Eight-seven moderately to severely depressed psychiatric outpatients were randomly assigned to 12 weeks of cognitive therapy (CT) (n = 24), pharmacotherapy (n = 24), CT plus pharmacotherapy (n = 22), or CT plus active placebo (n = 17). Seventy patients completed the treatment protocol. Seventeen dropped out before the end of the treatment period. Patients who completed treatment showed significant improvement in mean scores on two common measures of severity of depression (the Beck Depression Inventory and the Hamilton Rating Scale for Depression) between evaluation and termination. Improvement did not differ as a function of the different treatment modalities. Inclusion of dropout patients' end-point scores did not alter these results. Treatment gains in all groups were maintained at one-month follow-up assessment. A portion of this study replicated an earlier study. While the results were not identical, they indicated that either CT or antidepressant drug treatment can be effective in treating outpatients with primary, nonbipolar depression of moderate or greater severity. Combining treatments did not lead either to additive effects or negative interactions.