782 STUDIES IN AUSTRALIAN NKUROPTERA, vi., 



species, but may be very weakly indicated between Sc, R, and 

 Rs, and also in the position of the gradate series (these, though 

 shown in my figure of P^^otopsychopsis, are so faintly indicated, 

 that I have been doubtful how far they really existed). The 

 oldest forms of Prohemerohiidce were only of small to moderate 

 size, with only moderately numerous branches of Rs; specialisa- 

 tions in this family appear to have led to the development of 

 forms with an immense number of closely-set branches of Rs. 

 The Upper Jurassic forms are, on the whole, considerably larger 

 than the Liassic forms. 



As regards the Kalligrammatidcp, these gigantic insects differ 

 from the Prohemerobiidce only in the development of a close 

 series of connecting cross- veins all over the disc; in the beginning 

 of an anastomosis, by connection of Sc with R distally by means 

 of a short cross-vein, very much as in Psychopsis; and in the 

 development of the large eye-spots on the wings, recalling those 

 of Alegapsychops illidgei. 'J'he rise and fall of this wonderful 

 family was probably very quick; for. if their larvae, as is pro- 

 bable, were at all like those of Psychopsin, they could scarcely 

 have attained to dominance before the rise of the Birds must 

 have quickly exterminated them. 



Here let it be noted that Megapsychops illidgei shows, in the 

 possession of a media with more than two branches, a character 

 that was evidently possessed by the Kalligrainmatidce (the true 

 M of this family is labelled Cu by Handlirsch), and by certain 

 of the Liassic Prohemerohiidce, if not by all. Combining this 

 fact with another obvious one, viz., that the development of 

 numerous branches of Rs, occupying a very large area of the 

 wing, is a specialisation confined to the Planipennia, and must, 

 of necessity, have caused a squeezing out and consequent reduc- 

 tion of the media, we are bound to conclude that the ancestors 

 of the Planipennia approached more closely to the Panorpoid 

 type, as regards the structure of Rs and M, than they do to-day, 

 and that the many-branched media of Alegapsychops illidgei is an 

 archaic survival that stands to remind us of this fact. 



I regard the Psychopaidm as a Mesozoic remnant, considerably 

 isolated, at the present day, from all other existing families. Ii^i 



