OBJECTIVE: A previous comprehensive literature review indicated that suicide accounted for 18.9% of the deaths of 9,389 individuals with manic-depressive illness. The literature associates these deaths with the depressed phase of the disease. This study was designed to determine the rate and severity of suicidality among patients with pure and depressive mania. METHOD: The patients were 93 persons who met the Research Diagnostic Criteria (RDC) for bipolar I disorder (N = 75) or schizoaffective disorder (N = 18). All met the RDC for primary mania and the DSM-III-R criteria for bipolar disorder, manic or mixed. Patients with depressive mania met the RDC for mania and major depressive disorder concurrently. Severity of current suicidality was measured by using the Schedule for Affective Disorders and Schizophrenia suicide subscale. Differences in the mean suicidality scores between any two groups were assessed with the Kruskal-Wallis test. Relationships of age, gender, type of affective illness (bipolar I versus schizoaffective disorder), psychosis, race, and mania subtype to suicidality were assessed by using multivariate logistic regression analysis. RESULTS: One (2.0%) of the 49 patients with pure mania was suicidal. In contrast, 24 (54.5%) of the 44 patients with depressive mania were suicidal. This difference was highly significant. Gender and psychosis were not related to suicidality. African-Americans were less likely to be suicidal than Caucasians. Subtype of mania had the strongest relationship to suicidality. CONCLUSIONS: A subgroup of manic patients are severely suicidal. Presentation in the manic state is an indication for careful assessment of depressive symptoms and suicidality.