Predator feeding strategy and prey importance: a new graphical analysis.

Diet may be analysed by numerical, volumetric and gravimetric methods (Hynes, 1950; Windell & Bowen, 1978; Hyslop, 1980). These analyses may include the frequency of occurrence and relative quantity (abundance, volume or weight) of a prey in the diet. The merits of each analysis depend on the comparability of the prey as regards food value per item, and ease of capture and handling. The relative importance and dominance of the prey is generally left to personal judgement (Hyslop, 1980). Results of the analyses are usually given in tabular form. As a graphical representation of data is generally easier to interpret, Mohan & Sankaran (1988) developed two compound indices which could be presented graphically. However, the graphical method presented here uses more straightforward descriptors of diet, namely the frequency of occurrence and relative abundance of the prey in the diet (Fig. 1). In addition, it depicts the relative importance of the prey (dominant or rare) and the degree of homogeneity of prey selection in the predator population. Percent occurrence is defined as the number ofpredators in which a prey taxon occurs as a percentage of the number of predators with prey. Percent abundance is defined as the percentage a prey taxon comprises of all prey items in all predators. If prey volume or weight is considered more appropriate for the species under study, it may be substituted for abundance. Each point on the graph represents the percent occurrence and abundance for a prey taxon (Fig. 1). Prey points positioned close to 100% occurrence and 100% abundance are considered the dominant prey taxa. A cluster of prey points positioned close to 100% occurrence and 1% abundance indicate the predator took many different prey taxa in low abundance, i.e. a generalized diet. Conversely, points close to 1% occurrence and 100% abundance indicate a specialization on certain taxa by some predators. Diagonals can thus be drawn on the plot to represent prey importance and predator feeding strategy. If the points are spread along and below the diagonal originating at the origin (labelled ' prey importance '), this suggests that the feeding was homogeneous amongst the predators, as would seem likely to occur in a shoal of fish. Conversely, if the points are spread along and below the' feeding strategy ' axis (Fig. 1) it suggests that feeding was more heterogeneous. When feeding is heterogeneous, different prey taxa may be important because of their great abundance in a few predators or low abundance but high frequency of occurrence in many predators. A hypothetical example is presented in Fig. 2. The predator population A shows a heterogeneous diet, a few predators having concentrated on prey taxa 1 and 2, while prey taxon 8 was eaten relatively rarely but by most predators. The causes of the heterogeneity may be apparent if the size or age distribution of the predators, and temporal and spatial aspects of the sampling, are included in further analyses. In contrast, predator population