We observed the effect of application of n-PTFE on mandibular bone defects in the rabbit. GTR is based on the hypothesis that during wound healing, non-osteogenic cells are mechanically blocked from entering a bone defect. This allows the slower-migrating mesenchymal cells from the surrounding bone and marrow, having osteogenic potential, to move the defect site. Bilateral through & through defects, 8mmx4mm in size, involving the inferior border of mandible, were created in the mandible angle of 15 rats. The experimental side was applied with high-density n-PTFE membrane, with the opposite side serving as a control. In the first week of the early experimental period, the bone defects of experimental groups were filled with highly vascularized fibrous connective tissue showing prominent osteoblastic activity with osteoid formation, whereas the defects of control were replaced by dense fibro-muscular tissue without osteoblastlf activity. After 3 weeks, the experimental group revealed well formed bone trabeculae and fibro-vascular marrow within surrounding membrane. With time the amount of new bone was decreased with increase of hematopoietic marrow, and the cortical plate composed of dense mature bone was more thicker. In the whole experimental periods, the inflammation and foreign body reaction were not found around the inserted n-PTFE membrane. After removal of membrane there was marked remodeling of newly formed bone by active osteoclasts in periosteal soft tissue, which decreased the bone volume apparently. These findings suggest that n-PTFE has biocompatibility, flexibility, and rigidity and offer rapid bone regeneration within limited area by the membrane. On the other hand, the early removal of membrane caused reduction of bone volume by the osteoclastic resorption. Thus, the time of removal may be more important factor for maintenance of regenerated bone volume.