The habitat preferences and activity patterns of Glossina swynnertoni Austen were studied using Siamese traps, electrified screens and vehicle patrols in Aitong, southwestern Kenya in three habitats: large thicket, wooded grassland and an Acacia community. Except in the Acacia community, G. swynnertoni occurred in association with Glossina pallidipes Austen. Apparent density varied significantly among the three habitats, being highest in the Acacia community, intermediate in wooded grassland and lowest in large thickets. Density also varied seasonally, with higher densities occurring in the rainy season than in the dry season in all three areas. Vehicle patrol was a more effective and rapid method for sampling G. swynnertoni than either Siamese traps or electrified screens. However, vehicle patrol catches were strongly biased in favour of males (4:1). Diurnal activity in the Acacia community was unimodal for both sexes, with peaks of activity occurring at 1100–1200 h for males and at 1400–1500 h for females. Both sexes remained active in the afternoon but activity declined rapidly towards dusk (1700–1800 h).

Various trap designs were tested in Congo in order to assess their cost vs. effectiveness for the control of human trypanosomiasis vectors. These preliminary trials indicated that it is necessary to compare the different trap designs in several African countries. Furthermore, studies on the attractiveness of locally available fabrics must be carried out in order to make traps taking into account the sizes of locally available fabrics. At present, the ‘Vavoua’ trap is sail the most efficient but further investigations need to be carried out to reduce the cost of new traps without significantly affecting their efficiency. The ‘black/blue/black’ screen can be used successfully by rural communities in Congo for tsetse control.

Several tsetse control campaigns have been carried out in Senegal. During the 1970s a ground spraying campaign was carried out in the Niayes region (relic of Guinean forest infested with Glossina palpalis gambiensis, Vanderplank, 1949). At the beginning of the 1980s, the chemical control was strengthened by use of screens and biconical traps. As a result no fly has been caught in the region since 1983.

An assessment of the second method, done in Kolda (south of the country), has shown a reduction rate of 98% against G. palpalis gambiensis.

Four types of trap, yellow water, yellow sticky board, bottle with 20% protein hydrolysate, and funnel with 10% casein hydrolysate, were field-tested to control Liriomyza huidobrensis (Blanchard) in the coastal region of Peru. The yellow sticky boards were the most effective in trap capture with a mean of 1193.92 L. huidobrensis per week. The male:female capture ratio was 16:1.

Five parasitoid species, Halticoptera palellana, Chrysocharis phytomyzae, Chrysocharis sp., Diglyphus sp. and Opius sp., were also trapped. A significant negative correlation (r = −0.607 at P < 0.05) was observed for female leafminer capture vs temperature. Wind velocity was positively correlated with parasitoid capture (r = 0.509 at P < 0.05). The population of adult L. huidobrensis was significantly lower (P < 0.05) in plots with yellow sticky boards. Yields in plots with yellow sticky boards and those treated with insecticides were similar and significantly higher (36%) than check plots.

Field studies were done to elucidate the effectiveness of buffalo urine as a tsetse attractant under a variety of test conditions. When traps were baited with the test sample of urine placed in large-necked dispensers allowing evaporation of 2 to 2.9 ml/hr, they performed better than traps baited with similar samples placed in narrow-necked dispensers with evaporation rate of 0.2 to 0.4 ml/hr. Buffalo urine was compared with samples of cow urine: the catch size increased 1.8 times in traps with cow urine when compared to that of the control trap, whereas buffalo urine was 9.6 times more attractive than control. Acetone added to the test sample, even in very small quantity, boosted the effectiveness, increasing the catch size 2.8 times more. Urine samples that had stayed on the laboratory shelves for some weeks or months were more effective than freshly obtained ones.