A vitamin D-independent paracellular and not saturable mechanism important for intestinal calcium absorption turned out to be less effective in lactose-free diets. Twelve normal term babies with identical repletion levels of 25-hydroxyvitamin D (25D) were fed sequentially with starting formulas that differed only in their carbohydrate sources: lactose (control) and polycose. A 3-day test was performed after a 1-week adjustment to the new diet. Fecal excretion of calcium was significantly higher in the lactose-free formula period compared to the control one: 56.6 (13.9) versus 39.6 (12.8) mg/kg/day (p less than 0.02); subsequently, the net retention was considerably higher and statistically significant (p less than 0.01) for the control period versus the lactose-free formula period: 99.4 (15.3) versus 78.1 (14.3). Plasma 25D values were 21.8 (9.1) and 23.7 (7.1) ng/ml at the beginning of the lactose and polycose periods, respectively, and 1,25-dihydroxyvitamin-D values were 64 (62) and 75 (45) pg/ml, not significant. Results suggest that formulas containing polycose as the sole carbohydrate source may reduce the intestinal absorption of calcium in term newborns, but determining long-term biological significance requires further observation.