Limited information has been available about the influence of loading density on the performances of Scophthalmus maximus, especially in recirculating aquaculture systems (RAS). In this study, turbot (13.84±2.74g; average weight±SD) were reared at four different initial densities (low 0.66, medium 1.26, sub-high 2.56, high 4.00 kg/m2 ) for 10 weeks in RAS at 23±1℃. Final densities were 4.67, 7.25, 14.16, and 17.47 kg/m2 , respectively, which translate to 82, 108, 214, and 282 percent coverage of the tank bottom. Density had both negative and independent impacts on growth. The final mean weight, specific growth rate (SGR), and voluntary feed intake significantly decreased and the coefficient of variation (CV) of final body weight increased with increase in stocking density. The medium and sub-high density groups did not differ significantly in SGR, mean weight, CV, food conversion rate (FCR), feed intake, blood parameters, and digestive enzymes. The protease activities of the digestive tract at pH 7, 8.5, 9, and 10 were significantly higher for the highest density group, but tended to be lower (not significantly) at pH 4 and 8.5 for the lowest density group. The intensity of protease activity was inversely related to feed intake at the different densities. Catalase activity was higher (but not significantly) at the highest density, perhaps because high density started to induce an oxidative effect in turbot. In conclusion, turbot can be cultured in RAS at a density of less than 17.47 kg/m2 . With good water quality and no feed limitation, initial density between 1.26 and 2.56 kg/m2 (final: 7.25 and 14.16 kg/m2 ) would not negatively affect the turbot cultured in RAS. For culture at higher density, multi-level feeding devices are suggested to ease feeding competition. Limited information has been available about the influence of loading density on the performances of Scophthalmus maximus, especially in recirculating aquaculture systems (RAS). In this study, turbot (13.84 ± 2.74g; average weight ± SD) were reared at four different initial densities (low 0.66, medium 1.26, sub-high 2.56, high 4.00 kg / m2) for 10 weeks in RAS at 23 ± 1 ° C. Final densities were 4.67, 7.25, 14.16, and 17.47 kg / m2, respectively, which translate to 82 Density had both negative and independent impacts on growth. The final mean weight, specific growth rate (SGR), and voluntary feed intake significantly decreased and the coefficient of variation (CV) of final body weight increased with increase in stocking density. The medium and sub-high density groups did not differ significantly in SGR, mean weight, CV, food conversion rate (FCR), feed intake, blood parameters, and digestive enzymes. The protease activities of the diges tive tract at pH 7, 8.5, 9, and 10 were significantly higher for the highest density group, but ten to toe lower (not significantly) at pH 4 and 8.5 for the lowest density group. The intensity of protease activity was inversely related to Catalase activity was higher (but not significantly) at the highest density, perhaps because high density started to induce an oxidative effect in turbot. In conclusion, turbot can be cultured in RAS at a density of less than 17.47 kg / m2. With good water quality and no feed limitation, initial density between 1.26 and 2.56 kg / m2 (final: 7.25 and 14.16 kg / m2) would not negatively affect the turbot cultured in RAS. For culture at higher density, level feeding devices are suggested to ease feeding competition.