0. 1331. 



DRAGON-FLY WING VENATION— NEEDHAM. 



749 



Agrioninse. — This group contains a larger number of genera and 

 ;pecies than any other of equal homogeneit}'^ of venational character- 

 •stics. The radial sector leaves vein 3f^ near the nodus (sometimes 

 j'ollowing vein J/g in its migration along vein 31^ toward the stigma), 

 'md nodus and quadrangle tend to close approximation. Cross veins 

 fire generally matched in transverse lines, and the stigma is generally 

 krongl}' braced. 



[ Two minor lines of development may be briefly indicated in passing: 

 i3ne, tending toward the loss of the branches of the cubitus — Palximia 

 •Plate LIIl, tig. 2), Platysticta (Plate LIU, tig. 3), DisjMroneura (Plate 

 LIV, fig. 2), Idioneura (Plate LIV, fig. 3), and Oamoneura (Plate LIV, 

 'ig. 4); and a second, including nearly the whole of the group which 

 lacked this tendency, but in which nodus and quadrangle become more 

 uid more approximate, and the veins J/j, R^, and M.^ migrate sepa- 

 rately along vein J/j from their accustomed places toward the stigma 

 '[Plate LllI, tigs. 4, 5, and 6), or in which progress has consisted in 



Fig. 38.— Wings of Hcniiphkbia mirabilis Selys. 



mere reduction of cross veins and better matching of them in transverse 

 lines, and in the perfecting braces at the nodus and elsewhere. 



This series furnishes in the genus Ileiiuphleh'ia (tig. 38) a striking 

 [example of the loss of a cross vein that is elsewhere very constant — the 



III, figs. 16 and 21). I think Thaumatoneuria belongs here; it has all the essential 

 venational characters of this group, and surely these are sufficiently distinctive. It 

 is more generalized than Megaloprcpus, in that vein M., has made less progress along 

 vein 3fi away from the nodus, some vestiges of the primeval extra antenodal cross 

 veins are preserved, most of the interpolated sectors are still unattached to the veins, 

 and the stigma is larger and better preserved. It is a curious and probably significant 

 fact that in the two series of Zvgoptera— Vestalin.e and Anormosti(;.matin.e — in 

 which the long sector between veins J/j and M.2, parallels M.^ should lie the only ones 

 in which the stigma progressively dwindles and disappears. It is conceivable that 

 the stigma might lose importance for want of the support that this sector would give 

 if approximated to it at the apical bend. Perhaps the concavity on the costa in 

 Mecistogader lucretia (Plate LI, fig. 8) and the conjoining and the sharp bending 

 backward of the veins beliind it may be a unique way of supplying such deficiency of 

 bracing, preserving the utility of the stigma as a weighted striking point toward the 

 end of the cutting edsfe of the wing. 



