Thirty patients undergoing thoracotomy for lung resection were entered in a randomized, double-blind trial comparing the effects of epidural (E) versus intravenous (iv) morphine on postoperative pain and pulmonary function. Postoperatively the patients were given repeated doses of either 5.0 mg of morphine epidurally or 0.05-0.07 mg/kg morphine intravenously until there were no further spontaneous complaints of pain. Two, 8, and 24 h postoperatively, the following indices were measured: linear analogue pain score, somnolence score, vital signs, arterial PaO2, PaCO2, and pH, forced vital capacity (FVC), forced expiratory volume in the first second (FEV1), and peak expiratory flow rate (PEFR). Patients receiving epidural morphine had significantly less pain at 2 h (P less than 0.01) and 8 h (P less than 0.004) postoperatively. There was no difference in vital signs except for significantly slower respiratory rates at 2 h (P less than 0.04), 8 h (P less than 0.02) and 24 h (P less than 0.01) in the epidural group. No significant differences occurred in the somnolence scores or blood-gas measurements, which were within normal limits.(ABSTRACT TRUNCATED AT 250 WORDS)