Variability of the Structural Coloration in Two Butterfly Species with Different Prezygotic Mating Strategies.

Abstract

Structural coloration variability was investigated in two Blue butterfly species that are common in Hungary. The males of Polyommatus icarus (Common Blue) and Plebejus argus (Silver-studded Blue) use their blue wing coloration for conspecific recognition. Despite living in the same type of habitat, these two species display differences in prezygotic mating strategy: the males of P. icarus are patrolling, while P. argus males have sedentary behavior. Therefore, the species-specific photonic nanoarchitecture, which is the source of the structural coloration, may have been subjected to different evolutionary effects. Despite the increasing interest in photonic nanoarchitectures of biological origin, there is a lack of studies focused on the biological variability of structural coloration that examine a statistically relevant number of individuals from the same species. To investigate possible structural color variation within the same species in populations separated by large geographical distances, climatic differences, or applied experimental conditions, one has to be able to compare these variations to the normal biological variability within a single population. The structural coloration of the four wings of 25 male individuals (100 samples for each species) was measured and compared using different light-collecting setups: perpendicular and with an integrating sphere. Significant differences were found in the near UV wavelength region that are perceptible by these polyommatine butterflies but are invisible to human observers. The differences are attributed to the differences in the photonic nanoarchitecture in the scales of these butterflies. Differences in the intensity of structural coloration were also observed and were tentatively attributed to the different prezygotic mating strategies of these insects. Despite the optical complexity of the scale covered butterfly wings, for sufficiently large sample batches, the averaged normal incidence measurements and the averaged measurements using an integrating sphere are in agreement.