Summary
Several major migrations and population turnover events during the later Stone Age (after c. 11,000 cal. BP) are believed to have shaped the contemporary population genetic diversity in Eurasia. While the genetic impacts of these migrations have been investigated on regional scales, a detailed understanding of their spatiotemporal dynamics both within and between major geographic regions across Northern Eurasia remains largely elusive. Here, we present the largest shotgun-sequenced genomic dataset from the Stone Age to date, representing 317 primarily Mesolithic and Neolithic individuals from across Eurasia, with associated radiocarbon dates, stable isotope data, and pollen records. Using recent advances, we imputed >1,600 ancient genomes to obtain accurate diploid genotypes, enabling previously unachievable fine-grained population structure inferences. We show that 1) Eurasian Mesolitic hunter-gatherers were more genetically diverse than previously known, and deeply divergent between the west and the east; 2) Hitherto genetically undescribed hunter-gatherers from the Middle Don region contributed significant ancestry to the later Yamnaya steppe pastoralists; 3) The genetic impact of the transition from Mesolithic hunter-gatherers to Neolithic farmers was highly distinct, east and west of a “Great Divide” boundary zone extending from the Black Sea to the Baltic, with large-scale shifts in genetic ancestry to the west. This include an almost complete replacement of hunter-gatherers in Denmark, but no substantial shifts during the same period further to the east; 4) Within-group relatedness changes substantially during the Neolithic transition in the west, where clusters of Neolithic farmer-associated individuals show overall reduced relatedness, while genetic relatedness remains high until ~4,000 BP in the east, consistent with a much longer persistence of smaller localised hunter-gatherer groups; 5) A fast-paced second major genetic transformation beginning around 5,000 BP, with Steppe-related ancestry reaching most parts of Europe within a 1,000 years span. Local Neolithic farmers admixed with incoming pastoralists in most parts of Europe, whereas Scandinavia experienced another near-complete population replacement, with similar dramatic turnover-patterns also evident in western Siberia; 6) Extensive regional differences in the ancestry components related to these early events remain visible to this day, even within countries (research conducted using the UK Biobank resource). Neolithic farmer ancestry is highest in southern and eastern England while Steppe-related ancestry is highest in the Celtic populations of Scotland, Wales, and Cornwall. Overall, our findings show that although the Stone-Age migrations have been important in shaping contemporary genetic diversity in Eurasia, their dynamics and impact were geographically highly heterogeneous.
Introduction
It is argued that genetic diversity in contemporary western Eurasian human populations was largely shaped by three major migrations in the Stone Age: hunter-gatherers occupying the area since c. 45,000 BP; the Neolithic farmers expanding from the Middle and Near East c. 11,000 BP; and Steppe pastoralists coming out of the Pontic steppe c. 5,000 BP, signalling the final stages of the Stone Age and the beginning of the Bronze Age 1–5. However, due to a paucity of genomic data from skeletons older than 8 ka, knowledge of the population structure in the Mesolithic period and how it was formed is limited, and compromise our ability to understand the subsequent demographic transitions. Also, most ancient DNA (aDNA) studies have thus far been restricted to individuals from Europe, hampering our ability to understand the wider impact of these events. The spatiotemporal mapping of population dynamics east of Europe, including Siberia, Central- and Northern Asia during the same time period is limited. In these regions the local use of the term ‘Neolithic’ typically refers to new forms of lithic material culture, and/or the presence of ceramics6. For instance, the Neolithic cultures of the Central Asian Steppe possessed pottery, but retained a hunter-gatherer economy alongside stone blade technology similar to the preceding Mesolithic cultures7. The archaeological record testifies to a boundary, ranging from the eastern Baltic to the Black Sea, east of which hunter-gatherer societies persisted for much longer than in western Europe8. However, the possible population genomic implications of this phenomenon is not known. Another enigma in the neolithisation debate is that of Scandinavia 9. The introduction of farming reached a 1,000-year standstill at the doorstep to Southern Scandinavia before finally progressing into Denmark around 6 ka. It is not known what caused this delay, and whether the transition to farming in Denmark was facilitated by the migration of people (demic diffusion), similar to the rest of Europe10–13 or mostly involved cultural diffusion14–16. Lastly, although analyses of ancient genomes have uncovered large-scale migrations from the Pontic Steppe both into Europe and Asia around 5 ka, the details of this transforming demographic process has remained largely unresolved.
To investigate these formative processes of the early Eurasian gene pools, we conducted the largest aDNA study on human Stone Age skeletal material to date. We sequenced the genomes of 317 radiocarbon-dated (AMS) primarily Mesolithic and Neolithic individuals, covering major parts of Eurasia, and combined them with published shotgun-sequenced data to impute a dataset of >1600 diploid ancient genomes. Genomic data from 100 AMS-dated individuals from Denmark supported detailed analyses of the Stone Age population dynamics in Southern Scandinavia. When combined with genetically-predicted phenotypes, proxies for diet (δ13C/δ15N), mobility (87Sr/86Sr) and vegetation cover (pollen) we could connect this with parallel shifts in phenotype, subsistence and landscape.
Results/Discussion
Samples and data
Our primary data consists of genomes from 317 ancient individuals (Fig 1, Extended data fig. 2, Supplement Table I). A total of 272 were radiocarbon dated within the project, while 39 dates were derived from literature and 15 by archaeological context. Dates were corrected for marine and freshwater reservoir effects (Supplementary Note 7) and ranged from the Upper Palaeolithic (UP) c. 25,700 calibrated years before present (cal. BP) to the mediaeval period (c. 1200 cal. BP). However, 97% of the individuals (N=309) span 11,000 cal. BP to 3,000 cal. BP, with a heavy focus on individuals associated with various Mesolithic and Neolithic cultures.
Geographically, the sampled skeletons cover a vast territory across Eurasia, from Lake Baikal to the Atlantic coast and from Scandinavia to the Middle East, deriving from contexts that include burial mounds, caves, bogs and the seafloor (Supplementary Notes 5-6). Broadly, we can divide our research area into three large regions: 1) central, western and northern Europe, 2) eastern Europe including western Russia and Ukraine, and 3) the Urals and western Siberia. Samples cover many of the key Mesolithic and Neolithic cultures in Western Eurasia, such as the Maglemose and Ertebølle in Scandinavia, the Cardial in the Mediterranean, the Körös and Linear Pottery (LBK) in SE and Central Europe, and many archaeological cultures in Ukraine, western Russia, and the trans-Ural (e.g. Veretye, Lyalovo, Volosovo, Kitoi). Our sampling was particularly dense in Denmark, from where we present a detailed and continuous sequence of 100 genomes spanning the Early Mesolithic to the Bronze Age. Dense sample sequences were also obtained from Ukraine, Western Russia, and the trans-Ural, spanning from the Early Mesolithic through the Neolithic, up to c. 5,000 BP.
Ancient DNA was extracted from dental cementum or petrous bone and the 317 genomes were shotgun sequenced to a depth of 0.01X to 7.1X (mean = 0.75X, median = 0.26X), with >1X coverage for 81 genomes. We utilised a new method optimised for low-coverage data17 to impute genotypes using the 1000 Genomes phased data as a reference panel. We also applied this to >1300 previously published shotgun-sequenced genomes (Supplemental Table VII), resulting in a dataset of 8.5 million common SNPs (>1% Minor Allele Frequency (MAF) and imputation info score > 0.5) for 1,664 imputed diploid ancient genomes (see Supplementary Note 2). To validate imputation, 42 high coverage (5X to 39X) genomes were downsampled for testing (see also Mota et al. submitted1*. While 1X genomes provided remarkably high imputation accuracy (closely matching that obtained for modern samples, Extended Data Fig. 1A-D), African genomes had lower accuracy due to poor representation of this ancestry in the reference panel. Imputation accuracy was influenced by both MAF and coverage (Supplementary Fig. S2.3). We found that coverage as low as 0.1X and 0.4X was sufficient to obtain r2 imputation accuracy of 0.8 and 0.9 at common variants (MAF≥10%), respectively. We conclude that ancient European genomes can be imputed confidently from coverages above 0.4X and highly valuable data can still be obtained from genome coverage as low as 0.1X when using specific QC on the imputed data (although at very low coverage a bias arises towards the major allele, see Supplementary Note 2). For most downstream analyses, samples with low coverage or sites with low MAF were filtered out (depending on the specific data quality requirements) yielding 1,492 imputed ancient genomes (213 sequenced in this study) after filtering individuals with very low coverages (<0.1X) and/or low imputation quality (average genotype probability < 0.8) and close relatives. Overall, this dataset allows us to characterise the ancient cross-continental gene pools and the demographic transitions with unprecedented resolution.
We conducted a broad-scale characterization of this dataset using principal component analysis (PCA) and model-based clustering (ADMIXTURE), recapitulating and providing increased resolution into previously described ancestry clines in ancient Eurasian populations (Fig. 1; Extended Data Fig. 2; Supplementary Note 3d). Strikingly, inclusion of the imputed ancient genomes in the inference of the principal components reveals much higher variance among the ancient groups than previously anticipated using projection onto a PC-space inferred from modern individuals alone (Extended Data Fig. 2). This is particularly notable in a PCA of West Eurasian individuals, where genetic variation among all present-day populations is confined within a small central area of the PCA (Extended Data Fig. 2C, D). These results are consistent with much higher genetic differentiation between ancient Europeans than present-day populations, reflecting lower effective population sizes and genetic isolation among ancient groups.
