The study verified the bond strength in simulated dental restorations of silorane- or methacrylate-based composites repaired with methacrylate-based composite. Methacrylate- (P60) or silorane-based (P90) composites were used associated with adhesive (Adper Single Bond 2). Twenty-four hemi-hourglass-shaped samples were repaired with each composite (n=12). Samples were divided according to groups: G1= P60 + Adper Single Bond 2+ P60; G2= P60 + Adper Single Bond 2 + P60 + thermocycling; G3= P90 + Adper Single Bond 2 + P60; and G4= P90 + Adper Single Bond 2 + P60 + thermocycling. G1 and G3 were submitted to tensile test 24 h after repair procedure, and G2 and G4 after submitted to 5,000 thermocycles at 5 and 55 ?#61616;C for 30 s in each bath. Tensile bond strength test was accomplished in an universal testing machine at crosshead speed of 0.5 mm/min. Data (MPa) were analyzed by two-way ANOVA and Tukey's test (5%). Sample failure pattern (adhesive, cohesive in resin or mixed) was evaluated by stereomicroscope at 30?#61655; and images were obtained in SEM. Bond strength values of methacrylate-based composite samples repaired with methacrylate-based composite (G1 and G2) were greater than for silorane-based samples (G3 and G4). Thermocycling decreased the bond strength values for both composites. All groups showed predominance of adhesive failures and no cohesive failure in composite resin was observed. In conclusion, higher bond strength values were observed in methacrylate-based resin samples and greater percentage of adhesive failures in silorane-based resin samples, both composites repaired with methacrylate-based resin.