One modern people, the Ossetians of the Caucasus, trace their descent from the Scythians via the Alans and speak an Eastern Iranian language.1Scythians, Sarmatians, Alans : Iranian-Speaking Nomads of the Eurasian Steppes Autonomous University of Barcelona, 7-10 May 2007: abstracts. They have the highest level so far discovered of Y-DNA haplogroup G. This led some to conclude that the Scythians were high in G. Yet no G at all has been found in ancient DNA from Scythians and their presumed ancestors, who almost all carry Y-DNA R1a1a.2C. Bouakazeet al, First successful assay of Y-SNP typing by SNaPshot minisequencing onancient DNA, International Journal of Legal Medicine vol. 121 (2007), pp.493-499; C. Keyser, Ancient DNA provides new insights into the history of south Siberian Kurgan people, Human Genetics online May 16, 2009. Nor is G notably high overall in the steppe regions once home to the Scythians. Instead levels decline as one moves away from the Caucasus. Nasidze and colleagues came to the conclusion that, although the mtDNA of the Ossetians indicated an Iranian origin, their Y-DNA was the result of inter-mixture with their neighbours in the Caucasus.3I. Nasidze et al, Mitochondrial DNA and Y-Chromosome Variation in the Caucasus, Annals of Human Genetics, vol. 68, (2004), pp. 204-21; I. Nasidze et al, Genetic Evidence Concerning the Origins of South and North Ossetians, Annals of Human Genetics, vol. 68 (2004), pp. 588-99.
Anthony and Mallory refrain from arguing the case for mass migration spreading Indo-European languages. Instead they see language change triggered by Yamnaya bands establishing themselves as elites. However there are genetic clues to a much larger impact on the population than we would expect from a thinly-spread elite. The subclades of Y-DNA R1 are pre-eminent both in Europe and India. R1a1a dominates northern India and is also found strongly in Eastern Europe, particularly in Slavic and Baltic populations, while R1b1b2 dominates the rest of Europe. That is a geographical match to the distribution of Indo-European languages.4Pointed out by Richard Stevens, administrator of the Family Tree DNA R1b-L21 project.
The connection between R1a1a and the speakers of Indo-European languages was spotted in the late 1990s. From that Spencer Wells and colleagues deduced that R1a1a [M17] arose in the Pontic-Caspian steppes. 5R.S. Wells et al, The Eurasian Heartland: Acontinental perspective on Y-chromosome diversity, Proceedings of the National Academy of Sciences of the United States of America, vol. 98 no.18 (2001), pp. 10244-10249; The connection between Y-DNA R-M17 and the spread of Indo-European languages was first proposed by T. Zerjal et al, The use of Y-chromosomal DNA variation to investigate population history: recent male spread in Asia and Europe, in S.S. Papiha, R. Deka and R. Chakraborty (eds.),Genomic Diversity: applications in human population genetics(1999), pp. 91–101, and supported by L. Quintana-Murci et al., Y-Chromosome lineages trace diffusion of people and languages in Southwestern Asia, American Journal of Human Genetics vol. 68 (2001),pp.537–542. Studies of ancient DNA have followed the cultural trail eastwards from the European steppe, confirming the spread of R1a1a in a succession of cultures culminating in the Indo-European-speaking Scythians. Skeletons from Andronovo and its successor cultures carried more Western Eurasian than Eastern mtDNA,6C. Lalueza-Fox et al., Unravelling migrations in the steppe: mitochondrial DNA sequences from ancient Central Asians,Proceedings of the Royal Society of London B, vol. 271 (2004), pp.941–947. and were mainly blue (or green)-eyed, fair-skinned and light-haired people.7C. Bouakaze et al., Pigment phenotype and biogeographical ancestry from ancient skeletal remains: inferences from multiplexed autosomal SNP analysis, International Journalof Legal Medicine(2009). Of the 10 males, 9 carried Y-DNA R1a1a [M17]. Fairly close matches were found between the ancient DNA STR haplotypes and those in living persons in both eastern Europe and Siberia.8C. Bouakaze et al, First successful assay of Y-SNP typing by SNaPshot minisequencing on ancient DNA, International Journal of Legal Medicine vol. 121 (2007), pp. 493-499; C. Keyser et al, Ancient DNA provides new insights into the history of south Siberian Kurgan people,Human Genetics, vol. 126, no. 3 (September 2009), pp. 395-410. Mummies in the Tarim Basin also proved to carry R1a1a and were presumably ancestors of Tocharian speakers.9Chunxiang Li etal., Evidence that a West-East admixed population lived in the Tarim Basin as early as the early Bronze Age, BMC Biology, vol. 8, no. 15(2010).
