October 06, 2009

Settlement of Americas: one migration with recurrent gene flow

Molecular Biology and Evolution, doi:10.1093/molbev/msp238

A statistical evaluation of models for the initial settlement of the American continent emphasizes the importance of gene flow with Asia

N. Ray et al.

Abstract

While there is agreement in that the Bering Strait was the entry point for the initial colonization of the American continent, there is considerable uncertainty regarding the timing and pattern of human migration from Asia to America. In order to perform a statistical assessment of the relative probability of alternative migration scenarios and to estimate key demographic parameters associated with them, we used an Approximate Bayesian Computation (ABC) framework to analyze a dataset of 401 autosomal microsatellite loci typed in 29 Native American populations. A major finding is that a single, discrete, wave of colonization is highly inconsistent with observed levels of genetic diversity. A scenario with two discrete migration waves is also not supported by the data. The current genetic diversity of Amerindian populations is best explained by a third model involving recurrent gene flow between Asia and America, after initial colonization. We estimate that this colonization involved about 100 individuals and occurred some 13,000 years ago; in agreement with well established archeological data.

Link

28 comments:

terryt said...

"The current genetic diversity of Amerindian populations is best explained by a third model involving recurrent gene flow between Asia and America, after initial colonization".

That is the most sensible comment regarding America's settlement I've ever seen. Probably true of everywhere else as well.

eurologist said...

You beat me to it. Just common sense - same for out-of-Africa, and same for into-Europe. Of course, the details on the number of people involved and the time frame may be quite different.

For example, the high diversity of SW Asia and India, the separate haplogroups of Egypt, Arabia, and the Levant, and the much lower diversity of Europe is easier to explain if South India continued to receive more population input while a separation into a northwestern population was already under way.

German said...

As some readers of this blog know, I've developed the 4th scenario for the origin of American Indians (see Dziebel, The Genius of Kinship, 2007). It's not one migration in, not two migrations in, not one migration with gene flow, but two migrations out of the Americas (at 40+K and at 10-12K) with no gene flow between Old World and New World but with a lot of range expansions and gene flow in the Old World.

Unfortunately, geneticists continue to repeat the same old mistakes in the interpretation of genetic data, and this team is no exception. Only an interdisciplinary alignment of data can give us a good idea of prehistoric migrations. American Indians harbor 2/3 of world linguistic diversity, their kinship systems (my primary expertise) have preserved ancestral structures better than Old World kinship systems and their culture and mythology constitute a very diverse pool of forms (motifs, etc.) showing very specific links with such widely separated areas as Papua New Guinea, Australia, Europe, Dravidian-Munda India (not to mention Asia, of course). This is inconsistent with any recent discrete migration out of Siberia with or without gene flow.

Finally, American Indian population genetic structure seems to better preserve original Pleistocene demography: small, isolated demes subject to genetic drift and lineage extinction. American Indian populations show the highest Fst levels on a worldwide scale (Tishkoff et al. 2009), they are very divergent from Old World populations (Hofer et al. 2008) and show little-to-no population growth (Zhivotovsky 2003). This is why they have lower levels of genetic diversity. It's not because they're recent but because they've been small for a long time. Africa has low levels of linguistic diversity (only 20 independent genetic units as opposed to 140-150 in America), derived kinship systems associated with large population size, low levels of diversity in mythological motifs and signs of population amalgamation that explain high levels of genetic diversity better than "great age" (Poloni 2009).

Matt said...

German,

Why do you think Amerind linguistic and kinship diversity is analogous to a long term buildup of neutral diversity removed by a bottleneck, rather than genetic drift (created and supported by small, subdivided populations)? New Guinea would suggest that linguistic diversification is drift-like, supported by small and subdivided populations. Do you actually have good reasons to support the contrary position?

