Purpose: Previous studies have suggested that BDNF has a role in plasticity and survival following spinal cord injury and treadmill exercise increases BDNF levels in the normal brain and spinal cord. We attempted to determine whether swimming exercise improve motor function following experimental contusive spinal cord injury and whether motor outcome is associated with BDNF expression. Methods: Thirty six Sprague-Dawley rats (weight, 250 to 300 g) were divided into control (n=18) and experimental swimming group (n=18). Spinal cord injury was produced using NYU-spinal impactor at the eleven thoracic levels in both groups. Swimming exercise started $7^{th}$ day from SCI operation, lasted 5 min per day, 5 days a week for 4 weeks and then exercise times a day were increased in one number to each week. Motor functional recovery was determined by the Basso-Beattie-Bresnahan (BBB) locomotor rating scale, modified inclined board plane test, histological findings, H&E and BDNF expression observed at $1^{th}$, $3^{rd}$, $7^{th}$, $14^{th}$, $21^{st}$ and $28^{th}$day after injury. Results: 1. The BBB scores were higher in experimental group than control group at $14^{th}$, $21^{st}$ day (left hind limb) and at $21^{th}$ day (right hind limb) (p<0.05) after injury. 2. The inclined board plane test were significantly greater in experimental group than control group at $7^{th}$ day (p<0.05), $14^{th}$ and $28^{th}$ day (p<0.01) after injury. 3. The BDNF expression was severe revealed in experimental group than control group at $7^{th}$, $14^{th}$ and $28^{th}$ day after injury. Conclusion: This study suggests that swimming applied from the early phase after spinal cord injury be beneficial effects in motor functional recovery.