To obtain a finer-scale characterization of genetic ancestries across space and time, we assigned imputed ancient genomes to genetic clusters by applying hierarchical community detection on a network of pairwise identity-by-descent (IBD)-sharing similarities18 (Extended Data Fig. 3; Supplementary Note 3c). The obtained clusters capture fine-scale genetic structure corresponding to shared ancestry within particular spatiotemporal ranges and/or archaeological contexts, and were used as sources and/or targets in supervised ancestry modelling (Extended Data Fig. 4; Supplementary Note 3i). We focus our subsequent analyses on three panels of putative source clusters reflecting different temporal depths: “deep”, using a set of deep ancestry source groups reflecting major ancestry poles; “postNeol”, using diverse Neolithic and earlier source groups; and “postBA”, using Late Neolithic and Bronze Age source groups (Extended Data Fig. 4).
Deep population structure of western Eurasians
Our study comprises the largest genomic dataset on European hunter-gatherers to date, including 113 imputed hunter-gatherer genomes of which 79 were sequenced in this study. Among them, we report a 0.83X genome of an Upper Palaeolithic (UP) skeleton from Kotias Klde Cave in Georgia, Caucasus (NEO283), directly dated to 26,052 - 25,323 cal BP (95%). In the PCA of all non-African individuals, it occupies a position distinct from other previously sequenced UP individuals, shifted towards west Eurasians along PC1 (Supplementary Note 3d). Using admixture graph modelling, we find that this Caucasus UP lineage derives from a mixture of predominantly West Eurasian UP hunter-gatherer ancestry (76%) with ~24% contribution from a “basal Eurasian” ghost population, first observed in West Asian Neolithic individuals19 (Extended Data Fig. 5A). Models attempting to reconstruct major post-LGM clusters such as European hunter-gatherers and Anatolian farmers without contributions from this Caucasus UP lineage provided poor admixture graph fits or were rejected in qpAdm analyses (Extended Data Fig. 5B,C). These results thus suggest a central role of the descendants related to this Caucasus UP lineage in the formation of later West Eurasian populations, consistent with recent genetic data from the nearby Dzudzuana Cave, also in Georgia20.
We performed supervised admixture modelling using a set of twelve possible source clusters representing Mesolithic hunter-gatherers from the extremes of the HG cline, as well as temporal or geographical outgroups of deep Eurasian lineages (Fig 2A). We replicate previous results of broadscale genetic structure correlated to geography in European hunter-gatherers after the LGM12, while also revealing novel insights into their fine-scale structure. Ancestry related to southern European hunter-gatherers (source: Italy_15000BP_9000 BP) predominates in western Europe. This includes Denmark, where our 28 sequenced and imputed hunter-gatherer genomes derive almost exclusively from this cluster, with remarkable homogeneity across a 5,000 year transect (Fig. 3A). In contrast, hunter-gatherer individuals from the eastern and far northern reaches of Europe show the highest proportions of Russian hunter-gatherer ancestry (source: RussiaNW_11000BP_8000BP; Fig. 2B, D), with genetic continuity until ~5,000 BP in Russia. Ancestry related to Mesolithic huntergatherer populations from Ukraine (source: Ukraine_10000BP_4000BP) is carried in highest proportions in hunter-gatherers from a geographic corridor extending from south-eastern Europe towards the Baltic and southern Scandinavia. Swedish Mesolithic individuals derive up to 60% of their ancestry from that source (Fig. 2C). Our results thus indicate northwards migrations of at least three distinct waves of hunter-gatherer ancestry into Scandinavia: a predominantly southern European source into Denmark; a source related to Ukrainian and south-eastern European huntergatherers into the Baltic and southern Sweden; and a northwest Russian source into the far north, before venturing south along the Atlantic coast of Norway21 (Fig. 2). These movements are likely to represent post glacial expansions from refugia areas shared with many plant and animal species22,23.
Despite the major role of geography in shaping European hunter-gatherer structure, we also document more complex local dynamics. On the Iberian Peninsula, the earliest individuals, including a ~9,200-year-old hunter-gatherer (NEO694) from Santa Maira (eastern Spain), sequenced in this study, show predominantly southern European hunter-gatherer ancestry with a minor contribution from UP hunter-gatherer sources (Fig. 3). In contrast, later individuals from Northern Iberia are more similar to hunter-gatherers from eastern Europe, deriving ~30-40% of their ancestry from a source related to Ukrainian hunter-gatherers24,25. The earliest evidence for this gene flow is observed in a Mesolithic individual from El Mazo, Spain (NEO646) that was dated, calibrated and reservoir-corrected to c. 8,200 BP (8365-8182 cal BP, 95%) but context-dated to slightly older (8550-8330 BP, see26). The younger date coincides with some of the oldest Mesolithic geometric microliths in northern Iberia, appearing around 8,200 BP at this site26. In southern Sweden, we find higher amounts of southern European hunter-gatherer ancestry in late Mesolithic coastal individuals (NEO260 from Evensås; NEO679 from Skateholm) than in the earlier Mesolithic individuals from further inland, suggesting either geographic genetic structure in the Swedish Mesolithic population or a possible eastward expansion of hunter-gatherers from Denmark, where this ancestry prevailed (Fig. 3). An influx of southern European hunter-gatherer-related ancestry in Ukrainian individuals after the Mesolithic (Fig. 3) suggests a similar eastwards expansion in south-eastern Europe12. Interestingly, two herein reported ~7,300-year-old imputed genomes from the Middle Don River region in the Pontic-Caspian steppe (Golubaya Krinitsa, NEO113 & NEO212) derive ~20-30% of their ancestry from a source cluster of hunter-gatherers from the Caucasus (Caucasus_13000BP_10000BP) (Fig. 3). Additional lower coverage (nonimputed) genomes from the same site project in the same PCA space (Fig. 1D), shifted away from the European hunter-gatherer cline towards Iran and the Caucasus. Our results thus document genetic contact between populations from the Caucasus and the Steppe region as early as 7,300 years ago, providing documentation of continuous admixture prior to the advent of later nomadic Steppe cultures, in contrast to recent hypotheses, and also further to the west than previously reported12,27.
Major genetic transitions in Europe
Previous ancient genomics studies have documented multiple episodes of large-scale population turnover in Europe within the last 10,000 years1,2,4,8,11–13,24,28–31. The 317 genomes reported here fill important knowledge gaps, particularly in northern and eastern Europe, allowing us to track the dynamics of these events at both continental and regional scales.
Our analyses reveal profound differences in the spatiotemporal neolithisation dynamics across Europe. Supervised admixture modelling (using the “deep” set) and spatiotemporal kriging32 document a broad east-west distinction along a boundary zone running from the Black Sea to the Baltic. On the western side of this “Great Divide”, the Neolithic transition is accompanied by large-scale shifts in genetic ancestry from local hunter-gatherers to farmers with Anatolian-related ancestry (Boncuklu_10000BP; Fig. 3; Extended Data Fig. 4, 6). The arrival of Anatolian-related ancestry in different regions spans an extensive time period of over 3,000 years, from its earliest evidence in the Balkans (Lepenski Vir) at ~8,700 BP12 to c. 5,900 BP in Denmark. On the eastern side of this divide, no ancestry shifts can be observed during this period. In the East Baltic region (see also33), Ukraine and Western Russia local hunter-gatherer ancestry prevailed until ~5,000 BP without noticeable input of Anatolian-related farmer ancestry (Fig. 3; Extended Data Fig. 4, 6). This Eastern genetic continuity is in remarkable congruence with the archaeological record showing persistence of pottery-using hunter-gatherer-fisher groups in this wide region, and delayed introduction of cultivation and husbandry by several thousand years (Supplementary Note 4).
From approximately 5,000 BP, an ancestry component appears on the eastern European plains in Early Bronze Age Steppe pastoralists associated with the Yamnaya culture and it rapidly spreads across Europe through the expansion of the Corded Ware complex (CWC) and related cultures1,2. We demonstrate that this “steppe” ancestry (Steppe_5000BP_4300BP) can be modelled as a mixture of ~65% ancestry related to herein reported hunter-gatherer genomes from the Middle Don River region (MiddleDon_7500BP) and ~35% ancestry related to hunter-gatherers from Caucasus (Caucasus_13000BP_10000BP) (Extended Data Fig. 4). Thus, Middle Don hunter-gatherers, who already carried ancestry related to Caucasus hunter-gatherers (Fig. 2), serve as a hitherto unknown proximal source for the majority ancestry contribution into Yamnaya genomes. The individuals in question derive from the burial ground Golubaya Krinitsa (Supplementary Note 3). Material culture and burial practices at this site are similar to the Mariupol-type graves, which are widely found in neighbouring regions of Ukraine, for instance along the Dnepr River. They belong to the group of complex pottery-using hunter-gatherers mentioned above, but the genetic composition at Golubaya Krinitsa is different from the remaining Ukrainian sites (Fig 2A, Extended Data Fig. 4). We find that the subsequent transition of the Late Neolithic and Early Bronze Age European gene pool happened at a faster pace than during the neolithisation, reaching most parts of Europe within a ~1,000-year time period after first appearing in the eastern Baltic region ~4,800 BP (Fig. 3). In line with previous reports we observe that beginning c. 4,200 BP, steppe-related ancestry was already dominant in individuals from Britain, France and the Iberian peninsula.13,28,34. Strikingly, because of the delayed neolithisation in Southern Scandinavia these dynamics resulted in two episodes of large-scale genetic turnover in Denmark and southern Sweden within roughly a 1,000-year period (Fig. 3).