If there was an Out-Of-America scenario, then it would be a more complex model which would have to involve absorption of the "wave of advance" by Eurasian and African populations (continuously accreting genetic diversity back to Africa), and a bottleneck in South East Asia/Australia (to explain the pattern of genetic diversity). It's unclear to me what would drive this wave of advance, as no cultural advances (which have subsequently swept) or anatomical changes (anatomical modernity) or means to increase population size (e.g. demic diffusion of agriculture) are rooted in the Americas in the archaeological record. Your model would also have the Americas having increased linguistic diversity as evidence they are a source, while having New Guinea having increased levels of linguistic diversity (relative to the mainland) even though they aren't.

terryt said...

"same for out-of-Africa, and same for into-Europe".

And that's why there is so much disagreement as to the actual date of the OoA. In other words there was no sudden expansion of a genetically and intellectually improved (and separate) group.

"American Indians harbor 2/3 of world linguistic diversity, their kinship systems (my primary expertise) have preserved ancestral structures better than Old World kinship systems and their culture and mythology constitute a very diverse pool of forms (motifs, etc.) showing very specific links with such widely separated areas as Papua New Guinea, Australia, Europe, Dravidian-Munda India (not to mention Asia, of course)".

Matt's adequately done the job with the linguistic side. As to the remainder; it could well be true that the indigenous Americans preserve more ancient forms, but surely this is most likely the result of widespread changes within Eurasia. The regions of connection you mention are mostly marginal regions where changes in the centre of the human species' distribution are less likely to reach. Same is probably true for linguistic diversity. Africa is known to have had extensive population movements at times.

Gioiello said...

Of course I don’t believe to an “Out of America” nor an “Out of Asia” (where are there the most ancient mtDNA like L in Africa?), but that Africa has had back migrations from Asia: hgs. E and J and also R1b1*. My hypothesis is to think again not only to the Afro-Asiatic family as an Asiatic-African one, but also to the Bantu languages, derived from the semi-Bantu languages of Cameroon (where it is the most important concentration of R1b1*). Also Yoruba (and its mythology) has clear traces of a derivation from Egypt, and probably before from further, and further Egypt there is Asia.

pconroy said...

Gioiello,

That's why I suggested some time ago, that West Africans and Bantu in particular may be a hybrid population, comprised of Western Asia males and Pygmy females for the most part - thus giving them higher genetic diversity??

terryt said...

"That's why I suggested some time ago, that West Africans and Bantu in particular may be a hybrid population".

I'm reasonably sure that all populations are actually hybrid populations. In fact the phenomenon goes right back to Australopithecus times. I'm certainly very happy to accept substantial gene flow back into Africa at times, again probably going back to when H. erectus first emerged.

German said...

There's no contradiction between America being composed of small, sub-divided populations subject to drift and them being old populations. (Compare the crux of the out-of-Africa model is that African populations are the oldest and largest in terms of effective population size.) On the linguistic side, Amerindian linguistic diversity is associated with archaic grammatical structures (Nichols 1992), whereas limited linguistic diversity in Africa and Europe is associated with derived grammatical structures. Archaeology documents that Africa and Europe were regions of population growth from 45-40,000 YBP on. This explains the high levels of genetic diversity and low levels of linguistic diversity in these areas. In genetic literature, Amerindian populations are often used as a model for the earliest human population structure. Compare: "South American Indians are considered a reference for microsatellite variation in an ancient African ancestor because their population size is low and might be compared with that estimated for an African ancestor" (Zhivotovksy 2001, 704-5, with reference to Cavalli-Sforza)
Papua New Guinea is another area outside of Africa that confirms that populations outside of Africa are older than African populations and that Amerindian linguistic diversity and population structure are not result of recent local evolution.

As far as the supposed antiquity of L lineages is concerned, there's no direct evidence that they're ancient. Ancient remains from Taforalt (North Africa) at 12,000 YBP showed no L lineages. Cro-Magnon DNA at 28,000 YBP shows no evidence of L lineages either, although one would expect to find L lineages outside of Afica at low frequencies in modern populations and at higher frequencies in ancient remains outside of Africa. That's all we really know. The rest are speculations.

