Journal of Mammalogy, 95(6):000–000, 2014

Diet and prey selectivity of three species of sympatric mammalianpredators in central Australia

EMMA E. SPENCER,* MATHEW S. CROWTHER, AND CHRISTOPHER R. DICKMAN

Desert Ecology Research Group, School of Biological Sciences, University of Sydney, Sydney, New South Wales 2006,Australia

* Correspondent: espe7638@uni.sydney.edu.au

The diets of predators and their selection of prey often shape prey community dynamics. Understanding how

different predators select their prey could enable ecologists to predict their impact on specific prey

populations. Here, we investigate the diets of the feral cat (Felis catus), red fox (Vulpes vulpes), and dingo

(Canis dingo) in the Simpson Desert of central Australia, over a 1-year period between 2011 and 2012, and

compare the selectivity of these predators for small mammalian prey. We found that cats showed the greatest

consumption of small mammals, whereas dingoes consumed larger prey, thus indicating preferences for

different prey sizes. High occurrence of small mammals in the diets of all predators probably reflected high

abundances of small mammals in the environment; rodents declined after an irruption, but were still abundant

at the time of sampling. The cat exercised greatest selectivity for small mammal species, whereas the dingo did

not positively select for any species. Positive selection by predators for the long-haired rat (Rattusvillosissimus) and negative selection for the spinifex hopping-mouse (Notomys alexis) may reflect inefficient

and well-developed escape strategies by these 2 prey species, respectively. High selectivity by the cat for

Forrest’s mouse (Leggadina forresti) suggests that conservation of this rare rodent may depend on effective cat

management.

Key words: Canis dingo, cat, desert ecology, dingo, Felis catus, fox, introduced predator, predator scats, small mammals,

Vulpes vulpes

� 2014 American Society of Mammalogists

DOI: 10.1644/13-MAMM-A-300

Feeding and food selection are fundamental ecological

processes that enable consumers to meet their energy

requirements, but they also play an important role in shaping

prey community dynamics (Sih et al. 1998). To maximize rates

of energy gain, many predators are selective consumers.

Instead of consuming prey items in proportion to their

abundance, they prey selectively on particular organisms or

on specific life stages of these organisms (e.g., Janzen et al.

2000). Selective predation may be based on prey size, sex, and

the activity or behavior of the prey or predator species (e.g.,

Karell et al. 2010; Klare et al. 2010). For instance, ambush

predators may prey upon more evasive prey individuals in

comparison to mobile predators, because of the ability of these

predators to avoid detection by prey (Cooper et al. 1985). A

quantitative understanding of how different predators select

their prey could enable ecologists to predict the impact of

predators on specific prey populations. This understanding may

be of particular value in situations where the predators are

invasive and native prey species are at risk (Salo et al. 2007,

2010).

The feral cat (Felis catus), European red fox (Vulpes vulpes),

and dingo (Canis dingo [see Crowther et al. 2014]) are the

dominant mammalian predators in Australia and play an

important role in shaping prey communities across the

continent. The dingo has been present for at least 3,000–

5,000 years (Oskarsson et al. 2012; Sacks et al. 2013); it is not

thought to have highly detrimental current effects on native

prey species, although it may have impacted negatively on

native marsupial carnivores soon after arrival (Johnson and

Wroe 2003). On the other hand, the fox and cat were

introduced during the past 150 years and their subsequent

spread across Australia has been linked with the decline and

extinction of a range of native wildlife species (Burbidge and

McKenzie 1989; Dickman 1996a). High selectivity of

vulnerable small mammals by these introduced predators,

specifically by the feral cat (i.e., compared to the dingo), may

w w w . m a m m a l o g y . o r g

0

partially explain why these predators have had such a

considerable impact on native prey species in Australia

(Paltridge et al. 1997).

The diets of the dingo, red fox, and feral cat vary with

prey abundance and seasonal and environmental conditions

(Marsack and Campbell 1990; Corbett 1995; Paltridge 2002;

Cupples et al. 2011); that of the dingo also varies with pack

structure or size (Thomson 1992). In general, however, cats

specialize on small mammals with a mean prey size

equivalent to 41 g (Pearre and Maass 1998), and generally

less than 200 g (Dickman 1996b). The dingo favors

medium- to large-sized mammalian prey including common

wombats (Vombatus ursinus) and macropods (Thomson

1992; Letnic and Crowther 2013). By contrast, the fox

prefers small- to medium-sized mammals (Triggs et al.

1984). Although prey selectivity by dingoes, cats, and foxes

is often linked to prey size (and energetic return), it also

likely reflects the hunting strategies and habitat use of each

species (Glen et al. 2011). For example, cats are able to

specialize on small rodents even when prey is sparse because

of their ability to stalk prey and use cover to advantage

(Dickman 2009).

Over the past 100 years at least 21 mammal species have

become extinct in Australian desert regions, with 9 of these

species disappearing only from the arid and semiarid parts of

their ranges (Menkhorst and Knight 2001; McKenzie et al.

2007). Despite this, patterns of prey selectivity by mammalian

predators remain poorly understood in these environments. We

investigated the diets of the feral cat, dingo, and red fox in the

Simpson Desert, southwestern Queensland, by examining the

contents of predator scats. In particular, we aimed to explore

patterns of selectivity for small mammalian prey by the 3

predators in relation to prey availability, and discuss potential

links between these patterns and prey size preferences by the

predators, among other possible factors. Based on current

understanding of the predator species and prior studies in this

environment (Pavey et al. 2008a; Cupples et al. 2011), we

predicted that:

1) The diets of cats, foxes, and dingoes will contain more

small mammalian prey items than any other prey type;

2) There will be a preference for smaller mammals by cats,

larger mammals by dingoes, and larger and smaller

mammals by foxes; and

3) Cats will show the strongest selectivity for small mammals,

for example, selecting for species that weigh the least

among those available, whereas dingoes and foxes will be

more opportunistic in their selection of small mammal

species.

Although dietary studies cannot quantify the impacts of

predators on prey populations, they provide a simple 1st step to

identifying potential impacts and can highlight potentially

vulnerable, or selectively depredated, prey species. If patterns

in prey selectivity are discovered, targeted control action then

can be taken to mitigate risks and threats to these prey

populations (Medina et al. 2006).

MATERIALS AND METHODS

Study system.—The study area spanned 2 private

conservation reserves, Ethabuka and Cravens Peak Reserves,

and 1 cattle grazing property, Carlo, located in the Simpson

Desert, southwestern Queensland, central Australia (238460S,

138828 0E; 23816 0S, 138817 0E; and 23829 0S, 138832 0E,

respectively). The regional climate is highly irregular and

driven by the El Nino Southern Oscillation (Letnic and

Dickman 2006). Annual mean rainfall at Boulia, on the

northern edge of the desert, is 302 mm (Bureau of Meteorology

2013); however, the year-to-year rainfall is highly variable.

Prior to this study, on-site rainfall was low, with a total of 34

mm of rain falling from January to November 2012. However,

from January 2010 to April 2011, 950 mm of rain was recorded

(rainfall data taken from a preexisting Environdata weather

station located on Ethabuka Reserve).

