Wild Mammals in Captivity: Principles and Techniques for Zoo ...

35 downloads 104 Views 598KB Size Report
mans only to the point where they have minimal fear of hu- mans (tameness) but still .... health and “happiness” on observed behavior (Wolf and Tym- itz ). ... field Zoo (Brookfield [Chicago], Illinois) visitors in front of. exhibits—an ...
25 The Importance of Maintaining Natural Behaviors in Captive Mammals M. Elsbeth McPhee and Kathy Carlstead INTRODUCTION Behavior, like morphology and physiology, evolves in complex environments to increase an individual’s survival and reproductive success in its native habitat. Captive mammals, however, are living in environments widely different from that in which they evolved. In response, they adjust their behavior to cope with their environment, potentially resulting in genetic and phenotypic divergence between captive and wild populations (Darwin ; Price , ; Lickliter and Ness ; McPhee a, b, ). These responses can occur on  levels. First, an individual can change its behavior to meet an immediate specific need, e.g. conforming to feeding schedules or conspecific groupings. Second, growing up in a captive environment that is more restrictive than the wild can alter how an animal learns and change how it responds to future events. These changes occur within an individual, but build as the animal develops. The third level of response comprises many individual changes, but is expressed across a population. Within a captive population, certain behaviors will confer greater survivorship on the individuals who express them, e.g. greater tolerance of loud noises. These behaviors will be passed on genetically from generation to generation, resulting in a distribution of traits within the captive population that is distinct from distributions observed in wild populations. Maintaining natural behaviors in captive-bred mammals is a top priority for zoo biologists, because all  levels of change may compromise ex situ and in situ conservation efforts for endangered mammal species. For an individual animal, the presence of normal, species-specific behaviors, similar to those observed in the wild, is one potential indicator that its needs are being met, its captive environment is optimal, and it has good health and well-being. In exhibited animals, natural behavior is also a signal to the zoo visitor that the animal is a viable representative of its wild counterparts. Visitors who witness captive animals displaying abnormal behaviors are more likely to perceive those animals as “unhappy” (McPhee

et al. , unpublished data), and increased aberrant behaviors can detract from the educational message of the exhibit (Altman ). Such negative experiences with captive animals can potentially cause visitors to reject the concept that zoos are authorities in the preservation of that species and of biodiversity in general. In addition, formation of natural species-specific behaviors during growth and development is necessary for successful reproduction. Reproductive behaviors, such as mate choice, courtship, mating, and rearing of offspring, are significantly influenced by the captive environment (see Swaisgood and Schulte, chap. , this volume; Thompson, Baker, and Baker, chap. , this volume). For many mammal species, development of natural reproductive behaviors requires normal learning experiences during early development and appropriate socialization throughout development. Therefore, to maintain sustainable captive populations, the effects of captive conditions on the development of behavior must be given serious consideration in all captive breeding programs (Kleiman ; Carlstead and Shepherdson ). Finally, for captive populations to represent their wild counterparts accurately, overall trait distributions must be similar. Captive populations are exposed to selective pressures that, over generations, shape behaviors adaptive to the captive environment. These pressures alter behavioral expression and trait distribution within a population, resulting in captive populations that are behaviorally and morphologically distinct from wild populations. The captive population may thus evolve toward one lacking individuals able to survive in a wild environment, severely limiting the ability of captive populations to contribute to the recovery of wild populations. In this chapter we will discuss the importance of addressing all  levels of behavioral change within the contexts of individual animal well-being, development of natural reproductive behavior, and behavioral diversity of captive populations, and how these relate to in situ and ex situ conservation efforts. 303

304

the imp orta nce of maintaining natural behaviors in captive mammals

ANIMAL WELL-BEING An ethical and operational imperative of every zoo is to optimize an animal’s well-being (see Kagan and Veasey, chap. , this volume). Psychological well-being is an individual-level phenomenon, for it depends on how an animal perceives its own ability to respond to changes in its environment and react to its constraints. Dawkins (, ) suggests that determining whether the welfare of animals is good or bad requires answering  questions: is the animal healthy, and does it have what it wants? Poor physical health is an obvious sign of poor well-being, but other measures of health are less straightforward, such as reduced food intake or depressed immune function. Behavioral scientists are increasingly interested in studying stress responses of captive animals in order to link animal well-being to animal health. IS THE ANIMAL HEALTHY? STRESS ASSESSMENT To keep an animal physiologically and behaviorally healthy, the demands of stress must be kept within a tolerable range. Chronic stress is an undisputed cause of poor welfare in captive mammals (Broom and Johnson ). Prolonged periods of high levels of hypothalamic pituitary-adrenal activity in response to repeated or chronically present stressors may have costly biological consequences, such as immunosuppression and disease, atrophy of tissues, decreased reproductive function, or maladaptive behavior (Engel ; Henry ; Bioni and Zannino ; Blecha ; Elsasser et al. ; Whay et al. ). Stress can be good or bad for an animal depending on how long the stress lasts, how intensive it is, and how many options the animal has for responding to its situation (Ladewig ; McEwan ; Wielebnowski ). There are few zoo studies, however, that relate actual measurements of stress to consequences for animal health; stress is often presumed rather than measured, based on coincidental occurrence of precipitating events. For example, Cociu et al. () reported that Siberian tigers, Panthera tigris altaica, at the Bucharest zoo developed gastroenteritis due to a failure to adapt to unfamiliar quarters. Also, a persistent high noise level lasting several months caused by repairs in an adjacent courtyard was thought to induce gastroenteritis in some of the tigers. There is more recent direct evidence for a correlation between elevated corticoids and biological costs among some zoo-housed species. Carlstead and Brown () provide evidence that mortality in black rhinoceroses, Diceros bicornis, and reproductive failure in white rhinoceroses, Ceratotherium simum, are associated with high variability in the secretion of stress hormones (glucocorticoids) from the adrenal cortex. They interpreted high variability in fecal corticoids over a one-year collection period as indicating an inability of some rhinoceroses to adapt or maintain homeostatic levels of adrenal activity. There is also an association of high average glucocorticoid concentrations with hair loss in polar bears, Ursus maritimus (Shepherdson, personal communication), and with conspecific trauma, neoplasia, renal disease, and adrenal cortical histopathology in clouded leopards, Neofelis nebulosa (Terio and Wielebnowski, personal communication). To address the causes and reduce the deleterious effects of

