Variability in calcium absorption was estimated in three groups of normal subjects in whom Ca absorption was measured by standard isotopic-tracer methods at interstudy intervals ranging from 1 to 4 mo. Fifty absorption tests were performed in 22 subjects. Each was done in the morning after an overnight fast with an identical standard breakfast containing a Ca load of approximately 250 mg. Individual fractional absorption values were normalized to permit pooling of the data. The coefficient of variation (CVs) for absorption for the three groups ranged from 10.57 to 12.79% with the size of the CV increasing with interstudy duration. One other published study presenting replicate absorption values was analyzed in a similar fashion and was found to have a CV of absorption of 9.78%. From these data we estimate that when the standard double-isotope method is used to measure Ca absorption there is approximately 10% variability around any given absorption value within an individual human subject and that roughly two-thirds of this represents real biological variability in absorption.