In Reply.—
Zimmerman and Coryell have suggested that by not "faithfully" applying the diagnostic criteria for melancholia (DSM-III) and for endogenous depression (RDC) in our recent family study of depressive subtypes,1 we cannot validly assess these competing diagnostic systems. We disagree. They correctly point out that the diagnostic algorithms we used to generate these subtype diagnoses were not exactly identical to those of the DSM-III and RDC. As we indicated in the "Subjects and Methods" section of our report, subtype assignments were based on the proband's or relative's response to specific queries during a direct interview using a modified version of the Schedule for Affective Disorder and Schizophrenia. The modifications included questions about depressive symptoms that were not included in the original schedule, such as earlymorning awakening, lack of reactivity, and morning worsening. As a result, most, but not all, of the specific criteria for the melancholic and endogenous subtypes were