C 2005) International Journal of Primatology, Vol. 26, No. 6, December 2005 (
DOI: 10.1007/s10764-005-8853-y

Intergroup Relationships in Western Black-and-White Colobus, Colobus polykomos polykomos Amanda H. Korstjens,1,2,3,4,6 Estelle C. Nijssen,1,2 and Ronald Noe¨ 2,3,5 Received August 23, 2004; revision October 15, 2004; accepted November 3, 2004

We investigated patterns of intergroup relationships in western black-andwhite colobus, Colobus polykomos, in Ta¨ı National Park, Cote ˆ d’Ivoire, between 1993 and 1999. They live in one-male multifemale units, and demonstrate male dispersal and occasional dispersal by females. Solitary males and all-male bands are absent or very rare. Our aim was to investigate the function of female and male aggression during intergroup interactions. The species is particularly interesting because, in contrast to predictions from socioecological models, female aggression occurs during intergroup interactions in combination with female dispersal. Home ranges of neighboring groups overlapped considerably and groups lacked an area of exclusive access. Intergroup interactions occurred once every 6.6 observation days. Encounters were either peaceful (12%), or involved displays and threats (25%) or chases and fights (63%). Females interacted in 74% and males in 98% of aggressive intergroup encounters. We found little to no indication that male and female aggression correlated with the presence of food, importance of a location, or presence of infants or receptive females. However, females were more often aggressive during the months when the group depended strongly on seeds 1 Department

of Behavioural Biology, Utrecht University, P.O. Box 80086, 3508 TB, Utrecht, The Netherlands. 2 Ta¨ı Monkey Project, CSRS, 01 B. P. 1303, Abidjan 01, Cote ˆ d’Ivoire. 3 Max-Planck-Institut Seewiesen, Postfach 1564, D-82305 Starnberg, Germany. 4 School of Biological Sciences, Biosciences Building, The University of Liverpool, P.O. Box 147, Liverpool L69 3BX, United Kingdom. 5 Ethologie des Primates, CEPE (CNRS UPR 9010), Universite ´ Louis-Pasteur, F-67000 Strasbourg, France. 6 To whom correspondence should be addressed; e-mail: [email protected] or mandy [email protected]. 1267 C 2005 Springer Science+Business Media, Inc. 0164-0291/05/1200-1267/0


Korstjens, Nijssen, and Noe¨

1268

from Pentaclethra macrophylla. We also observed forays by males to other groups. Forays occurred on average once every 20 observation days. In 75% of the forays, the intruding male chased members of the target group. In 25% of the forays 1–3 females joined their male but females never attacked the target group. Our main study group was the target of such forays significantly more often when young infants were present in the group than when not. We conclude that female aggression between groups was related to food procurement and that male forays might be related to infanticide. KEY WORDS: colobus monkeys; dispersal; intergroup interactions; loud calls; mate and resource competition.

INTRODUCTION Intergroup relationships can strongly affect such aspects of the social system as intragroup social relationships, dispersal, and ranging patterns (Isbell, 1991; Isbell and Young, 2002; Koenig, 2002; Sterck et al., 1997; van Schaik, 1989, 1996; van Schaik and van Hooff, 1983; Wrangham, 1980). Wrangham (1980) suggested that females aggregated to defend resources against other female groups. He argued that, thanks to the inclusive fitness benefits related to kin selection, groups are most stable and individual group members are most likely to join in group efforts if females are closely related. Therefore, when intergroup resource competition is high, female group members are predicted to be philopatric. Subsequently, van Schaik and van Hooff (1983) argued that females aggregated to reduce predation (Alexander, 1974) but they agreed with Wrangham (1980) that strong intergroup contest competition would result in female philopatry. In subsequent versions of the last model, researchers also used the kin-selection argument to predict female philopatry in groups in which intragroup contest competition is high and cooperation among female group members allows them to monopolize valuable resources (Sterck et al., 1997; van Schaik, 1989). According to these 2 theoretical pathways, females disperse and forgo the benefits of philopatry only when dispersal costs are low and intergroup competition is low (Isbell, 1991; Isbell and Young, 2002; Wrangham, 1980) or both intergroup and intragroup competitions are low (Sterck et al., 1983, 1997; van Schaik, 1989). In contrast with these hypotheses, various primate species have strong intergroup aggression among females in combination with female dispersal (Temminck’s red colobus monkeys: Starin, 1991; mantled howlers: Glander, 1992; muriquis: Strier et al., 1993; red howlers: Pope, 2000). Isbell and van Vuren (1996) and Isbell (2004) suggest that occasional female dispersal can be expected, even in species in which female intergroup competition is strong, if costs of staying in the natal group are

Intergroup Relationships in Colobus polykomos

1269

high, for example, when the only breeding male in the group is closely related to a female, and costs of dispersal are relatively low, for example, when home ranges overlap extensively. Because female reproductive success in mammals is strongly dependent on access to food, female aggression during intergroup conflicts is generally seen as an indicator of strong contest competition between groups over important resources (Cheney, 1987). Contest competition generally arises over patchily distributed defendable resources of high quality (Chapman et al., 1995). This means that group contests can be expected over isolated trees or patches of a few trees that bear highly prized food and that can feed the entire group (van Schaik, 1989). Females may also show aggression during intergroup encounters because they try to fight off immigration attempts or harassment from intruding males (Starin, 1991; Steenbeek, 1996). Therefore, it is also important to understand male strategies in order to explain female behavior during intergroup interactions. Males that associate with a group of females often try to maintain exclusive mating access to them by chasing away intruding males (Cheney, 1987). Males can also defend resources or territories either in their effort to keep males from other groups at a distance or as a strategy to attract females (resource defense sensu Emlen and Oring, 1977). This reduces the effort that females need to put into resource defense. Furthermore, intergroup interactions offer a male a chance to mate with females ¨ 2004; Reichard, outside his own group (Digby, 1999; Korstjens and Noe, 1995; Sicotte and MacIntosh, 2004) or to persuade individual females to join his group (Cheney, 1987). The latter form of male behavior can be costly to the recruited females if it takes the form of female harassment or infanticide (Smuts and Smuts, 1993). Females can reduce the effects of male harassment and male infanticide by associating only with strong protector males (Steenbeek, 2000; Sterck et al., 1997; van Schaik, 1996; Watts, 1989) and to disperse when their group’s male(s) is (are) no longer able to protect them (Harcourt et al., 1976; Marsh, 1979; Sterck and Korstjens, 2000). We investigated intergroup interactions in western black-and-white colobus, Colobus polykomos polykomos, a species in which groups have 1–3 adult males and 4–6 adult females (Galat and Galat-Luong, 1985; Korstjens, 2001). In Ta¨ı (Korstjens, 2001) and in Tiwai, Sierra Leone (Dasilva, 1989; Oates, 1994), home ranges of conspecific groups of Colobus polykomos overlap extensively, intergroup relationships range from peaceful to aggressive, and males and females are active during intergroup conflicts. This contrasts to the limited involvement by females in intergroup conflicts in many other colobine species (eastern red colobus: Oates, 1977; Struhsaker, 1975; langur species: van Schaik et al., 1992; C. guereza: Fashing, 2001). Some well studied exceptions among

