energy and water use by invasive goats (capra hircus) in an australian rangeland, and a caution...

Comparative Biochemistry and Physiology, Part A 161 (2012) 216–229

Contents lists available at SciVerse ScienceDirect

Comparative Biochemistry and Physiology, Part A

j ourna l homepage: www.e lsev ie r .com/ locate /cbpa

Energy and water use by invasive goats (Capra hircus) in an Australianrangeland, and a caution against using broad-scale allometry topredict species-specific requirements

A.J. Munn a,b,⁎, C.E. Cooper c, B. Russell d, T.J. Dawson e, S.R. McLeod f, S.K. Maloney g

a Institute of Conservation Biology and Environmental Management, School of Biological Sciences, University of Wollongong, Australiab School of Biological Sciences, The University of Sydney, Australiac Department of Environment and Agriculture, Curtin University, Western Australia, Australiad Pest Management Unit, NSW Department of Environment, Climate Change and Water, Australiae School of Biological, Earth and Environmental Sciences, The University of New South Wales, Australiaf New South Wales Department of Industry and Investment, Orange Agricultural Institute, New South Wales, Australiag Physiology: School of Biomedical and Chemical Science, The University of Western Australia, Australia

⁎ Corresponding author at: Institute of ConservatioManagement, School of Biological Sciences, UniversTel.: +61 2 42214459.

E-mail address: [email protected] (A.J. Munn).

1095-6433/$ – see front matter © 2011 Elsevier Inc. Alldoi:10.1016/j.cbpa.2011.10.027

a b s t r a c t
a r t i c l e i n f o
Article history:Received 21 July 2011Received in revised form 28 October 2011Accepted 29 October 2011Available online 3 November 2011

Keywords:AllometryField metabolic rateWater turnoverGrazingInvasive species

Feral goats (Capra hircus) are ubiquitous across much of Australia's arid and semi-arid rangelands, wherethey compete with domestic stock, contribute to grazing pressure on fragile ecosystems, and have been im-plicated in the decline of several native marsupial herbivores. Understanding the success of feral goats inAustralia may provide insights into management strategies for this and other invasive herbivores. It hasbeen suggested that frugal use of energy and water contributes to the success of feral goats in Australia,but data on the energy and water use of free-ranging animals are lacking. We measured the field metabolicrate and water turnover rate of pregnant and non-pregnant feral goats in an Australian rangeland duringlate summer (dry season). Field metabolic rate of pregnant goats (601±37 kJ kg−0.73 d−1) was 1.3 timesthat of non-pregnant goats (456±24 kJ kg−0.73 d−1). The water turnover rate of pregnant goats (228±18 mL kg−0.79 d−1) was also 1.3 times that of non-pregnant goats (173±18 kg−0.79 d−1), but the differencewas not significant (P=0.07). There was no significant difference in estimated dry matter digestibility be-tween pregnant and non-pregnant goats (mean ca. 58%), blood or urine osmolality, or urine electrolyte con-centrations, indicating they were probably eating similar diets and were able to maintain osmohomeostasis.Overall, the metabolic and hygric physiology of non-pregnant goats conformed statistically to the predictionsfor non-marine, non-reproductive placental mammals according to both conventional and phylogeneticallyindependent analyses. That was despite the field metabolic rate and estimated dry matter intake of non-pregnant goats being only 60% of the predicted level. We suggest that general allometric analyses predictthe range of adaptive possibilities for mammals, but that specific adaptations, as present in goats, result inecologically significant departures from the average allometric curve. In the case of goats in the arid Austra-lian rangelands, predictions from the allometric regression would overestimate their grazing pressure byabout 40% with implications for the predicted impact on their local ecology.

© 2011 Elsevier Inc. All rights reserved.

1. Introduction

What makes invasive species successful? Putatively, invasive spe-cies share features that afford success in novel environments(Blackburn et al., 2009; van Kleunen et al., 2010; Zerebecki andSorte, 2011), particularly where they persist in the absence of preda-tors (e.g. Newsome et al., 2001). Features such as behavioural andphysiological plasticity that help to make a successful invader, may

n Biology and Environmentality of Wollongong, Australia.

rights reserved.

also help that species to withstand or adapt to changes in climate(Kolar and Lodge, 2001; Chown et al., 2007), and to adapt to manage-ment strategies aimed at controlling their population or impacts (e.g.manipulating water access to manage herbivores; Underhill et al.,2007; Fensham and Fairfax, 2008). Information on the resource re-quirements of invasive species offers an opportunity to evaluatetheir success relative to non-invasive introduced or native species.We investigate here the field metabolic rate and water turnoverrate of feral goats (Capra hircus) in an Australian rangeland.

Domestic goats were introduced to Australia with European set-tlement in the late 1700's (Parkes et al., 1996). Subsequently, escapedor released goats established themselves as one of the most signifi-cant invasive herbivores in Australia (Parkes et al., 1996; McLeod,

http://dx.doi.org/10.1016/j.cbpa.2011.10.027

mailto:[email protected]

http://dx.doi.org/10.1016/j.cbpa.2011.10.027

http://www.sciencedirect.com/science/journal/10956433

217A.J. Munn et al. / Comparative Biochemistry and Physiology, Part A 161 (2012) 216–229

2004; Coutts-Smith et al., 2007; West and Saunders, 2007). Feralgoats compete with domestic stock (sheep, Ovis aries and cattle, Bostaurus), contribute heavily to land degradation and overgrazing, andhave been implicated in the decline of native fauna such as theyellow-footed rock wallaby (Petrogale xanthopus; Wilson et al.,1976; Dawson and Ellis, 1979; Harrington, 1986; Lim and Giles,1987). The establishment of large and persistent populations of feralgoats has been coincident with control of the dingo (Canis lupusdingo), which has limited predation pressure that otherwise mightsuppress goat populations (Newsome et al., 2001). Further, a broadand flexible diet as a generalist herbivore may contribute to thegoats' success (Squires, 1980; Harrington, 1986). Anecdotal reportssuggest that goats can sustain reproductive output even duringprolonged drought when forage availability and quality are poor(Parkes et al., 1996). However, fundamental information on the fieldenergy and water requirements of feral goats in Australia is lacking.

We used the doubly labelled water method to investigate the fieldmetabolic rate (FMR) andwater turnover rate (WTR) of feral goats inha-biting a typical Australian rangeland.Wemeasured the FMR andWTR ofboth pregnant andnon-pregnant goats, aswehypothesise that pregnantgoats will have higher energy and water requirements and thus greaterecosystem impacts. We have also examined blood and urine concentra-tions and urine electrolyte concentrations of goats as indicators of diet,and estimated their diet digestibility to estimate the dry matter intake(DMI) required to meet their FMR. On a broader scale, we comparedthe FMR, estimated DMI and WTR of non-pregnant feral goats to thoseof non-reproductive, terrestrial placental mammals using allometricscaling, to determine whether the energy requirements of feral goatsdiffer from a ‘typical’ placental mammal, thereby testing the hypothesisthat feral goats have lower food (energy) and water requirements thanplacentals generally,whichmay contribute to the goats' success as an in-vasive herbivore in Australia's rangelands.We then evaluated the effica-cy of using allometric scaling to predict species-specific resourcerequirements and the implications of using this approach for predictingfiner-scale, local impacts of invasive or other species.

2. Materials and methods

2.1. Study site and climatic conditions

The study was conducted at Fowlers Gap (31°05′ S, 141°43′ E), theArid Zone Research Station of the University of New South Wales,112 kmnorth-east of BrokenHill, New SouthWales, Australia. The sta-tion covers approximately 39,200 ha and operates as a commercialsheep station. Vegetation at the study site is dominated by lowwoody shrubs (b1 m), chiefly of the family Chenopodiaceae. Topog-raphy on the western half of the station includes mainly hilly regionsof the Barrier Ranges (c. 300 m above sea level), whilst flood plains(140–170 m a.s.l.) cover the remainder of the property. Rainfall atthis site is variable, with a yearly average (±SEM) of 236.7±20.4 mm p.a. and a co-efficient of variation of 54% (1969–2007 inclu-sive; SILO Patched Point Dataset, Bureau of Meteorology and NHM

Table 1Body mass, doubly labelled water kinetics, and field metabolic rate (FMR) of non-pregnant

Body mass(kg)a

Body mass change(% d−1)

K 18O/2H

Non-pregnant 28.8±2.0 0.99±0.32 1.18±0.0Pregnant 42.7±1.5

(40.8±1.4)1.16±0.26 1.20±0.0

F 31.2 (25.8) 0.2 0.2P 0.001 0.68 0.648Mean (±SE) – 1.01±0.19 1.19±0.0

a Average of initial and final body masses, including foetal mass for pregnant animals (avb A sample size of n=4 was used here for pregnant animals as one individual had a NH:c Values in parenthesis are relative to foetus-free body mass.

QLD; data patched for 1971, and February and April 2000). Thisstudy was conducted during a mild late summer between the 25thFebruary and 12th March, 2008 (Table 1).

2.2. Study animals

Free-ranging feral goats (n=4 non-pregnant; n=5 pregnant)were mustered between 0600 h and 0800 h in the western part ofthe study site. The animals were moved to purpose-built holdingyards before being transferred via cage-trailer to a large (16 ha) en-closure. The enclosure had not been grazed for several years and veg-etation within the enclosure was markedly higher than outside theenclosure (A. Munn et al., pers. obs). Water was available within theenclosure via a water trough, from which all animals were observedto drink, and abundant shade was available via scattered acaciatrees and shrubs. The animals were acclimated to the enclosure for10 d prior to experimentation. The reproductive state of the goatswas determined post-mortem at the conclusion of the study. Of thefive pregnant females, all but one carried twins, and so the combinedfoetal body mass was used throughout analysis. Individual foetalmasses were used to estimate gestational age (McDonald et al., 1988).

2.3. Measurement of FMR and WTR

Wemeasured the fieldmetabolic rate (CO2 production; L d−1) andwater turnover rate (WTR; L d−1) of goats using the doubly labelledwater method (Lifson and McClintock, 1966; Speakman, 1997). Ani-mal numbers were limited both by the high cost of the doubly-labelled water for such large animals and the size of the enclosure.Goats were captured by muster and blood samples (ca. 4 mL) wereobtained from the jugular vein for the measurement of backgroundlevels of the isotopes. Goats were then injected intraperitoneallywith 0.3 g kg−1 18O (>98% enriched water) and 0.15 g kg−1 deuteri-um (2H; >95% enriched water) from separate syringes (isotopes fromRotem Industries, Israel) into the same location. The isotopes wereallowed to equilibrate with body water for 6–8 h before a secondblood sample (ca. 4 mL) was obtained from the jugular vein. Duringthe equilibration period the animals weremaintained in a small, shad-ed pen (ca. 10 m2) without access to feed or water. The goats werethen released into the enclosure and allowed to range freely for 8 d,after which they were mustered and held quietly in the shaded penuntil they were shot and a final blood sample obtained. Background,equilibration and final blood samples were analysed for 18O and 2Hby isotope ratio mass spectrometry after vacuum distillation to obtainpure water (Speakman, 1997; Metabolic Solutions, Nashua, NH, USA).

Pool sizes (NH or NO; moles) were estimated after Lifson andMcClintock (1966), Speakman (1997) and Gessaman et al. (2004) as

Nint ¼ Ids–Idistð Þ � Mdist=MWð Þ � Minj= Iint−Ibð Þ� �

=18:02; ð1Þ

where Nint = initial intercept pool size of the isotope in the animal,Ids = concentration of isotope in isotopically-labelled water solution,

(n=4) and pregnant (n=5) feral goats grazed together in rangeland in arid Australia.

Dilution space ratio(NH:NO)b

FMR(kJ d−1)

FMR(kJ kg−0.73 d−1)c

3 1.01±0.02 5278±343 455.9±24.23 1.04±0.01 9225±595 600.9±36.6

(615.5±36.5)1.5 26.2 9.7 (11.7)0.270 0.001 0.017 (0.01)

3 1.02±0.01 – –

erage foetal mass=1.9±0.3 kg).NO value markedly higher than all others (see Materials and methods text for details).