L lineages are not found in extant populations outside of Africa, especially in such test regions as Australia or Andaman islands. There's no evidence for early offshoots of M and N lineages in Africa. L lineages are African specific and autapomorphic, hence probably the result of local African development. Whether the rate of mutation is higher in African lineages, or African diversity is the result of early bottlenecks, population expansions and continent-wide gene flow is for geneticists to figure out.

Out of Africa is a hypothesis that needs to be proven using interdisciplinary means. Unfortunately it has been marketed as an example of infallible science. I would encourage to consider other options.

Gioiello said...

I thank you for your response. I am not a conformist or an anti-Copernican. You are a serious scholar and I think your theory merits to be examined and discussed, but you must give a new theory about mtDNA if L isn’t for you the origin. Actually it is hardly believable that American languages are differentiated within 15,000 years, but little tribes occupied a large space, differentiating themselves and not having contact except those which remained close. Anyway if they arrived there before, we should find some rest of them. Then any theory is legitimate, but what resolves are proofs.

terryt said...

"On the linguistic side, Amerindian linguistic diversity is associated with archaic grammatical structures (Nichols 1992), whereas limited linguistic diversity in Africa and Europe is associated with derived grammatical structures".

Isn't the lack of diversity in Africa and Europe likely to be the result of more recent expansions of single or closely related languages? As for the linguistic diversity of America. That may be more a matter of perception than fact. Everyone is still more than happy to accept Greenberg and Ruhlen's cassification of African languages, but there is widespread unwillingness to accept their classification of American languages. Why the diiference? They used the same method for both sets of languages.

German said...

Gioiello: your point is well-taken. Any hypothesis needs to be proven. I'm working on developing an optimal interpretation of the genetic evidence. I haven't settled on any single one yet. However I don't see L lineages as that of a problem. Many Amerindian foraging populations are more diverse than African hunter-gatherers such as !Kung (see Ward et al. 1991). It's the large African agriculturalists and pastoralists that skew worldwide distribution of variation toward Africa. If you toss off mtDNA lineages most frequent among Niger-Congo speakers (the most recently expanded populations), Africa will lose its specificity.
The African region is an amalgam of 20-40 small populations that went through bottlenecks, as they entered the continent, and then underwent a series of re-expansions with a widely ranging gene flow. It's possible that 9 bp deletion frequently found against the L background among the Pygmies (see Soodyall et al. 1996) is an old mutation shared with Asia and America.

Terry: In fact Greenberg's classification of African languages has been disproved. Now, linguists work with 20 independent genetic units in Africa instead of Greenberg's 4. See "Language Ecology and Linguistic Diversity on the African Continent" by Dimmendaal in Language and Linguistics Compass 2/5 (2008): 840–858.

So, we have 140 language families and isolates in America vs. 20 in Africa. Both classifications were developed on the basis of a single comparative methodology that has been around for 200 years.

I don't see why chances that Africa experienced a recent reduction of linguistic diversity (within the past 10,000 years) should be considered much higher than those in America or elsewhere. There're large, widely-spread families in America as well (Na-Dene, Algic, Oto-Manguean, Tupi-Guarani, Cariban). The same for Papua New Guinea, where the spread of Trans-New-Guinean family may have obliterated a dozen of older stocks. Moreover, the actual extent of Amerindian linguistic diversity remains unknown as hundreds of languages were wiped out after 1492. Some of them may have been independent stocks. Amazonia has the highest number of language isolates in the world (more than Papua New Guinea). There's a good chance many isolates perished throughout history, including past 500 years.

Kosmo said...

German,

You're suggesting that modern humans are the result of an Out of America expansion starting ~ 40+K ya?

It's been too long since I've been truly shocked by a post. You caught me with that one though. I'm always open to having my preconceptions blown out of the water... but, with respect, I have two issues with your idea that I just can't get past: genes and bones.

I don't see how you could place L anywhere but at the base of the mitochondrial tree. And also, what about the fossil record in Africa? The oldest modern human remains are found there.

German said...

