The Demographic History Of Southeast Asia

Analysis of uniparental DNA clades and linguistic evidence in modern Southeast Asian populations and archaeology had supported this conclusion a decade ago. The migrationist paradigm continues to rule. Now, ancient DNA confirms it. 

Two distinct population models have been put forward to explain present-day human diversity in Southeast Asia. The first model proposes long-term continuity (Regional Continuity model) while the other suggests two waves of dispersal (Two Layer model). 

Here, we use whole-genome capture in combination with shotgun sequencing to generate 25 ancient human genome sequences from mainland and island Southeast Asia, and directly test the two competing hypotheses. 

We find that early genomes from Hoabinhian hunter-gatherer contexts in Laos and Malaysia have genetic affinities with the Onge hunter-gatherers from the Andaman Islands, while Southeast Asian Neolithic farmers have a distinct East Asian genomic ancestry related to present-day Austroasiatic-speaking populations. 

We also identify two further migratory events, consistent with the expansion of speakers of Austronesian languages into Island Southeast Asia ca. 4 kya, and the expansion by East Asians into northern Vietnam ca. 2 kya. These findings support the Two Layer model for the early peopling of Southeast Asia and highlight the complexities of dispersal patterns from East Asia.

Hugh McColl, et al., "Ancient Genomics Reveals Four Prehistoric Migration Waves into Southeast Asia" BioRxiv (March 8, 2018).

Some first blush thoughts.

Two(+2) early waves or three (+2)?

Was there are distinct pre-Hoabinhian wave of pre-Papuan/Australian folk whom Onge-like hunter-gatherers (largely) replaced? If there was a replacement, what gave the newcomers an edge and when did that replacement happen?

The first paragraph of the paper impliedly raises this question but it doesn't seem to get resolved. It says:

The population history of Southeast Asia (SEA) has been shaped by interchanging periods of isolation and connectivity. Anatomically modern humans first colonized SEA at least 70,000 years ago. Within SEA, the complex topography and changes in sea level promoted regional expansions and contractions of populations. By the late Pleistocene/early Holocene, a pan-regional lithic technological culture was established across mainland SEA, named Hoabinhian. Hoabinhian foragers are thought to be the ancestors of present-day SEA hunter-gatherers, sometimes referred to as ‘Negritos’ because of their comparatively darker skin colour and short stature. Today, however, the majority of people in SEA are believed to be descendants of rice and millet farmers with varying degrees of East Asian phenotypic affinity, suggesting that human diversity in SEA was strongly influenced by population expansions from the north. Yet, the extent to which the movements from East Asia (EA) impacted on the genetic and cultural makeup of the people of SEA remains controversial.

The early Holocene is 10,000 years ago, leaving a 60,000 year gap between the first settlers and the Hoabinhian culture. There are limits to what the ancient DNA can do to resolve this because the ancient DNA samples in the study are not too old (the oldest is no more than 8,000 years old at more than two sigma). But, on closer examination, it appears that this question was examined but not resolved.

Unlike all other ancient samples, the two Hoabinhian samples (which also happen to be the oldest samples in our study) - Pha Faen, Laos (La368 - 14 140 C 7,888 ± 40) and Gua Cha, Malaysia (Ma911 - 14 C 4,319 ± 64) - designated as Group 1, cluster distantly from most East and Southeast Asians in the PCA and position closely to present-day Onge (Figure 1A). Group 1 individuals also contain a mixture of several different ancestral components in the fastNGSadmix plot, including components shared with Onge, the Pahari and Spiti from India, Papuans and Jehai (a Malaysian ‘Negrito’ group), which are markedly different from the other SEA ancient samples. This possibly results from our modeling of ancient populations as a mixture of components inferred in present-day populations, via fastNGSadmix, and from the fact the ancient samples are likely poorly represented by a single present-day group. 

The rest of the ancient samples are defined primarily by East and Southeast Asian components that are maximised in present-day Austroasiatic (Mlabri and Htin), Austronesian (Ami) and Hmong (indigenous to the mountainous regions of China, Vietnam, Laos and Thailand) populations, along with a broad East Asian component. . . . 

We used D-statistics of the form D(Papuan,Tianyuan,X,Mbuti), where X is a test population, to explore the relatedness of ancient and present-day Southeast Asians to two highly differentiated groups: Papuans and an ancient northern East Asian individual (Tianyuan - a 40 kya-old sample from Northeastern China). The values of this D-statistic are consistent with present-day and ancient SEA mainland samples being more closely related to Tianyuan than to Papuans. This applies to present-day northern EA populations, and - more weakly - to most populations of ancient and present-day SEA. However, this D-statistic is not significantly different from 0 in present-day Jehai, Onge, Jarawa and Group 1 - the ancient Hoabinhians. While the Onge’s relationship with Papuans and Tianyuan is unclear, D-statistics of the form D(Onge,Tianyuan,X,Mbuti), where X is a test population, show that Jarawa, Jehai and the ancient Group 1 share more ancestry with Onge than with Tianyuan. Like the Onge, both Group 1 samples carry mtDNA haplogroups from the M lineage, thought to represent the coastal migration to Australasia. 

To assess the diversity among the remaining ancient individuals, we computed a new PCA including only EA and SEA populations that did not have considerable Papuan or Ongelike  ancestry. . . . 

Group 2 samples from Vietnam, Laos, and the Malay Peninsula are the oldest samples after Group 1, and range in age from 4.2 to 2.2 kya. 

At K=6, Group 2 individuals, the present-day Mlabri and a single Htin individual are the only MSEA samples in the fastNGSadmix analysis to lack the broad EA component (dark green) maximised in northern EA , with the exception of the Malaysian ‘Negritos’ and ‘Proto-Malays’ (Temuan). 

