46 STUDIES m AUSTRALIAN KEUROPTERA, ii., 



this character is based essentially on the archaic position of Rs, 

 viz., originating from R close to the wing-base, it seems to me an 

 excellent character for tlie main subdivision of the family, and 

 marks off the large complex of forms included in the Deiidroleon- 

 tiiKH (to which the great majority of our Australian species 

 belong) as definitely more archaic than the My rmeleontince, in 

 which the radial sector of the hindwing has progressed much 

 further along R. 



In subdividing the mass of forms in the Dendrn/eontinrc, it 

 seems to me that too much stress has been laid on the presence 

 or absence of tibial spines. In the dichotomous tallies so far 

 published by Banks and Petersen, this character is used for 

 effecting the main dichotomy, and only later are venational 

 characters brought into play. Now, I would urge an alteration 

 of this procedure for two reasons; firstly, because it seems pro- 

 bable that some forms, at least, that lack spurs (if not all) did 

 originally possess them,* and are really closely related to those 

 forms (e.g., Glenoleon) which possess very short spurs, hut which, 

 on the present method of dichotomy, are driven right to the 

 other end of the table; and secondly, because the wing-venation, 

 particularly of the forewing, offers us more important and better 

 understood characters, and a far better chance of making natural 

 divisions. I would propose that, in no ca.se, sh(juld the absence, 

 or (if present) the size of the spurs be used for divisions greater 

 than of generic value, since we cannot ignore the strong pro- 

 bability of convergent reductions of these organs in widely 

 separated groups. 



Turning, then, to the venation of the forewing, we can select, 

 without hesitation, as a natural group, those peculiar genera in 

 which Cuj runs parallel with Cuj -I- h\., for a considerable distance. 

 This is clearly an archaic character, carried over without change 

 from Nyniphid-sLUcestors. These genera form a distinct tribe, 

 which I propose to name I'rotoplectrini. This tribe is represented 



* The presence of these spurs almost universally in the Neuroptera, 

 Triflioptera. Lepidoptera. and Meeopteia is a fact that cannot lie ignored 

 in our attempts at classification. 



