598 PROCEEDINGS OF THE NATIONAL MUSEUM. vol. xxix. 



THE (X)STAL. AREA. 



The costal area is represented by the costa, which is present in the 

 hind wings of practically all Tenthredinoidea as a strong vein thick- 

 ened at base. So far as observed costa is wanting onl}^ in the genera 

 Of'i/ssu.s (tig. 97) and Stromhoceron (fig. 50). 



On the front margin of the hind wings of all Hymenoptera there is 

 a series of hooks for fastening the two wings of a side together so 

 that they will move as a unit during flight. These hooks are of vary- 

 ing extent, in the Siricida there is a prominent group at the base 

 and another near the apex of Scj with isolated hooks between; this 

 same condition is found in certain Lydidte while in others only the 

 basal and apical areas are preserved. The Xyelidte and Xiphydriidi© 

 also have basal and apical areas. In the Oryssida? there is an apical 

 area with four or five isolated hooks just before the apical area. In 

 most Tenthredinoidea there is only the apical area, which is likewise 

 characteristic of the higher Hymenoptera. The apical area is of 

 about the same extent in all the groups except the Cimbicinse, where 

 it extends from one-half to two-thirds the length of the cell Rj+g. 



THE SUBCOSTAL AREA. 



All vestige of the subcosta has disappeared except in the single 

 family Lydidte (figs. 36-43), where in the genera Lyda (fig. 37) and 

 Bactroceros (fig. 41) it is as fully preserved as it is in the fore wing 

 of the Lydid?e and Xyelid^. Subcosta and its continuation, Scj, 

 extend as a straight vein from the base of the wing to near the apex 

 of the vein Scg + Rj, The basal free part of Scg is a short vein only 

 three or four times as long as broad and in some genera, as Bactro- 

 ceros^ Neurotoma (fig. 36), and Pamjjhilkis (fig. 39), it is only about 

 as long as broad, while in the genus Cxnolyda (fig. 38) it is entirely 

 wanting. When present it is generally situated about midway of the 

 vein Rj, making the cells Sc and Scj subequal in length. The only 

 exception observed is in the genus Neurotoina^ where the free part of 

 Sc2 is much nearer the apex of the wing, the cell Sc^ being less than 

 one-half the length of the cell Sc. The apical free part of Sca has 

 been obliterated b}' its coalescence with R^ to the margin of the wing. 

 In the genera Pamphllius^ Neurotortia^ Cephalela (fig. 42), Itycorda 

 (fig. 40), and Lyda a considerable portion of the subcosta found 

 between the base of the wing and the free part of Scj has completely 

 atrophied, the amount varying in the difl'erent genera. The condi- 

 tions found in the genera just named go to show that the reduction of 

 the subcosta in the hind wings has proceeded in a very different way 

 from what it has in the fore wing, where the modification is clearly 

 due to coalescence. The cell laying between costa and Sc + R + M, 

 C + Sc + Scj, is broad and well marked in all the specialized Tenthre- 