stress on well-being and health, measurement of stress in zoo animals is necessary. How do we assess stress? A definition of chronic, or “bad,” stress is “when environmental demands tax or exceed the adaptive capacity of an organism, resulting in psychological and biological changes that may place a person or animal at risk for disease” (Cohen, Kessler, and Underwood Gordon , ). Thus, to measure stress in zoo animals we must assess and integrate at least  factors: () environmental conditions and changes, () physiological and behavioral responses to these changes, and () biological consequences to health, reproduction, and disease processes. In zoos today, stress assessment is a rapidly growing field that necessitates the integration of these biological data collected sequentially on individual animals, for individuals perceive stressors differently as a function of their age, sex, background, personality, or reproductive condition (Benus, Koolhaas, and van Oortmerssen ; Suomi ; Mason, Mendoza, and Moberg ; Cavigelli and McClintock ). The environmental and social events that cause stress in captive mammals vary depending on the species. Morgan and Tromborg () provide an admirable review of factors that elicit stress responses in captive mammals. These include sound and sound pressure, olfactory stimulation from predators and chemicals, and space restriction (see also Kagan and Veasey, chap. , this volume). Unstable social groups (DeVries, Lasper, and Detillion ) or unnaturally high group densities, denied concealment, removal of scent marks, induced feeding competition, or forced proximity to zoo visitors also cause problems (Glatston et al. ; Hosey and Druck ; Chamove, Hosey, and Schaetzel ; Thompson ). In a multi-institutional study of  clouded leopards, Wielebnowski et al. () found that public display, proximity of predators, frequent changing of keepers, and lack of enclosure height are environmental factors associated with high levels of fecal corticoids and aberrant behavior. Similarly, black rhinoceroses in enclosures with a large percentage of the perimeter exposed to public viewing had higher fecal glucocorticoids than those in enclosures with more restricted viewing (Carlstead and Brown ). Responses to stressors may include various combinations of protective or defensive behaviors and neuroendocrine responses, depending on the stressor (Matteri, Carroll, and Dyer ; Moberg ). Physiological stress responses may occur transiently in response to normal situations such as courtship, copulation, or routine husbandry events, or they may be more sustained or variable as a result of more difficult challenges. A primary avenue for measuring stress, therefore, is to pair changes in behavior with changes in physiological measures, and then examine the precipitating environmental and/or social events. Measurement of glucocorticoids in feces and urine has become the primary means of studying stress responses in zoo animals and wildlife, because the collection method is noninvasive and represents a pooled sample of corticoid output over a period of several hours (Whitten, Brockman, and Stavisky ; Möstl and Palme ; Hodges, Brown, and Heistermann, chap. , this volume). In contrast, sampling blood for corticoid measurement needs to be conducted more frequently under rigidly controlled conditions in order to control for the naturally pulsatile excretion of corticoids and for circadian rhythms. This makes measur-

m. elsbeth mcphee and kathy carlstead

ing basal levels and changes associated with environmental stressors more difficult. Behavior is the animal’s “first line of defense” in response to environmental change; it is what animals do to interact with, respond to, and control their environment (Mench ). Therefore, behavioral changes must be monitored when assessing stress in zoo animals. The increased occurrence of repetitive pacing, aggression, excessive sleep or inactivity, or fear behaviors can be indicative of stress. In a group of socially housed captive gorillas, Gorilla gorilla gorilla, periods of social instability were indicated by elevated glucocorticoids, aggressive displays, and fighting (Peel et al. ). Wielebnowski et al. () found that elevated glucocorticoids in clouded leopards correlated with fur plucking, extensive pacing, and hiding behavior. A singly housed male orangutan, Pongo spp., had higher glucocorticoid levels on days when the keepers recorded his temperament as being “upset” and he refused to shift between enclosures (Carlstead , unpublished data). In a zoo-housed giant panda, Ailuropoda melanoleuca, there was a positive association between glucocorticoids, locomotion, and door-directed scratching behavior and loud noise levels (Owen et al. ). Conversely, the frequent occurrence of positive and/or natural behaviors may indicate low levels of stress and, presumably, good welfare. Leopard cats, Felis bengalensis, living in virtually barren cages had chronically elevated glucocorticoids and high levels of stereotypic pacing that decreased dramatically when they were given the opportunity to hide or conceal themselves in newly provided complex furnishings and vegetation. They also increased the amount of time spent exploring their cages (Carlstead, Brown, and Seidensticker ). Falk (personal communication) reported that, at one zoo, opening the shift doors for polar bears and giving them the choice to go inside or outside greatly reduced stereotypic pacing. Individual white rhinoceroses that keepers assess as being most adapted to their captive environment based on how “friendly” they are toward the keeper have lower mean glucocorticoids than conspecifics who seem less relaxed around keepers (Carlstead and Brown ). DOES THE ANIMAL HAVE WHAT IT WANTS? In zoos, absence of stress and abnormal behaviors are, by themselves, insufficient criteria to ensure animal well-being. Animals are motivated to perform behaviors (see Shepherdson, chap. , this volume), and animal welfare science seeks to identify which behaviors the animal wants/needs to do most (e.g. hunting, foraging, swimming). Captive mammals should also be active at levels similar to those in the wild (Kagan and Veasey, chap. , this volume). Thus, they need to be able to carry out highly motivated behaviors, especially those that result in a level of behavioral proficiency that supports the goals of the program for which they were raised (e.g. reintroduction, social living, captive breeding, program animal docile to humans, or exhibit animal displaying normal behavior). Motivations change with the season and reproductive condition; e.g. an American black bear, Ursus americanus, had a seasonal pattern of stereotypic pacing where the pacing changed temporally and spatially depending on whether it was the mating season or the prewintering season. Carl-

305

stead and Seidensticker () interpreted this result to indicate a change in what the bear wanted—mating or foraging opportunities. Two of the most common approaches to determine whether an animal has what it wants are () using environmental modifications and enrichment to reduce experimentally negative behaviors such as stereotypy, aggression, and excessive inactivity, and to increase activity levels and behavioral diversity, and () choice tests through which animals may indicate which behaviors they are most highly motivated to perform. One explanation for negative behaviors such as stereotypies is that appetitive motivation to seek something missing in the environment causes the animal to channel motivation toward repetitive locomotion (reviewed in Mason and Latham ; see Shepherdson, chap. , this volume). In other words, the animal cannot find an outlet for its desired behavior. As an example, feeding in the wild constitutes a large portion of some species’ activity budget. Foraging and hunting are processes that include searching, acquiring (harvesting or capturing and killing), and consuming food items (Lindburg ). Captive animals, however, rarely have opportunities to do anything but consume, and often that is a compromised experience, as food is reduced from a complex (e.g. an intact prey item) to an overly simple form (e.g. preprocessed meat). In such cases, enrichment in the form of provisioning with whole carcasses increases handling time of food and reduces time spent in stereotypic pacing (ibid.; McPhee ). Choice tests can suggest what the animal “wants” to be doing for pleasure, comfort, or satisfaction and have demonstrated that animals usually prefer challenges and engagement over passive rewards (Mench ). Animals of many species prefer to work for food, such as running a maze or pushing a lever, as opposed to getting food for free. Likewise, they prefer to explore novel environments and objects. Modifications to animals’ environments or changes in feeding methods frequently reduce negative behaviors and increase activity and behavioral diversity (for a review, see Swaisgood and Shepherdson  and Shepherdson, chap. , this volume). Wiedenmayer () hypothesized that stereotypic digging develops in captive Mongolian gerbils, Meriones unguiculatus, because the stimuli that normally cause the animal to cease digging were lacking in the captive environment. He kept young gerbils on a wood-chip substrate with access to an artificial burrow and found less digging behavior than in gerbils that were able to dig in a dry, sandy substrate which was incapable of supporting a burrow. In many captive species, stereotypies occur mainly before feeding time, when the animal is motivated to perform food acquisition behaviors such as foraging or hunting. Winkelstraeter () describes a female ocelot, Leopardus pardalis, that ran in a circular path before feeding. Similarly, Geoffroy’s cats, Felis geoffroyi, paced for  to  hours before feeding time (Carlstead ). Stereotypic behaviors in large felids decrease with the provisioning of intact prey items, while foodhandling time and other consumptive behaviors increase (e.g. crouch, stalk, pounce, leap, swipe, bite, hold, eat) (McPhee ; Bashaw et al. ). An American black bear paced less and explored/foraged more before feeding time when food was scattered throughout its enclosure (Carlstead, Seidensticker, and Baldwin ). Gould and Bres () had some

