SIMULIIDAE OF AFRICA 13 



Simulium s.l. lie mostly with the Holarctic Regions ; certainly none of these sub- 

 genera suggests any close relationship with the Neotropical fauna, with the exception 

 of some possible affinity between Pomeroyellum and Psilopelmia. The remaining 

 trio of subgenera, Anasolen, Freemanellum and the Malagasy Xeno simulium are 

 much more problematical since they have few if any features that strongly suggest 

 affinity with the Holarctic fauna ; on the other hand they have conspicuous 

 resemblances to some South American forms, and there may — though this must be 

 entirely conjectural — with this group of subgenera be a zoogeographical connection 

 between Ethiopian Africa and the Neotropical Region. 



Finally, it may be helpful in this section to draw attention to some characters 

 which are either always present, or alternatively which never occur, in the Simuliidae 

 of the Ethiopian Region : — 



<J head always holoptic (none dichoptic, as are a few Neotropical forms). 



Antenna always 11-segmented. 



Head always without postocular bulla. 



Wing- vein C112 never straight (cf. South American Gigantodax Enderlein). 



Cocoon present, always covering at least the pupal abdomen. 



Larvae always with cephalic fans. 



Larval anal sclerite always present, X-shaped. 



SIMULIIDAE Newman 

 SIMULIITES Newman, 1834, Ent. Mag. 2 : 387. Type-genus : Simulium Latreille, 1802. 



A definition of the Simuliidae among the families of Diptera, and a more detailed 

 consideration with diagnoses of the subfamilies, will be given in a forthcoming paper 

 on the classification of world forms, but it may be briefly mentioned here that — in 

 agreement with Smart (1945) — it is considered that only two subfamilies should be 

 recognized. The first is the Parasimuliinae, containing only the single remarkable 

 genus Parasimulium Malloch from western United States, and the other — the 

 Simuliinae — comprising all other known black-flies. The subfamily Gymnopaidinae, 

 which Rubzov (1956, 1959-1964) proposed and treated as distinct from the 

 Simuliinae for the Holarctic genera Gymnopais Stone and Twinnia Stone & 

 Jamnback, does not seem to me to be sufficiently distinct from other simuliids to 

 justify subfamily rank, despite the remarkable larvae (in which all instars lack 

 cephalic fans and the abdomen has a Y-shaped in place of the normal X-shaped anal 

 sclerite) ; in fact, as Shewell (1958) points out, Twinnia is so close to Prosimulium 

 Roubaud in many of its characters that there is doubt as to whether separate generic 

 status is justified. Certainly Twinnia interconnects Gymnopais and Prosimulium, 

 and makes it impossible justifiably to separate the first two off as a distinct sub- 

 family : hence Gymnopais and Twinnia are here regarded as forming part of the 

 tribe Prosimuliini (the more primitive one of the two tribes into which the subfamily 

 Simuliinae is here divided). Neither Gymnopais nor Twinnia occurs in Africa and 

 these genera are not considered further at this time. 



Two tribes are here recognized in the Simuliinae, the Prosimuliini containing the 



