Purpose: To evaluate the effect of chemical and microbiological methods of caries induction on microtensile bond strength (μTBS) of current adhesive systems to primary dentin. Materials and Methods: Flat dentin surfaces from 36 primary molars were assigned to 3 groups according to the method of inducing caries-affected dentin: (1) control (sound dentin); (2) pH cycling; and (3) microbiological. In both methods, teeth were submitted to caries induction for 14 days, and the sound dentin was stored in distilled water for the same period. Specimens were then randomly reassigned according to adhesive system: a two-step etch-and-rinse adhesive (Adper Single Bond 2) or a two-step self-etching system (Clearfil SE Bond). Composite buildups were constructed and the teeth were sectioned to obtain bonded sticks (0.8 mm2) to be tested for microtensile bond strength. The μTBS means were analyzed by two-way ANOVA and Tukey's tests (α = 0.05). Failure mode was evaluated using a stereomicroscope (400X). Results: Both methods of caries induction resulted in lower μTBS values (with no significant difference between them) than those obtained for sound dentin. Adhesive systems showed similar bond strength values. The percentage of premature failure was higher in the microbiological group, regardless of adhesive system. Conclusion: Microbiological and pH-cycling methods are both suitable for simulating caries-affected dentin for bonding evaluations in primary teeth. Keywords: caries-affected dentin, artificial caries, pH cycling, microbiology, microtensile