Hi there friends,
Recently, it dawned on me that I haven't published a single blog post in six months, with my last post dating back to 2023. I apologize for this extended period of inactivity. As I was contemplating my next topic, an intriguing preprint of an upcoming article related to the Scythians caught my attention on BioRxiv. This was uploaded just two weeks ago, so you may have already come across it. If not, you can check it out here:
North Pontic crossroads: Mobility in Ukraine from the Bronze Age to the early modern period
Lehti Saag, Olga Utevska, Stanislav Zadnikov, Iryna Shramko, Kyrylo Gorbenko, Mykola Bandrivskyi, Dmytro Pavliv, Igor Bruyako, Denys Grechko, Vitalii Okatenko, Gennadi Toshev, Svitlana Andrukh, Vira Radziyevska, Yurii Buynov, Viktoriia Kotenko, Oleksandr Smyrnov, Oleg Petrauskas, Borys Magomedov, Serhii Didenko, Anatolii Heiko, Roman Reida, Serhii Sapiehin, Viktor Aksonov, Oleksii Laptiev, Svyatoslav Terskyi, Viacheslav Skorokhod, Vitalii Zhyhola, Yurii Sytyi, Mari Jarve, Christiana Lyn Scheib, Kyriaki Anastasiadou, Monica Kelly, Mia Williams, Marina Silva, Christopher Barrington, Alexandre Gilardet, Ruairidh Macleod, Pontus Skoglund, Mark G Thomas
doi: https://doi.org/10.1101/2024.05.24.595769
Before I get into it I want to congratulate the authors! This is a good article, and when published it will give us some great additional insight into the genomic structure in Ukraine across various periods. Although I do think there are some changes the authors should make to the article since it is still in pre-publication. Personally I am unsurprisingly the most interested in their antiquity/iron age data as it relates to the Scythians and will discuss this further.
In addition, in the last few months there have been some more insights into the genetics of Scythian peoples through various pre-prints, abstracts and data uploads. 2023 was a bit of a dry spell for the Scythophiles, but 2024 and 2025 seems to be a feast for us. I will be discussing some of this as well in this blog entry.
North Pontic crossroads
In short, with this article we get new data from the late neolithic, bronze age, iron age, antiquity, the middle ages and the early modern period. Awesome. I think there is something in there for all different sorts of history nerds:
Samples from the iron age Lusatian culture, hard to come by given the prevalence of cremations within this population
North pontic Greek samples
Early medieval Slavs
Medieval Alans and Bulgars on the Pontic Steppe
Cumans
Nogai Horde
There is a lot in there, but I will focus on the Scythians in particular despite many of the other samples also being of high interest for me. I will not be reviewing the whole article as I have done with others in the past, as the article is pretty straight forward so I will just focus on some section of interest.
Scythia, Scythian and The Scythians
One important part when reading Histories by Herodotus is that the name Scythia refers to a region encompassing various ethnic groups. Due to similarities in looks and culture, all the non-Greek populations in this region are referred to as Scythians, but Herodotus is clear to distinguish the various peoples from one another, either through mentioning their features, subsistence economies, customs or language. In particular, the nomadic Scythian tribes were distinguished the various neighbouring peoples such as the Nueri, Geloni, Budini, Tauri, Maeotians, Black Cloaks, but also from other nomadic Iranic speaking populations such as the Sauromatae to their east.
As time progressed, the term Scythian became widely employed for the various nomadic steppe nomads in antiquity. It also developed into a catch-all term for all sorts of barbarians, with various Germanic people being referred to as Scythians by the time of the Romans or medieval Turkic peoples in the Byzantine sources.
In contemporary times, the archaeological remnants of peoples in the past are often classified as Scythian on the basis of Scythian items, primarily ornaments and weaponry, rather than their livelihoods, primary material culture and presumed ethnolinguistic backgrounds. This has led to a bit of a loose concept of “Scythian” in an archaeological sense, quite reminiscent of how Scythian became a loose term in antiquity. This usage of the term Scythian begs the question, what are Scythians? Are Scythians simply people who used akinakes and barbed arrows, or were they a population with their own identity based on tribal origins, language and livelihood, was it simply a shared culture across various unrelated populations?
In my opinion, in a perfect world there would be a two-fold use of the term scythian: one as a broader term for the various Iranic speaking nomadic populations identified by historical, archaeological and genetic sources. The second, in the context of eastern Europe, specifically referring to the Iranic speaking nomads known as the Scythians.
After all, the term Scythian comes from the Greek exonym Skýthai, itself derived from an early variation of the Scythian ethnonym Skuδatā. In the Pontic Scythian language there was a soundshift and the name became Skulata, Herodotus also attests this contemporary form as Skolotoi. This name is more or less the Scythian etymological equivalent of “Shooter”(archer), with both words deriving from the same PIE root: *(s)kewd-.
To get back to genetics, I think the authors did a good job of distinguishing the various groups from one another, and nicely highlight how different groups of “Sythians” were in fact different populations with differing livelihoods and archaeological material culture. However, I do feel a stronger emphasis on “Scythia” being a realm of ethnically diverse populations with attested names, rather than all being just various distinct groups of “Scythians”, could be emphasised a bit more. I think this is relevant as the majority of the Scythian samples come from the nomadic periphery rather than from a core nomadic zone.
The various groups include:
UkrEIA_Scythian_RightDnipro_IllThr
UkrEIA_Scythian_LeftDnipro_IllThr
UkrEIA_Scythian_LeftDnipro_LocAgr
UkrEIA_Scythian_LeftDnipro_LocEl
UkrEIA_Scythian_LeftDnipro_LocAgr
UkrEIA_Scythian_SivDon_Nom
UkrEIA_Scythian_SivDon_NomEl
UkrEIA_Scythian_SivDon_NomEl_3
UkrEIA_Scythian_NBlaSea_Nom
UkrEIA_LateScythian_Cri_Nom
From the article:
Scythian period individuals (nine of whom are radiocarbon dated to 798–199 cal BCE) were divided into groups based on their geographic location: right (i.e. west) or left (i.e. east) bank of Dnipro and Siversky Donets basin in the forest-steppe, Northern Black Sea region in the steppe.
The groups were further divided based on the demographic association inferred from archaeological context: IllirianThracian, local or nomad, agriculturalist or elite. The groups are named using the following structure, separated by underscores: time period, cultural association, geographic location, demographic association, genetic subgroup (where relevant) (Figure 2, Table 1, Table S1, Methods).
