Rates of progress up odour plumes by tsetse flies: a mark‐release video study of the timing of odour source location by Glossina pallidipes

Abstract

Abstract. The arrival of individually marked Glossina pallidipes Austen at a host odour source after their video‐timed release from 30–75 m downwind was measured in the field in Zimbabwe. In the absence of odour, the proportion recaptured was <2% (= ‐ random expectation); when synthetic ox odour was released, the probability of recapture at the source increased with proximity of release, from 6% at 75 m to 21% at 30 m (about twice this number arrived within ∼2 m of the source). There were two distinct distributions of recaptures: a ‘fast’ cohort which found the source within 40 s, and a ‘slow’ cohort which took from one to >20 min, with ∼50% of the flies in each cohort. The fastest flies probably reached the source in a single, mainly straight flight from take‐off, at an overall average (straight line) displacement speed of 2.8‐4.5 ms‐1 (i.e. close to the preferred flight speed of ∼5 m s‐1). The flies apparently maintained their ground speed largely independent of the wind speed they headed into. The ‘slow’ cohort had a constant probability of arrival at the source, presumably after losing and re‐contacting the plume, and after having stopped at least once on the way. There were no marked correlations with wind parameters, although the probability of recapture increased slightly with the directness of the wind from the source, and the probability of ‘slow’ flight increased slightly with wind speed. It is inferred that a repeated sequence of anemotactic ‘aim‐then‐shoot’ orientation at take‐off plus optomotor‐steered in‐flight correction of direction is used as a form of biassed random walk to bring the flies close to the odour source, rather than the use of moth‐type anemotactic zigzagging.