Pair-bonding among nonhuman primates is rare and the possible selection pressures at work to maintain this type of social grouping have been discussed at great length (Kleiman, 1977; Wittenberger, 1980; Kinzey, 1987; Palombit, 1999; Fuentes, 1999, 2002; Chambers, 2002; Reichard, 2003; van Schaik and Kappeler, 2003). While the behavioral ecology of pair-bonded species has been relatively well studied across radiations, there are fewer studies that examine the nuances of social behavior between pair-bonded individuals and how social behavior is affected by ecological variables such as changes in food availability and feeding competition (but see Curtis and Zaramody, 1997; Bartlett, 2003; Fietz, 2003; Curtis, 2004; Schulke, 2003, 2005). This inhibits researchers' ability to fully evaluate the two main competing hypotheses, mate defense and resource defense (Wrangham, 1980; Dunbar, 1988), that have been put forward to explain the evolution of pair-bonding. Of these two sets of hypotheses, mate defense models have received more attention and empirical support (van Schaik and Dunbar,

Deborah J. Overdorff • Stacey R. Tecot • Department of Anthropology, University of Texas at Austin

1990; van Schaik and Kappeler, 1997, 2003; Palombit, 1999). The main conclusion of these studies is that mate guarding and infanticide prevention are the main forces that influence the formation of pair-bonded groups in a wide range of species (van Schaik and Kappeler, 1997; Palombit, 1999; Brotherton and Komers, 2003).

Resource defense models, however, may merit further consideration for some groups of primates. For example, in his review of pair-bonding patterns, Fuentes (2002) concluded that pair-bonds, especially among Malagasy strepsirhines, may result due to "male defense against predators and/or defense against resource competition from conspecifics and other species" (p. 958). In this scenario, female reproductive success depends upon receiving help from the male, primarily to maintain exclusive access to food resources. Male aid can also come in the form of care for infants as seen in some of the small New World Primates such as owl and titi monkeys and callitrichids (Wright, 1990; Goldizen, 2003) or siamangs (Palombit, 1996). Although solitary individuals have only themselves to feed, they are likely to be more susceptible to predation, do not gain adequate access to higher-quality food because they are likely to be displaced by social groups, and are likely to accrue little to no gain in reproductive fitness. The addition of more than one male decreases a male's individual reproductive success, increases paternal confusion, and leads to less male investment in range defense and care for infants (Wright, 1990; Goldizen, 2003). Additional females would require a larger home range area to support their increased energetic needs due to reproduction which in turn decreases the group's ability to maintain exclusive access. Pair-bonding is the result of a balance that is struck between the costs and benefits of living a solitary life or group living.

Malagasy strepsirhines in particular are an excellent model taxon to examine the nature of pair-bonding and resource defense for two reasons. First, there is a relatively high proportion of pair-bonded species within the radiation compared with anthropoids (Heymann and Kappeler, 1996; Jolly, 1998) and these species occupy a wide variety of niches. Second, Malagasy primates must cope with severe environmental challenges due to poor food quality, small food patch size, and unpredictable resource patterns, that are influenced by extreme weather patterns (Ganzhorn, 1995; Ganzhorn et al., 1999, Gould et al., 1999; Wright, 1999), and may have evolved traits such as small group sizes to maximize resources and conserve energy (Wright, 1999). As a result of the environmental challenges present in Madagascar, the potential for contest competition for resources within and between species could be quite high (see Mutschler et al., 2000; Schulke, 2003, 2005), thereby placing an upper limit on how many individuals a group can support and explaining the prevalence of pair-bonding within the Malagasy lemurs.

Several predictions can be made if pair-bonding in Malagasy lemurs is driven by resource competition and the need to defend food resources (loosely following Fuentes, 2002). First, the pair-bond will be the common group composition observed within species; these groups will be stable across reproductive seasons and periods of fluctuation in available food. Second, the exchange of affiliative behaviors (nearest neighbor association, grooming, and scent-marking) should be strongest between the adult male and adult female. Third, within-group contest for food should be minimal and rates of aggression should be low. Fourth, between-group contest for food will be more common than within-group contest for food and evidenced through conspecific and intraspecific contest over food. Fifth, both males and females are expected to be active participants in range defense. Finally, given the extreme changes in food availability patterns that have been documented in southeastern Madagascar (Hemingway and Overdorff, 1999), it is likely that within-group and between-group interactions will vary with availability patterns although within-group affiliative behaviors should remain stable (Curtis and Zaramody, 1997; Schulke, 2005).

