Captive-rearing of animals for reintroduction into the wild not only involves propagation of individuals, but should also be concerned with ensuring that animals destined for release have the skills necessary to thrive in the wild. In recent years greater attention as been dedicated to preparing captive-reared individuals for life post-release. Predation is a major cause of mortality of reintroduced animals, but techniques to prepare captive-reared animals to cope with predators are often perceived to be difficult to design and implement. Here I show results of attempts to condition the black stilt (Himantopus novaezealandiae), a critically endangered wading bird species, to be wary of cats. Annually, about 30 black stilts are raised in captivity for release into the wild. A major source of post-release mortality is thought to be mammalian predators. The ability of juvenile captive-reared black stilts to recognise and learn to be wary of the cat (Felis catus), was studied in two experiments conducted between November 1992 and August 1994. Cats are not historical enemies of black stilts as mammalian predators are a relatively new introduction to New Zealand (in the last 100 years). However, captive-reared juvenile black stilts responded cautiously to the cat model upon their first encounter, suggesting some genetic recognition of mammalian predators already existed. Antipredator-training by conditioning involved structured presentations of a moving cat model (conditioned stimulus) paired with alarm calls (unconditioned stimulus), and was conducted at various ages. In Experiment I, two thirds of the birds received antipredator-training and one third did not. Few differences were found between antipredator responses of the two groups, and post-release survival was-the same for trained and untrained birds. In Experiment II recognition of cat and control models by juveniles that had been anti-predator trained at different ages was investigated. Results indicated that captive-reared black stilt juveniles were capable of learning to be more wary of cats after training, although some decrement of predator recognition behaviour occured over time, possibly due to factors such as habituation and extinction of responding. I suggest that conditioning captive-reared animals about predators is a valuable addition to existing reintroduction programmes, and make suggestions for efficient introduction of antipredator-training into the current black stilt management programme.