Effects of woodchips and buried food on behavior patterns and psychological well-being of captive rhesus monkeys.код для вставкиСкачать
American Journal of Primatology 23:141-151 (1991) RESEARCH ARTICLES Effects of Woodchips and Buried Food on Behavior Patterns and Psychological Well-Being of Captive Rhesus Monkeys GAYLE DIGREGORIO BYRNE AND STEPHEN J. SUOMI Laboratory of Comparative Ethology, NICHD, Bethesda, Maryland The effects of adding woodchip litter to bare-floored pens, burying monkey chow in the woodchips, and scattering sunflower seeds in woodchips was studied in 2 stable social groups of rhesus macaques (Macaca mulatta) to ascertain the effects of these manipulations on levels of foraging, exploration, abnormal behavior, social interactions, and urinary cortisol levels. The addition of woodchips increased exploration and feeding levels and decreased social interactions. Burial of regular monkey chow in woodchips had little effect on behavior beyond that of the woodchips alone, increasing exploration and decreasing passivity. The addition of sunflower seeds to the woodchips encouraged increased feeding and exploration and led to decreases in passivity and social interaction. There was little discernible effect of woodchip enrichment on urinary cortisol values. In contrast to some previous studies, there was no effect of wood &ips or sunflower seeds on the occurrence of agonistic interactions, play, or abnormal behavior patterns. Key words: rhesus, enrichment, behavior, foraging, cortisol INTRODUCTION There is currently a growing interest in the “enrichment” of captive environments and the influence of environmental variables on captive animals. The revised Animal Welfare Act of 1985 mandates a “physical environment adequate to promote the psychological well-being of primates”; the debate continues in the research community on the necessary and sufficient conditions t o provide such a n environment. One approach to a definition of psychological well-being makes the assumption that well-being is enhanced by providing animals with opportunities to perform the full range of behaviors seen in wild populations, at species-normative levels. Captivity alone increases the frequency of some social interactions in many species over levels seen in the wild [Rowell, 19671, as does provisioning of wild animals [Southwick, 1969; Marriott, 19881. Behavior patterns that have become labeled “abnormal” are also artifacts of captivity [Erwin & Deni, 19791. There have been many recent efforts to promote behavior patterns in captive nonhuman primates Received for publication February 14,1990; revision accepted October 17, 1990. Address reprint requests to Gayle Byme, NIH Animal Center, Building 112, Poolesville, MD 20837. 0 1991 Wiley-Liss, Inc. 142 / Byrne and Suomi that are characteristic of wild populations by enrichment of the physical environment. Some of these efforts have involved stimulation of behavior by providing a suitable substrate in which to forage [Chamove & Anderson, 1979; Chamove et al., 1982; Anderson & Chamove, 1983; Tripp, 1985; Westergaard & Fragaszy, 1985; Maple & Finlay, 1986; Kleiman et al., 1986; Bloomsmith et al., 1988, Boccia, 19881. Most of these studies have reported changes that are commonly regarded as beneficial to well-being, such as increases in activity and exploration, decreases in abnormal behavior, increases in play, and decreases in aggression. Thus, the imposition of a foraging requirement in some studies seems to negate some of the artificial effects of captivity and can be hypothesized to enhance psychological well-being. Wild environments are not without their own stresses; e.g., from predation, fluctuations in food availability, and intraspecific aggression. A limit must be imposed on approximations to wild patterns by balancing psychological well-being with physical well-being. Another approach to the assessment of well-being lies in monitoring behavioral and psychological correlates of stress. While it is recognized that many physiological systems respond to stress [e.g., Mason, 19741, one of the most widely studied is the pituitary-adrenocortical system. Emotionally arousing stimuli that result in some degree of novelty or uncertainty are potent activators of the adrenocortical system [Henry & Stephens, 1977; Hennessy et al., 