PRIMATES,41(3): 321-335, July 2000

321

Matrilineal Rank Inheritance Varies with Absolute Rank in Japanese Macaques NOBUYUKI KUTSUKAKE

The Universityof Tokyo ABSTRACT. In many cercopithecine primates, females form linear dominance hierarchies based on kinship. It is known that female rank follows the rules of matrilineal rank inheritance (MRI): (1) maternal rank inheritance, (2) maternal dominance, and (3) youngest ascendancy among sisters. Although, several studies have highlighted inter-species, inter-population, and intra-group variation in MRI, the factors determining such variation remain largely unknown. In this paper, I investigate the dominance relationships of 69 adult (> 6 yr old) female Japanese macaques (Macacafuscatafuscata) in a free-ranging provisioned troop living in Shiga-Heights (Nagano Prefecture, Japan) and report new evidence of intra-group variation. Dominance relationships among high-ranking females followed MRI within kin units, those among low-ranking females did not. Maternal rank inheritance and youngest ascendancy operated between mother/daughter dyads and sister dyads of high-rank, but not in the dyads of low-rank. The dominance ranks of females from low-ranking kin units were dispersed and less predictable. These findings suggest that MRI varies with absolute dominance rank, and are discussed in relation to other asymmetries between high- and low-rank Key Words: Japanese macaques; Dominance rank; Matrilineal rank inheritance; Youngest ascendancy; lntra-group variation; Support; Provisioning.

INTRODUCTION Many cercopithecine species form female-bonded societies, where females remain in their natal troop and males emigrate to other troops (PUSEY & PACKER, 1987), Among adult females, a linear dominance hierarchy based on kinship exists, and a female's rank is determined by a social process of rank inheritance. KAWAI(1958) and KAWAMURA(1958) produced the first studies of female rank in a cercopithecine species, Japanese macaques (Macaca fuscata fuscata). KAWAMURA(1958) reported rules of matrilineal rank inheritance (MRI) that determine the dominance relationships among adult females. The rules of MRI are threefold: (1) maternal rank inheritance: a daughter's rank depends on her mother's rank and a daughter becomes dominant to all females who were subordinate to her mother; (2) maternal dominance: a daughter ranks below her mother; and (3) youngest ascendancy: among sisters, a younger sister becomes dominant to her elder sisters, resulting in an age-inversed hierarchy among sisters. Consequently, related females form cohesive units in a dominance hierarchy and dominance relationships among matrilineal units (inter-matriline dominance relationships) are automatically formed (KOYAMA, 1967). Following KAWAMURA,dominance relationships among females were intensively studied in many cercopithecine species with frequent confirmation of the principle of matrilineal linear dominance (Japanese macaques: KOYAMA, 1967, 1970; MORt et al., 1989; TAKAHATA, 1991; NAKAMICHIet al., 1995; other macaques: MlSSAKIAN, 1972; DE WAAL, 1977; GOUZOULES, 1975; SILK et al., 1981; BERNSTEIN, 1969; PAUL • KUESTER, 1987; baboons: CHENEY, 1977; HAUSFATERet al., 1982; geladas: DUNBAR,1980; vervet monkeys: BRAMBLETTet al., 1982, reviewed in CHAPAIS, 1992; HILL t~ OKAYASU,1996).

