I1 1 8 2 0 1 B Univerza v Ljubljani jjjjQjg Filozofska Fakulteta Oddelek za Arheologijo Documenta Praehistorica XXVII Editor Mihael Budja DOCUMENTA PRAEHISTORICA XXVII 7th NEOLITHIC STUDIES/NEOLITSKE STUDIJE 7 Editor/Urednik: Mihael Budja, miha.budja@uni-lj.si Editorial board/Uredniski odbor: Dr. Tatjana Bregant, Dr. Mihael Budja (Univerza Ljubljana), Dr. Vida Pohar (Univerza Ljubljana), Dr. Katherine Willis (University of Oxford) © 2000 Oddelek za arheologijo, Filozofska fakulteta - Univerza v Ljubljani, SI -1000 Ljubljana, P.B. 580, tel.: 386 01 241 15 70 English advisor/Jezikovni pregled angleskega besedila: Philip James Burt Published by/Zalozila: Filozofska fakulteta, Oddelek za arheologijo Technical editor and DTP/Tehnicno urejanje in DTP: CAMBIO d.o.o., Ljubljana Printed by/Tisk: SCHWARZ d.o.o., Ljubljana Number printed/Naklada: 750 izvodov Natisnjeno s podporo Ministrstva za znanost in tehnologijo ter Ministrstva za kulturo Republike Slovenije in Znanstvenega instituta Filozofske fakultete. Contents Mihael Budja V Preface Mirjana Roksandic 1 Between Foragers and Farmers in the Iron Gates Gorge: Physical Anthropology Perspective Djerdap Population in Transition from Mesolithic to Neolithic Vesna Dimitrijevic 101 The Lepenski Vir Fauna: Bones in Houses And Between Houses Give Bonsall, Gordon Cook, Rosemary Lennon, Douglas Harkness, Marian Scott, Laszlo Bartosieivicz, Kathleen McSweeney 119 Stable Isotopes, Radiocarbon and the Mesolithic-Neolithic Transition in the Iron Gates Nina Kyparissi-Apostolika 133 The Mesolithic/Neolithic Transition in Greece as Evidenced by the Data at Theopetra Cave in Thessaly Laurens Thissen 141 Thessaly, Franchthi and Western Turkey: Clues to the Neolithisation of Greece? Burcin Erdogu 155 The Problems of Dating Prehistoric Axe Factories and Neolithisation in Turkish Thrace Onur Ozbek 167 A Prehistoric Stone Axe Production Site in Turkish Thrace: Hamaylitarla Eszter Bdnffy 173 The Late Starcevo and the Earliest Linear Pottery Groups in Western Transdanubia Wolfram Schier 187 Measuring Change: the Neolithic Pottery Sequence of Vinca-Belo Brdo Paraskevi Yiouni 199 Painted Pottery fom East Macedonia, in North Greece: Technological Analysis of Decorative Techniques Zhou Guoxing 215 On the Bailiandong Culture Feng Zhang 225 The Mesolithic in South China Chaohong Zhao and Xiaoliong Wu 233 The Dating of Chinese Early Pottery and a Discussion of Some Related Problems Jalal Rafifar 241 Anthropological Approaches in Neolithic Studies in Iran Hasan Talai 251 New Painted Pottery Assemblage at Ismailabad, Late Neolithic Site in the Central Plateau of Iran Adam from Lepenski Vir (D. Srejovic and L. Babovic. Umetnost Lepenskog Vira. Beograd 1983) Preface The prologue to this anthology was the seventh Neolithic Seminar held at the Department of Archaeology, University of Ljubljana in May 2000. The Seminars provided an opportunity for a more balanced exchange of research data of the processes of transition to farming, i.e. neo lithisation in Europe and Asia. The participants of the Seminar and the authors of the papers in this volume were selected as individuals who were active in field research dealing with the Mesolithic and the Neolithic in Europe and Asia, and who at the same time being involved in developing the revised perspectives for understanding the processes of transition to farming. As far as content is concerned, we have maintained the link with cultural, periodic and typological paradigms, although the focus has been shifted to questions in regard to the processes of globalisation of the Neolithic palaeoeconomy. Two concerns were emphasised at the first Seminar, how and when agriculture spread to Europe. With the growing body of data from different regions, it has become apparent later that agriculture developed independently in many areas of the world, and that modes of transition to food production from food gathering were specific for each individual geographic entity. It is clear from recent studies that the introduction of farming to Europe was not the monothetic consequence of the "wave of advance". It is broadly accepted that contacts between foragers and farmers, occurring within an agricultural frontier zone must have had a direct effect on the nature an the rate of the transition, and may have acted as a delaying mechanism in the process of the transition on the regional level. However, one of the most important points is that playing an active part as individuals and as communities, hunters and gatherers contributed to the generation of a different kind of Neolithic through their own communities and their influence on the established farming settlement. It has been already pointed out that in many parts of south-eastern Europe, there were clusters of Mesolithic settlements capable and ready to serve as a promotion centres of agro-pastoral farming in the course of which process these communities could be expected to develop or to adopt and to modify agro-pastoral practices and pottery production and integrate them with existing subsistence strategies. However, Feng Zhang has noted in this volume that many Chinese archaeologists agree that transition to farming in Eastern Asia correlates with the transition from the Palaeolithic to the Neolithic and, as Chaohong Zhao and Xiaohong Wu showed, it was pottery that appeared first. Besides the analyses of the transition to farming in Europe and Asia, special attention was paid to technological analyses of the Neolithic painted pottery in North Greece and Central Iran as well as to pottery sequence of the Neolithic tell site of Vinca. Radiocarbon dates given in this volume using the convention bp and be refer to uncalibrated dates. BP and BC are used to indicate calibrated radiocarbon dates unless otherwise noted by the authors. Ljubljana, december 2000 UDK: 903(497.11 )"633/634":572.1 /.4 Documenta Praehistorica XXVII Between Foragers and Farmers in the Iron Gates Gorge: Physical Anthropology Perspective Djerdap Population in Transition from Mesolithic to Neolithic1 Mirjana Roksandic Max Planck Institute for Demographic Research, Rostock, Germany Roksandic@demogr.mpg.de ABSTRACT - The research presented here aims at discerning possible interactions between Mesolithic hunter-gatherers of the Iron Gates Gorge (Serbia-Romania) and the surrounding Neolithic farmers during the 7th and the 6th millenniums BC. In order to examine the interactions of communities with different modes of subsistence (foraging and farming respectively), the nonmetric anatomical variants of the skull and postcranial skeletons ivere examined on the sites with the largest number of individuals buried. Another set of analyses, aimed at discerning environmental (occupation/nutrition) changes that could have affected the population in transition was performed on metric variables of postcranial skeleton. The combination of these two sets of analyses argues for local continuity within the region, with high degree of initial heterogeneity, and temporal ordering as the most likely explanation for the pattern of change. IZVLECEK - Nas namenje spoznati mocne medsebojne vplive med mezolitskimi lovci-nabiralci v Zeleznih vratih (Srbija-Romunija) in sosednjimi neolitskimi kmetovalci v 7. in 6. tisocletju BC. Da bi ugotovili medsebojne vplive skupnosti z razlicnima nacinoma prezivljanja (lov-nabiralnistvo in kmetovanje), smo raz-iskali nemetricne anatomske razlicice lobanj in postkranialmh skeleton z najdisc, kjer je najvec individu-alnih pokopov. Opravili smo tudi analizo metricnih razlicicpostkranialnih skeletov. Z njo smo zeleli ugo-toviti spremembe okolja (naselitev/prehranjevanje), ki bi lahko vplivale na skupnosti v tranziciji. Obe vrsti analiz kazeta na lokalno kontinuiteto z veliko zacetno heterogenostjo in na casovno strukturiranost kot najbolj verjetno razlago vzorca spremembe. KEY WORDS - Mesolithic; Neolithic; Iron Gates Gorge; non-metric traits; Lepenski Vir; Starcevo Dedication: To peoples and cultures, past and present, sacrificed to others' understanding of progress. 1. INTRODUCTION The transition from foraging to agriculture can be regarded as one of the most fundamental cultural-ecological transformations that has occurred in the human career, as it enabled large scale sedentism, subsequent population growth, and the appearance of an urban way of life. Recent revival of interest in this phenomenon is witnessed by a number of edited volumes that discuss concepts of domestication and plant cultivation, the origin and spread of agricultural practices in different regions, as well as popu-lational and social implications of the transition (Cohen and Armelagos 1984; Gregg 1991a; Harris 1996a; Price and Gebauer 1995). With the growing body of data from different regions, it has become apparent that agriculture developed independently in many areas of the world, and that modes of transition to food production from food gathering were specific for each individual geographic entity. The study presented here aims to understand the patterning of the change from foraging to farming in the Lower Danube as reflected in the transition from Mesolithic "Lepenski Vir" culture to Neolithic "Starcevo" complex in the Iron Gates Gorge in Serbia 1 This paper is based on the Doctor of Philosophy Thesis defended at Simon Fraser University before the Committee: Dr. Ruth Tringham, Dr. Christopher Meiklejohn, Dr. Jonathan Driver, Dr. Jack Nance, and Dr. A. Catherine D'Andrea. Fig. 1. Iron Gates Gorge in May 2000. On the left side of the photo the clearing of the submerged Hajducka Vodenica terace. The other sites are located further upstream. (Fig. 1). Since a long period of coexistence of the Me-solithic and Neolithic ways of life in this region has been proposed (Radovanovic 1996b), this research will attempt to reconstruct the extent and mode of interactions between farmers and foragers through the examination of skeletal material from four of the most important sites excavated in the region: Lepenski Vir, Hajducka Vodenica, Vlasac and Padina. 1.1. Theoretical background It is generally accepted that methods of food production, together with the cultigens, were introduced to Europe from the Near East via Greece and the Balkan Peninsula. The mode of transition was either the transfer of farming techniques to indigenous populations with practically no genetic admixture, the migration of farmers themselves, or both. The archaeological data alone do not permit the distinction between these modes even if they were mutually exclusive. Based on genetics of living European populations, Cavalli-Sforza proposed the model of "demic diffusion", which argued for the spread of agriculture by migration of people from the southeast to the northwest, at an average pace of 1 km a year (Ammerman and Cavalli-Sforza 1984). In later publications, this model came to incorporate the transfer of technology as a part of the process (Cavalli-Sforza 1996). Assimilation of foragers by farmers, through deforestation and acceptance of agricultural practices (competition and acculturation) and marriages of hunter-gatherer women to farmers (acculturation), would have created the "gradient" observed in the Principal Component Analysis (PCA) of genetic data. The underlying assumption of the model, that the Mesolithic population was sparse throughout Europe, has been questioned by recent paleodemographic studies (Meiklejohn et al. 1997). In the extremes of the European periphery (Scandinavia, Portugal, Spain, Baltic States) this model has been questioned by the survival of Mesolithic cultures in proximity to incoming Neolithic ones (Price 1996; Thomas 1996; Zvelebil 1996a). The cranio-metric analyses of European prehistoric samples failed to provide support for the model of population replacement (Harding et al. 1989). Further, coexistence and mutualistic exchange that was proposed recently for Central Europe, as opposed to confrontation and acculturation, provides a more likely and flexible context for understanding the early interactions of farmers and foragers (Gregg 1988, 1991b). Given the fact that the first farmers were moving north into essentially unknown habitats, and progressively more marginal climatic conditions for the production of their newly domesticated plants, while the foragers were long time inhabitants with excellent knowledge of seasonal food availability, it is unlikely that the neolithisation of Europe could have been as swift without involvement of local Mesolithic inhabitants (Prinz 1987). 1.2. Relevance of this study Most of the studies that propose models for Europe as a whole, are based on the spread of the "Linear Bandkeramik" culture that is limited to Central Northwest Europe and, in terms of European perio-disation, contemporaneous with the Late Neolithic cultures of the Southeast (Vinca, Dimini, see Table 1). The first manifestations of the Neolithic in south- eastern Europe precede Linear Bandkeramic culture and show significantly different economic and social patterns. Neither the spread of "Cardium-impresso" culture in the circum-Mediterranean region, nor the apparent processes in the Aegean or Balkans conform to the demic diffusion model. The comparatively understudied Balkans are rarely discussed in theoretical literature except for the absolute dates, which are usually applied without reference to the "cultural" or socio-economic background and are used to argue the direction and the pace of migration. In order to build larger geographic models, it is crucial that areas of Europe that have not been carefully examined become the focus of a critical study. Further, as neolithisation of southeastern Europe precedes the neolithisation of the rest of the continent, understanding the processes and those attributes of Mesolithic in the area that allowed for the rapid neolithisation is of extreme importance. The goal of this research, and of the ongoing work by the team of researchers from the Institute of Archaeology and the Departement of Archaeology in Belgrade, is to make the Iron Gates Gorge material accessible for comparison with that from other regions of Europe and thus provide a basis for building sensible models of neolithisation of the continent. Archaeological and anthropological study of the material from the Iron Gates Gorge (Figs. 1 and 2) is supported on a number of bases: • the neolithisation of this area preceded the neolithisation of all other parts of Europe except the Aegean; • the extent of excavations of a number of sites on both banks of the Danube has been unparalleled in the last three decades as it has exposed sub- stantial living surfaces amenable to detailed archaeological analyses; • the sites provide insight into both cultural and po-pulational consequences of the transition through abundant architectural features, artefacts, faunal, palynological and osteological remains and; • since the coexistence of the Mesolithic foragers, belonging to the Lepenski Vir culture, and Neolithic Starcevo farmers has been established (Boro-neanfetal. 1995; Radovanovic 1996a; 1996b), the degree and mode of interaction between them can potentially be traced in both archaeological and osteological remains. 1.3. The Lepenski Vir culture Material remains of the Mesolithic Lepenski Vir culture have been uncovered on a number of sites in Iron Gates Gorge and dated from 7500 to 5800 BC. The culture is characterised by a sedentary or semi-sedentary foraging economy based on varied and abundant resources provided by the Danube and adjacent mixed growth forests. A long period of cultural stability, complex social organisation and developed religious beliefs can be postulated from remarkable architectural achievements and expressive monumental sculpture. Most of these sites are at least partly contemporaneous with farming communities of the Gura Baciului, Anzabegovo, Starcevo and Kara-novo type (Gimbutas 1976;Jovanovic 1984; Radovanovic 1992; Srejovic and Letica 1978). However, throughout the period when contact was possible, Lepenski Vir culture did not change significantly, and the pressure from the outside served more to seal the bonds between different Iron Gates Gorge sites, than to undermine their cultural or economic unity {Srejovic 1978; Radovanovic 1996c; 1996d). Iron Karpathian Macedonia Bulgaria Thessaly Central Date Gates Pannonian Europe BC ca. Gorge Balkans Vinca Vinca Karanovo IV Dimini LinearBand Keramik 5000 Starcevo MB Starcevo Koros Cris Vrsnik Karanovo III Early Neolithic III (Sesklo) 5500 Lepenski Vir Starcevo Koros Cris Vrsnik Karanovo II Early Neolithic III (Sesklo) 6000 Lepenski Vir Gura Baciului Anzabegovo1 Karanovo 1 Early Neolithic II (Proto Sesklo) 6500 Lepenski Vir 7500 Tab. 1. Schematic representation of the chronological relationships between geographic regions in Southeast Europe. Dates BC only approximate. Osteological material that is central to ray research comprises 362 or more individuals from four sites of the Lepenski Vir culture: Padina, Hajducka Vodenica, Vlasac and Lepenski Vir. These four sites, situated on the south bank of the Danube and characterised by sophisticated architectural remains, were chosen for their large number of burials. This is one of the two largest skeletal series that span the Mesolithic to Neolithic transition in Europe. The time period they cover is over 1500 years. The other large series comes from sites in Russia and the Baltic States. In addition, the restricted regional distribution makes this sample more meaningful and amenable to statistical analyses than the Russian and Baltic material. Since basic metric analyses have been done, at least for Lepenski Vir and Vlasac (Mikic 1981a; 1981b; Nemeskeri and Szathmary 1978a; 1978b; 1978c; 1978d; 1978e), and odontometrics for Vlasac (y'Ed-nak and Fleich 1983), and in the view of partial inhumations and unequal preservation of individual skeletons, I have decided to concentrate on the distribution of non-metric anatomical variants as the most relevant measure of population distances. 2. THE PROBLEM OF DISTINGUISHING MESOLI THIC FROM NEOLITHIC Although the Mesolithic is a well-established term in archaeological literature, its exact meaning remains susceptible to differing interpretations. Since the focus of this work is the Mesolithic population of the Iron Gates Gorge and its presumed contact with Neolithic peoples, and since the debate over the meaning of this term has historically played an im- portant role in discussions between principal investigators of the Iron Gates Gorge sites (Boroneanf et al. 1995; Jovanovic 1972; Srejovic 1971; 1979; Sre-jovic and Letica 1978), it is important to provide clear definitions of both Mesolithic and Neolithic as they are used here. 2.1. Definition of terms 2.1.1. Mesolithic In 1865, John Lubbock divided Prehistory into the Old Stone Age - "Palaeolithic" characterised by flaked stone, and the New Stone Age - "Neolithic" -characterised by the introduction of polished stone implements. This division was formal and typological. It enabled archaeologists to assign finds into two global categories of prehistoric technology. The first to coin the term "Mesolithic" was Hodder M. West-ropp who in 1872 used it to denote everything from "Reindeer period caves" until the introduction of agriculture, therefore both Upper Palaeolithic and Mesolithic as we use it today (Rowley-Conwy 1996). At about the same time M. Reboux, independently introduced the term in France to describe an industry intermediate between those that he defined as producing only flakes (such as Levallois and Mouste-rian in general) and those that produced polished axes, essentially including the same time span and typology as Westropp (Orliac 1988). Although his classification was not accepted, the term "Mezolithi-que" came to live on in the work of Archibald Car-lyle who applied it to an industry of "small geometric flints" found in India (quoted in Orliac 1988). The temporal notion was introduced in 1893, by an antiquarian Allen Brown, who used the term to describe those cultures of the Holocene that existed Fig. 2. Satellite map of the Pannonian Plain, Karpathian Basin and the Balkan Mountains. Flowing through the Pannonian Plain, the Danube enters the Karpathian Mountains and forms the Iron Gates Gorge (outlined by the white square). The Early and Middle Neolithic cultures that are partially contemporatieous with Lepenski Vir culture are outlined in the centers of their respective areas. fAdapted from NCARTA 1998J. before the introduction of agriculture (Pedersen pers. comm.). According to Orliac (1988.686), the widely spread definition that viewed as "Mesolithic all the industries between Magdalenien and Neolithic" was introduced by J. de Morgan in 1909. Current definitions can be classified in two major groups: one typological and the other chronological. For the proponents of the typological definition the characteristics of flint industry (such as the appearance of microliths) have the most decisive value. Orliac proposes that those industries situated between Palaeolithic and Neolithic that possess "characters sufficiently different from those of the industries of the two periods" (Orliac 1988.686) should be determined as Mesolithic. Although it may be appropriate for western European archaeology, even though "sufficiently different characters" remain essentially arbitrary and ambiguous, the definition falls short in other areas of the world. For proponents of chronological definition all hunter-ga-therers of the Holocene are regarded as Mesolithic, regardless of whether they show differences from the Palaeolithic hunter-gatherers. Since it would include a number of communities in the Holocene that continue with essentially the same mode of food procurement and mobility patterns as their Pleistocene counterparts as well as significantly different groups, the need to introduce 'Mesolithic' as a different term is not as obvious. Introduction of economic parameters "that the [term] never had historically" (Orliac 1988.686) has, in a number of cases, led to economic determinism, which includes an open or subdued notion of evolutionism. Distinction is made between the Epipaleolithic - in which Holocene adaptation does not produce any changes in way of life and lithic technology (.Kozloivski and Kozlowski 1986; Leroi-Gourhan 1965) and the Mesolithic - with its substantial changes in economy, ecology, and material culture (.Kozlowski and Kozloivski 1986). The latter would be found only in innovation zones leading to food production (Leroi-Gourhan 1965), or enabling change from food collection to food production (Clark 1980). This definition supposes a unidirectional evolution towards food production and contradicts the data from large areas of the world where substantial changes in economy, ecology and mate- rial culture did not lead to introduction of agriculture (e.g. West Coast of Canada). In order to overcome chronological and typological ambiguity, as well as economic determinism characterised by an implicit evolutionist basis, Radovano-vic (1996a. 14) argues that a qualitatively different phenomenon, capable of distinguishing archaeolo-gically Mesolithic groups from those of the Palaeolithic and Epipaleolithic, can be found in the appearance of formal disposal areas for the burial of the dead. Formal disposal areas need not be a phenomenon separate from the habitation site, as that would exclude all western and central European sites (Meiklejohnpers. comm.), with the exception of the newly excavated Mesolithic necropolae in France (Duday and Courtaud 1998) and Belgium (Cauwe 1998). They are determined as "...areas of continuous, ceremonial, mortuary disposal" (Rado-vanovic 1996a. 14). Further, they are an archaeolo-gically visible phenomenon that is interpreted as arising from the need to lay claim on the territory by its ideological integration (Chapman 1981)2. The need to claim territory, in turn, would arise from a combination of linear rather than hexagonal arrangement of units within one hunter-gatherer group or higher than usual population densities (Gamble 1986.52-53), and a hunter-gatherer economy based on intensive exploitation of a vital resource, or a greater variety of resources in the vicinity, with semi-sedentism or sedentism. This would result in a structural complexity of the social unit (Srejovic 1979) usually expressed through developed ancestral and mortuary rituals. While the appearance of formal disposal areas for the dead does not necessarily arise exclusively from the concerns of territoriality, and while they are not associated exclusively with the Mesolithic, their appearance is a clear sign of changing times in prehistory associated with changes in the social arena (Chapman 1993)■ However, although they are an archaeologically visible element, they are not the only one that enables recognition of a site as Mesolithic. Furthermore, as they allow for different interpretation and understanding of what constitutes a formal disposal area, they are not necessarily the best element for classification. The introduction of economic parameters and, even more importantly, mobility patterns, once evolutio- 2 Although this monocausal explanation that was applied to all the mortuary monuments in Neolithic Europe is overly simplistic, and reveals more about the preoccupations of modern-day western scholars than prehistoric inhabitants of Europe (cf Cullen 1995.286), it remains one of the possible, and even plausible reasons, but can not be perceived as the only cause of the arising importance of mortuary ritual in the period (Masset 1993). nist connotations are removed, has the potential to make this term more meaningful and appropriate for regions where the distinction of Mesolithic from Palaeolithic and Neolithic, based on flint, stone and bone industries, is not as straight-forward as in France (for example, since microliths in Africa appear as early as 70 000 BP), or where ceramics (traditionally associated with the Neolithic) appear earlier than agriculture, as for example in the Jomon (Imamura 1996) or in Scandinavia (Werbart 1998. 37, and quoted literature). This innovation and change in adaptation is usually linked with intensification of exploitation of one or more abundant resources (r-resource) as opposed to exploitation of a ^-selected resource (see Gamble 1986.41) that characterises mobile hunter gatherers. The availability of an abundant and stable resource that can be exploited in the relative vicinity of the camp has been linked to reduced mobility. This combination seems to be determinative of Mesolithic settlements. For example, the specific climatic conditions of the Iron Gates Gorge, the refugial character of its flora, and the great variability of plants without a dominant species where 44% of the 371 species of plants are useful in human, and 59% of them useful in animal nutrition, has enabled intensification of both settlement and exploitation. However, throughout the period, the terrestrial animal component (K-resources) remains dominant (Radovanovic 1996a, 37) and although fish is deemed an important resource providing the bulk of the protein during the Mesolithic (Bonsall et al. 1997), Radovanovic argues that its role is more vital than dominant, thus providing a buffering mechanism in periods of scarcity, rather than being a year round staple. Either way, its availability becomes the key factor enabling different spatial distribution of settlements and other features of Mesolithic organisation. Structural complexity, seen as a segmentation or an increase in the parts that make up the whole (Kent 1989a. 10), could arise from changes in mobility patterns and increased sedentism which results in population increase and the need for an arbiter in settling disputes (Lee 1972a; 1972b). However, it can not be taken for granted and must be demonstrated by independent data in many areas of the world where it has been assumed (Brinch-Petersen and Meiklejohn in press). Moving in full circle from cause to the effect, that in its turn becomes a cause, rises a complex picture of interacting forces of environmental productivity, se- dentism motivated by linear distribution of resources or their availability, and causing social restructuring. Structural complexity arissing from sedentism, causing further environmental changes. Together with sedentism, the latter influences mortuary ritual, while mortuary ritual in turn influences both sedentism and social structure. In the context of this study, and for the area in question, the Mesolithic is best defined as primarily based on intensified exploitation of food resources on a limited territory with reduced mobility. Settlement distribution and mobility patterning, interacting with more intensive exploitation of r-resources (plant or fish), and aggregation necessary for tasks demanding cooperative effort, can be regarded as both determinative of the Mesolithic and as providing sufficient archaeological visibility. Within the Iron Gates Gorge, regardless of whether fish played a key role as staple or as a vital resource, its availability is the sine qua non of reduced mobility (sensu Kent 1989a) and the change in its exploitation provides clear evidence of the Mesolithic economy and social organisation. 2.1.2. Neolithic Proceeding from the typological classification of prehistory, the Neolithic in Europe has been regarded as a period when polished stone was introduced, alongside ceramic production. This typological distinction was questioned with further developments in archaeology, especially the excavations of "pre-pottery Neolithic" in the Levant (Wright 1992). An economic definition in which "the shift in mode of subsistence to agro-pastoral farming remains the only process that is relatively clearly defined, geographically widespread and sufficiently archaeologically detectable" to act as a signature of the Neolithic (Zve-lebil 1996b. 625) is widely accepted and both ethno-logically and archaeologically traceable. Preceding and subsequent changes in social structure, ideology or any other aspects of life need to be examined on a regional basis. The other two questions, that of how much evidence of plant manipulation and animal husbandry constitutes enough evidence (sensu Harris 1996a and quoted literature), and whether a horticultural stage of farming economy - characterised by lack of impact on the environment (Willis and Bennet 1994) - can be perceived as agriculture, also need to be regarded on a local scale. In Southeast Europe, the introduction of cultigens and domestic animals from the Near East solves this problem, as these are not found in the wild. Their introduction indicates a shift in subsistence strategy and reliance (at least partial) on imported cultigens. 2.2. The Iron Gates Gorge context In terms of the Iron Gates Gorge, this semantic discussion is by no means unimportant. Interpretations for Lepenski Vir - Schela Cladovei culture range from Epipaleolithic (Boroneant 1969; Lazarovici 1979; Letica 1971), Protoneolithic (Srejovic 1968), Epipaleolithic in its early and Protoneolithic in its late phase (,Srejovic 1979), Mesolithic (Prinz 1987; Srejovic 1989; Voytek. and Tringham 1989), to Mesolithic and Early Neolithic (jovanovic 1972; 1974). This variety stems to a great degree from the above mentioned definitions: one chronological, in which Holocene hunter-gatherers are differently viewed as Mesolithic or Epipaleolithic; and the other based on material culture where microliths are taken as a telltale sign of the Mesolithic, while ceramics, polished stone axes and adzes are used as markers of Neolithic (for discussion and appreciation of the theoretical positions in these different approaches see Radovanovic 1996a). Littoral distribution of the sites in the Iron Gates Gorge that resulted in reduced mobility or possibly even sedentism, increased social complexity evidenced by specialisation of ritual vs. domestic activities (,Srejovic 1979. but see Chapman 1993) and increased population aggregation (Jack.es et al. 2000) are all very prominent features of Mesolithic components on the sites of this culture. Neolithic in the region is characterised by introduction and reliance, at least partial, on the Near Eastern cultigens and livestock. Although hunting, fishing and gathering remain important in the region, the proportion of domesticates vs. wild fauna and flora is sufficient to argue for the introduction of Neolithic economy. Crucial to our understanding of the Iron Gates Gorge Mesolithic/Neolithic transition is the period of the coexistence of these two modes of life in the immediate proximity (Fig. 1). It is evidenced by both l4C dates and Starcevo ceramics and flint blades within Mesolithic strata of the Lepenski Vir culture. This period witnesses the coexistence, communication, trade and interaction between Lepenski Vir hunter-gatherers and Starcevo farmers. It is in this light that the semantic discussion of the two terms becomes increasingly important for understanding archaeological data3. Therefore, if intensification of exploitation of food resources, on a limited territory with reduced mobility, characterises the Lepenski Vir Mesolithic, then all strata within these sites that do not have evidence of food production and/or introduced domesticates (above 5% as suggested by Zvelebil 1996a) should be regarded as Mesolithic. If we accept that Neolithic in the region is characterised by introduced cultigens and livestock, then evidence for food production and appearance of these cultigens in the strata should be regarded as crucial for identifying them as Neolithic. In this perspective, the sporadic appearance of Starcevo (Neolithic) type ceramics and Starcevo blades in Mesolithic strata, i. e. Mesolithic type house with ceramics in situ (Jovanovic 1984), requires explanation. A porous agricultural frontier with transfer of knowledge, material items, and individuals across the board (Zvelebil 1996a) offers a reasonable model, if we accept that this exchange could have involved both different direction and different form over this long period of coexistence. 2.3. Forager - farmer's interactions: possible scenario(s) for Europe Historically, the role of local forager populations has received little attention. This is due to the perception of human culture as "developing" over certain immutable stages, through which every society would eventually have to pass (Childe 1936; 1951; Engels 1972 /1884, see also: Earle 1994, for an overview of different classificatory models), and since the spread of farming into Europe was modelled on the spread of European agriculturists into the New World (Atley and Findlow 1984), these populations, more than their Palaeolithic predecessors, cave painters and big game hunters, were perceived as insignificant because of an almost complete lack of art and their microlithic industry {Clark 1978.2). A considerable amount of evidence in the last quarter of the century has shown that many prejudices toward hunter-gatherers in general, and Mesolithic peoples in particular, were unjustified4. Hunter-ga- 3 It became apparent in my numerous discussions with Dr. Jovanovic at the Institute of Archaeology that we were using Mesolithic to denote quite different phenomena. Once we passed this barrier, the largests gaps in our respective interpretation were bridged. 4 Although they are only rarely explicitly present in up-to date archaeological literature, it is not so for other disciplines: see Rodri-gue (1992) for a critique of current perception of domestication by cultural geographers who regard ritual sacrifice as major impetus for domestication of animals and portray preagricultural populations as:"roving and hungry hunters and gatherers," while farming is perceived '"as creating leisure time" (p. 417 and quoted literature, my italics). It is possible to observe that there is practically no interaction between human geography on one hand, archaeology, and anthropology on the other, or else this statement could hardly be understood given recent insight into different subsistence strategies and their work-loads. therers around the world "have for many millennia routinely manipulated plant and animal populations in diverse ways to optimise their use of them" (.Harris 1984; Harris and Hillman 1989; and examples quoted in Dennell 1985). Farming and foraging are considered as "overlapping, interdependent, contemporaneous, coequal and complementary domains in the Amazonia" (Sponsel 1989.37). Hunter-gatherers and agriculturists coexisted in many parts of Europe for several centuries, and in some areas, for millennia. Early farming communities were scattered amongst predominantly forager populations, and, depending on many of the historical contingencies, the spread of agriculture was sometimes rapid, sometimes gradual. It even retreated at times or showed no change over long periods of time {Dennell 1985-113)• The interactions between these populations were inevitable, and our perception of their frontiers, communications and avoidance strategies has been largely improved from information gathered by cultural anthropology on recent foragers and farmers. It is very important to stress, when dealing with past human populations, that our perception of possible interactions needs to remain open-ended so that interactions can be perceived as changing and fluctuating. Primarily, this is important since patterns of recent interactions are at least partly determined by the growing marginalisation of both subsistence farmers and foragers in the global industrialised economy. Secondly, the archaeological record preserves only a very limited portion of total interactions, namely, those that concern exchange, and can therefore only rarely provide insight into other aspects of it. Further, archaeology operates within time frames that largely surpass our personal experiences: since interactions between local foragers and farmers in the Iron Gates Gorge, in Portugal, or in Denmark, spread over almost one millennium, it would be unreasonable to think that they remained the same throughout that period and that they were always characterised by either mutualistic exchange or warfare. Also, the perception of cultural unity, as shown by recent research on interactions of foragers with farmers, does not necessarily stem from the same mode of life and "models of interaction which make insufficient allowance for the lack of coincidence between ethnic grouping and mode of production or the influence of different societies on each other may need to be revised" (Jolly 1996.234). The archaeological record produced by people belonging to two clearly distinct ethnic groups, one pastoralist, the other hunter-gatherer, can leave very similar material remains, which are determined neither by economy nor ecology, but by intention or anticipated mobility (Kent and Vierich 1989). Perception of subsistence farmers as dominant over hunter-gatherers in recent small-scale societies has been fairly well documented (Speth 1991 and quoted literature). This current perception could result from the actual physical and numerical dominance of agriculturists in the modern world and would be irrelevant for the patterns of perception in prehistory. In all instances observed in contemporary populations, cultural traits move from dominant farmers to their neighbouring foragers (Hodder 1982), while women from foragers groups move into their neighbouring dominant farmer units (Speth 1991; Zvelebil 1996a). Although many mythological and ritual practices have unrestricted and bi-directional flow from one society to the other (Lewis-Williams 1996 and quoted literature), hunter-gatherers participate in farmers' initiation rituals 'and not vice versa (.TurnbuU 1961; Jolly 1994). Current patterns of domination could stem from the fact that modern farmers are less dependent on forest products than foragers are on carbohydrates (Speth 1991 and quoted literature). The situation could have been very different at the time of the spread of agriculture into Europe. Therefore, extrapolating current perception as relevant for past human societies unreasonably assumes that agriculture had greater objective value than foraging, and would therefore be necessarily perceived as a better cultural adaptation by both farmers and foragers of the past. In prehistoric Europe it could be expected only within the context of the late Neolithic/Chalco-lithic, when farming starts having a serious impact on the environment (Willis and Bennet 1994). Only in that context could hypergyny at an agricultural frontier result in the acculturation of foragers (as suggested by Zvelebil 1996a.338). Before that, this perception would fluctuate as determined by local and chronological dependence relationships between foragers and farmers. Hypergyny, or the practice of "women of low socio-economic status to move up the status hierarchy, marrying less fortunate men in higher social categories who must marry down to find a mate" (Speth 1991.20), commonly observed in these interactions today, would not necessarily mean that hunter-gatherer women would marry into farmers' villages. At least in certain regions where foragers would be numerous, sedentary, and with sophisticated social organisation, as in the Iron Gates Gorge, they could have been perceived as dominant by small-scale farmers. Hypergyny would, then, take on a different direction. Optimal forager models for subsistence transitions argue that "sharper growth is not associated with broader diet but with subsequent increases in handling efficiency associated with practices that result in domestication" (Hawkes and 0'Cornell 1992 and quoted literature). Population growth rates would, therefore, decline with expanding diet breadth in the Mesolithic and would increase with improvements in the handling during Neolithic (Hawkes and O'Connell 1992). This conclusion, however, disregards the role that sedentism had on population growth, changes in subsistence and social complexity (Kent 1984; 1989a). It is becoming increasingly evident that both hunter-gatherers and farmers can and do have different mobility patterns that are both ecologically and socially sensitive (Kent 1989b), and that these patterns cannot be classified as a simple dichotomy. This pattern is primarily, though not exclusively, determined by the distribution and seasonality of plant (or anadromous fish) resources, not animals (Kent 1989a. 11). Intensification and localisation of resource exploitation leads to further sedentism, and importantly, aggregation which results in population increases (Kent 1989a. 10 and quoted literature). Sedentary aggregations result in the need for an arbiter, as disputes can no longer be settled by mobility, as for example in Hadza (Woodburn 1968.106). The immediate consequence of arbiter becomes incipient political differentiation concomitant with changes in social organisation. Since it is also known that many sedentary horticulturists hunt and manipulate plants for higher yields on a regular basis (Kensinger 1989; Speth and Scott 1989 and quoted literature), a clear-cut distinction between hunter-gatherers and small scale farmers can not always be made. This is no less true for modern societies than for the Early Neolithic of Europe. If we accept that agricultural practices are only those that "create(s) agroecosystems, which limit subsistence choice because of environmental transformation or labour demands" (Harris 1989; 1996b; Spriggs 1996), they could include the early Neolithic of southeastern Europe as evidence for agriculture, although they had little observable impact on environment. 2.4. Neolithisation of Europe As already stated, the transition from foraging to agriculture can be regarded as one of the most fun- damental cultural-ecological changes not only because it enabled large scale sedentism, subsequent population growth, and the appearance of the urban way of life, but also as it represents a conceptual shift in the perception of the world. Recent revival of interest in this phenomenon has brought about the understanding that agriculture developed independently in several areas of the world, and that modes of transition to food production from food gathering are specific for each individual geographic entity, and dependent on a number of particular conjunctions of circumstances in particular places at particular times (Harris 1996b.552). This revival has also stressed the importance of Upper Palaeolithic and Mesolithic adaptations for the development of the incipient agriculture, since small-scale cultivation is practised by many hunter-gatherers (Harris 1989; 1996b; Spriggs 1996). It is generally accepted that methods of food production, together with cultigens, were introduced to Europe from the Near East via Greece and the Balkan Peninsula. Einkorn, emmer, naked wheat, six row barley, lentils and peas, all imported corps, show up in Early Neolithic settlements in Macedonia, Serbia and Bulgaria (Zohary and Hopf1988.191- 193), while evidence for the introduction of domestic sheep predates full agropastoral economies in the Western Mediterranean (Donahue 1992; Geddes 1985). If the problem of origin of cultigens and knowledge of agriculture is solved, the mode of transition and the respective roles of hunter-gatherers and farmers remain unclear. Two major models are proposed: one involving the spread of farmers themselves, and the other based on appropriation of the new method of food production by indigenous foragers. Since the beginnings of European archaeology, the transition from foraging to farming has been regarded as a replacement of European Mesolithic cultures and populations by Neolithic ones, that spread through colonisation by Near Eastern farmers (from Lubbock 1865; to Harris 1996c). The Epipaleolithic and Mesolithic population was perceived as sparse to the point that until the 1950s it was generally believed that no important developments occurred between the Upper Palaeolithic and Neolithic in Europe (Clark 1980). Childe's (see for example Childe 1958) evolutionist understanding of the benefits of agriculture culminated in Ammerman and Cavalli-Sforza's (1971) theory which was to become one of the most influential models: wave of advance or "demic diffusion." This interpretation is based on the comparison of the available l4C dates and the genetics of living European populations (Ammerman and Cavalli-Sforza 1971). The authors argue for the spread of agriculture by the more or less continuous migration of people from the southeast to the northwest, at the average pace of 1 km a year (or 25 km per generation) with continuing population growth immediately following the advancing front of agricultural settlement (Ammerman and Cavalli-Sforza 1984). In later publications, the model came to incorporate the transfer of technology as a part of the process (Cavalli-Sforza 1996). Assimilation of foragers by farmers, through deforestation and acceptance of agricultural practices (competition and acculturation) and marriages of hunter-gatherer women to farmers (acculturation), would have created the "gradient" observed in the Principal Component Analysis (PCA) of genetic data. Authors such as van Andel and Runnels (1995) accepted and attempted to moderate this view. Since early Neolithic settlements in both Anatolia (Roberts 1991), and southeastern Europe (van Andel and Runnels 1995) were characterised by small scale and locally intensive cultivation (Sherratt 1980) and situated in areas, at least in the Balkans, that were not occupied by indigenous Mesolithic populations, they argued that migration occurred in discrete steps, "the interval dictated by geography and by the population growth in each of a slowly rising number of parent areas" (van Andel and Runnels 1995.497). Cavalli-Sforza argues for expansion rather than migration, as the former involves both population growth and replacement (Cavalli-Sforza 1996.56). He also endorses Renfrew's equation of this population as Indo-European speaking (Renfrew 1996) 5. In a recent critique, Fix (1996.627-628) has shown that the parameters that Cavalli-Sforza and colleagues used are far from well defined and fixed. He argues that clinal patterns such as those observed in European populations also can be produced by temporal gradients in natural selection. A similar opinion was forwarded in early 1980s by Meiklejohn (Meiklejohn 1985). The selective factor would be increasing disease intensities brought about by the "diffusion of agriculture, and especially, by the association between humans and newly domesticated animals" (Fix 1996.625). Although the cause of clinal distribution would be agriculture, the mechanism of its spread need not be movement of popula- 5 For an exellent critique see Sergent (1995). tion. It could, just as well, result from natural selection acting on a local population that has changed (through transfer of ideas and otherwise) its cultural (subsistence) practices (Fix 1996.641). Fix is aware of the problems associated with the proposed model: namely it is very difficult to demonstrate small fitness differences for many loci, and prove the association of the origin of certain diseases with domestication of animals, but his model is important as it proposes another look at the data and argues that it is "mistaken to use loci such as HLA, for which there is strong evidence of selection, as neutral markers of population movement" (Fix 1996.641). The two models discussed contrast most sharply in the demographic mechanisms for generating the clines and therefore Fix concludes that "knowledge of the population characteristics of the Mesolithic and Neolithic peoples could decide the issue" (Fix 1996. 641). Further, the craniometric analyses of European prehistoric samples failed to provide support for the model of population replacement (Harding et al. 1989; but see Petersen 1997). If, as proposed by Cavalli-Sforza (1996), not only farming, but farmers themselves originated in the Near-East and then expanded into Europe, it would require that the Near East was a "demographic cistern" that overflowed its surplus to Europe (Dennell 1985.119). Early and Middle Neolithic farming communities are rare and small in size in southeastern Europe, and even in the Near East early farming populations seem to have been very low in numbers. Inclusion of local foragers into the genetic pool through hypergyny and acculturation would only have marginal importance in this initial phase according to the clinal distribution of PCA and the iso-chronic map that is used to support the "demic diffusion" model. The Mesolithic population of Europe was interpreted as highly homogenous, showing a high degree of similarity with preceding Upper Palaeolithic people (Henke 1989.541) with clinal distribution and a continuous gradual change over time. This was suggestive of intensive gene flow between Late Pleistocene and Early Holocene populations in Europe. In Henke's view, "due to a low population density there were continuously, partly overlapping mating networks without any greater barriers to gene flow" (Henke 1989.560). The low population densities would increase only later in the Mesolithic with a tribal level of social organisation that was either of a short duration, or was not prevalent and therefore did not lead to genetic isolation (Constandse-Wester-mann and Newell 1989; Constandse-Westermann et al. 1984). Bocquet-Appel (1985) suggested that small populations could avoid extinction only by means of high migratory flow, involving interpopu-lation gene flow and exchange over large geographic areas. Because of this model and the supposed scarcity of resident hunter-gatherer populations, their role in the process of introduction of agriculture "has been disregarded or minimised" (Meiklejohn and Zvelebil 1991.129). That the Mesolithic population was sparse throughout Europe, has been questioned by recent paleodemographic studies (fac-kes et al, 2000; Meiklejohn et al. 1997;Jackes et al. 1997). However, a careful reading of the argument by Constandse-Westermann and Newell points to a greater regional sedentism in the Late Mesolithic and may not be in contradiction with current archaeological or demographic evidence, although Danish material seems to cast considerable doubt on this concept (Meiklejohnpers. comm.). As with many other lines of inference, recent paleodemographic studies of Portuguese and Danish material have shown that the picture is not only complex but also region-spe-cific (Meiklejohn et al. 1997; Jackes et al. 1997). Given the fact that the first farmers would be moving north into essentially unknown habitats, and progressively more marginal climatic conditions for the production of their newly domesticated plants, while the foragers were long time inhabitants with excellent knowledge of seasonal food availability, it is unlikely that the neolithisation of Europe could have been as swift without involvement of local Mesolithic inhabitants (Prim 1987). Coexistence and mutualistic exchange that was proposed recently for Central Europe, as opposed to confrontation and acculturation, therefore provides a likely and more flexible context for understanding the early interactions of farmers and foragers (Gregg 1988; 1991b). As Dennell points out (1985.118) lack of evidence for defence structures around early farming villages argues for peaceful interactions rather than raiding and warfare. This is not only in sharp contrast with later societies in temperate Europe but also strongly contradicts the attempt to equate incoming farmers with Indo-Europeans as Renfrew (1996) does. However, idealistic peaceful interactions have been contested by L. Keeley (1997) who argues that the transition to agriculture in central and northern Europe offers evidence for substantial amount of violent relationships, especially in the western realm of the Linearbandkeramik (LBK) spread. The diffusionist point of view, so influential in English-speaking archaeological tradition was criticised as early as the 1970s by Garasanin (1973; 1974b), and greater emphasis was placed on understanding local Mesolithic populations as active participators in the process of neolithisation (see also Guilaine 1976). Price's conclusion (1983.771) that "the end of the Mesolithic is not brought about by an advance of invading farmers but rather reflects a period of readaptation and adjustment to changing environments and new subsistence practices, often within the context of existing societies," stresses the importance of adaptations of Mesolithic foragers that enabled the transition to agriculture. It precedes the recent shift towards placing greater emphasis on the role of Mesolithic populations in Europe and their incorporation of farming techniques as a mode of transition to agriculture (Barker 1985; Dennell 1984; 1985; Hodder 1990; Thorpe 1996; Zvelebil 1996a; 1996b). As Barker (1997) noted, what was heresy in the academic core in the early 1980s has become orthodoxy in 1990s and in this new development the role of population movement has been underplayed. In light of this shift in direction, the appreciation of incipient indigenous cultivation and domestication that was proposed for Lepenski Vir by Srejovic fails to be as unlikely as it was back in 1972 when it was first published in English (Srejovic 1972). It is important to note here that, although the positions discussed above propose models for Europe as a whole, or imply continent wide implications in their titles, most of the studies of the spread of agriculture are based on the evaluation of the spread of the LBK culture limited to Central and West-Central Europe and, in terms of European periodisation (Tab. 1), contemporaneous with the Late Neolithic cultures of the Southeast: Vinca-Tordos and Dimini (Demonic 1988; Garasanin 1980a; 1980b; Lichardus and Lichardus-Itten 1985). The spread of "Cardium-impresso" ceramics in the circum-Mediterranean region, with little evidence of agriculture but with even earlier pre-pottery evidence for domestic Caprinidae (Batovic 1966; Benac 1979; Donahue 1992; Lichardus and Lichardus-Itten 1985), or "La Hoguette" pottery in the South-Western and Western Europe for which an African - more specifically Central and Eastern Saharan - origin was recently proposed (Wi-niger 1998), contradict the above model. The evidence provided by southeastern Europe, and the Balkans in particular, is rarely discussed in the theoretical literature or syntheses except for absolute dates that are used to argue the direction and pace of migration or spread of cultural influence. The explanation for the neglect of patterns of interaction and change in these "marginal" European regions is mostly due to searching for a "general processual model" so typical of 1960s, 1970s and even 1980s (as pointed out by Harris 1996c) to which these regions do not provide a good fit, but also to the fact that the majority of data are published within local research traditions even when they are published in English. "Indigenous archaeology" (Evans and Ras-son 1984) which has a long tradition in Balkans, is perceived as devoted exclusively to typological studies and as lacking the more sophisticated economic perspective and scientific approach of the "New archaeology". With the demise of the New archaeology in Anglo-American archaeology, that was rejected by Balkan researchers because of its apparent mechanicism (Garasanin 1996), "indigenous archaeology" has gained new importance for non-local archaeologists. This is apparent in the work of authors such as G. Barker and J. Chapman who not only use available data from the published reports but also discuss and sometimes incorporate the ideas and interpretations of local researchers. In conclusion to this section, we can state that replacement is not demonstrated, although it is neither impossible nor improbable, and that the change in subsistence practices was not as uniform as previously believed. As pointed out by Van Gijn and Zve-lebil (1997.3), "both the Mesolithic and the Neolithic were internally far more heterogeneous than we have recognised." The long coexistence of farming and foraging communities, that has been demonstrated or proposed for different regions in Europe, provides the general framework for this study. However, the cultural and biological identity of farmers needs to be examined on a regional basis. Based on archaeological evidence, we cannot assume that the same population was responsible for the neolithisa-tion of the Balkans as for the neolithisation of the Circum-Mediterranean, Central European, or Baltic regions, nor can we argue that modes of interactions between these populations would have been the same over the entire continent. Therefore, building meaningful continent-wide models has to repose on well documented regional occurrences. Further, comparisons between regions should be made on the basis of archaeological sequences and data, rather than testing preconceived models on local data, as there is an incipient tendency in the latter to search for adequate data in local manifestations that would fit the (usually) monocausal model (as is the case with Cavalli-Sforza's argument). This however, does not mean that comparisons and general models are not possible or meaningful. On the contrary, they become possible when the same level of insight is obtained for different regions and periods. The process of learning about the past, although embedded in the present, has to overcome this determinism, and rather then being unilinear, rooted in either local data or global theories, must incorporate both in order to transcend the present and reveal more about past than it does about our current agendas, which are implicit in Shanks and Til-ley's (1987) approach (see also a critique by Van Gijn andZvelebil 1997). In that respect, some basic premises of this work are: • Understanding the processes of neolithisation for each region must incorporate understanding of the Mesolithic substrate and its response to (among other factors): availability of agricultural knowledge or contact with agricultural settlements. • Responses of past foragers to contact with farming were determined by both economic and ideological strength of the local Mesolithic culture. In any of the regions they could have taken a number of forms that were not necessarily paralleled in other regions. Further, the mode of contact need not (or was even unlikely to) have remained the same over long periods of time in which these different forms of subsistence were practised by respective populations. • Introduction of material and symbolic (as much as we can perceive them) elements of one culture into the other, if they did not substantially change the subsistence base, can only be used as evidence for contact and not as evidence for acculturation. • Regardless of the ultimate origin of agriculture, the farming community with which local Mesolithic inhabitants were coming into contact, could have been at the time: of different geographic origin or of the same geographic origin. • This population in either of the above cases could have been: morphologically and/or genetically distinguishable or morphologically and/or genetically indistinguishable. 2.5. The choice of study area The first manifestations of the Neolithic in Southeastern Europe and the Balkans precede the LBK culture and show significantly different economic, social and ideological patterns (Benac 1979; Garasanin 1979; Srejovic 1979). Three major complexes are distinguished within the Neolithic of Southeastern Europe. Each covers relatively vast geogra- phic areas and includes groups that are more or less related. These are the Balkano-Anatolian complex of the Early Neolithic, the Balkano-Karpathian complex and the Occidental Mediterranean complex (Garasa-nin 1980). The Balkano-Anatolian complex includes (in the European part) Thessaly, South-East and South of the Balkan Peninsula (Thracia south of Sta-ra Planina and Macedonia), as well as an important enclave that spreads north into the Pannonian plain and the Karpathians. It includes groups defined as Protosesklo in Thessaly, Anzabegovo Vrsnik I in Macedonia, Karanovo I in Thrace and Gura Baciului in the Karpathian basin (Garasanin 1980.58). Although based primarily on the typological and stylistic analysis of ceramic production and habitation, it coincides with economic parameters. The major distinction between the Balkano-Anatolian and Balkano-Karpathian complex is that goat/sheep herding predominates in Thessaly (Bokonyi 1974; Barker 1985. 63), while cattle are characteristic of Starcevo-Cris-Koros farmers (Bokonyi 1974). Bokonyi argues for local domestication of cattle in Argissa Magula (Mi-lojcicetal. 1962) and Nea Nikomedeia (Higgs 1962) that was soon replaced by animal husbandry based on caprovines, and that gave rise to the Starcevo-Cri§-Kords complex of the Early and Middle Neolithic. Unfortunately no faunal data are published for Gura Baciului, and the fauna of the closest related settlement (that of Lepenski Vir III) is published without regards to LV Ilia being synchronous with Early Neolithic (Gura-Baciului, Thessalian tradition) and LV Illb with Middle Neolithic, classical Starcevo. In the Lepenski Vir III settlement wild animals predominate (74,5%) while bovines are most common among the domesticates (15,83%) (Bokonyi 1972). Goat/sheep remains follow and are also attested at the Mesolithic site of Padina although their exact provenience is not clear (Clason 1980). Also, the fact that in the Balkans there is a paradoxical absence of palynological and other evidence for agricultural impact on the landscapes until c. 4000 BC (Willis and Bennet 1994), points to a different scale and importance of agriculture in this period than for the Late Neolithic Dimini-Vinca-LBK agriculture. The fact that new cultigens and (some) domestic animals are introduced to Southeast Europe together with the spread of ceramic and polished stone axes can be conveniently used as a sign of a moving agricultural frontier. The question is how to perceive and study this frontier, and the interactions that "Mesolithic foragers" and "Neolithic farmers" could have had. Dennell states that from 5300 to 4300 BC there was a rapid expansion of agriculture by pot- tery-using communities living in "large, permanent settlements" across much of Southeast and Central Europe. "Thereafter, agricultural expansion into northern and western Europe was more gradual and seldom associated with large, year-round settlements until much later." {Dennell 1985.121). What he refers to as "large, permanent settlements" are "tells" such as Anzabegovo (Garasanin 1974a; 1979; Garasanin and Garasanin 1961; Gimbutas 1976), Karanovo, (Georgiev and Cickova 1981), or Argissa Magula (Milojcic et al. 1962) that could have resulted from non-permanent but repeated use (see also Bailey 1997). Starcevo I (Early Neolithic) and Starcevo II (Middle Neolithic) settlements that characterise the Central Balkans present a different picture both in terms of architecture and spatial organisation. They do not have the complex vertical stratigraphy of the "magulas" or "tells" and were most probably not re-occupied after being abandoned. Permanent occupation of both types of sites is questionable, since the extensive (shifting) agricultural practices require that new land is found whenever the one currently being used becomes too poor in minerals and necessary elements and therefore requires either enlarging the radius of exploitation (with rising cost of transport and protection of the crops) or moving a settlement. The latter seems to be characteristic of both the Anzabegovo and Starcevo types of settlements. Only with the Vinca intensive agriculture (Garasanin 1979) and LBK introduction of crop rotation {Wilierding 1980) does it become possible to have permanent settlements. Dennell's (1985) appreciation of size and permanence (and associated mobility) of different groups in Balkans prehistory, leads him to assume that hun-ter-gatherers would perceive the agriculutralists as having "more substantial houses, novel items such as pottery, polished stones and so forth" and would therefore be more inclined to observe them as "better off', which would result in hypergyny and a loss of the Mesolithic population to incoming farming groups. Although this model does not fit all the archaeological data of the region (especially in terms of permanence of the Early Neolithic settlement and perception of farmers by foragers), determining the frontier for the interactions of early farmers and local foragers as porous, allowing transfer of people, resources and techniques across the border, seems appropriate in at least some of the early interactions and has been a crucial breakthrough in our understanding of forager-farmer interactions (Zvelebil 1996a and quoted literature). Since hun- ter-gatherers would have a larger radius of movement, they would more often come in contact with local farmers near the latter's villages, and could thus appropriate their knowledge and techniques, or be appropriated (one way or another) by farmers. 2.6. Why Mesolithic population of the Iron Gates Gorge? Both the Epipaleolithic and Mesolithic of the Balkans, with the exception of the Iron Gates Gorge and some smaller sites in Montenegro and eastern Serbia, are understudied. However, the Iron Gates Gorge has been excavated extensively and has produced evidence for all of the Mesolithic economy and structure (as discussed above). The sites in the area offer a unique possibility to observe possible (and probable) interactions between foragers and farmers over a substantial time span. These interactions can be further traced in the skeletal record and potentially some of the biological features of them could be delineated. Ideally, we would have as large a collection of Neolithic sample from the region, but it is not the case. We could lump all of the osteological material from early and middle Neolithic of the Balkans together, but this could cause the general trends of the region to obscure the specifics of interaction and therefore, a more regionally restricted approach was deemed better. With the stated problems in mind, the region was chosen for several reasons: • The neolithisation of this area preceded the neo-lithisation of all other parts of Europe except the Aegean. • The extent of excavations on a number of sites on both banks of the Danube has been unparalleled in the last three decades as they have exposed substantial living surfaces amenable to detailed archaeological analyses. • Through abundant architectural features, artefacts, faunal, plant and osteological remains, the sites provide insight into both cultural and populatio-nal consequences of contact. • Since the archaeological coexistence of the Mesolithic Schela Cladovei-Lepenski Vir culture and Neolithic Starcevo culture has been established (Bon-sall et al. 1997; Boroneanf et al. 1995; Radovano-vic 1992; 1994; 1996a; 1996b; 1996c) the degree and mode of interaction between the bearers of these cultures can potentially be traced in human osteological remains. • Since this is one of the largest Mesolithic skeletal samples in Europe, information provided by skeletal data can be crucial in our understanding of the resulting biological change in the population under transition and potentially elucidate its causes. 2.7. Questions This research attempts to reconstruct the extent of interactions between farmers and foragers through examining two aspects of skeletal material from four of the most important sites excavated in the region: Lepenski Vir, Hajducka Vodenica, Vlasac and Padina. Here I propose a list of questions that this work will attempt to answer. The choice of material, theoretical premises and methods, are discussed in subsequent Chapters (3 and 4). The results of the analyses are presented and discussed in Chapter 5 and a discussion and conclusion offered in Chapter 6 and 7 respectively. 1st question: In light of a proposed porous frontier between Mesolithic and Neolithic cultures in the studied region, traceable in the archaeological evidence, can we presume the interactions between bearers of these respective cultures? 2nd question: If there was an interaction between foragers and farmers, are we able to perceive it through the study of anthropological material? 3rd question: If the spread of agriculture was a consequence of cultural transfer alone, is it likely that the biological profile of the population will change significantly? In which direction? 4th question: If the Early Neolithic population of the Balkans is different from the Iron Gates Gorge Mesolithic population, can we trace the effects of contact on the latter? 5th question: If we can argue for a change in the Iron Gates population as a result of contact with Neolithic population WHEN did this change occur? Did it happen at the time of the first contact, which did not bring about the change in subsistence, or later, when the subsistence changes sufficiently to determine Iron Gates sites as Neolithic? 6th question: In either case, what explanation can we propose for the change: genetic admixture, replacement or changing ecology (occupation/ nutrition)? Hajducka Vodenica Cuina Till Schela Cladovei Ilisova . Velesmi 3. THE LEPENSKI VIR CULTURE This chapter is an attempt to summarise the archaeology of the Mesolithic to Neolithic transition in the Iron Gates Gorge, through some of its major features. In keeping with the goal of examining the possible biological (populational) effects of contact of Mesolithic foragers with Neolithic farmers, the chronology and the evidence of contact in the archaeological record were examined. The four sites from the right bank of the Danube are presented together with details of chronological assessment of each of the skeletons within the sites. Because of the fact that the Mesolithic component is present on all four of the sites, while Neolithic (as defined earlier) could be ascertained only at Lepenski Vir, these interactions are examined from the Mesolithic perspective. Accordingly, considerable space is allocated to the description of the basic features of the Iron Gates Gorge Mesolithic. The Neolithic component at the Lepenski Vir site fits well within the Balkano-Karpa-thian complex as described by Garasanin (1979). Therefore, it will only be discussed briefly in reference to pottery as the most prominent evidence of the contact at the sites. After the examination of the proposed methodology (Ch. 4) and the presentation of the results (Ch. 5), discussion on the morphological and biological affinities of each of the examined populations is proposed in Chapter 6. 3.1. Research, publications and interpretation history As recently as 1950 the Central Balkan was deemed to be uninhabited during the Mesolithic period (Srejovic 1989). The surveys and excavations that were undertaken before the building of the dam on the Danube downstream from the Iron Gates Gorge during the 1960s unearthed a number of Holocene sites in the Gorge that were ultimately assigned to the Lepenski Vir-Schela Cladovei cultural group. On the Yugoslav bank of the Danube (Fig. 3), Padina (Jovanovic 1972; 1974), Stubica, (Jovanovic 1971; 1974), Lepenski Vir {Srejovic 1968; 1969; 1971; 1979; Srejovic and Babovic 1983), Vlasac {Srejovic and Letica 1978), Hajducka Vodenica (Jovanovic 1984a), Ve-lesnica (Vasic 1986b), and Kula (Slavic 1986) were excavated over a 20 year period. On the Romanian bank of the Danube, Privod, Alibeg, Ilisova, Razvrata, Os-trovul Banului, Schela Cladovei, (Boroneant 1973) Vodneac, Cuina Turcului (Boroneanfn.d'.), Climen-te I and II, (Boroneant n.d.; Radovanovic 1981), Veterani terrace, Icoana, Ostrovul Corbului {Mogosa-nu 1978), Ostrovul Mare 875, and 873 {Boroneanf 1980) were excavated in successive campaigns from 1964 to 1973. From the 1970s onwards, a number of important volumes appeared: an English language compilation of the known data by Tringham {1971), a monograph on Vlasac by Srejovic and Letica {1978), the Cuina Turcului final report {Paunescu 1978), the papers from the conferences on the problems of neolithization held in Sarajevo in 1977 and Krakow 1979 (Benac 1978; 1980; Garasanin 1978; Srejovic 1978; 1980; Mikic 1980; Kozlowski and Kozlowski 1978), the conferences on the Mesolithic in Potsdam, Edinburgh {Boroneanf 1989; Srejovic 1989; Voytek and Tringham 1989; Chapman 1989), Grenoble 1995 {Radovanovic 1995; Boroneanf et at. 1995), and conferences on the Mesolithic and the chipped stone industries of the earliest farmers at Krakow {Kozlowski 1982; Kozlowski and Kozlowski 1987; Paunescu 1987). As well, a number of analyses were published on the chipped stone industry: from Padina {Radovanovic 1981), Vlasac (Prinz 1987), and Lepenski Vir {Kozlowski and Machnik 1980) as well as on floral and faunal remains (Cla-son 1980). Syntheses also appeared at the time: in Praistorija Jugoslovenskih Zemalja [Prehistory of Yugoslavia] by Srejovic, Benac and Garasanin (1979), then in Esquisse d'une Prehistoire de la Romanie {Mogosanu 1983; Dumitrescu etal. 1983), followed Ostrovul Banului Fig. 3• The map of the Iron Gates Gorge. The sites analysed in the present study are outlined in bold. (Adapted from NCARTA 1998). by a synthesis on Banat in the Neolithic by Lazarovi-ci (1979)• Lepenski Vir was briefly discussed in La Protohistoire de L 'Europe (Lichardus et al. 1985), as well as in Domestication of Europe (Hodder 1990), in Europe in the Neolithic (Whittle 1996), and in a number of articles by Chapman (Chapman 1989; Chapman and Dolukhanov 1993)• Radovanovic provides an excellent historical and critical overview of the phases of research and publications in the region (1996a. 2-10). The monograph by Radovanovic is the first synthesis based on published works, field documentation and a new approach to a combination of stylistic, chronological, and material analysis for the entire region. While she draws heavily from all local resources available, Radovanovic manages to incorporate a new understanding of a variety of issues into her appreciation of the data collected in the field. Benefiting from firm knowledge of the regional archaeology, access to field documentation, as well as a wide array of theoretical perspectives, the volume successfully incorporates different intellectual traditions into a comprehensive study. More importantly, this volume provides a wealth of information on particulars of the regional Mesolithic so often lacking in theoretical discussions by non-local scholars. A number of analyses that appeared in the 1980's on chipped stone assemblages from Lepenski Vir, Vlasac and Padina established continuity between the Iron Gates Epipaleolithic and Mesolithic. However, different synchronisations among phases of the sites were proposed (Radovanovic 1996a.9 and quoted literature). Late Palaeolithic finds from Cui-na Turcului I were perceived as predecessors of the Lepenski Vir culture by Srejovic (1989.54) and he offered the following periodisation: ® Late Palaeolithic, (Cuina Turcului I); © Epipaleolithic (Cuina Turcului II, Ostrovul Banu-lui I—II); %£&* -Ocxvut? ' »°<,xxix aatebJ'Sc.,.^ XVIIjJ jV^ <9. ...p® • VII ' vn^T _„ ^ ix IV Syf XII ■sf XXXV <"> •XIII > O f'7. /T2] ? s \xiv. i fe vVir-^-i. XXVl x"kvn 1 xxiv XXII! n- Fig. 8. Composite plan of the Lepenski Vir II settlement. ^Adapted from Srejovic 1969-78-79, Fig. 18). of rebuilding activity, and the house 57 stands out as the largest house in the whole sequence. Lepenski Vir was abandoned for a period of time (testified by a thin layer of brown loessic sand observed at places between the LV I and LV II settlements (Srejovic 1966.13; Radovanovic 1996a.ll3). There are significant differences in the organisation of the settlement, with supporting walls and artificial terraces. However, the houses are often at the same spot as the houses of the previous phase {Srejovic 1969.78-81). Lepenski Vir III is not considered by Radovanovic as she deals only with Mesolithic strata. It is divided into LV Ilia perceived by Srejovic as early Neolithic and LV Illb - Middle Neolithic Starcevo settlement. If the evidence for contacts with Neolithic communities from LV I is taken into account, the designation of the LV Ilia into early Neolithic is somewhat problematic. It is supposedly synchronous with Gura-Baciu-lui and arguably different from the classical Starcevo complex - at least based on its ceramic production. Since the ceramics from Lepenski Vir have not yet been thoroughly examined in the light of new interpretation of Starcevo development (Tasic 1998) further discussion is fruitless. However, the high percentage of game and fish, as discussed earlier, argues for substantial influence of local tradition. Absolute dates for Lepenski Vir were obtained from charcoal samples and have caused considerable discussion, especially on the dates of the Neolithic settlement. Since these are not crucial to our analysis, only the dates reported by Bonsall etal. {1997), derived from human bone collagen are examined here. Unfortunately, R. Lennon, who collected the bone samples in 1989, made only Lepenski Vir III material available for processing, although many more specimens were collected 8. The dates (Tab. 5) have quite a wide range (from 6993 to 5593 BC) although they are all derived from the Neolithic phase of the settlement. These dates do not coincide with Radovanovic's and Srejovic's archaeological determination of the site sequence and present a considerable problem. However, since they are only few and all of them are from a single phase in the sequence, the phase that seems to have been the most readily discerned by the field crew and since a possibility of them being contaminated with the 11C from the ground water, Skeleton number Lab ID ™C age BP cal BC age 1a cal BC age 2a 31a OxA-5827 7770+90 6621-6462 6993-6414 44 OxA-5830 7590+90 6463-6267 6552-6189 32 OxA-5828 7270+90 6178-5990 6229-5897 88 OxA-5831 7130+90 6018-5970 6159-5763 35 OxA-5829 6910+90 5840-5667 5954-5593 Tab. 5. Absolute dates. Lepenski Vir III settlement ffrom Bonsall et al. 1997.Tab. 6). 8 As evidenced by bones from all three periods that have been severely damaged by collection of unnecessarily large samples, more dates could have been available. it is hard to incorporate these dates in the present study. Until more dates are available their value remains tentative. Human remains examined in this study come from all three settlements. They comprise 190 individuals from 134 graves plus 34 adults, five subadult individuals and three newborns from different unidentified contexts. Of note is that many of the "scattered" adult remains bear traces of ochre. However, until a more detailed study of taphonomy is done and full access to field data is available, the explanation for these bones cannot be offered. Some of them are no less well preserved than some of the remains with associated grave numbers. However, since none of them have sufficient features for either demographic, morphometric or non-metric analysis they are excluded from further discussion. MNI for the site as a whole amounts to 190 individuals (as at least theoretically these 42 "scattered human remains" could have belonged to them). Skeletal representation ranges from small fragments to complete skeletons. Of them 101 individuals were buried in single, 58 in double graves, three graves contained three individuals, while four, five, six and seven individuals were buried in one instance each. Only 25 individuals were assigned sex based on pelvic morphology (same procedures as described for Padina). Females are represented by 11 positive determinations and males by 14. The site has a particularly high proportion of neonatal skeletons: 51 individuals or 27%. These burials are most often associated with house construction (underneath the floor) but also they are found within adult burials as well, often represented by only one bone fragment. A further 33 skeletons belong to children of different ages: 2-5 years old by seven individuals, 6-11 years old by 13 individuals, 12-18 years old by eight individuals, while for five subadults it was impossible to determine age. The total for adults is almost identical as for subadults (83 compared to 84) while no age could be assigned to 23 individuals. As with the collection from other sites, previously non-identified individuals were given the same number as the noted grave with the additional serial number in brackets. According to the presented chronology of the site and division of the strata into the Mesolithic, Mesolithic/Neolithic contact and Neolithic periods, skeletal remains were assigned to one of the following phases according to site documentation (kindly provided by Prof. Srejovic, the principal investigator) for the Neolithic burials or Radovanovic's analysis of Mesolithic burials (1996a. 174-189): The following 32 skeletons have been assigned to the Mesolithic period: single burials: 3; 21; 22; 46; 60; 61; 67; 69; 110; 111; 112; 113; 117: 118; 119; 120; 121; 132; 133; double burials: 50, 50(1); 64,64(1); 99,99(1); 102, 102(1); 109,109a; multiple burials: 101, 101(1), 101(2). It is interesting to note that 17 of these individuals are newborn babies found underneath the house floors (all of the numbers above 100). To the Mesolithic/Neolithic contact period belong following 68 individuals: single burials: 11; 12; 15; 16; 17; 23; 26; 28; 46; 68; 70; 90; 91; 92; 94; 95; 96; 97; 100; 103; 104; 105; 115; 116; 122; 126; 127; 128; 129; 130; 131; 134; double burials: 7a, 7b; 13, 13(1); 14,14(1); 45a, 45b; 63, 63(1); 74, 74(1); 93, 93(1); 98, 98a; 99, 99(1); 106, 106(1); 107, 107(1); 108, 108(1); 114,114(1); 123,123(1); 124,124(1); 125,125(1); multiple burials: 54a, 54b, 54c, 54d, 54d(l), 54e. Newborns represent 29 of these burials, most of them found underneath the house floors once the site was removed onto a higher terrace, as evidenced by their high sequence number (>100) assigned during excavation. To the Neolithic period Lepenski Vir Ilia and Illb belong the following 40 individuals: single burials: 1; 4; 5; 6; 8; 9; 20; 31a; 33; 35; 37; 38; 39; 42a; 43; 44; 48; 51; 53; 56; 57; 59; 66; 71; 88; double burials: 18, 18(1); 19, 19a; 32a, 32c; 52, 52a; 55a, 55b; 73,73(1); multiple burials: 83a, 83a(l), 83b. Unfortunately, for 48 burials there was not enough information to provide chronological assignment: single burials: 2; 10; 24; 25; 29; 36; 40; 41; 49; 58; 62; 65; 72; 75; 76; 78; 86; double burials: 77, 77(1); 80; 81; 82; 84, 84(1); multiple burials: 27a (27a+e), 27(27b), 27 (27C), 27 (27d), 27(1), 27 (271), 27(2); 34a, 34b, 34c; 79a, 79b, 79c; 85, 85a, 85b/85(l)/, 85b; 87, 87(1), 87(2), 87(3), 87(4); 89a, 89b, 89b(l). The skeletal material has received little detailed publication. Presented in Srejovic's book (1969.239-257) by Nemeskeri in a preliminary report it gives little information on the structure of the population. Nemeskeri, in keeping with the traditional approach of Central European anthropology was most interested in demographic profiles and typological determination of sub-populations. He distinguishes 2 major types further divided into 2 subtypes each, and concludes that for the differences between early and late population of Lepenski Vir (from the phase I to that of Ilia) to have evolved in loco, it must have taken 125 generations, or 2500 years. He therefore discarded the possibility of local evolution and argues for abrupt population change (Nemeskeri 1969. 255). Mikic has dealt with the entire Iron Gates Gorge series in his works on the process of neolithi-sation in Iron Gates Gorge (Mikic 1981a; 1989) and argues for the in loco evolution. Zoffmann (1983) has made an important contribution to anthropological publication on Lepenski Vir, and although sex and age determinations for individual skeletons were not reported, I was able to use the original documentation (kindly provided by Prof. Srejovic in 1996) in which sex and age determinations were given by Zoffmann. However, in the following season, in keeping with revision of sex determination for osteological material from other sites, I have reassessed sex using a more conservative approach based exclusively on pelvic morphology. It was interesting to note that differences in sex assignment were least important between my assessments and hers, while they differed considerably between both my and Nemeskeri's results, and my and Zivano-vic's results. The major reason for this could be that the two later authors based most of their conclusions on cranial remains. 3 5 3. Vlasac Vlasac was found at the very end of the campaign in the late summer of 1970. In the autumn of 1970, 432m2 of this site, situated in the Upper Djerdap Gorge downstream from Padina and Lepenski Vir, were excavated. A further 208 m2 were excavated in 1971, right before the inundation by the accumulation lake of the Djerdap Hydro-plant. In less than four months, the team of archaeologists, geologists, architects, and students unearthed 43 dwelling structures, 87 graves and more than 35 000 mobile objects. The monograph of the site was published in 1978 and is the most comprehensive publication on archaeological, environmental and anthropological data on any individual site of the Lepenski Vir culture (Bokonyi 1978; Bnczko el al, 1978; Carciumaru 1978; Srejovic and Letica 1978). The graves are treated and presented individually with relevant data on position, orientation, age and sex, and accompanied by drawings and pictures (Figs. 9, 10 and 11). The anthropological report is extensive and besides chapters on methodology, demography, pathology, dating, and sex and sexualisa-tion (sexual dimorphism change over time) gives individual data for each of the skeletons (Nemeskeri 1978; Nemeskeri and Lengyel 1978a; 1978b; Nemeskeri and Szathmary 1978a; 1978b; 1978c; 1978d; 1978e). It is of extreme importance to any of the metrical analyses, and also has all the relevant information on the dates obtained from human bones, useful in comparisons of stratigraphic assignments by researchers with Radovanovic's (1992; 1996a) chronology discussed below. However, a revision of the osteological material has shown numerous discrepancies between Nemeskeri's and my assessment of MNI and sex. During the 1996 campaign this difference started to appear, first and foremost in the number of individuals per grave. My first impression was that poor storage conditions had caused some mixing of the material. In 1998 campaign, this conclusion was dropped for a number of reasons: the mixing of the material had to be considerable to allow for such large discrepancies, the "extra" individuals were represented either by fragments of long bones, or very small fragments of skull. At least in one case (grave no. 7) a decorated bone implement (Fig. 12) was found with the post-cranial remains. In at least one case a whole coxal bone could be reconstructed where Nemeskeri assigned sex on the basis of the skull (grave no. 4a). The same coxal bone with embedded fragment of bone projectile was not discussed in his chapter on paleopathology (Roksandic 2000a). These instances have supported the conclusion that different results that Nemeskeri and I found in respect to both MNI and sex assessment stem from different weight accorded to the skulls and the postcra-nial skeleton in both of the analyses and also points out the benefits of detailed reconstruction of skeletons that was undertaken in the 1998 field season. More relevant information on the burial ritual is expected from the forthcoming analysis of skeletal representation and taphonomy of the material. According to the building horizons, Srejovic and Letica have divided the site into three chronological phases of the Mesolithic settlement (Vlasac I, II, and III) and one of the Neolithic (Vlasac IV). Since no human skeletal remains were associated with the lat- Fig. 9. Vlasac West. Position of graves ^Adapted from Srejovic and Letica 1978.Fig. 57/ ter, it will not be discussed in detail. Radovanovic has observed significant changes in burial practices over time and has proposed a division of formal disposal areas into two chronological phases. Based on the published material, as well as field documentation (kindly provided by late Prof. Srejovic, the principal investigator), she was able to distinguish an important change in burial practices that began to occur at the end of the Vlasac I but certainly were present in the Vlasac II phase (Radovanovic 1996a. 187-218). In the early Vlasac I phase she distinguishes a formal disposal area in the upstream Western Sector (Fig. 9) with parts of the Central Sector (Fig. 10) and a habitation area in the Eastern Sector. The situation changes in the Horizon II when habitations are clearly present on the border of the rocky plateau facing the river at both the Eastern (Fig. 11) and Western sector of the settlement. Burials are distributed along the whole settlement area. Some of the burials from Vlasac II are associated with the early and some with the later formal disposal areas. However, both Vlasac I and II would belong to the Mesolithic pre-contact period and only Vlasac III would chronologically belong to the period when the contact with the Neolithic populations was possible. All of the cases where Radovanovic could not certainly distinguish between the Vlasac II and III burials are treated as Mesolithic/Neolithic contact. However, a separate test was run with these individuals included in the Mesolithic group since contacts between Lepenski Vir and surrounding farming population^) is least attested in Vlasac of all of the sites: no pottery was found in these layers, and all of the Pre-Balkan Plateau flint was explained as intrusive (Srejovic and Letica 1978; Kozloivski and Kozlowski 1982). Five l4C dates (Tab. 6) derived from human skeletal remains are given in Bonsall et al. (1997) and they are well in accordance with Radovanovic's (1992; 1996a) determination of the burials phases since only Grave 24 is determined as belonging to the Contact period. Another set of dates was calculated on the basis of nitrogen and fluorine content (Buczko et al. 1978). The authors acknowledge that the content of nitrogen and fluoride is also dependent on climatic changes and therefore propose two different values for each of the skeletons. These values all fall within the range of 5250±100 to 5900+200 for the column A and between 6250+150 and 7400+300 in the column B. They also provide a set of more probable dates based on the relative - archaeological chronology. Since the whole process depends on the archaeological data, these absolute dates are perceived as uninformative and are not discussed further. All of the human skeletons from the site belong to the Mesolithic settlement. Human skeletal remains comprise 164 individuals from the 84 reported graves. This differs significantly from the 119 individuals reported by Nemeskeri (38% increase). These "extra" individuals are represented by very small cranial or postcranial fragments. No scattered human remains were reported among the faunal remains. The skeletal part representation varies from fragments of bones to whole skeletons. Of these, 45 were buried in single graves, 44 in 22 double graves, 13 graves contained remains of three individuals each (39); five graves had four individuals each (20); two graves contained remains of 5 individuals each (10); and one grave contained six individuals. Adults represent the majority of the sample: 108 individuals or 66%. Young adults are represented by 21 individuals, and old adults by 40; for 47 of them it was possible only to state that they were adults. Of the 47 subadult individuals 26 are of neonatal (or perinatal) age (16% of the total sample or 56% of the subadult sample), one was a child between 2-5 years, 8 children were between 6-11 and 6 between 12-18 years old. For six of the subadult skeletons the age could not be determined. In nine cases it was not possible to determine if the individual reached adulthood. As with other samples from the series, the approach to age was very conservative and age was assigned in deliberately broad categories. Positive sex determination was possible in 41 cases of the total adult sample (38%) of which 26 were determined as females and 15 as males. A further 16 were determined as possible females and 31 as possible males on the basis of robusticity. The larger number of determinable females could be due either to more elements (such as preauricular sulci) being significant to the female pelvic morphology or to a cultural agent. It will be discussed later with data on size and robusticity. In keeping with the marking of the other sites, "extra" individuals within graves were given a grave number from existing documentation and publication and an additional in brackets. According to the presented chronology the following 125 individuals are determined as Mesolithic pre-contact burials: single burials: 7; 8; 10, 11; 13, 20; 25; 28, 30; 31; 32; 33; 34; 37; 38; 39; 40; 41; 44; 59; 61; 63; 68; 72; 79; 81; double burials: 9, 9(1); 12a, 12b; 19, 78a •78 0 1 2 3m 71 69 70 69a 72 ff 58 67 58a» tiSb 43 Fig. 10. Vlasac Central Section. Position of graves ^Adapted from Srejovic and Letica 1978.Fig. 58). 0 I 3m 15 fW o ^.--25 o—•=«. 