The sanatorium course of 200 patients was studied for comparison with published series and for information concerning the effect of prior corticosteroid treatment. Seventy-four patients achieved complete remission or major improvement. Fifty patients advanced from severe or complete functional incapacity to minor incapacity or full functional ability. These results are superior to those reported for comparable groups. Prior administration of corticosteroids was associated both with disease duration (known to be importantly related to severity and prognosis of rheumatoid arthritis) and with the age of patients on admission. Steroid toxicity contributed significantly to the numbers of patients making unsatisfactory progress. There was no evidence that corticosteroids had any extended benefit on the course of rheumatoid arthritis.