The present paper analyses the epidemiologic effects of mass breast cancer screening. Mass mammography may possibly achieve a breast cancer mortality reduction in relative risk terms. However, this does not necessarily represent a net benefit. It is argued that the benefits and adverse effects of a screening programme must be measured in terms of absolute risks. According to this measure, the mortality reduction achieved by a mass breast screening programme is only one death per approx. 15,000 women-years. Many thousands of mammograms are needed to prevent one cancer death, and for each woman who can derive a direct benefit in terms of a prevented breast cancer death, hundreds of women have to suffer the anxiety of a positive screening mammography. Moreover, it is possible that adverse effects of breast cancer screening may contribute to mortality from other causes. Even with the assumption that screening can save lives, the net health effect of mass breast cancer screening is questionable and appears to be rather detrimental. It may be an error to recommend mass breast screening.