Objective: The purpose of this study was to test the hypothesis that remineralized dentin lesions induced by glass ionomer are less vulnerable to subsequent acid challenge. Method and Materials: Baseline demineralized (BDe) lesions were created in 50 bovine dentin slices immersed for 3 weeks in acid solution. A resin-modified glass ionomer (RMGI) was applied to the specimens before immersion in salivalike solution to allow remineralization (Re) for 2 weeks (V2W) or 6 weeks (V6W). Resin-modifed glass ionomer was coated to block further ion release before a 3-week immersion in acid solution as a second demineralization (SDe). Control groups (C2W and C6W) were done simultaneously without RMGI. In another group, RMGI was left undisturbed (V2W+) to allow continuous ion release during SDe. Mineral content was converted from microradiographs, after BDe, Re, and SDe steps. Changes in mineral content were calculated and compared between groups. Results: The remineralized surface zone was maintained after SDe, but the lesion body became deeper. Percentage of mineral loss from SDe was not significantly different between V2W and C2W (t test, P > .05). When remineralization was extended to 6 weeks, V6W showed significantly less mineral loss than C6W. Mineral loss was lowest when RMGI was not blocked (V2W+). Conclusion: Dentin lesions remineralized in the presence of RMGI maintained the highly mineralized surface zone when subjected to second demineralization. However, the remineralized surface zone could not prevent advance of the lesion body. The subsequent demineralization was markedly reduced by extending the remineralization period or by leaving the RMGI undisturbed to resume ion release. Keywords: artificial caries, demineralization, dentin, glass ionomer, remineralization, subsequent acid challenge