ZOOARCHAEOLOGY AND TAPHONOMY OF THE MIDDLE-UPPER PALEOLITHIC IN BONDI CAVE, REPUBLIC OF GEORGIA

ARCHAEOLOGY,

ETHNOLOGY

& ANTHROPOLOGY

OF EURASIA

Archaeology Ethnology & Anthropology of Eurasia 42/3 (2014) 2–13

E-mail: [email protected]

Copyright © 2014, Siberian Branch of Russian Academy of Sciences, Institute of Archaeology and Ethnography of the Siberian Branch of the Russian Academy of Sciences. Published by Elsevier B.V. All rights reserved.doi:10.1016/j.aeae.2015.04.002

2

Introduction

The Middle Paleolithic (MP)/Upper Paleolithic (UP) transition in the southern Caucasus is important for understanding the nature of the replacement of

R. Yeshurun1, M.-H. Moncel2, D. Pleurdeau2, R. Pinhasi3, N. Tushabramishvili4, T. Agapishvili4, 5, and D. Lordkipanidze5

1Zinman Institute of Archaeology, University of HaifaHaifa, 31905, Israel

E-mail: [email protected] de Paléontologie Humaine, Muséum National d’Histoire Naturelle,

1 rue René Panhard, Paris, 75013, FranceE-mail: [email protected]

3University College Dublin,Bel eld, Dublin 4, Ireland

E-mail: [email protected] State University,

Cholokashvili 3/5, Tbilisi, 0162, GeorgiaE-mail: [email protected]

5Georgian National Museum,Purtseladze 3, Tbilisi, 0105, Georgia

ZOOARCHAEOLOGY AND TAPHONOMY OF THE MIDDLE-UPPER PALEOLITHIC IN BONDI CAVE,

REPUBLIC OF GEORGIA

We present a detailed zooarchaeological and taphonomic study of Bondi Cave, western Georgia, containing Middle Paleolithic and Upper Paleolithic levels. The faunal assemblages are dominated by Caucasian tur (Capra caucasica) and European bison (Bison cf. bonasus). High proportions of specimens bearing cut marks and percussion marks indicate that human activity was the main depositional agent. Carcass transport was selective according to the animal’s size. Assemblage characteristics suggest short occupations of the cave by mobile Upper Paleolithic foragers. The general composition of the fauna probably re ects a more open setting for the Upper Paleolithic human occupations in this region, compared to the Holocene.

Keywords: Late Pleistocene, Caucasus, hunting, paleoecology.

Neanderthals by Anatomically Modern Humans in a region of potential interaction (Adler et al., 2008; Pinhasi et al., 2012; Moncel et al., in press). Recent studies provided a wealth of data on human life-ways across this major transition, as well as the pace of this

PALEOENVIRONMENT. THE STONE AGE

R. Yeshurun et al. / Archaeology, Ethnology and Anthropology of Eurasia 42/3 (2014) 2–13 3

change. It seems that in both the northern and southern Caucasus regions the demise of Neanderthals and associated MP industries took place a few millennia before the onset of UP industries, widely considered to be produced by Anatomically Modern Humans (though direct skeletal evidence is generally lacking) ruling out direct interaction (Pinhasi et al., 2011, 2012). This break is also evident in the technological (lithic) discontinuity and some aspects of material culture, e.g., the appearance of worked bone artifacts during the early UP (Adler et al., 2006; Pleurdeau et al., 2007; Bar-Yosef et al., 2011; Moncel et al., 2013). On the other hand, recent studies suggested that in western Georgia subsistence and obsidian gathering patterns of Neanderthals during the MP were quite similar to those practiced by Anatomically Modern Humans during the UP (Adler et al., 2006; Adler, Bar-Oz, 2009; Bar-Oz, Adler, 2005; Le Bourdonnec et al., 2012). Both populations narrowly focused their hunting on ungulates with a preference for prime-age animals, and developed hunting capacities allowing the interception of Caucasian tur (Capra caucasica) along its seasonal migration routes. Patterns of carcass transport, butchery and consumption were nearly identical in the late MP and UP, best evidenced at the stratigraphic sequences of Ortvale Klde and Dzudzuana caves (Adler et al., 2006; Adler, Bar-Oz, 2009; Bar-Oz, Adler, 2005; Bar-Oz et al., 2002; Bar-Oz et al., 2008; Bar-Oz et al., 2012; Moncel et al., 2013) (see also for the Northwestern Caucasus

(Hoffecker, Cleghorn, 2000)). As these studies are still based on a limited set of sites the analysis of additional MP and UP assemblages from western Georgia provides the means by which we can assess whether the observed patterns were typical to most or perhaps all Neanderthal and Anatomically Modern Humans populations in this region.

