Re: A fully developed creature can evolve?


"Anon." <bob.ohara@xxxxxxxxxxxxxxxxx> wrote in message news:dg84u0$2grd$1@xxxxxxxxxxxxxxxxxxxxxx
> With a population that is extensively sub-structured, natural selection
> may not have enough of an effect to be able to purge bad alleles. I
> mention this partly because a couple of the good empirical work showing
> this was done in Finland: on butterflies and Daphnia.

Interesting! I would never have guessed this. But now that I think
about it, I can imagine one possible mechanism. The subgroups are
small enough so that the bad allele (or for that matter, the good one)
becomes fixed in many subgroups. NS will have no effect on the
frequency of the allele within these subgroups. Furthermore, classical
group selection (against subgroups harboring the bad allele) can't
actually extinguish these groups due to density-dependent selection.
Do I have it basically right?

Hmmm. Aparently that old standard pop gen assumption of random mating
is more important than I thought, particularly for genes with selection
coefficients low enough so that they become 'neutral' in a small subgroup.

But my intuition is acting up again. It tells me that the mechanism
I suggest would only work if the gene flow among populations is less
than about one migrant per generation. But this means that each subgroup
is an incipient species!


.

Loading