Given that, like most neo-Darwinians, we accept the orthodox theory that speciation starts with geographical isolation, what should we expect to see in the fossil record?
Recall the hypothetical population of shrews, with a new species diverging on the far side of a mountain range, then eventually returning to the ancestral homelands and, quite possibly, driving the ancestral species extinct. Suppose that these shrews had left fossils; suppose even that the fossil record was perfect, with no gaps due to the unfortunate omission of key stages. What should we expect these fossils to show us? A smooth transition from ancestral species to daughter species? Certainly not, at least if we are digging in the main landmass where the original ancestral shrews lived, and to which the new species returned. Think of the history of what actually happened in the main landmass. There were the ancestral shrews, living and breeding happily away, with no particular reason to change. Admittedly their cousins the other side of the mountains were busy evolving, but their fossils are all on the other side of the mountain so we don't find them in the main landmass where we are digging. Then, suddenly (suddenly by geological standards, that is), the new species returns, competes with the main species and, perhaps, replaces the main species. Suddenly the fossils that we find as we move up through the strata of the main landmass change. Previously they were all of the ancestral species. Now, abruptly and without visible transitions, fossils of the new species appear, and fossils of the old species disappear.
The 'gaps', far from being annoying imperfections or awkward embarrassments, turn out to be exactly what we should positively expect, if we take seriously our orthodox neo-Darwinian theory of speciation. The reason the 'transition' from ancestral species to descendant species appears to be abrupt and jerky is simply that, when we look at a series of fossils from any one place, we are probably not looking at an evolutionary event at all: we are looking at a migrational event, the arrival of a new species from another geographical area. Certainly there were evolutionary events, and one species really did evolve, probably gradually, from another. But in order to see the evolutionary transition documented in the fossils we should have to dig elsewhere - in this case on the other side of the mountains.
