! AFRICAN SPECIES OF THE GENUS BACTRA STEPHENS (LEPIDOPTERA, TORTRICIDAE) BY A. DIAKONOFF Rijksmuseum van Natuurlijke Historie, Leiden After having published surveys of the species of the genus Bactra Stephens from tropical Asia (1950, 1956), and from the Palaearctic and Mediterranean Regions (1956, 1959 and 1962), I now present a preliminary survey of the species from the Aethiopian Region, chiefly South Africa. Does the genus show a rich speciation in tropical Asia, in South Africa this abundance is almost exceeded. At first I had the impression that India might represent the country of origin, a zoocentre, of Bactra species. The present results, however, suggest that we may be dealing with a very old group which may have originated in some ancient central region of the Old World Tropics, somewhere between the Asiatic and African continents, to spread east and west. This might explain the partly very puzzling distribution of certain species groups, e.g., the boschmai group: in New Guinea, Africa and the West Indies (!), the coronata group: in Australia, South Asia, and South Africa, etc. The species of Bactra can easily be grouped in five natural subgenera, four of which have been defined by me previously (1956), while the fifth is described below. The general character of these subgenera, their taxonomie "behaviour", is not at all similar. So the species of the most extensive subgenus, Chìloìdes Butler, possess strongly specialized genitalia in the two sexes and show marked differences from one another. The group offers the greatest diversity of genital structure which makes its study very fascinating. Chìloìdes contains many species occurring throughout the tropical and subtropical regions of the Old and the New World. These features suggest a great antiquity of the group. On the other hand, the two following subgenera, Bactra and Nannobactra, are different from Chìloìdes, but are rather similar to each other with regard to this taxonomie behaviour, viz., within each subgenus the species show but minor dif-ferences of genitalia; especially the females are rather uniform in each of these two subgenera. While those of Bactra Stephens may mostly be discriminated without difficulty, the females of Nannobactra Diakonoff often are so similar that their separation becomes a hazardous or even an impossible task. These two subgenera must be much younger than Chìloìdes, although of unequal origin: while Bactra must have developed from some simplified Chìloìdes-ìike ancestors through the loss of the valvula, Nannobactra may be regarded as a quite different off -shoot of the Chìloìdes stock, having originated through the speciation of the cucullus portion of the valva and of its armatures, but with the retention of the valvula. 285
BioStor