306

the imp orta nce of maintaining natural behaviors in captive mammals

success in reducing regurgitation behavior in captive gorillas by feeding browse, which increased time spent handling and ingesting food. Such studies can be interpreted as meaning that animals benefit from the opportunity to “work” for their food (Markowitz ). Another fundamental question is whether an animal has as much space as it wants or needs in a zoo enclosure. For carnivores and other wide-ranging species, home range size predicts the level of pacing in captivity (Clubb and Mason ). With primates, generally the smaller the cage, the more some individuals perform stereotypies (Paulk, Dienske, and Ribbens ; Prescott and Buchanan-Smith ). By increasing the size of the area available to an animal, the behavior can sometimes be eliminated or altered (Draper and Bernstein ; Clarke, Juno, and Maple ). Exactly how much space an animal needs is unclear, particularly because the critical factor for well-being is often the quality of the space (Berkson, Mason, and Saxon ). Polar bears have the largest home ranges of all carnivores in the wild and presumably need the most space in zoos—more space, in fact, than can be provided by any zoo. Shepherdson’s recent multi-institutional study of polar bear stereotypy found that enrichment and training is highly negatively correlated with time spent in stereotypy, implying that giving bears something to do continually can be compensation for lack of space (Shepherdson, personal communication). Perkins () and Wilson () found that cage size was not as important for stimulating higher levels of activity in groups of zoo-housed great apes (gorillas and orangutans) as was the number and type of furnishings. Odberg () compared the behavior of bank voles, Myodes glareolus, in small, rich environments and in large, sparse ones, and found less stereotypic jumping in the former. Similarly, spatial confinement was not a causal factor of stereotypies in Mongolian gerbils (Wiedenmayer ). Such evidence lends credence to Hediger’s () statements that the quality of a confined animal’s space is more important than the quantity. Choice tests are common in animal welfare research of intensive farming practices to determine how much effort animals are willing to expend to achieve access to opportunities for preferred activities. Dawkins () has used a consumer demand model derived from economic theory that requires animals to work to acquire commodities. For example, Mason, Cooper, and Clarebrough () found that mink, Mustela vison, expend more energy pressing a weighted door to acquire access to a swimming pool than to gain access to toys, novel objects, or alternative nest sites. When deprived of pool access they exhibited corticoids elevated to a level similar to the corticoid response when they are food deprived. The authors concluded that farmed mink are still highly motivated to perform aquatic activities despite being bred in captivity without access to pools of water for  generations. Choice test experiments are not commonly used with zoo animals, because standardizing environments sufficiently to offer  equal choices is difficult. However, studies of space utilization that divide enclosures into grids can assess animal preferences to some degree, particularly when behavior is also assessed. For example, Mallapur, Qureshi, and Chellam () found that almost all  leopards, Panthera par-

dus, in Indian zoos used the edge zone of their enclosure for stereotypic pacing and the “back” zone, farthest from visitors, for resting. Of additional interest for future research are the individual differences when using operant conditioning to train captive animals: learning and performing behaviors could be an indication of willingness to work for food rewards or social interaction with the keeper (see Mellen and MacPhee, chap. , this volume). Certainly, there are individuals who choose not to be trained, or do so only for their most highly valued rewards. BEHAVIORAL DEVELOPMENT AND REPRODUCTION Maintaining natural reproductive behaviors in captive-bred animals is vital to the establishment of self-sustaining captive populations and maintaining genetic diversity within those populations. Historically, however, many captive mammals have reproduced with difficulty or not at all (Wielebnowski ). Common problems include inability to court and choose mates, copulate successfully, or rear viable offspring, often due to a lack of normal species-specific interactions with the environment as an animal matures. As with animal welfare, such developmental problems are observed at the individual level. Interactions between the developing organism and its environment start before birth, since the hormonal state of the mother affects the uterine environment of the growing fetus. There are numerous reports of the effects of stress experienced by mothers during pregnancy on the behavior of their offspring, including increases (Ader and Blefer ) or decreases (Thompson, Watson, and Charlsworth ) in emotionality in a novel environment (open field), alterations of exploratory behavior in rats (Archer and Blackman ), and reductions in attack and threat behavior in male offspring in mice (Harvey and Chevins ). Male offspring of mother rats stressed daily in the last week of gestation showed reductions in attempted copulations and ejaculation responses as adults (Ward ). Such studies imply that prenatal stressors specific to a captive environment can cause postnatal behavioral changes that reduce reproductive viability of offspring. In most mammals, the mother-infant relationship is critical to the future development of offspring, affecting future defensive responses and reproductive behavior (Cameron et al. ). Subtle aspects of the parent-infant or juvenilepeer relationship may affect later sexual preferences and competence, and researchers speak of an extended period of socialization occurring during infant and juvenile stages (Aoki, Feldman, and Kerr ). A disturbed mother-infant relationship may deprive the young animal of specific stimulation essential for the development of normal emotional regulation, social interaction, and complex goal-directed behaviors, in particular, maternal and sexual behaviors. Deprivation of maternal licking when pups are young has been shown to affect the timing of sexual behavior patterns in male rats when grown; intromissions were more slowly paced and the rats took longer to ejaculate (Moore ). Relationships with peers are also important, especially for great apes. Maple and Hoff () found that young gorillas

m. elsbeth mcphee and kathy carlstead

maintained with little or no conspecific contact are often sexually dysfunctional as adults. Captive female chimpanzees, Pan troglodytes, are better mothers when they have social experience with nonrelated infants or other mothers with infants (Hannah and Brotman ). As an animal matures in captivity, many elements influence it that are unique to the captive environment. Close contact with humans can produce a range of behavioral characteristics not found in a wild-reared animal, depending on when it occurs during the animal’s development and how long it is sustained. The most significant effects of human contact on overall behavior likely occur as a result of contact early in life in lieu of caregiving by the natural mother (i.e. hand rearing) (see also Thompson, Baker, and Baker, chap. , this volume). In a study of western lowland gorillas, Ryan et al. () found that mother-reared females had more offspring and were more likely to become successful mothers than hand-reared females. Early socialization with humans can affect species-specific social skills in both positive and negative ways. Among ungulates with precocial young, filial imprinting, in which the young learn to follow the mother (rather than objects and individuals that do not resemble the mother), occurs within the first day or two of life (for reviews see Bateson ; Hogan and Bolhuis ). Guinea pigs, Cavia porcellus, also exhibit characteristics of filial imprinting (Sluckin ; Hess ). Removing a young animal from the mother during this sensitive period may result in following responses being elicited by human caregivers, as is commonly seen in sheep and goats, but has also been reported in the American bison, Bison bison, zebra, Equus spp., African buffalo, Syncerus caffer, mouflon, Ovis musimon, and vicuña, Vicugna vicugna (see Hediger ). Mellen and MacPhee (chap. , this volume) recommend caution in training young male ungulates, because as they mature they tend to show aggression toward humans. Ideally, captive-born mammals should be socialized with humans only to the point where they have minimal fear of humans (tameness) but still recognize them as heterospecifics. Many reproductive deficiencies derive from an inappropriate developmental environment, while others occur through lack of opportunities in captivity. In the wild, many mammals can choose their mate based on cues ranging from chemical odors to complicated courtship displays. Based on such cues, wild individuals tend to choose a mate such that inbreeding is decreased, pathogen resistance is increased, and ultimately, survival and reproductive success of offspring are maximized (Grahn, Langefors, and von Schantz ). Many species have evolved mechanisms by which they choose mates that are genetically dissimilar, which reduces the risk of inbreeding (Blouin and Blouin ) and increases the ability of offspring to resist disease. House mice, Mus musculus, prefer mates that are genetically dissimilar at the major histocompatibility complex (MHC), a suite of genes responsible for immune function, thus conferring increased immune response and disease resistance in their offspring (Penn and Potts ). Grahn, Langefors, and von Schantz () suggest that allowing greater mate choice in captive animals might increase disease resistance in captive populations through increased diversity of the MHC. Specifically, they suggest allowing cap-