Most of the Scythian individuals from the right bank of Dnipro with Illirian-Thracian associations (UkrEIA_Scythian_RightDnipro_IllThr), one individual from local agricultural tribes of the left bank of Dnipro (UkrEIA_Scythian_LeftDnipro_LocAgr_2), some of the non-elite nomad individuals from Siversky Donets basin (UkrEIA_Scythian_SivDon_Nom_2) and one elite nomad individual from the same area (UkrEIA_Scythian_SivDon_NomEl_3) are similar to previous Early Vysotska and Lusatian individuals and also modern Ukrainians in both PCA and ADMIXTURE (Figures 3B, 4, Figures S1, S2B, S3, S4). The rest of the Scythian individuals from the right bank of Dnipro with Illirian-Thracian associations (UkrEIA_Scythian_RightDnipro_IllThr_2) and from local agricultural groups of the left bank of Dnipro (UkrEIA_Scythian_LeftDnipro_LocAgr) are more similar to Western Steppe individuals (including previously published Scythian-related individuals from the region) (Figures 3B, 4, Figures S1, S2B, S3, S4).
The same is true for most of the non-elite nomad individuals from Siversky Donets basin (UkrEIA_Scythian_SivDon_Nom) and one elite nomad individual from the same region (UkrEIA_Scythian_SivDon_NomEl_2), the individual from the left bank of Dnipro with IllirianThracian associations (UkrEIA_Scythian_LeftDnipro_IllThr), as well as the four steppe nomads from the Northern Black Sea region (UkrEIA_Scythian_NBlaSea_Nom) (Figures 3B, 4, Figures S1, S2B, S3, S4).
Local elite individuals from the left bank of Dnipro (UkrEIA_Scythian_LeftDnipro_LocEl) have greater genetic affinity to Southern European individuals (somewhat similarly to Scythians from Moldova) (Figures 3B, 4, Figures S1, S2B, S3, S4). Most of the elite nomads from the Siversky Donets basin (UkrEIA_Scythian_SivDon_NomEl) (n=3) share highest similarity with individuals from the Caucasus (Figures 3B, 4, Figures S1, S3, S4).
Scythian samples from Mamai Hora
The two samples are UKR013 and UKR014, both dating to 400 - 300 BC. UKR013 was a female individual with mtDNA haplogroup HV2a3. UKR014 was a male with Y-chromosome haplogroup R1a-Z645 (probably R-Z93 but lacking further coverage) and mtDNA haplogroup I4a. In the models supplied by the authors the samples have an additional layer of EEF ancestry, but it is difficult to clearly determine the source from the graphs.
Gilded Akinakes found at Mamai Gora
Also interesting to note is that we already have a Scythian DNA sample from Mamai Hora, MJ-46 from Jarve et al. 2019. The genetic profile of this individual can best be described as a Pontic Scythian with a significant amount of southern/southeastern european ancestry. It might be the case that the two samples UKR13 and UKR14 will turn out similar to MJ-46 on a genomic level.
Don Scythian samples
The samples from the Don region are the most informative about the Scythians in my opinion, which is perhaps a bit of a paradox as they are geographically on the Scythian periphery rather than the core steppe zone. The sample size from the region makes up for that however.
The presence of two samples with Q-L330 attest to the eastern origins of the iron age Scythian populations. Also interesting to note that two samples from this region belonged to R-YP5018, and it is going to be interesting to see how the Y-DNA of these samples relates to modern carriers of this Y-chromosome haplogroup.
There also is quite some representation of R1b-M269 > L23 > Z2103 in the Don Scythian samples which is quite interesting as R1b-Z2103 is a minority haplogroup in Scythians compared to R1a-Z93 or the various Q clades. The haplogroup is present in both elite and commoner graves of the nomadic population in the Don region.
The UkrEIA_Scythian_SivDon_NomEl_2 cluster is modelled as being just Yamnaya + East Asian, which might indicate that those samples do not have significant ancestry from regions north or west of the Pontic steppe. If so this could mean that MJ-32/MJ-15, also from the Don region, is not the only “unmixed’ Scythian DNA sample anymore.
Also interesting is that UKR095, one of the Don local agriculturalist samples, carried haplogroup R1a-Z93. It is likely that this individual descends from an iron age nomad or Scythian which got assimilated into the agricultural communities adjacent to the steppe zone. I wonder which dynamics led to said assimilation, because the archaeological site this sample comes from has shown to have suffered violence at the hand of steppe nomads in several different periods:
Kolomak (Коломак) (Archaeology – V. Radzievska; text – S. Zadnikov, I. Shramko)
The settlement is located on the cape of an unnamed tributary of the Kolomak River (a tributary of the Vorskla River). The settlement consisted of the main yard (5.4 hectares) and suburb area (8 hectares). Excavations revealed that it was inhabited by the local forest-steppe population of the Scythian period (the second half of the 6th to 4th c. BCE). The territory of the settlement was densely populated. During excavations, 14 dwellings, numerous pits and a well were found. A big number of objects of material culture were found, namely knives, sickles, hoes, arrowheads, clay figurines, fragments of local moulded ware, fragments of Greek amphorae. Features of attacks were traced. The first attack dates back to the 6th–5th c. BCE, when the settlement suffered from the raids of the steppe Scythians. After that it was restored and strengthened with an additional rampart and ditch on the eastern side. In the second half of the 5th c. BCE the territory of the settlement decreased. At the border of 4th to 3rd c. BCE the settlement was again defeated by nomads and ceased to exist.
Crimean Scythians
Beyond the samples from the Don region, the article also included three Scythian samples from the Crimean peninsula. The Scythian period in Crimea is interesting as it did not have a large Scythian presence despite being part of the Scythian domain, with nomads in the northern flatlands and the Taurians in the hill country of the south.
However, at the end of the Scythian dominion on the Pontic-Caspian steppe, they suffered territorial losses to the Celts and Dacians in the west, and to the Sarmatians in the Pontic steppes, the latter crossing the Don and conquering Scythian steppe regions in the 3rd and 2nd centuries BC.
The “seat of power” of the Scythians then switched to the Crimean Peninsula, with Neapolis becoming their new capital city. Although the archaeology at Ak-Kaya is interesting as it showcases a Scythian seat of power in the Crimea prior to the expulsion of the Scythians from the European steppes, and this is likely the location the Scythian dynasty ruled from prior to Scythian Neapolis. At Ak-Kaya site you can also somewhat track the development of the late Scythian culture, the site attesting to the cultural transition spurred on by subsistence economy changes and significant cultural influences from Taurians and Greeks [1].
