Our understanding of recent extinctions and declines of Australian mammals is poor, particularly where there is a paucity of data to quantify change. The swamp antechinus (A. m. maritimus) has a fragmented, coastal distribution in south-east Australia. Although long-term studies (1975–2007) of this vulnerable species were conducted in the eastern Otways, its current status was unclear. We assessed the success of live trapping and camera trapping (2013–17) at 42 sites, 19 where the species was trapped previously. Between 2013 and 2015 A. m. maritimus was recorded at only 6 sites (n = 8), but at none in 2016–17. Assessment of long-term changes found that high-density populations occurred after above-average rainfall, and both low- and high-density populations collapsed after wildfire, after low rainfall, and in fragmented habitat. The species may now be restricted to very small populations in refuges such as coastal dunes, and predicted low rainfall and increased burning frequency pose major threats to the species’ survival. Recovery is unlikely without targeted management, including predator control and protection from inappropriate fire regimes and habitat fragmentation. If similar declines have been experienced across the species’ range, prevention of extinction of the species will require similar management strategies.

The conversion of natural forest to agroforestry plantations and annual cropping systems alters the soil habitat and food resources for biota, including earthworms. Native earthworm species may disappear whereas exotic species with greater tolerance of disturbance and less niche specialisation may thrive. The objective of the study was to compare the earthworm diversity in managed forest and agroforestry systems, which were cultivated for mixed plantation and annual crop production on Mount Gede, Indonesia. All the habitats in the study area were impacted by humans. The forest habitat was a managed forest, with a permanent tree cover, whereas mixed plantation had a partial shrub cover. Meanwhile, homogenous plantation was cultivated with annual crops. Among 3787 individuals collected during July–October 2012, five Oriental earthworm species were identified in the soil communities of Mount Gede: Drawida nepalensis, Notoscolex javanica, Pheretima pura-group, Polypheretima moelleri, and Polypheretima sempolensis. Also, 18 species were found that are reported to be non-Oriental in origin. Anthropogenic disturbance of forests on Mount Gede, due to conversion into plantations, alters the earthworm environment by increasing soil water content, temperature and total phosphorous content, while decreasing organic carbon. N. javanica was the only native species to survive this deforestation, while the exotic Ocnerodrilus occidentalis and Pontoscolex corethrurus thrived, becoming the eudominant species. From the forest area to the mixed and homogenous plantations, the predicted decreasing diversity is evidenced by the lowering trend of Shannon’s diversity index. In conclusion, the land-use change into mixed plantations and annual croplands has reduced earthworm diversity in this region of Mount Gede, Indonesia.

Habitat use is the most common dimension along which sympatric species partition resources to reduce competition. We conducted faecal pellet counts at Wilsons Promontory National Park, Victoria, to examine habitat use by an assemblage of mammalian herbivores with disparate evolutionary histories and varying body size: introduced European rabbit (Oryctolagus cuniculus) and hog deer (Axis porcinus), and native eastern grey kangaroo (Macropus giganteus), swamp wallaby (Wallabia bicolor) and common wombat (Vombatus ursinus). Overlap in habitat use was low between four pairs of species, suggesting spatial partitioning of resources to reduce the potential for interspecific competition. More generally, however, overlap in habitat use was high, particularly between native and introduced grazers. These results indicate the potential for competition if resources were limiting and suggest that assemblages of species with independent evolutionary histories have inherently less resource partitioning to facilitate coexistence than assemblages of species with common evolutionary histories. Despite evidence of high overlap in habitat use between native and introduced species at a broad scale, and variation in the competitive ability of species, coexistence was likely facilitated by niche complementarity, including temporal and fine-scale partitioning of spatial resources. There was no relationship between body size and the diversity of habitats used. In contemporary assemblages of native and introduced species, evolutionary history is likely to have a strong influence on resource partitioning.

