An 8-week feeding trial was conducted to evaluate compensatory growth of juvenile black sea bream, Acanthopagrus schlegelii (initial weight 9.56 ± 0.12 g) in fifteen 300 L indoors flow-through circular fibreglass tanks. Feeding regimes was designed as follows: the control group (fed continuously), and S1, S2, S3 and S4 groups experienced 1, 2, 3 and 4 days of feed deprivation and then refeeding for the remaining days per week respectively. Changes in body weight, specific growth rate (SGR), feed intake (FI) and feed efficiency ratio (FER) were examined biweekly during the trial. At the end of the cyclical feeding periods, survival was not significantly affected by feeding strategy (P > 0.05). Final body weight of fish in S1 group was significantly higher than the control group (P < 0.05) after an 8 weeks trial, and fish in the S2 group reached the same body weight of the control fish (P > 0.05), however, the growth data in the S3 and S4 groups could not catch-up with the control treatment (P < 0.05). In the first 4 weeks, SGR values of fish with feed deprivation more than 2 days per week were inferior to those of control group (P < 0.05); however, no significant differences of SGR were observed among the groups for the last 4 weeks (P > 0.05). FI increased significantly with the starvation days increasing during the whole feeding trial (P < 0.05). At the periods 2, 4 and 6 weeks, FER values increased significantly with increasing feed-deprivation days up to S2 group and then levelled off (P < 0.05); however, FER in S2 group was only higher than that in S0 and S1, but no significant differences were found among the other treatments at the end of week 8. Significant differences were found in apparent digestibility coefficients (ADCs) of dry matter and crude protein among the treatments, while ADCs of crude lipid were unaffected. Protein and lipid contents in whole body and dorsal muscle showed declining tendency with increasing food deprivation days, while moisture contents tended to increase (P < 0.05). Serum parameters were markedly affected by feeding regimes except for total protein concentration and thyroxine level. The present results indicated that starvation for 1 and 2 days per week of juvenile black sea bream could achieve over-compensation and complete compensation respectively. However, in case of longer term feed restriction regime, fish failed to obtain good growth performance.