Long, parallel sand dunes characterize this landscape and

run north-northwest–south-southeast, about 0.6–1 km apart,

rising to 8–10 m in height (Purdie 1984). The dune swales are

dominated by spinifex grass (Triodia basedowii), which

provides ground cover of 30–40%; gidgee trees (Acaciageorginae) also occur in patches of clay soil in the dune

valleys. The dune crests are sparsely vegetated but support

several species of ephemerals and shrubs (e.g., Crotalaria spp.,

Calotis erinacea, Tephrosia rosea, Goodenia cycloptera, and

Grevillea stenobotrya). Herbs are abundant after heavy rain,

but vegetation cover recedes quickly during drought (Letnic

and Dickman 2006). In this dry, sparsely vegetated environ-

ment, prey species are usually scarce. After heavy rain,

however, populations of small mammals—particularly ro-

dents—irrupt, in turn attracting mobile predators (foxes, cats,

dingoes, and letter-winged kites [Elanus scriptus]) into the

system. These predators may impose considerable per capita

predation pressure on prey during irruptions and also afterward

as prey numbers decline (Letnic et al. 2005; Pavey et al.

2008a).

Scat collection and analysis.—Fecal remains of predators

were collected opportunistically along unsealed roads, around

bores, and in dune swales and on dune crests, between

September and November 2012. Most collection sites had been

searched a year previously, so scats were not expected to be

older than 1 year. However, any samples that appeared

decomposed or broken were judged to be older than 12

months and were removed from the analysis. We identified

scats as fox, cat, and dingo according to their size, shape,

smell, and color (Triggs 1996), and any unidentified samples

were discarded. Samples were then packaged individually in

paper bags and dried in an oven at 608C for 5 h to kill parasites.

Once dried, we washed the samples, leaving only the

indigestible fragments of prey (hair, teeth, bones, skin,

scales, feathers, vegetation, and invertebrate exoskeletons).

We then placed fragments on a sorting tray with quarter

divisions, with 1 quarter split into 5% divisions to allow for

visual sorting and to aid in estimating the percentage volume of

prey items in each sample, which was approximated by eye.

0 Vol. 95, No. 6JOURNAL OF MAMMALOGY

All samples were processed; indigestible items were

identified and then verified by an independent expert (G.

Story, Scats About, Pty., Majors Creek, Australia). Private

reference collections of bones, hairs, and arthropod exoskel-

etons were used for reference in the identification process. We

identified larger bone and exoskeleton remnants under a

dissecting microscope. Hair samples were identified using

cross-sectional analysis and whole-mount techniques, compar-

ison to reference samples, and the program Hair ID methods

(Brunner and Triggs 2002). For each sample, remains were

identified to the lowest taxonomic level (species) whenever

possible. In general, mammalian prey items were classified to

species, reptiles to species or family, birds to order, and

arthropods to class. Plant material was grouped into broad

categories including ‘‘seeds and fruit’’ and ‘‘grass.’’ When

material could not be identified under any of these categories, it

was classified as ‘‘other.’’Estimating small mammal abundance.—To gauge dietary

selectivity by a consumer, it is necessary to compare the prey

eaten with the prey items that are potentially available to be

eaten. Here, because small mammals were the focus of study,

abundance estimates were made for rodent species prior to scat

collection (and thus at about the same time that the predators

were hunting and consuming prey) on 8 preexisting 1-ha

trapping grids (. 1 km apart) in the study region. Grids

comprised 36 pitfall traps made from polyvinyl chloride pipes

(diameter ¼ 160 mm, depth ¼ 600 mm) buried flush with the

ground, with each trap on grids spaced 20 m apart. Traps were

furnished with a 5-m-long drift fence made from aluminum fly

wire to increase trapping efficiency (Friend et al. 1989) and

checked for 3 consecutive mornings. All captured animals were

identified to species and ear-notched. Trapping was conducted

on 4 occasions between September 2011 and 2012, and

animals handled in accordance with guidelines approved by the

American Society of Mammalogists (Sikes et al. 2011).

Research was approved by the University of Sydney Animal

Ethics Committee (approval L04/4–2009/3/5020).

Statistical analyses.—We classified prey items into the

following categories: small mammals (, 500 g), medium-sized

mammals (500–6,999 g), large mammals (. 7,000 g), birds,

reptiles, arthropods, vegetation, rubbish, and other. Mammals

were sorted into small, medium, and large based on the

maximum weights listed in Menkhorst and Knight (2001). To

determine whether predator diets had been adequately sampled,

we plotted the cumulative diversity of prey items identified in

each sample against the number of scats examined. To

calculate diversity we applied the Brillouin index, using the

equation:

H ¼lnN! �

Xlnni!

N;

where H is the dietary diversity of the predator, N is the total

number of individual prey recorded, and ni is the number of

individual prey items of the ith type (Brillouin 1956). We used

the Brillouin index of diversity because prey items were

sampled nonrandomly (Pielou 1975).

Several analytical methods can be used to assess the diets of

predators based on scats, although percent frequency of

occurrence (the percentage of the total number of scat samples

containing a specific food item) is most commonly applied

(e.g., Pavey et al. 2008a; Cupples et al. 2011). We calculated

the percent frequency of occurrence of individual prey items in

predator scats, and also the percent frequency of occurrence of

each prey type, because prey items of different predators are

often compared based on taxon and size classes (Paltridge

2002; Mitchell and Banks 2005; Glen and Dickman 2008).

Because frequency of occurrence may overestimate the

importance of small food items in the diet (Corbett 1989;

Klare et al. 2011), we also reported the percent volume of prey

types in scats (the volume of a particular food type in scats

expressed as a percentage of the total volume of all food types

in the scats). Volumetric measurements also have limitations

and may underestimate prey items that are easily digested

(Klare et al. 2011). For this reason, it is best to consider both

frequency of occurrence and volume measurements in dietary

studies (Glen and Dickman 2006; Klare et al. 2011). We used

this approach here to test our 1st hypothesis.

To address hypothesis 2, we investigated compositional

divergences in the diets of dingoes, foxes, and cats, using

frequency of occurrence and volume measurements. For

volume, we used nonmetric multidimensional scaling based

on a Bray–Curtis similarity matrix (Clarke and Warwick

1994) and statistically compared the predators using analysis

of similarities (ANOSIM). If ANOSIM was significant (P ,

0.05), we used SIMPER to identify which prey types

contributed to differences in the predators’ diets (Clarke and

Warwick 1994). We used PRIMER 6 for Windows (Clarke

and Gorley 2006) for all analyses. Although Bray–Curtis

measures of difference in community composition can be

problematic, particularly with the inclusion of rarer species

(Chao et al. 2006), our method sought to examine relative

differences between scat samples for broad categories of taxa.

Further, this approach is comparable to methods used in prior

studies on predator selectivity in our desert system (Pavey et

al. 2008a; Cupples et al. 2011), and hence allows results to be

compared.

We did not rely solely on Bray–Curtis measures to compare

predator diets, and also addressed hypothesis 2 by investigating

differences in the frequency of occurrence of particular prey

types in predator diets. For this, we used logistic regression

with predator species as the categorical explanatory variable

and the presence or absence of a food type as the response

variable:

logitðpiÞ ¼ aþ bjXij;

where pi is the probability that any observed scat will contain a

given food type, a is the intercept value, bj is the influence of

the type of scat (fox, cat, dingo) on pi, and Xij is an indicator

variable representing which type of predator (j) produced scat i.These models were built using R (R Development Core Team

2012).

