BY R. J. TILLYARD. 



249 



On the whole, then, we are unable to establish a direct con- 

 nection between Belmontia and the Triassic Paratrichoptera. 

 The latter may possibly be derived rather from the Protomeco- 

 ptera, which, though only so far known from the Ipswich Trias, 

 must certainly also have existed in the Permian. 



(iv.) Affinity with the Diptera. 

 Concerning the arculus in the Diptera, Comstock (i ; p. 358) has 

 remarked that "owing to the reduced condition of the trachea- 

 tion in this Order, the elements that enter into its formation 

 cannot be definitely deter- 

 mined." Nevertheless, it 

 seems to me that the con 

 dition of this part of the 

 wing in the Stratiomyiidw 

 (Text-fig.6), as well as in 

 some archaic genera of 

 other families, points de- 

 finitely to the conclusion 

 that it is formed on the 

 same plan as in the Tri- 

 choptera and Lepidoptera. 

 The close affinity of the Diptera with the Trichoptera in many 

 other respects is generally recognised, and there can be little 

 doubt also of the close general resemblance of the venational 

 schemes in the two Orders. It is, however, necessary to point 

 out that the usually accepted limits of Cu in this Order are 

 incorrect. The vein usually designated Cuj is undoubtedly M 4 ; 

 this can be proved by reference to the trichiation in the genus 

 Hhyphus, as I have already pointed out in a previous paper ( 5 

 p. 612, note), and by the fact that it actually comes off from the 

 median cell itself in some archaic genera of Tipulidce, such as the 

 Australian Gynopliatia. Consequently, the true Cuj is a strong, 

 usually straight, convex vein, without any distal forking, and 

 the part usually labelled Cu 2 is really the distal continuation 



Basal portion of wing of an undetermined Stratiomyiid fly, ( x 18). 

 Hornsby, X.S.W. ; recent. For lettering, see Text-fig.2, p.238. 



Text-fig.6.' 



20 



