvialis emlosmaradvanyok Bajot eddig ismeretlen barlangjabol.
Le~ resles mammiferes diluviennes de la caveme de Baj6t jusqu 'a present inconnue. A11n.i\1us.Nat Hung.26. 1929. 1- 21.
Gaal, L: A pezsmatulok M ngyaronmig diluvialis fOldjen. P6tfilz. a Tenn. Tud.
K.Ozl. 65. Kotetehez 189. Filz. l29-l3l.
Gaal, 1.: A banhldai Szelim-batlang asatasa. Potfilzet a Term. Tud. Kozl. 67.
Koterehez 1935. Ftig. 49-63.

Gaal, 1.: Cs1llngaszau sziimioisokat igazol6 fOldtani megfigyelesek. Term.

Tud. KozL73 koL S. Sz 1119. Ffiz. 1941 190-205.
Gaal, L: Ablmbidai Szelim-barlang 'hicnas. rcteg"-~. FOldt. Ktiz l. 73. 1943.
GAAL 1943b
430-448.
M. Gasparik: Late Pleistocene gastropod and vertebrate fauna from Tokod
GASPA.RJ.K 1993
tNE Transdanubia. Htmgary/.Fragtn. Min. et Palaeom. 16 Budapest 1993.
89-116.
Gn~l>a rik , M.: A tokodi edesvi2i meszk6 osszlethez kapcsol6d6 felso
GASPARl.K l998
pk isztoccn !cont gerinces fauna vizsgalmaaak t:1j eredmeny,ei. Eloadas kivonal
lll.. l. Magyar Oslenytani Vt\ndorgyi.l]es 1998. majus 8-9. Tata. 5.
V. Gilbori-Cslink: La Grotte Remete ..Felso" /Supeneu.re/ et le "Szeletien de
GABORI-CSANK 1983
Transdanibie". Acta Arch. Hung. 5 1983. 249-285 .
V. Gabori-Csank: La fankovicbien. Unc civilisation paleolith.ique en Hongrie.
GABORI-CSANK 1993
ERAUL 53. 1993 Liege. 198.
Halavat'S, Gy.: A Domahidi es Merki osemlos leJetek. Foldt. K ozl. 28. 1.898.
HA.LA.VATS 1898
207-209.
HELLEBRANDT ET AL l976 M. Hellebrnndb-L. Kordos -L. T6th: A Di6sgy0r-Tapolca-barlang asatasanak eredtnenei. Ergeb.nisse der Ausgrabungen in der D i6sgyor-TapolcaHohle. Hetman O.Ml'.1z.Evk. 15. 1976. Miskolc 7-36.
Hillebrand, J.: A Balla-barlangban 1911. evben vegzelt asatasok eredmfoyei .
J-irLLEBRAND 1912

GAAL 1941

A Magyarhoni :FoldL Kozl. 42. 1912. 763-772.

HILLEBRAND 1913
HlLLEBRA ND 1915
HILLEBRAND 1919
HILLEBRAND 1935
HlR 1990
JANOSSY 1953

Hillebrand, J.: A Kiskcvelyi barlang probaasatasa11ak eredmenyei. Barlangkutatas. 1. 1913. 20-21.

Hillebrand. J.: A baj6Li Jankovich-barlangban 1914. ei; 1915. evekben vegzett
1rutatasok eredmenyei. Barlangkutlltas. 3. 34. L915 . 129-14. 1.
Hillebrand, J.: Az l9 L7. evbei1 vegzatt asatasaim eredmenyei. Barlangkut. 7.
Budapest 1919 6-13 .
J. Hillebrand: Magyarorszag oskokora. Die altere Steinzeil Ungams.
Arc.Hu.og, 17. 1935. 42.
Hir, J.: Retegazonosit6 asatas a Pesko-barlangban. Layer identifying excavation in I.be Pesko Cave. Karszt es Barlang 1990. 2. Budapest, 97- l 00.
Janossy, D .: Ritlcibb emLosok /Sicista, Apodemus, Asinus/ a dorogi es a

1118.riaremetei kesoi pleisetoeenbol. FOldr. Kozl. 83 1953. 41 ~36.

208

JANOSSY 1955

D. Janossy: Die Vogel-und Saugctierreslc der spiltpleistoranen Schicten der

JANOSSY 1962

Htihlevoulsr.<i.116sko. ActaArchHung5, 3-4.1955 149-181

D. Janossy: VorUtufige Ergebnissc der Ausgrabungcr in der Felsnische R.ejtc'k
l. /Bi.lkkgebi:rge, gem. Repashutal K.arszt- es Barlangkut:uas. 3. 119611 1962

49-58.
JANOSSY 1963
JANOSSY 1964

JANOSSY 1975

J ANOSSY 1976
JANOSSY 1979
JANOSSY-KORDOS 1976
JANOSS'r-VOROS 1979

JANOSSY-VOROS 1985
JANOSSY ET AL 1973
JUGOVJCS- KRETZOI-

CS.ANK 1954

KADlC 1911

KADIC 1914
KADIC 1916

KADLC 1934
KADLC 1935
KADIC 1940

KADlC 1944
KADIC- KRETZOl 1927

K.ADIC-MOTTL 1938
KEZ 1934
KOCH 1900

KORl\11.0S I9 l2

Jnnossy, D.: Letzinterglaziale Vertebraten-Fauna aus der Kalman LambrechtJHihJe (Bilkk-Gcbirge. Nord-Ost Ungam). Acta ZooLHung. 9. 293- 331
D. Janossy: Ll!t.zlinterglaziale Vcrtebralcnfauna aus der Krumim-Lambrecht
lltihle (Bill<l<-Gcbirge, Nordost Ungam). Il. Acla Zool.Hung. 10. 1964
139-197.
Janossy, D.: TarcaJ-Citrombanya oskOkon lel6hely gerinces faunaja. fol Arch.
25. 1975 26.
D. Jaoossy: Die Felsnische TarkO. Bcschreibung der Fundstelle ond der Vertebratenfauna. KarsZt- cs Barlangl"Ul. 8. 1976 3- 106.
.16nossy, D. A rnagyarors?.agi pleisztocen i.agolasa gerinces faunak alapjan.
Budapest I 979. 207.
D. Janossy-L. Kordos: Pleistocene-Holocene Mollusc and Venebrate Fauna
of two Caves in Hungary. Ann. Jlist-nal. Mus. Nat. Hung. 68. 1976. 5 29.
D. Janossy-1. Voros: Grossaugcr - Streufunde aus dem Pleistozan Ungars.
Fragm. Min. Cl Pal. 9. 1979 21-26.
D. Janossy-L Voros: Revision der mitcelpleistozanea Sauget1erfauna des
"Felsdacb.es" Szuhogy-Csorbak() /Nordw1gamt. Ann. Hist.-nat. Mus. Nat.
Hung. 77. L985. 69-95.
D . Janossy- L. Kordos- E. Krolopp-Gy. Topal: The Porlyuk Cave of J6svaf'o. Karszt- es Barlangkutauis. 7.1197211973. 15-59.
L. Jugovics-M. Kretzo1- E. CsAnk: Fclsojegkori entlosmaradvanyok a
Badacsony bazaltkUpjar61. Restes de mammiferes du pleistocene supericur du
cone basaltiqe du Badacsony. MAFL Evi Jel. 1953. 1 /1954/ 89--95
0 . Kadtc: Die fossile Silugetierfauna der Umgebung des Balatonsees. A Balaton videkcnek fossil is emlos maradvanyai. Bal.Tud.Tan. Er.. 19l1. I kot L.r.
Palcont. Fi.igg. 4. K<>L 11. 1911. 1- 24.
0 Kaclic: Az 1913. evbcn vegzett barlangkutatasaim eredmenyei. Resuh.ate
meiner Hohlcnforscbungen im Jahre 19l3. Darlang.kutatas. 2.4. 191 4.
18:5-191, 217- 223.
Kadic. 0.: A Herman On6-barlang Hamor k.o7..seg hat3niban. Barlangk."llt. 4
Budapest 19 16 6-17.
Kadic, 0.: A jegkor emberc Magyarorszagon. Der Mensch zur Eiszeit in Ungaro. MKFI. Evk. 30. J 1934. 148.
Kadic, 0 .: A Jlesko-barlangban eddig vegzen asatasok eredmenyei. Barlangvilag 5. 3-4. 1935 49-54.
Kadic. 0 .: Cserepfalu videkenek barlangjai. Barlangkutatas. 16. 2. 1940.
141 - 274.
Kadic, 0.: Az eszaknyugati Blikk ba:rlangjai. Barlangkutatas. 17. 1. 1944.
1- Ll l.
0 .. Kadic-M. Kretzoi: Elozetcs jelentes a csakvan sziklaU.regben veg-hell
asarasokr61. Vorl!iufiger Bericht Uber die Ausgrabungen in der Csakvar
Hohlung. Barlangkuunas. 14--L5. 1-4, 1926-1927. 11927/ l- 19. 40-60.
0 . Kaclic-M. Mott! : FelsoUirkany vidcikenek barlangjai. Barlangkutalas. 6. I.
1938. 8--89.
Kez. A.: A Duna gyor-bodapesti szakaszanak lcia1akulasar61. foldr Ktizl 62
10-l2. 193-t.. 1-19.
Koch. A.: A magyar korona orszagai kovUlt gcrlnczesallat maradvanyai
rendszeres atnczete. Magy. Orv. Es Tenn. Vizsg 30. Vandorgy. MWlk.
Szabadka. 1900. Budapest 526-560.
Kormos, T.: A tataJ osk6kortelep. MKFl Evk. 20. I 1912. 3- 66.

209

KORMOS 1914a
KORMOS 1914b
KORMOS L915
KORMOS 1925
KORMOS-LAMBRECHT
1914
KRETZOI 1927
KRETZOI l 942a
KRETZOJ l 942b
KRETZOl l954a
KRETZ011964b
KRETZOI 1968a
KRETZ01- VARR01< 1957
KUBACSK.A l 92 7
MOTI'L 1936

MOTIL 1938a

MOTTL 1938b

MOTTL 1938c
MOTTL 1940
MOTTL 1941
MOTTL 1942a
MOTTL 1942b
MOTTL l944
MOTTL 1945

210

Kormos, T. : A Lillafil.redi szil<laiireg faumija. Barlangkutatas. 2.4. 1914.

202-203.
Kormos, T.: A remelehegyi sziklafillke es faunaja. ln.: Kom1os,T.-Lambrecl1l,
K: A remetelilegyi sz.iklafillke es postglacialis faw1aja. MK.f J. Evk. 22. 6.
1914. 351 - 365.
Kormos, L.: A pilisszantOi kofillke, retegej es faunilja. ln.: Kormos. l.: A
pilisszimt6i kofilllce. Tanulmanyok a postglacialis kor geol6giaja, osipara. es
faunaja korebol. MKfl. Evk. 23, 6. 19 LS. 310-43 J.
Kormos, T. : A sottol fom\sm6szk0-kompJe.x1s faunaja. Allattani Kozl. 22.
3-4.159- 175.
Kormos, T. -Lambrecl1t, K.: A baj6ti Oregko nagy barlangjanakfau.naja. Barlangkutatas. 2, 2. l 914. 77- 80.
Kretzoi, M.: Elozeres jelentes a Biidospest faunajar61. Kezirat 1927.
M. K.retzoi: Spelaeus-Fauna aus dem Mecsek-Gebirge ohne Hl:ihlenbiiren.
Foldt. Kozl. 72. 1942. 364.
M. Krel2.oi: Die Fauna der Mexfoo-HOhle bei Di6sgyor im BO.kk_g.ebirge /Ungarn/. Foldt. Kozl. 72. 1942. 267-268.
M. K.re12oi: Befejezo jelentes a Csakvari barlang oslenytani fe11arasar61. Rapport final des fouillc:s paleontologiques dans la Grotte de Csa.kviir. .M.Afl. Evi
Jel. 1952. evrol /'1954/ 37-69.
M. Kretzoi: Die Wirbclllerfauna des l'ravertin-Komplexes von Tata. In.: L.
Vertes ed.: Tata. Eine mittelpal!iolithische Travertin-Siedlung in Ungarn,
Arch..Rung. 43. 1964. 105-126.
M. Kretzoi: Etude .Paleontologiqe. In.: V. Gabori--Csank: La stalion du
paleolithiquemoyen d' Erd-Hongrie. Mon. Hise. Budapeslinensia 3. 1968. Budapest 59- 103,
Kretzoi-S. Varr6k: Holozane und Pleisto7iifle Fauna der Hoblensedimenl.
ln.: D. Janosy et al .: Forsehun_gen in det .Bivakhlihle. Ungam. Eisze1talter u,
Gegenwart8 t957. 21-25
Kubacsk.a, A.: A solycruiri sziklaOreg pleisztocen csontlelete. Barl.angv1Jag I.
1926 1-4. (L927) 17-26.
M. Mottl: A bervavolgyi-sz!klaureg allatvilaga, J...illon6s tekiotettel a hazai
magdalenienre. Die Fauna der Bervavolgyer H"Ohlung mit besonderer Berticksichtigung des ungarisc:ben Magdalen.ien . Foldt. Kozl. 46. l- 3. L936.
148--l57.
Mottl, M.: A Bervavoll:,ryi sziklailreg. 6s1enytani ered.menyek. In.: Kaclic, 0 .Mottl, M.: Fels0tarkany videkenek barlangjai. Barlangkutat:as 16. 1. 1938
25-34.
M . Mottl: Faunen, flora und Kultur des ungarischen Solutreen. Quarta.r 1.
1938. Berlin 36-54.
Motll, M.: A lerak6dasok allatvilaga. ln.: Barhicz, L. et al.: A cserepfalusi
Mussoiini-barlang /Subal)rukl. Geol.Hung. Ser. Palaeont 14. 1938. 207- 308.
Monl, M. : Jelentes az 1932-35 evi barlangkutarasr6l es az osgerinces-osztaly
mUkodeserol. MI<Fl. Evijelentese az 1933-35. evrol 4. 11940/ 1899- 1972.
Mottl, M.: Az interglacialisok es mterstadialisok a magyarorszagi emlosfauua
tiikreben. Ftiggelek a MKfI L941 . Evi jolentesenek. 194 l 5-42.
M. Mottl: Das Aurignacie-11 in Ungam . Quartiir 4. Freiburg l 942. 82-108.
M. Mettl: A..datok a hazai 6- es ujpleisztocen foly6temlSZOk emlosfow1ajahoz
Beitrage zur Saugetierfau111a der ungarischeo Alt- w1d Jungpleistozanen Flussterrassen. MKFL Evk. 36. 2. 19421-70.
Moltl. M.: 6slenytani eredmenyek. ln.: Kadic, 0 .; Az eszaknyugati Bilkk barLangjai. Barlangkutatas. 17 J. 1944. Poska 9-22, lstall6sko 41-50, Olyveski\

62- 67.
Mottl. M.: Jelentes az 1936/38. evi iisatasok eredmenyeirol es az osgerinces
osztaly mllk.Odeserol. MA.Ft Evi jelentesei az L93(r.18. cvekrol 4. Kor. / 1945/
1513- 1585.

RlHMER L935

R1hmer. L.: A Pecsi Jpecsbanyalelepi/ mammut Gcogr. Pann. 12. 1935. Pees

SAAD-GAAL 1934

Saad, A. -Gatil, I : Elozetcs jelent~s a dj6sgyori barlangban vegzett asatasokr61 Barlangvilag 4 2. 1934. 12 19.
Saad, A.- Gaal, L: A Di6sgyori-barlang felso-diluvi<ius koeszkbzei es
faunaja. Dolgozatok 11. L- 2. Szeged 1935. 57-69.
Sebos, K.: A K.in\lykuti zsomboly a BUkk-b.egysegben. Barlangvalag. 4. J.
1934. 8- 14.
Slancrik., I.: Szolnok megyel regeszeti adatok. Hild Vik.tor Jegyzctc1bol. I .
Oslenytani anyag. Szolnok L975 8 15.
Varr6k, S.: A7. 1950-53. evi bakooyi asatasok oslenytaru eredmenye1 MAH
Evi Jd. 1953.evr612. /l955/ 491- 502.
Vertes, L.: Nehany liJ oskokori Lel6hclyilnkrol. Fol.Arch. 6 1954 9- 21 .
L Yertes: Neuere Ausgrabungea und pal!lolithischc Funde in der Hohle von
Istall6sko. Acta. Arch. Hung. 5. 3-4. L9:55. I I l- 131.
L. Yertes: Ausgrabungen in der Peteoy1- und Pcsk6-Hohlc / Btlkk-Gcbirge/
Fol.Arch. 8. 1956. 3- 11.
Vcrtes, L.: Az oskokor cs az atmcneti kokor emlekei Magyarorsr..Ugon. A
Magyar Regeszct Kezikonyve l. Budapest 1965 385.
l. Verres. The Upper Palaeolithic Site on Mt. Henye at Bodrogkeresltur. Ac ta
Arch.Hung. 18. 1966 3-14 .
L. Vertes-M. Kretzoi- .K. Bertalan: Jungpl~istozanc Funde aus emer Fclsmsche bet Goromboly-Tapolca. Karszt- es Barlangkut. 4. / 19621 1965. 81-85
V<iros. I.: A Karpal-medenoc Elephantidae-lelecanyagAnak attekintesc
Egyetemi Szakdolgo=a1. Debrecen 1974. Kezirol
Vor0s, 1.: Magyarorszag fosszilis Elephantidai. l. Eszak-Magyarorszag Elcphantidac lcletei. Fol.H.ist.-nat.Mus. Ma.tr. 6. 1980. 13-49.
l Voros: The skulls of Malllll'lolh in Hungary Fragm. Min. et Palaeont. 10.
1981 97 106.
V<iros. I.: A Pilismar6t-Di6s felsopaleoht telep alla1csontlclctei Comm. Arch.
Hung. 1981. 22- 24 .
I Voros: Fauna! remains from the Gravcuian Reindeer hunter's campsite at
Sagviir. Fol. Arch. 33. 1982 43-71.
I. Voros: Animal bone finds from the -Pilismar6t-Palret/ l 982/ Upper PaJacolithic Site. Act.a. Arch. Hung. 35. 3--4. 1983. 299- 304.
l. Voros: Hunted mmnmals from lhe Aurignacian cave bear hunter's site in the
1st.3116sko Cave. Fol.Arch 35. 1984. 7- 31 .
1. Voros: Animal remams of the Pilisszanr6 Rock-Shelter II. In.: V. T.
Dobosi- l. Vorlis: Chronological revision oflhe P1lisszant6 - Rock-Shelter ll.
Fol. Arch. 37 1986. 32--42.
L Voros: Large mammalian fauna! changes during the Late Upper Ple1srocenc
and Early Holocene Times in lhe Carpathian Basin. In.: M . Pecsi ed.: Pleistoceoe environment in Hungary. 1987. Budapest 81- 101.
I. Voros: The mammal fauna of the Pilisszruu6 I Rockshelter. In.: V. T.
Dobosi- 1. Voros: The Pilisszant6 t. Rockshelter. Rrevis1on. Fol.Arch 38.
1987. 31- 50.
L. Voros: Middle Pleistocene mammalian fauna from the Solymar Cave.
fol.Arch. 39. 1988. 4L- 58.
I. Voros: Animal bone remains from the Nadap Upper Palaeolithic Settlement.
ln .: V. T. Dobosi-B. Jun~bert-A Rmger- 1. Voros: Palaeolithic Seulement
in Nadap. Fol.Arch. 39. 1988. 32- 39.
Voros. l.: Madaras-Teglaveto fetsopleisz1ocen em16s maradvanyai. Cumania
11 Kecskemel 1989. 29-43
I. Voros: Felso plcisztocen Equus achenheimensis Nobis 1971 koponya Pil1smar6t-Basaharcr6I. Eqtms achenheimens1s Nobis 1971 Sch::idel aus dem

156.

