The majority of information available today indicates that the most efficient and accurate method of screening women to detect early-stage breast cancer is an aggressive program of patient self-examination, physical examination by well-trained, motivated personnel, and high-quality x-ray mammography. There are two important factors in the implementation of mammographic screening. The first is the availability of facilities to perform high-quality, low-dose mammography, which is directly related to the second factor: the expense to society for support of this large-scale effort. Cost-benefit analysis is beyond the scope of this review. In 1979 Moskowitz and Fox attempted to address this issue, using data from the Breast Cancer Detection Demonstration Project in Cincinnati, but additional analysis is required. The cost for each "curable" cancer that is detected must be compared with the psychological, social, and personal losses that accrue, as well as the numerous medical expenses incurred, in a frequently protracted death from breast cancer. All other imaging techniques that have been reviewed should be regarded as adjuncts to rather than replacements for mammographic screening (Table 1). Ultrasound and computerized tomography are helpful when the physical examination and mammogram are equivocal. Other techniques, such as transillumination, thermography, and magnetic-resonance imaging, should be considered experimental. In patients with clinically evident lesions, x-ray mammography is helpful to evaluate the suspicious area, as well as to "screen" the remaining tissue in both breasts and to search for multicentric or bilateral lesions. Mammography is the only imaging technique that has been proved effective for screening. The low doses required by present-day mammographic technology pose a possible risk that is so small it is not measurable. The image quality has improved considerably over the past decade, and data supporting the benefits of mammography are increasing. As a result, the American Cancer Society has recently modified its recommendations to include mammographic screening of asymptomatic women beginning at the age of 40 years (Table 2). Before any new system can be considered a replacement for mammographic screening, carefully executed trials are necessary to prove efficacy beyond anecdotal claims.