Sunday, November 4, 2012

For What They Were ... We Are on H1 in North Africa

Published in 2010 was the paper:

Mitochondrial DNA H1 in North Africa: 
An Early Holocene Arrival from Iberia
PLoS ONE 2010. (Open access (link))

Maju's very interesting comments on this paper are posted here: (link)

I would agree with Maju that the arrival of H1 in North Africa predates the Early Holocene.  I would also agree with comments on the association of H1 in North Africa with R-V88. 

What I would note is that the distribution of R-V88, with its association with Italy, Corsica, Sardinia and Africa, suggests a Strait of Sicily crossing rather than Iberia.

Regarding Maju's statement that the genesis of the Oranian culture is of an H1/R-V88 origin, I note that the transistion between the Aterian and Ounanian in North Africa is a current topic of interest regarding human origins.  A recommended reference is:

Modern Origins:  A North African Perspective
Editors:  Jean-Jacques Hublin, Shannon P. McPherron
(link)

Update:  The Oranian (wiki link) lithic culture to which Maju refers is not the same as the Ounanian lithic culture that is described in the Drake et al paper (link).  See comments in this post.  I believe the Oranian to which Maju refers is discussed as part of the late Aterian culture in the chapter "The Identity and Timing of the Aterian in Morocco" in the "Modern Origins: A North African Perspective", which is written by A. Bouzouggar and R. N. E. Barton.  That is not to say that the Aterian, Ounanian and Oranian are not related.

22 comments:

  1. Why "Ounanian" (sic)? Never read that before. First I though it might be Arabic for Oran but it's actually Wahran.

    The culture was known initially as Iberomaurusian because the discoverers though it was the product of an Iberian Solutrean migration or influence into North Africa. Later it was argued that it could originate in Egypt instead and a new name proposed (that was Oranian because there are several sites near Oran - not very imaginative but more conceptually neutral than Iberomaurusian for sure).

    But eventually C14 revealed itself stubborn in proclaiming Western sites in Morocco as the oldest ones (what totally demolishes the Egyptian origin hypothesis). The genetic discoveries of this decade, with some 30% of North African mtDNA probably derived from Iberia or France (nearly all H found in NW Africa: H1, H3, H4 and H7, and probably also V) also seem to support an Iberian origin for the Oranian because:

    1. It is very difficult to imagine another period of major genetic flow from SW Europe to NW Africa (Megalithism but it does not fit well for other reasons).

    2. The relative lack of European Y-DNA (although more common in the Canarian mummies) suggests a later Africa-born wave that replaced more intensively the male lineages than the female ones. That one was probably the Capsian, this one indeed originated from the Nile area.

    But I think that R1b1c-V88 (the Central African and Sardinian clade) has nothing to do with Oranian but actually spread from West Asia to both Balcans-Italy and Sudan-Central Africa in an unrelated process. It is R1b1a (P297, L320) which is the one residually found in NW Africa, more than V88. However it only makes c. 10% of the modern Y-DNA pool in the region, a mere fraction of the "European" mtDNA (H and V).

    ReplyDelete
  2. OK. Thanks for the history. I accept your use of the word Oranian.

    My use of the word Ounanian comes from the Drake et al paper:
    http://www.linearpopulationmodel.blogspot.com/2012/10/ancient-watercourses-and-biogeography.html

    But, in fact, Ounanian as defined in this paper, it shown on the southern edge of the Atlas mountains.

    I'm currently reading the Chapter "The Identity and Timing of the Aterian in Morocco" in the "Modern Origins: A North African Perspective", which is written by A. Bouzouggar and R. N. E. Barton. They argue for a continuum of the Aterian in Morocco from 111ka to 20ka. I'm still reading the chapter, but regarding the genesis of the Aterian in Morocco, they have this to say:

    "In our opinion, the tangs and other forms of preparation (eg. bifacially thinning) on Aterian points and other tools are indicative of a developed hafted technology. The origins of this may not lie exclusively in the Aterian. For example, the thinning of points has also been described in the Mousterian assemblage at Jebel Irhoud and elsewhere in North Africa, including at Oued El Akarit, Tunisia. Indeed, this may imply connections between the Aterian and a local Mousterian . . ."

    So there is a very complex, continuous interaction in Morrocco, along the Atlas Mountains, extending east to Tunisia.

