The development of Upper Palaeolithic China: new results from the Shuidonggou site.

Li, Feng ; Gao, Xing ; Chen, Fuyou 等

[ILLUSTRATION OMITTED]

Introduction

The replacement of archaic populations by anatomically modern humans, and the process of the Middle-Upper Palaeolithic transition in Eurasia during Oxygen Isotope Stage 3 (OIS 3) are heavily debated in the scientific community (e.g. Mellars 1990; Bar-Yosef & Pilbeam 2000; Mellars et al. 2007). Much discussion focuses on the age of blade technology, which is considered by many as a marker of modern humans, and its diffusion across Eurasia.

Shuidonggou Locality 1, located in northern China, has yielded what has been described as an initial Upper Palaeolithic assemblage with large blades produced by Levallois-like technology (Brantingham 1999; Brantingham et al. 2001). This site occupies a unique position in early prehistoric China (e.g. Jia et al. 1964; Zhang 1990, 1999a; Li 1993; Lin 1996; Gao et al. 2002, 2004) and historically has been aligned with the Eurasian Palaeolithic (Boule et al. 1928; Bordes 1968; Brantingham 1999; Brantingham et al. 2001). Given that there are few other well-studied, securely dated assemblages in China that resemble the early Eurasian Upper Palaeolithic (Lin 1996; Gao 1999), Shuidonggou plays an essential role in discussions of the diffusion of blade technology and even population migration across Eurasia from west to east.

The phase of research at Shuidonggou that began in 2003 focuses on the dating, depositional context, lithic industries, and behaviour patterns of several localities in the Shuidonggou Basin. Shuidonggou Locality 2, the subject of this paper, is significant for its unusually long sequence of seven distinct, well-stratified Palaeolithic layers and an abundance of archaeological material. The results from the investigation provide a new perspective on the origins and age of macroblade industries in the region.

The site

The Shuidonggou Basin is located in northern China, 18km east of the Yellow River on the margins of the Ordos Desert (Figure 1). It lies in an arid to semi-arid transition zone which is strongly seasonal and has a continental climate, dominated by the winter monsoon. The site cluster at Shuidonggou was first located and investigated by Emile Licent and Pierre Teilhard de Chardin in 1923 (Licent & Teilhard de Chardin 1925). Teilhard de Chardin initially noted five distinct localities in the Shuidonggou Basin. In the course of subsequent studies, another seven Palaeolithic localities have been identified (Zhang 1999b; Gao et al. 2004, 2009; Liu et al. 2008).

The Palaeolithic deposits in this area cover a time span of roughly 41-10 ka cal BP (Table 1). Several technological complexes have been identified, marked by the presence of large blade technology, simple core-flake technology and microblade technology. More specifically, Localities 1 and 9 and the earliest layers at Locality 2 yield assemblages with large blade production incorporating aspects of Levallois technology, for which Shuidonggou is best known. Most of the layers at Localities 2, 7 and 8 contain assemblages with simple core-flake technology. Evidence of microblade technology was discovered at Locality 12 (Liu et al. 2008; Gao et al. 2009), where it is dated to 11 ka by optically stimulated luminescence (OSL) dating (Liu et al. 2008). Microblades and cores were also found at the surface of Locality 6 (Zhang 1999b). To date, the assemblages from other localities are either small or difficult to classify.

Shuidonggou Locality 2 was one of the five localities originally identified in 1923 by Licent and Teilhard de Chardin (1925). Madsen et al. (2001) and Gao et al. (2002) conducted some radiocarbon (AMS [sup.14]C) dating work in 1999 and 2000 based on samples from around hearths exposed in the natural profile (Madsen et al. 2001; Gao et al. 2002). They placed the occurrence of blade technology in this area at an age of between 29 ka and 24 ka ([sup.14]C BP) based on the dates from Locality 2, and suggested that large blade technology spread from north to south during the Upper Palaeolithic. From 2003 to 2007 Gao et al. excavated several localities (Gao et al. 2006, 2008a; Pei et al. 2012), Locality 2 being one of the most intensively studied sites.

[FIGURE 1 OMITTED]

Stratigraphy and chronology at Locality 2

Locality 2 has been excavated over an area of almost 100m2 (Figures 2 and 3), revealing seven cultural layers (CL) containing several hearths, thousands of stone artefacts, bone fragments and some ostrich eggshell beads. Eleven hearths or depositional features related to hearths were identified: two of them in CL1, seven in CL2, one in CL3 and one in CL4. All are flat to slightly basin-shaped unprepared hearths ranging in diameter from 0.2-1m and in depth from 40-100mm. The hearths are surrounded by charcoal fragments, stone artefacts and bones (Figure 3). Fire-cracked pebbles were found in or immediately adjacent to most of the hearths. Most of the bone fragments were discovered in hearths and close to them suggesting that meat preparation and consumption was concentrated around the fireplaces (Guan et al. 2011).

