舊石器時代晚期初段色楞格河人類的擴散路線

摘要：色楞格河（Selenga river） 作為貝加爾湖的關鍵輸入河流，蜿蜒流經(jīng)蒙古國北部以及俄羅斯西南部的外貝加爾區(qū)域。本文針對色楞格河流域舊石器時代晚期早段的遺址展開了深入研究。這些遺址的石器工業(yè)可歸入舊石器時代晚期初段（IUP） 石葉技術體系范疇，且在西伯利亞南部和中亞有著廣泛的分布。IUP 工業(yè)于色楞格河現(xiàn)身的時間大致在 45000 年前，區(qū)域內石器類型的變遷以及遠距離外來原料的輸入，彰顯此地人群擁有發(fā)達的交換網(wǎng)絡，并通過高流動性來獲取所需原料。該區(qū)域存有兩種遺址類型，即蒙古國北部的原料采集地以及外貝加爾地區(qū)的一般性活動場所。盡管動物化石材料有限，但可解讀出的信息暗示了專業(yè)化的狩獵行動，獵物主要為遷徙的草原動物種類。遺址大多分布于中海拔區(qū)域以及色楞格 - 鄂爾渾（Orkhon） 地理區(qū)域的景觀之內，該地區(qū)地貌的同質性也為 IUP 期間人類區(qū)域間的流動提供了有力支撐。

關鍵詞：中亞；舊石器時代晚期初段；遷徙；原料；狩獵

1 Introduction

Initial Upper Paleolithic （hereafter， IUP） technocomplexes provide the earliest evidence ofthe emergence of Upper Paleolithic traditions in southern Siberia and eastern Central Asia. IUPlithic assemblages were primarily focused on producing large and medium-sized blades and wereaccompanied by the simultaneous emergence of human symbolling behavior.

Common traits of these industries include the dominance of subprismatic bidirectionalcore reduction， large numbers of large and medium blades， regular small blade and bladeletproduction from specialized cores， specific techniques of platform preparation， and the presenceof indicator tool-types that display distinctive typologies and morphologies together with apersistent set of baseline tools. IUP industries fall chronologically within the first half of MarineIsotope Stage （MIS） 3 （ca. 57-29 kaBP） and occur stratigraphically directly above Final MiddlePaleolithic assemblages[1-5]. The question of which hominin populations were responsible for theproduction of Initial Upper Paleolithic assemblages in southern Siberia and Central Asia remainsunanswered. The earliest anatomically modern humans in western Siberia （represented by theUst’-Ishim individual from Omsk Oblast） are ca. 45 kaBP[6]. Human fossil remains reliablycorrelated with IUP industries are still unknown from eastern Central Asia.

Between 50 kaBP and 45 kaBP cal， the first evidence of a complete and well-developedtechnological， typological and symbolic suite typical of the IUP and related to the Trans-EurasianIUP technocomplex appeared in the Altai Mountains of southern Russia. The lowermost lithiccomplex associated with occupation level 6-5/UP2 at the Kara Bom site overlies a MiddlePaleolithic Levallois-laminar industry in Layer M2/MP2 dated to final MIS-4 or early MIS-3.This complex constitutes a complete technological， typological and symbolic suite， related tothe Trans-Eurasian IUP technocomplex[7]. IUP technocomplexes in western Eurasia are reliablyassociated with anatomically modern humans， but in eastern Eurasia there are no human fossilremains yet associated with the IUP technocomplex[8-10].The geographical core of IUP complexesin southern Siberia and Central Asia includes several sub-regions where the earliest and mosttypical IUP industries have been found： the Altai Mountains， eastern Kazakhstan and Dzungaria（northern Xinjiang）， northern Mongolia， and the Transbaikal region， or Dauria， east of LakeBaikal. Peripheral areas include central Mongolia， the Gobi Altai and Cis-Baikal regions，the Middle Yenisei Basin， and the Ordos Plateau in China[11]. The outermost points of IUPdistribution are located in Yakutia in the east （Ibid.） and on the Tibetan Plateau in the south[12-13]，where the IUP appears later， and where lithic industries， while exhibiting traits similar to"complexes from the IUP’s geographical core， have fewer typological and technological featurescharacteristic of the IUP.

The IUP varies temporally in different regions of its distribution. In general， the lowerand upper chronological boundaries of this technocomplex fall between 50 and 35 BP cal（radiocarbon dates are hereafter calibrated to IntCal20）. It is possible to reveal a spatialtransgression of radiocarbon dates from the geographical core of the Karabom-type IUP： theearliest chronological range （50–45 kaBP cal） is found in the Altai Mountains at the Kara-Bomsite; in the Transbaikal region and northern Mongolia， the IUP technocomplex first appearedabout 45 kaBP cal （Kamenka， Podzvonkaya， Khotyk， Tolbor-4， -16， -21）; and the probable ageof IUP Layer 7 at the Ushbulak Site in eastern Kazakhstan falls between 46–43 kaBP cal[14-22].

These human groups had to contend with and adapt to various landscapes and geographicalzones with changing ecological conditions and widely varying degrees of availability of water，lithic raw materials， and faunal resources along their presumed migration routes， which suggeststhat these populations had significant adaptive flexibility[19，23]. In order to establish the ways inwhich human ancestors adapted to such changing conditions， one needs to identify the natureof their settlement systems and subsistence strategies. The subsistence systems of Early UpperPaleolithic populations remain poorly known and， with only a few exceptions[5，24-29]， there are nopublished zooarchaeological studies from relevant sites apart from lists of taxa recovered. Thepossible methods by which animals were hunted remain inadequately understood as well – thereare only two direct lines of evidence for Pleistocene hunting in this immense region， and both arefrom Late Upper Paleolithic contexts （ca. 15–13 kaBP）： projectile points embedded in mammothvertebrae at the Lugovskoe site on the central West Siberian Plain[30] and in a bison scapula fromKokorevo I on the Yenisei River in southern Siberia[31]. Zooarchaeological and spatial analysesconducted at these sites provide the bulk of the data for studying human-animal interactions，including hunting and butchering， in this region.

It is easier to understand the mechanisms of rapid distribution of IUP culture in southernSiberia and Central Asia within a single vast geographical region， limited by natural boundaries.The Middle Selenga Basin， situated in the region between the Siberian boreal physiographicprovince， the Arctic Ocean drainage system and interior Inner Asian arid mountain ranges anduplands， can be considered one such region.

This overview identifies specific aspects of human subsistence in the Russian Transbaikalregion and Mongolia， which are the easternmost regions where assemblages of the laminarIUP have been found （Fig.1）. This region was located at the crossroads of migration routesand the nexus of possible mutual influences from populations inhabiting the contact zonebetween Central and East Asia. Available information on the composition of Paleolithic faunalassemblages， spatial organization visible at sites and， in some cases， assemblages of lithicartifacts， will be used. Based on studies of possible selection of prey species， the methods of theirdelivery to prehistoric occupation sites and carcass processing， as well as spatial features of the"sites themselves， we reconstruct subsistence systems and their potential impact on the mobilityof Pleistocene human populations. Stratified Initial Upper Paleolithic sites dating to MIS-3 inMongolia and the southwestern Transbaikal region will be used as corroborative evidence.

Several natural migration routes， including the main course along the Selenga River （that flowsinto Lake Baikal） and its basin connect the southwestern Russian Transbaikal region with northernand central Mongolia， which have similar landscapes and climate. This territory is part of the montanebelt of southern Siberia and Central Asia. The landscape is dominated by mountains with an averageelevation of 800–1500 m above sea level and includes systems of intermontane depressions. Thisregion is located at the junction of two important ecological provinces – the Siberian taiga and CentralAsian steppe – which fosters the highly mosaic nature and diversity of these landscapes.

The densest concentration of IUP sites has been found on the middle reaches of the SelengaRiver. Over one hundred Paleolithic sites， both stratified and unstratified， are known along a 150 kmstretch of the Selenga River and in the valleys of its tributaries in north-central Mongolia[19，32]. Sucha high density of sites can be explained by their strategically advantageous geographical position，which made it possible to traverse the valley of the main river and its tributaries， connecting theTransbaikal region with northern and central Mongolia. While the earliest emergence of the IUP（ca. 50–45 kaBP） has been identified in the Altai Mountains， the IUP technocomplex appeared ca.45 kaBP in the Transbaikal region and northern Mongolia. The distribution of radiocarbon datesreveals the very rapid spread of IUP populations， and the earliest IUP assemblages in Mongolia andthe Transbaikal region are separated by no more than 1000～3000 years from IUP assemblages inthe Altai Mountains[1，2，21].

2 Lithic technology， site planigraphy and subsistence methods among Paleolithic populations of Mongolia and the southwestern Transbaikal region

The principal Final Middle Paleolithic and Initial/Early Upper Paleolithic stratified sites inMongolia are located in the valleys of two small rivers， the Kharganyn Gol and Tolbor， 8-12 kmfrom their confluence with the Selenga River， and along the Selenga’s tributary， the Orkhon River，located 300 km to the south.

Tolbor River localities are situated in the Middle Selenga Basin in similar geomorphologicalcontexts： gentle piedmont slopes with southern exposures on the western banks of rivers，dissected by erosional gullies cutting through Late Pleistocene – Holocene polygenetic sediments.The distance from the modern courses of the Tolbor and Kharganyn Gol rivers ranges between200 m to 1000 m. All known Upper Paleolithic sites in the Tolbor and Kharganyn Gol valleys arelocated near outcrops of metamorphically-altered sedimentary rocks， belonging to the PermianTolbor Formation[33]. This deposit includes silicite， a sedimentary silica-rich cryptocrystalline rockof chalcedony-quartz composition， namely an exposed belt of which extends through the valleysof the Selenga River and its tributaries[4]. Only one case of IUP raw material transport， or， ratherevidence of extended human networks is known： Kharganyn Gol 5 Layer 5 has yielded a perforatedpiece of muscovite， apparently transported/exchanged over a distance of roughly 500 km from thenorthern Selenga Basin in western Transbaikalia[34].

The regional cultural sequence begins with Middle Paleolithic assemblages recoveredfrom Layers 7 and 6 at Kharganyn Gol 5， dated to at least 46～52 kaBP cal and characterized byLevallois-laminar and convergent knapping[4].

A Levallois assemblage from the Orkhon 1 site in central Mongolia is technologically andchronologically very similar to Kharganyn Gol 5. But， in the same valley， flake and non-Levalloisindustries have been excavated from early-middle MIS-3 Layers 5 and 7 at the Orkhon 7 site[23，35].The upper chronological boundary of the North Mongolian Middle Paleolithic is slightly olderor synchronous with dates for IUP complexes in Mongolia and southern Siberia outside of theRussian Altai district[17-21]. The earliest dated IUP archaeological complex in the Tolbor valley，is derived from the Horizon 6 at the Tolbor 16 site， and falls between 45～44 kaBP[21]. Industriesfrom Horizons 5/5a and 6 at Tolbor 4 are related to the same stage[36]. The culturally distinctFinal IUP， dated to 42～40 kaBP， is present in Horizon 5 at Kharganyn Gol 5 and Horizon 4at Tolbor-21. Archaeological Horizon 3 at Tolbor-21 （39～40 kaBP） contains an Early UpperPaleolithic industry[20].

Upper Paleolithic industries initially appeared about 45～42 kaBP in this region， representedby “classical” IUP reduction technology， especially in the Altai Mountains and eastern Kazakhstan.In northern Mongolia， reduction of most IUP cores was undertaken within two related paradigmsof subprismatic parallel knapping （Fig.2）. Two main approaches were utilized to generate flake"removal surfaces： use of the asymmetrical triangular cross-section of blanks and reductionof symmetrical concave flaking surfaces. Bladelet production was a persistent technologicalcomponent employing burin-cores and small subprismatic nuclei （Fig.2）. Bladelets comprise17%～32% of all laminar blanks. Bidirectional blade production played a pivotal role in theearly IUP where bidirectional blades comprise 27%～60% of all laminar blanks. The earliest IUPindustries at the Tolbor 4 and 16 sites contain 35%～62% laminar blanks. This parameter remainshigh （40%～50%） in the late IUP. The principal changes noted at the IUP-EUP boundary lie inthe transition from bidirectional to unidirectional blade technology and a decrease in the size ofblanks， such as in EUP Horizon 4/4b at the Tolbor 4 site， dated to ca. 30 kaBP， at the end of MIS-3.

Tolbor-21 is currently the only Pleistocene site in Mongolia associated with extensivespatial information and zooarchaeological studies. The Initial Upper Paleolithic occupationsurface at Tolbor-21 （Horizon 4， Excavation Unit 2; 42～40 kaBP） was located in the easternpart of the site and provided access to sunlight throughout the entire day and an unobstructedviewshed of the river valley. Repeated cycles of settlement associated with the use of fire andconsumption of large ungulates， dominated by horses， occurred there. Butchery of animalcarcasses for procuring meat， bone marrow， and fat was carried out on-site， and was identifiedfrom the distribution and nature of traces of human impact on bones. The overwhelming majorityof faunal remains， including all anthropogenically modified bones， was found near hearths. Stonetools which could have been used for hunting （diverse points， including bifacial tools） and meatprocessing （flakes with extensively retouched laterals and wide striking platform， convenientfor hafting） were discovered in the same area. Short-term visits were made to the upper partof the site slope （e.g.， Horizon 4， Excavation Unit 4）， evidenced by preforms and cores in theearly stage of reduction and， probably， large blades that were brought in to the site. Specializedactivities associated with the use of large， heavy pebble tools were carried out there where largeand medium-sized ungulates were butchered （Tab.1）. All these activities took place aroundan oval lined stone feature. All four species of large mammals identified at the Tolbor-21 site（woolly rhinoceros， horse， kulan or onager， and yak） are members of a steppe-adapted fauna[20].

A cluster of stratified Paleolithic sites has also been found in the Orkhon River Valley， the maineast tributary of the Selenga River in central Mongolia. All sites are located on the terraces of the right，higher bank of the Orkhon River with a commanding view of the broad steppe on the river’s west bank.The raw material of these lithic complexes was based exclusively on Orkhon channel alluvium locatedin the immediate vicinity of the site. This alluvium provided poor and medium quality raw materialwhich nevertheless made it possible to produce blades and employ Levallois reduction. The mandibleof a Baikal yak （Poephagus baikalensis） exhibiting traces of anthropogenic modification was found inMiddle Paleolithic Horizon 3 at the Orkhon-1 site[13]. The Orkhon-7 site contained the richest faunalcollection among known Pleistocene open-air sites in Mongolia， principally yielding vertebrate speciesassociated with open environments， as has been the case with other Central Asian faunal assemblages.We interpret Orkhon-7 as a butchering site reflecting multiple short-term occupational episodes[23]. A"technologically simple pebble industry is associated with remains of butchered large ungulates， such asPoephagus baikalensis and Equus sp. indet. （likely E. hemionus） and mountain sheep or argali （Ovisammon）[13]. Multiple charcoal lenses uncovered in the lower part of archaeological horizon 4 mayrepresent the remains of a wildfire. One additional isolated and clearly defined charcoal lens was foundin Layer 8， associated with lithic and osteological remains[41].

Southwestern Transbaikalia is the vast mountainous territory east of Lake Baikal， south to theRussian state borders with Mongolia and China that linked northern Mongolia with several potentialprehistoric human migration routes. Foremost， it is the corridor associated with the SelengaRiver Valley， crossing the Buteeliin Nuruu range on the Russian-Mongolian border and flowingthrough the Dzhida and Malkhan ranges. Transbaikalian Paleolithic sites mentioned in the text aresituated on fan-shaped piedmont slopes with extensive viewsheds of valley floors or are locatedin geological semi-cirques， providing protection from wind and access to medium-size ungulatesthat prefer rocky upland topography. All such lithic industries were based on medium and highquality fine-grained and cryptocrystalline raw material. Generally， the acquisition of raw materialsat distances greater than 1 km was characteristic of the Pleistocene populations of Transbaikalia. The"settlers of Kamenka A（C） and Varvarina Gora （2-3） used mostly high quality silicified tuff （rhyolite） forknapping; 70% of the industries[15]： Table 17. Outcrops of this raw material are located in the Mukhor Talatuff outcrops 40 km east of Kamenka and about the same distance from Varvarina Gora. It indicates thatthe behavior of these populations is an exceptional example of predominant use of distant raw materialin the Middle Paleolithic and Early Upper Paleolithic of Central Asia.

The lowermost part of the Khotyk site profile contains Middle Paleolithic flake industries（Levels 6-4/2） that do not share common traits with IUP assemblages from Level 3. A smallMiddle Paleolithic Levallois-blade component is situated between in Level 4/1. IUP industriesat Kamenka A（C） and Khotyk are dated to ca. 45～41 kaBP. These assemblages are typical of theearliest IUP， and are direct analogs of the earliest known stone industries of Mongolia[42，15，43]. Theperiod between 44 to 41 is the most reliably identified range for all sites exhibiting laminarIUP assemblages in Transbaikalia. The period of large blade industries ended about 34 kaBPin Transbaikalia[39，37，18，44，11]. The earliest IUP industries in Transbaikalia are characterized bymarginal presence of Levallois centripetal and convergent methods and by much more commonbidirectional subprismatic technology （Fig.3）. After they appeared in Transbaikalia during theearliest stage in complete form， this technological suite of knapping methods persisted duringthe entire IUP. Blade industries there exhibit high laminar indices： from 23 （eastern complex ofthe Podzvonkaya site） to 46 （Kamenka A） and 58 （Tolbaga）. The largest number of bidirectionalblades was recorded at the Kamenka （43.3%）; much less in other complexes， 20%-30%[18，42，11].

Two principal mammalian species， horse （Equus caballus） and Mongolian gazelle or zeer（Proсapra gutturosa）， were hunted by Paleolithic humans inhabiting the Kamenka site （ComplexA（C）） in the Transbaikal[24]. Only 8.1% of recovered bones were intact; 17.1% of faunal remainsexhibited probable traces of human impact; 1% of those bones had cut marks （although accordingto the reference [28]， the figure is much higher： 9.5%）. Judging by the location of the cuts， theywere the result of carcass butchery and skinning. Traces of thermal alteration （burning， charring）were identified on 1.3% of bones. The highest concentration of bones was found in proximity totwo hearths which probably functioned at different times. This distribution of finds may indicateshort duration human activities at the site. Portions of gazelle skeletons appear in full， indicatingthat the animals were killed near the site and were transported back for butchering as whole bodies.Carcasses of Equus caballus were delivered to the site in an already reduced form. Based on data onthe condition of the epiphyses in the postcranial skeletons of Mongolian gazelles， it was possible toestablish that the site functioned during the late summer/autumn or early winter， when the animalswere in good physical condition[26， 47]. Planigraphic analysis of the Kamenka A（C） site has revealedthin layers （3–5 cm; in pits， up to 30 cm deep）， suggesting closed production and residential cycles，workplaces， semicircular stone features， hearths， and utility storage pits （？）[15]. Thus， Kamenka may bedescribed as having a sophisticated planigraphic structure associated with repeated short-term visits.

The Varvarina Gora site is located in the same inter-montane depression as Kamenka. Ithas yielded a similar ratio of prey species with the only difference that Procapra dominated"the faunal assemblage， followed by horse. Other taxa typically inhabiting open spaces （e.g.，Coelodonta） and alpine species dominated by argali sheep （Ovis ammon） have also beendiscovered at Varvarina Gora. Cut marks appear on 8.8% of bones recovered. Stone features，hearths， and supposedly utility storage pits have been identified at the site[39， 15， 28].

Mammalian species associated with open environments， including the Mongolian gazelle（Proсapra gutturosa; 18 individual animals）， horse （Equus sp. indet.; 5 individual animals）， rhinoceros（Coelodonta antiquitatis; 5 individual animals）， and argali （Ovis ammon; 2 individual animals）were predominant in the Upper Paleolithic complex of Horizon 2 at the Khotyk site， just as at theKamenka site. Analysis of anatomical locations of bone remains indicates utilization of the carcass asa whole. Horses， rhinoceroses， Mongolian gazelles， and mountain sheep are represented by animalsof different biological ages. Identification of the approximate age of young Procapra gutturosa makesit possible to speak of year-round gazelle hunting in the vicinity of Khotyk. Horizon 3 at Khotykalso revealed an orientation toward hunting gazelles and horses. The bones were heavily fragmented.Animals must have been procured near the site， since whole carcasses were utilized in the occupationarea. The seasonal nature of gazelle hunting is clearly manifest， since 67% of young animals wereprocured during the summer， while the rest were obtained at the end of winter– early spring. Level3 of the Khotyk site revealed clusters of artifacts with clear boundaries confined to features madefrom vertically set stone slabs. Bones of the principal prey species were numerous in such clusters[47].

Eight hearths concentrated in an area of ca. 40 m2 were discovered in one excavation unit inthe Eastern complex at the Podzvonkaya site （Horizon 3）， on the Russian-Mongolian border[18]. Itis assumed these hearths were used during several short-term， periodic seasonal visits to the site.Part of a large stone structure was been found in Podzvonkaya Horizon 3; faunal remains of theusual suite of Transbaikal region ungulates was discovered within the structure.

The Southeastern complex at the Podzvonkaya site is similar in chronological and spatialterms to the Eastern complex， and contained evidence of hunting prey species， including horses，argali， bison， and Mongolian gazelles （in descending order of MNI frequency）， that inhabiteddry steppes and rocky landscapes. In the deepest， fourth level where these finds occurred， boneswere found in areas with hearths， ashy spots and burned sediments， and in a separate area ofbone accumulation， where they were mixed with charcoal-laden sediments. Based on recoveredanatomical elements， primary butchery of carcasses was carried out on-site. Animals weretransported from hunting places located within a single day’s travel from the site[29].

Among the localities discussed thus far， the Tolbaga site is located farthest east， and issituated on a vast sloping terrain. Small flat areas populated at different times were preferred foroccupation[27， 44]. Stone structures interpreted as dwellings[46]， areas of lithic reduction activity，and hearths associated with numerous remains of humanly modified bones have been found inthese areas. The composition of the Tolbaga faunal assemblage， which includes prey speciessuch as rhinoceros， argali sheep， and horse， indicates that animals were utilized on-site， where afull spectrum of processing， from butchering carcasses to consuming meat， was carried out. The"percentage of bones exhibiting traces of thermal alteration was very high （39.2% of all artificiallymodified bones）. Split bones accounted for 50.2%， while bones bearing cut marks were scarce，amounting only to 1.1% of the assemblage. Hunting mostly took place near the site， an interpretationsubstantiated by the relatively large number of bones recovered from the distal portions oflimbs and fragments of skulls[40]. Seasonal use of the Tolbaga site during the warm， springthrough-autumn period is likely， as evidenced by the presence of newborn rhinoceroses （springsummer）and the abundance of argali， the hunting of which was most productive in the autumn.

3 Discussion

The Initial Upper Paleolithic sites in the Middle Selenga Basin are situated within the Orkhon-Selenga mid-altitude physiographic province. The modern environment of the Middle SelengaBasin comprises a forest-steppe province with mid-altitude ridges and rugged landscapes withintermountain basins and river valleys. North-facing mountain slopes are covered by Larix taiga，while steppe prevails on southern slopes and in valley bottoms. This ecological zone borders theUpper Selenga forest-steppe province with its mostly steppe landscape. The zones of interface ofthe steppe and forest-steppe ecologies continually wandered due to climate changes in MIS-3. IUPsites in southwestern Transbaikalia are situated across two ecological provinces exhibiting thesame types of topographies – taiga and forest-steppe in the Lower Selenga River Basin and thesteppe zone of the valleys and basins， becoming the landscapes of the Middle Selenga Basin[48].

The Initial Upper Paleolithic sites of the Transbaikal region and Mongolia are located in abelt of low and medium-sized mountains at elevations ranging from 500 to 1900 m above sea level.Sites are typically located along the tributaries of large rivers or in inter-montane depressions. Mostfrequently， sites are located on promontory aggradation slopes with southern exposures and limitederosional ravines of temporary streams， often adjacent to rocky ridges providing protection fromprevailing winds with an unobstructed viewshed over surrounding plain-steppe terrain. Thus， suchsites were located at the junction of various ecotones; landscapes typically had vertical zonation.Steppe landscapes with thicknesses of snow cover not exceeding 10 cm and fostered ungulategrazing， have been reconstructed for the later Pleistocene Selenga Basin during MIS-3 period[49， 50， 38].

The taphocoenosis of faunal assemblages from the Transbaikal region and Mongolia wasinfluenced by anthropogenic factors; the role of predators in the accumulation of animal bones inthese sites was insignificant[28， 20]. None of these sites suggest any orientation toward specializedhunting. Prey species formed a triad which included argali sheep， equids， and yaks in Mongolia，and argali sheep， equids， and Mongolian gazelle in the Transbaikal region. Rhinoceroses andbison were somewhat less common[23]. Thus， the faunal composition of these sites demonstratesan orientation toward hunting medium-sized ungulates – inhabitants of open and dry steppe areas，and alpine species （Tab.1）. The location of sites at the interface of significantly differing landscapes"could have been determined by the ecology of the prey species. At a number of sites， one canassume the existence of seasonal hunting， generally during the spring-through-autumn period， andthe associated migrations of animals， for example， vertical， altitudinal， or spatial migrations ofargali; under certain ecological conditions， both bison and Mongolian gazelle also migrated[29， 51].Gazelle are best described as “permanent migrants” with the largest transhumant areas traversed inthe winter when they cover hundreds of kilometers in Mongolia， and smaller migrations during thesummer and during the calving period[52]. Largest herd sizes occur in winter and during the calvingperiod[53]. Most likely， gazelle could be most effectively hunted precisely during those periodswhen animals gathered in large herds. Based on the composition of skeletal elements recoveredarchaeologically， it is likely that hunting， which was focused on delivering entire carcasses to homebases， would not have taken place far from occupation sites[27， 28， 29]. Remains of small mammalsin these sites are few in number and represented mainly by marmots （Marmota spp.）. They couldhave also constituted prey species， but in the absence of traces of butchery on their bones and thelarge number of krotovina disturbing the sites’ deposits， marmot remains may best be considereda natural rather than anthropogenic inclusion in the faunal assemblage. We conclude that InitialUpper Paleolithic and Early Upper Paleolithic hunter-gatherers pursued a seasonal high-calorie dietwhich entailed hunting large mammals. Such animals were sufficient in smaller numbers and weremore nutritious， but required more mobility from their human pursuers[54].

Where it was possible to unequivocally establish， the intensity of meat processing andhuman modification of bone was low. Intrasite areas of settlement activity exhibiting highdensities of artifacts， the presence of faunal remains bearing traces of anthropogenic impact， and，much less often， bones displaying thermal alteration have been observed at many IUP sites in theRussian Transbaikal region. Traces of sub-rectangular and oval stone structures， sophisticatedhearths， and simple fireplaces have been found in the same depositional contexts. Several suchareas can be distinguished at one site; fireplaces are often separated stratigraphically by thinsedimentary layers， which obviously reflects a sequence of short-term visits. Peripheral zoneswith lower artifact and bone occurrences and yielding no evidence of human-built structures havebeen observed at these sites， the majority of which also indicate the in-transport of lithic rawmaterials from workshops to settlement locations. Population activities focused on providing theresources necessary for functioning of the site and provisioning of place[55].

Based upon human activity reflected in faunal remains， it is difficult to determine functionsof Mongolian sites following the functional typology developed， for example， Paleoindian sites inthe Americas： base camp， kill site， kill and butchering site， storage[56， 57]. Initial Upper Paleolithichunter-gatherers definitely achieved “high specialization” in terms of their focus on large game，but their settlement systems and mobility deviated significantly from patterns developed forPaleoindian subsistence strategies. In the Tolbor Valley， the presence of equid， ovi-caprid， andbovid teeth and mandibles together with post-cranial bones allows us to interpret those sites asbase camps with butchering activity with butchering sites occurring in the Orkhon Valley.

However， it is possible to identify the peopling of the Selenga Basin by IUP huntergatherersnot only from their habitation areas and use of cultural landscapes， but also from theirdependence on outcrops of suitable lithic raw material and certain species of ungulates. Thus，seasonal hunting aimed at yielding a high calorie diet can be assumed. Among other things， ahigh calorie diet is more efficient in terms of a cost-benefit ratio and time expenditure. The mostcommon type of IUP sites can be described as a short-term occupations reflecting activitiesaimed at primary butchery and processing animal carcasses， as well as reduction of lithic corestransported in to the site， and the refurbishment of a mobile set of stone tools. Many sites provideevidence of multiple cycles of short-term visits， carried out over at least several millennia.

4 Conclusions

The laminar IUP appears in complete form in the middle and lower reaches of the SelengaRiver in northern Mongolia and Transbaikalia about 45000 BP cal and its form persists overthe span of its existence. Connections with the poorly understood local Middle Paleolithic areephemeral or lacking. There are clear technological and typological associations between twomain distribution areas of IUP assemblages situated in the lower and middle Selenga RiverBasin. These lithic assemblages appeared and existed contemporaneously and may be considereda single cultural expression.

Settlement systems adapted to the landscape， climate， and fauna of the montane belt andfoothills where Pleistocene human groups settled， can be reconstructed for the Final MiddlePaleolithic through Early Upper Paleolithic of southern Siberia and Central Asia. Evidence ofrepeated visits to the same locations over protracted periods of time may indicate the diachronicpredictability of hunting in these places， which led to the acquisition and in-transport of lithicraw materials and blanks intended for further use to those sites. In the IUP， the principal type ofhuman movement was residential mobility associated with changing activity loci during routinemovements in a bounded catchment area[58].

Current evidence indicates that Initial Upper Paleolithic populations occupied clearly delimitedterritorial ranges. The laminar technology they used restricted them to searching for high-qualityraw material outcrops： most archaeological assemblages are associated with such outcrops within5 km radii of those occupation sites[59]. However， data from several sites in Transbaikalia indicatesignificantly greater geographical ranges of human activity， including raw material acquisition fromdistances of 40 km. The same situation has been revealed in northern Mongolia where one case oftransportation of raw material from Transbaikalia over a distance of 500 km has been documented，providing important evidence of connections between the two main sub-regions of the Selenga-Orkhon mid-altitude zone[60]. Identified regional variability and access to exotic raw materialsfrom significant distances indicate the potential existence of exchange networks and possible long"distance population movements as part of established patterns of lithic raw material acquisition.

In the Russian Transbaikal region and in northern and central Mongolia， nodal clustersof sites occupied either a section of river valley， several closely adjacent valleys， a closedinter-montane depression， or system of depressions. Within those topographic boundaries，movements of social groups probably followed the known habitation areas of ungulate prey.According to the available， admittedly limited zooarchaeological evidence， animals were huntedin proximity to occupation sites which usually had outcrops of suitable stone raw materialslocated nearby. Human settlement systems were associated with routine movement along intermontanedepressions and adjacent river valleys supported by cycles of hunting migratoryherd animals. The location of analyzed sites in the Selenga-Orkhon mid-altitude zone and thegeomorphologically homogeneous landscapes of this region spanning taiga and steppe ecologiesis one reason for interregional migration processes in the Initial Upper Paleolithic， whoseancestral region is the Altai Mountains of southern Siberia.

Acknowledgments： We would like to thank the editors of AAS for the invitation to participate in thisspecial issue， Daria Marchenko for assistance in preparing illustrations， to the anonymous reviewer andJohn Olsen for editorial suggestions and invaluable help， as usual.

References

[1] Zwyns N. Laminar Technology and the Onset of the Upper Paleolithic in the Altai， Siberia. Studies in Human Evolution[M]. Leiden：Leiden University Press， 2012

[2] Rybin EP. Tools， beads， and migrations： specific cultural traits in the Initial Upper Paleolithic of southern Siberia and central Asia[J].Quat Int， 2014， 347： 39–52

[3] Kuhn SL， Zwyns N. Rethinking the Initial Upper Paleolithic[J]. Quat Int， 2014， 347， 29–38

[4] Khatsenovich AM， Rybin EP， Gunchinsuren B， et al. New evidence for Paleolithic human behavior in Mongolia： The Kharganyn Gol 5 site[J]. Quat Int， 2017， 442： 78-94

[5] Rybin EP， Khatsenovich AM. Middle and Upper Paleolithic Levallois technology in eastern Central Asia[J]. Quat Int， 2020， 535： 117–138

[6] Fu Q， Li H， Moorjani P， et al. Genome sequence of a 45，000–year–old modern human from western Siberia[J]. Nature， 2014， 514： 445–450

[7] Derevianko AP， Rybin EP. The earliest evidence of symbolic hunan behaviour in Mountain Altai[J]. Archeol Ethnol Anthropol Eurasia， 2003， 3： 27-50

[8] Hublin JJ， Sirakov N， Aldeias V， et al. Initial Upper Palaeolithic Homo sapiens from Bacho Kiro Cave， Bulgaria[J]. Nature， 2020， 581： 299–302

[9] Douka K， Bergman CA， Hedges REM， et al. Chronology of Ksar Akil （Lebanon） and implications for the colonization of Europe by anatomically modern humans[J]. PLoS One， 2013， 8： e72931

[10] Hajdinjak M， Mafessoni F， Skov L， et al. Initial Upper Palaeolithic humans in Europe had recent Neanderthal ancestry[J]. Nature，2021， 592（7853）： 253-257

[11] Rybin EP. Regionalnaya variabelnost kamennykh industriy nachala verkhnego paleolita v Yuzhnoi Sibiri i vostochnoi chasti Tsentralnoi Azii（Regional Variability of Initial Upper Paleolithic Lithic Industries of Southern Siberia and Eastern Central Asia）[M]. Novosibirsk： IAET SB RAS（Doctor habil Thesis）， 2020： 1-549

[12] Zhang XL， Ha BB， Wang SJ， et al. The earliest human occupation of the high-altitude Tibetan Plateau 40 thousand to 30 thousand years ago[J]. Science， 2018， 362（6418）： 1049-1051

[13] Khatsenovich AM， Rybin EP， Bazargur D， et al. Middle Palaeolithic human dispersal in Central Asia： new archaeological investigations in the Orkhon Valley， Mongolia[J]. Antiquity， 2019， 93（370）： e20

[14] Goebel T， Derevianko AP， Petrin VT. Dating the Middle-to-Upper Paleolithic transition at Kara Bom[J]. Curr Anthropol， 1993， 34： 452-458

[15] Lbova LV. Paleolithic of the Northern Zone of theW estern Transbaikal[M]. Publication of the Buryat Science Cente，r Ulan-Ude， 2000： 1-240 （In Russ）

[16] Kuzmin YV， Orlova LA， Zenin VN， et al. Radiocarbon Dating of the Palaeolithic of Siberia and the Russian Far East： Materials for 14C Data Catalogue （as of the End of 2010）[J]. Stratum Plus， 2011， 1： 171-200 （In Russ）

[17] Tashak VI. Khronologiya rannego etapa verkhnego paleolita zapadnogo Zabaikalia （po materialam Podzvonkoi） （Chronology of Early Upper Paleolithic of western Transbaikalia （on the basis of Podzvonkaya））[J]. Rossiyskiy arkheologicheskiy ezhegodnik，2011， 1： 100-110 （In Russ）

[18] Tashak VI. East Complex of Paleolithic Settlement Podzvonkaya in the Western Transbaikal Region[M]. Irkutsk： V. B. Sochava IG SO RAN Publ， 2016 （In Russ）

[19] Rybin EP， Khatsenovich AM， Gunchinsuren B， et al. The significance of the specific stone raw material in occupation of Middle Selenga by the Paleolithic population[J]. PAEAS， 2016， 22： 159-163 （In Russ）

[20] Rybin EP， Paine CN， Khatsenovich A M， et al. A new Upper Paleolithic occupation at the site of Tolbor-21 （Mongolia）： site formation， human behavior and implications for the regional sequence[J]. Quat Int， 2020， 559： 133-149

[21] Zwyns N， Paine CH， Bolorbat Т， et al. The Northern Route for human dispersal in Central and Northeast Asia： New evidence from the site of Tolbor-16， Mongolia[J]. Sci Rep， 2019， 9： e11759

[22] Anoikin AA， Pavlenok GD， Kharevich VM， et al. Ushbulak： A new stratified Upper Paleolithic site in northeastern Kazakhstan[J].Archeol Ethnol Anthropol Eurasia， 2019， 4（47）： 16-29

[23] Khatsenovich AM， Vishnevskaya IA， Bazargur D， et al. Late Pleistocene paleoenvironments and episodic human occupations in the Orkhon Valley of central Mongolia[J]. Archaeol Ethnol Anthropol Eurasia， 2021， 2（49）： 3-22

[24] Germonpré M， Lbova LV. Mammalian remains from the Upper Paleolithic site of Kamenka， Buryatia （Siberia）[J]. J Archaeol Sci， 1996， 23： 35–57

[25] Wrinn PJ. Middle Paleolithic settlement and land use in the Altai mountains， Siberia[A]. In： Conard NJ， Delagnes A （Eds.）.Settlement Dynamics of the Middle Paleolithic and the Middle Stone Age， Volume III[M]. Tübingen： Kerns Verlag， 2010： 163-194

[26] Klementiev AM. About possible hunting meaning of dzeren in the Paleolithic of western Transbaikal[A]. In： Current Problems of Eurasian Paleolithic Study[C]. Novosibirsk： IAET SB RAS Publ， 2001： 171-176 （In Russ）

[27] Vasiliev SG， Rybin EP. Tolbaga： Upper Paleolithic settlement patterns in the Transbaikal region[J]. Archaeol Ethnol Anthropol Eurasia， 2009， 37： 13–34

[28] Turner II CG， Ovodov ND， Pavlova OV. Animal Teeth and Human Tools： A Taphonomic Odyssey in Ice Age Siberia[M].Cambridge： Cambridge University Press， 2013： 1-490

[29] Antonova YE， Tashak VI， Kobylkin DV. Palaeoenvironment and hunting activity of the Upper Palaeolithic population in Western Transbaikalia： A case study on the Podzvonkaya Settlement， South Siberia[J]. Int J of Osteoarchaeol， 2020， 30（2）： 131-144

[30] Zenin VN， Leshchinskiy SV， Zolotarev KV， et al. Lugovskoe： geoarchaeology and culture of Paleolithic site[J]. Archaeol Ethnol Anthropol Eurasia， 2006， 1（25）： 41-53

[31] Abramova ZA， Grechkina TY. About hunting and hunting weaponry in the Late Paleolithic of Eastern Siberia[J]. KSIA， 1985， 181：44-49 （In Russ）

[32] Seitsonen O， Houle JL， Broderick LG， et al. Lithics in the steppe landscape： An off-site spatial analysis of stone artefacts in the Khanuy Valley， Mongolia[J]. Archaeol R Asia， 2018， 16： 1-13

[33] Rybin EP， Shelepaev RA， Khatsenovich AM. A study of raw material sources for Paleolithic objects in the Ikh-Tulbariin-Gol and Kharganyn-Gol River valleys in northern Mongolia[J]. PAEAS， 2014， 20： 87-90 （In Russ）

[34] Khatsenovich AM， Rybin EP， Shelepaev RA， et al. The impact of raw material sources on the distribution of Initial Upper Paleolithic humans in Central Asia[A]. In： 20th Congress of the INQUA. Book of Abstracts[M]. 2019： O-0056

[35] Derevianko AP， Kandyba AV， Petrin VT. Paleolithic of Orkhon[M]. Novosibirsk： IEAT SB RAS， 2010： 1-384 （In Russ）

[36] Derevianko AP， Zenin AN， Rybin EP. The technology of early Upper Paleolithic lithic reduction in northern Mongolia： the Tolbor-4 site[J]. Archaeol Ethnol Anthropol Eurasia， 2007， 29： 16-38

[37] Lbova LV， Rezanov IN， Kalmykov NP， et al. Environment and Humans in the Neopleistocene （Western Transbaikal and South-Eastern Cis-Baikal））[M]. Ulan-Ude： Publication of the Buryat Science Center， 2003： 1-174 （In Russ.）

[38] Klementiev AM. Landscapes of the Uda River （Transbaikalia） in the Late Pleistocene （on the fauna of large mammals）[D]. Tomsk：Tomsk State University， 2011 （In Russ）

[39] Goebel T， Aksenov M. Accelerator radiocarbon dating of the Initial Upper Paleolithic in Southeast Siberia[J]. Antiquity， 1995， 69： 349-357

[40] Ovodov ND. Fauna paleolithicheskikh poseleniy Tolbaga i Varvarina Gora d `zapadnom Zabaikalie （Fauna of Tolbaga and Varvarina Gora Paleolithic sites in western Transbaikalia）[A]. In： Prirodnaya Sreda I Chelovek V Pozdnem Antropogene（Environment and Human in Late Anthropogene）[M]. Ulan Ude： Geolog Institute BF SB AN USSR， 1987： 35-50 （In Russ）

[41] Derevianko AP， Nikolaev SV， Petrin VT. Geologiya i arkheologiya paleo- liticheskikh pamyatnikov Hangaya （Mongolia） （Geology and archaeology of Paleolithic sites of Khangai （Mongolia）[A]. In： International Symposium “Paleoecology and Ancient Human Occupation in Northern Asia and America”[M]. Krasnoyarsk， 1992： 79-83 （In Russ）

[42] Zwyns N， Lbova LV. The Initial Upper Paleolithic of Kamenka site， Zabaikal region （Siberia）： A closer look at the blade technology[J]. Archaeol R Asia， 2018， 17： 24–49

[43] Hughes S， Hayden TJ， Douady CJ， et al. Molecular phylogeny of the extinct giant deer， Megaloceros giganteus[J]. Mol Phylogenet Evol， 2006， 40（1）： 285-291

[44] Buvit I， Izuho M， Terry K， et al. Radiocarbon dates， microblades and Late Pleistocene human migrations in the Transbaikal， Russia and the Paleo-Sakhalin-Hokkaido-Kuril Peninsula[J]. Quat Int， 2016， 425： 100–119

[45] Abramova ZA Paleolithic of Northern Asia[A]. In： Paleolithic of the World： Caucasus and Northern Asia[M]. Leningrad： Nauka，1989： 145-244 （In Russ）

[46] Konstantinov MV. The Stone Age of the Eastern Part of Baikal Asia[M]. Chita – Ulan-Ude： Chita State Pedagogical Institute –Buryat Science Center， 1994： 1-179 （In Russ）

[47] Rybin EP， Lbova LV， Klementiev AM. Tool kit and settlement specifics of Early Upper Paleolithic complexes in western Transbaikal[A]. In： Paleolithic cultures of the Transbaikal and Mongolia （new sites， methods， hypothesis）[M]. Novosibirsk： IAET SB RAS Publ.， 2005： 69–80 （In Russ）

[48] Kasimov M. Environmental Atlas-monograph “Selenga-Baikal[M]. Moscow： Faculty of Geography， MSU， 2019： 1-288 （In Russ）

[49] Lbova LV， Kolomiets VL， Dergacheva MI， et al. Environment and climate during the Late Upper Pleistocene in the Western Transbaikalia region[J]. Archaeol Ethnol Anthropol Eurasia， 2005， 2（22）： 2–17

[50] Tashak VI， Savinova VV. Paleoenvironment of the south of Western Transbaikal in final Pleistocene–Holocene （on the materials of Podzvonkaya）[J]. Reports of the Laboratory of ancient technologies， 2009， 7： 39–51 （In Russ）

[51] Shaposhnikov FD. Wild sheep of Western Tian Shian[J]. Priroda， 1956， 1： 109-110 （In Russ）

[52] Lhagvasuren B， Milner-Gulland EJ. The status and management of the Mongolian gazelle， Procapra gutturosa population[J].ORYX， 1997， 31： 127-134

[53] Leimgruber P， McShea WJ， Brooks CJ， et al. Spatial patterns in relative primary productivity and gazelle migration in the eastern steppes of Mongolia[J]. Biol Conserv， 2001， 102 （2）： 205–212

[54] Foley RA. Another Unique Species： Patterns in Human Evolutionary Ecology[M]. London： Longman， 1987， 1-313

[55] Binford LR. Willow smoke and dogs’ tails： Hunter-gatherer settlement systems and archaeological site formation[J]. Am Antiquity， 1980， 45： 4–20

[56] Frison CG， Stanford DJ. The Agate Basin Site[M]. NY： Academic Press， 1982： 1-398

[57] Frison CG. Prehistoric Hunters of the High Plains[M]. NY： Academic Press， 1991： 1-532

[58] Wiesniewski A， Lauer T， Ch?on M， et al. Looking for provisioning places of shaped tools of the late Neanderthals： a study of a Micoquian open-air site， Pietraszyn 49a （southwestern Poland）[J]. Comptes Rendus Palevol， 2019， 18： 367–389

[59] Rybin EP， Kolobova KA. The structure of the lithic industries and the functions of Paleolithic sites in the Altai Mountains[J].Archaeol Ethnol Anthropol Eurasia， 2004， 4（20）： 20-34

[60] Khatsenovich AM， Shelepaev RA， Rybin EP， et al. Long distance transport and use of mica in the Initial Upper Paleolithic of Central Asia： An example from the Kharganyn Gol 5 site （northern Mongolia）[J]. J Arch Sci： Rep， 2020， 31： e102307