Remineralization of eroded enamel by dentifrices containing similar sources/concentrations of fluoride was investigated in situ. Fifty-three subjects completed a double-blind crossover study with 3 randomly assigned dentifrice treatments: placebo (0 ppm F, PD); reference (1,450 ppm NaF, RD) and test (1,450 ppm NaF + 5% KNO3, TD). Fluoride availability for each dentifrice was analyzed in vitro by standard tests (1-min fluoride release rate and enamel fluoride uptake). The subjects wore palatal appliances holding bovine enamel specimens previously eroded in vitro. Surface microhardness was determined before and after the in vitro erosive challenge, after in situ remineralization and after a second in vitro erosive challenge. ANOVA and pairwise comparisons were performed (α = 0.05). TD was superior to RD in the fluoride release tests, but similar to RD in the enamel fluoride uptake test. The mean percent surface microhardness recovery was 21.9 (standard deviation 8.0) for PD, 28.6 (8.0) for RD and 36.0 (8.0) for TD. The mean percent relative erosion resistance change was –58.8 (12.7) for PD, –31.3 (12.7) for RD and –27.3 (12.6) for TD. Both fluoride-containing dentifrices provided superior remineralization (p < 0.001) and erosion resistance (p < 0.001) compared to PD. The percent surface microhardness recovery demonstrated by the TD was significantly greater than for the RD (p < 0.001). There was no significant difference (p = 0.073) between TD and RD in relative resistance to further erosive challenge. The results suggest that fluoride availability may be different in dentifrices with similar sources/concentrations of fluoride, providing different levels of remineralization of eroded enamel.