A fish meal and a soybean meal sample were bioassayed for available
amino acid content according to five different published techniques. The techniques consisted of: 1) a two week chick growth assay (GA) for available
lysine; 2) a digestibility assay performed by total fecal collection (TFC); 3) a digestibility assay based on the analysis of ileal contents
IA); 4) a procedure based on partial fecal collection and analysis of a dietary marker using gnotobiotic chicks (GF); and 5) procedure 4) using conventional chicks (CONV). The GA gave high
lysine availability estimates for both
proteins, but values for fish meal were significantly reduced when the level of dietary incorporation was increased. The TFC assay proved difficult in collecting adequate sample and, as a consequence, abnormally high availability values were obtained. Modification of this collection procedure resulted in values comparable to the IA which gave the lowest average values of all the assays (90.5 per cent for fish meal and 91.1 per cent for soybean meal). Mean availabilities calculated from GF and CONV assay were significantly higher than the modified TFC and IA techniques. Significant differences were obtained in several individual
amino acids when the GF and CONV assay were compared. A growth assay, using diets formulated on the results of these assays, showed that the available
amino acid profile obtained from the CONV assay resulted in significantly improved
weight gains and feed conversions as compared to a similar diet based on the chemical analysis of
amino acids alone. All diets utilizing available
amino acid data resulted in improved feed conversions.