Parasplnal muscle action was included in the simulation of flexlon and extension of the lumbar spine in an in-vitro biomechanical study. An axial preload was applied anteriorly to a lumbar spine specimen to simulate the body weight with balance being maintained by a pair of cables attached to the posterior of the specimen, simulating the action of the erector spinae muscle in vivo. In each simulated posture, the three-dimensional orientation of each vertebra and intradiscal pressures of intervertebral discs were determined and analyzed. Six human cadaveric specimens were tested. The results showed that the L4-5 segment was the principle contributor to flexion while in extension the L5-S1 level contributed most. The relative contribution of each segment to the total flexion and extension of the lumbar spine varied with level. These data suggested that inter-segmental mobility of each level in flexion and extension ranges should not be combined as a single quantity I)ut should be studied separately. This study also Indicates that muscle simulation is important for cadaveric study of the lumbar spine. The magnitudes of intradiscal pressures recorded in this study were found to be similar to that measured in vivo of normal lumbar discs. Although the authors recognize that the conditions simulated in this investigation are not truly physiological, they maintain that the spinal motions obtained in this study are closer to those occurring in vivo than would he the case if experiments were conducted without muscle simulation.