1)62 THE PANORPOID COMPLEX, iii-, 



course, very materially from that of the Hepialidae. The essen- 

 tial points to be noted are the following: — 



(1) R x gradually approaches the stronger trachea Sc, and 

 fuses with it distally. The result, in the imago, is the forma- 

 tion of a single strong vein Sc+Rx distally, continuing the line 

 of Sc basally; while the basal part of R x itself appears as an 

 oblique vein considerably shorter in length than its correspond- 

 ing trachea in the pupa. Thus "vein 8" of systematic Lepi- 

 dopterists is proved to be a composite vein, Sc+Rj*; and I was 

 in error in my previous paper (13) in calling this B 1 . It 

 should be noted that, while this formation is to be found in the 

 hindwings of all Heteroneura, nevertheless the share taken in 

 the formation of the vein Sc+R x by its two components differs 

 very greatly in different families ; see, for instance, Wingia 

 (Text-fig. 90), in which R x is dominant, and Euschemon (Text- 

 fig. 99), in which Sc is dominant. 



(2) Rs is a single unbranched trachea. This is the condition 

 to be found in the early pupae of all the older Heteroneurous 

 types. Later on in pupal life, it frequently happens, as in 

 Xyleutes itself (Text-fig. 87, b), that a trachea of small calibre 

 grows out from Rs towards M. lf traversing the cross-vein r-m. 

 This is correlated with the outgrowth of other tracheae along the 

 cross-vein im and the vein M 4) so that the zig-zag veins that 

 are destined to form the distal termination of the basal cell 

 (called collectively the disco cellular s) become well supplied with 

 tracheae, by invasion of the region of M both above from Rs and 

 below from Cu la . 



(3) 1A loops up strongly towards Cu 2 , but does not quite 

 succeed in fusing with it. The condition is, however, very 

 similar to that seen in the Homoneura, and the failure to com- 

 plete the fusion may well be due to the very evident increased 

 breadth of the anal area in this wing. In the narrower wing of 

 Wingia ( Text-fig . 90), the fusion is completed. 



(4) After diverging from Cu 2 , 1A approaches 2 A towards 

 the middle of its length, and fuses with it for a short distance, 

 afterwards becoming of weak calibre, diverging slightly away 

 from 2A, and failing to reach the wing-border. The result of 

 this, in the imaginal venation (Text-fig. 88) is a small basal 

 Y-vein formation, not easy to make out in Xyleutes, but evi- 



*This specialisation was pointed out by Comstock and Needham as 

 pong ago as 1899 (14). 



