1919] Walker: Structure of Orthopteroid Insects 305 



and obviously represents the ventral valvulae. The other pair, 

 from its lateral position and comparatively broad form is doubt- 

 less the dorsal valvulae. Since the dorsal valvulae represent the 

 coxites of segment nine, the divided sternal plates, which are 

 also present cannot be correctly termed coxites. They are 

 more nearly comparable to the valvifers. 



In some genera, such as Anisolabis (Fig. (35) there is also 

 a small paired tenth sternum, but this is absent in Forficula. 

 The two large sub-triangular plates (Figs. 66, 67), which are 

 closely united with the margins of the tenth tergite and form 

 part of the articulation with the cerci, were regarded by Ver- 

 hoeff and Zacher as the coxites of segment ten, while Crampton 

 identified them with the paraprocts. Crampton's view is the 

 more probable in my opinion. Coxites are absent from segment 

 ten in all other primitive insects, so that on a priori grounds 

 we should not expect to find them in the Dermaptera. Para- 

 procts, on the other hand, are almost invariably present, and 

 although these plates appear to belong to segment ten, there is 

 no reason why, in such a highly modified group as the Der- 

 maptera, they should not have been developed from the para- 

 procts, as these are usually closely connected with the margins 

 of the tenth tergite, as, e. g., in the Blattoidea. 



The spiracles all occupy the pleural membrane, the last 

 pair, as in all the groups discussed, belonging to the eighth 

 segment. 



The unsegmented form of the cerci has probably been 

 developed within the course of evolution of the order, as in 

 immature stages of certain primitive forms {Diplatys, Karschiella 

 and Bormansia) they are segmented. 



Embiidina (Embiidas). 



This small group, which is undoubtedly of ordinal rank, has 

 been considered by some writers (Enderlein and others) to be 

 nearly allied to the Isoptera, while others (Crampton, West- 

 wood, MacLachlan) find closer relations with the Plecoptera 

 and Dermaptera. The latter view, is in the present writer's 

 opinion, much nearer the truth. The resemblance to the 

 Isoptera is largely due to the retention in both groups of many 

 primitive characters and to parallel development along certain 

 lines, such as the form and venation of the wings, the two- 



