Objectives. Evaluate the effect of the food ration on the accumulation of total ammonia nitrogen and the survival of juveniles of D. latifrons under experimental conditions. Materials and methods. A completely randomized experimental design was implemented, without 72 h water replacement, 10 fish per replicate and three replicates per treatment, with feed rations of 2, 4, 6 and 8 % in relation to biomass and 35 % protein balanced shrimp feed. Water quality parameters were recorded daily, and dead organisms were counted. Results. Total ammonia nitrogen concentrations ranged from 1.05 to 7.80 mg*L-1 and non-ionized ammonium ranged from 0.33 to 2.27 mg*L-1. The feed ration at 2 % showed values of total ammoniacal nitrogen (5.16 ± 1.70 mg*L-1) and non-ionized ammonium (1.56 ± 0.48 mg*L-1) significantly (p < 0.05) higher than the other treatments; however, it showed the highest survival (100 – 56 %). Survival was reduced in all treatments; rations of 6 and 8 % recorded mortality from 24 h. LD50 % at 72 h was estimated with a food ration of 5.54 % with respect to biomass. Conclusions. Dormitator latifrons resists concentrations of non-ionised ammonium above 1 mg*L-1. In intensive cultures of D. latifrons, a food ration less than 4 % of biomass should be used, with water replacements at least every 48 h to avoid mortalities.