May 9, 2013

Echoes from the past (May-9-2013)

I am getting updated with a rather long backlog, so I will speed things up placing here in nearly telegraphic style the informative snippets that require less work. This does not mean that they are less interesting, not at all, just that I have to adapt to that elusive quality of time...

Middle Paleolithic

Toba supervolcano only had short-term climate effectBBC.

Research on Lake Malawi's sediments shows that the climate-change effect of the catastrophic eruption was limited. Droughts previously believed to be from that period have been revised to be from at least 10,000 years before, corresponding to the end of the Abbassia Pluvial rather than to Toba super-eruption.

Upper Paleolithic

Altai rock art and early astronomy from 16,000 BPSiberian Times, Daily Mail.

Sunduki (Khakassia), here there are what are surely the oldest rock art of Northern Asia, representing people hunting or interacting among them, which are from just centuries ago, however other petroglyphs are apparently much older like this horse:

Prof. Vitaly Larichev (Institute of Archeology and Ethnography, Russian Academy of Sciences) has detected a whole astronomical structure implemented in the landscape.

He claims to have found 'numerous ancient solar and lunar observatories around Sunduki'.

'This square pattern of stones on the ground shows you the place', he told visiting author Kira Van Deusen. 'I knew there would be an orientation point, but we had to search through the grass for a long time to find it.

'Now look up to the top of that ridge. You see a place where there is a crack between the rocks? If you were here on the summer solstice, you would see the sun rise right there. Or you would if you were here 2,000 years so. Now the timing is slightly differen'.

High on one cliff wall is a rock engraving showing dragon heads in one direction, and snake heads in the other.

'If the sun were shining, we could tell the time,' he said. 'In the morning the shadow moves along the snake's body from his head to his tail, and in the afternoon it comes from the other direction along the dragon.

'From the same observation point you can determine true north and south by sighting along the mountains'.


Vietnam: early cemetery dug in Thahn HoaAustralian National University.

Some 140 human remains of all ages have been unearthed at the site of Con Co Ngua, estimated to be 6-4000 years old. Cemeteries of this size and age were previously unknown in the region. The site has also revealed a dearth of artifacts. 

The people were buried in fetal position with meat cuts of buffalo or deer.


India: 4000 y.o. stone tools unearthed in Bhopal (Madhya Pradesh, Narmada river)India Today.

  • Some of them are decorated with aquatic animals.
  • 150x200 m. mound in Birjakhedi
  • Terracotta game pieces
  • Pottery (incl. jars, pots, dishes)
  • Stone and ivory beads

Bell Beaker rich lady's burial unearthed in Berkshire (England)Wessex Archaeology.

The middle-aged woman wore a necklace of tubular golden beads, amber buttons on her clothes and a possible lignite bracelet. She was accompanied by a bell-shaped beaker of the "corded" type (oldest and roughest variant, of likely Central European inception).

The chemical signature of the gold beads is coherent with deposits from Southern Britain and SE Ireland. 

Giza pyramid construction's logistics revealed Live Science.

Caesar beat the Gauls.
Was there not even a cook in his army?

Bertolt Brecht (A Worker reads History)

Now we know that at the very least the famed early pharaohs Khafra, Khufu and Menkaure, who ordered the massive pyramids of Giza to be built as their tombs did have some cooks in charge of feeding the many workers who actually built them, stone by stone. 

These workers were housed in a village some 400 meters south of the Sphinx, known as Heit el-Ghurab. In this place archaeologists have found a cemetery, a corral with apparent slaughter areas and piles of animal bones. Based on these, researchers estimate that more than 2,000 kilograms of meat were eaten every day during the construction of Menkaure's pyramid, the last and smallest one of the three geometric mounds. 

The figures estimated for such a logistic operation border disbelief: 22,000 cows, 55,000 sheep and goats, 1200 km² of grazing land (roughly the size of Los Angeles or 5% of the Nile Delta), some 3500 herders (adding up to almost 20,000 people if we include their families). 

A curious detail is that most of the beef was destined to the building of the overseers, while the common workers were mostly fed sheep or goat instead. Another settlement to the East of apparently local farmers ate most of the pork. There were also temporary tent camps closer to the pyramids.

Iron Age

Late Indus Valley Civilization was overcome by violenceNational Geographic.

Harappa (CC by Shephali11011)
The Late Indus Valley Civilization (Cemetery H cultural layer, usually attributed to the Indoeuropean invasions) was, unlike in previous periods, quite violent, new evidence highlights. 

The evidence from the bones also highlights the arrival of many non-local men, who apparently married local women. But the most shocking element is the striking evidence of widespread violence:

The skull of a child between four and six years old was cracked and crushed by blows from a club-like weapon. An adult woman was beaten so badly—with extreme force, according to researchers—that her skull caved in. A middle-aged man had a broken nose as well as damage to his forehead inflicted by a sharp-edged, heavy implement. Of the 18 skulls examined from this time period, nearly half showed serious injuries from violence ...

Gaming pieces of Melton Mowbray (England)Science Daily.

Excavation of a hillfort at Burrough Hill revealed ancient gaming pieces, among other materials. 

Funerary chamber found near the original location of the Lady of Baza (Andalusia)Paleorama[es].

(CC by P.A. Salguero Quiles)
The tomb has an access gate and is estimated to be from the 5th or 4th centuries BCE (Iberian culture) and, unlike most burials of the time, the corpse was not incinerated. 

The finding highlights the need for further archaeological work in all the hill but the severe budgetary cuts threaten this development. 

Baza (Granada) hosts a dedicated archaeological museum inaugurated in 2011. 

Tocharian mummy buried with marijuana hoardPaleorama[es].

Some 800 grams of the psychedelic plant, including seeds, were found at the burial place of a Tocharian man, presumably a shaman, at Yanghai (Uyghuristan), belonging to the Gushi culture and dated to at least 2700 years ago. The plant belongs to a cultivated variety.

Some of the oldest cannabis evidence are also from that area (Pazyrk culture c. 2500 years ago) and also from Nepal (Mustang, similar dates). Later in Southern Central Asia it was used in combination with opium and ephedra, from where soon migrated to South Asia and many other parts of Eurasia.


New device radically reduces costs and time in DNA extractionScience Daily.

Researchers from the University of Washington and NanoFacture Inc. have developed a device, which looks like a kitchen appliance, able to extract DNA from tissues (like saliva or blood) in minutes at low cost and without using the toxic chemicals habitual in the field.

The prototype is designed for four samples but can be scaled for the lab standard of 96 samples at once.


  1. * The IVC finding is very notable. People have been asking for a more than a century how there could be a Harappan-Vedic transition without any evidence of violence. Now, we finally have the evidence right where it should be at the point of transition.

    * The Tocharian marijuana find is also notable in light of ongoing historical questions over what kind of drug was referred to in the Vedic scriptures presumably having antecedents in Proto-Indo-European culture. While this doesn't answer the question, it does firmly establish that marijuana use was present in a very basal branch of IE culture that is probably more conserved than other known IE cultures which would have had more significant substrate influences.

    * The Lake Malawi study shows that the Toba erruption impact wasn't all that great in East Africa, but that doesn't mean that Toba couldn't have had far more profound effects in SE Asia - possibly opening a door to an influx of modern humans by decimating whatever flora or fauna had provided a barrier pre-erruption for a long enough window of time to allow modern humans to fill the niche before the barrier was restored. East African impact from Toba matter much less when modern humans have already landed in South Asia by then as we know that they did, rather than having to be an Out of Africa driver (which it was too late to have been).

    1. Agreed the two first points. However on the third I must say that what the Lake Malawi research does is to reject or declare almost nimious the climate change effects of Toba, not the direct effects of the super-volcano, which may still have been important in SE Asia (direct effects of the eruption) or South Asia (thick ash layer). It's not so much discussing the effects of Toba in Africa as the effects of Toba in the global climate, which were, it seems now, quite limited in scope.

    2. Now that I think of it, Andrew, your comment about "what kind of drug was referred to in the Vedic scriptures presumably having antecedents in Proto-Indo-European culture"... may not make much sense. You're probably thinking of the Soma, whose nature has yet to be properly discerned (lotus anyone?). This, if anything would be an ancient Indo-Iranian tradition and not necessarily nor likely "Proto-Indo-European", unless you can muster similar evidence regarding other IE cultures. Between PIE and Indo-Iranian there are several millennia.