307

tive males to display for females and using female reactions to guide breeding decisions. Captive mammals, however, are not usually given the opportunity to assess multiple potential mates and ultimately choose an individual suitable to them. Indeed, little is known about how mate preferences develop or are lost in captive populations. For example, species differ in whether familiarity is a good or a bad feature, and in some, the introduction of an unfamiliar individual as a mate results in high levels of aggression. Yamada and Durrant () found that clouded leopards need to be paired while still sexually immature; in pairs established later, males show extreme aggression toward females, often resulting in serious or fatal injuries. Female kangaroo rats, Dipodomys heermanni arenae, were less aggressive toward and more responsive to familiar males with whom they had prior experience (Thompson ). By contrast, cheetah, Acinonyx jubatus, and white rhinoceroses are anecdotally known to be more likely to breed with newly introduced, unfamiliar mates. BEHAVIORAL DIVERSITY Animal managers must also be aware of how small, individual changes affect behavioral trait expression at the population rather than the individual level. Selection in captivity can affect the expression of behavioral traits in multiple ways; however, the selective pressures associated with captivity are vastly different from those in the wild environments in which species have evolved (Hediger ; Price ; Frankham et al. ; Soulé ; Soulé et al. ; Price ; Seidensticker and Forthman ). Captivity can thus impose novel selective pressures—either intentionally or inadvertently (Price , ; Endler )—and, over generations, result in changes in important life history and behavioral traits that affect functional relationships between behavioral, morphological, and physiological traits (McDougall et al. ). Understanding and identifying the ultimate mechanisms behind behavioral change can be difficult, because different selective pressures do not occur in isolation of one another. One trait may experience relaxed selection, while another experiences directional. The overall expression of traits is therefore the result of complicated synergistic effects. For the purposes of this chapter, we will focus on directional and relaxed selection. Directional selection occurs when the expression of traits at one end of the distribution is favored (Endler ). In this case, mean trait expression will shift, but variance around the mean will not necessarily change (fig. .a). For example, time spent foraging is a trade-off with predator avoidance for many wild species. More time foraging means more food, but it also means higher predation risk. On the other hand, reduced foraging time might decrease probability of predation, but the animal will have fewer resources than its bolder counterparts. In the captive environment, that trade-off does not exist. Some animals that are given foraging opportunities may increase the amount of time spent foraging—and as a result have healthier offspring and higher reproductive success than individuals who continued to balance foraging and vigilance. In this case the trait foraging time has been pushed in an increasing direction.

308

the imp orta nce of maintaining natural behaviors in captive mammals A

B

Fig. 25.1. Directional and relaxed selection. The solid curve represents the original distribution of a trait; the dashed curve represents how the distribution changes due to either directional (A) or relaxed (B) selection. With directional selection, the mean shifts, but the shape of the distribution does not change; with relaxed selection, the mean remains the same, but the distribution flattens out, including more values at either extreme.

Relaxed selection can occur when captive conditions permit expression of certain behavioral traits that otherwise would have been selected against in the wild, resulting in an increase in genotypic and phenotypic trait variability (Endler ; fig. .b). Consider the time necessary to respond to a predator. In the wild, animals have limited time to detect a predator and flee. In a captive environment, however, there is typically no predation and individuals can respond immediately, or never, to perceived threats with no effect on reproductive success. In this case, relaxed selective pressures in the captive environment result in an increase in variance in response time. In one of the few investigations of population-level behavioral processes affecting captive mammals, McPhee (a, b, ) tested for directional and relaxed selection in captive-bred oldfield mice, Peromyscus polionotus subgriseus, and found that behavioral and morphological divergence from the wild population increased with generations in captivity, primarily due to relaxed selection. Trait variance increased significantly for burrow/refuge use, activity level, response time to predators, and skull shape. To our knowledge, this is the only explicit test of these hypotheses in mammals. We recommend more research in this area to identify interspecific and trait-specific patterns, as well as linkages between traits. Initially, these increases in behavioral trait variance may seem counter to the large body of literature that shows that, with generations in captivity, genetic variance decreases. Two views can potentially resolve this difference. First, phenotypes may be the primary trait on which selection acts (WestEberhard ), and minor genetic changes can result in profound phenotypic changes, which are then selected for (or not). Second, reduced genetic variance resulting from inbreeding is not natural selection—it is the reduction of genetic variance based on what genes are available. Selection then acts on the phenotypic expressions that result from in-

breeding, a process that may ultimately appear to reduce genetic variance. Artificial selection, be it directional or relaxed, can be intentionally or inadvertently imposed. The most common form of intentional artificial selection is domestication (Price ). Humans have attempted, but failed, to domesticate many species. Not all animals are amenable to domestication, because certain behavioral characteristics are more favorable for domestication than others. Easily domesticated species generally live in large, hierarchical social groups in which the males affiliate with female groups, and mating is promiscuous. Young are precocial and experience a sensitive imprinting period during development. They are also generally adapted to a wide range of environments and dietary habits rather than to highly specialized conditions (ibid.). Species targeted for domestication are subjected to strong selective pressures designed to produce a certain outcome, e.g. increased milk production in dairy cows or long, pointed snouts in rat-hunting dogs; but many behavioral changes in domesticated populations are the indirect consequence of selection for other morphological and/or physiological attributes. For example, in Belyaev’s famous selection experiments with silver fox, Vulpes vulpes, animals selected over generations exclusively for tameness also exhibited a shift from one to  annual estrous cycles (Belyaev and Trut ) as well as phenotypic traits such as floppy ears and curly tail that are typical of domesticated dog species (Trut ). Although nondomestic captive populations do not undergo such strong intentional selection, they do experience unintentional selective pressures. For example, temperament in captive mammals is likely shaped by directional selection (Arnold ; Frankham et al. ; McDougall et al. ). In captivity, docile individuals are easier to handle, transport, and medicate than their more aggressive counterparts. In a human-controlled environment, docility might translate into better survival and reproduction, which could lead to unconscious artificial selection for docility and tractability in mammals—selection that may eventually make captive populations divergent from wild populations. On the other hand, Kunzl et al. () compared behavior and physiological responses of domestic guinea pigs, Cavia aperea f. porcellus, wild-caught cavies, and captive-bred cavies, Cavia aperea, bred with no purposeful selection. There were no significant differences between the wild-caught animals and those bred in captivity for  generations, suggesting that purposeful selection for specific traits may be necessary to produce domesticated animals. CONSERVATION IMPLICATIONS Maintaining natural behaviors in captive-bred animals is vital to the success of conservation efforts that rely on those animals, such as zoo education and reintroduction of captivebred animals into their native habitat. VISITOR EXPERIENCE Maintaining good animal welfare is important for not only the individual animal, but also the zoo visitor. Zoos often justify the exhibition of exotic species by highlighting the

