September 24, 2011

Mysterious 'East Asian' mtDNA in early European farmers

Critical update (Jul 17 2014): all the N9a found in ancient remains from Hungary is actually from the Sarmatian or Medieval period (interesting but not such a mystery anymore). 

This was reported by Eszter Bánffy et al. in 2012 in the Journal of Human Genetics (letter available at but I was not aware until today. There was some most unfortunate lack of communication between archaeologists and geneticists here.

Once the chronology has been revised the truly Early Neolithic samples of the Gubba 2011 study are reduced to four: one C5 (old nomenclature) and three likely H. Intriguingly the C5 sample is from Starcevo, while the likely H ones are from the Älfold Linear Pottery group. 

In the Late Neolithic (Tisza culture), two samples are H(?) and another one is D1 or G1a1. This indicates some sort of East Asian link so early in time but the lineage (particularly if confirmed as D1) is more common and is known to have existed in West Siberia since at least Chalcolithic times (although it became much more common later in the Bronze Age).

While N9a may still be somewhat anomalous even in the Sarmatian period (raising some questions about the origins of Sarmatians or some of their subpopulations) it is much more logical to find such Oriental elements already in the Iron Age.

(Thanks to Eva. Follows original unedited entry:)

This was going to go into a general compilation post but it has already stirred some debate in an unrelated entry, so I thought I'd open a proper space for debate here.

The issue is that a new research paper (Z. Gubba et al., Science 2011, pay per view) has found that many of the mtDNA HVS-I sequences at Hungarian Neolithic sites appear to be what we would usually call East Asian haplogroups, namely N9a (3/11, each from a different site), C5 (1/11), D1/G1a1 (1/11). In addition there are two mysteries: M*/R24 (1/11), maybe related to South Asia, and a novel R* lineage (1/11). Finally there are some more familiar cases N1a (1/1) and R-CRS, interpreted as H (3/11). See this chart provided by Waggg for the details.

The method of testing only HVS-I  has many limitations and one would have hoped that it would not be used anymore. If you are going to damage a valuable archaeological bone, you'd better produce valuable, solid, results and HVS-I does not serve that purpose. But the ego of researchers is infinite, or at lest bigger than my patience. 

Still the N9a lineages seem quite solid. While the 16261 site seems hypervariable (and hence suspicious) 16257 only shows mutation at the root of N9a in all the mtDNA phylogenetic tree as we know it. Its presence along other "Asian" lineages seems to reinforce the consistency of this finding.

However it is still most surprising. Even if C and D do exist in modern Europe (at low levels and mostly towards the NE, ref. - h/t to Waggg again), they have always been thought as rather late arrivals related to Uralic expansion probably. As far as I know N9a has never been reported (certainly not at any meaningful levels) in Europe before now. Not in Neolithic peoples either. 

So it's the kind of thing that makes you raise both eyebrows and drowns you in doubts. Feel free to debate in the comments section.


  1. I agree that extraordinary finds require extraordinary proof.

    However, perhaps with the exception of N9a (which could possibly be misidentified), I find the presence of eastern haplogroups not that surprising. If such movements are documented to have occurred in the neolithic and during historic times, why not between the LGM and the onset of neolithic?

    Between LGM and the Younger Dryas, there was a spillway in the planes between the northern Caspian Sea and the Black Sea - an ideal route for north-eastern people to follow. A short time later, you all-of-a-sudden find the Ahrensburg culture with novel bow-and-arrow hunting along the northern ice sheet from Germany through Poland.

    The water receded again during the Younger Dryas, just to rise by about 100m in ~1,000 years ~10,000 ya when climate got warmer and wetter, again. This corresponds to a loss of land of about 200km inland in the flatter shelf regions. So, before the beginning of the neolithic in the region, you would have had these people displaced from rich low-water fishing on the shelf, who then likely traveled the rivers (mainly Dniepr and Danube)upstream to seek a life.

    I don't see Europe as rotating its haplogroups every few thousand years. In the epipaleolithic I visualize pockets of relatively different local haplogroup composition, much of which has survived today, but is much more homogenous, now.

  2. I'd like that instead of conjecturing over hypothetical routes (the steppe could have been used as corridor at any time: AFAIK Altai keeps in fact West Eurasian affinities through all the Paleolithic) we could come up with archaeological data that could support such conjectures.

    AFAIK that data does not exist. And there is no reason to think of East Asian influxes in the Neolithic of Central Europe of all places.

    On the other hand there is, following Bauchet 2007, a minority "Finnish" genetic affinity scatter through much of Europe that looks too common to be just of "Viking" or "Germanic" arrival (West Irish and Italians have it in apportions that do not make sense if recent, even if Indoeuropean-scattered it looks odd). Finns have some Siberian admixture (c. 10%), so I guess that it is consistent to ponder some very minor Siberian introgression into many Europeans (it'd explain the occasional partial epicanthic fold). This is apparent, albeit very thinly, in fig. 1 of said paper, where the "Altai" orange component shows as very minor precisely in the same populations that display clear "Finnish" component admixture in the European-only graph (fig. 4)

    So genetically makes some sense. But archaeologically I'm amiss and I would not have expected it in Neolithic peoples of all them (either Paleolithic or Chalcolithic Indoeuropeans would have been my preferences - but seems not).

