182 R. E. SNODGRASS. 



terior wall of the genital chamber, just above the guard of the penis, 

 two elongate, free, chitinous rods. Each of these is very commonly 

 forked. The word gonapophyses has been used by various describers 

 of insect anatomy to designate four free rods that arise from the wall 

 of the genital chamber, two above and two below the base of the 

 penis, and project backward within the chamber. The application 

 of the name "gonapophyses" to these rods is, very likely, a misno 

 mer from the standpoint of homology. However, since it has been 

 so used the writer will not here attempt to use a new word. 



The guard of the penis is, as already stated, very often a solid 

 structure with simply a lengthwise groove above ; but, in a great 

 many cases, it consists of two slender plates set on edge side by side 

 and united by membrane along their lower margins. This structure 

 would suggest that it may have been formed from two plate-like 

 processes, extending backward from beneath the penis, whose lower 

 edges became connected by membrane'. In this case there would 

 have originally been a pair of processes projecting caudally above the 

 penis and another pair below it. This would be exactly the condi- 

 tion that exists in many forms of Diptera, e.g. in the Muscidse. 

 Hence, we can imagine, at least, that the guard of the penis and the 

 two gonapophyses of the Tipulidse represent the four gonapophyses 

 of some other families. For this reason the Tipulid gonapophyses 

 will be referred to in this paper as the posterior or second gonapo- 

 physes. The simple relationship of gonapophyses, penis and guard 

 is seen best in the Tipulid genera below the group Tipulina, where 

 the penis is a short, straight, rod-like tube. 



As will be seen from the specific descriptions given beyond, and 

 in the conclusion at the end of this paper, the structure of the hypo 

 pygium affords a good basis for determining not only the relation- 

 ships of the larger groups of the family to one another, but also that 

 of genera within the groups and of species in the genera. Further- 

 more, the minor features of the hypopygium, and especially the 

 shape of the apical appendages, make most excellent specific char- 

 acters. In the collection of Tipulidse, from which this paper is pre- 

 pared, a large number of specimens of the genus Antocha were 

 placed together and labelled A. opalizans. Externally they are all 

 very similar, except that some (PI. VIII, fig. 5) have the pleural 

 plates a little larger than others. However, it was found that in the 

 former the guard of the penis has the form shown in fig. 8, while in 



