Lacan et al., who tested the Neolithic site of Treilles in Southwest France (G2a + I2a) published a new study on Neolithic Spain. The samples are 2000 years older than in Treilles, and identified both G2a and E-V13. This could either confirm a Neolithic origin of E-V13 or, my recent proposal that E1b1b crossed from Africa to Europe before the Neolithic. Spain is indeed the most likely point of entry from North Africa, along with South Italy.
Considering that a substantial percentage of Neolithic mtDNA in Iberia is African (L1b1, L2, L3, respectively in Andalusia, Navarra and Valencia), this could indeed be a sign of a direct migration from North Africa. My hypothesis is that this migration was due to the desertification of the Sahara at the end of the last Ice Age.
Another interesting point is that they mention that the mitochondrial lineages are mostly pre-Neolithic. If Neolithic farmers kept marrying local girls (from the hunter-gatherer community), that would explain why West Asian autosomal genes diminish gradually as we move further away from the Middle East. That may be why there is so little West Asian admixture in Iberia and Sardinia despite the relatively high percentage of G2a.Originally Posted by Lacan et al.
EDIT : The individuals results are as follow :
- Y-DNA : five G2a men and one E-V13. All confirmed by SNP test except one G2a.
- mtDNA : three K1a, two T2b, one H3 and one U5.
The E-V13 man is the mtDNA U5.
I can't see why Lacan et al. think that these maternal lineages are pre-Neolithic, except for the U5 and H3.
Mt-haplogroup K, and K1a in particular, is so far the most overrepresented among Neolithic samples (16%) compared to the present-day population. This is exactly the frequency observed today in the Levant and in Kurdistan (Georgia is also high with 11%). This makes of mtDNA K one of the most distinctive Near Eastern marker, the maternal equivalent of G2a.
Mt-haplogroup T has never been found in Europe prior to the Neolithic (except for one sample in the Pitted Ware Culture, which is actually contemporary of the Late Chalcolithic and early Bronze Age elsewhere in Europe). There is an interesting inversion of frequency between mtDNA hg J and T in the early and late Neolithic. Based on the current data (247 samples), the frequency of hg T dropped from over 19% in the Early Neolithic to only 4.5% the Late Neolithic, then up again to 10.5% in the Bronze Age (similar level to today).