Recent excavations in Bondi Cave, Imereti, Republic of Georgia, produced a faunal assemblage in the context of MP and especially UP cultures (Tushabramishvili et al., 2012). This paper presents a detailed zooarchaeological and taphonomic study of the Bondi Cave faunas, adding an additional set of data for the Paleolithic of the region (Bar-Oz et al., 2012), and providing insights on late Pleistocene hunting and subsistence patterns along the southern margins of the Caucasus ridge.

The site and its setting

Bondi Cave lies in the basin of the Rioni and Kvirila rivers, Imereti region, northwestern Georgia, at an altitude of 477 m asl (Fig. 1). This region today is covered by forest, hilly, and cut by deep gullies. The cave opens to the south onto the slope of a small valley lying approximately 30 m above the Tabagrebi River, 2–4 km from the cave sites of Ortvale Klde and Dzudzuana (Fig. 2).

During the 2007–2010 eld seasons, twelve one by one meter squares were dug at the front part of the cave,

Fig. 1. Map showing the location of Bondi Cave and other sites mentioned in the text.

0 10 km

0 100 km

4 R. Yeshurun et al. / Archaeology, Ethnology and Anthropology of Eurasia 42/3 (2014) 2–13

Fig. 2. Bondi Cave (a) and its setting (b). July 2011. Fig. 3. Plan of excavations in Bondi Cave. a – 2007; b – 2009; c – 2010.

b

near its present-day drip-line. The excavation revealed a sedimentary sequence more than 2.8 m thick with eight distinct archaeological horizons which are interbedded between several levels of boulders and debris from major collapses of the cave roof (Fig. 3). Bedrock has yet to be reached (Tushabramishvili et al., 2012). Excavations have yielded two cultural complexes, a final MP and a UP sequence. Layer VII (the uppermost layer with MP af nities including thick blades, elongated akes,

and evidence of Levallois core technology) is dated to between 38.7 to 35 ka 14C BP (43–40 ka cal BPHulu). Layers V to III (UP levels indicating an increase in the bladelet ratio throughout time associated to numerous blades, with one Homo sp. remain from layer V) are dated from 24.6 to 14 ka 14C BP (29–17 ka cal BPHulu) (Ibid.). The UP layers would thus be contemporaneous in part with the Dzudzuana UP (units D to B) and the Ortvale Klde UP (layers 4d and 4c) (Moncel et al., in press).

0 2 m

b

c


Animal bones and lithic artifacts are present in all archaeological layers. In total, more than 7000 lithic artifacts were recovered. Layers II, IV and V contain the richest concentration, with nearly 90 % of the sample of artifacts recovered within these layers. The lower layers VII and VIII yielded much less artifacts but nonetheless contain thick elongated lithic products which contrast markedly with the thin and elongated blades found in UP layers (Ibid.). Evidence of re has been recovered throughout the sequence, demonstrated by burnt int and micro-charcoal fragments. The Bondi Cave fauna was previously studied from a paleontological perspective, focusing on taxonomic identi cation, by A. Vekua and M. Buchsianidze (Tushabramishvili et al., 2012).

Methods

Zooarchaeological and taphonomic data were collected following the procedures outlined in Bar-Oz and Adler (2005) and Bar-Oz et al. (2008). All bone bags from the 2007–2010 excavations were scanned for identi able elements. Units that could not be con dently assigned to a level or at least period were omitted. Skeletal element identification was as inclusive as possible and included articular ends and shaft fragments of long bones, teeth, cranial fragments, carpals, tarsals, appendicular elements, and other recognizable and quanti able bone fragments. Taxonomic identi cations were made using a virtual comparative collection of recent skeletons and osteological catalogues (Schmid, 1972; Pales, Garcia, 1981). Distinction of bovine taxa (Bos vs. Bison) followed S.J. Olsen (1960), and distinction of caprines (Capra vs. Ovis) followed M.A. Zeder and H.A. Lapham (2010). Species-level identi cations were checked and veri ed against the faunal lists made by A. Vekua and M. Buchsianidze, who used a comparative collection at the Georgian State Museum, Tbilisi. Skeletal elements that could not be assigned to species were grouped according to body size classes (i.e., small mammal, small ungulate, medium ungulate, and large ungulate).

