Search

For more than a half-century, obsidian provenancing has underpinned many archaeological investigations of peoples of the past. The pace of obsidian studies in this regard has gathered significantly since around 2007, and we review the literature to gain a sense of where this momentum has come from, and what it heralds. In part, there is a data revolution underway, arising thanks to the capabilities for rapid survey and analysis enabled by field-portable analytical equipment. Obsidian studies are also gaining a stronger foothold in regions of the world where the approach was previously under-exploited. Our survey spans progress made in obsidian studies in the Mediterranean, Central Europe, the Near East, the Caucasus, Northeast Asia and Tibet, the Eurasian Arctic and Alaska, Southeast Asia, the Americas, Oceania, and Africa and Arabia. We also consider methodological issues related to compatibility of differing geochemical analytical techniques, and the state of the art in obsidian geochemical classification. The proliferation of new observations brings opportunities in terms of development of regional and global databases, as well as challenges of calibration and validation of analyses made by different scientists and laboratories employing diverse instrumentation. Obsidian provenancing demonstrates the astonishing ranges of our ancestors’ interactions and networks, sometimes exceeding 1000 km and involving maritime transport.

Since the 1990s, a characteristic obsidian geochemistry has linked widespread archaeological assemblages spanning the Russian Far East, Korean Peninsula, and Northeast China. Referred to as PNK1, the source of this material has yet to be identified. As a contribution to solving this enduring puzzle, we report here analyses of a commercial specimen of obsidian exported from Chongjin in DPR (North) Korea. A combination of Neutron Activation Analysis (NAA) and Potassium Argon dating enable us to compare this piece with a large obsidian database for Northeast Asia. We find that the “Chongjin sample” is identical to PNK1 lithics from archaeological collections. While the exact source of the “Chongjin sample” remains unknown, we can more confidently locate the primary source for PNK1 lithics in DPR Korea. Based on an exhaustive literature review of the geology and geochemistry of volcanic glasses and other volcanic rocks in the northern part of the Korean Peninsula, and drawing on our own unpublished data, we suggest that the PNK1 source is most likely located south of Paektusan Volcano. This corroborates the existing evidence for the long-range transport of the material.

Geochemical analysis of 102 obsidian artifacts from the lower reaches of the Kolyma River was performed to understand the provenance of the raw material; previously, there were no such studies in this region. Sites under investigation belong to the Arctic Neolithic, generally dated to ca. 6000–1500 BP. Based on the data for potential obsidian sources in Northeastern Siberia and neighboring territories, available to us, it was found that all obsidian artifacts originated from the Lake Krasnoe source in Chukotka, with a straight-line distance of ca. 800–1100 km from archaeological sites of the Kolyma River. This is a remarkable example of long-distance exchange/transport of obsidian in Northeastern Siberia during the Stone Age. The Lake Krasnoe locale was the primary obsidian source for prehistoric populations in this vast region, including Chukotka, the Kolyma River basin, and Okhotsk Sea coast; this obsidian was also identified at some Alaskan sites near the Bering Strait.

Patterns for the emergence of pottery-making in greater East Asia based on radiocarbon dates associated with the earliest pottery assemblages are presented. According to a critical evaluation of the existing evidence, the oldest centers with pottery in East Asia are located in South China (dated to ca. 18,000 cal BP), the Japanese Islands (ca. 16,700 cal BP), and the Russian Far East (ca. 15,900 cal BP). The claim for earlier pottery in South China at the Xianrendong Cave, supposedly dated to ca. 20,000 cal BP, cannot be substantiated. The appearance of pottery in other parts of greater East Asia was a slow process, without clear diffusion from any of these centers toward the periphery. In neighboring Siberia, the oldest pottery dated to ca. 14,000 cal BP is known from the Transbaikal.

Several new finds of the Late Pleistocene porcupine (Hystrix brachyura vinogradovi) in the Altai Mountains of southern Siberia and the Urals occur far north of previously assigned range for porcupine. These finds have necessitated a renewed study of this species's chronology and spatial distribution. We conclude that the oldest records of this porcupine in the Ural Mountains date to MIS 5e, and its geographic range possibly included also the Altai at that time. Directly radiocarbon-dated porcupine bones in the Altai fall in MIS 3 (ca. 32,000–41,000 cal BP). It is the northernmost record of this species and the youngest find outside its current geographic range.

