Cane toads

Assessing the impact of the cane toad, Bufo marinus, on the terrestrial fauna of the East Kimberley

Graham G. Thompson 1 and Philip C. Withers 2

1 Centre for Ecosystem Management, Edith Cowan University , Email: g.thompson@ecu.edu.au

2 Zoology, School of Animal Biology , University of Western Australia , Email: philip.withers@uwa.edu.au

Abstract

Cane toads (Bufo marinus) will invade the east Kimberley of Western Australia in the next few years and will have a devastating negative impact on the fauna and the ecosystems of region. Species identified as being at serious risk include: the Northern Quoll , Sand Goanna, Yellow Spotted Goanna, Spotted Tree Goanna and the Water Monitor. Many other species have been identified as ‘probably’ at risk, but we are unsure of their status because of a lack of information.

A plethora of investigations have assessed the impact of cane toads on single species or taxa but only two studies have assessed the impact on broad scale terrestrial ecosystem function. Both of these studies have serious limitations due to small sample sizes and an inability to accommodate normal temporal variations. We therefore have little understanding of the impact that cane toads have on ecosystem function, and on all but a few significant species. Our study will address this issue and will be the first study in Australia that comprehensively sets out to measure the impact of cane toads on ecosystem function on a regional scale. We also intend to investigate, and if feasible, implement a program that relocates goannas onto islands in Lake Argyle , to preserve aspects of the Kimberley biodiversity considered to be most at risk as a consequence of the cane toad invasion.

We propose a comprehensive four year survey program that will:

• establish quality baseline data on small vertebrate assemblages in the east Kimberley , to assess the long-term impact that cane toads have on these assemblages;
• validate and refine our Degradation Index to quantitatively measure the impact that cane toads have on the functional ecosystems of the Kimberley ;
• provide an improved strategy of assessing the small vertebrate terrestrial fauna in the wet-dry tropics of WA (essential for the purpose of preparing EIA statements);
• establish monitoring protocols for key vertebrate species thought to be at risk; and
• explore, and if feasible, develop a relocation and breeding program for a number of goanna species on islands in Lake Argyle .

We will employ two basic survey strategies to collect the necessary data:

a) a comprehensive trapping program over three years in three of the major habitat types typical of the east Kimberley ; and

b) a series of transects that target key species (e.g. goannas, birds of prey, frogs, mammals in riparian habitats and fish).

These basic survey data are absolutely essential if management plans are to be developed to minimise the impact that cane toads will have on the Kimberley . It will also have two other important benefits,namely:

• providing quality base line data on the faunal assemblages of the east Kimberley ; and
• providing a tool that measures changes in ecosystem function as a consequence of disturbance (e.g. agriculture, cane toads, mining).

Introduction

Cane toads (Bufo marinus) are rapidly moving across northern Australia and will establish a significant presence in the east Kimberley of Western Australia (WA) in the next few years (Sessional Committee on Environment and Sustainable Development 2003) . Currently, there are populations of cane toads about 150 km west of Katherine, in the Northern Territory (NT) with isolated populations closer to the border, and they are projected to reach the Kimberley in about 4-5 years (based on movement data provided in van Dam et al. (2002) .

The history of cane toads invading the northeast coast and the wet-dry tropics of the NT is that they will have a devastating negative impact on the fauna and the ecosystems of the Kimberley region (Freeland 2004, Lever 2001, Sessional Committee on Environment and Sustainable Development 2003, van Dam et al. 2002) . We know that cane toads can rapidly increase their numbers (females can lay 4,000 – 36,000 eggs per clutch (van Beurden & Grigg 1980, Zug & Zug 1979) . They will out-compete a number of other frog species (Covacevich & Archer 1975, Crossland 1997) , they forage in a diverse range of riparian and terrestrial habitats, and they breed in a variety of water bodies throughout the year (Freeland 2004, van Dam et al. 2002) . They can have densities that are orders of magnitude higher than similar native species (up to 600-800 m -2 for tadpoles, and 2,000 ha -1 for adults adjacent to water bodies) and they have excellent dispersal capabilities in the wet-dry tropics (van Dam et al. 2002) . They feed mostly on arthropods, but will eat almost any vertebrate they can catch and devour (Freeland 2004) .

