ANIMAL BEHAVIOUR, 2006, 72, 1059e1068 doi:10.1016/j.anbehav.2006.02.016

Social functions of allogrooming in cooperatively breeding meerkats N O BU Y U KI K U T S UK AK E*† & TI M H . CLUTT ON- B ROCK *

*Large Animal Research Group, Department of Zoology, University of Cambridge yDepartment of Biological Science, Graduate School of Sciences, University of Tokyo, Japan (Received 19 September 2005; initial acceptance 25 October 2005; final acceptance 25 February 2006; published online 7 September 2006; MS. number: 8679)

In social mammals, grooming may be used in care of offspring, to maintain pair bonds, or to placate dominant individuals. We examined grooming patterns in groups of the cooperatively breeding meerkat, Suricata suricatta. Dominant females produce over 80% of litters, but older subordinate females occasionally breed. Grooming between dominant individuals was the most common and symmetrical interaction. The dominant female received more and gave less in grooming interactions with subordinates. The dominant female groomed younger subordinates more frequently than older subordinates, suggesting that grooming by dominant females represents parental care, and also reflects the reproductive conflict between females. Older subordinates groomed the dominant female more frequently than did younger subordinates. Subordinates that were frequently attacked by the dominant female groomed her for longer durations than those that were not, and the duration of dominant female grooming by subordinates increased as birth of the dominant female’s pups approached. These results support the idea that subordinates use grooming to placate the dominant female. Analysis of ‘immediate reciprocity’ (whether the groomee returned grooming of the groomer within the grooming bout) showed that subordinate females reciprocated more frequently than subordinate males when the dominant female initiated grooming. However, the dominant female reciprocated subordinate females less frequently than she did subordinate males. This suggests that the need to placate the dominant female may be higher for subordinate females than for subordinate males, possibly because of the risk of eviction caused by reproductive conflict between females. Ó 2006 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd. All rights reserved.

Social grooming is a common feature of many animal societies (reviewed in Spruijt et al. 1992). The primary function of grooming may be hygienic (removing parasites), providing benefits for the groomee by reducing parasite load (Hart et al. 1992; Tanaka & Takefushi 1993). Grooming also functions to relax the groomee or groomer as it stimulates beta-endorphin release (Keverne et al. 1989) and reduces the heart rate (Feh & de Mazieres 1993; Aureli et al. 1999). In social primates, grooming is usually observed to occur between related individuals, including grooming of dependent offspring by parents, which suggests that the behaviour represents nepotism or parental care (Gouzoules & Gouzoules 1987; Goldizen 1989; Schino 2001). Grooming is exchanged for grooming itself (reciprocity; wild

Correspondence and present address: N. Kutsukake, Laboratory for Biolinguistics, RIKEN Brain Science Institute, Wako-shi, Saitama, Japan (email: [email protected]). T. H. Clutton-Brock is at the Large Animal Research Group, Department of Zoology, University of Cambridge, Cambridge CB2 3EJ, U.K. 0003e 3472/06/$30.00/0

impala, Aepyceros melampus: Hart & Hart 1992; chimpanzee, Pan troglodytes: captive: Hemelrijk & Ek 1991; wild: Mitani et al. 2000; Watts 2002) or for ‘social commodity’ benefits other than grooming, such as increasing the probability that a groomee will tolerate the groomer (captive bonnet macaque, Macaca radiata: Silk 1982; captive chimpanzee: Koyama & Dunbar 1996), acquire information about the reproductive state of females (captive wood mouse, Apodemus sylvaticus: Stopka & Macdonald 1999), share food with the groomer (captive chimpanzee: de Waal 1997), allow the groomer access to its infant (wild patas monkey, Erythrocebus patas: Muroyama 1994; wild chacma baboon, Papio cynocephalus ursinus: Henzi & Barrett 2002) and form an alliance with the groomer (wild vervet monkey, Chlorocebus aethiops: Seyfarth & Cheney 1984; captive longtailed macaque, Macaca fascicularis: Hemelrijk 1994; captive chimpanzee: Hemelrijk & Ek 1991; wild chimpanzee: Watts 2002; reviewed in Cords 1997; but see Noe¨ & Hammerstein 1995; Barrett et al. 1999). Although the structure and function of social grooming relationships have been studied in many social primates,

1059 Ó 2006 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd. All rights reserved.

1060

ANIMAL BEHAVIOUR, 72, 5

relatively little is known of the behaviour in other animals in which individuals are identifiable. In particular, the function of grooming is poorly understood in cooperatively breeding vertebrates, in comparison with species living within multimale/multifemale breeding systems of low reproductive skew. We investigated the grooming relationships of dominant females in stable groups of the cooperatively breeding meerkat, Suricata suricatta, during the breeding season. In this species, related individuals form a family group in which one dominant breeding female produces more than 80% of the litters (Griffin et al. 2004). Subordinate females, in particular older individuals, reproduce occasionally (Doolan & MacDonald 1997a,b; O’Riain et al. 2000; Clutton-Brock et al. 2001a). A dominant female is usually tolerant of related males, but aggressively evicts subordinate females which commonly react submissively within the group in the later stages of the dominant female’s pregnancy (Doolan & MacDonald 1996, 1997a,b; Clutton-Brock et al. 1998, 2001a; Kutsukake & Clutton-Brock 2006). Analyses of aggression have shown that dominant females attack older subordinate females more frequently than they do the younger subordinate females (Kutsukake & Clutton-Brock 2006). Our primary aim was to determine whether the pattern of grooming reflects the quality of social relationships within a group. Specifically, we tested the following hypotheses regarding the social function of grooming: sexual bond, parental care, provision of staying or peaceful incentives, reproductive conflict and placation. In the sexual bond hypothesis, as proposed for common marmoset, Callithrix jacchus, grooming is commonly exchanged between dominant individuals, suggesting that grooming functions to maintain and reinforce the sexual relationship (Digby 1995). If this is the case, dominant females should groom the dominant male (sexual partner) more often relative to subordinates. Grooming between dominant individuals should be exchanged symmetrically, and the probability of reciprocation in dominant pairs should be higher than between dominant and subordinate individuals. We also examined the pattern of grooming interruption. Occasionally, meerkats aggressively interrupt grooming between other individuals. Supplanting the grooming partner is sometimes observed in wild cercopithecine primates (Cheney & Seyfarth 1990; Henzi et al. 2003), and this behaviour may represent competition between social partners. If grooming reinforces sexual bonds, both of the dominant individuals should most frequently interrupt grooming bouts with other partners, in particular when the partner grooms with unrelated opposite-sex individuals. In the parental care hypothesis, grooming from parent to offspring may represent parental care (e.g. saddleback tamarin, Saguinus fuscicollis; Goldizen 1989). If this is the case, dominant females should groom younger subordinates or natal offspring more frequently than older subordinates or unrelated immigrant individuals. The hypotheses of provision of ‘staying incentives’ or ‘peaceful incentives’ are based on the fact that, in cooperatively breeding species, a dominant individual grooms or preens more frequently than do subordinates

