"Depends a bit on where you look for differentiation. Whereas allopatric populations may not beocme differentiated morphologically if selection pressures stay the same, they will become differentiated at the molecular level, given enough time. The question is whether we should only acknowledge this once differentiation affects morphology, or whether consistent molecular differences (e.g., separate mtDNA haplotype clades) are enough."

Well, I figure any species concept has to be evolutionary (ironically, this is my main objection to the "evolutionary species concept" ) and therefore only differences that can reasonably be assumed to be evolutionarily important should be considered in species boundaries. This would mean, essentially, differential adaptation to environmental factors or sufficient differentiation in neutral genetic characters to preclude interbreeding.

That you have two mtDNA clades tells you that the two sets of organisms have been isolated in some form. It doesn't tell you anything about differential adaptation, and only at fairly high levels would it imply reproductive isolation. Basically, mtDNA clades tell you something about the past, but rarely say anything about the present or future. Burbrink's split of Pantherophis obsoletus provides a good example; he has nice mtDNA clades, but no evidence for differential adaptation and no evidence for current or future isolation of the clades.

"The reverse is of course also true - what do we do with allopatric populations that have undergone rapid morphological differentiation within a very short timespan, leaving them largely undifferentiated at the molecular level."

You call them separate species. Or subspecies if you're uncertain.

"Very often, what a taxonomist does will simply depend on the marker he happens to use or favour, or on what evidence is available. Hence the diversity of opinions among taxonomists."

True enough.

My feeling at this point, though, is that molecular data are excellent for phylogenies, but cannot, of themselves, tell us much of anything about taxon boundaries.

Patrick Alexander

[ Reply To This Message ] [ Subscribe to this Thread ] [ Show Entire Thread ]