Asmerom et al. (2018) recently obtained U-series dates for the Obi-Rakhmat Grotto, a key Paleolithic site in Central Asia, with ages of up to ca. 107–109 ka. However, their paper contains numerous errors and inconsistencies which affect the conclusions, and they are discussed.

Yue et al. (2021) have proposed a model of two Upper Palaeolithic industries in Northeast China called the “Initial Microblade Industry” and the “Northern Microblade Industry”, with the latter originating from the former, and its spread to the neighbouring regions of Northeast Asia. We have found several inconsistencies in their factual data and methodology, and therefore the conclusions reached by Yue et al. (2021) cannot be accepted at face value. Using the most reliable evidence, the oldest true microblade assemblages are known from the Korean Peninsula. We suggest that this region is the more probable centre of diffusion of microblade technology in mainland Northeast Asia, including China and the Russian Far East.

The freshwater reservoir effect (FRE) was estimated for the northern part of West Siberia for the first time, based on securely dated samples from the late medieval town of Mangazeya existed mainly in AD 1601–1650. Twelve specimens of six species of freshwater fish were selected for 14C dating, and C and N stable isotopes analysis. The FRE varies from ca. 310 to ca. 1970 14C years, and there is no consistency in terms of relationship to feeding habits or trophic level of fish; the FRE fluctuations for Mangazeya, as well as for other regions in northern Eurasia, are noteworthy. This information should be taken into account when 14C dating of bones of humans and domestic animals, who consume significant amount of freshwater fish, is conducted. This also applies to 14C dating of food crusts on the surface of pottery.

For the first time, an obsidian provenance study is performed for artifacts from the Indigirka River basin (Northeast Siberia). The non-destructive ED–XRF analysis of seven obsidian items from the Buolumuna-Taasa site, dated to the Neolithic (ca. 5000–1000 cal BC), shows that all of them originated from the Lake Krasnoe source, situated ca. 1300 km away. This is a remarkable example of super-long-distance transport / exchange of obsidian in prehistory of Siberian Arctic, and it testifies in favor of extensive contacts of ancient population in Northeast Siberia since the Mesolithic, ca. 6900 cal BC, or even earlier.

Issues related to the identification of Late Pleistocene dogs from different sites in Eurasia, triggered by recent publications (see Germonpre et al., 2009 , 2012 ; Ovodov et al., 2011 ), are discussed. The main focus is the problem of how to distinguish wolves from early dogs on the basis of skull and teeth morphology. The studies by Germonpre et al., 2009 , Germonpre et al., 2012 , reporting so-called ‘Palaeolithic dogs’ from Předmosti, Goyet, and other sites in Eastern and Central Europe, have some serious deficiencies. In our opinion, more work needs to be done to understand the biological mechanisms involved in wolf domestication and until then, it is premature to classify these Palaeolithic canids as fully domesticated dogs or even incipient dogs.

The paper presents the first paleontological data (pollen, plant macrofossil, diatom and ostracod) on the sediments of Lake Beloye, which is situated in the forest-steppe zone of Western Siberia. Our study contributes to better understanding of the Middle–Late Holocene regional to local environmental changes. Regionally, we recognized a cold stage of 3.4–2.3 calibrated ka BP and a dry stage of 2.8–1.7 ka BP. The dry stage coincides with the climate-driven lowered level of the lake at 2.6–1.5 ka BP. Later, the climate changed to warmer and wetter, which resulted in a highest stand of the lake at 1.3–0.2 ka BP and was followed by the lowering of the lake. Forest-steppe persisted in this area during the whole period under consideration, but probably shifted southward. Such a shift is reflected in the environment around Lake Beloye similar to that of modern taiga, which existed at 3.2–1.3 ka BP, however, compared to the colder and wetter climate of taiga, the conditions were colder and drier. The local plant associations and the lake ecosystem were affected by those events of climate and lake level changes. The initial eutrophic sedge-reed swamp evolved to the oligo-mesotrophic sphagnum bog, which existed around the lake at 3.2–1.3 ka BP. Later, the bogging process was interrupted by the lake transgression, after which the dried bog around the lake was occupied by the birch forest. The lake ecosystem was strongly affected by the 3.4 ka BP climate change: the low-alkaline eutrophic lake with abundant water higher plants and diatoms changed to the alkaline oligotrophic lake with calcium-fixing macrophytic algae and angiosperms. This resulted in total change of lake's bio-geochemical and sedimentation patterns.

