60 Annals Entomological Society of America [Vol. XV, 



The retention of this primitive character together with 

 others, such as the median cerciform appendage, jointed coxites, 

 the numerous moults, occurrence of a subimago stage, etc., 

 stamp the Ephemerida as far removed from other orders and 

 indicate that their line of descent had separated off before those 

 of the other orders had become differentiated. 



Next to the Ephemerida the Odonata are the most clearly 

 circumscribed group, having no near allies among recent 

 insects. The 9th sternum of the male preserves clear traces 

 of the original division into sternite and coxites, and the lateral 

 gonapophyses of the female are unique among Pterygote 

 insects in retaining the styli in the adult stage. They also 

 resemble the Thysanura (Machiloidea and Lepismoidea) some- 

 what in the form of these structures and their relation to the 

 other two pairs of gonapophyses. In their venation and 

 general structure of the body they show no near relationship 

 to the other groups here considered and may be regarded as 

 having probabl}^ separated off next in order after the 

 Ephemerida. 



In the remaining orders we may recognize two general 

 trends of evolution. In the one the tarsi became 3- jointed (or 

 remained thus), the ovipositor w^as reduced and subsequently in 

 most cases wholly lost, the coxites of the male were fused with 

 the 9th sternite, and the styli disappeared in both sexes. 



One of the branches from this stock gave rise to the 

 Plecoptera and probably from the same branch arose the 

 Embiidina. From another branch the Dermaptera were 

 evolved. The cerci were originally segmented in all of these 

 groups, but lost this primitive character in the Dermaptera. 

 The ovipositor was probably reduced in the common ancestor 

 of all these orders and completely disappeared in all but a few 

 of the Dermaptera. 



The terminal abdominal structures, however, throw little 

 light on the relationships of these groups and the views expressed 

 here are based largely on the work of Crampton. 



The other trend of evolution among Orthopteroid insects 

 shows itself in the presence of 5-jointed tarsi, a well developed 

 ovipositor, the retention of separate styligerous coxites in the 

 male and the loss of the double penis. In most of the modern 

 descendants of this branch, however, the coxites of the male 

 have fused with the 9th sternite, or with one another, in many 



