Wednesday, September 16, 2020

Domestic horses were introduced into Anatolia and Transcaucasia during the Bronze Age (Guimaraes et al. 2020)

Over at Science Advances at this LINK. This is a very important paper because it basically eliminates West Asia as the source of the modern domestic horse lineage, which leaves the Pontic-Caspian steppe in Eastern Europe as the only viable option.

It also corroborates the linguistic theory that the Proto-Indo-European homeland was located on the Pontic-Caspian steppe. That's because the horse is a key animal in the Proto-Indo-European pantheon, and it appears in Indo-European mythology in intricate roles. This suggests that the speakers of Proto-Indo-European weren't just familiar with the horse but also managed to domesticate it. From the paper:

Abstract: Despite the important roles that horses have played in human history, particularly in the spread of languages and cultures, and correspondingly intensive research on this topic, the origin of domestic horses remains elusive. Several domestication centers have been hypothesized, but most of these have been invalidated through recent paleogenetic studies. Anatolia is a region with an extended history of horse exploitation that has been considered a candidate for the origins of domestic horses but has never been subject to detailed investigation. Our paleogenetic study of pre- and protohistoric horses in Anatolia and the Caucasus, based on a diachronic sample from the early Neolithic to the Iron Age (~8000 to ~1000 BCE) that encompasses the presumed transition from wild to domestic horses (4000 to 3000 BCE), shows the rapid and large-scale introduction of domestic horses at the end of the third millennium BCE. Thus, our results argue strongly against autochthonous independent domestication of horses in Anatolia.
Guimaraes et al., Ancient DNA shows domestic horses were introduced in the southern Caucasus and Anatolia during the Bronze Age, Science Advances 16 Sep 2020: Vol. 6, no. 38, eabb0030, DOI: 10.1126/sciadv.abb0030

See also...


  1. 1. All the same, there are no statements about where the domestication of the horse took place in Europe and when, it could have been in the steppe, or it could have been in the forest-steppe, or it could have been in Central Europe. They just ruled out everything south of Europe.

    2. They have a very controversial assumption that the horse enters Western Asia through the Caucasus, and not through the Balkans. A way through the Caucasus is possible, but still the earliest horses in Anatolia are Troy IV around 2000 BC and it is somehow completely unconvincing that they came from the Caucasus. They completely ignored Troy. In general, here they are not at all convincing, they are judged by indirect signs associated with the magnitude of the statistical error in small samples.

    1. There are elite horse burials in the South Caucasus though, as well as evidence of wagons, wheeled chariots, and metal horse bits dating to sometime between 2400-2000 BCE.

  2. Interesting for the implication that if ancient Anatolian speakers entered Asia Minor circa 3500-3000 BCE, they didn't bring any domestic horses with them, which would be mysterious.

    Is this a strong hint that Anatolian speakers didn't enter Asia Minor until well after they left the steppe?

    I'm in favor of an early separation of Proto-Anatolian, but it seems from this that the ancestral Anatolians either first arrived in Asia Minor without horses or they stayed in the Balkans for millennia before crossing over. That in turn raises questions about linguistic separation times. Something to think on.

  3. The history of horses & men are not going to be parallel all the time.
    The current paper favours an entry of horses into Anatolia pred. via the Caucasus. They seem to single out the Majkop culture ''The abundance of horse bones and images of horses in Maikop culture settlements and burials of c. 3300 BCE in the northern Caucasus led to the suggestion that horseback riding began in the Maikop period''
    At face value, Majkop doesn;t strike me as a horse-culture, but of course it did have many trappings of the SPR.
    However, the flourescence of multiple lineages c. 2000 BC seems to parallel the establishment of the Karum period, caravans, and need for transortation. So a holistic interpretation is warranted

  4. I'm not sure why they single out the Maikop culture, because there's no secure evidence for domestic horses and certainly not for horse riding in the Maikop culture.

    Horse riding seems to have become widespread in the Andronovo culture.

    And the parallels between the histories of the early Indo-Europeans and domestic horses are interesting, but what really matters is the prominent role of the horse in Proto-Indo-European religion.

  5. Majkop culture symbolises bulls and other powerful traction animals for erecting their large kurgans. Yes, Im not seeing the claims of the paper. From Reinhold et al 2017 - Despite the acknowledged lack of horse remains in Maikop contexts,..''

  6. So the horses entered with no riders? We already know there was no Bronze Age population turnover in Anatolia.

  7. @Romulus

    The domesticated horses came with people, who mixed into the local population.

    You'll see this in new samples from Copper and Bronze Age Anatolia.

  8. Conclusions
    Overall, our large-scale genomic analysis reveals two major genetic events. First, during the Late Neolithic, gene pools across
    Anatolia and the Southern Caucasus mixed, resulting in an
    admixture cline. Second, during the Early Bronze Age, Northern
    Levant populations experienced gene flow in a process that
    likely involved yet to-be-sampled neighboring populations from
    Mesopotamia. Even though we could detect subtle and transient
    gene flow in Arslantepe, we acknowledge that disentangling
    questions related to local-scale population dynamics within the
    homogeneous Chalcolithic and Bronze Age Anatolian gene
    pool might be beyond the resolution of current analytical tools.
    Furthermore, while our sampling expands in number and
    geographic range on previous studies, the critical area of Mesopotamia remains unsampled; thus, although our picture of the
    genetic landscape of the Near East is highly suggestive, it remains incomplete. Nevertheless, the cumulative genetic dataset
    of Anatolia, the Southern Caucasus, and the Northern Levant between the Early and Late Bronze Age indicates that, following the
    genetic events of the Late Neolithic and Early Bronze Age, there
    was no intrusion of genetically distinct populations in this region.
    This conclusion is of great importance with respect to our understanding of the formation of complex Bronze Age socio-political

    -Genomic History of Neolithic to Bronze Age
    Anatolia, Northern Levant, and Southern Caucasus
    Eirini Skourtanioti 2020

  9. @Romulus

    There's new data coming soon from Anatolia, and it backs up what I said in these blog posts.

  10. Well, Mycenaeans do have steppe ancestry (while Minoans don't), and there are new Mycenaean samples coming with even more steppe ancestry than the current ones.

    So no, I'm not grasping at straws.

    1. @Davidski

      Any ideas when the new Mycenaean samples are going to be published?

  11. @ Synome
    That in turn raises questions about linguistic separation times. Something to think on.

    There is not much to think on. The idea that Anatolian represents some archaic PIE is just a product of imagination of historical linguists who love making things up and later treating them as undeniable facts.

    They know that the end of this fairy-tale is coming, so they're already preparing the counter-attack by forging new "evidence" aka Anatolian names in Ebla.

    They're doing this, because if archaeogenetics will show that Anatolians separated from other Indo-Europeans e.g. in the Bronze Age, it will be obvious for everyone that PIE reconstruction and historical linguistics (at least in its current form) is just a manipulation that has nothing to do with real science, so their careers will be in danger.

  12. The conclusions by Skourtanioti et al are surprising in the face of their own data which shows population turnover in central Anatolia

  13. @Mayuresh Madhav Kelkar

    Please quit posting links to files on your computer.

    No one can access these files but you.

  14. that's not what this suggests

    Anahit Hovhannisyan 2020

  15. Davidski:

    "Please quit posting links to files on your computer."

    Noted. My apologies if that is the case. However this one has come from a forum directly.

    It might be on one Johanne Derite's computer and it worked fine for me.

    In in any case, this is a 2 year old conversation. So people here may already know about it. Just post the link if you like, but please do not post this entire message.

    Try this one too:

    If you decide not to post the above please post the following on my behalf. Thank you.


    "Please quit posting links to files on your computer."

    Noted. My apologies

  16. off-topic
    Neolithic outlier VK531 from Norway is pre-R-Y13200 (11 SNPs +, 2 SNPs -).

  17. @Romulus

    R1b levels in Anatolia today are around 12-18%. Only a single sample was R1b from the ancient results of Anatolia determined so far, and it appears to be linked to the Kura-Araxes culture as a minority lineage integrated from the steppes. Now I have no clue who the Hittites were paternally, or other IE speakers of western Anatolia, but I can be certain there was significant gene flow of Z2103 that was not linked to modern day Central Asian "Turks", as these lines appear to have already been in the BA Balkans.

  18. The R1b-Z2103 line penetrated through the Caucasus to Anatolia around 2000 BCE. this is the Trialetti culture of the Caucasus. Apparently they brought the domesticated horses to Anatolia.
    Are they the Hittites?

  19. I think that the Proto-Hittites and related groups moved into central Anatolia via the Balkans and Troy.

    Other Indo-Europeans moved into the Near East via the Caucasus, but their languages went extinct. I wrote about this expansion here...

    You'll see this reflected in new ancient DNA data, with the spread of steppe ancestry via Dagestan into Iran and surrounds.

  20. In theory, one may not exclude the other. After all, the division of Anatolians into Hittites and luvians takes place.

  21. @Vladimir
    Luwian and Hittite languages are forming a single branch. It is extremely unlikely that they followed different paths from the Steppe. Moreover, Luwian dialects are more influenced by non IE languages in the East than in the West, which is pleading for a diffusion from the West.

    This paper is rather coherent with an arrival of IE Anatolians from the Balkans in several waves during the IIId millenium.

  22. @Arza

    „There is not much to think on. The idea that Anatolian represents some archaic PIE is just a product of imagination of historical linguists who love making things up and later treating them as undeniable facts.”

    They have built history on XIX century fake news. And they still believe it.
    It is a pity that XIX century fake news caused so much confusion in linguistics and the history of culture and religion.

    As a result we have no idea what PIE religion was like. Some tried to link it to Indo-Slavic religion but surely Indo-Slavic religion was much later. It originated when steppe tribes started to arrive in Poland and mixed with TRB and other farmer groups mixed with HG. Then with Corded Ware its influences expanded to India, to Thrace and Greece. And with religion some linguistic influences as well:

    But what was pure PIE language and pure PIE religion without Indo-Slavic influences we don’t know.


  23. The main mainstream theory about the penetration of the Hittites-Luwians into Anatolia has always been that they did not penetrate there before Troy II. Namely, Troy II was occupied by the Hittites, Troy IV was occupied by the Luwians, and the Hittites had not bred horses yet, and the Luwians had already bred horses, apparently both for meat and for carts. The Hittite capture of Central Anatolia took place during the time of Troy VI.

  24. Here are those guys who introduced horses.

    Early Kurgan burial from Armenia with horse sacrifices.

    Notice current timing of layers is this.

    3600-2400BC Kura Araxes
    2400-2200BC Early Kurgans of Transcaucasia.Burials with horse.
    2200-1600BC Trialeti Vanadzor culture.

    Trialeti Vanadzor is obviously the continuation of that Early Kurgans.

