In the past thirty years, the mortality rate from prostate cancer has not declined. This is due in part to the small number of tumors diagnosed while they are still curable by surgery or radiation. Recent studies have demonstrated that transrectal sonography can detect nonpalpable tumors, which are usually potentially curable. As a result, some physicians now recommend routine screening of all men over age fifty. The main goal of a cancer screening test is to help reduce mortality. To date, it has been shown that screening can increase early detection and survival, but there is no experimental evidence that screening reduces mortality. Therefore, screening for prostate cancer may not be worthwhile. The large number of prostate tumors found on autopsy series indicates that only a small fraction of patients with prostate cancer actually die of it. The risk of screening is that many insignificant tumors may be diagnosed and patients may be treated unnecessarily, resulting in significant morbidity, mortality, and expense. For now, it may be appropriate to let patients choose whether or not to have the test performed, after informing them screening can increase early detection, but could also result in overdiagnosis and treatment. However, screening should not be advocated across-the-board until a randomized, carefully controlled, large-scale clinical trial has shown that it saves lives.