The emergence of bone-working and ornamental art in the Caucasian Upper Palaeolithic.

The emergence of bone-working and ornamental art in the Caucasian Upper Palaeolithic. [ILLUSTRATION OMITTED] The Upper Palaeolithic in the Caucasus: old views and recentdiscoveries The place of the Caucasian Upper Palaeolithic (UP) has been underdiscussion since the 1930s. Zamiamin (1935) described it as generallyAurignacian in character and drew broad analogies with theMediterranean--from North Africa and Italy, across to Syria andPalestine. Later, Formozov (1959) argued for more geographically limitedaffinities with the nearest regions, particularly Syria, Palestine andIraq. Bader (1966) noted similarities between the UP of the westernCaucasus and the Baradostian and Zarzian industries of the ZagrosMountains (Iran). Amirkhanov (1986) emphasised theAurignacian-Perigordian character and also the rarity of bone tools as adistinguishing feature of the Caucasian UP. More recently, Kozlowski(1998) has argued for a bilinear development of the Early UpperPalaeolithic (EUP) in Georgia, and linked the Georgian EUP industries tothe Early Ahmarian and Aurignacian of the Levant. The chronology and homogeneity of the Caucasian UP has also been anissue of debate, and stratigraphic inconsistencies have affected ourunderstanding of both the timing and typological nature of thisindustry. To address these problems, Liubin (1989) carried out anoverview and critical analysis of key UP stratified sequences in theCaucasus. In addition, the first absolute dating of some Georgian UPcontexts was done in the 1990s (Nioradze & Otte 2000). Amirkhanov(1994) reviewed the UP materials from earlier excavations and concludedthat many of these artefact assemblages were probably of mixedMousterian and UP provenance. He attempted to divide the Caucasian UPinto two stages, based on climate-chronological data and peculiaritiesof lithic industries. Earlier, Meshveliani (1986) had concluded that UPmaterials from older excavations in Georgia were probably mixed witholder material. The new chronology Beginning in the mid 1990s, new data has begun to emerge, changingour understanding of the character and origin of the arrival andbehaviour of early Modern humans in the Caucasus. Modern excavationtechniques, including total sediment water screening and an expandedseries of absolute dates (Tables 1 & 2) from three recentlyexcavated sites (Figure 1)--Mezmaiskaya Cave (north-western Caucasus,Russia), Dzudzuana Cave and Ortvale Klde Rockshelter (southern Caucasus,Georgia)--have revolutionised the perception of the UP in this region,with important implications for our understanding of the development andspread of the EUP in Eurasia (for details see Bar-Yosef et al. 2006,Golovanova et al. 2006 and Adler et al. 2008). The EUP industries at Dzudzuana and Ortvale Klde are similar. AtDzuduana, the EUP Layer D dates are between 26 and 32 ka UP (uncal), andthe lowest EUP layers are not yet dated (Meshveliani et al. 2004;Bar-Yosef et al. 2006). At Ortvale Klde, the EUP begins at 38 100 ka UP([+ or -] 935 uncal) in Layer 4d, and continues to between 23 500 and 27ka BP (uncal) in Layer 4b (although Adler et al. (2008: 14) note thatthe single early estimate for Layer 4d 'must be treated withcaution'). Adler et al. (2008) calibrated AMS results from OrtvaleKlde using the CalPal_2007_HULU method (and all calibrated datesreported here use this method). These authors set the calendric agelimits of the EUP in the southern Caucasus between ~30 ka UP (Layer 4b)and ~39 500 ka UP (Layer 4c). At both sites, the EUP sequence ends at~26-28 ka BP (cal), the Late Upper Palaeolithic (LUP) dates between26-28 and 25 ka BP (cal), and at Dzudzuana (Unit B), the Epipalaeolithic(EPP) dates between 11 500 ka BP ([+ or -] 75 uncal) and 13 830 BP ([+or -] 100 uncal). Mezmaiskaya Cave in the north-western Caucasus is well known as aMiddle Palaeolithic site (Golovanova et al. 1998, 1999; Golovanova &Doronichev 2003). In 1997, ten years after excavations had begun, threeEUP layers (1C, 1B & 1A) were first recovered here (Golovanova etal. 2006). According to radiocarbon dates on Mezmaiskaya Layer 1C, theEUP begins in the northern Caucasus by ~33 ka BP (uncal or ~37 ka cal),and possibly as early as 36 100 ka BP ([+ or -] 2300 uncal or 40 128 [+or -] 2317 cal). This fits into the range given for Ortvale Klde. In2001, two EPP layers (1-4 and 1-3) were excavated at Mezmaiskaya. Layer1-4 (the lower EPP level) represents an eroded deposit apparentlycorresponding to the Last Glacial Maximum. The age of the youngest EPPlayer (1-3) at Mezmaiskaya falls between 13 and 14 ka BP (uncal) andcoincides with the EPP at Dzudzuana (Tables 