OBJECTIVE Calcium glycerophosphate (CaGP) was added to fluoride varnishes to analyze their preventive effect on initial enamel erosion and fluoride uptake: potassium hydroxide (KOH)-soluble and KOH-insoluble fluoride bound to enamel. METHODS This study was carried out in two parts. Part 1: 108 enamel samples were randomly distributed into six varnish groups: base varnish (no active ingredients); Duraphat® (2.26%NaF); Duofluorid® (5.63%NaF/CaF2); experimental varnish 1 (1%CaGP/5.63 NaF/CaF2); experimental varnish 2 (5%CaGP/5.63%NaF/CaF2); and no varnish. Cyclic demineralization (90 s; citric acid, pH = 3.6) and remineralization (4 h) was made once a day, for 3 days. Change in surface microhardness (SMH) was measured. Part 2: 60 enamel samples were cut in half and received no varnish (control) or a layer of varnish: Duraphat®, Duofluorid®, experimental varnishes 1 and 2. Then, KOH-soluble and KOH-insoluble fluoride were analyzed using an electrode. RESULTS After cyclic demineralization, SMH decreased in all samples, but Duraphat® caused less hardness loss. No difference was observed between varnishes containing CaGP and the other varnishes. Similar amounts of KOH-soluble and insoluble fluoride was found in experimental varnish 1 and Duofluorid®, while lower values were found for experimental varnish 2 and Duraphat®. CONCLUSIONS The addition of CaGP to fluoride varnishes did not increase fluoride bound to enamel and did not enhance their protection against initial enamel erosion. CONCLUSIONS We observe that the fluoride varnishes containing CaGP do not promote greater amounts of fluoride bound to enamel and that fluoride bound to enamel may not be closely related to erosion prevention.