The analgesic properties of intravenous pethidine and pethidine plus levallorphan were compared in a group of seventy-two fit obstetric patients. The assessment was made by graded pressure on the tibia. A double blind four-point assay plus placebo technique was used. The trial failed to distinguish between the analgesic effect of the narcotic and the placebo. It is concluded that this method of assessment is inapplicable to the pregnant patient. The commonest side dect noted was dizziness but the incidence of side effects was related more to increased dosage of drugs rather than Werences between pethidine and pethidine plus levallorphan.