December 2014 0SPENCER ET AL.—PREY SELECTIVITY BY SYMPATRIC PREDATORS

To determine selectivity for specific small mammal species

by each predator (hypothesis 3), we used 2 methods. First, we

calculated Ivlev’s electivity index (D) as modified by Jacobs

(Jacobs 1974):

D ¼ r � p

r þ p� 2rp;

where D is prey selectivity, r is the proportion of a given prey

species in the predator’s diet (from percent frequency of

occurrence), and p is the proportional availability, or

abundance, of the prey species in the study area. Prey

abundance was estimated by determining the numbers of each

small mammal species captured per hectare per night using

pooled data collected from 4 trapping periods between

September 2011 and 2012. Pooling over 1 year was justified

because scats were , 1 year old and thus only likely to

represent the prey that was present over this period. The index

value indicates whether a prey species is selected proportion-

ately more (þ1) or less (�1) than its availability in the

environment. Index values of 0 indicate that prey species are

consumed in proportion to their abundance. Second, we used

the Bonferroni simultaneous-confidence-interval approach

(Byers et al. 1984; Cherry 1996) to test for differences

between the relative availability of small mammals in the

environment (based on the proportional abundance of individ-

uals captured on the 8 trapping grids, over the 3 test nights, and

4 separate sampling periods) and their representation in the

diets of the predators (based on proportional percent frequency

of occurrence of the species in the scats). If the proportional

availability of a prey category fell above or below the

confidence levels estimated for its occurrence in the diet, we

concluded that the prey item was actively avoided (above

confidence level) or selected (below confidence level).

RESULTS

We analyzed 42 cat, 90 fox, and 72 dingo scats. Although

fewer cat scats were discovered, the cumulative diversity curve

of prey items reached an asymptote for each predator species,

indicating that sampling effort was sufficient (Fig. 1). At least 9

small, 3 medium, and 3 large mammal species were identified

in the diets of the predators, as well as at least 15 other prey

types in the reptile, arthropod, bird, vegetation, and ‘‘other’’categories (Table 1). Trapping between September 2011 and

2012 yielded 87 captures of the long-haired rat (Rattusvillosissimus), 154 of the sandy inland mouse (Pseudomyshermannsburgensis), 338 of the spinifex hopping-mouse

(Notomys alexis), 18 of the house mouse (Mus musculus), 3

of the hairy-footed dunnart (Sminthopsis hirtipes), 14 of the

lesser hairy-footed dunnart (Sminthopsis youngsoni), and 3 of

the Wongai ningaui (Ningaui ridei); no Forrest’s mice

(Leggadina forresti), desert mice (Pseudomys desertor), or

stripe-faced dunnarts (Sminthopsis macroura) were captured,

despite occurring in the predator scats.

Hypothesis 1: small mammals will form the major prey ofthe dingo, fox, and cat.—Small mammals composed the main

prey of all predators in terms of frequency of occurrence and

volume (Table 1). R. villosissimus was represented most

frequently in the diets of the predators (25–49%) and

Sminthopsis species, the least (Table 1). For the cat, R.villosissimus was the single most important prey item in terms

of frequency of occurrence and volume (Table 1). P.hermannsburgensis, N. alexis, and L. forresti also had high

frequencies of occurrence in cat’s diet (16–26%). Of these

species, L. forresti had the greatest volumetric contribution

(13%; Table 1). For the red fox, N. alexis had the 2nd-greatest

volume and frequency of occurrence, followed by P.hermannsburgensis. After R. villosissimus, this latter species

also had the greatest frequency of occurrence and volume in

the diet of the dingo (Table 1).

Reptiles, birds, and arthropods had moderately high

frequencies of occurrence (33–42%) in the diet of the cat;

however, only birds contributed substantially (10%) to dietary

volume (Table 1). Medium- and large-sized mammals,

vegetation, and other prey types did not appear in the diet of

the feral cat (Table 1). Arthropods, birds, reptiles, and

vegetation occurred frequently in fox feces (20–51%) and also

composed a moderate volume of fox diet (8–11%). Large- and

medium-sized mammals also were present in the fox’s diet, but

were relatively minor components (Table 1). In the dingo’s

diet, reptiles, arthropods, large mammals, birds, and vegetation

occurred at high frequencies (19–51%). In terms of volume,

large mammals and reptiles contributed most substantially

(16% and 23%, respectively), and birds, arthropods, and

vegetation to a lesser extent (5–7%). Medium-sized mammals

and other prey types occurred at lower frequencies and

volumes (� 5%; Table 1). Large mammals were primarily

Macropus spp. Of the reptilian species, the inland bearded

dragon (Pogona vitticeps) was the most frequently occurring

species in fox and dingo scats; however, unidentified scincids

appeared at high frequencies in all predator scats. Medium-

FIG. 1.—Species accumulation curves showing the cumulative

diversity (Hk) of prey taxa in scats of dingo (Canis dingo; dashed line),

fox (Vulpes vulpes; solid black line), and cat (Felis catus; solid gray

line) with increasing sample size of scats (k) collected in the Simpson

Desert, central Australia.

0 Vol. 95, No. 6JOURNAL OF MAMMALOGY

sized mammals included 1 echidna (Tachyglossus aculeatus) in

a dingo scat.

Hypothesis 2: predator dietary preferences will be based onmammalian size classes.—There was separation between the

diets of the predators in terms of volumetric composition (Fig.

2; global R¼ 0.04, P¼ 0.013). The differences were between

the diets of the fox and dingo (R ¼ 0.103, P ¼ 0.001) but not

between those of the cat and dingo (R ¼ 0.034, P ¼ 0.109) or

the cat and fox (R ¼�0.086, P ¼ 0.998). Fox and dingo diets

deviated most in terms of their representation of small

mammals, vegetation, reptiles, and large mammals, with the

former 2 occurring more in fox diets and the latter 2 more in

the diet of dingoes (Table 2). There also were marked

differences between predators in the frequency of occurrence

of major prey groups in their diets. Cats ate small mammals

more often than did foxes and dingoes. Dingoes ate more

reptiles and large mammals than did foxes, but foxes ate more

small mammals than did dingoes (Table 3).

Hypothesis 3: cats will show stronger selectivity for certainsmall mammal species compared to foxes and dingoes.—

Notomys alexis was the most abundant prey species in the

trapping record, followed by P. hermannsburgensis and R.

TABLE 1.—A) Percent occurrence and B) percent volumetric composition of prey types in dingo (Canis dingo), fox (Vulpes vulpes), and cat

(Felis catus) scats collected in the Simpson Desert, central Australia, with average prey mass indicated in parentheses next to each of the small

mammal species. Percent volumetric composition of prey types is equivalent to the approximate sum of corresponding prey items, because

percentages for prey types and items were calculated separately and each rounded to a whole number. Percent occurrence of each prey type in

predator scats is not a summation of the corresponding prey items, because multiple prey items may have been observed in a single scat.