14 13 I7&1/2, supraorbital notch snl/snr multiple notches supraorbital foramen sfl/sfr absent/ present/ multiple marginal tubercle tuberculum marginale tzl/tzr present/ absent infraorbital suture isl/isr absent/ partial/ complete absent/ internal division multiple infraorbital foramina mifl/mifr two foramina/multiple zygomatico-facial foramina zffl/zffr absent/ large/ small/ multiple parietal foramen pf absent/ parietal/ sutural epipteric bone ebl/ebr absent/ present coronal ossicle cbl/cbr absent/ present bregmatic bone breg absent/ present sagittal ossicle sag absent/ present apical bone apic absent/ present lambdoid ossicle laml/lamr absent/ present asterionic bone astl/astr absent/ present ossicle in occipito-mastoid suture occml/occmr absent/ present parietal notch bone parnl/parnr absent/ present absent/ complete single/ inca bone inca bipartite/ tripartite/ partial condylar canal concl/concr non patent/ patent absent/ partial internal/ divided hypoglosal canal hypl/hypr partial within canal/complete internal/ complete within canal flexure of superior sagittal sulcus flex right/ left/ bifurcate foramen ovale incomplete foil/foir absent/ partial/ no foramen foramen spinosum incomplete fsil/fsir absent/ partial/ no foramen pterygo-spinous bridge psbl/psbr absent/ trace/ partial/ complete pterygo-alar bridge pabl/pabr absent/ trace/ partial/ complete tympanic dehiscence tdl/tdr absent/ foramen/ full defect auditory exostosis audtl/audtr absent/<1/3/1/3-2/3/>2/3 large s no d/ no s and deep d / suprameatal spine and depression pael/paer small s and no d/ small s small d/ large s small d/ large s deep d mastoid foramen location mffl/mffr absent/ temporal/ sutural/ occipital/ sutu. and temp./ occ. and temp. mastoid foramen number mfnl/mfnr absent/1/2/>2 sutura squamo mastoidea ssml/ssmr present/absent mental foramen mefl/mefr absent/1/2/>2 mandibular torus matl/matr absent/ trace/ moderate/ marked maxillary torus maxl/maxr absent/ trace/ moderate/ marked palatine torus pal absent/ trace/ moderate/ marked absent/ near mandibular mylohyoid bridge location mhbll/mhblr foramen/ center of grove/ both with hiatus/ both no hiatus mylohyoid bridge degree mhbdl/mhbdr absent/ partial/ complete Tab. 8. Cranial traits examined in the study (I = left and r = right for paired bones). POSTCRANIAL CHARACTERS WITH CODE AND SCORING SCHEME character CODE scoring atlas bridging posititon abpl/abpr absent/ lateral/ posterior atlas bridging degree atlas abdl/abdr absent/ partial/ complete facies articularis condilaris partitum faal/faar absent/ partial/ complete dens axis isolated denai fused/ unfused fovea costo-clavicularis deep fccl/fccr absent/ present shallow/ semicircular / >2/3 suprascapular foramen or notch ssfl/ssfr notch/ foramen accessory acromial articular facet aaafl/aaafr absent/ present unfused coracoideus uncol/uncor fused/ unfused glenoid fossa extension gfel/gfer absent / present ligament teres in cavitas glenoidalis Itcgl/ltcgr absent / present absent/ small perforation/ perforatio fossae olecrani pfol/pfor multiple sp/1-2mm/ 2-5mm/ > 5mm supratrochlear spur stsl/stsr absent/ present fossa bicipitis radii fbrl/fbrr absent/ present unfused processus olecrani upol/upor fused/ unfused fossa faciei lunatae ffll/fflr absent/ present Allen's fossa alfl/alfr absent/ present third throcanter ttl/ttr absent/ present Poirier's facet of extension pofl/pofr absent/ present Vastus notch vnl/vnr absent/ <607 60-90Y>90° Squatting facets on distal tibia sfl/sfr absent/ present Squatting facets (talus) superior surface, anterior to the articular sftl/sftr absent/ present facet for the tibia Shape of the talar articular ctasl/ctasr discrete facets/ anterior and surface on calcaneus middle joined/ all 3 joined Tab. 9. Postcranial characters examined in the study (I = left and r = right for paired bones). example, unfused processus olecranii can be a non-metric trait in adults, while in subadults it is associated with a certain stage of development of the skeleton. This has reduced the total number of individuals examined from 438 (MNI) from all four sites to 259 (MNI) adult individuals. 2nd step - Since the chance of purely random significant correlation occuring on the tested samples becomes greater with the number of correlation tests performed (:Talligpers. comm.), the first step in the procedure was to remove all of the variables that could not be observed (both as absent or present) on at least 10% of the examined adult sample. This resulted in the elimination of the following variables with the number of possible observations in brackets: ISL (20), ISR (23), MIFL (18), MIFR (20), CONCL (25), CONCR (27) FOIL (7) FOIR (12), FSIL (13) FSIR (8), PABL (17), PABR (16), TDL (4) TDR (12) in cranial traits and ABPL (12), ABPR (9), ABDL (12) ABDR (9), FAAR (12), DENAI (24), SSFL (4), SSFR (2), AAAFL (11), AAAFR (11), UNCOL (27), UNCOR (18), GFER (26), FFLL (21), FFLR (16), in postcranial, or a total of 32 variables. 3rd step - In the studied population, a number of traits had very low incidence of positive values across the sample (less than 5). As, depending on the sample size, a small absolute number of occurrences can produce biased results, the following 16 variables were excluded even before their frequencies within subpopulations were examined: MET (2), EBL (0), EBR (1), CORL (2), CORR (2), BREG (1), OCCML (1), OCCMR (3), INCA (4), GFEL (0), STSL (4), STSR (4), FAAL (2), FBRL (2), FBRR (3), FCCL (4), FCCR (2), UPOL (0), UPOR (0). 4th step - Of the remaining 55 variables another group of characters, those with low overall frequencies, were checked against chronological and spatial subpopulations in order to assess their overall variability. If the traits show low variability within the population, they will tend to reduce the interpopula-tion difference in statistical analysis, as they have a negative, reducing effect on the variance of the MMD (Sjm-old 1977; Molto 1983-113)■ Sj0vold (1977) recognises two types of low variability traits: those that have reached fixation in every sample studied, and others that have very low uniform incidence in any set of population samples. Rather than using the j} or Fisher's exact between samples test to exclude the variables for which the significant difference is not obtained in at least one pairwise comparison (as suggested by Sjovold 1977), the empirical results that Molto (1983-114) reported for an Ontario Iroquois sample were applied. In Mol-to's study (1983-115) the largest range of frequencies among the traits that had low variability was 7.1 (for example, 0.0% in one sample to 7.1% in another). Molto has excluded these traits from further consideration and kept those with minimum range in any of the samples equal or greater than 10% (e.g. 21% in one and 31% in another). By using this observation as a rule of thumb in the present study, rather than increasing the possibility of finding statistical significance (where there might be none) through a large number of tests performed, following traits were determined as having low overall variability and excluded: ASTL, (14.29-20), ASTR (17.86-21.74), PARNL (8.5-16.67), PARNR (8.7-14.29), PAEL (0.00-6.25) and PAER (0.00-3.03). rature, there is no definite pattern of preference according to sex, but the general trend of predominance in females is reported (Hasuer and de Ste-phano 1989 and quoted literature). The trait was excluded from further consideration. 7th step - Since the number of variables thus obtained was still sufficiently large and in view of the poor preservation of the sample, it was felt that recording frequencies in individuals rather than sides, as well as pooling sides, would result in reducing bias, especially in the very restricted Neolithic sample. Tests of side correlation were performed on all of the pairwise traits. The ones that showed correlation were excluded. In doing so the risk of increasing the probability of false correlation was ignored, as potential benefits in increasing the number of observations outweighed the concerns. 8. Remaining traits - Of the remaining 26 traits, further comparisons have eliminated coronal ossicle (No. 7 on the list of traits) because of very low variation (0-2.86%) in frequency. Traits that had less than 9 observations on left and right side combined in any of the subsamples (see further discussion on the sides recording of the traits) were also excluded. Only 17 traits that were used in the analysis are described in detail and their recording presented here. These are presented in Table 10. As can be seen only two traits of the postcranial skeleton are included in the final analyses: the septal aper- 5th step - The following variables were excluded because of the high inter or intra observer error: MFLL/MFLR, MFLN/MFLR, MHBLR/MHBLL, calculated from the observations recorded in 1996 and those recorded in the 1998 field season on a randomly chosen subsample. 6th step - The Fisher exact test of significance was performed in order to check for possible correlation of traits with sex. Only one variable pair was found to be potentially correlated with sex: the mandibular torus (MATL/ MATR). In the published lite- trait name - common Latin code trait no. Marginal tubercle tuberculum marginale (TZ) 1 Squamomastoid suture sutura squamomastoidea (SSM) 2 Supraorbital notch incisura supraorbitalis (SN) 3 Supraorbital foramen supraorbital foramen (SF) 4 Zygomatico facial foramen foramen zygomatico-faciale (ZFF) 5 Parietal foramen foramen parietale presens (PF) 6 Coronal ossicle (COR) 7 Lambda ossicle (LAM) 8 Auditory torus torus auditivus (AUDT) 10 Mental foramen foramne mentale (MEF) 11 Maxillary torus torus maxilaris (MAX) 12 Mylohyoid bridge ponticulus mylohyoideus (MHBD) 13 Septal aperture perforatio fossae olecranii (PFO) 15 Third trochanter trochanter tertius (TT) 17 Apical bone (APIC) 23 Inca bone os inca (INCA) 24 Palatine torus torus palatinus (PAL) 26 Tab. 10. Traits used in various combinations in the final analyses. Sides pooled, ture (No. 15) for the humerus and the third trochanter (No. 17) of the femur. All other traits had to be excluded due to poor preservation of the relevant areas of the bone, especially in the Neolithic period. Therefore, inclusion of postcranial metrics seems even more complementary to the analysis of non-metric traits. 4.1.4. Description of traits and scoring procedures The following description and discussion of traits relies largely on Hauser and de Stefano (1989) and quoted literature. Marginal tubercle (No. 1) (Fig. 14): Hauser and de Stefano (1989.226-230, PL XXXII, Fig. 36) - This feature is differently known as tuberculum marginal, processus marginalis, apophysis pyramidalis, processus Somme-ringi. or tuberculum zygomaticum. It is a tubercle or a projection on the temporal border of the frontal processus of the zygomatic bone. This feature was observed as early as XIXth century and Luschka (von Lu-schka 1869 quoted in Hauser and De Stefano 1989) ascribed its formation to the insertion of the temporal fascia. Although no specific studies on the time of onset of the formation are known, it is observed in newborns. No inheritance studies have been carried out so far. The occurrence of the trait is symmetric. Peri-zonius (1979) has found it to show, on the largest European sample studied this far, a slight preponderance in males (36.7% compared to 30.0% in women). No correlation with sex was observed in the Iron Gates Gorge sample. There are not enough data on frequencies in different populations to allow comparisons. It seems to be a fairly common trait in European populations (Perizonius 1979). In order to determine the presence of a marginal tubercle, "a line is drawn from the most temporal point of the frontozygomatic suture, tangential to the deepest point of the curve on the superior temporal edge of the zygomatic bone" (Hauser and De Stefano 1989.227). This is done by using a small transparent ruler. If a part of the frontal process projects beyond the margin of the ruler, a marginal tubercle is present. The trait was coded as present or absent without distinguishing finer categories proposed by Hauser and De Stefano (1989). Squamomastoid suture (No. 2) (Fig. 15): (Hauser and De Stefano 1989.206-207, Fig. 32). Known also as: sutura squamomastoidea, sutura (fissura) mastoidea squamosa, sutura petrosquamosa, mastoid notch. The junction between the anterior part of the mastoid process, characterised by a smooth surface, and the posterior part roughened by muscle insertions, presents a suture in newborn and early childhood. If this suture, or parts of it, persist in the adult, it is recorded as a nonmetric variant. No genetic studies have been reported to date. There is not Vlasac 41 Padina 18 Vlasac 51 Vlasac 54 Vlasac 16 Fig. 14. Different expressions of the marginal tubercle. Fig. 15. Squamomastoid suture. Lepenski Vir 26. enough information on population incidence of the trait. Supraorbital notch (No. 3) and Supraorbital foramen (No. 4) (Fig. 16): (Hauser and De Stefano 1989.50-58, Fig. 10 PI. VIII). Supraorbital notches are also known as supratrochlear notch, incisura supratrochlearis, supraorbital medial notch, incisura supraorbitalis medialis, frontal notch, incisura frontalis, sulcus supraorbitalis, supraorbital foramen incomplete, supraorbital lateral notch, incisura supraorbitalis lateralis, incisura supraorbitalis. Supraorbital foramen is also known as: foramen supratrochlear, trochlear foramen, supratrochlear canal, foramen supraorbital mediate, supraorbital notch closed, medial supraorbital canal, canalis supraorbitalis, frontal foramen, foramen frontale, foramen supraorbital laterale, accessory foramen, supraorbital lateral canal, canalis supraorbitalis lateralis, canalis supraorbitalis. The supraorbital margin of the orbit is formed entirely by the frontal bone, which in this region might show either notches or foramina or both in varying positions and numbers, and of varying size. The notches may have blurred or acute margins. The foramina correspond to external orifices of canals perforating (piercing) the margin of the orbital roof. Care should be taken not to confuse these with external orifices or nutrient canals or large porosities. Simply stated they have to pierce through the bone to be recognised as such (Hauser and De Stefano 1989. 51, PI. VIII). In the study of prematurely born infants the notches and foramina were observed as early as the 25th gestation week (Hauser and Bergman 1984). There is an observable increase in canals and formation of a second notch Fig. 16. Supraorbital notch (a) and foramen (b). Lepenski Vir 8. later in development. However, age dependency ceases in adulthood and these traits remain constant throughout the adult years (Berry 1975; Perizonius 1979). The early manifestation of these traits suggests a strong genetic base. Sjovold (1984) recorded presence and absence of the trait in a number of skeletons of families of known sex, age and origin, and came to the same conclusion. The fact that both the shape and the number of notches and canals vary suggests both different growth patterns and different morphology of nerves and vessels. The number of canals can relate to bifurcation modalities i.e. a nerve can bifurcate before it enters the supraorbital margin or after and will produce a different result. No statistically significant differences were observed between males and females and there is apparently no side preference (Dodo 1987; Mouri 1976; Hauser and De Stefano 1989). There seems to be a general increase in frequency between puberty and adulthood (Berry 1975; Cesnys 1982; Hauser and De Stefano 1989; Perizonius 1979). There is a number of scoring procedures for these traits. Hauser and De Stephano (1985; 1989) distinguish supratrochlear, medial and lateral notch, as well as the supratrochlear foramen as a separate trait, noting the number of occurrences. In the present study no distinction was made between trochlear, medial and lateral supraorbital notch (lateral was not encountered). Supratrochlear and medial supraorbital notches are easily confounded and they differ more in the degree and position than in position alone, and therefore it is likely that a medial supraorbital notch with less than half of the structure occluded by spicules can be confounded with a supratrochlear notch by different observers, as can be seen from both the diagram and the photos of the traits provided by Hauser and De Stefano (1989.Fig. 10, 54, PL VIII c, 52). Following Buikstra and Ubelaker (1994), the distinction was made between notches and foramina and the coding was done in the following manner: Supraorbital notch: values: 0 - absent; 1 - present < 1/2 occluded by spicules; 2 - present > 1/2 occluded by spicules; 3 - present degree of occlusion unknown; 4 - multiple notches; 9 - unobservable. In the final analysis 0 = absent and 1, 2, 3, 4 = present. Supraorbital foramen: values: 0 - absent; 1 - present; 2 -multiple foramina; 9 - unobservable. In the final analysis 0 = absent and 1, 2 = present. Zygomaticofacial Foramen (No. 5): (Hauser and De Stefano 1989.224-6, Fig. 35). Also known as foramen zygomaticofacial, zygo-facial foramen. On the facial surface of the zygomatic bone one or more foramina are usually present. Rarely, however, this foramen is absent. Generally they appear 5-8 mm below the orbital border, but may vary significantly in position. Also, multiple foramina may occur. These foramina represent the external aperture of a canal whose internal orifice is situated in the orbit. The numbers of the former and the latter need not correspond. Sjovold (1984) reported a low heritability estimate for the absence of the foramina. Significant difference between sexes with higher incidence in females was reported by Cesnys (1982), and Corru-ccini (1974), others have noted only tendencies for higher incidence in either males or females. No sex correlation was found in this study. Incidences vary from 8.6% in Modern Japanese (Mouri 1976) to 99-1% in medieval Serbian populations (Zivanovic 1979a). Scoring differs among different authors. Berry and Berry (1967) note absence only, while Hauser and De Stefano (1989) suggest a more detailed scheme including (a) number: absence, one, two, three or more; (b) size of the largest: small = 0.3 mm wire enters, medium = 1 mm wire, large = 1.2 mm wire, excessive = 2 mm wire; (c) position: on the corpus, on the frontal process. Buikstra and Ubelaker (1994) suggest the following scheme: 0 = absent, 1 = 1 large, 2 = 1 large plus smaller, 3 = 2 large, 4 = 2 large plus smaller, 5 = 1 small, 6= multiple small. In the following study only the number of foramina was recorded in the following manner: value 0 - absent; 1 - one small; 2 - one large; 3 - two; 4 - more than two. In the final analyses 0 = absent 1, 2, 3 and 4 = present. Parietal foramen (No. 6): (Hauser and De Stefano 1989.78-82, PI. XII). Also known as foramen parietale, emisarium parietale, foramina parieta-lia permagna. One or two, rarely more foramina pierce the parietal near or in the sagittal suture in the obelion area. They vary in position, size and number. In l/3rd to l/6th of the population they are absent. Embryologically, the lateral angles of the bilateral clefts of the fontanella obelia may be perforated by vessels and so give rise to foramina when the ossification is complete (Gisel 1964). Large foramina known as foramina permagna are thought to represent a defective ossification of the parietal bones (Pamperl 1919 quoted in Hauser and De Stefano 1989.81) and are subsequently noted as a separate trait. In the present series none of the foramina permagna were recorded. The heritability estimate for absence of foramina parietalia is estimated to be high by Sjovold (1984) in an extensive pedigree study. Berry and Berry (1967) scored presence only. Following Hauser and De Stefano (1989) and Buikstra and Ubelaker (1994), both presence and the position were scored. Since both sides were examined for presence, if the position is sutural, the score 2 was given for both sides to facilitate comparisons. Size has not been recorded. An extensive literature on sex differences (see Hauser and De Stefano 1989 and quoted literature) shows that there are no significant differences in the frequencies for males and females. Slight increase is observed up to 3 years. Ossenberg (1969) and Cesnys (1985) reported some increase from childhood to adolescence, while the trait seems to be stable throughout adulthood. Scoring procedures: values: 0 - absent; 1 - present, on parietal; 2 - present, sutural; 9 - unobservable. In the final analysis 0 = absent, 1 and 2 = present. Sutural and fontanelle ossicles - Sutural bones: Surnumerary bones are present in a number of sutures of the skull. According to Sjovold (1984) the heritability of these traits is very moderate. None of the surnumerary ossicles have a known or suspected medical relevance. Although a great degree of inter-correlation was reported for these surnumerary bones (Hertzog 1968) they are considered as reliable by Buikstra and Ubelaker (1994). Lambdoid ossicle and apical bone did not show any intercorrela-tion in the present study and were accordingly retained in the final analyses. Lambdoid ossicle (No. 8) (Fig. 17) - One or more surnumerary bones can be situated within the lambdoid suture. Sex differences are not consistent as certain authors have found a significantly higher incidences in males (Ossenberg 1969; Berry 1975; Pe-rizonius 1979a; Molto 1983) while Czarnetzki (1975) noted a tendency for more frequent occurrence in females (Hauser and De Stefano 1989.93)• No sex correlation was found within the studied sample. Fig. 17. Lambdoid ossicles a, band c. Lepenski Vir 48. Apical bone (No. 23) - Also known as ossicle at lambda this surnumerary bone is located at lambda, within the posterior fontanelle. Hauser and De Ste-fano (1989.88, Fig. 15a-d) propose that not only presence or absence but also size and number be recorded. Also degree of protrusion into either parietal or occipital bone can be noted. In the present study no multiple bones were noted and only presence or absence were recorded. No sex predominance was recorded for this trait as well as no changes with age or artificial deformation of the skulls. Auditory exostosis (No. 10) (Fig. 18): (Hauser and De Stefano 1989.186-189, PL XXVIIa-c). Also known as auditory torus, torus auditivus, aural exostosis, ear exostosis, auditory exostosis, exostosis of the external auditory meatus, torus accusti-cus, and torus tympanicus - a bony growth situated within or protruding from the external auditory meatus, essentially evolving from the tympanic part or occasionally also from squamous portion. Two different types of bony hyperostosis can be distinguished: the superficial hyperostosis in the outer type of the meatus, which is strictly speaking a pathological benign tumour with genetic predisposition; and the deep hyperostosis that has no genetic predisposition but is caused by prolonged irritation in cold water (Kennedy 1986). However, the distinction between superficial and deep meatal type is very difficult, and the present study follows the recommendation included in Buikstra and Ubelaker (1994), that is essentially the same as the three degrees of expression recommended by Hauser and De Stefano (1989.187). The reports on both sex and side dependence vary and frequencies vary from 0.0% in prehistoric Siberians (Konzitsev 1972), Canadian Inuit (Dodo and Ishida 1987), and modern Caucasoid North Americans (Corruccini 1974a), to 26.2% in medieval Serbian population (Zivanovic 1979a). An important article on auditory exostosis in the Vlasac material was published by Frayer (1988) who ascribes this trait to hand netting as a method of fishing the large cat-fish reported among the fau-nal material at the site (Bokonyi 1978). Although Zoffmann (1983) has reported lack of auditory exostoses in Lepenski Vir, 3/28 individuals (10.7%) were recorded on that site. While Frayer reported 13/38 individuals (34.2%), 17/46 (36.9%) were found in the present study. This could be due to the fact that Frayer has been able to examine only whole skulls from the site and not the fragmented material found mixed with the postcrania. But the incidence found in the current study does not differ significantly from his findings. At the site of Padina for which Zivanovic (1975) has stated that the auditory exostoses "are always present and very large" the incidence is even greater where 10/19 individuals had this trait (52.6%), while at Hajducka Vodenica only 3/13 individuals had the trait (23%). Frayer's conclusion that the auditory exostoses can be related to the evidence for fishing, needs to be examined in more detail, since evidence for large fish in Lepenski Vir I is comparable to that of Vlasac, and no simple equation can be drawn between the two. However, there is a strong possibility that the majority of the occurrences of the tori are related to pathological rather than genetic condition and this concern has to be taken into account in the final analysis. Scoring was done as follows: values: 0 - absent; 1 - < 1/3 canal occluded; 2 - 1/3-2/3 canal occluded; 3 - > 2/3 canal occluded; 9 - unobservable. Mental foramen (No. 11) (Fig. 19): foramen mentale (.Hauser and De Stefano 1989.230-3, Plate XXXIIIe-h). A foramen situated on the exomandibular surface on each side of the mandible, generally in the area below the premolars and most often below the apex of the second premolar. The foramen may vary in shape and size, it may be double or multiple with varying distances between the apertures and in rare instances even absent. No genetic studies have been reported to _ date. Since the formation of Fig 1% Mental foramen. Padina 2. the mental foramina happens before birth, there might be a fair amount of gene- volve the buccal side of the molars resulting in hy-tics involved. In the present study, only the number pertrophy of the alveolar margin. In rare cases it of foramina was recorded. In case of the inner divi- can extend to PM4 or even a canine. There is a dis-sion of the foramen (doubled foramen) it was recor- agreement about its aetiology, and since it occurs ded as 2 foramina. Reported frequencies of accessory more often in skulls with palatine torus, the same foramina vary from 4.7 in modern Indians (Gerhen- function and interaction between genetics and en-son et al. 1986) to 38.8 in Modern Blacks from Bra- vironmental factors can be proposed. The published zil (Wijsman and Neves 1986). There is no consen- results on incidence by sex are inconclusive (Hauser sus on predominance acceding to sex, as it varies and de Stefano 1989.183) and need to be checked from one population to another. It occurs asymme- against each population. No sex correlation was trically more often but there is no general prefe- found in this series. There seems to be no preference rence of the side. Scoring procedures: values: 0 - ab- for side expression and either no change with age sent; 1 - 1 foramen; 2-2 foramina or 1 foramen with (De Villiers 1968) or a slight increase between young complete inner division; 3 - more __— than 2 foramina; 9 - unobservable. In the final analysis, since there were no instances of absence of foramina, the trait was treated as present only if 2 or more foramina or an inner division were present. If there was only one foramen, the trait was treated as absent. Maxillary torus (No. 12) (Fig. 20): torus maxillaris (Hauser and De Stefano 1989.180-3, Tab. XXVII d-g). Also known as maxillary hyperostosis, torus alveolaris maxillaris. Both the irregular bony nodules of varying size and a mound like thickening of the lingual margin of the alveolar process in the molar area of the maxilla is referred to as maxillary torus. These protrusions may also in- Fig. 20. Maxillary torus. Vlasac 78a. and old adults (Van den Broek 1945). The frequencies reported for different populations vary from 0.0% in recent Dutch (Perizonius 1979) and Italians from Sardinia (Cossedu etal. 1979) to 52.9% in western Australians {Milne et al. 1983). No inter-correlation with the palatine torus was found in the Iron Gates Gorge material. Scoring procedures: values: 0 - absent; 1 - trace (can palpate but not see); 2 - moderate: elevation 2mm - 5 mm; 3 - marked: elevation > 5mm; 9 - unobservable. In the final analysis, 0 = absent, and 1, 2 and 3 = present. Sufficient replicability was obtained both between observers and in intraobserver test to warrant inclusion of trace presence. Palatine torus (No. 26): toruspalatinus {Hauser and De Stefano 1989.174-180, Tab. XXVI). Also known as torus palatinus sagittalis, exostosis medio-palatina. The trait consists of paramedian, rarely median, bony protuberance of varying size, form and extent situated along the median suture of the hard palate. It may extend from the incisive foramen to the posterior border of the palatine bones. It may be short and restricted to a part of the hard palate. It is mostly situated in the middle, less commonly occupying the posterior, and very rarely in the anterior position. It varies also in the degree of expression and can be found either on both sides or only unilaterally (on either side of the median suture). Only the degree of expression was noted in the present study following Buikstra and Ubelaker {1994). Although various authors observed familial occurrence, and high concordance in monozygotic twins, others favoured functional explanation. The latter observation is based on the reduced frequencies in edentulous group and after the third decade observed by some authors. Hauser and De Stefano (1989) favour Schreiner's (1935) suggestion that a genetically determined strong osseous response to irritation leads to the formation of a palatine torus. The occurrence of palatine torus is already observed by a later foetal stage and in newborns. There is a marked age dependency in late infancy and during the first three decades of life both incidence and size continue to increase. Although Hauser and De Stefano (1989.178-9) report higher incidence in females than males in most of the series, there is too much variability to build a straightforward picture. No sex dependence was observed in the present study. Generally, the torus is rarely expressed before five years of age, there is a steady increase with age until the 3rd decade and subsequently a decrease which has been attributed to loss of teeth by Axel-sson and Hadegaard (1985). Fig. 21. Mylohyoid bridging. Lepenski Vir 47. There is a disagreement on the correlation of the palatine, maxillary and mandibular tori, and these features have to be compared within the series itself. As noted for the maxillary torus, no intercorrelation between the two traits was observed. Scoring procedures: values: 0 - absent; 1 - trace (can palpate but not see); 2 - moderate: elevation 2mm - 5 mm; 3 -marked: elevation > 5mm; 4 - excessive covers most of the palate; 9 - unobservable. As in the case of maxillary torus, only 0 was recorded as absent, 1, 2, 3 and 4 were recorded as present. Mylohyoid bridge (No. 13) (Fig. 21): ponticidus mylohyoideus (Hauser and De Stefano 1989.234-237, PL XXXII). Also known as canalis mylohyoideus, arcus mylohyoideus, mylohyoid bridging. The mylohyoid groove descends downward and anteriorly from the mandibular foramen endomandibula-rily. This groove can be covered by an osseous roof of varying length, and is thus transformed into a canal. The formation of this canal can begin at the upper or central part of the groove or more rarely both. The two can exist with an intermediate uncovered part. Although there have been no studies on the heritability of the trait, the pattern of regional and group variability suggests strong genetic basis. It is usually scored according to its location and degree. There are no conclusive results on the influence of sex and side symmetry and no correlations with either were found in the present study. According to Ossenberg (1969) it rarely achieves expression before adolescence and shows rapid increase into adulthood, but remains relatively stable in adult years. Frequencies between population vary from 5.8% in modern Japanese (.Mouri 1976) to 33-7% in Aleuts (Dodo and Ishida 1987). Scoring procedures: values: 0 - absent; 1 - partial; 2 - complete; 9 - unob-servable. Septal aperture (No. 15) (Fig. 22): Perforatio fossae ollecranii (After Sounders 1978). The trait consists of any number of smaller or larger perforations between coronoid and olecranon fossae at the distal end of the humerus. Saunders (1978.105-127) notes both side and sex correlation for this trait. Both Fin-negan (1973) and Gaherty (1970) have found important correlation with sex. However, no correlation with either side or sex was found in the present study. This is not uncommon, since studies differ in terms of results for correlations as has been shown in cranial traits. Apart from the possibility that correlations would occur randomly in the case of a great number of tests performed, and the possibility that the trait is simply spurious and lacking in biological significance (Saunders 1978.121), two other explanations are possible: (a) the trait's correlation with both side and sex differs among populations, and (b) that in order to get reliable results for trait correlation we need a greater sample size than in the current population. Although the results by Saunders are derived from much larger populations, the size of the population examined in this study was not negligible and the trait was subsequently retained. Scoring procedures: values: 0 - present; 1 -1 small perforation with "thinning"; 2 - multiple small perforations; 3 - small perforation between Fig. 22. Septal aperture. Vlasac. Coding value "5". 1-2 mm; 4 - perforation 2-5 mm; 5 - large perforation > 5 mm. In the final analysis 0 = absent; 1, 2, 3, 4 and 5 = present Third trochanter (No. 17): trochanter tertius - A rounded, conical tubercle at the superior end of the gluteal tuberosity of the femur. The third trochanter appears as a separate trochanter-like entity, reasonably easy to distinguish, even from a very large gluteal tuberosity. According to Saunders (1978.115, Tab. 5) there is no side preference for this trait, and no correlation with sex in any of the separate samples studied. The trait was scored only as present or absent in the study. There is no inter-correlation for these two postcranial traits, nor are they correlated with any of the cranial traits in the present study. 4.2. Size and robusticity analyses During a pilot study of the material in 1996, significant difference in size and robusticity between certain individuals became evident (Fig. 23). This observation is not new, as size and robusticity data have been used by Nemeskeri (1978), Zivanovic (1975) Mikic (1981a), Zoffmann (1983), and Schwi-detski and Mikic (1988) to argue for different processes. Arguments were based on the degree of gra-cilisation pertaining to the skull. This preference for the skull in previous reports was partly responsible for concentrating on postcranial remains in the present study. Another reason for this choice is that morphometric changes in skull often illustrate changes in skull shape and robusticity at the same time. Postcranial measurements are far simpler and although changes in shape (often expressed as indices) are common due to a number of possible causes, size changes are more readily visible than in the case of the skull. The list of variables was selected to provide the most information on size and, to a degree, on robusticity (as reflected in different indices). For description of the measurements refer to Buikstra and Ubelaker (1994). These variables are: for clavicle - CML (maximal length); CAD (anterior-posterior diameter at midshaft); and CSD (superior-inferior diameter at midshaft); for humerus - HML (maximal length); HEB (epicon-dylar breadth); HVD (vertical diameter of the head); HMXD (maximum diameter at midshaft); HMND (minimum diameter at midshaft); for radius - RML (maximum length); RAPD (anterior- posterior diameter at midhsaft); RMLD (medial-lateral diameter at midshaft); Fig. 23. Comparison of these two clavicles shows the striking extent of sexual dimorphism in the Iron Gates Gorge series. for ulna - UML (maximum length); UMC (minimum circumference); for femur - FML (maximal length); FBL (bicondylar length); FEB (epiconylar breadth); FMDH (maximum head diameter); FAPSD (anterior-posterior subtrochanteric diameter); FMLSD (medial-lateral subtrochanteric diameter); FAPM (anterior-posterior midshaft diameter); FMLM (me-dial-lateral midhsaft diameter); FMC (midshaft circumference); for tibia - TL (length); TPEB (maximum proximal epiphyseal breadth); TMDB (maximum distal epiphyseal breadth); TMDNF (maximum diameter at the nutrient foramen) TTDNF (transverse or medial-lateral diameter at nutrient foramen); TCNF (circumference at the nutrient foramen); for calcaneus - CCML (maximal length); CCMB (maximal breadth). While examining the output of descriptive statistics, many of the variables were found to have too few observations. Only variables with more than 60 observations (25% of the adult sample) were retained for the initial metric statistics. These are CAD (6l), CSD (60), HEB(62), HMXD (78), HMND (79), RAPD (71), RMLD (71), FMDH (60), FASPD (94), FMLSD (94), FAPM (90), FMLM (89), FMC (84), TMDNF (64), TTDNF (63). Ideally this analysis aimed at providing a different template on the basis of which to redefine our subgroups. For each of the four sites a subsample of robust and a subsample of gracile individuals would be obtained. Then, frequencies of non-metric traits would be calculated for each of the subsamples, and compared. If the robust subsamples would cluster together and the gracile together, we would have a strong case for po-pulational differences in robusticity, with the incoming population more gracile. If, on the other hand, they would cluster in a different pattern, the scenario of gracilisation as the result of changed subsistence and lifestyle would be proposed. However, this requires that a number of well-chosen measurements, providing the best separation either through PCA or discriminant analysis would be present for most of our adult specimens. Unfortunately, this was not the case. While single measurements could never be found in common on more than 90 skeletons, a combination of any two measurements (regardless of the fact that they were or were not correlated) could not be found on more than 71 individuals. When the number of measurements was increased to three, the number of comparable individuals fell to 46. Obviously, although anthropologists never expect an ideal situation, dividing 46 individuals into males and females, and then into 4 distinct sites and further into two robusticity groups, made the goal set out in the beginning impossible. However, size/robusticity analyses were still performed in order to see whether any distinct changes in size could be recognised between chronological periods described earlier. 4.2.1. Sex determination and consideration of sexual dimorphism It has already been noted that in previous studies of the material, sex determination was based on both pelvic characteristics and some of the skull features associated with greater robusticity in males than females (.Nemeskeri 1978). This can potentially create a problem, as same or intecorrelated features would be used for both sexual and populational distinction within the series. To avoid this methodological problem as well as ascertain to what degree we can determine sex based on size and robusticity data, only those individuals with pelvic remains sufficiently preserved to determine sex were assigned sex as males (m) or females (f). For all of the individuals where sex determination was based on any feature expressing secondary sexually characteristics (Workshop of European Anthropologists 1980), a question mark was added to the designation creating "m?" for males and "f?" for females. As presented in the above diagrams the two sexes clearly separate on the basis of simple size measurements presented as Probability Plots (Fig. 24 a-f). Probability plots were performed using SYS-TAT 7 PPLOT command. They present a powerful visual display of the distribution of data. The values of the variable are plotted against the corresponding percentage points of a theoretical distribution (Gna-nadeskan 1977; Wilkinson 1990.345). In this case the theoretical distribution is normal, and the data should, if normally distributed, lie on a straight line. The interesting feature of the presented plots is their extreme bimodality corresponding to sexes. These probability plots show clear separation between males and females, and even more importantly they classify the "?" cases into their respective groups. Accordingly, the cases based on robusticity were considered as accurately determined in terms of sex to warrant their inclusion in the analysis. The "t" tests comparing sexes run on SYSTAT 7, showed significant difference for each variable examined. Based on the above, the individuals assigned to m? and f? were added to their respective groups in order to get a more representative sample size. Further analyses were then based on thus assigned sexes. They are presented in the chapter 5.4. after the non-metric analyses. 5. ANALYSIS AND RESULTS This chapter provides the results of different statistical analyses performed on non-metric and metric data. The first three sections of the chapter (5.1. to 5-3) present discussion on the statistical treat- 0.999 Q (Q £ o Q 0.001 1000 1100 1200 1300 1400 1500 1600 1700 CML a. clavicle maximal lenght SEX v n a m? + m x f? o f 0.999 O "ro E o z 5 m ro Q 0.001 80 90 100 110 120 130 140 150 160 170 CSD SEX v n a m? + m X f? o f b. clavicle superior-inferior diameter '■5 0.999 jzi CO b E o z o ra ro O 0.001 T i .............- "O Jr F r .<* "O '■O 1-7 «*o 1 0.999 X LU 500 600 700 800 SEX v n a m? + m * f? o f Q E q u 0.001 1 O" 1........... '■y O" Op. 1" JsL - fl H" 'C i 1 f I 80 120 HEB 160 200 HMND 240 280 SEX v n a rn? + m X f? o f c. humerus epicondylar breadth d. humerus maximum diameter of the head Fig. 24 a-d. Probability plots of different measurements. Overlaid diagrams for sexes. Note the linearity of the plots showing normal distribution, as well as the alignment of "?" individuals with their respective sex. "n" individuals remain unclassified. Fig. 24 e-h. Probability plots of different measurements. Overlaid diagrams for sexes. Note the linearity of the plots showing normal distribution, as well as the alignment of "?" individuals with their respective sex. "n" individuals remain unclassified. 0.999 q "to £ o z CI.001 1 o raV O" § ■ aO "JT+ >-Y .ro 50 100 150 RAPD 200 SEX v n a m? + rn * f? o f g. radius anterior-posterior midshaft diameter 0.999 Q 75 e D oi- O • o» o- h . i*" 'jifl? I»-> r "t- 0.001 X ixi 110 20 130 140 150 160 170 180 190 200 RMLD SEX v n a m? + m x f? o f h. radius medial-lateral midshaft diameter 0.999 o E Q 0.001 200 0.999 300 FASPD 400 SEX v n a m? + m X f? o f Q t> 0.001 i i O" ......1................. V» On •"Y O' A- 1™ t' -C ,W7 .rV 1 1 f 1 1 80 120 160 200 HMND 240 280 SEX v n a m? + m X f? o f e. femur anterior-posterior subtrochanteric diameter f. femur midshaft circumference ment of the non-metric data set. In Chapter 5.4. the results of analyses of the size/robusticity data are discussed. Statistical analyses are crucial in examining large quantities of numeric data. They tend to become more complex in archaeological studies because of the problem of small sample sizes, representativeness and appropriateness of statistical methodology in studying archaeological data. Apart from samples being small and inadequate, they are often skewed with outliers and usually fraught with potential problems of archaeological context (Key and fantz 1990). Statistical analyses often rely on a number of assumptions that may or may not be valid for the ar- chaeological samples. Two assumptions are made in this thesis: (D The non-metric traits have a strong genetic basis. © The sample is representative of the populations we are trying to compare. Let us consider the first assumption. Although the influence of changing environment (occupation, habitat, nutrition) cannot be excluded for most traits, this assumption is reasonably well founded in the research on the genetics of non-metric traits. We are examining the population structure and not the genetic make-up of the individuals, and therefore even if the influence of environment on the expression of traits (due to their threshold character) can not be disregarded, the validity of the population comparisons is not reduced. The assumption of the representativeness of the sample that is examined can rarely be proved in the archaeological sample. As already discussed in Chapter 3, burial samples are often (if not always) biased. Even if we disregard the fact that excavations rarely expose the entire burial site, hoping that the excavation design has taken sufficient care to provide us with a representative picture, we have to keep in mind that buried individuals almost never reflect the living population. Since not everyone gets buried in a cemetery, and since the mode of deposition is strongly dependent on the social persona of the individual (.Masset 1993; Roksandic 2000 and quoted literature), it is unreasonable to assume that the sample studied is unbiased. Furthermore, the direction of bias can be discerned only rarely, after a thorough study of all of the social, biological and ta-phonomic aspects of the skeleton. Although we can not assume the representativeness of the sample for the purpose of studying the mortuary ritual and its social implications, there is hope that the populational biology (or the genetic makeup) of a changing population will still be represented adequately to discern it in our sample. Only under the circumstances of a completely different burial ritual for the local and the supposed incoming population, that would obliterate one or the other from our sample, the assumption of the representativeness could not be sustained. Although unlikely, this possibility had to be considered in the present study. Since burial ritual in the Mesolithic varies greately and becomes more or less canonised only in the Lepenski Vir Illb period which belongs to the Middle Neolithic (Antunovic 1990), and since inhumation is a demonstrated pattern for both of the periods, there is no reason to suspect total obliteration of any of the hypothetical groups in the current sample. 5.1. Statistical analysis of non-metric traits Berry and Berry's (1967) article was a turning point in non-metric trait analysis for a number of reasons. It asserted the value of non-metric traits in population studies, it provided a lengthy list of cranial traits that were subsequently commonly used by many osteologists and, most importantly, it drew attention to the Smith-Grewal statistic for calculation of average distances between sample populations (Saunders 1989.98). Smith's Mean Measure of Divergence (MMD) has further been investigated and developed by Sj0vold (1977) and serves, with minor modifications, as the major statistic used for examining the inter-sample distance. Stated simply the Mean Measure of Divergence (MMD) is the summed divergence between two samples, divided by the number of traits included in the analysis. For this type of analysis, proportions of the sample exhibiting a trait are given as theta (9) values symmetrical around 0, such that the incidence of 50% equals a theta of zero (Jackes et al. 1997.645). Sjo-vold has determined that the Anscombe formula is the best modification for calculation of 9, most accurately transforming the incidences of traits and stabilising the variance well, except in cases when incidences are extremely high or low (Sjovold 1977). If the sample sizes are small and incidences are accordingly low, the Freeman-Tukey transformation is judged to provide somewhat better variance stabilisation than Anscombe (Jackes et al. 1997.645). The actual formulae used in this study were taken from Jackes et al. (1997). The programming as well as the running of some of the data sets was done by Professor Mary Jackes on the QuatroPro spreadsheet program at the University of Alberta in Edmonton. Others were run by myself on MicrosoftExcel program provided by Professor Mary Jackes. 5.1.1. Formulae The following formulae were used: • Freeman Tukey transformation: appropriate for small sample sizes: theta(9) = (0.5*arcsin(l-(2*k)/(N + 1))) + + (0.5*arcsin ((l-(2*k + 1))/N + 1))), where k is the number of skeletal elements showing the trait and N is the number of elements observed (where observations were possible). • Anscombe transformation: theta (9) = arcsin[l-(2*(k + 0.375)/(N + 0.75))] • Mean measure of divergence: the summed divergence between two samples, divided by number of traits: (MMD) = l/r*I[(91-92)2 -V], where r is the number of traits analyzed and V = (l/Nj+0.5)) + (l/N2+0.5)), where N is the number of observations for each trait; • the variance of the MMD = a2 = 2/r2*I(V2) • The Z statistic (appropriate for any variable with normal distribution, mean of 0 and the variance of 1, which is the case of both theta and MMD) provides the significance for the MMD (Jackes and Gao in press)-. Z = V(2*S) - V((2*r) - 1), where S = I[(9re2)2/V] • Degree of isolation (DI) is calculated in order to confirm the Z value: DI = MMD - (g*2) An MMD value that is more than twice its standard deviation is significant at .05: thus a positive DI value is significant. All of the tables provide the statistics calculated on the basis of these formulae. The Z statistic was used further to develop the dendrograms that show the relationships between different sites. 5.2. The organisation of the presentation of the analyses Since many features in the chapter will be repeated from one analysis to the next, it is appropriate here to propose a number of explanations that will make the reading of different tables and figures easier. Each analysis is presented through three tables and two figures that respect the same order and that are numbered with the table/figure number and the analysis number in brackets. The first number is a sequential number of the table or figure while the numbers in brackets refer to the analysis number and one of the following: (1) for the data table, (2) for the computer output table, (3) for the Z matrix table. Table 22 therefore would indicate the Z matrix table of the second analysis. The same labelling is used for figures where numbers indicate: (1) dendrogram from cluster analysis, (2) multidimensional scaling plot. Data tables - All of the data tables (see for example Table 11) presented in this section of the chapter follow the same outline. The shaded area on the left represents the trait number as given on Table 10 in the previous section. In the upper shaded row, the number refers to the site number, to which the name of the site is given in the following column (e.g. 1 for Hajducka Vodenica, 2 for Lepenski Vir, 3 for Padina, and 4 for Vlasac). These numbers are important in understanding the Z matrix tables, "k" refers to the number of positive observations (trait present) while "N" refers to the total number of cases where the observation was possible (sum of trait absent and trait present). Output tables (Tabs. 12, 13, 14, 15) - The first 4 columns on the left refer to the subsamples compa- red in the first two columns (names of sites, chronological units or combinations of the two that will be explained separately for each of the analysis) and numbers given to these subsamples in the next two columns (see for example Table 16). Following statistics are represented on all of the output tables (for formulas refer to section 5.2.1.): mmdFT - mean measure of divergence; sdFT - standard deviation; standFT - standardisation value: mmd/sd. It is used when the sample sizes are different. This value is highly correlated with Z statistic. Z statistic is preferred by Jackes because it is more correlated with di, total n - an average of the number of observations possible across traits for the units compared; ZFT - provides the measure of significance for the mind: di - (mmd - 2*sd) is strongly correlated with the Z and shows correlations as significant whenever this value is positive; S - is used to calculate Z and is based on 0. The formula for this statistic is given in the section 5.2.1. formula - FT stands for Freeman Tukey and signifies that the output is based on this transformation rather than Anscombe (which was also run). These columns and values are consistently presented for each analysis. In analysing the distance between the populations, it is possible to use MMD, Z and stand. The choice here is based on the fact that Z is a way of standardising the minimal measure of divergence in case of unequal sample sizes and is more explicitly correlated with both MMD and DI, the latter being the measure of significance of the distance (Jackes and Gao in press). 'Z' matrix tables (see for example Tab. 12) - These are regular distance matrices. Upper and left shaded rows present the units of analysis either as numeric and textual (upper row) or only numeric (left row). The numbers are derived directly from the output. Figures - Two figures are provided for each analysis: a Dendrogram derived from cluster analysis, and a Multidimensional Scaling (MDS) plot. Although it can be argued that MDS plots are a more appropriate way of presenting distance relationships (Nance pers. comm.), dendrograms are retained as they are commonly provided for these types of analyses in the literature and since some of the relationships are more readily visualised through them. The MDS plots definitely outline the relationships between all components in a more appropriate way so that most of the discussions are based on them. Labelling followed the pattern for tables. Dendrograms (see for example Fig. 26) - Dissimilarity matrices were used to produce dendrograms on SYSTAT 7.0. The Linkage method is Complete as the most appropriate linkage for dissimilarity and similarity matrices. For comparisons of performance of different linkage methods see Wilmink and Uytter-schaut (1984). The Complete linkage calculates the distance from every distance in the sample and thus avoids the pooling of the cases towards either the largest or the smallest distance provided. The distance measure used with the method is Euclidean, the SYSTAT default. Dendrograms are labeled as Figures with two serial numbers in brackets. The first number refers to the number of the analysis and the second number is always 1 (for dendrogram). Multidimensional Scaling (see for example Fig. 27) - Since dendrograms can link the samples only in one direction, a spatial distance between different samples can be better appreciated by the information provided by the multidimensional scaling. To produce MDS plots, the same distance matrices as for dendrograms were used. Scaling is Monotonic, Kruskal Stress (measuring how well the curve fits all the points), and two dimensions because of the small number of points plotted were found the most appropriate. In each of the MDS plot figures, captions include the scores for the two dimensions, the Kruskal stress of final configuration - that should be less than .1 on a "good fit" (Wilkinson etal. 1996.667) -and a proportion of variance expressed, is presented. 5.2.1. The discussion of sides Since no side correlations were found for any of the traits used in the analyses (Tab. 10), several approaches were possible: Method 1 - to select only one side per individual and use the record for that side only (favoured by Saunders 1978) Method 2a - to pool sides for each individual. The incidence is calculated as the number of individuals with trait present on one or both sides/number of individuals. Proponents of the first approach argue that it is more reasonable to treat individuals rather than sides as members of the breeding unit. Further, because of the age dependency of the proportion of bilateral occurrence, the side method exaggerates the effect of age-regression in variant incidence, the side method artificially inflates sample size and introduces redundant information deriving from strong positive left-right correlation in trait presence (Malta 1983-133)- The individual approach is favoured by Buikstra (1972) and Suchey (1975). The rationale behind the 'individual' method is that, since non-metric traits are threshold characters, any expression of a trait should be treated as 'trait present' and therefore, if a trait is expressed on either of the sides, it is regarded as present. This leaves us with a number of cases in which not both of the sides are sufficiently preserved to warrant determination. These cases could be included only when present, while, when absent, they would be excluded from the analyses since we can not ascertain whether they were expressed on the other side. This would drastically reduce the number of observations (already low at some sites and periods) and would accordingly - because of the small sample sizes, substantially bias the frequencies, which in turn would make comparisons with any other material impossible. Method 2b - pooling sides by randomly selecting one or the other in case their expressions differ. Because the discussed 'individual' method was not operational as it selects against poorly preserved skeletons with trait absent, an attempt to overcome this problem was made by selecting the sides (in cases where the expression differs) randomly. In this way it was possible to retain 'individuals' as units of observation, while avoiding the problem of increasing incidence in a sample of small size due to unequal representation of sides. Method 3 - adding sides and treating the material by elements and not by individuals. In this case the incidence is calculated as number of skeletal elements (regardless of the side) with traits present/ total number of elements. This method is preferred by Ossenberg (1978 quoted in Motto 1983-136-137) who proposed the theoretical explanation as to why 'side' method should be more successful than 'individual' method. She argues that the observed correlation between the intensity of trait incidence and the proportion of bilateral expression reflects genetic factors since an individual expressing a trait bilaterally has a stronger dose of trait positive alleles than an individual with unilateral expression of the trait. Therefore, computing the frequency of a discrete trait on the basis of pooled sides quantifies the genetic potential in the population better than does the individual count. This way of recording has the benefit of expressing the underlying threshold character of the traits as it takes into account the total genetic potential for the trait expression within population. It also increases the sample sizes in many cases without violating the biological bases of the trait expression. Accordingly, sides were added in the following manner: k/N L + k/N R = k/N or (2/5 + 3/8 = 5/13) where k is the number of instances in which the trait was recorded as present, while N is the total number of possible observations. In order to demonstrate how similar these two methods are in their outcome, a series of analyses were performed using 'individual' (with sides pooled by 2b method) and 'side' method where skeletal elements are treated as discrete units. Here only one of the two pairs is presented as illustration. Since they differ very little in the significance of the results and resulting distance measures, side method was used as it allowed for increased sample size. 5.2.2. Analyses based on the individuals (method 2b) for sites (Tab. 11) In this analysis only 10 traits which had sample sizes of 5 or more were used. Although sample size of five is far from desirable, insisting on more representative sample sizes would have made comparisons with Hajducka Vodenica impossible for most of the traits. trait 1 HVod 2 LVir 3 Padina 4 Vlasac no. k N k N k N k N 4 37 24 49 26 42 17 46 4 47 8 31 9 17 19 47 9 26 16 45 Tab. 11. "k" and "N" values for sites based on individuals. 5.2.3. Analyses based on the elements (method 3) for sites (Tab. 13) As can be seen from the above figure, the multidimensional scaling produces the same spatial relationships between the four sites regardless of the me- thod of pooling the side information. The differences in the positioning of Vlasac and Lepenski Vir on the diagram (Fig. 25) result from the difference in the "Z" statistic (Tabs. 12 and 14), that has increased twice because of the greater sample size. The actual relationship between the sites has remained the same. Accordingly, only the analyses performed by element are presented in further discussions. Z(ft) 1 2 3 4 1 0 2 1.5643 0 3 -0.0926 3.2085 0 4 1.1091 1.5259 2.4937 0 Tab. 12. Z matrix based on individuals. trait 1 HVod 2 LVir 3 Padina 4 Vlasac no. k N k N k N k N 2 4 9 14 46 13 24 4 62 3 8 14 24 50 22 27 41 82 6 7 10 31 60 23 28 44 77 10 7 22 5 48 17 33 27 76 11 1 21 2 61 3 30 6 82 12 1 9 6 41 1 13 15 59 13 2 14 12 18 4 25 16 32 17 0 10 12 29 4 13 13 35 Tab. 13. "i ments. V and "N" values for sites based on ele- Z(ft) 1 2 3 4 1 0 2 3.1996 0 3 0.7411 6.1667 0 4 2.8046 3.3144 5.9511 0 Tab. 14. Resulting "Z" matrix based on elements. 5.3. Results of the analyses of non-metric traits In subsequent analyses, the Iron Gates Gorge population was divided into subsamples based on sites, chronology, and combination of sites and chronology. According to the discussion of the meaning of Mesolithic and Neolithic in the context of the region, the chronological division comprises three periods: Mesolithic, Mesolithic/Neolithic, and Neolithic. Mesolithic presumes lack of contact with farming populations, Mesolithic/Neolithic, the period when the contact, even if it did not take place, was possible, and the Neolithic, when the change in the subsistence base is evidenced on one of the sites. 2 2 5 8 26 8 16 3 5 9 14 28 13 16 6 5 6 16 31 12 16 10 3 13 3 28 10 19 11 1 15 0 35 2 19 12 1 6 3 25 1 8 13 1 9 7 11 2 14 15 1 5 5 15 3 11 17 0 8 7 18 3 8 25 1 5 9 28 2 12 cm l C o to c Q Fig. 28. Dendrogram showing relationship between the Iron Gates Gorge sites and Franzhausen I. Based on dissimilarity matrix, Euclidean distance and Complete linkage. Dimension-1 Fig. 29. Multidimensional scaling for the sites with Franzhausen I. Based on dissimilarity matrix. Dimensions (1, 2): HV(.16, -.14); LV(1.00, .39); PA (.12, -1.00); VL (.23, .69); FR (-1.51, .05); Kruskal Stress of final configuration: 0.0270. Proportion of variance: 0.9941. farming communities in the region becomes possible. This is similar to the porous frontier of Dennell (1985) or availability phase of Zvelebil (1996a). Neolithic is, primarily, characterised by greater importance of domesticates in the economic base (>5%). Evidence of adoption of cultural elements of the surrounding farmers of the Balkano-Anatolian and Balkano-Karpathian basin (ENCB and MNCB of Tasic 1998) although considered, was not taken as sufficient for determining the find as Neolithic. A significant feature of the dendrogram (Fig. 30) is clustering of Mesolithic/Neoli-thic (Contact) period with Neolithic period, which is in contradiction with the Trait Mesolithic 1 Meso/Neo 2 Neolithic 3 no. k N k N k N 1 30 33 20 45 1 8 2 10 55 16 66 8 14 3 40 60 43 77 6 14 4 9 66 14 65 2 12 5 23 42 29 43 4 9 6 42 67 48 76 8 18 7 0 62 4 70 0 18 8 16 62 23 73 5 16 10 30 50 26 85 0 15 11 6 81 4 82 1 18 12 12 46 7 51 2 13 23 8 33 3 32 0 10 Tab. 20. "k " and "N" values for tra its used in the analysis chronological units. site site Site 1 Site 2 mmd sd stand total Z di S formula 1 2 name name FT FT FT n FT FT FT 1 2 Meso MN 0.1298 0.0155 8.3884 119 4.8968 0.0988 46.9740 ft 1 3 Meso LVn 0.4667 0.0398 11.7260 69 6.4826 0.3871 63.6011 ft 2 3 MN LVn 0.1017 0.0388 2.6230 78 2.4289 0.0242 26.0986 ft Tab. 21. The output of the analysis of chronological units. z (ft) 1 Mesolithic 2 Meso/Neo 3 Neolithic 1 0 2 4.8968 0 3 6.4826 2.4289 0 Tab. 22. Matrix of Z values for chronological units. Significant relationships are outlined in bold. MESO CONT NEO- i-1-1-1-1-1-1-1 0 1 2 3 4 5 6 7 Distances Fig. 30. Dendrogram showing internal relationship of chronological units. Derivedfrom dissimilarity matrix, Euclidean distance, Complete linkage. wave of advance model for the neolithisation of this region. In order to demonstrate the spread of Neolithic farmers themselves, and not only their domesticates and/or knowledge, the result should show a slight to non-existent change in the Contact period (as some exchange of genes could be expected) and an abrupt change with the advent of Neolithic. This pattern would argue for an insurgence of people with different genetic make-up who brought about the change in economic base (as proposed by Cavalli Sforza 1996). It would also imply that the farmers are genetically different from the foragers in the region. The outcome presented in the dendrogram (Fig. 30) shows that more change (regardless of its cause) happens between the Mesolithic and Contact period than Contact and Neolithic which cluster together. If indeed the new population moves in at the time of Mesolithic/Neolithic, it does not bring about an immediate change in the economic base and can not be understood in accordance with the "wave of advance" model. All of the relationships are significant according to the "di" value (Tab. 21). It is noteworthy that the distance between Mesolithic and Contact is more than twice the distance between Contact and Neolithic. This is even more suggestively shown by the Multidimensional scaling plot (Fig. 31). If there is, indeed, an exchange of genes as well as goods at the time of the first contact, it does not destabilise the Mesolithic society and ideology, as can be seen from the continuation of architectural elements, mobiliary art and general organisation of the sites. Even more importantly, as the basis of subsistence remains hunting, gathering and fishing, this supposed exchange of genes does not bring about a fully developed farming economy. It is important to note that a certain amount of change in the genetic make-up, as evidenced in the non-metric traits, would be expected due to secular trends. However, for secular trends to be the only source of change, the distances between different periods would need to be approximately the same. The diagram in Figure 31 strongly suggests a change in the population structure at the time of Mesolithic/ Neolithic period. There is an indication that, apart from the obvious secular trend reflected in the alignment of the units, a greater amount of change happens between Mesolithic and Contact periods. The introduction of an outlier in the next analysis is aimed to clarify how important this difference was in the amount of genetic change. 5.3.2a. Subsamples based on chronology with Franzhausen I (Tabs. 23, 24 and 25) With the introduction of Franzhausen, further removed in time and space from the subsamples in the studied region, a better appreciation of distance is possible. As is evident from the Table 24, the difference between Contact and Neolithic periods ceases to be significant when an outlier is introduced (as shown by a negative di' value). Furthermore, distances between Mesolithic and Contact, and Contact and Neolithic, according to the Z statistic, become almost equal. The ensuing dendrogram shows strong clustering of Mesolithic and Contact regardless of the fact that the "di" value determines this relationship as significantly different. Part of the responsibility might lay in the larger "sd" and smaller sample CN i c o CO c CD O -2 —MESO O oCONT ONEO -2 -1 0 Dimension-1 1 Fig. 31. Multidimensional scaling plot of chronological units based on dissimilarity matrix. Dimensions: (1, 2): Meso (-1.34, .00); Cont. (.27, .00); Neo. (1.07, .00); Kruskal Stress of final configuration: 0.00. Proportion of variance: 1.00. size. There is very little actual difference between Mesolithic, Contact and Neolithic an indication of continuity, with accumulated changes through time resulting in significant difference between Mesolithic and Neolithic. The introduction of Franzhausen shows that the distances on the local scale become less obvious and that in general they follow the secular trend. However, the unidimensionality of dendrogram (Fig. 32) obscures some of the very important information, and a look at the Figure 33 explains the incompatibility of the clustering information to that provided by the "di" statistic. While Contact period is almost equidistant from the Mesolithic and Neolithic periods, the two are diffe- Trait Meso 1 M/N 2 Neo 3 Franz I 4 no. k N k N k N k N 2 10 55 16 66 8 14 8 588 5 23 42 29 43 4 9 90 451 6 42 67 48 76 8 18 211 425 8 16 62 23 73 5 16 27 190 10 30 50 26 85 0 15 130 638 11 6 81 4 82 1 18 46 530 12 12 46 7 51 2 13 8 365 13 16 39 15 42 1 5 28 446 23 8 33 3 32 0 10 16 194 Tab. 23. "k" and "N" values for traits used in the analysis of chronological units with Franzhausen I. site site mmdFT sd stand total Z di S formula 1 2 FT FT n FT FT FT 1 2 Meso MN 0.0411 0.0187 2.1905 114 2.0405 0.0036 18.9949 ft 1 3 Meso N 0.3244 0.0530 6.1157 66 5.1131 0.2183 42.6539 ft 1 4 Meso Frl 0.3561 0.0111 31.9687 478 12.7567 0.3339 142.4636 ft 2 3 MN N 0.1024 0.0522 1.9597 74 2.4522 -0.0021 21.6171 ft 2 4 MN Frl 0.2874 0.0103 27.8134 486 12.0167 0.2668 130.2460 ft 4 3 Frl N 0.3327 0.0448 7.4201 438 5.8392 0.2430 49.6241 ft Tab. 24. The output of the analysis of chronological units with Franzhausen I. rently positioned in respect to Franzhausen (Bronze Age). Namely, the Contact, Neolithic and Franzhausen are to be found on the same axis, while the Mesolithic period forms a different pattern and is situated on a different axis with Contact period. Z matrix 1 Mesolithic 2 Meso/Neo 3 Neolithic 4 Franz I 1 0.0000 2 2.0405 0.0000 3 5.1131 2.4522 0.0000 4 12.7567 12.0167 5.8392 0.0000 While secular trend is a definite factor in the change from the Mesolithic to the Neo- Hansen lithic and Bronze Age, a significant change, that cannot be explained solely by temporal trend, occurs at the time when the contact with Neolithic populations becomes possible in the Iron Gates Gorge. 5.3.3- Subsamples based on combined chronology and sites (Tabs. 26, 27 and 28) Subsamples that respected both sites and chronological determination were analysed in order to provide a more fine-grained understanding of the relationship among them. The Padina sample, when divided into Mesolithic and Mesolithic/Neolithic, made com- Matrix of Z values for chronological units with Franz-I. Significant relationships are outlined in bold. parisons almost impossible because of the small numbers of observations in almost all variables. Since Padina clusters consistently with Hajducka Vodenica, and since most of the individuals from Padina belonged to the same chronological unit (Mesolithic/ Neolithic) those were assigned to Hajducka Vodenica subsample and thus form the HVPmn (Hajducka Vodenica-Padina Mesolithic Neolithic). The remaining T MESO CONT NEO FRANZ 5 10 Distances 15 Fig. 32. Dendrogram showing the changing relationship with the introduction of Franzhausen I. Based on dissimilarity matrix, Euclidean distance and Complete linkage. CM 1 c o en C 0) E O 0 - MESO O CONT o OFRANi ONEO -1 - -2-10 1 2 Dimension-1 Fig. 33- Multidimensional scaling plots for Chronological units with Franzhausen I. Based on dissimilarity matrix. Dimensions (1, 2): Meso (-.88, .48); Cont, (-.75, -.35); Neo. (.15, -.27); Franz I (1.48, .14). Kruskal Stress of final configureation: 0.00. Proportion of variance: 1.00. individuals from Padina that belong to the Mesolithic were assigned to Lepenski Vir Mesolithic subsample (forming LVPm - Lepenski Vir Padina Mesolithic). This was done in order to strengthen the Mesolithic sample of Lepenski Vir after a careful examination of frequencies. Although the frequencies do not show substantial differences, this should be kept in mind in the analyses and interpretation. According to the distance matrix (Tab. 28) produced here, the most similar are Hajducka Vodenica/Padi-na Contact group with Lepenski Vir Neolithic. This points to a strong continuity between the two peri- Trait HVP 1 LVP 2 LV 3 LV 4 V 5 V 6 no. mn m mn n m mn k N k N k N k N k N k N 1 6 11 12 15 2 11 1 8 18 28 12 23 2 11 24 7 19 4 15 8 14 3 36 1 26 3 16 26 20 22 8 19 6 14 20 48 19 32 4 4 19 2 20 5 18 2 12 7 46 5 28 5 7 12 10 12 5 9 4 9 13 30 17 22 6 17 24 19 22 12 22 8 18 23 45 19 30 7 0 17 0 20 2 18 0 18 0 42 2 25 8 4 17 2 18 4 18 5 16 14 44 15 28 11 4 34 0 27 0 20 1 18 6 54 0 28 13 5 24 4 19 6 6 1 5 12 20 4 26 23 1 9 2 9 0 9 0 10 6 24 2 14 24 1 11 1 10 2 11 0 9 0 28 0 15 Tab. 26. "k" and "N" values for traits used in the analysis of subsamples based on site/chronology combination. ods. Along these lines is the similarity between Lepenski Vir Contact with the Lepenski Vir Neolithic. But Hajducka Vodenica/Padina group also shows little difference from the Lepenski Vir Mesolithic sub-sample. At Lepenski Vir itself, the change is pronounced at the time of Mesolithic/Neolithic transition and very restricted between the Contact and the Neolithic. It is interesting to note that Lepenski Vir Mesolithic is most different from Vlasac Mesolithic and Lepenski Vir Mesolithic/Neolithic and less, but still significantly different from Lepenski Vir Neolithic. Lepenski Vir Neolithic shows little difference from Vlasac Mesolithic and somewhat more from Vlasac Mesolithic/Neolithic. The dendrogram in Figure 34 shows that Mesolithic components at Vlasac and Lepenski Vir/ Padina are almost the most removed from each other, which is confirmed by the MDS plot in Figure 35. Vlasac Mesolithic and Mesolithic/Neolithic appear to be virtually identical along the second dimension and different along the first dimension where they are pooled by similarity to Lepenski Vir Neolithic. The general outline argues for temporal trend and continuity within the sample with greater variability in the Mesolithic/Neolithic pe- site site mmd sd stand total Z di S formi 1 2 FT FT FT n FT FT FT 1 2 HVPmn LVPm 0.0205 0.0520 0.3935 37 1.0880 -0.0836 17.3097 ft 1 3 HVPmn LVmn 0.2698 0.0584 4.6233 34 3.0129 0.1531 30.4878 ft 1 4 HVPmn LVn -0.0104 0.0630 -0.1647 32 -0.1692 -0.1364 10.7029 ft 1 5 HVPmn Vm 0.0959 0.0381 2.5196 56 2.6350 0.0198 27.6085 ft 1 6 HVPmn Vmn 0.0875 0.0437 2.0041 44 2.5068 0.0002 26.6643 ft 2 3 LVPm LVmn 0.5018 0.0586 8.5607 32 5.3937 0.3846 51.9135 ft 4 2 LVn LVPm 0.3109 0.0632 4.9183 30 4.0290 0.1845 38.9387 ft 2 5 LVPm Vm 0.2715 0.0382 7.1145 55 5.6937 0.1952 55.0150 ft 2 6 LVPm Vmn 0.1599 0.0438 3.6504 43 3.5494 0.0723 34.8213 ft 3 4 LVmn LVn 0.1905 0.0718 2.6525 27 1.2510 0.0469 18.2819 ft 3 5 LVmn Vm 0.2242 0.0459 4.8892 52 3.3118 0.1325 32.8670 ft 3 6 LVmn Vmn 0.3885 0.0505 7.6860 39 4.1027 0.2874 39.5918 ft 4 5 LVn Vm 0.1661 0.0507 3.2744 50 2.4544 0.0647 26.2831 ft 4 6 LVn Vmn 0.1712 0.0553 3.0968 37 3.0302 0.0606 30.6230 ft 6 5 Vmn Vm 0.1217 0.0300 4.0608 62 3.2961 0.0617 32.7397 ft Tab. 27. The otitput of the analysis for site/chronology combination. HVPMN VM LVPM LVMN VMN LVN 0 Distances Fig. 34. Dendrogram showing internal relationships of the site/chronology units within the Iron Gates Gorge. Based on dissimilarity matrix, Euclidean distance and complete linkage. riod. While not a definite evidence of insurgence of some new genes in the time of Contact, this pattern presents a strong argument for it. 5.3-3(i■ Subsamples based on Chronology and Sites with Franzhausen I (Tabs. 29,30 and 31) Analysing the site/chronology combination with Franzhausen aimed to clarify the extent of the importance of differences between these different sub-samples, and more specifically, the way in which these different groupings are related to each other once an outlier is introduced. According to the "di" values in Table 30, several distances are non-significant: Lepenski Vir Mesolithic/ Neolithic shows little distance from the Neolithic period at the same site. This is, in itself, a strong argument for local continuity at the site in the period of shift in the economic base. Another feature of interest is the association (continuity) between the two periods at Vlasac. Such a strong association raised doubt that the chronologi- 1 - CM l C o CO c c ® FRO Q LVMN a QVM °VMN -1 - -2-10 1 2 Dimension-1 Fig. 37. Multidimensional Scaling showing internal relationships betiveen site/chronology units with the introduction of Franzhausen. Based on dissimilarity matrix. Dimensions (1, 2): HVPM (.46, .42); LVPM (1.10, .31); LVMN. (-.25, .14); LVN(-.42, .77); VM (.15, -.77); VMN (.67, -.69); Franz I (-1.70, -.19). Kruskal Stress of final confl guration: 0.064. Proportion of variance: 0.973■ 5.3-4. Contribution of traits Theta values obtained in the Site/Chronology analysis (Ch. 5.3-3.) were submitted to the Principal Component Analysis (Tab. 33, Fig. 38). This, somewhat unorthodox approach was suggested by M. Jackes (pers. comm.) based on a published analysis by Christensen (1997). Simply stated, '9' values are treated as ordinal values and submitted to the PCA in order to ascertain which of the traits contributed the most to the observed pattern. The site chronology analysis being the one on which the interpretation is mostly based, it was deemed unnecessary to subject results of other analyses to the same procedures. Figure 38 shows the plot of Factor 1 and Factor 2 for the PCA of the '0' values obtained for the analysis 5.3.3. The output in Table 33 show that the trait 1 (marginal tubercle 0.90), trait 3 (supraorbital notch 0.92), trait 6 (parietal foramen 0.89) and trait 23 (apical bone 0.80), contribute the most to the first dimension. In real ordinal data this dimension represents the size, here it determines the traits as those with high frequencies. Trait 4 exhibits a strong negative association with the dimension one (supraorbital foramen -0.89), while trait no. 13 (mylohyoid bridge -0.59) shows negative association of a lesser extent. On the second dimension, trait 24 (Inca bone 0.89) has a strong positive association, while trait no. 11 (double mental foramen -0.69) has a strong but negative association. The two components explain 66.6 of the total variation within the sample. In Varimax rotation (Tab. 34), applied to reduce the number of variables on the "size" axis, Variables 1 and 23 show even stronger positive association (0.97 and 0.98 respectively) while variable 4 shows strong negative association. On the second component, variable 11 shows even stronger negative association, while variable 5 show the strongest positive association with this component. These two components explain 55.7% of the variation in the sample. Another interesting feature of the PCA analysis is the scatterplot of factor scores 1 and factor scores 2 for the Site/Chronology combination (Fig. 39)- More than any other diagram this scatterplot of factor scores shows a "horseshoe" pattern which is characteristic of temporal ordering. Starting with Vlasac Mesolithic in the upper left corner, through Le- penski Vir Mesolithic in Lower left corner, Vlasac Me-solithic/Neolithic in the lower centre, Lepenski Vir Mesolithic/Neolithic in lower right corner and Lepenski Vir Neolithic in Upper right corner. Except for Hajducka Vodenica and Padina Mesolithic/Neolithic that is positioned centrally, all other units follow the temporal pattern. This indeed can be regarded as strong indication of continuity. In conclusion to the chapter and as a summary of the analyses, the following interpretation is offered. In general terms, both dendrograms and multidimensional scaling with or without Franzhausen show a strong temporal trend in the data. This is clearly visible in Figure 33 that shows the relationship of the three chronological units with the Bronze Age site of Franzhausen I, and in Figure 39- A significant amount of change within the examined population may be due to non-directional microevolution that is expected for a series covering a time-span of 1500 years. However, as shown by diagrams in Figures 31, 35 and 37, and the position of Hajducka Vodenica and Padina in Figure 39, the position of different sites/chronology units can not be interpreted as demonstrating a straightforward temporal change. Figure 31 shows significantly more change occuring between Mesolithic and Contact period (due to avai- Latent Roots (Eigenvalues) 1 2 3 4 5 6 5.0401 2.9539 2.3817 1.1441 0.4803 0.0000 7 8 9 0.0000 0.0000 0.0000 Component loadings 1 2 3 4 V1 0.9061 -0.1902 0.0914 0.3483 V2 -0.0838 0.2467 -0.8662 -0.4264 V3 0.9282 0.3265 -0.0888 -0.1543 V4 -0.8026 0.4243 0.0552 0.0716 V5 0.7571 0.4489 0.4066 -0.2393 V6 0.8946 0.3982 -0.0725 0.0127 V7 -0.4109 0.6277 0.6539 -0.0315 V8 -0.3348 -0.4634 0.7356 -0.3070 V11 -0.2278 -0.6903 -0.5614 0.1728 V13 -0.5891 0.4175 0.0267 0.6585 V23 0.8072 -0.3803 0.1785 0.4010 V24 0.0366 0.8978 -0.3536 0.1874 Tab. 33- PCA output for Theta values of traits analyzed in Chapter 5-3-3- CN 0) J3 CO > c 25, showing differences between males and females in particular measurements (for codes refer to Chapter 4.2 - Size and robusticity analyses). Categorical values encountered during processing for variable 'SEX' are: (3 levels) f, m, n. 5 0 SEX a. clavicle, anterior-posterior diameter 735 cd lli I 519 SEX b. humerus, epicondylar breadth o cl < ct. 393 Q oo SEX c. radius, anterior-posterior diameter 283 f m SEX d. femur, medial-lateral subtrochanteric diameter 969.0 914.8 O O SEX e. tibia, transverse diameter at nutrient foramen 698.0 f. calcaneus, maximal length Fig. 40. Sex differences expressed through a series ofANOVA graphs based on the analyses in Table 35. parate. The first set of analysis was performed on male sample (Tab. 36). Since there were no males with measurable post-cranial remains in the Neolithic, the test included only two chronological units: Mesolithic and the Contact (Mesolithic/ Neolithic). It is obvious that none of the changes from Mesolithic to Contact period is significant even without reducing the 'p' by Bonferroni procedure. The only two variables that show a slightly more important change from Mesolithic to Contact period are TTDNF (0.0637) and FAPM (0.0748), but neither the transverse diameter at nutrient foramen for tibia nor the anterior-posterior midshaft diameter for femur show a statistically signifi cant change between the two periods. Dep Var N Sq multR sourc df df F-ratio P outliers CAD 22 0.0049 CHRO 1 20 0.0986 0.7567 59 CSD 22 0.0100 CHRO 1 20 0.2027 0.6574 HEB 28 0.0172 CHRO 1 26 0.4560 0.5055 HMXD 33 0.0228 CHRO 1 31 0.7225 0.4018 HMND 33 0.0012 CHRO 1 31 0.0376 0.8476 RAPD 27 0.0099 CHRO 1 25 0.2495 0.6218 RMLD 27 0.0426 CHRO 1 25 1.1125 0.3016 UMC 21 0.0049 CHRO 1 19 0.0936 0.7630 28 FMDH 23 0.0658 CHRO 1 21 1.4795 0.2373 FASPD 34 0.0032 CHRO 1 32 0.1022 0.7512 FMLSD 33 0.0128 CHRO 1 31 0.4027 0.5303 FAPM 36 0.0904 CHRO 1 34 3.3786 0.0748 FMLM 36 0.0318 CHRO 1 34 1.1166 0.2981 7 FMC 33 0.0008 CHRO 1 31 0.0243 0.8771 TMDNF 22 0.0820 CHRO 1 20 1.7869 0.1963 34 TTDNF 23 0.1543 CHRO 1 21 3.8309 0.0637 20 Females were subjected to the same test (Tab. 37). In the first run of the analysis all three chronologi- Tab. 36. AN OVA output for chronological units. Categorical values encountered during processing are CHRO: (2 levels): M. M/N. Males only. cal periods were kept and the tests were run with 2 degrees of freedom. The aim of the test was to show whether there was any significant change in size variables among females over 1500 year time-span. Since there were three periods in question a post-hoc Tukey test (Wilkinson et al. 1996) was run together with the ANOVA in order to ascertain, for those variables that showed significant differences, between which periods the difference appears. Several variables had p values at a level of significance without the Bonferroni procedure. In the case of examining chronological units, 19 variables with more than 20 measurable individuals were subjected to the test. When the reduced value of p' was applied (0.05/19 = 0.0026) none of them were significant. They were however examined in order to avoid the Type II error. These are RAPD (Tab. 38) (0.0121), FMLSD (Tab. 39) (0.0211), FAPM (Tab. 40) (0.0533), and CCMB (Tab. 41) (0.0237). For codes refer to the listing of variable labels in Chap- Dep Var N Sq multR sourc df df F-ratio P outliers CML 20 0.0330 CHRO 1 18 0.6143 0.4434 35 CAD 34 0.0921 CHRO 2 31 1.5729 0.2236 66 CSD 33 0.0278 CHRO 2 30 0.4296 0.6547 HEB 28 0.0939 CHRO 2 25 1.2952 0.2916 20, 43 HMXD 36 0.0938 CHRO 2 33 1.7085 0.1968 20 HMND 36 0.0866 CHRO 2 33 1.5635 0.2245 RAPD 37 0.2288 CHRO 2 34 5.0441 0.0121 RMLD 37 0.1174 CHRO 2 34 2.2612 0.1197 FMDH 27 0.1165 CHRO 2 24 1.5824 0.2263 20 FASPD 44 0.0559 CHRO 2 41 1.2135 0.3076 20 FMLSD 45 0.1679 CHRO 2 42 4.2374 0.0211 49 FAPM 40 0.1466 CHRO 2 37 3.1777 0.0533 FMLM 39 0.1268 CHRO 2 36 2.6142 0.0871 FMC 37 0.0616 CHRO 2 34 1.1163 0.3392 63 TMDNF 33 0.0319 CHRO 2 30 0.4947 0.6146 35 TTDNF 31 0.0623 CHRO 2 28 0.9308 0.4061 TCNF 28 0.0680 CHRO 2 25 0.9123 0.4146 CCML 24 0.0865 CHRO 2 21 0.9947 0.3866 CCMB 27 0.2680 CHRO 2 24 4.3929 0.0237 Tab. 37. ANOVA output for chronological units. Categorical values encountered during processing for variable CHRO'are: (3 levels): M, M/N, N. Females only. ter 4.2. While searching for significance is not necessarily an invalid method, the inference based on obtained significance is not as straightforward as it would be with hypothesis testing that involves only a limited number of variables (Moore 1991.420). Upon examining the graphs produced with these variables in Figure 41 it becomes obvious that Neolithic females exert strong influence on the results. This finding in itself would be extremely significant for the present study, but for the fact that the number of Neolithic females with measurable postcranial bones is at most 4. This restricted number cautions against potential bias, and more analyses were deemed necessary. The results of the Tukey "post-hoc" tests on ANOVA show the same. It is also important to note that two of the four significant variables (Tabs. 39, 40, Figs. 41 b, c) show differences only be- Post Hoc test of RAPD. Using model MSE of 91.224 with 34 df. Matrix of pairwise mean differences: 1 M, 2 M/N, 3 N 1 2 3 1 0.0 2 0.9848 0.0 3 -17.8485 -18.8333 0.0 Tukey HSD Multiple Comparisons. Matrix of pair- wise comparison probabilities: 1 2 3 1 1.0000 2 0.9556 1.0000 3 0.0124 0.0118 1.0000 Tab. 38. Post Hoc test of radius anterior-posterior diameter. Females only. Post Hoc test of FAPM. Using model MSE of 1122.611 with 73 df. Matrix of pairwise mean differ- ences: 1 2 3 1 0.0 2 -9.9414 0.0 3 -51.8333 -41.8919 0.0 Tukey HSD Multiple Comparisons. Matrix of pair- wise comparison probabilities: 1 2 3 1 1.0000 2 0.4180 1.0000 3 0.0320 0.1005 1.0000 Tab. 40. Post Hoc test of femur anterior-posterior diameter. Females only. Females only. tween periods that are the most removed temporally, Mesolithic and Neolithic. Only two of these measurements show significant differences between Contact and Neolithic period: RAPD, and CCMB. Given that the number of females who belong to Neolithic with these variables measurable is three and two respectively, we can not argue that the results are meaningful. Even if these variables were different between periods without any reasonable doubt, explaining these differences in any of the proposed terms is scientifically dubious. While the pattern for the four variables shows a significant decrease in size from Mesolithic pre-contact to Neolithic population, they should be considered with great care due to the very restricted number of Neolithic individuals. These trends are observable Post Hoc test of FMLSD. Using model MSE of 929.631 with 42 df. Matrix of pairwise mean diffe- rences: 1 2 3 1 0.0 2 -18.8375 0.0 3 -47.1884 -28.3509 0.0 Tukey HSD Multiple Comparisons. Matrix of pair- wise comparison probabilities: 1 2 3 1 1.0000 2 0.1265 1.0000 3 0.0405 0.3028 1.0000 Tab. 39. Post Hoc test of femur medial-lateral diameter. Females only. Post Hoc test of CCMB. Using model MSE of 705.559 with 24 df. Matrix of pairwise mean differences: 1 2 3 1 0.0 2 2.9191 0.0 3 -56.7059 -59.6250 0.0 Tukey HSD Multiple Comparisons. Matrix of pair-wise comparison probabilities: 1 2 3 1 1.0000 2 0.9645 1.0000 3 0.0229 0.0237 1.0000 Tab. 41. Post Hoc test of calcaneus maximal breadth. Females only. Females only. Fig. 41. ANOVA graphs of chronological differences. Variables showing statistical levels of significance. Females only. o Cl t cr 126 116 - 106 96 - 76 M M/N N CHRO a. radius, anterior-posterior midshaft diameter 338 310 | 282 254 226 M N M/N CHRO b. femur, anterior-posterior midshaft diameter o OD 372.0 338.5 305.0 271.5 238.0 T 466.0 432.6 m 5 O O 399.2 365.6 332 4 299.0 _L M N M/N CHRO c. femur, medial-lateral midshaft diameter M M/N N CHRO d. calcaneus, maximum breadth in females and need not follow the same pattern in males. In conclusion, a certain amount of size reduction in several variables shows a secular trend from Mesolithic to Neolithic. These changes could be due to small sample sizes and un-representativeness of the population, or they can be indicative of some degree of size reduction over time. 5.4.2. Principal Component Analyses for metric data ANOVA tests performed in search of significant differences in variables are an effective tool for appreciating the amount of difference between certain subgroups in the population. They are more appro- priate and informative when testing a tight hypothesis, and where the results are more or less consistently significant (as in the case of sex differences in the first test). They should not be used to search for a pattern of grouping or change (Moore 1991). For exploration of the pattern of grouping within the population, a more appropriate method is Principal Component Analysis (Baxter 1993; Shennan 1988). PCA is commonly used by archaeologists and anthropologists for morphological (or typological) analysis. The most obvious advantage of the PCA and other multivariate techniques is the ability to provide us with information based on the analysis of more than one variable at the time. PCA also reduces the number of dimensions in which a series of vectors (derived from correlations of variables) can be represented and thus makes observations of patterning of distances between individuals amenable to graphic representation in two or three-dimensional plots. The conceptual problem that has to be kept in mind in interpreting data derived from these analyses is that, since it is obtained by a series of mathematical operations, that can always be carried out, it need not represent any true patterning (Baxter 1993.49). Care should be taken, as with many statistical procedures, not to over-interpret the resulting diagrams. Lack of explicit patterning, especially if the grouping variable is an archaeological observation, could be considered as a strong indication of continuity across chronological periods or homogeneity between cultural groups examined. A brief explanation of the selection of variables, number of factors and resulting scatterplots is offered. Since the number of skeletons that have all of the variables measured is very low, several analyses were run in order to maximise the number of individuals contributing to the factor scores as well is represented on the scatterplots. The first plot (based on Table 42, Figure 42) recapitulates the already obvious distinction between males and females and is offered here only as an example of a plot with an obvious pattern of grouping. The first set of analysis was run with variable "sex" as grouping variable on the available upper limb bone measurements. Here, of the total computer output, only eigenvalues and variable scores are pre- 2 Latent Roots (Eigenvalues) 1 2 3 4 5 4.6516 0.7521 0.5287 0.4605 0.2675 6 7 0.2162 0.1234 Component loadings 1 2 CAD 0.7783 0.2270 CSD 0.7150 -0.5177 HEB 0.8656 0.1492 HMXD 0.7186 0.5663 HMND 0.9076 -0.2511 RAPD 0.8931 0.0176 RMLD 0.8047 -0.1620 Variance Explained by Components 1 2 4.6516 0.7521 Percent of Total Variance Explained 1 2 66.4521 10.7443 -2-10 1 2 FACTOR( 1) Fig. 42. Scatterplot offactor scores for upper limb bones. Grouping variable sex: "f'for females and "m "for males. Tab. 42. PCA output for upper limb bone measurements. Grouping variable "sex". sented, together with the amount and percent of variance explained by the 2 factors. An important feature of the scatterplot in the Figure 42 is almost perfect separation of males and females on the factor (1), and practically no separation on the factor (2) except for the male individual in the lower right corner, Vlasac 26. It is evident that the size (factor 1), plays a major role in the separation of sexes. Following this analysis, the individuals were labelled according to the chronological unit - M (Mesolithic), C (Contact) and N (Neolithic). Three sets of data were examined: a combination of variables of upper limb bones, lower limb bones and a selection of variables that had shown significant difference between Mesolithic and Neolithic in a separate ANOVA test. As opposed to the previous scatterplot (Fig. 42), the one in Figure 43 (based on analyses in Table 43) does not show any clear separation of individuals by period. The overlap is strong and argues for continuity in respect to upper limb bone size. Table 44 presents the results for femur and tibia measurements. As visible from the Figure 44, the measurements of the lower limb show even less patterning. Latent Roots (Eigenvalues) 1 2 3 4 5 4.6516 0.7521 0.5287 0.4605 0.2675 6 7 0.2162 0.1234 Component loadings 1 2 CAD 0.7783 0.2270 CSD 0.7150 -0.5177 HEB 0.8656 0.1492 HMXD 0.7186 0.5663 HMND 0.9076 -0.2511 RAPD 0.8931 0.0176 RMLD 0.8047 -0.1620 Variance Explained by Components 1 2 4.6516 0.7521 Percent of Total Variance Explained 1 2 66.4521 10.7443 Tab. 43. PCA output for upper limb bone measurements. Grouped by "chronology." PCA analysis with variables selected on the basis of significant results in ANOVA tests (Tabs. 36 and 37) on Mesolithic and Neolithic, result - unexpectedly -in a picture of patterning (Tab. 45 and Fig. 45a). In order to ascertain whether there is indeed a pattern of distribution that argues for a change between Mesolithic and Contact, the scatterplot based on these variables labelled by chronology units is presented together with the one of the same variables labelled by sex in Figure 45a and b. The superposition of these two scatterplots clearly shows that the pattern of separation along both axes results from small sample size and unequal representation of males and females that have the selected variables measured in the two chronological units. The only apparent patterning occurs among males, who seem to scatter far more than females. Although the total number of individuals presented by any of these graphs does not warrant overly sophisticated conclusions, this pattern cannot be neglected. The two outliers present among males in Figure 45 - Vlasac 26 and Vlasac 78 - require explanation. They were scrutinised for reconstructed or substituted measurements, since only one (preferably left) side of the individual was recorded. It was substituted with the right bone measurement only in instances where the left bone was missing. This is especially important in the upper limb where lateralisation can induce significant differences between paired bones (Buikstra and Ubelaker 1994). During the initial statistical screening I have checked for comparability of left and right bones and since no significant differences were observed in any of the tests, decided that the substitution of left with right bone measurements was acceptable. Both individuals that are definite outliers in the graph have significant number of measurements substituted. Other individuals on the graph, that form a much more homogenous picture, also have substituted measurements, but they differ in respect to the type of the bone. Among outliers, Vlasac 78 has radius and femur substituted, while in Vlasac 26 all three bones that are analysed are substituted. Among those that scatter more consistently Vlasac 50a and Vlasac 17 have all femoral measurement substituted while others have all left bones present. The two outliers were, consequently, removed from further consideration. However, the males still showed less homogeneity than females. Unfortunately, this argument cannot be furthered and explored in more detail, as the number of individuals amenable to this analysis is too restricted. The results of both ANOVA and PCA analysis suggest heterogeneity of the population in both Mesolithic Latent Roots (Eigenvalues) 1 2 3 4 5 6 6.5071 0.5354 0.3590 0.1980 0.1429 0.1138 7 8 0.1127 0.0311 Component loadings 1 2 FMDH 0.8433 0.4106 FASPD 0.8697 -0.2063 FMLSD 0.8703 -0.3814 FAPM 0.9733 0.0919 FMLM 0.9273 -0.1372 FMC 0.9373 -0.1645 TMDNF 0.9222 0.0619 TTDNF 0.8636 0.3473 Variance Explained by Components 1 2 6.5071 0.5354 Percent of Total Variance Explained 1 2 81.3382 6.6924 Tab. 44. PCA outputfor lower limb bone measurements. Grouped by "chronology." -1 0 1 FACTO R( 1) Fig. 43■ Scatterplot offactor scores for upper limb bones. Grouping variable chronology: "M"for Mesolithic, "C"for Mesolithic/Neolithic contact period and "N"for Neolithic period. and Mesolithic/Neolithic Contact period. It is difficult to draw any conclusions about the Neolithic population considering that only females could be examined, however, on the basis of the analyses presented, there is no reason to suppose a different pattern for the Neolithic. Further, no clear distinction be- Latent Roots (Eigenvalues) 1 2 3 4 5 5.5315 0.6272 0.3505 0.2028 0.1394 6 7 0.0937 0.0548 Component loadings 1 2 HMXD 0.8351 0.3499 RAPD 0.9252 -0.0891 RMLD 0.7289 -0.6480 FMDH 0.9283 -0.1310 FASPD 0.8948 0.0969 FMLSD 0.9415 0.1766 FMLM 0.9476 0.1385 Variance Explained by Components 1 2 5.5315 0.6272 Percent of Total Variance Explained 1 2 79.0214 8.9606 Tab. 45. PCA output for a combination of significant measurements. Grouping variable "chronology. " Both males and females. 