We evaluated the potential of a pyrophosphate-fluoride solution to affect the remineralizing-demineralizing equilibrium, i.e., caries-inhibiting/-promoting effects on enamel in vivo. Fifteen subjects carried dentin blocks and enamel thin sections with artificial caries lesions in removable partial dentures for periods of two weeks, during which time they rinsed twice daily in a double-blind, randomized cross-over design with solutions containing (a) 90 ppm F, (b) 90 ppm F and 1% pyrophosphate, or (c) no active agents (placebo). A severe cariogenic challenge provided to the lesions by plaque under a stainless steel mesh cover resulted in the placebo-treated lesions losing 70.2% +/- 72.1% mineral. The pyrophosphate rinse with fluoride held the mineral loss to only 28.1 +/- 52.8%, while the fluoride rinse without pyrophosphate held the loss to 24.2 +/- 50.1%. The differences between the fluoride and placebo results were significant (p less than 0.01), but the difference between the two fluoride groups was not. A large amount of fluoride was deposited in the dentin lesions. In the surface layer, the concentration was increased from 1000 ppm F to more than 2300 ppm F by both fluoride treatments. The concentration of fluoride in both groups of fluoride-treated lesions further increased to more than 3500 ppm F in the approximate center of the lesion before declining in deeper layers toward the level found in the placebo-treated group. The findings from both mineral change and fluoride uptake phases of this study show that in the presence of 90 ppm F, 1% pyrophosphate did not promote demineralization of artificial caries lesions.