No abstract

A laboratory bioassay was developed to allow blind testing of panels of odours for their effect on target-orientated behaviour of tsetse (Glossina spp.) in the laboratory. Landing responses of G. m. morsitans Westwood on black and blue targets were increased up to four-fold in the presence of carbon dioxide, but no significant effect of any other odours could be demonstrated. 2-methoxy phenol gave an apparent increase in landing behaviour in the laboratory, but this substance diminished rather than increased landing of G. pallidipes Austen and G. m. morsitans on targets in the field, as well as repelling tsetse from the target vicinity. Black, blue and red targets elicited strong landing behaviour in male G. m. morsitans in the laboratory, and white and yellow targets elicited little or none, whether carbon dioxide was present or not. In the field, high ultraviolet reflectivity increased landing behaviour of G. pallidipes, but only when there was a proportion of transmitted light through the target. Ultraviolet reflectivity always reduced overall attraction of tsetse to a single-coloured target. Two-coloured targets incorporating ultraviolet-reflecting white cloth obtained strong landing on the white panels, but caught fewer flies overall than all-black, or blue-and-black targets.

Glossina longipennis Corti was studied in Galana Ranch, Kenya over a four year period, in two areas (Tank E and Lali) where the species was abundant and other species were absent or scarce. There was active transmission of trypanosomiasis to cattle in both areas, the parasite species being Trypanosoma vivax Ziemann and T. congolense Broden. Mean infection rates of the G. longipennis were 1.1% and 0. 55% for T. vivax and T. congolense respectively at Tank E, and 0.88% and 0.15% at Lali. Experimental transmission studies showed that cattle in fly-proof enclosures challenged with wild G. longipennis collected from Galana became infected with both trypanosome species. A tsetse control operation in one area (Tank E) using targets impregnated with deltamethrin in an oil formulation reduced the population of G. longipennis by 98% over one year, despite evidence of re-invasion. Populations of G. longipennis in the other area (Lali) were relatively stable over the whole study period. The effect of tsetse control on the incidence of cattle trypanosomiasis at Tank E was less clear than that on tsetse numbers, probably due to the lack of a sustained reduction in tsetse numbers. However, a significant relationship was demonstrated between fortnightly incidence measurements and electric net catches of G. longipennis at Tank E. A further significant predictor of incidence was rainfall in the previous four to seven weeks. This study confirms the importance of G. longipennis as a vector of bovine trypanosomiasis in areas where it is the predominant tsetse present.