Bone-surface modifications were inspected macroscopically on every recorded specimen using a low-resolution magnifying lens (×2.5). Modi cations caused by humans (butchery, burning, and bone fragmentation), animals (rodent gnawing and carnivore puncture, scoring, and digestion marks) and post-depositional attrition agents (weathering, trampling, bleaching, abrasion, and root activity) were recorded (e.g., (Behrensmeyer, 1978; Fisher, 1995)). Long bone shaft fragments with ancient fractures were examined for the morphology of fracture planes. Fracture angle and outline were assessed in order to determine whether the bone was fractured in a fresh or dry condition (see (Villa,

Mahieu, 1991) for a detailed typological description of bone fractures). The degree of completeness of the long bone shaft circumference was also recorded (i.e., complete, more than half, or less than half of the complete circumference) as an index of fragmentation for limb bones (following (Bunn, 1983)). Additionally, aging data were collected based on epiphyseal fusion of long bones, and dental eruption and wear of mandibular teeth (Grant, 1982; Stiner, 1990).

Results

The 2007–2010 excavation seasons in Bondi Cave produced a medium-sized animal bone assemblage, composed mainly of long-bone shaft fragments of medium and large ungulates. The assemblage is fragmented, although some complete bone elements and intact jaws do occur. Bone surfaces are mostly well-preserved. Approximately 5900 bone fragments larger than 1 cm in length were recovered, yielding 12 identi ed specimens (NISP) in the MP levels and 277 identi ed specimens in the UP levels.

The dominant taxa in both periods are Caucasian tur (Capra caucasica) and bison, most likely European bison (Bison cf. bonasus) (Table 1). Other ungulates, which are represented by isolated specimens, are red deer (Cervus elaphus) and roe deer (Capreolus capreolus). Carnivore remains are present in both periods, identi ed as wolf (Canis lupus) and bear (Ursus sp.). Judging by the relatively small size of the bones, the latter remains possibly belong to brown bear (U. arctos), but the currently available measurements are too few to verify this (Table 2). Bones identi ed to size-class rather than species form a large part of the assemblage, due to the preponderance of limb bone shaft fragments. The proportional division of size classes follows the species-speci c elements in that the large ungulate (probably all bison) and medium ungulate (dominated by tur but includes also red deer) classes overwhelmingly dominate the assemblage. The presence of aurochs (Bos primigenius) was reported from nearby contemporaneous sites (Bar-Oz et al., 2008), but was not positively identi ed in this study, as all the bovine specimens that could be attributed to species displayed the morphological characteristics of bison.

While the MP sample is very small (probably because the excavation has yet to reach beyond the top of the MP levels) an important difference was noted in taxonomic composition between the MP and UP levels (Fig. 4). The remains of large carnivores (wolf and bear) are signi cantly more numerous in relation to ungulates in the MP levels ( 2 = 10.14, p = 0.001).

The UP levels were grouped to three parts: lower (level VI), middle (levels IV and V), and upper (levels I–III). Along this sequence the proportions of deer


Table 1. Taxonomic composition of the Bondi Cave assemblages

Note: NISP, number of identi ed specimens; MNI, minimum number of individuals for mutually exclusive taxa.

Table 2. Anatomical measurements of the Bondi Cave assemblages

TaxonMiddle Paleolithic Upper Paleolithic

NISP MNI NISP MNI

Ungulates

Bison cf. bonasus 2 1 34 4

Ungulate large – – 66 –

Capra caucasica 1 1 38 3

Cervus elaphus 1 1 2 1

Cervidae 1 1

Ungulate medium 3 102

Capreolus capreolus – – 1 1

Ungulate small – – 10 –

Carnivores

Ursus sp. 2 1 7 1

Canis lupus 2 1 8 3

Carnivore medium – – 1 –

Carnivore small – – 4 –

Birds

Bird medium – – 3 –

Specimen number Season Square Subsquare Depth, cm Period Level Bone Measurements, mm

1 2 3 4 5 6 7 8 9

Ursus sp.

190 2010 B4 Ab 255–265 UP V, VI Max. canine L > 99.7; B = 32.3*

192 2010 B4 C 395–400 MP VIII Canine L > 47.1; B = 13.7*

88 2009 C4 B 205–215 UP V P4 A–P = 12.7; ML = 8.7

194 2010 B4 A 375–380 MP VIII Radius BD = 67

230 2007 73 … UP Phalanx 1 BP = 18.2; BD = 15.1; GL = 43.2; SD = 12.1

Canis lupus

174 2010 B3 B 370–375 MP VII P4 B = 10.4; GB = 14.1; L = 24.7

167 2010 A3 B 390–395 UP VI P4 L = 16.6; B = 7.5

278 2010 D’2–E’2 360 » … » L = 21.2; B = 8.2

218 2007 80 … » … M1 L = 24; B = 9.8

125 2009 C4 bd 195–205 » V » L = 30.1; B = 11.7

276 2010 B’3 C 255–265 » V (?) M2 + M1 L = 9.9; B = 7.3; L = 22.2; B = 8.9; Lmolarrow = 40

131 2009 C4 bd 185–195 » V Phalanx 2 BP = 10.3; BD = 9.4; GL = 25; SD = 7

Cervus elaphus

195 2010 B4 A 365–375 MP VII Calcaneus BP = 21.9; GL = 98.3


Notes: Method and abbreviations after (Driesch, 1976).*Measurement taken at the enamel-dentine junction.

