( 119 ) V. On the classification of some families of the Tineina. By E. Meyrick, B.A. [Read December 6th, 1882.] Contrary to my preconceived opinion I have satisfied myself that the family classification of the Lepidoptera, as at present existing, is in the main wholly unsatis-factory. Based as it is entirely upon European types, its deficiencies only become readily apparent when it is attempted to apply the scheme to the fauna of a distinct region. One discovers then how vague and ill-defined the principal families are ; the newly-added species destroy all remaining semblance of distinction, and the conclusion is inevitable, that without the aid of a system founded on well-marked and definite characters, the whole science will speedily become involved in hopeless confusion. Two causes have principally contributed to this result, over and above the narrowness of the field of research, viz., over-reliance on superficial characters, and under-estimation of the value of neuration. As a striking instance of the former, take the group of the Tortricina, the classification of which, as at present understood in England, is scientifically quite worthless. The group presents a remarkably small range of variation in the superficial characters of form, marking, and colour, and consequently the same general appearance frequently recurs in distinct genera, and especially frequently in allied genera. Structure has consequently been dis-regarded in the attempt to bring together discordant species, and has even come to be more or less despised as an untrustworthy indication of affinity. Yet the group, which has been considered one of great difficulty (as indeed any group is, if classified on such principles), is really one of the easiest to arrange on a natural system. I have elsewhere (Proc. Linn. Soc. of New South Wales, 1881) given in full my views on the classi-fication of this group, to which I need not further allude TRANS. ENT. SOC. 1883. PART II. (JUNE.)
BioStor
