From retrospective studies, there is substantial evidence that birthweight and the rate of weight gain during early infancy are associated with increased risk for adverse health outcomes later in life. Birthweight is the marker of the integrative effects of the prenatal environment, while the rate of weight gain after birth reflects both genetic potential and external postnatal influences. The adulthood-to-infancy associations constitute the basis for the 'fetal origins' and 'catch-up growth' hypotheses for some diseases. However, these findings are based on the assumption that anthropometric-based indices reflect body composition during both time periods, with the body mass index (weight/stature2) being the most frequently used index. More direct measures of body composition were simply not available at the time of the births of the adults participating in these studies. Nowadays, there are a number of in vivo techniques that can be used to examine body composition in infancy. In particular, what does the body mass index reflect in terms of body composition for the infant? Is it an adequate index?