Cannabinoid-receptor antagonists have shown some promise as treatments capable of reducing abuse and relapse to a number of abused drugs. In rodents, such effects have been observed with methamphetamine self-administration. However, the effects of cannabinoid receptor antagonists on methamphetamine self-administration and relapse have not been studied in primates. In the present study, rhesus monkeys were trained to respond on a three-component operant schedule. During the first 5-min component, fixed-ratio responses were reinforced by food, during the second 90- or 180-min component fixed-ratio responses were reinforced by i.v. methamphetamine. The third component was identical to the first. There was a 5-min timeout between each component. The effects of the cannabinoid receptor antagonists AM 251 and rimonabant were tested at various doses against self-administration of 3 μg/kg/injection methamphetamine, and 1 mg/kg AM 251 and 0.3 mg/kg rimonabant were tested against the methamphetamine dose-effect function. The 1 mg/kg dose of AM 251 was also tested for its ability to alter reinstatement of extinguished self-administration responding. The cannabinoid receptor antagonist AM 251 was found to reduce methamphetamine self-administration at doses that did not affect food-reinforced responding. The cannabinoid receptor antagonist rimonabant had similar, but less robust effects. AM 251 also prevented reinstatement of extinguished methamphetamine seeking that was induced by re-exposure to a combination of methamphetamine and methamphetamine-associated cues. These results indicate that cannabinoid receptor antagonists might have therapeutic effects for the treatment of methamphetamine dependence.