The sources of high quality volcanic glass (obsidian) forarchaeologicalcomplexes in the Amur River basin of the Russian Far East have been established, based on geochemical analyses by neutron activation and X-ray fluorescenceofboth‘geological’(primarysources)and‘archaeological’(artifactsfromtheNeolithicandEarly Iron Age cultural complexes) specimens. A majorobsidian source identified as the Obluchie Plateau, located in the middle course of the Amur River, was found to be responsible for supplying the entire middle and lowerpartsoftheAmurRiverbasinduringprehistory.Thesourcehasbeencarefullystudiedandsampledfor the first time. Minor use of three other sources was established for the lower part of the Amur River basin. Obsidian from the Basaltic Plateau source, located in the neighboring Primorye (Maritime) Province, was found at two sites of the Initial Neolithic (dated to ca.11,000e12,500 BP). At two other sites from the same timeperiod, obsidian fromastillunknown sourcecalled“Samarga”wasestablished.At theSuchuIslandsite oftheEarlyNeolithic(datedtoca.7200e8600BP),obsidianfromthe‘remote’sourceofShirataki(Shirataki-A sub-source) on Hokkaido Island (Japan) was identified. The range of obsidian transport in the Amur River basinwas from 50 to 750 kmwithin the basin, and from 550 to 850 km in relation to the ‘remote’ sources at the Basaltic Plateau and Shirataki-A located outside the Amur River valley. The long-distance transport/ exchange of obsidian in the Amur River basin in prehistory has now been securely established.

The human–environment interaction in the southern part of the Russian Far East is considered, based on current archaeological, chronological, palaeoenvironmental, zooarchaeological, and archaeobotanical data. The major branches of the economy and its dynamics throughout the final Late Pleistocene and the Holocene are reconstructed on the basis of primary indicators (animal and plant remains from cultural complexes). The main stages in the process of human–environment interaction are distinguished, with principal boundaries at ca. 4500 BP (appearance of hoed agriculture), ca. 3000 BP (beginning of animal breeding), and ca. 1500 BP (emergence of plough agriculture and intensive cattle breeding). In some regions, such as Sakhalin Island and the Kurile Islands, communities of hunter-fisher-gatherers continued to exist for a long time, up to the 17–18th centuries AD. The relationship between cultural (maritime adaptation and agriculture) and natural (climatic coolings and warmings, and sea level changes) processes was not direct, and the palaeoeconomy in the Russian Far East was not environmental-driven; migrations and exchange played a certain role in the introduction of productive economy. The history of human–environment interaction in the region under study is closely related to more general features of this process in greater East Asia.

Pottery tempered with rice chaff from the early Iron Age cemetery of Ban Non Wat site, northeast Thailand, has been subjected to direct AMS 14 C dating, using low temperature combustion with oxygen as originally developed by authors. The carbon yield (0.2–0.5%) testifies the suitability of this pottery for dating. However, not all the results are in agreement with expected archaeological ages and other 14 C dates from the studied site and neighboring site of Noen U-Loke. This calls for a thorough analysis and interpretation of pottery temper dates from the region.

Two sites were recently studied in the southern part of the Aral Sea. In borehole M1, several 14C dates were generated. Two major low stands of the Aral Sea, at ca. 1000–1500 BP (ca. 700–1200 cal BP) and ca. 4800–5400 BP (ca. 5300–5900 cal BP), and a higher level at ca. 1500–4800 BP (ca. 1200–5300 cal BP), were determined. Our data are generally consistent with previously received results, besides high level at ca. 4200–6600 cal BP contra to Boomer et al. (2000) [8] and extremely low level at ca. 1400–1900 cal BP contra to Tarasov et al. (1996) [16] and Boomer et al. (2000) [8] . Further 14C dating should clarify these discrepancies. In the mean time, it is obvious that catastrophes of the Aral Sea ecosystem, similar to the modern episode, which began in the 1960s, happened several times in the past.

