Purpose: This in vitro study evaluated marginal integrity, 2-body wear, and fracture behavior of an array of bonded and nonbonded posterior restorative materials after thermomechanical loading (TML). Materials and Methods: Eighty-eight MOD cavities with one proximal box beneath the CEJ were prepared in extracted human third molars according to a well-established protocol. Direct restorations were made using the following materials: amalgam (Dispersalloy), Ketac Molar Quick, Surefil One (with or without light curing), Activa, AdheSE Universal/Heliomolar, Fuji II LC improved, Equia Forte, Scotchbond Universal/Filtek Supreme, Xeno V+/CeramX.mono+, Prime&Bond active/Spectra ST CeramX HV, Prime&Bond elect/Spectra ST CeramX HV. Before and after thermomechanical loading (2500/5000/12,500 thermocycles between 5°C and 55°C + 100,000/ 200,000/500,000 x 50 N), marginal gaps and 2-body wear depths were analyzed on epoxy resin replicas using SEM and CLSM. Fractures were observed under a light microscope (20X). Results were analyzed with Kruskal-Wallis and Mann-Whitney U-tests (p 0.05). Results: For marginal quality, Surefil One showed promising in vitro behavior close to that of resin composite bonded with a self-etch adhesive (p > 0.05). For wear, amalgam and resin composites with recent filler technology were still superior (p 0.05), but Surefil One LC outperformed Activa, Ketac Molar Quick, Equia Forte Fil, and Fuji II LC (p 0.05). When Surefil One was occlusally light cured, no fractures occured, even after 500,000 cycles of TML. Conclusion: The novel self-adhesive posterior restorative Surefil One did not exhibit superior outcomes for all evaluated aspects. However, it showed stable fracture behavior, good marginal quality, and acceptable wear resistance in vitro. Schlagwörter: amalgam alternatives, resin composites, resin-modified glass-ionomer cements, self-adhesive materials