218 A.J. Munn et al. / Comparative Biochemistry and Physiology, Part A 161 (2012) 216–229

Idist = concentration of isotope in water distilled from body fluid,Mdist = mass of distilled water, MW = mass of isotopically-labelledwater used in dilution, Minj = mass of isotopically-labelled waterinjected into the animal, Iint=the intercept or equilibration concen-tration of isotope (18O or 2H) and Ib = concentration of isotope(18O or 2H) in water distilled from the background blood sample. Ini-tial dilution space ratios (NH:NO) were not significantly different be-tween pregnant and non-pregnant goats, and so a mean (±SE)group ratio of 1.02±0.01 was used to calculate WRT and FMR (seebelow). Of note, one pregnant animal had a NH:NO value of 1.15,and was considered sufficiently different to all others to exclude itfrom the group mean NH:NO; that individual's FMR and WTR werecalculated using its specific NH:NO.

Total body water (TBW; % initial live mass) was estimated fromthe dilution space (N) for 18O for all goats, because 2H dilution typi-cally overestimates body water in ruminants (Fancy et al., 1986). Be-cause of the potential for body water content to change throughoutthe experiment, particularly in pregnant animals (McDonald et al.,1988), we estimated WTR (rH2O; mol d−1) for linearly changingbody water contents according to:

rH2O ¼ kHN �H=R−displace� �

= f 1 � Xð Þ þ 1–Xð Þ½ �; ð2Þ

where, kH=2H flux (turnover rate) during the experiment (seeEq. (3)),NH = the mean isotope pool size calculated from 2H dilutionand for linearly changing body water (see Eqs. (4), (5) and (6)below), R-displace = mean group displacement ratio for bodywater pools, estimated from initial pool sizes for 2H and 18O (i.e.NH:NO; Midwood et al., 1994), f1=fractionation constant for 2H2O va-pour relative to 2H2O liquid (assumed to be 0.93; Lifson andMcClintock, 1966; Nagy and Costa, 1980; Speakman, 1997), X = esti-mate of the proportion of total water loss that is fractionated (as-sumed=0.25; Speakman, 1997).

Deuterium flux during the experiment (kH) was estimated as:

kH ¼ lnHint–lnHfinalð Þ=t; ð3Þ

where ln= natural log of initial (Hint) and final (Hfinal) concentrations(ppm) of 2H in body water after correction for background levels, andt = time (days).

The mean isotopic pool size (N) for 18O and 2Hwas calculated for alinearly changing body water pool according to:

N1 ¼ MB1 � Nint=MB1; ð4Þ

N2 ¼ MB2 � Nint=MB1; ð5Þ

N ¼ N1 þ N2ð Þ=2; ð6Þ

whereN1 andN2 are the initial andfinal isotopic pool sizes (i.e. for 18O or2H), MB1 and MB2 = body mass at the beginning and end of the experi-ment, Nint = initial isotope pool size (18O or 2H), assuming that bodywater content remained a constant fraction of body mass (i.e. Nint/MB1)for each animal (Nagy and Costa, 1980; Gessaman et al., 2004).

The production of CO2 (rCO2; mol d−1) was estimated using equa-tions validated for ruminants after Midwood et al. (1994) as:

rCO2 ¼ kO � NO� �

– rH2O � X � f 2ð Þ þ 1−Xð Þ � rH2O½ �� =2f 3; ð7Þ

where kO = 18O flux (turnover rate) during the experiment (seeEq. (3), substituting 2H for 18O),NO = the mean isotope pool size cal-culated from 18O dilution and for linearly changing body water(Eqs. (4)–(6)), rH2O=as per Eq. (2), X = estimate of the proportionof total water loss that is fractionated (assumed=0.25; Speakman,1997), f2 = fractionation constant for H2

18O vapour relative to H218O

liquid (assumed to be 0.99; Speakman, 1997), and f3 = the fraction-ation constant for H2

18O2 gas relative to H218O liquid (assumed to be

1.039; Lifson and McClintock, 1966; Speakman, 1997). Carbon diox-ide production was then converted to a field metabolic rate (FMR;kJ d−1) assuming energy equivalents of 21.7 kJ L−1 CO2 (Nagy et al.,1999).

2.4. Potential errors of DLW method in ruminants

The use of the doubly labelled water method in ruminants may becomplicated by the large amount of vegetative material in the gut,which provides a substrate for deuterium exchange with plant fibreand subsequent faecal loss, in addition to the incorporation of deute-rium as methane during methanogenic fermentation (Fancy et al.,1986; Midwood et al., 1989, 1993, 1994). Therefore, some of the deu-terium introduced as labelled water may be lost via avenues otherthan water, possibly elevating the estimation of the deuterium fluxand thus water flux (rH2O), and therefore underestimating the CO2

production via the difference between 18O and 2H fluxes (Fancyet al., 1986; Midwood et al., 1989, 1993, 1994). Because of these ave-nues CO2 production can be underestimated by between 3% and 12%from non-growing (stable body mass) ruminants (Fancy et al.,1986; Midwood et al., 1989, 1993, 1994). However, more recent val-idation of the use of deuterium-labelled water to measure WTR ofsheep and goats found no significant differences between isotope-kinetic predictions and those estimated from water intake (combinedwith preformed sources; Al-Ramamneh et al., 2010; see also Junghanset al., 1997). Moreover, errors of the DLW method described by theearlier validations in ruminants (e.g. Fancy et al., 1986; Midwoodet al., 1989, 1993, 1994) were within the ranges reported from valida-tion trials with non-ruminants (e.g. Sparling et al., 2008). Correctingfor deuterium losses in faeces and methane in free-ranging ruminantsis difficult and rarely attempted (but see Williams et al., 2001). Wecould not measure methane or faecal output from our goats, and sohave presented unadjusted values for WTR and FMR. We later discussthe likely impacts of deuterium loss through faeces and CH4 on ourresults, based on data for other ruminants (Midwood et al., 1993,1994).

2.5. Osmolalitiy of blood and urine and urine electrolytes

Urine samples were taken from the goat bladders immediately fol-lowing post-mortem evisceration. These samples were immediatelystored on ice in an insulated box and were frozen within one hourof collection. Urine sub-samples were later thawed and analysed forosmolality, along with fresh whole-blood samples collected viaheart puncture of deceased animals using lithium heparinised tubes.The osmolality of urine and blood was determined using a freezing-point depression osmometer (Gonotec Osmomat 030; Gallay Scientif-ic, Melbourne, Australia). Concentrations of electrolytes in urine, in-cluding sodium (Na+), potassium (K+), magnesium (Mg++) andcalcium (Ca++) were quantified using Inductively Coupled PlasmaOptical Emission Spectrometry (Perkin Elmer 5300DV ICP-OES; Syd-ney Analytical Services, Seven Hills, NSW, Australia), and concentra-tions of Cl− were determined using an Ag/AgS Ion Specific Electrode(Sydney Analytical Services). Electrolyte concentrations were notavailable for blood as the samples were used for labelled wateranalysis.

2.6. Diet digestibility

Apparent digestibilities of dry matter (DM) from the rumen wereestimated using manganese (Mn) as a naturally occurring indigestiblemarker (Nagy, 1977; Bersényi et al., 2002; see also Fadely et al., 1990and references therein). Absorption and secretion of Mn in the gut ofvertebrates are negligible and it has been used as a digestibility mark-er for numerous species (e.g. Nagy, 1977). Because the amount of Mnshould not change along the gut, the digestibility of the diet was


estimated using Mn concentrations from forestomach samples takenadjacent to the oesophageal opening at the cardia and comparedwith that in faeces collected as formed pellets from the distal colon.Digestibility was estimated according to:

Apparent digestibility ð%Þ¼ 1�Md

Mf

� �·100; ð8Þ

whereMd = concentration of Mn in the forestomach sample (per unitDM) and Mf = concentration of Mn faeces (per unit DM). Rumen (asabove) and faecal (distal colon) sub-samples (ca. 70 g wet mass)were collected at dissection then immediately stored on ice and fro-zen within 1 h. Forestomach material and faeces (ca. 70 g wet mass)were later dried at 60 °C to constant mass and then milled througha 1 mmmesh (Glen Creston c. 580 μ hammer mill, Glen Creston, London,UK). Sub-samples (0.6–1.0 g DM) of ground material were then digestedin nitric acid (10 mL; 70%) using a Milestone Microwave Digestion Sys-tem (Milestone MLS-1200 MEGA; Program 1) according to the manufac-turer's instructions. Digesta were then weighed, diluted to 25 mL withdeionised water, allowed to settle overnight, then the supernatant wasdrawn off and analysed for Mn content using an Inductively CoupledPlasma Atomic Emission Spectrometer (ICP-AES; Vista AX, Varian; CA,USA).

2.7. Dry matter intakes

Daily DMI (g d−1) was estimated as the amount of dry materialneeded to satisfy an animal's FMR (kJ d−1) according to the metabo-lisable energy content of that animal's diet (Emet; kJ g−1 DM), as FMR/Emet (Nagy et al., 1999). For herbivore diets the gross energy contentof herbaceous material generally ranges from 16.3–21.3 kJ g−1 DM(Robbins, 2001), which is comparable to that reported for perennialgrasses and saltbushes typical of our study site (Corbett, 1990; seealso Golley, 1961). Using our estimates of the dry matter digestibilityof goat rumen material (ca. 58%), we predicted the digestible energycontent of the goat's diet to be 12.2±0.4 kJ g−1 DM (assuming agross energy content of 21 kJ and that energy digestibility was com-parable to dry matter digestibility). Other studies at the same fieldsite found comparable digestibility coefficients using in-vitro acid-pepsin digestions of forbs, grasses and shrubs (range of means was9–14 kJ g−1 DM for all plant types from winter and summer;McLeod, 1996). However, neither our estimate of digestible energycontent nor the in-vitro estimates of energy digestibility (McLeod,1996) account for energy losses as urine or methane, which were un-known for our goats. Therefore, we have assumed that the metaboli-sable energy content of our goats' diets (Emet) was 11.5 kJ g−1 DM(Nagy et al., 1999), consistent with other investigations of ruminantmetabolisable energy content for forage (Nagy et al., 1999).

2.8. Allometry of FMR, DMI and WTR

The FMR and WTR of non-pregnant goats were compared withother adult, non-marine, non-reproductive (i.e. non-lactating/non-pregnant) placental mammals (n=64 species for FMR and n=37species WTR respectively; Appendix A1). These data include speciescovering four orders of magnitude of body mass. Data for WTR wasused only from species for which FMR was simultaneously measured.When data were available from more than one season, we selectedthe lowest values for FMR and WTR, typically using data collectedduring a species' dry season; FMRs reported during other seasons,or following rainfall, were generally higher (e.g. Mutze et al., 1991;Nagy and Gruchacz, 1994; Covell et al., 1996; Degen et al., 1991,1997; Williams et al., 1997, 2001; see Appendix A1). Importantly, allprevious allometric analyses of mammalian FMR (e.g. Anderson andJetz, 2005; Capellini et al., 2010; Speakman and Krόl, 2010) have in-cluded data from animals that were pregnant and/or lactating, were

growing juveniles, or were otherwise confounded (see AppendixA2). We have therefore collated the most conservative dataset forplacental mammal FMR and WTR, using minimal seasonal valueswhere known, which should provide a reasonable estimate of theminimum free-range resource requirements of placental mammalsgenerally.

Resource requirements of our goats were further compared withthose of eutherian mammals by converting our FMR dataset to grossDMIs, according to each species' diet and respective metabolisable ener-gy contents of their diet (i.e. Emet; Nagy et al., 1999; Appendix A1).Whilstthe metabolisable energy content of dry food probably varies consider-ably, broad patterns may be attributable to mammalian dietary guildsaccording to whether the species is an insectivore (Emet=18.7 kJ g−1

DM), a nectarivore (20.6 kJ g−1 DM), a carnivore (16.8 kJ g−1 DM), afrugivore (6.6 kJ g−1 DM), a granivore (16.9 kJ g−1 DM), a hindgut fer-menting herbivore (10 kJ g−1 DM), a foregut fermenting herbivore(mainly ruminants; 11.5 kJ g−1 DM), or an omnivore (14 kJ g−1 DM);after Nagy et al., 1999 (seeAppendix A1). To investigate the potential im-pact of aridity on FMR, WTR and DMI, we further classed animals as ei-ther desert or non-desert species (Nagy et al., 1999; Appendix A1).