The exposed strata reached a total thickness of 12.5m (Figure 4). The sedimentary sequence from unit two, the more complete of the two trenches, is described and interpreted as mainly lacustrine deposits by Liu et al. (2009). The sediments at the base are fine sand and gravel; these give way successively to a greyish-black peat deposit (CL7), then light greyish-green silt, and finally light greyish-yellow silt (see Liu et al. 2009 for complete and detailed descriptions of stratigraphy). A total of 18 substrata are described (Figure 4), seven of which contain relatively concentrated debris from Palaeolithic occupations. A few stone artefacts were also collected from other substrata.

[FIGURE 2 OMITTED]

The combined radiocarbon dates (Table 1) show that the first cultural layer (CL7) falls within the period from 41.5-34.4 ka cal BP; the second and third (CL6 and 5) are expected to date from 34.4-32.6 ka cal BP (based on ages of strata above and below); the fourth and fifth (CL4 and 3) 32.6-31.4 ka cal BP; the sixth (CL2), 31.3-29.9 ka cal BP; the seventh (CL1) 20.3 ka (OSL BP) (Li et al. 2013).

Technological and typological features of the assemblages

The range of materials collected from Locality 2 during the recent excavation includes lithic artefacts, animal fossils and ostrich eggshell beads. The three-dimensional coordinates of specimens discovered in situ were recorded with a Total Station. All sediments from 20-50mm artificial levels were dry-sieved through fine mesh (c. 2mm.). The sample from Locality 2 is thus relatively complete, and is also large enough for our analysis (> 15 000 stone artefacts).

Table 2 summarises some of the basic technological characteristics of the lithic assemblages from different cultural layers from Locality 2. Most of the artefacts are manufactured from quartz sandstone, low-quality chert and silicified dolomite obtained as well-rounded pebbles from nearby river banks. Based on the size of the artefacts, the pebbles selected appear mostly to have ranged from 30-150mm in diameter. A small proportion of artefacts produced from black and grey high-quality chert in the assemblage from CL2 preserve white, chalky cortex, showing that they were obtained directly from a source of chert nodules, rather than from secondary alluvial deposits. Unfortunately, this primary source has not yet been identified.

[FIGURE 3 OMITTED]

In terms of the retouched tool inventory, all assemblages from Locality 2 are clearly flake-based, although there is some variation among the various layers. Blade and blade-like flakes as blanks for tools are extremely rare, but CL7 and CL5a yielded two large blade cores (Figure 5), the only ones from the entire sequence. Overall, the majority of stone tools from CL6, CL5b and CL4-CL1 exhibit clear features of the small flake tool tradition of northern China (Zhang 1990, 1997, 2002).

The proportion of cores is quite small in every cultural layer. According to the morphological and technological traits, all cores from CL6 to CL1 except for CL5a exhibit features of simple flake manufacture with free-hand percussion (Figure 6); the two cores from CL7 and CL5a that were clearly used for systematic blade production are obvious exceptions (Figure 5). The core from CL5a is a Levallois-like flat-faced bidirectional core with two faceted platforms, and the other, from CL7, is an edge-faceted blade core with platforms on two opposite ends. These two cores are regionally distinctive but exhibit obvious similarities to cores from the larger assemblage at Shuidonggou Locality 1, which has been described as an initial Upper Palaeolithic industry (Brantingham 1999; Brantingham et al. 2001). Cores from CL6, CL5b and CL4-CL1 were all exploited to produce simple flakes and show no preparation of the platforms and working surfaces. Bipolar cores and flakes were found in CL5 to CL1 (Figure 6), but their number and proportion both increase dramatically in the most recent assemblage (CL1a). In most cases, hard hammer percussion seems to have been the dominant technique for detaching flakes. However, several flat blanks from CL2 flaked on black chert from the primary (non-local) source exhibit traces of soft-hammer percussion, including small or invisible platforms, a distinct lip on the ventral edge of the platform, and evidence of preparation by grinding at the exterior of the platform (e.g. Kuhn 2004). There are no counterpart cores showing evidence of soft-hammer percussion from this layer, although we might not expect to find them if the raw material source were far away.

