Previous investigations have demonstrated the ability of atropine and chlorpromazine pretreatment to increase intravenous cocaine self-administration in rhesus monkeys. These results were interpreted as resulting from either an antagonism of cocaine's reinforcing effect or an effect of cocaine which may interfere with its self-administration. The generality of the postulated drug antagonism was tested in monkeys exhibiting stable FR10 food-reinforced behavior. Intramuscular pretreatment with 0.25-4.0 mg/kg of chlorpromazine significantly depressed this behavior as did pretreatment with 0.25-2.0 mg/kg of atropine. These same dosages had been shown to increase cocaine self-administration. Fifteen-minute influsions of cocaine (1.0-10 mg/kg), which began 15 min after session onset, also resulted in a depression of food reinforced behavior. When chlorpromazine pretreatment and cocaine infusion were combined, the decrement in food reinforced responding was lessened. However, when cocaine infusion and atropine pretreatment occurred concurrently, the depression in responding was increased. These data indicate that chlorpromazine may nonspecifically antagonize the behavioral effects of cocaine; whereas, atropine may enhance these effects.