Enclosure environment affects the activity budgets of captive Japanese macaques (Macaca fuscata).код для вставкиСкачать
American Journal of Primatology 70:1133–1144 (2008) RESEARCH ARTICLE Enclosure Environment Affects the Activity Budgets of Captive Japanese Macaques (Macaca fuscata) M. FIROJ JAMAN1,2 AND MICHAEL A. HUFFMAN1 1 Department of Ecology and Social Behavior, Primate Research Institute, Kyoto University, Inuyama, Japan 2 Wildlife Biology and Management Section, Department of Zoology, University of Dhaka, Dhaka, Bangladesh Individuals adapt to changes in their environment, such as food availability and temperature, by adjusting the amount of time spent in different behavioral activities. These adjustments in behavior should vary across age–sex class according to specific physiological and social needs. We studied the activity budgets of three social Japanese macaque groups inhabiting either vegetated or nonvegetated enclosures in order to compare the effects of access with vegetation, as both food and substrate on resting, feeding, grooming and moving activities over a 12-month period. Daily access to natural foods for monkeys in the vegetated enclosure seems to be largely responsible for the differences in daily time budgets of these three groups. Resting time in all three groups was longer than the time devoted to other activities. Resting and moving time in the two nonvegetated enclosures was significantly longer than in the vegetated enclosure. In contrast, feeding and grooming time was significantly longer in the vegetated enclosure. Seasonal variation in time spent feeding, resting and grooming was significantly effected by enclosure type. In all three enclosures, immatures, particularly females, spent more time feeding and moving, whereas adults spent more time resting. Significant monthly variation in time spent by age–sex class was noted only for feeding and resting. Interestingly, in the vegetated enclosure, time spent feeding on natural vegetation was equal to the amount of time spent feeding on provisioned food. This suggests that factors other than energetic and nutritional needs may be important determinants of the activity budget of the species. These results have important implications for the enrichment of captive primates and our understanding of the maintenance of activity patterns by primates in the wild. Am. J. Primatol. 70:1133–1144, 2008. r 2008 Wiley-Liss, Inc. Key words: enrichment; captivity; seasonality; age–sex differences; activity budget; Macaca fuscata; vegetated enclosure INTRODUCTION The allocation of time for multiple activities has significant effects on the survival of primates. For example, in winter months Japanese macaques (Macaca fuscata) undergo an energy crisis, and seasonal changes in their daily activities correspond to the seasonal fluctuation of food supply and atmospheric temperature [Nakayama et al., 1999; Yotsumoto, 1976]. Seasonality of time budgets can be affected by seasonal changes in food availability [O’Brien & Kinnaird, 1997; Overdorff, 1996; Strier, 1987; Sussman, 1974]. Thus, the activity budgets of primates can be highly flexible in response to seasonal variation [Vasey, 2005]. Activity budgets are related to important physiological and environmental factors such as energy balance, body size and food availability [CluttonBrock, 1974; Coelho, 1986; Harrison, 1985; Kurup & Kumar, 1993; Li, 1992; Marsh, 1981; Menon & Poirier, 1996; Watanuki & Nakayama, 1993]. Diet and habitat structure are known to affect time spent in activities because of trade-offs between obtaining r 2008 Wiley-Liss, Inc. food and the costs in energy of doing so [Altmann, 1974b; Oates, 1987; Post, 1981]. Habitat quality affects food availability and primates respond variably to changes in food availability [Defler, 1995; Li & Rogers, 2004]. In the wild, Japanese macaques decrease traveling and feeding time when temperature decreases, but when food is abundant, food distribution and feeding speed increase while time spent feeding and traveling decreases [Hanya, 2004]. Feeding is considered to have priority over all Contract grant sponsor: Graduate Studies Scholarship ‘‘Monbukagakusho’’ (Ministry of Education, Culture, Sports, Science and Technology, Japan). Correspondence to: Michael A. Huffman, Department of Ecology and Social Behavior, Primate Research Institute, Kyoto University, 41-2 Kanrin, Inuyama 484-8506, Japan. E-mail: firstname.lastname@example.org Received 22 February 2008; revised 11 August 2008; revision accepted 12 August 2008 DOI 10.1002/ajp.20612 Published online 10 September 2008 in Wiley InterScience (www. interscience.wiley.com). 1134 / Jaman and Huffman nonessential activities. Studies predict that food distribution and food quality determine activity budgets [Agetsuma, 1995; Nakagawa, 1989a; Suzuki, 1965; Watanuki & Nakayama, 1993]. Temperature influences a plant’s rate of growth and its nutritional content. This has both direct and indirect effects on the seasonality of the activity budget of species dependant upon plants for food [Dunbar, 1992]. Activity budgets in wild Japanese macaques can be affected by seasonal changes in temperature and this can vary with age and sex [Agetsuma, 1995; Jaman, 2007; Maruhashi, 1981; Nakagawa, 1989b; Watanuki & Nakayama, 1993]. It is expected therefore that age–sex class activity patterns should be different. Immatures spend more time feeding and playing [Hanya, 2003; Watanuki & Nakayama, 1993], as well as moving [Maruhashi, 1981], than other age–sex classes throughout the year. Furthermore, females spend more time engaged in social grooming and feeding, and are much more active than males. However, we still do not understand the relationship between behavioral activities and how such factors as age class, sex and environment influence the behavior of captive macaques. Our understanding about the dynamics of Japanese macaques’ activity budgets comes from free-ranging provisioned and nonprovisioned populations [Hanya, 2004; Maruhashi, 1981; Watanuki & Nakayama, 1993]. Controlled experiments (or manipulations) are an ideal method for deciphering how specific mechanisms affect behavior, but rigorous controls are often impossible under natural conditions. Thus, captive populations provide an opportunity to more closely monitor behavioral variations in response to factors such as food availability and