  25. They did find some cheekpieces in Maykop.

    "It is thus doubtful that this very striking but mysterious culture had local roots. The settlement material alone would lead to this conclusion, but the complex of sites as a whole compels us to consider far m,ore complex and diverse formation processes. The metal sources, for example, point to the ore sources of central Transcaucasia. Parallels for this precious objects in the very rich burials lead us still further south, to Mesopotamia and Iran. The Maikop warriors were horsemen. This is demonstrated by the bronze psalia (cheek-pieces from bridles) that were found in a number of kurgans. Maybe their mounted detachments traveled far to the south in order to rob their rich neighbours, and brought back the highly artistic objects which we find in the burial grounds of their leaders in the Northern Caucasus. On the other hand, the combination of rude settlements with sumptuous burial sites indicates cultural closeness with the group of pastorialist communities of the northern steppe zone."

    Some points,
    Regarding Chernykh's doubt of Maykops "local roots", autosomally Maykop differed little from modern Caucasians, but they do show some deeper Mesopotamian roots which later became standard in whole of Caucasus. It is now also known that they had a symbiotic relation with Steppe maykop, who in turn were related to Botai, the first known horse domesticators.

    All in all I think that the "mounted divisions" are exaggerated, but if Maykop ancestry did appear in Arslantepe, one has to wonder how they got there and if they indeed had a role in introduction of domestic horses into Anatolia.

  26. @Aram

    That horse burial does not belong to the Kura-Arax culture time 3600-2400BC.

    quote "the horse was dated to the Hellenistic period (119 BC–26 AD), while
    the human skeleton was dated to 1886–1730 BC"

    CrM said...
    "They did find some cheekpieces in Maykop."

    These are not cheekpieces, it is generally incomprehensible that it has no analogues, moreover, it was found outside the context of time and burial, so it may even be Scythian. In general, this is essentially just a rumor, which has no confirmation, because it was found somewhere in the garbage dumps.

  27. @ Davidski

    Where does the Salzmünde Tabiano coloured Horse of +-3100 BC fit into this picture ? If Tabiano already was found so far West at roughly the same time that it appears near the Caucasus it could basically rule out Maykop as epicenter of domestication...or at least as the founder of the Tabiano Colour which is associated with domestication.

  28. The relatively late introduction of Tabiano coloured Horses into Anatolia is also interesting...Late Bronze Age migration or trade maybe from the Balkans ?

  29. DAVID

    Have you looked to that enormous paper about the Vikings in total?

    It seems a huge article but somehow is fuzzy. Also the leading author's name does not ring a "nice ring in the bell".

    After this post have you considered maybe checking out the article?

  30. The first wave of Indo-Europeans into Anatolia wouldn’t be the elite, who possessed the steppe derived horses. It would be the explorers, artisans, low level soldiers and peasants looking for opportunities. Only until horses became more common to the middle level of society, mostly likely brought from Balkanic tribes related to proto-Phrygians and Kaska, who defeated the Hittites once, and then again to the point of collapse during the Sea People’s era.

  31. Why is it that the Botai Horses seemed to have had all kinds of colour variations while their descendants that I have seen are predominantly Dun coloured ? Or do they still carry some of those genes ? I still think the Botai horses have some hidden story to tell...and I think that story will come from the West.

  32. "Donkeys of the Mountains" This kind of suggests to me that horses were mainly used in the Caucasus for the transporting of goods on their backs...maybe they had protruding backbones back then making them just as uncomfortable to ride as donkeys...

  33. Quick plot from supplementary spreadsheet of specific samples where X: date (Y is just random position to visually separate samples from same date):

    Green dots are pre-2500 BCE mt haplogroups, black dots post-2500 BCE mt haplogroups.

  34. @Sofia Aurora

    Have you looked to that enormous paper about the Vikings in total?

    Any ideas when the new Mycenaean samples are going to be published?

    Probably this year.

  35. @Ric Hern
    "Donkeys of the Mountains"

    This is a Sumerian word combination, rather means "Donkey from behind the Mountains".

  36. @Vlad

    How do you know this? The Armenian R1b probably entered via the western side of the Caspian flatlands. This is a separate source from the 15-18% of R1b in in modern western Anatolia (most of which is Z2103) R1b is actually lowest in eastern Anatolia ~12% which supports a western intrusion in Bronze or later periods.


  37. Another thing they completely missed in their study is that no later than 1600BC, Mitanni with horses from Central Asia come to Eastern Aanatolia, therefore, the hypothesis about the spread of horses through the Caucasus may be just seeming, they simply forgot about Central Asia.

  38. Toponyms in Greece suggest that you likely had Anatolian presence, possibly Luwians in the islands of the Aegaean as well as mainland Greece.

    Not too far from the Aegaean sea there was a Srubnaya-related man with R1a-z93 buried from 1700 bc or so.

    If Greeks only entered Greece relatively late after 2000 bc, they probably were hiding out in the Balkans which then could've had trading connections with early Indo-Iranians (Graeco-Aryan?) who were spreading across the Pontic-Caspian steppe, leading to the Srubnaya culture.

    We know Indo-Iranians were all about the horse, so I'd imagine there was a bit of a horse trading market going on. Perhaps that's why that's what that R1a-z93 guy in Bulgaria was up to.

    An issue would be that the Srubnaya culture only really takes hold from 1900-1800 bc onwards, and this article suggest Anatolia had horses by 2000 bc. But perhaps it's not an issue if horse merchants from further east were trading out west before settlers started moving into that direction.

    If Indo-Iranians were trading horses in southern Bulgaria where that Srubnaya male was found, it would be very close to where you could have Anatolian peoples, both in Asia minor as well as the northern Aegaean coastline.

    I think this is certainly more plausible than a Maykop related entry via the Caucasus, especially if the Troy site had the earliest domesticated horses in Anatolia.

  39. Whether the Proto-Hittites brought horses with them to Anatolia or not isn't very important. The Anatolian branch is linguistically an outlier compared to other Indo-European branches, so Anatolian speakers were likely to be Indo-European outliers in a lot of other different ways too.

    The important thing here is that modern domestic horses don't come from Anatolia or the South Caucasus, but it's very likely that they do come from the Proto-Indo-European homeland. So take from that what you will.

  40. @Copper Axe

    Yet R1b-Z2103 is found among EBA Serb and Bulgarians now, as well as one of the 3 core groups of Albanians, who weren't as heavily affected by later migrations to the Balkans. While I suspect R1a-Z93 had some presence in one of the steppe groups, as it seems to appear commonly among former Magna Graecia. I suspect Z2103 has as large, or a larger story to tell, but obviously we need to see a lot more data. Z2103 is also common in both mainland and Pontic Greeks, something that is not shared by many other haplogroups. Whether these were IE speakers, who knows? But why not? Since their obvious root is the PC steppes.

  41. “ The route across the Bosporus has been postulated on the basis of the earliest zooarchaeological evidence for domestic horses in the southern Balkans at the Early Bronze Age site of Kanligeçit around 2600 to 2300 BCE (8). The coat colors of 10 horses from this site were genotyped and revealed a highly biased distribution of coat color mutations with 6 of the 10 homozygous black horses (a/a), 4 of which also show the leopard spotting (LP), plus 2 bay-colored horses with leopard spotting; no chestnut mutation was detected (72). This pattern strongly contrasts with our results in Anatolia and the Caucasus, where chestnut (e/e) is the earliest coat color variant, while black and leopard mutants remain very rare (only 2 of 25). These differences argue against the introduction of a domestic horse population similar to that found at Kanligeçit. Moreover, there is no archaeological evidence for horse management in western Anatolia in the third millennium BCE, providing little additional support for the hypothesis of an early introductory route across the Bosporus (73, 74).”

    Mariya A. Balabanova
    Volgograd State University, Volgograd, Russian Federation
    Ivan I. Marchenko
    Kuban State University, Krasnodar, Russian Federation
    Natalya Yu. Limberis
    Kuban State University, Krasnodar, Russian Federation
    The article also deals with the possible relationship between the tribes of the Novotitorovka culture and the Azov-Black Sea sites of Catacomb culture. This conclusion is based on the results of intergroup comparison by the method of canonical analysis. The studied skull of the Novotitorovka culture has a morphological complex that characterizes the groups of burials of the Catacomb culture localized on the terraces of the Ingul river and on the terraces of the Don river left bank. This conclusion calls into question the archaeologists' hypothesis on the connection of the the Novotitorovka culture with the tribes of the Novosvobodnenskaya culture and the Maykop culture.

  43. A. N. Gey. Russian Academy of Sciences. Institute of archaeology: "It is assumed that the migration (or series of migrations) of some carriers of the late stage of the Novotitarovka culture to the South was one of the main reasons for the disappearance of the Kuro-Araks culture and the formation of a number of" Kurgan " cultures of the middle bronze age in Transcaucasia (the bedene culture and the trialet culture formed on its basis).”

  44. @Copper Axe

    "Not too far from the Aegaean sea there was a Srubnaya-related man with R1a-z93 buried from 1700 bc or so.
    An issue would be that the Srubnaya culture only really takes hold from 1900-1800 bc onwards, " ...

    What has the Srubnaya culture to do with it? This sample does not belong to the Srubnaya culture in any way. Horses were tamed long before the Srubnaya culture, they were already tamed before to KMK, Abashevo, Sintashta and were in them. The Srubnaya culture in the problem of the distribution of horses is not at all important at all. Any market is irrelevant here, horses spread along with the migrating people.

    @Vladimir, they absolutely did not give a single argument in favor of spreading into Anatolia because of the Caucasus, there is not a single genetic or archaeological fact, they only ignore Troy and Central Asia(Mitanni).

  45. Archi

    I didn't say that horses were found in Kura Araxes. Even authors had a doubt about one radiocarbon date being to old falling in KA period.

    As for one horse from tomb N9 being from Hellenistic period so what? There were other tombs with horses.

    As usual You are in denial of self evident things.

    Btw previous paper also confirms high diversity of horses in South Caucasus and near by.

    A Georgia MBA horse was close to Sintashta horse without Iberian admix. A horse from Hasanlu IA? did had an Iberian admix like the one in Hungary. So so.

  46. @Aram
    "As usual You are in denial of self evident things."

    What are the self-evident things? You are giving incorrect data, you wrote horses in the Kuro-Araxes culture 3600-2400BC. None of those horses belong to the Kuro-Araxes culture.

    "I didn't say that horses were found in Kura Araxes."

    "Aram said...

    Here are those guys who introduced horses.

    Early Kurgan burial from Armenia with horse sacrifices.

    Notice current timing of layers is this.

    3600-2400BC Kura Araxes
    September 17, 2020 at 3:33 AM

  47. I think the lower R1b in Eastern Anatolia is probably because Kurds have replaced Armenians. I doubt it was always the case. Especially since NW/N Iran have plenty of R1b-Z2103.

    1. Tigran- R1b-z2103 can be broken down into many, many more branches. Keep an eye on interaction between L584 z2108 Z2110.

  48. @Archi

    I only mentioned Srubnaya as it's the most proximate material culture that would've housed people such as I2163, hence the Srubnaya-related (as denying a relation would be kind if ludicrous), in the grand scheme of things it doesn't really matter if this guy was a Srubnaya male or not.

    Likewise, it also does not really matter what role the Srubnaya played, nor did I say they were responsible for the spread of domesticated horses.