1 0 FACTOR( 1) Fig. 44. Scatterplot of factor scores for lower limb bone measurements. Grouping variable chronology: "M" for Mesolithic and "C" for Mesolithic/ Neolithic contact period. tween the three periods can be made on the basis of either single measurements or a combination of measurements. Although examining different sets of variables and rotating the PCA could produce an explicable pattern eventually, as even random numbers will take on some form of patterning in repeated trials, the presented evidence does not warrant such an exercise since the picture presented across different methods is consistent. 6. DISCUSSION In the following sections comparisons will be made between the results presented here and previous research on morphometric analysis (Ch. 6.1). The influence of archaeological interpretations on conclusions reached by anthropologists is discussed in Chapter 6.2. and 6.3. reviews observations of biological phenomena that were derived independently of archaeological interpretation. Interpretation based on several lines of inference is offered in the Chapter 7.1 and a review of planned future research is made in Chapter 7.2. 6.1. Insights from previous anthropological research As already mentioned, most of the previous anthropological research in the Iron Gates Gorge was based on comparisons of metric data for the two sites that had yielded larger numbers of measurable cranial Fig. 45. Scatterplots offactor scores for a combination of variables with grouping variables a) chronology and b) sex. M 0 ct o u_ c*1 M M mMM M M C C M -1 0 1 FACTOR( 1) a) by chronological unit Si ct o a: o 5-1 LL -2 -3 rn m m rn f f f ft rn m rn m rn rn _L -2 -1 'V26 0 V78 1 FACTOR( 1) b) by sex ffl remains, namely Vlasac and Lepenski Vir. One of the first syntheses came from Nemeskeri and his colleagues (Nemeskeri 1969; 1978; Nemeskeri and Lengyel 1978b; Nemeskeri and Szathmary 1978a; 1978b; 1978d; 1978e). Nemeskeri's research was very influential and remains one of the most comprehensive studies of the Vlasac material. Here, the conclusions of his research are presented in light of the questions investigated in the present study. In a comprehensive study of Vlasac osteological material, Nemeskeri and Szathmary (1978a. 178) conclude that, in view of the time-interval covered by the series, the variations of quantitative and qualitative traits do not indicate any significant heterogeneity. They, however, observe certain differences between chronological sub populations in which heterogeneity is more pronounced within the male and far less pronounced in the female group. In this work, the same results are indicated by the PCA scores scatterplot for a selected number of measurements in Figures 42 and 45b. These figures show greater spread of males along both axes while females tend to cluster more tightly. The measurements and indices of the facial skeleton in females show homogeneity while in males they do not. An inverse pattern, with females more variable than males, is observed only in the case of the neurocranium. Although the authors rightly caution against too strong a reliance on the analysis based on such a restricted number of individuals (1.5 to 2.7 individuals per generation, not all of which were sufficiently preserved to allow all observations and measurements), they proceed with distinguishing two major taxonomic units with further subdivisions. These taxonomic units, described as "A" and "B," follow the general chronological outline proposed by Srejovic and Letica (1978). While the first sub group was represented by an Upper Palaeolithic local population similar to Briinn-Predmosti type (A-l) with two distinct local variants, Cro-Magnon characteristics were present in sub group A-2. The B type is characterised by two phenomena, (1) mixed occurrence of type A-l and A-2 characters; and (2) a significant and gradually growing tendency for graci-lisation. Further subdivision of type B gives the B-l form, parallel to the A-2 more archaic form, and B-2 form, later in the sequence, that "further developed the peculiarities of the local form A-2 but in a gra-cile manifestations" (Nemeskeri and Szathmary 1978b. 180). In Nemeskeri's comparisons with Lepenski Vir, which was viewed by Srejovic (1972) as presenting phase displacement compared to Vlasac, the local A-l group would not be present, the A-2 group is of more definite (unmixed) Cro-Magnon type, and the gracilisation is more evident. Furthermore, according to the blood type analysis, females from Vlasac would be more closely related to both males and females from Lepenski Vir than to males from Vlasac (Nemeskeri and Lengyel 1978a.276) In conclusion, Nemeskeri and Szathmary state that the archaic A-l type might have been the "initial ethnic stratum at Vlasac" followed by a migration of another local type. The formation of the A-2 subtype (characterised as Cro-Magnon "race") in the Iron Gates population preconditions the intensification of the ethnic contact. From it a mixed type B arises, and in later phases becomes more gracile. At Lepenski Vir, which starts slightly later in the Vlasac sequence, the first local Briinn-Predmosti type would be missing and the population would be characterised by the "Cro-Magnon race and its subtypes" (Ne-meskeri and Szathmary 1978b, 182). Given recent understanding of problems associated with "racial taxonomy" (Jacobs et al. 1996), these conclusions could be either rejected or incorporated into a different understanding of population genetics and micro evolutionary changes. Given the great experience and wealth of knowledge (albeit within a different paradigmatic approach) of Nemeskeri, it would be presumptuous to disregard his findings. If his findings are red without the taxonomic labels that he has put on them, a clearer picture can be gained from his conclusions. Several characteristics of the population can be identified: • Great heterogeneity within the local Mesolithic population. This coincides with the conclusions of independent examination of sites/chronology units in 5.3.3a and is well demonstrated in Figure 37. • Greater heterogeneity within the male group and greater homogeneity within the female group. Similar conclusions can be derived from the metric analysis of the postcranial skeleton in Figures 42 and 45b. • Temporal trend of gracilisation. Although it could be indicated by some of the results in the ANOVA tests (Tab. 37 and Fig. 41), the findings of this research do not support the conclusion. Further, the small sample size in Neolithic does not permit any firm conclusion. • Trend toward homogenisation of the population in later phases that Nemeskeri attributes to greater inter-group gene flow. This could be supported by stronger clustering of Mesolithic/Neolithic components from different sites in Figure 34. In this generalised form, Nemeskeri's findings correspond, to a great extent, to the results of the current research. However, the fine-grained distinctions, that the authors made in the discussion on the basis of such restricted material evidence - even within the framework of the "anthropotypology" - are not as convincing. Nemeskeri and Szathmary's conclusions were based on analysis of Vlasac, regarded as a sub population within the Iron Gates population. Concerning the Lepenski Vir material, in a preliminary report Nemeskeri (1969) concludes that the Mesolithic strata contain the finds of Cro-Magnon type, while the Neolithic strata show most probably three distinct types belonging to the Mediterranean taxon. The first synthesis on the Iron Gates material comes from Mikic (1980; 1981a; 1981b; 1988; 1992). Following general divisions into A and B types outlined by Nemeskeri and Szathmary, Mikic developed an explanatory scheme that accounts for possible processes that could have induced the change within the series (Mikic 1981a. 104, Fig. 1). In his first synthesis of the material, strongly influenced by Srejo-vic's appreciation of indigenous Lepenski Vir cultivation and domestication, Mikic proposes that micro evolutionary trends at Lepenski Vir could account for gracilisation as a consequence of neolithisation. He has introduced another set of "types," all based on generalised "Mediterranean" morphology: 'Longheaded Mediterranean', 'Lepenska variety of Mediterranean', and 'Robust Narrow-headed Mediterranean.' All of these were derived through micro evolutionary processes from the "Cro-Magnon" type. This micro evolution occurs within the layer II of Lepenski Vir, and subsequent changes in both shape and size of the skeletons occur without interruption into the Neolithic, eventually producing 'Gracile Mediterranean' and 'Generalised Mediterranean' types. The introduction of the 'Mediterranean' label, however, does not imply the movement of Mediterraneans into the region, and he argues for local evolution from one "type" to another. Abandoning the typological classification in his later works (Mikic 1988; 1992), the author argues for local continuity and isolation stressing the morphological similarities between the Palaeolithic Climente specimen and Late Mesolithic and even Neolithic individuals from Lepenski Vir (Mikic 1992.40). Padina and Hajducka Vodenica were very summarily treated by Zivanovic and most of the conclusions were based on archaeological interpretation of the sites. Thus, on the basis of morphological examination, he singles out seven skeletons belonging to the Lepenski Vir culture on the site of Hajducka Vodenica, attributing all others to the Iron Age stratum (Zivanovic 1976c). Considering that in this research Mesolithic/Neolithic finds from Padina and Hajducka Vodenica cluster together in all of the analyses performed, and that Hajducka Vodenica does not show any deviation from other sites in the region when the material is treated as a whole, there are no grounds for this morphological separation. Although livanovic claimed that he has recognised a number of more robust skeletons belonging to the "Padina racial sub-group" and substantially more gracile ones belonging to a much later population (Zivanovic 1976c. 124), I was unable to make any such distinction once the skeletons were sexed. This Padina sub racial group' is viewed as autochthonous, different from all other known groups and best described as "Obercassel type of the Dinaric race" (Zivanovic 1975a; 1975b). Zivanovic ascribes to Nemeskeri the conclusion that Lepenski Vir belonged to the same "Proto-Dinaric" population, however, Nemeskeri disclaimed this quote (Nemeskeri and Szathmary 1978b. 180). In summary, both Hajducka Vodenica robust individuals and Padina in general are very similar. This finding is confirmed by all of the analyses performed in this research. As opposed to Ziva-novic's interpretation, apart from pronounced sexual dimorphism, no evidence of an extremely gracile group was found at Hajducka Vodenica. While Zivanovic does not discuss the series as a whole, the other authors, working within the paradigm of racial typology, agreed on one important aspect of the material, essential continuity within the region. All of the changes were attributed to mi-croevolution towards more gracile forms with intensification of contacts and admixture at the time of Vlasac II/Lepenski Vir I phases. None of the authors perceives any abrupt change in the Neolithic populations of the region. Schwidetzky and Mikic (1988) argue that gracilisation cannot be assumed to coincide with Neolithic adaptation. They reach the conclusion that the high rates of change support the mi-croevolutionary processes in the Iron Gates rather than abrupt change in population (Schwidetzky and Mikic 1988.117). It is hard to see how the observed greater degree of gracilisation in Neolithic Lepenski Vir as compared to some other anthropological series, in itself demonstrates continuity. Even more problematic is the grouping of Lepenski Vir II and III into a single unit (comprising 13 measurable skeletons) and a small number of measurable individuals attributed to Lepenski Vir I as only four measurements could be taken on all four individuals attributed to the period. For the other 18 measurements the representation is even worse: 13/28 could not be taken on any individual and 5/18 varied between 1-3 individuals. Quite a different conclusion was reached by Menk (Menk and Nemeskeri 1989). While he also claimed a sharp decrease in robusticity between the Termi- nal Mesolithic and Early Neolithic of the series, as well as considerable change in shape, the author concluded that the change cannot be explained by local evolution, but rather by a progressive replacement of the population. Menk has applied PCA to cranial and postcranial measurements provided by Nemeskeri (presented here in Figure 46). Missing values were reconstructed by estimation from iterative regression. This approach is problematic as it reconstructs, on the basis of regression, those elements that it sets to distinguish as potentially different. Although it gives more specimens for which the observations can be made, it can either accentuate or distort the observed difference. After computation of the 'z' scores for individuals, the sexes were pooled. The Lepenski Vir material is divided into Mesolithic and Neolithic samples while Vlasac was divided into five samples: Vlasac 1', 'Vlasac 1?', Vlasac 2/3' and two 'undetermined' samples. As Menk notes, the Lepenski Vir Neolithic sample "fractions into three parts with a remarkable gap in the central part of its area" (Menk and Nemeskeri 1989. 534). In itself, coupled with a small sample size and a problematic pooling of the sexes, this finding can invalidate the analysis since the central value of the Neolithic Lepenski Vir population is derived from the strong dissimilarity of the individuals of which it is made. Similarly, this phenomenon is shown even by those ANOVA plots that do show significant differences among the three periods examined in this work in Figure 41, if we observe the variance around the least square means for the Neolithic sample. In two of the four examples presented (anterior-posterior midshaft diameter and medial-lateral midshaft diameter of the femur) the spread of the values around the means overlaps with the spread of the previous period. The other two measurements (an-terior-posterior midshaft diameter of radius and maximal breadth of calcaneus) show a slight increase in size in the Contact period and accentuate the problem of small sample size. A look at the PCA graphic output for components 1, 2 and 3 that Menk and Nemeskeri offer (Fig. 46), shows strong differentiation along axis 1 (corresponding to size) for Lepenski Vir Mesolithic and Lepenski Vir Neolithic. But, the same is true for the distance between Vlasac 1' and '1?' and 'Lepenski Vir Mesolithic'. However, the distance between the two on the 2, 3 axis (measuring some form of shape differentiation) is small. It is, in effect, much smaller than between Vlasac 1' and Vlasac 1?'. Furthermore, the -10 *Mugem Starcevo-Cris Neolithic sample seems -20 to be less, removed on the size axis from Mesolithic Lepenski Vir and * Karanovo practically identical with it on the shape axis (axis 2, 3). The only actual difference between different sub- c 11.1 Q -1 i 1 1 MENEFO oMESOF -MESOM O - OMEN Bill 1 i 1 -1 0 Dimension-1 1 Fig. 48. Multidimensional scaling plots for units based on a combination of sex and chronology. Based on dissimilarity matrix. Dimension (1, 2): MESOF (1.33, 24); MESOM (-1.22, .04); MENEF (- 33, .34); MENEM (.25, -.62). Kruskal Stress of final configuration: 0.00. Proportion of variance explained = 1.00. MesoFI MesoM2 MeNeF3 MeNeM4 MesoFI 0 MesoM2 3.4185 0 MeNeF3 2.4915 1.6662 0 MeNeM4 2.1880 2.4225 1.9211 0 Tab. 48. Matrix of Z values for units based on a combination of sex and chronology. Significant relationships are outlined in bold. males from Mesolithic cluster with females from Contact and males from Contact cluster with females from Mesolithic. Mesolithic females show as slightly more different from Contact males, than Mesolithic males are from Contact females. Figure 48 shows much more clearly the greater homogeneity of females who, although they are as much removed on the dimension 1 as the male sample, appear practically indistinguishable in dimension 2. Males, on the other hand, are much more distant on the second dimension. This points towards a more homogenous picture for females in the two periods and could argue for matrilocality. The fact that Mesolithic males and females are the most removed on the first dimension, and only slightly removed on the second as well as very similar to Contact females, while Contact male sample seems to be the most removed from all of them could suggest that: • some males from a more distinct male group moving in at the time of contact and • greater homogenisation at the time of contact. 7. CONCLUSION 7.1. Combining the Lines of Interpretation In view of the proposed porous frontier between Mesolithic and Neolithic cultures in the studied region, can we presume interactions between bearers of these respective cultures? What forms did these interactions take? Was neolithisation their ultimate consequence? Interactions, understood at large, involving any amount of change within a population and resulting from the availability of contact or presence of another population, can be assumed even without any specific explanatory mechanism. It is improbable that two populations existing in a relatively restricted geographic area would never interfere or interact with one another (see Gregg 1991a for an overview of scope of proposed interactions). Beyond assumptions, the contact between groups with distinct material culture, which in the case of Balkan archaeology correspond well to subsistence groups, is evidenced on many of the sites in the region through the exchange of trade items. The question is therefore centred more on the nature and consequences of this contact than on its existence. First it is important to stress that this contact need not be uniform and could have been site specific. For example, while there is no evidence for ceramics at the Contact period in Vlasac, Hajducka Vodenica is rich in potsherds, and ceramics were found in situ in Padina houses. Is the close clustering of two periods at Vlasac indicative of greater isolation of Vlasac as a specific locality? Or, is it a consequence of poor chronological separation of a number of skeletons? On the basis of repeated analyses performed on different chronological assignment for the studied individuals, which gave the same results, the latter suggestion seems unlikely, however, it still remains a possibility and argues for more direct AMS dates for the whole series. In terms of anthropological change within the period, some regrouping of the population is evident. Vlasac seems to be very closed and little population admixture occurs at the time of availability of contact. Similarities between Padina and Hajducka Vodenica and Lepenski Vir seem to point towards greater mobility within the group as a result of possible pressure from the outside. In Radovanovic's terms (Radovanovic 1996b; 1996c; 1996d; Radovanovic and Voytek 1997), this period is a phase of consolidation of the Lepenski Vir culture, of greater ideological integration and most probably associated social differentiation. This ideological consolidation and realisation of some form of unity among the previously dispersed and distinct sites is evidenced by greater insistence on art and ritual. Accordingly, all of the sites on the right bank examined in this study prosper during this period, and the increase in number of graves seems to reflect this greater care for rituals and ancestors. Can the observed difference between Mesolithic and Mesolithic/Neolithic populations be attributed to admixture between the existing local sub populations of the Mesolithic Iron Gates Gorge, or does it provide evidence for the influx of other, more remote, genes? The observed differences between the chronological units examined seem to be largely due to a secular trend. At the time of the first contact with Neolithic population a more important change in the genetic profile of the population occurs. This indicates higher levels of admixture with a non-local population. It could have been brought about by an influx of non-local foragers, by an influx of surrounding farmers, or both. In order to answer this question with certainty, a better understanding of local Neolithic populations as well as a wider base of the Mesolithic Iron Gates populations (both from the Romanian side of the Danube as well as from the sites situated more inland) would be needed. Neither was available for this study. The Neolithic site of Velesnica contained only three female skeletons, while Ajmana (with 17 individuals) was not available for study at the time of this research. The published report by Radosavljevic-Krunic (1986) does not give enough information for the inclusion in any of the performed statistical analyses. On the Romanian side, only Schela Cladovei has yielded a significant number of individuals (62) that are still under study (Sweeney at al. 2000), while a survey beyond the banks of the Danube on both Romanian and Serbian sides is yet to be undertaken. Based on the data presented here, the distances between Mesolithic components of Vlasac and Lepenski Vir and Padina seem to be important. The great heterogeneity of the population observed by other researchers also support this finding. However, a simple trend towards homogenisation in the Contact period, would have resulted in pooling of the Mesolithic/Neolithic component in these sites somewhere towards the equidistance from the earlier components. This is not the case. As stated earlier, Vlasac seems to remain the most isolated while an important degree of similarity is observed between Lepenski Vir Mesolithic and Hajducka Vodenica and Padina Contact periods. Some, although minor, introduction of new genes is possible. Ascertaining either that they come from the surrounding Neolithic people or other people moving as a consequence of neo-lithisation of the surrounding region would be over-interpreting the scant evidence. It is notable that demographic analysis (Jackes et al. 2000) strongly suggest migration at the time of Contact where a slight over-representation of adults among the dead can be observed. The Mesolithic/ Neolithic sample could indicate a fall in fertility consequent upon a period of instability associated with cultural change and an influx of adults from outside. This would lead to an apparent over-representation of adults. Such an influx could result in a drop in fertility: the drop could be actual, as a result of the changing and unstable conditions, or it could be perceived, resulting from an unbalanced sex ratio among the migrants (an excess of males). Furthermore, although demographic analysis show increase in fertility in Neolithic sample, when combined with Con- tact period this sample argues for a stable population with total fertility approaching the foragers and not the farmers pattern. The migrants, according to morphometry and non-metric data mostly males, did not bring about the change in the economy. If hyperginy is regarded as necessarily favouring the farmers, percieved in contemporary societies as dominant (but see discussion in Ch. 2.3), these migrant males could not have been members of agricultural societies. Furthermore, if we do accept that Neolithic brings about the change in the quality of nutrition and consequently, size reduction (Cohen and Amerlagos 1984; although see Jackes et al. 1997), the lack of significant reduction in size of the bones between the two periods, would argue against the Neolithic population moving in. Even if we accept that they would have been different in size, their small number and specific mortuary patterns could account for underrepresentation of these supposed "Neolithic" individuals: the original individuals moving into the community would not necessarily be accorded the same ritual status, and the nutritional and occupational habits would account for the lack of distinction in the subsequent generations. Their genetic input would, however, be reflected in the increase of change between the two periods examined. Although both of the lines of reasoning point towards migrants as most probably the more remote Mesolithic groups moving into the ideological centre or under the pressure from the farming communities, neither direction of hyperginy, nor the size change caused by neolithisation can be regarded as unequivocal evidence. The identity of the migrants will have to be resolved by comparisons to other populations in the region. If some influx of new genes is probable in the Contact period, the Neolithic in the region, in terms of population biology, represents the continuation of the local Mesolithic. This is evident in both the non-metric traits - where Neolithic helps make the "horseshoe-shaped" curve typical of temporal ordering (Greenacre 1984), and in metrics where there is practically no significant difference between the Contact and the Neolithic. Again, a small number of measurable individuals (all of them females) in the metric analysis, and the fact that this period is present only at the Lepenski Vir site, cautions against too strict adherence to this interpretation. It is, however, the most plausible explanation based on the above data. In conclusion, large-scale population admixture can not be demonstrated from the above data. Some "seeping in" of the population suggested by Menk (Menk. and Nemeskeri 1989.531), but without the successive replacement that he argued for, can be proposed on the bases of current research. This "seeping in" happens more perceivably at the beginning of the contact, rather than at the time of change in subsistence. Once this change in subsistence does occur, it is not complete. Fishing and hunting still account for the major portion of the animal assemblage in both Neolithic sub phases at Lepenski Vir. Although the reasons for the change in subsistence are beyond the scope of this research, it can be stated on the basis of the anthropological information that it is not brought about by an incoming population. It must be regarded as a consequence of cultural and social factors operating within the Mesolithic of Lepenski Vir itself, which brought about its disintegration. Mesolithic Lepenski Vir culture is based on the rich riverine environment that tends to support the richer societies and these are not "among the first to make the transition to food production. Rather they appear to be late lasting in historic terms" (Brinch Petersen and Meiklejoh n in press). The Lepenski Vir Mesolithic successfully paralleled local Neolithic developments over a long period of time. The contacts with the Neolithic population in the region seem to have helped to form an ideological unity of sites and thus bring into full expression the artistic achievements of an already affluent society. Ideological integration evidenced at the time of possible Contact could have resulted from the growing wealth of the sites based on trade in salt preserved foodstuffs (fish from Iron Gates Gorge and wild game from Gura Ba-ciului) as proposed by Tasic (1998). Internal conflicts, overexploitation of the environment and innumerable other factors may have played a role in the disintegration of the Lepenski Vir tradition. The biological descendants of Lepenski Vir culture remained at the locality, in smaller groups, and adopted a different material culture and architecture, but retained the same burial practices, and to a great extent the hunting and fishing economic base. The greater percentage of domestic animals and definite use of domesticated varieties of cereals classifies them as a Neolithic group, but in many respects this population remained unchanged. Only within the fully developed Starcevo phase (Lepenski Vir Illb) and with the change in burial ritual towards more canonised forms (Antunovic 1990) did this population finally integrate itself into a larger Neolithic community. 7.2. Future research Many questions remain unanswered in the Iron Gates Gorge. The ritual praxis associated with burials awaits an analysis based on thorough examination of taphonomic and stratigraphic factors and their integration into understandings of Cognitive archaeology. More AMS dates, as well as an attempt to directly study genetic relationships through DNA analysis, is planned for near future, as well as the re- examination of paleodemography and paleopathology. Re-evaluation of the zooarchaeological evidence, with questions of seasonality of the sites as well as nutritional habits, is currently under way. A thorough examination of the ceramics within their site context is planned and we are hoping that more research can be done on plant remains. A survey of the foothills on both banks of Danube is a necessary step toward a more balanced picture of the subsistence, demography and meaning of these very specific sites. - ACKNOWLEDGEMENTS - For making the material availablefor the present research, I am indebted to Dr. Borislav Jovanovic and late Professor Dragoslav Srejo vic, principal in vestigators of the Iron Gates Gorge sites studied. I am indebted to Christopher Meiklejohn of the University of Winnipeg, and Jonathan Driver andJack Nance at Simon Fraser University for their guidance and support. To Mary Ja-ckes (University of Alberta, Edmonton) for specific analyses and prompt response to all my requests and queries. To Dr. Borislav Jovanovic and Dr. Ivana Radovanovic (Institute of Archaeology, Belgrade) for unfailing confidence in my research and numerous discussions, to Dr. Zivko Mikic (University of Belgrade) for logistic help and expertise during my research in Belgrade. I am grateful to my late Professor Dragosla v Srejovic, for sharing his insights and enthusiasm for Lepenski Vir culture with me. To my students and colleagues Julija Kelecevic, Gor-dana Grabez, Miki and many others in Belgrade for all the work, they have done and Tracy Rogers from SFU for discussions. To Lynda Przybyla, Shannon Wood and Ann Sullivan at SFU for tech nical help. The research was made possible by Wen-ner-Gren Foundation for Anthropological Research Grant no. 6250, Ada and Isabelle Steele Memorial Scholarship (SFU), and Graduate Travel A wards (SFU). I am indebted to Koshtana and Ivan for providing a meaning. REFERENCES AMMERMAN A. J. and CAVALLI-SFORZA L. L. 1971. Measuring the Rate and Spread of Early Farming in Europe. Man 6: 674-688. 1984. The Neolithic Transition and the Genetics of Populations in Europe. Princeton University Press, Princeton, New Jersey. ANTOUCHINE D. N. 1878. Sur la conformation du pterion chez diverses races humaines et les primates. Bull. Soc. Anthrop. 1:330-333- ANTUNOVIC M. 1990. Anthropological and Archaeological Survey Concerning Mortuary Practices in the central Area of Balkan Peninsula During Early and Middle Neolithic. In D. Cahen and M. Otte (eds.), Ru-bane et Cardial. E.R.A.U.L. 39: 39-50. ATLEY S. P. and FINDLOW F. J, (eds.) 1984. Exploring the Limits: Frontiers and Boundaries in Prehistory. International Series. vol. S223. British Archaeological Reprots, Oxford. AXELSSON G. and HADEGAARD B. 1985. Torus palatinus in Icelandic School Children. Am. J. Phys. Anth. 67: 105-112. BABOVIC L. 1998. Polozaj i funkcija svetilista na Le-penskom Viru (Location and Function of the Sanctu- aries at Lepenski Vir). In Uzdarje Dragoslavu Srejo-vicu Antidoron Dragoslavo Srejovic: 97-108. BAILEY D. W. 1997. Impermanence and Flux in the Landscape of Early Agricultural South Eastern Europe. In J. Chapman and P. Dolukhanov (eds.), Landscapes in Flux. Central and Eastern Europe in Antiquity, vol. 3: 41-58. BARKER G. 1985. Prehistoric Farming in Europe. New Studies in Archaeology. Cambridge University Press, Cambridge. 1997. Bugs and Bottlenecks: Approaches to the Transition to Farming. Antiquity 71: 480-482. BATOVIC S. 1966. Stariji Neolit u Dalmaciji. Zadar. BAXTER M. J. 1993. Exploratory Multivariate Analysis in Archaeology. Edinburgh University Press, Edinburgh. BENAC A. 1978. Les theses fondamentales sur l'ori-gine du neolithique dans les Balkans et les regions avoisinantes. Godisnjak Centra za Balkanoloska Is-pitivanja XVI: 9-29. 1979- Prelazna Zona In M. Garasanin (ed.), Prais-torija Jugoslovenskih Zemalja II, Nolit: 363- 470. 1980. Conditions geographiques du processus de neolithisation dans certaines regions de l'Europe. In J. K. Kozlowski and J. Machnik (eds.), Proble-mes de la neolithisation dans certaines regions de l'Europe: 35-48. BERRY A. C. 1975. Factors Affecting the Incidence of Non-metric Skeletal Variants. American Journal of Physical Anthropology 40: 345-358. BERRY A. C. and BERRY R. J. 1967. Epigenetic Variation in the Human Cranium. Journal of Anatomy 101:361-379. BERTELLI D. 1892. Foramin mentonieri nell'uomo ed in altri mammiferi. Monit. Zool. Ital 3: 52-55. BOCQUET-APPEL J. P. 1985. Small Populations: Demography and Paleoanthropological Inferences. fournal of Human Evolution 14: 683-691. BOLOMEY A. 1973- An Outline of the Late Epipalaeo-lithic Economy at the "Iron Gates": the Evidence of Bones. Dacia N.S. XVII: 41-52. BOKONYI S. 1969. Kicmenjaci. In D. Srejovic (ed.), Lepenski Vir. Nova praistorijska kultura u Podu-navlju: 224-229. 1972. The Vertebrate Fauna In D. Srejovic (ed.), Europe's First Monumental Sculpture: New Discoveries at Lepenski Vir: 186-190. 1974. History of Domestic Mammals in Central and Eastern Europe. Akademiai Kiado, Budapest. 1978. The Vertebrate Fauna of Vlasac. In D. Srejovic and Z. Letica (eds.), Vlasac, Mezolitsko naselje u Djerdapu: 34-65. SANU Posebna izdanja. Vol. 2. BONSALL C., LENNON R, McSWEENEY K., STEWART C., HARKNESS D., BORONEANT V., BARTOSIEWICZ L., PAYTON R. and CHAPMAN J. 1997. Mesolithic and Early Neolithic in the Iron Gates: a Palaeodietary Perspective. fournal of European Archaeology 5(1): 50-92. BORONEANf V. 1969. Decouverte d'objets d'art epi-paleolithique dans la zone des Portes de Fer du Danube. Rivista de Science Prehistoriche XXIV(2): 283-298. 1973- Recherches archeologiques sur la culture Schela Cladovei de la zone des "Portes de Fer". Dacia, N.S CVII: 5-39. 1980. Probleme ale culturi Schela Cladovei-Lepen-ski Vir in lumina noilor cercetari. Drobeta 4:27-42. 1989. Thoughts on the Chronological Relations between the Epi-Paleolithic and the Neolithic of the Low Danube. In C. Bonsall (ed.), The Mesolithic of Europe: Papers Presented at the Third International Symposium, Edinburgh 1985: 475-480. n.d. Paleolithique superieur final et Epipaleoli-thique dans la zone des Portes de Fer. BORONEANJ V., BONSALL K. C., McSWEENEY R. and PAYTON M. M. 1995. New Research on the Mesolithic of the Iron Gates Region: Excavations at Schela Cladovei, Romania Paper presented at the Epipaleo-lithique et mesolithique en Europe, Paleoenvirone-ment, peuplement etsystemes culturels, 5eme Con-gres International U.I.S.P.P., Commision mesolithique, Grenoble. BRINCH PETERSEN E. and MEIKLEJOHN C. in press. Historical Context of the Term 'Complexity' in the Scandinavian Mesolithic. In L. Janik, S. Kaner, A. Ma-tsui and P. Rowley-Conwy (eds.), From the fomon to Star Carr. Holocene Hunters and Gatheres in Temperate Eurasia. BUCZKO C. M, NEMESKERI J. and SAS L. 1978. Age Dating of Bones from Vlasac Based on their Nitrogen and Fluorine Contents. In D. Srejovic and Z. Letica (eds.), Vlasac. Mezolitsko naselje u Djerdapu: 75-76. Posebna izdanja SANU. vol. 2. BUIKSTRA J. and UBELAKER D. 1994. Standards for Data Collection from Human Skeletal Remains 44. Fayetteville. BUIKSTRA J. E. 1972. Techniques for Coping with the Age Regressive Nature of Non-metric Traits. Am. f. Phys. Anthrop. 37: 431-432. BURLEIGH R. and ZIVANOVIC S. 1980. Radiocarbon Dating of a Cro-Magnon Population from Padina, Yugoslavia with Some General Recommendations for Dating Human Skeletons. Zeitschr. fur Morphol. und Anthrop. 70(3): 272. CARCIUMARU M. 1973. Compte rendu de l'analyse pollenique des coprolithes d'Icoana - Portes de Fer. Paper presented at the Actes de du VIIle Congres International des sciences Preehistoriques et Pro-tohistoriques, Beograd. 1978. L'analyse pollenique des coprolithes de la stations archeologiques de Vlasac. In D. Srejovic and Z. Letica (eds.), Vlasac. Mezolitsko naselje u Djerdapu: 69-74. Posebna izdanja SANU. vol. 2. CAUWE N. 1998. Sepultures collectives du mesoli-thique au neolithique. In J. Guilaine (ed.), Sepultures d'Occident etgenese des megalithismes (9000-3500 avant notre ere): 11-25. CAVALLI-SFORZA L. L. 1996. The Spread of Agriculture and Nomadic Pastoralism: Insights from Genetics, Linguistics and Archaeology. In D. R. Harris (ed.), The Origins and Spread of Agriculture and Pastoralism in Eurasia. Vol. 1: 51-69. CESNYS G. 1982. Relation of Non-metric Cranial Traits to Sex and Age in the 1st-2nd millenia A.D. Lithuanian Craniological Materials. Humanbiol. Budapest 10: 19-24. 1985. Variability of Discrete Cranial Traits in the Skulls of Lithuanian Children. Anat. Anz. Jena 159: 43-54. CHAPMAN J. 1989- Demographic Trends in Neother-mal South-East Europe. In C. Bonsall (ed.), The Mesolithic in Europe: Papers Presented at the Third International Symposium, Edunburgh 1985:500-515. 1993. Social Power in the Iron Gates Gorge. In J. Chapman and P. Dolukhanov (eds.), Cultural Transformations and Interactions in Eastern Europe: 71-121. CHAPMAN J. and DOLUKHANOV P. (eds.) 1993. Cultural Transformations and Interactions in Eastern Europe. CHAPMAN R. 1981. The Emergence of Formal Disposal Areas and the "Problem" of Megalithic Tombs in Prehistoric Europe. In R. Chapman, I. Kinnes and K. Randsborg (eds.), The Archaeology of Death: 71-82. CHILDE V. G. 1936. Man Makes Himself. Watts, London. 1951. Social Evolution. Watts, London. 1958. The Prehistory of European Society 22. Penguin, Harmondsworth. CLARK J. G. D. 1978. Neothermal Orientations. In P. Mellars (ed.), The Early Post-Glacial Settlement of Northern Europe. 1980. Mesolithic Prelude. The Paleolithic-Neoli-thic Transition in Old World Prehistory. Edinburgh University press, Edinburgh. CLASON A. T. 1980. Padina and Starcevo: Game, Fish and Cattle. Palaeohistoria XXII. 141-73- COHEN M. N. 1977. The Food Crisis in Prehistory: Overpopulation and the Origins of Agriculture. Yale University Press, New Haven, Conecticut. COHEN M. N. and ARMELAGOS G.J. 1984. Paleopathology at the Origins of Agriculture: Editors' Summation In M. N. Cohen and G. J. Armelagos (eds.), Paleopathology at the Origins of Agriculture. CONSTANDSE-WESTERMANN T. S. and NEWELL R. R. 1989- Social and Biological Aspects of the Western European Mesolithic Population Structure: a Comparison with the Demography of North American Indians. In C. Bonsall (ed.), The Mesolithic in Europe: Papers Presented at the Third International Symposium, Edunburgh 1985: 106-115. CONSTANDSE-WESTERMANN T. S„ NEWELL R. R. and MEIKLEJOHN C. 1984. Human Biological Background of Population Dynamics in the Western European Mesolithic. Proceedings of the Koninklije Neder-landse Akademie van Wetenschappen Series B, 87: 139-223. CORRUCCINIR. S. 1974. An Examination of the Meaning of Cranial Discrete Traits for Human Skeletal Biological Studies. Am. J. Phys. Anthrop. 40:425-446. 1976. The Interaction Between Non-Metric and Metric Cranial Variation. Am. J. Phys. Anthrop. 44: 285-294. COSSEDU G. G, FLORIS G. and VONA G. 1979- Sex and Side Differences in the Minor Non-metrical Variants. Journal of Human Evolution 8: 685-692. CRUBEZYE. 1991. Caracteres Discrets et Evolution. Example d'une Population Nubienne: Missiminia (Soudan). These de Sciences. Universite de Bordeaux I. CULLEN T. 1995. Mesolithic Mortuary Ritual at Fran-chthi Cave, Greece. Antiquity 69: 270-289. CVIJICJ. 1987/1922. Balkanskopoluostrvo. Sabra-na dela knjiga 2. Srpska knjizevna zadruga, Beograd. CZARNETZKI A. 1972a. Epigenetische Skelettmerk-male im Populationsverleich. I. Rechts-links-Unter-schiede Bilateral Angelegter Merkmale. Z. Morph. Anthrop. 63: 238-254. 1972b. Epigenetische Skelettmerkmale im Populationsverleich. II. Frequenzunterschiede zwischen den Geslchlechtern. Z. Morph. Anthrop. 63:341-350. 1972c. Epigenetische Skelettmerkmale im Populationsverleich. III. Zur frage der Korrelation zwi-chen den Grossen des epigenetischen Abstandes und dem Grad der Allopatrie. Z. Morph. Anthrop. 64: 145-158. 1975. On the Question of Correlation Between the Size of the Epigenetic Distance and the Degree of Allopatry in Different Populations. J. Hum. Evol. 4: 483-489. CZARNETZKI A, KAUFMANN B, SCHOCH M. and XI-ROTIRIS N. 1985. Definition der anatomichen Va-rianten, Unterlagen zur Diskussion. Basel. DEMOULE J.-P. 1988. Neolithique Europeen. In A. Le-roi-Gourhan (ed.), Dictionnaire de la Prehistoire: 744-745. DE VILLIERS H. 1968. The Skull of the African Negro. Witwaters University Press, Johannesburg. DENNELL R. W. 1984. The Expansion of Exogenous-based Economies across Europe: the Balkans and Central Europe. In S. P. Atley and F. J. Findlow (eds.), Exploring the Limits: Frontiers and Boundaries in Prehistory: 93-H5. International Series, vol. S 223. 1985. The Hunter-Gatherer/Agricultural Frontier in Prehistoric Temperate Europe. In S. W. Green and S. M. Perlman (eds.), The Archaeology of Frontiers and Boundaries: 113-139. DODO Y. 1987. Supraorbital Foramen and Hypoglossal Canal Bridging: the two Most Suggestive Non-me-tric Cranial Traits in Discriminating Major Racial Groupings of Man./ Anthrop. Soc. Nippon 95: 19-35- DODO Y. and ISHIDA H. 1987. Incidences of Nonme-tric Cranial Variants in Several Population Samples from East Asia and North America./. Anthrop. Soc. Nippon 95: 161-77. DONAHUE R. E. 1992. Desperately Seeking Ceres: A Critical Examination of Current Models for the Transition to Agriculture in Mediterranean Europe. In A. B. Gebauer and T. D. Price (eds.), Transitions to Agriculture in Prehistory: 73-80. DUDAY H. 1985. Nouvelles observations sur la decomposition des corps dans un espace libre. Paper presented at the Methode d elude des sepultures, Saint-Germain en Laye. DUDAY H. and COURTAUD P. 1998. La Necropole mesolithique de La Vergne (Charante-Maritime). In J. Guilaine (ed.), Sepultures d Occident etgenese des megalithismes (9000-3500 avant notre ere): 25-39- DUMITRESCU V., BOLOMEY A.and MOGOSANU F. 1983. Esquisse d une Prehisotire de la Romanie ju-squ'd la Fine de T Age du Bronze. Editura Stiintifi-ca si Enciclopedica, Bucarest. EARLE T. 1994. Political Domination and Social Evolution In T. Ingold (ed.), Companion Encyclopaedia of Anthropology: 940-961. ENGELS F. 1972 /1884. The Origin of Family Private Property and State. In K. Marx and F. Engles (eds.), Selected Works in One Volume. International Publishers, New York. EVANS R. K. and RASSON J. A. 1984. Ex Balcanis Lux? Recent Developments in Neolithic and Chalcolithic Research in Southeast Europe. American Antiquity 49(4): 713-741. FALCONER D. S. 1965. The Inheritance of Liability to Certain Disease, Estimated from the Incidence among Relatives. Ann. Human Genet. 29: 51-76. FINNEGAN M. 1973. Discrete Non-Metric Variation of the Postcranial Skeleton in Man. American fournal of Physical Anthropology 40:135-136. FISCHER A. (ed.) 1995. Man and Sea in the Mesolithic. Coastal Settlement above and below Present Sea Level. 53- FIX A. G. 1996. Gene Frequency Clines in Europe: Demic Diffusion or Natural Selection./. Roy. Anth. Inst. (N.S.) 2: 625-643- FRAYER D. 1988. Auditory Exostoses and Evidence for Fishing at Vlasac. Current Anthropology 29(2): 346-349. 1989- Oral Pathologies in the European Upper Paleolithic and Mesolithic In I. Hershkovitz (ed.), People and Culture in Change: 255-282, Vol, 29. GAHERTY G. 1970. Skeletal Variation in Seven African Populations. Ph.D., University of Toronto. GAMBLE C. 1986. The Paleolithic Settlement of Europe. Cambridge University Press, Cambridge. GARASANIN M. 1973- Rapport generaux. Paper presented at the Actes du VHIe Congres de I'UISPP, Belgrade. 1974a. Die Grabung Anzabegovo und das Problem der Zugehorigkeit der jungsteinzeitlichen Bevolkerung des Balkans. Thracia III: 57-79. 1974b. Les rapports entre le Sud-Est Europeen et la Mediterranee orientale a l'epoque prehistori-que. Paper presented at the Ille Congres de IAIESEE, Belgrade. 1978. La neolithisation dans le sud-est et dans le centre de la peninsule Balkanique. Godisnjak Centra za Balkanoloska Ispitivanja XVI: 31-44. 1979- Centralno-balkanska zona In M. Garasanin (ed.), Praistorija fugoslovenskih Zemalja II. Neo-litsko doba, Vol. II. (V vols.). 1980. Les origines du neolithique dans le bassin de la Mediterranee et dans le sud-est Europeen. In J. K. Kozlowski and J. Machnik (eds.), Problemes de la neolithisation dans certaines regions de 1'Europe: 57-72. 1996. Razgovori 0 arheologiji. Razgovarali ipri-redili: Stasa Babic i Miodrag Tomovic. 3T, Beo-grad. GARASANIN M. and GARASANIN D. 1961. Neolitska naselba "Vrsnik" kraj selo Tarinci. Zbornik na Stip-skiot narodniet muzej II: 7-40. GEDDES D. S. 1985. Mesolithic Domestic Sheep in West Mediterranean Europe, fournal of Archaeological Science 12: 25-48. GEORGIEV G. I. and CICKOVA M. 1981. Cultures pre-historiques en Bulgarie. Izvestja Sofia: 63-188, vol. 36. GERHENSON A., NATHAN H.and LUCHANSKY E. 1986. Mental Foramen and Metnal Nerve. Acta Ana-tomica 126: 21-28. GIGOV A. 1969. Analiza polena In D. Srejovic (ed.), Lepenski Vir. Nova praistorijska kultura u Podu-navlju: 203-207. GIMBUTAS M. 1976. Neolithic Macedonia. As reflected by Excavations atAnza, Southeast Yugoslavia. Monumenta Archaeologica I. University of California Press, Los Angeles. GISEL A. 1964. Uber die Emissaria parietalia et ma-stoidea des Neugebornen. Anal. Anz.fena 114:371-373- GNANADESIKAN R. 1977. Methods for Statistical Data Analysis of Multivariate Observations. John Willey and Sons Inc., New York. GREENACRE M. J. 1984. Theory and Application of Correspondence Analysis. Academic Press, London. GREENFIELD H. 1984. Analyses of the Vertebrate Fauna from Hajducka Vodenica; an Early Neolithic site in the Iron Gates, Yugoslavia. Unpublished manuscript. GREGG S. A. 1988. Foragers and Farmers: Popida-tion Interaction and Agricultural Expansion in Neolithic Europe. University of Chicago Press, Chicago. 1991a. (ed.) Between Bands and States. 9- CAI Southern Illinois University, Carbondale. 1991b. Indirect Food Production: Mutualism and the Archaeological Visibility of Cultivation. In S. A. Gregg (ed.), Between Bands and States: 203-215, vol. 9. GRGA D. 1998. Abrazija zuba u Dejrdapskojpopu-laciji. Ph.D., University of Belgrade. GRUNEBERG H. 1952. Genetical Studies on the Skeleton of the Mouse IV. Quasi-continuous Variations. f. Genetic 51:95-114. GUILAINE J. 1976. Premiere bergers et pay sans de I'Occident Mediterranneen. Mouton, Paris-La Haye. HARDING R. M., ROSING F. W. and SOKAL R. R. 1989. Cranial Measurements do not Support Neolithization of Europe by Demic Expansion. Homo 40: 45-58. HARRIS D. R. 1984. Ethnohistorical Evidence for the Exploitation of Wild Grasses and Forbs: its Scope and Archaeological Implications. In W. van Zeist and W. A. Casparie (eds.), Plants and Ancient Man: Studies in Paleoethnobotany: 63-69. 1989- An Evolutionary Continuum of People-Plant Interaction. In D. R. Harris and G. C. Hillman (eds.), Foraging and Farming: The Evolution of Plant Exploitation: 733■ One World Archaeology, vol. 13. 1996a. Introduction: Themes and Concepts in the Study of Early Agriculture. In D. R. Harris (ed.), The Origins and Spread of Agriculture and Pastoralism in Eurasia: 1-9. 1996b. The Origin and Spread of Pastoralism in Eurasia; an Overview. In D. R. Harris (ed.), The Origins and Spread of Agriculture and Pastoralism in Eurasia: 552-573- 1996c. (ed.) The Origins and Spread of Agriculture and Pastoralism in Eurasia. UCL Press, London. HARRIS D. R. and HILLMAN G. C. (eds.) 1989. Foraging and Framing: The Evolution of Plant Exploitation. 13- Unwin Hyman, London. HAUSER G. and BERGMAN P. 1984. Some Biological and Methodological Problems of Assymetrical Development Illustrated with Reference to Sutural Bones. Anthrop. Ariz. 42: 101-116. HAUSER G. and DE STEFANO G. F. 1985. Functional and Methodological Aspects of Supraorbital Osseous Structures. Arquivo. Anat, Anthrop. 39: 23-36. 1989- Epigenetic Variants of the Human Skull. Schweizerbart, Stuttgart. HAWKES K. and O'CONNELL J. 1992. On Optimal Foraging Models and Subsistance Transitions. Current Anthropology 33(1): 63-65- HENKE W. 1989- Biologial Distances in Late Pleistocene and Early Holocene Human Populations in Europe. In I. Hershkovitz (ed.), People and Culture in Change: 541-563- vol. 508 (ii). 2 vols. HERTZOG K. P. 1968. Association between Discontinuous Cranial Traits. American Journal of Physical Anthropology 29: 397-403. HIGGS E. S. 1962. The Fauna of the Early Neolithic Site at Nea Nikomedeia (Greek Macedonia). Proceedings of the Prehistoric Society XXVIII: 271-274. HODDER I. 1982. Symbols in Action. Cambridge University Press, Cambridge. 1990. The Domestication of Europe. Social Archaeology. Basil Blackwell Ltd., Oxford. IMAMURA K. 1996. Prehistoric Japan, New Perspectives on Insular East Asia. UCL Press, London. JACKES M. and GAO Q. in press. Jiangzhai and Ban-Po (Shaanxi, PRC): New Ideas from Old Bones. In L. Janik, S. Kaner, A. Matsui and P. Rowley-Conwy (eds.), From the Jomon to Star Can: Holocene Hunters and Gatheres in Temperate Eurasia. JACKES M. and LUBELL D. in press. Human Skeletal Biology and the Mesolithic-Neolithic Transition in Portugal In P. Bintz (ed.), Epipaleolithique etMeso-lithique en Europe: peuplements, systeme culturels et paleoenvironnement. JACKES M, LUBELL D. and MEIKLEJOHN C. 1997. Healthy but Mortal: Human Biology and the First Farmers of Western Europe. Antiquity 71(273): 639-658. JACKES M., ROKSANDIC M., MEIKLEJOHN C. and LUBELL D. 2000. The Demography of the Iron Gates Samples and its Relationship to other European Mesolithic Samples. Paper presented at the Conference on Iron Gates Gorge, Edinbourgh. JACOBS K, WYMAN J.and MEIKLEJOHN C. 1996. Pitfalls in the Search for Ethnic Origins: a Cautionary Tale regarding the Construction of "Anthropological Types" in Pre-Indo-European Northeast Europe. In K. Jones-Blay and M. E. Huld (eds.), The Indo-Europea-nization of Northern Europe. Journal of Indo-European Studies Monograph No. 17. JOLLY P. 1996. Symbiotic Intraction between Black Farmers and South-Eastern San. Implications for Southern African Rock Art Studies, Etnographic Analogy, and Hunter-Gatherer Cultural Identity. Current Anthroplogy 37(2): 277-288. JOVANOVIC B. 1966. Hajducka Vodenica, praistorij-sko nalaziste. Arheoloski pregled 8: 102-3. 1968. Padina, naselje starijeg neolita i starijeg gvo-zdenog doba. Arheoloski pregled 10: 89-93- 1969. Padina, naselje starijeg neolita i starijeg gvo-zdenog doba. Arheoloski pregled 11: 129- 136. 1970. Padina - naselje mezolita, starijeg neolita i starijeg gvozdenog doba, rimska straza i ostava iz XIV veka. Arheoloski pregled 12: 78-84. 1971. Chronological Frames of the Iron Gate Group of the Early Neolithic Period. Archaeolo-gia Iugoslavica X: 23-38. 1972. The Autochthonous and the Migrational Components of the Early Neolithic in the Iron Gates. Balcanica III: 49-58. 1974. Praistorija Gornjeg Djerdapa. Starinar N.S. XXII: 1-22. 1984a. Hajducka vodenica, praistorijska nekropo-la. Starinar N.S. XXXIII-XXXIV: 305-313■ 1984b. Padina, naselje mezolita i starijeg neolita. Starinar N.S. XXXIII-XXXIV: 159-167. 1984c. Stubica, naselje starijeg neolita. Starinar N.S. XXXIII-XXXIV: 177-178. 1987. Die Architektur und Keramik der Siedlung Padina B am Eisener Tor, Jugoslawien. Germania 65(1): 1-16. KEELEY L. H. 1997. Frontier Warfare in the Early Neolithic. In D. L. Martin and D. W. Frayer (eds.), Troubled Times. Violence and Warfare in the Past: 303-320. War and Society. Vol. 3- KENNEDY G. E. 1986. The Relationship between Auditory Exostoses and Cold Water: A Latitudinal Analysis. American fonmal of physical Anthropology 71: 401-16. KENSINGER K. M. 1989- Hunting and Male Domination in Cashinahua Society. In S. Kent (ed.), Farmers as Hunters: The Implication of Sedentism: 18-26. KENT S. (ed.) 1984. Analyzing Activity Areas: An Et-noarchaeological Study of the Use of Space. 1989a. Cross-cultural Perception of Farmers as Hunters and the Value of Meat In S. Kent (ed.), Farmers as Hunters: The Implication of Sedentism: 2-17. 1989b. (ed.) Farmers as Hunters: The Implication of Sedentism . Cambridge University Press. KENT S. and VIERICH H. 1989- The Myth of Ecological Determinism - Anticipated Mobility and Site Spatial Organisation In S. Kent (ed.), Farmers as Hunters: The Implication of Sedentism: 96-130. KERCKRING T. 1670. Spicilegium anatomicum. Andreas Frisius, Amsterdam. KEY P. J. and JANTZ L. R. 1990. Statistical Measures of Intersample Variability. Hu. Evol. 5: 457-469. KONZITSEV A. G. 1972. Non-metrical Traits on Crania of 1 thousand B.C. from Miunusinsk hollow. Arch. Anat. Histl. Embryol, Leningrad 62: 53-59■ KOZLOWSKI J. K. (ed.) 1982. Origin of the Chipped Stone Industries of the Early Farming Cultures in Balkans. Zeszyt 33. KOZLOWSKI J. K. and KOZLOWSKI S. K. 1978. Le substrat mesolithique dans la partie nord de ['Europe occidentale et le processus de la Neolithization. Godisnjak, Centarza Balkanoloska Ispitivanja XVI: 93-104. 1982. Lithic Industries from Multi-layer Site Vlasac in Yugoslavia In J. K. Kozlowski (ed.), Origin of the Chipped Stone Industries of the Early Farming Cultures in Balkans: 11-109. vol. Zeszvt 33- 1986. Foragers of Central Europe and their Acculturation In M. Zvelebil (ed.), Hunters in Transition: Mesolithic Societies of Temperate Eurasia and their Transition to Agriculture: 95-108. 1987. (eds.) Chipped Stone Industries of the Early farming Cultures in Europe. KOZLOWSKI J. K. and MACHNIK J. (eds.) 1980. Pro-blemes de la neolithisation dans certaines regions de IEur ope. KRENZER U. 1996. Anthropologischer Beitrag zur mesolitisch-neolithischen Transition. HOMO 47(1): 223-247. LAZAROVICI G. 1979- Neolticul Banatului. Napoca, Cluj. LAZIC M. 1988. Fauna of Mammals from the Neolithic Settlements in Serbia In D. Srejovic (ed.), The Neolithic of Serbia, Archaeological Research 1948-1988: 22-30. LEE R. B. 1972a. The Intensification of Social Life Among the !Kung Bushmen In B. Spooner (ed.), Population Growth: Anthropological Implications: 343-350. 1972b. Population Growth and the Beginnings of Sedentary Life among the !Kung Bushmen. In B. Spooner (ed.), Population Growth: Anthropological Implications: 329-343- LEKOVIC V. 1985. The Starcevo Mortuary Practicces - New Perspectives. Godisnjak. Zavoda za Balkano-loska Ispitivanja XXIV: 157-172. LENGYEL I. 1978a. Laboratory Examination on the Vlasac Bone Finds. In D. Srejovic and Z. Letica (eds.), Vlasac. Mezolitsko naselje u Djerdapu: 261-284. Posebna izdanja SANU. vol. 2. LEROI-GOURHAN A. (ed.) 1965. La Prehistoire. LETICA Z. 1971. Vlasac, nouvel habitat de la culture Lepenski Vir a Derdap. Arcaheologia Iugoslavica X: 7-12. LEWIS-WILLIAMS D. 1996. Comment on Jolly 1996. Current Anthropology 37(2): 289-291. LICHARDUS J. and LICHARDUS-ITTEN M. 1985. La Prehistoire de I'Europe: Le Neolithique et Chalcoli-thique. Nouvelle Clio, L'histoire et ses problemes. Presses Universitaires de France, Paris. LUBBOCK J. 1865. Pre-historic Times, as Illustrated by Ancient Remains and the Manners and Customs of Modern Savages. Wiliams and Norgate, London. LYMAN R. L. 1994. Vertebrate Taphonomy. Cambridge Manuals in Archaeology. Cambridge University Press, Cambridge. MASSET C. 1993. Les dolmens: societes neolithiques et pratiques funeraires: les sepultures collectives d'Europe occidentals. Editions Errance, Paris. MEIKLEJOHN C. 1985. Review. American Anthropology. MEIKLEJOHN C. and DENSTON B. 1987. The Human Skeletal Material: Inventory and Initial Interpretation In P. A. Mellars (ed.), Excavations on Oronsay: Prehistoric Ecology on a Small Island: 290-300. MEIKLEJOHN C, WYMAN J. J, JACOBS K. and JACKES M. K. 1997. Issues in the Archaeological Demography of the Agricultural Transition in Western and Northern Europe: A View from the Mesolithic. In R. R. Paine (ed.), Integrating Archaeological Demography: Multidisciplinary Approaches to Prehistoric Population: 311-326. MEIKLEJOHN C. and ZVELEBIL M. 1991. Health Status of European Populations at the Agricultural Transition and the Implication for the Adoption of Farming. In H. Bush and M. Zvelebil (eds.), Health in Past Populations: 129-145. BAR International Series. vol. 567, Oxford. MENK R. and NEMESKERI J. 1989- The Transition from Mesolithic to Early Neolithic in Southeastern and Eastern Europe: An Anthropological Outline. Paper presented at the People and Culture in Change, Tel Aviv. MIKIC Z. 1980. Anthropologische Typen der Djerdap (Eisernen-Tor)-Serie. In J. K. Kozlowski and J. Mach-nik (eds.), Problemes de la neolithisation dans cer-taines regions de I'Europe: 151-162. 1981a. Djerdapska serija kao antropoloski model neolitizacije. Glasnik Etnografskog instituta SANU XXX: 103-130. 1981b. Stanje iproblem!fizicke antropologije u Jugoslaviji - Praistorijskiperiod. Posebna izdanja LIII9- Centar za balkanoloska ispitivanja, Sarajevo. 1988. Die Anthropologischen Funde vom Eiser-ner Tor als Neolithisationsmodell. Berytus Archaeological Studies XXXVI: 45-52. 1992. The Mesolithic Population of the Iron Gates Region. Balcanica XXIII: 4-12. MILNE N., SCHMITT L. H. and FREEDMAN L. 1983. Discrete Trait Variation in Western Australian Aboriginal Skulls .Journal of Human Evolution 12:157-168. MISIC V. 1981. Sumska vegetacija klisura i kanjo-na Istocne Srbije. Institut za bioloska istrazivanja "Sinisa Stankovic", Beograd. MILOJCIC V., BOESSNECK J. and HOPF M. 1962. Die deutchen Ausgrabungen auf der Argissa-Magula. Das praekeramische Neolithikum soivie Tier-und Pflanzenreste. BAM 2, Bonn. MOGOSANU F. 1978. Mesoliticul de la Ostrovul Cor-bulu, o nua asezare de tip Schela Cladovei. Studii si Cercetari de Istorie veche si Arheologie 29: 335-351. 1983. Paleolithique et epipaleolithique. In V. Du-mitresku, A. Bolomey and F. Mogosanu (eds.), Es-quisse d une Prehisotire de la Romanie jusqu a la Fine de /' Age du Bronze: 29-55. MOLTO J. E. 1983. Biological Relationships of Southern Ontarion Woodland Peoples: the Evidence of Discontinuous Cranial Morphology. National Museum of Man Mercury Series 117. National Museums of Canada, Ottawa. MOORE D. S. 1991. Statistics, Concepts and Controversies. 3rd edition ed. W. H. Freeman and Company, New York. MOURIT. 1976. A Study of Non-metrical Cranial Variants of the Modern Japanese in the Kinki District. J. Anthrop. Soc. Nippon 84: 191-203. NANDRIS J. 1977. The perspective of Long-Term Change in South-east Europe. In F. W. Carter (ed.), A Historical Geography of the Balkans: 25-57. NEMESKERI J. 1969- Stanovnistvo Lepenskog Vira (Prethodni izvestaj). In D. Srejovic (ed.), Lepenski Vir: 239-262. 1978. Demographic Structure of the Vlasac Epipaleolithic Population In D. Srejovic and Z. Letica (eds.), Vlasac. Mezolitsko naselje u Djerdapu: 97-134. Posebna izdanja SANU. vol. 2. NEMESKERI J. and LENGYELI. 1978b. The Results of Paleopathological Examinations. In D. Srejovic and Z. Letica (eds.), Vlasac. Mezolitsko naselje u Djerdapu: 231-260. Posebna izdanja SANU. vol. 2. NEMESKERI J. and SZATHMARY L. 1978a. Analysis of the Variation of Quantitative Traits. In D. Srejovic and Z. Letica (eds.), Vlasac. Mezolitsko naselje u Djerdapu: 157-176. Posebna izdanja SANU. vol. 2. 1978b. Anthroposcopic and Epigenetic Variation In D. Srejovic and Z. Letica (eds.), Vlasac. Mezo- litsko naselje u Djerdapu: 135-156. Posebna izdanja SANU. vol. 2. 1978c. Individual Data on the Vlasac Anthropological Series. In D. Srejovic and Z. Letica (eds.), Vlasac. Mezolitsko naselje u Djerdapu: 285-426. Posebna izdanja SANU. vol. 2. 1978d. Sex and Sexualisation. In D. Srejovic and Z. Letica (eds.), Vlasac. Mezolitsko naselje u Djerdapu: 77-96. Posebna izdanja SANU. vol. 2. 1978e. Taxonomical Structure of the Vlasac Mesolithic Population. In D. Srejovic and Z. Letica (eds.), Vlasac. Mezolitsko naselje u Djerdapu: 177-229. Posebna izdanja SANU. vol. 2. NICOLAESCU-PLOPSOR C., COMSA E. and PAUNESCU A. 1957. Santierul arheologie Baile Herculane (reg. Timisoara, r. Almas). Materiale 3: 51-58. ORLIAC M. 1988. Mesolithique. In A. Leroi-Gourhan (ed.), Dictionnaire de la Prehistoire: 686. OSSENBERG N. S. 1969. Discontinuous Morphological Variation in the Human Cranium. Ph.D., Univ. Toronto. 1976. Within and Between Race Distances in Population Studies Based on Discrete Traits of the Human Skull. Am. J. Phys. Anthrop. 45: 701-716. 1978. A Theoretical Argument for the Use of Total Side Frequencies of Bilateral Discrete Traits in Population Distance Analysis. Paper presented at the Canadian Association of Physical Anthropology, Niagara-on-the-Lake. PAMPERL R. 1919- Foramina parietalia permagna. Dt. Z. Chir. 148: 91-110. PAUNESCU A. 1978. Cercetarile archeologice de la Cuina Turcului-Dubova. Tibiscus Istorie Volume In-cchginat Celei de-a 60 Aniversari a Unirii: 11-56. 1987. Les industries lithiques du Neolithique an-cien de la Roumanie et quelques considerations sur l'inventaire lithique des cultures du Neolithique moyen de cette contree. In J. K. Kozlowski and S. K. Kozlowski (eds.), Chipped Stone Industries of the Early Farming Cultures in Europe: 74-94. PETERSEN H. C. 1997. A discriminant Analysis Approach to Morphological Regionalization in the European Late Mesolithic .Anthrop. Anz. 55(1): 33-42. PERIZONIUS W. R. K. 1979- Non-metric cranial traits: sex difference and age dependence./ Hum. Evol. 8: 679-684. POP E, BOSCAIU N. and LUPSA V. 1970. Analiza spo-ro-polinicaa sedimentelor de la Cuina Turcului. Stu-dii si cercetari de Istorie veche 21(1): 31-34. PRICE T. D. 1983. The European Mesolithic. American Antiquity 48(4): 761-778. 1996. The First Farmers of Southern Scandinavia In D. R. Harris (ed.), The Origins and Spread of Agriculture and Pastoralism in Eurasia: 346-362. Vol. 1. (1 vols.). PRICE D. T. and GEBAUER A. B. (eds.) 1995. Last Hunters-First Farmers. PRINZ B. 1987. Mesolithic Adaptation on the Lower Danube. BAR International Series 330. B.A.R., Oxford. RADOSAVLJEVIC-KRUNIC S. 1986. Resultats de l'etu-de anthroplogique des squelettes provenant du site Ajmana. Djerdapske sveske - Cahiers des Portes de Fer III: 51-58. RADOVANOVIC I. 1981. Ranoholocenska kremena industrija sa lokaliteta Padina u Djerdapu. Centar za arheoloska istrazivanja, Beograd. 1992. Mezolit Djerdapa. Unpublished Ph.D. thesis, Beogradski Univerzitet. 1994. A Review of Formal Disposal Areas in the Mesolithic of Europe. Starinar, N.S. XLIII-XLLV (1992/1993): 93-102. 1995. Mesolithic and Early Neolithic in the Iron Gates Region, Settlements, Subsistence and Chronology. Paper presented at the Epipaleolithique et mesolithique en Europe, Paleoenvironement, peuplement et systemes culturels, 5eme Congres International U.I.S.P.P., Commision mesolithique, Grenoble. 1996a. The Iron Gates Mesolithic. International Monographies in Prehistory. University of Michi-gen Press, Ann Arbor. 1996b. Kulturni identitet djerdapskog mezolita. Zbornik radova Narodnog muzeja XVI(l). 1996c. The Lepenski Vir Culture: a Contribution to Interpretation of its Ideological Aspects. In Zbornik radova posvecenih D. Srejovicu. Centar za arheoloska istrazivanja Filozofskog fakulteta u Beogradu, Beograd. 1996d. Mesolithic/Neolithic Contacts: a Case of the Iron Gates Region. Porocilo 0 raziskovanju pa-leolitika, neolitika in eneolitika v Sloveniji XXIII: 39-48. RADOVANOVIC I. and VOYTEK B. 1997. Hunters, Fishers or Farmers: Sedentism, Subsistence and Social Complexity in the Djerdap Mesolithic. Analecta praehistorica Leidensia 29: 19-31■ RENFREW C. 1996. Language Families and the Spread of Farming In D. R. Harris (ed.), The Origins and Spread of Agriculture and Pastoralism in Eurasia: 70-92. Vol, 1. (1 vols.). ROBERTS N. 1991. Late Quaternary Geomorphologi-cal Change and Origins of Agriculture in South Central Turkey. Geoarchaeology 6: 1-26. RODRIGUE C. M. 1992. Can Religion Account for Early Animal Domestications? A Critical Assessment of the Cultural Geographic Argument Based on Near Eastern Archaeological Data. Professional Geographer 44(4): 417-430. ROKSANDIC M. 2000. Interpreting ritual behaviour in mortuary context. How can forensic anthropology help? In W. D. Haglund and M. H. Sorg (eds.), Forensic taphonomy. the postmortem fate of human remains: vol. II. 2000a. Vlasac 4a. An Example of Warfare in Mesolithic Djerdap. Paper presented at 6th EAA. ROWLEY-CONWY P. 1996. Why didn't Westropp's 'Mesolithic' catch on in 1872. Antiquity 70:940-4. ROWLEY-CONWY P., ZVELEBIL M. and BLANKHOLM H. P. (eds.) 1987. Mesolithic Northwest Europe: recent Trends. SAUNDERS S. R. 1978. The Development and Distribution of Discontinuous Morphological Variation of the Human Infracranial Skeleton. National Museum of Man Mercury Series 81. National Museums of Canada, Ottawa. SAUNDERS S. S. 1989- Nonmetric Skeletal Variation In M. Y. Iscan and K. A. R. Kennedy (eds.), Reconstruction of Life from the Skeleton: 95-108. SCHREINER K. E. 1935. Zur Osteologie der Lappen 1. Harvard Univeristy Press, Oslo. SCHULTZ A. H. 1919- Der Canalis cranio-pharyngeus persistens beim Mensch und bei den Affen. Gegen-baurs Morph, Jb. Leipzig 50: 417-426. SCHWIDETZKY I. and MIKIC Z. 1988. Lepenski Vir und das Grazilisationsproblem in der Anthropologic. Godisnjak Centra za Balskanoloska ispitivanja ANUBiH 24: 113-123. SERGENT B. 1995. Les Indo-Europeens: Histoire, langues, rnythes. Bibliotheque historique Payot. Pa-yot & Rivages, Paris. SHANKS M. and TILLEY C. 1987. Social Theory and Archaeology. Cambridge University Press. SHENNAN S. 1988. Quantifying Archaeology. Edinburgh University Press, Edinburgh. SHEPHERD F. J. 1893. Symmetrical depressions on the exterior surface of the parietal bone. J. Anat. Physiol. 27:501-504. SHERRATT A. 1980. Water, Soil and Seasonality in Early Cereal Cultivation. World Archaeology 11: 313-330. SJ0VOLD T. 1977. Non-metrical Divergence Between Skeletal Populations. Ossa 4(supplement I): xii + 133- 1984. A report on the Heritability of Some Cranial Measurements and Non-Metric Traits. In G. N. van Vark and W. W. Howells (eds.), Multivariate Statistical Methods in Physical Anthropology: 223-246. SLADIC M. 1986. Kula pre Mihajlovac - un site pre-historique. Djerdapske sveske III: 432-42. SPETH J. D. 1991- Some Unexplored Aspects of Mu-tualistic Plains-Pueblo Food Exchange. In K. A. Spiel-mann (ed.), Farmers, Hunters, and Colonists. Interactions between the Southwest and the Southern Plains: 18-35. SPETH J. D. and SCOTT S. L. 1989. Horticulture and Large-mammal Hunting: the Role of Resource Depletion and the Constraints of Time and Labor. In S. Kent (ed.), Farmers as Hunters: The Implication of Sedentism: 71-80. SPONSEL L. E. 1989- Farming and Foraging: a Necessary Complementarity in Amazonia? In S. Kent (ed.), Farmers as Hunters: The Implication of Sedentism: 37-46. SPRIGGS M. 1996. Early Agriculture and What Went Before in Island Melanesia: Continuity or Intrusion? In D. R. Harris (ed.), The Origins and Spread of Agriculture and Pastoralism in Eurasia: 524-537. SREJOVIC D. 1966. Lepenski Vir, Boljetin - neolitsko naselje. Arheoloski Pregled 8: 94-96. 1968. Lepenski vir (Boljetin) - predneolitska i neolitska naselja. Arheoloski Pregled 10: 85-87. 1969- Lepenski vir. Nova praistorijska kultura u Podunavlju. SKZ, Beograd. 1971. Predneolitske i neolitske kulture u Djerda-pu. Materijali VIII kongresa arheologa fugosla-vije VI: 21-. 1972. Europe's First Monumnetal Sculpture: New Discoveries at lepenski Vir. Themes and Hudson, Aylesbury. 1978. La neolithisation de la regions des Portes de Fer. Godisnjak. Centra za Balkanoloska Ispitivanja XVI: 21-29. 1979- Protoneolit - kultura Lepenskog Vira. In Praistorija Jugoslovenskih zemalja: 49-53■ vol. 2. (5 vols.). 1989- The Mesolithic of Serbia and Montenegro. In C. Bonsall (ed.), The Mesolithic in Europe: Papers Presented at the Third International Symposium, Edunburgh 1985: 481-491. SREJOVIC D. and BABOVIC L. 1981. Lepenski Vir, Menchenbilder einer friiher europdischer Kultur. Romisch-Germanisches Museum, von Zabern, Mainz. 1983. Umetnost Lepenskog Vira, Beograd. SREJOVIC D. and LETICA Z. 1978. Vlasac mezolitsko naselje u Djerdapu. SANU Posebna izdanja 1 - Ar-heologija. 2 vols. SANU, Odeljenje istorijskih nauka, Beograd. SREJOVIC D, KOZLOWSKI J. K. and KOZLOWSKI S. K. 1980. Les industries de Vlasac et de Lepenski Vir (position taxonomique et correlations). In J. K. Kozlowski and J. Machnik (eds.), Problemes de la neolithisation dans certaines regions de lEurope: 195-205- STALIO B. 1986. Le site prehistorique Ajmana a Mala Vrbica. Djerdapske sveske - Cahiers des Portes de Fer III: 27-35. SUCHEY J. M. 1975. Biological Distance of Prehistoric Central California Populations Derived from Non-metric Traits of the Cranium, Ph.D., Univer. California, Riverside. TASIC N. 1997. Apsolutna hronologija neolita cen-tralnog Balkana. Ph.D., Belgrade University. 1998. U potrazi za solju. IQ 3: 12-38. TASIC N. N. 1989- Apsolutna hronologija neolitskih kultura Centralnog Balkana na osnovu fizicko-he-mijskih metoda. M.A., University of Belgrade. THOMAS J. 1996. The Cultural Context ot the First Use of Domesticates in Continental Central and Northwest Europe. In D. R. Harris (ed.), The Origins and Spread of Agriculture and Pastoralism in Eurasia: 310-322. vol, 1. (1 vols.). THORPE I. J. 1996. The Origins of Agriculture in Europe. Routledge, London, New York. TRINGHAM R. E. 1971. Hunters, Fishers and Farmers of Eastern Europe 6000-3000 BC. Hutchinson University Library, London. TURNBULL M. 1961. The Forest People. Simon and Schuster, New York. VAN ANDEL T. H. and RUNNELS C. N. 1995. The Earliest Farmers in Europe. Antiquity 69: 481-500. VAN DEN BROEK A. J. P. 1945. On Exostoses in the Human Skull. Acta Neerland Morph. Norm, Path. 5: 95-118. VAN GIJN A. and ZVELEBIL M. 1997. Stone Age, Ideology and Scaling the Ladder of Inference. Analecta Prehsitorica Leidensia 29: 3-12. VAN VARK G. N. and SCHAAFSMA W. 1992. Advances in the Quantitative Analysis of Skeletal Morphology. In S. R. Saunders and M. A. Katzenberg (eds.), Skeletal Biology of Past Peoples: Research Methods. VASIC R. 1986. Compte-rendu des fouilles du site prehistorique a Velesnica 1981-1982. Djerdapske Sveske, Cahiers des Portes de Fer III: 271-285- VON LUSCHKA H. I869. Der Processus marginalis des menschlichen Jochbeins. Arch. Anat. Physiol, Wiss. Med. 1869: 326-330. VOYTEK B. A. and TRINGHAM R. E. 1989. Rethinking the Mesolithic: the Case of South-East Europe. In C. Bonsall (ed.), The Mesolitic in Europe: Papers Presented at the Third International Symposium, Edinburgh 1985: 492-499. WERBART B. 1998. Subneolithic: What is it? - 'Sub-neolithic' Societies and the Conservative Economies of the Circum-Baltic Region In M. Zvelebil, R. Den-nell and L. Domanska (eds.), Harvesting the Sea, Farming the Forest. The Emergence of Neolithic Societies in the Baltic Region: 37-44. WHITTLE A. 1996. Europe in the Neolithic: the Creation of New Worlds. Cambridge World Archaeology. Cambridge University Press, Cambridge. WIJSMAN E. and NEVES W. A. 1986. The Use of Non-metric Variation in Estimating Human Population Admixture: a Test Case with Brazilian Blacks, Whites and Mulattos. American fournal of Physical Anthropology 70:395-405. WILLERDING U. 1980. Zum Ackerbau der Bandkera-miker. Materialheft Ur und Fruhgeschihte Nieder-sachens 16: 421-456. WILLIS K. J. and BENNET K. D. 1953. The Neolithic Transition - Fact or Fiction? Palaeoecological Evidence from the Balkans. The Holocene 4:326-330. WILKINSON L. 1990. SYGRAPH: The System for Graphics. SYSTAT. Evanston, IL. WILKINSON L, BLANK G. and GRUBER C. 1996. Desktop Data Analysis with SYSTAT. Prentice Hall. NJ. WILMINK F. W. and UYTTERSCHAUT H. T. 1984. Cluster Analysis, History, Theory and Applications. In G. N. van Vark and W. W. Howells (eds.), Multivariate Statistical Methods in Physical Anthropology. WILTSCHKE-SCHROTTA K. 1992. Das fruhbronzezeit-liche Graberfeld von Franzhausen I, Niederoster-reich. 3- Untersuchung der epigenetischen Merkmale. Anthrop. Anz. 50(1-2): 27-49. WINIGER J. 1998. Etnoarchaologische Studien zum Neolithokum Sudwesteuropas. BAR International Series 701. Hadrian Books, Oxford. WOODBURN J. 1968. Stability and Flexibility in Hadza Residential Groupings. In R. Lee and I. DeVore (eds.), Man the Hunter: 103-110. WORKSHOP OF EUROPEAN ANTHROPOLOGISTS. 1980. Recommendations for Age and Sex Diagnosis of Skeletons .Journal ofHuman Evolution 9:517-549■ WRIGHT G. A. 1992 /1971/. Origins of Food Production in Southwest Asia: A Survey of Ideas. Current Anthropology 33 (Supplement): 109-139. Y'EDNAK G. and FLEICH S. 1983. Microevolution and Biological Adaptability in the Transition from Food -collecting to Food-production in the Iron Gates of Yugoslavia./. Hum. Evol. 12: 279-296. ZOFFMANN Z. 1983. Prehistoric Skeletal Remains from Lepenski Vir (Iron Gates, Yugoslavia). Homo 34: 129-148. ZOHARY D. and HOPF M. 1988. Domestication of Plants in the Old World: The origin and Spread of Cultivated Plants in West Asia, Europe, and the Nile Valley. Oxford University Press, New York. ZVELEBIL M. 1996a. The Agricultural Frontier and the Transition to Farming in the Circum-Baltic Region. In D. R. Harris (ed.), The Origins and Spread of Agriculture and Pastoralism in Eurasia: 323-345, vol. 1. (1 vols.). 1996b. What's in the Name: the Mesolithic, the Neolithic and Social Change at the Mesolithic-Neo-lithic Transition In M. Edmonds and C. Richards (eds.), Understanding the Neolithic of Northwestern Europe: 1-35- ZVELEBIL M., DENNELL R. and DOMANSKA L. (eds.) 1998. Harvesting the Sea, Farming the Forest. The Emergence of Neolithic Societies in the Baltic Region. ZIVANOVIC S. 1975a. Mesolitic Population in Djer-dap Region. Balcanica VI: 1-9. 1975b. A note on the Anthropological Characteristics of the Padina Population. Z. Morph, Anthrop. 66(2): 161-175. 1975c. Prvo saopstenje 0 rezultatima antropolos-kog proucavanja skeletnih ostataka sa Padine u Djerdapu. Starinar N.S. (IIIV-XXV/1973-1974): 139-153■ 1976a. Cromagnon in the Iron Gate Gorge of the Danube. Nature 260(5551): 518. 1976b. The Masticatory Apparatus of the Mesolithic Padina Population. Glasnik Antropoloskog drustva Jugoslavije 3- 79-96. 1976c. Ostaci ljudskih skeleta iz praistorijskog na-lazista na Hajduckoj Vodenici. Starinar, N.S. XXVI. 124-129. 1979a. Antropoloski nalazi sa srednjevekovne ne-kropole Trujane. Zbornik Narodnog Muzeja IX-X: 163-175. 1979b. Apsolutna starost skeletnih ostataka kro-manjonaca sa Padine u Djerdapu.- 26-30. 1988. The Temporal Region and the Supramastoid Ridge in Mesolitic Skulls from Padina in the Iron Gates Gorge of Danube. Human Evolution 3(5): 329-334. UDK: 903(4-924.56)"633/634' Documenta Praehistorica XXVII The Lepenski Vir Fauna: Bones in Houses and Between Houses Vesna Dimitrijevic Institute of Regional Geology and Palaeontology, Faculty of Mining and Geology, Belgrade, YU e-mail: vesnadim@beotel.yu ABSTRACT - Besides its monumental stone sculpture and peculiar architecture, and its dubious position between the Mesolithic and Neolithic, Lepenski Vir culture is claimed as one of the first in Europe in which dog domestication occurred. There are notes of other domestic species' bones found in a context originally interpreted as belonging to a fisher-hunter-gatherer society. It is presumable that a subsistence strategy itself related primarily to animal exploitation, be it of tame or wild, mammalian or non-mammalian species, inspired the foundation of settlements on the Danube's banks in the Iron Gates. In this regard, the first results ofpreviously un-analysed osteological material from the locality are presented. The material originates from the Lepenski Vir excavation campaigns of 1968-1970, from the floors of houses, beneath them, and from the spaces extending between houses. IZVLECEK - Kultura Lepenskega Vira je znana po svojih monumentalnih kamnitih plastikah, nena-vadni arhitekturi in nejasni umescenosti med mezolitik in neolitik. Razen tega naj bi bila ta kultura ena odprvih v Evropi, kjer so udomacili psa. Obstajajo tudi zapisi o kosteh drugih udomacenih vrst, ki sojihprvotno interpretirali kot dapripadajo ribisko-lovsko-nabiralni skupnosti. Domnevamo, da je prav nacin prezivljanja, povezan pretezno z izkoriscanjem zivali - bodisi divjih ali udomacenih, sesatcev ali drugih vrst, vplival na ustanovitev naselbin na bregovih Donave v Zeleznih vratih. V tern clanku predstavljamo prve rezultate dosedaj se neanaliziranega osteoloskega materiala z naj-disca Lepenski Vir, izkopanega med leti 1968 in 1970. Material izvira s hisnih tal, plastipod njimi in iz prostora med hisami. KEY WORDS - Lepenski Vir; fauna! remains; Mesolithic; Neolithic; Neolithisation INTRODUCTION The faunal remains to be described in this paper originate from material collected in the course of the 1968-1970 excavation campaigns at Lepenski Vir. Excavations at the site of Lepenski Vir were rescue excavations, as part of an intensive archaeological research into the Iron Gates region that was initiated by the construction of the dam near Klado-vo. They began in 1965, as small-scale excavation, and with rather modest expectations (Srejovic 1966), but soon revealed a settlement with a new kind of architecture, and monumental stone sculptures, providing the foundations for the establishment of a new archaeological culture (Srejovic 1966a). The excavation area was enlarged to incorporate the whole area of the settlement, and work continued until 1970, when the houses of Lepenski Vir were relocated to a site above the rising water level (Canak-Medic 1970). Constant pressure imposed by a time limit, due to the construction of the dam, determined the excavation strategy, and the collecting of animal remains certainly could not have been the priority. Consequently, a rather small sample of animal bones was collected. Animal remains from the earlier excavation campaigns have been analysed and published in the form of a preliminary report (Bokonyi 1969). The sample consisted of 2999 identified specimens, of which only 630 were from Lepenski Vir I and II, with the balance deriving from Lepenski Vir III phases. A selection of this sample, as well as animal remains collected in succeeding campaigns, are stored in the National Museum in Belgrade. It was through the kind- ness of Ljubinka Babovic, a curator at the National Museum, that I learned of this material, and she entrusted me with analysing it, for which 1 am sincerely grateful. Animal bones stored in the National Museum in Belgrade are packed in bags, with labels designating their origin. The better part of the material originates from the spaces related to the houses, and usually only the number of the house is marked, or whether bones come from beneath the house floor. For the material occurring outside the architectural features, a digging layer is specified, and a square. Since the squares are 4x4 m, and digging layers are not easy to correlate with the building horizons, the position of bones occurring outside the houses is not very precise. The state of preservation of the bones is very good. There are fragile skeleton parts, bones in articulation, and bones from neonate animals, all indicating that physico-chemical agencies and soil quality did not alter the quantity of bones to a great extent, if at all. We can conclude that the fragmentation of bones is due to predepositional factors, while the amount of bones collected is determined by the excavators' decisions. These were strongly selective, therefore creating a biased sample. Hand collecting resulted in the under-representation of small animals and a small parts of large animal skeletons; furthermore, the vertical and horizontal distribution of fau-nal components cannot be reconstructed, and the collecting of bones belonging to particular units does not mean that all the bones related to particular features or indicated spaces were actually collected. In this respect, I have decided to present particular units that offer most promise of enlightening particular spots in the settlement area at Lepenski Vir, and enable the reconstruction of particular patterns of man/animal relations. Among the units presented, two are related to houses: house 40, at the eastern end of the settlement, and house 28, at the western end, while the two units derive from the area below the floors of houses 47' and 31, and were in fact, related to the spaces stretching between the houses (Fig. 1). The remains of the following species have been analysed: wild swine, (Sus scrofa Linnaeus) from the floor of house 40; red deer (Cervus elaphus Linnaeus) from the floor of house 28, and from the area below the floors of Fig. 1. Lepenski Vir Iplan of settlement (after Radovanovic 1996/ The position of houses 40 and 28, where the bones described in this paper are found on the house floors is marked by shading, and position of houses 47 and 31 where the bones are found below the house floors is covered by mosaic pattern. houses 47' and 31; brown bear, (Ursus arctos Linnaeus) below the floors of houses 47' and 31; beaver, (Castor fiber Linnaeus), roe deer, (Capreolus capreolus Linnaeus), and a large species of deer, Cervi-dae indet, from the area below the floor of house 47'. The wild boar carcass in house 40 House 40 (Fig. 2) is a non-superimposed house found at the downstream end of the settlement. It is interpreted as belonging to Lepenski Vir I, Phase Ic (Srejovic 1969.71, Fig. 15), i.e. Lepenski Vir I, phase I (Radovanovic 1996.176). A sculpture was found on the floor in the rear part of this house, and a grave (grave 61) beneath the floor, with its head just below the sculpture. There was an altar above the fireplace, and behind it, an adult female mandible ("grave 21") with a stone plaque within it. Another grave was interred behind the left frontal corner of the house (grave 60). None of the graves was noted to contain grave goods (Radovanovic 1996.176). Animal bones were found on the house floor, along its longer right wall (B-C (after Srejovic 1969), as it was labelled during the excavation), mostly of wild boar. As indicated in the subtitle, these are not just fragments showing the presence of the species, but complete or better parts of bones, indicating the deposition of a large section of carcass within the house at the time when its use was suspended (Figs. 3, 4). This could have been either suddenly and unwillingly, or as a prepared departure, with the intention of permanent or temporary abandonment. The following questions could bring us closer to under- / / 0v. \ Fig. 2. House 40, Lepenski Vir I (after Srejovic and Babovic 1983). Fig. 3- Wild boar skeleton with bones found in house 40 shaded. standing the house abandonment event: whether the bones were stripped of meat in the process of food preparation, and in such a way left behind in the house, or carcass parts were left, either as fresh meat, or dried; whether these remains belonged to a single animal or several different animals; what body parts were present, i.e. exactly what amount of carcass. The insight into the skeleton parts' distribution shows that those parts that have most meat are present: the sacral part of the backbone and the final lumbar vertebra, a fragment of atlas, parts of shoulder and pelvic girdles, and mostly the upper parts of fore and hind legs (Fig. 4). Since none of the skeleton elements were found represented by more then a single specimen, the assumption is appearing that we are dealing with the remains of a single animal. To confirm this, we should find all the skeletal elements showing the same age, and fitting by their size. Regarding the first question, we can notice that the epiphyseal lines are fused in all the bones except one, the thigh bone, in which the epiphyseal line of distal articulation is clearly observable along its whole length (Figs. 4.2a,b). Although thighbone distal articulation fuses late in ontogenetic development, this is also true for the articulations of proximal and distal ulna and distal radius (Habermehl 1975; Bull and Payne 1982). These are completely fused in the ulna and radius from house 40 (Figs. 4.4, 4.5), thus indicating older individual age. It is possible, and even probable, that the thighbone did not belong to the same animal as the other bones. Another consideration in establishing whether all the bones belonged to a single animal is their size (Tab. 1). Measurements after Driesch, 1976, except depths of radius proximal and distal end, and tibia distal end, which are perpendicular to their breadths; the measurements are greater than or close to the maximal values for wild boar at Vlasac {Bokonyi 1978) and Padina (Clason 1980), and clearly correspond to the male group, if compared with the Neolithic wild boar remains from Opovo (.Russell 1993)• hi any case, remains of probably two wild boars were found in house 40, possibly more then two, which may indicate that in the distribution of the kill, this house for some reason received a good part. Fig. 4. Sus srofa scrofa, house 40:1. sacrum, last lumbar vertebra and fragments of right and left pelvis; 2. femur sin. dist., a. cranial, b. medial view; 3. scapula sin.; 4. radiusprox. dext.; 5. ulna and radius sin.; 6. left astragalus, calcaneus and distal fibula in articulation, a. dorsal, b. plantar view; 7. a. same bones, the arrow points to the breakage of fibula, b. fibula diaphysis, c. left distal tibia, the arrow points to the rough surface articulating with fibula; 8. cut nuirks at calcaneus distal end; 9. cut marks at astragalus medial side. The tarsal bones, talus and calcaneus, together with the distal calf bone, remained in articulation (Figs. 4.6a,b; 7a,b,c). There are no traces of synarthrosis, which would indicate pathological fusing, so it is certain that they were buried in situ as articulated, and not removed from the site afterwards, when the organic tissue had decomposed. The quality of the deposit in which they were embedded, probably enriched by carbonates, made their connection firm even after excavation. In addition, a distal shinbone (Fig. 4.7c) and calf bone shaft (Fig. 4.7b) were found, fitting perfectly the talus and calf bone fragment, indicating that those bones were a continuation of the joint described. breadth of the distal fibula, which is the greatest breadth of the distal end, and the calcaneus' greatest breadth, which is measured in the anteroposterior direction. To reach a conclusion as to whether all the bones, or all the bones except the thighbone, are from the same individual, we need a comparison with measurements taken on a large sample of skeletons to find out the individual and sexual variation in the proportion of skeletal elements. Presently, I am not aware that these kinds of data exist. However, it is possible to conclude that all the bones from house 40 originate from a rather large male individual. Particular The final lumbar vertebra, as well as left and right pelvic girdle fragments, were attached to the sacrum (Fig. 4.1), while two long bones of the left front leg, the ulna and radius, were also in articulation (Fig. 4.5). Another important observation could be made on this part of the skeleton. The ulna is complete, except for damage to its central portion, while the upper part of the radius is broken. The breakage of both bones was caused by a strong blow, and happened while the tendons, and probably also muscular tissue, were still holding the bones together. The single fragment of long bones in the right front leg, the proximal radius, points to a similar breakage (Fig. 4.4), a consequence of the patterned manipulation of the carcass parts. We should stress here that we are by no means certain that all the bones found at the site were collected. On the contrary, the excavation technique at Lepenski Vir mainly entailed the taking of only a selection of bones, and this selection was made by an excavator unfamiliar with osteological material. So it is quite possible that we are missing some parts of the carcass originally placed in house 40. In fact, they were primarily collected due to a certain notion in relation to the graves of this house, since the original label bore the legend "animal and human bones". Finally, we should consider whether the bones of wild boar found in house 40 were left in the house as a complete carcass, as meat carcass parts, or just as bones stripped of meat, and if we are dealing with meat carcass parts, whether they were raw, dried or otherwise prepared for consuming. First, it could be safely concluded that there was no complete carcass, since there are proofs of butchering. These are clearly observable on the tarsal bones, in the form of several deep transversal Lepenski Vlasac Padina Vir Bokonyi Clason house 40 1978 1980 sacrum cranial articular breadth 47.00 scapula glenoid process length 51.45 29.0-49.0 glenoid cavity length 38.70 glenoid cavity breadth 33.55 29.0-34.5 scapula neck length 34.05 32.0-33.0 ulna length 279.45 olecranon length 84.20 processus anconaeus depth 55.55 olecranon depth 44.70 coronoid process breadth 33.85 26.0 left radius distal end breadth 49.45 41.0-47.0 35.0 distal end depth 39.40 32.5-36.0 right radius proximal end breadth 40.55 39.5-41.0 36.5 proximal end depth 28.85 26.0-30.0 femur distal end breadth 61.00 tibia distal end breadth 40.55 32.5-44.0 37.0-40.0 distal end depth 34.80 29.0-38.0 fibula distal end breadth 21.45 astragalus lateral half length 50.00 47.0-55.0 50.0-53.0 distal breadth 31.85 29.0-35.0 29.0-31.0 calcaneus length 102.90 90.0-107.5 71.5-108.0 greatest breadth 41.60 37.0-40.5 38.5-44.0 grooves on the calcaneus (Fig. 4.8), and few short grooves on the talus (Fig. 4.9), made in the same direction, and in same action indicating disarticulation of the lower hind leg. There are no cut marks on other bones, but there are indications that other butchering techniques besides cutting might have been applied, such as a blow to the middle of the radius and ulna shafts. The left and right pelvic girdle fragments attached to the sacrum were probably broken in the course of the disarticulation of the left and right flanks. As we have already concluded that primary butchering was performed, we could further ask whether the butchering process was continued to the point of completely stripping the meat from the bones. In this respect it is important to note that there are no traces of filleting, which would be very difficult if not impossible to perform and avoid the contact of sharp artefacts with the bone necessary for this operation. This is especially true of the shoulder blade, on which filleting leaves a characteristic long longitudinal groove (Binford 1981.Fig. 4.06). This leads to the assumption that there were several meat parts present, before the house was abandoned, either as raw meat, or dried, or prepared for drying. A part of the left pelvis fragment was burnt, which could indicate contact with fire or hot smoke while the meat was dried, but we have no other signs which would certify this method of food preparation. Tab. 1. Sus scrofa, house 40, measurements (mm). Why was the meat placed within the house? I consider it is less probable that it was left because of the sudden abandonment of the house. There is evidence that animal parts, specifically red deer antlers, were left in many houses, which certainly was not unplanned and unintentional. An example of antlers in house 28 will be described later in the text. It is more likely that the placement of the wild boar carcass parts in house 40 reflects a common tradition performed in connection with an abandonment event in house 40 or the settlement itself. The meaning of this custom could have been to make offerings to the house spirits, because of the quitting of the house for a certain period, or because the use of the house was suspended, while life continued in other houses. Otherwise, it may be a case of food storage for the settlement's inhabitants, or even a sympathetic offering to a chance passenger. Bone raw material collection under the floor of house 47' House 47' (Fig. 5) is at the edge of the lower terrace and is covered by another house of similar outline, labelled as house 47 (Srejovic and Babovic 1983■ 138). House 47' is built above two smaller houses: the rear part of house 53 and the base of house 58. Both houses 58 and 53, and house 47' belong to Lepenski Vir I (Srejovic 1969.Fig. 8, 14, 16; Radovanovic 1996). A large, non-superimposed house, 54, on the left side of this group of houses, and a very small house, 49, to their right, are interpreted as belonging to phase c of Lepenski Vir I (Srejovic 1969. Fig. 15). The animal bones collected are from below the floor of house 47'. A more precise position is not noted, but they were deposited most probably in the space between houses 58 and 53 (along the west side of house 58, or in front of its left corner, or behind house 53). The bone assemblage is diverse, both in the species and the skeletal elements present (Tab. 2) (Figs. 6-9). There are the remains of at least 7 animals, belonging to 5 different species (beaver, bear, and three species of deer) so fur and meat animals, large, medium and small are present. There are broken and complete mandibles of several species, two shoulder blades and many fragmented metapodials. Among the unidentified fragments, long bones and Fig. 5. House 47 (overlaying house 47') and adjacent houses fafter Radovanovic 1996/ metapodial splinters dominate. The composition of the assemblage itself indicates