The role of delta opioid receptors in the self-administration and discrimination of cocaine by rhesus monkeys was evaluated using the delta opioid antagonist naltrindole and the delta opioid agonist BW373U86 [(+/-)-4-((alpha R*)-alpha-((2S*,5R*)-4-Allyl-2,5-dimethyl-1- piperazinyl)-3-hydroxybenzyl)-N,N-diethylbenzamide dihydrochloride]. In the self-administration studies, monkeys were trained to respond for food (1 g banana pellets) or cocaine (0.032 mg/kg/injection, i.v.) under a second-order FR4(VR16:S) schedule of reinforcement during multiple daily food and drug sessions. Substitution of saline or other unit doses of cocaine (0.001-0.1 mg/kg/injection) for the maintenance dose of cocaine yielded typical inverted-U shaped dose-effect curves relating unit dose of cocaine to the number of saline or drug injections per day. The unit dose of cocaine available during drug sessions had little effect on food-maintained responding. In the drug discrimination studies, monkeys were trained to discriminate 0.4 mg/kg cocaine (i.m.) from saline in a two-lever, food-reinforced drug discrimination task. Cumulative dosing with cocaine (0.004-1.3 mg/kg) produced a dose-dependent generalization to the training dose of cocaine without consistently altering response rates. Initial experiments with naltrindole (0.1-3.2 mg/kg, i.v. or i.m.) produced effects in some monkeys suggesting that naltrindole may have antagonized the effects of cocaine under both procedures. In the drug self-administration studies, treatment with at least one dose of naltrindole for 10 consecutive days decreased self-administration of a unit dose of cocaine at the peak of the ascending limb of the cocaine dose-effect curve (0.01 mg/kg/injection) in three of four monkeys. In the drug discrimination studies, treatment with at least one dose of naltrindole shifted the cocaine dose-effect curve 1/2 log unit to the right in two of four monkeys. However, naltrindole did not alter the effects of cocaine in all monkeys under either procedure. Furthermore, in the monkeys in which naltrindole was effective, the cocaine-antagonist actions of naltrindole were not replicable and were not monotonically related to dose. When the delta agonist BW373U86 (0.001-0.032 mg/kg/inj, i.v.) was substituted for cocaine in one monkey in which naltrindole decreased cocaine self-administration, BW373186 was not self-administered. In addition, BW373U86 (0.0032-0.32 mg/kg, i.m.) neither generalized to cocaine nor potentiated the effects of cocaine in the drug discrimination procedure. These results suggest that delta opioid receptors play, at best, a minor and inconsistent role in the reinforcing and discriminative stimulus effects of cocaine in rhesus monkeys.