At K=7, a bright green component is maximised in these populations, and this component is also found in present-day Indonesian samples west of Wallace’s Line. The two ancient Indonesian samples (Group 5; 2.2 to 1.9 kya) represent a mix of Austronesian- and Austroasiatic-like ancestry, similar to present-day western Indonesians. Indeed, after Mlabri and Htin, the closest populations to Group 2 based on outgroup-f3 statistics are the western Indonesian samples (from Bali and Java) reported to have the highest amounts of ancestry from mainland SEA. 

These lines of evidence suggest Group 2 are possible descendants of an “Austroasiatic” migration that expanded southward across MSEA and into island SEA (ISEA) by 4 kya. We also observe a gradient in “Austronesian-like” vs. “Austroasiatic-like” ancestry in the PCA: while PC1 separates populations found in SEA and those found in northern EA, PC2 distinguishes population based on their amounts of Austronesian-like ancestry (pink component in Figure 1 - lower panel) versus Austroasiatic-like ancestry (bright green component in Figure 1 - lower panel).

The Supplemental Materials note that:

Relationship between Papuan, Tianyuan and EA/SEA/Ancients

We find support for Australians and Bougainville islanders forming a clade with Papuans, to the exclusion of Tianyuan (Table S14). In turn, many EA and SEA form a clade with Tianyuan, to the exclusion of Papuan (Table S13). Onge, Jarawa and Jehai do not form a clade with either Papuans or Tianyuan (Table S13, S14), but have a stronger affinity to Papuans than to Tianyuan (Z = 3 - 4.2, for D(Onge/Jarawa/Jehai, Tianyuan; Papuan, Mbuti)).

Relationship between Onge, Tianyuan and SEA

We find Onge, Jarawa and Jehai form a clade with Onge to the exclusion of Tianyuan, but no other EA or SEA population form a clade with Onge, to the exclusion of Tianyuan (Table S17).

Relationship to Surui and Mixe

We tested for a specific affinity in the Surui to our ancient samples, as was previously detected in Papuans, Onge and Tianyuan. For the 2240k panel, we find that D-statistics of the form D(Mixe, Surui, Group 1 individual, Mbuti) are high but non-significant (Z = -2.18 and -2.5, using Ma911 and La368, as the Group 1 representative, respectively) (Table S19).

Relevant data is also contained in a paper looking at the genetics of modern Indonesia.

What cultural and technological and ecological impact did each of these waves have? Were there climate or other events that drove these transitions?

One plausible possibility in my mind is that the Tianyuan individual, the Onge, Mainland Southeast Asian Negritos, and pre-admixture Ancestral South Asians are all part of a wave of Asian migration after an initial pre-Papuan wave ca. 70,000 years ago, but at least somewhat before the Tianyuan individual ca. 40,000 years ago. In that scenario, these second wave hunter-gatherers may have been able to conquer pre-Papuan first wave hunter-gatherers by virtue of the fact that they had domesticated dogs at their disposal, while the pre-Papuans did not (or if they did, didn't bring them with them on their maritime colonization journey). Certainly, we know that the founding population of the Americas which would have arrived in Beringia more than 20,000 years ago had dogs (a conclusion that tends to trump estimates for dog domestication at just 15,000 years ago or less). I previously explored this hypothesis at greater length here as a way to help explain the dilution of Denisovan ancestry in mainland Southeast Asia and parts of Island Southeast Asia that were previously part of Sundaland.

This time frame is a fairly good fit the genetically and archaeologically estimated time frame in which dog were domesticated: One scholar has argued at book length that dog domestication was a key factor in the Upper Paleolithic revolution and also in giving modern humans an edge over the Neanderthals in Europe. See also here. There have even been some studies that located the domestication of the dog event (if there was just one primary one) in Southeast Asia, although there is not a consensus on that point.

It also isn't too far afield from estimates of wave of Asian migration in the Upper Paleolithic based upon analysis of the phylogeny of uniparental markers in modern populations in light of known mutation rates.

Archaic hominins in Southeast Asia

We know that archaic Homo erectus was present in Southeast Asia before any modern humans. We know that Homo floresiensis was present in Southeast Asia before any modern humans. We are quite confident that Homo floresiensis were not Denisovans because their physical anthropology is too archaic even relative to Homo erectus.

In the modest straightforward scenario by which Papuan/Australian folk receive Denisovan admixture, there were Denisovans in Southeast Asia before any modern humans.

We don't know with any great confidence which archaic hominins were present to have first contact with modern humans, although Denisovans must have been among them.

The paper has this to say about Denisovan admixture:

We find that the genetic diversity found in present day SEA populations derives from at least four prehistoric population movements by the Hoabinhians, an “Austroasiatic-like” population, the Austronesians and, finally, additional EA populations into MSEA. We further show that the ancient mainland Hoabinhians (Group 1) shared ancestry with present-day Onge of the Andaman Islands and the Jehai of peninsular Malaysia. These results, together with the absence of significant Denisovan ancestry in these populations, suggest that the Denisovan admixture observed in Papuans occurred after their ancestors split from the ancestors of the Onge, Jehai and the ancient Hoabinhians. This is also consistent with the presence of substantial Denisovan admixture in the Mamanwa from the Philippines, which are best modeled as resulting from an admixture between Austronesians and Papuans, not Onge.

I think a model in which Papuans comes first, and entirely separate wave of modern humans akin to the Onge arrive next is a better description of the most plausible inference, even though they may be technically equivalent.

Why don't researchers integrate more kinds of evidence?

