The population ecology of the house mouse (Mus musculus) on Mana Island

Abstract

A live-trapping study on Mana Island, Cook Strait, carried out from March 1981 to February 1982 allowed estimates of house mouse (Mus musculus) abundance to be derived from six estimating equations. The estimating equations included one deterministic model (Weighted Mean), two stochastic models (Jolly-Seber & Manly-Parr), two regression methods (Zippin & Marten) and a nonparametric method (Burnham-Overton). All estimates were compared with the minimum number of mice known to be alive. The stochastic models give higher estimates for the initial months (April to June). All methods give similar estimates during the middle period of trapping when population numbers were low. The results suggest that the probability of capture differs amongst members of the population, and changes after capture. The effect of this on the population estimates is discussed. Four hundred and fifty eight specimens of M. musculus collected in 12 consecutive monthly samples (March 1981 to February 1982), from a variety of habitats on Mana Island, were measured, weighed and autopsied to provide information on sex ratio, age structure and reproduction of the population. Reproductive capabilities are high, yet the population is limited to an eight month breeding season. The diet from 282 autopsied rodents is described as being primarily leptidopteran larvae and seeds. The possible effects of this on breeding activity are discussed. High population densities are indicated by high snap trap returns, high population size estimates and small home ranges (mean size 0.26 ha), derived from footprint tracking and live trapping data. Five parasites were found on M. musculus, consisting of a cestode (Hymenolepis diminuta), two nematodes (Mastophorus muris and Physaloptera getula) and two species of Siphonaptera (Leptopsylla segnis and Nosophylla londiniensis londiniensis). Despite a high degree of nematode parasitism, general body condition is considered to be good, and average body measurements are shown to be larger than those reported by New Zealand mainland studies.