Theoretical genetics of batesian mimicry II. Evolution of supergenes

In this paper, we describe an extension of the computer model explained in paper I to simulate the genetics of two loci, and have used it to study the selection of genes modifying the effects of mimicry genes, either by improving the mimicry towards the same model, or by giving some mimetic resemblance to a new model.

The results suggest a possible explanation of the phenomenon of imperfect mimicry. They also predict that, as appears to be the case in Papilio dardanus, populations will either become fixed for the mimicry gene or else eliminate any non-specific modifiers that are not extremely closely linked to the mimicry locus. This leads us to a consideration of the evolution of mimicry supergenes, which we conclude is unlikely to have occurred by means of unlinked or loosely-linked genes being brought together into one region of the chromosome.