climate. However, one criticism might be that enclosure conditions are too artificial. One possible solution is to use captive populations housed in seminatural, vegetated enclosures that closely or partially mimic natural conditions. Unfortunately, we do not know to what degree environmental enrichment benefits captive Japanese macaques or whether their behaviors approach those of wild populations in response to enrichment. Although both captive rearing and enrichment of Japanese macaques have received considerable attention, few studies have compared activity budgets of monkeys inhabiting different captive environments [Boinski et al., 1999; Kerridge, 2005]. These comparisons are important for our understanding of inter- and intra-specific behavioral variations, could be helpful in devising enrichment strategies, and may illuminate the adaptive nature of variability in temporal activity patterns in primates in general [Kinnaird, 1990; Robinson, 1986; Terborgh, 1983]. In this light it can be argued that studying the activity budgets of captive monkeys, especially those in semi-naturally vegetated enclosures, is important in understanding Am. J. Primatol. how captive macaques use both provisioned and natural foods, when both are available, and how this affects age–sex class at different times of the year. The main objectives of our research are to provide baseline data on the factors influencing variations in activity budgets between captive populations of Japanese macaques in outdoor nonvegetated and semi-natural, vegetated enclosures. We compare the results of this study with known behavioral patterns from wild populations. Here we test the following predictions based on the assumption that time spent in each behavioral activity varies according to enclosure type, age–sex class, and season: (1) the relative proportion of time spent in various activities should vary according to enclosure type; (2) in general, captive monkeys should spend less time feeding and more time resting because they have to work less to obtain abundant, high-energy food than do their wild counterparts; (3) the presence of natural vegetation in an enclosure should increase the amount of time spent feeding; (4) time spent feeding and resting should vary seasonally according to food availability to meet nutritional and energy demands; and (5) immatures should spent more time feeding and less time resting than adults in order to obtain sufficient amounts of nutrition for maturation and high-energy demands. METHODS AND MATERIALS Study Subjects and Facilities We observed three socially living groups of Japanese monkeys at the Primate Research Institute (PRI), Kyoto University, Japan. Two of these groups (Takahama and Wakasa-3) were kept in nonvegetated enclosures. Takahama troop (48–53 individuals during this study) lives in a 960 m2 outdoor enclosure with metal climbing structures, one open-side 12 m2 wooden-roofed shelter and two metal shelter boxes with an entrance from the floor for protection from adverse weather. Wakasa-3 troop (20–22 individuals) is kept in a 496 m2 outdoor enclosure with an 8 m high climbing structure located in the middle of the enclosure. The structure consists of two stories with platforms and a smaller-roofed shelter on top. There is also a 15 m long artificial stream with a small pond and a small patch of grass on the ground. The Wakasa-5 troop (32–41 individuals) is kept in a 6400 m2 naturally vegetated enclosure with three wooden platforms, 2–3 m off the ground that monkeys daily climb up and rest on. Flora in the vegetated enclosure consists of over 60 species of trees, shrubs, grasses and climbers. The four most common tree species are Eurya japonica (53.88% of total treesZ1.5 m high in the enclosure), Cleyera japonica (12.97%), Quercus glauca (10.13%) and Photinia glabra (9.18%). Combined they account for over 86% of the individual trees in the enclosure. Enclosure Environment and Activity Budgets / 1135 All study groups are provisioned daily by husbandry staff with commercial monkey chow (AS, Oriental Yeast Co. Ltd., Tokyo, Japan) to meet basic developmental and reproductive requirements. The main difference between these groups was that monkeys in Wakasa-5 enclosure are able to supplement their diet freely with natural vegetation. The other two groups are dependent almost entirely upon the commercial diet. Provisioning occurred once a day between 9:00 and 11:00. Infrequently, they were provisioned for a second time in the afternoon, and this occurred often in Wakasa-3. As a result, food remained on the ground until the latter part of the day, allowing this group to feed on provisioned food until sunset. Food was evenly dispersed in the enclosure to avoid food competition. During the study period, subjects were provisioned with monkey chow, sweet potatoes, wheat, and occasionally cucumber and bamboo shoots. This research was conducted in accordance with all national and institutional guidelines for the care and management of primates established by PRI. Temperature Data Collection Temperature was recorded automatically every 10 min throughout the year using PRI’s weather recording device placed in a central location near the enclosures. From these data, average daily and monthly temperatures were calculated for the daily study period between 7:00 and 17:00 hr. Temperature data collected from the summer months of two different years (2005 and 2006) were compared in order to see whether the maximum, average and minimum temperatures of June, July and August were similar. There was similarity in temperatures between the two years and we therefore combined the behavioral data of these two different years into one ‘‘summer season’’ for the analysis of behavioral correlates. Observations and Behavioral Data Collection Data collection began in August 2005 and continued through July 2006. From a total of 816 observation hours, we collected 3,264 focal sessions from 34 focal subjects on ten behavioral activities. Before this, six months were spent habituating the monkeys in each troop to our presence, establishing individual identification of each monkey and to identify and catalogue plants in the vegetated enclosure. In both the nonvegetated enclosures observations were made from both inside and outside. In the vegetated enclosure however observations could only be made from inside owing