Fig. 4. Assemblage composition in the MP vs. UP of Bondi Cave.a – ungulates; b – carnivores.

Fig. 5. Taxonomic frequency trends during the UP of Bondi Cave.

b

1 2 3 4 5 6 7 8 9

Capra caucasica

142 2009 A3 D 285–295 UP V Humerus BD = 53.5; HDH = 20.7

177 2010 B3 B 305–315 » VI Same BT = 47.9; HTH = 22.3

99 2009 B4 A 255–265 » V Radius BP = 45

140 2009 A3 Cd 275–285 » V Metacarpus BP = 34.5

208 2007 74 … » … Phalanx 2 BP = 16.7; BD = 13.5; GL = 33.9; SD = 12.2

264 2010 A’4 D 205–215 » V (?) Phalanx 3 DLS = 46.7; MBS = 10.2

Bison cf. bonasus

139 2009 A3 Cd 275–285 » V Scapula BP = 54.5; GLP = 81.7; LG = 70.1

222 2007 109 … » … Phalanx 2 BP = 42.2; BD = 35.9; GL = 52.5; SD = 34.4

223 2007 94 … » … Phalanx 3 DLS = 73.6

123 2009 C4 bd 195–205 » V Same DLS = 76.5; LD = 65.5; MBS = 20.9

Table 2 (end)


Table 3. Bone surface modi cation data and bone fragmentation data for each taxonomic group in the Upper Paleolithic of Bondi Cave

*Number of determinable specimens hereafter is indicated after the slash.

Fig. 6. Selected bone modi cations from the UP of Bondi Cave.

a – a medium ungulate tibia midshaft (No. 107) displaying evidence of hammerstone percussion (broad conchoidal notch, an internal ake scar, and spiral fracture); b – a tur humerus (No. 177) displaying carnivore gnawing (crenellated edges, tooth scores, and a narrow notch

lacking an internal ake scar).

and large carnivore are always low, and all levels are dominated by tur and bison (Fig. 5). However, the proportions of the latter two taxa change over time, with tur dominating the lower and middle levels and bison dominating in the upper levels (upper vs. middle levels: 2 = 7.76, p = 0.051).

The taphonomic origin of the MP faunal remains is hard to asses, given the small sample. One cut-marked bone (a red deer calcaneus) and one specimen bearing carnivore gnawing marks (a medium ungulate tibial fragment) were found, providing evidence for both human and carnivore involvement with the fauna. Much more data exists on the taphonomy of the UP assemblage (Table 3). An anthropogenic origin for the ungulate remains is indicated by high proportions of cut marks and hammerstone percussion marks. Cut marks were found on all skeletal regions of tur and bison and indicate all stages of butchery (skinning, dismemberment, lleting and periosteum cleaning). The impact of hammerstone percussion on the assemblage is demonstrated also with limb-bone fracture patterns. The majority of limb shafts retain less than half of their original circumference, frequently accompanied by “green” fracture edge indicative of deliberate fracture to remove intra-bone nutrients (Fig. 6, a). Strangely, burnt bones are

0 2 cm

b

Feature Large carnivoreUngulate

small medium large

NISP 14 11 141 100

Burning 0 0 1 (0.7 %) 0

“Green” fracture 0 0 15 18

Dry fracture 0 3 11 6

Intermediate 0 0 8 8

Shaft circumference:

<50 % 0 3 40 40

>50 % 0 0 3 0

100 % 2 2 3 0

Weathering (stages 3–5) 0/7* 0/10 6 (4.8 %)/126 7 (9.1 %)/77

Cut marks 0 0 11 (8.7 %) 8 (10.4 %)

Percussion marks 0/6 0/5 6 (9.7 %)/62 10 (20.8 %)/48

Gnawing (carnivore) 0 2 (20.0 %) 5 (4.0 %) 7 (9.1 %)

