IN THIS issue of the ARCHIVES, Murphy et al^1,2 and Eaton et al³ provide new data to address current issues involving the distinction between psychopathology and normal mental states as well as the assessment of specific mental disorders in community settings.^4-7 The 40-year Stirling County Study¹ and the 13-year follow-up of the Baltimore Epidemiologic Catchment Area (ECA) study³ provide a valuable longitudinal perspective on the clinical course of persons identified with mental disorders in epidemiologic surveys. Both studies have the added dimension of comparing diagnoses made by nonclinicians using highly structured instruments with diagnoses made by clinicians using structured clinical interviews.