SAAD-GAAL 1935
SJ:.BOS 1934
STANCZIK 1975
V/\.RROK 1955
VERTES 1954

VERTES 1955

\' ERTES 1956

VERTES 1965
vERTES 1966
VFRTES- KRETZOlBERTALAN 1965
VOROS 1974

VOROS 1980
VOROS 1981b
VOROS 198lc
VOROS 1982
V6R0S 1983d

VOROS 1Q84

VOROS 1986b
VOROS l987b
VOROS 1987c
VOROS 1988a

VOROS I 988b
VOROS 1989
VOROS 1990a

211

ober~n

VOROS 1991b
VOROS !991c
VOROS 1994

VOR6S 1998

212

Pleistoziin von Pilismar6l-Basaharc. Dunai Regeszcu Kozl. 3. /1989/

1990. 5- 10.
I. Vt>ros: Large mammal remams from the Upper Pa1eolilbic site at Es'l\ergom-Gyurgyalag. Acta. Arch. Hung. 43. 1991. 261 -264.
L Voros: A versegi paleolit telep emlosfaunaja. Die Saugclierfauna der Pa1eolithsiedlung von Verseg. Stud.Com. 2L. 199l. Szent.endre 85- 96.
1. Voros: Bird and Mammal fau11al revision of Kiskevely Cave. Tn. V. T.
Dobosi-1. Voros: Material and ohronologicaJ revision oflheKiskevely Cave.
Fol.Arch. 43. 1994. 21-36.
V6ros, 1.: A Duna-kanyar baJ partjaoak pleisztocen nagyemlos lcletei. Kezirat

Loess- and PaJaeopedestratigraphkal Data on the Boclrogkeresztur-Henye

Upper Palaeolithic site

A.rpad Ringer
The MiddJe Danube Basm is one of the regions in
Europe where surface loess cover lli very important.
The North-Eastern Hungarian pa11s within the basin
are specific for their vanablc, generally shallow loesslike sediments which comprise, however, so-called
"'typical loess sequences" only rarely (PECSI 1993).
One of these most remarkable occurrences is located jus1 in the Bodrogkereszliir basin and the
Kopasz I Iii! at Tokaj, in the vicinity of the Henye PaJaeolil111c settlement.
Opposed to the rich and significant archaeological
and palaeontological evidence, the stratigraphical potentials of the site are .relatively modest.

According to the study trench of the Heoye hill excavated in 1982. the volcamc base rock is overlain by
loessy cover in a layer of some ten ems thickness. Tbc
division of the sediments, however. is clear and W'lifonn along the complete surface anaJysed. Starting
from the surface downwards. under the Holocene soi!
we can find a loess layer followed by a palaeo-pcdocomplcx divided into several parts, overlying the bedrock.
The typical and suitably 1hick layer sequence of lhc
sice was sampled from the Western wall of tnal trench
nr. 8 ( 1982). ln the following, the division of this section is presented (Fig. I , Table 1.)-

Table I Granulometrical characteristics of the layers at the BodrogkeresztUr- Henye Upper Palaeolithic site

l
H

Nr Of

soinple

C11C03

0.86
4 12

Oram ~itt distnbunon mm 0 gr 1t

11 ~.

o.oos
o' 002 .,, I 0.0010,005

0,02

O.Q:!O.Q5

3S.S3

0,65

20.Sa

7,56

6.77

18.81

0.21
0.21

20.1 1

8,)S

6,9.S

15.69

g. 1~

7,68

17,26

1
6,43 I
22.4
./\ Cla) I Loom I. Loess H: Sand
I

0,01-

0.0 1

22.6-1
34.59

I O.QS
0.1

0.1--0.2 0.2-0,s

10.70
8,8S

0,.31

9,15

A "la

J .,

L '

O.S<

II~.

0.IS

0,67

26,12

11 83

0.82

~8.40

lS.S8
22,64

35,53

I.JO

lS.19

22,itI

26~

0,3b

0..28

0.32

34,16

2194

30,59

10, 11

Fig. I . Sttatigrapbical sequence oftbe Henye-hill Upper

m
0,0 .....,....-r-,........"T"T"m.......,
0 ,1

Palaeolnhic site
l. Recent soil

0,2
2. Locs

0.5
3. Three-layered palaeopedocomple.x
0.7

0.8

+
+

+
..

..

t;\+ +
-t~ +
+

..

4. Volcanic bedrock

I. 0,0-0,25 m. Sbal!ow skeletal soil ofbrownish grey

colour with anthropogenc interferences. At the
usual depth of 0.2 m. the base level of ploughing
mixture connected with cultivation of the ground is
observable. No traces of deep ploughing were
found here (Table l , Sample 1)
2. 0,25-0.40 m Greyish yellow, slightly loamy sandy
loess (A 1==28,46%, H2=26,06%, L3=222.64<1fo).
Concretions of calcium-magnesium carbonate indicate that large, (upper) parts of the loess were
eroded prior to recent soil formation processes. The
habit of these concretions is probably due LO
secretion from large bunches of loess and not leachmg of the original Holocene (forcsl) soil (Table I,
Sample 2.)
3. 0,4-1 ,0 m. Palaeope-docomplex comprising several
members Only the first soil boriz.on was suitable
for sampling (Tablel, Sample 3.). This is a light
brownish grey soil, comprising the archaeological

I A .,agyag". c lay fraction

:? H ..homok". sand fracuoo
3 L ,.IOsz", locss fraction

213

culmral layer as welJ, formed on a loess comain:ing

more clay than. the recent so.il (A= 34,l6%) and Less
sand than its cover (H=l0,11%). Under this layer,
over the volcanic bedrock and partly in its fissures,

remains of a darker grey and a reddish brown soil

could be observed.
Jn the past few-years, a more or less concordant palaeopedostratigraphy was elaborated for North-Eastern Hungary in respec~ of the Upper Pleistocene
,,.,_

....

':

:i::

(i)o

:c a;

~
N

I\,)

SzO 1-3

-~ ............~............

Denekamp

.::::::.,.----==-= ~~-

(;)

;JZO

I\,)

(RINGER, l993). This pa1aeopcd0s1.ratigraphy is supported by archaeological and palaeontological correlation at several po.ints, paralleled also by the Pl.cistoce:ne cave seclimeotS of the Biikk Mts. (Fig. 2.) fn the
u1>per third of the palaeopedostratigraphical sequence
there is a palaeopedocomplex comprising three soil
horizons. These soils are marked accordingly: MS3
brown soil with pseudogley(?), MS2 grey forest] soil
and MSl steppe soil. This sequence was worked out ai
Sajoszentpeter-Margit kapu site.

----.

,,.,_
0

- -- ---

..

MS1
=-MS 4.:._

S1
SzS2

MS3

(.J

53

~
N

O>

84-5

U'I

O>

~
.....

O>
0

........

;' "'cte
....,, ,.l#ct.

r:1'
0

.........

-~

.....

U1

Cl>

~A~~~stoo
!.,,

00

0.01

m
3

:::J

.......

.... 1:
,,.,_
...............:!
0

com >

-3
111_

o.=
--: ID
al :::i

:tJ

""" ~

.......

I
I

S6
S7

I
t~'
1 I I
I

I I I l
I

--- --

. B rf6rup
- - -r -+
- _\_
...... .

CD

O>

"'f....Oct.

(JI

01

Cl>

:::.-:::.::....,.

--- ---------

M1

SI 1-3

Ma 1

..

X'

'<

' -

Fig. 2. Chronostra,tigraphi~I system ofNorthem 1lltnRarian

Upper P!elstoccne loess and cave sediments
Key of symbols.
Subaeric sedunents L.
Ma IMlil)'l clay ex1roction pt!S. soil I.
Ml .Miskolc, I Palacopedocomplex;
i\llA I + MA2 Mis.kolc~Avas I -Avas 2 palneosoil;
SM~ Saj6szentp6ter- Mar~lL kapu site, palaeosoil 4:
SM3 Sajosr.:ntpeter- Margit kapu sire, palat:osoil 3;
M'S2 MiskolcSOJ6s1.cmpc!ter, ptJlaeoson 2;
1'.IS I M1skok-Sa;6s2entpeter, palaeosoil I;
Sz.0 1,2, 3 Sz1nnabcsenyo-6nod, paJ'aeosoils 1-3.

214

Cave sc<ltmcnts 2.
Sl 1-3 cave soil ofSubalyuk 1-3 layers:
L soil complex of layers 4-5 of Lambrecht Kalman cave;
S6- S7 double cave soil complex of layers 10 and 12 of Suhalyuk
cave:
S4 double (grey) cave soil of the "pipe" in Subalyuk C<J,Ve (lay\lrs 8
w1d 10)
53 brown cave soil of the 'pipe" in Subnlyuk cave (layer 6)
SzS2 tlark grey cave soils oftbe ca\'CS Szekm1 (layertl u.cco:rding to
KADlC 1916 st.nuigrc1phical section) and Subalyuk Hayer 'I of
the upipe")
S l cave soil oflaycr 3 "pipe" in the Suha]yt1k cave and layer
5 in Szclcta cave (accordjng to KADJC 1916 straligrnphical section)

The palacopedocomplex contained a Middle Palaeolithic /Upper Palaeolithic. transitional industry, over
this, an Aungnaco-Gravetuan industry could be observed. The series is closed by a younger Upper
Palaeolithic horizon (RlNGER, 1993 73- 77, figs.
U-1.4).
This palacopedocomplex with llueemembers could
be paralleled with the interstadials Hengelo + ArcyStillfried B + Kessell in Western European young
Pleistocene chronostratigraphy {fig. 2.).
On the basis of evidence discussed rn other pans in
this monograph tC - l 4 date and archaeological analysis by Dobosi) as well as tJ1e soil sequence presented
above we can connect the culrural layer to the lop
member oflhe three-layer palacopedocomplex, around
the period ofStillfried B + Kcsselt lnterstadials.
The settlement must have been lying m the loess
forming the substratwn of soil, which turned to soil
only later. Tilis is demonstrated by the fossilisation of
the palaeontological material which could not have
been accomplished in a soil. Thus the age of the seWement can be placed witli highest certainty at the cooler

climaltc phase between the interstadtals Sullfried B t

Kessell.
The raw material type spectrum is also typical for
this period. Wh.ile on otlier Gravettiau sites. fels1tic
quartz porphyry is seldom occurring, especially typjcal Gravettian forms made of this material arc rare,
these occur in considerable amounts a1 rhe Henye-rull
industry. According to personal experiences of the
author, Gravette points and microgravcnes mnde of
felsitic quart:z porphyry were also found 111 the Saj6szentpeler aurignaco-gravettian collection. in die palaeosoil MS2 correlated with Stillfricd B (RlNGER
1993. fig. 15) in layer 2 of the Di6sgyor-Tapolca cave
paralleled witli this soil type on ~e basis ~f the ori~i
nal documentation of A. SAAD (GAAL-SAAD
1935), as well as the layers 4. and 5. of the Szeleta
cave(KADIC 1916).
This characteristic "Early Gravetlian" horizon can
be connected with, most probably, a a.arrow horizon of
time around the interstadials Stillfried B + Kesselt.
Probably Lbe Henye hill Palaeolithic settlement can be
bound to Uus phase.

REFERENCES
GAAL- SAAD 1935

KADJC 1916
PECSl 1982

PECSI 1993
PECSl-RINGER 1987
RINGER 1988
RlNGER 1989

RLNGER 1993

Gaal L- Saad A. di6sgyori barlang felso diluvialis koeszkozei es faw1aja

[Upper Diluvial stone tools and launa of the D16sgyor cave}- Dolgozatok 11.

56-69.
Kadic 0.: A Szeleta-barlang lcutatasanak eredmcnyei [Results of the investigations of the Szelela-cave]- M.Kir. F6ldr. lnL Evk., 23/4 155- 278.
Pecsi M.: Magyarorsuigj 1osz6k es losszerii Uledckek tipusai. [Types of Hungarian loesses and loess-like sediments]. M= l : 1250000, map published by the
Geographical Research lnstitute of the 1 lAS.
Pccsi M.: Negycdkor ~s 16szkutams [Quaternary period and loess studies].
Akademiai Kiad6, Budapest 1- 375.
Pecsi M. -Ringer A.: Barlangi kepzod.menyek- Magyarorszag tajfcildrajza
[Cave fonnations - regional geography ofHungaryl Akademiai Kiado, Budapest 118- 121.
Ringer A.: Possible correlations beLween loess and cave deposit stratigrapbies
for the Upper Pleistocene in Hungary - Paleogeogrnphy of Carpathian Regions. G.eogr.Res.lnst. of the HAS, Budapest 65-85.
Ringer A. The role of cave sites and lheir chronostratigraphy in the research
of the Palacolith1c of Hungary. Karsz1 es Barlang, special issue 5 l- 57.
R.mger A.: Eszakk.elet-magyarocszagi gcomorfo16giai szintck es regeszeti
adataik. Felso-plei.sztocen foly6tcraszok:, loszok es barlangi uledekek
kronosztrnligrafiai rendsz.ere. [T11e chronostratigraphica.I system or the g,eomorbological layers and their correspond.mg archaeological data of the UpperPlcisroccne cave and loess deposils and river terraces ofNorth-fatstern llungruy] Dissertation, manuscript l- 163.

215

'

'r/.

I
l"

llrli

al .,

II

"'.:II

l"tl

- ~

r ,.

'

'

,.

..

II

. r

.....
l'J

.t

--,.

I I

....

::T

,, .......
L
I

...

:I
....

HI

..,

..... I

'

I
I
~

.-

JI

"

t',
.,,

JI

'
I

[.~

q
r1

.,

~L

. .. JI.,

-.

.'-n.!
!..

.,.,
L

u!

:l

Stratigraphical and palaeottological investigation of the fossil soil comprising

Upper Palaeolithlc tools at BodrogkercsztUr- Henye
P. Sumegi-E. Rudncr- L Beszeda

l1tlroducrion
Between J988 and 1994, several sampling was
perfonned on the loess areas at the Kopasr-HiJl of
Tokaj and its environs for palaeontological. Quaternary gcologicaJ, s tratigraphictil and palaeoecological
invesligauons. Apart from artificial and natural outcrops (deep roads, cellar cuts, walls of brickyards,
slide walls, grape plantations). boreholes and excavated trenches were also made. Most of the latter were
deepened at the plateau Dereszla and Henye at BodrogkercsztW-, because there were no natural outcrops
available on either sites.
The send.age deepened at the l lenye-plateau nenr
the spot height opened a fossil soil layer. This soil
layer was dark brown, slightly carbonatic, containing
buml charcoal, bone morsels, small fragments of
quartzite and obsiclian. It was settled over a base of
Upper Sarmatian rbyolite effusive body in loessic
sediments. The development of the fossil soil was considerably varied. Tts thickness was about 2-4 cm depending on local geomorphological conditions. On its
surface, a recent soil layer was fom1ed, probably already in Lhc Rolocen~. Due to the position of the recent soil borizon, the fossil soil layer is intersected by
roots of the recent vegetauon. Al the same time, the
contemporary am! rece111 animal and vegetal effects
(bioturbation etc.) and human interference (felling of
the forests, ploughing, grazing, plant cultivation ecc.)
Ute recent and the fossil soil got complete ly mixed at
several places. Due to erosion of the recent soil layer,
fo::;i.il soil is located immediately under the presem
surface. ln spite of this, the recent and fossil soil layers could be clearly distinguished macroscopically in
the sections deepened till the base rock, both on the
plateau and the side of the hill.
lnvestJgations were aimed at the date aod condiuons of formation of the fossil soil and its lithological,
slratigl'aphical connections with other fossil soil sections opened at other locations around the Kopasz-bill
ofTokaj.
Experimental methods

In each 5 cm layer of lhe geological sections, grain

size composition was detennined by aerometric hydrometry (VENDEL 1959).
From the level ofth.c fosstl ~oil, some 5 kg material
was collected for Quaternary malacological analysis.
The studied sample proved to be sterile in respect of

Molluscs. Al the same time. it contained significam

amount of charcoal which was suitable after special
cleaning procedure for radiocarbon analyses. Charcoal
pieces had to be cleaned from recent roots which were
intersecting the fossil wood. Without cleaning these
young organic components would distort the measurement results. The measurements were rnade aiming at
datmg the fossil soil. Radiocarbon measurements were
performed by E. Hertelcndi and co-workers in the
.Light Element Isotope Laboratory of the Debrccen
Nuclear Research lnstit:ute of the l LAS according to
the methods described by HERTELENDT et al. 1989.
Results are given below:
Site

llodrogkeri!$4Ut-

Henye

Depth

o.s. 0,6

Age

26318+

I J6SJ:ars

1'c1PDB)