Prey item

Feral cat Red fox Dingo

A B A B A B

Small mammals 95 82 81 60 56 38

Muridae, unidentified rodent 12 3 8 1 10 1

Rattus villosissimus, long-haired rat (150 g) 49 40 34 30 25 19

Pseudomys desertor, desert mouse (25 g) 7 4 2 1 6 2

Pseudomys hermannsburgensis, sandy inland mouse (11 g) 26 10 19 8 19 6

Notomys spp. 19 9 26 16 15 5

Notomys alexis, spinifex hopping-mouse (30 g) 16 8 26 15 12 5

Leggadina forresti, Forrest’s mouse (20 g) 23 13 3 1 6 1

Mus musculus, house mouse (17 g) 0 0 4 2 1 , 1

Sminthopsis spp. 9 3 6 2 4 2

Sminthopsis youngsoni, lesser hairy-footed dunnart (9 g) 0 0 2 1 3 1

Sminthopsis hirtipes, hairy-footed dunnart (16 g) 5 1 1 , 1 0 0

Sminthopsis macroura, striped-faced dunnart (20 g) 2 1 1 1 1 1

Medium-sized mammals 0 0 4 1 7 5

Tachyglossus aculeatus, echidna 0 0 0 0 1 , 1

Felis catus, cat 0 0 2 , 1 1 1

Vulpes vulpes, fox 0 0 1 1 4 4

Large mammals 0 0 3 1 23 16

Macropus spp. 0 0 2 1 25 16

Macropus rufus, red kangaroo 0 0 0 0 7 5

Macropus robustus, common wallaroo 0 0 2 1 11 7

Camelus dromedarius, feral camel 0 0 1 , 1 1 , 1

Reptiles 33 5 30 8 51 23

Agamidae, unidentified dragon 2 , 1 1 4 3 1

Pogona vitticeps, inland bearded dragon (250 g) 0 0 10 4 32 16

Varanus spp. 0 0 3 1 4 2

Varanus gouldii, goanna 0 0 0 0 1 1

Scincidae, unidentified skinks 30 4 18 3 14 3

Serpentes, unidentified snakes 0 0 0 0 3 1

Testudines, unidentified turtles 0 0 0 0 1 , 1

Birds 33 10 21 8 22 5

Arthropods 42 4 51 11 51 6

Arthropoda, unidentified arthropods 42 4 42 10 51 7

Unidentified larvae 2 , 1 3 1 0 0

Vegetation 0 0 20 11 19 7

Unidentified vegetation 0 0 4 3 3 2

Seeds or fruit 0 0 13 6 11 3

Grass 0 0 4 2 6 2

Other 0 0 0 0 1 , 1

Rubbish 0 0 0 0 1 , 1

December 2014 0SPENCER ET AL.—PREY SELECTIVITY BY SYMPATRIC PREDATORS

villosissimus (Fig. 3). The abundance of these 3 rodent species

declined substantially from October 2011 to September 2012;

however, comparable proportions of N. alexis and P.hermannsburgensis were captured across all 4 trapping

periods (Fig. 4). R. villosissimus was captured only in

September and October 2011 (Fig. 4). All predators

underconsumed N. alexis relative to its abundance, and both

foxes and cats selected for R. villosissimus. Cats also consumed

L. forresti more than its availability in the environment. The

dingo did not show positive selection for any small mammal

species (Fig. 3).

DISCUSSION

Our study provides a snapshot in time of the diets and

selectivity for small mammal prey of 3 sympatric predators in

the Simpson Desert. The prey remains in predator scats

compared to the abundance of those prey in the field suggested

some preference for rodent species that were irrupting, or just

declining from peak numbers, at the time in the study area

(Greenville et al. 2013). This was not unexpected, and prey

preferences that we had anticipated to be based on size also

were shown by each of the 3 predators. In terms of small

mammal prey selection, the cat displayed the strongest

selection of small mammal species, including R. villosissimus

and L. forresti. As anticipated, the dingo displayed the least

selectivity, and did not positively select for any small mammal

species. On the other hand, the positive selection for R.villosissimus and apparent avoidance of N. alexis by cats

contradict our hypothesis, which states that selectivity by the

cat should be based on its mean preferred prey size. This

finding may indicate effective (or ineffective) antipredator

behavioral adaptations, although further studies are required to

substantiate this possibility. We discuss the results in more

detail based on each of the study’s hypotheses, and then

address the implications of high selectivity by predators for

rare prey species, in particular L. forresti, by feral cats, below.

Hypothesis 1: small mammals will form the major prey ofthe dingo, fox, and cat.—All predators preyed predominantly

on the rodent species that were irrupting at the time; that is, R.villosissimus, P. hermannsburgensis, and N. alexis (Greenville

et al. 2013). This result is consistent with prior studies that

have examined the diet of the 3 predators in other parts of the

Simpson Desert (Pavey et al. 2008a; Cupples et al. 2011) and

in other Australian arid environments that experience similar

rodent irruptions (Paltridge 2002). However, L. forresti also

contributed to a large part of the diet of the feral cat, despite not

being found in the trapping record. We discuss the implications

of this finding later.

Nonmammalian prey types including reptiles and arthropods

also were very important in sustaining predator populations in

the Simpson Desert. Previous studies have found reptiles to

contribute substantially to the diets of cats in arid Australian

areas (Bayly 1976; Paltridge et al. 1997; Newsome et al. 2014).

However, the frequency of occurrence of reptiles in the diet of

the fox in Australia is generally less than 15% (Marlow 1992)

and, in many studies, reptiles represent , 2% of prey items

consumed by dingoes (Corbett 1995). Prior work in the study

FIG. 2.—Nonmetric multidimensional scaling (nMDS) ordination

showing the dietary overlap of foxes (Vulpes vulpes; þ), dingoes

(Canis dingo; D), and cats (Felis catus; u) in the Simpson Desert,

central Australia, using volume composition data. Stress ¼ 0.12.

TABLE 2.—The results of SIMPER analysis of predator diets in the Simpson Desert, central Australia. Average similarity values show the mean

similarities between the contents of each scat for dingoes (Canis dingo) and foxes (Vulpes vulpes). Percentage contribution indicates the average

contributions that individual prey categories make to dissimilarity between diets of the 3 predators. Cumulative percent is the cumulative sum of

the percent contribution.

Predator species Average similarity Prey category Average dissimilarity % contribution Cumulative %

Cat 71.33 Small mammal 25.29 36.03 36.03

Fox 42.73 Reptile 12.79 18.22 54.24

Dingo 26.41 Large mammal 8.44 12.02 66.27

Vegetation 8.00 11.40 77.66

Arthropod 6.70 9.55 87.21

Bird 5.90 8.40 95.62

TABLE 3.—Significant results of logistic regression analyses of

predator diets in the Simpson Desert, central Australia, for different

prey types. Regression coefficients are shown, with standard errors in

parentheses.

