308 Annals Entomological Society of America [Vol. XII, 



with in certain Libellulid dragonflies. There is no other trace 

 of the ovipositor in the Plecoptera, so far as I am aware, the 

 ninth sternum being quite simple. 



The tenth segment is well developed, often little smaller 

 than the ninth. The tergite is often somewhat prolonged 

 behind, concealing or partly covering the small supra-anal 

 plate, with which it may be adherent. The paraprocts are 

 large, generally well chitinized and usually intimately fused 

 with the bases of the cerci so that the latter appear to arise 

 from them. The cereal basipodites are thus not distinct from 

 the paraprocts. 



The cerci are typically multiarticulate, but are very variable 

 in respect to both the number and the form of the segments. 



These characters, taken by themselves, do not throw much 

 light on the affinities of this group, but point to a very gen- 

 eralized structure, with secondary loss of the ovipositor. The 

 primitive multiarticulate cerci are approached b}^ those of the 

 Grylloblattoidea, more closely than any other order, with the 

 possible exception of some of the Ephemerida. 



Ephemerida. 



Although the Ephemerida and Odonata can hardly be called 

 " Orthopteroid " insects, and their lines of descent from the 

 Paleodictyoptera are undoubtedly quite distinct from any of the 

 others considered, they deserve a few words, on account of their 

 having retained certain very primitive characters. 



The females of Ephemerida are chiefly remarkable from the 

 fact that the oviducts open separately, and behind the seventh, 

 instead of the eighth sternum. In some forms the seventh 

 sternum is prolonged backwards into a spout-like structure, 

 which apparently functions as an ovipositor (Morrison, '19)^* 

 but this, of course, has no homology with any part of the 

 Orthopterous ovipositor. The elongate, uniformly segmented 

 abdomen, miultiarticulate cerci and cerciform caudal filament, 

 borne by the eleventh tergite are all marks of primitive structure. 

 The anal valves are membranous and very slightly developed. 



2* Morrison, Emily Reed, Can. Ent., Vol. LI, No. 6, pp. 139-146 (1919). 



