Boldness towards novelty and translocation success in captive-raised, orphaned Tasmanian devils

RESEARCH ARTICLE

Boldness Towards Novelty and Translocation Successin Captive‐Raised, Orphaned Tasmanian DevilsDavid L. Sinn,1,2,3* Lisa Cawthen,1,2 Susan M. Jones,2 Chrissy Pukk,1 and Menna E. Jones1,2

1Department of Primary Industries, Parks, Water, & The Environment, Hobart, Tasmania, Australia2School of Zoology, University of Tasmania, Hobart, Tasmania, Australia3Department of Psychology, University of Texas at Austin, Austin, Texas

Translocation of endangered animals is common, but success is often variable and/or poor. Despite its intuitive appeal, little isknown with regards to how individual differences amongst translocated animals influence their post‐release survival, growth,and reproduction. We measured consistent pre‐release responses to novelty in a familiar environment (boldness;repeatability¼ 0.55) and cortisol response in a group of captive‐reared Tasmanian devils, currently listed as “Endangered” bythe IUCN. The devils were then released at either a hard‐ or soft‐release site within their mothers’ population of origin, andindividual growth, movement, reproduction (females only), and survival across 2–8 months post‐release was measured. Sex,release method, cohort, behavior, and cortisol response did not affect post‐release growth, nor did these factors influence thehome range size of orphan devils. Final linear distances moved from the release site were impacted heavily by the releasecohort, but translocated devils’movement overall was not different from that in the same‐agewild devils. All orphan females ofreproductive age were subsequently captured with offspring. Overall survival rates in translocated devils were moderate(�42%), andwere not affected by devil sex, release method, cohort, release weight, or pre‐release cortisol response. Devils thatsurvived during the study period were, however, 3.5 times more bold than those that did not (effect size r¼ 0.76). Our resultssuggest that conservation managers may need to provide developmental conditions in captivity that promote a wide range ofbehaviors across individuals slated for wild release. Zoo Biol. 33:36–48, 2014. © 2013 Wiley Periodicals, Inc.

Keywords: reintroduction; animal personality; stress physiology; endangered species

INTRODUCTION

Translocation of threatened or endangered species is animportant tool used by many conservation programs tosupplement already existing populations or to return animalsto sites where they were previously extirpated. Unfortunately,survival rates of translocated animals (especially carnivores)raised in captivity are usually low [Beck et al., 1994; Fischerand Lindenmayer, 2000; Jule et al., 2008; Molinari‐Jobinet al., 2010]. Factors such as pre‐release conditioning [Dobsonand Lyles, 2000], time spent in captivity [Fernández‐Moránet al., 2004; McPhee, 2004; Devineau et al., 2011], releasemethod [Ewen and Armstrong, 2007], cohort effects [Rödelet al., 2009], age [Aaltonen et al., 2009], sex[Wedekind, 2002], and social factors [Gelling et al., 2010;Shier and Swaisgood, 2012] may all have an impact ontranslocation success, but for many endangered or threatenedcarnivores, if and when these different factors impact onsuccess (or interact) are generally unknown. In cases whereconservation programs implement translocation on new

species, use of adaptive management to refine future decisionscan be implemented [Rout et al., 2009]. Given uncertaintysurrounding a new translocation strategy, adaptive manage-ment involves systematic collection of information onreleased individuals and outcomes of their success todetermine the effectiveness of different management methods[Sarrazin and Barbault, 1996].

Of course, information‐gathering by conservation pro-grams is often restricted logistically and/or financially, and itcan be unclear in cases where new species are translocated as to

David L. Sinn and Lisa Cawthen contributed equally to this paper.

�Correspondence to: David L. Sinn, Department of Psychology, TheUniversity of Texas at Austin, 108 E, Dean Keeton Stop A8000, Austin,TX 78712‐1043. E‐mail: [email protected]

Received 19 November 2013; Accepted 27 November 2013

DOI: 10.1002/zoo.21108Published online 28 December 2013 in Wiley Online Library(wileyonlinelibrary.com).

© 2013 Wiley Periodicals, Inc.

Zoo Biology 33: 36–48 (2014)

which components of a translocation strategy should beconsidered. In addition, working with an endangered speciesmay necessarily restrict sample sizes available to test multipleputative factors of interest. In cases such as these, in addition toconsidering more obvious (and easily measured) features suchas an individual’s sex [Letty et al., 2000] or its history oftemporally‐specific social and environmental conditions [e.g.,cohort effects: Linklater et al., 2012], variation in pre‐releasebehavior and physiology between captive animals may need tobe considered, as these are thought to be the most likelypredictors of an individual’s post‐release health, survival, andreproduction [Kleiman, 1989; Sarrazin and Barbault, 1996;Sarrazin and Legendre, 2000; Teixeira et al., 2007; Juleet al., 2008; Dickens et al., 2010]. Individual variation inbehavior and physiology is ubiquitous in nature [i.e.,“personalities,” “coping” styles, and “behavioral syndromes”:Koolhaas et al., 1999; Wingfield, 2003; Sih et al., 2004].Individual variation in behavior and/or stress physiology maylargely affect an individual’s foraging behavior [Randset al., 2003; van Oers et al., 2005], habitat choice [Stampsand Swaisgood, 2007; Stamps et al., 2009], antipredatorresponse [Sih et al., 2003; Quinn and Cresswell, 2005], socialstrategy [Sinervo, 2001; Sih and Watters, 2005], and survival[Dingemanse et al., 2004]. In other words, individualdifferences in behavior and stress physiology in a releasegroupmay be the salient feature when considering translocatedanimals’welfare, reproduction, and mortality, and thus overalltranslocation success [Teixeira et al., 2007]. Unfortunately,there is a dearth of information currently available on intrinsicindividual‐level variation and subsequent translocation successin endangered carnivores [Bremner‐Harrison et al., 2004;Watters and Meehan, 2007].

The Tasmanian devil (Sarcophilus harrissii), theworld’s largest extant carnivorous marsupial, is restrictedin its range to the island state of Tasmania, where it has facedsignificant population declines since the 1990s [Hawkinset al., 2006] due to a contagious fatal disease, Devil FacialTumor Disease [DFTD; McCallum and Jones, 2006;McCallum, 2008]. DFTD, a transmissible cancer, has hadsignificant impacts on population structure [Lachishet al., 2007; 2009], life history [Jones et al., 2008; Lachishet al., 2011], and resulting population dynamics. In somelocal populations devil numbers have decreased by 90% sincethe late 1990s [McCallum et al., 2009], resulting in a listing ofthe species by the IUCN as “Endangered” (http://www.iucnredlist.org/). Management approaches aimed at therecovery of devil populations in the wild are currentlylimited, but include reintroduction following extinction in thewild from an “insurance” metapopulation and geneticrestoration involving translocations [Jones et al., 2007]. Aninsurance metapopulation of Tasmanian devils currentlyexists in a number of captive facilities, including intensivecaptive pens in wildlife parks and zoos, in free‐rangeenclosures, and in a semi‐wild population on an island [Joneset al., 2007; Sinn et al., 2010a]. These populations aremanaged with the idea of maintaining 95% genetic diversity

in captivity for 50 years, but to date, little has been quantifiedconcerning how captive devils may behave, reproduce, andsurvive upon translocation to the wild.

Our overall aim in this paper is to explore whether pre‐release behavior and physiology, sex, cohort, and releasemethod influence the post‐release growth, movement,reproduction, and survival of translocated captive‐rearedTasmanian devils. A secondary aim is to compare patterns ofstress physiology, sex, movement, reproduction, and survivalof the translocated orphan devils with similar aged and sexdevils from the wild. Our available sample size of orphandevils for translocation was limited in this endangered speciesconservation program, necessitating rigorous reduction of thenumber of independent variables prior to analysis and carefulinterpretation of results. This paper represents the firstexploratory analysis of the factors that may influencetranslocation success of Tasmanian devils, a key componentof the effective conservation of this species.