Kosmo,

Thanks for chiming in. I have two doctorates, I participated in archaeological digs, worked in a genetic lab, measured skulls, worked on etymological reconstructions, studied the ways in which cultural beliefs may affect scholarly research, all this on top of my central concern, namely the structure and evolution of human kinship systems. In addition to the book on kinship, I wrote a lengthy post here (http://anthropology.net/2008/05/12/the-genius-of-kinship-human-kinship-systems-and-the-search-for-human-origins/). In this post I outlined the origin and the current state of the out of America idea.

My book is not so much about human origins, as about kinship systems. I'm preparing to write a book on the out of America scenario, and in the meantime engaging in skirmishes with bloggers here and there.

I just don't buy into the conventional out of Africa story. I could never understand how bones and lithics can possibly generate a convincing theory of human evolution and prehistory. I can see that there're 140 language families in the Americas and only 20 in Africa. American Indians are very diverse culturally, they share important cultural variables with the rest of the globe (including Australia and Europe) and have unique and ancestral kinship systems. If I were a student of ancient human dispersals, that's where I would start my thinking. This suggests great antiquity even without much digging around for bones and stones.

In the meantime, archaeology and paleobiology have dominated our perceptions of the human past for more than a hundred years and heavily influenced the interpretation of mtDNA (and later Y-DNA) results. The first mtDNA paper authored by Johnson et al in 1983 suggested not one but two ways to interpret global variation: out of Africa and out of Asia/out of America. Since then, the potential antiquity of Amerindians and the relative recency of Africans have been swept under the carpet, although all the genetic data show the same tree topology as the original paper by Johnson et al. This happened in response to the pressure by the fossils in Africa supposedly telling a linear story of continuity and evolution in Africa, with subsequent replacement in Europe and Asia.

What's being offered as an archaeological proof for this scenario? There's no evidence for replacements in the archaeological record of Asia or Sahul. Instead there's a smooth transition from Mid-to-Upper Paleolithic. European Upper Paleolithic is more complicated, but again a replacement is not the best way to express what's going on. Continuity is a better explanation, according to some recent critiques of the out of African theory. Anatomically modern humans in Africa and Levant aren't associated with modern human behavior (all the tools are Mousterian), plus many of these skulls and pieces of skulls fall outside of the craniometric variation found in existing human populations.

What we seem to know is that only at 45-40,000 YBP there's unequivocal presence of modern morphology and behavior in Europe, Africa, Asia, South Siberia and Australia/Papua New Guinea. Plus a bunch of poorly described and controversial finds in the Americas. But, most importantly, we don't find any archaeological evidence of humans entering America and bringing with them typical Siberian/Northeast Asian lithics. And this is all after 100 years of believing that humans came to America at a late date.

German said...

I continue:

I can go on and on, as you can see. I've done this in the past. If you're really interested in developing an alternative viewpoint on human origins, I can explain everything in a greater detail and cite some literature. If you firmly believe that L lineages are the oldest and that African fossils furnish answers, rather than raise questions, then we may end up wasting our time.

If you have any specific questions, I'll try to answer them. Otherwise, I'll have to ask you a very specific question: why do you believe African lineages are the oldest, while Amerindian lineages are the youngest? I assume you know that allele diversity is a function of effective population size. Why can't you think of Amerindians as a huge refugium that has preserved small population size characteristic of our ancestors, and of Africans as an amalgam of populations that first went through bottlenecks but subsequently re-expanded to achieve the greatest effective population size?

terryt said...

"I just don't buy into the conventional out of Africa story".

I'll confess that I'd guess it was more complicated than usually portrayed, but I believe it's fundamentally correct.

"If you firmly believe that L lineages are the oldest and that African fossils furnish answers, rather than raise questions, then we may end up wasting our time".

I think the evidence that L lineages are the oldest is pretty convincing but ... Do we really know that the woman who first carried mtDNA L0 was already 'anatomically modern', or are we simply assuming that she was? As I said, the situation may actually be very complicated.

"There's no evidence for replacements in the archaeological record of Asia or Sahul. Instead there's a smooth transition from Mid-to-Upper Paleolithic".