Skin toxins from adult cane toads are known to be lethal for at least 27 terrestrial vertebrate species, and eggs and tadpoles are known to be highly toxic to many aquatic invertebrates, some fish species and many frog species (Lever 2001) . Nine native mammal species, many frog-eating snakes, and large varanids are known to die after ingesting or mouthing cane toads. The indirect impacts of this invasive toad will be devastating, and there will be many cascading consequences for the many ecosystems of the Kimberley .

When it was apparent that cane toads were about to invade the NT, Environment Australia and the Supervising Scientist (NT) undertook an extensive review of the literature to assess the potential impact of cane toads on Kakadu National Park (van Dam et al. 2002) . More recently, the WA Department of Industry and Resources commissioned HLA-Envirosciences Pty Ltd to undertake a similar assessment but focussed on the cane toad impact on the east Kimberley (Freeland 2004) . These two reports provide the most recent comprehensive summaries of the potential impacts of the cane toad in the wet-dry tropics of Australia . A summary of current studies on cane toad impact on native fauna is provided in the Legislative Assembly of the Northern Territory’s, Sessional Committee on Environment and Sustainable Development report on, Issues associated with the progressive entry into the Northern Territory of Cane Toads (2003).

Cane toads are the most comprehensively studied vertebrate species in Australia (Freeland 2004) . Despite the many research investigations and thousands of pages written on the ecology and negative impact that cane toads have had, and will have on the many near pristine ecosystems of the Kimberley, we have no demonstrated effective strategy to minimise their impact, let alone manage, control or limit their invasion of the region (Freeland 2004; Sessional Committee on Environment and Sustainable Development 2003, van Dam et al. 2002) . At best, we might be able to slow their progress with tighter quarantine controls.

Burnett (1997) provided evidence to indicate that populations of Dasyurus hallucatus (northern quoll) and Varanus gouldii (sand goanna), V. panoptes (Yellow Spotted Goanna), V. scalaris (spotted tree goanna) and V. mertensi (water monitor) are at serious risk of being decimated by cane toads. There are other goanna species that Burnett did not review that would also be very susceptible because of their preferred habitat, foraging behaviour and diet (e.g.V. mitchelli, V. acanthurus, V. glauerti). Other species assessed to be at risk ‘probably’ including the ornate frog (Limnodynastes ornatus), northern dwarf tree frog (Litoria bicolour), desert tree frog (Litoria rubella), blue-tongued skink (Tiliqua scincoids), Death Adder (Acanthopis praelongus), Mulga Snake (Pseudonaja australis), western brown snake (P. nuchalis), carpet python (Morelia spilota), brown tree snake (Boiga irregularis), slaty-grey snake (Stegonotus cucullatus), Black Bittern (Ixobrychus flavicollis) and blue-winged kookaburra (Dacelo leachii) (van Dam et al. 2002) . Numerous other species of reptiles, mammals and birds were recorded as ‘possibly’ at risk. There are other data and assessments that support these conclusions (Freeland 2004, van Dam et al. 2002, Watson & Woinarski 2003) .

A plethora of investigations have assessed the impact of cane toads on single species or taxa (see recent reviews in Freeland 2004, Lever 2001, Sessional Committee on Environment and Sustainable Development 2003, van Dam et al. 2002) . Only two studies have assessed the impact on broad scale terrestrial ecosystem function. Catling et al. (1999) assessed the short-term effects of cane toads on native fauna in the vicinity of Roper River, NT. They reported that there was little evidence that cane toads had a significant effect on the diversity or abundance of many of the native fauna in the short term, with only three species (Roth’s tree frog, Gilbert’s dragon lizard and the dingo) being seriously affected. However, the findings of this study are seriously compromised by their limited survey effort and subsequent small samples that take little account of normal temporal variations. Sessional Committee on Environment and Sustainable Development (2003) reported that monitoring studies are currently being undertaken in Kakadu National Park by the Parks and Wildlife Commission of the NT; however, no data are publicly available. The report by Watson and Woinarski (2003) is currently being acquired so it might be assessed.

