Nature News (the story appears in other outlets, as well) is reporting a new discovery of an anatomically modern skull cap in Israel, at Manot Cave. The skull was found on a shelf and the associated deposits have been securely dated by uranium/thorium. The skull is unquestionably modern, with the characteristic “house” shape. The maximum cranial breadth is high on the parietals and overall shape is rounded, rather than flattened, as in Neandertals. Further, the rear of the vault is lacking the occipital bun present in Neandertals and the cranium is much shorter, having the length of a modern human. Further compound the issue is that the Neandertals in the region from the sites of Amud and Kebara are, based on the best estimates, coeval in time with the Manot Cave specimen.
The Nature story expounds thus:
What happened between 90,000 and 45,000 years ago, a period ten times the length of recorded history? Only the fossils can tell us, and they are few. It seems that the earliest modern humans got to the Levant and no farther. Mount Carmel in Israel hosts caves, such as Qafzeh and Skhul, where H. sapiens remains appear in levels older than those occupied by Neanderthals, Homo neanderthalensis. The replacement of our own species by Neanderthals seems to be an affront to our prejudices. So how did humans eventually make it to Europe?If so, then these human/Neandertal hybrids would have been the group that colonized Europe and that is why we (some of us, anyway) see archaic traits in the earliest modern humans from Europe that date to between 35 and 25 thousand years ago. These traits are present mostly in the shape and contours of the vault in the Mladeč and Předmosti specimens from central Europe. As these humans spread west, the archaic traits, which may have been under negative selection, diminished. Arguing against this hypothesis is that the Mladeč and Předmosti specimens are more archaic than the Manot Cave fossil. Alternatively, the Manot Cave individual may represent the modern end of the spectrum of a population that does represent Neandertal/early modern hybridization and that contained a high degree of morphological variability. Much is unclear.
The partial H. sapiens skull from Manot Cave goes some way towards providing an answer, as well as hinting at how complicated our early history might have been. It looks much more modern than skulls from Qafzeh and Skhul. It is also much younger, suggesting that the hominin was closer, genetically and evolutionarily, to the earliest known European representatives of our species. This skull, the simple answer would suggest, represents modern humans poised to expand out of Africa and colonize the rest of the world.
Here comes the ‘but’. Our modern genomes contain Neanderthal DNA. At some point, our ancestors bred with Neanderthals before they became extinct. Does the Manot skull represent that moment? We simply do not know. Welcome, Manot skull, to messy reality.
The authors seem to think that there was not continuous evolution from the Skhul/Qafzeh humans, which date to around 100k BP, to the population represented by the Manot Cave individual and that the Manot Cave 1 specimen represents a later migration of moderns into the area, although this is based largely on the fact that there aren't fossils that span that time period. The problem with this argument, of course, is that up until this this year, we didn't have the Manot Cave fossil, either. Perhaps, the more persuasive argument is that the Skhul/Qafzeh fossils are quite a bit more archaic than Manot Cave 1 and are morphologically closer to late archaic Homo sapiens in the shape plot they provide (this mirrors my own research in this area). On the other hand, a lot can happen in fifty thousand years and the required evolution could certainly have taken place.
That scenario is simplistic, but may explain some things that have confounded palaeoanthropologists for over fifty years.
After reading the paper, however, I am left with one question: What about the Herto cranium? It is a reasonably complete African cranium dating to around 150k that would provide a fantastic comparison and answer even more questions.