The aims of this study were to test the tensile bond strength of a recently developed veneer. Sound premolar teeth (120) extracted for orthodontic purposes were divided into two experimental and two control groups. In one experimental group (V1) 4-META/MMA-TBB resin (4META) was used on the surface veneer prepared with micro particle filled resin (MFR) as an adhesive for bracketing and in the second group (V2) 4META was applied on the surface veneer with the trial resin. For the controls, in group R 4META was used on the enamel surface without veneer and in group G light-cured glass ionomer cement was applied. The 30 samples in each group were divided into three groups of 10 samples and thermal cycled (TC) at 3000, 10,000 or left uncycled. Tensile testing was carried out using an Instron machine. After tensile testing the bond failures in the experimental groups were recorded using a stereomicroscope. Statistical analysis was performed using ANOVA. In group V2 the resin veneer was able to maintain sufficient bond force to enamel during clinical use. The bond strength of group V1 was significantly higher than that of groups R (P < 0.05) and G (P < 0.01) at TC 0, but for both TC 3000 and 10,000, the bond strength of group V1 was lower than groups R and G, respectively. There were significant differences between groups V1 and R (P < 0.01) for TC 3000, and between groups V1 and R and G (P < 0.01) at TC 10,000. The bond strength of group V2 was almost equal to that of group R at TC 0. At TC 3000, group V2 showed significantly lower bond strength than group R (P < 0.05), but no significant difference was found compared with group G. At TC 10,000, there were no significant differences between groups V2, R or G. When comparing groups V1 and V2, the bond strength of group V1 was significantly higher than that of group V2 (P < 0.01) at TC 0, but the bond strength of group V1 was significantly lower than that of group V2 for both TC 3000 (P < 0.05) and TC 10,000 (P < 0.01). Comparison between groups R and G, showed that the bond strength of group R was significantly higher than that of group G for both TC 0 (P < 0.01) and TC 3000 (P < 0.01), but no significant difference was found for TC 10,000. In group V2, nine samples showed adhesive failure between the veneer surface and bracket adhesive before thermal cycling. There were significant differences between the MFR and both trial resin and glass ionomer cement (P < 0.01) when examining thermal expansion. No significant difference was found between the trial resin and glass ionomer cement. It is suggested that application of a resin veneer prior to bracket bonding is suitable for clinical application to protect the teeth and to prevent decalcification and caries.