Here I present a critical evaluation of the analysis conducted by Graf [Graf, K.E., 2009. “The Good, the Bad, and the Ugly”: Evaluating the radiocarbon chronology of the middle and late Upper Paleolithic in the Enisei River valley, south-central Siberia. Journal of Archaeological Science 36, 694–707] of the Paleolithic radiocarbon (14C) dataset for the Upper Paleolithic sites in the Yenisei River basin of Central Siberia. Graf applied a rating system to the corpus of existing 14C data for the region, and announced some new 14C dates. The results obtained, however, are highly biased due to several factors, including prejudice concerning the higher accuracy of the AMS technique in 14C dating and that 14C dates from the same cultural component should overlap with plus–minus two sigmas; ambiguities with the subdivision of the Lateglacial period; artificially high sample selection criteria; incomplete factual material; and unjustifiable and misleading statements on the 14C dating of fossil bones and the issue of Last Glacial Maximum human presence in Siberia. As a result, the conclusions made by Graf [Graf, K.E., 2009. “The Good, the Bad, and the Ugly”: Evaluating the radiocarbon chronology of the middle and late Upper Paleolithic in the Enisei River valley, south-central Siberia. Journal of Archaeological Science 36, 694–707] are unconvincing. New analysis is necessary in order to improve the quality of treatment of the original data for the Upper Paleolithic 14C chronology in the Yenisei River basin.

A study of the movement of people within Northeast Asia at the end of the Pleistocene is critical for understanding how and when some of the first human populations entered North America. Chemical source studies of obsidian may provide the evidence necessary to document people's migrations between these regions. Sixty two obsidian artifacts from the late Pleistocene and Holocene Ushki Lake sites in Kamchatka Peninsula were analyzed by instrumental neutron activation analysis (INAA). Data generated demonstrate that multiple obsidian sources throughout Kamchatka were exploited by the inhabitants of Ushki Lake, and allow us to document long-distance population movements during the late Pleistocene and Holocene. It is reasonable to expect that obsidian from Kamchatka might have been transported to Alaska. This is true for the Chukotka region of Northeastern Siberia; obsidian from Chukotka has been found in late Holocene archaeological sites in Alaska. Ultimately, an expanded study that includes all areas of Northeast Asia and Alaska may provide the data necessary to document the earliest movements of people in these regions.

Human remains from the Xarusgol Valley, Ordos Plateau, northwestern China, have been considered to date to the Late Pleistocene. In order to ascertain their true age, direct AMS 14 C dating of a femur collected in the early 1920s was conducted. The results demonstrate that the femur is very young, with one sample of ‘post-bomb’ age and the other sample c. 200 years old. This first direct dating of a Chinese fossil hominid underscores the need to apply the same methodology to other Chinese modern human fossils currently believed to be of Pleistocene age. 2006 Elsevier Ltd. All rights reserved.

This paper reviews the history of the woolly mammoth (Mammuthus primigenius Blum.) in Siberia and adjacent northern Asia. The particular emphases are the chronology and environment of mammoth existence and extinction, based on about 530 radiocarbon dates from about 230 localities with mammoth remains and palaeoenvironmental records of the last 50,000 years. Until ca. 12,000 radiocarbon years ago (BP), mammoths inhabited all of northern Asia, from the High Arctic to southern Siberia and northeastern China. Since ca. 12,000 BP, mammoth disappeared from major parts of Siberia and adjacent northern Asia, and survived mainly in the Arctic regions of Siberia, north of 69° northern latitude. However, recently, it was found that some mammoth populations continued to exist in central and southern Western Siberia until ca. 11,100–10,200 BP. ‘Normal’ size mammoths became extinct in mainland Siberia at the Pleistocene–Holocene boundary, ca. 9700 BP. On Wrangel Island in the High Arctic, small-sized mammoths survived into the Middle–Late Holocene, ca. 7700–3700 BP. Compared with previous studies, it is now possible to reveal the complex nature of the process of final mammoth extinction in Siberia, with some small populations surviving outside of the Arctic until ca. 10,000 BP. The extinction of mammoth was most probably caused by a combination of factors, such as global warming in the Late Glacial (since ca. 15,000 BP) and the disintegration of landscapes suitable for mammoths throughout the Upper Pleistocene, such as light forests with vast open spaces occupied by meadows and forest tundra. The expansion of forest vegetation after the Last Glacial Maximum in Siberia, including its northeastern part, created unsuitable habitats for herbivorous megafauna, especially for mammoths. However, the Holocene environment of Wrangel Island was not of ‘glacial’ type and this requires further studies. The relationship between mammoths and Upper Palaeolithic humans is also considered. The role of humans in the process of mammoth extinction was of secondary importance. The lack of direct evidences of mammoth hunting limits the estimation of its role in Upper Palaeolithic human subsistence. Siberia is undoubtedly the area where the final extinction of mammoth occurred, and the future study of this process is important to understand the patterns of Pleistocene megafaunal extinction in the Northern Hemisphere.