The allometry of FMR, WTR and DMI was examined using conven-tional and phylogenetically independent linear regression of log10-transformed physiological data and body mass. Allometric relation-ships for desert or non-desert species were compared by ANCOVA.Allometrically-predicted FMR, WTR and DMI were determined fromthe conventional regressions using the maximum variance unbiasedestimate (MVUE) of Hayes and Shonkwiler (2006, 2007), and confor-mity to conventional and phylogenetically-independent allometric re-lationships was assessed using the 95% prediction limits for theregression (Cooper and Withers, 2006). For phylogenetically-independent analyses, data were rendered independent of phylogenyby autoregression (Cheverud and Dow, 1985; Rohlf, 2001) using themammalian phylogenetic tree of Bininda-Emonds et al. (2007). A phy-logenetic distance matrix was obtained from the original Nexus filepublished by Bininda-Emonds et al. (2007) using the APE module forthe package R. The species in our dataset were extracted from this dis-tance matrix. The nexus file (and the distance matrix generated fromit) provided branch lengths. We were unaware of any polytomies inthe database. Of note, our programme for phylogenetic analysis re-quired the phylogenetic tree to be in the format of a distance matrix,rather than the Newick format presented by Bininda-Emonds et al.(2007), so it was necessary to convert the Nexus file to a distance ma-trix. Extracting species required for this study from the overall mam-mal distance matrix was done using custom-written Visual Basic(V6) programmes (P.C. Withers). The strength (K*) and significance(P) of the phylogenetic signal for each phylogenetic variable, and forbody mass, were determined after Blomberg et al. (2003) andWithers et al. (2006).

2.9. Statistical analysis

For between-group comparisons of pregnant and non-pregnantgoats we used ANOVA, unless otherwise stated. Assumptions forANOVA were tested using the Kolmogorov–Smirnov test for normal-ity (α=0.05) and Levene's test for homogeneity of variances(α=0.05). Proportional data were arcsine transformed for analysis.Urine electrolyte concentrations (Na+, K+, Cl− and Ca++) werelog10 transformed to normalise their distribution. Arid and non-aridallometric relationships and those for FMR and DMI were comparedusing ANCOVA. Diet effects were examined using ANOVA on the re-siduals of the allometric relationship for each physiological variable.For those variables with a significant diet effect, the allometric resid-ual for goats was compared to those of other herbivores with a one-sample t-test. The significance of phylogenetic correction on the allo-metric relationships for FMR, DMI and WTR was determined using anF-test comparing the mean square error (MSE) for the conventional

Table 3Apparent digestibility of rumen dry matter (DM) and dry matter intakes (DMI)a esti-mated for non-pregnant (n=4) and pregnant (n=5) feral goats grazing together inrangeland in arid Australia.

Apparent DMdigestibility (%)

DMI(g d−1)

DMI(g kg−0.76 d−1)b

Non-pregnant 54.4±2.9 459±30 35.9±1.9Pregnant 61.2±2.3 802±52 46.4±2.9

(47.9±2.8)F 3.32 29 8.7 (10.8)P 0.111 0.001 0.03 (0.01)Mean (±SE) 58.2±2.1 – –

a Estimated DMI needed to meet daily FMR, assuming metabolisable energy contentof forage of 11.5 kJ g−1 DM (see text).

b Values in parenthesis are relative to foetus-free body mass.


allometric regression with the MSE for the phylogenetically correctedregression, after Withers et al. (2006). Statistical analyses were per-formed using Minitab 15 and StatistiXL V1.8. Autocorrelation (autore-gression), sub-sampling of the distance matrix of Bininda-Emondset al. (2007), calculation of the MVUE and determination of K* andthe significance of phylogenetic signals were achieved usingcustom-written Visual Basic (V6) programmes (P.C. Withers). Valuesare presented as mean±standard error unless stated otherwise.

3. Results

Pregnant (42.7±1.5 kg) goats were 50% heavier than the non-pregnant goats (28.8±2.0 kg; Table 2). There was no significant dif-ference in the average body mass change during the experiment(mean change +1.01±0.19%d−1; Table 1). Four of the five pregnantgoats carried twins. The youngest twin-pair weighed 400 g and 480 g,respectively, and were approximately 95 d old (animal G2),approaching the latter stages of the second trimester (i.e. 51–100 d;McDonald et al., 1988). The remaining foetuses ranged in ages from120 to 127 d (mass range 1000–1300 g), estimated to be mid-waythrough their third trimester (i.e. 101–150 d; McDonald et al., 1988).

Equilibration blood samples (~8 h after injection) indicated thatgoat body fluids were enriched above background levels to 388±12 ppm for 18O and 192±10 ppm for 2H. Final blood samples (ap-proximately 8 d following equilibrium) had concentrations of 18Oand 2H substantially above background levels for all goats (range65–111 ppm above background for 18O, and 39–92 ppm for 2H).The dilution space ratio and the kinetics of the labelled isotopes inthe body water of our goats were within ranges acceptable to estab-lish field metabolic and water turnover rates (Table 1; Speakman,1997).

The FMR of pregnant goats was significantly higher (approximate-ly 1.75 times) than the FMR of non-pregnant goats (Table 1), and itwas 1.32 times even after accounting for the difference in theirbody masses by allometry (i.e. kJ kg−0.73 d−1; Table 2). Similarly,the WTR of the pregnant goats was 1.3 times that of non-pregnantgoats after accounting for body mass differences (i.e. mL H2-

O kg−0.79 d−1), though this difference was not statistically significant(P=0.066; Table 2). Total body water content (61.5% of body mass)was not different between the pregnant and non-pregnant goats(Table 2). It is important to note that there is potential for incorpora-tion of labelled isotopes, particularly deuterium, into growing foe-tuses and as uterine water. However, amniotic fluid volume remainsconstant in natal goats between 110 and 130 d of pregnancy, and al-lantoic fluid increases only by around 10 g d−1 over the same period(McDonald et al., 1988). Thus, changes in uterine water contentduring our 10-day trial are unlikely to have had a large effect on theresults (mean body mass change of the pregnant animals was+1.16±0.26%d−1 of initial body mass, which was not significantlydifferent from the change in the non-pregnant animals; Table 1).

The apparent digestibility of dry matter from pregnant and non-pregnant goat rumens was not significantly different, averaging58.2% (Table 3). Therefore, consistent with their higher FMR, thepregnant goats in our study had daily DMIs that were significantly

Table 2Total body water content (TBW; %), rate of water turnover (WTR) of non-pregnant(n=4) and pregnant (n=5) feral goats grazing together in rangeland in arid Australia.

TBW (%) WTR (L d−1) WTR (mL kg−0.79 d−1)a

Non-pregnant 60.7±0.7 2.42±0.17 173.0±17.5Pregnant 62.0±1.1 4.43±0.39 228.1±17.7

(236.5±19.3)F 0.9 18.6 4.7 (5.6)P 0.367 0.004 0.066 (0.049)Mean (±SE) 61.5±0.7 – –

a Values in parenthesis are relative to foetus-free body mass.

higher than those on non-pregnant animals after accounting forbody mass differences via allometry (i.e. g DM kg−0.76 d−1;Table 3). There were no significant differences between the preg-nant and non-pregnant goats with respect to blood or urine osmolal-ity or urine electrolyte concentrations (Table 4).

The conventional allometric relationship for FMR for non-marineplacental mammals was highly significant (R2=0.948; F1,62=1121,Pb0.001), with a scaling exponent of 0.73. The FMR of non-pregnant goats was 60% of predicted, but it conformed statisticallyto the relationship, being well within the 95% prediction limits for afurther datum (Fig. 1). The significant allometric relationship forFMR remained (R2=0.883; F1,62=468, Pb0.001) after accountingfor the significant and exaggerated phylogenetic signal in both bodymass (K*=1.31, Pb0.001) and FMR (K*=1.21, Pb0.001), and thegoats still conformed to this relationship. Accounting for phylogenydid not reduce the variability in the allometric relationship for FMR(MSE conventional regression=0.044, MSE for phylogenetically in-dependent regression=0.047, F62,62=0.948, P=0.583).

Drymatter intake bynon-marine placentalmammals scaled similar-ly to FMR, with a DMI scaling exponent of 0.76 (slope comparison withFMR, F1,124=0.846, P=0.360). The DMIs of our goats conformed to theallometric relationship derived for non-marine placental-mammals(conventional analysis; R2=0.947; F1,62=1098, Pb0.001), being wellwithin the 95% prediction limits for a further datum (Fig. 1), despitethese non-pregnant goats having a DMI just 62% of that predicted. Thesignificant allometric relationship for DMI for non-marine placentalmammals remained after accounting for the significant and exaggerat-ed phylogenetic signal in DMI (R2=0.882; F1,62=463, Pb0.001;K*=1.14, Pb0.001), and our goats still conformed to this relationship.Accounting for phylogeny did not reduce the variability in the allome-tric relationship for DMI (MSE conventional regression=0.049, MSEfor phylogenetically independent regression=0.054, F62,62=0.911,P=0.642).

The WTRs of all goats conformed statistically to the generalplacental-mammal allometric relationship (R2=0.866; F1,36=232,Pb0.001), being well within the 95% prediction limits for a furtherdatum (Fig. 1). Goat WTR was 94% of the allometricaly predictedvalue. After accounting for a significant phylogenetic signal in WTR(K*=1.06, Pb0.001), the statistically significant allometric rela-tionship for scaling WTR with body mass remained (R2=0.707;F1,36=87, Pb0.001), and the goats conformed to this relationship(Fig. 1). Accounting for phylogeny did not significantly reduce thevariability in the allometric relationship for WTR (MSE conventionalregression=0.178, MSE for phylogenetically independent regres-sion=0.146, F35,35=1.226, P=0.275).

Therewas a significant difference in the slope of the conventional al-lometric relationship for FMR between desert (0.77±0.03) and non-desert (0.69±0.03) placental mammals (F1,60=4.72, P=0.034;Table 5), precluding examination of an elevation difference. The FMRof non-pregnant goats conformed equally well to both the desert andnon-desert datasets. When the data for FMR were phylogenetically

Table 4Mean (±SEM) urine electrolytes and urine and blood osmolality for non-pregnant (n=4) and pregnant (n=5) feral goats grazing together in rangeland in arid Australia.

Sodium(mmol L−1)

Potassium(mmol L−1)

Calcium(mmol L−1)

Magnesium(mmol L−1)

Chloride(mmol L−1)

Urine osmolality(mOsmol kg−1)

Blood osmolality(mOsmol kg−1)

Non-pregnant 82±11 20±5 0.52±0.20 5.8±1.3 47±17 476±83 337±4Pregnant 196±77 36±13 0.88±0.3 6.2±1.0 87±32 751±132 333±5F 2.88 1.57 1.12 0.1 1.33 1.78 0.00P 0.13 0.25 0.25 0.81 0.29 0.224 0.95Mean (±SE) 144±45 29±8 0.72±02 6.0±0.8 70±20 629±107 335±4


corrected there was no significant difference in the allometric slopebetween desert and non-desert placental mammals (commonslope=0.70; F1,60=2.63, P=0.11) but the intercepts differed(F1,61=5.44, P=0.023) with the desert species having a lowerphylogenetically-independent FMR than non-desert species. Our goatsconformed more closely to the desert-mammal line than the non-desert line (Fig. 1).

Desert mammals had a lower DMI than those frommoremesic en-vironments, with the regressions for desert and non-desert species

Fig. 1. Conventional (main) and phylogenetically independent (insets) allometric relationshipreproductive placental mammals (see Appendices A1 and A2 for raw data): open circles = aridgrey diamond=pregnant goat; dashed line = arid regression, solid line = non-arid regressionregression.

having the same slope (F1,60=3.0, P=0.09), but with the interceptfor desert species being significantly lower (F1,61=14.8, Pb0.001).The DMI of non-pregnant goats was 66% of that predicted for a desertspecies, and 62% of that predicted for a non-desert species, but wasnot statistically significantly lower than either the desert or non-desert groups. The significant difference between desert and non-desert mammals' intercepts for DMI remained when the datawere subject to phylogenetically-independent analysis (F1,61=8.95,P=0.004).

s for field metabolic rate, water turnover rate and dry matter intake for non-marine, non-zone species, black circles are non-arid zone species; grey triangle = non-pregnant goat,; insets are phylogenetically independent residuals with symbols as for the conventional

Table 5Allometric regressions for field metabolic rate (FMR), dry matter intake (DMI) andwater turnover rate (WTR) for non-marine, non-reproductive placental mammals, byconventional and phylogenetically independent methods. All regressions were statisti-cally significant (Pb0.001).

