BY R. J. TILLYARD. 559 



remains concave, while M 3+4 becomes slightly convex. Mean- 

 while, Cuj remains convex for its whole length, and thus we 

 have two adjacent main veins, M 3 , 4 and Cu'j both of them 

 convex. This points to the fact that a concave vein must have 

 been suppressed between them. If the upper arm of the Y- 

 vein be truly M 5) then this interpretation fits in exactly with 

 the present convexity and concavity of the veins; for M 5 is a 

 concave vein, and must be the basal portion of the concave vein 

 missing further distally between M 3+4 and Cuj. 



Thus we see that, on all five counts, the upper arm of the 

 Y-vein is shown to be M 5) and not a true cross-vein. 



It is impossible to decide whether the distal forking of Cui 

 is due to incomplete fusion of this vein with M 5 or not. The 

 latter being a concave vein, it may well be that it had already 

 become too weak distally, before the basal fusion was accom- 

 plished, to be able to take any part in the more distal forma- 

 tion. An examination of the distal forking of Cuj shows that 

 both branches are more or less convex; this evidence, then, so 

 far as.it goes, would indicate that both belonged originally to 

 Cuj. I propose, therefore, to continue the notation Cu la and 

 Cu lb for these two branches, while leaving it quite open for 

 anybody to consider the upper branch as the distal end of M' 5 . 

 not completely fused with Cu v if he prefers that interpreta- 

 tion. That M 5 took some part in the formation of the main 

 stem of the Y-vein there can be no doubt whatever, seeing how 

 strongly formed this vein is at the fork of the Y. Conse- 

 quently the correct notation for this vein is M 5 -|-Cui in this 

 fossil. If this be borne in mind, there does not seem to be 

 any reason why the notation Cuj for the corresponding vein in 

 recent Orders should not be maintained, seeing that the in- 

 fluence of M 5 in its formation has undoubtedly gradually de- 

 creased ; until, at the present day, the only trachea supplying 

 it in most pupal wings is that of Cuj itself. 



We may now sum up the above evidence as follows : — 



In the Permian fossil Order Paramecoptera, the cubito-median 

 Y-vein is strongly formed, and consists of an upper arm, M 5 

 a lower arm, Cu 1? and a distal main stem Mg-j-Ciij, which does 

 not extend to the wing-border, but forks again distally into 

 Cu la and Cu lb . 



The further discussion of the evolution of the distal forking 

 of Cuj will be found in Section iv., dealing with the cubitus. 