1270

Korstjens, Nijssen, and Noe¨

the colobines are Hanuman langurs (Borries, 1993; Koenig and Borries, 2001) and Temminck’s red colobus (Starin, 1991, 1994). In Hanuman langur species, females fight over food and food abundance predicts intergroup contest competition (Koenig, 2000). We investigated which ecological factors could predict intergroup competition of females Colobus polykomos. In contrast to predictions from the theoretical considerations of van Schaik and colleagues and Wrangham, females disperse in Colobus polykomos at least occasionally despite apparently strong inter-group contest competition. The extent of female dispersal in Colobus polykomos still requires further investigation, but in one study group of 4–6 breeding females in Ta¨ı, 2 females immigrated and the investigators thought that 3 young females emigrated between 1993 and 1999 (Korstjens, 2001). Aggression rates in the group were higher during the 2 mo preceding the disappearance of these 3 young females than afterwards and they had received relatively more aggression from females in the group than any of the other females in the 2 mo preceding their disappearance (E. C. N., unpublished report; (Korstjens et al., in press). In Tiwai, 1 femaleimmigrated or attempted to immigrate into the main study group after the end of the 1-year study period (Dasilva, 1989). If female aggression during intergroup interactions is related to food competition, then female aggressiveness should change according to the characteristics of the foods the monkeys consume. Colobus polykomos forages in trees of a wide variety of sizes and has a frugi–folivorous diet with relatively high percentages of seeds in the diet during several months of the year (Korstjens, 2001; Korstjens et al., 2002). Intragroup competition among females is stronger when they forage on food items that require a long processing time such as seeds that are concealed in hard husks, for example, those of the large tree, Pentaclethra macrophyla, than when they consume leaves (Korstjens, 2001; Korstjens et al., 2002). The trees that Colobus polykomos selects are relatively rare and patchily distributed compared to those selected by the closely related Procolobus badius badius (western red colobus) that forages in the same area (Korstjens et al., 2002). The food sources of Colobus polykomos seem to have the characteristics of resources hat evoke strong contest competition between groups (Chapman et al., 1995; van Schaik, 1989), at least during the months they feed mainly on fruits that grow on large trees. Consequently, if females are fighting over food, female aggressiveness during intergroup interactions should increase with the consumption of fruits, especially those from Pentaclethra macrophylla trees. However, if female aggression is related to male harassment of infants, we predict that female aggressiveness correlates to the presence or absence of infants.

Intergroup Relationships in Colobus polykomos

1271

We also paid special attention to agonistic interactions among males during intergroup interactions because this is the only time when adult males from 1-male units compete with other males in the species, in which most groups only have 1 or 2 adult males. A corresponding pair of predictions can be formulated for the level of male aggression during such interactions: (1) it varies with the relative importance of a location and the contestability of food sources because its sole purpose is the defense of food sources for the females and themselves. Alternatively, (2) it is related to the presence of receptive females or vulnerable infants or both because it serves as mate defense or for mate acquisition. METHODS The Study Site We conducted the study in the Ta¨ı National Park (located between ˆ d’Ivoire. The forest is clas5◦ 10 N to 6◦ 20 N and 4◦ 20 W to 6◦ 50 W), Cote sified as a tropical evergreen seasonal lowland forest (Stoorvogel, 1993). From 1995 until 1999 we measured a mean annual rainfall of 1820 ± 300 mm (±SD) and a mean daily temperature of 28.7◦ C (range 18–34◦ C). Boesch and Boesch-Achermann (2000) and McGraw (1996) have described the site in detail. Study Subjects We monitored 3 groups (Pol1, Pol2, and Pol3). We identified adult males and females (AM and AF, respectively), nulliparous females (NF), subadult males (SM), and immatures. Adults were sexually active individuals with fully developed secondary sex characteristics. AM weigh ca. 9.9 kg and AF 8.3 kg (Oates et al., 1990). It was difficult to distinguish NF from AF in size but the former had not yet reproduced. SM were the same size as AF but did not produce a complete roar. We regarded juvenile males as having reached the subadult stage when it became difficult to distinguish them from adult females in size. One female reproduced in Pol1 at approximately 5–6 yr old (She had immigrated with her mother when ca. 1–2 yr old). We characterized females in the study group as NF at the age of 3.5 yr until 6 mo before their first infant was born (average gestation time is 170 days; Harvey et al., 1987). Females and males could be distinguished from birth based on the shape of their ischial callosities. In Pol1 and Pol3 we recognized each individual on the basis of physical characteristics within a year after the start of observations on the group.

Korstjens, Nijssen, and Noe¨

1272

Nine different researchers followed Pol1, the main study group, from January 1993 to July 1999. The total group size fluctuated from 12 to 16 as a result of births, immigrations, and disappearances. The group had 2 AM and 1 SM at the beginning of the study. One AM and 1 SM disappeared in April 1996. The number of AF ranged from 4 to 6. One AF disappeared in 1993 and 2 AF joined the group in 1994. One adult female died in October 1994 and 1 disappeared in December 1996. The number of NF ranged from 0 to 4 as juvenile females (JF) matured and 3 SF disappeared in February 1998 (Korstjens, 2001). The fourth SF was the daughter of 1 of the immigrants and she reproduced in the group in November 1998. This adds up to 1 disappearance of an AM in 134 AM-mo and 1 disappearance of a SM in 45 SM-mo. We observed 3 disappearances of AF (1 confirmed death) and 2 immigrations in 453 AF-mo and 3 disappearances of NF in 194 NF-mo. Five different researchers followed Pol3 from January 1998 to July 1999. The total group size increased from 17 to 19 as a result of births. Pol3 had 1 AM, 6 AF, and 2 NF during the study. We had only partly habituated individuals in Pol2. Three different observers followed the group in 1999, and it had ca. 16 individuals. The group had 1 AM, 2 SM, 4 AF, and 5 NF. Because we followed Pol2 and Pol3 less frequently, we do not describe their intergroup interactions.

