340 



ENTOMOLOGICAL NEWS [Oct., ' 



The fact that the sterna are widely separated, is character- 

 istic of the Isoptera; and the triangular outline of the pro- 

 sternum (bs l , Fig. 3) is also strongly suggestive of the Isoptera. 

 The smaller sternal sclerites fs and jj were very hard to dis- 

 tinguish, and I am not certain that their outlines are correctly 

 portrayed, but as far as I could make out, they seem to re- 

 semble those of the Isoptera. The larger sternal plate, bs, of 

 the meso- and metathorax is unlike that of any other insect I 

 have examined. 



The coxae (Fig. 2, ex} are extremely like those of the Isop- 

 tera, save for the fact that the meso- and metathoracic coxae 

 of Grylloblatta are not divided by a vertical suture into two 

 parts, as in the Isoptera, and in this respect they are more sim- 

 ilar to the meso- and metathoracic coxae of the Blattidae. The 

 trochanters are similar to those of the Isoptera and Blattidae. 

 The femora on the other hand are strikingly similar to those 

 of the Dermaptera. The tibiae also resemble those of the 

 Dermaptera, somewhat, although certain features present in 

 them are suggestive of the Isoptera and Blattidae. The four 

 basal tarsal segments are as much like those of the Blattidae 

 as any, but the terminal tarsal segment is surprisingly like that 

 of the Dermaptera not only in shape but also in the absence 

 of pulvilli, and in the character of the claws it bears. 



It is therefore apparent from the study of the thoracic scler- 

 ites, that the Grylloblattidae combine in themselves characters 

 which occur in the Dermaptera, Isoptera (and the closely re- 

 lated Blattidae) and Gryllidae. There are also certain fea- 

 tures, which "presage" (so to speak) some of the characters 

 which reach a fuller development in the crickets and katy-dids. 

 It would be impossible, however, to form any definite conclu- 

 sions concerning the affinities of the Grylloblattids from the 

 study of the thoracic structures alone, and it is therefore nec- 

 essary to take into consideration the head and terminal abdom- 

 inal structures as well. (The ideal method would be to make 

 a comparative study of the internal structures, the embryologi- 

 cal development of the various organs, etc., in addition to the 

 study of the external morphology, but in this case, such an 



