The aim of this study was to evaluate the efficacy of collagen membranes, either alone or combined with a human demineralized freeze-dried bone allograft (DFDBA) or natural bovine bone graft, in bone defects around dental implants with an SLA (sand-blasted, large grit, acid-etched) surface. The experiments were carried out in three beagle dogs using a split-mouth design. On one side of the jaw, three implants were placed and intra-bony defects were created and covered with a collagen membrane, randomly combined in two of the defects with human DFDBA or inorganic bovine bone graft. A control implant, without membrane covering or defect filling, was also placed. On the other side of the jaw, three implants were placed and the bone defects were treated in a similar fashion, but without membrane covering. The studied variables were the percentage of bone-to-implant contact within the limits of the initial bony defect and percentage of the original bony defect occupied by bone tissue. Although no statistically significant differences were found in this study between the membrane and nonmembrane groups, bone defects augmented with anorganic bovine bone and membranes showed the most promising results from a histological and histomorphometric perspective.