The ileal digestibility of tryptophan for growing pigs was determined for cottonseed, meat-and-bone and soya-bean meals. Tryptophan in the food and digesta was measured by two analytical procedures (NaOH hydrolysis and colorimetric estimation (method 1) and LiOH hydrolysis and HPLC determination (method 2)). The results were respectively: cottonseed meal 0.46, 0.81; meat-and-bone meal 0.55, 0.65; soya-bean meal 0.74, 0.90. In the first experiment the values for method 1 were shown to be inapplicable to pigs. In a second experiment three tryptophan-deficient diets (005 g ileal digestible tryptophan/MJ digestible energy (DE)) were formulated using values from method 2 for cottonseed meal, meat-and-bone meal plus L-tryptophan and soya-bean meal respectively as the only sources of tryptophan in the diets. This experiment was terminated after 28 d as overall growth performance of the pigs was very low. A third experiment was conducted in a similar manner to Expt 2 except that the diets were formulated to 0.065g ileal digestible tryptophan/MJ DE and growth responses and tryptophan retention were assessed over the 20–45 kg growth phase. Growth rates (g.d) of the pigs given the three diets were significantly different (P < 0.01): cottonseed meal 393, meat-and-bone meal plus L-tryptophan 531, soya-bean meal 437 (SED 39.0). Tryptophan retention (as a proportion of ileal digestible tryptophan intake) was significantly different (P < 0.05): cottonseed meal 0.51, meat-and-bone meal plus L-tryptophan 0.49, soya-bean meal 0.41. These results indicate (1) that the colorimetric technique for assessing tryptophan was inapplicable and (2) that ileal digestible values for tryptophan were not suitable for formulating diets containing heat-processed proteins, possibly due to absorption of some of the tryptophan in a form that was non-utilizable, and/or to underestimation of total tryptophan in the protein concentrates.