1979; Sassenrath, 1983; Levine, 19851. Feeding is a vital activity in any animal’s life, and a challenge to expectancies concerning the procurement of food can be expected to affect emotional and physiological responses, including adrenocortical activity. There have been few studies to relate changes in environmental enrichment or foraging opportunities to cortisol levels in nonhuman primates [see Line et al., 1987; Novak & Drewsen, 1989, for exceptions]. The present study was designed to investigate the effects of woodchips and foraging on behavior patterns and cortisol secretion in captive, stable groups of rhesus macaques. The effects of physical enrichment of the environment were tested by adding woodchips to the bare floors of animals’ pens. The effects of foraging enrichment were investigated by burying monkey chow in the woodchips. Foraging demand was manipulated further by scattering sunflower seeds in the woodchips. Finally, in one group, the regular ration of monkey chow was increased to test the effects of a change in food availability. Changes in activity levels, affiliative and agonistic patterns, abnormal behaviors, and urinary levels of cortisol were all monitored with changes in environmental conditions. MATERIALS AND METHODS Subjects Two groups of rhesus macaques (Macaca mulatta) were used in this study. Group 1consisted of an adult male, 3 adult females, 5 juvenile offspring (2-3 years old), and a 4 year-old male adopted into the group at 5 months of age. One infant was born to group 1 during week 12 of the study. Data involving interactions with this infant were excluded from analysis. Group 2 consisted of an adult male, 2 adult females and 3 offspring (2-4 years old). The adults in this study had been raised in 2 peer groups from the age of 1 month, and these groups had been together for 9 years at the time the study began. Each group was housed in an indoor-outdoor pen with bare floors of tile (inside) and concrete (outside). The indoor pen measured 2.4 x 3.0 x 2.2 m and contained 3 resting shelves; the outdoor pen measured 2.5 x 3.0 x 2.5 m and contained a single isolation cage which was used to collect urine samples. The animals were maintained on a 12L:12D light schedule indoors, and were fed mon- Effects of Woodchips on Behavior I 143 TABLE I. Schedule of Experimental Conditions* Group 1 2 Experimental Condition B W F F' W' F' B' INC S B w1 F w2 W' F' B' S Week 1-10 11-14 15-17 18-22 23-21 28-31 32-35 36-37 38-39 1-8 9-12 13-16 17 18-25 26-29 30-33 34-37 "B = bare floors, regular feeding (B' = fed 2 x /day); W = woodchips, regular feeding (W' = fed 2 x /day); F = woodchips, buried food (F' = fed 2 x /day); S = woodchips, regular feeding (2 x /day), sunflower seeds scattered in bedding; INC = woodchips, regular feeding (2 x /day), increased food ration. key chow either once or twice a day (see below). Water was available ad libitum through automatic watering spouts in the indoor pen. For the first 5 months of the study (June to November) the animals had access to the indoor and outdoor pens both day and night. During the winter months (November to March) animals were locked inside overnight, and were occasionally kept locked in on extremely cold days. Indoor and outdoor pens were cleaned every morning in the bare-floored condition. During the conditions in which woodchips were added to the indoor pens, the woodchips were replaced and the pens were cleaned once a week. Procedure This study employed a within-subjects reversal design under three main environmental conditions: 1)Baseline (B)-bare floors, food thrown in pen and scattered on bare floor; 2) Woodchips (W)-5 cm layer of woodchips on floor of indoor pen, food thrown on top of woodchips; 3) Foraging (F)-woodchips on floor, food buried beneath woodchips. Each of these conditions was presented under two different feeding regimes: 1)food given primarily once a day, in the afternoon, and 2) food given both in the morning (0700) and in the afternoon (1400-1500). This change was brought about at the reversal point of the study when a change in animal care personnel a t the NIH Animal Center inadvertently resulted in a change in feeding schedule and a small increase (10%) in total food ration. The schedule of conditions for each group is shown in Table I. The order of conditions was the same for