I remain puzzled by the reluctance of investigators to more heavily integrate distinct evidence from the modern mix of human uniparental markers, modern autosomal DNA in humans, ancient and modern plant genetics, ancient DNA, linguistics, and archaeology into a single comprehensive analysis which would be so much more powerful. It isn't as if these are lone wolf investigators who can't play well with others. These papers are done by large teams with lots of well informed feet on the ground. Some of this may flow from the incentives to publish in the smallest publishable unit and to have relatively short papers in science relative to the humanities or law. But, one can get much more powerful conclusions by integrating all of the available evidence.

The paper does outline that "Two layer model" that its data eventually confirms:

[T]he Two Layer model advocates for two major dispersal waves into SEA, where EA farmers replaced the original Hoabinhian inhabitants across SEA through a major demographic southward expansion ca. 4 kya. The exception to this would be the isolated populations of the Andaman Islands, peninsular Thailand/Malaysia and the Philippines which are considered the primary descendants of Hoabinhian hunter-gatherers. Under this model, the migratory wave of farmers originated in present-day China, where rice and millet were fully domesticated in the Yangtze and Yellow River valleys between 9-5.5 kya, and paddy fields developed by 4.5 kya. Farming practices are thought to have accompanied these populations as they spread southward through two main routes – an inland wave associated with the expansion of Austroasiatic languages, and an island-hopping route associated with Austronesian languages which eventually reached the Pacific. Within mainland SEA (MSEA), exchanges with EA appear to have continued in the recent past, however, the extent to which these expansions had a genetic impact on the indigenous populations is unknown.

Some of this gap is filled in with efforts like Razib's excellent post on the topic which contextualizes this paper's findings in a larger cultural context and also ties it in to relevant aspects of South Asian pre-history like the origins of the Munda people of India. With respect to the two waves of migration after the "first farmers" ca. 4 kya, he explains:

The authors also detect migrations into Southeast Asia besides that of the Austro-Asiatics and Austronesians. One element seems correlated with the Tai migrations, and another with Sino-Tibetan peoples, most clearly represented in Southeast Asia by the Burmans. The excellent book, Strange Parallels: Volume 1, Integration on the Mainland: Southeast Asia in Global Context, c.800–1830, recounts the importance of the great migrations of the Tai people into Southeast Asia ~1000 A.D. Modern-day Thailand was once a flourishing center of Mon civilization, an Austro-Asiatic people related to the Khmers of Cambodia. The migrations out of the Tai highlands of southern China reshaped the ethnography of the central regions of mainland Southeast Asia. The Tai also attempted to take over the kingdoms of the Burmans. Though they failed in this, the Shan states of the highlands are the remnants of these attempts (tendrils of the Tai migrations made it to India, the Ahom people of Assam were Tai). Vietnam, shielded by the Annamese Cordillera, came through this period relatively intact. It is also well known that Cambodia’s persistence down to the present has much to do with the shielding it received from France in the 19th century in the wake of Thai expansion.

What Does This Mean For South Asian Pre-History?

With respect to the Munda of India, Razib notes that:

They detect shared drift between Austro-Asiatic people and tribal populations in northeast India. This is not surprising. A 2011 paper found that Munda speaking peoples, whose variant of Austro-Asiatic is very different from that of Southeast Asia, are predominant carriers of Y chromosome O2a. This is very rare in Indo-European speaking populations, and nearly absent in Dravidian speaking groups. Additionally, their genome-wide patterns indicate some East Asian admixture, albeit a minority, while they carry the derived variant of EDAR, which peaks in Northeast Asia. 

One debate in relation to the Munda people is whether they are primal and indigenous, or whether they are intrusive. The genetic data strongly point to the likelihood that they are intrusive. An earlier estimate of coalescence for O2a in South Asia suggested a deep history, but these dates have always been sensitive to assumptions, and more recent analysis of O2a diversity suggests that the locus is mainland Southeast Asia. 

Now that archaeology and ancient DNA confirm Austro-Asiatic intrusion into northern Vietnam ~4,000 years ago, I think it also sheds light on when these peoples arrived in India. That is, they arrived < 4,000 years ago. As widespread intensive agriculture came to Burma ~3,500 years ago, I think that makes it likely that Munda peoples arrived in South Asia around this period. 

I now believe it is likely that the presence of Austro-Asiatic, Dravidian, and Indo-Aryan languages in India proper was a feature of the period after ~4,000 years ago. None of the languages of the hunter-gatherer populations of the subcontinent remain, with the possible exception of isolates such as Nihali and Kusunda.

The recent origins of Austro-Asiatic, Indo-Aryan and Tibeto-Burmese languages in India are all established beyond what I would consider to be reasonable doubt through multiple lines of evidence.

The case of Dravidian, which is correlated, especially in lower caste populations, with Ancestral South Indian genetics that autochthonous to India and for whom the closest modern population is the Onge population of the Andaman Islands, is trickier, and one I've explored before. 

At a minimum, the close linguistic relations between the Dravidian languages suggest that this language family underwent a linguistic bottleneck, possibly sometime after the arrival of the Indo-Aryan invasion of India, from which all modern Dravidian languages derive, even if the pre-agricultural people of India spoke languages that were part of a family that includes proto-Dravidian.

It is also very plausible that the rise of the Dravidian languages, whether or not my bottleneck conjecture for the Dravidian languages is correct, is associated with the South Indian Neolithic revolution ca. 4500 years ago (i.e. around 2500 BCE), only a thousand years or less before the Indo-Aryan invasion. There is ample precedent of language replacement in favor of a dominant farmer language in association with the expansion of a newly food producing culture. 

