Published online by Cambridge University Press: 09 March 2007
An incomplete ring of electric nets was evaluated as a means of estimating trap efficiency for Glossina spp. This methodology assumes flies approach the trap directly, and then enter, or leave directly in random directions. These results showed that the ratio of the number of flies intercepted on the outside of the electric nets to the number on the inside was lower than predicted by this single-approach behavioural model. Moreover, an incomplete ring of nets around a trap reduced trap catch more than the model predicted. These inconsistencies were greater early in the day, and greater for females than for males. It is suggested that flies may make several approaches to a trap before entering or departing. This mixes arriving and departing flies on each side of the electric nets. Use of a complete ring of nets around a trap to estimate trap efficiency entails fewer behavioural assumptions. Catches at a complete ring around a trap were compared to catches in a trap without nets, replicated in a cross-over design. The efficiency of an odour baited NG2G trap was estimated to be 58% for males and 37% for females. Biconical traps were much less efficient. Both trap types were less efficient in the early morning, suggesting entry response is temperature dependent. The NG2G trap was more efficient for non-teneral nulliparous females than for other ages. For both trap types there was little difference between mean fat content of approaching and trapped males, but the mean fat content of trapped females was lower than that of approaching females.