that the reason for its being collected at the site was not the use of the animals' soft parts (meat, skin), but the bones themselves. In fact, there is quite convincing evidence that they were collected as raw material. Fragmented metapodials are the most representative skeletal part in this respect. The metapodial bones of ruminants, especially of cervids, because of their shape and structure, were the most desirable skeletal part for artefact production throughout the prehistoric periods. A distal metatarsal of roe deer, and 11 fragments of red deer metapodials were found in this assemblage. The red deer metapodials belonged to at least 3 individuals, since among 7 fragments of metatarsals (Fig. 6.3) it was possible to identify two right proximal metatarsal fragments bearing the same elongated facete for articulation with the centrotarsal bone (Figs. 6.3a,c). These fragments would have originated from two individuals. Two remaining proximal fragments are from the left metatarsal, and probably from a single bone (Figs. 6.3b,d). We cannot determine from the proximal parts whether they belonged to fully grown animals, since the proximal epiphysis fuses earlier, but from the two fragments of distal metacarpals, left and right, with fused epiphysis (Figs. 6.1a,b), there is evidence for the existence of at least one fully grown animal. On the other hand, it is not possible to differentiate distal unfused epiphyses and to identify whether they belong to metacarpals, or metatarsals, but since they are of different sizes (Figs. 6.2a,b), and the surface for fusing in the larger specimen is much more compact, revealing a better degree of ossification, it is possible to conclude that they belonged to juvenile individuals of different ages, or in any event, to two different animals. Consequently, the red deer metapodials in the assemblage are from at least three animals, one adult, and two juveniles. All the fragments show certain degree of weathering, having been exposed to atmospheric influence for some time. Their fragmentation started with lengthwise splitting, but continued in various ways. Two fragments belonging to the same left metatarsal (Figs. 6.3b,d) show that the bone was first split lengthwise, which resulted in the separation of these two fragments, but thereafter the larger fragment was broken transversally. The left distal part of the metacarpus (Fig. 6.1a) was modified by flaking its rim. On the diaphysis fragments there are small depressions made by multiplied pressure on the bone, probably in the course of artefact production. The reason that carpal and tarsal bones of reed deer (Figs. 6.4, 6.5) were found within this assemblage is probably not related to the intention of their further modification, but is the consequence of their being articulated with those bones that were of interest -metapodials. Perhaps someone who had the skill of making artefacts took a part of a skeleton that he knew to be useful for his purpose from a site where primary butchering was taking place, took it to a work area, and disarticulated it as the first step of the work. There are three carpal bones, two of them, the left intermedium and left radiale, probably belonged to the same animal, since their articulations fit perfectly (Figs. 6.4a,b). Another one, the left intermedium, belonged to another animal, probably a young one, as the structure of the bone shows less ossification (Fig. 6.4c). A single phalanx also belonged to a young animal, with unfused proximal epiphysis, and peculiar traces on its diaphysis caused by rodent teeth (Fig. 6.6). It is highly probable that the toolmaker was not always on its working place, so during his absence, a small rodent was sharpening its incisors on this bone while it was exposed on the surface. Species Skeletal part NIS MNI Castor fiber (beaver) mandible 2 1 Ursus arctos (brown bear) mandible 2 1 Cervus elaphus (red deer) antler 1 3 mandible 4 upper molar 1 scapula 1 distal humerus 3 distal femur 1 distal metacarpal 2 proximal metatarsal 3 metatarsal diaphysis 4 distal metapodial p unfused epiphyses carpals 3 tarsals 1 second phalanx 1 Capreolus capreolus (roe deer) distal metatarsal 1 1 Cervidae indet. (a species of large deer) scapula 1 1 NIS - number of identified specimens; MNI - minimal number of individuals. Tab. 2. Species and skeletal parts distribution from the area under the floor of house 47'. The roe deer distal metatarsal (Fig. 7) is also a remnant of lengthways splitting. Its proximal part might have been used for making bone artefacts as well, but this fragment itself shows traces of modification and use. The obliquely broken diaphysis is polished, while its dorsal side has many scratches which are use traces. Its pointed end was broken, perhaps in the course of a working process at the site itself. The other possibility is that the artefact was brought into the workshop to be repaired. Two shoulder blades found in the assemblage, one from a large species of deer, and another from a red deer, bear clear evidence of filleting in the form of sharp longitudinal furrows (Figs. 8.1c,2b). There are, also, further modifications on the blades, and it is possible to assume that they belonged to carcasses butchered elsewhere and brought to the site as a raw material. These further modifications are best observed on the surface of the large deer shoulder blade (Fig. 8.1c), in the form of multiplied pit-like bone damage, made probably not by direct, but indirect blows or pressure by some implement pressing on the blade, and showing successive movements of that implement, in the course of manufacturing. Two of those groups of pit-like damage lie over the filleting marks, thus demonstrating the sequence of work performed on the bone. Somewhat lower, in the area of the shoulder blade neck, there are two irregular, semi-circular cuts, while the whole surface of the bone is covered with tiny scratches. The question of identification of the large deer shoulder blade is intriguing. Cervid species of a stature larger than red deer are not recorded in the Postglacial of the region up to date. However, a giant deer, Megaloceros sp, and elk, Alces alces (Linnaeus), inhabited the central Balkans in the course of the Pleistocene epoch (Dimitrijevic 1983; 1997), and survived in the Carpathian basin even in the Postglacial period (Bartosiewicz 1999). Identification is made more difficult by the fragmented state of the distal articulation, although its circular form and the position of the coracoid process clearly indicate a deer (Fig. 8.1a), and exclude cattle species. The only measurement obtainable, the diameter of the neck of the scapulae (46.1 mm), is greater than the variation range of red deer (33.0-44.0 mm at Vlasac (Bokonyi 1978); 27.0-40.0 mm at Padina (Clason 1980)). It is possible that the blade belonged to a young animal. Although the co- Fig. 6. Cervus elaphus, lower extremital bones: 1. metacarpus dist., dorsal view, a. sinistralis, arrows point to the flaked rim of the fragment; b. dextralis; 2. distal unfused epiphyses, lateral view, a. larger specimen, probably from a slightly older animal, b. smaller specimen, probably from a younger animal; 3■ metatarsal fragments, a, c proximal dextralis; b, dproximal sinistralis; e,fg diaphyses; 4. carpals, a. radia-le sin., b. intermedium sin., c. intermedium sin,; 5. centrotarsale, plantar view; 6. phalanx II, lateral view, arrow indicates rodent teeth marks. jaw. There is a conspicuous glow in the simphyseal region, both in the front part, below the incisors, and on the inner sides, on the simphyseal joint itself, which is a possible consequence of its use as a grinder. The presence of beaver mandibles in this context is important, since at the other sites in the Upper Gorge of the Danube, mandibles are overwhelmingly represented among beaver remains (36 mandible fragments against only 5 postcranial bones at Padina [Dimitrijevic and Boric, in preparation}', 15 mandible fragments among 71 beaver remains at Vlasac, after Bokonyi 1978), and there remains the question of the pattern of use of this animal in the Gorges. The incisors are still in place, while they are mostly lacking in the mandible fragments at Padina. racoid process is fused, the structure of the bone inside the glenoid cavity is not very compact, but porous, which is characteristic of young animals. For the remaining items in the assemblage, represented by skeletal parts not so commonly used in the process of tool making, and not bearing exact modification traces, the intention of including them in this collection of raw material is not so clear. The collection of lower jaws of different species, a deer, a beaver and a bear, is interesting. There are 4 fragments of a red deer's lower jaw, probably belonging to a single branch (Figs. 9.4a~d), thus indicating an in situ breakage. The fragments are heavily weathered and the largest bears cuts and multiple, grouped scratches on the surface. On the other hand, the beaver left and right jaws have almost complete horizontal ramie, and the bear's left mandible even has the incisor part complete, which is the easiest to break down. The beaver left and right mandible branches (Figs. 9-3a,b) belonged to a single lower The bear mandible belonged to a young adult animal (Fig. 9-1)-There are several short, deep, and sub-parallel cuts in the middle of its basal margin, grouped in two places (Fig. 9-la). These cuts indicate primary butchering, after which the assumption is that the bone was brought to this site. Besides this left mandible, there is a fragment of a right bear mandible (Fig. 9-2), in the form of a single small basal fragment of a horizontal branch, with old breaks. This could only originate from a heavily fragmented mandible. So, there are two bear lower jaws at the site, one very well preserved, if we exclude excavation breaks, and the other, for some reason, greatly fragmented. The flat mandible surfaces might have had a role in tool manufacturing, but might also have been of interest because of the teeth it contained, among which incisors and canines would have been especially popular for amulet production. Young deer and brown bear below the floor of the house 31 House 31 is found in the group of houses of Lepenski Vir I positioned in the Middle Terrace (Fig. 10). It is older than houses 18 and 23, since its left front wing is cut by house 18, while the rear of house 18 is covered by house 23, but probably later than house 19 (Radovanovic 1996). The bone assemblage to be described was collected from below the floor of house 31, so it was accumulated at an open area of an early settlement phase. The bone assemblage contained large mammal bones, fish teeth and bones, a long diaphysis bird bone, and several human bones, probably in relation to grave 97. The complex stratigraphic situation evokes suspicion concerning the assemblage's unity. However, the remains of two skeletons, one of a young deer (Fig. 12), and another of an adult brown bear (Figs. 11, 13, 14), confirm the unity of the assemblage. Both animals were presumably butchered on the spot, showing in that specified process differences, in respect to their age and size, which influenced the butchering method itself. The following skeleton parts of the young deer were found (Figs. 12.1-13): the lower right mandible, atlas, front legs bones (proximal humerus, radius distal epiphysis, a carpal bone, proximal metacarpals), a pelvic fragment, hind leg bones (distal femur diaphysis and a related, unfused distal epiphysis, proximal parts of the left and right tibia dia-physes, and related right unfused epiphysis, the distal part of the right tibia diaphysis, astragalus, and calcaneus with unfused tuber calcanei), as well as a single third phalanx characterised by porous bone structure indicating incomplete ossification. All the skeleton parts indicate a young, growing animal. The lower jaw bears milk dentition, the long bones have both proximal and distal epiphyses unfused, and short bones have a porous structure. It is possible to conclude convincingly that these are the remains of a single animal, since all the skeleton parts indicate a similar age. The most precise age is given by the mandible (Fig. 12.1). It is a right mandibular branch, broken orally at the diastema, while the aboral processi of the vertical ramus are also broken. The first and second decidious molars are in alveoli, while the third milk molar was lost post mortem, but its alveoli are fully preserved. The first milk molar has no traces of crown wear, but on the second, some slight wear is observable on the tips of the crown. There are no traces of the formation of permanent premolar crypts. Behind the third milk molar alveoli, the first permanent molar is erupting, and behind it, the crypt is opened where the germ of the second molar was developing. The mineralisation of the first permanent molar crown is complete, and the infundibu-lum is well formed. The mandible is broken through the first molar alveoli, and the tooth itself damaged in its lower portion, so it is not possible to observe whether root formation had started, but, since the eruption began, it is to be expected that root formation had also started. This stage of development is corresponding to the age of 4 months [Brown and Chapman 1991). This also gives us the season of this animal's death, as well as the season of this particular hunting and butchering episode at Lepenski Vir. Since deer give birth in May/June (Biitzler 1986), this means that the animal was hunted in the early autumn, most probably in late September or early October. The epiphyseal fusion is not so accurate for ageing, and it is not studied in detail like tooth eruption, but finds like these remains of a young deer below house 31 could be very important if such a study is going to be made in the future, since they give us a set of data direct from prehistory. It is important to note that elements of both the left and right front legs were found (fragments of the left and right metacarpals) (Fig. 12.6), as well as those of the left and right hind legs (left and right proximal tibia) (Fig. 12.9), in considering what part of the body was actually present at the site. Though we cannot quite exclude attritional processes, the bones are well preserved, in spite of their porous structure, and the presence of both diaphyses and ■ I V i I mm , WW;*"'" '• i Vjft^-lX-J 11 1 ■ ■ Fig. 7. Capreolus capreolus, distal metatarsal, a. dorsal, b. plantar view, arrow indicates broken polished end. where they are not so clearly observable as on compact, mature bones. In addition, since this young deer skeleton was found with the remains of another butchered animal, a brown bear, we should not search for any other reason for its deposition. The brown bear butchered at the site was an adult animal, of a size comparable to representatives of the same species in the region (Tab. 3)- According to the skeleton parts' distribution (mandible, vertebral column and rib cage elements represented, both the front and hind legs, as well as both the left and right side of the body) (Figs. 11,13,14), the whole carcass was probably butchered on site, excluding possibly only the skull. Unlike the juvenile deer from the same place, traces of primary butchering, as well as traces of further processing of the disarticulated skeleton parts are easily detectable, and they are found on expected places on the bones, and rather easy to interpret. Fig. 8. Scapulae with filleting marks: 1. Cervidae indet., scapula sin., la. distal view, lb. lateral vieiv, 1 c. medial view, detail, upper two arrows point to pitlike depressions, arrow in the middle to longitudinal filleting mark and pitlike depression below crossing over it, and the one at the bottom points to halfcircular incisions; 2. Cervus elaphus. scapula dext., lateral view, 2a. caudal view, detail, arrow points to filleting mark. epiphyses of unfused long bones (distal femur, proximal and distal tibia) (Figs. 12.8,9,10) and calcaneus (unfused tuber calcanei) (Figs. 12,13a,b), show that they were not spread out further. It is more probable that not all the bones were collected by the excavators, and it is reasonable to suppose that the whole body, except the head of the animal was at the site. There are no traces of butchering, but we have to bear in mind that for the butchering of a young animal much less effort is needed, and much fewer interventions made by artefacts. Moreover, cuts made by artefacts on young bones remain mainly in the layer of cartilage, because of which they are not detectable in fossil material, or on porous bone surfaces, The articular end of the mandible bears distinct traces of cutting (Figs. 13-la,b), most probably the consequence of the disarticulation of the mandible from the skull after the masticatory muscle had been already removed. From the vertebral column, two vertebrae are preserved which are mutually articulated, since one is the last thorasic (Fig. 13-2), wearing the anterior demifassete for rib articulation, humerus ML distalis 87,9 femur ML proximalis 86,1 ML distalis 73,8 radius ML proximalis 37,6 metatarsus II length 69,6 metatarsus V length 82,9 ML - medio-lateral breadth Tab. 3■ Ursus arctos, extremital bones (mm). Fig. 10. House 31 and adjacent houses Rafter Rado- Fig. 11. Bear skeleton with bones found in house vanovic 1996/ 11 shaded. and the second has the anterior processi (praezigapophysys) well fitting with the posterior processi of the thoracal vertebrae, so this is probably the first lumbar vertebrae (Fig. 13-3). The cut-marks (several short and shallow parallel cuts) are found at the base of the lumbar dorsal spine and derived from the removal of tenderloin (comparable to Binford 1981.Fig. 4.21). Remarkable cut marks are found on the hip-girdle and long bones, originating both from disarticulation of the long bones from the girdle, and from the further processing of meat parts. At the left hip-girdle (Fig. 13.5), cuts are positioned at the ileum and ischium portions. There are two short parallel cuts at the ileum (Fig. 13.5b), at its narrowest part, while many artefact traces cover the ischium. They were made, if not by different artefacts, then in the course of "operations" of varying intensity: there are short and shallow parallel cuts extending trans-versally, afterwards a very deep and long single furrow, crossed by a short, deep cut (Fig. 13.5a), while the third group of traces are represented by pit-like notches made by multiple chiselling. At the right hip-girdle fragment the tuberosity at the acetabulum rim is knocked off (Fig. 13.6), which might indicate butchering by blows. There are two more fragments of hip-girdle, impossible to reconstruct with larger fragments, since parts of the broken bone are missing. One reveals longitudinal Fig. 9■ Mandibles: 1. Ursus arctos, mandible sin., outer view, arrow points to the position of cut marks, la. cut marks at basal part of horizontal branch; 2. Ursus arctos, mandible dext., fragment of basal part of horizontal branch; 3- Castor fiber, a. mandible dext., b. mandible sin.; 4. Cervus e/aphus, mandible dext., a. fragment of horizontal branch with diastema and fragmented P3 and P4 in alveoli, b. fragment ofM2, c. mandible fragment with M3, 4d. coronoid process. shallow furrows, most probably filleting marks, and the other is covered by oblique cuts. From the front leg long bones, a distal humerus and three fragments of a radius were found. The cuts on the humerus are numerous. There are many oblique, sub-parallel cuts, seemingly made with a single artefact and during a single operation of butchering. They are positioned at the medial epicondyle, starting from its base and spreading to the diaphysis uplift (Fig. 13.8a). At the lateral epicondyle crest there are again oblique sub-parallel cuts, but they are Fig. 12. Cervus elaphus. infantile: 1. right mandible with milk molars (D2 and D3), and Ml erupting, inner view; 2. atlas, dorsal view; 3• humerus dext. prox., lateral view; 4. radius sin., distal epiphysis; 5. fragment of capitato-trapezoid sin., fragmented, palmar view, articulation fitting to proximal left metacarpal shown in 6b; 6. metacarpus, a. dext. prox., b. sin. prox., c. same bone from the inner aspect showing longitudinal spliting; 7. pelvis dext., fragment acetabulum with fragment of ischium; 8. femur sin., medial view, a. distal end of diaphysis and b. unfused distal epiphysis; 9■ tibia dextralis, a. unfused proximal epiphysis, cranial view, b. proximal end of diaphysis. cranial view, c. unfused proximal epiphysis, proximal view, d. tibia sinistra-lis, proximal end of diaphysis. crista tibiae damaged, cranial view; 10. tibia dext.. distal en-fused diaphysis; 11. Ph III; 12. astragalus dext., dorsal view, articulation fitting to calcaneus; 13. calcaneus dext., a. unfused tuber calcanei, b. corpus calcanei, medial view. the diaphysis in two of these three fragments, were probably made when the radius was already separated from the ulna. deeper (Fig. 13-8b) and seemingly made with much more effort, which could derive from cutting the tendon binding the humerus/ulnar articulation in the freshly killed animal (comparable to Binford 1981. Fig. 4.30. b, Fig. 4.31). At the trochlea itself, there are no cut-marks, which means that further disarticulation was easy after the tendons binding the humerus and ulna were severed. For the same reason, there are no cut-marks at the radius. Besides the proximal end of the radius (Fig. 13-9), two more radius fragments were found, all with old breaks, showing that further processing was performed on site after primary butchering. Small cuts, found on From the hind leg long bones, a proximal right (Fig. 13-7) and a distal left part of the femur (Fig. 13.10) were found. At the proximal right femur, damage caused during the dismemberment from the hip-girdle is observable: two cuts on the caput femori, and cuts below the caput on the femur neck (Fig. 13.7a). The interesting feature are several cases of dimple damage below the trochanter minor, which all bear scratches on their bottoms, originating from a multi-pointed artefact. The remaining part of the diaphysis bears other longitudinal and oblique scratches and cuts, and the end of the fragment was transversely cut after the bone was split longitudinally. On the other hand, the distal end of the left femur bears filleting marks in the form of longitudinal shallow cuts on its diaphyseal part (Fig. 13.10a). There was Fig. 13■ Ursusarctos, below the floor of the house 31: 1. mandible sin., articular process, a. inner view, b. outer view, arrow points to cut marks made below the articular condyle; 2. last thoracal vertebra, lateral view, cranial side turned to right, arro w points to demi-facet for rib articulation; 3-first lumbar vertebra, lateral view, cranial side turned to right, arrow indicates the position of cut marks; 4. a.,b., c. three fragmented costae, 4d. cut marks at proximal part of costa shown at 4b, inner side; 5. pelvis sin., arrows point to position of cut marks, 5a. at distal part of ischium, 5b. at narrowest part of ilium; 6. pelvis dext. with damaged tuberosity at the acetabulum rim; 7. femur dext. proximo-lis, caudal view, la. cut marks at caput femoris and neck; 8. humerus dext. dist., caudal view, arrows point to positions of cut marks, 8a. cut marks at medial side, 8b. cut marks at lateral side; 9- radius dext. prox., caudal view; 10. femur sin. dist., cranial view, 10a. same bone, caudal view, detail of the surface covered by filleting marks; 11. fragment of femur diaphysis covered by filleting marks, probably in the continuation of distal femur shown at this figure, 10a. another fragment of diaphysis belonging to the same distal femur, and showing the same longitudinal grooves, which were obviously a continuation of the grooves found on the larger fragment (Fig. 13.11). Since the breaks are old, it is apparent that after filleting, further breakage of the bone happened on site. For the assumption that the whole animal was butchered on site, it is important to note the presence of short paw bones from the left anterior leg, and left hind leg. There are 4 carpal bones (Figs. 14.2a-d), well-preserved and not fragmented, belonging to the left front leg, and the second and fourth metatarsal belonging to the hind limb (Figs. l4.1a,b). Two first, three second, and four third phalanges (Figs. 13.3a-h) were found, also well preserved and not fragmented, except a single broken third phalanx. It is not possible to say whether they belong to the front or hind leg. In addition, three sesamoid bones were found (Fig. I4.2e). These bones are small (with a maximum length of 12 mm in this case), bean-like structures, not jointed with other bones, except for muscular tissue and tendons, showing again that the bones of the skeleton did not accumulate independently, but as parts of the skeleton, which primary disarticulation, forced by men, as well as final natural disarticulation, happened at the place. The deer skull with antlers in house 28 House 28 (Fig. 15) is a non-superimposed house found at the upstream end of the settlement. Its peripheral position and size are similar to house 40, as Fig. 14. Ursus arctos, belotv the floor of the house 31, metapodial and short bones: 1. metatarsalia sini-stralis, dorsal view, a. metatarsus II, b. metatarsus IV; 2. a-d. carpalia sinistralis, a. ulnare, b. carpale 1, c. carpale 2, d. carpale 3, e. three sesamoid bones; 3. phalanxes, a-b. first phalanxes, c-e. second phalanxes, f-h. third phalanxes. well as the arrangement of stone art objects (Radovanovic 1996). Two sculptures were found, on the right and left sides of a large stone slab in the rear of the house, and because of the large rocks behind it, the house was named "the sanctuary below the rocks" (Srejovic and Babovic 1983). A red deer skull with antlers was found on the floor. The antlers, both left and right, were preserved at a length of over 50 cm (Fig. 16), and attached to the skull, must have been a voluminous item, whose find itself shows the tradition of the placement of particular skeletal parts of animals in the course of house abandonment. The skull was damaged when lifted from the position where it lay to such an extent that it is not possible to reconstruct it. A delicate structure of bones, rather thin and unfused Fig. 15. House 28, Lepenski Vir I (after Srejovic and Babovic 1983/ skull roof bones, and fragile processes contributed to this. However, it is possible to observe that the frontal, temporal and occipital parts of the skull were present, together with the left and right upper jaws. The bones of the face, the nasals and praema-xillars, were not identified. The animal's age is clearly indicated by its upper teeth, which are in the last stage of milk/permanent dentition replacement and with the last permanent teeth erupting (Fig. 17). Both third milk molars are still holding above the crowns of permanent fourth premolars, although half of the crown of the one on the right side of the jaw is worn out (Fig. 17b); it would have been a matter of days before they fell out. The second and third premolars are erupting, as well as the third molar, preserved only in the left jaw. This stage of dental development in the upper jaw should be analogous to that in the lower jaw, and related to an age of 27 months (Brown and Chapman 1991). Since deer give birth in May/June (Butzler 1986), this means that the animal was hunted in the early autumn, most possibly in the late September or early October. The antlers are asymmetrical: the right one consists of a single branch, with just a slight protuberance at the site of the brow tine, while the left antler has a brow tine and simple crown with two tines. Due to their age they are extremely thin (right burr circumference 126 mm, left burr circumference 123 mm, right column circumference, 10 cm from the burr, 80 mm; the same measurement in the left column 76 mm). Although variability in red deer antlers is well known (Dragicic 1957), antlers from house 28 should be characterised as unusual in their length and asymmetry. Fig. 16. Cervus elaphus. house 28, left and right antler, frontal view. the assemblage from house 40 is important. This assemblage, originating from a part of a wild boar carcass, left in the house either as fresh or dried meat, indicates also a situation at the time of the house being abandoned. There is a possibility that the departure in this case was sudden and unwilling. However, a thoughtful and prepared departure might be a more plausible scenario, especially considering the symmetrical position, similar size and arrangement of stone art objects in relation to house 28, with deer skull and antlers. Whether there is a temporal connection between these two house abandonment events, or at least whether houses 28 and 40 belonged to the same building horizon might be enlightened only by absolute dating. The other two units described in this paper also depict particular events: episodes in the communal life of the settlement. The assemblage found in the area later covered by house 31 contains the bones two animals. They were brought into the settlement after hunting, and butchered on the spot, possibly, but not necessarily at the same time. The remains of one of them, a young deer, point to the same hunting season, early autumn, as well as the deer skull from house 28. The identification of when a particular animal died leads us to the expectation that through CONCLUSION - THE ARCHAEO ZOOLOGY OF TROPHIES The osteological material described here is not what we are generally accustomed to as animal remains on archaeological sites - a tiresome piles of bone fragments, the remnants of meals and activities difficult to reconstruct, when the main discussion after a thorough analysis is related to measurements and statistics of those many fragments arranged by species and skeletal parts. It is rather what one might describe as a collection of trophies. It seems justifiable to use this expression, since the hunted animals presented, and their particular importance for the society dwelling at Lepenski Vir, and because the units described here related to houses and defined areas between houses, are so precious from the ar-chaeo-zoological point of view, as trophies are for hunters, and finally, because the placement of animal remains in houses in particular instances, especially of red deer antlers, had meaning beyond their economic value. Red deer skulls with antlers have been found in at least 13 houses at Lepenski Vir (Bokonyi 1969). As has been stressed by the inspired doyen of Lepenski Vir culture research, Srejovic (1969.137), red deer antlers symbolise connections of death and renewed birth, in relation to their unique feature of growing and rejecting yearly cycle, and their amazing ability to grow bigger, stronger and more beautiful every year. The voluminous antlers on deer skulls positioned on the house floor point further to an item very important for understanding the "life cycle" of the houses themselves, if not the entire settlement. Their deposition marks a time when the house fell into disuse, the house abandoned, and shows that this event was, besides its practical connotations, also of symbolic importance. It also suggests it was prepared. A question remains as to whether house abandonment was a patterned ritual unique to every house, or whether it changed diachronically through the succeeding phases of Lepenski Vir settlement. Although all the antlers and skulls found in these contexts were not collected during the excavation, because of their volume and apparent meaning, they could not have remained unobserved; they were noted and described. This was not necessarily the case with other osteological finds deposited in the houses, especially if they were smaller skeleton elements, or remains of smaller animals. In this respect, the analysis of animal remains, such as large mammals, but also fish, and even birds, might connect exploitation of certain animal species with a specific time of the year. This would undoubtedly help in understanding the "life cycle" of houses and the settlement itself. Another assemblage, found at an area later covered by house 47', indicates an activity that is to be expected in a settlement of the period: work on bone as a raw material. The bones found in the assemblage are not the final products of a bone workshop, which we are used to seeing reflected in an inventory of bone artefacts from an archaeological site, but mainly products of the initial phases of working on bone material. There are parts of skeletons of various species that were brought to the spot and disarticulated as the first step in the working process, while particular bones were further split, flaked or modified in another way. Enlightening utilitarian activities in the open areas within the settlement, presented in the osteological material found below the floors of houses 31 and 47', remind us of the general importance of animals in the subsistence strategy, but also to the probability that the foundation of settlements on the Danube's banks in the Iron Gates was primarily initiated by animal exploitation. The importance of animal resources for subsistence continued throughout the development of the Lepenski Vir culture, and is Fig. 17. Cervus elaphus, house 28, upper jaw, arrows point to third decidious premolars hanging above forth permanent premolars: a. P2-M3 dext., b. P2-M2 sin. reflected in shared activities exemplified in the archaeological record. But the animals also played a role in the sphere beyond economic importance, and their remains are found in symbolic relation to important events, like the abandonment of a house. -ACKNOWLEDGEMENTS I thank Ivan Stefanovic for conservation of antlers and bones, Dragan Milovanovic for kind help in making photographies and Dusan Boric for valuable comments on the paper. REFERENCES BARTOSIEWICZ L. 1999- The emergence of holocene faunas in the Carpathian Basin: a review. In Benecke N. (ed.), The Holocene History of the European Vertebrate Fauna, Modern Aspects of Research, Workshop, 6th to 9th April 1998, Berlin. BINFORD L. 1981. Bones, ancient men and modern myths. Studies in Archaeology. BOKONYI S. 1969 Kicmenjaci (prethodni izvestaj). In Srejovic D. (ed.), Lepenski Vir, Nova praistorij-ska kultura u Podunavlju: 224-228. 1978. The vertebrate fauna of Vlasac. In Srejovic D. and Letica Z. (eds.), Vlasac Vol. II: 35-65. BROWN W. and CHAPMAN N. 1991- Age assessment of red deer (Cervus elaphus): from a scoring scheme based on radiographs of developing permanent mo-lariform teeth./ Zool, Lond, 225: 85-97. BULL D. and PAYNE S. 1982. Tooth eruption and epiphysial fusion in pigs and wild boar. In Wilson B., Grigson C. and Payne S. (eds.), Ageing and Sexing Animal Bones from Archaeological Sites, BAR British Series 109: 55-71. BUTZLER W. 1986. Cervus elaphus Linnaeus, 1758 -Rothirsch. In Niethammer J. and Krapp F. (eds.), Handbuch der Saugetiere Europas, Band2/II: 107-139. CLASON A. 1980. Padina and Starcevo: game, fish and cattle. Palaeohistoria XXII: 141-173. CANAK-MEDIC 1970. Radovi na spasavanju Lepenskog Vira u 1970. godini. Saopstenja IX: 5-15. DIMITRIJEVIC V. 1983. Pleistocenski cervidi okoline Beograda. Diplomski rad, Rudarsko-geoloski fakul-tet, Univerzitet u Beogradu, Beograd. 1997. Upper Pleistocene mammals from cave deposits of Serbia. Geoloski anali Balkanskogpolu-ostrva 61, 2: 179-370. DIMITRIJEVIC V. and BORIC D. in preparation. Animal bones from the site of Padina in the Iron Gates, Serbia. In B. Jovanovic (ed.), Padina. DRAGISIC 1957. Jelen. Lovacka knjiga. Zagreb. DRIECH von den A. 1976. A Guide to the Measurement of Animal Bones from Archeological Sites. Peabody Museum Bulletin 1. Harvard University, Cambride, USA. HABERMEHL K-H. 1975. Die Altersbestimmung bei Hans - und Labortieren. Paul Parey, Berlin. RADOVANOVIC I. 1996. The Iron Gates Mesolithic. Ann Arbor: International Monographs in Prehistory, Archaeological Series 11. RUSSELL N. 1993- Hunting, Herding and Feasting: Human Use of Animals in Neolithic Southeast Europe. University of California at Berkeley. SREJOVIC D. 1966. Lepenski Vir, Boljetin - neolitsko naselje. Arheoloski pregled 8:94-96. 1966a. Lepenski Vir - a new prehistoric culture in the Danubian region. Archaeologia Iugoslavica VII: 13-17. 1969. lepenski Vir, Novapraistorijska kultura u Podunavlju. Srpska knjizevna zadruga, Beograd. SREJOVIC D. and BABOVIC L. 1983. Umetnost Lepenskog Vira. Beograd. UDK: 581.13+591.13:902.65/4-924.56)"633/634' Documenta Praehistorica XXVII Stable Isotopes, Radiocarbon and the Mesolithic-Neolithic Transition in the Iron Gates Clive Bonsall1, Gordon Cook2, Rosemary Lennon3, Douglas Harkness4, Marian Scott5, Laszlo Bartosiewicz6, Kathleen McSweeney7 department of Archaeology, University of Edinburgh, UK, CBonsall@ed.ac.uk 2Scottish Universities Environmental Research Centre, UK, g.cook@surrc.gla.ac.uk 3 Department of Archaeology, University of Edinburgh, UK 4Scottish Universities Environmental Research Centre, UK, d.harkness@surrc.gla.ac.uk 5Department of Statistics, University of Glasgow, UK, marian@stats.gla.ac.uk 6Institute of Archaeological Sciences, Lorand Eotvos University, Hungary, h10459bar@ella.hu 7Department of Archaeology, University of Edinburgh, UK, kath.mcsweeney@virgin.net ABSTRACT - The results of stable carbon and nitrogen-isotope analyses of human bone collagen from the Iron Gates sites of Lepenski Vir, Vlasac and Schela Cladovei are reconsidered in the light of recent developments in stable isotope palaeodietary research and new information on chronology. The revised data have implications for the interpretation of Lepenski Vir and Vlasac, and the timing of the Mesolithic-Neolithic transition in the Iron Gates. IZVLECEK - Vclanku smo preucili rezultate analiz stabilnih izotopov ogljika in dusika iz kolagena cloveskih kosti, ki izvirajo iz najdisc Zeleznih vrat: Lepenski Vir, Vlasac in Schela Cladovei. Pri tem smo upostevali najnovejse izsledke raziskovanja paleoprehrane s stabilnimi izotopi in nove krono-loske podatke. Nanovo pregledanipodatki vplivajo na interpretacijo Lepenskega Vira in Vlasca ter na casovno umestitev mezolitsko-neolitskega prehoda v Zeleznih vratih, KEY WORDS - Iron Gates; stable isotopes; radiocarbon; palaeodiet; Mesolithic; Neolithic; Lepenski Vir; Schela Cladovei; Vlasac INTRODUCTION The Iron Gates has an abundant and continuous record of human occupation in open-air settlements from the Late Mesolithic into the Early Neolithic, c. 8500-6500 BP (7500-5450 cal BC). While early farming settlements are well represented in other parts of the Balkan peninsula, Mesolithic sites are uncommon and there are few if any sites that were inhabited continuously from one period to the next. Therefore, the Iron Gates is arguably the only area of southeast Europe where the transition from Mesolithic to Neolithic can be studied in detail. The evidence for changes in subsistence practices across the Mesolithic-Neolithic transition in the Iron Gates was reviewed by Bonsall et al. {1997). Their assessment was based largely on the results of stable carbon and nitrogen isotope analyses of human bone from Lepenski Vir, Vlasac and Schela Cladovei. The purpose of the present paper is to re-examine the Iron Gates stable isotope data and their significance, in the light of better information on food sources and chronology. DIETARY RECONSTRUCTION FROM STABLE ISOTOPES: SOME BASIC CONSIDERATIONS Stable isotope analysis of carbon and nitrogen in bone collagen has become a standard technique for palaeodietary studies. The underlying principles may be summarized briefly as follows: • stable isotope ratios in bone collagen reflect those in diet • ratios of stable isotopes vary naturally between major food sources • therefore the importance of different foods in human diets can be estimated from the isotopic composition of collagen. Table 1 lists "typical" 813C and 815N values of bone collagen for various major food sources. With each step along the food chain, there is fractionation of one isotope relative to another, resulting in a change in ratio. Most workers assume a slight enrichment in 813C (up to 1%0) and an enrichment in 815N of 3-4%o between the bone collagen of the food source and that of the consumer. Because aquatic food webs are more complex than terrestrial food webs, this results in much higher 815N values at the top of the aquatic food chain. These factors lead to the "expected" values in bone collagen of humans shown in Table 2. Food source 813C (%0) 515N (%<,) C3 terrestrial herbivores -21.0 +5.0 freshwater fish -20.0 +11.0 marine fish -13.0 +13.0 Tab. 1. "Typical" &3C and 8'5N values in bone collagen of three major animal food sources available to humans. The figures for stable isotope ratios in food sources and estimates of trophic level effects cited above should be regarded as "global" averages. There can be significant variation between ecosystems. Therefore, precise dietary reconstruction requires detailed knowledge of the isotopic compositions of local food resources. Humans feeding on: S13C (%<») 815N (%») C3 terrestrial herbivores -20.0 +8.0 freshwater fish -19.0 +14.0 marine fish -12.0 +16.0 Tab. 2. "Expected" &3C and 5,3N values in bone collagen of humans feeding exclusively on each of the food sources listed in Table 1. Other factors need to be taken into account when interpreting stable isotope data. Bone collagen in adult humans is estimated to have a turnover (replacement) rate in a range of 10-30 years (Mays 1998). Therefore, stable isotope ratios are a reflection of average diet over decadal timescales. However, in children the turnover rate is probably more rapid (Kleppinger 1984). Furthermore, the nitrogen isotopic composition of bone collagen is thought to reflect mainly the protein component of the diet as virtually all nitrogen in food comes from protein. Carbon in collagen can be derived from protein, fats or carbohydrates. In high protein diets, the carbon in collagen is thought to come mainly from protein, but in low protein diets a significant proportion of the carbon is probably derived from carbohydrates (Ambrose 1993)- REVIEWING THE IRON GATES DATA Since the original study of the Iron Gates stable isotope data (Bonsall et al. 1997), new information has become available that makes it possible to refine some aspects of the interpretation. This includes information on the isotopic composition of the major food sources, and more accurate age estimates for the human bone samples from Lepenski Vir, Vlasac and Schela Cladovei. There is also a larger data set for Lepenski Vir and Vlasac that can be considered. Food sources In the original study, because of a lack of detailed information on the isotopic composition of local food sources, the human bone stable isotope results from Lepeaski Vir, Vlasac and Schela Cladovei were plotted against data for North American food sources published by Schwarcz {1991), with allowance for fractionation effects (Fig. 1). From this it was concluded that the diets of Mesolithic and Early Neolithic populations were a mixture of foods drawn from two major sources, freshwater fish and terrestrial herbi-vores/C3 plants. It is true that, when compared against the North American data, average 815N values for Mesolithic skeletons from the Iron Gates appear unusually high for a population that subsisted mainly on freshwater fish (cf. Fig. 1). This has led some other researchers (Hedges et al, 1998; Schulting 1999) to infer that the Iron Gates Mesolithic diet must have included a high proportion of Danube-caught anadromous fish from a marine environment, i.e. the Black Sea. The possibility that anadromous fish were the source of the high 815N values was also considered by Bonsall et al. (1997) but was rejected because there was no corresponding enrichment in the 813C values, and because average S15N and 813C values for Mesolithic adults at Schela Cladovei (where there is abundant evidence for Mesolithic exploitation of Fig. 1. S'}C and 5'5N profiles of human populations from Schela Cladovei (A), Vlasac (B) and Lepenski Vir (C) plotted against the ranges of North American aquatic (A), marine (M) and terrestrial (T) food sources, derived from Schwarcz (\ 991 )■ Redrawn from Bonsall et al. (1997). sturgeon) appeared very similar to those of their counterparts at Lepenski Vir and Vlasac, where no sturgeon remains were identified by Bokonyi (1969; 1970; 1977). Although he did not claim to have been a fish expert, masses of large sturgeon remains would not have escaped his attention. Moreover, he did identify Neolithic sturgeon at the site of Mihajlo-vac-Knjepiste (Bokonyi 1992), downstream of Schela Cladovei and the gorge. In this context it is worth noting that research by Ryan etal. (1997) strongly implies that the Black Sea was a freshwater lake until 6700 BP, when there was a rapid influx of salt water as the Mediterranean broke through the Bosporus "dam". If their hypothesis is correct, then all Danube fish exploited during the Mesolithic were freshwater fish, and Neolithic people could not have had access to marine fish until after 6700 BP. This in turn implies that freshwater fish are the source of the very high 815N values recorded in Mesolithic skeletons from the Iron Gates. Existing stable isotope data for aquatic food sources from the Iron Gates are limited to analyses of collagen from three fish bones and an otter bone from Lepenski Vir. These show no consistent pattern, with 813C values varying between -26.3 and -15.7%o and 815N values varying between +8.2 and +12.9%o (.Bonsall et al. 1997). With hindsight, these results may not be unusual. Recent research suggests that 815N and 813C values of freshwater fish can be highly variable (Fig. 2). Moreover, the values can vary quite considerably for different species from the same freshwater system, and for the same species from different freshwater systems. For example, Du- four et al. (1999) report inter-species differences from Lake Constance of ~7%o for 813C and ~8%o for 815N. In the case of the 813C values for the Iron Gates, the spread is almost 11%0. While this spread appears very large, it could be due to a number of factors: ® The number of species in the River Danube could be greater. From the work of Dufour et al. (1999) it can be observed that for each lake under study, the inter-species differences were much greater than the intra-species differences. © The age range of the fish from the study of Dufour et al. (1999) was limited (3-5 years) while those from the Iron Gates could be much greater, as is shown by the evidence of bones from large sturgeon as well as mature carp of extremely large sizes. ® The study by Dufour etal. (1999) was effectively a snap-shot in time, while the samples analyzed from the Iron Gates sites could conceivably span several thousand years and there could have been changes in the freshwater ecosystem within this time-span. ® Most importantly, there could have been a shift to anadromous fish when the Bosporus was breached and the Black Sea became a marine system. The fish specimen with a 813C value of -15.7%o and a 815N value of +12.9%o is certainly not inconsistent with this hypothesis. If a +3-4%o trophic level shift is employed between freshwater fish and human bone collagen (Minaga-wa and Wada 1984), then it would require average 815N values for Danube fish of approximately +10.5%o, or greater, to produce human bone colla- gen values of >+l4%0, which are characteristic of Mesolithic people from the Iron Gates. While this type of data for fish from the Iron Gates is limited, such values are not uncommon. Iacumin etal. (1998) and Pate (1998) report 515N values of about +12%o for Lake Nasser and South Australia fish, respectively. Dufour etal, (1999) report 815N values >+13%o for fish from Lake Geneva and Lake Constance, while Hobson and Welch (1995) report 815N values for large char collected from a high Arctic lake in Canada of >+l4%0. The enrichment of any species will of course depend on the complexity of the food web and its trophic level within the web. Moreover, there is evidence for certain freshwater species that the 815N value increases with the age/size of the fish. This is related to the fact that as a fish grows, it tends to feed at higher trophic levels - in effect, it becomes increasingly carnivorous - and beyond a certain growth stage may start to feed on smaller members of the same species. This phenomenon has been reported for arctic char from Canada (Hobson and Welch 1995). It may also be characteristic of species such as carp, catfish and sturgeon, which dominate fish bone assemblages from the Iron Gates. Many Mesolithic specimens of these species from the Iron Gates sites were very large and, in comparison to their numbers, such fish may have made a disproportionately large contribution to the food supply. 25 20- 4 Freshwater Fish 15- 10- Marine Fish ;T2_...................j Terrestrial Herbivores -5- -40 —rrjTT- -35 -30 -25 -20 I i i i •15 -10 T -5 6I5C (%o) Fig. 2. &3C and S'SN natural variations in freshwater fish (Fl), terrestrial herbivores (Tl) and marine fish from Eurasia Rafter Dufour et al. 1999) compared against North American freshwater fish (F2) and terrestrial herbivores (T2) Rafter Schwarcz 1991) and terrestrial herbivores (T3), river fish (*) and an otter (*) from the Iron Gates fafter Bonsall et al. 1997/ It will be evident from the foregoing discussion and Figure 2 that the North American data on freshwater fish published by Schwarcz (1991) are not necessarily an appropriate model for the Danube, and that locally derived data are to be preferred. While the locally derived data are limited to three fish bone collagen analyses, the isotopic signature for the otter (813C, -19.8%o; 81SN, +10.7%o) effectively provides an average signature for small fish from the Iron Gates, since these will be its primary food source, and arguably, this provides the most reliable model for the Mesolithic human population of the Iron Gates (Fig. 2). An otter's diet comprises primarily small fish, but includes other aquatic animals and small land mammals. Compared to an otter, the bone collagen of a human feeding mainly on much larger fish (and occasionally otters) from the same freshwater ecosystem could be expected to show a slight enrichment in 813C and an enrichment of 3-4%o in 8HN. The data for Mesolithic people from Lepenski Vir, Vlasac and Schela Cladovei are perfectly consistent with this model. While further research is needed into the isotopic composition of aquatic food resources available to Mesolithic and Neolithic peoples in the Iron Gates, new information is available for terrestrial food sources. Collagen values for ungulate bone samples from Mesolithic and Neolithic contexts at Lepenski Vir and Schela Cladovei can be substituted for the North American herbivore data used in the original study (Bonsall et al. 1997). Figure 2 compares the two data sets (T2, T3). Although not all the Iron Gates ungulate samples could be identified to species, probably they derive mainly from deer and cattle. The spread of values on the 813C axis is significantly less for the Iron Gates data set and, while the spread on the 815N axis is similar to the North American sample, the median value (+5.3%o) is significantly higher. Taking into account the various lines of information on the isotopic composition of aquatic and terrestrial food sources, it seems reasonable to continue to use the 815N value as a measure of freshwater versus terrestrial food intake in Iron Gates stone age populations (cf. Cook et al, in press). An end point of +17%o for a 100% aquatic diet is assumed, which is the highest 815N value measured in an adult from the Iron Gates region (Bonsall et al, 1997). For a 100% terrestrial diet a value of +8%o is assumed. This is based on studies by Ogrinc (1999) and Mays (1998) supported by local data for herbivores, as discussed above. Chronology In their original study, Bonsall et al. (1997) published new AMS l4C age measurements on human bones from Lepenski Vir, Vlasac and Schela Cladovei. For all three sites, the human bone ages were older than expected on the basis of existing dates on charcoal - for detailed discussion, see Cook et al. (in press, and forthcoming). From this, it was suggested that the bone collagen of humans who had ingested large quantities of freshwater fish may be depleted in 14C as a consequence of the consumption of material from a reservoir that differed in l4C specific activity from the contemporary atmosphere, thus resulting in l4C ages that are "too old". It was further suggested that this possibility could most easily be tested by comparing radiocarbon age measurements on human bones with those on artefacts of terrestrial animal bone found in the same graves. Schela Cladovei provided material ideal for investigating this problem in the form of bone projectile points, made from long bone splinters of artio-dactyls, found in direct association with skeletons. These were either embedded in human bone or found immediately adjacent to bones of articulated skeletons (which may originally have been embedded in the soft tissue surrounding the bones). In all cases, the bone points may have been the actual cause of death. Cook et al. (in press) obtained AMS 14C dates on paired human and ungulate bone (projectile point) samples. Systematic differences were found between the two sets of ages, demonstrating the existence of a freshwater reservoir effect, and its magnitude was calculated as 540±70 radiocarbon years. From this, and using the 815N value as a measure of the proportion of the diet derived from aquatic foods, it is possible to apply a correction to the human bone ages from all the sites (Cook et al. in press, and forthcoming). The corrected ages are given in Table 3. The effect of the correction is to make the human bone ages significantly younger by approximately 200-500 years depending on the 815N value. However, the reservoir-corrected ages are less precise, i.e. have larger error terms. The expanded data set Bonsall et al. (1997) reported 813C and 815N measurements on 70 individual skeletons - 33 from Lepenski Vir, 29 from Vlasac, and 8 from Schela Cladovei. All the skeletons were those of adults of (supposedly) known chronological context. The Schela Cladovei skeletons were a burial group from below and adjacent to a Mesolithic "house", and seven of them were directly dated by AMS to the Late Mesolithic (Tab. 3). The skeletons from the other two sites had all been recorded as belonging to specific phases of Mesolithic or Early Neolithic occupation -Vlasac I—III ("Mesolithic"), Lepenski Vir I—II ("Mesolithic") and Lepenski Vir III ("Neolithic"). To the original data set can now be added the results from a further 46 skeletons, all from Lepenski and Vlasac. They comprise both adults (25 from Lepen- Site Laboratory Burial 14C Age BP Corrected Calibrated Age ID 14C age BP Range (2o) BC Schela OxA-4384 M52 8570±105 - - OxA-4379 M43 8550±105 8070+122 7450-6645 OxA-4385 M55 8510±105 8090±118 7465-6653 OxA-4382 M49 8490±110 8046+124 7448-6615 OxA-4380 M46 8460±110 8046±123 7448-6640 OxA-4378 M42 8415±100 7971±116 7295-6512 OxA-4381 M48 8400±115 7932+130 7289-6466 OxA-4383 M50 8290±105 7834±120 7061-6439 Vlasac OxA-5824 72 10240+120 9850±130 9949-8843 OxA-5822 51a 8760±110 8380±120 7600-7080 OxA-5827 83 8200+90 7810±105 7049-6441 OxA-5823 54 8170±100 7750±115 7032-6401 OxA-5826 24 8000±100 7600±115 6647-6625 Lepenski Vir OxA-5827 31a 7770±90 7310±108 6404-5926 OxA-5830 44 7590±90 7150±106 6225-5797 OxA-5828 32 7270±90 7040±95 6156-5721 OxA-5831 88 7130±90 6960±93 6011-5644 OxA-5829 35 6910+90 6720±93 5772-5479 Tab. 3. Human bone radiocarbon ages from Lepenski Vir, Vlasac and Schela Cladovei, corrected for the freshwater reservoir effect using method 1 of Cook et al. ^forthcoming). All 14C ages are expressed in conventional radiocarbon years BP (before 1950 AD). The errors are expressed at the one-sigma level of confidence. Calibrated age ranges were determined using CALIB 4.2 fStuiver and Reimer 1993; Stuiver et al. 1998). ski Vir, and 2 from Vlasac) and children (12 from Lepenski Vir, and 7 from Vlasac). The adults are individuals whose chronological context is either unknown or not recorded, or which (in the case of five samples from Lepenski Vir) were assigned to post-Neolithic occupations. These new data are only presented here in graphic form (Figs. 4-7). Full details will be presented in a later publication. All analyses were carried out by the sealed tube combustion method described in Bonsall et al. (1997)• Briefly, this comprises the combustion of small collagen samples in evacuated quartz tubes containing copper oxide as the source of oxygen and a small quantity of silver wire to remove halide contaminants. The CO2 and N2 are then cryogenically separated and analyzed by stable isotope mass spectrometry. The authors consider this to be the most precise and accurate method to determine stable isotope ratios in human bone collagen, and would advise that these and continuous flow measurements should not be combined. In the original study, identification of groups was done primarily by visual inspection of bivariate scat-terplots. In this paper, a variety of exploratory and formal statistical methods, including exploratory cluster analysis, linear discrimination techniques and hypothesis tests have been used to explore groupings and to assess evidence for pre-defined archaeological groups. Statistical analysis was carried out in MINITAB v. 13. DISCUSSION Adults Lepenski Vir The original data set from Lepenski Vir comprised measurements on 33 adults from the various Stone Age occupation phases that were recognized by Srejovic (1969; 1972; Zoffmann 1983) - Proto-Lepen-ski Vir, Lepenski Vir I, II and III. A bivariate scatter-plot of the data (Bonsall et al. 1997; cf. Fig. lc) suggested that there were two groups. One group exhibited 513C and 815N values that were similar to Mesolithic individuals from Schela Cladovei and Vlasac (cf. Figs, la and lb), indicative of diets with a high input of protein from aquatic sources, and the other showed much lower 815N values suggesting diets with increased levels of protein from terrestrial food sources. Cluster analysis of the original data set (Fig. 3) broadly supports these two groupings. The enlarged data set of 58 adults comprises the 33 individuals that had been attributed to Mesolithic and Neolithic contexts, plus 5 individuals "dated" to the Chalcolithic and Medieval periods, and 20 individuals that were not assigned to any occupation "phase". Dietary end points of +8%o and +17.0%o were adopted as representing 100% terrestrial and 100% aquatic diets respectively (Cook etal. in press). Cluster analysis of this revised data set suggests that there are at least three groups, distinguished primarily on the basis of the 815N value (Fig. 4A). The individuals in group 1 have 815N values ranging between +14.4 and +17.0%o, which implies that this group had diets in which 71-100% of the protein was derived from aquatic sources. Group 3 individuals have 515N values between +9.3 and +11.2%o. which implies that the bulk (64-86%) of the protein came from terrestrial sources. Group 2 skeletons have 815N values ranging from +11.8 to +I4.0%o, intermediate between groups 1 and 3, indicating diets in which protein was derived from aquatic and terrestrial sources in similar proportions (42- 67%). The provisionally identified groups (clusters) do not correspond to archaeological (Srejovic) phases and groups contain individuals of diverse age at death and both sexes. However, there appears to be a link between groups and radiocarbon age. Of the five radiocarbon ages currently available (Tab. 3), two lie in group 1 (7310±108 BP, 7150+106 BP), one falls in group 2 (7040+95 BP), and two lie in group 3 (6960+93 BP, 6720+93 BP). The reservoir-corrected ages form a more or less continuous series, and suggest that the three groups relate to different phases in the use of the site. The dendrogram (Fig. 4A) and scatterplot (Fig. 4B) show that sub-groups may exist within the main groups 1 and 3, but their identification is based on only small numbers of individuals and so remains unconfirmed. For example, in group 1 there are seven individuals with very low 813C values relative to 8 15N, and with virtually identical 815N values (Fig. 4B, "sub-group lb"). Of these, five are male/probably male, one is probably female, and one is of indeterminate sex (Roksandic 1999). The female is an elderly individual (>40). Given the age/sex composition of this "sub-group" and the fact that there is at least one child with a similar 813C isotopic signature (see below), it would be difficult to see these as "outsiders" who had married into the Lepenski Vir community (cf. Bonsall et al. 1997). A more likely expla- Similarity 52.23 - 68.15 - 84.08 - 100.00 I ti n h m Vlasac Cluster analysis of the enlarged data set of 35 individuals from Vlasac suggests a division into four groups (Fig. 5A). Groups 1 and 4 are separated on the basis of 515N. Groups 2 and 3 are distinguished on the basis of both 8I3C and S15N (Fig. 5B). 1 2 6 3 4 5 7 8 9 10 1517 11 12 1316)4 18 19 202122 2324 25 262728 2931 32 3330 Observations GROUPS Fig. 3■ Dendrogram grouping 33 skeletons from Lepenski Vir (d. Bonsall et al. 1997^ according to 5l3C and 8I5N values. nation is that they represent a specific phase in the occupation of the site when people had access to aquatic food sources that were relatively depleted in 13C (compared to other group 1 individuals). Variations in the isotopic composition of Danube fish may have occurred through time as a result of natural changes in the freshwater ecosystem. A similar explanation may be proposed for four individuals in group 3 who have unusually enriched 813C relative to 515N values (Fig. 4B, "sub-group 3b"). They evidently consumed larger amounts of S13C-enriched terrestrial (and possibly aquatic) foods compared to other group 3 individuals, which may indicate that they belong to a different phase in the occupation of the site. Theoretically, changes in the natural environment and/or economic practices could have raised average 813C levels of some important food sources. Such changes include (i) an increase in grazing herbivores (including domesticated cattle) at the expense of browsers (deer), as agriculture expanded and woodland cover was reduced; (ii) the introduction of C4 millet (Panicum miliaceum) into the food chain during the Neolithic (or a subsequent increase in its use) either directly as human food or indirectly as grown forage for livestock, and (iii) the appearance of true marine fish (anadro-mous) in the Danube after 6700 BP when the Black Lake was converted into the Black Sea. The suggestion that within their respective groups, "sub-groups" IB and 3B are chronologically distinct is a working hypothesis that requires confirmation from radiocarbon dating. Again, the provisional groups (clusters) do not correspond to archaeological phase (cf. Srejovic and Letica 1978) and groups contain individuals of diverse age at death and both sexes. Of the five radiocarbon ages currently available, four lie in group 1 (9800+108 - 7768+113 BP) and one lies in group 2 (7598±113 BP). The single group 2 age measurement is in trend the youngest and raises the possibility that the two clusters represent different periods in the use of the site. However, to confirm any time relationship between the two groups would require further 14C measurements with improved precision. No 14C age measurements are currently available for groups 3 or 4. Groups 1-3 at Vlasac have 813C and S15N values similar to group 1 at Lepenski Vir, while Vlasac group 4 has §15N values in the range of Lepenski Vir group 2. Vlasac and Lepenski Vir occupy almost identical riverside locations just a few kilometres apart that, presumably, gave access to essentially the same food resources. Therefore, it may be suggested that Vlasac group 4 belongs to the same time-range as Lepenski Vir group 2, and is later than Vlasac groups 1-3 and Lepenski Vir group 1. Children Isotopic analyses are available for 10 children (under 15 years old) from Lepenski Vir. These are compared against the adult ranges in Figure 6. The overall distribution is similar to that of the adults. Six children have 815N values in the range of the group 1 adults, including one with a 813C relative to 815N value reminiscent of the adult "lb sub-group". One child has a 815N value in the range of the group 2 adults, and there are three children whose 815N values are similar to group 3 adults. It is interesting that the first group of children have 815N values that are, on average, l.l%o higher than the corresponding group of adults (+l6.5%o versus +15.4%o). A tendency toward more positive 5I5N values in children compared to adults has been observed in some previous studies (e.g. Pate 1997; Ogrinc 1999). This is usually attributed to the fact that, during infancy, children ingest their mothers' milk and thus, in effect, feed at a higher trophic level. After weaning, it is supposed that bone collagen turnover would result in the progressive loss of this "nursing signal" (cf. Pate 1997). A "nursing effect" is not apparent for the other Lepenski Vir children, perhaps because of the very small number of individuals involved. Nor is it evident among the children from Vlasac (Fig. 7), although the possibility cannot be excluded. Of the seven Vlasac children analysed, five have 813C and 515N values similar to group 1 adults, while another has a 515N value that falls at the bottom of the range for group 2 adults. Assuming a 1.1%0 difference between the average 515N values of adults and children, it is possible that some of the Vlasac children are the offspring of group 4 females, and others are the offspring of group 1 and 2 females. Similarity 43.61 62.41 - 81.20- 100.00 Observations GROUPS B. 18- 16- 1o 14- 12- 10- • Group 1 o Group 2 » Group 3 • • 1b ... 0° 0 o o° 0° A . '' A; A A /A A 3b -22 -21 -20 -19 -18 -17 -16 The remaining child from Vlasac has 515N and 813C values of+12.7%o and -19.8%o, respectively. These are the lowest values recorded for any individual (child or adult) from Vlasac, and fall within the range of the group 2 individuals from Lepenski Vir. This evidence appears to confirm the presence at Vlasac of individuals with "intermediate" diets, and it is not inconceivable that this child was that of a female who had a predominantly terrestrial diet (cf. Lepenski Vir group 3). Because of the possibility of systematic differences between the isotopic signatures of adults and children, it was decided that separate statistical analyses be carried out. Dietary change and the timing of the Mesolithic-Neolithic transition in the Iron Gates The direct AMS 14C age measurements on human bones from Lepenski Vir and Vlasac (Tab. 3) are evi- 5i3c (%o) Fig. 4. A. Dendrogram grouping 58 skeletons from Lepenski Vir according to S'-'C and 5nN values. B. Scatterplot of 5'3C versus 815N for 58 adults from Lepenski Vir, and groupings suggested by cluster analysis. Note the exaggerated scale on the S' s€ axis. dence of human occupation of that part of the Iron Gates gorge from c. 9800-6700 BP. Bone collagen stable isotope analyses indicate fundamental changes in diet during that time range. Humans dated before 7150+106 to 7040+95 BP on the reservoir-corrected time-scale have 815N values of >+14.4%o, indicating diets in which the bulk (>67%) of the protein was derived from Danube fish. This phase is represented by the group 1 individuals at Lepenski Vir and by Vlasac groups 1, 2 and 3. Since there is no evidence for the keeping of domesticated animals (other than dog) prior to that time, it seems reasonable, using this criterion, to describe the pre-7150-7040 BP inhabitants of Lepenski Vir and Vlasac as "Mesolithic". If so, when did the people of the Lepenski Vir-Vlasac section of the Iron Gates become "Neolithic" farmers? A change in diet is evident at Lepenski Vir between 7150-7040 BP. The group 2 individuals from Lepenski Vir show a significant reduction in average 815N values, consistent with an increase in the amount of protein derived from terrestrial food sources and a corresponding decrease in protein from aquatic sources. The same change may be represented at Vlasac by the group 4 adults. At Lepenski Vir this can be seen as the beginning of a trend that culminated in the adoption of a predominantly terrestrial diet by c. 6960 BP, represented by the group 3 individuals. The timing of this dietary change corresponds quite closely with the appearance of Neolithic farmers in the regions surrounding the Iron Gates, represented by the earliest Starcevo-Cris-Koros settlements, and it is reasonable to infer that the two events are connected. There are two hypotheses that can plausibly account for the changes observed at Lepenski Vir (and possibly Vlasac) after 7150+106 BP. The first is that Mesolithic people of the Lepenski Vir-Vlasac area adopted farming more or less as soon as it became available to them, and gradually increased the amount of agricultural products in their diets at the expense of traditional aquatic resources. The second is that the local population did not become farmers immediately, but traded with neighbouring farmers for agricultural products for a period of decades to centuries before eventually taking up livestock raising and cultivation. Similarity 51.60 67.74 83.87 - 100.00 3 9 15 17 5 27201128 6 10 4 1626 1819 1 14121330 2 2531243423 8 2229 7 21333235 Observations i_i i_i i_i i_i 1 GROUPS 17.0 B. 16.5- 16.0- 15.5 J 15.0- tO 14.5- 14.0- 13.5- 13.0- 12.5 x Group 1 ■ Group 2 o Group 3 • Group 4 o o x -20.0 -19.5 -19.0 -18.5 -18.0 813C (%O) Fig. 5- A. Dendrogram grouping 35 skeletons from Vlasac according to ft^C and 8nN values. B. Scatterplot of S*3C versus S'\\for 35 adults from Vlasac, and groupings suggested by cluster analysis. This latter possibility has been suggested by several authors, most notably Voytek and Tringham (1989). On the existing radiocarbon evidence, an "availability phase" (cf. Zvelebil and Rowley-Conwy 1984; 1986) during which Mesolithic people in the Iron Gates increased their intake of terrestrial protein through trade or exchange with farmers, could have lasted between a few decades and approximately 600 years. The same radiocarbon evidence suggests that it would have ended by c. 6800 BP at the latest. Two individuals from Lepenski Vir group 3 with reservoir-corrected radiocarbon ages of 6960+93 BP and 6720+93 BP have S15N values of+10.9%o and +11.2%o, respectively, indicating predominantly (64-68%) terrestrial diets. In the context of the Iron Gates, it is difficult to see how such high levels of terrestrial protein intake could have been sustained without a direct investment in agriculture. 18 \C - 16- 14- - 12 CO 10 • 1b.......... ......•• • * # ha 2........ J ° 3a" .....A •* A (...-•"3b -22 -21 -20 r -19 513C (%O) -18 -17 -16 Fig. 6. 5l3C versus 815N for 10 children from Lepenski Vir, plotted against ranges of adult groupings suggested by cluster analysis. Other explanations could be proposed for the initial reduction in average 8HN values at 7150-7040 BP (cf. Lepenski Vir group 2, Vlasac group 4). They include a long-term increase in the consumption of wild animal and/or plant resources, a reduction in the average size of freshwater fish caught, or even a change in the type of fish caught. However, there is no convincing supporting evidence from either Lepenski Vir or Vlasac. Even if there were, such changes are most unlikely to account for the strongly "terrestrial" isotopic profile of the group 3 individuals from Lepenski Vir. Stable isotopes and the dating of houses at Lepenski Vir Since the publication of the account of the Lepenski Vir excavations (,Srejovic 1969; 1972) there has been considerable controversy over the age and cultural context of the trapezoidal-plan houses of Lepenski Vir I and II. While most authors now accept the radiocarbon ages on charcoal from the houses as valid, there is still disagreement on whether the houses should be interpreted as Mesolithic (cf. Radovanovic 1996) or Neolithic (cf. Ehrich 1974; Milisauskas 1978). The radiocarbon and stable isotope measurements on human bones from Lepenski Vir, discussed in this paper, have a critical bearing on the issue. Charcoal samples from the LV I—II houses produced ]4C ages between 6560 and 7430 BP (Quitta 1972; Boric 1999). These are similar to the reservoir-corre-cted 14C ages on human bone (Tab. 3). If the 14C ages of the charcoal and human bone samples are accepted as being correct, then the houses and the human bones can be considered as belonging to approximately the same time-range. Since the human bones appear to span the transition from a Mesolithic to a Neolithic economy, it would be reasonable to conclude that the same applies to the houses. However, as Cook et al. (in press and forthcoming) have pointed out, the charcoal samples were from long-lived tree species (oak and elm). Such samples can yield l4C ages that are several hundred years older than the archaeological events they purport to date - often referred to as the "old wood" problem. Therefore, it is conceivable that the houses are significantly younger than the radiocarbon ages of the charcoal samples, and all the houses post-date the change in diet between 7150 and 7040 BP, i.e. they belong to the time-range of the group 2 and 3 humans. This would be consistent with the presence of Starcevo pottery in several of the houses that were dated (Budja l999.Fig. 7). These are house 54 (7l6l±56 BP - weighted mean of five 14C measurements), house 1 (6860+100 BP) and house 16 (6820+100 BP). Given the uncertainties over the interpretation of the charcoal-based radiocarbon ages, the question may be asked: can it be shown that any of the houses at Lepenski Vir belongs to the period of the group 1 ("Mesolithic") humans? It was suggested above that the division of the human remains from Lepenski Vir into three groups, according to the 813C and 815N values, represents a time-series. For convenience, this may be characterised as: Period 1 ("Mesolithic") dating before 7040± 95 BP and comprising individuals with 815N values of —+l4.4%o. Period 2 ("transitional Mesolithic-Neo-lithic") dating c. 7040+95 BP and represented by individuals with SHN values ranging between +11.8 to +I4.0%o. Period 3 ("Neolithic") dating after 7040± 95 BP and comprising individuals with S15N values of <+11.2%o. If the "phasing" of the human remains based on 515N is reliable, where there is a clear stratigraphic relationship between a human skeleton and a house, it follows that the bone collagen 515N value can be used to infer the age of the house. However, this would only apply in the case of articulated skeletons. Many of the "skeletons" uncovered in the Lepenski Vir excavations appear to be groups of disarticulated bones. These could represent delayed or secondary burials of individuals who had died some time previously, and hence there could be a significant "age offset" between the time of death of the individual and the time of final burial. Few details of the stratigraphic relationships of burials and houses are provided in published accounts of the Lepenski Vir, but there is a limited amount of photographic evidence that can be considered. Published photographs of house 21 (Srejovic 1969. PL 69; Radovanovic 1996.Fig. 4.3) show the articulated skeleton of an adult female lying below the floor of the house. It is clear from other photographs published by Radovanovic (1996.Figs 3-14, 3-31, 3-32) that the burial was inserted through the plaster floor. This relationship indicates that the burial must have been emplaced after the plaster floor was laid, and therefore (presumably) post-dates construction of the house. The skeleton (7b or 7/1) has a 815N value of +15.8%o, placing it firmly within Period 1 ("Mesolithic"). Unless this is a case of delayed burial following excarnation, which seems highly improbable, the evidence implies that house 21 is also Mesolithic. According to Srejovic {1969; 1972; Srejovic and Ba-bovic 1983) and Radovanovic (1996) house 21 is superimposed upon houses 22, 29 and 30 (Fig. 8), which therefore places them also in period 1. Thus, there are at least four houses at Lepenski Vir that, on the combined evidence of stratigraphy and bone collagen isotopic data, can be argued to pre-date the change in diet between 7150 and 7040 BP. 17.0- 16.5- 16.0- 15.5- ^_s o a? 15.0- z VI 14.5- CO 14.0- 13.5- 13.0- 12.5- 3 1 f • • • 4 '''■•■... • -20.0 -19.5 -19.0 Sl3C (%o) -18.5 Fig. 7. 5I3C versus 8'5N for 7 children from Vlasac, plotted against ranges of adult groupings suggested by cluster analysis. It is significant that none of the four probable period 1 houses discussed appears to have been associated with Starcevo pottery (cf. Budja 1999.Fig. 7). This would be consistent with a Mesolithic context and an age prior to 7040 BP. The photograph in Srejovic (1969.Pl. 67; Srejovic 1972.PI, 58) shows the articulated skeleton of an adult female (54e) lying directly above stone slabs apparently set into the floor of house 65 (Fig. 8). This skeleton has a 815N value of +13.2%o and belongs to period 2, post-dating the dietary change at 7150-7040 BP. Leaving aside the question of whether the corpse was deliberately left exposed on the floor of the house, or placed beneath a cairn (a possibility suggested by other photographic evidence), or buried in a grave pit dug from a higher level, the position of the skeleton with respect to the floor suggests that the house is older and could belong to either period 1 or period 2. A period 2 (post-7150-7040 BP) age would be consistent with the presence of pottery inside the house (cf. Budja 1999-Fig. 7). It was suggested by Srejovic that house 65 contained an earlier burial (54d), which is represented by disarticulated bones around skeleton 54e (Srejovic 1969.Pl. 67; Srejovic 1972.Pl. 58). One of these bones gave a 815N value of +15.3%o suggesting an age prior to 7150-7040 BP. It was claimed that the bones of 54d had been disturbed by burial 54e. However, given the disarticulated nature of 54d and the fact that the bones may be from more than one individual, they are more plausibly interpreted as a secondary burial and cannot therefore be used to "date" house 65. The stable isotope data also have implications for Radovanovic's architectural phasing of the Lepenski Vir houses (Radovanovic 1996). As noted above, there is good evidence that houses 21, 22, 29 and 30 are Mesolithic and predate 7150-7040 BP. In Radovanovic's scheme, houses 21 and 22 are assigned to phase 1.2, and houses 29 and 30 to phase 1.1 (Fig. 8). If her phasing of the houses were correct, one would have to conclude that the shift away from a traditional Mesolithic diet began either during or after phase 1.2, and that all houses assigned to phase 1.1 are Mesolithic. However, this interpretation is inconsistent with radiocarbon evidence from three other houses attributed to phase 1.1 (houses 1, 9 and 37) suggesting -18.0 65601100 BP 66 (phase 1.3) under 66201100 BP XXXII, 6859164 BP 7430+160 BP 6820+100 BP 6860+100 BP 7335171 BP 6845+100 BP 6970+60 BP uncertain Danube ^,66301100 BP IX ^ 72101200 BP Fig. 8. Lepenski Vir I site plan showing architectural phases and radiocarbon ages of houses based on charcoal samples (after Radovanovic 1996, with acknowledgement to Boric 1999 and Bailey 2000/ The UC ages of houses 36, 37 and 54 are weighted means of several measurements. The locations of two Lepenski Vir II houses (IX and XXXII) and corresponding 14C ages are also shown. that they post-date the dietary change at 7150-7040 BP (Fig. 8). The use of stable isotope data as a proxy dating tool may also contribute to a better understanding of the evidence from Vlasac. Five charcoal samples from Srejovic and Letica's (1978) phase lb gave 14C ages of 6805-7000 BP. These ages were rejected because they were out of sequence with radiocarbon determinations for the succeeding phases II and III, and because they were not in accord with the excavators' belief that the contexts dated were Mesolithic. However, in European archaeology radiocarbon ages have often proved more reliable than chronologies derived from archaeological observations. The ages for "Vlasac lb" are consistent with the presence of Early Neolithic (Starcevo) remains on the site. These have always been considered a very minor component of the archaeological record. However, as noted above, stable isotope evidence indicates that a significant proportion of the humans buried at Vlasac - the group 4 adults, representing 23% of the samples analyzed - had diets similar to the group 2 adults from Lepenski Vir, and may therefore belong to the same time-range of c. 7040±95 BP. CONCLUSIONS Reappraisal of a larger stable isotope data set for Lepenski Vir and Vlasac demonstrates a shift from a Mesolithic-type dietary regime, based largely on aquatic resources, through an intermediate phase, to one based largely on terrestrial resources that probably included a major agricultural component. Radiocarbon evidence suggests that the transition centred around 7040±95 BP (6156-5721 cal BC), and that agriculture was being practised in the Lepenski Vir-Vlasac area by 6800 BP (c. 5700 cal BC). The stable isotope and radiocarbon data coupled with evidence of the stratigraphic relationships between burials and houses suggest that the trape- zoidal plan houses of "LV I—II" span the time-range of the dietary change. This is contrary to previous interpretations of the houses as either exclusively Mesolithic or exclusively Neolithic. Moreover, the stable isotope evidence suggests that Lepenski Vir (and possibly Vlasac) was occupied continuously from the Mesolithic into the Early Neolithic. For both Lepenski Vir and Vlasac, this paper has also highlighted apparent conflicts between archaeological sequences and radiocarbon based chronologies supported by stable isotope analyses. Future research must be directed toward resolving the issues that have been raised here. REFERENCES AMBROSE S. H. 1993- Isotopic analysis of paleodiets: methodological and interpretive considerations. In Sandford M. K. (ed.), Investigations of Ancient Human Tissue: Chemical Analyses in Anthropology: 59-130. BAILEY D. W. 2000. Balkan Prehistory. London: Routledge. BONSALL C., LENNON R., McSWEENEY K., STEWART C., HARKNESS D., BORONEANT V., BARTOSIEWICZ L, PAYTON R. and CHAPMAN J. 1997. Mesolithic and Early Neolithic in the Iron Gates. A palaeodietary perspective, fournal of European Archaeology 5: 50-92. BORIC D. 1999- Places that created time in the Danube Gorges and beyond, c. 9000-5500 BC. Docu-menta Praehistorica XXVI, 6"' Neolithic Studies: 41-70. BOKONYI S. 1969. A Lepenski Vir-i oskori telep ge-rinces faunaja (The vertebrate fauna of the prehistoric settlement at Lepenski Vir). Archeologiai Erte-sfto 96:157-160. 1970. Animal remains from Lepenski Vir. Science 167, 3926: 1702-1704. 1977. Vlasac und die Fragen der mesolithischen Domestikation. Mitteilungen des Archaologischen Instituts der Ungarischen Akademie der Wissen-schaften 7: 85-92. 1992. Animal remains of Mihajlovac-Knjepiste, an Early Neolithic settlement of the Iron Gate Gorge. (Hommage a Nikola Tasic a l'occasion de ses soix-ante ans), Balcanica XXIII: 77-87. BUDJA M. 1999. The transition to farming in Mediterranean Europe - an indigenous response. Docu- menta Praehistorica XXVI, 6th Neolithic Studies: 119-141. COOK G. T, BONSALL C., HEDGES R. E. M., McSWEENEY K., BORONEANT V., PETTITT P. B. In press. A freshwater diet-derived 14C reservoir effect at the Stone Age sites in the Iron Gates gorge. Radiocarbon 43: accepted for publication. COOK G. T, BONSALL C., HEDGES R. E. M., McSWEENEY K, BORONEANT V, BARTOSIEWICZ L. and PETTITT P. B. Forthcoming. Problems of dating human bones from the Iron Gates. Submitted to Antiquity. DUFOUR E„ BOCHERENS H. and MARIOTTI A. 1999. Palaeodietary implications of isotopic variability in Eurasian lacustrine fish Journal of Archaeological Science 26: 617-627. EHRICH R. W. 1974. Book review of Hunters, Fishers and Farmers of Eastern Europe 6000-3000 BC, by Ruth Tringham. American fournal of Archaeology 78: 435-436. HEDGES R. E. M., PETTITT P. B, BRONK RAMSEY C. and VAN KLINKEN G. J. 1998. Comment (Lab.), p. 440. In Radiocarbon dates from the Oxford AMS System: Archaeometry datelist 26. Archaeometry 40(2): 437-455. HOBSON K. A. and WELCH H. E. 1995. Cannibalism and trophic structure in a high Arctic lake: insights from stable-isotope analysis. Canadian fournal of Fisheries and Aquatic Science 52: 1195-1201. IACUMIN P., BOCHERENS H., CHAIX L. and MARIOTH A. 1998. Stable carbon and nitrogen isotopes as dietary indicators of ancient Nubian populations (Northern Sudan), fournal of Archaeological Science 25: 293-301. ' KLEPPINGER L. L. 1984. Nutritional assessment from bone. Annual Review of Anthropology 13: 79-96. MAYS S. 1998. The Archaeology of Human Bones. London: Routledge. MILISAUSKAS S. 1978. European Prehistory. New York: Academic Press. MINAGWA M. and WADA E. 1984. Stepwise enrichment of 15N along food chains: further evidence and the relation between 815N and animal age. Geochi-mica et Cosmochimica Acta 48:1135-1140. OGRINC N. 1999. Stable isotope evidence of the diet of the Neolithic population of Slovenia - a case study: Adjovska jama. Documenta Praehistorica XXVI, 6th Neolithic Studies: 193-200. PATE F. D. 1997. Bone chemistry and paleodiet: reconstructing prehistoric subsistence-settlement systems in Australia, fournal of Anthropological Archaeology 16: 103-120. 1998. Bone collagen stable nitrogen and carbon isotopes as indicators of past human diet and landscape use in southeastern South Australia. Australian Archaeology 46: 23-29. QUITTA H. 1972. The dating of radiocarbon samples. In D. Srejovic (ed.), Europe's First Monumental Sculpture. New Discoveries at Lepenski Vir: 205-210. RADOVANOVIC I. 1996. The Iron Gates Mesolithic. Ann Arbor: International Monographs in Prehistory, Archaeological Series 11. ROKSANDIC M. 1999. Transition from Mesolithic to Neolithic in the Iron Gates Gorge: Physical Anthropology Perspective. PhD Thesis, Simon Fraser University. RYAN W. B. F, PITMAN W. C., MAJOR C. 0, KAZIMIE-RAS S„ MOSKALENKO V., JONES G. A., DIMITROV P., GORUR N., SAKINg M. and YUCE H. 1997. An abrupt drowning of the Black Sea shelf. Marine Geology 138: 119-126. SCHULTING R. J. 1999- Slighting the sea: stable isotope evidence for the transition to farming in northwestern Europe. Documenta Praehistorica XXV, 5th Neolithic Studies: 203-218. SCHWARCZ H. P. 1991- Some theoretical aspects of paleodiet studies Journal of Archaeological Science 18: 261-275. SREJOVIC D. 1969. Lepenski Vir - Novapraistorij-ska kultura u Podunavlju. Beograd: Srpska knjizev-na zadruga. 1972. Europe's First Monumental Sculpture. Neiv Discoveries at Lepenski Vir. London: Thames and Hudson. SREJOVIC D. and BABOVIC L. 1983- Umetnost Le-penskog Vira. Belgrade: Yugoslavija. SREJOVIC D. and LETICA Z. 1978. Vlasac. Mezolit-sko naselje u Djerdapu (Vlasac. A Mesolithic Settlement in the Iron Gates). Beograd: Serbian Academy of Sciences and Arts Monograph DXII. STUIVER M. and REIMER P. J. 1993. Extended 14C database and revised CALIB radiocarbon calibration program. Radiocarbon 35: 215-230. STUIVER M, REIMER P. J., BARD E., BECK J. W, BURR G. S., HUGHEN K. A., KROMER B, McCORMAC F. G, VAN DER PLICHT J. and SPURK M. 1998. INTCAL98 radiocarbon age calibration, 24,000-0 cal. BP. Radiocarbon 40: 1041-1083- VOYTEK B. A. and TRINGHAM R. 1989- Rethinking the Mesolithic: the case of southeast Europe. In C. Bonsall (ed.), The Mesolithic in Europe, 492-499. ZOFFMANN Z. K. 1983- Prehistorical skeletal remains from Lepenski Vir (Iron Gate, Yugoslavia). Homo 34: 129-148. ZVELEBIL M. and ROWLEY-CONWY P. 1984. Transition to farming in northern Europe: a hunter-gath-erer perspective. Norwegian Archaeological Review 17: 104-128. 1986. Foragers and farmers in Atlantic Europe. In M. Zvelebil (ed.), Hunters in Transition: 67-93■ UDK: 903(495)"633/634' Documenta Praehistorica XXVII The Mesolithic/Neolithic Transition in Greece as Evidenced by the Data at Theopetra Cave in Thessaly Nina Kyparissi-Apostolika Ephorate of Paleoanthropology and Speleology, Ministry of Culture, Athens, Greece nkyparissi @ mailbox.gr ABSTRACT - The Mesolithic period was sparsely documented in some littoral sites in Greece until 1992, when it was first testified in Theopetra Cave, western Thessaly, in central Greece. Excavation data in Theopetra indicate a very normal and natural transition from the Mesolithic to the Neolithic, which is documented by (a) the presence of unbaked masses of clay in the Mesolithic layers and a few atypical and monochrome sherds found at the same layers could show the very initial samples ofpottery technology, (b) The presence of domesticated sheep and goat in the Mesolithic zoo-archaeological material, and a good percentage of wild fauna in the Neolithic deposit, indicating a rather similar way of life in both periods; (c) the presence of cultivated hulled six-row barley and of wild eincorn, both testifying farming as a local development and not as knowledge that came from the Near East to Greece, (d) The biological homogeneity which derives from the DNA analysis of human bones of successive periods. IZVLECEK - Vse do leta 1992je bilo mezolitsko obdobje v Grciji redko dokumentirano v nekaj obreznih najdiscih. Takratpa smo ga prvic dokazali v jarni Theopetra, ki lezi v zahodni Tesaliji v osrednji Grciji. Izkopavanja v Theopetri kazejo, da je bilprehod iz mezolitika v neolitik zelo normalen in nara ven. To do-kazujejo: (a) navzocnost nepecenih kepgline v mezolitskih plasteh in nekaj netipicnih in enobarmih cre-pinj, ki smo jih nasli v istih plasteh, hi lahko kazalo na prve zacetke izdelovanja keramike; (b) navzocnost udomacene ovce in koze v mezolitskem zoo-arheoloskem materialu in precejsen odstotek divje fav-ne v neolitskih plasteh, kar kaze na dokajpodoben nacin zivljenja v obeli obdobjih; (c) navzocnost kulti-viranega sestvrstnega jecmena in divjega eincorn zita, kar oboje prica, da seje kmetovanje razvilo lokal-no in da znanje o tern v Grcijo niprislo izBliznjega vzhoda. (d) Bioloska homogenost, ki izhaja iz DNA analiz cloveskih kosti, iz naslednjih obdobij. KEY WORDS - Theopetra, Thessaly, Mesolithic, transition to farming. INTRODUCTION The Mesolithic is the least investigated archaeological period in Greece. Two cave sites with Mesolithic finds (Zaimis and Ulbrich in Attica and the Pelopon-nese respectively) were excavated in 1920 by A. Markovits (1933), but until thel950's the Neolithic was considered almost the back end of Prehistory in Greece, as the two above caves were never included in discussions concerning the Mesolithic in Greece (Galanidou in press). After Milojcic's Palaeolithic finds in the banks of the Peneios River in Thessaly in the same decade (Miloj-cic et al. 1965), the demand for a Mesolithic background that would complete the prehistoric chain in Greece had become a permanent claim of the inves- tigations that followed during the 60's and the 70's. The deepest Neolithic layers, where sparse sherds were found, were then named "aceramic" in the hope that these would lead to the even earlier period, the Mesolithic, which was expected to be found with the characteristics of a trapezoid and microlithic technique known already from Europe. At the same time (1964), a littoral Mesolithic site, Sidari, on the island of Corfu yielded indeed microlithic finds (Sor-dims 1969). Then, in the 70's, at Franchthi Cave, near the eastern coast of the Peloponnese, one more site with a Mesolithic deposit was attested (Jacobsen 1976), and so the picture of the prehistoric record in Greece led to the view that the Mesolithic was entirely a marine story. Thessaly was, of course, exclu- tied from the map of the Mesolithic (Perles 1988; 1989; 1994; Runnels 1988; 1993)• When the Mesolithic was found in Theopetra Cave as an intermediate deposit between the end of the Palaeolithic and the beginning of the Neolithic, it was first confronted with scepticism by many scholars, as it suddenly overturned the theory that made Thessaly deserted at the beginning of the Holocene, a model that was reinforced by a recent survey of Thessaly in 1987 carried out by the American School of Classical Studies at Athens under C. Runnels {Runnels 1988), according to which caves were uninhabited during the Palaeolithic in Thessaly, and there was a gap of some 23 Kyr in the area before the Neolithic. The presence of the Mesolithic period in Theopetra confirmed the belief of some Greek prehistori-ans (Theocharis 1967; Kotsakis 1992), who had laid down the idea of inseparable continuity between the Pleistocene and Holocene in Greece. 