I. A slope-ratio assay was developed with growing pigs to determine the availability of lysine in five protein concentrates. The basal diet contained 5.2 g lysine/kg and six levels of lysine, in 500 mg/kg increments, were used to determine the pig's response to standard lysine. The protein concentrates were incorporated into the basal diet to provide five levels of total lysine, again in 500 mg/kg increments, at the expense of wheat starch. A daily feeding scale based on live weight was used to ensure similar nutrient intakes. Pigs were fed at three-hourly intervals to ensure the utilization of free amino acids in the diets. Four pigs were allotted to each dose level and response was assessed over the 20-45 kg growth phase. 2. Potency estimates for available lysine in the five protein concentrates varied, depending on whether live-weight gain or carcass gain was used as the criterion of response. Carcass gain was considered more appropriate as it was not influenced by variation in gut fill. Availability of lysine in the five proteins, using carcass gain/d were (proportion of total) cottonseed meal 0.39, fish meal 0.89, meat-and-bone meal 0.50, skim-milk powder 0.88 and soya-bean meal 0.87. 3. Rat slope-ratio assay results for available lysine in the five protein concentrates were in general agreement with those from the pigs. In contrast, thedifferences in available lysinewere not detected by thechemical Silcock available-lysine test (Roach et al. 1967) nor by the direct fluorodinitrobenzene procedure (Carpenter, 1960). © 1979, The Nutrition Society. All rights reserved.