BY R. J. TILLYARD. 199 



dichotomous branchings are quite different from those of any 

 known form. 



Panorpoid Insect incertjE Sedis. 



Specimen No. 1146 is a badly-preserved wing which appears 

 to belong to one of the Panorpoid Orders, but which cannot be 

 determined with certainty. Parts of Sc, R, M, and Cuj are 

 visible. R 3 is forked distal ly, Rs dichotomously forked many 

 times over, M apparently five branched, and Ciij straight and 

 unforked. Length of fragment, 95 mm.; greatest breadth, 3*5 

 mm. The costal space was moderately wide, and shows portions 

 of an archedictyon; there are also some slight signs of the same 

 kind of meshwork in other parts of the wing. 



This insect may perhaps be related to the Protomecoptera; but 

 is so poorly preserved that I do not propose to name it. 



Order PARATRICHOPTERA, ordo nova. 



Triassic insects with wings in which the venation is on the 

 Trichopterous plan, but differing from the true Trichoptera in 

 two very important points, as follows : — 



(1 ) The three anal veins remain quite separate in the fore wing. 

 (Tn all true Trichoptera these three veins are looped up together 

 in a very typical manner). 



(2) Ci^ is a strong, straight, convex vein, without any apical 

 fork. (In all archaic genera of the true Trichoptera, Ciij is 

 forked apically). 



A separate costal vein may be present. Rs and M both four- 

 branched, as in archaic genera of recent Trichoptera; both the 

 radial and median cells closed 



The discovery of the almost complete and remarkably well- 

 preserved wing of Aristopsyche, n.g., shows that I was in error, 

 in No. 1 of this 8eries(3), in restoring the forewings of Mesopsyche 

 and Triassopsyche on the typical Trichopterous plan, as regards 

 the looping-up of the anal veins. I may also have been wrong 

 in determining the presence of the wing-spot (a difficult thing to 

 be certain of, in any fossil) since neither of the two new genera 

 here dealt with shows any signs of it. 