m. elsbeth mcphee and kathy carlstead

zoo’s and the species’ educational value. A zoo is likely the only opportunity most people have to see such animals live, and most of a visitor’s education while at a zoo comes from watching the exhibited animals (see Routman, Ogden, and Winsten, chap. , this volume). Zoo visitors often base their evaluation of an animal’s health and “happiness” on observed behavior (Wolf and Tymitz ). In general, visitors tend to be more engaged in front of an exhibit with an active animal (Altman ; Margulis, Hoyos, and Anderson ). When that activity is the display of obviously repetitive and abnormal behaviors, however, visitors perceive animals to be “unhappy” and “bored” (McPhee , unpublished data). Through nature documentaries and other media sources, visitors are coming into zoos with more background information than ever before, e.g. knowledge of how animals in the wild behave. This knowledge, however, can create false expectations. Nature programs often show carnivores capturing and killing prey. Yet, when visitors see a captive carnivore, it is often sleeping (a very natural behavior), and they may think it is “bored” and has “nothing to do” (Wolf and Tymitz ; McPhee , unpublished data). Exhibit type also influences visitor perceptions of natural behavior. McPhee et al. () surveyed over  Brookfield Zoo (Brookfield [Chicago], Illinois) visitors in front of  exhibits—an outdoor barren grotto, an outdoor vegetated grotto, an indoor immersion exhibit, and an outdoor traditional cage—and found that visitors were more likely to perceive animals observed in more naturalistic enclosures as behaving naturally and thus being “happy,” compared with animals in traditional cagelike enclosures. Perceptions of animal well-being also strongly affected the educational power of a zoo exhibit. Visitors that observe behaviorally healthy animals are more likely to walk away with an appreciation for that species’ biological significance and need for conservation. These data suggest that, ultimately, an animal’s behavior is the most powerful communicator of that individual’s health and well-being as well as its natural history and conservation value. REINTRODUCTION Though most zoo animals will spend their entire lives as captive representatives of their wild counterparts, a very small percentage of individuals are targeted for release into their native habitat. Due to changes in important life history and behavioral traits, however, individuals from an established captive population are often at a disadvantage upon reintroduction. Evaluation of reintroduction programs indicates that many deaths of reintroduced animals are due to behavioral deficiencies (Kleiman ; Yalden ; Miller, Hanebury, and Vargas ; Biggins et al. ; Britt, Katz, and Welch ). When golden lion tamarins, Leontopithecus rosalia, were first reintroduced into the coastal rain forests of Brazil, captive-born animals had deficient locomotor skills; they could not orient themselves spatially; and they were unable to recognize natural foods, nonavian predators, and dangerous nonpredaceous animals (Kleiman et al. ; Stoinski and Beck ). Similarly, in Madagascar, reintroduced blackand-white ruffed lemur, Varecia variegata variegata, failed

309

to avoid predators, find food, negotiate a complex arboreal environment, and recognize appropriate habitat (Britt, Katz, and Welch ). In , all known wild black-footed ferrets, Mustela nigripes, were brought into captivity for breeding and eventual release of offspring. The initial releases of captivebred ferrets resulted in high mortality due to predation, suggesting antipredator deficiencies (Reading et al. ). Since many of these problems stem from the fact that the animals developed and matured in a captive environment, such behavioral problems might be eliminated by training animals targeted for reintroduction for certain skills before release. Most work in this area has been done with predator response behaviors. For juvenile prairie dogs, Cynomys ludovicianus, prerelease training consisted of exposure to either a live black-footed ferret, a live prairie rattlesnake, Crotalus viridis, or a mounted red-tailed hawk, Buteo jamaicensis. All exposures were coupled with the appropriate conspecific alarm call, but there was no aversive stimulus associated with the predator presentation. This training was sufficient to enhance survival for at least the first year postrelease (Shier and Owings ). Similarly, captive-bred Siberian polecats, Mustela eversmanni, exposed to predator models (e.g. great horned owl, Bubo virginianus, and badger, Taxidea taxus) coupled with mildly aversive stimuli, exhibited heightened antipredator responses (Miller et al. ). McLean, Lundie-Jenkins, and Jarman () trained hare-wallabies, Lagorchestes hirsutus, to be cautious in the presence of new predators. Prerelease training, especially in orientation and locomotor behavior, for golden lion tamarins was not as effective as hoped in increasing survival (Kleiman ; Beck ; Beck et al. ). However, pairing captive-bred individuals with experienced wild-caught animals did increase the reintroduced animals’ survival rate (Kleiman ). Prerelease training can also be used to establish hunting and foraging behaviors. For example, red wolves, Canis rufus, were exposed to carcasses and live prey before release, and swift foxes, Vulpes velox, were given natural foods in the form of road-killed prey (ungulates and beavers) and chicks from a local hatchery (USFWS  and Scott-Brown, Herrero, and Mamo , as cited in Kleiman ). Black-footed ferrets that had prerelease conditioning with prey were more efficient at locating and killing prey than those without prey experience (Vargas and Anderson ). At the population level, understanding how selection has altered the distribution of key behavioral traits within the population may allow reintroduction biologists to compensate for the observed changes. If the distribution of a trait or traits has shifted significantly, the proportion of individuals in a population that exhibits natural behaviors will also shift. Mortality will then increase in the release population, because fewer individuals will exhibit behaviors adapted for the wild environment. If a program’s goal is to release  individuals, the question becomes how many need to be released such that  individuals will fall within the natural behavioral range? To address this question, McPhee and Silverman () developed the concept of the release ratio, a calculation that uses trait variances in release and wild populations to determine the number of release individuals needed such that the targeted number of individuals exhibits natural behaviors.

310

the imp orta nce of maintaining natural behaviors in captive mammals

For example, McPhee’s (a, b, ) behavioral and morphological measurements of oldfield mice showed that various traits, such as time to burrow and time in a refuge after having been exposed to a predator, exhibited significant increases in variance. Thus, if a representative sample of the captive population were released into the wild, there would be more individuals in the tails of trait distributions than would be seen in a wild population, resulting in a higher mortality rate than would otherwise be expected. Based on these data, McPhee and Silverman () calculated release ratios of about . for the oldfield mouse, which means that for every  mice targeted to exhibit wild-type behaviors,  individuals need to be released. CONCLUSION In this chapter we highlighted  areas in which understanding and maintaining captive mammals’ natural behavior is key to the success of propagation programs. We commonly see behavioral change in captive and captive-bred mammals, and these changes can occur on a number of levels. First, individuals adapt their behavior to captive conditions. Chronic or repeated stress due to inappropriate environmental conditions may result in poor health for a captive mammal. In addition, animals need to be able to carry out a diversity of species-appropriate behaviors. Thus, zoos and aquariums have a responsibility to promote a range of natural behaviors through appropriate exhibit design, husbandry, and enrichment programs. Second, as an individual develops and matures, behaviors emerge that are the result of interactions between the individual’s genetic makeup and its captive environment. In some cases, such changes can have strong negative effects on an individual’s ability to reproduce successfully, thus affecting the probability of maintaining a sustainable ex situ population. Finally, captivity can exert directional or relaxed selective pressures on behaviors that will affect the frequencies of those behaviors in future generations. Therefore, as individual behavior shifts, the distribution of traits within a population will also shift over generations. One of the primary missions of zoos is conservation education through animal exhibitry, outreach programs, on-site tours and courses, and exhibit graphics (Hutchins, Willis, and Wiese ; Routman, Ogden, and Winsten, chap. , this volume). Behaviorally healthy captive-bred animals enhance in situ conservation primarily by educating the public about and instilling an appreciation for the importance of conservation. For many visitors, the zoo setting will be their only opportunity to see most species of wildlife. What that visitor takes away from the experience in terms of appreciation for the species and understanding of its natural history is largely dependent on what the visitor experiences—an animal in a barren cage displaying aberrant behavior or a behaviorally healthy animal in a naturalistic enclosure, doing what it does naturally in the wild. Another conservation imperative of zoos is providing viable animals for reintroduction into native habitat. If captive breeding is to be a successful conservation tool, we must understand how captivity affects behavior developmentally and genetically, and how we can counter those effects if they are deleterious. More immediately, however, we need to maintain