We next investigated fine-grained ancestry dynamics underlying these transitions. We replicate previous reports2,5,11–13,31,35 of widespread, but low-level admixture between immigrant early European farmers and local hunter-gatherers resulting in a resurgence of HG ancestry in many regions of Europe during subsequent centuries (Extended Data Fig. 7). The resulting estimated hunter-gatherer ancestry proportions rarely exceed 10%, with notable exceptions observed in individuals from south-eastern Europe (Iron Gates), Sweden (Pitted Ware Culture) as well as herein reported Early Neolithic genomes from Portugal (western Cardial), estimated to harbour 27% – 43% Iberian hunter-gatherer ancestry (Iberia_9000BP_7000BP). The latter result, suggesting extensive first-contact admixture, is in agreement with archaeological inferences derived from modelling the spread of farming along west Mediterranean Europe36. Neolithic individuals from Denmark show some of the highest overall hunter-gatherer ancestry proportions (up to ~25%), mostly derived from Western European-related hunter-gatherers (EuropeW_13500BP_8000BP) supplemented with marginal contribution from local Danish hunter-gatherer groups in some individuals (Extended Data Fig. 7D; Supplementary Note 3f). We estimated the timing of the admixture using the linkage-disequilibrium-based method DATES37 at ~6,000 BP. Both lines of evidence thus suggest that a significant part of the hunter-gatherer admixture observed in Danish individuals occurred already before the arrival of the incoming farmers in the region (Extended Data Fig. 7), and further point towards Central Europe as a key region in the resurgence of HG ancestry. Interestingly, the genomes of two ~5,000-year-old male individuals (NEO33, NEO898) from Denmark were entirely composed of Swedish hunter-gatherer ancestry, and formed a cluster with Pitted Ware Culture (PWC) individuals from Ajvide on the Baltic island of Gotland (Sweden)38–40. Of the two individuals, NEO033 also displays an outlier Sr-signature (Fig. 4), potentially suggesting a non-local origin matching his unusual ancestry. Overall, our results demonstrate direct contact across the Kattegat and Öresund during Neolithic times (Extended Data Fig. 3, 4), in line with archaeological findings from Zealand (east Denmark) showing cultural affinities to PWC on the Swedish west coast41–44.
Further, we find evidence for regional stratification in early Neolithic farmer ancestries in subsequent Neolithic groups. Specifically, southern European early farmers appear to have provided major genetic ancestry to Neolithic groups of later dates in Western Europe, while central European early farmer ancestry is mainly observed in subsequent Neolithic groups in eastern Europe and Scandinavia (Extended Data Fig. 7D-F). These results are consistent with distinct migratory routes of expanding farmer populations as previously suggested45. For example, similarities in material culture and flint mining activities could suggest that the first farmers in South Scandinavia originated from or had close social relations with the central European Michelsberg Culture46.
The second continental-wide transition from Neolithic farmer ancestry to Steppe-related ancestry differs markedly between geographic regions. The contribution of local farmer ancestry to the incoming groups was high in eastern, western and southern Europe, reaching >50% on the Iberian Peninsula (“postNeol” set; Extended Data Fig. 4, 6B, C)24. Scandinavia, however, portrays a dramatically different picture, with a near-complete replacement of the local population inferred across all sampled individuals (Extended Data Fig. 7B, C). Following the second transition, farmer-related ancestry remains in Scandinavia, but the source is now different. It can be modelled as deriving almost exclusively from a genetic cluster associated with the Late Neolithic Globular Amphora Culture (GAC) (Poland_5000BP_4700BP; Extended Data Fig. 4). Strikingly, after the Steppe-related ancestry was first introduced into Europe (Steppe_5000BP_4300BP), it expanded together with GAC-related ancestry across all sampled European regions (Extended Data Fig. 7I). This suggests that the spread of steppe-related ancestry throughout Europe was predominantly mediated through groups that were already admixed with GAC-related farmer groups of the eastern European plains. This finding has major implications for understanding the emergence of the CWC. A stylistic connection from GAC ceramics to CWC ceramics has long been suggested, including the use of amphora-shaped vessels and the development of cord decoration patterns47. Moreover, shortly prior to the emergence of the earliest CWC groups, eastern GAC and western Yamnaya groups exchanged cultural elements in the forest-steppe transition zone northwest of the Black Sea, where GAC ceramic amphorae and flint axes were included in Yamnaya burials, and the typical Yamnaya use of ochre was included in GAC burials48, indicating close interaction between these groups. Previous ancient genomic data from a few individuals suggested that this was limited to cultural influences and not population admixture49. However, in the light of our new genetic evidence it appears that this zone, and possibly other similar zones of contact between GAC and Yamnaya (or other closely-related steppe/forest-steppe groups) were key in the formation of the CWC through which steppe-related ancestry and GAC-related ancestry co-dispersed far towards the west and the northcf. 50. This resulted in regionally diverse situations of interaction and admixture51,52 but a significant part of the CWC dispersal happened through corridors of cultural and demic transmission which had been established by the GAC during the preceding period53,54.
Fine-scale structure and multiproxy analysis of Danish transect
We present a detailed and continuous sequence of multiproxy data from Denmark, from the Early Mesolithic Maglemose, via the Kongemose and Late Mesolithic Ertebølle epochs, the Early and Middle Neolithic Funnel Beaker Culture and the Single Grave Culture, to Late Neolithic and Bronze Age individuals (Fig. 4). To integrate multiproxy data from as many skeletons as possible we made use of non-imputed data for the admixture analyses (Supplementary Note S3d) which were not restricted to the >0.1X coverage cut-off used elsewhere. This provided genetic profiles from 100 Danish individuals (Fig. 4), spanning c. 7,300 years from the earliest known skeleton in Denmark (the Mesolithic “Koelbjerg Man” (NEO254, 10,648-10,282 cal. BP, 95% probability interval) and formerly known as the “Koelbjerg Woman”55), to a Bronze Age skeleton from Hove Å (NEO946) dated to 3322-2967 cal. BP (95%). Two shifts in genomic admixture proportions confirm the major population genetic turnovers (Fig. 4) that was inferred from imputed data (Fig. 3). The multiproxy evidence, however, unveils the dramatic concomitant changes in all investigated phenotypic, environmental and dietary parameters (Fig. 4).
During the Danish Mesolithic, individuals from the Maglemose, Kongemose and Ertebølle cultures displayed a remarkable genetic homogeneity across a 5,000 year transect deriving their ancestry almost exclusively from a southern European source (source: Italy_15000BP_9000BP) that later predominates in western Europe (Fig. 2). These cultural transitions occurred in genetic continuity, apparent in both autosomal and uniparental markers, which rules out demic diffusion and supports the long-held assumption of a continuum of culture and population56–58. Genetic predictions (see Supplementary Note 3i) indicate blue eye pigmentation with high probability in several individuals throughout the duration of the Mesolithic, consistent with previous findings 1,13,35. In contrast, none of the analysed Mesolithic individuals display high probability of light hair pigmentation. Height predictions for Mesolithic individuals generally suggest slightly lower and/or less variable stature than in the succeeding Neolithic period. However, we caution that the relatively large genetic distance to modern individuals included in the GWAS panel produces scores that are poorly applicable to Mesolithic individuals and are dependent on the choice of GWAS filters used (see Irving-Pease et al. submitted2*). Unfortunately, only a fraction of the 100 Danish skeletons included were suitable for stature estimation by actual measurement, why these values are not reported.
Stable isotope δ13C values in collagen inform on the proportion of marine versus terrestrial protein, while δ15N values reflect the trophic level of protein sources59,60. Both the Koelbjerg Man and the second earliest human known from Denmark, (Tømmerupgårds Mose – not part of the present study; see61) showed more depleted dietary isotopic values, representing a lifestyle of inland hunterfisher-gatherers of the Early Mesolithic forest. A second group consisted of coastal fisher-hunter-gatherers dating to the late half of the Maglemose epoch onwards (Supplementary Figs. S9.1 and S9.2). During this period global sea-level rise gradually changed the landscape of present-day Denmark from an interior part of the European continent to an archipelago, where all human groups had ample access to coastal resources within their annual territories. Increased δ13C and δ15N values imply that from the late Maglemose marine foods gradually increased in importance, to form the major supply of proteins in the Ertebølle period61,cf. 62. Interestingly, broadly consistent 87Sr/86Sr isotope ratios throughout the Mesolithic indicate limited longer-range mobility, in agreement with the evidence for genetic continuity reported here and modelled in previous work63,64Fig.3, and/or dietary sources from homogeneous environments.
The arrival of Anatolian farmer-related ancestry at c. 5,900 BP in Denmark resulted in a population replacement with very limited genetic contribution from the local hunter-gatherers. The shift was abrupt and brought changes in all the measured parameters. It is a clear case of demic diffusion, which settles a long-standing debate concerning the neolithisation process in Denmark9,46,65,66, at least at a broader population level. The continuing use of coastal kitchen middens well into the Neolithic67,68 remains, however, an enigma. It is unknown if these sites represent local remnants of Mesolithic groups that survived in partly acculturated form, or middens taken over by the newcomers. Concomitant shifts in both autosomal and uniparental genetic markers show that the migration by incoming farmers was not clearly sex-biased but more likely involved family units. Diet shifted abruptly to a dominance of terrestrial sources evidenced by δ13C values around −20 ‰ and δ15N values around 10 ‰ in line with archaeological evidence that domesticated crops and animals were now providing the main supply of proteins (Supplementary Note 9). Isotope values remained stable at these levels throughout the following periods, although with somewhat greater variation after c. 4,500 BP. However, five Neolithic and Early Bronze Age individuals have δ13C and δ15N values indicating intake of high trophic marine food. This is most pronouncedly seen for NEO898 (Svinninge Vejle) who was one of the two aforementioned Danish Neolithic individuals displaying typical Swedish hunter-gatherer ancestry. A higher variability in 87Sr/86Sr values can be seen with the start of the Neolithic and this continues in the later periods, which suggests that the Neolithic farmers in Denmark consumed food from more diverse landscapes and/or they were more mobile than the preceding hunter-gatherers (Supplementary Note 10). The Neolithic transition also marks a considerable rise in frequency of major effect alleles associated with light hair pigmentation69, whereas polygenic score predictions for height are generally low throughout the first millennium of the Neolithic (Funnel Beaker epoch), echoing previous findings based on a smaller set of individuals35,70.
