Western Australia’s tropical bat fauna comprises two communities; both are strict ensembles. One involves up to 19 species and occupies mangrove forest; the other, its landward counterpart, involves up to 22 species. Each includes habitat-generalist, productivity-dependent and cave-dependent species, so they achieve their full diversity in mangrove or riparian ecosystems associated with cavernous landscapes. Periods of significantly reduced rainfall are recognised in the climate record of the last ∼20 000 years, during which mangrove and upland riparian forests were the only highly productive ecosystems remaining in tropical Western Australia. During glacial cycles, these two refugia for mesic-adapted bats become geographically separated by the exposed continental shelf, a flat coastal plain more than 100 km wide. We review mangrove forests as evolutionary refuges during the Pleistocene, and use information on geographic range, endemism and habitat specificity to investigate the role of climatic and sea-level fluctuations in generating bat community structure. We conclude that mangrove forests provided a glacial refuge for all ensemble components; that the separation drove the evolution of two different ensembles. Northern Territory populations of Western Australia’s landward and mangrove specialists occupy landward as well as mangrove ecosystems, implying that mangrove refugia were not accessible during glacials.

Australia was isolated for approximately 40 million years from the presence of eutherian predation until the introduction of the dingo (Canis familiaris; 4000 years ago), foxes (Vulpes vulpes; 1871) and feral cats (Felis catus; post-1788). The arrival of these invasive species coincides with the decline and extinction of many native mammals, specifically within the critical weight range (35–5500 g). These extinctions are likely a result of competition and predation, where locomotor performance and the associated behaviours contribute largely to overall fitness. We used the population responses of native fauna in the presence of introduced predators to establish a research framework. Introduction/extinction timelines, predator diets, and prey occurrence were used to identify invasive/native relationships where predation may define the population outcome. We then examined the locomotor performance of these species using current data (maximum speeds). Consumption of prey items does not seem to be associated with the probability of the predator encountering the prey. Dingoes had the most variable mammalian prey of all invasive predators, likely due to higher maximal speeds. Feral cats favour Dasyuridae and smaller species, preying upon these prey groups more than dingoes and foxes. The role of locomotor performance in invasive ecology is not well understood; we identified relationships for further exploration.

Parasite load can vary with seasonality, but this is rarely quantified. The garden skink (Lampropholis guichenoti) is host to multiple species of endoparasite. To measure seasonal effects of parasite transmission we established three captive groups of wild-caught individuals in which 2 of 16 individuals (12.5%) were initially infected with nematodes. We collected three faecal samples from each lizard, a sample at the beginning and at the end of the non-activity season and at the end of the following activity season. We measured parasite load (ascarid group) by counting parasite eggs per gram of faeces using a microscope. We found that parasite load was significantly higher in the activity season than in the non-activity season. The prevalence of parasites increased from 15.9% in the non-activity season to 72.5% in the activity season. The activity season is characterised by greater host activity and warmer ambient temperatures, which promote parasite egg survival in the environment as well as egg development. Taken together, this facilitates parasite transmission and could ultimately explain the higher parasite load during the activity season.