    What I think matters is that Indo-Iranian related people (Potopovka, Sintashta, Srubnaya, Abashevo etc.) were probably moving into the direction of the Balkans without majorly settling there. Perhaps the migration of the Babino/KMK towards the Balkans is related to this, but their identity is far from secure. If I look at the Kikkuli texts or the adoption of the horse and chariot in China, I think the spread of horses to Anatolia might not be as simple as "people migrating, they bring horse" particularly since there are early steppe influxes in Anatolia yet the horse doesn't show up around 2000 bc or so. Which people, acquianted with horses migrated from Europe into Anatolia around 2000 bc? Seems more likely that the Anatolians acquired their horses from Indo-Iranian related merchants, either via the west, or the east (via the Mitanni).

  49. @Copper Axe

    We have no reason to associate I2163 with the Srubnaya culture, for historical reasons it can only be associated with Babino / KMK.

    We only have information that the horse came to Anatolia together with the Luwians through Troy. How the Luwians got the horses is another matter, borrowing technology does not imply trade at all, in general, here the issue of trade is not at all important. Simply, the Luwians raised their own horses. You can never prove that they got horses through trade.

  50. @Archi

    So you're proposing the Luwians only migrated towards Anatolia around 2000 bc then or something? Because if they lived there before that timeframe that doesn't match up very well with what this article is suggesting.

    Also why exactly would the sample have to be associated with the KMK? I know of their migration towards the Balkans, but we don't really know yet if the Babino were R1a-z93 folks or Z2103 folks, perhaps a bit of both, but I think it's more probable that they were related to the Catacomb culture.

    This sample is pretty much dated just after the Srubnaya have a big expansion across the Pontic-Steppe, and it's probably around that time we have lot's of Indo-Iranians horsing around everywhere, and if they were smart, which they probably where, they would specialize in horse breeding.

    But yeah this is all just speculation, not a hill worth dying on lol.

  51. We now know that Botai's horses aren't the source of our domesticated horses (and as far as I know, the horses of Antiquity aren't from Botai's stock either). We also know that Przewalsky horses are "feral Botai's horses", and not a major source of our horses.

    So who, when and where about ours horses origin? We did know for sure Sintashta people have horses, but probably horse itself was domesticated at least a thousand years before Sintashta's chariots. As I see it, Botai people could have influenced PIE pre-Yamnaya people about horse domestication (they saw it and guessed it was a really good idea to have horses), but some PIE people had domesticated horses by their own.

  52. @Siegfried

  53. @ Siegfried

    The Colour of Prezwalski Horses does not make sense when looking at the diversity of colours in Botai horses. Did Botai people prefer to eat coloured horses and only keep the Duns ? Or did coloured horses come from West of the Urals ?

  54. @Copper Axe

    "So you're proposing the Luwians only migrated towards Anatolia around 2000 bc then or something?"

    Namely, we have no evidence that they were in Anatolia before that time.

    "Also why exactly would the sample have to be associated with the KMK? but I think it's more probable that they were related to the Catacomb culture."

    Anthropologists are adamantly against this. Babinians and Catacombians differ from each other more the Arabs are from the Germans. Babinians did not differ from Fatyanovians at all, and Fatyanovians are R1a-Z93.

    "This sample is pretty much dated just after the Srubnaya have a big expansion across the Pontic-Stepp"

    Srubnaya never reached those areas to which Babino reached. It is impossible for such a young culture as Srubnaya to reach the south of Bulgaria by that time. In Babino, they also rode horse-drawn chariots like everyone else in the steppe at that time.


    Horse experts strongly deny that Przewalski's horses were ever domesticated. In fact, there is no real evidence that there were domesticated Przewalski horses in Botai, they have never been domesticated by anyone.
    The domestication of the horse took place just before the time of the Babino / Abashevo / Sintashta cultures (CWC time), as soon as it was domesticated, cheekpieses and chariots were immediately invented, because only such carts could be pulled by horses.

  55. @ Archi

    I think you forgot Poledrags and Sleds which preceded the carts with wheels....

  56. Isn't there a study that was supposed to come out analyzing the remains of Repin horses? Whatever happened to all that?

  57. @Copper Axe

    ''So you're proposing the Luwians only migrated towards Anatolia around 2000 bc then or something''

    up to date reports by field archaeologists clearly outline ''Judged by the archaeologically attested external relations of Troy with neighboring regions, the intensity
    of exchangeduring periods IV (2200BC) and V (2000BC) ...considerable trans-regional links can be determined with the northwestern, central, and southeastern Anatolian regions – as part of an introduction to Near Eastern innovations ''

    So, come 2000 BC, the main allogenic arrival was from southeast, diffusion of ideas &
    merchants associated with the developing caravan route & merchant colonies. As part of this, horses might have been introduced

    In northwestern Anatolia, as in parts of the Eastern Balkans and central Greece, the major demographic shift introducing a large corpus of IE speakers capable of sustaining their coherence and linguistic legacy for thousands of years occurred from ~ 3300 BC, after the period of demographic decline which occurred ~ 4000 BC (also throughout vast areas of S/E Balkans, and Anatolia ? due to seismic activities & flooding). So these people might have moved as family groups with wagons and bulls & the question of the horse in this regard is an entirely separate matter.
    {Concurrently, or perhaps earlier, groups linked with the Majkop phenomenon and K-A also diffused toward central & western Anatolia, bringing with them Arsenical copper technology}

    Given that the former are an earlier offshoot and might have been geneto-demographically distinctive to the later major wave of IE expansion, the established position of Anatolian in the family tree dovetails

  58. Olga
    There are some other native horses in western Europe, not related with the Przewalski horse but similar in looks, that live in a semi feral fashion.
    Basques call them Pottokas, and they are thought to be related with the horses painted in paleolithical caverns in the basque mountains.

    In basque the word for horse is Zaldi.
    Curiously, accordingo to Asko Parpola in his paper about "The formation of the Indoeuropean and Uralic language families in the light of archeology" the Hittite word for Chariot is Zalti coming from the Luwian "Zalal" meaning Wagon or Chart.

  59. Archi

    I posted those dates just for information to better understand the text. Maybe I should have been more descriptive.

    My point is that new horse mitogenomes were introduced in MBA via Daghestan and when Mesopotamian people coined the term ANSE KURRA they were referring to those horses. So here authors are right.

    While little bit later new mitogenomes came from Andronovo. Mitanni and later Iranians. Here I agree with You.

    All this events increased the diversity of mitogenomes in Near East.
    I abstain to comment about Balkanes because of lack of data.

  60. While IE languages seem to be from steppe, this paper isn't the slam-dunk for domestic horse origin in the steppe IMHO. Rather, it is evidence for the expansion and flourishing of equine domestication that began in Armenian Highlands.

    First of all, looking at mtDNA can only tell us so much because like today, stallions were probably used to "spread the right genes" to other horse populations. We needed Y-DNA to really solve this mystery and we didn't get it. At least not prior to the Bronze Age.

    Just going by mtDNA, look at Q which dominated the Armenian Highlands. Thousands of years later, it makes up a disproportionate share of the mares in Anatolia. It didn't just hang on, like P, it prospered. That indicates that this haplogroup was a part of the early domesticated population that prospered.

    The influx of diverse mtDNA linages probably did come from the steppe, but that doesn't mean that the domestic horse originated there. Rather, stallions from the original Armenian Highland domestic herds were taken to the steppe and spread their genes to the diverse population of wild mares in the steppe, very much like what the authors of the paper describe happening with mtDNA "P" in Anatolia.

    While the ancient Armenians may have used horses, they were not a horse culture until the IE came back over the mountains with a diverse and large horse population. Maybe it was the IE who came up with the idea of riding the horses themselves, and not just using them as a pack animal. This would be similar to the situation with gunpowder, invented in the east but thoroughly exploited in the west.

  61. @Mark Moore

    Rather, it is evidence for the expansion and flourishing of equine domestication that began in Armenian Highlands.

    There was no equine domestication in the Armenian Highlands.

    This is just your wild imagination.

  62. Can't wait for the corresponding study on Bell Beaker and Corded Ware horse mtDNA, oh wait there won't be since they didn't have any horses.

  63. David do you know if there are any upcoming preprints with paleolithic samples?

  64. @TLT

    Yes, but I won't be posting anything about them here before they're released.

  65. @Mark Moore

    "Rather, it is evidence for the expansion and flourishing of equine domestication that began in Armenian Highlands."

    Sheer nonsense. Learn to read. This article unequivocally proved that the horse came to the Armenian Highlands already in a domesticated form after 2000BC, even a wild horse did not exist there.

    "Just going by mtDNA, look at Q which dominated the Armenian Highlands."

    Do not walk nonsense and untruth. Learn to read. There is no Q in the Armenian Highlands, this Q is found in North Georgia at about ~2500BC, although the dating is obviously too old.

    You wrote a deliberate untruth. You will not be able to deceive anyone and pass off black for white.

  66. @ Romulus

    If I remember correctly there were Bell Beaker horses in Hungary...some distantly related to the Iberian Horses...

  67. @Davidski

    Not requesting any sensitive info, but is it likely that one such preprint will be made public before 2021 or no?

  68. I tried to find mtdna data from this paper where Duk2 from Hungary was presented but it seems in their excel files they don't give mtdna haplogroups in a usable form.

    1. Data Availability
      Sequencing data aligned against the horse reference genome EquCab2.0 (Wade et al., 2009) can be found on the European Nucleotide Archive, ENA: PRJEB31613

  69. @TLT

    The ASHG meeting usually releases some aDNA abstracs doesn't it? Maybe something Paleolithic from there in the coming month.

  70. Just to clarify some of my comments

    - ''In northwestern Anatolia, as in parts of the Eastern Balkans and central Greece, the major demographic shift introducing a large corpus of IE speakers capable of sustaining their coherence and linguistic legacy for thousands of years occurred from ~ 3300''

    But I do not imply proto-Greeks cam so early as 3000 BC, but rather widespread settlement shifts occurred at the nearly same time across those regions, but whose subsequent replacements were different in Anatolia c.f. Greece

  71. I'm not sure, but maybe a new article Kristian Kristiansen The Archaeology of Proto-Indo-European and Proto-Anatolian: Locating the Split
    “Dispersals and Diversification: Linguistic and Archaeological Perspectives on the Early-Stages of Indo-European (eds. M. Serangeli & Th. Olander), Leiden – Boston: Brill, 2020., 2020.

  72. Maykop is the natural choice for the Proto-Indo-European culture if you really want to stick the Proto-Indo-European homeland in the Caucasus.

    But why stick the Proto-Indo-European homeland in the Caucasus?

    I don't see any good reasons to do that. And, of course, ancient DNA tells us that there wasn't a strong relationship between the steppe-derived cultures that really made an impact, and are plausibly Indo-European, with Maykop or even Steppe Maykop.

    See here...

  73. As I understand it, Christiansen is saying that the proto-Indo-European-Anatolian language did not exist. A certain tribe from the North Caucasus was a native speaker of a certain language. Part of this tribe went to Anatolia and brought there the Anatolian languages, the other part already in the Pontic steppe mixed with other tribes from which the proto-Indo-European language arose. If I understand correctly, the role of this base tribe is claimed by CHG.