1 must say here that, after Theopetra, two or three more Mesolithic sites were found in Greece, one in a cave on the island of Yioura in Sporades (Sampson 1996a), one in a cave at the gorge of Klissoura in mainland Argolid, the east Peloponnese, not too far from the coast (.Koumouzelis et al. 1996), while one more at the Cycladic island of Kythnos (Sampson 1996b) is faced as Mesolithic by the excavator without 14C dates yet. The Mesolithic in Yioura was found in sequence with the Neolithic but without Pala- eolithic background, while in Klissoura the Mesolithic consists the roof of the Pleistocene deposits. In Franchthi and in Theopetra the Mesolithic is found between the Upper Palaeolithic and the Neolithic, but in Theopetra additionally there is a more complete Pleistocene sequence including Middle Palaeolithic assemblages, being for the moment the only site in Greece with all this sequence of deposits (Kyparissi-Apostolika 1999a; 1999b; in press). THE CASE OF THEOPETRA Location of the site The cave of Theopetra is located on the north side of a limestone formation on the right of the road leading from Trikala to Kalambaka (Prefecture of Trikala, west Thessaly) and 3 km outside the latter (Figs. 1, 2). Its altitude is about 60 m above the plain and 300 m above the sea level. It lies between the edge of the Thessalian plain and the foothills of the east Pindus Mountains, being the natural border between Thessaly and Epirus. It has a roughly quadrilateral shape and measures somewhat less than 500 m2. The entrance is large (17x3 m) and arched, oriented towards to the Byzantine monasteries of Meteora in Kalambaka. It is the westernmost prehistoric settlement of Thessaly (Papathanassopoulos 1996.Map 9, No. 406). The excavation data The Mesolithic deposit in Theopetra is recognised as a distinct yellowish-brown (Munsell 10YR 3/4-4/4) humid sediment, interrupted partly by fire remains, and it possibly reflects a rather humid climate. This deposit was not found in all the area of the cave and was absent from the central area, where large-scale erosion took place due to the invasion of large volumes of water from carstic aquifers which repeatedly filled the cave, eroding a huge amount of the autochthonous sediment. The above described yellowish-brown sediment lies at the surface of a consolidated sediment reflecting the last glaciation, and contains the end of the Palaeolithic after the last glaciation at its deepest deposition, and the Meso- Fig. 1. Relief map of Thessaly with the site of Theopetra indicated. lithic at the rest, that is to say, the end of the Pleistocene and the beginning of the Holocene1. At first glance, this sediment seems homogeneous, reflecting a normal climatic sequence. However, it was interrupted by a harsh climatic episode, the so-called younger Dryas, which affected a burnt layer dated to 11 500 BP. After this episode, the Mesolithic in Theopetra begins (Karkanas 2000; Karkanas in press). A good number of 14C dates so far fall into the Mesolithic period, with the oldest boundary touching the 9721± 390 BP (DEM 142) and the youngest the 7995+73 BP (DEM 360) (9940-8550/7060-6780 BC) (Fig. 3), cove- FJfJd View ring about 1700-2000 years (Faco-rellis and Maniatis in press). Among them there is the date of a Mesolithic skeleton dated from the bones to 8070+60BP (7050-7010 BC) (CAMS 21773, Fig. 3) being a clear Upper Mesolithic date. The skeleton belongs to a young woman, 18-20 years old, buried in a semiflexed position in a shallow pit (Fig. 4). The head looked straight in front and was at a higher level than the body, which was turned to the right of the dead, looking at the entrance of the cave. No morphological indications of pathological alterations were found macroscopically, while from the cranial radiography arose mild porotic hyperostosis, possibly as a result of iron depletion, but no acute anaemia, which could lead to death. The dental examination shows a healthy individual (Stravopodi etal. 1999). Four flint implements were found beside the skeleton, but it is difficult to assess whether they were intentionally placed with the body or were coinciden-tally part of the infill of the pit. Charcoal selected from the immediate vicinity of the skeleton gave two 14C dates, 9274+75 and 9348±84 BP (8610-8340 and 8740-8470 BC) (DEM 315,316), while the dating of the bones already mentioned puts the buried individual about 13 centuries later, so the selected charcoal obviously belongs to the infill of the pit, which was dug into an older Mesolithic deposit. The finds of the Mesolithic deposit in Theopetra (li-thic industry, bones, carbon) are dense compared to those from the last glaciation deposit, but sparse compared to the overlying Neolithic finds. of the limestone balk where the cave of Theopetra is The lithic industry relied heavily on the locally available radiolarite, while local materials were also employed. There are large numbers of flakes, retouched forms including truncations, notches and flakes with alternate retouch, but no bladelets, no backed bla-delets, no geometric microliths and possibly no practice of the microburin technique (Adam 1999). It must be stressed here that the analysis of the Mesolithic industry is in progress, the case though doesn't seem to be that of a typical Mesolithic one as it is known from Europe. It appears to be closer to the lithic phase VII of Franchthi (Perles 1990; 1999), showing dissimilarities from phase VIII (the reappearance of backed and truncated elements) (Adam 1999). From the study of the Neolithic industry (Skonrto-poulou in press) there arises a clear continuity with the Mesolithic: (a) in the use of the same raw material, chocolate thessalic radiolarian flint, its sources found in the Pindus Mountains or as a secondary material (pebbles) found in alluvial deposits; (b) in the technology of the implements - the Neolithic is to a high degree a flake industry, as is the Mesolithic, and the cores show similarities. The technological continuity is also observed in the types of implements (a high percentage of notches and denticu-lates worked in flakes in dimensions bigger than the usual). Despite the aforementioned disturbance and erosion of the deposit in Theopetra, which does not allow any observations that derive usually from normal stratigraphy, it is obvious that elements known 1 In Theopetra and in the neighbouring Pindus Mountains is the southernmost point of Europe where the last glaciation is attested with characteristics of such severe climatic conditions (Bailey and Gamble 1991; Karkanas 2000). Fig. 3■ list of radiocarbon dates concerning the Mesolithic and the Mesolithic/Neolithic transition. Lab code Pit Depth Age Calendar Standard (yr BP) age deviation DEM- -918 A8, Pass 4, Brown layer 0.24-0.37 m 7901±29 6818-6663 BC 7028-6649 BC 1a 2c DEM- -360 H6, Pass 10, Layer B 1.15-1.23 m 7995±73 7060-6780 BC 7080-6660 BC 1a 2a DEM- -583 111, Pass 3, Western region Neolithic-Mesolithic boundary 0.57 m 8014±49 7060-6829 BC 7073-6706 BC 1a 2a DEM- -576 111, Pass 3 Neolithic-Mesolithic boundary 0.77 m 8060±32 7078-6866 BC 7137-6829 BC 1a 2a CAMS-21733 H6, Human skeleton, 0.30 m 8070±60 7180-6830 BC 1a burial in situ 7300-6770 BC 2a DEM- -120 110 Eastern region 1.04-1.17 m 8524±57 7590-7540 BC 7650-7480 BC 1a 2 a DEM- -578 111, Pass 8, South-eastern region 1.37 m 8547±71 7650-7520 BC 7750-7480 BC 1a 2a DEM- -587 111, Layer 1, Central region 0.60-0.71 m 8558±37 7599-7549 BC 7641-7529 BC 1a 2a DEM- -125 110, Western region 1.56 m 8673±76 7780-7590 BC 7950-7580 BC 1a 2 a DEM- -589 111, Pass 6 1.18 m 8863±119 8210-7830 BC 8270-7650 BC 1a 2a DEM- -207 T9 2.13 m 9093±550 9120-7590 BC 10130-6820 BC 1a 2 a DEM- -590 111, Pass 7 1.27 m 9150±112 8530-8270 BC 8720-7970 BC 1a 2 a DEM- -586 111, Layer 2 0.80 m 9188±86 8520-8290 BC 8620-8250 BC 1a 2a DEM- -315 H6, Layer B, On human skeleton 0.73 m 9274±75 8610-8340 BC 8720-8290 BC 1a 2 a DEM- -316 H6, Layer B, On human skeleton 0.73 m 9348±84 8740-8470 BC 9090-8300 BC 1a 2a DEM- -577 111 1.37 m 9370±93 8780-8470 BC 9110-8300 BC 1a 2a DEM- -588 111 1.23 m 9461±129 9120-8610 BC 9220-8350 BC 1a 2a DEM- -142 110, Eastern region 1.17 m 9721±390 9940-8550 BC 10690-8210 BC 1a 2a from earlier industries are present in the Neolithic material in general. One of the most important finds of the Mesolithic deposit in Theopetra is the presence of some unbaked clay masses, as well as of some monochrome atypical sherds within them. Their technological characteristics (the shaping and finishing of the surface) are primitive compared to the Early Neolithic examples, while in some cases they are slightly baked. At the beginning, we regarded as intrusive from the overlying Neolithic deposit, but as they continued to be found down to 45 cm below the Neolithic deposit, in parallel with unbaked masses of clay, and given the fact that the chromatic and component sedimentation of the Mesolithic deposit is absolutely distinct from that of the Neolithic, any intrusion from the one deposit to the other could be safely perceptible. My assessment is that they represent a very early pottery at the boundary of the Mesolithic/ Neolithic transition (Kyparissi-Apostolika in press). Additionally, the presence in the Neolithic deposit of pottery characterized in the bibliography (Theocha-ris 1967.127-143) as "primitive painted pottery" belonging to the initial Early Neolihic before Sesklo, reinforces the possibility that we have here the very beginning of pottery technology in Greece. TL dating would clarify the period to which these sherds petra (drawn by M. Deilaki). belong. The presence of unbaked clay is also noticed in the late Upper Palaeolithic deposits of Theopetra, while some cylindrical clay objects up to 7-8 cm long and 2-3 cm thick are also observed in the same layers, coming though from Mesolithic layers which eroded the Palaeolithic ones (Kyparissi-Apostolika 1999 in press and Karkanas in press). The presence of clay, either in unbaked condition or as sherds, is referred to also from other Pre-Neolithic sites in Greece (Theocharis 1958; Perles 1999; Vitel-li 1993) and in Europe (Baku 1978; Vandiver etal, 1990), and so the beginning of pottery technology must be re-examined for its starting boundaries. The archaeo-botanical material of Theopetra cave, the oldest ever recovered from excavations in Greece, plays a key role in the study of the Mesolithic, as well as in arguments concerning the Neolithization of Greece. The presence of cultivated Hordeum vidgare ssp. exastichum (hulled six-row barley) and of Tri-ticum boeticum (wild eincorn), as well as of some wild legumes such as Lens sp. (lentil), Vicia ervilia and so on in the Mesolithic material of Theopetra is definite proof that farming must have started here. The same plants are found as cultivated species in the Neolithic material among other kinds of crops and fruits. The absence of the above plants from the Greek data in the past was one of the strongest arguments in the discussion that sought to regard farming as the result of Near Eastern population movements. Of course, we would not deny that people who were familiar with farming could have come from the Near East and possibly spread this knowledge to some population on Greek territory as well, but this territory, after Theopetra's finds, does not seem to be Thessaly. According to palynological investigations (Bottema 1979), at the beginning of the Mesolithic the steppe vegetation was replaced by forest vegetation. This fits with the finds of the anthracological analysis in Theopetra (Ntinou in press) (the palynological analysis is not ready to give information yet), according to which deciduous species like litmus sp., Prunus sp., Quercus and Fraxinus sp. are present in the Mesolithic material, probably related to the development of forests in the early Holocene. The picture fits also with the fauna species present in the Mesolithic of Theopetra, which are woodland species (wild boar and wild cat - Felis sylvestris) (Rowley-Conwy and Neivton in press). The large carnivores that were present in the Palaeolithic strata disappear in the Mesolithic. The Mesolithic assemblage, however, is dominated up to 40% by small ovi-caprids, indistinguishable from domesticated sheep and goat, and it is possible that early Holocene hunter-gatherers acquired domesticated ovi-caprids from neighbouring groups with herd animals (Newton in press and forthcoming). The same question arises with the fauna of the Mesolithic strata in Yioura Cave, where ovi-caprids also seem to be domesticated (Trantalidou in press). A rather high percentage, around 11% of the Neolithic material in Theopetra belongs to the wild fauna, and some of the species (wild cat, red deer, wild boar and hare) feature also in the Mesolithic. This perhaps indicates a certain continuity of habitat type between the two periods. From the domesticated fauna of the Neolithic material, over 70% are ovicaprids, of which 70% are sheep, while pig and bovids are present with 8% each (Hamilakis in press). Last, but not least, I would like to emphasize the biological homogeneity which derives from the DNA analysis of the Palaeolithic, Mesolithic and Neolithic human bones of Theopetra, and hence the possibility that the genetic gradient in Europe may not have originated from the Near East with the spread of farming, but may have been in place as early as the Upper Palaeolithic at least (Evison et al, in press). CONCLUSION Theopetra Cave, at the western end of the Thessa-lian plain and equidistant from the Pindus Mountains and the plain, is a hidden site not accidentally perceptible by travelers, as could be maintained for littoral cave-sites like Franchthi or the cave of Cy-clope in Yioura. Hence, the hypothesis that it could be used for a while and then be abandoned for some thousand years and again be found by chance later and so on, in my opinion, must be excluded. Within this perspective, it is likely possible that once it was discovered, it was used by the same population and their descendants, who either used it as a permanent base or periodically, the kind of use changing according to the reasons that ruled the one or the other way. As no vegetation changes that could reflect climatic changes are referred to in the palynoligical investigations (Bottema 1979) for the end of the Mesolithic and the beginning of the Neolithic, the transition from the one period to the other should be seen (a) in the establishment of pottery technology, (b) in farming knowledge, (c) in the faunal and floral sequence and, (d) in the biological continuity or discontinuity of the population. As we have seen from the data of Theopetra Cave, all the above parameters lead to the hypothesis that they are the result of long-term attempts starting from the Palaeolithic onwards, and not knowledge that was taught to the population of Thessaly by people from the Near East. If we did yet not have proofs that Thessaly was populated before the Neo- lithic, as was estimated in previous decades, the model of exogenous new experience would work well. However, with the new data from Theopetra Cave it would be beyond any common sense to accept that the Palaeolithic and the Mesolithic people who lived in Theopetra Cave and exploited the environment for their survival had not observed nature and the properties which plants and soil and all other natural elements have, namely the same observations that led the population of the Near East to knowledge of farming. And it is possible that this is why, in western Thessaly (where Theopetra is located), there are the oldest Early Neolithic settlements, such as Prodromos and Magoulitsa (Chourmouziadis 1971; 1972; Papadopoulou 1958, respectively), while during later periods they spread to eastern Thessaly, closer to the coasts of the Aegean (i.e. Dimini, Pev-kakia) (Halstead 1980). These first settlements were probably established by the descendants of Theope-tra's population, and some of these people continued to live in Theopetra during the Neolithic also. In my opinion, the poor presence of the Mesolithic in the Greek peninsula is the result of wrongly directed research, as it has been oriented to European and Anatolian models for the settlement pattern and the lithic industry typology. It is now time to turn to different models harmonised to the Greek environmental and climatic conditions of that period, which, I believe, will lead us to more Mesolithic installations. I believe that further research on the material of Theopetra, as well as new excavations that will follow, will prove definitively a model of indigenous Neolithic civilisation in Thessaly, a picture that derives from the excavation in Theopetra Cave. REFERENCES ADAM E. 1999- Preliminary presentation of the Upper Palaeolithic and Mesolithic stone industries of Theopetra Cave, western Thessaly. In G. Bailey, E. Adam, E. Panagopoulou, C. Perles and K. Zachos (eds.), The Palaeolithic Archaeology of Greece and Adjacent Areas, Proceedings of the ICOPAG Conference, British School at Athens Studies 3:266-270. BAHN P. 1978. Palaeolithic pottery - the history of an anomaly. Anthropos V. 5: 98-110. BAILEY G. and GAMBLE C. 1991- The Balkans at 18000 BP: the view from Epirus. In C. Gamble and 0. Soffer (eds.), The world at 18000 B. P.: 148-167. BOTTEMA S. 1979- Pollen analytical investigations in Thessaly (Greece). Palaeohistoria 21: 19-40. CHOURMOUZIADIS G. 1971. Dio neai egatastassis tis Archaioteras Neolithikis is tin ditikin Thessalian. Ar-chaiologika Analekta ex Athinon 4: 164-174. 1972. Anaskafi is ton Prodromon Karditsis. Ar-chaiologikon Deltion 27.8: 394-396. EVISON M., KYPARISSI-APOSTOLIKA N., STRAVOPODI E., FIELLER N. and SMILLIE D. in press. An ancient HLA type from a Palaeolithic skeleton from Theopetra cave, Greece. In N. Kyparissi-Apostolika (ed.), Pro- ceedings of the Conference "Theopetra Cave: Twelve years of Excavation and Research" Trikala, 6-7 November 1998. FACORELLIS Y. and MANIATIS Y. in press. Evidence for 50000 years of Human activity in the cave of Theopetra by 14C (English abstract). In N. Kyparissi-Apostolika (ed.), Proceedings of the Conference "Theopetra Cave: Twelve years of excavation and Research", Trikala 6-7November 1998. GALANIDOU N. in press. Reassessing the Greek Mesolithic: the pertinence of the Markovits collection. In N. Galanidou and C. Perles (eds.), The Greek Me-solithic-Problems and Perspectives. British School at Athens Studies. HALSTEAD P. and JONES G. 1980. Early neolithic economy in Thessaly - some evidence from excavations at Prodromos. Anthropologika 1: 93-117. HAMILAKIS Y. in press. Zooarchaeology of Neolithic Theopetra (English abstract). In N. Kyparissi-Aposto-lika (ed.), Proceedings of the Conference "Theopetra Cave: Twelve years of excavation and Research", Trikala 6-7November 1998. JACOBSEN T. W. 1976. 17000 years of Greek Prehistory. Scientific American 234: 76-87. KARKANAS P. 2000. Theopetra: Ena spileo dini pli-rofories gia tin exelixi tou klimatos. National Geographic (Greek edition) Vol. 4, No. 4: 136-139. KARKANAS P. in press. Lithostratigraphy and Diage-nesis of Theopetra cave deposits, Kalambaka (English abstract). In N. Kyparissi-Apostolika (ed.), Proceedings of the Conference "Theopetra Cave: twelve years of excavation and Research", Trikala 6-7 November 1998. KOTSAKIS K. 1992. 0 Neolithikos tropos paragogis: ithagenis i apikos? Praktika Diethnous Synedriou gia tin arhaia Thessalia sti mnimi D. R. Theocha-ri, Volos 1987: 120-134. KOUMOUZELIS M, KOZtOWSKIJ, NOWAK M., SOB-CZYK K., KAZCANOWSKA M„ PAWLIKOWSKI M., PAAZDUR A. 1996. Prehistoric settlement in the Kli-ssoura gorge, Argolid, Greece (excavations 1993-1994). Prehistoire Europeenne 8:143-173. KYPARISSI-APOSTOLIKA N. 1999a. The Palaeolithic deposits of Theopetra Cave in Thessaly. In G. Bailey E. Adam, E. Panagopoulou, C. Perles and K. Zachos (eds.), The Palaeolithic Archaeology of Greece and Adjacent Areas. Proceedings of the ICOPAG Conference: 232-239. 1999b. The Neolithic Use of Theopetra Cave in Thessaly. In P. Halstead (ed.), Neolithic Society in Greece, Sheffield Studies in Aegean Archaeology 2: 142-152. in press a. The Neolithic period of Theopetra Cave (English abstract). In N. Kyparissi-Apostolika (ed.), Proceedings of the International Conference "Theopetra Cave: Tivelve years of Excavation and Research", Trikala 6-7 November 1998: Ch. 14. in press b. The Mesolithic of Theopetra Cave: new data for a debated period of Greek Prehistory. In N. Galanidou and C. Perles (eds.), The Greek Meso-lithic-Problems and Perspectives. British School at Athens Studies. MARKOVITS A. 1933. Die Zaimis-Hohle (Kaki Skala, Megaris, Griechenland), Spelaologisches fahrbuch 13/14 (1932-33): 133-146. MILOJCIC V, BOESSNECK J., JUNG D. and SCHNEIDER H. 1965. Palaeolithikum urn Larissa in Thes-salien. Bonn. NEWTON S. in press. The Mesolithic Fauna from Theopetra. In N. Galanidou and C. Perles (eds.), The Greek Mesolithic-Problems and Perspectives. British School at Athens Studies. forthcoming. Theopetra cave and the Palaeolithic-Mesolithic transition in Southern Europe. In G. Gamble, P. Halstead, Y. Hamilakis and E. Kotzam-bopoulou (eds.), Zooarchaeology in Greece: recent Advances. London, British School at Athens Studies. NTINOU M. in press. Preliminary results of the an-thracological analysis from Theopetra Cave. In N. Kyparissi-Apostolika (ed.), Proceedings of the Conference "Theopetra Cave: Twelve years of Excavation and Research". Trikala 6-7November 1998. PAPADOPOULOU M. 1958. Magoulitsa, neolithikos synikismos para tin Karditsan, Thessalika A: 39-49. PAPATHANASSOPOULOS G. (ed.) 1996. Neolithikos Politismos stin Ellada. Athens: Goulandris Foundation. PERLES C. 1988. Chipped stone assemblages as an index of cultural discontinuity in early Greek prehistory. In E. French and K. A. Wardle (eds.), Problems in Greek prehistory. Bristol: Classical Press: 477-488. 1989. La Neolithisation de la Grece. In 0. Auren-ce and J. Cauvin (eds.), Neolithisations. Proche et Moyen Orient, mediterranee orientate, Nord de /' Afrique, Europe meridionale, Chine, Amerique du Sud, Oxford, British Archaeological Reports 516:110-127. 1990. Les industries lithiques de Franchthi (Ar-golide, Grece), Tome II, Les industries du Meso-lithique et de Neolithique initial, Excavations at Franchthi Cave, Greece, Fascicule 5, Bloomington and Indianapolis: Indiana University Press. 1994. Les debuts du Neolithique en Grece. La Recherche 266, V. 25: 642-649. 1999. Long-term perspectives on the occupation of the Franchthi Cave: continuity and discontinuity. In G. Bailey, E. Adam, E. Panagopoulou, C. Perles and K. Zachos (eds.), The Palaeolithic Archaeology of Greece and Adjacent Areas. Proceedings of the ICOPAG Conference, British School at Athens Studies 3- 311-318. ROWLEY-CONWY P. and S. NEWTON in press. Late Palaeolithic and Mesolithic animal bones from Theo-petra cave. Preliminary report. In N. Kyparissi-Apostolika (ed.), Proceedings of the Conference "Theo-petra Cave: Twelve years of Excavation and Research", Trikala 6-7 November 1998. RUNNELS C. 1988. A prehistoric survey of Thessaly: new light on the Greek Middle Palaeolithic Journal of Field Archaeology 15: 277-290. 1993. The Lower and Middle Palaeolithic of Thessaly, Greece. Journal of Field Archaeology 20: 299-317. SAMPSON A. 1996 a. La grotte du Cyclope. Un abri de pecheurs prehistoriques? Archeologia No 328: 55-59. 1996 b. New evidence of the Mesolithic in the Greek Area. Archaiologia (Athens) v. 61: 46-51. in press. The Mesolithic of the Aegean islands: Subsistence Strategies and Contact networks. In N. Galanidou and C. Perles (eds.), The Greek Me- solithic-Problems and Perspectives. British School at Athens Studies. SKOURTOPOULOU K. in press. Chipped stone production in Theopetra cave during the Neolithic (English abstract). In N. Kyparissi-Apostolika (ed.), Proceedings of the Conference "Theopetra Cave: Twelve years of Excavation and Research", Trikala 6-7 November 1998. SORDINAS A. 1969. Investigations of the prehistory of Corfu during 1964-1966. Balkan Studies 10: 393-424. STRAVOPODI E, MANOLIS S. and KYPARISSI-APOSTOLIKA N. 1999- Palaeoanthropological findings from Theopetra Cave in Thessaly: a preliminary report. In G. Bailey, E. Adam, E. Panagopoulou, C. Perles and K. Zachos (eds.), The Palaeolithic Archaeology of Greece and Adjacent Areas. Proceedings of the ICOPAG Conference: 271-281. THEOCHARIS D. 1958. Ek tis prokeramikis Thessa-lias - Prossorini anaskafiki ekthessis. Thessalika A: 70-86. 1967.1 avgi tis Thessalikis Prohistorias. Thessalika Meletimata No 1, Volos. TRANTALIDOU K. in press. Faunal Remains from the Earliest Strata of the Cave of Cyclope, Youra, Alon-nessos. In N. Galanidou and C. Perles (eds.), The Greek Mesolithic-Problems and Perspectives. British School at Athens Studies. VANDIVER P., SOFFER 0., KLIMA B. and SVOBODA J. 1990. Venuses and Wolverines: the origins of ceramic technology, ca. 26 000 B. P. In W. D. Kingery (ed.), The Changing Roles of Ceramics in Society: 26000 B. P. to the Present. VITELLI K. D. 1993- Franchthi Neolithic Pottery. Vol. II: Classification and phases 1&2. Excavations at Franchthi Cave, fascicule 9, Bloomington, Indianapolis: Indiana university Press. UDK: 903(495+560)"634' Documenta Praehistorica XXVII Thessaly, Franchthi and Western Turkey: Clues to the Neolithisation of Greece? Laurens Thissen Amsterdam, Netherlands laurens@thissenl.demon.nl ABSTRACT - Potential pathways towards neolithisation are discussed for two regions: Thessaly and the Peloponnese (Franchthi). Differences between North and South Greece in settlement patterns, subsistence and social structure are argued to reflect similar variations in a hypothesised West A natolian Aceramic Neolithic. It is proposed to seek, the neolithisation of Greece in an ultimate stress-si-tuation in specific inland plains of West Anatolia. Traditional contacts of sites here with settlements along the West Turkish seaboard may ha ve pro vided information on new land, the stimulus to consider migration as a possible solution, and the practical means of crossing the Aegean. IZVLECEK - Vclanku got 'orimo o moznih poteh neolitizacije v dveh regijah: Tesaliji in na Pelopone-zu (Franchthi). Razpravljamo o tem, da razlike med severno in juzno Grcijo odsevajo podobne spre-membe hipoteticnega zahodnoanatolskega akeramicnega neolitika tako glede vzorca naselitve, naci-iia prezivljanja kot tudi druzbene zgradbe. Predlagamo, da zacetke neolitizacije Grcije iscemo v skraj-no stresnih razmerah v ravnicah v notranjosti zahodne Anatolije. Tradicionalni stiki med tukajsnji-mi nfijdisci in naselbinami v zahodnoturskem primorju so morda prinesli podatke o novi dezeli, spro-zili razmisljanja o migraciji kot mozni resitvi in zagotovili nacin za preckanje Egejskega morja. KEYWORDS - Anatolia; culture contact; Greece; migration; neolithisation INTRODUCTION The often-observed difference between North and South Greece (for instance in settlement patterns, in material culture or as to pathways towards neolithisation - cf. Perles 1987-34; Demoule and Perles 1993-364, 370; Halstead 1994) may give clues to the nature and origins of the first farming communities on European soil. In an earlier paper I stated that the neolithisation of Europe was, in its initial stages, an "Aegean phenomenon," meaning that the actual impulse to establish permanent farming villages in Greece resulted from a long-lasting Aegean interaction (likewise, Halstead 1996.299). On the basis of the accumulated evidence acquired from the recent work carried out in North-western Turkey (Oz-dogan 1999; Roodenberg 1999a; 1999b), I likewise proposed that the assumed bridge function of that area vis-a-vis the neolithisation of Europe had not much to credit it - the area, at the present state of research, being peripheral both to the developments taking place in Central and Southwest Anatolia and to those in Greece (Thissen 2000a). Here, I would like to put forward some hypotheses concerning the origin of the Greek Early Neolithic, integrating the evidence from Thessaly and Southern Greece (notably from Franchthi), and that from West Turkey. If the Greek North-South difference for the EN period is accepted, we may perhaps extrapolate this difference backward in time, e.g., to the Aceramic Neolithic, which at Franchthi at least was not disruptive to the preceding Mesolithic stage (cf. Chapman 1994.136; Halstead 1996.300). THESSALY Evidence concerning the nature and the dating of the earliest Neolithic in Thessaly (inclusive of a PPN phase) is rather conflicting and not generous in hard facts. The conflicts appearing in the debate on the validity of a PPN stage (cf. Nandris 1970; Theocha- ris 1973; Bloedow 1991; Bloedow 1992-1993) and, the absence of solid data particularly felt in the limited exposures and absolute chronological backing. The possibility of an autochthonous process of plant cultivation in Thessaly has recently been ventured on the basis of the Theopetra Cave data (Budja 1999.132; cf. also Kyparissi-Apostolika 1998; 1999. 238). Here, on the NW edge of the Thessalian Plain, in the Mesolithic deposit, several wild seeds and pulses have been identified, including hordeum vul-gare subsp. spontaneum, and triticum boeticum (Kyparissi-Apostolika 1998.249; 1999.237). According to the excavator, the Mesolithic deposit also contained some sherds in situ {Kyparissi-Apostolika 1998.249). Unfortunately, at Theopetra there is a huge gap of 800 calendar years in the local sequence of the Mesolithic-Neolithic, at least as far as it has been fixed in 14C dating (Kyparissi-Apostolika 1999. 236-239) (Fig.l). It is, therefore, impossible to check whether the knowledge and use of wild seeds and pulses led to domestication here; and whether the sherds suggest an independent early invention of the craft of pottery making potentially much along the same lines as hypothesised by Vitelli for 'Aceramic Franchthi,' viz. as representing a "rare and precious" product (Vitelli 1993)• Barring the as yet preliminary data from Theopetra, it is presently safer to as- Atmospheric data from Stuiver et al. (1998); OxCal v3.5 Bronk Ramsey (2000); cub r:4 sd:12 prob usp[chron] Sequence THEOPETRA Phase Mesolithic DEM- 142 9722±390BP DEM-207 9093±55(IBP DEM- 316 9348±84BP DEM- 315 9275±75BP DEM- DEM- DEM- 125 8674±76BP 120 8525±57$P 360 7995±73BP Phase Neolithic DEM-455 6890±43BP DEM- 454 6563±68BP DEM' 361 6326±94BP DEM- 122 6222±38BP sume that the Mesolithic-Neolithic sequence in Thessaly is disruptive in time. And I share the view of several authors (Demoule and Perles 1993-364- 365; Van Andel and Runnels 1995) that the EN in Thessaly was disruptive also in the cultural sense - being a foreign intrusion by migrant farmers. Simultaneously, this might not have been the case for Franchthi (vide infra). While the question of a PPN phase in Thessaly rests on unsteady grounds (largely due to the restricted areas excavated), it is a fact that the small component of much less sophisticated pottery occurring in the basal layers of Achilleion, Argissa, Gendiki, Nes-sonis, Sesklo and Soufli (the Fruhkeramikum, or the Early Neolithic I) decreases in time, coinciding with an increase in technical ability in overall pottery manufacture (cf. Wijnen 1981.33-34). As Wij-nen rightly assumes, these crudely made vessels themselves do not represent the first pottery-mak-ing stage, but are merely part of that initial stage (Wijnen 1981.34). The fact that a 'beginners' stage' and an 'advanced stage' are not archaeologically separable (in the chronological sense - hence the 'mix'), would rather point to rapidity on the part of the potters in mastering the different levels of expertise required. Archaeologically visible is the intense level of experimentation apparent from the EN Thessalian pottery concerning shaping, the use of slips and paints, and firing (Wijnen 1993-323; and contra Bloedow 1991-43)- JIU 14000CalBC 12000CalBC 10000CalBC 8000CalBC 6000CalBC 4000CalBC Calibrated date Fig. 1. Theopetra Cave radiocarbon dates for the Mesolithic-Neolithic, calibrated individually. In line with the foregoing, it seems fair to assume that the idea to start manufacturing pottery was developed by the settlers upon founding the sites in Thessaly. Put otherwise, pottery was more or less a local invention, and possibly part of adaptation processes to cope with new environments and circumstances of living (pots used to "underline the social significance of hospitality," as Hal-stead has suggested [1994. 206]). If locally invented, then the knowledge of pottery, or the notion of its need, cannot have been part of the cultural world of those who ultimately risked the move towards Thessaly. Fol- lowing Theocharis (1967.173-174) and Wijnen (.1981.97; 101-102), Perles in 1989 also suggested that the pottery of EN Greece was developed locally, and, consequently, proposed that the first Greek Neolithic be established during a pre-ceramic stage (Perles 1989.119). Certainly the evidence from the PPN sites in the Near East makes it clear that there is no direct relationship between farming and the origins of pottery, and people had, of course, built up long experience in cooking foodstuffs without the knowledge or the need of containers made of baked clay (cf. Pavlu 1997.28ff; Bjork 1998.44). The theory of Vitelli that the EN pottery was not used for cooking, but was instead non-utilitarian and high-status (Vitelli 1989; Perles 1993.377; cf. Hal-stead 1994.206), is probably correct, viewing the absence of soot traces, and the dominant presence of ring- and pedestal bases (cf. Wijnen 1981.33 for Sesklo). gest different ways of handling, positioning and using pottery. Open dishes of the Thessalian kind are not in general use in the Central and NW Anatolian assemblages, while, alternatively, Anatolian oval-mouthed shapes (possibly referring to original wooden or gourd prototypes), do not seem to have been present in Thessaly. Also, the possibly earliest pottery from SW Turkey, viz. that found in the Pam-phylian site of Bademagaci and datable to the second half of the 7th millennium cal BC, appears to be based on a more diversified vessel repertoire and a handle system different from the Thessalian pottery (see Duru 1999.Figs. 33~38, 42). The date for the beginning of permanent farming villages in Thessaly cannot be satisfactorily established with the Achilleion and Sesklo radiocarbon dates, which do not allow as fine-grained a resolution as one would like for this key phase in Euro- The earliest Thessalian pottery consists of only a few different categories, including dishes and deep globular bowls, which typologically merge into holemouth bowls (Fig. 2). The use of vertically- or horizontally pierced knobs is limited to the bowls, while other handle types are not attested. Vessels are not larger than medium size, with neither rim diameters nor general height extending over 20 cm. Ring- and low ring bases are common. If we accept the general date for the Thessalian EN as starting at about 6300 cal BP1 at the earliest (see below), then a correspondence with contemporary pottery concepts and use in the wider world (to be specific: Anatolia) is farfetched. The Konya Early Neolithic pottery (e.g., Catalhoyuk East levels VI-0, the Bey§ehir and Sugla Lakes sites), and by extension the NW Anatolian wares (Demircihuyuk, Mente-§e, Ilipinar and the Fikirtepe sites) differ in a major way from the Thessalian assemblages in their discrimination of different types of cooking > c,> A / rfl S;!haped h7u'the ?rfTmce FiS- 2- Early Neolithic I pottery from Sesklo and Achilleion (marked tor Hat bases, and the use of alterna- «A"). Dishes, bowls and holemouth bowls/pots (after Wijnen tive handle types in Anatolia, all sug- 1981.26 Fig. 11). 1 Calibrations throughout this paper are made with help of the latest version of the OxCal program (v3.5) (Bronk Ramsey 2000), dependent on the most recent calibration curve INTCAL98 (Stuiver et al, 1998). pean prehistory. At Sesklo, combining the stratigra-phic evidence collected so far from three trenches all located in the NE-sector of the Sesklo Acropolis,2 the entire EN period including the PPN stage does not seem to have comprised more than three to four building levels (Tab. 1). Virgin Soil l4C dates come from several different Sesklo trenches, but only Trench 2 yields a larger, though by no means sufficient body of dates (Fig. 3).3 When calibrated, agreement of the posterior distributions of the six Trench 2 PPN and EN dates is far below the statistically acceptable (34.8% where the threshold is set at 60.8%) (Fig. 4). A date much earlier than the 6300 cal BC threshold would, however, conflict in my view with the small number of individual building levels counted at PPN/EN Sesklo and EN Achille-ion. For Achilleion, after a reanalysis of the stratigra-phic sequence (Thissen 2000b), only two building levels appear to belong to the EN period. Achilleion yields a larger series of 14C dates (nine for the combined levels la and lb, eight for the combined levels Ila and lib) (Fig. 5). When we combine the probability distributions of the calibrated dates of Achilleion Ia-Ib, assuming that the samples stem from a single event or from events occurring within a short period, then the earliest possible range at 2a is set at 6240-6160 cal BC (Fig. 6).4 A pre-6300 cal BC date for the onset of the Thessa-lian PPN/EN would further be in disaccord with the most likely date for the beginning of the MN period at about 6000 cal BC. Finally, the limited thickness of the EN deposits at Sesklo and Achilleion does not suggest a very large time span for these levels. If we further know that the earliest possible range for the beginning of settlement at Nea Nikomedeia, based on a combination of the probability distributions of the calibrated dates, can be put at 6230-6150 cal BC (at 2a), then again a pre-6300 cal BC beginning of permanent villages in Sesklo and Achilleion is not warranted (Figs. 7 and 8). The Nea Nikomedeia dates, likewise, conform rather perfectly to those of basal Achilleion, suggesting roughly contemporaneous events. In this respect, when acknowledging EN Thessaly as a coherent, culturally cohesive society (Ha/stead 1994. 207) with social barriers to external contacts (apart from those established and maintained by tradition - see below), this may stand in the way of seeing Thessaly as a root area for renewed colonization of the regions further north, notably of Macedonia. Indeed, cultural variance between Thessaly and Macedonia is visible in pottery, in settlement patterns and in commitment to the land (cf. Thissen 2000a, 194; Fotiades et at, 2000.217; for a contrary view, however, see Wilkie and Savina 1997). Franchthi In contrast to Thessaly, Southern Greece, or at least Franchthi, reflects a mobile, non-static society, not intent on exploiting the land, but the boundless sea. Franchthi Cave was used over an extremely long period, but discontinuously and fluctuating in intensity. An important place, as Chapman argues, for those who used the cave (Chapman 1994.137), it must have been only one of several (cf. Ulbrich Cave, Zai-mis Cave) during the Palaeolithic, Mesolithic, and Aceramic (Initial) Neolithic periods. It is probably only a matter of time before similar sites will be found on the Turkish shores of the Aegean. It is even possible that this sense of mobility is still present in the first pottery Neolithic at Franchthi Cave and at Paralia, the small open-air site on the coast.5 Thessalian patterns of tradition and place and of self-containment are not conspicuously visible at period building number of thickness method building levels of deposit EN III settlement burnt EN ll/lll pise or mud brick 1-2 (?) 20 - max. 85 cm EN I like PPN 2-3 floors 40 - 50 cm PPN single stone 1 30 - 60 cm foundations/pise (total thickness: 90 - max. 195 cm) Tab. 1. Sesklo. Stratigraphic evidence from Trench 2, NE-sector Acropolis (after Wijnen 1981.12, Fig. 5; Wijnen 1992). 2 A trial trench of 2.5x2.5m, excavated in 1956 and 1957 (Wijnen 1981.9. Fig. 0; a trench dug in 1962, possibly trench Thita (Bio-edow 1991.23, Fig. 8); and trench 2, excavated during 1963 and 1965 (Wijnen 1981.12, Fig. 5). 3 I am greatly indepted to Mies Wijnen for allowing me to use the Groningen data of Sesklo. 4 The nine Achilleion Ia-Ib dates are exclusive of LJ-4449 and UCLA-1896A which come from test pit east. In addition, I have reassigned level II samples LJ-3328, LJ-3186 and LJ-3325 to level I instead on stratigraphic grounds. 5 cf. Jacobsen (1984), hypothesizing such patterns of mobility for the MN period. Atmospheric data from Stuiver et al. (1998); OxCal v3.5 Bronk Ramsey (2000); cub r:4 sd:12 prob usp[chron] Sequence SESKLO PPN Phase P PN P-1681 7755±97BP P-1682 7483±72BP P-1680 7300±93BP Phase E P-1679 GrN-168 N I 7611±83BP EN, Trench 2 41 7520±30BP Phase EN ll-l GrN-16i 42 7250±25BP thic settling of the NE Peloponnese, not so much the traditional search for new fertile land. It is only during the Final Neolithic and the Early Bronze Age that people in the Ar-golid oriented themselves towards their hinterland: only then were the best soils of the region settled (Van Andel and Runnels 1987. 81-85). 8000CalBC 7500CalBC 7000CalBC 6500CalBC Calibrated date Fig. 3- Sesklo Trench 2 radiocarbon dates, calibrated individually. Sequence Sesklo PPN/E Franchthi, or for that matter, in the other Neolithic sites in the Peloponnese and Central Greece (cf. De-moule andPerles 1993364,370). Tell built-up, with its long-term association to localised space (cf. Chapman 1989) is quite rare in Southern Greece, and if occurring, is seemingly restricted to sites in key positions in relation to the sea (e.g. Old Corinth, Go-nia, Lerna, Halai or Franchthi Cave itself in a sense) (cf. also Cherry et al. 1988). Moreover, settlement locations differ from Thessaly - in Southern Greece rocky preeminences are favourite spots instead of floodplains, terraces and fens (Demoule and Perles 1993-362; Van Andel and Runnels 1995; but also Wilkie and Savina 1997.201). As Van Andel and Runnels have pointed out on the basis of their extensive surveys, the Argolid was "very thinly settled" during the EN period (and, indeed, during the ensuing MN and LN periods as well) (Van Andel and Runnels 1987. 67). They also make clear that the EN inhabitants of Franchthi did not exploit their environment to the full. In stark contrast to Thessaly, the EN settlers in the Peloponnese "(...) failed to spread out to fill the space available (...)" (Van Andel and Runnels 1987. 69; cf. 75, Map 13). Van Andel and Runnels consider the region's geographic setting "(...) well placed to maintain trade contacts throughout the southern Aegean and across the Peloponnese" (1987.73) - as the first reason for the Early Neoli- A reanalysis of the Franchthi Cave sequence, a thorough treatment of which falls outside the limits of this paper (Thissen 2000b), has 6000CaiBC 5500CaiBC led me to the following synopsis: O Franchthi phase Int 0/1 is acera-mic, following Jacobsen (1969-352), but in contrast to Vitelli (1993-39). Domesticated plant and animal species are already known. A set of five consistent 14C dates makes it possible to date this stage somewhere within a range of 7000-6600 cal BC (Fig. 9). Franchthi Int 0/1 is roughly equivalent to Perles' "phase lithique X" (Perles 1990.115ff) and to Hansen's "botanical zone V/VI and VI" (Hansen 1991.163). © Franchthi phase Int 0/1 probably did not lead into FCP 1, a point which is confirmed by the pertaining 14C dates, which show a gap of 200-400 years (at la) between Int 0/1 and FCP 1. Given this discontinuity at the site, the knowledge of domesticates during Interphase 0/1 may have remained an isolated phenomenon, not leading to continued ex- ploitation. It is, however, unlikely that the settling Atmospheric data from Stuiver et al. (1998); OxCal v3.5 Bronk Ramsey (2000); cub r:4 sd:12 prob usp[chron] Phase PPN P-1681 P-1682 P-1680 Phase EN I P-1679 GrN-16£ Phase E GrN-168 M {A= 100.2% 101.9% 33.7% 39.0% 41 57.1% N ll-lll 42 98.3% 34.8%(A'c= 60.0%)} AjL 8000CalBC 7500CalBC 7000CalBC 6500CalBC Calibrated date 6000CalBC 5500CalBC Fig. 4. Sesklo Trench 2 radiocarbon dates, showing posterior distributions (in black). LJ-; of Franchthi, and the introduction of domesticates at the site, including the absolute date at which the latter allegedly took place, are phenomena that are applicable or contemporary to other Greek regions, notably to Thessaly. The immediate density of sites in the Thessalian Plain, the continuity evident from their individual histories and the coherence shown by their material culture all point to a strong, rather sudden and lasting impact on the land. Franchthi phase Int 0/1, at the present state of research, would antedate the Thessalian PPN/EN by some 400 calendar years. © Being perhaps a trial event, the NE Pe-loponnese with Franchthi lacked occupation for many centuries. Only by the 60th-59th century cal BC were pottery-Neolithic sites established in the Argolic Gulf: at Franchthi the Paralia site was founded, while the old Cave site was reused as well, as evidenced by contemporary deposits. The total duration of occupation in the cave as well as at Paralia during Franchthi phase FCP 1 may have been fairly short, given the shallow deposits and the absence of thick and consecutive occupation horizons, and given the absence of any development within the ceramic assemblage. 0 In view of the gradual transition attested both in the pottery- and in the lithics development from FCP 1 over FCP 2, as well as similar patterns in faunal remains over FCP 1-2, the EN period at Franchthi most likely is not as early as suggested in the literature. Instead, FCP 1 could well immediately predate FCP 2, i.e. roughly at about 5900 cal BC. © On the basis of the radiocarbon evidence, the MN period at Franchthi, represented by the FCP 2 and FCP 3 stages, appears to be of short duration as well - the absolute dates suggesting the 58th and 57th centuries cal BC (Fig. 9)- © The Franchthi FCP 1 pottery resembles rather perfectly the 'EN' assemblage retrieved from Old Corinth. There, what Weinberg classified as 'red monochrome' and 'coarse monochrome' wares have strict parallels in technique (paste, colour, firing) and form (including decoration and location of vertically pierced knob handles below the rim) with FCP 1. Also at Corinth continuity is noted for EN-MN (Lavezzi 1978.427). 0 As pointed out by Lavezzi (I.e.). the EN-MN development at Corinth is comparable to Franchthi Cave, Atmospheric data from Stuiver et al. (1998), OxCal v3.5 Bronk Ramsey (2000): cub r:4 sd:12 prob usp[ch Sequence ACHILLEION Early Neolithic Phase la + lb UCLA-1882B 7360±155BP P-2j18 7471±77BP GrN-7437 7440±55BP GrN-7438 7390+45BP LJ-3329 7370±50BP LJ-3184 7320±50BP LJ-3328 7310±50BP 186 7300+50BP LJ-3325 7290±i)0BP Phasi e Ma + lib LJ-3180 7550±60BP UCLA-1896C 7330±100BP r L P-2120 7342+68BP LJ-3326^7290±80BP GrNh7436 7295+703P P-2117 7273±76BP LJ-3201 7210±90BP LJ-3181 7250±$0BP 8000CalBC 7500CalBC 7000CalBC 6500CalBC 6000CalBC 5500CalBC Calibrated date Fig. 5. Achilleion radiocarbon dates from the Early Neolithic levels (exclusive of LJ-4449 and UCLA-1896A from test pit east), calibrated individually. Lerna, Phlius and Asea, and even to Central Greece, i.e. Elateia and Halai. Similar EN-MN pottery groups have been acknowledged by Howell surveying Eastern Arcadia (Howell 1970.103-108). © If the update of the Franchthi Neolithic (FCP 1) is correct, it opens the road to reconsidering the EN period of Southern Greece in general. Given the tight correspondences in the pottery assemblages of EN Franchthi and Lerna (cf. Vitelli 1974), as well as the links with other EN sites in the region, we have, I think, to reconsider the current temporal equation of the Southern Greek EN period with the Thessalian EN sequence. While, unfortunately, 14C dates from EN sites in Southern Greece are lacking (except for the rather unreliable ones from Elateia [cf. the remarks on these dates by Vogel and Waterbolk 1963-182-183]), Franchthi EN would rather date to a time frame during which, in Thessaly, the Middle Neolithic Sesklo period had already begun. Interpretation and explanation of the misleading archaic aspect of the Southern Greek pottery, well represented by Franchthi, of simple vessel forms (for example, deep Atmospheric data from Stuiver et ai. (1998); OxCal v3.5 Bronk Ramsey (2000); cub r- = 0.8 ; o.6 ® 0.4 ; 0.2 0.0 Combine la + lb 68.2% probability 6230BC (37.7%) 621OBC 6190BC (1.3%) 6180BC 6170BC (16.2%) 6160BC 6130BC (13.0%) 6110BC 95.4% probability 6240BC (69.1%) 6160BC 6140BC (26.3%) 6100BC Agreement 57.9% evidence, and are tentatively assigned to the last centuries of the 7th millennium cal BC (Fig. 10). _i_ 6500BC 6400BC 6300BC 6200BC Calendar date 6100BC 6000BC Fig. 6. Achilleion radiocarbon dates from the Early Neolithic levels (exclusive ofLJ-4449 and UCLA-1896A), the probability distributions of level Ia-Ib combined. hemispherical bowls), 'Early Neolithic' handle shapes such as vertically pierced knobs, and a limited number of ceramic categories, may profit from re-evalua-ting it from the perspective that we have at least two different pottery traditions: a Thessalian one and a Southern Greek one, neither related in time nor in origin. These different traditions are possibly nothing more than a reflection of the different pathways that led to the neolithisation of both regions (see further below). Evaluating the present data on early site location in West Turkey, it is the diversity that is striking. Several sites are immediately on the Aegean seaboard, to note: Karaagagtepe (Fig. 10, site 12) on the southern tip of the Gelibolu peninsula, Kumtepe (site 13), Co§kuntepe (site 11), Ayio Gala cave, Killiktepe (site 8), Liman-tepe (site 33), Milete (site 9); or on islets (Tavsan Adasi [site 10]) and small peninsulas (Sapli Adasi [site 5]). The orientation of these sites was evidently towards the sea, their position not on the edge of fertile alluvial plains suggesting that agriculture may not have been the dominant subsistence strategy. Several of these coastal sites are situated on rocky outcrops (e.g. Co§kunte-pe, Tavsan Adasi). By contrast, the inland sites yielding similar material culture assemblages are concentrated in several alluvial plains and side valleys of the Gediz and Biiyiik Menderes rivers: e.g., the Akhisar and Manisa Plains (Fig. 10, sites 40, 41, 44, 48, 49) (French 1965; Ding 1997), the Torbali Plain (sites 19, 20, 22, 26, 32, 35, 36) (Merig 1993), the Akgay Plain (sites 6, 7) (Akdeniz 1997), or even the Ala§ehir Plain (sites 43, 45-47, 50) (Merig 1993). WESTERN TURKEY Even compared to the scarcity of data on EN Thessaly and Southern Greece, West Anatolia is worse off, hard evidence (14C dates, excavations) being virtually non-existent. Little work is done here, and consists' almost solely of surface surveys. If I will, nonetheless, treat this area as a potential key region in the neolithisation of SE Europe, cq. Greece, I can only defend my position with the current adage that absence of evidence is not evidence of absence. It should be said beforehand that aceramic sites have not yet been attested in West Anatolia. The sites mentioned here are, on the basis of the surface pottery, cross-dated with the Hacdar Atmospheric data from Stuiver et al. {1998}; OxCal v3.5 Bronk Ramsey (2000); cub r:4 sd:12 prob usp[chron] Nea Nikomedeia EN Combine Nea Nikomedeia EN [n=8 A=113.5%(An= 25.0%; OxA-16Q5+OxA-4282 7400±64BP OxA-3876 OxA-3874 QxA-1606 7370±90BP 7370±80BP OxA-4283 7324±67BP OxA-3873 7300±80BP OxA-3875 7280±90BP. P-1203A 7281±74BP QxA-1604 OxA-4281 7223±64BP Combine Nea Nikomedeia EN 7000CalBC 6500CalBC 6000CalBC Calibrated date 5500CalBC Fig. 7. Nea Nikomedeia radiocarbon dates (exclusive of Q-655, GX-679, P-1202 and OxA-l6()3+OxA-4280), calibrated individually. From the density of early sites, these plains obviously represent key areas for the Neolithic-Early Chalcolithic periods in West Anatolia, and might very well have done so earlier, their success being dependent on the exploitation of the alluvial plain and the mountains around. Atmospheric data from Stuiver et al. (1998); OxCal v3.5 Bronk Ramsey (2000); cub r:4 sd:12 prob usp[chron] 1 0.8 ro | 0.6 CL o> 0.4 ^0.2 al 0.0 Combine Nea Nikomedeia EN 68.2% probability 6220BC (42.9%) 6160BC 6140BC (25.3%) 6100BC 95.4% probability 6230BC (55.3%) 6150BC 6140BC (40.1%) 6080BC Agreement 113.5% The obsidian from the West Anatolian settlements was probably all imported from Melos, although this assumption rests on the two analysed pieces from the site of Morahlar (Fig. 10, site 48) in the Akhisar Plain {Renfrew, Cam and Dixon 1965. 235). The lithics industry appears based on simple blades, but seems highly exploited as evidenced by Coskuntepe (Seeker 1990.11, 13-Fig. 2:11-16). Indeed, the mountains have been used thoroughly for raw materials (cf. the use of pumice, volcanic stone from the area around Kula, and silex in Morahlar [French 1965. 15; Ding 1997.266-267]). While solid data are still lacking on the Turkish side, several correspondences between Thessaly and West Anatolia can tentatively be pointed out in support of a shared cultural background. If the survey data from West Anatolia are be trusted, individual regions of this large area were rather densely settled at least in the final centuries of the 7th millennium cal BC, both in the coastal areas and in large alluvial plains in the hinterland. A dualism in orientation, on the one hand to the sea, on the other hand to solid farming away from the Aegean, hidden behind coastal mountain ranges, is equally present both in Thessaly and in West Anatolia. At the same time, dependence on the sea was possibly felt also in the hinterland, if we may believe the Melian obsidian at Morahlar. West Anatolian sites, being usually not much larger than 100 m in diameter (cf. Hoca (Je§me or Coskuntepe), would compare both in settlement location and in size to Thessalian EN villages. The picture sketched by Halstead for EN Thessaly, viz., that of a thickly wooded region studded with small, but many sites may well be applicable to the West Anatolian plains (cf. Halstead 1981; 1989). His interesting point, that sheep/goat were foremost kept for their meat, as evidenced by the high death rate of young sheep, hence discarding pastoralism as a means of 6500BC 6400BC 6300BC 6200BC Calendar date 6100BC 6000BC 5900BC Fig. 8. Nea Nikomedeia radiocarbon dates (exclusive of Q-655, GX-679, P-1202 and OxA-1603+OxA-4280), the probability distributions combined. subsistence (Halstead 1989), is equalled and confirmed at least at Ilipinar. Also here, sheep/goat were bred purely for meat (Buitenhuis 1989-1990. 117-118), while most of the animals were killed as sub-adults or young adults (I.e.). It is likely that similar patterns will become available when archaeological research finally focuses on contemporary sites in West Anatolia. In line with Halstead's findings for Thessaly, West Anatolian inland sites might also have relied primarily on arable farming to provide for the energy requirements of the population. In contrast, the West Anatolian coastal sites very probably acted as base sites for the Aegean navigators-fishers, and it is likely that several such locations did represent points of reference within a Transaegean network of places, much as might have been the case for Franchthi (cf. Chapman 1994. 137). These coastal sites may only have depended on farming in a very limited way and, to extend the speculation, may have depended on the inland villages for agricultural products in exchange for obsidian and raw materials from the sea. Concerning parallels in material culture, both regions may eventually demonstrate affinity, although here again the lack of West Anatolian data is felt. Discussing the Thessalian evidence above, I have ventured the idea that Thessalian ceramic procedures were developed on the spot and not as part of the baggage of the immigrants. We must begin to reconcile, as pleaded for by Cauvin (1994), the facts of colonization with the possibility of cultural variation. However, in the chipped stone assemblages of both Thessaly and West Turkey a rapprochement may exist in the parallel preference for a flake/blade industry. Neither the Thessalian toolkit, nor, for instance, that of Hacilar in the SW Anatolian Lakes District has much affinity with the sophisticated assemblages of the Konya Plain (e.g., Catalhoy uk East). In addition, despite the fact that the obsidian of Ha-cilar may have been retrieved from the Acigol source, Mortensen, in his analysis of the Hacilar obsidian, was not able to relate it to the obsidian industry in the Konya region or to Mersin (Mortensen apud Mellaart 1970.156-157). Both in technology, in type range and in quantity, the Hacdar obsidian yielded highly different results. Lamellar pressure-flaking, well known at Catal, was rare at Hacdar. While the Hacdar industry is based on blades (as that of Kurugay, very close to Hacdar, see Baykal-Seeher apud Duru 1994.108), at (Jatalhoyiik fifty different types of tools and weapons have been dis- Atmospheric data from Stuiver et al. (1998); OxCal v3.5 Bronk Ramsey (2000); cub r:4 sd:12 prob usp[chron] Sequence Franchthi Cave Phases 0/1-2 Phase Interphase 0/1 P-2095 7980±110BP P-1526 8020±80BP P-2094 7930±100BP P-1527 7900±90BP P-1392 7790±140BP Phase FCP P-1525 770(D±80BP P-1667 7280±90BP P-2093 6940±90BP Phase FCP 2.2 P-1824 6670±70BP Phase FCP 2.3 P-1537 6650+80BP Phase FCP 2.3-2.4 P-1922 6790±90BP P-1922A 6730±70BP Phase FCP 2.5 1-6128 6855±190BP tinguished, forty-three for Levels III—II alone (Bialor 1962; Mellaart 1975.103). At Hacdar, flint dominated the tool kit, versus 42% of obsidian (Catal: 95% obsidian in levels III—II). In view of the cultural coherence of Western Anatolia and the Lakes Region, the remark by Mortensen that the flint and obsidian objects surveyed by French from Morahlar "bear a considerable resemblance to the chipped stone industry of Hacilar" (Mortensen apud Mellaart 1970. 157) gains in importance. DISCUSSION The following discussion rests on two basic assumptions, first, the existence of an Aceramic Neolithic culture in West Anatolia depending for a large part on farming, but to be distinguished in an inland area, and a coastal area, with different commitments and subsistence bases, and second, the existence of a body of Transaegean navigators (hunters-fishers?) acting as know-how transmitters, suppliers of Me-lian obsidian and ultimately as a medium in transferring West Anatolian aceramic farmers to the Thessalian Plain. 9000CalBC 8000CalBC 7000CalBC 6000CalBC 5000CalBC Calibrated date Fig. 9. Franchthi Cave radiocarbon dates for Phase Int 0/1, Phase 1 and Phase 2, calibrated individually. In seeking for a possible origin of the Thessalian settlers, Perles' remark that they deliberately ignored the local raw material sources (meaning those for quality flint, obsidian and jasper, not so much the directly utilitarian ones) is significant (Perles 1992.121, 124, 128). Living in an obviously strongly socialised context, the first Thessalian farmers, in order to evade conflicts, depended on local exchange mechanisms (Perles 1992.121). The avoidance of conflict, the success of which is archaeologically visible in the coherence of Thessalian EN culture, in the close proximity of sites over centuries and in the scarcity of burnt destruction horizons, and an unwillingness to engage in conflict might both find their basis in the to all appearances peaceful, non-ag-gressive milieu, with only a minor hunting component (cf. Halstead 1994.206-207; 1996.304-305). *12 All Lesbos ^AyioGala Chios *r f i f . jr '/ J 4 29 37 38 40 31 24 A*50 423 33 A3&0 45 A A"6 439 *S1 Samos 2^3236 KucuhMenderei *28 *9 *10 A17 *18 ..{JW ) Jm » liF-JK. BuyukMondsres •■4S • Fig. 10. Map showing Late Neolithic -Z?flr(y Chalcolithic sites in Western Anatolia. 1. Ak Hoyiik. 2. Boz Hoyuk-Dinar. 3. Dinar Hoyiik. 4. Afrodisias. 5. Akbiik-Saphadasi. 6. Hamidiye (Toygartepe). 7. Kavakhkahve. 8. Killiktepe. 9. Milete. 10. Taiwan Adasi. 11. Co§kuntepe. 12. Ka-raagaftepe. 13■ Kurntepe. 14. (bandar I. 15. (andar II. 16. Karakurt. 17. Otner-koy. 18. Siirmeli Hoyiik. 19. Altmtepe. 20. Arapkahve. 21. Araptepe-Bekirlerte-pe. 22. Aslanlar. 23. Barbaros. 24. Bor-nova. 25. Bozkoy. 26. Bulgurca. 27. (al-tidere. 28. Gokfealan. 29. Helvaei-Hoyu-cek. 30. Hoyiicek II. 31. Kiiftik Yaman-lar. 32. Lembertepe. 33- Limantepe. 34. Nemrut Hoyiigu. 35. Ogfananasi. 36. Te-pekoy. 37. Ulucak. 38. Yenmis. 39. Ak-makca. 40. Alibeyli. 41. Arpah II. 42. Qerkeztevflkiye. 43. Gavurtepe (Alasell ir). 44. Kayislar. 45. Ketnaliye. 46. Kil-lik. 47. Mersinli. 48. Morali (Moraltlar Hoyiigii). 49. Nuriye. 50. Yuvacah. 51. Alifli Hoyiik. Conflict-evasion might be prompted by the initial foreignness to the land and by the concomitant necessity to keep together. Early Neolithic Thessalian society thus offers a picture of a densely occupied land of peaceful, undoubtedly hard-working (see Sahlins 1972.Ch. 1) farming villages or hamlets, keeping in close contact with each other and making use of each others', overlapping, raw material source areas. Not only in a material sense, but also socially, all villages are thus linked through a network of reliable integrative mechanisms maintained through local exchanges (Perles 1992.121). Though internally dynamic, this society is self-contained and static externally except for a few important and specific, direct alliances established by tradition. The self-containment, and the probable intention on the part of the settlers to "make it" in the new land (an intention which would live on over the generations and find its consolidation and justification in the success of the exploit) would generate what Chapman has called the concept of "cyclical, or reversible, time" (1994.139). In such a concept of time -denying linear progression - tradition, and the maintenance of tradition, will become the yardstick for life, instead of time; tradition which causes to remain to the land, to the village and to the building plot (cf. Chapman, I.e.). As Perles argues, in such a society there are social barriers to engage and maintain the circulation of goods and/or people over long distances, adding that such a society presupposes "a socially more neutral trading system, such as one based on recognised middlemen" (Perles 1992. 121). Direct alliances, whether or not mediated by "neutral" middlemen, might very well form the basis for the import of Melian obsidian so conspicuously and permanently present on Thessalian sites. Perles argues convincingly that the Melian obsidian was probably not acquired through local initiative, as seems confirmed by the small amount of these materials on each EN site not being in proportion to the exertions of such distant trips. There is, furthermore, absence of local variation, the incoming material arriving in a worked state, while, additionally, the specialised know-how needed to circumnavigate the Aegean may not have been present (Demoule and Perles 1993-383). A similar constellation might well have existed in West Turkey: inland sites depending on middlemen for providing the Melian obsidian (and perhaps other "marine" resources); middlemen supplying information about the Aegean, about "available land" across the sea. If the existence of an aceramic farming society in West Anatolia is assumed (albeit not yet proven), this society, with its long ancestry, must have differed from what, in a later stage, was implanted in Thessaly; in fact, it is improbable that the social structure of the Anatolian inland communities were the same as those in Thessaly. Put otherwise: the Anatolian colonists did not apply the traditional social structure (perhaps viewed as one of the causes leading to migration, and therefore not to be reiterated) in the new land, where it was decided to 'stick together.' If we above followed Halstead, see- ing the Thessalian pottery as structuring the laws of hospitality (Halstead 1994:206), it is perhaps this decision that stood at the basis of the willingness to make and use pottery, so indeed, following Perles and Vitelli, primarily as a social construct, to tie the bonds between the different groups of settlers. It is even imaginable that the aceramic West Anatolian inland farmers knew about the new invention, either from the Konya Plain area to the East of them, or from the lakes area to the Southeast of them, but did not find any immediate use for them. From Perles' analyses and on the basis of the evidence discussed in this paper, several suggestions may be advanced, testable through excavations of some key sites both in inland West Anatolia, and along the Turkish Aegean seaboard: • The EN Thessalian settlers are not identical to those who explored the Aegean, catching tunny-fish and exploiting Melos obsidian. • These EN Thessalian colonists are farmers pur sang, not much depending on hunting, as well as not much acquainted with the sea and with marine life. Their ignorance or dislike of the sea and marine food may find its corroboration in their food habit patterns: most of the EN Thessalian sites yield very meagre evidence of the use of seafood (cf. Wijnen 1981.54; Schwartz apud Wijnen 1981.112 for Sesklo). A major exception is the site of Pyrasos sitting immediately on the coast (Wijnen 1981.57). • It is, among others, the Franchthi people who navigated the Aegean, who possessed the expertise to cope with the currents and winds and who may have acted as providers of the Melian obsidian, either directly to Thessaly, or more probably, by way of middlemen. The information supplied by Wijnen (1981.78) concerning the site of Nea Makri, on the coast of Attica, and the only site thus far yielding obsidian implements including cores and waste flakes, leads her to suggest that "people from this area shipped obsidian from Melos, knapped blades and then transported them over the country or exchanged them for other goods."6 • The people frequenting Franchthi Cave may have represented the filter through which the existence of fertile land in Thessaly became known in the region of origin of the Thessalian farmers. Moreover, the knowledge that the fertile land was roughly similar environmentally to the root country must have filtered through as well. 6 Clear evidence of obsidian working is attested also at Neolithic Halai (Coleman 1992.274). • The most logical option for the root country of the Thessalian colonists may be the plains in the West Anatolian hinterland, The farmers living here stood in contact with sites on the West Anatolian seaboard, ergo with the Aegean navigators, as proven by the Melian obsidian from Morali. • Given the absence of know-how on the part of the colonists in seafaring, it is tempting to assume that the transporting of the migrants was in the hands of the Aegean navigators. • If the Thessalian farmers had their roots in West Anatolia, then the fact that they developed pottery only upon arrival in Thessaly, implies that they left West Anatolia at a time when they were still 'aceramic.' The development of pottery in Thessaly cannot therefore be related to cooking, as techniques of cooking without clay containers were known. • By identifying the Thessalian settlers as rooting in West Anatolia, we may assume a long-lasting process of development in West Anatolia before the move across the Aegean was decided upon. A longer process of development is required in order to comply with the time needed to raise to conflicts, to become so overwhelmingly embedded in farming and in order to take the decision to solve the conflict by migration. This assumption, incidentally, might explain why the Thessalian Early Neolithic appears so balanced and mature (as both practising animal husbandry and plant exploitation, with hunting playing a minor role). This longue-duree perspective also would provide a framework within which to position the age and complexity manifest in the exploitation of the Melian obsidian (cf. Perles 1989.117). -ACKNOWLEDGEMENTS lam very grateful to Mihael Budja and the University of Ljubljana for their invitation to the 7th Neolithic Seminar "The Processes of Neolithisation in Eurasia," held in Ljubljana 22. -27. May 2000, and for their subsequent hospitality. REFERENCES AKDENIZ E. 1997. 1995 yili Biiyiik Menderes Ovasi ve cevresi yiizey ara§tirmalan. Arastirma Sonuglari Toplantisi 14(11): 233-254. BIALOR P. 1962. The chipped stone industry of Ca-tal Hiiyiik. Anatolian Studies 21: 67-110. BJORK C. 1998. A comparative outline of the Early Neolithic cultures in China and in the Near East. Do-cumenta Praehistorica 25:37-52. BLOEDOW E. 1991. The 'Aceramic' Neolithic phase in Greece reconsidered. Mediterranean Archaeology 4: 1-43. 1992-1993. The date of the earliest phase at Argissa Magoula in Thessaly and other Neolithic sites in Greece. Mediterranean Archaeology 5/6: 49-57. BRONK RAMSEY C. 2000. OxCal program t'3.5; online: http://rlaha. ox. ac. uk/orau. html. BUDJA M. 1999. The transition to farming in Mediterranean Europe - an indigenous response. Docu-menta Praehistorica 26: 119-141. BUITENHUIS H. 1989-1990. [Investigations at Ilipi-nar] The faunal remains. Anatolica 16: 111-119. CAUVIN J. 1994. Naissance des divinites. Naissance de I'agriculture. La revolution des sym-boles au Neolithique. Paris: CNRS Editions. CHAPMAN J. 1989. The early Balkan village. In S. Bokonyi and P. Raczky (eds.), Neolithic of Southeastern Europe and its Near Eastern connections. International conference 1987. Szolnok-Szeged: 33-53- 1994. The origins of farming in South East Europe. Prehistoire Europeenne 6: 133-156. CHERRY J., DAVIS J., DEMITRACK A., MANTZOURANI E., STRASSER TH. and TALALAY L. 1988. Archaeological survey in an artifact-rich landscape: a Middle Neolithic example from Nemea, Greece. American fournal of Archaeology 92:159-176. COLEMAN J. 1992. Excavations at Halai, 1990-1991. Hesperia 61: 265-289. DEMOULE J.-P. and PERLES C. 1993. The Greek Neolithic: a new review, fournal of World Prehistory 7: 355-416. DINC R- 1997. Kulaksizlar mermer idol atolyesi ve gevre ara§tirmalari. Arastirma Sonuglari Toplantisi 14(11): 255-282. DURUR. 1994. Kurugay Hoyiik I. 1978-1988 kazi-lartn sonuglari, Neolitik ve Erken Kalkolitik (lag yerlesmeleri. Results of the excavations 1978-1988. The Neolithic and Early Chalcolithicperiods. Ankara: Tiirk Tarih Kurumu. 1999- The Neolithic of the Lake District. In M. Oz-dogan and N. Ba§gelen (eds.), Neolithic in Turkey. The cradle of civilization. New discoveries: 165-191. FOTIADES M., HONDROYANNI-METOKI A., KALOGI-ROU A. and ZIOTA C. 2000. Megalo Nsi Galanis (Ki-trini Limni Basin) and the Later Neolithic of Northwestern Greece. In S. Hiller and V. Nikolov (eds.), Karanovo III. Beitrdge zum Neolithikum in Sud-osteuropa: 217-228. FRENCH D. 1965. Early pottery sites from western Anatolia. Bulletin of the Institute of Archaeology 5: 15-24. HALSTEAD P. 1981. Counting sheep in Neolithic and Bronze Age Greece. In I. Hodder, G. Isaac and N. Hammond (eds.), Patterns of the past. Sudies in honour of David Clarke: 307-339. 1989. Like rising damp? An ecological approach to the spread of farming in South East and Central Europe. In A. Milles, D. Williams and N. Gardner (eds.), The beginnings of agriculture: 23-53■ 1994. The north-south divide: regional paths to complexity in prehistoric Greece. In C. Mathers and S. Stoddart (eds.), Development and decline in the Mediterranean Bronze Age: 195-219. 1996. The development of agriculture and pasto-ralism in Greece: when, how, who and what? In D. Harris (ed.), The origins and spread of agriculture and pastoralism in Eurasia: 296-309. HANSEN J. 1991- The palaeoethnobotany of Franchthi Cave. [Excavations at Franchthi Cave, Greece, Fasc. 7]. Bloomington & Indianapolis: Indiana University Press. HOWELL R. 1970. A survey of eastern Arcadia in prehistory. Annual of the British School at Athens 65: 79-127. JACOBSEN T. 1969. Excavations at Porto Cheli and vicinity, preliminary report, II: the Franchthi Cave, 1967-1968. Hesperia 38: 343-381. 1984. Seasonal pastoralism in Southern Greece: a consideration of the ecology of Neolithic Urfirnis pottery. In P. Rice (ed.), Pots and potters. Current approaches in ceramic archaeology: 27-43. KYPARISSI-APOSTOLIKA N. 1995. Prehistoric inhabitation in Theopetra Cave, Thessaly. In J.-C. Decourt, B. Helly and K. Gallis (eds.), Thessalia, dekapente chronia archaiologikis ereunas, 1975-1990. Apote-lesmata kai prooptikis. Praktika diethnous sune-driou Lyon, 17-22 apriliou 1990/La Thessalie, quinze annees de recherches archeologiques, 1975-1990. Bilans et perspectives Actes du colloque international Lyon, 17-22 avril 1990: 103-108. 1998. The significance of Theopetra Cave for Greek prehistory. In M. Otte (ed.), Prehistoire d'Anatolie. Genese de deux mondes. Actes du colloque international de Liege (28 avril-3 mai 1997): 241-252. 1999- The Palaeolithic deposits of Theopetra Cave in Thessaly (Greece). In G. Bailey etal. (eds.), The Palaeolithic archaeology of Greece and adjacent areas: 232-239. LAVEZZI J. 1978. Prehistoric investigations at Corinth. Hesperia 47: 402-451. MELLAART J. 1970. Excavations at Hacilar. Edinburgh: Edinburgh University Press. 1975. The Neolithic of the Near East. London: Thames and Hudson. MERig R. 1993. Pre-bronze age settlements of West-Central Anatolia. Anatolica 19: 143-150. NANDRIS J. 1970. The development and relationships of the earlier Greek Neolithic. Man N.S. 5:192-213 OZDOGAN M. 1999. Northwestern Turkey: Neolithic cultures in between the Balkans and Anatolia. In M. Ozdogan and N. Basgelen (eds.), Neolithic in Turkey. The cradle of civilization. Neiv discoveries: 203-224. PAVLU I. 1996. Pottery origins. Initialforms, cultural behavior and decorative styles. Prague: Karoli-num. PERLES C. 1987. Les industries du Neolithique "pre-ceramique" de Grece: nouvelles etudes, nouvelles interpretations. In S. K. Kozlowski (ed.), Chipped stone industries of the early fanning cultures in Europe: 19-39. 1989- La neolithisation de la Grece. In 0. Auren-che and J. Cauvin (eds.), Neolithisations. Proche et Moyen Orient, Mediterranee orientate, Nord de l Afrique, Europe meridionale, Chine, Ame-rique du Sud: 109-127. 1990. les industries lithiques taillees de Franchthi (Argolide, Grece). Tome II. Les industries du Mesolithique et du Neolithique Initial. [Excavations at Franchthi Cave, Greece, Fasc. 5]. Bloomington & Indianapolis: Indiana University Press. 1992. Systems of exchange and organization of production in Neolithic Greece. Journal of Mediterranean Archaeology 5: 115-164. RENFREW C, CANN J. and DIXON J. 1965. Obsidian in the Aegean. Annual of the British School at Athens 60: 225-247. ROODENBERG J. 1999a. Investigations at Mente§e Hoyiik in the Yenisehir Basin (1996-97). Anatolica 25:21-36. 1999b. Ilipinar, an early farming village in the Iznik Lake basin. In M. Ozdogan and N. Ba§gelen (eds.), Neolithic in Turkey. The cradle of civilization, New discoveries: 193-202. SAHLINS M. 1972. Stone age economics. Chicago: Aldine. SEEHER J. 1990. Co§kuntepe - anatolisches Neolithi-kum am Nordostufer der Agais. Istanbuler Mittei-lungen 40:9-15. STUIVER M., REIMER P., BARD E„ BECK J., BURR G., HUGHEN K„ KROMER B., MCCORMAC G., VAN DER PLICHT J. and SPURK M. 1998. INTCAL98 radiocarbon age calibration, 24000-0 cal BP. Radiocarbon 40: 1041-1083. THEOCHARIS D. \%l.laugi tis Thessalikis proisto-rias [The dawn of Thessalian prehistory (Origins and early evolution of the Neolithic)]. Volos. 1973. Neolithic Greece. Athens: National Bank of Greece. THISSEN L. 2000a. A chronological framework for the neolithisation of the Southern Balkans. In S. Hil-ler and V. Nikolov (eds.), Karanovo III Beitrage zum Neolithikum in Sudosteuropa: 193-212. 2000b. Early village communities in Anatolia and the Balkans, 6500-5500 cal BC. Studies in chronology and culture contact. Unpublished PhD dissertation, University of Leiden. VAN ANDEL T. and RUNNELS C. 1987. Beyond the acropolis. A rural Greek past. Stanford: Stanford University Press. 1995. The earliest farmers in Europe. Antiquity 69: 481-500. VITELLIK. 1974. The Greek Neolithic patterned Ur-firnis Ware from the Franchthi Cave and Lerna. Michigan: University Microfilms International/Ann Arbor Xerox. 1989. Were pots first made for food? Doubts from Franchthi. World Archaeology 21: 17-29. 1993. Franchthi Neolithic pottery. Vol. 1: Classification and Ceramic Phases 1 and 2. [Excavations at Franchthi Cave, Greece, Fascicle 8]. Blo-omington & Indianapolis: Indiana University Press. VOGEL J. and WATERBOLK H. 1963. Groningen radiocarbon dates IV. Radiocarbon 5: 163-202. WIJNEN M.-H. 1981. The Early Neolithic I settlement at Sesklo: an early farming community in Thessaly, Greece. Leiden:Leiden University Press. 1992. Building remains of the Early Neolithic period at Sesklo. In Praktika (Diethnes sunedrio gia mnimi ton Dimitri R. Theochari): 55-63- 1993. Early ceramics: local manufacture versus widespread distribution. Anatolica 19:319-331- WILKIE N. and SAVINA M. 1997. The earliest farmers in Macedonia. Antiquity 71: 201-207. UDK: 903(560)"634":621.968-032.5 Documenta Praehistorica XXVII The Problems of Dating Prehistoric Axe Factories and Neolithisation in Turkish Thrace Bur^in Erdogu Department of Archaeology, University of Durham, UK Burcin.erdogu@durham.ac.uk ABSTRACT - Prehistoric axefactories or manufacturing areas have been found in the Sarkoy region of Turkish Thrace. So far, they are unique in the prehistoric record of the Balkans and Anatolia. A typological and petrological analysis of the stone axe factories and investigations of their distributions are in progress. Early results show that all the axes are manufactured from the same rock -metabasite. The source of metabasite is the Western outcrops of Ganos Mountain. At the early Neolithic settlement of Hoca Ce§me, stone axes were made of the same rock as the Sarkoy sources. In this paper, I discuss the problems of dating prehistoric axe factories, together with the wider problems of the early Neolithic period in Turkish Thrace. IZVLECEK - Vturski Trakiji v regiji Sarkoy so bile odkrite prazgodovinske delavnice oziroma podroc-ja, kjer so izdelovali kamnite sekire. Zaenkratje to edinstvenprimer vprazgodovinskem zapisu Bal-kana in Anatolije. Vteku so tipoloske in petroloske raziskave kamnitih sekir izdelavnic in raziskave njihove razprostranjenosti, Raziskave zaenkrat kazejo, da so bile use sekire izdelane iz iste kamnine - metabazita, Povrsinska nahajalisca te kamnine lezijo na zahodnem delu hriba Ganos. Tudi v zgodnjeneolitski naselbini Hoca (.esme so bile kamnite sekire izdelane iz enake kamnine. Vclan-ku razpravljamo o tezavah pri datiranju prazgodovinskih delavnic sekir ter o sirsih vprasanjih zgod-njeneoiitskega obdobja v turski Trakiji. KEY WORDS - stone axe production; neolithisation; Thrace; Western Anatolia INTRODUCTION According to R. Wagner, "in learning how to use tools, we are secretly learning how to use ourselves" (Wagner 1975.77). He claims that tool use is about the objectiflcation of our skills as the controls which tools place on the relationship between humans and the environment. In prehistoric times, stone served as the main material for making tools. Only stones that met certain technological requirements were used and they were deliberately sought out. Stone is most intractable and the most difficult material to work on. Each tool took so much labour to produce that it was among the most valued of a person's possessions. The polished stone axe is a very significant tool type, especially in the Neolithic period. The polished stone axe provided a central symbol within Neolithic society because it effectively linked a whole range of spheres of human activity (Tilley 1996. 114). The axe was a basic tool in subsistence, an important exchange item linking together communities, personal status and prestige items in a community. The stone axes which circulated within society had a worth which would have been related to debt and kinship, and to the articulation of relationships between persons and groups (Thomas and Tilley 1993-290). According to Tilley: "the axe provided a durable symbolic medium for creating and maintaining social ties and dependencies through ritual and everyday activities" (Tilley 1996.114). K. Kristiansen has argued that the axe links together agricultural production, exchange and ritual consumption and feasting (Kristiansen 1984.79). Ethnogra- phical studies show that the leader in lineage groups in the Pokou, Ussiai and Matankol people of the Admiralty Islands is in possession of the axe/adze and can also pass it on to his successor (Ohnemus 1998.152). He holds the axe/adze in his hand while speaking and dancing in ceremony. On a sad occasion, such as death, the leader appears without his axe/adze. The axe/adze is also used in peacemaking talks or punishment. It stands for law and order, peace and joy. Among Australian Aboriginal societies, the stone axe was prominent in interpersonal relations, in the totem system and in the wider belief system (Tagon 1991.194). Axes probably had important roles in ceremonial activities. In the Papua New Guinea highlands, the largest axes were valued especially for ceremonial and display purposes (White and Modjeska 1978. 29). During the mortuary feast of the Sabarl Islanders of Papua New Guinea, the dead paternal clan publicly presents five ceremonial axes to its maternal clan heirs. In absolute secrecy, the axes are used to construct in effigy the corpse of the honoured dead (Battaglia 1983-291). The axes were placed next to the dead against one another, with the heads facing in the same direction. They rest on their blades and points, resembling angels in the air. They are said to represent a human body reclining in its grave. The axes and the dead become intertwined in the grave. Then, the deceased was raised as it were from the grave and re-installed at the centre of reproductive life. This marks the beginning of his life as an ancestor and establishes him as a source of economic and spiritual aid for the living. The corpse is magically endowed with the power to reproduce axe blades; it becomes more than a representation of the ancestor, it becomes a concrete substitute for the 'child' as a reproductive unit of his/her society tBattaglia 1983-298). The axes may serve as points of reference for broader belief systems. In Neolithic chamber tombs in Brittany, the deposition of particular types of stone axes is relatively restricted, especially of those obtained from great distances. By passing from hand to hand, over the distance from their sources, each axe would have built up its own genealogy, as myths became attached to them (Kristiansen 1984.79). The tomb may act to fix all of those myths in one location. Axes were so deeply connected with the person that the history of axe and person becomes intertwined. Thus the burial of the axes introduced the presence of this person to the depositional context (Thomas and Tilley 1993-293)- In Neolithic chamber tombs in Brittany, some of the axes were deliberately broken (Thomas and Tilley 1993-290-91). Axes may be regarded as having biographies, like persons. They are born (produced), exchanged and destroyed (die). As Chapman argued, the relationship between fragmented objects and persons is an important, interpretative link (Chapman 1996. 214; 2000). Axes were deeply connected with the person, and when the body died, the axes were rit-ually destroyed. In the centre of the chamber at Mane-er-Hroek, Brittany, a large ring of jadeite and a huge axe was arranged so that its butt penetrated the ring. Behind the blade of the axe were two beads, and behind this were a perforated axe and a further bead. All these axes and beads are set along a north-south axis. According to Thomas and Tilley, the sexual symbolism is here quite explicit that all axes represent phalluses (Thomas and Tilley 1993-291-293)-Among the Australian Aboriginal groups in the Yir Yoront of North Queensland and Western Arnhem Land, stone axes and other tools were recognised as belonging to men, especially older men, and embodying their ancestral power (Tagon 1991-194-195)-The women and young had to borrow the axe from the older male. In the borrowing, the status, position and power of older males were reinforced. Aborigines also believed that the axes were formed from ancestral bones. In Sabarl Island society, the axes are personified persons and identified with the bodies of the persons making them (Battaglia 1983-295). The axe blade is called "Hinona" - the "content" or "vital substance" of the valuable. In the context of the physical person, "Hinona" is the term for "genitals" and "right hand", a symbolism associated with economic and biological reproduction. The axe blade broadly represents the reproductive potential of a singular person (Battaglia 1990.133)- Factories or manufacturing areas are places where craft specialists perform a limited set of activities on a frequent, perhaps regular basis in order to produce items for exchange with other groups of people. Stone axe factories or manufacturing areas were recently found in Turkish Thrace. Although the field data are not complete, typological and petrological investigations of prehistoric stone axe factories show us the operational chain for prehistoric axe manufacture and the raw material from which the axes are made. In this article, I would like to discuss the problems of the dating of stone axe factories, together with the wider problems of the early Neolithic period in Turkish Thrace. Bayrarpic --Sofitk&> KtfMBM Kauai* HAMAVLiT- MASLAtf r. FENER KARADUTLAR |THE SEA OF MARMARA In all examples, roughouts, flakes and hammer stones were found in and around the prehistoric settlements. In three sites, all axes are manufactured from the same rock, metabasite, and the operational chain for prehistoric axe manufacture is the same. The stone axe factories of Hamaylitarla (Buruneren) and Fener Karadutlar are situated at the rock source. The source is in western outcrops of the Ganos Mountains. However, Yartarla is about 3 km away from the source. Problems of Dating Prehistoric Axe Factories Although hundreds of stone axes are discovered at excavations each year in the Balkans and Anatolia, until now no prehistoric axe factories have been found. However, at the site of Divostin in Serbia, numerous unfinished axe specimens indicate the method of manufacture. In Divostin phase II, a working floor with roughouts, drilling pieces, flakes and also a large pit filled with the flakes of roughouts were found (.Prinz 1988.257-259 and plan Ilia). All the axe factories were found associated with prehistoric settlements. Fig. 1. Location map for Prehistoric axe factories in Turkish Thrace. Prehistoric Axe Factories in Turkish Thrace In 1989, a large number of roughouts was sold to Istanbul Museum by a farmer from the Sarkoy region. Scholars working in Eastern Thrace were for a long time looking for the site from which these roughouts came. In 1995, the stone axe factory of Yartarla was found by M. A. Isin, director of Tekirdag Museum, and he demonstrated that the roughouts held in Istanbul Museum come from Yartarla. Later, two more axe factories or manufacturing areas, Hamaylitarla (Buruneren) and Fener Karadutlar, were found by 0. Ozbek in the Sarkoy region. A geo-ar-chaeological project since 1997 has focused on the typology and petrology of axes and the wider questions regarding these sources {Ozbek in this volume). However, since there are no intensive archaeological surface surveys in the Sarkoy region yet, it is possible that other such sites exist. As a result of the investigation of prehistoric axe factories, two topographical locations can be distinguished. The stone axe factory of Yartarla is located ca. 14 km North-West of Sarkoy, ca. 3 km north east of the village of Sofukoy. It is situated on a high terrace of the Kavak Suyu River. The Kavak Suyu River rises in the Ganos Mountain and descends westwards to the Gulf of Sa-roz. It has a flat, marshy, alluvial mouth. The Kavak Suyu runs through wide gorges, with steep sides that in some places rise vertically from the river, reaching a height of 200-250 m, at which Yartarla was formed. Hamaylitarla (Buruneren) and Fener Karadutlar are situated on well-watered lowlands at the southern foot of Mounts Helvaci and Sarikayalar. Hamaylitarla (Buruneren) is located ca. 17 km West of Sarkoy and ca. 1 km west of the village of Kizilcaterzi. Fener Karadutlar is situated on Cape Ince, on the northern shore of the Sea of Marmara, ca. 1 km northwest of Hamaylitarla. The southern foot of Mount Helvaci, Kazanagzi stream and a number of small seasonal streams run into the Sea of Marmara, constituting flat, fertile cultivated land (%')• ___ Viiluce This concentration, Sector B, seems to indicate an area where stone axes were manufactured. Divostin phase II is dated to the Late Vinca Culture (4lh milen-nium BC). In Obre II in Bosnia, the regular shapes of sixteen stone axes were found between two stone slabs in sounding D, together with two big flint knives, three bone awls and two round baked clay objects. This has been interpreted as an axe-making area (Benac 1973-82 and Fig 13a). A similar axe-making area was also found in sounding VII at Obre II (Benac 1973-82). Obre II, sounding D, is dated to the Classic Butmir Culture (4th millennium BC). However, the dates from sounding VII in Obre II fall within 5th millennium BC. In Bosnia, at the site of Kalosevic-Malo Brdo a large number of flaked stone axe roughouts was discovered (Chapman 1976.146). The pottery on the site was found to date to the Late Vinca Culture. Kalosevic-Malo Brdo is probably an axe-manufacturing site; however, there are as yet no detailed investigations. The dating of axe factories in Eastern Thrace is problematic. No complete axes were found in the factories. In the settlement of Yartarla, Late Chalcolithic and Early Bronze Age sherds were collected. The settlement of Fener Karadutlar was completely destroyed by the building of a Byzantine church; only a few Early Bronze Age sherds were found. At the site of Hamaylitarla (Buruneren), early Neolithic sherds together with a few Early Bronze Age sherds were collected. It seems evident that, without excavations, it is difficult to date these stone axe factories. The petrological investigation of polished stone axes from excavated sites and surface collections in Eastern Thrace is still in progress. On the other hand, early results from the Early Neolithic site of Hoca (]e§me, near the town of Enez, showed that the polished stone axes of Hoca (Je§me was made from me-tabasite, probably from the Sarkoy region. Pottery Grid Neolithic E. B. A Chipped Weight [g] no. pottery pottery stone (Neo. only) 1 136 2 0 1500 2 33 1 1 500 3 21 1 0 300 4 76 2 0 800 5 120 27 1 1200 6 10 12 0 200 7 6 2 0 60 8 19 6 6 300 9 10 9 2 200 Tab. 1. Summary results of Hamaylitarla. similar to that of Hoca Ce§me was also found in Hamaylitarla. The following discussions will focus on materials from the site of Hamaylitarla. Hamaylitarla (Buruneren) and Its Relations Hamaylitarla was first discovered by M. A. Isin, director of Tekirdag Museum at the beginning of the 1990's and initially dated to the Early Bronze Age. In 1997, 0. Ozbek and the author visited the site and found stone axe rough-outs, flakes and hammer stones together with Early Neolithic pottery. As a result of the geo-archaeological project of 0. Ozbek, we understand that Hamaylitarla is an axe factory associated with prehistoric finds. The site of Hamaylitarla measures about 120x120 m. The stone axe factory is spread over 250 square meters. An area of 70x70 m was investigated, using alternately spaced 10x10 m grids (Tab. 1 and Fig. 3). The vast majority of the Hamaylitarla pottery, up to 90%, is red slipped and burnished. Black and brown burnished sherds were found in smaller quantities. All pottery is handmade, thin-walled and with an abundant use of grit and sand-temper. A little chaff is usually present in the paste. A number of sherds are tempered with chaff only. Mica is rare or absent. The paste colours are black, buff or cream. The different tones of the red and pink slip (mainly 2,5 YR 6/6 Red) are applied on both surfaces or on the exterior surface only. Sometimes the firing was irregular, causing mottling and smoke staining on the surface. The thickness of the application varies greatly. Surfaces are usually burnished. Sometimes the interior surfaces are smoothed only. The range of shapes consists of deep bowls with S profiles, bowls with plain rims and flaring sides, straight-sided bowls, bowls with angle-necks and hole-mouth vessels. Vertically placed tube-like and knob-like tubular lugs, as well as crescentic lugs, are characteristic. Flat and ring bases were found. For decoration, bands in relief occur (Figs. 4 and 5). Pottery similar to Hamaylitarla was also noted at Kaynarca, near the town of Gelibolu (Ozdogan 1986; 1999-Fig. 43). Recent surveys in Western Anatolia have revealed new Early Neolithic sites, such as Te-pekoy, Araptepe, Hoyiicek II, Nemrut (Meric 1993), Coskuntepe (Seeher 1990), Tepeiistii-Barbaros, Kyme-Ege Giibre and Bergama-Pasakoy. Similar wares and shapes occur especially at the sites of Araptepe, Tepeiistii-Barbaros, Kyme-Ege, Giibre and Bergama-Pa-sakoy (personal observation). Hamaylitarla was dated to the Classic Phase of the Fikirtepe Culture by Ozdogan (/997.21; 1999.214). Before talking about dating Hamaylitarla, I would like to discuss briefly the Fikirtepe Culture. In the Marmara region, the Fikirtepe culture is the earliest Neolithic Culture in the regional sequence. According to Ozdogan and Efe, three evolutionary phases were distinguished on the basis of pottery (Ozdogan 1997. 19; 1999.213; Efe 1996.51). The earliest phase of the Fikirtepe Culture, called the Pendik phase or archaic phase, is known from the lower layers of the Fikirtepe and Pendik excavations (Ozdogan 1997.21). Pottery from this phase comprises brown-grey, dark grey and sometimes pale orange, reddish brown coloured, burnished wares. The most common shapes are bowls and jars with either simple convex sides or with a slight "S" curve. Hole-mouth vessels, exaggerated large lugs and vertically perforated knobs are also common. Decoration is rare, mainly consisting of incised lines. The most common motifs are parallel lines, triangles, squares and hatching. The second phase is the Classic Fikirtepe Phase. It is best represented at Pendik and the upper horizon of Fikirtepe. According to Ozdogan, the transition between the first and the second phases is difficult to define (Ozdogan 1997.21). There is a gradual development in the pottery. The most common form is a bowl with "S" curved profiles and an oval mouth. Besides the heavy lugs, there are also tubular lugs. Four-footed rectangular vessels or boxes are very characteristic. There are also lids. The decoration is the same as in the previous phase, but the designs are more complex. During the Classic Fikirtepe phase, the red slipped, burnished wares began to appear. According to Ozdogan, Ilipinar level X represents the transition between the first and the second phases (iOzdogan 1997.21; 1999.213). Classic Fikirtepe pottery was also found in the Ktitahya-Eskisehir Region, Inner Western Anatolia (Efe 1995; Ozdogan 1997.21). The last phase of the Fikirtepe Culture is called Developed Fikirtepe or the Yarimburgaz 4 phase. This phase is characterized by elaborate decoration made by wedge-like excisions, often set directly behind one other or else set in zigzags. The designs are more complex, which Ozdogan called textile-like designs ('Ozdogan et. at. 1991). Dark faced wares are common. The surfaces of vessels are mostly burnished, and occasionally a dark slip is applied. The red slipped and burnished wares rarely occur. The most characteristic shapes are short-or tall-necked jars with squat globular bodys. Developed Fikirtepe type sherds were also noted in Ilipinar level VIII. The sites of Demirci Hoytik (Seeker 1987), Orman Fidanli-gi and Kanlitas (Efe 1989/90; 1996) in the Eskisehir region include typical developed Fikirtepe sherds. Recently, L. Thissen proposed that differences in the main vessel shapes between sites on the Eastern Marmara coast and Ilipinar X may be related to differences in the subsistence base rather than indicating chronological variety (Thissen 1999.32). This means there could be considerable regional and chronological variation. According to Thissen and Roodenberg, the Iznik-Yenisehir region was settled by early farmers migrating from Central Anatolia (Thissen 1999; Roodenberg 1993)■ However, for Thissen, Fikirtepe sites on the Eastern Marmara coast show the simultaneous adaptation of farming techniques and pottery, probably as a result of contact with the Iznik-Yenisehir region (Thissen 1999.38). Ozdogan also agrees that Epi-palaeolithic populations on the Eastern Marmara coast adapted Neolithic elements (Ozdogan 1998.450; 1999.215). Comparisons between the Iznik-Yenisehir region and the Eastern Marmara coasts show that the buildings of Fikirtepe and Pendik are oval huts with depressed floors and wattle and daub walls. However, the buildings of Ilipinar and Mentese are rectangular, constructed in wattle and daub. The subsistence of Ilipinar was mostly dependent on domesticates, while Fikirtepe and Pendik were based on mixed hunting, fishing and a stock breeding economy, with some agriculture (Roodenberg 1995.167-168; Ozdogan 1989.203). The chipped stone industries of both Fikirtepe and Pendik are both similar to the preceding Epi-palaeolithic tradition. Although Ozdogan argued that the chipped stone industry of Ilipinar is different from those of Fikirtepe and Pendik (1997.23), recent work shows that Ilipinar represent a continuation of a local Epi-palaeolithic tradition analogous to Fikirtepe and Pendik (Thissen 1999-37). Moreover, the chipped stone industries from Fikirtepe-type settlements in the Eskisehir region, such as Fin-dik Kayabasi and Asarkaya, are also similar to the Epi-paleolithic tradition (Efe 1995-108). 