(Gaston 1977; Rasa 1987; Zahavi 1990; Lazaro-Perea et al. 2004; but see Rabenold 1990; Digby 1995). Therefore, dominant females may use grooming to provide ‘a staying incentive’ in a group to discourage subordinates from dispersing (e.g. common marmoset; Lazaro-Perea et al. 2004). Grooming may also signal the relative dominance of the groomer and discourage challenges by the subordinates (‘peaceful incentive;’ e.g. birds: Harrison 1965; dwarf mongoose, Helogale undulata: Rasa 1987). From these hypotheses, it is predicted that dominant females will groom subordinates more frequently and for longer durations than vice versa. If grooming functions to provide a staying incentive (Lazaro-Perea et al. 2004), the dominant female should groom older subordinate males more frequently than younger males because the former are more likely to disperse (Clutton-Brock et al. 2002). Similarly, if a dominant female offers peaceful incentives to subordinates by grooming (Harrison 1965; Rasa 1987), a dominant female should groom older subordinate females more frequently than younger females because the former are more likely to reproduce (Clutton-Brock et al. 2001a), and the dominant female may need to signal her dominance to them more frequently than to younger subordinate females. In the reproductive conflict hypothesis, grooming frequency in cooperatively breeding species reflects conflict within a group. Examples include the decrease in grooming frequency that occurs during reproductive conflict caused by loss of a dominant female or between a receptive daughter and her mother (common marmoset: Saltzman et al. 1997; Lazaro-Perea et al. 2000). If grooming involving a dominant female reflects reproductive conflict, a dominant female should initiate and reciprocate grooming less frequently with those subordinate females that are likely to reproduce (i.e. old, heavy and non-natal females) than to noncompetitive younger subordinate females. The total amount of grooming and the probability of grooming reciprocation by the dominant female may also be higher with subordinate females than with subordinate males that have no reproductive conflicts with the dominant female. Finally, dominant individuals may ‘police’ the possibility of subordinate reproduction by interrupting grooming between unrelated pairs of males and females. In the placation hypothesis (for grooming by subordinates), grooming functions to reduce social tension in groomees (Feh & de Mazieres 1993; Aureli et al. 1999). Because the dominant female meerkat attacks subordinate males and females, subordinates at risk of receiving aggression may placate the dominant female by grooming. If this is the case, subordinates should groom the dominant female more frequently and for longer durations than vice versa. Given that older subordinate females are most likely to be attacked and the dominant female’s aggressiveness against subordinate females increases in the latter stages of her pregnancy (Kutsukake & Clutton-Brock 2006), older subordinate females may groom the dominant female most frequently, and grooming by subordinates should increase in the latter stages of the dominant female’s pregnancy. Finally, these females should be more active in reciprocating grooming to the dominant female than subordinate males.

KUTSUKAKE & CLUTTON-BROCK: MEERKAT ALLOGROOMING

By investigating the social function of grooming, this study may further our understanding of reproductive skew in social mammals. Reproductive skew models offer different interpretations for reproduction by subordinates in cooperatively breeding societies. The ‘concession’ model attributes subordinate reproduction as a result of the dominant individual’s concession of reproduction to a subordinate (Vehrencamp 1983a, b; Keller & Reeve 1994), whereas the ‘tug-of-war’ model attributes it to the failure of reproductive suppression by the dominant individual (Cant 1998; Clutton-Brock 1998; Reeve et al. 1998; reviewed in Johnstone 2000). Although this study was not designed to test these models directly, analyses of grooming may elucidate whether the nature of social relationships between dominant and subordinate females is hostile and competitive.

METHODS

Study Animals and Field Site This study was conducted in the Kalahari, South Africa, close to Vanzylsrus (26 580 S, 21 490 E) between September and December 2003. The study site consisted of the dry riverbeds of the Kuruman River, herbaceous flats and vegetated dunes. The ecological conditions and climate of this region are described elsewhere (Clutton-Brock et al. 1999a, b; Russell et al. 2002). The study population comprised 198 individuals living in 13 social groups. When this study was conducted, one individual meerkat in each group had been fitted with a radiocollar, thereby allowing groups to be located on any predetermined date (Scantlebury et al. 2002). All individuals were habituated to close observation (i.e. from <1 m) and could be identified by a unique pattern of hair dye marks on the fur, which we maintained while individuals were resting without disturbing or capturing them (Clutton-Brock et al. 1999b; Sharpe et al. 2002). The ages of most individuals were known to within a few days because they had been observed since birth. Groups were visited at least once every 3 days to collect demographic and behavioural data. Most individuals were habituated to a weighing balance, which allowed us to collect weight data regularly by supplying water or pieces of hardboiled egg (<0.5 g). The study was conducted with permission of the Northern Cape Conservation Service, South Africa.

3 h 33 min; sessions lasting <1 h were excluded. N. K. conducted 22 focal observations, on average, of each group (range 7e33, N ¼ 608 observation-hours). During observations, the grooming interactions involving a single dominant female were recorded. Meerkats frequently groom each other during the morning sunning time soon after emergence from a sleeping burrow, or before going down the sleeping burrow in the evening. Grooming was also observed during foraging. In addition to grooming interactions involving a dominant female, N. K. tried to record all grooming interactions among group members. Because a digital video camera was used to store images of grooming interactions among several individuals engaged in the behaviour, few interactions were overlooked. A ‘session’ was defined as 5 s of grooming. A new grooming interaction was defined as commencing when grooming was separated by intervals of more than 1 min. During sessions, N. K. identified a grooming initiator, its partner, the duration of grooming for each individual, whether the recipient of grooming reciprocated (i.e. groomed back) during a grooming session, and how the interaction ended. For polyadic grooming, in which three or more animals groomed together in a huddle, all interactions were recorded as dyads. Grooming was occasionally accompanied by a subordinate female’s submission to the dominant female (i.e. high-pitched vocalization and grovelling); these interactions were seen mainly in the older subordinate females (Kutsukake & Clutton-Brock 2006) and, because they represent submissions, were not included in our analysis. In total, 5339 grooming interactions were observed. During the focal observation, all social interactions involving the focal dominant female, i.e. all aggression received and performed and the identity of the interacting partners, were recorded. Aggressive behaviour was classified as follows: charge (running directly at the subordinate), hip-slam (slamming its hip against the side of a subordinate), chin mark (rubbing its chin on a subordinate), hit (swatting a subordinate with one paw), chase and physical contact by biting. Aggression was excluded when it was a result of competition over a food item. During each focal observation, all individuals that came within a 1-m radius of the dominant female were recorded; hereafter, these are referred to as ‘encounters’.

