Attempts to stimulate periodontal regeneration in the past have focused on either filling the defect with some type of material or providing a space for host cells to repopulate the site and elicit new tissue. In some cases, these approaches have been combined with the assumption that the filler material will help maintain the space necessary for the host cells to invade the area. Growth stimulating substances such as growth factors and other proteins have also been used to encourage periodontal tissue regeneration and histological evaluation supports the use of these substances. Thus, the role for and the necessity of a certain amount of space maintenance for periodontal regeneration is not exactly understood. In addition, it is not known if there is some critical size required for space maintenance or for exactly how long the space must be maintained in order for the host cells to stimulate new cementum, periodontal ligament, and bone. The goal of this study was to evaluate periodontal regeneration in intrabony defects of various sizes treated with a combination of enamel matrix proteins and autogenous bone graft.Periodontal defects ranging in size from 1 to 6 mm were randomized and created bilaterally beside three teeth in the mandibles of baboons. Plaque was allowed to accumulate around wire ligatures placed into the defects. After 2 months, the wire ligatures were removed, the teeth and roots scaled and root planed, and a notch was placed with a chisel at the base of the defect. On one side of the mandible, neutral ethylene diamine tetracetic acid and enamel matrix derivative (EMD) were first used to treat the defect. Autogenous bone taken from the same surgical site was treated with enamel matrix derivative in a dampen dish and then added to the EMD-treated defects. The other side of the mandible served as control with neutral ethylene diamine tetracetic acid and scaling and root planing. Flaps were sutured and the animals were allowed to heal without oral hygiene procedures. After 5 months, the animals were sacrificed and the teeth were processed for histological evaluation.The results revealed new cementum, periodontal ligament with Sharpey's fibers, and new bone tissue similar to native periodontal tissues. Remnants of the autogenous bone chips were still present at this 5-month post-healing period. Thus periodontal regeneration occurred in all sizes of the periodontal defects. In general, EMD plus autogenous graft treatment resulted in greater tissue formation than controls. In fact, in many cases, very dramatic tissue formation occurred far coronal to the base of the defects in the EMD plus autogenous graft-treated lesions. In addition, horizontal bone fill occurred in the defects and was prominent in the 4 or 6 mm wide lesions. When evaluating the combined 1 and 2 mm defects, the height of new cementum with EMD plus graft was 3.88 mm versus 2.03 mm in the controls, a statistically significant (P0.005) difference. In the wider (4 and 6 mm) lesions, this difference was not significant and was much less between treated and control lesions with 2.78 and 2.57 mm of new cementum respectively. In the case of new bone height, in the smaller lesions EMD plus graft resulted in 4.00 mm new bone versus 2.22 mm in the controls, again a statistically significant (P0.005) difference. In the larger lesions, EMD plus autogenous bone graft had 3.24 mm new bone height compared to 2.71 mm in the controls, a difference that was not statistically significant. Additionally, in the smaller lesions, new cementum width at the level of the notch was twice as great (statistically significant, P0.015) in the EMD plus graft sites compared to control. The width of the periodontal ligament at the coronal aspect of the new bone tissue was similar in the smaller lesions between treated and control sites. The results from the wider defects must be interpreted cautiously as the interproximal bone heights were remodeled adjacent to the wider defects and likely limited the potential for regeneration.The combination of enamel matrix derivative plus autogenous bone graft stimulated statistically significant periodontal regeneration in the more narrow 1 and 2 mm lesions. No statistically significant difference was observed in the wider 4 and 6 mm lesions. In many cases, dramatic amounts of new cementum, Sharpey's fibers, periodontal ligament, and bone tissue were formed far above the notch placed at the base of the contaminated defects. This was especially significant considering the width of some of the defects and the fact that no oral hygiene was performed over the 5-month healing period. This periodontal regeneration occurred in the absence of exogenous growth factors or barrier membranes. In summary, the combination of enamel matrix derivative and autogenous bone represents a therapeutic combination that can be highly effective in stimulating significant amounts of periodontal regeneration.