An 8-week feeding experiment was conducted in a water flow-through system (26–28 °C) to determine the dietary threonine requirement of fingerling Labeo rohita (3.90±0.03 cm; 0.58±0.02 g). Growth, feed utilization and body composition of fish fed test diets (40% crude protein; 17.9 kJ g−1 gross energy) with graded levels of l-threonine (0.75%, 1.0%, 1.25%, 1.50%, 1.75% and 2.0% dry diet) to apparent satiation were response variables used to assess threonine adequacy. Diets were made isonitrogenous and isoenergetic by adjusting the levels of glycine and dextrin. The amino acid profiles of the test diets were formulated to that of 40% whole chicken egg protein except for threonine. The performance of fish fed experimental diets was evaluated using calculated values for weight gain (g fish−1), feed conversion ratio (FCR), protein efficiency ratio (PER) and protein productive value (PPV) data. Maximum weight gain (g fish−1) (1.79), lowest FCR (1.39), highest PER (1.76) and PPV (0.33) were recorded at 1.50 g per 100 g dietary threonine. Statistical analysis of weight gain, FCR, PER and PPV data reflected significant differences (P<0.05) among treatments. Except for reduced growth performance in fish fed threonine-deficient diets, no deficiency signs were noted. Weight gain, FCR, PER and PPV data were also analysed using second-degree polynomial regression analysis to obtain a more accurate threonine requirement estimate, which was found, using each response variable, to be at 1.70, 1.63, 1.65 and 1.51 g per 100 g of dry diet, corresponding to 4.2, 4.07, 4.12 and 3.77 g per 100 g of dietary protein respectively. Based on the second-degree polynomial regression analysis of the live weight gain, FCR, PER and PPV data, the optimum dietary level of threonine for fingerling L. rohita was found to be in the range of 1.51–1.70 g per 100 g of the dry diet, corresponding to 3.77–4.2 g per 100 g of dietary protein.