This invitro study designed to investigate the effect of bonding agent on amalgam repaired with either spherical or admixed alloy. Material and Methods: Cl I cavity preparations were made on the occlusal surface of hundred and twelve extracted human premolars. The teeth were randomly divided into two groups, 56 cavities filled with spherical amalgam and the other 56 cavities filled with admixed amalgam, after aging of restorations, buccal half of the previously placed restorations were removed and each group was divided into two subgroup according to the lining material used. The final number were eight groups of 14 teeth in each. All specimens were thermocycled for 500 times with 15 second dwell time. The samples were stained with 2% methylene blue, sectioned and evaluated for dye penetration. The data were analyzed by Fisher Freeman Halton test.Results: There was a significant differences in microleakage of bonded and non-bonded specimens, the use of bonding agent significantly increase the microleakage of repaired amalgam. Repairing amalgam with similar alloy type significantly reduce the microleakage.Conclusions: Within the limitation of this study, repairing amalgam with bonding agent is acceptable but significantly less microleakage can observed when amalgam repaired without bonding agents. malgam alloy can repaired with different types of alloy, but better marginal seal accomplished when the same types of alloy used for repair