A chick bioassay procedure' has been used to determine the relative potency of several animal protein concentrates based on live-weight gains, food conversion efficiency (FCE) or nitrogen retention. In general, the assays were statistically valid and the majority had 95 % fiducial limits as narrow as 88-1 14%. Expressing response as FCE or nitrogen retention did not give better precision than using weight gain alone. The results showed that there are wide differences in available methionine and SAA in fish meals marketed under same the name.