The purpose of this study was to assess, in vitro, the microleakage of Class V composite resin restorations following sealing of the margins with a fissure sealant or the resin component of a dentinal bonding agent, and to compare it to the microleakage of an unsealed control group. All groups of teeth showed some degree of leakage at the gingival margin. Statistical analysis showed that there was significantly less leakage when the fissure sealant (P .01) and bonding agent (P .001) were used than there was in the control group, in which the margins had not been sealed. There was no statistically significant difference between the two experimental groups. There was no leakage at the enamel margin in any of the teeth.