The effects of continuous i.v. phencyclidine (PCP) infusion and withdrawal on operant behavior were studied in rhesus monkeys. The monkeys were trained to lever press under a fixed-ratio 100 schedule of food presentation. They responded under this schedule during four daily 30-min periods conducted every 6 hr. After at least 5 days of continuous saline infusion through indwelling i.v. catheters, the subjects received 10 days of continuous infusion of 0.05 mg/kg/hr of PCP. During chronic PCP, rates of responding increased above saline control values. When saline was substituted for PCP, responding was suppressed markedly. This suppression of responding occurred within 8 hr of saline substitution and lasted several days. Mild signs of withdrawal were seen within 3 hr after saline substitution and dissipated by 48 hr. These signs included muscle tremors, oculomotor hyperactivity and increased aggressiveness. After responding during saline infusion had returned to control levels, continuous PCP infusion was resumed. Withdrawal effects on behavior decreased in intensity after repeated withdrawals and it was necessary to raise the infusion dose to produce consistent withdrawal disruption of behavior. Withdrawal-induced disruption in responding was reversed by acute pretreatment with PCP (0.01-0.3 mg/kg i.m.) in a dose-related fashion. Presession administration of naloxone (0.1-1.0 mg/kg i.m.) during chronic PCP infusion failed to precipitate withdrawal signs or disrupt operant responding, suggesting that PCP dependence is not of the opioid type. The results of this study indicate that operant behavior is disrupted during withdrawal from chronic PCP and can be used as evidence of behavioral dependence.(ABSTRACT TRUNCATED AT 250 WORDS)