SEARCH

SEARCH BY CITATION

Keywords:

  • anurans;
  • invertebrates;
  • invasive species;
  • mechanisms;
  • predation

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The ways in which invasive organisms influence native ecosystems remain poorly understood. For example, feral cane toads Bufo marinus have spread extensively through tropical Australia over the last 70 years, but assessments of their ecological impact remain largely anecdotal. We conducted experimental trials to examine the effect of cane toad presence on invertebrate fauna in relatively small (2.4 × 1.2 m) outdoor enclosures on a floodplain near Darwin in the wet–dry tropics. Toads significantly reduced invertebrate abundance and species richness, but only to about the same degree as did an equivalent biomass of native anurans. Thus, if toads simply replaced native anurans, the offtake of invertebrates might not be substantially different from that due to native anurans before toad invasion. However, our field surveys suggest that toads cause a massive (fourfold) increase in total amphibian biomass. The end result is that cane toads act as a massive nutrient sink in the floodplain ecosystem because they consume vast numbers of invertebrates but (unlike native frogs) are largely invulnerable to predation by frog-eating predators.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

It is relatively straightforward to enumerate the direct pathways by which invasive species can modify the attributes of native ecosystems, ranging from direct predation through to competition and disease transfer (Jaksic, 1998; Simon & Townsend, 2003; Lobos & Jaksic, 2005). Indirect (secondary) effects may also be important; for example, the invasive taxon might increase the mortality rates of predators that had previously influenced the abundance of prey (Spiller & Schoener, 1988; Flecker & Townsend, 1994). However, quantifying the magnitude of even the simplest direct effects, or understanding their exact nature, poses formidable logistical obstacles. Thus, even in the case of intensively studied invasive species, we often have only indirect (and, usually, merely inferential) evidence concerning the mechanisms by which they impact natural systems. To understand fully the effects of exotic organisms, we need more specific information on these potential pathways. Perhaps the best place to start is with one of the simplest, most obvious possibilities: that the invading taxon modifies the attributes of a natural system by consuming more food, or different kinds of food, than that taken by native species.

This simple mechanism of impact seems highly plausible in the case of many invasive organisms, and has been widely suggested as a probable mechanism of impact for the species that is the subject of our own research: the cane toad Bufo marinus. Toads were brought to Australia specifically to prey on insect pests of sugar cane, a testimony to their renown as voracious foragers (Mungomery, 1936; van Beurden, 1978, 1979, 1980). These putatively high feeding rates suggest that toad invasion might substantially reduce the abundance and/or diversity of invertebrate assemblages in tropical Australia (van Dam, Walden & Begg, 2002). To test this prediction, we need to quantify the attributes of invertebrate assemblages as a function of the presence or absence of toads. Remarkably, only one previous field study has attacked this issue, and has done so by surveying invertebrates in areas with and without cane toads (Catling et al., 1999). Such an approach has logistical benefits and can be conducted at broad spatial scales, but is poorly suited to identifying actual mechanisms of impact. An experimental approach allows greater confidence in this respect, but entails an inevitable reduction in spatial scale and a more formidable set of logistical obstacles.

In the present paper, we describe the results of an experiment using field enclosures to examine the effects of toad presence and toad body size on the diversity and abundance of invertebrates on a tropical floodplain. To place these results into a broader perspective, our experiment also included treatments with native frogs. In this way, we can not only establish whether or not toads influence invertebrate assemblages but also compare the magnitude and nature of any such effect with that generated by an equivalent biomass of native anurans. To apply the resulting information on offtake rates of toads and native frogs to the consequences of toad invasion, we also need to (1) extrapolate to much larger spatial and temporal scales than are feasible with experimental enclosures and (2) quantify the total biomass of anurans before and after toad invasion. Toads may have a large impact even if their consumption rates are low per toad, if the invasion of toads results in a higher total anuran biomass than was the case pre-invasion. Alternatively, if toad invasion greatly reduces the numbers of native frogs, these feral animals might reduce rather than increase the net predation pressure from anurans. Hence, we conducted preliminary surveys to assess the total anuran biomass before and after toad invasion.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Study organisms

Cane toads Bufo marinus

Native to Central and South America, these large, ground-dwelling anurans were introduced to north-eastern Queensland in 1935 in an unsuccessful attempt to control agricultural pests (beetles), and have since spread across most of northern Australia (Zug & Zug, 1979). Their current range extends from northern New South Wales through most of northern and eastern Queensland and the northern part of the Northern Territory. Bioclimatic analyses suggest that toads will ultimately spread to the northern part of Western Australia (Phillips, Brown & Shine, 2003). In Australia, cane toads have been recorded as growing to 240 mm, although sizes of 90–150 mm are more common (Tyler, 1991). Cane toads are habitat generalists, occupying most non-arid habitats within their Australian range (Freeland, 1984; Seabrook, 1993).