Slope Intercept R2

ConventionalFMR All (64) 0.73±0.022 0.65±0.057 0.95

Arid (30) 0.77±0.029 0.46±0.078 0.96Non-arid (32) 0.69±0.027 0.81±0.068 0.96

DMI All (64) 0.76±0.023 −0.56±0.060 0.95Arid (30) 0.80±0.03 −0.75±0.08 0.96Non-arid (32) 0.73±0.03 −0.40±0.070 0.96

WTR All (37) 0.79±0.05 −0.30±0.14 0.86Arid (22) 0.90±0.04 −0.75±0.11 0.96Non-arid (15) 0.66±0.08 0.24±0.22 0.83

Phylogentically independentFMR All (64) 0.69±0.032 0.002±0.027 0.88

Arid (30) 0.74±0.046 −0.064±0.040 0.90Non-arid (32) 0.65±0.041 0.052±0.034 0.89

DMI All (64) 0.73±0.034 0.004±0.029 0.88Arid (30) 0.79±0.050 −0.082±0.044 0.90Non-arid (32) 0.70±0.041 0.077±0.034 0.91

WTR All (37) 0.65±0.071 0.052±0.063 0.71Arid (22) 0.82±0.077 −0.072±0.068 0.85Non-arid (15) 0.50±0.106 0.168±0.099 0.63

Values are ±standard error, with sample sizes (N) in parenthesis.


The allometric relationships from conventional analysis of WTRfor desert and non-desert mammals were not significantly dif-ferent with regard to slope (F1,34=0.61 P=0.420) or intercept(F1,35=0.32, P=0.576). However, the slope of the regression forphylogenetically-independent analysis for the desert and non-desertmammalWTRs was significantly different (slope=0.82±0.08 for de-sert species, and 0.50±0.11 for non-desert species; F1,33=6.18,P=0.018), which precluded an analysis for differences in intercept.

Diet had no statistically significant effect on the FMR or WTR ofnon-marine placental mammals, either before or after phylogeneticanalysis. However, we found a statistically significant effect of dieton the DMI of non-marine placental mammals, both before(F6,57=2.85, P=0.017) and after accounting for phylogenetic history(F6,57=2.47, P=0.034). Notably, the DMI of our goatswas significant-ly lower than that predicted for other herbivores by both conventional(T19=4.9, Pb0.001) and phylogenetically independent analyses(T19=2.3, P=0.032).

4. Discussion

The success of feral goats surviving, indeed thriving, in arid andsemi-arid Australia does not seem to be related to a remarkablewater economy, at least when surface water is freely available as itwas in this study. The WTR we measured for non-pregnant goatswas as predicted for a non-desert species, and similar to the WTRsreported by Dawson et al. (1975) for goats at the same study site.On the basis of those results, Dawson et al. (1975) suggested thatgoats do not generally exhibit the physiological specialisations forwater economy that some African desert ungulates do. However,daily water turnover is dependent on many factors, including theavailability of free water, the distance to water from foraging sites,forage mineral content, forage water content, thermal challengesthat may increase an animal's water requirements, and urine concen-trating ability (Silanikove, 2000; Cain et al., 2006). Consequently,evaluating the daily water turnover of desert animals under non-water-stressed conditions is not likely to provide much insight forevaluating the benefits of adaptations related to water use. For exam-ple, the ability to withstand dehydration appears to be a key adapta-tion of large desert mammals (Cain et al., 2006) and studies on water-restricted, free-range animals would be useful. Nonetheless, we did

identify a significant difference in the allometric slope for WTR be-tween desert and non-desert mammals, indicating that smaller(b100 g), but not the larger, desert species exhibited lower dailyWTRs compared with non-desert species. Adaptation for water con-servation appears more important for smaller than larger species,presumably because smaller species have larger surface area tobody volume ratios and higher mass-specific metabolic rates, leadingto relatively greater cutaneous and respiratory water losses. It mustbe noted however that larger desert species are overly representedin the data set available. Further data on comparably sized mesic-zone mammals are required to provide a truly ‘general’ data set forcomparison with our goats or other species.

Daily energy and food (dry matter) requirement of our goats weretypical of desert-adapted animals, and importantly were significantlylower than the requirement of non-desert species. Feral goats aredescended from the bezoar, Capra aegagrus (Naderi et al., 2008),which is naturally found in arid habitats in the Middle East(Weinberg et al., 2008). Overall, reduced energy and therefore foodrequirements appear to be a general feature of desert species, includ-ing goats, presumably aiding their survival in low productivity habi-tats (Silanikove, 2000). Therefore, the success of feral goats in theAustralian arid zone is probably attributable to a combination of fea-tures that support their low energy and food requirements, includingflexible feeding behaviours, and their ability to select the most nutri-tious and digestible plant parts from a variety of forages (e.g. buds,leaves, fruits and flowers; Hoppe et al., 1977; Huston, 1978; Warrenet al., 1984; Harrington, 1986; Lu, 1988).

Our data on apparent dry matter digestibility of rumen material(ca. 58%) suggests that the goats were selecting good quality diets,and achieving digestibility comparable with high-quality diets fed incaptivity (Freudenberger and Hume, 1992). Similarities in blood andurine osmolality and urine electrolyte concentrations of our goats tovalues reported for free-range goats at the same study site (Dawsonet al., 1975; Dawson and Ellis, 1996), suggest that the vegetationeaten by our goats was of similar composition to that ingested bytruly free-range animals. Both the free range goats of Dawson et al.(1975) and those from our study were capable of maintainingosmo-homeostasis. Our data for the WTR, FMR and DMI of feralgoats are therefore likely to be broadly applicable to goats throughoutthe rangelands, where they subsist mainly on trees and shrubs, and tolesser extent flat- and round-leaf chenopods (Dawson et al., 1975;Dawson and Ellis, 1996). Of note, the water-content of the availablevegetation may also impact food selection and energy intake bygoats, for example by altering their preference towards forage withhigher water, rather than energy, content. But the goats in ourstudy had ad libitum access to water. This is not unusual for free-range goats at our study site, or indeed across much of Australia'seastern rangelands where the density of artificial watering pointsfor stock is high. In fact, in Australia's sheep rangelands it is difficultto move beyond 10-km of free water (Landsberg et al., 1997), andwater-point spacing of 2–3 km is not unusual in stocked regions(James et al., 1999).

As mixed-feeders, feral goats can adapt readily to seasonal andgeographical variations in diet availability (Lu, 1988), and their toler-ance of bitter plants (Bell, 1959) enables them to feed on vegetationthat is unpalatable to other herbivores. Moreover, the proliferationof artificial watering points (Fensham and Fairfax, 2008) and the re-moval of dingoes (Newsome et al., 2001) from much of Australia'sarid and semi-arid regions, thereby supporting and protecting pasto-ral stock, may have released goats from water limitation and preda-tion pressure, respectively. It is likely that a combination of all ofthese factors explains the feral goat's success as an invasive speciesin the Australian rangelands.

Whilst the FMR and DMI we measured in non-pregnant goatswere only ca. 60% of those predicted for a placental mammal (con-ventional analysis), they were not statistically lower than predicted


by allometry (Fig. 1). Such an outcome raises issues about the utilityof using the allometry of placental FMR for making ecological predic-tions. For example, using the general placental FMR-allometry to pre-dict the energy and food requirements of feral goats in an Australianrangeland would overestimate their local grazing pressure. This hasmajor implications for using allometric datasets to infer or predictmicro-ecological phenomena from a macro-ecological perspective(see also conclusions of Capellini et al., 2010). Interestingly, account-ing for phylogeny did not improve the predictive power of our allo-metric relationships, despite significant phylogenetic signals in bodymass and physiological parameters. This suggests that the phyloge-netic signal in body mass and physiological traits has a similar pat-tern. Accounting for phylogeny for allometric relationships for arange of physiological variables (basal metabolic rate, body tempera-ture, thermal conductance and evaporative water loss) also failed toreduce the variability of these relationships in another mammalgroup, the marsupials (Withers et al., 2006). In these examples, pre-dictions from conventional allometric analyses were at least as robustas those from allometric analyses that account for phylogenetichistory.

Whilst we constrained our analysis to the most conservative dataavailable, our FMRs, DMIs and WTRs were collated for species from abroad range of habitats and circumstances. Notably, these includemany animals from extreme habitats (e.g. desert-dwelling Arabianoryx Oryx leucoryx, Williams et al., 2001) and/or with extreme life-styles (e.g. three-toed sloth Bradypus variegates, Nagy andMontgomery, 1980; namib desert golden mole Eremitalpa grantinamibensis, Seymour et al., 1998). Consequently, the prediction limitsdescribed by the allometry of FMR and DMI (and to a lesser extentWTR) may well represent boundaries for what is physiologically orecologically possible for non-reproductive placental mammals. Ifthat is indeed the case, then fundamentally no species will statistical-ly be different from the allometric regressions (see Cooper andWithers, 2006), and it may not be possible for any species to differstatistically from the current dataset.

Because the full dataset represents the range of physiology that ispossible, it then becomes informative to determine the basis of thevariability making up those possibilities. A more meaningful compar-ison is revealed by the separate allometries for desert and non-desertspecies. The different scaling exponents for FMR between the desertand non-desert species hampers conventional comparisons of theirallometry, but when the influence of phylogenetic history is removed,desert-adapted species have FMRs significantly lower than non-desert species (Fig. 1). Further, the FMR that we measured in goatsresident in the arid rangelands conformed better to the desert rela-tionship than the non-desert relationship. These data suggest thatferal goats in Australia are similar to desert-adapted placentals gener-ally with regards to their energy requirements. However, compari-sons of resource use by feral goats with different but sympatricspecies under comparable environmental conditions will likely pro-vide better insights for predicting and managing local herbivore graz-ing pressures on Australia's rangelands.

The field metabolic rate of our non-pregnant goats(456 kJ kg−0.73 d−1; Table 3) was comparable with that reportedfor other goat breeds under free-range or near free-range conditions(c.f. 455–550 kJ kg−0.73 d−1; Animut et al., 2005; Lachica andAguilera, 2003, 2005). This level of daily energy use was less thanhalf that measured in domestic Merino sheep (956 kJ kg−0.73 d−1)grazed at the same location and under comparable environmentalconditions (Munn et al., 2009). Sheep grazing for wool and meat pro-duction is the dominant industry in eastern Australia's rangelands.Our data suggest that the grazing pressures imposed by sheep areconsiderably higher than those by feral goats, but the reasons forthis are unclear. The higher energy turnover of the Merino sheepcompared with goats could be related to differences in their minimalor maintenance metabolic rates, differences in their activity levels or

the energetic cost of activity, or to differences in their productivitystate. Merino sheep are bred for wool production, which is an energycost not present in feral goats. Further studies comparing goats withMerino sheep under more stringent dietary intakes (e.g. at mainte-nance rations) would help distinguish how these species partition en-ergy turnover and how this relates to their free-range FMRs andWTRs.

It is noted that the FMR of Merino sheep measured by Munn et al.(2009) was adjusted for deuterium losses in CH4, assuming produc-tions of 0.422 L CH4 kg−1 body mass d−1 (Midwood et al., 1989)and by reducing WTR by 1.02 g d−1 for each litre of CH4 produced(Midwood et al., 1989). Similar adjustment would have increasedthe FMR of our non-pregnant goats by 3%, comparable with CH4-re-lated errors reported for other ruminants (Fancy et al., 1986;Midwood et al., 1993, 1994). Nonetheless, CH4 output is heavily de-pendent on diet composition, which was unknown for our goats,and the 1.03 correction factor (i.e. +3%) could lead to an overesti-mate of goat FMR if the CH4 production of goats was less thansheep. Trees and shrubs (Dawson et al., 1975) typically containmore tannin than commercial concentrates or legume forages (asused in the validation studies on sheep by Midwood et al., 1989),and tannins reduce CH4 output from goats (Puchala et al., 2005).Even using a 1.03 correction factor, the FMR of our goats was withinthe range of data from other arid-adapted ungulates (Fig. 1).

Deuterium losses via faecal solids can also complicate compari-sons of ruminant FMR (Midwood et al., 1989, 1993, 1994; Williamset al., 2001). For ruminants, faecal-deuterium losses derived from iso-tope exchanges with bulky vegetable matter (mainly in the rumen)may lead to overestimates of water flux by a further 3–5% abovethose associated with CH4. The range of errors quantified via respi-rometry validations is from b1% to ca. 15% (Fancy et al., 1986;Midwood et al., 1993, 1994). Therefore, assuming that the combinedlosses of CH4 and faecal-deuterium resulted in an underestimationof our reported CO2 production by a liberal value of 10% (i.e. 5% forCH4 and a further 5% for faecal losses; Fancy et al., 1986; Midwoodet al., 1989; 1993, 1994), then the FMR for our non-pregnant feralgoats may have been as high as 502 kJ kg−0.73 d−1. This value isstill within the range of other goat breeds, and is just 52% of theFMR of domestic Merino sheep grazed at the same site (Munn et al.,2009). Thus, even allowing for the maximum errors in the technique,it remains that goats have an energy requirement typical of desert-adapted placentals.

