Purpose: The purpose of this study was to assess the marginal integrity of composite and glass-ionomer restorations as a function of finishing technique, restorative material and margin location. Materials and Methods: Forty extracted third molars free of defects were assigned to four groups (N=10) according to finishing instruments (aluminum oxide discs, fluted carbides, fine diamonds, and coarse diamond). Each specimen received standardized Class V preparations on the facial and lingual surfaces with occlusal margins on enamel and gingival margins on dentin. Each preparation was randomly assigned to be restored with either resin-based composite (RBC) or resin-modified glass ionomer cement (RMGIC). Specimens were finished with standardized pressure at approximately 0.15 N and evaluated at a magnification of 600X using an environmental scanning electron microscope (ESEM). Occlusal and gingival margins were analyzed using an imaging software and means for all measured gaps were calculated. Data were analyzed using a linear regression using generalized estimating model. Result: There were no statistically significant differences among the four types of finishing instruments used in the study. . RBC-restored specimens exhibited significantly smaller mean marginal gaps (1.70 μm, 7.56 μm) than RMGI-restored specimens (5.24 μm, 14.24 μm) in enamel and dentin margins, respectively. There was a statistically significant difference between enamel and dentin with regards to marginal gap formation. Conclusion: Under the conditions of this study, marginal gap formation was not affected by finishing technique. Resin-based composite margins exhibited significantly less marginal gap than did resin-modified glass ionomer margins, while enamel margins resulted in significantly less marginal gap than did dentin margins.