We argue that the invasion of cane toads in to the Kimberley provides a unique opportunity to develop an understanding of the impact that these amphibians have on vertebrate ecosystems. We outline below a research project that will provide baseline data on the vertebrate assemblages of the east Kimberley that could be used to measure the affect that cane toads have once they have established themselves in the area. This proposed project has a number of significant additional benefits that are also outlined below.

Project objectives

 The primary objective of this research project is to monitor the impact that cane toads will have on native terrestrial vertebrates and ecosystem function in the east Kimberley . In undertaking this investigation we will achieve three important related secondary objectives. The specific objectives for this study are to:

• Establish high quality baseline data on small vertebrate assemblages in the east Kimberley to assess the long-term impact that cane toads will have on these assemblages (primary objective) and, in doing so, establish monitoring protocols for key amphibian, reptile, bird and mammal species that are likely to be heavily impacted by cane toads:

1. amphibians – e.g.Cyclorana australis (giant frog) and Limnodynastes ornatus (ornate frog);
2. reptiles – e.g.Varanus mertensi (water monitor ), V. gouldii (sand goanna), V. panoptes (yellow-spotted goanna), V. mitchelli (Mitchell’s water monitor), V. scalaris (spotted- tree goanna), and Tiliqua scincoides (northern blue tongue lizard ), Acanthophis praelongus (northern death adder);
3. birds – e.g.Todiramphus sanctus (sacred kingfisher), Dacelo leachii (blue-winged kookaburra), Cracticus nigrogularis (pied butcherbird); and
4. mammals – e.g.Hydromys chrysogaster (water-rat), Isoodon macrourus (northern brown bandicoot);

• Refine and validate our Degradation/Rehabilitation Index (Thompson 2004) as a quantitative tool to measure the impact that cane toads have on the functional ecosystems of the Kimberley;

• Provide an improved strategy for assessing the small vertebrate terrestrial fauna in the wet-dry tropics of WA for the purposes of preparing EIA statements, and in doing so:

1. document temporal variations in the trappable small vertebrate fauna in the wet-dry tropics so that ‘one-off’ future fauna surveys can be more adequately interpreted;
2. develop ‘typical’ averaged species accumulation curves for terrestrial fauna surveys in the wet-dry tropics so that small vertebrate assemblage structures might be better understood and the trapping effort required to adequately survey areas better appreciated;
3. describe ‘typical’ small vertebrate assemblages so that data from future surveys might be more adequately interpreted in a regional context; and
4. assess trap-type bias, and temporal changes in catch rates on a daily and a seasonal basis so that better trapping protocols can be employed in fauna surveys;

and

• Explore and, if feasible, develop a relocation program for a number of goanna species (e.g.V. mertensi, V. gouldii, V.panoptes, V. mitchelli) on islands in Lake Argyle , such that they will be protected from the impacts of the cane toad.