Gnawing (rodent) 0 0 0 0

Root marks 2 (28.6 %) 3 (30.0 %) 11 (8.7 %) 12 (15.6 %)

Trampling striations 0 0 7 (5.6 %) 7 (9.1 %)

Abrasion 0 2 (20.0 %) 7 (5.6 %) 2 (2.6 %)


nearly absent (NISP = 1). The carnivore remains do not display anthropogenic modi cations. Carnivore activity in the cave is attested by several gnawing marks on ungulate bones (Table 3; Fig. 6, b). However, the small number of carnivore remains and moderate abundance of their gnawing modi cations suggest that their role in the collection and modi cation of the UP faunal assemblage was not great. The assemblage suffered from mild effects of abiotic destruction processes, namely weathering, trampling, root activity, and abrasion (Table 3).

The skeletal element profiles for the medium and large ungulate groups in the UP (all levels combined) were reconstructed based on MNI (minimum number of individuals) tallies for the most abundant skeletal element in each anatomical region (Table 4). The results should be interpreted with preservation bias in mind. Speci cally, the assemblage appears to have undergone marked density-mediated attrition. Shaft fragments outnumber epiphyseal ends of the same limb bones in both the medium ungulate (NISP shaft = 48, NISP end = 27) and the large ungulate (NISP shaft = 41, NISP end = 8) classes. This indicates that taphonomic processes acted on skeletal elements differentially, according to bone mineral density, hence deleting the spongy epiphyses and preserving the cortical diaphyses of the same bone elements.

The results of the skeletal element analysis point to a near-complete representation of medium ungulate (tur) anatomical regions, along with a more selective body-part pro le for the large ungulate (bison) in the UP sample (Table 4; Fig. 7). Speci cally for bison, the mandible and intermediate limbs are present, while toes and lower-limb elements are markedly underrepresented. The low representation of axial parts in both size classes probably stems from the marked density-mediated attrition, acting more forcefully on ribs and vertebrae. Overall, it seems that tur were transported complete or nearly so to the cave, while the much larger bison were initially butchered in the eld and some high-utility (i.e., meat- and fat-bearing) and low-bulk body parts were preferentially transported.

The sample does not allow the reconstruction of detailed mortality pro les. For the UP levels, the dP4-worn M3 sequence suggests that at least two adult tur individuals were procured. In contrast, judging by the eruption and wear state of the lower P4, at least two juvenile (unworn teeth) and two adult bison (worn teeth) individuals were hunted. The bone fusion results for limb bones and phalanges indicate a prevalence of adult individuals for tur and bison alike; proportion of unfused out of all fused and unfused elements is 14 % (total NISP = 35) for

Table 4. Occurrence of parts of skeletons of medium and large ungulates in the Upper Paleolithic layers

Skeleton elementMedium ungulate Large ungulate

NISP MNI NISP MNI

Skull 7 3 4 2

Mandible 7 2 22 4

Cervical 2 – – –

Thoracic/lumbar 11 1 1 –

Rib 20 – 2 –

Scapula – – 1 –

Humerus 9 3 9 2

Radius 7 2 5 1

Ulna 1 – 1 –

Carpal 5 – – –

Pelvis 4 – 2

Femur 5 – 7 2

Tibia 9 3 18 5

Tarsal 7 – 1

Metapodial 24 2 6 1

Phalanx I 11 2 6 –

Phalanx II 5 – 3 –

Phalanx III 3 – 2 –

Sesamoid 2 – 10 –


Note: LBSF, long bone shaft fragment.

Fig. 8. Bone tools from Bondi Cave. a – WB-3, an awl; b – WB-2, an awl; c – WB-4, possible awl fragment; d – No. 160, a modi ed antler tip; e – WB-1, possible awl fragment.

Table 5. Worked-bone specimens from Bondi Cave

0 1 cm

0 2 cm 0 2 cm

0 2 cm0 2 cm

b c

de

Fig. 7. Skeletal element pro le of large ungulate (a) vs. medium ungulate (b) in the UP of Bondi Cave (all levels combined).

b

Sample number Provenience Period Level Bone Taxon Length,

cm Description

160 C3b, 245–255 UP V Antler tip

Cervidae 53 Cut marks, probably, from processing

WB1 C4bd, 135–145 » III–V LBSF … 16 Probably, awl fragment. Shaved longitudinal striations

WB2 E’2, 580 … » Large ungulate 52 Awl tip (complete)

WB3 B’4b, 225–235 UP » … 30 Awl (almost complete)

WB4 B’4d, 210–225 » » Medium mammal 37 Awl fragment. Shaved-longitudinal striations


Capra and 20 % (total NISP = 10) for Bison. Thus, for tur there are two lines of evidence, tooth eruption and wear sequences and bone fusion, which suggest procurement of mainly adult individuals, while for bison the picture is more ambiguous and awaits larger samples.

