Olfactory responses of the omnivorous generalist predator Dicyphus hesperus to plant and prey odours

This paper describes a method for studying the neighbourhood (dispersal) movement of pest insects that overcomes the major problems associated with release/recapture experiments. The method has been developed using the carrot fly, Psila rosae (Fab.) (Diptera: Psilidae), as the experimental insect. It involves building up a large population of insects in an area free of other host plants, and then, as the insects disperse in the spring, monitoring the fall‐off in numbers of insects with increasing distance. Estimates of the fall‐off in numbers of the carrot fly were obtained by using sticky traps to monitor changes in fly numbers in 11 small plots of carrots drilled 130–1300 m away from the site at which the flies emerged (emergence site) in the spring. A strong linear relationship (r² = 0.96) was obtained between the log₁₀ numbers of flies caught in each plot and the log₁₀ distance the plot was from the emergence site. Few flies were caught more than 1 km from the emergence site. A linear relationship (r² = 0.86) was also obtained between the date when 50% of the flies were caught in each plot and the distance the plot was from the emergence site. The date of 50% capture was delayed by 1 day for each 100 m the given plot was away from the emergence site. This suggests that when carrot flies move to find new crops, the population moves about 100 m day^?1. About one million flies would have to be released in conventional release/recapture experiments to obtain data as robust as those collected using the current method. The major breakthrough of this method is that dispersal can be estimated for an insect species that, although a major pest of field crops of carrots and parsnips, is exceptionally difficult to rear, even in relatively small numbers, under controlled‐environment conditions.