Out of Africa and into the Levant: replacement or admixture in Western Asia?

Late Pleistocene Israel is the region in which issues of population mixture or competition at the time of the emergence of modern humans are most like ly to be solved. For those who believe that modern humans first arose in Af rica and subsequently spread throughout the world replacing archaic populat ions, the Levant would be the first region where such archaic populations w ere encountered. For those who regard the Levantine Neandertal populations as late emigre from a glaciated and inhospitable Europe, the Levant is the place where it is most likely that Neandertals encountered other human popu lations. If ever there was a time and place where we can examine the questi on of whether European and African populations exchanged ideas and mates, o r competed with each other without genetic exchanges, this is it! In this p aper we test the null hypothesis of a single human species occupying the Le vant at the onset of the Late Pleistocene. An inability to delineate two di stinct groups among the Levantine hominids would support the null hypothesi s, while a demonstration of the presence of two morphs would lead to its re futation. We use non-metric traits to examine the eight most complete adult Levantine human crania to try to refute the contention first proposed by M cCown and Keith (1939. The Stone Age of Mount Carmel: the Fossil Human Rema ins from the Levalloiso-Mousterian, Vol. II. Clarendon Press, Oxford), that the Levant "Neandertals" (Amud, Tabun) were the same species as the "early modern humans" (Qafzeh III, VI, IX; Skhul IV, V, IX). To test this hypothe sis we use individual specimens as "operational taxonomic units", and asses s it using phylogenetic analysis as a heuristic clustering procedure. While our analyses produce many different trees, none of the most parsimonious o nes reveal a separate Neandertal clade. Furthermore, we conducted a pairwis e difference analysis of these data, which also failed to reveal a unique r elationship between the Neandertal crania that would be expected if these h ominids were a different species from that represented by Qafzeh and Skhul. We acknowledge that the bases for refutation are necessary but not indispe nsably sufficient conditions, and yet nevertheless, our findings fail to re fute the null hypothesis. Instead our results suggest that the traditional "Neandertal" versus "modern human" groupings in the Levant may not be as di stinct as often thought. This would imply that as populations left Africa, they interbred with the Late Pleistocene inhabitants of the Levant, and sug gest that as different populations moved or expanded their range, subsequen t human evolution be viewed as a consequence of the continued mixing of ide as and genes. (C) 2001 Elsevier Science Ltd and INQUA. All rights reserved.