We investigated the effects of exclusion criteria in depression treatment studies by applying simulated exclusion criteria to a relatively unselected sample. The criteria were borderline personality disorder, drug/alcohol dependence, and exclusion criteria from the Treatment of Depression Collaborative Research Program and the Maintenance Therapies for Recurrent Depression Protocol. Each set of criteria substantially distorted demographic characteristics and Axis I and II comorbidity but did not yield higher rates of study completion or, in general, response to tricyclics. Excessive exclusion criteria in depression treatment studies have many adverse effects and should be applied only when overwhelmingly compelling reasons exist.