Leggadina Forresti and Pseudomys

Home , Desert mouse, Hopping mouse, Leggadina, Pseudomys, Sandy inland mouse, Silky mouse

Approved Recovery Plan

Forrest’s Mouse (Leggadina forresti) and Sandy Inland Mouse (Pseudomys hermannsburgensis) Recovery Plan

Leggadina forresti

Pseudomys hermannsburgensis

July 2002 © NSW National Parks and Wildlife Service, July 2002. This work is copyright. Apart from any use as permitted under the Copyright Act 1968, no part may be reproduced without prior written permission from NPWS. NSW National Parks and Wildlife Service 43 Bridge Street (PO Box 1967) Hurstville NSW 2220 Tel: 02 95856444 www.npws.nsw.gov.au

For further information contact Threatened Species Unit, Western Directorate. NSW National Parks and Wildlife Service P.O. Box 2111 Dubbo NSW 2830 Tel (02) 6883 5358

Cover illustrations: Leggadina forresti: Photographer: Andrew Henley, ã Nature Focus Pseudomys hermannsburgensis: Photographers: H&J Beste, Ó Nature Focus

This Plan should be cited as follows; NPWS (2002). Forrest’s Mouse (Leggadina forresti) and Sandy Inland Mouse (Pseudomys hermannsburgensis) Recovery Plan. NSW National Parks and Wildlife Service, Hurstville NSW.

ISBN: 0731365151 NSW National Parks and Wildlife Service Recovery Planning Program

Forrest’s Mouse (Leggadina forresti) and Sandy Inland Mouse (Pseudomys hermannsburgensis) Recovery Plan

Prepared in accordance with the New South Wales Threatened Species Conservation Act 1995

July 2002 Acknowledgements

Information and assistance that was useful in the preparation of the plan was obtained from Sandy Ingleby, David Read, David Croft, Lars Kogge, Kim Piddington, Rebecca Montague-Drake, Gerry Swan, Murray Ellis, Dani Ayers, Michele Cooper, David Monahan, Robyn Molsher, Rita Enke, Silvia Ricci, Paul Mahon and Terry Dawson.

ii Foreword

The conservation of threatened species, populations and ecological communities is crucial for the maintenance of this State’s unique biodiversity. In NSW, the Threatened Species Conservation Act 1995 (TSC Act) provides the framework to conserve and recover threatened species, populations and ecological communities through the preparation and implementation of recovery plans.

The preparation and implementation of recovery plans are identified by both the National Strategy for the Conservation of Australia’s Biological Diversity and the approved NSW Biodiversity Strategy as a key strategy for the conservation of threatened flora, fauna and invertebrates. The object of a recovery plan is to document the research and management actions required to promote the recovery of a threatened species, population or ecological community and to ensure its ongoing viability in nature.

This plan describes our current understanding of Forrest’s Mouse and the Sandy Inland Mouse, documents research and management actions undertaken to date and identifies actions required and parties responsible to ensure ongoing viability of the species in the wild.

NSW National Parks and Wildlife Service has prepared the Forrest’s Mouse and Sandy Inland Mouse Recovery Plan with the assistance of a number of people. I thank these people for their efforts to date and look forward to their continued contribution to the recovery of the species.

BOB DEBUS MP Minister for the Environment

iii Executive Summary

Legislative Context

The Threatened Species Conservation Act 1995 (TSC Act) is New South Wales’ most comprehensive attempt at establishing a legislative framework to protect and encourage the recovery of threatened species, populations and communities. Under the TSC Act, the Director-General of National Parks and Wildlife has certain responsibilities including the preparation of recovery plans for threatened species, populations and ecological communities. This Recovery Plan has been prepared in accordance with the provisions of the TSC Act.

Preparation of Plan

This Recovery Plan has been prepared with the assistance of interested parties with relevant expertise. Components within the plan do not necessarily represent the views nor the official positions of all the individuals or agencies consulted. The information in this Recovery Plan was accurate to the best of the NPWS’ knowledge on the date it was approved.

Current Conservation Status

Forrest’s Mouse

Forrest’s Mouse is listed as vulnerable on Schedule 2 of the Threatened Species Conservation Act 1995 (TSC Act). The species is not listed on the Commonwealth Environment Protection and Biodiversity Conservation Act 1992 (ESP Act), but is listed as “lower risk, near-threatened” in the IUCN (1996) Red List of Threatened Species.

Sandy Inland Mouse

The Sandy Inland Mouse is listed in Schedule 2 of the Threatened Species Act 1995 (TSC Act) as vulnerable. The species is not listed in the Commonwealth Environment Protection and Biodiversity Conservation Act 1992 (ESP Act) or in the IUCN Red List of Threatened Species.

iv Objectives of the Recovery Plan The overall objective of this Recovery Plan is to secure extant populations of Forrest’s Mouse and Sandy Inland Mouse in New South Wales through appropriate management.

Specific objectives are to:

1) monitor the populations of Forrest’s Mouse and Sandy Inland Mouse to determine their status;

2) protect the populations of Forrest’s Mouse and Sandy Inland Mouse from the threatening processes of overgrazing by herbivores and predation by feral animals;

3) survey areas of potential habitat for new extant populations of Forrest’s Mouse or Sandy Inland Mouse;

4) promote appropriate grazing practices and feral animal control where Forrest’s Mouse or Sandy Inland Mouse occurs on non-Service estate through cooperation with landholders and other relevant stakeholders;

5) improve current knowledge of the biology and ecology of both Forrest’s Mouse and the Sandy Inland Mouse.

Recovery performance criteria Recovery criteria are that:

1) an ongoing system of trapping and release is implemented on and off Service estate to monitor the populations of small mammals;

2) the control and monitoring of grazing herbivores and feral predators is continued on the Service estate;

3) appropriate management is instituted where Forrest’s Mouse or Sandy Inland Mouse occurs on areas of non-Service estate through the provision of information and, where appropriate, assistance to landholders and other relevant stakeholders;

4) areas of potential habitat for Forrest’s Mouse and Sandy Inland Mouse are identified and surveyed for new extant populations of the species;

5) a greater understanding of the biology and ecology of Forrest’s Mouse and Sandy Inland Mouse in NSW is achieved through appropriate research.

v Biodiversity Benefits

The native rodent fauna of Western New South Wales has experienced a severe decline since European settlement. Of the seventeen native rodents which have been recorded in the arid zone of Western New South Wales, eleven are now regionally extinct, three (including Sandy Inland Mouse) are rare, two (including Forrest’s Mouse) are sparse, and one is common in limited habitat (Dickman, 1993). In view of this decline, the biodiversity value of the remaining species is considerable.

As well as aiding the recovery of Forrest’s Mouse and Sandy Inland Mouse, the actions contained in this Recovery Plan will have benefits for biodiversity in general. Actions such as control of feral predators and reduction of grazing pressure have benefits for a range of native species (Dickman, 1993), while surveys and ongoing monitoring will yield data which will be useful in the management of other native species.

Other threatened species that will benefit from actions detailed in this Recovery Plan include the Flame Spider-flower (Grevillea kennedyana), Australian Bustard (Ardeotis australis), Flock Bronzewing (Phaps histrionica), Squatter Pigeon (Geophaps scripta), Long-haired Rat (Rattus villosissimus), Inland Blind Snake, (Ramphotyphlops endoterus), Centralian Blue-tongue Lizard (Tiliqua multifasciata), Gunther’s Skink (Cyclodomorphus branchialis) and Stimson’s Python (Liasis stimsoni).

BRIAN GILLIGAN Director-General

vi Table of Contents

Acknowledgements

Foreword

Executive Summary

1 Introduction ...... 1

2 Species Information - Forrest's Mouse ...... 1 2.1 Current Conservation Status...... 1 2.2 Description...... 1 2.2.1 Taxonomy ...... 1 2.2.2 General ...... 2 2.3 Distribution ...... 2 2.4 Ecology ...... 4 2.4.1 Life Cycle...... 4 2.4.2 Diet...... 4 2.4.3 Habitat...... 5 2.5 Threats and reasons for decline ...... 6

3 Species Information - Sandy Inland Mouse...... 8 3.1 Current Conservation Status...... 8 3.2 Description...... 8 3.2.1 Taxonomy ...... 8 3.2.2 General ...... 8 3.3 Distribution ...... 9 3.4 Ecology ...... 9 3.4.1 Life Cycle...... 9 3.4.2 Diet...... 11 3.4.3 Habitat...... 12 3.5 Threats and reasons for decline ...... 13

4 Relevant Legislation...... 13 4.1 Threatened Species Conservation Act 1995...... 13 4.2 Environment Protection and Biodiversity Conservation Act 1999 ...... 14 4.3 National Parks and Wildlife Act 1974...... 14 4.4 Environmental Planning and Assessment Act 1979.....14 4.5 Native Vegetation Conservation Act 1997...... 14

5 Management Issues ...... 15 5.1 Scientific and taxonomic value ...... 15 5.2 Biodiversity value...... 15 5.3 Social and economic considerations ...... 15

6 Previous Actions Undertaken ...... 16 6.1 Reduction of grazing pressure...... 16 6.2 Control of feral predators ...... 16 6.3 Fauna surveys ...... 17

7 Species Ability to Recover ...... 18

8 Recovery Objectives and Performance Criteria ..... 18 8.1 Objectives of the Recovery Plan ...... 18 8.2 Recovery performance criteria ...... 19

9 Recovery Actions...... 19 9.1 Action 1 - Monitor known extant populations...... 19 9.2 Action 2 - Development of effective monitoring programs ...... 20 9.3 Action 3 - Targeted surveys for Forrest's Mouse and Sandy Inland Mouse in western NSW...... 21 9.4 Action 4 - Maintain and review the Service's ongoing program of feral animal management...... 22 9.5 Action 5 - Involve and inform relevant stakeholders...24 9.6 Action 6 - Promote further biological and ecological research ...... 25

10 Alternative Management Strategies...... 25 10.1 Management of fire regime ...... 25 10.2 No management action taken ...... 26

11 Implementation...... 26 12 Preparation details ...... 27 12.1 Date of last amendment ...... 27 12.2 Review date...... 27

References ...... 28 1 Introduction

The native rodent fauna of western New South Wales has experienced severe decline since European settlement. Of the seventeen species that have been recorded in the arid zone of New South Wales, only one, the Water Rat (Hydromys chrysogaster) remains common. Eleven are now considered extinct in New South Wales. The remaining five species are listed as threatened under the New South Wales Threatened Species Conservation Act 1995.

