Laboratory-based research in germ line mutations associated with breast cancer susceptibility is rapidly being integrated into clinical practice with profound implications. A Medline search was performed for all relevant articles published since 1990. Where appropriate, historical articles referenced in those identified were also reviewed. The results suggested that while mutations in the BRCA1 and BRCA2 genes are the most clinically relevant, much of the data on which clinical decisions are based must be interpreted with wide confidence intervals. Between 1 in 152 and 1 in 833 individuals carry such mutations. They account for less than 5% of all breast cancer, but up to 10% of cancers in those under the age of 40 years. Founder mutations are responsible for a larger proportion of breast cancer cases within certain inbred communities. Phenotypic expression and penetrance of different mutations is not currently predictable and estimates of penetrance are largely based on highly selected populations. BRCA1 mutations are more commonly associated with ovarian cancer than BRCA2 mutations. BRCA1 cancers tend to have more distinct pathological features and are usually oestrogen receptor (ER)-negative. To conclude, the evidence in this review suggests that caution should be exercised when translating scientific progress in breast cancer germ line genetics into clinical practice. Most of the available data are derived from studies on highly selected populations. The importance of other less penetrant, but more prevalent, germ line mutations may be realised in the future.