both groups, but presentation of conditions t o one group was staggered 2 weeks behind the other group. Thus, as each group entered a new condition the other group was still in a prior condition 144 I Byrne and Suomi under identical variables of day length, temperature, etc. Wherever possible, these conditions were presented in 4-week blocks. However, the unplanned change in feeding schedule at week 19 cut short the conditions in effect a t that time and necessitated the extension of the succeeding conditions i n order to get back on a staggered schedule. After the replications of the three main experimental conditions were completed, two shorter conditions were added. After the last bare-floored baseline condition, the woodchips were reinstated and, in addition to the regular food ration placed on top of the chips, sunflower seeds were scattered every morning to further induce foraging. This condition is identified as “S” in Table I, and was the final condition under which observations were performed for both groups. In group 1one other condition preceded the S condition. In this condition (INC) the woodchips were reinstated and the total daily ration of food was increased approximately 25% and placed on top of the woodchips. This group was almost twice the size of group 2, and this manipulation was intended to offset the increased competition for food in the larger group. The increased food ration was kept in effect during the S condition which followed. Observational scoring. Within each environmental condition each animal was observed for 10 minutes twice each week, once from 0900 to 1000 and once from 1300 to 1400 hours. Within each 10 minute focal animal sample, 10-second point samples of a number of behaviors were recorded (Table I1 lists the behaviors that were scored). Order of observation of animals was randomized prior to the session. Urine sampling. A single isolation cage was installed in each group’s outdoor pen. The animals were trained to urinate either inside or on top of the cage by reinforcement with preferred foods (raisins, grapes, etc.). The urine was then aspirated with a syringe and plastic collection tube from the metal pan underneath the cage and frozen to -20°C. Attempts were made to obtain samples from both groups two to three times during each observation day, at regular times between 0900 and 1200. A subset of all the samples was selected such that there was a maximum of four samples for any one subject in any one condition. The urine samples were analyzed for excreted levels of cortisol by radioimmunoassay (Ciba Corning Magic‘@”Cortisol radioimmunoassay). Samples were also analyzed for creatinine in order to correct for variables such as body weight and dilution of the urine. Creatinine values were determined via kinetic alkaline picrate assay. Cortisol values were divided by their corresponding creatinine values to give an index of cortisol excretion. Data analysis. Behavioral scoring data were analyzed using repeated measures ANOVAs, with environmental conditions as the within-subject factor. Tukey HSD tests were used to identify significant differences among means when the overall F was significant at the .05 level. Data from the two groups were analyzed separately, as the differences in size and composition of the groups precluded any between-groups comparisons. The behavioral categories listed above (Social, Abnormal, Agonistic, Active, Feeding, and Passive) served as the behavioral measures used i n analyses. Means of each subject’s scores in each category were calculated for each environmental condition. Post hoc tests first compared the levels of behavior in the replications of each experimental condition, to test for effects of changes in season and indooroutdoor access and to ascertain if the unplanned difference in feeding schedule significantly affected behavior (e.g., B was compared to B’, W to W‘, and F to F’). If there were no significant differences between the replications they were combined to yield a mean value for each condition, and ANOVAs were performed Effects of Woodchips on Behavior I 145 TABLE 11. Behavioral Categories and Definitions Social Contact-in physical contact with another animal, not engaged in any other scored social behavior with that animal Proximity-within 30 cm of another animal, not touching or