This would be a few centuries older than Razib's casual estimate, but still reflects the same basic theme that the pre-agricultural hunter-gatherer languages of India are probably now entirely lost or are represented by only a couple of nearly moribund language isolates while untold scores or hundreds of other hunter-gatherer languages of pre-agricultural India (since hunter-gatherer civilizations appear to have had more linguistic diversity than early farmer civilizations in the fertile areas that first adopted farming) have been forever lost.

It is far less obvious, however, whether Dravidian's ultimate source (possibly a source prior to a bottleneck that impairs the usual linguistic methods of dating it) was home grown, becoming dominant while stamping out its neighbors, or if Dravidian has its source in the same population that brought the crops that made the South Indian Neolithic revolution possible as none of the core crops in that food production package are native to India. Some of those crops have origins in the Fertile Crescent and others arose in the wild and were domesticated in the African Sahel.

I've also explored potential genetic markers of an outside source for Dravidian, with Y-DNA T looking like a particularly promising marker in light of its geographical distribution within India, the fact that it is almost certainly invasive to India, and the fact that it was present in a place that would have been on the path of African Sahel crops to India, but I don't have data good enough to confirm or rule out those hypotheses definitively. Without more detailed sub-haplogroups of Dravidian Y-DNA T bearers, it is hard to date their antiquity, their diversity, and the place from which they made an invasive appearance in India. If it is a distinctly Indian and very basal clade of Y-DNA T, my hypothesis that it arrived with African Sahel crops is probably wrong. If it shows affinities in particular to clades of Y-DNA found in Yemen and Somalia and Ethiopia, and shows a star-like pattern of sub-haplotypes specific to India around 2500 BCE (per this paper by Dorian Fuller), it would strongly confirm my hypothesis.

The much anticipated ancient Harappan DNA evidence that should appear in published work any day now won't be much help because, as I have recent argued at length elsewhere, the Harappan language was unlikely to be a Dravidian language and while it may have shared some areal linguistic features with Dravidian, may not have even been in the same language family.

Termites Are Social Cockroaches

Ten and a half year after a scientific paper reached this conclusion the U.S. scientific body that handles the naming of insect species has agreed that termites are a social sub-type of cockroach. As Science News explains:

The Entomological Society of America is updating its master list of insect names to reflect decades of genetic and other evidence that termites belong in the cockroach order, called Blattodea. 

As of February 15, “it’s official that termites no longer have their own order,” says Mike Merchant of Texas A&M University in College Station, chair of the organization’s common names committee. Now all termites on the list are being recategorized. 

The demotion brings to mind Pluto getting kicked off the roster of planets, says termite biologist Paul Eggleton of the Natural History Museum in London. He does not, however, expect a galactic outpouring of heartbreak and protest over the termite downgrade. Among specialists, discussions of termites as a form of roaches go back at least to 1934, when researchers reported that several groups of microbes that digest wood in termite guts live in some wood-eating cockroaches too. 

Once biologists figured out how to use DNA to work out genealogical relationships, evidence began to grow that termites had evolved as a branch on the many-limbed family tree of cockroaches. In 2007, Eggleton and two museum colleagues used genetic evidence from an unusually broad sampling of species to publish a new tree of these insects (SN: 5/19/07, p. 318). Titled “Death of an order,” the study placed termites on the tree near a Cryptocercus cockroach. 

The paper that spurred this reassessment is:

D. Inward, G. Beccaloni and P. Eggleton. Death of an order: a comprehensive molecular phylogenetic study confirms that termites are eusocial cockroaches. 3 Biology Letters 331 (June 22, 2007) doi: 10.1098/rsbl.2007.0102.

Coming Attractions

A number of really good anthropology and genetic papers have crossed my computer screen recently, but I have been out of pocket with work commitments, and also not quite sure what take to give them.

Sometimes soon, I will either cave and just do a "quick hits" post to acknowledge the new developments for future reference, or will put together some of them into a richer post with more analysis.

The Dynamics Of The Crater II Dwarf Satellite Galaxy Are Inconsistent With The Dark Matter Hypothesis

Crater II is a dwarf galaxy that orbits the Milky Way at a distance of about 360,000 light years, which while distant, is still within the gravitational field of the Milky Way. It is the "coldest" galaxy ever observed.

In the current iteration of MOND, its velocity dispersion was predicted to be 4 km/s if it were isolated in space away from any gravitational effect.

But, due to an "External Field Effect" of the current iteration of MOND, which violates the "Strong Equivalence Principle" of General Relativity that arises from the fact that it is still in the Milky Way's gravitational field in MOND, its velocity dispersion was predicted in advance of a measurement of this quantity to be 1.5 km/s to 3.0 km/s.

Crater II is an unusual object among the dwarf satellite galaxies of the Local Group in that it has a very large size for its small luminosity. This provides a strong test of MOND, as Crater II should be in the deep MOND regime (gin≈34km2s−2kpc−1≪a0=3700km2s−2kpc−1). Despite its great distance (≈120 kpc) from the Milky Way, the external field of the host (gex≈282km2s−2kpc−1) comfortably exceeds the internal field. Consequently, Crater II should be subject to the external field effect, a feature unique to MOND. This leads to the prediction of a very low velocity dispersion: σefe=2.1+0.9−0.6kms−1.

Stacy S. McGaugh, "MOND Prediction for the Velocity Dispersion of the `Feeble Giant' Crater II" (November 3, 2016).

This is a big deal because under a wide range of dark matter hypotheses, the velocity dispersion could have been no lower than 5 km/s and was expected to be more like 11 km/s to 24 km/s.

The actual velocity dispersion of Crater II was measured with the latest and greatest telescopes in a result first announced six and a half weeks after this prediction was made on December 19, 2016. What did they find?

