In a randomised double-blind trial, postoperative analgesia and side effects of intrathecal morphine 0.1 mg and intrathecal diamorphine 0.25 mg were compared. Sixty women were randomised to receive intrathecal injection of 12.5 mg hyperbaric bupivacaine and 12.5 microg fentanyl with either morphine 0.1 mg (group M), or diamorphine 0.25 mg (group D). All women received 100 mg diclofenac rectally at the end of surgery and were given intravenous morphine via a patient controlled analgesia (PCA) system. Pain, PCA morphine usage and side effects were assessed at 2, 4, 8 and 24 h after spinal anaesthesia. The two groups were comparable for quality of intraoperative analgesia. There were no significant differences between the groups in time to first PCA demand, morphine consumption or oral analgesic use in each time period. Significantly more patients in group M were nauseated at 4 h whilst at 24 h nausea was significantly worse in group D. There was no significant difference between the groups in the number of women vomiting in the 24-h period. The two groups were comparable for pruritus and drowsiness. We conclude that 0.25 mg subarachnoid diamorphine is a suitable alternative to 0.1 mg morphine for post caesarean section analgesia.