Evidence for dynastic succession among early Celtic elites in Central Europe


Active member
Reaction score
Y-DNA haplogroup
This study has 35 Hallstatt samples from Baden-Württemberg. There are ten R1b, seven G2a-L497, one I2a-L38 and one J2b2a-L283>Z597 (aka Z628). G2a makes up 37% of the haplogroup, which may seem huge compared to modern Germany, but this study focuses on dynastic successions, so several of them could be the same lineage. The terminal SNP isn't the same simply because the quality of the samples varies and some have lower resolution.

What is more unexpected is to find an R1b-U106>L48>Z9>Z347 in southern Germany at the time. It's also unfortunate that the other samples weren't tested for SNP beyond P311 or P312. There should be at least some U152.

Sample IDSite NamemtDNA haplogroupY-DNA HaplogroupY-DNA SNPArchaeological Dating
APG001Asperg "Grafenbuehl"J1b1a1R1b1a1b1a1a2P312500-480 BCE
..Asperg "Grafenbuehl"......
APG002Asperg "Grafenbuehl"J1c3e1....480-450 BCE
APG003Asperg "Grafenbuehl"X2bR1b1a1b1a1a2P312480-450 BCE
..Asperg "Grafenbuehl"......
HEU001HeuneburgU5b2b....550-530 BCE
HEU002HeuneburgW6a....500-450 BCE
HOC001Eberdingen-Hochdorf "Biegel"J1b1a1R1b1a1b1a1P311530-520 BCE
..Eberdingen-Hochdorf "Biegel"......
HOC002Eberdingen-Hochdorf "Biegel"K1a4c1*1G2a2b2a1a1b1a1a2aCTS4803530-500 BCE
HOC003Eberdingen-Hochdorf "Biegel"K1a2aG2a2b2a1a1b1CTS9737530-500 BCE
HOC004Eberdingen-Hochdorf "Biegel"U4c2aG2a2b2a1a1b1a1~Z753530-500 BCE
LAN001Langenenslingen "Alte Burg"H30b1R1b1a1bCTS12972400-200 BCE
LWB001Ludwigsburg "Römerhügel"U4a1aR1b1a1b1a1a2P312530-450 BCE
LWB002Ludwigsburg "Römerhügel"H2a2a1R1b1a1b1a1aL151500-450 BCE
LWB003Ludwigsburg "Römerhügel"K1b2bG2a2b2a1a1b1a1a2a1b~FGC8304530-500 BCE
MBG001Magdalenenberg......616-530 BCE
MBG002Magdalenenberg..R1b1a1b1a1a1c2b2bZ347616-530 BCE
MBG003MagdalenenbergH1c9R1b1a1bPF6527616-530 BCE
MBG004MagdalenenbergH1c....616-530 BCE
MBG005MagdalenenbergT2+16189....616-530 BCE
MBG006MagdalenenbergK1a+195J2b2a1a1a~Z628616-530 BCE
MBG007MagdalenenbergH3....616-530 BCE
MBG008MagdalenenbergHV0+195I2a1b2aS2550616-530 BCE
MBG009MagdalenenbergH1c9....616-530 BCE
..Magdalenenberg......616-530 BCE
MBG010MagdalenenbergT2b4G2a2b2a1a1b1a1aZ1823616-530 BCE
MBG011MagdalenenbergK1a....616-530 BCE
MBG012MagdalenenbergH2NANA616-530 BCE?
MBG013MagdalenenbergI2R1b1a1b1a1L52616-530 BCE
MBG014Magdalenenberg..R1b1a1bCTS12478616-530 BCE
MBG015MagdalenenbergH5....616-530 BCE
MBG016MagdalenenbergU5a1a2bG2a2b2a1a1bL497616-530 BCE
MBG017MagdalenenbergH27G2a2b2a1a1b1a1a2a1b~FGC8304616 BCE
SCN001Ditzingen-SchöckingenH3g....500-450 BCE

R1b-P312, G2a-L497 and I2-L38 have long been associated with Alpine Celts. There is a particularly strong correlation between the distributions of R1b-U152 (aka S28) and G2a-L497, which goes beyond Hallstatt and La Tène Celts and also include Etruscans and Italic tribes.



Another surprising element from the paper is what they say here:

Interestingly, we find that individuals with haplogroup G2a-L497 (for example, MBG017, MBG016 and HOC004) exhibit significantly more southern European ancestry than individuals carrying haplogroup R1b-M269 (for example, HOC001, APG001 and MBG003)

This is odd because it would mean that G2a-L497 lineages would be recent migrants to southern Germany who hadn't had time to mix with the mainstream population. They all date from around 600 to 500 BCE. We could imagine that Etruscans could have provided those lineages, but the problem with that is that they had vastly more R1b-U152 than G2a-L497. There isn't really any other region of southern Europe that was close enough and where G2a-L497 was a significant lineage. In fact I doubt that there would be any outside of Italy. All the other regions of southern Europe (Iberia, Greece) where L497 is found today owe its presence to Celtic and Roman colonisation and wouldn't have had any L497 around 600 BCE.

