The preparation of radioactive vitamin D3 and its absorption in the rat have been described by Norman and DeLuca (1) and by Schachter, Finkelstein, and Kowarski (2). Using tied intestinal loops and animals with artificial lymph fistulae, Schachter and his associates (2) have shown that maximal absorption of tritium-labeled vitamin D3 takes place in the mid-jejunum and that its transfer into the blood is mainly via the lymph. Little information exists on the absorption of vitamin D in man, except for the observations of Kodicek (3), who found that between 13 and 23% of an oral dose of vitamin D2-14C was recoverable from the feces of infants within 3 days.The present paper deals with the preparation, purification, and radiochemical behavior of vitamin D3 after random labeling with tritium and with its use in human subjects. The labeled vitamin D was purified by methods essentially similar to those previously described (2), with the main exception that the vitamin was recovered in crystalline form without preliminary esterification. Vitamin D absorption was assessed in control subjects and patients with various forms of intestinal malabsorption by measuring their plasma and fecal radioactivity after oral doses of vitamin D3-3H. Identification. The labeled crystalline material was identical to authentic vitamin D3 in its mobility on thinlayer chromatography when using either chloroform or 10% vol/vol acetone in hexane as solvents, in its ultraviolet absorption spectrum, which showed a peak at 265 msu, and on quantitative estimation with Nield, Russell, and Zimmerli's reagent (4). In addition, bioassay was carried out at two dilutions in paired rats from each of four rachitic litters, healing being assessed radiologically (5). The labeled vitamin was fully active when compared with a vitamin D3 standard.On thin-layer chromatography the main contaminant in the unpurified material moved with the same mobility as a precalciferol 3 marker, prepared by refluxing crystalline vitamin D3 in benzene (6). This radioactive compound had an absorption peak at 262 my and reacted with Nield's reagent similarly to vitamin D3. It was biologically inactive, but on storage at 40 C it was gradually converted into vitamin D3-3H. These characteristics suggest that it was precalciferol 3, an isomer of vitamin Ds (7).Radiochemical behavior. The highest initial specific activity obtained with any batch of vitamin D3-3H was 54.6 /Ac per mg. Further measurements of specific activity were performed at intervals after repeated repurification of the labeled vitamin by thin-layer chromatography. During the 5 days after crystallization the specific activity of a solution of this batch of vitamin D3-3H in benzene decreased rapidly to 18.8 jAc per mg, and this was followed by a more gradual decline in specific activity during the next 10 days (Figure 1). From the