322

N. KUTSUKAKE

However, not all of these studies confirmed MRI and many reported disagreements with some of the rules of MRI. These disagreements could be classified into three types. In the first type, referred to as intra-group variation, the rules of MRI were kept by most of the females in the population but some females did not conform. For example, a female's alliance with a highranking individual, such as an alpha male, may allow her to outrank females she should have been subordinate to (GouZOULES, 1980; TAKAHATA, 1982; MORI et al., 1989; CHAPAIS, 1985; CHAPAIS & LECOMTE,1995). Mother-daughter rank reversals are rare but reported sporadically (MISSAKIAN,1972; CHIKAZAWAet al., 1979; SILK et al., 1981; HAUSFATERet al., 1982; MORI et al., 1989; TAKAHATA, 1991). It is also known that orphans sometimes cannot attain their expected matrilineal rank (HASEGAWA& HIRAIWA, 1980; WALTERS, 1980). The second type - inter-population variation - comprises cases where MRI was confirmed in some populations but not in other populations of the same species. For example, MRI was strictly adhered to in the populations of provisioned or captive Japanese macaques but maternal rank inheritance and youngest ascendancy were not confirmed in troops of wild Japanese macaques (FuRUICHI, 1983; HILL & OKAYASU, 1995). The third type, referred to as inter-specific variation, includes cases where some rules of MRI could not be confirmed in some species even in captive or provisioned populations (e.g. captive bonnet macaques: SInK et al., 1981; provisioned Barbary macaques: PAUL & KUESTER, 1987; PRUD'HOMME & CHAPA1S, 1993a). From these studies, it becomes obvious that the rules of MRI are not universal and discrepancies not rare in cercopithecine species. However, the causes of such variation are still not well understood. From the ultimate perspective, inter-species variation may be related to the speciesspecific dominance styles (DE WAAL & LUTTRELL, 1989; THIERRY, 1990). From the proximate perspective, several factors seem to influence the dominance relationships of females. First, some studies indicate that demographic factors, such as mother's age or age difference between sisters, affect dominance relationships (CHIKAZAWAet al., 1979; HAUSFATERet al., 1982; PAUL & KUESTER, 1987), but some suggest that demographic factors are additional rather than primary factors determining female dominance relationships (HILL & OKAYASU,1995; PRUD'HOMME & CHAPAIS, 1993a). Further, on the basis of observations of wild Japanese macaques on Yakushima Island, HILL and OKAYASU (1996) proposed that ecological conditions determine dominance relationships among females, since they affect rates of aggression and opportunities for support. One convincing factor that could explain such variation is agonistic support during rank inheritance. There is much observational evidence that support, mainly from close relatives such as mothers, is crucial in rank inheritance (CHENEY, 1977; KURLAND, 1977; BERMAN, 1980; WALTERS, 1980; HORROCKS & HUNTE, 1983; SILK, 1982; DATTA, 1988; reviewed in CHAPAIS, 1992; but see also SILK et al., 1999). Experimental evidence was presented by CHAPAlS and his colleagues who created subgroups of female Japanese macaques designed to induce rank reversals and noted the importance of alliances with kin (CHAPAIS, 1988) and non-kin (CHAPAISet al., 1991, 1994; CHAPAlS & ST-PIERRE, 1997) in the process of rank inheritance. These proximate factors are not completely independent, therefore, several factors may simultaneously affect the determination of dominance relationships among females. To reveal factors that cause such variation, it is necessary to study female dominance relationships and the rules of MRI both in multiple species and in several troops of the same species. In the latter respect, Japanese macaques are especially good candidates for analysis of the effects of the rules of MRI. Japanese macaques have been the subjects of long-term studies covering many groups living in various conditions, resulting in many reports of intra-group and inter-popula-

Matrilineal Rank Inheritance in Japanese Macaques

323

tion variation, enabling clarification of the factors that produce such variation. For example, NAKAMICHIet al. (1995) reported that females of a provisioned troop based in Katsuyama from high-ranking kin units formed cohesive units, while, females in middle- and low-ranking kin units formed more dispersed hierarchies. Thus, they concluded that the effects of MRI vary with dominance rank. In their study, however, there were many kin units (17) and few females (1-8 females: mean 4.5 females) in each kin unit; a demographic composition which did not facilitate a detailed study of the differences between high- and low-ranking kin units. In this study, I investigated the dominance relationships among female Japanese macaques in a provisioned troop, Shiga A-1 troop. This troop contains only two matrilines and one matriline contained 94% of the adult females, which enables us to study details of differences between high- and low-rank. The aim of this paper is to investigate agreements/disagreements with MRI between high- and low-rank kin units in a provisioned troop of Japanese macaques of a distinct composition relative to other troops, providing new evidence of intra-group variation in Japanese macaques.