It's probably just a coincidence that the samples with more southern European ancestry all happen to be G2a-L497. They could just as well have been R1b-P312 or U152. The authors tried to infer their place of origin and it is indeed northern Italy, though not necessarily Etruria. That's interesting because it shows that there were still considerable population exchanges (at least among the elite) across the Alps around 600 to 500 BCE, which is at least 600 to 700 years after the Urnfield/Hallstatt expansion into Italy.

Last edited:
Here under some extracts written in "italics" - comments are of mine for some blog, don't mind.
1 – on their PCA we see the relatively loose S-Germany Celtic cluster spans a little bit over the today Germans cluster (I suppose SW and Southern ones) and surely over what I suppose to be today Northern Spanyards. It’s true the centrum seems rather around today Southern France, so a it different from the Champagne IA Celts. So « separate in genetic space from present-day Germans » is not accurate, spite the German centroid todate is far enough, effectively. It would confirm if general, the same profile as among Alsace IA Celts, rather « southern » compared to other Celts, with an input of a Y-G2a rich element. Gretzinger : Interestingly, we find that individuals with haplogroup G2a-L497 (for example, MBG017, MBG016 and HOC004) exhibit significantly more southern European ancestry than individuals carrying haplogroup R1b-M269 (for example, HOC001, APG001 and MBG003) - < Urnfields times ??? -

2- That there had been created a continuum at those times is surprising in no way, History explains it.. Identity is another thing, similarity or affinity is an unprecise term.

Interestingly, this third inter-site group of relatives exhibits significantly more southern European ancestry than the rest of our analysed individuals (93.6 ± 1.9% versus 59.9 ± 3.9%; two-sided Wilcoxon rank-sum exact test; W = 0, P = 0.0002259) and, consequently, significantly more Early European Farmer (EEF) ancestry (55.6 ± 0.9% versus 48.4 ± 1.1%; two-sided Wilcoxon rank-sum exact test; W = 0, P = 0.0002259) (Supplementary Fig. 2.8) (for details on EEF ancestry decomposition, see Methods and Supplementary Notes 2 and 4). This might indicate a non-local, southern European origin of the ancestors of the Magdalenenberg elite. Consequently, we applied MOBEST35 to perform spatiotemporal interpolation of their genetic affinity to ~5,660 previously published ancient genomes, obtaining similarity probabilities across early Iron Age Europe that can be interpreted as proxies for geographical origin (Supplementary Note 2). We detect for all four of these samples (MBG004, MBG016, MBG017 and HOC004) a putative transalpine origin in northern Italy, while all other tested Hallstatt individuals’ origins are located north of the Alps, close to their respective sites (Fig. 3a,b). Remarkably, these individuals feature excess EEF ancestry on the X chromosome in comparison with the autosomes (83.5 ± 9.9% versus 55 ± 1.1%). Applying the formula described in Mathieson et al.36, we find evidence that the EEF admixture was significantly female biased (Z  = −2.86), suggesting an excess of females over males with south-European origin among their ancestors. In contrast, we detect no difference in EEF ancestry on the X chromosome and the autosomes in the rest of the sampled Hallstatt population (43.6 ± 5.7% versus 49 ± 0.6%) and, consequently, no evidence for sex-biased admixture in the main group (Z = 0.93).

3- Spite evoking the input of a component or its increase into Lech valley BA pop Gretinger doesn’t compare it to a turnover when he says : … we find the Iron Age individuals to be separate in genetic space from present-day Germans and falling closer to present-day French and other southern European individuals (Supplementary Fig. 4.1). Compared with contemporaneous data, the Hallstatt individuals cluster homogeneously intermediate between Iron Age samples from present-day France and the Czech Republic, together with Bronze Age samples from the Bavarian Lech valley within the present-day French variation. The genetic affinity between our Hallstatt individuals from southern Germany and individuals from Bronze and Iron Age France is part of a broader genetic continuum spanning from Iberia to the Balkan peninsula, featuring a common genetic ancestry component.

(BTW this continuum concept can hide a complicated process of multidirectionnal local admixtures, when the writing gives the impression of a big almost homogenous move of some component)-
some other PCA pics
I’m assuming there are no G25 coordinates yet?

This thread has been viewed 1089 times.