414 TWO FOSSIL IKSECT WINGS IK OOLLEt'TIOX OF MR. JOllX MITCHELL, 



is presei've<l right in the middle of one such iroud, as may be seen Irom Plate 

 XXXV., fig. 2. Owing to the material liaving been burnt, the impressions, both 

 of the wings themselves and of the Glosaopteris fronds, are very hai'd and verj' 

 clearly marked. The embankment from which the,^■ were taken is at Merewether 

 Beach, about four miles south of Newcastle, N.S.W. 



These two wings were lent to me by Mr. Mitchell and have been in my 

 possession for about two years, during which time I have studied them fre- 

 quently, but found it difficult to determine with certainty their systematic posi- 

 tions. I should doubt if two more unique types of insect wing have ever before 

 been found. Moreover, the vei-j' simplicity of their venation adds to the difli- 

 culty of placing them. On the one hand, they stand far apart from all known 

 Carboniferous forms ; while, on the other, they show no close affinities with any 

 known Mesozoic, Tertiary or Recent forms. I have finally decided to place both 

 wings within the Order Hemiptera, of which early representative types have 

 already been discovei-ed in the Upper Permian of Newcastle and Belmont in 

 the shape of the two genera Permofidgor and I'ermoscarta. If my placing of 

 these two fossils is correct, then it still remains true that, of all the wings so 

 far discovered in these beds, there is not a single type tliat does not belong 

 either to the Hemiptera or to the Mecoptera. 



Wliile further discoveries in tliese beds should most certainly increase the 

 number of Orders represented in Australia at that period, one cannot but admit 

 that it already apj^eare evident that the dominant Insect Fauna of the Period 

 was a mixed one of Plant Hoppers and Scoi-pion-Hies. Sucli a combination can 

 only be found in Australia, at the present day, in the damp bottoms of shady 

 gullies, or along the shady southern slopes of steep hillsides, where dews fall 

 heavily and remain long undispersed by the sun's rays. In such localities, 

 Cockroaches should most certainly occur, if they liad reached Australia at the 

 time these insects were fossilized; so also should Megaloptera, and perhaps some 

 early forms of Planipennia. Coleoptera, if evolved by then, would most cer- 

 tainly have been present also ; the absence of this Order from the Upper Permian 

 of New South Wales is one of the most striking facts about this horizon, con- 

 sidering how abundant they are, and liow highly evolved, in the Up)ier Trijis of 

 Ipswich, Q. 



Before going on to describe the fossils themselves, I wish to thank Mr. 

 Mitchell for the opportunities he has given me for studying these and other 

 fine insect fossils in his collection, and also to draw attention to the magnificent 

 work which he has accomplislied, and is still accomplishing, in spite of his 

 advanced years, in the difTicult and painstaking exploration of the Belmont and 

 Newcastle Beds, — in which, as I know from personal experience, one has often 

 to work for days on end without getting a single wing as a reward. I have 

 also to thank Mr. W. C. Diivies, Curator of the Cawthron Institute, for tlie two 

 excellent photographic enlargements of the fossil wings, wliicli are reproduced 

 in Plate xxxv., figs. 1 and 2. 



Order HEMIPTERA. 



Suborder PalaiSohemiptera. 



Familv PROSBOLIDAE. 



Genns M l tcit e i. l o x k !• r .\, n.g. (Plate xx.w., fig. 1; Text-fig. 1.) 



Tlindwing: Shape moderately broad and rounded at apex; veins very 

 strongly l)uilt, but the wing membrane delicate. Sc short, fused with Ri, ending 

 up at ab<nif half-way along tlie costal margin. Ri very strongly formed, slightly 



