A laboratory feeding trial was conducted for 80 days using rohu, Labeo rohita, fingerlings (5.67 ±0.25 g) to evaluate the optimum dietary carbohydrate level. Five semipurified diets containing 30, 35, 40, 45, and 50% carbohydrate (diets D1 to D5) were formulated and fed to triplicate groups of fish. All the diets were isonitrogenous (35% protein) and almost isocaloric. Fish performance in terms of average live weight gain (%), SGR (%/day) and FCR was best with 40% carbohydrate level. Significantly (P < 0.05) poor growth of fish also was noticed in the groups reared on diets with 30 and 35% levels of dietary carbohydrate, however, no significant difference in PER and ANPU (%) was obtained with diets D2 and D3. Although no significant difference in apparent protein digestibility was recorded up to 45% dietary carbohydrate level, lipid, ash, dry matter, and energy digestibilities were found to be significantly (P < 0.05) higher for the diets D2 and D3. These two dietary treatments also resulted in higher deposition of protein and lipid and lower moisture and ash contents in the fish muscle. Amylolytic and proteolytic enzyme activities also were estimated to be higher in the groups of fish fed 35% and 40% carbohydrate diets. The study indicated that a minimum level of 40% dietary carbohydrate is required (protein content being 35%) for optimum growth, feed conversion and nutrient utilisation in rohu, Labeo rohita, fingerlings.