742 



STUDIES IN AUSTRALIAN NEUROPTERA^ 1. 



We must also remark that the excessively short lower branch 

 of Cu in the venation, usually labelled Cu.., does not really repre- 

 sent the full limits of that vein, but is formed basally fi'om the 

 stem of Cuo, and distally from the first branch descending from 

 it. The rest of Cu.2 is represented by the irregular series of cross- 

 veins up to five or six cells' width distad from Cu,,. In Proto- 

 jjlectron, we can see a moi-e normal arrangement, Cuo running along 

 under, and parallel to Cu^ for a considerable distance. In this 

 respect, therefore, we must count Frotoplectron and allies as 

 amongst the most archaic of Myrineleonid(v. In Myrmdeon uni- 

 seriatus, the supplementary vein running under and parallel to 

 Cuj + Mg is a purely venational development, quite unrepresented 

 in the tracheation. It is clearly foi-med by alignment of the first 

 series of crossveins below Cu, + Mg, and may be called the cubito- 

 median suppleinent. 



Text-fig. 7. — Wing-tracheation from the right hind wing-sheath of a pupa 

 of 31. uniseriatus ( x 19), to show correct naming of tracheae. Com- 

 pare the hindwing in Text-fig. 3, and PI. Iviii., fig. 3. (Original). 



The venation of the hindwing in Myiineleonidoi presents no 

 difficulties at all, as may be seen by comparing Text-figs. 3 and 7. 

 Both M and Cu are unbranched, and run straight through the 

 wing. The waviness of M in the venation, and the very weak 

 fork of Cu, are slight venational specialisations which appear to 

 indicate a weak attempt at convergence towards the form of the 

 forewing venation, by a different route. The weak Cug which, 

 in the hindwing, joins lA distally, is indicated in the trachea- 

 tion only as a weak branch of Cu of a calibre barely equal to 