  74. Majkop remains a very interesting phenomenon. However, it's impact in Anatolia becomes swamped by the later expansion of Kura-Araxes phenomenon, which expanded via Arslanetepe into the Malatya plain, then further west as far as Greece. The Lefkandi ware which appears ~ 2500 BC there is influenced by the Red-Black burnished ware of KAC.

  75. Did some Steppe Eneolithic Tribes survive somewhere North of the Caucasus (Dagestan ?) while the majority migrated Westwards ? Maybe remaining hyper conservative for a significant time Linguistic wise while adopting certain aspects of their neighbours (Maykop and Steppe Maykop) through trade.

  76. Sounds like we are getting Trojan and more Greek + Aegean DNA soon from Harvard.

  77. @Vladimir

    All this is nonsense. All of this is based on the mistake who ascribed names to Hatti to the Hittite language for no reason at all. This is an inappropriate mistake for a scientist. What do we know? None of the names he listed have Indo-European etymology. The ending of these names does not coincide with the Hittite-Luwian names, but only similar in one sound of suffix. All known Hittite names derive from the Hatti names, as the Hittites took the Hatti names and gods. Finally, he does not find the names of the Hatti in the texts of Armi, not a single one, which is nonsense, since the Hatti certainly lived practically in the same place, it is simply impossible that they there were none, but he allegedly discovers the Hittite-Luwian names, while none of them has Indo-European etymology.
    And after that, writing that there the names of Anatolian IE is simply an unscientific heresy. They were required to provide the Hatti names and prove that those names were not Hatti, but the Hittites. Until then, all their reflections are just linguistic freaky.

    I repeat once again, no IE Anatolian names in 2500BC were found from the word at all, it was just an inappropriate assumption about some similarity of something expressed there in an indistinct form even without the slightest proof, which everyone perceived as some kind of discovery.

  78. @Romulus

    Really?Did They found time for ancient DNA?Now with the American elections...Btw where is that guy Lazaridis?I think i am going to create an amber alert for him hahaha

  79. @Rob

    I agree.But i am waiting samples from Trialeti.This culture is they key for what happened pretty much in eastern anatolia-armenian highlands during LBA/EIA.

    The Trialeti pottery style is believed to have developed into the Late Bronze Age Transcaucasian ceramic ware found throughout much of what is now eastern Turkey. This pottery has been connected to the expansion of the Mushki.

  80. @Davidski
    Do you plan on upgrading the G25 at some point to some more robust numerical method like this one:

  81. The Proto-Indo-European split: steppe Maykop groups introduced a superior cattle-based pastoral economy and transportation technology to the preYamnaya groups living in the steppe in the mid to late 4th millennium BC. This
    is reflected in the earliest burials with nose rings to control the bulls, pairs of
    oxen in burials, while in the steppe the wagon is more often deposited (Reinholdt et al. 2017: Figure 8.7). Sabine Reinholdt and her team document and
    discuss this transmission of skills to the steppe environment that by the late 4th
    millennium BC led to the formation of the Yamnaya social formation. I suggest
    that this transmission also included a language transmission corresponding to
    the formation of Proto-Indo-European, as a sister language to Proto-Anatolian.
    Once the new economy and technology was adopted to the steppe environment it was followed by a fast demographic and geographic expansion, whose
    final manifestation was the Yamnaya Culture starting around 3000 BC. By this
    time the speakers of pre-Tocharian separated from the remaining groups.
    This scenario implies that the mother language of both Proto-Indo-European and Proto-Indo-Anatolian should be pushed further back in time, perhaps
    into the mid to later 5th millennium. As stated in Wang et al. 2018:

    “Perceiving the Caucasus as an occasional bridge rather than a strict border during the Eneolithic and Bronze Age opens up the possibility of a homeland of PIE south of
    the Caucasus, which itself provides a parsimonious explanation for an early
    branching off of Anatolian languages.” Here I would replace PIE with ProtoIndo-Anatolian, and I would locate it rather in the northern Caucasus, but with
    possible cultural links south of the Caucasus that did not have lasting linguistic

    I think that's a really interesting hypothesis. One problem I see is that if Tocharian branched off early just as Proto-Anatolian, then shouldn't Anatolian and Tocharian be highly similar? But I believe that the opposite is true and they are starkly different given the basal appearance of Anatolian and the Centum features of Tocharian.

    On the other hand though the historical records of these two differ by 2000 years or so, making it hard to compare.

  82. @Mary

    No, because I've looked at the results in that paper, and they look kind of crap to what I'm getting with the Global25.

  83. @Romulus

    Which Indo-European groups actually have Maykop or Steppe Maykop ancestry? Can you list them?

    And how friendly was the relationship between Steppe Maykop and Yamnaya, if they hardly ever or maybe never mixed?

  84. Here you still need to understand what the author means by "early Maykop culture". In the past, this was called the 4700-4000 BCE period. Now this period is called the "steppe Eneolithic" or the culture of ringed pearl ceramics.

  85. Wonder if the equid remains of Alikemek-Tepesi were tested ? Will they ever test Horse remains from Samara and Sredny Stog ?

  86. What happened to Maykop is a mystery.Personally i do not believe this culture left any genetic impact somewhere in eurasia.Steppe Maykop on the other hand were just random individuals(You can call them outliers thought).A tiny minority lets say!Maykop completely vanished.West Caucasians and even North Caucasians require something more CHG shifted than Maykop.Dolmen might be the answer?There is not any serious reason left for people focusing in Maykop/Steppe Maykop.Catacomb also shows typical Steppe autosomal and nothing weird.

  87. Its kind of funny that these so-called Maykop steppe colonists were more likely the (un)friendly Yeniseian pastoralists Davidski once joked the western steppe herders were. Aside from the fact that there is no relation to the Maykop, there is nothing to suggest that these populations introduced a cattle based herding economy to the western steppe herders, who seemed more focused on the Balkans in the early stages of their development from hunter-gatherer to farmer/pastoralist.

  88. I was surprised about the word wagon in Hittite. Well because there are opinions that Hittites didn't have any word for wheel and possibly any word related to wagon making.
    So just checked various words related to wagons in Sumerian and here one interesting

    zardu [~WAGON] wr. ĝeš zar-du3 "a wagon part" Akk. zardû


    Imho Hittites separated before wagons were invented.

    Btw most Anatolian's words for horse also suspiciously look like loanwords from an Indo Aryan source.

  89. @Aram

    "Well because there are opinions that Hittites didn't have any word for wheel and possibly any word related to wagon making."

    There is no such opinion.

    Hittite: hurki- c. 'wheel', Tokharian: A wärkänt, B yerkwanto 'wheel'

  90. @Archi
    I did some research and apparently there were several findings of what people once assumed were cheek pieces in Maykop, but nowadays the general consensus is that those so called cheek pieces are actually nose rings for controlling bulls.

    "culture of ringed pearl ceramics"
    You mean Pricked Pearls Pottery culture? Isn't that essentially Darkveti-Meshoko, specifically the latter?

    @Anatolian Farmer
    "Personally i do not believe this culture left any genetic impact somewhere in eurasia."
    Apparently they did, assuming it's not a fluke because I imagine other contemporary West Caucasian cultures would have Maykopian-like ancestry proportions.

    In my opinion one of Maykops legacy is the linguistic similarity between Kartvelian and PIE, sam'qura being a good example.

    I think that Novosvobodnaya specifically spoke some para-Kartvelian language, while other Maykop branches (whoever OSS001, MK5001, MK5004 and probably SIJ001, SIJ002, SIJ003 represent) spoke something else.
    IMO the Meshoko branch of Darkveti were ancestral to Dolmen culture (and thus to NWC) and to non-Novosvobodnaya branches of Maykop. While West Georgian Darkveti to Proto-Kartvelians and to Novosvobodnaya.


  91. Hi Archi,
    Hitite: hurki Basque : Gurpil Eng: Wheel Sp:Rueda

    Sumerian zardu [~WAGON] wr. ĝeš zar-du3 "a wagon part" Akk. zardû
    Basque (horse) Zaldi/Zaldu

    Wagon: Sp Carro En : Cart
    Horse: Sp Caballo Horse, colt

    There are many pre historical facts we ignore.

  92. @olga

    You just quote some words that seem similar to your fantasy. Nothing that you cited is related, except Carro En : Cart. I'm referring to kindred words, not just similar words.

    But the words are related:
    Luwian: zalal "chariot" < PIE: *kʷel-"wheel"

    Hittite: h.l. aś(u)was, pl. aśuwai "horse", Lyk esbedi "horse troup" < PIE: *ek'w-"horse"
    Hittite: h.l. śuwanis "dog" < PIE: *k'wen-"dog"
    Hittite: śuppi "pure" < PIE: *k'ubh- > Old Indian śubhá- "pure"
    etc. (phonetic law: PIE: *k'u- > Hittite śu, PIE: *kʷ- > Hittite ku)

  93. It is beyond me why some people equate the use of a bit as signs of first domestication. As far as I understand the Bitless Bridle were used in Mesopotamia as early as 1400 BC. This was basically the basic plan which they most probably inherited from the earliest horse domesticators.

    Once upon a time I broke in a Pony for my friend using a self made bridle and thick leather strap bit I invented myself. It worked for some time until be could buy some nice ones.

    That is why I don't understand the obsessiveness with a bit and domestication.

  94. @CrM
    “ You mean Pricked Pearls Pottery culture? Isn't that essentially Darkveti-Meshoko, specifically the latter?”

    Yes. This is a very interesting culture. Now it is also divided into two phases. The first, which arose under the influence of the Ubaid culture of Mesopotamia, the second phase has similar features to the Sredniy Stog culture and the Khvalynsk culture. It is probably during this period that the CHG penetration into the steppes occurs and the CHG/EHG/WHG population begins to form.
    What else is interesting. All these processes involve R1b-V1636

  95. @Vladimir

    There are samples from the Neolithic (~5,000 BCE) from the Don/Volga area well north of the Caucasus already with ~50% CHG.

  96. Thanks. Is this new data from a study on the don river region?

  97. I wasn’t aware of Meshoko culture having any direct relation to the Ubaid.
    It rather stems from Darkveti , with rather crude handmade pottery . It’s high CHG affinities is consistent with that

  98. Archi

    Kozyntsev was saying that Hittites didn't knew wheel.

    And aren't You surprised that the so called Hittite (or Luwians?) word for horse is satemised?

    Btw Hittite were using the Sumerogram ANSE.KUR.RA for denoting the word horse which means we can't know it's exact phonetic pronunciation.


    Hittite: 𒀲𒆳𒊏𒍑 (ANŠE.KUR.RA-us /ekkus/) (only attested with Sumerogram)
    Cuneiform Luwian: 𒊍𒍪𒉿𒀸 (az-zu-wa-as /azzuwas/)
    Hieroglyphic Luwian: 𔑮𔐓𔗵𔗬 (EQUUSá-sù-wa/i)
    Lycian: 𐊁𐊖𐊂𐊁 (esbe)

  99. @Crm

    Maykop is a culture created by immigrants from K.A or Leyla Tepe or something in between.Their roots have eastern anatolian/armenian highlands/Azeri Highlands or even Northern Mesopotamian origins.In this area pretty much as a whole.Meshoko and Darkveti or even the later Dolmen have their origins somewhere in the W.Caucasus or NW Caucasus parts.The reason i am saying that Maykop culture didn't left a big genetic impact to modern Caucasians is that we seeing Iran N to be limited to almost absent.I am waiting samples from Eastern Georgia,Caucasian Albania and Trialeti culture as well to have a better view for west asians.

