The objective of the present work was to compare colorimetric and distillationÈtitration methods to determine the ammonia-nitrogen con-(NH 3
-N) centration in sheep and goat urine samples. Colorimetric methods used were based on the indophenol reaction, whereas titrimetric methods were based on the alkali distillation of ammonia from the urine and its titrimetric determination with a standard acid. Colorimetric methods were only reliable when urine samples were diluted at least 1 : 20. Both colorimetric and alkali-distillation methods gave quantitative recoveries with standard solutions, but when NH 3 -N was determined in urine samples there was a signiÐcant discrepancy NH 3 -N between analytical methods on the measured concentrations. These were between 1É3 and 10-fold (on average 2É6-fold) higher with alkali distillation than with the colorimetric method. The di †erence between concentrations measured by both analytical methods was signiÐcantly related (R2 \ 0É990 ; P \ 0É001) to the concentration of urea in the urine samples. To study the e †ect of urea concentration, standard solutions containing variable concentrations of urea and were prepared, and concentration determined by both methods. NH 3 -N NH 3 -N Concentrations measured by colorimetry were similar to the actual concentrations irrespective of the urea concentration of the solutions. In urea-free solutions, alkali-distillation methods gave values similar to the actual concentrations, but when urea was present observed values were consistently higher than the expected concentrations. The overestimation increased with the urea concentration of the solutions, with a signiÐcant (P \ 0É001) relationship between both variables. Colorimetric methods were more reliable to measure concen-NH 3 -N trations in urine samples containing high urea concentrations.