Quantification of Grooming Observation Methods Continuous focal observations (Altmann 1974) of the dominant female were conducted by N. K. in nine groups of 9e36 individuals at the beginning of the observation period (median ¼ 19). Morning focal observations were conducted when the group appeared from the sleeping burrow, and continued after the group left the burrow to forage, until they became inactive at midday. Evening focal observations were conducted when the group was located, and ended when the dominant female entered the evening sleeping burrow. Observations ceased when the pups were born. On average, each focal session lasted

We quantified dyadic grooming interactions in three ways, and treated them as response variables. (1) Frequency of grooming: the number of grooming bouts observed between the dominant female and each partner during one focal session, regardless of bout duration. (2) Duration of grooming: the total duration of grooming observed between the dominant female and each subordinate in one focal session. (3) Probability of ‘immediate reciprocity’: the probability that a groomee subsequently groomed the groomer within one grooming bout. This reflects a readiness to return favours given by the actor.

1061

1062

ANIMAL BEHAVIOUR, 72, 5

Statistical Analysis We used separate generalized linear mixed models (GLMMs; Schall 1991) to examine predictors of the grooming response variables. Mixed models allow both fixed and random terms to be fitted to a model. Random terms take into consideration repeated sampling. Unless otherwise noted, the group, the identity of the litter and the identity of each subordinate female were set as random terms. Following Crawley (2002), we included all likely independent terms and possible interactions in the maximal model, and excluded terms sequentially until the model included only those terms whose elimination would significantly decrease the Akaike’s Information Criterion (AIC) of the model.

Grooming by dominant females To investigate whom the dominant female groomed most frequently, or who groomed the dominant female most frequently, we calculated a total frequency and duration of grooming for each statuseageesex combination on 1 observation day. Following Clutton-Brock et al. (1999b), we classified individuals <12 months old as juveniles. We set each statuseageesex class (dominant adult males, subordinate adult males, subordinate adult females, subordinate juvenile males and subordinate juvenile females) as levels of a categorical independent factor. To correct for different numbers of dyads for each ageesex combination in each group and focal observation time, we set these variables as covariates. We included identity of the group as a random term in this analysis.

Symmetry of grooming To investigate whether the frequency and duration of grooming between the dominant female and each grooming partner were correlated, we used GLMMs to compare grooming that the dominant female received from and gave to a grooming partner. If grooming was reciprocated in a dyad, the total number of grooming bouts given and received within a dyad or the total duration of grooming exchanged within a dyad should be positively correlated. We also investigated the symmetry/asymmetry of the grooming exchanged within a dyad by comparing the total number of bouts or total duration of grooming that a dominant female performed on and received from each grooming partner. The total number of bouts or total duration of grooming was set as a dependent term with a Poisson error distribution.

Grooming between dominant female and subordinates To investigate grooming between the dominant female and subordinates, we set the total number of bouts or total duration of grooming for each subordinate individual received or given as an independent term with a Poisson error distribution. To investigate variables affecting the probability of immediate reciprocation in grooming between the dominant female and subordinates, we set whether or not grooming was reciprocated during a grooming session as a dependent term with a binomial error distribution.

Independent variables We analysed the predictor variables listed below, which are known to have important effects on meerkat behaviour (Clutton-Brock et al. 1998, 2001a, b; Young 2003; Kutsukake & Clutton-Brock 2006): (1) the number of days before the dominant female gave birth; (2) the sex of each subordinate; (3) the ages of each subordinate, calculated in days, all of which were accurately known; (4) the frequency of aggression by the dominant female in each observation session, calculated as (the number of times that each subordinate was attacked by the dominant female)/(the number of times that each subordinate approached to within 1 m of a dominant female) (Kutsukake & Clutton-Brock 2006). Note that the aggression frequency was significantly associated with the age of the subordinates: older subordinates were attacked more often than those that were younger (Kutsukake & Clutton-Brock 2006). To deal with the collinearity between the age of the subordinate and the aggression frequency, we did not include the two terms together in models; the more significant term was retained in the model (Gilchrist et al. 2004). The length of focal observation in each observation session was fitted as a covariate in the analysis of grooming frequency and duration. The duration of grooming was fitted as a covariate in the analyses of immediate grooming reciprocity because it is likely that the probability of reciprocation is high in grooming sessions of long duration. In addition to the independent terms listed above, we investigated three independent terms listed below, which are known to have important effects on meerkat behaviour. However, we do not report the results of these factors here for simplicity because they did not affect the grooming interactions (all P > 0.15). (1) The weight of each subordinate, calculated by averaging all morning weights for each female during the observation period. Because age and weight are correlated, we used age-controlled weights by calculating an average weight for each age class and subtracting this from the actual weight of each subordinate female. (2) Whether the subordinate was an offspring of the dominant female. Although we could not find statistically significant effects of this variable, the power of the analysis was weak because the number of nonoffspring was low (4/66 males and 5/87 females), with few repeated observations. (3) The number of helpers on the day of observation. We considered all group members to be helpers because the youngest individuals were more than 6 months old during the study period, and previous studies (CluttonBrock et al. 2001b, 2002; Russell et al. 2003) have shown that individuals older than 6 months contribute to pup care more often than do individuals of less than 6 months, and that the contribution level of the older helpers is comparable to that of adult subordinates.

