Adaptation and adaptability in wild-living House mice (Mus musculus)



House mice (Mus musculus L.) have successfully colonized a wide variety of habitats throughout the world. In part this is due to their ability to adapt genetically to new situations; in part to their physiological flexibility. The most important part of this individual adaptability is the response to low temperatures.

Cold tolerance seems to be largely metabolic in mice, and can be detected by an increase in the oxygen carrying capacity of the blood and an ability to mobilize body reserves, to some extent indicated by the amount of brown fat. In addition, the relative surface area of the body, and the amount of heat radiating tissue (especially tail length) are important. The paper presents data from both feral and commensal populations of mice, and from a range of habitats which are likely to be stressful to the animals. Most of these (from Macquarie Island in the Australian sub-Antarctic, from islands in the Faroe, Shetland and Orkney groups, from Skokholm off the Welsh coast, and localities on the British mainland) are based on single samples, but we have also included data from one genetically closed population (the Isle of May) where we have sampled the same population on a number of different occasions throughout the year.

All the haematocrit values except one (a mid-winter value from the Isle of May) fall within the range in different inbred mouse strains, although the haemoglobin concentrations were mostly higher than those in laboratory mice. However the wide range of haematological values in genetically different strains kept in a constant environment shows the fallacy of using absolute physiological values as measures of adaptability. This in turn highlights the importance of using a range of phenotypic traits to indicate response to environmental pressures–both physiological and genetical. We obtained a greater accuracy of allocation of individuals into age, genetical, or survivorship groups when using a number of traits in a multiple discriminant analysis, than when only considering single characters.