Reproduction, relative abundance, and variability in North American arvicoline rodent populations

Abstract

The ecological and life history characteristics of North American arvicoline rodents vary greatly.  A general model suggests that changes in reproduction, as a response to changes in climatic harshness and habitat type, likely affect variation in relative abundance of arvicoline populations.  Previous work indicated that variability in abundance does not always increase with mean litter size or with latitude, but litter size does tend to increase as the length of the breeding season decreases.  I therefore propose the reproductive potential (RP) hypothesis which states that under favorable conditions, populations with higher reproductive potential can grow more rapidly and can reach higher densities during the breeding season, which leads to greater variability in abundance because very high populations eventually decline to low densities.  I define reproductive potential as the maximum number of offspring a typical female could produce during a year and calculate it as the product of mean litter size (m) and length of the breeding season in weeks (b) divided by 3 (RP = mb/3).  I then review the problems associated with estimation of the necessary parameters and indicate my criteria for accepting data.  To test the RP hypothesis I correlate RP to a measure of variability in abundance for populations with at least 10 years of data, and I compare populations of the same or different arvicoline species at the same and different sites.  Most results did not support the RP hypothesis.  However, three species had different litter sizes in habitats with different vegetation, and all three reached higher maximum densities where reproductive potential was greater.