Introduced small mammals and invertebrate conservation in a lowland podocarp forest, South Westland, New Zealand.

This study presents the results of an investigation into invertebrate predation by introduced small mammals and the population dynamics of the ship rat and stoat in lowland podocarp forest. The study area was Okarito Forest, South Westland, New Zealand. The invertebrate faunas of four study sites were compared using pitfall traps and found to be similar, justifying the assumption that small mammals trapped within any of these sites would have access to a similar range of invertebrate food items. Drainage was found to be the most important environmental variable influencing invertebrate distribution. Diet of the stoat, ship rat and possum were investigated using gut contents obtained from animals trapped between June 1994 and February 1995. The diet of stoats during 1994/95 included a much larger invertebrate component than expected (close to 50% of volume). It is proposed that this is due to stoats switching to invertebrates in a year of low rat abundance. Stoats appeared to selectively consume freshwater crayfish and the weta, Gymnoplectron sp.. Other invertebrates made up an insignificant part of stoat diet. Invertebrates predominated in ship rat diet. It is proposed that ship rats are not foraging in a particularly selective manner, but consume invertebrates in relation to size and availability. Invertebrates made up a small but significant part of possum diet. Large invertebrates were apparently preferred and possums were found to selectively consume large invertebrates in times of increased abundance or activity. As a result of a study of population dynamics, utilising trapping and autopsy information from previous years, a system very similar to that described for New Zealand's beech forests is proposed. In the majority of years Okarito Forest is thought to provide at best a marginal habitat for the ship rat. However a rimu mast year leads to increased rat abundance, which in turn results in increased stoat numbers. Data indicates that the trapping season of 1993/94 was a mast year and that of this study (1994/95) a crash year. Although this study does not provide enough data to determine the effects of predation on the invertebrate fauna, some general comments are made regarding invertebrate conservation and the effects of mammalian predation on the invertebrate fauna of mainland New Zealand