The 24-hour motor activity pattern was evaluated in 26 inpatients with major depression at treatment onset and after 4 weeks of antidepressant therapy. Clinical state, depression, and psychomotor retardation, as well as motor activity level and circadian rhythm, were simultaneously assessed. Treatment responders and nonresponders were also considered. Diurnal hypoactivity and reduced 24-hour rhythm amplitude were found at treatment onset. Activity level increased significantly on discharge. The rest-activity cycle for each depressed patient fit a cosine function of 24-hour periodicity. Data tended to show no phase shift but a large intragroup phase variability. Preliminary findings of a negative correlation between basic activity level and clinical improvement, and a trend toward responders having a lower activity level than nonresponders, suggest that activity could be used to predict therapeutic response.