Giant burrowing frogs Cyclorana australis

These frogs are distributed in riparian habitats through northern Queensland, the far northern Northern Territory and northern Western Australia. They are found in a wide range of habitats from coastal forests, floodplains, monsoon forests and woodlands (Seabrook, 1993; Cogger, 1996). With an average length of 100 mm, this species is the largest native ground-dwelling anuran in our study sites. Giant burrowing frogs are more similar to cane toads in size, appearance and general habits than are any other sympatric anurans.

Dahl's aquatic frog Litoria dahlii

Distributed from Cape York in north Queensland to the north-western part of the Northern Territory, these semi-aquatic frogs grow to an average length of 70 mm and can attain very high densities in suitable habitats (Barker, Grigg & Tyler, 1995). This species is also likely to overlap extensively with cane toads in habitat use (billabong margins) and prey.

Experimental enclosures

The wet–dry tropics of northern Australia are hot year-round (daily maximum temperatures average >30°C in every month) but exhibits highly seasonal precipitation regimes. The area experiences a relatively brief (December–March) ‘wet season’ with heavy monsoonal rains, followed by a much longer (April–November) ‘dry season’ (see Brown, Shine & Madsen, 2005 for details). Our study was conducted during the wet season, when vegetation cover is most dense and anuran and invertebrate activity is at its peak. Beatrice Hill Farm (131°18′12′′E, 12°37′16′′S) is a farming property 60 km east of Darwin with a long history (>30 years) of use for grazing buffalo and agistment cattle. It encompasses part of the Adelaide River floodplain as well as surrounding higher ground. The first cane toads to reach this site did so midway through our study (January 2005); thus, at the time of construction of the enclosures (November 2004) no cane toads had been in the area. However, the close proximity of the invasion front allowed us to sample the local anuran fauna both in front of and behind the invasion front, and to obtain both toads and native frogs from nearby areas. Our experimental enclosures were constructed on a flat, grassy area in open forest dominated by Eucalyptus miniata, bordering the floodplain. Each enclosure was made from sheets of galvanized iron and was rectangular in shape (1.2 m wide, 2.4 m long, 80 cm high), with the lower edge of each iron sheet sunk 10 cm into the ground. The enclosures were covered with 10 mm netting to exclude avian predators.

Each enclosure contained a mixture of native and introduced pasture grasses (Pangola Digitaria eriantha subsp. pentzii, Sida Sida acuta and Tully Urochloa humidicola) and small herbaceous plants (buffalo clover Alysicarpus vaginalis, Calapo Calopogonium mucunoides and Senna Senna obtusifolia). This vegetation grew naturally after initial construction of the enclosures and closely resembled ‘natural’ vegetation in the surrounding area. Each enclosure contained a single shelter (40 × 40 cm ‘Hardie-flex’ plasterboard) and a water bowl. Procedural controls (see below) were also constructed from sheets of galvanized iron. A barbed wire fence around the enclosures and procedural control enclosures excluded grazing buffaloes.

We captured both species of native frogs on Beatrice Hill Farm, in areas more than 5 km away from the sites in which surveys of amphibian biomass were conducted. Toads were collected (also by hand) 10–70 km from our main study site. From their capture until 2 days before use in experiments, frogs and toads were maintained in enclosures similar to those described above. Water was available ad libitum, and to supply food we attracted insects with a light source (a light bulb above the holding pens). Two days before the experiment began, experimental subjects (frogs and toads) were placed in separate plastic bins (10–12 individuals per bin) in a laboratory. No food was provided to the animals, but water was available ad libitum.