  100. @Rob

    On this occasion, a little is written in the work of Korenevsky S. N. " Common for these settlements are stucco round-bottomed vessels, similar to the ceramics of Yaseneva Polyana (Meshoko culture), reconstructed in a similar way from the Hearth grotto and vessels with an S-bent Corolla and flattened bottom. However, their forms are quite simple. They find a wide range of analogies, but more in the cultures associated with Ubaid-Uruk influence in the Caucasus.
    Thus, round-bottomed vessels in the South Caucasus before the Kuro-Araks culture appeared were found in the lower layer at the Berikldeebi settlement. They are typical of the East Anatolian monument Arslan-Tepe VII0'.
    But older forms of a similar configuration are found in the North- Mesopotamia layer Ubeid period of the Yarym-Tepe III monument, which N. Ya. Merpert and R. M. Munchaev refer to the time of Tepe-Havre XVIII-XVII, that is, an earlier period than the Uruk time in Mesopotamia.
    Therefore, we can raise the question that the origins of this form in the Western Caucasus tend to the North Mesopotamian traditions of the Ubeid period. But whether they appeared as a result of influences at this time from the South or for other reasons is not very clear. In this case, it is hardly legitimate to exclude the possibility of alien migration from the South."

  101. @Anatolian Farmer

    Yes, whenever Maykop left a genetic trace unto modern Caucasian populations is a tough question, and I agree with you on that one.
    I also agree with Maykop-Leyla Tepe link but I disagree with KAC, in any case we have LT samples and they are very poor in CHG ancestry (much like every sample so far in Azerbaijan) to be the sole continuous link between LT and Maykop. Evidently LT needs something more CHG-rich like Darkveti-Meshoko in order to form the autosomal profiles of Maykop.
    The haplogroup L,T and possibly G2a2 that are found in Maykop likely stem from Leyla Tepe or Shulaveri Shomu, while J2 from Darkveti-Meshoko. I'm curious when G2a1 makes its first appearance in the Caucasus, so far the oldest Caucasian G2a1 is found in Iron Age Koban culture, but the clade must have been present in the Caucasus even prior to that.

  102. @Aram

    I do not know if Kozintsev said such nonsense or not, but this is definitely a mistake. Kozintsev is not a specialist, he is not an authority it this, he is a very old anthropologist and only an anthropologist who does not understand other issues, especially in the Hittite language and culture.

    It is not satemized, it is a phonetic law that I gave you. Hittite and Luwian have many words with its examples. Hittitologist V.V. Ivanov so simply asserts that the Luwian language is Satem.

    Wikipedia is not about anything.

  103. @ Crm

    Shulaveri Shomu is Copper Age.After that you have something different like K.A and Maykop and their autosomal is not exactly the same.Shulaveri Shomu is like a more west asian shifted BA Anatolian version.Farmers from Anatolia pour out to caucasus and west asia.In Kura-Araxes we seeing an increase in CHG/Iran N components especially to the former.You are right in some way that Shulaveri Shomu played a role to the formation of K.A/Maykop and even to L.T but they are not identical.Their farmer admixture might be related to Shulaveri Shomu but the components of CHG/Iran N looking to have arrived there from different directions.CHG probably come from northwest while Iran N probably come from an Iran Chalcolithic source.After all Kura-Araxes is half Iran C related.So,with a few words these cultures after the Copper Age(Shulaveri Shomu) are a mix of different sources from west to east.I agree with your statement about the yDNA.G2a is definitely associated with farmers from Anatolia and i am pretty sure some lineages arrived there during LBA/EIA.

  104. @ Vlad
    Thanks, Ubeid influences via Eastern Anatolia to West Caucasus makes perfect sense.

  105. @CrM

    Are you speaking of the autosomal components of "ALX002", "POT002" and "MTT001"? Do you know where I can find the frequencies for the autosomal data?

    Regarding your comments on the Y-haplogroups. The Tell Kurdu (Halaf and Ubaid) individuals were very close to the Chalcolithic and PPNB individuals from the Levant. The 3 Tell Kurdu males belonged to H2, G2a2, J1a2a. And the Chalcolithic Peqi'in Cave individuals (whom are autosomally very close to the Tell Kurdu individuals) belonged to T. The PPNB Levantine individuals had H2 and T. And with more individuals tested we surely will see L among these groups.

    Could it be that the Halaf and the Ubaid people were the native people from Northern Mesopotamia with Y-haplogroups J1a2, H2, G2a2, L, and T? And the pre/proto Darkveti-Meshoko period is defined by J2 with high amounts of CHG?
    Could it be that the Halaf/Ubaid people with Y-haplogroups J1a2, H2, G2a2, L, and T (with lower amounts of CHG) came into the Caucasus and formed the Leyla Tepe and the Maykop cultures? As you said, except H2, we can see J1a2, G2a2, L, and T among the tested Maykop individuals.

  106. What do people think of Areni Chalc ? Seems a bit out of place in South Caucasus

  107. @Sarah

    That's interesting. I wonder why the majority of paternal lineages in ancient West Eurasian DNA mostly derive from East Eurasians (H, LT, K2b/PQR)?

  108. @Sarah
    Here are the G25 models:

    Autosomally ALX002, POT002 and MTT001 are much closer to Chalcolithic Hajji Firuz samples, who are representative of the Dalma culture. The culture's material presence is attested in the Caucasus. Dalma in turn is a mix of Mesopotamians and Zagrosian farmers, the source of many Caucasian haplogroups could stem from both of those sources, including L, T, G2a2 and most likely G2a1 (it was found in a late Neolithic Iranian farmer from Seh Gabi). When it comes to J2, some (if not all?) of the clades in Darkveti-Meshoko later made an appearance in Maykop.

  109. @Rob
    Areni Steppe ancestry is a mystery, but without it it's akin to the samples from Arslantepe.

    Distance to: Areni_noSteppe
    0.03678099 TUR_Arslantepe_LC
    0.03857428 TUR_Arslantepe_EBA
    0.04106972 TUR_Camlibel_Tarlasi_LC
    0.04183460 TUR_Ovaoren_EBA
    0.04314770 TUR_Kaman-Kalehoyuk_MLBA
    0.04331244 TUR_Alalakh_MLBA
    0.04364118 TUR_Ikiztepe_LC
    0.04580644 IND_Roopkund_B_o
    0.04765671 Levant_LBN_MA_o3
    0.04790440 TUR_Isparta_EBA
    0.04912346 SYR_Ebla_EMBA
    0.04959762 TUR_Titris_Hoyuk_EBA
    0.04989031 AZE_Caucasus_lowlands_LN
    0.05318802 TUR_Barcin_C

    I remember FrankN stating "Besides, the Eneolithic Dniepr Rapids area has yielded finds of Armenian obsidian, so there definitely existed a (pre-Maykop) trans-caucasian trade link during the Eneolithic."
    Armenian and Georgian obsidian deposits were connected with North Caucasus through trade. Darkveti-Meshoko, Areni and possibly Leyla Tepe have traces of Steppe ancestry, and the former used to built the "chalcolithic fortresses" to fend off people that were "more native, more primitive" than them (Progress, Vonyuchka?). Could be that the Steppe ancestry reached South Caucasus through slave trade from further North.

  110. @ CRM

    Thanks. I don;t think the steppe ancestry is the mystery. If there's south Caucasian ancestry in north of the mountains, then there's going to be steppe Eneolithic ancestry south of them . And of course, Progress _Eneolithic fits the bill

    It is rather the high Anatolian-related ancestry & relatively low CHG in Areni_C. which strikes me. I guess its relation to Arslantepe could explain it, although in reality its probably a more complex interaction network betweewn Central & Eastern Anatolia and the south Caucasus. These seem like non-local groups which arrived from E Anatolia, and then rapidly engaged in contact with steppe communities

  111. Could Areni_Cl haplogroup L1a really be immigrants from Mesopotamia?

  112. @Rob
    Yep. Most of their non-Steppe ancestry seems to be derived from SSC, with some extra East Anatolia. There's one outlier who has a lot of Hajji Firuz C and no SSC.

    Here are the coordinates if anyone is curious.


  113. In the recent bronze age serbia paper, there is:

    1x J2b-L283
    4x R1b-Z2103

    The J2b-L283 is in a prestigious grave buried with a stone battle ax head. If i recall correctly the R1bs are not in prestigious graves.

    If there is a movement of chg into steppe, is it possible J2b-L283 is the mediator?

    J2b-L283 and R1b-Z2103 seem to have come from the steppe together to the balkans that is pretty evident from the serbia paper and distribution in Albanians.

  114. @Dita

    I'm not aware of J2b-L283 in currently or soon to be published samples from Yamnaya, Corded Ware and closely related cultures from the steppe.

    It seems to me that J2b-L283 was incorporated into steppe-derived populations in the Balkans or maybe just north somewhere.

  115. Vladimir

    I think You speak about Chaff faced ware. Right? It was found in Berikldeebi also.

    We have four samples from Arslantepe VII layer. Chaff period in Arslantepe. 04,05,39,42
    And the interesting thing is that Arslantepe 039 seems to have some minor Steppe ancestry. Maybe she got it from Areni. Because that Areni also had Chaff faced ware.

    And I agree with comments that there is no puzzle for the Steppe in Areni. It is from those Steppe Neolithic Progress type people. Not surprising we already have two V1636 from Near East. Both _after_ Chaff faced ware period.

  116. KDC001 from Balkaria, North Caucasus is L283. I can see it being a part of late Catacomb lineages. The Kubano-Tersk samples have some noticeable Bronze Age Caucasian input in them.

  117. Vladimir

    The L1a in Areni has definitively more southern origin but is it from North Mesopotamia or just from neighbouring NW Iran is less clear.

    On the other hand here and in other places I made the case that E-M84 found in MBA Armenia is an old migrant from North Mesopotamia even before KAC.
    And now we have E-M84 in LC Arslantepe.

    So there was a probably genuine influence from North Mesopotamia in the formation of that Chaff faced ware ( Arslantepe VII, Areni, Azerbaijan LC) but the bulk of ancestry of this people was still local. Anyway the J2 CTS900 found in Meshoko was also found in Arslantepe but I wouldn't bet much on it because we don't know how widespread was CTS-900 in those time. It is quite possible that CTS-900 was just a local S Caucasian foragers lineage who learned farming.

  118. Concerning the motivations why should Progress related people appear in South Caucasus.
    It is hard to me imagine a slave trade in Eneolithic but is more easy to imagine that the arrival of Q rich herders from Central Asia who in most likelihood had more militaristic culture would had a serious impact on locale Progress V1636 people. They were simply forced to move away.
    Other option is the trade.

