The purpose of this study was to investigate whether a grafting technique using either periosteum or bone marrow as an adjunct, would reconstitute more favorable tendon anchorage morphology with improved tensile strength in a bone tunnel model. We hypothesized that autogenous bone marrow aspirate can enhance the tendon-bone attachment as well as a freshly harvested periosteum, because both tissues contain pluripotent cells. Thirty-six skeletally mature New Zealand white rabbits were utilized. For the tendon graft healing in a bone tunnel model, the extensor digitorum longus tendon was detached from its femoral insertion and transplanted through a bone tunnel into the proximal tibia. Three groups were compared. For the group P (periosteum), a periosteum-wrapped tendon was fixed into the tunnel through the proximal tibial metaphysis. For the group BM (bone marrow), instead of periosteum augmentation, fresh bone marrow was injected into the tendon graft that would sit inside the tunnel. For the group C (control), the limb underwent a similar operation with neither the periosteum enveloping nor bone marrow injecting the tendon. At 6 and 12 weeks after surgery, two rabbits were used for light and electron microscopic examinations, and ten rabbits were used for biomechanical tests in each group. The interface tissue between bone and tendon was thicker and less organized in group C compared to groups P and BM at 6 weeks. Ultra-structurally, the interface tissue was loosely organized in group C, compared to others. Bone ingrowth into tendon was more obvious in groups P and BM, compared to group C. The proliferation of cartilage islands was observed within bone tunnels of both groups P and BM; but a well-defined fibrocartilage zone was noted only in group BM at the interface at week 12. Biomechanical findings: (1) at 6 weeks, the average failure load of group P was significantly higher than the others (P < 0.01). At same time point, in terms of stiffness, while group P was significantly higher than the other groups (P < 0.01), group BM was also significantly higher than that of group C (P < 0.05); (2) at 12 weeks, in terms of failure loads, there was a statistical significant difference only between groups BM and C (P < 0.05). At the same time point, stiffness values were not statistically different among the three groups. Based on the histological and biomechanical findings, the present study demonstrated that periosteum had a positive effect when compared to bone marrow and control groups on the tendon-to-bone healing at an early time point (6 weeks), and bone marrow was also effective at 12 weeks time point compared to the control group in an extra-articular bone tunnel in rabbits. The presence of pluripotent cells in both the bone marrow and the periosteum may be the possible mechanism for enhanced healing. Periosteum had a positive effect at an early time point (6 weeks). Bone marrow was more effective at 12 weeks. Therefore, it is possible that a combination of wrapping periosteum and injecting bone marrow to the tendon graft would have a synergistic effect (early and strong). To prove this hypothesis, future studies which would combine both methods are needed.