Biological and chemical methods to predict the level of plant-available N in animal manure were investigated under laboratory and growth-chamber conditions. Two biological methods (maize cultivation in pots and incubation of soil-waste mixtures) and four chemical methods (N extraction by autoclaving, 0.5 N KMnO4, pepsin, and 6 N HCI) were compared for their accuracy in determining the availability of N in 10 samples of animal manure applied to soil. The autoclaving, permanganate, and pepsin methods were able to predict N availability in this group of wastes. Total N mineralized in a soil amended with different samples of animal manure ranged from 0 to 31 1 mg N kg-1 soil. Expressed as a percentage of organic N added to the soil, mineralized N range from 0 to 39%. The poultry manure samples gave higher mineralization rates than the other types of manure tested. In general, mineralized N became immobilized during the incubation process, except for the second poultry manure sample, which showed an initially rapid then a a slow release of mineral N, and pig manure sample 2, which showed a slow initial rate, followed by a rapid increase, and then a slow rate of N release. The first-order exponential model used was able to describe the pattern of N mineralization in pig manure sample 2, poultry manure sample 2 but not the other samples.