Several studies by Indian geneticists offered the alternative view that R1a1a spread from India,10S. Sahoo et al, A prehistory of Indian Y chromosomes: Evaluating demic diffusion scenarios,PNAS, vol.103 (2006), no. 4, pp. 843-848. but this is not supported by more recent dating of R1a1a samples from across its range. Common ancestors of samples from European countries mainly fell between 4,050 and 4,825 years ago. The skeletons from Andronovo and succeeding cultures shared an R1a1a haplogroup with their oldest common ancestor calculated at between 5,050 and 1,800 years ago, while the oldest common ancestor found from India was calculated as 3,675 years ago. So a migration can be traced through its genetic trail from the border of Europe to Northern India.11A. Klyosov, DNA Genealogy, Mutation Rates, and Some Historical Evidences Written in Y-Chromosome, Nature Precedings(2008). Klyosov calculated the common ancestor of R1a1a males in the western Balkans lived 11,600 years ago. However this date seems to be the result of sampling error. Recalculation from the same data gives a result of 5,050 years ago (Prof. Kenneth Nordveldt personal communication.)
R1b1b2 (M269) provides the other half of the story. It has been dated to 5,000-8,000 years ago and appears to be the first R1b subclade to enter Europe. Although the highest densitiesof R1b1b2 and subclades today are in north west Europe (which initially led to the idea that it spread from Iberia), its highest diversity is in Asia Minor and the Caucasus. 12P. Balaresque et a., A Predominantly Neolithic origin for European paternal lineages, PLoS Biology, vol. 8, no. 1 (January 2010); B. Arredi, E. S. Poloni and C. Tyler-Smith, The peopling of Europe, in M. Crawford (ed.),Anthropological Genetics: Theory, methods andapplications (2007), p. 394. Equally telling is the fact that predecessor and brother haplogroups appear in western Asia. Subsequent mutations further down the line produced two huge sub-clades with clusters of offspring mutations - the sign of a population in rapid growth and spread. So a more recent study of the haplogroup suggested that it spread into Europe with the first farmers from Anatolia.13N.M Myres et al., A major Y-chromosome haplogroup R1b Holocene era founder effect in Central and Western Europe,European Journal of Human Genetics, vol. 19, no. 1 (January 2011), pp. 95-101; Fulvio Cruciani et al., Strong intra-and inter-continental differentiation revealed by Y chromosome SNPs M269, U106 and U152,Forensic Science International: Genetics, (advance online publication, 22 August 2010). There are several problems with that idea. We now know from a study of the route of Neolithic cultivars that the earliest farmers spread into Europe from the Levant, not Anatolia.14Coward, F., S. Shennan, S. Colledge, J. Conolly and M. Collard, The spread of Neolithic plant economies from the Near East to Northwest Europe: a phylogenetic analysis, Journal of Archaeological Science, vol. 35, no. 1 (2008), pp. 42-56. Its island-hopping progress to Greece alone took several thousand years. Overall from Cyprus (8,500 BC) to Scotland (4,000 BC) and Scandinavia (3,500 BC) farming took 5,000 years to spread over Europe. Ignoring Cyprus and counting only the burst from 6,200 BCslashes that estimate almost in half. Even so, that scarcely matches the great burst of U106 and P312 around 3,000 BC. Furthermore carriers of R1b1b2+ share with carriers of R1a1a a high level of lactose tolerance, which appears to have arisen with dairy farming, rather than in the earliest farmers. In fact the type common in Europe probably arose on or near the Pontic-Caspian steppe. (SeeLactase persistence - 13910T.)
It seems that among the steppe peoples, some tribes were dominated by haplogroup R1a1a, and others by R1b1b2. Judging by the end results, the Volga-Ural region, whence sprang the Afanasievo and Andronovo Cultures, was strong in R1a1a, while the region around the Sea of Azov was strong in R1b1b2. Since earlier forms of R1b appear in the Levant and seem connected to the spread of the Neolithic to Africa, we may hazard a guess that R1b1b2 had fed into the steppe with pastoralists who had developeddairy farming around the Sea of Marmara. R1b-L23 could have entered south-eastern Europe with dairy farmers, who contributed to the Cucuteni-Tripolye Culture adjacent to the steppe. In its late stages the Cucuteni-Tripolye Culture and Yamnaya Horizon cross-fertilised each other and merged to some extent.
