The hypothesis that a long-term increase of behavioural responses in snails (over a period of days) might be due to environmental conditioning was examined. Training consisted of delivering electric shocks non-contingently with test stimuli twice per day for 5 days to freely moving snails on a ball floating in water. After training, a significant difference in amplitude of a withdrawal reaction to tactile test stimulation appeared between shocked and control snails. Responses were significantly facilitated in shocked animals for up to 12 days after training, but only if the animals were tested in the environment used for training. Testing of the same groups of animals crawling freely on the glass lid of a tank in which they lived between experimental sessions revealed no difference in responses to the same stimuli between shocked and control snails. Injection of the neurotoxin 5,7-dihydroxytryptamine, which selectively impairs serotonergic cells, eliminated the differences between shocked and control animals. Changing the pH of the water in which the ball floated, by addition of citric acid, led to a significant selective increase of responsiveness in snails sensitized in this environment relative to the responsiveness of the same snails with normal water in the tank. The results suggest that the long-term sensitization of withdrawal reactions observed is at least in part a manifestation of an associative process, namely environmental conditioning.