[Aloysius88]

small pop. = fewer mutations
large pop. = difficulty in mutation spreading

I realize what you are talking about but the problem is a multi-variable problem with time, population, mutation frequency and harm, and information complexity all having to be dealt with. If you start crossing off variables you can make the problem seem solvable.

occasional beneficial mutations fixing in one population or the other are the icing on the cake.

The theory seems to call for isolation of the new population to fix the new beneficial mutation and so seems to tend toward ever more isolated smaller populations thereby trending away from a density that would ensure greater mutational frequency.

[Thatcherite]

The theory seems to call for isolation of the new population to fix the new beneficial mutation and so seems to tend toward ever more isolated smaller populations thereby trending away from a density that would ensure greater mutational frequency.

Once isolation has occurred (into just 2 populations) each can go its merry way. They may both be vast in number, even if one (animal) population starts with just one family or a single pregnant female that has for example got across a mountain range.

Beneficial mutations will tend to spread through the entire population over many generations even if the advantage they confer is tiny. Natural selection sees to that. The 2 isolated populations will experience different beneficial mutations and different drift (which in itself opens up new possibilities for beneficial and harmful mutations) and the process of speciation begins. THis will tend to accelerate if for some reason any aspect of the environment of the two populations is markedly different, because then the selection pressure operating on the two populations will not be the same. A beneficial mutation in one environment might be harmful or neutral in the other, for instance.

We can see this "in process". Try googling on "ring species" for examples where we can see 3 or more isolated groups and group A and C can still interbreed with group B, but not with each other. Are they fully speciated yet? It depends on your definition. The theory of evolution predicts that the definition of species is hard to nail down for reasons like this, and indeed it is.