1 & 2). Thus, we nowhave a better understanding of the timing and pace of the UP sequence inthe Caucasus. [FIGURE 1 OMITTED] The evidence from Mezmaiskaya, Dzudzuana and Ortvale Klde alsofundamentally changes our understanding of the origin and industrialpeculiarities of the Caucasian UP. All three sites lack a period oftransition from the Middle to the Upper Palaeolithic, and instead showthe rather abrupt appearance of a fully-developed EUP lithic and boneindustry suggesting the arrival of a new biological population (i.e.Homo sapiens) and population replacement of local Neanderthals byanatomically Modern humans (Golovanova 2000; Bar-Yosef et al. 2006).Among EUP assemblages in neighbouring regions, the EUP industries atMezmaiskaya, Dzudzuana and Ortvale Klde are most similar to the EarlyAhmarian of the Levant in its highly developed blade/bladelet technologyand tool sets, while differing from the Aurignacian (Meshveliani et al.2004; Bar-Yosef et al. 2006; Golovanova et al. 2006, 2007). Like theCaucasian EUP, the Levantine Early Ahmarian is dated between ~40 and 33ka BP (Belfer-Cohen & Goring-Morris 2007). The recent data show thatit is highly unlikely that the Caucasian EUP industries developed intothe classic Aurignacian of Europe. New features of the Upper Palaeolithic in the Caucasus: bone toolsand ornaments Several recent articles discuss the newly published UP data, mostlyin the context of the Middle to Upper Palaeolithic transition, theorigin of EUP, and the chronology of these events in the Caucasus(Bar-Yosef et al. 2006; Golovanova et al. 2006, 2010; Adler et al.2008). A significant aspect of the recent research, little discussed todate, is the discovery of a rich and varied set of bone tools, decoratedbone artefacts and personal ornaments. The recent UP excavations in theCaucasus showed that fully-developed bone-processing technologiesappeared abruptly in the EUP, constituting a revolution in materialculture, and were further developed throughout this epoch. Moreover,these bone tools and personal ornaments, together with new manufacturingtechniques using novel materials (i.e. bones, teeth and shells)emphasise the cultural originality of the Upper Palaeolithic industriesin the Caucasus. Here we present the first detailed publication of these bone toolsand ornaments from the UP in Mezmaiskaya Cave. We report the details ofthis inventory from our excavation at Mezmaiskaya (Table 3), and providea comparative analysis between EUP and EPP levels at this site, as wellas a comparison with other UP occurrences in the Caucasus. A brief review of the bone implements Points are the most significant group of bone tools in the EUPlayers at Mezmaiskaya Cave. The majority of points (11 items) were foundin the earliest EUP layer, 1A. Six of these artefacts are pointfragments and vary from 12 to 45mm in length (Figure 2: no. 2, Figure 3:nos. 7 & 8). Five of them have rounded cross-sections. Among themore complete points, all four are double points with roundedcross-sections, a maximal thickness at the middle of a tool, and lengthsvarying from 82 to 62mm (Figure 3: no. 6). In addition, we recoveredanother point type: a small (23 x 5 x 2mm) tip fragment of a point witha flattened cross-section (Figure 2: no. 1, Figure 3: no. 3). In the overlying EUP layers, 1B and 1A, only point fragments arefound. Among four such fragments in Layer 1, three are small (20, 25 and38mm) tip fragments with rounded cross-sections (Figure 3: nos. 4 &5). The fourth tip fragment is 29mm in length and has a flat-convexcross-section. In Layer 1B, four point fragments between 27 and 34mm inlength have rounded cross-sections. Three of these are tip fragments andone is a shaft portion (Figure 4: nos. 14-16). Layer 1B also contained avery large (120 x 14 x 10mm) point fragment, broken at both ends, withan oval cross-section (Figure 3: no. 9). In the EPP layers atMezmaiskaya, points are also the dominant group of bone tools, and theyare all fragmented. Layer 1-4 included five point fragments with roundedcross-sections (Figure 3: nos. 10-11). Four were tip fragments (14, 18,23 and 33mm in length) and one was a shaft fragment (30mm). One point(48 x 5 x 2mm) made from mammoth tusk has a flat-convex cross-section(Figure 4: no. 8). In Layer 1-3, two point fragments (13 and 26mm) haverounded cross-sections, and three fragments (between 39 and 50mm inlength) have flat-convex cross-sections. One of these has two engravedincisions converging at the tip (Figure 3: no. 17). [FIGURE 2 OMITTED] Bone needles are mostly found in the EUP layers 1A and 1B. In Layer1C, only two fragments of needle were found--an eyed needle fragmentwith a fiat base (13 x 3 x 1mm) and part of a