The respiratory depressant capacities of intravenous nalbuphine, a potent analgesic of the narcotic antagonist type, and of morphine were compared in 23 healthy subjects using displacement of CO2 response by a steady-state method as the index of respiratory depression. At equianalgesic doses of 10 mg/70 kg, respiratory depression by nalbuphine was equal to that by morphine. When increments of 10 mg/70 kg were given hourly the dose-effect curve for respiratory depression by nalbuphine was flatter than that of morphine, and maximum respiratory depression occurred after 30 mg/70 kg. In a separate study of 10 subjects nalbuphine was administered in 10 mg/70 kg increments to a total dose of 60 mg/70 kg; doses in excess of 30 mg/70 kg failed to increase respiratory depression beyond that induced by morphine 20 mg/70 kg. A ceiling effect for respiratory depression previously known to exist only for nalorphine was thereby demonstrated to apply to nalbuphine. The respiratory depression of nalbuphine was readily antagonized by naloxone 0.4 mg, nalorphine 10 mg, and levallorphan 1.0 mg. Subjective effects of nalbuphine were milder than those of morphine, and dysphoria suggestive of the psychotomimetic effects of narcotic antagonists was reported only 4 times in 24 subject exposures. The ceiling effect for respiratory depression by nalbuphine provides a unique safety factor among potent analgesics.