92 patients suffering from endogenous depression who had been treated with 150 mg of imipramine daily (in the course of several double blind studies), were investigated for predictive criteria of clinical response to imipramine: 1. Measured by means of 12 scales of the AMP-system, responders and non-responders to imipramine showed no difference in their psychopathological base level. 2. There was no difference in the course of treatment with regard to success between men and women. 35 men and 57 women have been investigated by means of a two way analysis of covariance for repeated measurements. Thus, women reacted just as well to imipramine as men. 3. Of 19 anamnestic and social variables 3 only showed significant correlation with the effect of imipramine treatment within 20 days: – patients whose first episode has occurred long ago react better to imipramine treatment than “new” patients, – patients whose episode has been precipitated by psychological factors react more readily than if no precipitation is apparent, – patients whose youth was spent in a broken home atmosphere respond badly. 4. When interpreting our results it must be taken into consideration that they deal exclusively with the prognosis of pharmacotherapeutic success or otherwise with imipramine in freshly admitted hospitalized depressives. Our results are predominantly negative, i.e. for all practical purpose we have not found any clinical variables of prognostic value. The few positive findings are, for the time being, not more than hypotheses which ought to be subjected to cross validation.