Juvenile red seabream (mean body weight 29.0 g) were reared in recirculating culture systems with three different biofilter media, sand (SF), polystyrene microbeads (PF), and Kaldnes beads (KF). The efficiencies of the three different biofilter media were also tested. The SF was fluidized, and the PF and KF were trickled. All treatments were duplicated. The volumetric removal rates of total ammonia nitrogen by SF, PF, and KF were 193.8, 183.9, and 142.6 g $m^{-3}day^{-1}$, respectively, and those of nitrite nitrogen ($NO_2$-N) were 113.4, 105.9, and 85.8 g $m^{-3}day^{-1}$, respectively. The TAN and $NO_2$-N removal rates of KF were lower than those of SF and PF (P < 0.05), but there was no significant difference in these rates between SF and PF (P > 0.05). Among the biofilters used, only KF showed total suspended solid (TSS) removal capacity. The TSS removal efficiencies of SF and PF were negative. The growth rates of fish in SF were significantly higher than those in KF but not higher than those in PF. There was no difference in growth rate between fish in PF and KF. The specific growth rate and feed conversion efficiency of red seabreams in KF were lower than those in SF and PF, but there were no significant differences between SF and PF. These results indicate that sand and polystyrene microbeads are recommended for red seabream culture in a recirculating system.