to the larger size of the enclosure and abundant vegetation that reduced visibility. All focal individuals were identified using a tattoo number and/or individually distinct morphological characteristic. Data were collected by M. F. J. in the form of 15 min continuous focal animal samples [Altmann, 1974a; Martin & Bateson, 1993]. We classified all focal subjects into one of two age classes: immature (male: r4 years and female: Z3 years) and adult (male: 45 years and female: 44 years). Three focal subjects were selected from each of the four age–sex classes for every group except Wakasa-3, where only one adult male was present, for a total of 34 individuals. Focal subjects were chosen for similarity of age across groups. Observations were made in four different time periods; early morning (EM: 7:00–9:30), late morning (LM: 9:30–12:00), early afternoon (EA: 12:00–14:30) and late afternoon (LA: 14:30–17:00). An equal number of samples were collected in each timeperiod of the day for each focal subject in each troop per week to facilitate monthly, seasonal and group wise comparisons of behavioral data. Two focal samples per week were collected for each subject totaling 68 samples/week for 34 focal individuals. We constructed and employed an ethogram during observations, which included feeding, foraging, moving, resting, grooming, aggression and submission, object manipulation, play, vigilance and object licking. Closely related behaviors were later combined and arranged to make seven behavioral categories (dominance interactions, feeding, grooming, moving, object manipulation and play, resting, vigilance) for easier comparisons. Eating, foraging and licking were combined into feeding behavior. ‘‘Eating’’ is defined as the intake of solid food or water into the mouth, followed by chewing and/or swallowing. ‘‘Foraging’’ is uninterrupted search for food items on the ground or in vegetation where food is typically found, often but not always, ended by ingestion of food or water. ‘‘Licking’’ is defined as licking any object or surface. ‘‘Moving’’ is defined as terrestrial or arboreal locomotion, lasting a minimum of 5 sec, from one place to another, except when searching for food or another animal. ‘‘Resting’’ starts when no other behavior occurs regardless of posture. ‘‘Grooming’’ is defined as the act of picking out, scratching or removing debris, ectoparasites or other objects from the hair or skin, regardless of whether the focal subject is grooming itself, another individual or being groomed. ‘‘Aggression’’ and ‘‘submission’’ are combined into ‘‘dominance interaction.’’ ‘‘Aggression’’ is defined as any aggressive physical contact, gesture or vocalization (supplanting, grabbing, hitting, chasing, biting, stare threat, head bobbing, threat bark, etc.) typically directed toward a subordinate individual. ‘‘Submission’’ includes any submissive behavior, gesture or vocalization (retreat, avoidance of eye contact, crouching, grimace, cry, etc.) in responses to aggressive behaviors received from a more dominant individual. ‘‘Object manipulating’’ is a behavior in which monkeys pick up objects such as stones, sticks, tree branches, etc. and manipulate them with their hands. ‘‘Playing’’ is defined as jumping, running, Am. J. Primatol. 1136 / Jaman and Huffman hanging on, mounting on the back of mother or another individual or grasping others in a nonthreatening context. ‘‘Vigilance’’ is defined as watching something for more than 10 sec while not engaged in other actions. and tests were done using SYSTAT 11 and SPSS [version 13.0]. RESULTS Data Analysis General Differences in Activity Between Groups Normality and equal variance tests on the data failed and therefore analyses were done using square-root n10.5 transformed data [Sokal & Rohlf, 1969]. We analyzed the duration (time in seconds) of behavioral category (see description above) with multivariate analysis of variance (MANOVA) using age–sex class (adult male, adult female, immature male and immature female), month (12 months), enclosure (Takahama, Wakasa-3 and Wakasa-5) and their interaction as explanatory variables. Post hoc pair wise contrasts were then used to compare the average time spent on each behavioral category between enclosures using the MANOVA main model error term as there were no significant differences in the variance between the enclosure populations. For the four most frequent behaviors (feeding, resting grooming and moving), post hoc pair wise contrasts were also used to compare time spent for each behavior across age–sex class in each group. We examined the relation between daily temperatures and time spent on each activity per group and per age–sex class using a Pearson correlation. Finally, we analyzed time spent feeding using analysis of variance (ANOVA) with enclosure type (vegetated, nonvegetated) and food type (provisioned, natural food) as explanatory variables. Post hoc pair wise contrasts were used to compare the average time spent feeding on provisioned and natural foods within and between enclosure types. Statistical significance for all analyses was set at Po0.05, There were significant differences in the average time spent between the seven behavioral activities regardless of the month in the year of the study (F6, 1848 5 1777.54, Po0.0001), and there was a significant interaction between behavioral activity and group (F12, 1848 5 43.29, Po0.0001) (Table I and Fig. 1). Post hoc pair wise contrasts show that there was no significant difference between each group regarding time spent in dominance interactions or vigilance activity. In addition, there were no significant differences in time spent grooming and object manipulation and play between the two groups in nonvegetated enclosures. However, for each of these behavioral categories, there were significant differences between each of the two nonvegetated enclosure groups and the vegetated enclosure group (grooming: Takahama vs. Wakasa-5, F1,1848 5 7.99, P 5 0.005; Wakasa-3 vs. Wakasa-5, F1,1848 5 13.58, P 5 0.0002; Object manipulation and play: Takahama vs. Wakasa-5, F1,1848 5 11.29, P 5 0.001; Wakasa-3 vs. Wakasa-5, F1,1848 5 12.81, P 5 0.005; Resting: Takahama vs. Wakasa-5, F1,1848 5 59.94, Po0.0001; Takahama vs. Wakasa-3, F1,1848 5 10.75, P 5 0.001; Wakasa-3 vs. Wakasa-5, F1,1848 5 104.11, Po0.0001). Time spent feeding was significantly different between all