The daily water turnover rates (WTR) of our goats were compara-ble with those measured in free-ranging goats at the same study siteand under similar environmental conditions (i.e. ca. 3.0 L d−1;Dawson et al., 1975). Further, the WTR of our non-pregnant goatswas comparable with that predicted for a general placental mammalof equivalent body mass (Fig. 1). More striking was that the WTR ofour non-pregnant goats (173 mL kg−0.79 d−1) was just 33% of theWTR of non-pregnant Merino sheep grazed at the same site undercomparable environmental conditions (Munn et al., 2009). Differ-ences between the water use of our non-reproductive feral goatsand Merino sheep are further accentuated if their WTRs are adjustedusing crude corrections for CH4- and faecal-deuterium losses, assum-ing that the raw WTRs for these animals are overestimated by up to10% (errors from validations studies range from 2% to 10%;Midwood et al., 1989, 1993, 1994; Haggarty et al., 1998). Therefore,assuming a water-flux correction factor of 0.9 (to correct an overesti-mate of 10%), the WTR of our non-pregnant goats would be156 mL kg−0.79 d−1, around 84% that predicted for a general placen-tal and just 30% that of Merino sheep grazed at the same site (i.e.sheep=521 mL kg−0.79 d−1, after Munn et al., 2009).

Whilst analysis of dry-goats provides an insight into adaptationsthat might enhance frugal resource-use to support the persistenceof invasive species, that persistence requires population stabilisationor expansion and the energy and water requirements of these


animals under more intense life history stages, such as during repro-duction, will better quantify their environmental impact. Anecdotalreports indicate that in Australia's arid rangelands, feral goats main-tain breeding even during prolonged drought (Parkes et al., 1996;see also Silanikove, 2000). Such an observation suggests that goatshave lower energy and/or water requirements for reproduction com-pared with other species. However, we found that the additional en-ergy requirement of pregnant goats relative to non-pregnant goatswas comparable to the difference reported for other species (i.e.1.2–1.3 times higher in pregnant animals, Robbins, 2001; Table 2).The pregnant goats in our study were mostly carrying 3rd trimestertwins, and the 3rd trimester is the peak of energy expenditure duringpregnancy in placental mammals (e.g. Goldberg et al., 1993; Pekinset al., 1998). The higher energy requirements of pregnant mothersat that stage are associated with numerous factors, including foetalgrowth, elevated basal metabolic rates, elevated maintenance costsfor placental and uterine tissues, and increased activity costs associat-ed with carrying additional mass (Gittleman and Thompson, 1988;Schoeller and Fijeld, 1991; Goldberg et al., 1993; Robbins, 2001).That theWTRs of our pregnant goats were not statistically significant-ly higher than those of non-pregnant goats was likely a Type II errorand lack of power as a consequence of small samples size (n=4 ver-sus n=5; P=0.066). The 1.3 times higher WTR of the pregnant goatswas very likely of biological significance, suggesting that access towater may be important for reproductive animals compared withdry females. Data on the FMR and WTR of lactating goats would beuseful in appreciating the costs of reproduction in feral goats andhow this might influence their population dynamics and survival inAustralia's rangelands.

5. Conclusions

We caution against using broad-scale allometry of FMR to predictlocal-scale species-specific impacts or requirements. This approachhas been used in other studies, for example to predict the impactsof feral cats on seabirds (Keitt et al., 2002), to infer prey communitydynamics as driven by predicted wolf FMRs (Gazzola et al., 2007), to

Body mass (BM; g), field metabolic rate (FMR; kJ d−1), dry matter intake (DMI; g d−1), watF = frugivore, G = granivore, H = hindgut fermenting herbivore, H (R) = foregut fermentimarine placental mammals, collated from non-reproductive animals under free-range or neused when available.

Genus species Common name BM(g)

FMR(kJ d−1)

Saccopteryx bilineata Sac-winged bat 7.9 16.2Sorex araneus Common shrew 8.19 58.1Plecotus auritus Brown long-eared bat 8.5 27.6Glossophaga commissarisi Commissaris's long-tongued bat 8.7 45.7Gerbillus henleyi Pygmy gerbil 9.7 22.2Anoura caudifer Flower-visiting bat 11.5 51.9Macrotus californicus California leaf-nosed bat

(Big eared bat)12.9 22.1

Peromyscus crinitus Canyon mouse 13.4 39.3Mus domesticus (musculus) Wild house mouse 15.5 57.1Clethrionomys rutilus Northern red-backed vole 16 48.96Syconycteris australis Southern blossom bat 17.36 76.87Perognathus formosus Pocket mouse 17.7 40.5Apodemus sylvaticus Wood mice 18.1 64.4Carollia brevicauda Silky short-tailed bat 18.2 50Peromyscus maniculatus Deer mouse 18.4 46.3Peromyscus leucopus White-footed deer mouse 19.4 36.5Eremitalpa granti (namibensis) Namib desert golden mole 20.67 12.46Gerbillus allenbyi Allenby's gerbil 22.8 35.6

Appendix A1

evaluate the ecology of restoration for mammalian assemblages(Gorman, 2007), and to assess macroecological phenomena such asthe energy equivalence rule in birds (Russo et al., 2003). Because al-lometry averages a broad range of adaptive possibilities, using allom-etry to make species-specific predictions is unlikely to be biologicallyor ecologically relevant, at least with regard to local phenomena. Inour case the potential grazing pressure of feral goats would havebeen overestimated by 40% if we relied on allometric predictions oftheir energy and food requirements. Consequently, broad-scale al-lometries of animal resource use may not be an appropriate tool forevaluating adaptive management protocols, either for invasive spe-cies or for other situations such as recovery programmes for endan-gered species. We further question the usefulness of broad-scaleallometry for predicting species-specific responses to climate changesor imposed animal management strategies (e.g. water-point closure).In short, there is no substitute for comparative, on-ground studies.

Acknowledgements

We thank Fowlers Gap Arid Zone Research Station, The University ofNew SouthWales, and particularly Gary Dowling for his field assistance.We thank Andrew Lothian for field assistance. Thanks to Alexander Riekand Ken Nagy for discussion of the DLW method in ruminants. Thanksto Jonathan Gilbert, Don Bradshaw and Jenny Bryce for providing rawdata for the body masses of animals for which they measured FMRand WTR. Thanks to P.C. Withers for access to software and discussionof the phylogenetic methods used herein. This project was funded inpart by The NSW Department of Environment, Climate Change andWater, and was partly supported by ARC Grant (LP0668879) to A.J.M.with Professors Chris Dickman and Michael Thompson (School of Bio-logical Sciences, University of Sydney) and with support from NSW De-partment of Environment and Climate Change Kangaroo ManagementProgram (N. Payne), the SA Department for Environment and Heritage(L. Farroway), the WA Department of Environment and Conservation(P. Mawson), the NSW Western Catchment Management Authorityand the NSW Department of Primary Industries. Animal ethics approv-al: UNSW ACEC 06/85A.

er turnover rate (WTR; mL d−1), diet (I = insectivore, N = nectarivore, C = carnivore,ng herbivore, and O = omnivore), and habitat (D = desert, ND = non-desert) for non-ar free-range conditions; the most conservative seasonal data for FMRs and WTRs were

DMI(g d−1)

WTR(mL d−1)

Diet Habitat Reference

0.9 I ND Voigt et al. (2001)3.1 20.24 I ND Ochocińska and Taylor (2005)1.5 I ND Speakman and Racey (1987)2.2 18.5 N ND Voigt et al. (2006)1.3 1.1 G D Degen et al. (1997)2.5 13.4 N ND von Helversen and Reyer (1984)1.2 2.46 I D Bell et al. (1986)

2.8 O D Mullen (1971a)4.1 3.74 O D Mutze et al. (1991)4.9 5 H ND Holleman et al. (1982)3.7 30.91 N ND Geiser and Coburn (1999)2.4 G D Mullen and Chew (1973)3.8 7.5 G ND Corp et al. (1999)7.6 29.1 F ND Voigt et al. (2006)3.3 O D Hayes (1989)2.6 O ND Randolph (1980)0.7 2.27 I D Seymour et al. (1998)2.1 2.78 G D Degen et al. (1992)

(continued)

Genus species Common name BM(g)

FMR(kJ d−1)

DMI(g d−1)

WTR(mL d−1)

Diet Habitat Reference

Microtus agrestis Field vole 26.5 72.7 7.1 H ND Meerlo et al. (1997)Gerbillus pyramidum Greater egyptian gerbil 31.8 45.2 2.7 3.24 G D Degen et al. (1992)Pseudomys albocinereus Ash-grey mouse 32.6 62.2 4.4 O ND Nagy (1987) (after Nagy

and Morris, Pers. Obs.)Dipodomys merriami Desert-dwelling kangaroo rat 33.9 44.0 2.6 3.1 G D Nagy and Gruchacz (1994)Acomys cahirinus Common spiny mouse 38.27 51.7 3.7 5.04 O D Degen et al. (1986)Pseudomys nanus Barrow Island mouse 40.9 25.5 2.6 5.27 H D Bradshaw et al. (1994#)Sekeetamys calurus Bushy-tailed jird 41.23 44 3.1 5.89 O D Degen et al. (1986)Microgale dobsoni Dobson's shrew tenrec 42.6 77.1 5.5 I ND Stephenson et al. (1994)Microgale talazaci Talazac's shrew tenrec 42.8 66.5 4.8 I ND Stephenson et al. (1994)Zyzomys argurus Rock rats 43.78 14.3 1.0 5.4 O D Bradshaw et al. (1994#)Acomys russatus Golden spiny mouse 45.04 47.8 3.4 5.65 O D Degen et al. (1986)Dipodomys microps Chisel-toothed kangaroo rat 56.96 89.2 6.4 O D Mullen (1971b)Praeomys natalensis Multi-mammate mouse 57.3 86.6 6.2 O ND Green and Rowe-Rowe (1987)Microcebus murinus Grey mouse lemur 65.6 109.2 7.8 16.2 O ND Schmid and Speakman (2000)Meriones crassus Sundevall's Jird 77.6 55.7 3.3 3.81 G D Degen et al. (1997)Phyllostomus hastatus Greater spear-nosed bat 80.8 146 10.42857143 O ND Kunz et al. (1998)Ammospermophilus leucurus White-tailed antelope squirrel 86.25 97.5 7.0 O D Karasov (1981)Tamias striatus Eastern chipmunk 96.3 143 10.2 O ND Randolph (1980)Thomomys bottae Botta's pocket gopher 103.8 127.4 12.7 4.67 H ND Gettinger (1984)Heliophobius argenteocinereus Silvery mole rat 161 121.8 12.18 H ND Zelova et al. (2011)Psammomys obesus Fat sand rat 165.6 146.3 14.6 30.2 H D Degen et al. (1991)Octodon degus Degus 182.2 125 12.5 H D Bozinovic et al. (2003)Spermophilus saturatus Cascade golden-mantled

ground squirrel256 226.5 22.7 H ND Kenagy et al. (1989)

Tamiasciurus hudsonicus Red Squirrel 322 347 18.9 G ND Bryce et al. (2001#)Cavia magna Greater guinea pig 466 381.6 38.2 132.4 H ND Künkele et al. (2005)Sciurus carolinensis Eastern grey squirrel 588 574 31.2 G ND Bryce et al. (2001)Suricata suricatta Meerkat 697 468 25.0 I D Scantlebury et al. (2002)Bassariscus astutus Ringtailed cat 752 472 28.1 113.3 C D Chevalier (1989)Lepilemur ruficaudatus Red-tailed sportive lemur 774.2 411.5 41.2 H ND Drack et al. (1999)Vulpes cana Blanford's fox 936.5 627.5 33.6 112.4 I D Geffen et al. (1992)Martes americana American Martin 975 709 50.6 O ND Gilbert et al. (2009#)Vulpes rueppelli Rüppell's fox 1750 1014.7 72.5 123 O D Williams et al. (2002)Vulpes velox Swift fox 1990 1488 106.3 205 O ND Covell et al. (1996)Marmota flaviventrus Yellow bellied marmot 3190 2434 243.4 H ND Salsbury and Armitage (1994)Bradypus variegatus Three-toed sloth 4385 611.9 61.2 154 H ND Nagy and Montgomery (1980)Vulpes vulpes Red fox 5597 1681 120.1 251 O ND Winstanley et al. (2003)Alouatta palliata Mantled howler monkey 7116 2542 254.2 H D Nagy and Milton (1979)Proteles cristata Aardwolf 8543 1845 98.7 292 I D Williams et al. (1997)Lycaon pictus African wild dog 25,170 15,300 910.7 C D Gorman et al. (1998)Antidorcas marsupialis Antelope Springbok 42,100 13,100 1139.1 3180 H (R) D Nagy and Knight (1994)Odocoileus hemionus Mule deer 45,400 26,700 2321.7 5500 H (R) ND Nagy (1987) (after Nagy

and Jacobson, Pers. Obs.);see also Nagy et al. (1990)

Lama pacos Alpaca 48,000 14,050 1221.7 3850 H (R) D Riek et al. (2007)Ovis aries Sheep 50,200 16,664 1449.0 11,500 H (R) D Munn et al. (2009)Rangifer tarandus Reindeer 61,000 15,980 1389.6 H (R) ND Gotaas et al. (2000)Oryx leucoryx Arabian oryx 81,500 11,081 963.6 1310 H (R) D Williams et al. (2001)Cervus elaphus Red deer 107,300 24,050 2091.3 11,963 H (R) ND Haggarty et al. (1998)

Note: #Relevant body masses supplied by authors.

