Purpose: To evaluate the effect of a polymerization accelerator on the microtensile bond strength (µTBS) of etch-and-rinse and self-etch adhesives to eugenol-contaminated dentin. Materials and Methods: Sixty flat dentin surfaces were prepared from human molars. Half of the specimens were restored with zinc oxide eugenol temporary cement (IRM) (eugenol-contaminated group) and the other half remained without restoration (control group). After 24-h storage, the cement was mechanically removed. Then the specimens in each group were further divided into three subgroups based on the application procedure of a polymerization accelerator (p-toluenesulfinic acid sodium salt; Accel): no application, 10-s application, or 30-s application. After air drying, the dentin surfaces were bonded with either a three-step etch-and-rinse adhesive (OptiBond FL) or a two-step self-etch adhesive (Clearfil SE Bond) and restored with composite. After 24-h water storage, the bonded specimens were subjected to the µTBS test. Data were analyzed by three-way ANOVA and Dunnett's T3 test (p 0.05). Results: The eugenol-contaminated groups had significantly lower µTBS than the control groups with both types of adhesives (p 0.05), and the application of Accel significantly increased the compromised µTBS to eugenol-contaminated dentin. Optibond FL presented significantly higher µTBS to eugenol-contaminated dentin than did Clearfil SE Bond (p 0.05). Conclusion: The application of a polymerization accelerator on eugenol-contaminated dentin prior to adhesive resin application increased the μTBS of both the three-step etch-and-rinse and two-step self-etch adhesive. Keywords: eugenol, polymerization accelerator, adhesive resin, bond strength