    ReplyDelete
    Replies
    1. Actually the description of Ounanian (Saharan Neolithic, arguably linked to Nilo-Saharan expansion) does not seem related at all with Oranian (Upper Paleolithic of NW Africa, largely coastal). If anything I'd relate Ounanian with Capsian (although IMO Capsian culture is Afroasiatic, proto-Berber, in language), not as identical but rather as parallel E->W Epipaleolithic/Neolithic flows.

      "They argue for a continuum of the Aterian in Morocco from 111ka to 20ka".

      I'd say it's very possible because there seem to be some Aterian influences in Oranian and even in the late Gravetto-Solutrean of Iberia. At least one would think that tanged points may be conceptually derived from Aterian (and AFAIK did not exist previously in Europe).

      This has led me to imagine the Solutrean-Oranian episode as possible source for Iberian genetics of North African origin: specially E1b-M81 but also mtDNA lineages like U6 or L(xM,N), all them concentrated in Western Iberia, including Asturias and parts of Cantabria (important UP population areas). The alternative episode would be Neolithic, with Cardium Pottery "bouncing" in North Africa before reaching Portugal.

      "So there is a very complex, continuous interaction in Morrocco, along the Atlas Mountains, extending east to Tunisia".

      I have no problem with that at all. The Oranian actually only begins c. 20 Ka., so it is the next phase. The last phase would be Capsian which should be original from Egypt or nearby areas (Nubia, Libya) and have brought Afroasiatic language and some lineages like J1 and E1b-M78. A re-expansion of U6a can also be associated with Capsian but not of all U6 because the greatest basal diversity is in Morocco, followed by Canary Islands, Iberia, etc. This would imply first an U6 flow from West to East before the backflow, maybe with Oranian.

      Still I have some blanks, notably U6 and the apparent lack of Aurignacoid industries West of Cyrenaica is a problem for my model. If the Dabban industries or some other Aurignacoid offshoot would have reached Morocco or at least Algeria, then the distribution and apparent history of U6 would be easier to explain. Probably the archaeological reconstruction of NW African Prehistory surely still holds some secrets.

      Delete
    2. "Actually the description of Ounanian (Saharan Neolithic, arguably linked to Nilo-Saharan expansion) does not seem related at all with Oranian . . ."

      According to Blake et al, the Ounanian is associated with elegant shouldered arrow points: "The “Ounanian” of Northern Mali, Southern Algeria, Niger, and central Egypt at ca. 10 ka is partly defined by a distinctive type of arrow point (37). These arrowheads are found in much of the northern Sahara (Fig. 3) and are generally considered to have spread from Northwest Africa."

      See also: http://www.aggsbach.de/2011/02/the-tenerian-and-the-vanshing-of-a-green-sahara/

      Regarding lithic definitions in Northwest Africa, sounds like a new project for me. There are honestly *chapters* of different views regarding dating, naming, and classifying Northwest African lithics in this book "Modern Origins: A Northwest African Perspective". So, until I've had a chance to think about it, I am not going to get myself into hot water trying to discuss with you the merits of Aterian vs. Oranian vs. Solutrean for Northwest Africa.

      "The last phase would be Capsian which should be original from Egypt or nearby areas (Nubia, Libya) and have brought Afroasiatic language and some lineages like J1 and E1b-M78."

      Ah, I sense hot water.

      Regarding the "Aterian" and U, we're talking about a time period in Northwest Africa that spans almost 100,000 years. So, plenty of time for progenitors of U to expand north or northwestward across the Sahara into the Maghred and Morocco and then for U to expand northward and eastward.

      "Still I have some blanks, notably U6 and the apparent lack of Aurignacoid industries West of Cyrenaica is a problem for my model." Why the Aurignacian in northwest Africa? I don't see that term used in the literature. The terms used are Mousterian, Aterian, Levallois, Upper Paleolithic and Maghrebine Middle Paleolithic. These are the terms used for lithic industries up to the Last Glacial Maximum.

      The Nilo-Saharan barbed point tradition, the Iberomaurusian, the Capsian and the Ounanian traditions occur after the Last Glacial Maximum, no?

      Delete
    3. R-V88, old R1b1c, spread with the Neolithic, from the Near East or thereabouts. That is recognized now but was not a few years ago. H1 is now thought to have spread with the Neolithic as well - by both the Mediterranean and central European routes. This is not pre-Holocene.