With this transition, many of the Scythians in the Crimea settled down and intermixed with the Taurians and to a lesser extent the Greeks, and this synthesis led to the development of the Tauri-Scythians. The kings of the Crimean Scythian dynasty were of the same dynasty of the Pontic Scythian kingdom, extending the continuation of the Scythian kingdom for some centuries. However, the rule of the Scythians would come to an end with the Pontus kingdom.
The supposed tomb of King Skilurus at Scythian Neapolis
Relief depicting King Skilurus and his son
The lads from AncestralWhispers did a facial reconstruction of King Skilurus a while back which I enjoyed quite much:
In the period after the Scythian rule the Tauri-Scythian sedentary population persisted in the region for a few centuries, although they were being assimilated into their neighbouring Sarmatian peoples. The Crimean Scythians were last attested in the 3rd century AD with the Gothic invasions in the region.
The DNA samples we have are seemingly from the early phase of the Scythian presence in Crimea, in the period where they became sedentary and received significant Greek cultural input. Interestingly, the site they come from is close to the ancient Greek settlement site Maslyny.
There seems to be a bit of a contradiction in relation to the dates of the settlement and burials. The archaeological descriptions mention a phase of 4th-3rd century BC. The dates for the samples, although not RC dated, were 150 - 1 BC in the article.
Adding to this, Masynly sees a destruction layer around the beginning of the 2nd century BC [2]. which could be seen as the date Scythians settled in the area and erected burials there, although the date around 200-150 BC could also be an indication of conflict within Crimea as the Scythians established power in the region or an attack from incoming Sarmatian populations as 150 BC is around the time they first start showing in the region. Hopefully there is some further clarification on this matter in the final article.
Getting back to genetics, the three samples in question are:
UKR051 - XX - U7a3a*
UKR052 - XY - R1a1a - R1b-M269
UKR053 - XX - HV1a1
Unfortunately with the model provided it is hard to say what the exact genetic composition of these samples are. They score a high amount of Yamnaya but there is an excess of ANF/EEF ancestry. This could be either from southeastern European populations from the Eastern Hallstatt sphere or from Greek colonists. Perhaps an input of Caucasus-related ancestry is also present but I can’t currently retrieve this from the DNA data.
I guess we'll just have to wait until the genomes get published to figure this one out.
UKR066 - the “Cimmerian”
One sample of particular interest to me is UKR066, an early iron age nomad found in Kumy, Kharkiv region. Radiocarbon dating places this individual between 1195 and 919 cal BC.
UKR066. Kumy, kurgan 6, burial 5. The inlet burial of a Cimmerian nomad. The burial was found at a depth of 0.7–0.9 m above the level of the reference point. The contours of the pit were not traced. The skeleton of an adult man (35–36 years old) was lying on his right side, with his head to the west. The bones of the right hand were slightly bent at the elbow and were stretched along the body. The left hand was missing. There were no artefacts. Chronology according to archaeology: 10th–9th с. ВСE. Chronology according to 14C dating: 1195–919 cal BCE (2865±39BP).
A DNA sample of a historical Cimmerian, fascinating right? Well about that…
The phase of the nomadic iron age prior to the rise of the Scythians is referred to as the Pre-Scythian period, but it is perhaps more commonly known as the Cimmerian period. I personally find the Cimmerian period an unfortunate name as it associates the whole pontic nomadic horizon with the historical Cimmerians, who had a more limited range in between the Azov sea and the Volga river going by ancient toponyms, hydronyms and historical records.
The groups west of the Cimmerians are not attested historically in this period but there are some reasons to presume a presence of the Agathyrsi on the Pontic Steppe, prior to their migration to the Carpathian basin and ultimate assimilation into the Daco-Thracian sphere.
To be specific, the Pre-Scythian/Cimmerian horizon is divided into two material cultures:
- Chernogorovka culture ( 900 - 700 BC)
- Novocherkaskk culture (700 - 600 BC)
The Chernogorovka culture is the first phase of the Pre-Scythian iron age. The formation of the culture can best be described. While some archaeologists favoured a local origin of the Chernogorovka culture, it is clear that with the sudden appearance of derived Karasuk-style weaponry and new burial rites with a distinct supine position, a new population had arrived to the Pontic-Caspian steppe. The same population appeared all over the western Eurasian steppe belt, originating in the Circum-Altai region.
The skeleton of sample I20086 from Lazaridis et al. 2022, a pre-Scythian period male from Glinoe, Moldova (kurgan 110, burial 1) is a good example of this new burial position:
This map of Central-Eastern Europe in the 9th century BC shows the rough distribution of the Chernogorovka culture [3]:
1: Middle Danubian Urnield culture. – 2: Kyjatice Urnield group. – 3: Mezőcsat Group. – 4: Gáva Culture. – 5: different groups using stamped and incised pottery (Gornea -Kalakača; Ostrov-Insula Banului; Babadag; Pšeničevo, Cozia -Sacharna, Černoles). – 6: Černogorovka cultural group. – 7: Koban culture (after Mezner -Nebelsick 2010, 139 ig. 5a).
The Chernogorovka develops in a western and eastern variant, the rough distribution you can see in the following map [4]:
The origins of the Novocherkassk culture seems to be the eastern part of the Chernogorovka culture, and the distribution of the Novocherkassk culture is more in line with the location of the Cimmerians prior to their journey into West Asia as it is primarily present in the Don steppe region However, the Novocherkassk culture also spreads further west into the Pontic Steppe prior to the the migration of the Cimmerians into the Near East.
To make matters more complex, the archaeological imprint of the Cimmerian presence in West Asia cannot be tied to the Chernogorovka or Novocherkakk culture directly, but seems to be an early variant of the Scythian material culture [5]. The Scythians and Cimmerians in West Asia more or less had the same material culture, which then subsequently became the archaeological basis for the Scythian material culture in Europe. It is quite ironic that material cultures of historically unknown nomads are considered Cimmerian but the material culture of the historical Cimmerians is part of the Scythian culture.
UKR066 would have belonged to the early phase of the Chenogorovka culture, and going by the map shared above he would have been within the western horizon, although given the proximity to the eastern horizon an origin in this region could also be possible. In any case, referring to this sample as “a Cimmerian” as the authors have done is unfortunately not accurate. The sample falls outside the geographic region of the historical Cimmerians, belongs to a material culture not directly associated with the historical Cimmerians, and lived more than two centuries prior to the first attestations of the Cimmerians, which raises the possibility that UKR066 predates the ethnogenesis of the Cimmerians as a distinct population.
