REHN AND HEBARD 23 



adpressed character of the cereal tooth giving it a rather unique 

 position, although in general its relationship to the other two 

 species is readily perceived. This group (C) is quite divergent 

 from group B, its probable ancestral type. Group D is probably 

 a link connecting groups B and E, but distinct enough in char- 

 acter from either of these to be given an independent position. 

 In Groups E and F the elongation of the cercus is progressively 

 pronounced, much resembling that found in one species of Group 

 G and one of Group I, which, however, are members of other 

 phyla of the genus. The ovipositor in these groups shows the 

 extreme development of the elongate arcuate type. Group F 

 was apparently derived from a Group E-like ancestor, and in it 

 we find the extreme development of the cercus in elongation 

 (equalled in unispina of Group I), this also being gently inbowed 

 distad, the tooth proximal and distinctly directed proximad, 

 while the ovipositor is straight and elongate. Group G is very 

 distinct in character, being sharply defined by the pecuUarity 

 of the speculum of the male tegmina, and the straight dorsal 

 margin of the ovipositor (this in volantum resembling that of 

 gladiator of Group B, but this is probably due to convergence 

 caused by the use of similar oviposition sites). In this group (G) 

 the cercus is elongated, the tooth is distinctly proximal, although 

 the direction of the tooth is different in the two included species. 

 Group H occupies a peculiar position, showing a number of fea- 

 tures of relationship to Group I and some apparently superficial 

 resemblance to Group G, but the greater affinity is with Group I. 

 Group I shows a decided tendency toward Conocephalus, but in 

 general it is distinctly a member of the genus Orchelimum. The 

 unispinose caudal genicular lobes of the species of Groups H and 

 I readily separate them from those of the other groups. Group 

 I has its extreme condition in unispina with its obsolete humeral 

 sinus. 



Key to the Syecies 



The following key is largely artificial, particularly in the female 

 sex, but it will be found to separate the majority of the species 

 with little difficulty. Some few forms which are easily distin- 

 guishable in the male are difficult to separate in the opposite sex 

 and vice versa. In case any difficulty is encountered in forming 



TRANS. AM. ENT. SOC, XLI. 



