The oropharyngeal feeding mechanism of the carp was analysed as a case study for cyprinids. Light and X‐ray cinematography combined with electromyography allowed a detailed analysis of the external and internal events during processing of the following food types: radtopaque pellets, earthworms, barley, tubificids, cladoceran suspensions and food‐soil mixtures. Ten patterns of head movements serve 12 feeding actions: paniculate feeding and gulping for intake; rinsing, repositioning, selective retention and spitting for selection; gathering from the branchial sieve, transport, loading of the teeth, crushing, grinding and deglutition. Muscular cushions in the pharyngeal roof (palatal organ) and floor (postlingual organ) permit postcapture selection between food and non‐food and transport. Protrusion of the upper jaw is crucial in food processing and serves different aims in particulate intake, gulping and internal selection. The mechanism of each single pattern and its effects in manipulating the flow and particles is discussed. Restrictions for processing different types of food are formulated. Tentative limits are set to the feeding on the available food types in the environment. The feeding apparatus appears to be unsuitable for exploiting very small particles (< 250 μm), plant and other materials of fibrous content. Only slow and immobile food particles with a diameter up to c. 4% of the carp's body length are effectively processed. The carp appears to be a generalist in its diet, with specializations for the exploitation of food and non‐food mixtures from the bottom, even if the contained food is of considerable density and hardness. The distinct elements of feeding behaviour are considered to be stereotyped action patterns. They are released and steered according to the actual size, distribution, consistency and contamination of the food and integrated into varied probing and feeding sequences. Different food types require different movement patterns and ‘handling times’. Protrusion with closed mouth appears to be a core pattern in food handling as it is basic to several feeding actions (repositioning and back‐washing during purification; gathering of retained food from the branchial sieve). Protrusion and the palatal and postlingual organs in this lower teleost are basic to the substrate feeding habits of many cyprinoids and are discussed in relation to (i) the hypertrophy of the pharyngeal masticatory apparatus, (ii) the recruitment of body power for mastication, and (iii) the evolutionary loss of toothed upper pharyngeal transporting bones. A scheme connects the unique character set of cypriniform fish to the origin and evolution of their feeding mechanism. The cooperation between functional morphology, ethology and ecology for the study of niche separation between species is emphasized.

A new model for filter feeding in bream (Abramis brama, Cyprinidae) is presented based on the three dimensional architecture of the branchial sieve. Transverse ridges on the upper surface of the gill arches form a system of channels in which food particles appear to be retained. These ridges are formed by a fleshy interconnection between the middle part of the gill arch and the bony parts of its gill rakers. Muscles attached to the rakers, present only on the lateral edge of the gill arch, indicate movability of the lateral bony raker element. If the fish is foraging on particles smaller than the channel diameter, movement of these gill rakers probably adjusts the sieve by reducing the channel diameter of the opposite channel. Selectivity of bream depends on available size classes of zooplanktons and changes in selectivity are attributed to adjustment of the branchial sieve. The channel model has been tested with feeding experiments and X-ray cinematography. The reconstructed paths of marked food particles show that particles follow the hypothesized path. Particle retention occurred mainly at the expected medial site of the arches. Our study strongly supports the channel model of particle retention.

During development, form and function (behaviour) change while the match between them must be maintained. The quality of this match determines the importance of morphological parameters in constraining behaviour. If the match is close, the morphology of organisms will be more constraining to the behaviour than when there is a large reserve capacity that creates a certain flexibility. This leads to two questions: (1) How good is the match between form and function during development? The quality of the match necessarily changes during development because changes in structural capacity often cannot proceed at the same speed as changes in functional demand. The evidence for these changes is discussed. (2) What are the mechanisms that maintain the match between form and function during developmental and evolutionary changes? Two mechanisms for maintaining the match are discussed: (a) reserve capacity and (b) flexible muscle activity patterns. Special emphasis is given to fish examples throughout this review.

The technique of X-ray cinematography was used to study pharyngeal movements in Abramis brama (L.). The theoretical and practical problems in X-ray cinematography of feeding fish are discussed, as well as criteria for the selection of images suited for detailed measurements.Respiration and filter-feeding on Daphnia pulex (length c. 1 mm) show different gill arch movement patterns in bream. Slits between gill-arches are kept smaller during filter-feeding. In addition, during filter-feeding, this inter-arch distance decreases considerably in a posterior direction. The hypothesis that particle retention occurs on the slits formed between adjacent gillarches and their gill-rakers is not supported by the present results.