This multi-species Recovery Plan defines the means by which Forrest’s Mouse and Sandy Inland Mouse populations within the arid zone of New South Wales are to be managed and conserved.

2.0 Species Information – Forrest’s Mouse

2.1 Current Conservation Status

Forrest’s Mouse (Leggadina forresti) is listed as vulnerable in Schedule 2 of the Threatened Species Conservation Act 1995 (TSC Act). The species is not listed in the Commonwealth Environment Protection and Biodiversity Conservation Act 1999 (EPBC Act), but is listed as “lower risk, near-threatened” in the IUCN (1996) Red List of Threatened Species.

Ashby and Lunney (1985) list the status of Forrest’s Mouse in New South Wales as rare and Dickman (1993) lists it as sparse. Within western New South Wales, Dickman et al. (1993) classes its distribution as limited, its abundance as sparse, and its survival status as vulnerable. At the national level, Dickman et al. (1993) wrote that its distribution is widespread, its abundance is very sparse, and its survival status is secure.

2.2 Description

2.2.1 Taxonomy

Scientific Nomenclature: Leggadina forresti (Thomas, 1906) Family: Muridae Common Name: Forrest’s Mouse Other Common Names: Melrose Desert Mouse, Waite’s Mouse, Berney’s Queensland Mouse, Short-tailed Mouse, Southern Short-tailed Mouse, Forrest’s Territory Mouse Recent Synonyms: Mus forresti Pseudomys / Leggadina messoria Pseudomys / Leggadina waitei Gyomys / Leggadina berneyi

1 Forrest’s Mouse was first described by Thomas (1906) and assigned the name Leggadina forresti. Several authors (e.g. Tate, 1951; Troughton, 1967; Ride, 1970) have placed species of Pseudomys within the genus Leggadina or placed Forrest’s Mouse in the genus Pseudomys. In a study of penis morphology, Lidicker and Brylski (1987) placed hermannsburgensis, delicatula and forresti in the genus Leggadina. However, Watts (1976) maintained that Leggadina was a separate genus from Pseudomys, consisting of L. forresti and L. lakedownensis. This position is supported by studies involving electrophoretic analysis and albumin microcomplement fixing (Baverstock et al., 1981; Watts et al., 1992). Sperm morphology places L. forresti in the Hydromyinae. Further study is required to clarify the taxonomy, but convergent evolution may explain the morphological similarities between Leggadina and other genera (Watts et al., 1992).

Populations of Leggadina vary greatly in body size and dental morphology. There has been confusion in the past over the identity of some populations of Leggadina and a suggestion that a third species - the "Thevenard Island Mouse" - should be recognised (Cooper et al., unpubl.). However, a recent electrophoretic study by Moro et al. (1998) found that the Thevenard Island population belongs to L. lakedownensis. It is suggested, however, that this population be assigned to a new subspecies. Populations of Leggadina occurring in central Australia were found to be L. forresti, and populations from north-east Queensland to the Pilbara in Western Australia were L. lakedownensis (Moro et al., 1998).

2.2.2 General

Forrest's Mouse is easily recognised by its small size (15-25g) and short tail which is 60-70% of the combined length of the head and body (Morton, 1974; Dickman, 1993; Reid and Morton, 1995). The fur ranges from light grey to yellowish brown with a grizzling of darker hairs above, and is white below. It has relatively small ears and eyes and a short, broad muzzle (Dickman, 1993). Reid and Morton (1995) state that identification is also aided by thick, short, coarse hair. Appearance is variable across the range of the species (Reid and Morton, 1995).

2.3 Distribution

Prior to 1996, Forrest's Mouse had been recorded in only four locations within New South Wales. These were Fowlers Gap (north of Broken Hill), Sturt National Park, Mutawintji National Park (as subfossil remains), and in a construction trench to the south-east of Sturt NP (Morton, 1974; Denny, 1975; Ayers and Wallace, 1997; Ellis, in press). Since 1996, the Australian Museum has obtained specimens of Forrest’s Mouse from a number of properties in the Tibooburra area, adjacent to Sturt NP. Forrest's Mouse is also known from two locations in South Australia, just west of the New South Wales border (Dickman, 1993). Figure 1 shows the distribution of Forrest's Mouse in New South Wales.

2 Recent biodiversity surveys in western New South Wales have yielded several specimens of Forrest’s Mouse and identified a new population in the Paroo River overflow. The Woody Weeds Biodiversity Study, conducted in late 1999 and early 2000, collected two individuals from between Wanaaring and Louth (D. Ayers, pers. comm.). The most recent record of Forrest’s Mouse was obtained during the NSW NPWS Darling Riverine Plains Biodiversity Study, in November 2000, where a single Forrest’s Mouse was captured on a property, 15km north of Wilcannia (M. Cooper, pers. comm.). These most recent records represent a substantial extension of the known range of the species in NSW.

Forrest’s Mouse is widespread in Queensland, South Australia, Western Australia and the Northern Territory (Van Dyck and Longmore, 1991; Reid and Morton, 1995). Read (1995) states that Forrest’s Mouse could occur in the floodplain of the Great Anabranch of the lower Darling River, approximately 200 km south of its known distribution. The distribution of the species prior to European settlement is uncertain, but Reid and Morton (1995) consider that the range has probably not been greatly altered.

Figure 1: The distribution of Forrest’s Mouse in NSW.

3 2.4 Ecology

2.4.1 Life Cycle

Reproduction in Forest’s Mouse is thought to peak during winter and spring (Read, 1984), although some breeding may occur in autumn (Reid and Gillen, 1988). Reid and Morton (1995) suggest that reproduction may be more related to rainfall than to season. Litters usually consist of three or four young (Philpott and Smyth, 1967). Gestation is probably between 30 and 40 days, and the young weigh around 2g at birth. Suckling lasts for one month and the young are weaned when weighing around 9g (Watts and Aslin, 1981). The age at which individuals become sexually mature is unknown, as is their reproductive output (Dickman, 1993). Longevity is estimated by Watts (1982) to be 1.5 years, but may be greater than 2 years (D. G. Read, pers. comm.). Individuals have been observed to enter torpor (a state of reduced metabolic rate) both in captivity and in the wild (D. G. Read, pers. comm.).

Little is known about the social organisation or population dynamics of Forrest's Mouse (Dickman, 1993). They are thought to be mainly solitary and occur in low densities regardless of habitat suitability (Finlayson, 1941, 1961; Morton, 1974; Watts and Aslin, 1981). The simple structure of burrows supports the view that they are largely solitary (Dickman, 1993). Lactating females may be intolerant of other adults (Watts and Aslin, 1981).

Unpublished radio-tracking studies by C. R. Dickman show that Forrest's Mouse does not stay within a fixed home range and may travel linear distances of more than 1km in a night. These findings may explain the observation by Read (1984) that recaptures of the same individual did not occur. The species is thought to be strictly nocturnal and terrestrial (Dickman, 1993).

Analysis of Barn Owl pellets has provided evidence that populations of Forrest’s Mouse increase following extended periods of rainfall (Morton et al., 1977; Morton and Martin, 1979). However, the magnitude of such increases is unknown, and it is unclear whether the increases are due to a localised increase in breeding or to immigration from surrounding areas. Other than rainfall, factors regulating populations of Forrest’s Mouse are unknown (Dickman, 1993). The species is not thought to reach the high numbers typical of some inland rodents (e.g. Pseudomys hermannsburgensis (Predavec, 1994)), which plague during periods of favourable conditions (Reid and Morton, 1995).

2.4.2 Diet

Limited dietary studies suggest that Forrest’s Mouse is an omnivore (Dickman, 1993; Murray et al., 1999). In assessing the frequency of occurrence of various food items, Read (1984) found that the major item in the stomachs of three individuals was seed (47%) followed by arthropods (27%). The remainder (26%) consisted of roughly equal proportions of stem and leaf material. Arthropods found in the diet included beetles, isopods and spiders (Read, 1984). Watts (1972) and 4 Murray et al. (1999) have made similar dietary observations, each based on a single individual. Murray et al. (1999) also report finding a small quantity of fungal material (1.5%) in the stomach of one individual. It is thought that Forrest's Mouse, like other well-adapted desert rodents, obtains sufficient moisture from its food and therefore does not rely on standing water for drinking (Reid and Morton, 1995). Although this observation is untested, Dickman (1993) considers it plausible given the distribution of the species in areas with little or no standing water.

2.4.3 Habitat

Forrest’s Mouse is found in a variety of habitats, ranging from sand plains or ridges characterised by spinifex, to chenopod shrubland, Mulga woodland, savanna woodland, monsoon forest and claypan (Philpott and Smyth, 1967; Watts, 1972; Reid and Gillen, 1988; Menkhorst and Woinarski, 1992; Woinarski et al., 1992). The first two records of Forrest's Mouse in New South Wales came from Fowlers Gap, 110km north of Broken Hill (Figure 1) (Morton, 1974). The area has an annual rainfall of around 200mm and the animals were caught on a flat, red, treeless plain dominated by Bladder Saltbush (Atriplex vesicaria) (Morton, 1974). Degradation by sheep had led to invasion by several species of Bassia (Milthorpe, 1972). Patches of Black Bluebush (Kochia pyramidata) and other species of saltbush and grasses also occurred. These captures occurred at a time of considerable plant growth in a high rainfall year (Morton, 1974). The recent record of Forrest’s Mouse from near Wanaaring was from sandy chenopod dunes and the Wilcannia specimen was obtained from an area of Bladder Saltbush (Atriplex vesicaria) on cracking soils (pers. comm. M. Cooper). This later record is the first occasion that the species has been collected from a riverine plains environment in New South Wales.

