AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 000:000–000 (2007)

Late Pleistocene Human Remains From Wezmeh Cave, Western Iran Erik Trinkaus,1* Fereidoun Biglari,2 Marjan Mashkour,3 Herve´ Monchot,4 Jean-Louis Reyss,5 He´le`ne Rougier,6 Saman Heydari,7 and Kamyar Abdi8 1

Department of Anthropology, Washington University, St. Louis MO 63130-4899 Center for Paleolithic Research, National Museum of Iran, Iran Cultural Heritage and Tourism Organization, P.O. Box 11365/4364, Tehran, Iran 3 De´partement d’Ecologie et Gestion de la Biodiversite´, UMR 5197/CNRS, Baˆtiment d’Anatomie Compare´e, Muse´um national d’Histoire naturelle, 55 rue Buffon, 75005 Paris, France 4 De´partement de Pre´histoire, UMR 5198/CNRS, Institut de Pale´ontologie Humaine, Muse´um national d’Histoire naturelle, 1 rue Rene´ Panhard, 75013 Paris, France 5 Laboratoire des Sciences du Climat et de l’Environnement, UMR CEA-CNRS, Domaine du CNRS, 91198 Gif-sur-Yvette cedex, France 6 Laboratoire d’Anthropologie des Populations du Passe´, UMR 5199 PACEA, Universite´ de Bordeaux 1, Avenue des Faculte´s, 33405 Talence cedex, France 7 ¨ ltere Urgeschichte und Quarta¨ro¨kologie, Institut fu¨r Ur-und Fru¨hgeschichte und Archa¨ologie des Abteilung A Mittelalters, Universita¨t Tu¨bingen, Schloss Hohentu¨bingen, 72070 Tu¨bingen, Germany 8 Department of Anthropology, 6047 Silsby Hall, Dartmouth Colleg, Hanover NH 03755 2

KEY WORDS

human paleontology; modern humans; Neandertals; dentition; carnivores; Zagros

ABSTRACT Paleontological analysis of remains from Wezmeh Cave in western Iran have yielded a Holocene Chalcolithic archeological assemblage, a rich Late Pleistocene carnivore faunal assemblage, and an isolated unerupted human maxillary premolar (P3 or possibly P4). Species representation and U-series dating of faunal teeth place the carnivore assemblage during oxygen isotope stages (OIS) 3 and 2, and noninvasive gamma spectrometry dating of the human premolar places it at least as old as early OIS 2. The human premolar crown mor-

phology is not diagnostic of late archaic versus early modern human affinities, but its buccolingual diameter places it at the upper limits of Late Pleistocene human P3 and P4 dimensions and separate from a terminal Pleistocene regional sample. Wezmeh Cave therefore provides additional Paleolithic human remains from the Zagros Mountains and further documents Late Pleistocene human association with otherwise carnivore-dominated cave assemblages. Am J Phys Anthropol 000:000– 000, 2007. V 2007 Wiley-Liss, Inc.

Despite long term interest and excavation of sites in the Zagros Mountains of western Iran and eastern Iraq, Paleolithic human remains have rarely been found. The principal exception is the 10 partial Neandertal skeletons excavated in Shanidar Cave (Iraq) between 1953 and 1960 (Solecki, 1963; Trinkaus, 1983; Cowgill et al., 2007). Of the Middle Paleolithic fragmentary remains from Bisitun and Tamtama caves (Iran) excavated by Coon (1951), only the Bisitun 1 radial diaphysis is human (Trinkaus, 2006; Trinkaus and Biglari, 2006). Fragments of earlier Upper Paleolithic humans have also been noted from Gar Arjeneh (Iran) but are undescribed (Hole and Flannery, 1967). It is in this context that we report here an isolated human premolar from Wezmeh Cave, Iran, associated with a rich Late Pleistocene faunal assemblage dominated by carnivores.

of the Qazivand Mountains, about 60 m above the valley floor on a steep slope with a dip of ca. 368. From the entrance, for 12 m back, the cave is relatively straight. At ca. 15 m from its mouth, the cave makes a sharp turn towards the west and extends for about 5 m in which the height of the ceiling gradually increases. The floor of this section also has a 138 slope downwards, which led to a large pit (ca. 4 3 2 m2) excavated by clandestine diggers. This section of the cave extends for 5 m and then makes a turn towards the southwest and extends for another 7 m with a steep slope downwards. The cave was discovered in 1999 during archeological survey of the Islamabad Plain, when it was realized that