      Delete
    4. Thanks. Could you provide the reference for this.

      Delete
  3. "[Ounanian] spread from Northwest Africa."

    Fair enough. I misinterpreted the paper. Probably because I always though that Nilo-Saharan languages and R1b-V88 had their origin at the Nile, not in the West.

    I have to chew on that now.

    "Regarding the "Aterian" and U, we're talking about a time period in Northwest Africa that spans almost 100,000 years".

    But mtDNA U (U6 almost exclusively in North Africa) is an Eurasian backflow (N > R > U > U6), so it just cannot be Aterian. It must have arrived later on, possibly with the colonization of most West Eurasia (non-Arabian West Asia, Central Asia, Europe and, by extension, parts of North Africa also) c. 55-40 Ka. ago. All Eurasian lineages in Africa (except for the Malagassy) are West Eurasian (also and mainly found in West Eurasia) so they must have arrived with the West Eurasian colonization process or after it.

    U6 is one of the main examples, others are X1, M1, etc. (with different distributions). In the Y-DNA side we have specially J1, T and R1b-V88 if my memory is correct.

    "Why the Aurignacian in northwest Africa?"

    Not "Aurignacian" but Aurignacoid. I and others use the wider Aurignacoid label to refer to a set of cultures that share many elements with Aurignacian but are not it, for example Palestine's Emirian and Ahmarian (arguably the West Asian direct relatives of proto-Aurignacian and Bohunician), also the recently researched Altai "Aurignacian" (not really Aurignacian either but close enough) or the Dabban industries of Cyrenaica, very similar to their Palestinian counterparts for what I've read.

    U6 must be very old. Although the exact date can hardly be estimated IMO, a Neolithic or Epipaleolithic time frame should be discarded. It could still have arrived from Europe with the Solutrean-Oranian episode. But, while Iberia has high diversity of the lineage, it's concentrated in the areas of relatively high North African ancestry in the West, suggesting it arrived from Africa with the other markers (E1b, L(xM,N)).

    So this leaves U6 in Morocco at the beginning of Oranian as native clade. And Aterian cannot be a source - at least not a logical one.

    "The Nilo-Saharan barbed point tradition, the Iberomaurusian, the Capsian and the Ounanian traditions occur after the Last Glacial Maximum, no?"

    Yes. Most of them (all but the Oranian) already in the Holocene.

    What I say is that, based on the presence of U6, with an apparent coalescence area in Morocco prior to the Oranian, I still expect to see "Aurignacoid" industries like Dabban found in NW Africa. I don't think that the region is sufficiently well researched to discard that.

    "Absence of evidence is not evidence of absence" and there are other indications (U6 specially) that do suggest some sort of "Aurignacoid" penetration in NW Africa prior to the LGM.

    But I accept that the evidence West of Cyrenaica is so far missing.

    ReplyDelete
  4. Just so we have it straight, the Ounanian culture, according to Blake et al, spreads southwestward from Northwest Africa sometime before the Holocene Climatic Optimum.

    The Nilo-Saharans spread westward across the Sahara from the Nile near Atbara sometime before the Holocene Climatic Optimum.

    Carriers of R-V88 and H1 cross the Strait of Sicily during the last ice age, but before the Holocene Climatic Optimum, and head south.

    Carriers of the E Y-haplogroup cross from Tunisia to Sicily and other points north during the last age, and also head into the Levant at about the same time. That accounts for E in Europe and the Levant.

    In addition to the above mentioned North and West African groups, there are other cultures, not mentioned, including the Moroccans on the north side of the Atlas Mountains, and the forebearers of the Niger-Kordofanian culture on the Niger and Volta river.

    For clarity, I don't think that U6 is carried by a back migration from Europe. I think it is indigenous to North Africa insofar as something being local to an area for more than 50,000 years can be called indigenous. U in Europe, by the way, arrives at the moment that we see the earliest evidence for modern humans in Europe.

    Regarding Aurignacian ... Solutrean, I just don't see those terms being used by archaeologists working in the field in North Africa. I've put the reference up, so you can check that yourself.




