The determination of total nitrogen in soy sauce by the Kjeldahl method using traditional digestion and distillation, or using block digestion and steam distillation, has been studied as a candidate to be proposed to the Codex Alimentarius Commission. In order to obtain the optimum digestive condition, the added amounts of H2SO4, K2SO4 and CuSO(4)center dot 5H(2)O catalysts, the way of adding the catalyst, and the boiling period after the liquid became clear were examined. The optimum conditions for the digestion of 1 ml soy sauce were found to be adding 10 ml of H2SO4, 10 g of K2SO4, and 1 ml of 20% CuSO4 center dot 5H(2)O, and boiling for 80 min after a clear liquid is obtained in traditional digestion. Eight grams of K2SO4 were found to be appropriate in block digestion. Samples were measured by weight, and the total nitrogen was calculated on a % (w/w) basis to avoid the effect of the soy sauce viscosity. The nitrogen content was converted into % (w/v) by using the density of each sample. These sampling and digestion methods were compared with the JAS grading method for three kinds of soy sauce, lysine and (NH4)(2)SO4 solutions. The results yielded a significantly higher nitrogen content in full-bodied so), sauce; also, the recoveries (98 to 100%) for lysine and (NH4)2SO4 solutions, had lower RSD values than did that of the JAS grading method, and were regarded as being more accurate.