Grooming interruption To investigate the variables predicting grooming interruption, we set whether or not a grooming dyad was

KUTSUKAKE & CLUTTON-BROCK: MEERKAT ALLOGROOMING

RESULTS

Grooming by Dominant Females

Table 1. Variation in grooming frequency and grooming duration between dominant females and other group members b  SE

t

df

P

Grooming by dominant female Frequency Dominant male 0 Adult male 1.330.32 Adult female 1.520.32 Juvenile male 1.290.35 Juvenile female 1.290.35

4.18 4.79 3.69 3.69

166 166 166 166

<0.0001 <0.0001 0.0003 0.0003

Duration Dominant male Adult male Adult female Juvenile male Juvenile female

0 1.560.25 1.910.25 1.570.29 1.60.29

6.16 7.23 5.36 5.47

166 166 166 166

<0.0001 <0.0001 <0.0001 <0.0001

Grooming of dominant female Frequency Dominant male 0 Adult male 0.080.19 Adult female 0.090.19 Juvenile male 0.840.32 Juvenile female 1.960.55

0.4 0.46 2.57 3.55

540 540 540 540

0.69 0.65 0.01 0.0004

Duration Dominant male Adult male Adult female Juvenile male Juvenile female

2.13 0.58 2.09 3.22

540 540 540 540

0.034 0.57 0.037 0.001

0 0.490.23 0.130.23 0.730.35 2.640.82

Frequency bouts/h

(a)

Grooming given

0.3

Grooming received

0.2 0.1 0 20

DM

AM

AF

JM

JF

Sub

AM

AF

JM

JF

Sub

(b)

16 12 8 4 0

Grooming exchanges were a distinctive feature of relationships between the dominant female and male in each group. Dominant females groomed dominant males more frequently and for longer periods than they groomed subordinates (GLMMs after controlling for observation time and number of possible dyads; Table 1, Fig. 1a, b). The frequency of grooming of the dominant female by the dominant male differed from that by juvenile subordinate males (GLMMs after controlling for observation time and number of possible dyads; Table 1) and juvenile subordinate females, but not from that by adult subordinate males and adult subordinate females (Fig. 1a). The dominant male groomed for longer than adult subordinate

Grooming partner

0.4

Duration (s/h)

interrupted as a dependent term with a binomial error structure. To investigate which grooming dyad a dominant female interrupted, we classified grooming dyads in which a dominant female was not a participant as follows: (1) whether the dominant male participated, (2) the sex combination of the grooming dyads, and (3) the possibility of reproduction (i.e. unrelated or not). In an analysis of grooming interruption including a dominant female, the status/sex of the grooming partner (dominant male, subordinate male, and subordinate female) was set as a categorical independent factor. We included the group identity as a random term in these analyses, and excluded data from the group in which the dominant male was absent.

DM

Figure 1. Variation in (a) grooming frequency and (b) grooming duration between dominant females and other group members. Group members were classified according to their status, age and sex. DM: dominant male; AM: adult males; AF: adult females; JM: juvenile males; JF: juvenile females; Sub: all subordinates combined. Data were pooled across nine groups, and are individual means per dyad þ SE.

males and juveniles, but not for longer than adult subordinate females (Table 1, Fig. 1b).

Immediate Reciprocity Between Dominants Dominant females were most likely to reciprocate grooming initiated by dominant males than grooming initiated by subordinate females and males. The mean probability of immediate reciprocation by a dominant female was 79% when a dominant male groomed her (Fig. 2a), compared to 33% with subordinate females (b ¼ 0.55, t106 ¼ 2.28, P ¼ 0.02) and 43% with subordinate males (b ¼ 0.10, t106 ¼ 0.43, P ¼ 0.67). On the other hand, the dominant males were not more likely to reciprocate grooming by the dominant female than other individuals were. The mean probability of immediate reciprocation by a dominant male during grooming by the dominant female was 85%, but this was not significantly different from the probability of reciprocation by subordinate males (68%; b ¼ 0.04, t90 ¼ 0.17, P ¼ 0.87) or subordinate females (75%; b ¼ 0.03, t90 ¼ 0.14, P ¼ 0.89; Fig. 2b).

Grooming Symmetry: Dominants Grooming was exchanged symmetrically in a dominant pair. The total amount of grooming exchanged did not differ between partners in a dominant pair (GLMM: duration: b ¼ 0.35, t207 ¼ 1.45, P ¼ 0.15; frequency: b ¼ 0.21, t207 ¼ 0.86, P ¼ 0.39). There was no evidence that grooming within a dominant pair varied according to the female’s reproductive status (GLMM: duration: b ¼ 0.007, t96 ¼ 0.48, P ¼ 0.63; frequency: b ¼ 0.001, t96 ¼ 0.10, P ¼ 0.92).

1063

ANIMAL BEHAVIOUR, 72, 5

100 (a)

Male Female

Table 2. Variables affecting a grooming interaction between a dominant female and subordinates Independent term

b  SE

t

df

P

80 Grooming of subordinate by dominant female Frequency Sex (M ¼ F) 0.300.18 1.68 1632 0.09 Age 0.00070.0003 2.30 60 0.02y Days to birth 0.00700.004 1.57 1632 0.12

60

Duration Sex (M ¼ F) Age Days to birth

40

% Reciprocation

20

<1

1–2

2+

Dominant male

100 (b) 80

Duration Sex (M ¼ F) Aggression frequency Days to birth Aggression frequency*days to birth

60

40

20

0.430.24 1.77 1632 0.08 0.00070.0003 1.88 60 0.06y 0.00080.0051 0.163 1632 0.87

Grooming of dominant female by subordinate Frequency Sex (M ¼ F) 0.780.43 1.80 1629 Age 0.00190.0008 2.52 60 Days to birth 0.00120.012 0.10 1629 Age*sex 0.00170.0007 2.31 1629 Age*days 0.000040.00002 2.22 1629 to birth

0.71 0.015y 0.92 0.02y 0.027y

0.230.31 10.763.90

0.73 1574 0.47 2.76 1575 0.006y

0.0400.008 0.290.16

4.82 1575 0.0001y 1.84 1575 0.07y

yVariables included in the final models.

<1

1–2

2+

Dominant male

Age of subordinate (years) 0.25 Frequency (bouts/h)

Figure 2. (a) The relation between probability of immediate reciprocation by dominant females and the age and sex of subordinate individuals, after grooming of dominant females by subordinates. (b) The relation between probability of immediate reciprocation by subordinate individuals and the age and sex of subordinate individuals, after grooming by dominant females. For comparison, the probability of immediate reciprocation by the dominant females to the dominant males, after grooming of dominant females by dominant males is shown. Data were pooled across nine groups and are individual means  SE.