WEZMEH CAVE Wezmeh Cave is a small cave (Figs. 1–3), with a surface area of ca. 45 m2, a depth of 27 m, and a current opening of 2 m wide by 1.2 m high. It is located ca. 12 km southeast of the town of Islamabad-e Gharb and 3.5 km northeast of the village of Tajar-e Akbar (348 030 20@ N and 468 380 42@ E). The cave faces north at an elevation of 1,430 m a.s.l.,
100 m lower than the summit C 2007 V

WILEY-LISS, INC.

C

Grant sponsor: Muse´um national d’Histoire Naturelle. *Correspondence to: Erik Trinkaus, Washington University, Department of Anthropology, Campus Box 1114, One Brookings Drive, St. Louis, MO 63130-4899. E-mail: [email protected] Received 26 June 2007; accepted 28 September 2007 DOI 10.1002/ajpa.20753 Published online in Wiley InterScience (www.interscience.wiley.com).

2

E. TRINKAUS ET AL.

Fig. 1. Map of the central Zagros Mountains showing the locations of known Upper Pleistocene sites that have yielded human remains.

Fig. 2. General view of the narrow valley at the Qazivand Mountain where the Wezmeh Cave is located. The site is indicated by a circle.

part of the cave’s contents at the rear chamber had been displaced onto the slope in 1995 by local people looking for artifacts (Abdi et al., 2002). The survey discovered Early and Middle Chalcolithic material mixed with a large mammal fauna and some human remains. The majority of the fragmentary human remains appear preservationally and morphologically recent, and they most likely derived from the Chalcolithic occupation of the cave. Three lithic artifacts were found, one amorphous flake, one simple flake, and a rough blade, but they

could not be attributed to any Paleolithic industry. The faunal remains, however, consist of two samples, one clearly Holocene based on its preservation, and the other distinctively more patinated and mineralized and therefore appearing substantially older [Mashkour et al. (submitted)]. In light of these findings, in 2001 a short field season was undertaken, during which faunal remains out of context were collected from the exterior slope, a 3 3 3 m2 trench was excavated to bedrock on the terrace im-

American Journal of Physical Anthropology—DOI 10.1002/ajpa

WEZMEH HUMAN PREMOLAR

3

Fig. 3. Plan of the Wezmeh Cave with the areas of excavation (after Abdi et al., 2002). The area of the clandestine excavation is indicated at the top, as are the positions of the six test pits within the cave and the larger trench at the entrance to the cave.

mediately outside of the cave entrance, and six test pits at 2-m intervals were dug to bedrock within the cave (Fig. 3). The excavated trench and other small test pits yielded Chalcolithic material consisting of sherds, two flint blades, beads, bone tools, and some faunal remains. But the surface collecting on the exterior slope and the

excavations on the terrace and within the cave yielded an abundant faunal assemblage, particularly rich in carnivore remains and including a human premolar of possibly Pleistocene age and the three lithic artifacts, [Mashkour et al. (submitted)] (Mashkour, in Abdi et al., 2002).

American Journal of Physical Anthropology—DOI 10.1002/ajpa

4

E. TRINKAUS ET AL.

Fig. 4. Minimum number of individuals (MNI) for the Wezmeh Cave Late Pleistocene faunal assemblage, showing the predominance of carnivore species (numbers 1–11 to the left). 1: Crocuta crocuta; 2: Ursus arctos; 3: Panthera leo; 4: P. pardus; 5: Caracal/Lynx/Felis; 6: Felis silvestris; 7: Canis lupus; 8: Vulpes vulpes; 9: Meles meles; 10: Mustela putorius; 11: Martes martes (foina?); 12: Herpestes sp.; 13: Dicerorhinus sp.; 14: Equus caballus; 15: Hemionus/E. asinus; 16: Bos primigenius; 17: Sus scrofa; 18: Cervus elaphus; 19: Gazella sp.; 20: Ovis orientalis; 21: Capra aegagrus.

