Re: Many mysteries
"It all falls down of course when you learn that a very few female mules can be fertile, when mated to a purebred horse or donkey."
Indeed. Does the same apply to genets (the male/female parental crossover of mules)? Now, a viable mule/genet cross, that would be something! Pretty much a return to a common ancestor currently believed to be extinct. Maybe one could cross a mule/donkey hybrid with a mule/horse hybrid?
"The situation in the avian world is much more complex and controversial."
There is a wonderful example in a certain population of seagulls. In Europe they appear as two separate species incapable of crossbreeding, and were long thought to be so. But as you go round the world one way, one of the species slowly mutates, and as you go round the other way, so does the other. When you get back to where you started, you find that there is one global population that interbreeds with its neighbours but, by the time it has spread round the world and met its other end, it can no longer interbreed. One species or more?
Speciation is not a binary on/off process. Classification into species was already in trouble before DNA analysis came along.