Methods

To achieve objectives 1, 2 and 3, a number of long-term terrestrial vertebrate fauna survey sites will be established. It is proposed that these surveys can be repeated in years to come after cane toads have established themselves in the area and the ‘before’ and ‘after’ data can be used to assess the impact of cane toads on the small vertebrate fauna of the east Kimberley. Reptiles, birds and mammals activity patterns vary appreciably on a daily, seasonal and year-to-year basis (Thompson & Thompson in press) . As a consequence, there will be significant seasonal and year-to-year variations in trapping data, overlain with the unavoidable ‘sampling error’. Therefore, surveys will need to be undertaken with sufficient intensity during each survey period so that sample sizes provide an adequate representation of the trappable population during each survey period. We will use averaged species accumulation curves to assess the adequacy of our field survey data during each period. Surveys will need to be repeated multiple times in a single year to address seasonal variations, and over a number of years to accommodate year-to-year variations. Activity patterns of many of the reptiles, amphibians and small mammals are at a peak during the commencement of the monsoon season, and they generally progressively decline through the wet season as prey availability increases and energy resources are replaced after the extended dry season. Many of the reptiles and amphibians appreciably reduce their activity levels during the dry season, and some small mammals also seem much less active and abundant. Year-to-year variations in catch rates for reptiles and mammals vary appreciably in semi-arid and mesic habitats (Thompson & Thompson in press).It is felt that this pattern will also be evident in the wet-dry tropics, although we could find no evidence that it had been quantified on a large scale for reptile, amphibian and mammal assemblages. Two basic survey strategies will be adopted:

• a trapping program in three of the major habitat types typical of the east Kimberley ; and
• transect surveys.

We would have preferred to use the BACI (before, after, control, impact) model in assessing the impact of cane toads. However, it is highly unlikely that any area will be protected (e.g. fenced) from cane toads in the east Kimberley and it will therefore not be possible to establish quality control sites. We have therefore adopted the approach of developing a high quality data set that provides an excellent assessment of the existing faunal assemblages (species richness and abundance) and that takes account of temporal variations before the cane toad invasion. We will use this as baseline data against which to assess the cane toad impact.

Trapping program

The trapping protocol will include pit-traps (pipes and buckets), funnel traps, wire cage and Elliott traps. Each trapping array will be laid out as shown in Figure 1.

Fifteen trapping arrays using the configuration as shown in Figure 1 will be laid out approximately 50 – 100 m apart in three different habitat types representative of the region and accessible during the wet season. In each habitat type we will select three areas separated by as much distance as is logically possible and install five arrays in each of these areas. This will provide us with three repeats, in each of three habitats. Each site will be simultaneously surveyed on three occasions each year:

a) at the end of the dry season (Nov/Dec);

b) at the beginning of the wet season (Jan/Feb); and

c) at the end of the wet season (Apr – May)].

Each survey period will last for a minimum period of 21 days, or longer, until the averaged species accumulation curve for the reptile and mammal assemblages at each site has plateaued (providing a clear indication that the trapping effort is adequate for that site). The anticipated trapping effort is shown in Table 1. We treat a pair of funnel traps (one either side of the drift fence) as equivalent to a single pit-trap (e.g. 20 L PVC bucket) when comparing capture success for different trap types.

Figure 1. A single trapping array consisting of three 20 L PVC buckets, three 150 mm PVC pipes, three pairs of funnel traps aligned along a 300 mm high drift fence, two Elliott cages and one wire cage trap.

Table 1. Trap types and trapping effort to be applied to three habitat types over a three year period

This is the trapping protocol that we have used at other sites (e.g. Ora Banda, Bungalbin, Cataby) and we are confident that it will provide the data that we require to produce an averaged species accumulation curve that plateaus, a good appreciation of temporal variations (season and year-to-year), an excellent understanding of the small vertebrate assemblages (i.e. species richness and abundance) in the three habitats, catch rate variations over time, and differences between trap-types when the data are used in appropriate multi-variate analyses. If we presume the catch rate, in pit and funnel traps combined, is 8% at the beginning of the wet season (Jan/Feb), 6% at the end of the wet season (Apr – May) and 4% at the end of the dry season (Nov/Dec), and we plot the likely captures on averaged species accumulation curves for semi-arid and mesic sites (Figure 2), then we can indicate the proportion of the total species in a biotope likely to be caught during a single trapping period in each biotope. We have been conservative in our estimates here, so we are confident that we will catch more than 80% of the species in each habitat type during each survey. The trapping program proposed generally far exceeds that employed anywhere in Australia to assess small vertebrate community structure, as we have enough data to indicate that most trapping programs have not captured a high proportion of the species in a given biotope.

