The aim of this in vitro study was to evaluate the influence of occlusal finish line configuration on microleakage of indirect composite inlays. Forty-five recently extracted premolars were randomly assigned into three groups with different types of occlusal finish line preparations: bevel, chamfer, and butt joint. All restorations were made in Artglass (Heraeus Kulzer) and luted according to the manufacturer's recommendations. The teeth were sealed and then thermocycled 500 times between 5°C and 55°C. After that they were immersed in a 2% methylene blue (pH 7.2) solution for 24 hours. Each tooth was bisectioned in the buccal/lingual direction and then examined microscopically at each section to evaluate the microleakage score. The data were analyzed with chi-square and Fisher's Exact tests. There were no significant differences between the buccal margins of maxillary premolars with occlusal bevel or butt joint; the chamfer had the worst performance. In the buccal margin of mandibular premolars, the bevel and chamfer showed significantly smaller amounts of leakage than the butt joint. No significant differences were observed between the three types of occlusal cavosurface preparations in the lingual margin of maxillary and mandibular premolars. The finish line configuration, the position of the teeth in the arch, and the margin (buccal vs lingual) influenced microleakage. Whenever microleakage was observed in this study, it was always observed at the tooth-cement interface.