Invertebrate sampling

To capture invertebrates, we used pitfall traps (plastic vials 50 mm in diameter and 120 mm deep) dug in so that the top of the trap was level with the ground surface. Each experimental enclosure contained two such traps, with another 48 traps outside the enclosures to assess possible artifacts from the enclosures themselves. Our 24 ‘procedural controls’ were pitfall traps placed the same distance from an upright metal sheet as were the traps inside the enclosures, and separated from each other by the same distance (1.2 m) as were those in the experimental enclosures. These controls were designed to mimic the enclosure traps in terms of proximity to the upright metal sheets, but without any artifacts arising from the enclosure walls restricting ingress or egress by invertebrates. Our 24 ‘field control’ pitfall traps were placed >10 m from the nearest enclosure or procedural control. These field controls thus provide estimates of the floodplain invertebrate fauna with no (or minimal) influence of the enclosures, whereas the procedural controls enabled us to evaluate artifacts arising from enclosure construction (Dayton & Oliver, 1980; Benedetti-Cecchi & Cinelli, 1997).

Each pitfall trap contained a mixture of 60% ethylene glycol and 10%‘Homebrand’ washing detergent to a depth of 40 mm. Traps were left open throughout the study, but were changed at 5-day intervals over three successive sampling periods. At the same time that pitfall traps were changed in each enclosure, a sweep-net sample was taken from that enclosure by passing a sweep net through the vegetation in the enclosure five times. The same method was repeated at procedural controls and adjacent to field control pitfall traps. All samples were stored in 70% ethanol and later sorted to morphospecies (Oliver & Beattie, 1996) using a dissecting microscope. Additionally, we scored and recorded body size (length and width from a dorsal aspect, to the nearest mm) of each captured invertebrate, number of individuals and whether or not the animal was capable of flight. For analysis, we combined data from pitfall trap and sweep-net samples to provide an overview of the invertebrate fauna within each enclosure (or control).

Experimental treatments

Cane toads span a wide range of body sizes both within and among populations; for example, females grow larger than males, and animals in long-established populations are typically smaller than those at the invasion front (van Beurden, 1979; Freeland, 1986; Lee, 2001). Thus, we included toads of two different body sizes in this experiment. Each ‘large B. marinus’ enclosure contained a single toad weighing c. 230 g, whereas ‘small B. marinus’ enclosures each had one toad of about 120 g. Native frogs are smaller than toads; therefore, we used multiple frogs per enclosure to provide a total anuran biomass similar to that in the ‘small toad’ treatment (i.e. 120 g). Thus, each C. australis enclosure required three to four individuals, whereas each L. dahlii enclosure required nine to 12 animals. There were 12 replicates of each of these four treatments, plus 12 control enclosures.

Field surveys of toad and frog abundance and biomass

For surveys of toad and frog abundance, we selected three sites behind the rapidly advancing invasion front, where toads were known to have been present for at least 12 months: (1) Adelaide River Township 100 km south of Darwin along the Stuart Hwy, on the Adelaide River (131°6′29″E, 13°14′26″S); (2) Mary River Park 100 km east of Darwin along the Arnhem Hwy (131°38′56″E, 12°54′25″S); and (3) Corroboree Park 70 km from Darwin along the Arnhem Hwy, and 5 km from the Mary River (131°29′10″E, 12°45′56″S). For comparison, we also surveyed three sites in advance of the front: (4) Beatrice Hill Farm, at least 10 km from where any collecting was done for our experiments (131°18′12″E, 12°37′16″S); (5) Middle Point Village 10 km north of Beatrice Hill Farm, 5 km from the Adelaide River (131°18′50″E, 12°34′55″S); and (6) Fogg Dam Nature Reserve 15 km north of Beatrice Hill Farm and 3 km from the Adelaide River (131°17′45″E, 12°33′26″S). All six sites were similar in habitat structure and composition, consisting of open woodland dominated by eucalypts (Eucalyptus tetradonta and E. miniata). The three ‘non-toad’ sites were all colonized by toads in March 2005, confirming their suitability as toad habitat.

Surveys were conducted on foot between 20:00 and 22:00 h Central Standard Time (CST) in January and February 2005. At each location, one of us (M. J. G.) searched for anurans for 60 min, and recorded the number of native frogs and cane toads. Individuals were located by their calls or by sight. Each site was surveyed on three occasions, in random order. Water bodies ranged in size from c. 10 m × 30 m to c. 20 m × 50 m. Surveys included single water bodies at each site and the surrounding area, within 30 m of the water. Body masses of a sub-sample of individuals of each species were recorded for later calculation of mean body sizes and, thus, overall biomass per species per site. This sub-sample consisted of the first 10 individuals of each species encountered during the surveys.

