8 ■ SCUDDER ON THE DEVONIAN 



form of the wing, which is presumed to be much as outlined on the plate. The wing 

 was probably more than GO mm. in length, and about 27 mm. in breadth ; the alar 

 expanse was therefore at least 125 mm., and probably 135 mm., and the two figures 

 have been so placed as to indicate this expanse. 



This is more than double the ordinary size of the larger Ephemeridae and the largest 

 mentioned in Eaton's paper on these insects has an expanse of only 78 mm., and the 

 largest of the Jurassic species only 65 mm. 



The costal margin is very gently arcuate ; the apex probably somewhat pointed, toward 

 which the upper veins are directed without additional arcuation ; the greatest breadth 

 was probably a little before the middle of the wing, and the outer perhaps half as long 

 again as the anal margin. The marginal vein runs close to but does not form the margin 

 of the wing, the latter being indicated in the figures on the plate by a dotted line. 



The mediastinal vein runs as close as possible to the margin, and is not connected with 

 it by cross veins ; these two veins apparently run side by side to the apex, when the 

 marginal disappears and the mediastinal takes its place close to the border. The scapular 

 vein runs sub-parallel to the mediastinal, but at double the distance from it apically as 

 basally, the change occurring rather abruptly near the middle of the preserved portion of 

 the wing ; it is connected with the vein above by straight cross-veins at tolerably regular, 

 rather frequent intervals. 



As usual in this family, the externomedian vein is apparently divided, probably not far 

 from, or at the base, into two stems, and the upper of these stems is again divided, prob- 

 ably at some distance from the base, into two principal branches ; the main portion of the 

 upper branch runs parallel to, but somewhat distant from the scapular vein, approaching 

 it, however, apically, and is everywhere connected with it by cross-veins, very much as in 

 the mediastino-scapular interspace; it throws off from its inferior surface several inequi- 

 distant feeble offshoots ; the first originate a little before the middle of the wing, and 

 run irregularly but with a gentle downward curve to the outer margin ; they have between 

 them and between the outermost and the main branch a number of equally irregular 

 intercalary nervules, all of which are connected together l)y cross-veins, and thus form 

 over the whole area a mesh work of irregular but usually hexagonal and longitudinally 

 elongated cells, making it impossible to distinguish between normal and intercalary veins, 

 since the latter are as prominent as the former, and invariably arise from cross-veins ; 

 while whatever nervules lie next the main branch are united with it by frequent and, 

 equally irregular cross-veins falling from the main branch quite in the manner of the off- 

 shoots proper, and forming cells only slightly larger than the others, although generally 

 transversely elongated ; together there are about nine rows of cells between the main 

 upper branch and its first offshoot. The lower branch of the upper stem is simple and, 

 originating apparently near the middle of the basal lialf of the wing, diverges at first 

 slightly fi-om the upper branch, afterwards a little more rapidly, and in its apical fourth 

 curves downward considerably, and is somewhat irregular in its course ; its direction is in 

 general parallel to the oflshoots, and especially the nearer offshoots of the upper branch, 

 and on the border it is separated from the apex of the upper branch by nearly one-third 

 of the outer margin of tlie wing ; in its smiplicity this branch resembles the same nervure 

 in Tricorythus, which is peculiar in this particular among modern Ephemeridae. As in 



