Six of the eight deceased persons were identified based on various methods of comparing antemortem and postmortem material. This included implant comparisons in one case, bone comparisons in three cases, and implant plus bone comparisons in two cases, with additional antemortem healing fractures found in two of these cases. In the two cases in which identifications were not made, inquiries by the police to the manufacturers of the implants were unsuccessful in obtaining identifying information. In addition, the skeletal material associated with these implants was limited, consisting of fragments of long-bone shafts only. This restricted the usefulness of an anthropological evaluation for sex, age, or other individual characteristics. The authors advise that although radiographs of implanted medical devices and unique bone characteristics may be critical in establishing the identity of deceased persons, as was shown in the reported cases, significant problems remain. The validity of comparisons has been questioned, and the lack of individual identifiers on orthopedic prostheses limits the identification of deceased individuals from medical records. The selection of the eight cases derived from an examination of the files of Forensic Science SA in Adelaide, South Australia, from May 1994 to April 2004. The search focused on cases in which the identification of the deceased person was complicated by processes of decomposition, skeletonization, and/or incineration, or by the finding of fragmented incomplete skeletons. In the eight cases selected, the bones of the individuals were examined radiographically for individual features and orthopedic implants. 9 figures and 29 references