    ReplyDelete
  5. Correction to the above:

    "the Ounanian culture, according to Blake et al, spreads southwestward from Northwest Africa"

    Should read "spreads southeastward from the Northwest of Africa"

    ReplyDelete
  6. Regarding the phylogeny preceeding U (N > R > U > U6), it's an unsettled question as to whether mtDNA "N" and mtDNA "R" have their genesis in Asia or Africa. If you look at the pluvial phases in the Sahara (J. R. Smith, "Spatial and Temporal Variation in the Nature of the Pleistocene Pluvial Phase Environments Across North Africa", Chapter 3 in "Modern Origins: A North African Perspective") you can see that the Northern Sahara was much greener than today in a window beginning about 160kya and ending about 75kya. So plenty of time for the N > R > U expansion through North Africa, some time in that window, beginning in East Africa and heading to the Maghreb/Morocco. Modern humans do not appear in Europe until 55kya. Jebel Irhood, on the other hand, is dated to 160,000 years ago.

    ReplyDelete
    Replies
    1. "... it's an unsettled question as to whether mtDNA "N" and mtDNA "R" have their genesis in Asia or Africa".

      Not at all. There are some very confused (or even maybe dishonest) people, who for what I've seen tend to abound in some Africanist forums spreading their confusion, who make absurd claims on the matter, also for M. But from an empirical viewpoint there is no doubt: the basal diversity of M, N and R in Africa is ridiculously low and always derived from very specific West Eurasian haplogroups.

      The L3 node was in Africa, no doubt, but the M, N and R nodes must be located in Southern Asia (senso lato), possibly in the Indian subcontinent but could also be in some cases SE Asia or (some claim, without any solid support IMO) in West Asia.

      M has 40-something basal subhaplogroups and only a subclade of a subclade of one of them, M1, is found in Africa (shared with West Asia, where it appears to be more diverse). N has some 14 basal haplogroups and none of them appear to have coalesced in Africa (the closest one would be X, which is probably from Levant, X1 may be already coalesced in Egypt), R (the largest N subclade) has some 16 basal subhaplogroups and the only ones found in Africa are those found and dominant in West Eurasia: U, R0 (incl. HV) and JT, all of which appear to have coalesced in West Asia.

      In my understanding all three super-haplogroups coalesced in Southern Asia without doubt.

      A greener or drier Sahara is rather irrelevant for all this, as we should only be looking at where the various basal haplogroups are located.

      It's been a long time since I last discussed this matter but you may want to check my explanations in the following links:

      http://forwhattheywereweare.blogspot.com/2010/10/on-high-mobility-of-mtdna-macro.html

      http://leherensuge.blogspot.com/2010/02/reconstruction-of-mtdna-spread-in.html

      Not only mtDNA and Y-DNA emigrated out of Africa to return much later, notably to North Africa, but also bacteria in our guts:

      http://forwhattheywereweare.blogspot.com/2012/06/h-pylori-also-made-it-out-of-africa.html

      "Modern humans do not appear in Europe until 55kya. Jebel Irhood, on the other hand, is dated to 160,000 years ago".

      Indeed but, honestly if Jebel Irhoud left descendants among modern North Africans, we have yet to spot the genetic markers.

      As you surely know, I have defended repeatedly (in line with Behar 2008 in part) that the presence of some L(xM,N) lineages in North Africa and Arabia date from the time of the OoA (but not before), which should be roughly the time of the formation of Aterian prior to 100 Ka ago; there are even some rare Y-DNA lineages, shared at low frequencies between West and NW Africa, that could be even older (but how old in NW Africa specifically?); I have also spotted a very distinct autosomal component in Southern Morocco that I believe should be an Aterian genetic remnant... but it amounts to 14% among Southern Moroccans and is residual (<2%) among the rest.

      So, unless you show me other evidence, I must state that the genetic legacy from Aterian and older layers is quite tenuous and hard to detect. I do think it is there but I also find it quite minor. The rest is mostly West Eurasian genetics with some Tropical African influence also. North African specific components other than the "Aterian" one, when they show up, are clearly akin to West Eurasian components by Fst distances.

      Also I must add a mention to the latest Pereira paper, which claims, like I do, a West Eurasian origin for the precursor of U6 (undifferentiated U, obviously): http://forwhattheywereweare.blogspot.com.es/2010/12/new-paper-on-mtdna-u6.html

      I don't agree in her denial of all kind of Aterian continuity but on the central matter (U6) I agree fully (except molecular clock, which I always despise).

