304 first- and second-year medical students were prospectively assessed for depression with a monthly Beck Depression Inventory (BDI). Students scoring above nine on the BDI and a control group were then interviewed with the NIMH Diagnostic Interview Schedule. The incidence of major depression or probable major depression by DSM-III criteria during the first two years of medical school was 12%. The lifetime prevalence was 15%, three times greater than the rate in the general population. An episode of depression prior to medical school was much more common among the depressed students (69 vs. 8%, P less than 0.001) as was a family history of treated depression (46 vs. 21%, P less than 0.025). The elevated rate of depression during medical school does not appear to be a result of the medical school experience alone. Rather, it suggests a positive bias of unknown nature in the selection of students predisposed to depression.