This sample also highlights a phenomenon I have been pointing out for years. The bronze-to-iron age transition and the onset of the Pre-Scythian period across the Eurasian steppes in the traditional Soviet chronology is underestimated by roughly one century. Samples with a 10th-9th century B.C context are often a century older going by C14 dates.
This also has consequences for the date of the formation of The Scythian material culture, which is typically placed somewhere around 650 - 600 BC. The Cimmerians were already present in West Asia by 714 BC [6] and migrated to Anatolia prior to 679 BC, and since both the material cultures of the Scyhians and Cimmeria in West Asia were of the Scythian archaeological culture it is highly unlikely that this material culture developed after 679 BC, and it likely developed decades prior to said date.
Depiction of Cimmerians on an Assyrian relief
As mentioned before, the Chernogorovka culture arose through the rapid migration of early nomads from the greater Altai-Sayan area. As they settled onto the European steppes there was an assimilation of the preceding populations. This can also be noted in the partial continuation of flexed burials in the iron age. This mixture of eastern newcomers, local populations and additional gene flow from neighbouring peoples ultimately shaped the genetic profiles of the historical iron age nomadic populations such as the Cimmerians and Scythians.
DNA from Bilsk Fortress - Capital of the Geloni?
The UkrEIA_Scythian_LeftDnipro_LocAgr, UkrEIA_Scythian_LeftDnipro_IllThr and UkrEIA_Scythian_LeftDnipro_LocEl samples come from the Bilsk fortress in Poltava. The Bilsk fortified settlement dates from the 8th century BC until the 4th century BC and is the largest fortified settlement of Europe in the early iron age.
The samples from various burials near Bilsk. Samples UKR087, UKR088, UKR089 and UKR090 come from the elite Skorobir mounds, and sample UKR091 comes from the elite Pereshchepyno burial.
The samples from the Bilsk fortress in Poltava seemed to have quite a bit of E1b1b, 3 out of 4 of the Y-DNA haplogroups were reported to be E-V13 with one reported R1a-M198. Keep in mind that the E1b1b were all relatives of one another, with two of them being from the Skorobir mounds and one from Pereshchepyno. It might also be that the E-V13 classification was incorrect and that it is actually a more basal clade of E-L618, going by the comments of Pribislav on GenArchivist.
The ancestry of the populations seems intermediate between northeastern and southeastern europeans with some additional admixture from Scythians. What I find interesting is that the archaeology of Bilsk suggested a significant influence from the eastern Hallstatt sphere [7], which seems corroborated by genetics now.
Hallstatt type sword found near the Bilsk fortress.
Archaeologist Denis Grechko recently published another article on the materials at these mounds and how they stand within the context of the “Scythian military invasion of Central Europe”, an event which unfortunately has not left any historical records but can be noted by the simultaneous destructions of forts and appearance of scythian arrowheads in different parts of Central-Eastern Europe in the 6th century BC.
Furthermore, the Bilsk fortress is one of the primary archaeological sites identified with Gelonus, the wooden city which was the capital of the Geloni, one of the tribes said to inhabit Scythia but considered distinct from the Scythians.
"The Budini are a great and populous nation; the eyes of them all are very bright, and they are ruddy. They have a city built of wood, called Gelonus. The wall of it is three and three quarters miles in length on each side of the city; this wall is high and all of wood; and their houses are wooden, and their temples; [2] for there are temples of Greek gods among them, furnished in Greek style with images and altars and shrines of wood; and they honor Dionysus every two years with festivals and revelry. For the Geloni are by their origin Greeks, who left their trading ports to settle among the Budini; and they speak a language half Greek and half Scythian. But the Budini do not speak the same language as the Geloni, nor is their manner of life the same.
…
The Budini are indigenous; they are nomads, and the only people in these parts that eat fir-cones; the Geloni are farmers, eating grain and cultivating gardens; they are altogether unlike the Budini in form and in coloring. Yet the Greeks call the Budini too Geloni; but this is wrong. [2] Their whole country is thickly wooded with every kind of tree; in the depth of the forest there is a great, wide lake and a marsh surrounded by reeds; otter is trapped in it, and beaver, besides certain square-faced creatures whose skins are used to trim mantles, and their testicles are used by the people to heal sicknesses of the womb."
The size of the Bilsk fortified settlement, coupled with its geographical location in the Poltava region, positions the settlement as a strong candidate to be identified as Gelonus. There are no other fortified settlements remotely comparable to this one in the region, which means the identification is not the strongest as it relies on absence of evidence of comparable fortresses. This theory also has plenty of opponents so the identification of Bilsk as Gelonus is far from a done deal.
If this identification of Gelonus is correct, then the mention of Dionysus is interesting here. Dionysus cult has a connection to Thracian religion and culture. It had once been assumed Dionysuss was of Thracian origin and then got incorporated into Greek religion, but the attestation of Dionysus in Mycenaean times makes this a bit untenable. Nevertheless, there is an identification of Dionysus with Sabazios as well as a spread of Dionysus cult-like practices in the Daco-Thracian cultural sphere. The mention of Dionysus in particular could indicate a connection between the Geloni and the southeastern European regions west of the Black Sea. Alternatively, it spread directly to the Geloni via the Greeks on the black sea coasts.
I cannot make any claims for the ethnolinguistic affinity of the Geloni, but I do suspect that perhaps Herodotus’ claim of their language being somewhat in between Greek and Scythian should not be taken too literally. Rather than a creole, Herodotus may have just described an unfamiliar (Indo-European) language with some recognizable vocabulary, but a language which was not Thracian, Iranian or Greek. Hopefully with the release of the samples we can uncover more about the genomics of those samples and how it could potentially tie to known ethnolinguistic groups of the iron age and subsequent periods.
Some additional thoughts on the Eastern Hallstatt
The Eastern Hallstatt complex is an overarching material culture for various groups in eastern Europe around the Carpathian Basin, Danube river and western black Sea regions. Argued to have arisen from the preceding Urnfield groups of the region, there was an significant expansion of the eastern Hallstatt material culture in the early iron age. This expansion had been connected to the predecessors of Getae and Thracians from the iron age/antiquity by quite some archaeologists in the past. These new findings seem to back up the connection between Eastern Hallstatt and the Thracian peoples to a degree, as Y-DNA haplogroup E-V13 appears in both populations.