  2. We must not forget that according to Dienekes blog
    Indo-European Caucasoid Tocharians discovered in the Iron Age Pazyryk burials carried the mtdna haplogroups D (4), C (2), A (1), G2 (1), K (3), U5a (2), J (1), HV6 (1) and T1 (1). This means that their maternal composition was 50% East-Asian and 50% West-Eurasian. I am quite sure that they did not look like modern Europeans or Vedic Aryans. Probably, they were culturally and genetically a mixture between Eastern and Western elements.

    This Pazyryk burial seems to be from Iron Age, so I do not know if this 2500 years old cannabis discovery was found somewhere else. However, a bag containing cannabis seed was found in this Iron Age burial.

    1. Why does it matter how Ancient Aryans from India looked like? I would imagine that cultural affinities or differences would matter much more.

      However, now that I think on it, by this time Indo-Aryans and Iranians were already separate groups even in religious terms (Zoroastrianism vs Hinduism). Yet this may not have been enough barrier for cultural propagation, much less of "interesting" drugs like cannabis, so important even today in the Hindu spiritual tradition. Sect diversity and the tolerance and syncretism proper of ancient religions probably helped to this cultural transmission.

      "... a bag containing cannabis seed was found in this Iron Age burial".

      Surely the same guy.

  3. @Kristiina: "Indo-European Caucasoid Tocharians discovered in the Iron Age Pazyryk burials"

    The Pazyryk culture was not Tocharian, they were related to the Saka and thus probably Iranic-speaking (there are tracks or Iranic hydronymy in this general region region) BTW). South Siberia was first inhabited by the Afanasevo cukture then it became part of the Andronovo horizon around 1,700 BCE IIRC).

    The Tocharians were living in the Tarim Basin not in south Siberia, even if Chunxiang Li et al, 2010 (about some of the earliset Tarim mummies whose hgs shows they were admixed to some point and copming from south Siberia) hints that the Tocharian speakers of the Xinjiang might have come from south Siberia, as an offshoot of the Afanasevo culture which was apparently the result of a migration from the Pontic steppes roughly at the time of Yamnaya culture.

    As for the Haplogroups of that south Siberian region, the mtDNA was 90% west Eurasian (hgs such as H, U5, T1, T2, I4, etc...) during bronze age and about 67% during Iron age according to Kayzer et al, 2009 which is confirmed by the morphology of the skeletons.

    To be precise, concerning this marijuana thing it's actually an old news. I heard about it several years ago (unless it's a new case and this case and it's still not something we just learnt about).

    1. There's a more recent study:

      According to it Pazyrk burials from Western Mongolia were 50-50 of West/East affinity (mtDNA) in the Iron Age and 100% Oriental in the Bronze Age. The paper also discusses other sites of Central Asia, showing how, while there was a quite abrupt divide in the Bronze Age at the Alati-Mongolia border, this divide vanished in the Iron Age as everyone became wildly admixed. This happened centuries before the Turkic expansion in the Scythian and Tocharian proto-historical period.

  4. Maju, you are right, my physiognomic comment was a bit irrelevant, but I agree with you 100% that inventions of this kind can move very quickly from one culture to another. I just wanted to underline the cross-cultural character of these ancient societies - East meets West, North meets South. Anyway, I prefer to think that cultures belong to the people who live in the area and that there is a certain continuum between ancient times and modern times in terms of genetics and culture.

    Yes Wagg, what I said about Tocharian was not true and it was deliberate. Tocharian texts seem to date from the sixth to the eighth century AD and they were found in the Tarim Basin. Tarim mummies which date from 1800 BCE to 200 CE carried R1a, which means that one can argue that they spoke a pre-Tocharian language. The mtdna of these people seem to be predominantly hg C. Wagg, do you mean with “that south Siberian region” the Pre-Iron Age burial in Kazakhstan (3400-3000BP)? When you said that Pazyryk people might have spoken an Iranian language, it is true that T1 and HV6 seem to be frequent among the Iranians, but also in Eastern Europe.

    However, these Central Asian burials and archaeological finds are really amazing! I hate cold but it conserves well.

    1. "Wagg, do you mean with “that south Siberian region” the Pre-Iron Age burial in Kazakhstan (3400-3000BP)?"

      He probably means other Pazyrk burials West and North of Mongolia, in Altai, see:

      Not all Pazyrk was exactly the same thing, it seems.

  5. These maps, that you surely have seen many times, show to some extent the distribution between Eastern and Western haplogroups, although Pazyryk seems not to be there:

    1. If they are the same sites as in the other map (same dates, different paper), Pazyrk culture is there. Pazyrk does not refer only to an specific site but to a culture spanning through the region in the Metal Ages.

  6. This is a very good list, if you understand Russian!
    Древние мт-гаплогруппы на современной тер-ии восточной части ЕурАзии:

    1)Тарим, Xiaohe кладбище,бронзов. век или пред. эпохи -С4,М*,К,R*,H, 4000 BP.
    2)Тарим, Xinjing(Синзянь) Hami кладбище,бронзов. век или пред. эпохи-С,D,Z,G2a,F1b,M7e,W,U2e,U5,U4,4000 BP.
    3)Монголия,древний Xiongnu,железный век- A,B,C,D,G2a,F1b,M,U5,J1,2000 BP.
    4)юж.Cибирь,бронзов. век или пред. эпохи-Z,U2,U4,U5,T,H,K, 3800-2800 BP.
    5)Алтай,железный век-C,F1b,G2a,N9a,T,H,HV,I, 2800-1600 BP.

    The map is here toward the end of the page
    6)KZ,железный век- A,M4,G2a,M*,H,HV,T,U1,W,U*,2800-1500 BP.
    7)KZ,бронзов. век или пред. эпохи-U5,T,H,I,HV,3400-3000 BP.
    8)оз.Байкал,бронзов. век или пред. эпохи-A,C,D,G2a,U5a,4960-3590 BP.
    9)оз.Байкал,бронзов. век или пред. эпохи-A,C,D,G2a,F,U5a, 7250-6040 BP.

  7. At this same site
    there were also interesting Mtdna haplogroup networks of Uygurs and Kazakhas. While looking at these maps, I started thinking that what if A and and C (and Z) arose in Northern China - Mongolia and went South only later on. In this scenario they would be originally connected to ydna K and P in Eurasia.
    R1a may have brought U4, U5, U2, W and H+HV to Altai, or it happened, at least in part, already with Q or even K. I am wondering how old these Western Mtdna lines are in Central Asia... Are the oldest U2 and H1. How about T. X should be old as well, as it is found in the Americas. I would say that mtdna D and G came to the North with ydna C and D. B and F could be connected to what happened in China.

    1. "... if A and and C (and Z) arose in Northern China - Mongolia and went South only later on. In this scenario they would be originally connected to ydna K and P in Eurasia".

      Probably not, excepted certain subclades, namely NO (East Asian) and Q subclades related to the Siberian UP spread ("proto-Amerindians"). There's no evidence of East→West flow except surely in the far North, related to Y-DNA N1c before the Iron Age.

      "R1a may have brought U4, U5, U2, W and H+HV to Altai, or it happened, at least in part, already with Q".

      I can agree with that but do not forget J and other Western or South Asian (H) Y-DNA lineages also present in Central Asia at significant levels. Don't forget X2.

      "... or even K".

      Nah. K as such seems too old to me, rather belonging to the southern Asian (and Near Oceanian) flows after the OoA.

      Essentially: F (SA) → IJK (SA?) → K (SA) → MNOPS (SA/SEA) → P (SA) → Q (WA) & R (SA) → R1 (WA), where SA=South Asia, SEA=SE Asia and WA=West Asia (incl. Central Asia). Only since the Q & R episode we can talk of clear Westward flows, although the matter of the genesis of IJ remains a bit debatable.

      "I am wondering how old these Western Mtdna lines are in Central Asia..."

      I'd say that they date to the early UP, at least in most cases. My opinion in any case. H1 may have arrived from Europe in the Gravettian period for example while U2 is clearly documented at similar dates in Kostenki (Russia).

