This study was designed to evaluate the effect of race, specifically African-American, on electroencephalographic (EEG) sleep and clinical symptom profile in unipolar major depression. A clinical research database was used to identify appropriate subjects and a double-matched historical case-control design was implemented. African-American depression patients were double matched within protocol to Euro-American depressed patients. Age, sex, and protocol of origin were matching variables. African-American depressed patients had less total sleep, less slow wave sleep, more stage 2 sleep, longer rapid eye movement (REM) sleep latency, less REM sleep, and lower REM density than Euro-American depressed patients. African-American depressed patients did not differ from Euro-American patients in symptom severity, age of onset, number of episodes, socio-economic status, and, as planned, did not differ in age and sex distribution. Depressive symptom constellation also did not distinguish the two groups. African-American depressed patients demonstrated differences in EEG sleep profile, with less total sleep, overall lighter nonREM sleep, and relatively preserved REM sleep, despite a clinical symptom profile that did not differ from Euro-American depressed patients. The sleep profile appeared to be consonant with the sleep findings in chronic insomnia. The pathological implications of these differences remained to be explored in careful prospective studies of African-American depressed patients and in well-characterized, racially matched normal control comparisons.