No. 64] DIPTERA OF CONNECTICUT : MORPHOLOGY 87 



develoiMiieiit of the cerci in both sexes of the mosquitoes, finds that 

 there are true cerci in the male as well as in the female, and there is 

 every reason to believe that the cerci of other male Diptera in general 

 are homoloo-oiis with the cerci of the females, as is the case in the 

 mosquitoes. Furthermore, the cerci of female mosquitoes (and their 

 homologues in the male) are clearly homologous with those of other 

 primitive female Diptera in which the cerci are borne on a reduced 

 eleventh segment (as in Fig. 9, A), comparable in every way to the 

 cerci of female Mecoptera, in which the cerci are borne on a distinct 

 eleventh segment (although the eleventh segment also unites with the 

 tenth in male Mecoptera), as they are in both sexes of the lower 

 insects, in some of which, however, the cercus-bearing eleventh seg- 

 ment also tends to unite with the tenth segment even in these primi- 

 tive forms, so that it is not surprising that the eleventh segment 

 tends to unite with the tenth in male Diptera ! 



The cerci are one-segmented in male Diptera, and are usually 

 relatively smaller in male Nematocera than in male Cyclorrhapha, in 

 which they are associated with the surstyli in mating, and apparently 

 have a sensory function. They are frequently hugely developed in 

 male Dolichopodidae (Fig. 10, D). and may be strikingly modified in 

 these insects. 



Inversion and Torsion of the Male Terminalia. There are two 

 principal types of permanent inversion of the genital segments (with 

 the proctiger) in male Diptera, in both of which types the ninth ster- 

 nite takes up a dorsal position, and the ninth tergite assumes a ven- 

 tral position, but the direction in which the aedeagus, etc., project 

 after the inversion is quite different in the two types, because of the 

 difference in the ]3rocesses by which the two types of inversion are 

 brought about. 



The first type of inversion may be illustrated by the male mos- 

 quito shown in Fig. 11, B, in which the inversion takes place when the 

 parts rotate about the long axis of the body, through 180 degrees; 

 and the direction of the rotation is indicated by the looping up of the 

 ejaculatory duct from left to right over the top of the hindgut. The 

 designation "rotation" should be restricted to such a turning of the 

 segments about the long axis of the body, and the resulting inversion 

 may be spoken of as a rotary inversion (or "transinversion"). After 

 a rotary inversion has taken j^lace, the inverted aedeagus, genital 

 forceps, proctiger, etc., are still directed approximately straight back 

 along the long axis of the body. This type of inversion takes place 

 just after the insect emerges from the pupal stage in certain Tipu- 

 lidae, Culicidae, etc. 



Edwards (1924) describes a permanent inversion of the ninth 

 segment (with the proctiger) in the tipulid Sty ring omyia, and also 

 states (on page 265) that "the occurrence of this torsion is rare in 

 the Tipuliclae, occurring only, so far as I am aware, in the genera 

 Molophilus and Amphineurus, and in an incomplete form in Micro- 

 mastix." The writer has observed an inversion of the ninth segment 

 (and proctiger) in the primitive tanyderid ProtopJasa jitchii, and an 



