OBJECTIVE The purpose of this study was to evaluate, by shear bond strength (SBS) testing, the influence of different types of temporary cements on the final cementation using conventional and self-etching resin-based luting cements. METHODS Forty human teeth divided in two halves were assigned to 8 groups (n=10): I and V (no temporary cementation); II and VI: Ca(OH)2-based cement; III and VII: zinc oxide (ZO)-based cement; IV and VIII: ZO-eugenol (ZOE)-based cement. Final cementation was done with RelyX ARC cement (groups I to IV) and RelyX Unicem cement (groups V to VIII). Data were analyzed statistically by ANOVA and Tukey's test at 5% significance level. RESULTS Means were (MPa): I - 3.80 (+/-1.481); II - 5.24 (+/-2.297); III - 6.98 (+/-1.885); IV - 6.54 (+/-1.459); V - 5.22 (+/-2.465); VI - 4.48 (+/-1.705); VII - 6.29 (+/-2.280); VIII - 2.47 (+/-2.076). Comparison of the groups that had the same temporary cementation (Groups II and VI; III and VII; IV and VIII) showed statistically significant difference (p<0.001) only between Groups IV and VIII, in which ZOE-based cements were used. The use of either Ca(OH)2-based (Groups II and VI) or ZO-based (Groups III and VII) cements showed no statistically significant difference (p>0.05) for the different luting cements (RelyX ARC and RelyX Unicem). The groups that had no temporary cementation (Groups I and V) did not differ significantly from each other either (p>0.05). CONCLUSIONS When temporary cementation was done with ZO- or ZOE-based cements and final cementation was done with RelyX ARC, there was an increase in the SBS compared to the control. In the groups cemented with RelyX Unicem, however, the use of a ZOE-based temporary cement affected negatively the SBS of the luting agent used for final cementation.