6Y R. J. TILLYAKt). 685 



seventh from the basal to the distal side, is very clearly \ isil)le 

 in whatever direction the light may be arranged, and is evidently 

 much thicker than the others. Further, it is very distinctly 

 concave. It would seem, therefore, very likely that this vein 

 is really part of Cuj, so that the forking of the cubitus takes 

 place at the posterior basal angle of the triangle (the point d in 

 the figure). If this be so, then the continuation of the proximal 

 side must be C.u^; while Cuj, after leaving d, crosses to meet the 

 long distal side at the point e, where it turns sharply downwards 

 to continue the line of that side posteriad. It must be noted, 

 howevei-, that this continuation remains highly convex. In all 

 recent Anisoptera, Cui remains concave throughout its length, 

 whereas Cn^ becomes convex after leaving the triangle. That 

 the convex vein does not continue to the wing-border in our 

 fossil, is quite evident, since, on the other side of the fracture, 

 one finds a strongly concave vein running to the border, just in 

 the usual position for Cu,. We have, therefore, two alterna- 

 tives : either («) if the cubital fork is at d, then the local con- 

 vexity of Cu, just below the triangle must be attributed to the 

 influence of the strong distal side of the triangle dominating it 

 for some distance, though it soon sinks into the usual concavity; 

 or (6) the vein de may be only a strengthened cross-vein, and the 

 distal side may continue below it to meet the proximal side at 

 some point just within the obliterated area, so that the triangle 

 is really triangular, and not quadrangular; if so, then the cubital 

 fork is at this latter point, and the two branches of Cu may very 

 well possess their usual characters from the very start. A study 

 of Handlirsch's figure of this part of the hind wing of ^EschnidiuDi 

 deusum Hagen, (2, Atlas, Plate xlvii., fig. 16) fails to help us here, 

 though it is quite likely that an examination of the actual fossil 

 might give us the necessary clue. Thus the actual shape of the 

 triangle, and the position of the cubital fork in our fossil, must 

 remain doubtful. It is, indeed, just possible that the triangle 

 may remain open posteriorly, as the subtriangle appears to do 

 also in this remarkable wing-form. 



The suhtriaugle is a wide, irregularly quadrangular area, 



