Hard evidence for allopatric speciation

A question was posed to me, and I have been thinking about it on and off for a couple of days. Peter and Rosemary Grant has been studying Darwin's Finches (Geospiza) for decades, and at the talk they gave at Caltech last Thursday, he presented this map.



It details how the finches originally arrived at San Cristóbal from the mainland, and then radiated from there to Española, Santa Maria, Santa Cruz, and back to San Cristóbal, during which time they had speciated, resulting in two different species on San Cristóbal. The inference from the observations of living birds is that the finches speciated allopatrically. That is to say, that the genetic differences resulting from mutations took place in two different populations of finches, between which there was no gene flow (i.e. no mating between the two populations). This is the mode of speciation that is most easy to comprehend: two pupolations of the same species are reproductively isolated from each other by some physical barrier (such as a large body of water between San Cristóbal and Española), and as mutation and natural selection (or genetic drift) cause the two populations to diverge from each other, they eventually become different species.


Different species of Darwin's Finches have different beaks adapted to different types of foods.

My question to Peter Grant was what hard evidence there is that the mode of speciation was allopatric, and if sympatric speciation could be ruled out. Peter's reply was to ask me what I thought would constitute "hard evidence", and that is a good question indeed.

Sympatric speciation is a mode of speciation where there is no cessation of gene flow between any individuals - everyone continues to be able to mate with everyone else (of the opposite sex). Recent work on computer simulations have shown that this mode of speciation is possible with one or more of limited dispersal, assortative mating, and competition for resources. Since Darwin's Finches have specialized feeding on different foods, by changing the depth of their beak to match the seeds etc., it is at least possible to imagine that sympatric speciation occurred. Can that be ruled out given the evidence available? Is there any hard evidence for allopatric speciation in this case, or is that a conclusion arrived at because it is so much easier to understand allopatric speciation?

With that I am left pondering what evidence would satisfy me that this is a case of allopatric speciation.

My answer so far is to say that if among remnants of dead birds on San Cristóbal from the time that they are thought to have speciated, only finches morphologically similar to the birds that lived there at that time are found, but the remnants of birds on Española show signs that they changed there, then I would say that's pretty solid evidence of allopatric speciation. On the other hand, if it is found that among remnants of the first population if birds on San Cristóbal there is a continuous change in morphology, then I would consider that as evidence for sympatric speciation.

My question to you, dear reader, is whether you can think of anything else that would provide evidence either way?