METHODS STUDY GROUP AND TROOP COMPOSITION

This study was conducted on Shiga A- 1, a free-ranging troop of Japanese macaques living in Shiga-Heights, Nagano Prefecture, Japan. Demographic records (i.e. births, deaths, immigrants, and emigrants) have been kept since 1962 and all individuals are individually identifiable. Barleycorn and soy beans were scattered three times (09:00, 12:00, and 15:00) and apples one time (16:30) daily by the staff of Jigokudani Monkey Park. Tourists were prohibited from feeding them. The troop's ecological conditions are described in more detail by WADA and IcntKl (1980). Data were collected on a total of 94 days from January (mating season) until October in 1998 (non-mating season) intermittently. Observations were taken from 08:00 to 17:00. Total observation time was 652.5 hr. During the study period, the troop consisted of 199 - 218 individuals (19 infants were born). At the beginning of April 1998, there were 21 adult males (>6 yr old), 69 adult females (>6 yr old), and 128 juveniles and infants. This study focused on 69 adult females of which 20 were nulliparous, 9 primiparous, and 40 multiparous when the study began. Shiga A-1 troop stemmed from Shiga A troop which had five matrilines when first observed in 1962. As a product of several bouts of troop fission, t w o o f these five matrilines remain in the present Shiga A-1 troop. Figure 1 shows the genealogy of 69 adult females in this troop. Since 94.2% (65/69) of the adult females came from one matriline, I divided this matriline into four kin units (R, K, N, and T) so that the daughters of Keg become the founders of each kin unit. Following this procedure, five kin units (R, K, N, T, and M) exist in the troop. Since observation began, females of M matriline were subordinate to the matriline that included the four present kin units. Based on the rules of MRI, the dominance order among the five kin units can be predicted to be R>K>N>T>M. However, since Keg died when Kro and Krl were still immature, it was possible that Kro and Krl failed to inherit their mother's rank (see Introduction). Thus, I calculated the average rank in each kin unit to assess the relative rank of matriline. From this procedure, the dominance rank among kin units was determined as T>M>N>R>K. Hereafter, I refer to this female's rank as predicted by rules of MRI as 'predicted' rank, and the female's rank as determined in the present study to 'absolute' rank.

324

N. KUTSUKAKE 1970

1975

!

Keg

1980

1985

1990

I

!

.'

I

KroKrl

i! ,

r---- * - ~

,

--

--

O

]

I

i

O

O

OO... O

T kin unit

ITL

T w

w

A w

--9

(~

O

N kin unit

0 _- _-

Rkinunit

~

K kin unit

O

Mkinunit

Fig. 1. The genealogy of Japanese macaques in Shiga A-1 troop. A circle indicates an adult female, and a black circle indicates the adult females whose rank disagreed with the rules of MRI. As indicated by the dotted line, R, K, N, and T kin units were from the same matriline. Kin units (except for M kin unit) were arranged from top to bottom vertically in the order of the dominance. An arrow refers to the females who emigrated to the neighboring troop (see Discussion). The asterisk indicates the sister dyad in which the younger sister became subordinate to her elder sister due to their mother's death (see Method).

OBSERVATIONMETHODS

Dominance relationships were usually observed by the ad lib sampling method, although the focal animal sampling method (ALTMANN, 1974) was occasionally employed where ad lib sampling failed to provide sufficient data. Ignoring the effect of other individuals, female A was considered to be dominant to the female B if (1) As a consequence of A's approaching B, A supplanted B, or B expressed a 'bared-teeth display' (DE WAAL 8Z LUTTRELL, 1985); or (2) A unilaterally attacked B. Soybeans were occasionally given to the subjects in order to induce (1) and (2). However to minimize the effect of artificially induced aggression, soybeans were used in only 28 (1.1%) dominance interactions covering 27 dyads (1.2% in the whole). As consorting males influence dominance relationships between estrous and other females (TAKAHATA,1982), I ignored the dominance interactions of consorting females. A dominance hierarchy was constructed such that the number of interactions below the diagonal of the matrix were minimized (MARTIN & BATESON, 1990). The linearity index, h' (DE

