We develop a food-fish model (FFM), which quantitatively relates properties of aquatic food types (size, shape, escape velocity, habitat, mechanical properties and chemical quality) to feeding structures of cyprinid fish. The model is based on functional morphology and experiments on search, capture, selection, and internal processing of food by fish. The FFM shows which food properties are most critical in feeding and how fish can optimise coping with them. Relative food size imposes the highest demands, followed by prey velocity, food habitat and mechanical properties. These overrule taxonomic affinities of food types. Highly demanding food types (large, fast prey, suspensions of plankton, benthic prey and mechanically tough items) impose incompatible morphological requirements on fish. We apply the FFM to the endemic Barbus species flock of Lake Tana (Ethiopia), since the structural diversity of its 14 species reflects recent adaptations to trophic niches. We predict their potentials in utilising different food types by quantitative comparisons of 35 parameters, measured for each species, with the values for each food specialist derived from the FFM. These diet predictions are tested against gut contents from 4,711 fish, sampled over seasons and habitats. Gut contents and predictions show a good overall fit. The value of the model is shown by its resolution in predicting resource partitioning among the barbs. For the 14 barbs a trophic hierarchy with six major trophic groups is reconstructed which closely matches the predictions. Trophic specialists (> 65␋y volume of a single food type) are also structurally specialised, whereas less extreme anatomical structures characterise trophic generalists, allowing them to switch between feeding modes. Trophic generalists are best defined by behavioral flexibility, since feeding modes integrate both fish and food characters. The FFM is of practical use in evaluating the role of morphological diversity in an ecosystem and enables the analysis of trophic interactions in fish communities and of the cascading effects by environmental change. Such an approach can be instrumental in the development of management strategies for fisheries and in conservation of biodiversity.