The kappa-opioid agonists U50488H, bremazocine, and BRL52537, and the mu-opioid agonist morphine were compared in their ability to modify spontaneous motor activity in male NMRI mice. Higher, analgesic doses of the kappa-agonists reduced rearing, motility, and locomotion in nonhabituated mice. These effects, as well as the analgesic action of U50488H, were blocked by the selective kappa-opioid antagonists nor-binaltorphimine and DIPPA. In contrast, lower, subanalgesic doses (1.25 and 2.5 mg/kg for U50488H; 0.15 and 0.075 mg/kg for bremazocine, and 0.1 mg/kg for BRL52537) time dependently increased motor activity. The stimulatory effects of U50488H and bremazocine were not observed in habituated animals and were reduced by dopamine depletion. Surprisingly, the stimulatory effects of U50488H and bremazocine were not blocked by nor-binaltorphimine and DIPPA but they were completely eliminated by naloxone (0.1 mg/kg). The effects of morphine were dose-dependent; an initial limited suppression was followed by increased motility and locomotion (but not rearing) with a peak effect at 20 mg/kg both in habituated and nonhabituated mice. The selective mu-opioid antagonist beta-funaltrexamine blocked morphine-induced motor stimulation and analgesia but failed to affect the analgesic and motor stimulatory effects of U50488H. The results indicate that kappa-opioid agonists interact with different functional subtypes of opioid receptors. A stimulatory, naloxone-sensitive but nor-binaltorphimine- and DIPPA-insensitive subtype of opioid receptor appears to operate only when the dopamine system is tonically active in nonhabituated animals. At higher doses, kappa-agonists produce analgesia and motor suppression, effects mediated by a "classic" (inhibitory) kappa-opioid receptor.