Macquarie Island is a Nature Reserve under the Tasmanian National Parks and Wildlife Act 1970 and also a World Heritage Area but it has been modified significantly as the result of the introduction and establishment of exotic species including the house mouse, Mus musculus (Brothers and Copson 1988). Current attitudes favour the reversal of changes caused by such introductions, however, to date, ... efforts on the island have concentrated on the control of rabbits (Oryctolagus cuniculus) and cats (Felis catus). Although cats were extirpated some few years ago, this was followed by a considerable increase in the rabbit population. Control of both rabbits and rodents is currently being addressed (Anon 2007).
Invertebrates are rarely considered in conservation decisions even though invertebrate interactions have been established as playing an integral role in maintaining ecosystem function emphasizing their ecological importance (Majer 1987; Wilson 1987; Kremen et al. 1993). Examples of their various roles are their importance in soil aeration and drainage, litter decomposition and nutrient cycling, pollination, seed distribution and survival and herbivory (Majer 1987). Comparative studies of secondary production by insects and vertebrates invariably show that insects are greater producers of biomass and conduits of energy through communities than vertebrates (Price 1984). In the subantarctic environment, where many of these processes occur at a low rate much of the time (Hnatiuk 1993), altering the composition of invertebrate communities could have a significant impact on ecosystem processes (Hanel and Chown 1998). Moreover, macroinvertebrates have been shown to be responsible for most litter decomposition on subantarctic Marion Island (Chown and Smith 1993; Smith 1993; Hanel and Chown, 1998).
Introduced rodents have the potential to indirectly alter ecosystems of subantarctic islands through their impact on the invertebrate fauna (Crafford 1990). On Macquarie Island, Copson (1986) found that spiders made up a significant proportion of the diet of the house mouse. Of 108 stomach contents examined, spiders were recorded in 84% of stomachs and were common or abundant in 49% of those. The three spider species that occur on Macquarie Island (Greenslade 2006) are probably the major predators of small invertebrates. It is possible therefore that alteration of spider density has significant flow-on effects in both the invertebrate community and the systems of which they are a part. It is not clear however if mouse predation is important in the regulation of spider densities.
The aim of this study was to test the hypothesis that predation by M. musculus affects the densities of the three spider species, Myro kerguelensis O. P. Cambridge, Parafroneta marrineri (Hogg) and Haplinis mundenia (Urquhart) present on Macquarie Island. An exclusion experimental design was used.
This work was completed as part of ASAC project 104 (ASAC_104).
The experiment was located in grassland, predominantly Poa foliosa, on Wireless Hill (100m asl) (North Head), Macquarie Island (54o39's, 158o50'E) and operated from January to March 1993. The experiment comprised three plot treatments with five replicates, each plot being 2m x 2m in area (Fig. 1). The first (caged) type of plot was fenced to exclude ... mice, and spider densities here were increased to 150% of background densities by transferring spiders from outside the exclosure. This was done because it was not possible to conduct the experiment for a long enough period to allow spider densities to increase naturally if the relaxation of predation pressure was going to have an effect. The second (control) treatment was not fenced but spider densities also were increased to 150% of background densities. These plots were to control for the effect of increasing population densities in the cage plots. The third (natural) treatment was not manipulated and so untreated. Plots were arranged in a randomized block design with five replicates for each treatment and with treatments randomly assigned to plots within blocks (Fig. 1).
Fences were set up in early January, 1994 and consisted of two layers of wire mesh, offset so that the largest hole was about 6 mm in diameter. The fence was installed 30cm into the ground to prevent mice from burrowing underneath it and they extended to a height of 1m. To minimise disturbance to the plots from this process, intact blocks of grass and soil were carefully removed from the perimeter of the plot, the fence installed, and the blocks replaced.
Destructive sampling was used in five plots measuring 1m x 1m to estimate spider densities outside the exclosures. This involved pulling up all vegetation in the plot and hand collecting all spiders found in the extracted material. At the beginning of the experiment, these spiders were added to the centre of the caged and control plots.
Breakback traps for mouse capture were set with butter as bait inside both caged and control plots for the duration of the experiment. Traps were checked regularly. Initially skuas (Stercorarius antarctica) removed the bait from the breakback traps so we placed the traps inside 3cm lengths of 13 cm diameter PVC pipe.
The experiment ran for two months. In early March, we took a destructive sample from a 0.5m x 1m (0.5m2) area at the centre of each plot. Again, this involved tearing up all vegetation and collecting all spiders encountered. Five pitfall traps, 1.8cm in diameter three quarters filled with alcohol, were then set in each plot for seven days after which the fences were taken down.
In the laboratory, all spiders caught in the pitfalls were identified, sexed and measured. Lengths were measured from the anterior surface of the head to the cribellum.
We used analysis of variance (ANOVA) to test for an effect of mouse predation on the density of the three spider species.
The data are available for download from the provided URL.
van Klinken R., Green A. (1992), The first record of Oniscidea (terrestrial Isopoda) from Macquarie Island., Polar Record, 28, 166, 240-242, doi:doi:10.1017/S0032247400020763
Greenslade P. (1992), New records of Mesaphorura (Collembola: Onychiuridae, Tullbergiinae) species from Australia, Macquarie Island and the Antarctic., Transactions of the Royal Society of South Australia, 116
Davies K.F., Melbourne B.A. (1999), Statistical models of invertebrate distribution on Macquarie Island: a tool to assess climate change and local human impacts., Polar Biology, 21, 240-250
Davies K.F., Greenslade P., Melbourne B.A. (1997), The invertebrates of sub-Antarctic Bishop Island, Polar Biology, 17, 455-458
Greenslade P. (1990), Notes on the biogeography of the free-living terrestrial invertebrate fauna of Macquarie Island with an annotated checklist., Papers and Proceedings of the Royal Society of Tasmania, 124, 1, 35
Greenslade P. (1990), Terrestrial invertebrates (excluding parasites and mites) of Macquarie Island., Subantarctic Macquarie Island: Environment and Biology, 264