Routine measurement of adipose tissue composition by repeated biopsy invokes both ethical and practical difficulties, limiting long-term serial studies of adipose tissue composition. In vivo13C nuclear magnetic resonance (NMR) spectroscopy has been applied as a non-invasive alternative, although it has not as yet been fully validated. In this study we critically assess in vivo13C NMR spectroscopy and gas–liquid chromatography for the analysis of adipose tissue composition. The advantages and drawbacks of both methods are discussed, in particular to the study of adipose tissue during dietary manipulation and development. Our results show that the NMR measurements of adipose tissue composition are highly reproducible, but they can significantly differ from those obtained by gas–liquid chromatography (GLC) from the same volunteer. We show that the discrepancy between these two techniques arises from inherent limitations of both 13C NMR spectroscopy and GLC. Finally, we show that 13C NMR spectroscopy remains a useful non-invasive tool for the study of adipose tissue, and will enable us to perform long-term serial studies to further our knowledge of lipid metabolism.