14C dates from Yarimburgaz Cave (Ozdogan et. at, 1991), Mentese (Thissen 1999; Roodenberg 1999) and Ilipinar (Roodenberg et. at 1989/90; 1995) are seen in Table 2. The Fikirtepe Culture can be dated to c. 6200-5700 cal BC. Now I shall discuss some observations about the early Neolithic period in Western Anatolia and the Marmara Region. I believe that these observations are directly related to the dating of Hamaylitarla and also correlations between Western Anatolia and the Marmara Region (Fig. 2). ® 90% of the Hamaylitarla assemblage is red slipped and burnished. The Fikirtepe Culture is marked by dark monochrome pottery. During the Classic Fikirtepe phase, red slipped and burnished wares began appearing. According to Ozdogan (1999.213), in the Classic Fikirtepe phase, red sherds comprise six to ten percent of the total assemblages. On the other hand, the excavations of Ilipinar and Mentese have not revealed red slipped and burnished sherds (personal communication with L. Thissen). In Western Anatolian sites, red slipped and burnished sherds similar to those of Hamaylitarla are found. This type of pottery is very common in Western Anatolia (Me-rig 1991; Harmankaya et. al. 1991). Lab. No. Level 14C Age BP Cal BC (1a) Grn- -15529 Yarimburg. 4 7330±60 6231 (6216,6167,6164) 6084 Grn- -18745 Yarimburg. 4 6650±280 5797 (5615,5585,5561) 5322 Grn- -24463 Mentese 7260±60 6213(6158,6143,6082) 6028 Grn- -24461 Mentese 7170±60 6156 (6018) 5931 Gm- -24462 Mentese 7050±35 5986 (5975,5950,5916) 5844 Gm- -17046 Ilipinar X 7100±30 6006 (5988,5940,5929) 5920 Grn- -15085 Ilipinar X 7100±50 6012 (5988,5940,5929) 5960 Gm- -15087 Ilipinar X 7070±50 5992 (5981,5946,5921) 5844 Gm- -17045 Ilipinar X 7025±30 5979 (5890) 5841 Grn- -17048 Ilipinar X 7025±90 5992 (5890) 5794 Grn- -17047 Ilipinar X 6925±70 5890 (5792)5724 Grn- -15084 Ilipinar X 6440+50 5475 (5466,5444,5401, 5382) 5325 Grn- -15077 Ilipinar IX 7020±50 5982 (5889,5846,5845) 5810 Grn- -16144 Ilipinar IX 6935±35 5840 (5835,5834,5799) 5735 Gm- -15078 Ilipinar IX 6920±70 5867 (5787) 5722 Grn- -16145 Ilipinar IX 6800±90 5736 (5711,5678,5672) 5624 Grn- -16146 Ilipinar IX 5330±80 4320 (4221,4163,4118, 4055) 4003' Grn- -17052 Ilipinar VIII 6995±45 5973 (5869,5861,5842) 5805 Grn- -17054 Ilipinar VIII 6990±30 5890 (5866,5864,5841) 5807 Gm- -17055 Ilipinar VIII 6980±45 5957 (5840,5816,5815) 5795 Grn- -17051 Ilipinar VIII 6960±45 5879 (5838, 5822,5809) 5749 Grn- -17056 Ilipinar VIII 6950±45 5870 (5837,5826,5806) 5742 Gm- -16149 Ilipinar VIII 6890±90 5841 (5734) 5671 Grn- -17053 Ilipinar VIII 6750±65 5718 (5658,5651,5640) 5565 © Vertically-placed tubular lugs characteristic for the Lake District as well as Western Anatolia are attested at Hamaylitarla. Vertically placed tubular lugs do occur rarely in the Classic Phase of the Fikirtepe Culture (only one published example: Ozdogan 1999. Fig.33, D.231), but are not characteristic elements of the Fikirtepe Culture. © There are some similarities between Hamaylitarla sherds and the early phases of Hoca (Je§me. Before explaining these similarities, I would like to discuss Hoca Qe§me. Excavations at Hoca (Je§me-Enez, conducted by M. Ozdogan between 1990 and 1992, suggest the existence of a different Early Neolithic culture in Eastern Thrace, called Hoca Cesme Culture by Ozdogan (Ozdogan 1991). Hoca Cesme is a small mound on a natural rise overlooking the delta of the Merig River, ca. 5 km east of the district centre of Enez. It measures about 80 x 70 m and the archaeological deposit is about 2 111 thick (Ozdogan 1993-182; 1998; 1999. 211-219). Four phases were discovered in Hoca (Je§me. Phase IV is the earliest phase at Hoca (Jennie. The architectural remains of this phase were built immediately 011 the bedrock. Houses are oval, wattle and daub, hut-like structures, cut into bedrock some 30 cm deep. Their diameter varies from 5 to 6111. The settlement was surrounded by a massive stone fortification wall around 1.20 m thick, and it sits right on the bedrock. Post-holes found just behind the wall indicate that the fortification wall was supported by a wooden structure (Ozdogan 1998. 439). The pottery of this phase is characterized by well burnished, thin walled red or black wares. Deep bowls with "S"-curves, vertically placed tubular lugs, crescent-shaped lugs, bead rims and flat bases are common elements of this phase. There are also a few zoomorphic vessels. Decoration is rare, mainly consisting of fine curvilinear or vertical bands in relief. There are also some grooved and incised sherds. Ref. University of Washington. Radiocarbon Calibration Program 2000 Tab. 2. The Fikirtepe Culture dates. Hoca (Je§me phase III consists of two architectural layers. Houses the black sea > flC^kM : 39 • Fig. 2. Distribution of Early Neolithic Settlements in Western and North-Western Turkey: 1. Asagi Pinar, 2. Bulgar Kaynagi, 3. Maya Baba, 4. Hoca Cepne, 5. Hamaylitarla, 6. Kaynarca, 7. Ugurlu, 8. Karaaga-ctepe, 9- Yarimburgaz, 10. Fikirtepe, 11. Pendik, 12. Tuzla, 13. Ilipinar, 14. Hoyu-cek, 15- Marmaracik, 16. Mentese, 17. Ye-nisehir II, 18. Aktopraklik, 19. Taracci, 20. Yilanlik, 21. Calca, 22. Coskuntepe, 23. Pasakoy, 24. Caltidere, 25. Kyme-Ege Gubre. 26. Araptepe, 27. Hoyucek II, 28. Kayislar, 29. Nuriye, 30. Alibey, 31- Mora-li, 32. Kucuk Yamanlar, 33- Ulucak, 34. Nemrut, 35- Agio Gala, 36. Tepeustu-Bar-baros, 37. Tepekoy, 38. Akmakca, 39- Asar-kaya, 40. Orman Fidanligi, 41. Demirci Hoyuk, 42. Kanlitas, 43. Keskaya, 44. Fin-dik Kayabasi. are again oval in plan and the fortification wall still exists, but with some renovations. On the north western edge of the settlement, one house is different from the others. It is a big oval hut of 7 m diameter, and its floor was paved with small pebbles coated and painted in red. The pottery of phase III shows a gradual development in fabric and decoration. All the ware types of Phase IV continue, although they are slightly coarser and thicker. Red coating on black burnished ware appears. There are also red-black, and light cream-red-black mottled sherds. Vessel shapes are similar to the previous phase. However, the profiles are now more carina-ted and necked jars are slightly increasing. X Data Fig. 3. Contour plan of Early Neolithic pottery distribution, Hamaylitarla. Phase II consists of three architectural layers. This phase is marked by a change in the plan and construction techniques of the buildings. The houses are rectangular in plan, with plastered walls. There are domed ovens on raised platforms; round or rectangular bins and working platforms were found inside the houses. The fortification wall was still in use. The red and black wares of the previous phases were now noted in lesser amounts. In phase II, there is an increasing amount of reddish-brown and matt black sherds. The sherds are notably thicker. Some new shapes are attested, such as footed rectangular or triangular vessels with excised or incised decoration, and tall-necked jars sometimes with small handles. The decoration of the preceding phases continues. Fluting and intentional mottling also occur. There are also some red on cream, red on black, white on black and white on red painted sherds. According to Ozdogan, houses and pottery, especially white on red sherds of phase II, are strongly reminiscent of Karanovo I period of Bulgaria (Ozdogan 1997; 1998.448). A few red on buff painted sherds in Phase II are also similar to Early Sesklo painted sherds (Ozdogan 1998.449). Phase I deposits have been considerably eroded by agricultural activity. In this phase, Toptepe phase I and the Kumtepe Ia-Besiktepe type of pattern-burnished bowls were found together with Karanovo III-IV types of sherds. A number of l4C dates are available at Hoca gesme (iOzdogan 1997.28; 1998). They are presented in Table 3- Although there are some technological differences between the pottery of Hoca Ces-me and Hamaylitarla, a basic similarity between the sites cannot be denied. Deep bowls with S curves, vertically placed tubular lugs, crescent-shaped lugs and bead rims constitute links between both sites. However, some of the forms and decorations at Hoca Cesme are absent in Hamaylitarla. On the other hand, Hamaylitarla pottery is slightly coarser than that of Hoca Cesme. The pottery of Hoca ge§me is elaborately made and the surfaces are lustrously burnished. It is not yet clear whether these differences in pottery are due to chronological factors (i.e. Hamaylitarla is earlier than Hoca ge§me), cultural differences (i.e. the site of Hamaylitarla belongs to the Fikirtepe Culture, while Hoca Qe§me does not) or social variation (i.e. Hamaylitarla is a manufacturing site occupied by craft specialists only). In addition, Ozdogan compares Hoca ge§-me to Western Anatolian sites (Ozdogan 1997). Above, I also compare Hamaylitarla to Western Anatolian sites, such as Tepeti-stu and Araptepe. © According to Ozdogan, Hoca ge§me is an Anatolian colony in Eastern Thrace (Ozdogan 1997; 1998. 450). The pottery, small finds, the lithic technology, and the domesticates are unmistakably of Central Anatolian origin' (Ozdogan 1997.26). There is a close similarity in the pottery between early Hoca Lab. No. Level 14C Age BP Cal BC (1cQ Bln-4609 IV Grn-19779 IV Grn-19355 IV Grn-19357 III Grn-19780 III Grn-19311 III Grn-19781 III Grn-19310 II Grn-19782 II Grn-19356 II Ref. Univ. of Washington. Table 3■ Hoca (!e§me dates. (Je§me and Hacilar IX-VI and Kurucay 11-13. According to Ozdogan, the lithic technology is said to have characteristic traits of the Central Anatolian cultures (Ozdogan 1997). Connections with Anatolia are also documented by figurines and pseudo-stamp seals {cf. Hacilar: Mellaart 1970.Fig. 187). An analysis of the animal bones of the lower levels at Hoca Cesme determined that all the animals were domesticated (Buitenhuis 1994). However, round building structures of the early Hoca gesme different from those in Central Anatolian settlements. In the course of our survey on the island of Gokceada, c. 20 km west of the coast of the Gelibolu Peninsula, we found a new early Neolithic site, Ugur-lu, with pottery exactly similar to Hoca ge§me (Fig. 6). Ugurlu is a low mound c. 900 m northeast of the village of Ugurlu in the western part of 7637±43 6473 (6459) 6439 7360±35 6233 (6224) 6110 7200±180 6229 (6056,6042,6028) 5845 7135+270 6234 (6005,6003,5994) 5728 6920±90 5886 (5787) 5718 6960±65 5955 (5838,5822,5809) 5734 6900±110 5886 (5741) 5665 6890±280 6019 (5734) 5535 6890±60 5837 (5734) 5718 6520±110 5609 (5478) 5369 Radiocarbon Calibration Program 2000 Fig. 5. Pottery from Hamaylitarla. the island. In Ugurlu, red and black slipped, well burnished, thin-walled sherds were found similar to those of Hoca Ces-me IV. Most of the forms are deep bowls with S profiles and bead-rims. There are a significant number of sherds with vertically placed tubular lugs, crescent-shaped lugs and ring-bases. There is also a piece of a zoomorphic vessel. In Ugurlu, red-black or light cream-red-black mottled sherds were also found as Hoca Cesme III type. However, absent at early Hoca Ces-me are very long vertically placed tubular lugs. These types of lugs were also found at Ayio Gala (Hoocl 1981; Fig. 6.13, 14), Coskuntepe and Tepeiistii (personal observation). Ugurlu indicates that more Hoca (Je§me-type settlements exist. It is obvious that without any detailed knowledge of Western Anatolia, it is very difficult to interprete Hoca AEGEAN SEA <3 Fig. 1. Location of Hamaylitarla and the other important neolithic sites in the region: 1. Hoca ^e§me; 2. Agagli; 3- Ya-rimburgaz; 4. Hamaylitarla; 5. Kaynarca; 6. Fi-kirtepe; 7. Iferenkdy; 8. Pendik; 9. Yalova-Gdztepe; 10. Tepetarla; 11. Mtislu (lepne; 12. Uyiicek; 13■ Tarafct; 14. Yilanlik; 15. Gavur-tarla; 16. Anzavurtepe; 17. (Mica; 18. Coskuntepe. According to our laboratory analyses the material of the polished stone axes*" was of a single type of rock: metabasite. The occurrence of the metamorphic outcrops on the slopes of Ganos Mountain was not abundant, despite some of the outcrops on a limited number of loci. On some 3 The site was already under protection by the local museum thanks to the efforts of M. Akif Isin, the director of this museum. 4 As this paper is not intended to discuss the petrologic analysis carried out on the samples obtained from the postulated quarry sites in the region, the author gives brief information on the matter. See Ozbek and Erol in press for the petrologic analysis. 5 See mainly Petrequin and Petrequin (1993) for a general assessment of ethnological studies in Indonesia and Petrequin et al. (1993) for a brief history of the recovery of the Neolithic quartz mudstone axe quarries (Plancher-les-Mines) in France. 6 The term 'axe' is used here in a general sense, putting the axes and the adzes in one group: wood-working implements, whether they are used for tree felling or in carpentry in the settlements. chance to saw the material. On the contrary, it is sound to say that pecking would take less time than sawing this rock. One can say at first glance that the forms of the roughouts (Figs. 3, 4) reflect a close resemblance to the Lower Paleolithic quartz hand axes in general. Axe roughouts Their weights vary from 400 grams to 1000 grams. Most of them are waste material, as they were probably broken during the production stage (Fig. 3). If we take into account their mean sizes, we can estimate that they were planned to be 20-25 centimeters when finished. Most of them should be regarded as roughouts of big adzes, rather than symmetrical axes. Their cutting edges were left untouched after being flaked, and never pecked. Fig. 2. Loca tion of the production sites in Gelibolu peninsula. parts of these inclination surfaces, one could easily notice the presence of metabasites as boulders, some weighing 1 kilogram and some 100 kilograms. The Hamaylitarla production site seems to cover an area of 400 m2, while the mound seems to spread over a 120 x 120 m field. We collected many roughouts, flakes, hammers and blocks from the locality, and following our laboratory analyses, we saw that they were of the same material. Hammers Their weights vary from 150 grams to 800 grams. They are usually broken axe roughouts, transferred to hammers after they were broken. We did not see any of the sphere shaped hammers we came across at the other sites. Flakes The flakes are 50 to 200 gram pieces, and very difficult to notice during the collecting of the material (Fig. 3-5, Fig. 4.1). The quantity of flakes collected on this mound is also too small to make a statistical analysis. 1 Hamaylitarla . - - "' Brooks As the mechanical properties of the stone played a big role in the production technique, what we observed was the intense practice of knapping and pecking. According to our experimental studies, we can say that with this raw material it is very difficult to orient any of the edges of an axe model. As the rock itself had no orientation in either the macro or micro (mineralogical) basis, it was also impossible to expect to obtain roughout blades right after the flaking process. The makers of these roughout axes also had no Fig. 3- The roughout manufacture of Hamaylitarla mound, Axe roughouts are broken during the production stage (No. 1, 3, 4). There are also a limited number offtakes (No. 5). SEA of MAKMAKA AEGEAN SEA is evident from the many studies held in different parts of the world with many different cultural con-texts7, the forms of axes do not change much with time. However, it is interesting to note that the finished axe material of the neighbouring excavated Neolithic site of Hoca Qe§me bears a close resemblance to the Hamaylitarla material, according to penological analysis. The 20-25 cm (1242 gr.) long forms unearthed from this prehistoric mound generally fit the planned models of the Hamaylitarla roughouts in weight and in shape. We expect that it will be possible to make a broader study of the region in the future. In addition, we will have more information about the paleogeography and geology of the Gelibolu Peninsula. This project will also increase our databases on the Neolithic settlements in the area. - ACKNOWLEDGEMENTS - The author wishes to express his thanks to the director of the Tekirdag Museum, M. Akif Isin for providing logistic support and for his active interest in our study. The survey team, headed by Onur Ozbek, were M. Akif Isin, Iht rcin Erdogu and Kenan Erot. Above all, our thanks are especially due to Marie-Claire Cauvin and Mehmet Ozdogan for encouraging the realization of this study. Fig. 4. The roughout manufacture of Hamaylitarla mound. Broken axe roughouts during production (No. 2,3). Some of the flakes ivere knapped (No. 1). CONCLUSION The dating of our material is one of the most important problems in this study. As we could not open test trenches or start an excavation on the site due to the lack of an excavation team and sponsor problems, we can not discuss the material in a stratigraphic context for the moment. However, in the near future, we await a multinational excavation on the site. We would like to express the fact that it is unwise to expect great help from typology in 'axe studies'. As 7 It is worth mentioning the studies by Buret (1983) of the Neolithic sites of Switzerland, and Moundrea-Agrafioti on the Neolithic sites of Greece (1981). REFERENCES BURET C. 1983- L'industrie de la pierre polie an Neolithique moyen et recent a Auvernier, canton de Neuchdtel (Suisse). Ph.D. Thesis, University of Paris X. MOUNDREA-AGRAFIOTI H. A. 1981. La Thessalie Du Sud-EstAu Neolithique: Outillage Lithique et Osseux. Ph.D. Thesis, University of Paris X. OKAY A. I. 1984. Distribution and characteristics of the northwest Turkish blueschists. In Robertson A. H. F. and Dixon J. E. (eds.), The geological evolution of the eastern Mediterranean: Geological Society of London Special Publication 17: 455-466. OKAY A. I, DEMIRBAG E„ KURT H, OKAY N. and KU-SCU 1.1999- An active, deep marine strike-slip basin along the North Anatolian fault in Turkey. Tectonics 18/1: 129. OZBEK 0. and EROL K. 2000. (in press) Etude pe-trographique des haches polies du Hamaylitarla et Fener-karadutlar (Turquie). Anatolia Antiqua IX (2001), De Boccard, Paris. OZDOGAN M. 1986. Prehistoric sites in the Gelibolu Peninsula. Anadolu Arastirmalari X: 51-66. PETREQUIN P. and PETREQUIN A.-M. 1993. Ecologie d'un Outil. La hache de pierre en Irian Jaya (Indo-nesie). Monographie du CRA 12. CNRS Editions. PETREQUIN P., JEUDY F. and JEUNESSE C. 1993. Neolithic quarries, the exchange of axes and social control in the southern Vosges. In E. C. Searre and F. Healy (eds.), Trade and Exchange in Prehistoric Europe. Proceedings of a conference held at the university of Bristol, April 1992: 45-60. UDK: 903(389-11)"634' Documenta Praehistorica XXVII The Late Starcevo and the Earliest Linear Pottery Groups in Western Transdanubia Eszter Banffy Archaeological Institute of the Hungarian Academy of Sciences, Budapest banffy@abklaw-patent.hu ABSTRACT - Although the Western Part of the Carpathian Basin, Transdanubia must have been one of the most important areas as neolithisation is concerned, research has failed to clarify some key factors. In this paper, possible traces of the Late Mesolithic forager groups are collected, indirect hints of the existence of a population that could have had an influence both on the northernmost limit area of the late Starcevo culture and on the formation of the oldest Transdanubian Linear Pottery Cidture. The hunter-gatherer groups are assumed to have controlled the prehistoric flint mine at Szentgal, Northern Transdanubia, supplying late Starcevo villages as well as early Linear Pottery settlements at a great distance with red radiolarite raiv material. Besides a new late Starcevo site at Ba-barc, found in 1997, the systematic excavations at Pityerdomb in Western Transdanubia are discussed in detail, which might be of essential importance in understanding the process of neolithisation around Lake Balaton, IZVLECEK - Cepravje zahodni del karpatskega bazetia (Transdanubija) eno najpomembnejsih po-drocij povezanih z neolitizacijo, pa dosedanje raziskave niso uspele pojasniti nekaterih kljucnih de-javnikov. V tern clanku smo zbrali mozne sledi o poznomezolitskih lovsko-nabiralniskih skupnostih, to je posredne namige o obstoju popidacije, ki hi lahko vplivala tako na skrajno severno mejo po-drocja kulture Starcevo, kot tudi na oblikovanje najstarejse transdanubijske kulture LTK. Predvide-va se, da so lovsko-mbiralniske skupnosti obvladovale prazgodovinski rudnik kremena v Szentgalu v severni Transdanubiji in so vasipozne kulture Starcevo kot tudi zgodnje naselbine LTK na velike razdalje oskrbovale s surovino - rdecim radiotaritom. Podrobno obravnavamo novo najdisce pozne kulture Starcevo v Babarcu, ki so ga odkrili leta 1997, in sistematicna izkopavanja v Pityerdombu v zahodni Transdanubiji, kar bi lahko bilo bistvenega pomena za razumevanje procesa neolitizacije v okolici Blatnega jezera. KEY WORDS - Transdanubia; neolithisation; boundary; new transitional site INTRODUCTION When regarding the neolithisation process of the Carpathian Basin (Fig. 1), we must face the old problem first which occurs in the research of each prehistoric period. Namely, the area east of the Tisza River has been traditionally much more investigated during the last century, than Transdanubia. The Neolithic heritage of the Alfold region has always been more spectacular, with the enormous number of rich settlements and find assemblages, not to speak of the earliest tell mounds and the rich grave goods in the Tisza region, This abundance of information is also valid for the earliest phase of the Neolithic. The intensive occupation of the river meanders of the Tisza, Koros and Berettyo by the Koros culture people also implies extremely rich find assemblages: one single settlement pit might contain several ten thousand pieces of pottery, even though the internal chronology of the Koros culture is still problematic. It might not be mere chance that the first thorough investigations concerning the late Mesolithic brought success precisely in the Northern Alfold, along the area of the Koros culture's northern limit, where a possible Mesolithic/Early Neolithic contact had been theoretically presumed earlier. The settlement of Jaszte-lek I., excavated and published by R. Kertesz, is the first real hint of the possibility of contacts between indigenous and newcomer groups also in the Carpathian Basin (Kertesz et al. 1994; Kertesz 1996). oldest linear pottery. oldest linear pottery k6r6s starcevo early v1nca The situation in the Western part, in Transdanubia, is far less understood. The term 'Early Neolithic' means here, as in the Alfold region, at least two integer phases: the life of the Starcevo culture in the Southern part and the formulation of the oldest Linear Pottery ware culture in the Northern part. Concerning the Mesolithic presence in Transdanubia, we are in a less advantageous position than in Eastern Hungary. That is, our knowledge is still based mainly on scattered surface finds, which, especially in the supposed later Mesolithic phase, may well belong to the earliest Neolithic find assemblages of destroyed settlement features, where the coarsely fired pottery had already diminished. According to colleagues with a good knowledge of flint typology, the finds from Kaposhomok, south of Lake Balaton could perhaps be attributed to Late Mesolithic groups (Pusz-tai 1957. Fig. 2). This can be assumed mainly from their geometric microlithic character, which can be ranged to the Tardenoisien, and is considered one of the latest Mesolithic assemblages in the area (Dobo-si 1972.41- 42). Be that as it may, there remains the great problem of how the Northern Transdanubian hills were settled during the late Mesolithic. If we take the numerous indirect hints of their existence into consideration, the problem becomes even greater, as we shall see below. THE LATE STARCEVO CULTURE In the Transdanubian Early Neolithic, the identification and description of the Starcevo settlements as well as the research into their chronological position versus the earliest TLP groups were the greatest steps forward. This was thanks to N. Kalicz and J. Makkay, who carried out several small sondages and field research in the late sixties, although their term 'Medina-phase' for the transition between the two cultures today belongs to the forgotten and outworn categories (Kalicz 1978-79b; Kalicz, Makkay 1972). Fig. 1. The distribution of the Koros-Starcevo and early Linear Pot tery cultures in the Carpathian Basin. As it is known from the detailed publications of N. Kalicz and also from his monograph published in 1990, the settlement pattern of the Starcevo culture cannot be compared to that of the Koros, as the former is far less intensive (Kalicz 1977-78; 1978-79a; 1983; 1990.39-40). This is reflected not only in the number of settlements, but also in their extent. Kalicz concluded by assuming a few smaller population groups who never stayed in one place for long. According to his observations, two important consequences can be drawn about the Starcevo culture: (a) it appeared from the South in the Linear B phase, according to the periodisation by Dimitrije-vic, and survived until the final, Spiraloid B phase of the culture, even though only four sites could be then dated to this phase; (b) its area distribution area reached only to the southern banks of Lake Balaton. In recent years, both statements have had to be corrected. That is, two new settlements of the Starcevo culture have been excavated: Gellenhaza-Varosret and Vors-Mariaasszonysziget (Simon 1994; 1996; Kalicz, Virdg, Biro 1998). Both belong to the latest phase on the one hand, and both lay on the northern borderland of the culture on the other. Moreover, the site at Gellenhaza-Varosret has modified this border by some 50 km to the north and west, as it lies near Zalaegerszeg, in the Zala hills in Western Transdanubia. So the number of Spiraloid B settlements has increased from 4 to 6. Two years ago a settlement of the same phase, Ba-barc (No. 10), came to light in South Eastern Transdanubia (Figs. 3, 4). The settlement pits of Babarc Fig. 2. Kaposliomok - lithic finds (after R. Pusztai, re-drawn by T. Marton). are again unimportant because of their large quantity, rather than because each new site might become a piece in the chain needed for a better understanding of the complicated processes within the early Neolithic (Bdnffy in press). However, some interesting problems emerge also on the basis of their dating. First, the site of Babarc lies next to the village of Lanycsok, where until now the most large-scale excavations of the culture have taken place (Kalicz 1977-1978). However, all the Lanycsok features, including the famous four-headed altarpiece can be dated to the so-called classical (Linear B, "Ghirlan-doid", Spiraloid A) phase, and none of them survived in the late phase. By that time they must have moved and built their new settlement somewhere in the vicinity. Thus, it is not impossible that in Babarc the heritage of the Lanycsok people was found, from a period of some generations later. The other point of interest is that, as has been said before, we do not know many late Starcevo settlements in Transdanubia. The four original ones lie to the north and west: Kaposvar-Deseda, Dombovar-Ka-pospart and Harc-Nyanyapuszta are located in South Eastern Transdanubia, but not far from the Southern banks of Lake Balaton, while Becsehely lies at the westernmost edge of Southern Transdanubia. Not to speak of the two newly found sites, Vors and Gellen-haza, which have even modified the northern and western distribution limits of the Starcevo distribution area. Now, on the basis of the new finds, two territorial groups of the late Starcevo culture might be drafted in Transdanubia. Babarc belongs to a group which is strictly bound to its southern relatives beyond the Drava River: its best parallels can be found in Croatian sites such as Podgorac or Vin-kovci-Gradska Zona (Minichreiter 1992b.43-49). Many typological features from Kaposvar and Dom-bovar still resemble this southern typed version of the late Starcevo. In contrast to the above stylistic and typological features, the two late Starcevo settlements in the Northwest, Vors-Mariaasszonysziget and Gellenhaza-Varosret, seem to belong to a slightly different group, with less evidence of direct Balkan contacts. As to the excavators, a number of these features become typical in the oldest Linear Pottery culture (e.g. deeply incised linear patterns), which occurs in an uncommonly high quantity, compared to the whole Starcevo area (Kalicz, Virag, Biro 1998. 163-164). Similarly, in Gellenhaza, the character of the pottery and some find groups strongly resemble those of the earliest Linear Pottery in the vicinity, as discussed below (Simon 1996; Bdnffy 2000.376). In the case of a cultural formation such as the Starcevo culture, which remained almost identical over a vast geographic area, from Macedonia to the Pan-nonian hills, these differences observed at the northwestern boundary cannot be neglected! However, the differences could be hard to analyse without a new settlement from the westernmost part of Hungary, which seems to be the first site of the transitional phase between the Starcevo and the earliest Transdanubian Linear Pottery Ware culture and thus, gives a new aspect to research into neolithisation in Western Transdanubia. PITYERDOMB In the course of our third micro-regional programme, a small plain along the upper flow of the River Kerka and the surrounding hills were investigated, close to the Slovenian border, north of the town of Lenti (Fig. 5). Among the sixty new sites ranging from the Neolithic to late medieval times, the early Neolithic period proved the greatest surprise. East of this, that is, a microregion which already been investigated (No. 2 in Fig. 5) was totally uninhabited by Linear pottery people. We explained this gap by the distribution area of the culture, and thought we had gone beyond it to the west. Remarkably enough, in the Kerka valley, even farther to the west, twelve of the sixty sites belong to Linear pottery culture. The settlement pattern must have been so dense up to this period that the inhabitants of one village at a hilltop could well catch sight of the next settlement built on a neighbouring hill. What is more, off-site evidence of agriculture and land use was distributed between all the Linear Pottery settlements, in the form of lost flint tools and small household refuse, probably taken with farmyard manure to the cultivated fields. Today, the region consists of small, cultivated areas and grazing land, surrounded and divided by large forests. This area lies at the foot of the Eastern Alps, so even the summers are humid and cool, unlike the average climate in Hungary. In spite of the continuous forests, the largest in Hungary, the first traces of the destruction of indigenous forests could be assigned to as early as the beginning of the West Transdanubian Neolithic, roughly the middle of the 6th millennium BC (Kertesz-Sumegi 1999■ 18). One of the settlements, Pityerdomb, near the modern village of Szentgyorgyvolgy, was excavated between 1995-1998, and provided important new data on the problem of neolithisation in Western Trans-danubia. The site is located on the top of the hill in the hilly area between the Zala hill-country and the Alps, which consists of old, acidous clay with a very thin layer of humus. The small settlements were con- centrated on the top of the hill and on its northwest- Fig. 3. New finds of the late Starcevo culture from Babarc. ern slope, at a height of some 220 m, not far from the Szentgyorgy creek, which flows at the foot. The highest area has been eroded. The excavated area, nearly 1000 m2, which covered almost the whole site - i.e. the part that has been preserved - brought the traces of two houses to light, lying a certain distance (some 30 meters) from each other (Fig. 6). Between the houses, archaeological features were almost totally absent. This probably means there were two focuses to the settlement. As we shall see below, on the basis of the finds, it was impossible to tell whether the two farming units differed in age. Nevertheless, according to measurements in other Central European Linear Pottery sites, the distance between the Pityerdomb houses does not differ from the average, and might be even less. In Langweiler 8, for example, the distance between two coeval houses was measured at not less than 66 m, while in Langweiler 9 almost double this distance could be measured! (Liming 1982.147-148; See also Coudart 1998.108). Fig. 4. New finds of the late Starcevo culture from Babarc. Thanks to the lack of deep ploughing, the features at 25-30 cm below the present surface remained undisturbed. The upper layer consisted of heavily burnt wattle and daub, which covered the rich finds: pottery and lithic material, as no bones survived in the acidous clay. Although the floor within the houses has not been burnt, in some places a certain walking level could be observed, not to speak of the finds which lay on this floor and helped identify it. Inside the houses we could also observe shallow pits, or rather small deepenings, in which the remains of large storage vessels were found (Fig. 7). Fireplaces were found both inside and outside the houses, but larger and deeper storage pits were always dug outside. A round pit, located to the southwest of House 2, (Feature Nr. 17) can probably be considered a workshop for flints. Both houses were of similar size. In spite of the imperfect circumstances, according to both the site location and financial means, the forms and the sizes can well be estimated: both between 8.50-10.0 times some 13.0-14.50 meters. This shape and size can be perhaps considered one of the most archaic types in Linear pottery architecture, where we do have the central part of later long houses, while the two outer rooms are missing. It is to be noted that some Ko-ros houses are known, e.g. one belonging to its late phase from the Middle Tisza region, which appear to be close parallels to the shorter Linear pottery typed houses in Pi-tyerdomb (e.g. Tiszajeno-Szarazer-part: Selmeczi 1969). Given the present state of our knowledge, it is hard to tell whether the northern Starcevo groups also built similar houses, since only pits or systems of pits containing workshops are known, while no dwellings have come to light as yet (Petrolic 1986-87; Minichreiter 1992a; 1992b. 11-38). All features were oriented precisely towards north. This was the case also with the most typical feature type in Pityerdomb: long ditches alongside the house's walls (Fig. 8). In the filling of these long pits the stratigraphy of the settlement could be well observed. The profiles showed that soon after having dug the pit, the lower part was buried quickly. Above this more-or-less sterile layer there is one layer with plenty of finds. Finally, in the course of a serious fire, the burning parts of the roof and the walls fell in and covered not only the area of the house itself, but also the ditches. The direction of their fall can still be seen, showing the in situ character of the assemblages (Fig. 9). The pottery was generally fired at a low temperature. Consequently, the profile of the wall is red-black-red: the usual characteristic of the early Neolithic throughout South East Europe. The organic tempering was almost always completed with sand. The vivid red colour sometimes occurs with dark greyish spots, similar to late Starcevo house ware in Northern Transdanubia. Black topped pottery is also typical: important chronological information. Pedestalled vessels occur in a larger number, belonging to two types: one is higher and conic, while the Pots and other storage vessels are often covered with 'Schlickwurf bar-botine type or with the barbotine arranged with fingers in lines or different patterns. Buckles and twin-buck-les are often finger-pressed on their top, or divided by cuttings. Linear patterns made with nail imprints are also a frequent form of coarse ware decoration. However, linear motifs, which can be three parallel lines or spiraloid, "voluted" motifs, are extremely rare. Although the pottery surface was considerably worn by the acidic soil, a fairly high ratio of thin-walled fine ware is clearly observable. As to their types, most are small bowls and mugs, both often carinated. The upper part of the vessel is quite frequently concave. Both the inner and outer surfaces of the fine ware are polished, wherever it these survived. It is to be noted that a high percentage of these vessels show a kind of highly polished wine-red slip, a most typical characteristic of the Balkan Karanovo I—II, the Koros-(]ri§ and the Starcevo cultures (Fig. 10). Another important type of decoration on the surface of fine vessel was preserved in better condition. This is a group of finely polished lines (einpolierte Ware), often in the form of concentric circles, semicircles or small lines on the corner point of biconic, carinated bowls. According to N. Kalicz, this decoration occurs only in Fig. 5. The three microregional research areas in Western Trans-danubia (1: Little Balaton area, 2: Hahot valley, 3: Kerka valley). Key site: Szentgyorgyvolgy-Pityerdomb. Fig. 6. The excavated area with two houses at Pityerdomb. the late Koros-Starcevo millieu and the earliest Trans-danubian Linear pottery culture (.Kalicz 1994.68; 1995). An important difference from other early LP settlements in Transdanubia is the ratio of linear decorated vessels. In contrast to the Bicske phase and also in the newly published site at Budapest-Arany-hegyi ut (Makkay 1978; Kalicz-Kalicz-Schreiber 1992; Kalicz 1995), in Pityerdomb linear, decoration occurs on no more than 0.5 percent of the whole pottery assemblage. Black painting was present in the Pityerdomb pottery. Given the worn surfaces, mentioned earlier, they can hardly be seen on the potsherds after they dry, but they were present on the wet finds, especially when they lay deeper than 50 cm below the present surface. In some cases, the spiraloid motifs could also be perceived, although usually in the form of an imprint in the soil visible after the lifting of the pottery fragment. There are two more indicators of the use of black paint: one voluted vessel was a container for black paint, with a thick layer inside and smudged traces of it appearing even on the outside (Fig. 11). Finally, a small clay leg, probably broken off from a four-legged vessel, was used other is quite low, resembling a foot-ring. Fig. 7. Pityerdomb, finds at the floor surface. secondarily as a pintadera: the same black paint covered its broken surface.1 Among finds that can be associated with cult life at the settlement there was a spiraloid, "voluted" fragment, which has a handle in the form of an upraised human hand. The whole pot, probably having two hands placed symmetrically on both sides, is probably an anthropomorphic vessel, typical of the earliest Linear pottery culture (Fig. 12). On the other hand, another fragment, representing a human leg, with a smoothed surface and delicate linear decoration, can be compared with legged human vessels from the Starcevo tradition (Mostonga I, II and Do-nja Branjevina, Karmanski 1977.Pl. 33; 1990.Pls. 1/1, 4/1-9; Circea, Nica 1977.Fig. 12/3; Ostrovul Golu, Lazarovici 1979.Pl, X/27). The most unique find from Pityerdomb is an almost intact clay figurine of a bovine-type animal, probably an ox (Fig. 13)- The finely elaborated, asymmetric linear decoration, together with the early Neoli- thic typological features and the wine-red, polished body surface might perhaps symbolise the formulation of the Centra] European "Neolithic type of thinking" in the context of early Neolithic Balkan traditions (Stankovic 1989-90; Ciobotaru 1998.Pls. 1/9, 10). The immense quantity of lithic finds all come from the prehistoric mine of Szentgal near Veszpre'm, in North Eastern Transdanubia (Fig. 14). The character of the assemblage is microlithic.2 The high number of cores and splits, as well as the feature south-west of house 2, probably a workshop for flints, suggests that flint must have been imported in the form of raw material from the Szentgal mine to Pityerdomb, where the chipped stone artefacts were prepared. The kit was probably made for each household. The same raw material, the red radiolarite, is typical also of coeval and somewhat younger sites of early LP settlements in Western Transdanubia. Similarly, the Szentgal radiolarite also occurs in the same period in Eastern Austria and Southern Germany. This Fig. 8. Pityerdomb, the long ditch along House 1. 1 The chemical analysis was partly made in the MAFI, Budapest, where the results speak about a certain organic, resin-like material, probably from a tree. A more detailed analysis was promised from Salzburg. 2 Oral communication by Katalin T. Biro. The lithic finds will be evaluated by her. domb. What is more, a series of assemblages found earlier, and partly published, can also be placed in this category (Fig. 15). In his contribution to the first Hungarian Topography volumes, N. Kalicz found and/or identified numerous surface finds coming from a well definable site or from small sondages, all of which he dated to the earliest phase of the TLP. The finds from Balatonszepezd, Revfiilop, Von-yarcvashegy, Sarmellek, Zalavar all lie along the northern bank of Lake Balaton or close to it. To the same group might belong the pit from Garabonc-ofalu, in the Little Balaton region, which was found and excavated later, in the course of our first Trans-danubian microregional programme. Fig. 9- Pityerdomb, burnt wattle and daub remnants fallen down, in situ. fact clearly shows that the Szentgal stone was a raw material of high value at the time of the earliest Neolithic period; its long distance trading had a strong influence on the direction and intensity of the neolithisation process in Central Europe. Along the Trans-danubian rivers the L-pottery traders reached the Danube and also regions such as Moravia and the Munich Basin. However, the L-pottery population was not the first to use this raw material in the region, as we shall see below. The detailed and final proceedings of the Pityerdomb assemblages have not been completed, like the finds of Gellenhaza and Vors. So all I am going to say now are not conclusions, but much more some tentative statements which must be confirmed or dismissed later. First, there is a strong resemblance between the pottery of Gellenhaza and that of Pityerdomb. This concerns the method of firing, the tempering and the surface of the vessels, the plastic and slightly incised decoration (einpolier-te Ware), the majority of the biconic, strongly carinated forms, the polished fine ware and also the coarse ware. What is more, there is a unique clay weight, having an amorphous pear form, which occurs in a greater number both in Gellenhaza and in Pityerdomb, but to my knowledge, nowhere else yet. This weight has a hole in each end, but is never completely perforated. These typological parallels as well as the similar geographical preferences suggest not only a possible synchronicity, but also live contacts between the late Starcevo people and the inhabitants of Pityer- The sites at Vorosto and Mencshely in the Southern Bakony mountains, near Szentgal are of special importance, because the lithic assemblages earlier found on the surface can well be ranged to the final Mesolithic, on the basis of to their microlithic Tarde-noisien character (Meszaros 1948; Dobosi 1912). Only after a small-scale rescue excavation did it become clear that at least a part of the flint show sickle glow, hence they must have belonged to some early Neolithic farming groups (T. Biro 1991.55). However, there is a slight contradiction between this latter statement and the Linear Pottery finds found together with sickle blades in the course of the rescue excavation, since the potsherds published belong for the most part to the developed, so-called 'Kesz-thely' phase of the culture (Regenye 1991). I cannot exclude an earlier population group at Mencshely and Vorosto, whose settlement traces were destroyed by the 'Keszthely' phase site. In my opinion, the question of whether the archaic types in the lithic assemblage belong to the 'Epialeolithic' or the early Fig. 10. Pityerdomb, dark red slipped and polished ware, with fine polished incisions. Fig. 11. Pityerdomb, a pintadera, containing black paint. Neolithic is almost irrelevant, since even in the early Linear Pottery community the use of exactly these types must have been adopted from local earlier, hunter-gatherer groups. This opinion might be supported by the evaluation by J. Regenye and K. T. Bird, as follows: 'A small part of the Meszaros collection is reminiscent, typologically, of Epipaleolithic forms as well. This impression would coincide with the existence of rich and very early LBC (Linear Pottery, E. B.) groups in the Balaton highlands, a region unusually rich in stone tools within Hungary.' (Biro-Regenye 1991.352). Finally, the finds from Tapolca might also belong here, including the altar-piece fragment known to the excavators as 'the wheat-eyed Goddess' (Sa-gi-Torocsik 1990; Eory-Sagi-Torocsik 1991). Indeed, the eyes of the human head are formed with two grains of triticnm dicoccum, a domesticated plant. this group, which seems to be an earliest Linear Pottery settlement. Here, some features such as the house designs resemble the Linear Pottery types, but the eponymous linear decoration on the pottery is missing (Lenneis 1995.14-16; Stadler 1999). This is not to exclude the notion that the differences in the two territorial late Starcevo groups and the transitional assemblages in the northern periphery respectively are caused by a certain presence of late Mesolithic (Mencshely-Vorosto?) groups there. This presence cannot be proven yet in a direct way, but the above-mentioned criteria suggest their participation in the neolithisation process north of the Starcevo distribution area. As is more or less known, especially since the research work carried out by D. Gronenborn, all the earliest Linear Pottery groups in Transdanubia, but also to a not inconsiderable extent thoss in Central Europe, appreciated and used the Transdanubian Szentgal raw material (Gronenborn 1994; 1999). The red radiolarite from the Bakony Mountains, in Northern Transdanubia must have become actually as precious in Western Hungary, and more than one thousand kilometres to the northwest, as the Tokaj obsidian for the Alfold region. As mentioned above, the approximately one and a half thousand flints and nuclei in Pityerdomb all come from the Szentgal mine, about 200 kilometres distant. The excavator, P. Stadler claims, this is also the case at the Brunn settlement, at an even greater distance from Szentgal. The people of the late Starcevo settlement at Gellenhaza also used the same raw material, and the same flints are found in those Transdanubian Starcevo settlements where lithic material is used. (See the article by K. T. Bird in Kalicz, Virag, Biro 1998.) Concerning this, it is important to mention that Szent- It is perhaps also possible to range the site of Brunn II, near Vienna, to Fig. 12. Pityerdomb, vessel with a human hand application. Fig. 13. Pityerdomb, zoomorphic figurine. gal lies far outside, to the north of any modified Starcevo area. Consequently, if the Starcevo people had access to this raw material, they must have known about it from a group of people about whom we hardly have any firm knowledge yet, apart from some guesses, as in the case of Mencshely and Voro-sto. We can consider that they ruled over the northern Bakony forests, conducted some sorts of of exchange with the Starcevo inhabitants and also, that they must have had an influence on the character of the earliest Linear Pottery culture. Thus, it is likely that traces of an important boundary can be identified along the Balaton coast and westwards in Trans-danubia. CONCLUSION To sum up, it is likely, that the Koros-Starcevo-Cri§ culture, i.e. all formations of the early Balkan Neolithic which reached the Carpathian Basin, developed according to geographic differences with a different rhythm, but meanwhile each of them had similar basic features in later life. As is known, the area which was most thoroughly investigated, but meanwhile provoked the greatest debate, was the Middle Tisza region in connection with the so-called 'Proto-Vinca' problem and the formation of the earliest ALP, the Szatmar II group. This debate was and still is certainly not independent of the uncertainties in the Starcevo-Vinca transition in the core area. The two most important opinions reflect totally antagonistic views: the first emphasises local development, without any external impact (Lekovic 1990); while the other regards the southern impulses as the most important element, not excluding even migrations (Garasanin 1979; Lichardus-Lichardus-Itten 1989-90). I regard the opinion of J. Chapman as one of the simplest and not independently, the most logical opinion of a somewhat combined model, which he formulated in his monograph some decades ago (Chapman 1981. 34). This model might also sound likely because it has numerous analogies within the prehistory of Europe. Last, but not least, I should like to draw attention to an ecological model published recently by R. Kertesz and P. Stimegi (Kertesz, Siimegi 1999). This model finally makes clear why the Koros and the Starcevo cultures stopped at a borderline which does not coincide with any natural geographical barriers, neither in the Szolnok-Berettyo line in the centre of the Tisza region, nor at or around Lake Balaton in Transdanubia. In their publication there is a map showing the northern limits of the common climatic, petrological and ground soil potentialities that together were necessary for Koros-Starcevo-typed food production (Kertesz, Siimegi 1999.18, Fig. 3). The line is not straight because of some environmental mosaics, but roughly it represents the real borders of the early Neolithic distri- Fig. 14. Pityerdomb, lithicfinds. Fig. 15- The late Mesolithic, Starcevo and Earliest Linear Pottery sites in Southern and Western Transdanubia. Sites mentioned in the text: 1. Szentgdl (prehistoric mine), 2. Kaposhotnok (probably late mesolithic), 3- Mencshely (late mesolithic, or this tradition), 4. Vorosto (late mesolithic, or this tradition), 5. Ld-nycsok (classical Starcevo), 6. Be-csehely (late Starcevo), 7. Harc-Nya-nyapuszta (late Starcevo), 8. Kapos-vdr-Deseda (late Starcevo), 9. Dom-bovdr-Kapospart (late Starcevo), 10. Babarc (late Starcevo), 11. Vors-Mdriaasszonysziget (late Starcevo), 12. Gellenhdza-Varosret (late Starcevo), 13• Medina (earliest LP), 14. Szentgydrgyvdlgy-Pityerdomb (earliest LP), 15- Balatonszepezd (earliest LP), 16. Revfulop (earliest LP), 17. Tapolca (earliest LP), 18. Vo-nyarcvashegy (earliest LP), 19. Sdr-mellek (earliest LP), 20. Zalavdr (earliest LP), 21. Garabonc Ofalu (earliest LP), 22. Kethely (earliest LP). km bution area in the Carpathian Basin. This line should represent the 'Central European agro-ecological barrier', north of which it was impossible to continue the southeast European mode of food production. Thus, the early Neolithic groups of southern origin slowed down and finally had to stop. This pause for breath might have given time to the local, indigenous Mesolithic groups living to the north of them in the contact zone to learn most of the Neolithic inventions without a total assimilation and absorption into the Koros-Starcevo civilisation. It might be too early to draw any important conclusions from the above model from the archaeological point of view. Nevertheless, it is not unlikely that the settlements of the Pityerdomb type reflect just this mode of neolithisation: the adoption of the Neolithic way of life whilst preserving some traces of the old values. In my opinion, this sketchy hypothesis will have a good test, which is different from the certainly necessary and useful natural scientific analyses. This would be an analysis of the few hints reflecting something of the way of thinking and the symbolism of the Starcevo and the Pityerdomb-typed early Linear Pottery archaeological material. It is to be hoped that this will help build a bridge between many of the problems associated with the neolithisation of the Carpathian Basin. * ■ x T - the Szentgal mine; - the Mesolithic sites; - the Lanycsok site; - the late Starcevo sites - the oldest Linear Pottery sites. REFERENCES BANFFY E. 2000. Neue Daten zur Entstehung der Bandkeramik. In S. Hiller, V. Nikolov (eds.), Karano-vo III. Beitrdge zum Neolithikum in Siidosteuropa: 375-382. in press. Neue Funde der Starcevo-Kultur in Trans-danubien. In F. Drasovean (ed.), Festschrift Ghe-orghe Lazarovici. BIRO K. and REGENYE J. 1991. Prehistoric workshop and exploitation site at Szentgal-Tuzkoveshegy. Acta. Arch. Hung 43:337-375. BIRO T. K. 1991- Mencshely-Murvagodrok koanyaga (The lithic assemblage from Mencshely-Murvagodrok). A Tapolcai Varosi Muzeum Kozlemenyei 2: 51-60. CHAPMAN J. C.1981. The Vinca culture of South East Europe. Studies in chronology, economy and society. BAR IS 1171-II. CIOBOTARU D. L. 1998. Plastica neolitica din aseza-rea de la Foeni-Salas (Jud. Timis) (Neolithic artistic and cult artifacts from the Foeni-Salas site /Timis County/). Analele Banatului 6: 73-82. COUDART A. 1998. Architecture etsociete neolithi-que. L'unite et la variance de la maison danubi-enne. Documents d'Archeologie Francaise. Paris DOBOS1 T. V. 1972. Mesolithische Fundorte in Un-garn. Aktuelle Fragen der Bandkeramik. Szekesfe-hervdr: 39-60. EORY B, SAGI K, TOROCSIK Z. 1991. A Dunantuli Vonaldiszes Keramia „Tapocai csoportjanak" legu-jabb lelohelyei a Balaton kornyekerol (New sites of the „Tapoca group" of the Transdanubian Linear Pottery ware culture around the Balaton). A Tapocai Varosi Muzeum Kozlemenyei 2: 7-30. GARASANIN M. 1979- Starcevacka grupa. Praistorija Jugoslavenskih zemalja II. Sarajevo: 115-143- GRONENBORN D. 1994. Uberlegungen zur Ausbrei-tung der bauerlichen Wirtschaft in Mitteleuropa -Versuch einer kulturhistorischen Interpretation al-testbandkeramischer Silexinventare. Praehistorische Zeitschrift 69/2:135-151- 1999. A variation on a basic theme: The transition to farming in Southern Central Europe. Journal of World Prehistory 13/2:123-210. KALICZ N. 1977-78. Friih- und spatneolithische Fun-de in der Gemarkung des Ortes Lanycsok (Vorbe-richt).JPME 22:137-158. 1978_79a. Ausgrabungen: Becsehely. Mitt. Arch. Inst. 8-9: 119-120. 1978_79b. Funde der altesten Phase der Linien-bandkeramik in Stidtransdanubien. Mitt. Al 8/9: 13-46. 1983. Die Koros-Starcevo-Kulturen und ihre Be-ziehungen zur Linearbandkeramik. Nachr. Nieder-sachsens Urgesch. 52:91-130. 1990. Fruhneolithische Siedlungsfunde aus Siid-westungarn. IPH Budapest 1994. A Dunantuli (kozep-europai) vonaldiszes keramia legidosebb leletei es a korai Vinca-kultura (Die altesten Funde der transdanubischen (mittel-europaischen Linienbandkeramik und die fruhe Vinca-Kultur). A kokortol a kozepkorig - Von der Steinzeit bis zum Mittelalter. Festschrift 0. Trog-mayer: 67-84. 1995. Die Alteste transdanubische (mitteleuropa-ische) Linienbandkeramik. Aspekte zur Ursprung, Chronologie und Beziehungen. Acta. Arch, Hung. 47: 23-59. KALICZ N. and KALICZ-SCHREIBER R. 1992. Die erste neolithische Fundstelle in Budapest. Hommage a Nikola Tasic. Balcanica 23: 47-76. KALICZ N. and MAKKAY J. 1972. Siidliche Einflufte im friihen und mittleren Neolithikums Transdanubi-ens. Alba Regia 12: 93-105- KALICZ N, M. VIRAG ZS., BIRO T. K. 1998. The northern periphery of the early neolithic Starcevo culture in south-western Hungary: a case study of an excavation at Lake Balaton. Documenta Praehistorica 25: 151-187. KARMANSKI S. 1977. Katalog antropomorfne i zoo-morfne plastike iz okoline Odzaka. Odzaci KERTESZ R. 1996. The Mesolithic in the Great Hungarian Plain: A survey of evidence. In At the fringes of three worlds. Hunters-gatherers in the Middle Tisza valley. Szolnok 1996:5-39. KERTESZ ET AL. 1994. KERTESZ R., SUMEGI P., KO-ZAK M, BRAUN M, FELEGYHAZI E. 1994. Mesolithi-kum im nordlichen Teil der ungarischen Tiefebene. JAME 36: 15-61. KERTESZ R. and SUMEGI P. 1999- Teoriak, kritika es egy modell: miert allt meg a Koros-Starcevo kultura terjedese a Karpat-medence centrumaban? (Theories, critiques and a model: why did the expansion of the Koros-Starcevo culture stop in the centre of the Carpathian Basin? Tisicum 11: 9-23- MAKKAY J. 1978. Excavations at Bicske. I. The Early Neolithic - The earliest Linear band Ceramic. Alba Regia 16:9-60. LAZAROVICI G. 1979- Neoliticul Banatului. Cluj-Na-poca. LEKOVIC V. 1990. The Vincsanization of Starcevo culture. In D. Srejovic, N. Tasic (eds.), Vinca and its world: 67-74. LENNEIS E. 1995. Altneolithikum: Die Bandkeramik. Jungsteinzeit im Osten Osterreichs. St. Polten: 11-56. LICHARDUS J., LICHARDUS-ITTEN M. 1989-90. Der Komplex mit Schwarz-, braun und graupolierter Ke-ramik und der Beginn des Mittelneolithikums in Siidosteuropa. Starinar N.S. 40-41: 43-49. LUNING J. 1982. Forschungen zur bandkeramischen Besiedlung der Aldenhovener Platte im Rheinland. In J. Pavuk (ed.), Siedlungen der Kultur mit Linear-keramik in Europa. Kolloquium Nove Vozokany: 125-156. MESZAROS G. 1948. A Vazsonyi-medence mezolit-es neolitkori telepulesei. (The mesolithic and neolithic settlements in the Vazsony Basin). Veszprem MINICHREITER C. 1992a. Ranoeneoliticka arhitektu-ra sjeverne Hrvatske (Fruhneolithische Architektur Nordkroatiens). Porocilo 0 raziskovanju paleoliti-ka, neolitika in eneolitika v Sloveniji 20:17-26. 1992b. Starcevacka kultura u Sjevernoj Hrvat-skoj. Zagreb. NICA M. 1977. Nouvelles donnees sur le neolithique ancien d'Oltenie. Dacia N.S. 21: 13-54. PETROVIC J. 1986-87. Zemunica u naselju starcevac-ke kulture na Golokutu (Erdhiitten in der Siedlung der Starcevackaer Kultur auf Gologut). Rad Vojvo-danskog Muzeja 30:13-28. PUSZTAI R. 1957. Mezolitikus leletek Somogybol. JPME: 95-105. REGENYE J. 1991. Neolitikus leletek Mencshely ko-rnyekerol (Neolithic finds from Mencshely). A Tapol-cai Vdrosi Muzeum Kozlemenyei 2:31-47. SAGI K, TOROCSIK Z. 1990. A Dunantuli Vonaldf-szes Keramia Japolcai csoportja" (The „Tapolca group" of the Transdanubian Linear Pottery culture). A Tapolcai Vdrosi Muzeum Kozlemenyei 1:29-72. SELMECZI L. 1969. Das Wohnhaus der Koros-Gruppe von Tiszajeno. Neuere Angaben zu den Haustypen des Fruhneolithikums. MFME 1969/2: 17-22. SIMON H. K. 1994. Fruhneolithische Kultgegenstande bei Gellenhaza, Kom. Zala. In G. Lorinczy (ed.), A M-kortol a kozepkorig. Von der Steinzeit bis zum Mit-telalter. Studien zum Geburtstag von 0. Trogma-yer: 53-65. 1996. Ein neuer Fundort der Starcevo-Kultur bei Gellenhaza (Kom. Zala, Ungarn) und seine siidli-che Beziehungen. The Vinca culture, its role and cultural connections: 59-92. STADLER P. 1999- Die altestlinearbandkeramische Fundstelle von Brunn am Gebirge, Flur Wolfholz (5620-5200 v. Chr.). Fuhrer zur Austellung in Brunn im August 1999: 1-13■ STANKOVIC S. 1989-90. Bull representations in the Early Neolithic. Starinar N.S. 40-41:35-42. UDK: 903.2:902.65(4-924.57)"633/634' Documenta Praehistorica XXVII Measuring Change: the Neolithic Pottery Sequence of Vinca-Belo Brdo Wolfram Schier Institut fur Archaologie, Universitat Wurzburg, Germany wolfram.schier@mail.uni-wuerzburg.de ABSTRACT - This paper examines the pottery sequence of the famous neolithic tell site ofVinca, Yugoslavia. In earlier research its stratigraphy of 9 m had been considered to be of only limited chronological value, since the excavator failed to document the position offinds in regard to actual layers, and recorded only the respective spit of the excavation. Theoretical considerations and a statistical analysis attempt to evaluate which chronological resolution can be achieved using the old find material. Recent radiocarbon dating of stratified samples from the old excavation is used to establish an absolute time scale for the stratigraphy. The stratigraphic information is used to reduce the statistical error of these dates caused by the calibration process. Finally, the quantitative change in pottery type frequencies is compared with the interpolated rate of sedimentation. This comparison makes it possible to distinguish between innovation or stagnation phases in the production ofpottery and differences in the accumulation rate of settlement debris. IZVLECEK - V clanku se ukvarjamo s keramiko iz znamenitega neolitskega tel najdisca Vinca v Jugoslaviji. V zgodnejsih raziskavah so smatrali, da ima 9-metrska stratigrafija najdisca le ome-jeno kronolosko vrednost, saj izkopovalci niso dokumentirali lege najdb glede na dejanske ptasti, ampak so zapisovali le odgovarjajoce reznje. S teoreticnim razglabljanjem in statisticnimi anal-izami poskusamo oceniti, kaksno kronolosko locljivost lahko dosezemo z najdbami starih izkopa-vanj. Z novejso radiokarbonsko datacijo stratiftciranih vzorcev iz starih izkopavanj smo izdelali absolutno casovno skalo stratigrafije. Stratigrafske podatke uporabljamo, da zmanjsamo sta-tisticno napako teh datacij, ki jih povzroca kalibracijski proces. In koncno, kvantitativno spre-membo v pogostosti tipa keramike primerjamo z interpolirano stopnjo sedimentacije. Ta primer-java omogoca locevanje med inovativnimi in mirujocimi fazami v proizvodnji keramike ter med razlikami v stopnji akumulacije crepinj v naselbini. KEY WORDS - Vinca; Neolithic; stratigraphy; seriation; radiocarbon dating THE SITE The tell site ofVinca Belo Brdo near Belgrade (Yugoslavia) is among the best-known archaeological sites of south-eastern Europe. Not only has it become the type-site of the Vinca culture, with its approximately 9 metres of cultural deposit, it also scores among the longest stratigraphic sequences of the European Neolithic. Since the excavator, M. M. Vasic, published his four volume Praistorijska Vinca between 1932 and 1936 (Vasic 1932; 1936a; 1936b; 1936c), Vinca has remained not only a key point of reference for the research on Balkan Neolithic, but also an object of controversial debate. The publication of Vasic, while being well ahead of its time with regard to the thoroughness of the documentation and classification of finds, failed to describe with equivalent accuracy the structural remains uncovered during the excavation. The excavation method consisted in removing horizontal levels of 10 cm thickness and marking on most finds the vertical distance from an arbitrarily chosen zero point. However, none of the finds and only a few observed houses were recorded in their horizontal position. Thus subsequent research concentrated mainly on chronological and typological analyses of the find material (Holste 1939; Milojcic 1943; 1949a; 1949b; Garasanin 1951; 1979), whereas few studies attempted to reconstruct the position of structures and the sequence of building phases (Korosec 1953; Jovanovic I960; 1984; Chapman 1981; Stalio 1984). By grouping the artificial levels of the type-site in intervals of 1-1.5 in, several phasing systems have been developed which were gradually regarded and used as being valid for the whole area of the Vinca culture. The two main chronological systems of M. Garasanin (1951; 1979) and V. Milojcic (1949) concur in their major stratigraphic divisions at v 8.0, 6.0 and 4.1 m, while minor subdivisions are assumed at v 7.5, 7.0 and/or 6.5 m respectively. The type spectrum in their periodisations comprises almost exclusively material published by M. Vasic (Vasic 1932-36), but neglects his personal selection of "relevant" types and artefact attributes for publication. The diagnostic types of Vinca culture in the Garasanin and Milojcic system thus appear as a rather arbitrary selection from the bulk of restored vessels from the type-site. For their selection, neither absolute frequency nor stratigraphic distribution are specified. Especially in a stratigraphic sequence of contested reliability, as in Vinca-Belo Brdo, conclusions based on single occurrences of types and elements may be quite misleading if their overall distributional characteristics have not been analysed beforehand. a _________________ A systematic quantitative study of pottery types and elements in their stratigraphic distribution thus appeared a promising approach for reassessing the chronological validity of the Vinca sequence. On more general methodological grounds the aim was to test which chronological "resolution" can be achieved in a stratigraphic analysis restricted to given 10 cm-levels without reference to the real sedimentation and construction layers. REAL AND ARTIFICIAL STRATIGRAPHY A model was developed to analyse theoretically the distorting effects of a find recording system in regular vertical steps superposed on a realistic settlement stratigraphy (Figs, la, lb). The unknown percentage of material that differs in age from the bulk of finds in each 10 cm-level is defined in the present study as "stratigraphic contamination". Theoretically, we can distinguish two types of contamination: the first is caused by the unintentional overlapping of adjoining layers (Fig. lb: A, B), the second by intrusions of far later (unrecognised deep pits) or far older artefacts, which result, for example, from finds in sediment re-used as building material (Fig. lb: pit from layer 2 in level C). Obviously, the remaining chrono- i stratigraphic contamination of 1st kind B stratigraphic contamination of 2nd kind Fig. 1. Idealised tell stratigraphy (a) and enlarged section (b), showing possible superposition of layers and excavation units. b logical value of a contaminated stratigraphy depends on both kind and quantity of contamination. Figure 2 illustrates the likely effects of stratigraphic contamination on an idealised vertical distribution pattern of diagnostic types. Contaminations of the first kind will result in a flattening or shifting of uni-modal distributions of the types concerned (Fig. 2b). A certain amount of intrusive sherds (contamination of the 2nd kind, Fig. 2c) will show up as a bimodal or even polymodal stratigraphic distribution of some types. Beyond a certain critical threshold contaminations of both degrees will distort the unimodal type distribution patterns up to a total loss of stratigraphic information - the state of chronological entropy. Inversely, a pattern of overlapping unimodal type distributions will only be detected in a given stratigraphy when there is comparatively little contamination. STATISTICAL ANALYSIS OF STRATIGRAPHIC UNITS Since the same model of overlapping unimodal frequencies is the methodological basis of seriation techniques, our approach applies seriation as a tool for testing the stratigraphy of Vinca-Belo Brdo. However, unlike its usual applications, seriation is used here in reverse: instead of chronologically ordering a number of closed find units, we examine the chronological "closedness" of a number of find complexes, the sequence of which is predetermined stra-tigraphically. The seriation technique chosen is correspondence analysis (CA), a powerful statistical tool (Greenacre 1989; Madsen 1988; Djindjian 1991.181-186) sensitive in the detection of distorting factors in a generally seriable data matrix. Before being submitted to seriation, however, the pottery sample of Vinca-Belo Brdo required reclassification and further statistical pre-treatment. A sample of about 3400 pottery fragments served for a new classification of vessel shapes, decoration and handle types. Among the most numerous vessel category of bowls, amounting to 80 % of all fragments, 180 types could be distinguished, organized into 23 type groups. It is well known that archaeological types are artificial groupings not (necessarily) corresponding to functional or aesthetic classes in the potter's mind. Such a highly differentiated classification system, as developed here, intends to resolve a morphological continuum in the smallest discrete entities attainable in order to reproduce quantitatively the time scale in terms of pottery change. However, a classification should allow for an average type frequency that is still statistically meaningful. The types of pottery shape which are used in this study show a minimum frequency of 5 and an aver- frequency of pottery types idealised type distribution in original (unknown) stratigraphic sequence frequency of pottery types expected effects of 1st kind stratigraphic contamination on type distribution frequency of pottery types expected effects of 2nd kind stratigraphic contamination on type distribution Fig- 2. Effects of stratigraphic contamination on vertical type distribution. age of 10, with many of the chronologically most significant types exceeding 20 occurrences. A sub-sample of 950 vessel fragments was subjected to a metric analysis: the coordinates of crucial profile points and the length, angle and curvature of the intermediate profile sections were recorded. On the basis of these measurements and derived proportions a number of cluster, variance and discriminant analyses were performed to verify the visual classification of the whole sample. Vessel decoration was classified, separating its components into (1) decorative technique, (2) decorative motif/pattern, (3) composition of one or several decorations (in zones, repetitive, alternating, limiting) and (4) position of the decoration on the vessel (rim, shoulder, lower part, interior). All four components could be shown to vary throughout the stratigraphic sequence, i.e. to have some chronological value. As a further step, the analysis of stratigraphic frequency distribution provided information for the assessment of the chronological significance of shape types and decorative elements. Obviously, not every type or variant of vessel shape or decoration can be expected to show significant chronological variability. Many formal or technical details of pottery production may vary for functional or simply individual reasons. Even in a schematic stratigraphy, as in Vinca, the vertical distribution of types and elements can be regarded as a coarse indication of their variability in time. In order to minimize the effects of stratigraphic contamination types with obvious bi- or polymodal distribution were excluded. Only those shape types were included in the CA, whose interquartile range (central part of stratigraphic distribution without lowermost and uppermost 25% of occurences) did not exceed 1.4 m, a value that was chosen on the basis of dispersion diagrams. A similar selection was performed on decorative motifs and techniques, where only few types passed the interquartile criterion. In general, morphological variability of Vinca pottery proved to be a far better indicator of time than decorative variability. The seriation of a stratigraphic sequence can be visualised as the attempt to sort a number of find boxes (each containing the material of a single stratification unit) whose identification labels have been lost. Such an experiment can only be successful if (a) the stratigraphy encompasses sufficient time to allow for substantial change in artefact types, (b) the type classification is detailed enough to reveal slight and gradual change, (c) the sample size is sufficient to enable differences in the type percentages to attain statistical significance, (d) the chronological variation in the given units is considerably smaller than between them. In other words, they should show a sufficient degree of chronological homogeneity. The seriation matrix used in the present study measured the relative frequency of 204 pottery types and decorative elements in 39 stratigraphic units. These units comprise pottery samples of the excavated 10 cm levels from v 9-3 to 5.0 m, some of which in our analysis had to be paired or grouped to compensate for the small sample sizes. A number of pits discovered in the lowermost horizon at Vinca were also included, which rather traditionally than convincingly have been interpreted as semi-subterranean dwellings (Korosec 1953-40; Stalio 1984.34-36). Figure 3 shows the relative frequency of shape and decoration types in the stratigraphic units as ordered by the first eigenvector of correspondence analysis. Not only is a general pattern of shifting stratigraphical distributions obvious, but there are also differences in their specific range. Decoration techniques or motifs (M..., DT...) have much longer lifespans than pottery shapes (S..., A.., F.., K.., T..). While the central part of the frequency distribution is well ordered in a diagonal way, some extremely early or late isolated occurrences suggest a certain amount of stratigraphic contamination of second degree. Figure 4 represents the distribution of unit scores in the plane of the first two eigenvectors of CA. The diagram shows a fairly symmetrical arrangement of the units in the shape of a parabola, a statistical indication of a data matrix that can be diagonalised very well (Greenacre 1989.226-231). Obviously, a pattern of overlapping unimodal type frequency distributions can be discovered in the sequence of 10 cm levels at Vinca. Looking more closely we see that the seriation sequence starts in the lower left corner with pit Z (an almost pure complex of the preceding Starcevo culture), followed by a group of other pit contents. Very closely clustered are the lowermost levels above the sterile loess subsoil from v 9.3 to Y 9.0 m. Projected on the first eigenvector which we interpret as time scale there is almost no chronological difference. Separated by a gap, the pits W, M and K are grouped around the (reversed) levels v 8.9 and v 8.7 m. After another interruption the levels from v 8.5 to v 8.0 m appear in perfect stratigraphic order, while the levels from v 7.8 to v 7.1 m cluster in two Fig. 3. Abundance matrix of pottery types and attributes in stratigraphic units, seriated by Correspondence Analysis. groups rather than in stratigraphic sequence. From v 7.1 m up to v 5.0 m the stratigraphic units are once again arranged by CA in correct order, with the exception of level v 6.2, being slightly misplaced. Apart from the levels between v 7.8 and v 7.1 m, which will be discussed below, the given stratigraphic sequence of Vinca Belo Brdo can thus be reproduced statistically on the basis of type combinations only. Furthermore, the ability of CA to space data points in a two-dimensional plane according to their similarity/ dissimilarity allows borderlines between cultural phases to be defined empirically rather than to be drawn arbitrarily. Thus we can distinguish phases 1 to 7 in Vinca with a threefold subdivision of phase 5 (a, b, c) and a possible subdivision of phase 2 (Fig. 4). These supposed phase boundaries derived from gaps in the eigenvector plot were subjected to further statistical testing. A one-way analysis of variance confirmed that there are significant differences in type frequency between all of the phases except 2a and 2b. These two sub-phases, however, could be separated by a discriminant analysis, which reproduced the given groups (stratigraphic units combined to phases) perfectly (100% correctly classified), using three canonical functions calculated from type frequencies, Since stratigraphy and type seriation represent functions of real time, we may compare them in a cor- 2.0 1 .O O.O -1 O -'2.04 -3.0 Phase 5a Bp- Phase 3 Phase 4 ^ 8,0 U> 8.1 8,4 P 8,3 ► \ 8.5 \ \ \W Phase 2b RV9V ► 9,1 9.0 o9,2 ^9,3 Phase 2a B A + 9,3 Phase 5b \ \ 7,6 ► \ \ 7-6* J?-7-3 6,9, 7.7 ► ; 7f7 5 ► |>( 7,0 O ► 6,6 Phase 5c ►6,5 ► 6,4 ► 6,1 6.2P ► 6,0 z Phase 1 B> 5,5-5,9 > 5,0-5,4 Phase 7 ■ i .5 -i .0 -0.5 0.0 0.5 l .0 l .5 2.0 ► Content of pit ► 10 cm level [•> Combined 10 cm levels Fig. 4. Stratigraphic units in the plane of the first tivo eigenvectors of CA. relation diagram (Fig. 5). A polynomial regression curve was calculated which suggests a very high nonlinear correlation (r = 0.99) between stratigraphic position and similarity in type composition expressed by the first eigenvector of CA. Especially interesting are the residuals, i.e. the points lying at a greater distance from the regression curve. In the lower left corner of Figure 5, three groups of pits (shaded triangles) are discernible: the pits at y 9-3 and y 9-2 m as well as A and B according to their type composition appear earlier than their stratigraphic position would suggest. The pits W, M and K, despite being discovered at almost the same depth, contain considerably later types. An intermediate position is assigned by the first eigenvector to pits T and R. Statistical analysis thus suggests that the pits in the lowest level of Vinca-Belo Brdo do not form a chronologically homogenous "pit horizon" as was assumed hitherto. Instead, several pits were probably dug from different levels (accordingly with different fill), which the excavator presumably recognised only when the lighter sterile loess subsoil had been reached (between y 9-0 and y 9-4 m). The correlation diagram also shows several adjoining stratigraphic units which are indistinguishable on the basis of their type contents and therefore deviate from the regression curve: the levels of y 9.2-9-0 m, y 7.8-7.7 m, v 7.5- 7.1 m and v 6.8-6.6 m in Figure 5 appear piled up, instead of being spread along the regression curve. Several conclusions can be drawn from these results. Firstly, the chronological resolution of Vasic's schematic, 10 cm levels is far clearer than generally assumed; in large patches of the stratigraphy even a difference in depth of only 10 cm is chronologically meaningful. Secondly, the contamination effects caused by unrecognised pits and/or the crosscutting of sedimentation layers which do not run horizontally (cf. Fig lb) cannot be considered serious enough to discredit the whole stratigraphic sequence. Thirdly, when based on CA and statistically tested, the grouping of stratigraphic units into phases can be regarded as methodologically sound. CULTURAL DYNAMICS AND THE GROWTH OF TELL SETTLEMENTS Chronological blocks based on a combination of stratigraphy and typology can thus be resolved into a pattern of gradually shifting type composition by seriating the artificial stratigraphic units. Since correspondence analysis can not only sort find complexes on an ordinal scale, but reproduces geometrically in few dimensions the overall (dis)similarity pattern of the units (and types), we can attempt to analyse a problem which is crucial to many archaeological find sequences: the problem of cultural dynamics. The refined chronological resolution enables a quantification of the change in find composition between adjoining units on a stratigraphic scale. This stratigraphic scale, however, is proportional to the real time scale only if we have evidence to assume an uninterrupted continuous accumulation of settlement debris at a constant rate. Unless tell accumulation can be shown to be a linear function of time, differential change between adjoining stratigraphic units can always be interpreted in an ambiguous way: Greater dissimilarity can be caused by cultural innovation, or by lower accumulation rate. Inversely, greater similarity in the type composition of two layers can result from cultural stagnation or from in- creased accumulation of sediment. The only way to resolve this ambiguity is to establish an independent time scale by means of absolute dating. In contrast to its central importance for relative chronology, until now the type site contributed very little to the absolute dating of Vinca culture (Breu-nig 1987.107; Todorovic and Cermanovic 1961. 101-102). In 1991, a number of unworked bone and antler finds with documented stratigraphic positions were subjected to radiocarbon analysis. A promising, mathematically sophisticated approach has recently been published as an attempt to reduce the additional statistical error caused by the calibration process (Buck, etal 1991; Buck, Litton, Smith 1992). Baye-sian probability theory is used to incorporate archaeological information in the calibration procedure. An application of this calibration approach, using the program OXCAL (Ramsey 1994), appears in Figure 6, in which 13 radiocarbon dates are arranged in reverse stratigraphic order. The open areas signify the probability distributions of all samples calibrated independently. The comparatively broad spread of their estimated ages results not from the measurement procedure itself, but from marked wiggles in the calibration curve between 5300 and 5000 cal BC. The application of a posteriori probabilities means Q. (D -o CO o -C Q. TO i_ O) 03 •+—> C/3 -5 -5.5 -6 -6.5 --7 --7.5 -8 -8.5 -9 -9.5 -10 -10.5 5,4-5,0T, T / 5,9-5,5 7.4 ro 7.5 - V, 76 7.9 8.0- 9.0- --L I phase 7 __1 phase 6 I . - I 1____I "L The rate of change in pottery composition is expressed by the difference between adjoining stratigraphic units in their first component score of correspondence analysis. For this purpose the same matrix of type frequencies as in Figure 3 was submitted to a detrended CA (Greenacre 1989.232) in order to express the "typological distance" by one vector only. The shaded histogram in Figure 8 represents the degree of dissimilarity between neighbouring stratigraphic units. Negative values signify that stratigraphic units have been reversed as a result of the seriation. Once again, the coincidence of the phase boundaries with peaks of typological dissimilarity is obvious. The central issue of Figure 8, however, is the correlation between sedimentation rate and typological change, which leads us once again to the question of cultural dynamics and/or accelerated settlement growth. Starting from the bottom, we can see a coincidence of slow sedimentation (according to dating model 2) with a high rate of pottery change between.Y 8.7 and 8.5 m. Increased sedimentation above this level correlates with decreasing change in the type assemblages until y 8.1 m. Between y 8.0 and 7.8 m, the highest degree of dissimilarity of all adjoining levels can be observed, followed by a marked drop. For the overlying 70 cm of tell accumulation, only small dif- phase 5c phase 5b phase 5a I phase 4 n phase 3 phase 2b phase 2a ■, T, T-, 1", 1-, p r-> ,-r it |-r 70 mm average sedimentation rate (mm/year) sequence calibration model 2 (10 dates) sequence calibration model 1 (13 dates) Fig. 8. Correlation of interpolated sedimentation rate and typological change. ferences between neighbouring levels occur, which made it difficult for the seriation to sort these units in the correct order with respect to their type composition. Above Y 7.1 m, a similar pattern reappears as below y 8.0 m: low sedimentation correlates with higher rates of change. Accelerated sedimentation, occurring around y 6.4 m, is met by a peak in typological change. Three basic patterns can be observed in Figure 8: the rate of sedimentation and typological change between stratigraphic units can be negatively correlated as between y 8.4 m or around y 7.0 m. Such an inverse relation would be considered normal, as artificial stratification units from slowly sedimented layers comprise more time and therefore greater dif- ferences in type composition. Between V 8.4 and v 7.1 m sedimentation and typological change are uncorrelated: in the lower part, moderate change coincides with a constantly high sedimentation rate. Between v 8.0 and v 7.8 m sedimentation remains almost constant, the exceptional dissimilarity in the type composition of these two levels can thus only be explained as an innovation horizon. It coincides with the end of the first building phase, represented by 6 houses discovered between v 8.3 and v 7.9 m whose horizontal position is unknown (Stalio 1968; 1984.35-37). The overlying 70 cm of sediment appear to have been accumulated at still the same speed as before, but there was little change. The type composition remains fairly homogenous and the stratification units are therefore difficult to distinguish statistically. So apparently this period has to be interpreted as a stagnation phase rather than as compressed time caused by rapid sedimentation. Even more difficult to interpret is the pattern above v 6.6 m. Obviously, there is a certain amount of innovation occurring around v 6.5 m, the effect of which is, however, counterbalanced by increased sedimentation above v 6.4 m. The interpolation of settlement growth is probably not accurate enough here, because it is based on only two radiocarbon dates from the levels v 6.4 and v 4.1 m. On the other hand, however, some of the few known and published house inventories come from the levels v 6.8 to 6.5 m, and during the following 50 cm until v 6.0 m a major replacement of pottery types is well documented (cf. Garasanin 1979.150-152; 1973-95-96; 199313-15). This could plausibly have resulted from a destruction and levelling horizon after the end of the settlement around v 6.6 m, which accumulated half a meter of debris in a comparatively short time. On top of it, probably the first houses of the next building phase were constructed, the positions and contents of which unfortunately are not documented. The premature occurrence of new pottery types below v 6.0 m could result from pits belonging to the later settlement, which were not recognised during excavation. The combined statistical analysis of the gradual change in pottery type frequencies and recent radiocarbon results may thus help to distinguish between cultural dynamics and the accumulation process of settlement debris. Interestingly, some discontinuities in the development of the pottery, which CA revealed, correspond quite well with building phases, which were reconstructed with reference to the unpublished notes and sketches of the excavator (Ko-rosec 1953; Stalio 1968; 1984). Phases 5b and 5c of the present study (Fig. 4) coincide with B. Stalio's settlement III, while her settlement IV covers both phases 6 and 7. The stratigraphic boundaries of phase 5a to 7, as defined here, show surprisingly good accordance with Korosec's layers Ila, lib, lie, lid and III (Korosec 1953-41-44). Many problems and open questions about the sequence of building phases in Vinca-Belo Brdo can be solved only by means of new excavations, which would require large areas and a corresponding investment of funds. But for the time being, a statistical analysis of both relative and absolute chronology, based on the old finds, can improve our understanding of the settlement history of one of south-eastern Europe's largest tell sites. - ACKNOWLEDGEMENTS - The results presented in this article were part of a research project supported by the Deutsche Forschungsgemeinschaft. I wish to tha nk Dr. Bernd Kroner, Heidelberg for performing the radiocarbon analyses and his information about the OxCalprogram. Prof. Dr. Irwin Scollar, Remagen supported the project by providing test versions of his archaeological statistics program WinBASP. PD Dr. Paid Yule, Bonn was kind enough to go over the English version of this text. REFERENCE BREUNIG P. 1987. 14C-Chronologie des vorderasia-tischen, siidost- und mitteleuropaischen Neolithi-kums. Fundamenta A 13- Koln. Bohlau. BUCK C. E„ KENWORTHY J. B, LITTON C. D. and SMITH A. F. M. 1991. Combining archaeological and radiocarbon information: a Bayesian approach to calibration. Antiquity, 65: 808-821. BUCK C. E, LITTON C. D., SMITH A. F. M. 1992. Calibration of radiocarbon results pertaining to related archaeological events .Journal of Archaeological Science 19: 497-512. CHAPMAN J. 1981. The Vinca Culture of South-East Europe. Studies in Chronology, Economy and Society. Oxford: British Archaeological Reports International Series 117. DJINDJIAN F. 1991. Methodes pour I'archeologie. Paris: Armand Colin. GARASANIN M. 1951. Hronologija vincanske grupe. Arheoloski Seminar. Univerza v Ljubjani. 1973. Praistorija na ttu SR Srbije. Srpska Knji-zevna Zadruga. Beograd. 1979. Centralno-balkanska zona. In: Praistorija jugoslavenskih zemalja. II. Neolit:79-212. Aka-demija Nauka i Umjetnosti Bosne i Hercegovine, Centar za balkanoloska ispitivanja, Sarajevo. 1993. Zu den Problemen der Vinca-Gruppe in Ru-manien. Balcanica 24: 7-20. GREENACRE M. J. 1989. Theory and Applications of Correspondence Analysis. London, New York: Academic Press. HOLSTE F. 1939- Zur chronologische Stellung der Vinca-Keramik. Wiener Prdhistorische Zeitschrift 26: 1-21. KOROSEC J. 1953. Delitev vincanske kulturne plasti. Arheoloski Vestnik 4: 5-46. JOVANOVIC B. I960. Stratigrafska podela vincan-skog naselja. Starinar N.S. 11: 9-20. 1984. Stratigrafija. In D. Srejovic et al. (eds.), Vinca upraistoriji isrednjem veku. Galerija Srpske Akademije Nauka i Umjetnosti: 23-34. MADSEN T. 1988. Multivariate statistics and archaeology. In T. Madsen (ed.), Multivariate Archaeology. Numerical Approaches in Scandinavian Archaeo- logy. futland Archaeological Society Publications 21: 7-27. MILOJCIC V. 1949a. Chronologie derjungeren Stein-zeitMittel- und Sudosteuropas. de Gruyter. Berlin. 1949b. South-Eastern Elements in the Prehistoric Civilization of Serbia. Annual of the British School at Athens 44: 258-306. RAMSEY C. B. 1994. OXCAL v2.0. A Radiocarbon Calibration and Analysis Program, Radiocarbon Accelerator Unit, Research Lab for Archaeology. Oxford. STALIO B. 1984. Naselje i stan. In D. Srejovic et al. (eds.), Vinca upraistoriji i srednjem veku: 34-41. TODOROVIC J., CERMANOVIC A. 196I. Banjica - Naselje vincanske kulture. Povremena izdanja sveska 2. Beograd: Muzej Grada Beograda. VASIC M. 1932. Preistorjiska Vinca I. Industrija ci-nabarita i kosmetika u Vinci. Izdanje i stampa Dr-zavne stamparije Kraljevine Jugoslavije. Beograd. 1936a. Preistorjiska Vinca II: Oblici grobova. -Misticne oci. - Igra na tabli. - Datovanje Vince. Izdanje i stampa Drzavne stamparije Kraljevine Jugoslavije. Beograd. 1936b. Preistorijska Vinca III: Plastika. Izdanje i stampa Drzavne stamparije Kraljevine Jugoslavije. Beograd. 1936c. Preistorjiska Vinca IV: Keramika, Izdanje i stampa Drzavne stamparije Kraljevine Jugoslavije. Beograd. UDK: 903.2(469-117)"634":66.3/.7 Documenta Praehistorica XXVII Painted Pottery from East Macedonia, in North Greece: Technological Analysis of Decorative Techniques Paraskevi Yiouni Archaeological Museum of Kavala, Kavala, Greece euri@hol.gr ABSTRACT - Neolithic pottery from East Macedonia displays a considerable variety of surface finishing, ranging from through burnishing, painted, applied, incised, excised and impressed decoration. Methods of surface treatment, although widely used as the main criterion for the classification of ceramic wares, have received little attention. Pottery is usually described in terms of decorative motifs, colour variation or colour combination, but the technological processes responsible for these effects are, in most cases, ignored. The usual assumption is that similar results can be obtained by means of similar technological processes. This assumption is questioned by the present work, on the basis of a study of a large sample ofpainted ceramics from East Macedonia covering the whole of the Neolithic period. The material has been studied ma-croscopically and microscopically (polarising microscope and scanning electron microscope). A number of reflring tests were also carried out. The analysis so far indicates that there was great variation firstly, in the raw materials and techniques used by the potters for the production ofpainted motifs, and secondly, in the conditions prevailing during firing. IZVLECEK - Neolitska keramika iz vzhodne Makedonije kaze veliko raznolikostpri obdelavi povrsine, ki obsega tehnike okrasevanja, kot so loscenje, slikanje, apliciranje, vrezovanje, izrezovanje in vtiskovanje. Ceprav se metode obdelovanja povrsine na siroko uporabljajo kotglavni kriterij za klasifikacijo keramic-nih posod, jim niso posvecali veliko pozornosti. Keramiko obicajno opisujejo glede na motive okrasevanja, barvo ali barvne kombinacije, vecinoma pa prezrejo tehnoloskiproces, s katerim so izdelovalci do-segli omenjene ucinke. Obicajno domnevajo, da s podobnim tehnoloskim procesom dobimo podoben uci-nek. Vclanku to predpostavko postavljamo pod vprasaj na osnovi raziskave velikega vzorca slikane keramike iz vzhodne Makedonije skozi vse neolitsko obdobje. Material smo raziskovali makroskopsko in mi-kroskopsko (polarizacijski mikroskop in vrsticni elektronski mikroskop (SEM')). Opraviti smo tudi stevil-ne teste sponovnim zganjem. Analize zaenkrat kazejo, da gre za veliko raznolikost tako v surovini kot v tehnikah, ki so jih loncarji uporabljali za slikanje motivov. Velike razlikepa smo nasli tudi v nacinu zganja. KEY WORDS - Neolithic Macedonia; painted pottery; technological analysis; firing of pots; petrographic study; chemical analysis; SEM; refiring test INTRODUCTION East Macedonia is the area between the Strymon and Nestos Rivers, in Northern Greece (Fig. 1). The earlier Neolithic levels excavated so far are dated to 5500-5200 BC (the late Middle Neolithic period). This phase is represented by the lower levels of the site of Limenaria on the island of Thassos (.Malami-don and Papadopoidos 1993), as well as those of Sitagroi (Sitagroi I: Renfrew et al. 1986) and Dimi-tra (Dimitra I: Grammenos 1997). Late Neolithic I, dated to ca. 5200-4800 BC, is represented by Sitagroi II, Dimitra II, Dikili Tash I (for Dikili Tash see Treuil 1992) and the lower levels of Krioneri (Mala-midou 1997) and Promachon-Topolnitsa (Koukouli etal. 1996a). Late Neolithic II, starting around 4800/ 4600 BC, is represented by Sitagroi III, Dimitra III, Dikili Tash II, Paradeisos (for Paradeisos see Hel-strom 1987), the upper levels of Krioneri and Pro-machon-Topolnitsa, and the site of Kastri in Thassos (Koukouli 1972). The end of this phase is estimated to be around 3800 BC. Apart from the excavated sites mentioned so far, ceramic material attributable to the Late Neolithic has been found, through surveys, Fig. 1. Map of East Macedonia and Adjacent areas shotting the Neolithic sites discussed in the text: 1. Kastri Srimonikou, 2. Zervochori, J. Monovri-si, 4. Chriso, 5. Toumba, 6. Fakistra, 7. Prof. Ilias Pentapolis, 8. Gradiskos. 