  3. I understand your criticism - at times I tend to have rather elaborate hypotheses.

    On the flip side, I did mention several testable points that conform to current research (two of which you agree with).

    - Archeology: Ahrensburg culture does come out of nowhere with unprecedented technology for Europe: bow-and-arrow hunting

    - mtDNA: well, we have it right here, locally.

    - autosomal: as you mentioned. there seems to be a remnant. Many autosomal studies I have looked at have a NE Asian background of 1-2% at high K in much of Europe. Could be noise, but perhaps it is real.

    - linguistics: the Urheimat of the Uralic languages may very well lie west of the Urals.

  4. "Ahrensburg culture does come out of nowhere with unprecedented technology for Europe: bow-and-arrow hunting".

    I doubt it comes "out of nowhere". Most Paleolithic cultures "come out of nowhere" on first shallow look and then they happen to be product of local evolution with transitions documented in many sites.

    Also I doubt it is related to this East Asian stuff.

    If I recall correctly Ahrensburg or rather some precursor is one of the last pockets of true Aurignacian tradition in Europe and is probably related to the origins of Magdalenian, which after all is nothing but an advanced Aurignacian.

    One possibility could be Lepenski Vir, which I think it's craniometrically associated to East European Epipaleolithic. But what is very strange is that N9a (and other "Oriental" lineages) are so common in Hungary's Danubian and then non-existent elsewhere in the whole culture.

    "the Urheimat of the Uralic languages may very well lie west of the Urals".

    Unsure about the language (it could well be a distorted very early PIE) but the Y-DNA lineages (N) come from NE Asia without doubt, as do come some associated mtDNA lineages (CZ, D). The fact that Uralic is spoken quite far in NE Asia by peoples of East Asian complexion strongly suggests to me that the direction of this ethno-linguistic flow was from East to West, from Asia into Europe. Not just genetics is strongly consistent with that but there is some logic in Siberian peoples already adapted (culturally) to extreme cold colonizing the tundra lands freed by the end of the Ice Age in Europe (there was never a ice sheet in most of Asia: just permafrost) and admixing here with local women (and some men as well).

  5. - AFAIK, there is absolutely nothing that would relate Ahrensburg to Magdalanian, at this point

    - it seems to "come out of no-where" in the local epipaleolithic context

    - Re: Uralic, if you think about it just a few millennia earlier, it fits the East --> West, and then very late western expansion picture very well.

  6. I think you must read this paper on the origin of Badegoulian (proto-Magdalenian) in Central Europe.

    The issue is still under debate but it is very possible that Badegoulian coalesced in Germany first from late Aurignacian survivals and then spread to SW Europe, where it is probable more recent. It was in France indeed where it coalesced as full Magdalenian and from where it would re-expand to Iberia and Central Europe.

    Whatever the case, the NW European lowlands where Ahrensburgian coalesced were not in contact with any Eastern culture, because east of the Elbe there was nothing but a thick ice sheet. I have yet to find any source specific on the origins of the Ahrensburgian but it must be from some of the cultural complexes south of it: Aurignac-Badegoule-Magdalenian or Gravettian or Solutrean - because elsewhere around them there were just ice or sea.

    Other papers relevant to understand the Badegoulian and the non-Solutrean genesis of Magdalenian may be this one from Gamble and this one from Banks (needs membership).

  7. Btw, I do not think you are correct at all by claiming that arrow use began only then. I have always thought it was a Solutrean invention, yet I also thought that of the needle and, lo!, it was a Kostenki one after all (or maybe even older).

    Whatever the case arrowpoints similar to those of Ahrensburgian surely existed in at least Iberian Gravetto-Solutrean, where the North African trend of back-tipped winged points reaches Europe.

    BTW, here there is a PPV paper that proposes the "Benelux" as origin for Ahrnesburgian, which would be consistent with my proposal of R1b1a2a1a1a (M405/S21/U106) being from that origin.

  8. Thanks for the links - I will study them when I have time.

    Use of bow-and-arrow seems to be on-and-off for many cultures, but some took it to the limit, to succeed. Ahrensburg seems to be one of them; with many finely worked pieces and shafts with resource-saving, pre-designed break-points found.

    Here is a video about the original amateur Ahrensburg archeologist. Quite impressive perseverance, notwithstanding debatable political affiliation.


    1. Thanks a lot, Eva. So in the end all the N9a was "recent" Sarmatian misidentification? The title of the article says it all: "appeal for real interdisciplinarity" between archaeologists and geneticists. The D1/G5 from Tisza culture (early Chalcolithic) and the C5 from Starcevo (early Neolithic) seem within the expected parameters and suggest that the Carpathian basin was always somewhat exposed to Eastern European influences but not in a the non-consistent way that so much N9a so early suggested.

      I'll see to update the entry with the new info later in the day. Thanks again.


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).