Résumé : Sleep abnormalities have been repeatedly demonstrated in major depression. However, the respective influences of age, severity, adaptation and gender have never been clearly disentangled. In a retrospective study, full polysomnograms of 67 male depressive patients and 67 carefully age-matched male healthy control subjects were analyzed. The usual differences associated with the sleep of depressed patients were observed. However, in contrast to most reports, REMS was also found to be reduced; although no comparisons between sexes can be made in this all-male study, one interpretation of this finding is that reduction of REMS is a marker of male depression. Age was found to influence most sleep variables, but not the order of their association with depression. Depression severity was found to be associated with Wake After Sleep Onset (WASO), REMS, and Non-REMS (NREMS). No residual adaptation effect was observed. One of the main markers of depression was in fact the absence of sleep, whether observed as long delays prior to entering sleep, or excessive intermittent awakenings. This sleep reduction affected both REMS and NREMS, in comparable percentages. This supports the hypothesis of a hyperarousal possibly linked to stress.