Grooming Symmetry: Dominants and Subordinates The frequency and duration of grooming exchanged between a dominant female and each subordinate were positively correlated (frequency: b ¼ 1.189, t1533 ¼ 19.43, P < 0.0001; duration: b ¼ 0.019, t1533 ¼ 16.89, P ¼ 0.01). However, dominant females were groomed for longer than they groomed others (b ¼ 0.39, t168 ¼ 2.46, P ¼ 0.015), whereas grooming frequency was symmetrical (b ¼ 0, t168 ¼ 0.92, P ¼ 0.36).

0.2 0.15 0.1 0.05

12

Grooming of subordinates by the dominant female was affected by the age of the subordinate individuals. Younger subordinates were groomed more frequently than were older subordinates (Table 2, Fig. 3a). Similarly, younger

<1

1–2

2+

(b)

10 8 6 4 2 0

Grooming of Subordinates

(a)

0

Duration (s/h)

1064

<1

1–2

2+

Age of subordinate (years) Figure 3. The relation between the subordinate’s age and (a) grooming frequency and (b) grooming duration by a dominant female. Data were pooled across nine groups and are individual means  SE.

KUTSUKAKE & CLUTTON-BROCK: MEERKAT ALLOGROOMING

subordinates were groomed for longer durations than were older subordinates (Table 2, Fig. 3b). The sex of subordinate individuals affected the probability of immediately reciprocating grooming by the dominant female. Subordinate females were more likely to reciprocate grooming by the dominant female than were subordinate males (Table 3, Fig. 2b).

0.66

(a) 0.25

2+ 1–2 <1

0.2 0.15

Older subordinates groomed the dominant female more frequently than did younger subordinates, and the effect of age on grooming frequency differed with the gender of subordinates and the gestation stage of the dominant female (Table 2, Fig. 4). The duration of grooming of the dominant female by subordinates was positively related to how frequently the dominant female was aggressive to each subordinate and also increased as the birth of pups approached (Table 2, Fig. 5a, b), although the total duration of grooming of other group members by subordinates decreased as parturition approached (GLMMs after controlling for observation time: duration: b  SE ¼ 0.014  0.007, t1599 ¼ 1.8, P ¼ 0.06). The sex of a subordinate individual affected the probability of immediate reciprocation by the dominant female. When a dominant female was groomed by a subordinate, the dominant female reciprocated with subordinate males more frequently than with subordinate females (Table 3, Fig. 2a).

Grooming Interruption Grooming interruption by dominant females was more likely in bouts involving the dominant male (11%; 59/ 540) than in those between subordinates (6.5%; 163/ 2517; GLMM: b ¼ 0.58, t3021 ¼ 3.61, P < 0.001). The probability of interruption did not vary with the sex of the subordinate with which the dominant male was engaged in grooming (b ¼ 0.48, t530 ¼ 1.65, P ¼ 0.10), nor by whether the female grooming partner was related to the dominant male (b ¼ 0.14, t183 ¼ 0.36, P ¼ 0.72). In seven of 59 dyads in which a dominant female interrupted grooming of a dominant male, the interrupting Table 3. Variables affecting the probability of grooming reciprocation within a grooming bout Independent term

b  SE

t

df

P

Dominant female initiated grooming of subordinates Grooming duration 0.0200.003 5.3 281 0.0001y Sex (F > M) 0.870.32 2.7 53 0.01y Age 0.00080.0008 0.94 43 0.35 Days to birth 0.0100.009 1.17 279 0.24 Subordinates initiated grooming of dominant female Grooming duration 0.0080.005 1.65 190 0.1 Sex (M > F) 2.470.75 3.31 193 0.011y Age 0.00040.0010 0.42 37 0.67 Days to birth 0.0010.007 0.22 189 0.83 Age*sex 0.0060.002 2.83 193 0.005y yVariables included in the final models.

Frequency (bouts/h)

0.1

Grooming by Subordinates

0.05 0

0.25

80–61

60–41

40–21

20–0

(b)

0.2 0.15 0.1 0.05 0

80–61

60–41 40–21 Days to birth of pups

20–0

Figure. 4. Changes in grooming frequency by (a) subordinate females and (b) subordinate males in each age group (years) with respect to the number of days until birth of the dominant female’s pups. Data were pooled across nine groups and are individual means  SE.

female subsequently initiated grooming of one of the animals previously engaged in the behaviour. A dominant female groomed a dominant male in five cases, and a subordinate individual in two cases. In grooming bouts involving the dominant female, the interruption frequencies were similar for the dominant male (4.1%: 5/121 cases) and for subordinate animals (5.3%: 54/ 1025 cases; GLMM: b ¼ 0.04, t1079 ¼ 0.08, P ¼ 0.93). Grooming between the dominant female and unrelated immigrant males was never interrupted (N ¼ 15). A dominant male interrupted grooming between the dominant female and a subordinate in 46 of 54 cases, and the remaining eight interruptions were made by subordinate individuals. After initiating a grooming interruption, the dominant male groomed the dominant female in eight of 46 cases, but grooming of the subordinate after an interruption by the dominant male was never observed. When subordinate males and females groomed, whether or not grooming individuals were unrelated (and had a potential to mate) did not affect the probability of interruption by the dominant female (unrelated dyads: 3/22; related dyads: 61/1077; GLMM; b ¼ 0.88, t1092 ¼ 1.39, P ¼ 0.16). DISCUSSION For dominant females, grooming with dominant males was distinct from that with subordinates. Dominant females groomed the dominant males more frequently, and for longer, than subordinates. The total amount of

1065

ANIMAL BEHAVIOUR, 72, 5

12

(a)

10 8 6 4 2 Duration (s/h)

1066

0

80–61

60–41

40–21

20–0

Days to birth of pups 12

(b)

10 8 6 4 2 0

0–10

11–20

21–30 31–40 % Aggression

41–50

Figure 5. The relation between and (a) days until the birth of the dominant female’s pups and (b) the duration of grooming by subordinates the frequency of aggression by the dominant female. Data were pooled across nine groups and are individual means  SE.