Appendix A1 (continued)

Published reports for field metabolic and water turnover rates for marine (M) and non-marine placental mammals (D = desert, ND = non-desert) omitted from our allometricanalyses and a priori justifications for their omission.

Genus species Common name Habitat Omission justification Reference

Pipistrellus pipistrellus Pipistrelle ND Pregnant or lactating Racey and Speakman (1987);Speakman (1997)

Myotis lucifugus Little brown bat ND Pregnant or lactating Kurta et al. (1987)Eptesicus fuscus Big brown bat ND Pregnant or lactating Kurta et al. (1990)Mus domesticas feral house mouse ND Body mass not specified Rowe-Rowe et al. (1989)Microtus pennsylvanicus Meadow vole ND Not free range, 25 m2 enclosure,

high density; home range=400–800 m2 (Getz, 1961)

Berteaux et al. (1996)

Lemmus trimucronatus Brown lemming ND n=1 male, n=2 females, likelypregnant and each carryingtransmitter—not validated for load

Peterson et al. (1976)

Arvicola terrestris Water vole ND Authors caution that re-breathingof isotopes may have occurred

Grenot et al. (1984)

Vulpes macrotis Kit fix ND Mixed sexes and reproductive states,specific body masses not available

Girard (2001)

(continued on next page)

Appendix A2


(continued)

Genus species Common name Habitat Omission justification Reference

Lepus californicus Black-tailed jackrabbit D Authors note DLW results werenot comparable with measuredfeed intakes, may not be sufficientlyfree-range in a 300 m2 enclosure,home ranges=1–3 km2 (Smith, 1990)

Shoemaker et al. (1976)

Arctocephalus gazella Antarctic fur seal M Lactating Costa et al. (1989);Costa et al. (1985);Costa and Trillmich (1988)

Arctocephalus galapagoensis Galapagos fur seal M Lactating Costa and Trillmich (1988)Callorhinus ursinus Northern fur seal M Lactating and pups Costa et al. (1985); Costa and

Gentry (1986)Neophoca cinerea Australian sea lion M Lactating Costa and Gales (2003)Zalophus californicus California sea lion M Lactating Costa et al. (1985)Phocarctos hookeri New Zealand sea lion M Lactating Costa and Gales (2000)Mirounga angustirostris Northern elephant seal M Pups Kretzmann et al. (1993)Phoca vitulina Harbour seal (common seal) M N=1 adult male# Reilly and Fedak (1991)Odobenus rosmarus Warlrus M N=2 adult males# Acquarone et al. (2006)

#We have omitted these data because there is no evidence that males of such large species are representative of both sexes, and because the scaling of FMR in marine and terrestrialmammals may differ (Speakman and Krόl, 2010).

Appendix A2 (continued)


Appendices A1 and A2 ReferencesAcquarone M., Born E.W. and Speakman J.R. 2006. Field metabolic

rates of walrus (Odobenus rosmarus) measured by the doubly labelledwater method. Aquat Anim 32: 363–369.

Bell G. P., Bartholomew G.A. and Nagy K.A. 1986. The roles of ener-getics, water economy, foraging behavior, and geothermal refugia inthe distribution of the bat, Macrotus californicus. J Comp Physiol B156: 44 1–450.

Berteaux D., Thomas D.W., Bergeron J.M. and Lapierre H.L. 1996.Repeatability of daily field metabolic rate in female meadow voles(Microtus pennsylvanicus). Funct Ecol 10: 751–759.

Bozinovic F., Gallardo P.A., Visser G.H. and Cortés A. 2003. Seasonalacclimatization in water flux rate, urine osmolality and kidney waterchannels in free-living degus: molecular mechanisms, physiologicalprocesses and ecological implications. J Exp Biol 206: 2959–2966.

Bradshaw S.D., Morris K.D., Dickman C.R., Withers P.C. andMurphyD. 1994. Field metabolism and turnover in the golden bandicoot(Isoodan auratus) and other small mammals from Barrow Island,Western Australia. Austr Zool 42: 29–41.

Bryce J.M., Speakman J.R., Johnson P.J. and Macdonald D.W. 2001.Competition between Eurasian red and introduced eastern greysquirrels: the energetic significance of body-mass differences. ProcRoyal Soc London B 268: 1731–1736.

Chevalier C.D. 1989. Field energetics and water balance of desert-dwelling ringtail cats Bassariscus astutus (Carnivora Procyonidae). AmZool 29: A827.

Corp N., Gorman M.L., Speakman J.R. 1999. Daily energy expendi-ture of free-living male Wood Mice in different habitats and seasons.Funct Ecol 13: 585–593.

Costa D.P. and Gales N.J. 2000. Foraging energetic and diving be-haviour of lactating New Zealand Sea Lions, Phocarctos hookeri. J ExpBiol 203: 3655–3665.

Costa D.P. and Gales N.J. 2003. Energetics of a benthic diver: sea-sonal foraging ecology of the Australian Sea Lion, Neophoca cinerea.Ecol Monogr 73: 27–43.

Costa D.P. and Gentry R.L. 1986. Free-ranging energetics of north-ern fur seals. In: R.L. Gentry and G.L. Kooyman, eds. Fur seals: mater-nal strategies on land and at sea. Princeton University Press,Princeton, N.J., Pp. 79–101.

Costa D.P., Croxall J.P. and Duck. C.D. 1989. Foraging energetics ofAntarctic fur seals in relation to changes in prey availability. Ecology70: 596–606.

Costa D., Thorson P., Feldkamp S., Gentry R., DeLong R., AntonellisG. and Croxall J. 1985. At sea foraging energetics of three species ofpinniped. Fed Proc 44: 1000.

Costa D. P. and Trillmich F. 1988. Mass changes and metabolismduring the perinatal fast: a comparison between Antarctic and Gala-pagos fur seals. Physiol Zool 61: 160–169.

Covell D.F., Miller D.S. and Karasov W.H. 1996. Cost of locomotionand daily energy expenditure by free-living swift foxes (Vulpes velox):a seasonal comparison. Can J Zool 74: 283–290.

Degen A. A., Kam M., Hazan A. and Nagy K.A. 1986. Energy expen-diture and water flux in three sympatric desert rodents. J Anim Ecol55: 421–429.

Degen A.A., Khokhlova I.S., Kam M. and Nagy K.A. 1997. Body size,granivory and seasonal dietary shifts in desert gerbilline rodents.Funct Ecol 11: 53–59.

Degen A.A., Hazan A., Kam M. and Nagy K.A. 1991. Seasonal waterinflux and energy expenditure of free-living fat sand rats. J Mammal72:652–657.

Degen A.A., Pinshow B. and Kam M. 1992. Field metabolic ratesand water influxes of two sympatric Gerbillidae: Gerbillus allenbyiand G. pyramidum. Oecologia 90: 586–590.

Drack S., Ortmann S., BührmannN., Schmid J.,Warren R.D., HeldmaierG. and Ganzhorn J.U. 1999. Fieldmetabolic rate and the cost of ranging ofthe red-tailed sportive lemur (Lepilemur ruficaudatus). In: Newdirectionsin lemur studies, B. Rakotosamimanana, H. Rasaminanana, S.M. Good-man, J.U. Ganzhorn (eds), Kluwer Academic/Plenum Publishers, NewYork, Pp. 83–90.

Geffen E., Degen A.A., Kam M., Hefner R., Nagy K.A. 1992. Daily en-ergy expenditure and water flux of free-living Blanford's foxes(Vulpes cana), a small desert carnivore. J Anim Ecol 61: 611–617.

Geiser F. and Coburn D.K. 1999. Field metabolic rates and wateruptake in the blossom-bat Syconycteris australis (Megachiroptera). JComp Physiol B 169: 133–138.

Gettinger, R.D. 1984. Energy and water metabolism of free- rang-ing pocket gophers, Thomomys bottae. Ecology 65: 740–751.

Getz, L.L. 1961. Home Ranges, territoriality, and movement of themeadow vole. J Mammal 42(1): 24–36.

Gilbert J.H., Zollner P.A., Green A.K., Wright J.L., KarasovW.H. 2009.Seasonal field metabolic rates of American martens in Wisconsin. AmMidl Naturalist 162: 327–334.

Girard I. 2001. Field cost of activity in the Kit Fox, Vulpes macrotis.Physiol Biochemical Zool 74: 191–202.

Gotaas G., Milne E., Haggarty P. and Tyler N.J.C. 2000. Energy ex-penditure of free-living reindeer estimated by the doubly labelledwater method. Rangifer 20: 211–219.

Gorman M.L., Mills M.G., Raath J.P. and Speakman J.R. 1998. Highhunting costs make African wild dogs vulnerable to kleptoparasitismby hyaenas. Nature 391: 479–81.


Green B. and Rowe-Rowe D.T. (1987). Water and energy metabo-lism in free-living multi-mammate mice (Praomys natalensis) duringsummer. S Afr Zool 22: 14–17.

Grenot C., Pascal M., Buscarlet L., Francaz J.M. and Sellami M. 1984.Water andenergy balance in thewater vole (Arvicola terrestris Sherman)in the laboratory and in the field (Haut-Doubs, France). Comp BiochemPhysiol A 78: 185–196.

Haggarty P., Robinson J.J., Ashton J., Milne E., Adam C.L., Kyle C.E.,Christie S.L. and Midwood A.J. 1998. Estimation of energy expenditurein free-living red deer (Cervus elaphus) with the doubly-labelledwater method. Br J Nutr 80: 263–272.

Hayes J.P. 1989. Field and maximal metabolic rates of deer mice(Peromyscus maniculatus) at low and high altitudes. Physiol Zool 62:732–744.

Holleman D.F., White R.G. and Feist D.D. 1982. Seasonal energyand water metabolism in free-living Alaskan voles. J Mammal 63:293–296.

Karasov W. H. 1981. Daily energy expenditure and the cost of ac-tivity in a free-living mammal. Oecologia 51: 253–259.

Kenagy G.J., Sharbaugh S.M. and Nagy K.A. 1989. Annual cycle ofenergy and time expenditure in a golden-mantled ground squirrelpopulation. Oecologia 78: 269–282.

Kretzmann M.B., Costa D.P. and Le Boeuf B.J. 1993. Maternal ener-gy investment in the elephant seal pups: evidence for sexual equality.Am Nat 141: 466–480.

Künkele J., Kraus C. and Trillmich F. 2005. Does the unusual lifehistory of the precocial cavy (Cavia magna) translate into an excep-tional field metabolic rate? Physiol Biochem Zool 78: 48–54.

Kunz T.H., Robson S.K. and Nagy K.A. 1998. Economy of haremmaintenance in the greater spear-nosed bat, Phyllostomus hastatus. JMammal 79: 631–42.

Kurta A., Johnson K.A. and Kunz T.H. 1987. Oxygen consumptionand body temperature of female little brown bats (Myotis lucifugus)under simulated roost conditions. Physiol Zool 60: 386–397.

Kurta A., Kunz T.H. and Nagy K.A. 1990. Energetics and water fluxof free-ranging big brown bats (Eptesicus fuscus) during pregnancyand lactation. J Mammal 71: 59–65.

Meerlo P., Bolle L., Visser G.H., Masman D. and Daan S. 1997. Basalmetabolic rate in relation to body composition and daily energy ex-penditure in the field vole,Microtus agrestis. Physiol Zool 70: 362–369.