Figure 2. Proportion of species likely to be caught in each habitat type during each survey period based on catch rates and averaged species accumulation curves for semi-arid and mesic habitats.

Each of these three sites will be established as permanent trapping sites and used on a long-term basis to assess the impact of cane toads in years to come.

Data from this trapping program will also be used to fine tune and validate our Degradation Index so it can be used as an effective tool to measure the impact of cane toads on the functional ecosystems in each of the three habitats.

Transect surveys

Transect surveys will be used to target and collect base line data on particular species or taxa. Twelve initial transects will be established (e.g. 2 for reptiles, 1 for birds of prey, 1 for birds in Lake Argyle, 4 for frogs, 2 for mammals).

Reptile Transects

Goannas, particularly large bodied goannas, are very vulnerable to cane toads, as are freshwater crocodiles and Gilbert’s dragon lizard. Two transects will target specifically the relative abundance of V. mertensi, but will also collect data on the relative abundance of V. panoptes, V. gouldii, Lophognathus gilberti and Crocodylus johnstoni. These will be carried out along the gravel track that runs adjacent to the M1 and Packsaddle irrigation channels around Kununurra. These five species of reptiles are frequently seen along the edges or in these irrigation channels. We already have survey data collected over a two year period for the M1 irrigation channel for V. mertensi and C. johnstoni. During these searches, we often saw both V. panoptes,V. gouldii and L. gilberti. For the lizards, surveys will be conducted at the same time each morning (while reptiles are basking) from the roof of a slow moving vehicle for 5 days (every second day) on four occasions each the year, and repeated for each of the three years. Surveys of C. johnstoni will be undertaken on the alternate days along M1 irrigation channel, at the same time each night, again on four occasions during the year, and repeated for each of the three years.

Birds of Prey and Kingfisher Transects

The available data on the impact of cane toads on birds of prey and kingfishers is ambiguous, and could best be described a ‘possible’ (van Dam et al. 2002) . Packsaddle Road ( » 6 km) runs through the irrigated farming areas on the south side of the Ord River and contains an unusually high abundance of raptors. These raptors either soar or perch on the very high power lines or power poles that run parallel to Packsaddle Road . We have also regularly seen small kingfishers in this area. Surveys will be conducted just before sunset from a slow moving vehicle for 5 consecutive days on four occasions during the year, and repeated for each of the three years. We believe the presence of the high number of birds of prey and kingfishers in this area reflects the availability of small reptiles, birds and frogs found in the adjacent areas. A change in the relative abundance of these raptors could be directly linked with the cane toad invasion. Other bird species seen along this transect will also be recorded.

Lake Argyle Bird Survey

Lake Argyle is an important breeding habitat for aquatic and semi-aquatic birds in the east Kimberley . Many of the bird species that frequent the waterways of the Ord and Bow Rivers, and the shallow areas of Lake Argyle are the species most likely to feed on cane toads (van Dam et al. 2002) . We are aware that Rio Tinto’s Argyle Diamond mine operation has supported an annual bird survey in the area of the southern end of Lake Argyle for a number of years. We wish to build on these data, and tighten the protocols, so that the information can be used to assess changes in bird abundances. Surveys will be conducted annually at the end of the wet season and when the lake edges are beginning to dry out. We will involve the local community in these surveys.

Frog Transects

Many frog species appear to be at risk, although the extent of that risk appears not to have been adequately assessed (van Dam et al. 2002) . We will again use Packsaddle Road and other bitumen roads to survey frog abundance. We have already undertaken preliminary surveys on Packsaddle Rd for frog species for other purposes. Surveys will be conducted an hour after sunset from a slow moving vehicle for 5 consecutive nights on two occasions during the year, and repeated for each of the three years. These occasions will coincide with the first heavy rains in the monsoon season (there is normally a couple of preliminary thunderstorms before the wet season establishes itself). A rainfall event greater than 30 mm after November will bring most of the burrowing frogs (e.g. Cyclorana australis) to the surface and a number of the arboreal species (e.g.Litoria rothii) will form large choruses in ponds adjacent to the road. Our experience in the last three years is that a large number of at least three species of frogs are present, easily identified and counted on this road at this time of the year. Two other transects will be established adjacent to temporary ponds known to contain at least three species of frogs. These latter two transects will be walked on 5 consecutive nights on three occasions during the year, and again repeated for each of the three years.

