Opioid withdrawal can produce a constellation of physiological and behavioral signs, including an increase in opioid self-administration. Different mechanisms mediate different withdrawal signs, and the present study used pharmacologic tools to assess mechanisms underlying withdrawal-associated increases in opioid reinforcement. Five rhesus monkeys were rendered heroin dependent via daily 21-h heroin self-administration sessions. One hour after each heroin self-administration session, monkeys chose between heroin (0-0.1 mg/kg per injection) and food (1 g pellets) during 2-h choice sessions. Under these conditions, heroin maintained a dose-dependent increase in heroin choice, such that monkeys responded primarily for food when low heroin doses were available (0-0.01 mg/kg per injection) and primarily for heroin when higher heroin doses were available (0.032-0.1 mg/kg per injection). Periods of spontaneous withdrawal were intermittently introduced by omitting one 21-h heroin self-administration session, and test drugs were administered during these withdrawal periods. Untreated withdrawal robustly increased heroin choice during choice sessions. Withdrawal-associated increases in heroin choice were completely suppressed by the mu opioid agonist morphine (0.032-0.32 mg/kg/h, i.v.), but not by the alpha-2 noradrenergic agonist clonidine (0.01-0.1 mg/kg/h, i.v.), the dopamine/norepinephrine releaser amphetamine (0.032-0.1 mg/kg/h, i.v.), or the kappa-opioid antagonist 5'-guanidinonaltrindole (1.0 mg/kg, i.m.). The corticotropin-releasing factor 1 antagonist antalarmin (1.0-10 mg/kg per day, i.m.) produced a morphine-like suppression of withdrawal-associated increases in heroin choice in one of three monkeys. These results suggest that mechanisms of withdrawal-associated increases in the relative reinforcing efficacy of opioid agonists may be different from mechanisms of many other somatic, mood-related, and motivational signs of opioid withdrawal.