Read (1984) reports that most animals captured more recently at Fowlers Gap (twenty-one of twenty-four) were from tussock grassland (predominantly Mitchell Grass) on deep red massive loam. The remaining three individuals were captured in chenopod shrubland on deep yellow-red massive loams. A specimen from Sturt NP was captured on a flat-topped outcrop with a "silcrete type duricrust surface, cracking soil and shallow depression similar to gilgais" (Denny, 1975). Forrest's Mouse is frequently associated with areas where the soil has cracks (Dickman, 1993) which are believed to be used as shelter (Morton, 1974; Denny, 1975). The animals also inhabit gibber plains (Ayers et al., 1996).

Forrest’s Mouse burrows observed in Western Australia by Philpott and Smyth (1967) were 15cm deep and 37cm long, leading to a nest chamber lined with grass and daisy flowers. Several blind tunnels led from the nest chamber. Burrows observed by Dickman (1993) in the Simpson Desert either descended steeply to a single chamber around 80cm deep or ran horizontally 20cm under the surface for up to 2.5m with little branching. Finlayson (1941) reports that the animals dig shallow burrows at the base of spinifex tussocks. All observations of burrow-sharing have involved a lactating female and her offspring (Philpott and Smyth, 1967).

5 2.5 Threats and reasons for decline

Forrest's Mouse is a prey species for Cats (Felis catus) and Foxes (Vulpes vulpes), as well as Barn Owls (Tyto alba) (Morton et al., 1977; Dickman, 1993; Mahon, 1999). The species may also be adversely affected by stock grazing in the Western Division, particularly in favoured areas of tussock grassland. Grazing stock affect native rodents by removing shrubs and long grass which provide the rodents with food and shelter and by powdering and compacting the topsoil, making burrowing difficult or impossible (Dickman, 1993; Dickman et al., 1993). Dickman (1993) lists introduced competitors as a contributing factor in the decline of native rodents. It is therefore likely that exotic species such as the House Mouse (Mus domesticus) present a threat to Forrest’s Mouse.

Inappropriate fire regimes are listed by Dickman et al. (2000b) as a threatening process for native rodents in general, however the biology of Forrest’s Mouse is poorly known making it difficult to assess whether this species is adversely affected by current fire regimes. It is not known whether Forrest’s Mouse has specific requirements or preferences in terms of fire history. Further research is required to determine whether current fire management practices are inappropriate for the conservation of Forrest’s Mouse.

The use of 1080 (sodium monofluoroacetate) as a means of pest control is of serious concern to both Forrest’s Mouse and Sandy Inland Mouse (Dickman 1993). There is very little information published as to the impacts of 1080 poisoning in Forrest’s Mouse. The scant information available suggests that Forrest’s Mouse and Sandy Inland Mouse could be poisoned during some pest management operations.

In view of the reputation that 1080 has as a rodenticide, McIlroy (1982) undertook a study to determine the sensitivity of 14 Australian rodents. McIlroy (1982) suggested that most rodents would face a high degree of risk should they consume pellet baits (0.5mg 1080 g–1) or grain baits (0.4mg 1080 g-1) that are standard use in Pig control. It was found that all 14 species tested would have to consume less than 8% of their body weigh in pellet or grain baits to ingest an LD50. The same study found Dingo meat baits (0.014 mg 1080 g-1) to be safer as the dosage rates were sufficiently lower than pellet or grain baits. The Sandy Inland Mouse would need to consume more than their body weight of Dingo meat bait to ingest an LD50. This study had several limitations. The dosing was administered orally therefore not indicative of bait consumption and some rodents, including Forrest’s Mouse, were not examined.

Later studies demonstrated that under laboratory conditions both Forrest’s Mouse and Sandy Inland Mouse ate crackle baits and meat baits (Eastman and Calver, 1988; Calver et al., 1989). However variation occurred in the proportion of individuals that ate baits when alternative food sources were available and between animals from differing populations selected for the study (Calver et al., 1989).

6 Eastman and Calver (1988) supported McIlroy (1982) in suggesting meat baits to be relatively safe to most non-target mammal species however found crackle baits were of concern as they contain 0.85mg 1080 g-1. McIlroy (1986) has confirmed that the majority of non-target species tested ate pellet or grain baits. The implications for this are highlighted by an observation in Victoria where the use of pellet baits killed all adult residents in a marked population of the Silky Mouse (P. apodemoides) (McIlroy, 1982). Therefore some pest management practices in New South Wales that involve the use of 1080 poison may be detrimental to the Forrest’s Mouse and Sandy Inland Mouse.

The use of pesticides for the control of locusts may also represent a threat, however direct or secondary poisoning in native ground mammals has yet to be conclusively demonstrated. Forrest’s Mouse occurs in a habitat that is regularly utilised by the Australian Plague Locust (Choroicetes terminifera) particularly during plague events. At these times aerial spraying of pesticides may be used to control outbreaks. The habitat of the Sandy Inland Mouse is not preferred by any of the locust species for which authorities undertake aerial spraying to control. Therefore, it is considered unlikely that locust control measures will have an impact upon this species. The Australian Plague Locust Commission in conjunction with the University of Wollongong is undertaking a long-term study into the effect of locust pesticides on vertebrate fauna. Completion of this study will yield additional information that will allow for more informed management of this threat if necessary.

Overgrazing caused by introduced (domestic stock and feral Goats, Rabbits and Pigs) and native herbivores (kangaroos) may pose a threat to small mammals in western New South Wales. No study has adequately measured the overall impact of total grazing pressure on native vertebrate species. However, watering points act as focal points for herbivores and modification of the vegetation as a result of grazing is often greatest around these locations. Several comprehensive long-term grazing studies are currently under way in far-western New South Wales (D. Croft pers. comm.). Completion of these studies will greatly increase our knowledge and will allow for more effective abatement of grazing impacts.

A number of aspects of the biology and ecology of Forrest’s Mouse may render this species subject to decline. These include omnivorous diet, ground-dwelling or burrowing lifestyle, use of shrubland habitat, presence in arid or semi-arid regions (Lunney et al., 1997) and small body weight (Dickman et al., 2000b).

7 3.0 Species Information – Sandy Inland Mouse

3.1 Current Conservation Status

Sandy Inland Mouse (Pseudomys hermannsburgensis) is listed as vulnerable in Schedule 2 of the Threatened Species Conservation Act 1995 (TSC Act). The species is not listed in the Commonwealth Environment Protection and Biodiversity Act 1999 (ESP Act) or in the IUCN (1996) Red List of Threatened Species.

Dickman et al. (1993) state that the distribution and abundance of the species in western New South Wales is moderate, and the survival prognosis in the region is listed as vulnerable. Dickman (1993) describes the status of the Sandy Inland Mouse in New South Wales as rare. At the national level, distribution, abundance and survival prognoses are all considered secure (Dickman et al., 1993). However, in studying rodent populations in Queensland Dickman et al. (2000) found that neither abundance nor size of geographical range provides immunity from decline. They therefore contend that even large, widely distributed populations may be susceptible to decline.

3.2 Description

3.2.1 Taxonomy

Scientific Nomenclature: Pseudomys hermannsburgensis (Waite, 1896) Family: Muridae Common Name: Sandy Inland Mouse Recent Synonyms: Mus hermannsburgensis Leggadina hermannsburgensis Leggadina hermannsburgensis brazenori

The species was first described by Waite (1896) as Mus hermannsburgensis. It was subsequently assigned to the Genus Pseudomys by Troughton (1932). Leggadina hermannsburgensis and L. hermannsburgensis brazenori Troughton, 1937 are synonyms of Pseudomys hermannsburgensis (Bannister et al., 1988). Prior to its redescription by Kitchener et al. (1984), Pseudomys bolami was considered to be a subspecies of P. hermannsburgensis.

3.2.2 General

The Sandy Inland Mouse is superficially similar to the introduced House Mouse (Mus domesticus) but is more slender, has relatively long ears and tail and more prominent eyes. It lacks the notched incisors and distinctive musty odour of M. domesticus. The two species may also be distinguished by the pattern of the footpads (Cooper, 1993).

The Sandy Inland Mouse is greyish-brown to sandy-brown above and off-white below and is also similar to Bolam’s Mouse (P. bolami). It has smaller ears and 8 hind feet than Bolam’s Mouse, and the tail is shorter and less heavily furred (Kitchener et al., 1984). Adults weigh 9-15g, head and body length is 55-80mm and tail length 70-90mm (Dickman, 1993).

3.3 Distribution

The distribution of the Sandy Inland Mouse is described by Ayers et al. (1996) as wide but very sparse in arid and semi-arid zones of central, southern and western Australia. Within New South Wales, it is known only from the far north and west (Figure 2). It has been recorded in Sturt NP, at Fowlers Gap Station north of Broken Hill, and as subfossil remains in Mutawintji NP (Ellis, in press). The species has also been found near Kajuligah Nature Reserve north of Ivanhoe (prior to 1870) (Dickman, 1993), in the Enngonia area north-east of Bourke, in a construction trench south-east of Sturt NP (Ayers et al., 1996), and at several locations in the Tibooburra area (Australian Museum records). The NPWS Atlas of New South Wales Wildlife also contains one record of Sandy Inland Mouse from just outside the eastern boundary of Mutawintji NP.

The Sandy Inland Mouse is also present in Queensland, South Australia, the Northern Territory and Western Australia (Stanger et al., 1998). Dickman (1993) states that it probably does not occur much below 33º South, although past confusion with P. bolami leads to some uncertainty.