THE VERTEBRATE FAUNAL CONTEXT Age of the faunal assemblage The mixed faunal remains were sorted into those that appeared Holocene (or Chalcolithic) in age and those that were likely to be considerably older. Among the latter (Fig. 4) were three species that suggested an early or pre-Holocene age for the assemblage. The hyena represented (the spotted hyena, Crocuta crocuta) underwent a significant decrease in prevalence during the last glacial and disappeared from southwestern Asia (including the Iranian Plateau) around 13–11 ka cal BP (Stuart, 1991; Stiner, 2004). The most recent southwest Asian occurrence of the steppe rhinoceros (Dicerorhinus) is from the oxygen isotope stage (OIS) 2 Kebaran layers of Hayonim Cave (Davis, 1982), and it is generally known (Tchernov, 1988) to be late Middle and Late Pleistocene in age in the region. The Ursus arctos dentition present at Wezmeh is substantially larger than those of the recent U. arctos in the region (Kurte´n, 1965). To further evaluate the age of these faunal remains, several carnivore and ungulate teeth were dated using U-series analysis by alpha spectrometry. The dates range between ca. 70 and ca. 11 ka BP [Mashkour et al. (submitted)]. The necessary condition for the applicability of the U-series dating, that the samples behave as closed systems during their burial history, is often violated in fossil tooth samples. Nevertheless, we hypothesize a uranium uptake and assimilation into the teeth soon after burial (‘‘early uptake’’), which implies that they behaved as though they were in a closed system. On the basis of these assessments, it may be inferred that the majority of the more fossilized faunal assemblage from Wezmeh Cave derives from the Late Pleistocene, principally from OIS 3 and 2, but possibly extending back into OIS 4.

Fig. 5. Mesial (M), distal (D), buccal (B), lingual (L), and occlusal views of the Wezmeh 1 maxillary premolar. Scale in millimeters. [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.]

(N 5 1,734) of the NISP and 74.1% of the total minimum number of individuals (carnivores and ungulates). This ratio is always at least 20% in carnivore dens; in contrast, it is less than 10% in archeological assemblages (Klein and Cruz-Uribe, 1984). Among the carnivores it is the four cave-denning species, C. crocuta, U. arctos, Canis lupus and Vulpes vulpes, which dominate the assemblage. These four denning species have NISP to MNI ratios of 25.9–43.7, reflecting the high level of skeletal part representation for these individuals; they are nonetheless approached by several of the Artiodactyl species, especially Sus scrofa (24.3) and Ovis orientalis (21.2), even though there are far fewer individuals for those species. The carnivore remains include all anatomical regions of the skeleton, whereas the herbivore remains consist principally of the distal portions of the extremities, suggesting scavenging of the latter. In addition, there were a few hyena coprolites in the cave deposits. Yet, the bones in general show little evidence of gnawing or other carnivore damage, most of the breaks resembling those from dry bone compaction and trampling. These observations, in addition to more detailed aspects of the assemblage taphonomy [Monchot et al. (submitted); Mashkour et al. (submitted)], imply that this cave served principally as a carnivore den. Given the U-series dates on the teeth, it is likely that it was principally a brown bear den ca. 60–70 ka cal BP, but it then served more as a hyena den during later OIS 3 and OIS 2, with additional use of the cave by wolves, foxes, badgers and occasional other carnivores.

THE WEZMEH 1 HUMAN PREMOLAR Among the disturbed faunal remains and Holocene archeological remains was a human maxillary premolar (Fig. 5), which appeared to be both large and fossilized compared with the Holocene human remains scattered on the surface. Since Pleistocene human remains are poorly known in the Zagros Mountains outside of Shanidar Cave, we attempted to date the tooth noninvasively using gamma spectrometry.

The faunal assemblage In terms of MNI and especially NISP, the assemblage is dominated by carnivores [Fig. 4; see Mashkour et al. (submitted)]. The carnivores overall make up 79.7%

Gamma spectrometry dating Given the small size of the Wezmeh human premolar, direct dating is limited to nondestructive gamma spec-

