Re: 'Out of Africa' still only a theory?!?!?

henry harpending (hxh5@psu.edu)
30 Sep 1995 20:27:44 GMT

They are talking about (nuclear) repeat polymorphisms, also called
microsatellites. These do indeed have a very high mutation rate. They
described linkage of one of these to a nuclear rflp site, which is a regular
point mutation in the nuclear DNA. African diversity is greater in the
repeats but not in the nuclear RFLP sites. My take on the evidence is that
since diversity accumulates in large populations, African ancestors got
larger first and so have accumulated more repeat diversity. But the nuclear
RFLP sites accumulate diversity over much longer time scales, tens of
millions of years, so there is no signature in them or in other low mutation
polymorphisms of the earlier African expansion. These nuclear RFLP sites
behave much like the classical markers (so called) that are the subject of
Cavalli's monograph. The diversity in these sites must be really ancient.

All these show that the biggest split in humanity is between Africa and
non-Africa, and they take this to indicate an earlier split. Another
hypothesis is simply that the Sahara has been a great isolating mechanism and
there has been less gene flow across the Sahara than elsewhere in the world.
In other words the ASSUMPTION that gene differences tell us about a tree-like
history is the basis of these reconstructions, and there is nothing to
support the tree interpretation. There is some evidence to suggest it is
wrong. A good paper is N. Takahata, 1993, Allelic Genealogy and Human
Evolution, Molecular Biology and Evolution 10:2-22.

In message <bcatDFqByt.613@netcom.com> - bcat@netcom.com (Bearcat) writes:
:>Sarah Tishkoff and Ken Kidd of Yale chose nuclear DNA, which is inherited
:>from both parents, in order to account for the possibility of men and
:>women becoming separated, for instance, after a hunt, and thus reflecting
:>population movement more accurately. Their study revealed a three-way
:>population division: Sub-Saharan Africans, northeastern Africans, and
:>non-Africans. South of the Sahara, there is *tremendous* genetic
:>variation. In northeast Africa, among Ethiopian Jews and Egyptians there
:>is less variation, and outside of Africa, there is less variation still.
:>This is because, Tishkoff and Kidd assert, the small population that
:>left Africa altogether did so in the very recent past, compared to the
:>thousands of generations we *all* lived in Africa. "The migration
:>was recent enough so that mutations and recombination haven't yet
:>reintroduced variability to the pattern. It's very consistent with a
:>recent spread of modern humans out of Africa - it's hard to imagine
:>anything else that could explain the data," says Kidd.
:>