This does not rule out the possibility of an entry via the Caucasus. People of the Maikop Culture could have carried it. (Some of the direct descendants of the tribes around the Sea of Azov remained on the steppe after others left for western Europe. They appear in history as the Cimmerians, who were driven out of the steppe into Anatolia and up the Danube in the Iron Age. It is therefore unlikely that we shall find their descendants in the Ukraine today.)
There is a much older link between the Volga-Ural region and the southern Caspian basin. The Yangelskaya Culture which appears in the former area around 9000 BC is virtually identical with finds in the latter area. Contacts between the two areas continued even into the Neolithic.15G. Matyushin, The Mesolithic and Neolithic in the southern Urals and Central Asia, chapter 10 in M. Zvelebil (ed.), Hunters in transition: Mesolithic societies of temperate Eurasia and their transition to farming (1986), p. 146. Here we have a clue that Mesolithic people carrying R1 may have moved between summer quarters on the steppe and winter quarters in the more sheltered forest fringing the southern Caspian - known as the Hyrcanian refuge. Transport by boat was within their power. Indeed images of rowing boats, though of later date, appear among the famed petroglyphs of Gobustan beside the Caspian in South-East Azerbaijan. We can deduce that R1a arose among those of their descendants who settled on the steppe, while R1b appeared among those descendants who favoured the southern homeland, and became involved in agriculture earlier. So far this can only be speculative, in the absence of ancient DNA.
The deductive process becomes even more tenuous in attempting to track the movements of R1b through the early farming cultures. One slender clue is that in the heartland of the Neolithic, farmers such as those at Çatalhöyük had begun to build rectangular houses of mud-bricks by 7,400 BC. Round huts on stone foundations were common earlier. So two farming cultures with round houses which appear after that and are otherwise technologically equal to or even advanced from Çatalhöyük, are interesting. One is the Ancient Fikirtepe Culture beside the Sea of Marmara and the other the Halaf Culture of the Levant. Could their origin point be the same? Could it be somewhere around the origin point of R1b? If so that could explain how R1b arrived in the Levant in time to contribute to the farming colonisation of North Africa and thence southward. We can deduce that they were speakers of Afro-Asiatic languages. If R1a and R1b spoke the same language in the Mesolithic, those days had past it seems. Sheer chance brought R1b and R1a carriers together again on the steppe millennia after their ancestors had parted ways. Economic fortunes eventually led to the dominance of the steppe people and hence their language.
This is not to say that the Indo-Europeans were all descended from the R1 founder. Nor is the R1a1a/R1b1b2 division so neat that there is no overlap. The two could travel together. R1a1a appears in areas that have never been Slavic, such as Scandinavia and West Germany.
A group of Bronze Age skeletons found in Lichtenstein cave, in Lower Saxony, provide a real life example of the mixture of haplogroups within one band. The men included two of Y-DNA R1a1, one of R1b, but no less than twelve of I2a2b [L38/S154] (formerly known as I2b2).16F. Schilz: MolekulargenetischeVerwandtschaftsanalysen am prähistorischen Skelettkollektiv derLichtensteinhöhle, Dissertation, Göttingen (2006). The last two haplogroups still reflect the connection shown in the cave. The present-day distributions of I2a2b and R1b-L21 both flow along the Rhine and into the British Isles.17H. De Beule, Origins of HgI-L38 (I2b2) Subclades (April 2009) and Early Bronze Age Origin and Late Iron Age (La Tène) Migrations of I-L38 (November 2009). The mtDNA haplogroups from the cave were dominated by H at 47%, which is close to the average for present-day Europe. Others individuals carried U5b, T2, and J*, a mix of Mesolithic and Neolithic markers.18F. Schilz: Molekulargenetische Verwandtschaftsanalysen am prähistorischen Skelettkollektiv der Lichtensteinhöhle, Dissertation, Göttingen (2006). The spread of mtDNA H2a mimics that of Y-DNA R1a1a to some extent, which may or may not be coincidental. 19E.-L. L. Loogväli et al, Disuniting Uniformity: a pied cladistic canvas of mtDNA haplogroup H in Eurasia, Molecular Biology and Evolution vol. 21, no. 11 (2004), p. 2017. As mentioned above, mtDNA H5a seems to have sprung from the western Caucasus and dispersed widely some time after farming, suggesting that it spread with the Indo-Europeans. It has been found in the DNA of one man of the Tagar Cultureof the eastern steppe, a precurser to the Scythians.
