«Item type Thesis or dissertation Authors Davis, Nicolas Citation Davis, N., Schaffner, C. M., & Smith, T. E. (2005). Evidence that zoo visitors ...»
The second prediction, that adult males would be the most frequent actors of severe and lethal aggression and that the juvenile males would be the most frequent targets was supported. Adult males were overwhelmingly responsible for severe and lethal aggression. Males committed 9 of the 11 cases of severe aggression. In addition, juvenile males were more likely than expected to be the targets of severe and lethal aggression. The proportion of aggressive incidents that resulted in severe injuries (22.0%) is high compared to data from several long-term studies of captive primates in which no severe aggression was reported (Bernstein, Williams, & Ramsay, 1983; Fuentes, Malone, Sanz, Matheson, & Vaughan, 2002; Ren, et al., 1991; Thierry, 1985; Zucker, 1994), although it is likely that minor injuries were under reported as they could only be based on direct observations. Males were responsible for all six cases of lethal aggression.
Although lethal aggression has been reported in captive chimpanzees (Pan troglodytes) (de Waal, 1986a) and captive golden lion tamarins (Leontopithecus rosalia) (Inglett, et al., 1989), generally reports of lethal aggression are rare. For example, in a two year study of a captive group of chimpanzees involving 219 conflicts only three incidents resulted in observable injuries (0.14%) (Fuentes, et al., 2002), in a four month study of two groups of captive golden monkeys (Rhinopithecus roxellanae roxellanae ), where 130 agonistic encounters were observed, no injurious aggression was recorded (Ren, et al., 1991). In a much larger study of macaques 1322 aggressive interactions were recorded in rhesus macaques (Macaca mulatta), 570 in a group of long-tailed macaques (M. fascicularis) and 682 in a group of tonkean macaques (M. tonkeana) over a 12 month period, and only 56 (2.2%) agonistic interactions with bites were recorded (Thierry, 1985). In a group of free ranging patas monkeys (Erythrocebus patas) 1353 agonistic interactions were recorded over seven months, of which 6.13% were bites (Zucker, 1994), and in a year long study of agonistic behavior in large groups of captive rhesus macaques (M.
mulatta), stumptail macaques (M. arctoides), pigtail macaques (M. nemestrina), Sulawesi black crested macaques (M. nigra) and sooty managabeys (Cercocebus atys) adult males participated least of any age-class in any agonistic encounters and seldom involved in any forms of contact aggression (Bernstein, et al., 1983).
Importantly no lethal aggression was reported in any of these studies.
Field studies however have revealed cases of lethal intragroup aggression illustrating that aggression does occur in the wild. Itani’s (1982) review of intragroup lethal aggression revealed killing occurred in 13 species of nine genera. He separated them into two categories including infanticide, which accounted for the vast number of instances, and rare events of killing among adults. Recent studies reveal that lethal intragroup aggression also occurs in white handed gibbons (Hylobates lar) (Palombit, 1993), chimpanzees (Pan troglodytes) (Fawcett & Muhumuza, 2000;
Nishida, 1996; Watts, 2004; Watts, Muller, Amsler, Mbabazi, & Mitani, 2006), capuchin monkeys (Cebus capucinus) (Gros-Louis, Perry, & Manson, 2003) and spider monkeys (Ateles spp) (Campbell, 2006b; Valero, et al., 2006). Thus, there appears to be something unusual about the intensity of spider monkey aggression that occurs in zoos, with such a high percentage of aggressive incidents which involve individuals receiving substantial and potentially life-threatening or lethal injuries.
There may also be a possible difference between the four species with general aggression much more prevalent in the A. g. geoffroyi and to a lesser extent A. g.
rufiventris. However, this may be explained by the fact they were generally kept in larger groups and so had an increased potential for aggression between conspecifics Finally, I predicted that females would be the least frequent actors of aggression. The data supported this prediction as only eight incidents of aggression were attributable to adult females, and all but one incident was categorised as minor.
