Fifty outpatients with dysthymic disorder (DSM-III) were divided by double-blind randomized assignment into three groups given ritanserin, imipramine, and placebo, respectively. The trial was of 7 weeks' duration; by week 6, imipramine was clearly superior to placebo, whereas by week 7, both drugs caused significantly more improvement than the placebo. Although imipramine had slightly greater efficacy than ritanserin, it also had significantly more side effects. This was particularly evident in the higher dropout rate with imipramine. The efficacy and side effect profile of ritanserin makes it well tolerated and acceptable with dysthymic patients who, although they do not respond as quickly as patients with major depressive disorder, do show significant improvement, given sufficient time.