grooming exchanged between dominant partners was symmetrical, and the dominant female reciprocated most frequently with the dominant male. The pattern of grooming interruption also showed a bond between dominant individuals. The dominant female interrupted grooming between the dominant male and subordinates more frequently than that between two subordinate partners. Grooming involving a dominant female was interrupted most frequently by the dominant male. Furthermore, in most cases after an interruption, a dominant individual groomed the dominant partner in the interrupted pair. Given that the relationship between dominant individuals may be the most stable in cooperatively breeding meerkat societies, these results suggest that grooming functions to maintain and reinforce the sexual bond between dominant individuals. Frequent grooming in a dominant pair has also been reported in marmosets, Callithrix jacchus (Digby 1995). Grooming of subordinates by the dominant female was affected only by the age of subordinates, that is, dominant females groomed younger subordinate females more than older subordinates (Table 2, Fig. 3). Because societies of cooperatively breeding species are composed of related individuals (where subordinates are the offspring of a dominant pair), grooming of young subordinates by a dominant female may represent parental care (Goldizen 1989). The low level of grooming of older subordinates does not support the notion that the dominant females provide ‘a staying incentive’ or ‘a peaceful incentive’ to subordinates (Harrison 1965; Gaston 1977; Rasa 1987;

Rabenold 1990; Zahavi 1990; Lazaro-Perea et al. 2004), but suggests instead that reproductive conflict between females is reflected in grooming interactions (Saltzman et al. 1997; Lazaro-Perea et al. 2000). Previous studies on meerkats have clarified the competitive relationship between a dominant female and older subordinate females. Old subordinate females occasionally reproduce, and are more likely to be evicted from a group (Clutton-Brock et al. 1998, 2001a). The frequency of aggression shown by the dominant female towards subordinates, and the submission frequency each subordinate female showed in response to aggression, increased with the age of the subordinate female (Kutsukake & Clutton-Brock 2006). The pattern of grooming by subordinates supports the placation hypothesis because dominant females were groomed by subordinates whose relationship with the dominant female appeared to be competitive. For example, the grooming frequency was affected by the age of subordinates (older subordinates groomed dominant females more frequently than did younger subordinates). The duration of grooming was positively related to the frequency of aggressive behaviour by the dominant female towards subordinates, and increased in the later period of the dominant female’s pregnancy. In meerkats, subordinate males voluntarily disperse from the group, whereas subordinate females are aggressively evicted. Given that grooming functions to reduce the tension of the groomer and the groomee (Feh & de Mazieres 1993; Aureli et al. 1999), subordinates, in particular subordinate females, may attempt to reduce their social tension or agonistic tendencies and the risk of eviction from the group by the dominant female. This may be reflected in the asymmetrical distribution of grooming between dominant and subordinate females (the subordinates groom dominants more). Our observation of grooming by subordinates to placate the dominant female is inconsistent with previous studies showing that individuals in conflict avoid grooming (Saltzman et al. 1997; Lazaro-Perea et al. 2000). Our analyses of immediate reciprocity clarify sex differences in the probability of grooming reciprocation. Subordinate females reciprocated a dominant female more frequently than did subordinate males when the dominant female initiated grooming. However, when subordinates initiated grooming, the dominant female reciprocated grooming with subordinate females less frequently than with subordinate males, and the probability of grooming reciprocation with all subordinates was lower than with the dominant male. These observations suggest that the need to placate the dominant female was higher for subordinate females than for subordinate males. In summary, this study shows that grooming in meerkats has important social functions, such as the maintenance of sexual pair bonds between dominant individuals, parental care, reflection of reproductive conflict, and placation of dominants by subordinates. As far as we know, this study is the most detailed examination of interactions between individually identified social carnivores among studies of the distribution and function of grooming in nonprimates. Although previous studies on cooperatively breeding mammals and birds have shown that dominant

KUTSUKAKE & CLUTTON-BROCK: MEERKAT ALLOGROOMING

individuals are more frequent groomers than are subordinates (Gaston 1977; Rasa 1987; Zahavi 1990; Lazaro-Perea et al. 2004), we found a ‘despotic’ pattern of grooming, such as asymmetry of grooming, where the behaviour responded to the frequency of aggression by the dominant female, with different levels of immediate reciprocation or amounts of grooming according to the gender of subordinates. Other than the hypotheses investigated in this study, grooming may function to convey social communication via scent. For example, rodents use autogrooming to advertise their scent and increase their sexual attractiveness (e.g. meadow voles, Microtus pennsylvanicus: Ferkin et al. 1996). Grooming among meerkats may function similarly to distribute the scent contained in their saliva and may be used as a communicative signal. Although this hypothesis requires further testing, we doubt that meerkats rely on grooming for scent marking because they commonly use anal glands for marking objects and group members (personal observation). Grooming interactions between the dominant female and subordinate females may reflect competitive relationships in the reproductive conflict between females. These results have an important implication for the reproductive skew model. There is a controversial proposal that subordinate reproduction is a result of the dominant individual’s concession of reproductive opportunity to a subordinate individual (Vehrencamp 1983a, b; Keller & Reeve 1994), or the failure of reproductive suppression by the dominant individual (Cant 1998; Clutton-Brock 1998; Reeve et al. 1998; reviewed in Johnstone 2000). For meerkats, Clutton-Brock et al. (1998, 2001a) reported intense reproductive conflicts among females, and suggested that dominant females may less easily control reproductive attempts by older subordinate females (see Kutsukake & Clutton-Brock 2006). Our findings on the despotic and asymmetrical grooming relationship between the dominant female and older subordinate females provide supporting evidence for competitive relationships between females and the limited control model. Acknowledgments We thank S. English, K. Moyes, Z. Fry, F. Ballantyne, M. Baker, T. Flower, A. King, A. Ross-Gillespie, C. Walker, K. Skinner, M. Hill, K. Golabek, H. Johnson, P. Minting, A. Thornton, A. Turbe, G. Spong, M. van der Vyver, M. Scantlebury, M. Ridley, N. Raihani, and L. Hollen for support over the course of this study. We especially thank N. Jordan, L. Sharpe, A. Russell, S. Hodge, A. Young, and S. Matsumura for supporting every stage of the fieldwork and discussions. This study was supported by JSPS Research Fellowships. References Altmann, J. 1974. Observational study of behavior: sampling methods. Behaviour, 49, 227e265. Aureli, F., Preston, S. D. & de Waal, F. B. M. 1999. Heart rate responses to social interactions in free-moving rhesus macaques (Macaca mulatta): a pilot study. Journal of Comparative Psychology, 113, 59e65.