American Journal of Physical Anthropology—DOI 10.1002/ajpa

5

WEZMEH HUMAN PREMOLAR trometry dating. Gamma spectrometry (Yokohama and Nguyen, 1981) measures the daughter nucleids of 230Th, 235 Pa, and 238U, respectively 226 Ra, 227Ac, and 234Th. The sample was placed in the well of a very low background, high efficiency germanium detector in the ‘‘Laboratoire Souterrain de Modane,’’ which is protected from cosmic interactions by a 1,700 m rock overburden (Reyss et al., 1995). The measured specific activities were: 234Th 5 2.15 6 0.10 dpm/g, 226Ra 5 0.52 6 0.02 dpm/g, 227Ac 5 0.041 6 0.012 dpm/g, which correspond to a 230Th/234U age of ca. 20 ka cal BP and a 231Pa/235U age of ca. 25 cal ka BP. Taking into account the numerous hypotheses involved in such an analysis and the substantial data uncertainties (Yokohama and Nguyen, 1981), these dates are approximate and may be only minimum ages. Yet, these data are sufficient to assign the human tooth to either OIS 3 or early OIS 2, generally Upper Paleolithic in age.

TABLE 1. Crown and cervix measurements for the Wezmeh 1 Late Pleistocene human maxillary premolar Crown mesiodistal diameter Crown buccolingual diameter Cervix mesiodistal diameter Cervix buccolingual diameter Buccal crown height Lingual crown height

8.1 11.1 5.8 9.7 10.3 8.8

0.8 mm in diameter and a larger one 1.5 mm in diameter on the tip of the lingual cusp. These five pits represent minor developmental defects of the enamel (or dental enamel hypoplasias). Similar or more pronounced ones are commonly found on Late Pleistocene (and recent) human teeth.

Morphology Preservation, identification, age-at-death, and paleopathology The Wezmeh 1 premolar retains a complete crown and ca. 7 mm of the root. It is an unerupted tooth without occlusal or interproximal wear and with incomplete formation of the root. The only postmortem changes to the crown are two vertical fine line cracks on the mesial side, one running from the cervix to the developmental groove and the other from the cervix towards the apex of the buccal cusp, and one fine line crack distally from the cervix onto the buccal cusp. The root is largely complete to its developing edge, but chips have been lost from the thin margin buccally, lingually, and distally. On the basis of crown occlusal morphology, it is a maxillary right premolar. The buccal cusp is 17.0% higher than the lingual one (Table 1). The occlusal outline is slightly narrower lingually, as a result of the buccal cusp extending further distally. The lingual cusp is set mesially on the crown. The distal marginal ridge is relatively thick, but there is a developmental groove mesially. There is no ‘‘canine fossa’’ or concavity on the mesial crown, since it is straight to minimally convex. The root is incompletely formed, but the distal side has a marked sulcus implying that the root would have bifurcated had it formed completely. Following Hillson (1996), these criteria suggest that it is a first maxillary premolar, or P3. It nonetheless remains possible that it is a second premolar, or a P4. Since the tooth is incompletely formed and would have been in its crypt at the time of death, the age-at-death of the individual can be approximated. The amount of root formation is between R1/4 and R1/2 (Moorrees et al., 1963). Applying Smith’s (1991) data for mandibular premolars provides mean ages of 7.8–9.3 years (males) and 7.4–8.7 years (females). Comparison with Ubelaker’s revised chart (Ferembach et al., 1980) suggests an age of 8–9 (6 2) years. Wezmeh 1 therefore represents a late juvenile individual. As an unerupted tooth, there is no attrition or caries on the tooth, and the root exhibits no abnormalities. There are three very small pits on the sides of the crown, each one ca. 0.5 mm in diameter; they are located on mesiobuccal corner 3.2 mm from the cervix, on the distal side of the buccal face 7.1 mm from the cervix, and just distal of the second one but 8.1 mm from the cervix. There is also a pit on the tip of the buccal cusp

The Wezmeh premolar has a single (nonbifurcated) essential crest on the buccal cusp, which is separated from the lingual cusp crest by a distinct transverse groove; the groove continues mesially to bisect the mesial marginal ridge. The distal marginal ridge is the larger of the two, but it too is partly bisected by a mesiodistal groove. The crown bulges buccally above the cervix, and there is a distinct swelling with a shallow sulcus towards the occlusal surface ca. 6.0 mm from the cervix on the mesiobuccal corner of the crown. This latter feature is probably a minimally developed paramolar structure (or tuberculum molare). None of these features is unusual on P3s or P4s for either recent humans or any of the Late Pleistocene western Eurasian (late archaic or early modern) humans (Hillson, 1996; Bailey, 2006). On the distal side, the cervix curves convexly into the crown above the buccal cusp and then is straight above the mesial cusp. Mesially, the cervix has an even curve convex into the crown. On the buccal half of the mesial root, ca. 2.0 from the cervix, there is a distinct pit opening away from the cervix, 2.2 mm in diameter buccolingually. Similar pits have been observed on early Neandertal P3s from Biache-Saint-Vaast and Krapina (Rougier, 2003, per. observ.), but their distribution across relevant comparative samples is unknown.