3.4 Ecology

3.4.1 Life cycle

The Sandy Inland Mouse is a nocturnal species and shelters in burrows during the day (Ayers et al., 1996). Burrows, which are up to 50cm deep and 1m long, are often constructed at the base of shrubs and small trees (Dickman, 1993). Multiple burrow occupancy is common and as many as 22 individuals have been found in a burrow (Watts and Aslin, 1981). Large congregations such as this occur only outside breeding periods (Watts and Aslin, 1981; Ayers et al., 1996). Small groups of four or five individuals are found during breeding (Ayers et al., 1996). Finlayson (1941) reports finding Sandy Inland Mouse in burrows with Knob-tailed Geckos. Spencer (1896) reported rings of stones around burrow entrances. However, these may have been the burrows of Pebble-mound Mice (C. R. Dickman, pers. comm.). Philpott and Smyth (1967) state that the burrows are characterised by the lack of a soil mound at the entrance.

Predavec (1994b) found that breeding is opportunistic rather than seasonal, and Breed (1990) observed pregnant females in all months except January, although only three individuals were sampled in that month. Breeding is triggered by rainfall (Predavec, 1994b), although some breeding has been observed in early summer during arid conditions, suggesting a combination of seasonal and opportunistic breeding strategies (Breed, 1990).

9 Gestation in Sandy Inland Mouse is 29-34 days and litter size ranges from 1-6 with an average of 3. The oestrus cycle lasts 7-13 days and the female re-enters oestrus following the weaning of a litter (Watts, 1979; 1982; Dickman, 1993). At birth, the young are naked and weigh 2g. Independence is reached at 30 days and reproductive maturity at 3 months. Breeding females are intolerant of other adults (Watts and Aslin, 1981).

Individuals are highly mobile, having been observed to cover linear distances of up to 14km, and it is likely that still longer movements occur (Dickman et al., 1995). Long-range movement has been observed in both sexes and all age groups, with no relationship to reproductive condition. In western Queensland and Western Australia, movements were found to occur more frequently during and after rain, with the animals moving towards areas of higher rainfall (Dickman et al., 1995). Southgate and Masters (1996), however, did not observe long-range movements in Central Australia, and report a high rate of recaptures, suggesting a sedentary population.

Figure 2: The distribution of the Sandy Inland Mouse in NSW.

Early maturity, aseasonal reproduction and long-range movement allow localised increases in numbers during periods of favourable conditions (Dickman, 1993). Consequently the Sandy Inland Mouse has been classified by Newsome and Corbett 10 (1975) as an r-strategist (a species whose breeding strategy involves rapid reproduction). Marked fluctuations in population sizes have been observed. For example, Southgate and Masters (1996) report a tenfold variation in numbers in Central Australia over a six-year study period, with density peaking at five individuals per hectare. In western Queensland, Predavec (1994b) observed 40-fold fluctuations in population sizes.

Irruptions are correlated with rainfall indices and seed availability, although various time lags have been reported (e.g. Predavec, 1994b; Southgate and Masters, 1996; Dickman et al., 1999). Southgate and Masters (1996) found that the highest correlation between abundance and rainfall index occurred with a 15-month time lag. Dickman et al. (1999) report that populations irrupt 3-10 months after rain. Irruptions observed in these studies were attributed principally to in situ reproduction and partly to immigration. This is in concurrence with the findings of Reid et al. (1993) and Predavec (1994b). Populations may be limited by seed availability (Predavec, 1994b), and irruptions may also be dampened by lower fecundity or increased predation (Southgate and Masters, 1996).

Dickman (1993) stated that the availability of food is likely to be the principal factor influencing populations of Sandy Inland Mouse. This is supported by Predavec (1994a), who showed experimentally that Sandy Inland Mouse increased in abundance when supplementary food was provided.

Predavec (1997) reports large variations in the basal metabolic rate (BMR) of Sandy Inland Mouse over time. Individuals have been observed to enter a state of reduced metabolic rate, or torpor. Possible reasons for lowering BMR in this manner include conservation of energy and water and increased longevity, all of which are adaptations for an unpredictable environment. BMR may be raised for reproduction, or variation may be due to changes in the ambient temperature (Predavec, 1997).

3.4.2 Diet

The Sandy Inland Mouse was thought until recently to be principally granivorous. Finlayson (1941) described the diet as comprising seeds, grass roots and small tubers. Further studies by Watts (1970; 1972), Watts and Morton (1983) and Murray and Dickman (1994a) found that the diet consisted of seed, plant material and invertebrates. In the latter study, the diet was found to vary seasonally. Seeds were important in summer and particularly in winter, but invertebrates constituted up to 60% of the diet in autumn (98% in the Tanami Desert during April 1986 (Murray et al., 1999)). Further, it was found that animals preferred invertebrates to plant material or seeds in laboratory trials. A small quantity of fungal hyphae was also observed in the diet, although these may have been ingested incidentally with plant root material.

The most common invertebrates found in the diet were spiders, although beetles and beetle larvae were also eaten. Plant material consumed included leaf, stem, flowers and roots. Types of seed eaten varied over time and included Grevillea, 11 Sida and Triodia. While seeds were common in the diet, plant material constituted less than 10% of the diet (by volume) in nearly 90% of individuals. Invertebrates were present in over 80% of stomachs (Murray and Dickman, 1994a). The Sandy Inland Mouse lacks adaptations to granivory shown in North American rodents such as cheek pouches, seed caching and excavation of buried seeds. Furthermore, analysis of the digestive tract suggests a diverse diet (Murray et al., 1995). Based on these observations, Murray and Dickman (1994a) contend that Sandy Inland Mouse is an omnivore. Murray and Dickman (1994b) suggest that an omnivorous diet, which includes a broad range of foods, is well suited to an unpredictable environment.

Standing water is not required for drinking, and individuals may select seeds on the basis of moisture content, as well as energy and nitrogen content. This allows the animals to gain sufficient water from their food (Dickman et al., 1995; S. Ricci, pers. comm.). The production of concentrated urine is a further adaptation to the arid zone (MacMillen et al., 1972).

Foraging is principally on the ground surface. However, the animals may climb up to 1m above the ground in shrubs or trees, or may obtain food underground (Dickman, 1993).

3.4.3 Habitat

The Sandy Inland Mouse is found in a range of habitats including hummock grasslands, Mulga flats, alluvial flats and gibber plains (Finlayson, 1941; Watts and Aslin 1974; Burbidge et al., 1976; Gibson and Cole, 1988). In the Tanami Desert, it is most often found on sand plains, dunes and gravelly rises and less frequently in riverine woodland, on rock outcrops and salty drainage systems (Gibson, 1986). In New South Wales, it has been found in open Mitchell Grass, Coolibah woodland and in a grassy creek bed (Denny, 1975; H. E. Parnaby, D. G. Read pers. comm. in Dickman, 1993). Individuals observed in the Simpson Desert, south-west Queensland showed a preference for dense cover of spinifex. Other cover used included Grevillea, Acacia, dead wood and small annuals and perennials (Murray and Dickman, 1994a).

Ayers et al. (1996) describe the habitat as comprising a variety of open vegetation types, including Coolibah or Acacia woodlands, tall open shrublands and hummock grasslands. Friable soils are preferred, mainly sands and sandy loams on plains and dunes, although the species has also been recorded on cracking earth soils (Watts and Aslin, 1981; Ayers et al., 1996). The Sandy Inland Mouse has recently been found in an area of Silver-leaved Ironbark and spinifex north-east of Bourke (see Figure 2). In this area, it was more abundant in country burnt four years previously than in country not burnt for fifteen years (Ayers et al., 1996).

12 3.5 Threats and reasons for decline

The greatest threat to populations of Sandy Inland Mouse is the modification of habitat as a result of grazing (Dickman, 1993). However, the species is also scarce in areas unaffected by grazing, suggesting that other factors may also depress populations. Predation by Barn Owls (Tyto alba), Cats (Felis catus) and Foxes (Vulpes vulpes) may pose a threat, particularly in areas where vegetation cover is reduced by grazing (Dickman, 1993). Remains of the Sandy Inland Mouse were frequently found in Barn Owl pellets from western Queensland (Morton et al., 1977) and have been found in Cat and Fox scats from the same area (Mahon, 1999). The Sandy Inland Mouse has also been recorded in the diet of Cats in the Flinders Ranges of South Australia (Holden, 1999).

The Sandy Inland Mouse does not appear to benefit from the introduction of standing water, whereas birds and introduced mammals may benefit from artificial water points. By giving an advantage to potential predators and competitors, artificial water points may represent an indirect threat to arid-zone rodents such as Sandy Inland Mouse (Kotler et al., 1998). Altered fire regimes may also be deleterious, as the species may have a preference for particular stages of post-fire succession (Ayers et al., 1996).

The use of 1080 (sodium monofluoroacetate) as a means of pest control is of serious concern to the Sandy Inland Mouse (Dickman, 1993). See Section 2.5 for discussion on effects of 1080 baiting on the Sandy Inland Mouse. Also discussed, is the likely impact of pesticides used in the control of locust. Due to differences in habitat preference between the Sandy Inland Mouse and locust species that plague, the use of pesticides for these species is not considered a significant threat.

Lunney et al. (1997) list various ecological attributes which are associated with decline in the fauna of New South Wales. A number of these are characteristic of the biology of Sandy Inland Mouse. These include omnivorous diet, presence in arid or semi-arid regions, and ground-dwelling or subterranean lifestyle. The small body weight of the Sandy Inland Mouse may also render the species more susceptible to decline (Dickman et al., 2000b).

4 Relevant Legislation

4.1 Threatened Species Conservation Act 1995

Forrest’s Mouse and Sandy Inland Mouse are listed on Schedule 2 of the Threatened Species Conservation Act 1995 (TSC Act) as vulnerable. It is an offence to harm, pick or damage the habitat of a threatened species unless the damage is the result of activities which have been licensed under Section 91 of the TSC Act, or have otherwise gained approval under the Environmental Planning and Assessment Act 1979.