[FIGURE 4 OMITTED]

[FIGURE 5 OMITTED]

[FIGURE 6 OMITTED]

Overall, the assemblages from Locality 2 reveal two broadly different core reduction technologies. Cores from CL7 and CL5a demonstrate clear features of large blade technology which connects these layers with Locality 1. The dates from CL7 (41-34ka cal BP) and CL5a (> 32.6 ka cal BP) are also in reasonably good agreement with dates from the layers (34 ka and 38 ka, U-Th BP) containing products of a similar core reduction sequence at Locality 1 (Li et al. 2013). Cores from other main cultural layers (CL6, CL5b and CL4-CL1) at Locality 2 show the simple free-hand core reduction and bipolar reduction which are very common at contemporary late Pleistocene Palaeolithic sites in northern China (e.g. Zhang 1990, 1997, 2002).

No retouched tools were recovered from the layers associated with large blade cores at Locality 2. The retouched tools from CL6, CL5b and CL4-CL1 are typologically and technologically characteristic of the northern Chinese Late Pleistocene Palaeolithic (Table 3). The frequency of retouched tools is very low, as is the intensity of modification on each specimen. The most abundant retouched tools from CL6, CL5b and CL4-CL1 are side-scrapers (Figure 7), most of which are manufactured on relatively flat flakes. End-scrapers manufactured mainly on flakes are the second most common artefact class in CL2 and CL1a, while endscrapers are nearly absent from earlier layers. Other tool forms, including points, notches, burins, drills and choppers occur in small numbers in the assemblages from every cultural layer at Locality 2.

For a variety of reasons related to both conditions of preservation and human activities, a small sample of faunal remains was retained from the 2003-2007 excavation, but unfortunately most of the fauna elements are small bone fragments for which it is difficult to make taxonomic determinations. Based on tooth counts the mammal component of the fauna from CL2, the largest faunal assemblage, is dominated by Equus hernionus and Antelopina. While the use of ostrich (Struthio sp.) for food is unclear, a handful of ostrich eggshell beads (Struthio andersoni, from analysis on the collections from the surface) (Wang et al. 2009) were recovered. All beads come from CL2, as does one bone needle fragment.

[FIGURE 7 OMITTED]

Discussion

The reduction sequences indicated by cores from CL7 and CL5a at Shuidonggou Locality 2 fit well with the general characteristics of initial Upper Palaeolithic technology from Locality 1. The manufacture of blades by hard-hammer percussion from prepared cores seems to be an especially good indicator of this phase. Not surprisingly, distinctive 'index fossils' of other initial Upper Palaeolithic or 'transitional' assemblages from the Near East--chanfreins, Emireh and Uum et Tlel points (Kuhn et al. 1999)--are not found in the Shuidonggou Upper Palaeolithic layers. However, the Shuidonggou materials are similar to the earliest Upper Palaeolithic assemblages from south Siberia, such as Kara Born, and from Mongolia, such as Chikhen Agui (Brantingham et al. 2001), suggesting the existence of a regional initial Upper Palaeolithic in central Asia, south Siberia, Mongolia and north-west China.

Re-examination of the dates of initial Upper Palaeolithic layers at Shuidonggou Localities 1 and 2 places this stage within a rough time span of 38-34 ka (Li et al. 2013). Taking account of these dates for Levallois-like blade technology in northern China, the hypothesis that the technology spread from north to south and that "Shuidonggou is the latest initial Upper Palaeolithic assemblage yet known in all of Eurasia" proposed by Madsen et al. (2001: 715) should be re-evaluated. The radiocarbon ages cited by Madsen et al. (2001) were derived from samples from hearths located higher in the sequence, probably CL2; no such fireplaces were identified below CL4. The estimate of 41-34 ka for CL7 and the age determination of >32.6 ka (CL3) for CL5a place them within the same time range as the Mongolian sites with macroblade technology. Although dates for the Siberian assemblages with initial Upper Palaeolithic blade technology are considerably older (e.g. Derevianko 2011), the temporal gap between Shuidonggou and the Siberian sites is shorter than previously estimated, suggesting a more rapid spread of techniques or populations from the north and west. Unfortunately, for the time being at least, the absence of well-dated sites in adjacent parts of north-west China, such as Xinjiang Province and Inner Mongolia, make it impossible to chart the introduction and spread of blade technology into north-west China in greater detail.