natural behaviors in captive animals because it is ethically imperative from the perspective of animal well-being. REFERENCES Ader, R., and Blefer, M. . Prenatal maternal anxiety and offspring emotionality in the rat. Psychol. Rep. :–. Altman, J. D. . Animal activity and visitor learning at the zoo. Anthrozoos :–. Aoki, K., Feldman, M. W., and Kerr. B. . Models of sexual selection on a quantitative genetic trait when preference is acquired by sexual imprinting. Evolution :–. Archer, J., and Blackman, D. . Prenatal psychological stress and offspring behavior in rats and mice. Dev. Psychobiol. :– . Arnold, S. J. . Monitoring quantitative genetic variation and evolution in captive populations. In Population management for survival and recovery, ed. J. D. Ballou, M. E. Gilpin, and T. J. Foose, –. New York: Columbia University Press. Bashaw, M. J., Bloomsmith, M. A., Marr, M. J., and Maple, T. L. . To hunt or not to hunt? A feeding enrichment experiment with captive large felids. Zoo Biol. :–. Bateson, P. P. . The characteristics and context of imprinting. Biol. Rev. Camb. Philos. Soc. :–. Beck, B. B. . Reintroduction of captive-born animals. In Creative conservation: Interactive management of wild and captive animals, ed. P. J. S. Olney, G. M. Mace, and A. T. C. Feistner, –. London: Chapman and Hall. Beck, B. B., Castro, M. I., Stoinski, T. S., and Ballou, J. D. . The effects of pre-release environments and post-release management on survivorship in reintroduced golden lion tamarins. In Lion tamarins: Biology and conservation, ed. D. G. Kleiman and A. B. Rylands, –. Washington, DC: Smithsonian Institutions Press. Belyaev, D. K., and Trut, L. N. . Some genetic and endocrine effects of selection for domestication in silver foxes. In The wild canids, ed. M. W. Fox, –. New York: Van Nostrand Reinhold. Benus, R. F., Koolhaas, J. M., and van Oortmerssen, G. A. . Individual differences in behavioural reaction to a changing environment in mice and rats. Behaviour :–. Berkson, G., Mason, W. A., and Saxon, S. V. . Situation and stimulus effects on stereotyped behaviors of chimpanzees. J. Comp. Physiol. Psychol. :–. Biggins, D. E., Vargas, A., Godbey, J., and Anderson, S. H. . Influence of prerelease experience on reintroduced black-footed ferrets (Mustela nigripes). Biol. Conserv. :–. Bioni, M., and Zannino, L. G. . Psychological stress, neuroimmunomodulation, and susceptibility to infectious diseases in animals and man: A review. Psychother. Psychosom. :–. Blecha, F. . Immune response to stress. In Biology of animal stress: Basic principles and implications for animal welfare, ed. G. P. Moberg and J. A. Mench, –. Wallingford, UK: CABI Publishing. Blouin, S. F., and Blouin, M. . Inbreeding avoidance behaviors. Trends Ecol. Evol. :–. Britt, A., Katz, A., and Welch, C. . Project Betampona: Conservation and re-stocking of black and white ruffed lemurs (Varecia variegata variegata). In th World Conference on Breeding Endangered Species: Linking zoo and field research to advance conservation, ed. T. L. Roth, W. F. Swanson, and L. K. Blattman, –. Cincinnati: Cincinnati Zoo. Broom, D. M., and Johnson, K. G. . Stress and animal welfare. London: Chapman and Hall. Cameron, N. M., Champagne, F. A., Fish, C., Ozaki-Kuroda, K.,

m. elsbeth mcphee and kathy carlstead and Meaney, M. J. . The programming of individual differences in defensive responses and reproductive strategies in the rat through variations in maternal care. Neurosci. and Biobehav. Rev. :–. Carlstead, K. . Determining the causes of stereotypic behaviors in zoo carnivores: Toward appropriate enrichment strategies. In Second nature: Environmental enrichment for captive animals, ed. D. J. Shepherdson, J. D. Mellen, and M. Hutchins, –. Washington, DC: Smithsonian Institution Press. ———. . Addressing and assessing animal welfare. In AAZPA Annual Conference Proceedings, –. Wheeling, WV: American Association of Zoological Parks and Aquariums. Carlstead, K., and Brown, J. L. . Relationships between patterns of fecal corticoid excretion and behavior, reproduction, and environmental factors in captive black (Diceros bicornis) and white (Ceratotherium simum) rhinoceros. Zoo Biol. :–. Carlstead, K., Brown, J. L., and Seidensticker, J. C. . Behavioural and adrenocortical responses to environmental change in leopard cats (Felis bengalensis). Zoo Biol. :–. Carlstead, K., and Shepherdson, D. J. . Effects of environmental enrichment on reproduction. Zoo Biol. –. Carlstead, K., and Seidensticker, J. C. . Seasonal variation in stereotypic pacing in an American black bear (Ursus americanus). Behav. Process. –. Carlstead, K., Seidensticker, J. C., and Baldwin, R. . Environmental enrichment for zoo bears. Zoo Biol. –. Cavigelli, S. A., and McClintock, M. K. . Fear of novelty in infant rats predicts adult corticosterone dynamics and early death. Proc. Natl. Acad. Sci. U.S.A. :–. Chamove, A. S., Hosey, G. R., and Schaetzel, P. . Visitors excite primates in zoos. Zoo Biol. :–. Clarke, A. S., Juno, C. J., and Maple, T. L. . Behavioral effects of a change in physical environment: A pilot study of captive chimpanzees. Zoo Biol. :–. Clubb, R., and Mason, G. J. . Captivity effects on wide-ranging carnivores. Nature :–. Cociu, M., Wagner, G., Micu, N. E., and Mihaescu, G. . Adaptational gastro-enteritis in Siberian tigers. Int. Zoo Yearb. : –. Cohen, S., Kessler, R. C., and Underwood Gordon, L. . Strategies for measuring stress in studies of psychiatric and physical disorders. In Measuring stress: A guide for health and social scientists, ed. S. Cohen, R. C. Kessler, and L. Underwood Gordon, –. New York: Oxford University Press. Darwin, C. R. . The variation of animals and plants under domestication. Baltimore: Johns Hopkins University Press. Dawkins, M. S. . From an animal’s point of view: Motivation, fitness and animal welfare (with commentaries). Behav. Brain Sci. :–. ———. . Using behavior to assess welfare. Anim. Welf. : S–S. ———. . A user’s guide to animal welfare science. Trends Ecol. Evol. :–. DeVries, A. C., Lasper, E. R., and Detillion, E. E. . Social modulation of stress responses. Physiol. Behav. :–. Draper, W. A., and Bernstein, I. S. . Stereotyped behaviour and cage size. Percept. Mot. Skills :–. Elsasser, T. H., Klasing, K. C., Filipov, N., and Thompson, F. . The metabolic consequences of stress: Targets for stress and priorities of nutrient use. In Biology of animal stress: Basic principles and implications for animal welfare, ed. G. P. Moberg and J. A. Mench, –. Wallingford, UK: CABI Publishing. Endler, J. A. . Natural selection in the wild. Princeton, NJ: Princeton University Press. Engel, G. L. . A psychological setting of somatic disease: The giving up-given up complex. Proc. R. Soc. Med. :–.

