Two hypotheses have been proposed for the origin of insect wings. One holds that wings evolved by modification of limb branches that were already present in multibranched ancestral appendages and probably functioned as gills1-5. The second proposes that wings arose as novel outgrowths of the body wall, not directly related to any pre-existing limbs6. If wings derive from dorsal structures of multibranched appendages, we expect that some of their distinctive features will have been built on genetic functions that were already present in the structural progenitors of insect wings, and in homologous structures of other arthropod limbs. We have isolated crustacean homologues of two genes that have wing-specific functions in insects, pdm (nubbin) and apterous. Their expression patterns support the hypothesis that insect wings evolved from gill-like appendages that were already present in the aquatic ancestors of both crustaceans and insects.