Five worked bone specimens were found in Bondi Cave (Table 5; Fig. 8), all in UP levels except for one specimen from an uncertain context. Four specimens are awls or awl fragments made of ungulate long bone shaft fragments, and one is a modi ed deer antler. All items but one of the awls were found incomplete, and none were burnt. The types and preferred blanks of the bone tools are reminiscent of the worked items found in Dzudzuana Cave (Bar-Yosef et al., 2011: Fig. 5, 6) though the Bondi assemblage is less varied, perhaps due to the smaller sample.

Discussion and conclusions

Bondi Cave yielded MP and UP faunal assemblages dominated by tur and bison, in accordance with similar sequences from western Georgia, especially the nearby cave of Dzudzuana and Ortvale Klde rock shelter (Bar-Oz, Adler, 2005; Adler et al., 2006; Bar-Oz et al., 2008; Adler, Bar-Oz, 2009; Bar-Yosef et al., 2011; Moncel et al., 2013). The faunal composition is also reminiscent of the Late MP cave-sites in the northwestern Caucasus (Hoffecker, Cleghorn, 2000). Our analysis focused on the much larger UP sample, which provided a satisfactory sample size. The UP faunal remains were demonstrated to result from human hunting, transport, and butchery for meat and marrow. Carcass transport was selective, as medium-sized ungulates (tur) were carried complete or nearly so to the site, while the much larger bison underwent some eld butchery. It is possible that large quantities of bison meat were nonetheless transported to the site as lleted cuts (Monahan, 1998). The discarded remains were ravaged by carnivores (notably wolf) subsequent to the human occupations and then suffered from some abiotic destruction processes, leaving a faunal assemblage characterized by density-mediated attrition.

The near absence of bone burning is interesting. Excavations in the nearby Paleolithic sites yielded higher proportions of burnt specimens (Bar-Oz , Adler, 2005; Bar-Oz et al., 2008), possibly resulting from accidental charring of preexisting refuse. This coupled with the evidence of carnivore activity and the moderate numbers of bones and individual animals hint at short occupations of the cave by mobile UP foragers. Judging by the high number of carnivores in the uppermost MP levels, it seems that MP human occupation was even less intensive. Nevertheless, body-part profiles and butchery patterns indicate that some (for bison) or all

(for tur) of the butchery and consumption sequence was performed onsite. The near-complete representation of tur body parts suggests that the hunting grounds may not have been very far from the cave. This conclusion ts well with the MP lithic assemblage, which is small

and represents partial reduction sequences, suggesting seasonal short-term occupations with importation of already worked products. However, the density of UP lithic assemblages rather suggests debitage workshops in situ using the local int and possibly another nature of occupation of the cave. These data contribute to the discussion on the type of land-use performed by MP and UP groups in this area.

The end of the UP sequence is characterized by procuring more bison than tur. A similar trend of diminishing tur proportions and elevated bovine proportions during the UP was observed also in Dzudzuana Cave, where it was suggested to re ect change in the season of site occupation (Bar-Oz et al., 2008, 2011). More data are needed to determine if this trend is indeed related to seasonality, perhaps in conjunction with changes in hunting gear or natural availability.

The two predominant animals hunted during the Bondi Cave sequence – Caucasian tur and European bison – thrive in mountainous environments containing both forest and open habitats (Adler et al., 2006; Julien et al., 2012). The late Pleistocene macrofaunal sequence in Bondi Cave may attest to lowered mountainous belts and more open environments in western Georgia compared to the Holocene, as the Mesolithic (early Holocene) faunal composition of the region is markedly different. In the Mesolithic of the nearby site of Kotias Klde, tur and bison are absent, whereas the dominant animals are brown bear, red and roe deer, and wild boar (Sus scrofa) (Meshveliani et al., 2007; Bar-Oz et al., 2009). The rarity or absence of the three latter ungulates in Bondi Cave and contemporaneous sites probably re ects a signi cantly more open setting for the late MP and UP human occupations in this region. Within the UP sequence, major environmental uctuations are not observed by the fauna, perhaps because the interim latitude of the cave and its location in western Georgia were habitable during much of the late Pleistocene, in contrast with high-elevation or more inland localities (Vekua, Lordkipanidze, 1998; Bar-Oz et al., 2012).

