In three previous communications based on a study of 633 cases of manic-depressive psychosis in the private practice of Dr. Hugh T. Patrick, I1 showed that these cases differed materially from this disorder as described in the literature. These differences were principally the presence of many attacks lasting from only a few hours to a few days, practically equal sex distribution and a tendency to a much later onset, shorter attacks and longer intervals. The existence of these differences was explained by the fact that the literature on manic-depressive psychosis has been developed almost entirely by physicians working in institutions for the insane; they saw only institutional cases, that is, the most malignant forms, and based their descriptions on these. In Dr. Patrick's material there were many cases so mild that hospitalization was not necessary, a type of case that physicians in institutions seldom saw and hence did not