To determine the efficacy of epidural morphine during labor and delivery, 40 healthy parturients who requested epidural analgesia were randomly given either a single injection of morphine sulfate (2 mg, n = 9; 5 mg, n = 10; or 7.5 mg, n = 11) or 0.5% bupivacaine (n = 10). Bupivacaine provided excellent analgesia in all patients. Morphine (2 or 5 mg) did not produce adequate analgesia and needed to be supplemented by local anesthesia. Morphine (7.5 mg) gave satisfactory analgesia in 7 of 11 patients until the end of the first stage of labor, and, in an additional patient, for almost 13 hr, at which time cephalopelvic disproportion was diagnosed. In all patients given epidural morphine, local anesthesia was needed for the second stage of labor if instrumentation or episiotomy was required. The only side effect was pruritus, which occurred in three patients. Epidural morphine did not produce neonatal depression, as evidenced by Apgar scores and neurologic and adaptive capacity scores. We conclude that 7.5 mg of epidural morphine can give satisfactory pain relief during labor but not delivery, and that 2 or 5 mg of morphine is ineffective. Although 0.5% bupivacaine provided better anesthesia, epidural morphine might prove useful for selected patients.