bV R. i. TILLYAKD. 271 



since the two new genera forming it difier most strikingly from 

 all known Neuroptera. 



The forms studied included also the beautiful genus Psi/chopsis, 

 usually placed in the Hemerobiidce, but already recognised by N. 

 Banks and Handlirsch as entitled to at least subfamily rank. 

 The discovery of the complete life-history of one species of this 

 genus, and the opportunity of examining the habits of the living 

 larvae, pupje, and imagines, has so sti'engthened the conviction 

 (which I had already gained from a study of the venation), that 

 this group has nothing whatever to do with the Hemerobiidce, 

 that I have cut it out of this paper, preferring to deal with it as 

 a separate family, PsycJiopsidce, in a monograph to follow later. 



The only true allies of the Hemerobiidce, as restricted by me, 

 are the Ithonidce and Dilaridce, the latter not found in Australia. 

 These three families might well be placed together as constituting 

 the Sub-Order Hemerobioptera, constituting the only remains 

 of a single phyletic line of descent, defined by the exceedingly 

 ancient character of the possession of more than one radial 

 sector in the forewing. This character, though it may have 

 occurred more frequently in the past, in groups now extinct (as, 

 for example, in the Protodonata), appears to be quite lost in 

 other recent Insecta, a reduction to a single radial sector being 

 the almost universal rule. Its persistence in the Hemerobioptei'a 

 is correlated with the retention of an ancient wing form and 

 venational scheme. Narrowing or lengthening of the wing 

 would require the elimination of the extra sectors; but the 

 Hemerobioptera on the whole, though undergoing, through the 

 course of ages, extreme reduction in size, have retained a very 

 uniform and unspecialised venational pattern. 



Distinguished from these by the possession of a single radial 

 sector in the forewing, the whole of the rest of the Order Neu- 

 roptera stand out as an Osmyloid stock, and might fittingly form 

 a Sub-Order Osmyloptera. These insects, though probably not 

 in the main aquatic in their life-histories, have been continually 

 throwing off aquatic or semi-aquatic remnants, while the great 

 mass of forms progressed rapidly onwards along the more suc- 

 cessful lines ofFei'ed by the rapacious, terrestrial, carnivorous. 