Data Collection On most observation days (≥65%), we followed the groups from 07:00 h until 17:30 h before 1998 and until 18:00 h thereafter. On other days, we followed the groups from the moment we located them until the evening or the moment when the observer lost contact with the group. Two observers collected data on behavior of individuals by instantaneous sampling (Martin and Bateson, 1993). Scan samples had a maximum duration of 20 min and we collected them at 1-h intervals. We recorded the first activity that lasted more than 5 s for each individual in sight. We used individual recognition to avoid sampling individuals twice during one scan. When an individual was foraging we noted the specific name of the tree or liana, its diameter-at-breast height, and the item eaten. Every hour on the hour, we mapped the location of the center of mass of the groups onto a grid system of 2 km2 with cells of 1 ha. We calculated the relative importance of a grid cell as the number of times a group was recorded in that grid cell on the hour divided by the total number of hourly observations in that time period. In addition, we determined the relative importance of a grid cell as a feeding location by calculating the percen-tage

Intergroup Relationships in Colobus polykomos

1273

of feeding scans recorded in the cell in a specific time period (in this case the positions of individuals plotted on the map). We assumed data on the relative importance of a grid cell in a specific month to be reliable when ≥100 locations were recorded during that month, i.e., ca. 10 observation days. All observers recorded every interaction between individuals of different groups that they observed. We recorded an intergroup interaction (IGI) when we observed members of different groups at <50 m from each other (Fashing, 2001; Oates, 1977; Stanford, 1991; Steenbeek, 1999). IGI include encounters between entire groups, that is, intergroup encounters (IGE) and encounters between a group and some members of another group, that is, participants in forays. We classified IGE into 3 categories according to the observed behaviors: aggressive, display, and peaceful encounters. During peaceful IGE, both groups either ignored each other or intermingled peacefully. During display IGE, groups did not mingle and members of both groups displayed to each other or we heard loud calls or both. During aggressive IGE, we observed chases or even fights between members of different groups. To minimize the effect of inter-observer differences, we analyzed data collected by Korstjens and Noe¨ between December 1996 and November 1998 (99 observation days from dawn to dusk representing 990 point samples on the hour) separately. During observation days of ¨ Noe/Korstjens, the number of forays and peaceful intergroup encounters recorded was higher than that during observation days by other observers but the relative involvement of females and males was comparable between observers. It is unclear whether this difference is the result of IGI missed by the other observers in the dense forest, or to a higher rate of IGI during the ¨ Therefore, both the values from the observation period of Korstjens/Noe. ¨ Noe/Korstjens data set as well as those from the data set of other observers are provided. To measure tree density, Korstjens and C. Kremer laid out 3 line transects 1 km long (north–south) and 25 m wide through the middle of the home ranges of the study groups (C. Kremer, unpublished report). We divided each transect into quadrants of 25 × 25 m. For every tree (girth >20 cm) or liana (girth of the largest stem >10 cm) in each quadrant we recorded the girth at breast height and the specific name. The median girth at breast height of trees that Colobus polykomos used is 138 cm (Korstjens, 2001). Three pairs of local assistants measured tree phenology every 2 wk between 1993 and 1999; 2 of them were working simultaneously from 1995 to 1999. We located 3 (1993–1997) – 5 (1997–1999) mature individuals of each food tree species within the study groups’ home ranges. For each tree, observers studied the presence of fruit (ripe or unripe), leaves (young or mature), and blossom (flowers or flower buds).

Korstjens, Nijssen, and Noe¨

1274

Observers recorded the percentage of the crown in which each of these food items was present on a 4-point scale: 0–25%, 25–50%, 50–75%, and 75–100%. We predicted that male mate competition would be stronger when the group contained potential mates, that is, females that were not pregnant and were not nursing infants, but had regular cycles.We regarded a female to be cycling during the 7th–4th mo preceding the birth of her infant. This is a very conservative measure based on an average gestation time of 5.6 mo (Harvey et al., 1987). All statistical tests are nonparametric and 2-tailed, with the α-level set at 0.05. When we conducted multiple tests on the same data set, we calculated α using the sharper Bonferroni correction method (Hochberg, 1988). We performed most tests using SPSS 11.0 (following instructions in Siegel and Castellan, 1988). When applicable, we calculated the expected values in a chi-square test on the basis of the number of observation hours collected in a specific time period. In analyses in which aggressiveness of individuals is tested, we compared the aggressive IGE to the peaceful IGE because they formed 2 clearly distinct extremes, and ignored IGE with only displays. RESULTS Male and Female Behavior During Intergroup Encounters We recorded 83 IGE involving group Pol1 (Table I). 63%of them were aggressive, 12% involved displays only, and 25% were peaceful (Table I). Of the 15 IGE that Korstjens and Noe¨ recorded, 33% were aggressive, 27% involved displays only, and 40% were peaceful. We observed 1.02 IGE/100 h, or ca. 1 every 10 observation days. Korstjens and Noe¨ recorded an IGE once every 6.6 full observation days (n = 99 days) or 1.5 IGEs/100 h while other observers recorded 0.95 IGEs/100 h. AM were involved in all displays and in all but one aggressive IGE. During IGE that involved only displays, males threatened the other groups with stiff-leg posture (sensu Marler, 1972), staring, bouncing around in the trees, or sometimes roaring. During aggressive IGE, the males of either group chased and fought one another and sometimes the males chased females belonging to the other group. AF chased and fought in 76–80% of aggressive IGE and threatened or roared in 67% of IGE with only displays (Table I). Females never fought alone during IGE with female participation. At least 2, but usually the majority of the females—AF and NF— joined in the aggression. This made it hard to determine who fought whom, but we did not observe females fighting males.