Predator comparison Prey category Estimate Z P

Fox–cat Small mammal �1.69 (0.77) �2.20 0.028

Dingo–cat Small mammal �2.72 (0.76) �3.57 0.000

Fox–dingo Small mammal 1.03 (0.35) 2.94 0.003

Fox–dingo Large mammal �2.33 (0.64) �3.62 0.000

Fox–dingo Reptile �0.89 (0.33) �2.72 0.007

0 Vol. 95, No. 6JOURNAL OF MAMMALOGY

region has, nonetheless, found that reptiles can contribute

considerably to fox and dingo diets (Pavey et al. 2008a;

Cupples et al. 2011). In some arid environments, reptiles can

be the most important food item for cats, dingoes, and foxes,

especially when small mammals are sparse (Paltridge 2002).

Arthropods also may be particularly useful in arid environ-

ments because they contain high ratios of water and fat to body

mass (Konecny 1987). Insects are generally a major component

in the diets of foxes, and dingoes will feed opportunistically on

invertebrate prey (Saunders et al. 2004; Allen et al. 2012). In

general, arthropods occur relatively infrequently in the diet of

cats (Fitzgerald and Turner 2000), but their contribution may

increase when avian and small mammal prey become sparse

(Paltridge et al. 1997). In this region, invertebrate prey

occurred in similarly high proportions when small prey were

not as abundant (Cupples et al. 2011) and in much lower

proportions during booms in small mammal population

numbers (Pavey et al. 2008a).

Hypothesis 2: predator dietary preferences will be based onmammalian size classes.—Dingoes ate more large mammals

and fewer small mammals in comparison to foxes and,

especially, cats. This result is consistent with previous

studies in the region (Pavey et al. 2008a; Cupples et al.

2011); other studies also have observed similar partitioning by

prey size between foxes, cats, and dingoes (Paltridge 2002;

Mitchell and Banks 2005; Pavey et al. 2008a). Dingo scats also

contained a much higher frequency of reptilian prey items than

did fox scats. This result has been observed in previous desert

studies (Paltridge 2002; Pavey et al. 2008a), but in some cases

reptiles have occurred more frequently in fox scats (Cupples et

al. 2011). In the latter work, large mammals (specifically

macropods) occurred most frequently in the diet of dingoes and

were probably selected in preference to reptiles. In our study,

and in others with similar results, large mammals made a

relatively smaller appearance in the diet of the dingo. The

different frequencies of large mammals in dingo diets probably

FIG. 3.—Proportional representation of small mammal prey items in

the diet (based on percent frequency in scats) of a) feral cat (Feliscatus), b) red fox (Vulpes vulpes), c) dingo (Canis dingo) (mean 6

95% confidence interval [95% CI]; white bars) and in the environment

(based on abundance of individuals captured per hectare in traps; gray

bars), Simpson Desert, central Australia. Jacobs’ index values

(diamonds) indicate preference (. 0) and avoidance (, 0) of prey

species. Percent volumetric composition of prey types is equivalent to

the approximate sum of corresponding prey items, as percentages for

prey types and items were calculated separately and each rounded to a

whole number. Percent occurrence of each prey type in predator scats

is not a summation of the corresponding prey items, because multiple

prey items may have been observed in a single scat.

FIG. 4.—Average numbers of Notomys alexis (gray diamonds),

Rattus villosissimus (black diamonds), and Pseudomys hermannsbur-gensis (white diamonds) captured per hectare over 4 trapping periods

in the Simpson Desert, central Australia.

December 2014 0SPENCER ET AL.—PREY SELECTIVITY BY SYMPATRIC PREDATORS

reflect the availability of prey in each study, but could,

alternatively, reflect differences in pack size or structure;

dingoes catch larger mammals generally only if in packs

(Thomson 1992).

Feral cats showed the greatest preference for small

mammals, and there was a much higher frequency of

occurrence of this prey type in cat scats compared with fox

and dingo scats. The preference for small mammals, in

particular rodents, by the cat conforms to many studies that

have examined diets of cats in Australia (Paltridge et al. 1997).

However, in the only other study that has examined cat scats in

the study region, a similar composition of rodents (. 95%)

was found in both fox and cat diets (Pavey et al. 2008a). Our

results probably reflect a time when relatively fewer small

mammals were present compared with the study by Pavey et al.

(2008a). Foxes are less adept at stalking small vertebrate prey

than are cats (Bayly 1976). When rodent abundance is low, the

fox may be unable to successfully hunt enough of this prey

type to sustain itself, and so relatively fewer small mammals (in

comparison to the cat) will appear in its diet.

Hypothesis 3: cats will show stronger selectivity for certainsmall mammal species compared to foxes and dingoes.—Our

study is the first to compare selectivity of small mammalian

prey species by foxes, cats, and dingoes in this desert

environment. Our results indicate that, as predicted, cats

show the strongest selectivity for small mammal species of the

3 predators. However, selection of small mammal species,

including R. villosissimus and N. alexis, by cats was not based

on its optimum preferred weight range of prey species, as also

predicted in hypothesis 3. We discuss the positive selection of

R. villosissimus and apparent avoidance of N. alexis by cats,

below.

In earlier studies, R. villosissimus has been largely absent

from trapping records because it is an irruptive species that

increases only following flooding rains. Therefore, selection of

R. villosissimus by different predators, and compared with

other prey species, has not been previously examined. The high

proportion of R. villosissimus in the diet of cats is at first

surprising. Cats generally specialize (and hunt) animals

weighing less than 200 g, with a preference for mammals

weighing considerably less than this (Dickman 1996b). In

terms of its overall diet, this finding appeared to hold: cats

consumed no medium- or large-sized animals, ate small skinks

and arthropods, and, probably, small birds (we found no large

bird bones or feathers in cat scats). However, R. villosissimuscan weigh up to 280 g, well above cats’ preferred prey weight.

Nonetheless, selection of R. villosissimus also was seen in

another recent study, which found 44 of 73 cat stomachs to

contain the remains of this animal (Yip et al. 2014). This work

suggested that broadscale consumption of unusual prey types

probably occurs when those prey are very abundant and easy to

hunt (Fitzgerald and Turner 2000; Denny and Dickman 2010),

as may have occurred in the present study.

Notomys alexis was selected by all predators much less than

its abundance in the field might have suggested; it was more

abundant than R. villosissimus and any other small mammal

species in the trapping record. A study examining small

mammal selectivity by the letter-winged kite (E. scriptus) in

the Simpson Desert showed a similar pattern to our results,

because this bird consumed N. alexis considerably less in

proportion to its availability (Pavey et al. 2008a). The authors

suggested that kites took fewer N. alexis than were potentially

available because it is bipedal and likely to possess superior

sprint speed and less predictable escape behavior in the open

than sympatric quadrupedal rodents (see also Kotler et al.

1994; Spencer et al. 2014). R. villosissimus is quadrupedal and

does not have any specific adaptations for escaping predators.

Compared to N. alexis it is conspicuous, probably easy to hunt,

and yields a high energy return, which might explain why it

was selected so consistently by all predators. If correct,

predation (specifically by cats and to a lesser extent also by

foxes) could be important in driving population crashes of R.villosissimus in arid environments (Newsome and Corbett

1975), and in doing so more quickly than in other rodent

species (Fig. 4).

Population declines of R. villosissimus also may reflect

resource scarcity (Predavec and Dickman 1994). R. villosissi-mus can survive for only 13 days without green vegetation and

water (Baverstock 1976) and as animals become weak with

hunger, they are likely to be highly susceptible to predation.