Laboratory

25,96

dcb-2SS5

c:odc

The measurements were: made m 1993. No calibrated dates are avaHable as yet, so the above dera
should be interpreted accordingly
Apart from the radiocarbon analysis of the charcoal, it was su.iLable also for aruhracotomical (xylotomical, hislologioal) analysis
On the basis of histological analysis of the charcoal
remains we can determine the species it onginatcs
from . Such studies can help in reconstrucling fonner
local vegetation.
After macroscopic and binocular stereo111icroscop1c
selection and taxonomical derennination, Ai\1RA Y
scanning electron microscopic analyses were mnde for
a more detailed analysis of the histological marks
(wood and phloem elements, thickening, resin canals)
on tangential, radial longitudinal and transversal fracture surfaces (SZ06R et al. 1992. RUDNER 1994).
TI1is analytical series aimed at reconstructing the former arboreal vegetation, its compos111on and do 1l 1he
best possible way by the help of h1stologicaJ marks.
Different uixonomic monographs were used
(GREGUSS 1972; SCHWElNGRUBER 1978. 1990)
as well as technical literature on arboreal analOrny.
Identifying 1>lanrs on the basis of histological marks
can be d1fficul:t, ometimes impossible on the species
level (BARTHOLfN 1979).
Charcoal was separated from 5 kg sediment by s ilting. Sieve aperture was 0,4 mm, the residuals were
217

dried at room temperature. This was necessary to

assure minimal fracture. Wet charcoal disintegrates
easily an by this we loose information. Small pieces
do not necessarily con.ta.in aU diagnostic elements and
also the relation oft.he pieces belonging to one twig or
more will be lost. After drying. the weight of the char
coal was measured and compared it to the total
amount of the soil sample from the fossi I soil 'This da
rum can give a hint on the density of former arborefil
vegetation coverage. According to the pedologieal-anthraootomfoal studies by CARCAfLLET and THlNON ( 1996), made in the French Alps, S g charcoal in
100 g sed.iroellt means not a closed forest, but a rather
open, forestal steppe vegetation with groups of trees.
AnaJysis re,sults

Rough rock sill (0.02-0,06 mm diameter grains)

fraction in the grain size composition of the fossil soi]
is about 40-50%. Fine rock sill (0.0-35%. At the same
time, clay fraction was also considerable, between
I.0-14%, denoting a more intensive weathering in
connect.ion with soil formation. In the loess layer basemenL of fossil soil sections around BodrogkeresztUr,
Tokaj and Tarcal, whic11 could be related to BodrogkeresztUr-Heoye plateau fossil soil level. clay contenr
never swpassed l 0%; on average, values of 5-6%
could be demonstrated. These values indicate that the
fossil soil of Henye plateau, thougb overlying rhyolite
rock basem.ent, was formed on loess. thus the fonnation of the fossil soH was preceded by that of a loess
layer which turned to sol I later. Soil ronnation took
place on a more humid and tnilder climate than the
formation of Joess (sedimentation and diogenesfa).
Result of radiocarbon analysis (from the 5 kg sample collected originally for mal.acological analysis,
proved sterile for Molluscs but containing great
amount of charc-0al instead) gave a date of263 J8::365
BP (deb-2555). Stereemicroscopic analysis of all
charcoal samples indicated pine wood (Fig. I .} which
could be assigned to the Larix-Picea group (STIEBER
1967, 1968). At the same time. scanning electron
microscopy could demonstrate the narrow annual
rings, the sharp, undulate boundaries among lhe rings,
the largelyLhick walled tracheids and the lack of I.Win
bordered pits, the bordered pits arranged in one row
and the presence of resin canals (GREGUSS 1972,
Schweingmber. 1978, 1990) indicate the presence of
sotne Picea species (Picea sp .. Figs. 2., 3., 4., 5., 6.
qnd 7.), Consiaerable part of tl1e samples showed
traces of burning. These pieces, however. were found
not in one concentration in the sectitm but dispersed
(features indicating forest-fire rather than fireplace
embedded in the sediments). The mass ofthe charcoal
calculated for l 00 g sediments ( l-2 wt %) showed 1ha1
the territory must have been populated not by a closed
218

Fig. I. Sleteomicrogruph ofpim:-wood charcoul from the

Bodrogkeresznlr fossil soil (Magnification: 62.Sx)

Fig. 2. Scanning electron micrograpb tangential framurc surl"atte of Pfcea :;p. charcual rrotn the Bodrogk~rci;ztur lb:;sil :;oil
(Magnification: 344x)

Fig. 3. Scannlng electron micrograph radial fracture surface of

P1cea sp. charcoal from the Bodro{!keresztirr fossil soil
(Magnificsalion: 344x)

rig. 4. Scannmg electron m1crogmph tangential and radial

fracture surfaces or l'tcea sp. chu.rooal from the
Bodrogk~rcszrur fossil soil (Magnification 605x)

Fig. 1 Scannmg electron micrograph radial fracture surface

of P1cea ~p. cho.rcoal from the Bodrogkcrci;ztiir fossil soil
(Magnificnuon I320x)

Fig. 5. Scanning electron micrograph radial rracturc surface

of 1'1cea sp. charcoal from the Bodrogkc:reszuir fossil soil
(Magnification 4lli)

pine forl!st but a sLeppean taiga or a Laig_a wilh sceppean elements in Lhe time of the formation of the fossil
soil. The recenl ecological demands and resistivity of
Picea type forests (NIKOLOV-HELMISARJ J 992) 1s
characterised by tolerating well long and cold winters
and able to live m shon, bul long day-Lit vegetation periods. Compared lo Larlt species, they require milder
climate but more cold than members of the Pmus genus. Picea type forests are typically. fonncd on
leached acidic podsol type soil. Currently, members of
the Picea genus are spread on the Nonhem hemisphere, in tempera le and cold cl 1matk zone; m
Europe, the plains of the Northern parts and the mountainous regions of flie central parts of the continent.
Picea species arc also spread in the Carpathians,
between the altitude 600 and 1900 meters; moreover, a
"Picea belt'' has also developed between 1000- 1500
meters aJlitude.

Summary

Fig 6. <kanning electron micrograph radial fracture 'urtacc

of P1cea sp chun:oal from the Bodrogkcn.:sztiir fossil soil
lMagmlica1ion: 360x)

Al the Henye plateau, a fosstl soil horizon was

opened (sondage by the authors) which was formed on
loess some 26.000-27.000 years ago and from a
vegetation dominated by arboreal plants of Picco type.
in a forestal steppe environment. These data sho\\
good agreement with former chronological data on the
Bodrogkereszrur-Henyc site (GEYH et al. 1969) and
anthracotomical invesagalions on charcoal (STIEBER
1968), supposing tl1e fom1ation of a lanx-P1C:ea type
taiga here 26.000-32.000 years ago (GXO- l 95:
28.700 + 3000 BP) Scanmng electron nm:roscopical
studi~ could further reline the compos1uon of the
,-egeiation described before.
The fossil soil opened at Hcnye-pluteau can be paralleled. from a lithostratigraphical point of view, by
the younger levels of the Mende Upper Soil complex

219

described by PECST (1975. 1993), lhe so-called MF L

horTu'.on (rndi.ocafuon dales of the ~pe locality: 29.800
T 600 BP, 27 200 + ]400 BF. 27 ,855 + 1589 BP). It
shouJd be noted, b0.wever, lhat the .steppeao {PECSl

suUs of Stieber (STIEBER 1968) on the Mende type

loealily, who found Charcoal originating from wood
belonging to the Picea~Larix group in the younge
le\1els of tb.e Mende Upper Soil Complex ( MF1).

1992) or chernozem. (PECSI et al., 1979) type soHde-

ChronologicaUy, the formation of the fossil soil level

at Heoye Plateau can be placed to a more temperate
aoa humid period ofWtinn2-3 interstadia.I or, .in other
oam.es, the interplenjglacial phase.

scribed for the Mende Upper Soil complex s11ould be

revised on U1e basis of Lbe arboreal vegetation opened
from it This view is supj1)orted by anthraco1omical re-

REFERENCES

BARTHOLfN 1979
T. Bartholin:Thc Picea-Larix problem, I.AWA Bulletin 1979/ l, 7-10.
CARCAlLLET-THlNON 1996 C. Carcaillel-M. Thinon :Pedorulthracological contribut1on to the s tudy of the
evolution of the upper lrc-eline in the Maurienne Valley (Nortll French Alps):
methodology and preliminary data. Rev. PaJaeobot Palyn. 9 'I , 399~ 16.
M. A. Geyh-F. Schweitzor-L. Vcrtes-f. C. Vogel:, /\ magyarorszAgi
GEYH et al l969
wurmi eljcgescdes uj kronol6giai adalai. Foldrajzi Kor.lem~nyck 18. 5-18.
GREGUSS 197'2.
P. Greguss: XyJotomy of Living Conifers. Akadeinia Kiad6, Budapest.
HERTELENDl et a1. l989
E. Hertelendi- E1. Csongor- L. Z8borszky- J. Molnar-J. Gal~M. Gyorffy-S . Nagy: Cownlug system for high prec ision C-14 dating. Radiocarbon
31, 399-408.
N.LK.OLOV- .HELMISAAlU
N. Nik.olov-H. Helmissari: Silvies of lbe circL1mpoJar boreaJ foresl tree spe1992
cies. Pages U-84. in: Shuga:n H. R., Leemans R., Bonan G .B ., (eds.) A system analysis of U\e globaJ boreal fotesL Cambridge Univer.sity Press, Cambridge, UK.
M . Pecsi: Lilhost:ratigrapbtcal subdivision o.f the loess sequences in Hungary.
PECSJ 1975
f6ldrajzi KOzlemenyek 23, 228-239.
P~csi M.: Negyedkor es loszk\itat;As. Akadetn.4ii Kiad6, Budapest
PECSI 1993
Rudner E. Z.: Felse ple1s.ztocen vegetaci6 rekonstr.uk.ci6ja Magyaro.tszagon
RUDNER l994
ibszeoelemzesek alapjan (Upper pleistoceoe vegetation reconstruction in
Hungary on the basis of anthracologicaJ investigations.) Thesis for Msc, Asvany- es Foldrani Tanszek, Kossuth Lajos U niversity, Debrecen.
F. H. Schweingruber: Microscopic \Vood Anatomy. Zurcher AG, CH-6301
SCHWEINGRUBER 1978
Zug.
F. H. Schweingruber: Anatomy of European j Woods.Zilrchcr AG , CH-630 I
SCHWEINGRUBER 1990
Zug.
Stieber J.: A magyarorszagi felsopleisztocen vegetaci6-tortenete az anthrakoSTIEBER 1967
l6miai eredmenyek tOkreben. (Vegetatfon history of the Upper Pleistoeene in
Hungary on the basis of anlhracolom1cal data.) Foldtani Kozlony 97. 306-3 17.
Stieber J.: Antiuakot6mia, kvarterkto11ol6gia es a hazai pleisztocen. vegetaci6.
STIEBER l968
(Athracotomy, quarterchrortology and the Hungarian pleistocene -vegetation)

Thesis for Acad. Doct. Depamnent of Planr Anatomy, Eorvos L6rand UniverVENDEL 1959
VERTES 1966

220

sity, Bud~pesl.
Vendcl M.: A kozeth.ararozas 1116dszerraoa. Akademia Kind6, Budapest.
L. Vertes: The Upper Palaeolithic Site on Mo11nt Henye at Bodro.gkereszt6r.
Acta Archl-iung.18, 3- l4.

Instrume ntal analysis I.:

The Carpathian sources of raw mate rial for o bsidia n tool-making.

(Neutron activation and fission track analyses on the Bodrogkeresznir-Henye Upper Palaeohlhic artefacts)

G. Bigazzi-K. T. Bir6-M. Oddone

AbstracJ
The characteristics of the Carpatluan obsidians have been analysed by fission track dating methoJ and by instrumental
ncut.ron acuvarion analysis. Cl1cmtcat analyses and gcolog.ical dating techniques together discriminate the sources of
tile TokaJ Mountains (Carpathian I. II) from other sources located in the Mediterranean and m Anatolia. Although part
of the original primary sources cannot be located today, the best quality glass prcferenually used by prehistonc man
for tool-making comes in al I probability from Eastern Slovakia. Prehistoric exploitation of the Tokaj obsidian sources
started m early rimes, and the Upper Palaeolithic site Bodrogkereszt.ilr-Henye had a remarkable role m this process.

Introductory notes
In the Mediterranean basin a11d adjacent regions,
obsidian was widely used by prehistoric men for roolmakmg. Thus the identification of natural sources of
obsidian artefacts provides archaeologisLS a peculiar
opportw1ity of de1ecting trade routes and cultural interactions. Two factors make obsidian an ideal matenal for provenance studies: ( l) this volcanic glass was
recognised in the form of lava flows, domes or as a
component of pyroclastic products in distinct volcanic
regions an<l (2) its physical and/or ohemtcal properties
appear distinctly characteristic of the place of origin.
For these reasons, provenance studies of obsidian artefacts are among the most popular archaeomctricaJ research subjeccs.
l\vo potential factors may fonn an obslacle in obsidian provenance studies: (1) surface morphological
changes (natural or produced by more recent Jmman
acuviues) may have made rnacccss1ble some ancient
sources and (2) p rehistoric man may have exhausted
some occurrences.
Among the several approaches used for characterisation and provenancing volcanic glass, chemical
analysis of the major element and/or trace e lements
appears the most popular technique for connecting artefacis with natural sources.
Since the routine applicauon of fission track dating
(FT), several authors have shown that age and trock
densities are effic1ent, source specific discriminants
of obsidian (SUZUKI, 1969; DURRANI et al., 1971;
ARIAS RAD1 et al., 1972: BIGAZZI and BONADONNA, l 973; WAGNER et al.. 1976). Application
in different geographical areas such as Japan
(SUZUKI, 1969). Europe (ARIAS RADl et al. 1972;
AR.LAS et al.. 1986; BIGAZZJ et al. l 990), Latia
America (MTLLER and WAGNER, 1981; BIGAZZI
et al. 1992) proved the potentials of this mechocl

More recent..ly BlGAZZl et ol. ( 1986, l 993aJ

showed that the combination of different techniques
based on iodcpendent parameters, such as the trace
element composition by instrumental neutron activation analysis (lNAA) and FT dating, is an cfficiem
diagnostic tool, especially in case of ambiguous
source identification.
The archaeometric signilicance obsidian greatly
contributed to stimulate scientists to srudy the potential natural sources and lo refine analytscal techniques. Therefore. t.>speclally in l11c last 20 years
proYenance studies of obsidian artefacts yielded solid
results in several sectors of earth sc1ences.
This work reports on tJ1e results of an interdisciplinary approach, combination of CNAA and FT dating.
applied to samples from Hungary and Slovakia in order to characterise the Carpathian obsidians and compare them '"ilh I.be potential natural sources located m
the Mediterranean and in Anatolia.

or

Potential obsitlian sources i11 the Jl1editerra11ean and

adjacent regiom

Since the early sixties several authors studied the

nansral porential sou~ce& of lhe obstdian prehistoric artefacts in tl1e Mediterranean and surrounding areas
(Fig. l). The data-set (location. stratigraphy, chemical
and physical properties) related co the Italian, the
Aegean and the Carpathian occurrences and the documentalion of Lheir prehistoric use have reached considerable resulcs. ln contrast, knowledge on the occurrences located in AnaLolia is still inadeq1.U1te, especially in Its western and eastern territories (KELLER
and SElFRIED, l990: QZDOGAN, 1996).
In t:hc following, the well known Mediterranean obsidians will be just menuoncd. Some details will be
221

ZEMP Lt N

TOKAJ

Fig. I. ~acion of volcanics bcarmg obs1dians tn the Mediterranean and adjacent regions. 1-4: Monte Arc1, Palmarala, Lipan,
Pantelleria; 5-7: Metos, Antiparos. Giali; 8: the Galntean massif; 9: the Central Anatollan volcanic district; I 0-16: Kars, San-

kamis, Pasinler, Erzurum, Erzlncan. Bing61; 17: Mus; 18: Lake Van :sources; 19: F~ 20 Gordcs. 21: Qlnllkkllle; 22: Kalabak
Full square reprcscuts the area shown in Fig. 2.

given for the less known Anatolian obsidians and for

the Carpathian obsidians, which are the subject of this
work.
TI IE MEDITERRANEAN SOURCES.

In the Italian sector, obsidian sources are located in

the Monce Arci volcanic complex, in Sardinia island,
in Pal.marola Island, in Lipari island and in Pantelleria
Island (COR.NAGGIA et al., 1962, 1963; CANN and
RENFREW, 1964). Ln lhe Aegean, classical names are
Milos (Adhamas and Demenegaki), Giali and Antiparos (RENFREW et al.. 1965).
THE ANATOLIAN SOURCES

Since I.be pioneering studies by RENEREW et al.

(1966. 1968), several aulhors contributed to document
location, characteristics and prehistoric use of the
Anaroliao obsidfans (WIUGHT and GORDUS, 1969;
FORNASERJ et al., 1977; BLACKMAN, 1984). An

222

exhaustive outline on the present status of knowledge

on Lhese obsidians is given in the recent book edited
by CAUVIN et al. ( L998). Nevertheless, several factors indicate that more detailed field work and analytical investigations are needed in this region to register
all existing sources or to detect the exhausted ones.
These factors are:
fragmentary volconological surveying,
inadequate informarion on exac1 locations and
varieties of nomenclatures adopted and topono-

maslic uncertainties which make it difficult to compare data by different authors.

