230 R. E. SNODGRASS. 



we assume that the structure of this organ is most primitive in such 

 genera as Antocha, Dicranomyia and Limiiobia. 



The variation of the internal features of the hypopygium is less 

 important than that of the external. In all the genera below the 

 Tipulina the penis is a, straight, or slightly curved, tubular rod 

 arising from the floor of the genital chamber and projecting straight 

 backward. In the Tipulina it arises from the roof of the genital 

 chamber. Its base is swollen to produce the central vesicle, while 

 the rest is a slender, often hair-like tube curving anteriorly, ven- 

 trally and then posteriorly, making often a long loop forward. Its 

 tip is in all cases protected by a guard. This latter structure is 

 typically composed of two slender, longitudinal plates set close 

 together on edge, with their lower margins united by membrane. 

 In the lower genera often the entire penis lies in the groove of the 

 guard. In the Tipulina it protects the tip when retracted, and 

 guides the penis when being protruded. From above the base of 

 the penis or the base of the guard, especially in the lower genera, 

 there projects caudally a pair of arm like processes which are often 

 forked. If we count the guard as two processes, since it is so often 

 composed of two slender plates united by their lower edges, then, 

 with the dorsal pair, we have four processes arising about the base 

 of the penis, one pair being above its base and the other below. It 

 is evident that we may regard these structures as the homologues of 

 the four free arms that arise about the base of the penis in a similar 

 manner in some other Dipteran families. These have been termed 

 the male gonapophyses, and so we may call the dorsal processes in 

 the Tipulidse the second or upper gonapophyses. 



No attempt has been made in this paper to speculate upon the 

 function of the parts of the hypopygium, or upon the cause of their 

 evolution into such varied forms. The latter would afford a most 

 interesting field for investigation. With all the striking diversifica- 

 tion of the male parts, there is almost no variation in the genital 

 parts of the females. Throughout the entire family the females 

 present one type of structure,* of which there is but little modifi- 

 cation, and certainly none to correspond with the great variety of 

 specific diflferences found in the genitalia of the males. 



* Jourual N. Y. Eotom. Soc. Terminal segments of female Tipididw (MS.) E. 

 E. 8nodgrass. 