We do not know how the Mesolithic Ertebølle population disappeared. Some may have been isolated in small geographical pockets of brief existence and/or adapted to a Neolithic lifestyle but without contributing much genetic ancestry to subsequent generations. The most recent individual in our Danish dataset with Mesolithic WHG ancestry is “Dragsholm Man” (NEO962), dated to 5,947-5,664 cal. BP (95%) and archaeologically assigned to the Neolithic Funnel Beaker farming culture based on his grave goods71,72. Our data confirms a typical Neolithic diet matching the cultural affinity but contrasting his WHG ancestry. Thus, Dragsholm Man represents a local person of Mesolithic ancestry who lived in the short Mesolithic-Neolithic transition period and adopted a Neolithic culture and diet. A similar case of very late Mesolithic WHG ancestry in Denmark was observed when analysing human DNA obtained from a piece of chewed birch pitch dated to 5,858–5,661 cal. BP (95%)73.
The earliest example of Anatolian Neolithic ancestry in our Danish dataset is observed in a bog skeleton of a female from Viksø Mose (NEO601) dated to 5,896-5,718 cal. BP (95%) (and hence potentially contemporaneous with Dragsholm Man) whereas the most recent Danish individual showing Anatolian ancestry without any Steppe-related ancestry is NEO943 from Stenderup Hage, dated to 4,818-4,415 cal. BP (95%). Using Bayesian modelling we estimate the duration between the first appearance of farmer ancestry to the first appearance of Steppe-related ancestry in Denmark to be between 876 and 1100 years (95% probability interval, Supplementary Note 8) indicating that the typical Neolithic ancestry was dominant for less than 50 generations. The arrival of steppe-related ancestry signals the rise of the Corded Ware derived cultures in Denmark (i.e. Single Grave Culture), followed by the Dagger epoch and the Bronze Age cultures. While this introduced a major new component in the Danish gene pool, it was not accompanied by apparent shifts in diet. Our complex trait predictions indicate an increase in “genetic height”, which is consistent with Steppe individuals (e.g., Yamnaya) being genetically taller on average35; c.f. 70,74. These major population turnovers were accompanied by significant environmental changes, as apparent from the high-resolution pollen diagram from Lake Højby in Northwest Zealand reconstructed using the Landscape Reconstruction Algorithm (LRA75, Supplementary Note 11). While the LRA has previously been applied at low temporal resolution regional scale e.g. 76,77, and at local scale to Iron Age and later pollen diagrams e.g. 78,79, this is the first time this quantitative method is applied at local scale to a pollen record spanning the Mesolithic and Neolithic periods in Denmark. Comparison with existing pollen records show that the land cover changes demonstrated here reflect the general vegetation development in eastern Denmark, while the vegetation on the sandier soils of western Jutland stayed more open throughout the sequence (Supplementary Note 11). We find that during the Mesolithic (i.e. before c. 6,000 BP) the vegetation was dominated by primary forest trees (Tilia, Ulmus, Quercus, Fraxinus, Alnus etc.). The forest composition changed towards more secondary, early successional trees (Betula and then Corylus) in the earliest Neolithic, but only a minor change in the relationship between forest and open land is recorded. From c. 5,650 BP deforestation intensified, resulting in a very open grassland-dominated landscape. This open phase was short-lived, and secondary forest expanded from 5,500 to 5,000 BP, until another episode of forest clearance gave rise to an open landscape during the last part of the Funnel Beaker epoch. We thus conclude that the agricultural practice was characterised by repeated clearing of the forest with fire, followed by regrowth. This strategy changed with the onset of the Single Grave Culture, when the forest increased again, but this time dominated by primary forest trees, especially Tilia and Ulmus. This reflects the development of a more permanent division of the landscape into open grazing areas and forests. In contrast, in western Jutland this phase was characterised by large-scale opening of the landscape, presumably as a result of human impact aimed at creating pastureland80.
Finally, we investigated the fine-scale genetic structure in southern Scandinavia after the introduction of Steppe-related ancestry using a temporal transect of 38 Late Neolithic and Early Bronze Age Danish and southern Swedish individuals. Although the overall population genomic signatures suggest genetic stability, patterns of pairwise IBD-sharing and Y-chromosome haplogroup distributions indicate at least three distinct ancestry phases during a ~1,000-year time span: i) An early stage between ~4,600 BP and 4,300 BP, where Scandinavians cluster with early CWC individuals from Eastern Europe, rich in Steppe-related ancestry and males with an R1a Y-chromosomal haplotype (Extended Data Fig. 8A, B); ii) an intermediate stage where they cluster with central and western Europeans dominated by males with distinct sub-lineages of R1b-L51 (Extended Data Fig. 8C, D; Supplementary Note 3b) and includes Danish individuals from Borreby (NEO735, 737) and Madesø (NEO752) with distinct cranial features (Supplementary Note 5); and iii) a final stage from c. 4,000 BP onwards, where a distinct cluster of Scandinavian individuals dominated by males with I1 Y-haplogroups appears (Extended Data Fig. 8E). Using individuals associated with this cluster (Scandinavia_4000BP_3000BP) as sources in supervised ancestry modelling (see “postBA”, Extended Data Fig. 4), we find that it forms the predominant source for later Iron and Viking Age Scandinavians, as well as ancient European groups outside Scandinavia who have a documented Scandinavian or Germanic association (e.g., Anglo-Saxons, Goths; Extended Data Fig. 4). Y-chromosome haplogroup I1 is one of the dominant haplogroups in present-day Scandinavians, and we document its earliest occurrence in a ~4,000-year-old individual from Falköping in southern Sweden (NEO220). The rapid expansion of this haplogroup and associated genome-wide ancestry in the early Nordic Bronze Age indicates a considerable reproductive advantage of individuals associated with this cluster over the preceding groups across large parts of Scandinavia.
Hunter-gatherer resilience east of the Urals
In contrast to the significant number of ancient hunter-gatherer genomes from western Eurasia studied to date, genomic data from hunter-gatherers east of the Urals remain sparse. These regions are characterised by an early introduction of pottery from areas further east and were inhabited by complex hunter-gatherer-fisher societies with permanent and sometimes fortified settlements81 (Supplementary Note 4).
Here, we substantially expand the knowledge on ancient Stone Age populations of this region by reporting new genomic data from 38 individuals, 28 of which date to pottery-associated huntergatherer contexts e.g. 82 between 8,300-5,000 BP (Supplementary Table II).The majority of these genomes form a previously only sparsely sampled37,83 “Neolithic Steppe” cline spanning the Siberian Forest Steppe zones of the Irtysh, Ishim, Ob, and Yenisei River basins to the Lake Baikal region (Fig. 1C; Extended Data Fig. 2A, 3E). Supervised admixture modelling (using the “deep” set of ancestry sources) revealed contributions from three major sources in these hunter gatherers from east of Urals: early West Siberian hunter-gatherer ancestry (SteppeC_8300BP_7000BP) dominated in the western Forest Steppe; Northeast Asian hunter-gatherer ancestry (Amur_7500BP) was highest at Lake Baikal; and Paleosiberian ancestry (SiberiaNE_9800BP) was observed in a cline of decreasing proportions from northern Lake Baikal westwards across the Forest Steppe83 (Extended Data Figs. 4, 9).
We used these Neolithic hunter-gatherer clusters (“postNeol” ancestry source set, Extended Data Fig. 4) as putative source groups in more proximal admixture modelling to investigate the spatiotemporal dynamics of ancestry compositions across the Steppe and Lake Baikal after the Neolithic period. We replicate previously reported evidence for a genetic shift towards higher Forest Steppe hunter-gatherer ancestry (SteppeCE_7000BP_3600BP) in Late Neolithic and Early Bronze Age individuals (LNBA) at Lake Baikal83,84. However, ancestry related to this cluster is already observed at ~7,000 BP in herein-reported Neolithic hunter-gatherer individuals both at Lake Baikal (NEO199, NEO200), and along the Angara river to the north (NEO843). Both male individuals at Lake Baikal belonged to Y-chromosome haplogroup Q1, characteristic of the later LNBA groups in the same region. (Extended Data Fig. 3, 6A). Together with an estimated date of admixture of ~6,000 BP for the LNBA groups, these results suggest gene flow between huntergatherers of Lake Baikal and the south Siberian forest steppe regions already during the Early Neolithic. This is consistent with archaeological interpretations of contact. In this region, bifacially flaked tools first appeared near Baikal85 from where the technique spread far to the west. We find its reminiscences in Late Neolithic archaeological complexes (Shiderty 3, Borly, Sharbakty 1, Ust-Narym, etc.) in Northern and Eastern Kazakhstan, around 6,500-6,000 BP86,87. Our herein-reported genomes also shed light on the genetic origins of the Early Bronze Age Okunevo culture in the Minusinsk Basin in Southern Siberia. In contrast to previous results, we find no evidence for Lake Baikal hunter-gatherer ancestry in the Okunevo83,84, suggesting that they instead originate from a three-way mixture of two different genetic clusters of Siberian forest steppe hunter-gatherers and Steppe-related ancestry (Extended Data Fig. 4D). We date the admixture with Steppe-related ancestry to ~4,600 BP, consistent with gene flow from peoples of the Afanasievo culture that existed near Altai and Minusinsk Basin during the early eastwards’ expansion of Yamnaya-related groups1,84.