Obsidian occurrences arc known from
areas:

severa I

Nonhern Anatolta. ln the GaJatean massif KJ::.L-

LER et al. (1976) recognised two hitherto unknown

obsidian sources, the Yaglar group and Sakaeli, in the
volcanic region between the towns of Bolu

~lfiocene

11nd Cankin (N and NE of Ankara). KELLER and

SEIFRlBD (1990) reported on an unknown source
called "Galata-X". Homogeneity of age (KELLER et
al .. 1996) with Yaglar and Sakaeli and geographical
location (soulh side of the Galatean massit), suggcsr
that Galata-X was within Lhe same volcanic complex.
Cen1ral Anatolia The main sources located in Cappadocia. in the Aksaray-Ncvsehir-Nigde triangle are
the best known in Turkey. These are the large caldera
along the Acigol-Nevseh1r road (Acigol obsidians)
and the GollO Dagi rhyolil11ic volcau() (<;1ntik obsidian, after the name of the town located a few kmS-SW
from Gollli Dagi). Recently, BIGAZZl cl al. (1993b)
published detailed geological maps for these two areas
and geochronological data by FT dating.
The Acigol obsidians have been strarigraphically
di:.linguished by ERC/\N et al. ( 1989) in two groups:
(I) Bogukoy obsidians - these are located on the
caldera wall (and then predate 11s collapse) N of the
Acigo1-Nevsehir road 7-8 km E of Acigol and 1- 2 km
SE of Bogazkoy. They include the "WTHD source"
(white tuIT north of Hotamisdag of KELLER and
SEtFRJED, L990) According to them, this occurrence
1:orresponds to the Acigol source described by RENFREW et al. (1966) and Fornaseri et al. (1977). Keller
and Sc:ifried's '"SE Bogaz:.kiJy obsidian" also belongs to
llus group.
(2) Taskes1ktepe obsidians - these are several obsidian tlows located in the domes inside the calderu:
the Kocadag (or Hotamisdag) dome, E of Karacaoren
village, and lhe GUney Dagi and Korudag domes, in
the wesl side of the caldera (BJGAZZI et al., l 993b)
The Cifllik obsidians (more properly ''Glillil Dagi''
obsidians) are several flows which outcrop I km N,
between I and 2 km S-SW and 4 km NW of K6milrcil
village and I km E ofBozkoy village.
Although not all the mapped sources have been analysed, the available data-sec on Ac1g.ol and Ciftlik obsidians seems to be rather adequate.
Other sources in Central Anatolia are located SW
of Catkoy village (few km N of Lhc Acig61 caldera), in
the Nenez1 Dagi dome (BATUM. 1975). E ofBekarlar
\'tilage. between Acigol and <;iftlik, and in the Hasan
Dagi volcano. SW of 1.he Ciftllk town.

Eastem A11ato/w

Large volcumc areas bear obsi-

dmn in this region. Classical somccs, introduced in the

l1Lcrarure .since the 60's (RENFREW et al., 1966.

1968; DIXON et al., 1968) and revisited by ERENTOZ and KETIN (1974) and FORNASERT et al.
(1977} arc Kars, 13ingol, Nemrut Dagi and, more reccnlly. Sankamis, J:rzmcan (Bas L979), Stiphan Dagi
and Ziyaret I Mcydan Dagi. Other sourcc:s recently
~urvcyed nre ErLurum, Pasinlcr and Mw; (ERCAN et
nl 1996: BIGAZZI el al.. 1997). A shon description

following an ideal arc from E to \\/. SW and E is given

below.
Kars - This name, introduced by Renfrew et al.
(1966), is used by Ercan cl al. ( 1996) for se"eral ob-

sidian beds located S of Digor village (40 km SE of

Kars) and N of Kagizman (50 km S-SE of Kars).
Other obsidian occurrences were recognised I 0 km
SW of Kar~. Relationship between Rcnfrew's and Ercan 's Kars obsidian is not clear.
Sarikamis - A volcanic field bearing several obsidian beds extends from - 15 km to - 35 km S-SE of
Sarikamis, along the Sarikamis-Karakurl road. A detailed map is now available (ERCAN et al.. 1996;
BIGAZZl et al., 1997). KFLLER and SEIFRIED
(J990) quote also an obsidian occurrenc;e W of Sankamis.
Rize - Obsidiao beds asi.ociated lo tuffs were recognised near the Tkizdere village, 30 km S of Ri1e
(TANER. 1977).
Pa:.inler - This name, introduced by ERCAN et al
( 1996), refers to a sequence of outcrops in tu fTs in volcanic cones, near the Tizgi village (NW of Pasinler).
In a similar stratigraphic sequence, obsidian beds were
recognised near Eri:urum and Erzincan:
Erzurum-AL Tabya Dagi and K1bletepe. near Tambura village, - 20 km SW of Er.curum and, very recently, - 20 km SE ofErzurum (ERCAN et al., 1996)
Erzincan At Degirmenlepe and Boztepe. few km
E of Erzincan (BAS, 1979).
Bingo!, Solhan - A large volcanic field beanng obsidians is locaced in lhe Bingol-Solhan-K.arliova triangle. Their knowledge and documentation is rather
poor: OZDOGAN (1996) and M. C. CAUVlN (personal communication) refer of frequent obsidian occurrences that they surveyed along the Bing<>l-Solhan
rood. No detailed geological map of the area LS avail able as yet.
Mus- These are new sources, located - 30 km NE
of Mus (near the Mercimekkale village) and NW (near
Yaygin village) of Mus.
Finally, the best known obsidian sources in Anatolia are located in the Quaternary volcano series
along the coast of Lake Van. Important obsidian beds
were documented by several authors in Nemrut Dag1,
Sllr>han Dag1 a.od Meydan Dagi (called also Ziyarcl
Dagi) volcanoes (R.ENFRl:.W et al. 1966, 196~;
DIXON et al l968: LNNOCENTI et al. 1976. 1980;
FORNASERI et al. 1977). Several maps arc available:
recently. new maps were prepared by RCAN et ::ii.
(1996) and BIGAZZI et al.. (I 997).
Western Anamliu. This is a rather unknown region
wttb regard to obsidtan occurrences. Ercan et nl.
( 1996) quote I 0- 15 cm thick obsidian beds m NC'ogcne deposits n11ar the Alanyurt trnin i>tation (7 km E
of Kiitabya) and thin beds intercalated with perlnes m

223

Foya (W. of lzmir). Another obsidian occurrence was

recognised recently E of Gordes village. TI1e material
of these three occurrences seems to be nol suitable for tool-making. Another potenlfal source was
found near Qanakka]e (p.c. by Mehmet Ozdogan.
Univ. Istanbul).

WAGNER~WEINER

(1987) quote

another source. Kalaba.k, located fow kms W of

Eskisehir.
T HE CARPATHlAl'< SOURCbS

TI1e prehistoric exploitation of Carpathian obsidian

sources is well documented. They are (or, more properly, we.re) located in a volcanic district covering a

relatively wide area (- 2000 k:m2) across the Hu.oga.cian-Slo\'akian border within the Tokaj-Prefov molUltain range. In Lbe Hungarian side, volcanic activity
started in the Late Badenian and continued through Lhe
whole Sannatian tiJJ Upp.er Miocene rimes. Sev.eral
eruptions .from different centres produce~ large
amounts of rhyodacite ash flow tuJT and a wide variety
of volcanic rocks of acid and intermediaLe compositi (GYARMAT1 1977, PERLAI-....ry 1972).
ln Lhe adjacent Slovakian area, the acid volcanism
lasted some million years, from Middle-Lene Badenian up to Middle-Late Sannatian t.imes. Availabl.e KAr ages are shown in Fig. 2.

Geologic al samples

Ar~ifacls

0
l

6 Km
I

12.2 t0.IO

HUNGARY
Fi& 2. Map showing the areal distribution of I.he chyolire bodies i11 the Toka.j-Prdov Mounlnins (Gynrmati. 1977; Vass cl al..
1978). Location oft.he obsidians (geological samples ru1dancfaro) analysed by !he.FT dating method mid by JNAA and published K-Ar ages (Ma) are also shown (Vass et al 1978: Pccskay. 1986). B.1-1. refers to borehole samplt!S. From Biga;a;i et al. 1990.

224

The Carpathian sources yielded relatively smaU

amount of volcanic glass useful for tool-making compared to other sources of I.he Mediterranean area. Still
they have great importance because these are the
unique natural sources of obsidian present Ln Central
Europe and. altogether in Mainland (Conunenml)
Europe. It means ~bat before regular sea-faring, this
was the only obsidian source region available for
EuropennPalacolilh.Jc communities (BJR6 1984).
The Carpatl1ian sources were mentioned. mainly
through archaeological evidence 111 the classical srudies by optical emission spectroscopy performed by
Renfrew and co-authors (RENFREW et al. 1966),
based on arcbaeologicaJ material (.bdeleny-BorsodDerekegyhaza). Later 011, provenance studies wore
supported by source collected evidence in the late
seventies, by a British team usmg neutron activaL1on analyses (WILLTAMS-NANDlUS 1977. WILLIA.\llS-THORJ>E 1984) as well as Hungarian research. using main component analyses by AAS (Biro
L981) and later EDS-~RF (BIR6- POZSGAf L984,
BIRO et a1. 1986, BIRO et al. 1988). At Lhc end of the
80's the obsidian sources of the region were surveyed
by a Hunganan- Slovak1an-llahan campaign, which
was planned to get new geochemical and geocbronolog1ca1 evidence on the obsidian sources of die
Mediterranean and the Anatolian sources. Results
were briefly reported in BIGAZZl e1 al. ( 1990). The
Carpathian obsidian sources can be basically divided
into two regional I chemical groups, in 1.he southern
part of I.he Tokaj Mountains (Mad, Erdobenye,
Tolcsva villages) and to the North of I.be Jlungarian I
Slovakian border (Vinicky. Maia Bara: in the vicinuy
of archaeolog1cal sites Ccjkov and Ka!ov). 1be primary separation or Carpathian I (Slovakian) and Carpathian ll (Hungarian) obsidian has been suggested by
0. Williams; B1R6 et aJ (1986) discovered further
sub-groups within U1e Hunganan material Most recently. new subgroups seem 10 emerge within the
Slovakian obsidian sources as weU (TOTH et al.
1999) The geochronological study confirmed the
~xistence of substantial d11ierenccs m age as well as
c hemical properties between the main groups Carpathian 1. and ll.
Lsually ob~idian occurs in association with rltyolnc
flows (rbyodacites). Two Ltulln rhyollte bodies e>.."tending S<>Uth of the Tclk.ibanya-Nagyb6zsva li11e and
Nonh of the Erdobeoye-Tolcsva line. rcspect1vcly,
were recognised in lhe TokaJ Mountains. Several
smaller rbyolite occurrences are scattered through the
region. szADECZKY (1886) described obsidian Outcrops still existing m his lime.
Due mosUy 10 human aclivnies since pretustonc
times, obsidia11 fragments are scattered everywhere in
the rhyolite areas. Bands of agricultural grow1d showing rich concentration of obsidian pebbles (size up to

5-1 0 cm) are located few hundred meters NW as wc[J

as SW of Tolcsva. S of Erdobenye and E of Mad at
Kakashegy.This source is closest to Bodrogkereszt:Ur.
at the same time, it is the richest and most accessible

source within Carpathian ll (Hungarian) obsidians.

Similar secondary geological occurrences were observed in Slovakia at Maia Bara and Vinicky. These
are 111rerpreted as remains of originaJ outcrops or obsidian horizons.
1n conclusion. the Carpathian obsidian source region experienced radical changes in lhc last ten thousand years: the original outcrops bearing raw material
useful for tool-m aking were heavily exploited Prehistoric exploitation conm"buled to a great extent to lbc
scarcity of large obsidian nodules. Already S1adeczky
concluded at the end of the last century that the good
quality glass of many prehistoric artefacts was not represen1ed on the existing outcrops and I.hat lhc sources
containing the best material for tool-making had been
exhausted by prehistoric men

Fission track dating ofl'o/canit g la.ts: 0 11 alternative

nietlrod for provenance studies ofnbsidia11 artefacts
Glas!. 1s among the most suitable materials for FT
dating: its importance has been recently reconfirmed,
because glass 1s the only datable pha..::e of many reph.ra
(WALTER, 1989). N> a matter orract, FT dating gave
a solid co11tribution to tephrochronological studil!S
(sec, for example, WESTGATE, 1989). .\1.oreover, although due to relatively low uraruum content of natural glass (usualJy between 2- 3 pPM up to 30 pPM)
low track densities can be accumulated in shon times,
large surfaces can be prepared for track counung in
obsidian. Therefore, FT dating can be applied to few
thousand year old samples under favourable conditions. Recenlly BIGAZZI el al. ( l 993b) have shov.'tl
the potentiality of FT dating of obsidian for reconstrncting cbronostm1igraphy in very young volcanic

areas.
Due 10 low stability of fission tracks in glass O\'Cr
genlogical times, the spontaneous Lracks arc u.<>ually
partially annealed. FT ages arc in Lllli case rumunum
ages (called "apparent ages"), unless a technique for
correcttng lherrnally lowered ages 1s applied STOR
ZER and WAGNER (1969) and STORZER and PO LPEAU ( 1973) proposed llie s1z.e correcuon method"
and the "plateau method", respectively.
The lirst technique is based on esumate or track
densiry loss by track size measurcmems. The spontaneous to Lnduced {assumed as 1.m<l1slurbed reference
tracks) track size ratio is related to the partial annealing amount of the spontaneous tracks and to t.be con-esponding track areal density decreasing.
225

The plateau method consists in ce-establishing by

laboratory therma l trcatmenls an identical etching efficiency of i;pontaneous and induced tracks.
Experimenlal evidence fodicates Lhat these techniques produce equjvalent results and that corrected
FT ages an glass are commonly reliable formation
ages (ARlAS et al.. 1981; NAESER et al.. 1981;
STORZBR and WAGNER. 1982; WESTGATE,
1989).
As mentioned above, I.be PT meLhod proved to be
an efficient tool for coaela:tion of prehistoric artefacts

wilh natural sources. Results from pluriennual application.of this tecbnique can be summarised as follows:
(J) usually artefa.cts are replicas of geeiogical samples from the sources where they or.iginated from.
(2) as artefacts experienced a different thermal history during the lasl fow thousand years, semetimes
they show larger spontane.ous aac.k partial annealing

amounts than. U1eir mother-rocks. Application of correction techniques producei. again Lhe foanation ages
and aJlows source identification
(3) few artefac(s (,_ 5%) suffered intense then.ual
events whicb erased pre-existing tracks. Track accumulation started again after FT clock reset: if this accj-

dent can be related with lhe human activity which produced the artefact. il.s :FT age corresponds to iLs use

and is caUed "archaeological age".

FT analysis of obsidians in the Carpatltian region
Tbe FT analysis of Carpathian obsidians was pcrfom1ed uiling the techniques described in BIGAZZI et
aL (1990); see thls paper for technical detai ls. Results
are shown in Tables 1 (geological samples) r:md 2 (artdacLS). Tue distribution of the size-corrected ages
shown in Fig. J, indicates that the FT ages of geological samples from the Slovakian (Carpathian 1.) sources
are systematically older (by - 5 Ma) than those of
samples from Lue Southern part of the Tokaj Mts. T his

e\lidence does not corroborate the geocb.ronologicaJ

framework inferred from K-Ar dating whic1) does not
delecl a s ubsra.ntial difference of age between the two
rhyolite body groups ln Slovakia and Hungary, respccLiwely (PECSK..l\Y et al. L986. VASS et aJ. 1978). Corrected Ff ages (Table l) appear generally older and
younger tbanK-Ar ages, suggesting a larger time span
for the volcanic activity of the region.

10
(./}

tLI
_J

D...
~
~
(./}

u.

et:
LU

8
7
6

5
I.

ID
~
::::i

T T T
T T T
5

10

15

20

AGE Ma

Fig. 3. D1s tnbulmn of corrected FT ages of the Carpathian obsidians. geological samples and ancfacts (.shaded), shown in Tnble.s
I and 2. T (Z) indicates samples c-0lle<?ted in the To~aj MQunrains. Age v1,dues tlisttibute tu two distinct peaks. Most artefacts
belong to Carpalhian 1 (Slovakian.) obsidmn peak. From Bigazzi et al. 1990.

226

Tahle I. Fission Track dating of Carpathian obsidian

p,

Sample

N,

PI

41

N1

D~'Dt

CF

App. Age

I Ag~{-!-~
1>13-079

7.~mplm h1/11

f'M ll6/I

2'.000

1,431/

821.000

2.351

5 63

0 81

0,69

10;2

f'M8f.~-1

:nooo

636

I 144.000

1,377

718

0 77

0.63

10 43

l b ~D

PM 36/1-?
1h 250 "C

27.400

1,075

900,000

l.091

H4

0 81

069

10 18

14 7o I '- 0 QO

25,100

l.11 S

549.000

998

544

15 6-1 !- 0.6'8

PM S61J.I

13,800

1,195

S1S,000

1,527

7 18

0.74

060

IOol

lh250 '(;

10,SOO

279,000

1.258

7 IS

PM 11()132
PM 86...t-1

14.800

l.OIJ7
1.27.S

463,000

l,238

5 44

0 77

Q63

10.85

28,400

667

978.000

1,096

5A4

0 79

067

9o9

14 Jo,.,_ 0.95

28,o!OO

1.233

l,f77

5.4"

IS. 15 +l-0 63'

26,900

633

602,000
l.l29,000

1,384

7 18

0,82

() 71

9 66

13.71 t/ 0.8'.!

7,1190

31>.S

430.000

648

563

0.63

635

I016L0]8

8,630

195

+44,000

'17:?

