From a clinical standpoint the problems of the etiology and treatment of pellagra remain unsolved, and the mortality rates in general hospitals range between 20 and 50 per cent. In an effort to determine the most promising fields for further inquiry into the etiology of pellagra and also to determine the causes of such persistently high death rates, the records of 440 patients with this disease were analyzed. When patients were admitted to the hospital more than once, the new data were added to those previously recorded; no patient was counted twice. The interval over which the observations were made covered the sixteen years between Jan. 1, 1919, and Dec. 31, 1934. Several reviews of pellagra have appeared during recent years, notably those of Turner,1 Boggs and Padget,2 Mulholland and King3 and Garrett.4 It was felt that our series, being considerably larger than those previously reviewed