The importance of protein-induced hypercalciuria and the variables contributing to this effect were assessed in a survey of 150 men, 18-56 years of age. Each participant kept a five day dietary record and collected their total urinary output for the last three days of the five. Urinary calcium, magnesium, sulfate and nitrogen were determined on these samples. There was a significant correlation of r = .31 (p < .0001) between dietary protein and urinary calcium. Linear regression analysis (y = 1.22x + 98) revealed that the slope of the line representing this relationship was similar to that of well controlled experimental studies. Calcium intake did not appear to alter this relationship, as the correlation between dietary protein and urinary calcium changed only slightly when calcium intake was statistically controlled. Dietary protein and phosphorus were so interrelated r = .85 (p < .0001) that it was not possible to separate the effects of phosphorus from protein on urinary calcium. The correlation between urinary sulfate and calcium was r = .50 (p < .0001) and step-wise addition multiple regression analysis indicated that urinary sulfate was the best predictor of urinary calcium. Urinary magnesium was not affected by dietary protein or urinary sulfate. However, urinary magnesium and urinary nitrogen were correlated r = .43 (p < .0001) and urinary nitrogen was found to be the best predictor of urinary magnesium. Neither body weight and creatinine excretion nor the amount kilocalories consumed confounded any of the correlations. These results indicate that protein-induced hypercalciuria may be a real effect in this population and that urinary sulfate is the most important determinant of this effect. / M.S.