Many vole and lemming populations fluctuate cyclically reaching peak populations every 3-4 years. Since 1959 I have been trying to find out what drives these cycles. Dennis Chitty suggested in 1960 that cyclic changes in social behaviour and population genetics might hold the key to understanding why cycles occur in so many different species of voles in such a variety of habitats. My students and I have done a series of field experiments on the deer mouse (Peromyscus maniculatus), Townsend's vole (Microtus townsendii), and the creeping vole (Microtus oregoni) around Vancouver since 1971 to see what factors are important in causing populations to fluctuate. In 1973 we began work at Kluane Lake in the southwest Yukon to see if our findings could be extended to more northern populations. The dominant theme that has emerged from our work on rodents is that aggressive behaviour, particularly of breeding females, is a major cause of juvenile losses in rodents. We do not know yet if aggressive behaviour is genetically controlled or if it changes only in response to crowding. In 1976 I began studies on the ten-year cycle of snowshoe hares at Kluane Lake, and I continue research on snowshoe hares, voles and mice. In 1987 I began to look at lemming population dynamics on the arctic coast of the N.W.T. This work is now completed. In 1991 I began a collaboration with CSIRO Rodent Research Group in Canberra, Australia on the causes of house mouse outbreaks in the wheat growing areas of southeastern Australia. This work is now completed. In 1999 I joined the Swedish Tundra Northwest Expedition in the Canadian arctic and began to assemble data on community dynamics in arctic Canada. This work continued with the International Polar Year research that occurred in collaboration with Don Reid and Gilles Gauthier from 2007 to 2009. I retired from teaching at the Zoology Department of UBC on December 31, 2001. I continue to carry on research on population dynamics of snowshoe hares, mice and voles at Kluane Lake in the southwestern Yukon.