According to the conditioned anorexia hypothesis, conditioned taste aversions occur when flavour stimuli are classically conditioned to the anorexigenic or hypodipsic effects of drugs. The effects on water intake of a range of doses of amphetamine and of several related compounds have therefore been examined in an attempt to correlate their known potentices in tate aversion experiments with their hypodipsic potencies (+)-Amphetamine was more potent than (-)-amphetamine in suppressing water intake but under similar experimental conditions, the isomers were equipotent in the conditioning of taste aversions. Methamphetamine and p-chloromethamphetamine were equipotent in suppressing water intake, but the latter was a more potent agent for conditioning taste aversions. Furthermore, fenfluramine produced taste aversions at doses well below those which suppressed water intake. It was concluded that the ability of the drugs to induce taste aversion was not related to their unconditioned, hypodipsic effects. However, it was confirmed that when drugs with different durations of action are compared for anorexic or hypodipsic potency, the outcome can be greatly influenced by the time over which measurements are made.