      "X should be old as well, as it is found in the Americas".

      Not necessarily because its survival in the "proto-Amerindian" branch may have been a mere fluke (drift in a very small population); it is clear in any case that they mixed massively with East Asian women once and again, until they look very similar to East Asians in their autosomal DNA pool in spite of the almost unquestionable Western origin of Y-DNA Q.

    2. Correction: ignore the last paragraph on X, because I misunderstood you, Kristiina. I thought you wrote "X should be older", when you clearly do not say that at all. My bad.

  8. Maju: "this divide vanished in the Iron Age as everyone became wildly admixed" / "This happened centuries before the Turkic expansion in the Scythian and Tocharian proto-historical period. "

    Definitely, the early Iron age (roughly at 700 BCE) is the moment of an east Asian flow in all this region. Lalueza-fox et al, 2006 and Kayzer et al, 2009 clearly showed that (respectively for Kazakhstan and south Siberia). And this is indeed visible in skeletons as well.

    Kristiina: "Tarim mummies which date from 1800 BCE to 200 CE carried R1a, which means that one can argue that they spoke a pre-Tocharian language. The mtdna of these people seem to be predominantly hg C"

    Yes, in the Xiaohe people (Chunxiang Li et al, 2006), they were mostly mtDNA C4 which points to south Siberia in itself. There were only 2 west Eurasian hgs in these samples (the mtDNA H has matches in Europe up to Iceland (via Great britain I assume since there are also 4 mofdern matches there).
    This study alleges (with good reasons) that these early Tarim settlers came from south Siberia and thus we can expect that the Haplogroups of the other early settlers, easter, reflects the set of haplogroups found in south Siberia Bronze and Iron age's aDNA (clearly the Afanasevo substrate) in Kayzer et al 2009 which were almost fully R1a1a and with many "western" female lineages hence the (ancient and modern) Tarim lineages found throughout the diverses studies (mtDNA H, I4, T, K, U5, etc...).

    "do you mean with “that south Siberian region” the Pre-Iron Age burial in Kazakhstan (3400-3000BP)?"

    I was speaking of that region in general: the Afanasevo culture (an offshoot from the Pontic region, more or less at the time of early Yamnaya (Afanasevo starts around 3500-3200 BCE IIRC) where Europoids settled roughly around the Altai-Sayan (even though there were some settlements in north-eastern Kazakhstan). This culture became part of the Andronovo horizon around 1700 BCE. The pazyryk culture (around 700 BCE IIRC) were, AFAiK, derived from the Scythian (used as a generic term here, that englobes all the people like the Sakas, Alans, Sarmatians, etc...) culture.

    "When you said that Pazyryk people might have spoken an Iranian language, it is true that T1 and HV6 seem to be frequent among the Iranians, but also in Eastern Europe. "

    I'm not saying they came from Iran, I said they might have spoken some Iranic language.
    In the Kurgan theory, Iranic language is not considered as appearing in Iran itself.
    Proto-Indo-Iranian , indo-aryan and Iranian language families are considered natives of central Asia (which would be supported by the fact that the Finno-ugric languages got loanwords from all these groups and stages of this language family (some Indo-aryan (or closely related) loanwords were even apparently identified in languages east of the Urals as well) - and BTW, even some old IE loanwords (not Indo-iranian) were also identified, which certainly gives some weight to the Kurgan theory, as the region south of the Urals is not exactly the center of the antique world - one wonders why there would be so much IE loanwords from different groups there).

  9. "in the Xiaohe people (Chunxiang Li et al, 2006)"

    Oops. Chunxiang Li et al, 2010... Sorry.

  10. I was in Delhi in February and I looked incredibly different there, in particular my albino-approaching friend was touched and photographed by local people (and we thought that we would be the photographers and not the objects). I do not believe that India gave rise to almost all the Eurasians. If IJK arose in India and started expanding from there, there would be much more mtdna M in Eurasia. I think that IJK arose somewhere in Iran and P might have arisen in Afghanistan or Pakistan.

    As for mtdna C1, I found a recent study on it ( - Chapter 2). It seems that there is this very old European branch of C1 found, for example, in Ancient Karelians, Icelanders and Canarians (!).

    Then I found another study
    which confirms the existence of M8, M8a, Z and a big variety of C among the Altaians and other South-Siberian populations. They say that the coalescence time of all haplotypes to the root of haplogroup C was estimated as 38400 B.P. Then I found out that there is yet another Northern sister branch of C and Z in Udegey, Ulchi and Tofalar.

    The story behind this haplogroup seems very exciting to me.

    1. Isn't that argument way too superficial? Pigmentation is the less important of differentiating physical traits, precisely because it has been subjected to very strong adaptive selection (vitamin D is generated by photosynthesis in the human skin and is crucial for early brain development). It's like considering that a red maple and a green one are more different than a green maple and a pine, just because both are green!

      If you look at all other physical traits the differences are quite minor, more so when compared with other more distant populations like East Asians, Papuans or Tropical Africans. Anyhow the divergence between West Eurasians and South Asians (barring possible Neolithic remixing) dates to c. 50-60 Ka ago, while the divergence with other "Eurasians" like Papuans or East Asians would date to just "a bit" earlier: 60-80 Ka ago.

      The why of relative lack of mtDNA M must be understood as a founder effect. Anyhow most Western lineages are subclades of R, which has its greatest diversity in South Asia. So I would imagine a group back-migrating from SE Asia or even just Bengal to the West (say AfPak), from where they began the conquest of "the Neanderlands".

      Regardless of where IJK arose, it's clear that K and MNOPS migrated Eastwards before they back-migrated Westwards. K has two sublineages: LT, centered in Pakistan, and MNOPS, centered in SE Asia; P has three known subclades: P* centered near Bengal (and possibly including more than just one P sublineage), Q centered around Iran/Central Asia and R centered around maybe Pakistan (assuming that R1 coalesced West of it, as R2 seems very much South Asian). So, ignoring the quite dubious IJK question, all the precursors of R1 (F, K, MNOPS, P and R) look South Asian or from farther East (MNOPS).

      Being "quasi-albino" is anyhow not the norm in Europe but an extreme.

      "... confirms the existence of M8, M8a, Z and a big variety of C among the Altaians and other South-Siberian populations".

      Sure. What's new in this? They are all East Asian lineages are not documented in "Soviet" Central Asia, including Altai, before the Iron Age. They could be further North in Siberia however and that was surely the case, because we do find some of them in NE European Epipaleolithic and later.

      It's just obvious: if the East Asian lineages were not present in Central Asia until the Iron Age but they were present much earlier in NE Europe (Finno-Ugric peoples), then they first migrated westward through the Asian Far North (reindeer hunter specialists, later recycled to reindeer husbandry in some but not all cases) and only later they penetrated Central Asia.

  11. You are right that my comment might have been a bit superficial. And it is true that I might feel more at home in the North East than you do. I also knew that you believe in out-of-India expansion. I am changing my mind all the time with respect to migration patterns.

    You are right that P is found in Eastern India. According to Wikipedia, it is found in India, Iran and South Caucasus. In many studies you can find the marking P-M45, e.g. in Uzbeks (from Wells et al. 2001), listed like this: 5.5 P-M45 (xQ1a1-M120, Q1a31-M3, R1-M173, R2-M124). Which haplogroup is it in your opinion? Is it a Q line?

    Yes, C has been detected around Lake Baikal 7250-6040 BP, but also in Bronze Age Kurgan burials in Ukraine!
    and C1 in Carelia at Yuzhnyy Oleni Ostrov 7700–7300 BP, as you know.
    I do not know if you checked the network representation of C1 (p. 151). According to it, 2 European clusters (one European cluster has even spread to India) and Lebanese cluster are not derived from the East-Asian one but are clustered around the root. C1 has not been found in Finno-Ugric people, but e.g. in Bashkirs.,d.bGE

    I am still wondering who brought this line to Europe and from where. This Northern reindeer hunter may not have achieved all this.

