BY R. J. TILLYARD. 673 



Great efforts were made to secure larvae and pupae of this 

 genus, but were frustrated by the severe drought of last summer. 

 My thanks are due to Mr. E.J. Dumigan for a large supply of 

 fresh imagines of a small (undetermined) species from South 

 Queensland, and to Mr. G. Lyell for named examples of several 

 species, including Western Australian forms. 



The pupal tracheation probably does not differ very much 

 from the venational scheme given in Text- fig. 94. The media 

 is complete in both wings; and, as in the Hepialidae, the point 

 mf is placed close to the base of the wing. The stalked ar- 

 rangement of the branches of Rs in the forewing is a specialisa- 

 tion peculiar to Synemon; the genera Castnia and Gazera have 

 the usual dichotomic arrangement, with the radial cell, or areole, 

 closed by the cross-vein ir, as figured by Turner' (32, p. 175) . 

 Important specialisations are to be noted in the following 

 points : — 



(1) M 5 ca^i just be made out as a very short vein at the 

 base of the forewing, but appears to have become completely 

 suppressed in the hind. 



(2) In both wings, Mj has moved a considerable distance 

 away from M 2) and has evidently been captured by Rs, through 

 the outgrowth of a special trachea along r-m. (Cf. Wingia, 

 p. 669) . The importance of this will become more evident when 

 we come to consider the next two groups. 



(3) In the hindwing, there is a small basal Y-vein, formed 

 by the fusion of 1A and 2 A, as in Carpocapsa and other Tor- 

 tncina. 



The peculiar form of the basal cell of the hindwing, which 

 remains incomplete and of very irregular shape, is very note- 

 worthy, and shows that this family is far more archaic than 

 the Tortricina and Tineina in this respect. 



Taking all these points into consideration, it must be clear 

 that the conclusion of Handlirsch, Turner and others, that this 

 family stands far apart from other Lepidoptera, and is only to 

 be derived as an isolated line of development from the original 

 ancestor of the Heteroneura, is fully justified. The venational 

 scheme shows no possible avenue by which the Butterflies could 

 have evolved from this family; and we can only conclude that 

 the adoption of the day-flying habit, and the correlated develop- 

 ment of clubbed antenna?, have been independently gained in 

 the two o T0U ps. 



