Purpose: To estimate the in vitro reliability of typical self-etching and etch-and-rinse adhesives of various application protocols. Materials and Methods: The following adhesives were applied on flat dentin surfaces of extracted human teeth (n = 223): self-etching two-step adhesives: AdheSE (AH), Clearfil SE Bond (CL), OptiBond SE (OS); one-step adhesives: Adper Prompt L-Pop (ADP), Adper Prompt (AD), and Xeno III (XE); all-in-one adhesive: iBond (IB); etch-and-rinse threestep adhesives: OptiBond FL (OF), two-step Gluma Comfort Bond (G), Excite (E) and Prime & Bond NT (PB). Composite buildups were prepared using a microhybrid composite, Opticor New. Shear bond strength was determined after 24 h of storage at 37°C in distilled water. The results were analyzed with a nested ANOVA (adhesive, type of adhesive) followed by the Fisher post-hoc tests of group homogeneity at α = 0.05. A two-parameter Weibull distribution was used to calculate the critical shear bond strength corresponding to 5% probability of failure as a measure of system reliability. Results: ANOVA revealed a significant decrease (p 0.001) in the mean shear bond strength as follows: AH=CL=OS=G=E=OF>AD=IB=XE>PB=ADP, but no significant difference (p > 0.48) between the etch-and-rinse and self-etching adhesives. The corresponding characteristic bond strength of Weibull distribution ranged between 24.1 and 12.1 MPa, Weibull modulus between 8.3 and 2.1, and the critical shear bond strength varied from 16.0 to 3.0 MPa. Conclusion: Pronounced differences in the critical shear bond strength suggest reliability variations in the adhesive systems tested, which originate from chemical composition rather than type of adhesive. Keywords: bond strength, dentin adhesive system, reliability, Weibull analysis