The status of Pseudantechinus roryi relative to its congeners has been determined from DNA sequences obtained from both nuclear and mitochondrial gene loci. Although all other recognised species of Pseudantechinus form reciprocally monophyletic lineages in phylogenetic analyses, individuals identified in museum collections as Ps. roryi (including type specimens) were indistinguishable from those identified as Ps. macdonnellensis. Ps. roryi is thus considered to be a synonym of Ps. macdonnellensis. Neighbour-joining network analyses failed to reveal any clear biogeographic differences between populations of Ps. macdonnellensis other than some evidence of isolation by distance.

For many years Phryssonotus novaehollandiae Silvestri, 1923 was thought to be the only species of this genus present in Australia. Specimens collected from three geographically separated populations had similar taxonomically important morphological characters and body lengths. However, their body scale trichomes exhibited three distinctive patterns: a trapezoid, T-shape, or dark banding, and the length-to-width ratios of their trichomes also differed. These differences, as well as results of phylogenetic analysis using 18S and COI gene regions from representatives of the three populations, suggested that more than one Phryssonotus species were present. Specimens of Phryssonotus preserved in ethanol can prove difficult to distinguish because the colour patterning gradually fades. Examination of the colour patterns of live specimens, other morphological characters and DNA analysis together, were found to be the most reliable method of correctly distinguishing between these three species. P. novaehollandiae Silvestri, 1923, the first described species collected from South Australia, remains, with its range extended to inland Victoria, and two new species (P. australis, sp. nov. and P. occidentalis, sp. nov.) are described.

Islands provide the last refuge for many Australian species that have succumbed to range contractions since European settlement. These species have heightened vulnerability when developments are planned for the islands. The burrowing bettong (Bettongia lesueur), a threatened macropod, was once widespread across Australia but now occurs naturally only on Barrow Island and two other Western Australian offshore islands. Here, we document the demographic, social and genetic characteristics of the Barrow Island population, using nuclear microsatellites, mtDNA and demographic parameters from 286 individuals trapped around 29 warrens. The Barrow Island population utilises complex warren systems scattered across the landscape and exhibits continuous breeding, even during extensive dry conditions. Males (759 g) were significantly heavier than females (735 g), with the latter reaching sexual maturity at ∼600 g. Warrens varied markedly in size and number of individuals, with 20% of females and 25% of males moving between nearby warrens. There were two deep mtDNA haplotype clades that coalesce around 298 000 years ago, indicating that this island population has maintained an effective population size that has permitted the retention of one marked feature of its preisolation genetic diversity. Dispersal and gene flow between warrens was severely constrained. Both mtDNA and 11 nuclear microsatellites showed a strong isolation by distance effect, with genetic differences between warrens increasing linearly with geographic separation. While this was marked in both sexes, it was stronger in females and consistent with shorter dispersal distances for females than for males. The trapping and genetic data are concordant, with between-warren sociality correlated with genetic similarity. Long-term conservation, management and translocation programs will benefit from cognisance of the strong fidelity to natal areas, as evidenced by both recapture data and genetic analyses.

Habitat heterogeneity can have considerable effects on gene flow and migration across a region of parapatry. Describing habitat across a region of parapatry is important for the development of eco-evolutionary theory. Two subspecies of thick-billed grasswren (Amytornis modestus) share a region of parapatry between the South Australian salt lakes, Lake Eyre and Lake Torrens. While the two subspecies remain morphologically diverged outside the region of parapatry, it is not known what factors within the region of parapatry may affect migration and gene flow. In this study, we test associations between habitat differences and subspecies distributions and discuss whether ecological barriers could play a role in mitigating gene flow between the subspecies. We compare dominant plant species (1) between the allopatric ranges of the subspecies and within their region of parapatry, and (2) in relation to presence or absence of grasswrens within their region of parapatry. We found that the dominant plant species differed between grasswren subspecies in their allopatric range and in their region of parapatry, and also differed in the region of parapatry at sites with or without grasswrens. Specifically, grasswrens were absent in vegetation that is typical of sand dunes. These findings are discussed in light of evidence for secondary contact and hybridisation between A. m. indulkanna and A. m. raglessi, and susceptibility to introgression.