Evaluation of two counterflow traps for testing behaviour-mediating compounds for the malaria vector Anopheles gambiae s.s. under semi-field conditions in Tanzania

Background Evaluation of mosquito responses towards different trap-bait combinations in field trials is a time-consuming process that can be shortened by experiments in contained semi-field systems. Possible use of the BG Sentinel (BGS) trap to sample Anopheles gambiae s.s. was evaluated. The efficiency of this trap was compared with that of the Mosquito Magnet-X (MM-X) trap, when baited with foot odour alone or combinations of foot odour with carbon dioxide (CO2) or lemongrass as behaviour-modifying cues. Methods Female An. gambiae s.s. were released in an experimental flight arena that was placed in a semi-field system and left overnight. Catch rates for the MM-X and BGS traps were recorded. Data were analysed by fitting a generalized linear model to the (n+1) transformed catches. Results Both types of traps successfully captured mosquitoes with all odour cues used. When the BGS trap was tested against the MM-X trap in a choice assay with foot odour as bait, the BGS trap caught about three times as many mosquitoes as the MM-X trap (P = 0.002). Adding CO2 (500 ml/min) to foot odour increased the number of mosquitoes caught by 268% for the MM-X (P <0.001) and 34% (P = 0.051) for the BGS trap, compared to foot odour alone. When lemongrass leaves were added to foot odour, mosquito catches were reduced by 39% (BGS, P