DISCUSSION AND CONCLUSIONS (continued)

5. How might evolving Heliconius jump from one adaptive peak to another without crossing fitness valleys below?

Single allele changes in regulatory genes (such as those controlling shutters and shutter color in the MCS clade) can create a strikingly novel pattern. Because major pattern themes are varied qualitatively by so few genes in this clade, hybridization of pattern genotypes having strikingly different phenotypes can produce qualitative variety among offspring resembling macro-mutation-like variants. Only a few such changes can instantly "beam" hybrids onto or near new peaks of mimetic protection without necessitating gradual evolution that puts between-peak variants at risk . Sheppard (1962), developed such ideas in reference to "major gene" mutation.

Examples of the potential of the MCS tool box to promote colonization of new peaks of protection can be found in synthetic hybrid zones. In one example only two steps were required for hybridization of distinct cydno races to produce a near perfect mimic of a toxic moth (Figure 7 lower left) and in another, the F1 of a cydno X ismenius cross show close mimicry of H. hecalesia, an ESS clade species distinctly patterned from the parents of the cross (Figure 7 middle right). These are but two of many examples that have I have noticed in the course of watching hybrid populations.

I previously mentioned examples from nature in which mimetic adaptation by both cydno and melpomene may have been enabled by genes crossing the species boundary. Specifically, I hypothesize that the steps that allowed cydno to rapidly colonize the black FW erato peak of protection in Colombia (Linares 1989, 1997a) and to obtain the variation that allows some populations of Heliconius to simultaneously track two distinct Müllerian comodels as polymorphic populations under frequency dependent selection in Ecuador (Kapan 1998) arose from introgression with melpomene.