Aims: a) To review donor site morbidity and record problems coming out during bone harvesting, b) to evaluate bone graft adequacy and effectiveness. Methods: Senior author operated 127 patients using cancellous bone autografts (from May 1988 to September 2001). Bone grafts were obtained from iliac crest (83 patients), proximal tibial metaphysis (39 patients) and other donor sites (5 patients). In all cases intraopera-tive problems and early postoperative problems were recorded. Late postoperative problems were evaluated either subjectively (questionnaire) or clinically for 109 patients. Minimum length of follow up was 12 months. Results: In all the cases, the goals of graft harvesting were accomplished. Generally, grafts were adequate in quantity and quality (apart from 2 cases in which grafts were taken from proximal tibial metaphysis). Blood loss from the proximal tibia was less than iliac bone. Iliac pain was more severe than tibial, resulting to delayed mobilization and pain could persist for a long period of time. Conclusions: Proximal tibial metaphysis, as a donor site, is associated with far less morbidity than iliac and it should be chosen in patients with lower-extremity fractures or nonunions. Harvesting from tibia is adequate and easier comparing to iliac. Tibial bone grafts were as effective as iliac, despite that iliac bone graft is widely used and generally referred as a better quality autologus bone graft.