Mullen R. K. 1971a. Energymetabolism of Peromyscus crinitus in itsnatural environment. J Mammal 52: 633–635.

Mullen R.K. 1971b. Energy metabolism and body water turnoverrates of two species of free-living kangaroo rats, Dipodomys merriamiand Dipodomys microps. Comp Biochem Physiol A 39: 379–390.

Mullen R. K. and Chew R.M. 1973. Estimating the energy metabo-lism of free-living Perognathus formosus: a comparison of direct andindirect methods. Ecology 54: 633–637.

Munn A.J., Dawson T.J., McLeod S.R., Croft D.B., ThompsonM.B. andDickman C.R. 2009. Field metabolic rate and water turnover of redkangaroos and sheep in an arid rangeland: an empirically-deriveddry-sheep-equivalent for kangaroos. Austr Zool 57: 23–28.

Mutze G.J., Green B. and Newgrain K. 1991. Water flux and energyuse in wild house mice (Mus domesticus) and the impact of seasonalaridity on breeding and population levels. Oecologia 88: 529–38.

Nagy K.A. 1987. Field metabolic rate and food requirement scalingin mammals and birds. Ecol Monogr 57: 111–128.

Nagy K.A. and Gruchacz M.J. 1994. Seasonal water and energy me-tabolism of the desert dwelling kangaroo rat (Dipodomys merriami).Physiol Zool 67: 1461–1478.

Nagy K.A. and Knight M.H. 1994. Energy, water, and food use byspringbok antelope (Antidorcas marsupialis) in the Kalahari Desert. JMammal 75: 860–872.

Nagy K.A. and Milton K. 1979. Energy metabolism and food con-sumption by wild howler monkeys (Alouatta palliata). Ecology 60:475–480.

Nagy K.A. and Montgomery G.G. 1980. Field metabolic rate, waterflux and food consumption in three-toed sloths (Bradypus variegatus).J Mammal 61: 465–472.

Nagy K.A., Sanson G.D. and Jacobsen N.K. 1990. Comparative fieldenergetics of two macropod marsupials and a ruminant. Wildl Res17: 591–600.

Ochocińska D. and Taylor J.R.E. 2005. Living at the physiologicallimits: field and maximum metabolic rates of the common shrew(Sorex araneus). Physiol Biochem Zool 78: 808–818.

Peterson R.M., Jr., Batzli G.O. and Banks E.M. 1976. Activity and en-ergetics of the brown lemming in its natural habitat. Arctic Alpine Res8: 131–138.

Racey P. A. and Speakman J. R. 1987. The energetic costs of preg-nancy and lactation in heterothermic bats. Symp Zool Soc London,57: 107–125.

Randolph J.C. 1980. Daily energy metabolism of two rodents (Per-omyscus leucopus and Tamias striatus) in their natural environment.Physiol Zool 53: 70–81.

Reilly J.J. and Fedak M.A. 1991. Rates of water turnover and energyexpenditure of free-living male common seals (Phoca vitulina). J Zool223: 461–468.

Riek A., Van Der Sluijs L. and Gerken M. 2007. Measuring the ener-gy expenditure and water flux in free-ranging Alpacas (Lama pacos)in the Peruvian Andes using doubly labelled water technique. J ExpZool A 307: 667–675.

Rowe-Rowe D.T., Green B. and Crafford J.E. 1989. Estimated impactof feral house mice on Sub-Antarctic invertebrates at Marion Island.Pol Biol 9: 457–460.

Salsbury C.M. and Armitage K.B. 1994. Home-range size and ex-ploratory excursions of adult, male yellow-bellied marmots. J Mam-mal 75: 648–656.

Scantlebury M., Russell A.F., McIlrath G.M., Speakman J.R. andClutton-Brock T.H. 2002. The energetics of lactation in cooperativelybreeding Meerkats Suricata suricatta. Proc Royal Soc London B 269:2147–2153.

Schmid J. and Speakman J.R. 2000. Daily energy expenditure of thegrey mouse lemur (Microcebus murinus): a small primate that usestorpor. J Comp Physiol B 170: 633–641.

Seymour R.S., Withers P.C. andWeathers W.W. 1998. Energetics ofburrowing, running, and free-living in the Namib Desert golden mole(Eremitalpa namibensis). J Zool 244: 107–117.

Shoemaker V. H., Nagy K.A. and Costa W.R. 1976. Energy utiliza-tion and temperature regulation by jackrabbits (Lepus californicus)in the Mojave Desert. Physiol Zool 49: 364–375.

Smith G.W. 1990. Home range and activity patterns of black-tailedjackrabbits. Great Basin Nat 50: 249–256.

Speakman J. 1997. Factors influencing the daily energy expendi-ture of small mammals. Proc Nutr Soc 56: 1119–1136.

Speakman J.R. and Krόl E. 2010. Maximal heat dissipation capacityand hyperthermia risk: neglected key factors in the ecology of endo-therms. J Anim Ecol 79: 726–746.

Speakman J.R. and Racey P.A. 1987. The equilibrium concentrationof oxygen-18 in body water: implications for the accuracy of thedoubly-labelled water technique and a potential newmethod of mea-suring RQ in free-living animals. J Theoret Biol 127: 79–96.

Stephenson P.J., Speakman J.R. and Racey P.A. 1994. Field metabol-ic rate in two species of shrew-tenrec, Microgale dobsoni and M. tala-zaci. Comp Biochem Physiol A 107: 283–287.

Voigt C.C., Helversen O.V., Michener R. and Kunz T.H. 2001. Theeconomics of harem maintenance in the sac-winged bat, Saccopteryxbilineata (Emballonuridae). Behav Ecol SocioBiol 50: 31–36.

Voigt C., Kelm D.H. and Visser G.H. 2006. Field metabolic rates ofphytophagous bats: do pollination strategies of plants make life ofnectar-feeders spin faster? J Comp Physiol B 176: 213–222.

von Helversen O.V. and Reyer H.U. 1984. Nectar intake and energy ex-penditure in aflower visiting bat (Anoura caudifer). Oecologia 63: 178–84.


Williams J.B., Anderson M.D. and Richardson P.R.K. 1997. Seasonaldifferences in field metabolism, water requirements, and foraging be-havior of free-living aardwolves. Ecology 78: 2588–2602.

Williams J.B., Lenain D., Ostrowski S., Tieleman B.I. and Seddon P.J.2002. Energy expenditure and water flux of Rüppell's foxes in SaudiArabia. Physiol Biochem Zool 75: 479–488.

Williams J.B., Ostrowski S., Bedin E. and Ismail K. 2001. Seasonalvariation in energy expenditure, water flux and food consumptionof Arabian oryx Oryx leucoryx. J Exp Biol 20: 2301–2311.

Winstanley R.K., Buttemer W.A. and Saunders G. 2003. Field met-abolic rate and body water turnover of the red fox Vulpes vulpes inAustralia. Mammal Rev 33: 295–301.

Zelová J. Šumbera R., Okrouhlík J., Šklíba J., Lövy M. and Burda H.2011. A seasonal difference of daily energy expenditure in a free-living subterranean rodent, the silvery mole-rat (Heliophobius argen-teocinereus; Bathyergidae). Comp Biochem Physiol A 158: 17–21.

References

Al-Ramamneh, D., Riek, A., Gerken, M., 2010. Deuterium oxide dilution accurately pre-dicts water intake in sheep and goats. Animal 4, 1606–1612.

Anderson, J.J., Jetz, W., 2005. The broad-scale ecology of energy expenditure of endo-therms. Ecol. Lett. 8, 310–318.

Animut, G., Goetsch, A.L., Aiken, G.E., Puchala, R., Detweiler, G., Krehbiel, C.R., Merkel, R.C.,Sahlu, T., Dawson, L.J., Johnson, Z.B., Gipson, T.A., 2005. Grazing behaviour and energyexpenditure by sheep and goats co-grazing grass/forb pastures at three stockingrates. Small Rum. Res. 59, 191–201.

Bell, F.R., 1959. Preference thresholds for taste discrimination in goats. J. Agric. Sci. 52,125–129.

Bersényi, A., Fekete, S., Hausenblas, J., Emese, A., 2002. Lótápok emészthetősége. Mag-yar Állatorvosok Lapja 124, 579–585.

Bininda-Emonds, O.R.P., Cardillo, M., Jones, K.E., MacPhee, R.D.E., Beck, R.M.D., Grenyer,R., Price, S.A., Vos, R.A., Gittleman, J.L., Purvis, A., 2007. The delayed rise of present-day mammals. Nature 446, 507–512.

Blackburn, T., Cassey, P., Lockwood, J.L., 2009. The role of species traits in the establish-ment success of exotic birds. Global Change Biol. 15, 2852–2860.

Blomberg, S.P., Garland Jr., T., Ives, A.R., 2003. Testing for phylogenetic signal in com-parative data: behavioral traits are more labile. Evolution 57, 717–745.

Cain III, J.W., Krausman, P.R., Rosenstock, S.S., Turner, J.C., 2006. Mechanisms of ther-moregulation and water balance in desert ungulates. Wildl. Soc. Bull. 34, 570–581.

Capellini, I., Venditti, C., Barton, R.A., 2010. Phylogeny and metabolic scaling in mam-mals. Ecology 91, 2783–2793.

Cheverud, J.M., Dow, M.M., 1985. An autocorrelation analysis of genetic variation dueto lineal fission in social groups of Rhesus macaques. Am. J. Physiol. Anthropol.67, 113–121.

Chown, S.L., Slabber, S., McGeoch, M.A., Janion, C., Leinass, H.P., 2007. Phenotypic plas-ticity mediates climate change responses among invasive indigenous arthropods.Proc. R. Soc. B 277, 2531–2537.

Cooper, C.E., Withers, P.C., 2006. Numbats and aardwolves—how low is low? A re-affirmation of the need for statistical rigour in evaluating regression predictions.J. Comp. Physiol. B 176, 623–629.

Corbett, J.C., 1990. Standing committee on agriculture, ruminants subcommittee. Feed-ing Standards for Australian Livestock: Ruminants. CSIRO, Melbourne.

Coutts-Smith, A.J., Mahon, P.S., Letnic, M., Downey, P.O., 2007. The Threat Posed by PestAnimals to Biodiversity in New South Wales. Invasive Animals Cooperative Re-search Centre, Canberra.

Covell, D.F., Miller, D.S., Karasov, W.H., 1996. Cost of locomotion and daily energy ex-penditure by free-living swift foxes (Vulpes velox): a seasonal comparison. Can. J.Zool. 74, 283–290.

Dawson, T.J., Ellis, B.A., 1979. Comparison of the diets of yellow-footed rock-wallabiesand sympatric herbivores in western New South Wales. Aust. Wildl. Res. 6,245–254.

Dawson, T.J., Ellis, B.A., 1996. Diets of mammalian herbivores in Australian arid, hillyshrublands: seasonal effects on overlap between euros (hill kangaroos) sheepand feral goats, and on dietary niche breadths and electivities. J. Arid. Environ.34, 491–506.

Dawson, T.J., Denny, M.J.S., Russell, E.M., Ellis, B.A., 1975. Water usage and diet prefer-ences of free ranging kangaroos, sheep and feral goats in the Australian arid zoneduring summer. J. Zool. London 177, 1–23.

Degen, A.A., Hazan, A., Kam, M., Nagy, K.A., 1991. Seasonal water influx and energy ex-penditure of free-living fat sand rats. J. Mammal. 72, 652–657.

Degen, A.A., Khokhlova, I.S., Kam, M., Nagy, K.A., 1997. Body size, granivory and season-al dietary shifts in desert gerbilline rodents. Funct. Ecol. 11, 53–59.

Fadely, B.S., Worthy, G.A.J., Costa, D.P., 1990. Assimilation efficiency of northern furseals determined using dietary manganese. J. Wildl. Manage. 54, 246–251.

Fancy, S.G., Blanchard, J.M., Holleman, D.F., Kokjer, K.J., White, R.G., 1986. Validation ofthe doubly labelled water method using a ruminant. Am. J. Physiol. 20, R143–R149.

Fensham, R.J., Fairfax, R.J., 2008. Water-remoteness for grazing relief in Australian arid-lands. Biol. Conserv. 141, 1447–1460.

Freudenberger, D.O., Hume, I.D., 1992. Ingestive and digestive responses to dietaryfibre and nitrogen by two macropodid marsupials (Macropus robustus erubescensand M.r. robustus) and a ruminant (Capra hircus). Aust. J. Zool. 40, 181–194.