From around 3,700 BP, individuals across the Steppe and Lake Baikal regions display markedly different ancestry profiles (Fig. 3; Extended Data Fig. 4D, 9). We document a sharp increase in non-local ancestries, with only limited ancestry contributions from local hunter-gatherers. The early stages of this transition are characterised by influx of Yamnaya-related ancestry, which decays over time from its peak of ~70% in the earliest individuals. Similar to the dynamics in western Eurasia, Yamnaya-related ancestry is here correlated with GAC-related farmer ancestry (Poland_5000BP_4700BP; Extended Data Fig. 9G), recapitulating the previously documented eastward expansion of admixed Western Steppe pastoralists from the Sintashta and Andronovo complexes during the Bronze Age1,37,88. However, GAC-related ancestry is notably absent in individuals of the Okunevo culture, providing further support for two distinct eastward migrations of Western Steppe pastoralists during the early (Yamnaya) and later (Sintashta, Andronovo) Bronze Age. The later stages of the transition are characterised by increasing Central Asian (Turkmenistan_7000 BP_5000BP) and Northeast Asian-related (Amur_7500BP) ancestry components (Extended Data Fig. 9G). Together, these results show that deeply structured huntergatherer ancestry dominated the eastern Eurasian Steppe substantially longer than in western Eurasia, before successive waves of population expansions swept across the Steppe within the last 4,000 years, including a large-scale introduction of domesticated horse lineages concomitant with new equestrian equipment and spoke-wheeled chariotry1,37,88,89.
Genetic legacy of Stone Age Europeans
We inferred distributions of Stone Age and Early Bronze Age ancestry components across modern populations by performing chromosome ‘painting’90 on UK Biobank individuals, using a pipeline adapted from GLOBETROTTER91(Supplementary Note 3h), and admixture proportions estimated by Non-Negative Least Squares. We selected a total of 24,511 individuals of typical ancestral backgrounds from 126 countries from the UK Biobank using a custom pipeline (Supplementary Note 3g).
The various hunter-gatherer ancestries are not homogeneously distributed amongst modern populations (Fig. 5). WHG-related ancestry is highest in present-day individuals from the Baltic States, Belarus, Poland, and Russia; EHG-related ancestry is highest in Mongolia, Finland, Estonia and Central Asia; and CHG-related ancestry is maximised in countries east of the Caucasus, in Pakistan, India, Afghanistan and Iran, in accordance with previous results92. The CHG-related ancestry likely reflects both Caucasus hunter-gatherer and Iranian Neolithic signals, explaining the relatively high levels in south Asia93. Consistent with expectations94,95, Neolithic Anatolian-related farmer ancestry is concentrated around the Mediterranean basin, with high levels in southern Europe, the Near East, and North Africa, including the Horn of Africa, but is less frequent in Northern Europe. This is in direct contrast to the Steppe-related ancestry, which is found in high levels in northern Europe, peaking in Ireland, Iceland, Norway, and Sweden, but decreases further south. There is also evidence for its spread into southern Asia. Overall, these results refine global patterns of spatial distributions of ancient ancestries amongst modern populations.
The availability of a large number of modern genomes (n=408,884) from self-identified “white” British individuals who share similar PCA backgrounds96 allowed us to further examine the distribution of ancient ancestries at high resolution in Britain (Supplementary Note 3h). Although regional ancestry distributions differ by only a few percent, we find clear evidence of geographical heterogeneity across the United Kingdom as visualised by assigning individuals to their birth county and averaging ancestry proportions per county (Fig. 5, inset boxes). The proportion of Neolithic farmer ancestry is highest in southern and eastern England today and lower in Scotland, Wales, and Cornwall. Steppe-related ancestry is inversely distributed, peaking in the Outer Hebrides and Ireland, a pattern only previously described for Scotland97. This regional pattern was already evident in the Pre-Roman Iron Age and persists to the present day even though immigrating Anglo-Saxons had relatively less Neolithic farmer ancestry than the Iron-Age population of southwest Briton (Extended Data Fig. 4). Although this Neolithic farmer/steppe-related dichotomy mirrors the modern ‘Anglo-Saxon’/‘Celtic’ ethnic divide, its origins are older, resulting from continuous migration from a continental population relatively enhanced in Neolithic farmer ancestry, starting as early as the Late Bronze Age98. By measuring haplotypes from these ancestries in modern individuals, we are able to show that these patterns differentiate Wales and Cornwall as well as Scotland from England. We also found higher levels of WHG-related ancestry in central and Northern England. These results demonstrate clear ancestry differences within an ‘ethnic group’ (white British) traditionally considered relatively homogenous, which highlights the need to account for subtle population structure when using resources such as the UK Biobank genomes.
Sociocultural insights
We used patterns of pairwise IBD sharing between individuals and runs of homozygosity (ROH) within individuals (measured as the fraction of the genome within a run of homozygosity f(ROH)) to examine our data for temporal shifts in relatedness within genetic clusters. Both measures show clear trends of a reduction of within-cluster relatedness over time, in both western and eastern Eurasia (Fig. 6). This pattern is consistent with a scenario of increasing effective population sizes during this period99. Nevertheless, we observe notable differences in temporal relatedness patterns between western and eastern Eurasia, mirroring the wider difference in population dynamics discussed above. In the west, within-group relatedness changes substantially during the Neolithic transition (~9,000 to ~6,000 BP), where clusters of Anatolian farmer-associated individuals show overall reduced IBD sharing and f(ROH) compared to clusters of HG-associated individuals (Fig. 6A,C). In the east, genetic relatedness remains high until ~4,000 BP, consistent with a much longer persistence of smaller localised hunter-gatherer groups (Fig. 6B,D).
Next, we examined the data for evidence of recent parental relatedness, by identifying individuals harbouring > 50cM of their genomes in long (>20cM) ROH segments100. We only detect 39 such individuals out of a total sample of 1,540 imputed ancient genomes (Fig. 6E), in line with recent results indicating that close kin mating was not common in human prehistory31,92,100,101. With the exception of eight ancient American individuals from the San Nicolas Islands in California102, no obviously discernible spatiotemporal or cultural clustering was observed among the individuals with recent parental relatedness. Interestingly, an ~1,700-year-old Sarmatian individual from Temyaysovo (tem003)103 was found homozygous for almost the entirety of chromosome 2, but without evidence of ROHs elsewhere in the genome, suggesting an ancient case of uniparental disomy. Among several noteworthy familial relationships (see Supplementary Fig. S3c.2), we report a Mesolithic father/son burial at Ertebølle (NEO568/NEO569), as well as a Mesolithic mother/daughter burial at Dragsholm (NEO732/NEO733).
Conclusion
The large-scale nature of our genomic study allows for fundamental insights into the population dynamics of Stone Age Eurasia in unprecedented detail. We demonstrate the existence of a clear east-west genetic division extending from the Black Sea to the Baltic, mirroring observations from Stone Age material culture, and persisting over several millennia. By imputing >1600 ancient genomes we recover the full scale of ancient genetic diversity in prehistoric Eurasia and show that previous estimates have been severely underestimated. We show that the genetic impact of the Neolithic transition was highly distinct east and west of this boundary. We identify a hitherto unknown source of ancestry in hunter-gatherers from the Middle Don region contributing ancestry to the Yamnaya pastoralists, and we document how the later spread of steppe-related ancestry into Europe was very rapid and mediated through people showing admixture related to the Anatolian farmer-associated Globular Amphora Culture. Finally, we report two near-complete population replacements in Denmark within just 1,000 years, concomitantly with major changes in material culture, ruling out cultural diffusion as a main driver and settling generation-long archaeological debates. Our study reveals that the devil lies in the detail: although the migrations during the Stone Age were crucial in shaping contemporary Eurasian genetic diversity as argued in recent years 104–106, their mode and impact varied extensively across time and space.
Extended Data Figures
Data availability
All collapsed and paired-end sequence data for novel samples sequenced in this study will be made publicly available on the European Nucleotide Archive, together with trimmed sequence alignment map files, aligned using human build GRCh37. Previously published ancient genomic data used in this study is detailed in Supplementary Table VII, and are all already publicly available. Bioarchaeological data (including Accelerator Mass Spectrometry results) are included in the online supplementary materials of this submission.
Code availability
All analyses relied upon available software which has been fully referenced in the manuscript and detailed in the relevant supplementary notes.
