The carving of a complex amalgam restoration may occasionally result in light proximal contact with the adjacent tooth. The purpose of this study was to investigate the strength of complex amalgam restorations repaired with a proximal slot amalgam preparation. Extracted human third molars of similar coronal size were sectioned 1 mm apical to the height of the contour using a saw and were randomly distributed into 9 groups of 10 teeth each. One pin was placed at each line angle of the flattened dentinal tooth surface. A metal matrix band was placed and an admixed alloy was condensed and carved to create a full crown contour but with a flat occlusal surface. A proximal slot was prepared with or without a retention groove and repaired using a single-composition spherical amalgam 15 minutes, 24 hours, one week, or six months after the initial crown condensation. The specimens were stored for 24 hours in 37°C water before fracture at the marginal ridge using a round-ended blade in a universal testing machine. The control group was not repaired. The mean maximum force in newtons and standard deviation were determined per group. Data were analyzed with a 2-way analysis of variance as well as Tukey and Dunnett tests (α=0.05). Significant differences were found between groups based on type of slot preparation (p=0.017) but not on time (p=0.327), with no significant interaction (p=0.152). No significant difference in the strength of the marginal ridge was found between any repair group and the unrepaired control group (p>0.076). The proximal repair strength of a complex amalgam restoration was not significantly different from an unrepaired amalgam crown. Placing a retention groove in the proximal slot preparation resulted in significantly greater fracture strength than a slot with no retention grooves. Time of repair had no significant effect on the strength of the repair.