engaged in any other scored social behavior with that animal Grooming-one animal pickinglbrushing through the fur of another animal-may be scored either as “groom” or “groom receive”, according to the focal animal’s role Play-animated chasing, wrestling, sham-biting, may be accompanied by open-mouth play face Sex-either male or female sexual postures-includes mounting, thrusting, presenting Abnormal Abnormal-includes stereotypic pacing (repeated at least three cycles) and other repetitive abnormal behavior patterns, e.g., self-biting or mouthing, “saluting”, etc. Abnormal sex-masturbation by self-stimulation of genitals, either manually, orally, or through disoriented mounts of other animals. Agonistic Withdraw-deliberately move out of the path of a n approaching animal Aggression-includes noncontact open-mouth threat and/or stare, grabbing, biting, chasing of another animal (not in play), bouncing display Submission-includes fear grimace, lipsmacking, crouching, screeching, screaming, flight from a threatening or attacking animal Active Locomotion-moving verticaly or horizontally at least 15 cm, not engaged in any social behavior or self-play Explore environment-oral, olfactory, or tactile manipulation or investigation of a nonfood object or structure-includes exploration of woodchips but excludes searching through woodchips for food, excludes manipulation of object while engaged in social or self-play Self-play Feeding Forage-search for and/or handle food item, includes searching through woodchips for food, picking up and investigating food item. Eat-put food item in mouth and chew, also includes transfer of food from cheek pouches to mouth Drink-take i n water in any way Passive Sitting, standing, or lying alone, not engaged in any other scored behavior across the main conditions. If significant differences were revealed between the replications, additional ANOVAs were performed which included the differing replications a s separate conditions. Differences in relative cortisol levels over the experimental conditions were also tested in ANOVAs. RESULTS Focal Animal Samples ANOVAs were first carried out across all of the environmental conditions. An alpha level of .05 was adopted for all analyses. Post hoc tests were used to compare 146 / Byrne and Suomi Baseline (B), Woodchips (W), and Foraging (F) conditions both before and after the midpoint in the study. In most cases the levels of behavior seen within the replications of the three main conditions were not significantly different. There were two exceptions, however. In group l,the level of Active in the first F' condition was significantly lower than that in the preceding condition F (F(8,72) = 18.00). When the first F' was considered separately from the other combined F conditions, it was revealed that Active levels in that phase were significantly lower than in the combined W, F, or (A) SOCI,4L (A) FEEDING 401 m 2 in rn m 0 z m 0 15 z a W 4 = B 0 (B) Ln Y a a W F INC S 41 3 n 15 W F INC F INC ACTIVE 45 30 m m 0 0 5 z 15 4 W W = 0 B (C) W F INC S S 6o 0 10 m I B 0 v) in 4 2oL i (El) !n 0 2 I 10 ABNORMAL 20 !n y W 30 o 1 S B W AGONlSTlC in 2 B W F INC S B W F INC S Fig. 1. Mean occurrence of social, abnormal, and agonistic in the focal samples over the main experimental conditions. Open bars = p o u p 1;Hatched bars = group 2. Vertical bars denote standard errors. Fig. 2. Mean occurrence of feeding, active, and passive in the focal samples over the main experimental conditions. Open bars = group 1;Hatched bars = group 2. Vertical bars denote standard error. Effects of Woodchips on Behavior I 147 S conditions, and significantly higher than those in INC (F(5,45) = 22.82). In group 2, Feeding levels in the first two presentations of the woodchip conditions significantly differed from each other-Feeding in W2 was much lower than in W 1 (F(7,35) = 8.30). Examination of the data revealed a Feeding level of zero during W2; none of the animals in this group were ever observed eating, foraging, or drinking during this condition. The replications of the main conditions were then combined, and behavior scores across the main conditions were examined (Figs. 1, 2). Social. The number of samples in which Social was scored was significantly lower in S than in B, F, or INC in group 1 (F(4,36) = 8.04), and lower in S than in