Breast cancer, or its effects, may be preventable. Childbearing at an early age will confer some protection. Until there are other methods of primary prevention the physician must concentrate on secondary prevention, including screening. In presenting risk factors as criteria for screening we have attempted to make rational the use of screening technology. We should not easily accept screening for breast cancer as of proved value. Rather we must encourage clinical trials of various screening methods. In the meantime, while we engage in a moderate amount of screening we do so under the Scottish verdict of "not proven".