The purpose of this study was to compare the conventional acid-etch technique with an air abrasion surface preparation technique. Eighty freshly extracted noncarious human premolar teeth were randomly divided into the following 4 groups: (1) acid etched with 37% phosphoric acid for 15 seconds (Express Dental Products, Toronto, Canada), (2) sandblasted with 50 mu aluminum oxide by a microetcher (Danville Engineering Inc, Danville, Calif), (3) polished with pumice followed by acid etched with 37% phosphoric acid for 15 seconds, (4) sandblasted with 50 mu aluminum oxide by a microetcher followed by acid etched with 37% phosphoric acid for 15 seconds. All the groups had stainless steel brackets (Dentaurum, Standard Edgewise) bonded to the buccal surface of each tooth with no-mix adhesive (Express Dental Products, Toronto, Canada). A Lloyd testing machine (Lloyd Instrument LR 30K; Segensworth West, Foreham, UK) was used to determine tensile bond strengths at a crosshead speed of 0.5 mm/sec. The mean bond strength values of these are as follows: the only sandblasted group was 38.05 +/- 9.93 N; the only acid-etched group was 62.72 +/- 11.44 N; the group that was polished with pumice followed by acid etched was 69.78 +/- 14.87 N; and the group that was sandblasted followed by acid etched was 89.31 +/- 13.34 N. The statistical analysis was done by an analysis of variance and Scheffe test. The sandblasting followed by acid etching group had significantly higher bond strength values when compared to the other 3 groups, This study showed that sandblasting should be followed by acid etching to produce enamel surfaces with comparable bond strength. The current findings indicate that enamel surface preparation using sandblasting with a microetcher alone results in a significantly lower bond strength and should not be advocated for clinical use as an enamel conditioner.