The pea aphid, Acyrthosiphon pisum, exhibits several environmentally cued, discrete, alternate phenotypes (polyphenisms) during its life cycle. In the wing polyphenism, female progeny develop as either winged or unwinged depending on the extent of crowding or host plant quality experienced by the mother. Males also have the ability to develop as either winged or unwinged, but this is genetically determined by a single locus on the X chromosome and is thus referred to as a wing polymorphism. In order to gain insight into the patterns of gene expression that underlie the wing polyphenism and polymorphism we have used a pea aphid cDNA microarray to examine gene expression in winged and unwinged females and males. Results suggest that winged and unwinged morphs exhibit systemic differences in gene expression and that many of these differences are shared between the wing polyphenism and polymorphism (i.e., between females and males). In addition, adult winged and unwinged males exhibit pronounced differences when compared to adult females and fourth instar males, as well as to each other.

Insect dispersal dimorphisms, in which both flight-capable and flightless individuals occur in the same species, are thought to reflect a balance between the benefits and costs of dispersal. Fitness costs and benefits associated with wing dimorphism were investigated in the male pea aphid, Acyrthosiphon pisum (Harris) (Hemiptera: Aphididae). In one-on-one mating competitions in small arenas between winged and wingless males, the winged aphids obtained most of the matings with virgin females. In contrast, during competition experiments in larger cages with multiple individuals of each morph, the winged males no longer had a clear mating advantage over wingless males. In the absence of competition, wingless males had marginally higher lifetime reproductive success than winged males, probably because mating winged males tended to die faster than wingless males. In the absence of females, winged males survived longer than wingless males and this difference disappeared under starvation conditions. Mating males of both morphs died significantly faster than males without access to females. There does not appear to be a direct tradeoff of dispersal ability with life history characteristics in pea aphid males, suggesting that the advantages of producing winged males may result from outbreeding.

What is the relationship between variation that segregates within natural populations and the differences that distinguish species? Many studies over the past century have demonstrated that most of the genetic variation within natural populations that contributes to quantitative traits causes relatively small phenotypic effects. In contrast, the genetic causes of quantitative differences between species are at least sometimes caused by few loci of relatively large effect. In addition, most of the results from evolutionary developmental biology are often discussed as though changes at just a few important 'molecular toolbox' genes provide the key clues to morphological evolution. On the face of it, these divergent results seem incompatible and call into question the neo-Darwinian view that differences between species emerge from precisely the same kinds of variants that segregate much of the time in natural populations. One prediction from the classical model is that many different genes can evolve to generate similar phenotypes. I discuss our studies that demonstrate that similar phenotypes have evolved in multiple lineages of Drosophila by evolution of the same gene, shavenbaby/ovo. This evidence for parallel evolution suggests that svb occupies a privileged position in the developmental network patterning larval trichomes that makes it a favourable target of evolutionary change.