Author Information
These authors contributed equally: Morten E. Allentoft, Martin Sikora, Alba Refoyo-Martínez, Evan K. Irving-Pease, Anders Fischer, William Barrie & Andrés Ingason
These authors equally supervised research: Thorfinn Korneliussen, Richard Durbin, Rasmus Nielsen, Olivier Delaneau, Thomas Werge, Fernando Racimo, Kristian Kristiansen & Eske Willerslev
Affiliations
Lundbeck Foundation GeoGenetics Centre, Globe Institute, University of Copenhagen, Copenhagen, Denmark
Morten E. Allentoft, Martin Sikora, Alba Refoyo-Martínez, Evan K. Irving-Pease, Andrés Ingason, Jesper Stenderup, Fabrice Demeter, Maria Novosolov, Ruairidh Macleod, Rasmus A. Henriksen, Tharsika Vimala, Hugh McColl, Lasse Vinner, Gabriele Scorrano, Abigail Ramsøe, Anders Rosengren, Anthony Ruter, Anne Birgitte Gotfredsen, Charleen Gaunitz, Fulya Eylem Yediay, Isin Altinkaya, Lei Zhao, Peter de Barros Damgaard, Kurt H. Kjær, Thorfinn Korneliussen, Rasmus Nielsen, Thomas Werge, Fernando Racimo, Kristian Kristiansen & Eske Willerslev
Trace and Environmental DNA (TrEnD) Laboratory, School of Molecular and Life Sciences, Curtin University, Perth, Australia
Morten E. Allentoft
GeoGenetics Group, Department of Zoology, University of Cambridge, Cambridge, UK
William Barrie, Alice Pearson, Ruairidh Macleod & Eske Willerslev
Department of Historical Studies, University of Gothenburg, Gothenburg, Sweden
Anders Fischer, Karl-Göran Sjögren, Bettina Schulz Paulsson, Malou Blank & Kristian Kristiansen
Institute of Biological Psychiatry, Mental Health Services, Copenhagen University Hospital, Roskilde, Denmark
Andrés Ingason, Vivek Appadurai,
The National Museum of Denmark, Ny Vestergade 10, Copenhagen, Denmark
Lasse Sørensen, Poul Otto Nielsen, Morten Fischer Mortensen, Peter Rasmussen, Peter Vang Petersen,
Sealand Archaeology, Gl. Roesnaesvej 27, 4400 Kalundborg, Denmark
Anders Fischer
Department of Genetics, University of Cambridge, Cambridge, UK
Alice Pearson & Richard Durbin
Wellcome Sanger Institute, Wellcome Genome Campus, Cambridge, UK
Richard Durbin & Eske Willerslev
Department of Computational Biology, University of Lausanne, Switzerland
Bárbara Sousa da Mota & Olivier Delaneau
Swiss Institute of Bioinformatics, University of Lausanne, Switzerland
Bárbara Sousa da Mota
Department of Integrative Biology, University of California, Berkeley, USA
Alma S. Halgren, Peter H. Sudmant & Rasmus Nielsen
Center for Computational Biology, University of California, Berkeley, USA
Andrew Vaughn, Aaron J. Stern, Peter H. Sudmant & Rasmus Nielsen
Laboratory of Biological Anthropology, Department of Forensic Medicine, University of Copenhagen, Copenhagen, Denmark
Marie Louise Schjellerup Jørkov & Niels Lynnerup
Muséum National d’Histoire Naturelle, Paris, France
Fabrice Demeter
Research department of Genetics, Evolution and Environment, University College London, London, UK
Ruairidh Macleod
Section for Evolutionary Genomics, GLOBE Institute, University of Copenhagen, Copenhagen, Denmark
Ashot Margaryan, Theis Zetner Trolle Jensen, Hannes Schroeder, Enrico Cappellini
Centre for Evolutionary Hologenomics, University of Copenhagen, Copenhagen, Denmark
Ashot Margaryan
Anthropology Department, University of Utah, USA
Melissa Illardo
Department of Geology, Lund University, Lund, Sweden
Anne Birgitte Nielsen
Department of Archaeology and Ancient History, Lund University, Lund, Sweden
Lars Larsson
Tårnby Gymnasium og HF, Kastrup, Denmark
Mikkel Ulfeldt Hede
Department of Health Technology, Section of Bioinformatics, Technical University of Denmark, Kongens Lyngby, Denmark
Gabriel Renaud
Department of Archaeology and Heritage Studies, Aarhus University, Aarhus, Denmark
Niels N. Johannsen, Rikke Maring
Institute of Biological Psychiatry, Mental Health Centre Sct Hans, Copenhagen University Hospital, Denmark
Andrew J. Schork, Anders Rosengren & Thomas Werge
Department of Clinical Medicine, University of Copenhagen, 2200 Copenhagen N, Denmark
Thomas Werge
Neurogenomics Division, The Translational Genomics Research Institute (TGEN), Phoenix, AZ, USA
Andrew J. Schork
Terra Ltd., Letchik Zlobin St. 20, Voronezh, 394055, Russian Federation
Andrey Skorobogatov, Ruslan Turin
Department of Archaeology, University of Exeter, Exeter, UK
Alan K. Outram & Catriona J. McKenzie
Institute of Archaeology and Ethnography, Siberian Branch of the Russian Academy of Sciences, Novosibirsk, Russian Federation
Aleksey A. Timoshchenko, Dmitri V. Pozdnyakov, Liudmila N. Mylnikova, Marina S. Nesterova & Vyacheslav I. Molodin
Institute of Ethnology and Anthropology, Russian Academy of Sciences, Moscow, Russian Federation
Elizaveta V. Veselovskaya, Sergey Vasilyev
Centre for Egyptological studies, Russian Academy of Sciences, Moscow, Russia
Sergey Vasilyev
Research Institute and Museum of Anthropology, Lomonosov Moscow State University, Mokhovaya str. 11, Moscow, Russian Federation
Alexandra Buzhilova, Natalia Berezina, Svetlana Borutskaya
Department of Environmental Biology, Sapienza University of Rome, Rome, Italy
Alfredo Coppa, Dušan Borić
Peter the Great Museum of Anthropology and Ethnography (Kunstkamera), Russian Academy of Sciences, Saint Petersburg, Russian Federation
Alisa Zubova & Vyacheslav Moiseyev
CIAS, Department of Life Science, University of Coimbra, Coimbra, Portugal
Ana Maria Silva
UNIARQ, University of Lisbon, Lisbon, Portugal
Ana Maria Silva & João Zilhão
Kostanay Regional University A. Baitursynov, Kostanay, Kazakhstan
Andrey V. Logvin, Irina V. Shevnina
Vesthimmerlands Museum, Søndergade 44, Aars, Denmark
Bjarne Henning Nielsen
Museum Nordsjælland, Frederiksgade 9, 3400 Hillerød
Per Lotz, Søren Anker Sørensen, Thomas Jørgensen
Museum Vestsjælland, Klosterstræde 18, 4300 Holbæk, Denmark
Per Lotz
Vendsyssel Historiske Museum, DK-9800 Hjørring, Denmark
Per Lysdahl, Sidsel Wåhlin
Moesgaard Museum, Moesgård Allé 15, Højbjerg, Denmark
Søren H. Andersen
Grupo EvoAdapta, Departamento de Ciencias Históricas, Universidad de Cantabria, Santander, Spain
Borja González-Rabanal
Natural Sciences Museum of Barcelona (MCNB), Barcelona, Spain, Barcelona, Spain
Carles Lalueza-Fox
Institute of Evolutionary Biology, CSIC-Universitat Pompeu Fabra, Barcelona, Spain
Carles Lalueza-Fox
ICREA, University of Barcelona, Barcelona, Spain
João Zilhão
Departamento de Prehistoria y Arqueología Department, Universidad Autónoma de Madrid, Madrid, Spain
Concepción Blasco, Corina Liesau, Patricia Ríos
Department of Biology, University of Rome “Tor Vergata”, Rome, Italy
Cristina Martinez-Labarga, Olga Rickards
Department of History, Humanities and Society, University of Rome “Tor Vergata”, Rome, Italy
Mario Federico Rolfo
Instituto Internacional de Investigaciones Prehistóricas de Cantabria, Universidad de Cantabria, Santander, Gobierno de Cantabria, Santander, Spain
David Cuenca-Solana, Igor Gutiérrez Zugasti & Manuel González-Morales
Centre de Recherche en Archéologie, Archeosciences, Histoire (CReAAH), UMR-6566 CNRS, Rennes, France
David Cuenca-Solana
Georgian National Museum, Tbilisi, Georgia
David O. Lordkipanidze
Tbilisi State University, Tbilisi, Georgia
David O. Lordkipanidze
IPND, Tyumen Scientific Centre, Siberian Branch of the Russian Academy of Sciences, Tyumen, Russian Federation
Dmitri Enshin, Olga Poshekhonova, Svetlana N. Skochina
Departament de Prehistòria, Arqueologia i Història Antiga, Universitat de València, València, Spain
Domingo C. Salazar-García, J.Emili Aura Tortosa
Department of Geological Sciences, University of Cape Town, Cape Town, South Africa
Domingo C. Salazar-García
Department of Anthropology, New York University, New York, USA.