    My point is that M8 is quite rare in the south. According to this study, the age estimation of M8a in Han Chinese is less than 20,000 years and, as I said before, it is estimated as 38400 B.P in South Siberia. If it arose from M in the North, I am wondering what is the ydna. Ancient Kanto Jomon had some M8a, Hokkaido Jomon none and Okhots people had some C. Jomon were typically N9b, D4, G1 and M7 and Okhotsk people Y, G1 and N9b. Udegey who have also this rare CZ sister branch have over 30% of C and M8, Nivkh and Negidal none, and Udegey carry also 40% of ydna P (M45), R1a (M17) and K-M9. If M8CZ is not connected to ydna C and D, we have either P or NO left. N is not old enough and O is problematic, because M8aCZ is clearly a Northern cluster.

    1. Rather than "out-of-India" I'd talk of "Southern Asia" not defining any boundaries too strictly because we really do not know enough for that. However when you see the line K-MNOPS-P swing between SE Asia/Oceania and West/Central Asia, it seems obvious that the Indian subcontinent is at the center of that region and must have played a major role, if nothing else as "corridor".

      "Which haplogroup is it in your opinion? Is it a Q line?"

      I don't know. It can be upstream Q (actually detected in most of those populations in other studies, which AFAIK did not detect P* anymore) or it could also be other P minor lineages. I lean for upstream Q but always with caution.

      Q itself could well be from South Asia, if we are to follow Wikipedia's info (ref. link broken), which mentions Q(xQ1) only in South Asia and Afghanistan.

      The data we have is not 100% conclusive but it's quite clear that MNOPS coalesced in or near SE Asia (brutal diversity over there), so P must have back-migrated Westward at some point and the only available and credible corridor is South Asia, especially when we look at clearly related mtDNA R and the lack of AMH in Altai before the "Aurignacian" early UP, as well as of any sign of East→West migration via Altai before the Iron Age.

      "Yes, C has been detected around Lake Baikal 7250-6040 BP, but also in Bronze Age Kurgan burials in Ukraine!"

      I did not know that but I guess it's easy to attribute it to the expected interactions of earliest Indoeuropeans (and maybe also Dniepr-Don peoples) with early Finno-Ugric peoples north of them. The Indo-Uralic linguistic paradigm, even if as mere sprachbund, seems very consolidated, and the same that Uralic peoples became genetically Europeans by means of repeated interaction and admixture with other Eastern European peoples, these must also have acquired some of the Uralic genetic baggage themselves. We can picture a time when there was a, primarily female-mediated, inter-flow between patrilocal early Finno-Ugrians and Indoeuropeans at the Volga region, which also influenced their respective languages in both directions.

      "This Northern reindeer hunter may not have achieved all this".

      Seems pretty obvious to me that they did.

      ... "M8 is quite rare in the south"...

      Should be. It probably originated 'around Beijing' in the early times of the Eurasian expansion. Like other NE Asian lineages (mtDNA D, A, Y-DNA C3, N) their main open areas for expansion were towards the Far North, maybe taking advantage of the sharp climatic changes that the area must have experienced in the late Pleistocene and later.

      "the age estimation of M8a in Han Chinese is less than 20,000 years and, as I said before, it is estimated as 38400 B.P in South Siberia".

      M8a also? If so, and assuming everything is correct, then it suggests (diversity rather than just accepting the guesstimate in years as such, which I do not) that the area of origin of M8 overall is rather "North of Beijing" than "Near Beijing", as I said before.

      "If it arose from M in the North, I am wondering what is the [related] ydna"...

      C3 and N seem the most likely candidates. I suspect that C3 is older in NE Asia than N and was there when the "Q clan" arrived from Altai, mixing with them in a long coalescent process. My hunch is that N expanded mostly, especially in the Far North, after the LGM.

    2. There is hg Q in Norway, its share is even 4-5% in some studies. In FT Scandinavia project I could find some Q (Q, Q1, Q1a3) in Finland, Norway and Sweden. I am enclined to think that this 7,500 years old Yuzhnyy Oleni Ostrov C1 could be connected to ydna Q. This could resolve in part the identity of people along this very early Arctic corridor to the Scandinavian North.

      Age estimation 38400 B.P is of whole C in South Siberia not of M8. The Li study about Tarim Basin mummies says about C that "the ancestral haplotype was found mainly in South Siberian whose diversity of haplotypes C is very high. Therefore, the original source of haplogroup C was inferred to South Siberian. In this Kurgan Burial study of Ukraine, all three Bronze Age individuals carried C, and they say that the type was near the root.

      According to Zhong et al. study, the age of STR variation of C3 is c. 19.000 and divergence time c. 33.000, and "the Han Chinese display a high STR diversity, especially those in the eastern coastal region as well as other eastern populations (Korean, Japanese), whereas populations in the north and west show low diversities ... [above] support the subsequent northward migration of Hg C along the coastline of mainland East Asia", and as I already said mtdna C is not widespread on the East Coast and it is not found in ancient Hokkaido, Tohoku and Kanto Jomon burials.

      N is said to be only c. 20,000 old. NO is of course older, and it is another candidate. To me that seems to imply that NO was in South Siberia c. 40.000 ago.

      Yes, I know that it is still possible that mtdna C arose with ydna C on the East coast, but there are however indications that point to another direction and the existence of ydna Q in China, Korea, Japan and Tibet is also highly exciting.

    3. According to Wikipedia, Scandinavian Q is mostly Q1a2a1a2 (L804) and Q1a2b1 (L527), so it is not the Near-Eastern M378. Q1a2a1a2 seems to be quite close to M3.

  12. Then I found another interesting study ( on R1b and Q in Siberia. “There is hg R1b1b1-M73 in the whole series of Turkic-speaking populations, including Shors, Teleuts, Khakassians,Tuvinians, Altaians, as well as Kalmyks. Its coalescence age in South Siberia is estimated as about 18.2±10.5Ka. The ages of subclusters A and B are amounted to 4.4±1.5 and 5.6±4.0Ka. Analysis of published data demonstrates that these two subclusters are present simultaneously not only in Siberia but also in different ethnic populations of China and the Caucasus.

    Previous studies have shown that haplogroup R1a1-M17 is present in South Siberia from the Holocene period (11.3±3.2Ka), while the oldest age estimates dating back to Mesolithic times (approximately 18Ka) were obtained for this haplogroup in South India and South Pakistan. High estimation of the age has been received for hg R1b1b1-M73 (22.9±9.3Ka) in populations of the Caucasus, South Ural and East Europe. A much younger age was found for hg Q1a3*-M346 in South Siberia (4.0±1.25Ka). One should note that the age of Q1a3*-M346 calculated from haplotypes detected in populations of China and North Pakistan appears to be much higher—17.8±4.1 Ka according to Zhong et al.

    Haplogroup R1b1b1-M73 is characterized by the high intra-haplogroup variability and appears to be very ancient, being dated at about 40Ka. We should note, however, that the calculated age of R1b1b1-M73 is so high probably due to very high repeat variance at locus DYS390. Coalescence time estimate for this haplogroup in South Siberia corresponds to about 20Ka, although its two phylogenetic STR-clusters (A and B) appear to be young (o4Ka), suggesting that their entrance in South Siberia occurred in relatively recent historical times or that evolution of this Y chromosome lineages in South Siberia was associated with recent population bottlenecks.

    According to study by Zegura et al.,20 divergence time estimate for the Altaian and North Asian Q* haplotypes versus the Native American Q1a3a-M3 haplotypes was 17.2±4.6Ka. We have also calculated divergence time estimate between Q1a3*-M346 haplotypes found in South Siberians and Q1a3a haplotypes revealed in northern Native Americans. As a result, we found that, despite the low current diversity of haplogroup Q1a3*-M346 in South Siberia, divergence time between Q1a3*-M346 and Q1a3a-M3 is equal to 13.81±3.88Ka, thereby pointing to a relatively recent entry date to America. One should note that similar divergence time was estimated for the other Amerindian founder haplogroup C3b-P39— 13.9±3.2Ka.20.

    Four Koryak individuals (10.3%) from the same region belong to paragroup Q1a* (xQ1a1, Q1a2, Q1a3, Q1a4, Q1a5, Q1a6). They are all characterized by similar Y-STR profiles and probably belong to a certain haplogroup, yet, unidentified by SNP marker ... it is likely that the range of Q1a*-MEH2 may cover a distance of about 1000km between the coasts of East Siberian Sea and Sea of Okhotsk. Coalescence age of the Koryak/Yukaghir Q1a*-MEH2 haplotypes is about 3.5±1.5Ka, that is within the bounds of the Saqqaq culture dating.”

