NOTE 20F: There are several striking examples where caterpillars have just this kind of plasticity and grow up with the mimetic phenotype that best camouflages them on the host they find themselves on. See Figure P20.5.
NOTE 20G: If it is not possible for a “mixed strategy” to evolve, involving phenotypic plasticity (see, e.g., Fig. P20.5), then a genetic polymorphism may evolve. However, that may not perfectly match the ESS: For example, heterozygotes or recombinants might be produced that do not specialize perfectly. Then, there may be selection favoring speciation. This is discussed on pages 650–654.
NOTE 20H: We assume that the PU females can find TP males however rare they are.
NOTE 20I: As we saw in Problem 19.9, this is a general result for the change in frequency of a neutral allele that is associated with a selected allele.
NOTE 20J: This approximate result is not especially important in itself. However, this simple model does illustrate the components of Fisher’s runaway process: An allele that is favored by sexual selection will become associated with any alleles that cause females to prefer it, and so both will increase together. This problem follows an influential analysis of this model that included natural selection, and arbitrary strength of preference, was made by Kirkpatrick (1982).