During population declines of R. villosissimus it is also not

uncommon to find these rats intact and dead on the ground,

presumably having succumbed to starvation (Predavec and

Dickman 1994). Because predator scats sampled in this study

were produced at a time when R. villosissimus was declining in

the system, the scavanging of dead rats may in part explain the

high frequency of this species in predator scats, rather than just

the active predation of live rats. To ascertain the true effect of

predators on R. villosissimus, manipulation experiments that

involve the removal of these predators from the environment

are required.

The cat’s positive selection of R. villosissimus is interesting;

however, this rat is relatively abundant across the Australian

continent and is probably not threatened by feral cats. On the

other hand, L. forresti often is scarce even in apparently

suitable habitat. It is classified as vulnerable on Schedule 2 of

the Threatened Species Conservation Act 1995 in New South

Wales, and is suspected to be declining in abundance and

distribution in Queensland (Dickman et al. 2000). L. forrestialso appears to be uncommon in our study region, having been

captured very rarely in regular trapping conducted from 1990

to 2013. Selection by cats for L. forresti may reflect apparent

competition. That is, where large populations of predators are

subsidized by common or abundant prey, species that are

relatively scarce are subject to intensified per capita predation

pressure and can experience sharp population declines as a

result (Sinclair et al. 1998). If this is the case, positive selection

of L. forresti by the cat could be highly detrimental to the

species, especially during the population booms of small

mammal species such as R. villosissimus or N. alexis.

Leggadina forresti is not typically found in the spinifex

dunefields, which characterize our study area (McFarland

0 Vol. 95, No. 6JOURNAL OF MAMMALOGY

1992). Thus, the appearance of this rodent in predator scats

here was surprising. Although it is possible that L. forresticould expand its habitat use in exceptional conditions, such as

those following the high rainfall in 2011, there are 2 other

possibilities that might explain the absence of this species in

our trapping record. First, it is possible that the species was in

fact not present in our study area, and was incorrectly identified

in predator scats. Morphological analysis of hair may not have

reliably distinguished L. forresti from other rodents (Lobert et

al. 2001). However, this is unlikely. The hair of L. forresti is

quite distinct from that of the other rodents in the study area,

and we obtained independent confirmation of identity from a

professional hair analyst. In addition, we also found many bone

fragments in the predator scats that reliably confirmed this

species. Analyzing DNA in hair samples might further increase

the reliability of accurate identification, although current

technology is limited by high expense and the risk of

genotyping errors from low-quality DNA that is extracted

from hair samples that have gone through the gut of predators

(Piggott and Taylor 2003).

Second, and most likely, trapping may have missed small

populations of L. forresti to which cats had gained access,

especially because this rodent has been trapped around water

holes in the study region some 25–30 km from our study sites

(M. Tischler, Bush Heritage Australia, pers. comm.). In a study

of the diet of the letter-winged kite and barn owl (Tyto alba) in

the western Simpson Desert, L. forresti was observed in the

scats of both predators, although it was not recorded during

trapping (Pavey et al. 2008b). Pavey et al. (2008b) concluded

that trapping was probably not conducted in the correct

locations and that the predators were likely to be more effective

at sampling L. forresti than were their traps. Increasing the

trapping area, while also accounting for the home ranges of the

predators in the trapping design, should increase the accuracy

of population estimates and thus the reliability of selection

indexes.

Trapping efficiency also may have impacted the relative

abundance estimates for other rodent species, such as N. alexisand R. villosissimus. Smaller, faster N. alexis may be more

likely to fall into pitfall traps than the slower R. villosissimus,

which have a head–body length that can exceed the width of

our pitfall traps. Footprints of N. alexis and R. villosissimus are

relatively easy to identify, and the sand surrounding pitfall

traps provided a good indication of whether animals may have

detected and then avoided traps. There were, however, few

instances where prints of N. alexis and R. villosissimus were

identified around empty traps, suggesting that it is unlikely

there was a difference in trappability between the 2 species.

Further, previous surveys in our study area have revealed

similar proportions of different rodent species at sandplots,

which measured the activity of rodents using footprint

identification, and in pitfall traps (e.g., Dickman et al. 2011;

Tesoriero 2011).

Conclusions and future studies.—Our study shows that

small mammals form a large part of the diet of foxes and feral

cats in the study area, and highlights the possibility that these

predators could exert considerable impact on small mammals if

they occur at high density. The cat’s high selectivity for small

mammals in particular may drive prey populations to low

levels, and study of its impacts on L. forresti and R.villosissimus would be rewarding. In particular, experimental

removal of feral cats conducted during and following rodent

irruption events should provide the most useful data on the

impacts of this predator on rodent species present in the

system. The more limited selection by the 3 predators of N.alexis also warrants further study to uncover behaviors or other

strategies that allow this species to persist in the presence of

high predator activity (e.g., Spencer et al. 2014). Finally,

longer-term surveys of predator diet and selectivity, which

encompass multiple rodent irruption stages, will allow for the

extrapolation of potential predation risk effects past this single

rodent irruption event.

ACKNOWLEDGMENTS

Many volunteers assisted with scat collection in the field. Particular

thanks go to G. Story, who conducted the scat dissection for this

project. Thanks also go to the Desert Ecology Research Group,

including C. L. Beh, B. Tamayo, A. Greenville, and G. Wardle, who

provided constant logistical and moral support. Funding was provided

by the Australian Research Council and permission to use the study

sites was given by landowners, Bush Heritage Australia.

LITERATURE CITED

ALLEN, L. R., M. S. GOULLET, AND R. PALMER. 2012. The diet of the

dingo (Canis lupus dingo and hybrids) in north-eastern Australia: a

supplement to the paper of Brook and Kutt (2011). Rangeland

Journal 34:211–217.

BAVERSTOCK, P. 1976. Water balance and kidney function in four

species of Rattus from ecologically diverse environments. Austra-

lian Journal of Zoology 24:7–17.

BAYLY, C. P. 1976. Observations on the food of the feral cat (Felis

catus) in an arid environment. South Australian Naturalist 51:22–

24.

BRILLOUIN, L. N. 1956. Science and information theory. Academic

Press, New York.

BRUNNER, H., AND B. TRIGGS. 2002. Hair ID—an interactive tool for

identifying Australian mammalian hair. CSIRO Publishing, Mel-

bourne, Australia.

BURBIDGE, A. A., AND N. L. MCKENZIE. 1989. Patterns in the modern

decline of Western Australia’s vertebrate fauna: causes and

conservation implications. Biological Conservation 50:143–198.

BUREAU OF METEOROLOGY. 2013. Climate statistics for Australian

locations. http://www.bom.gov.au/climate/averages/tables/

cw_038003.shtml. Accessed 14 April 2013.

BYERS, C. R., R. K. STEINHORST, AND P. R. KRAUSMAN. 1984.

Clarification of a technique for analysis of utilization–availability

data. Journal of Wildlife Management 48:1050–1053.

CHAO, A., R. L. CHAZDON, R. K. COLWELL, AND T. J. SHEN. 2006.