    I think now that not only R1a, but also R1b may have carried some Western haplogroups to Siberia c. 5 000 ka, even before R1a, but they are not old enough for mtdna CZ. It is very interesting that it seems that Q1a3*-M346 was in China before it went to South Siberia (if I understood it correctly) and to America, but there is also another Q line that is detected in this ancient Saqqaq individual. Under this Q1a there seems to be also Q-M120, found in China, Japan, Korea and Tibet, and I now wonder if this Q-M120 is connected to the first Q wave to North East Asia. It is a pity that there is no age estimate for Q-M120. Perhaps, this proto M8CZ was carried to the North with this Q1a. I have also another theory, but I do not go into it now.

    1. The only thing I can say is that age estimates from different studies are not directly comparable, because each author probably used a different method. We should compare other more objective parameters like haplotype diversity and such.

      "I think now that not only R1a, but also R1b may have carried some Western haplogroups to Siberia c. 5 000 ka"...

      It's plausible, maybe even earlier. Western Siberians (Khanty but with extension to Central Asia) seem to be the product of very old admixture (although slightly leaning to the East) in Autosomal DNA. So it's probable that before admixture happened in Eastern Europe there was a preliminary stage of complex E-W admixture in Western Siberian early Uralic peoples. The Komi and the Chuvash at least keep a minor presence of this Western Siberian "mixed" component (also Finns and Russians but very thinly so).

  13. This comment has been removed by the author.

  14. @Kristiina:
    "Previous studies have shown that haplogroup R1a1-M17 is present in South Siberia from the Holocene period (11.3±3.2Ka)"

    It's not likely at all - the mtDNA lineages of the south Siberian aDNA (Kayzer et al, 2009) go against that idea. If R1a1a didn't bring the Ponatic/Yamnaya culture and technology (and west Eurasian mtDNA) in the Altai-Sayan and have insted to be associated with the few pre-chalcolithic east Asian hunter-gatherers from the this region, what y-DNA lineage brought it there then? Who brought the north European autosomal component that we see in ancient and modern people of that region? Seriously, I don't think there is much doubt on all this.

    I think you simply read an age estimation for the apparition of the clade found in the Altai, but this doesn't mean in anyway it appeared there...

  15. So, you protest against the idea that "R1a1-M17 is present in South Siberia from the Holocene period"?

    If you want to see that only R1a1a brought the Ponatic/Yamnaya culture and technology to the Altai-Sayan, it is still possible. R1a is a very widely distributed group and it allows a big variety of locations and sub-cultures developing here and there and interacting with other groups.

    However, contacts between East and West have been going on for several millennia in Central Asia and Siberia, so it is not necessary to postulate that North European autosomal component was brought there specifically through Ponatic/Yamnaya culture.

    1. Just to be precise: the Holocene period is right now. Without a more defined date it means nothing. I was actually wondering where you got that date of c. 11 Ka BP, which is almost exactly the beginning of the Holocene. From Wikipedia:

      "The Holocene is a geological epoch which began at the end of the Pleistocene (around 12,000 to 11,500 C-14 years ago)".

    2. I just copied the text as it is. In the study I cite above, reference is made to Derenko, M., Malyarchuk, B., Denisova, G., Wozniak, M., Dambueva, I., Dorzhu, C. et al. Contrasting patterns of Y chromosome variation in South Siberian populations from Baikal and Altay-Sayan regions. Hum. Genet. 118, 591–604 (2006).

      In the abstract of this Derenko et al study they say that "the population of the Baikal region demonstrates the prevalence of Central/Eastern Asian lineages, whereas in the populations of Altai and Sayan regions the highest paternal contribution resulted from Eastern European descent is revealed. Yet, our data on Y-chromosome STRs variation demonstrate the clear differences between the South Siberian and Eastern European R1a1-lineages with the evolutionary ages compatible with divergence time between these two regional groups."

    3. Kristiina: "So, you protest against the idea that "R1a1-M17 is present in South Siberia from the Holocene period"?"

      Well, I'm merely following the apparent course of events from the data I'm aware of: Europoids of the same physical type than the mesolithic population in the general Pontic region suddenly appear there during the Chalcolithic, apparently roughly around 3300 BCE, with Pontic customs, technology and artefacts - in stark contrast with what was found prior.

      That being said, maybe things were more complicated than what they look like.
      It's still possible that some western lineages + R1a1a arrived before, even if spontaneously I would think it's more likely _most_ the west eurasian lineages and R1a1a of the south Siberian aDNA (and most of the north European autosomal "tracks") arrived in the Chalcolithic.

    4. I just read the study ”Human migrations in the southern region of the West Siberian Plain during
      the Bronze Age: Archaeological, palaeogenetic and anthropological data” (available at least through a link on Dienekes blog)

      In this study we move East of Urals in the steppe zone between the Ob and Irtysh rivers. It might have been inhabited by Khantys who originated from the south Ural steppe and moved into their current location about 500 AD. Some consider Khanty's ancestors to be the prehistoric metalworking Andronovo Culture.
      This study is interesting for many reasons and not only if you want to insist on the late arrival of R1a and R1b in Siberia.

      The oldest burials are from 4th millennium BC (Ust Tartas). They have 50% U2e, U4 and U5a and 50% of C, Z, A and D. People are anthropologically defined as Northern Eurasian Anthropological Formation. From the 3rd millennium BC this first culture is followed by the Odinovo culture with bronze casting and weapons of the Seima-Turbino type. Mtdna seems to be D, C, Z and U5a (from the most frequent to the less frequent). At the end of the 3rd and beginning of the 2nd millennium BC Odinovo is followed by Krotovo culture with more intense contacts with Central Asia, but anthropologically and genetically people are like the earlier inhabitants.

      At the beginning of the 2nd millennium BC, the Andronovo migration starts and Krotovo populations are partially displaced to the north and partly assimilated. ”The changes observed in males from the Late Krotovo group could be attributed to the influence of Andronovo representatives. Some Mongoloid features, which were not related to the previous autochthonous substrate, appeared in females from the Late Krotovo group. The Europoid component observed in these females does not fall within the generally accepted criteria for the Andronovo anthropological type.” The most important feature is the appeareance of hg T. The other remarkable feature is that most mtDNA samples belong to haplogroups, such as A and C and one G2a. Instead, the transition from the Bronze to the Early Iron Age marks the arrival of more Western Eurasian haplogroups, to such an extent that in Chicha Culture there is a dominance of Western Eurasian hgs and the near absence of East Eurasian hgs. New haplogroups are U1a, U3, U5b, K, H, J and W. However, later on, U1a and U3 nearly disappear, but H becomes one of the most frequent lineages.

    5. It is safe to say that the oldest culture is a complete mixture of Eastern and Western elements: probably ydna N and mtdna C, Z, A, D (note that this mixture is close to the native american mtdna composition). The first culture bears resemblance to northeast European Pit–Comb Ware culture, so the other ydna could be I1, although today it is rare in Siberia. In Finland we have this ydna mixture, N1c + I1. Andronovo culture would indeed mark the arrival of R1a. Mtdna T arrives in the area at this time and is still very frequent in Tatars, Russians, Mordvins and in particular in Udmurts, but it is noteworthy that with Andronovo culture the Eastern (female) component increases. IMO Chicha Culture could mark the arrival of R1b. Khanty carry 19% of R1b, 76% of N and 4% of R1a. In this light, they are not descendants of Andronovo culture but instead a mixture between the earlier culture and the early Iron Age culture carrying ydna R1b. In addition to the oldest lineages, modern Khanty have 14% of H, 12% of J, 7% of T, very little of N1a, M, F and V.

      And Maju, in Chicha remains we have H6a1!

      After all this, I must insist on Davidski’s earlier comment on his blog according to which “I think this ancient Amerindian-like admixture in Europe is something Central Siberian or Central Asian, and it's shared by Europeans, Amerindians and East Asians. It might well be the same thing as what the proto-Uralics were before they mixed with Europeans and East Asian EDAR carrying eastern Siberians.” However, I would add that early Indo-European men (if Andronovo men spoke an Indo-European language) seem to have liked women with East Asian markers. I now wonder what was the linguistic identity of these R1b men.