A velocity dispersion of 2.4 km/s to 3.0 km/s. 

This was perfectly consistent with MOND adjusted for the external field effect at the one sigma level.

It is inconsistent with MOND without the external field effect by about four sigma. 

It is inconsistent at more than 7 sigma with even the most conservative estimate based upon dark matter, and is inconsistent at the 27 sigma level with a prediction at the low end of more typical dark matter assumptions.

Crater II is a rejection of the dark matter hypothesis that is particularly notable because it is independent of pretty much all of the other empirical evidence that shows that the dark matter hypothesis is inconsistent with the data, and because it is a very clean measurement that the dark matter hypothesis once again has failed to predict accurately.

I'll sum up all of the leading ways that the dark matter hypothesis contradicts empirical evidence in future post recapping the state of the dark matter debate when you consider Crater II with all of the over evidence. But, for now, the take away point is that multiple rejections of the cold dark matter hypothesis by multiple different methods makes the conclusion that the cold dark matter hypothesis is wrong much more robust and reliable. 

Incidentally, while the "external field effect" has no place in conventional general relativity as it violates the strong equivalence principle,* it makes lots of sense in a graviton based quantum gravity theory where there is a physical difference (a higher density of gravitons per cubic meter of space)  between the situation when there is an external field than when there is not an external field.

* as the link explains, general relativity arguably does not perfectly honor the strong equivalence principle either, but the deviation is negligible in weak gravitational fields with objects moving slowly relative to the speed of light which is the regime which Crater II inhabits. But see here (arguing that conventional general relativity does obey the strong equivalence principle).

In particular, this difference of graviton density could have an effect as a consequence of the fact that the importance of graviton self-interactions (which are the source of MOND-like effects in Deur's work) relative to the overall strength of the graviton field.

It will take time for the conclusion that dark matter phenomena are due to some form of gravity modification relative to conventional applications of classical general relativity, probably associated with quantum gravity effects that only manifest in weak gravitational fields. This modified/quantum gravity explanation is a better fit than some new undiscovered form of non-baryonic matter, possibly in association with some new undiscovered force that exists only in interactions involving this non-baryonic matter with other non-baryonic matter and/or between non-baryonic matter and ordinary matter.

Realistically, a generation of astrophysicists are going to have to die before MOND receives universal acceptance.

But, the evidence is already there and is already sufficiently weighty to make it clear that some form of MOND and not some form of dark matter and dark forces, is the answer. What is left is the sociology of science, not the science itself.

MOND will probably be the leading explanation of dark matter phenomena by the time I have grandchildren old enough to study the topic in a serious and scientific way, at the very least. I'm confident of this because, unlike high energy physics, astronomy data continues to produce torrents of new data, and when scientists have data rather than speculation and theory to help them answer a question, they usually get the right answer sooner or later.

Effective Population Sizes For Select (Mostly Native American) Populations

The following data is from the Supplemental Materials to a paper examining an ancient Taino DNA sample in PNAS.

It is a good example of the difference between census population and effective population for genetics purposes, and shows the relative levels of genetic diversity in a variety of Native American populations as well as the Han Chinese population. (A separate post on the findings of the Taino paper's main conclusions may or may not follow later.) The number between the population label and the effective population size is apparently the sample size used to make the determination and ranges from one to nine whole genomes per population.

Table S13.​ Effective population size estimates (Ne ) for selected Native American populations. 

Population n N​e 

Aleut 2 2,663
Athabascan 2 1,095
Aymara 1 1,409
Chane 1 3,462
Clovis 1 539
Eskimo 9 1,330
Han 3 19,189
Huichol 1 1,176
Karitiana 3 472
Mayan 2 1,967
Mixe 3 1,383
Piapoco 2 1,023
Pima 2 1,019
Quechua 3 3,191
Surui 2 340
Taino 1 1,634
Wichi 4 698
Yukpa 1 727
Zapotec 2 3,823

A Short History of Demographic Change In Britain

Britain has seen five or more rounds of near total population replacement, in addition to other more modest tweaks to its gene pool (and new cultural eras that had surprisingly little demographic impact).

The most notable less than complete population replacements have been the Anglo-Saxon migration, the Viking migrations, and modern immigration, each of which has been more heavily concentrated in some geographic regions than others. The Normans had little genetic impact outside the British aristocracy.

There is overwhelming evidence of a great Celtic cultural impact, but the demic impact of the Celts was not obviously great. But, there are methodological problems with determining what their demographic impact was on Britain because the Celts would have been genetically and physically quite similar to native Britons. Romans and Punic people meanwhile, definitely had little demographic impact.

1. Pre-Neanderthals Hominin occupation of Britain was intermittent in pre-history.

The first members of the genus Homo in Britain were pre-Neanderthal archaic hominins who had arrived by 814,000 years ago, and were forced out by an ice age about 200,000 years ago, leaving Britain without any members of the genus Homo for the next 100,000 years.

Early pre-Neanderthals inhabited Britain before the last ice age, but were forced south by a previous glaciation about 200,000 year ago. When the climate warmed up again between 130,000 and 110,000 years ago, they couldn't get back because, similar to today, the Channel sea-level was raised, blocking their path.

Homo heidelbergensis arrived in Britain around 500,000 years ago and used Acheulean flint tools, but then left during a severe ice age from 478,000 years ago to 424,000 years ago.  Pre-Neanderthal hominins were then present intermittently for the next 200,000 years or so.

2. Neanderthals Starting around 100,000 years ago, Neanderthals arrived in Britain. Early Neanderthals or "pre-Neanderthals" were also present from 230,000 years ago to around 200,000 years ago.