Bones of three human individuals, found at the Upper Palaeolithic site of Sungir (central European Russia) in graves with unique adornments and other rich burial goods, have recently been 14C-dated at the NSF-Arizona AMS Facility. Collagen for dating was extracted using slow demineralization of whole bone fragments in weak hydrochloric acid. For the Sungir 1 burial, a 14C value of ≈19 200 BP was obtained; for the double burial Sungir 2 and 3, 14C values of ≈26 200–27 200 BP (average ≈26 400 BP) were obtained. This contradicts earlier 14C dates made at the Oxford AMS Lab, with collagen extracted following Longin's technique: ≈22 900 BP for Sungir 1; and ≈23 800–24 100 BP (average ≈23 900 BP) for Sungir 2 and 3. The use of two different techniques of collagen extraction for the Tucson and Oxford dates might be the cause of the age differences. Additional studies need to be conducted to resolve these 14C age differences.

Riedel et al. (2011) discuss the processes of foraminifera transportation and colonization, and their implication to the Holocene history ofWestern Siberia and Aral Sea using work by Gusskov et al. (2008) and other papers (see below) as case studies. Upon acquaintance with the arguments presented and their evaluation, it is necessary to give comments and replies to Riedel et al. (2011) due to their views and misrepresentation of our work (see Gusskov et al., 2008). There are two main topics which deserve discussion: 1) transport of living foraminifera by birds and colonization of lacustrine environments; and 2) geological history of the Caspian and Aral seas in relation to the spread of benthic organisms, primarily foraminifera and mollusks. The major issue raised by Riedel et al. (2011) is the possibility of foraminifera transport by birds as concluded (Riedel et al., 2011:109). This is based on studies by different authors (Resig, 1974; Cann and De Deckker, 1981; Patterson, 1987; Almogi-Labin et al., 1992, 1995; Perthuisot, 1995; Boudreau et al., 2001; Abu-Zied et al., 2007; Wennrich et al., 2007; see Riedel et al., 2011:105). In our opinion, the migration of foraminifera by water birds is unproven. First, nobody has ever observed the transport of living foraminifera by waterfowl, to the best of our knowledge, and this is in accord with Riedel et al. (2011:109). Second, the examples of foraminifera existence in inland lakes and other reservoirs used by Riedel et al. (2011) are not convincing, and they can be sub-divided into two groups. As for the first group, in the references cited the transportation by birds of foraminifera is not discussed; these are Resig (1974), and Cann and De Deckker (1981). In the second group, the authors are convinced that birds transported foraminifera; however, they do not consider other ways of their appearance in lakes and other basins, and this should be briefly commented on. Information presented by Patterson (1987) shows that the foraminifera complex in Lake Tecopa (California, USA) does not represent a natural assemblage. This is supported by the distribution of foraminifera in the lake section; the majority of species are represented by single shells. The degree of shell preservation is quite poor, and is most

Instrumental neutron activation analysis was performed on 79 obsidian tools and flakes from 35 sites on Sakhalin Island dating from Upper Paleolithic ( c . 19,000 bp ) to Early Iron Age ( c . 2000–800 bp ). Due to the absence of volcanic glass on Sakhalin Island, raw materials from the nearest obsidian sources on Hokkaido Island, such as Oketo, Shirataki, Tokachi-Mitsumata, and Akaigawa, were also analysed. A strong correlation between the chemical compositions of obsidian artefacts from Sakhalin and volcanic glass sources from Hokkaido was discovered. In particular, the Oketo and Shirataki sources were used for tool manufacturing throughout all of Sakhalin Island's prehistory. The distances between sources and archaeological sites range from 200–1000 km. The intensive exchange of raw materials continued and even intensified after the appearance of the La Perouse (Soya) Strait between Hokkaido and Sakhalin about 10,000–8000 bp . The Sakhalin Island populations were deeply involved in the obsidian exchange network centered on Hokkaido.

Both the AMS radiocarbon technique and TL dating have been used to study the age of organic-tempered pottery from Gasya, one of the earliest Neolithic sites in the Russian Far East. The results obtained are consistent with the previous set of radiocarbon dates from the site, that were obtained for charcoal, and show that the hypothesis that the earliest pottery from the Amur River basin dates to before 10,000 BP (calendar years) is valid. The organic-tempered pottery from the Russian Far East therefore looks promising for future dating by the AMS radiocarbon and TL methods.