The Y-DNA R1a1a subclade defined by marker M458 is strongly correlated with the spread of Slavic languages,20P.A. Underhill et al., Separating the post-Glacial coancestry of European and Asian Y chromosomes within haplogroup R1a,European Journal of Human Genetics, advance online publication (4 November 2009). This paper uses the evolutionary effective mutation rate, which generally overestimates ages by a factor of three. The authors therefore do not recognise the correlation between M458 and Slavic language dispersal. as is haplogroup Y-DNA I2a1b1 [L69.2/S163.2] (formerly known as I2a2a). Though the genetic and linguistic map of South-East Europe has been churned up again and again by historic migrations, the focus of I2a1b1 in Bosnia led Vincenza Battaglia and his colleagues to see its origins in the western Balkans. They dated its spread to the Neolithic period, despite computing a variation age of only 4,000 years.21V. Battaglia et al, Y-chromosomal evidence of the cultural diffusion of agriculture in southeast Europe, European Journal of Human Genetics, vol. 17, no 6. (June 2009), pp. 820-30.
It seems more likely that the Cucuteni-Tripolye farmers had absorbed some local hunter-gathers of haplogroup I, and then contributed I2a into the composite Yamnaya-with-Cucuteni-Tripolye culture. If I2 was associated with Usatovo and the villages along the Dniester, that would explain why I2a2b [L38/S154] appears alongside R1a1a after apparently migrating up the river and around the Carpathians into present-day Germany (Lichtenstein Cave). Meanwhile some I2a1b1 [L69.2/S163.2] people could have been living in the Late Cucuteni-Tripolye towns of the Middle Dnieper. If their descendants chose to remain in what became the Proto-Slavic heartland, (together with a few R1a1a men, in one of whom the M458 mutation arose), until population growth pushed them outward in all directions in the early Middle Ages, that would explain the pattern we see today. Yet we may guess that at least one man carrying I2a2a travelled with a Yamnaya band up the Danube and all the way to the British Isles, since his descendants are found in Scotland and Ireland.
A related lineage - I2a1b2 [L161] - may spring from an earlier migration. It appears in the North European Plain and the British Isles. If the early farmers around the Danube did indeed absorb some I2a, then a man of that haplogroup may have been carried north with the LinearBandKeramik (LBK) Culture or with a later wave of dairy farmers. We may picture a descendant somewhere in what is now Germany or nearby in whom the mutation L161 appeared around 3,000 BC, to be carried eventually into Britain.22Research on the ages and locations of haplogroup I by Prof. Kenneth Nordtvedt, who is not responsible for my conclusions.
The discovery of four family graves in Eulau, Saxony-Anhalt, Germany, of the Corded Ware culture, provided an opportunity to use an array of research techniques. They were all victims of a violent attack, which killed men, women and children. Family relationships were traced through DNA. One Y-DNA haplogroup R1a1a father was buried with his mtDNA K1b wife and their two sons. Other mtDNA haplogroups present were H, I, U5b and the rare X2. Strontium isotope results showed that the men and children had grown up in the local area, whereas the women had spent their early lives some distance away, which suggests a patrilocal society.23W. Haak et al., Ancient DNA, Strontium isotopes, and osteological analyses shed light on social and kinship organization of the Later Stone Age, PNAS. Online before print November 17, 2008.
Another clue to the impact of pastoralists from the steppes is the fact that most European adults can drink milk. That is the result of a helpful genetic mutation that confers lactase persistence. Lactase is the enzyme in the intestinal system that can metabolise lactose, a sugar found in milk and other dairy products. Mammals produce this enzyme as infants, so they can digest their mother's milk. The production of lactase shuts down automatically after infancy in all mammals except some human beings. East Asians and many Africans are lactose intolerant as adults, while most Europeans, Pakistanis and those African and Middle Eastern populations historically associated with pastoralism can digest milk as adults.