      Delete
    2. Maju,

      "the basal diversity of M, N and R in Africa is ridiculously low and always"

      We've now gone through two Saharan pluvial phases since the diversification of L3->M,N, so it is hardly reasonable to compare the genetic diversity in Southwest Asia with the genetic diversity of the Sahara. Different processes are at work on different timeframes.

      "A greener or drier Sahara is rather irrelevant for all this, as we should only be looking at where the various basal haplogroups are located."

      Well, I'm willing to keep on open mind regarding the geography of N->R->U.

      I do think that mtDNA M has its locus of expansion in Asia. If you look at the mtDNA tree (Behar 2012), mtDNA M and N split at about 70ky, right at the end of a Saharan pluvial phase. So, probably at this point, M is driven east. N, possibly, takes refuge somewhere in North Africa at this time.

      "Indeed but, honestly if Jebel Irhoud left descendants among modern North Africans, we have yet to spot the genetic markers."

      L1 is present in West Africa and is dated near the dates for Jebel Irhoud. L0 is more common in East and South Africa, but that might only be because L0 is attached to particular hunting strategies that migrated to East and Southern Africa from a West African origin. If you look at the distribution of grassland bovidae in Africa, it shows a continuum from the Atlas mountains, then to west Africa, then to east and south Africa. For very mobile hunter gatherer groups such as the San, it is plausible that they moved out of West Africa to their present locations in East and Southern Africa.

      "So, unless you show me other evidence, I must state that the genetic legacy from Aterian and older layers is quite tenuous and hard to detect."

      Well, the matter is under study by the Max Planck Institute, so I think I will entrust them with the analysis.

      Regarding the connection and timing of U5 and U6, I'd like to read more about this. For now, I will keep and open mind.

      Delete
    3. "I have also spotted a very distinct autosomal component in Southern Morocco that I believe should be an Aterian genetic remnant..."

      Have a look at the eustatic sea level 150 kya:

      http://linearpopulationmodel.blogspot.com/2011/05/refining-eustatic-sea-level-curve-since.html

      The Max Planck researchers are calling Jebel Irhoud "Mousterian":

      http://www.eva.mpg.de/evolution/files/irhoud.htm

      I'm not sure if your genetic remnant has something to do with this, but as the sea level shows, there's probably a pattern of interaction between Europe and North Africa across the Strait of Gibraltar and the Strait of Sicily which allowed the "Neanderthals" to reach Europe in the first place more than 500,000 years ago. An easy window of opportunity to cross back into Africa was 150 kya.

      Delete
    4. It is the number of BASAL haplogroups: the diversity coalesced just under the founding node (N in our case). There's not a single N clade specifically African.

      Also North Africa is not the Sahara (but a fertile Mediterranean region) and should have kept most of the lineages (and in fact did to a small amount with some presumably Aterian genetics but only to a small amount, what means that further major waves of colonization took place over the Aterian substrate).

      "mtDNA M and N split at about 70ky"

      Regardless of age estimates (I strongly dislike the confusion induced by molecular-clock-o-logy, which is not really scientific), pre-M and pre-N split from the L3 node and that happened in East Africa, very probably between the Nile and the Red Sea.

      But, while the other L3 subclades, which coalesced as such in Africa, show signs of immediate re-expansion under L3, often just one mutation away from the mother node, M and very specially N show signs of quite longer coalescence periods: M has three coding region mutations in the stem between L3 and M, while N has five. That means that they had some time before re-expansion, before they became M and N as we know it.

      This extra time is precisely what is needed to more easily allow for a long migration as small groups through non-hospitable lands, like the OoA through an arid and maybe already settled Arabia.

      Delete
    5. "L1 is present in West Africa and is dated near the dates for Jebel Irhoud".

      Fine. But how much distinct NW African L1 exists today? AFAIK zero, at least at top tier levels. What we find is some derived L1, which probably arrived more recently from West Africa.

      Of course further surveys can expand our knowledge but Behar 2008 only spotted: L1c3b2 in Morocco (but also Gabon, South Africa) and L1b1a2 in Egypt (also Negev Bedouins, Ethiopia).

