The effects of intrathecal (IT) administration of two doses of morphine (Group 1: 2 mg, n = 9; Group 2: 5 mg, n = 10) were studied in 19 patients after upper abdominal surgery. The ventilatory variables and occlusion pressure (P0.1) were recorded during room air breathing and during CO2 rebreathing tests prior to surgery, 24 h after surgery before IT morphine (n = 12), and 3, 5, 7, 11, and 24 h after injection. During room air breathing, minute ventilation (VE) did not change significantly in Group 1 and decreased significantly 3, 5, 7, and 11 h after injection in Group 2. During the rebreathing tests, there was a significant shift to the right of the ventilatory response to CO2 in both groups. The peak of the ventilatory depression was delayed, occurring 7 h and 11 h postinjection in Groups 1 and 2, respectively. Two patients in Group 2 developed clinically significant ventilatory depression. The shallow breathing observed after surgery was not changed after analgesia. In group 2, 5, mg IT morphine was responsible for a significant decrease in f60 (respiratory frequency for a PETCO2 of 60 mmHg). P0.1 increased markedly after surgery during both room air breathing and the rebreathing tests. After IT morphine, compared with the postoperative preanalgesic values, P0.1(60) (P0.1 at a PETCO2 of 60 mmHg) did not change in Group 1 and decreased significantly in Group 2. It is concluded that IT morphine is responsible for a ventilatory depression that is delayed and seems to be dose related and that analgesia does not abolish the shallow breathing observed after upper abdominal surgery.