This study evaluated the effectiveness of a dentin bonding agent as a barrier to prevent coronal microleakage and examined the effect of a eugenol-based sealer on the sealing ability of this resin adhesive. Fifty-one extracted human mandibular molars were incorporated in a model system using an oral streptococci as a microbial marker. Group 1 consisted of 15 teeth that were obturated with only gutta-percha and received a coronal barrier of Clearfil Liner Bond 2V. Group 2 was identical to group 1, but included the use of a eugenol-based sealer in the obturation. Group 3 consisted of 15 teeth that were obturated with gutta-percha and sealer, but did not receive a coronal barrier. Six teeth served as controls. Bacterial penetration was monitored for 90 days. Results were analyzed after 30, 60, and 90 days with Fisher's exact test (p < 0.05). All controls behaved as expected. Neither group 1 nor group 2 exhibited any bacterial leakage. Eleven of the 15 specimens in group 3 leaked between 15 and 76 days. The coronal barriers in group 1 and group 2 were significantly better in preventing coronal microleakage at 60 days (p = 0.002) and 90 days (p = 0.00005). The presence of eugenol in the sealer had no significant effect on the sealing ability of Clearfil Liner Bond 2V (p = 1).