This first controlled psychotherapy trial for seasonal affective disorder (SAD) compared SAD-tailored cognitive-behavioral therapy (CBT), light therapy (LT), and their combination to a concurrent wait-list control. Adults (N = 61) with major depression, recurrent with seasonal pattern, were randomized to one of four 6-week conditions: CBT (1.5-hr twice-weekly group therapy), LT (10,000-lux for 90-min/day with administration time individually adjusted), combined CBT + LT, or a minimal contact/delayed LT control (MCDT; LT following 6 weeks of monitoring). CBT, LT, and CBT + LT significantly and comparably improved depression severity relative to MCDT in intent-to-treat and completer samples. CBT + LT (73%) had a significantly higher remission rate than MCDT (20%). Using prospectively measured summer mood status to estimate the "functional" population, CBT + LT also had a significantly larger proportion of participants with clinically significant change over treatment compared with MCDT. The LT condition outcomes virtually replicated results from prior trials. CBT, alone or combined with LT, holds promise as an efficacious SAD treatment and warrants further study. If replicated, CBT + LT's remission rate would represent a clinically meaningful improvement over the 53% observed across LT studies.