Propranolol, 60 mg or less, was administered daily between 5:30 and 6:00 a.m. to 33 patients with winter depression. After open treatment with a mean dose of 33 mg/day, 24 patients (73%) met the remission criteria; 23 completed double-blind continuation or placebo substitution. Subjects who continued to receive propranolol had a mean increase in Hamilton depression score of 3.5, whereas patients switched to placebo had an increase of 11.2; the difference was statistically significant. These findings are consistent with the hypothesis that duration of nocturnal melatonin secretion is the critical seasonal time cue in humans.