Repetitive transcranial magnetic stimulation (rTMS) is an emerging potential treatment for depression, but the data supporting its efficacy have not been systematically reviewed. The purpose of this study was to conduct a meta-analysis of rTMS trials in the treatment of depression. A search for all published and unpublished sham-controlled studies of left or right prefrontal cortical rTMS in the treatment of depression evaluated by the Hamilton Depression Rating Scale (HDRS) was conducted using no language restrictions. Fixed- and random-effects meta-analyses were performed on 12 studies comparing the decrease in HDRS scores achieved with rTMS and sham stimulation. Initial results with a fixed-effects analysis failed homogeneity testing; thus, a random-effects analysis was used to calculate all results. In 12 studies (16 individual effect sizes), the weighted mean effect size was 0.81 (95% CI: 0.42-1.20, P < .001). For studies using left dorsolateral pre-frontal cortex (DLPFC) stimulation (11 studies, 14 effect sizes), the weighted mean effect size was 0.89 (95% CI: 0.44-1.35, P < .001). For studies using left DLPFC stimulation in a parallel-groups design (seven studies, nine effect sizes), the weighted mean effect size was 0.88 (95% CI: 0.22-1.54, P < .01). No study showed a mean decrease in HDRS scores of > 50%, and the number of responders to rTMS (defined as a > 50% decrease in HDRS scores) across studies was relatively small (13.7% with rTMS versus 7.9% with sham stimulation). rTMS is statistically superior to sham stimulation in the treatment of depression, showing a moderate to large effect size. However, the clinical significance of these results is modest. The differences in response to rTMS across studies are not clearly explained, and, therefore, more research is needed.