Intuitively, breast tissue mass should be directly related to a woman's risk of breast cancer, simply because having more cells at risk would seem to increase the potential for malignant transformation. However, studies attempting to link breast size with breast cancer risk have been inconsistent. Limitations include crude measures of breast size, the inability to distinguish glandular from adipose tissue, and the confounding influence of co-factors such as obesity. A nationwide study in Denmark was undertaken to investigate the effect of breast reduction surgery on the subsequent risk of breast cancer, including an evaluation of the patterns of risk by age and time since surgery. The Danish Hospital Discharge Registry was used to identify women who underwent reduction mammaplasty between 1977 and 1992. Linkage based on personal identification numbers with the Danish Cancer Registry provided information on cancer incidence. Expected numbers of cancers were calculated from rates in the general population. Among 7,720 women whose breasts were surgically reduced, 182 cancers were subsequently observed cf 209 expected (standardized incidence ratio [SIR] = 0.9;95 percent confidence interval [CI] = 0.7-1.0). Breast cancer was significantly reduced by nearly 50 percent (29 observed cf 53.9 expected, SIR = 0.5, CI = 0.4-0.8), and accounted for the overall deficit in cancer. The risk reductions were related inversely to age at surgery, with significant deficits apparent only among women 40 years of age and older at surgery and especially among those over age 50 (SIR = 0.3). No clear trend was apparent with increasing years post-surgery. The findings indicate that breast reduction surgery among women over age 40 is associated with a lower subsequent risk of breast cancer, but the surgery and presumably glandular mass appear less closely related to breast cancer risk among younger women.