It almost seems like there are two separate research project under way regarding the evolutionary history of Homo sapiens. One focuses on recent humans, tends to use DNA as a major source of information, and from this base projects back into the past. This approach tends to confirm the idea that humans share an African origin with a subsequent spread from Africa, with various degrees of complexity in that series of historical events. The other focuses on early human remains, sometimes including remains that would be placed by some in a separate species or sub species. This sort of approach typically results in a similar conclusion regarding the African origin.
But the two populations ... modern or recent humans and humans several tens of thousands of years old do not necessarily share a history that has been coherently assembled by researchers. Of course, they shared an actual history, but do the lines linked together to make reconstructed phylogentic trees from these two data sources match up or integrate in any sensible way? Not really.
A new paper coming out in the Journal of Human Evolution shares some light on how we might connect these similar but unintegrated reconstructions.
Michael A. Schillaci states in the abstract of Human cranial diversity and evidence for an ancient lineage of modern humans:
This study examines the genetic affinities of various modern human groupings using a multivariate analysis of morphometric data. Phylogenetic relationships among these groupings are also explored using neighbor-joining analysis of the metric data. Results indicate that the terminal Pleistocene/early Holocene fossils from Australasia exhibit a close genetic affinity with early modern humans from the Levant. Furthermore, recent human populations and Upper Paleolithic Europeans share a most recent common ancestor not shared with either the early Australasians or the early Levantine humans. This pattern of genetic and phylogenetic relationships suggests that the early modern humans from the Levant either contributed directly to the ancestry of an early lineage of Australasians, or that they share a recent common ancestor with them. The principal findings of the study, therefore, lend support to the notion of an early dispersal from Africa by a more ancient lineage of modern human prior to 50 ka, perhaps as early as OIS 5 times (76-100 ka).
What this new research is telling us is something that has been emerging as the most likely scenario for some time now. Homo sapiens arose in Africa and spread around that continent, and into southern regions of the rest of the Old World (best documented in southwest Asia) including Australia. Subsequently, there were significant and widespread and fairly complex movements. This means, for instance, that the skeletal remains found several tens of thousands of years ago in, say, Europe, may bear much less relationship to modern Europeans than they do to modern people somewhere else. Upper Paleolithic Europeans together with current or recent humans can be seen as one closely related group that does not include the ancestors of early Australasians. The two larger groups do, of course, share an earlier common ancestor.
This study also finds that the earlier group is (physically) marginaly closer to Neanderthals than with Upper Paleolithic Europeans. This would not be surprising if either there was intermixing between early "modern humans" and Neanderthals (not strongly suggested by the DNA in other studies) or if the traits being studied were simply retained in both populations, with subsequent changes in traits (from the Upper Paleolithic onwards) being somewhat more rapid (which does conform to other studies).
Here is the simplified results of the detailed analysis presented in this paper:
From Figure 4: Linearized neighbor-joining trees describing the phylogenetic relationships among groupings of the genus Homo based on (a) size/shape and (b) shape distances. EA, early Australasians; EMH, early modern humans from the Levant (Skhul/Qafzeh); UPE, Upper Paleolithic Europeans.
The observed pattern of relationships between the early modern humans from the Near East and modern human population samples from elsewhere is consistent with the early-dispersal hypothesis described in the recent literature ... assuming the Levantine early humans represent either a source population for the initial dispersal out of Africa or a population that experienced significant gene exchange with the source population prior to dispersal. The observed relationship could have also resulted from recent common ancestry for the Levantine early humans and an African source population. Morphological convergence seems to be an unlikely explanation for the observed pattern of relationships given the concordance of results from size/shape and size-corrected shape data sets. Nevertheless, the possibility of convergence or homoiology must be considered.
The results presented here hint at the possibility of the existence of two lineages of modern humans (H. sapiens) outside of Africa, one more ancient than the other. The notion of a distinct ancient lineage of modern humans expanding across southern Asia into Australasia is supported directly by recent ancient mtDNA work describing a very ancient sequence in a sample of early Australians ... In addition, research on the coevolution of human parasites, including head lice, tape worms, bed bugs, and the human T-cell lymphotropic virus (HTLV), also seem to indicate that there may have been two lineages of humans throughout much of our evolutionary history
SCHILLACI, M. (2008). Human cranial diversity and evidence for an ancient lineage of modern humans. Journal of Human Evolution DOI: 10.1016/j.jhevol.2007.10.010