While the posterolateral technique is the accepted fusion method for chronic mechanical back pain, the results are variable and the indications are narrow. In an attempt to resolve this problem the role of internal fixation has been explored. Relative stiffness and strength under flexion loading of four lumbosacral fixation systems-( I ) translaminar facet joint screws, (2) Luque rectangular box, (3) Luque fixation to the pelvis via the Galveston technique, and (4) two-part pelvic spinal rod systemwere evaluated and compared to the normal spine with and without posterior ligaments. The stiffness of the normal intact spine was found to be nonlinear: the bending moment is a function of the square of the flexion rotation. Internal fixation resulted in a twofold increase in stiffness. Facet screws were 20% stiffer and the pelvic-spinal system 2.3 times stiffer (p = 0.001) than the Luque systems but all had similar strength. A retrospective study of 135 consecutive posterolateral lumbosacral spine fusions was conducted to determine what factors improved the fusion rate, clinical results, and time to fusion. Diagnosis and previous surgical treatment, discectomy or fusion, were of no significance. The radio-graphic and clinical results were highly correlated (p = 0.0001). Electrical stimulation failed to improve the results. Internal fixation with facet joint screws or rods to the pelvis was found to statistically decrease the pseudarthrosis rate and reduce the time required for spine fusion (p = 0.02). The surgical technique consists of translaminar facet screws and segmental fixation to the pelvis using a new implant system.