The Neanderthals were probably weakened by climate factors partially related to a string of extreme volcanic eruptions in Europe and possibly also by improving anatomically modern human capabilities both cultural and individual. Still, Neanderthals in Europe kept anatomically modern humans at bay for about 32,000 years after modern humans expanded to West Asia and South Asia, and at least 82,000 years after modern humans first left Africa.

Neanderthals persisted in Jersey (and probably also Doggerland) until about 42,000 years ago.

3. Cro-Magnon Neanderthals were completely replaced in Britain by the first wave on anatomically modern humans in Europe, the Cro-Magnon, who first appeared in Britain around 43,000 years ago. There was probably some admixture at that time, but most Neanderthal admixture in the Cro-Magnon probably pre-dated their arrival in Britain rather than occurring in situ. Neanderthals and Cro-Magnon typically overlapped in any one place for about 1,000 years at most before Neanderthals were replaced.

Britain was not a refugium during the last big ice age, however. Its entire Cro-Magnon population was eliminated in the run up to the Last Glacial Maximum as glaciers covered Britain. The Last Glacial Maximum was 20,000 years ago.

Very few relic populations in refugia during the ice age that including the Last Glacial Maximum had British Cro-Magnon migrants among them. These refugia had an effective male population as small as 30 men. Thus, any admixture between Neanderthals and Cro-Magnon that took place in Britain was eliminated in the last great ice age.

4. Mesolithic Western Hunter-Gatherers Then, Britain was repopulated in the Mesolithic era by Western Hunter-Gathers like Cheddar Man (from about 9500 years ago) over a period somewhere in the range of about 14,500 to 6,000 years ago. 

Despite coming from a much more restricted gene pool, Western Hunter-Gathers were actually pretty similar in terms of Y-DNA and mtDNA, and even, to a lesser extent, autosomal genetics, to Cro-Magnon population, whose relic populations in refugia like the Franco-Cantrabrian refuge and Italy repopulated Europe from a very restricted founding population after the Last Glacial Maximum, even though direct continuity was absent in Britain.

Also notably, around 6200 BCE, a megatsunami driven by runoff from melting glaciers suddenly flooded an inhabited land bridge between Britain and continental Europe called Doggerland (which had been shrinking with rising sea levels since 9000 BCE). The Dogger Bank, however, an upland area of Doggerland, remained an island until at least 5000 BCE.

5. Neolithic Farmers Around 6000 years ago (about 4000 BCE), in the Neolithic Revolution in Britain, early European Farmers, probably more Mediterranean Cardial Pottery folk than LBK farmers with more direct links to Anatolia, largely (90%+) replaced Western Hunter-Gatherers. Farming supports more than an order of magnitude greater population density than a hunter-gatherer lifestyle does, and as an isolated island with relatively little megafauna after the Last Glacial Maximum, Britain was probably not the most abundant place for Western Hunter-Gatherers to try to survive in, so the hunter-gatherer to farmer population surge may have been particularly great.

One key subtlety genetically is that early European Farmers were themselves a mix of European hunter-gatherers and Fertile Crescent farmers, and the hunter-gatherers that the original farmers admixed with were only modestly genetically drifted from Western hunter-gatherers. So, crude ancestry estimates overestimate the extent to which British or Western European hunter-gatherers admixed with local hunter-gatherer populations.

This said, on the European continent, there was significant enrichment of local hunter-gather admixture following first wave Neolithic collapse before Bell Beaker and Corded Ware people emerged onto the European scene.

But, the first wave Neolithic farming civilization collapsed, in my view, most likely as a result of crop failures from some combination of soil exhaustion due to poor farming practices of first wave Neolithic farmers (repeated everywhere first wave Neolithic farmers went) and climate, returning Britain to a predominantly hunter-gatherer-herder society with a much lower population density. There is some evidence a wave of plague that swept Europe at this time as well, but disease is often an effect rather than a cause of famine.

6. Bell Beaker People The semi-hunter-gatherer/herder ancestors of the first wave Neolithic Britons were then almost entirely replaced or overwhelmed demographically (93%+), in perhaps a few centuries or less (in a period starting around 2400 BCE and ending before 2000 BCE, see also here suggesting 2500 BCE to 2100 BCE), by the Bell Beaker people who brought a more sophisticated Copper/Bronze age farming package with them that endured.

The Bell Beaker people who colonized Britain were genetically very similar to the Bell Beaker people of Continental Europe, with significant steppe ancestry and regionally specific Y-DNA R1b clades and mtDNA H clades, rather than like the Iberian Bell Beaker people who were genetically more similar to the Neolithic people of that region with only a sprinkling of the steppe genetic ancestry that is predominant in other European continental Bell Beaker people. This was a quite surprising discovery, because in terms of ceramics and other physical relics, the Bell Beaker culture appears to have originating in Iberia, and in particular in Portugal, which is the least like European continental Bell Beaker people genetically.

Also, despite the heavy rate of population change associated with the appearance of the Bell Beaker people, both in Britain and in Europe, the change must not have been complete, because there was, for example, a continuity of architectural styles and religious practices to some extent, between the descendants of the first wave Neolithic people and the Bell Beaker people. For example, Stone Henge was built by the Neolithic people, but their Bell Beaker successors continued to use it.

The Bell Beaker colonization of Britain was its last nearly complete population replacement through the present. Today's British people are on average perhaps 80% identical to the Bell Beaker people genetically (in terms of ancestry percentages, which ignore the large portion of the genome in which all humans and all Europeans which are basically fixed; in a raw, model independent genetic overlap the percentage similarity is much, much higher) with Germanic admixture making up the balance. The successive Neolithic and Bell Beaker waves of replacement left only about 1% or less of the British gene pool attributable to the Mesolithic Western Hunter-Gatherers of Britain. Previous estimates from the early 2000s that concluded that most British ancestry was traceable to the Mesolithic era, or to the Neolithic revolution in Britain, have been revealed by ancient DNA evidence to be incorrect.

