358 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 84 



There appear to be good specific differences in the shape and size 

 of the bursa copuhitrix, but, while this organ has been carefully 

 figured in each case, I have not attempted to use it to define species. 

 There are plenty of other more obvious characters in the female 

 genitalia, and the bursa is so subject to distortion in preparation, 

 so difficult to see in balsam, and subject to so much individual varia- 

 tion in size or shape that the attempt to use it in classification of 

 species would confuse rather than help our definitions. 



KEY TO THE SPECIES OF RUPELA 



Males 



1. Gnathos with central area caudally produced and strongly 



sclerotized (beaklike). Anal tuft white 2 



Gnathos with central area not caudally produced, thin (band- 

 like). Anal tuft yellow 13 



2. Dorsal element of anellus spined 3 



Dorsal element of anellus unspined 9 



3. Uncus laterally compressed at apex 4 



Uncus not laterally compressed at apex 5 



4. Aedeagus finely scobinate on venter near apex leucatea (p. 360) 



Aedeagus sphied on lateral margins at apex segrega (p. 3G6) 



Aedeagus with pronounced lateral spur at apex pallidula (p. 365) 



5. Tegumen with projecting spur from each ventrolateral 



margin scitula (p. 374) 



Tegumen simple 6 



6. Penis bearing a line of minute cornuti liberta (p. 364) 



Penis without cornuti 7 



7. Aedeagus with ventral scobinations near apex. One pair of 



long, stout spines on anellus cornigera (p. 371) 



Aedeagus with one lateral margin near apex weakly serrate. 



Spines on anellus numerous, small, scattered albinella (p. 362) 



Aedeagus with apex smooth and labeose ; anellus with two or 



three pairs of minute .spines labeosa (p. 36.3) 



Aedeagus with sclerotized manica, otherwi.se simple ; .spines 



on anellus stout, rather short in a single cluster or in a 



pair of dense combs 8 



8. Spines of anellus a single cluster at one side of dorsal mem- 



branous part g-ibbera (p. 367) 



Spines of anellus arranged as an opposing pair of dense, dark 

 combs saetigera (p. 367) 



9. Sacculus of harpe produced at apex into a long, stout spine 



(clasper) 10 



Sacculus not .so produced 11 



10. Clasper a straight spine nivea(p. 370) 



Clasper a curved spine vexativa (p. 371) 



11. Basal part of imcus scobinate and produced backward (cowllike) 12 



Basal part of uncus without spines or scobinations, not pro- 

 duced backward tinctella (p. 368) 



12. Aedeagus with apical half greatly narrowed (rodlike) . Cucullus 



of harpe narrow sejuncta (p. 373) 



