EN
The cyclicity of Arctic populations of small rodents is a subject with a long history and a large literature (Batzli,1992) in which the question ‘What drives the cycle?’ has received many answers, among them that the source of the cycle is either rodent interaction with food or the interaction with predators or both. Another question concerns the confinement of the cycle to Arctic conditions. The paper by Gårding (2000) presented a simple mathematical model of the combined predator-prey-food interaction based on a general eater-food interaction in which cycle length is an explicit decreasing function of the average birth rate of eaters. In the combined interaction, the cycle length is the same function of the sum of the average birth rates of predators and prey. Numerical fits of these models make it possible to answer the questions above. The results are that the short 3–5 year cycles of the Arctic rodents: lemming (Lemmus lemmus) and vole (Microtus agrestis) are mainly driven by interaction with food while the ten year cycle of the Canadian snowshoe hare (Lepus americanus), is driven by interaction with its predator – lynx. Rodents in the Arctic live and breed in burrows and experience predation pressure when surfacing. This explains their interaction with food. The greater variety and easier availability of food in a temperate climate accounts for a missing rodent interaction with food. The paper starts with a presentation of the eater-food interaction model itself, its simple but unfamiliar mathematics and its points of credibility. At the end of the paper some current hypotheses about the nature of the rodent cycle are seen in the light of the model used here.