I would like to comment on one of your initial assumptions regarding speciation. Using a strictly genetic definition of species - IN VITRO mixing of gametes fails to produce viable offspring - one can still imagine that speciation need not happen "suddenly one morning" even if it occurs as a consequence of long, incremental change. Consider a group of immortal lab technitions, mixing 100 pairs of gametes (from two physically/behaviorally separate groups of the same species) in 100 separate test tubes, and repeating this experiment at intervals of X years. A series of test results might well follow a simple pattern of decreasing % of mixed gamete tubes resulting in a viable offspring - 90% viable the 1st test, 75% viable the next, 40% viable and so on. Thus, you don't "suddenly one morning" get two species - you get a statistically defined, probabilistic (and thus gradual) divergence into two distinct species over a period of, say, 20X years. Differential selection pressure and genetic drift could create enough genetic difference to gradually produce a parallel gradual decrease in gamete compatability. Just a thought I had about speciation some time ago. Hope it's of interest.