What stands out to me is not just the E-L618/V13 associated with the Eastern Hallstatt sphere, but also the rare ANF-derived lineage C1a-Z38888/C-Y83490 carried by UKR002 from Kartal. Entering the realm of speculation, this relative diversity of rare ANF clades in addition to the EEF-rich profiles of these populations might to suggest an ethnogenesis in a region where primarily EEF-derived populations would have been somewhat shielded from the population expansions of the third millennium BC and were able to “hide out” so to say. Perhaps the Transylvanian Plateau was of importance here? An abstract from an upcoming article seems to hint at such a process taking place in that region:
Looking at a map of the eastern Hallstatt expansion below [8], I could certainly see a case where a significant portion of the genetics is derived from such a putative population.:
However, a genetic connection between the two must be proven and we currently have no insight into the genetic data of these Transylvanian populations, so it is impossible to say if this is the region to look for. Wherever it may have come from, there is a population with EEF-rich ancestry and EEF derived paternal lineages which played a substantial role in the formation of the Eastern Hallstatt complex of iron age southeastern Europe, distinguishing it from the western Hallstatt complex which wa genetically on the western/central European cline formed by the convergence of Corded Ware/Bell Beaker and late neolithic European farmer ancestries.
Furthermore, the lack of genomes from other areas of this eastern Hallstatt complex or the Urnfield predecessors make it difficult to compare the degree of genetic relation between various groups considered to be part of the Eastern Hallstatt complex. We can however see a variable imprint of genetic impact related to the eastern Hallstatt influence in Ukraine, with the samples from Kartal, Medvyn and Bilsk.
The samples from Kartal from a proper Eastern Hallstatt context seem primarily southeastern European whereas the samples from Medvyn site, which has connections to the Zhabotin-Chernoles group (which received E. Hallstatt/Basarabi influence hence the “Illirian-Thracian” designation in this article), shows a genomic profile similar to LBA/EIA samples from Northwestern Ukraine, which could indicate a Balto-Slavic nature of this population. The samples from Bilsk seem intermediate between these groups on a genetic level.
Admixture models
One aspect of the preprint which I found disappointing was the choices of populations utilised for the admixture modelling. In order to properly model iron age Scytho-Siberian populations one must account for all these broad ancestry components; Steppe_MLBA, West-Siberian HG, Iran_N/Iran_chl and East Asian ancestries. For a model utilising fewer and more proximate sources, something akin to Steppe_MLBA+BMAC+Khovsgol_LBA should cover the “steppe” side of the equation. Gnecchi-Ruscone et al. 2022 had some nice models and I would suggest the authors take some inspiration from it:
Admixture modeling of IA steppe populations
Genetic ancestry modeling of the IA groups performed with qpWave and qpAdm confirmed that the steppe_MLBA groups adequately approximate the western Eurasian ancestry source in IA Scythians while the preceding steppe_EBA (e.g., Yamnaya and Afanasievo) do not (data file S4). As an eastern Eurasian proxy, we chose LBA herders from Khovsgol in northern Mongolia based on their geographic and temporal proximity. Other eastern proxies fail the model because of a lack or an excess of affinity toward the Ancient North Eurasian (ANE) lineage (25). However, this two-way admixture model of Khovsgol + steppe_MLBA does not fully explain the genetic compositions of the Scythian gene pools (data file S4). We find that the missing piece matches well with a small contribution from a source related to ancient populations living in the southern regions of the Caucasus/Iran or Turan [we use the term “Turan” for consistency with (7), only its geographical meaning, designating the southern part of Central Asia; Fig. 3A]. The proportions of this ancestry increase through time and space: a negligible amount in the most northeastern Aldy_Bel_700BCE group, ~6% in the early Tasmola_650BCE, ~12% in Pazyryk_Berel_50BCE, ~10% in Sargat_300BCE, ~13% in Saka_TianShan_600BCE, and ~20% in Saka_TianShan_400BCE (Fig. 3A), in line with f4-statistics (table S2). Sarmatians also require 15 to 20% Iranian ancestry while carrying substantially less Khovsgol and more steppe_MLBA-related ancestry than the eastern Scythian groups.
Granted many of the European Scythian samples would have limited ancestry from those sources, but looking at several Iron age nomadic samples being modelled as purely Steppe_EMBA/Yamnaya + East Asian, it seems the lack of accounting for Central Asian + Siberian geneflows weaken the models of the authors.
But what I find more important is that articles act as a diffusion point for new information. What will undoubtedly happen if the article is published is that some readers will take that model too literally. Do not be surprised if the wikipedia article about the Scythians will state that they were just western steppe herder + east asian in origin sourcing these models, when their real genetic composition was far more complex.
There already is this belief out there that Pontic Scythians and Eastern Scythians have distinct origins with Pontic Scythians being derived from Z2103 Yamnaya groups, which is based on incorrect Y-DNA haplogroup calls from Krzewinska et al. 2018. Take for example Mary et al. 2019:
“Haplogroup R1a-M173 was previously reported for 6 Scytho-Siberian individuals from the Tagar culture (Keyser et al. 2009) and one Altaian Scytho-Siberian from the Sebÿstei site (Ricaut et al. 2004a), whereas haplogroup R1a1a1b2-Z93 (or R1a1a1b-S224) was described for one Scythian from Samara (Mathieson et al. 2015) and two Scytho-Siberians from Berel and the Tuva Republic (Unterländer et al. 2017). On the contrary, North Pontic Scythians were found to belong to the R1b1a1a2 haplogroup (Krzewińska et al. 2018), showing a distinction between the two groups of Scythians.
…
Thus, these results are similar to what we have described for maternal lineages with Scytho-Siberians carrying more paternal lineages associated with Eastern Eurasia than Western Eurasia. The absence of R1b lineages in the Scytho-Siberian individuals tested so far and their presence in the North Pontic Scythians suggest that these 2 groups had a completely different paternal lineage makeup with nearly no gene flow from male carriers between them.”
It would thus be unfortunate if the distal models employed in this article would further add to this incorrect notion.
If possible I really hope the authors make some amendments and include realistic proximate models. Another interesting attempt would be to model Scythians without a direct east asian source and utilising samples from the pre-Scythian period as the source of the eastern ancestry.