Matrilineal Rank Inheritance in Japanese Macaques

325

VRIES, 1995) was used to test the linearity of dominance hierarchy. The h' index ranges between

0 and 1 and if the linearity is complete, h' equals 1. I excluded four M kin unit females from some analyses because they frequently formed alliances with the alpha male (KuTSUKAKE, in prep.) and had unstable dominance relationships (TAKAHATA, 1982). I also removed four 'orphans' whose mothers died before they reached 4 yr of age, because orphans may not achieve their matrilineal rank (see Introduction). There was also one case, at the top of the dominance hierarchy in the T kin unit, in which the mother died before her youngest daughter reached 4 yr of age, and this youngest daughter became subordinate to her elder daughter. After that, the dominance relationship between the sisters remained stable and those ranks were inherited by their daughters. This case was not considered in a disagreement with the rules of MRI (indicated by * in Fig. 1). I divided the 61 females into 2 classes: 30 females of the highest half as "high-rank" and the 31 females of the other half as "low-rank." By this definition, high-ranking females include females of part of the T kin unit, and low-ranking females include the remaining females of the T kin unit plus all females of N, R, and K kin units.

RESULTS COMPARISON BETWEEN PREDICTED RANKS AND ABSOLUTE RANKS

Two thousand four hundred and fifty (2450) dominance interactions were observed among 1824 dyads, involving 77.7% of all the dyads. Twenty-seven dominance interactions (1.1% of interactions) occurring in 25 dyads (1.4% of the observed dyads) contradicted the absolute rank hierarchy. I observed a single dominance interaction in 1308 dyads, and two or more interactions in 516 dyads. Of these 516 dyads, 10 dyads (1.9%) were tied; there were equal numbers of both wins and loses. This suggests that dominance relationships were stable, and that observation of one interaction was sufficient to define the direction of dominance in a relationship. A linear dominance hierarchy was found (h'=0.59, p-=0.001) and there was a strong positive correlation between predicted rank and absolute rank (Fig. 2, r=-0.93, p<0.0001). AGREEMENTS]DISAGREEMENTSWITH RULES OF MRI

Within Kin Units In the top and third highest ranking kin units, T and N, there were positive correlations between predicted and absolute ranks (Table. 1). In the fourth highest-ranking kin unit, R, there was a negative but non significant correlation. In the second highest and bottom ranking kin units, M and K, statistical analysis were not possible because of the small sample size. Overall it appears that the rules of MRI held within the high-ranking kin units but not within the lowranking kin units. Overall Disagreements Disagreements with MRI were found for 26 of the 61 females (42.6%) in 119 dyads (6.5%). For high-rankers, disagreements were found for 4 of 30 females (13.3%) in 13 dyads. For lowrankers disagreements were found for 22 of 31 females (71.0%) in 106 dyads (Fig. 3). A significant difference was found between high- and low-rank in the number of females that disagreed

326

N. KUTSUKAVm

70

60-

[]

50~D 0

[] T kin unit 9 M kin unit N kin unit 9 R kin unit x K kin unit

40

[],%r [ ] d . "~-

D

30[]

9

2010 0

9 . . . . . . . . . . .

0

10

I

. . . .

I

. . . .

I

. . . .

20 30 40 50 predicted rank

I

. . . .

60

70

Fig, 2. Comparison of the females' absolute and predicted ranks. On the dotted line, predicted rank is equal to absolute rank, in which case the rules of MRI would hold.

Tablel. Agreements/disagreements with MR| within kin units. Correlation coefficient No. of dyads contradictKin No. of betweenpredicted ranks ing MRI (% of dyads unit females* and absolute ranks p value within kin unit) T 41 0.96 <0.0001 33 (4) M 4 0.4 ** 2 (33) N 9 0.88 0.0125 4 (l 1) R 8 -0.69 0.0677 17 (61) K 3 -0.5 ** 2 (67) * Excluding orphans; ** sample size is too small for statistical analysis.