Statistical analysis

For the enclosure experiments, we first conducted a repeated-measures ANOVA with sampling period (first, second or third) as the repeated measure to see whether or not any treatment effects changed through time within the period of the experiment. This analysis showed no significant effect of time nor any significant interactions between time and the main effects. Thus, we pooled the results of our invertebrate counts for the three periods, and analyzed these data using single-factor ANOVAs with treatment (presence, size and identity of anurans within enclosures) as the factor. The dependent variables were number of invertebrates collected, number of species, mean body size and proportion of individuals belonging to flying versus non-flying taxa. Fisher's protected least significant difference (PLSD) posthoc tests were applied to significant results from the ANOVAs. For field survey data, we used single-factor ANOVAs to determine whether there were significant differences in mean species richness, abundance or total biomass of native frogs between sites with versus without cane toads.

The inclusion of multiple variables to describe invertebrate and anuran assemblages in these tests raises the possibility that statistical tests on such data are not independent, and hence that one should ‘correct’ the level of significance by Bonferroni or related procedures. Unfortunately, the decision as to which group of tests are ‘related’ is highly subjective (Cabin & Mitchell, 2000), and hence we report here uncorrected values. In practice, our results were so clearcut that no realistic application of Bonferroni correction within any ‘related group’ of tests would move a result from P<0.05 to P>0.05.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Experimental studies

We found significant differences among treatments for the numbers and mean body sizes of individual invertebrates, and the number of species in our samples (Table 1). There was no significant difference among treatments in the proportion of flying versus non-flying invertebrates (Table 1). Both for the overall number of individual invertebrates and for the number of species of invertebrates, Fisher's PLSD post hoc tests revealed no significant differences between the three types of controls (i.e. enclosure control=field control=procedural control: see Fig. 1a, b). Thus, enclosure artifacts appear to be minimal for these variables. However, both the number of individual invertebrates and their species richness were significantly reduced below these levels in the enclosures containing either toads (of both size classes) or native frogs [Fisher's PLSD post hoc test; (all controls) > large toads=small toads=Litoria=Cyclorana]. Thus, the presence of cane toads significantly reduced invertebrate abundance and species richness within our experimental enclosures, and did so to about the same degree as did an equal biomass of native frogs (see Fig. 1a, b).

Table 1.  Results of statistical tests of the influence of anurans on abundance, species richness and mean body size of invertebrates in outdoor enclosures and associated procedural controls on a tropical floodplain
Variabled.f.F-valueP-value
  1. The table shows results from a one-factor ANOVA, with the factor being treatment (presence, size and identity of anurans and three types of controls). Significant (P<0.05) results shown by bold font.

Number of individual prey items collected in enclosures after treatment applied6,775.134<0.001
Number of species of prey collected in enclosures after treatment applied6,7712.51<0.001
Mean body size of prey (mm2) collected in enclosures after treatment applied6,774.0980.001
% flying prey collected in enclosures after treatment applied6,770.8730.519
image

Figure 1. Effects of the presence and body size of anurans on the composition of samples of invertebrates collected from outdoor enclosures and surrounding areas on a tropical floodplain. (a) Number of individual invertebrates captured; (b) species richness of invertebrates captured; (c) mean body size of invertebrates captured, all from pitfall traps and sweep-net sampling over a 15-day sampling period. These samples thus indicate the numbers, types and sizes of invertebrates remaining after predation by anurans.

Download figure to PowerPoint

Mean body sizes of invertebrates also differed significantly among treatments (Fig. 1c). On average, invertebrates captured from enclosures containing cane toads (of either size class) were significantly larger than those from enclosures containing native anurans or from control enclosures (Fisher's PLSD post hoc test; large toad=small toad>either native anuran=procedural control=field control). This result suggests that toads selectively remove smaller invertebrates from the enclosure than do native frogs.