all other conditions in group 2 (F(3,15) = 7.70 (Fig. 1A)). In addition, Social levels in group 1 were significantly lower in W than in B. Further analyses examined the levels of the behaviors comprising the Social category. Changes in Social were due to changes in contact, proximity, and grooming; there were no significant differences in levels of play or sex among any of the experimental conditions. Abnormal. There were no significant changes across conditions in Abnormal behavior in either group (group 1-F(4,36) = 0.22; group 2-F(3,15) = 2.27; Fig. 1B). Agonistic. There were no significant changes across conditions in Agonistic behavior in either group (group 1-F(4,36) = 0.89; group 2-F(3,15) = 1.70; Fig. 1C). Feeding. The number of samples in which animals were observed Feeding in group 1 increased significantly in INC and S from those seen in B, W, or F (F(4,36) = 13.27; Fig. 2A). Feeding in group 2 was significantly higher in S than in B (F(3,15) = 4.40) and higher in W than in B when W2 was excluded from the W condition (see above). Active. In group 1 levels of Active in W, F, and S were greater than those in B. Active in INC was less than in F and S (F(4,36) = 29.41; Fig. 2B). In group 2 scores for Active were greater in the S condition than in any of the other conditions (F(3,15) = 15.13). In addition, analysis of environmental exploration (the main component of Active scores) revealed that the level of this behavior was greater in F than in B (F(3,15) = 15.60). (Significant changes in Active scores were always due to changes in environmental exploration; no significant changes in locomotion or self-play were observed.) Passive. The number of samples in which Passive was scored was significantly lower in S than in B and W for both groups (group 1-44,36) = 9.19; group 2-F(3,15) = 6.25; Fig. 2C). In group 1 levels of Passive in F and INC were also lower than those seen in Baseline. Cortisol Values An ANOVA based on data from 6 subjects in group 1 for the replications of the main conditions revealed that cortisol levels in the second F’ condition were significantly higher than those in B, W, or F (F(6,30) = 2.75). There were no significant differences within replications for group 2 in the 3 subjects with adequate samples (F(6,12) = 0.47). An analysis which combined replications of the three main conditions for all subjects revealed no significant differences among the combined B, W, or F conditions in either group. There were not enough samples from conditions INC or S to include in most of the statistical analyses. When the INC condition was included for 3 subjects in group 1 with adequate samples, cortisol levels in INC were significantly lower than in B or F (F(3,6) = 6.31). 148 I Byrne and Suomi DISCUSSION The results of this study with rhesus macaques were, to some extent, in agreement with those of previous studies with other species, but several discrepancies emerged a s well. Enrichment with woodchips increased activity and led to decreases in social interactions, echoing results from Chamove and Anderson [19791, Chamove et al. r19821, and Westergaard and Fragaszy 119851. Burying food in woodchips had little effect on behavior beyond that of the woodchips alone; the addition of buried food was associated in both groups with increased environmental exploration, and, in group 1,with decreased levels of Passive in comparison to Baseline levels. The burial of food in the F condition was intended to affect the amount of effort required to find food. In reality, however, simply burying food in 5 cm of woodchips did not constitute a particularly difficult foraging task. The large pieces of monkey chow were easily uncovered, either deliberately or by accident, and were often examined and discarded, leaving a n assortment of unburied food available for other animals. Sunflower seeds scattered in the woodchips led to even greater changes in behavior, with increases in activity and foraging and decreases in passivity and social interaction. These findings again concur with Chamove’s studies as well as with later studies by Bloomsmith et al. 119881 and Boccia . If approximation to patterns of behavior seen in the wild is indicative of psychological well-being, then the S condition i n the present study came closest to fulfilling this criterion. Captive rhesus may