The purpose of this study was to assess the bond strength of two self-etching primers (SEP) to ground and unground enamel. Seventy-two bovine incisors were used in this study. The buccal enamel surface of 36 teeth was ground flat to resemble freshly cut enamel. The rest of the teeth were left intact. Two SEPs - Clearfil SE Bond, Kuraray (CSE) and Prompt L-Pop (3M ESPE) (LP) - and a conventional adhesive system, Scotchbond Multi-Purpose (3M ESPE) (SBMP) as a control, were used to bond a composite button to prepared and unprepared enamel. Microtensile test specimens were trimmed, resulting in a cylindrical cross-sectional area (0.21 mm2 to 0.47 mm2). These specimens were subjected to a tensile force at 1 mm/min until failure. Differences between adhesives and surface preparation were determined by two-way ANOVA. The samples were observed under SEM to evaluate the mode of failure. Bond strength values in MPa (SD) obtained from pooled data in descending order were: SBMP ground 44.54 (5.96), LP unground 42.97 (7.90), CSE unground 41.67 (11.28), LP ground 41.07 (12.07), CSE ground 38.56 (8.78), and SBMP unground 37.60 (9.55). No statistically significant differences were found (p = 0.5061) between surface preparation or adhesive systems. The mean in MPa (SD) of all the specimens that failed at the adhesive joint were: LP unground 47.13 (14.65), SBMP ground 45.28 (7.33), CSE unground 41.40 (11.07), SBMP unground 41.1 (10.04), CSE ground 39.96 (11.83), and LP ground 39.92 (15.45). No statistically significant differences were found (p = 0.5863). Failure occurred mainly at the adhesive interface. Surface preparation and adhesive treatment had no influence on resin composite microtensile bond strength to bovine enamel.