( 277 ) XV. On the classification of the Australian Pyralidina. By E. Meyrick, B.A. [Read April 2nd, 1884,] The present instalment includes the Musotimidce, Boty-dida, and Scopariadce, with some additional species of the two families i^reviously treated. There remain the Hydrocamjnda and Pterophoridce, which will form the subject of a third paper ; and the Cjximbidce, Phycididce, and Galleriadce, which have been already described elsewhere. I think that this family subdivision will allow of the development of the group being properly understood. Its essential principle is, as I have explained before, that one family is not bound to be absolutely separated from another simply by the presence or absence of a single character, but by a majority of several distinguishing points ; which is the most that can be expected from a system really arranged on natural lines. It seems to me useless to attempt to judge of the value of characters for classification, without strict reference to the principles of evolution. I think it might be laid down as an axiom, that when an organ has wholly disappeared in a genus other genera which originate as offshoots from this genus cannot regain the organ, although they might develoj) a substitute for it. Thus in the Geometrina we have a number of genera in which the larvae have wholly lost three pairs of abdo-minal legs, the character being proved to be very per-sistent ; it must be held that no genera derived from any of these could recover the lost pairs of legs, and there-fore the Geometrid genera with 12-legged larvEe must be ancestral types, and not derivative ; and for the same reason the Geometrina must be regarded as a terminal development, a group which ends in itself and has given rise to no other groups. The character of the absence of ocelli may be applied in the same way. So, in the present group, the Botydidce are specially cha-racterised by the absence of the uncus in the male TEANS. ENT. SOC. LOND. 1884. — PART III. (OCT.)
BioStor