Gazzola, A., Avanzinellia, E., Bertellia, I., Tolosanob, A., Bertottob, P., Mussoc, R., Apollonio,M., 2007. The role of the wolf in shaping a multi-species ungulate community in theItalian western Alps. Ital. J. Zool. 74, 297–307.

Gessaman, J.A., Newgrain, K., Green, B., 2004. Validation of the doubly-labeled water(DLW) method for estimating CO2 production and water flux in growing poultrychicks. J. Avian Biol. 35, 71–96.

Gittleman, J.L., Thompson, S.D., 1988. Energy allocation in mammalian reproduction.Am. Zool. 28, 863–875.

Goldberg, G.R., Prentice, A.M., Coward, W.A., Davies, H.L., Murgatroyd, P.R., Wensing, C.,Black, A.E., Harding, M., Sawyer, M., 1993. Longitudinal assessment of energy ex-penditure in pregnancy by the doubly labeled water method. Am. J. Clin. Nutr.57, 494–505.

Golley, F.B., 1961. Energy values of ecological materials. Ecology 42, 581–584.Gorman, M.L., 2007. Restoring the ecological balance to the British mammal fauna.

Mammal Rev. 37, 316–325.Haggarty, P., Robinson, J.J., Ashton, J., Milne, E., Adam, C.L., Kyle, C.E., Christie, S.L.,Midwood,

A.J., 1998. Estimation of energy expenditure in free-living red deer (Cervus elaphus)with the doubly-labelled water method. Br. J. Nutr. 80, 263–272.

Harrington, G.N., 1986. Herbivore diet in a semi-arid Eucalyptus populnea woodlandwoodland. 2. Feral goats. Aust. J. Exp. Agric. 26, 423–429.

Hayes, J.P., Shonkwiler, J.S., 2006. Allometry, antilog transformations, and the perils ofprediction on the original scale. Physiol. Biochem. Zool. 79, 665–674.

Hayes, J.P., Shonkwiler, J.S., 2007. Erratum: allometry, antilog transformations, and theperils of prediction on the original scale. Physiol. Biochem. Zool. 80, 556.

Hoppe, P.P., Qvortrup, S.A., Woodford, M.H., 1977. Rumen fermentation and food selec-tion in East African sheep, goats, Thomson's gazelle, Grant's gazelle and impala.J. Agric. Sci. 89, 129–135.

Huston, J.E., 1978. Forage utilisation and nutrient requirements of the goat. J. Dairy Sci.61, 988–993.

James, C., Landsberg, J., Morton, S.R., 1999. Provision of watering points in Australianarid zone: a review of effects on biota. J. Arid. Environ. 41, 87–121.

Junghans, P., Derno, M., Gehre, M., Höfling, R., Kowski, P., Strauch, G., Jentsch, W., Voigt,J., Hennig, U., 1997. Calorimetric validation of 13C bicarbonate and doubly labelledwater method for determining the energy expenditure in goats. Z. Ernährungswiss.36, 268–272.

Keitt, B.S., Wilcox, C., Tershy, B.R., Croll, D.A., Donlan, C.J., 2002. The effect of feral catson the population viability of Black-vented Shearwaters (Puffinus opisthomelas)on Natividad Island, Mexico. Anim. Conserv. 5, 217–223.

Kolar, C.S., Lodge, D.M., 2001. Progress in invasion biology: predicting invaders. TrendsEcol. Evol. 16, 199–204.

Lachica, M., Aguilera, J.F., 2003. Estimation of the energy needs in the free-ranging goatwith particular reference to the assessment of its energy expenditure by the 13C-bicarbonate method. Small Rum. Res. 49, 303–318.

Lachica, M., Aguilera, J.F., 2005. Energy needs of the free-ranging goat. Small Rum. Res.60, 111–125.

Landsberg, J., James, C.D., Morton, S.R., Hobbs, T.J., Stol, J., Drew, A., Tongway, H.,1997. The effects of artificial sources of water on rangeland biodiversity.Final report to the Biodiversity Convention and Strategy Section of the Biodi-versity Group, Environment Australia, January 1997. Canberra: EnvironmentAustralia.

Lifson, N., McClintock, R., 1966. Theory of use of the turnover rates of body water formeasuring energy and material balance. J. Theor. Biol. 12, 46–74.

Lim, T.L., Giles, J.R., 1987. Studies on the yellow-footed rock-wallaby, Petrogale xanthopusgray (Marsupialia: Macropodidae) III. Distribution and management in western NewSouth Wales. Aust. Wildl. Res. 14, 147–161.

Lu, C.D., 1988. Grazing behavior and diet selection of goats. Small Rum. Res. 1, 205–216.McDonald, B.J., O'Rourke, P.K., Connell, J.A., Hoey, W.A., 1988. Prenatal growth and es-

timation of fetal age in the Australian feral goat. Aust. J. Agric. Res. 39, 729–734.McLeod S (1996) The foraging behaviour of the arid zone herbivores the red kangaroo

(Macropus rufus) and sheep (Ovis aries) and its role in their competitive interaction,population dynamics and life-history strategies. Dissertation, The University of NewSouth Wales, Australia.

McLeod, R., 2004. Counting the Cost: Impact of Invasive Animals in Australia 2004. Co-operative Research Centre for Pest animal Control, Canberra.

Midwood, A.J., Haggarty, P., McGaw, B.A., Robinson, J.J., 1989. Methane production inruminants: its effect on the doubly labelled water method. Am. J. Physiol. 25,R1488–R1495.

Midwood, A.J., Haggarty, P., McGaw, B.A., 1993. The doubly labeled water method: er-rors due to deuterium exchange and sequestration in ruminants. Am. J. Physiol.264, 561–567.

Midwood, A.J., Haggarty, P., McGaw, B.A., Mollison, G.S., Milne, E., Duncan, G.J., 1994.Validation in sheep of the doubly labeled water method for estimating CO2 produc-tion. Am. J. Physiol. 266, R169–R178.

Munn, A.J., Dawson, T.J., McLeod, S.R., Croft, D.B., Thompson, M.B., Dickman, C.R., 2009.Field metabolic rate and water turnover of red kangaroos and sheep in an aridrangeland: an empirically derived dry-sheep-equivalent for kangaroos. Aust. J.Zool. 57, 23–28.

Mutze, G.J., Green, B., Newgrain, K., 1991. Water flux and energy use in wild house mice(Mus domesticus) and the impact of seasonal aridity on breeding and populationlevels. Oecologia 88, 529–538.

Naderi, S., Rezaei, H., Pompanon, F., Blum, M.G.B., Negrini, R., Naghash, H., Balkiz, O.,Mashkour, M., Gaggiotti, O.E., Ajmone-Marsan, P., Kence, A., Vigne, J., Taberlet, P.,2008. The goat domestication process inferred from large-scale mitochondrial


DNA analysis of wild and domestic individuals. Proc. Natl. Acad. Sci. U.S.A. 105,17659–17664.

Nagy, K.A., 1977. Cellulose digestion and nutrient assimilation in Sauromalus obesus, aplant-eating lizard. Copeia 2, 355–362.

Nagy, K.A., Costa, D.P., 1980. Water flux in animals: analysis of potential errors in thetritiated water method. Am. J. Physiol. 238, R454–R465.

Nagy, K.A., Gruchacz, M.J., 1994. Seasonal water and energy metabolism of the desertdwelling kangaroo rat (Dipodomys merriami). Physiol. Zool. 67, 1461–1478.

Nagy, K.A., Montgomery, G.G., 1980. Field metabolic rate, water flux and food con-sumption in three-toed sloths (Bradypus variegatus). J. Mammal. 61, 465–472.

Nagy, K.A., Girard, I.A., Brown, T.K., 1999. Energetics of free-ranging mammals, reptiles,and birds. Annu. Rev. Nutr. 19, 247–277.

Newsome, A.E., Catling, P.C., Cooke, B.D., Smyth, R., 2001. Two ecological universes sep-arated by the dingo barrier fence in semi-arid Australia: interactions betweenlandscapes, herbivory and carnivory, with and without dingoes. Rangelands J. 23,71–98.

Parkes, J., Henzell, R., Pickles, G., 1996. Managing vertebrate pests: feral goats. AustralianGovernment Publishing Service, Canberra.

Pekins, P.J., Smith, K.S., Mautz, W.W., 1998. The energy costs of gestation in white-tailed deer. Can. J. Zool. 76, 1091–1097.

Puchala, R., Min, B.R., Goetsch, A.L., Sahlu, T., 2005. The effect of a condensed tannin-containing forage on methane emission by goats. J. Anim. Sci. 83, 182–186.

Robbins, C.T., 2001. Wildlife Feeding and Nutrition, 2nd edition. Academic Press, NewYork.

Rohlf, F.J., 2001. Comparative methods for the analysis of continuous variables: geo-metric interpretations. Evolution 55, 2143–2160.

Russo, S.E., Robinson, S.K., Terborgh, J., 2003. Size-abundance relationships in anAmazonian bird community: implications for the energetic equivalence rule.Am. Nat. 161, 267–283.

Schoeller, D.A., Fijeld, C.R., 1991. Human energy metabolism: what have we learnedfrom the doubly labelled water method? Annu. Rev. Nutr. 11, 355–373.

Seymour, R.S., Withers, P.C., Weathers, W.W., 1998. Energetics of burrowing, running,and free-living in the Namib Desert golden mole (Eremitalpa namibensis). J. Zool.244, 107–117.

Silanikove, N., 2000. The physiological basis of adaptation in goats to harsh environ-ments. Small Rum. Res. 35, 181–193.

Sparling, C.E., Thompson, D., Fedak, M.A., Gallon, S.L., Speakman, J.R., 2008. Estimatingfield metabolic rates of pinnipeds: doubly labelled water gets the seal of approval.Funct. Ecol. 22, 245–254.

Speakman, J.R., 1997. Doubly Labeled Water. Chapman & Hall, New York NY, USA.Speakman, J.R., Krόl, E., 2010. Maximal heat dissipation capacity and hyperthermia

risk: neglected key factors in the ecology of endotherms. J. Anim. Ecol. 79,726–746.

Squires, V.R., 1980. Chemical and botanical composition of the diets of oesophageallyfistulated sheep, cattle and goats in a semi-arid Eucalyptus populnea woodlandcommunity. Austr. Rangeland J. 2, 94–103.

Underhill, S., Grigg, G.C., Pople, A.R., Yates, D.J., 2007. A physiological assessment of theuse of water point closures to control kangaroo numbers. Wildl. Res. 34, 280–287.

van Kleunen, M., Dawson, W., Schlaepfer, D., Jeschke, J.M., Fischer, M., 2010. Are in-vaders different? A conceptual framework of comparative approaches for assessingdeterminants of invasiveness. Ecol. Lett. 13, 947–958.

Warren, L.E., Ueckert, D.N., Shelton, J.M., 1984. Comparative diets of Rambouillet,Barbado, and Karakul sheep and Spanish and Angora goats. J. Range Manage.37, 172–180.

Weinberg, P., Jdeidi, T., Masseti, M., Nader, I., de Smet, K., Cuzin, F., 2008. Capra aega-grus. IUCN 2010. IUCN Red List of Threatened Species Version 2010.4.

West, P., Saunders, G., 2007. Pest Animal Survey: A Review of the Distribution, Impactand Control of Invasive Animals Throughout NSW and ACT. NSW Department ofPrimary Industries, Orange, Australia.

Williams, J.B., Anderson, M.D., Richardson, P.R.K., 1997. Seasonal differences in fieldmetabolism, water requirements, and foraging behavior of free-living aardwolves.Ecology 78, 2588–2602.

Williams, J.B., Ostrowski, S., Bedin, E., Ismail, K., 2001. Seasonal variation in energy ex-penditure, water flux and food consumption of Arabian oryx Oryx leucoryx. J. Exp.Biol. 204, 2301–2311.

Wilson, G.R., Gerritsen, J., Milthorphe, P.L., 1976. The yellow-footed rock-wallaby, Pet-rogale xanthopus (Macropodidae), in western New South Wales. Aust. Wildl. Res. 3,73–78.

Withers, P.C., Cooper, C.E., Larcombe, A.N., 2006. Environmental correlates of physio-logical variables in marsupials. Physiol. Biochem. Zool. 79, 437–453.

Zerebecki, R.A., Sorte, C.J.B., 2011. Temperature tolerance and stress proteins as mech-anisms of invasive species success. PLoS One 6, e14806 doi:10.1371.

energy and water use by invasive goats (capra hircus) in an australian rangeland, and a caution...

Documents