18 R. W. CROSSKEY 



small number of Simuliids in the fauna of North America and the USSR (where 

 most work has been done) that form a distinctive group recognized by the presence of 

 a fork in vein Rs, by lacking the calcipala, and by having the costa and radial veins 

 haired without any development of spiniform macrotrichia, as well as by a number of 

 other rather distinctive features in the immature stages. So long as only the fauna 

 of the Holarctic area is considered these characters hold good for the delimitation of a 

 rather conspicuous and easily characterized genus, but when forms from the whole 

 world fauna are taken into consideration it no longer becomes possible to delimit 

 Prosimulium at all clearly by the old characters that worked well for a limited area : 

 for example, forms from southern Africa or from Western Australia, which on overall 

 balance of characters have a completely Prosimulium-like facies, simply fail to fit 

 with the old definitions or alternatively to show combinations of characters among 

 themselves that could serve to exclude them satisfactorily from Prosimulium. For 

 examples of the difficulties met may be mentioned a species from Australia (tonnoiri 

 Drummond) in which vein Rs sometimes has a definite fork while at other times it is 

 simple, or a species from Africa (morotoense McCrae & Prentice) in which the costal 

 vestiture is hair-like in the female but incipiently spiniform in the male. 



It seems to me that it is no longer possible in the interests of a balanced world 

 classification to maintain Prosimulium in the old narrow sense favoured by workers 

 in North America, Europe and USSR (and as used for example in the recent publica- 

 tions of Rubzov, 1956, 1959-1964 ; Stone, 1963, 1964, 1965) and I prefer to treat 

 Prosimulium in a broad sense but divided into a number of named subgeneric 

 segregates — thus establishing a concept for Prosimulium that equates with that 

 accepted by most workers for Simulium s.l. and balances the classification of the 

 more primitive forms in segregates more equivalent to those recognized for the more 

 advanced forms. A detailed account of the subgenera of Prosimulium s.l. thus 

 recognized on a world basis will be presented later, and here it is only necessary to 

 put forward keys to, and diagnoses of, the three subgenera that occur in Africa (note 

 that Prosimulium s.l. in the African area is confined to continental Africa and is 

 absent from the Malagasy Region and the other islands). 



The genus is only known from a few scattered African localities, and very little 

 material is yet available in collections from Africa, but future collecting will almost 

 certainly ' fill in ' some of the apparent breaks in distribution ; until recently the 

 forms here placed in Prosimulium (previously placed in Cnephia Enderlein by 

 Freeman & de Meillon (1953)) were believed to be confined to the extreme south of 

 Africa (Cape Province and South- West Africa) but the recent description of species 

 from Uganda by McCrae & Prentice (1965) and from Rhodesia by Crosskey (1968) 

 has shown that there is not nearly such a wide gap in distribution of Prosimulium- 

 like forms in the Palaearctic and southern Africa as was previously supposed. 

 Careful collecting in future in areas of outcropping granite massifs and inselbergs 

 (which seem to provide a favoured environment, even if the streams are only inter- 

 mittently flowing) may link the distribution still more closely to that of the Palae- 

 arctic, for there are no reasons at all for supposing that Prosimuliine forms in 

 southern Africa had a southern route of entry : they are more likely remnants 