rounded needle (16 x 3 x3mm) (Figure 4: no. 1). In Layer 1, three needles (42 x 4 x 2mm, 34 x 4x 2mm and 22 x 3 x 2mm) have flat bases and rounded tips (Figure 4: nos.5 & 6) and one needle (31 x 4 x 3mm) has a rounded base. In Layer1A, all three needles (40 x 2 x 2mm, 36 x 2 x 2mm and 37 x 3 x 3mm) anda needle fragment (23 x 3 x 3mm) have rounded cross-sections (Figure 4:nos. 3 & 4). Only one fragment of a rounded needle (26 x 3 x 3mm)was recovered from the EPP Layer 1-4, and three needle fragments werefound in the overlying Layer 1-3--one with a flat base (31 x 4 x 1mm)and two converging engraved lines (Figure 4: no. 2), and two roundedneedle fragments (23 x 4 x 4mm and 20 x 4 x 3mm). Bone awls are most frequently found in EUP layers 1C and 1A. InLayer 1C, five awls have massive cross-sections (Figure 2: nos. 3 &4, Figure 3: nos. 1 & 2). These are made from large fragments oflong bones or ribs (142, 129, 110, 111 and 63mm in length) and preserveoriginal bone morphology with one end modified into a massive roundedpoint. One massive awl has spots of ochre. The shaped end of a micro-awlfrom Layer 1C (Figure 4: no. 7) is similar to the needles in dimension(32 x 5 x 4mm), while retaining a massive base. One long, thin awl (78 x11 x 11 mm) made from bird bone was found in EUP Layer 1B (Figure 3: no.16, Figure 5: no. 3). In Layer 1A, there are four thin awls (81, 65, 57and 48mm in length) made from long bone fragments (Figure 3: nos. 12-15,Figure 5: nos. 2-4) and two massive awls (54mm in length), one of whichis fragmented (Figure 4: nos. 10 & 11, Figure 5: no. 1). There are few bone awls in the EPP layers 1-4 and 1-3. A long(70mm), massive awl made from a long bone was recovered from Layer 1-4(Figure 4: no. 13). Two thin awls--one complete (78mm) and anotherfragmented--were recovered from Layer 1-3 (Figure 4: no. 9). A fragment of a bonepolisher/abrader (102 x 15 x 5mm) from Layer 1Chas a heavily polished external bone surface and an awl-like, partiallyfragmented point (Figure 2: no. 5). A fragment of a polisher (46 x 17 x7mm) from Layer 1B is heavily polished and rounded on the plane end. Asimilar polisher (44 x 18 x 6mm) was found in Layer 1-3. [FIGURE 3 OMITTED] [FIGURE 4 OMITTED] Other bone tools. A large (195 x 44 x 14mm) spatula-like bone toolwas found in EUP layer 1C. This artefact has small facets and polishingon the transverse end and on the distal part of the lateral side,probably resulting from use. Another tool from this level resemblesflint chisel-like tools (Figure 4: no. 17). This was made from a longbone fragment (118 x 28 x 12mm), and its wide transverse end has aseries of elongated facets forming a straight edge. In Layer 1B, a large(115mm) fragment of the mammoth tusk has wear that indicates it was usedas a polisher, and there are two small rib fragments with polished andscraped surfaces. Large amounts of bone flakes and chips were recoveredfrom all UP layers at Mezmaiskaya. Most of the tooth pendants from the EUP layers at Mezmaiskaya aremade from caprid incisors. However, two pendants made from red deer milkteeth were found in Layer 1A (Figure 6: no. 1). A pendant from EUP layer1C is unique in having an eye produced by a cutting action (Figure 6:no. 2). In Layer 1A, a slightly fragmented pendant has an eye drilledfrom one side (Figure 6: no. 1), and a second pendant has a deeplyincised groove surrounding the base of a broken eye (Figure 6: no. 3).There are significant differences in manufacture technique between theEUP and EPP. All pendants from EPP layers 1-4 and 1-3 have eyes drilledfrom two sides in a biconical technique not found in the EUP layers(Figure 6: nos. 4 & 5). Additionally, pendants from Layer 1-3 haveV-shaped notches in tooth crowns (Figure 6: no. 4). [FIGURE 5 OMITTED] Two flat mammoth ivory strip beads with roughly square outlines (10x 8 x 1mm and 9 x 7 x 1mm) were found in Layer 1A. The holes in bothhave similar diameter (3 and 4mm), and are drilled from one side (Figure6: nos. 6 & 7). One more strip bead (11 x 6 x 1 mm) found in mixedUP sediments of Lens Y is similar to the ivory beads in form, but ismade from bone and has an incised, rather than drilled, hole. A needle case with fine geometric ornamentation is a unique findfor the Caucasian UP. This artefact was found in the upper part of Layer1B, in the zone of contact with Layer 1A. The needle case is made from asmall long bone (48 x 10 x 10mm) and is broken on one end. On thepreserved area of the artefact, there are four narrow relief engravedplatens, each 3mm wide and about 3mm apart. Each platen is decoratedwith a varying number of slanting daggers. Heavy polish on the platenlocated on one preserved edge of the needle case is