three groups (Takahama vs. Wakasa-3: F1,1848 5 12.48, Po0.0003; Takahama vs. Wakasa-5: F1,1848 5 275.12, Po0.0001; Wakasa-3 vs. Wakasa-5: F1,1848 5 126.56, Po0.0001), but the two nonvegetated groups spent less time TABLE I. MANOVA Result for the Effect of Group Type, Month and Age–Sex Class on Activity Budgets (Time Spent in Seconds by Focal Animal) Factor Activity Group Month Age–sex Groupactivity Monthactivity Age–sexactivity Monthtroop Age–sextroop Monthage–sex Monthtroopactivity Monthage–sexactivity Age–sextroopactivity Age–sextroopmonth Monthtroopage–sexactivity Error Statistically significant P values appear in bold. Am. J. Primatol. DF MS F P 6 2 11 3 12 66 18 22 6 33 132 198 36 66 396 1,848 101,502 13.02 245.19 2,028.93 2,472.47 421.55 4,266.08 20.21 79.27 23.06 132.89 83.23 358.43 19.56 67.62 57.1 1,777.54 0.23 4.29 35.53 43.29 7.38 74.71 0.35 1.39 0.4 2.33 1.46 6.28 0.34 1.18 o0.0001 0.79 o0.0001 o0.0001 o0.0001 o0.0001 o0.0001 0.99 0.21 0.99 o0.0001 o0.0001 o0.0001 1.0 0.01 Enclosure Environment and Activity Budgets / 1137 Fig. 1. Comparison of time spent for seven behavioral activities between Takahama, Wakasa-3 and Wakasa-5 enclosure monkeys. Bars with different letters represent significant differences in pair wise comparisons, Po0.05. Comparisons were done for each behavior separately. feeding than the vegetated group (Fig. 1). Wakasa-3 may have spent more time feeding than Takahama, the other nonvegetated group, most likely owing to the seasonal presence of a small patch of grass that grows inside the enclosure in spring and summer. Takahama spent more time moving than either Wakasa-3 or Wakasa-5 (F1,1848 5 42.6, Po0.0001; F1,1848 5 64.07, Po0.0001, for each comparison, respectively). But no significant variation for moving was found between Wakasa-5 and Wakasa-3 groups (P 5 0.53). In Takahama the movement of one group of individuals sometimes resulted in the supplanting of lower ranking family groups, which led to a chain reaction of many individuals moving around the periphery of the enclosure. In Wakasa-3, individuals or family groups were able to avoid each other by resting and interacting on different levels of the stratified climbing platform. Likewise in Wakasa-5, the dense vegetation and several wooden platforms scattered throughout the enclosure provided the opportunity for individuals to avoid such interactions. Time spent feeding, resting and moving was the only behavioral activities that varied significantly between the two nonvegetated groups. Nonetheless, the remaining analyses in this article are directed at generally comparing vegetated (Wakasa-5) vs. nonvegetated (Takahama and Wakasa-3) enclosure types to elucidate the influence of natural vegetation on the activity patterns of captive primates. Differences between Vegetated and Nonvegetated Enclosures Time spent resting in both vegetated and nonvegetated enclosures was longer than the time devoted for other activities (Fig. 2). Feeding, resting, grooming and moving accounted for approximately 90% of the activity budget of either enclosure type (Fig. 2). We focused on comparing these four most frequently observed activities between ‘‘semi-naturally’’ vegetated and nonvegetated enclosures to assess whether access to vegetation as food and as substrate affects the overall activity budget of primates housed in these conditions. The effects of group, month and age-class on the time spent on behavioral activity are summarized in Table I. Effect of group Whether the enclosure was vegetated or not significantly affects the time spent for different activities by each group (F12,1848 5 43.29, Po0.0001) (Table I). The time spent resting and moving by groups in nonvegetated enclosures (Takahama and Wakasa-3) was longer than the group in the vegetated enclosure (Wakasa-5) (Fig. 2). Conversely, time spent feeding and grooming was longer in the vegetated enclosure (Fig. 2). Effect of month Regardless of group, monthly variation of time spent in all activities varied significantly (F66,1848 5 7.38, Po0.0001) (Table I). Monthly variation of time spent and time spent by groups did not interact significantly (F22,1848 5 0.35, P 5 0.99) (Table I). In the vegetated enclosure, monkeys spent more time feeding in the spring (March–May), resting in summer (June–August), grooming in summer and moving in autumn (September–November) (Fig. 3). In spring, new leaves and grasses appeared. In the nonvegetated enclosures, feeding time peaked in winter (December–February), resting peaked in summer and moving peaked in autumn (Fig. 3). Am. J. Primatol. 1138 / Jaman and Huffman Effect of age– sex class Regardless of group, all behavioral categories showed significant age–sex class differences in time spent (F18,1848 5 74.71, Po0.0001) (Table I). Age–sex class and enclosure type significantly affect activities (F36,1848 5 6.28, Po0.0001). Post hoc pair wise con- trasts showed the significant variation of time spent feeding in all age–sex classes except between adults and immatures of both sexes in Takahama and adult males vs. adult females in Wakasa-5 group (Table II). Time spent resting in all age–sex classes also varied significantly in all groups except adult males vs. adult females in both nonvegetated enclosure groups and immature males vs. immature females in the Wakasa-3 group. In all groups, immatures of both sexes spent significantly more time feeding and less time resting, whereas adults spent significantly less time feeding and more time resting (Table II and Fig. 4). Time spent moving varied significantly only between adult females and immature females in the vegetated enclosure (Wakasa-5) and in one nonvegetated enclosure (Wakasa-3 group) (Table II). Time spent moving did not vary in any of the age–sex classes in Takahama group (nonvegetated). In the nonvegetated enclosures, however, females spent more time feeding and less time resting and moving than males (Fig. 4). Time spent grooming did not vary significantly by groups according to their enclosure type for all age–sex classes (Fig. 4), except between adult males and adult females in Takahama and Wakasa-3 (both nonvegetated) groups, between immature males and immature females in Takahama group and between adult females and immature females in Wakasa-3 (Table II). In Wakasa-5 group, grooming was only significantly different between adult males and immature males (Table