222 STUDIES IN AUSTRALIAN NEUROPTERA, iii., 



unbranched media, a forked cubitus, and two or three short anal 

 veins. Right througli the family there is so little variation, 

 except in minor details, that the limits of these veins, having 

 been assigned without any sufficient reason to one species, soon 

 became applied equally readily to all. From one point oi view, 

 it may be argued that it did not matter that the determinations 

 of the venation wei-e wrong, since they were all conawfently wrong, 

 and so the numerous descriptions of new species all conform to a 

 single plan, and are easy to follow. 



Why is it necessary to disturb them] The answer is a very 

 obvious one. It is not only that false homologies are not to be 

 tolerated by anyone with the true scientific spirit, but also that 

 they .actually seldom work well in practice. It is only by 

 pure chance that some genus of Chrysopid(P has not turned up 

 which would not have fitted in with the accepted venational plan, 

 and would thus have started a series of evasions and explanations, 

 on whose heels confusion might have followed fast. More than 

 that, again, we can never hope to understand the phylogeny of 

 such a group as the Chrysopidfp, unless we really understand its 

 wing-structure, and its relationship to the general venational 

 plan in the Neuroptei'a Planipennia. 



The present paper was undertaken in the strong belief that 

 the accepted venational scheme for the Chrysojndw was wrong in 

 certain particulars. I was not prepared to admit (after my ex- 

 perience in the Myrmeleontidoi) that the media, in the forewing 

 at any rate, was an unbranched vein. I could not find any 

 satisfactory explanation of the persistence of the peculiar 

 "divisory veinlet" in the forewing (Text-fig. 4, dv), since it is 

 obviously not in a position of maximum effect as a support to any 

 main vein. It seemed to me that this veinlet must be some 

 remnant of the lost lower branch of the media, and its absence 

 in the hindwing further strengthened this belief. Finally, I 

 realised that the connection between the venational plans of the 

 Chrysopidce and Apochrysid(v was still entirely missing, and that 

 we could never hope to offer any adequate phylogeny of the 

 Chrysopidoe until this was cleared up. Such beliefs and sus- 



