In the first experiment four food-deprived rats developed high levels of adjunctive water drinking during daily sessions of intermittent food pellet delivery. When the water was removed and a solution of d-amphetamine sulfate (0.01 mg/ml) put in its place, adjunctive drinking was disrupted towards the end of each session although the rats ingested doses of approximately 0.5 mg/kg daily for over 40 sessions. Consumption of the d-amphetamine solution was increased by injections of several doses of alpha-methyl-p-tyrosine (AMPT). In a second experiment injections of d-amphetamine (0.25, 0.5, 1.0, 2.0 mg/kg) were found to reduce adjunctive water consumption in six rats. It was also found that the actions of the two highest doses of d-amphetamine were reduced by pretreatment with a dose of AMPT (100 mg/kg), which itself slightly reduced levels of drinking. These results suggest that, although adjunctive drinking may be a useful technique for inducing rats to self-administer d-amphetamine, the amount of drug consumed is limited by a direct action of the drug on drinking.