To test these predictions, we use detailed data on social interactions within three wild groups of red-bellied lemurs (Eulemur rubriventer) from a 14-month study and monthly censuses on these groups between 1989 and 1994 (Overdorff, 1993a,b, 1996a,b), and supplement these data with comparable demographic data collected from another long-term study by Tecot (in preparation). In general, red-bellied lemurs appear to maintain small groups, are highly frugivorous, defend discrete home ranges which overlap little with other conspecific groups, and males actively care for infants (Merenlender, 1993; Overdorff, 1993a,b, 1996a,b; Durham, 2003).

First, we address the predictions outlined above by describing the social organization of red-bellied lemurs using data on group composition, affiliative behaviors (grooming, scent-marking, and nearest neighbor patterns), and agonistic patterns within and between groups. In addition, we present data on directionality of social behaviors and how adult males and females participate in interactions with conspecifics and other sympatric species. Finally, we evaluate how food availability and food scarcity patterns influence each of these variables to determine what variables might contribute to the selection for and maintenance of pair-bonding in this species.

METHODS Study Site

The study site was located in the Ranomafana National Park (RNP) region, a large (43,500 ha) southeastern rainforest in Madagascar (see Wright, 1992). RNP is located between 47"18"-47"37" and 21"02"-21"25' S and ranges from montane cloud forest (1500 m) to lowland rainforest (500 m). The 3.5 km2 site for this study, Vatoharanana, is approximately 5 km south of the Talatakely Research Station and is a high montane rainforest (altitude: 1200 m). Annual rainfall averages from 1500 mm to over 4000 mm (Overdorff and Wright, unpublished data). Ten sympatric species of prosimian primates are found in the area in addition to the study species.

Phenological patterns have been documented previously in Hemingway and Overdorff (1999) and Overdorff et al. (unpublished). Because this species is highly frugivorous (Overdorff, 1993a), it was assumed fruit availability would have the most impact on behavior. Fruit availability typically peaks between August and February and is at its lowest between March and July. The food availability period corresponds to the birth season (mid-September to mid-October) and lactation (through March) while the food scarcity period corresponds to the mating season (mid-May to mid-June) and gestation (mid-June to mid-September/October). Seasonal differences in social behavior, demography, and conspecific and interspecific aggression were examined using these two distinct categories of food availability.

Study Groups and Demography

Two groups of red-bellied lemurs were followed from dawn to dusk at least 8 days a month from July 1988 through August 1989 (1500 observation hours). A third study group was observed on an opportunistic basis. Monthly censuses on each of these study groups continued from September 1989 to December 1994 (see Overdorff, 1993a,b, 1996a,b, for further details).

As part of the long-term nature of this study, groups were censused once a month between September 1989 and December 1994. Additional information on group movements, emigrations, and immigrations was available as other researchers have worked at the site on separate projects (Strait and Overdorff, 1995; Tecot, in preparation). Data from Tecot's study which took place from September 2003 through March 2005 (4800 observation hours), were used to provide supplemental data on two groups at this site, and three groups at the adjacent Talatakely site.

Data Collection

A combination of continuous focal animal sampling, point sampling, and all occurrences sampling was used to quantify social behavior (Overdorff, 1996b). An adult male and female focal animal were each followed by one observer exclusively on each sampling day (8-10 hours in length) and focal samples were balanced between all adult individuals in each study group. In addition, individuals were marked with colored collars and pendants or radio collars. At 5-minute intervals the focal animal's nearest neighbor within 5 m was noted; if two animals were equidistant from the focal, then both animals were recorded. A sign test (Sokal and Rohlf, 1995) was used to determine independence of scan samples and it was determined that samples were independent at 30-minute intervals. These data points were then used to calculate the percent of samples individuals spent near each group member.

