OBJECTIVE The purpose of this study was to evaluate the resin-dentin interface morphology of 7 resin-based dentin adhesive systems (Solid Bond, EBS-Multi, PermaQuik, One Coat Bond, Gluma One Bond, Prime & Bond NT/NRC, and Clearfil Liner Bond 2V). METHODS Fourteen dentin disks 1.0 mm thick were obtained from superficial occlusal dentin of extracted human third molars, and finished with wet 600-grit silicon carbide paper. Two dentin disks were bonded using each of the adhesives above according to the manufacturers' instructions, and a thin layer of flowable resin composite was applied. The specimens were kept in tap water for 24 h at 37 degrees C, and then assigned to one of two observational techniques: a fracture technique and an acid-base technique. Fracture technique: shallow grooves were cut, fixed in 10% buffered formalin, and dehydrated in an ascending ethanol series up to 100%, critical-point dried, and fractured along the prepared grooves. Acid-base technique: the specimens were embedded in epoxy resin, sectioned through the center, polished with diamond paste down to 0.25-micron particle size, and treated with 10% orthophosphoric acid for 10 s and 5% sodium hypochlorite for 5 min. All the specimens were mounted on aluminum stubs, gold sputter coated, and observed using field-emission scanning electron microscopy (FE-SEM). RESULTS All the dentin adhesive systems showed hybrid layer formation, but the thickness varied depending on the bonding system used. The self-etching priming systems (Prime & Bond NT/NRC and Clearfil Liner Bond 2V) showed the thinnest hybrid layer at 1 to 2 microns, whereas the "single-bottle" system (Gluma One Bond) exhibited the thickest hybrid layer at 8 to 16 microns. CONCLUSIONS The ultramorphological structures of dentin bonding systems are determined by the composition of each system. Characterization of the interface of the adhesive system using the fracture technique provides additional information regarding the pattern of resin infiltration in some dentin bonding systems.