I've long been aware of the problem of lack of replacement through much of Asia. In fact I've just recently failed to convince Maju of this, in spite of my providing a large amount of evidence in favour of it. On the other hand he is completely prepared to accept continuity within his beloved India going a long way back. His solution to the problem is to argue that Y- and mtDNA haps are much older than usually claimed, although Dienekes has argued convincingly that the opposite is true.

"Anatomically modern humans in Africa and Levant aren't associated with modern human behavior (all the tools are Mousterian)".

Not completely relevant. Technology has often been relatively independent of genetic change. And I don't think anyone would argue that the Upper Paleolithic began in Africa, or is necessarily associated with the first expansion of modern humans.

German said...

These are good caveats, Terry. However we do need to have a recognizable, recurrent pattern to talk about things. Lack of correspondence between morphology and behavior at the crucial point of argument for out of Africa deprives this theory of facticity. Yes, we can hypothesize that morphology came first, behavior came next, but we have to add an ad hoc interpretation to the data in order to make out of Africa work. I'm looking for a more factual and parsimonious way of explaining things when you don't need to read an interpretation into data in order to prove a theory. I don't care what kind of data it is - archaeological or linguistic/kinship - but it should be data, not our speculations. If the behavioral transition to modernity happened in Africa, we would expect to find a greater linguistic and cultural diversity there. But in fact we find it on the other end of the world - in America.

What I hypothesize (and this may change) is that at 45-40K a population different from "anatomically modern humans" advanced into Europe and Africa from the east. This population was very cerebral: it spoke our languages, had kinship systems and other means of classification, and told lots of myths around their campfires. They relied on soft technologies (bones, snares, etc.) to procure food and build shelters. They replaced Neanderthals and "anatomically modern humans" but also adopted a lot from their lithic inventory thus creating an illusion of lithic continuity. The origin of these true Homo sapiens sapiens was in America based on the data coming from modern populations (genetics, linguistics, kinship systems and mythological data). (In the end Homo sapiens sapiens is probably an offshoot of Asian Homo erectus that sneaked across the Bering Strait around 100K and speciated there. (An ideal place to speciate, gentlemen, isn't?) At about 50K it migrated back into the Old World as "us." For the genetic support of this interpretation see Reed et al. 2004 on lice DNA.) The craniological and dental continuities between Asian Homo erectus and modern Asian populations emphasized by Wolpoff and more recently Habgood are found at high frequencies in modern Native Americans and are attested in ancient remains such as Kennewick man. (We do need more archeological and craniological data from Asia and America to substantiate/test this claim.) In Europe and Africa we find the progressive decline of these Asian-American craniomorphological traits. The pattern of decline is consistent with the same pattern of trait decline observed in kinship terminologies.

What archaeology is telling us is that Africa and Europe, from 45-40K on, experienced population growth. We don't see a signal of population growth in Africa before this. Again, one would expect the onset of demographic expansion in Africa to predate Europe or Asia. We would expect this growth to be visible in the archaeological record of Africa from 100K on. We don't have it. See Klein, in Evolutionary Anthropology 2008.

Alternatively population size and density in Asia, Australasia and America was depressed until 20-10K. This is consistent with migration to Africa and Europe with subsequent re-expansion and with lack thereof in Asia, Australasia and America.

In a word, I think contemporary studies of human origins are essentially wrong (not out of Africa but out of America) but they're accurate in a great number of details. One just need to identify where data is solid and where it's not.

Kosmo said...

"I have two doctorates, I participated in archaeological digs, worked in a genetic lab, measured skulls, worked on etymological reconstructions, studied the ways in which cultural beliefs may affect scholarly research, all this on top of my central concern, namely the structure and evolution of human kinship systems."

-- German, well, as far as answers go, that has to rank right up there as one of the best starts ever. I'll give you that. Coupled with everything else you've written, it's obvious that you really know your stuff.

I will certainly not discount your ideas, (Though the mitochondrial tree is still a sticking point. It's not doctrine for me, but the way the mutations unfold, building one on another, seem to lay out a pretty convincing path from L upward.)

