There is an increasing body of research suggesting that low body weight is associated with scoliosis, but this is based on case-control studies, which are prone to bias. No studies have investigated the components of body weight: fat and lean mass. We have therefore carried out the first population-based prospective study of the association between fat and lean mass at age 10 years assessed by dual-energy X-ray absorptiometry (DXA), with presence of scoliosis at age 15 years using the Avon Longitudinal Study of Parents and Children (ALSPAC). All children with scoliosis at age 10 years were excluded. Of 5299 children at age 15 years, 312 (5.9%) had scoliosis. Our results show a negative association between body mass index (BMI)/body weight at age 10 years and scoliosis at age 15 years, with a 20% reduced risk of scoliosis per SD increase in BMI (odds ratio [OR], 0.80; 95% confidence interval [CI], 0.70-0.92; p ¼ 0.001). This association with BMI/body weight reflects associations with both fat mass and lean mass. After adjustment for age, gender, leg length, and fat mass per SD increase in lean mass, there was a 20% reduced risk of scoliosis (OR, 0.80; 95% CI, 0.65-0.97) and per SD increase in fat mass there was a 13% reduced risk of scoliosis (OR, 0.87; 95% CI, 0.74-1.03). In terms of adipocyte function, an inverse association was seen between leptin at age 10 years and scoliosis (OR for scoliosis per SD increase in leptin of 0.78; 95% CI, 0.63-0.99), and a positive association between adiponectin at age 10 years and scoliosis (OR for scoliosis per SD increase in adiponectin of 1.44; 95% CI, 0.99-2.10). This is the first study to address the association between the individual components of body weight and scoliosis in a prospective cohort study, and shows altered body composition that is present before the onset of clinically detected scoliosis.