Dušan Borić
Institute of Humanities, Ivanovo State University, Ivanovo, Russian Federation
Elena Kostyleva
Saryarka Archaeological Institute, Buketov Karaganda University, Karaganda, Kazakhstan
Emma R. Usmanova
South Ural State University, Chelyabinsk, Russia
Emma R. Usmanova
The Saxo Institute, University of Copenhagen, Copenhagen, Denmark
Erik Brinch Petersen, Rune Iversen
Museum Østjylland, Stemannsgade 2, Randers, Denmark
Esben Kannegaard, Lutz Klassen
Soprintendenza Archeologia Belle Arti e Paesaggio per la Città Metropolitana di Bari, Via Pier l’Eremita, 25, 70122, Bari, Italy
Francesca Radina
Soprintendenza per i Beni Archeologici delle Marche, Via Birarelli 18, 60100, Ancona, Italy
Mara Silvestrini
Soprintendenza Archeologia, Belle Arti e Paesaggio per la provincia di Cosenza, Cosenza, Italy
Paola Aurino
UMR 5199 PACEA, CNRS, Université de Bordeaux, 33615 Pessac, France
Henri Duday, Patrice Courtaud
Institute of Archaeology, National Academy of Sciences of Ukraine, Kyiv, Ukraine
Inna Potekhina
National University of Kyiv-Mohyla Academy, Kyiv, Ukraine
Inna Potekhina
Collège de France, 75231 Paris cedex 05, France
Jean Guilaine
Odense City Museums, Overgade 48, Odense, Denmark
Jesper Hansen
Museum Sydøstdanmark, Algade 97, 4760 Vordingborg, Denmark
Kristoffer Buck Pedersen
Institute of Archaeology and Ethnology, Polish Academy of Sciences, Kraków, Poland
Krzysztof Tunia, Piotr Włodarczak
CNRS UMR 5608, Toulouse Jean Jaurès University, Maison de la Recherche, 5 Allées Antonio Machado, 31058 Toulouse, Cedex 9, France
Laure Metz & Ludovic Slimak
ARGEA Consultores SL, C. de San Crispín, Madrid, Spain
Jorge Vega, Roberto Menduiña
Institute of Molecular Biology, National Academy of Sciences, Yerevan, Armenia
Levon Yepiskoposyan
Russian-Armenian University, Yerevan, Armenia
Levon Yepiskoposyan
Institute of Archaeology and Ethnography, National Academy of Sciences, Yerevan, Armenia
Ruben Badalyan
HistorieUdvikler, Gl. Roesnaesvej 27, DK-4400 Kalundborg, Denmark
Lisbeth Pedersen
Department of history and cultural heritage, University of Siena, Siena, Italy
Lucia Sarti, Mauro Calattini
Centre d’Anthropobiologie et de Génomique de Toulouse, CNRS UMR 5288, Université Paul Sabatier, Toulouse, France
Ludovic Orlando
Västergötlands museum, Stadsträdgården, Skara, Sweden
Maria Vretemark
Cabinet of Anthropology, Tomsk State University, Tomsk, Russian Federation
Marina P. Rykun
Faculty of Archaeology, Adam Mickiewicz University in Poznań, Poznań, Poland
Marzena H. Szmyt
Institute of Archaeology, Jagiellonian University, Ul. Gołębia 11, 31-007, Kraków, Poland
Marcin Przybyła
Zoological Institute of Russian Academy of Sciences, Universitetskaya nab. 1, 199034, St. Petersburg, Russian Federation
Mikhail Sablin
Department of Anthropology, Czech National Museum, Prague, Czech Republic
Miluše Dobisíková
Department of Health and Nature, University of Greenland, Greenland
Morten Meldgaard
The Viking Ship Museum, Vindeboder 12, Roskilde, Denmark
Morten Johansen, Otto Christian Uldum
Archaeology Institute, University of Highlands and Islands, Scotland, UK
Nick Card
Laboratory for Experimental Traceology, Institute for the History of Material Culture of the Russian Academy of Sciences, Dvortsovaya nab., 18, 191186, St. Petersburg, Russian Federation
Olga V. Lozovskaya
Scientific Research Center “Baikal region”, Irkutsk State University; 1, K. Marx st., Irkutsk, 664003, Russian Federation
Nikolai A. Saveliev
Paleoecology Laboratory, Institute of Plant and Animal Ecology, Ural Branch of the Russian Academy of Sciences, Ekaterinburg, Russian Federation
Pavel Kosintsev
Department of History of the Institute of Humanities, Ural Federal University, Ekaterinburg, Russian Federation
Pavel Kosintsev
Centre for the Study of Early Agricultural Societies, Department of Cross-Cultural and Regional Studies, University of Copenhagen, 2300 Copenhagen, Denmark
Peder Mortensen
Museum of Cultural History, University of Oslo, P.O. Box 6762. St. Olavs Plass NO-0130 Oslo, Norway
Per Persson
ArchaeoScience, GLOBE Institute, University of Copenhagen, Copenhagen, Denmark
Pernille Bangsgaard
Department of History, University of Santiago de Compostela, Spain
Pilar Prieto Martinez
Lipetsk Regional Scientific Public Organisation “Archaeological Research”, Lipetsk, Russian Federation
Roman V. Smolyaninov
Laboratory for Archaeological Chemistry, Department of Anthropology, University of Wisconsin-Madison, Madison, USA
T. Douglas Price
Nizhny Tagil State Socio-Pedagogical Institute, Nizhny Tagil, Russia
Yuri B. Serikov
Institute for History of Medicine, First Faculty of Medicine, Charles University, Prague, Czech Republic
Vaclav Smrcka
Centre for Archaeological Research Toraighyrov University, Pavlodar, Kazakhstan
Victor Merz
Malmö Museer, Malmöhusvägen 6, Malmö, Sweden
Yvonne Magnusson
Institute of Statistical Sciences, School of Mathematics, University of Bristol, Bristol, UK
Daniel J. Lawson
Novo Nordisk Foundation Centre for Protein Research, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen N, Denmark
Simon Rasmussen
MARUM, University of Bremen, Bremen, Germany
Eske Willerslev
Contributions
M.E.A., M.S., A.R.-M., E.K.I.-P., A.F., W.B., and A.I. contributed equally to this work. M.E.A., M.S., T.S.K., R.D., R.N., O.D., T.W., F. Racimo, K.K. and E.W. led the study. M.E.A., M.S., A.F., C.L.-F., R.N., T.W., K.K. and E.W. conceptualised the study. M.E.A., M.S., H.S., L.O., T.S.K., R.D., R.N., O.D., T.W., F. Racimo, K.K. and E.W. supervised the research. M.E.A., L.O., R.D., R.N., T.W., K.K. and E.W. acquired funding for research. A.F., J.S., K.G.S., M.L.S.J., M.U.H., A.A.T., A.C., A.Z., A.M.S., A.J.H, A.G., A.V.L., B.H.N., B.G.R, C.B., C.L., C.M-L., D.V., D.C.-S., D.L., D.N., D.C.S.-G., D.B., E.K., E.V.V., E.R.U., E. Kannegaard, F. Radina, H.D., I.G.Z., I.P., I.V.S., J.G., J.H., J.E.A.T., J.Z., J.V., K.B.P., K.T., L.N., L.L., L. M., L.Y., L.P., L. Sarti, L. Slimak, L.K., M.G.M., M. Silvestrini, M.V., M.S.N., M.P.R., M.H.S., M.P., M. C., M. Sablin, N.C., O.P., O.R., O.V.L., P.A., P.K., P.C., P. Ríos, P. Lotz, P. Lysdahl, P.P., P.B., P.d.B.D., P.V.P., P.P.M., P.W., R.V.S., R. Maring, R. Menduiña, R.B., R.T., S.V., S.W., S.B., S.N.S., S.A.S., S.H.A., T.D.P., T.J., Y.B.S., V.I.M., V.S., V.M, Y.M. and N.L. were involved in sample collection. M.E.A., M.S., A.R.-M., E.K.I.-P., W.B., A.I., J.S., A.P., B.S.d.M., M.I., L.V., A.J. Stern, C.G., F.E.Y, D.J.L., T.S.K., R.D., R.N., O.D., F. Racimo, K.K. and E.W. were involved in developing and applying methodology. J.S., C.G. and L.V. led the DNA laboratory work research component. K.G.S. led bioarchaeological data curation. M.E.A., M.S., A.R.-M., E.K.I.-P., W.B., A.I., A.P., B.S.d.M., B.S.P., A.S.H., R.A.H., T.V., H.M., A.M., A.V., A.B.N., P. Rasmussen, G.R., A. Ramsøe, A.S., A.J. Schork, A. Rosengren, C.J.M., I.A., L.Z., R.Maring, V.S., V.A., P.H.S, S.R., T.S.K., O.D. and F. Racimo undertook formal analyses of data. M.E.A., M.S., A.R.-M., E.K.I.-P., A.F., W.B., A.I., K.G.S., D.J.L., P.H.S., T.S.K., and F. Racimo drafted the main text (M.E.A. and M.S. led this). M.E.A., M.S., A.R.-M., E.K.I.-P., A.F., W.B., A.I., K.G.S., A.P., B.S.d.M., B.S.P, A.S.H., R. Macleod, R.A.H., T.V., M.F.M., A.B.N., M.U.H., P. Rasmussen, A.J. Stern, N.N.J., H.S., G.S., A. Ramsøe, A.S., A. Rosengren, A.K.O., A.B., A.C., A.G., A.V.L., A.B.G., C.J.M., D.C.S.-G., E. Kostyleva, E.R.U., E. Kannegaard, I.G.Z., I.P., I.V.S., J.G., J.H., J.E.A.T., L.Z, L.Y., L.P., L.K., M.B., M.G.M., M.V., M.P.R., M.J., N.B., O.V.L., O.C.U., P.K., P. Lysdahl, P.B., P.W., R.V.S., R. Maring, R.B., R.I., S.V., S.W., S.B., S.H.A., T.J., V.S., D.J.L., P.H.S., S.R., T.S.K., O.D. and F. Racimo drafted supplementary notes and materials. M.E.A., M.S., A.R.-M., E.K.I.-P., A.F., W.B., A.I., G.G.S., A.S.H., M.L.S.J., F.D., R. Macleod, L. Sørensen, P.O.N., R.A.H., T.V., H.M., A.M., N.N.J., H.S., A. Ramsøe, A.S., A.J. Schork, A. Ruter, A.K.O., B.H.N., B.G.R., D.C.-S., D.C.S.-G., I.G.Z., I.P., J.G., J.E.A.T., L.Z., L.O., L.K., M.G.M., P.d.B.D., R.I., S.A.S., D.J.L., P.H.S., T.S.K., R.D., R.N., O.D., T.W., F. Racimo, K.K. and E.W. were involved in reviewing drafts and editing (M.E.A., A.F., K.G.S., F.D., R. Macleod, H.M. and T.V. led this). All co-authors read, commented on, and agreed upon the submitted manuscript.