13 4.2 Commonwealth Environment Protection and Biodiversity Conservation Act 1999

Forrest’s Mouse and the Sandy Inland Mouse are not listed in the Commonwealth Environment Protection and Biodiversity Conservation Act (EPBC Act) 1999. The EPBC Act regulates actions that may result in a significant impact on nationally listed threatened species and ecological communities. It is an offence to undertake any such actions in areas under State or Territory jurisdiction, as well as on Commonwealth-owned areas, without obtaining prior approval from the Commonwealth Environment Minister. As well as regulating Commonwealth agencies and areas, the EPBC Act will govern matters of National Environmental Significance such as nationally threatened flora and fauna, World Heritage properties, protected wetlands, internationally protected migratory species and nuclear actions.

4.3 National Parks and Wildlife Act 1974

Forrest’s Mouse and Sandy Inland Mouse have been recorded in Sturt National Park, an area gazetted under the National Parks and Wildlife Act 1974 (NPWS Act) and in the care and management of the NSW National Parks and Wildlife Service.

4.4 Environmental Planning and Assessment Act 1979

Land use and development on leasehold land in New South Wales is subject to evaluation in accordance with the Environmental Planning and Assessment Act 1979 (EP&A Act). Threatened species are to be taken into account by consent authorities when they are considering development applications under Part 4, and by determining authorities undertaking or approving activities under Part 5 of the Act. Under the Western Lands Act 1901 the Department of Land and Water Conservation is the determining authority for activities on Western lands leases.

4.5 Native Vegetation Conservation Act 1997

The clearing of vegetation in New South Wales is subject to consent from the Department of Land and Water Conservation in accordance with the Native Vegetation Conservation Act 1998. The Act is integrated with the Environmental Planning and Assessment Act 1979, and requires that threatened species are taken into account by the consent authority when considering clearing applications under Part 4 of the EP&A Act.

14 5 Management Issues

5.1 Scientific and taxonomic value

Leggadina is a genus of considerable taxonomic interest. The precise relationships between the species of this genus, and between Leggadina and other genera, are not clearly understood. Forrest’s Mouse is therefore of high significance in the study of taxonomy and convergent evolution.

The taxonomy of the Sandy Inland Mouse, in particular its similarity to Bolam’s Mouse (P. bolami) (Kitchener et al., 1984) render this a species of considerable scientific interest. The differences in geographic distribution and breeding strategies between these similar species provide a potential insight into the adaptations of Australian rodents to arid, semi-arid and mesic (moderately moist) habitats.

5.2 Biodiversity value

The native rodent fauna of Western New South Wales has experienced severe decline since European settlement. Of the seventeen native rodents which have been recorded in the arid zone of Western New South Wales, eleven are now regionally extinct, three (including Sandy Inland Mouse) are rare, two (including Forrest’s Mouse) are sparse, and one is common in limited habitat (Dickman, 1993). In view of this decline, the biodiversity value of the remaining species is considerable.

5.3 Social and economic considerations

Many of the records of Forrest’s Mouse and Sandy Inland Mouse in New South Wales are from areas that are currently managed for the purposes of wildlife conservation and research. No significant social or economic consequences are likely to arise from recovery actions undertaken in these areas. 15 Where these animals occur on areas of leasehold land, overgrazing may present a threat to the species. This recovery plan does not propose the regulation of grazing practices. Rather, when populations of these species are located on leasehold land, a cooperative approach to management is advocated. This fact, as well as the limited known occurrence of both species at present, suggests that the social and economic costs to landholders will be minimal.

Control of feral predators may also be required in areas where Forrest’s Mouse or Sandy Inland Mouse is located. However, such control is common practice and beneficial on grazing land and, as such, the targeting of feral predator control for the conservation of Forrest’s Mouse and Sandy Inland Mouse may represent a minimal economic cost to landholders.

Native fauna and habitat provide aesthetic value for members of the public with an interest in the natural environment. Distribution of information about Forrest’s Mouse and the Sandy Inland Mouse has the potential to increase public awareness of threatened species, and rodents in particular.

6 Previous Actions Undertaken

6.1 Reduction of grazing pressure

In Sturt National Park, regeneration of plant communities has been undertaken by controlling rabbits and closing down watering points to reduce the intensity of grazing by kangaroos (NSW NPWS, 1996). Rabbits have been controlled in the area by the introduction of rabbit calicivirus. Feral pigs and goats are occasionally encountered in the park, and are removed by shooting (L. Kogge, pers. comm.).

At Fowlers Gap, rabbit control has been implemented, including warren ripping and the introduction of rabbit calicivirus and myxomatosis (T. J. Dawson, pers. comm.). Feral goats are rounded up and removed when large groups (around 30 or more) are encountered. Although Fowlers Gap is an operational sheep station, light stocking levels are maintained and stock is excluded from some areas (T. J. Dawson, pers. comm.).

6.2 Control of feral predators

A program of fox control is currently in operation within Sturt National Park, with bait stations established along approximately 200km of roads. Baiting is usually conducted between autumn and spring and the bait take is monitored as an index of fox density. Some trapping of feral cats has also been undertaken in conjunction with the fox baiting program (L. Kogge, pers. comm.).

16 Control of feral cats and foxes has not been undertaken at Fowlers Gap, although a postgraduate study has been conducted on the ecology of foxes in the area (T. J. Dawson, pers. comm). NSW NPWS has acquired Beulah Station, north-west of Enngonia, where limited pest management is being undertaken (R. Enke, pers. comm.). Pest control effort at Beulah is likely to increase when the property is gazetted as a reserve under the NPW Act.

6.3 Fauna surveys

Several short-term biodiversity studies or long-term ecological research projects have been conducted on the distribution of Forrest’s Mouse and the Sandy Inland Mouse in NSW. These projects have provided most of the current knowledge of Forrest’s Mouse and the Sandy Inland Mouse in NSW to date.

The Australian Museum undertook an extensive survey for vertebrate fauna in and around Sturt National Park in 1995 and obtained numerous records of both Forrest’s Mouse and the Sandy Inland Mouse. The Australian Herpetological Society has coordinated an extensive survey for herpetofauna in Sturt National Park with annual surveys that commenced in 1997. As well as yielding valuable new information on the threatened reptiles in Sturt National Park they have also obtained records for both Forrest’s Mouse and Sandy Inland Mouse (Gerry Swan, pers. comm.). A program of pitfall trapping has recently been commenced in Sturt National Park as part of a University of NSW postgraduate study. The study aims to determine the effects of artificial watering points on biodiversity. Trapping commenced in 1999 and has yielded captures of Forrest’s Mouse and Sandy Inland Mouse (R. Montague-Drake, pers. comm.). This project has been partially funded by the NSW NPWS.

Sampling of small mammal populations has been conducted at Fowlers Gap Research Station for many years. The Station is managed by the University of NSW and used for teaching purposes, and for honours and postgraduate studies. No system of ongoing sampling exists for Forrest’s Mouse or Sandy Inland Mouse, however, the species are occasionally detected during other studies (T. J. Dawson, pers. comm.).

The Rangelands Liaison Group sampled an area east of Enngonia (north-east of Bourke), known as Ledknappers Spinifex, in 1993. This survey yielded several records of the Sandy Inland Mouse from areas of spinifex. A notable feature of this survey was their ability to compare burnt and unburnt areas of spinifex showing the species had a successional post-fire usage of the spinifex. The NSW NPWS has recently acquired Beulah Station, east of Enngonia, which was one of the locations from which the Sandy Inland Mouse has been recorded.

Surveys undertaken as part of the Woody Weeds Project, funded by West 2000, occurred north of Ivanhoe, west of Cobar and between Louth and Wanaaring. The surveys between Louth and Wanaaring yielded two juvenile Forrest’s Mouse on

17 private lands (D. Ayers, pers. comm.). The most recent survey activity undertaken in north western New South Wales has been the Darling Riverine Plains Biodiversity Study undertaken by NSW NPWS. This study sampled approximately sixty sites within the catchment, concentrating on the east in the areas of higher agricultural development. Surveys in the Wilcannia and Tilpa areas yielded one capture of Forrest’s Mouse. These two studies indicate that a population of Forrest’s Mouse exists in the Paroo and represents a substantial range extension for the species in New South Wales.

7 Species Ability to Recover

It is difficult to assess the ability of Forrest’s Mouse to recover as there is uncertainty as to whether the species has declined significantly in range (Reid and Morton, 1995). The population biology of the species is also poorly understood, although Forrest’s Mouse is thought to occur naturally at low densities (Dickman, 1993; Finlayson, 1941; 1961; Morton, 1974; Watts and Aslin, 1981). Given that predation by feral animals and overgrazing by herbivores are considered to be threatening processes, there may be potential for the species to recover if these threats are alleviated.

Several biological attributes of the Sandy Inland Mouse suggest that it may have the ability to recover significantly in terms of abundance and distribution. Rapid, aseasonal reproduction and the ability to disperse over long distances allow marked increases in populations (Dickman, 1993; Predavec, 1994a; 1994b; Dickman et al, 1995; Southgate and Masters, 1996). As populations of the Sandy Inland Mouse are likely to be limited by threatening processes such as overgrazing and feral predators, the species may have considerable ability to recover if these threats are alleviated.

8 Recovery Objectives and Performance Criteria

8.1 Objectives of the Recovery Plan

The overall objective of this Recovery Plan is to secure extant populations of Forrest’s Mouse and Sandy Inland Mouse in New South Wales through appropriate management.

Specific objectives are to:

1) monitor the populations of Forrest’s Mouse and Sandy Inland Mouse to determine their status;

18 2) protect the populations of Forrest’s Mouse and Sandy Inland Mouse from the threatening processes of overgrazing by herbivores and predation by feral animals;

3) survey areas of potential habitat for new extant populations of Forrest’s Mouse or Sandy Inland Mouse;

5) improve current knowledge of the biology and ecology of both Forrest’s Mouse and the Sandy Inland Mouse.

8.2 Recovery performance criteria

Recovery criteria are that:

1) an ongoing system of trapping and release is implemented on and off Service estate to monitor the populations of small mammals;

2) the control and monitoring of grazing herbivores and feral predators is continued on Service estate;

4) areas of potential habitat for Forrest’s Mouse and Sandy Inland Mouse are identified and surveyed for new extant populations of the species;

5) a greater understanding of the biology and ecology of Forrest’s Mouse and Sandy Inland Mouse in NSW is achieved through appropriate research.

