If you go to Creation/Evolution: The Eternal Debate and do a search for articles on "Yeast", you'll find some from March & May of 2003 on the experiments where they manipulated two distinct yeast species to become much more interfertile between each other. They showed that a chunk of a particular chromosome reversed itself, which helped force the original speciation.
I suppose that is why I hedged with the words usually and remote. I suppose, given enough time and the right contingencies, anything is reversible.