There is a continuing controversy regarding the value of estimating degree of intra-tumor vascularity to predict prognosis in breast cancer. In order to resolve this controversy, primary tumors from a cohort of 519 women with breast cancer were analysed to determine whether association exists between degree of vascularity and prognosis. Tumor vascularity was estimated by immunohistochemistry using a monoclonal antibody to the antigen CD31. The tumor area showing the highest degree of vascularity was chosen to score the number of microvessels per unit area. Issues such as the reproducibility of the microvascularity score and its association with tumor parameters including size, histological grade and hormone receptor levels were investigated. Although previously agreed criteria were used, consensus between two pathologists' estimations of the degree of vascularity was only moderate. There was no statistically significant association between tumor vascularity score and other currently established parameters of prognosis. After a median follow up of 71 months for axillary node negative patients, there was no association between tumor vascularity score and increased risk of relapse or death from breast cancer. In axillary node positive patients, tumor vascularity score was associated with increased risk of relapse and death from breast cancer. This association was not however independent of other established parameters of prognosis.