A bioavailable nutrient has the potential to be utilized when ingested. Ideally, the nutrient requirements of animals and the nutrient concentrations of feedingstuffs should be expressed in terms of bioavailable units. Bioavailability is an abstract concept which cannot be measured but it can be estimated. The purpose of this review is to describe, evaluate and compare procedures used to estimate the concentrations of bioavailable amino acids (BAA) in feedingstuffs for pigs and poultry. Accurate and precise measurement of total amino acid (AA) concentrations is of fundamental importance to many assays. Problems are encountered both in making the AA available for measurement (hydrolysis) and in the measurement procedures. There is need for standardization of methodology but until this is achieved caution must be exercised when comparing data obtained by different techniques or even from different laboratories. In vitro assays for BAA are relatively fast and can be made without access to animals. Enzymic digestion has the potential to be a good estimator of BAA but is constrained by a lack of information about the bioavailabilities of the peptides which are liberated. Chemical assays and dye binding produce data which sometimes, but not always, correlate with biological estimates of BAA. The future of such assays seems to be in quality control where comparisons are made among samples within ingredients. Among the indirect biological assays, those based on microbial growth have received the most attention. The major weakness of such assays resides in the difference between animal and microbe in the ability to utilize peptides. Plasma free amino acids are not satisfactory estimators of BAA although they can help to identify limiting AA in feedingstuffs. Nitrogen digestion lacks the sensitivity to be a predictor of BAA. Growth assays are thought by many to be the ultimate measure of bioavailability but in fact they measure utility which is not the same. Nevertheless, growth assays can provide valuable estimates of BAA providing that their limitations are understood. The assays are expensive and slow and the results can be affected by many variables including changes in nutrient concentrations when a test material is added to an assay diet. Much of the review is devoted to balance experiments which require great care and attention to detail if reliable, reproducible data are to be obtained. Procedures have been refined during many years of use and with the advent of new equipment and techniques. Concern is currently focussed on corrections for metabolic and endogenous losses and on the effects of the microflora of the alimentary tract. Recent developments have increased assay efficiency, particularly with poultry. The balance experiment provides BAA estimates, that can be used effectively in diet formulation, at a much faster rate than growth assays. The final section of the review makes comparisons among various methods of estimating BAA. Interspecies comparisons are useful because they provide guidance when attempting to apply data derived with one species to formulate diets for another. The search for a single assay which provides data applicable to several species is a worthwhile objective. In the meantime, there are many assays used to estimate BAA which have specific applications and limitations. Key words: Pigs, poultry, amino acids, assay methodology, feedingstuffs, review
