The purpose of this study was to determine the effect of regular exercise during dexamethasone injection on the body weight, weight of hindlimb muslces, myofibrillar protein content and glutamine synthetase activity. 180-200g female Wistar rats were divided into four groups : control, exercise, dexamethasone injection (dexa), and exercise during dexamethasone injection(D+E) group. The dexa group received daily subcutaneous injection of dexamethasone at a dose of 4mg/kg body weight for 7 days. The exercise group ran on a treadmill for 60min/day(20minutes every 4 hours) at 10m/min and a 10$^{circ}$grade. The control group received daily subcutaneous injection of normal saline at a dose of 4mg/kg body weight for 7 days. The D+E group ran on a treadmill for 60min/day(20minutes every 4 hours) at 10m/min and a 10$^{circ}$ grade during dexamethasone injection. Body weight of the control group increased significantly from days of experiment, that of the dexa group decreased significantly from day 4 of the experiment resulting in a 82.4% decrease compared to the first day of the experiment. Body weight of the D+E group decreased significantly from day 5 of experiment resulting in a 81.77% decrease comprared to the first day of the experiment. Body weights, muscle weight and myofibrillar protein content of the plantaris and gastrocnemius decreased significantly and muscle weight of the soleus tended to decrease with dexamethasone injection. Glutamine synthetase activity of the hindlimb muscles increased significantly with the dexamethasone injection. The relative weight of the soleus was comparable to the control group and that of plantaris decreased significantly and that of gastrocnemius tended to decrease compared to that of the control in the dexa group. Body weight and muscle weight of the plantaris and gastrocnemius of the excrcise group were comparable to the control group, and the muscle weight of soleus showed a tendencey to increase. The relative weight of the soleus increased significantly and that of the plantaris and gastrocnemius were comparable to the control in the exercise group. Myofibrillar protein content of the soleus and plantaris increased significantly and there was no change of GS activity of the hindlimb muscles compared to the control in the exercise group. Body weight of the D+E group was comparable to the dexa group, muscle weight of the plantaris increased significantly and that of the soleus and gastrocnemius showed a tendency to increase. The relative weight of the hindlimb muscles increased significantly. Myofibrillar protein content of the soleus and plantaris increased significantly and that of the gastrocnemius tended to increase compared to the dexa group. Body weight and muscle weight of the plantaris and gastrocnemius of the D+E group did not recover to that of the control group. Muscle weight of the soleus recovered to that of the control group. The relative weight and of myofibrillar protein content of the hindlimb muscles recovered to that of the control group. From these results, it is suggested that regular exercise during dexamethasone injection might attenuate the muscle atrophy of the hindlimb muscles.