The aim of the present paper is to review recent information on food consumption rates of individual fish and to explore the ways in which values for individual food consumption can be used in studies of fish behaviour, nutrition and physiology. There are two main ways of carrying out nutritional studies in fish in which the aim is to investigate how the amount or the composition of the diet influences growth rate. One involves feeding tanks of animals and measuring the growth rates of the groups. This method stresses the importance of the group response to dietary manipulation and the ground rules for carrying out such studies have recently been clearly expounded (Cho, 1992;Cowey, 1992). An alternative method is to measure the food consumption of the individual fish and to construct from the data on individual animals food consumptiongrowth rate relationships for the species. In some species of fish which can be held and fed individually, e.g. cod (Gadus morhua L.; Houlihan et al. 1989) or minnow (Phoxinus phoxinus L.; Cui & Wootton, 1989), there is not a problem in determining food consumption and growth rate relationships. However, in fish feeding in groups a major problem has been to develop a reliable method to make repeated measurements of an individual's food consumption. Early attempts involved direct observations of feeding activity or the examination of gut contents in order to estimate consumption. These techniques have proved unsatisfactory, as the methodologies involved are timeconsuming, stressful or invasive and periods of pre-or postprandial starvation were necessary (for review, see Talbot, 1985). In the 1980s, two non-invasive methods were developed to measure consumption rates of individual fish, held in groups, which employed either feed labelled with the radioisotope I3lI (Storebakken et al. 1981) or with an X-ray opaque particulate marker (Talbot & Higgins, 1983). These techniques permitted repeated measurements of food consumption rates of fish held in groups without any alteration to the feeding regimen. However, for health and safety reasons X-radiography has been the preferred technique (for review, see Talbot, 1985).
T H E USE OF RADIOGRAPHY T O MEASURE INDIVIDUAL CONSUMPTION RATES OF FISHA number of studies have reported on food consumption rates of individual fish in groups using X-radiography (Jobling et al.