Field surveys of toad and frog abundance and biomass

Neither the numbers of frogs captured nor the estimated total biomass of native frogs recorded per survey differed significantly between sites where toads were present as opposed to those where they were absent (numbers: ANOVA, F1,16=0.702, P=0.41, Fig. 2a; total biomass: ANOVA, F1,4=0.003, P=0.96; see Fig. 2c). We recorded a total of 15 anuran species during field surveys, with more rather than less species of native frogs found at sites where toads were present versus those without toads (ANOVA, F1,16=17.65, P<0.001, Fig. 2b). Thus, the arrival of toads was correlated with a minor increase in anuran species richness (likely reflecting local habitat characteristics rather than an effect of toads), but no significant change in total abundance nor biomass of native frogs. Thus, the total biomass of anurans was about four times higher after toad invasion because toads attained higher numbers, and thus biomass (>1000 g of toads per survey), than did native frogs (<300 g: see Fig. 2c). Using nested ANOVA with sites nested within toad presence/absence (and testing the toad effect against the nested term), the arrival of toads significantly increased both the total number of anurans seen (F1,4=10.24, P<0.03) and total anuran biomass (F1,4=9.50, P<0.04).

image

Figure 2. Comparison between areas recently invaded by cane toads Bufo marinus, and nearby areas yet to be invaded, in terms of species richness, abundance and total biomass of anurans encountered during replicated standardized surveys (see text for details).

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Inevitably, experiments such as those described above are not performed under entirely ‘natural’ conditions, raising the possibility that some conclusions may be unreliable. In the case of the work above, the most likely problem involves the restriction of experimental animals to relatively small enclosures. However, data from sampling at procedural controls and field controls indicate that the enclosures did not significantly affect invertebrate biomass. Thus, the experimental anurans were exposed to natural levels of prey. In the field, the densities of all three of these anuran species regularly exceed the densities used in the enclosures. For example, Freeland & Kerin (1988) reported that toads could attain densities of up to 5000 ha−1. Toads have now invaded the floodplain containing experimental arenas, and are breeding at a site only 100 m away (G. P. Brown, pers. obs.). Thus, the habitat is suitable for toads, densities are within the range of field values, and the invertebrate fauna was not unduly affected by experimental enclosures. Hence, we believe that we can attempt to extrapolate our experimental results to the impact of toads on tropical floodplains. Issues of spatial scale necessarily weaken such inferences, but speculation about toad impact based on rigorous (albeit small-scale) experiments is surely preferable to the current situation, where such speculation is based only on correlation, intuition and anecdote.

Why did the presence of toads influence (increase) the mean body size of invertebrates remaining in our enclosures, as well as reduce overall numbers and species richness? Although some authors have stressed the catholic food habits of toads (Alexander, 1964; Grant, 1996), detailed studies have reported a trend for toads to take a higher proportion of ants than do sympatric anurans of other species (Bailey, 1976; Strussmann et al., 1984). Heavy reliance upon ants as food has been documented in the toads' native range within South America (Zug & Zug, 1979) as well as in their introduced range (Bailey, 1976; Grant, 1996). Presumably, one major effect of this bias will be to remove small insects selectively, thus increasing the mean body size of the remaining invertebrates. Studies on the proximate cues that elicit feeding responses in toads have shown that a broad range of stimuli can elicit feeding (Ingle & McKinley, 1978). For example, auditory as well as visual cues can elicit hunting by toads (Jaeger, 1976). The most important visual stimuli may comprise color (toads prefer dark-colored prey items), shape (toads prefer objects elongated along the axis of movement) and movement (toads are stimulated to attack by moving objects; Ingle & McKinley, 1978; Zug & Zug, 1979; Freeland, 1984). Other biases in the types of prey consumed may reflect toad activity patterns (adult toads are crepuscular and nocturnal, and so will take prey active at these times) and microhabitat use (toads are clumsy climbers, and thus will feed mostly on invertebrates close to the ground surface).

Cane toads have been spreading across Australia for more than 70 years, and throughout that entire period scientists have speculated that these invasive amphibians might influence natural ecosystems (e.g. see reviews by Lever, 2001; van Dam et al., 2002). The idea that direct predation by toads upon native invertebrates could be one potential mechanism of impact was expressed in some of the earliest published discussions of this topic, and has been repeated in numerous subsequent reports and discussions (e.g. van Beurden, 1979; Freeland, 1984; Freeland, Delvinqueir & Bonnin, 1986; van Dam et al., 2002). Given that massive public concern over toad invasion has stimulated very high levels of funding for research on these animals (to date, >$ 9 000 000: D. MacRae, pers. comm.), it is surprising that our study is the first attempt to measure such an impact. The closest previous approaches have involved surveys on either side of the toad invasion front (Catling et al., 1999): however, the inference of causation remains indirect for such correlative data. In contrast, we worked at a site not yet exposed to toad predation, and (at least in relatively small field enclosures) detected strong effects of toad presence on invertebrate numbers and species richness.

