Mb NEW PERMIAN INSECTS FBOM BELMONT, N.S.W., 



each case (Mi aud Ms) is again forked distally, while the lower (M2 and M4) 

 remains simple, so that there are altogether six branches of Mi-4 ending on the 

 wing-margin. The median cell (toc) is closed by a strong cross- vein, and both 

 forks are sessile upon it. The main stem of the cubito-median Y-vein, Mg+Cui, 

 is a strong convex vein resembling that of Belmontia, but ivithout any distal 

 forking; in its connections with M4 and Cu2, it strongly recalls the formation 

 of Cui in the Upper Triassic P'aratriehoptera, such as Aristopsyclie. Cu2 is 

 curved distally, as in Bslmontia, and connected with the straight, strong lA by 

 a single cross-vein, cu-a. 2A' is little more than half as long as lA, is bent dis- 

 tally as in Belmontia, and is connected with lA above the bend by a single cross- 

 "\-ein, ia. 3A is a short, much-curved vein, isolating, between itself and the border, 

 a small convex jugal area. 



Genotype, Parahelmontia perrmana, n.sp. 



Closely related to Belmontia, from which it differs chiefly in the unforked 

 Cui, the different number of branches of Rs and M, the somewhat different ar- 

 rangement of the cross- veins, and the different position of the forking of R4+5, 

 which, in Belmontia, is placed very close up to the primary forking of Rs. 



The origin of the Triehoptera and Lepidoptera from forms Resembling Bel- 

 montia has already been dealt with in a previous paper {op. cit. pp. 242-8). In. 

 the same work (pp. 248-250) I also discussed the affinities of; BeZ?7?.ow*ia with the 

 Paratrichoptera and Diptera, and concluded that there was no direct ancestral 

 connection between them. In the case of Parahelmontia, we - can say with cer- 

 tainty, owing to the formation of Cm, that this genus is not ancestral to the 

 Triehoptera and Lepidoptera; but, for the same reason, it comes into the direct 

 ancestral line of the Paratrichoptera and Diptera, in wliich the foiToation of Cui 

 is very closely similar. If we postulate the existence of other types, closely allied 

 to Parahelmontia, but with the short costal vein still not fully reduced to a vein- 

 let {hm), as seen, for instance, in the genus Aristopsyclie of the Paratrichoptera, 

 then we could say with certainty that, from such a type, this latter Order is de- 

 rivable simply by reduction of the number of branches of Rs and M to four 

 each; and it woiild follow that the Diptera were also to be derived from it by 

 further reduction. Or we may regard Aristopsyclie as an archaic side-branch, 

 which preserved the separate short costal vein, long after the rest of the group 

 had lost it, and may then derive those of the Paratrichoptera which have no 

 costal vein direct from forms like Parahelmontia, and the whole of the Diptera 

 from those more specialised Paratrichoptera themselves. 



F'aeabelmontia permiana, n.sp. (Plate xxxiii., fig. 2; Text-flg. 3.) 



An almost perfect forewing; total length 17 mm., greatest breadth 7 mm. 

 The impression is the reverse or mould of the wing, as is proved by the fact 

 that Ri, Cui and lA appear as deep furrows; as the apex lies to the right, the 

 impression must be the cast of a left forewing. The wing lies upon the smooth 

 surface of a grey cherty shale, the whole wing being pale ochreous in colour, and 

 stained fulvous along the costa and posterior border of clavus. There are slight 

 abrasions of the apex and tornus, and the course of the distal margin is very 

 faint in consequence; otherwise the venation is practically perfect. The system 

 of cross-veins, as will be seen from Text-flg. 3, is very like that of Belmontia. 

 The closure of the small cell on Rs may perhaps be considered a specific rather 

 than a generic character. 



Type, Specimen No. 54 in Mr. Mitchell's Collection. 



Horizon. — Upper Permian of Belmont, N.S.W. 



