BY R. J. TILLYARD. 583 



jority of the genera of this group showing- no trace of it in 

 the imaginal venation. 



This same Y-vein formation may be found at the base of 

 the hindwing in the Castniidae (Text-fig. 94) and the Thyrididae 

 (Text-fig. 98). It is absent, however, from the higher groups, 

 as, for instance, in the Psychina, Pyralididae and the Butterflies 

 (Text-figs. 97, 100, 102). An examination of the pupal trachea- 

 tion in this last group shows that, as in the ease of the Tor- 

 trieina, it must have been the trachea 1A which became aborted. 

 For the only trachea left of the two is placed well apart from 

 Cn 2) and slmws no tendency to approach it near the base. Thus 

 we reach the unexpected result that, in the Butterflies, and 

 probably also in the other highly specialised groups, the re- 

 duction of the number of anal veins has not been by loss of 3A, 

 as hitherto supposed, but by loss of 1A. It should be noted also 

 that, in all these higher groups, although trachea Cu 2 persists 

 in the pupal wing, no chitinisation is formed around it; and 

 so vein Cu 2 is also absent. Thus the only veins lying below 

 the basal cell in the hindwing are 2 A and 3A. 



We see, then, that the rise and fall of the anal Y-vein of 

 the hindwing is a line of evolution entirely confined to the 

 Heteroneura. Its origin is most certainly to be traced back t<> 

 a form in which the approach of 1A to 2A took place at the 

 point where these two veins were joined by the cross-vein iai, 

 as m the Trichoptera and Jugo-frenata (Text-tig. 54a). This 

 cross-vein must have gradually become shortened, and finally 

 aborted, leaving the two veins in contact. The wdiole series 

 of evolutionary forms is indicated in Text-fig. 54. 



In concluding this account of the anal group of veins, I should 

 like to call attention to the fact that it appears to be only an 

 assumption on ('< linstock's part that the three so-called anal veins 

 are really three separate veins, each of them equivalent to one 

 of the other more anterior main veins. In the Odonata, there 

 is only one anal vein, which gives off a number of branches, 

 all of which are convex. In the Orthoptera and Perlaria, 2A 

 and 3A come off from the same stem, though 1A is slightly 

 separated from them. In the Cicada, of which I have examined 

 the tracheation in the wings of several freshly-turned nymphs 

 in situ, the anal group of veins comes off from a common stalk 

 in the forewing. as in Odonata, and in the hindwing the con- 