probably the resultof use. The remaining surface of the needle case is covered with deepparallel concentric grooves engraved at 3mm intervals. Between the firstand second grooves there are small (1mm in length) cross hatch marks atabout 1mm intervals (Figure 7: nos. 1a & b). [FIGURE 6 OMITTED] [FIGURE 7 OMITTED] [FIGURE 8 OMITTED] Other ornamented bones include several fragmented bone artefactswith simple geometric ornamentation. One of them is a fragment of flatbone (84 x 40 x 11mm) from Layer 1A. It has a slanting gridornamentation formed by shallow cut marks in one end (Figure 7, nos. 3a& b). Another piece originates also from Layer 1A. It is a fragmentof burned bone (37 x 39 x 7mm) with three parallel-engraved incisionsgoing through the whole bone surface (Figure 7: no. 2). A fragment ofornamented long bone (108 x 15 x 7mm) is found in Layer 1-3. It has arhombus-like feature created by convergent engraved incisions on theexternal surface of the bone (Figure 7: no. 4). In addition, small bonefragments with short parallel cut marks are found in many layers. There are a few other modified bones of interest from Mezmaiskaya.In EUP layer 1C, an herbivore phalanx with an incompletely drilled holein the mid-shaft area has been made into a tube by the completeclearance of trabeculae from the interior cavity. A scraped and heavilypolished bone tube made from the shaft portion of a small (61 x 4 x 3mm)long bone was found in EPP layer 1-3. Shell beads (species identified by A. Chepaliga). A few marinegastropod shells (all commonly known from the Black Sea) are found inthe EUP layers (1A, 1B & 1C) only. All EUP layers include the small(7 to 13mm in diameter) marine shells from Trophon muricatus (Mtd.) orNassarius reticulates (L). Four such shells were recovered from each oflayers 1B and 1C, and six were found in Layer 1A. Only two marine shellsfrom layers 1A and 1B have holes (Figure 6: nos. 9 & 10).Additionally, a pendant made from the marine shell of a Cyclope neritea(Linne), which seems to have a drilled hole, was found in Layer 1A(Figure 6: no. 8). The shells of small terrestrial gastropods are found in the EPPlayers (1-3 and 1-4) at Mezmaiskaya. The terrestrial taxa includeHelicidae gen., Succinidae gen. (Succinaea sp.), and Pupillidae gen.,and most of these shells are broken into small or very small fragments.However, there are 22 unbroken shells with perforations--9 pieces inLayer 1-3 and 13 pieces in Layer 1-4. In EPP layers 1-3 and 1-4, thereare two concentrations of perforated shells in the excavated area. InLayer 1-3, all perforated shells were found in adjacent quadrants O-14(5 pieces) and O-13 (4 pieces). In the underlying Layer 1-4, aconcentration of perforated shells (Figure 6: no. 11) occurred inquadrants O-14 (8 pieces) and H-14 (5 pieces). However, these quadrantsalso show evidence of an intrusive pit from Layer 1-3. This pitcontinues into part of the EUP levels. Although a few perforated shellswere recovered from the bottom part of the pit within the EUP layers,these finds are clearly intrusive materials into the EUP deposits. [FIGURE 9 OMITTED] Comparative examples In the EUP sequence at Ortvale Klde, Adler et al. (2006: 10, figs.11a and b) also describe bone tools. They report three fragmentedbone/antler points, two polished bone/antler abraders, an awl and apolished bone with a series of parallel linear incisions (Figure 8: nos.1 & 2). In Dzudzuana Cave, bone tools were found in each UP unit(195 items in total), including simple points, awls, pendants and somepieces with geometric ornamentation (Meshveliani et al. 2004; Bar-Yosefet al. 2006; Figure 8: nos. 3-6). Most of the pendants from Dzudzuanaare made from caprid teeth (T. Meshveliani pers. comm.). Although bone tools were found in earlier excavations of UP sitesin the southern Caucasus, such finds were rare. Thin and massive boneawls, as well as complete double points with rounded cross-sections wererecovered at Samertskhle Klde, Sakajia, Gvardjilas Klde and Satsurblia(Nioradze & Otte 2000; Figures 9, 10, 11: nos. 1-8, 12: nos. 7-12,13: no. 2). Pendants made from ungulate teeth (Figure 12: nos. 3, 5& 6) and stone (Figure 12: nos. 1, 2 & 4) were found inSatsurblia Cave. A bone pendant with an incised groove around the eye(Figure 11: no. 10) and a fragment of a decorated bone tool (Figure 11:no. 9) were found in Gvardjilas Klde Cave. Bone pendants with geometricornamentation (Figure 13: nos. 3 & 4) and an ornamented bone needle(Figure 13: no. 1) were reported from Sakajia Cave. Most of these siteshave been dated to the final stage of the UP (Epipalaeolithic), from 15to 10 ka BP (uncal, Nioradze & Otte 2000). [FIGURE 10 OMITTED] In the northern Caucasus, bone