    Anyway both that branch of Q from Steppe Maykop and that R1b V1636 are practically absent from modern north Caucasian people. But are found in South Caucasus. Which means that some patterns of migrations repeated in Caucasus more than one time.

  119. L. S. Klein wrote about the origin of the Shulaveri- Shomutepe culture: "This culture arose shortly after a sharp cold snap that broke out around 6200 BCE and lasted for about 2 centuries. It probably destroyed the previous Neolithic culture. Round mud buildings of shulaveri-shomutepe, ceramics with plastic ornaments, female figurines, long obsidian knife – shaped plates indicate the relationship of the culture with the cultures of the middle Eastern Neolithic – Hassuna and Khalaf and suggest that the population after the death of the local Neolithic came from there, making up at least part of the Eneolithic population of the Caucasus."

  120. @ CRM

    '' Most of their non-Steppe ancestry seems to be derived from SSC, with some extra East Anatolia. There's one outlier who has a lot of Hajji Firuz C and no SSC.'

    I wonder how representative the Menteshe tepe and Areni C. are of Shuvaleri-Shomu processes. Menteshe tepe and the other Azeri lowland samples look different, more Mesopotamian than Areni. But then again, the Azeri lowland could have been part of a distinctive Iran-LN zone.
    Guess we'd have to get some Syrian Neolithic & Armenian Shuvaleri samples too

    @ Dita

    ''The J2b-L283 is in a prestigious grave buried with a stone battle ax head. If i recall correctly the R1bs are not in prestigious graves.''

    Also, the chief with a bronze dagger was I2a1b. Seems different dynamics were operating in SEE

  121. @All

    Preliminary G25 coords for most of the new Viking samples are available here.

  122. Interesting, those Vikings from Poland_Bodzia are not Scandinavian-like. They're rather Slavic. The closest modern samples for comparison:

    0.02973276 Russian_Smolensk

    0.02862656 Slovakian

    0.03935087 Polish

    0.03666865 Polish

    0.04852882 Cossack_Ukrainian

    Local Slavs who have adopted Viking culture?

  123. With Poland_Sandomierz and Poland_Cedynia:VK212 it's similar:

    0.02992285 Belarusian

    0.02415055 Belarusian

    And strangely, Poland_Cedynia:VK211 could be even some kind of German:

    0.04130935 French_Seine-Maritime
    0.04159328 Swiss_German
    0.04299900 French_Nord
    0.04350448 French_Alsace

  124. A sensible model for the Normand French from Seine-Maritime:

    Target: French_Seine-Maritime
    Distance: 2.8228% / 0.02822769

    41.6 Picardy_La-Tene
    37.4 Denmark_Langeland
    21.0 Dutch

    Picardy_La-Tene reflects the Belgic ancestry from the Caletes and Veliocasses, Denmark_Langeland the Danish Viking admixture and the modern Dutch are used as a surrogate for the Franks.

  125. Using scaled data from Davidski's upthread of Viking samples, run all those samples through Vahaduo distance with all other G25 samples as source:

    Top 25:
    Lowest Overall Distance Only:
    Lowest Overall Distance Only, Present Day References Only:

    Of samples; Best overall matches to present day countries are in descending order: Swedish: 67, English: 31, Orcadian: 22, Norwegian: 13, Irish: 12, Icelandic: 10, Dutch:10, Latvian: 10, French: 9, Polish: 8, Belgian: 7, Belarusian: 6, German: 6, Scottish: 6, Estonian: 5, Finnish: 4, French_Brittany: 3, Ukrainian: 3, Lithuanian: 2, Russian: 2, Slovakian: 2, Austrian: 1, Croatian: 1, Hungarian: 1, Italian_NE: 1, Welsh: 1.

    Bear in mind that this will favour matching to countries with larger numbers of samples in G25 somewhat! E.g. if 64 Irish, 41 Polish, 32 English 22 Swedish samples, only 10 Orcadians and 7 Norwegians, that favours the more numerous sets to be matched. Plus standard blurb about whether representative etc.)

    Will show which samples are close to already known samples, and which are a bit distant, and those which might be location atypical.

    Scaled match is generally about 0.02-0.03 distance.

    Regional subsamples and where they visually plot on Vahaduo's Europe1 PCA: . Samples seem to overlap most frequently to Britain, Ireland, Norway, Netherlands area of plot (I expect Danish vikings would show better overall match to Denmark but not on G25). There's definitely some trailing down into Northern and Central France though. There's definitely some trailing into Eastern Europe as well, though that's mostly accounted for by the Gotland samples and by VK sampled from Eastern Europe. In most regions there is some overlap with present day people - least so in the Eastern Europe VK samples; Estonia, Poland, Russia, where some overlap but not very much (although samples do have some eastern shift, typically, relative to the set as a whole and quite a bit of dispersal!).

  126. It would be also cool to have some specific samples from Varangians.The people who ruled 'Kievan Rus'.Kinda funny if we thing that the biggest royal houses during middle ages and even later have Viking/Norman roots.These people might not left anything big 'civilization' etc,but the conquered the old world.From west to east they left their impact, also assilimating other folks adopting their culture,traditions and battle skills.

  127. Davidski, are you going to create an open thread so people can share results for Viking G25 samples?

  128. Is the ANA component considered to be a SSA component or a sister branch of Eurasians? I've seen claims on Anthrogenica that both ANA and Basal Eurasian are both SSA components.

  129. @Samuel

    Yes, once I have the whole dataset. But I'm still missing about a third of the samples, and the official dataset hasn't even been released yet.

  130. David, I don't see samples with Slavic clades on Matt's PCA. Are they in the missing third?

  131. Global 25 North Europe PCA show East –West Europe division along PC1. What % of variation PC1 and PC2 explain?
    Looks like all ‘Slavic’ Vikings are a mix of Swedes and Finns or Swedes and Slavs or are pure Slavs:

    It will not solve Rurik problem. There are three theories:

    1. Rurik was Norman I1
    2. Rurik was Finn N1c
    3. Rurik was Slav from Rethra R1a

  132. @Simon_W, re; Polish VK samples actually have to modify my post above, I think you're right about that.

    Visually the Polish VK samples looked a bit north positioned relative to present day Poles in G25, but barring one outlier (VK211*) the distances to closest Polish sample are not too different (median: 0.033, range: 0.025 - 0.045) from e.g. closest VK samples in UK to present day UK samples (median: 0.026, range: 0.017 - 0.033).

    On outlier VK211 (who as you note seems closest to the French) he is actually classified in the supplement as Medieval from 11-13 centuries CE rather than Viking era, e.g. "We also sampled a post-VA grave, No. 435, for comparative purposes. Burial in a wooden coffin with no grave goods 104,105. An adult male, oriented east-west, facing east (a typical Christian orientation of that time) and buried in a supine position, on the back, arms along the body, legs straight. The individual dates to the Middle Ages (11th-13th centuries CE).". He also has R1b1a1b. No grave goods, but in Christian Medieval period it would be unusual to be buried with weapons or anything like this. Possibly some kind of far-migrating merchant / vassal warrior from France? (Somewhat unexpected / lucky for them to find that by chance though!).

    Interestingly a very elite grave from same site from Viking Age, Grave 558, is rather Belarusian like in G25. They note: The individual has also been subjected to a strontium isotope analysis. Its isotopic signature falls within the range of 87Sr/86Sr values occurring in the postglacial lowland areas of the southern shore of the Baltic Sea. It is likely therefore that the individual was of local origin, from Cedynia itself or from elsewhere in West Pomerania. and also The individual was facing east (a typical Christian orientation of that time) and buried in a supine position, on his back, arms along the body, legs straight.. A local elite guy who was part of a group that had converted to Christianity, and Scandinavian Viking customs with it? He has R1a1a1b.

    Actually would note, not that may but a fair few of the VK samples are from outside Viking period. So it may be worth using the supplementary table to sift out the non-Viking samples before we develop too many ideas about Viking Age from samples!

  133. Quick look at where some of the lower frequency (for north) y-dna haplogroup samples from Viking Age sit in PCA:

    (Treating E1b, G2a, T1a, J2a as lower frequency here; there are also N1, I2 results which could be explored. Also someone could look into rare Mtdnas as well, but I don't have a good sense of what those would be).

    Sometimes they plot in unusual positions (probably reflecting recent ancestry from outside core areas), and sometimes in more expected positions.

    Separate plots of all male and all female Viking Age samples on Vahaduo Europe1: . There is some slight tendency for the male samples to concentrate on a narrow band around Norway/North Dutch today (and perhaps around Denmark, if I had Danish samples on here), but the distributions of samples are generally similar.

  134. @Simon_W

    Seems quite Germanic, makes you wonder about the rest of Northern France.


    Based on evidence from this blog, Rurik was N1c. But I wouldn’t expect you to believe it, even if Dave said it himself.

  135. @Gabriel

    The oldest Rurikid Gleb Svyatoslavich who lived in the XI century (died in 1078) has a purely Slavic haplogroup I2a1a2b1a1a. No N1c was found in ancient DNA of Rurikid remains.

  136. @Matt @Simon_W

  137. @G2a-M406 Anatolian Farmer

    From the Margaryan's work we have "Rurikids"
    Gleb Svyatoslavich - I2a-L621 (> Y3120)
    Izyaslav Ingvarevich - R1a-M458> L1029
    From another work Yaroslav Osmomysl - E-V13, plus there are also rumors that there is DNA data from Chelm, a possible Daniil Galisi's son, and that none of the DNA directly indicates a relationship with the modern Rurikids.
    In Ladoga, the "Rurikids" haplogroup was not found also.
    The thing is, the Russian chronicles were repeatedly edited, and the legend about the "calling" of the Varangians and Rurik was later inserted. Early sources do not know Rurik.

  138. @EastPole

    PC1 57.9%
    PC2 10.5%

    But these values are only true for the full sample set that was used to create the PCA.

  139. @Simon_W

    I also noticed this, but by haplogroups. They were attributed to the Vikings, most likely because they (Bodzia) were buried in the chamber graves. This is a good counter-argument for the statements "only Scandinavians can be buried in chamber graves, they were the elite, and the Slavs are poky farmers". The tradition of chamber graves in Poland came from Rus'. I wonder from where Rus' inherited chamber graves burial tradition? The thing is, if this rite was brought from Scandinavia, then it should have been in Ladoga, but there were boat burials. Yes, there is a mound No. 11 in the Plakun burial ground with an early dating, but no inventory was found, and the dendrodata of logs became the only argument for the early dating of this burial ground. However, there are numerous examples where the logs was used for a second time. Apparently, in this case, the felling date is mistakenly used to determine the time of the burial itself.
    So now there are doubts about the early dating of this burial, and even the attribution of this burial.
    Chamber graves burial tradition go in the opposite direction, first they are recorded in the South (Kiev, Gnezdovo), then in the North, in Ladoga. Conversely, there are no boat burials in Kiev.

  140. @Matt

    "On outlier VK211 (...) he is actually classified in the supplement as Medieval from 11-13 centuries CE rather than Viking era, e.g. (...) Possibly some kind of far-migrating merchant / vassal warrior from France?