0 71

6.76

9 59

~.570

412

3~8.000

637

073

0 S8

S.38

9 lS

:?h.250 ''C

6.400

31)(l

212.000

865

563
563
563

077
082

PM Bti Ila-~

8,S90

411

478,000

720

s 63

l PMS6 15

7.700

214

3!>-4.000

144

5.63

PM &ti.16

6.300

439

383,000

6KO

S.63

7,001)

S4K

282.000

SOS

S.(13

9,280

:!42

376.000

402

544

lh250 c

PM 86:4-2
Tukaj '.11.f

-l'M 86:10-:?
-P:\186111
J>M 86 12

1112.SO c
l'M861J9

I ?2

I 13

IH3~

16 b4 .,_ 0.69
17 22 </- (l.96

110

u 71

0.77

062

644

w 38-1-0.77

0.86

0.79

654

8 27 .'-069

0.78

065

5.68

aao-1 o6s

8 25

10 18 -.-091

0.88

9 JS -0 56

ll.Sti
0 81

048 J

Table 2. Fission Track dating of obsidian artefacts from the Carpath.ian area

p,

'iAmpk

N,

PI

N1

11

D/01

CF

App Age

Age( I la}

H11drtigkuen1tir-HMye

33,600

1,513

34,900

3,082

0165

0 74

0.59

9011

16 70 +t- L.:50

3:?,700

I Ail

19,l(l()

1.167

0.16S

I 00

I7 40 +- I.SO

-c.,;km

4,400

111

8,200

91

0165

0.87

079

57

7 '.?I -t-1.90

l'\.186/SI

37.100

1,28:?

%4,000

1,384

on

1:?81\

16 71 + - ( 08

31,900

1.110

654.000

1.192

37,300

1,380

1.069,000

l,228

5.44
5 44
5 44

0 86

2h250 c

38,700

1,262

J,047,UOO

1, 196

36JOO

1.105

875,000

1,109

PM M

l0,500

1489

.t!J,000

1,351

244

4h 270 "('

:?3.~~

290

:?23,000

1,628

.__ .H 4

PM 869-1

4,2f.0

197

I 1.029,000

J,147

l'M 86'9-2~

)3,900

941

1,007,000

PM 86.'9'.?b

38,700

1,075

PM ~f>'93

H 700

1,247

2h.2SO "C

26.800

1.166

PM 869-l

6,900

.,

PM 86152

---

PM 1!6fS-l
f>M 86118

-A:aJ<J>
-

lh.33 t 1-0 68

II 66

IS4S-l-I 10

0 SS
O 8S

076
0 76

12 78

16 81-1-0 95

0.91

0 86

1388

II\ 14 I' J 02

086

068

10 78

IS85-L.U97

1582 +'-I 07

S.44

OZQ

0.09

1 38

l.ISS

S.63

0. 87

079

1165

l .OS9,000

1,218

5.63

0.87

0.69

l~.24

15.4'1 t-1-0.88

1,064.00fl

1,21

5A4

0.80

0.68

1059

IS 61) /- I.OJ

S66,000

l,031

54-1

371

5~

- --- -

501,000

Jlt.I

llSS.000

1.159

544

() 39

0.16

563
544

'

rareof

---

MddKako.1 hat>'

I""""'" ---

10

28

1.00

~ 16 30

2.30

14 75 I- 0. 86

1ssa

I
2.60

.,.,_ o .6l!

16.:io -1-

1 .~o

0 ()?86

oooso

227

Table 2 (con.}
p,

N,

PI

N1

41

Di/DJ

C.F

A pp, Age

PM 86.12aI

38.500

1,336

951.000

1.032

S.63

0.9 1

.0.86

13 98

PM So. 11- 1

38,CJOO

1.352

978,000

2,l.57

563

0,92

0.87

.13 75

2hl50"C

)4,500

1. 199

781),()0Q

819

0 77

063

10 48

lo.63 +! 1.35

31.200

1.086

1,037,tm
0111.0QO

563
)63

31.400

1. 133
74J

16.15 !stmbol
01.48
15.81 !sm1bol
o I 22
15.29 .o./ 0.63

1.306

563

IS.OS ' 0.b2

-

18

Qn,oon

338

5.44

--

l'M 86,Ji-3

PM86174

:n.ooo

955

114~.cxio

1,l9S

I)

~:?SO C
I PM Sfi, 11.5

lll,800

l,03 J

544.000

21.100

l.100

94),00tl

5,44

0.83

0.72

7-!7.

i lh::?snc

20,300

1,136

681,000

l.S68
1,343
1,707

5.44
S.44

8h :!50 "(:

16,500

715

Sll6,000

1.047

Sam11lc

Age1 ti\-. lm

'folt:.1\'IJ

-PM 86:17-2

2h

2~0 "C

--

The reason of this contradiction is not clear: IT

ages on glass usually agree with K-Ar ages on the
same rocks, excepted in some cases. It has to be
pointed 011t that a standard for FT dating of glass does
not exist yet; so calibration problems may be at least
partially responsible for the discrepancy. As the c-0rrelauon or obsidian arccfacrs wuh natural sources was
based on the comparison of homogeneous FT paramcrers, the quoted disagreement did not invalidate the

83

I)

5.44

8 75

00()62~-

72

5.44

O.tXl31__J

11. lS t-1 0.13

12.15+( Cl.49
I 11122 I j. 0.!14
9 97 11 O.J7

39 ""'"' 0.46

archaeometric meaning of the r\!sulls presented on

Tables 1 and 2. This can be summarised as follows:
( 1) Comparison of parameters referring tO Carpathian area sources with those available for Mediterranean and Anatolian obsidians (Table 3) confirms
reliability of the FT dating analysis for characterisation of potential sources of raw malerial. Ages and
track densiLies appear efficient tools for discrim1nauo11
(Table 3.)

Tahle 3 FT parameters of obsidians from the Mecliterraneru1 basin and adjacent regions
111Lcn>"1l

tcm2 t (x1113)

FT np_p. age {Ma)

Joi form. ngc {M:Q

ISl!Itlu:ua

22 2 6

36-1.6

2000 -5000

10-2Jn

LlJ.HLtJ

rf'almarolll

o.9 LS

L6

I
I
I
I

3,000-6,000

180-300

(2}, l_UJ)

So11rtt

Pt

Reference

\Italy

Llp:trl G11bcllontl
Pantellcna

0.0085

0 05-0 14

007--0.l~

40

280

li!!. lu.t.J

150-atiO

150-200

(2J

LS- 1.7

650-1400

2:>-60

(2)

0.030

50

100

(2)
(2)

Atgun
Milos
Gmli

I~

1\nUpnras

l.O 31

4 !i-52

2&,000-30,000

550-750

q7- ll'J.8

13 7-17.8

I 0,500-30 .SOO

80- 180

~J),

S.4-S.2

8>-102

5,5008.90!-'

65-85

(3).( L" J

1~ 1 8

'.!O 23
23

40,000
62,000

ISO

(4)

18

210

(51

0 11-'J IS

0.1 1 0.18

(4 )

0.075

25Cl-J50
IOll

130-160

80

(4)

00 1~020

0020

50

180

l4)

0.85-1.0

1.0-1.3

1,9()(}-2, 700

125-170

N)

17

Caq1tt1hian "'11

Zcmphn Rill
!1u11a1

w..

(\ 1~.J

1
:>l. Amuulta

Sakaeh
1Y~gl;u

(' An(JtQf11:r

Ac1gi\l Bogiizkoy
~l-T:iskes1ktepe(od

Acrgul-'l'a.skes!ktn:pc (.Yd t

<;il\ltk

228

I
I

Table 3 (con.)
FT app. age ('.\f)

Sourtt

IT ronn. gl' (MA)

P1 lt m-2J

Pl (cm-2] (:L!03)

Refr.rcncr

2.~

Knrs-01gur
Ker.;

36

---1

Jn

J,600

110

(61

40

K.000

160

(6)

Sariknous

:! 3-3 9

3.6-S.O

4 ,800-6,800

95 120

(6 1

Pusmter
Enurum
B1ngot

s sn

6. l--6.6

12,000 15,000

18> 210

(II)

5.1

69

9.000

100

0.74.0

46

t ,SOO-t t .200

7.S-170

(0)

16)

'.!,()

2700

SQ

Nemrut Dagt

0024

65

170

(6)

Suph~n

00b8

li5

16)

O.OC'--0.70

11.60 090

75
200-ISOO

l)S-185

(2), (6)

Mus

Dn31

~\1311 n.ig1

Nevertheless. the present status of knowledge of

the sources located in Anatolia does not allow to de:finnively exclude existence of an outcrop with c haracteristics recalling the Central European raw material
yet. Several occurrences from that region are not represented
Table 3. Presence of obsidian with Carpathian ages appear incompatible with geological evidence in Northern, Cen1ral and Eastern Anatolia. The
chronology of the volcanic rocks located in Western
Anatolia is still fuJJ of blanks: so, the mentioned interaction may be possible although highly unlikely. At
che same time, Carpathian obsidians are essentially
different from other ages in the MeditetTanean region.
(2) The artefacts excavated from loess at Cejkov
(samples PM 86/5) represent a homogeneous group
constituted by very clear glass pieces showing a high
track density and a rel~tively low annealing amount of
spontaneous tracks in relationship wilh their age. The
artefacts PM 86/9-2a and PM 86/9-2b also belong to
this group. BO- I. PM 8619- I , PM 86/9-3 and PM
86117-2 show very similar induced tracl.. density (proportional to U content), and originated probably from
the same source. The only difference is a higher annealing amount, which may have been produced due
to different environmental conditions in the last thousand years. TI1e unusually high annealing amount of
PM 86/9-1 Ui more consistenl with a recent thermal
event chan anomalous room temperature of lhc source.
Three artefacts, P 186/12a-1, PM 86/ 17-1 and PM
86118 (collected on a hill 500 m from Cejkov, and reported together with the artefacts from that cxcavaiio11
in Table 2), constirute a further homogeneous group
characterised by a slightly lower U-content and by unLISUally low annealing amount. PM 8, which snows 11
higher anneaHng amount, belongs probably to the
same group.
Quality of glass, age and U-content indicate that the
sources corrcspondJng to lhesc cwo groups of anefacts
are (more properly. were) located in the Slovakian
pans of the Tokaj-Pr~ov Mts.; their characteristics

E. A11atulla

16)

slightly diITer from those of the geological samples

shown in Table l.
(3) Two artefacts only, B0-2 and PM 86/9-4, arc
consistent witb a Southern Tokaj Mountains source.
The induced crack density only was detennined on the
Inst one, due to presence of a high concentration of
microliths which prevented a reliable identification of
the spontaneous tracks. This is a common feature for
most geological samples from the Carpathian ll (Hungarian) sources.
(4) Artefacts PM 86117-4 and PM 861175 differ
from the remaining artefacts of Table 2. Good quality
ofglass and U-content point to the Slovakian parts
of the Tokaj-Prefov Mts.; but they yielaed younger
ages.
(5) Two artefacts have very interesting analytical
dat.a. Sample PM 86/10-la shows a bimodal size distnoution of spontaneous tracks (Fig. 4a). Sucb a distribution 1s found very rarely (see. for example,
MILLER and WAGNER, 1981) and indicate~ that the
artefact experienced a strong thennaJ event during its
history. The accident produced a consistent annealing
amount of pre-existing tracks (the "small" tracks in
Fig. 4a). Tracks formed afterwards (the "large" tracks
in Fig. 4a) remained undisrnrbed and show normal
sizes after etching. Counting both types of tracks (Table 2.), it was possible to calculate a geological'' age
(fonnation age) and an "archaeological age" for the
artefact. The resulting "archaeologicaJ age" is Upper
Palaeolithic, fits well to absolute chronologicaJ dates
on the Bodrogkcresztiir Upper PalaeolitWc site. Tb is
piece, found at Mad-Kakashegy was both macroscopi
cally and analytically identified as coming from the
Curp41thian I (Slovakian) sources; the "archaeological
age" measured (0.0286 +!- 0.0050 Ma, i.e. 28.6 kyear
B.P.) allows us, with just a little bit of fiction, associate it with the people of the Bodrogkcresztlir-Henye
site who might have lost it while surveying for
obsidian (Carpathian 11) at theMad-Kakasbegy source
region.

229

80

a)
70
60

70

SPOHTMIEOUS

b)
Sf'ONlA!tEOUS

60

/ sMAl.l IRACl<S
Ill

x 50

so

...
...
a:

x
u 40
~
c:

(0

I
I

r-~

20

10

...

r-. ./

CD

'
''

'
I

L,

...

INDUCED TRACK S

a: JO
w

x
J
z

/ INDUCED
TRACKS

VI

lO

a:

SPONTANEOUS
"LARGE" lRACl<S

rJ

IU

CD

.~

.20

::>

10

2
'
6
8
TRACK S1ZE Hm

JO

l4

2
'
6
8
TRACK SIZE - pm

10

,,

Fig 4. The "anomalous" sponianeou!l track size disl'ributioil of artefact PM 86/10-1 ii (a). The undjsllJrbed induced trackS (shaded)
are also shown. The bimodality of die distribution indfoates that a relatively intense thermal event produced a strong partlal
annealing of pre-existing t+acks (the "small" tracks). Afimwards, the new tracks were stored normally (the "large" tracks). "L.arge"
track.density is proportional to the age of the lhetmal event, which is very probably com:latcd with the prehistoric use of the
artefact. "Small" track density 1dentified the geological source. i.e.. Carpathian I (Slovakian) obsidian. For comparison,
spontanoous and induced track size di~bution are atso sbown for a "nonnal" sample (PM 86!2-l. b).

The other artefact, PM 86117-3, reveals only very

few !large! tracks and the m easured age is Neolithic.
In this case the the1mal event, very probably related
with human activity, fully erased pre-eicisting
tracks.
The source of PM 86/10-la could be tbe same of the

second group established in (2), but the U-c0ntenL

points t0 the Vini~ky or to the Mata Bara obsidians
also. T11e provenance of PM 86/ 17-3 is more doubtful.
Besides the above ,potential sources, the same origin
might be hypothescd for PM 86/ 17-4 or PM 86/17 -5 as
well.

The results of the application of the Fr method to

Carpathian obsidians confirm that this technique is an
exceltent tool for provenance studies of prebisLOric
obsidian artefacts: potential sources of a large area including Europe and Near East appear fuUy cliscriminated. FT dating produced a self-consisteDt cJutonological framework although there ate problems of disagreement with K-Ar dating of other volcanites from
the region.
Sampling in the Carpathian area requires careful
selection. nue to dispersion of obsidian pieces everywhere during pr-ehistoric times, it is possible to pick up
botl1 geological samples and "imported" artefaclS at
the same locality. This was the oase ef samples PM
86/ lO-la and PM 86/12a-L (identified as artefacts
from the Carpathian I (Slovakian) source.s) that were
230

collected on well-known Hungarian obsidian source

regions, Mad-Kakashegy and Tolcsva environs.
Jn the present collection of samples, it appears that
prehistoric men mostly favoured the Carpathian 1
(Slovakian) obsidians for tool-making, very .likely for
the good quality of the glass (see also Fig. 3). Most artefacL~ colJected in the Tokaj Mountains, even the ones
witb poten.tial sources nearby, originated from the Carpcuhian I (Slovakian) s.ources.
Comparison of analytical data on attcfactS and geological samples confirms the conclus.ion deduced by
macroscopioal observations by Szadeczky {1886): part
of th.e ancient sources af obsidian are at present inaccessible, due lo morphological changes and I or intense exploitation dur:ing prehistoric limes.
Finally, data referring to artefacts B0-1 anc;I PM
86110-la appear very significant. B0-1., odginated in
the Carpathian I (Slovalcian) sources, was collected in
the BodrogkeresznirHenye Upp~r Palaeollibic site,
dated 28~0003,000 by C-14 method cvER:rES,
1966). Th.e "archaeological age" ofanefact PM 86/ l Ola, collected from Mad-Kakas hegy, although burdened with a relatively large experimental error, agrees
with the mentioned C-14 date. These data confirm importation of Carpathian I (Slovakiam) obsidian by the
inhabitants of Lhe Upper Palaeolithic site BodrogkeresztU.r-Henye as well as their prospecting aolivity on
the (inferior.quality) Hungaroiru1;0bsidiansources.

Application of lNAA to Carpathian obsidians

Samples from rhe same obsidian beds and most of
the artefacts studied by the FT method were analysed
by IN AA in the Radiochemistry Laboratory of the Department of General Chemistry of the University of
Pavia.
About 10 g (when available) from each sample
were ground, powdered (~ I 00 mesh) and homogenised Three randomly collected subsamplcs (from 0.1
up to l .O g in weight) from each sample investigated
were submitted to homogeneity tests (JAFFREZIC,
1976). The variance analysis of the results, assumed as
an evaluation of sample variability. showed that the
aliquots of - 150 g are quite homogeneous and representative of the investigated obsidians.
Since most computer y-ray data reduction codes
proved lo be very reliable especially when both samples and reference standards have similar a spectral
shape as weU as relative inteosi1ies, it is generally recommended to use a standard analogous in chemical
composition to the samples to be analysed. NIST
SJlM 278 "Obsidian Rocle" standard of the N"ational
Institute of Standards & Technology (NlST) was used
for quality control.
Rare earth elements (La, Ce, Nd, Sm. Eu, Gd, Tb,
Dy, Tm, Yb and Lu) and Sc, Fe, Rb, Cs, Ta. Th and U
were determined. Samples of 300 mg in weight were
irradiated with neutrons for 2 hours in the University
of Pavia Triga Mark II reactor at a thermal flux of approximately I x 10 12 n cm -2 s -1 . Induced activity
was measured by y-ray spectrometry with a Ge hYPer
pure detector coupled to a multichannel analysercomputer system, 2 days after the end of irradiation
for Dysprosium determination. Aflerwards. the samples were irradiated again for 25 h in the same cond itions. induced activity was measured 3, I 0, 20 and
60 days after the end of the second irradiation.
Nuclear data of investigated elements are reported in
Table4.

Table 4 Nuclear data of the investigated elements

Elemcnr

I lalf-lifc of radionuclide fonncd

1.6R d

Energy of measured
y-ruy. keV

LA

140La

Ce
Nd

l>l!Ce__

IJ7Nd

Sm

153Sm

I 96 d

JQ3

Ell

152 Eu

122y

)44.liWll

'

- - - --IS)Gd

Cd

33 d
II Id

1595,487
145

9 1,531

242 d

98

Th

160Tb

73d

1179

Dy
-no

16SDy

2 38 h

95

166Ho

30y

80,SIO

l'm

170Tm

129 d

84

Yb

169Yb

326d

198

Lu

177 l.u

Sc
f'e

46Sc
- 59Fc
-

----~---

6.75 d
83 9d

888.1120

4S.I d

1098,1292

1077

- -

66Rh

18.6 d

I:l4 Cs

207y

1'11

1&2Ta

IJS Id

Rb

?08

79S

- -

1221

Tb

233 Pa

2711

31 1

ll

239Np

l.J5d

1711

The NIST-SRM 278 "Obsidian Rock" (Table 5)

and aqueous solutions of the above mentioned elements were also irradiated as reference standa rds.