often spend less than 20% of their time feeding [Bernstein, 1971; Post & Baulu, 19781, whereas percentages are typically higher for wild, unprovisioned animals [Malik, 1986; Seth & Seth, 19861. In the present study, the percent of samples in which eating and foraging occurred was small, ranging from 1 to 7 i n the combined B, W, or F conditions. During the S condition this figure increased to 35% in group 1 (which also had a n increased food ration) and 13% in group 2. Social behavior levels, which may be artificially increased by captivity, were lowered in the S condition; nonsocial exploration and activity increased as the incentive to attend to the physical aspects of the environment increased. These changes resulted in a time budget more characteristic of natural populations without compromising animals’ physical well-being. In contrast to previous studies with other species [e.g., Chamove & Anderson, 19791, there were no changes i n abnormal behavior associated with any of the environmental manipulations i n the present study. Abnormal behavior was relatively infrequent to begin with, occurring i n less than 10% of the samples, and was most often exhibited by the adults, who had been raised from infancy in peer groups. It has been suggested that peer-reared rhesus macaques are more reactive to stress throughout their lives than mother-reared ones [e.g., Suomi & Harlow, 19753; the enrichment i n the present study may simply not have been sufficient to reduce the incidence of abnormal behavior patterns in these animals. The lack of effect of environment on levels of agonistic behavior or play was probably due to the long-term stability and structure of the subject groups. The importance of social enrichment, as contrasted with physical enrichment, has been well recognized [Reinhardt, 1988; DeWaal, 19891. The nature of the social groups in the present study was a comparatively enriched one; there were plenty of opportunities for play t h a t were intrinsic to the social environment. Conversely, levels of agonistic behavior were relatively low and were probably affected more by the nature and stability of the social environment than the physical one. Effects of Woodchips on Behavior / 149 Urinary Cortisol There were few significant changes in cortisol values throughout the study. In group 1 there was an increase in cortisol over several days during the second F’ condition, but it is difficult to account for this increase. This period occurred in late December, after the groups had begun to be locked in overnight due to cold. This restriction of freedom of movement may have been related to the increase in cortisol levels. However, there was no increase at that time of cortisol levels in the other group, or in any other phase during the winter months. In general, cortisol values gave little reliable indication of animals’ emotional reactions to experimental manipulations of the housing environment, in contrast to earlier work using 24 hour urine samples [e.g., Mason et al., 1957; Mason, 19591. It may be that single urine samples, even taken repeatedly, are not as effective as 24 hour aggregate collections in detecting changes in overall cortisol excretion, although Novak and Drewsen , using single samples, did find changes in urinary cortisol levels with changes in home cage lighting conditions in rhesus macaques. It is also possible that the present environmental manipulations simply did not affect cortisol levels in the long term. Under more controlled conditions of sample collection, animal care, feeding, and indoor-outdoor access more conclusions might be drawn as to the utility of these values. Implications for Psychological Well-Being The use of woodchip litter and the addition of sunflower seeds is a promising strategy for enhancing psychological well-being in social groups of rhesus macaques, as has been suggested in studies with other macaque species. The fact that behaviors such as aggression and play showed no change in the present study emphasizes the importance of a stable social environment in interpreting levels of social behavior. Psychological well-being is encouraged above all by access to conspecifics; new animal welfare regulations call for increased use of social groupings of nonhuman primates. It must be recognized, however, that there may be