Intergroup Relationships in Colobus polykomos

1275

Table I. Summary of different types of intergroup encounters (IGEs) and relative male (M) and female (F) involvement

Total no. IGEs aggression Total no. IGEs displays Total no. IGEs peaceful Total no. IGEs F-displays/all IGEsa F-aggression/all IGEsa M-displays/all IGEs M-aggression/all IGEs No. observation-hours Total no. forays Exc. with only M aggression Exc. with M + F aggression Exc. with only M displays Exc. with M + F interm. displays

All IGI

Food

Rec. AF

Inf. ≤6 mo

52 10 21 83 4/81 38/81 6/83 51/83 8147 16 10 2 2 2

12 1 2 15 0/14 10/14 1/15 12/15

21 3 10 34 1/32 13/32 3/34 21/34 3556 6 3 1 1 1

39 7 16 62 2/57 27/57 7/62 38/62 5871 16 10 2 2 2

3 3 0 0 0

Note. M: male; F: female; IGE/I: inter-group encounter/interaction; Food: food recorded at IGI-site; Rec. AF: receptive AF present in Pol1; Inf. ≤6 mo: infant of ≤6 mo present in Pol1. no. of IGEs for which we recorded female involvement.

a Total

Aggression During Intergroup Encounters Relative to Their Location We considered the IGE to involve a food source, when either of the groups was feeding or sitting in a tree that they had fed in during the preceding hour. Females interacted in 71% and males in 86% of these encounters (n = 14) versus 47% and 76%, respectively, when food was not at stake (χ21 (females) = 2.6, p = 0.11; χ21 (males) = 2.1, p = 0.15, respectively; Table I). The home ranges of Pol2 and Pol3, that is, 2 of the 5 neighboring groups, covered the home range of Pol1, 114 ha, for 47%. IGI occurred throughout the range of Pol1 (Fig. 1). IGE with the highest level of aggression by females (n = 28) took place in grid cells that the study group visited significantly less than those (n = 24) in which the females did not interact (Mann Whitney U: U = 202.5, p = 0.014, α = 0.017; Fig. 2a). There is no significant relationship between female aggression during IGEs and the frequency with which the group fed in the quadrant where the encounter occurred (nno F interactions = 18 and nF−aggression = 26, U = 193; p = 0.31; Fig. 2a). Nor is there a significant relationship between male aggression during IGE and the frequency of visits to a cell or the frequency with which the group fed in that cell (nno M interactions = 15 and nM−aggression = 36, U = 241.5, p = 0.56, and nno M interactions = 11 and nM−aggression = 32, U = 174, p = 0.95, respectively; Fig. 2b). Further, there is no significant interaction between aggression during encounters and the importance of

1276

Korstjens, Nijssen, and Noe¨

Fig. 1. Home ranges of Pol1 (demarcated with a thick line), Pol2 (demarcated with a thin line), and Pol3 (demarcated with a double line) and the location of IGIs involving Pol1 (demarcated with ∗ for 1 or more IGIs), plotted on the grid system in the research area (1-ha cells).

cells based on lumping all data (grid cell visitation rates based on all observations throughout the study: females: nno F interactions = 34 and nF−aggression = 38, U = 584, p = 0.48; males: nno M interactions = 22 and nM−aggression = 50, U = 470, p = 0.33) or on the monthly rate lumping the data for corresponding calendar months of different years (females: nno F interactions = 24 and nF−aggression = 28, U = 536, p = 0.21, males: nno M interactions = 15 and nM−aggression = 36, U = 514, p = 0.66). The number of cases in the tests varies because female involvement was uncertain in some IGE. Further, in the tests that assess data from each observation month separately, we included only those months in which the group was observed for ≥100 h. In addition, we collected data on diet until 1998, not 1999. We excluded from the analyses IGE in which individuals only displayed to each other.

Temporal Distribution of IGE The number of IGE per month was not equally distributed over the calendar months (χ211 = 27.7, p = 0.004; α = 0.025; Fig. 3). An increase in the IGE-rate can be a direct result of an increase in the chance of encountering other groups rather than avoidance or attraction between groups. Indeed, there is a positive association between IGE-rate and average dayjourney-length (n = 34, rs = 0.545, p = 0.001; Fig. 4). The IGE rate does

Intergroup Relationships in Colobus polykomos

Fig. 2. The % of location points (median, 5th, 10th, 25th, 75th, 90th, and 95th percentiles) or feeding scans recorded in a grid cell in which an IGE occurred plotted against the level of female (F: a) and male (M: b) involvement in the IGE.

1277

1278

Korstjens, Nijssen, and Noe¨

Fig. 3. Average monthly intergroup encounter rate, for all IGEs and only IGEs with aggression, and average monthly rainfall (between 1993 and 1999).

not correlate to monthly rain (n = 37 mo, rs = −0.2, p = 0.2) or percentage of fruit in the diet (n = 37 mo, rs = 0.27, p = 0.1). The lag between rainfall and visible effect on the vegetation is often several months. Therefore, we also calculated the correlation with rainfall 2 mo before the encounter, but still there is no significant relationship (n = 37, rs = 0.20, p = 0.2). The proportion of IGE that was aggressive does not vary significantly between mo (χ211 = 7.5, p = 0.76; Fig. 3). To reveal seasonal patterns in the proportion of IGE in which males and females were aggressive, we summed the data for corresponding calendar months over all years of the study (i.e., n = 12) because the frequency of IGE per month was too low to allow for an analysis for each year separately. Neither the proportion of IGE with female nor the proportion of IGE with male aggression correlates with the amount of rainfall per month (rs = 0.24, p = 0.5 and rs = 0.14, p = 0.7 respectively) or with the percentage of fruit in the diet (rs = 0.09, p = 0.8 and rs = −0.09, p = 0.8). The proportion of female aggression, but not that of male aggression, correlates with rainfall 2 mo earlier (rs = 0.73, p = 0.007 and rs = 0.51, p = 0.092, respectively) as well as with the estimated quantity of fruit in the Pentaclethra macrophylla in the phenology data set (rs = 0.68, p = 0.015

Intergroup Relationships in Colobus polykomos

1279

Fig. 4. IGE rate in a specific month with ≥100 observation hours and in each calendar month throughout the years, plotted against the average day-journey length in that calendar month.