It is also clear that the late Pleistocene sequence from Shuidonggou is more diverse than previously characterised. Assemblages from CL6 to CL1 (except for CL5a) at Locality 2 are similar if not identical to many late Palaeolithic assemblages in northern China predating the emergence of microblade technology, and fit technologically and typologically within the so-called flake-tool cultural tradition (Zhang 1990, 1997, 2002). These assemblages share a number of general features including local raw material exploitation, free-hand percussion, amorphous or variable cores, irregular flakes, high proportions of chunks and debris, and informal tools with little retouch, sometimes combined with bipolar products. Some behavioural changes are also observed among different layers at Locality 2, including variable intensity of occupations (based on densities of finds), different patterns of raw material procurement in CL2 and an increase in bipolar reduction products in CL1a. Assemblages from main cultural layers (CL6, CL5b and CL4-CL1) share little with the initial Upper Palaeolithic and appear to have their roots in the Late Pleistocene Palaeolithic industries of northern China. The existence of such different technological systems in successive layers runs counter to the common impression of Shuidonggou as a site containing only Levallois-like blade technology. Levallois-like blade technology was practiced in this area for one or more relatively brief periods during the Upper Palaeolithic, after which more typical small flake-tool technologies were produced for almost 14 000 years. Moreover, the early blade technology had no obvious impact on the practices of subsequent occupants according to the archaeological materials, probably indicating that two different populations, representing different cultural traditions, occupied the Shuidonggou area successively during OIS 3.

Madsen et al. (2001) proposed a hypothesis based on the findings at Shuidonggou Locality 2 that microblade technology originated from a combination of blade technology and bipolar technology. Examination of the bipolar products at Locality 2, especially the cores, cannot tell us about the details of that evolutionary process. Even though the detached products are similar in dimension, bipolar technology involved very distinct flaking procedures from microblade technology, which requires a systematic preparation of the platform and working surface and usually employs a pressure flaking technique (Kuzmin et al. 2007). Moreover, cultural layers at Locality 2 which yielded artefacts made by bipolar technology yielded no blade products, implying that blade technology did not influence bipolar technology at all. The abundance of bipolar products in CL1a is not a precursor to the development of microblades, but instead represents a response to some functional and economic requirement for very small flakes, the nature of which is currently unknown.

Conclusion

The varied lithic technology from Locality 2, combined with Shuidonggou's geographic position between arid and semi-arid areas, should lead to a better understanding of the western and eastern Eurasian Upper Palaeolithic sequences, and the possible interactions between these two areas. The Upper Palaeolithic in northern China/East Asia is regionally very distinct from the blade-dominated Upper Palaeolithic of Western Europe, and Western Europe should not be seen as typical of the processes that occurred in East Asia. As Shuidonggou Locality 2 demonstrates, a distinctive form of macroblade technology was introduced into northern China, probably from Mongolia or Siberia, as early as 40 ka, but subsequently disappeared, to be replaced by local flake-based production systems. The scale of variation among these areas should stimulate scholars who are interested in the Palaeolithic in East Asia to propose a unique Palaeolithic system for East Asia, not only in terms of stages of the Palaeolithic (Gao 1999; see also Gao & Norton 2002), but also the behavioural patterns and adaptive strategies in East Asia.

Received: 10 February 2012; Accepted: 10 May 2012; Revised: 11 July 2012

Acknowledgements

The Shuidonggou Project was initiated by Professor Gao Xing, from the Institute of Vertebrate Paleontology and Paleoanthropology (Beijing), and Professor Wang Huimin, from the Archaeological Institute of Ningxia Hui Autonomous Region. The authors are grateful to Luo Feng, Zhong Kan, Feng Xingwu, Luo Zhigang, Mei Huijie, Wang Chunxue, Peng Fei, Zhou Zhenyu, Ma Ning and Yi Mingjie for their significant contributions to the SDG fieldwork, and to Niu Dongwei, Xu Xin, Wei Yi, Xu Ting, Xing Luda and Li Jingshu for their assistance with the collection of artefact metric data. The authors would like to thank Mary C. Stiner and John W. Olsen for discussion of some issues in this research. We also thank two reviewers for valuable comments on the manuscript. Li Feng thanks the China Scholarship Council (PRC) that allowed him to carry out part of his dissertation research at the University of Arizona. This work was supported by the CAS Strategic Priority Research Program (Grant No. XDA05130202) and National Natural Science Foundation of China (Grants No. 41102016 and No. 41272032).

References

BAR-YOSEF, O. & D. PILBEAM (ed.). 2000. The geography of Neanderthals and modern humans in Europe and the greater Mediterranean (Peabody Museum Bulletin 8). Cambridge (MA): Peabody Museum of Archaeology and Ethnography, Harvard University.

BORDES, E 1968. The OM Stone Age. New York (NY): McGraw-Hill.

BOULE, M.H., E. BREUIL, E. LICENT & P. TEILHARD DE CHARDIN. 1928. Le Paleolithique de la Chine. Paris: Archives de l'Institut de Paleontologie Humaine.

BRANTINGHAM, P.J. 1999. Astride the Movius Line: Late Pleistocene lithic technological variability in northeast Asia. Unpublished PhD dissertation, University of Arizona.

