Take a look at this paper from the last issue of Nature. From the abstract:
“Here we confirm, using species-level molecular phylogenies for five families of non-passerine birds, that colour polymorphism is associated with accelerated speciation rates in the three groups in which polymorphism is most prevalent. In all five groups, colour polymorphism is lost at a significantly greater rate than it is gained. Thus, the general rarity and phylogenetic dispersion of colour polymorphism is accounted for by a combination of higher speciation rate and higher transition rate from polymorphism to monomorphism”.
It would be interesting to hear Brian Moore’s perspective on the phylogenetic analysis. And of course I wonder whether the same pattern would hold outside of birds.