Comparative morphometrics The Wezmeh premolar appeared to be relatively large compared with both Holocene and Late Pleistocene P3s and P4s. To assess this, we compared its buccolingual diameter (breadth) of 11.1 mm to samples of Late Pleistocene humans from western Eurasia. These samples included late archaic humans (Neandertals), southwest Asian Middle Paleolithic modern humans (Qafzeh and Skhul), European earlier Upper Paleolithic modern humans (Aurignacian and Gravettian), European late Upper Paleolithic modern humans, and southwest Asian Epi-Paleolithic humans (Natufians) (Table 2). The first four samples are from personal measurement and primary paleontological descriptions; the last sample is from Glantz (2004, pers. comm.). For both tooth positions, ANOVA across the five samples provides a P \ 0.0001; the first four samples provide an ANOVA P 5 0.0006 for the P3 (produced mostly by the smaller late Upper Paleolithic P3s), and 0.6195 for the P4.

American Journal of Physical Anthropology—DOI 10.1002/ajpa

6

E. TRINKAUS ET AL.

TABLE 2. Comparative buccolinguyal crown diameters for the Wezmeh 1 premolar and samples of Late Pleistocene western Eurasian humans, in millimeters P3 Wezmeh 1 Neandertals Qafzeh-Skhul Earlier Upper Paleolithic Late Upper Paleolithic Natufian Comparative sample ANOVA ANOVA without Natufians

11.1 10.3 6 0.7 10.4 6 0.4 9.9 6 0.6 9.5 6 0.8 8.6 6 0.8 \0.0001 0.0006

P4 (28) (9) (29) (18) (72)

11.1 10.0 6 0.7 10.2 6 0.8 9.9 6 0.6 9.6 6 0.6 8.7 6 0.9 \0.0001 0.6195

(21) (10) (25) (16) (75)

Mean 6 standard deviation (N).

The buccolingual breadth of 11.1 mm for Wezmeh 1 is at the tops of the Late Pleistocene Middle and Upper Paleolithic ranges of variation and exceeds the Natufian ranges of variation (Fig. 6). Using z-scores adjusted for comparing a specimen to a sample (Sokal and Rohlf, 1981), Wezmeh 1 has values of 1.12, 1.66, 1.97, 1.95, and 3.10, respectively, relative to the P3 comparative samples. Among Late Pleistocene specimens, as a P3 it is matched or exceeded only by the Cariguela 1, La Quina 5, and Spy 2 Neandertals. Among Pleistocene early modern humans, it is approached (breadth  11.0 mm) by Arene Candide 1, Arancio 1, Cisterna 1, and Qafzeh 9. If it is considered a P4, its z-scores are 1.54, 1.07, 1.96, 2.43, and 2.65, and it is approached only by Le Moustier 1 (11.0 mm) among the Middle Paleolithic humans, and approached or exceeded (breadth  11.0 mm) by CroMagnon 4 and Arene Candide 1 and 5, plus one Natufian from Eynan, among the Upper Paleolithic humans. The Wezmeh 1 is therefore among the largest known Late Pleistocene maxillary premolars, whether it is considered to be a P3 or a P4.

DISCUSSION Assuming that the Wezmeh 1 premolar dates to late OIS 3 or early OIS 2, based on the gamma spectrometry dating, it is likely that it derives from an Upper Paleolithic early modern human. The youngest Neandertals in southwest Asia are probably 40 ka cal BP (Vogel and Waterbolk, 1963; Valladas et al., 1999; Akazawa et al., 2002), and the oldest post-OIS 5 early modern humans in the region may well be ca. 40 ka cal BP (Bergman and Stringer, 1989). If, on the other hand, in the unlikely case that the gamma spectrometry dates are minimum ages and the tooth is substantially older, it could derive from an earlier OIS 3 Neandertal, or even less likely, an OIS 5 Middle Paleolithic early modern human. Unfortunately, the morphology of the tooth is insufficient to assess whether it derived from a late archaic or an early modern human. It is nonetheless a large tooth, falling at the upper limits of crown buccolingual diameter variation for the available Late Pleistocene samples and above a terminal Pleistocene (Natufian) sample. Moreover, assuming that Wezmeh Cave served solely as a carnivore den during the Late Pleistocene, given the abundance of such mammals and rare undiagnostic lithic artifacts from the deposits, it is probable that the late juvenile individual represented by this tooth was killed, or had its remains scavenged, by one of the carnivore species represented. Carnivore consumption of human remains in the Late Pleistocene has been docu-