Our enclosure experiments provide the strongest empirical evidence to date that the arrival of feral cane toads can modify invertebrate diversity, abundance and composition (body size distributions) of the invertebrate assemblage on tropical floodplains. Nonetheless, our data also suggest that toads are not too different in these respects to native frogs, or at least to the two species we studied, which are the most similar to toads in size and feeding habits. These two native anuran taxa also were similar to each other in these effects. Although some differences between the impact of toads versus the two species of frogs were apparent (e.g. on mean body size of invertebrates remaining in the enclosures), overall the similarities were more striking than the differences (Fig. 1).

If we are prepared to extrapolate from field enclosures to a landscape scale, do the above results mean that toad invasion will have no real effect on the abundance and diversity of native invertebrates? The answer to this question will depend upon the biomass of anurans after toad invasion, and this might be increased (if toads are simply ‘additional’ components) or reduced (if toads compete with or prey upon native anurans, thereby decreasing their numbers). Our survey data suggest the former effect (at least in the short term), with an overall fourfold increase in anuran biomass following toad invasion (Fig. 2). Previous studies similarly found no evidence of declines in native anurans following the arrival of toads (Alcala, 1957; Zug, Lindgren & Pippet, 1975; Freeland & Kerin, 1988). Doubtless the exact magnitude of impact also depends upon habitat characteristics. It seems likely that toads will attain their highest densities at sites where their closest ecological analogs (native frogs) are also (and apparently remain) abundant, a pattern documented for many other non-native taxa also (e.g. Levine, 2000; Sax, 2002; Crall et al., 2006).

Because toads averaged much larger individual mass than native frogs, and attained higher densities, the overall effect of toad arrival was to increase total anuran biomass greatly (by about fourfold: Fig. 2c), and hence increase the rate of offtake of invertebrates in this system. The massive biomass attained by this invasive species, rather than any species-specific attributes of foraging biology, may thus prove to be the most significant factor determining the rates at which toads remove invertebrates from floodplain habitats. Apart from reducing the food source for insectivores, depletion of invertebrate biomass may interrupt significant ecological pathways ranging from decomposing organic matter through to dispersing seeds (Wallace & Trueman, 1995; Dahl & Greenberg, 1997; Lindsey & Washok, 2000; Graca, 2001; Park et al., 2001).

The ecological impact of toads will depend not only on the rate at which they remove invertebrates via predation but also on the rate at which nutrients in the bodies of those invertebrates are transferred to higher trophic levels within the ecosystem. Toads differ significantly from native frogs in this respect, as well as in their higher standing-crop biomass. Toads are invulnerable to most predators in Australian systems, whereas frogs are major dietary items for a diverse array of other anurans, reptiles (e.g. snakes, varanid lizards), birds and mammals (Strahan, 1983; Barker & Vestjens, 1989; Cogger, 1996). In consequence, adult toads likely experience much lower mortality rates than do most native frogs, and may achieve considerable lifespans; some individual toads have been recorded to live for >15 years in captivity (Pemberton, 1949). Thus, toads may influence ecosystem function by a previously unrecognized pathway: by reducing rates of nutrient recycling. Essentially, the massive standing crop of toads acts as a major nutrient sink, whereby nutrients previously incorporated into invertebrates (and occasional small vertebrates) are tied up in a form (the bodies of toads) unavailable to higher trophic levels. Thus, the invasion of cane toads can not only reduce the species richness and abundance of invertebrates on tropical floodplains, but may also modify ecological processes within these complex systems in ways that we do not yet understand.