Mammal Transects

Hydromys chrysogaster (water-rat) and Isoodon macrourus (northern brown bandicoot) live in riparian habitats likely to be occupied by cane toads once they invade the east Kimberley . Both species are likely to eat frogs and could be susceptible to cane toad toxins, although the data from the east coast suggests that the water rat does not find the cane toad toxic. The northern brown bandicoot is rated by van Dam et al. (2002) as ‘possibly’ at risk. Six repeats of 20 cage traps will be set in riparian habitats known to contain both the Water-rat and the Northern Brown Bandicoot. These six transects will be surveyed on three occasions each year (immediately after the three trapping programs referred to above) for 5 consecutive nights for each of the three years.

We are also aware that the northern quoll (Dasyurus maculatus) is in the very high risk category. The Northern Quoll has not been found in the east Kimberley for a number of years; however, this may be because of a paucity in trapping effort in the area. During a recent visit to Kununurra, a resident indicated that a small group of northern quolls live near a caravan park on the outskirts of the town and are frequently seen foraging in the park. These sightings have not been confirmed. If they exist, we may catch them in one of our other trapping programs.

Varanus mitchelli Transects

Varanus mitchelli is a semi-aquatic, arboreal species that appears to have high site fidelity. Although not specifically mentioned in other studies as a species at high risk, we believe Varanus mitchelli could be at risk because it inhabits the riparian environments which cane toads prefer, eats frogs and is a widely foraging predator capable of eating both tadpoles and juvenile cane toads. We have frequently found them in the crevices of the small bridges over the irrigation channels and waterways scattered throughout the area. We have also trialled ‘arboreal’ pit-traps and had some success in trapping V. mitchelli. We will document the location of at least 20 of these individuals (marked with fish tags) to assess the extent to which they can be re-sighted on a regular basis, and can be used as a bio-indicator of the cane toad impact. We propose to explore further the use of ‘arboreal pit-traps’ as a trapping strategy with a view to using it as a tool to assess the abundance of this and other species. If V. mitchelli prove to be a useful bio-indicator, then a protocol will be developed and implemented.

Relocation Program

Environment Australia in its submission to the NT Legislative Assembly’s Sessional Committee on Environment and Sustainable Development (2003) on the invasion of cane toads into the NT indicated that it was supportive of conserving breeding populations of species threatened by cane toads through translocations. We believe there is potential to translocate some of the goanna species that are likely to be seriously threatened by cane toads on to islands in Lake Argyle with the view to establishing viable breeding populations. Many of these islands are large enough to support viable populations and are far enough from the ‘mainland’ to minimise the possibility that cane toads will be able to invade the islands. Clearly this will need to be regularly monitored. Therefore, to achieve objective four (4) we will describe the habitat requirements for V. mertensi, V. mitchelli, V. gouldii and V. panoptes and any other goanna species (e.g. Varanus acanthurus, V. kingorum, V. glauerti) considered likely to be under threat by the cane toad invasion of the east Kimberley . We will then search the large islands (> 10 ha) in Lake Argyle to determine which of these islands have habitats that could be suitable for translocation of one or more species. In conjunction with the Department of Conservation and Land Management we will prepare a detailed plan for the translocation of one or more of these species onto the islands. This plan will include how many individuals (ages and sexes), their origins, when and where they will be translocated, and how their survival will be monitored. The survey work to assess the suitability of the islands will be undertaken during the first year of operation, a report prepared and circulated within the relevant government agencies for consideration. Animals will be released during the fourth year of the project.

