This article reviews all the prospective, double-blind controlled studies that have evaluated the prediction of response to imipramine hydrochloride and amitriptyline hydrochloride in depressed patients. Despite widely divergent methodologies, an attempt is made to extract clinically useful conclusions from these data. Critiques of each study and the criteria used in their evaluation are presented, with suggestions for future research included. The predictors of positive response to imipramine and amitriptyline are as follows: upper socioeconomic class, insidious onset, anorexia, weight loss, middle and late insomnia, and psychomotor disturbance. The predictors of poor response are the following: neurotic, hypochondriacal, and hysterical traits, multiple prior episodes, and delusions. Pretreatment urinary 3-methoxy-4-hydroxyphenylglycol (MHPG) levels may someday be useful in predicting to which of these two tricyclic antidepressants a patient will respond.