XVI] CONDITIONS AFFECTING TRANSMISSION t 251 



becoming infective than if they are starved after a meal of 

 infected blood. 



Probably temperature exercises a greater influence than any 

 other known factor on the development of trypanosomes in the 

 tsetse-fly. Kinghorn and Yorke have shewn that T. rhodesiense 

 can only complete development in its invertebrate host, 

 G. morsitans, at a temperature of at least 75 to 80 F. Similarly 

 Kleine and Fischer on Lake Victoria were quite unable to 

 transmit T. gambiense by means of G. morsitans, whilst on the 

 warmer shores of Tanganyika, Taute, employing the same 

 species, succeeded without much difficulty (vide p. 306). 



The observations on the natural infections of tsetse-flies 

 have shewn that flies of different species frequenting the same 

 district are not infected in the same proportions, nor in the 

 same manner. Moreover these proportions, even for the same 

 species, vary in different localities, as for example in West 

 Africa. Thus Bouet and Roubaud found that in Lower 

 Dahomey T. cazalboui predominated in G. longipalpis and 

 palpalis, T. dimorphon in longipalpis and tachinoides, and 

 T. pecaudi in longipalpis. On the other hand, in Upper 

 Dahomey, T. pecaudi was most predominant in G. morsitans, 

 and in Casamance T. dimorphon in morsitans. The number of 

 G. palpalis becoming infected with T. cazalboui when fed on an 

 animal suffering from this infection, was 40 per cent, in 

 Middle Dahomey, and nil in Upper Casamance, although 

 repeated attempts were made to infect the flies. These experi- 

 mental results were confirmed by an examination of natural 

 infections. In Middle Dahomey one fly in thirty was naturally 

 infected with T. cazalboui, whilst in Upper Casamance out of 

 560 flies examined only one shewed infection of the pro- 

 boscis. 



Souma (T. cazalboui) exists in both regions, but it is evident 

 that some other species of tsetse besides palpalis must be 

 responsible for its transmission in Casamance. From these 

 data Roubaud deduces that the receptivity of a given species 

 of Glossina for any particular virus, is not uniform throughout 

 the region of the flies' distribution. In other words a virus is 

 only endemic where there are receptive races of tsetse-flies. 