      However, and I'm not sure if this is what you have in mind, the most recent genetic data, which detected many rare L1b1a lineages in Europe (Iberia especially but also other areas).

      Still only one such lineage is found in NW Africa proper: L1b1a9 (shared with Spain, Italy and France). It's truly difficult to imagine with that data in hand a Jebel Irhoud continuity. However I'd love to see an in-depth mtDNA survey of NW Africa (and other transition regions like Arabia also) focused precisely in L subclades - because for sure that there is even more information hiding there.

      I'll keep an open mind in this (as in most other things) but I'm not betting that we will ever find the lost lineage(s) of Jebel Irhoud. In any case, whatever legacy it left, it should be very residual nowadays.

      "L0 is more common in East and South Africa, but that might only be because L0 is attached to particular hunting strategies that migrated to East and Southern Africa from a West African origin".

      I do not think so. Applying geographical inference to African mtDNA scatter, the origin of humankind should be in the Upper Nile or Central-East Africa, in an arch between the CAR and Malawi, most likely towards Uganda, South Sudan, Ethiopia...

      L1b in my understanding shows signals of slowdown before expansion and boomed only late surely, up to the point that I'm not showing it in the maps of the link, while its "sister" L1c is instead.

      It's true that recent Y-DNA refinement appears to suggest a Western African oldest Y-DNA and that autosomal studies suggest a Southern African origin instead. So caution is required but I still feel that mtDNA is the most reliable marker with Y-DNA doing weird things because men move forth and back and kill or displace each other while women tend to stay and survive much better in the long run.

      "Well, the matter is under study by the Max Planck Institute"...

      I'm glad to hear that. It'd be interesting to know.

      "Regarding the connection and timing of U5 and U6, I'd like to read more about this".

      The connection is obvious, you just have to browse PhyloTree or Wikipedia: U had four "daughters": U1 (rare: centered in West Asia), U5 (West Asia, Europe), U6 (North Africa, Iberia) and U2'3'4'7'8'9 (with many branches in South and West Eurasia).

      The timing is debatable but it's been considered an old lineage for long. Pre-LGM for sure and IMO c. 55 Ka.

      Delete
  7. The other possibility is that the mtDNA L3->N transition occurs in Northwest Africa and then is carried across the Sahara eastward.

    Some N->R->U remains in place in Northwest Africa. Eventually, U expands in place, and also northward into Europe.

    There is, at low frequency, mtDNA N in Northwest and West Africa.

    The only impediment to this line of reasoning is that there is no mtDNA R in West or Northwest Africa.

    ReplyDelete
    Replies
    1. Even if the transition happened as you suggest, the carrier of N still had to travel all the way to Bengal (more or less) before it began having daughters who left a matrilineal legacy. But that's not likely because between L3 and N there are actually five transitions, each of which implies IMO at least one millennium, probably several.

      In any case just compare a clearly African node like L3, from which 5/7 basal subclades are almost exclusively African, with a clearly Asian node like N, from which 16/16 subclades are clearly non-African or mostly so. R is very similar, M is even more dramatic because it has more than 40 basal subclades.

      That's how things are: most N sublineages are in East Asia and Australia. In fact I usually consider N to have coalesced in SE Asia (or maybe Bengal), not West Asia as many Eurocentrists do. With a centroid estimated near Bangkok and not a single African-specific sublineage, an African origin for N is simply absurd.

      "... there is no mtDNA R in West or Northwest Africa".

      Actually most NW African mtDNA is R (U6, H, V, K, etc.) but all are clearly derived from West Asia (U, of which K is a highly derived subclade) or Europe (H and V).

      The West Eurasian haplogroups (often also found in North and even East Africa) are essentially:

      From N(xR):
      ·N1'5 > N1 (> I)
      ·N2 > W
      ·X > X1 and X2 (X1 is surely an Egypt-coalesced lineage but X2 is found in West and Central Asia and even as far as Native North America)
      ·(Other 13 haplogroups are almost only found from India to the East)

      From R:
      ·R0 > HV > H and V (both found in North Africa but clearly as derivatives from SW Europe)
      ·U > several subclades among which only U6 is African-specific (with generous overflow to Iberia)
      ·JT > J and T found in Africa but as derivatives from West Asia AFAIK
      ·(Other 13 haplogroups are almost only found from India to the East)

      From M:
      ·M1'20'51 > M1 found specially in West Asia and East Africa
      ·(Other 40-50 haplogroups are almost only found from India to the East, excepted "recent" Austronesian arrivals to Madagascar)

      So the pattern is demolishing and I would like you to have it very clear: M, N and R are Asian lineages by origin without any doubt. Their spillovers into Africa all belong to West Eurasian branches, all went through West Asia and in some cases also Europe.