II). Relation between Temperature and Activity Fig. 2. Distribution of diurnal behavioral activities of monkeys living in vegetated (Wakasa-5) and nonvegetated captive enclosures (Takahama and Wakasa-3). In the vegetated enclosure group (Wakasa-5), we found a negative correlation between time spent feeding and temperature (Pearson correlation, r 5 0.228, Po0.05, N 5 101). Resting time was not correlated with temperature in this enclosure. However, regarding the two nonvegetated enclosures, in Wakasa-3 only the time spent resting was positively correlated with temperature (r 5 0.420, Po0.001, N 5 50), whereas in Takahama no significant corre- Fig. 3. Comparison of monthly average time spent for diurnal activity budgets between vegetated and nonvegetated captive monkeys. Am. J. Primatol. Enclosure Environment and Activity Budgets / 1139 TABLE II. MANOVA Post Hoc Results for the Comparison of Time Spent in Activities Across Age–Sex Class Within Three Groups Takahama Activity Age–sex Feeding Adult~–adult # Adult~–immature~ Adult#–immature# Immature~–immature# Adult~–adult # Adult~–immature~ Adult#–immature# Immature~–immature# Adult~–adult # Adult~–immature~ Adult#–immature# Immature~–immature# Adult~–adult # Adult~–immature~ Adult#–immature# Immature~–immature# Resting Grooming Moving F1, P 1848 17.33 29.56 102.6 35.34 5.91 16.61 4.11 Wakasa-3 F1, 0.87 o0.0001 o0.0001 0.2 0.08 o0.0001 o0.0001 0.01 o0.0001 0.3 0.3 0.004 0.9 0.07 0.7 0.2 1848 17.14 33.18 32.55 12.58 31.02 22.77 6.89 31.95 12.66 Wakasa-5 P o0.0001 o0.0001 o0.0001 o0.0001 0.06 o0.0001 o0.0001 0.12 0.008 o0.0001 0.09 0.6 0.5 0.004 0.17 0.5 F1, P 1848 147.5 52.33 12.26 6.74 238.07 48.16 34.73 12.1 6.13 o0.0001 o0.0001 0.0005 0.009 o0.0001 o0.0001 o0.0001 0.39 0.2 0.0005 0.2 0.9 0.01 0.3 0.14 Statistically significant P values appear in bold. lations were found between time spent in any activities and temperature. Regardless of group, overall time spent feeding and resting was significantly correlated with temperature only in adult females (Table III). Time spent moving was only significantly correlated with temperature in immature females, whereas grooming in immature males was negatively correlated with temperature (Table III). Moreover, in the vegetated enclosure, time spent feeding, grooming and moving was significantly correlated with temperature only in adult females (r 5 0.291, Po0.01, N 5 96; r 5 0.316, Po0.01, N 5 96 and r 5 0.235, Po0.05, N 5 96, respectively). Between the two nonvegetated enclosures, time spent grooming was significantly correlated with temperature only in Takahama and only for immature males (r 5 0.251, Po0.05, N 5 96). In Wakasa-3, time spent feeding and resting was significantly correlated with temperature only in adult females (r 5 0.329, Po0.05, N 5 50 and r 5 0.595, Po0.001, N 5 50 respectively). Also in this group, time spent resting was significantly correlated with temperature only in adult males (r 5 0.437, Po0.01, N 5 50). No other activities were significantly correlated with temperature in any age–sex class. Inter- and Intra-Group Comparison of Feeding on Provisioned and Natural Foods The overall time spent feeding on provisioned food did not differ significantly by enclosure type (vegetated, nonvegetated) (ANOVA: F(1, 64) 5 0.123, P 5 0.727). The Takahama group spent a nearly equal amount of time feeding on natural foods as was spent feeding on provisioned foods (6.2, 6.3 min/hr, respectively). However, in the nonvegetated enclosure groups, considerably more time was spent feeding on provisioned foods than on natural vegetation (Takahama: 5.1, 0.6 min/hr; Wakasa-3: 7.2, 1.5 min/hr, respectively), as natural food was not available to the extent that it was in the vegetated enclosure. Taking the average amount of time spent feeding on natural and provisioned foods by the two groups in nonvegetated enclosures, (Takahama, Wakasa-3), the amount of time spent feeding on provisioned foods was highly significant (F(1, 64) 5 84.81, Po0.001). DISCUSSION Variation in the structure of activity budgets (feeding, resting, moving, socializing, etc.) is attributed to fluctuating conditions of the environment in which an animal lives. It is assumed that individuals adjust their behavior in response to these conditions based on a combination of age- and sex-specific physiological factors. Naturally, an increase in one behavior necessitates decreases in others. Kurup and Kumar  found that time devoted to feeding is inversely related to that of resting. Temporal variations in temperature, especially high midday temperatures during the summer or dry season, may cause an increase in time spent resting and a decrease in social behavior [O’Brien & Kinnaird, 1997; Post, 1981; Robinson, 1986]. Activity budgets are directly related to metabolism and reproduction, and thus vary as energy requirements change across the seasons [Halle & Stenseth, 2000]. We analyzed these relationships in three captive Japanese macaque groups living in differing outdoor Am. J. Primatol. 1140 / Jaman and Huffman Fig. 4. Age–sex class variation of feeding, resting, grooming and moving in the vegetated and nonvegetated enclosures (M—Male, F—Female). TABLE III. Correlation Between Daily Average Temperature and Daily Activity (Sec.) Budgets Across Age–Sex Classes (all Groups) Age–sex Adult male Adult female Immature male Immature female Activity Feeding Resting Grooming Moving Feeding Resting Grooming Moving Feeding Resting Grooming Moving Feeding Resting Grooming Moving Statistically significant P values appear in bold. Am. J. Primatol. Correlation (r 5 ) 0.111 0.044 0.062 0.018 0.248 0.156 0.065 0.059 0.019 0.035 0.156 0.025 0.069 0.006 0.111 0.141 Significance level (P 5 ) Sample size (N 5 obs. days) 0.085 0.492 0.335 0.784 o 0.000 0.015 0.313 0.358 0.767 0.584 0.015 0.697 0.288 0.927 0.084 0.028 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243 Enclosure Environment and Activity Budgets / 1141 enclosure types (vegetated and nonvegetated) and compared trends with the existing primate literature, in particular that of wild Japanese macaques on the sub-tropical island of Yakushima [Hanya, 2004; Marushashi, 1981]. Prediction 1 was supported by our results. As summarized in Figure 5, we found that the relative proportion of time spent in the four major activities—resting, grooming, feeding and moving—varied according to groups living in