Non-adult females were never the actors of aggression. This is not surprising as very low rates of female-female aggression are reported in wild populations of spider monkeys (Fedigan & Baxter, 1984; van Roosmalen & Klein, 1988), with the exception that resident females do target aggression toward new immigrant females (Asensio, et al., 2008). However, the analogous context, the introduction of a new female into an existing group, only occurred once in my study groups and did not result in any female-female aggression.
One explanation for the pattern of aggression identified is that male-male aggression in zoo-housed spider monkeys may reflect natural behaviour that occurs in wild communities. Intragroup aggression between males has been reported in the wild and recent studies from two field sites (Campbell, 2006b; Valero, et al., 2006) and indicates that male-male aggression, albeit rarely observed, may be more serious than the more commonly reported male-female aggression (Fedigan & Baxter, 1984;
van Roosmalen & Klein, 1988). Valero et al. (2006) reported on a single lethal attack, while Campbell (2006b) reported on three separate attacks, which all resulted in serious injuries and in two cases had a presumed lethal result. Campbell reported all three attacks involved a coalition of resident adult males attacking a single nonadult resident male. Such intragroup lethal aggression is surprising given that males are philopatric, are likely to be related and have the strongest social bonds. For example, males affiliate with each other more frequently (Ahumada, 1992) and travel in all-male subgroups more than other age/sex classes (Shimooka, 2005). In addition, males defend their territory during inter-group encounters (L. L. Klein, 1974;
McFarland Symington, 1990), and make joint raids into neighbouring communities (Aureli, Schaffner, Verpooten, Slater, & Ramos-Fernandez, 2006). There is also negligible sexual dimorphism in spider monkeys, which has been linked to low levels of male-male competition (Dixson, 1998; Fedigan & Baxter, 1984).
Intermale relationships among primates in intra- and intergroup competition are highly variable across species (Kappeler, 1999) but are generally shaped by female distribution and competition for fertilizations. This has typically been characterised by competition, intolerance and clear dominance relations with agonistic interactions common among males in primates (Kappeler & van Schaik, 2002). Spider monkeys are unusual in that they do not demonstrate a clear dominance hierarchy and instead form strong bonds and demonstrating well developed coalitionary behaviour (Aureli & Schaffner, 2008), a characteristic that seems to be linked with male philopatry. The recent discoveries of severe and lethal aggression however also suggest a degree of complexity in this relationship.
Further support for the finding that males are responsible for the majority of aggression in zoos was provided by questionnaire respondents. Zoo keepers indicated that the most frequent context of aggression was ‘tension’ between males. These descriptions indicated that this context represented a long-standing situation between two males in which they had repeated conflicts that were not resolved. In addition, this tension was particularly noted to occur between non-adult and adult males. The remaining contexts of aggression, feeding by zoo visitors and reintroducing group members or introducing new group members are well-established sources of shortlived aggression in a variety of primate taxa (Honess & Marin, 2006b; Hosey, 2005).
There was also a consensus by the respondents that aggression in spider monkeys differs to that of other primate species. In particular, spider monkey aggression was characterised as infrequent, involving non-adult males more often and was more severe than what the keepers observed in other primate species.
Several explanations might account for the higher than expected levels of aggression between males reported in zoological parks. Firstly, one proximate mechanism in promoting better relationships among males is the need to cooperate in defending their home territory (Aureli, et al., 2006). In a zoo setting, the absence of rivals may reduce the value of male social relationships (Aureli, Cords, & Van Schaik, 2002) and lead to a greater degree of male intolerance. A second explanation may lie in the management practices of zoos. Males are regularly relocated between zoo groups (Pierre Gay personal communication), which does not correspond to the pattern of immigration in wild groups where females disperse (Ahumada, 1992;
Fedigan & Baxter, 1984; L. L. Klein & Klein, 1971; McFarland Symington, 1990;
van Roosmalen & Klein, 1988). Such practices inevitably lead to more unrelated males housed together than would occur in the wild, which could be further a source of tension (see Eisenberg & Kuehn, 1966). The practice of relocating males and not females (Durlot & Gay, 1998; Newland, 1999) could influence rate and intensity of aggression, although this factor was not explicitly examined in the questionnaire.
