The components of each sample were dried to constant weight in the laboratory at Seattle (Table 1). The entire diet for each individual was then homogenized with added water in a high-speed blender, to a fine powder. fat, protein, dried at 98°C and pulverized Subsamples of the powder were taken for carbohydrate and radiochemical analyses. Forty to 250-gm portions were wet ashed with HNO3 and H202 and the ash dried in 250-ml beakers for gamma ray spectroscopy. The gamma-counting equipment consisted of a three-inch thallium-activated sodium iodide crystal used in conjunction with a 256~-channel analyzer with a digital print-out. total counts per minute under the photopeak The were calculated by summing counts per minute of all channels included in the peak and subtracting the background counts. The counting efficiency for the gamma energy measured was determined by calibrating the instrument with standards with an error of f 10 per cent. Counts per minute were converted to disintegra- tions per minute by a factor derived from the counting efficiency and the fraction of disintegrations giving rise to the gamma radiation counted. zinc-65, te factors used for cobalt-60, manganese-54 and cesium-barium-137 were 27.4, 18.8 and 16.5 respectively. 54, The correction factor used for sample size as compared to a point source, determined