Corresponding authors
Correspondence to Morten E. Allentoft (morten.allentoft{at}curtin.edu.au), Martin Sikora (martin.sikora{at}sund.ku.dk), Eske Willerslev (ew482{at}cam.ac.uk).
Ethics declarations
Competing interests
The authors declare no competing interests.
Acknowledgements
This publication, the culmination of a research effort lasting over a decade, is dedicated to the memory of Pia Bennike, who was part of a small core team that initiated the project. Sadly Pia passed away in 2017 but this study, and many others, testifies to her tremendous efforts and knowledge on Danish prehistoric skeletal material.
We are deeply indebted to former and present staff members at the National Museum, the Anthropological Laboratory, the regional museums and citizen scientists of Denmark, who for many generations carefully collected, recorded and curated the prehistoric skeletal remains that form a key component of this study. We are equally thankful to curators of the many other institutions across major parts of Eurasia, who to the benefit of following generations curated human skeletal remains and gave us access and permission to sample this precious material. We also thank all the former and current staff at the Lundbeck Foundation GeoGenetics Centre and the GeoGenetics Sequencing Core, and to colleagues across the many institutions detailed below. We are particularly grateful to Line Olsen as project manager for the Lundbeck Foundation GeoGenetics Centre project, and to Pernille Selmer Olsen for assisting with sample processing. We thank UK Biobank Ltd. for access to the UK Biobank genomic resource. We are thankful to Illumina Inc. for collaboration. We also thank Sturla Ellingvåg for assistance in sample access. EW thanks St. John’s College, Cambridge, for providing a stimulating environment of discussion and learning.
The Lundbeck Foundation GeoGenetics Centre is supported by the the Lundbeck Foundation (R302-2018-2155, R155-2013-16338), the Novo Nordisk Foundation (NNF18SA0035006), the Wellcome Trust (UNS69906), Carlsberg Foundation (CF18-0024), the Danish National Research Foundation (44113220) and the University of Copenhagen (KU2016 programme). This research has been conducted using the UK Biobank Resource and the iPSYCH Initiative, funded by the Lundbeck Foundation (R102-A9118 and R155-2014-1724). This work was further supported by the Swedish Foundation for Humanities and Social Sciences grant (Riksbankens Jubileumsfond M16-0455:1) to KK. M.E.A. was supported by Marie Skłodowska-Curie Actions of the EU (grant no. 300554), The Villum Foundation (grant no. 10120) and Independent Research Fund Denmark (grant no. 7027-00147B). W.B. is supported by the Hanne and Torkel Weis-Fogh Fund (Department of Zoology, University of Cambridge); AP is funded by Wellcome grant WT214300, B.S.d.M and O.D. by the Swiss National Science Foundation (SFNS PP00P3_176977) and European Research Council (ERC 679330); M.N. by the Human Frontier Science Program Postdoctoral Fellowship (LT000143/2019-L4); R. Macleod by an SSHRC doctoral studentship grant (G101449: ‘Individual Life Histories in Long-Term Cultural Change’); G.R. by a Novo Nordisk Foundation Fellowship (gNNF20OC0062491); N.N.J. by Aarhus University Research Foundation; B.S.P. by an ERC-Starter Grant ‘NEOSEA’ (grant no. 949424) H.S. by a Carlsberg Foundation Fellowship (CF19-0601); G.S. by Marie Skłodowska-Curie Individual Fellowship ‘PALAEO-ENEO’ (grant agreement number 751349); A.J. Schork by a Lundbeckfonden Fellowship (R335-2019-2318) and the National Institute on Aging (NIH award numbers U19AG023122, U24AG051129, and UH2AG064706); A.V.L. and I.V.S. by the Science Committee, Ministry of Education and Science of the Republic of Kazakhstan (AP08856317); B.G.R. and MGM by the Spanish Ministry of Science and Innovation (Project HAR2016-75605-R); C.M.-L. and O.R. by the Italian Ministry for the Universities (grants ‘2010-11 prot.2010EL8TXP_001 Biological and cultural heritage of the central-southern Italian population trough 30 thousand Years’ and ‘2008 prot. 2008B4J2HS_001 Origin and diffusion of farming in central-southern Italy: a molecular approach’); D.C.-S. and I.G.Z. by the Spanish Ministry of Science and Innovation (Project HAR2017-86262-P). D.C.S.G. acknowledges funding from the Generalitat Valenciana (CIDEGENT/2019/061) and the Spanish Government (EUR2020-112213); D.B. was supported by the NOMIS Foundation and Marie Skłodowska-Curie Global Fellowship ‘CUSP’ (grant no. 846856); E.R.U. by the Science Committee, Ministry of Education and Science of the Republic of Kazakhstan (AP09261083: “Transcultural Communications in the Late Bronze Age (Western Siberia - Kazakhstan - Central Asia)”); E.C. by Villum Fonden (17649); J.E.A.T. by the Spanish Ministry of Economy and Competitiveness, (HAR2013-46861-R) and Generalitat Valenciana (Aico/ 2018/125 and Aico 2020/97); P.K. by the Russian Ministry of Science and Higher Education (Ural Federal University Program of Development within the Priority-2030 Program) and acknowledges the Museum of the Institute of Plant and Animal Ecology (UB RAS, Ekaterinburg). L.Y. acknowledges funding by the Science Committee of the Armenian Ministry of Education and Science (Project 20TTSH-006, L.V. by ERC Consolidator Grant ‘PEGASUS’ (agreement no. 681605); M. Sablin by the Russian Ministry of Science and Higher Education (075-15-2021-1069); N.C. by Historic Environment Scotland; S.V. by the Russian Ministry of Science and Higher Education (075-15-2020-910); V.M. by the Science Committee, Ministry of Education and Science of the Republic of Kazakhstan (AR08856925). V.A. is supported by a Lundbeckfonden Fellowship (R335-2019-2318); P.H.S. by the National Institute of General Medical Sciences (R35GM142916); S.R. by the Novo Nordisk Foundation (NNF14CC0001); R.D. by the Wellcome Trust (WT214300); R.N. by the National Institute of General Medical Sciences (NIH grant R01GM138634); F. Racimo by a Villum Fonden Young Investigator Grant (no. 00025300). T.W. and V.A. are supported by the Lundbeck Foundation iPSYCH initiative (R248-2017-2003).
Footnotes
↵§ Joint first authors
↵@ Joint last authors
Added high-resolution Extended Data figures
↵1* B.S.d.M., Simone Rubinacci, Diana Ivette Cruz Dávalos, Carlos Eduardo G. Amorim, M.S., N.N.J., M.H.S., P.W., Anita Szczepanek, M.P., H.S., M.E.A., E.W., Anna-Sapfo Malaspinas, O.D.,
↵2* E.K.I-P, A.R-M, A.I., A.P., A.F., W.B., K.G.S, .A.S.H., R.Macleod, F.D., R.A.H, T.V., H.M., A.V., L.V., A.J. Stern, G.S., A. Ramsøe, A.J. Schork, A. Rosengren, K.K., P.H.S., D.J.L., R.D., T.S.K., T.W., M.E.A., M.S., R.N., F.R., E.W., The Selection Landscape and Genetic Legacy of Ancient Eurasians. (Submitted)
References
- 1.↵
- 2.↵
- 3.
- 4.↵
- 5.↵
- 6.↵
- 7.↵
- 8.↵
- 9.↵
- 10.↵
- 11.↵
- 12.↵
- 13.↵
- 14.↵
- 15.
- 16.↵
- 17.↵
- 18.↵
- 19.↵
- 20.↵
- 21.↵
- 22.↵
- 23.↵
- 24.↵
- 25.↵
- 26.↵
- 27.↵
- 28.↵
- 29.
- 30.
- 31.↵
- 32.↵
- 33.↵
- 34.↵
- 35.↵
- 36.↵
- 37.↵
- 38.↵
- 39.
- 40.↵
- 41.↵
- 42.
- 43.
- 44.↵
- 45.↵
- 46.↵
- 47.↵
- 48.↵
- 49.↵
- 50.↵
- 51.↵
- 52.↵
- 53.↵
- 54.↵
- 55.↵
- 56.↵
- 57.
- 58.↵
- 59.↵
- 60.↵
- 61.↵
- 62.↵
- 63.↵
- 64.↵
- 65.↵
- 66.↵
- 67.↵
- 68.↵
- 69.↵
- 70.↵
- 71.↵
- 72.↵
- 73.↵
- 74.↵
- 75.↵
- 76.↵
- 77.↵
- 78.↵
- 79.↵
- 80.↵
- 81.↵
- 82.↵
- 83.↵
- 84.↵
- 85.↵
- 86.↵
- 87.↵
- 88.↵
- 89.↵
- 90.↵
- 91.↵
- 92.↵
- 93.↵
- 94.↵
- 95.↵
- 96.↵
- 97.↵
- 98.↵
- 99.↵
- 100.↵
- 101.↵
- 102.↵
- 103.↵
- 104.↵
- 105.
- 106.↵