Résumé
We reconsider the taxonomy of a group of closely related Ostrinia spp., illustrating how useful Mayr's biological species concept remains for studying speciation patterns and processes. We review and re-analyse recent data on Ostrinia scapulalis, Ostrinia nubilalis, Ostrinia narynensis and Ostrinia orientalis, along with those obtained over > 45 years in the former Soviet Union. The ten species of the 'trilobed uncus' group in the Ostrinia genus are classified into subgroups according to male mid-tibia morphology. However, none of the characters that further discriminate between them (female sex pheromones, male genitalia and calling time) varies together with male mid-tibia morphology, and neither do molecular markers. Moreover, male mid-tibia morphology appears to depend on only two diallelic loci and seems to be unrelated to reproductive isolation between Ostrinia taxa. By contrast, reproductive isolation is strongly related to host-plant type. In accordance with Mayr's species concept, we thus propose a revision of the trilobed uncus Ostrinia spp. based primarily on host-plant type. We propose that O. narynensis Mutuura & Munroe, 1970 (syn. nov.) and O. orientalis Mutuura & Munroe, 1970 (syn. nov.) be synonymized with O. scapulalis (Walker, 1859). We further demonstrate that O. nubilalis auctt. pro parte feeding on mugwort, hop, and several other dicotyledons (previously called the 'O. nubilalis mugwort-race' in France) also belongs to O. scapulalis. Consequently, we propose that only O. nubilalis specimens feeding on maize (the former French 'O. nubilalis maize-race') belong to O. nubilalis (Hubner, 1796). The implications of this revision are discussed