9. Agio Pneuma, 10. Neo Souli, 11. Promachon, 12. Tholos, 13. Airi Bairi, 14. Dimitra, 15. Fidokoriphi, 16. Krioneri, 17. Hill 133 Amfipolis, 18. Mikro Souli, 19. Galipso, 20. Lakkovikia, 21. Akropotamos, 22. Podochori, 23. Agista, 24. Nea Bafra, 25. Megalokabos, 26. Maara Cave, 27. Petrous, 28. Bournar Basi, 29. Milopotamos, 30. Zoodochos, 31■ Kalos Agros, 32. Sitagroi, 33■ Sikia, 34. Kalabak Tepe, 35. Do-xat Tepe, 36. Arkadikos, 37. Xiropotamos, 38. Ka-lifitos, 39. Kirgia A, 40. Kirgia B, 41. Kefalari, 42. Dikili Tash, 43. Polistilo, 44. Sibolo Cave, 45. Nymphs Cav., 46. Kara Orman, 47. Paradisos, 48. Maries Cave, 49. Limenaria, 50. Kastri. at a large number of tell sites (i.e. Polystylo, Mikro Souli, Podochori, Toumba etc., see Appendix D in Grammenos 1991). Most of the East Macedonian settlements are located in the well-drained fertile plains of the region. Common to all sites is their proximity to a water source, which is usually a river or lake (Andreou et al. 1996). Since most of the excavations are small-scale operations, information on the layout of the settlements is scarce. An exception to this is the site of Dikili Tash, where a number of parallel, rectangular houses with internal ovens and platforms, an abundance of pots, stone and bone tools have been revealed by the recent excavations (Koukouli-Chrys-santhaki et al. 1996b). The houses, separated by small lanes, are dated to the LN II period. A very clear picture of the organisation of the Neolithic household is also offered by the architectural remains of the LN I strata of the same site (Koukouli-Chryssanthaki et al, 1996b; Treuil and Tsirtsoni 2000). Furthermore, important data on the organisation of a Middle Neolithic settlement have been obtained through the recent excavations at the site of Limenaria (Malamidou and Papadopoulos 1993)• A characteristic of the Neolithic pottery from East Macedonia is the great variation of painted vessels. Late Neolithic pots in particular, when compared with contemporary material from Central or West Macedonia show a much greater variation and elaboration in the colour of the motifs and the surface of the vessels, as well as in the decorative patterns and the forms of the pots (Grammenos 1997). De- spite this, the methods of surface treatment have received little attention. The usual assumption is that similar results can be obtained by similar technological processes. This assumption is questioned by the data discussed in the present work. In fact, study of the ceramic material from a number of East Macedonian sites has shown that a great variety of raw materials and techniques, some of which are rather complicated, were used by the Neolithic potters to decorate their pots. Data for the present work were obtained through the macroscopic examination of a large sample of ceramics, the microscopic examination of 51 thin sections, the refiring of 75 sherds, and the chemical analysis of 9 samples1. The presentation of the data will follow a chronological order: material from the Late Neolithic I phase will be presented first, followed by the LN II data. LATE NEOLITHIC I During the early part of the Late Neolithic (LN I) a whole new range of painted wares emerge. These vessels, amounting to 10%-12% of the ceramic assemblage, are mainly painted with a brownish pigment on a buff-coloured surface. Despite their rather limited appearance, vessels display a great variation in the colour of their pigment, background and fabric. Thus a number of wares have been identified by researchers to classify the material: brown-on-cream, brown-on-buff coloured surface and brown-on-white slip (Keighley 1986; Grammenos 1997). Except for 1 Chemical analysis and examination with the Scanning Electron Microscope (SEM) was conducted by Dr. V. Kilikoglou (Laboratory of Arcahaeometry Institute of Material Science NCSR "Demokritos", Athens) the vessels decorated with dark pigment on a clear-coloured surface there are, in minor quantities, pots decorated with red or purplish-brown pigment on a red-coloured background (orange-on-orange and red/brown-on-red ware respectively: Keighley 1986. 353)- Still, these wares are considered very general and many sherds could not be assigned to any category, due to their colour variation (Keighley 1986. 354-357; Grammenos 1997.Cat, 12 in Tab. 1). It will be argued in the following paragraphs that this polymorphy of painted vessels results from the variety of techniques and raw materials used by the potters to decorate the pots. Another criterion used by researchers for the classification of the material is the decoration of the vessels. Fragments with very thin linear motifs were assigned to the brown-on-cream and red/brown-on-red ware (Akropotamos style: Keighley 1986.352-353; Grammenos 1997.Categs. 2,6,7,8,9 in Tab. /). Motifs are in narrow lines, about 3-4 mm wide, and there is a rich display of spirals, concentric circles, parallel lines and ladder elements (Keighley 1986. 352) (Fig. 2). All examples are made from fine-textured fabrics. Brown-on-buff or brown-on-white sherds are reported as being decorated exclusively with motifs of broad thickness (6-9 mm). Among the design characteristics are thick wavy lines, parallel lines, concentric circles, thick arcs etc. (Keighley 1986.354-356) (Fig. 3). The fabric of these vesseis can vary from fine to medium-coarse textured. Vessels decorated with a brown pigment on a grey/buff-coloured background Two elements are of importance for the decoration of these vessels. Firstly, the colour of the pigment, and secondly, the colour of the background bearing the decoration. The pigment is basically dark brown, but colours vary from light brown to black. Research for the present work has shown that in the production of the dark colour of the decoration two different methods were used: a. Vessels were painted with an iron-based pigment that acquired a dark colour by being fired in reducing conditions (iron reduction technique). This complex technique can be summarised as follows: during firing, both the decoration and the body of the vessels turn black as they are fired in a reducing atmosphere. Under these conditions the sintered, iron-based pigment will retain the dark colour, in contrast with the body which, by being more porous, is re-oxidised (Farnsworth and Simmons 1963-391). It is the first time that this technique has been identified in Neolithic East Macedonian pottery, p. Vessels were painted with a manganese-based pigment, a material that can produce a dark colour irrespective of firing conditions. In comparison to the iron reduction technique, the use of a manganese-based pigment is a rather simple technique. Recent research, however, has shown that the distinction of the vessels according to the width of their motifs is over-simplistic and conveys a biased picture of the material (Tsir-tsoni 2000). Vessels can have their entire surface decorated either with bold or thin-lined motifs (Figs. 2.1, 5). In other cases, however, a single vessel can combine both types of decoration (Fig. 2.2). In the present work, the term "Akropotamos style" will be used to describe the thin lined motifs, but it should be kept in mind that the results of the vessels with Akropotamos style decoration may be applicable to vessels with bold motifs and vice-versa. Fig. 2. Late Neolithic Painted Vessels: 1. Brown-on-cream from Dimitra ^Grammenos 1997.Fig. 10.88/; 2. brown-on-cream from Dikili Tasli (Tsir-tsoni 2000.Fig. 4); 3- red/brown-on-redfrom Sitagroi ^Keighley 1986.Figs. 11.12, 8); 4. red-on-red from Dimitra ^Grammenos 1997.Fig. 21.320); 5. brown-on-cream from Sitagroi ^Keighley 1986.Figs. 11.12, 6); 6. red-on-red from Dimitra ^Grammenos 1997-Fig. 21.31.3/ Fig. 3. Late Neolithic Painted Vessels: 1. brownon-grey from Dimitra fGram-menos 1997-Fig. 20.304); 2. brown-on-grey from Dimitra fGrammenos 1997.Fig. 20.3083. brown-on-grey from Dimitra fGrammenos 1997.Figs. 21.310^; 4. vessel decorated with a post-firing organic coating from Dimitra fGrammenos 1997-Fig. 20.306). As for the whitish background, various raw materials such as kaolin or calcareous based slips were used by the early Late Neolithic potters (see the section below). In other cases, vessels remained unsli-pped, but acquired a buff-grey coloured surface through the control of the firing atmosphere. Iron-based pigment We can begin our presentation with the data on the iron reduction technique. The data were obtained through refiring tests, chemical analysis and macroscopic examination of the material. The refiring of 30 sherds with brown motifs (20 Ak-ropotamos-style, 10 with bold motifs) was carried out in a Labertherm furnace in an oxidising atmosphere from 25°C to 900°C (Tab. 1). After refiring, the bold motifs (Munsell soil colour notations 10YR 3/1; 10YR 3/2) of four sherds from the sites of Mi-kro Souli and Podochori acquired a clear red colour (Munsell soil colour notations 2.5YR 4/8: see Fig. 4). In the remaining sherds the motifs retained their dark colour. According to the characteristics described above, it appears that the decoration of the four vessels from Mikro Souli and Podochori was produced through the iron reduction technique. The change of colour, from brown to red, indicates that the iron-based pigment was not thoroughly reduced. For a successful result, two variables must be controlled by the potters: first, the quality of the pigment, and second, the firing conditions. The preparation of fine-textured pigments is very important because coarse and grainy pigments are difficult to vitrify (Aloupi and Mani-atis 1990) and during the oxidising phase of firing they will re-oxidise more easily than the fine textured ones. At the same time, the pigment must be applied in thick layers, since such layers sinter more effectively than the thin ones (Jones 1986. 765). Equally important for the vitrification of the pigment is the control of firing conditions, mainly temperature and duration of firing. The duration of reducing conditions is also of great importance. Analysis of two of the refired sherds (one from Mikro Souli and one from Podochori) showed that their pigment is fine textured, but it remained porous since it was thin (5-10 |im), and the firing temperature (800°C-850°C) and perhaps the duration of firing were not enough for its complete vitrification. As for the twenty-six sherds retaining their dark motifs after refiring in oxidising conditions, the results of the refiring test are not conclusive. The sherds were either decorated with a manganese-based pigment, or with a thoroughly reduced ferruginous material. Mikro Souli and Podochori are situated in the western provinces of East Macedonia, not far from Dimitra. Macroscopic examination of the painted vessels from Dimitra indicates that the iron reduction technique was also used there: quite often the brown motifs of brown-on-buff or brown-on-white pots are, in places, red. At Dikili Tash, a site further east, similar features have been noticed both on vessels with broad motifs (Fig. 5) and on those with the Akropo-tamos-style decoration. Since manganese based pigments do not shade to red or brown (Farmvorth and Simmons 1963-394), it has to be concluded that the vessels were decorated with an iron-based pigment fired in reducing conditions. The discoloration of the motifs can be attributed to local draughts of air that did not permit the maintenance of fully reducing conditions. The non-uniform thickness of the pigment material may also be responsible for such features. Motifs, mainly the bold ones, are quickly executed, and often the relief of the decoration is varied on a single vessel. As already mentioned, thicker layers can retain their colour better than thin. Manganese-based pigmen t The brown-black pigment was also produced through the application of manganese-based materials. Mn02 was identified by the chemical analysis of four sherds with Akropotamos style motifs (two from Dikili Tash and Mikro Souli, respectively: see Tab. 1). Similarly, Gardner has identified the brown pigment of the Akropotamos style sherds from Sitagroi, as an iron oxide-manganese based material (Gardner 1980. 109). The use of manganese pigment for the production of the bold brown-black motifs of the East Macedonian sites is not documented, but on the basis of the discussion in the previous section it cannot be excluded. Pure manganese coatings do not ordinarily contain sufficient fluxing materials to sinter or vitrify, and so if they are applied to the surface of the vessels they will be fugitive. But the brown-black motifs of the Neolithic vessels have been permanent. According to Shepard (1976.42) the successful bonding of a manganese pigment could be due to the following reasons: (1) polishing, (2) the presence of impurities, mainly clay, (3) protection by a post-firing coat (lac-quer-like substances or resins of various plants). The use of a post-firing, protecting coat is not reported from any Neolithic site. The burnishing lustre of the Late Neolithic I vessels varies from good to poor. It seems, however, that the durability of the dark pigment of the vessels should be attributed mainly to the presence of clay. The sherds analysed in the present study (including the LN II black-on-red ware: see section below) were painted with an extremely fine-textured material containing MnOj, FeiOi and high amounts of alumina (that is, a clay-based material). Although iron was most probably naturally present in the clay used for the preparation of the pigments (i.e. a ferruginous clay), manganese-ore had to be added by the potters. Buff-grey background The background of the LN I painted vessels ranges in colour from grey-cream, white, very pale brown or buff. Such colours can be produced either by the application of a slip or by controlling the firing conditions, both of which methods were often used in combination by the East Macedonian potters to produce their painted pots. Sample studied macroscopically Thin sections Retired sherds Chemical analysis and SEM Akropotamos LN 7 (5 Akropotamos 2 Bold br. decor) Dikili Tash LN 31 (23 undecorated 2 B l/red 1 Bl/white 5 Graphite) 17 (5 Akropotamos 2 Bold br. dec.) 9 Graphite 1 "Bitumen" 3 (2 Akropotamos 1 Br/white) Dimitra LN 4 (2 Bl/red 2 Graphite) 1 Bl/red Galipsos LN 2 Bl/red 6 (2 Akropotamos 1 Polychrome) Kalambaki LN 4 (2 Bl/red 2 Graphite) 5 (3 Bl/red 2 Graphite) Kalifitos LN 1 Graphite Mikro Souli LN 2 Bold br. decor 3 (2 Akropotamos 1 Bold br decor) Nea Bafra LN 2 Bl/red 5 (2 Bl/red 3 Graphite) 1 Bl/red Podochori LN 2 Bold br decor 1 Bold br decor Promachon- Topolnitsa LN 10 (3 Akropotamos 1 Graphite 1 "Bitumen" 2 incised decor 1 black topped 2 undecorated) 6 (3 Graphite 3 "Bitumen") Sitagroi MN and LN 10 Graphite Toumba LN 2 Bl/red 10 (5 Akropotamos 2 Bold br decor 2 Bl/red 1 bichrome) —- The clay fabric of the Akropota- Tab. 1. The ceramic material and the analytical methods employed in mos-style vessels is cream-grey, the present work. very light brown, or reddish-yel- red a clear reddish-yellow colour. Five were covered with a thin slip, which macroscopically appears similar to that applied on the Akropota-mos-style vessels. Four sherds with Akropotamos-style decoration (two from Dikili Tash and two from Mikro Souli) were analysed by V. Kiliko-glou. Three were covered with a calcareous-rich slip, ranging in thickness from 10-20 |am. The fourth calcareous slip was thicker, reaching 30-40 jam. Fig. 4. Sherds from Mikro Souli and Podochori decorated by the iron reduction technique. After refiring, in an oxidising atmosphere at 900°C, the broum motifs acquired a clear red colour. low. In the first case, the motifs appear to have been applied to an unslipped, carefully burnished surface. Vessels made from very light brown, or reddish-yellow coloured materials usually have a thin, cream-beige slip. In most of the cases the slip is worn out, being preserved only in small spots. Gardner, in her analysis of the Sitagroi material, identifies this white layer as kaolin slip (Gardner 1980.109). Twenty Akropotamos-style sherds with cream-grey and very pale brown fabric (the colour of surfaces and fabric ranging from Munsell soil colour notations 10YR7/2 and 7/3, 7.5 YR 7/2 to 10YR 6/3) were refired, for the present study, in oxidising conditions at 900°C (see Tab. 1 and the previous section on iron deduction technique). After refiring, their fabric acquired a reddish-yellow colour (Munsell soil colour notations 7.5 YR 7/6 and 6/6 and 5YR 6/8) indicating that the grey, very pale brown fabric was produced by firing the pots in conditions that did not permit their full oxidation. Furthermore, re-firing showed that all sherds had a thin, whitish, matt slip, except for two sherds which were only burnished. Similar results were obtained after refiring ten sherds with bold brown motifs on a buff, very light brown background. All sherds acqui- The whitish slip, even when well preserved, is usually thin and cannot conceal the reddish colour of the fully oxidised body. It seems then that despite the use of a slip, firing the vessels under conditions which did not permit their full oxidation was important for the successful production of a cream-grey, very light brown background. Chemical analysis of four of the refired sherds showed that the pigment used for their decoration is manganese-rich (see previous section on manganese-based pigments). Thus, firing in non-oxi-dising conditions was not aimed at the production of the dark motifs, but at the production of the desired background of the decoration. Fig. 5. Browttron-buff vessel from Dikili Tash. Both motifs and background are, at places, red. This discoloration can be attributed to local drafts of air that did not permit the maintenance of fully reducing conditions. Vessels decorated with a red pigment on a red-coloured background Pots with red, red/brown motifs on a red coloured surface (orange-on-orange and red/brown-on-red wares) are present, although in minor quantities, in Late Neolithic I levels (Figs. 2.3, 4, 6). For the production of the red motifs an iron-based pigment was used which acquired a clear red colour by firing in oxidising conditions. Under the same conditions the fabric of the vessels turned red and was used as the background of the decoration. Vessels with red/brown (orange-on-orange and red/ brown-on-red wares) and dark brown decoration (brown-on-cream, brown-on-buff and brown-on-white wares) have comparable motifs (thin or bold lines), similar forms and fabrics (Keighley 1986. 353; Grammenos 1997.39-41). They differ mainly in the colour of their decoration. Based on this, someone could propose that the red/brown motifs were the result of firing accidents: potters were aiming at the production of brown-on-buff effect, but since they could not control the firing conditions, produced the red-on-red vessels. Although this may be the case for a few vases, it seems that the clear red colours of most of the red, red/brown-on-red vessels should not be attributed to firing accidents. Sometimes, iron-based and manganese-based pigments were combined by the potters to produce a polychrome effect (red and black motifs on a red background). Such vessels are rather rare: a few sherds have been found at Dimitra, Toumba, Galipsos, and Promachon-To-polnitsa (Grammenos 1997; Koukouli et al. 1996a). Some of these sherds (from Dimitra, Toumba and Galipsos) were examined under stereoscopic microscope. In all cases the red and black pigments appear to have comparable relief. This seems to exclude the use of the iron reduction technique, in which, by varying the thickness of an iron-rich pigment and changing from a reducing to oxidising atmosphere, a dichromatic effect can be produced (for a theory on this see fones 1986.765). Support for this conclusion lies in the fact that after refiring (in the same conditions as the vessels with brown-black motifs) the motifs of two sherds from Toumba and Galipsos did not lose their original colours. Vessels decorated with an organic material applied after firing Such vessels have been found in the recent excavations of Promachon-Topolnitsa, a site on the Greek-Bulgarian border (Koukouli et al. 1996a. 753). The decoration, which consists of stripes, wavy lines or groups of parallel lines, was executed with a soft, thick, black-coloured material, which adheres well to the surface of the pots. Three decorated sherds were refired, in an oxidising atmosphere at 320°C for 30 minutes. After firing, the motifs were still present, but assumed a very pale grey colour. After firing at 450°C for 30 minutes, the decoration disappeared (Fig. 6; Yiouni et al. 1994). Since this temperature is low (far below the tempe-rature-range at which ceramics are fired), it can be concluded that the black colouring material was an organic coating applied after the original firing of the vessels. Macroscopically, the material appears as bitumen, but this has to be verified by further analysis2. Fig. 6. Refiring of sherds from Promachon-Topolnitsa decorated with an organic coating. In the clear coloured chips of two of these sherds the organic material disappeared after refiring at 450°C, for 30 minutes. The pink-red slip of the third sherd was not affected by the firing. 2 Bitumen is either an originally mineral pitch or any of several hard or semisolid materials obtained as asphalitic residue in the distillation of coal tar, wood tar, petroleum etc. Similarly decorated vessels were also identified among the ceramic material from two other settlements: Dikli Tash and Dimitra (Fig. 3-4; Yiouni 2001). These are the first documented instances for the use of this technique in Macedonia. It is very probable, however, that this practice was more widespread. Sherds from the Late Neolithic Servia (in West Macedonia) are decorated with a colouring material which, according to the excavator's description, has the characteristics of an organic coating applied after firing (Yiouni 2001). It should also be pointed out that the scarcity of this type of decoration might not reflect the original distribution of this ware. Ta-phonomic processes or the vigorous cleaning of the sherds can easily destroy the post-firing application. Vessels with similar decoration have also been found at two Bulgarian settlements situated north of Pro-machon: Damyanitsa and Balgarcevo (20 and 80 km respectively to the north: Grabska-Kulova 1993; Pernicheva 1995). Analysis of one sherd from Damyanitsa showed that the colouring material was bitumen (Wagner and Graf1993)- Deposits of bitumen exist in the area around Damyanitsa, but Wagner and Graf cannot exclude the possibility that the material was obtained as a residue in the distillation of organic substances. Firing of Late Neolithic I painted pottery According to the discussion so far, it is clear that firing played a major role in the final appearance of the LNI painted vessels from East Macedonia. By controlling the firing conditions, potters could modify the colour of the motifs and/or the background of the decoration. In the present section we will try to reconstruct the firing techniques used by the potters on the basis of the macroscopic examination of the material, the estimated firing temperatures of the analysed sherds and the available remains of firing structrures. Starting our discussion with the vessels decorated with brown motifs, it is easy to conclude that these vessels were not fired in open firings. In such a procedure the positioning of fuel and vessels before firing can affect the flow of air (Rye 1981.98), but during firing it is very difficult to control the atmosphere and to change it at will from oxidising to reducing. It is true that the control of the firing conditions by the LN I potters was not very strict and the changes in atmosphere were not always successful, but still the overall appearance of the vessels is not compatible with an open firing. So vessels were fired either in pits (where fuel and vessels are not separated), or in kilns (where fuel and vessels are separated). When pits are simple depressions in the ground they do not offer any significant advantages compared to open firings. More sophisticated structures, however, such as a circular wall or a three or four-sided enclosure, are known ethnographically. The fuel and vessels are often placed in alternating layers, and air access may be provided by holes in the walls at or near ground level and by passages let through the setting (Rye 1981. 98). Thus, pits may achieve higher temperatures and sustain them longer than open firings. Moreover, pits provided with an air inlet offer better control since air can be excluded easily during firing (Shepard 1976.216-217). In a kiln firing, of course, the firing atmosphere and the rate of heating can be controlled more effectively. Macroscopic examination of the painted vessels indicates that they were, most probably, fired in pits. The basic indication for such a firing is the discoloration of their motifs. The change in colour, on a single vessel, from brown to red is compatible with a pit firing because it suggests that the firing conditions were not strictly controlled, as would have been expected in a kiln firing. Moreover, the painted vessels have often grey-coloured, smoked areas, a feature suggesting that, during firing, the vessels were not separated from the fuel. The available estimated temperatures of the analysed sherds are also within the range attained by pit firings. As can be seen from Table 2, the mean temperature is 850°C, although some vessels were fired at even higher temperatures (1000°C-1100°C). So far, the discussion has been restricted to the data obtained from the analysis of fired ceramics. But what do we know about the firing structures themselves? An oven from Dikili Tash, similar in shape to the domestic ovens from the site, has been identified as a structure used for firing vessels since it contained a number of pots mixed with charcoals (Sefe-riades 1983.643, Fig. 6). In this case, vessels and fuel were placed in the same chamber. Re-examination of the data, however, and comparison with the architectural remains from the recent excavations of the site question this interpretation (Tsirtsoni 2000). According to the new interpretation, it appears that the vessels were not standing on the floor of the oven, but on a near-by clay platform (a common feature accompanying the ovens at the site of Dikili Tash). Another structure that could have been used for firing pots was found at Krioneri. A cylindrical Site/Ware Temperature Scientific Method Range (°C) of Estimation Dikili Tash 1 Akropotamos-style 800-900 Scanning Electron Microscope (V. Kylikoglou) Dikili Tash 2 1100 Akropotamos-style Mikro Souli 2 Akropotamos-style 1050-1080 » Mikro Souli 3 Akropotamos-style 800-900 jj Mikro Souli 1 850 jj Bold. Br. Decor Podochori 1 Bold. Br. Decor 800-850 » Dimitra X-Ray Diffraction 5 sherds of 750-850 (Kessissogloy and Akropotamos-style Mirtsou 1997) Dimitra 2 sherds of 900-950 jj Akropotamos-style Tab. 2. Firing temperatures of Late Neolithic I painted vessels. pit (1 meter in diameter), with an opening at the side (0.40 m wide), was dug by the Neolithic inhabitants at the edge of the settlement (Malamidou 1997.515, Fig. 5). It contained some undecorated sherds mixed with ashes and charcoal. The interior of the pit was burned, whereas the opening was blocked with stones and earth. Although the contextual data are not fully conclusive for the use of this pit for firing pots (i.e. absence of whole vessels inside the pit and lack of pottery wasters in the surrounding area), theoretically it could have provided the conditions needed for the reduction of the iron-based pigments. Late Neolithic II (Tab. 3). This phase, however, is characterised by the presence of two distinctive wares: the black-on-red and graphite decorated vessels. Black-on-red ware is characteristic of the East Macedonia, since it is restricted mainly to this region. Appearing sporadically in LN I (Grammenos 1997. Tab. 1; Keighley 1986. 358), black-on red vessels are more common in LN II (Tab. 4). In Central and West Macedonia black-on-red vessels are sporadically present (Grammenos 1991.126). In Thrace, this ware is found at the settlements of Paradeisos (Hellstrom 1987) and Paradi-mi [Bakalakis and Sakellariou 1981), being rare at Makri (Efstratiou 1991.600). Graphite, the mineral form of carbon, has an early appearance in the East Macedonian region. Grey lustre and grey-channelled wares are present in the late Middle Neolithic and early Late Neolithic (Sitagroi I and II levels). These vessels are covered by graphite, which produces a smooth, glittering and "soapy" surface (Keighley 1986.346). Graphite, painted and excised with graphite (vessels combining painted and excised motifs) are characteristic of the LN II levels. They are present at all East Macedonian settlements, increasing in frequency as we move eastwards. At Dimitra, for example, such vessels amount to 5%-17% of the decorated pottery, but they are far more common at Sitagroi and Dikili Tash, where they comprise 75% or more of the decorated pottery (Tab. 4). In Thrace, graphite-decorated The contemporary vessels decorated with red motifs (orange-on-orange and red/ brown-on-red ware) could also have been fired in similar structures, although an open firing cannot be excluded. The same can be proposed for the vessels decorated with an organic coating applied after firing. LATE NEOLITHIC II Vessels painted with a dark brown or red pigment (brown-on-cream, brown-on-buff, brown-on-white, orange-on-orange and red/ brown-on-red wares) continued to be produced, although in decreasing frequency, in Site/Ware Temperature Scientific Method Range (°C) of Estimation Dimitra X-Ray Diffraction 8 Bl/Red sherds 900-950 (Kessissogloy and Mirtsou 1997) Dimitra 1 Bl/Red sherd 750-850 » Dimitra 1 Scanning Electron 1 Bl/Red sherd 850-950 Microscope (Kilikoglou) Dikili Tash 3 1 Bl/White sherd 1000-1100 » Nea Bafra 1 1 Bl/Red sherd 850-950 jj Tab. 3■ Firing temperatures of black-on-red and related Late Neolithic IIpainted vessels. vessels are common at Paradeisos and Paradimi, but rare at Makri (Efstratiou 1991-600). Only sporadic examples are reported from a few Central Macedonian sites (.Heurtley 1939.133, no. 128-9). In contrast to the black-on-red pottery, which has a very limited spatial distribution, graphite decorated vessels are characteristic of the Gumelnitsa-Karanovo VI cultural complexes of Bulgaria and Romania (Demo-ule 1993-382). Black-on-red The black motifs were painted either on the clear red, fully oxidised surface of the pots, or on an iron-based slip. The complex decoration of the vessels has been divided into two styles, I and II. Style I, displayed on a wider range of vessels, consists of various curvilinear and rectilinear patterns, various filled motifs (rectangles, triangles) and combinations of all three (Evans 1986.400) (Figs. 7, 8). Style II consists of broad curvilinear lines which often give a floral appearance (Fig. 9). Black-on-red vessels are usually open, large-sized pots with rounded or flaring walls. Jars are also present. All the analyses so far show that the black decoration was executed with a manganese-based pigment. This material decorated the vessels from Sitagroi and Dikili Tash (Gardner 1980.123; Courtois in press). Similarly, manganese has been identified in the chemical analysis of two black-on-red sherds from Nea Bafra and Dimitra (Tab. 1). During the macroscopic examination of the pottery from East Macedonia an interesting feature was noticed among the black-on-red pottery from Dimitra and Krioneri. Black-on-red vessels sometimes have a milky-white, transparent layer, which was applied on top of the painted decoration. Traces left on the surface of one sherd from Dimitra which was not carefully covered with this material, indicate that the transparent slip was wiped in various directions with a soft material (probably cloth). Furthermore, whereas the white layer is micaceous, both the red slip and the black pigment are free from mica. Chemical analysis of this sherd, showed this thin (3-5 (xm), fine-textured layer was applied before the firing of the vessel. Neolithic ceramic material. Perhaps it was restricted to sites near the river Strymon (both Dimitra and Krioneri are situated there), since it seems to be absent from settlements further east (i.e. Dikili Tash). What are the reasons for the application of this extra layer? Colour contrast and the reduction of porosity should be excluded as possible explanations as the material was applied to slipped and painted surfaces. Nor can it be proposed that it was used to protect the black pigment, since chemical analysis of the sherd from Dimitra showed that the colouring material is not a pure manganese coating, but a clay-based solution enriched with manganese. One probable explanation is that, by applying this extra layer, the potters could rapidly produce a lustrous surface. Black-on-red vessels are often open, large-sized pots. Burnishing these pots is more time-consuming than simply covering their surface with a liquid suspension. Graphite decorated pottery Graphite is also applied on large vessels with flaring or carinated walls (Evans 1986.398). Closed pots, often smaller in size, were also painted with graphite. The decorative style varies from rather simple lines to various complicated combinations of straight or curvilinear motifs: spirals, meanders, circles, triangles and lozenges (Evans 1986.397) (Figs. 10.1, 3, 4). In the vessels excised with graphite, painted motifs are combined with excised linear patterns filled with white or red-coloured paste (Fig. 10.2). Graphite is a soft, grey, laminar form of pure carbon occurring in high-grade metamorphic rocks as a final product of the carbonisation of organic substances. Due to its silvery appearance graphite pigment can be relatively easily recognised by macroscopic examination. So far, this material has been identified, by X-ray diffraction, on specimens from Karanovo in Bulgaria, Dikili Tash and Sitagroi in East Macedonia Site Graphite Painted Black-on-Red Vessels with dark (% of decorated (% of decorated brown motifs pottery) pottery) (% of decorated pottery) Dikili Tash 75% 1-5% - Sitagroi 79% 21% - Paradeisos 90% 10% - Krioneri 15% 60% 15% Dimitra 5-17% 28-58% 52-9% It is the first time that such a practice has been reported for a Greek Tab. 4. Relative amount of black-on-red and graphite decorated vessels. (Data from Demoule 1993; Evans 1986.Tab. 2.1; Hellstrom 1987; Malamidou 1997; Grammenos 1999-Tab. I). (Jones 1985.768, Tab. 9.6a). In her study of the material from Sitagroi, Gardner concluded that two different methods were used for the execution of the decoration: graphite was either rubbed on the surface of vessels (like a crayon), or it was applied, with a brush, in a liquid suspension (Gardner 1980.124; Evans 1986.397). These differences must have been dictated to a great extent by the nature of the raw material. If it is pure, graphite can be used as a crayon. Otherwise, it has to be refined before it can be used for decorative purposes (Yiouni 2001). A refiring test carried out for the present study indicates that pure graphite was rarely used by Neolithic potters. Thirty sherds with graphite decoration from a number of East Macedonian sites (Tab. 1) were refired at 850°C, in an oxidizing atmosphere. The maximum temperature was retained for 30 to 60 minutes. Apart from two sherds (both from the site of Promachon) unaffected by the firing, in all other cases the pigment appears as a white residue, occasionally preserving some metallic sheen. Since graphite is burned at rather high temperatures, it seems that the colouring material of the refired vessels was not pure graphite. Graphite decoration occurs mainly on black or dark brown-coloured surfaces (Evans 1986.397; Seferia-des 1983-653)- Thus, the silver colour of the pigment is more pronounced. The dark coloured background was not achieved by the application of a slip, but through the control of firing conditions. This Fig. 8. Black-on-red vessel from Dimitra (Grammenos 1997. Fig. 30. 457> Fig. 7. Black-on-red vessels from Sitagroi (Evans 1986.Fig.l2.8). conclusion is based on the following observations: the black colour of the surface(s) extends deep into the vessel's cross-section; in other cases the entire cross-section is dark. Furthermore, after refiring in an oxidising atmosphere, all sherds (30) acquired clearer surfaces and cross-sections. The linear patterns of the vessels excised with graphite are filled with a white or red-coloured paste. At Sitagroi both calcium carbonate and kaolin were used as a white infill, the red paste being ochre (Gardner 1980.128). Potters from Dikili Tash were using a paste rich in calcium oxide (CaO), sometimes containing white mica (Courtois in press). These materials were most probably applied after the original firing of the vessels (Gardner 1980; Yiouni 2001). So, the production of graphite-de-corated vessels required a combination of various techniques. Extreme examples of this practice are the pots combining graphite painted patterns, with excised motifs, the production of which can be summarised as follows: first the potters painted the motifs, and with a sharp tool produced the excised patterns. The painted sections of the pots were burnished. During firing, the vessels acquired a dark-coloured background, and after firing, the paste filling the excised motifs was applied. The firing of Late Neolithic II painted vessels In her study of the Sitagroi material Gardner, distinguished two groups (Type I and II) of graphite decorated ware (Gardner 1979). Type I is found north of the Rhodope Mountains, and Type II, south of the mountains, on the Drama Plain in east Macedonia. According to Gardner, pots from the southern region (Type I) were fired at rather high temperatures (1000°C or even more, according to refiring tests) and the quality of their pigment is very good (grainy, with a crystalline metallic sheen). In contrast, vessels found in Bulgaria were painted with a thin, fugitive pigment, and were low fired (lower than 750°C, according to Garner's refiring tests). Subsequent research on graphite-decorated vessels from Greek Neolithic sites does not support Gardner's conclusions. The estimated firing temperature of two sherds from Dikili Tash is lower than 750°C (Maniatis and Tite 1981.57-9). Both vessels were fired in non-oxidising conditions. Kessissoglou and Mirtsou reached the same results after the refiring of a number of sherds from Dimitra (Kessissoglou and Mirtsou 1997.89, Tab. 1). As mentioned above, thirty sherds from a number of east Macedonian Neolithic sites were refired by the author. In most cases the graphite pigment was Fig. 9■ Black-on-red vessel from Sitagroi (Evans 1986.Fig. 12.9.3;. burned off when fired at 850°C, or even at lower temperatures. The fabric of the sherds was also affected, since seven sherds (from Nea Bafra, Sitagroi and Dikili Tash) were vitrified. Graphite-decorated vessels were most probably manufactured from non-calcareous clays and, in contrast with the refiring test, were originally fired in a non-oxidising atmosphere. Since the firing of non-calcareous clays in reducing conditions lowers (by 50°C) the temperatures at which vitrification starts (Maniatis and Tite 1981.61), it is quite probable that some of the sherds were originally fired at temperatures lower than 850°C. Thus the analyses of the graphite-decorated pots suggest that these vessels were regularly fired in a non-oxidising atmosphere at rather low temperatures. A pit firing could facilitate the production of black coloured surfaces because of the ease of excluding air. It should be pointed out, however, that open firing of the vessels cannot be excluded. Instead, open firing can easily be excluded for the contemporary black-on-red ware. These vessels have clear red-coloured surfaces and cross-sections. When tapped, they produce a clear, crystalline, sound. These characteristics indicate that the pots were thoroughly fired in oxidising conditions at rather high temperatures. Indeed, the estimated firing temperature from a number of analysed sherds is around 900°C (Tab. 3). An interesting feature of the black-on-red vessels is the rarity of smoked areas on their surfaces. According to these characteristics, an open firing can be excluded, but is it possible to propose that these vessels were fired in kilns? Kilns represent a major advance toward ensuring success in firing pots. Updraft kilns are simple, enclosed firing chambers in which the heat moves upward from underneath the pots and is then vented outward. While many types of complex kilns exist, simple ones used by traditional potters usually have open tops through which the kiln is loaded. The maximum temperatures these kilns attain usually range from 900°C to 1000°C (Rice 1987.160). Apart from sustaining temperature as long as is needed, the other main advantage of a kiln firing is that both the atmosphere and the rate of temperature rise can be controlled (Rye 1981.98). The remains of structures that could be interpreted as kilns are absent from the Greek Neolithic sites. Taking into consideration the limited scale of the excavations, this absence is not a sound argument against the existence of kilns. Turning to the fired vessels themselves, it can be seen that the consis- tency of high firing temperatures and the control of the firing atmosphere are compatible with kiln firing. The technology of the vessels is also supportive of such a firing. As mentioned above, black-on-red vessels are often large and have complex shapes (Figs. 7, 8, 9). Furthermore, black-on-red vessels were regularly made from extremely fine-tex-tured fabrics (.Renfrew et al. 1986.155). Firing such vessels can be very risky. In a kiln firing, however, pots can be heated slowly and evenly to complete the drying process. The need for temper in the clay is not so great because, the rate of shrinkage is more controlled (as the rate of temperature rise can be controlled). In contrast, graphite decorated vessels were regularly made from medium to coarse-textured fabrics (Renfrew et al. 1986.158). Such fabrics facilitate the building and drying of large, complex forms and can withstand abrupt changes in firing temperature. Fig. 10. Graphite decorated vessels: 1. and 4. from Sitagroi fEvans 1986. Figs. 12.5, 1 and Fig. 12.4); 2. excised with graphite from Dikili Tasli; 3-from Dimitra ^Grammenos 1997-Fig. 19, 2(>1). ing conditions was imperative for the achievement of the desirable grey-buff colour. Red motifs, less common than the dark brown, were produced by the application of an iron-based pigment fired in oxidising conditions. CONCLUSIONS During LN I there was great variation in the methods and raw materials used to decorate pots. Dark brown, the preferred colour of decoration, was produced either through the use of manganese-based materials or through the reduction of iron-based pigments. More rarely, the dark motifs were produced with an organic material applied to the surface of the vessels after firing. Manganese-black and the iron reduction technique are widely known in East Macedonia. In contrast, the post-firing organic coating, despite its sporadic appearance at Dikili Tash and Dimitra, is characteristic mainly of the site at Promachon-Topol-nitsa (a site divided by the Greek/Bulgarian border) and the neighbouring Bulgarian site of Damyanitsa (being rare at Balgarcevo). The grey-buff colour of the background of the decoration was produced either through the application of a slip (kaolin or calcareous material) or by firing the vessels in non-oxidising conditions. Since in many cases the slip was very thin, the manipulation of fir- In all cases, the manipulation of firing conditions was decisive for the successful appearance of the decoration. The reduction of iron-based pigments, in particular, is a complex method requiring the careful preparation of the colouring material and organisation of firing conditions (control of atmosphere, temperature and duration of firing). These vessels were fired in pits, most probably provided with an air inlet (as the structure excavated at Krioneri). The discoloration of motifs from brown to red indicates that the potters were not always successful in their attempts. Characteristic of LN II are black-on-red and graphite-decorated vessels. Although these wares have many similarities in vessel form and decoration, they have pronounced differences in the fabric used for their manufacture, the raw materials used for decoration and the firing sequence. Graphite-decorated vessels were made from medium to coarse-textured clays which could not withstand firing at high temperatures. In contrast, the fine-tex- tured fabrics of the black-on-red vessels were regularly fired at rather high temperatures (around 900°C). This means that, for each ware, there existed specific recipes of clay fabrics, and potters systematically collected raw materials that would produce the desired results. The same is true of the earlier LN I painted vessels, since potters collected clays that could acquire, by being fired in slightly reducing conditions, a grey-buff colour. In the LN II period, however, there was a strict dichotomy between the clay fabrics used in the black-on-red and graphite decorated vessels. Completely different raw materials were also used for the decoration of these two wares: manganese-based pigment and an iron-rich slip for the decoration of the black-on-red vessels. The graphite-based pigment was often combined with a variety of post-firing pastes (ochre, kaolin or calcareous-based paste). It is interesting to note that manganese and graphite are common in the mountains surrounding the Drama Plain and that very often they are found in juxtaposition3. Despite this, the black background of the graphite-decorated vessels was produced through the manipulation of firing conditions and not by the application of a manganese-based slip. A pit or even an open firing could provide the conditions for the successful firing of graphite-decorated vessels. A kiln firing is proposed for the black-on-red vessels. Due to the complexity of firing techniques this proposal must be considered as preliminary and has to be confirmed by future research. It is undeniable, however, that the firing sequence of black-on-red vessels was carefully organised and strictly controlled. Given the basic differences outlined above and taking into consideration that the whole technological sequence is very complex, it could be proposed that these two wares were not manufactured by the same potters changing at will from one ware to the other (see also Courtois in press). As can be seen from Table 4, both wares are present at all east Macedonian sites, but their distribution follows an opposite pattern. Graphite-decorated vessels are rare at Krioneri and Dimitra (that is, in the western part of east Macedonia), but increase dramatically in frequency as we move eastwards. In contrast, black-on-red vessels are more common at the western sites. In the western region, vessels decorated with dark brown motifs are also present. At Dimitra, in particular, such 3 Dr. I. Chatzipanagis (Institute of Geological and Mining Research, Thessaloniki), pers. comm. 1998. vessels are extremely common in the early levels of LN II (amounting to 52% of the decorated pottery). Black-on-red and vessels with dark brown motifs have many similarities in the texture and colour of the clay fabrics used for their manufacture. It should be stressed, however, that the pattern of production of the Late Neolithic painted wares may not has been so simple (that is, the production of black-on-red in the west and of graphite decorated pots in the east) as appears from the data presented above. It has already been seen that the black-on-red decoration of the vessels from Krioneri and Dimitra was sometimes covered with an extremely thin (3-5 |im), whitish slip. This feature seems to be absent further east. A research program combining the pe-trographic and chemical analysis of a large sample of ceramics and raw materials is needed to better define the organisation of the production of the Late Neolithic painted vessels from east Macedonia. -ACKNOWLEDGEMENTS- I wish to thank Prof. Mihaet Budja for his invitation to come in Slovenia and for his unforgettable hospitality. For the preparation of this work pottery from a number of East Macedonian sites has been studied. I would like to thank the excavators of the sitesfor permitting me to examine the ceramic material and to take samples for a further study. In particular I would like to thank the Ephor of Antiquities of East Macedonia, Dr. Ch. Koukouti-Chrysanthaki, Prof. R. Treuil, Dr. D. Grammenos, and my colleagues Miss D. Mala-midou, Miss M. Valla, Dr. R. Christidou, Dr. Z. Tsirtsoni and Dr. S. Papadopoutos. Many thanks to Dr. V. Ki-likoglou for analysing the samples (SEM and chemical analysis) and for his helpful suggestions while I was preparing this work. I am grateful to Dr. K. Ko-tsakis, Dr. S. Andreou and Mr. A. Vargas for providing the facilities for the petrographic analysis and the re-firing tests at the Laboratory for the Toumba excavations, at the Aristotelian University of Thessaloniki. My thanks are also due to Mr. Ch, Romanidis for the drawings of figures. Finally, I wish to acknowledge the financial assistance of the Ecole Francaise dAthe-nes (the Dikili Tash Project), for the chemical analysis of the ceramic samples. REFERENCES ALOUPI E, MANIATIS Y. 1990. Investigation of the technology of manufacture of local LBA Theran pottery: the body and the paint. Thera and the Aegean World III. ANDREOU S., FOTIADIS M., KOTSAKIS K. 1996. Review of Aegean Prehistory V: The Neolithic and Bronze Age of Northern Greece. American Journal of Archaeology 100: 537-597. BAKALAKIS G, SAKELLARIOU A. 1981. Paradimi. Mainz. COURTOIS L. C. in press. Les techniques de la cera-mique. In Treuil R. (ed.), Dikili Task: Village Prehis-torique de Macedoine Orientate, I: Fouilles de J. Deshayes. Vol, II. Bulletin de Correspondance Helle-nique Supplement. DEMOULE J-P. 1993. Neolithique et Chalcolithique de Macedoine: Un etat des questions. APXAIA MA-KEAONIA V (1): 373-389. EFSTRATIOU N. 1991 Neoteres endixis gia tin proisto-riki egatastasi sti Thraki, Archeolgiko Ergo stiMake-donia ke Thraki 5: 425-437. EVANS R. K. 1986. The Pottery of Phase III. In C. Renfrew, M. Gimbutas and E. S. Elster (eds.), Excavations at Sitagroi: a Prehistoric Village in Northeast Greece Vol, 1: 393-428. FARNSWORTH M., SIMMONS 1.1963. Colouring Agents for Greek Glazes. American Journal of Archaeology 67:389-96. GARDNER E. J. 1979- Graphite painted ceramics. Archaeology 32(4): 18-23. 1980. The Pottery Technology of the Neolithic Period in Southeastern Europe. Los Angeles. GRABSKA-KULOVA M. 1993- Neolitische bemalte Ke-ramik aus der Siedlung Damjanica im mittleren Stru-matal. Saarbrucker Studien und Materialien sur Altertumskunde 2:121-55- GRAMMENOS D. 1991 • Neolithikes Ereunes stin Ken-triki kai Anatoliki Makedonia. Neolithiki Makedo-nia. HELLSTROM P. 1987. Paradeisos. A Late Neolithic Settlement in Aegean Thrace. Stockholm. HEURTLEY W. A. 1939- Prehistoric Macedonia. An Archaeological Reconnaissance of Greek Macedonia (West of the Struma) in the Neolithic, Bronze, and Early Iron Ages. Cambridge. JONES R. E. 1986. Greek and Cypriot Pottery: A review of scientific studies. London. KEIGHLEY J. M. 1986. The Pottery of Phases I and II. In C. Renfrew, M. Gimbutas and E. S. Elster (eds.), Excavations at Sitagroi: a Prehistoric Village in Northeast Greece Vol. 1: 345-90. KESSISSOGLOU M. D., MIRTSOU E. A. 1997. Ceramic Technology in Dimitra during the Neolithic Period. Appendix in D. Grammenos Neolithiki Makedonia: 84-94. KOUKOULI-CHRISANTHAKI C. 1972. Proistoriki iki-smi. Archeologiko Deltio 27 (B): 524-526. KOUKOULI-CHRISANTHAKI C, TREUIL R„ MALAMI-DOU D. 1996a. Proistorikos ikismos Philippon "Dikili Tash". Deka chronia anaskafikis ereunas. Archeolgiko Ergo sti Makedonia ke Thraki 1 OB: 681-704. KOUKOULI-CHRISANTHAKI C, TODOROVA H, ASLA-NIS Y, BOJADZIEV J., KONSTANDOPOULOU F, VAJSOV I., VALLA M. 1996b. Promachonas-Topolnitsa. Neoli-thikos ikismos Ellinobulganikon sinoron. Archeolgiko Ergo sti Makedonia ke Thraki I OB: 745- 768. MALAMIDOU D. 1997. Anakafi ston proistoriko iki-smo "Krioneri" N. Kerdillion. Archeolgiko Ergo sti Makedonia ke Thraki lli: 509-519. MALAMIDOU D„ PAPADOPOYLOS E. 1993. Anaskafi-ki ereuna ston proistoriko ikismo Limenarion Tha-sou, Archeolgiko Ergo sti Makedonia ke Thraki 7: 559-572. MANIATIS Y., TITE M. S. 1981. Technological Examination of Neolithic-Bronze Age Pottery from Central and Southeast Europe and from the Near East. Journal of Archaeological Science 8:59-7. PERNICHEVA L. 1995- Prehistoric Cultures in the Middle Struma Valley: Neolithic and Eneolithic. In D. W. Bailey and I. Panayotov (eds.), Prehistoric Bulgaria: 99-140. RENFREW C., GIMBUTAS M., ELSTER S. 1986. Excavations at Sitagroi, A Prehistoric Village in Northern Greece. Vol. 1. Los Angeles RICE P. M. 1987. Pottery Analysis. A sourcebook. Chicago and London. RYE 0. S. 1981. Pottery Technology: Principles and Reconstruction. Washington. SEFERIADES M. 1983- Dikili Tash: Introduction a la Prehistoire de la Macedoine Orientale. Bulletin de Correspondance Hellenique 107: 635-76. SHEPARD A. 0. 1976 Ceramics for the Archaeologist. Washingthon. TREUIL R. 1992 (ed.). Dikili Tash- Village Prehisto-rique de Macedoine Orientale, I: Eouilles de f. De-shayes. Vol, 1. Bullettin de Correspondance Hellenique, Supplement 24. TREUIL R, TSIRTSONI Z. 2000. Late Neolithic houses at Dikili Tash: a contextual approach. In S. Hiller and V. Nikolov (eds.), Karanovo Bannd III, Beitrage zum Neolithikum in Sudosteuropa: 213-216. TSIRTSONI Z. 2000. Les potteris du debut du Neoli-thique Recent en Macedoine: 1 Les types de receptions. Bullettin de Correspondance Hellenique 124 (in press). WAGNER H., GRAF R. 1993- Untersuchung keramis-cher Bemalung an ausgewahlten Scherben aus Dam-janica. Appendix in Grabska-Kulova Neolitische be-malte Keramik aus der Siedlung Damjanica im mit-tleren Strumatal. Saarbrucker Studien und Mate-rialien sur Altertumskunde 2: 153-55- YIOUNI P. 2001. Surface treatment of the Neolithic vessels from Macedonia and Thrace, in Greece. Annual of the British School at Athens 96 (in press). YIOUNI P., KOUKOULI-CHRISANTHAKI C., PLOUMIS P. 1994. Techonologiki analisi tis Neolithikis kerami-kis Promachonos-Topolnitsa. Archeolgiko Ergo sti Makedonia ke Thraki 8:343-348. UDK: 903(512)"632/634" Documenta Praehistorica XXVII On the Bailiandong Culture Zhou Guoxing Beijing Natural History Museum, Beijing, China yuanren@fm365.com ABSTRACT - The Bailiandong cave site was discovered in 1956, and was excavated and a preliminary study made of it by Beijing Natural History Museum and Liuzhou Municipal Museum from 1980 to 1982. A supplemental study from 1991 to 1993, funded by the Chinese National Science Foundation, found that its deposit included five continuous cultural strata from the Late Palaeolithic to the Neolithic with a transitional phase, and established a new framework for the Bailiandong Cultural Series. The Bailiandong cave site has incalculable value in attempts to explore how the Palaeolithic was transformed into the Neolithic in Southern China. IZVLECEK - Najdisce v jami Bailiandong so odkrili leta 1956, ga izkopali in med leti 1980 in 1982 pod vodstvom Prirodoslovno-zgodovinskega muzeja iz Pekinga ter Mestnega muzeja izLiuzhouja ob-javilipreliminarne rezultate. Nadaljne raziskave med leti 1991 in 1993je financiral Kitajski nacio-nalni sklad za znanost. Odkrili smo, daje medjamskimi depozitipet kontinuiranih kulturnih plas-ti odpoznega paleolitika do neolitika, ki vkljucujejo tudiprehodno fazo. Izdelali smo nov okvir za bailiandongsko kulturno zaporedje. jama Bailiandong je neprecenljive vrednosti pri raziskovanju preoblikovanja paleolitika v neolitik na juznem Kitajskem. KEY WORDS - China; Guangxi Province; Bailiandong cave; Palaeolithic; Neolithic; transition INTRODUCTION Bailiandong (White Lotus Cave) is located 12 km southwest of the city of Liuzhou (109°20' E, 24°15' N) and 2 km from the famous Liujiang Man site in Guangxi Province (Fig. 1). The Bailiandong cave is situated on Mount Baimian (White Face) and was discovered in 1956 when Pei Wenzhong (Pei Wen-chung) was surveying near the Gigantopithecus cave site. Isolated cultural remains were found in disturbed layers in Bailiandong cave, and were classified as late Palaeolithic by Jia Lanpo (Chia Lan-po) and Qiu Zhonglang in I960 (Chia and Qiu 1960). The Liuzhou Municipal Government designated it in 1961 as being among the key relics under municipal protection. From 1973 to 1980, Liuzhou Municipal Museum studied the relics several times and made a number of small-scale excavations, which recovered both, cultural and fossil mammalian remains (Zhou and Yi 1982). In 1980 and 1982, the Beijing Natural History Museum and Liuzhou Municipal Museum jointly conducted excavations and a preliminary study, which led to the conclusion that the cave site includes five continuous cultural strata from the late Palaeolithic to the early and middle Neolithic, with a transitional phase. The archaeological materials from the cave are referred to as the Bailiandong Cultural Series (Zhou 1984; 1986). In 1985, the Bailiandong Cave Science Museum was founded and opened. In 1991 funding was secured from the China National Natural Science Foundation (CNNSF) for further investigations over three years in an effort to explore the dating and ancient ecological environment of the cave's sediments and to establish a new framework for the Bailiandong Cultural Series. These investigations have resulted in a number of breakthroughs (Yi etal. 1994; Zhou 1994). THE GEOLOGICAL STRATIGRAPHY The cave is located on the southern slope of Mount Baimian, 152 meters above the erosional plain of the ground surface. It has a half-hidden entrance facing south and a long inner passage to the north. The 3 metre thick deposits can be divided into eight layers in the eastern part of the cave and ten layers in the western part (Liu and Xie 1994). Layer 7 of the eastern deposit is a thick calcareous concretion. In the western deposit, Layer 3 and the upper part of Layer 4 form another thick calcareous concretion, which merges with that of Layer 7 of the eastern deposit in the centre of the cave. This concretion has been dated to 18 500-20000 BP based on i*C dates, and represents the driest and coldest period of the Late Glacial period (Figs. 2-4). By means of uranium-series and l4C, especially AMS l4C determination, a number of quite valuable dating figures have been obtained (Yuan and Gao 1994) which amply prove that the deposit inside Bailiandong cave is composed of regularly successive strata (Tab. 1). Fig. 1. The Localities of Fossil man in the Liuzhou Region: 1. Gigan-topithecus Cave Site of Liucheng, 2. Liujiang Man Site, 3■ Bailiandong Cave Site, 4. Douledong Cave Site, 5. Ganqianyan Cave Site, A. Liuzhou, B. Saikwan, C. Lonmon, D. Tobo, E. Shantu, F. Qinde. The stratigraphic study and a spore-pollen analysis (Kong et al. 1994) have fully revealed the tendency of changes in the ancient ecological environment and climate of Liuzhou region since 30 000 BP which was synchronised with global changes in the ancient climate (Tab. 2). In fact, the Bailiandong deposits and relics have clearly demonstrated that the cave is a type of rare reservoir of global climatic information in the southern subtropics since the Late Pleistocene Glacial period. The recovery of data on the ancient ecological environment of Bailiandong cave is actually a background to the research of exploring the birth of agriculture in South China. East 1 7080±125 BP Holocene Atlantic Time East 2 9520±90 BP Boreal Time West 1 10310±290 BP East 3 11160+580 BP Late Pleistocene Late Glacial Epoch Wurm Alerad Time East 4 13550±590 BP Boiling Time West 2 12 775±180 BP East 5 13905±250 BP East 6 14 650±230 BP Wurm II West 3 18450±410 BP East 7 19645±200 BP West 4 19910±180 BP East 8 20 240±660 BP West 5 West 6 28000±2000 BP Paudorf West 7 West 8 West 9 West 10 37 000+2000 BP Wurm I Tab. 1. Dating Figures of the Main Layers of Bailiandong Cave. Age Western Deposit West 1 (0.2-0.56 m) Greyish-brown mildciay Upper 10310±290 bp Lower 17680±300 BP West 3 (0.15-0.35 m) Yellowish-brown thick flowstone (upper sideward stone dike) 18450*410 BP West 2 (0.4 m) Milk-yellow flowstone Top Layer 12780±180 BP Bottom Layer 19145±180 BP West 4 (0.5 m) Yellowish-brown thick flowstone (lower sideward stone dike) Top Layer 19910±180 BP 21 575±150 BP Bottom Layer 26 680±625 BP West 5 (0.3-0.5 m) Redish-brown miidclay West 6 (0.1 m) Yellowish flowstone 28 000±2000 BP West 7 (0.18 m) Yellowish-brown miidclay West 8 (0.1 m) Greyish-yellow flowstone West 9 (0.12 m) Brown mildciay West 10 (0.15 m) Milk-yellow flowstone 37 000±2000 BP (Bedrock not yet reached) Eastern Deposit East 1 (0.2-0.32 m) Flowstone with pottery flakes Calcareous miidclay 7080±t25 bp East 2 (0.3 m) Milk-white flowstone 7140±60 bp and calcareous miidclay 9520±90 bp East 3 (0.3-0.37 m) Greyish-yellow miidclay n i6o±58o bp East 4 (0.38 m) Yellowish-brown miidclay 13905±250 BP East 5 (0.01-0.04 m) Greyish-white flowstone 13905±250 bp East 6 (0.48 m) Brown miidclay 14650±230 bp East 7 (0.44 m) Yellowish-brown thick flowstone (upper-sideward stone dike) 11 670±150 BP 19645±200 BP East 8 (>1 m) Red-brown miidclay 20 240±660 BP (Bedrock not yet reached) Ancient Climate and Ecological Environment Great warm epoch of the Holocene. The appearance of many subtropical evergreen broadleaf forest, graminaeceous, fern and water plants. Subtropical evergreen broadleaf forest, the climate was beginning to turn warm and humid. Appearance of the subtropical evergreen broadleaf forest and modern fauna. The climate was cold and dry and characteristic of the maximum period of the Late Glacial epoch. Mixed forests of temperate mountain conifer and subtropical broadleaf in the low lands. Cold or temperate conifer forest in the mountains. Warmth-loving mammals migrate to the south. The climate was beginning to turn cold. Temperate broad leaf forest. Stegodon-Ailuropoda fauna. Atlantic Time Boreal Time Alerad Time Boiling Time Wurm II Paudorf Wurm Tab. 2. Schematic Chronology based on Geological Stratigraphy and Paleoenvironmental data of Bailiandong Cave Site THE ARCHAEOLOGICAL DATA Large quantities of mammalian fossils have been found in the cave's deposits. The bones are very fragmentary and may represent the food remains of the prehistoric occupants. One thousand five hundred bones and teeth were found in the eastern deposits and 2000 in the western. Most of the teeth are isolated specimens: 150 derive from the eastern deposits and 240 from the western. The mammalian fossils identified were from 23 species. These are: Rhizomes sp., Hysterics subcristata, Macaca sp., Homo sapiens sapiens, Martes sp., Paguma larrvata, Vulpes cf. vulgaris, Ursus sp., Arctonyx collaris, Ailuropoda melanoleuca, Sus scrofa, Bubalus sp., Pseudaxis sp., Muntiacus sp., Cervus sp., Lijiangocerus speciosus, Ovis sp., Rusa unicolor, Rhinoceros sinensis, Stegodon sp., Ele-phas sp, Muridae gen. etsp. in del and Vespertilio-nidae gen. etsp. indet. Side stone dike Steep slope Stalagmite Breccia Deposit and strata number Lower hole Fig. 2. Plan of the Bailiandong cave Site. A number of extinct species and two isolated human teeth were recovered from Layer 7 of the western deposit. The human teeth, which comprise the right molar of a young female and the left third molar of a middle-aged male, are attributable to Homo sapiens sapiens. The faunal assemblage from the western deposit represents the fauna of Stego-don-Ailuropoda with Homo sapiens sapiens, while the eastern deposit consists solely of modern mammalian specimens. Shells from five species of snail ('Viviparus dispiralis, Bellamys leei, Helix sp., Semosulcospira sp. and Unio douglaside) were found in the upper layers of the cave. Layers 5 and 7 in the western deposit contain only a few snail shells. In addition, two species of fish (Cyprinus carpio and Mylopharyngodon Fig. 4. Section of Western deposits of the Bailiandong Cave Site. piceus) were also found along with frog (Rana sp.), turtle (Testudinidae indet.) and bird limb bone remains (BSCM and BNHM 1987). Of the 500 chipped stone artefacts found at the site, 258 have been studied in detail. These specimens comprise cores, unused and used flakes and modified tools. The large tools were made on cobbles and pebbles, while the small tools were manufactured on flakes of black flint. Most of the small tools were found below the western thick calcareous concretion and a few pieces derive from the snail shell layers in the eastern part of the deposit. Bone, antler and polished horn tools have also been recorded. Three ground stone tools, including two completely ground and one ground only at the edge, and two perforated decorative objects were found in the upper layer of the eastern deposit. In the eastern deposit a few fragments of crude cord-marked pottery were found in the top layer. The small flint tools were not made with the typical indirect percussion manufacturing technique {Chen 1983; Chia 1978; Zhou 1974), but with the anvil technique applied in the Danawu tradition in Yunnan (Zhou and Zhang 1984). There are three broken weight stones - perforated pebbles that were found separately in Layer 1 of the western deposit, Layer 6 and Layer 3 of the eastern deposit. One specimen in Layer 3 is completely ground. Two hearths were found in Layer 5 of the western deposit (Figs. 5-9). Fig. 6. Stone tools from West layer 5:1. and 3• sra-pers; 2., 5. and 6. utilized flakes; 4. thumbnail scraper (black flint). Fig. 5. Stone tools from West layer 7:1. arrowhead (black flint); 2. utilized flake; 3■ scraper; 4. sraper; 5. chopper. The artefact layers of the cave can be subdivided into five cultural strata as follows: Cultural Stratum 1 - Layers 1 and 3 of the eastern deposit are grayish-yellow mildclay and flow-stone and contain pottery flakes. They also contain snail shells, ground stone tools, perforated decorative objects, weight stones and pottery. Between Cultural Stratum 1 and Cultural Stratum 2 there is a thin calcareous concretion in Layer 3 dated to 11160± 580 BP. Cultural Stratum 2 - Layer 4 of the eastern deposit is yellowish-brown mildclay. It contains snail shells, stone tools and horn tools with polished edge and point. Cultural Stratum 3 - Layer 6 of the eastern deposit is a brownish mildclay. It contains snail shells, ash, red burnt clay, burnt bones and stones. This layer also includes one crude weight stone, crude pebble tools and hematite powder. This layer is always found overlying thick, hard travertine and overlain by another distinct travertine (calcareous concretion). Cultural Stratum 4 - Layer 4 of the western deposit is a shallow yellowish-brown mildclay; the upper portion is a calcareous concretion. Large numbers of very small flint tools, including an arrowhead, a small crude polished cutting tool and some spoke-shaves were found here as well as crudely made pebble tools. Cultural Stratum 5 - Layers 5 and 7 of the western deposit are red-brown and yellowish-brown mild- Fig. 7. Stone tools from West layer 2: 1. - 4. choppers (black flint artifacts); 5. and 6. srapers; 1. graver; 8. core; 9■ arrowhead; 10. point. occurred during the Late Pleistocene to the beginning of the Early Holocene, perhaps over a period of several thousand years. Phase III - Early and Middle Neolithic (Cultural Stratum 1): In this phase riverine adaptation is still well developed, but the appearance of completely ground stone artefacts including weight stones and crude cord-marked pottery represents evidence of the Neolithic at Bailiandong. The patterns of development of human adaptation shown in the cultural phases and data on the ancient climate and ecological environment at Bailiandong are of incalculable value in attempts to explore how the Palaeolithic transformed into the Neolithic. In addition to justifying the conclusions reached by former researchers, such as that the Bailiandong site contains continuous cultural strata from the Palaeolithic to the Neolithic with a transitional phase, using the most recent research results on the Bailiandong Cultural Series it has become possible to establish a new framework (Tab. 3). After inspecting a number of other sites of similar age in South China, including Yunnan, (Aigner 1981; Hu 1977; Lin and Zhang 1978; YPM 1977; Zhang et al. 1978), Guizhou, (Cao 1982a; 1982b; Li and Zhang 1981; Li and Cai 1986), Guangxi (Chang 1977; GZARM 1983; He and Qing 1985; Chia and Woo 1959; Li and You 1975; Li et al, 1984; Pei 1935; Pei 1965; Wang et al. 1982; Wu et al. 1962; Zhao et al. 1981) Guangdong (GPM1959; Huang et clay. Layer 5 contains a few snail shells. Very small flint artefacts and pebble tools in Layers 5 and 7 are typical of the Late Palaeolithic. The artefacts found in the deposits of the cave can be put in the following order, from the oldest to the most recent: (a) Typical Palaeolithic stone tools, (b) Small and very small flint tools, (c) Crude pebble tools, (d) Crude weight stone, (e) Crude ground tools and (f) Crude pottery. According to the lithic, faunal, spore-pollen analyses and the stratigraphic dates, the Bailiandong Cultural Series may be divided into three cultural phases. Phase I - Late Palaeolithic (Cultural Stratum 4 and 5): The lithic artefacts are predominantly characterised by specimens typical of the Late Palaeolithic. The main subsistence mode was hunting large game and gathering plant food. Phase II - Mesolithic (Cultural Stratum 2 and 3): This is a transitional phase (from Palaeolithic to Neolithic) characterised by unifacially worked pebble tools (choppers) and crudely made ground tools. The appearance of large numbers of snail shells and chopping tools, which may have been used to crush the shells, is evidence of the intensification of riverine resource procurement. The occurrence of a weight stone is related to primitive agricultural activities. This transformation of subsistence patterns Fig. 8. Stone tools from East layer 2: 1. chopping tool; 2. perforated pebble (weight stone); 3- cores; 4. cores; 5- pebble for grinding hematite powder; 6. chopping tool; 7. chopper with high-back. Fig. 9. Antler and stone tools from East layer 4:1. end-ground an tler; 2. edge-ground flat pebble; 3. chopper. al. 1982; Qiu et al, 1986; Song et al. 1981), Sichuan, (Pel and Woo 1957; Yang 1961; Zhang 1977) and Jiangxi provinces (JACR 1963) (Tab. 4), it can be concluded that none of these contain evidence of three succeeding cultural phases beginning at more than 30 000 years ago. CONCLUSION Based on the cultural framework established for Bailiandong, the author has conducted a synthetic study of the relevant contemporary archaeological materials of South China and drawn up a possible picture of the Mesolithic age of this region: O The faunal assemblages according to Layers 4 and 6 of the eastern deposit of the Bailiandong cave mainly consist of modern mammalian species such as deer and buffalo. The few extinct species include taxa such as elephant and rhinoceros. At the time of human occupation, the flora was temperate broad-leaf forest in the lowlands and boreal conifer forest in the mountains. This is also reflected in the colour of this deposit, which is brownish instead grey (like the upper layers). © The lithic industry of the Mesolithic in South China was predominantly characterised by two types of technique. One technique was the manufacture of crude choppers and chopping-tools for which quartzite was the main material selected. The second technique produced finely made flint tools, axes with polished cutting edges and a number of terminal pieces of deer antler exhibiting traces of polish. Many of the tools were apparently used for specific functions, including crude agricultural tools (such as weight stones), tools for grinding food (seeds and roots) and crushing snail shells, and tools for grinding hematite powder. At a few sites (such as the Miaoyan - Temple Cave near Guilin) in South China, the earliest pottery is dated to 15 000 BP (Yuan et al. 1995) © Based on the evidence from Phase 2 of the Bailiandong Cultural Series, the Mesolithic of South China is dated to 18000-12 000 BP and it may be divided into two sub-phases: the first (earlier) is significantly characterised by the occurrence of crude perforated pebbles (e.g. weight stone), and the second (later) by partly polished stone, bone and horn tools. The palaeo-climate and environment varied dramatically during the terminal Pleistocene and the beginning of the Holocene, and witnessed the emergence of new subsistence patterns. 0 The social economy began to transform during the Mesolithic from a hunter-gathering subsistence pattern to one of agriculture. During this transfor- Cultural Phases Phase 1 Phase 2 Phase 3 Components 5th & 4th Cultural Strata 3rd & 2nd Cultural Strata 1st Cultural Stratum Layer West 7,5 & 4 East 6 & 4 East 3 & 1 Time Span 30 000-18 000 years ago 18 000-12 000 years ago 12000-7000 years ago Cultural Stages Late Palaeolithic Transitional (Mesolithic) Early & Mid Neolithic Periods of Each Cultural Phase A. Subphase B. Subphase A. Subphase B. Subphase A. Subphase B. Subphase Occurrence of Typical Articles of Each Subphase A. Chipped stone implements with Palaeolithic traits & tiny flint stone artefacts. B. Primitive ground articles & arrowheads. A. Roughly made pebble tools & primitive holed pebbles, and hematite powder. B. Edge- or end- ground tools A. Overall ground stone tools & primitive pottery. B. Pottery flakes Tab. 3■ New Framework of Bailiandong Cultural Series. s Bailiandong Cultural Series cn c Phase 1 Phase II Phase III "O c <0 Cultural Strata 5 and 4 Cultural Strata 3 and 2 Cultural Stratum 1 m Upper Palaeolithic Mesolithic Early & Middle Neolithic LiuJiang Liujiang Man Laibin Qilinshan Wannian Xianrendong LiuJiang Gangqinyan Liuzhou Sidouyan Wuyuan Qingtan Duan Jiulengshan Liujiang Chenjiayan Guilin Zengpiyan Guilin Baojiyan Chongzuo Aidong Liuzhou Dalongtan t/i CD Hanyuan Fulin Yangchun Dushizidong Xinyi (Upper Cultural Level) cn O) Chenggong Longtanshan Fengkai Huangyandong Naihai Maomaodong c 'u c Tongliang Zhangretan (1-3 Sites) Wannian Hsichiaoshan o Q. to Baise Shangsong Wuming Baqiao Xianrendong o O 0) Village Baxun Leipin Chilinshan Tengxiang cn w o Tongliang Zhangrentang Guilin D-Cave Changgong Longtanshan Miaoyan Tzeyang Wangshanqi Liuzhou Dalongtan Tobo Ganqiangyan LiJiang Puding (Lower Cultural Level) Mofaiqiao Baiyanjiao Tab. 4. Bailiandong Cave and Possible Corresponding Sites. mative age of the Mesolithic, subsistence was still predominantly characterised by hunting and gathering, although the procurement of riverine food resources was well developed and incipient agriculture may have occurred. © Mesolithic settlements were located near karst caves and rock shelters in the vicinity of streams or rivers. © The human remains of the Mesolithic of South China often exhibit morphological features of the Oceanic Negritos. For instance, the Dalongtan homi-nids that lived in the Dalongtan shelter-like cave which consists of two cultural strata. The human fossils were excavated from the upper most layer of the lower cultural stratum, and have been dated to 12 000+220 BP, which means that the Dalongtan specimen represents a hominid transitional between Liujiang Man of the Late Palaeolithic (Woo 1959) and Zengpiyan Man of the Neolithic (Zhang et al. 1977). The Dalongtan cranium No. 2, which is relatively well preserved, belongs to a male about 30 years of age. Morphologically it displays the main traits of Mongoloids, and also shows some racial traits typical of Oceanic Negritos. Observation of the morphological features and analysis of the measurements have concluded that the traits of Dalongtan Man probably resulted from the genetic mixing of southern Chinese hominids with people from farther south (Zhou and Zhang 1994). Fig. 10. Stone tools from East layer 3: 1■ utilized flake; 2., 4. and 5. choppers; 3■ chopping tool; 6. perforated pebble (ground weight stone); 7. and 8. perforated decorated objects; 9- ground cutting tool; 10. fragments of cord-marked pottery. REFERENCES AIGNER S. 1981. Archaeological Remains in Pleistocene China. Munchen: Verlag C. H. Beck. BSCM - Bailiandong Science Cave Museum and BNHM - Beijing Natural History Museum 1987. Archaeological Finds in the Bailiandong Cave Site of Stone Age. Southern Ethnology and Archaeology 1: 143-160. CAO Z. 1982a. On the Palaeolithic Artefacts front Maomaodong (the Rock-Shelter Site), Guizhou Province. Vertebrata Pal. Asiatica 20: 36-41. 1982b. The Preliminary Study of Bone Tools and Antler Spades from the Rock-Shelter Site of Maomaodong. Acta Anthropologica Sinica 1:18-19. CHANG K 1977. The Archaeology of Ancient China. 3rd edition. New Haven: Yale University Press. CHEN C. 1983. Preliminary Exploration of the Typology and Technology of Microcores in China - Also the Culture Relationship between Northeast Asia and Northwest America. Acta Anthropologica Sinica 2:331-341. CHIA L. 1978. On the Phase, Origin, and Tradition of Micro tool Industry in China. Vertebrata Pal. Asiatica 16: 137-143. CHIA L. and QIU Z. i960. On the Age of the Chipped Stone Artefacts in Kwangxi Caves. Palaeovertebrata et Palaeoanthropologia 2: 64-68. CHIA L. and WOO R. (Wu, Rukang) 1959- Fossil Human Skull Base of Late Palaeolithic Stage from Ch'i-linshan, Leipin District, Kwangxi, China. Palaeovertebrata et Palaeoanthropologia 1: 16-18. GUANGDONG PROVINCE MUSEUM 1959- Stone Tools Excavated at Hsich'iao-shan, Nan-hai County, Guangdong. Acta Anthropologica Sinica 4:1-15. GUANGXIZHUANG AUTONOMOUS REGION MUSEUM 1983. The Discovery of Human Teeth and Mammalian Material from Du'an, Guangxi. Vertebrata Pal. Asiatica 11: 221-223. HE H. and QING S. 1985. On the Mesolithic of Guangxi and Guangdong. Acta Anthropologica Sinica 4: 308-313- HU S. 1977. Palaeolithic Discoveries in Chenggong County, Yunnan. Vertebrata Pal. Asiatica 15: 225. HUANG W„ LI C, WANG H, and HUANG Y. 1982. Reexamination of a Microlithic Site at Xiqiaoshan, Nan-hai City, Guangdong. Current Anthropology 23:487-492. (JACR) Jiangxi Administration of Cultural Relics 1963. Preliminary excavation at the Xianrendong cave site, Wanian, Jiangxi. Kaogu Xuebao 63:1-15. KONG Z., DU N., SONG C. and ZHANG J. 1994. Pa-lynological Analysis of the Bailiandong Cave Site and its Significance in Palaeovegetation and Palaeoenvi-ronment. In Zhou G., Zhang Y. and Zhang X. (eds.), International Symposium Relationship between Chinese and Japanese Ancient Men and Prehistoric Cultures: 176-201. LI Y. and ZHANG S. 1981. On Palaeolithic Culture of Tongliang County. Vertebrata Pal. Asiatica 19:359-371. LI Y. and CAI H. 1986. A Study of the Typology of Stone Tools from the Cave of Baiyanjiao, Puding, Guizhou. Acta Anthropologica Sinica 5:317-324. LI Y. and YOU Y. 1975. On the Discovery of Palaeolithic in Baise, Guangxi. Vertebrata Pal Asiatica 13: 225-228. LI Y., WU M, PENG S. and ZHOU S. 1984. Human Tooth Fossils and some Mammalian Remains from Tobo, Liujiang, Guangxi. Acta Anthropologica Sinica 3: 322-329. LIN Y. and ZHANG X. 1978. Mammalian Fossils and Palaeolithic from Lijiang Basin, Yunnan. Professional Papers of Stratigraphy and Palaeontology 7:80-85. LIU X. and XIEJ. 1994. Stratigraphic Geology of Bailiandong. In Zhou, Guoxing, Zhang, Yan, and Zhang, Xiaohua (eds.), International Symposium Relationship between Chinese and Japanese Ancient Men and Prehistoric Cultures: 156-174. PEIW. (Pei Wen-chung) 1935. On a Mesolithic Industry of the cave of Kwangsi. Bulletin of the Geological Society of China 14: 393-412. PEI W. 1965. Excavation of Liucheng Gigantopithe-cus Cave and Exploration of Other Caves in Kwang-si. Memoirs of the Institute of Paleontology and Paleoanthropology 7. PEI W. and WOO J. (Wu R.) 1957. Tzeyang Man. Memoirs of the Institute of Paleontology and Paleoanthropology Series A 1: 1-171. QIU L., SONG F. and WANG L. 1986. A Fossil Human Tooth front Fengkan, Guangdong. Acta Anthropolo-gica Sinica 4:308-313■ SONG F„ QIU L. and WANG L. 1981. Preliminary Investigation of the Huangyandong Site, Fengkan, Guangdong. Vertebrate Pal. Asiatica 19: 98-101. WANG L., PENG S. and CHEN Y. 1982. On the Human Fossils and Stone Artefacts Found in Baojiyan Cave, Guilin. Acta Anthropologica Sinica 1: 30-35. WOO J. (Wu, Rk.) 1959. Human Fossils Found in Liu-kiang, Kwangsi, China. Palaeovertebrata et Palaeo-anthropologia 1: 97-104. WU X., ZHAO Z., YUAN Z. and SHEN J. 1962. Report on Palaeoanthropological Expedition of the Northeastern Part of Kwangsi. Vertebrata Pal. Asiatica 6: 408-418. YANG L. 1961. Discovery of the Palaeolithic from Fu-linchen, Hanyuan, Sichuan. Vertebrata Pal. Asiatica 6: 139-142. YI G., ZHANG X. and ZHOU G. 1994. The Discovery - Earlier Stage of Research and Later Progress Made in Study of Bailiandong Site. In Zhou, Guoxing, Zhang, Yan, and Zhang, Xiaohua (eds.), International Symposium Relationship between Chinese and Japanese Ancient Men and Prehistoric Cultures: 147-155. YUAN S. and GAO S. 1994. Supplemental Study on the Ages of the Deposits of Bailiandong Site. In Zhou, Guoxing, Zhang, Yan, and Zhang, Xiaohua (eds.), International Symposium Relationship between Chinese and Japanese Ancient Men and Prehistoric Cultures: 175- YUAN S, ZHOU G, GUO Z. and GAO S. 1995. l4C AMS Dating the Transition from the Palaeolithic to the Neolithic in Southern China. Radiocarbon 37: 101-105. YPM - Yunnan Province Museum 1977- Notes on Li-jiang Man's Skull from Yunnan. Vertebrata Pal. Asiatica 15: 157-161. ZHANG S. 1977. On Fulin Culture. Vertebrata Pal. Asiatica 15: 14-27. ZHANG X., HU S. and ZHENG L. 1978. The Late Pleistocene Human Teeth from Kunming, Yunnan. Vertebrata Pal. Asiatica 16: 288-289. ZHANG Y., WANG L. and DONG X. 1977. The Human Skulls from the Zhengpiyan Neolithic Site at Guilin, Guangxi. Vertebrata Pal. Asiatica 15: 4-13- ZHAO Z„ LIU X. and WANG L. 1981. Human Fossils and Associated Fauna of Jiulenshen Hill, Guangxi. Vertebrata Pal. Asiatica 19: 45-54. ZHOU G. 1974. Stone Age Remains from Lingjiang, Xuchang, Henan. Kao Gu (Archaeology) 2: 91-98. 1984. The Discovery of the Bailiandong Cave Site and its Significance. Prehistoric Investigation 2: 34-37. 1986. On Bailiandong Culture. Memoirs of Beijing Natural History Museum 40: 1-25. 1994. The Supplemental Study on the Bailiandong Cultures. In Zhou, Guoxing, Zhang, Yan, and Zhang, Xiaohua (eds.), International Symposium Relationship between Chinese and Japanese Ancient Men and Prehistoric Cultures: 203-262. ZHOU G. and YI G. 1982. On the Remains from Liu-zhou Region, Guangxi. Memoirs of Beijing Natural History Museum 20: 1-23- ZHOU G. and ZHANG X. 1984. The remains of Micro-lithic tools found in the Yuanmou Basin. In Zhou, Guoxing and Zhang, Xingyong (eds.), Yuanmou Man: 113-123- ZHOU G. and ZHANG Z. 1994. A Study on Remains of Liuzhou Dalongtan Man of Mesolithic to Early Neolithic. In Zhou, Guoxing, Zhang, Yan, and Zhang, Xiaohua (eds.), International Symposium Relationship between Chinese and Japanese Ancient Men and Prehistoric Cultures: 59-88. UDK: 903(512)"632/634' Documenta Praehistorica XXVII The Mesolithic in South China Zhang Feng Dept. of Anthropology, Zhongshan University, Guangzhou, P. R. China zhfeng@public.guangzhou.gd.cn ABSTRACT - The history of the study of the Mesolithic in China is longer than sixty years. In south China many cave sites relating to the Mesolithic have been found. Typological implements excavated in these sites imply that the cultural transition in this period was closely related to climatic changes that cause environmental diversification, and inevitably lead to changes in humans' mode of subsistence. Although the discussion of the existence of Mesolithic culture in China is still a controversial topic to many Chinese archaeologists, the author insists that archaeologists pay more attention to the subsistence mode of ancient people in this transitional period, than become immersed in the traditional historiographic orientation, IZVLECEK - Zgodovina raziskav mezolitika na Kitajskem je dolga ze vec kot sestdeset let. Vjuzni Ki-tajski so bila odkrita stevilna jamska najdiscu, ki so povezana z mezolitikom. Izkopani predmeti iz teh najdisc kazejo, da je bil kulturni prehod tega obdobja tesno povezan s klimatskimi sprememba-mi, ki so povzrocite spremembo okolja, kar je neizogibno vodilo v spremenjen nacin prezivljanja ta-kratnega cloveka. Ceprav je razpravljanje o obstoju mezolitske kulture na Kitajskem za mnoge kitaj-ske arheologe se vedno sporno, avtor clanka vztraja pri mnenju, naj arheologiposvecajo vecpozor-nosti nacinu prezivljanja starodavnih tjudi v tem prehodnem obdobju, in naj se ne poglabljajo toli-ko v tradicionalno zgodovinsko usmeritev. KEY WORDS - South China; Mesolith ic; pebble tools; hunting-gathering; broad-spectrum food INTRODUCTION: MESOLITHIC STUDIES IN SOUTH CHINA1 The history of the studies of the Mesolithic in China has not been more than seventy years since the archaeologists began to pay attention to this topic in the mid-1950s (Zheng 1936.20, 54). In 1934, Prof. Pei, one of the earliest Chinese archaeologists, surveyed three limestone caves, namely Baqiao, Baxun, and Tengxiang, in Wuming County, and D cave in Guilin, Guangxi Province2 in south China, and found some pebble artefacts. He thought some of these stone implements bore a few characteristics similar to those of the Hoabinhian Culture, the famous Me- solithic in North Vietnam, and implied these could belong to the Mesolithic (Pei 1935.393-412). Also, these sites were further identified as Mesolithic by the archaeologist An twenty years later (An 1956. 36). From then on, the study on Mesolithic in China never stopped. In 1960s and 1970s, more findings relating to the Mesolithic were found in South China. These include: Gaitou Cave, Chenjia Cave in Liujiang, Aidong Cave in Congzuo County, Guangxi Zhuang Autonomous Region (f ia et al, 1960.64-68)-, Qing-tang in Yingde County, Guangdong Province (Peng 1 According to the accepted common practice, 'South China' geographically refers to the area of Pearl River Valley, which covers from eastern Guangdong in the east to the eastern edge of the Yungui Plateau in the west, and from Wuling Mountains in the north to Hainan Island in the south. Accordingly, South China in this article includes the whole area of Guangdong Province and Guangxi Zhuang Autonomous Region, the main body of Pearl River Valley, which is about 0.41 million km2. 2 In this article Guangxi Province refers to as the same place as Guangxi Zhuang Autonomous Region. 1961.585-588); Dongyan Cave in Guangxi (Wit, Xin-zhi 1962.408-411); Zengpiyan Cave in Guilin, Guangxi (Working Team For Cultural Relics of Guangxi Zhuang Autonomous Region 1976.20). After 1980, more systematic surveys were done in south China, in which more Mesolithic sites were found, such as Dushizai in Yangchun, Huangyan Cave, Dongzhongyan and Luojiyan in Fengkai County, Guangdong (Song et al. 1981.292-293; 1991.1-12); Miaoyan Site in Guilin, Liyuzui Site in Liuzhou, Guangxi (Liuzhou Museum 1983.769-774); II second stratum in Bailian Cave in Guangxi (Yang 1991.154; Kong et al, 1994.147-155)] Luo-sha Cave in Guangdong (Zhang 1994.300-308), and Niulan Cave in Yingde, Guangdong (Qiu, Li-cheng et al, 1999.1-111). Most of these sites are limestone caves located at branches of Xijiang River Valley and Beijing River Valley (Fig. 1). Generally, all of these sites are located in valleys and small alluvia basins of perennial rivers. Most of them are found in caves of limestone hills, which are topographically common in mountainous areas in south China. The typological findings of these limestone caves are pebble tools including chipped scrapers, choppers, stone gravers, and holed stone, bone implements, remains of mussels and shells, shell tool, and animal bones. At some site are found a few stone arrows and flint microlithics (Fig. 2). Radio-carbon dating shows the earliest date of these Mesolithic remains is more than 15 000 BP (Bailian Cave) and the latest date is earlier than 9000 BP (Zengpiyan Cave), which means the cultures of these sites exist between the last stage of the late Pleistocene and the first stage of the early Holocene. THE SUBSISTENCE OF MESOLITHIC PEOPLE IN SOUTH CHINA A systematic and dynamic review of Mesolithic findings at these sites in south China exposes the socioeconomic structure and the subsistence style in this period. The culture of this period obviously not only carries some old characteristics of Palaeolithic culture, but also some new elements for Neolithic culture, which is mostly shown in the composition of implement typology. In south China, the dominance of large and medium sized pebble tools had existed for a long time since the early Palaeolithic. One of the examples is that of pebble tools found in the terraces along the You River in Guangxi, which is dated to as early as 700 000 BP (Huang et al. 1990.105-112). The pebbles were selected by ancient people from a nearby riverbed. The technology of these pebbles is simple, since most of them are one-side chipped and have wide and flat tops and deep flake scars. Choppers, scrapers, hammers, and drills are common types of pebble implements. While pointed tool, tools for sculpture are seldom found. The composition of these tools had lasted from the early to late Palaeolithic in this area. While in Mesolithic times, the skills of making pebble tools were not only inherited, but also obviously improved: pebble rngvang Fig. 1. Localities of Mesolithic Sites in South China: 1. Winning County (Baqiao, Baxun and Tengxiang Caves); 2. Gaitou Cave; 3■ Chengjia Cave; 4. Aidong; 5. Qingtang; 6. Dongyan Cave; 7. Zengpiyan Cave; 8. Dushizai; 9. Huangyan Cave; 10. Dongzhongyan Cave; 11. Luojiyan; 12. Miaoyan Cave; 13- Liyuzui; 14. Bailian Cave; 15. Niulan Cave. Fig. 2. The main Findings in some Mesolithic Sites in south China. Site Location Stratified Deposit Main Findings Radio-carbon dating (BP) Liyuzui Liuzhou City, Guangxi Province Lower layer Chipped pebble tools, stone artefacts with polished edge, holed stones, a few bone tools, animal bones such as needle, awl, and knife, mussels, a little piece of cored sandy pottery 12 880 Zengpiyan Cave Guiling City in Guangxi Province Early stratum Chipped pebble tools, stone artefacts with polishing edge, hole stone, grinding stones, mussel bones, animal bones, sandy pottery shards More than 9000 Bailian Cave Liuzhou City, Guangxi Province Middle layer (II period) Chipped pebble tools including choppers, scrapers, holed stones, stone artefacts with polishing edge, flint microlithics including arrows, points, two-sided scrapers, one-sided scrapers, and stone gravers, shells, animal bones 15910 Huangyan Cave Fengkai County, Guangdong Province Middle layer Quantity of pebble scrapers, pebble choppers, and hammers, stone awl stone artefacts with polishing edge, a few holed stones, shells, animal bones 10950 Dushizai Yangchun County, Guangdong Province Upper layer Chipped pebble tools, stone artefacts with polishing edge, bone tools, shells, bone arrow 11 500 Middle layer Chipped pebble tools, holed bone tools, bone tools, shells, animal bones 14260-15 350 Niulan Cave Yingde County, Guangdong Province Middle layer Chipped pebble tools, stone artefacts with polishing edge, holed bone tools, shell net-weights, shells, animal bones, silicinized remains of rice 10450 tools were usually made in oblique, straight, and sharp edge. Almost all the angles of the oblique-edged pebble tools are more than 750. Meanwhile, the chipped and flake scars on the oblique-edged tools are more systematically distributed than those on the pebble tools of Palaeolithic. In Huangyan Cave and Dushizai site, hundreds of such artefacts are excavated, and dominate the stone artefacts (Fig. 3). Interestingly, the dominance of and the manufacturing skills for the pebble stone are very similar to those of Hoabiahian culture in Southeast Asia. Another interesting phenomenon in some sites in south China is the emergence of holed stones. Hole stone artefacts are obviously part of composite implements, and can be regarded as the model of some advanced tools in Neolithic. These pebble tools are closely related to the ecology at that time if we look at climate changes. In the early part of the Holocene, the last ice age ended, and as the glaciers slowly melted away the weather became hot and humid in the subtropical region including south China and Southeast Asia. The widely distributed and most appropriate materials available for these tools in hot and wet south China were bamboos, woods, and lianas. In view of the function of traditional tools in the Palaeolithic, large pebble choppers were not suitable for processing these raw materials. Instead, people tended to use a different pebble tool, a pebble similar to that in the Palaeolithic, but with an oblique edge. We guess that these kinds of tools were used as intensive and efficient choppers to process bamboos, woods and liana Another factor of socio-economic development manifested is the large amount of remains of shells of oysters, clams, and mussels. In almost every cave, accumulated mounds of discarded shells were excavated. This reflects that people at that time had learned to intensively utilise aquatic resources by gathering shellfish along lakesides, streams, and coastlines. Fishing in the environment in Mesolithic times was easier than hunting, since seasonal changes did not have too much influence on aquatic molluscs in the environment, and consequently did not affect human engagement in fishing. Conversely, hunting-gathering production is dependent on seasonal changes. But a question arises here: how could ancient people fish easily with their relatively undeveloped technology? Contemporary ethnographical data can help answer this. In northeast China, some minorities nowadays merely use a harpoon made from the branch of a tree, or a small wicker basket to fish easily in shallow brooks. We can imagine that ancient people in the Mesolithic understood how to catch fish in a simple way or by using naturally occurring tools, such as branches with forks, as human do now. Consequently, people began to use bones or shells as implements, such as mussel knives and bone knives. Furthermore, this hunter-gatherer economy brought about the budding stage of agriculture. As is well known, there are several theories/hypotheses about the origin of agriculture and animal domestication. Whatever they are, all of them need some pre-conditions. Objectively, there must be an appropriate ecology that offers food resources and an environment for the domestication of plants; subjectively, population, human skills in getting food, primitive thoughts, values and customs, and social organisations can affect human attitudes to domestication. With the changes of climate and ecology in south China in the period between the late Pleistocene and early Holocene, more food resources were available and there was surplus food. It is not surprising that hunters and gatherers thought to apply their learning of the regulations of plant and animal rearing in different K.P89119 X 1/6 K.P89355 X 1/6 KP893597 X 1/6 KP89328X 1/6 KP89256 X 1/6 K.P89176 X 1/6 Fig. 3. Choppers from Huangyan Cave, Guangdong Province. plants. Hence, the oblique-edged pebble tools are evidence of the adaptation of human beings to the new ecology in Mesolithic times. In the early Holocene, humans had an opportunity for foraging since more aquatic species were reproduced as the sea level and water level inland in south China rose along with a global rise in temperature. As a result, ancient people were attracted to living in caves which were not very high in relative altitude from the ground (the relative altitude of the entrances of the caves in Figure 1 ranges from several metres to twenty metres) and close to rivers, lakes, and the seashore. Thus, gradually, fishing and gathering were developed. season. Thus, human practice laid the foundations for the domestication of plants and animals. Recent findings in the Pearl River Valley support this assumption. In the middle stratum of Niulan Cave in Guangdong Province, archaeologists found small holed pebbles used as fishnet weights, shells, mussels, fish bones, and tortoise shell. All these remains suggest people at that time lived in a pleasant environment with rich food resources. The most exciting find is the silicinized remains of rice, of which the 14C is dated as early as 12 000 BP. The rice was analysed by scientists and recognised as neither Indica nor Japonica. Some archaeologists insist that south China is one of the key zones where ancient people in Mesolithic times began to domesticate rice {Ding 1957; Tong 1984.21-30). The rice remains in Niulan Cave add further evidence to suggest that ancient people might have tried to cultivate rice 12 000 years ago. Also, the density of these sites in the Mesolithic period is greater than that in the Paleolithic. In Fengkai County, west Guangdong Province, three caves of this period are found in the area of 2 km-' in a small river valley. In of these, Huangyan Cave, more than 900 pebble tools were excavated in an area of 300 m2 (Fig. 3). Similar finds were also made in Niulan Cave in Yingde County, Guangdong. The large mount of pebbles implies that ancient people in this region lived in groups of considerable size and for a long time. Actually, in the Mesolithic period, south China was covered in tropical and subtropical forests, where rich resources of plants and animals for food selection were available, and this attracted groups of people to stay in place for a longer time. They gradually understood that they could have enough food without seasonal migration. Changes in mobility consequently caused changes in patterns of settlement and social organisation. Women had more energy and time than before to raise children, which reduced the probability of infant mortality. The result of this was an increase in population. Also, they had more time to work together when they settled down. This offered them the opportunity for a division of labour between men and women and the old and young. All in all, we can imagine ancient people had to endure a long, complicated, and tortuous process to acquire the necessary experience of plant and animal husbandry. However, the evidence of human behaviour, thoughts, and religion in south China at that time is not so encouraging. To understand better the socio-economic organisation in the context of the transition to farming we need more archaeological data. DISCUSSION AND CONCLUSION For more than half a century, the concept of the Mesolithic has been a controversial topic in China. Some Chinese archaeologists do not agree that there was a Mesolithic Period in China. They have two reasons for believing the Neolithic evolved directly out of the Palaeolithic (Jia 1991.53-54; Zhang 2000.6). One reason is that there are no representative Mesolithic artefacts found in China. While in Europe, the Mesolithic has been well recognised for a long time as microlithic and arrows are regarded as the representative tools. Actually, a few microlithic tools are found in some sites, such as the lower layer of Bailian Cave (2668-2800 BP) and the fourth layer of Liyuzui (18 388-21 217 BP) in Guangxi Province, although they do not dominate the composition of stone artefacts. However, the pebble tools make up a large percentage of the artefacts and this may be the crucial difference between the Mesolithic in South China (as well in Southeast Asia) and that in Europe, which was due to the different climate and ecology of the two regions. We cannot deny the existence of the Mesolithic in South China just because there are not so many microlihtics and arrows commonly found at all the sites. Another reason is that a few pottery sherds are found in some Mesolithic strata, such as in Liyuzui site and Zengpiyan Cave site, where pottery are dated earlier than 8000 BP. Traditional Chinese archaeologists define any of these findings as Neolithic culture if they are associated with the pottery. But archaeological contexts have shown the appearance of pottery production in South China before the Neolithic. At Miaoyan Cave in Guangxi (Fig. 1, site 12) five pottery fragments were found. The thermo-luminescence dating of these fragments is as early as 15 000 BP {Chen 1999.156-157; Qi 2000.54). Obviously, the Neolithic could not have been identified about 15 000 years ago, if the appearance of pottery is judged as its symbol. The appearance of pottery means nothing more than a revolutionary technology of human beings to make more portable artefacts and their engagement in settlement. Only by studying Mesolithic times can we study how pottery originated. Some archaeologists make the criticism that this close and short-term view to negate the existence of Mesolithic in China comes the traditional Chinese cultural-historical methodology, which usually emphasises the importance of the origin, distribution and relationship of archaeological findings, especially the typology of implements, and pays little attention to the dynamic of human culture and events {Chen 2000.11-22). Analysing the process of the transition from the Palaeolithic to the Neolithic, archaeologists should know clearly that the Neolithic revolution constitutes a profound change from the specialised hunting of herd animals to a broad-spectrum economy. People adopted a mixed resource strategy involving plant collecting, hunting and fishing. Namely, the nature of this transition is the subsistence mode of ancient people, regardless of whether there are lithics, arrows, and pottery. Actually, Chinese archaeologists never stop their studies on the transition from Palaeolithic to Neolithic. Due to new archaeological finds in the last decade and the re-analysis of old materials, the study of Mesolithic culture in south China has made a breakthrough, which exposes a trace of a cultural transition from Palaeolithic to Neolithic. Fortunately, more and more Chinese archaeologists are beginning to criticise to the traditional historio-graphical orientation, and are turning their eyes to this research area with new perspectives. At a Conference on Mesolithic Culture, the first seminar on this topic in China, held in Yingde City, Guangdong Province, in December 1999, archaeologists reported their new findings about the transitional culture from the Palaeolithic to the Neolithic and other related issues. Many of them agreed that more intensive research and co-operation on how Palaeolithic culture shifted to Neolithic culture are necessary. More active excavation should be done, and re-analysis and systematic research on the old findings should not be neglected. Obviously, archaeologists focusing upon Mesolithic culture in China have a long and difficult way to go, but their prospects will inevitably be bright. REFERENCES AN Z. 1956. Guanyu woguo zhongshiqi shidai de jige yizhi (Some Mesolithic Sites in China). Kaogu tong-xun {Journal of Archaeology) 2: 2. CHEN C. 2000. Research and Some Ideas on the Mesolithic Period. Nongye Kaogu {Agricultural Archaeology) 1: 11-22. CHEN S. 1999- Guilin miaoyan dongxue yizhi de fa-jue yu yanjiu (The Excavation and Study on Miaoyan Cave in Guilin). Zhongshiqi wenhua ji youguan wenti yantaohui lunwenji (Treatises of International Conference on Mesolithic Culture): 150-164. DING Y. 1957. Zhongguo zaipei zhongdao de qiyuan jiqi yanbian (Origin of Rice Domestication and its Development in China). Nongye xuebao {Journal of Agriculture) 8: 3- HUANG W. et al. 1990. Guangxi baise shiqi cengwei he shidai de xin renshi (Re-thinking of the layer and dating of stone artifacts fund in Baise). Renlei Xuebao (Journal of Anthropology) 2: 105-112. JIA L. et al. i960. Guangxi dongxue zhong daji shiqi de niandai (On the time of the Chipped Stone Tools in the Caves of Guangxi). Vertebrata Palasiatica 1: 64-68. 