Objective: The aim of this study was to compare the effects on the resistance to demineralization by the iontophoresis and the conventional fluoride application (CFA) in vitro. Methods: Eighty four bovine enamel specimens were embedded in acrylic resin with the labial surfaces exposed. The specimens were divided into 3 groups: no fluoride treatment, CFA, and iontophoresis. The experimental groups were divided into 3 different subgroups (1.23% APF gel, 2% NaF gel, and 2% NaF solution). The specimens in the control group were not treated with fluoride and were placed in the remineralizing solution (RS) for 10 days. However, all the experimental groups were immersed in the RS for 2 days after the fluoride treatment for 4 minutes. This cycle was repeated 5 times. All the specimens were exposed to a pH-cycling model which consisted of demineralization (6 hours) and remineralization (18 hours) for 5 days. The Vickers surface micro-hardness number (VHN) of all the specimens was measured using microhardness tester and the specimen surfaces were observed by scanning electron microscope (SEM), confocal laser scanning microscope (CLSM). The results were analyzed using one-way ANOVA followed a Tukey' s multiple comparison at a significance level of 0.05 (SPSS Inc, Chicago, USA). Results: According to the fluoride application method, although iontophoresis groups showed higher VHN (difference in mean = 3.8±1.8) than those of the CFA groups, there were no significant differences between two method groups. After the modified pH-cycling, the 2% NaF solution with the iontophoresis group showed the best resistance to demineralization (p < 0.05). While 1.23% APF gel with the iontophoresis group showed the lowest resistance (p < 0.05). These results were also confirmed by SEM and CLSM. Conclusion: The fluoride iontophoresis method did not show any significant superiority to the conventional fluoride application in terms of the resistance to demineralization.