I am especially interested in the genetic evolution of an adaptive mimicry polymorphism in the species H. numata, an Amazonian species which has up to 7 forms segregating in a single locality, each of which is a precise mimic of local species and races of Melinaea butterflies (Nymphalidae: Ithomiinae) [figure, left]. Crosses and Mendelian genetics of the mimicry variation in H. numata has shown that all elements of wing patterns variation are controlled at a single locus P; recombinants between wing pattern elements are occasionally found, indicating P is a supergene, i.e. a tight cluster of genetic elements that hardly ever recombine. We have shown that this supergene is a positional homologue of one of the several independent colour-pattern loci which control geographic variation in other species of Heliconius; In other words H. numata produces its entire diversity of forms using one single tool in the Heliconius toolbox of wing pattern loci, where other species use a diversity of tools. My research is now focused in a large part to characterise the genetic identity, the genomic organisation, and the comparative evolution of this condensed genetic architecture, to understand how variation at a master switch genes can determine the maturation and spatial arrangement of pigmented scale cells on the wing, leading to adaptive variation in phenotype.

Warning colour and mimicry have been the subject of intense theoretical and experimental work in the laboratory, but it is still unclear how an ecological phenomenon that reduces diversity and selects for phenotype convergence can in fact be expressed in the most bewildering diversity of warning patterns in nature. It seems, therefore, that diversity in warning colours and mimicry might be a sort of inevitable by-product of the way species evolve warning patterns and mimicry in the first place. Müllerian mimicry evolves under strong positive density dependence, producing a form of selection that strongly favours local monomorphism, but also has a propensity to produce spatial mosaics of selection and conserve diversity at the regional scale. We are interested in theory as well as ecological and molcular data showing how spatial and populational processes can influence selection for mimicry and the maintenance of diversity from the local to the landscape scale. This may also help understand the biogeography and history of diversification in some of the most diverse communities on Earth, the tropical Andes.

Mate choice and mating success are strong determinant of sexual selection. In Bicyclus anynana (Nymphalidae: Satyrinae), we showed that inbreeding strongly influences male mating success, suggesting inbreeding depression may affect either behavioural performance and chemical signalling during courtship. However, we found that this effect is much stronger in semi-natural, free-flying conditions compared to lab experiments, suggesting much of the effect is certainly in the behavioural performance of the males. I am now particularly interested in chemical signalling in mimetic butterflies, which have a wide diversity of male androconial hair pencils or patches on their wings, and varied chemical compounds at different taxonomic level. How the diversity of pheromone cocktails correlates with genetics, with phenotypic variation, and with mimicry, are questions we are interested to tackle.