NO. 143S. WINGS OF THE TENTHREDINOIDEA—MacGILTJVBAY. 037 



found in the Schizocerinss, where this cell is appendicuhite in Lahi- 

 darge (fi^. 78), and broadly open at apex in Scohhut and Diclocerus 

 (fig. 79); in fact, in Scohinu this cell is not appendiculate in either wing. 

 If the results of our studies on the other groups of this family be of 

 any value, then the variation in the characters just cited must indicate 

 ver}' different lines of ascent, and be of more value than for the mere 

 differentiation of genera. The medio-cubital cross-vein is generally 

 joined to Sc+R+^I Jit or very near the origin of media. Although it 

 is not true of all the genera, yet in certain genera there seems to be a 

 marked tendency for the migration of the free parts of R^ and R^ and 

 the transverse part of M^ toward the ))ase of the wing, thus greatly 

 increasing the size of the apical cells. The second anal cell is wanting 

 in the Perrejnn^, Init according to descriptions of genera may be either 

 present or absent in both the Hylotomina? and Schizocerinte. This 

 character is not of any phylogentic value, since, so far as it is con- 

 cerned, these subfamilies are imdergoing a progressive reduction of 

 this cell. In the Hylotomina' and Schizocerinjv the hind margin of 

 the cell Mj is a fairly straight vein, the free part of M^+Cuj joining it 

 near its middle, and is either perpendicular to it or inclined toward the 

 apex of the wing. In most Perreyiniv the hind margin of the cell M^ 

 is deeply curved, the free part of M^+Cu, is joined to it at or before 

 the middle, and is always strongly inclined toward the l)ase of the 

 wing. This is the only venational character thus far discovered that 

 is of any value in separating the Perreyinte from the other two sub- 

 families. In the hind wings it is only with rare exception that either 

 the free part of R^ or the transverse part of M.^ are wanting in the 

 Hylotomina? and Schizocerina?, and when either of them are wanting 

 it is always the latter. In marked contrast to this is the almost entire 

 absence of the transverse part of M, in the Perreyina', the free part of 

 Rij being always present, so far as can be judged from figures of wings. 

 Here, just as in the front wings, there is a marked tendency toward 

 the migration of the free part of R^ and the transverse part of M2, 

 when it is present, toward the base of the wing. In many Tenthredi- 

 nida? the transverse part of M^ is either interstitial, or nearly so, with 

 the free part of R^, l)ut in this group it has migrated toward the base 

 of the wing vmtil it is near, or sometimes even before, the free part 

 of M^. There is also in many Tenthredinida^ a prominent angle oppo- 

 site the anterior end of the free part of M.,, })ut in this group this 

 angle has l)een straightened out and cul)itus appears to extend directl}' 

 to the margin of the wing. The Perreyinie are frequently separated 

 from all the preceding subfamilies by the loss of the free part of the 

 second anal vein. It has been pointed out above that this vein is also 

 wanting in the Phyllotominje and Fenusint^, and the same condition 

 is found in certain Schizocerinii?and Hylotomiufp. This is a character 

 that has arisen several times in widely separated groups, and does not 