No. of females contradicting MRI (% of females within kin unit) 9 (22) I (25) 2 (22) 4 (50) 2 (67)

with M R I (Fisher's exact probability test, p=0.0037) and in the number of females that were expected to be subordinate in predicted rank but were actually dominant (Mann-Whitney U-test, nHigh= 30, nLow=31, U=181.0, p<0.0001). These results indicated that disagreements with MRI occurred more frequently among low-rankers. This tendency did not change when I excluded disagreements with the rule of maternal dominance and that o f youngest ascendancy (Fig. 4). A significant difference was found between high- and low-rank in the number of females that disagreed with MRI (Fisher's exact probability test, p<0.0001) and in the number of females that were expected to be subordinate in predicted rank but were actually dominant (Mann-Whitney U-test, nHigh= 30, nLow=31, U=225.5, p<0.0001).

Rule of Maternal Rank Inheritance Usually, daughters rank just below their mother. In this group, there were five mother-daughter dyads in which females other than the daughter's sisters ranked between them. All of these separated mother/daughter dyads were low rankers (Fig. 5).

Matrilineal Rank Inheritance in Japanese Macaques T kin unit zx N kin unit 9 Rkinunit

327

o

14 ~2

E

X

~

10

0

8

E

high-rank ~.d

~w-rank

K kin unit

o Q

l i l

6 0

0

00. ~

10

20

O0 , ~

A ~__,

...~ ,

A --

XX m m

40

30

6O

50

a b s o l u t e rank Fig. 3. The absolute rank of females that disagreed with MRI, and the discrepancy between the female's predicted and absolute ranks in terms of the number of females ranking in between.

14 12 E t.i.., O

n T kin unit z~ N kin unit 9 R kin unit

high-rank ~..~.~ w - r a n k ' +

X K kin unit

o

tO

13

0 1:3

8

E

6

r..,

0

4 2 0

I~ 0 0 .--.:'z.--..--.~U,:'z..~..~..--...":.--..-z..~.~:'z.~."z.~,.--. ~.'~..--..'z.~ , ~ .--.~

10

20

30

~ ,-,.-.~.'."

40

9 ~_~_

,~--~X

50

,

60

absolute rank Fig. 4. The absolute rank of females that disagreed with MRI excluding mother ascendancy and youngest ascendancy, and the discrepancy between the female's predicted and absolute ranks in terms of the number of females ranking between them.

Rule of Maternal Dominance Mothers are usually dominant to their daughters. In the present group, 4 of 31 mother-daughter dyads (12.9%) wcrc composed of daughtcrs dominant to their mothcrs (Fig. 6). One dyad was high-ranking, and three were low-ranking. Although there was insufficient data for statistical analyses, disagreements with maternal dominance did not appear to bc rclated to absolute rank.

328

N. KUTSUKAKE

high-rank

[] /, 9 It

low-rank

12

II

T kin unit N kin unit R kin unit K kin unit

10

E

s

%.,

o

6

~

4

~

2

[]

DO

9

~ M M M M M ~ M

A A A A

~

mother-daughter dyads a~anged from high-rank Fig. 5. The absolute rank of mother-daugther dyads in which the daughter did not rank just below her mother, and the number of females that ranked between the mother and her daughter [excluding the daughter's sister(s)].

high-rank ~_.J

[] T kin unit zx N kin unit 9 R kin unit tt K kin unit 9 M kin unit

low-rank 1

..

daughter>mother

r-i

I /',

9

X

I I I I I

mother>daughter

I.,I I - 1 t.,.I I..I t . d I--I I..I 6 1

"~

I-,1 I . . I I..I I..I 6 J I,.I 6 n . d

I . . I I-.11-.I

..........

/,.a g . . x / . ~ E

illllw mu~

,ww

mother-daughter dyads arranged from high-rank Fig. 6. Dominance relationships between mothers and daughters.

