Fear is now widely regarded as an undesirable state of suffering by many members of the public, the scientific community, welfare and policy groups, as well as by a growing number of farmers. At first glance this view appears to be inconsistent with one based on biological significance, in which fear is regarded as an adaptive state with fear behaviour functioning to protect the animal from injury. Fear is often adaptive in ideal circumstances but neither we nor the animals in our care live in an ideal world. In reality, many farming systems prevent the animals from responding in an adaptive fashion to potentially threatening stimuli. In these circumstances, fear can be a powerful and potentially damaging stressor, particularly if it is intense or persistent. Both acute and chronic fear can seriously harm the welfare and performance of poultry. Certain fear responses, such as violent escape or panic, are inap-propriate in intensive systems and can cause injury, pain or even the death of birds. High levels of underlying fearfulness have also been negatively associated with plumage condition, egg production, egg shell quality, growth, and food conversion efficiency. Furthermore, because fear inhibits all other motivational systems, its elicitation is likely to impair the ability of birds to adapt to environmental change, to interact successfully with each other or with the stock person, and to utilize new resources. Therefore, high fear is clearly undesirable from the bird's and the farmer's viewpoint and its reduction is of major importance. This objective had been hampered because of the complex-ity of the concept of fear and the controversy over its measurement. However, operational definitions of fear and its components are given in this review and a number of tests which have been used to measure fear in poultry are described. Strong intra-individual correlations found between the fear scores afforded in many of these tests suggested that they are measuring the same intervening variable, presumably underlying fearfulness, rather than purely stimulus-specific responses. Two of the commonest and potentially most frightening events encountered by poultry are sudden changes in their physical or social environment and exposure to humans. Other more specific types of alarming stimuli are also listed. Hitherto, the most promising ways of reducing underlying fearfulness and increasing adaptability, at least in the laboratory, involve one or more of the following approaches: environmental enrichment, regular handling or related treatments, vitamin C supplementation of the diet or drinking water, and genetic selection. The potential benefits and short-comings of these remedial measures are discussed, with attention to their practical relevance as well as to their more fundamental implications. Other measures to mask and/or minimize the incidence of frightening events are also briefly mentioned. Our objective should be not only to achieve a balance between maintaining an emotional state that facilitates the avoidance of danger while reducing the likelihood of overreaction and the expression of inap-propriate responses, but also to provide a stimulating, safe but economically viable environment. Continued investigation should enable recommendations to be made concerning the most effective programme, in terms of environmental modification, human-animal interaction, nutritive manipulation, and selective breeding, in order to achieve the optimal levels of fear in poultry, whatever the future trends in the industry.