The outcome of surgical treatment of idiopathic scoliosis in forty-five adults was studied with special attention to pain, function, self-image, and pulmonary function. All of the patients were more than twenty-five years old at operation and had been followed for more than three years. Every patient who was operated on by one of us (J. E. H.) and who met these criteria was evaluated. The magnitude of the curves averaged 66 degrees. Standardized gradations of pain and function showed improvement over-all, but significant impairment remained. There was a reduction in the levels of peak and constant pain, but no change in the frequency of peak pain after operation. The number of patients who were pain-free after surgery was not increased. Functional impairment due to the scoliosis was lessened, and the ability to perform the common activities of daily living was improved, but no important changes in occupation or recreational activity were recorded. Correlations of pain or function, or both, and the changes in either, were found with only two parameters: age at follow-up and physical occupation. Pulmonary function, as measured, did not change. Eighteen (40 per cent) of the patients had a minor complication and ten (20 per cent), a major complication; there was one death, due to pulmonary embolism, of a patient who was excluded from the series. In view of the high rate of complications, the limited gains to be derived from spinal fusion should be assessed and clearly explained to patients before the procedure is undertaken.