ECOLOGY OF THE STOAT IN NOTHOFAGUS FOREST - HOME-RANGE, HABITAT USE AND DIET AT DIFFERENT STAGES OF THE BEECH MAST CYCLE

We studied the ecology of a high-density population of stoats in Fiord land, New Zealand, in the summer and autumn of 1990-91 following a Not hofagus seeding in 1990. Results are compared with findings from the s ame area in 1991-92, a period of lower stoat density. In the high-dens ity year, minimum home ranges (revealed by radio-tracking) of four fem ales averaged 69 ha and those of three males 93 ha; range lengths aver aged 1.3 km and 2.5 km respectively. Neither difference was statistica lly significant. For combined sexes, average range area in the high-de nsity year was significantly less, and range length was significantly shorter, than in the following year. When we compared stoat diet in th e high-density year with that in the following two years, there were n o significant differences in the frequencies of occurrence of birds or invertebrates in stoat guts. Overall, bird remains were found in 56% of guts, and invertebrates in 28%. Possum remains occurred in 6% of ma le stoats but were never found in females. Mice were only detected in stoats in the high-density year, when they occurred in 54% of guts. La gomorphs occurred significantly more often in the guts of stoats durin g lower-density years (26%) than the high-density year (7%). Seedfall in Nothofagus forest is synchronous and periodic. Following seedfall, mouse density rises dramatically, followed by a sharp rise in stoat nu mbers. It has been suggested that mice feed on the abundant seed and t hat stoats in turn increase because of the large numbers of mice avail able to them. We suggest that the situation is more complex and that i ncreases in not only mouse, but also bird (and possibly invertebrate), densities may contribute to the high productivity of stoats in the ye ar following a Nothofagus seedfall.