Purpose: The aim of this in vitro study was to compare the dentin bond strength of directly and indirectly inserted composite restorations. Materials and Methods: Twenty extracted human mandibular molars were prepared to form a flat occlusal dentin surface using an Isomet saw under water. Five-mm-thick composite restorations (Filtek Z 250 and Clearfil AP-X) were placed on the dentin surfaces by direct or indirect techniques. The direct restorations were bonded with Single Bond and Clearfil SE Bond bonding agents. Indirect restorations were cemented with two composite luting cements (3M Opal Luting Cement and Panavia F). All restorative procedures were applied under simulated pulpal pressure (15 cm H2O). Each tooth was then vertically sectioned with an Isomet saw through the composite buildups and the dentin. Thus, for microtensile testing, 0.65 x 0.65 mm specimens were obtained from pulpal and remote dentin regions. The results were analyzed according to the regional dentinal surfaces (pulpal and remote dentin). For statistical analyses, two-way ANOVA and Duncan post-hoc tests were used (p 0.05). Results: Direct application of Clearfil AP-X showed that tensile bond strengths were higher than those obtained for directly and indirectly applied Filtek Z 250 and indirectly applied Clearfil AP-X. No significant difference in tensile bond strengths was demonstrated between pulpal and remote dentin (p > 0.05). Conclusion: It is concluded that the bond strengths of resin luting cements to dentin need further improvement. Keywords: microtensile bond strength, indirect inlay, direct posterior composite, pulpal pressure