Acknowledgments

We thank A. Vekua and M. Buchsianidze for their help in taxonomic identi cation, and G. Bar-Oz for his assistance in the course of this research and his comments on the manuscript. We also thank the French Ministry of Foreign Affairs and the National Museum of Natural History (Paris) for their nancial support for eldworks and analyses.


References

Adler D.S., Bar-Oz G. 2009Seasonal patterns of prey acquisition during the Middle and

Upper Palaeolithic of the southern Caucasus. In The Evolution of Hominid Diets: Integrating Approaches to the Study of Palaeolithic Subsistence, J.-J. Hublin, M.P. Richards (eds.). Dordrecht: Springer, pp. 127–140.

Adler D.S., Bar-Oz G., Belfer-Cohen A., Bar-Yosef O. 2006Ahead of the game: Middle and Upper Paleolithic hunting

behaviors in the Southern Caucasus. Current Anthropology, vol. 47: 89–118.

Adler D.S., Bar-Yosef O., Belfer-Cohen A., Tushabramishvili N., Boaretto E., Mercier N., Valladas H., Rink W.J. 2008 Dating the demise: Neandertal extinction and the

establishment of modern humans in the southern Caucasus. Journal of Human Evolution, vol. 55: 817–833.

Bar-Oz G., Adler D.S. 2005 Taphonomic history of the Middle and Upper Palaeolithic

faunal assemblage from Ortvale Klde, Georgian Republic. Journal of Taphonomy, vol. 3: 185–211.

Bar-Oz G., Adler D.S., Meshveliani T., Tushabramishvili N., Belfer-Cohen A., Bar-Yosef O. 2002 Middle and Upper Paleolithic foragers of the southwest

Caucasus: New faunal evidence from western Georgia. Archaeology, Ethnology and Anthropology of Eurasia, vol. 4: 45–52.

Bar-Oz G., Belfer-Cohen A., Meshveliani T., Djakeli N., Bar-Yosef O. 2008Taphonomy and zooarchaeology of the Upper Palaeolithic

Cave of Dzudzuana, Republic of Georgia. International Journal of Osteoarchaeology, vol. 18: 131–151.

Bar-Oz G., Belfer-Cohen A., Meshveliani T., Jakeli N., Matskevich Z., Bar-Yosef O. 2009Bear in mind: Bear hunting in the Mesolithic of the southern

Caucasus. Archaeology, Ethnology and Anthropology of Eurasia, vol. 37, No. 1: 15–24.

Bar-Oz G., Weissbrod L., Gasparian B., Nahapetyan S., Wilkinson K., Pinhasi R. 2012Taphonomy and zooarchaeology of a high-altitude

Upper Pleistocene faunal sequence from Hovk-1 Cave, Armenia. Journal of Archaeological Science, vol. 39: 2452–2463.

Bar-Yosef O., Belfer-Cohen A., Meshveliani T., Jakeli N., Bar-Oz G., Boaretto E., Goldberg P., Kvavadze E., Matskevich Z. 2011 Dzudzuana: An Upper Palaeolithic cave-site in the Caucasus

foothills (Georgia). Antiquity, vol. 85: 331–349.Behrensmeyer A.K. 1978 Taphonomic and ecological information from bone

weathering. Paleobiology, vol. 4: 150–162.Bunn H.T. 1983 Evidence on the diet and subsistence patterns of Plio-

Pleistocene hominids at Koobi Fora, Kenya, and Olduvai Gorge, Tanzania. In Animals and Archaeology, J. Clutton-Brock, C. Grigson (eds.). Oxford: BAR, pp. 21–30. (BAR Internat. Ser.; No. 163).

Driesch A., von den 1976A Guide to the Measurement of Animal Bones from

Archaeological Sites. Cambridge, MA: Peabody Museum Press.

Fisher J.W. 1995 Bone surface modi cations in zooarchaeology. Journal of

Archaeological Method and Theory, vol. 2: 7–68. Grant A. 1982 The use of tooth wear as a guide to the age of domestic

animals. In Aging and Sexing Animal Bones from Archaeological Sites, B. Wilson, C. Griggson, S. Payne (eds.). Oxford: BAR, pp. 91–108. (BAR Internat. Ser.; No. 109).

Hoffecker J.F., Cleghorn N. 2000 Mousterian hunting patterns in the northwestern Caucasus

and the ecology of the Neanderthals. International Journal of Osteoarchaeology, vol. 10: 368–378.

Julien M.A., Bocherens H., Burke A., Drucker D.G., Patou-Mathis M., Krotova O., Péan S. 2012 Were European steppe bison migratory? 18O, 13C and Sr

intra-tooth isotopic variations applied to a palaeoethological reconstruction. Quaternary International, vol. 271: 106–119.

