The mechanisms of the antinociceptive, depressor and bradycardic responses produced by intravenous (i.v.) administration of morphine were examined in rats lighyly anesthetized with pentobarbital sodium. Intravenous administration of 0.1, 0.25, 0.5, 1.0 or 2.5 mg/kg of morphine produced dose-dependent inhibition of the nociceptive tail flick (TF) reflex, hypotension, and bradycardia. Bilateral cervical vagotomy (CVAG) significantly attenuated the antinociception produced by i.v. morphine and the degree of attenuation was inversely related to drug dose. CVAG had no effect on the depressor response produced by lesser doses of morphine (0.1 or 0.5 mg/kg), but at greater doses converted the depressor response into either a pressor response (1.0 mg/kg) or an initial pressor response followed by a depressor response (2.5 mg/kg). Morphine-induced bradycardia was blocked by CVAG at all drug doses tested (0.1, 0.5, 1.0 and 2.5 mg/kg). In selective tests of either 0.5 or 2.5 mg/kg of i.v. morphine, prior administration of the peripherally acting opiod receptor antagonist naloxone methobromide (NMB) attenuated the antinociception to the same degree as CVAG. NMB also completely blocked the depressor and bradycardic responses of these doses of morphine. Bilateral subdiaphragmatic vagotomy (SDVAG) resulted in a marginal attenuation of antinociception at 0.5 mg/kg but not 2.5 mg/kg of morphine, and the attenuation produced by SDVAG was delayed in onset following morphine administration relative to that produced by CVAG. Bilateral sino-aortic deafferentation (SAD) had no significant effect on the antinociception in tests with 0.5 mg/kg of morphine. SDVAG and SAD had little effect on cardiovascular responses produced by these doses of morphine. The spinal antinociceptive systems activated by vagal afferents following i.v. morphine administration were characterized with the 0.5 mg/kg dose. Spinal cold block significantly antagonized the antinociception, hypotension and bradycardia produced by this dose of morphine. Intrathecal administration of naloxone (1.5, 15 or 30 μg) significantly antagonized the antinociception compared to saline controls, whereas intrathecal administration of methysergide (30 μg), phentolamine (30 μg), or the combination of methysergide with phentolamine (30 μg each) had no significant effect on the antinociception. These intrathecal doses of naloxone also antagonized the depressor and bradycardic responses produced by morphine. However, the antagonism produced by 1.5 μg of intrathecal naloxone was not due to spread to the systemic circulation, since i.v. administration of 1.5 μg of naloxone did not significantly affect either the antinociceptive or cardiovascular responses produced by morphine. These findings indicate that vagal afferents play a significant role in the antinociception produced by i.v. administration of morphine. Lesser doses of i.v. morphine produce antinociception primarily by activating vagal afferents that engage a spinopetal inhibitory system using spinal opioids. At greater doses of i.v. morphine, the vagal influence is masked to a large extent, presumably by direct spinal/supraspinal actions of morphine. Depressor responses produced by lesser doses of i.v. morphine are not due to activation of vagal afferents, although at greater doses of morphine, the vagi are clearly important for the expression of morphine-induced changes in arterial blood pressure. Bradycardic responses produced by i.v. morphine appear to reflect activation of vagal afferents. © 1991.