artefacts from earlier excavationsthat are comparable to the finds reported here are derived from poorlydated contexts. Bone points are known (but not radiometrically dated)from Layer 1 in Gubs Rockshelter 1 (Amirkhanov 1986). UP excavations atSatanai Rockshelter uncovered bone awls (Figure 14: nos. 1, 2 & 6),pendants from horse incisors (Figure 14: no. 3), a bone polisher, andlarge (14-17cm in length) bone double points with flat-convexcross-sections (Figure 14: nos. 4 & 5) that have no known analogies(Amirkhanov 1986.). At Satanai, the lack of both secure dates and adetailed stratigraphy within the thick (up to 1.5m) Upper Palaeolithicsediments, make it difficult to securely correlate these finds withMezmaiskaya Cave. By contrast, preliminary results from recentexcavations at Korotkaya Cave indicate that bone awls are present inUpper Palaeolithic layers, dating from 24 ka BP (uncal) and earlier(Blajko 2001). Conclusions These recent findings significantly contribute to our understandingof UP technology within the northern Caucasus, and the relationship ofthe Caucasus to the surrounding Palaeolithic landscape. The UPassemblages from Mezmaiskaya Cave not only illustrate the abruptintroduction of symbolic and ornamental technology but subsequentchanges in bone tool production and symbolic behaviour from the EUPthrough the EPP in the Caucasus. Within pendant production, for example,there appears to be a shift from incised holes (Layer 1C) to holesdrilled from one side (Layer 1A), to bi-conically drilled holes (layers1-4 and 1-3). Among needles, there is a shift from flat bases (Layer1C), to fully rounded cross-sections similar to thin awls (Layer 1A).Ornamentation of bone artefacts is absent in the earliest EUP (Layer1C), but appears in the later EUP (layers 1B-1 and 1A), and in the EPP(Layer 1-3). This last result is consistent with data from the Georgiansites. [FIGURE 11 OMITTED] [FIGURE 12 OMITTED] [FIGURE 13 OMITTED] At Dzudzuana and Ortvale Klde, bone implements with geometricornamentation are not reported from the oldest EUP levels. In general,at Mezmaiskaya, it appears that the bone tool industry and personalornaments become more variable from the EUP to the EPP. In addition,there is a shift in some raw material sources at Mezmaiskaya from marinegastropod beads/pendants in the EUP to the terrestrial shell pendants inthe terminal Palaeolithic. Within the broader context of the full Caucasian region, new datafrom Dzudzuana, Ortvale Klde, and particularly Mezmaiskaya change ourunderstanding of the timing, origin and character of the UP (Meshvlianiet al. 2004; Adler et al. 2006, 2008; Bar-Yosef et al. 2006; Golovanovaet al. 2006, 2010). These data allow us to draw several importantconclusions. Firstly, the appearance of the EUP in the Caucasus can now be moresecurely dated between ~40 and ~35 ka cal BP (in Ortvale Klde andMezmaiskaya; see Table 2). This suggests that EUP technology moved intothe region after the Campanian Ignimbrite (CI) eruption from thePhlegrean Fields in southern Italy at ~40 ka cal BP (Golovanova et al.2010). Secondly, the new data demonstrate the abruptness of thechronological break between the end of the Middle Palaeolithic and thebeginning of the UP in the Caucasus (Bar-Yosef et al. 2006; Golovanovaet al. 2006; Adler et al. 2008). In Mezmaiskaya Cave, this break is alsocorrelated with large contemporaneous volcanic eruptions in Italy (CIeruption) and Caucasus. These eruptions seem to be a major naturalcatastrophic event that may have completely destroyed the ecologicalniche of certain populations of Neanderthals and caused their rapiddecline (Golovanova et al. 2010). Thirdly, it is now clear that the EUP industry appears in theCaucasus as a fully developed technological tradition. Comparisons ofthe Caucasian assemblages with the EUP in nearby regions reveal somesimilarities. In particular, the earliest fully-fledged UP industry inthe Levant, the Early Ahmarian, dating from ~40 to ~33 ka BP(Belfer-Cohen & Goring-Morris 2007), appears to be the most similarto the Caucasian EUP industry. We conclude that the time period between40 and 30 ka BP was significant for the dispersal of essentially new EUPmicro-laminar (bladelet) industries distinguishing by developed bladeand bladelet technologies, together with numerous and variable bladelettools across a broad region including the Zagros, Levant and Caucasus(Golovanova et al. 2007). [FIGURE 14 OMITTED] While there are strong similarities with non-Caucasus industries,the inter-assemblage variability within the Caucasus and neighbouringregions is also becoming obvious. For example, the EUP industries fromMezmaiskaya Cave contain no typical el-Wad points