Abstract: Objective To evaluate the effect of demineralized cancellous bone (DCB) seeded with allogeneic chondrocytes for repairing articular osteochondral defects in rabbits. Methods Articular chondrocytes were isolated from a 1-month-old male New Zealand rabbit for primary culture. The passage 1 chondrocytes were seeded onto prepared rabbit DCB scaffold to construct tissue-engineered cartilage and cultured in vitro for 2 weeks. Full-thickness articular osteochondral defects (3 mm both in diameter and depth) were created on both sides of the femoral medial condyles in 30 New Zealand white rabbits (age 4- 5 months). In 20 of the rabbits, the defects were filled with the tissue-engineered cartilage on the right side (group A) and with DCB only on the left side (group B); the remaining 10 rabbits did not receive any implantation in the defects to serve as the control (group C). At 1, 3, and 6 months after the implantation, tissue samples were collected from the defects for macroscopic observation and histological examination with Toluidine blue (TB) and collagen type II staining. The effect of defect repair using the tissue-engineered cartilage was assessed at 6 months based on the histological scores. Results The prepared DCB had a spongy 3D structure with open and interconnected micropores of various sizes and showed good plasticity and mechanical strength. DCB began to degrade within 1 month after implantation and was totally absorbed at 3 months. At 6 months after implantation, the defects filled with the chondrocyte-seeded DCB were repaired mainly by hyaline-like cartilage tissues, which were well integrated to the adjacent cartilage without clear boundaries and difficult to recognize. The chondrocytes were located in the lacunate and arranged in vertical columns in the deep repaired tissue, where matrix proteoglycans and collagen type II were distributed homogeneously close to the normal cartilage. The subchondral bone plate was reconstructed completely. The defects implanted with DCB only were filled with fibrocartilage tissue, as compared with fibrous tissue in the control defects. The histological scores in group A were significantly superior to those in group B and C (P<0.05), but the scores for subchondral bone plate reconstruction were comparable between groups A and B at 6 months. Conclusion DCB is a good scaffold material for preparing tissue-engineered cartilage, and chondrocyte- seeded DCB can repair articular osteochondral defects by inducing the generation of hayline-like cartilage.