Barrett, L., Henzi, S. P., Weingrill, T., Lycett, J. E. & Hill, R. A. 1999. Market forces predict grooming reciprocity in female baboons. Proceedings of the Royal Society of London, Series B, 266, 663e670. Cant, M. A. 1998. A model for the evolution of reproductive skew without reproductive suppression. Animal Behaviour, 55, 163e169. Cheney, D. L. & Seyfarth, R. M. 1990. How Monkeys See The World. Inside The Mind of Another Species. Chicago: University of Chicago Press. Clutton-Brock, T. H. 1998. Reproductive skew, concessions and limited control. Trends in Ecology and Evolution, 13, 288e292. Clutton-Brock, T. H., Brotherton, P. N. M., Smith, R., McIlrath, G. M., Kansky, R., Gaynor, D., O’Riain, J. M. & Skinner, J. D. 1998. Infanticide and expulsion of females in a cooperative mammal. Proceedings of the Royal Society of London, Series B, 265, 2291e2295. Clutton-Brock, T. H., MacColl, A. D. C., Chadwick, P., Gaynor, D., Kansky, R. & Skinner, J. D. 1999a. Reproduction and survival of suricates (Suricata suricatta) in the southern Kalahari. African Journal of Ecology, 77, 69e80. Clutton-Brock, T. H., Gaynor, D., McIlrath, G. M., MacColl, A. D. C., Kansky, R., Chadwick, P., Manser, M., Brotherton, P. N. M. & Skinner, J. D. 1999b. Predation, group size and mortality in a cooperative mongoose, Suricata suricatta. Journal of Animal Ecology, 68, 672e683. Clutton-Brock, T. H., Brotherton, P. N. M., Russell, A. F., O’Riain, M. J., Gaynor, D., Kansky, R., Griffin, A., Manser, M., Sharpe, L., McIlrath, G. M., Small, T., Moss, A. & Monfort, S. 2001a. Cooperation, conflict and concession in meerkat groups. Science, 291, 478e481. Clutton-Brock, T. H., Russell, A. F., Brotherton, P. N. M., Sharpe, L., McIlrath, G. M., White, S. & Cameron, E. Z. 2001b. Effects of helpers on juvenile development and survival in meerkats. Science, 293, 2446e2449. Clutton-Brock, T. H., Russell, A. F., Sharpe, L. L., Young, A. J., Balmforth, Z. & McIlrath, G. M. 2002. Evolution and development of sex differences in cooperative behavior in meerkats. Science, 297, 253e256. Cords, M. 1997. Friendships, alliances, reciprocity and repair. In: Machiavellian Intelligence II: Extensions and Evaluations (Ed. by A. Whiten & R. W. Byrne), pp. 24e49. Cambridge: Cambridge University Press. Crawley, M. J. 2002. Statistical Computing: An Introduction to Data Analysis Using S-Plus. Chichester: J. Wiley. Digby, L. J. 1995. Social organization in a wild population of Callithrix jacchus: II. Intragroup social behavior. Primates, 36, 361e375. Doolan, S. P. & MacDonald, D. W. 1996. Dispersal and extra-territorial prospecting by slender-tailed meerkats (Suricata suricatta) in the south-western Kalahari. Journal of Zoology, 240, 59e73. Doolan, S. P. & MacDonald, D. W. 1997a. Band structure and failure of reproductive suppression in a cooperative breeding carnivore, the slender-tailed meerkats (Suricata suricatta) in the south-western Kalahari. Behaviour, 134, 827e848. Doolan, S. P. & MacDonald, D. W. 1997b. Breeding and juvenile survival among slender-tailed meerkats (Suricata suricatta) in the south-western Kalahari: ecological and social influences. Journal of Zoology, 242, 309e327. Feh, C. & de Mazieres, J. 1993. Grooming at a preferred site reduces heart rate in horses. Animal Behaviour, 46, 1191e1194. Ferkin, M. H., Sorokin, E. S. & Johnstone, R. E. 1996. Self-grooming as a sexually dimorphic communicative behaviour in meadow voles, Microtus pennsylvanicus. Animal Behaviour, 51, 801e810. Gaston, A. J. 1977. Social behaviour within groups of jungle babblers (Turdoides striatus). Animal Behaviour, 25, 828e848. Gilchrist, J. S., Otali, E. & Mwanguhya, F. 2004. Why breed communally? Factors affecting fecundity in a communal breeding

