Comparisons have been made of [6_ 3H]-, [3-3H]-, and [2-3H]glucose with [U-14C]glucose for measuring parameters of glucose metabolism in sheep given their daily ration in 24 equal amounts at hourly intervals. The specific radioactivity (R) of plasma glucose was measured at frequent intervals from 0 to 10 hr and 0 to 32 hr after the start of a constant infusion. or single injection, of mixtures of these isotopes respectively.Estimates of glucose pool size by consideration of either the initial rectilinear slope (monoexponential analysis) or the multiexponential components of the log R-time curve after single injections of [3H} and [14C]glucose were similar.The mean irreversible loss of plasma glucose estimated by using [3-3H}, and [2-3H]glucose was approximately 79, 70, and 103 respectively. Further release of tritium to body water apparently occurred between 7 and 32 hr following injections of [6-3H} and [3-3H]glucose.Resynthesis of glucose from blood bicarbonate was equivalent to approximately 2% of the irreversible loss of glucose carbon, whereas resynthesis of carbonbound hydrogen of plasma glucose from body water was negligible, although about 91 % of this hydrogen was apparently derived from body water.It was suggested that 20-30% of the intermediates of glucose metabolism were utilized in synthetic reactions and that [2-3H]glucose with [U-14C]glucose was the most useful mixture of labelled glucoses for studies of glucose metabolism, particularly for estimating the extent of glucose resynthesis.