METHODS

Subjects

We had access to two different cohorts of orphanedTasmanian devils that were slated for release in their mothers’population of origin, as well as a small same‐age wild cohortof devils. Twenty‐eight subadult devils contributed data toour study. Four subjects (two males and two females) wereselected from the wild population of devils in the release areaon the Forestier Peninsula in southeast Tasmania (42° 030 5300

S, 148° 170 1400 E). Twenty‐four subjects had been hand‐reared within the Department of Primary Industries, Parks,Water, & Environment (DPIPWE) orphan care program,Tasmania, Australia, following removal as pouch young fromdiseased females that were euthanized as part of a diseasesuppression trial on the Forestier Peninsula [see Lachishet al., 2010]. Hand‐reared orphaned devils were from twodifferent cohorts. The first cohort (birth year 2006) consistedof 13 devils (six females, seven males) from seven differentlitters. The 2006 cohort came into orphan care duringJuly 2006—August 2006, where they were hand‐reared(bottle‐fed) together in litters or groups of no less than twoindividuals. The 2006 cohort was moved from individualcarers to four adjacent group‐holding pens at a wildlife park(Tasmanian Devil Conservation Park, Taranna, TAS) inApril 2007, where they remained in mixed groups of three tofour individuals until release. The 2007 cohort came intoorphan care during July 2007–August 2007, and consisted of11 devils (six females, five males) from five different litters.The 2007 cohort differed from the 2006 one in that they didnot spend time in the mixed group‐holding pens. The 2007cohort was released at a younger age directly from individualorphan care and so data were unavailable for pre‐releasebehavioral and cortisol responses (see Data Analysis section).Since sample sizes were small, we used all available data

Translocation of Tasmanian Devils 37

Zoo Biology

from the two orphan cohorts where possible. All devils wereuniquely tagged with individual microchip transponders(Allflex©, Palmerston North, New Zealand) implantedsubcutaneously under the skin to allow for individualidentification.

Release Sites and Methods

Release of the 2006 cohort occurred from 29November 2007 to 10 December 2007, when individualswere approximately 20 months old. All 2007 cohort devilswere released from 11 January to 19 May 2008 directly fromorphan care, when devils were 10–14 months of age.Typically, male and female devil offspring are weaned anddisperse from their natal home at 9 months of age [Lachishet al., 2011]. Orphaned devils were released at three differentsites, all within DFTD‐affected areas but also within theorphans’ population of origin (Fig. 1). The ForestierPeninsula site offered several additional useful features: (a)it is large (160 km2); (b) it has a high degree of naturalisolation being surrounded by water on all sides andconnected to “mainland” Tasmania only by a narrow roadbridge across a boating channel and to the Tasman Peninsulato the south by a narrow isthmus; and (c) there was an alreadyexisting intensive quarterly trapping program over the entirepeninsula [Lachish et al., 2010]. The northern third of theForestier peninsula comprises a mixture of dry eucalypt forestand improved pasture on a large livestock property thatsupports a high‐density devil population. All three releasesites were in the northern third of the peninsula. The lowertwo‐thirds of the peninsula are primarily forests managed fortimber harvest which grade from dry into wet eucalypt forestat the southern end that supports lower densities of devils[Pukk, 2005].

Translocated orphan devils were given either a “soft‐”or “hard‐release.” Hard release involved transporting thedevils in traps to the release site in the late afternoon. Trapswere placed within 100m of each other and left open for the

devils to leave on their own free will. The hard release sitewas located within forest in the northeastern part of thepeninsula just south of the main area of pasture, and waschosen to maximize the distance (5 km) from the mainhighway on the peninsula. Soft release involved holdingdevils in pens on‐site for 1 week prior to release, andproviding supplemental food after release. Supplementalfeeding involved placing carcasses of native herbivores atseveral sites within 1 km from the release site, three times aweek for 4 weeks, then twice a week for 4 weeks, followed byonce a week for 4 weeks. The soft release site for the 2006cohort was located in a mosaic of pasture and forest in thenorthwestern part of the peninsula along a drainage systemthat channeled animal movement away from the highway(1 km distant), and was chosen for logistical reasons (i.e.,staff access for supplemental feeding). For the 2007 cohort, anew soft release site was chosen on the pastured at the easternedge of the peninsula, 5 km north of the hard release site and8 km from the highway.

Five individuals from the 2006 orphan cohort showedstereotypical pacing of pen walls and were retained in thewildlife park as display animals. The remaining devils fromthe 2006 cohort were assigned to hard or soft release sitesafter consideration of relatedness, sex, and size of individuals.Eight individuals (four males, four females) were equallysplit across hard and soft release sites with related individualsand rearing groups (2006 cohort) being split among treat-ments. The majority of the 2007 cohort was released at thesoft release site (four males, five females); two individuals(one male, one female) from the 2007 cohort were released atthe hard release site (Table 1).

Pre‐Release Behavioral Assay

Behavioral assays were performed on the 13 devilsfrom the 2006 cohort prior to release; assays were notperformed on the 2007 cohort due to logistics. The test arenafor behavioral observations was a holding pen that devils had

Fig. 1. Location of the Forestier peninsula in Tasmania, Australia. All three orphan release sites were located in the northern third of theForestier Peninsula in the mixed forest/pasture habitat that had the highest densities of wild devils.

38 Sinn et al.

Zoo Biology

already been exposed to during group housing at the wildlifepark, 5m� 5m in size, which consisted of a 1.2m highperimeter fence made of corrugated iron placed on naturalground, skirted with 0.5m wide chicken wire to discouragedigging near fences. The test arena was modified by fixing alarge pink exercise ball (1.2m diameter) with rope at one endof the enclosure and a mirror (0.8m� 1.4m) at the other.None of the devils had been exposed to these objectspreviously; thus, we considered their reactions to theseobjects to be “boldness” with regards to novelty (the beachball and mirror) in a semi‐familiar environment [Réaleet al., 2007]. Each devil was given the behavioral assay twice;once during the second week of November 2007 and thenagain 6 days later. Testing order of individuals on each testday was haphazard (i.e., groups of devils were targeted fortesting on a particular day, but the order of testing individualswithin target groups was randomized). Due to logistic reasons(i.e., volunteers were only able to feed devils during the day),the 2006 cohort devils were day active at the time of thebehavioral assays, so assays were given during daylight hours(0800–1700). Preliminary observations indicated that after ashort acclimation period, devils did not react to the presenceof human observer standing outside the test arena as long asthe observer remained quiet and restricted their movements.

At the start of each behavioral assay, the subject was placed ina standardized location in the test arena in a hessian sack, andwas allowed to emerge from the sack in its own time. A30min observation period began when the devil emergedfrom the sack (all devils climbed out of sacks unassisted,mean time¼ 67.8 sec, SD¼ 83.6 sec).

Five behaviors were recorded during behavioralassays: latency to approach within 1min of the mirror (s),latency to approach within 1min of the beach ball (s), numberof times the subject devil touched the mirror (frequencycount), number of times the subject devil touched the beachball (frequency count), and total time spent moving (sec). Forfrequency counts, we used an a priori “5 second rule,”whereby a behavior was scored as a multiple frequency ifthere was at least a 5 sec break between occurrences. Forexample, if a subject touched the beach ball, subsequenttouches were counted only if they occurred 5 sec after theprevious touch. The first author performed all behavioralobservations. During the second week of testing a secondobserver independently recorded behaviors: inter‐observeragreement (Pearson’s r) on four of the five recordedbehaviors was higher than 0.53. Pearson’s r betweenobservers for total time spent moving was low (0.12), thisvariable was therefore removed from further analyses. Only

TABLE 1. Summary information for 24 captive‐raised orphan and four wild Tasmanian devils

Devil Cohort SexReleasedate

Releasemethod

Behavioralassay (Y/N)

ACTHchallenge (Y/N)

Radiocollared(Y/N)

Survival toJuly 08

Captive‐raised orphan devils that were releasedBlinky 2006 M 10/12/07 Soft Y Y Y NPixie 2006 F 29/11/07 Hard Y Y Y NBuzzy 2006 M 1/11/07 Soft Y Y Y NFanou 2006 F 10/12/07 Soft Y Y Y NDutchess 2006 F 29/11/07 Hard Y Y Y NVictor 2006 M 29/11/07 Hard Y Y Y YJoe 2006 M 29/11/07 Hard Y Y Y YQueenie 2006 F 10/12/07 Soft Y Y Y YMozart 2007 M 11/01/08 Soft N N N YTosca 2007 M 11/01/08 Soft N N N NSweetie Pie 2007 F 1/03/08 Soft N N N YHamburger 2007 M 23/02/08 Soft N N N NFreddy 2007 M 24/02/08 Soft N N N NMinty 2007 F 26/03/08 Soft N N N YMurf 2007 F 26/03/08 Soft N N N YMyrna 2007 F 4/04/08 Hard N N N YLevi 2007 M 4/04/08 Hard N N N NLyrra 2007 F 19/05/08 Soft N N N NAshka 2007 F 19/05/08 Soft N N N N

Captive‐raised orphan devils retained in captivityOrion 2006 M N/A N/A Y Y N N/ADraco 2006 M N/A N/A Y Y N N/ALady 2006 F N/A N/A Y Y N N/AWilma 2006 F N/A N/A Y Y N N/ADerek 2006 M N/A N/A Y Y N N/A

Wild devilsMasikus 2006 F N/A N/A N Y Y YJulia 2006 F N/A N/A N Y Y YBHoliday 2006 M N/A N/A N Y Y YB Bevan 2006 M N/A W N N Y N

Translocation of Tasmanian Devils 39

Zoo Biology

behavioral observations made by DS were used in furtherstatistical analysis.

Pre‐Release Physiological Assay

The 2006 orphan cohort and four wild same‐age cohortdevils were given an ACTH challenge to test the functionalityof the hypothalamic pituitary axis, and to provide pre‐releasemeasures of the capacity of the animals tomount an appropriatecortisol response to a standard stressor. The ACTH challengewas a single dose of Synacthen Depot (Novartis), 1mg/ml.One milligram Synacthen corresponds to approximately 100international units of ACTH so a 5 kg devil receiving 0.5mlSynacthen receives a dose equivalent to 10 iu/kg body weight.The ACTH challenge was given to the captive orphans duringthe fourthweek of November 2006, after the second behavioralassay, but prior to release. Thewild devils were tested similarlyin January 2008 during a trapping trip. Captive and wildanimals were trapped and blood sampling occurred before12:00 hr. The subject devil was gently removed from the trap toa hessian sack; 600ml of blood was obtained by pricking theperipheral ear vein with a lancet and collecting the blood intoheparinizedmicrohematocrit tubes. The devil wasweighed andan intramuscular injection of ACTH was given in the rump orthigh at a dosage rate of 20ml per kilogram body weight; theanimal was then returned to the trap. At 30min after the ACTHinjection, a second blood sample was taken following the sameprocedure. Blood samples were stored on cold packs in thefield and centrifuged that night; the plasma component wasseparated and stored at �20C until assayed.

Plasma cortisol in blood samples was measured using aspecific radioimmunoassay as in [Jones et al., 2005] but usinga different antiserum. Briefly, 25ml plasma was extractedwith 1ml absolute ethanol (A.R. grade); 100ml extract wasassayed in duplicate and commercial control sera wereincluded to confirm accuracy. Standards were set to 12.5–800 pg authentic cortisol in 50ml absolute ethanol. Fiftymicroliters of tritiated cortisol (�4,500 cpm: Amersham,diluted in absolute ethanol) were added to each sample orstandard tube and dried down before 200ml of antiserum(Bioquest SiroSera C‐3368, diluted 1:20,000 in phosgelbuffer) was added to all tubes except the non‐specific bindingtubes. The assay was incubated overnight at 4°C. Aliquots of300ml supernatant were counted in 2.5ml scintillation fluid(Ecolite1) for 5min using a Beckman LS 5801 liquidscintillation counter. Cortisol concentrations were calculatedusing a log‐logit plot, and corrected for the extractionefficiency of 80%. Assay accuracy and precision wereassured by including three levels of a commercially availablehuman control serum (CON‐6W; Diagnostic ProductsCorporation, Los Angeles, California, USA) in each assay.

Post‐Release Monitoring of Survival, Growth,Reproduction, and Movement

The Forestier Peninsula site had been trapped as threecontiguous trapping regions up to four times a year since

January 2006. For this study, we examined trapping data(GPS location and body weight (kg) for all individuals, andreproductive activity in released females) from four trips tothe region in December 2007, and January, April, andJuly 2008. Trapping periods typically included 40 traps setfor 10 nights in each trapping region simultaneously, exceptfor December 2007 when a single trip was performed forseven nights using 35 traps. During field trips, PVC pipetraps (diameter 315mm� length 875mm) were placedapproximately 250m apart at landscape features inlocations that would increase the likelihood of capturesuccess. Traps were baited with native herbivore prey(wallaby Macropus rufogriseus and pademelon Thylogalebillardierii) sourced from private culling programs on site,and checked daily commencing in the early morning. DFTDtransmission was minimized by following a protocoldeveloped by veterinary practitioners; this involved thesterilization of all equipment and traps with VirkonTM

(Antec International, Du Pont Animal Health Solutions,Sudburg, UK), an antibacterial, antiviral, antifungal, DNAdenaturing solution, and subsequent burning of hessiansacks and any remaining bait.

We used trapping records to measure survival toJuly 2008, post‐release growth of orphaned devils ((lastknown weight� release weight)/number of days betweenmeasures (kg/day)), and to observe whether orphaned femaledevils successfully reproduced through observations offemale pouches. We considered devils that were not foundpost‐mortem or were not trapped after release as “unsuccess-ful”; we assumed these individuals had most likely died. It ispossible that devils that were not found had left thereintroduction area, but we consider this unlikely given thesize and geographic isolation of the Forestier Peninsula.

Data on devil movements were also collected usingGPS data loggers (Televilt electronics) attached to leathercollars that were placed on the animals prior to release (eightcaptive‐raised individuals from the 2006 cohort) or on wildindividuals from the same age cohort (four wild individuals,the same wild individuals that experienced the ACTHchallenge). All collars were removed or collected byJanuary 2008. All GPS collars had an inbuilt magnetic pre‐programmed time‐release; collars were timed to release andfall off devils during the night to decrease the likelihoodcollars would be lost underground in dens. Unfortunately, weobtained reliable data from collars for only one captive raisedorphaned devil and three wild individuals due to acombination of collar malfunction, loss, and difficulty ofobtaining fixes. Therefore, we combined all available radiocollar, trapping location, and location of mortality data(Table 2) to calculate two animal movement metrics usingArcView GIS 3.3: the linear distance from the release site tolast known location (from collar or trapping data), and theindividuals’ 95% minimum convex polygon (MCP) using allavailable data points. Final linear distance was not calculatedfor the four wild‐caught devils, but 95% MCP home rangemetrics were.

40 Sinn et al.

Zoo Biology

Data Analyses

Due to our small sample size and large number offactors, we first attempted to reduce the number of potentialpredictor variables used in subsequent analyses. For each ofthe four observed behaviors, we averaged observations acrossweeks within individuals [Epstein, 1983;Fleeson, 2001; 2004] and examined the subsequent correla-tion matrix of the four averages (N¼ 13 for all pairwisecomparisons; Table 3). We also performed principalcomponents analysis [PCA, N¼ 13: Tabachnick andFidell, 1996] on the standardized average behaviors. Thefour behaviors (latency to touch mirror, latency to touch ball,(�) frequency of mirror touches mirror, (�) frequency of balltouches) were strongly inter‐correlated (averaged r usingFisher’s r to z transformation¼ 0.78, L95%CI¼ 0.40, U95%CI¼ 0.93), and loaded together on a single principalcomponent explaining 84.3% of the variation (Table 4).

Aggregate behavior scores were computed for eachdevil for each test time by summing the four observedstandardized measurements (“latency to touch mirror”[reverse keyed], “latency to touch ball” [reverse keyed],“frequency of mirror touches” and “frequency of balltouches”). We conducted standardization according to themean and SD of the global behavior observed at both timeperiods (i.e., N¼ 26, each individual occurred in two rows),allowing for variation in scores between time points. Higherboldness scores describe devils that had shorter latencies toinitially touch novel objects and touched them morefrequently during test periods; lower scores described devilswith larger latencies and higher overall frequencies of

touching the novel objects. Boldness scores were repeatablefrom time 1 to time 2 (intraclass correlation coefficient: 0.55,F(12,13)¼ 3.093, P¼ .03), so we summed the first and secondstandardized boldness scores to arrive at a single sumboldness score per devil. Variation in between‐individualaggregate responses to novel objects in familiar environmentsis commonly considered an index of “boldness” in nonhumananimals [Réale et al., 2007]; we adopt this terminology here.Only the single sum boldness score was used in subsequentanalyses, hereafter referred to as “boldness.”

In a further attempt to reduce the number of factorsneeded for prediction we tested whether boldness wasexpressed in a sex‐specificmanner using a t‐test (N¼ 13), andexamined the correlationmatrix of boldness, the three cortisolmeasures from the ACTH challenge, and initial release bodyweight (N for all pairwise comparisons¼ 13). Basal cortisolconcentrations were non‐normally distributed (Shapiro–Wilknormality test:W¼ 0.82, P¼ 0.004), so we used Spearman’srank for correlations that included this variable.

We tested whether the ACTH challenge was successfulusing a one‐way ANOVA with sex as the between‐subjectsfactor and time (basal and 30min post‐treatment concen-trations) as the within‐subjects factor (N¼ 17). Due to samplesizes and unbalanced design we did not include “source” (i.e.,wild or orphan) in this linear model. Instead, we graphicallycompared means and 95% confidence intervals of basalcortisol and 30min cortisol responses between orphan andwild devils [Cumming et al., 2007].

Standardized values of boldness, basal cortisol, andrelease weight were independent of one another (Table 5).Given this, we used main effects models to examine how sex,cohort, release method, boldness, and basal cortisol impactedon growth, movement, and survival. We tested whether sex,release method, and cohort predicted post‐release growthusing a general linear model (GLM; N¼ 16). We thenexamined the relationship between post‐release growth,boldness, and basal cortisol (N¼ 7 for both comparisons)using Spearman‐rank coefficients (post‐release growthwas notnormally distributed, Shapiro–Wilk W¼ 0.67, P¼ 0.0001).

TABLE2. Number ofGPS fixes using radiotelemetry, trapping,and the locations of dead individuals on the Forestier Peninsula,Tasmania

Group Radiotelemetry Trapping Mortality Total

Soft releaseMale (n¼ 6) 0 16 2 18Female (n¼ 7) 3 20 0 23

Hard releaseMale (n¼ 3) 2 52 2 56Female (n¼ 3) 0 15 1 16

WildMale (n¼ 2) 12 14 0 26Female (n¼ 2) 54 12 0 66

Total 71 129 5 205

The two separate cohorts of devils are collapsed here.

TABLE 3. Correlation matrix of the four averaged behavioralmeasurements taken during standardized tests (N¼ 13)

Behavior 1 2 3 4

Latency to touch mirror (1) 1.00Latency to touch ball (2) 0.65 1.00Frequency of mirror touches (3) �0.83 �0.80 1.00Frequency of ball touches (4) �0.80 �0.65 0.85 1.00

Spearman’s rank coefficients are given.

TABLE 4. Principal component loadings of standardizedaverage behaviors observed during behavioral tests (PCA;N¼ 13)

Behavior PCA 1—“Boldness”

Latency to touch mirror �0.91Latency to touch ball �0.89Frequency of mirror touches 0.93Frequency of ball touches 0.94KMO 0.86Bartlett’s test of sphericity x2ð6Þ ¼ �35:5, P< 0.0001

The Kaiser–Meyer–Olkin (KMO) statistic compares the observedcorrelations and partial correlations among the original variables;correlation matrices with KMO< 0.5 are inappropriate for PCA,those with KMO below 0.6–0.7 should be treated with caution, andthose with KMO> 0.70 are adequate [Budaev, 2010]. Bartlett’s testof sphericity has the null hypothesis that all correlations are zero.

Translocation of Tasmanian Devils 41

Zoo Biology

We tested whether sex, release method, and orphancohort predicted the final linear distance animals traveledusing a GLM; we also tested whether these same threeindependent variables predicted 95% MCP home range oftranslocated devils using a GLM (N¼ 9 for both models).Using the 2006 cohort data only, we examined therelationship between boldness, basal cortisol, and the twomovement metrics using Pearson’s and Spearman‐rankcorrelations (N¼ 6 for all tests). Finally, we examined meansand 95% confidence intervals to examine whether 95%MCPhomes ranges were different between reintroduced orphandevils and their wild counterparts (N¼ 10 for orphans, N¼ 4for wild‐caught devils).

For survival, we first fit a binary logistic model withsurvival to July 2008 as the dependent variable, with sex,release method, cohort, and release weight as independentpredictors (N¼ 19). Binary logistic models with boldness andbasal cortisol from the 2006 cohort would not converge,likely due to complete separation in the data (Fig. 2, Panel C)and/or sample size (N¼ 8). Instead, we examined mean‐leveldifferences in boldness and basal cortisol concentrations in2006 cohort devils that survived to July 2008 and those thatdid not, using two separate two‐tailed t‐tests (N¼ 8 for eachtest).

All analyses were performed using R version 3.0.1. Forall GLMs we assessed linearity and homogeneity of residuals[Field et al., 2012]; no violations were found.

RESULTS

Variable Reduction and Relationships AmongstPredictors

There was no effect of sex on boldness (t(10.63)¼ 0.02,P¼ 0.98), nor for sex on circulating cortisol (F(1)¼ 0.06,P¼ 0.81) or a time by sex interaction on ACTH response(F(1)¼ 0.37, P¼ 0.55) in the 2006 cohort. However, therewas an overall strong 30min response in circulating cortisolin response to the ACTH challenge (F(1)¼ 57.0, P< 0.001;Cohen’s d¼ 2.79). Examination of means and 95% confi-dence intervals of cortisol concentrations in wild and orphandevils revealed that orphaned devils did not have a

significantly different pattern of response to the ACTHchallenge than wild devils did, but wild devil baselinereadings tended to be higher than those observed in orphans(P� 0.05; Fig. 3).

Growth

None of the three independent variables (sex, releasemethod, and cohort) successfully predicted post‐releasegrowth of captive‐raised devils (R2¼ 0.11, F(3,11)¼ 0.45,P¼ 0.72), and there was no relationship between anindividual’s boldness (Spearman’s r(7)¼�0.03, P¼ 0.96),basal cortisol (Spearman’s r(7)¼�0.31, P¼ 0.50), and post‐release growth.

Movement

In total, 205 location fixes were obtained from trapping,mortality location, and GPS collar fixes (Table 2). In the maineffects model sex, cohort, and release method successfullypredicted final linear distance from release site (R2¼ 0.91,F(3,5)¼ 16.16, P¼ 0.005). However, cohort was the onlysignificant parameter, with devils from the 2007 cohorttraveling almost five times greater distance from their releasesites than devils from the 2006 cohort (Table 6; Fig. 4). Forthe captive‐raised 2006 cohort, there was no relationshipbetween boldness and final linear distance (Pearson’sr(5)¼ 0.28, P¼ 0.64). The rank‐order estimate for therelationship between basal cortisol concentrations and finallinear distance was negative and moderate, but not statisti-cally significant (Pearson’s r(5)¼�0.61, P¼ 0.27).

There was no relationship between sex, cohort, andrelease method on 95% MCP home ranges (R2¼ 0.31,F(3,5)¼ 0.75, P¼ 0.57). There was a trend for bolder devils tohave larger 95% MCP ranges, but this relationship did notreach statistical significance (Pearson’s r(6)¼ 0.68,P¼ 0.14). The relationship between 95% MCP range andbasal cortisol was strong and positive (Spearman’sr(10)¼ 0.68, P¼ 0.03). Mean 95% MCP ranges betweencaptive raised (5.71� 3.74 km2) and the four wild devils(6.64� 5.02 km2) had a large degree of overlap of 95%confidence intervals.

Reproduction

Of the captive raised released females that survived toJuly 2008 (two from the 2006 cohort, three from the 2007cohort), the two females of typical reproductive age (i.e., 2‐year‐old age class) from the 2006 cohort were both found tobe carrying four pouch young each in July 2008. None of thethree 2007 cohort females had reproduced by July 2008. Bothwild females used as the comparison females also had fourpouch young detected during the April 2008 trip.

Survival

Of the 19 orphan devils released from the 2006 and2007 cohorts, eight devils were alive as of July 2008, six were

TABLE 5. Correlation matrix of five potential predictorvariables

BoldnessBasalcortisol

30mincortisol

Releaseweight

Boldness 1Basal cortisol 0.00 130min cortisol 0.42 �0.32 1Release weight �0.03 0.10 0.10 1

N¼ 13 in all comparisons. Correlation coefficents are Pearson’s rexcept for those comparisons that involved basal cortisol circulatingconcentrations, in which case they are Spearman’s rank coefficients.None of the estimates reached statistical significance.

42 Sinn et al.

Zoo Biology

confirmed dead (five confirmed roadkill), and five individualswere not trapped or located by radiotelemetry after release(these individuals were presumed dead), resulting in anoverall survival rate of 42.1%. During the same period oftime, there was a 75% survival rate for the four wild‐caughtindividuals, although in July 2008 one wild individual

included in our study was found with signs of DFTD (M.E.Jones, unpublished data). No released orphans were foundwith any signs of DFTD as of July 2008. From the 2006cohort, three individuals were alive as of July 2008, threewere confirmed dead (all roadkill), and two individuals werenot recaptured, resulting in a known survival rate of 37.5%.From the 2007 cohort, five individuals were alive as ofJuly 2008, three confirmed dead (two from roadkill), andthree individuals were not recaptured or found, resulting in aknown survival rate of 45.4%.

None of the four independent variables (sex, releasemethod, cohort, and release weight) successfully predicted

0.0 0.2 0.4 0.6 0.8 1.0

-3-1

13

A

Survival to July 2008

Mea

n an

d 95

% C

I

No Yes

-3-1

13

0.0 0.2 0.4 0.6 0.8 1.0

-20

020

4060

B

Survival to July 2008

No Yes

-20

1550

-1 0 1 2

0.0

0.4

0.8

C

Boldness

Sur

viva

l

0 10 20 30 40 50 60

0.0

0.4

0.8

D

Basal Cortisol (ng/ml)

Fig. 2. Error bar (panels A and B) and scatterplots (panels C and D) for boldness, basal cortisol concentrations, and survival until July of2008 for translocated orphan Tasmanian devils. The y‐axis in Panel A is the aggregate boldness z‐score; the y‐axis in Panel B is basal cortisolconcentration (ng/ml). In Panels C and D, survival until July 2008 is plotted on the y‐axis, with boldness z‐score (Panel C) and basal cortisolconcentration (Panel D) on the x‐axis. Complete separation of boldness data occurred for the two groups of devils (survive/did not survive;panel C). N for all panels¼ 8.

Fig. 3. Means and 95% confidence intervals for baseline and30min cortisol response to an ACTH challenge in orphanedtranslocated devils (N¼ 13) and same‐age wild‐caught devils(N¼ 4). Two means with 95% confidence intervals overlap lessthan one‐half of each other can be considered prospectivelysignificantly different at P� 0.05 [Cumming et al., 2007].

TABLE 6. Parameter estimates from a general linear modelfitting final linear distance traveled according to an individual’ssex, cohort, and release method assigned (N¼ 9)

Parameter Estimate Standard error t P‐value

Intercept 765.7 1474.2 0.52 0.63Sex 2254.9 1002.7 2.25 0.07Cohort 5540.2 1124.7 4.93 0.004Release method �433.8 1194.6 �0.36 0.73

Parameter estimates for sex is females relative to males, for cohort is2006 relative to 2007, and for release method is soft‐relative to hard‐release.

Translocation of Tasmanian Devils 43

Zoo Biology

survival of the captive‐raised orphans to July 2008 over andabove the constant‐only logistic model (x2ð1Þ ¼ �1:10,P¼ 0.90). There was complete separation of boldness scoresbetween 2006 cohort devils that survived and those that didnot (Fig. 2), with devils having survived until July 2008being, on average, more bold towards novelty than those thatdid not survive (t(5.50)¼�3.35, P¼ 0.02; effect sizer¼ 0.76). There were no mean‐level differences in basalcortisol concentrations between successful and unsuccessfulreintroduced devils (t(4.72)¼ 0.53, P¼ 0.62; Fig. 2).

DISCUSSION

Here we provide a first exploratory report of severalputative factors that may have an impact on translocation ofcaptive‐reared Tasmanian devils. We measured responses tonovel objects in a familiar environment [“boldness”: Réaleet al., 2007] and hormonal response to an adrenocorticotropichormone (ACTH) challenge prior to release in hand‐raisedorphaned juvenile devils. Subjects were then released ateither a hard‐ or soft‐release (i.e., food supplemented) sitewithin their mothers’ population of origin, and we monitoredrelease sites and measured individual survival, growth,reproduction (females only), and movement of the releasedanimals. In addition, we monitored survival, reproduction(females only), hormonal response to an ACTH challenge,and animal movement of a same age cohort of wild devils tocompare with wild fitness at release sites [Mathewset al., 2005]. Overall, we found little evidence for an effectof an individual’s sex, cohort, and release method on itssubsequent post‐release growth, movement, and survival.There was also no detected relationship between anindividuals’ sex, its pre‐release boldness, cortisol, and releaseweight. Orphans and wild devils had similar cortisol profilesin response to the ACTH challenge, but wild devils hadhigher baseline cortisol concentrations overall. The younger

cohort of orphaned devils (2007) moved greater distancesthan the older cohort of orphans (2006), but there were nodifferences in measured home range sizes between orphancohorts andwild devils. Orphaned devils with higher baselinecortisol concentrations had larger resultant home rangesduring our study. Female orphan devils of reproductive agewere all found to be carrying pouch young. Overall, survivalof orphans during our study was moderate (�42%), andagain, neither sex, cohort, release method, or cortisolpredicted survival. Instead, we found that devils that survivedduring our study were three and a half times as bold in pre‐release behavioral assays than those than did not. Below wediscuss these results in light of our sample sizes andsubsequent translocation efforts in this and other endangeredcarnivore translocation studies.

We detected little relationship between sex, pre‐releaseweight, and boldness, and as well, these factors along withcohort and release method did not predict post‐releasemeasures of growth. In addition, sex and pre‐release weightwere not found to impact on our measure of post‐releasesurvival. An individual’s sex can often have a large impact onlife history [e.g., Crews et al., 1998], physiological stressresponses [e.g., While et al., 2010], growth [Mangel andStamps, 2001], and subsequent behavior [e.g., Hedrick andKortet, 2012]. While we consider our negative results in thisregard to be preliminary and in need of further testing (seebelow), sex, behavior, physiology, and growth traits that arephenotypically independent of one another implies that futuretranslocation efforts using captive Tasmanian devils may beable to select for particular univariate phenotypes withoutaffecting other individual traits that also may be important[e.g., Veenema et al., 2003; Dickens et al., 2009]. In otherwords, selection of release candidates may be possible basedsolely on sex, behavior, size, or physiology (or anotherindependent character thought to be salient), and not onmultivariate profiles of these traits simultaneously. Clearly,future tests using larger sample sizes are needed to confirmthese ideas, especially since correlated suites of traits withinindividuals are commonly observed [e.g., Koolhaaset al., 1999; Murn and Hunt, 2008; Pinter‐Wollman, 2009;Pinter‐Wollman et al., 2009; Carere et al., 2010]. Geneticstudies using current insurance population captive individu-als may also be useful in determining any genotypiccorrelations between these putative traits of interest, asphenotypic correlations can often mask actual genotypiccorrelations between traits [van Oers and Sinn, 2013].

Related to the idea that none of our independentpredictors successfully predicted growth is the idea of hard‐versus soft‐release methods. Results from other translocationefforts that have compared hard‐ versus soft‐release methodsare mixed. In some cases (i.e., burrowing owls [Mitchellet al., 2011] and Canadian lynx [Devineau et al., 2011]), soft‐release methods significantly improve post‐release survival;in other cases (i.e., Tawny owls [Griffiths et al., 2010] andhare wallabies [Hardman and Moro, 2006]) soft‐versus hard‐release method has little to no influence on survival of

0.2 0.8

040

0060

0010

000

Cohort

Mea

n (m

) and

95%

CI

2006 2007

050

0010

000

Fig. 4. Means and 95% confidence intervals for final lineardistancemoved between two cohorts of orphaned translocated devils(2006 cohort: N¼ 5; 2007 cohort: N¼ 4). Two means with 95%confidence intervals that do not overlap can be consideredsignificantly different at P� 0.05 [Cumming et al., 2007].

44 Sinn et al.

Zoo Biology

released animals.While there is some evidence that the lengthof time animals are allowed access to soft‐release conditionsmay influence survival rates [Hamilton et al., 2010; Roucoet al., 2010], evidence is also accumulating that factors suchas release cohort size and release‐site habitat quality may bemore salient to translocation success [Robinette et al., 1995;Linklater et al., 2012; Mat�ej 8u et al., 2012; Shier andSwaisgood, 2012]. In the current example, habitat quality atboth our hard‐ and soft‐release sites may have beenequivalent, and resulted in the lack of effect of releasemethod and pre‐release weight on post‐release growth. Ourdata appears to support this hypothesis: all translocatedindividuals regardless of sex, release method, cohort,behavior, and physiology tended to have similar post‐releasegrowth. Future translocations using soft‐ versus hard‐releasemethods on larger cohorts of translocated devils are needed toclarify these issues. In addition, reliable measures of habitatquality at release sites (along with measures of temporal andspatial heterogeneity) may need to be developed if apredictive translocation strategy is desired for Tasmaniandevils. In the meantime, our results suggest that hard‐releasemethods are sufficient during translocation of Tasmaniandevils, but perhaps only when combined with suitable high‐quality habitats [e.g., Ewen and Armstrong, 2007; Watterset al., 2003].

One of the main aims of most translocation programs isthat animals remain and settle at the release site [Christieet al., 2011; Stamps et al., 2009; Stamps andSwaisgood, 2007]. We found that 91% of the variation inlinear devil movement could be explained by an individual’ssex, its cohort, and release method. Specifically, devils fromthe 2007 cohort traveled five times farther than those from the2006 cohort. The 2006 orphan cohort were released just 4–5months prior to their first breeding season (2‐year‐old femalesbreed about 1 month later than mature adults; breeding indevils is annual with a mean birth date across Tasmanianpopulations of 20 March [Hesterman, 2008]). This 2006cohort, therefore, were at the end of their subadult year; in thewild, they would have dispersed and settled in their adulthome range by this time. In contrast, the 2007 cohort werereleased at dispersal age, in the months around and followingweaning (weaning is late Jan to mid‐Feb in the wild), whichmay help explain their increased movement. Taken together,these results are encouraging, and suggest that the additionaltime spent in captivity did not impact on the 2006 cohort’sability to settle and establish home ranges in the wild. Indeed,females from the 2006 cohort successfully bred during theirfirst breeding season after translocation (due to their age, the2007 cohort females would not have been expected to breedin their first year in a high density population such as theForestier Peninsula [Jones et al., 2008]).

While behavior, sex, pre‐release weight, and releasemethod did not predict final linear distances or home rangesize, devils that tended to have higher basal circulatingcortisol pre‐release also had larger 95% MCP core rangesthan devils with lower circulating cortisol. In addition, orphan

devils overall had lower baseline circulating cortisolconcentrations than wild cohort devils. While differencesin baseline cortisol between wild and orphan devils was alsoconfounded by time (i.e., wild devil cortisol readings weretaken several months after orphan cortisol measures), theseresults provide some preliminary evidence that captivity maybe altering stress physiological phenotypes of devils [Joneset al., 2005], perhaps due to a lack of predation threat incaptivity. If captive conditions are indeed altering howindividuals select, settle into, and retain different habitats orhome ranges via hormonal mechanisms, further study onstress physiology in both healthy wild devils as well as theircaptive counterparts will be necessary to fully delineate anycaptive adaptations in stress responses.

Previous to this study there was no scientific evidenceregarding translocation survival in devils. We found thatoverall survival rates in our translocated devils weremoderate(�42%) and were similar to that observed in many othertranslocation programs [e.g., Beck et al., 1994; Juleet al., 2008]. Survival rates of male and female translocateddevils were not different from one another, and were close topreviously documented survival rates for wild devils of 50%for both the sexes [Lachish et al., 2007; 2009; McCallumet al., 2009]. Intuitively, we expected that spending longerperiods of time in captivity (i.e., the 2006 cohort), a lack ofrelease support (i.e., hard release devils), and lower pre‐release bodyweights should have lowered survival rates [e.g.,Letty et al., 2000]. Perhaps surprisingly then, we detected nodifference in survival between release method (hard vs. soft)or a cohort effect, and no influence of pre‐release weight ondevil survival. As well, our measures of hormonal stressresponses did not have a strong relationship to post‐releasesurvival.

Instead, we found a strong effect of an individualdevil’s responses to novelty in a familiar environment on itspost‐release survival during our study. Indeed, the 2006cohort devils that survived were three and a half times as boldas their counterparts that did not survive, and this was perhapsthe strongest pre‐release predictor of post‐release outcomesdocumented here. Boldness is often a large component of anindividual’s fitness [Réale et al., 2007; Smith andBlumstein, 2008], and recent evidence has begun to suggestthat individual variation in behavior may be a largedeterminant of translocation success [Bremner‐Harrisonet al., 2004; Mathews et al., 2005; Pinter‐Wollman, 2009;Pinter‐Wollman et al., 2009]. While our results are promisingin terms of offering a phenotypic character that could be usedtomanipulate the composition of release groups, we also offercaution that the direction of the relationship between aparticular type of behavior and individual fitness (i.e., thefitness landscape of particular behavioral types) is mostalways unknown. It is also likely that fitness landscapes ofbehavior change spatially and temporally [Dingemanseet al., 2004, 2011; Sinn et al., 2010b; Mathot et al., 2012].In other words, while we found that increased boldnessimproved survival in one cohort during translocation, it is

Translocation of Tasmanian Devils 45

Zoo Biology

possible that this relationship could change given differentspatial or temporal circumstances (i.e., individuals that areless bold could be favored in some cohorts in some years insome locations). Given the uncertainty of knowing whichbehavioral types may succeed in any given year, weemphasize that management efforts would do well to bothmaintain habitat variability in the wild as well as behavioralvariability in the current Tasmanian devil insurance popula-tion [Watters et al., 2003; Watters and Meehan, 2007]. Ourresults suggest that developing measures of habitat quality atrelease sites along with continued measures of individualvariation could be a primary focus of future translocationefforts in this and other endangered carnivores.

Inferences and Sample Sizes

Our study, similar to other work in other endangeredspecies programs [Soulé, 1986; Kattan, 1992], was necessar-ily restricted by the limited sample sizes of orphan devilsavailable for release. Due to the small sample sizes, it is worthnoting that a lack of an effect of several of the putative factorstested here may indeed have an impact on translocationgrowth, movement, or survival, but we did not have thestatistical power here to detect such effects. Similarly, ourstrong result of boldness on survival may have been affectedby sample sizes—if our sample of orphans was not trulyrepresentative of the larger captive‐reared Tasmanian devilpopulation we may have detected an effect of boldness whenin actuality there is none. Despite these caveats, publicinformation reporting of endangered species actions is acritical component of any conservation science [Caro andSherman, 2011, 2013], and is extremely useful for adaptivemanagement techniques that require prior information onwhich tomake future management decisions. Indeed, it is alsoa valid inference that since we detected a strong effect ofboldness on survival despite our small sample sizes, the effectof variation in behavior on subsequent survival in trans-located devils may indeed be very strong. Again, replicationis a basic scientific inference tool, applicable to all scientificstudies [Ryan, 2011], and the need for replication of ourresults is not unique in this respect.

Management Implications

Currently, the only option for saving Tasmanian devilsfrom extinction is an “insurance metapopulation,” whichincludes maintaining a combination of disease‐free popula-tions across intensive captive situations for up to 50 years,until DFTD causes extinction in the wild or evolution of thetumor or the devil leads to coexistence of host and pathogen[Jones et al., 2007]. However, housing animals in captiveconditions for such a large number of generations willinvariably result in unintended genetic and phenotypic effects[McPhee, 2004; Jones et al., 2005; Kelley et al., 2006;Christie et al., 2012], with unknown consequences forreintroduction to the wild. Together, we believe a correctinference of our results is that responses to novelty (boldness)

matters to translocation success in Tasmanian devils, and thatthe other factors measured here (sex, cortisol response, pre‐release growth, cohort, and release method) may impact ontranslocation outcomes in Tasmanian devils, but futuretranslocation efforts using larger sample sizes combined withlonger‐term monitoring are needed [Ewen andArmstrong, 2007; Burbidge et al., 2011]. Studies designedto understand the factors that promote behavioral diversityamong the insurance metapopulation, such as habitat qualityand variation, resource competition, and appropriate levels ofstress normally experienced in the wild (such as non‐lethalexperiences with potential predators) are urgently needed.

CONCLUSIONS

1. Factors such as stress [Teixeira et al., 2007], behavior[Bremner‐Harrison et al., 2004], early experience [Stampsand Swaisgood, 2007], and release site and method [Linklateret al., 2012] continue to be common considerations for manycaptive‐release and translocation programs, but there is asurprising lack of experimental evidence of the relativeimpacts of these different factors on translocation success[Mathews et al., 2005; Seddon et al., 2007; Armstrong andSeddon, 2008; Sheean et al., 2012].

2. Our study provides some of the first evidence of theimportance of individual‐level variation in behavior, or“personalities,” on translocation outcomes.

3. In the face of uncertainty regarding the fitness landscape ofbehaviors in the wild, conservation managers should continueto attempt to maintain habitat diversity in the wild and providedevelopmental conditions in captivity that promote a widerange of behaviors.

4. An increased focus on measurements of habitat quality atrelease sites along with a focus on behavioral andphysiological development in captive individuals shouldcontinue to be a primary focus of any translocation orreintroduction efforts for endangered carnivores.

ACKNOWLEDGMENTS

The authors would like to thank John Hamilton andTom, Cynthia, and Matthew Dunbabin for allowing access totheir properties during the study period. Several wildlifecarers and volunteers made significant contributions to caringfor orphaned devils pre‐release, and the public community onthe Forestier Peninsula provided invaluable information onpost‐release sightings of study animals. Special thanks areextended to Patsy Davies of DPIPWE’s orphan care program.Clare Hawkins provided access to GPS collars, Billy Lazenbyprovided assistance with fitting GPS collars, and AngelaAnderson provided key field assistance. All methodsdescribed here complied with all state and federal laws andDPIPWE agency policies (DPIPWE Animal Ethics Commit-tee approval number: 9/2007‐08). Regular veterinary checkswere performed by Kim Skogvold on all individual devils

46 Sinn et al.

Zoo Biology

during orphan care and during holding at Taranna (2006cohort only).

REFERENCESAaltonen K, Bryant AA, Hostetler JA, Oli MK. 2009. Reintroducingendangered Vancouver Island marmots: survival and cause‐specificmortality rates of captive‐born versus wild‐born individuals. Biol Conserv142:2181–2190.

Armstrong DP, Seddon PJ. 2008. Directions in reintroduction biology.Trends Ecol Evol 23:20–25.

BeckBB, Rapaport LG, Stanley PriceMR,WilsonAC. 1994. Reintroductionof captive‐born animals. In: Olney PJS, Mace GM, Feistner ATC, editors.Creative conservation: interactive management of wild and captiveanimals. London, UK: Chapman and Hall. p 264–386.

Bremner‐Harrison S, Prodohl PA, Elwood RW. 2004. Behavioural traitassessment as a release criterion: boldness predicts early death in areintroduction programme of captive‐bred swift fox (Vulpes velox). AnimConserv 7:313–320.

Budaev SV. 2010. Using principal components and factor analysis in animalbehaviour research: caveats and guidelines. Ethology 116:472–480.

BurbidgeAH,MaronM,ClarkeMF, et al. 2011. Linking science and practicein ecological research and management: how can we do it better? EcolManag Restor 12:54–60.

Carere C, Caramaschi D, Fawcett TW. 2010. Covariation betweenpersonalities and individual differences in coping with stress: convergingevidence and hypotheses. Curr Zool 56:728–740.

Caro T, Sherman PW. 2011. Endangered species and a threatened discipline:behavioural ecology. Trends Ecol Evol 26:111–118.

Caro T, Sherman PW. 2013. Eighteen reasons animal behaviourists avoidinvolvement in conservation. Anim Behav 85:305–312.

Christie K, Craig MD, Stokes VL, Hobbs RJ. 2011. Movement patterns byEgernia napoleonis following reintroduction into restored jarrah forest.Wildl Res 38:475–481.

Christie MR, Marine ML, French RA, Blouin MS. 2012. Genetic adaptationto captivity can occur in a single generation. Proc Natl Acad Sci 109:238–242.

Crews D, Sakata J, Rhen T. 1998. Developmental effects on intersexual andintrasexual variation in growth and reproduction in a lizard withtemperature‐dependent sex determination. Comp Biochem Physiol CPharmacol Toxicol Endocrinol 119:229–241.

Cumming G, Fidler F, Vaux DL. 2007. Error bars in experimental biology. JCell Biol 177:7–11.

Devineau O, Shenk TM, Doherty PF, White GC, Kahn RH. 2011. Assessingrelease protocols for Canada lynx reintroduction in Colorado. J WildlManag 75:623–630.

Dickens MJ, Delehanty DJ, Romero LM. 2009. Stress and translocation:alterations in the stress physiology of translocated birds. Proc R Soc B BiolSci 276:2051–2056.

Dickens MJ, Delehanty DJ, Michael Romero L. 2010. Stress: an inevitablecomponent of animal translocation. Biol Conserv 143:1329–1341.

Dingemanse NJ, Both C, Drent PJ, Tinbergen JM. 2004. Fitnessconsequences of avian personalities in a fluctuating environment. Proc RSoc Lond Ser B 271:847–852.

Dingemanse NJ, Bouwman KM, van de Pol M, et al. 2011. Variation inpersonality and behavioural plasticity across four populations of the greattit Parus major. J Anim Ecol 81:116–126.

Dobson A, Lyles A. 2000. Black‐footed ferret recovery. Science 288:985–987.

Epstein S. 1983. Aggregation and beyond: some basic issues on theprediction of behavior. J Pers 51:360–392.

Ewen JG, Armstrong DP. 2007. Strategic monitoring of reintroductions inecological restoration programmes. Ecoscience 14:401–409.

Fernández‐Morán J, Saavedra D, Ruiz de la Torre JL, Manteca‐Vilanova X.2004. Stress in wild‐caught Eurasian otters (Lutra lutra): effects of a long‐acting neuroleptic and time in captivity. Anim Welf 13:143–149.

Field A, Miles J, Field Z. 2012. Discovering statistics using R. Los Angeles,CA, USA: Sage Publications Inc. 957 p.

Fischer J, Lindenmayer DB. 2000. An assessment of the published results ofanimal relocations. Biol Conserv 96:1–11.

Fleeson W. 2001. Toward a structure‐ and process‐integrated view ofpersonality: traits as density distributions of states. J Pers Soc Psychol80:1011–1027.

Fleeson W. 2004. Moving personality beyond the person–situation debate.Curr Dir Psychol Sci 13:83–87.

Gelling M, Montes I, Moorhouse TP, Macdonald DW. 2010. Captivehousing during water vole (Arvicola terrestris) reintroduction: does short‐term social stress impact on animal welfare? PLoS ONE 5:e9791.

Griffiths R, Murn C, Clubb R. 2010. Survivorship of rehabilitated juvenileTawny Owls (Strix aluco) released without support food, a radio trackingstudy. Avian Biol Res 3:1–6.

Hamilton LP, Kelly PA, Williams DF, Kelt DA, Wittmer HU. 2010. Factorsassociated with survival of reintroduced riparian brush rabbits inCalifornia. Biol Conserv 143:999–1007.

Hardman B, Moro D. 2006. Optimising reintroduction success by delayeddispersal: is the release protocol important for hare‐wallabies? BiolConserv 128:403–411.

Hawkins CE, Baars C, Hesterman H, et al. 2006. Emerging disease andpopulation decline of an island endemic, the Tasmanian devil Sarcophilusharrisii. Biol Conserv 131:307–324.

Hedrick A, Kortet R. 2012. Sex differences in the repeatability of boldnessover metamorphosis. Behav Ecol Sociobiol 66:407–412.

Hesterman H. 2008. Reproductive physiology of the Tasmanian devil(Sarcophilus harisii) and spotted‐tailed quoll (Dasyurus maculatus).Hobart, TAS: University of Tasmania.

Jones SM, Lockhart TJ, Rose RW. 2005. Adaptation of wild‐caughtTasmanian devils (Sarcophilus harrisii) to captivity: evidence fromphysical parameters and plasma cortisol concentrations. Aust J Zool53:339–344.

Jones ME, Jarman PJ, Lees CM, et al. 2007. Conservation management ofTasmanian devils in the context of an emerging, extinction‐threateningdisease: devil facial tumour disease. EcoHealth 4:326–337.

Jones ME, Cockburn A, Hamede RK, et al. 2008. Life‐history change indisease‐ravaged Tasmanian devil populations. Proc Natl Acad Sci USA105:10023–10027.

Jule KR, Leaver LA, Lea SEG. 2008. The effects of captive experience onreintroduction survival in carnivores: a review and analysis. Biol Conserv141:355–363.

Kattan GH. 1992. Rarity and vulnerability—the birds of the cordillera centralof Colombia. Conserv Biol 6:64–70.

Kelley JL, Magurran AE, Garcia CM. 2006. Captive breeding promotesaggression in an endangered Mexican fish. Biol Conserv 133:169–177.

Kleiman DG. 1989. Reintroduction of captive mammals for conservation.Bioscience 39:152–161.

Koolhaas JM, Korte SM, De Boer SF, et al. 1999. Coping styles in animals:current status in behavior and stress‐physiology. Neurosci Biobehav Rev23:925–935.

Lachish S, Jones M, McCallum H. 2007. The impact of disease on thesurvival and population growth rate of the Tasmanian devil. J Anim Ecol76:926–936.

Lachish S, McCallumH, JonesM. 2009. Demography, disease and the devil:life‐history changes in a disease‐affected population of Tasmanian devils(Sarcophilus harrisii). J Anim Ecol 78:427–436.

Lachish S, McCallum H, Mann D, Pukk CE, Jones ME. 2010. Evaluation ofselective culling of infected individuals to control Tasmanian devil facialtumor disease. Conserv Biol 24:841–851.

Lachish S, Miller KJ, Storfer A, Goldizen AW, Jones ME. 2011. Evidencethat disease‐induced population decline changes genetic structure and altersdispersal patterns in the Tasmanian devil. Heredity 106:172–182.

Letty J, Marchandeau S, Clobert J, Aubineau J. 2000. Improvingtranslocation success: an experimental study of anti‐stress treatment andrelease method for wild rabbits. Anim Conserv 3:211–219.

Linklater WL, Gedir JV, Law PR, et al. 2012. Translocations as experimentsin the ecological resilience of an asocial mega‐herbivore. PLoS ONE 7:e30664.

Mangel M, Stamps JA. 2001. Trade‐offs between growth and mortality andthe maintenance of individual variation in growth. Evol Ecol Res 3:583–593.

Mat�ej 8u J, �Rí�canováŠ, Poláková S, et al. 2012.Method of releasing and numberof animals are determinants for the success of European ground squirrel(Spermophilus citellus) reintroductions. Eur J Wildl Res 58:473–482.

Mathews F, Orros M, McLaren G, Gelling M, Foster R. 2005. Keeping fit onthe ark: assessing the suitability of captive‐bred animals for release. BiolConserv 121:569–577.

Mathot KJ, Wright J, Kempenaers B, Dingemanse NJ. 2012. Adaptivestrategies for managing uncertainty may explain personality‐relateddifferences in behavioural plasticity. Oikos 121:1009–1020.

Translocation of Tasmanian Devils 47

Zoo Biology

McCallum H. 2008. Tasmanian devil facial tumour disease: lessons forconservation biology. Trends Ecol Evol 23:631–637.

McCallum H, Jones M. 2006. To lose both would look like carelessness:Tasmanian devil facial tumour disease. PLoS Biol 4:1671–1674.

McCallum H, Jones M, Hawkins C, et al. 2009. Transmission dynamics ofTasmanian devil facial tumor disease may lead to disease‐inducedextinction. Ecology 90:3379–3392.

McPhee ME. 2004. Generations in captivity increases behavioral variance:considerations for captive breeding and reintroduction programs. BiolConserv 115:71–77.

Mitchell AM, Wellicome TI, Brodie D, Cheng KM. 2011. Captive‐rearedburrowing owls show higher site‐affinity, survival, and reproductiveperformance when reintroduced using a soft‐release. Biol Conserv144:1382–1391.

Molinari‐Jobin A, Marboutin E, Wolfl S, et al. 2010. Recovery of the Alpinelynx Lynx lynx metapopulation. Oryx 44:267–275.

Murn C, Hunt S. 2008. An assessment of two methods used to release redkites (Milvus milvus). Avian Biol Res 1:53–57.

Pinter‐WollmanN. 2009. Spatial behaviour of translocated African elephants(Loxodonta africana) in a novel environment: using behaviour to informconservation actions. Behaviour 146:1171–1192.

Pinter‐WollmanN, Isbell LA,Hart LA. 2009. Assessing translocation outcome:comparing behavioral and physiological aspects of translocated and residentAfrican elephants (Loxodonta africana). Biol Conserv 142:1116–1124.

Pukk C. 2005. The habitat use of the Tasmanian devil (Sarcophilus harrisiilaniarius) across natural and pastoral mosaics [honours thesis]. Hobart,Australia: University of Tasmania.

Quinn JL, Cresswell W. 2005. Personality, anti‐predation behaviour andbehavioural plasticity in the chaffinch Fringilla coelebs. Behaviour142:1377–1402.

Rands SA, Cowlishaw G, Pettifor RA, Rowcliffe JM, Johnstone RA. 2003.Spontaneous emergence of leaders and followers in foraging pairs. Nature423:432–434.

Réale D, Reader SM, Sol D, McDougall PT, Dingemanse NJ. 2007.Integrating animal temperament within ecology and evolution. Biol Rev82:291–318.

Robinette KW, Andelt WF, Burnham KP. 1995. Effect of group size onsurvival of relocated prairie dogs. J Wildl Manag 59:867–874.

Rödel HG, Holst Dv, Kraus C. 2009. Family legacies: short‐ and long‐termfitness consequences of early‐life conditions in female European rabbits. JAnim Ecol 78:789–797.

Rouco C, Ferreras P, Castro F, Villafuerte R. 2010. A longer confinementperiod favors European wild rabbit (Oryctolagus cuniculus) survivalduring soft releases in low‐cover habitats. Eur J Wildl Res 56:215–219.

Rout TM, Hauser CE, PossinghamHP. 2009. Optimal adaptive managementfor the translocation of a threatened species. Ecol Appl 19:515–526.

Ryan MJ. 2011. Replication in field biology: the case of the frog‐eating bat.Science 334:1229–1230.

Sarrazin F, Barbault R. 1996. Reintroduction: challenges and lessons forbasic ecology. Trends Ecol Evol 11:474–477.

Sarrazin F, Legendre S. 2000. Demographic approach to releasing adultsversus young in reintroductions. Conserv Biol 14:488–500.

Seddon PJ, Armstrong DP, Maloney RF. 2007. Developing the science ofreintroduction biology. Conserv Biol 21:303–312.

Sheean VA, Manning AD, Lindenmayer DB. 2012. An assessment ofscientific approaches towards species relocations in Australia. Austral Ecol37:204–215.

Shier DM, Swaisgood RR. 2012. Fitness costs of neighborhood disruption intranslocations of a solitary mammal. Conserv Biol 26:116–123.

Sih A, Watters JV. 2005. The mix matters: behavioural types and groupdynamics in water striders. Behaviour 142:1417–1431.

Sih A, Kats LB, Maurer EF. 2003. Behavioural correlations across situationsand the evolution of antipredator behaviour in a sunfish–salamandersystem. Anim Behav 65:29–44.

Sih A, Bell AM, Johnson JC, Ziemba RE. 2004. Behavioral syndromes: anintegrative overview. Q Rev Biol 79:242–277.

Sinervo B. 2001. Selection in local neighborhoods, the social environment,and ecology of alternative strategies. In: Dugatkin LA, editor. Modelsystems in behavioral ecology: integrating conceptual, theoretical, andempirical approaches. Princeton, NJ: Princeton University Press. p 191–226.

Sinn DL, Macnab K, Sharman A. 2010a. Save the Tasmanian devil free‐range project, annual report January–December 2009: initial results fromthe first year’s breeding season and recommendations for future planning.Save the Tasmanian devil program. Hobart, TAS, Australia: ISBN: 978‐7246‐6528‐0.

SinnDL,Moltschaniwskyj NA,Wapstra E, Dall SRX. 2010b.Are behavioralsyndromes invariant? Spatiotemporal variation in shy/bold behavior insquid. Behav Ecol Sociobiol 64:693–702.

Smith BR, Blumstein DT. 2008. Fitness consequences of personality: a meta‐analysis. Behav Ecol 19:448–455.

Soulé ME. 1986. Conservation biology: the science of scarcity and diversity.Sunderland, MA, USA: Sinauer Associates.

Stamps JA, Swaisgood RR. 2007. Someplace like home: experience, habitatselection and conservation biology. Appl Anim Behav Sci 102:392–409.

Stamps JA, Krishnan VV, Willits NH. 2009. How different types of natalexperience affect habitat preference. Am Nat 174:623–630.

Tabachnick BG, Fidell LS. 1996. Using multivariate statistics. 3rd edition.New York, NY, USA: HarperCollins.

Teixeira CP, Schetini de Azevedo C, Mendl M, Cipreste CF, Young RJ.2007. Revisiting translocation and reintroduction programmes: theimportance of considering stress. Anim Behav 73:1–13.

van Oers K, Sinn DL. 2013. The quantitative and molecular genetics ofanimal personality. In: Carere C, Maestripieri D, editors. Animalpersonalities: behavior, physiology, and evolution. Chicago, IL, USA:The University of Chicago Press. p 149–200.

van Oers K, Klunder M, Drent PJ. 2005. Context dependence ofpersonalities: risk‐taking behavior in a social and a nonsocial situation.Behav Ecol 16:716–723.

Veenema AH, Meijer OC, de Kloet ER, Koolhaas JM, Bohus B. 2003.Differences in basal and stress‐induced HPA regulation of wild house miceselected for high and low aggression. Horm Behav 43:197–204.

Watters JV, Meehan CL. 2007. Different strokes: can managing behavioraltypes increase post‐release success? Appl Anim Behav Sci 102:364–379.

Watters JV, Lema SC, Nevitt GA. 2003. Phenotype management: a newapproach to habitat restoration. Biol Conserv 112:435–445.

Wedekind C. 2002. Sexual selection and life‐history decisions: implicationsfor supportive breeding and the management of captive populations.Conserv Biol 16:1204–1211.

While GM, Isaksson C, McEvoy J, et al. 2010. Repeatable intra‐individualvariation in plasma testosterone concentration and its sex‐specific link toaggression in a social lizard. Horm Behav 58:208–213.

Wingfield JC. 2003. Control of behavioural strategies for capriciousenvironments. Anim Behav 66:807–816.

48 Sinn et al.

Zoo Biology