1067

1068

ANIMAL BEHAVIOUR, 72, 5

mammal: the banded mongoose (Mungos mungo). Behavioral Ecology and Sociobiology, 57, 119e131. Goldizen, A. W. 1989. Social relationships in a cooperatively polyandrous group of tamarins (Saguinus fuscicollis). Behavioral Ecology and Sociobiology, 24, 79e89. Gouzoules, S. & Gouzoules, H. 1987. Kinship. In: Primate Societies (Ed. by B. B. Smuts, D. L. Cheney, R. Seyfarth, R. W. Wrangham & T. T. Struhsaker), pp. 299e305. Chicago: University of Chicago Press. Griffin, A. S., Pemberton, J. M., Brotherton, P. N. M., McIlrath, G. M., Gaynor, D., Kansky, R. & Clutton-Brock, T. H. 2004. A genetic analysis of breeding success in the cooperative meerkat (Suricata suricatta). Behavioral Ecology, 14, 472e480. Harrison, C. J. O. 1965. Allopreening as agonistic behaviour. Behaviour, 24, 161e209. Hart, B. L. & Hart, L. A. 1992. Reciprocal allogrooming in impala, Aepyceros melampus. Animal Behaviour, 44, 1073e1083. Hart, B. L., Hart, L. A., Mooring, M. S. & Olubayo, R. 1992. Biological basis of grooming behaviour in antelope: the body size, vigilance and habitat principles. Animal Behaviour, 44, 615e631. Hemelrijk, C. K. 1994. Support for being groomed in long-tailed macaques, Macaca fascicularis. Animal Behaviour, 48, 479e481. Hemelrijk, C. K. & Ek, A. 1991. Reciprocity and interchange of grooming and ‘support’ in captive chimpanzees. Animal Behaviour, 41, 923e935. Henzi, S. P. & Barrett, L. 2002. Infants as a commodity in a baboon market. Animal Behaviour, 63, 915e921. Henzi, S. P., Barrett, L., Gaynor, D., Greefe, J., Weingrill, T. & Hill, R. A. 2003. Effect of resource competition on the long-term allocation of grooming by female baboons: evaluating Seyfarth’s model. Animal Behaviour, 66, 931e938. Johnstone, R. A. 2000. Models of reproductive skew: a review and synthesis. Ethology, 106, 5e26. Keller, L. & Reeve, H. K. 1994. Partitioning of reproduction in animal societies. Trends in Ecology and Evolution, 9, 98e102. Keverne, E. B., Martensz, N. & Tuite, B. 1989. Beta-endorphin concentrations in cerebrospinal fluid of monkeys are influenced by grooming relationships. Psychoneuroendocrinology, 14, 155e161. Koyama, N. F. & Dunbar, R. I. M. 1996. Anticipation of conflict by chimpanzees. Primates, 37, 79e86. Kutsukake, N. & Clutton-Brock, T. H. 2006. Aggression and submission reflect reproductive conflict between females in cooperatively breeding meerkats. Behavioral Ecology and Sociobiology, 59, 541e548. Lazaro-Perea, C., Castro, C. S. S., Harrison, R., Araujo, A., Arruda, M. F. & Snowdon, C. T. 2000. Behavioral and demographic changes following the loss of the breeding female in cooperatively breeding marmosets. Behavioral Ecology and Sociobiology, 48, 137e146. Lazaro-Perea, C., Arruda, M. F. & Snowdon, C. T. 2004. Grooming as a reward? Social function of grooming between females in cooperatively breeding marmosets. Animal Behaviour, 67, 627e636. Mitani, J. C., Merriwether, D. A. & Zhang, C. 2000. Male affiliation, cooperation and kinship in wild chimpanzees. Animal Behaviour, 59, 885e893. Muroyama, Y. 1994. Exchange of grooming for allomothering in female patas monkeys. Behaviour, 128, 103e119. ¨ , R. & Hammerstein, P. 1995. Biological markets. Trends in EcolNoe ogy and Evolution, 10, 336e339. O’Riain, M. J., Bennett, N. C., Brotherton, P. N. M., McIlrath, G. M. & Clutton-Brock, T. H. 2000. Reproductive suppression and inbreeding avoidance in wild populations of cooperatively breeding meerkats (Suricata suricatta). Behavioral Ecology and Sociobiology, 48, 471e477.

Rabenold, K. N. 1990. Campylorhynchus wrens: the ecology of delayed dispersal and cooperation in the Venezuelan savannah. In: Cooperative Breeding in Birds (Ed. by P. B. Stacey & W. D. Koenig), pp. 157e196. Cambridge: Cambridge University Press. Rasa, O. A. E. 1987. The dwarf mongoose: a study of behavior and social structure in relation to ecology in a small, social carnivore. Advances in the Study of Behavior, 17, 121e161. Reeve, H. K., Emlen, S. T. & Keller, L. 1998. Reproductive sharing in animal societies: reproductive incentives or incomplete control by dominant breeders? Behavioral Ecology, 9, 267e278. Russell, A. F., Clutton-Brock, T. H., Brotherton, P. N. M., Sharpe, L. L., McIlrath, G. M., Dalerum, F. D., Cameron, E. Z. & Barnard, J. A. 2002. Factors affecting pup growth and survival in cooperatively breeding meerkats Suricata suricatta. Journal of Animal Ecology, 71, 700e709. Russell, A. F., Sharpe, L. L., Brotherton, P. N. M. & Clutton-Brock, T. H. 2003. Cost minimization by helpers in cooperative vertebrates. Proceedings of the National Academy of Sciences, U.S.A., 100, 3333e3338. Saltzman, W., Severin, J. M., Schultz-Darken, N. J. & Abbott, D. H. 1997. Behavioral and social correlates of escape from suppression of ovulation in female common marmosets housed with the natal family. American Journal of Primatology, 41, 1e21. Scantlebury, M., Russell, A. F., McIlrath, G. M., Speakman, J. R. & Clutton-Brock, T. H. 2002. The energetics of lactation in cooperatively breeding meerkats Suricata suricatta. Proceedings of the Royal Society of London, Series B, 269, 2147e2153. Schall, R. 1991. Estimation in generalized linear models with random effects. Biometrika, 78, 719e727. Schino, G. 2001. Grooming, competition and social rank among female primates: a meta-analysis. Animal Behaviour, 62, 265e271. Seyfarth, R. M. & Cheney, D. L. 1984. Grooming, alliances and reciprocal altruism in vervet monkeys. Nature, 308, 541e543. Sharpe, L. L., Clutton-Brock, T. H., Brotherton, P. N. M., Cameron, E. Z. & Cherry, M. I. 2002. Experimental provisioning increases play in free-ranging meerkats. Animal Behaviour, 64, 113e121. Silk, J. B. 1982. Altruism among female Macaca radiata: explanations and analysis of patterns of grooming and coalition formation. Behaviour, 79, 162e188. Spruijt, B. M., van Hooff, J. A. R. A. M. & Gispen, W. H. 1992. Ethology and neurobiology of grooming behavior. Physiological Reviews, 72, 825e852. Stopka, P. & Macdonald, D. W. 1999. The market effect in the wood mouse, Apodemus sylvaticus: selling information on reproductive status. Ethology, 105, 969e982. Tanaka, I. & Takefushi, H. 1993. Elimination of external parasites (lice) is the primary function of grooming in free-ranging Japanese macaques. Anthropological Science, 101, 187e193. de Waal, F. B. M. 1997. The chimpanzee’s service economy: food for grooming. Evolution and Human Behavior, 18, 375e386. Watts, D. P. 2002. Reciprocity and interchange in the social relationships of wild male chimpanzees. Behaviour, 139, 343e370. Vehrencamp, S. L. 1983a. A model for the evolution of despotic versus egalitarian societies. Animal Behaviour, 31, 667e682. Vehrencamp, S. L. 1983b. Optimal skew in cooperative societies. American Zoologist, 23, 327e355. Young, A. J. 2003. Subordinate tactics in cooperative meerkats: helping, breeding and dispersal. Ph.D. thesis, Cambridge University. Zahavi, A. 1990. Arabian babblers: the quest for social status in a cooperative breeder. In: Cooperative Breeding in Birds (Ed. by P. B. Stacey & W. D. Koenig), pp. 103e130. Cambridge: Cambridge University Press.