We characterized single upwind surges of flying male Helothis virescens moths in response to individual strands of pheromone generated experimentally in a wind tunnel. We then showed how this surge functions in this species as a basic 13.4-cm, 0.38-sec-long building block that is strung together repeatedly during typical male upwind flight in a normal pheromone plume. The template for a single iteration, complete with crosswind casting both before and after the straighter upwind surging portion, was exhibited by males flying upwind to pheromone and experiencing ifiament contactsjust frequently enough to produce successful upwind flight to the source, as hypothesized by an earlier model. Also as predicted, with more frequent filament contact by males, only the straightest upwind portions of the surges were reiterated, producing direct upwind flight with little crosswind casting. Electroantennogram recordings made from males in free Tfight upwind in a normal point source pheromone plume further support the idea that a high frequency offilaments encountered under the usual pheromone plume conditions promotes only these repeated straight surges. In-flight electroantennogram recordings also showed that when filament contacts cease, the casting, counterturning program begins to be expressed after a latency period of 0.30 sec. Together these results provide a plausible explanation for how male and female moths, and maybe other insects, fly successlly upwind in an odor plume and locate the source of odor, using a surging-casting, phasictonic response to the onset and disappearance of each odor strand.In the quest for understanding how male moths fly upwind and locate females (1-5) there have been suggestions (6, 7) that all odor-mediated flight in moths, including host plant location by females, may be explained by two programs, optomotor anemotaxis (2) and self-steered counterturning (8), that are turned on and modulated by odor fluctuations. It has also been pointed out (9, 10) that many other kinds of insects flying upwind to odor exhibit behavior somewhat similar to moths', and these other insects may also use these same mechanisms in odor-source location. Knowledge about the mechanisms used by moths should therefore be important for understanding the neuroethology of olfaction, the evolution of pheromone and host-plant-insect systems, and the potential application of pheromones in insect control.The physical structure of a pheromone plume created by a point source of odor is not a time-averaged homogeneous cloud. Turbulence causes the plume to break up into strands of odor-laden air (filaments) interspersed with pockets of clean air where little or no odor is present (11,12). The physical intermittency of these filaments is a requirement for male moths to sustain their upwind progress; they will not continue to fly upwind upon entering a homogeneous cloud of pheromone (4, 13) but will if this cloud is alternated with swaths of clean air (14).Recently, results from experiments investigating the highspeed beh...