Records of patients who had been systematically evaluated at the Center for Anxiety and Depression were reviewed. Included in this report are records of patients who had a diagnosis of bipolar II or unipolar affective disorder as determined by one of the authors and who also underwent a Structured Clinical Interview for DSM-III (SCID). A second study was undertaken with L.K.T. observing the clinical interview of D.L.D. and both clinicians making a diagnosis for the patient. There was reasonable diagnostic agreement between D.L.D. and the SCID for diagnosis of bipolar II disorder. Twelve of 34 patients clinically diagnosed as bipolar II were diagnosed by the SCID as unipolar. These 12 patients failed to demonstrate clinical factors, which could help explain the diagnostic difference between SCID and the clinical diagnosis. There was excellent agreement on 34 patients as to the presence or absence of hypomania, comparing two clinicians. The SCID interview underestimates the diagnosis of hypomania, a condition that can be diagnosed reliably by clinicians trained to make this diagnosis. This finding has some implications for DSM-IV, where there is a proposal for inclusion of bipolar II disorder as a parallel entity with bipolar disorder.