Part I of the evaluation of a prototype bioabsorbable physical barrier composed of a copolymer of lactide and glycolide for treatment of bone defects in the guided bone regeneration procedure indicated that the prototype bioabsorbable physical barrier did not possess sufficient spacemaking characteristics to prevent collaspe of the barrier into the defect or against the threads of the titanium implants. The purpose of Part II was to evaluate this bioabsorbable physical barrier in combination with a supporting material to prevent barrier collaspe. Posterior mandibular teeth in three dogs were extracted and allowed to heal for 3 months. This produced localized alveolar ridge defects with a narrow buccolingual width. Six titanium threaded implants were placed in the right and left mandibles of each dog so that nonspacemaking dehiscence-type defects were produced. Two defects in each animal were randomly treated with the prototype bioabsorbable physical barrier and decalcified freeze-dried bone allograft; two defects were treated with a nonbioabsorbable expanded polytetrafluoroethylene barrier with decalcified freeze-dried bone allograft; and one defect each was treated with prototype bioabsorbable physcial barrier alone or by flap access with no barrier or bone replacement graft. The results demonstrated that both the bioabsorbable and the nonbioabsorbable barrier combined with decalcified freeze-dried bone allograft produce comparable amounts of new bone with percent bone-to-implant contact, height, width, and area. Defects treated with the prototype bioabsorbable physical barrier alone or no barrier demonstrated unfavorable results. It is suggested that a bone replacement graft is indicated when treating defects with a nonspacemaking morphology.