Expected Project Outcomes

This project will yield the following specific outcomes:

• Established field trapping sites with documented methodologies for undertaking further terrestrial fauna surveys once the cane toads have reached the east Kimberley;
• Established transects for surveying susceptible reptile, mammal, bird and freshwater fish species;
• Provide baseline data on terrestrial vertebrate faunal assemblages in three habitats of sufficient quality that can be used to measure the impact of the cane toad on these assemblages;
• Documented a Degradation Index that will assess the impact of cane toads on the terrestrial vertebrate ecosystem in the east Kimberley , and elsewhere in the State;
• Translocated susceptible goanna species on to islands in Lake Argyle ;
• provide detailed descriptions of the trappable vertebrate terrestrial fauna assemblages in three of the major habitats in the east Kimberley ;
• Provide detailed descriptions of the daily, seasonal and year-to-year variations in small trappable vertebrate fauna assemblages in the wet-dry tropics of Australia ;
• Provide averaged species accumulation curves for small trappable terrestrial vertebrate fauna in the wet-dry tropics;
• Assess trap-type bias in the trapping small terrestrial vertebrates in the wet-dry tropics;
• Provide advice to the Department of Conservation and Land Management, Environmental Protection Authority, environmental consultants and researchers on the trapping effort required to adequately survey the small trappable vertebrate fauna in the wet-dry tropics.

Funding

As it is the primary responsibility of State and Commonwealth government conservation agencies to protect the biodiversity of the State, an application has been made to the Department of Conservation and Land Management to financially support this project and an application has also been made to the Commonwealth government for funding.

References

Burnett S. (1997). Colonizing cane toads cause population declines in native predators: reliable anecdotal information and management implications. Pacific Conservation Biology3: 65-72.

Catling P.C., Hertog A., Burt R.J., Wombey J.C. & Forrester R.I. (1999). The short-term effect of cane toads (Bufo mariuns) on native fauna in the Gulf Country of the Northern Territory. Wildlife Research 26: 161-185.

Covacevich J. & Archer M. (1975). The distribution of the cane toad, Bufo marinus, in Australia and its effects on indigenous vertebrates. Memoirs of the Queensland Museum17: 305-310.

Crossland M.R. (1997). Impact of the eggs, hatchlings and tadpoles of the introduced cane toad, Bufo marinus (Anura: Bufonidae) on native aquatic fauna in Northern Queensland, Australia. Unpublished PhD thesis James Cook University, Townsville.

Freeland B. (2004). An Assessment of the Introduced Cane Toad's (Bufo marinus Lineaus) Impacts on the Native Australian Fauna, with particular Reference to the Eastern Kimberley Region. Perth: HLA-Envirosciences Pty Ltd.

Lever C. (2001). The Cane Toad: the History and Ecology of a Successful Colonist. West Yorkshire: Westbury Publishing.

Sessional Committee on Environment and Sustainable Development. (2003). Issues associated with the progressive entry into the Northern Territory of Cane Toads. Darwin: Legislative Assembly of the Northern Territory.

Thompson S.A. (2004). Mine site rehabilitation index using reptile assemblage as a bio-indicator. Unpublished PhD thesis Edith Cowan University, Perth.

Thompson S.A & Thompson GG. Temporal variation in reptile assemblages in the Goldfields of Western Australia. Journal of the Royal Society of Western Australia. IN PRESS.

van Beurden E.K. & Grigg G.C. (1980). An isolated and expanding population of the introduced toad Bufo marinus in New South Wales. Australian Wildlife Research7: 305-310.

van Dam R., Walden D., Begg G. (2002). A preliminary risk assessment of cane toads in Kakadu National Park. Darwin, NT: Supervising Scientist Report 164, Supervising Scientist.

Watson M. & Woinarski J. (2003). Vertebrate Monitoring and Re-sampling in Kakadu National Park, 2002. Report to Parks Australia, Darwin.

Zug G.R. & Zug P.B. (1979). The marine toad, Bufo marinus: a natural history resume of native populations. Smithsonian Contributions to Zoology284: 1-54.