Provide a proof of concept for the potential of using a novel RF resonant cavity device for accurately and repeatedly measuring fat and fat-free masses in phantom infants. Design, construct, and characterize an RF resonant cavity with dimensions compatible to holding an infant. The cavity was characterized using spherical phantoms of 0%fat, 50% fat, and 100% fat to empirically calibrate shifts in resonant frequency. The phantoms were constructed using emulsions of bovine lard, water, and dish soap inside spherical containers which do not interact with the electric field. The calibration phantoms were compared with a phantom of a test sample to assess the ability of the resonant cavity perturbation technique for measuring body composition. Phantoms of distinct %fat (0%, 50%, and 100%) were used to calibrate the resonant cavity for measuring body composition. The calibration phantoms were used to create calibration lines of unique %fat and were compared to a 475-mL sample of unknown %fat as a measure of how accurate the resonant cavity technique is for measuring body composition. A 475mL test sample was used to examine the robustness of the RCP technique. The sample was 25% fat and had a fat mass of ( )g. The measured fat mass from the RCP technique was g, or a 2% difference. The resonant cavity perturbation technique provides an accurate and repeatable measurement of fat mass in spherical phantoms and suggests the technology might be an effective obesity research tool for infants. Future studies will focus on extending the work to more complex anthropomorphic shapes.