Table 5. Iron (%) and trace elemenls (gig} abundance

ofNlST-SRl\1278 "Obsidian Rock"
l:!lerncl)I

Th~\

work

P~VIOllS

La
Ce

346-1-04

31 9 l

65 I +r- 02

62 22

Nd

33.10 +0 02

331

Sm

s 14

sn

! 0.03

0.841
s 32

Eu

0&3 1 004

Od

$41 +1- 00S

Th

1.04-1-003

I 02

Dy

6 49 ...1-0.()4

llo

I 21 ...,_ 004

"11
1.213

Tm

US-'-f- 001

O.JS3

Yb

4.49T/-0.04

4.52

Lu

0 94 +/-0 01

0 712

Se

s 41 t l 0.03

s ll

Fe

I 49 +l- 0.06
128.9 t 80

I 412

Rb
Cs

541.,./-0 OJ

Ta

l JI ;./ 0.06

127 S2

S.52
i 21

Th

13 l H 0,Q9

12 42

4 56 .;,/- 0.04

4.582

lron and trace element contents of the analysed

Carpathian obsidians are shown in Tables 6- 8; indlvidual data are the average of three or more independent determinations. Precision of m ost element
contents resulted around 4%, excepted for Holmium,
Terbium and Lutetium (precision beLween LS % and
24 %) whose abundance is quite low. Accuracy may
be evaluared by comparing the determined and certified or previously published abundances for NISTSRM 278 "Obsfrlian Rock" as reported in Table 5; a
general agreement within experimental errors can be
observed.
In order to find similarities among th.e investigated
samples, a dtmension reduction was applied tu trace
e lement abundances, using pattern recognition procedures by computer processing with the help of the
BMDP statistical package (Engelman, L990). The
data-set of previously analysed obsidians from the
Mediterranean Basin and Anatolia was also i11cloded.

231

Table 6. Cron (%) and trace elements ( gig} abundance in obsidian rock sampfos from the Carpathian area
8612(.Ml)

8612(1\12)

86/l(MJ;)

86/J(J'\12)

3~.9

130

25.0

81i/4fMI
297

8614(1\12)

40.3

Ce

63.2

58.9

38.4

31.8

46.S

Nd

4S. I

4-1.4

Jo.6

45.2

50.2
47.2

Sm

5.12

l::.LI

M6

488
0.4 1

0.73
7.2.1

31:8
4.12
034
7.39

l.32

136

L47

1.27

8.80

9.'20

950

ICW

F.lcmcnl
Ln

4 41

34.2

7 20

4.31

0.87

0.39

11.91

6.73

Gtl

7.81

Tb

1. 18

7-36
L27

Dy
Ho

K.20

7.80

I.fit

1.40

154

1.54

1-75

1.61

Tm

0.60

0.51

0.54

OST

0.57

3.S2

3.46

Yb

2.72

2.73

2.78

l'.l.57
2.74

t Lu

0.3'1

0.35

0.37

('J.36

0 .46

0:45

Sc

3.71

3.32

2.96

2.87

3..5:5

3.22

Fe

0.95

O.&l

0.82

0.91

0 76

0.75

Rb

180

207

217

200

217

196

Cs

1036

1L40

9.58

9,')3

13.7!

I0.95

Tn

1.22

1.27

0.91

0.95

0.8.7

1.33

Th

20.0

18,J

11.&

113

16.6

17.S

11.9

12. J

69

5.5

12.5

12.7

Table 7 lron (%)and trace elements (,gig) abundance in obsidian rock samples from the Carpathian area
86/.1()

(Mn

86110
(M2l

La

612

55 I

c~

78.8

78.6

Element

s&to

86/ll (M)

8f\/l.2
(Ml )

86!ll

85114

86114

86115

(M21

(Ml}

(M2)

(Ml)

S3i6
70.1

52...2

50.6

37.4

37..5

33.6

J2.8

35.2

52_8

52.&

50.1

72.8

73.S

60.S

60.9

516

47.8

53.6

74.0

83.3

84.9

86/JS

(M2.\

86/l6
(Ml )

(M2)'

&6!19
(Ml)

86/19
(M2)

Nd

SS.8

58.2

48.6

~1.4

43 8

45.9

-l.3

44.4

32.5

35.6

50.JI

570

57 6

Sm

6. 16

5.98

S.82

4 10

4.J:?

380

J 50

HO

5.90

6 00

Ell

0 .51

0 51

6 70
047

O.SJ

4 18
0.'11

0.43

0.4 (

0.41

0.42

0.33

Oc53

059

6.10
0,56

Gd

8.71

It JO

'6 .ll

8.46

6.18

7.l l

69.J

6 90

6.&5

6.93

9.15

8.12

8.15

Tb

1.27

I J.2

1,27

I 08

1.22

1.21

111

119

1.20

1.31

Dy

8.20

7 80

I 08
7 10

8.50

6 80

755

7.80

7.80

8.16

8.84

8. 16

Ho

1.40

147

I 40

1.54

161

140'

I 47

780
I 47

I 2&
7 14

147

154

1.54

1 33

1.33

lm

0.60
l,55

0.58

0.59

0.59

0.58

0.52

052

0 39

0.49

0.52

o:s2

0 5'.?

052

3.Sl

3.98

2.13

3. 4?

3 08

3.01

2..59

1.93

H O

J.18

3 02.

3,00

0.47

0.J7

0.50

OJ7

0,4?

048

047

0.36

0 33

0 4()

0.4it

0.111

04.5

Sc

4.89

4 70

4.43

4.4 1

l.4.S

310

3.0J

4,34

5.03

5 14

l.lj()

1..38

093

1.15

0.88

' OSS

0 77

1.09

1 34

130

Rb
Cs

222

259

203

203

139

140

127

218

231

232

t4.'19

13.60

1515

1612

7.26

10 86

J0.52

1095

10.56

0.89

0.81

0.93

o,~1

Olt6-

0.73

0.87

0.97

094

Tb

24.8

25 1

31 8

179

l~.3

16.3

168

-p.7

2.48

25 l

6.3

69

7 .7

0 94
253
6.6

1001
09 1
17.9

8.26

T11

2.04
IS.04
0.93

0 80
127

2.98
0.83
ISO

3 IS

Fe

3.24
0 85

4-.4

J4

34

69

73

66

7 .3

Vb
Lu

2s,1
6.2

9.36
0,94

Tn order to perform a correct application of the procedure, prelim~ tests were appl.ied for checking
~vhethcr U1a internal and instrumenuil variabilities of
Lhe single chemical eJeme.nt<; were less than those observed between different isample.s. The latte r resulted
significantly larger lhan the former for all elemenrs.
The cluster analysis BM OP statistical procedure selecled .the chemical elements that displayed lower in~
232

1.29

strumenuil and internal/inter-sample variability ratios.

Ce, Nd, Eu, Gd, Tb, Ho~ Cs. Yb and Ta resulted the
most discriminant elements. The resulting dissimilarity dendr0gram is shown in Fig. 5. The cluster analysis
fu lly discriminate the investigated potential sources of
raw material located in Lbe Mediterranean and adjacent areas. The Carpathian I- TI obsidians appear
grouped in distinct c lusters and sub-clusters. AU a.rte-

Ol.li
PweUt11a1JT

,,..,.u..,.11
P,,,1d1....10

i'wolcllutl

J'MW2(MI)
P~HQ!(MI)

IWU6'4(1dl)
J'lofl&-l(M2)

PMw.l(M!)
pj.IJISl9..)

PMNr.{Ml)
l'Sd!Nll-1
PMW10..I
PM!!ltl?.l

PMW).I
~2

STR.
PMINl7_.
PMWl7.S
l'MWl7-J
Polmon>lo
Uj>or1~V

L>l"'IAA
L1pWtllll
1...,.or

j_, PMl<lll"'(Ml)
PMWll(M)
l'MWIO(MI)
PMl&IO(Ml)
PMWl~l l

""""'~'

~MW ll(MI )

PM161J9()1U)
~fWIJ(Ml)

l't-Wlt.1).12)

PM!<.'lf",J.il)
l'MWll(1'CI)

PMWl.l().C)
l'Mf/W'4
PMU/12(M:ll
~IYIJ(MI )

s.i..tlll
S.Wc.111
s.l.tt.1)

s........
GaBolul
Gcrok floWl
)llS
llQS

Nl6

'Ill
Nq.j

1m
NI
!>'Ill

t:Ol
NOJ

Niii
Nl)6

"''

IM

"rn
~

McloolA)
M10<(DJ
MoNt.Arcil
MoalArri-1
M,... Aml

Mono Mia

l
0

Dissunilarity (%)

100

Fig. 5. Dissunilariiy dendrogram produced by the BMDP (Engelman, 1990) data reduction statistical process using the Cc, Nd,
Eu, Gd, Tb, Ho, Cs, Yb and Ta contents. Italian, Aegean nnd Anatolian obsidians are s.hown also. NOl-N16: obsidians from
Central Anatolia (NO l-N03 and N07: Bogazk!Sy obsidians; N04-06, N08-NI I: Taskesiktcpe obsidians; Nl 2-N 13, N 16: <;1ftlik
obsidians; N 15: Nenezi Dagi). Cankiri Orta and Gercde Bolu: obsidians from Northern Anatolia.

233

Table 8. Iron (%) and trace elements (gig) abundance in obsidian artefacts from lhc Carpathian area
86193

8619-4
S90
623

t-:d

370
520
49 I

Srn

959

7.-17

6 21

Eu

131

Gd

699

121
7.59

Th
Dy

9 40

Ho

115

089
6 39
I 15
8 30
I 70

Trn

0.S7

Yb

3 SI
0.71

l!lcment

l.3
Ce

l.51

Lu

Sc
1-c
Rb

351
080

Cs

12-70

211

Ta

151

lh

19.4
12.1

SS I

136
8 lO
1.69
0.51
3.82
0.63
4.10
I 13
222
1230
I 42
274
64

86118

86. 18

86.rl7l

86"17-2

861173

36 0

399
51 9

43.0
49,6

42 8

52 l
J9 2

SJ 4
51 l
677

470

37.2

500

520

s 1(1

59,I
46 8
4 9J

O.S9
6.79
137

062

0.4-0
7.36

041
1 14

0 33
761

1.20

I J'.!
8 80

159

I 41

11t0

8.70

8.90

l 47
0 SS
3 62
0.46

IAO
058
3 74
047

3 67
0 84

3.87

1 71
0.56
3.S<l
078
3.32
078
21.l

491
6 75
087
7 03

129
8.30
I 71
053
J91
074
361
0.75

0.51
3R6
073
357
0.81

48 2

6.-1 9
0.79
691
141
8-20
170
0 54
4 29
0 77
365
0 8S

8.60
I 6Q

0 SS
433
091

(Ml)

(~U}

3-10

34.9

35 I

660

S4 -I

568

690
I 17
BIO

l 70
061
.387

231

232

9.31

1130
I 31

211
1090
110

II 00

I JS

037
J 34
111
239
10.11
128

22.6

20.4

21 ,J

18 7

12.7

13.5

l::?-S

IC> 7

I 21
216
11.7

333
0 .97
?31

facOi are correlated with the Carpalhian I (Slovakian)

obsidians, except PM 86/9-4. which is grouped to the
Carpathian rl (Hungarian) obsidians.
The Thorium/Uranium, Th/U. ratio (Fig. 6) appears
an excellent marker of lhe Carpathian l (Slovakian)
obsidians also. Th/U resulted 2.5 for all the studied
sources (3.5 for the Carpathian U (Hungarian) sample.~) . whereas is 2 for the Carpathian I (Slovakian)
obsidians. 1n Fig. 5 aJI the ancfacts are correlated with
these last sources, wilh the exception of PM 86/9-4.
Comparison hehvee11 INAA and FT
dating

Bolh FT dating as well as INAA indicate that the

main sources of the prehistoric artefacts originated in
lhe Carpathian area were the rhyolithic bodies distributed in Eastern Slovakia. Prehil.toric exploitation and
transportation of lhis raw material started ar least as
early as the Upper Palaeolithic. I
Potentiality of an interdisciplinary study, where
techniques based on different parameters are used, is
especially revealed by peculiar cases whel."e the two
methods corroborate eacll otbcr or yield different information. For examples, fNAA identified the Carpathian 1. sources as provenance area of all artefacts except one. based on cbemicaJ affinity. Discrimination
berween the different occurrences located in that area
1

STR rw

86/10-111

More ancient use of obsidian

LS

115
7 81
I.SS
0.65
4.00

OA3
380

091
248
13.33
153

201
144

384

7.83
154

0.65
3.9~

047
3.53
08.I

86.IS(IM) 86i5C!M)

- 38,9

200

232

13 19
I 6o
194

12 29
193
193
133

14.2

8(>' 17-~

RCI 175

35 u
55 0

26,7

JH

42 8

47 3

456

39.2

8.56

601

H-1
631

032

0.58
6 71

6.91

6.80
I 17

0.89
230
IS 79
I 89
18J
169

0 S:?

l ()Q
057
357
0 77
-

357
Q 86

189

:!JO
11 SO
121
:1 3

t:?.I

11 7

II 30
1.06

un.d attribution of individual a1tefacts to peculiar

sources remained open.
On the conirary, FT dating identified at least three
groups possibly corresponding to different occur
rences, and found lhese clusters not fully correlatable
with the cxiSting remains of original outcrop . However, the identification of two samples (PM 86117-4
and PM 86/17-5) remamed uncertain. Confirma11on
given by INAA resulted substantial for a definili\ cattnbution to Carpathian I (Slovakian) sources.
Moreover, some artefacts may lu1ve experienced intense thermal events which total ly (or partially) erased
pre-existing lission tracks (two examples are given in
this work). Tn lhis case. source identification may be
doubtful because one of the parameters, the formation
age of the glass, was cancelled. Nevertheless. information extracted from lhe tool is also significant, determining the age of its use Ln such cases. chemical
composition may be essential for Lhe identification of
provenance.
General con clusiom;

Interdisciplinary research by INAA and Fi daung

on provenance of prehistoric obsidian artefacts rn ltnly
and in Anatolia allows to contribute to trace lhe southern border of lhe diffusion area of the Carpathian
sources. Some hundred artefacts from several sites

documented in Subalyuk cave (Middle Palaeolithic) and the curly Upper Palaeol11h1c as-

semblages of Szclcta cave and Istall6sko cave, also tested by EDS-XR'F (Bir6 I984, Biro eL al. 1986, 19.88)
234

40

Tok aj

Zemplin
o Artif acts

1 Th/U=3 . 5

I
I

11. M

1Th/U=2

30

9'-

I
/16M1

19M2~0
W 16M2
12M1

Th

10M1

ppm

l0M2
I 9M1

20

12 M 2 I

I 17. 5

/~7
.1
10.la

l~o
11. 2M1

o 18 M 2
O 5.1

14M2 .,=--14M1 !:M 2 1.M

15M2 I
/
el.Ml
15 M 1 I
I
11. 4
/
I STR.
I
I
/ JM 1

17.2

I 3M2

05.2

e e

10

18M1

I I
I I
I I
I I

1/
I

10

20
U-ppm

Fig. 6. The obsidians collected in the TokaJ Moumams (Carpathian I (Slovakian) and Carpathian II (Hungarian) sources) from the
hcdli interpreted as remains of origmal occurrences (full circles) appear fully discnminated in a Th-U diagram. All artefacts (open
ctrclcs). excepted PM 86/9-4, are attribured to Carpathian I (Slovakian) obsidians. STR: see explanation to Table 8.
Tbe prefix PM 86 has been omitted m this figure.

235

dated from Early Neolithic op to Bronze Age from Ttaly were analysed since l972; no sowce ou!side Italy
was identified as yet. The only case of Carpathian raw
material was reported by WlLLLAMS-THORPE et al.
(1979) referring to an artefact from Grotta della Tartaruga site (recent studies asSign this site to MesoUthic, but l do not have direct information and I do not
know whether there are also Neolithic levels and to
those levels obs.idian is associated, therefure it is better
LO om1L Neolithic), located in North-Eastern ftaly, near
Trieste. Two further artefacrs from the sa.tne site were
more recently analysed and attributed to Palmarola
and Lipari sources, respectively (BlGAZZ1 et al.
1986). Carpathian artefacts reached lhe Adriatic sea
eost seemingly as an isolated exception.
A similar phenomenon was observed in the supply
zone of the Aegean (Melian) obsidian in Thessaly. At
the Neolithic site Mandato, N Greece, occurrence of
Carpathian J obsidian was reported (BASS1AKOS et
al. 1993 , KlLlKOGLOU et al. 1984). Recently

BIGAZZI et al. (l 993a) published several analyses on

artefacts from four Neolithic sices (Domali. Fikirtepe,
Pendik and Uipinar) located in Norlh-westem Anatolia, in the Istanbul area. None of these samples rcsillted made by a non-Anatolian glass. Evidence collecced so far seem to indicate that Carpathian obsidian
did not cross the Bosporus. On the con Lrary, C:ifllik
obsidian (Central Anatolia) penetrated in South-Eastern Europe (ASPINALL et al. 1972).

ACKNOWLEDGEMENTS

We are grateful Lo P. Marton, L. Kaminska. T. Sasvari and P. Szabo: lheir help was esscnrinl for a successful joint sampling campaign and for interpretation
of data. L. Banesz, Y. De Michele. V. Dobosi and E.
Matyas kindly provided samples. Finally, a special
mention is dedicated LO the memory of L. Roz.loznik:,
who coordinated field work in Slovakia.

Explanations to Tables

Table 1 and Table 2: pa(p1): spontaneous (induced)

track areal density (cm-2);.Na(N1): spontaneous (induced) track.counted; cJ:>: neutronfluence (x 1015 cm-2),
referred to the SRM 962a NIST glass standard; Da/D1:
spontaneous to induced track. size ratio; C.F.: corre
spending correction factor of the apparent age obtained by a correction curve drawn for the standard
etching conditions used in this work (120 .s in 20% HF
at 40 C); App. Age: apparent age (Ma); Age: size
corrected plateau age (Mn); 2 h 250 C: heating step
used for plateau age detem1ination.
Paramecers used for age calculation: cr=S.802 x 1022 crn2; A:=l.55125 x io-10 a-1; '1.p7.03 x 10-11 a-1 .