9 Recovery Actions

9.1 Action 1 - Monitor known extant populations

Small mammal populations in Sturt National Park and Fowlers Gap should continue to be monitored. The University of NSW expects to continue the existing regime of ecological research undertaken in Sturt National Park and Fowlers Gap, until 2007 and most likely beyond (D. Croft pers. comm.). Current research programs are yielding sufficient captures to allow for limited monitoring of the species at both locations. Additional funding from the Recovery Plan for the monitoring of both the Sturt National Park and Fowlers Gap populations is not 19 expected to be required. Logistical support for the research at both locations could be provided by NPWS where appropriate.

Outcome The size and viability of these populations can be assessed, leading to a better understanding of the conservation status of these species. Further insight can be gained into the ecology of Forrest’s Mouse.

Action 1 2002/2003 2003/2004 2004/2005 2005/2006 2006/2007 Monitoring Nil Nil Nil Nil Nil TOTAL Nil*

* University of NSW expects to continue the small mammal surveys at Sturt National Park and Fowler’s Gap. Therefore, no additional funding associated with the Recovery Plan is required.

Agency responsible for implementation NSW National Parks and Wildlife Service.

Funding source Not required.

9.2 Action 2 – Development of effective monitoring programs

Effective monitoring of both species will be a significant component of this Recovery Plan. Monitoring will be necessary to determine the species true status in western NSW and any associated trend. Monitoring over the long-term will allow for an assessment of the effectiveness of the Actions contained within this Recovery Plan.

Monitoring of arid zone rodents is difficult as they generally occur at very low densities. Trap success for a specific species can often be a single animal per thousand trap nights. Programs aimed at targeting species in the arid zone generally require a very intensive effort in order to yield sufficient results. Therefore, a monitoring program aimed at determining status and trend for Forrest’s Mouse and Sandy Inland Mouse in western NSW is likely to be expensive and time consuming. It is important that a thorough review of existing information is undertaken in order to design and implement an effective and cost efficient program.

This review will aim to establish the minimum effort and most appropriate methods required to obtain sufficient information to allow some assessment of population status and trends. The review should also identify the best locations in NSW to undertake such monitoring.

20 Outcome The review will enable NSW NPWS to undertake efficient and cost effective monitoring for both species which is critical in determining the effectiveness of recovery actions.

Action 2 2002/2003 2003/2004 2004/2005 2005/2006 2006/2007 Review $2 500 Nil Nil Nil Nil TOTAL $2 500

Agency responsible for implementation NSW National Parks and Wildlife Service.

Funding source NSW National Parks and Wildlife Service.

9.3 Action 3 – Targeted surveys for Forrest’s Mouse and Sandy Inland Mouse in western New South Wales

There are very few records of Forrest’s Mouse and Sandy Inland Mouse in New South Wales. The NSW NPWS Wildlife Atlas has only seven records of Forrest’s Mouse and twelve records of Sandy Inland Mouse. These records are clustered in several areas, which are the location of major biodiversity surveys or research in the west of the State. These locations are Sturt National Park, Fowlers Gap, Mutwintji National Park and Culgoa floodplains.

The small number of records has resulted in a limited comprehension of the species habitat preferences and the clustered nature of the locations have resulted in a poor understanding of the species distribution at a State level.

Further surveys are therefore required to provide a more accurate picture of the species distribution at a local level and across the State, as well as provide additional information on the ecology of both species. An additional benefit of targeted surveys in the first years of the Plan will be to better inform the development of a long-term monitoring program.

Targeted surveys should be conducted on the Paroo and Culgoa floodplains as well as Mutwintji, Gundabooka and Peery National Parks. Targeted surveys would require methods such as pitfall and Elliott trapping and analysis of predator scats and raptor pellets (Dickman, 1993). Surveys should span an extended period (Dickman (1993) suggests a minimum of ten days) and should be carried out two or three times over a period of one year, preferably in spring, late summer and winter (Dickman, 1993). Where possible, sampling should include mesic areas such as soaks which are likely to act as refuges for rodents (Dickman, 1993). Where possible, additional information should be gathered to increase our knowledge of these species in NSW.

21 Outcome Improved knowledge of the distribution and ecology of Forrest’s Mouse and Sandy Inland Mouse in New South Wales. Improved knowledge of other ground fauna, including other threatened species, in areas sampled.

Action 3 2002/2003 2003/2004 2004/2005 2005/2006 2006/2007 Survey $38 700 $37 400 $18 700 Nil Nil TOTAL $94 800

Agency responsible for implementation NSW National Parks and Wildlife Service.

Funding source NSW National Parks and Wildlife Service.

9.4 Action 4 – Maintain and review the Service’s ongoing program of feral animal management

Introduced herbivores and predators should continue to be controlled on Service estate. If possible, control efforts should be conducted in collaboration with neighbouring properties.

The use of 1080 poison to control pest species has been identified as potentially detrimental to both Forrest’s Mouse and Sandy Inland Mouse. Pellet and grain baits containing 1080 used in the control of feral pigs has been shown to be particularly lethal to rodents. The use of 1080 pig baits should be avoided until such time as a protocol is developed that will minimise any adverse impacts associated with the technique.

The program of goat control on Service estate should continue. The general practice of letting a single contract per estate area should be reviewed and its effectiveness to achieve long term suppression of goat numbers determined. The review should also compare alternative strategies such as ‘competitive harvesting’ through the letting of multiple contracts simultaneously, target driven contracts, aerial culling, fencing or a combination of methods.

Small rodents, such as Forrest’s Mouse and Sandy Inland Mouse, are likely to be at greater risk from predation by cats than by foxes (Mahon, 1999). Several studies have shown that foxes may limit cat abundance and distribution across the landscape (Risbey & Calver 1998; Molsher 1999; Risbey et al. 1999). Further studies involving fox specific control have indicated that a reduction of fox numbers may result in an increase in cat numbers and their distribution in the landscape (Molsher 1999; Risbey et al. 2000). This fact raises serious concerns about the impact of feral predator control programs that target foxes but not cats. Such programs may in fact be detrimental to Forrest’s Mouse and Sandy Inland Mouse. 22 Several studies have demonstrated that rabbits provide a substantial proportion of the diet of foxes and cats (Risbey et al. 1999; Mahon 1999). An increase in rabbit numbers has the potential to result in an increase in fox and cat numbers and intensify competitive effects during and particularly after rabbit eruptions (Pech et al. 1992; Williams et al. 1995).

Integrated pest management strategies should be developed for the Service estate where it does not currently exist. This technique involves the effective control or removal of foxes, cats and rabbits to avoid compensatory responses by one or several of the control species which in turn may not alleviate pressure on native species or at worst increase pressure on specific groups.

A management technique targeting foxes but not cats (e.g. baiting with Foxoff â or dried meat baits) is therefore not considered sufficient for the protection of Forrest’s Mouse or Sandy Inland Mouse. At present, techniques are not available in New South Wales which provide effective, long-term reductions in the density of feral cats without an intensive, continuous effort. Reductions may be achieved with an intensive program of shooting and trapping (e.g. Mahon 1999), however, the labour costs of such a strategy can be substantial. Intensive periods of shooting and trapping cats following irruptions of prey species may provide a viable alternative to a continuous effort (P. S. Mahon, pers. comm.). Novel control techniques such as new bait types should be evaluated as they arise and utilised where appropriate. Monitoring of control species dynamics is an essential part of the integrated pest management approach. Therefore regular surveys should be conducted to ascertain the density of feral animals in the area in order to assess the efficacy of the control measures used.

Feral animal management should also be considered a priority in areas where Forrest’s Mouse or Sandy Inland Mouse is located, both on and off conservation estate. The provision of assistance to private landholders to conduct appropriate feral animal management on their properties should also be considered when such practices are intended for the conservation of Forrest’s Mouse or Sandy Inland Mouse.

Outcome This action should maintain a measurable reduction in the density of introduced predators and herbivores on Service estate. This will prevent populations of Forrest’s Mouse and Sandy Inland Mouse from being suppressed by predation or habitat destruction, thereby increasing the likelihood of recovery.

Action 4 2002/2003 2003/2004 2004/2005 2005/2006 2006/2007 Pest Nil Nil Nil Nil Nil Control TOTAL Nil*

23 * A program of feral animal management currently occurs in National Parks and in most Nature Reserves. Additional costs associated with feral animal control on Service estate will only be necessary if changes are required to achieve effective integrated pest management. Provision of assistance to landholders carrying out pest control would also be additional.

Agency responsible for implementation NSW National Parks and Wildlife Service.

Funding source NSW National Parks and Wildlife Service.

9.5 Action 5 - Involve and inform relevant stakeholders

Populations of Forrest’s Mouse and Sandy Inland Mouse are protected from the threats of feral animals and grazing by stock in Sturt National Park, and a light stocking regime is implemented at Fowlers Gap. However, many of the recent records of these species come from leasehold land. Furthermore, the Sandy Inland Mouse is known to move over large distances in order to track localised rainfall events (Dickman et al., 1995). Forrest’s Mouse is also known to travel large distances and is thought not to occupy a fixed home range (C. R. Dickman, unpublished data). For this reason, it is important that areas of non-Service estate are also managed for the protection of these species.

A cooperative management approach is advocated to encourage appropriate management on these properties. This will require the provision of information to, and consultation with land managers. Leaflets containing information about Forrest’s Mouse and Sandy Inland Mouse will be prepared for distribution to landholders and other relevant stakeholders.

Outcome Increased awareness of Forrest’s Mouse and Sandy Inland Mouse, and support for conservation of the species within New South Wales.

Action 5 2002/2003 2003/2004 2004/2005 2005/2006 2006/2007 Survey $1 650 Nil Nil Nil Nil TOTAL $1 650

Agency responsible for implementation NSW National Parks and Wildlife Service.

