580 PROCEEDINOS OF THE NATIONAL MUSEUM. vol. xxix. 



the stigma. Its posterior end in the families Xyelidte (tigs. 31-35) 

 and Cephid;e (Hgs. 93-lH)) is attached near the middle of the cell Rj; 

 in the Megalodontidie (tig. 92) and Siricidie (tigs. 86-91), except the 

 genera Tremex (tig. 91) and Teredon (tig. 90), where it is joined to the 

 cell Rj, it is attached just bej^ond the middle of the cell Rj; in 

 the L3^did{e (tigs. 3t)— 43) it varies in position from just beyond the 

 middle of the cell Rg to the apex of this cell, and in some forms is 

 interstitial with the free part of the vein Rr,, and in all other Tenthre- 

 dinoidea it is attached near the middle of the cell R^, except in certain 

 species of Tenthredopsis^ Scolioneura (tig. T5), Monoj^hadnus^ Lycaota 

 (tig. 55), and Blen.nocampa (tig. 72), where it is interstitial with the 

 free part of the vein R^, and in certain species of the genus Kaliosys- 

 jpliinga (tig. 73), where it is attached to the cell R3, but this latter 

 change is not due to a shifting of the position of the radial cross-vein, 

 but to a migration of the free part of the vein R^ toward the base of 

 the wing. The r-adial cross-vein is present in the wings of all Hyme- 

 noptera where the base of the radial sector is present other than those 

 of a few groups of Tenthredinidae, Lophvrinte (tig. 45), Nematinae 

 (tig. 68), Perreviinffi (tig. 80), Pergina? (tig. 84), and Pterygopherinre 

 (tig. 81). 



The radio-medial cross-vein is rarel}" wanting, though in many 

 genera it is so completely covered by a large clear spot or bulla that 

 it is often difficult to determine whether it is present or not. This 

 condition is well shown in many Nematina?, where all stages from a 

 distinct well-marked cross-vein to its total disappearance can be 

 found. In other genera, as Trichiosoma (tig. 59), Clavellavin (fig. 60), 

 OrysKHs. (tig. 97), KaHosysphinga (fig. 73), Acordidecera (fig. 83), and 

 Blastlcotoma (fig. 44), all trace of the cross-vein has disappeared, 

 while in Monoctenus (fig. 67) only the posterior half is wanting. In 

 those genera, where this cross-vein is retained, it always appears as a 

 transverse vein extending between the stem of the radial sector and 

 the stem of media. In the Xyelida (figs. 31-35) the medial end has 

 swung toward the base of the wing so that it appears to be a contin- 

 uation of the radial sector, while a portion of the stem of the radial 

 sector appears to be the cross-vein. In certain genera of the Siri- 

 cida?, as Sirex (fig. 87-88) and Tremex (fig. 91), the medial end has 

 swung around stiW farther toward the base of the w^ing so that it arises 

 from the angle made by the transverse and longitudinal parts of the 

 stem of media where it is joined by the medio-cubital cross-vein, 

 and in some species arises distinct!}' from the transverse part of 

 media. 



The free part of R, is wanting in only a very few genera, as Dolerus 

 (fig. 49), Lode7'u><, Euura^ and Tremex (fig. 91). 



The free part of R^ so far as observed is never wanting in this super- 

 family. It is not so constant in position as R5; in the Xyelidre (figs. 