The presence of Asiatic ostrich in Central Asia in the later Cenozoic time is well-documented; nevertheless, a few direct age determinations existed until recently. We performed AMS 14C dating of ostrich eggshells found in Mongolia, Transbaikal, and Tuva. It shows that ostriches existed throughout the second part of Late Pleistocene, until the Late Glacial time (ca. 13,000–10,100 BP). It seems that Asiatic ostrich went extinct in Central Asia just before or even in the Holocene.

This paper presents reservoir age determinations of pre-bomb marine mollusc shells from Sakhalin Island, Sea of Japan and Okhotsk Sea. The samples were collected from Japanese waste disposal sites created between 1905 and 1945. Radiocarbon analyses of the shells are used to establish marine reservoir age corrections for each site. The ΔR value for the Sea of Japan, +95 ± 60 years, is consistent with previous data. The ΔR value for the Okhotsk Sea is found to be +578 ± 50 years. The large difference in reservoir age between the two locations results from significantly different sources of surface water. The water source for the Sea of Japan is the Tsushima Current, a branch of the Kuroshio Current that originates in the equatorial Pacific and has a ΔR value close to the mean ocean value. The primary water source for the Okhotsk Sea is the Oyashio Current, which transports water from the Bering Sea in the open Northern Pacific. This source is depleted with 14C, as compared with waters of the equatorial Pacific. The southern Kuriles (Zeleny and Yuri islands) reflect a mixture of Oyashio and Kuroshio waters, with a ΔR value of +354 ± 23 years.

Russian Far East as a part of Northeast Asia is one of the centres of early plant domestication, particularly with respect to millet. New radiocarbon AMS dates associated with the pollen of cultivated cereals (Cerealia) from Late Neolithic layers on the Boisman 2 and Novoselische 4 sites allow estimates of the beginning of foxtail millet (Setaria italica(L.) Beauv.) cultivation about 4200-3700 BP (2900–2000 cal- BC ), and possibly up to 4900 BP (4100 cal- BC ) in Primorye (southern Russian Far East).

Eastern Siberia is located between 40° and 75° N latitude, and 100° E and 170° W longitude. In this region, approximately 60 archaeological sites chronometrically date to the period of Pleistocene–Holocene transition (38 Upper Paleolithic sites older than 10,000 BP and 22 Mesolithic sites that date between 10,000 and 8000 BP). This study examines the nature and timing of Upper Paleolithic and Mesolithic culture change during the Pleistocene–Holocene transition. Cultural changes consist of four distinctive chronoclines in the archaeological record: (1) late Pleistocene big game animals disappeared from the archaeological record gradually from one species to the next, the largest first and the smallest last; (2) blade, core, and flake tools faded with the extirpation or extinction of large gregarious Pleistocene herbivores; (3) microblade technology appeared with the exploitation of a new suite of smaller, less gregarious game animals; and (4) the development of ground stone tools occurred after subsistence strategies shifted away from big game animals. The latter pattern was most likely associated with an increase in the procurement of plant food resources available in terminal Pleistocene and early Holocene forests.

This paper critically evaluates the published radiocarbon data obtained for 46 Eastern Siberian Paleolithic sites (older than 8000 BP ). The sites were characterized by both radiocarbon age and geographic location (i.e. latitude and longitude), and rated according to the materials dated and their stratigraphic context. Multiple regression statistics (r2,P[two tails]) were used to examine the relationships between the variables age and location. The results were then illustrated as three-dimensional linear surface plots. A plot based on the bulk of14C dates (N=150) shows that the age of sites decrease in a west-to-east direction. A plot created from the highest rated14C dates (N=63) demonstrates that the principal chronocline is from south-to-north. Statistically, both stratigraphic context and sample composition affect the accuracy of radiocarbon dates. Age determinations made on charcoal collected from undisturbed sediments are the most reliable. This new approach provides a more precise view of human colonization and may be used to examine time transgressive cultural dispersion elsewhere.

Recently obtained radiocarbon accelerator mass spectrometry dates from the Gasya and Khummi sites (lower Amur River basin, the Russian Far East), on charcoal associated with pottery, fall within the interval 10345 ± 110 to 13260 ± 100 radiocarbon yr BP. Now both Russian Far East and southern Japanese Islands present evidence of the earliest pottery-making technology in the world starting about 13 000 BP.