Methods: The facial and lingual surfaces of forty caries-free human molars were wet ground flat with 1200-grit SiC paper. The teeth were randomly assigned into five groups. Each adhesive system was applied dentin and the respective composite resin was applied using a Teflon mold. The specimens were sheared at a crosshead speed of 1 mm/min after 24 h of storage at 37°C in water. Five additional teeth were prepared for SEM.
Results: Values in MPa were: Group I: 33.23 ±12.67; Group II: 32.41 ±9.90; Group III: 17.50 ±4.24; Group IV: 21.37 ±5.87; and Group V: 30.68 ±4.08. The statistical analysis by Kruskal-Wallis and Wilcoxon Rank Sum tests, revealed that there was no significant difference (P > .05) between the groups I, II and V. The groups III and IV were different from the others and to each other (P < .05). The debonded areas were observed in stereomicroscope and the fracture modes were mostly cohesive in resin. SEM analyses of the resin-dentin interface showed a homogeneous gap-free hybrid layer for all observed groups. Conclusions: The adhesives systems tested showed goods results with the shear bond strength test and higher values to those with two steps of application.