Effects of gradual salinity changes on the gill NKA activities, plasma, gill, and kidney osmolarity and weight losing rate from starvation of Trachinotus ovatus juvenile reared in the deep water net-cage were studied in order to understand the mechanism of osmoregulatory adaptation in this fish. Experimental fish with the mean total length of (8.058±0.843) cm and body mass of (9.010±1.537) g were reared in the glass aquariums (80 cm×40 cm×50 cm) with five salinity treatments in triplicate, including 5, 15, 25, 30 and 35 ‰. The water temperature and dissolved oxygen were maintained at (30±3) ℃ and (6.5±0.3) mg/L throughout the experiment, and the pH varied from 8.0 to 8.8 as the changes of salinity. On 12 days after transferring to the target salinities, the survival rates of the juveniles in treatments of salinity 5, 25 and 35 were 87.14%, 95.71% and 88.86%, respectively, and no mortalities were recorded in the treatment of salinity 15 and 30. Gill NKA activities of fish from the salinity of 5, 25, 30 and 35 decreased at the beginning of the trial and then increased and decreased afterwards, and maintained at stable levels. After accumulated in the target salinities two days, the gill NKA activities followed a "U"-shape developmental trends with salinity change at the same time. NKA activities were found stable in both control (salinity 30) and the salinity of 25 treatments, indicating a stronger adaptive capacity of fish to lower salinity than that to the higher salinity. Plasma osmolarity was sensible to salinity changes, increased at the beginning of the trial and then decreased and increased afterwards and then decreased, and maintained at stable levels at the same salinity. Significant changes in plasma osmolarity were found in all the salinity treatments at the first two days of the experiment (P < 0.05) varying within wide limits in high salinity treatment than that in low ones. The variations of the plasma osmolarity were significantly higher in the salinity treatments of 30 and 35 (P < 0.05). Afterwards, the relationship between plasma osmolarity and the salinity were positively correlated in all the treatments at the same time. Osmoregulation in gills was sensitive to lower salinity changes, and the variations were found bigger in the salinity of 5 and 15, but relative stable in the salinity of 25, 30 and 35. This may suggest that gill play an important role in osmoregulation of fish in the lower salinity environment than that in higher salinity. Osmolarity in gill was gradually stabilized after one-day accumulation in the target salinity, and showed "U" - shape development trends with the changing of salinity, and no significant differences were found in all the treatments (P > 0.05). Kidney was more sensitive to salinity changing than gill in terms of osmoregulation. Osmolarity in the kidney was stable after two day's change of salinity, and was positively correlated to the changing of salinity at the same time. Osmoregulation was coordinated by both gill and kidney in lower salinities (< 30‰), and kidney played a leading role in the regulation at higher salinities (> 30‰). Salinity changes significantly affect the weight losing rate in the starved T. ovatus. Minimum weight losing rate was recorded at the salinity of 25, and maximum weight losing rate was found in the 5 and 35 salinity treatments. In conclusion, the salinity osmoregulatory ability of juvenile T. ovatus was relatively strong, and can be adapted in the salinity ranging from 5 to 35. Stabilization can be established within 1-2 days after salinity changed, and low salinity waters are preferable in juvenile T. ovatus.