The Relationships between Beech (Nothofagus Sp.) Seedfall and Populations of Mice (Mus musculus), and the Demographic and Dietary Responses of Stoats (Mustela erminea), in Three New Zealand Forests

Abstract

SUMMARY (1) There was a significant relationship between seedfall and populations of mice in all forests studied. Good seedfalls were followed by increases in density of mice and changes in population structure not seen in years when the seedfall failed. Predictability of these changes decreased with increasing diversity of forest composition. (2) Stoats responded both numerically and functionally to increases in mice. There was a significant relationship between the densities of mice and of stoats in summer in two forests. (In a third, this relationship was modified by the presence nearby of many lagomorphs, an important alternative prey for stoats.) The numerical response was due to increased survival, in the uterus or the nest, of young stoats born in spring when mice were abundant, not to increased fecundity of adult female stoats. In all three forests, stoats ate significantly more mice in the summers that mice were abundant. (3) Ship rats (Rattus rattus (L.)) were most abundant in the most diverse of the three forests. The density index for R. rattus increased there after the 1976 seedfall, but not after that of 1979. A few kiore (R. exulans (Peale)) coexisted with R. rattus in this forest. (4) Stoats did not eat significantly fewer birds when there were plenty of mice, either in summer or autumn, but nevertheless there was a significant negative correlation between the proportions of birds and mice in the diet in autumn. In summer this correlation was very weak and probably non-existent. (5) A 'bird predation index', which takes into account both the numbers of stoats present and what they ate, suggests that more birds are eaten in summers of peak numbers of mice and stoats, because of the strong numerical response by stoats to mice and lack of 'buffering' of birds by mice. In autumn this effect is weakened by a regular seasonal switch by stoats from birds to mice, especially in the autumn of a mouse peak year. (6) The indirect effects of mouse irruptions on populations of forest birds in New Zealand should be further examined. Control of stoats in mouse peak summers could prevent a temporary increase in predation on birds by stoats during the highest risk period, though whether this would benefit the birds is unknown.