Predation By the Three-Spined Stickleback (Gasterosteus Aculeatus L.): the Influence of Hunger and Experience

Abstract

[Predatory behaviour in a small fish, the three-spined stickleback (Gasterosteus aculeatus L.), has been analyzed quantitatively by continuous observation of individual fishes for food (generally one Tubifex worm) in a large maze during experiments of 40 min. Properties of environment and prey (e.g. their number and size) were kept as constant as possible, in order to study the effects on prey risk of changes in the internal state of the predator, viz. its hunger and its experience with the maze and the prey type in question. By variation of the length of the foregoing deprivation time under specified conditions of temperature, time of day, feeding-deprivation schedule, and kind of food eaten before — the sticklebacks were brought into well defined hunger states. A number of elements of their feeding behaviour were found to increase and decrease with time of deprivation and feeding, respectively. These parallel fluctuations of different and independently measured elements of the behaviour argue a common internal mechanism, the state of which may be called the hunger of the stickleback. Hunger, then, may be measured by anyone of these behaviour elements fluctuating in parallel. Hunger of the stickleback severely affected the risk of their prey. Risk is defined as the number of prey destroyed per unit time by one predator per unit prey density. Higher hungers caused higher risks by both a higher swimming activity and higher proportions of complete responses to discovered prey (viz. greater proportions of the discovered prey were grasped and greater proportions of the grasped prey were eaten). Encountering efficiency (i.e. the ratio of the number of prey encountered to the distance swum or to the number of cells of the maze visited) proved to be independent of hunger. It increased in all individual fishes used (though to a different extent) with their growing experience with the maze by their increasing ability to explore it more systematically, which ability was achieved by turning less. The chance of an encountered prey to be discovered proved to be affected primarily by the distance to which the fish approached it. This chance at a given distance was dependent on the visual acuity of the stickleback in question. It was dependent on the hunger of the fish only at very low levels of hunger, not normally reached during 40 min.-experiments at unit prey density. Prey risk was higher according as the fish was larger. Nearly all sticklebacks were able to get satiated towards the end of the 40 minutes, having ingested then about 5% of their body weight. As their swimming activity increased much less with their weight than their feeding capacity, the larger fishes got satiated at a slower rate than the smaller ones. The dependence of prey risk on predator size could be explained by these differences in hunger decay. At higher than unit prey density the fishes did not eat more. Prey risk was then inversely proportional to prey density. The fishes became satiated at a faster rate at the higher prey density. These differences in hunger could largely explain the observed relationship between prey density and prey risk. Introduction of a new type of prey caused — after some delay — a rapid increase of its risk. This searching image occurred only if the fish in question judged the new prey sufficiently palatable. Palatability was measured as the proportion of grasped prey eaten at specified hunger states. During image formation only one component of the prey risk, viz. the chance of an encountered prey to be discovered, was consistently affected. Responsiveness to inedible objects, which looked like the new prey, was conspicuously raised at the same time. Sticklebacks are polyphagic. Once a new type of prey was well known, its presence negatively affected the risk of alternative prey. Larvae of Drosophila, which some fishes judged nearly as palatable as Tubifex worms, affected the latters risk like a mere increase of the density (expressed in weight units) of Tubifex themselves would do. Addition of the highly palatable pieces of Enchytraeus had more drastic consequences : all prey were encountered at a higher rate by an increase of swimming activity. Nevertheless, the risk of Tubifex decreased dramatically by a selective inhibition of the responses to encountered prey of this type. Some evidence points to a learning process, involving the refusal of Tubifex worms at high hungers in the expectation of the more palatable pieces of Enchytraeus. The characteristics of the stickleback's predatory behaviour have been compared with premises of various mathematical models on predator-prey interactions. It is concluded that all existing models should be modified on essential points in order to represent the characteristic traits of the behaviour of predators like the stickleback., Predatory behaviour in a small fish, the three-spined stickleback (Gasterosteus aculeatus L.), has been analyzed quantitatively by continuous observati