88 CONNECTICUT GEOL. AND NAT. HIST. SURAEY [BlllL 



inversion of the ])arts has been recorded for some dixids. chironomids, 

 culicids, phlebotomine psychodids, mycetophilids, etc., among the 

 Nematocera, while in the orthorrhaphous Brachycera, such asilids as 

 DasylUs and Laphria have the eighth and ninth segments (with the 

 proctiger) temporarily inverted during copulation. 



Tn tlie second type of inversion, which may be spoken of as a 

 torsion inversion (or "retroinversion"), the inversion results from a 

 torsion or Avinding of the parts forward and around along the sagittal 

 l)lane of the body, as in the insects shown in Fig. 12, E and I, (com- 

 pare also Fig. 12, B), as though it Avere an arrested stage in the so- 

 called circum version of the ninth segment (with the proctiger),_ which 

 takes place in the pupal stage of such Cyclorrhapha as GaUifhora. 

 In tliis type of inversion, the aedeagus and other structures are direct- 

 ed more nearly downward and forward than backward (as they are 

 in the rotary type of inversion). 



The torsion process is apparently initiated by the setting out of, 

 the ninth segment (with the proctiger) to undergo a "circumversion", 

 which probably takes place by a turning of the parts forward and 

 around along the sagittal plane of the body, and the process is con- 

 tinued until the ninth segment and proctiger almost complete the 

 circumversion, although they do not completely revert to their orig- 

 inal position, since the aedeagus, etc., project downward and forward 

 (as in T^'igs. 13 and 14) instead of backward, as they do in the lower 

 Diptera shown in Fig. 12, C, G, etc., in which the parts remain in their 

 original position. 



As the ninth segment Avinds around, it apparently drags up the 

 eighth sternite into the ir.sect's left side, and on over into the dorsal 

 region where it becomes inverted in the distorted post abdomen of 

 male (Cyclorrhapha. The seventh sternite is apparently dragged up 

 into the insect's left side, following the eighth sternite, and becoming 

 strongly asj^mmetrical. Finally, the sixth sternite may become dis- 

 placed into the insect's left side, but the sixth tergite is less art'ected 

 by the torsion process than the seventh and eighth tergites are, since 

 these may become vestigial or atrophied, although their ultimate fate 

 is not clear from the material at hand. 



A torsion of the parts occurs in some Dolichopodidae and Em- 

 pidae, and in certain Bombyliidae and Cyrtidae, etc., as well as in the 

 Phoridae and Lonchopteridae which approach (or may be included 

 among) the Cyclorrhapha, in which the torsion tendency reaches its 

 Ihuil culmination and affects all of the males of this suborder. The 

 torsion exhibited by such dolichopodids as the one shown in Fig. 12, B, 

 (com})are also Fig. 10, D) is particularly suggestiA'e of a foreshadow- 

 ing of the more ])ronounced torsion occurring in male Cyclorrhapha, 

 and a study of these dolichopodids is of considerable interest since 

 it suggests how the torsion process began in the lower forms. 



Since the twisting of the terminalia is always from left to right 

 (as is shown by the looping up of the ejaculatory duct from left to 

 right over the top of the hindgut) in the lower as well as in the 

 higher Diptera, it is possible that Ave are dealing Avith a fundamental 