and rs = 0.38, p = 0.2 respectively; α = 0.017 for all tests on females). Female aggressiveness was more than expected in July–December, when P. macrophylla was fruiting (Fig. 5). However, during September, in the middle of the season, female aggressiveness was temporarily down again to the level before the fruiting season (Fig. 5). The minimum diameter-at-breast height (DBH) of P. macrophylla used by the monkeys is 31 cm (mean ± SD 171 ± 71.3 cm). We counted 2.7 P. macrophylla ha on our transect, and 2.1 trees/ha of ≥30 cm DBH. P. macrophylla constituted 0.36% of the total number of trees recorded on the transect. Effect of Presence of Infants and Receptive Females There was no effect of the presence of infants or potentially cycling females in the study group Pol1 on male or female aggressiveness during intergroup encounters (Table I). Males were aggressive in 38 IGE in which Pol1 had ≥1 young infant ≤6 mo old (χ21 = 0.06, p = 0.8) and in 21 IGE in which a cycling female, (i.e., one that gave birth 4–7 mo later), was present (χ21 = 0.12, p = 0.7). Only 3 IGE occurred ≤10 days from an

Korstjens, Nijssen, and Noe¨

1280

Fig. 5. Monthly variation in the proportion of IGE that contained female (F) or male (M) aggression, average rainfall, and the average fruiting score for the Pentaclethra macrophylla trees.

observed copulation (1997–1998 data only). In one case the male was aggressive. Females were aggressive in 27 IGE in which Pol1 had ≥1 young infant ≤6 mo old (χ21 = 0.79, p = 0.4). Forays Forays occurred once every 20 days on average (based on 6 events ob¨ All observers together recorded 16 forays. served by Korstjens and Noe). Members of another bisexual group visited the study group Pol1 12 times and members of Pol1 paid a visit to another group 4 times. We observed chases or fights during 10 visits to Pol1 and 2 visits by Pol1. During the other forays the visiting individuals approached the group and watched it from a short distance without attacking or being attacked. The group members that did not join in the foray continued their activities apparently undisturbed at a distance of ≥100 m from the visited group. Females joined their AM during a foray in 2 of the 12 aggressive forays and 3 of the 4 forays without aggression (Table I). Female Ma, which had a 10-mo-old infant (not yet weaned), and females Ta and Ir, which each had

Intergroup Relationships in Colobus polykomos

1281

a ±1-mo-old infant, accompanied the male on a foray without aggression in November 1997. Female Sa, which had a juvenile offspring and was not pregnant (she gave birth in August 1998), accompanied the male on a foray without aggression in December 1997. Female Ma, which had just given birth in January 1999, accompanied the male on an aggressive foray to Pol3 then. For the other foray the identity of the female was not recorded. When the incoming AM chased females, they fled without fighting until the AM of their own group chased off the intruding AM. On one well documented occasion, the members of the group an AM was visiting did not detect his presence until he attacked a female (without offspring) that was foraging at ≥10 m away from the resident male. She fled while the resident male chased the intruder away. All 16 forays occurred when ≥1 infant of ≤6 mo was present in Pol1 (binomial test on 12 forays to Pol1: expected probability based on observation time for each category, 2276 h without and 5871 h with infants: p = 0.02). We did not record any direct attack on infants. Only 5 of the 12 forays to Pol1 occurred when a female in the group could be receptive (binomial test: expected probability based on 4319 h without and 3556 h with such a female: exact p = 0.2; Table I). Only 3 forays occurred while one of the groups was at a food site; each visit was aggressive and only males executed them (Table I).

DISCUSSION Female involvement in intergroup conflicts is generally taken to indicate strong contest competition between groups over food (Cheney, 1987). Female Colobus polykomos are regularly involved in intergroup conflicts (Dasilva, 1989; this study), while C. guereza and C. vellerosus are rarely involved in such conflicts (Marler, 1969; Oates, 1977; Sicotte and MacIntosh, 2004). Two detailed studies on the last 2 species indicate no relationship between female aggressiveness and resource characteristics (Fashing, 2001; Sicotte and MacIntosh, 2004). At Ta¨ı female aggressiveness varied seasonally whereas we detected no such pattern in male aggressiveness. The seasonality in female aggressiveness was related to rainfall 2 mo before the encounter and it was highest when an important food source, fruits of Pentaclethra macrophylla was available. The seeds of this legume have a high protein-to-fiber ratio and high oil content (Dasilva, 1994; Sicotte and MacIntosh, 2004). Intragroup aggression was higher when the monkeys fed on the pods, which require long handling times (Korstjens et al., 2002). The Pentaclethra macrophylla that the monkeys used were large, and the entire group generally fed in 1

1282

Korstjens, Nijssen, and Noe¨

or 2 (intertwined) trees. According they are large defendable resources that provide high-quality food, and thus they fit the characteristics of resources that are thought to evoke contest competition between females of different groups (van Schaik, 1989). Therefore, the general prediction that defendable resources are the cause for intergroup contest competition among females is supported by the observation that females were more likely to be aggressive when fruits of Pentaclethra macrophylla were an important part of the diet. The observed negative correlation between female aggressiveness and the use of a grid cell is difficult to explain. Possibly, the lesser used cells were those that contained more temporally variable small resources or the monkeys used them less because other groups used them intensively. Our analyses revealed no link between female aggression and the presence of infants in the group. Females always fled when an intruding male attacked them. Female Colobus polykomos in our study site could probably depend on resident males to protect their offspring against harassment by intruding males because (1) male tenure in Pol1 was ≥7 yr (≥1 yr in the other study groups) and 8 of 10 of the groups in the population had only one adult male (Korstjens, 2001), so the dominant male was the most likely father of the offspring; and because (2) a lack of all male bands in our study site and the rare occurrence of multimale groups, harassment generally came only from a single individual that could be easily chased off by the adult male in the group single-handedly. During the study period we observed only a solitary male twice and no all-male group. Considering the number of observers in the study area, including researchers studying other primate species), it is safe to say that solitary males and all-male groups are very rare. Male involvement in intergroup encounters is generally a male mate defense or male resource defense or a combination of the 2 factors (Cheney, 1987). At Ta¨ı, male Colobus polykomos were more active during intergroup encounters than females were Males were not more likely to be aggressive when a potentially receptive female or young infant was present. Female receptivity is not seasonal in the population and females have no external sign of receptivity other than proceptive behavior (Korstjens (unpubl. data)). Therefore, a male is probably not able to detect receptiveness in females of another group unless he gets very close. Likewise, Sicotte and MacIntosh (2004) did not find a relationship between male aggressiveness and the occurrence of copulations on a given day in Colobus vellerosus. Fashing (2001) found a link between male aggressiveness, but not female aggressiveness, and importance of a location for foraging in