Rule of Youngest Ascendancy Among Sisters Usually, younger sisters dominate their older sisters. Disagreements with the rule of youngest ascendancy were found for 7 females in 9 dyads of the 35 sister dyads (25.7%). While there was just one case o f disagreement among 21 high-ranking sister-dyads (4.8%), there were 8 cases among the 14 low-ranking dyads (57.1%; Fig. 7). These data indicated youngest ascendancy was adhered to by high-rankers, but not by low-rankers (Fisher's exact probability test, p=0.021). However, this does not mean that older sisters usually dominated their younger sister (binomial test, p--0.79).

Matrilineal Rank Inheritance in Japanese Macaques

329

high-rank

young>old

low-rank

I

old>young

[] O D D

L . I L . J L..J L . J L . J L a t = J t . 4 L . J L U l _ J l _ , . . I

t..IL.I k.J L.I k..Ik.JL.I

[] ,x 9 It 9

I

mli~

A

...... L..I I....IL..W

T kin unit N kin unit R kin unit K kin unit M kin unit

9

/ , ~ t . ~ t...X

sister dyads arranged from high-rank Fig. 7. Dominance relationships between sisters. The three sister dyads linked by a line indicates that one female was subordinate to her three older sisters.

RANKS OF FEMALES IN M KIN UNIT

Four females in the M kin unit were excluded from the above analyses since they formed alliances with the alpha-male and had unstable dominance ranks. However, analyses in the context of within-kin unit were available. There were no mother-daughter rank reversals (Fig. 6). In one M kin unit sister dyad, the older sister was dominant to her younger sister (Fig. 7). RANKS OF ORPHANS

Three of the four orphaned females had a lower absolute rank the one predicted, from 14th (predicted rank) to 30th (absolute rank), 37th to 55th, 55th to 69th. One female, from 35th to 38th, seemed to retain her mother's rank. The three females who fell in rank were outranked by all their (respectively 4, 2, and 1) older sisters. EFFECTS OF DEMOGRAPHICFACTORS

As noted in the Introduction, it is known that demographic factors, especially age differences between sister's and mother's age, affect dominance relationships among females. To test whether disagreements with predicted rank resulted from demographic factors, I investigated demographic differences between high-/low-rank and between the dyads that agreed/disagreed with MRI. There was no statistical difference in the age difference of sisters who followed and those who did not follow the rule of youngest ascendancy (Agree=5.27 yr, Disagree=3.2 yr; MannWhitney U-test, nAgree=26, nDisagree=lO, U=83.00, p=0.09). There was also no statistical difference between the age of mother who followed or failed to follow the rule of maternal dominance (Agree=l8.5 yr, Disagree=l8.5 yr; Mann-Whitney U-test, nAgree=27, nDisagree=4, U=41.50, p=0.46). No statistical difference was found in the age difference between mothers

330

N. KUTSUKAKE

and daughters who followed and those who did not follow the rule of maternal dominance (Agree=8.55 yr, Disagree=7.25 yr; Mann-Whitney U-test, nAgree=27,nDisagree=4, U=45.50, p=0.49). There was also no statistical difference between the age of mothers that agreed and disagreed with maternal rank inheritance (Agree=18.30 yr, Disagree=17.2 yr; Mann-Whitney U-test, nAgree=26, nDisagree=5, U=61.50, p=0.85). Since agonistic support, mainly by close relatives, is important in the dynamics of MRI (see Introduction), I also tested the possibility that the difference between high- and low-rank was a product of a difference in the number of close relatives. In this paper, I defined close relatives as r_~0.25 (CHAPAlS et al., 1997). There was no statistical difference in the number of close relatives between high-rankers and lowrankers had (High=2.67, Low=2.26; Mann-Whitney U-test, nHigh=30, nLow=31, U=396.0, p=0.32). In summary, I was unable to find any effect of demographic factors on MRI. Special attention should be paid to maternal dominance. Some studies have shown that old mothers of weakened physical condition tended to be outranked by their daughter (see Introduction). However, in this study, statistical analyses failed to uncover a relationship between mother's age and mother/daughter dominance relationships. Furthermore, two of the four outranked mothers bore infants during April 1998, suggesting that in this population, mother/daughter rank reversals are not related to mother's age, reproductive condition or absolute rank.

