We determined whether spinal manipulation could prevent and/or reverse the decrease and increase in paraspinal muscle spindle responsiveness caused respectively by lengthening and shortening histories of the lumbar muscles. Single unit spindle activity from multifidus and longissimus muscles was recorded in the L6 dorsal root in anesthetized cats. Muscle history was created and spinal manipulation delivered (thrust amplitude: 1.0mm, duration: 100ms) using a feedback-controlled motor attached to the L6 spinous process. Muscle spindle discharge to a fixed vertebral position (static test) and to vertebral movement (dynamic test) was evaluated following the lengthening and shortening histories. For the static test, changes in muscle spindle responsiveness were significantly less when spinal manipulation followed muscle history (p<0.01), but not when spinal manipulation preceded it (p>0.05). For the dynamic test, spinal manipulation did not significantly affect the history-induced change in muscle spindle responsiveness. Spinal manipulation may partially reverse the effects of muscle history on muscle spindle signaling of vertebral position.