Abundance-based similarity indices and their estimation when there

are unseen species in samples. Biometrics 62:361–371.

CHERRY, S. 1996. A comparison of confidence interval methods for

habitat use–availability studies. Journal of Wildlife Management

60:653–658.

December 2014 0SPENCER ET AL.—PREY SELECTIVITY BY SYMPATRIC PREDATORS

CLARKE, R. K., AND R. N. GORLEY. 2006. PRIMER v6.1.11: user

manual/tutorial. PRIMER-E Limited.

CLARKE, R. K., AND M. R. WARWICK. 1994. Similarity-based testing for

community pattern: the two-way layout with no replication. Marine

Biology 118:167–176.

COOPER, S. D., D. W. SMITH, AND J. R. BENCE. 1985. Prey selection by

freshwater predators with different foraging strategies. Canadian

Journal of Fisheries and Aquatic Sciences 42:1720–1732.

CORBETT, L. K. 1989. Assessing the diet of dingoes from feces: a

comparison of 3 methods. Journal of Wildlife Management 53:343–

346.

CORBETT, L. K. 1995. The dingo in Australia and Asia. University of

New South Wales Press, Sydney, Australia.

CROWTHER, M. S., M. FILIOS, N. COLMAN, AND M. LETNIC. 2014. An

updated description of the Australian dingo (Canis dingo Meyer,

1793). Journal of Zoology (London) 293:192–203.

CUPPLES, J. B., M. S. CROWTHER, G. L. STORY, AND M. LETNIC. 2011.

Dietary overlap and prey selectivity among sympatric carnivores:

could dingoes suppress foxes through competition for prey? Journal

of Mammalogy 92:590–600.

DENNY, E. A., AND C. R. DICKMAN. 2010. Review of cat ecology and

management strategies in Australia. Invasive Animals Cooperative

Research Centre, Canberra, Australia.

DICKMAN, C. R. 1996a. Impact of exotic generalist predators on the

native fauna of Australia. Wildlife Biology 2:185–195.

DICKMAN, C. R. 1996b. Overview of the impacts of feral cats on

Australian native fauna. Australian Nature Conservation Agency,

Canberra, Australia.

DICKMAN, C. R. 2009. House cats as predators in the Australian

environment: impacts and management. Human–Wildlife Conflicts

3:41–48.

DICKMAN, C. R., A. C. GREENVILLE, B. TAMAYO, AND G. M. WARDLE.

2011. Spatial dynamics of small mammals in central Australian

desert habitats: the role of drought refugia. Journal of Mammalogy

92:1193–1209.

DICKMAN, C. R., L. K.-P. LEUNG, AND S. M. VAN DYCK. 2000. Status,

ecological attributes and conservation of native rodents in Queens-

land. Wildlife Research 27:333–346.

FITZGERALD, B. M., AND D. C. TURNER. 2000. Hunting behaviour of

domestic cats and their impact on prey populations. Pp. 151–175 in

The domestic cat: the biology of its behaviour (D. C. Turner and P.

Bateson, eds.). Cambridge University Press, Cambridge, United

Kingdom.

FRIEND, G. R., G. T. SMITH, D. S. MITCHELL, AND C. R. DICKMAN. 1989.

Influence of pitfall and drift fence design on capture rates of small

vertebrates in semi-arid habitats of Western-Australia. Wildlife

Research 16:1–10.

GLEN, A. S., AND C. R. DICKMAN. 2006. Diet of the spotted-tailed quoll

(Dasyurus maculatus) in eastern Australia: effects of season, sex

and size. Journal of Zoology (London) 269:241–248.

GLEN, A. S., AND C. R. DICKMAN. 2008. Niche overlap between

marsupial and eutherian carnivores: does competition threaten the

endangered spotted-tailed quoll? Journal of Applied Ecology

45:700–707.

GLEN, A. S., M. PENNAY, C. R. DICKMAN, B. A. WINTLE, AND K. B.

FIRESTONE. 2011. Diets of sympatric native and introduced

carnivores in the Barrington Tops, eastern Australia. Austral

Ecology 36:290–296.

GREENVILLE, A. C., G. M. WARDLE, AND C. R. DICKMAN. 2013. Extreme

rainfall events predict irruptions of rat plagues in central Australia.

Austral Ecology 38:754–764.

JACOBS, J. 1974. Quantitative measurement of food selection.

Oecologia 14:413–417.

JANZEN, F. J., J. K. TUCKER, AND G. L. PAUKSTIS. 2000. Experimental

analysis of an early life-history stage: selection on size of hatchling

turtles. Ecology 81:2290–2304.

JOHNSON, C. N., AND S. WROE. 2003. Causes of extinction of

vertebrates during the Holocene of mainland Australia: arrival of

the dingo, or human impact? Holocene 13:941–948.

KARELL, P., N. LEHTOSALO, H. PIETIAINEN, AND J. E. BROMMER. 2010.

Ural owl predation on field voles and bank voles by size, sex and

reproductive state. Annales Zoologici Fennici 47:90–98.

KLARE, U., J. F. KAMLER, AND D. W. MACDONALD. 2011. A comparison

and critique of different scat-analysis methods for determining

carnivore diet. Mammal Review 41:294–312.

KLARE, U. N. N., J. F. KAMLER, U. T. E. STENKEWITZ, AND D. W.

MACDONALD. 2010. Diet, prey selection, and predation impact of

black-backed jackals in South Africa. Journal of Wildlife

Management 74:1030–1041.

KONECNY, M. J. 1987. Food habits and energetics of feral house cats in

the Galapagos Islands. Oikos 50:24–32.

KOTLER, B. P., J. S. BROWN, AND W. A. MITCHELL. 1994. The role of

predation in shaping the behavior, morphology and community

organization of desert rodents. Australian Journal of Zoology

42:449–466.

LETNIC, M., AND M. S. CROWTHER. 2013. Patterns in the abundance of

kangaroo populations in arid Australia are consistent with the

exploitation ecosystems hypothesis. Oikos 122:761–769.

LETNIC, M., AND C. R. DICKMAN. 2006. Boom means bust: interactions

between the El Nino/Southern Oscillation (ENSO), rainfall and the

processes threatening mammal species in arid Australia. Biodiver-

sity & Conservation 15:3847–3880.

LETNIC, M., B. TAMAYO, AND C. R. DICKMAN. 2005. The responses of

mammals to La Nina (El Nino Southern Oscillation)–associated

rainfall, predation, and wildfire in central Australia. Journal of

Mammalogy 86:689–703.

LOBERT, B. O., L. F. LUMSDEN, H. BRUNNER, AND B. TRIGGS. 2001. An

assessment of the accuracy and reliability of hair identification of

south-east Australian mammals. Wildlife Research 28:637–641.

MARLOW, N. J. 1992. The ecology of the introduced red fox (Vulpes

vulpes) in the arid zone. Ph.D. dissertation, University of New

South Wales, Sydney, Australia.

MARSACK, P. R., AND G. CAMPBELL. 1990. Feeding-behaviour and diet

of dingoes in the Nullarbor region, Western-Australia. Wildlife

Research 17:349–357.

