A study of five methods of determining burnup in fuel test specimens has been made to determine the confidence which can be placed in the burnup numbers so obtained. The five methods compared are (1) uranium-235 depletion, (2) cobalt monitoring, (3) cesium-137 production, (4) plutonium production, and (5) calorimetry. The study includes a comparison of data obtained on portions of the specimen with that obtained on the complete specimen. It has been found that all of the methods give burnup values that are within ± 5% of the “best” burnup value, the “best” value being defined as the unweighted average of all the available results on a particular sample. The limitations and pitfalls of all the methods are discussed in some detail. It has further been found that a complete cross-section of the test specimen, approximately ½in. long, is sufficient sample to give representative burnup data. The integration from this to the complete specimen is relatively straightforward. Any sample not comprising a complete cross-section involves radial as well as longitudinal integration, giving results that are relatively uncertain, particularly in samples of high heat rating where migration of some species is a distinct possibility.