720 



PROCEEDINGS OF TFIE NATIONAL MUSEUM. 



Differentiation between fore and hind wings ha.s been most com 

 pletely carried out in the Libelkdidae/' and, fortunatel}^, almost all thi 

 steps are still to ])e seen in living genera. In this family two ver' 

 different })roce.5ses have operated in the two wings: 



A. TJui precession <>f ihe trlan(jle in the f(n'<> irimj. — In all hut th 

 more generalized Li])elhdida^ the anal vein seems to run directly t 

 the antero-internal angle of the triangle, while retaining in the hini 

 wing its usual course to the hind angle. This is a puzzle; but the ke 

 to the puzzle is found in the behavior of the second cubito-anal cros.' 

 vein and the anal vein. The accompanving diagram (tig. 15) epitcl 

 mizes what has taken place. 



The anal vein formerly went directly to the hind angle of th 

 triangle in the fore wing. Then, through the simultaneous dellectio 

 of crossvein and anal V(Mn, the latter came to appear forked — a stagi 



it has reached and still mairi 

 tains in most Gomphina 

 Then the l)asal part of th 

 anal trunk was carried foi 

 ward into direct line wit 

 the still further deflecte 

 second cubito-anal cros.' 

 veins, so that the latter ap 

 pears as the continuation c 

 the vein, and not at all as 

 crossvein, while the distr 

 end of the anal vein is almo? 

 lost among the crossveins o 

 the proximal side of the tr 

 angle. Thus it is that th 

 anal vein seems to join tl: 



Fig. 15.— Diagram illustrating the procession (jf the 

 triangle and the deflection of the anal vein and 

 second cubito-anal crossvein in the fore wings of 

 libellulid.e. a, the first, and 6, the second cubito- 

 ANAL crossveins: 1, 2, 3, AND 4, SUCCESSIVE POSITIONS. 



front angle of the triangh 

 With the ac(iuisiti<>n of new responsibilities, the second cubito-an: 

 crossvein has waxed stronger and longer, and has pushed the iniK 

 angle of the triangle toward the outer, making the triangle narrov 

 and bringing about by this means its apparent remoteness from tl 

 arculus. Such genera are illustrated in Plates XLIII, XLV, an 

 XLIX, and show the steps in this process. Simultaneously tl 



«Their differentiation in function was proved experimentally by Girard (186 

 Note sur divers experiences relatives a la fonction des ailes chez les insectes, An 

 See. Ent. France, (4), II, pp. 153-162). His results that are of most interest he 

 were gotten by simply clipping off one pair of wings at a time. Thus he showed th 

 in A(/rion, with fore and hind wings practically alike, flight is fairly well sustaint 

 with either pair of wings alone: Li))elluhi can fly with the fore wings only, but wi 

 the fore wings removed the hind wings alone will not sustain it in the air. I ha 

 veriiied these results, using Enalhujtna and Sumpclndii. 1 have also made stuuent 

 exiierinients to determine the utility of certain structures. Brief reference will i 

 made to my results farther on. 