311

Frankham, R., Hemmer, H., Ryder, O., Cothran, E., Soulé, M. E., Murray, N., and Snyder, M. . Selection in captive populations. Zoo Biol. :–. Frankham, R., Manning, H., Margan, S. H., and, Briscoe, D. A. . Does equalization of family sizes reduce genetic adaptation to captivity? Anim. Conserv. :–. Glatston, A. R., Geilvoet-Soeteman, E., Hora-Pecek, E., and van Hooff, J. . The influence of the zoo environment on social behavior of groups of cotton-topped tamarins, Saguinus oedipus oedipus. Zoo Biol. :–. Gould, E., and Bres, M. . Regurgitation and reingestion in captive gorillas: Description and intervention. Zoo Biol. : –. Grahn, M., Langefors, A., and von Schantz, T. . The importance of mate choice in improving viability in captive populations. In Behavioral ecology and conservation biology, ed. T. Caro, –. New York: Oxford University Press. Hannah, A. C., and Brotman, B. . Procedures for improving maternal behavior in captive chimpanzees. Zoo Biol. :–. Harvey, P. W., and Chevins, P. F. D. . Crowding pregnant mice affects attack and threat behavior of male offspring. Horm. Behav. :–. Hediger, H. . Wild animals in captivity. New York: Dover Publications. Henry, J. P. . The relation of social to biological processes in disease. Soc. Sci. Med. :–. Hess, E. H. . Imprinting. New York: Van Nostrand Reinhold. Hogan, J. A., and Bolhuis, J. J. . The development of behaviour: Trends since Tinbergen (). Anim. Biol. :–. Hosey, G. R., and Druck, P. L. . The influence of zoo visitors on the behaviour of captive primates. Appl. Anim. Behav. Sci. :–. Hutchins, M., Willis, K., and Wiese, R. J. . Strategic collection planning: Theory and practice. Zoo Biol. :–. Kleiman, D. G. . The sociobiology of captive propagation. In Conservation biology: An evolutionary and ecological perspective, ed. M. E. Soulé and B. A. Wilcox, –. Sunderland, MA: Sinauer Associates. ———. . Reintroduction of captive mammals for conservation: Guidelines for reintroducing endangered species into the wild. BioScience :–. Kleiman, D. G, Beck, B. B., Baker, A., Ballou, J. D., Dietz, L., and Dietz, J. . The conservation program for the golden lion tamarin, Leontopithecus rosalia. Endanger. Species Updat. : –. Künzl, C., Kaiser, S., Meier, E., and Sachser, N. . Is a wild mammal kept and reared in captivity still a wild animal? Horm. Behav. :–. Ladewig, J. . Chronic intermittent stress: A model for the study of long-term stressors. In Biology of animal stress: Basic principles and implications for animal welfare, ed. G. P. Moberg and J. A. Mench, –. Wallingford, UK: CABI Publishing. Lickliter, R., and Ness, J. W. . Domestication and comparative psychology: Status and strategy. J. Comp. Psychol. :–. Lindburg, D. G. . Enrichment of captive mammals through provisioning. In Second nature: Environmental enrichment for captive animals, ed. D. J. Shepherdson, J. D. Mellen, and M. Hutchins, –. Washington, DC: Smithsonian Institution Press. Mallapur, A., Qureshi, Q., and Chellam, R. . Enclosure design and space utilization by Indian leopards (Panthera pardus) in four zoos in southern India. J. Appl. Anim. Welf. Sci. :–. Maple, T. L., and Hoff, M. P. . Gorilla behavior. New York: Van Nostrand Reinhold. Margulis, S. W., Hoyos, C., and Anderson, M. . Effect of felid activity on visitor interest. Zoo Biol. :–.

312

the imp orta nce of maintaining natural behaviors in captive mammals

Markowitz, H. . Behavioral enrichment in the zoo. New York: Van Nostrand Reinhold. Mason, G. J., Cooper, J., and Clarebrough, C. . Frustrations of fur-farmed mink: Mink may thrive in captivity but they miss having water to romp about in. Nature :–. Mason, G. J., and Latham, N. R. . Can’t stop, won’t stop: Is stereotypy a reliable animal welfare indicator? Anim. Welf. : –. Mason, W. A., Mendoza, S. P., and Moberg, G. P. . Persistent effects of early social experience on physiological responsiveness. In Primatology today, ed. A. Ehara, T. Kimura, D. Takenaka, and M. Iwamoto, –. Amsterdam: Elsevier Sciences Publishers. Matteri, R. L., Carroll, J. A., and Dyer, D. J. . Neuroendocrine responses to stress. In Biology of animal stress: Basic principles and implications for animal welfare, ed. G. P. Moberg and J. A. Mench, –. Wallingford, UK: CABI Publishing. McDougall, P. T., Reale, D., Sol, D., and Reader, S. M. . Wildlife conservation and animal temperament: Causes and consequences of evolutionary change for captive, reintroduced, and wild populations. Anim. Conserv. :–. McEwan, B. S. . Protective and damaging effects of stress mediators: The good and bad sides of the response to stress. Metabolism :–. McLean, I. G., Lundie-Jenkins, G., and Jarman, P. J. . Teaching an endangered mammal to recognise predators. Biol. Conserv. :–. McPhee, M. E. . Intact carcasses as enrichment for large felids: Effects on on- and off-exhibit behaviors. Zoo Biol. :–. ———. a. Effects of captivity on response to a novel environment in the oldfield mouse (Peromyscus polionotus subgriseus). Int. J. Comp. Psychol. :–. ———. b. Generations in captivity increases behavioral variance: Considerations for captive breeding and reintroduction programs. Biol. Conserv. : –. ———. . Morphological change in wild and captive oldfield mice Peromyscus polionotus subgriseus. J. Mammal. : –. McPhee, M. E., Foster, J. S., Sevenich, M., and Saunders, C. D. . Public perceptions of behavioral enrichment: Assumptions gone awry. Zoo Biol. :–. McPhee, M. E., and Silverman, E. . Increased behavioral variation and the calculation of release numbers for reintroduction programs. Conserv. Biol. :–. Mench, J. A. . Why it is important to understand animal behavior. ILAR J. :–. Miller, B., Biggins, D., Wemmer, C., Powell, R., Calvo, L., Hanebury, L., and Wharton, T. . Development of survival skills in captive-raised Siberian polecats (Mustela eversmanni) II: Predator avoidance. J. Ethol. :–. Miller, B., Hanebury, L. R., Conway, C., and Wemmer, C. . Rehabilitation of a species: The black-footed ferret (Mustela nigripes). In Wildlife rehabilitation, ed. D. Ludwig, –. Edina, MN: Edina Printing. Miller, B., Hanebury, D., and Vargas, A. . Reintroduction of the black-footed ferret (Mustela nigripes). In Creative conservation: Interactive management of wild and captive animals, ed. P. J. S. Olney, G. M. Mace, and A. T. C. Feistner, –. London: Chapman and Hall. Moberg, G. P. . Biological response to stress: Implications for animal welfare. In Biology of animal stress: Basic principles and implications for animal welfare, ed. G. P. Moberg and J. A. Mench, –. Wallingford, UK: CABI Publishing. Moore, C. L. . Maternal contributions to the development of masculine sexual behavior in laboratory rats. Dev. Psychobiol. :–.

