290 NEW PERMIAN INSECTS FROM BELMONT, N.S.W., 



distal fusion of Sc with Ri, which, though apparently a specialisation, may be 

 due, as in the Peiiaria, to a partial fusion only of Sc2 with Ri, and may there- 

 fore be the original condition in the Planipennia, the wing before us is an abso- 

 lutely generalised Planipennian, with primitive terminal twigging of the veins, 

 primitive pectination of R2+3, an absolutely primitive condition of M, primitive 

 pectination of Cui, and even two characters which one could scarcely have sus- 

 pected ever to have been present within the Order, \dz. the closure of the radial 

 and median cells by special cross-veins. The wing also stands very close to the 

 Archetype of the Megaloptera, but the terminal twigging places it definitely 

 within the Planipennia, as also does the position of the piimary cubital fork 

 very close to the base of the wing. It is to be distinguished from the more 

 densely veined Corydalid types, of which Protohermes davidi Weele (op. cit., p. 

 696) is a good example, by the much smaller number of costal veinlets and inter- 

 radial cross-veins, as well as by the much more basal position of the primary 

 cubital fork. It is, however, possible to derive the whole of the Megaloptera as 

 well as the whole of the Planipennia from this wing-type, provided we assume 

 that the Megaloptera are an aquatic offshoot from the very base of the terrestrial 

 Planipennia, and that the Corydalidae are an older type than the Sialidae. These 

 assumptions are, however, scarcely justified, and it seems more logical to assume 

 that definite Megalopterous types were in existence in the Upper Permian, though 

 not necessarily in Australia, and that the present fossil is a true Planipennian, 

 from which the Mesozoic Prohemerobiidae, and consequently the whole Order as 

 we know it at ijresent, are easily derivable by further specialisations. The re- 

 lationship of the present-day Ithonidae of Australia to the fossil type is so 

 evident that I have selected a generic name for the fossil to indicate it as the 

 Permian ancestor of that family; but it is scarcely less easy to derive from it 

 such families a's the Psychopsidae, Berothidae and Hemerobiidae, not to mention 

 the Dilaridae, which do not occur in Australia. 



Peemithonb belmontensis, n.sp. (Plate xxxiii., fig. 3; Text-fig. 6.) 



The specimen is the obverse or east of a left forewing, showing Ri as a 

 strongly formed convex vein, Cui as a slightly less strong, similar vein. The 

 impression is on cherty shale stained with iron (ochreous), the wing itself being 

 of an ochreous colour, shading to fulvous along the distal portion of the posterior 

 margin, from half-way up to apex. The anal area is missing, and there is a 

 slight overfold of the membrane a little below the apex. This appears to have 

 been brought about by a tearing of the wing from near the end of Mi across R4+B, 

 followed by a slight buckling of the apical portion, so that the lower side of 

 the tear came to overlap the upper slightly. In Text-fig. 6, I have restored the 

 wing, adding the missing veins covered up by the overlapping, and turning the 

 apex to the right. Total length 9.4 mm.; greatest breadth 4 mm. 



It should be noted that the terminal furrows so characteristic of the Order 

 Planipennia, situated between the terminal twigs of the main veins, are clearly 

 to be seen in this fossil around and above the apex, as are also the swollen 

 bases of the tufts of hairs situated along the wing-margin between the twigs. 



Type, Specimen No. 52, in Mr. Mitchell's Collection. Label "New Insect 

 Wing, Belmont, N.S.W., Coll. Mitchell." 



Mr. Mitchell is heartily to be congratulated on this wonderful find, which 

 brings the record of the Lacewings right back from the Upper Triassic of Ipswich 

 to the Upper Permian. We may express the hope that other representatives of 

 the Order may yet be found at Belmont. 