There have been at least six separate mutations which cause the lactase switch-off to fail. Lactase-persistence genes of East African origin are 13907G and 14010C. A compound allele (13915G and 3712C) found in Saudis and East Africans probably originated in the Middle East.24N.S.Enattah et al, Independent introduction of two lactase-persistence alleles into human populations reflects different history of adaptation to milk culture, American Journal of Human Genetics, vol. 82 (2008), no. 1, pp.57-72; C.J.E. Ingram et al, Lactose digestion and the evolutionary genetics of lactase persistence, Human Genetics, vol.124, no.6 (January 2009), pp. 579-591. A rarer lactase-persistence allele (13913C) was discovered in two cases in Italy, and subsequently reported in Cameroon, Sudan, Ethiopia, and the Bedouin population in Saudi Arabia.25I. Järvelä et al, Molecular genetics ofhuman lactase deficiencies, Annals of Medicine, vol. 41, no. 8 (2009), pp. 568-575; A.Piepoli et al., Genotyping of thelactase-phlorizinhydrolase C/T-13910 polymorphism by means of a new rapiddenaturing high-performance liquid chromatography-based assay in healthysubjects and colorectal cancer patients, Journal of Biomolecular Screening,vol. 12, no. 5 (2007), pp.733-739. The dominant mutation in Western Eurasia and South Asia is 13910T (rs4988235(T)).26J. Babuet al., Frequency of lactose malabsorption among healthy southern and northern Indian populations by genetic analysis and lactosehydrogen breath and tolerance tests, American Journal of Clinical Nutrition, Vol. 91, No. 1 (January 2010), pp. 140-146. This allele is found within different haplotype backgrounds i.e. the stretches of DNA code either side of it. One of these is common, as we shall see, but the others (which originate from the same ancestral haplotype) are found only in populations living on the Pontic-Caspian steppe. 27N.S.Enattah et al., Evidence of still-ongoing convergence evolution of the lactase persistence T-13910 alleles in humans, American Journal of Human Genetics, vol. 81 no.3 (2007), pp. 615-25. So the allele may have originated in that area, or at least been there long enough to acquire a diversity of haplotype background. The 22018A (rs182549(C)) mutation was first recognised in Finns. While it generally correlates with 13910T in Europeans, it can appear as an independent cause of lactase-persistence, for example in Pakistanis and notably in the Kazaks of Northern China.28N.S.Enattah et al., Identification of a variantassociated with adult-type hypolactasia,Nature Genetics, vol. 30,no. 2 (2002), pp.233-7; C.J.E. Ingram et al, Lactose digestion and the evolutionary genetics of lactase persistence,Human Genetics, vol.124, no.6 (January 2009), pp. 579-591; T. Bersaglieri et al., Genetic signatures of strong recentpositive selection at the lactase gene, American Journal of HumanGenetics, vol. 74, no. 6 (June 2004), pp. 1111–1120; Lidan Xu et al., The -22018A allele matches thelactasepersistence phenotype in northern Chinese populations,Scandinavian Journal of Gastroenterology, vol. 45, no. 2 (February 2010), pp.168-17.
The haplotype containing these two alleles (13910T and 22018A) is common in Northern European-derived populations (77% in European Americans). It is also largely identical over nearly 1 cM. Such lengthy stretches of DNA indicate recent origin; older DNA has had more time to be broken up by recombination each generation. The haplotype could not have risen quickly to such high frequency without the aid of natural selection. Todd Bersaglieri and colleagues calculated that strong selection occurred within the past 5,000–10,000 years, consistent with an advantage to lactase persistence in the setting of dairy farming. Even more recent estimates (1,625–3,188 years ago) were obtained for a Scandinavian population, suggesting stronger and more recent selection there.29T. Bersaglieri et al., Genetic signatures of strong recent positive selection at the lactase gene,American Journal of Human Genetics, vol. 74, no. 6 (June 2004), pp. 1111–1120.