Clearly, many more studies will be needed to test and extend the ideas that have arisen from the current study. For example, the densities and body sizes of cane toads vary considerably among sites and seasons, and the distinctive diurnal habits, microhabitat preferences and small body sizes of metamorph toads (Freeland & Kerin, 1988) doubtless result in impacts on invertebrate prey very different from those of the larger toads that were the subjects of our own study. Similarly, decreases in mean adult body sizes and densities of toads as a function of time since colonization (Freeland et al., 1986; Lever, 2001) presumably have substantial impacts on the resource offtake by toad populations. Nonetheless, these problems are accessible to study, and our work with field enclosures represents the first small step toward replacing decades of speculation about toad impact with an understanding based on empirical evidence.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

We thank Eric Cox and the staff at Beatrice Hill Farm for logistical support, Emma Henderson and Sebastian Iglesias for assistance in the field, S. Dusty and J. Brown for advice on experimental design, and the Australian Research Council for funding.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Alcala, A.C. (1957). Philippine notes on the ecology of the giant marine toad. Silliman J. 4, 9096.
  • Alexander, T.R. (1964). Observations on the feeding behaviour of Bufo marinus (Linne). Herpetologica 20, 255259.
  • Bailey, P. (1976). Food of the marine toad, Bufo marinus, and six species of skink in cacao plantation in New Britain, Papua New Guinea. Aust. Wildl. Res. 3, 185188.
  • Barker, J., Grigg, G.C. & Tyler, M.J. (1995). A field guide to Australian frogs. Sydney: Surrey Beatty and Sons.
  • Barker, R.D. & Vestjens, W.J.M. (1989). The food of Australian birds. I. Non-passerines. Melbourne: Melbourne University Press.
  • Benedetti-Cecchi, L. & Cinelli, F. (1997). Confounding in field experiments: direct and indirect effects of artifacts due to the manipulation of limpets and macroalgae. J. Exp. Mar. Biol. Ecol. 209, 171184.
  • Van Beurden, E.K. (1978). Report on the results of stage 1 of an ecological and physiological study of the Queensland cane toad, Bufo marinus. Report to ANPWS, Canberra.
  • Van Beurden, E.K. (1979). Report on the results of stage 2 of an ecological and physiological study of the Queensland cane toad, Bufo marinus. Report to ANPWS, Canberra.
  • Van Beurden, E.K. (1980). Report on the results of stage 3 of an ecological and physiological study of the Queensland cane toad, Bufo marinus. Report to ANPWS, Canberra.
  • Brown, G.P., Shine, R. & Madsen, T. (2005). Spatial ecology of slatey-grey snakes (Stegonotus cucullatus, Colubridae) on a tropical Australian floodplain. J. Trop. Ecol. 21, 605612.
  • Cabin, R.J. & Mitchell, R.J. (2000). To Bonferroni or not to Bonferroni: when and how are the questions. Bull. Ecol. Soc. Am. 81, 246248.
  • Catling, P.C., Hertog, A., Burt, R.J., Wombey, J.C. & Forrester, R.I. (1999). The short-term effects of cane toads (Bufo marinus) on native fauna in the Gulf Country of the Northern Territory. Wildl. Res. 26, 161185.
  • Cogger, H.G. (1996). Reptiles and amphibians of Australia. Sydney: Reed Books.
  • Crall, A.W., Newman, G.J., Stohlgren, T.J., Jarnevich, C.S., Evangelista, P. & Guenther, D. (2006). Evaluating dominance as a component of non-native species invasions. Biodivers. Distrib. 12, 195204.
  • Dahl, J. & Greenberg, L. (1997). Foraging rates of a vertebrate and an invertebrate predator in stream enclosures. Oikos 78, 459466.
  • Van Dam, R.A., Walden, D.J. & Begg, G.W. (2002). A preliminary risk assessment of cane toads in Kakadu National Park. Scientist report 164, Supervising Scientist, Darwin.
  • Dayton, P.K. & Oliver, J.S. (1980). Problems in the experimental analyses of population and community patterns in benthic marine environments. In Marine benthic dynamics: 93120. Tenore, K.L. & Coull, B.C. (Eds). Columbia: University of South Carolina Press.
  • Flecker, A.S. & Townsend, C.R. (1994). Community wide consequences of trout introduction in New Zealand streams. Ecol. Appl. 4, 798807.