Intergroup Relationships in Colobus polykomos

1283

Colobus guereza. Sicotte and MacIntosh (2004) detected no difference in male aggression in grid cells in which the group fed for >5% versus <5% of its feeding time in Colobus vellerosus. At Ta¨ı, male aggressiveness did not increase with importance of a location or the presence of food, and it did not differ significantly between months. Therefore, although males were the main defenders in the group, our analyses could not reveal what they were defending. An important conclusion from studies on Colobus polykomos has to be that this is yet another species in which females are aggressive during intergroup interactions and females, at least occasionally, leave their natal groups (Koenig, 2002). This is in contrast to the original hypotheses on female social relationships in primates (Isbell and Young, 2002; van Schaik, 1989; Wrangham, 1980). The dispersal and behavior of Colobus guereza generally fit the pattern predicted for species in which mothers are “incomplete suppressors” in the model proposed by Isbell (2004). One aspect of this category of species is that home range expansion in response to increasing group size is costly because the females depend on knowledge of their home range. Isbell (2004) stated that goal-directed travel patterns are indicators of strong dependence on knowledge of the home range. Colobus polykomos travels through the area in a less constant travel speed than Procolobus badius, which supports the general impression in the field that Colobus polykomos generally travels in a goal-directed fashion between food sources while P. badius forages in a more paripatetic manner (Korstjens et al., 2002). They may depend on knowledge of the home range because this allows them to minimize energetic costs. Dasilva (1992) suggested that Colobus polykomos is a low-energy strategist. Another reason for limitations to an increase in the number of females per group may be that larger groups involve increased risk of male harassment or infanticide because larger groups may attract males (Isbell, 2004). The last argument is probably true only if the minimum group size that is set by predation risk has been reached. A second aspect of this category of females is that they receive targeted aggression from other females before they disperse. The costs of dispersing for young females in this category are predicted to be high because they depend on knowledge of the home range. Isbell (2004) predicted for this category that home range overlap is minimal. If home range overlap is minimal, females cannot move from a social group while remaining in a home range that is at least partly familiar (Isbell and van Vuren, 1996), which raises the costs of dispersal for them unless they can join females that have knowledge of the area. In Colobus polykomos home range overlap was extensive; this could mean that dispersal has lower costs for females in this

1284

Korstjens, Nijssen, and Noe¨

population if females move to neighboring groups. However, the 3 females that disappeared from our group were not found in any of the neighboring groups that share the home range with our study group. Third, Isbell (2004) predicted that because of the costs of home range expansion in response to increasing group size, young females will be forced to leave the group and targeted aggression will occur. The increased aggression toward the 3 young ¨ unpublished females before they disappeared from our study group (Noe, report; Korstjens et al., in press) supports this idea. For species that fit these characteristics, Isbell (2004) predicts that groups consist of small kin-groups from which females disperse occasionally because home range expansion or other costs of increased group size do not allow all females to remain in their natal group. At Ta¨ı, 2 of the 5 adult females were immigrants that immigrated when the number of females in the group was relatively low (4 resident females (1) of which died a few mo later). Dasilva (1989) also reported female immigration and intergroup conflicts between females in Colobus polykomos in Sierra Leone. There are indications of occasional female dispersal also in, Colobus guereza (Fashing, 2001) and C. satanas (Fleury and Gautier-Hion, 1999), though females generally seem to stay in their natal groups and female aggression between groups is rare in them. Another important observation in the Ta¨ı study is the occurrence of regular interactions between a group and only 1–4 individuals of another bisexual group. Similar forays or intrusions by males from bisexual groups were described for Presbytis thomasi, by Steenbeek (1999, 2000), Colobus. vellerosus in Ghana by Sicotte and MacIntosh (2004), and C. guereza in Uganda by Oates (1977) Only subadult males performed the intrusions in the population. Male forays in Presbytis thomasi and Colobus vellerosus often resulted in attacks on infants and, although not necessarily associated with the forays, infanticide occurred in both populations. Infanticide was also occurred in Colobus. guereza in Kibale, Uganda, after a male immigrated into a group (Onderdonk, 2000) and at least once during an intergroup interaction (Harris and Monfort, 2003). Steenbeek (2000) explained the forays as a strategy of males to advertise their strength to females from the visited group and to demonstrate the inability of the resident male to protect the females and their offspring. If a male is able to harm infants or females, the females may decide to leave the resident male and join the intruding male. Forays into Pol1 occurred when they had infants, suggesting that males targeted females with infants or the infants themselves. However, we never observed direct attacks on infants. In our limited, analyses there is no indication that the presence of food influenced on the frequency of forays.

Intergroup Relationships in Colobus polykomos

1285

What distinguishes the observations at Ta¨ı from those in other species is that females sometimes accompanied their male during a foray. The females we recognized during forays were multiparous females at varying stages of the reproductive cycle: pregnant, with a young infant, or potentially receptive. Our limited data set did not allow us to draw conclusion on the function of female involvement in the forays. Like Sicotte and MacIntosh (2004), we propose that at least some of the forays may function to sample the strength of neighboring groups or to intimidate neighboring groups. This hypothesis would be supported if the outcome of a foray influences the outcome of a subsequent intergroup encounter between the same 2 groups. In conclusion, we observed that Colobus polykomos differ from other Colobus spp. because in the latter females from different groups regularly fight and females occasionally disperse (Oates, 1994). This does not fit well with the predictions from original socioecological models on primate sociality (Isbell and Young, 2002). This issue has been raised by Isbell (2004), Koenig (2002), Pope (2000) and Starin (1994). We found some indications that the female intergroup conflicts were most likely when resources were contestable. We think that female Colobus polykomos disperse only when the costs for remaining in their natal groups, that is, inbreeding, are simply too high for females to bear, despite potential advantages to remaining with kin and the costs of dispersing (Koenig, 2002). The fact that the females that immigrated and the females that disappeared did not do so alone suggests that there may still be advantages of living with kin. Furthermore, the 3 females that disappeared were >6 yr old, while the 2 females that reproduced in the group did so before 6 yr. Which indicates that the females that disappeared seemed to have postponed reproduction by remaining in their natal group after they reached sexual maturity. Male Colobus polykomos were the main defenders of the group as they are in other Colobus spp. (Oates, 1994). Also in accordance with other studies on black-and-white colobus species (Oates, 1994; Fashing, 2001), male aggression could not be related to the presence of receptive females or infants in the group. We found no indication that males were more likely to be aggressive in relation to the importance of resources as Fashing (2001) observed in Colobus guereza. Therefore, our analyses do not allow us to detect any link between male aggression during encounters between entire groups and mate competition, male resource defense, or male strategies to attract or coerce females. Confrarily, encounters between a few individuals of one group and all individuals of another group, were more likely to occur when infants were present in the visited group, which this suggests that this

Korstjens, Nijssen, and Noe¨

1286

type of aggression might be related to harassment of infants or females with infants.