I will admit that I, too, have had issues with apparent continuity in Asia.

German said...

Kosmo: "Though the mitochondrial tree is still a sticking point. It's not doctrine for me, but the way the mutations unfold, building one on another, seem to lay out a pretty convincing path from L upward."

German: This is a very good formulation, Kosmo. If everything else lined up nicely, I wouldn't question it. But I think it's good to offer a different perspective on things - a single origin but not out of Africa - so that scholars routinely test both scenarios as they're working through their data. Judging from the fact that in so many publications Amerindians aren't even discussed (because people assume we know when and how America was peopled), I don't know what the mtDNA tree would've looked like now had archaeology made a case for the antiquity of Amerindians. Amerindians are very diverse in a narrow set of loci (see Ward et al. 1991 and Bandelt 2007 with coalescence dates for the lineages found in a single Amerindian tribe up to 80,000 years ago) and very uniform in a broad set of loci. Could we read this pattern in the opposite way from the conventional view, since we have good support for Amerindian diversity furnished by linguistics?

The Y-DNA tree is even more controversial than the mtDNA tree, with derived YAP+ E haplotypes found in more than 50% of African Y chromosomes and with a possibility that YAP+ originated in Asia where DE is found next to D, C and F clades, which are non-African. If the majority of African Y-DNA types came from Asia, then we're only left with A and B largely restricted to Khoisan and Pygmies as exclusively African and potentially archaic. Again lineages loss in small populations in Asia and America and lineage retention in Africa could explain this rather plausibly without even reversing the tree.

German said...

I continue:

Finally if we look at X chromosome (bi-parental, slow mutating, larger effective population size), the phylogenetic tree further loses indicators of an African origin. The basal haplotype B006 is found at high frequencies in Amerindians and at low frequencies in Africans. This is the same pattern observed in the distribution of the so-called "nuclear insert" in human mtDNA (Zischler et al. 1995). This insert was used most recently as an out-group for the Australian Mungo man mtDNA from 40,000 YBP (Adcock et al. 2001). The Mungo man sequence was more divergent than African L lineages, although the study hasn't been replicated, hence remains questionable.

So it's possible that mtDNA represents recent events better, while Y-DNA and especially X chromosome are better in preserving the earliest stages in the genetic history of our species.

Kosmo said...

Thanks, German. You've explained your views very clearly, and I can now see the logic that you are using to inform your perspective.

In the end, like a lot of long-standing anthropological disagreements, this one will have to come down to Occam's Razor-- but where that razor cuts may be different for different people. Thanks for posting your ideas. I found them fascinating.

I'm particulary glad to see you brought up Mungo Man. I admit I was wondering about your interpretation of his DNA results--being as they have been essentially thrown out for no other reason than their wild divergence from expectation.

German said...

Well put. Thank you, Kosmo.

terryt said...

"The origin of these true Homo sapiens sapiens was in America based on the data coming from modern populations".

Isn't it more likely that the origin was somewhere in North/Central Asia? I don't think there's any evidence for human presence across the Bering strait going back 100k, is there? Whereas there is certainly evidence for human presence at least 55 degrees north through Central and East Asia going way back beyond then. To alter your own comment a little, 'lineage loss in small populations in Asia and lineage retention in America could explain this rather plausibly'.

"What I hypothesize (and this may change) is that at 45-40K a population different from 'anatomically modern humans' advanced into Europe and Africa from the east".

I could go along with elements of that. In fact it ends up being remarkably similar to the 'regional continuity' theory. In other words human evolution is a product of backwards and forwards movements going back to when H. erectus (or something similar) first emerged. There was no sudden expansion of anatomically modern humans, just a series of genetic, cultural and technological expansions. It's ultimately impossible to draw a line and say, 'that's when we first became modern human'. Hence the huge disagreement as to when 'anatomically modern humans' actually emerged from Africa.

German said...

