BY R. J. T1LLYAKD. 571 



base of the wing; and the two branches, Cuj and Cu' 2 , extend 

 nearly parallel to one another, without any further branching. 

 The anterior branch is, as always within the Complex, a strong, 

 convex vein, easily picked out as of almost equal calibre with 

 R, and much more strongly built than any other main vein. 

 The posterior branch, Cu2, on the other hand, is a very weak, 

 concave vein; in the more specialised t}'pes, it tends to sink 

 completely into the anal furrow, and may cease to become 

 ehitihised, either altogether or in part, as in many Lepidoptera. 



It is necessary to note here that Comstock admits this inter- 

 pretation of the limits of Cu for the Mecoptera. There is, of 

 course, no doubt about the matter at all in that Order. 



Text-fig. 4:6b shows the form of Cu in the Orders Paramecop- 

 tera, Protomecoptera, Megaloptera, Trichoptera and Lepidoptera. 

 One of these Orders, viz. the Megaloptera, has holotracheate 

 wings ; and hence we are able to examine the structure of the 

 cubital trachea in the pupa. This corresponds exactly with 

 the imaginal venation. Comstock himself figures the trachea- 

 tion of the pupal wing of Cluadiodes (15, fig. 170) with the 

 three branches, labelling them Cu 1; Cu la and Cu 2 . But, owing 

 to the fact that he treats the Megaloptera as a portion of the 

 Order Xeuroptera, the other part being the Planipennia, he fails 

 to give due prominence to this Order, and has missed the valu- 

 able evidence contained in its special type of wing-venation. 

 The form of Cu in the Permian Paramecoptera agrees exactly 

 with that of the Megaloptera, and so does that found in the 

 Triassic Protomecoptera, as far as it is preserved. Thus there 

 can be no doubt as to the antiquity of the extra branch of Cu', 

 which dates back to Permian times. As we have already seen 

 in Section iii., the forking of Cu x may have been originally 

 due to incomplete fusion of M s with Cu, distally. But, wdiether 

 that be so or not, the fact remains that the true primary cubital 

 fork (cuf) is not this fork at all, but the point at which Cu 

 originally forked, i.e. much nearer to the base of the wing. All 

 this is admitted by Comstock, as shown by his notation of the 

 wings of M' j <raloptera (15, fig's. 163-169), 



Let us now turn to the case of the Trichoptera and Lepi- 

 doptera. Text-fig. 47 shows the complete venations of the fore- 

 wings of the Megalopteron Austrosialis and the Trichopteron 

 Rhyacophila. The former wing is holotracheate, the latter 

 merotracheate : hence we cannot compare their tracheations. 