Fig. 6. Boxplots of maxillary premolar buccolingual diameters, P3 above and P4 below. WZM: value for Wezmeh 1–11.1 mm; Nean: Neandertals; MPMH: Middle Paleolithic modern humans; EUP: European earlier (early and middle) Upper Paleolithic modern humans; LUP: European Late Upper Paleolithic humans; Natuf: southwest Asian terminal Pleistocene (Natufian) humans.

mented for Neandertals [e.g., Guattari (Giacobini, 1990– 91; White and Toth, 1991), Rochelot (Tournepiche et al., 1996), Rochers-de-Villeneuve (Beauval et al., 2005), Les Pradelles (Mann et al., 2005), and Oliveira (Trinkaus et al., 2007)], and carnivore activity may well account for the fragmentary condition of other late archaic and early modern humans (cf., Trinkaus et al., 2000). In addition, there are numerous cases of later Paleolithic sites, with or without human remains, in which the human occupations were variably intense and the sites also served as large carnivore dens, (e.g. Marquet et al., 1979; Clot, 1987; Stiner, 1991; Pa˘unescu, 2000, 2001; Monchot et al., 2000; Michel, 2004; Rabinovich et al., 2004; Villa et al., 2004; Monchot, 2006). The absence of carnivore damage to the Wezmeh premolar means that any such inference must be based on context.

CONCLUSION Wezmeh Cave is one of the few Late Pleistocene sites in the Zagros Mountains, and it has yielded an isolated human immature P3 (or P4) from a strictly paleontological context. It is a large, but otherwise morphologically unexceptional, tooth. On the basis of its context in a car-

American Journal of Physical Anthropology—DOI 10.1002/ajpa

WEZMEH HUMAN PREMOLAR nivore den, it is hypothesized that the juvenile individual may have been prey of those carnivores or had its remains scavenged after death. It may relate to roughly contemporaneous Upper Paleolithic occupations on the neighboring plain of Kermanshah (Olszweski and Dibble, 1984; Biglari and Taheri, 2000; Shidrang, 2005).

ACKNOWLEDGMENTS We are deeply indebted to Dr. M. Azarnoush, former Director of the Iranian Center of Archaeological Research, and Mr. M. Beheshti, Director of the Iranian Cultural Heritage Organization and Tourism (ICHTO), for having authorized this study, and to Mr. Bairanvand, Director of ICHTO at Kermanshah, Y. Moradi and A.M. Bisotuni for the help and logistical support. The UMR 5197 of the Muse´um national d’Histoire Naturelle (Paris) facilitated the use of comparative osteological collections. M.M. Glantz provided her original Natufian dental measurements. To all we are grateful.

