Catalepsy is an important symptom because, although it occurred in slightly less than a third of our cases, it seems to be a peculiarity of the stupor reaction found but rarely in other benign psychoses. It seems never to occur without there being some evidence of mental activity, and, consequently, we are forced to conclude that it is of mental rather than of physical origin. Just what it means psychically it is impossible to state without much more extended observations. We conjecture tentatively, however, that the retention of fixed positions is in part merely a phenomenon of perseveration, and in part an acceptance of what the patient takes to be a command from the examiner, and sometimes a distorted form of muscular resistiveness. The intellectual processes suffer more seriously in stupor than in any other form of manic-depressive insanity. Not only do the deep stupors betray no evidence of mentation during the acme of the psychosis, but retrospectively they usually speak of their minds being a blank. Incompleteness and slowness of intellectual operations are highly characteristic features of the partial stupors and of the incubation period of the more profound reactions. The features of this defect are a difficulty in grasping the nature of the environment, a slowness in elaborating what impressions are received, with resulting disorientation, poor performance of any set tests and incomplete memory for external events when recovery has taken place. At times the thinking disorder may develop with great suddenness or improve as quickly, and a tendency to isolated evidences of mental acuity is another example of the inconsistency which is so highly characteristic of stupor. We should note, however, that these sporadic exhibitions of mentality are always associated with brief emotional awakening. When we turn to examine the fragmentary utterances of stupor patients, we are surprised by the narrowness and uniformity of the ideational content. It seems to be confined to thoughts of death or closely related conceptions. Thirty-five out of thirty-six consecutive cases at one time or another referred literally to death. It is commonest during the onset, as all but five of these patients spoke of it during the incubation of their psychoses. Hence we conclude that death ideas and stupor are consecutive phenomena in the same fundamental process. As two-thirds of the series interrupted the stupor to speak of death or to attempt suicide, we assume that this relationship persists. Only a quarter gave any retrospective account of these fancies, so we presume that their psychotic experiences were repressed with recovery.