with the fine lateralretouch characteristic of the Levantine Early Ahmarian assemblages(Belfer-Cohen & Goring-Morris 2007). On the contrary, typicalGravettian points with straight backs made by blunted retouch are themost common point type in Mezmaiskaya. Likewise, various bone tools,which are poorly represented in the Early Ahmarian (possibly due to badbone preservation), are characteristic of the EUP of Mezmaiskaya. Our fourth significant observation is that the EPP industries inthe Caucasus (dating from ~16 to ~11 500 ka uncal BP) are generallycharacterised by highly developed bladelet technologies and largequantities of backed bladelets in tool sets. Additionally, geometriclithics (segments, trapezes and triangles) and shouldered points arefound in many of these industries. It is worth noting that the EPP layer1-3 at Mezmaiskaya and Unit B at Dzudzuana are separated from theearlier UP levels by an erosional hiatus corresponding to the LastGlacial Maximum (Table 2). The EPP in Mezmaiskaya is similar to EPPindustries in Georgia and in the Southern Russian Plain (i.e. atKamennaya Balka) (Leonova et al. 2006). Finally, the recent data show that the Caucasian Upper Palaeolithicincludes a wide assortment of bone implements, decorated bone artefactsand personal ornaments, as well as new techniques of bone processing(e.g. abrasion, drilling). Among them, pendants made from ungulate teethand implements with geometric ornamentation are both especiallycharacteristic of the Upper Palaeolithic in the Caucasus. In addition,the split-base bone points so typical of the Aurignacian are not foundin this region. Thus, the recent UP excavations in the Caucasus showthat fully developed organic technologies and decorations abruptlyappeared in the EUP, constituting a revolution in material culture.These bone tools and ornamented items are the key features of culturaloriginality throughout the all UP in Caucasus, alongside the distinctivelithic industry. Acknowledgements The authors wish to thank the L.S.B. Leakey Foundation and theWenner-Gren Foundation for the long-term financial support of theexcavations at Mezmaiskaya Cave. We are grateful to the Management forPreservation, Restoration and Exploitation of History-Cultural Heritageof the Culture Department of Krasnodar Krai and Krasnodar StateHistory-Archaeological Museum-Reservation for their help in organisingthe fieldwork at Mezmaiskaya Cave. We appreciate support provided by theFulbright Scholar Exchange Program, the Glassman Holland Foundation, theW.F. Albright Institute of Archaeological Research and the NationalGeographic Society for our research. We thank O. Bar-Yosef, A.Belfer-Cohen, H. Dibble T. Meshveliani, M. Nioradze and N.Tushabramishvili for helpful discussions and V. Plicht and G. Burr fortheir aid in dating of the Upper Palaeolithic in Mezmaiskaya Cave. Received: 29 December 2008; Revised: 23 September 2009; Accepted:29 September 2009 References ADLER, D.S., A. BELFER-COHEN & O. BAR-YOSEF. 2006. Between arock and a hard place: Neanderthal-Modern human interactions in thesouthern Caucasus, in N.J. Conard (ed.) When Neanderthals and Modernhumans met: 89-118. Tubingen: Kerns. ADLER, D.S., O. BAR-YOSEF, A. BELFER-COHEN, N. TUSHABRAMISHVILI, E.BOARETTO, N. MERCIER, H. VALLADAS & W.J. RINK. 2008. Dating thedemise: Neanderthal extinction and the establishment of Modern humans inthe southern Caucasus. Journal of Human Evolution 55:817-33. AMIRKHANOV, H.A. 1986. Verkhniy paleolit Prikuban'ya [TheUpper Palaeolithic of the Kuban River]. Moscow: Nauka. --1994. K probleme evolutcii i periodizatcii verhnego paleolitaZapadnogo Kavkaza [To a problem of evolution and periodisation of theUpper Palaeolithic of the western Caucasus]. Rossiiskaya Arheologiya 4:9-24. BADER, N.O. 1966. Razlichyia mejdu verkhnepaleoliticheskimikul'turami Zakavkaz'ya i Blijnego Vostoka [Differences amongthe Upper Palaeolithic cultures of the Transcaucasus and Near East], inN.Y.A. Merpert & P.M. Kojin (ed.) Arheologiya Starogo i Novogo Sveta[Archaeology of the Old World and New World]: 135-43. Moscow: Nauka. BAR-YOSEF, O., A. BELFER-COHEN & D.S. ADLER. 2006. Theimplications of the Middle-Upper Paleolithic chronological boundary inthe Caucasus to Eurasian prehistory. Anthropologie 44(1): 49-60. BELFER-COHEN, A. & N. GORING-MORRIS. 