    6. "And Maju, in Chicha remains we have H6a1!"

      Interesting. Could you include a relevant link? I ask because searching for "Chicha culture" I just get info about Latin America (chicha is a kind of alcoholic beverage over there, while in European Spanish it means 'meat' or 'brawn' colloquially).

    7. In google, Chicha culture is found with the 3 key words "chicha culture siberia"

      we read :
      "Russian Federation: Chicha
      Investigations on one of the most important sites of the transitional period between Late Bronze Age and Early Iron Age in the forest-steppe region of Western Siberia.

      Chicha, Western Siberia: a fortified settlement of the Late Bronze Age / Early Iron Age in the forest-steppe zone

      This is an interdisciplinary research project. Geomagnetic measurement was used for the prospection of the site. Since 2000 a group of Russian geophysicists is working on the development and testing of instruments and techniques. Archaeozoological and palaeobotanical analyses as well as some radiocarbon datings have been carried out in the laboratories for natural sciences of the DAI. Computer equipment and GIS are in use for documentation and elaboration. The functional analysis of house structures is aided by detailed phosphate mapping. Other participating methods with special issues are archaeometric measurements and analysis of organic and anorganic materials, for example remains of paint or food on ceramics, and stable isotope analysis of bone collagen in order to shed light on the dietary habits of the Chicha community.

      The settlement was existing during the transitional period from the final Bronze Age to the Early Iron Age. At about that time the inhabitants of the steppes are told to have changed their way of life from formerly farmers to nomadic people with a great importance of horses under the influence of climatical change. The situation of Chicha in the contact zone from northern (Taiga) non-nomadic with southern (steppe) nomadic people just at the edge of the Bronze Age to the Iron Age makes it a site of great scientific value for the investigation of that process. Interdisciplinary research with modern methods will do a great contribution to the understanding of the cultural development in Siberia.

      History of Research
      he inner part of the settlement is known since the late Seventies. In 1979 V.I. Molodin excavated in this area a so-called "Poluzemljanka" - a house with a half-sunken floor dating to the Late Irmen´ Culture (end of 2nd millennium B.C.). Since that time Chicha has been one of the major archaeological sites for the transitional period from the LBA to the EIA in the forest-steppe of Western Siberia. "

    8. Thanks Palamede, very informative.

      "In google, Chicha culture is found with the 3 key words "chicha culture siberia"".

      Heh! How did I not think of that myself? :)

  16. It is a pity that we do not know the haplotype of Tarim mummies more specifically. Then we would be much wiser.

  17. Maju,

    I had included above the list where the first two are the following:
    1)Таrim, Xiaohe, Bronze Age С4,М*,К,R*,H, 4000 BP.
    2)Таrim, Xinji(a)ng, Hami Bronze Age С,D,Z,G2a,F1b,M7e,W,U2e,U5,U4,4000 BP.

    On Eurogenes Blog ( we have a list of ancient ydna from China, including
    Xiaohe, Xinjiang, 3500-4000 YBP, 11 R1a1a, 1 K*
    Tianshan Beilu, Hami, Xinjiang, 3300-4000 YBP, 5 N, 1 C

    Do you think that mtdna and ydna correspond respectively?

  18. Maciamo says on Eupedia page that "Gao et al. tested the mtDNA os remains from the Hami site dating from the Bronze Age and early Iron Age. They identified the Eastern lineages A, C, D, F, G, Z and M7, and the Western lineages W, U2e, U4, and U5a. The Western lineages are typical of European Russia, probably the Volga-Ural region. U2e, U4 and U5 were all found in Bronze-age Andronovo sites that had exclusively R1a1 male lineages. No surprise here."
    I now wonder if it is these Hami people who had 5 N and 1 C. It is interesting that in Volga Ural region we have today ydna R1a and N.

    Then Maciamo goes on saying that "The second study by Li et al. tested the much more recent site of Heigouliang (2000 years old). While the Eastern lineages were all identical (except for M7 missing), the West Eurasian haplogroups were all different (H, K, J, M5) and much more Middle Eastern in appearance. M5 is actually South Asian. This points at the arrival of a new group of people, perhaps from Iran via Afghanistan. This is in line with Hemphill & Mallory's morphological description of the skeletons from this period, which were described as Eastern Mediterranean, in sharp contrast to the Bronze Age Proto-European Steppe people."

    On Eurogenes blog page on Chinese ydna the list includes this:
    Heigouliang, Xinjiang, 2000 YBP, 6 Q1a*, 4 Q1b, 2 Q

    Q1a could be indeed Q-M346 that has been found in Pakistan,Saudi Arabia, the United Arab Emirates, India and Tibet

    According to Wikipedia, Silk Road originated during the 1st century BC, but had precursors in nephrite jade trade from the 2nd millennium BC. The area was surely highly attractive from the trading point of view.

    1. I'm sorry but I'm getting a bit lost, where is Hami. Also can you include links when you reference sources?

    2. It is all said in the comment below:
      Ancient DNA reveals the population origin of the Eastern Xinjiang SHIZHU GAO, HONGJIE LI, CHUNXIANG LI and HUI ZHOU
      College of Life Science, Jilin University, College of Pharmacia Science, Jilin University, Research Center for Chinese Frontier Archaeology, Jilin University
      It is probably not available in English
      The Hami site is located in the East of Tian-Moutain in Xinjiang.

    3. So this Hami site would be one of the first E-W admixed populations in Central Asia, assuming it's all the same in the Bronze and Early Iron, what may not be the case.

      It's interesting but does not seem to dispel the general notion of a rather sharp E/W genetic divide in the region before the Iron Age. An exception does not make the rule.

  19. In the meeting abstract including this Gao study "Ancient DNA reveals the population origin of the Eastern Xinjiang SHIZHU GAO2, HONGJIE LI1, CHUNXIANG LI1 and HUI ZHOU1", they say that The Hami site is located in the East of Tian-Moutain in Xinjiang, so the above "Таrim, Xinjing, Hami" is probably not written correctly.

  20. The study of R1a by academics is a total mess at the moment. These people aren't even aware of 10% of the SNPs that make up its structure. I wouldn't bother to quote any peer reviewed papers on R1a.

    The vast majority of Asian R1a is too young to be anything but intrusive from Europe during and after the Chalcolithic. The main Asian subclade, R-Z93, broke off from the main European subclade, R-Z283, no earlier than the late Neolithic.

    I have a summary of the latest R1a data from the 1000 Genomes Project below, and it paints a very convincing picture of an Iron Age invasion of South Asia by populations carrying a lot of European-derived R1a.

    1. "Asia" is a way too loose concept, David. Could we not be more strict and differentiate between South Asia (which is what you seem to be talking about), Central Asia and West Asia? In this other thread it was evidenced that the basal diversity of R1a is high in West Asia (all R1a*, most R1a1*) for example, what makes it a plausible origin for the lineage as a whole R1a.

      With the data I have now, I cannot exclude (but rather the opposite) that R1a expanded from West Asia, maybe in the Neolithic (or earlier or both).

    2. Yes, there are differences between West Asia and South and Central Asia in this context.

      West Asia has almost all of the ancestral and rare R1a mutations that Europe has. What this might mean is that R1a expanded into Europe from West Asia during the Neolithic.

      But at the moment the data doesn't preclude a European origin of R1a. Indeed, it might well be that West Asian R1a is so diverse and old because West Asia is close to Europe. In any case, South Asian R1a looks young, and unlikely to have arrived there during the Neolithic from West Asia. An entry from Central Asia during the metal ages makes more sense.

    3. But R1a* is only found in West Asia.

    4. I documented it in the thread mentioned above (update). You were present in the discussion leading to that update and yourself provided some of the relevant information.

      Following the discussion, I see that the main source was Underhill 2010.

    5. From that data it'd seem that the IE-specific lineage would be not R1a but R1a1a1b1 (S339/Z283), although there are many question marks between R1a1* (West Asia and Europe) and R1a1a1b1. This lineage might also be Neolithic instead of IE/Bronze in South Asia (we do not know for sure).