different enclosure types, reflecting differences and similarities in their respective behavioral modifications in relation to their living conditions. In both vegetated (Wakasa-5) and nonvegetated (Takahama and Wakasa-3) enclosures, monkeys spent more time resting compared with other activities. This was greater than the time spent resting by wild Yakushima macaques [Hanya, 2004; Maruhashi, 1981]. In contrast, the proportion of time spent resting by the vegetated group was similar to that of a group of wild lion-tailed macaques (33% of the day time) in the highlands of southern India [Kurup & Kumar, 1993]. In all activities but moving, the activity budget for the vegetated group was closer than the nonvegetated group to that of wild Yakushima macaques. Surprisingly, although the time spent moving by Yakushima macaques was expected to be higher, the nonvegetated groups actually spent a relatively similar proportion of time moving, and spent more time moving particularly, by Takahama group, than the vegetated group (Wakasa-5) (Fig. 5). This may be partly explained by the observation that in Takahama group the movement of dominant individuals sometimes resulted in the supplanting of lower ranking family groups, which led to a chain reaction of many individuals moving around the periphery of the enclosure. In the other two groups (Wakasa-3 and Wakasa-5), the structural or natural vegetative Fig. 5. Relative proportion of time spent in resting, grooming, feeding and moving in PRI captive groups and wild Yakushima Japanese macaques. The values for Yakushima monkeys are the mean values derived from Maruhashi  and Hanya . enrichment provided individuals with the opportunity to avoid such types of interactions, limiting this kind of movement. Maruhashi  found that the time devoted to social grooming (27.9%) by Yakushima macaques is exceptionally high in relation to other activities when compared with other primates. Time devoted to this activity was even higher than the PRI captive groups reported here (nonvegetated, 13.2% and vegetated, 16.1%). Prediction 2 was supported by our results. Monkeys in all enclosures spent more time resting than feeding (Fig. 1). Furthermore, it was found that monkeys in the two nonvegetated enclosures spent more time resting than feeding (Fig. 2). This may be a behavioral modification on their part to conserve needed energy brought about by access to provisioned foods only. Prediction 3 was supported by important differences in the amount of time spent feeding between the vegetated and nonvegetated enclosure groups. That is, monkeys in the two nonvegetated enclosures spent less time feeding than monkeys in the vegetated enclosure (Fig. 2). Agetsuma and Nakagawa  predicted that when food availability increases, monkeys increase feeding time. We found that the presence of natural food in the vegetated enclosure resulted in the monkeys spending more time feeding. Interestingly, the amount of time spent feeding by the vegetated group (27.1%) was intermediate to that of two troops on Yakushima (23.5 and 38%, Maruhashi, 1981; Hanya, 2004, respectively, see Fig. 5). In the vegetated enclosure time spent feeding on natural food was equal to the amount of time spent feeding on provisioned food, and overall they spent twice as much time feeding as did monkeys in the nonvegetated enclosures. In spite of the fact that the provisioned diet is supposed to adequately provide for basic growth and reproduction, vegetated monkeys devoted more time to feeding. The motivation to feed more may be based on factors other than nutritional or energetic deficit alone. Three possible explanations are proposed for this difference; (1) regardless of nutritional or energetic needs, when given the opportunity to feed more, captive primates continue to do so until satiated, (2) monkeys prefer a greater diversity in their diet than that available from provisioned foods alone, or (3) other nonnutritive elements missing in the provisioned diet are available in natural vegetation, and when given the opportunity, monkeys seek to fulfill these needs. Analysis of the diet’s chemical and nutritional content is needed to further evaluate these possibilities. Feeding and moving are energy-consuming activities [Coelho et al., 1976]. Food availability is expected to fluctuate seasonally as energy expenditures are expected to fluctuate with changes in temperature [Halle & Stenseth, 2000]. Japanese Am. J. Primatol. 1142 / Jaman and Huffman macaques living in the high-altitude region of Yakushima decreased time spent moving and feeding with decreasing temperature [Hanya, 2004]. At temperate Takagoyama, the time spent feeding and resting fluctuated seasonally [Yotsumoto, 1976]. In winter, when the least amount of food was available, time spent feeding dropped to its lowest levels of the year [Yotsumoto, 1976]. For Japanese macaques living in the lowland sub-tropical areas of Yakushima, time spent feeding and resting is constant between months, whereas time spent grooming increases from October to –December and moving decreases toward winter [Maruhashi, 1981]. Vasey  noted that for red-ruffed lemurs in the lowland coastal rain forest of Madagascar, time allotted to feeding and resting mainly depended on food availability. In Geoffroy’s marmosets, time spent resting and foraging showed a significant difference between the dry and wet seasons and is related to the availability of food, i.e. insects [Passamani, 1998]. Hill  also noted seasonal trends for Japanese macaques on Yakushima related to the availability of plant and insect foods. Therefore, according to Prediction 4, we expect that time spent feeding and resting should vary seasonally according to food availability to meet nutritional and energetic demands. Prediction 4 was supported by our results. We found that the time spent feeding was significantly correlated with temperature in the vegetated enclosure (Wakasa-5). However, between the two nonvegetated enclosures, time spent resting was significantly correlated with temperature only in the Wakasa-3 group. This is likely owing to the availability of more shaded resting places provided by the multi-floored tower, not found in Takahama group enclosure, allowing them to rest for longer duration during hot periods of the year. Time spent resting was greatest in summer and shortest in autumn (Fig. 3). In the