At the very least, an inclusion of Iran_N/CHG populations as one of the sources would help round out the models quite a bit, and this applies to most of their samples in the iron age and beyond, not just Scythians.
Taurians - descendants of Cimmerians?
If I’m being pedantic already, I might as well point this bit from the supplementaries I came across:
The Late Scythian culture encompassed remnants of various ethnic groups that inhabited Taurica at that time: Scythians, Tauri (descendants of Cimmerians), Thracians, and Sarmatians. The formation of the culture is associated with the gradual sedentarisation of the barbarian population, accompanied by changes in economic activities.
The bolded part about Taurians being descendants of Cimmerians is quite an unsubstantiated claim to put in the supplementary I think. Since we have no DNA from the historical Taurians in this article or preceding ones, I assume this was based primarily on archaeology and/or history.
We have no historical mention of Cimmerians from West Asia returning to Europe and settling in Crimea, nor do we have evidence of a Cimmerian population continuing to exist in the region.
The archaeology of the pre-scythian Crimea has a peculiar material culture called the Kizil-Koba culture, which is highly distinct from the Cimmero-Scythian material culture and is distributed in the Crimean mountainous region, precisely where the historical Taurians were mentioned.
Connections to the Caucasus and Eastern Hallstatt have both been mentioned in literature. I think a Caucasian connection could make sense considering the ancient tradition of dolmen burials in the Caucasus, which might have a relation to the burial structures of the Kizil-Koba population:
What I think happened here is that there is an antiquated theory of linking the Tauri to the local Srubnaya, which have been connected to the historical Cimmerians by various authors, mistakenly. And thus the idea of Taurians being descendants of Cimmerians sneaked into this article.
In any case, even leaving aside the archaeological/historical issues with declaring the Tauri as descendants of the Cimmerians, there is no genetic evidence for this in this article since it does not contain DNA samples from the Taurians. Since there is no necessity for this claim to be included in the supplementary, my advice for the authors would be to press the backspace button a few times in this section.
So all in all, this article by Saag and co. will be an interesting article with fascinating DNA samples when finally published. That said, I do think there are some changes the authors should make to the article for the final publication, with the primary focus on adding more realistic and better fitting models in line with previously published research on Scythian samples.
But wait, there is more! In addition to this article we also got:
Chinge-Tey I genome
Another Scythian related item I want to highlight is related to the Chinge-Tey burial complex in Tuva/ A while ago this genome was uploaded to National Center for Biotechnology Information:
Ancient DNA analysis of elite nomadic warrior
This a sequence for an upcoming article about the Chinge-Tey burial complex from Tuva. Chinge-Tey is a mound complex situated in the “Valley of Kings”, a region in Tuva which houses some of the most impressive royal Scythian mounds unearthed. The oldest example of a Scythian royal burial (Tunnug 1) has been uncovered in this valley [9].
Chinge-Tey-I Mound
Map showing burials in the “valley in kings” [10]
The burial in question is burial 9, which is a rather wealthy burial of a presumed prince/king of the Aldy-Bel culture [11]. This site was discovered in 2022 and burial 9 is insightful as the burial was undisturbed. Although most organic material perished, remnants of the skeleton have laid here unmoved for thousands of years and there are many gold objects. You can see the burial contents below:
Note that image B comes from a slab from Arzhan-2 rather than the Chingi-Tey burial.
Unfortunately the quality of the sample is really low so we will not get much useful data out of this individual. A bit disappointing, but fortunately we already have genomes from the royal Arzhan-2 grave in the valley of Kings, samples I0576 and I0577 from Unterländer et al. 2017.
A presentation mentioned that this sample had haplogroup Q and had the same mtDNA as the “Arzhan queen”, which would mean mtDNA haplogroup H. The autosomal ancestry seems similar to other central and eastern Scythian populations.
McColl 2024 - Stone age genomics of West Eurasia
A few months a rather massive pre-prepint came out:
Steppe Ancestry in western Eurasia and the spread of the Germanic Languages
The main focus on this preprint is focused on the Germanic people and their origin, which naturally interests me since I’m from the Netherlands.This article is very daring with novel IBD methods and their claims, and given that it is still in the pre-publication phase it is impossible to verify many of the different claims made by the authors. But that's not what we are here for today, McColl’s article also provides new data which is relevant for Scythian peoples.
IBD cluster 0_2_4_CEEuNAs
One of the IBD clusters in this article is 0_2_4_CEEuNAs. Most of the samples are Steppe_MLBA populations or iron age/historical populations derived from Steppe_MLBA such as Scythians/Sarmatians, Iron age southern Central Asians and the Swat valley samples from Pakistan.
These populations are linked based on markers used to track identity-by-descent, and the inclusion in said IBD cluster is not indicative of the genomic profile of the current individual, but it does illustrate ancestral ties.
In this article, the following samples were part of this IBD cluster:
Crimea - Suvlu-Khaye
CGG017681 - XX - I1
CGG017682 - XY - U4 - R1a1a1b2a2b2b
CGG017683 - XY - HV0a - G2a2a1a2a1a1b2~
These three samples come from the Suvlu-Khaye settlement and date from the late Sarmatian (CGG017682) and Gothic periods (CGG017681 and CGG017683) of Crimea. All three samples from this site were included in this cluster.
The Suvu-Khaye burial site features two burial phases; One was a phase of underground burials, dating to the 3-4th century AD. These remains belonged to Sarmatians, mostly men and many of the remains show signs of physical trauma [12].
The second phase featured burials in crypts, and dated to the 4th-5th century AD. A significant number of people with craniological similarities to contemporary individuals from Scandinavia and the Chernyakov culture, although it also featured people of different origins. Based on the trauma rate the population in this time seemed to have undergone less violence and may have been comparatively healthier than the individuals from the preceding phase.
CGG017682
The sample I am most interested in is CGG017682, the individual from the Sarmatian period at Suvlu-Khaye. During this period there was a significant influx of Sarmatians into the Crimea, attested archaeologically as well as attestations of names in the Bosporus kingdom. This individual was likely one of those Sarmatians based on his archaeological context. Given the time period I’d imagine this was an Aorsi or Alan individual. With mtDNA haplogroup U4 and Y-DNA haplogroup R-Y57 you more or less have one of the most common y/mtDNA haplogroup combinations of Sarmatian people, so genetics suggests the same.
I always find it interesting when we can have a look at the physical remains of DNA samples, unfortunately I only have a side profile for this one (first on the left):
As you can see CGG017682 rocked the iconic artificial deformed cranium, a hallmark of late Sarmatian populations.
