In antiquity what we term Tunisia and Tripolitania were part of "African province." Just as "Asia" originally referred to the margins of what we now term Asia (regions of Anatolia), so "Africa" originally denoted a subset of the northwest fringe of the continent which became Africa. In biogeography this segment of the continent is actually not part of Africa (it is part of the Palearctic ecozone). And yet the vicissitudes of early modern cartography are such that continent had to be bounded by water on as many sides as possible, and today we clumsily make recourse to the term "Sub-Saharan Africa" to distinguish that region from the northern littoral, which is really part of the Mediterranean world. This context is somewhat relevant when we evaluate a new PLoS ONE paper, North African Populations Carry the Signature of Admixture with Neandertals. This paper makes little sense unless you've read an earlier one, Genomic Ancestry of North Africans Supports Back-to-Africa Migrations. In that paper authors make the case that a majority of the ancestry of modern Maghrebis seems to date back to before the Neolithic, from a late Pleistocene migration of Eurasians. Though the indigenous North African element is naturally somewhat diverged from other West Eurasian components, it nevertheless comes out of the West Eurasian milieu, on the order of 10-40,000 years before the present. In the PLoS ONE paper the authors show that the Neandertal admixture in the modern Mahgrebis is consistent with the ancient event on the order of ~60,000 years before the present. I was curious to see that the paper was careful to observe that the balance of scholarship is still shifting toward a Eurasian admixture model, instead of ancient population structure within Africa. And these findings push it further in that direction, in particular showing that Neandertal admixture is not contingent upon European or West Asian admixture, but was found in large proportions even among predominantly North African groups. There are some qualms I have about the number of SNPs, but I can't say much more until I try and replicate (it may be that 50-100,000 SNPs is sufficient for their questions). What I am confused about is what they expected. After all, it is well known that the peoples of North Africa are part of greater Eurasia. What Sub-Saharan African admixture exists seems to be mostly a function of the Islamic era. As such, we'd expect to see Neandertal admixture, in particular because a previous group already argued that the indigenous substrate of North Africa derives from an Ice Age migration out of Eurasia. More importantly the "true answer" for the question of weather the affinity to Neandertals among Eurasians is due to ancient African population structure (so that the African ancestors of Eurasians happened to be closer to Neandertals than the Africans left behind) will require Sub-Saharan data sets. With whole genome coverage and thicker sampling of the continent I'm moderately optimistic. I do know that there is some evidence of Neandertal admixture in Kenyan Masai. This is probably another clue to an ancient back migration rate, though at some point someone will try and see where African and non-African Neandertal ancestry relate to each other on a phylogenetic true (preliminary data I've heard about seem to indicate that it's intrusive to the continent).