2007. The shift from theMiddle Palaeolithic to the Upper Palaeolithic: Levantine perspectives,in M. Camps & C. Szmidt (ed.) The Mediterranean from 50,000 to25,000 BP: turning points and new directions: 87-97. Oxford: Oxbow. BLAJKO, A.V. 2001. Issledovanie Korotkoi pesheri na severo-zapadnomKavkaze [The study of Korotkaya Cave in the north-western Caucasus].Archaeological Discoveries 2000:121-2. FORMOZOV, A.A. 1959. Etnokuhurnie oblasti na territorii Evropeiskoichasti SSSR v kamennom veke [Ethno-cultural areas in the territory ofthe European part of the USSR in the Stone Age]. Moscow: Nauka. GOLOVANOVA, L.V. 2000. Rubej srednego i pozdnego paleolita nasevarnom Kavkaze [The Middle-Upper Palaeolithic interface in thenorthern Caucasus]. Stratum plus 1:158-77. GOLOVANOVA, L.V. & V.B. DORONICHEV. 2003. The MiddlePaleolithic of the Caucasus. Journal of World Prehistory 17(1): 71-140. GOLOVANOVA, L.V., J.F. HOFFECKER, S.A. NESMEYANOV, G.M.LEVKOVSKAYA, V.M. KHARITONOV, G.P. ROMANOVA & YU. SVEJENCEVE. 1998.Un site micoquien Est-Europeen du caucase du Nord (Resultatspre1iminaires de l'etude de la grotte Mezmaiskaya, les fouilles desannees 1987-1993). L'Anthropologie 102(1): 45-66. GOLOVANOVA, L.V., J.F. HOFFECKER, V.M. KHARITONOV & G.P.ROMANOVA. 1999. Mezmaiskaya cave: a Neanderthal occupation in thenorthern Caucasus. Current Anthropology 40(1): 77-86. GOLOVANOVA, L.V., N.E. CLEGHORN, V.B. DORONICHEV, J.F. HOFFECKER,G.S. BURR & L.D. SULERGIZKIY. 2006. The Early Upper Paleolithic inthe northern Caucasus (new data from Mezmaiskaya Cave, 1997 excavation).Eurasian Prehistory 4(1-2): 43-78. GOLOVANOVA, L.V., V. DORONICHEV & N. CLEGHORN. 2007. Gettingback to basics: a response to Otte 'Comments on Mezmayskaya'.Eurasian Prehistory 5(1): 131-6. GOLOVANOVA, L.V., V.B. DORONICHEV, M.A. KULKOVA, N. CLEGHORN &T.V. SAPELKO. 2010. Significance of ecological factors in the Middle toUpper Paleolithic transition. Current Anthropology (accepted). KOZLOWSKI, J. 1998. The Middle and the Early Upper Paleolithicaround the Black Sea, in T. Akazawa, K. Aoki & O. Bar-Yosef (ed.)Neandertals and Modern humans in Western Asia: 461-82. New York: Plenum. LEONOVA, N.B., S.A. NESMEYANOV, E.A. VINOGRADOVA, O.A. VOEIKOVA,M.D. GVOZDOVER, E.V. MIN'KOV, E.A. SPIRIDONOVA & S.A. SYCHEVA.2006. Paleoecologiya Ravninnogo Paleolita (Na primere kompleksaverhnepalealiticheskih stoyanok Kamennaya Balka v SevernomPriazov'e) [The palaeoecology of the Plains Palaeolithic (TheKamennaja Balka Upper Palaeolithic sites north of the Sea of Azov)].Moscow: Nauchniy Mir. LIUBIN, V.P. 1989. Paleolit Kavkaza [Palaeolithic of the Caucasus],in P.I. Boriskovsky (ed.) Paleolit Kavkaza i Severnoi Azii [ThePalaeolithic of the Caucasus and northern Asia]: 9-142. Leningrad:Nauka. MESHVEHANI, T. 1986. O rannem etape verhnego paleolita ZapadnoyGruzii [About the early stage of the Upper Palaeolithic of westernGeorgia]. Trudi Gosudarstvennogo Muzeya Gruzii 1986:116-23. MESHVELIANI, T., T.K. DJASHLI, O. BAR-YOSEF & A. BELFER-COHEN.2002. A renewed archaeological expedition in Dzudzuana (2000 results).Academia 3:29-31 (in Georgian). MESHVELIANI, T., O. BAR-YOSEF & A. BELFER-COHEN. 2004. TheUpper Paleolithic in western Georgia, in P.J. Brantingham, S.L. Kuhn& K.W. Kerry (ed.) The Early Upper Paleolithic beyond WesternEurope: 129-43. Berkeley (CA): University of California Press. NIORADZE, M.G. & M. OTTE. 2000. Paleolithique superieur deGeorgie. L'Anthropologie 104: 265-300. ZAMIATNIN, S.N. 1935. Novie dannie po paleolitu Zakavkaz'ya[New data on the Palaeolithic in the Transcaucasus]. Soviet Ethnography2:116-23. Liubov V. Golovanova (1), Vladimir B. Doronichev (1) & Naomi E.Cleghorn (2) (1) Laboratory of Prehistory, 14 Linia 3-11, St Petersburg 190034,Russia (Email: lprehist@peterstar.ru) (2) Department of Integrative Biology, 3060 Valley Life SciencesBuilding, University of California Berkeley, Berkeley, CA 94720, USA(Email: cleghorn@berkeley.edu)Table 1. Chronometric results from Mezmaiskaya Cave, DzudzuanaCave and Ortvale We Rockshelter. Data calibrated with CalPal-online(CalCurve: CalPal_2007_HULU). All data are taken from Golovanova etal. 1999, 2006, Bar-Yosef et al. 2006 and Adler et al. 2008.Layer Lab no. Method Material Mezmaiskaya--EPP sequence1-3 GrA--25965 AMS bone GIN--12900 [sup.14]C bone1-4 GIN--12901 [sup.14]C bone GrA--25933 AMS bone Mezmaiskaya--EUP sequence1A AA-41855 AMS bone CURL-5757 AMS wood charcoal1B CURL-5756 AMS wood charcoal CURL-5759 AMS wood charcoal Beta-113536 AMS wood charcoal CURL-5762 AMS wood charcoal1C CURL-5760 AMS wood charcoal CURL-5761 AMS wood charcoal GIN-10946 [sup.14]C bone AA-41856 AMS bone Dzudzuana--EPP sequenceUnit B RTA-3282 AMS RTA-3278 AMS Dzudzuana--LUP sequenceUnit C RTA-3434 AMS bone RTA-3433 AMS bone RTA-3435 AMS bone RTT-4339 AMS RTT 4341 AMS RTA-3823 AMS bone Dzudzuana--EUP sequenceUnit D RTT-4336 AMS RTT-4340 AMS RTA-3436 AMS RTA-3437 AMS RTT-4338 AMS RTT 4701 AMS RTA-3438 AMS Ortvale Klde--LUP sequence2 RTT-3824 AMS bone3 AA-38195 AMS charcoal AA-38196 AMS charcoal Ortvale Klde--EUP sequence4b RTT-3964 AMS charcoal RTT-3825 