Fifty-five patients with winter seasonal affective disorder (SAD) were treated with a light visor, a newly developed portable light-delivery system, in a controlled parallel design. A dim (400 lux) visor was compared with a bright (6000 lux) visor for either 30 or 60 minutes in the morning for 1 week. Response rates for these two treatments were 36% and 56%, respectively; the duration of treatment sessions did not affect outcome. There was no evidence that the brighter visor was superior in efficacy to the dimmer one. Significantly greater relapse occurred following withdrawal of the dimmer visor. Alternative explanations for these findings are that the light visor is acting as a placebo or that it is equally effective over a wide range of intensities.