Le Bourdonnec F.-X., Nomade S., Poupeau G., Guillou H., Tushabramishvili N., Moncel M.-H., Pleurdeau D., Agapishvili T., Voinchet P., Mgeladze A., Lordkipanidze D. 2012Multiple origins of Bondi Cave and Ortvale Klde (NW

Georgia) obsidians and human mobility in Transcaucasia during the Middle and Upper Palaeolithic. Journal of Archaeological Science, vol. 39: 1317–1330.

Meshveliani T., Bar-Oz G., Bar-Yosef O., Belfer-Cohen A., Boaretto E., Jakeli N., Matskevich Z. 2007Mesolithic hunters at Kotias Klde, western Georgia:

Preliminary results. Paléorient, vol. 33: 47–58.Monahan C.M. 1998The Hadza carcass transport debate revisited and its

archaeological implications. Journal of Archaeological Science, vol. 25: 405–424.

Moncel M.-H., Pleurdeau D., Pinhasi P., Mgeladze A., Yeshurun R., Tushubramisvili N., Agapishvili T., Jennings R., Higham T., Lordkipanidze D. (in press)New insights into the Lower and Middle Palaeolithic in

Georgia. In V.P. Liubin Festschrift.Moncel M.-H., Pleurdeau D., Tushabramishvili N., Yeshurun R., Agapishvili T., Pinhasi R., Higham T.F.G. 2013Preliminary results from the new excavations of the Middle

and Upper Palaeolithic levels at Ortvale Klde-north chamber (South Caucasus Georgia). Quaternary International, vol. 316: 3–13.

Olsen S.J. 1960Post-cranial Skeletal Characters of Bison and Bos. Cambridge,

MA: Cambridge Univ. Press. (Papers of the Peabody Museum of Archaeology and Ethnology; vol. XXXV, No. 4).

Pales L., Garcia M.A. 1981Atlas ostéologique pour servir à l’identification des

mammifères du quaternaire. Paris: CNRS.


Pinhasi R., Higham T.F., Golovanova L.V., Doronichev V.B. 2011 Revised age of late Neanderthal occupation and the end of

the Middle Paleolithic in the northern Caucasus. Proceedings of the National Academy of Sciences, vol. 108: 8611–8616.

Pinhasi R., Nioradze M., Tushabramishvili N., Lordkipanidze D., Pleurdeau D., Moncel M.-H., Adler D.S., Stringer S., Higham T.F.G. 2012New chronology for the Middle Palaeolithic of the southern

Caucasus suggests early demise of Neanderthals in this region. Journal of Human Evolution, vol. 63: 770–780.

Pleurdeau D., Tushabramishvili N., Nioradze M., de Lumley H., Lordkipanidze D. 2007Les industries lithiques du Paléolithique moyen de Géorgie.

L’Anthropologie, vol. 111: 400–431.Schmid E. 1972 Atlas of Animal Bones for Prehistorians, Archaeologists

and Quaternary Geologists. Amsterdam: Elsevier Publishing Company.

Stiner M.C. 1990The use of mortality patterns in archaeological studies of

hominid predatory adaptations. Journal of Anthropological Archaeology, vol. 9: 305–351.

Tushabramishvili N., Pleurdeau D., Moncel M.H., Agapishvili T., Vekua A., Bukhsianidze M., Maureille B., Muskhelishvili A., Mshvildadze M., Kapanadze N., Lordkipanidze D. 2012Human remains from a new Upper Pleistocene sequence in

Bondi Cave (Western Georgia). Journal of Human Evolution, vol. 62: 179–185.

Vekua A., Lordkipanidze D. 1998 The Pleistocene paleoenvironment of the Transcaucasus.

Quaternaire, vol. 9: 261–266.Villa P., Mahieu E. 1991Breakage patterns of human long bones. Journal of Human

Evolution, vol. 21: 27–48. Zeder M.A., Lapham H.A. 2010Assessing the reliability of criteria used to identify

postcranial bones in sheep, Ovis, and goats, Capra. Journal of Archaeological Science, vol. 37: 2887–2905.

Received January 20, 2014.Received in revised form January 24, 2014.

ZOOARCHAEOLOGY AND TAPHONOMY OF THE MIDDLE-UPPER PALEOLITHIC IN BONDI CAVE, REPUBLIC OF GEORGIA

Documents

Transcript of ZOOARCHAEOLOGY AND TAPHONOMY OF THE MIDDLE-UPPER PALEOLITHIC IN BONDI CAVE, REPUBLIC OF GEORGIA