LITERATURE CITED Abdi K, Biglari F, Heydari S. 2002. Islamabad project 2001. Test excavations at Wezmeh cave. Archa¨ol Mitt Iran Turan 34:171–194. Akazawa T, Muhesan S, Dodo Y, Ishida H, Konda O, Yoneda M, Griggo C. 2002. A summary of the stratigraphic sequence. In: Akazawa T, Muhesen S, editors. Neanderthal burials. Excavations of the Dederiyeh cave, Afrin, Syria. Kyoto: International Research Center for Japanese Studies. p 15–32. Bailey S. 2006. Beyond shovel-shaped incisors: Neandertal dental morphology in a comparative context. Period Biol 108:253– 267. Beauval C, Maureille B, Lacrampe-Cuyaube`re F, Serre D, Peressinotto D, Bordes JG, Cochard D, Couchoud I, Dubrasquet D, Laroulandie V, Lenoble A, Mallye JB, Pasty S, Primault J, Rohland N, Pa¨a¨bo S, Trinkaus E. 2005. A late Neandertal femur from Les Rochers-de-Villeneuve, France. Proc Natl Acad Sci USA 102:7085–7090. Bergman CA, Stringer CB. 1989. Fifty years after: Egbert, an early Upper Palaeolithic juvenile from Ksar Akil, Lebanon. Pale´orient 15:99–111. Biglari F, Taheri K. 2000. The discovery of Upper Paleolithic remains at Mar Kuliyan and Mar Dalan Cave, Rawansar. In: Taheri K, editor. Essays on the archaeology, geology, geography and culture of Rawansar area. Kermanshah: Taq-e Bostan Publications. p 7–27 (in Farsi). Clot A. 1987. La Grotte de Gerde (Hautes-Pyre´ne´es). Bagne`resde-Bigorre: Socie´te´ Ramond. Coon CS. 1951. Cave explorations in Iran 1949. Philadelphia: Museum Monographs, University Museum, University of Pennsylvania. Cowgill LW, Trinkaus E, Zeder MA. 2007. Shanidar 10: a middle Paleolithic immature distal lower limb from Shanidar Cave, Iraqi Kurdistan. J Hum Evol 53:213–223. Davis SJ. 1982. Climatic change and the advent of domestication: the succession of ruminant artiodactyles in the Late Pleistocene, Holocene in the Israeli region. Pale´orient 8:5–15. Ferembach D, Schwidetzky I, Stloukal M. 1980. Recommendations for age and sex diagnoses of skeletons. J Hum Evol 9:517–549. Giacobini G. 1990–91. Hyenas or cannibals: fifty years of debate on the Guatteri Cave Neandertal cranium. Quatern Nova 1:593–604. Glantz MM. 2004. Late Pleistocene-Holocene human morphological trends and the evolution of subsistence adaptations: Natufian skeletal remains in context. In: Delage C, editor. The last hunter-gatherer societies in the Near East. Oxford: John and Erica Hedges (BAR International Series). p 229– 248. Hillson S. 1996. Dental anthropology. Cambridge UK: Cambridge University Press.