vegetated enclosure, monkeys spent more time feeding in spring, when low quality foods such as young leaves and grasses were most abundant; presumably eating more to satisfy nutritional needs from available sources. In autumn, the vegetated group (Wakasa5) spent relatively less time feeding than other seasons (Fig. 3). Autumn is a period when availability of nutrition-rich foods in the vegetated enclosure increases, in particular a variety of fruits and nuts became available, e.g. C. japonica, E. japonica, P. glabra, Q. glauca. In winter, monkeys in all enclosures spent relatively more time feeding, assumedly to maintain body heat, resulting in the reduction in time spent in other activities, such as resting and moving in winter months. Adult females spent significantly more time feeding and less time resting in colder temperatures, probably owing to food availability in the vegetated enclosure and perhaps, equally important, owing to the need to accumulate fat during the early reproductive stage. Am. J. Primatol. Prediction 5 was supported by our results. We found that immatures spent more time feeding year round than did adults. This was more pronounced for the vegetated enclosure, where monkeys have access to both provisioned and natural foods at the same time. Immatures of both sexes spent more time feeding and less time resting than adults, particularly in the vegetated enclosure, throughout the year (Fig. 4). Immature monkeys spent more time feeding throughout the year, perhaps because they are more active and need more nutrients for physical development and maturation [Watanuki & Nakayama, 1993]. Juvenile Yakushima macaques spent more time playing and moving whereas females spent more time grooming and feeding, and were more active than males [Maruhashi, 1981]. Immatures may be accumulating more energy reserves for sexual maturation by spending more time feeding and less time resting, regardless of season. In the vegetated enclosure, unlike provisioned food, natural food items are small in size (fruits and flowers), sometimes few in number, and not always easy to process. Therefore, monkeys, and particularly immatures, eat constantly by extending their feeding time to satisfy energy demands [Hanya, 2004; Nakagawa, 1989b]. In the vegetated enclosure, immature individuals, particularly females, spent more time feeding on natural foods, partly because they were often chased away by adults from the site of provisioning when food was first given. Dominant individuals tend to monopolize high-quality food patches until they are satiated [Post et al., 1980]. Watanuki and Nakayama  reported that adult male Japanese macaques spent less time grooming than adult females. Similar trends are noted for males of other primate species [CluttonBrock, 1974; Smith, 1977; Waser, 1977]. The possible reasons for this are: (1) they do not face the energy costs of pregnancy and lactation; (2) in most primate species, they are dominant to females and juveniles and thus have access to the places of maximum food availability; and (3) they tend to feed faster than females and juveniles [Clutton-Brock, 1977]. In this study, we found that adult males and females in Wakasa-5 and Takahama groups spent more time in social grooming than immatures of both sexes. Young spent more time in feeding and engaged in other social interactions, such as play. In conclusion, the activity budgets of Japanese macaques in ‘‘semi-naturally’’ vegetated and nonvegetated enclosures reflect adaptations related mainly to food availability and seasonal changes of temperature, largely by adjusting time spent feeding and resting. Daily access to natural foods for monkeys in the vegetated enclosure seems to be largely responsible for the differences in daily time budgets of the two enclosure types. Activity budgets also varied across age–sex class. These data support the idea that captive groups can be used to Enclosure Environment and Activity Budgets / 1143 experimentally test behavioral adaptations to environmental conditions, and to evaluate species relevant food resources and provisioning schedules for the betterment of their living conditions. One approach, if the goal is to provide captive primates with conditions as close to those in the wild as possible, would be to experiment with the amount, type and timing of provisioning across seasons. Furthermore, we are currently investigating the nutritional properties of plant foods selected by different age–sex classes in the vegetated enclosure to verify the nutritional and potential medicinal effect of natural vegetation on the health of individuals. This is needed to evaluate and compare the differences between a traditionally provisioned diet and the one enhanced by natural vegetation. For primates in captivity, evaluating the benefits of this type of environmental enrichment concerning the physiological and mental well-being of the group is an important and essential process in the formulation and refinement of management practices. ACKNOWLEDGMENTS We express our gratitude to the Center for Human Evolution Modeling Research for maintaining the enclosure facilities and caring for the monkeys, and to the ‘‘research and animal care’’ committee for permission to conduct our research in the enclosures at PRI. We thank our colleagues Drs. A. Mori, T. Furuichi, G. Hanya, H. Sugiura, C. Hashimoto, H. Takemoto and others at PRI for their valuable suggestions and comments during the research period, oral presentations and manuscript preparation. We thank Dr. K. Zamma and Dr. A.D. Hernandez for their guidance in analyzing the data. We thank A. D. Hernandez and A. MacIntosh for reading and commenting on an earlier draft of the manuscript. We are grateful to participants of the annual conference of the Primate Society of Japan, from whom we benefited in their many constructive comments and advice. M. F. J. was supported by a Graduate Studies Scholarship ‘‘Monbukagakusho’’ (Ministry of Education, Culture, Sports, Science and Technology, Japan). Finally, we are grateful to associate editor Dr. W. D. Hopkins and two anonymous reviewers for their helpful comments. REFERENCES Agetsuma N. 1995. Dietary selection by Yakushima macaques (Macaca fuscata yakui): the influence of food availability and temperature. Int J Primatol 16:611–627. Agetsuma N, Nakagawa N. 1998. Effects of habitat differences on feeding