The context date of 400 - 700 AD seems really late and would be a century after the death of Attila himself if you take the midpoint of 550 AD, yet the burial is attributed to the Sarmatian period. The sample is in probability younger because CGG017681 dates to 254 - 402 AD and it has the same context date - this sample was attributed to the Gothic period and came from a crypt burial which should postdate the Sarmatian phase, meaning that this individual probably dates to the 2-3rd century AD as the authors suggested.
CGG017681
The article which featured the remains also mentioned that the individual from burial 7, CGG017681, was brachycephalic with medium facial width and a high nasal bridge. On a craniometric PCA she was close to three sarmatian individuals and one individual from the Kizil-Koba culture; the authors suggested that the samples from the crypts possibly had a contribution from Pre-Greek individuals. In light of these suggestions it would be very interesting to see what the autosomal profile of this sample will be like.
Unfortunately there was no mention of burial 32 containing sample CGG017683, who seems to have a haplogroup which I’d imagine to be of Greek origin. Given the paternal haplogroup, the gothic society this individual lived in and the inclusion in a Scythian-Sarmatian related IBD cluster, I’d imagine this sample to be quite a mix of various populations.
Crimea - Chersonesus
CGG021473 - XY - U5b1b1+@16192 - E1b1b1a1b1a2~
CGG021474 - XX - W1c
Chersonesus was a Greek colony in the Crimea, founded around 500 BC, the necropolis the samples were retrieved from are from the 5th-4th century BC. Two out of the five samples from this necropolis were grouped in the0_2_4_CEEuNAs IBD cluster.
It would be interesting to compare the profiles of these individuals to the other samples from Chersonesus in order to see to which degree there was Scythian-related admixture present in these two samples.
Russia - North Caucasus Alans
CGG021451 - XY - I1b - R1b1a1b1b
CGG022000 - XX - K1a12a
This article included seven samples from the Sarmatian/Alan period in the north Caucasus, although only two were included in the CEEuNAs cluster. Most of the samples came from Zaragizh, an archaeological site in Kabardino-Balkaria which featured a fortified settlement and several burial groups. Sample CGG021451 came from Ekazhevo site in Ingushetia.
Alanic word from the early medieval period at Zaragizh
The Alanic period is a bit of a generalised archaeological term, perhaps a bit unfortunate as the ethnic makeup of the North Caucasus region in the Alanic period was very complex, and simply referring to them all as Alans obfuscates this diversity in my opinion.
The Alanic period is another one of those archaeological terms which is unfortunate due to the usage of an ethnonym for the designation of the culture. During the Alanic period the ethnic makeup of the North Caucasus was as complex as it is today, and simply referring to all populations within the zone of the North Caucasian Alanic culture as Alan obfuscates this diversity.
I always get a chuckle out of this description of the North Caucasian populations by Strabo, which was written some centuries prior to the Alanic period:
The same Dioscurias is the beginning of the isthmus between the Caspian Sea and the Euxine, and also the common emporium of the tribes who are situated above it and in its vicinity; at any rate, seventy tribes come together in it, though others, who care nothing for the facts, actually say three hundred. All speak different languages because of the fact that, by reason of their obstinacy and ferocity, they live in scattered groups and without intercourse with one another. The greater part of them are Sarmatae, but they are all Caucasii. So much, then, for the region of Dioscurias.
The other samples from Zaragizh were in different IBD clusters:
CGG022002 - XY - R1a - G2a1a1a1a1a1a1b~ - 0_2_3_1_Armenia
CGG022003 - XY - T2b36 - R1b1a1b1b - 0_2_3_1_Armenia
CGG022004 - XY - U1a1a - J2a1a1a2a2a - 0_2_3_1_Armenia
CGG022005 - XY - T1a1 - R1b1a1b1b - 0_2_3_1_Armenia
CGG021996 - XY - H5 - G2a1a1a1a1b - 0_4_3_1_WMed
All samples but CGG021996 were grouped within the 0_2_3_1_Armenia cluster, with CGG021996 being included in the 0_4_3_1_WMed cluster.
It might be that these individuals all had Sarmatian-related ancestry but due to their caucasian-related ancestry were included in the 0_2_3_1_Armenia cluster, or perhaps these samples did not have Sarmatian ancestry, or a combination of both.
Hungary - Sarmatian period
CGG021896 - XX - W4d
CGG021897 - XY - T2c1a - E1b1b1a1b1a
CGG021898 - XX - J1c2k
CGG021908 - XY - U2e2a1a2 - Q1b1b1~
CGG021909 - XY - J2a2b2 - R1a1a1b~
CGG021913 - XY - H8c - R1a1a1b~
CGG021922 - XY - K1a3 - R1a1a1b2a2b2b~
These were the Hungarian samples in different clusters:
CGG021899 - XY - H2a - I2a1b1a1b1a1a~ - 0_2_1_4_EWEu
CGG021912 - XY - T2d1b2 - J1a2a1a2~ - 0_4_3_2_SCEEu
CGG021916 - XX - T2b2b - 0_4_3_2_SCEEu
CGG021917 - XY - U5a1+@16192 - Q1b1b1~ 0_1_2_SouthScan
CGG021918 - XY - U4 - E1b1b1b2a1a6d2~ - 0_1_1_Baltic
The samples come from the Madaras and Kiskundorozsma cemeteries and date to the Sarmatian period. The timeframe of the is between 100 and 500 AD, with most samples falling in the 2nd-4th century AD range. The Sarmatian period of Hungary is another case of an archaeological period more intricate as its name suggests, which seems to be the running theme of this blog entry.
Skeleton of sample CGG021909 from Madaras, quite the short king standing at 161 cm.
The Sarmatian period gets its name from the Sarmatians, primarily the Iazyges, which migrated into Hungary in the first centuries AD. The issue here is that aside from Sarmatians, Hungary was inhabited by various Pannonian tribes, Celtic people, Germanic tribes. In the later phases of the migration era the ethnic situation got even more complex, which means that not every population within the domain of the Iazyges was Sarmatian. The Iazyges themselves became genetically mixed with all these populations as well as Romans and Balto-Slavic related peoples, meaning their genetic profiles were all over the place compared to the regular and fairly homogenous Sarmatian genetic cluster.
