GENERA OF AFRICAN LYCAENIDAE 



223 



characters are sometimes so poorly defined that it is difficult to decide whether one 

 is dealing with a distinct species, a geographical race or even an individual variation. 

 The species can be grouped according to variation in the valves, but as in certain 

 species I had only a few specimens for dissection, I cannot rest satisfied with such an 

 arrangement. To illustrate what I mean I give a figure (Text-fig. 194) of the male 

 genitalia of major, which had been described as a subspecies of noquasa, but which is 

 certainly a valid species. 



H. notobia Trimen (Text-fig. 195). In this species uncus, subunci, tegumen and 

 penis are of the same type as in reginaldi, but the two arms of the lower fultura are 

 strongly curved and their apices are connected by a membrane which forms a com- 

 plete ring around the penis, and the valves are very long, digitate, without any distal 

 widening. 



Judging by their male genitalia the species of Harpendyrens form a homogeneous 

 genus. The markings on the underside of the hind wing of reginaldi remind one of 

 Uranothaunta antinorii Oberthur. All the other species have a white transverse 

 streak, more or less pronounced, like the one in Cyclyrius webbianus and it is this 

 trivial unimportant character that led some authors to include them, rather arbi- 

 trarily, in the genus Cyclyrius. 



In Eastern Africa Harpendyreus is distributed from Kivu, Ruanda and Ruwenzori 

 to Natal, Transvaal, Basutoland and the Cape, but it is discontinuous, the various 

 scattered species and subspecies being restricted to specific mountain masses, though 

 clearly having a common origin. 



The early stages of notobia and noquasa have been described by Murray (1935 : 

 163, 186). 



Fig. 194. Harpendyreus major (Joicey & Talbot), <J genitalia. 



