The aim of this study was to evaluate the marginal adaptation of high-viscosity resin composite restoratives bonded to dentin in a cylindrical cavity model. The buccal enamel of 64 human premolars was removed and cylindrical cavities 3 mm in diameter and 1.3 mm in depth were prepared on each dentin surface. The cavities were divided into 8 groups of 8 cavities each and restored according to the manufacturers' instructions with the following adhesive/composite systems: Bond 1/Alert, Stae/Glacier, OptiBond Solo/Prodigy Condensable, One-Step/Pyramid, Solidbond/Solitaire, Prime&Bond NT/Surefil, One Coat Bond/Synergy, and Scotchbond 1/Z250. The composite surfaces were pressed against mylar strips, covered with cover slips, and photopolymerized in a single increment for 40 s. The restorations were polished with wet SiC papers of 320 to 1000 grit size to expose dentin margins. The marginal adaptation was evaluated immediately after photopolymerization and again after 1 week of storage in water at 37 ± 1°C. Evaluation was performed under a metallographic microscope at 200X magnification by recording the frequency of gap-free restorations (GF), the percentage length of the debonded margins relative to the cavity periphery (DM), the width of the maximum marginal gap (MG), and the marginal index (MI = MG x DM / 100). The results were statistically analyzed with one-way ANOVA and the Mann-Whitney U-test at α = 0.05. No incidence of gaps was found in 62.5% of One Coat Bond/Synergy and 37.5% of OptiBond Solo/Prodigy Condensable restorations. All the other restorative systems exhibited restorations with gaps. One Coat Bond/Synergy, Scotchbond 1/Z250, and OptiBond Solo/Prodigy Condensable were the groups with the lowest DM values, while Stae/Glacier showed the highest DM values. One Coat Bond/Synergy and OptiBond Solo/Prodigy Condensable revealed the lowest MI values and Stae/Glacier the highest. No statistically significant differences were recorded between immediate evaluation and evaluation after 1 week of water storage. The results showed that all the adhesive/composite systems examined failed to achieve gap-free margins with dentin. The restorative systems One Coat Bond/Synergy, Optibond Solo/Prodigy Condensable, and Scotchbond 1/Z250 presented the best marginal adaptation of all the systems tested. No improvement was recorded on the quality of marginal adaptation following 1 week of water storage. Keywords: high-viscosity resin composites, marginal adaptation, dentin bonding