increased risk to well-being when social groups are newly formed or if the composition of groups is constantly changing t o meet experimental requirements. Proposed regulations for nonhuman primates which call for varied types of food and methods of feeding recognize the potential for enrichment that these techniques afford. Beneficial effects may be obtained solely with supplemental foraging tasks, (e.g., sunflower seeds), and it may not be necessary or advantageous to require animals to forage for their full ration. Indeed, some of Rosenblum’s studies in bonnet macaques [e.g., Rosenblum & Sunderland, 1982; Rosenblum and Paully, 19841 have shown that such demanding requirements may be detrimental to group relations and infant development. In addition, extreme variation in feeding schedules or methods may impart an atmosphere of unpredictability or uncontrollability which may affect group social relations and physiological stress responses, and may actually be counterproductive. CONCLUSIONS 1. Enrichment of rhesus macaques’ bare-floored pens with woodchips led to increases in environmental exploration and feeding and decreases in social interactions, echoing results of studies with other species of nonhuman primates. 2. Burying chow in woodchips had little effect on behavior beyond that of the woodchips alone; buried food was associated with increased environmental exploration, and decreased passivity. 150 I Byrne and Suomi 3. Sunflower seeds scattered in the woodchips led t o e v e n greater changes in behavior, w i t h f u r t h e r increases in activity and feeding and decreases in passivity and social interaction. 4. There w a s little systematic c h a n g e in urinary cortisol values w i t h different environmental conditions. The l a c k of significant variation in cortisol m a y represent a failure of adequate sample collection and experimental control. It is also possible that the environmental m a n i p u l a t i o n s did n o t affect cortisol levels in the long term. 5 . There were n o c h a n g e s in levels of agonistic behavior, play, or a b n o r m a l behavior in this study. The stability and composition of the social group m a y influence these behaviors more than do aspects of the physical environment. ACKNOWLEDGMENTS The authors w i s h to acknowledge the contributions of E v a n B y r n e and Charles Snowdon for editorial comments and statistical consultation, and Bob Brown for technical expertise. T h i s research w a s conducted as part of the Intramural Res e a r c h P r o g r a m , Laboratory of Comparative Ethology, NICHD, Bethesda, MD, and constituted part of the first author’s doctoral dissertation at the University of Wisconsin-Madison. Data on u r i n a r y cortisol values in this study m a y be obtained from the first a u t h o r . REFERENCES Anderson, J.R.; Chamove, A S . Allowing captive primates to forage. Pp. 253-256 in STANDARDS OF LABORATORY ANIMAL MANAGEMENT: PROCEEDINGS OF SYMPOSIUM, VOL. 2. Potters Bar, England, Universities Federation for Animal Welfare, 1983. Bernstein, 1.23. Activity profiles of primate groups. Pp. 69-106 in BEHAVIOR OF NONHUMAN PRIMATES, VOL. 3. A.M. Schrier; H.F. Harlow; F. Stollnitz, eds. New York, Academic Press, 1971. Bloomsmith, M.A.; Alford, P.L.; Maple, T.L. Successful feeding enrichment for captive chimpanzees. AMERICAN JOURNAL OF PRIMATOLOGY 16:155-164, 1988. Boccia, M.L. Preliminary report on the use of a natural foraging task to reduce aggression and stereotypies in socially housed pigtail macaques. LABORATORY PRIMATE NEWSLETTER 28:3,4, 1988. Chamove, AS.; Anderson, J.R. Woodchip litter in macaque groups. JOURNAL OF THE INSTITUTE OF ANIMAL TECHNICIANS 30(2):69-74, 1979. Chamove, A.S.; Anderson, J.R.; MorganJones, S.C.; Jones, S.P. Deep woodchip litter: hygiene, feeding and behavioral enhancement in eight primate species. INTERNATIONAL JOURNAL FOR THE STUDY OF ANIMAL PROBLEMS 3:308318,1982. DeWaal, F.B.M. Social nature of primates. In PROCEEDINGS OF THE FIRST CONFERENCE ON THE PSYCHOLOGICAL WELL-BEING OF CAPTIVE PRIMATES. M.A. Novak; A.J. Petto, eds., in press, 1989. Erwin, J.; Deni, R. Strangers in a strange land: abnormal behavior or abnormal environments? Pp. 1-28 in CAPTIVITY AND BEHAVIOR: PRIMATES IN BREEDING COLONIES, LABORATORIES, AND ZOOS. J. Erwin; T. Maple; G. Mitchell, eds. New York, Van