We examined the relation of anthropometric variables and breast cancer risk in the Epidemiological Follow-up Study of the first National Health and Nutrition Examination Survey, a cohort study based on a sample of the United States population. A total of 7149 women, 25 to 74 years of age, who were examined during the period 1971 through 1975 were included in the analysis. Stature, sitting height, elbow width, weight, and subscapular and triceps skinfold measurements were collected during the baseline interview and examination. Breast cancer cases (N = 121) were identified through hospital records or death certificates. The median follow-up period for this cohort was 10 years. Women who developed breast cancer were taller and had greater frame size (elbow width) than women who remained free of breast cancer during the follow-up period. After controlling for the effect of potential confounders, the relative risk of breast cancer was 1.9 (95% confidence interval, 1.1 to 3.1) and 2.2 (95% confidence interval, 1.3 to 3.8) among women in the fourth quartiles of stature and elbow width, respectively. Body size defined by weight, relative weight, or skinfold measurements was not associated with increased risk of breast cancer. The positive association of stature and frame size to risk of breast cancer suggests a potential role of early nutrition in cancer etiology.