A previous study showed that a two-solution fluoride (F) rinse deposited significantly more loosely-bound F on the tooth surface than did a sodium fluoride (NaF) rinse with the same F concentration (12 mmol/L). In the present study, this experimental rinse was evaluated for its ability to cause remineralization of enamel lesions in an in vitro pH-cycling model. Caries-like lesions were formed in the enamel of extracted human molars by means of a pH 4 demineralizing solution. Fifty-one approximately 120-microns-thick sections containing lesions were randomly divided into (1) control, (2) NaF rinse, and (3) two-solution F rinse groups. With the cut surfaces protected, the control samples were immersed in a pH 7 remineralizing solution for 12 days, and twice daily the sections were also exposed to a pH 4 demineralizing solution for 30 min. Samples in the NaF group received an additional one-minute rinse with a NaF (12 mmol/L) solution twice daily. Samples in the two-solution rinse group received the rinse treatment with a 12 mmol/L F solution prepared by combination of a Na2SiF6 and phosphate-containing solution with a calcium solution just before use. The mineral contents of the lesions were assessed by quantitative microradiography. The results showed that (1) no significant de- or remineralization was detected in the controls; (2) a 46% decrease in mineral loss (delta Z) of the lesion was produced by the NaF rinses; and (3) a 94% decrease in delta Z and a 20-microns-thick, mineral-dense surface-coating were produced by the two-solution F rinse treatment.