BRANTINGHAM, P.J., A.I. KRIVOSHAPKIN, J.Z. LI & Y. TSERENDAGVA. 2001. The initial Upper Paleolithic in northeast Asia. Current Anthropology 42: 735-47.

DEREVIANKO, A.P. 2011. The Upper Paleolithic in Africa and Eurasia and the origin of anatomically modern humans. Archaeology, Ethnography and Anthropology of Eurasia 39: 2-31.

GAO, X. 1999. A discussion of the Chinese Middle Paleolithic. Acta Anthropologica Sinica 18: 1-16 (in Chinese).

GAO, X. & C.J. NORTON. 2002. A critique of the Chinese 'Middle Paleolithic'. Antiquity 76: 397-412.

GAO, X., J.Z. LI, D.B. MADSEN, P.J. BRANTINGHAM, R.G. ELSTON & R.L. BETTINGER. 2002. New 14C dates for Shuidonggou and related discussions. Acta Anthropologica Sinica 21:21 l-18 (in Chinese).

GAO, X., S.W. PEI, H.M. WANG & K. ZHONG. 2004. A report on Paleolithic reconnaissance in Ningxia, north China. Acta Anthropologica Sinica 23:307-25 (in Chinese).

GAO, X., H.M. WANG, S.W. PEI, EY. CHEN & X.W. FENG. 2006. Tracing the footprints of human groups from Shuidonggou, in Institute of Archeology of Yunnan Province (ed.) Collected works for "The 40th Anniversary of Yuanmou Man Discovery and the International Conference on Palaeoanthropological Studies". 152-58. Kunming: Yunnan Science and Technology (in Chinese).

GAO, X., H.M. WANG, S.W. PEI, X.W. FENG & F.Y. Chen. 2008a. Progress of the reconnaissance and excavation in Shuidonggou area, in X. Gao, J.M. Shi & X.W. Feng (ed.) Essays in honor of Prof. Jia Lanpo an the hundredth anniversary of his birth: 230-39. Beijing: Science (in Chinese).

GAO, X., B.Y. YUAN, S.W. PEI, H.M. WANG, EY. CHEN & X.W. Feng. 2008b. Analysis of sedimentary-geomorphologic variation and the living environment of hominids at the Shuidonggou Paleolithic site. Chinese Science Bulletin 53: 2025-32.

GAO, X., H.M. WANG, D.C. LIU, S.W. PEI, EY. CHEN, X.L. ZHANG & Y. ZHANG. 2009. A study of fire-use activities at Shuidonggou Locality 12. Acta Anthropologica Sinica 28:329-36 (in Chinese).

GUAN, Y., X. GAO, H.M. WANG, EY. CHEN, S.W. PEI, X.L. ZHANG & Z.Y. ZHOU. 2011. Spatial analysis of intra-site use at a Late Palaeolithic site at Shuidongou, northwest China. Chinese Science Bulletin 56: 3457-63.

JIA, L.P., R GAI & Y.X. LI. 1964. New materials from the Palaeolithic site of Shuidonggou. Vertebrata PalAsiatica 8:75-86 (in Chinese).

KUHN, S.L. 2004. Upper Paleolithic raw material economies at Ucagizlicave, Turkey. Journal of Anthropological Archaeology 23: 431-48.

KUHN, S.L., M.C. STINER & E. GULE 1999. Initial Upper Palaeolithic in south-central Turkey and its regional context: a preliminary report. Antiquity 73: 505-17.

KUZMIN, Y.V, S.G. KEATS & C. SHEN (ed.). 2007. Origin and spread of microblade technology in northern Asia and North America. Vancouver: Archaeology.

LI, F., S.L. KUHN, X. GAO & F.Y. CHEN. 2013. Re-examination of the date of large blade technology in China--a comparison of Shuidonggou Locality 1 and Locality 2. Journal of Human Evolution 64: 161-68. doi: 10.1016/j.jhevol.2012.11.001

LI, Y.X. 1993. A discussion of the Upper Paleolithic industries of China. Acta Anthropologica Sinica 12: 214-23 (in Chinese).

LICENT, E. & P. TEILHARD DE CHARDIN. 1925. Le Palaeolithique de la Chine. L'Anthropologie 25: 201-34.

LIN, S.L. 1996. Comparisons of Chinese and western Palaeolithic technological models. Acta Anthropologica Sinica 15:1-20 (in Chinese).

LIU, D.C., EY. CHEN, X.L. ZHANG, S.W. PEI, X. GAO & Z.K. Xia. 2008. Preliminary comments on the paleoenvironment of the Shuidonggou Locality 12. Acta Anthropologica Sinica 27:295-303 (in Chinese).

