The aim of this study was to examine the possible influence on survival of delays prior to presentation and/or treatment among women with breast cancer. Duration of symptoms prior to hospital referral was recorded for 2964 women who presented with any stage of breast cancer to Guy's Hospital between 1975 and 1990. Median follow-up is 12.5 years. The impact of delay (defined as having symptoms for 12 or more weeks) on survival was measured from the date of diagnosis and from the date when the patient first noticed symptoms to control for lead-time bias. Thirty-two per cent (942/2964) of patients had symptoms for 12 or more weeks before their first hospital visit and 32% (302/942) of patients with delays of 12 or more weeks had locally advanced or metastatic disease, compared with only 10% (210/2022) of those with delays of less than 12 weeks (P < 0.0001). Survival measured both from the date of diagnosis (P < 0.001) and from the onset of the patient's symptoms (P = 0.003) was worse among women with longer delays. Ten years after the onset of symptoms, survival was 52% for women with delays less than 12 weeks and 47% for those with longer delays. At 20 years the survival rates were 34% and 24% respectively. Furthermore, patients with delays of 12-26 weeks had significantly worse survival rates than those with delays of less than 12 weeks. Multivariate analyses indicated that the adverse impact of delay in presentation on survival was attributable to an association between longer delays and more advanced stage. However, within individual stages, longer delay had no adverse impact on survival. Analyses based on 'total delay (i.e. the interval between a patient first noticing symptoms and starting treatment) yielded very similar results in terms of survival to those based on delay to first hospital visit (delay in presentation).