  • Freeland, W.J. (1984). Cane toads: a review of their biology and impact on Australia. Technical report no. 19. Darwin: Conservation Commission of the Northern Territory.
  • Freeland, W.J. (1986). Populations of cane toad Bufo marinus in relation to time since colonization. Aust. Wildl. Res. 13, 321330.
  • Freeland, W.J., Delvinqueir, B.L.J. & Bonnin, B. (1986). Food and parasitism of the cane toad, Bufo marinus, in relation to time since colonization. Aust. Wildl. Res. 13, 489499.
  • Freeland, W.J. & Kerin, S.H. (1988). Within-habitat relationships between invading Bufo marinus and Australian species of frog during the tropical dry season. Aust. Wildl. Res. 15, 293305.
  • Graca, M.A.S. (2001). The role of invertebrates on leaf litter decomposition in streams – a review. Int. Rev. Hydrobiol. 86, 383393.
  • Grant, G.S. (1996). Prey of introduced Bufo marinus on American Samoa. Herpetol. Rev. 27, 6769.
  • Ingle, D. & McKinley, D. (1978). Effects of stimulus configuration on solicited prey catching by the marine toad (Bufo marinus). Anim. Behav. 26, 885891.
  • Jaeger, R.G. (1976). A possible prey-call window in anuran auditory perception. Copeia 1976, 833834.
  • Jaksic, F.M. (1998). Vertebrate invaders and their ecological impacts in Chile. Biodivers. Conserv. 7, 14271445.
  • Lee, J.C. (2001). Evolution of secondary sexual dimorphism in the toad, Bufo marinus. Copeia 2001, 928935.
  • Lever, C. (2001). The cane toad. The history and ecology of a successful colonist. Otley, West Yorkshire: Westbury.
  • Levine, J.M. (2000). Species diversity and biological invasions: relating local process to community pattern. Science 288, 852854.
  • Lindsey, A.V. & Washok, K.A. (2000). Seed removal by ants, birds and rodents in a woodland savanna habitat in Zimbabwe. Afr. Zool. 35, 295299.
  • Lobos, G. & Jaksic, F.M. (2005). The ongoing invasion of African clawed frogs (Xenopus laevis) in Chile: causes of concern. Biodivers. Conserv. 14, 429439.
  • Mungomery, R.W. (1936). A survey of the feeding habits of the giant toad (Bufo marinus), and notes on its progress since its introduction into Queensland. Proc. Qld Soc. Sugar Cane Tech. 1936, 6374.
  • Oliver, I. & Beattie, A.J. (1996). Invertebrate morphospecies as surrogates for species: a case study. Conserv. Biol. 10, 99109.
  • Park, K.J., Robertson, P.A., Campbell, S.T., Foster, R., Russell, Z.M., Newborn, D. & Hudson, P.J. (2001). The role of invertebrates in the diet, growth and survival of red grouse (Lagopus lagopus scoticus) chicks. J. Zool. (Lond.) 254, 137145.
  • Pemberton, C.E. (1949). Longevity of the tropical American toad, Bufo marinus. Science 110, 512.
  • Phillips, B.L., Brown, G.P. & Shine, R. (2003). Assessing the potential impact of cane toads on Australian snakes. Conserv. Biol. 17, 17381747.
  • Sax, D.F. (2002). Native and naturalized plant diversity are positively correlated in scrub communities of California and Chile. Biodivers. Res. 8, 193210.
  • Seabrook, W. (1993). Habitat use of the cane toad Bufo marinus: implications for assessment of impact and control strategies. PhD thesis, University of Sydney.
  • Simon, K.S. & Townsend, C.R. (2003). Impacts of freshwater invaders at different levels of ecological organization, with emphasis on salmonids and ecosystem consequences. Freshwater Biol. 48, 982994.
  • Spiller, D.A. & Schoener, T.W. (1988). An experimental study of the effect of lizards on web-spider communities. Ecol. Monogr. 58, 5777.
  • Strahan, R. (1983). The complete book of Australian mammals. Sydney: Australian Museum, Angus and Robertson.
  • Strussmann, C., Do Vale, M.B.R., Meneghini, M.H. & Magnusson, W.E. (1984). Diet and foraging mode of Bufo marinus and Leptodactylus ocellatus. J. Herpetol. 18, 138146.
  • Tyler, M.J. (1991). Australian frogs. A natural history. Sydney: Reed New Holland.
  • Wallace, H.M. & Trueman, S.J. (1995). Dispersal of Eucalyptus torelliana by the resin-collecting stingless bee, Trigonia carbonaria. Oecologia 104, 1216.
  • Zug, G.R., Lindgren, E. & Pippet, J.R. (1975). Distribution and ecology of the marine toad, Bufo marinus, in Papua New Guinea. Pac. Sci. 29, 3150.
  • Zug, G.R. & Zug, P.B. (1979). The marine toad, Bufo marinus: a natural history résumé of native populations. Smithsonian Contrib. Zool. 284, 158.