Non-contingent "priming" drug injections and conditioned stimuli associated with drug injections led to reinstatement of responding after a period of extinction. Rats implanted with intravenous catheters were trained to self-administer cocaine 1 mg/kg/injection), and then given daily test sessions consisting of a period of self-administration followed by extinction conditions. Test drug injections or conditioned stimuli were presented during extinction and the latency to the first response and the total number of responses following the treatment were measured. Cocaine injections of 0.5, 1.0, and 2.0 mg/kg restored responding during extinction, regardless of the duration of the extinction period (between 10 min and 180 min) since drug self-administration. Amphetamine, apomorphine, and morphine but not ethanol, heroin, or methohexital reinstated previously cocaine-reinforced responding. Amphetamine, cocaine, and morphine did not increase responding in animals trained to bar press only for food reinforcement, suggesting that the reinstatement effect is specific to drug-reinforced responses. The statement effect is specific to drug-reinforced responses. The final experiment showed that a tone that had been paired with drug infusions acquired a statistically significant tendency to facilitate responding when tested during extinction but this effect disappeared after the first test presentation of the tone.