Sleep measures were obtained in 16 depressed and 21 control adolescents following 1 week of adherence to a uniformly imposed and strictly enforced sleep/wake schedule. Three nights of baseline electroencephalographic (EEG) sleep on the same 10:00 PM to 7:00 AM schedule revealed prolonged sleep latency and reduced rapid eye movement (REM) latency in the depressed adolescents. Following baseline measures, sleep was restricted for 2 nights (10:00 PM-4:00 AM) and measures of recovery sleep were obtained showing further sleep latency differences. There was no evidence for delta sleep changes or sleep continuity differences in depressed adolescents. These results suggest that control over sleep/wake schedules is an important methodological issue in adolescent sleep studies. Furthermore, the findings are consistent with a larger body of evidence indicating that dysregulation near sleep onset represents a primary psychobiological change in early-onset depression.