The Iazyes when migrating into Pannonia were quite impoverished, this is apparent when comparing the wealth of the burials in this period to those of the Sarmatians on the steppes [13]. Over time the subsistence economy of the Iazyges changed too, becoming more sedentary. Other archaeological changes which can be attributed to cultural influence from neighbouring peoples such as Dacians, Celtic and Germanic tribes are also notable.
IBD Cluster 0_3_2_EAsia
The second cluster relevant for Scythians in this article is0_3_2_EAsia. Previously published samples included in this IBD cluster are mostly ancient samples from East Asia, but also Scythian-period individuals from Central Asia and Siberia, as well as Finno-Ugric populations from Europe. Interestingly, the Yana RHS and Ust-Ishim samples also were included in this IBD cluster.
Aymyrlyg - Russia
CGG021493 - XX - A8a1
CGG021494 - XX - H15b1
CGG021496 - XY - U5b2a1a2 - R1a1a1b2a
CGG021497 - XX - H14b2a
The samples of relevance come from Aymyrlyg, an archaeological site in Tuva housing burials of the Sycythian and Hunno-Sarmatian periods. The Scythian remains belong to the later phase of the Sagly-Uyuk culture, dating to 300 - 100 bc. Physical remains at Aymyrlyg have shown evidence of scalping [14], a Scythian practice I covered in one of my previous blog entries.
Based on the description in the supplementary, the the first skeleton to the left is probably that of CGG021494, given that she had a relatively undisturbed skeleton and her head was on a stone slate:
Log House Tomb 6, the grave of sample CGG021494
We already have quite a bit of ancient DNA samples from the Uyuk culture and these new samples will probably not be revolutionary in terms of our understanding of the genetics of eastern Scythian populations, but the more samples the merrier.
So McColl’s upcoming article will be a very significant one for the study of Scythian peoples despite it not being the primary focus of the article. It is interesting to note the strength of the IBD segments as some of these previously published samples do not have much Scythian/Sarmatian ancestry at all yet are still included in this cluster due to their descent from it. This new angle of aDNA research is highly insightful, but I think some care has to be taken with the interpretation of some of the results.
I wonder what the precise methodology for assigning samples to clusters is, this part is not clear to me yet. Some of the samples in Cluster 0_3_2_EAsia are from populations with a small amount of east asian autosomal ancestry such as iron age Estonian samples. The Aymyrlyg and Central/Eastern Steppe samples all have 0_2_4_CEEuNAs ancestry, and all the Scythian samples have Cluster 0_3_2_EAsia ancestry, yet there is a bit of a division between European and Asian Scythian samples in regards to their IBD clustering. Something similar also applies to the Hungarian Sarmatian period samples, CGG021917 has Y-chromosome Q1b1b1 (which should be Q-YP4000), yet is grouped in the 0_1_2_SouthScan IBD cluster.
To round it up,
We still have to wait a while before we can get our hands on these new batches of Scythian samples and pick them to the bone as the folks in the online DNA community do. We had to wait a bit since the last DNA articles which covered the Scythians, but I am very content with the prospect of these samples being published soon.
References
Ivantchik, Askold. (2019). The Scythian Kingdom in the Crimea in the 2nd Century BC and Its Relations with the Greek States in the North Pontic Region. Ancient Civilizations from Scythia to Siberia. 25. 10.1163/15700577-12341351.
Котенко Виктория Владимировна (2014). Поселение Маслины в Северо-Западной Таврике (общая характеристика). Stratum plus. Археология и культурная антропология, (3), 325-341.
Metzner-Nebelsick, C. (2017). At the crossroads of the Hallstatt East. Connecting Elites and Regions: Perspectives on Contacts, Relations and Differentiation During the Early Iron Age Hallstatt C Period in Northwest and Central Europe.
А.Ю. Алексеев, Н.К. Качалова, С.Р. Тохтасьев (1993) - Киммерийцы: этнокультурная принадлежность.
Ivantchik, Askold. (2001). The Current State of the Cimmerian Problem. Ancient Civilizations From Scythia To Siberia. 7. 307-339. 10.1163/15700570152758043.
Who Were the Cimmerians, and where Did They Come From?: Sargon II, the Cimmerians, and Rusa I- Anne Katrine Gade Kristensen Munksgaard, 1988
Shramko, I., & Zadnikov, S. (2021). The Bilsk fortified settlement and the Hallstatt world. The Bilsk Fortified Settlement and the Hallstatt World . https://doi.org/10.34616/ssa.2021.63.123.148
Kashuba, Maya & Zanoci, Aurel. (2010). Locuinţele comunităţilor hallstattiene din spaţiul est-carpatic în secolele XII-VIII a. Chr. (tradiţii, deosebiri culturale şi perspectivele cercetării comparative). In: Tyragetia s.n. IV/1, 2010, p. 49-102. IV. 49-102.
Caspari, G., Sadykov, T., Blochin, J., & Hajdas, I. (2018). Tunnug 1 (Arzhan 0) – an early Scythian kurgan in Tuva Republic, Russia. Archaeological Research in Asia, 15, 82–87. https://doi.org/10.1016/j.ara.2017.11.001
Oleszczak, Łukasz & Pokutta, Dalia & Chugunov, K & Götherström, Anders. (2023). The food culture of the Iron Age nomadic elite from the 'Valley of the Kings' in Tuva: radiocarbon dating, stable carbon and nitrogen analysis of the Chinge Tey barrows (Turan-Uyuk Basin, Russia). Journal of Archaeological Science Reports. 51. 104186. 10.1016/j.jasrep.2023.104186.
Chugunov, K., & Sutiagina, N. (2022). An elite burial of the early Scythian period from the Chinge-Tey I kurgan: preliminary publication and research perspectives. Camera Praehistorica, 8–25. https://doi.org/10.31250/2658-3828-2022-2-8-25
Kazarnitski, A. (2016). Антропологическая экспертиза скелетных материалов из позднеантичного могильника Сувлу-Кая (Юго-Западный Крым). Казарницкий А. А. Антропологическая Экспертиза Скелетных Материалов Из Позднеантичного Могильника Сувлу-Кая (Юго-Западный Крым) // Краткие Сообщения Института Археологии, 2016, №243. С. 203-218.
Harmatta, J. (1970). Studies in the History and Language of the Sarmatians.
Murphy, E. M. (2003). Iron Age Archaeology and Trauma from Aymyrlyg South Siberia: An examination of the health diet and lifestyles of the two Iron Age populations buried at the cemetery complex of Aymyrlyg. https://doi.org/10.30861/9781841715223
