62 Annals Entomological Society of America [Vol. XV, 



as in the Orthoptera and lost their styli. The development 

 in the female of a large subgenital plate from the 8th sternum 

 also recalls the Orthoptera. These last features together with 

 the rather strongly Orthopteroid wing-venation make the 

 position of the order doubtful. It is perhaps equally well 

 placed at the base of the Orthopteran stem, in which case the 

 resemblance of the penis to that of the other orders in which 

 it is asymmetrical must be looked upon as a case of con- 

 vergence.* 



In the Grylloblattoidea the ovipositor is retained in its 

 primitive exposed form, there being no distinct subgenital 

 plate. The coxites in the male remain, separate from the 9th 

 sternite, and the supra-anal plate, though small, is distinct in the 

 adult. The Plecopteroid form of body is retained, with many 

 primitive features, e. g., in the cervical sclerites, the propleura, 

 and separate latero-sternites of the meso- and metasterna. 

 The legs, however, are Blattoid and the coxae large and close 

 together like all of the Blattoid groups. 



In the Blattoidea, Mantoidea and Isoptera so many common 

 features are present that we need have no hesitation in con- 

 sidering them as belonging to one common stock. The ovi- 

 positor is concealed by a large subgenital plate, the modified 

 sternum of the 7th abdominal segment. It is reduced in the 

 Blattoidea and is vestigial or absent in the Isoptera. The 

 coxites of the male have fused with the ninth sternite, but the 

 styli are retained. The penis, except in the Isoptera, in which 

 it has disappeared, is remarkable for its extreme asymmetry of 

 structure, which as in Gryllohlatta, may be shared by neighboring 

 parts. The cerci tend to shorten, being generally reduced in 

 the Isoptera to two segments. The supra-anal plate is replaced 

 by the 10th tergum. Among other peculiarities is the form of 

 the cervical sclerites, the ventral pair of which meet in the 

 middle line. The tarsi are primitively 5-jointed, even in the 

 Isoptera, in which the number of segments is commonly 

 reduced. 



The exact relationship of the Isoptera to the other two 

 orders is very uncertain. The form of the body and position of 

 the head is more "plecopteroid" and hence apparently more 



* The abdominal musculature, according to Miss Ford, shows a nearer rela- 

 tionship to the Blattoid-Mantoid type than to the Orthopteran type. 



