316 MR. H. BOLTON ON INSECT-EEMAINS FROM [Sept. I9I2, 



ends upon the margin in four rami, three of which arise from the 

 inner of the two members of the first division. 



The branches of the median and of the radial sector occupy all 

 the distal half of the inner wing-margin. 



The cubital vein is strongly arcuate, dividing near the base into 

 two branches, the outer of which forks twice and the inner once. 

 The cubital vein therefore ends upon the inner margin in five rami. 



Only a portion of the anal area is preserved, but traces of five 

 anal veins can be distinguished. The first anal is a strong vein 

 which forks at the middle of its length. The remainder are simple. 



The branches of the cubital and anal veins occupy the proximal 

 half of the inner margin. The wing is somewhat plicated, the veins 

 lying in shallow angular troughs. No definite trace of intercalary 

 venation can be made out, owing to the existence of a complex 

 series of irregular wrinkles, some of which stretch across several 

 of the marginal veins, while others pass between them. This 

 wrinkled area is restricted to the distal half of the inner margin 

 of the wing. 



By projecting the terminal branches of the veins towards the base 

 of the wing, some attempt has been made to reconstruct the whole. 

 The wing was undoubtedly very broad in comparison with its length, 

 and, notwithstanding its short broad outline, was very delicate. 

 Both inner and outer margins were convex, and both bent round at 

 the tip of the wing to meet almost at an angle, at the point where 

 the outermost ramus of the radial sector reached the margin. 



Affinities. — The wing is typically Palaeodictyopterous in 

 character, and agrees remarkably closely with Dr. Handlirsch's type 

 figure.^ It differs from that form in the greater division of the 

 cubitus, which ends in five rami instead of three. The greatest 

 depth of the wing was also, in all probability, nearer the base than 

 in his species. 



With the genus Atliyinodictya " the relationship is even closer. 

 As in that genus, the costal and subcostal are close together, the 

 radial sector arises low down and is divergent from the radius, 

 the median is a simple vein ending in three branches, while the 

 cubitus is almost identical in its divisions, the difference being that 

 the first forking arises at a higher point than that in Atliymodictya 

 parva, and that the inner simple ramus of the outer branch comes 

 off a little below the middle of the wing. The anal veins number 

 four in A. parua, as against five in the form under consideration. 

 One point to which I attach some importance is that the outer 

 margin in both is almost straight, bending sharply round distally 

 to meet the more convex inner margin in the neighbourhood of the 

 outermost ramus of the radial sector. In both cases the wings are 

 comparatively small. 



^ ' Die Bedeutung der Fossilen Insekten fur die Geologie ' Mitt. Gaol. 

 Gesellsch. Wien, vol. iii (1910) p. 505, fig. 1. 



^ A. Handlirseh, ' INew Palasozoic Insects from the Vicinity of Mazon Creek, 

 IllinoiB ' Am. Journ. Sci. ser. 4, vol. xxxi (1911) p. 298. 