7

Hole F, Flannery KV. 1967. The prehistory of southwest Iran: a preliminary report. Proc Prehist Soc 33:147–206. Klein RG, Cruz-Uribe K. 1984. The analysis of animal bones from archaeological sites. Chicago: University of Chicago Press. Kurte´n B. 1965. The carnivora of the Palestine caves. Acta Zool Fennica 107:3–73. Mann AE, Beauval C, Costamagno S, Bordes JG, Bourguignon L, Couchoud I, Lacrampe F, Laroulandie V, Marquet JC, Meignen L, Texier JP, Vandermeersch B, Maureille B. 2005. Carnivore and Neandertal interactions at the Les Pradelles site (Charente, Southwest France). PaleoAnthropol A 2005:28. Marquet JC, Byrne N, Delpech F, Macaire JJ, Prat F, Rage JC. 1979. Un grand site pre´historique en Touraine: la Roche Cotard a` Langeais. Grotte et station du Pale´olithique moyen. Tours: Laboratoire de Ge´ologie, Universite´ Franc¸ois Rabelais. Michel P. 2004. La grotte d’Unikote´ (Iholdy, Pyre´ne´es Atlantiques, France): une grotte-repaire d’hye`nes des cavernes avec des indices de pre´sence humaine dans des niveaux wu¨rmiens. Rev Pale´obiol Gene`ve 23:741–760. Monchot H. 2006. Un assemblage original au pale´olithique moyen: le repaire a` hye`nes, porc-e´pics et hominide´s de la grotte Geula (Mont Carmel, Israe¨l). Pale´orient 31:27–42. Monchot H, Marchal F, Cauche D, Desclaux E, Livache M. 2000. Un probable repaire d’hye`nes du de´but du wu¨rm les Auzie`res II (Me´thamis, Vaucluse). Bull Arche´ol Provence 28:75–83. Moorrees CFA, Fanning EA, Hunt EE. 1963. Age variation of formation stages for ten permanent teeth. J Dent Res 42:1490–1502. Olszweski DL, Dibble HL. 1984. The Zagros Aurignacian. Curr Anthropol 35:68–75. Pa˘unescu A. 2000. Paleoliticul s i Mezoliticul din Spatiul Cuprins ˆıntre Carpati s i Duna˘re. Bucharest: Editura AGIR. Pa˘unescu A. 2001. Paleoliticul s i Mezoliticul din Spatiul Transilvan. Bucharest: Editura AGIR. Rabinovich R, Bar-Yosef O, Vandermeersch B, Horwitz LK. 2004. Hominid-carnivore interactions in the Paleolithic site of Qafzeh, Israel. Rev Pale´obiol Gene`ve 23:627–637. Reyss JL, Schmidt S, Legeleux F, Bonte´ P. 1995. Large, low background well-type detectors for measurements of environmental radioactivity. Nucl Instrum Methods Phys Res Sect A 357:391–397. ´ tude Descriptive et Comparative de BiacheRougier H. 2003. E Saint-Vaast 1 (Biache-Saint-Vaast, Pas-de-Calais, France). The`se de Doctorat, Universite´ de Bordeaux 1. Shidrang S. 2005. Survey of the rockshelter site of Warkaini near Kermanshah. Bastanshenasi 1:76–79 (in Farsi with English Abstract). Smith BH. 1991. Standards for human tooth formation and dental age assessment. In: Kelly MA, Larsen CS, editors. Advances in dental anthropology. New York: Wiley-Liss. p 143–168. Sokal RR, Rohlf FJ. 1981. Biometry, 2nd ed. San Francisco: W.H. Freeman. Solecki RS. 1963. Prehistory in Shanidar valley, northern Iraq. Science 139:179–193. Stiner MC. 1991. The faunal remains from Grotta Guattari: A taphonomic perspective. Curr Anthropol 32:103–117, 132–138. Stiner MC. 2004. Comparative ecology and taphonomy of spotted hyenas, humans and wolves in Pleistocene Italy. Rev Pale´biol Gene`ve 23:771–775. Stuart AJ. 1991. Mammalian extinctions in the Late Pleistocene of northern Eurasia and North America. Biol Rev 66:453–562. Tchernov E. 1988. The biogeographical history of the Southern Levant. In: Yom-Tov Y, Tchernov E, editors. The zoogeography of Israel. Dortrecht: Dr. W. Junk Publishers. p 159–250 Tournepiche JF, Couture C, Guadelli JL, Michel P. 1996. Les restes ne´andertaliens du repaire d’hye`nes de la grotte de Rochelot (Saint Amant de Bonnieure, Charente, France). C R Acad Sci Paris Se´rie IIa, 322:429–435. Trinkaus E. 1983. The Shanidar Neandertals. New York: Academic Press. Trinkaus E. 2006. The purported human femur from Tamtama Cave, Iran. Bastanshenasi 2:11–12.

American Journal of Physical Anthropology—DOI 10.1002/ajpa

8

E. TRINKAUS ET AL.

Trinkaus E, Biglari F. 2006. Middle Paleolithic human remains from Bisitun Cave, Iran. Pale´orient 32:105–111. Trinkaus E, Maki J, Zilha˜o J. 2007. Middle Paleolithic human remains from the Gruta da Oliveira (Torres Novas), Portugal. Am J Phys Anthropol 134:263–273. Trinkaus E, Svoboda J, West DL, Sla´dek V, Hillson SW, Drozdova´ E, Fisˇa´kova´ M. 2000. Human remains from the Moravian Gravettian: morphology and taphonomy of isolated elements from the Dolnı´ Ve˘stonice II site. J Archaeol Sci 27:1115–1132. Valladas H, Mercier N, Froget L, Hovers E, Joron JL, Kimbel WH, Rak Y. 1999. TL dates for the Neanderthal site of the Amud Cave, Israel. J Archaeol Sci 26:259–268.

Villa P, Castel JC, Beauval C, Bourdillat V, Goldberg P. 2004. Human and carnivore sites in the European Middle and Upper Paleolithic: similarities and differences in bone modification and fragmentation. Rev Pale´obiol Gene`ve 23:705– 730. Vogel JC, Waterbolk HT. 1963. Groningen radiocarbon dates IV. Radiocarbon 5:163–202. White TD, Toth N. 1991. The question of ritual cannibalism at Grotta Guattari. Curr Anthropol 32:118–124, 135–138. Yokoyama Y, Nguyen HV. 1981. Datation directe de l’homme de Tautavel par la spectrome´trie c, non destructrive, du craˆne humain fossile Arago XXI. C R Acad Sci Paris 292:741–744.

American Journal of Physical Anthropology—DOI 10.1002/ajpa