The diagnostic efficiency of the Minnesota Multiphasic Personality Inventory (MMPI), Millon Clinical Multiaxial Inventory-II (MCMI-II), and Rorschach Comprehensive System for detecting depressive and psychotic disorders were compared. Studies examining the diagnostic performance of these measures published in the Journal of Clinical and Consulting Psychology, Psychological Assessment, and the Journal of Personality Assessment were included in this review. The methodological limitations of these studies are identified and suggestions for improving future research are offered. Although only limited conclusions can be reached because of the small number of existing studies and methodological problems, several preliminary observations were made. First, the MMPI, MCMI-II, and Rorschach are comparable in their sensitivity to detecting depression. However, high MMPI and MCMI-II scores are not specific to depression and misclassify a substantial percentage of nondepressed patients as being depressed. The Rorschach has a lower rate of false positives and higher specificity than the MMPI and the MCMI-II. Second, the Rorschach was more sensitive and specific to psychotic disorders than either the MMPI or MCMI-II. The MCMI-II did not detect psychosis at a rate above chance and frequently inaccurately classified psychotic patients as nonpsychotic. Overall, these preliminary findings suggest that Rorschach indices are promising psychometric markers for depression and psychotic disorders. The clinical implications of these preliminary findings are discussed.