Infancy is the period of most rapid postnatal growth and is accompanied by major changes in body composition (BC). There are many challenges to successfully measuring BC of infants in vivo, which include the inherent limitations in the underlying assumptions for each technique. The small body mass and rapid nonuniform changes in body parts, that is, the components of BC during infancy, can strain the technical limits of all methods. Many techniques for in vivo BC measurement used in older people have been applied to infants. However, the vast majority of them either are difficult to adapt for widespread use in infants, or the roles and limitations for using them during infancy are ill-defined because of limited or no critical validation and cross-calibration studies. Based on validation data from animals, well-defined methodological issues in data acquisition and analyses, availability of normative data, and pertinent accuracy and precision of the technique to allow us to determinate clinically relevant changes in BC within a reasonable time interval, three techniques appear to be most suitable for in vivo BC measurement in infants. Anthropometric measurements can be used in field studies or for group comparisons, and total body electrical conductivity (TOBEC) and selected dual-energy X-ray absorptiometry (DXA) measurements can be used to compare BC in individual infants. DXA has the advantages of being able to measure bone mass and the potential to be adaptable to the widely available existing instruments. However, regardless of the techniques used in measuring BC in infants, meticulous attention to details in data acquisition and data analysis, and a knowledge of the limitations of the particular technique are the prerequisites for generating valid data.