164 NORTH AMERICAN BLATTIDAE 



Subfamily BLATTINAE 



The following features are considered diagnostic. Head with 

 vertex usually exposed. Ocellar spots rarely absent. Tegmina, 

 when present, usually semicoriaceous. Male subgenital plate 

 rotundato-rectangulate and symmetrical, with a pair of slender 

 styles, of equal length, set in similar sockets on the distal margin. 

 Male concealed genitalia very complex. Subgenital plate of 

 female with meso-distal portion valvular; a transverse, weakly 

 arcuate suture bounding these valves proximad. Femora with 

 ventral margins heavily spined. 



The valvular female subgenital plate, with basal suture of valves 

 transverse arcuate, is a distinctive feature, found only in the Blat- 

 tinae and occurring in all of the species of this subfamily. Shel- 

 ford,-*^*^ following Brunner, however, was not fully correct in his 

 comments on this character, as a valvular female subgenital plate, 

 though different in structure, is characteristic of the genera .4 wa- 

 plecta, Tlolocompsa, Compsodes and Ilomoeogamia, members of the 

 Ectobiinae, Corydiinae and Polyphaginae. 



We do not, however, conform with Caudell-''^ in merging the 

 Psuedomopinae and Blattinae^^s on the ground that they appear 

 to differ in this feature alone. The present subfamily is in our 

 opinion, fully as satisfactory as the others in the Blattidae. In 

 no case oan these subfamilies be separated by one or two absolutely 

 diagnostic characters, and we are forced to rely rather upon a 

 general appearance of consanguinity added to a certain usual 

 combination of characters. We can not too strongly emphasize 

 the assertion that subfamilies are artificial groupings to facilitate 

 systematic work, and, as they are such, exceptions must be ex- 

 pected. In fact the laws of organic evolution prohibit the devel- 

 opment, within a family like the Blattidae, of subdi\isions, such 

 as we term subfamilies, with distinctions clear cut and constant. 

 The species of the Blattidae are too numerous, and far too many 

 primitive types still sur\ive, for this to be possible. 



2«sGcn. Ins., Fasc. 109, Orth., Blattidae, Blattinac, p. i, (1910). 



26" Proc. U. S. Nat. Miis., xliv, p. 600, (1913). 



268 That author states "I have united Periplanetinae with Blattinae." Instead of 

 Periplanetinae, Pseudomopinae was intended, the equivalent of which has been Phyllo- 

 dromiinae. 



