The aim of this study was to investigate the intra- and intertooth microtensile bond strength variability of two adhesive systems applied to dentin. Twenty human third molars were ground to expose a flat dentin surface. Two adhesive systems (Single Bond and One Step) were applied according to the manufacturers' instructions and composite resin crowns were constructed incrementally. After 24 h in distilled water at 37°C, each tooth was longitudinally sectioned in both "x" and "y" directions under 300 rpm to obtain sticks with a cross-sectional area of 0.8 mm². The sticks from each tooth were tested at a crosshead speed of 0.5 mm/min. The fracture pattern of each stick was microscopically analyzed after debonding (400X). A mixed model with adhesive as a fixed variable, tooth as a random variable, and bonded area included as a covariate, was employed for the data treatment. In this model, all observations have the same variance, which is the sum of two components: the variance of the random effect (intertooth) and the variance of the errors (intratooth). The intratooth variability was greater than the intertooth variability. The intratooth variability is higher than the intertooth variability. Sticks from the same tooth cannot be considered as an experimental unit, since it does not fulfill all ANOVA requirements. Keywords: adhesive systems, sticks, regional variation, microtensile bond strength, statistical analysis