All occurrences of the following behaviors were noted: mutual grooming, allo-grooming, self-grooming, aggression within groups, aggression between groups, aggression between species, and all forms of scent-marking. The initiator and recipient of grooming bouts and event behaviors were identified and the context of behaviors was also recorded (feed, rest, travel). Detailed ad libitum notes were recorded on the adult male and female's behavior when group members interacted with conspecific groups or other species. Supplemental data collected by Tecot followed a similar protocol. Chi-squared tests and Mann-Whitney U tests were used to compare differences between food availability periods. Significance level was set at p < 0.05.

RESULTS Group Composition

During both studies, each study group contained only one adult male and one adult female (Table 1). With the exception of infant births which occurred during mid-September through mid-October, group compositions remained stable. During subsequent censuses we noted that neither males nor females were philopatric and all natal juveniles left social groups between 2.5 and 3 years of age. One of these juvenile females was rediscovered in June 1994 with an adult male and offspring in a home range adjacent to her natal range and was still occupying that range as of September 2003 (Tecot, personal observation).

Replacement of resident adult females by nongroup females has been witnessed four times and we have yet to observe resident males being replaced by nongroup

Table 1. Group compositions for Overdorff's study and Tecot's study

# adult # adult # infants and Group ID females males juveniles


A 112

C 113

1 113

2 112 3 111


# immigrations # emigrations # deaths

0 12

1 12

0 10

1b 1b 1a a Emigration between 2.5 and 3 years of age (natal juveniles). b Forced emigration by another female, died 2 weeks later. ? Emigrated or died.

males. In three cases, a resident female died or disappeared from her group and a new female joined the remaining male within 3 to 5 weeks. In May 1994, another female from Overdorff's study (Group II) was actively evicted from her group by a nonresident female over a 24-hour period. The first time the nonresident female had been observed was during the week prior to the eviction while she was following the group at a distance (50-100 m). On the day the resident female was evicted, the nonresident female repeatedly chased her, cuffed her, pulled her hair, and attempted to bite her; other group members did not intervene. The day following the eviction the resident female was observed on a trail outside of Group II's home range, approximately 2 km from where she had been seen last. She had sustained several bite wounds on her back and neck and her body was found 3 weeks later. Based on forensic evidence such as puncture wounds to the skull, it was presumed she had been predated on by a Cryptoprocta fossa (Overdorff and Strait, 1994).

Affiliative Patterns Nearest Neighbor Patterns

Red-bellied lemurs almost always had a nearest neighbor and were observed to be alone (i.e., no nearest neighbor within 5 m) in less than 10% of the scan samples. Males and females were alone in 9.1 and 8.7% of the scans, respectively, and offspring were rarely without a nearest neighbor (3%). Males and females were observed slightly less often with each other (42.9% of scans) than with their offspring (male-offspring 48.05%; female-offspring 48.15%) but the difference was not significant. The adult males and females in each group remained in close proximity to each other regardless of season. However, nearest neighbor patterns varied seasonally among other age and sex classes (Figure 1). During food scarcity, the adult male and female were near their offspring less often (Mann-Whitney U test, »1=11, n2=14, male-offspring Z =4.10, p<0.0001, female-offspring Z=3.61, p < 0.0003), and all individuals spent more time alone (Mann-Whitney U test male alone Z=2.85, p <0.004; female alone Z=2.41, p <0.02; offspring alone Z=3.7, p <0.0002, Figure 1).


A total of 377 mutual-grooming (MGR), 344 allo-grooming (AGR), and 839 self-grooming (SGR) bouts were recorded. Overall, rates of social grooming between adult males and females and their offspring did not vary significantly although adults tended to initiate allo-grooming toward offspring at slightly higher rates than toward each other (Table 2). Social grooming rates (mutual-groom and

How To Bolster Your Immune System

How To Bolster Your Immune System

All Natural Immune Boosters Proven To Fight Infection, Disease And More. Discover A Natural, Safe Effective Way To Boost Your Immune System Using Ingredients From Your Kitchen Cupboard. The only common sense, no holds barred guide to hit the market today no gimmicks, no pills, just old fashioned common sense remedies to cure colds, influenza, viral infections and more.

Get My Free Audio Book

Post a comment