      Delete
    2. As I stated above, I believe that the L3->M transistion occurred somewhere in Asia, so no argument there.

      "That's how things are: most N sublineages are in East Asia and Australia." So, if mtDNA N could go all the way to Australia, by say, 50,000 years ago, as it is argued, why on Earth couldn't N->R->U make across the measly Strait of Gibraltar?

      Regarding the separation of U from other African lineages, it could have something to do with the Atlas Mountains, hunting preferences and adaptation to Saharan pluvial/dry phases.

      Delete
    3. U does not separate from "other African lineages" but from other Asian (or Eurasian) ones in fact. The link of U with "other African lineages" (i.e. the L3-derived "cousins thrice removed") is long and convoluted.

      "So, if mtDNA N could go all the way to Australia, by say, 50,000 years ago, as it is argued, why on Earth couldn't N->R->U make across the measly Strait of Gibraltar?"

      Geographical reconstruction does not suggest anything related to the strait of Gibraltar nor anywhere in the vicinity. The basal centroid of N is in SE Asia, the basal centroid of R is around Bengal and the basal centroid of U is in West Asia. So:

      N > R > U

      reads in terms of geography:

      SE Asia > South Asia > West Asia

      (more or less)

      We don't see anything near Gibraltar until the U6 node itself. What means that we must add a geographic reading:

      West Asia > NW Africa

      And that's what needs explanation and has been tentatively linked to the Dabban industries... but really in wait of more archaeological data which could explain it well.

      Delete
  8. Just as a point of reference, the mtDNA L1 split from L is dated at about 155 kya according to the paper "A "Copernican" Reassessment of the Human Mitochondrial DNA Tree from its Root", Behar et al, Figure 2.

    There's plenty of L1 mtDNA in West Africa.

    ReplyDelete
    Replies
    1. I'd agree that it's as old or maybe even older but West Africa is not NW Africa.

      IMO L1 coalesced near Lake Chad (Megachad?), representing one of the first three branchings of early Humankind (H. sapiens), the others being L0 (East Africa initially) and L2"6 in the Upper Nile (~Sudan-Ethiopia). L1c seems to represent expansion to Camerun, Gabon, East Nigeria, Congo (and is indeed related to West Pygmies - but not only), L1b may be linked to a later expansion into West Africa, which, with my personal way of looking at the molecular "tick-tock", happened only at a later phase, even after the OoA. While L1 is a most old lineage as such, L1b does not seem that old (because it has an extremely long basal stem, indicating an extremely long period of existing as a mere small 'private' lineage - or otherwise radically "bottlenecked").

      The age of L1c itself is only comparable to that of L3, L2 and the OoA. Most of the mtDNA-detectable expansion in Africa itself happened soon before the OoA most likely. Not just I estimated that myself (repeated ref. link) but recent studies seem to confirm it for at least for L3 (Soares 2011).

      So we should accept that, for practical purposes, L1-derived lineages are like anything else, not for beginning with that letter-number combo a lineage is automatically older in this or that part of the world. Only that it belongs to a branch that began separating very early on in the ultimate all-human phylogenetic tree, which surely coalesced in a relatively small area of Africa around the Upper Nile.

      Linking that with Jebel Irhoud seems difficult. However the new Y-DNA A "elder" branches might be related.

      Delete
    2. Here's what I'm getting at . . .

      Jebel Irhoud is a *Mousterian* industry that takes advantage of the lower sea level at Gibraltar 150kya.

      Homo Sapiens, carrying mtDNA L or L1, comes into contact with Jebel Irhoud somewhere in northwest Africa.

      The timeline post dates Jebel Irhoud but predates the L3 split.

      Other details unknown at this time.

      Delete

Comments have temporarily been turned off. Because I currently have a heavy workload, I do not feel that I can do an acceptable job as moderator. Thanks for your understanding.

Note: Only a member of this blog may post a comment.