Using a stereoscopic clearing protocol and scanning electron microscopy, we investigated the extent and nature of microleakage in a total-etch, current-generation dentin adhesive by a wet-bonding technique under different handling conditions. The hypotheses were that inadequate light curing of the primer or incomplete drying of the primer solvent might adversely affect the sealing ability of an acetone-containing adhesive system. The study consisted of three experimental groups: (I) a control group with an adequate light source and with the primer solvent completely dried; (II) an "inadequate-light" group; and (III) an "incomplete evaporation of primer solvent" group. The extent of microleakage after silver staining and clearing of the specimens was scored based on a modified five-point scale. Nonparametric statistical analysis (Kruskal-Wallis ANOVA) followed by a multiple comparison test (Dunn test) indicated significant differences among the three groups (p < 0.05). SEM examination of the restorative interface revealed that microleakage appeared to be initiated from the bonding resin-hybrid layer interface in all three groups, representing the weak link in the adhesive system. In addition, microleakage was characterized by 5 zones, each delineating a stage in a continuous array of progressively deleterious microleakage patterns variously distributed among the three groups. It was suggested that, while the bonding resin-hybrid layer interface represented the intrinsic weakness in an already much improved dentin adhesive, extrinsic factors such as the adequacy of the curing light and, more importantly, complete removal of the primer solvents can and should be avoided to preserve the structural integrity of the marginal seal.