The present experiment was performed to determine the influence of Bio-Oss on hard tissue formation at sites that, following implant installation, presented a 1-1.25 mm wide marginal defect. Four Labrador dogs were used. The premolars and first molars on both sides of the mandible were extracted. After 3 months, mucoperiosteal flaps were elevated and three experimental sites were prepared for implant installation in each side of the mandible. A step drill was used to widen the marginal 5 mm of the canal. Thus, following the placement of the implant (3.3 x 10 mm, SLA surface, Straumann AG, Waldenburg, Switzerland) a circumferential gap, about 1-1.25 wide and 5 mm deep, remained lateral to the titanium rod. The test sites in the left side of the mandible were first filled with a deproteinized cancellous bone mineral (Bio-Oss). The defect sites in the right side of the mandible (control sites) were left for spontaneous healing. A resorbable barrier membrane (Bio-Gide) was placed to cover the implant and the bone tissue in two sites of each quadrant, while the third site was left without membrane placement. The flaps were repositioned to cover all defect sites and were sutured. After 4 months of healing, block biopsies of each implant site were dissected and processed for ground sectioning. It was demonstrated that at 4 months, all types of defects were filled with newly formed bone and that the biomaterial placed in the marginal defect in conjunction with implant installation during healing became incorporated in the newly formed bone tissue. A high degree of contact was established between the Bio-Oss particles and the newly formed bone. Bio-Oss became integrated with the newly formed bone. In the model used, Bio-Oss did not enhance the process of bone formation and defect closure.