We investigated the temporal relationship among the biomechanical, radiographic, and histological properties of a posterolateral spinal fusion mass to elucidate the changes in load-sharing of the spinal instrumentation and that of the fusion mass throughout the healing process. Destabilization of the posterior spinal column and transpedicular screw fixation at the segments between the third and fourth and the fifth and sixth lumbar vertebrae was performed in twenty-four sheep. A posterolateral spinal arthrodesis with use of autologous corticocancellous bone graft was done randomly at one of the two segments; the other segment (without bone graft) served as the instrumented control. Six animals each were killed at four, eight, twelve, and sixteen weeks postoperatively. Biomechanical testing showed that the posterolateral fusion mass had increased mechanical stiffness after the fourth week. The strain on the hardware, measured with use of rods instrumented with strain-gauges, decreased significantly (p < 0.01) beginning at eight weeks. Radiographically, three independent observations of each of the six animals at each time-period showed that, although all of the fusion masses were considered solid unions at sixteen weeks, bridging of trabecular bone was noted during only ten of eighteen observations at twelve weeks, three of eighteen observations at eight weeks, and none of eighteen observations at four weeks. Computerized tomography and histomorphometric analyses demonstrated that mineralization in the fusion mass increased in a linear fashion even after eight weeks. Histologically, the fusion mass consisted predominantly of woven bone at eight weeks; thereafter, it was gradually trabeculated.CLINICAL RELEVANCE: We found a great discrepancy between biomechanical stability and histological maturation of the posterolateral fusion mass. The biomechanical properties of a stable spinal fusion preceded the radiographic appearance of a solid fusion by at least eight weeks, suggesting that immature woven bone provided substantial stiffness to the fusion mass. The spinal instrumentation was subjected predominantly to bending stress rather than to axial stress, and the load-sharing of the spinal instrumentation decreased concurrently with the development of the spinal fusion.