There was significant population exchange and trade between Bell Beaker Britain and Bell Beaker areas in continental Western Europe for pretty much the entire Bell Beaker area until Bronze Age collapse (ca. 1200 BCE).

In terms of physical traces of culture ,and probably language as well, based upon the time depth of the Celtic languages, the Bell Beaker derived cultures collapse, and are replaced by recognizably Celtic cultures, within a few centuries of Bronze Age collapse, which was a climate driven collapse of many cultures over a geographic range from Britain to Egypt to the Indus River Valley (at least).

The language(s) spoken by the Bell Beaker people of Britain remains an open issue that may never be definitively resolved. The oldest historically attested linguistic layer in Britain is Celtic with discernible impacts from later populations that were historically present in Britain that are discussed below.

7. Celts, Romans and Punic People Subsequently, the Celts (coinciding with the British Iron Age ca. 800 BCE), Romans (43 CE to 410 CE plus an earlier invasion in 55-54 BCE), and heavy maritime trade handled by Punic people from Northwest Africa (Iron Age ca. 1100 BCE to early Middle Ages ca. 800 CE) arrived in Britain. Each of these peoples had a powerful toponym impact and a significant cultural impact on Britain, but none of them had much of a long term population genetic impact. The demographic impact of the Celts was fairly minor (although difficult to determine by genetic means because continental and British Bell Beakers people were genetically very similar and exchanged people for the entire Bell Beaker era) and the Romans and Punic traders had only a negligible long term demographic impact on Britain. 

Only about 4% of the population of Roman Britain was Roman, and some of those would have been foreign soldiers on a short term tour of duty, rather than permanent settlers.

Roman Britain had an estimated population between 2.8 million and 3 million people at the end of the second century. At the end of the fourth century, it had an estimated population of 3.6 million people, of whom 125,000 consisted of the Roman army and their families and dependents. 

The urban population of Roman Britain was about 240,000 people at the end of the fourth century. The capital city of Londinium is estimated to have had a population of about 60,000 people. Londinium was an ethnically diverse city with inhabitants from across the Roman Empire, including natives of Britannia, continental Europe, the Middle East, and North Africa. There was also cultural diversity in other Roman-British towns, which were sustained by considerable migration, both within Britannia and from other Roman territories, including North Africa, Roman Syria, the Eastern Mediterranean, and continental Europe.

The limited Roman demic impact coincides with the shallowness of its cultural impact. For example, while Christianity arrived in Britain first from the Romans, as did Roman law, both nearly completely died out, with Christianity re-emerging from Irish and continental European missionaries, and Roman legal concepts returning only with the Norman elites.

The permanent Punic population appears to have been confined to expatriot neighborhoods in some select port cities.

The certainty with which we can say that the demographic impact of the Celts was small, however, is fairly weak. As noted above, the significant population exchange between Britain and Western Europe in the Bell Beaker era means that the population genetic makeup of the Celts may have been very similar to that of the British, to the point where limited ancient DNA samples, and population genetic studies of modern samples three thousand years removed, may not be able to discern the differences between the populations as distinct ancestral groups. If invading Celts were genetically very similar to Bell Beaker era Britons who were invaded, even a major population shift might have been almost invisible.

In the past, I have estimated Celtic demographic impact by assuming that the Celtic elite was initially mostly Y-DNA R1a and was a male dominated migration, and then assuming that the percentage of men with Y-DNA R1a (of the Northern European clade) in Western Europe is roughly half the percentage change in population due to Celtic migration. This was supported by ancient Urnfield Y-DNA R1a. Y-DNA R1a rates in men range from 0% to 8% in the British Isles, suggesting an average on the order of 4% which in turn suggests a 2% population turnover with the Celts.

But, the trouble with this hypothesis is that the R1a distribution in the British Isles appears to be a better fit geographically for Angles, Saxon and Viking demographic impacts than it does for Celtic demographic impact. Ireland, Scotland and Wales, which should be higher than average if Celts are the source of Y-DNA R1a against a Bell Beaker R1b source, are actually lower in Y-DNA R1a than England (Ireland is about 1% and Wales is 1%-2%) except on islands where maritime invaders would have had an edge, suggesting that Y-DNA R1a in Britain has a source that is more likely mostly Germanic than Celtic. Likewise, the Y-DNA R1a frequency in France which was historically Celtic before Romance languages replaced Celtic languages (excluding French Basque for which the percentage is 0%) is only about 2%, again disfavoring a hypothesis that even Celt elites had Y-DNA R1a.

Ancient dental remains also support a primarily cultural diffusion model of Celtic culture, rather than a mass migration, although if the populations are genetically and physically similar, that degeneracy may also be hard to resolve,

This doesn't detract from, however, and indeed reinforces, the possibility that degeneracy in population genetic makeup between the Bell Beaker people and the Celts could cause us to underestimate to the extent to which the Celtic cultural transition in Britain involved a mass migration of people from Europe to Britain.

It is also notable that: Celtic parts of the U.K. (presumably Scotland, Wales, Northern Ireland), have more steppe ancestry than Southern and Eastern England proper, presumably because Norman invaders ca. 1066 CE had less steppe ancestry than the pre-existing residents of the U.K. The modern residents of England proper also have less steppe ancestry than Anglo-Saxon ancient DNA. Keep in mind, however, that this is a subtle difference that is discernible only because of a huge sample size (N=113,851) in a generally very homogeneous population.

