660 



THE PANORPOID COMPLEX, 111., 



which these costal veinlets might be restored in their entirety, 

 supposing that some other change, such as loss of scales in a 

 clear-winged form, might make such a restoration of value to 

 the species? 



The forewings of Xyleutes and the Hepialidae also agree in 

 the absence of the cross-vein ir and the presence of im. But 

 they differ in the details of the forking of Rs. In Hepialidae, 

 the original condition seen in Belmontia (Text-fig. 63) and all 

 archaic Trichoptera is retained, viz., that R 4+5 forks close to 

 rf, R2+3 much more distad from it. In Xyleutes, the latter 

 trachea forks at a level slightly nearer rf' than does the former; 

 and this difference is greatly increased in the imaginal venation 

 (Text-fig. 86). In the pupal wing of Xyleutes, rf lies only 

 just distad from rf; but Rs remains separate from R 1; though 

 very close to it. In the imaginal wing, a slight fusion of the 

 basal part of Rs with Rj takes place, resulting in the appear- 

 ance of R 2+ 3 and R 4+5 arising as slightly separated from one 

 another upon R lt This formation is of great interest, since it 

 is a very clear illustration of the manner in which two apparent 

 radial sectors can be evolved from the original single one. 

 Similar formations may be seen in many families of Lepi- 

 doptera ; ,and, in some cases, notably the Butterflies, the change 

 has extended to the pupal tracheation, and has become a fixed 

 character there . Carried to its fullest extreme, the same charac- 

 ter may be seen in the pupal and imaginal wings of the family 

 Hemerobiidae (Order Planipennia) . 

 In the pupal forewing of Xyleutes, R2 + 3 an d R 4+ 5 remain 



1A. Cu z 



Text-Fig. 86. 

 Venation of forewing of Xyleutes eucalypti (Scott), imago (xO-8). 

 Lettering as on p. 535. 