    I kind of suspected it, that he is post-Viking era. 11-13th c. CE makes sense: the German settlement in this area started between 1200 and 1250 CE. I would guess he may be a Fleming. Linguistic and other evidence suggests an appreciable Flemish component in Brandenburg and along the lower Oder:

  141. @Gabriel

    "Seems quite Germanic, makes you wonder about the rest of Northern France."

    Since Viking admixture in France matters only in parts of the Normandy, I obtain comparable models for other northern French regions using just Picardy_La-Tene and the modern Dutch as sources; in that case the French_Nord are 42.6% Dutch-like, the French_Pas-de-Calais 53.8%, and the French_Paris 34.0%.

    Also, subtracting the Viking admixture from the French_Seine-Maritime in order to calculate the pre-Viking-admixture proportions yields 33.5% Dutch-like, which is similar to Paris, and lower than in Pas-de-Calais and the Nord. Which makes sense, as it's farther away from the Frankish homeland.

    Possible Roman admixture may complicate things however, and it's somewhat unfortunate that with the Dutch I have to resort to a modern sample. I had tried DEU_MA for the French_Seine-Maritime, but Vahaduo didn't pick it up.

  142. For sake of completeness, all Viking Age male samples by y haplogroup, on Vahaduo Europe PCA: , Pastebin: .

    Some disjoint in distributions / density.

  143. @Tigran

    What makes you think that LT is east Eurasian?

  144. @Simon_W,

    There's lots of variation in the Iron age samples from France. So we have to take any population average created by it with a grain of salt. I'd rather use modern North French, like France Nord, as proxy for the non-Viking non-Dutch ancestry in modern Normans.

  145. @Draft Dozen
    "The thing is, the Russian chronicles were repeatedly edited, and the legend about the "calling" of the Varangians and Rurik was later inserted. Early sources do not know Rurik."

    This is just not true. They have not been edited, and the earlier ones do not exist.

  146. I'll have a look at the IA French individuals later. But modern French_Nord of course includes ancient Frankish admixture on top of the ancient Belgic layer. But that said, I guess it may work as a surrogate for pre-Viking admixed Normans.

  147. @ Draft Drozen

    Rurik the name itself sounds Germanic/North Germanic to me.I am not sure if the dynasty's origins is Germanic or Slav but when it comes to autosomal DNA Eastern Slavs do not show Scandinavian admixture.So,we are talking probably for a small 'elite' warrior class group and nothing more.Their impact to Eastern Slavs was limited to almost none.

  148. @Archi

    Shakhmatov A.A. had a different opinion

  149. @ Simon

    The big question here is how Franks were genetically when they invade the Gallo-Roman world.They were Scandinavian like?Or they already had Celtic affinities?We need samples from Clovi's Franks and even before the battle of Soissons.Something is telling me that they were already in contact with Gauls before their conquest in Roman Gaul.I would say the same for other germanic barbarian tribes during that period like bavari,alemanni, Thuringii etc..they definitely mixed with southern groups on their way so they started to lose their Scandinavian like profile.

  150. Taking up Sam's suggestion:

    Target: French_Seine-Maritime
    Distance: 2.0704% / 0.02070430

    77.2 French_Nord
    22.8 Denmark_Langeland

  151. @G2a-M406 Anatolian Farmer

    "The big question here is how Franks were genetically when they invade the Gallo-Roman world.They were Scandinavian like?Or they already had Celtic affinities?We need samples from Clovi's Franks and even before the battle of Soissons.Something is telling me that they were already in contact with Gauls before their conquest in Roman Gaul."

    Yes, some admixture has to be expected, as the Salian Franks had been living for about a century as Roman foederati in Toxandria before they started their conquests.

    As for the other continental West Germanic tribes, they were still quite northern, judging from the Baiuvaric evidence:

    STR_486 = Danish-like
    STR_480 = German-like
    STR_316 = Icelandic-like
    NW_255 = Danish-like
    ALH_3 = Dutch-like
    ALH_10 = Orcadian-like
    ALH_1 = Dutch-like
    AED_249 = Danish-like
    AED_1135 = Afrikaner-like
    AED_106 = Swedish-like

    while their average coordinates are Dutch-like. However note that there is no North_German sample in the Global25 sheet, but the Eurogenes K15 sheet does have it. And according to Eurogenes K15 data the average of the same Baiuvari is closest to North Germans; the Dutch follow on the second place. So they were not exactly Scandinavian-like, but still North-German like and thus quite northern.

  152. Is anyone else having trouble with umap or is it just me? The samples are where they should be but the map itself isn't loading.

  153. @Anotolian farmer that's indeed a big question they came from the Rhine Delta and the Sallian ones just above the Rhine in the Netherlands from Salland. They were Northern European but Scandic like mmmmm????

  154. Early Franks are probably similar to the Anglo Saxon samples, which if I remember correctly had a unique North Euro composition including an unexpected affinity to Balts.

  155. @Tigran,

    "Is the ANA component considered to be a SSA component or a sister branch of Eurasians? I've seen claims on Anthrogenica that both ANA and Basal Eurasian are both SSA components."

    ANA is neither a SSA nor a Eurasian component but something in between. However, I suppose that ANA is a precursor of the Eurasian lineage. ANA people seem to have split from non-Africans after the ancestral populations of Sub-Saharan Africans did, but before the non-African genetic bottleneck. That being said, there is an affinity between ANA and SSA that is probably due to the fact that SSAs harbor some ANA admixture.

    Basal Eurasians are not an SSA component either. What defines Basal Eurasians as non-African is their participation in the Out of Africa bottleneck. Furthermore, after Basal Eurasians split from other Non-Africans they remained in the Middle East without mixing with Neanderthals.

  156. @ Romulus

    The anglo-saxon sample has something like native 'briton' in it.It is definitely not pure Germanic.

  157. @ Σιμον

    Hmm interesting.I will agree with it.But still we have not enough info about what populations inhabit the regions of modern belenux,central germany etc.Well,we have hallstatt/La Tene in the southern regions,we also have the nordic BA/IA and later proto-germanic homeland but what we know about the Jastorf culture or even the Nordwestblock culture?They were a combination of Celts and Germanics or something like beakers?It is something we need to know in the future if we want to have a better view about Celto-Germanic genetics.

  158. @G2a-M406 Anatolian Farmer

    Jastorf could have been early West Germanic, while Nordwestblock could have been a late Beaker survival.

    Also Franks probably had some Celtic admixture but not too much, could have been Central Dutch-like when they arrived in France. Also there were Saxons in Northern France too.

  159. @ Gabriel

    Yes Jastorf could have been west Germanic speaking.But when it comes to autosomal DNA?What it would be the diffrences with Scandinavians/North Germanics?Less WHG/Funnelbeaker autosomal and more ANF or Steppe?Do you really believe all the Germanics had the same DNA?Even about early East Germanics i have my doubts.It is very likely east Germanics to had a Slavo drif since they were not that far away from Pomeranian culture.About Nordwestblock it seems to be a Beaker related culture but we should wait for samples thought.About Franks i think their origins are between Jastorf and Nordwestblock cultures.These 2 cultures are the key for Frankish genetics i guess.Now i have a serious question to you guys and also to David.Witch european ethnic group represents better the Bell Beakers?I mean do we all agree here that British Beakers are probably the only survival Beaker related ethnic group as a whole in Europe?Ofc south England has something Celtic/Belgian that pulls them more south witch is also visible in yDNA terms but overall is UK and British isles in general the only hideway of B.B?I was also thinking about Dutch people and how related they are to Dutch Beakers,but checking a little bit their autosomal profile south Dutch and Belgians are kinda southern shifted i guess,while North Dutch have a Scandinavian like profile.Well, what you guys think of this?

  160. Maybe West Germanics were remnants of Single Grave/Bell Beaker groups on the North European Plain who did not cross the Rhine and absorbed some later migrations from the North and West of the Elbe ?

  161. @G2a-M406 Anatolian Farmer

    I think they weren’t very different overall, maybe slightly more southern but no big difference. Like North Dutch.

    I think Franks were like Central Dutch, mostly Germanic with a minor Celtic (and Nordwestblock) substrate.

    @Ric Hern

    I don’t think so, due to drift and the presence of I1 in West Germanic groups, so I think they came from southern Scandinavia.

  162. @ G2a-M406 Anatolian Farmer

    I think you will find more closely related Bell Beaker genetics in pockets all over Europe from Southern parts of Central Europe all the way to Ireland and Spain/Portugal.

    All depends on which Bell Beaker group ended up where and admixed with whom but still retained their Bell Beaker Culture for a while afterwards.

    Irelands West I think maybe retained the most of the original Bell Beaker genes in Northwestern Europe. In Southwestern Europe maybe the Basque etc...

  163. @ Gabriel

    I see.For Some reason non-North Germanic speaking Germanics are a little bit more Southern than Scandinavians/Nordics.It is might be like you mention above a Beaker related culture who pulled them more South or even Single Grave Genetics might have a different genetic profile. And yes North Dutch and North Germans are not 100% Scandinavian/Nordic like.

    @Ric Hern

    Indeed. Basques, Some areas of Britain and Irish are good example of B.B remains. I am not sure about The province of Brettany In France. If i am not mistaken They are not that far away from Brits.But overall In EU The influences of Celts and later Germanics are much bigger i think.

  164. Preliminary G25 coords for most of the new Viking samples are available here.

    If you want to transform the scaled data into unscaled data, you can do it in this way:

    1. Prepare a text file file with the 25 eigenvalues column wise/vertical. Store it with the name 'eigenvalues_G25.txt'
    read this file with:
    eig <- as.numeric(scan('eigenvalues_G25.txt', sep=',', what='', quiet=T))

    2. Now if the scaled data are in the computer memory named as scaled
    you can do:
    unscaled <- t(apply(scaled, 1, function(x) x/sqrt(eig)))

    you can restore the scaled version with:
    scaled <- t(apply(unscaled, 1, function(x) x*sqrt(eig))).

  165. The Franks had most probably no connection with Jastorf. The Franks are Istvaeones, Rhine-Weser Germans, Jastorf was right of the Weser....

  166. @G2a-M406 Anatolian Farmer

    There is also Halberstadt_LBA. Politically Halberstadt is in eastern Germany, but looking at the geography it's roughly in the middle of the southern fringe of the north German plain. Even though his yDNA is R1a, autosomally he's closest to England_IA, Scotland_MBA and French Bell Beakers, in that order. In any case rather northwestern than Germanic. But this became part of the substrate that makes the difference between north and west Germanics (except the northernmost west Germanics like Angles and Jutes).

  167. See:
    Target: DEU_MA
    Distance: 1.2828% / 0.01282773

    55.6 SWE_IA
    25.8 DEU_Halberstadt_LBA
    18.6 CZE_Hallstatt_Bylany:DA111

  168. @ Weure

    If Franks are not related to Jastorf then would you say that they are to Nordwestblock culture?

    @ Simon

    The DEU_MA seems kinda southern compared to Proto-Germanics.Do you believe it represents some Germanic linguistic group?

  169. I think West Germanics just have higher TRB ancestry, for example Danes have that profile and I don’t think they descend from a northwards migration that was Halberstedt_LBA-admixed.

