Purpose: To evaluate the effect of saliva contamination on the microleakage of an etch-and-rinse adhesive and a selfetching adhesive. Materials and Methods: Class V cavity preparations at the cementoenamel junction were made on the buccal and lingual surfaces of 40 freshly extracted human molars. The teeth were randomly assigned into two groups for different adhesives: Group I, Single Bond (two-step etch-and-rinse adhesive) and Group II, Futura Bond NR (one-step self-etching adhesive). The adhesive materials were applied according to their manufacturers' recommendations. The specimens were then further subdivided into five subgroups (A-E; n=8): group A, contamination before adhesive curing; group B, contamination before adhesive curing and re-application of adhesive; group C, contamination after adhesive curing; group D, contamination after adhesive curing and re-application of adhesive; and group E, no contamination (control). Contamination of adhesive surfaces was performed with fresh natural saliva and the saliva was removed by a gentle air blast. The specimens were restored with Z250 hybrid composite. After thermocycling (500x, 5°C to 55°C) and immersion in 0.5% basic fuchsin, the dye penetration was evaluated under a stereomicroscope. Statistical analysis was performed with the Kruskal-Wallis test at p 0.05. Results: No statistically significant differences were found among the contamination subgroups on the enamel or on the dentin for either group (p > 0.05). There was no significant difference in leakage between Group I and Group II for any subgroup (p > 0.05). Conclusion: Contamination of adhesives with saliva before and after adhesive curing did not worsen the microleakage of the two-step etch-and-rinse adhesive Single Bond or the one-step self-etching adhesive Futura Bond NR. Keywords: saliva contamination, etch-and-rinse adhesive, self-etching adhesive, microleakage