The purpose of this laboratory study was to evaluate the tensile bond strength of a new generation of universal bonding systems to zirconia ceramic and to compare the results with the bond strength of a clinically-established bonding system. Eighty zirconia ceramic test specimens (e.max ZirCAD) were air-abraded and bonded to Plexiglas tubes, filled with an aliphatic dimethacrylate filling material (Clearfil F II), using three so called universal bonding systems of a new generation with different compositions (Monobond Plus/MultilinkAutomix, NX3, Scotchbond Universal/RelyX Ultimate). The latter was used also without the phosphate monomer containing primer Scotchbond Universal. A clinically established phosphate monomer containing adhesive cement served as control group (Panavia F2.0). The specimens were stored in water at 37°C for 3 or 150 days and the long-term storage series were additionally thermal cycled between 5 and 55°C for 37,500 times to simulate oral conditions. All specimens underwent tensile bond strength testing. The statistical analysis was performed using Kruskal-Wallis and Wilcoxon-Test with a Bonferroni-Holm correction for multiple testing. After 150 days the median bond strength of RelyX Ultimate, with and without Scotchbond Universal, and Panavia F2.0 did not differ statistically (range: 21.7-28.8MPa), while the bond strength of Monobond Plus/Multilink Automix was significantly lower (15.4MPa), and that of NX3 the lowest (6.6MPa). After 150 days of water storage with thermal cycling, all adhesive system showed significantly reduced tensile bond strengths compared to that after 3 days. Only RelyX Ultimate was comparable to the established bonding system Panavia F2.0. The additional use of Scotchbond Universal did not result in a significant effect.