Objectives: Evaluate the influence of composite viscosity, adhesive systems curing mode and artificial aging on marginal sealing of composite resin restorations. Methods: 240 cavities were prepared in bovine incisors. Teeth were divided in two groups: GR - GrandioSO (Voco), conventional viscosity and GF - GrandioSO Heavy Flow (Voco), low viscosity. Each group was divided into 6 subgroups: FM - Futura Bond M (Voco), FDCL and FDCC - Futura Bond DC (Voco), light and chemical cured, CS - Clearfil S-3 Bond (Kuraray), CDCL and CDCC - Cleafil DC (Kuraray), light and chemical cured. Half of all specimens were subjected to mechanical (300.000) and thermal cycling (1.000). Marginal gap was measured in an optical microscope. For microleakage, specimens were stained in 50% ammoniacal silver nitrate. Results: For marginal gap mean (m), significant differences were shown for all factors (p=0.000). Results for resin were GR: 28.1 and GF: 40.6; for adhesive CDCC: 28.67, CDCL: 32.25, CS: 34.12, FDCL: 36.10, FM: 36.50 and FDCC: 38.46; and for artificial aging WITHOUT: 29.05 and WITH: 39.65. For microleakage mean (mm), there were also differences for all factors (p=0.000). Results for resin were GR: 2.68 and GF: 3.46; for adhesive FM: 2.85, CDCC: 3.00, FDCL: 3.08, CDCL: 3.15, CS: 3.16 and FDCC: 3.19; and for artificial aging WITHOUT: 2.66 and WITH: 3.48. Conclusion: Composite with conventional viscosity resulted in narrower marginal gaps and less microleakage. CDCC showed less marginal gap formation. The presence of thermo-mechanical cycling increased the values of gap and microleakage.