OBJECTIVE: The purpose of this study is to develop an experimental spinal cord injury spasticity model using rabbits and to evaluate a quantitative spasticity measure. METHOD: After an general anesthesia, 19 rabbits out of total 24 rabbits were laminectomized posteriorly and the spinal cords contused with 15 g 20 cm weight drop around 11th and 12th thoracic cord. After two weeks, behavioral analysis and clinical measurements of hindlimb spasticity were assessed. After that, rabbits' triceps surae was dissected and the length-tension plot was obtained by stretching it 5 mm, 2 mm/sec. And then the stiffenss of stretch reflex was determined from the length-tension plot. The stiffness of stretch reflex of the five normal control group was measured. RESULTS: 1) The measure was possible among 15 out of 19 spinal cord injured rabbits. One out of 19 rabbits was died from urinary infection. Two rabbits was died from intraoperative bleeding at 2 weeks. Another one was excluded from the measurement due to joint contracture. 2) All of 15 rabbits showed clinical spasticity of hindlimb in 2 weeks. The spasticity was increased after the 4th day. 3) On the length-tension plot, the stiffness of stretch reflex of 15 rabbits significantly increased more than that of 5 normal group. 4) The total stiffness is correlated with the deep tendon reflex of knee and muscle tone of ankle. The stiffness of stretch reflex is weakly correlated with the deep tendon reflex of knee, digital hyperreflexia and ankle clonus. CONCLUSION: The experimental spinal cord injury model and the quantitative spasticity measure method would contribute to measure the effect of the new therapy of spasticity more accurately, and to establish a principle of the new therapy. In addition, the study is expected to contribute to establish pathophysiology of spasticity.