1991- Zhongshiqi shidai shifou cunzai (Whether Mesolithic Period exists). Treaties in Commemoration of the 30"' Anniversary of the Discovery of Huangye Cave Site: 52-54. KONG ZHAO-CHEN et al. 1994. Palyeogical Analysis of the Bailiandong Cave Site and its Significance in Palaeovegitation and Palaeoenvironment. In Zhou Guo-xing et al. (eds.), International Symposium Relationship betiveen Chinese and Japanese Aticient Men and Prehistoric Cultures: 147-155■ LIUZHOU MUSEUM 1983. Liuzhoushi dalongtan liyu-zui xinshiqi shidai beiqiu yizhi (On Liyuzui Site, a Neolithic Shell Site in Dalongtan, Liuzhou City). Kaogu (Archaeology) 9: 769-774. PEI W. 1935. On a Mesolithic (?) Industry of the Caves of Kwangsi. Bulletin of Geological Society of China 14: 413-425. PENG R. 1961. Guangdong wengyuanxian qingtang xinshiqi shidai yizhi (Qingtang Neolithic Site in Wengyuan County, Guangdong). Kaogu (Archaeology) 11: 585-588. QI Z. 2000. Guidongbei lijiang liuyu de shiqi shidai dongxue yizhi jiqi fenqi (The Cave Sites in Northeastern Lijiang Valley and Their Perionds). Nongye Kaogu (Agricultural Archaeology) 2000/1: 47-54. QIU L. et al. 1999- Yingde yunling niulandong yizhi (Niulan Cave Site in Yunling, Yingde). Yingde shiqian kaogu baogao (Reports of Prehistoric Sites in Yingde): 1-111. SONG F. et al. 1981. Guangdong Fengkai and Huaiji yanrong dongxue diaocha baogao (Investigation Report on Limestone Caves in Fengkai and Huaiji County, Guangdong Province). Vertebrata Palasiati-ca 19: 292-293■ 1991- Guangdong Fengkai huangyandong yizhi zongshu. Treaties in Commemoration of the 30"' Anniversary of the Discovery of Huangye Cave Site: 1-12.' TONG E. 1984. Luoshu dongnanya ji zhongguo nan-bu nongye qiyuan de ruogan wenti (On Some Problems of the Origin of Agriculture in south-east Asia and China). Nongye Kaogu (Agricultural Archaeology) 2: 21-30. WORKING TEAM FOR CULTRAL RELICS OF GUANGXI ZHUANG AUTONOMOUS REGION 1976. Guangxi Zengpiyan dongxue de shijue (Trial Excavation in Zengpi Cave, Guiling, Guangxi). Kaogu (Archaeology) 3: 20. WU X. 1962. Guangxi dongbei diqu diaocha baogao (Report on Palelanthropological Expidition of the North-eastern Part of Guangxi. Vertebrata Palasiati-ca 6: 408-411. YANG Q. 1991. Liuzhoushi bailiandog yizhi de chubu yanjiu (A Preliminary Study on Bailian Cave Site in Liuzhou). Shiqian Yanjiu (Prehistoric Research) 1990-1991: 154. ZHANG Z., ZHANG F. et al. 1994. Guangdong feng-kaixian luoshayan dongxue yizhi diyiqi fajue jianbao (Preliminary Report on the Excavation of Luoshayan Site in Fengkai, Guangdong). Renleixue xuebao (Journal of Anthropology) 4: 300-308. ZHANG Z. et al. 1999- Yingde yunling niulandong yizhi (On Niulan Cave in Ynling, Yingde). In Yingde City Museum et al. (eds.), Yingde shiqian kaogu (Reports of Preh istoric Sites in Yingde): 1-112. ZHANG S. 2000. Zhongshiqi wenhua guoji xueshu yantaohui bimuci (Conclusion Speech at the International Conference on Mesolithic Culture). Nongye Kaogu (Agricultural Archaeology) 1: 3-7. ZHENG S. 1936. Zhongshiqi shidai zhi yanjiu (The Study on Mesolithic). Xueshi shijie (Academic World) V. 1, No. 1: 54; No.2: 20. UDK: 902.65(512)"633/634" Documenta Praehistorica XXVII The Dating of Chinese Early Pottery and a Discussion of Some Related Problems1 Chaohong Zhao, Xiaohong Wu Department of Archaeology, Peking University, China wuxh@pku.edu.cn ABSTRACT - This paper examines the radiocarbon dating of Chinese early pottery. It is suggesting that the earliest pottery in eastern Asia can be dated to 17000 years BP, in the period of the last Pleistocene glaciations in Eurasia. IZVLECEK - Vclanku raziskujemo radiokarbonske datacije zgodnje kitajske keramike. Menimo, da lahko najzgodnejso keramiko v vzhodni Aziji datiramo v cas 17 000 BP, to je v obdobje zadnje pte-istocenske poledenitve Evrazije. KEY WORDS - China; Palaeolithic; Neolithic;• pottery; AMS radiocarbon dating INTRODUCTION The invention of pottery is a milestone in the history of human civilization. The uses of pottery made a great improvement in the conditions of ancient human life and increased the adaptability of humanity to nature. Therefore, prehistorians and archaeologists always pay great attention to the discoveries of early pottery. Some archaeologists consider the appearance of pottery as a boundary between the Palaeolithic and Neolithic. Sites of early pottery have been unearthed in Japan, Russia, Mongolia, China and elsewhere in Asia since the corded pottery with an age greater than 10 thousand years was excavated at the Fukui Cave site in Japan in the 1960's. The early pottery found in different sites in China gave earlier radiocarbon ages than those in other countries and attracted the particular attention of acade-mia. This article gives a brief introduction to the discoveries and dating of Chinese early pottery and discusses some related problems. DISCOVERIES OF CHINESE EARLY POTTERY Many sites of early pottery have been excavated in China successively since Xianrendong site was unearthed at Wannian County in Guangxi Province in the 1960's (Fig. 1 and Tab. 1). All this pottery proved to be older than 10 thousand years according to different dating measurements. The sites of Zengpi-yan, Liyuzui, Miaoyan, Yuchanyan, Xianrendong, Diaotonghuan, Nanzhuangtou and Yujiagou have very clear stratigraphical and periodical sequences. The early pottery discovered on Yuchanyan, Miaoyan, Xianrendong and Diaotonghuan sites have the same technological characteristics of those found in South China. Nanzhuangtou and Yujiagou pottery represent the typical styles of North Chinese pottery production. The cave site of Yuchanyan, also called Toad Cave, is located in Baishizhai village in Dao Xian, at the north side of Nanling Mountain, in the Southwest of Hunan Province. This cave is 5 meters high above the modern surface and faces southeast, with a capacious hall at the entrance of the cave, which admits much sunlight. There is an open plain in front of the cave. The site was excavated twice, in 1993 and 1995, over an area of some 100 m- and to a depth of 1, 2 to 1, 8 metres. The cultural continuity from Palaeolithic to Neolithic has been determined. The artefact assemblages consist of (a) stone tools -scrapers, cutters, points, etc., (b) bone tools - chip- 1 This project was supported by the National Science Foundation of China, No. 49771074. ped horn shovels, polished bone shovels and awls, (c) perforated shells and notched teeth. Traces of burning have been detected on the artefacts. However, the most important discoveries in Yuchanyan site are early pottery - coarse sandy wares, dark brown in colour and rice husks. The largest inclusion was identified as sandy grain, about 2 cm in diameter. The pottery discovered in 1995 was restored as a wide-mouth cauldron with pointed round bottom. Decoration consists of a corded pattern in a rough texture inside and outside of vessels. Four grains of rice were collected in 1995 being identified as a domestic species with the character of a wild species. Abundant rice phytoliths were also recovered from the soils of the cultural deposit, accompanied by a large quantity of animal and plant fossils, including 28 species of mammal, 27 birds, 5 fishes, 33 kinds of shells, terrapins, insects and so on, and more than 17 species of plant (Yuan 1996). Xianrendong and the Diaotonghuan are two cave sites, 800 metres distant from each other. They are located in Dayuan Xiang Village in Wannian Xian County, in northern Jiangxi Province. They are situated within a small, swampy Dayuan basin. The Xian- rendong cave faces southeast. A small river flows from east to west in front of the cave. Four excavations were carried out, in I960, 1964, 1993 and 1995. The Diaotonghuan cave is situated at the top of a small limestone hill about 60 metres high. A full-scale excavation was carried out in 1995. Both of the sites have abundant cultural deposits, belonging to the period of the Late Palaeolithic to the Early Neolithic. A considerable quantity of archaeological remains were discovered at the two sites in 1995, including 625 stone objects, 318 bone artefacts, 26 perforated shells, 516 pottery fragments, thousands of animal and a number of human bones. The stone tool assemblage consists of scrapers, points, choppers, multi-edge blades and micro-blades made from flint and quartz chips. Bone and antler tools comprised spades, awls, needles, arrowheads and fish darts. The pottery fragments were fired at low temperature, with brown coloration. Large grains of feldspar and quartz mark the fabrics. The basic shape looks like a cylinder jar with round bottom and upright mouth or something like a round-bottomed vessel with swelling belly and slightly wide flared mouth. Both the inside and outside of the vessel are decorated with stripes in a basket pattern. Most of the animal bones from these sites belong to deer, making up 80% of the assemblage, followed by pig and fowl (including chicken) bones. The analysis of pollen and phytoliths give evidence of the existence of wild rice and cultivated rice at both sites (Zhang and Liu 1996). At Diaotonghuan site the prospecting trench at a depth of 5 metres was divided into 16 stratigraphic zones, from zone A to zone P (MacNeish and Libby 1995). Table 1 provides the representative cultural remains for the six upper zones (Zhao 1997 (1998)). Zones Representative Cultural Remains B Pottery with geometric impressions. Predominance of ground stone tools. C Early pottery made by the coiling method. D Early pottery made by the section modelling technique. E Primitive ceramics. F Ground stone tools. But chipped stone tools and bone artefacts are dominant, which continues through Zone C. G Chipped stone tools and bone artefacts. Tab. 1. Cultural remains for the six upper zones of Diaotonghuan site. The Miaoyan cave is located on a small limestone hill, situated in an eastern suburb in Guilin, in Guangxi province. This cave is about 150 metres above sea level and 13 metres above the ground. An archaeological team from Guilin carried out an excavation in 1988. An area of 50 square metres was excavated in the cave. A total of 6 well-defined strati-graphic zones were identified in a 2.4-2.9 meters thick cultural deposit, which covered the remains of the transitional period from the Palaeolithic to the Neolithic. There were abundant stone and bone tool assemblages deposited. The early pottery was documented in the layer 5. There were five undecorated fragments in grey and brown colour deposited. Some of them had soot on the surface and quartz and carbon granules mixed in the clay matrix. Nanzhuangtou site was found in Xushui County, He-bei Province, located at the eastern foot of Mount Taihangshan, on the western edge of Northern China fluvial plain, at an altitude of 214 metres (Baoding Institute of Cultural Relics 1992). It is an Early Neolithic site covering the area of 300 square metres. The site was excavated in 1986 and 1997. Besides ash pits and fire evidence, certain cultural remains, such as a grinding stone and saddle-querns, bone awl, bone arrow, antler-awl, perforated crabstick, and wood pieces were found. There were also identified large quantities of animal and bird bones, snail and clamshells, foliage, seeds etc. Sand and mica have been identified as main inclusions in 20 pottery fragments. The bodies are about 0.8-1 cm thick. They were made at low temperature with a loose texture. Most of them are grey, and a few are red and brown. Most of them are jars with flat bottoms on which we can see the evidence of firing and smoking, and some look like little bowls (Li 1998). Yujiagou Site was found in Hutouliang County, He-bei Province, located on a northern branch of Sang-gan River. From 1995 to 1997 the Department of Archaeology at Beijing University carried out excavations at this site in cooperation with the Institute of Cultural Relics in Hebei Province. We found overlapping cultural layers from the Late Palaeolithic to the Early and Middle Neolithic within the 7 metres of deposit. There were plenty of stone tools, such as scrapes, points, arrows and very many microliths. Pottery and bone tools were also found, accompanied by a large quantity of animal bones. The pottery was found in a grey and yellow-silver sand layer 3- The fragments are sandy ware, of red and brown colour, among which the largest piece seems to be the bottom of a flat object (Institute of Cultural Relics in Hebei Province 1998; Zhou 1999). DATING OF CHINESE EARLY POTTERY The dating of potsherds and related archaeological strata has been completed at the Laboratory of Archaeometry and Conservation at Beijing University and the Laboratory in the Institute of Archaeology at the Academy of Social Sciences. These potsherds were excavated at sites in Yuchanyan, Miaoyan, Xian-rendong, Diaotonghuan, Nanzhuangtou and Yujiagou, and others. The results of series dates are list in Tables 2, 3, 4, 5, and 6. The dating samples were collected from different layers at different sites and different kinds of samples were treated with different pre-treatment methods. For example, we used sherds to date directly the sites at Yuchanyan and Miaoyan. At first, we analysed carbon sources in the pottery and followed different procedures to collect different carbon components. For the pre-treatment procedures see Figure 2. The components of humic acid and residue were dated and the results of these dates agreed with the Crushed Potsherd Washed by Ultrasonication in Distilled Water Water Crushed Sherd (discard) Methyl Benzene and Alcohol Extraction Organic Solution Crushed Sherd (Lipid) Treated with Dilute HCI Soluble Insoluble (discard) Dilute NaOH Extraction Precipitate with HCI Soluble Insoluble (Residue) Precipitate (Humic Acid) Fig. 2. Pre-treatment procedure of potsherd for radiocarbon dating. dates of different kinds of samples from the same layers. See tables 2 and 3. All of the radiocarbon dates younger than 20 thousands years in this article were calibrated by OxCal v 3-5 with intcal 98. The calibrated age of the earliest Chinese pottery is about 17 thousands years BP. So far, potsherds older than 10 thousand years have been found at many sites in different countries in East Asia, such as China, Japan and Russia (Tab. 6). This fact gives us very important information for research on the origin of pottery in the world and the relationship between human activity and the environment in the late Pleistocene. DISCUSSION ON SOME RELATED PROBLEMS One of the important questions for discussion in the field of prehistoric historiography and archaeology is the origin of pottery. Until now, there have been many academic viewpoints. One is that the primitive people invented the basket container by spreading some earth and wattle from all kinds of foliage on it. By accident, the wattles were burned and a basket made of clay is left. That was the invention of pottery. The other standpoint is that the pottery is going with the origin of agriculture. Agriculture has close relations with residential settlement. Only settled residents have the need to use pottery. The appearance of pottery in western Asia happened after the invention of agriculture. From Chinese ancient literature we can also read "Shennong people invented agriculture earlier than pottery" (Song et al. 1983). Other scholars have stated firmly that the invention of pottery was not related to agriculture (Clark 1952), because in some areas pottery appeared before agriculture and some peasants did not use pottery (Child 1956). We believe that pottery is something that is closely related to people's daily life. And the clay used to made pottery can be found here and there. At present, pottery dating earlier than 10 000 years BP can be found not only in China, Japan and Russian, but also in western Asia and Africa. All this evidence proves that the invention of pottery did not happen in a single place. Regarding the primitive resident's use of early pottery, it is used not only by primitive peasants such as those living in western Asia, but also by people taking up collecting and hunting such as the early inhabitants of Japan in "a cord pattern cultural Age". This indicates that the invention of pottery as related to agriculture is not absolutely sure. Judging from the study of the early pottery found in China and Japan, the primitive people who invented the earliest pottery were still engaged in an economy of gathering and hunting. From the pottery excavated at the site of Yuchanyan, Lab. No. Zone Material Method 14C Age (BP) Calibrated Age (BC) Remains 5730 5568 BA95058 3E Charcoal AMS-14C 14080±270 13680±270 15350-14500 Potsherds, Grains of Rice BA95057a 3H Humic Acid from Potsherds AMS-14C 12320±120 11970±120 13150(0.45)12650 12500(0.55)12100 Potsherds BA95057b 3H Potsherds Residue AMS-14C 14810±230 14390±230 16150-15400 Potsherds Tab. 2. Radiocarbon Dating at the site of Yuchanyan at Dao Country in Hunan Province. Lab. No. Zone Material Method 14C Age (BP) Calibrated Age (BC) Remains 5730 5568 BA92030-1 2 Shell AMS-14C 12730±370 12370±370 13800-12300 BA92033-1 3M Shell AMS-14C 12630±450 12270±450 13700-12200 BA92034-1 4M Shell AMS-14C 13710+270 13320±270 14950-14100 BA92036-1 5L Shell AMS-14C 18140±320 17630±320 20150-19100 Three Potsherds from 5M BA92037-1 6L Shell AMS-,4C 20 920±430 20 330±430 Out of Range BA94137a 5 Humic Acid from Potsherd AMS-14C 15560±500 15120±500 17300-15900 Potsherds BA94137b 5 Potsherd Residue AMS-14C 15 660±260 15220+260 17200-16300 Potsherds ZK-2839 Right 2 Shell 14C 12707±155 12350+155 13700-13100 ZK-2840 Left 2 Shell 14Q 13547±168 13170±168 14650-14000 ZK-2841 5 Shell 14Q 17238±237 16750±237 19000-18100 Three potsherds from 5M Tab. 3. Radiocarbon Dating at the Site of Miaoyan at Guilin in Guangxi Province. Xianrendong, Diaotonghuan, Miaoyan, Liyuzhui (Liu-zhou City Museum 1993), Zengpiyan, Yujiagou and some other places, we can conclude that the early pottery was made by mixing some irregular quartz and granule of feldspar or dolomite. There is soot on the surface of some pottery. Whether their bottoms are round or flat, they were used for cooking, called guan or fu. Although we do not have enough evi- dence to prove what kind of things the cooked food were, we still can conclude that one of the main uses for this early pottery was cooking rice (in Northern China people collected millet and proso) from the evidences of rice phytoliths and grains of early cultivated rice found at Yuchanyan, Xianrendong and Diaotonghuan (Zhang, Yuan 1998; Zhao 1997 (1998)). Lab. No. Zone Material Method 14C Age (BP) Calibrated Age (BC) Remains 5730 5568 BK86120 Ti©~© Wood 14Q 9875±160 9600±160 9750-9150 Potsherds, Stone Tools & Bone Tools BK86121 Ti©~© Wood 14Q 9690±95 9420±95 9250(0.56)9110 9000(0.33)8890 8880(0.11)8830 Potsherds, Stone Tools & Bone Tools BK87093 Ti©~© Wood 14Q 9810±100 9530±100 9600(0.04)9560 9390(0.96)9140 Potsherds, Stone Tools & Bone Tools BK89064 1m East of Ti©~© Wood 14Q 9850±90 9570±90 9600(0.07)9560 9390(0.93)9210 Potsherds, Stone Tools & Bone Tools BK87086 On Middle of North Wall in T3© Mire 14Q 9980±100 9700±100 9690(0.04)9660 9630(0.96)9280 Potsherds, Flakes & Charcoal BK87075 Bottom of Ti© Ash Pit Charcoal 14Q 10510+110 10210+110 10900(0.96)10350 10300(0.04)10200 Potsherds, Antlers etc. BK87088 Bottom of Ti® Mire 14Q 10815±140 10510±140 11070(0.75)10840 10800(0.25)10690 Potsherds, Stone Tools & Bone Tools Tab. 4. Radiocarbon Dating at Nanzhuangtou at Xushui County in Hebei Province. Zone Sample Depth Soil Type TL Age No. (m) (KaBp) Top of Layer 2 Y-2-80 0.82 Grey-black Clay 2.13 Bottom of Layer 2 Y-2-16 2.08 Grey-black Clay 6.07 Top of Layer 3a Y-3a-54 2.60 Brown-yellow Fine Silt 6.95 Top of Layer 3b Y-36-23 4.28 Brown-yellow Pulps 11.12 Top of Layer 6 Y-6-42 5.60 Grey-green Calcareous Silt 12.19 ' The potsherds of early pottery were discovered at the top of forth layer. TL age of potsherd is 11,60KaBp. Tab. 5- Ages of TL at the Site of Yujagou at Yangyuan County in Hebei Province fZhengkai 2001). At the above sites, where we found rice phytoliths and paddy remains, a large quantity of snail and clamshells were excavated. According to the quantity statistics, at Yuchanyan the categories of snail and clam totalled 33 (Yuan 1999). Also, we found the same instances at the site of Zengpiyan, Liyuzhui and Huangyandong and some other cave sites. The appearance of a large quantity of snails can prove that cooking rammish and fresh aquatic foods such as snail, clam and fish, is one usage of early pottery. Eating the rammish aquatic food is acceptable for people living near the sea, but it absolutely cannot be accepted by for people living inland. However, it is reasonable to imagine that this pottery was used for cooking. Some scholars have pointed out that, "In Southern China the early food which needed to be cooked is not wild rice, plant seeds or roots, but aquatic food which is rammish and cannot be eaten without cooking such as snail and clam" (Zhou 1994). This opinion is not absolutely correct, but the appearance of early pottery going with large quantity of snails and clams in southern China can certainly provide us new clues for our discussion on the origin of pottery in this region. The appearance of early pottery provided us new clues for our discussion on the relations between humans and living conditions in the Late Pleistocene. However, it was quite a normal opinion among academics that pottery was invented with the origin of agriculture and stockbreeding in Eurasia in the warmer Holocene (Pei and An 1986). So far, the earli- Lab. No. Zone Material Method 14C Age (BP) Calibrated Age (BC) Remains AA-13393 Gasya, Lower layer, Upper part Charcoal 14Q 11190+90 10875±90 11400(0.26)11300 11260(0.74)11050 LE-1781 Gasya, Lower layer, Bottom Charcoal 14Q 13 340±120 12 960+120 14400-13750 Potsherds, 1.2-1.7 cm thick AA-13392 Khummi, Lower layer, Lower part Charcoal 14Q 13650±100 13260±100 14700-14150 AA-13391 Khummi, Lower layer, Middle part Charcoal 14C 10650±110 10345±110 10940(0.77)10670 10 530(0.23)10430 GaK-949 3rd layer in Fukui Cave at Nagasaki in Japan Charcoal 14C 12 760±350 12400±350 13900(0.57)13100 12 900(0.43)12 300 Cordoned Pottery See above Charcoal 14Q 13 070±500 12 700±500 14 500(0.83)13100 12 800(0.17)12400 Cordoned Pottery Odaiyamamo- to I Kanita town in Aomori Perfecture in Japan Charcoal AMS 14C 14180 13780 14570 Tab. 6. Radiocarbon Ages of Early Pottery in Russia and Japan. est pottery in eastern Asia can be dated to 17000 years BP, in the period of the last Pleistocene glacia-tions in Eurasia. It shows that the invention of pottery in eastern Asia was not connected with the warmer climate, at least. It could be the human adaptation to bad climate that accounts for the relations between the climate and the invention of pottery. However, it can be proved probably that even in 17 000 BP the ancient people at the end of the late Pleistocene had an extraordinary ability to acclimatize themselves to the environment and to made great progress in technology. REFERENCES BAODING INSTITUTE OF CULTURAL RELICS et al. 1992. The Brief Report On the Excavation at the Site of Xushui Nanzhuangtou in Hebei Province, Archaeology No. 11: 961-970. CHILDE V. G. 1956. Piecing Together the past, Presentation of Archaeological Data. Routledge and Kegan, London. CLARK J. G. D. 1952. Prehistoric Europe Economic Basis. London INSTITUTE OF CULTURAL RELICS IN HEBEI PROVINCE 1998. The Great Gain of Archaeological Excavation in Nihewan Basin. China cultural relics News, November 15"' (in Chinese). LI JUN 1998. The New Important Archaeological Discovery at the Site of Xushui Nanzhuangtou, China Cultural Relics News, February 11th, 1998, Beijing. LIUZHOU CITY MUSEUM 1983. The Neolithic Archaeological Shell Heap Site of Dalongtan Liyuzhui in Liuzhou City. Archaeology No. 9: 769-774 (in Chinese). MacNEISH S. R. and LIBBY J. G. (eds.) 1995. Origins of rice agriculture: the preliminary report of the Sino-American Jiangxi (PRC) project SAJOR. Sino-American Jiangxi Origin of Rice (Agriculture) Project. Publications in anthropology (El Paso Centennial Museum) 13: 19-22. PEI WENZHONG and AN ZHIMIN 1986. Prehistoric Archaeology. Chinese Encyclopaedia (Archaeology): 476-478 (in Chinese). China Encyclopaedia Press. SONG ZHAOLIN, LI JIAFANG, DU YAOXI 1983. The history of Chinese Primitive Society. Culture Relics Press: 170-172 (in Chinese). XIA ZHENGKAI 2001. Environmental Background of Cultural Transition from Paleolithic to Neolithic at Ni He Wan Basin in North China. Sciences in China, (in press in Chinese). YUAN JIARONG 1996. New Evidence on the origin of Rice at Yuchanyan, China Cutural Relics News: December 14th, 1997, Beijing (in Chinese). 1999- The Site of Yuchanyan. The Great Archaeological Discoveries in China. Culture Relics Press, Beijing (in Chinese). ZHANG CHI and LIU SHIZHONG 1996. The sites of Wannian Xianrendong and Diaotonghuan in Jiangxi Province. Monthly History: 47-49 (in Chinese). ZHANG WENXU, YUAN JIARONG 1998. The Preliminary Study on the Ancient Cultivated Rice at the Site of Yuchanyan in in Dao County in Hunan Province. Zuowu Xuebao 4/24 (in Chinese). ZHAO ZHIJUN 1997(1998). The Middle Yangtze region in China is one place where rice was domesticated: phytolith evidence from the Diaotonghuan Cave, Northern Jiangxi. Antiquity 72: 885-97. ZHOU GUOXING 1994. Discussion on Bailiandong Culture, Proceedings of International Seminar on the Relations of Ancient People and Prehistoric Culture Between China and Japan China Radio International Press: 203-264 (in Chinese). ZHOU KUNSHU 1999- The Inconformity of the latest Holocene in North China and the Early Neolithic Culture, Proceedings of the Seminar on Mesolithic Culture and Some Related Problems. Guangdong People's Press: 142-149. UDK: 903(55-15)"633/634' Documenta Praehistorica XXVII Anthropological Approaches in Neolithic Studies in Iran Jalal-Rafifar, Department of Anthropology, University of Tehran, Iran jrafifar@chamran.ut.ac.ir ABSTRACT - The Mesolithic/Neolithic transition in western Iran are reconsidered in the light of recent anthropological approaches in Neolithic studies in Iran and a brief review of the chronology of the Neolithisation process in Zawi-Chemi, Shanidar, Palegawa, Tepe Gary Dareh, Karim Shahir, Tepe Asiab, Tepe Abdol Hosien, Tepe Alikosh, Tepe Jarmo, Sarab and Tepe Guran are presented. IZVLECEK - Vclanku ponovnopretehtamo prehod iz mezolitika v neolitik v zahodnem Iranu v luci najnovejsega antropoloskega pristopa k raziskavam neolitika. Podajamo tudi kratek kronoloskipregled procesa neolitizacije v najdiscih Zawi-Chemi, Shanidar, Palegawa, Tepe Ganj Dareh, Karim Shahir, Tepe Asiab, Tepe Abdol Hosien, Tepe Alikosh, Tepe Jarmo, Sarab in Tepe Guran. KEY WORDS - Neolithisation processes; Western Iran; Ethno-archaeology INTRODUCTION After a long period, during which the subsistence of human society was founded on hunting and gathering, came a new era in human history marked by the first appearance of rural society. In the full sweep of human history, no development has had a greater effect than the introduction of agriculture. Approaches in historical anthropology claim culture to be the primary source of changes and transformations. Every change, taking place in different social fields such as politics, economic etc., is therefore a direct consequence of cultural transformation. This means culture and cultural transformation play fundamental roles in engendering development. Anthropological studies have also provided many useful illustrations, for example the first revolution in the history of mankind, known as the Neolithic revolution (child 1936). It was a transformation that led people from an economic system of hunting and gathering to a productive economic system of foraging and farming. As a result of this transition, new social organisations appeared; all this took place around 100 00 BC in the Middle East. The outcome was the first appearance of a more developed society, orga- nised as a village community (Braidwood 1975). The advent of a more developed community, which transpired some 4000 years after the appearance of villages and was followed by a series of cultural inventions such as the discovery of metal, the invention of writing and other technological advances, in fact took place in the wake of the development of village cultures. This mutation seems to have appeared for the first time in the Near East, especially in the hilly regions adjacent to the area of Near- and Middle East, that Breasted called the "Fertile Crescent" (Peake, Harold and Flenre 1927). The piedmont hills and lower inter-mountain valleys of the Zagros located in the Western part of Iran are an ideal location for animal and plant domestication. This area has been one of the favourite places of archaeologists and environmental specialists for at least three centuries. The zone suggested lies above hot and almost rainless flood plains, yet below the cold and damp mountains peaks. Ranging in elevation from 300 to 1500 meters, with rainfall between 250 and 500 mm per year, this region was ideal for naturally irrigated agriculture. According to Van Zeist's and Bottema's, as well as Wright's studies of four sites in the cen- tral Zagros at Lalabad, Nilofar, Zaribar and Mirabad, the climate of the region before 9000 BC was colder and drier than it is today (Wright 1977.281-381). The area that nowadays supports an open forest of oak and pistachio trees contained the late Pleistocene Artemisia Steppes (Van Zeist and Bottema 1977). As the climate of the Zagros region became warmer and wetter at the end of Pleistocene, different flora communities developed in the area of the oak and pistachio open woodland, which included the wild progenitors of potentially domestic wheat and barley. According to Wright (1977) the Zagros mountain region of diversified habitat was an attractive area for certain wild animals and for the people who hunted them. The Epipaleolithic and Neolithic civilisation in the Near and Middle East have been linked culturally and spatially; this enables us to observe the evolution of the Neolithic phenomenon in it's domestic regions in the Jordan Valley and Greater Mesopotamia (Cauvin 1978). However, one of the questions we have to face is the chronology and classification of the excavated sites that span through several millennia (10000-6000 BC). The systems of classification in Zagros were based on technological, economic and cultural criteria, as well as on absolute chronology. The traditional classification, which is founded on the evolution of stone tool technology (objectified as Palaeolithic-Mesolithic-Neolithic sequence), cannot answer all the questions concerning the detection and description of different Neolithic phases and stages of civilisation. Therefore, we intend first to define different periods relevant to this study and thereafter try to present some aspects of the principal stages of the Neolithization. THE PRINCIPLE TRAITS OF DIFFERENT SEQUEN CES AND A BRIEF REVIEW OF THE CHRONOLOGY OF THE NEOLITHISATION PROCESS IN IRAN Braidwood, after several seasons of excavating many Neolithic sites in the foothills of the Zagros mountains, formed a sequence founded on socioeconomic evolution to represent the different stages known since the upper Palaeolithic: (a) intensive food gathering, (b) incipient food production and (c) developed farming communities (Braidivood 1975-98). This perspective on evolution has great historical merit. For the first time, the relation between humans and their natural environment was accentuated. Braidwood also thought the key to all questions concerning Neolithisation would fit all the Near- and Middle East. But after I960, some intensive studies in the Levant (Palestine, Syria and Anatolia) provided new data on the transition from a hunting-gathering to a farming society. The excavations at Mureybet, As-wad, Jericho and Cayomi showed some layers dated with 14C that differed from the Zagros sites. It is perhaps necessary to point out that the Epipaleolithic (Zarzian) stone tool industry was based on micro-liths having only a few geometric forms, such as the crescent and the triangle. 9th to 7th millennium BC The most significant period in our region is between the 9th and 7th millennia BC. A uniform complex of economic and socio-cultural changes was completed during this period. A special brief period, called Zarzian, is of great importance, because it links the end of the Mesolithic with the beginning of the Neolithic. Around the 9th millennium BC the first signs of Neolithisation process appeared on Mt. Zagros, in Kor-destan, with the indisputable augmentation of wild sheep, the appearance of "grinding stones" and round stone structures. This phase is represented in Zagros by layers marked as Zawi Chemi B, Shanidar B1 and the upper layer of Palegawa. Information about the latter is scarce, but we have some significant data about Zawi Chemi and Shanidar. Zawi-Chemi - the multilayered open-air site, located on an extensive terrace, was explored in 1958 and I960 (Solecki 1980). The stratigraphy, more than two metres high, consists of a sequence of cultural horizons with depths ranging from 215 to 50 cm. The horizon at a depth of 120 cm provided a radiocarbon date of 10870+300 BP (W-1681). Perkins (1964) claimed that the sheep and goats were raised on the site especially in its later phase (100-50 cm). However, the new information allows us to create alternative explanations of the excessively large numbers of young individuals. The dominant sheep and goats are always accompanied by red deer, although the latter loses its significance towards the top of the sequence. The flint material was accompanied by the remnants of a rich ground stone industry and sheep and goat husbandry, which prompted Rose Solecki to determine the assemblage as Proto-Neolithic. According to Kozlowski, "further evidence" of such an evolution could be the presence of partially polished axes documented in the uppermost level of the stratigraphic sequence (Kozlowski 1996. 101-116; Hole 1994105). However, we failed to find a true axe in the Washington collection, although we examined most of the finds classified by Rose Solecki as such. Although the stone industry from Zawi-Chemi perfectly matches the early Neolithic Iraqi standard represented in the region of Nemrik, M'lefaat and Jarmo, there was a lack of the other elements of the PPN package. Tokens, clay figurines and stone vessels are absent, and we believe the lower part of Zawi Chemi's stratigraphic sequence should have been determined as the final Zarzian, more or les contemporaneous with Natu-fian civilisation in the Levant. Because of the incorrectness in identifying the stratigraphic sequence, it is impossible to date precisely the "proto-agricul-ture" phase and to identify the appearance of polished stone axes as well as the beginning of the processes of animal and plant manipulation in the region. However, we can be almost sure that all these innovations must belong to the upper part of level B at Zawi-Chemi. Shanidar - the cave site is located at the southern part of Baradost Mountain at an elevation of about 822 metres. It overlooks Shanidar valley and is a short distance from the Great Zab River. The cave is still occupied during the winter months by local villagers and their livestock, mainly goats (Redman 1978.62). It is large, having about 1000 square metres of floor space, and prehistoric deposits as much as 13 metres deep. The deposits have been divided into four major archaeological levels (A-D) spanning the past 100000 years. The oldest is Mousterian (level D) and the youngest is early Neolithic (level A). The Upper Palaeolithic layers (Baradostan) were superimposed by a deposit that has been divided into two cultural strata (B1 and B2). Rose Solecki has determined the lower layer as being Epipaleolithic or Zarzian (level B2) and the upper one as Proto-Neolithic (level Bl). There is some similarity between the tool assemblages of levels Bl and B2. The stone tool industry in general is similar to that found at the nearby open air site of Zawi-Chemi. However, the uppermost level of this site (A) contains materials that range from the Neolithic to the present. Information about the bone tools of Shanidar Bl is limited; what we do know is that the celts/ axes have not been found. This could be an argument to support the thesis that this layer is older than Zawi-Chemi B. The available 14C dates also support this hypothesis, placing Shanidar Bl level in the 9th millennium BC (10600+300 BP) (W-667). It is interesting that grinding stones, which could be associated with gathering, were more abundant in Shanidar Bl (Hole 1987). Palegawa - in the rock shelter of Palegawa, located very close to the Iraqi border (see map), a micro-blade industry of rectangles, triangles, trapezoids and lunates has been discovered in the Epipaleolithic layer. A few obsidian tools like in the Shanidar B2 assemblage have been identified, while the nearest obsidian source is located near Lake Van in Turkey, about 250 km away (Reedman 1978.64). The animal bone assemblage consists of gazelle, red deer, wild cattle, goat and probably sheep and domestic dog (Turnbull and Reed 1974). As far as we know, the final Zarzian in Kurdistan must have ended round 8000 BC. The contemporary sites of this period have not been found in Khuzistan, although we found the site Kuh-Banan further to the southeast, which might be coeval to Zarzian period in the western part of Iran (Rafifar 1993)• The site was unfortunately not excavated properly and the radiocarbon dates and data on subsistence strategies and architecture are not available. So, the stone tool assemblage provides the only information we have. The artefacts showed some Natufian characteristics, which could have been defined according to Huc-kried (1961) by the presence of "geometric micro-liths" and "sickles" to suggest harvesting activities. From 8000 BC to 7300 BC This period was the most important part of the Neolithisation process in this region, because it included the transition from a sub-nomadic to sedentary and agricultural society. All the sites of this period are situated in Zagros, five of them in the northern part in Kurdistan (Zawi-Chemi B, Zarzi A, Karim Shahir, Gird Chai and M'lefaat) and at least three in the central part in Kermanshah (Ganj Dareh, Asiab, Abdol Hosien) have been completely excavated. These eight settlements were used as seasonal camps. The economy of residents was based on sheep and goat breeding. There are some indications for initial agriculture, though hunting and gathering continued to be main economic resource. However, according to Van Zeist, grains of morphologically domesticated barley have been found on the base level of Ganj Dareh (Smith 1986.32) and, if the report is confirmed, the cultivation of barley in Zagros would appear earlier than animal domestication, as in the case of (/aydnu in Anatolia (Dollfus 1989.44). The structures discovered at these sites, with the exception of Tepe Ganj Dareh and Karim Shahir, display round buildings that stood until the end of the 8th and the beginning of the 7th millennium. One of the best examples is the construction of the first occupation phase in Tepe Abdol Hosien, which is round and up to one metre thick (Pullar 1979). This round and "primitive" architecture does not appear either in the rest of Iran or in Turkmenistan (Aurenche 1982). The appearance of real masonry is placed in the beginning of the 7th millennium. It is worth mentioning that the chronology of these sites is problematic and should be reconsidered. The end of the 8th and beginning of the 7th millennium BC This period is contemporaneous to PPNB in Levant. The number of known sites belonging to this period is higher than in the preceding period. They appear in all parts of Zagros and also in Khuzistan. Generally this period is characterised by first indications of agriculture, which is manifested for the first time in the "level D" of Ganj Dareh (Smith 1975). This event is associated with the appearance of the first rectangular houses that are made of rectangular pla-no-covex mud brick that show real complex and solid masonry. Tepe Ganj Dareh - was probably occupied for about 500 years without a longer period of desertion at least by one part of the inhabitants during the year. Four occupation levels (A-D) are known. In "level D" appear the first evidence of tools associated with harvesting: a sickle, grinding tools and especially a "receptacle" made of mud which is the principal innovation in plant processing and storage (Smith 1976; LeMiere 1986). Numerous clay figurines are found on the site: 65 anthropo-zoomorphic, 113 anthropomorphic and 812 animal figurines (Eygun 1992.110). According to Schmandt-Besserat (1974. 12) in the most ancient level (E) 6 figurines of sheep and goat were found, together with some geometric clay objects, spheres, discs and cylinders. However, Hole (1987) believes that the evidence of domestication of cereals, animals and the new type of architecture cannot be considered as an argument for permanent settlement. Also, the very high altitude of the settlement (1400 m) is unfavourable for the earliest permanent habitation. The chronology of different sequences of Ganj Dareh has not been determined. The radiocarbon sequence seems unreliable: the first 14C analysis dated the base level (E) to about 8450+150 BC (Gak-807) and the upper levels (A-D) to about 7300-7000 BC (Smith 1976). The analysis of four carbon samples performed in 1971 put the dates close to 6500 BC. Smith refused the dates and believes even the younger layers at Ganj Dareh belong to the 8th millennium BC (O.c.) Karim Shahir - an open-air site excavated in 1951, is located on an eroding escarpment in the Iraqi province of Kirkuk. The artefact assemblage consists of chipped and ground stones and organic remains, while pottery is completely absent. The sample of 896 cores, representing the pattern of simple forms dispersed in all levels and all around the settlement has been examined recently and attributed to a single occupational period (Hoive 1983; Hildebrand 1996.169). Human figurines are rare, only two having been found (Besserat 1974.11). Tepe Asiab - a PPN open air site in the Kermanshah region near Ganj Dareh, it was excavated by Howe in 1959-60 but remains unpublished. Excavations yielded some chipped and ground stones, charcoal and burned human bones as well as few intrusive pottery fragments (Hildebrand 1996). The basal deposit has been dated to 9755+85 BP (Howe 1983). It is worth noting that the artefacts deposited in different layers in southern pit have been treated as a consistent artefacts assemblage for analytical purposes. However, 110 cores from have been examined, analysis is still in progress and final results are yet to come. Human figurines of minute size are rare, asexual and extremely schematised. Amorphous figurines have outstretched arms; others have legs and arms like stumps, and a plaque-like face with a pinched nose and a scatter of circular reed incisions. The meaning of these figurines is unknown (Schmandt-Besserat 1974.12). The number of animal figurines is equally small; at best, perhaps four can be identified. One of them represents the head of an unidentifiable little animal, fashioned without a body, and having a pinched nose and ears. Tepe Abdol Hosien - is located about 55 km southeast of Ganj Dareh in Luristan, it is probably contemporary with the sites mentioned above. If so, it suggests that formally aceramic groups with or without mud brick buildings may already have been cultivating barley and emmer wheat. Tepe Abdol Hosien also yielded a large number of bullet-shaped cores, which is a sign of pressure-blade technology (Pullar 1990). Tepe Alikosli - is located on the Deh Loran plain in southwest Iran and was excavated by Hole in 1961-63- Two trenches and a large pit were divided into six stratigraphic zones (A1-C2) attributed to the "era of early dry farming and caprine domestication" (Hole, Flannery and Neely 1969). The earliest occupation phase (Boz Mordeh) was dated to the end of the 8th and the beginning of the 7th millennium BC (7500 to 6750 BC). It is characterized by simple mud brick rectilinear structures and by significant use of obsidian tools comparing to contemporary sites in Zagros (Asiab, Ganj Dareh and Abdol Hosien). The abundance of bullet cores at Tepe Ali-kosh shows a much greater development in pressure blade technique than sites of Zagros area. Whereas an elevation of 300 metres might be considered as a lower limit for the distribution of potential domesticates in this region and therefore for early dry farming, the elevation at Alikosh, which contains evidence of early plant and animal domestication, is fixed at 145 metres below the limit. The subsistence strategy was based on the combination of wild and domestic resources. A small percentage of the seeds found were of cultivated varieties of two-row hulled barley and emmer wheat, neither of which is indigenous to the region (Redman 1978.167). Sheep were herded in much smaller numbers than goats. Hunting and fishing presented another major component of the subsistence activities of these early villagers (L. c). However, the northern Khuzistan is an excellent area for winter grazing, a fact that may have had a great deal to do with the beginning of food production there. The 7th millennium BC The number of sites at Zagros increases in this period. There are four most important sites located in western Iran. The first is in Kurdistan, two others in central Zagros, and the last on the Deh Loran Plain. Tepe Jarmo - is situated in the Kurdish hills of western Iran at an altitude of about 800 m. It was the first village to be discovered and described (by Braid-wood) and therefore it became a kind of a prototype of the early village society in the Near East. Qulat Jarmo's deposit spanned about nine metres in depth, of which the remains of the Neolithic settlement are preserved to a depth of about 7 metres and cover about one fifth of a hectare. There were as many as twelve layers of architecture identified, representing a community of 150 people over a period of several hundred years. The number of people was obtained when substantial architecture, constructed largely of pressed mud, was estimated. The people of Jarmo grew barley and two different sorts of wheat. They made flint sickles to harvest the cereals, used mortars or querns in which to grind them, ovens in which they might have been parched, and stone bowls out of which they could eat their gruel. They domesticated goats, sheep, dogs and in the latest levels, pigs (Braidwood 1975.127). The buildings, which were rectilinear, consisted of several rooms, many of which had small courtyards. The walls of the houses were made of padded mud, often set on crude stone foundations. According to Braidwood, the Neolithic village probably looked much like a simple Kurdish farming village of today, with its mud-walled houses and low mud-on-brush roofs (Braidiuood 1975.129). Jarmo has been identified as a permanent settlement. The site has been dated to approximately 6750 BC. It is interesting that portable pottery does not appear until the uppermost settlement layers, and it was not distributed over the entire site. Pressure technology in making stone tools gradually replaced the old tradition of blades and microliths, which was still very strong. In the upper part of the settlement deposit, where the pot- tery appeared, the microlithic tools - geometrically shaped microliths - comprise almost 60% of the whole assemblage. The majority (97%) are trapezoids (Hole 1983). It is worth mentioning that round 6500 BC a specific obsidian tool appeared in Jarmo. We named it "fabrictor" and it is not known on any other contemporary site in Zagros. The point is that the nearest obsidian source is more then three hundred miles to the north. (Braidwood 1975-129). One of the most characteristic features of the Jarmo assemblage is the quantity and variety of clay objects. More than 5500 of them were discovered during three seasons of excavation. Part of the assemblage consists of anthropomorphic and zoomorphic clay figurines. Braidwood believed they favoured the figurine of a markedly pregnant woman, expressing some sort of fertility spirit. Clay pieces shaped into definite and recognizable forms occur in the earliest levels and persist throughout the settlement phases. The total number of preserved zoomorphic figurines is about 1100 pieces. The number of human figurines has not been reported. However, according to Vivian Broman Morales none of the clay figures was found in a context that could suggest pottery use or even production (cfr. Braidwood 1983.370). The Alikoshphase (6750 BC) of Tepe Alikosh has been determined as contemporaneous with the lower aceramic levels in Jarmo. The architecture developed into larger buildings that had more solid construction. There is evidence of a substantial community, perhaps with a greater degree of sedentism and bigger dependence on domesticates available. Sheep bones were abundant, though the hunting of large ungulates continued. Wheat and barley became more significant. Clay figurines increased in number and in variety of forms, a goat-like representation being very typical and widespread. Flint artefacts were abundant, especially blade tools and sickle blades. The microlithic tools grow scarce, although the quantity is still twice that of the microliths from the Zagros region (Jarmo aceramic phase). There are some geometric forms in the Alikosh assemblage that could differ significantly from the Deh-Loran and Kurdish industry. On the other hand, the "lamel-les a'dos" and "troncature" that are relatively rare in Deh-Loran (18%) represent nearly 75% of the microliths in Jarmo, and the bladelets with discontinued retouch which frequently appeared in Alikosh phase (80%) are completely absent in Jarmo (Rafifar 1996. 414). The microliths in Deh Loran are characterised by a high quantity of end scrapers, borers and sickle blades. Some of them are obsidian, but we should not forget that the nearest obsidian source to Deh Loran is more than 900 km away. Sarah - the middle and late Neolithic site of Sarab is located 7 km east of Kermanshan at an elevation of 1300m, which is higher than Jarmo (Braidwood, Howe and Reed 1961). Braidwood conducted the first excavations in 1961, but field documentation was unfortunately mislaid. Three small sondages were dug by the Mahidasht project in 1978. These confirmed that there were two periods, a middle Neolithic and a late Neolithic. The ceramic assemblage of the middle Neolithic consists basically of buff ware in a limited range of shapes. Some of the pottery is painted, usually in a variation of the "tadpole" design known from Jarmo and Guran. A small percentage is painted with geometric patterns along the rim and base, with the centre of the vessel left undecorated (Levine and Young 1986). According to J. Braidwood it was thought at first that Sarab might have been inhabited only seasonally, but the evidence for restricted seasonal settlement is not so clear (1975.130). Braidwood's excavations did not reveal substantial mud-walled architecture, but the evidence from animal bones suggested a year-round occupation of at least some of the village's inhabitants. Some of the people may have moved from one site to the other in pursuit of pastures, while the others remained at home to continue activities that could be carried out during the summer. Curiously, Jarmo, a Sarab-like site in Northern Zagros lacks some of the very characteristics that define Jarmo as a settlement. Sarab has flint and obsidian, and also pottery, clay figurines, stone bowls and bracelets, and even goats, sheep and wheat, but its architectural traces are of wretched reed huts at best. More than 2400 clay objects have been found at Sarab (Broman Morales 1990). The human figurines comprise the largest category (650 sp.) of realistically modelled pieces among the classified clay material. According to Vivian Broman Morales, this is partly due to the inclusion of three types of abstract form that she considers representing females (O. c. 16). The key example for this type is in the National Museum in Tehran. The legs or "shells" of the figures are beautifully decorated with parallel-line nail incisions which are followed by a row of tiny circular punctuates that were produced by a hollow stem like that of a straw. Twenty-one other figurines of this type are covered with different kinds of decora- tion. The category of animal figurines follows with three identifiable types: dog (33 sp.), pig (42 sp.) and horned animal (sheep or goat, 255 sp.) and three other types that are not identifiable, probably representing small animals (258 sp.). Some objects have geometric forms, and the presence of stone objects such as beads and labret like studs has resulted in a few imitations of these forms in clay. Tepe Guran - is situated in Lurestan at 950 m altitude and has been excavated by Mortensen {1974). This site yielded a long sequence of early village material: twenty-one levels have been identified, which make up from 6 to 7 meters of occupation debris dated roughly from 6500 to 5500 BC. The earliest settlers lived in wooden huts in which traces were found of what may have been matting on the floor. At this level we have some evidence of semi-permanent occupation, with domestic goat already present. No evidence for the existence of agriculture has been found. The three lowest levels at Guran do not contain ceramic remains. In later levels, mud-walled houses predominate and there is abundant evidence of both farming and gathering as well as of development of pottery production. Painted pottery appeared in higher layers (R, Q and P) of Tepe Guran (Mortensen 1974.22). This standard painted ware is decorated with red strokes that resemble small tadpoles, a style of painting also found on the pottery in the upper levels of Jarmo and in other early villages in Zagros. Flint and obsidian tools were made on flakes and blades, some of which were characterized as microliths (Redman 1978.172). Level "V" is defined as the oldest (8410+200 BP, k-1006); it contains 2% of obsidian. On the other hand, we do not have any obsidian in level "U". In level "T", dated to about 6350 BC, obsidian tools represent 45% (36 pieces of 80 in total) of the industry. In level "S" there were only 6% and in level "R" no obsidian tools have been identified (Renfew et al, 1966.58). Mohammad Jaffar phase During the latest occupational phase in Tepe Alikosh (6000-5600 BC), there were many innovations, including the introduction of pottery and, it is obvious that the number of sites in Zagros area increased. Building techniques improved and the agricultural tools specified depending on task they were meant for. Sickels are not often found in the aceramic layers of sites discussed above, though they became abundant together with numerous blades, bladelets, end scrapers and borers. There are also evidences for polished axes and domesticated sheep, goats as well the agriculture. In Zagros area appeared rectangular houses with several rooms as a frequent type of architecture. From now on the society of western Iran can be defined as rural village society. In the period round 6500 BC we notice the appearance of a special type of tool, so called "fabricator" (the distinctive uCay6nil tool") in Zagros. The tool, made from obsidian, was found at Jarmo in Zagros only. It might suggest that there were commercial and cultural relations between Zagros and eastern Turkey as well western Iraq, where the distinctive tools were produced at Magzaliyeh at the same time (Bader 1979). ETHNO ARCHAEOLOGY IN IRAN - CASE STUDY: LURISTAN As is well known, ethno-archaeology is often used to help archaeology explain the past with analogies derived from observation of the present (Freeman 1968; Watson 1979). In western Iran several ethno-archaeological studies have been made due to the long-term interest in the transition from foraging to food producing systems. However, F. Barth (1952) was one of the first ethnographers to investigate several tribal communities in southern Kurdistan. There was Braidwood's team, working in Iraqi Kurdish village in the years 1960-1961, but W. M. Sumner performed the first ethno-archaeological research in a village in the southern part of Zagros (Marv Dasht region). He believed that there is a correlation between population and settlement area, which could be used to estimate ancient populations by recent analogy (Sumner 1979.164-174). C. Kramer was working on household size and wealth in Shahabad village in western Iran. She states that "The Shahabad data are relevant to certain classes of archaeological materials from the Zagros region for periods in which we have evidence suggesting the existence of variations in socio-economic rank". Another point that she observed and discussed was the correlation between archaeological changes and family structure. She claims residential architecture is closely bound to the needs of nature and number of inhabitants and suggests that archaeologists and social anthropologists alike should further explore the relationships between architecture and the domestic cycle (Kramer 1979.139-163). Hole's ethno-archaeological project in Luristan in western Iran provides a broad overview of his data on contemporary nomadic camps and the pastoral type of material culture, suggesting a venue for future research with such groups (Hole 1979.192-218). Hole's visit was organised with help of an Iranian assistant S. Amanolahi. They travelled widely through Luristan, questioning people about the economic, social and cultural aspects of their life. They joined one camp of nomads making their annual migration from winter to summer pastures. Hole believes that we can learn much about changes in vegetation, dietary habits, social practices, technology and about influences from outside through interviews, and that pastoral nomads required essentially the same equipment as used by villagers today. The majority of this equipment was probably available by the time domestication began more than 10000 years ago. He thinks that the basic difference between tribal villagers and nomads today is in the amount of equipment and in the style of housing. He also suggests that nomadic treks were usually short and took advantage of closely juxtaposed areas of environmental diversity. Hole adds that modern nomads may be entirely independent of agriculturalists, which could mean that specialised stock-raising could have developed independently of agriculture. Our ethnographic study in several villages in Luristan yielded some interesting information about the traditional form of life of some nomadic and sedentary populations. One of the examples is a village that lies at about 1900 meters above sea level, approximately midway between Izeh (in the Eastern part of Khuzistan plain) and province of Ardel in the southwest part of Bakhtiyari region. The village itself covers an area of approximately 20 hectares extending along two sides of a river that flows throughout the year. It is set in a valley with excellent climate, and surrounded by the beautiful Bakhtiyari Mountain. The nearest communication is a third rate road almost 20 km away, which is only used at annual migrations. As often in the Zagros area, people depend on stock raising (cattle, sheep, goat, also donkeys) and agriculture (barley, wheat, rice, vegetables and orchards, limited to several wild fruits such as fig, pomegranate and grapes). Half of the population is supported by approximately 10 hectares of arable land. In 1989 the village was occupied by 135 people living in 33 families, every one of them descended from a clan of the Bakhtiyari tribe. As to architecture, two kinds of dwellings are recognisable. The first one, the winter residence with one or two rooms and a small storing place, is built of stone and mud with combinations of chineh. Wooden beams and twigs, capped with mud and rolled annually, are the most common roofing materials; most roofs are flat. Summer residences are black tents, one for each family, situated approximately 200-500 metres apart. These kinds of tens are also made by the people of Houfel of goat wool, which is weaved by the village women. This type of village can be placed in the category of simple societies. This classification is supported by indices showing primitive societies such as (a) absence of writing and literature (b) low population density, (c) simple productive system, (d) democratic tribal organisation, (e) lack of disorders caused by entropies (0 the existence of primitive culture which does not develop historically (Levi-Strauss 1969). It should be mentioned that this village never came in touch with any form of civilisation, such as religious, health, etc, and the majority of the inhabitants never left the area. Its presence in this area could be traced back some 200 years. The community continues to live only with help of entirely indigenous cultural resources (Rafifar and As-sgary 1989). REFERENCES AURENCHE 0.1982. La tradition architecturale a l'est de la Mesopotamie (Iran Turkmenistan) du 10 eme au 4 eme millenaire. In Symposium franco-sovietique de Dushanbe, Avril 1982. BADER N. O. 1979. Tell Magzalia, an Early Neolithic site in the northern Iraq. Sovetskaja Archeologija 2: 117-132. BARTH F. 1953. Principles of Social Organization in Southern Kurdistan. Universitets Etnografiske Museum Bulletin 7. BRAIDWOOD R.J. 1975. Prehistoric men, 8th edition, Scott, Foresman, Brighton. BRAIDWOOD L. S. 1983. Aditional remarks on Jarmo obsidian. In L. S. Braidwood et al. (eds.), Prehistoric Archeology along the Zagros flanks. Oriental Institute publications no. 105: 285-287. BRAIDWOOD L. S„ BRAIDWOOD R. J., HOWE B, REED C. A., WATSON P. J. (eds.) 1983- Prehistoric Archeology along the Zagros flanks. Oriental Institute publications no. 105. Oriental Institute of the University of Chicago. BRAIDWOOD R. J, HOWE B. and REED C. 1961. The Iranian prehistoric project. Science 133: 2008-10. BROMAN MORALES V. 1983. Jarmo figurines and other clay objects. In L. S. Braidwood et al. (eds.), Prehistoric Archeology along the Zagros flanks. Oriental Institute publications no. 105: 396-423. 1990. Figurines and other from Sarab and Ca-yonii. Oriental Institute Communication 25. University of Chicago Press. Chicago. CAUVIN J. 1978. les premiers villages de Syrie-Pa-lestine du IXe au Vile millenaire avant fesus-Christ. Travaux de La Maison de l'Orient 4, serie ar-cheologique 3. Maison de l'Orient. Lyons. CHILD V. G. 1936. Man makes himself. Watts, London. DOLLFUS G. 1989. Les processus de neolithisation en Iran-Bilan des connaissances. In 0. Aurenche et J. Cauvin (eds.), Neolithisations. BAR Interntional Series 516:37-65. EYGUN G. 1992. Les figurines humaines et animates du site neolithique de Ganj Dareh (Iran). Paleoori-ent 18/1: 109-117. FREEMAN L. G. 1968. A Theorical Framework for Interpreting Archeological Materials. In S. R. Binford and L. Binford (eds.), New Perspectives in Archeology. 262-267. HILDEBRAND E. A. 1996. Changes in Methods and Techniques of blade production During the Epipaleo-lithic and Early neolithic in the Eastern Fertile Crescent. In H. G. K. Gebel and S. K. Kozlovski (eds.), Neolithic Chipped Stone Industries of the Fertile Crescent. Studies in Early Near Eastern Production, Subsistence, and Environment 3:193-207. HOLE F. 1979- Rediscovering the past in the present: Ethnoarcheology in Luristan, Iran. In C. Kramer (ed.), Ethnoarchaeology: Implications of Ethnography for Archaeology: 192-218. 1987. Chronologies in the Iranian Neolithic. In Aurenche, J. Evin and Hours F. (eds.), Chronologies du Proche-Orient:16000-4000 BP. BAR International Series 379:353-379. 1994. Interregional Aspects of the Khuzestan Ace-ramic - Early Pottery Neolithic Sequence. In H. G. K. Gebel and S. K. Kozlovski (eds.), Neolithic Chipped Stone Industries of the Fertile Crescent. Studies in Early Near Eastern Production, Subssi-stence, and Environment 1: 101-116. HOLE F, FLANNERY K. Y. and NEELY J. 1969. Prehistory and human ecology of the Deh Luran Plain: An Early Sequence from Khusistan, Iran. University of Michigan Memoirs of the Museum of Anthropology no.l. Ann Arbor, University of Michigan Press. HOWE B. 1983. Karim Shahir. In L. S. Braidwood et al. (eds.), Prehistoric Archeology along the Zagros flanks. Oriental Institute publications no. 105: 23-135. HUCKRIED R. 1961. Jung-quartar und End-Mesolithi-kum in der Provinz Kerman (Iran). Eiszeitalter und Gegenwart 12: 25. KOZLOVSKI S. K. 1990 (ed.). Nemrik 9. Pre-Pottery Neolithic site in Iraq 1. 1996. From Zawi Chemi to M'Lefaat. In H. G. K. Gebel and S. K. Kozlovski (eds.), Neolithic Chipped Stone Industries of the Fertile Crescent. Studies in Early Near Eastern Production, Subsistence, and Environment 3:175-183- KRAMER C. 1979- An archaeology view of a contemporary Kurdish village: domestic architecture, household size and wealth. In C. Kramer (ed.), Ethnoarcheology: Implication of Ethnography for Archaeology: 139-164. LeMIERE M. 1986. Les premiere ceramiques du Mo-yen Euphrate. These de Doctors, Universite Lumiere-Lyon 2. LEVINE L. D. and YOUNG T. C. 1986. A summary of the ceramic assemblages of the central western Zagros from the middle Neolithic to the late third millennium BC. Colloques Internationaux Prehistoire de la Mesopotamie 17-18-19 Decembre 1984. Edition du CNRS. Paris LEVI-STRAUSS C. 1969. Conversations with Claude Levi-Strauss. In G. Charbonnier (ed.), Cape editions 34. Cape, London. MORTENSEN P. 1974. A Survey of Early Prehistoric Sites in the Holailan Valley in Lorestan. Proceeding of 2nd annual Symposium on Archaeological research in Iran 2:34-52. PEAK H. and FLEURE H. J. 1927. Peasants and potters. Oxford University Press. London. PERKINS D. 1964. The prehistoric fauna from Shanidar, Iraq. Science 144: 1565-1566. PULLARJ. 1979- Tepe Abdol Husein. Survey and excavation in Iran. Iran XVII: 153-155. 1990. Tepe Abdol Husien, a Neolithic site in Western Iran. BAR International Series 563- RAFIFARJ. 1993- Prehistoric Koh-banan (available in Farsi with English summary). Journal of the Faculty of Letter and Humanities of the University ofKer-man 4.2 (NS): 1-16. 1994. Nouvelle reflexion sur l'industrie du Deh-LOran, 9500-7500 BP (available in Farsi with French summary). Madjalle-ye-Bastanshenasi wa Tarikh. Iranian Journal of Archeology and History 7/1-2: 2-18. 1996. Remarques sur L'industrie lithic Neolithic de L'Iran. (10000-6000 BC). In H. G. K. Gebel and S. K. Kozlovski (eds.), Neolithic Chipped Stone Industries of the Fertile Crescent. Studies in Early Near Eastern Production, Subsistence, and Environment 3: 405-421. 1997. The Obsidian Route in ancient Iran. In Man and flint. Proceedings of the Vllth nternational Flint Symposium Warszawa-Ostrowice Swieto-krzyski September 1995: 279-285. RAFIFARJ. and ASSGARI A. 1989. Houfel Village. Na-mehe Oloume Edjtemai e. Journal of social sciences 3/1:35-59. REDMAN C. L. 1978. The Rise of Civilisation: From Early Farmers to Urban Society in the Ancient Near East, W. H. Freeman. San Francisco. RENFREW C„ DIXONN J. E. and CANNJ. R. 1966. Obsidian and early cultural contact in the Near East. Proceeding of the Preh istoric Society 32:30-72. SCHMANDT-BESSERAT D. 1974. The use of clay before pottery in the Zagros. Expedition 6/2: 11-17- SMITH P. E. L. 1975. Ganj Dareh Tepe. Iran 13:178-180. 1976. Reflections on four Seasons of Excavations at Tape Ganj Dareh. Proceeding of the 4th annual Symposium on Archeological Research in Iran 4: 11-22. 1986. Palaeolithic Archeology in Iran. American Institute of Iranian studies, Monographs 1. The University Museum. University of Pennsylvania. Philadelphia SOLECKI R. S. 1981. An Early Village Site at Zawi Chemi-Shanidar. Bibliotheca Mesopotamica 13. Un-dena Publications. Malibu, California. SUMNER W. M. 1979. Estimating Population by Analogy: An Example. In C. Kramer (ed.), Ethnoarcheo-logy: Implications of Ethnography for Archaeology: 164-175. TURNBULL P. F. and REED C. A. 1974. The fauna from the terminal Pleistocene of Palegawra Cave, a Zarzian occupation site in Northeastern Iraq. Anthropology 63: 81-146. VAN ZEIST W. and BOTTEMA S. 1977. Penological Investigation in Western Iran. Palaeohistoria 19: 9-85. WATSON P. J. 1979- The idea of Ethnoarchaeology. Notes and Comments. In C. Kramer (ed.), Ethnoar-cheology: Implication of Ethnography for Archaeology. Columbia University Press, New York. WRIGHT H. E. 1977. Environmental change and the origin of agriculture in the Old and New Worlds. In C. A. Reed (ed.), The Origins of Agriculture: 281-318. UDK: 903.2(55-191,2)"634" Documenta Praehistorica XXVII A New Painted Pottery Assemblage at Ismailabad; a Late Neolithic Site in the Central Plateau of Iran Hasan Talai Institute of Archaeology, University of Tehran, Iran ABSTRACT - Recent discoveries of Middle and Late Neolithic pottery assemblage at Ismailabad have been presented. Ismailabad painted pottery ivas the hallmark of a distinctive culture. Ornamental patterns are representationally sophisticated and conceptional, showing inspiration from an Iranian tradition. IZVLECEK - Predstavljamo zadnja odkritja srednje- in poznoneolitske keramike iz Ismailabada. Is-mailabadska slikana keramika nosijasen pecat kulture, katere vzorci okrasevanja so znacilno pre-finjeni in vsebinsko izdelani, in ki se navdihujejo v iranski tradiciji. KEY WORDS - Iran; Neolithic; painted pottery INTRODUCTION For many reasons the single-culture unit of the north-ern-central part of Iran presents one of the most provocative challenges for archaeologists working on Neolithic studies because (1) there are only a few sites in the area that have been excavated (Fig. 1), (2) most of the sites were excavated years ago, when stratigraphy was considered much less significant (Talai 1983a), (3) archaeological reports have mostly emphasised pottery sequences and therefore the problem of cultural sequence and its relation with other cultural zones in Iran, such as northwest, central west, southwest and northeast, are based on ceramic parallels (Dyson 1991), (4) it is still unclear when and how the Neolithic period in the area began in comparison to the Zagros region - the central north part of the Iranian plateau does not seem to have gone through a pre-pottery and early Neolithic phases. It is worth noting that there are still some problems in identifying the Iranian pre-pottery Neolithic if we do not take into account the very few sites indicating pre-pottery debris. Given all the available evidence, we can divide the Neolithic period in the central plateau into a middle and late Neolithic (6200-4300 BC). Yet the foundation for the determi- nation of the two phases is only based on a change in pottery tradition, the appearance of the black-on-red painted pottery which is most characteristic for the late Neolithic and appears first in Zagheh level 8 (Malek Shahmirzadi 1990) and Sialk 1-5 (Ghirsh-man 1938). Obviously this foundation is very insecure, and the writer is quite ready to discard his in- Hissar, 3■ Sialk, 4. Zagheh. Fig. 2. Zagheh, MN painted pottery. Iran, but we hesitate to adopt the popular explanation of descent from a common source. The MN pottery is almost always thick-walled, made of coarsely levigated gritty clay and badly fired. The fabric of the pottery is buff in colour, of medium coarseness and contains chopped straw. The surface was smoothed before the application of paint, which is usually brown, with the design always monochrome. The style of painting and fabric of the pottery differ from the LN pottery assemblage (Fig. 3); MN pottery is technically inferior, coarse in texture and rough surfaced, with a slip used infrequently. The motifs are predominantly geometric and show that the pain- Fig. 3■ Ismailabad, LN painted pottery. terpretation as soon as contradictory evidence is available. The Middle Neolithic period is represented in the lower levels of Zagheh I in the western part of the region and in Tepe Sialk level I in the eastern part, while the information from Cheshmeh Ali is scanty and inconclusive. The Late Neolithic period is represented in upper levels of Zagheh, in Sialk level II, in lower levels of Cheshmeh Ali and in a more recently excavated site at Ismailabad. However, on the bases of stratigraphy and chronology of Zaheh and Ismailabad, there seems to be a discontinuity in pottery sequence from the Middle to the late Neolithic. THE MIDDLE NEOLITHIC (MN) POTTERY ASSEMBLAGE An interesting feature of the MN is painted buff pottery (Fig. 2), which front a technical point of view is obviously more developed than the Early Neolithic pottery in Iran. The place of origin of MN pottery is a matter of considerable dispute. The original period of this assemblage is difficult to determine. It may have received some stimulus from the north-west of ters had a marked preference for straight lines over curves. Simple patterns, such as hatching, crosshatc-hing and zigzags are used very effectively, those made with multiple parallel lines and vertically painted zones being especially noticeable. Almost invariably the drawing is arranged in an orderly decorative scheme, and a feeling for the shape of the vessel is clearly absent. Another interesting characteristic of MN culture are containers for grain storage, indicating a well-developed storage system. The walls of these vessels are approximately 6 cm thick, made of baked straw-tempered clay, with diameters averaging around 60 cm. These vessels were lowered into underground pits and the spaces around them filled with loose debris. Many bins still held decayed chaff and carbonised grain when excavated. Biconical spindle whorls, usually of baked clay, and bone tools such as needles are common, but flints and obsidian blades are very rare. Tiny, drilled beads and stone pendants appear, sometimes made of agate, turquoise, limestone, hematite and unidentified stone. The presence of some objects within MN assemblages suggests the existence of long-distance trade (Talai 1999). Crude human and animal clay figurines (Negahban 1984) and plain tokens in the shape of cone disks are present in MN deposits, as are clay tokens used for counting and accounting units of goods (Sckmandt-Besserat 1997). MN deposits in the region show a well- developed society of people who built permanent dewellings. Tepe Zagheh is the largest site to yield architectural remains. As a result of several seasons of excavations a considerable part of the open and defenceless MN village has been revealed in the upper levels with the best-preserved houses with open courtyards. The settle- Fig. 4. Ismailabad, zoomorphic motives (after Maleki 1968). ment was probably at all times about the same size as modern villages in the neighbourhood. The total height of the mound from virgin soil is 6.7 meters. In the early stage pise is used a building material. But soon technical proficiency increased and they began to use long mud brick. Structural details indicate that there must have been a long tradition in MN architecture. The point is the appearance of Zag-heh painted temple, a large (117 m2) and complex structure with nine benches and a fireplace built inside. Walls were painted with simple designs. Large numbers of clay figurines were found inside the temple, which further indicates the building was a religious centre or was used for social gatherings (Ne-gahban 1979). Domesticated plants and unidentified animal species are documented. The people of MN buried their dead with grave goods under the floors of the roofed areas of houses. In most cases the skeletal remains showed intensive use of red ochre, even in the mouth, which, if we may judge by burial practices, suggests an idea of an afterlife (Ne- gahban 1979). As a whole, the assemblage depicts well the lifestyle and economy of MN people in this region, who reached a considerable level of technical development and created a flourishing culture. THE LATE NEOLITHIC (LN) POTTERY ASSEMBLAGE AND ARCHITECTURE Due to its distinguished architecture and pottery, Ismailabad remains the most important site of the Late Neolithic period. The stratigraphy of almost 7 meters of deposit and ten architectural levels of LN occupation is incomparable with any other site in the region because it shows a continuous development of black-on-red painted pottery, which is the common denominator of the Late Neolithic period, from light red to dark red. This fact makes it very likely that Ismailabad began its history as a result of migration very early in the sixth millennium BC. During the sixth millennium BC Late Neolithic villages Fig. 5. Ismailabad, geometric motives (after Maleki 1968J. appear widely on the central plateau and beyond, from Qazvin plain to Kashan (Sialk), and eastwards to Turkestan. It would be premature to assert on the present evidence that the wide expansion of the LN culture originated in the central plateau. The connections between LN settlements on the central plateau and eastern Iran seem, however, indisputable. It seems that the culture of the central plateau had a character of its own, influenced by local traditions. The culture persisted into the fifth millennium BC, with a break in continuity recognized in the region. After the final abandonment of Ismailabad, soon after c. 4300 BC, a culture appeared at Sialk III, developing slowly through the long succession of late Chalcolithic (Majidzadeh 1981). As stated earlier, red painted ware is the most characteristic of Late Neolithic pottery. The motifs are predominantly geometric representing stylised animals and plants (Fig. 4). Complex linear designs such as hatching, crosshatching, chevrons and zigzag are common and used effectively, those made of multiple parallel lines being especially noticeable (Fig. 5). The considerable use of various horned and un-horned animals and birds is most noteworthy. Fish, wild horse and various unidentified animals also appear, the animals being represented almost in horizontal rows. The horned animals and birds are orientated in different directions, which might indicate two different conceptions of those animals in the painter's mind (Figs. 6, 7). mm HTfrif fif Fig. 6. Ismailabad, the horned animals being represented in horizontal row. Negative design, which means the background is filled in with paint while the design areas are left in the colour of the pot surface, is frequently used (Fig. 8). It seems that painters were quite skilled at this technique, and they embellished it by adding painted motifs within negative areas or by placing painted motifs so as to produce the effect of a negative design. The fact that this technique appears earlier in the MN painted pottery assemblage is one of the reasons for considering LN pottery as local in character. Horizontal zones of continuous design, undoubtedly the most effective decoration for the round surface of a pot, predominate; elements in these design zones tend to be contiguous. Rarely, mostly on simple bowls and some pots, only one zone of design appears. Even on these vessels the broad zone is frequently accompanied by very narrow subsidiary zones on simple bands. The characteristic LN style of Fig. 7. Ismailabad, birds being represented in horizontal rows. Fig. 8. Ismailabad, negative design and horned animals. Fig. 9. Ismailabad, circidar surface decoration. decoration consists of multiple zones of design, usually rather narrow, extending from the exterior rim to, or just below, the point of the largest diameter. Generally, these are separated by plain narrow bands. The spacing, filling and balancing of these design zones is excellent, as well as the feeling for dark and light areas. The drawing tends to be in thin lines, but heavy ones are not avoided, and a notable balance of the bold and the fine is achieved. In all taller and more closed forms the main design is on the outside. Shallow bowls and plates bear their main design in the interior; it is in the composition of these interior designs that the painter achieved his best works. The decoration of a circular surface presents a difficult aesthetic problem. The success of the painters is demonstrable by a glance at their work: they have created a wonderful effect of life and movement and a harmonic balance by dividing the circular surface into four equal parts (Fig. 9). Drawing is arranged in and orderly decorative scheme. Feeling for the shape of the vessel is present and can clearly be seen in many examples. The lower part of vessel is differentiated from the upper part in many cases, often by a heavy festoon design, while the foot usually only bears plain vertical bands of paint (Fig. 4). Finally, there was a continuous use of designs such as swastikas and the rays of the sun throughout the whole sequence of the LN period (Figs. 10, 11). The painted ware takes many main forms (Figs. 4, 5, 9). The fabric is red in colour, of mostly fine clay, with the inclusion of chopped straw or fine sand. The surface was usually slipped before the application of paint, which is mostly black, while the design is always monochrome. There seems to have been a dual system of pottery production in the LN period. Everyday vessels were probably produced in small quantities in every village. The second mode of production, in which high quality ware was produced apparently by specialists for export to neighbouring settlements, is best known from Ismailabad. Studies have shown that pottery from this workshop was traded over a radius of about 200 km. If this was the case, then it is suggested that the high quality Ismailabad painted ware, which shows a uniformity of design and execution, was probably mass-produced, while other less sophisticated ware was produced locally (Maleki 1968; Van-den Berghe 1968). The characteristic site formation of the MN and LN periods in the region is a by-product of the main building material used. Unbaked mud brick is cheap, convenient and easy to make, but unless a mud brick building is kept in good condition with a proper roof and plastered interior and exterior ******* S3 Fig. 10. Ismailabad, "Sun rye" motive. Fig: 12. Ismailabad, level 5, architectural remains. Fig. 11. Ismailabad, swastika presentation on pottery. added to mud in order to improve the durability and consistency of the plaster. This mixture (straw and clay) is a traditional plaster in Iran, locally known as Kahgle. The lime plaster could be smoothed and polished to give a high quality finish, while the clay plaster was considerably rougher. Both clay and lime plastering is used in the LN period. Roofs were usually flat and were made much as they are today, the rafters of timbers being covered with mats layers of mud. With this limited range of material at their disposal, the LN builders mastered construction techniques (Fig. 12). We know very little about the tools and measuring equipment they used. With minimal equipment they were able to lay out relatively large buildings with accuracy and orient them to the cardinal points as tradition seemed to demand. The mathematics needed may not have been very sophisticated, but an understanding of basic engineering principles and of the properties of mud brick must have been essential. The architectural remains at Ismailabad belong to the LN period (our evidence relates mainly to the domestic buildings which are preserved well enough to allow us to discuss them). The private houses tell us relatively little about the everyday life of the people who lived in them. They have few distinctive features and, generally speaking, are poor in small finds. The plans of the houses suggests they were for self-contained and their importance as the basic unit in society. The positions of the houses at the centre of family life is emphasised by the presence of graves under the floors. Throughout the MN and LN periods, custom seems to have dictated that the dead were walls, it deteriorates very quickly and the bricks crumble back into a low heap of earth. This heap will often form the basis of later buildings and so, over a period of years, a small site is formed. This process, repeated on the ground scale, augmented by domestic debris, leads eventually to the formation of the sites which comprise the ancient settlements in much of the Iranian plateau. Mud brick and pise are the main building materials. Chopped straw and sand are usually added to improve the consistency, and the mixture is manually shaped. The size and shape of bricks changes through time and can sometimes be used as age indicators. For example, the MN period is characterised by the use of long rectangular handmade mud brick. The LN period is characterised by the use of so-called "plano-convex" brick, rectangular in plan, with a rough flat surface, dried in the sun. This odd shape seems to have been achieved by rounding off each brick by hand. Bricks were often laid in a conventional bonding fashion. They are distinctive and are found only in the LN period. The mortar used with the brick is usually of mud. The importance of protecting exterior walls from the weather has already been mentioned, and mud or lime plaster is usually used to achieve this. Chopped straw is buried with grave goods which were apparently intended to make life in the other world as comfortable as possible. Whatever the relationship between the central plateau cultures finally proves to be, it appears fairly clear that at least the cultural B sequence, during the sixth and mid-fifth millennium BC has been clarified, and indeed, in some degree modified by excavations at Ismailabad. It is no longer enough to rely on the well-known discoveries at Tepe Sialk. During the sixth millennium BC the first beginnings of a new and distinctive culture can be detected - that is, the Late Neolithic culture. The radiocarbon dates may be used tentatively to suggest a maximum time-span for the culture from c. 5000 until 4300 BC. Ismailabad painted pottery was the hallmark of a distinctive culture. Another criterion of this culture, on the evidence of Ismailabad, must be the architectural pattern, which was utterly alien to the Middle Neolithic Zagheh culture. Precisely who the first inhabitants of Ismailabad were, or when, whence or how they arrived are questions which may never be answered to the satisfaction of every specialist. The pottery shows an improvement in quality of clay, firng and decoration throughout the LN period. Patterns are representationally sophisticated and conceptio-nal, showing inspiration from an Iranian tradition. The eastward expansion of the LN culture extended as far as Turkestan (Malek Shahmirzadi 1977). REFERENCES DYSON R. H. 1991. The Neolithic period through The Bronze Age in Northeastern and North-Central Persia. In Ehsan Yarshater (ed.), "Ceramic" in Encyclopaedia Iranica, vol. V: 265-275- GHIRSHMAN R. 1938. Eouile de Sialk pres de Ka-shan, 1933-1934 1937. Paris MAJIDZADEH Y. 1981. Sialk III and the pottery sequence of Tepe Ghabristan. The coherence of the cultures of the central plateau of Iran 19:141-146. MALEK SHAHMIRZADI S. 1977. Tepe Zagheh: A sixth Millennium B.C. village in the Qazvin plain of the central Iranian plateau. PhD Dissertation, available at University of Pennsylvania. 1990. Private Houses at Zagheh: a Sixth Millennium B.C. Village in Iran. Bulletin of the Ancient Museum, Tokyo XI: 1-23. MALEKIY. 1968. Art Abstrait et Decor Animalier. Ar- cheologie vivante 1: 43-55■ NEGAHBAN E. 0. 1979- A Brief Report on the Painted Building of Zagheh. Paleorient 5: 239- 250. 1984. Clay Human Figurins of Zagheh. Iranica Antiqua XIX. 1-20. TALAI H. 1983. A Stratigraphical Squence and Architectural Remain at Ismailabad, the Central Plateau of Iran". Archaeolgische Mitteilungen aus Iran 16:57-68. 1999- Funeral Rite at Zagheh: a Neolithic site in the Qazvin plain. Documenta Praehistorica XXVI: 15-20. SCHMANDT-BESSERAT D. 1997. Accounting before writing in the ancient Near East. Documenta Praehistorica XXIV: 151-156. VANDEN BERGHE L. 1968. Richesse de Linspiration naturaliste dans la ceramique peinte Iranian. Arche-ologie vivante 1968: 20-34. Izdala in zalozila: Filozofska fakulteta, Oddelek za arheologijo Tehnicno urejanje in DTP: CAMBIO d.o.o., Ljubljana Tisk: SCHWARZ d.o.o., Ljubljana Naklada: 750 izvodov Leto izida: 2000 Documenta Praehistorica je vkljucena v naslednje izbrane indekse: SACHKATALOG DER PUBLIKATIONEN ZUR ALTERTUMSKUNDE - DAI - Germania ISBN 3-8053-1660-7; DYABOLA - DAI - Sachkatalog Der Publikationen Zur Altertumskunde - http://www.dyabola.de/rom_frm.htm; DYABOLA-CD-ROM - http://www. dyabola.de/BooknCD.htm; DYABOLA IM INTERNET - http://www.dyabola.de/dyaimwww.htm; GBV - http://www.gbv.de; IBZ- http://www.ub.unibas.ch; COPAC - http://copac.ac.uk; J0PAC - http://www.bibliothek.uni-regensburg.de/opac/opacind.htm; 0BN - PICA - http://www.pica.nl; 0PAC - http://www-opac.bib-bvb.de/; SUBITO - http://www-opac.bib-bvb.de/ subbvb/subhome.htm