Adult dog skull defects larger than 17 mm do not spontaneously heal. A quest for a potentially viable, cosmetically, mechanically, and technically acceptable template for human cranial reconstruction prompted a comparison of processed autogeneic and allogeneic bone implants with a fresh autograft control in the dog. Quantitative reproducible observations demonstrated that fresh calvarial bone autografts were superior to the nonviable implants in volume percent defect filled, mm2 new cortical bone, mm2 new and old cortical bone, and cortical bone porosity. Frozen autografts achieved 75%, antigen-extracted, autolyzed, partially demineralized auto- and allografts, 50% of the overall efficiency of fresh autografts. Fresh cancellous bone added to allografts did not improve long-term repair. Remodeling of all grafts appeared consistent with osteoconductive invasion by peripheral host endosteal and diploic elements; host external periosteum and dura contributed less. Central osteoinductive recruitment of mesenchymal cells from muscle or dura seemed not to occur in the adult dog. Partially demineralized dog calvarial grafts were resorbed without acting as a template for new bone formation. Surface demineralization, antigen extraction, and autolytic digestion of autografts and allografts, with or without fresh iliac bone, did not improve calvarial bone regeneration in adult dogs.