The purpose of this study was to evaluate the in vitro dentin bond strengths of two dentin bonding systems in Class I cavities following fatigue load cycling (FLC) with thermal cycling (TC). Class I cavities were prepared in 12 human molars and restored with either Clearfil SE Bond (SE) and Clearfil AP-X resin composite (AP-X) or Single Bond (SB)/AP-X according to the manufacturers' instructions. After water storage for 1 week, untreated controls and test samples were subjected to 50,000 FLC with TC from 5°C to 55°C for 625 cycles and stored in water for 1 week. Microtensile bond strength (MTBS) tests were measured on the floor of the cavities at a crosshead speed of 1 mm/min. Data were statistically analyzed using one- and two-way ANOVA and Fisher's PLSD test at a 5% level of significance. The MPa results were: SE 30.5 ± 12.8 (control) and 23.2 ± 10.8 (50,000 FLC); SB 5.1 ± 9.2 (control) and 4.7 ± 5.7 (50,000 FLC). Bond strengths were not statistically significantly affected by FLC (p > 0.05); however, they were influenced by the bonding system (p 0.05). The MTBS of SE was significantly higher than that of SB (p 0.05) under FLC and TC test conditions. Over half of the SB specimens debonded during sample preparation, showing blister-like structures, suggesting that an over-wet phenomenon occurred on the cavity floor. SE produced excellent dentin bonds that were much stronger than those produced by SB, which was shown to be a technique-sensitive bonding system.