High BMI is a well-known risk factor for postmenopausal breast cancer. There have been some reports of excess risk in association with weight gain and WHR, but little is known about the influence of body fatness per se. Using data from the Malmö Diet and Cancer Study, a prospective cohort study, 12,159 postmenopausal women (59.9 ± 7.7 years) were categorized by quintiles of baseline anthropometric and impedance measures and reported weight change since age 20. RRs from multivariate Cox regression models were calculated. All analyses were adjusted for age, height, smoking, alcohol consumption, occupation, marital status, parity, age at first pregnancy, age at menarche and current hormone use. During the 5.7 years of follow-up, there were 246 incident breast cancer cases. Weight, height, BMI and %BF were positively associated with risk of breast cancer (ptrend ≤ 0.02). %BF showed the strongest association, with an RR of 2.01 (95% CI 1.26–3.21) in the highest vs. lowest quintile. There was significant modification of this association by hormone use, suggesting a greater impact of body fatness among nonusers. Fat distribution was not independently associated with breast cancer risk. Women with weight gain >21 kg (top quintile) had an RR of 1.75 (95% CI 1.11–2.77) compared to women with low weight gain. Breast cancer risk in postmenopausal women is predicted by increased body fat and weight gain. %BF is a more discriminating risk factor for breast cancer incidence than the commonly used BMI. © 2002 Wiley-Liss, Inc.