The hypothesis that the Norman invaders had less steppe ancestry is consistent with the evidence that the Bell Beaker people, who had significant steppe ancestry, almost fully replaced the population in Britain but less fully replaced populations in Continental Europe, the balance of whom would have been predominantly early European farmers with little or no steppe ancestry, who were most similar to modern Sardinians and Basque people.

8. Anglo-Saxons, Vikings, Normans and Jews. There were subsequent waves of Angles and Saxons (early Middle Ages after the fall of Rome ca. 400 CE) who are the source of the Germanic Old English language, Vikings (late first millennium in the middle Middle Ages ca. 865 CE or 800 CE-950 CE) with a lasting genetic impact mostly limited to the Orkney Islands and a few other coastal and island localities in Northern Scotland and between Britain and Ireland, and Normans (in the late Middle Ages, conventionally 1066 CE) whose Norman French influences caused the transition from Old English to Middle English. 

Traces of these migrations are visible in modern British regional population genetics, despite the fact that Britain is actually very homogeneous in terms of population genetics, due to the large sample sizes and precision genetic sampling of individuals whose genomes were sampled in the latest genetic surveys of the British people, and despite the fact that Angles, Saxons, Vikings and Normans are all genetically only subtly different from the pre-existing mostly Bell Beaker and Celtic derived populations of Britain.

Anglo-Saxon demic impact may have been as high as 38% in Eastern England, although it declines with distance from that epicenter (other regional estimates are in the 10%-40% range with considerable regional variation).

As noted here: 

A study into the Norwegian Viking ancestry of British people found that there is evidence of particular concentrations in several areas; especially in Lowland and Eastern Scotland - and the North Sea islands Shetland and Orkney, Western Scotland and the Western Isles including Skye in Scotland, Anglesey in Wales, the Isle of Man and the Wirral, Mid-Cheshire, West Lancashire and Cumbria in England.

The percentage of modern British ancestry attributable to the Normans is more slippery to determine, and although it is not zero, it is closer in order of magnitude to the Viking and Roman contributions to British population genetics which are small. An article in the New York Times from 2007 references some historical information regarding this issue:

Dr. Oppenheimer . . . cites figures from the archaeologist Heinrich Haerke that the Anglo-Saxon invasions that began in the fourth century A.D. added about 250,000 people to a British population of one to two million, an estimate Dr. Oppenheimer notes is larger than his but considerably less than the substantial replacement of the English population assumed by others. The Norman invasion of 1066 A.D. brought not many more than 10,000 people, according to Dr. Haerke.

This would suggest a Norman genetic impact of about 0.5% or less on the British gene pool, which would make it almost impossible to discern outside the British aristocracy, many of whom hold hereditary titles traceable to those 10,000 or so Normans.

Britain does not have anything approaching the endogamous caste features in its gene pool that India does, but there are subtle enhancements of Norman ancestry in the British upper classes and there are some very subtle but traceable genetic connections between the modern British upper classes/lower classes and their ancestors many generations earlier, with corresponding traces in surnames.

The earliest evidence of Jews in England is from shortly after the Norman invasion in 1070 CE. Also notable from a population genetic perspective is the fact that in the post-Norman era in Britain (as a result of attitudes associated with Norman involvement in the Crusades) there was, in 1290 CE, an expulsion of every Jew from England and Gascony (4,000-16,000 people), except 128 Jewish converts to Christianity in a single communal building in London (following multiple prior massacres), by the Normans, which left the entire region without any Jews for the next 365 years (i.e. until 1655 CE), more than a century after the Church of England broke from the Roman Catholic Church in 1534 CE so the King could get a divorce.

9. Irish Travelers.  "Travelers" in Ireland, who emerged in the modern era (ca. 1650 CE), while culturally similar to European Roma with South Asian ancestral roots, are genetically pretty much indistinguishable from other native of Ireland. They make up about 0.1% of the population of the U.K.

10. Modern Era Migration. Modern era immigrants from the British empire began to arrive starting around the 16th century CE and has had a modest demographic impact on the British gene pool, at least regionally, but has never come remotely close to total replacement. 

Frequently, researchers try to screen out modern immigration when characterizing a country's gene pool, but in Britain this is more problematic than in most places, because Britain's maritime capability has been a global population draw (from Europe to India, Indonesia, Africa and China) for people from its empire for five or six hundred years, which is long enough to make these introgressions part of what is the indigenous Britain gene pool at this time.

Modern global migration have probably contributed more to the British gene pool, for example, than Western hunter-gatherers like Cheddar man have, and also more than the Romans or the Vikings or the Punic people. The impact of modern migration in Britain is closer to the population genetic impact of Anglo-Saxon migration to Britain in its magnitude and timing.

For example, about 14% of the current population of the U.K. is foreign born, and that percentage has never been less than 4% in the post-World War II period. About 13% of the population of the U.K. is non-white. About 7% of the population of the U.K. is Muslim, Hindu, Sikh or Buddhist. The largest share of the foreign born population is South Asian, followed by Chinese. About two-thirds of foreign born citizens of the U.K. have "Asian" ancestry and about a third have African ancestry (often Afro-Caribbean). These are rapidly growing sectors of the population relative to the native born population (mostly due to immigration). Several percent of people in the U.K. are native born and are not of British or Irish ancestry, roughly in accord with the historical 4% of the population was the foreign born in earlier times.

Modern migrants have settled mostly in Greater London (which has about 47% of the population that is white from the U.K. or Ireland, and about 60% white of any kind) and a few other major British cities.

UPDATE March 12, 2018:

A chart showing the population of the regions of the British Isles since the Norman Invasion: