Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Translocation stress in Stephens' kangaroo rats : how individual variation influences success Baker, Liv 2014

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2014_spring_baker_liv.pdf [ 1.6MB ]
Metadata
JSON: 24-1.0165863.json
JSON-LD: 24-1.0165863-ld.json
RDF/XML (Pretty): 24-1.0165863-rdf.xml
RDF/JSON: 24-1.0165863-rdf.json
Turtle: 24-1.0165863-turtle.txt
N-Triples: 24-1.0165863-rdf-ntriples.txt
Original Record: 24-1.0165863-source.json
Full Text
24-1.0165863-fulltext.txt
Citation
24-1.0165863.ris

Full Text

 TRANSLOCATION STRESS IN STEPHENS? KANGAROO RATS:  HOW INDIVIDUAL VARIATION INFLUENCES SUCCESS  by Liv Baker B.A., Mount Holyoke College, 1998 M.Sc., University of Massachusetts, Amherst, 2003   A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY in The Faculty of Graduate and Postdoctoral Studies (Applied Animal Biology)  THE UNIVERSITY OF BRITISH COLUMBIA (VANCOUVER)  January 2014  ? Liv Baker, 2014  ii Abstract   Wildlife translocation involves the relocation of free-ranging animals from one area to another. It is commonly used to combat species loss. However, its outcomes are poor; some reviews put the success rate as low as 10-25%. This is likely influenced by a lack of attention to individuals. Translocations involve some combination of stressors (e.g., capture, captivity, monitoring, environmental change). Although animals have evolved behavioural and physiological mechanisms to manage challenges, the combination of stressors during translocation can compromise these coping mechanisms. Personality should have significance for translocations as individuals with certain personality types and life experience may handle translocation stressors better than others. The aim of this thesis was to profile individual Stephens? kangaroo rats (SKR), Dipodomys stephensi, and to assess their responses to translocation. To do this a combination of behavioural and physiological measures were used. Personality types were identified using quantitative and qualitative measures from mirror-image stimulation and predator scent tests done while animals were held before release (Chapters 3 and 5). A radioimmunoassay specific for cortisol in SKR fecal extract was developed and adrenocortical activity (cortisol) in response to predator urine was reliably assessed in fecal samples (Chapter 2). Fecal cortisol concentration (FCC) was used to measure the effect of translocation stressors, including the use of radio transmitters (Chapters 4 and 5). Survival was affected by individual variation in behavioural and physiological responses (Chapter 5). Assertiveness, Excitability and Persistence were identified as three personality dimensions. Overall, FCC increased in response to temporary captivity. Radio transmitters caused a short-term elevation in FCC 6 days after  iii attachment but not at 30 days. Survival to 1 month was similar for animals with and without transmitters. SKR with lower Assertiveness and Excitability and with higher basal FCC had higher short-term survival. Higher Assertiveness was correlated with lower basal FCC. SKR that had a smaller change in FCC during captivity had higher long-term survival. This study lends convincing support that variation in personality affects how well an animal copes with translocation and has consequences for survival. Knowing how to manage different personality types may determine how successfully a translocated population establishes itself.    iv Preface   Liv Baker performed the study design, performance, statistical analyses, interpretation and write-up for all chapters in this dissertation, under the supervision of Dr. David Fraser. All research was performed in strict accordance with US federal (USFWS), state (California) and institutional (San Diego Zoo Global; University of British Columbia) standards. As the research was reviewed by the Institutional Animal Care and Use Committee (IACUC) at San Diego Zoo Global, the UBC Animal Care Committee accepted the IACUC proposal and approval certificate in lieu of a UBC certificate number. Federal permit number TE-142435-4; California state permit number SC-002508; San Diego Zoo IACUC number 09-007.      v Table of Contents  Abstract .............................................................................................................................. ii	 ?Preface ............................................................................................................................... iv	 ?Table of Contents .............................................................................................................. v	 ?List of Tables ................................................................................................................... vii	 ?List of Figures ................................................................................................................. viii	 ?Acknowledgements ........................................................................................................... x	 ?Dedication ........................................................................................................................ xii	 ?1.	 ? General introduction ................................................................................................. 1	 ?1.1 Translocation biology ........................................................................................... 4	 ?1.2 Stress physiology ................................................................................................ 12	 ?1.3 Animal personality .............................................................................................. 19	 ?1.4 Study species ....................................................................................................... 30	 ?2.	 ? Validation of a fecal glucocorticoid assay for the endangered Stephens? kangaroo rat using behavioural and pharmacological tests ................................ 33	 ?2.1 Introduction ......................................................................................................... 33	 ?2.2 Materials and methods ........................................................................................ 35	 ?2.3 Statistical analyses .............................................................................................. 42	 ?2.4 Results ................................................................................................................. 43	 ?2.5 Discussion ........................................................................................................... 52	 ?3.	 ? Composition of personality differences in a translocated population of Stephens? kangaroo rat ............................................................................................................. 56	 ?3.1 Introduction ......................................................................................................... 56	 ?3.2 Materials and methods ........................................................................................ 58	 ?3.3 Statistical analyses .............................................................................................. 63	 ?3.4 Results ................................................................................................................. 64	 ?3.5 Discussion ........................................................................................................... 72	 ?4.	 ? The effect of radio transmitters on fecal cortisol concentrations and survival in translocated Stephens? kangaroo rats .................................................................... 76	 ?4.1 Introduction ......................................................................................................... 76	 ?4.2 Materials and methods ........................................................................................ 78	 ?4.3 Statistical analyses .............................................................................................. 81	 ?4.4 Results ................................................................................................................. 82	 ?4.5 Discussion ........................................................................................................... 86	 ?5.	 ? Translocation success of Stephens? kangaroo rats: a behavioral and physiological profile of survivors ........................................................................... 89	 ?5.1 Introduction ......................................................................................................... 89	 ?5.2 Materials and methods ........................................................................................ 90	 ?5.3 Statistical analyses .............................................................................................. 96	 ?5.4 Results ................................................................................................................. 97	 ?5.5 Discussion ......................................................................................................... 103	 ? vi 6.	 ? General discussion and conclusions ..................................................................... 107	 ?6.1 Introduction ....................................................................................................... 107	 ?6.2 Chapter review .................................................................................................. 109	 ?6.3 General recommendations and future research ................................................. 116	 ?References ...................................................................................................................... 119	 ?        vii List of Tables  Table 1 Intra-class correlation coefficients (ICC) of term ratings for Stephens? kangaroo rats in mirror-image (MIS) and predator scent (PS) tests ...................................... 67	 ?Table 2 Principal Component Analysis term loadings for Stephens? kangaroo rats in mirror-image (MIS) and predator scent (PS) tests. Component designations (Assertiveness, Persistence, Excitability) are based on term loadings. .................. 68	 ?Table 3 Correlations between behaviours and personality components of Stephens?s kangaroo rats from mirror-image (MIS) and predator scent (PS) tests .................. 71	 ?Table 4 Behaviours from mirror-image (MIS) and predator scent (PS) tests associated with Stephens? kangaroo rats from two populations (P Lot and El Sol) .............. 101	 ?     viii List of Figures  Fig. 1 Immunochromatogram of HPLC fractions assayed with monoclonal mouse anti-cortisol (071210107, MP Biomedicals) for Stephens? kangaroo rat. a = cortisol; b = 5a-pregnane-3B,11B,21-triol-20-one; b = corticosterone; c = 5B-androstane-3a-ol-11-17-dione; d = testosterone; e = esterone sulfate; f = progesterone ............... 46	 ?Fig. 2 Immunochromatogram of HPLC fractions assayed with polyclonal rabbit anti-corticosterone (CMJ006; University of California, Davis) for Stephens? kangaroo rat. a = cortisol; b = 5a-pregnane-3B,11B,21-triol-20-one; b = corticosterone; c = 5B-androstane-3a-ol-11-17-dione; d = testosterone; e = esterone sulfate; f = progesterone ........................................................................................................... 47	 ?Fig. 3 Immunochromatogram of HPLC fractions assayed with polyclonal rabbit anti-corticosterone (071201123, MP Biomedicals) for Stephens? kangaroo rat. a = cortisol; b = 5a-pregnane-3B,11B,21-triol-20-one; b = corticosterone; c = 5B-androstane-3a-ol-11-17-dione; d = testosterone; e = esterone sulfate; f = progesterone ........................................................................................................... 48	 ?Fig. 4 Fecal cortisol concentrations (ng/g) of Stephens? kangaroo rat for each exposure treatment (predator urine, mirror and unexposed) at time of capture and after treatment. Data are shown as medians with the 25th and 75th percentile. Whiskers show the 5th and 95th percentile. Outliers are presented as dots. ............................ 49	 ?Fig. 5 Fecal cortisol concentrations ng/g [ln (FC+1)] over 24 h collection period of Stephens? kangaroo rats after dexamethasone (1?g/g) and saline injection in 2011. Data are shown as medians with the 25th and 75th percentile. Whiskers show the 5th and 95th percentile. Outliers are presented as dots. ................................................ 50	 ?Fig. 6 Fecal cortisol concentrations ng/g [ln (FC+1)] over 24 h collection period of Stephens? kangaroo rats after dexamethasone (8?g/g) and saline injection in 2012. Data are shown as medians with the 25th and 75th percentile. Whiskers show the 5th and 95th percentile. Outliers are presented as dots. ................................................ 51	 ?Fig. 7 Interaction of individual factor scores of Stephens? kangaroo rats for each personality dimension (Assertiveness, Excitability, Persistence), derived from the PCA for mirror-image (MIS) test. Assertiveness is represented as a categorical variable ................................................................................................................... 69	 ?Fig. 8 Interaction of individual factor scores of Stephens? kangaroo rats for each personality dimension (Assertiveness, Excitability, Persistence), derived from the PCA for predator scent (PS) test. Assertiveness is represented as a categorical variable ................................................................................................................... 70	 ?Fig. 9 Percent survival of Stephens? kangaroo rats at 1 month after release for Transmitter (N = 9) and No transmitter treatment groups (N = 10); difference is non-significant ................................................................................................................................ 84	 ?Fig. 10 Fecal cortisol concentrations for SKR in Transmitter and No transmitter (control) treatment groups. Wild and Captive periods were prior to transmitter outfitting. (Control, mean?SE: Wild, 16.9?2.1; Captive, 21.4?2.08: Day 1, 21.6?3.2; Day 6,  ix 18.1?3.3; 1 mo post-release, 21.2?2.7. Transmitter, mean?SE: Wild, 17.3?2.7; Captive, 24.2?3.1: Day 1.48.1?13.1; Day 6, 35.3?7.4; 1 mo post-release, 25.6?5.9.) Data are shown as medians with the 25th and 75th percentile. Whiskers show the 5th and 95th percentile. Outliers are presented as dots. ............................ 85	 ?Fig. 11 One-zero score (proportion of all sample intervals during which the behaviour occurred) of Stephens? kangaroo rats from 2 populations (P Lot and El Sol); Sampling occurred during captivity, before translocation.. ................................. 100	 ?Fig. 12 Fecal cortisol concentrations (mean?SE) at time of capture and during captivity, prior to translocation, for Stephens? kangaroo rats from two populations (P Lot and El Sol) ................................................................................................................... 102	 ?   x Acknowledgements   I am grateful for the support and guidance of supervisory committee members Debra Shier, Lance Barrett-Lennard and Nina von Keyserlingk. I am equally grateful to Dan Weary ? an unrivaled devil?s advocate and to Nina again for the many hats she wears so well. To my advisor, David Fraser, I am beholden. A truer mentor I could not have asked for. I was lucky to have received doctoral funding from the Killam scholarship and UBC?s four-year fellowship. I would like to thank Kai Chan and certainly Tom Sullivan for making my advancement to candidacy a challenging pleasure rather than a painful ordeal. I am better for the conversation. I was fortunate to have worked with remarkable students during my doctoral research. Many thanks to Mike Lawrence, Christina Tse, Mary Toews, Sean Kuling, Ivy Wu, Amy Maunsell, Nancy Chen, Yasmine Yavari and Alex Boo for loving the roo rats (almost) as much as I. My research divided my time between southern California and Vancouver - I was blessed with wonderful support in both places. In California the know-how and assistance of Christine Moen, Tom Ash and the staff at Lake Skinner Reserve were invaluable. At San Diego Zoo?s ICR Matt Milnes, Corinne Pisacane, Maryke Swartz and the incomparable Jeff Zuba made things just that much easier. I am especially grateful to Ron Swaisgood for his help with the collaboration. In Vancouver, the openness and compassion of the folks in the Animal Welfare Program make me privileged to be part of such a unique group of scholars. To Amelia ? I could not have asked for a smarter, funnier, better-dressed co-conspirator. Chris McGill ? without his facility, good spirit and generous time all would be lost. I thank my family for having the grace to no longer ask when I would finish. I am thankful for the friendship of Katie and Matt, forever bonded by heteromyids. My deepest gratitude to Amy (Shakes), Becca, Elizabeth, Jo, Kristen, Sara, and Susan ? women of beauty, drive and heart, all of who inspire me. And to Duke, of course. Thank you.    Chapter 2 I thank Corinne Pisacane, Alyssa Hall, and Alan Fetter for their assistance in the lab, and Michael Lawrence, Christina Tse, Angelique Herman and Sarah Brewster for their help with fecal collection and exposure trials. I thank Christine Moen for access to the Reserve and RCHCA for funding provided to Debra Shier for part of this research. I am grateful for the veterinary support and expertise of Dr. Jeff Zuba.  Chapter 3 I thank Corinne Pisacane and Alan Fetter for their assistance in the lab, and Amanda Lea and James Liu for help with trial recordings. I thank Christine Moen for  xi access to the Reserve and RCHCA for funding provided to Debra Shier for part of this research. I am grateful for the veterinary support and expertise of Dr. Jeff Zuba.  Chapter 4 I thank Christine Moen, Tom Ash and Dave Bircheff for assistance on the ground, and Michael Lawrence, Maryke Swartz, Amanda James and Juliana Trunzo for assistance with fieldwork. This work was supported by San Diego Zoo Global and a grant to Debra Shier from Riverside County Habitat Conservation Agency.  Chapter 5 I thank Corinne Pisacane and Alan Fetter for their assistance in the lab, and Michael Lawrence, Christina Tse, Angelique Herman and Sarah Brewster for their help with behavioural observations, personality tests and fecal collection. I thank Christine Moen for access to the Reserve and RCHCA for funding provided to Debra Shier for part of this research.        xii Dedication           For the beautiful fur balls in my life and always in my heart ~ Zingiber, David Christopher, Finn and Mulligan  And for the roo rats 1 1. General introduction  Animal welfare and conservation biology represent mission-oriented science (Soul?, 1985; Fraser, 2010). Each is responsive to present and emerging social concerns, and serves to guide policy and practice (Fraser, 2010). At a basic level these fields share a guiding ethic of the protection of animals; however, the fundamental objectives of these fields have been shaped by different concerns, which in turn have shaped their scope of protection. The welfare of wild, free-ranging animals, as it stands, has been largely excluded from the scope of either field. Animal welfare developed from concerns over the quality of life of domestic and captive animals. Rooted in these interests, animal welfare science has concentrated on animals under direct human care or management. Traditionally, this has included animals in food production, science, and entertainment (Fraser and MacRae, 2011). In keeping with this convention, the welfare of wild animals has been focused mainly on captivity and the sport and commercial aspects of activities such as hunting and trapping (Carlstead and Shepherdson, 2000; Warburton et al., 2008).  By contrast, the modern conservation movement developed from concerns over the loss of wilderness and extinction of species through exploitation. Within the confines of these concerns the protection of wild animals has historically functioned as an extension of wilderness preservation. Conservation biology, as a result, has focused not on the welfare of the individual, but on the health of populations, preservation of species and overall ecosystem biodiversity (Soul?, 1985; Norton, 1995; Rawles, 1997; Vucetich and Nelson, 2007; Fraser, 2010).  Thus, while the protection of animals may be an interface between animal welfare science and conservation biology, the ethical norms that have traditionally directed these fields are distinct. Recent changes in our understanding of human activity on wildlife, including the  2 intensification of conservation programs, warrant a reevaluation of this distinction. Translocation biology offers an important opportunity towards this end, primarily because the practice of translocation demands the direct care and management of wild, free-ranging animals as a means to species survival.  Modern wildlife translocation involves the deliberate capture and release of free-ranging animals from one area to another. It is a common conservation practice used to combat loss of species and habitat, and is an integral part of the recovery plans for many at-risk species. Despite the global popularity of translocation, its outcomes have been notoriously dismal (Kleiman, 1989; Griffith et al., 1989; Stamps and Swaisgood, 2007; Teixeira et al., 2007); some reviews put the success rate as low as 10-25% (Beck et al., 1994; Wolf et al., 1998).  Typical translocations involve some combination of stressors, such as capture, captivity, marking, monitoring, transport and handling, in addition to environmental and social disturbance (Letty, 2007; Teixeira et al., 2007). Although animals have evolved a repertoire of behavioural and physiological coping mechanisms to manage challenges throughout their lives, these adaptations are a function of the physical and social environment in which the animals live. Translocation stressors - particularly in combination - can compromise the mechanisms that manage vital events such as foraging, navigation, reproduction and predator avoidance, rendering what an animal has learned throughout its life useless when moved suddenly to a new environment (Teixeira et al., 2007; Swaisgood, 2010). If this is the case, translocations threaten individual welfare and actually conflict with the conservation goals of the practice. Thus, being able to pinpoint sources of stress, and knowing how to effectively measure stress for translocated animals, has great importance for animal conservation and welfare.  3 For decades, physiological stress responses have been an established component of animal welfare science, used to assess a variety of welfare concerns. For example, scientists have used stress responses to study pain associated with industry practices, frustration and distress caused by confinement and the physiological correlates of abnormal behaviour associated with housing conditions (Petrie et al., 1996; Korte et al., 1997; Anil et al., 2005). Recognizing the need to manage translocation stress, consideration of how and where stress emerges for translocated animals has gained currency in the field of conservation biology (e.g., McEwen and Sapolsky, 1995; Mendl, 1999; Letty et al., 2000; Teixeira et al., 2007; Dickens et al., 2010). Specifically, there has been a surge in the use of glucocorticoid (GC) levels as an indicator of stress in conservation-oriented studies (Busch and Hayward, 2009). While a growing number of studies show that concentrations of GCs (notably cortisol and corticosterone) are affected by translocation events (e.g., Hartup et al., 2005; Franceschini et al., 2008; Viljoen et al., 2008; Dickens et al., 2009; Pinter-Wollman et al., 2009; Gelling et al., 2012), these studies report only overall treatment effects and overlook the variation among individual animals. The individual variation could be important. Research reveals that stressors differentially affect individuals because of personality type. For example, variation in animals? aggression levels, motivation to explore and approach novel objects, as well as the underlying physiological mediators, cause individuals to be more or less vulnerable in a challenging environment (Verbeek et al., 1994; Verbeek et al., 1996; Dingemanse et al., 2002; Bell, 2004).  A few biologists have argued that personality is an important concept in conservation biology and should have significance for translocations as individuals with certain personality types and certain life experience may handle the stressors of translocation better than others (McDougall et al., 2006; Dickens et al., 2010; Swaisgood, 2010). But there are scant examples  4 from the translocation literature where consideration at the individual level has been put into practice.   Because of this lack, my research focused on understanding the differential stress effects of the translocation process on individuals, specifically of the endangered Stephens? kangaroo rat (SKR), Dipodomys stephensi. Ultimately, the aim of my dissertation is to provide a scientific foundation on which the animal welfare and conservation concerns of translocation biology can be addressed. Such a foundation should help increase survival, as we are better able to anticipate and reduce the negative impact of stressors on the individuals involved.   1.1 Translocation biology  Throughout human history, various forms of self-interest have motivated people to capture and relocate free-ranging animals. Primary interests have included sustenance, sport, pest control, and to a lesser extent ornamentalism and sentimentalism (Leopold, 1933; Allen, 1954; Laycock, 1966). Only recently has the conservation of threatened species become the primary reason for the relocation of animals. The relocation of wild, free-ranging animals has largely been focused on game species (Leopold, 1933; Laycock, 1966; Griffith et al., 1989). Historically, species were moved to foreign lands, or just outside their natural geographic range, simply to enhance hunting experience and opportunity; more contemporary examples include repeated introductions of the Asian ring-necked pheasant, Phasianus spp., to Europe and North America; the introduction of the Atlantic striped bass, Roccus saxatilis, to the West Coast; and red deer, Cervus elepahus, to New Zealand from England (Laycock, 1966). On occasion the interest in sport was coupled with conservation concerns. As recently as the 1960s, the New Mexico Game and Fish Commission  5 allowed the introduction of the threatened African kudu, Tragelephus strepsiceros, to the state?s thicketed riparian areas. As an extension of European game-keeping principles, areas of introduced game species were often restocked to help establish populations, and if accomplished, offset hunting and angling pressures (Seddon et al., 2007). By the early 20th century there began, at least in the Western world, a programmatic shift from the principles of European game keeping, to the more encompassing practice of game conservation. This is possibly best synthesized by Aldo Leopold, a founder of the science of wildlife management, in his 1933 publication Game Management. Leopold?s proposal was that game conservation, to be effective and enduring, must move beyond the notion of ?restrictive use?, embrace Theodore Roosevelt?s idea of ?conservation through wise use?, and apply biological science in this endeavour. At the time of publication, however, Leopold believed that science revealed more about ?how to distinguish one species from another than of the habits, requirements, and inter-relationships of living populations.? And he went on to suggest that the intentional movement of animals, despite its lengthy history, continued to be done with ?little guidance from either science or experience.? This paucity of knowledge notwithstanding, the intentional movement of animals persisted beyond the 1930s for both non-native and native species (Leopold, 1933; Griffith et al., 1989). Increasingly, however, efforts began to focus on the reintroduction of game species to areas where they had been extirpated (Nielsen, 1987; Griffiths et al., 1989). The mid 1930s saw attempts in North America, for example, to return moose, Alces alces, to the Upper Peninsula of Michigan (1934-37); wild turkey, Meleagris gallopavo, to Indiana countryside (1940s); white-tailed deer, Odocoileus virginanus, to South Carolina (1950s); and black bear, Ursus  6 americanus, to Arkansas?s interior highlands (1958) (Laycock, 1966; Smith and Clark, 1994; Latch and Rhodes, 2005; Beyer et al., 2011). In the decade after Rachel Carson?s Silent Spring (1962) and the passing of the Endangered Species Preservation Act (1966) in the United States, there was a noticeable diversification of reintroduction programs. While the vast majority of efforts still focused on extirpated game species, by the 1970s (a decade of landmark conservation legislation and reform) efforts to reestablish threatened non-game species, particularly birds of prey, intensified. The U.S. Endangered Species Act of 1973 and CITES, the Convention on International Trade in Endangered Species of Wild Fauna and Flora (1975), transformed the protection of threatened species and their habitat worldwide. In the three decades to follow, under the gloom of the extinction crisis, the intentional movement of animals has become a common conservation practice (Griffith et al., 1989). As translocation became an increasingly popular conservation strategy for threatened species, conservation practitioners increasingly became concerned about the standard methodology and high failure rate. Under the urgency of species loss and the regulations governing protected species, the scrutiny of translocation initiatives intensified (Griffith et al., 1989). From this, ?reintroduction biology? emerged as a recognizable field in the late 1990s. The IUCN Guidelines for Reintroductions, published in 1998, spotlighted this emergence. Under the auspices of the IUCN?s Species Survival Commission (SSC), the Reintroduction Specialist Group (RSG) prepared guidelines to address the increasing number of reintroduction initiatives worldwide. Specifically, they were drafted in response to ?the growing need for specific policy guidelines to help ensure that the reintroductions achieve their intended conservation benefit, and do not cause adverse side-effects of greater impact (IUCN 1998).?  7 In 1933, Aldo Leopold confessed that the successful translocation of a species was likely to be ?determined by factors as yet invisible and unknown to science [. . .].? In the transition from modern wildlife management to conservation biology, much of what we learned about translocations came largely from intuition, tradition and guesswork (Seddon et al., 2007). Translocation efforts were at the mercy of a damning trifecta: poor understanding of the animals involved, poor planning, and poor attempts to disseminate information. Over the past 20 years practitioners of translocations have made a concerted effort to attend to these issues (Seddon et al., 2007).  A number of publications influenced the practice of translocation. The IUCN?s 1987 statement (updated in 2012) was the first high-profile, unified summary of the state of translocation programs. While its consequence came more from its intent than from the depth of its content, the document did outline a basic approach to improve translocations; this included the importance of feasibility studies, as well as preparation, release and follow-up phases. Sadly, a contemporaneous review of translocations by Leon Nielsen (then Executive Director of the Wisconsin Humane Society) was largely overlooked. Nielsen?s, article, ?Definitions, Considerations, and Guidelines for Translocation of Wild Animals,? is a comprehensive and pragmatic overview of the critical issues involved in translocation programs. In the planning section alone, Nielsen addressed reasons for translocation, status of the wild population, status of the species range, overall translocation strategy, capture technique, post-capture processing, transportation, captivity, release strategy and post-release monitoring (Nielsen, 1988). His considerations could serve as a valuable reference for current practitioners.  In contrast, the 1989 Science article by Griffith et al., ?Translocation as a Species Conservation Tool: Status and Strategy,? has been cited nearly 600 times. It showed that  8 previous translocation programs were rife with failure. The authors surveyed translocations from 1973-1986. In that period, of the 80 translocations involving threatened and endangered species, 56% failed to establish self-sustaining populations. This paper showed the dismal record of translocations for animals of conservation concern. It also identified tangible factors of translocation success: free-living animals cope better than captive-reared ones; habitat quality matters; and releasing more animals increases survival. A self-sustaining population, the ultimate measure of translocation success, depends on the animals? survival, settlement, and reproduction at the release site (Gosling and Sutherland, 2000; Letty et al., 2003). Translocations can fail because of failure in one or more of these stages. After release, there may be an inadequate number of reproducing survivors; translocated animals may disperse without reproducing; or they may reproduce briefly and then languish as non-breeding adults (Leopold, 1933; Fischer and Lindenmayer, 2000; Armstrong and Seddon, 2008). More often than not, translocated animals die during the initial days or weeks after release (Kleiman, 1989; Short et al., 1992; Beck et al., 1994; Teixeira et al., 2007). By the mid 1990s, researchers were proposing that translocations were unsuccessful because of the limited understanding of the basic ecology and behaviour of the species involved. Several articles (Sarrazin and Barbault, 1996; Clemmons and Buchholz, 1997; Sutherland, 1998 and Caro, 1999) advocated that the principles of behavioural ecology be applied to conservation biology and translocations. The field of reintroduction biology is a tangible instance of this application.  Reintroduction biology is a multidisciplinary science drawing on ecology, population biology, population genetics, and systematics (Clemmons and Buchholz, 1997; Caro, 1999). Reintroduction biology has made good use of animal behaviour research. This was largely because behavioural studies were already being used to improve the captive breeding of animals  9 in translocation programs (Kleiman, 1989; Brittas et al., 1992; Biggins et al., 1999; Miller et al., 1998; Griffin et al., 2000). As an extension, in the 15 years since the formal call for a connection between conservation and behaviour, researchers have looked into, for example, dispersal, foraging, social and anti-predator behaviours, habitat requirements, social structure, and mating habits to enhance survival, settlement and reproduction of translocated animals (Letty, 2000; Gerber et al., 2003; Shier 2006). Nevertheless, the application of behaviour in translocations is limited (Caro, 2007; Swaisgood, 2010).   Since the IUCN?s 1987 position statement on translocation, documented successes remain scarce (Fischer and Lindenmayer, 2000; Swaisgood, 2010). Apart from the failure to establish viable populations, many translocated animals die within days or weeks of release (Beck et al., 1994; Stamps and Swaisgood, 2007; Teixeira et al., 2007). Not surprisingly, nearly every translocation paper calls for a more rigorous investigation of the factors involved in success and failure.   Because reintroduction biology arose from conservation biology, the methodological approach for translocations has been most influenced by disciplines with species and population level concerns rather than concern at the level of individual (Caro, 2007). If we consider what an animal experiences as it moves through the translocation process, we may see that in reality this top-down approach to translocation is not workable and a new conceptual framework is needed for the practice. Swaisgood (2010) eloquently describes this journey of the individual.  An individual animal may suddenly find itself confined in a trap, experiencing frustrated motivation to escape. Then the trapped animal is approached by humans (likely perceived as predators), handled, and transported to a novel environment, and  10 either held for a while in a human-constructed enclosure or perhaps it is released immediately (Griffith et al., 1989). In its home area it has invested considerable time and energy learning how best to exploit local resources (Stamps and Krishnan, 1999; Inglis et al., 2001). It has learned where to find food resources, including those that are seasonally limiting. It knows where predators lurk, how to avoid them, and where to find cover or refuge (Lima and Dill, 1990). It knows its place in the social group and is able to maintain its position with minimal escalated aggression (Huntingford and Turner, 1987; Archer, 1988). If territorial, it has established beneficial relationships with its neighbors that reduce conflict and defense costs (Temeles, 1994; L?pez and Mart?n, 2002). Forced to forego a lifetime of such gradually acquired situation-specific knowledge, the animal must deal with its post-release environment without many of the advantages it retained at home. Under stress and cognitively impaired (Teixeira et al., 2007), the animal ventures out into a dramatically novel and dangerous environment to fend for itself against all odds.   Because of the translocation process, an animal may experience multifarious hardships, having direct and indirect effects. Translocated animals are known to suffer, for example, from trapping injuries, capture and transport myopathy, excessive weight loss, stereotypy, anomalous predation, disease, sub-optimal foraging and navigation, reproduction depression, and isolation (Texeiria et al., 2007). The impact of the translocation process on individual animals has only recently received attention, notably by Letty et al. (2007) and Teixeira et al. (2007), who call attention to potential translocation stressors that affect an animal?s coping abilities and thus influence the survival of translocated animals. Since the publication of these papers several  11 studies have shown that translocation events do affect the levels of stress hormones in translocated individuals (Franceschini, 2008; Pinter-Wollman, 2009; Dickens et al., 2009 a&b; 2010). Moreover, current stress physiology research indicates that stressors differentially affect individuals because of personality types.   Rising concern for translocation stress has also focused attention on animal welfare. Animal welfare concerns were raised over captive breeding of animals for eventual translocation (e.g., Law and Tatner 1998; Carlstead et al. 1999). But aspects of animal welfare science have been applied only superficially to translocation studies (Garshelis and Siniff, 1983; Waas et al., 1999). Effectively, the implications of animal welfare science have neither been considered nor explored for reintroduction biology (but see Swaisgood, 2010).  Recent research has supported the contention that attending to individual, population, species and landscape-level factors should enhance the survival, settlement and reproduction of translocated populations. Specifically, there is compelling evidence that translocation success is influenced by individual health, behavioural habits and coping ability; population size, status and genetic integrity; species life-history traits, ecology and ethology; and habitat quality, composition and location (Bright and Morris, 1994; Reynolds et al., 2008; Pinter-Wollman et al., 2009; Dickens et al., 2010). Indeed, the few translocations that have come closest to taking these factors into account, notably the reintroductions of the golden-lion tamarin, Leontopithecus rosalia, in Brazil and the black-footed ferret, Mustela nigripes, in North America, have been most successful (tamarin: Bush et al., 1996; Kierulff and DeOliveira, 1996; Kleiman and Mallinson, 1998; ferret: Wisely et al., 2003; Wisely et al., 2008).      12  1.2 Stress physiology Walter Cannon (American physiologist, 1871-1945) was perhaps the first researcher to use stress as a physiological term. Through his work, he developed the famous ?fight or flight? theory, which suggests that animals react to stressors with a non-specific activation of the sympathetic nervous system (SNS), priming the animal for engagement or escape. Cannon observed that the SNS, in concert with the adrenal gland, manages the response by secretion of catecholamines ? leading to increased heart rate, blood pressure, glucose availability, perspiration, and decreased stomach motility (Nelson, 2000). Cannon?s research focused on the adaptive quality of the stress-response in emergencies, looking at how the body successfully copes with energetically demanding situations (Sapolsky, 1993).  As with Cannon, Hans Selye (Austrian-Canadian, 1907-1982) also focused his research on the nonspecificity of the stress-response (Sapolsky, 1993). Selye, in what he termed the ?general adaptation syndrome?, observed that a variety of stressors cause glucocorticoid (GC) secretion. With this observation, he discovered that in addition to the sympathetico-adrenal medullary system (SAM) that Cannon studied, a second major system was at play. This would be referred to as the hypothalmic-pituitary-adrenal cortex (HPA) system.   As discovered by Cannon and Selye respectively, catecholamines (epinephrine and norepinephrine) and glucocorticoids (cortisol and corticosterone) are the two key classes of hormone released during the stress-response. The sympathetic nervous system, as part of the autonomic nervous system, works in conjunction with its parasympathetic counterpart (PNS) to mediate reactions to stressors. While the PNS maintains, for example, normal breathing, heart rate and digestion, the SNS responds to a perceived stressor by activating the release of catelcholamines. Epinephrine is released from the adrenal medulla directly into the blood stream  13 and norepinephrine is relayed to other organs through spinal nerve endings (Sapolsky, 1993). Consequently, the classic signs, as noted by Cannon, occur. Simultaneously, the HPA cascade is activated as the brain perceives the stressor. Specifically, the hypothalamus releases corticotropin-releasing hormone (CRH), which in turn stimulates the anterior pituitary to release adrenocorticotropic hormone (ACTH). ACTH then promotes the release of GCs from the adrenal cortex. The effects of GCs are wide-ranging on the body, and in the short term have the net result of preparing the body for sustained exertion by reallocating energy.  The work of Cannon and Selye still provides the foundation for stress-response research. Naturally, however, the story has become more complex. In addition to the general responses they emphasized, we now understand that different stressors elicit some specificity of response (Tsigos and Chrousos, 2002).  The specific manner in which animals ? human and non-human alike ? perceive, respond, and cope with stressors is now known to be an important determinant of human and animal welfare. For example, researchers have learned that the effect of a stressor varies with context. Particularly, stressors elicit different ?hormonal patterning? between species and among individuals (Sapolsky, 1998), and the affective state of an individual alters its response to a given stressor (Sapolsky, 1998). Conceptual models of the stress-response have thus incorporated these advancements. The more comprehensive models include individual-specific psychological input, behavioural and biological responses to cope with a perceived stressor, as well as pathological outcomes when the adaptive coping mechanisms are not sufficient (e.g., see McEwen and Stellar, 1993; Cohen and Rodriguez, 1995; Moberg, 2000).  The stress-response model by Moberg (1985, 2000) is arguably the most influential with regard to animal welfare (Rushen et al., 2008). The model includes three overlapping stages: the  14 recognition of a threat, the stress response, and the consequences of the response. According to Moberg?s model, the individual has a repertoire of biological responses ? behavioural, autonomic, neuroendocrine, and immunological ? to cope with the stressor. If the cost of mounting a stress-response is too great, the individual will enter a pre-pathological state, with the potential to develop a full pathology. The model predicts that the nature of the stress-response itself and the toll it takes on the individual varies according to the animal?s energy reserves and its existing energy use (McLaren et al., 2007).  Moberg and others have argued that it is the stress-induced changes in biological function that signal an animal?s welfare state. Still, others claim that to fully appreciate the welfare implications of a stressor, we need to understand not only the biological response, but also how an animal perceives a stressor (Moberg, 2000; Rushen et al., 2008).  It is recognized that an animal?s cognitive appraisal of a potential stressor influences the behavioural and physiological character of its stress-response, such that an animal?s perception affects its ability to cope with a stressor (Sapolsky, 1998; Koolhaas et al., 1999; Fraser, 2008). Moreover, the ability to predict and control a stressor leads to more successful coping (Koolhaas et al., 1999). In the classic experiments by Jay Weiss (1972), rats that were given a warning signal were able to anticipate an electric shock and did not develop stomach ulceration to the degree that rats given no warning did. And rats that were given the added means of preventing the shock exhibited less stomach ulceration than a defenseless group of rats that received the same number of shocks. In essence, an animal?s own ability to control the situation thus influences its biological response. The neural recognition of a stressor activates an intricate cascade of behavioural, neurophysiological and peripheral physiological responses (Rushen et al., 2008). In this  15 activation, the stress-system involves all major and ancillary endocrine systems ? the cardiovascular, respiratory, digestive, immune and reproductive systems are each affected (Fraser, 2008). The general, adaptive stress-response causes an acceleration of cardiac and respiratory output, increase in blood glucose availability, a decrease in the production of sex and growth hormones as well as the sensitivity to them and a reduction of immune system activity (Sapolsky, 1993; Tsigos and Chrousos, 2002). The adaptive stress-response is designed well for coping with short-term, physical stress; evolutionarily this is thought to be what most animals would encounter most often (Sapolsky, 1993). However, as animals are managed more intensively and increasingly affected by human activity they are faced with novel stressors that they are not necessarily equipped to cope with.  According to Moberg (2000) the biological cost of most stressors is minimal because they are short-lived, but prolonged insults or repeated exposure to acute stressors may become a biological burden (Moberg, 2000). Moreover, the specificity of the adaptive response to a given stressor may be lost as the severity of the stressor increases (Tsigos and Chrousos, 2002). It is these chronic stressors that are connected to the emergence of stress-related diseases and conditions (Sapolsky, 1993; McEwen and Stellar, 1993; Moberg, 2000). For example, extended activation of the stress system can cause chronic hypertension leading to heart damage and weakened vessel walls, and through their inhibitory effects, GCs can cause gastric ulcers (Selye?s marker), suppressed reproduction, stunted growth, as well as immunosuppression (Sapolsky, 1993).  The stress-response is a complex of biological events, only conceptually isolated into discrete stages. Thus, the reliance on just a part of the overall response to evaluate animal welfare is problematic. That being said, our ability to assess welfare by measures of stress has  16 improved through our progressive understanding of these stages and thus their interaction. The animal welfare implications of a burdened stress system are largely uncontested. Rather, there is some debate about which indicators of stress are best used to assess welfare and the means by which they should be measured and interpreted (Ramos and Morm?de, 1998).  It is argued that to effectively assess stress, understanding how an animal perceives events is critical (Rushen et al., 2008). Avoidance, vigilance and communication behaviours have been typically used as indicators of stressor-perception in controlled settings (Rushen, 2000). Umvelt studies in behavioural ecology (sensory ecology) investigate how animals acquire and respond to information in their environments. Although applied less frequently to studies of stress, these studies offer a useful framework for how an animal appraises a potential stressor, particularly for wild animals (Swaisgood, 2010). Umvelt studies, for example, have revealed how species of sea turtle, bird and some insects respond negatively to artificial lights (Witherington, 1997).   However, some stressors cause physiological changes before major behavioural responses occur. Consequently, tracking hormone levels can provide insight into the early signs of compromised welfare (Wingfield et al., 1997). The physiological components of the stress-response provide a basis of welfare assessment by signaling that an animal may be undergoing a negative experience, particularly with regard to acute stressors, and such measures may predict long-term negative consequences for an animal (Lester et al., 1996; Rushen et al., 2008). As mentioned, the autonomic system and the HPA axis represent the foundations of the physiological stress-responses. However, much of the welfare implications of stress physiology research focuses on the HPA axis. In part, this is because the HPA axis, producing GCs, has  17 broader, longer-lived effects on the body and the GCs and their metabolites are easier to isolate than the catecholamine metabolites of the autonomic system (Moberg, 2000; Palme et al., 2005).  Sapolsky et al. (2000) distinguish the stress-related actions of GCs into two classes: modulating actions, which affect the animal?s response to an immediate stressor, and preparative actions, which affect an animal?s response to a future stressor.  The modulating actions of GCs are further differentiated into permissive, suppressive and stimulating modes (Sapolsky et al., 2000). Generally, permissive and stimulating GC actions enhance the effects of the immediate stress-response, while suppressive actions dampen the stress-activated reactions to return the body to homeostasis. The immediate stress-response is typically marked by release of catecholamines, hypothalamic release of CRH, pituitary secretion of ACTH, decreased hypothalamic release of gonadotropin releasing hormone (GnRH), pituitary release of prolactin (PRL), and pancreatic secretion of glucagon. The actions of both basal and elevated GC levels, through all modes of GC action, are integral to managing the stress-response and for homeostatic maintenance (Matteri et al., 2000; Sapolsky et al., 2000). Additionally, the self-regulation of GC action, via a negative feedback loop, minimizes tissue exposure to the catabolic, anti-reproductive, and immunosuppressive effects of these hormones (Tsigos and Chrousos, 2002). The physiology of GC action appears designed to respond to and recover from natural stressor-induced defense reactions such as those caused by a predation attempt; however, when the response and recovery phases are hindered GC pathology occurs (Sapolsky et al., 2000).  Bearing in mind that the adaptive physiology of animals is not well designed for anthropogenic stressors, concerned researchers have looked at links between GC concentrations  18 and animal welfare assessment. GC levels have thus been used as a measure of stress effects in a wide range of animals. GC levels have traditionally been measured through blood plasma analysis, but blood sampling brings with it a variety of limitations. Because of the pulsatile nature of GC secretion, plasma GC levels are representative of only the time at which the sample was taken (Harper and Austad, 2000; Keay et al., 2006), and the samples do not well represent average circulating GC (Lane, 2006). Moreover, the sampling procedure itself can produce a rise in GC levels (Cook et al., 2000). Of particular importance to questions of stress and animal welfare, non-invasive techniques to analyze GCs and their metabolites have been developed for an increasing number of domestic and wild species (Touma et al., 2004). Salivary collection for GC analysis is one non-invasive method that has become more common over the past 30 years. This method has the advantage of accurately representing circulating GC levels and the method of collection is less invasive than blood sampling. However, this method is more feasible when handling is of minimal concern (Lane, 2006). Consequently, welfare-concerned studies have refined species-specific standards for fecal glucocorticoid (FGC) measurement. The overall method of FGC measurement is both practically and empirically valuable. It is of practical value because the collection method is entirely non-invasive and the timing of sample retrieval is more flexible than other methods. It has empirical value because FGC levels represent an aggregate of GCs and their metabolites over a period of time (Keay et al., 2006). Thus FGC levels, when validated properly and monitored over an extended duration, can be useful in understanding how persistent stressors can affect the welfare of animals.  19 FGC levels have been examined in a growing number of species, but the biological relevance of this technique has been validated in only a handful of them (Touma and Palme, 2005). Because of the high variability in GC metabolism and excretion within and across species, validation steps are crucial to reliably assess adrenocortical activity for a given species. Important methodological considerations for validation include ensuring the appropriate extraction method, fully identifying GC and their metabolites through high performance liquid chromatography, and screening immunoassays for the appropriate antibody to avoid confounding cross-reactivity.  In the light of our increasing understanding of the role GCs have in the stress-response and the methodological and empirical benefits of the technique, the measurement of FGC has wide application for wild, free-ranging animals, with welfare and conservation implications.    1.3 Animal personality For much of the 20th century, the study of animal personality has predominantly been the interest of psychologists. By the latter half of the century, the field had attracted the attention of physiologists; and currently animal personality represents one of the fastest growing areas of research in behavioural biology and ecology. More slowly, researchers are recognizing that knowledge of animal personality has importance to animal welfare. Thus far, this link has been made most noticeably through the application of animal personality to conservation outcomes. Over the past century, the scientific view of individual variation in animal personality has shifted. Until the 1950s, researchers spanning fields of psychology, physiology and ethology, were openly interested in individual differences in personality and emotionality in various  20 species. But, since the 1950s, many scientists have been hesitant to attribute personality traits or emotions to animals (Gosling and John, 1999; Fraser, 2009).  The shift has been ascribed to the general influence of Positivism on scientific inquiry (Rollin, 1990; Fraser, 2009). Positivism, which distinguished the epistemology of science from that of non-scientific fields such as theology and metaphysics, maintained that science is concerned with material phenomena that can be observed and measured. Thus, psychological aspects of animals, such as cognition, emotions and even personality fell outside the scope of science.  Celebrated scientists, including American psychologist John B. Watson (1878-1958) and Dutch ethologist Nikolaas Tinbergen (1907-1988), reflected the influence of Positivism in their methodology. Despite steady criticism, their opinions effectively shaped the direction of their fields and of animal behavioural studies as a whole ? by all but excluding subjective experience (Fraser, 2009).  By the late 20th century, the influence of Positivism appeared to have weakened, and with this came a revival of subjective experience in scientific inquiry. Over the past two decades, there has been a renewed interest in animal personality among comparative psychologists, physiologists and animal behaviourists and a new interest among conservation biologists and animal welfare scientists (Gosling and John, 1999; Gosling, 2008; Baker, 2013).  Animal personality research is rooted in the tradition of comparative psychology. Traditionally, animals have been used as models to elucidate human dimensions of personality. As an early, if crude forerunner, Ivan Pavlov ? in his studies with dogs (1906; 1928)  ? identified four types of personality reminiscent of the ancient Greek system of personality based on the four humours: Choleric (excitable), Sanguine (lively), Phlegmatic (quiet), and Melancholic  21 (inhibited). Pavlov?s attention to individual variation informed his understanding of variation in nervous system function (Gosling, 2008; Carere and Locurto, 2011). Other early, but more enlightened contributors to the field include Robert Yerkes (American comparative psychologist, 1876-1956), Donald Hebb (Canadian Psychologist, 1904-1985) and Calvin Hall (American psychologist, 1909-1985). Yerkes, despite the influence of Positivism and the prevalence of Behaviourism amongst American psychologists, unapologetically studied subjective aspects of the animal mind. Through his study of primate intelligence, he came to recognize and admire individuality and personality differences in apes as well as other animal species. In a 1939 paper to the American Society of Naturalists about the life history and personality of the chimpanzee, Yerkes stated that there was ?no question about the reality of chimpanzee mind, individuality, personality.? And as early as 1917, in an essay entitled, ?Individuality, Temperament, and Genius in Animals? he, and coauthor Ada Yerkes, wrote,  Everyone who knows anything about primates, high or low, realizes that in them individuality is more conspicuous for the human observer than in most other organisms. But our results do not justify the conclusion that temperamental differences are more obvious or more important in monkeys, anthropoid apes, or man, than in crows, pigs, or rats. We have come to suspect that the popular opinion concerning the matter is due chiefly to similarity of structure and behavior?in a word, to felt kinship. It is simply because we are more like monkeys and apes that we more readily notice and more highly value their individual characteristics.   22 While more in the tradition of the Behaviourists, Hebb and Hall recognized that a rigid exclusion of subjective concepts had scientific disadvantages. In his seminal study on emotion in man and animal, Hebb (1946) came to see that reliance on behavioural descriptors was not sufficient to examine the nuances of emotions and temperament. Once he dispensed with that approach, he observed stable differences in fear, nervousness and shyness among individual chimpanzees.  Hall (1941) surveyed the literature on temperament in animals ? primarily from experimental psychology journals ? to gain insight into individual differences in a variety of personality traits, such as fearfulness, aggressiveness, and activity. To understand the bases for temperamental differences, Hall studied emotions in rats. His ?open-field? experiments, which placed rats in well-lit arenas, launched a sub-field of ?emotionality? research in laboratory rodents that lasted into the 1970s (Archer, 1973). Use of the open-field test reemerged with the new generation of animal personality studies.  Despite the prominence of these scientists and their work, animal personality research largely fell by the wayside until the 1970s (Gosling and John, 1999; Gosling, 2008). Then the personality assessment of primates by ethologically minded psychologists began to reassert the study of animal personality (e.g., Chamove et al., 1972; Buirski et al., 1973; Stevenson-Hinde et al., 1978; Stevenson-Hinde et al., 1980). These studies used trait-testing methodology developed in the field of human personality in the 1960s to identify broad personality dimensions such as anxiety, fearfulness, affection, confidence and sociability. Researchers primarily used codings of the animal?s behaviour and subjective ratings of traits to assess personality structure and individual differences within these traits (Gosling, 2001).  In the last decade of the 20th century a collection of animal personality studies emerged primarily from comparative psychology, primatology and applied behaviour. Studies ranged  23 from assessment of broad personality dimensions to correlational studies of personality with, for example, social tendency, rearing, and gender (e.g., Bolig et al., 1992; French, 1993; Mather and Anderson, 1993; Gold and Maple, 1994; Forkman et al., 1995; Budaev, 1997; Coren, 1998; Gosling, 1998; Capitanio, 1999). While studies involving animal personality had certainly increased during this time they were generally opportunistic and employed a widely varying set of methodologies that made them difficult to compare and generalize. Despite a growing interest and acceptance of individual variation during this time, progress in the field was held back by the absence of a unifying theoretical framework for understanding and explaining it. In a cross-species review Gosling and John (1999) used the human Five Factor Model (FFM) as a preliminary framework to unify the disparate literature. FFM is an empirical, data-driven research model, which uses factor analyses to correlate measures of the traits. In its purest form, the five spectral factors are Openness, Conscientiousness, Extraversion, Agreeableness, and Neuroticism. Gosling and John evaluated the major dimensions of personality in twelve species (primate, dog, hyena, cat, donkey, pig, rat, fish, cephalopod). Agreeableness, Neuroticism and Extraversion were more strongly conserved across species, followed by Openness. A Conscientiousness dimension was associated only with chimpanzees.  With evidence of evolutionary continuity of personality among humans and other animals Gosling (2001), in a seminal paper, further explored the diverse animal personality literature. In it he outlined general principles to guide the nascent field of animal personality. Of particular influence to a multidisciplinary program, Gosling emphasized the importance of an evolutionary and ecological approach.  As a psychologist Gosling was interested in how animal personality studies could inform human psychology. Nonetheless, his paper helped shape a field unto itself. As the field began to  24 take shape, a definition of personality emerged from its roots in human psychology to include the concerns of animal cognition, physiology and behaviour. For example, personality went from being defined as those characteristics of individuals that describe and account for consistent patterns of feeling, thinking and behaving (Pervin and John, 1997) to referring to stable, long-term behavioural, emotional, and physiological differences in suites of traits among individuals of the same species (Carere and Locurto, 2011).  As the study of animal personality was emerging from the field of psychology, ?coping styles? were beginning to be thought of in terms of personality (Gosling pers. comm., 2011). Like personality, coping style is defined by emotional, cognitive and behavioural components, but coping is specific to the management of stressful situations. A more strictly biological definition of coping focuses on the behavioural and physiological ways of dealing with stressors. Moreover, a particular coping style is further defined by a behavioural and physiological profile consistent throughout an individual?s life and under different situations (Koolhaas et al., 2001).  As with personality psychology the study of coping styles in animals can be traced back to Pavlov?s studies on dogs (1906; 1928; 1941). Dichotomous coping typologies, such as internalization and externalization as well as extraversion and introversion are comparable to Pavlov?s sanguine and phlegmatic types (Traue and Deighton, 2000; Lindsay, 2005). Contemporary thinking about coping styles, however, stems more directly from the work of physiologist, James P. Henry (Koolhaas et al., 2001).  Based on social stress research, Henry proposed two types of response ? active and passive. The active response was behaviourally characterized by high territoriality and aggression; in contrast animals with a passive response showed low levels of movement and aggression (Koolhaas et al., 2001). The active-passive response to environmental challenges  25 described by Henry (Henry and Stephens, 1977) led to a significant body of literature on coping styles in animals, particularly in laboratory rodents.  Initial work concentrated on behavioural responses in mice bred for different levels of aggression (Benus et al., 1990). Level of aggression (high and low) was predictive of how an individual mouse would behave in a suite of challenges. Generally, high-aggressive mice were characterized by quicker, cursory exploratory behaviour, lower neophobia, and higher risk-taking behaviours in contrast to low-aggressive individuals (Dingemanse and de Goede, 2004).  Investigators also looked at physiological reactions of high- and low-aggressive rodents responding to environmental challenges, with particular reference to the autonomic nervous system (sympathetic and parasympathetic) and HPA axis. Overall, more aggressive rodents displayed greater sympathetic reactivity (higher levels of catecholamines) and lower HPA activity (lower GC levels) than less aggressive rodents (e.g., Fokkema et al., 1988; Sgoifo, 1996; Sgoifo et al., 1998). The bipolar characterization of behavioural and physiological stress responses has come to be described as the proactive-reactive axis of coping style (Koolhaas et al., 1999).  Although coping style research promoted a dichotomous typology rather than a continuum of individual variation of the stress-response, its concern for individual difference, even at this level, influenced research in a number of biological fields ?  particularly in ethology, ecology and stress biology. In these fields, variability of response was typically viewed as problematic for biologists, and rarely were individual differences viewed as biologically meaningful (Sapolsky, 1994; Carere and Locurto, 2011). Two major themes, however, that came from coping style research changed the prevailing perception. One was the  26 adaptive value of different coping styles and the other was pathology development based on coping style (Koolhaas et al., 2001).  For stress physiologists, individual variability in the stress-response was an opportunity to better understand disease vulnerability and prevention (Sapolsky, 1993; Carere et al., 2010); for behavioural ecologists it was a chance to understand the fitness and evolutionary consequences of different life strategies (Archard and Braithwaite, 2010; Carere and Locurto, 2011). These questions have become the key challenges in personality research (Pawlak et al., 2008).  Following in the methodological tradition of coping style research, stress biologists have relied primarily on captive animals to elucidate the relation between coping style and disease vulnerability (Archard and Braithwaite, 2010). In rodent models, researchers have shown that reactive individuals were more susceptible to social stress and had increased incidence of tumor metastases (Wu et al., 2000; Azpiroz et al., 2007). Biologists have seen a similar connection in captive primates, linking susceptibility to social stress with disease proliferation. Capitanio et al. (1999), for example, saw that after inoculation with simian immunodeficiency virus (SIM) rhesus macaques that were less sociable had a higher incidence of SIM RNA than more sociable monkeys.  Additionally, researchers examined the neuroendocrine profiles of the animals, finding that higher sociability covaried with lower HPA activity, as reflected in lower GC levels. This general pattern, correlating behaviourally proactive individuals with lower GC levels and behaviourally reactive ones with higher levels, has been seen in host of other animals (e.g., Hessing et al., 1993; Wingfield and Romero, 2000; Orchinik et al., 2002; Cockrem, 2007; Stowe et al., 2010).  Biologists believe that personality traits interact with physiological stress-responses, such  27 as activity of the HPA axis, leading to changes in immune function and pathology development (Sapolsky et al., 2000; Segerstrom and Miller, 2004). Furthermore, neuroendocrine substrates, particularly GCs, have been identified as key mediators of personality (Carere et al., 2010).  At the time Gosling (2001) wrote his review of animal personality studies, much of the relevant research was devoted to understanding proximate mechanisms of variation, primarily in laboratory rodents and primates (Sih et al., 2004). In the last decade, however, individuals from over 100 species ? from insect to mammal ? have been shown to vary significantly from their conspecifics in both behaviour and physiology. This covariation suggests that there are mechanistic and functional links between the two (Carere et al., 2010). Seeking to explain why consistent differences in behaviour and physiology persist in a population, researchers hypothesize that there are fitness consequences to personality.   In the field of behavioural ecology, the focus of animal personality has been on consistent individual differences and suites of correlated behaviors, or what has been referred to as ?behavioural syndromes?. Any underlying mechanisms, such as genetic or physiological factors have been components of the questions of interest, rather than inherent in the definition of personality itself.  While broadly interested in the causes and consequences of behavioural diversity, behavioral ecologists had otherwise tended to neglect individual variation in behaviour. Instead, researchers have generally attended to changes in the average behavioural response to environmental variation (Sih et al., 2004). This was in large part because evolutionary theory predicted that when individual differences are heritable and linked to fitness, natural selection should select for optimal behavioural suites and minimize behavioural variation (Wilson, 1998; Bell et al., 2010). But recent studies suggest that personality differences are maintained by  28 frequency-dependent natural selection (Carere et al., 2010). Thus, the idea of studying differences in animal personality represented a significant shift in the traditional thinking of the field (Sih et al., 2004).  In the last decade, work by behavioural ecologists has found that fitness consequences of different personalities within a population change with changing environmental conditions. Researchers have shown that mortality, breeding success, dispersal, partner preference, energy use, immune system function, as well as disease and parasite vulnerability are correlated with personality type (R?ale and Festa-Bianchet, 2003; Dingemanse et al., 2004; Both et al., 2005; Korte et al., 2005).    It is hypothesized that different personality types are preserved within a population because different types are favored under naturally shifting environmental pressures, and that no single personality type is optimal under all conditions (Dingemanse et al., 2004; Wolf et al., 2007). The persistence of a wide complement of personality types, then, positively affects population health.   The direct and indirect effects of human activity, however, may disrupt the selective forces that maintain personality composition. In particular, certain personality types are almost certainly more susceptible to anthropogenic effects than others, disproportionately reducing individual fitness. As a result, the overall composition of personality types within a population is expected to shift (Smith and Blumstein, 2008).   For example, personality types may be differentially affected by the consequences of environmental pollution, habitat fragmentation, and global warming. Research has shown that different strategies of dispersal, migration, foraging, predation and breeding cause individuals to be more or less vulnerable in a compromised environment.   These findings should extend to certain conservation management practices as well,  29 particularly translocation programs (McDougall et al., 2006; Baker, 2013). Despite the intention of conservation biologists this practice may cause unintended disruption to personality composition and affect animal welfare.   Translocations have been described as ?exercises in forced dispersal for conservation purposes? (Stamps and Swaisgood, 2007). In addition to being artificially relocated from one habitat to another, the animals are also opportunistically assembled. In particular, those that are trapped for relocation are likely to include fewer trap-shy individuals, and hence the translocated group may be skewed toward a particular personality type. Moreover, it is unlikely that a certain personality type is favored in all events of a typical translocation. Translocated animals are faced with a range of unfamiliar conditions, often in a concentrated period. These events are intense and varied enough to undermine an animal?s repertoire of behavioural and physiological coping mechanisms.  Watters and Meehan (2007) argue that to offset the environmental variability and unfamiliarity of the release site, attempts should be made to compose release groups of the full complement of personality types that are typically represented in the species. But, the success of a translocation is likely contingent not only on the range of personality types released, but also on how well we mitigate translocation procedures for different personalities. Thus, knowledge of personality composition, and how certain personality types respond to various stressors, may help us better protect threatened populations as well as anticipate and reduce negative impacts on individuals.       30  1.4 Study species Stephens? kangaroo rat (SKR), Dipodomys stephensi, is one of 21 recognized Dipodomys species. It is a medium-sized heteromyid rodent weighing between 45 and 75g, with an average body length of 7 cm. Like other kangaroo rat species, it is nocturnal, uses ricochetal locomotion and dwells in burrows when not active above ground. While individuals predominantly live independently in burrows, these burrow systems are typically within close proximity, exposing SKR to frequent visual and auditory contact with one another. Documented home ranges vary from 0.06 to 0.10 ha and home range sizes appear to be a function of population density (Thomas, 1975). Population density can vary from 5 to 58 individuals per ha based on location and season (Bleich, 1977; McClenaghan and Taylor, 1993).  SKR is endemic to open grasslands and to regions of sparse coastal sage scrub in southern California (Bleich, 1977), where it forages primarily on seeds. Its physiology is adapted to this semi-arid environment, with reliance on metabolic water production for hydration. SKR is considered a keystone species as its soil disturbance and seed-caching activities significantly affect plant community structure (Brock and Kelt, 2004a).   Populations of SKR currently exist in three distinct areas of southern California: they are found primarily in western Riverside County, with fewer populations persisting in western and central San Diego County. The historic home range of SKR, which covers 2,870 sq. km., lies inland between the major cities of Los Angeles and San Diego. Populations are typically found at elevations between 55 and 1250 m above sea level (USFWS, 1997). Over the last 40 years, suburbanization and cultivation have become the major, most immediate threats to SKR survival. It was estimated that the historical habitat had been reduced by 60 percent by 1984 (Price and Endo, 1989). California Department of Fish and Wildlife (CDFW) listed SKR as rare in 1971  31 due to substantial loss of habitat throughout its range and later listed it as threatened in 1984. By 1988 the U.S. Fish and Wildlife Service (USFWS) classified SKR as endangered, which remains its current status. In 1997 USFWS drafted a species recovery plan, the final version of which is still in preparation.   The United States? Endangered Species Act (ESA) prohibits any taking of species federally listed as threatened or endangered; however, with federal review, incidental take permits may be issued contingent upon approval of mitigation plans. Larger landholders, such as counties, typically prepare Habitat Conservation Plans (HCPs) to meet federal and state requirements, and in so doing avoid the need to secure individual permits. In 1996 the Riverside County Habitat Conservation Agency (RCHCA) drafted an HCP, which included SKR conservation, mitigation and monitoring measures. As an extension of these measures, funds were set aside for managed translocation efforts.    In collaboration with RCHCA, the San Diego Zoo?s Institute for Conservation Research (ICR) spearheaded an intensive translocation project for SKR (Shier, 2009, 2010, 2011); before this time (2008) only a few translocations of SKR were attempted and no viable populations were established. Since that time, translocations have been conducted within the boundaries of the Southwestern Riverside County Multispecies Reserve (Reserve) in Temecula, California, USA (N33 34 54.78, W117 01 36.87). My research was conducted through this project.   To assess the differential stress of effects of the translocation process on Stephens? kangaroo rats, I used a combination of behavioural, physiological and psychological measures to profile individuals. Specifically I:  32 1)  Determined if adrenocortical activity can be assessed reliably through fecal samples of SKR (Chapter 2). 2) Measured the effect of the translocation process on fecal glucocortcoid levels (Chapters 4 and 5).  3) Evaluated behavioural budgets of SKR during captivity (Chapter 5). 4) Identified personality types within SKR using qualitative and quantitative measures (Chapters 3 and 5). 5)  Determined if the above measures or any correlations thereof were predictive of survival after release (Chapters 4 and 5).      33 2. Validation of a fecal glucocorticoid assay for the endangered Stephens? kangaroo rat using behavioural and pharmacological tests     2.1 Introduction  Much stress physiology research, particularly in non-human animals, has focused on refining species-specific standards for measuring the activation of the hypothalmic-pituitary-adrenal cortex (HPA) axis via glucocorticoid (GC) production (Moberg, 2000; Palme et al., 2005). Traditionally, adrenal cortex activity has been measured through analysis of GC concentrations in plasma, but blood sampling has drawbacks (Harper and Austad, 2000; Keay et al., 2006; Lane, 2006), including the increase in GCs that the sampling procedure itself may cause (Cook et al., 2000).  An alternative approach is to analyze GCs and their metabolites in feces (Touma et al., 2004). In this case, the collection method is non-invasive and hence less likely to affect GC production, and the timing of sample retrieval is more flexible than other methods (which is useful when direct access to animals is limited or not possible). In addition, fecal GC (FGC) levels represent an aggregate of GCs and their metabolites over a period of time (Keay et al., 2006).  Therefore, FGC levels can be useful in understanding how persistent stressors affect animals and their welfare. This use extends to assessing stress in animals that are part of conservation programs as these typically involve some combination of stressors, such as captivity, marking, monitoring, transport and handling, in addition to environmental and social disturbance. Mitigating stress responses to these procedures may ultimately be important for the success of conservation efforts (Teixeira et al., 2007; Dickens et al., 2010). As a result, conservation researchers have begun to investigate the effect on FGC levels of common  34 conservation practices (e.g., trapping, Harper and Austad, 2001; transport, Millspaugh et al., 2007; radio transmitters, Wells et al., 2003; Pereira et al., 2009; captivity, Rothschild et al., 2008). Many wildlife conservation projects in recent years involve the translocation of free-ranging animals. However, translocation as a practice has been historically associated with a high animal mortality rate (Griffith et al., 1989). As typical translocations involve at least all of the stressors aforementioned, a few studies have examined FGC levels in response to translocation events (Franceschini, 2008; Viljoen et al., 2008; Dickens et al., 2009a, 2009b; Pinter-Wollman, 2009). While FGCs and their metabolites have been examined in a growing number of species of conservation concern, the biological relevance of the technique has been validated in only a small percentage of the species (Touma and Palme, 2005). Because of the high variability in GC metabolism and excretion within and across species, validation steps are crucial to reliably assess adrenocortical activity for a given species (Hunt et al., 2004).  Stephens? kangaroo rat (SKR), Dipodomys stephensi, is an endangered species of heteromyid rodent (USFWS 1988), endemic to southern California. To varying degrees of success, researchers have attempted to restore SKR to areas within their historic range via translocation (O?Farrell, 1994; Spencer, 2003; Shier, 2009; Shier, 2010; Shier, 2011; Shier and Swaisgood, 2011). With proper validation, FGC analysis may offer an effective measure of the physiological status for translocated SKR and a means of assessing the stressfulness of translocation procedures so that they can be refined. Thus, the purpose of this study was to validate an assay to measure FGCs as an indicator of adrenocortical activity in SKR in order to characterize the excretion of FGCs as it pertains to translocation-related activities, such as temporary confinement, as well as to sex and time of day.  35  We attempted both behavioral and pharmacological challenges meant to modulate adrenocortical activity. Pharmacological tests have proved useful for validation. Researchers, however, recognize that rigorous pharmacological validation may not always be feasible with endangered species, largely due to handling constraints (Touma and Palme, 2005; Mart?nez-Mota et al., 2008). In the event of such constraints, it is recommended to take advantage of biologically relevant stressors (e.g., restraint, agonistic interactions, predator stimuli) to validate assay methods (Touma and Palme, 2005). As a behavioural challenge, we presented SKR with a naturally occurring stressor, predator urine (fox). Exposure to predator urine has been shown to stimulate adrenocortical activity in other animals (Masini et al., 2005; Moncl?s et al., 2006; Harris et al., 2012). The pharmacological tests included the administration of adrenocorticotropic hormone (ACTH) and dexamethasone (Dex). ACTH is secreted by the anterior pituitary gland as part of the stress response by the HPA axis. ACTH, in turn, promotes a sharp increase in GC secretion from the adrenal cortex, which has been detected in fecal assays for a variety of species after the injection of ACTH (Wasser et al., 1997; Palme et al., 1998). Dex is a synthetic GC; injection of this compound activates the negative feedback loop of the HPA axis, thus reducing the level of endogenous GCs secreted into the bloodstream. After Dex injection a marked dip in the concentration of GCs and/or their metabolites has been observed in the feces in a range of vertebrate species (Touma et al., 2005; Sheriff et al., 2010; but see Dehnhard et al., 2003).     2.2 Materials and methods 2.2.1 Animals and housing We conducted FGC validation experiments with Stephens? kangaroo rats during the summers of 2010, 2011 and 2012. Fieldwork (trapping, testing, housing and fecal collection) was  36 done at the Southwestern Riverside County Multispecies Reserve (Reserve) in Temecula, California USA and laboratory work (fecal steroid extraction, HPLC, radioimmunoassays) was performed at the San Diego Zoo Institute for Conservation Research (Escondido, CA). We exposed SKR to predator urine in August and September 2010. Dex experiments were conducted in summer 2011 and 2012. ACTH trials were attempted in summer 2011. SKR trapped for the predator urine tests came from two source sites, one located near the Reserve (N33 33 17.43, W117 02 13.49) and another within it (N33 35.207, W117 02.095). Dex- and ACTH-tested SKR were trapped at a separate site within the Reserve (N33 34 54.78, W117 01 36.87). These animals were the offspring of founders translocated to the area in 2008 and 2009. All SKR were captured with modified Sherman live traps (30.5 ? 7.6 ? 9 cm, with shortened doors to prevent tail injury). Traps, baited with a mixture of white millet seed and raw oats, were positioned just outside active burrow entrances. For Dex and ACTH tests, SKR were temporarily held in captivity for 24 hours. SKR exposed to predator urine were captured as part of a larger translocation program and held in captivity for 6-14 days. Animals were housed in a facility located within the Reserve. The temperature of the facility was maintained at 23 ? 2? C and kept at a seasonable photoperiod of 13 h light:11 h dark. All extraneous human activity was minimized and all activity conducted at night was done under red light. SKR were housed individually in plastic tanks (large Petco Keepers for small animals, 34.8 x 21 x 23.4 cm). Each tank was provided with sand (Quickrete Premium Play Sand no.1113), a polyvinyl chloride (PVC) ?burrow? covered at one end (9.6 x 14.4 cm), animal bedding (Carefresh Basic pet bedding), and cardboard tubing for chewing. Animals were fed 59 ml of an oats and millet  37 mixture. Seed was provided each evening and lettuce was given every other day for SKR held longer than 24 h.  2.2.2 Fecal collection  We regarded all pellets collected from an individual for a given time window or during a collection event as a single sample. For collection SKR were carefully removed from their tanks while in their PVC burrows and were placed in a temporary tank. This enabled recovery of all pellets in the layer of sand of the ?home? tank, with minimal disturbance to the animals.  Fecal samples were stored in 2-ml screw-cap micro tubes and placed immediately upon collection in a conventional freezer (set at -23? C) at the holding facility. Samples were later transferred to -80? C freezers prior to assaying.  2.2.3 Antibody characterization by high performance liquid chromatography (HPLC) Preliminary analyses by radioimmunoassay (RIA) indicated that SKR fecal extracts contain analytes that are reactive with cortisol and corticosterone antibodies (unpubl. data). This is consistent with previous findings in Merriam?s kangaroo rat (Dipodomys merriami), in which significant concentrations of both GCs were measured in plasma (Preston, 2001). In an effort to characterize and validate an antibody for the measurement of GC concentrations in SKR feces, pooled fecal extracts were separated by reversed phase HPLC followed by RIA of each fraction using three different antibodies: monoclonal mouse anti-cortisol (071210107; MP Biomedicals, Costa Mesa, CA), polyclonal rabbit anti-corticosterone (07120113; MP Biomedicals), and polyclonal rabbit anti-corticosterone (CMJ006; supplied by Coralie Munro; University of California, Davis). One fecal pellet from each of 5 male and 5 female SKR was lyophilized.  38 Freeze-dried pellets were then transferred to 12 x 75 mm borosilicate tubes and extracted by shaking in 1 ml of 80% methanol for 30 minutes. The 5 extracts from each sex were combined into a clean tube, dried down under vacuum, reconstituted in water, and then concentrated on C18 cartridges (WAT051910 Sep-Pak, Waters, Milford, MA) that had been conditioned with 10 ml each of 100% methanol and then water. The samples were then washed with 10 ml of water, eluted from the cartridges with 5 ml of 100% methanol, dried down under vacuum, and reconstituted in 0.2 ml of 100% methanol. Following solid-phase extraction, 20 ?l of each fecal extract pool was injected into a Beckman System Gold 3-piece unit (Programmable Solvent Modules 125/406 and Diode Array Detector Module 186, Beckman Coulter) and separated on a Nova Pak C18 column (WAT086344, 3.9 x 150 mm, Waters) as previously described (Harris et al., 2012).  Samples were separated along an acetonitrile gradient beginning with 2:98 (acetonitrile:water, v/v) and increasing to 75:25 over 75 minutes at a flow rate of 1 ml per minute. Fractions were collected at 1-minute intervals, dried down, and reconstituted in phosphate buffered saline (PBS, pH 7.0).  Each of the 75 fractions was prepared in duplicate for measurement by RIA (as described in 2.2.4) with each of the antibodies. The elution times of immunoreactive fractions were compared to the elution times of several commercially available (Steraloids, Newport, RI) steroids (corticosterone, cortisol, testosterone, estrone sulfate, and progesterone) and steroid metabolites (5?-androstane-3?-ol-11-17-dione and 5?-pregnane-3?,11?,21-triol-20-one) similarly prepared and detected by absorbance at 205 nm.    39 2.2.4 Fecal glucocorticoid assays Fecal pellets were lyophilized and weighed to the nearest 0.001 g and transferred to 12 x 75 mm borosilicate tubes for extraction. To extract fecal GCs lyophilized pellets were shaken in 1 ml of 80% methanol for 30 minutes. Following extraction, 0.9 ml of the extract was transferred to a clean tube, dried down under vacuum, and reconstituted in 0.5 ml of assay buffer (PBS). Cortisol standards were serially diluted in PBS from 0.16 ? 20 ng/ml and prepared in duplicate for each assay, along with quality control standards of approximately 0.8 and 8.0 ng/ml. In order to reduce nonspecific binding, 400 ?l of 0.4% bovine serum albumin (Fraction V; Fisher Scientific, Pittsburgh, PA) in PBS was added to each tube. Finally, tritiated cortisol (1,2,6,7-3H; Perkin Elmer, Waltham, MA; 10,000 cpm per 0.1 ml) and cortisol antibody (1:3000 in 0.1 ml of PBS) were added, bringing the final assay volume to 0.70 ml. Each assay was incubated overnight at 4oC, and bound-free separation was performed by adding 250 ml of 5% charcoal/0.5% Dextran in PBS, incubating at 4oC for 30 minutes, and centrifugating at 2000 x g for 15 minutes. The supernatant was combined with 3.5 ml of scintillation cocktail and counted on a Beckman LSC6500 scintillation counter. The concentrations of cortisol standards were plotted against the log-logit transformation of the % 3H-CORT bound. Fecal glucocorticoid concentrations were expressed as ng/g fecal dry weight.  2.2.5 Predator urine experiment  The predator urine exposure test was incorporated into an ongoing translocation event in August 2010. As a result, sample sizes for each treatment were contingent on SKR availability. Experimental animals were exposed to fox urine purchased from Predatorpee.com. In addition to a non-exposed control group, we also used a conspecific simulation test to offer SKR a stressor  40 less potentiating than predator scent (Williams, 1999). For this treatment we presented the animals with a mirror (Svendsen and Armitage, 1973). In total, 131 SKR (60 females, 71 males) were tested.  Specifically, 20 animals (11 females, 9 males) were exposed to urine, 39 (19 females, 20 males) were exposed to the mirror, and 72 (30 females, 42 males) were unexposed (these animals were left undisturbed in their tanks except during fecal collection). We conducted trials 48-56 hours after animals were brought into captivity; trials ran between 22:00 and 03:00, falling within SKR active period.  We collected fecal samples at time of capture (wild) and 2-3 nights after exposure treatment. Animals were trapped over a one-month period. The time of night that post-exposure samples were collected was matched to the time of collection of the wild samples. To ensure that fecal pellets were excreted during the desired time window, we moved animals to a new cage with fresh sand, and then returned them to their ?home? cage. To reduce the need for direct handling, animals were transferred in their PVC burrows.  Two identical arenas (96.5 x 67.3 x 35.5 cm) were constructed of black foam board to be used exclusively for one test (urine or mirror). The arenas were lined with black painter?s plastic to hold a bed of sand, 7.2 cm thick. At one end we cut out a rectangular opening (7.6 x 9 cm), with a re-closable flap for SKR entry. To prevent escape a fitted lid made from fiberglass window screen (Phifer 18 x 24 mesh) was placed atop the arena. For the predator scent exposure, we sprayed 4 complete spritzes of fox urine onto a rock. This was refreshed every 4th trial. The scented rock was placed in the far corner opposite the entry. For the conspecific simulation trials a mirror was positioned on the arena wall opposite the entry. To release an animal into the arena, we transferred it from its home-cage PVC burrow into another tube with a removable back. This tube was inserted into the entry and with a soft  41 plunger we gently guided the animal into the arena. Between trials the top 2-cm layer of the sand was removed and replaced with new sand. Arenas were lit with a 60-W red light placed 1.9 m above. Each trial was a total of 10 minutes.    2.2.6 Rationale for sequence of pharmacological tests  We attempted an ACTH challenge in July 2011, concurrent with the first of two Dex tests. Results for the ACTH test, however, were inconclusive. From the 8 SKR (males) trapped, fecal samples were excreted on average during only 5 (? = 2) of the possible 13 time windows. We injected ACTH (Cortrosyn, Fisher Scientific, North Carolina, USA) at a dose of 0.1?g/gm, which met the 2 IU/kg dose recommended by Wasser and colleagues (S. Wasser personal communication, 2012).  During the next season (August 2012), we prioritized Dex treatment over ACTH. The rationale for this was based on the limited access to SKR (both in trap days and trap success) and the inability to provide a habituation period as the USFWS permit limited holding to 24 hours, due to the risk of territory loss if animals were held longer. In the light of this and SKR being a prey species, we predicted that the HPA axis might reach its ceiling with trapping and handling alone. Thus, fecal validation of adrenal activity may be more readily observed via pharmacological suppression of the axis given the constraints.  2.2.7 Dex injection and sample collection   Dexamethasone (dexamethasone sodium phosphate 4mg/g) was injected at doses of 1 ?g/g (2011) and 8 ?g/g (2012). The total volume injected was 2 ?l/g. The two dosage treatments were conducted in different years because the limited availability of SKR in 2011 afforded a robust  42 sample size for only one dose. In 2011, injections were performed in the holding facility, which caused a lag time between capture and injection of approximately 30 ? 45 minutes. To minimize this in 2012 we injected in the field, at the time of capture, and then transported the animals to the holding facility. Total transport time (both years) was 5 minutes. In 2011 4 females and 2 males were treated with Dex, 4 females and 4 males were injected with normal saline (0.9% w/v NaCl). In 2012 4 females and 9 males were treated with Dex, 8 females and 11 males were treated with saline. Both saline and Dex were administered through intraperitoneal (IP) injection.  For injections, the animal was carefully scruffed while its legs were stabilized. Injection time took no more than 1 minute. Animals were immediately placed into PVC burrows and put into holding tanks. For all treatment groups we collected fecal samples, when available, at time of capture and every 2 hours post-treatment for 24 hours.  2.3 Statistical analyses 2.3.1 Predator urine experiment We analyzed the effects of the Exposure Treatment (urine, mirror, control) as well as the potential effects of body weight, age, and sex of the animal on measured cortisol levels using model selection methods (Burnham and Anderson 2003). Because the days between the collection of the first and second fecal samples varied slightly among individuals we also included this factor (Days) in the analysis. The full model included independent variables for all of these factors.  Models were compared using Akaike Information Criteria (AIC), and models having ?AIC <2 of the best model were considered to have similar support. In addition, analysis of variance (ANOVA) was used to test for an interaction between Exposure Treatment and Days and for a non-linear relationship with Days. Cortisol levels were log-transformed prior to  43 analysis, and two outliers having cortisol levels>100 ng/g were removed. The response variable was the difference in cortisol levels between the first and second fecal sample. All analyses were conducted with R (version 2.15, R Core Team 2013).  2.3.2 Dex experiment   Variation in fecal cortisol was analyzed using a mixed-effect linear model. Fecal cortisol (FC) measurements were log transformed [ln(FC + 1)] to improve normality before analysis. The full model included fixed effect terms for year of study (2011, 2012), experimental treatment (saline and Dex), sex, and time (hours since injection). To identify statistically significant effects, terms with the least explanatory power were removed from the model in a stepwise fashion to arrive at a final, best model. Alternative models were compared using AIC. To detect whether responses to experimental injections were dependent on circadian rhythms, the residuals from the best model were plotted against time (hour since injection). Analyses were conducted using R (version 2.15, R Core Team 2013) and its lme4 package.  2.4 Results 2.4.1 Antibody characterization and fecal assay  The immunoreactivity of fecal extract fractions following HPLC is summarized in Figs. 1-3. The cortisol antibody was highly specific for HPLC fractions corresponding to the retention time of cortisol (Fig 1). Approximately 67% and 35% of the total immunoreactivity in the female and male extract pools, respectively, corresponded to the retention time of cortisol. No other fractions exceeded 9% of the total cortisol antibody immunoreactivity. In contrast, both corticosterone antibodies lacked specificity among the fecal extract HPLC fractions. The  44 percentage of the total immunoreactivity for any particular fraction of the extract did not exceed 10% for either corticosterone antibody.   2.4.2 Response to predator urine Exposure to fox urine and days between samples influenced measured fecal cortisol compared to the unexposed and mirror-exposed groups. On average, urine-exposed SKR had a greater change in FC during captivity (untransformed mean?SE, urine-exposed 15.16?1.75; mirror-exposed 2.60?2.41; unexposed 2.28?1.1.6, Fig. 4). The ANOVA indicated that the influence of exposure treatment and days between fecal collection were similar in magnitude (Exposure, F=5.99; Days, F=4.93; both P<0.03). There were no statistical interactions between these factors and no significant nonlinear relationship between Days and FC. The covariates of body weight, age, and sex of the animal did not improve the explanatory power of the statistical models.  2.4.3 Response to Dex injection Dex administration did not result in a marked suppression in FC concentrations (FCC) in SKR over a 24-hour period with a dose of either 1 or 8?g/g. In 2011 and 2012, the animals injected with Dex exhibited a similar time course pattern to SKR from the control group (Figs. 5 & 6). In both groups, FCCs increased to peak around 16 hours after injection then declined (Figs. 5 & 6).  FCCs were on average 55% higher in 2012 (untransformed mean?SE, 5.174?0.202) than 2011 (3.342?0.123). Neither the experimental treatments nor sex provided useful explanatory information. There was no evidence that FC was affected by time of day.  Model residuals did not vary with respect to the hour of initial injection, thus there did not appear to be  45 an effect of circadian rhythm on FC after accounting for terms in the model (Pearson?s r = -0.026; P>0.51). Differences among individual animals were substantial, as estimated by random effects, and accounted for 15% of the variation in the data.    46 Fig. 1 Immunochromatogram of HPLC fractions assayed with monoclonal mouse anti-cortisol (071210107, MP Biomedicals) for Stephens? kangaroo rat. a = cortisol; b = 5a-pregnane-3B,11B,21-triol-20-one; b = corticosterone; c = 5B-androstane-3a-ol-11-17-dione; d = testosterone; e = esterone sulfate; f = progesterone          a b c d e f 0 20 40 60 80 100 120 0 100 200 300 400 500 600 700 800 1 6 11 16 21 26 31 36 41 46 51 56 61 66 71 Male cortisol (pg/fraction) Female cortisol (pg/fraction) Fraction Female Male  47 Fig. 2 Immunochromatogram of HPLC fractions assayed with polyclonal rabbit anti-corticosterone (CMJ006; University of California, Davis) for Stephens? kangaroo rat.    a = cortisol; b = 5a-pregnane-3B,11B,21-triol-20-one; b = corticosterone; c = 5B-androstane-3a-ol-11-17-dione; d = testosterone; e = esterone sulfate; f = progesterone          c a b d e f 0 10 20 30 40 50 60 70 80 90 1 6 11 16 21 26 31 36 41 46 51 56 61 66 71 76 Corticosterone (pg/fraction) Fraction Female Male  48  Fig. 3 Immunochromatogram of HPLC fractions assayed with polyclonal rabbit anti-corticosterone (071201123, MP Biomedicals) for Stephens? kangaroo rat. a = cortisol; b = 5a-pregnane-3B,11B,21-triol-20-one; b = corticosterone; c = 5B-androstane-3a-ol-11-17-dione; d = testosterone; e = esterone sulfate; f = progesterone         c d a b e f 0 5 10 15 20 25 30 1 6 11 16 21 26 31 36 41 46 51 56 61 66 71 76 Corticomsterone (pg/fraction) Fraction Female	 ? Male	 ? 49 Fig. 4 Fecal cortisol concentrations (ng/g) of Stephens? kangaroo rat for each exposure treatment (predator urine, mirror and unexposed) at time of capture and after treatment. Data are shown as medians with the 25th and 75th percentile. Whiskers show the 5th and 95th percentile. Outliers are presented as dots.            Exposure treatmentUrineMirrorUnexposedFecal cortisol (ng/g)6040200Post-exposurePre-exposure 50 Fig. 5 Fecal cortisol concentrations ng/g [ln (FC + 1)] over 24 h collection period of Stephens? kangaroo rats after dexamethasone (1?g/g) and saline injection in 2011. Data are shown as medians with the 25th and 75th percentile. Whiskers show the 5th and 95th percentile. Outliers are presented as dots.             51   Fig. 6 Fecal cortisol concentrations ng/g ([ln (FC + 1)]) over 24 h collection period of Stephens? kangaroo rats after dexamethasone (8?g/g) and saline injection in 2012. Data are shown as medians with the 25th and 75th percentile. Whiskers show the 5th and 95th percentile. Outliers are presented as dots.            52 2.5 Discussion The results show that the developed radioimmunoassay is specific for cortisol in Stephen?s kangaroo rat fecal extract.  Furthermore, the increase in fecal cortisol concentrations in Stephens? kangaroo rat after exposure to predator urine suggests that FCCs are indicative of adrenocortical activity. While we were unable to show the predicted response using the standard pharmacological methods, such as ACTH or Dex, the response to the fox urine is consistent with other studies that have shown a marked increase in GCs in response to perceived predator stimuli (Cockrem and Silverin, 2002; Masini et al., 2005; Moncl?s et al., 2006; Harris et al., 2012).   The results indicate that FC is a reliable indicator of adrenal activity, being detected by the cortisol antibody. However, we do not presume that cortisol is the most prevalent circulating form of glucocorticoid. Restrictions in the protocol prevented us from taking blood samples from the kangaroo rats, but previous research on Merriam?s kangaroo rat suggests that both cortisol and corticosterone are present in circulation (Preston, 2001). The lack of specificity of the two corticosterone antibodies relative to the cortisol antibody is likely due to the complexity of the extract and the origin of the antibodies. Fecal extract contains a mixture of parent compounds, the products of endogenous GC metabolism and microbial degradation, and artifacts of the extraction procedure. The polyclonal corticosterone antibodies reacted similarly with numerous HPLC fractions, limiting the utility of either antibody for our purposes.  On the other hand, the monoclonal antibody showed a reasonable degree of specificity for HPLC fractions corresponding to the retention time of cortisol. Although these results alone do not confirm cortisol as the immunoreactive analyte, when taken together with the increasing concentration following exposure to predator urine, they indicate that we have a valid measure of adrenal cortex activity.   53 Dex injection has suppressed the HPA axis in a number of species, including other rodents (Mostl et al., 1999; Touma et al., 2005; Sheriff et al., 2010); however, this suppression has not been observed in certain other animals, including pigs and chickens (Mostl et al., 1999; Dehnhard et al., 2003). In these studies the researchers have speculated that a long gut transit time, an inadequate sampling period or an ineffective dose were possible reasons for the unexpected results. For example, Mostl et al. (1999) suggest that the long gut transit time in pigs (as indicated by the interval between infusion of radiolabeled steroids and their fecal excretion) may have prolonged the excretion of FGCs such that the fecal cortisol signal was so dampened that a Dex-induced suppression was not detected. However, in one study with chickens FGC profiles ? which revealed no response to Dex over 35 hours ? corresponded with plasma GC profiles (Dehnhard et al., 2003, but see Etches, 1976; Vanmontfort et al., 1997). As plasma is strongly reflective of pharmacological challenge, the lack of response in the chicken plasma indicates that Dex may not induce suppression in all animals. In addition to the potential dampening of the signal, the aggregation of FC may be delayed as compared to plasma levels. For example, while corresponding patterns between the fecal and plasma responses to Dex injection have been observed in horses, the suppression was not detected in feces until 24 -36 hours after injection (Mostl et al. 1999). In SKR we were limited to a 24-hour fecal collection period. Thus, it is possible that our sampling period was too short to detect the expected cortisol suppression, particularly in a desert-adapted species such as SKR where there may be a longer transit time to allow for optimal moisture extraction from the feces. If so, this would differ from laboratory rodent species that have shown Dex-induced suppression in the feces 8 to 10 hours after injection, which corresponds with documented gut transit time in those species (Harper and Austad, 2000; Touma et al., 2003; Touma et al., 2004).  54 In an attempt to overcome the effect of dosage dependency, we administered a low and high dose of Dex (1?g/g and 8?g/g). In mice, for example, Touma et al. (2004) observed a more pronounced suppression ? in magnitude and duration ? at 3 ?g/g versus 1 ?g/g, and Harris et al. (2012) observed a suppression of plasma GCs in California mice at higher Dex concentrations of 5 and 10 ?g/g versus 0.5 ?g/g. Although we did not see a Dex-induced suppression in SKR feces at either 1or 8 ?g/g, the animals injected with the higher dose exhibited polyuria. Polyuria is a common outcome of excess cortisol; for example, it is often observed in human and non-human animals treated with cortisol for conditions such as hypoadrenocorticism (Addison?s disease). In some systems, cortisol manages norepinephrine secretion, which in turn influences the secretion of antidiuretic hormone (ADH) (Vanjonack et al., 1975). Excess cortisol can impair the regulation of ADH secretion from the posterior pituitary (Rothuizen et al., 1995). In SKR, it is possible that Dex did not affect the negative feedback loop of the HPA axis, but instead behaved as a source of excess cortisol and activated the diuretic pathway; the high dose of Dex could have promoted the secretion of norepinephrine and the suppression of ADH. Additionally, Dex has been shown to decrease aldosterone in a semi-arid lizard species, Tiliqua rugosa (Bradshaw and Grenot, 1976). A decrease in this mineralocorticoid would likely lead to a decrease in renal Na+ retention, thus to increased urination. The actions of adrenal corticosteroids, such as the ones described above, are mediated by tissue-specific glucocorticoid and mineralocorticoid receptors. Sapolsky et al. (2000) argue that synthetic GCs, such as Dex, may not be appropriate substitutes for natural GCs as their receptor-binding action may differ. Also, there is evidence to suggest that Dex has limited access to mineralocorticoid and glucocorticoid receptors in the brain (Wilckens, 1995; Cole et al., 2000). It is conceivable that exogenous Dex, while binding to receptors downstream from the brain and  55 suppressing the secretion of adrenal aldosterone (causing polyuria), could not affect upstream inhibition of the HPA axis (thus no FC suppression). If this is true, then Dex may be molecularly irrelevant for causing HPA suppression in SKR.  All things considered, pharmacological validation of adrenocortical activity may be too logistically challenging and impractical for an endangered species such as SKR, given the limited access to individuals ? both in terms of numbers necessary for robust sample sizes and the amount of time animals need to be removed from the wild. For species in this situation the use of a natural, biologically relevant, stressor may be better suited for validation. Further research might succeed in identifying why the pharmacological challenges failed in this species. For conservation purposes, however, and especially for endangered animals, the positive response to the natural challenge provides justification for using non-invasive fecal cortisol levels as an indicator of an HPA response in this species. It can thus be used as an indicative measure for reducing stress in translocation activities.          56 3. Composition of personality differences in a translocated population of Stephens? kangaroo rat    3.1 Introduction   The study of animal personality represents one of the fastest growing areas of research in behavioural biology and ecology. To date personality dimensions have been documented in many species across a range of taxa (Stamps and Groothuis, 2010). Beyond a theoretical understanding of the individual differences within species, the awareness of individual variation has applied and ethical implications.   Increasingly, biologists think that a range of personality types is preserved within a wild population because different types are favored under naturally shifting environmental pressures (Dingemanse and R?ale, 2005; R?ale et al., 2007). Variation in, for example, animals? aggression levels, motivation to explore and approach novel objects, as well as the underlying physiological mediators can influence life strategies (Verbeek et al., 1994; Verbeek et al., 1996; Dingemanse et al., 2002; Bell, 2004), causing individuals to be more or less vulnerable in challenging conditions. The persistence of a wide complement of personality types is thus seen as important for maintaining the adaptive resilience of a population (Dingemanse et al., 2004; Wolf et al., 2007).   In line with this thinking, a few biologists have argued that animal personality is an important concept in conservation biology and should be considered along with other phenotypic characteristics in wildlife translocations (McDougall et al., 2006; Watters and Meehan, 2007). A typical translocation involves some combination of capture, captivity, marking, monitoring, transport and handling, and release in a novel physical and social environment (Letty, 2007; Teixeira et al., 2007). Because these events are intense and varied it is unlikely that any single  57 personality type would be favored in all events. As animal personality shapes needs, preferences, and motivations, it should have consequences for how individuals respond to the stages of the translocation process. Consequently, individuals with certain personality types may handle the stressors of translocation better than others (McDougall et al., 2006; Dickens et al., 2010; Swaisgood, 2010). For example, Bremner-Harrison et al. (2004) observed that among Swift foxes (Vulpes velox) bolder individuals, characterized by their lower levels of neophobia, had higher death rates than more cautious individuals after release. Cote et al. (2010) have shown that the personality of a mosquito fish affects its dispersal decisions and these decisions are further influenced by the personality structure of the population in which it finds itself. Thus, knowledge of personality composition within a translocated group, and how certain personality types respond to translocation events, may help us improve translocation strategies and fitness following release as we are better able to anticipate and reduce negative impacts on individuals.   The translocation literature includes few examples where individual personality has been taken into consideration. In light of this need, we examined personality dimensions and composition in a translocated group of endangered Stephens? kangaroo rats (SKR), Dipodomys stephensi. SKR is a heteromyid rodent endemic to southern California, which researchers have attempted to restore to areas within their historic range via translocation (O?Farrell, 1994; Spencer, 2003; Shier and Swaisgood, 2012). Since 2008 San Diego Zoo?s Institute for Conservation Research (ICR), in collaboration with the Riverside County Habitat Conservation Agency (RCHCA) in California, spearheaded an intensive translocation project for SKR; this research was conducted through that project.    To assess personality we used subjective rating and behavioural coding, which have been used in tests to measure personality traits and dimensions in a variety of species (e.g., ratings:  58 Gosling, 1998 (hyena); Capitanio, 1999 (rhesus macacques); Fox and Millam, 2010 (cockatiels); codings: Bremner-Harrison et al., 2004 (swift foxes); Sinn et al., 2008 (dumpling squid); Highfill et al., 2010 (bushbabies)). For these tests SKR were exposed to mirror-image-stimulation (MIS) and predator scent (PS) trials while in temporary holding, before release. MIS has been used as a test of social and agonistic behaviour in a variety species (e.g., Svendsen and Armitage, 1973; Blumstein et al., 2006; Boon, 2007; Dochtermann and Jenkins, 2007). Predator scent tests are commonly used to measure boldness in animals (e.g., Dochtermann and Jenkins, 2007; Pruitt et al., 2010; Vainikka et al., 2011).  An increasing number of studies have shown that there is often a correlation between neuroendocrine activity and personality types (Koolhaas et al., 1999; Gosling, 2001; Byrne and Suomi, 2002; Martins et al., 2007). Hypothalamic-adrenal-cortex (HPA) axis activity, notably via glucocorticoid (GC) levels, has been identified as a key mediator of personality (Carere et al., 2010). With this in mind, we measured GC concentrations of the kangaroo rats before and during the translocation.   3.2 Materials and methods 3.2.1 Capture and captivity  As part of a larger translocation effort to move SKR from a disturbed area to protected, unoccupied habitat, we trapped 46 SKR from September 8-14, 2009 using modified Sherman live traps (30.5 ? 7.6 ? 9 cm with shortened doors to prevent tail injury). Traps were set just before sunset and were positioned just outside active burrow entrances. Traps were baited with a mixture of white millet seed and raw oats. Animals were trapped from a trail alignment within  59 the confines of Lake Skinner Recreational Area, located within the Riverside County Multispecies Reserve in Temecula, California (N33 35.207, W117 02.095).   SKR were held in captivity in a facility in the Multispecies Reserve (N33 34 54.78, W117 01 36.87) for 10?28 days before translocation at a temperature of 23 ? 2? C and a seasonable photoperiod of 13 h light:11 h dark. SKR were housed individually in plastic tanks (large Petco Keepers for small animals, 34.8 x 21 x 23.4 cm). Each tank was provided with sand, 7.2 cm thick (Quickrete Premium Play Sand no. 1113), a cardboard ?burrow? covered at one end (9.6 cm in diameter, 14.4 cm in length), animal bedding (Carefresh Basic pet bedding), and cardboard tubing for chewing. Animals were fed 59 ml of an oats and millet mixture each day and a lettuce leaf was given every other day.  We recorded sex and age for each SKR: adults were distinguished from juveniles through a combination of body weight and pelage, as adults have tawny-brown pelage and no molt line, in contrast to no molt line of juveniles. Animals were weighed upon capture and just before release using a 300 g Pesola medio-line spring scale.  3.2.2 Arena set-up and test procedure All 46 individuals slated for translocation (23 females, 23 males) were used for personality assessment tests. Each of these SKR was subjected, one time, to a mirror image stimulation (MIS) test and a predator scent (PS) test between September 19th and October 6th, 2009. Trials ran between 22:00 and 03:00, falling within the active period of the night for SKR. Individuals were given a rest period of 2 nights between tests. Animals were tested in two identical arenas (97 x 67 x 36 cm) constructed from black foam board, one used exclusively for the MIS test and the other for the PS test. The bottoms of  60 the arenas were lined with black painter?s plastic to hold a bed of sand 7.2 cm thick. One of the shorter walls had a rectangular opening (8 x 9 cm), with a re-closable flap for SKR entry. A fitted lid made from fiberglass window screen (Phifer 18 x 24 mesh) was placed atop the arena to prevent animals from escaping. At the entrance of the arenas was a refuge (a natural rock to which wooden sticks were affixed) where animals could seek shelter but remain visible to the observer. Trials lasted 13 minutes, comprising a 5-minute acclimation period and an 8-minute active period. A wooden barrier positioned parallel to the shorter walls was used to prevent access to the stimulus until the start of the active period. The barrier was positioned 64 cm from the entry for use during the acclimation period of the trials. After the acclimation period, we removed the barrier by sliding it to one side, through a slit in the arena wall. To release an animal into the arena, it was transferred from its home-cage burrow into another tube with a removable back. This tube was inserted into the entry and with a soft plunger the animal was gently guided into the arena. Between trials the top 2 cm layer of the sand was removed and replaced with new sand. After the trial, a PVC burrow was placed in the arena for the animals to retreat. They were then carried back to the their cages while in the burrow. In the arena used for MIS trials, the entire wall opposite the entry was covered with a mirror. This was cleaned with vinegar between trials. For the predator scent trials, we sprayed (4 complete spritzes) of fox urine (Predatorpee.com) onto a rock; this was refreshed every 4th trial. The scented rock was placed in the corner opposite the refuge. A fan was positioned to move the airflow away from the entry to minimize the chance of an animal detecting the scent during the acclimation period. Arenas were lit with a 60-W red light placed 1.9 m above. Tests were recorded with a handheld camcorder under the night mode setting (Sony Handycam HDR-SR12HD).  61 3.2.3 Personality assessment 3.2.3.1 Subjective ratings To assess personality two observers using a list of 16 pre-determined terms independently evaluated the recorded trials for MIS and PS tests (Table 1). For each term, observers rated the animal using an ordinal rating scale of 1 (trait absent) to 7 (trait extremely present) using 0.5 increments. We generated the list of terms based on the methods used by Gosling (1998) for assessment of hyena personality dimensions. The starting point for our terms was a list of 25 terms used for rhesus macaques (Stevenson-Hinde, 1980). Although raters were unfamiliar with the individual subjects, they were well versed in SKR behaviour and helped select the list of terms. While other personality studies have used clearly defined behaviours to define terms (via ethograms) (Stevenson-Hinde, 1980; Gosling, 1998; Petelle and Blumstein, 2014), raters assigned the scores based on their qualitative assessment of the animal?s behaviour throughout the trial (Wemesfelder, 2001). We reasoned that this approach would allow the raters to register subtle behaviours reflective of personality that might be obscured by a strict ethogram.   3.2.3.2 Behavioural codings In addition to the subjective ratings, the following behaviours for both MIS and PS tests were scored from the active periods of the recorded trials: time standing still, walking and running, as well as number of grooming events and number of jumping and digging behaviours. A jump was defined as a complete bipedal vertical leap. A single bout of digging was defined as use of the forearms to displace sand and lasted for a minimum of 2 seconds. We deemed a bout over if there was a 2 second cessation of the activity. We differentiated digging from sandkicking (the only other digging-like behaviour observed). Sandkicking was characterized by a sudden displacement of sand with the hind feet. Specific to MIS trials, we scored the latency to approach  62 the mirror, the time at the mirror, with eyes oriented toward it and the number of approaches to the mirror, all with the criterion of the animal being within 5 cm of the mirror. For PS trials the time spent in physical contact with the urine-covered rock and latency to approach the rock were also quantified. Behaviours were scored independently by two people (not the subjective raters) using the event-recording program JWatcher (Blumstein and Daniel, 2007).   3.2.4 Fecal collection We measured fecal cortisol concentration (FCC) in a subset of 26 kangaroo rats (13 female; 13 male) using a validated assay (see Chapter 2). We attempted to collect fecal samples at time of capture (t = 0) to represent baseline adrenocortical activity, and during the next 48 hours at t = 4, 8, 12, 24 and 48 hours after capture to examine adrenocortical reactivity, but kangaroo rats defecated reliably at only t = 0, 24 and 48 hours.  Baseline samples (t = 0) were retrieved from the trap at the time of capture. SKR were in the traps for a maximum of 1.5 hours, which indicates the maximum age of the fecal pellets. We pooled all pellets collected from an individual for a given time window as a single sample. To collect samples during captivity, kangaroo rats were carefully removed from their ?home? tanks while they were in their PVC burrows, and placed in a temporary tank. This enabled the recovery of all pellets in the layer of sand from the ?home? tank with minimal disturbance to the animals.  Fecal samples were stored in 2 ml screw cap micro tubes and placed immediately upon collection in a conventional freezer (set at -23? C) at the holding facility. Samples were later transferred to -80? C freezers prior to assaying.    63  3.3 Statistical analyses  All analyses were done using IBM PASW (version 20.0 for Mac; SPSS, Chicago, Illinois). MIS and PS trials were analyzed separately. To determine inter-rater reliabilities we used a two-way mixed intra-class correlation coefficient (ICC). We conducted principal components analyses (PCA) to reveal underlying personality dimensions from the ratings in both MIS and PS tests. We ran parallel analyses (Monte Carlo simulations) to determine the number of components to extract from the PCA. To do this, we used code in SPSS syntax written for random permutations by O?Connor (2000). Components were then subjected to varimax rotation for the PCA. We used Spearman?s rank correlation coefficient to examine the correlations between the factor scores of individual animals derived from the extracted components from each personality test (MIS and PS) as well as between these factor scores and behavioural observations. Spearman correlations were also used to test whether fecal cortisol levels were correlated with factor scores from the components derived from PCA. Spearman?s rank was used because it is more suitable to non-normally distributed data and more robust to outliers than Pearson correlation.    64 3.4 Results 3.4.1 Inter-rater reliability of ratings  For 11 of the 16 terms, ICCs (used to assess inter-rater reliability) were statistically significant for both the MIS and PS tests (Table 1). Coefficients for the MIS test ranged from 0.54 (for persistent) to 0.87 (active), with a mean reliability of 0.69; for the PS test, they ranged from 0.43 (curious) to 0.84 (timid), with a mean reliability of 0.60. The remaining 5 terms (aggressive, deliberate, anxious, tense and cautious) had significant ICCs for one test, and only 1 of the 5 (deliberate) had an ICC below 0.100 (0.04). We chose to retain this term, as it was unlikely to affect the principal components structure (see Kone?n? et al., 2008). Thus we used all 16 terms for the PCA.   3.4.2 Principal Components Analysis 3.4.2.1 Loading structure PCA extracted 3 components for each of the two tests (MIS and PS), representing 79.2% and 79.3% of the total variance, respectively. Table 2 shows the structure and loadings of the 16 terms for each of the components. To interpret the components we, as in other studies, attributed the term to the component with the greatest (both positive and negative) loading (e.g., Weiss et al., 2006, Kone?n? et al., 2008). Aside from 3 terms (aggressive, territorial, tense), there was correspondence between the cluster of salient terms for MIS and PS components. Confident, active, curious, exploratory, (not) apprehensive, (not) cautious and (not) timid, loaded onto the first component for both tests (MIS 1 and PS 1). In addition to this cluster of terms, (not) tense was salient for MIS 1, aggressive and territorial were salient for PS 1. Based on these loadings, we interpreted the first component for each test as Assertiveness. For the second component from the MIS test (MIS 2) and the third component from the PS test (PS 3), deliberate, persistent  65 and (not) permissive were clustered together, aggressive and territorial also loaded with these terms for MIS 2. We interpreted these components as Persistence. For the third component from the MIS test (MIS 3) and the second component from the PS test (PS 2), anxious, fearful, tense for PS and (not) calm loaded together. Based on these terms we described these components as Excitability.    3.4.2.2 Individual factor sores There was a strong consistency between the two tests in how individuals were scored. The individual factor scores generated by the PCA for each component had highly significant correlations between similarly interpreted components: Assertiveness (rs [46] = 0.69, p < 0.0001); Excitability (rs [46] = 0.46, p = 0.001); Persistence (rs [46] = 0.41, p = 0.005).  Thus, individuals were similarly characterized in each test. To examine the range of individual variation of personality along these three components we created a scatterplot for each test (Figs. 7 & 8). Although we chose to present Assertiveness as a categorical gradient in these scatterplots, regardless of dimension orientation, no obvious pattern of clustering of personality types emerged.  3.4.2.3 Behavioural and fecal correlations We examined the relationship between quantified behaviours and the factor scores from each test for the three personality dimensions (Table 3). Of the 11 behaviours that were assessed 9 correlated with at least one dimension. Individuals high on Assertiveness (in both MIS and PS tests) spent more time walking and less time immobile. Individuals high in Assertiveness in MIS tests were slower to approach the mirror and spent less time interacting with it. Additionally, individuals high in PS Assertiveness were quicker to approach the urine-coated rock. Individuals  66 high on Persistence (in both MIS and PS tests) performed more jumps and individuals high in Persistence in the PS test also exhibited more digging. High Persistence in the MIS test was also associated with more time interacting with the mirror, both in approaches and time spent at it. Individuals high in Excitability in the MIS test performed fewer jumps and approached the mirror less. Individuals with high scores for Excitability in the PS test spent more time running in the arena.  Baseline FCC ranged from 3.1?30.5 ng/g (mean?SE, 8.8?1.1). Baseline FCC was negatively correlated with Assertiveness components for both MIS and PS tests (MIS: rs [23] = -.46, p = 0.026; PS: rs [23] = -.46, p = 0.027). FC levels at 24 (range, 6.1?40.8 ng/g mean?SE, 15.6?2.1) and 48 hours (range, 2.4?60.8 ng/g mean?SE, 14.3?2.5) of captivity were not related to personality measures.    67   Table 1 Intra-class correlation coefficients (ICC) of term ratings for Stephens? kangaroo rats in mirror-image (MIS) and predator scent (PS) tests   ICCs in boldface had significant p-values. Italicized terms indicate significant ICC for only one test.      MIS   PS    Term ICC       p ICC     p   0.74 <0.0001 0.62 0.001 Confident 0.74 <0.0001 0.62 0.001 Aggressive 0.77 <0.0001 0.33 0.095 Territorial 0.79 <0.0001 0.50 0.011 Apprehensive 0.56 0.003 0.63 0.001 Cautious 0.17 0.271 0.50 0.011 Timid 0.83 0.322 0.84 <0.0001 Active 0.87 <0.0001 0.85 <0.0001 Curious 0.64 <0.0001 0.43 0.031 Exploratory 0.71 <0.0001 0.56 0.004 Calm 0.65 <0.0001 0.65 <0.0001 Anxious 0.25 0.167 0.56 0.004 Fearful 0.61 0.001 0.45 0.025 Tense 0.57 0.003 0.34 0.081 Deliberate 0.04 0.45 0.46 0.022 Persistent 0.54 <0.0001 0.57 0.002 Permissive 0.71 <0.0001 0.72 <0.0001  68 Table 2 Principal Component Analysis term loadings for Stephens? kangaroo rats in mirror-image (MIS) and predator scent (PS) tests. Component designations (Assertiveness, Persistence, Excitability) are based on term loadings.       Component    Term MIS 1 Assertiveness 48.2 %  Variance PS 1 Assertiveness 51.5%  Variance MIS 2 Persistence 20.6%  Variance PS 3 Persistence 6.3% Variance MIS 3 Excitability 10.4% Variance PS 2 Excitability 21.5% Variance Confident   .73   .61   .57   .46   .08  ?.50 Aggressive   .31   .63   .80  . 53    .35  ?.19 Territorial   .07   .61   .67   .34 ?.10  ?.28 Apprehensive ?.87 ?.83 ?.20 ?.11 ?.02    .34 Cautious ?.89 ?.78   .05 ?.33   .10    .30 Timid ?.79 ?.78 ?.47 ?.38 ?.15    .23 Active   .80   .76   .42   .48   .17    .06 Curious   .79   .88   .08   .18 ?.22    .07 Exploratory   .92   .83 ?.07   .13   .02    .03 Calm   .04 ?.04 ?.17 ?.13 ?.90  ?.88 Anxious   .05 ?.17   .15 ?.18   .94    .87 Fearful ?.55 ?.25 ?.21 ?.03   .69    .89 Tense ?.84 ?.25 ?.28 ?.03   .14    .89 Deliberate   .26   .37   .79   .82 ?.16  ?.08 Persistent ?.06   .22   .80   .89    .20    .03 Permissive ?.37 ?.47 ?.81 ?.55 ?.28  ?.54 Loadings in boldface were salient to the term Italicized, boldfaced terms reflect correspondence between MIS and PS loadings for a given component    69  Fig. 7 Interaction of individual factor scores of Stephens? kangaroo rats for each personality dimension (Assertiveness, Excitability, Persistence), derived from the PCA for mirror-image (MIS) test. Assertiveness is represented as a categorical variable              70 Fig. 8 Interaction of individual factor scores of Stephens? kangaroo rats for each personality dimension (Assertiveness, Excitability, Persistence), derived from the PCA for predator scent (PS) test. Assertiveness is represented as a categorical variable              71   Table 3 Correlations between behaviours and personality components of Stephens?s kangaroo rats from mirror-image (MIS) and predator scent (PS) tests   Component  Behaviour Assertiveness Persistence Excitability     Standing still (MIS)  ?.73 ?.27  .29 Walking (MIS)    .74  .12 ?.08 Jumping (MIS)    .15  .52 ?.47 Latency to approach mirror (MIS)   .38 .20   .21 Time at mirror (MIS)  ?.57  .43 ?.05 No. of mirror approaches (MIS)   .17  .10 ?.46 Standing still (PS)  ?.59 ?.26 ?.15 Walking (PS)   .54  .08  .18 Jumping (PS)   .11  .43 ?.20 Running (PS)   .10  .18  .46 Digging (PS)   .15  .30  .22 Latency to urine-coated rock (PS)  ?.55  .29 ?.02 Behavioural correlations significant for at least one component are shown. Correlations in bold  indicate significance at p < 0.05 or lower.    72 3.5 Discussion  These data provide evidence that an opportunistically assembled group of Stephens? kangaroo rats possesses a range of personality types. Across conspecific (MIS) and predator (PS) contexts we found evidence in SKR for three personality dimensions: Assertiveness, Persistence and Excitability. There was substantial correspondence between how the animals were scored in the MIS and the PS tests. This supports the view that these personality features are consistent over time and across situations. The notable exceptions were aggressive, territorial and tense which appeared to manifest differently in the tests, likely because these traits reflected the animals? different responses to conspecifics and predators.   The Assertiveness dimension, with its correlation to high activity, short latency to approach the urine-scented rock and low baseline FC, corresponds closely with the so-called proactive-reactive coping styles in response to environmental challenges as described in mice and rats (Gentsch et al., 1982; Steimer et al., 1997; Koolhaas et al., 1999). Within this dichotomy, kangaroo rats with a proactive coping strategy would be characterized by active exploration (walking) and active defense (short latency to approach urine-scented rock), in addition to a lower baseline FC levels; in contrast to animals with a reactive strategy, displaying immobility and passive defense in association with higher FC levels. The Assertiveness dimension of SKR possibly differs in one respect from the coping style typology ascribed to murine rodents. In mice and rats, aggression level has been shown to be positively coupled with level of exploration and defense (Benus et al., 1997; Koolhaas et al., 1999). In SKR, level of exploration and defense appear to be negatively coupled with degree of conspecific engagement. Similar to SKR, low-resistant pigs (characterized by response to manual restraint) showed less aggression in group-feeding tests (Hessing et al., 1994).  73 The usual assumption is that the behaviours observed in response to the mirror image offer a good measure of agonism (Gallup, 1968; Svendsen and Armitage, 1973). Aside from clear aggressive displays, even measures of latency have proved useful as an indication of agonism in rodents, including another kangaroo rat species (Svendsen and Armitage, 1973; Dochtermann et al., 2007). In confirmatory tests, Hargett (2006) found that latency to approach the mirror was negatively correlated with aggression levels and dominance status as determined by pair-wise interactions of Merriam?s kangaroo rats. While mirror-image stimulation may elicit agonistic behaviour in other rodents, including other heteromyids, it is possible that MIS does not elicit this in SKR. An alternative interpretation is that unlike in Merriam?s kangaroo rat, mirror interactions or shorter latency to approach the mirror, do not indicate increased agonism in SKR. Work with dogs has found that aggressiveness can arise in very different ways including a link to both assertiveness (?dominance aggression?) and to timidity (?fear aggression?) (Borchelt, 1983). It is likely that aggressiveness in rodents also has multiple causes and manifestations and needs to be treated as a complex variable.  In either case, the recorded behaviours in MIS tests reflect a social response, indicative of consistent behavioural variation. SKR is considered a more docile kangaroo rat species; in addition to our personal observations, there are few reports of intraspecific agonism (pers. obs.; Brock and Kelt, 2004; but see Shier and Swaisgood, 2012). Moreover there is evidence to suggest that SKR possesses a higher level of sociality than other kangaroo rat species (O?Farrell, 1990; Jones, 1993; Randall, 1993). Thus, it is conceivable that rather than agonism, we are seeing variation in social flexibility (Jones, 1993). Beyond Assertiveness and its parallels with proactive-reactive coping styles, Persistence and Excitability were also important in the analysis and can be expected to influence how SKR  74 respond to environmental and psychosocial events. Excitability shares features of  ?emotional reactivity? long observed in rats (e.g., Hall, 1936; 1938; Billingslea, 1941; Steimer et al., 1997). In fact, Steimer et al. (1997), using behavioural and neuroendocrine correlates, found evidence of emotional reactivity as an independent dimension in addition to coping style to explain individual vulnerability to anxiety in rats. In unfamiliar environments, there is believed to be a conflict between a rodent?s natural impulse to explore versus its fear of novelty (Steimer et al., 1997). Persistence may modulate these tendencies. In both MIS and PS tests, Persistence was correlated with jumping and digging; the nature of the jumping and digging behaviours were demonstrative of active attempts at escape. Although, there is a degree of correspondence between the proactive-reactive typology and the personality characterizations in SKR, the proactive-reactive dichotomy likely obscures what may be underlying causes and motivations of behaviour.  The scores for individual animals from the PCA show fairly continuous variation along the three dimensions rather than a clustering of scores into certain personality types (Figs 7 & 8). This may suggest that a diversity of personality types is relevant in the wild and that a population with a wide diversity may have a better chance of adapting to environmental change (Dingemanse et al., 2004; Wolf et al., 2007). The variation in SKR personality provides further support that individuals have different strategies for coping with their environments (Dochtermann and Jenkins, 2007). This in turn supports the idea that individuals will likely respond differently to the events of a translocation based on personality. Indeed, we now have preliminary evidence to suggest that more cautious, more social Stephens? kangaroo rats survive translocations better (see Chapter 5). We would argue, then, that the success of a translocation is likely contingent not only on the range of personality types released, but also on how well we  75 mitigate translocation procedures for different personalities. Thus, it is important to understand the personality structure within a population, including its adaptive significance to the species.      76 4. The effect of radio transmitters on fecal cortisol concentrations and survival in translocated Stephens? kangaroo rats   4.1 Introduction  Proper post-release monitoring is considered a crucial component for the success of conservation translocation programs. Despite this acknowledgement, failure to conduct monitoring as well as a failure to report the results has long been a criticism of the practice (Nichols and Armstrong, 2012). In fact, this lack of monitoring and reporting has been cited as one of the key shortfalls of the practice contributing to its low success rate (Lyles and May, 1987; Swaisgood, 2010).   Radio telemetry is an established technology for monitoring wildlife at the individual level and is increasingly used to monitor translocated animals. It can provide information on home range size, habitat use and selection, dispersal and settlement (Wauters et al., 2007; Rosalino et al., 2011; Caryl et al., 2012), which is important for improving translocation programs. A potential conflict arises, however, between obtaining the information that monitoring offers and minimizing the impact of translocation stress.  Stress is considered an unavoidable element of translocations (Teixeira et al., 2007; Dickens et al., 2010) as activities such as capture, handling, and release into a new environment are necessary to the process. Consequently, any additional stressor, such as monitoring, needs to be assessed; and concerted efforts should be made to minimize stressors during translocation (Teixeira et al., 2007; Dickens et al., 2010). Without these efforts, animal welfare and conservation goals can be undermined if the monitoring method itself can disturb the animals to such a degree that behavior and physiology are altered. Moreover, if behaviour or physiology is affected, information such as settlement times, dispersal rates or habitat use become difficult to  77 interpret, calling into question any conclusions drawn from the data collected (Walker et al., 2011; Jewell, 2013).  Despite the prevalence of its use, surprisingly few studies using radio telemetry have looked for any adverse effects of transmitter attachment on the animal, with the majority of these occurring in birds (Godfrey and Bryant, 2003; Pereira et al., 2009). Likely prompted by the obvious risk to flight, researchers have refined transmitter type, weight and attachment method for a variety of bird species (e.g., Gessaman and Nagy, 1988; Hooge, 1991) and have looked at the effects of transmitters on behaviors and time budgets, as well as growth rates, reproduction and survival (Carroll, 1990; Hubbard et al., 1998; Ackerman et al., 2004; Chimpan et al., 2007). Additionally, avian researchers have looked at the physiological effects of transmitters on stress hormone levels, using measures of fecal glucocorticoids (FGCs) (Suedkamp Wells et al., 2003; Schulz et al., 2005; Pereira et al., 2009) which together with their metabolites are commonly used as an indicator of adrenocortical activity in response to persistent stressors (Touma and Palme, 2005). In contrast to birds, many fewer studies have been conducted to assess the risk of transmitter attachment on small rodents, and the majority of these assess transmitters attached by collars (see, e.g., Webster and Brooks, 1980; Betreaux et al., 1994; De Mendon?a, 1999; Harker et al., 1999; Moorhouse and Macdonald, 2005). As part of a large translocation effort, backpack radio transmitters were used to monitor dispersal and settlement of the endangered Stephens? kangaroo rat (SKR) after release. In addition to the acquisition of behavioral data, we attempted to determine if the radio transmitters would function as an additional stressor that would influence the welfare and survival of the kangaroo rats. In an effort to determine this, we used a  78 validated fecal cortisol assay (see Chapter 2) to measure fecal cortisol concentrations (FCC) in response to transmitter outfitting and looked at the effect on survival after release.  4.2 Materials and methods  One hundred fifty-two SKR were translocated as part of larger translocation project (see Chapter 5 and Shier, 2011, for description of the methods). They were trapped from two parcels of privately owned land adjacent to the Multispecies Reserve and an overflow parking lot at the Lake Skinner campground. Kangaroo rats were translocated into 5 release plots with 6 different habitat conditions replicated within each plot. Thirty animals were soft-released at each site, 5 into each habitat. Of these 152, a subset of 60 adult SKR (30 males, 30 females) were assigned to one of two treatment groups (transmitter: SKR fitted with transmitter during translocation; control: SKR not fitted with transmitter during translocation). We selected 6 kangaroo rats per plot (one animal per habitat condition) on each of the 5 plots for a total of 30 adult SKR for the transmitter treatment.   4.2.1 Capture and captivity   Animals were captured with Sherman live traps (30.5 ? 7.6 ? 9 cm) modified with shortened doors to prevent tail severance. The traps were set just before sunset and baited with a mixture of white millet and raw oats.    SKR were temporarily held in captivity (? 14 days) before translocation in a temperature-controlled facility located within the Reserve. Animals were housed individually in either plastic tanks (large Petco Keepers for small animals, 34.8 x 21 x 23.4 cm) or ten-gallon glass tanks (Petco, 48.72 x 30 x 30.24 cm). We provided each tank with a PVC ?burrow? (9.6 x 14.4 cm),  79 animal bedding (Carefresh Basic pet bedding), and cardboard tubing for chewing. Additionally, the bottoms of the tanks were covered with 7.2 cm of sand. We fed animals each evening 59 ml of an oats and millet mixture and a leaf of lettuce every other day.   We weighed animals upon capture and just before release using a digital bench scale. We tagged animals using ear tags (Monel ? 1005-1, National Band and Tag) with color-coded reflective tape for individual identification at night.    4.2.2 Transmitter outfitting  The 30 SKR assigned to the transmitter treatment group were fitted with VHF radio transmitters (Holohil BD-2C). The total transmitter package weighed 1.8 g, which met the requirement that it not exceed 5% of the animal?s body weight (American Society of Mammologists, 1987). While there is an established record of using radio collars for various kangaroo rat species (Schroder, 1979; Jones, 1989; Shier and Randall, 2004; Cooper and Randall, 2007), a figure-8 backpack harness was used because SKR is bipedal and there is the risk of the forelegs becoming caught in a collar (see Shier and Swaisgood, 2012 for details). We excluded any individuals from the transmitter treatment that exhibited immediate resistance or discomfort to the apparatus (N= 4). We individually sized harnesses based on the animal?s weight at capture to accommodate for any loss of weight during captivity. The outfitting procedure ? from the time the kangaroo rat was removed from the cage to its return ? took no longer than one minute. Each kangaroo rat was monitored for 24 hours to ensure that the harness was loose enough to allow for the full range of movement and prevent abrasion, yet tight enough that the animal was unable to remove it. We applied transmitters 30-35 hours before animals were released (soft-released) into acclimation cages at the release site.  80 4.2.3 Acclimation period and release   Once all animals were captured (? 14 days), they were transferred individually to acclimation cages at the release plots for one week. The acclimation cages consisted of an underground nest chamber leading to an aboveground retention cage via flexible tubing. The animals were entirely confined to these cages for the week, but could move freely to and from the nest chamber and retention cage. After release from the acclimation cages, supplemental feeding was provided for 1 month. SKR with transmitters were tracked every day for 1 month; after that period, animals were re-trapped for transmitter removal. At this time, the animals were carefully assessed for signs of abrasion. We evaluated release success of SKR by re-trapping all release plots for 3-5 consecutive nights at 1, 6 and 12 months post-release.   4.2.4 Fecal collection   We attempted to collect fecal pellets from all 152 animals in the larger study at the time of initial capture (?wild sample?) and during captivity, acclimation and one-month post-release. We regarded all pellets collected from an individual for a given time window or during a collection event as a single sample. Upon collection, we stored fecal samples in 2-ml screw-cap micro tubes and placed them immediately in a conventional freezer (set at -23? C) at the holding facility. Samples were later transferred to -80? C freezers prior to assaying. Fecal samples were assayed for cortisol at the San Diego Zoo?s Institute for Conservation Research (see Chapter 2 for a complete methodology of SKR fecal cortisol assay).   The time of night that captivity samples were collected was matched to the time of collection of the wild samples. We collected these twice during an animal?s time in holding, at its mid point and on its final night. To ensure that fecal pellets were excreted during the desired time  81 window, we moved animals to a new cage with fresh sand from which the samples were retrieved. After collection the animals were returned to their ?home? cage. To reduce the need for direct handling, animals were transferred while in their PVC burrows.  During the acclimation period (soft-release), we collected fecal pellets from the cages after the first and final full day of acclimation (Day 1 and Day 6). Fecal samples collected one-month post-release were taken either directly from the animal if it defecated at the time of handling or from the trap bag which was cleared of feces between animals).   4.3 Statistical analyses We used a matched-pairs design to assess the effects of the transmitters. We performed logistic regression using log-transformed data. Control and treatment (transmitter or no transmitter) animals were matched by plot, release treatment and sex. To examine the effect of transmitters on stress, we required that fecal samples were collected for each treatment and control animal in a pair at multiple time points. We successfully collected fecal samples for 9 treatment + control pairs (18 animals), thus the analysis of the effect of transmitters on FCC was restricted to a sample size of 9 pairs.   We used the same matched-pairs design to examine the effects of transmitters on post-release survival at 1 month (the sample size was too small to analyze survival at 6 and 12 months using the matched-pairs design). Due to sex matching constraints, in one instance a kangaroo rat with a transmitter could not be matched with an animal without a one from the same plot and habitat treatment. Thus, the sample size was reduced to 29 pairs (58 animals) for this analysis. All matched-pairs design analyses were done using STATA (version 10 for Windows; Stata, College Station, Texas).   82 To compare the percent weight change (weight at one-month post-release minus weight at capture) between animals with and without transmitters, we used a Mann-Whitney U test. We removed juveniles and pregnant females from the subset of SKR with no transmitters for the analysis.  Because our methods of transmitter outfitting have been refined since the first SKR translocation (2008), we were interested in the effects of radio transmitters on animals from this study compared to those from the earliest translocations (2008/2009). With no means to explore physiological differences, we retrospectively looked at the comparative percent weight change (weight at one-month post-release minus weight at capture) between animals with transmitters from 2008/2009 and from this study (Mann-Whitney U test). All animals had transmitters in these earlier translocations, thus we were unable to compare the percent weight change to animals without transmitters. Mann-Whitney U tests were done using IBM PASW (version 20.0 for Mac; SPSS, Chicago, Illinois).  4.4 Results  The results indicate that the transmitters were not predictive of FCC in translocated SKR at Day 1 of acclimation (t (16) = 1.85, p = 0.08), nor at one-month post-release (t (8) = 0.43, p = 0.68, Fig. 10). They were, however, predictive of FCC at Day 6 (t (16) = 2.30, p = 0.03, Fig. 10). Logistic regression also suggests that transmitters had no effect on survival at 1 month (z (58) = -? 0.84, p = 0.40, Fig. 9). There was no difference between the percent weight change at 1 month after release comparing animals with and without transmitters (Mann-Whitney U, tied Z = ? 4.7, P = 0.641).  83 Animals with transmitters from this study had a significantly greater increase in body weight (8%) at one-month post-release compared to animals with transmitters (1%) from earlier translocations (2008/2009) (Mann-Whitney U, tied Z = ? 3.4, P = 0.001).    84 Fig. 9 Percent survival of Stephens? kangaroo rats at 1 month after release for Transmitter (N = 9) and No transmitter treatment groups (N = 10); difference is non-significant        0 10 20 30 40 50 60 70 80 90 no transmitter transmitter Percent survival 1 mo.  post-release  85 Fig. 10 Fecal cortisol concentrations for SKR in Transmitter and No transmitter (control) treatment groups. Wild and Captive periods were prior to transmitter outfitting. (Control, mean?SE: Wild, 16.9?2.1; Captive, 21.4?2.08: Day 1, 21.6?3.2; Day 6, 18.1?3.3; 1 mo post-release, 21.2?2.7. Transmitter, mean?SE: Wild, 17.3?2.7; Captive, 24.2?3.1: Day 1.48.1?13.1; Day 6, 35.3?7.4; 1 mo post-release, 25.6?5.9.) Data are shown as medians with the 25th and 75th percentile. Whiskers show the 5th and 95th percentile. Outliers are presented as dots.              86 4.5 Discussion The radio transmitters appeared to have a short-term effect on SKR fecal cortisol concentrations as seen during acclimation. The physiological response to the transmitters may have taken longer than 24 hours to appear in the feces, explaining why an increased FCC was not observed at Day 1 but was present at Day 6. Nevertheless, this effect did not appear to persist after release nor did transmitters affect survival estimates one-month post-release. Importantly, animals re-trapped at 1 month for transmitter removal were in good condition, as assessed by pelage quality, weight and overall appearance and showed minimal to no signs of abrasion from contact with the transmitter and harness.  Claims from the avian literature suggest that transmitter devices behave as an acute stressor with animals exhibiting a temporary rise in fecal corticosteroids, similar to the physiological response of capture or handling (Pereira et al., 2009; Schulz et al., 2005; Suedkamp Wells et al., 2003). But, when properly fitted they should pose no long-term threat (Pereira et al., 2009). A more recent meta-analysis of birds outfitted with transmitters, however, concluded that transmitters by and large impede animals and likely bias results (Barron et al., 2010). In some of the studies examining transmitter attachment in rodents, researchers reported altered dominance roles (meadow voles, Betreaux et al., 1994) increased parasite load (yellow-necked mice, De Mendon?a, 1999), skewed recruitment sex-ratio (water voles, Moorhouse and MacDonald, 2005) and lower survival (meadow voles, Webster and Brooks, 1980). In some cases, the authors provided mitigating measures, such as adjusting the weight of the apparatus, adding tick repellent, and avoiding the use of transmitters during seasons of higher predation risk.  87 Had this study been conducted when the methodology for transmitter outfitting was less refined, we might have seen a longer-lived effect of the transmitters on FCC and perhaps survival estimates. Upon recapture, during earlier SKR translocations in 2008/2009 we saw chaffing under the arms, swelling and restriction from the harnesses in a number of kangaroo rats; in a few rats, there was active bleeding (pers. obs.). The transmitters were removed when these or any signs of distress were observed. Moreover, animals with transmitters from 2008/2009 had a smaller percent increase in body weight compared to animals from this study. While there are confounding factors, such as year and food availability, between this and the earlier translocations, it does not rule out the effects that transmitters may have on foraging ability and metabolic condition.  It is common for SKR to lose weight in captivity. Seventy-six percent of the kangaroo rats in the transmitter treatment lost weight during captivity, on average losing 2% of their body weight. Because of this, we adjusted the reference point for harness size for this study. Rather than using the weight at time of outfitting, as we had in the past, we used the animal?s weight at capture. This allowed the harness to fit securely, but loose enough to allow for weight gain. We observed no difference in percent weight change between animals with and without transmitters.  In other attempts to minimize the harms posed by the transmitters, we removed them from animals that exhibited any initial signs of intolerance; this included struggle during outfitting or strained behaviours in their cages. Thus all 30 SKR used in the transmitter treatment were outfitted without resistance and performed unaffected behaviours immediately upon return to their cages. Additionally, we lessened the time to transmitter removal. In a previous SKR translocation, transmitters were left on for 3 months, but it was observed that kangaroo rats  88 settled in the new environment in fewer than 30 days, on average (Shier and Swaisgood, 2012). Thus, we considered that the transmitters had served their main purpose after 1 month.  While we have evidence to suggest that the use of radio transmitters, as fitted in our study, did not pose immediate threats to the animals? welfare or short-term survival, we do not know whether the transmitters had other effects, for example, we do not have data on activity budgets, energy expenditure or mating and reproduction in the short- or long-term.  Monitoring is clearly needed to measure the success of translocations. Without it, we have few means to examine survival, settlement and reproduction. As all methods pose some risk to the animal, the selection of a particular monitoring method is important and should be considered as carefully as other events of the translocation. The effect of radio transmitters on an animal?s welfare and fitness is influenced by many factors. Important considerations may include the physical attributes of the apparatus, method of application, the duration of placement, season of use, as well as selection of individuals.  When effective, transmitters can provide mortality information and range of movement that trapping and other indirect measures cannot. But, we know that translocation failure is in all probability linked to the effects of chronic stress and a primary aim should be to minimize stress (Dickens et al., 2010). Thus, we need to be especially cautious when adding any other stressor to the process, such as transmitters. As we attempted in this study, we would recommend tagging the fewest number of animals possible (see Walker et al., 2011). For small rodents, such as SKR, a combination of direct and indirect monitoring techniques may offer the best way of gathering the necessary information while minimizing disturbance to the animals.    89 5. Translocation success of Stephens? kangaroo rats: a behavioral and physiological profile of survivors  5.1 Introduction Wildlife translocation is a common conservation practice used to combat species loss and is incorporated into the recovery plans for many at-risk species. The global prevalence of the practice notwithstanding, its outcomes have been poor; some reviews estimate the success rate to be as low as 10-25% (Beck et al., 1994; Wolf et al., 1998). Broadly speaking, the stressors characteristic of a typical translocation may adversely affect an animal?s ability to survive in the wild (Molony et al., 2006; Teixeira et al., 2007; Swaisgood, 2010). While animals have evolved behavioural and physiological mechanisms to manage stress, these can be undermined when the quality of their environment shifts dramatically. The procedures involved in a typical translocation are intense and varied enough to cause such a shift as they include a combination of social and environmental disturbance (Teixeira et al., 2007; Dickens 2010; Baker, 2013).  With this in mind, the need to understand how and where stress emerges for translocated animals has been given increasing attention in the field of conservation biology (e.g., McEwen and Sapolsky, 1995; Mendl, 1999; Letty et al., 2000; Teixeira et al., 2007; Dickens et al., 2010). Researchers have shown that translocation events can affect stress hormone levels, body condition and overall survival (Franceschini et al., 2008; Dickens et al., 2009a; Pinter-Wollman et al., 2009). However, these and similar studies report only overall treatment effects and overlook the variation among individual animals. Such individual variation may be important because stressors can differentially affect individuals with different personality types and different life experience. For example, variation in animals? aggression levels, willingness to explore and approach novel objects, as well as the underlying physiological mediators of these  90 responses cause individuals to be more or less vulnerable in a challenging environment (Verbeek et al., 1994; Verbeek et al., 1996; Dingemanse et al., 2002; Bell, 2004). Thus, personality is increasingly seen as an important concept in conservation biology and should have significance for translocations as certain personality types may deal with the stress of translocation better than others (McDougall et al., 2006; Dickens et al., 2010; Swaisgood, 2010; Baker, 2013).  But there are few examples from the translocation literature where consideration at the individual level has been put into practice. Because of this lack, we looked at the differential stress effects of the translocation process on individuals within a translocated population of the endangered Stephens? kangaroo rat (SKR), Dipodomys stephensi. To do this we profiled individual SKR using behavioural and physiological measures and correlated these to survival. Specifically, we examined behavioural budgets during temporary captivity and assessed personality using conspecific and predator simulation tests (see Dochtermann and Jenkins, 2007; Vainikka et al., 2011). We measured activation of the hypothalamic-pituitary-adrenal (HPA) axis using fecal glucocorticoid concentrations as an indicator of physiological stress. In addition to being an indicator of stress, glucocorticoid levels have been identified as a key mediator of personality (Carere et al., 2010).  5.2 Materials and methods 5.2.1 Capture and captivity  As part of a multi-stage translocation effort, 152 Stephens? kangaroo rats (63 females and 89 males) were captured in August and September 2010 in Temecula, California. Animals came from two primary source sites. One site, ?El Sol?, was privately owned land located across from the Multispecies Reserve (N33 33 17.43, W117 02 13.49). Exotic grasses dominated this area;  91 the density of the grasses pushed SKR burrows near the dirt roads and along the road berms. The second site, ?Parking Lot? (P Lot), was the overflow parking lot at Lake Skinner campground (N33 35.207, W117 02.095). Aerial cover did not exist at this site and resident SKR were subjected to year-round noise and physical disturbance through campground activity (with an increase in summer months). To capture animals, we used modified Sherman live traps (30.5 ? 7.6 ? 9 cm) with shortened doors to prevent tail injury, baited with a mixture of white millet seed and raw oats. Traps were set just before sunset and were positioned just outside active burrow entrances. SKR were transported to the holding facility within the Sherman traps (5-15 minutes transport time).  SKR were temporarily held in captivity ?14 days before translocation, in a facility located within the Multispecies Reserve (N33 34 54.78, W117 01 36.87). Animals were housed individually in either plastic tanks (large Petco Keepers for small animals, 34.8 x 21 x 23.4 cm) or ten-gallon glass tanks (Petco, 48.72 x 30 x 30.24 cm). We provided each tank with sand, 7.2 cm thick (Quickrete Premium Play Sand no. 1113), a PVC ?burrow? (9.6 x 14.4 cm), animal bedding (Carefresh Basic pet bedding), and cardboard tubing for chewing. Animals were fed each evening 59 ml of an oats and millet seed mixture and a leaf of lettuce every other day.  We recorded sex and age for each SKR: adults were distinguished from juveniles through a combination of pelage condition and weight. Adults have a tawny brown pelage and no molt line in contrast to the gradated pelage and distinct molt line of juveniles. Animals were weighed upon capture and just before release using a digital bench scale.   On the second full day in captivity, each SKR was ear tagged (monel 1005-1) for individual identification and ear snipped for genetic analysis; this entailed removing a small fragment (1-2 mm in length, 0.5mm in width) of the pinna.   92  The temperature of the facility was maintained at 23 ? 2? C and a kept at a seasonable photoperiod of 13 h light:11 h dark. To conserve the light regime, human activity conducted during the dark cycle was done under red light.  5.2.2 Captive activity budgets   To assess behaviours in captivity we conducted behavioural observations on a subset of the total SKR translocated (N = 80; 35 adult females, 45 adult males). SKR held in captivity fewer than 7 nights were excluded from observations. Due to constraints of time and to limit disturbance, we used one-zero sampling as a recording method. Individuals were observed in 1-minute batches for 10 minutes every 4 hours on 3 different days in captivity. We recorded behaviour during the animal?s first full day in captivity, during the middle of captive holding and on its final day before release. Hours of observation were 21:00, 1:00, 5:00, 9:00, 13:00 and 17:00. We recorded the following behaviours: resting (outside the burrow), digging, sandbathing, chewing, seed-caching, feeding, grooming, jumping, and ?in the burrow?.    5.2.3 Personality tests   A subset of the SKR observed for captive activity budgets (N=60, 30 of each sex) was used in tests to assess personality. We used simulated conspecific and predator situations. Each SKR was subjected, one time, to a mirror image stimulation (MIS) test and a predator scent (PS) test while in temporary holding. MIS has been used as a test of social and agonistic behaviour in a variety species (e.g., Svendsen and Armitage, 1973; Blumstein et al., 2006; Boon, 2007; Dochtermann and Jenkins, 2007). Predator scent tests have been commonly used to measure boldness in animals (e.g., Dochtermann and Jenkins, 2007; Pruitt et al., 2010; Vainikka et al.,  93 2011). Forty SKR (19 females, 21 males) were tested first with the MIS; the other 20 SKR (9 females, 11 males) were tested first with the PS. We conducted trials in August and September 2010. We ran these trials during the animal?s 3rd and 6th nights in holding, thus individuals were given a rest period of 2 nights between tests. Trials were conducted between 22:00 and 03:00, falling within SKR active period. Trials were recorded for behavioral analysis with a handheld camcorder under the night mode setting (Sony Handycam HDR-SR12). Persons recording the trials were seated outside the sight of the kangaroo rats. To test animals, we constructed two identical arenas (96.5 x 67.3 x 35.5 cm) of black foam board to be used exclusively for one test (MIS or PS). The arenas were lined with black painter?s plastic to hold a bed of sand, 7.2 cm thick (Quickrete Premium Play Sand no. 1113).  A rectangular opening (7.6 x 9 cm), with a re-closable flap for SKR entry was cut into one end. A wooden barrier was positioned 64.3 cm from the entry for use during the acclimation period of the trials. To prevent escape a fitted lid made from fiberglass window screen (Phifer 18 x 24 mesh) was placed atop the arena. For both tests a seed-filled tray and a refuge (a rock with affixed twigs) were accessible during the entirety of the trial. The tray was positioned halfway along the edge of the arena and the refuge was placed in one of two corners near the entry. Test arenas were lit with a 60 W red light placed 1.9 m above. In the arena used for MIS trials, the entire wall opposite the entry was covered with a mirror. This was cleaned with vinegar between trials. For the PS trials, we sprayed (4 complete spritzes) of fox urine (Predatorpee.com) onto a rock; this was refreshed every 4th trial. The scented rock was placed in the corner far opposite the refuge. A fan was positioned to move the airflow away from the entry, in an attempt to prevent the kangaroo rat from detecting the scent  94 during the acclimation period. Behavioural differences between acclimation and active periods indicated this precaution was successful.  To release an animal into the arena, we transferred it from its home-cage burrow into another tube with a removable back. This tube was inserted into the entry and with a soft plunger we gently guided the animal into the arena. Between trials the top 2 cm layer of the sand was removed and replaced with new sand. Trials were 10 minutes, consisting of a 5-minute acclimation period and a 5-minute active period. After the acclimation period, we removed the barrier by sliding it to one side, through a slit in the arena wall. For both MIS and PS tests, video-recordings of the trials were viewed and scored for the following behaviours: latency to resume activity (active period), time standing still, time walking as well as number of maintenance events (grooming and sandbathing), jumping and digging. Grooming behaviour was defined as the cleaning of the body while the animal maintained an upright position. The animal stretching its body and rolling over to rub its dorsal gland along the substrate defined a bout of sandbathing. A jump was defined as complete bipedal vertical leap. A bout of digging was marked by use of the forearms to displace sand and lasted for a minimum of 2 seconds. A digging bout was deemed over if there was a 2 second cessation of this activity. Specific to MIS trials, scorers recorded latency to approach the mirror, the number of mirror approaches and the time at the mirror, all based on the criterion of the animal coming to within 5 cm of the mirror. For PS trials the latency to approach (within 1 cm) the urine-scented rock and time spent in physical contact with the rock were also quantified. Behaviours were scored independently by two people using the event-recording program JWatcher (Blumstein and Daniel, 2007).     95 5.2.4 Fecal collection   We collected fecal pellets, when available, from each animal (N= 152) at the time of initial capture (wild) and twice during captivity. We regarded all pellets collected from an individual for a given time window or during a collection event as a single sample. Fecal samples were stored in 2 ml screw cap micro tubes and placed immediately upon collection in a conventional freezer (set at -23? C) at the holding facility. Samples were later transferred to -80? C freezers prior to assaying. Fecal samples were assayed for cortisol at the San Diego Zoo?s Institute for Conservation Research (see Chapter 2 for a complete methodology of SKR fecal cortisol assay).   The time of night that captivity samples were collected was matched to the time of collection of the wild samples. We collected these twice during an animal?s time in holding: at its mid point and on its final night. Because SKR were trapped opportunistically over a 2-week period their length of time in captivity varied. As a result, the number of nights in captivity on which their samples were collected as well as the duration between those nights varied among individuals. For the majority of SKR, the mid point corresponded with their 5th or 6th night and the final fecal collection with their 10th, 11th or 13th night in holding. For other SKR mid and final collection corresponded with their 3rd or 4th and 5th or 6th nights. To ensure that fecal pellets were excreted during the desired time window, we moved animals to a new cage with fresh sand for the duration of the collection window, after which they were returned to their ?home? cage. To reduce the need for direct handling, animals were transferred in their PVC burrows.   5.2.5 Release   After a maximum captivity period of 2 weeks, SKR were ?soft-released? within the Riverside County Multispecies Reserve (N33 39.154, W117 00.077).  The kangaroo rats were  96 placed in acclimation cages at the release site for one week (see Shier and Swaisgood, 2011 for a description of this acclimation period). After release supplemental feed was provided for 1 month. Attempts were made to re-trap SKR for 3-5 consecutive nights at 1, 6 and 12 months after release.   5.3 Statistical analyses  All analyses were performed using IBM PASW (version 20.0 for Mac; SPSS, Chicago, Illinois). We ran full models with all relevant interaction terms. We removed any non-significant interactions from the final models and report only significant terms. FCC (wild and captive) were log transformed to improve normality. There was a strong correlation of FCC between the 2 samples taken during captivity (r = 0.678, p < 0.001). Because of this and the fact that we had a more complete sample for the first captive fecal collection period, we chose to use only these measures in the subsequent analyses. We conducted nonparametric bivariate correlations (because of their suitability to non-normally distributed data) to examine the relationship between wild FCC and the change in FCC between wild and captive levels. To determine the effect of capture and captivity on FCC we conducted a paired t-test comparing wild and captive levels. We used univariate general linear models (GLM) to analyze FCC. To examine FCCs (wild, captive and change between wild and captive levels) we included home environment (Parking Lot and El Sol), sex and age in this order in the models. Type I sum of squares was selected to manage for multicollinearity among the fixed factors. We conducted paired t-tests to examine the differences between wild and captive FCCs for each home environment separately.   97 We used binary logistic regression to examine predictors of early survival (not re-trapped vs. re-trapped 1, 6 or 12 months), early versus later survival (re-trapped only after 1 month vs. re-trapped 6 or 12 months), as well as to examine behavioural predictors of later survival (not trapped vs. re-trapped 6 or 12 months). Because home environment was a significant explanatory variable for survival and FCC, we examined the differences between the 2 populations in captive behaviour and personality analyses. We performed a univariate GLM to analyze the effect of home environment on captive activity levels, averaged across all behaviours and observation days. A multivariate GLM, which considered all 8 assessed behaviours, was initially conducted. In this model sex was included as a fixed effect in addition to home environment. Only population was a significant factor and significant across all behaviours. For personality tests, we used a multivariate GLM to examine the effect of home environment and test order (MIS first or PS first) on assessed behaviours.  5.4 Results 5.4.1 Captive activity budgets There was a significant difference between one-zero scores, across all behaviours, for SKR from El Sol and from P Lot (El Sol, mean?SE, 0.4?0.1; P Lot, mean?SE, 0.12?0.04; F1,78 = 81.71, p < 0.001). El Sol SKR (mean?SE, 7.9?0.18) were observed resting in the burrow less often than P Lot SKR (mean?SE, 9.3?0.09) and El Sol SKR (mean?SE, 1.2?0.15) were observed resting outside of the burrow more often than P Lot SKR (mean?SE, 0.12?0.03). The rank order for the three least common behaviours (grooming, seed-caching and sandbathing) was the same for both populations (Fig. 11).   98 5.4.2 Personality tests  There were significant differences between the two home environments in behavioural responses to both MIS and PS tests (Table 4). In both tests, P Lot SKR spent more time standing still (MIS, F1,55 = 4.40 p = 0.04; PS, F1,55 = 32.9, p < 0.001) and less time walking than El Sol animals (MIS, F1,55 = 8.71, p = 0.005; PS, F1,55 = 15.44, p <0.001). P lot SKR took longer to resume activity upon exposure to the mirror (F1,55 = 8.32, p = 0.006). However, P Lot kangaroo rats were quicker to approach the mirror (F1,55 = 3.02, p = 0.050) and spent more time interacting with it than El Sol animals (F1,55 = 4.47, p = 0.039). Additionally P Lot animals had longer latency periods to approach the urine-covered rock and spent less time in contact with it than El Sol SKR (latency to approach rock, F1,55 = 3.66, p = 0.050; time with rock, (F1,55 = 9.15, p = 0.004). Maintenance behaviours were seen more in El Sol animals. On average, El Sol SKR groomed and sandbathed more than P Lot SKR (grooming, MIS, F1,55 = 3.83, p = 0.05; PS, F1,55 = 5.55, p = 0.022; sandbathing, PS, F1,55 = 4.99, p = 0.030). While jumping and digging were not affected by home environment, SKR who had experience with the testing arena exhibited more of these activities on their second test than on their first (MIS, F1,55 = 5.7, p = 0.027; PS, F1,55 = 9.1.9, p < 0.001).  5.4.3 Fecal cortisol   Kangaroo rats with higher wild FCC tended to have higher captive levels (rs (131) = 0.428, p < 0.001), and exhibited a smaller increase in FCC during captivity (rs (131) = - 0.512, p < 0.001). FCC from captive samples (log transformed) were higher on average than from wild samples (t130 = -4.81, p < 0.001). This pattern held for both home environments (P Lot, t83  = -3.05, p < 0.003; El Sol, t48 = -4.15, p < 0.001; Fig. 12). Overall, P Lot kangaroo rats had higher  99 wild and captive FC levels than El Sol animals (wild F1,127 = 6.71, p = 0.011; captive, F1,127 = 6.47, p = 0.012). There were no age or sex differences in FCC.   5.4.4 Survival  A significantly higher percentage of P Lot SKR survived translocation to 1 month than El Sol SKR (Exp (B) = 2.263, 95% CI 1.067 - 4.799, p = 0.033). Of the 102 P Lot kangaroo rats released, 76 were re-trapped (75% survival), compared to only 26 of the 50 El Sol kangaroo rats re-trapped (52% survival). Home environment was a predictor of early survival (1 month), but not of later survival (6 and 12 months; (Exp (B) = 1.464, 95% CI 0.700 ? 3.060, p = 0.311).  Change in FCC (from captive to wild) was predictive of later survival (Exp (B) = 0.989, 95% CI 0.939 ? 1.001, p = 0.050). On average, animals with a smaller change in FCC were more likely to be alive at 6 and 12 months compared to individuals with a greater change in FCC. Probability estimates from the logistic regression analysis indicated that animals with a mean FCC change in the upper 75% quartile (24 ng/g) had only a 34% chance of survival compared to animals with a FCC change in the lower 25% quartile (-12 ng/g), which were estimated to have had a 62% chance of later survival. Animals who, on average, showed no change in FCC during translocation had a 50:50 chance of surviving beyond 1 month.     100 Fig. 11 One-zero score (proportion of all sample intervals during which the behaviour occurred) of Stephens? kangaroo rats from 2 populations (P Lot and El Sol); Sampling occurred during captivity, before translocation.         0 0.2 0.4 0.6 0.8 1 1.2 1.4 Resting out of burrrow Jumping Digging Feeding Chewing Grooming Seedcaching Sandbathing  Proportion of all sample intervals that behaviour occured P Lot El Sol  101 Table 4 Behaviours from mirror-image (MIS) and predator scent (PS) tests associated with Stephens? kangaroo rats from two populations (P Lot and El Sol)                        + denotes a positive correlation / ? denotes a negative correlation     Home Environment   Behaviour P Lot  El Sol    Standing still (MIS/PS)  + ? Latency to resume activity (MIS)  + ? Time at mirror (MIS)  + ? Latency to urine-coated rock (PS)  + ? Latency to mirror (MIS)  ? +  Walking (MIS/PS)  ? + Time at urine-coated rock (PS)  ? + Grooming (MIS/PS)  ?  +  Sandbathing (PS)  ? +  102  Fig. 12 Fecal cortisol concentrations (mean?SE) at time of capture and during captivity, prior to translocation, for Stephens? kangaroo rats from two populations (P Lot and El Sol)        0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00 Wild Captive Fecal cortisol (ng/g)	 ?P Lot El Sol  103 5.5 Discussion  The results of this study provide further evidence that translocation events are stressful to the animals and affect their ability to survive after release. Importantly, this study also shows that translocation stressors differentially affect individuals, revealing that certain personality types do indeed cope with translocations better than others. Furthermore, an animal?s original home environment may have important implications for its translocation success.   Taken together, the behavioural and physiological measures that we used to assess individual variation suggest that different biological repertoires may be advantageous for different aspects of post-release survival. In total, kangaroo rats with more cautious, less excitable, but more social behaviours, coupled with a more active HPA axis were associated with higher early survival. While animals surviving longer revealed no obvious behavioural differences, they did mount a significantly smaller stress response to translocation stressors compared to the rest of the translocated population.  Overall, FCC increased for kangaroo rats during captivity and remained elevated until their release (? 14 days). This suggests that the kangaroo rats did not acclimate to captivity and holding events, and that this period was a cause of chronic stress. It is well established that prolonged elevation of glucocorticoids can become a biological burden; extended activation of the stress system is known to affect immune system function, energy allocation as well as activation of the stress response itself (Saplosky, 1993; Moberg, 2000; Dickens et al., 2010). This may explain why kangaroo rats with the greatest increase in FCC tended to have poor long-term survival and animals with a less reactive HPA axis fared better over the long run. While a relatively small change in FCC was a strong predictor of longer-term survival, it appears that a less reactive HPA axis was not as important for immediate survival. There was no significant  104 difference in FCC change between kangaroo rats that survived to 1 month and those that were never re-trapped. Survival in the short-term, however, seemed to be strongly influenced by a particular personality type.   Kangaroo rats from the Parking Lot habitat had higher survivorship during the first month after release compared to El Sol animals. The behavioural and physiological profiles of these animals suggest several factors that may have contributed to their greater survival. In behavioural tests, P Lot animals were more cautious than animals from El Sol; they explored less and took longer to approach predatory stimuli. When observed in their cages P Lot SKR also spent most of their time in the available burrows, whereas El Sol SKR were predominantly found to be resting outside the burrows (Fig. 11). This is consistent with the results of Bremner-Harrison and colleagues (2004) that show that swift-foxes (Vulpes velox), exhibiting greater levels of cautiousness, had higher survival after release than bolder foxes. Bolder swift-foxes were quicker to approach novel stimuli, traveled farther and emerged from dens more quickly. Although basal FCC alone was not predictive of early survival there was a strong correlation between basal FCC and home environment. On average, El Sol animals had lower basal FCC than P Lot ones. This typological relationship between personality and HPA measures has been observed in other species, including humans. Lower basal level has been associated with less cautious, more excitable individuals, in contrast to higher levels in cautious, less excitable ones (Kagan et al., 1988; Hessing et al., 1994; Captianio et al., 2004). While the causal relationship is difficult to determine, it is thought that the higher basal levels indicate a greater physiological responsiveness (Capitanio et al., 2004).  The quality of habitat of the two home environments was quite distinct.  The Parking Lot was devoid of aerial cover and barn owls were often present (pers. obs.). El Sol, however, had  105 dense grass and shrub cover, which likely provided cover from owls. The consistent behavioural differences seen in captivity may reflect behaviour learned in their respective habitats. Familiarity with an open, high-predation-risk habitat could explain the reluctance of P Lot individuals to emerge from their artificial burrows during captivity and their general caution during personality tests. Such behavioural traits, in combination with a more responsive stress system, could well make these animals better equipped to avoid predation during the vulnerable period shortly after release into a new environment.  Despite their greater cautiousness, P Lot individuals exhibited more conspecific interest as shown by the MIS test (see Chapter 3 for a discussion on this interpretation). In the Parking Lot, burrow systems are quite close and the openness of the area puts these animals in more constant visual contact and increases the chance of interaction. Therefore, it may be adaptive to have a high level of social flexibility under such conditions (Schradin, 2013). Shier and Swaisgood (2012) have shown that SKR released with familiar individuals (neighbours) have higher survival than those released in unfamiliar groups. Among strangers, they witnessed increased agonistic behaviours. To the best effort, SKR from this study were translocated in neighbour groups. If indeed more socially flexible, the importance of neighbours may have been more important for P Lot animals than for El Sol SKR. Moreover a higher level of social flexibility may have made P lot animals less affected by strangers, reducing the energy expended towards agonistic interactions. This coupled with high cautiousness could be another reason for their success over El Sol SKR in the short term; individuals from the Parking Lot may have been better at predator avoidance and less affected by the social disruption caused by the translocation.  106   Longer-term survival seems to be attributable to other biological parameters. As mentioned, longer-term survivors showed a smaller change in FCC, but no apparent behavioural differences compared to animals that survived only to 1 month. As with the HPA measures associated broadly with the two home environments, a small HPA response may be linked to an altogether different personality type that our methods were not sensitive enough to discern. Rather than relying on behavioural scoring from the personality tests, using subjective ratings of the individuals may offer greater insight to the personality dimensions associated with various stages of translocation.   Increasingly, researchers believe that the presence of a wide complement of personality types is important for maintaining a resilient population, particularly to combat labile environments (Wingfield, 2004; Dingemanse et al., 2004; Wolf et al., 2007). In translocations we may be inadvertently selecting for a narrower range of personalities that can cope with the stress of translocation and thus establishing a population without a sufficient variation. A key question for translocations is how to reduce stress to improve individual welfare and survival for a wider range of personalities? In the case of Stephens? kangaroo rats modifications may be needed to housing and the period of captivity in order to reduce activation of the stress response. Possible solutions might include increasing the cage size to offer relief to more excitable individuals (Wielebnowski et al., 2004) and anti-predator training for less cautious animals, such as those from El Sol.   This study lends convincing support that variation in personality affects how well an animal copes with translocation, with consequences for survival. Knowing how to manage different personality types may determine how successfully a translocated population establishes itself.   107 6. General discussion and conclusions  6.1 Introduction Translocation biology has made significant strides over the past 30 years. Beyond the large, charismatic animals that once dominated conservation efforts, translocations are now part of the recovery programs for a greater diversity of species including reptiles (Nussear et al., 2012), amphibians (Germano et al., 2009), fish (Hammer et al., 2013), butterflies (Harris, 2008) and mollusks (Wilson et al., 2001). This may be a mark of dire ecological conditions, but it also represents a more inclusive perspective on species worth. Moreover, practitioners view translocation biology as a means to combat the effects of human-induced environmental change (Richardson et al., 2009; Seddon et al., 2009). For example, ?assisted migration? moves animals beyond their natural range to proactively protect them from anthropogenic threats (Vitt et al., 2009; but see, Ricciardi and Simberloff, 2009) and ?ecological replacement? fills a niche left vacant by extinction with a comparable species (Seddon, 1999).  Given that translocation continues to be a popular conservation tool with broadening application, it is encouraging that the practice faces increased scrutiny. It is not surprising then that the factors that need to be taken into account have evolved. Habitat suitability, long-term food availability, season of release, type of release, and source of animals (Jule et al., 2008) in addition to an understanding of behaviour, life history patterns and habitat use, are considered fundamental translocation criteria. Disease risk-assessment (Ewen et al., 2011; Sainsbury et al., 2012) and genetic diversity (Vucetich and Waite, 2000; Keller et al., 2011; Jamieson and Lacy, 2011) are receiving greater attention by practitioners as well.   Importantly, stress is now recognized as an element of translocation concern, as it is likely a major contributor to the high mortality that has plagued the practice (Teixeira et al., 2007;  108 Dickens et al., 2009b, 2010). But very few translocation programs assess success at the individual level (Jule et al., 2008; Parker et al., 2011). This is counterintuitive and counterproductive when the success of a translocation hinges on the survival of relatively few animals. While the definition of translocation success is still somewhat elusive, it must include post-release survival, settlement at the site and high recruitment (IUCN, 1987, 1998, 2012; Seddon 1999; Gosling and Sutherland, 2000; Teixeira et al. 2007). Because this establishment period is critical to the long-term viability of a translocated population, understanding how stress affects individuals and mitigating for these stressors must be incorporated into translocation protocols. Thus, attention to individuals cannot be overestimated. This is where animal welfare science is arguably useful for both ethical and practical purposes in conservation.  Despite its entrenched position in conservation and even with the advances made over the last few decades, translocation biology struggles to improve its practices. With the ongoing extinction crisis and the suffering that inferior translocations bring to animals, systematic improvement is urgently needed. Researchers such as Burbridge et al. (2011) and Parker et al. (2011) believe a multidisciplinary approach offers the best chance of success by merging expertise from a range of disciplines to manage the ecological, biological, logistical, and political concerns that arise. While this should be prescribed by the demands of translocation (Parker et al., 2011), a guiding framework still eludes the practice.  Animal welfare science methodology may provide the conceptual infrastructure to improve the immediate and ultimate survival of translocated individuals. Animal welfare science is multidisciplinary, drawing upon information and methodology from a host of other disciplines to address individual-level concerns (Fraser, 2010). The core disciplines ? veterinary medicine, biology, stress physiology, and animal behaviour ? are directly applicable to the translocation  109 process. Because the empirical work is grounded in value-based assumptions regarding the moral significance of animals and their quality of life, animal welfare science also has strong ties to psychology, philosophy, bioethics, sociology and political science (Fraser, 2010). An individual?s welfare, in a conservation context, may be defined by its ability to manage the translocation process, marked primarily by survival and a return to functional behaviours (to achieve what may be called a ?thriving survival?). In this case, higher-level concerns, such as the elements of landscape ecology, population biology and behavioural ecology, can be seen as essential to animal welfare science.    My dissertation research represents one of the first translocation projects to attempt an individual-based approach to understanding translocation success by systematically implementing animal welfare science methodologies. To understand where stress emerges for animals and how individual animals respond to translocation stressors, I took a multi-pronged approach using measures of stress physiology, animal behaviour, and psychology.   6.2 Chapter review 6.2.1 Validation of fecal cortisol assay (Chapter 2) In collaboration with an endocrinologist, I developed a radioimmunoassay to measure fecal cortisol in SKR. The radioimmunoassay was validated by its specificity for the fraction that corresponded to the retention time of cortisol. The predicted increase in fecal cortisol after exposure to fox urine was reliably detected by the radioimmunoassay. However, because I did not positively identify the compound (using mass spectrometry, for example), I cannot claim a fully validated assay. Moreover, many endocrinologists view adrenocorticotropic hormone (ACTH) and dexamethasone (Dex) tests as the gold standard for hypothalmic-pituitary-adrenal  110 Axis (HPA) manipulation. Since I did not observe the expected change in FC following Dex and ACTH tests, skeptics might reject the claim that our assay is validated. However, isolating an immunoreactive fraction corresponding to the analyte of interest in conjunction with subsequent physiological validation is considered strong evidence of a reliable assay (Touma 2005; Touma and Palme 2005) and has been demonstrated successfully in a number of different studies (Whitten et al., 1998; Terio et al., 1999; Harper and Austad, 2001; Wielebnowski et al., 2002).  In retrospect it may have been better to collect fecal samples at multiple and/or shorter intervals, rather than at only capture and 2-3 days after urine exposure. Because I was working with a group of SKR just before release, my intention was to minimize disturbance. Without an extended collection period I was unable to determine how quickly a change in HPA activity can be detected in feces and how long elevated adrenal activity lasts following exposure to a stressor. Without more frequent fecal samples following predator urine exposure and corresponding blood samples I also could not answer how the magnitude of change in FC compares to the magnitude of change in adrenal activity. Although such information would be needed in order to use the assay for acute stress responses, the assay should be useful for chronic changes in adrenal activity, for example, to compare HPA activity among translocated populations subjected to different management regimes, and to compare how individuals respond to translocation events, such as monitoring (Chapter 4) and captivity (Chapter 5). An advantage of using a FC assay in addition to other measures of translocation efficacy is that FCCs could be acquired before longer-term measures of success, such as survivorship and fecundity could be obtained.  6.2.2 Personality composition (Chapter 3) Principal component analysis of subjective ratings identified three key dimensions of SKR personality that I interpreted as Assertiveness, Excitability, and Persistence. Kangaroo rats  111 demonstrated stability of these personality traits across time and context: animals scored similarly in conspecific (mirror-image) and predator (fox urine) tests, conducted two days apart. As shown in other species, low Assertiveness was associated with high basal glucocorticoid activity (cortisol for SKR) and high Assertiveness was associated with low basal cortisol activity. The three personality dimensions that emerged from the factor analyses did not show any obvious clustering that would suggest discrete personality types. However, behavioural and physiological correlations suggested the presence of a proactive-reactive coping typology described for other rodent species (Koolhaas et al., 1999). Similar to other animals characterized as proactive, SKR high in Assertiveness were more exploratory, displayed riskier behaviour and had lower basal cortisol. In contrast, SKR low in Assertiveness were less exploratory, showed less risky behaviour and had a higher basal cortisol, similar to reactive animals. SKR revealed a negative correlation of exploration and riskier behaviour with measures of agonistic behaviour. This was unexpected, as species that have been characterized by proactive-reactive types tend to show a positive correlation between levels of exploration, risky behaviour and agonism (e.g., Gentsch et al., 1982; Steimer et al., 1997; Koolhaas et al., 1999).  In kangaroo rats, agonistic behaviour often involves rushing by one animal, kicking sand into the other?s face, leaping onto the other and erratic jumping to avoid a rush (Jones, 1993). Overt displays of these behaviours were not observed in mirror-image-stimulation (MIS) tests with SKR. However, the discrepancy I observed between correlations of agonism and boldness in SKR and other species could be because the behaviours I used to measure aggression were not accurate or the MIS test did not elicit agonistic behaviour in SKR. In either case, I have recorded behaviours reflective of other social patterns indicative of consistent behavioural variation.   112 The usual inference is that the behaviours observed in response to the mirror image offer a good measure of agonism (Gallup; 1968; Svendsen and Armitage, 1973). Certainly this is hard to dispute when an animal responds to the mirror with behaviours categorically characteristic of agonism. Stereotyped threat behaviour, in response to mirror image, has been seen in a variety of taxa (siamese fighting fish, Betta splendens, Lissman, 1932; Thompson and Sturm, 1965; male sticklebacks, Gasterosteus aculeatus, Tinbergen, 1951; towhees, Pipilo juscus petulans, Ritter and Bensen, 1934; California sea lions, Zalophus californicus, Schusterman et al., 1966). However, measures of latency have also proven useful as an indication of agonism (Svendsen and Armitage, 1973; Dochtermann et al., 2007). In confirmatory tests, Hargett (2006) found that latency to approach the mirror was negatively correlated with aggression levels and dominance status as determined by pair-wise interactions of Merriam?s kangaroo rats. On reexamination, it would have been useful to conduct preliminary tests for SKR to confirm the inferences generated by the MIS test. The standard for assessing personality is to show consistent differences across at least two contexts and over time (Sih and Bell, 2004). Our tests were separated by only two days, largely because of time constraints of the translocation. While this is a relatively brief period, strong test-retest correlations have been found irrespective of the time interval (Gosling, 2001), and a more recent meta-analysis by Bell et al. (2009) reveals that repeatability tends to be stronger at shorter intervals. Given the factors influencing behaviour over time, such as sexual maturity, environmental change and state differences, a shorter interval may be preferential to ensure that the same trait is being tested across contexts.  If I had the luxury of time it would have been valuable to record behaviours in the animals? home environments before capture and after release; doing so likely offers the most  113 complete insight into individual variation (Powell and Gartner, 2011). Moreover behaviours measured in the field were found to be more repeatable than those recorded in captivity (Bell et al., 2009). In addition to time constraints, however, individual identification is often difficult with SKR. We have had limited success with colour-coded reflective tape placed on ear tags; a reliable identification is often not possible because of distance and angle to the kangaroo rats, as well as their rapid movements.  6.2.3 Effects of radio transmitters (Chapter 4) I observed a short-term (1 week) increase in fecal cortisol in SKR outfitted with backpack transmitters, but cortisol levels had returned to captive levels by one month after release. Radio transmitters did not affect survival estimates at one month after release based on a match-pair design. When unrestricted by the match-pair design, survival estimates were somewhat lower, although not statistically significant, for SKR with radio transmitters at one, six, and twelve months after release. I do not know whether the difference would become significant with a larger sample size, but in any case the differences were modest. A meta-analysis of transmitter effects on avian behaviour revealed that transmitters caused a marked increase in energy expenditure and reduced nesting time (Barron et al., 2010). In situ observations after release would likely have given us more insight into the behavioural effects of the transmitters on SKR, perhaps elucidating if the slightly lower survival was because of the transmitters or sampling error.  Kangaroo rats were placed in acclimation cages at the release site a day after they were outfitted with transmitters. I collected fecal samples on the first and final full days in  114 acclimation. In retrospect, it would have been better to collect samples from each day during acclimation to obtain a more accurate picture of how quickly the transmitters affect cortisol levels and how long-lived the effect is. Given that fecal cortisol levels, for those with transmitters, were still elevated at one week into the acclimation period, it may be recommended to either extend the acclimation period or to put the transmitters on earlier in captivity to allow for the stress response to abate.    6.2.4 Translocation success (Chapter 5)  I determined that the events surrounding capture and captivity were a source of chronic stress for SKR as captive SKR had consistently high fecal cortisol levels.  Unexpectedly, a kangaroo rat?s home environment was a significant predictor of its ability to cope with translocation. Animals from the two different populations (Parking Lot and El Sol) displayed distinct behavioural and physiological differences. These differences conformed to the types of coping styles I identified in Chapter 3. Parking Lot SKR displayed less exploratory and risky behaviours in personality tests and tended to have higher basal fecal cortisol than SKR from El Sol. Parking Lot SKR also had higher survival rates at one month after release. However, SKR with a smaller change in fecal cortisol during captivity, regardless of population, had a greater chance of surviving to six or twelve months after release.  A great methodological frustration was not having the time or personnel to include each of the 152 SKR captured for translocation in behavioural and personality tests. (Of these 152, 80 were included in behavioural observations and 60 of these 80 in personality tests.) If I had been able to include all, I would have had behavioural and personality profiles, in addition to the cortisol profiles, for all of the survivors at each of the re-trapping periods (one, six, and twelve  115 months). With this information, the survivor profiles could have been even more revealing of the traits supportive of translocation success at various stages.  Although individual identification is difficult, for the reasons stated above, it would have been highly informative if I could have profiled the as-yet translocated animals while in their home environments. This would have provided a reference point for the behaviours observed during captivity; from this comparison I could have had another measure of coping ability.   6.2.5 Conclusions My thesis, while providing a bridge from theory to application to implementation, is ultimately descriptive, and stops short of providing tested mitigation methods. That being said, taken in its entirety, this thesis shows a compelling relationship among behaviour, physiology, and experience in Stephens? kangaroo rats, with significant consequences for translocation survival. These studies show that translocation stressors differentially affect individuals; revealing that SKR with certain personality types and certain life experience, do indeed cope with translocations better than others.   From this study I can conclude that in Stephens? kangaroo rats: (1) A complex personality structure, with behavioural and physiological correlates, exists.  (2) Chronic stress can be monitored non-invasively through fecal cortisol assays. (3) Translocation procedures are a source of chronic stress. (4) An animal?s home environment is correlated with personality type. (5) Personality and reactivity of an animal?s stress response are strong predictors of translocation survival.   116 6.3 General recommendations and future research Events such as, capture, captivity, transport and release are essential components of translocation, all of which are novel situations that come with a loss of control and predictability for the animals. This is a grave loss as the ability to control and predict one?s environment is vital to successful coping (see Chapter 1, p. 14). Thus, in addition to reducing the number of stressors imposed in a translocation, we need also to examine how to return control to the animals and ease the unpredictability of the process. This could include designing capture methods that reduce its predation-like effect. In addition to the physical attributes of the capture technique, duration in the trap and timing of capture are also concerns (O?Neill et al., 2008; Casper, 2009). For a species like SKR, we can use baited, walk-in traps that may offer some level of choice. While initial capture may pose minimal threat (although we have no evidence of this) as the kangaroo rats are usually occupied eating the seed, increased trap time may potentiate the stress response.  Determining which features of a holding cage are critical can also ease the loss of control and predictability. For example, cage shape appears to be important for certain bird species (e.g., European starlings, Asher et al., 2009). Providing an outlet for motivated behaviours is also likely to be important in captivity (Dawkins, 1990). Anecdotally, SKR in traps near dawn seemed more agitated than animals trapped earlier in the night. Moreover, in data not presented in the thesis, captive SKR were observed jumping significantly more near dawn than at other times of the day. Presumably as a nocturnal, burrow-dwelling species they are motivated to return to their burrows before sunrise. Although we provided the kangaroo rats with artificial burrows and sand in the holding cages, this may not be sufficient to satisfy their needs.   117 For amenable species, such as arthropods, researchers have devised methods to eliminate the need for trapping, handling and dramatic shifts in environments. For example, artificial refugia were placed in the habitat of a cricket-like insect (Hemideina thoracica) months prior to translocation. When the insects took up residency, the refugia were transported directly to the new site (Green, 2005). While the extent of this ?hands-free? approach may not be feasible for many species, including SKR, I think it is worth investigating how closely it can be achieved. This may entail trading off examinations to reduce handling time and to minimize exposure to novel situations. The diversity of animals, across and within taxa, makes it impossible to recommend boilerplate guidelines for translocation methodologies (Parker et al., 2011). Parker et al. (2011), however, rightly point out that there are themes common to the more successful translocations: (1) they are carefully planned and carried out by a multidisciplinary team (e.g., managers, scientists, animal husbandry experts; (2) planning and implementation is guided by an intimate knowledge (biological, ecological) of the species; and (3) stress is considered in the planning. Without attention to individual variation, however, we could not begin to understand where and how stress emerges for the animals in order to adjust our procedures appropriately. Thus, I would go one step further to recommend that assessment of individual variation and how variation influences translocation success should be included in translocation protocols. If it were properly implemented, it too would be an important theme common to successful translocations. Moreover, I would suggest that inclusion of personality would indeed be proof of careful planning, intimate knowledge of the species and consideration of stress. Asking species-specific questions such as, when is the best season for translocation, where is the best habitat for release,  118 and how shall we release the animals are crucial, but limited if we do not understand who we are translocating.      119 References  Ackerman JT, Adams, J, Takekawa JY, Carter HR, Whitworth DH, Newman SH, Golightly RT, Orthmeyer DL. 2004. Effects of radio transmitters on the reproductive performance of Cassin?s auklets. Wildlife Society Bulletin, 32, 1229-1241.  Allen DL. 1954. Our Wildlife Legacy. Funk and Wagnalls, NY, NY.   American Society of Mammalogists, 1987. Guidelines on acceptable field methods in mammalogy.  Anil L, Anil SS, Deen J, Baidoo SK. 2006. Cortisol, behavioral responses, and injury sores of sown housed in gestation stalls. Journal of Swine Health and Production, 14, 196-201.  Archard GA and Braithwaite VA. 2010. The importance of wild populations in studies of animal temperament. Journal of Zoology, 281, 149-160.  Archer J. 1988. The Behavioural Biology of Aggression. Cambridge University Press: Cambridge, UK.   Armstrong DP and Seddon PJ. 2008. Directions in reintroduction biology. Trends in Ecology and Evolution, 23, 20-25.   Asher L, Davies, GTO, Bertenshaw CE, Bateson, M. 2009. The effects of cage volume and cage shape on the condition and behaviour of captive European starlings. Applied Animal Behaviour Science, 116, 286-294.   Azpiroz A, De Miguel, Fano E and Vegas O. 2007. Relations between different coping strategies for social stress, tumor development and neuroendocrine and immune activity in male mice. Brain, Behavior and Immunology, 22, 690-698.  Baker, L. 2013. Why individuals matter: lessons in animal welfare and conservation. In: M. Bekoff (ed) Ignoring Nature No More. Chapter 12. University Chicago Press.  Barron, DG, Brawn, JD, Weatherhead, PJ. 2010. Meta-analysis of transmitter effects on avian behaviour and ecology. Methods in Ecology and Evolution, 1, 180-187.  Beck BB, Rapaport LG, Stanley Price MR and Wilson AC. 1994. Reintroduction of captive-born animals. In: P. Olney (ed) Creative Conservation: Interactive Management of Wild and Captive Animals. Chapman & Hall: London, UK. 265-286.   Bell AM. 2004. Behavioural differences between individuals and two populations of stickleback (Gasterosteus aculeatus). Journal of Evolutionary Biology, 18, 463-474.  Bell AM, Hankison SJ, Laskowski, KL. 2009. The repeatability of behaviour: a meta-analysis. Animal Behaviour, 77, 771-783.  120  Bell AM, Henderson L, Huntingford FA. 2010. Behavioral and respiratory responses to stressors in multiple populations of three-spined sticklebacks that differ in predation pressure. Journal of Comparative Physiology, 180, 211-220.  Benus RF, Den Daas SJ, Koolhaas JM, van Oortmerssen GA. 1990. Routine formation and flexibility in social and non-social behaviour of aggressive and non-aggressive male mice. Behaviour, 112, 176-193.   Bertreaux D, Duhamel R, Bergeron J-M. 1994. Can radio collars affect dominance relationships in Microtus? Canadian Journal of Zoology, 72, 785-789.  Beyer DE, Winterstein SR, Lederle PE, Maskey JJ. 2011. Moose in Michigan: history, biology and considerations for hunting Report to the Michigan Department of Natural Resources and Environment Moose Hunting Advisory Council.  Biggins DE, Vargas A, Godbey JL, Anderson SH. 1999. Influence of prerelease experience on reintroduced black-footed ferrets (Mustela nigripes). Biological Conservation, 89, 121-129.  Billingslea, FY, 1941. The relationship between emotionality and various other salient of behavior in the rat. Journal of Comparative Psychology, 31, 69-77. Bleich, VC, 1977. Dipodomys stephensi. Mammalian Species, 73, 1-3. Blumstein DT, Holland B-D, Daniel JC. 2006. Predator discrimination in Vancouver Island marmots. Animal Conservation, 9, 274-282.  Blumstein DT, Daniel JC, Evans CS. 2006. JWatcher 1.0.  Bolig R, Price, CS, O'Neill, PL, Suomi, SJ. 1992. Subjective assessment of reactivity level and personality traits of rhesus monkeys. International Journal of Primatology, 13, 287-306.  Boon AK. 2007. The interaction between personality, offspring fitness and food abundance in North American red squirrels. Ecology Letters, 10, 1094-1104.  Borchelt PL. 1983. Aggressive behavior of dogs kept as companion animals: Classification and influence of sex, reproductive status and breed. Applied Animal Ethology, 10, 45-46.  Both C, Dingemanse NJ, Drent PJ, Tinbergen JM. 2005. Pairs of extreme avian personalities have highest reproductive success. Journal of Animal Ecology, 74, 667-674.  Bradshaw SD, Grenot CJ. 1976. Plasma aldosterone levels in two reptilian species, Uromastix acanthinurus and Tiliqua rugosa and the effect of several experimental treatments. Journal of Comparative Physiology, 111, 71-76. Bremner-Harrison S, Prodohl PA, Elwood RW. 2004. Behavioural trait assessment as a release criterion: boldness predicts early death in a reintroduction programme of captive-bred swift fox  121 (Vulpes vulpes). Animal Conservation, 7, 313-320.   Bright, PW, Morris, PA. 1994. Animal translocation for conservation - performance of dormice in relation to release methods, origin and season. Journal of Applied Ecology, 31, 699-708.  Brittas R, Marcstr?m V, Kenward RE, Karlbom M. 1992. Survival and breeding success of reared and wild ring-necked pheasants in Sweden. Journal of Wildlife Management, 56, 368-376.  aBrock RE and Kelt DA. 2004. Keystone effects of the endangered Stephens? kangaroo rat (Dipodomys stephensi). Biological Conservation, 116, 131-139. bBrock RE and Kelt DA. 2004. Conservation and social structure of Stephens? kangaroo rat: implications for burrow-use behavior. Journal of Mammalogy. 85, 51-57. Budaev SV. 1997 "Personality" in the guppy (Poecilia reticulata): A correlational study of exploratory behavior and social tendency. Journal of Comparative Psychology, 111, 399-411.  Buirski P, Kellerraan H, Plutchik R, Weninge, R, Buirski N. 1973. A field study of emotions, dominance, and social behavior in a group of baboons (Papio anubis). Primates, 14, 67-78.  Burbridge AH, Maron M, Clarke M, Baker J, Oliver DL. 2011. Linking science and practice in ecological research and management: how can we do it better? Ecological Management and Restoration, 12, 54-60.  Burnham KP, Anderson DR. 2002. Model selection and multimodel inference: a practical information-theoretic approach, 2nd Ed., Springer-Verlag, New York. Busch DS, Hayward LS. 2009. Stress in a conservation context: A discussion of glucocorticoid actions and how levels change with conservation-relevant variables. Biological Conservation, 142, 2844-2853. Bush M, Beck BB, Dietz J. 1996. Radiographic evaluation of diaphragmatic defects in golden lion tamarins (Leontopithecus rosalia rosalia): implications for reintroduction. Journal of Zoo Wildlife Medicine, 27, 346-357  Byrne G and Suomi SJ. 2002. Cortisol activity and its relation to homecage behavior and personality ratings in tufted capuchin juveniles from birth to six years of age. Psychoneuroendocrinology, 27, 139-154.  Capitanio JP. 1999. Personality dimensions in adult male rhesus macaques: prediction of behaviors across time and situation. American Journal of Primatology, 47, 299-320.  Capitanio JP, Mendoza SP, Baroncelli S. 1999. The relationship of personality dimensions in adult male rhesus macaques to progression of simian immunodeficiency virus disease. Brain, Behavior and Immunology, 13, 138-154.  Capitanio JP, Mendoza, SP, Bentson KL. 2004. Personality characteristics and basal cortisol  122 concentrations in adult male rhesus macaques (Macaca mulatta). Psychoneuroendocrinology 29, 1300-1308.  Carere C, Caramaschi D, Fawcett TW. 2010. Covariation between personalities and individual differences in coping with stress: Converging evidence and hypotheses. Current Zoology, 56, 728-740.  Carere C and Locurto C. 2011. Interaction between animal personality and animal cognition. Current Zoology, 57, 491-498.  Carlstead K. and Shepherson D. 2000. Alleviating stress in zoo animals with environmental enrichment. In: GP Moberg and JA Mench (eds) The Biology of Animal Stress. New York: CABI publishing 337-354.  Carlstead K, Mellen J, Kleiman DG. 1999. Black rhinoceros (Diceros bicornis) in U.S. zoos: Individual behaviour profiles and their relationship to breeding success. Zoo Biology, 18, 17-34.  Caro T. 1999. The behaviour-conservation interface, Trends in Ecology and Evolution 14, 366-369.  Caro T. 2007. Behaviour and conservation: a bridge too far? Trends in Ecology & Evolution, 22, 394-400.  Carroll JP. 1990. Winter and spring survival of radio-tagged gray partridge in North Dakota. Journal of Wildlife Management, 54, 657-662.  Carson, Rachel. 2002. [1st. Pub. Houghton Mifflin, 1962]. Silent Spring. Mariner Books.  Caryl FM, Quine CP, KJ Park. 2012. Martens in the matrix: the importance of nonforested habitats for forest carnivores in fragmented landscapes. Journal of Mammalogy, 93, 464-474.  Casper RM. 2009. Guidelines for the instrumentation of wild birds and mammals. Animal Behaviour, 78, 1477-1483.  Chamove AS, Eysenck HJ, Harlow HF. 1972. Personality in monkeys: Factor analyses of rhesus social behavior. Quarterly Journal of Experimental Psychology, 24, 496-504.  Clemmons JR and Buchholz R. 1997. Behavioral Approaches to Conservation in the Wild, Cambridge University Press.   Cockrem JF, Silverin B. 2002. Variation within and between birds in corticosterone responses of Great Tits (Parus major). General and Comparative Endocrinology, 125, 197-206.  Cohen S and Rodriguez MS. 1995. Pathways linking affective disturbances and physical disorders. Health Psychology14, 374-380.   123 Cole MA, Kim PJ, Kalman BA, Spencer RL. 2000. Dexamethasone suppression of corticosteroid secretion: evaluation of the site of action by receptor measures and functional studies. Psychoneuroendocrinology 25, 151-167.  Cook CJ, Mellor DJ, Harris PJ, Ingram JR, Matthews LR. 2000. Hands-on and hands-off measurement of stress. In: GP Moberg and JA Mench (eds), The Biology of Animal Stress. CABI Publishing, New York, New York, USA, 123-146.  Cooper LD and Randall JA. 2007. Seasonal changes in home ranges of the giant kangaroo rat (Dipodomys ingens): a study of flexible social structure. Journal of Mammalogy, 88, 1000-1008.   Coren S. 1998. Why we love the dogs we do: How to find the dog that matches your personality. New York: Free Press.  Cote JC, Fogarty S, Brodin T, Weinersmith K, Sih A. 2010. Personality-dependent dispersal in the invasive mosquitofish: group composition matters. Proceedings of Royal Society B., 278, 1670-1678.  Dawkins MS. 1990. From an animal?s point of view:Motivation, fitness, and animal welfare. Behavioural and Brian Sciences, 13, 1-9. Dehnhard M, Schreer A, Krone O, Jewgenow K, Krause M. Grossmann R. 2003. Measurement of plasma corticosterone and fecal glucocorticoid metabolites in the chicken (Gallus domesticus), the great cormorant (Phalacrocorax carbo), and the goshawk (Accipiter gentilis). General and Comparative Endocrinology, 131, 345-352. De Mendon?a PG. 1999. Impact of radio-collars on Yellow-necked Mice, Apodemus flavicollis (Mammalia, Rodentia). Mammal Review, 29, 127-134.   aDickens MJ, Earle KA, Romero LM. 2009. Initial transference of wild birds to captivity alters stress physiology. General and Comparative Endocrinology, 160, 76-832.  bDickens  MJ, Earle KA, Romero LM. 2009. Stress and translocation: alterations in the stress physiology of translocated birds. Proceedings of Royal Society B, 276, 2051-2056.  Dickens MJ, Delehanty DJ, Romero LM. 2010. Stress: an inevitable component of animal translocation. Biological Conservation, 143, 1329-1341.  Dingemanse NJ, Both C, Drent PJ, van Oers K, van Noordwijk AJ. 2002. Repeatability and heritability of exploratory behaviour in great tits from the wild. Animal Behaviour, 64, 929-938.  Dingemanse NJ, Both C, Drent PJ, Tinbergen JM. 2004. Fitness consequences of avian personalities in a fluctuating environment. Proceedings of the Royal Society of London Series B, 271, 847-852.  Dingemanse NJ, De Goede P. 2004. The relation between dominance and exploratory behaviour is context-dependent in wild great tits. Behavioural Ecology, 15, 1023-1030.  124  Dingemanse NJ and R?ale D. 2005. Natural selection and animal personality. Behaviour, 142, 1159-1184.  Dochtermann NA and Jenkins SH. 2007. Behavioural syndromes in Merriam?s kangaroo rats (Dipodomys merriami): a test of competing hypotheses. Proceedings of the Royal Society B, 274, 2343-2349.  Etches RJ. 1976. A radioimmunoassay for corticosterone and its application to the measurement of poultry. Steroids, 28, 763-773.  Ewen JG, Acevedo-Whitehouse K, Alley M, Carraro C, Sainsbury AW, Swinnerton K, Woodroffe R. 2011. Empirical consideration of parasites and health in reintroduction. In: JG Ewen, DP Armstrong, KA Parker and PJ Seddon (eds). Reintroduction Biology: integrating science and management, Chapter 9. Wiley-Blackwell, Oxford, UK.   Fischer J and Lindenmayer DB. 2000. An assessment of the published results of animal relocations. Biological Conservation, 96, 1-11.   Fokkema DS, Smit K, van der Gugten J, Koolhaas J. 1988. A coherent pattern among social behavior, blood pressure, corticosterone and catecholamine measures in individual male rats. Physiology and Behaviour, 42, 485-489.  Fox RA and Millam JR. 2010. The use of ratings and direct behavioural observation to measure temperament traits in cockatiels (Nymphicus hollandicus). Ethology, 116, 59-75.  Forkman B, Furuhaug IL, Jensen P. 1995. Personality, coping patterns, and aggression in piglets. Applied Animal Behaviour Science, 45, 31-42. Franceschini MD, Rubenstein DI, Low B, Romero LM. 2008. Fecal glucocorticoid metabolite analysis as an indicator of stress during translocation and acclimation in an endangered large mammal, the Grevy?s zebra. Animal Conservation, 11, 263-269. Fraser D. 2008. Understanding Animal Welfare: the science in its cultural context. Wiley-Blackwell, Oxford.  Fraser D. 2009. Animal behaviour, animal welfare and the scientific study of affect. Applied Animal Behaviour Science, 118, 108-117.  Fraser D. 2010. Toward a synthesis of conservation and animal welfare science. Animal Welfare 19, 121-124.  Fraser D and MacRae AM. 2011. Four types of activities that affect animals: implications for animal welfare science and animal ethics philosophy. Animal Welfare, 20, 581-590.   French JM. 1993. Assessment of donkey temperament and the influence of home environment. Applied Animal Behaviour Science, 36, 249-257.  125  Gallup GG Jr. 1968. Mirror-image stimulation. Psychology Bulletin, 70, 782-793.  Garshelis DL and Siniff DB.1983. Evaluation of radio-transmitter attachments for sea otters. Wildlife Society Bulletin, 11, 378-383.  Gelling M, Johnson PJ, Moorhouse TP, Macdonald DW. 2012. Measuring animal welfare within a reintroduction assessment of different indices of stress in water voles, Arvicola amphibious. Plos One, 7, 1-8.  Gentsch C, Lichtsteiner M, Driscoll P, Feer H. 1982. Differential hormonal and physiological responses to stress in Roman High- and Low-Avoidance rats. Physiology and Behavior, 28, 259-263.  Gerber LR, Seabloom EW, Burton RS, Reichman OJ. 2003. Translocation of an imperiled woodrat population: integrating spatial and habitat patterns. Animal Conservation, 6, 309-316. Germano JM, Bishop PJ. 2009. Suitability of amphibians and reptiles for translocation. Conservation Biology, 23, 7-15. Gessaman JA and Nagy KA. 1988. Transmitter loads affect the flight speed and metabolism of homing pigeons. Condor, 90: 662-668.  Godfrey JD and Bryant DM. 2003. Effects of radio transmitters: Review of recent radio-tracking studies. In: Williams, M. (ed) Conservation applications of measuring energy expenditure of New Zealand birds: Assessing habitat quality and costs of carrying radio transmitters. Science for Conservation 214, 83-95.  Gold KC and Maple TL. 1994. Personality assessment in the gorilla and its utility as a management tool. Zoo Biology, 13, 509-522.  Gosling SD. 1998. Personality dimensions in spotted hyenas (Crocuta crocuta). Journal of Comparative Psychology, 112, 107-118.  Gosling SD and John OP.1999. Personality dimension in nonhuman animals: a cross-species review. Current Directions in Psychological Science, 8, 69-75.  Gosling SD. 2001. From mice to men: What can we learn about personality from animal research? Psychological Bulletin, 127, 45-86.  Gosling SD. 2008. Personality in non-human animals. Social and Personality Psychology Compass, 2, 98-1002.  Gosling LM and Sutherland WJ. 2000. Behaviour and Conservation, Cambridge University Press, Cambridge.   Green C. 2005. Using artificial refuges to translocate and establish Aukland tree weta Hemideina  126 thoracica on Korapuki Island, New Zealand. Conservation Evidence, 2, 94-95.  Griffin AS, Blumstein DT, Evans CS. 2000. Training captive-bred or translocated animals to avoid predators. Conservation Biology, 14, 1317-1326.  Griffith BJ, Scott JM, Carpenter JW, Reed C. 1989. Translocation as a species conservation tool: status and strategy. Science, 245, 477-480.   Hall, CS.1936. Emotional behavior of the rat. II. The relationship between need and emotionality. Journal of Comparative Psychology, 22, 61-88.  Hall, CS. 1938. The inheritance of emotionality. Sigma Xi Quarterly, 26, 17-27.  Hall, CS. 1941. Temperament: A survey of animal studies. Psychological Bulletin, 38, 909-943.  Hammer, MP, Bice, CM, Hall, A. et al. 2013. Freshwater fish conservation in the face of critical water shortages in the southern Murray-Darling Basin, Australia. 64, 807-821.  Hargett AC. 2006. Fecal testosterone and corticosterone levels in relation to dominance in an asocial species, Merriam?s kangaroo rat (Dipodomys merriami). Department of Biology, University of Nevada, Reno, Reno, NV.  Harker MB, GB Rathbun, CA Langtimm. 1999. Beaded-chain collars: a new method to radiotag kangaroo rats for short-term studies.Wildlife Society Bulletin, 27, 314-317. Harper JM and Austad SN. 2000. Fecal glucocorticoids: a noninvasive method of measuring adrenal activity in wild and captive rodents. Physiology and Biochemistry of Zoology, 73, 12-22. Harper JM and Austad SN. 2001. Effect of capture and season fecal glucocorticoid levels in deer mice (Peromyscus maniculatus) and red-backed voles (Clerionomys gapperi). General and Comparative Endocrinology, 123, 337-344. Harris JE. 2008. Translocation of the silver-studded blue Plebejus argus to Cawston Heath, Norfolk, England. Conservation Evidence, 5, 1-5. Harris BN, Saltzman W, Jong TR, De Milnes MR. 2012. Hypothalamic - pituitary - adrenal (HPA) axis function in the California mouse (Peromyscus californicus ): Changes in baseline activity, reactivity and fecal excretion of glucocorticoids across the diurnal cycle. General and Comparative Endocrinology, 179, 436-450. Hartup BK, Olsen GH, Czekala NM. 2005. Fecal corticoid monitoring in whooping cranes (Grus americana) undergoing reintroduction. Zoo Biology, 2, 15-28.  Hebb DO. 1946. Emotions in man and animals: An analysis of the intuitive process of  recognition. Psychological Review, 53, 88-106.  Henry JP and Stephens PM. 1977. Stress, health and the social environment: sociobiological  127 approach to medicine. New York: Springer Verlag.    Hessing MJC, Hagels AM, Van Beek JA, Wiepkema PR, Schouten WPG, Krukow R. 1993. Individual behavioural characteristics in pigs. Applied Animal Behavior Science, 37, 285?295.   Highfill F, Hanbury D, Kristiansen R, Kuczaj S, Watson S. 2010. Rating vs. coding in animal personality research. Zoo Biology, 29, 509-516.  Hooge PN. 1991. The effects of radio weight and harnesses on time budgets and movements of acorn woodpeckers. Journal of Field Ornithology, 62, 230-238.   Hubbard MW, Tsao LC, Klaas EE, Kaiser M, Jackson DH.1998. Evaluation of transmitter attachment techniques on growth of wild turkey poults. Journal of Wildlife Management, 62, 1574-1578.  Hunt, KE, Trites AW, Wasser SK. 2004. Validation of a fecal glucocorticoid assay for Steller sea lions (Eumetopias jubatus). Physiology Behavior,. 80, 595-601.  Huntingford F and Turner A. 1987. Animal Conflict. Chapman & Hall: London, UK.  IBM Corp. Released 2011. IBM SPSS Statistics for Mac, Version 20.0. Armonk, NY: IBM Corp.  Inglis IR, Langton S, Forkman B, Lazarus J. 2001. An information primacy model of exploratory and foraging behaviour. Animal Behaviour, 62, 543-557.  IUCN 1987 IUCN Position statement on translocation of living organisms. IUCN/SSC Available from http://intranet.iucn.org/webfiles/doc/SSC/SSCwebsite/Policy_statements/IUCN_Position_Statement_on_Translocation_of_Living_Organisms.pdf  IUCN 1998 IUCN/SSC Guidelines for Re-introductions.  Available from http://intranet.iucn.org/webfiles/doc/SSC/SSCwebsite/Policy_statements/Reintroduction_guidelines.pdf.   IUCN 2012, IUCN/SSC Reintroduction Specialist Group and Invasive Species Specialist Group. guidelines for reintroductions and other conservation translocations. Available from http://www.issg.org/pdf/publications/translocation-Guidelines-2012.pdf  Jamieson IG and Lacy RC. 2011. Managing genetic issues in reintroduction biology. In: JG Ewen, DP Armstrong, KA Parker and PJ Seddon (eds) Reintroduction: integrating science and management, Chapter 13. Wiley-Blackwell, Oxford, UK.  Jewell Z. 2013. Effect of monitoring technique on quality of conservation science. Conservation Biology, 3, 501-508.   128 Jones WT. 1989. Dispersal distance and the range of nightly movements in Merriam?s kangaroo rats. Journal of Mammalogy, 70, 27-34.  Jones WT. 1993. The social systems of heteromyid rodents. In: HH Genoways and JH Brown (eds) Biology of the Heteromyidae. Special Publication, The American Society of Mammalogists, 575-595.  Jule KR, Leaver LA, Lea SEG. 2008. The effects of captive experience on reintroduction survival in carnivores: a review and analysis. Biological Conservation, 141, 353-363.  Kagan J, Reznick JS, Snidman N, Gibbons J, Johnson MO. 1988. Childhood derivatives of inhibition and lack of inhibition to the unfamiliar. Child Development, 59, 1580-1589. Keay JM, Singh J, Gaunt MC, Kaur T. 2006. Fecal glucocorticoids and their metabolites as indicators of stress in various mammalian species: a literature review. Journal of Zoo and Wildlife Medicine?: official publication of the American Association of Zoo Veterinarians 37, 234-44. Keller LF, Biebach I, Ewing SR, Hoeck PEA. 2011. The genetics of reintroductions: Inbreeding and genetic drift. In: JG Ewen, DP Armstrong, KA Parker and PJ Seddon (eds). Reintroduction: integrating science and management Chapter 11. Wiley-Blackwell, Oxford, UK.  Kierulff MCM, DeOliveira PP. 1996. Re-assessing the status and conservation of the golden lion tamarin (Leontopithecus rosalia) in the wild. Dodo, 32, 98-115.   Kleiman DG. 1989. Reintroduction of captive mammals for conservation. BioScience, 39, 152-161.    Kleiman DG and Mallinson JJC. 1998. Recovery and management committees for lion tamarins: partnerships in conservation planning and implementation. Conservation Biology, 12, 27-38.  Konecna M, Lhota S, Weiss A, Urb?nek T, Adamov? T, Plu?cek J. 2008. Personality in free-ranging Hanuman langur (Semnopithecus entellus) males: subjective ratings and recorded behavior. Journal of Comparative Psychology, 122, 379-389.  Koolhaas JM, Korte SM, De Boer SF, van der Vegt BJ, van Reenen CG, Hopster H, De Jong IC, Ruis MAW, Blokhuis HJ. 1999. Coping styles in animals: current status in behaviour and stress-physiology. Neuroscience Biobehaviour Review, 23, 925-935.   Koolhaas J, de Boer SF, Buwalda B, van der Vegt BJ, Carere C, Groothuis AGG. 2001. How and why coping systems vary among individuals In: DM Broom (ed) Coping with challenge: welfare in animals including animals pp197-209. Dahlem University Press.   Korte SM, Beuving G, Reusink W, Blokhuis HJ. 1997. Plasma catecholamine and corticosterone levels during manual restraint in chicks from a high and low feather pecking line of laying hens. Physiology and Behavior, 62, 437-441.  129 Lane J. 2006. Can non-invasive glucocorticoid measures be used as reliable indicators of stress in animals? ?Animal Welfare, 15, 331-342. Latch EK and Rhodes OE. 2005. The effects of gene flow and population isolation on the genetic structure of reintroduced wild turkey populations: Are genetic signatures of source populations retained? Conservation Genetics, 6, 981-997.  Law G and Tatner P. 1998. Behaviour of a captive pair of clouded leopards (Neofelis nebulosa): Introduction without injury. Animal Welfare, 7, 57-76.  Laycock G. 1966. The alien animals Nat. Hist. Press, Garden City, NY.  Leopold A. 1933. Game management Charles Scribner?s Sons, NY, NY.   Letty J, Marchandeau S, Clobert J, Aubineau J. 2000. Improving translocation success: an experimental study of anti-stress treatment and release method for wild rabbit. Animal Conservation, 3, 211-219.  Letty J, Aubineau J, Marchandeau S, Clobert J. 2003. Effect of translocation on survival in wild rabbit (Oryctolagus cuniculus). Mammal Biology, 68, 250-255.  Lester SJ, Mellor DJ, Holmes RJ, Ward RN and Stafford KJ. 1996. Behavioural and cortisol responses of lambs to castration and tailing using different methods. New Zealand Veterinary Journal, 44, 45-54.  Lima SL and Dill LM. 1990. Behavioral decisions made under the risk of predation: a review and prospectus. Canadian Journal of Zoology, 68, 619-640.   Lindsay SR. 2005. Handbook of applied dog behavior and training vol. 3. Procedures and Protocols Blackwell Publishing, 466.  Lissman HW. 1932. Die Umvelt des Kampffishes (Betta splendens Regan). Zeitschrift fur Vergleichende Physiolgie, 18, 62-111.  L?pez P and Mart?n J. 2002. Chemical rival recognition decreases aggression levels in male Iberian wall lizards, Podarcis hispanica. Behavioral Ecology and Sociobiology, 51, 461-465.   Lyles AM and May RM. 1987. Problems leaving the ark. Nature, 326, 245-246.  Mart?nez-Mota R, Valdespino C, Robelledo JAR, Palme R. 2008. Determination of fecal glucocorticoid metabolites to evaluate stress response in Alouatta pigra. International Journal of Primatology, 29, 1365-1373.  Martins TLF, Roberts ML, Giblin I, Huxham R, Evan MR. 2007. Speed of exploration and risk-taking behavior are linked to corticosterone titres in zebra finches. Hormones and Behavior, 52, 445-453.  130 Masini, CV, Sauer S, Campeau S. 2005. Ferret odor as a processive model in rats: neurochemical, behavioral and endocrine evidence. Behavior and Neuroscience, 119, 280-292. Mather JA, Anderson RC. 1993. "Personalities" of octopuses Octopus rubescens. Journal of Comparative Psychology, 107, 336-340.  Matteri RL, Carroll JA, Dyer CJ. 2000. Neuroendocrine responses to stress In: GP Moberg and JA Mench (eds) The Biology of Animal Stress: basic principles and implications for animal welfare. CABI publishing,Wallingford 43-76.  McClenaghan Jr, LR and Taylor, E. 1993. Temporal and spatial demographic patterns in Dipodomys stephensi from Riverside County, California. Journal of Mammalogy 74, 636-645.  McDougall PT, R?ale D, Sol D, Reader SM. 2006. Wildlife conservation and animal temperament: causes and consequences of evolutionary change for captive, reintroduced, and wild populations. Animal Conservation 9, 39-48.   McEwen BS and Stellar E. 1993. Stress and the individual Archives of Internal Medicine. 153, 2093-2101.  McEwen BS and Sapolsky RM. 1995. Stress and cognition. Current opinion in neurobiology, 2, 205-216.  McLaren G, Bonacic C, Rowan A. 2007. Animal welfare and conservation: measuring stress in the wild. In: Macdonald DW, Service KM (eds) Key Topics in Conservation Biology. Oxford: Blackwell Publishing, 120-133.  Mendl M. 1999. Performing under pressure: stress and cognitive function. Applied Animal Behaviour Science, 65, 221-244.  Miller B, Biggins D, Vargas A. 1994. Reintroduction of the black-footed ferret (Mustela nigripes). In: Olney, PJS, Mace, GM, Feistner, ATC (eds) Creative Conservation: Interactive Management of Wild and Captive Animals. Chapman and Hall, London, 455-464. Millspaugh JJ, Burke T, Van Dyk G, Slotow R, Washburn BE, Woods RJ. 2007. Stress response of working African elephants to transportation and safari adventures. Journal of Wildlife Management, 71, 1257-1260. Moberg GP. 1985. Biological response to stress: key to assessment of animal well-being? In: GP Moberg (ed) Animal Stress, American Physiology Society, Bethesda, MD 27-49.  Moberg GP. 2000. Biological response to stress: implications for animal welfare In: GP Moberg and JA Mench (eds), The Biology of Animal Stress:basic principles and implications for animal welfare. CABI publishing, Wallingford 1-21.  Molony SE, Dowding CV, Baker PJ, Cuthill IC, Harris S. 2006. The effect of translocation and temporary captivity on wildlife rehabilitation success: An experimental study using European  131 hedgehogs (Erinaceus europaeus). Biological Conservation, 130, 530-537.  Moncl?s R, R?del H.G, Palme R, Von Holst D, de Miguel J. 2005. Non-invasive measurement of the physiological stress response of wild rabbits to the odour of a predator. Chemoecology, 16, 25-29. Moorhouse  TP and Macdonald DW. 2005. The indirect negative impacts of radio-collaring: sex ration variation in water voles. Journal of Applied Ecology, 42, 91-98. M?stl E, Messmann S, Bagu E, Robia C, Palme R. 1999. Measurement of glucocorticoid metabolite concentrations in faeces of domestic livestock. Zentralblatt fur Veterin?rmedizin A, 46, 621-31. Nelson RJ. 2000. An Introduction to Behavioral Endocrinology 2nd Ed. Sinauer Associates, Sunderland MA.  Nichols JD and Armstrong DP. 2011. Monitoring for reintroductions. In: JG Ewen, DP Armstrong, KA Parker and PJ Seddon (eds), Reintroduction Biology: integrating science and management, Chapter 7. Wiley-Blackwell, Oxford, UK.  Nielsen L.1988. Definitions, considerations, and guidelines for translocation of wild animals. In: N Nielsen and RD Brown (eds) Translocation of wild animals. Wisconsin Humane Society, Madison WI and Caesar Kleberg Wildlife Research Institute, Kingsville TX, 12-49.  Norton BG. 1995. Caring for nature: a broader look at animal stewardship. In: BG Norton, M Hutchins, EF Stevens, TL Maple (eds). Ethics on the ark: zoos, animal welfare, and wildlife conservation. Smithsonian Institution Press, Washington D.C. 375-395. Nussear KE, Tracy CR, Medica PA, Wilson DS, Marlow RW, Corn PS. 2012. Translocation as a conservation tool for Agassiz?s desert tortoise: Survivorship, reproduction and movements. Journal of Wildlife Management, 76, 1341-1353. O?Connor BP. 2000. SPSS and SAS programs for determining the number of components using parallel analysis and Velicer?s MAP test. Behavior Research Methods, Instrumentation, and Computers, 32, 396-402. O?Farrell, MJ. 1990. Stephens? kangaroo rat: natural history, distribution, and current status. Memoirs of the Natural History Foundation of Orange County, Newport Beach, California 3,78-84.  O?Farrell MJ. 1994. Final report: Stephens? kangaroo rat translocation, landfill expansion?Skinner reservoir. Metropolitan Water District of Southern California, Los Angeles.  O?Neill L, Wilson P, De Jong A, Rochford, J. 2008. Field techniques for handling, anesthetising and fitting radio-transmitters to Eurasian otters (Lutra lutra). European Journal of Wildlife Research, 54, 681-687.   132 Orchinik M, Gasser P, Breuner C. 2002. Rapid corticosteroid actions on behavior. In: DW Pfaff, AP Arnold, AM Etgen, SE Fahrbach, RT Rubin (ed). Hormones, Brain and Behavior. Amsterdam: Academic Press, 567-600.  Palme R, Robia C, Messmann S, Mostl E.1998. Measuring faecal cortisol metabolites: a non-invasive tool to evaluate adrenocortical activity in mammals. Advanced Ethology, 33, 27. Palme R, Rettenbacher S, Touma C, El-Bahr S.M, M?stl E. 2005. Stress hormones in mammals and birds: comparative aspects regarding metabolism, excretion, and noninvasive measurement in fecal samples. Annual NY Academy of Sciences, 1040, 162-71. Parker KA, Dickens, MJ, Clarke RH, Lovegrove TG. 2011. The theory of practice of catching, holding, moving and releasing animals In: JG Ewen, DP Armstrong, KA Parker and PJ Seddon (eds), Reintroduction Biology: integrating science and management, Chapter 4. Wiley-Blackwell, Oxford, UK. Pavlov IP. 1906. The scientific investigation of the psychical faculties or processes in the higher animals. Science, 24, 613-619.  Pavlov IP. 1928. The inhibitory type of nervous systems in the dog. In: IP Pavlov and  WH Gantt (eds), Lectures on Conditioned Reflexes: Twenty-five years of objective study of the higher nervous activity (Behaviour) of Animals. New York: Liverwright Publishing Corporation, 363-369.  Pavlov IP. 1941. Lectures on conditioned reflexes. vol. 2: Conditioned Reflexes and Psychiatry. W. Horsley Gantt, Translator. New York: International Publishers.   Pawlak CR, Ho YJ, Schwarting RKW. 2008. Animal models of human psychopathology based on individual differences in novelty-seeking and anxiety. Neuroscience Biobehavioral Reviews, 32, 1544-1568.  Pereira RJG, Granzinolli MAM, de Barros FM, Duarte JMB. 2009. Influence of radiotransmitters on fecal glucocorticoid levels of free-ranging male American kestrels. Journal of Wildlife Management, 73, 772-778.  Pervin L and John OP. 1997. Personality: Theory and research (7th ed.). New York: Wiley.  Petelle, MB and Blumstein DT. 2014. A critical evaluation of subjective ratings: unacquainted observers can reliably assess personality. Current Zoology (in press).  Petrie NJ, Mellor DJ, Stafford KJ, Bruce RA and Ward RN. 1996. Cortisol response of calves to two methods of disbudding used with and without local anaesthetic. New Zealand Veterinary Journal, 44, 9-14.  Pinter-Wollman N, Isbell LA, Hart LA. 2009. Assessing translocation outcome: Comparing behavioral and physiological aspects of translocated and resident African elephants (Loxodonta africana). Biological Conservation, 142, 1116-1124.  133  Powell DM and Gartner MC. 2011. Application of personality to the management and conservation of nonhuman animals In: M. Inoue-Murayama, S Kawamura, A Weiss (eds). From Genes to Animal Behaviour, Chapter 8. Tokyo:Springer.  Preston SD. 2001. Effects of decision making in the Merriam?s kangaroo rat (Dipodomy merriami). Ph.D. Dissertation. University of California, Berkeley.  Price, MV and Endo, PR. 1989. Estimating the distribution and abundance of a cryptic species, Dipodomys stephensi (Rodentia: Heteromyidae), and implications for management. Conservation Biology 3, 293-301.  Pruitt JN, Riechert SE, Iturralde G, Vega M, Fitzpatrick BM, Aviles L. 2010. Population differences in behaviour are explained by shared within trait correlations. Journal of Evolutionary Biology, 23, 748-756.  R Core Team. 2013. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria.  Ramos A and Mormede P. 1998. Stress and emotionality: a multidimensional and genetic approach Neuroscience. Biobehaviour Review, 22, 33-57.  Randall, JA. 1993. Behavioural adaptations of desert rodents (Heteromyidae). Animal Behaviour, 45, 263-287.  Rawles K. 1997. Conservation and animal welfare In: TDJ Campbell (ed) The philosophy of the environment. Edinburgh University Press. Edinburgh, 135-55.  R?ale D and Festa Bianchet M. 2003. Predator-induced natural selection on temperament in bighorn ewes. Animal Behaviour, 65, 463-470.  R?ale D, Reader SM, Sol D, McDougall PT, Dingemanse NJ. 2007. Integrating animal temperament withing ecology and evolution. Biological Reviews, 82, 291-318.  Reynolds MH, Seavy NE, Vekasy MS, Klavitter JL, Laniawe LP. 2008.Translocation and early post-release demography of endangered Laysan teal. Animal Conservation, 11, 160-168.  Ricciardi AR and Simberloff D. 2009. Assisted colonization, good intentions and dubious risk assessment. Trends in Ecology and Evolution, 24, 476-477.  Richardson DM, Hellman JJ, McLachlan JS, Sax DF, et al. 2009. Multidimensional evaluation of managed relocation. Proceedings of the National Academy of Sciences of the United States of America, 106, 9721-9724.  Ritter WE and Benson SB. 1934. "Is the poor bird demented?" Another case of "shadow  134 boxing." Auk, 51, 169-179.   Rollin BE. 1990. The Unheeded Cry. Oxford Univ. Press, Oxford, UK.  Rosalino LM, Ferreira D, Leitao I, Santos-Reis M. 2011. Usage patterns of Mediterranean agro-foresthabitat component by wood mice Apodemus sylvaticus. Mammalian Biology, 76, 268-273. Rothschild DM, Serfass TL, Seddon WL, Hegde L, Fritz RS. 2008. Using fecal glucocorticoids to assess stress levels in captive river otters. Journal of Wildlife Management, 2, 138-142. Rothuizen J, Biewnga WJ, Mol JA.1995. Chronic glucocorticoid excess and impaired osmoregulation of vasopressin release in dogs with hepatic encephalopathy. Domesticated Animal Endocrinology, 12, 12-24. Rushen J. 2000. Some issues in the interpretation of behavioural responses to stress. In: GP Moberg GP and JA Mench (eds), The Biology of Animal Stress. CABI Publishing: Oxon/NewYork, UK/USA, 23-42. Rushen J, de Passill? A, von Keyserlingk MAG, Weary DM. 2008. The Welfare of Cattle Springer.  Sainsbury AW, Vaughan-Higgins RJ. 2012. Analyzing disease risks associated with translocations. Conservation Biology 26, 442-452.   Sapolsky RM. 1993. Neuroendocrinology of the stress-response In: JB Becker, SM Breedlove, D Crews (eds) Behavioral endocrinology, MIT Press, Cambridge USA 287-323.   Sapolsky RM. 1994. Individual-differences and the stress-response. Seminars in Neuroscience, 6, 261-269.  Sapolsky RM. 1998. Why Zebras Don't Get Ulcers: An Updated Guide To Stress, Stress Related Diseases, and Coping. 2nd Rev Ed, W. H. Freeman. Sapolsky, RM. 2000. How do glucocorticoids influence stress responses? integrating permissive, suppressive, stimulatory, and preparative actions. Endocrinology Reviews, 21, 55-89. Sarrazin F and Barbault R. 1996. Reintroduction: challenges and lessons for basic ecology. Trends in Ecology and Evolution, 11, 474-478.  Schradin C. 2013. Intraspecific variation in social organization, genetic variation, developmental plasticity, social flexibility or entirely extrinsic factors. Philosophical Transactions of the Royal Society  B. 368, 1-10.  Schroder GD. 1979. Foraging behavior and home range utilization of the Bannertail kangaroo rat (Dipodomys spectabilis). Ecology, 60, 657-665.   135 Schulz JH, Millspaugh JJ, Washburn BE, Bermudez AJ, Tomlinson JL, Mong TW, Zhuoqiong H. 2005. Physiological effects of radiotransmitters on mourning doves. Wildlife Society Bulletin. 33, 1092-1100.  Schusterman RJ, Gentry R, Schmook J.1966, Underwater vocalization by sea lions: Social and mirror stimuli. Science, 1966, 154, 540-542. Seddon PJ. 1999. Guidelines for subspecific substitutions in wildlife restoration projects. Conservation Biology, 13, 177-184. Seddon PJ, Armstrong DP and Maloney RF 2007 Developing the science of reintroduction biology Conservation Biology 21, 303-312. Seddon PJ, Armstrong DP, Soorae P, et al. 2009. The risks of assisted colonization. Conservation Biology. 23, 788-789. Segerstrom SC and Miller GE. 2004. Psychological stress and the human immune system: A meta-analytic atudy of 30 years of inquiry. Psychology Bulletin, 130, 601-630.  Sgoifo A, de Boer SF, Haller J, Koolhaas JM. 1996. Individual differences in plasma catecholamine and corticosterone stress responses of wild-type rats: Relationship with aggression. Physiology and Behavior, 60, 1403-1407.  Sgoifo A, De Boer SF, Buwalda B, Korte-Bouws G, Tuma J. 1998. Vulnerability to arrhythmias during social stress in rats with different sympathovagal balance. American Journal of Physiology, 275, 460-466. Sheriff MJ, Krebs CJ, Boonstra R. 2010. Assessing stress in animal populations: Do fecal and plasma glucocorticoids tell the same story? General and Comparative Endocrinology, 166, 614-619. Shier DM. 2006. Effect of family support on the success of translocated black-tailed praire dogs. Conservation Biology 20, 1780-1790.   Shier DM. 2009. Behavioral ecology and translocation of the endangered Stephens? kangaroo rat (Dipodomys stephensi). Report to U.S.F.W.S. for the 2008 calendar year. San Diego Zoo Institute for Conservation Research, Escondido, California.  Shier DM. 2010. Behavioral ecology and translocation of the endangered Stephens? kangaroo rat (Dipodomys stephensi). Report to U.S.F.W.S. for the 2009 calendar year. San Diego Zoo Institute for Conservation Research, Escondido, California.  Shier DM. 2011. Behavioral ecology and translocation of the endangered Stephens? kangaroo rat (Dipodomys stephensi). Report to U.S.F.W.S. for the 2010 calendar year. San Diego Zoo Institute for Conservation Research, Escondido, California.   136 Shier DM and Randall JA. 2007. The use of different signaling modalities to communicate status by Heermann?s kagaroo rats (Dipodomys heermanni). Behavioral Ecology and Sociobiology, 61, 1023-1032. Shier DM and Swaisgood RR. 2012. Fitness costs of neighborhood disruption in translocations of a solitary mammal. Conservation Biology, 26, 116-23.  Short J, Bradshaw S, Giles J, Prince R and Wilson G. 1992. Reintroduction of macropods (Marsupialia: Macropodoidea) in Australia: a review. Biological Conservation 62, 189-204.  Sih A, Bell AM, Johnson JC. 2004. Behavioural syndromes: An ecological and evolutionary overview. Trends in Ecology and Evolution, 19, 372-378.   Sinn DL, Gosling SD, Moltschaniwskj NA. 2008. Development of shy/bold behavior in squid: context-specific phenotypes associated with developmental plasticity. Animal Behaviour, 75, 433-442.  Spencer W. 2003. Ramona airport improvement project: Stephens? kangaroo rat annual monitoring report for the years 2001-2002. County of San Diego Department of Public Works, San Diego, California.  Smith KG and Clark JD. 1994. Black bears in Arkansas: characteristics of a successful translocation. Journal of Mammalogy, 75, 309-320.  Smith BR and Blumstein DT. 2008. Fitness consequences of personality: a meta-analysis. Behavioral Ecology, 19, 448-455.   Soul? ME. 1985. What is conservation biology. Bioscience, 35, 727-734.  Stamps JA and Krishnan VV. 1999. A learning-based model for territory establishment. Quarterly Review of Biology, 74, 291-318.  Stamps JA and Groothuis TGG. 2010. The development of animal personality: relevance, concepts and perspectives. Biological Reviews, 85, 301-325.   Stamps JA and Swaisgood RR. 2007. Someplace like home: experience, habitat selection and conservation biology. In: RR Swaisgood (ed) Animal Behaviour, Conservation and Enrichment. Special Issue, Applied Animal Behaviour Science 102, 392-409.S   StataCorp. 2007. Stata Statistical Software: Release 10. College Station, TX: StataCorp LP.  Steimer T, la Fleur S, Schulz PE. 1997. Neuroendocrine correlates of emotional reactivity and coping male rats from roman high (RHA/Verh)-and low (RLA/Verh)-avoidance lines. Behavior Genetics, 27, 503-512.  Stevenson-Hinde, J, and Zunz, M 1978 Subjective assessment of individual rhesus monkeys  137 Primates 19, 473-482.  Stevenson-Hinde J, Stillwell-Barnes R, Zunz M. 1980. Individual differences in young rhesus monkeys: Consistency and change. Primates, 21, 498-509.  St?we M, Rosivall B, Drent PJ, M?stl E. 2010. Selection for fast and slow exploration affects baseline and stress-induced corticosterone excretion in great tit nestlings Parus major. Hormones and Behavior, 58, 864-871.  Suedkamp Wells KM, Washburn BE, Millspaugh JJ, Ryan MR, Hubbard MW. 2003. Effects of radio-transmitters on fecal glucocorticoid levels in captive dickcissels. The Condor, 105, 805-810.  Sutherland WJ. 1998. The importance of behavioural studies in conservation biology. Animal Behaviour, 56, 801-809.  Svendsen GE, Armitage KB. 1973. Mirror-image stimulation applied to behavioral studies. Ecology, 54, 623-627.  Swaisgood RR. 2010. The conservation-welfare nexus in reintroduction programmes: A role for sensory ecology. Animal Welfare, 19, 125-137. Teixeira C, Deazevedo C, Mendl M, Cipreste C, Young R. 2007. Revisiting translocation and reintroduction programmes: The importance of considering stress. Animal Behaviour, 73, 1-13. Temeles EJ. 1994. The role of neighbors in territorial systems: when are they ?dear enemies?? Animal Behaviour 47, 339-350.   Terio, KA, Citin SB, Brown JL. 1999. Fecal cortisol metabolite analysis for noninvasive monitoring of adrenocortical function in the cheetah (Acinonyx jubatus). Journal of  Zoo Wildlife Medicine, 30, 484-491.  Thomas JR Jr. 1975. Distribution, population densities, and home range requirements of the Stephens' kangaroo rat (Dipodomys stephensi). M.S. Thesis, California State Polytechnic Univ., Ponoma.   Thompson T and Sturm T. 1965. Classical conditioning of aggressive display in Siamese fighting fish. Journal of the Experimental Analysis of Behavior, 8, 397-403.  Tinbergen N. 1951.The study of instint. London. Oxford, University Press.  Touma C, Sachser N, Mostl E, Palme R. 2003. Effects of sex and time of day on metabolism and excretion of corticosterone in urine and feces of mice. General and Comparative Endocrinology, 130, 267-278. Touma C, Palme R, Sachser N. 2004. Analyzing corticosterone metabolites in fecal samples of  138 mice: a noninvasive technique to monitor stress hormones. Hormones and Behavior, 45, 10-22. Touma C and Palme R. 2005. Measuring fecal glucocorticoid metabolites in mammals and birds: the importance of validation. Annual NY Academy of Science, 1046, 54-74. Traue HC and Deighton RM. 2000. Emotional Inhibition In: G Fink (ed), Encyclopedia of Stress, Academic Press, 32-38.  Tsigos C and Chrousos GP. 2002. Hypothalmic-pituitary-adrenal axis, neuroendocrine factors and stress. Journal of Psychosomatic Research, 53, 865-871.  USFWS (U.S. Fish and Wildlife Service). 1997. Recovery plan for the Stephens? kangaroo rat (Dipdomys stephensi). USFWS, Portland, Oregon. 1-71.  Vainikka A, Rantala MJ, Niemal? P, Hirvonen H, Kortet R. 2011. Boldness as a consistent personality trait in the noble crayfish, Astacus astacus. Acta Ethologica, 14, 17-25.  Vanjonack WJ, Scott M, Yousef MK, Johnson HD. 1975. Corticosterone plasma levels in desert rodents. Comparative Biochemistry and Physiology, 51, 17-20. Vantmonfort D, Roo G, Brugggeman V, Rombauts L, Bergham LR, Verhoeven G, Decuypere E. 1997. Ovarian and extraovarian sources of immunoreactive inhibition in the chicken: Effects of dexamethasone. General and Comparative Endocrinology, 105, 333-343. Verbeek MEM, Drent PJ, Wiepkema PR. 1994. Consistent individual differences in early exploratory behaviour of male great tits. Animal Behaviour ,48, 1113-1121.   Verbeek MEM, Boon AK, Drent PJ. 1996. Exploration, aggressive behaviour and dominance in pair-wise confrontations of juvenile male great tits. Behaviour, 133, 945-963.  Viljoen JJ, Ganswindt A, du Toit JT, Langbauer WR Jr. 2008. Translocation stress and faecal glucocorticoid metabolite levels in free-ranging African savanna elephants. South African Journal of Wildlife Research, 38, 146-152.  Vitt P, Havens K, Hoegh-Guldberg O. 2009. Assisted migration: part of an integrated conservation strategy. Trends in Ecology and Evolution, 24, 473-474.  Vucetich JA and Nelson MP. 2007. What are 60 warblers worth? Killing in the name of conservation. Oikos, 11, 1267-1278.  Vucetich, JA and Waite TA. 2000. Is one migration per generation sufficient for the genetic management of fluctuating populations? Animal Conservation, 3, 261-266.  Waas JR, Ingram JR and Matthews LR 1999 Real-time physiological responses of red deer to translocations Journal of Wildlife Management 63, 1152-1162   139 Walker KA, Horning M, Mellish, J-AE, Weary D. 2011. The effects of two analgesic regimes on behavior after abdominal surgery in Stellar sea lions. The Veterinary Journal, 190, 160-164.  Warburton B, Poutu N, Peters D, Waddington P. 2008. Traps for killing stoats (Mustela erminea): improving welfare performance. Animal welfare, 17, 111-116.  Wasser SK, Bevis K, King G, Hanson E. 1997. Noninvasive physiological measures of disturbance in the Northern spotted owl. Conservation Biology, 11, 1019-1022.  Watters JV and Meehan CL. 2007. Different strokes: Can managing behavioral types increase post-release success? In: RR Swaisgood (ed) Animal Behaviour, Conservation and Enrichment. Special Issue, Applied Animal Behaviour Science, 102, 364-379.   Wauters LA, Preatoni DG, Molinari A, Tosi G. 2007. Radio-tracking squirrels: performance of home range density and linkage estimators with small range and sample size. Ecological Modelling, 202, 333-344.  Webster AB and Brooks RJ.1980. Effects of radiotransmitters on the meadow vole, Microtus pennsylvanicus. Canadian Journal of Zoology, 58, 997-1001.  Weiss J. 1972. Psychological factors in stress and disease. Scientific American, 226, 104-113  Weiss A, King JE, Perkins L. 2006. Personality and subjective well-being in orangutans (Pongo pygmaeus and Pongo abelii). Journal of Personality and Social Psychology, 90, 501-511.  Wells, KMS, Washburn BE, Millspaugh JJ, Ryan MR, Hubbard MW. 2003. Effects of radio-transmitters on fecal glucocorticoid levels in captive dickcissels. Condor, 105, 805-810.  Wemesfelder F, Hunter TEA, Mendl MT, Lawrence AB. 2001. Assessing the ?whole animal?: A free choice profiling approach. Animal Behaviour, 62, 209-220.  Wielebnowski NC, Fletchall N, Carlstead K, Busso JM, Brown JL. 2002. Noninvasive assessment of adrenal activity associated with husbandry and behavioural factors in the North American Clouded Leopard population. Zoo Biology, 21, 77-98.  Wilckens T. 1995. Glucocorticoids and immune function: Physiological relevance and pathogenic potential of hormonal dysfunction. Trends in Pharmacological Science, 16, 193.  Williams, J., 1999. Effects of conspecific and predator odors on defensive behavior, analgesia and spatial working memory. The Psychological Record. 49, 493?536.  Wilson DS.1998. Adaptive individual differences within single populations. Philosophical Transactions of the Royal Society of London B 353, 199-205.  Wilson, CD, Arnott G, Reid N, et al. 2011. The pitfall with PIT tags: marking freshwater bivalves for translocation induces short-term behavioural costs. Animal Behaviour, 81, 341-346.  140  Wingfield JC, Hunt K, Breuner C, Dunlap K, Fowler GS, Freed L and Lepson J. 1997. Environmental stress, field endocrinology, and conservation biology In: Behavioral Approaches to Conservation in the Wild Cambridge Univ. Press, UK 95-131.  Wingfield JC and Romero LM. 2000. Adrenocortical responses to stress and their modulation in free-living vertebrates. In:  BMcEwen (ed) Coping With the Environment. New York: Oxford University Press, 211-234.  Wisely SM, McDonald DB, Buskirk SW. 2003. Evaluation of the species survival plan and captive breeding program for the black-footed ferret. Zoo Biology, 22, 287-298.   Wisely SM, Santymire RM, Livieri TM, Mueting SA and Howard J. 2008. Genotypic and phenotypic consequences of reintroduction history in the black-footed ferret (Mustela nigripes). Conservation Genetics, 9, 389-399.  Witherington BE. 1997. The problem of photopollution for sea turtles and other nocturnal animals In: Behavioral Approaches to Conservation in the Wild Cambridge University Press, UK.  Wolf M, van Doorn GS, Leimar O, Weissing FJ. 2007. Life-history trade-offs favour the evolution of animal personalities. Nature, 447, 581-585.  Wu W, Yamaura T, Murakami K, Murata J, Matsumoto K, Watanabe H, Saik I. 2000. Social isolation stress enhanced liver metastasis of murine colon 26-L5 carcinoma cells by suppressing immune responses in mice. Life Sciences, 66, 1827-1838.  Yerkes RM and Yerkes AW. 1917. Individuality, temperament, and genius in animals Am Mus J. http://www.naturalhistorymag.com.  Yerkes RM. 1939. The life history and personality of the chimpanzee. American Naturalist, 73, 97-112.               

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0165863/manifest

Comment

Related Items