Although light therapy has been shown to be effective in the treatment of seasonal affective disorder (SAD), little research has been done to determine which light wavelengths affect treatment outcome. In this triple crossover study the authors compared 1 week of light therapy in which bright (2500 lux), full-spectrum fluorescent light, with and without blockade of the ultraviolet (UV) spectrum, was used with a dim (500 lux) light control in 11 SAD patients. The dim light condition had no significant antidepressant effects as measured by the Hamilton Rating Scale for Depression (HAM-D), the Beck Depression Inventory (BDI), and an atypical depressive symptom (ATYP) score. The UV-light condition significantly reduced HAM-D, BDI, and ATYP scores, whereas the UV-blocked condition significantly reduced only the ATYP score. These results suggest that the UV-spectrum in light therapy may have a differential effect on typical and atypical symptoms in SAD.