Statement of problem. Residual provisional cement and debris on prepared teeth may have the potential to negatively influence the performance of the definitive luting agent. Purpose. This in vivo study quantified the adherence of provisional cement to abutment teeth prepared with 1 of 2 textures (coarse or fine) and cleaned with 1 of 3 common cleansing techniques. Material and methods. One hundred ten provisional restorations in 22 patients were luted to maxillary anterior abutment teeth. The teeth were prepared for complete veneer restorations with either a coarse-grit diamond bur or a coarse-grit diamond bur followed by a fine-grit diamond bur. After removal of the restoration, abutment teeth were randomly cleansed with either a No. 23 dental explorer and air-water spray, a prophy cup with fine flour pumice, or a cotton pellet soaked in chlorhexidine gluconate, 0.12%. After the abutments were cleansed, they were evaluated intraorally under a light microscope (magnification, X64) by 2 blinded examiners. The teeth were given scores based on the number of specks of residual cement found on the surface after the cleansing techniques were performed. Pearson correlation coefficients (alpha=.96), 3-way analysis of variance, and post hoc Scheffe tests (P<.05) were used to analyze the data. Results. As determined with Pearson correlation coefficient, the interexaminer reliability was alpha=.96. Three-way analysis of variance revealed significant effects for the cleansing techniques but not for the interaction between cleansing techniques and preparation texture. A post hoc Scheffe test showed that the pumice cleansing technique (1.8 specks/tooth) was significantly better than the explorer or the cotton pellet/chlorhexidine gluconate technique (3.6 and 3.5 specks/tooth, respectively) (P<.05). Conclusion. Within the limitations of this study, abutment teeth cleansed with a prophy cup and flour pumice exhibited the least amount of residual provisional cement.