Morgan, K. N., and Tromborg, C. T. . Sources of stress in captivity. Appl. Anim. Behav. Sci. :–. Möstl, E., and Palme, R. . Hormones as indicators of stress. Domest. Anim. Endocrinol. :–. Odberg, F. O. . The influence of cage size and environmental enrichment on the development of stereotypies in bank voles (Clethrionomys glareolus). Behav. Process. :–. Owen, M. A., Swaisgood, R. R., Czekala, N. M., Steinman, K., and Lindburg, D. A. . Monitoring stress in captive giant pandas (Ailuropoda melanoleuca): Behavioral and hormonal responses to ambient noise. Zoo Biol. :–. Paulk, H. H., Dienske, H., and Ribbens, L. G. . Abnormal behavior in relation to cage size in Rhesus monkeys. J. Abnorm. Psychol. :–. Peel, A. J., Vogelnest, L., Finnegan, M., Grossfeldt, L., and O’Brien, J. K. . Non-invasive fecal hormone analysis and behavioral observations for monitoring stress responses in captive western lowland gorillas (Gorilla gorilla gorilla). Zoo Biol. :–. Penn, D. J., and Potts, W. K. . The evolution of mating preferences and major histocompatibility complex genes. Am. Nat. :–. Perkins, L. . Variables that influence the activity of captive orangutans. Zoo Biol. :–. Prescott, M. J., and Buchanan-Smith, H. M. . Cage sizes for tamarins in the laboratory. Anim. Welf. :–. Price, E. O. . Differential reactivity of wild and semi-domestic deermice (Peromyscus maniculatus). Anim. Behav. :–. ———. . Behavioral aspects of animal domestication. Q. Rev. Biol. :–. ———. . Behavioral genetics and the process of animal domestication. In Genetics and the behavior of domestic animals, ed. T. Grandin, –. San Diego, CA: Academic Press. ———. . Animal domestication and behavior. New York: CABI Publishing. Reading, R. P., Clark, T. W., Vargas, A., Hanebury, L. R., Miller, B. J., Biggins, D. E., and Marinari, P. E. . Black-footed ferret (Mustela nigripes): Conservation update. Small Carniv. Conserv. : –. Ryan, S., Thompson, S. D., Roth, A. M., and Gold, K. C. . Effects of hand-rearing on the reproductive success of western lowland gorillas in North America. Zoo Biol. :–. Scott-Brown, J. M., Herrero, S., and Mamo, C. . Monitoring of released swift foxes in Alberta and Saskatchewan. Final report. Unpublished report to the Canadian Fish and Wildlife Service, Edmonton, Alberta. Seidensticker, J., and Forthman, D. L. . Evolution, ecology, and enrichment: Basic considerations for wild animals in zoos. In Second nature: Environmental enrichment for captive animals, ed. D. J. Shepherdson, J. D. Mellen, and M. Hutchins, –. Washington, DC: Smithsonian Institution Press. Shier, D. M., and Owings, D. H. . Effects of predator training on behavior and post-release survival of captive prairie dogs (Cynomys ludovicianus). Biol. Conserv. :–. Sluckin, W. . Imprinting in guinea-pigs. Nature :–. Soulé, M. E. . Conservation biology and the “real world.” In Conservation biology: The science of scarcity and diversity, ed. M. E. Soulé, –. Sunderland, MA: Sinauer Associates. Soulé, M. E., Gilpin, M. E., Conway, W., and Foose, T. J. . The millennium ark: How long a voyage, how many staterooms, how many passengers? Zoo Biol. : –. Stoinski, T. S., and Beck, B. B. . Changes in locomotor and foraging skills in captive-born, reintroduced golden lion tamarins (Leontopithecus rosalia rosalia). Am. J. Primatol. :–. Suomi, S. J. . Genetic and maternal contributions to individual differences in rhesus monkey biobehavioral development. In Perinatal development: A psychobiological perspective, ed.

m. elsbeth mcphee and kathy carlstead N. Krasnegor, E. Blass, M. Hofer, and W. Smotherman, –. New York: Academic Press. Swaisgood, R. R., and Shepherdson, D. J. . Scientific approaches to enrichment and stereotypies in zoo animals: What’s been done and where should we go next? Zoo Biol. :–. Thompson, K. V. . Factors affecting pair compatibility in captive kangaroo rats, Dipodomys heermanni. Zoo Biol. :–. Thompson, V. D. . Behavioral responses of  ungulate species in captivity to the presence of humans. Zoo Biol. :–. Thompson, W. R., Watson, J., and Charlsworth, W. R. . The effects of prenatal maternal stress on offspring behavior in rats. Psychol. Monogr. :–. Trut, L. N. . Early canid domestication: The farm fox experiment. Am. Sci. :–. USFWS (U.S. Fish and Wildlife Service). . Red wolf recovery plan. Atlanta: U.S. Fish and Wildlife Service. Vargas, A., and Anderson, S. H. . Effects of experience and cage environment on predatory skills of Black Footed Ferrets (Mustela nigripes). J. Mammal. :–. Ward, I. L. . Prenatal stress feminizes and demasculinizes the behavior of males. Science :–. West-Eberhard, M. J. . Developmental plasticity and evolution. New York: Oxford University Press. Whay, H. R., Main, D. C. J., Green, L. E., and Webster, A. J. F. . Animal-based measures for the assessment of welfare state of dairy cattle, pigs, and laying hens: Consensus of expert opinion. Anim. Welf. :–.

313

Whitten, P. L., Brockman, D. K., and Stavisky, R. C. . Recent advances in noninvasive techniques to monitor hormone-behavior interactions. Yearb. Phys. Anthropol. :–. Wiedenmayer, C. . Effect of cage size on the ontogeny of stereotyped behaviour in gerbils. Appl. Anim. Behav. Sci. :–. Wielebnowski, N. . Contributions of behavioral studies to captive management and breeding of rare and endangered mammals. In Behavioral ecology and conservation biology, ed. T. M. Caro, –. Oxford: Oxford University Press. ———. . Stress and distress: Evaluating their impact for the well-being of zoo animals. J. Am. Vet. Med. Assoc. :–. Wielebnowski, N., Fletchall, N., Carlstead, K., Busso, J. M., and Brown, J. L. . Non-invasive assessment of adrenal activity associated with husbandry and behavioral factors in the North America clouded leopard population. Zoo Biol. :–. Wilson, S. F. . Environmental influences on the activity of captive apes. Zoo Biol. :–. Winkelstraeter, K. H. . Das Betteln der Zoo-Tiere. Ph.D. diss., Hans Huber, Berlin. Wolf, R. L., and Tymitz, B. L. . Studying visitor perceptions of zoo environments: A naturalistic view. Int. Zoo Yearb. : –. Yalden, D. W. . The problems of reintroducing carnivores. In The proceedings of a symposium held by The Zoological Society of London and The Mammal Society: London, nd and rd November , ed. N. Dunstone and M. L. Gorman, –. Oxford: Clarendon Press. Yamada, J. K., and Durrant, B. S. . Reproductive parameters of clouded leopards (Neofelis nebulosa). Zoo Biol. :–.