MCFARLAND, D. 1992. Fauna of the Channel Country biogeographic

region, south west Queensland. Queensland National Parks and

Wildlife Service, Department of Environment and Heritage,

Brisbane, Queensland, Australia.

MCKENZIE, N. L., ET AL. 2007. Analysis of factors implicated in the

recent decline of Australia’s mammal fauna. Journal of Biogeog-

raphy 34:597–611.

MEDINA, F. M., R. GARCIA, AND M. NOGALES. 2006. Feeding ecology of

feral cats on a heterogeneous subtropical oceanic island (La Palma,

Canarian Archipelago). Acta Theriologica 51:75–83.

MENKHORST, P. W., AND F. KNIGHT. 2001. A field guide to the

mammals of Australia. Oxford University Press, Melbourne,

Australia.

MITCHELL, B. D., AND P. B. BANKS. 2005. Do wild dogs exclude foxes?

Evidence for competition from dietary and spatial overlaps. Austral

Ecology 30:581–591.

0 Vol. 95, No. 6JOURNAL OF MAMMALOGY

NEWSOME, A. E., AND L. K. CORBETT. 1975. Outbreaks of rodents in

semi-arid and arid Australia: causes, preventions, and evolutionary

considerations. Pp. 117–153 in Rodents and desert environments (I.

Prakash and P. K. Ghosh, eds.). Junk, The Hague, Netherlands.

NEWSOME, T. M., G. A. BALLARD, P. J. S. FLEMING, R. VAN DE VEN, G.

L. STORY, AND C. R. DICKMAN. 2014. Human-resource subsidies

alter the dietary preferences of a mammalian top-predator.

Oecologia 175:139–150.

OSKARSSON, M. C., C. F. KLUTSCH, U. BOONYAPRAKOB, A. WILTON, Y.

TANABE, AND P. SAVOLAINEN. 2012. Mitochondrial DNA data

indicate an introduction through mainland Southeast Asia for

Australian dingoes and Polynesian domestic dogs. Proceedings of

the Royal Society, B. Biological Sciences 279:967–974.

PALTRIDGE, R. 2002. The diets of cats, foxes and dingoes in relation to

prey availability in the Tanami Desert, Northern Territory. Wildlife

Research 29:389–403.

PALTRIDGE, R., D. GIBSON, AND G. EDWARDS. 1997. Diet of the feral cat

(Felis catus) in central Australia. Wildlife Research 24:67–76.

PAVEY, C. R., S. R. ELDRIDGE, AND M. HEYWOOD. 2008a. Population

dynamics and prey selection of native and introduced predators

during a rodent outbreak in arid Australia. Journal of Mammalogy

89:674–683.

PAVEY, C. R, J. GORMAN, AND M. HEYWOOD. 2008b. Dietary overlap

between the nocturnal letter-winged kite Elanus scriptus and barn

owl Tyto alba during a rodent outbreak in arid Australia. Journal of

Arid Environments 72:2282–2286.

PEARRE, S., AND R. MAASS. 1998. Trends in the prey size-based trophic

niches of feral and house cats Felis catus. Mammal Review 28:125–

139.

PIELOU, E. C. 1975. Ecological diversity. John Wiley & Sons, Inc.,

New York.

PIGGOTT, M. P., AND A. C. TAYLOR. 2003. Remote collection of animal

DNA and its applications in conservation management and

understanding the population biology of rare and cryptic species.

Wildlife Research 30:1–13.

PREDAVEC, M., AND C. R. DICKMAN. 1994. Population dynamics and

habitat use of the long-haired rat (Rattus villosissimus) in south-

western Queensland. Wildlife Research 21:1–9.

PURDIE, R. W. 1984. Land systems of the Simpson Desert region.

CSIRO Division of Water and Land Resources, Melbourne,

Australia.

R DEVELOPMENT CORE TEAM. 2012. R: a language and environment for

statistical computing. R Foundation for Statistical Computing,

Vienna, Austria. http://www.R-project.org/. Accessed XX Month

Year.

SACKS, B. N., S. K. BROWN, D. STEPHENS, N. C. PEDERSEN, J.-T. WU,

AND O. BERRY. 2013. Y chromosome analysis of dingoes and

Southeast Asian village dogs suggests a Neolithic continental

expansion from Southeast Asia followed by multiple Austronesian

dispersals. Molecular Biology and Evolution 39:1103–1118.

SALO, P., P. B. BANKS, C. R. DICKMAN, AND E. KORPIMAKI. 2010.

Predator manipulation experiments: impacts on populations of

terrestrial vertebrate prey. Ecological Monographs 80:531–546.

SALO, P., E. KORPIMAKI, P. B. BANKS, M. NORDSTROM, AND C. R.

DICKMAN. 2007. Alien predators are more dangerous than native

predators to prey populations. Proceedings of the Royal Society, B.

Biological Sciences 274:1237–1243.

SAUNDERS, G. R., M. J. BERGHOUT, B. J. KAY, B. TRIGGS, R. VAN DE

VEN, AND R. K. WINSTANLEY. 2004. The diet of foxes (Vulpesvulpes) in south-eastern Australia and the potential effects of rabbit

haemorrhagic disease. Wildlife Research 31:13–18.

SIH, A., G. ENGLUND, AND D. E. WOOSTER. 1998. Emergent impacts of

multiple predators on prey. Trends in Ecology & Evolution 13:350–

355.

SIKES, R. S., W. L. GANNON, and the Animal Care and Use Committee

of the American Society of Mammalogists. 2011. Guidelines of the

American Society of Mammalogists for the use of wild mammals in

research. Journal of Mammalogy 92:235–253.

SINCLAIR, A. R. E., R. P. PECH, C. R. DICKMAN, D. HIK, P. MAHON, AND

A. E. NEWSOME. 1998. Predicting effects of predation on

conservation of endangered prey. Conservation Biology 12:564–

575.

SPENCER, E. E., M. S. CROWTHER, AND C. R. DICKMAN. 2014. Risky

business: do native rodents use habitat and odor cues to manage

predation risk in Australian Deserts? PLoS ONE 9:e90566.

TESORIERO, S. 2011. Mammalian predator–prey interactions in arid

Australia. Honours thesis, The University of Sydney, Sydney,

Australia.

THOMSON, P. C. 1992. The behavioural ecology of dingoes in north-

western Australia. III. Hunting and feeding behaviour, and diet.

Wildlife Research 19:531–541.

TRIGGS, B. 1996. Tracks, scats and other traces: a field guide to

Australian mammals. Oxford University Press, Melbourne, Austra-

lia.

TRIGGS, B., H. BRUNNER, AND J. M. CULLEN. 1984. The food of fox, dog

and cat in Croajingalong National Park, south-eastern Victoria.

Australian Wildlife Research 11:491–499.

YIP, S. J. S., C. R. DICKMAN, E. A. DENNY, AND G. M. CRONIN. 2014.

Diet of the feral cat, Felis catus, in central Australian grassland

habitats: do cat attributes influence what they eat? Acta Theriolog-

ica 59:263–270.

Submitted 11 November 2013. Accepted 4 April 2014.

Associate Editor was Chris R. Pavey.

December 2014 0SPENCER ET AL.—PREY SELECTIVITY BY SYMPATRIC PREDATORS