26 FOSSIL INSECTS OF THE BRITISH COAL MEASURES. 



backwards in its course to the wing-apex. The intercostal area is very wide 

 at the base, and diminishes in diameter towards the wing-apex, but it is doubtful 

 if it reached the latter. 



The subcosta is straight, but not well defined. The radius is fairly parallel 

 to the subcosta and gives off the radial sector low down in the base of the wing: 

 the radial sector comes off from the radius at an acute angle, and keeps parallel 

 with it as far as preserved. Its direction is such that it must have reached the 

 wing-apex. The median and cubital veins merit special consideration. In the 

 specimen three veins occupy the position of the normal median and cubitus. 

 Of these, two either arise from a common root, or so close together as to be 

 indistinguishable, the third having a separate origin. The innermost of the three 

 sends off from near its base a forwardly directed twig, which joins the second 

 of the two outer veins. 



It is necessary to resolve these veins into median and cubitus. To do this 

 we must consider recent research on living insects. Comstock and Xeedham 

 concluded that the primitive median vein was a four-branched structure; more 

 recently Tillyard, from further studies of the wing-venation of recent nymphs, 

 concludes that the primitive median had an initial dichotomy, of which the outer 

 branch divided up into four (" M " of Tillyard, ' Proc. Linn. Soc. N.S. Wales,' 

 vol. xliv, 1919, p. 552), and the second branch remained simple. Comstock and 

 Needham concluded that the primitive cubitus was two-branched; Tillyard considers 

 it (loc. fit., p. 553) as three-branched, the outer or first branch having a distal 

 forking into two feeble twigs, and an inner feeble branch which remains simple. 

 The primary cubital fork is situated near the base of the wing. 



Tillyard's views are not very different from those of Comstock and Xeedham, 

 as he admits a basal fork of the cubitus, but adds a secondary forking of the end of 

 the outer branch. 



Tillyard has studied the relation of the two veins in the various Orders, and 

 finds that the " posterior arculus " of Comstock, which is supposed to be a cross- 

 vein from the median to the cubitus, is, as shown in the fossil Order Paramecop- 

 tera, not a cross-vein, but a true branch of the median. The various stages by 

 which the connection becomes established and afterwards developed into a 

 combined vein from the point of union, are fully stated (loc. cit., p. 357), and the 

 compound vein is named the " cubito-median Y-vt'im" An examination of the 

 conditions observable in Dictyoneura higginsii shows that the united veins and 

 their single-stemmed prolongation are in position and character identical with 

 Tillyard's " cubito-median Y-^ein-" The three veins of the specimen, therefore, 

 are the outer median vein, the inner median uniting with the first cubital vein, to 

 form a " cubito-median Y-vein," and lastly, the second (inner) branch of the cubitus. 



Tillyard recognises the presence of the " cubito-median Y-vein " in the Permian 

 Order Paramecoptera, but its discovery in the Coal Measure genus Dictyoneuru 