Funding source NSW National Parks and Wildlife Service.

24 9.6 Action 6 - Promote further biological and ecological research

There is very little known about the specific requirements of both species in NSW. Thus, further research must be undertaken to acquire the information necessary to achieve successful management and recovery of both species. Where possible, the collection of information that will provide an insight into the ecology and biology of both the species should be integrated with any of the monitoring programs undertaken as part of this Recovery Plan. These monitoring programs will be designed to provide presence and abundance data and will have only a limited ability to provide ecological and biological data. In order to gather such information a research project, designed specifically to gather additional ecological and biological data should be undertaken and when possible integrated with additional monitoring programs. This project will be supported by financial assistance to the institution undertaking the work in the form of a scholarship.

Outcome A specific research project will increase the our current understanding of the biology and ecology of both species in NSW, which in turn will allow for more effective management of the species.

Action 6 2002/2003 2003/2004 2004/2005 2005/2006 2006/2007 Research $5 000 $5 000 $5 000 $5 000 $5 000 TOTAL $25 000

Agency responsible for implementation NSW National Parks and Wildlife Service.

Funding source NSW National Parks and Wildlife Service.

10 Alternative Management Strategies

10.1 Management of fire regime

As inappropriate fire regimes are listed by Dickman et al. (2000a, 2000b) as a threatening process for native rodents, one management alternative might be to manage the fire regime in Sturt National Park for the conservation of Forrest’s Mouse and Sandy Inland Mouse. However, there is insufficient knowledge to formulate specific fire management actions for the conservation of these species. The requirements of the species in terms of fire regime should be determined, and management recommendations made on this basis.

The sparse covering of vegetation prevalent during most years in Sturt National Park is not conducive to fire. Wildfire is infrequent, and fires are unlikely to spread 25 or become intense under normal conditions. The risk of fire to life and property is therefore considered to be low (NSW NPWS, 1996). For these reasons, a Fire Management Plan has not previously been prepared for Sturt National Park, however fire management is addressed in the Plan of Management (NSW NPWS, 1996).

10.2 No management action taken

The distribution of Forrest’s Mouse is not known to have declined significantly since European settlement, and both Forrest’s Mouse and Sandy Inland Mouse are widespread in other states. In view of this, an alternative strategy is not to undertake any management of these species.

This approach is not considered appropriate, as a stated objective of the TSC Act is the conservation of biodiversity and the recovery of threatened species within New South Wales. Also, a number of ecological attributes of Forrest’s Mouse and Sandy Inland Mouse, combined with existing threats, may render them susceptible to further decline. Finally, the lack of detailed knowledge of Forrest’s Mouse needs to be addressed in order to ensure its conservation.

11 Implementation

Table 1 allocates responsibility for the implementation of recovery actions specified in this plan to relevant government agencies for the period 2002 – 2007.

Table 1: Implementation schedule

Section Description Responsibility for Timeframe Cost Priority implementation 9.1 Population monitoring and NPWS Ongoing Nil High research 9.2 Review of monitoring NPWS 2002/2003 $2 500 High 9.3 Targeted Survey NPWS 2002/2003 $94 800 High 9.4 Feral animal management NPWS Ongoing Nil High 9.5 Community Education NPWS 2002/2003 $1 650 High 9.6 Research NPWS Ongoing $25 000 High TOTAL $123 950

26 12 Preparation details

Alistair Glen and Matthew Chambers of the Threatened Species Unit, Western Directorate NPWS, prepared this Recovery Plan. Matt Cameron (NPWS-Western) edited the plan and contributed to the development of the recovery actions.

12.1 Date of last amendment

This document is the first recovery plan for Forrest’s Mouse and the Sandy Inland Mouse. No amendments to the plan have been made.

12.2 Review date

This recovery plan, and the conservation status of Forrest’s Mouse and Sandy Inland Mouse, will be reviewed within five years of the date of publication.

27 References Ashby, E. D. and Lunney, D. 1985. Review of policies, priorities and programmes for nature conservation in New South Wales. Paper No. 41. Rodents. New South Wales National Parks and Wildlife Service, Sydney. Unpublished report.

Ayers, D., Nash, S. and Baggett, K. 1996. Threatened species of Western New South Wales. New South Wales National Parks and Wildlife Service, Hurstville.

Ayers, D. and Wallace, G. 1997. Pipeline trenches: an under-utilised resource for finding fauna. pp. 349-357 in P. Hale and D. Lamb (eds) "Conservation Outside Nature Reserves." Centre for Conservation Biology, University of Queensland.

Bannister, J. L., Calaby, J. H., Dawson, L. J., King, J. K., Mahoney, J. A, McKay, G. M., Richardson, B. J., Ride, W. D. L. and Walton, D. W. 1988. Vol. 5 Mammalia. In D.W. Walton (ed) "Zoological Catalogue of Australia." Australian Government Publishing Service, Canberra.

Baverstock, P. R., Watts, C. H. S., Adams, M. and Cole, S. R. 1981. Genetical relationships among Australian rodents (Muridae). Australian Journal of Zoology (29) 289-303.

Breed, W. G. 1990. Comparative studies on the timing of reproduction and foetal number in six species of Australian conilurine rodents (Muridae: Hydromyinae). Journal of Zoology (221) 1-10.

Breed, W. G. 1995. Sandy Inland Mouse, Pseudomys hermannsburgensis Waite, 1896. pp. 604-605 in R. Strahan (ed) "The Mammals of Australia." Reed Books, Chatswood.

Brunner, H. and Wallis, L. R. 1986. Roles of predator scat analysis in Australian mammal research. Victorian Naturalist (103) 79-87.

Burbidge, A. A., McKenzie, N. L., Chapman, A. and Lampert, P. M. 1976. The wildlife of some existing and proposed reserves in the Great Victoria and Gibson Deserts, Western Australia. Wildlife Research Bulletin of Western Australia. (5) 1- 16.

Calver, M. C., King, D. R., Bradley, J. S., Gardner, J. L. and Martin, G. R. 1989. An assessment of the potential target specificity of 1080 predator baiting in Western Australia. Australian Wildlife Research (16) 625-638.

Cooper, N. K. 1993. Identification of Pseudomys chapmani, P. hermannsburgensis, P. delicatulus and Mus musculus using footpad patterns. Western Australian Naturalist (19) 69-73.

28 Cooper, N. K., Adams, M. and Aplin, K. Unpubl. Morphological and genetic studies of Leggadina confirm existence of two species. Unpublished seminar abstract.

Denny, M. J. S. 1975. Mammals of Sturt National Park, Tibooburra, New South Wales. Australian Zoologist (18) 179-195.

Dickman, C. R. 1993. The biology and management of native rodents of the arid zone in New South Wales. Species Management Report Number 12. New South Wales National Parks and Wildlife Service, Hurstville.

Dickman, C. R. Leung, L. K. - P. and Van Dyck, S. M., 2000a. Status, ecological attributes and conservation of native rodents in Queensland. Wildlife Research (27) 333-46.

Dickman, C. R., Lunney, D. and Matthews, A., 2000b. Ecological attributes and conservation of native rodents in New South Wales. Wildlife Research (27) 347-55.

Dickman, C. R., Mahon, P. S., Masters, P. and Gibson, D. F. 1999. Long-term dynamics of rodent populations in arid Australia: the influence of rainfall. Wildlife Research (26) 389-403.

Dickman, C. R., Predavec, M. and Downey, F. J. 1995. Long-range movements of small mammals in arid Australia: implications for land management. Journal of Arid Environments (31) 441-452.

Dickman, C. R., Pressey, R. L., Lim, L. and Parnaby, H. E. 1993. Mammals of particular conservation concern in the Western Division of New South Wales. Biological Conservation (65) 219-248.

Eastman, A. and Calver, M.C. 1988. Consumption of dingo baits by non-target fauna from the pastoral areas of Western Australia. Australian Rangelands Journal (10:2) 106-8.

Ellis, M. in press. Rodents of the Western Plains of New South Wales. Wildlife Research

Finlayson, H. H. 1941. On central Australian mammals Part II. The Muridae. Transactions of the Royal Society of South Australia (65) 215-232.

Finlayson, H. H. 1961. On central Australian mammals Part IV - The distribution and status of central Australian species. Records of the South Australian Museum (14) 141-191.

Gibson, D. F. 1986. A biological survey of the Tanami Desert in the Northern Territory. Conservation Commission of the N.T. Technical Report (30) 1-258.

29 Gibson, D. F. and Cole, J. R. 1988. A biological survey of the northern Simpson Desert. Conservation Commission of the N.T. Technical Report (40) 1-205.

Holden, C. 1999. The impact of rabbit calicivirus disease on fauna of the Flinders Ranges and implications for threat abatement programs. Report for the Flinders Ranges RCD Monitoring and Surveillance Program 1997 – 1999. National Parks and Wildlife, South Australia.

Kitchener, D. J., Adams, M. and Baverstock, P. 1984. Redescription of Pseudomys bolami Troughton, 1932 (Rodentia: Muridae). Australian Mammalogy (7) 149-159.

Kotler, B. P., Dickman, C. R. and Brown, J. S. 1998. The effects of water on patch use by two Simpson Desert granivores (Corvus coronoides and Pseudomys hermannsburgensis). Australian Journal of Zoology. (23) 574-578.

Lidicker, W. Z. and Brylski, P. V. 1987. The conilurine rodent radiation of Australia, analyzed on the basis of phallic morphology. Journal of Mammalogy (68) 617-641.

Lunney, D., Curtin, A. L., Fisher, D., Ayers, D. and Dickman, C. R. 1997. Ecological attributes of the threatened fauna of New South Wales. Pacific Conservation Biology. (3) 13-26.

MacMillen, R. E., Baudinette, R. V. and Lee, A. K. 1972. Water economy and energy metabolism of the Sandy Inland Mouse, Leggadina hermannsburgensis. Journal of Mammalogy (53) 529-539.

