182 B- H. SCUDDER ON THE WINGS OF SOME FOSSIL NEUROPTERA 



The lower branch forks and reforks again several times. This vein, with the two previous, 

 occupy just about one half of the wing. The v. cxterno-media can with difficulty be distin- 

 guished from the v. scapularis next the base, since it runs for a short distance in very close 

 contiguity to it, after which it diverges and runs sub-parallel to the lowermost branches 

 of the previous veins, forking about the middle of its course, each of the forks dividing once 

 more. The v. interno-media, taking its origin at a distance from the v. externo-mcdia, curves 

 directly up into close proximity to that, then diverging runs nearly parallel to it, forking 

 once at no great distance from the base (but in Bittacus at the base), thereafter remaining 

 simple. The v. analis, except where it is atrophied, as in Bittacus, forks at the base. The 

 upper branch, curving like the v. interno-media and running parallel to it, remains simple. 

 The lower branch forks again immediately, the lower fork (forming the margin) dividing 

 once more, but with these exceptions remaining simple. All the areas, and the spaces 

 between the branches, but especially those beyond the middle of the wing, have conspicu- 

 ous but very infrequent cross-veins. There are none in the area extcmo-media previous to 

 the forking; of the v. externo-mcdia. 



o 



Phryganina. 



The v. marginalis is continuous. The v. mediastina is straight and runs into the margin in 

 the apical half of the wing. It is connected with the v. marginalis close to the base by a 

 strong cross-vein. The v. scapularis branches very near the base. The upper branch sim- 

 ple, and running parallel to the v. mediastina, reaches the margin before the apex. The 

 lower branch branches again before the middle of the wing, the branches connected half 

 way to the outer border by a bent cross-vein, which sends out a branch parallel to and 

 midway between these. The v. externo-mcdia bends up towards the v. scapidaris, then diverg- 

 ing from it forks not far from the origin, the branches united at half their distance to the 

 border by a cross-vein, at which point they both fork, the upper fork of the upper branch 

 again and again forking, each time approaching the lowermost branch of the v. scapidaris, 

 with which the last is joined by a cross-vein, which is continuous with the forks of this 

 vein and the cross-vein of the v. scapularis. The v. interno-media is single, and runs parallel 

 to the lower branch of the v. extcmo-media. It terminates at the border or at a cross-vein 

 which unites the lower branch of the v. externo-mcdia with the border. The v. analis forks at 

 the base, the upper branch running parallel to the v. interno-media, and united near the base 

 to the other branches, one of which forms the internal margin and the other is short and 

 unimportant. 



Since form dependent on general structure is a characteristic of families, as Agassiz has 

 well presented it, we might have properly anticipated what we now find from this review, 

 viz. : that distinctions of a general nature in the neuration of the wings correspond with 

 the family divisions ; for it is upon the structure of the wings of insects that their form 

 very much depends. Especially is this true in Neuroptera, where, as much as in any other 

 group except Lepidoptera, the form is presented most obviously in the contour of the 

 wings. No systematists, however, have used these characters in the Neuroptera to any 

 extent so far as I am aware, except Burmeister, who has treated of them in a general way, 

 but scarcely so as to allow of ready comparison between the families. Heer, also, in his 

 work on the Fossil Insects of (Eningen, has given in detail — and differing scarcely at all 

 from what I have here presented — the mode of this venation in the Termitina. Besides 



