The aim of this study was to evaluate the effect of breast size on dose heterogeneity. Twenty women underwent a planning CT scan of the thorax. A three-dimensional treatment plan was devised for each patient using a standard technique of isocentric medial and lateral wedged tangential fields. Three-dimensional dose distributions were derived using an equivalent path length (EPL) inhomogeneity correction and cumulative dose-volume histogram (DVH) data calculated for the breast. Analysis of the DVHs for each patient reveals that 0.2-23.8% of the breast received an absorbed dose outside the desired 95-105% of that prescribed at the isocentre. The degree of dose heterogeneity was most strongly correlated with breast volume (r = 0.70, 95% confidence interval (C.I.) 0.37-0.87). There was also a positive correlation for breast dose heterogeneity versus brassière (bra) cup size (Spearman rank correlation rho = 0.62), breast area (r = 0.39, 95% C.I. -0.06-0.71) and chest wall separation (r = 0.31, 95% C.I. -0.15-0.66). We conclude that breast size is an important determinant of dose heterogeneity within the breast.