  170. @Anatolian Farmer,

    The published Saxon samples from England seem to be pure Germanic. This is because they are basically identical to modern Norwegians and to ancient Lombards.

  171. I have a couple of newbie questions. First, as I understand it seems likely that the horse was first domesticated by the Botai people, with the Botai using them for livestock and presumably for riding as well.

    What was the genetic makeup of the Botai? I have read that they were highly ANE, although I haven't found a precise breakdown of their genetics. Were they entirely ANE, or ANE/East Asian?

    Interesting that Maykop was similar to the Botai. Were they related peoples? Steppe Maykop was mostly CHG. I have read in a previous comment section that Steppe Maykop had an influx of people genetically similar to Kenniwick Man. This is interesting.Possibly a Native American like genotype was once very common in Siberia/Central Asia and parts of Eastern Europe?

    Finally, a bit off topic but Eduard Vajda's development of the NaDene-Yeniseian hypothesis seems to equate the Yeniseians with the Duyktai culture.If this theory holds water, could the Botai be descendants, like the Yeniseians or the Paleo-Eskimos, of the Duyktai? Do we know how the Dyuktai would lean towards ANE?

  172. @KateFrey

    The Botai people may have been the first people to domesticate the horse, but the horse they domesticated was from the same clade as the Przewalski's horse. So someone else domesticated the horse that became the modern domestic (and also Indo-European) horse. Preliminary evidence points to the Pontic-Caspian steppe as the place where this horse lineage was first domesticated. See here:

    The Botai people were mostly of Western Siberian Hunter-Gatherer (WSHG) stock, but with slightly elevated East Asian-related admixture. So in terms of deeper ancestry they were largely ANE, but far from pure ANE.

    Also, just like their horses, the Botai people didn't contribute much, if any, DNA to Europeans, directly or indirectly.

    Steppe Maykop people were not mostly CHG. They were very far from being mostly CHG, because they were a ~50/50 mixture between something like Steppe Eneolithic and WSHG. Caucasus Maykop was different, and much more CHG-related.

    The Kennewick link was mentioned in the Wang et al. paper with the Maykop samples, but that was only because WSHG genomes weren't available yet when the Wang et al. manuscript was written. So it's best ignored.

  173. @ Wise Dragon

    Thanks for sharing your insight. if you piece together the arguments on the anthrogenica than they are basically saying West Eurasians are a combination of Common West Eurasians, an SSA Basal Eurasian population, another SSA ANA population and an East Asian population like Salkhit or Tianyuan with the latter two groups contributing the majority of ydna (SSA ANA- ydna E and East Asian Salkhit/Tianyuan- ydna K2b/PQR + K1/LT). So I'm confused how people can such varying opinions. What makes you think Ans were the precursors of Eurasians? Their geographical location?

    @KateFrey, Davidski

    Im not even sure the Botai Horses were domesticated.

  174. @Davidski

    //There was no equine domestication in the Armenian Highlands.

    This is just your wild imagination.//

    I've not imagined the data on Q boss. It gets BIGGER in the rest of Anatolia as a proportion of the domestic horse population. It doesn't get smaller, in absolute numbers, even in Armenia, even after the Steppe DNA sweeps over the region.

    What about the data is inconsistent with Domestication in Armenian Highlands/south Caucus but the folks in the Steppe developing a true horse culture by breeding stallions from the south with their large herds of wild stock? Sort of like how gunpowder was invented in the east but the west was where they took application to new heights.


    Calm down man, it is just horse talk. What I read in the paper is that the oldest sample THEY COULD GET DNA FROM was 2500 BC, not that there were no horses in the S. Caucuses/ Armenian Highlands before then.

  175. @Mark Moore

    There's no evidence that the founder stallions of the modern domesticated lineage came from Armenia or any part of West Asia.

    In fact, the evidence from ancient Y-chromosome lineages suggests that they came from the Pontic-Caspian steppe.

  176. @Davidski,

    Thank you for that link. The Wutke et al study on stallion diversity which was the first link on your piece. As predicted, stallion diversity dropped dramatically while mare genetic diversity remained much higher, consistent with what I said (and is commonly known to be true) about using stallions to breed herds.

    Maybe a part of why we are not seeing this the same way is the precise question being asked. I am focused on the origin (not just the process of genetic domination) of the extant population of domestic horses today. There is no doubt, and the Wutke study supports this, that Steppe and Eastern Europe populations were essential to spreading the Y-HT-1 chromosome of horses today and making it what it is- the last haplogroup of surviving domestic horses.

    But does the Wutke study support the idea that this haplogroup originated in the Steppe? It does not. This quote from the study makes that clear "Haplotype Y-HT-1, on the other hand, which dominates in present-day stallions, was only detected later than 2200 BCE. Although its estimated age roughly correlates with the onset of domestication (3500 BCE; Fig. 4), this haplotype only started to become more frequent during later periods. All analyses indicate low Y-HT-1 frequency before 2200 BCE (fig. S1). However, the first time bin (>2200 BCE) only includes samples from Europe, with the easternmost samples falling on the longitude of the Black Sea (Fig. 2). Accordingly, Y-HT-1 could have been present with higher frequencies at that time in populations from further East."

    They don't have the samples to say where the line leading to today's domestic horse (male side) originated. It was such a superior line that it is the sole surviving domestic haplogroup today, yet it lingered for over 1,000 years after its origin in a small population somewhere ("All analyses indicate low Y-HT-1 frequency before 2200 BCE"). After that it explodes and soon the other domestic lines are extinct, and it was in the Balkans early too. The expansion did not seem to be centered on the Steppe.

    To me, this evidence points to the Steppe cultures finding a superior line somewhere and using it to replace their existing stock. Maybe there were two independent domestication events, but once they found this obscure line that was a better domestic horse they quickly used it to replace the strain they had. I recall that this pattern is true for other domestic animals like the dog and perhaps the pig. There were separate domestication events and our modern types are mostly one kind over the other. In the case of the horse, the dominance of one domesticated strain became complete.

    The Armenian Highlands are right across the Caucus Mountains. That is what divides them from the Steppe. This is the general area in which first domesticated sheep, cattle, pigs, and goats are found. Is it likely that the people who originated our four biggest livestock mammals today never possessed a domesticated horse until the Steppe populations came through the region thousands of years after they domesticated these other mammals? The maternal line data, to the extent it says anything with such a paucity of samples, says this region had its own successful line of domestic horse.

    I think it is more likely that a horse was just less useful for a sedentary culture in the Armenian Highlands than it was for a group of mobile pastoralists in the wide-open Steppe. Maybe they only used the horse as a pack animal in the mountains and it was the Steppe people who first mastered life on horseback. That the Sumerians called the horse "the donkey of the mountains" supports this view. So the situation would be like what happened with gunpowder, even if invented in the east, its use was taken to new levels in the west.

    The bottom line is that the question of how Y-HT-1 came to be the sole line of today's domestic horse has been answered, and the Steppe cultures are a big part of that answer. What is still unclear is who developed that line.

  177. @ Mark Moore (Moderator)

    If horses were first domesticated around 3500 BC. does it make sense that the horse was domesticated in Armenia and kept secret for 1500 years from their nextdoor neighbours while the Salzmünde Tabiano Coloured Horse pops up at +-3100 BC., less than 400 years, thousands of Kilometres to Armenias West ?

  178. @ Mark Moore (Moderator)

    If the horse was first domesticated in Armenia isn't it strange that they did not spread in all directions while domesticated Cattle, Goats, Sheep and Pigs did spread in all directions from their initial area of domestication ?

  179. Of course horses weren't domesticated in Armenia.

    "Mark Moore" is talking shit.

  180. @RicHern

    //If horses were first domesticated around 3500 BC. does it make sense that the horse was domesticated in Armenia and kept secret for 1500 years from their nextdoor neighbours while the Salzmünde Tabiano Coloured Horse pops up at +-3100 BC//

    You can't just look at the distance. Look at the geography. A clan of folks tucked into some valley in eastern Anatolia could have very little contact with folks from the other side of the Caucus Mountains even if the Steppe Cultures had 1,000 KM trade networks over the open ground.

    But I will turn your question back 'round on you: If the Steppe cultures had the strain of Stallion that was so good that it has now replaced all other domestic lines in their midst from 3500 BC, is it likely that it would stay obscure for over 1,000 years?

    //If the horse was first domesticated in Armenia isn't it strange that they did not spread in all directions while domesticated Cattle, Goats, Sheep and Pigs did spread in all directions from their initial area of domestication ?//

    If you will look up a landmark paper by the Smithsonian's Melinda Zeder ( ) you will see the pattern for the spread of all four of our major livestock animals, Sheep, Pigs, Cattle, and Goats. They all originate at about the same time and about the same place (well before the domestic horse line Y-HT-1). There is sort of a crescent centered on SE Anatolia. From there they spread mostly east-west at first, including an odd early appearance on Cyprus (I think somebody got the idea to take their herds to a place where there were no natural predators). From there over the next 2500 years they work their way south to Sumer and the Levant, but not north. Perhaps something very similar to what happened with every major herd animal we use for meat happened with the horse, only in the reverse direction.

    So in the first place the premise of your question is incorrect. PS-the ancestors of the Chinese a bit after this separately domesticated the pig, and our pigs today are a mixture of theirs and the Anatolians (though the Anatolian strain was mixed with wild strains later on).

    Secondly, there is good a reason why the horse would not spread in the same manner as Sheep, Goats, Pigs, and Cattle. I've already touched on it. Before someone came up with the idea of riding the horses, they would basically serve as pack animals - a role the donkey filled in Mesopotamia and which would explain why the Sumerian word for horse was "donkey of the mountains".

    Animals that are superior eating are much more likely to be adopted than those who serve as pack animals when another pack animal (the donkey) is already available. The cattle could pull carts just as well so long as you were not in a race.

    Combining a nomadic pastoralism with the wide-open geography of the Steppe, the idea of adding horse-ridding to that is a great leap forward. That's the context in which the horse got valuable. (And after that as a weapon of war). I think they were doing that, and they found a better line of horse. I mean, doesn't what happened with Q mt-DNA in the one study point to the idea that there was a successful domestic horse population down there, even if it was overwhelmed by Steppe stock later? Isn't this the same region where all of these other animals were domesticated? Isn't it a known fact that some animals were domesticated more than once? If the Steppe cultures developed the Y-HT-1 line around 3500 B.C. isn't it odd that it took over 1,000 years for it to explode?

    I just don't see why there is such resistance to this idea. It fits so naturally with the answers to all of those questions.

  181. @ Mark Moore (Moderator)

    Well maybe it was not just the Y that created the perfect Stallion. Maybe it needed specific Autosomal-DNA to make that line pop out...or maybe the specific traits that that innitial stallion possessed was something very small that only a few people noticed and exploited...

    You surely can not propose geography as an excuse. There are just as much Mountains and Valleys to the North of Armenia as to its South...So I fail to see your point.

  182. @ Mark Moore (Moderator)

    Most commercial pig breeds today are almost totally descendants from European Wild Boar...


Read the rules before posting.

Comments by people with the nick "Unknown" are no longer allowed.

See also...

New rules for comments

Banned commentators list