264 H. STEMPFFER 



lower portion curved like a pistol-stock, widely open on its upper surface, apex pointed ; vesica 

 enclosing a short spine ; uncus densely clothed with fine hair ; just a few hairs on the upper 

 edge of the valves near the apex. 



The male genitalia of abottii (see Stempffer, 1938, Mission Omo 4 : 210) resemble 

 so closely those of phleaas that it is permissible to wonder whether it is a true species 

 or a geographical subspecies of phlaeas. Those of orus are of the same type except 

 that the apices of the valves, instead of being rounded, are obliquely truncate and 

 irregularly dentate, and the lower fultura consists of two narrow lamellae with 

 pointed apices. 



The biology of the European races of phlaeas is well known. The early stages of 

 orus have been described by Murray (1935 : 140, fig. 86) and Dickson (1947, /. ent. 

 Soc. sth. Afr. 10 : 127). 



One can understand how, during the colder geographical periods, the genus 

 Lycaena managed to reach South Africa by way of the Mountains of Abyssinia and 

 East Africa. It is much more difficult to explain how it reached New Zealand, 

 since in Asia it extends no further than the Himalayas and Sze Chuan being 

 effectively replaced by Heliophorus in further Asia. 



List of Species of Lycaena 



*Lycaena abottii (Holland), 1892. Fig. Holland, 1895. 

 Lycaena areas (Fabricius), see orus. 

 * Lycaena orus (Cramer), 1780. 



areas (Fabricius), 1787. 

 *Lycaena phlaeas pseudophlaeas (Lucas), 1866. 



phlaeas ethiopica (Poulton), 192 1 ; phlaeas menelicki (Thierry Mieg), 191 1. 

 Lycaena phlaeas ethiopica (Poulton), see phlaeas pseudophlaeas. 

 Lycaena phlaeas menelicki (Thierry Mieg), see phlaeas pseudophlaeas. 

 Lycaena phlaeas shima Gabriel, 1954, Exped. S.W. Arabia 1937-38 : 38. 



CONCLUSION 



Now that I have set out in detail the generic characters of the Ethiopian Lycaen- 

 idae, the time has come to draw conclusions from my studies, and to attempt an 

 outline of classification based on phylogeny, one which will, I hope, give a more 

 accurate picture of the family than that given by the existing classification. 



What are the essential characters that should form the basis of a classification ? 

 This poses the question that is at the root of the problem, because differing con- 

 clusions will result from differing choices of prime characters. Further, whatever 

 these results may be, in terms of major taxonomic units, many of the latter will not 

 be completely homogeneous, for not all the included species will display all the chosen 

 characters ; one can but hope to reach an approximation to a true phylogeny, 

 because the fossil record which could prove or disaprove it, is lacking. 



Aurivillius chose venation as the prime factor and I quote below the characters 



