Summary1. Native recolonization of abandoned farmland in the wheat-growing region of western Australia is slow to non-existent, even 45 years after abandonment. Instead, old-fields tend to be dominated by non-native annual grasses. The native vegetation in the region is highly fragmented. We predicted that recovery is limited by seed availability and recruitment failure. 2. We compared the seed rain, soil seed bank and extant vegetation of three old-fields and three adjacent native woodland remnants to test our predictions. 3. Seed dispersal limitation was evident at two old-fields: species richness of the native seed in the soil seed bank declined at 50-m increments (i.e. 0 m, 50 m and 100 m) into the old-fields. Species richness of recolonizing native vegetation also declined along this transect. 4. Recruitment limitation was evident at one old-field, probably linked to the few abundant non-native species that dominated the vegetation. These non-natives dominated the seed supply, and tended to germinate more rapidly than the native seeds. 5. In contrast, seed dispersal and recruitment had not limited native recolonization of the third old-field: 91% of the species present in the adjacent remnant had recolonized this old-field. Species richness and abundance of non-natives in the seed supply and extant vegetation were minor compared with that of the other old-fields. Therefore an adequate seed supply of natives, the lack of non-natives and chance natural disturbance events (i.e. cyclone and fire) had probably contributed to this result. 6. Synthesis and applications. Native recolonization of wheatbelt old-fields is contingent upon the unusual coincidence of seed availability, favourable conditions for recruitment and an absence of competitive non-natives. Grasses, annual herbs and Acacia shrubs are dominant among the few native species that have consistently recolonized these oldfields. In most cases, direct seeding and control of non-natives are the minimum requirements for restoration of wheatbelt old-fields to species-rich eucalypt woodlands.

No abstract

The ghost bat (Macroderma gigas) is Australia’s largest echolocating bat. It is restricted to several disjunct populations in the north of the continent, including a population in the Pilbara region of Western Australia. In 2016 the ghost bat was listed as Vulnerable under Australian federal legislation, owing to declining numbers across many regional populations. The most severe threat to ghost bats in the Pilbara region is the destruction and disturbance of habitat due to mining operations, but disturbance to their roosts from other infrastructure developments and changes to and loss of foraging habitat also pose significant threats. A set of research priorities for ghost bats in the Pilbara was developed during a workshop attended by mining industry representatives, environmental consultants, scientists and government regulators. Five research priorities were identified: (1) identify and characterise critical diurnal roosts and foraging habitat; (2) improve knowledge of the distribution, movement and dispersal patterns of ghost bats in the region; (3) improve knowledge of population size, persistence and long-term trends; (4) better understand the cumulative, direct and indirect impacts of mining and other development activities; and (5) better understand the threats posed by fence entanglements, cane toads and feral cats.