LIU, D.C., X.L. WANG, X. GAO, Z.K. XIA, S.W. PEI, F.Y. CHEN & H.M. WANG. 2009. Progress in the stratigraphy and geochronology of the Shuidonggou site, Ningxia, north China. Chinese Science Bulletin 54: 3880-86.

MADSEN, D.B., J.Z. LI, P.J. BRANTINGHAM, X. GAO, R.G. ELSTON & R.L. BERRINGER. 2001. Dating Shuidonggou and the Upper Palaeolithic blade industry in north China. Antiquity 75: 706-16.

MELLARS, P. (ed.). 1990. The emergence of modern humans: an archaeological perspective. Ithaca (NY): Cornell University Press.

MELLARS, P., K. BOYLE, O. BAR-YOSEF & C. STRINGER (ed.). 2007. Rethinking the human revolution: new behavioural and biological perspectives on the origin and dispersal of modern humans (McDonald Institute Monographs). Cambridge: McDonald Institute for Archaeological Research.

PEI, S.W., X. GAO, H.M. WANG, K. KUMAN, C.J. BAE, F.Y, CHEN, V. GUAN, Y. ZHANG, X.L. ZHANG, F. PENG & X.L. LI 2012. The Shuidonggou site complex: new excavations and implications for the earliest Paleolithic in north China. Journal of Archaeological Science 39:3610-26.

WANG, C.X., Y. ZHANG, X. GAO, X.L. ZHANG & H.M. WANG. 2009. Archaeological study of ostrich eggshell beads collected from Shuidonggou site. Chinese Science Bulletin 54: 3887-95.

ZHANG, S.S. 1990. Regional industrial gradual advance and cultural exchange of Paleolithic in north China. Acta Anthropologiea Sinica 9:322-34 (in Chinese).

--1997. On the problems of seeking the earliest (about 2 Ma) human remains in China. Acta Anthropologica Sinica 16:87-95 (in Chinese).

--1999a. On the important advancement of the Paleolithic archaeology in China since 1949. Acta Anthropologica Sinica 18:193-214 (in Chinese).

--1999b. A study of stone artefacts found at the Xiaokouzi prehistoric site. Acta Anthropologica Sinica 18:81-100 (in Chinese).

--2002. Advances in Chinese Paleolithic archaeology in the past 20 years and some comments. Quaternary Science 22:11-19 (in Chinese).

Feng Li (1), Xing Gao (1), *, Fuyou Chen (1), Shuwen Pei (1), Yue Zhang (1), Xiaoling Zhang (1), Decheng Liu (1), Shuangquan Zhang (1), Ying Guan (1), Huimin Wang (2) & Steven L. Kuhn (3)

(1) Key Laboratory of Vertebrate Evolution and Human Origin of Chinese Academy of Sciences, Institute of Vertebrate Paleontology and Paleoanthropology, Chinese Academy of Sciences, 142 Xizhimenwai Street, Beijing 100044, China (Email: lifengivpp@gmail.com; gaoxing@ivpp.ac.cn; chenfuyou@ivpp.ac.cn; peishuwen@ivpp.ac.cn; zhangyue@ivpp.ac.cn; zhangxiaoling@ivpp.ac.cn; liudecheng@ivpp.ac.cn; zhangshuangquan@ivpp.ac, cn; guanying@ivpp.ac.cn)

(2) Institute of Archaeology of Ningxia Hui Autonomous Region, 121 Limin Street, Yinchuan 750001, China (Email: huimin.wang123@163.com)

(3) Department of Anthropology, Building 30, University of Arizona, Tucson, AZ 85721-0030, USA (Email: skuhn@email.arizona.edu)

* Author for correspondence

Table 1. Dating results from Shuidonggou Locality 2.

