1919] Walker: Structure of Orthopteroid Insects 293 



a "larval" type, i. e., they are probably to be looked upon as 

 degenerate, tunctionless structures, whose development has 

 been arrested at an early stage, as is not uncommonly the case 

 in degenerate structures. The eighth sternum is but little 

 produced over the bases of the ventral valvulae, which are 

 slender and flexible with broader basivalvulas. The ninth 

 segment is shorter than usual in this order, but the sternum 

 is horizontal, its lateral parts, the valvifers, being larger and 

 having a larval position in relation to the dorsal valves of the 

 ninth tergite. They were erroneously regarded as basal seg- 

 ments of the dorsal valvulae (basivalvulas) by Crampton. The 

 dorsal valvulae are flexible, triangular flaps, whose bases meet 

 the inner edges of the valvifers exactly. The term "dorsal" 

 is here inappropriate, as they retain the larval position external 

 to the inner valves, with which they are fused to within a short 

 distance of the apices of the latter. Neither dorsal nor inner 

 valvulae are connected with the ventral pair. Having seen 

 only one specimen of Timema, which was kindly lent by Prof. 

 Crampton, I have not studied the internal relations of the 

 valvulae. 



The tenth tergite is rather long dorsally, but narrowed 

 ventro-laterally. Its apex is somewhat produced over the supra- 

 anal plate, which like the paraprocts, is rather long and pointed, 

 though but thinly chitinized. The cerci are large and stout. 



A less primitive ovipositor, but more typical of the order, 

 is that of the common stick-insect, Diapheromera femoraia Say. 

 (Figs. 41-43). The valvulae of this species are of considerable 

 size, but quite soft and flexible. They are flattened dorso- 

 ventrally and arched towards the anus. The inner valvulae 

 lie as in Timema in the primitive position between the dorsal 

 valvulae (although in some genera they lie dorsal to the latter), 

 with which they are united in a similar manner. This con- 

 nection seems to be characteristic of the Phasmoids and it was 

 for this reason that the inner valvulae were mistaken by Lacaze- 

 Duthiers ('52, loc. cit.) for a part of the dorsal valvulce. There 

 is little development of the intervalvular membrane and the 

 rami are represented only by a short chitinous process on each 

 side, at the angle of union with the dorsal valvulae, these pro- 

 cesses ending freely, there being no trace of either superior or 

 inferior intervalvulae. The underside of the inner valvulae 

 bears a very distinct though flexible ridge (Fig. 43), which fits 