Terry,

This may be the next version of my thinking but as of now the key aspect of out-of-America is the single origin idea. Out of America is a cross between Multiregional (regional continuity traits in Asia and their decline in Europe and Africa) and Out of Africa (genetic variation has a clinal shape connecting Africa and America). But I deny continuity in Africa and suggest replacement from America - two point in which I challenge both out of Africa and Multiregional.

There were several range expansions in the Old World going east and west, north and south but these happened after humans had left America. Hence, lower levels of allele diversity in America, which is a function of isolation and non-participation in wide ranging gene flow in the Old World.

There're controversial sites in America going all the way to 250,000 YBP. I don't base my inferences on archaeology, though. Darwin didn't have fossils at his disposal but he predicted that they would be found. I think the archaeological record in the Americas is complicated by a whole host of heterogeneous factors ranging from 1) cultural biases against Indians being an old population to 2) the use of European Paleolithic as a measuring stick in a region that may have involved an earlier stage in the evolution of human adaptation and that may have relied on perishables more than on lithics to finally 3) the low population size and and density of Amerindian populations.

There's a clear gradient of visibility of "modern human behavior" in the archaeological from Africa and Europe to Australasia, Asia and especially America. However, as the linguistic map attests, it's not the matter of Europe and Africa being older than Asia, Australia and America. It's the function of the relative exposure of incoming Homo Sapiens sapiens to the lithic traditions of indigenous hominins. There's a notable delay in the emergence of modern lithic technologies in Australasia and especially in America. Clovis points are a good example: essentially the same points as Solutrean in Europe, only 6-15K later. I attribute it to the differential role of pre-Homo Sapiens sapiens hominins (Neanderthals and Anatomicals) in the evolution of the Paleolithic human tool kit. In Europe and Africa, Homo sapiens sapiens learned a lot from the indigenous hominins, while in Asia these encounters were sparse. In America and Australia, there were no hominins to learn from, hence these are the two areas with the greatest delays in the emergence of lithic tools.

terryt said...

A few weeks ago someone posted a link on this blog which I followed up at the time. Sorry I can't remember what, who or where. But as a side issue the authors looked at the genes for Rh+ and Rh-. The particular link set me to thinking along your lines, because the author's interpretation of the series of mutations required seemed unlikely to me.

The gist of the relevant part of the article was that, because the authors assumed an African origin for the gene combination concerned, they were forced to claim that the mutation Dce was the original. From that assumption they were forced to derive a complex mixture of both dominant and recessive mutations to account for the modern situation.

I know from years of being associated with livestock and poultry breeding that genetic mutations are usually, in the first instance, recessive. If they were dominant it would make such breeding so simple. Oh that it were so. Anyway I considered the possibility that the original situation had been DCE, that all mutations had resulted in recessive genes and that the recessive and heterozygous combinations have since driven the DCE to extinction.

I quickly realised that we could now account for the present situation with just four simple regional mutations. If I remember correctly we finish up with DcE in Central Asia, DCe in SE Asia and the d mutation occurring in Europe, but having to occur in a gene already containing recessive e. The by now most common Dce mutation may in fact have happened in Africa but it occured on a gene already containing either e or c, presumably the DCe of Central Asia. So the obvious conclusion was that a population containg DCE had been spread from Central to SE Asia, but not necessarily through Africa, at an early date.

You may be able to follow up the Rh system and find something useful.

German said...

I also vaguely remember reading a recent paper that gave an example of Rh in support of the out of africa scenario. Thanks for the pointer, I'll look into it.

terryt said...

Silly me. I realised later that Dce need not have involved a separate mutation. It could have arisen through hybridism between DCe and DcE, East Asian and Central Asian populations. It was a hybrid population that moved into Africa, in relation to the Rh genes anyway.

German said...

I can totally see hybridization as common throughout African history. See Hofer 2009 for mtDNA in East Africa: a combination of drift and gene flow between once isolated populations can create an illusion of antiquity. We have M, U, X coming into Africa anyway as part of the conventional theory. 9 bp deletion is frequently found among the Pygmies - this could be another lineage.