Cultural
layer Original unit Context Material

SDG2-CL1 Strata 4 Profile Sediment

SDG2-CL2 Hearth 1 Profile Charcoal

SDG2-CL2 Hearth 2 Profile Charcoal

SDG2-CL2 Hearth 2 Profile Ostrich eggshell

SDG2-CL2 Hearth 3 Profile Charcoal

SDG2-CL2 Hearth 4 Profile Charcoal

SDG2-CL2 Hearth 5 Profile Charcoal

SDG2-CL2 Hearth 7 Profile Charcoal

SDG2-CL2 Hearth 10A Profile Charcoal

SDG2-CL2 Strata 6 Profile Ostrich eggshell

SDG2-CL2 Strata 6 Profile Charcoal

SDG2-CL2 Strata 6-2L3 In situ Charcoal

SDG2-CL2 Strata 6-L18 In situ Charcoal

SDG2-CL2 Strata 6-L20-H6 In situ Charcoal

SDG2-CL2 Strata 6-2L4 In situ Charcoal

SDG2-CL2 Strata 6-L20-H7 In situ Charcoal

SDG2-CL2 Strata 6-L21-H7 In situ Charcoal

SDG2-CL3 Strata 8-L27 In situ Bone

SDG2-CL3 Strata 8-L28 In situ Bone

SDG2-CL3 Strata 8 Profile Sediment

SDG2-CL4 Strata 10 Profile Sediment

SDG2-CL4 Strata 10-L30 In situ Charcoal

SDG2-CL5b Strata 13 Profile Sediment

SDG2-CL5b Strata 13 In situ Bone

SDG2-CL6 Upper part of Profile Sediment
 Strata 15

SDG2-CL6 Lower part of Profile Sediment
 Strata 15

SDG2-CL7 Upper part of Profile Sediment
 Strata 16

SDG2-CL7 Lower part of Profile Wood
 Strata 16

SDG2-CL7 Strata 16 In situ Wood

Cultural
layer Dating method Lab # Age (BP)

SDG2-CL1 OSL IEE1880 20 300 [+ or -] 1000

SDG2-CL2 AMS [sup.14]C Bata-132982 26 350 [+ or -] 190

SDG2-CL2 AMS [sup.14]C Bata-132983 25 670 [+ or -] 140

SDG2-CL2 AMS [sup.14]C Bata-132984 26 930 [+ or -] 120

SDG2-CL2 AMS [sup.14]C Bata-134824 26 830 [+ or -] 200

SDG2-CL2 AMS [sup.14]C Bata-134825 25 650 [+ or -] 160

SDG2-CL2 AMS [sup.14]C Bata-146355 26 310 [+ or -] 170

SDG2-CL2 AMS [sup.14]C Bata-146357 29 520 [+ or -] 230

SDG2-CL2 AMS [sup.14]C Bata-146358 23 790 [+ or -] 180

SDG2-CL2 AMS [sup.14]C Bata-207935 28 420 [+ or -] 160

SDG2-CL2 AMS [sup.14]C Bata-207936 28 330 [+ or -] 170

SDG2-CL2 AMS [sup.14]C BA110217 26 450 [+ or -] 120

SDG2-CL2 AMS [sup.14]C BA110218 30 360 [+ or -] 120

SDG2-CL2 AMS [sup.14]C BAl10219 25 090 [+ or -] 90

SDG2-CL2 AMS [sup.14]C BAl10220 26 040 [+ or -] 90

SDG2-CL2 AMS [sup.14]C BA110221 2 520 [+ or -] 30

SDG2-CL2 AMS [sup.14]C BA110226 895 [+ or -] 30

SDG2-CL3 AMS[sup.14]C BA110223 28 290 [+ or -] 110

SDG2-CL3 AMS [sup.14]C BA110222 27 190 [+ or -] 100

SDG2-CL3 OSL IEE1881 27 800 [+ or -] 1400

SDG2-CL4 OSL IEE1882 20 500 [+ or -] 1100

SDG2-CL4 AMS [sup.14]C BA110224 985 [+ or -] 30

SDG2-CL5b OSL IEE1883 29 200 [+ or -] 2100

SDG2-CL5b AMS [sup.14]C BA110227 20 280 [+ or -] 70

SDG2-CL6 OSL IEE1884 23 600 [+ or -] 2400

SDG2-CL6 OSL IEE1885 38 300 [+ or -] 3500

SDG2-CL7 AMS [sup.14]C BA07940 29 759 [+ or -] 245

SDG2-CL7 AMS [sup.14]C BA07943 36 329 [+ or -] 215

SDG2-CL7 AMS [sup.14]C BA110228 980 [+ or -] 30

Cultural
layer Cal (BP) * (95.4%) Reference

SDG2-CL1 Liu et al. 2009

SDG2-CL2 30 984 [+ or -] 152 Madsen et al. 2001;
 Gao et al. 2002

SDG2-CL2 30 519 [+ or -] 175 Madsen et al. 2001;
 Gao et al. 2002

SDG2-CL2 31 273 [+ or -] 88 Madsen et al. 2001;
 Gao et al. 2002

SDG2-CL2 31 239 [+ or -] 111 Madsen et al. 2001;
 Gao et al. 2002

SDG2-CL2 30 503 [+ or -] 197 Madsen et al. 2001;
 Gao et al. 2002

SDG2-CL2 30 966 [+ or -] 147 Madsen et al. 2001;
 Gao et al. 2002

SDG2-CL2 34 149 [+ or -] 342 Madsen et al. 2001;
 Gao et al. 2002

SDG2-CL2 28 607 [+ or -] 290 Madsen et al. 2001;
 Gao et al. 2002

SDG2-CL2 32 734 [+ or -] 330 Gao et al. 20086

SDG2-CL2 32 605 [+ or -] 344 Gao et al. 20086

SDG2-CL2 31 071 [+ or -] 92

SDG2-CL2 34 881 [+ or -] 124

SDG2-CL2 29 933 [+ or -] 199

SDG2-CL2 30 802 [+ or -] 142

SDG2-CL2 2 606 [+ or -] 77

SDG2-CL2 824 [+ or -] 53

SDG2-CL3 32 561 [+ or -] 300

SDG2-CL3 31 385 [+ or -] 94

SDG2-CL3 Liu et al. 2009

SDG2-CL4 Liu et al. 2009

SDG2-CL4 883 [+ or -] 48

SDG2-CL5b Liu et al. 2009

SDG2-CL5b 24 191 [+ or -] 151

SDG2-CL6 Liu et al. 2009

SDG2-CL6 Liu et al. 2009

SDG2-CL7 34 395 [+ or -] 328 Liu et al. 2009

SDG2-CL7 41 445 [+ or -] 213 Liu et al. 2009

SDG2-CL7 877 [+ or -] 47

* [sup.14]C dates were calibrated using Oxcal 4.1 online
software (IntCa109 curve).

Table 2. Technological features of assemblages from different
cultural layers at Shuidonggou Locality 2.

Cultural Blade Simple Bipolar Flake
layer core flake core core Flake fragment

CL1a -- 50 109 575 378
 -- 0.61% 1.33% 7.02% 4.61%

CL1b -- 4 4 8 6
 3.15% 3.15% 6.30% 4.69%

CL2 -- 17 13 780 312
 -- 0.80% 0.61% 36.90% 14.76%

CL3 -- 21 4 140 60
 -- 2.41% 0.46% 16.04% 6.87%

CL4 -- 2 2 25 14
 -- 2.44% 2.44% 30.49% 17.07%

CL5a 1 -- -- -- --
 10.00% -- -- -- --

CL5B -- 10 2 14 3
 -- 3.83% 0.77% 5.36% 1.15%

CL6 -- 2 -- 1 --
 -- 13.33% -- 6.67% --

CL7 1 1 -- 2 --
 6.67% 6.67% -- 13.33% --

Cultural Bipolar Fire-cracked
layer flake Chunk Manuport pebble

CL1a 831 6078 84 1
 10.14% 74.19% 1.03% 0.01%

CL1b 5 77 20 --
 3.94% 60.63% 15.75% --

CL2 68 858 11 7
 3.22% 40.59% 0.52% 0.31%

CL3 41 578 23 --
 4.70% 66.21% 2.63% --

CL4 5 31 2 --
 6.10% 37.80% 2.44% --

CL5a -- 1 8 --
 -- 10.00% 80.00% --

CL5B 3 150 68 --
 1.15% 57.47% 26.05% --

CL6 -- 11 -- --
 -- 73.33% -- --

CL7 -- 3 8 --
 -- 20% 53.33% --

Cultural Hammers/ Retouched Grinding
layer Anvil tool tool Total

CL1a 10 76 1 8193
 0.12% 0.93% 0.01%

CL1b 1 2 -- 127
 0.79% 1.58% --

CL2 -- 48 -- 2114
 -- 2.27% --

CL3 -- 6 -- 873
 -- 0.69% --

CL4 -- 1 -- 82
 -- 1.22% --

CL5a -- -- -- 10
 -- -- --

CL5B 3 8 -- 261
 1.15% 3.07% --

CL6 -- 1 -- 15
 -- 6.67% --

CL7 -- -- -- 15
 -- -- --

Table 3. Counts and frequencies of retouched tools from
different cultural layers at Shuidonggou Locality 2.

Cultural
layer Side-scraper End scraper Point Drill

CL1a 43 12 3 2
 56.58% 15.79% 3.95% 2.63%

CL1b 2 -- -- --
 100

CL2 28 8 -- 3
 58.33% 16.67% 6.25%

CL3 5 1 -- --
 83.33% 16.67%

CL4 1 -- -- --
 100

CL5a -- -- -- --

CL5B 6 2 -- --
 75.00% 25.00%

CL6 1 -- -- --
 100

CL7 -- -- -- --

Cultural Chopper/
layer Burin Notch chopping tool Other

CL1a 1 3 2 10
 1.32% 3.95% 2.63% 13.26%

CL1b -- -- -- --

CL2 -- 2 1 6
 4.17% 2.08% 12.50%

CL3 -- -- -- --

CL4 -- -- -- --

CL5a -- -- -- --

CL5B -- -- -- --

CL6 -- -- -- --

CL7 -- -- -- --