PURPOSE To evaluate in vitro any increase in size of Class V resin-based composite (RBC) restorations as a result of replacement. MATERIALS AND METHODS For group 1, 20 Class V cavity preparations were prepared in extracted incisor and canine teeth. The cavosurface margin was prepared as a butt joint 90-degree angle toward the occlusal wall. Impressions were made of all preparations with a polyvinyl siloxane material. The cavities then received a 1.5-mm 45-degree bevel at the cavosurface margin of the occlusal wall. Impressions were again made of the preparations. The cavities were then restored with a conventional multistep bonding system (Scotchbond Multi-Purpose, 3M Dental Products, St. Paul, Minnesota) and a microfilled RBC (Silux Plus, 3M Dental Products). As a second part of the study (group 2), another 20 Class V cavity preparations were prepared. Half of the cavities received a bevel at the cavosurface margin of the occlusal wall, and half were finished with a butt joint. Impressions were made from all 20 specimens followed by resin-based composite restoration in the same manner as previously described. All the restorations from both group 1 and group 2 were then removed with carbide burs. New impressions were made, and 100 stone models were fabricated. The perimeter and area of the cavity preparations, including the beveled area, were recorded using a video-based imaging system equipped with macro lens, at 10 to 15 times magnification. Both perimeter and area of the postoperative cavity preparation were compared to the initial cavity size for both butt joint and bevel finishing. RESULTS Intraclass correlation coefficient and paired t-test showed no significant difference between the preoperative bevel type of cavity preparation and its postoperative counterparts for both group 1 and group 2 (p > .05). However, a significant increase in the size of the cavity preparation was found preoperatively when a bevel was placed at the cavosurface margin. Postoperatively, a significant increase in size of the cavities was found for butt joint type finishing for both area (p = .0001) and perimeter (p = .0001), compared to initial cavity size.