PURPOSE To evaluate and compare the 3-year clinical performance of a silorane-based composite with that of a methacrylate-based composite in Class II restorations. MATERIALS AND METHODS Seventy-eight patients, each with two class II restorations under occlusion, were enrolled in this study. One hundred fifty-six restorations were placed, 50% for each material: a silorane-based composite, Filtek P90/ P90 System Adhesive and a methacrylate-based resin composite, Quixfil/ Prime &Bond NT. A single operator placed all restorations according to the manufacturers' instructions. Immediately after placement, the restorations were finished/polished. Clinical evaluation was performed at baseline and at yearly intervals after placement by two other independent examiners using slightly modified USPHS criteria. The changes in the USPHS parameters during the three-year period were analyzed with the Friedman test. The baseline scores were compared with those at the recall visits using the Wilcoxon signed rank test. The level of significance was set at p < 0.05. RESULTS All restorations were evaluated at 3 years. Post-operative sensitivity was observed in 6 patients (2 Filtek P90, 4 Quixfil) between 1 and 3 weeks. Seven failed restorations (4.5%) were observed during the follow up: 4 in the Filtek P90 (5.1%) and 3 in the Quixfil group (3.8%). This resulted in non-significantly different annual failure rates of 1.7% and 1.2%, respectively. Fracture of restoration was the main reason for failure. CONCLUSION After 3 years, no significant difference was seen in overall clinical effectiveness between the silorane- based and methacrylate-based composite restorative systems.