BY R. .1. TILLYARD. 431 



wrote this, I am not sure that lie knew of the great difference 

 in size between the two sexes in this family. It is, therefore, 

 possible that this specimen is really a large female of /. fnsca, 

 I should estimate that the expanse of the average female of 

 H. megacerca would be about 70 mm., while the difference in the 

 shape of the wings, especially the greater breadth towards the 

 apex of the forewing, ought to be even more strongly marked 

 than in the male. 



Type: Holotype £ in Coll. Tillyard, taken at light at Hornsby, 

 N.S.W., in November, 1913. A second male, slightly smaller 

 than the type (length, 24 mm.: expanse, 52"5mm.) is also in my 

 collection; it was taken at light at Hornsby in Nov., 1909. 



Genus Varnia Walker. (Text-fig. 4). 



Walker, Trans. Ent. Soc. London, I860, v., p. 197. (>8up 

 pressed by McLachlan, 1870, Ent. Mo. Mag., 1869-70, p. 26 :- 

 "Genus Va/rnia Walker = Ithone Newman. V. perloides is an 

 interesting second species of this curious genus." No reason at 

 all is offered for the suppression). Nespra Navas, Rev. Real 

 A.ad., Madrid, 1914, p.478. 



Characters as given in the key on p.423. 



Gen o t y p e, Varnia perloides Walker (Western Australia). 



Habitat : Sandy places in Western and Central Australia. 



The only specimens known of this genus, apparently, are the 

 type 9 of I'. perloides Walker, from Western Australia, and the 

 type of Nespra implexa Navas, from Central Australia, both in 

 the British Museum Collection. I have not seen these insects, 

 but it is quite clear, from Mr. Campion's study of them, that 

 they must be regarded as congeneric, seeing that they agree in 

 the three important characters of having the veins of the wings 

 marked with fuscous or black, the possession of more than one 

 apparent radial sector in the forewing, and the costal veinlets in 

 that wing much forked and connected together by means of 

 cross-veinlets. The only difference is that of size, which we now 

 know to be a characteristic difference between the sexes in this 

 family. I therefore have no hesitation, firstly in restoring the 



