Hydrodynamics of suction feeding in fish

Abstract

1. Suction feeding is the dominant way to obtain food in teleosts. Its high demands on structures due to the required velocities and forces leads to the expectation that a biophysical analysis will aid to reveal the adaptive significance of head and body structures related to this action. 2. A main factor in the development and perfection of mechanisms of prey suction is the pushing of water and prey by the approaching predator due to the high density of water. 3. To obtain the prey only the flow relative to the predator is important. This flow consists of components caused by suction, by forward motion due to suction and by swimming. Flow velocities are therefore calculated in the frame of the moving fish. They differ from flow measured in an earth-bound frame e.g. the aquarium. The distinction of these two systems of coordinates is an essential feature of the present approach. 4. A model is developed of flow in an expanding and compressing cylinder or cone, possessing an anterior opening (mouth aperture) and a posterior expanding region (opercular) with a moving branchiostegal- and opercular valve. The simplifications necessary in the fish to model transition are examined on their applicability. This is done for hydrodynamic (e.g. the role of friction) as well as for biological (e.g. influence of gills) abstractions. 5. The model features the interrelation between shape and movement of mouth aperture and opercular region, the moment of prey swallowing and valve opening, the influence of translation on the flow and the unsteady nature of the generate flow. The distinction of a point of zero flow in an earth-bound frame facilitated the construction of streamlines and is a point of reference for the calculation of pressures. 6. The analysis of actual fish head motion from high-speed movies and the use of these measurements in model simulations necessitated complex procedures of curve fitting. 7. The flow in front of the fish towards its mouth, inside its buccal cavity, behind the valves and around the rest of the fish is united in one comprehensive flow picture. Problems of boundary values which must be chosen to describe the frontal and caudal flow are discussed. 8. To capture the prey the optimization of the amount of sucked water and of the velocity of the water into the fish's mouth is requested, together with a directed suction flow reducing the amount of sucked water not containing the prey. The former is increased when continuing head expansion can be performed with closed valves thus maintaining an antero-posterior flow. This solution requires long and rotatable branchiostegal rays. The conditions for outflow through the opercular slits after valve opening are discussed. 9. The effect of leakage through the opercular valves and of inflow through notched mouth angles on the water velocities in the frontal mouth opening are presented. Morphological adaptations in these area's can be treated in their quantitative effect on flow velocity. 10. The internal form of the buccal cavity, the form of the promontory consisting of urohyal, sternohyoid muscle and pectoral girdle and the shape of the gills are explained as adaptations to a flow with only gradual changes in the direction of water accelerations. 11. A series of feeding types is distinguished based on the relative contributions of suction and forward motion with the time of valve opening as a third characteristic. Suction causes a flow to the mouth from all sides whereas swimming leads to a highly directed parallel flow. 12. Jaw protrusion greatly increases the flow velocity in the moving frame and has therefore similar effects as swimming to the prey. Due to protrusion the impulse added to the water can be kept at a low value. Protrusion besides increases the ecological versatility because it provides a highly directed suction flow in a habitat with e.g. many obstacles which prohibit fast swimming. 13. The accurate prediction of pressure changes in model studies is limited as pressures are proportional to the second derivatives of motion. Suction as well as forward motion contribute to pressure variations. Pressures and velocities calculated by simulating head length increase in the model demonstrate load increase during growth and suggest limits to which head structures are subjected. Combination of thus influenced factors suggest a restricted range of possible options. 14. Careful application of the above findings to existing schemes of the succession of types in the evolution of head construction in actinopterygian fishes strongly suggests that changes in jaw apparatus and opercular- and branchiostegal system improve the result of suction feeding and not only biting.