ACKNOWLEDGMENTS ` d’Enseignement Superieur ´ We thank the Ministere et Recherche Sci` d’Agriculture et Resources Animales, the Cenentifique, the Ministere tre Suisse de Recherche Scientifiques, the P.A.C.P.N.T., and the Centre ˆ d’Ivoire for support and permission de Recherche en Ecologie in Cote to conduct research in the Ta¨ı National Park. The Max-Planck Institut ¨ Verhaltensphysiology, the Deutsches Forschungs Geselschaft, Utrecht fur University, and the Dr. J. L. Dobberke Stichting and Stichting Fonds Dr. Christine Buisman provided financial support for the research. We thank ´ e, ´ R. Ble, ´ C. van der Hoeven, Isaac, C. Kremer, A. Moresco, A. Bity, F. Bel C. Paukert, and M. Schaaf for their contributions to the research. We greatly appreciate the support and advice by J. van Hooff and E. Sterck. We thank two anonymous reviewers and P. Sicotte for their valuable comments on earlier drafts of the manuscript.

REFERENCES Alexander, R. D. (1974). The evolution of social behavior. Annu. Rev. Ecol. Syst. 5: 325–383. Boesch, C., and Boesch-Achermann, H. (2000). The Chimpanzees of the Ta¨ı Forest: Behavioural Ecology and Evolution, Oxford University Press, Oxford, 316 pp. Borries, C. (1993). Ecology of female social relationships: hanuman langurs (Presbytis Entellus) and the van Schaik model. Folia Primatol. 61: 21–30. Chapman, C. A., Wrangham, R. W., and Chapman, L. J. (1995). Ecological constraints on group size: an analysis of spider monkey and chimpanzee subgroups. Behav. Ecol. Sociobiol. 36: 59–70. Cheney, D. L. (1987). Interactions and relationships between groups. In Smuts, B. B., Cheney, D. L., Seyfarth, R. M., Wrangham, R. W., and Struhsaker, T. T. (eds.), Primate Societies, The University of Chicago Press, Chicago, pp. 267–281. Dasilva, G. L. (1989). The Ecology of the Western Black and White Colobus (Colobus polykomos polykomos Zimmerman 1780) on a Riverine Island in Southeastern Sierra Leone [PhD], University of Oxford, Oxford, England. Dasilva, G. L. (1992). The western black and white colobus as a low energy strategist: activity budgets, energy expenditure and energy intake. J. Anim. Ecol. 61: 79–91. Dasilva, G. L. (1994). Diet of Colobus polykomos on Tiwai Island—selection of food in relation to its seasonal abundance and nutritional quality. Int. J. Primatol. 15: 655–680. Digby, L. J. (1999). Sexual behaviour and extragroup copulations in a wild population of common marmosets (Callithrix jacchus). Folia Primatol. 70: 136–145. Emlen, S. T., and Oring, L. W. (1977). Ecology, sexual selection, and the evolution of mating systems. Science 197: 215–223. Fashing, P. J. (2001). Male and female strategies during intergroup encounters in guerezas (Colobus guereza): evidence for resource defense mediated through males and a comparison with other primates. Behav. Ecol. Sociobiol. 50: 219–230.

Intergroup Relationships in Colobus polykomos

1287

Fleury, M.-C., and Gautier-Hion, A. (1999). Seminomadic ranging in a population of black colobus (Colobus satanas) in Gabon and its ecological correlates. Int. J. Primatol. 20: 491– 509. ˆ de Ta¨ı, Galat, G., and Galat-Luong, A. (1985). La communaute´ de primates diurnes de la foret ˆ d’Ivoire. Rev. Ecol. (Terre Vie) 40: 3–32. Cote Glander, K. E. (1992). Dispersal patterns in Costa Rican mantled howling monkeys. Int. J. Primatol. 13: 4415–4436. Harcourt, A. H., Stewart, K. S., and Fossey, D. (1976). Male emigration and female transfer in wild mountain gorilla. Nature 263: 226–227. Harris, T. R., and Monfort, S. L. (2003). Behavioral and endocrine dynamics associated with infanticide in a black and white Colobus monkey (Colobus guereza). Am. J. Primatol. 61: 135–142. Harvey, P. H., Martin, R. D., and Clutton-Brock, T. H. (1987). Life histories in comparative perspective. In Smuts, B. B., Cheney, D. L., Seyfarth, R. M., Wrangham, R. W., and Struhsaker, T. T. (eds.), Primate Societies, The University of Chicago Press, Chicago, pp. 181–196. Hochberg, Y. (1988). A sharper Bonferroni procedure for multiple tests of significance. Biometrika 75: 800–802. Isbell, L. A. (1991). Contest and scramble competition: patterns of female aggression and ranging behaviour among primates. Behav. Ecol. 2: 143–155. Isbell, L. A. (2004). Is there no place like home? Ecological bases of female disperseal and philopatry and their consequences for the formation of kin groups. In Chapais, B., and Berman, C. M. (eds.), Kinship and Behaviour in Primates, Oxford University Press, Oxford, pp. 71–109. Isbell, L. A., and van Vuren, D. (1996). Differential costs of locational and social dispersal and their consequences for female group living primates. Behaviour 133: 1–36. Isbell, L. A., and Young, T. P. (2002). Ecological models of female social relationships in primates: similarities, disparities, and some directions for future clarity. Behaviour 139: 177– 202. Koenig, A. (2000). Competitive regimes in forest-dwelling Hanuman langur females (Semnopithecus entellus). Behav. Ecol. Sociobiol. 48: 93–109. Koenig, A. (2002). Competition for resources and its behavioral consequences among female primates. Int. J. Primatol. 23: 759–783. Koenig, A., and Borries, C. (2001). Socioecology of Hanuman langurs: the story of their success. Evol. Anthropol. 10: 122–137. Korstjens, A. H. (2001). The Mob, the Secret Sorority, and the Phantoms. An analysis of the Socio-ecological Strategies of the Three Colobines of Ta¨ı [PhD], Utrecht University, Utrecht, 174 pp. ¨ R. (2004). The mating system of an exceptional primate, the olive Korstjens, A. H., and Noe, colobus (Procolobus verus). Am. J. Primatol. 62: 261–273. Korstjens, A. H., Bergman, K., Deffernez- , C., Krebs, M., Nijssen, E. C., van Oirschot, B. M. A., Paukert, C., and Schippers, E. P. (in press). How small-scale differences in food competition lead to different social systems in three closely related sympatric colobines. ¨ ¨ R. (eds.), The monkeys of the Ta¨ı Forest, Ivory In McGraw, S., Zuberbuhler, K., and Noe, Coast an African Primate Community, Cambridge University Press, Cambridge, UK. ¨ R. (2002). How adaptive or phylogenetically inert Korstjens, A. H., Sterck, E. H. M., and Noe, is primate social behaviour? A test with two sympatric colobines. Behaviour 139: 203–225. Marler, P. (1969). Colobus guereza: territoriality and group composition. Science 163: 93–95. Marler, P. (1972). Vocalizations of East African monkeys, II: Black and white colobus. Behaviour 42: 175–197. Marsh, C. W. (1979). Female transference and mate choice among Tana River red colobus. Nature 281: 568–569. Martin, P., and Bateson, P. (1993). Measuring Behaviour: An Introductory Guide, Cambridge University Press, Cambridge, UK, 222 pp. McGraw, W. S. (1996). Cercopithecid locomotion, support use and support availability in the Tai Forest, Ivory Coast. Am. J. Phys. Anthropol. 100: 507–522.

