Hi -
I recently sent a private message to a birding friend about discussions of lumping redpolls, based on relatively small genetic distances. I think the same issue is involved here.
DNA sequence data as described here are useful for estimating the length of time taxa have diverged in genetic isolation from each other. Substantial differences, particularly where multiple examples of each taxon are sequenced, provide evidence of extended independent evolutionary histories, which strengthens arguments for treatment as separate species.
However, not all speciation takes place through the gradual accumulation of small genetic differences during long periods of allopatry. This perhaps is the most common route to speciation in birds (debatable) but clearly is not the exclusive route. In other organisms speciation has been documented as occurring in response to strong bidirectional (disruptive) selection, and also through macrogenetic changes such as chromosomal re-arrangements (e.g. in the group of ground squirrels formerly called Townsend's). In such cases it is very possible for taxa to behave as "good" species long before the supposedly neutral parts of the genome show substantial differentiation.
It was reported recently (and discussed on this forum) that a chromosomal polymorphism is associated with the phenotypic differentiation in North American Junco taxa. I do not have a strong opinion on whether there should be more species of Junco recognized, but this research suggests that mechanisms for rapid speciation might be operative in some bird taxa.
I have a short list of North American bird taxa for which at least circumstantial evidence exists for rapid differentiation to the level of potential or documented reproductive isolation. This includes white-cheeked Branta, where "minima" has been known for the better part of a century to nest quite close to the nesting range of other white-cheeked geese now included within "hutchinsii," without interbreeding. The Branta bernicla also seems to show reproductive isolation out of proportion to its genetic differentiation. Also consider Loxia, Coccothraustes, and Tympanuchus. I am sure some of you can come up with others.
For some of these e know or can hypothesize some of the elements of the "special circumstances" that allow rapid phenotypic and behavioral differentiation. In Loxia, the combination of flocking behavior mediated by (learned?) contact calls, strong selection for morphological adaptation to different cone characteristics, seems to be at least part of the picture. In Branta, the combination of migration in family groups organized into flocks of close relatives, and pair formation in wintering areas within these flocks are probably important. And for White-cheeked Branta, the additional propensity for rapid morphological evolution, apparently in response to climatic and food-quality differences likely plays a significant role.
My views on this subject have been greatly influenced by my interest and study, over the past 14 years, in salmon biology, where selection can produce incredibly rapid responses in genetically based life history traits, as well as morphology, including rapid evolution of substantial levels of sympatric reproductive isolation. Salmon taxonomists prefer not to recognize these differentiated populations formally because that would complicate management, but the genetics lesson is clear.
So in summary, lack of gene flow, at least in moderately conservative portions of the genome, can be very useful in recognizing cryptic species, or species that remain moderately similar in morphology, but comparatively "recent" gene flow in these same portions of the genome should not necessarily preclude species recognition.
Wayne Hoffman