While there may be management difficulties keeping groups of spider monkeys in their ‘natural’ group structure its feasibility should be investigated, particularly as maintaining animals in abnormal social groupings is a known source of stress (Honess & Marin, 2006b; Hosey, 2005; Morgan & Tromborg, 2007). A policy that would include the transfer of females, and not males, between zoos would not be unprecedented as this is the general management policy for chimpanzees in zoological parks (Carlson, 2006; Fulk, 2000).
Finally, it is well-established that captive environments may alter the behaviour of individuals relative to their wild counterparts (Hosey, 2005; Kummer & Kurt, 1965). These include variations in enclosure size and complexity (Carlstead, et al., 1999; Van Keulen-Kromhout, 1978) and management routines (Bassett & Buchanan-Smith, 2007). Specifically, Hosey (2005) highlighted that for zoo-housed primates the presence of visitors, restricted space and management practices all impact behaviour, and social factors are likely to interact with these variables. For example, the social system of a given species, which can vary in the extent of group cohesion and mating patterns (van Schaik & van Hoof, 1983), could be influenced under captive conditions. Therefore, individuals that live in social systems characterised by a high degree of fission-fusion dynamics could be particularly affected given that in the wild they have the option of leaving a subgroup to reduce conflict (Aureli & Schaffner, 2007; Rebecchini in prep). This is relevant because although most of the zoo housed groups were small with a modal group structure of one male and one female, the groups which had more males had more incidents of aggression. While a positive relationship between the density of the spider monkeys in zoos and the number of aggressive episodes was not found this could be explained by the coping strategies adopted by other primates when restricted to a confined space (Aureli & de Waal, 1997; Caws & Aureli, 2003; Judge & de Waal, 1997).
However, small enclosures do reduce opportunities for fissioning, which may influence the intensity of aggression in zoo-housed spider monkeys.
The information from this questionnaire highlights a phenomenon that has not been previously reported in zoo-housed spider monkeys although has been recently discovered in the wild. The unusually high prevalence of serious aggression reported may be linked to the current social management practices in zoos.
Collectively, the lack of rival males, housing unrelated and unfamiliar males together and having no means to simulate fission are factors that may lead to patterns of male aggression that exceed what would be expected by chance alone. These factors also indicate the importance of providing suitable housing and husbandry practices for spider monkeys. Specifically, larger and more complex enclosures that allow opportunities for individuals to temporarily separate themselves from the larger social group, in order to simulate fission events in the wild, may reduce the frequency and severity of aggression (Caws, Wehnelt, & Aureli, 2008; Wehnelt, Bird, & Lenihan, 2006). The unnatural social and physical environment in which spider monkeys tend to be kept in zoos may exacerbate the propensity for male against male aggression in the wild. Therefore, managing zoo populations of spider monkeys should also entail the relocation of females rather than males.
THE IMPACT OF SOCIAL EVENTS ON THE
HYPOTHALAMIC-PITUITARY-ADRENAL AXIS OF
ZOO-HOUSED SPIDER MONKEYS (ATELES
5.1 Introduction Primates are highly intelligent sentient beings that display a complex behavioural repertoire and require a stimulating environment (McCann, et al., 2007).
Their cognitive sophistication makes them particularly susceptible to psychological stress arising from a variety of social and environmental factors, not only through actual stressors but also in anticipation of stress (Sapolsky, 2003, 2004). It is therefore essential that the appropriate conditions that satisfy their behavioural and social needs are met when in captivity to ensure their wellbeing (Boissy, Manteuffel, et al., 2007; Chapter 1).
Reducing stress levels is one way in which welfare of captive animals can be improved (Boissy, Manteuffel, et al., 2007; Chapter 1). Although the stress response to a variety of laboratory procedures is well documented in several primate species, it is still a relatively new area of research (Honess & Marin, 2006a). To date, such studies include investigations into environmental (Crockett, et al., 2000) and social factors (Abbott, et al., 2003) and reproductive status (Setchell, et al., 2008).