There is growing evidence that not only the total amount of fat, but also the distribution of body fat determines risks for metabolic and cardiovascular disease. Developmental studies on factors influencing body fat distribution have been hampered by a lack of appropriate techniques for measuring intraabdominal fat in early life. Sonography, which is an established method for assessing abdominal fat distribution in adults, has not yet been evaluated in infants. To adapt the sonographic measurement of abdominal fat distribution to infants and study its reliability. The Generation R study, a population-based prospective cohort study. We included 212 one- and 227 two-year old Dutch infants in the present analysis. Sixty-two infants underwent replicate measurements to assess reproducibility. We developed a standardized protocol to measure the thickness of (1) subcutaneous and (2) preperitoneal fat in the upper abdomen of infants. To this end we defined infancy specific measurement areas to quantify fat thickness. Reproducibility of fat measurements was good to excellent with intraclass correlation coefficients of 0.93-0.97 for intra-observer agreement and of 0.89-0.95 for inter-observer agreement. We observed a pronounced increase in preperitoneal fat thickness in the second year of life while subcutaneous fat thickness increased only slightly, resulting in an altered body fat distribution. Gender did not significantly influence fat distribution in the first two years of life. Our age specific protocol for the sonographic measurement of central subcutaneous and preperitoneal fat is a reproducible method that can be instrumental for investigating fat distribution in early life.