DISCUSSION In the present study, new evidence of intra-group variation in MRI dominance relationships was found. MRi characterized high-ranking females, but not low-ranking females. The detail of differences between high- and low-rank were summarized as follows: (1) dominance relationships within low-ranking kin units did not follow MRI; (2) maternal rank inheritance was not followed in low-ranking mother/daughter dyads; (3) youngest ascendancy was not followed in low-ranking sister dyads; and (4) the ranks of low-ranking females were dispersed and intermatrilineal dominance relationships became equivocal. Previous studies of provisioned Japanese macaques confirmed the strong effects of the rules of MRI, but also reported that MRI varies with absolute rank. For example, MORI et al. (1989) reported long-term changes of dominance relationships among adult females (number of females: minimum: 7; maximum: 49) over 29 years in Koshima group, noting that dominance ranks changed frequently among low-rankers but not high-rankers. NAKAMICHI et al. (1995) studied dominance relationships among 74 females in a provisioned troop at Katsuyama, and reported that females in high-ranking kin units formed cohesive units, while, females in middleand low-ranking kin units had more dispersed hierarchies. How did these differences between high- and low-rank arise? To answer this question, it is necessary to consider other asymmetries between high- and low-rank. ASYMMETRIES BETWEEN HIGH- AND LOW-RANK

Previous studies of cercopithicine species highlighted many asymmetries between high- and low-rank. For example, it is known that high-ranking females have more active and powerful allies than low-ranking females (CHAPAIS, 1992; PRUD'HOMME & CHAPAIS, 1993b) and that asymmetries cause strong kin unit cohesion in high-rank but weak cohesion in low-rank (CHENEY, 1983). NAKAMICHIet al. (1995) argued that rank dispersion in low-ranking kin units is

Matrilineal Rank Inheritance in Japanese Macaques

331

related to the weak kin unit cohesion, but failed to provide data, such as differences in grooming or spatial proximity to support their idea. In this study, there is some evidence supporting NAKAMICHI et al. idea, that is, that three females in R kin unit subsequently emigrated to a neighboring troop (KuTSUKAKE,pers. observ., see Fig. 1) both in the preliminary study period (November 1997) and after the study period (October and December 1998). In particular, one female left her mother and older (dominant) sister in the natal troop, abandoning benefits received from close relatives such as grooming and agonistic support, suggesting that relationships between relatives were of low value and kin unit cohesion weak in R kin unit. It is also known that lower-ranking females attempt to affiliate with higher-ranking females (CHENEu & SEYVARTH, 1990; CHAPAISet al., 1995; but see also DE WAAL & LUTTRELL,1986). Because low-ranking females do not have active and powerful allies, they may attempt to follow an individualistic strategy, such as, forming bridging alliances with high-ranking females (CHAPAIS, 1992) and/or males (GouZOULES, 1980; TAKAHATA,1982; MORI et al., 1989; CHAPAIS, 1985; CHAPAIS& LECOMTE, 1995), at the expense of kin unit cohesion and support of close relatives. Alternatively, it is possible that weak kin unit cohesion can encourage females of lowrank to adopt individualistic strategies. This strategy appeared to be adopted by females in M kin unit who formed alliances with the alpha male and succeeded in outranking females expected to be dominant. However, the fact that they were more widely dispersed in the dominance hierarchy than other kin units indicates that inter-matriline dominance relationships became equivocal (Fig. 2). Further, food distribution also seems to produce differences between high- and low-rankers. In provisioned conditions, food resources are highly concentrated at the feeding site producing relatively high aggression rates among individuals (MORI, 1977). Because they are attracted to clumped food resources, high-ranking females tend to concentrate in the central area of the feeding site, while, low-ranking females remain in the periphery. Consequently, low-ranking individuals feed on provisioned food less than high-ranking individuals and tend to rely more on the natural resources (SouMAH & YOKOTA, 1991). The distribution of natural resources is less dense than artificial food, and inter-individual distance increases reducing the rate of agonistic interactions and support from close relatives. It is possible that, in this study troop, some disagreements were the products of social factors and others of ecological factors. Therefore, these explanations are not mutually exclusive, and empirical data on agonistic support are needed to reveal details of the process causing disagreements with MRI. VARIATIONAMONGPROVISIONEDTROOPSOF JAPANESEMACAQUES MRI did not characterize the dominance relationships of low-ranking females in the provisioned troop of Japanese macaques at Shiga (this study), Koshima (MOR1 et al., 1989), and Katsuyama (NAKAMICHIet al., 1995). However, other studies did not report such an effect of MRI among low-rankers. For example, KOYAMA(1970) studied dominance relationships among 54 adult females in a provisioned troop of Japanese macaques in Arashiyama and found few disagreements with MRI both in the high- and low-rank. This suggests that a weak effect of MRI in low-rankers is not common in every provisioned troop of Japanese macaques, and variation among provisioned troops exists. However, each provisioned troop lives in a different environment, and has distinct troop composition, troop history, demography (e.g. WATANABEet al., 1992; KOYAMAet al., 1992; ITOIGAWAet al., 1992), and style of provisioning (HILL, 1999). Therefore, it would not be surprising that parts of social behavior were determined by those factors which are distinct to a troop, resulting in variation among provisioned troops.

