In light of the growing prevalence of obesity in the United States, and the health risks associated with childhood obesity in particular, it is critical to identify avenues for obesity prevention. This study tests the hypothesis that breastfeeding serves as one protective factor against children's subsequent development of obesity. We used linear-, logistic-, and sibling fixed-effects regression models to evaluate the association between infant feeding history and body mass index (BMI) in late childhood or adolescence (9-19 years, mean = 14 years). Complete data were available for 976 participants (488 sibling pairs) in the 2002 Child Development Supplement of the Panel Study of Income Dynamics, a nationally representative survey of families in the United States. In sibling pairs in which only one sibling was breastfed, the breastfed sibling had an adolescent BMI that was 0.39 standard deviations lower than his or her sibling, controlling for child-specific factors that may have influenced parents' feeding decisions. This effect is equivalent to a difference of more than 13 pounds for a 14-year-old child of average height. Furthermore, fixed-effects logistic regressions predicting overweight and obese status showed that breastfed siblings were less likely to reach those BMI thresholds. We therefore conclude that breastfeeding in infancy may be an important protective factor against the development of obesity in the United States. The application of a sibling fixed-effects model provides stronger evidence of a causal relationship than prior research reporting similar patterns of association.