Mahon, P. S. 1999. Predation by feral cats and red foxes, and the dynamics of small mammal populations in arid Australia. Ph.D. Thesis, University of Sydney.

Mahony, J. 1988. p. 176 in D. W. Walton (ed) “Zoological Catalogue of Australia. Vol. 5 Mammalia. AGPS, Canberra.

McIlroy, J.C. 1982. The sensitivity of Australian Animals to 1080 poison IV: Native and introduced rodents. Australian Wildlife Research (9) 505-17.

McIlroy, J.C. 1986. The sensitivity of Australian Animals to 1080 poison IX: Comparisons between the major groups of animals, and the potential danger non- target species face from the 1080-poisoning campaigns. Australian Wildlife Research (13) 39-48.

Menkhorst, K. A. and Woinarski, J. C. Z. 1992. Distribution of mammals in monsoon rainforests of the Northern Territory. Wildlife Research (19) 295-316.

30 Milthorpe, P. L. 1972. Vegetation of the Fowlers Gap - Calindary area. In "Lands of the Fowlers Gap - Calindary Area, New South Wales." Fowlers Gap Research Series No. 4. University of New South Wales.

Molsher, R. 1999. The ecology of feral cats, Felix catus, in open forest in New South Wales: Interactions with food resources and foxes. Ph.D. Thesis, University of Sydney.

Moro, D., Campbell, N. J. H., Elphinstone, M. S. and Baverstock, P. R. 1998. The Thevenard Island mouse: historic and conservation implications from mitochondrial DNA sequence variation. Pacific Conservation Biology (4) 282-288.

Morton, S. R. 1974. First record of Forrest's Mouse Leggadina forresti (Thomas, 1906) in N.S.W. Victorian Naturalist (91) 92-94.

Morton, S. R., Happold, M., Lee, A. K. and MacMillen, R. E. 1977. The diet of the Barn Owl, Tyto alba, in south-western Queensland. Australian Wildlife Research (4) 91-97.

Morton, S. R. and Martin, A. A. 1979. Feeding ecology of the Barn Owl, Tyto alba, in arid southern Australia. Australian Wildlife Research (6) 191-204.

Murray, B. R. and Dickman, C. R. 1994a. Granivory and microhabitat use in Australian desert rodents: are seeds important? Oecologia (99) 216-225.

Murray, B. R. and Dickman, C. R. 1994b. Food preferences and seed selection in two species of Australian desert rodents. Wildlife Research (21) 647-655.

Murray, B. R., Dickman, C. R., Watts, C. H. S and Morton, S. R. 1999. The dietary ecology of Australian desert rodents. Wildlife Research (26) 421-438.

Murray, B. R., Hume, I. D. and Dickman, C. R. 1995. Digestive tract characteristics of the Spinifex Hopping-Mouse, Notomys alexis and the Sandy Inland Mouse, Pseudomys hermannsburgensis in relation to diet. Australian Mammalogy (18) 93- 97.

Newsome, A. E. and Corbett, L. K. 1975. Outbreaks of rodents in semi-arid and arid Australia - causes, preventions and evolutionary considerations. pp. 117-151 in I. Prakash and P.K. Ghosh (eds). "Rodents in Desert Environments." Junk, The Hague.

NSW National Parks and Wildlife Service. 1996. Sturt National Park Plan of Management. NSW NPWS, Tibooburra.

31 Pech, R.P., Sinclair, A.R.E., Newsome, A.E. & Catling, P.C. 1992. Limits to predator regulation of rabbits in Australia: Evidence from predator-removal experiments. Oecologia (89) 102-12.

Philpott, C. M. and Smyth, D. R. 1967. A contribution to our knowledge of some mammals from inland Australia. Transactions of the Royal Society of South Australia (91) 115-129.

Predavec, M. 1994a. Food Limitation and Demography in Australian Desert Rodents. Ph.D Thesis, University of Sydney.

Predavec, M. 1994b. Population dynamics and environmental changes during natural irruptions of Australian desert rodents. Wildlife Research (21) 569-582.

Predavec, M. 1997. Variable energy demands in Pseudomys hermannsburgensis: possible ecological consequences. Australian Journal of Zoology. (45) 85-94.

Read, D. G. 1984. Diet and habitat preference of Leggadina forresti (Rodentia: Muridae) in Western New South Wales. Australian Mammalogy (7) 215-217.

Read, D. G. 1995. Fauna survey in the floodplain of the Great Anabranch of the lower Darling River. Australian Zoologist (30) 57-64.

Reid, J. and Gillen, J. 1988. The Coongie Lakes Study. Consultancy to the Department of Environment and Planning, Adelaide. Unpublished. report.

Reid, J. R. W., Kerle, J. A. and Baker, L. 1993. Mammals. pp. 69-78 in J. R. W. Reid, J. A. Kerle and S.R. Morton (eds). "Kowari: Uluru Fauna." Australian National Parks and Wildlife Service, Canberra.

Reid, J. R. W. and Morton, S. R. 1995. Forrest's Mouse, Leggadina forresti (Thomas, 1906). pp. 555-556 in R. Strahan (ed) "The Mammals of Australia." Reed Books, Chatswood.

Ride, W. D. L. 1970. "A Guide to the Native Mammals of Australia." Oxford University Press, Melbourne.

Risbey, D. & Calver, M.C. 1998. Can current control strategies against introduced predators endanger native fauna? Proceedings of the 11th Vertebrate Pest Conference, Banbury, Western Australia.

Risbey, D.A., Calver, M.C. & Short, J. 1999. The impact of cats and foxes on the small vertebrate fauna of Heirisson Prong, Western Australia. I. Exploring potential impact using diet analysis. Wildlife Research (26) 621-30.

32 Risbey, D.A., Calver, M.C., Short, J., Bradley, J.S. & Wright, I.W. 2000. The impacts of cats and foxes on the small vertebrate fauna of Heirisson Prong, Western Australia. II. A field experiment. Wildlife Research (27) 223-35.

Saunders, G., Coman, B., Kinnear, J. and Braysher, M. 1995. Managing Vertebrate Pests: Foxes. Australian Government Publishing Service, Canberra.

Southgate, R. and Masters, P. 1996. Fluctuations of rodent populations in response to rainfall and fire in a central Australian hummock grassland dominated by Plectrachne schinzii. Wildlife Research (23) 289-303.

Spencer, W. B. 1896. Mammalia. pp. 1-52 in W. B. Spencer (ed). "Report on the Work of the Horn Scientific Expedition to Central Australia Pt.2 - Zoology." Melville, Mullen and Slade, Melbourne.

Stanger, M., Clayton, M., Schodde, R., Wombley, J. and Mason, I. 1998. "CSIRO List of Australian Vertebrates: A Reference with Conservation Status." CSIRO Publishing, Collingwood.

Tate, G. H. H. 1951. The rodents of Australia and New Guinea. Bulletin of the American Museum of Natural History (97) 187-430.

Thomas, O. 1906. On mammals from northern Australia presented to the National Museum by Sir. Wm. Ingram, Bt., and the Hon. John Forrest. Proceedings of the Zoological Society London (1906) 536-543.

Thomas, O. 1925. A new Australian mouse of the genus Pseudomys. Annals and Magazine of Natural History (9) 669-671.

Troughton, E. le G. 1932a. Australian furred animals, their past, present and future. Australian Zoologist (7) 173-193.

Troughton, E. le G. 1932b. On five new rats of the Genus Pseudomys. Records of the Australian Museum (18) 287-294.

Troughton, E. Le G., 1967. "Furred Animals of Australia." 9th edn. Angus and Robertson, Sydney.

Van Dyck, S. M. and Longmore, N. W. 1991. The mammal records. pp. 284-336 in G.J. Ingram and R. J. Raven (eds) "An Atlas of Queensland's Frogs, Reptiles, Birds and Mammals." Queensland Museum, Brisbane.

Waite, E. R. 1896. Muridae. pp. 393-409 in B. Spencer (ed) "Report on the Work of the Horn Scientific Expedition to Central Australia. Pt. 2 - Zoology." Melville, Mullen and Slade, Melbourne.

33 Watts, C. H. S. 1970. The foods eaten by some Australian desert rodents. South Australian Naturalist (44) 71-74.

Watts, C. H. S. 1972. Handbook of South Australian Rodents and Small Marsupials. Field Naturalists Society of South Australia, Adelaide.

Watts, C. H. S. 1976. Leggadina lakedownensis, a new species of murid rodent from North Queensland. Transactions of the Royal Society of South Australia (100) 105-108.

Watts, C. H. S. 1979. Reproductive parameters of some Australian rodents. Australian Zoologist (20) 305-311.

Watts, C. H. S. 1982. Australian hydromyine rodents: maintenance of captive colonies. pp 180-184 in D. D. Evans (ed). The Management of Australian Mammals in Captivity. Zoological Board of Victoria, Melbourne.

Watts, C. H. S. and Aslin, H. J. 1981. The Rodents of Australia. Angus and Robertson, Sydney.

Watts, C. H. S., Baverstock, P. R., Birrell, J. and Krieg, M. 1992. Phylogeny of the Australian rodents (Muridae): a molecular approach using microcomplement fixing of albumin. Australian Journal of Zoology. (40) 81-90.

Watts, C. H. S. and Morton, S. R. 1983. Notes on the diets of Mus musculus and Pseudomys hermannsburgensis (Rodentia: Muridae) in western Queensland. Australian Mammalogy (6) 81-82.

Williams, K., Parer, I., Coman, B.J., Burley, J. & Braysher, M.L. 1995. Managing Vertebrate Pests: Rabbits. Australian Government Printing Service, Canberra, Australia.

Woinarski, J. C. Z., Braithwaite, R. W., Menkhorst, K. A., Griffin, S, Fisher, A. and Preece, N. 1992. Gradient analysis of the distribution of mammals in stage III of Kakadu National Park, with a review of the distribution patterns of mammals across north-western Australia. Wildlife Research (19) 233-262.

34 43 Bridge Street Hurstville 2220 (02) 9585 6444