This thesis investigated the nutritional consequences of processing chicken feed, in particular the loss of available amino acids as a result of the Maillard reaction, both in model protein systems and during poultry feed processing. Model systems containing RNase A and pure or feed-type carbohydrates were incubated at 37°C, 50°C or 70°C for up to 8 days. From lysine analysis, reduction in reactive amino groups in RNase A was shown to occur rapidly at room temperature, with no incubation period necessary. However, protein fragmentation that occurred during incubation, especially at higher temperatures, interfered with the quantification of lysine. SDS-PAGE showed crosslinking reactions of RNase A with the tested carbohydrates to be slow below 70°C. Incubation of RNase A with cyclotene or xylose produced the greatest rate of crosslinking, while starch and sucrose produced the least. A modified OPA method was developed such that the level of reactive lysine could be quantitated in barley flour proteins, in a manner that was technically straightforward, inexpensive and allowed good through-put of samples. This method was shown to have good agreement with the published ninhydrin method in the measurement of Maillard reacted lysine in barley flours. Up to 25% loss of reactive lysine was observed in barley flour that had not undergone processing beyond milling. Samples were taken before, during and after the pelleting of chicken feed. Lysine analysis of these samples showed loss in amino group content of at least 18% could occur during processing, although this was dependent on variations between pelleting runs. Protein fragmentation during processing potentially masked further losses. A growth trial assessed a novel method of lysine addition to the feed, via spraying on free lysine solution post-pelleting. Over the 2 week trial period, 10% of the lysine applied by this method remained in the uneaten fines. Lysine eaten by chickens aged day 8-14, as calculated from measurements taken using the OPA method, correlated well with bird growth. Increased feed intake was also seen as the lysine content of the diet increased. In chickens aged day 15-21, lysine addition above the first addition level produced no further significant gain in performance. Therefore, lysine was not growth limiting above this level. Results indicated that over the 2 week period of the trial, equivalent performance for birds on the standard feed could have been achieved with a 50% reduction in free lysine added to the feed formulation. No lysine loss occurred in the standard sample as a result of pelleting. Therefore, while the potential exists for lysine loss to occur during chicken feed pelleting, this thesis has shown that this is not a significant problem. As up to 25% lysine blockage in unprocessed barley flour was observed, obtaining feed ingredients with consistently low levels of Maillard reaction damage may be as important as maintaining ideal processing conditions.