      It'd seem in any case that R1a expanded from West Asia into Europe in one or several bouts and that one specific downstream variant (R1a1a1b1 (S339/Z283)) became most important and spread also to South Asia (R1a1a1b2 (S202/Z93)). These could be the IE lineage (or Neolithic?)

    6. Well, there's definitely R1a* in Europe.

      Also, the split between R1a-Z283 and R1a-Z93 is far too low down the R1a haplotree to have happened during the Neolithic. The two lineages are very closely related, and most likely split during the Copper Age.

      At that time R1a-Z283 expanded across Europe and swamped most of the other subclades, like R1a-Z93 and R1a-L664, while some R1a-Z93 moved into Central Asia, and expanded into South Asia from the north (as the latest SNP data from South Asia shows).

      BTW, the Corded Ware R1a from Eulau looks like it's Z283-, Z93- and L664-, which is interesting as Z283 now dominated that area almost completely.

    7. "Well, there's definitely R1a* in Europe".

      Source and specific locations please.

      "... the split between R1a-Z283 and R1a-Z93 is far too low down the R1a haplotree to have happened during the Neolithic".

      That obviously depends on how much you trust this or that molecular-clock-o-logic estimate. I generally distrust them all, so for me this means nothing without further evidence. R1a is way too common in South Asia for me to easily accept that it is the product of a bunch of nomadic warrior invaders, especially as it's not restricted to the upper castes at all. Therefore I find the Neolithic time frame a more reasonable window to have caused such a dramatic impact.

      In Europe it may be a bit different because most of the impact of R1a belongs to areas that had low densities in the Neolithic/Chalcolithic period. A more reasonable comparison with South Asia could be Turkey (Hittites), Kurdistan or Iran (or even the Balcans), where R1a is important but in minority.

      "BTW, the Corded Ware R1a from Eulau looks like it's Z283-, Z93- and L664-, which is interesting as Z283 now dominated that area almost completely".


    8. So, it seems that R1a-Z283 and R1a-Z93 may have come to the West from the Caspian Sea region and for the most part replaced the earlier R1a types in Europe. That would explain IE-languages pretty nicely.

      I studied more in detail the different N haplogroups and now I think that the eastern western division of the earlier Barabba cultures could also reflect the difference between N1b and N1c. Khanty have 38% of N1b and 38% of N1c, western and eastern mtdna were also divided in 1:1 ratio. N1c1 has a high diversity in Eastern Europe and its evolutionary time estimate in Eastern Europe is 9 ky. ETE for N1b is 6.7 ky. But there is also a division between eastern and Siberian N1c1a (age in Eastern Europe 9 kya) and N1c1(?) (before N3a2, age in Siberia 5 ky), respectively. The latter spread from Buryatia and Tuva to the West and it is typical in Komi and Mari but not further West. Thus, the Ust Tartas people might have carried European N1c1 and eastern N1c. Then, N1b might have arrived from the East with Odinovo culture and Seima-Turbino weapons, as Odinovos have for the most part eastern mtdna.

      Shors seem to be a central group in the spread of R1a in Siberia (see haplotype analysis
      and their ydna frequencies may give an indication of the pre-Andronovo and post-Andronovo eastern influence in Barabba steppe. Here you have their frequencies: P (Q?) 2%, R1* 19.6%, R1a1 (59%), N* 14%, N3 2%, C 2%, F 2%. Without R, the distribution between Q and N is 2 and 16 respectively. If, indeed, Andronovo culture spread to Barabba steppe from the East that would explain why they carried for the most part eastern mtdna (of which again 50% of C) and the first true ”mongoloid” features were detected only then in the skeletons of Barabba steppe. I am not arguing for the eastern origin for R1, but I propose that Andronovo R1 arrived to Altai from the south through Kazakhstan steppe and started spreading in the Taiga forest zone from the east. It is obvious that East-European R1a1 spread from the West and also reached Altai and, in fact, R1b and this East-European R1a1 (with, for the most part, western mtdna, including K, H, U3, W) may have arrived during the late Bronze Age). It seems to me that R1a of Volga Ural people has dual origins, partly from the West and partly from Northern Altai.

      As Andronovo people in Barabba steppe carried also mtdna T, I checked where you can find the highest frequencies of T and it is in Iran (Gilaki, 16%, Mazandarian 14%), Azerbaijan (15%), Caucasus (13%) and Turkmenistan (Kurd, 19%). In Russia the share of T is around 10%.

      As for the age, we all know that there is a complete disagreement. According to this evolutionary time estimate, the age of R1a1 in Siberia is 11 ky and it means Palaeolithic time depth. Also the recent Zhong et al. paper concluded that the age of Y-STR Variation of only East Asian R1a1*–M17 is 15.37 ky. By comparison, with the evolutionary effective mutation rate the divergence time estimates for the whole Q is 17 ky, for C 14 ky (Altai + North Asia versus Native Americans) and for M3 only 15 ky (Trichotomy for North, Central, and South Americans). I think that it is problematic if we always pick just those age estimates that are obtained with a method that best suits our own beliefs.

      However, IMO it is possible that haplogroup M3 arrived very late to the Americas and replaced the previous ydna which could have been Q1aMEH2 or even NO. That would explain why there is such an enormous morphological diversity in Native American languages.

      On the basis of this Barabba Steppe paper, it seems to me that N spread to the West and R1 to the East quite early.

    9. T at the small apportions found in Baraba could well have arrived from Europe, where it is found quite commonly since at least Neolithic times, or Central Asia as well.

      I hope to be able to discuss that paper today. Work never seems to stop piling up - right now I'm trying to build a drift+mutation simulator in order to discuss the issue of mtDNA "molecular clock" problems with more specific data (i.e. "I did 100 runs and got these chances"...) As I'm not any good at programming, I'm working with FreeBasic, which is almost the same as the Basic I was relatively good at in my youth (but need to recall a lot of rusty stuff). In addition to that more and more interesting "paleonews" show up almost every day and I still have some other papers to comment in the "to do" folder. But the Baraba one is a priority because it seems pretty straightforward in the data and conclusions and also deals with a region not studied for aDNA before AFAIK.

      "On the basis of this Barabba Steppe paper, it seems to me that N spread to the West and R1 to the East quite early".

      Probably correct, even if we are making Y-DNA inferences from mtDNA data. However it would also seem to me that south of Siberia proper, there was probably not significant East Asian genetics before the Iron Age, remaining these westerly offshoots restricted for long to the Subartic latitudes.

    10. Using just "common sense" I would argue that after the end of Ice Age R1 started spreading on the steppe aare which was getting more hospitable and similarly N was spreading West in the more and more hospitable Taiga zone. Haplogroup I probably did the same in Europe.
      I am waiting with excitment your runs! I also agree on T. That would strenghten the eastern connections of Andronovo culture in Baraba Steppe.

    11. I quote Wikipedia: Seima-Turbino Phenomenon
      "The Altai Mountains in what is now southern Russia and central Mongolia have been identified as the point of origin of a cultural enigma termed the Seima-Turbino Phenomenon. It is conjectured that changes in climate in this region around 2000 BC and the ensuing ecological, economic and political changes triggered a rapid and massive migration westward into northeast Europe, eastward into China and southward into Vietnam and Thailand across a frontier of some 4,000 miles. This migration took place in just five to six generations and led to peoples from Finland in the west to Thailand in the east employing the same metal working technology and, in some areas, horse breeding and riding. It is further conjectured that the same migrations spread the Uralic group of languages across Europe and Asia. However, recent genetic testings of sites in south Siberia and Kazakhstan (Andronovo horizon) would rather support a spreading of the bronze technology via Indo-European migrations eastwards, as this technology was well known for quite a while in western regions."
      I somewhat disagree on these migrations as the inventions can pass from one culture to another without any replacement of people.
      Anyway, the first Chinese Bronze Age is Majiayao c. 3000 and the first Bronze Age culture in India is early Indus Valley civilization 3300 BC and in Ancient Egypt, the Bronze Age begins in the Protodynastic period, c. 3150 BC. From this perspective, Indus Valley culture seems to be the oldest! Thus, it would be logical that Bronze metallurgy spread to Altai from India through the steppe and not from Russia.

