A series of experiments was conducted to assess the sensitivity of the acoustic startle response to chronic morphine administration and naloxone-precipitated withdrawal. Rats were implanted with two subcutaneous pellets containing either 75 mg each of morphine or containing only placebo. In experiment 1, withdrawal induced by 0.05-0.2 mg/kg naloxone dose-dependently decreased the magnitude of the startle response. Physical dependence was confirmed by a naloxone-induced acute weight loss seen in morphine-implanted rats, but naloxone had no effect on startle or body weight in non-dependent animals. In experiment 2, a modified procedure with fewer trials per session and fewer test days was employed. Naloxone (0.2 mg/kg) given 4-5 days after implantation induced large startle-response decreases in morphine-dependent rats while having no effect in placebo-implanted rats. Post-naloxone saline tests revealed no significant differences in startle between morphine and placebo groups. Startle scores were significantly higher in morphine-implanted rats than in placebo rats during a saline test given 3 days following pellet implantation. In a separate group of animals, however, acute IP injections of morphine from 0.3-10 mg/kg had no significant effect on startle amplitude. The effect of repeated pairings of withdrawal with the startle environment was assessed in experiment 3. Morphine-dependent rats startled significantly less if naloxone injections were given before the startle session than if they were administered 4 h later. Conditioned withdrawal effects, expressed during a final test session when all rats received saline, were observed for the body-weight measure but not for the startle response.(ABSTRACT TRUNCATED AT 250 WORDS)