Evidence of milking first appears around the Sea of Marmara c. 6500–5000 BC, but initially the milk was processed into yoghurt or cheese, which would make it more digestible for those without lactase.30R.P. Evershed et al., Earliest date for milk use in the Near East and southeastern Europe linked to cattle herding,Nature, vol. 455 (25 September 2008), pp. 528-31 (531). The lack of 13910T in Anatolia except for that part beside the Sea of Marmara suggests that the mutation did not occur in the early Neolithic, but among dairy farmers spreading into South-East Europe in the Late Neolithic. It could have moved up the Danube with dairy farmers, but also into the Cucuteni herders adjacent to the steppe and from there into the steppe peoples adopting pastoralism. The distribution of 13910T and 22018A suggests that they were spread by Indo-European-speakers both east and west. If so there must have been intermarriage between the Yamnaya tribes, to disseminate these genes thoroughly before the migrations. In a patrilocal system, in which wives moved into the husband's home (which has been deduced on linguistic evidence for Proto-Indo-European-speakers 31Laura Fortunato and Fiona Jordan, Your place or mine? A phylogenetic comparative analysis of marital residence in Indo-European and Austronesian societies,Philosophical Transactions of the Royal Society B: Biological Sciences, vol. 365, no. 1559 (12 December 2010), pp. 3913-3922.), such mixing would have no impact on which Y-DNA haplogroup was predominant within a family. 13910T is found both among nations in which R1a1a+ is dominant and those in which R1b1b2+ is dominant. A clue that it occurred first among R1b-dominant people is provided by its appearance among the Fulani, a Chadic-speaking tribe of pastoralists.32A. Inkeri Lokki et al., Lactase persistence genotypes and malaria susceptibility in Fulani of Mali, Malaria Journal, vol. 10, no. 9 (14 January 2011). R1b-V88 appears to have spread into Africa with farmers speaking an Afro-Asiatic language that developed into Chadic. See Mediterraneans: Farmers.
The 13910T mutation also crops up right across the far north of Europe and Asia, where Uralic languages are spoken. Proto-Uralic evidently evolved in contact with Proto-Indo-European (PIE), since it borrowed words from PIE. Its homeland is generally supposed to have been close to the Ural Mountains, whence its speakers spread eastward and westward in parallel with Indo-Europeans, but on a more northerly trajectory.33D. W. Anthony,The Horse, the Wheel, and Language (2007), pp. 93-4 summarises. For a much more detailed discussion see C. Carpelan and A. Parpola (eds),Early Contacts between Uralic and Indo-European: Linguistic and archaeological considerations (2001). Inter-marriage between PIE-speakers and Proto-Uralic speakers could explain the spread of 13910T to the latter.
Adults who could digest raw milk had an excellent source of food on the hoof. Cattle could go on turning grass into milk for years before slaughter for beef. It has been proposed that lactase persistence was the genetic edge that allowed the dairy pastoralist Indo-Europeans to spread. Dairy-farming produces five times as many calories per acre as raising cattle for slaughter.34G. Cochran and H. Harpending, The 10,000 Year Explosion: How civilization accelerated human evolution (2009), pp. 174, 181-86. The protein and calcium of milk certainly builds bones. Prehistoric dairy farmers tended to be taller than other farmers.35N. Koepke, Nutritional status in pre-historic and historic Europe, paper read at The Economic History Society Annual Conference, University of Durham, 26-28 March 2010. Online in the accompanying provisional volume, pp. 13-18; N. Koepke and J. Baten, Agricultural Specialization and Height in Ancient and Medieval Europe, Explorations in Economic History, vol. 42, no. 2 (2008), pp.127-46.
Analysis of ancient DNA has so far confirmed the deduction that 13910T spread with dairy farming. It has not been found in early European Neolithic (LBK) human remains, or late Neolithic remains from the south of France, but appears common in the Funnel Beaker Culture (TRB). A low level (5%) has been found among Pitted Ware hunter-gatherers who had contact with TRB farmers. Despite this evidence, some authors argue that this mutation arose in Central Europe and was spread by the LBK. Such an origin could scarcely explain why it crops up in India. 36J. Burger et al, Absence of the lactase-persistence-associated allele in early Neolithic Europeans,Proceedings of the National Academy of Sciences USA, vol. 104 (2007), no. 10, pp. 3736-3741; M. Lacan et al., Ancient DNA reveals male diffusion through the Neolithic Mediterranean route, Proceedings of the National Academy of Sciences of the USA, online before print May 31, 2011; A. Linderholm, Migration in Prehistory: DNA and stable isotope analyses of Swedish skeletal material (Doctoral thesis Stockholm 2008); H. Malmstrom et al., High frequency of lactoseintolerance in a prehistoric hunter-gatherer population in northern Europe, BMC Evolutionary Biology (online ahead of print 2010); Y. Itan et al, The Origins of Lactase Persistence in Europe, PLOS Computational Biology, vol. 5, no. 8(2009).
If you are using a browser with up-to-date support for W3C standards e.g. Firefox, Google Chrome, IE 8 or Opera, hover over the superscript numbers to see footnotes online. If you are using another browser, select the note, then right-click, then on the menu click View Selection Source. If you print the article out, or look at print preview online, the footnotes will appear here.