behaviors of Japanese monkeys: comparison between Yakushima and Kinkazan. Primates 39:275–289. Altmann J. 1974a. Observational study of behavior: sampling methods. Behaviour 49:227–267. Altmann J. 1974b. Baboons, space, time and energy. Am Zool 14:221–248. Boinski S, Swing S, Gross TS, Davis JK. 1999. Environmental enrichment of brown capuchins (Cebus apella): behavioral and plasma and fecal cortisol measures of effectiveness. Am J Primatol 48:49–68. Clutton-Brock TH. 1974. Activity patterns of red colobus (Colobus badius tephrosceles). Folia Primatol 21:161–187. Clutton-Brock TH. 1977. Some aspect of intraspecific variation in feeding and ranging behaviour in primates. In: CluttonBrock TH, editor. Primate ecology. London and New York: Academic Press. p 539–556. Coelho AM. 1986. Time and energy budgets. In: Mitchell G, Erwin J, editors. Comparative primate biology. Vol. 2A. Behavior, conservation, and ecology. New York: Alan, R. Liss. p 141–166. Defler TR. 1995. The time budgets of a group of woolly monkeys (Lagothrix lagotricha). Int J Primatol 16:107–120. Dunbar RIM. 1992. A model of the gelada socio-ecological system. Primates 33:69–83. Halle S, Stenseth NC. 2000. Introduction. In: Halle S. Stenseth NC, editors. Activity patterns in small mammals: an ecological approach. Berlin: Springer. p 3–17. Hanya G. 2003. Age differences in food intake and dietary selection of wild male Japanese macaques. Primates 44:333–339. Hanya G. 2004. Seasonal variation in the activity budget of Japanese macaques in the coniferous forest of Yakushima: effects of food and temperature. Am J Primatol 63:165–177. Harrison MJS. 1985. Time budget of the green monkey, Cercopithecus sabaeus: some optimal strategies. Int J Primatol 6:351–376. Hill DA. 1997. Seasonal variation in the feeding behavior and diet of Japanese macaques (Macaca fuscata yakui) in lowland forest of Yakushima. Am J Primatol 43:305–322. Jaman MF. 2007. Comparison of activity budgets of captive Japanese macaques (Macaca fuscata) in two different enclosure environments and some aspects of their feeding ecology. Master’s thesis, Primate Research Institute, Kyoto University, Aichi, Japan. 47p. Kerridge FJ. 2005. Environmental enrichment to address behavioral differences between wild and captive black-andwhite ruffed lemurs (Varecia variegata). Am J Primatol 66:71–84. Kinnaird MF. 1990. Behavioral and Demographic Response to habitat change by the Tana River Crested Mangabey (Cercocebus galeritus galeritus) (unpublished Ph.D. dissertation), University of Florida, Gainesville. Kurup GU, Kumar A. 1993 Time budget and activity patterns of the lion-tailed macaque (Macaca silenus). Int J Primatol 14:27–39. Li ZY. 1992. Time budgets of Presbytis leucocephalus. Acta Theriol Sin 12:7–13. Li ZY, Rogers E. 2004. Habitat quality and activity budgets of white-headed langurs in Fusui, China. Int J Primatol 25:41–54. Marsh CW. 1981. Time budgets of Tana River red colobus. Folia Primatol 35:30–50. Martin P, Bateson P. 1993. Measuring behavior: an introductory guide. Cambridge: Cambridge University Press. 222p. Maruhashi T. 1981. Activity patterns of a troop of Japanese monkeys (Macaca fuscata yakui) on Yakushima Island, Japan. Primates 22:1–14. Menon S, Poirier F. 1996. Long-tailed macaques in a disturbed forest fragment: activity pattern and time budget. Int J Primatol 17:969–985. Nakagawa N. 1989a. Feeding strategies of Japanese monkeys against deterioration of habitat quality. Primates 30:1–16. Nakagawa N. 1989b. Bioenergetics of Japanese monkeys (Macaca fuscata) on Kinkazan Island during winter. Primates 30:441–460. Nakayama Y, Matsuoka S, Watanuki Y. 1999. Feeding rates and energy deficits of juvenile and adult Japanese monkeys Am. J. Primatol. 1144 / Jaman and Huffman in a cool temperate area with snow coverage. Ecol Res 14:291–301. Oates JF. 1987. Food distribution and foraging behavior. In: Smuts BB, Cheney DL, Seyfarth R, Wrangham RW, Struhsaker TT, editors. Primate societies. Chicago: University of Chicago Press. Overdorff DJ. 1996. Ecological correlates to activity and habitat use of two prosimian primates: Eulemur rubriventer and Eulemur fulvus rufus in Madagascar. Am J Primatol 40:327–342. O’Brien TG, Kinnaird MF. 1997. Behavior, diet and movements of the Sulawesi crested black macaque (Macaca nigra). Int J Primatol 18:321–351. Passamani M. 1998. Activity budget of Geoffroy’s marmoset (Callithrix geoffroyi) in an Atlantic forest in southeastern Brazil. Am J Primatol 46:333–340. Post D. 1981. Activity pattern of yellow baboons (Papio cynocephalus) in the Amboseli National Park, Kynya. Anim Behav 29:357–374. Post D, Hausfater G, MacCuskey S. 1980. Feeding behavior of yellow baboons (Papio cynocephalus): relationship to age, gender and dominance ranks. Folia Primatol 34:170–195. Robinson JG. 1986. Seasonal variation in use of time and space by the wedge-capped capuchin monkey, Cebus olivaceus: implications for foraging theory. Smithsonian Contrib Zool 431:1–60. Smith CC. 1977. Feeding behavior and social organization in howling monkeys. In: Clutton-Brock TH, editor. Am. J. Primatol. Primate ecology. London and New York: Academic Press. p 97–126. Sokal RR, Rohlf FJ. 1969. Biometry, 2nd ed., New York: WH Freeman and Company. Strier KB. 1987. Activity budget of woolly Spider monkeys or muriquis (Brachyteles arachnoids). Am J Primatol 13:385–395. Sussman RW. 1974. Ecological distinctions in sympatric species of lemur. In: Martin RD, Doyle GA, Walker AC, editors. Prosimian biology. London: Duckworth. p 75–108. Suzuki A. 1965. An ecological study of wild Japanese monkeys in snowy areas focused on their food habits. Primates 6:31–72. Terborgh J. 1983. Five new world primates: a study in comparative ecology. Princeton, NJ: Princeton University Press. Vasey N. 2005. Activity budgets and activity rhythms in red ruffed lemurs (Varecia rubra) on the Masoala Peninsula, Madagascar: seasonality and reproductive energetics. Am J Primatol 66:23–44. Waser P. 1977. Feeding, ranging and group size in the mangabey, Cercocebus albigena. In: Clutton-Brock TH, editor. Primate ecology. London and New York: Academic Press. p 183–222. Watanuki Y, Nakayama Y. 1993. Age difference in activity pattern of Japanese monkeys: effects of temperature, snow and diet. Primates 34:419–430. Yotsumoto N. 1976. The daily activity rhythm in a troop of wild Japanese monkey. Primates 17:183–204.