1288

Korstjens, Nijssen, and Noe¨

Oates, J. F. (1977). The social life of a black-and-white colobus monkey, Colobus guereza. Z. Tierpsychol. 45: 1–60. Oates, J. F. (1994). The natural history of African colobines. In Davies, A. G., and Oates, J. F. (eds.), Colobine Monkeys, Cambridge University Press, Cambridge, UK, pp. 75– 128. Oates, J. F., Whitesides, G. H., Davies, A. G., Waterman, P. G., Green, S. M., Dasilva, G. L., and Mole, S. (1990). Determinants of variation in tropical forest primate biomass: new evidence from West Africa. Ecology 71: 328–343. Onderdonk, D. A. (2000). Infanticide of a newborn black-and-white colobus monkey (Colobus guereza) in Kibale National Park, Uganda. Primates 41: 209–212. Pope, T. R. (2000). Reproductive success increases with degree of kinship in cooperative coalitions of female red howler monkeys (Alouatta seniculus). Behav. Ecol. Sociobiol. 48: 253– 267. Reichard, U. (1995). Extra-pair copulations in a monogamous gibbon (Hylobates lar). Ethology 100: 99–112. Sicotte, P., and MacIntosh, A. J. (2004). Inter-group encounters and male incursions in Colobus vellerosus in Central Ghana. Behaviour 141: 533–553. Siegel, S., and Castellan, N. J. J. (1988). Nonparametric Statistics for the Behavioral Sciences, McGraw-Hill, New York, 399 pp. Smuts, B. B., and Smuts, R. W. (1993). Male aggression and sexual coercion of females in nonhuman primates and other mammals: evidence and theoretical implications. Adv. Stud. Behav. 22: 1–63. Stanford, C. B. (1991). Social dynamics of intergroup encounters in the capped langur (Presbytis pileata). Am. J. Primatol.: 35–47. Starin, E. D. (1991). Socioecology of the Red Colobus Monkey in The Gambia with Particular Reference to Female-Male Differences and Transfer Patterns [PhD], City University of New York, New York, 406 pp. Starin, E. D. (1994). Philopatry and affiliation among red colobus. Behaviour 130: 253– 270. Steenbeek, R. (1996). What a maleless group can tell us about the constraints on female transfer in Thomas’s langurs (Presbytis thomasi). Folia Primatol. 67: 169–181. Steenbeek, R. (1999). Tenure related changes in wild Thomas’s langurs II: Between group interactions. Behaviour 136: 595–650. Steenbeek, R. (2000). Infanticide by males and female choice in wild Thomas’s langurs. In van Schaik, C. P., and Janson, C. H. (eds.), Infanticide by Males and Its Implications, Cambridge University Press, Cambridge, pp. 153–177. Sterck, E. H. M., and Korstjens, A. H. (2000). Female dispersal and infanticide avoidance in primates. In van Schaik, C. P., and Janson, C. H. (eds.), Infanticide by Males and Its Implications, Cambridge University Press, Cambridge, UK, pp. 293–321. Sterck, E. H. M., Watts, D. P., and van Schaik, C. P. (1997). The evolution of female social relationships in nonhuman primates. Behav. Ecol. Sociobiol. 41: 291–309. Stoorvogel, J. J. (1993). Gross Inputs and outputs of nutrients in undisturbed forest, Ta¨ı Area, Cote ˆ d’Ivoire, The Tropenbor Foundation, Veenman Drukkers, Wageningen. Strier, K. B., Mendes, F. D. C., R´ımoli, J., and R´ımoli, A. O. (1993). Demography and social structure of one group of muriquis (Brachyteles arachnoides). Int. J. Primatol. 14: 513– 526. Struhsaker, T. T. (1975). The Red Colobus Monkey, University of Chicago Press, Chicago. van Schaik, C. P. (1983). Why are diurnal primates living in groups? Behaviour 87: 120– 144. van Schaik, C. P. (1989). The ecology of social relationships amongst female primates. In Standen, V., and Foley, R. A. (eds.), Comparative Socioecology. The Behavioural ecology of Humans and Other Mammals, Blackwell, Oxford, pp. 195–218. van Schaik, C. P. (1996). Social evolution in primates: the role of ecological factors and male behaviour. Proc. Brit. Acad. 88: 9–31. van Schaik, C. P., Assink, P. R., and Salafsky, N. (1992). Territorial behavior in Southeast Asian langurs: resource defense or mate defense. Am. J. Primatol. 26: 233–242.

Intergroup Relationships in Colobus polykomos

1289

van Schaik, C. P., and van Hooff, J. A. R. A. M. (1983). On the ultimate causes of primate social systems. Behaviour 85: 120–144. Watts, D. P. (1989). Infanticide in mountain gorillas: new cases and a reconsideration of the evidence. Ethology 1–18. Wrangham, R. W. (1980). An ecological model of female bonded primate groups. Behaviour 75: 262–300.