In order to reappraise the interrelationships of genera in the tribe Troidini, and to test the resultant theory of troidine evolution against biogeographical data a cladistic analysis of troidine genera was performed. Data were obtained mainly from immature stages, providing characters that appeared to be more reliable than many "traditional" adult characters. A single cladogram hypothesising phylogenetic relationships of the troidine genera was generated. This differs markedly from cladograms obtained in previous studies that used only adult characters. However, the cladogram appears to fit well biogeographical data for the Troidini in terms of vicariance biogeography, especially as this relates to the general hypotheses of Gondwanaland fragmentation and continental drift events advanced by recent geological studies. The genus Ornithoptera is shown to be distinct from Troides. Based on input data drawn equally from immature stages and adult characters, a single cladogram hypothesising the likely phylogeny of Ornithoptera species was generated. With minor weighting of a single important adult character (male forewing upperside sex-brand), a further two cladograms were generated, one of which is similar to hypotheses proposed by previous workers. Based on these findings, and on ecological data, notably larval foodplant relations with Aristolochiaceae, as well as present-day biogeographical data, a new theory of the origin and evolution of Ornithoptera is presented which fits well Gondwanan vicariance events ascertained by geological studies: essentially that Ornithoptera evolved on northward drifting Australia, allopatrically from Troides on the Indian plate, and therefore that Ornithoptera did not reach the Australian subregion via Troides-like ancestors in Southeast Asia as has been previously postulated.
抄録全体を表示