The reported experimental erroJ includes Poisson

counting errors and, for the size-corrected ages, the error introduced by the correction itc;elf (See Bigazzi et
al., 1990, 1993b for further technical derails).

Table 3. FT form. age: the size corrected or plateau

age, assumed liere as formation age, doesn't refer to
rhe full range of apparent ages for some Eastern Anatolian sources. For example, the plateau age reported
for Bingo) was deteanined for the occurrence whose
parameters (app. Age and track densities) are the upper limit of the range. The apparent age has not been
reported for obsidians Lhat did not show spontaneous
track partial annealing. (u.r.): unpublish_e d results;
(t.w.): this work; (1.): after Biga1..zi et al. 1988; (2.)r after Arias et al. 1986; (3.): after Bigazz1 et al. 1990;
(4.): after Bigazzi et al 1993b; (5.): after Keller et al.
1996; (6.): from Bi.gazzi et al. 1994.

236

Table 5: I : from Bower et al., 1982; 2 : from the

NIST - SRM 278 ''Obsidian Rocle" Cerificate, Washington, D.C., 1981; 3: after Bigazzi et al, 1986.

Table 6 and Table 7: M J, M2 and M indicate samples selected for lNAA. These are not necessarily rhe
same pieces selected from the same location for FT
dating (after Oddone et al., 1999).
Table 8: The artefacts reported here are exactly the
same analysed by FT dating, except for PM
86/18(M2). analysed by 1NAA only, and STR.: this
sample, labelled "Vmicky, Stteda nad Bodrogom,
was supplied by the Museum of Natural History of
Nlilan. No indication was given about its exact provenance. The high quality of glass. different from the
Vini~k:y obsidian and the large size (- 5 cm) exclude
Streda nad Bodrogom perlite q11arry as a source. TI1e
piece may have been an artefact. Artefact PM 86/18
(Ml) corresponds to artefact PM 86118 of Tabl.e 2.
The prefix -PM has been omitted in this table (atler
Oddone et al., 1999 and this work, L.w.).

Description of t he origin of artefacts <lnalysed

Bodrogkere.~rr-Henyehegy.

Samples from lhe

Bodrogkereszhrr-Henyehegy Upper Palaeolithic site,
supplied by V. Dobosi.
Cejkov. PM 86/5: samples from the excavation from
loess at Cejkov, supplied by L. Banesz. PM 86/ 18:

samples collected as surface findings on a hill - 500 m

from Cejkov.
KaAov. supplied by L. Bane.sz
Trucal. Samples collected in the archaeological site
Tarcal-Citrombanya?. Supplied by E. Matyas.
Mad-Kakas-hcgy. Sample PM 86110-La was collected
as surface finding at lhe same locality of samples PM
86/ 10. Its shape and characteristics of the glass
revealed that this piece was an " imported artefact".

Tolcsva. Anefact PM 86/ 12ivl was collected at the

sampliJ1g locality PM 86/12, and its history is the same
as of artefac1 PM 86110-1 a. Artefacts PM 86/17 were
collected as swface findings in a ogncultaral ground
along U1c Tolcsva-Erdohorvati road (on the eastcm
side). - 1 km north ofTolcsva. Tlie qualny of the glass
indicate U1at the pieces were nor related to the Tolcsva
source area.
Streda nad Bodrogom. Supplied by V. de Michele; see
footnote to Table 8.

REFERENCES
ARIAS etal. 1981
ARIAS et al. L986

ARlAS-RADL et al. l 972

ASPINALL et al 1972
BAS 1979

BASSIAKOS el aJ. 1993

BASSIAK.OS et al. 1998

BATUM 1975

BlGAZZl et al. 1992

BTGAZZI et al 1986

BlG AZZl et al.1987

BlGAZZl et al.1988

BJGAZZl et al.1990

Arias C., B1gazzi G ., Bonadonna FP. Size correction and plateau age in glass
shards Nucl. Tracks RadtaL Meas., 5. 129-J 36
Arias C., Bigazzi G., Bonadonna FP.. Cipolloni M., Hadler J.C., Lattes
C.M.G.. Radi G. Fission track dating lu archaeology. A useful application. l n:
Scientific Methodologies Applied to Works of Art. Edited by Paolo L. Parrini.
Montedison,.Progetto Culcura eds., Milan,ltaly. 151- 159.
Arias-R.adi G., Bigazzi G., Bonadonna FP. Le tracce di fissione. Un mctodo
per lo studio delle vie di commercio dell "ossidiana. Origini, 6, 155- 170.
Aspinall A., Feather S.W., Renfrew C. Neutron activallon analysis of Aegean
obsidians. Nature, 237, 333- 334.
Bas H. Petrologische und geochemische Untersuchungcn an subrezenten Yulkaniten der Nordanatolischen Stl>nmgszone (Abschniu: Erzincan-Niksar). ?h.
D. Diss. Univ. Hamburg, l 06 p.
Bas.siakos, Y.- Kilikoglou, V. -Souvatzis, K.- Grimanis, A.P., Provenance
Studies of Obsidian from Mandala in Macedonia, Greece Abstracts of Asmosia 1993, lllrd lnt.Conf. Athens 17- 23 May Demok:ritos, Ailiens NCSR
Demokritos 4.
Bassiakos. Y. -Bir6, K. -Kilikoglou, V. Obsidian trade - the South-Eastern
connection. Poster presented at 31 st ISA Symposium, Budapest. In: Bir6
K.-Horvluh T (t:ds.). 31 51 lntcrnational Symposium on Archaeomc1ry. Program and Abstracts Budapest 22.
Batum 1. Petrograpbische und geochemische Unrersuchungen in den Yulkaogebieten GBllOdag und Aci~ol (Zenttalanatolien/Turkei). Ph.D. Diss.,
Univ. Freiburg, IOI pp.
Bigazzi G., Coltelli M., Hadler N. J. C ., Osorio Araya A. M., Oddone M ..
Salu.ar E. Obsidian bearing lava flows and pre-Columbian artefacts from
Ecuadorian Andes: first new multidisciplinary data. J. South Am. Eanh Sci ..
6, 21- 32.
Bigazzi G., Meloni S .. Oddone M., Radi G .. Provenance study of obsidian artefacts: trace elements analysis and da1a reduction. 1. Radioanal. Nucl. Cbem..
98. 353- 363.
Bigazzi G .. Yegingil Z., Ercan T., Oddone ~ .. Ozdogan M.. Age determination of obsidian bearing volcamcs in eastern Anatolia using the fission track
dating method. Geological Bulletin of Turkey (English Edition), 401 57-73.
Bigazzi G., Dompnier E., Hadler Neto J. C., Poupeau G., Vulliez E ., A reactor
intercalibration for fission track. dating and urnnium micromapping. Nucl.
Tracks Radiat Meas., 15, 755- 757.
Bigazzi G., Marton P., Norelli P., Rozloznik L. ( 1990). Fission track dating of
Carpathian obsidians and provenance identification. Nucl. Tracks Radiat.
Meas., 17. 391- 398.
237

BIGAZZI et al. l993a

BlGAZZl et al.1993b
BIGAZZL et al.1994
BlR0198J
BIRO 1984

BlRO- POZSGAl 1984

BIRO el al. 1986

Big11ZZi G., Erca11 T., Odclone M., Ozdogan M., Yegingil Z.~ Application of
fission track dating to archaeometry: provenance srudies of prehist9ric obsidian artefacts. NucL Tracks Radiat. Meas, 22, 757-762.
Bigazzi G., Yc.gingil z. Ercan T., Oddone M., Ozdogan M. (1993q). Fission
track dating obsidians in Central and Northern Anatolia. Bull. Volcano!., 55,
588-595.
Bigazzi Q., YegingU Z., Ercan T., Odd~me M., Ozdogan M. Prove)1ance studies of prehistoric artefacts in Eastern Anatolia: fusueSults of an interdisciplinary research. Miner. Petrqgr. Acla, XXXVII, 307-326.
Bi:r6, K. T., A K.arpat medencei obszidianok vi~sgalata (lnvestigation of obsidian from the C:::arpathian B~in), Archaeologiai Ertesjt6 (Budapest) l 08
196-205.
Bir6, K. T., Distribution of obsidian from the Carpathian Sources on Central
European Palaeolithic and Mesolithic sites. Acta Archaeologica Carpath1ca
(Krak6w) 23 5-42.
Bir6, K. T.-Pozsgai L, Obszidian lelohely-azonositas elektronsugaras mik:-

roanalizis segitsegevel I Obsidian characterisation by electron micrnprobe

analysis lparregeszet - lndustrial Arcbaeology Veszprem MTA Yeszprem.i
Akademiai Bizottsaga 1.984, 1, 25-37.
Biro K.. T. Pozsgai 1, Vladar A. Electron beam microanalyses of obsidian
samples from geological and archaeological sites. Aeta Arch. Acad. Sci.
1

Hung., 38,257- 278.

BIRO et al. 1988

BIRO eL al. in press

BLACKMAN 1984
BOWER 1982
CANN et al. 1964

CAUVlN et al. 1998

CORNAGOIA et al. 1962

CORNAGGIA et al 1963

Bir6, K. T.--.Pozsgai, I.-Vladar, A., Central European obsidian studies. State

of affairs in 1987. ln: Jar6 M.-Kolto L. eds. , Archaeometncal Research in
Hungary Budapest National Centre of Museums 1988, 119-130.
Biro, K-Vladar., A.-Pozsgai, I., Characterisation of obsidian from the Carpathian region and Central Anatolia (manuscript in press)
Blackman M. J., Provenance .studies of Middle Eastern obsidian from sites i:n
highland in lran, Archaeol. Chem. 11, Adv. in Chem., 205, 19-50.
BGWer N. W., Lewis C. M., Neifert P. E., Gladney E. S. ( 1982). Elemental
coneentrati0n .in twenty NIST Standards of geochemical interest. Geostandards .Newsletter 1527-60.
Catm J. R., Renfrew C .. The characle1i5ati.on of obsidian and its application Lo
the Mediterranean region. Proc. Prehist. Soc., 30, 111-132.
Cauvin M-C., Gonrgaud A., Gran1zeB., Arnaud N., Poupeau G., Poidevin J.L. Eds.. L'Obsidienne au Proche et Moy en Orient. Du volcan a I' out1J. BAR
International Series 738. Archaeopress, Oxford. 388 pp.
Cornaggia 0., Castiglioui 0 . fussi F., D 'Agnolo. M. 1ndagine suJla provenienza dell 'ossidiana in uso nelle industrie preistoriche italiaoe. Atti della SocietUtahana di Scienze Naturali e del Museo Civico di Sroria Nanirale di Milano, 101, 12-18.
Comaggia 0., Castigli.oni 0., Fussi F., D'Agnolo M. fndagine suJla provenienza dell'ossidiana utilizzata .nelle industri.e preisl.oricbe del Medite.rraneo
occidental.e. Atti. delJa Societa ltaliana di Scienze Naturali e del Museo Civ:ico
di Storia Narurali e del Museo Civico di Storia Naturale di l\1.ilano, 102.
31.0:-312.

DIXON et al.1968
DURRANI et al. L971

ENGELMAN 1990
ERCAN et al. 1989

Dixon .T. E., Ca.n u J. R., Renfrew C. Obsidian and t.he origin of trade. Scientific
American, 21.8, 38-45.
DuuaniS. A., Khan H. A., Taj M, Renfrew C. Obsidian.source identification
by fission track analysis. Natur~. 233i 242-245.
Engelman L. (1990). Cluster analysis of cases. In: BMDP-statistical software.
University ofCalifomia pres.s, Los Angeles, USA, B17-B27.
Ercan T., Yegingil Z., Blgazri G. The obsidfans: definilion, characte11i.stics and
distribution in Anatolia; ge.ochemical features of Central Anatolian obsidians.
Jeom.Derg. (Bult of Geomorphology), 17, 71-83 (in Turkish).

238

ERCAN et al.1996

ERENTOZ-KETlN 1974

FORNASERl et al. L977

GYARMATI l 977
INNOCENTI et al.1976

INNOCENTI et aJ.1980

JAFFR.EZIC H. 1976
KELLER et aJ. 1976
KELLER et aJ. 1990

KELLER et al. 1996

Kl.LlK.OGLOU et al. 1994

MILLER- WAGNER 1981

NAESER et al. 1981

ODDONE et al. 1999
OZDOGAN 1996

PECSKAYeLal. 1986

PERLAKY 1972
RENFREW et al. 1965

Ercan T., Saroglu f., Kus!u J.. Features of obsidian beds formed by the aclivity of the volcanoes in Ana~ia since 25 million years B.P. Arcbaeometry 94.
The Proceedings of the 29 International Symposium on ATchaeometry,
Ankara 9-14May1994, 505-513.
Erentoz c .. Ketin I. TGrlciye jeoloji Hantasi. Explanatory text of lhe geological map of Turkey, I:500000, sheet Kars. MTA - General Directorate of Mineral Research and Exploration Eds., Ankara, Turkey. 57p.
Fomaseri M.. Malpieci L., Palmieri A. M., Taddeucci A. Analysis of obsidians from the late chalcolithic levels of Ars1antepe (Malatya). Paleorienl, 3,
231- 246.
Gyarmati P. Intermediate volcanism in the Tokaj Mountains. Ann. Hung.
Geol. Inst.. 1-195.
lnnocenli F., Mazzuoli R., Pasquar G., Radicati di Broz.olo F.. Villari L. Evolution of the volcanism in the area of interaction between the Arabian, Anatolian
and Iranian plates (Lake Van, Eastern Anatolia). J. Volcano!. Geoth. Res., I,
103-112.
lnnocenti F., Mazzuoli R., Pasquar G., Serri G., Villari L. Geology of the volcanic area north of Lake Y;'ln (Turkey). Geol. Reisch., 69, 292-323.
Jaffrezic H. L'6stimation de l'e rreur mtriduit dans le dosage des elements a I'
etat de craces dans le roches liee aux caracteristiques statisliques de Jeur repartition. Talanta, 23, 487-50 l.
KeUer J., Jung D.. Eck.hart F. J., Kreuzer H. Das Galaliscbe Andesitgebiet
(Massiv von Kizi.lyahamam, Tllrkei). Unpttb1. DFG Report, 34 p.
Keller J., Seifried C. The presenl status of obsidian source identification in
Anatolia and U1e Near East. Vulcanologje et Arcbaeolog1e. Actes des Ateliers
Europens de RaveUo, Louvain - La Neuve (F. Widemann and C. A1bore Livadie Eds.). PACT 25, 57-87.
Keller J .. Bigazzi G., Pemicka E. A combined major element- trace elemc.:nt fission track characterisation of an unknown source in Northeastem Anatolia:
the Galata-X source - some general observatiou,s from an obsidian short story.
Archaeometry 94. The Proceedings of the 29 TntemationaJ Symposium on
Archaeometry, Ankara 9-14 May L994, 543- 551.
Kilikoglou V., Bassiakos Y., Souvatzis K. and Grimarus A. P. Carpathian obsidian in Macedonia, Greece. Archneometry 94 - 29lh lut. Symp. on Archaeometry, 9- 14 May 1994, Ankara. Turkey. Abstracts, 46.
Miller D. S., Wagner G. A. Fission-t:r:ack ages applied to obsidian artefacls
from South America using the plateau-annealing and track-size age-correction
techniques. Nucl. Tracks Radiat. Meas. , 5, 147- 155.
Naeser C. W., Briggs N. D., Obradovich J. D., Izen G. A. Dating of Lephra.
Geochronology of Quaternary tephra deposits. S. Self & R.S.J . Sparks Eds.,
Tephra Studies, 13-47.
Oddone M., Marton P., Bigazzi G., Bir6 K. T. Chemical characcerisatioos of
CarpatWan obsidian sources by in~tn1mental and epithennal neutron activation
analysis. J. Radioanal. Nucl. Chem.. 240, 147-153.
Ozdogan M. Obsidian in Anatolia: an archaeological per~ective on the status
of research. Arebaeometry 94. The Proceedings oflhe 29 lntemational Symposium on Arcbaeometry, Ankara 9-14 May 1994, 423-431.
PCe-skay Z., Balogh K., Szeky-Fux V.. Gyannati P. Geochronological investigations on the Neogene volcanism of the Tokaj Mountains. Geol. Sb. Geol.
Carp., 37, 635- 655.
Perlaky, E.. A Tokaj-Hegyseg harmad.kori savan)'U vulkanizmusa (Tertiary
acidic volcanic activity in the Toja Mts.). PhD Diss., Budapest (in }fungarian). l-.
Renfrew C., Cann J. R., Dixon J. E. Obsidian in the Aegean. Annual B.S.A.
Athens, 60, 225-247.