    12. I also do not think that the Seima-Turbino phenomenon necessarily implied migrations, much less all from the same origin. For example in the Baraba area, we see that there is a change of haplogroup composition in that period, with clear increase of Oriental lineages (notably mtDNA D) and dramatic reduction of Western ones, however this is hardly compatible with an Altaian origin of the migration, so the connection must be oblique.

    13. So, Xinjiang, Hami Bronze Age С,D,Z,G2a,F1b,M7e,W,U2e,U5,U4,4000 BP.
      Tianshan Beilu, Hami, Xinjiang, 3300-4000 YBP, 5 N, 1 C

      In Hami Xinjiang we have 5 N and 1 C. Mtdna C,D,Z,U2e,U5 and U4 are just the same haplogroups that we have in Ust Tartas! Mtdna G2a (frequent in Buryatia and in Ryukuans) was found in Late Krotovo culture. Mtdna F1b is most frequent in Tuva. M7e may have come in Hami with ydna C (or there is no correlation at all), as the distribution of M7e is very southern (
      So, it seems that the tiny N1c clade in China may have spread with Seima-Tubino phenomenon, although I do not know what these Hami N are.

    14. In this Taiga zone, Eastern haplogroups are found in the West: at Bolshoy, 3500 yBP, we have 2xU4a1, 6xU5a, 1xT*, 6xC*, 2xC5, 3xD*, 3xZ1a. However, it is true that at Popovo site, 7000 yBP, we have only 2xU4 and at Yuzhnyy Oleni Ostrov site, 7500 yBP, we have western haplos: 2xU2e, 2xU4, 1xU5a, 1xH and surprisingly 3xC1. I would say that the European N1c1 is connected to U2e, U4 and U5a.

  21. Hey Maju, what do you think about my comments in the update at the bottom of this post?

    1. Short of time but will write on the Minoan DNA and other backlogged stuff soon. I don't think there's any reasonable connection with R1b, which could not really expand (no cultural vehicle) so late in time.

      There is clear continuity with modern Lashiti population, what is a surprise, because the population was supposed to have been deported and replaced by mainland Greeks (yet no mainland Greek population as so much R1b, as do not any other Cretans either.

      "... the plateau has been continuously inhabited since then, except a period that started in 1293 and lasted for over two centuries during the Venetian occupation of Crete. During that time and due to frequent rebellions and strong resistance, villages were demolished, cultivation prohibited, and natives were forced to leave and forbidden to return under a penalty of death. (...) Venetian rulers allowed refugees from the Greek mainland (eastern Peloponnese) to settle in the plain and cultivate the land again".


      Lashiti Plateau (just a minor inner district of Lashiti Prefecture/Region, do not confuse them both) was not any major Minoan center anyhow, as these were at the coasts. The most outstanding Cretan Y-DNA lineage is J2 (Mainland Greeks are instead dominated by E1b).

      So there's some mystery (fluke coincidence?) about the apparent mtDNA continuity between ancient and modern Lashiti Plateau peoples.

      Said that, it's true that in the Late Chalcolithic or so (exact time???) we see a growth in the representations of bulls in Iberian rock art, sometimes replacing the older representations of buck deers. This can reasonably be attributed, I guess to Eastern Mediterranean influences or maybe also to an increase of agriculturalism vs residual foraging. Rituals involving bulls or cattle ere known in many other peoples (for example the Dinka) and could be just attributed to cattle-herding culture. However the particular non-bloody bullfighting style of "recorte" shows some similitudes to ancient Cretan traditions (if we are to believe the frescoes and modern versions of the "recorte" style, which sometimes include jumping above the bull). Yet other traditions like the mainstream bloody bullfight don't seem to relate to Crete directly and instead may make better sense with Plato's mention of bull sacrifice ("hunt") rituals of Atlantis.

      In any case it does not look like the modern R1b from Lashiti Plateau can be easily related to ancient Minoan Crete (much less without knowing about downstream SNPs) and also the dominant Y-DNA lineage in most of Crete is J2, with only lesser presence of R1b, which is surely not of the Western varieties.

    2. Also, I do not see any realistic affinity of ancient Minoans with Western Europe when looking at specific haplotypes and the overall frequency of them. They look instead akin to Central Europe's Neolithic (LBK) populations.

    3. The Minoan mtDNA doesn't resemble mtDNA from LBK or even Cardium Pottery. It actually looks very Bronze Age southern and western European.

      Also, the central Mediterranean is one of the main areas of Europe where western and eastern R1b clades overlap. So it's not implausible that modern western European R1b is a subset of early Bronze Age central Mediterranean R1b.

    4. I'm going to discuss it today but my impression is that it looks a lot like LBK (with some differences of course).

      ... "the central Mediterranean is one of the main areas of Europe where western and eastern R1b clades overlap".

      The origin of 'Western' R1b-M269 looks Balcanic. 'Overlap' zones are around that Balcanic core. Also Cretan R1b is almost 50% of the "Celtic" subclade (R1b-U152) while some 40% is of the "ancestral" clades (R1b(xL11).

      The most important Cretan R1b sublineage is R1b-U152, which must have arrived from Western Europe, probably via Italy.

      In other words: the R1b of Crete does not look ancestral to Western R1b sublineages but at least half of it is derived from it, while the rest appears Balcanic.


      This opens the possibility that the perceived Cretan affinity with the West is caused by migration from the West in fact (mixed with local Balcanic/Anatolian elements).

  22. This comment has been removed by the author.

  23. Please check these theses now available online:

    Kurgan people of North-Pontic steppe in Ucraine carried in particular mtDNA haplogroup C4a and it seems to have developed in situ!

    Earlier Neolithic Dnieper-Donetz culture appears to be genetically more mixed with MtDNA from Europe, Siberia and Middle East (Caucasus, Iran?). Dnieper-Donetz people carried ancestral C, and their H haplotypes are found today around North Sea, in Bosnia, Armenia/Greece, Western Europe and in Druze. One H haplotype may be present in a Neolithic cemetary in Lake Baikal. Dnieper-Donetz T haplotypes seem to be at the root of the haplogroup. Other haplotypes found are U3 and U5a1a.

    1. Previous studies like the one by Desarkissian also found some presence of of C, D, A and F1b lineages among Scythians some 3-4 centuries before the conventional era. But much higher frequencies of C, D and Z1a among peoples further North which are probably at the origin of modern Finnics. But the majority of matrilineages (2/3) were Western.

      Again in this case they find a majority of Western lineages and some Oriental ones, curiously all belonging to C, which could well be the vanguard of Oriental lineages in Europe. The fact that some of them are as old as Neolithic and also C1 was found at important frequencies in Epipaleolithic Karelia (1/3), would suggest that the Uralic advance into NE Europe was already going on in the Epipaleolithic, what does not surprise me the least. After all those were peoples highly specialized for the coldest areas of North Eurasia, as the distribution of Y-DNA N1 evidences, and the Ice Sheet never covered the sub-Uralic area of Europe (as it did not with most of Siberia, too dry to be snowed), so they may have been around there even from late Upper Paleolithic times, although I do not know this extreme for sure.

      "Dnieper-Donetz people carried ancestral C, and their H haplotypes are found today around North Sea, in Bosnia, Armenia/Greece, Western Europe and in Druze".

      The H data makes sense to me but for the C alleged local ancestry I can't be certain in any way (it would only affect one of the later lineages anyhow). We have to be aware that there is no need for mtDNA to accumulate mutations at any regular rate: the apparent rate is way too slow for that, with the evolutionary steps showing rates of some 40 generations or more, which can perfectly be irregular. At most I'd say that they do seem related and allows for such speculation.

      "Dnieper-Donetz T haplotypes seem to be at the root of the haplogroup".

      They are described in italics, what means that the HVS-I sequence is incomplete. It's extremely unlikely that DD would be ancestral to all mtDNA T IMO.

      In any case it's an interesting piece, thanks a lot.


Please, be reasonably respectful when making comments. I do not tolerate in particular sexism, racism nor homophobia. Personal attacks, manipulation and trolling are also very much unwelcome here.The author reserves the right to delete any abusive comment.

Preliminary comment moderation is... ON (your comment may take some time, maybe days or weeks to appear).