Alan Templeton has followed up his work in Out of Africa again and again where he argued that the human genome holds evidence that is incompatible with an exclusively recent origin of modern humans. While human mtDNA and Y-chromosomes coalesce to a time less than 200ky and are thus contemporaneous with the emergence of anatomically modern humans, we also possess genes that are far older, and are best explained by the retention of ancestry from populations other than those emerging recently in the African continent.
Templeton used classical hypothesis testing to formally reject the Out of Africa with replacement model with a staggeringly low probability. It's possible that his procedure might be contested, but I just can't see how anyone can really support the replacement theory any more.
Templeton also believes that there were several Out of Africa movements, each of them leaving a "signature" in the genomic ancestry of modern humans. Different genes are dated to each of these three movements:
Templeton also finds that the three Out of Africa movements correlate quite well with archaeologically/anthropologically established movements:
These dates overlay well upon the fossil and archaeological record, with the first expansion corresponding to the original expansion of Homo erectus out of Africa into Eurasia (Aguirre and Carbonell, 2001; Bar-Yosef and Belfer-Cohen, 2001; Anto´n et al., 2002; Vekua et al., 2002) and the development of a culture capable of keeping impaired individuals alive for many years (Lordkipanidze et al., 2005). The second expansion corresponds to the spread of Acheulean culture into much of Eurasia after an earlier African origin (Asfaw et al., 1992; Hou et al., 2000) and the initiation of a substantial increase in cranial capacity (Ruff et al., 1997; Relethford, 2001b; Rightmire, 2004). The most recent expansion out of Africa corresponds to the spread of several anatomically modern traits into Eurasia after an earlier African origin (Stringer, 2002; White et al., 2003).
This certainly also seems to agree with Erik Trinkaus' recent statement:
Versions of the assimilation model have remained contenders for the interpretation of modern human phylogenetic emergence, if frequently overshadowed by the more polarized regional continuity (with gene flow) and (out of Africa with) replacement scenarios. The last two interpretations are finally intellectually dead. Both are contradicted by available evidence, and it is time for the discussion to move on. Yet, despite the general acceptance of some form of the assimilation model, issues remain.
Update... but Milford Wolpoff writes in the comments (which disappear after 4 months on Haloscan):
I just don't see how Templeton's excellent review somehow supports Trinkaus' comment. ET asserts that multiregional evolution is "finally intellectually dead" whereas Templeton describes an interpretation of human evolution that is compatible with multiregional evolution, as he has noted. Trinkaus would replace multiregional evolution with his "assimilation theory" - an interpretation that does not seem to differ from multiregional evolution except by name, which perhaps explains the problem.
New analysis shows three human migrations out of Africa (excerpt):
Feb. 2, 2006 — A new, more robust analysis of recently derived human gene trees by Alan R. Templeton, Ph.D, of Washington University in St Louis, shows three distinct major waves of human migration out of Africa instead of just two, and statistically refutes — strongly — the 'Out of Africa' replacement theory.
That theory holds that populations of Homo sapiens left Africa 100,000 years ago and wiped out existing populations of humans. Templeton has shown that the African populations interbred with the Eurasian populations — thus, making love, not war.
"The 'Out of Africa' replacement theory has always been a big controversy," Templeton said. "I set up a null hypothesis and the program rejected that hypothesis using the new data with a probability level of 10 to the minus 17th. In science, you don't get any more conclusive than that. It says that the hypothesis of no interbreeding is so grossly incompatible with the data, that you can reject it."
Templeton's analysis is considered to be the only definitive statistical test to refute the theory, dominant in human evolution science for more than two decades.
"Not only does the new analysis reject the theory, it demolishes it," Templeton said.
Templeton published his results in the Yearbook of Physical Anthropology, 2005.
American Journal of Physical Anthropology
Volume 128, Issue S41 , Pages 33 - 59
Haplotype Trees and Modern Human Origins (p 33-59)
Alan R. Templeton
A haplotype is a multisite haploid genotype at two or more polymorphic sites on the same chromosome in a defined DNA region. An evolutionary tree of the haplotypes can be estimated if the DNA region had little to no recombination. Haplotype trees can be used to reconstruct past human gene-flow patterns and historical events, but any single tree captures only a small portion of evolutionary history, and is subject to error. A fuller view of human evolution requires multiple DNA regions, and errors can be minimized by cross-validating inferences across loci. An analysis of 25 DNA regions reveals an out-of-Africa expansion event at 1.9 million years ago. Gene flow with isolation by distance was established between African and Eurasian populations by about 1.5 million years ago, with no detectable interruptions since. A second out-of-Africa expansion occurred about 700,000 years ago, and involved interbreeding with at least some Eurasian populations. A third out-of-Africa event occurred around 100,000 years ago, and was also characterized by interbreeding, with the hypothesis of a total Eurasian replacement strongly rejected (P < 10-17). This does not preclude the possibility that some Eurasian populations could have been replaced, and the status of Neanderthals is indecisive. Demographic inferences from haplotype trees have been inconsistent, so few definitive conclusions can be made at this time. Haplotype trees from human parasites offer additional insights into human evolution and raise the possibility of an Asian isolate of humanity, but once again not in a definitive fashion. Haplotype trees can also indicate which genes were subject to positive selection in the lineage leading to modern humans. Genetics provides many insights into human evolution, but those insights need to be integrated with fossil and archaeological data to yield a fuller picture of the origin of modern humans.