239

Renfrew C., DiJcon J. E . Cann J. R . Obsidian and early cultural contact m the
Near East. Proc. Prehist. Soc , 32, 30-6&.
Renfrew C., Dixon J.E. Cann J. R. Further anolys1s of Near Eastern obsidiRENFREW et al. 1968
ans. -Proc. Prehist. Soc., 34, 319- 33 1.
REPCOK er al. 1988
Repcok I., Kaliciak M ., 1Bacs6 Z. Age of some volcanicl> of Eastern Slovakia
(in Slovakian with abstract ia English). Zapadne Karpaty. ser. miner. petrogr.
metalogen., 11 , 75-88.
STORZER- POUPEAU 1973 Storzer D.. Poupeau G. Ages-plateau de m'ineraux el verres par la mechode des
traces de fission. C. R. Acad Sci. Paris, 276 sene 0 , 317 319.
STORZER-WAGNER l969
Storzer D., Wegner G. A. Corrcctionoftherrnally lowered fission track ages of
tektites. Earth Planet. Sci. Len., 5, 463-468.
STORZER-WAGNER 1982
Storzcr D .. Wagner G. A . (1982). The appLicaLion of fission crack dating in
stratigrapJ1y: a critical review. NumeticaJ dating io stratigraphy. Edited by
G.B. Odin, John Wiley & sons Ltd. Eds., 199-222.
SUZUKI 1969
Suzuki M. Fission track dating and uranium contents of obsidians. Daiyonki
Kenkyu, 8, 123-130.
szAn ECZKY 1886
Szadeczky G. A MagyarorS?.a_gi obs1dianok, kiilonos tekintettel geolog1a1
viszonyaikra (in Hungarian). Ertekezesek a termeszettudomanyok korebol.
XVV6, 1-64.
TANER 1977
Taner M . F. Etude geologique et petrographique de la region che Gi.iaeycerk:izdere, situee au sud de Rize (Pontides onentales, Turquie). Ph. D. Thesis.
University of Geneva, 180 p.
TOTH ET AL. 1999
Toth, A.-Bir6, K. T.-Dobosi, V. T., Gorog-magyar kozos obszidianvizsgaletok elozetes eredmenyeibol (Prehminruy results of Oreek-Hunganan
PetroarchaeologicaJ Collaboration Project). Poster presented in the Geoarcbaeology Meeting ac the BNM, .Budapest, 10-11 Mey 1999.
Vass D., To~er J., Bagdasarjan G. P., Kaliciak M., Orhcky 0., Durica D.
VASS et al. J 978
Chronology of volcanic events in .Eastern Slovakia on the grounds of isotopicaJ-palaeomagnetical reseru;ches {in Slovakian with abstract in English). Geol.
Prace, 7t, 77-88.
Vertes L. The Upper PaJaeoUthic sile on Mt. Henye at BodrogkeresztUr. Acla
VERTES 1966
Arch. Hung., 18, 3-14.
Wagner G. A., Weiner K. L. Deurches ArcbeoJogiscbes lnstitut Demircibeyek
WAGNER-WEINER 1987
- Die Ergebnisse der Ausgrabungen 1975-1978. Herausgegeben von M.
Korfmann. Band U; Naturwissenschaftliche Ontersuchungen. Verlag Philipp
von Zabem Eds., Mainz am Rhein, Gennany, 26-29.
Walter R. C. Application and limitation of fission-track geochronology to
WALTER 1989
Quaternary tephra. Quat. lntematl, l, 35-46.
Westgate J. A. (1989). 1sothellllal plateau fission track ages of hydraced glass
WESTGATE 1989
shards from silicic tepbra beds. Earth Planet. Sci. Lett., 95, 226-234.
WILLIAMS-NANDRIS 1977 Williams, 0 .-Nandris, J.; The Hungarian and Slovak sowces of Archaeological Obsidian: an Interim Report on further Fieldwork Journal of Archaeological Science 4 207-219.
WlLUAMS-THORPE 1984 Witliams-Thorpe, 0.-Thorpe, R. S. The distribution and sowces of archaeological pitchstone in Britain J. of Arch. Sci. 11 1- 34.
Wright G. A., Gordus A. A. Distribution and utilization of obsidian from Lake
WRlGHT-GORDUS 1969
Van sources between 7500 and 3500 B.C. Am. J. Arch.. 73, 75- 77.

RENFREW et al. 1966

240

lnstrume.ntal a nalysis D.
Ion beam analyses of artefa cts from the Bodrogkeresztur- Re.nye lithic assemblage

K.T. Biro-Z. Elekes - B. Graruze

-tbstract
ln frames of a collaborauon projec1 between the
llunganan National Museum ru1d the lnslitute of
Nuclear Research, Debrccen (ATOM K.1 I), ion beam
analyllcal techniques were used for provcnancing geological and archaeological samples of

a. obs1d1an
b. radiolarile

c, control samples of various other local materials

(limnic quartzilt!, "stone maffow").
PIGE and PIXE methods were used for analysis in
the ATOM.Kl; additionally. LA-TCP was used for the
analysis of obsidian samples in Orleans, France.
Identification of obsidian samples proYed to be effect1 ve as known for several analytical techniques already; analysis of radiolarite samples represent preliminary state of research with a lot of open questions.
Details of results on geological source areas and efficiency of characterisation are given elsewhere (Elekes
et al. 2000a, 2000b, Biro et al. in press).
ln this paper, the data relevant lo the Bodrogkeresztur Upper Palacolithlc site are presented.

Method

The investigations were mn111ly carried out using

non-destructive ion beam techniques, namely Particle
luduced Gamma-ray Emission (PlGE) and Particle Induced X-ray Emission (PIXE) methods. Utilizmg a
novel CJover-Ge-BGO system the deteanination of
light elemental composition via the detection of
gamma-rays could be made with extremely low detection Lunits because of the high Compton-suppression
factor. The minor and trace constituents of the samples
were srudied employing common PIXE technique.
fhe results achieved by ion beam methods were compared and contributed with the data obtained with Laser Ablarionlnducllvcly Coupled Plasma-Mass Spectrometry lLA-ICP-MS).
Objectfre~.-

oftlte st11dy

The collaboration project between ATOMKl and

Hi'IM, with participation in other international programs (COST. IGCP-442, GIIPCP) aims at using ion

beam analysis techniques for d1e study of archaeological artefacts. Several themes are investigated, among
them two are relevant to the monograph on the Upper
Palaeolithic site Bodrogkeresztfu-lienye: obsidian
studies and rndiolarile analyses. Both of these themes
included the study of authentic, source collected reference samples, aiming at a good coverage of Lhe
possible geological sources of the prehistoric people
as well as archaeological matenal for comparison and
provenance study. Details of these studies are presented elsewhere (obsiruan: ELEKES et al. 2000b, radioJarite: ELEK.ES et al. in press). Archaeolog1cal
samples in both themes included matenal from Bodrogkeresztil.r-ffenyc site, and we could use this example as an example of succesful source identification
(ELEK.ES et al. 2000a.). The method is also working
the other way round. 1hese results are sujtable to characterise the local, regional contacts of the site

S amples an alysed
'The material analysed from Bodrogkeresztur ii; presented on Fig. I. Samples were selected from the
Oakes/chips of the settlement, on macroscopical basis.
All of the studied flakes are in the collection of the
Hungarian National Museum. inventorised under Pb
64/363. For the analyses, a small piece of the flake
was taken from the flakes: control material will be
stored in the Lilhotheca collection.

Obsidian (Fig. l/4- 7):

363A - Bodrogkere-sztur-Henye. J>b 64/363 (o): small
decortication flake with hydrated cortex. Deep
black, non-Lransparent. Macroscopical clussification: Carpa1.hian 11 T ( l'olcsva) obsidian
363B - Bodrogkcreszti'.1r-Henyc Pb 64/363 (b) smaU
Oake. deep black with lighter stripes. non-cransparent. Macroscopical classjfication: Carpatluan 11 T
(Tolcsva) obsidian
3631 - Bodrogkeresztirr-Henye. Pb 64/363 " karpati
I"- small chip, transparent grey. Macroscopical
classification: Carpathian I (Slovakian) obsidian

OTKA T 02577 l
241

363E - Bodrogkeresztlir-Heuye. Pb 6413.63 "karpati 2

E"- small decortication flake with hydrated cortex.
Dark graphite grey, non-transparent. Macroscop1cal
classification: Carpathian JI E (Erdobenye-Miid)
obsidian.

Bk-5 - Boclrogkeresztfu-Henye. Pb 64/363 (5) Limnic

quartzite, thick blade-like flake, slightly translucent. blwsh white.

Results

''Rad10/arire '' (Fig.I . 1- 3, 8- 9):

Obsidian samples were analysed by PIGE for light

elements.
Bk-1 - BodrogkeresztUr-Henye. Pb 64/363 ( I) radfo1.ariLe, thick core-flake, bluish grey " Carpathian" radfolarite
Bk-2 - Bodrogkeresztlir-Henye. Pb 64/363 (2) radiolarite, tlat flake, bluish grey "Carpathian" radiolarite with red margin (this latter parrwas analysed)
Bk-3 - Bodrogkercszrur-Henye. Pb 64/363 (3) radiolarile, chip, dark red "CarpaLl11an" radiolarite
Bk-4 - Bodrogkeresztitr-Henye. Pb 64/363 (4) "stone
marrow" (kaolinitic hydroquartzire) blade, porcelauush white

Sample

36JA
,...._

N1~0

t)

~~

3. l7

15

JOJA
3638
3631

- S1

L1

796,3303
686,l68J
533,4743

81 .S7J2

73412,52

1 6451.1

347332..1

IOK, 1443

66047,45

26548,23

)3524~. t

104,2173

58428,94

25323.34

3SR12ll

Al

Na

and LA-ICP-MS for a wide range of elements including main components. accessory elements and some
trace elements:

.. ~

l<O

r"il

Mie

c..

Ti

Mn

i I

137

4,47

1036

599

8516

979

250

132: 1 ____.

1454

7116

562

306

10::?19

434
923

5779

307

342

6306

8178

967

13370

(5)

13.5

Ssmple

~~f

SI02

Obsidian PIGE (concentrations in ppm)

3638

J.4S

4,(A

J 11

14.3

75.2
76,2

798

363l

ND

4 14

SI :?
378

J632E
Sample

2,72

16,0

73,0

NU

4A

980

Rb

Sr

Zr

Nb

LSI

Ct

Tb

ll

363A

Zn
29

:?61
Nd

198

27

14 I

10

498

25

17

180

:26

IOI

IJ

62

2~

Ii

,,

3631

28

167

-17

16

51

483
335

36
33

66

3638

65
56

12

3632

48

187

60

27

142

10

4::?8

16

~~

.e

3:?

61)

l~

~3

The elemental d1St:ribution was compared lo geological source materials from the Carpathian 1., JI.
sources and further comparative materials from
Greece and other localities. As best discriminant, ratio
of Y Sr and Nb/Sr were used

sidered local (about 10 km to the NW from lhe settlement); Tolcsva is located a bit further on (20 km to N.
NW) and the Slovakian sources can already be considered mesolocal (regional (40 km to N).

As a result of chemical analyses, the samples were

classified into the following categories:

ples (limnic quartzite, stone marrow) were analysed

by PIGE + PIXE method.

Radiolarite and other siHceous raw material sam-

Radiolarite", PlGE (ppm)

363A

Mlid-Erdobcnyc

3o.>B

foksva

3631

Oirpnthian l

Sample

3032E

Mad-Erdobl!n)e

Bk~ I

Observacions on macroscopic grounds proved basically correct with the exception of 363A, which was
classified 10 Mad-Erdobenye type instead oCTolcsva.
The chemical analysis corroborated our view that
the people living at the Bodrogkeresztlir-Henye Upper
Palaeohtluc settlement knew and exploited aJI Carpathian obsidian sources, of which Mad can be con-

242

radiollln1e

f
17~.l l42

Li

Al

Si

1'11

l?l, 12911 33876,65 11 77,93 1 -100234.3

llk-2
rac:Jlolnrilc

88,6329

68,838

llk.J
ntdlola rirt

118,9960

57,270S

l 1631.S

t5J8,208 4SS374,4

Bl.-1
11cr11e m.i mi\\'

197.9 127 181 ..SS.5 1 30596.S

926,4845 399517,(>

B k--5

63,5.SOB

limnk qu.rtzlle

9,6343

8027..502 685.0:238 4 54t\74.6

3404,757 20-l.689-1 -16J280,9

'Radiolarite", PIXE (ppm)

St

465600

< 1600

< 160

4 70700

<JOOO

<160

---

491500

<.1600

357800

Ca

TI

CJ'

\In

238,6

2123

674,7

SH ,I

29.S

19

10 -I

380..S

1807

).LI

86,49

13

19

<160

2205

2784

1097

U6,ti

JS

1.7

<1600

< 160

1564

34900

1848

591.3

113.!l

36,91

47SH00

< l6U0

<160

2381

37S,8

138i<

17

68

SI

10.5

Fe

Co

"1

Cu

Z.n

Al

Rb

Sr

Zr

608.3

10.93

4.76J

7,714

7,.SlS

5,0M

6,49

:?8,94

<1

.]

1----

7~6.:2

< 12

<.S

9,648

2,S

7.168

13420

<:;12

<5

13, 16

9,76S

6.218

21.98

U.!)1
>----42..SS

I Bk-4
stunt ntu nO\\

5:?63

<12

<j

4S,73

3212

9.ll8

llS.3

' m.-5

1!Xl.1

5,!37

4.593

4,433

4,637

20.36

<.S

;sample

Ijruiolnrite
m..-1

__

Bk-2
rndiularilc

---- - - - ----

~huicc
81<-l

~nttnatTO"
-5
bmnlc quarttitt

~m11k
Bk- I

l nuhor..licc
Ill.-?
nidlnlarilc
Bk-3
nulJulluitc

~ilc q11amice

Radiolarite, as well as olher siliceous raw materials

have been only sporadically investigated by geochemical methods. Part of the studies were only semiquantitative (KOZLOWSKl et al. 1981); others not
representative in quantity of mate.rial (nr. of sources
compared to known occurrences, ns well as nr. of
pieces studied from the individual sources: BIR6PALOSJ 1986) Existing evidence has been collected
lil the volumes of Lithotheca (BJR6- DOBOSI l991 ,
BIR6-oOBOSl:_SCHLEDER 2000). The authors
performed special srudies for source detennination of
radiolarites within the Carpathian Basin (ELEK.ES et
al. in press: BLRO el al in press) and concluded that
further daaa are needed for reliable source identification.
Stone marrow and limnic quartzite samples were
mainly separable - on the basis of their very high I
very low K., Rb content from lhe radiolarite samples.
.Former geochemical analysis of hydrothcr"'!al siliceou. raw materials (BIRO e1 al. 1984, BIRO I 986,
VlCZlAN l986) proved a great variability within the
hydrochennal/limnic siliceous rocks. therefore we
cannot aim ac exactly identify mg che source of these
rocks on the basis of chemical composition. However,

<.l

-- - - -

\Is
I

62.58

~~

73,HI

,,

32,43

,7

19,-15

40.2

43,25

~9

<7

<1

they are distinctly sepnra~le from the radiolarite

group.
Co1rclusio11s

Ion bean1 analytical 1echniqi1es c~n be-appUed with

promising results on archaeological lithic material.
Obsidian - as a raw material with large arcbaeometrlcal cognisance and chem1oaily dIBtinctlvely dffferent composition per source can be effecuvely
characlerised by Lhe applied methodology
(PLGE/PIXE; LA-ICP-MS)
Radiolarite is a more difficult problem: chemical
composition less distinct, archeometrical, geochemical daLa rare. while source areas are mor numerous geographically and formation pcnod much
longer U1an tllat of obsidian.
Though on very small number of samples, radiolarites could be separated from siliceous raw materials of other (postvokanic) origin
Our results for Bodrogkereszt.ur mainly contributed
to characterisation of the local- mesolocal raw materials.

243

Fig. I. Samples analysed from Bodrogkereszti1r. inY. HNM Pb 64/363.

1 Ilk-I radiolarlte; 2. Bk-2 ra.diolarile; 3. Bk-3 radiolarire; 4. 3631 - ohSidian, 5. J63B- obsidian: 6. 363E - obsidian; 7. 363A obsidian: 8. Bk-5 - limnie quartzite; 9. Bk:-4- stone marrow"'.

REFERENCES

BJRO l986a

BIRO et al. 1984

BTRO et al. in press

Takacs-B ir6 K., The raw material stock for chipped stone artefacts in Lhe
Northern lv1id-Mountain$ Teajary in Hungary. Silmeg Papers (Vol. I) Buda-

pest KMJ Rota 1986 183-196.

Talc:acs-Bir6 C.-Siman K.-Szakall, S., On a characteristic &i02 Raw Material Type Group 4sed in Prehistoric Hungary. illrd lntemalional Seminar on
-Petroarchacology. P1ovdiv 1984 103-126.
Bir6, K. T.-Elekes, z..-Uzonyi, !.-Kiss A.z., Racliolarit mintlik vizsgalata
ionnyalab analitikai m6dszerekkel. In press for Folia Archaeologica 2000.

BIRO-DOBOS! 1991

Uiro, K . T.- Dobosi, V. T., LITHOTHECA - The Comparative Raw Material

Collection of the Hungarian National Maseum. Catalogue. ( 1991) Budapest

BTRO- DOBOSJ-

Bir6. K. T.-Dobosi, V. T.-SchJeder, Zs., LITTIOTHECA - The Comparauve Raw Material Collection of the Hungarian National Museum. Catalogue

1- 268.
SCHLEDER 2000

(ll) 1- 332.

244

BIRO- PALOSI 1986

ELEK.ES et al. 2000a
E LEKES et al. 2000b
EL EKES et al. in press

KOZLOWSKI et al. 198 l

VJCZl.AN 1986

Biro, K. T.- Palosi M .., A pattintott k6eszktlztlk nyersanyaganak forrasai

Magyarorszagon. MAFI EviJel az 1983-as evrol (1986) Budapest407-435 .
Elekes, Z., Biro, K. T .. Rajta I., Uzouyi, I., Gl"atuze, B. Kiss, A. z. Analyses of
Obsidian and Radiolarite Samples by Ion Beam Techniques. 32 SA Conference 2000 Mexico City.
Elekes, Z., Uzonyi, 1., Graruze, B., R6zsa, P., Kiss A. Z., Szoor Gy., Contribution of PIGE Technique to the Study of Obsidian Glasses, Nuclear losrrumcnts
and Methods B 161- l 63 (2000} 839.
EJekes, Z., Bir6, K . T., Uzonyi L Rajta L, Kiss A. Z., Geochemical analysis o f
radiolarite samples from the Carpalhian Basin, Jn press for Nuclear instruments and Methods in Pb.ysics .Research .
Kozlowski, J. K.- Manecki, A.- Rydlewski, J.-VaJde~Nowak. P. Mineralogico-geocbemical Characteristics of Radiolarites Used in cbe Stone Age in
Poland and Slovakia. Archaeologica Carpathica (Krakow) L981 21, L71-2 10.
Viczian I. X-ray diffraction investigation of silica rocks Silmeg Papers (Vol. l )
Budapest KMI Rota 1986 197- 200.

245