We reviewed evidence regarding breast self-examination (BSE) and screening for breast cancer. To our knowledge, no controlled prospective trial links BSE to lives saved from breast cancer. Compared with clinical breast examination and mammography, the estimated sensitivity of BSE is low (20% to 30%) and is lower among older women. The potential sensitivity of BSE should be higher because women can detect small lumps (0.3 cm) in silicone models. Instruction increases BSE frequency over the short term. Sensitivity also increases, but specificity decreases. The psychological effects of teaching and performing BSE are not yet clear. The cost of screening by BSE is unknown but depends on the accuracy of the test as well as the training method used. Breast self-examination has potential as a screening test for breast cancer, but many questions require scientific examination before this procedure can be advocated as a screening test for breast cancer.