AMS bone AA-38193 AMS charcoal AA-38197 AMS charcoal RTT-4207 AMS charcoal RTT-4210 AMS charcoal4c RTT-4209 AMS charcoal RTT-4208 AMS charcoal RTT-4211 AMS charcoal AA-45865 AMS charcoal RTT-4214 AMS charcoal RTT-4213 AMS charcoal AA-45864 AMS charcoal RTT-4212 AMS charcoal4d RTT-4725 AMS boneLayer Age [sup.14]C BP Weighted mean Mezmaiskaya--EPP sequence1-3 12 960 [+ or -] 60 13 860 [+ or -] 701-4 16 260 [+ or -] 100 21 050 [+ or -] 110/120 Mezmaiskaya--EUP sequence1A 28 510 [+ or -] 850 28 510 [+ or -] 850 32 000 [+ or -] 2501B 32 400 [+ or -] 240 32 278 [+ or -] 138 32 400 [+ or -] 230 32 010 [+ or -] 250 33 000 [+ or -] 260 32 766 [+ or -] 1271C 33 000 [+ or -] 240 33 100 [+ or -] 270 32 900 [+ or -] 900 33 325 [+ or -] 838 36 100 [+ or -] 2300 Dzudzuana--EPP sequenceUnit B 11 500 [+ or -] 75 13 830 [+ or -] 100 Dzudzuana--LUP sequenceUnit C 20 980 [+ or -] 150 21 220 [+ or -] 200 21 930 [+ or -] 190 22 490 [+ or -] 180 23 125 [+ or -] 175 23 240 [+ or -] 200 Dzudzuana--EUP sequenceUnit D 26 320 [+ or -] 260 26 925 [+ or -] 255 27 150 [+ or -] 300 27 400 [+ or -] 300 27 450 [+ or -] 275 32 140 [+ or -] 500 30 350 [+ or -] 400 Ortvale Klde--LUP sequence2 21 170 [+ or -] 140 21 170 [+ or -] 1403 21 580 [+ or -] 230 21 664 [+ or -] 159 21 740 [+ or -] 220 Ortvale Klde--EUP sequence4b 27 000 [+ or -] 260 27 000 [+ or -] 260 23 770 [+ or -] 200 23 770 [+ or -] 200 30 660 [+ or -] 430 30 486 [+ or -] 323 30 260 [+ or -] 490 31 900 [+ or -] 780 32 039 [+ or -] 213 31 700 [+ or -] 5004c 31 800 [+ or -] 400 32 200 [+ or -] 550 32 300 [+ or -] 550 32 510 [+ or -] 530 34 100 [+ or -] 800 34 188 [+ or -] 328 34 600 [+ or -] 600 33 700 [+ or -] 620 34 300 [+ or -] 6504d 38 100 [+ or -] 935 38 100 [+ or -] 935Layer Age cal BP Hulu Weighted mean Mezmaiskaya--EPP sequence1-31-4 Mezmaiskaya--EUP sequence1A 33 023 [+ or -] 765 33 023 [+ or -] 765 36 078 [+ or -] 3831B 36 937 [+ or -] 752 36 372 [+ or -] 310 36 935 [+ or -] 747 36 095 [+ or -] 390 37 473 [+ or -] 675 36 799 [+ or -] 2761C 37 475 [+ or -] 665 37 569 [+ or -] 689 37 583 [+ or -] 1350 38 228 [+ or -] 1166 40 128 [+ or -] 2317 Dzudzuana--EPP sequenceUnit B Dzudzuana--LUP sequenceUnit C Dzudzuana--EUP sequenceUnit D Ortvale Klde--LUP sequence2 25 329 [+ or -] 363 25 329 [+ or -] 3633 25 799 [+ or -] 534 25 885 [+ or -] 394 25 987 [+ or -] 583 Ortvale Klde--EUP sequence4b 31 742 [+ or -] 201 31 742 [+ or -] 201 28 694 [+ or -] 421 28 694 [+ or -] 421 34 889 [+ or -] 451 34 704 [+ or -] 316 34 520 [+ or -] 444 36 380 [+ or -] 1205 36 269 [+ or -] 344 35 785 [+ or -] 7234c 35 825 [+ or -] 632 36 701 [+ or -] 982 36 809 [+ or -] 966 37 020 [+ or -] 897 39 146 [+ or -] 1340 39 475 [+ or -] 559 39 781 [+ or -] 911 38 861 [+ or -] 1506 39 560 [+ or -] 9994d 42 714 [+ or -] 805 42 714 [+ or -] 805Table 2. Chronology of the Upper Palaeolithic at Mezmaiskaya Cave,Dzudzuana Cave and Ortvale Klde Rockshelter. Data calibrated withCalPal-online (CalCurve: CalPal_2007_HULU). All data are takenfrom Golovanova et al. 1999, 2006, Bar-Yosef et al. 2006 andAdler et al. 2008.Mezmaiskaya Cave Age AgeLayer [sup.14]C BP cal. BP Hulu (* Weighted (* Weighted mean) mean)EPP 1-3 12 960 [+ or -] 60 13 860 [+ or -] 70 1-4 16 260 [+ or -] 100 21 050 [+ or -] 120 Last Glacial. MaximumEUP 1A 28 510 [+ or -] 850 33 023 [+ or -] 765 1B 32 278 [+ or -] 138 * 36 372 [+ or -] 310 * 1C 32 766 [+ or -] 127 * 36 799 [+ or -] 276 * 36 100 [+ or -] 2300 40 128 [+ or -] 2317Ortvale Klde Rockshelter AgeLayer [sup.14]C BP cal BP Hulu (* Weighted (* Weighted mean) mean)LUP 2 21 170 [+ or -] 140 25 329 [+ or -] 363 3 21 664 [+ or -] 159 * 25 885 [+ or -] 394 *EUP 4b 23 770 [+ or -] 200 28 694 [+ or -] 421 27 000 [+ or -] 260 31 742 [+ or -] 201 4c 30 486 [+ or -] 323 * 34 704 [+ or -] 316 * 32 039 [+ or -] 213 * 36 269 [+ or -] 344 34 188 [+ or -] 328 * 39 475 [+ or -] 559 * 4d 38 100 [+ or -] 935 42 714 [+ or -] 805Dzudzuana CaveLayer Age 14C BP 11 500 [+ or -] 75EPP EPP Unit B 13 830 [+ or -] 100 Last Glacial Maximum LUP Unit C 20 980 [+ or -] 150 21 220 [+ or -] 200 21 930 [+ or -] 190 22 490 [+ or -] 180 23 125 [+ or -] 175 23 240 [+ or -] 200 EUP Unit D 26 320 [+ or -] 260 26 925 [+ or -] 255 27 150 [+ or -] 300 27 400 [+ or -] 300 27 450 [+ or -] 275 30 350 [+ or -] 400 32 140 [+ or -] 500Table 3. Bone artefacts from the Upper Palaeolithic layers inMezmaiskaya Cave (1997, 2001, 2002, 2004, 2006, 2007 excavations).Layers 1C 1B 1A 1-4 1-3 ToolsPoints 11 4 5 6 5Needles 2 4 4 1 3Awls 6 1 6 1 2Polishers 1 1 -- -- 1Other tools 2 2 -- -- 1 Ornamented items & personal ornamentsPendants 1 -- 2 1 2Stripes--beads -- -- 2 -- --Ornamented needle case -- 1 -- -- --Ornamented bone fragments -- -- 2 -- 1 VariaOther pieces 5 2 2 -- 5