A study was conducted to evaluate the ability of juvenile turbot Scophthalmus maximus to utilize different molecular complexity of carbohydrates in diets. The control diet without carbohydrates supplementation contained 40% protein and 18% lipid. Based on it, 15% of three kinds of carbohydrate (glucose, sucrose and dextrin) were respectively added into the diet. At the same time, dietary lipid content was adjusted to 12% to keep isoenergetic with the control diet. Triplicate groups of fish (initial body weight: 8.12 ? 0.04 g) were hand-fed one of the four diets to apparent satiation for 9 weeks. The results showed that survival rate, daily feed intake (DFI), the apparent digestibility coefficients (ADCs) of protein and lipid, contents of moisture, crude protein, ash and glycogen in muscle, contents of moisture, crude protein and ash in liver, condition factor (CF) and plasma glucose were not significantly affected by dietary treatments (P > 0.05). Weight gain rate (WGR) and specific growth rate (SGR) of fish fed dietary dextrin and the control diet were significantly higher than those of fish fed dietary glucose and sucrose (P < 0.05). Feed efficiency (FE) of fish fed the control diet and dietary dextrin was significantly higher than those fed dietary sucrose (P < 0.05). However, there was no significant difference in FE between fish fed dietary glucose and the other three groups (P > 0.05). The order of the ADC of carbohydrate was glucose > dextrin > sucrose. The ADC of energy was significantly highest in fish fed dietary glucose and lowest in fish fed dietary sucrose (P < 0.05). Fish fed the control diet had the significantly highest lipid contents in muscle (P < 0.05), and there was no significant difference in lipid content of muscle among the other three groups (P > 0.05). The order of the liver lipid content was control > dextrin > glucose > sucrose (P < 0.05). The order of the liver glycogen was glucose > sucrose > dextrin > control (P < 0.05). Total amino acid concentration in plasma of fish fed the control diet was significantly higher than those fed dietary glucose and sucrose (P < 0.05). Plasma insulin of fish fed dietary dextrin was significantly higher than those fed the control diet (P < 0.05). Total cholesterol (CHO) in plasma of fish fed the control diet was the highest one followed by dietary dextrin. The lowest was found in fish fed dietary glucose and sucrose (P < 0.05). Triacylglycerols (TAGs) in plasma of fish fed the control and dietary dextrin were significantly higher than those fed dietary glucose and sucrose (P < 0.05). In conclusion, the present study suggested that turbot can utilize dextrin more efficiently than glucose and sucrose.