Abstract: From June to August,thegrowth experiment was conducted in a recirculating rearing system on juvenileof the southerncatfish, Silurus meridionalis Chen with initial weight 12.93 ± 0.13g. Thewhite fishmeal, corn oil and gelatinized cornstarch wereused as protein, lipid and carbohydrate respectively. Fish were fed with six iso-nitrogenous (40% crude protein) and iso-lipidic(10%crude lipid) experimental diets contain 0 (control diet), 6, 12, 18, 24 and 30%of carbohydrate level. At each of the diets,4 groups of southern catfish (12 in each group) were fed at ad libitum ration level for8 weeks at 2715e. The digestibility of car-bohydrate was not significantly affected when dietary carbohydrate level did not exceed 18%, but decreased when dietary carbohy-drate level exceeded 18%. The feeding rate and specific growth rate (SGR) increased with dietary carbohydrate level at a level ofdietary carbohydrate not more than 18%, and decreased at a higher level. The feeding rate of 6% and 12% carbohydrate levelwere significantly higher than those of other levels, and the feeding rate of 30% carbohydrate level was significantly lower thanthose of other levels. The specificgrowth rate (SGR) of 6%and 12%carbohydrate level were significantly higher than those of 0,24 and 30%carbohydrate levels, and specificgrowth rate (SGR) of 30%carbohydrate level was significantly lower than thoseof 6,12,18 and 24% carbohydrate level. The results suggested the optimal dietary carbohydrate level for growth in this fish was12% ) 18%.