Predation by European red foxes is believed to be the major cause of the extinction and decline of a large number of native medium-sized terrestrial mammals in Australia. We examined the impact of poisoning of foxes on the relative abundance of a group of medium-sized mammals in an experiment conducted in three large forest blocks in south-eastern Australia. The blocks consisted of paired sites, as follows: one site where poison baiting was used to control foxes (treatment site) and one where foxes were not controlled (non-treatment site). At all six sites, the population responses of a range of mammals were measured, and compared between treatment and non-treatment sites. The relative fox abundance, as indexed by bait-take, declined during the course of the study at treatment sites and to a lesser extent at non-treatment sites. The decline in bait-take at non-treatment sites was most likely due to treatment sites acting as ecological traps, so that reduced intra-specific competition attracted foxes from non-treatment to treatment sites, where they were subsequently poisoned. There was a significant treatment effect for the abundances of total mammals, long-nosed potoroos, southern brown bandicoots and common brushtail possums, with higher abundances at treatment sites than at non-treatment sites. Common ringtail possums increased in abundance during the course of the study, with no significant difference between treatment and non-treatment sites. There was no significant effect of time or treatment on the abundance of long-nosed bandicoots. The increase in the abundance of native mammals at treatment sites was most likely due to a lower predation pressure by foxes brought about by fox control, and the smaller increase in abundance in non-treatment blocks was likely due to the ecological-trap effect because of fox baiting at treatment sites. The present study demonstrated that broad-scale fox control can lead to increases in the abundance of native mammals in forested habitats, without recourse to aerial baiting or fences. The study also demonstrated that the influence of fox control on the fox abundance can extend well beyond the perimeter of the area baited.

The home ranges, movement patterns and spatial organisation of spotted-tailed quolls (Dasyurus maculatus maculatus) were studied in rainshadow woodland in southern New South Wales, Australia. Fourteen individuals were radio-collared and simultaneously tracked. Home-range size estimates ranged from 621 ha to at least 2561 ha for males, and 88 ha to at least 653 ha for females. Mean home-range size was significantly greater for males (992 ± 276 ha) than females (244 ± 72 ha). The maximum straight-line distances between successive fixes for males over 24-h and 48-h periods were 2529 and 4430 m, respectively, while for females these distances were 1865 and 3085 m. Mean maximum straight-line distances between successive fixes for males over a 24-h period were not significantly different from mean maximum straight-line distances of females (1493 ± 918 v. 1034 ± 540 m). However, over 48 h, the mean maximum distances between successive fixes for males was greater than that of females (2911 ± 934 v. 1680 ± 918 m). The home ranges of males mostly overlapped with those of other individuals, regardless of sex. In contrast, home ranges of females tended not to overlap with those of other females, suggesting some form of spatial separation. Home-range estimates derived for spotted-tailed quolls in our rainshadow woodland study site are comparable to those obtained for the species in wetter vegetation types. From a conservation perspective, it seems that habitat structure and the availability of prey is more important than rainfall or vegetation type in determining spatial requirements of the species. Until more advanced telemetry systems are developed, caution should be applied when using current home-range data on the species to infer breeding systems and patterns of spatial organisation, particularly the issue of territoriality among female quolls.

Using a biomarker, we assessed the propensity of spotted-tailed quolls (Dasyurus maculatus) to encounter and consume non-toxic meat baits, ordinarily laced with the poison 1080 (sodium monofluoroacetate) and deployed for control of wild dogs (Canis lupus dingo, Canis familiaris and hybrids of the two) in southern Australia. In the first experiment, 60 unpoisoned meat baits injected with Rhodamine B were placed on the surface of the ground at 250-m intervals along two separate transects crossing an open woodland study area. One week after placement, a range of animals, including quolls, had removed all baits. Microscopic assay of whisker samples collected from live-captured quolls later revealed that 6 of 10 (60%) animals were positive for the biomarker, indicating that they had encountered and consumed baits. In the second experiment, conducted at the same site one year later, 150 similarly prepared meat baits were delivered aerially from a helicopter along the same transects, at a rate of one bait every 100 m. Eight of 17 quolls (47%) were found to have encountered and consumed at least one and up to five baits. Combined with previous studies, our results reaffirm that surface or aerial baiting operations for wild dogs may place local quoll populations at risk. However, further research is necessary to establish the relationship between this risk and actual mortality levels during such baiting operations since there are a number of factors that may influence the toxicity of baits for spotted-tailed quolls in a field situation as well as the danger those baits may pose.

No abstract