The Jour

can in (Cooter, 1989; Rainey, 1989). Within a given region of the swarm, locusts generally fly with their bodies parallel, and they can be separated by as little as 10 cm (Gunn et al. 1948; Rainey, 1976, 1989; Baker et al. 1984). Each locust’s wing-beat produces a rhythmic flow field that extends behind its body for a distance of more than 30 cm (Kutsch et al. 1994), raising the possibility that others flying within this flow field will encounter turbulence inimical to stable flight. In an earlier study, pairs of locusts flying in tandem in a wind tunnel were shown to couple their wing-beats intermittently, sometimes for periods of up to 1min. The major sensory cue responsible for this wing-beat coupling appears to be the rhythmic flow field produced by the front locust’s wing-beat and detected by the rear locust’s cephalic wind-receptive hairs (Kutsch et al. 1994). Three procedures, deafening both locusts, blinding the rear locust combined with fixing its antennae in place and darkening the room, failed to prevent wing-beat; ho er e responsible for coupling arose from measurements of the specific phase values at which coupling took place at different distances (Kutsch et al. 1994). Flying in a wind tunnel at 2m s21, a locust with a wing-beat frequency of 19–20Hz leaves behind a rhythmic flow field with a wavelength of approximately 7.5 cm. A tethered locust beating its wings at 20Hz (and thus producing 20 waves of its flow field over 2m) should produce a wavelength of 10 cm. However, the forces produced by tethered locusts are often much lower than those produced during free flight (Krüppel and Gewecke, 1985; Zarnack and Wortmann, 1989; Robertson and Johnson, 1993). For instance, lift is typically only 50–60 % of body mass. If thrust is also lower when the locust is tethered, its wing-beat would slow down the air flow of the wind tunnel and thus produce a wavelength shorter than the expected 10 cm, which could account for the measured wavelength of 7.5 cm (Kutsch et al. 1994)