Nostrand Reinhold Co., 1979. Hennessy, M.B.; Heyback, J.P.;Vernikos, J.; Levine, S. Plasma corticosterone concentrations sensitively reflect levels of stimulus intensity. PHYSIOLOGY AND BEHAVIOR 22321-825, 1979. Henry, J.P.; Stephens, P.M. STRESS, HEALTH AND THE SOCIAL ENVIRONMENT: A SOCIOBIOLOGIC APPROACH TO MEDICINE. New York, Springer-Verlag, 1977. Kleiman, D.G.; Beck, B.B.; Dietz, J.M.; Dietz, L.A.; Ballou, J.D.; Coimbra-Filho, A.F. Conservation program for the golden lion tamarin: Captive research and management, ecological studies, educational strategies and reintroduction. Pp. 959-979 in PRIMATES: THE ROAD TO SELFSUSTAINING POPULATIONS. K. Benirschke, ed. New York, Springer-Verlag, 1986. Levine, S. A definition of stress? Pp. 51-69 in ANIMAL STRESS. G.P. Moberg, ed. Bethesda, American Physiological Society, 1985. Line, S.W.; Clarke, AS.; Ellman, G.; Effects of Woodchips on Behavior I 151 Markowitz, H. Behavioral and physiologic pairing unfamiliar adult male rhesus monresponses of rhesus macaques to an envikeys for the purpose of environmental ronmental enrichment device. LABORAenrichment. LABORATORY PRIMATE TORY ANIMAL SCIENCE 37(4):508, NEWSLETTER 27(4):1-3, 1988. Rosenblum, L.A.; Paully, G.S. The effects of 1987. varying environmental demands on materMalik, I. Time budgets and activity patterns in free-ranging rhesus monkeys. Pp. 105nal and infant behavior. CHILD DEVELOPMENT 55:305-314,1984. 114 in PRIMATE ECOLOGY AND CONSERVATION. J.G. Else; P.C. Lee, eds. Rosenblum, L.A.; Sunderland, G. Feeding Cambridge, Cambridge University Press, ecology and mother-infant relations. Pp. 1986. 75-110 in PARENTING: ITS CAUSES Maple, T.L.; Finlay, T.W. Evaluating the enAND CONSEQUENCES. L.W. Hoffman; vironments of captive nonhuman primates. R. Gandelman; H.R. Schiffman, eds. Hillsdale, NJ, Lawrence Erlbaum Assoc., Pp. 479-488 in PRIMATES: THE ROAD 1982. TO SELF-SUSTAINING POPULATIONS. K. Benirschke, ed. New York, Springer- Rowell, T. A quantitative comparison of the behaviour of a wild and a caged baboon Verlag, 1986. group. ANIMAL BEHAVIOUR 15:499Marriott, B.M. Time budgets of rhesus mon509, 1967. keys (Macaca mulatta) in a forest habitat in Nepal and on Cay0 Santiago. Pp. 125- Sassenrath, E.N. Studies in adaptability: experiential, environmental, and pharmaco149 in ECOLOGY AND BEHAVIOR OF logical influences. Pp. 57-81 in HORFOOD-ENHANCED PRIMATE GROUPS. MONES, DRUGS AND SOCIAL J.E. Fa; C.H. Southwick, eds. New York, BEHAVIOR IN PRIMATES. H.D. Steklis; Alan R. Liss, Inc., 1988. A.S. Kling, eds. New York, Spectrum PubMason, J.W. Psychological influences on the lications, 1983. pituitary-adrenal cortical system. Pp. 345389 in RECENT PROGRESS IN HOR- Seth, P.K.; Seth, S. Ecology and behavior of rhesus monkeys in India. Pp. 89-103 in MONE RESEARCH. G. Pincus, ed. New PRIMATE ECOLOGY AND CONSERVAYork, Academic Press, 1959. TION. J.G. Else; P.C. Lee, eds. Cambridge, Mason, J.W. Specificity in the organization Cambridge University Press, 1986. of neuroendocrine response profiles. Pp. 68-80 in FRONTIERS IN NEUROLOGY Southwick, C.H. Aggressive behaviour of AND NEUROSCIENCE RESEARCH. P. rhesus monkeys in natural and captive Seeman; G.M. Brown, eds. Toronto, the groups. Pp. 32-43 in AGGRESSIVE BENeuroscience Institute, 1974. HAVIOUR. S. Garattini; E.G. Siggs, eds. Mason, J.W.; Harwood, C.T.; Rosenthal, N.J. Amsterdam, Excerpta Medica Foundation, 1969. Influence of some environmental factors on plasma and urinary 17-Hydroxycorti- Suomi, S.J.; Harlow, H.F. The role and reacosteroid levels in the rhesus monkey. son of peer friendships in rhesus monkeys. AMERICAN JOURNAL OF PHYSIOLPp. 153-185 in FRIENDSHIP AND PEER OGY 190:429-433,1957. RELATIONS. M. Lewis; L.A. Rosenblum, eds. New York, John Wiley, 1975. Novak, M.A.; Drewsen, K.H. Enriching the lives of captive primates: Issues and prob- Tripp, J.K. Increasing activity in captive orlems. In PSYCHOLOGICAL WELLangutans: provision of manipulable and edBEING OF CAPTIVE PRIMATES. E. ible materials. ZOO BIOLOGY 4225-234, Segal, ed. New Jersey, Noyes Publishing, 1985. 1989. Westergaard, G.C.; Fragaszy, D.M. Effects of Post, W.; Baulu, J. Time budgets of Macacu manipulatable objects on the activity of mulutta. PRIMATES 19:125-140, 1978. captive Capuchin monkeys (Cebus apella). Reinhardt, V. Preliminary comments of ZOO BIOLOGY 4:317-327, 1985.