332

N. KUTSUKAKE

MORI et al. (1989) speculated that the limited disagreements with MRI in Arashiyama group may relate to troop conditions, that is, that troop membership increased in the period of KOYAMA (1970). To support this, in Koshima group, changes of dominance relationships among females and disagreements with MRI increased during a period of population decline, mainly due to the restriction of provisioning (Morn et al., 1989; WATANABEet al., 1992). Increase and decrease of individuals in the group seems to be closely related to the provisioning style, such as type, amount and the density of provisioned food. Since the style of provisioning determines the intensity of competition on food, inter-individual distance during feeding, rate of aggression (HILL, 1999), and the opportunity for support, it is possible that the effects of MRI were affected by provisioning style. To test this possibility, detailed information about provisioning style and reproductive/demographic factors in each provisioned troop is indispensable. Unfortunately, reproductive/demographic factors of this study troop were not analyzed quantitatively, preventing further analysis and comparison with other provisioned troops. Future study should pay more attention to variation within provisioned troops of Japanese macaques, especially the relationships between reproductive/demographic factors, style of provisioning, and social behavior such as the effect of MRI on determining dominance relationships among females. IS YOUNGESTASCENDANCYINTRINSICTO JAPANESE MACAQUESOR A PRODUCTOF PROVISIONING?

In studies of MRI in cercopithecine species, youngest ascendancy between sisters has been a controversial issue, being confirmed in free-ranging provisioned and captive populations of Japanese and rhesus macaques, but not in provisioned Barbary and bonnet macaques (FuRuICHI, 1983; SILK et al., 1981; PAUL • KUESTER, 1987; PRUD'HOMME& CHAPAIS, 1993a). Concerning inter-population variation, youngest ascendancy between sisters was not confirmed in wild troops of Japanese macaques in Yakushima (HILL & OKAYASU, 1995) and Kinkazan (NAKAGAWA & TAKAHASHI,pers. comm., cited in HILL, 1999), suggesting that youngest ascendancy in Japanese macaques is a by-product of provisioning (HILL & OKAYASU, 1996). It is known that youngest ascendancy is a product of powerful, long-term support, mainly by close-relatives (DATTA, 1988; but see also SILK et al., 1999). Therefore, the absence of youngest ascendancy between low-ranking sisters in this study may result from the absence of support caused by weak kin unit cohesion due to social or ecological conditions. If the absence of youngest ascendancy between low-ranking sisters was caused by social asymmetries between high- and low-rankers (e.g. attraction to higher-ranker individuals), the way in which disagreements with MRI arose is opposite to that of wild troop: the former resulted from high frequency, complex social interactions, and the later from an absence of social interactions such as aggression and support. Similarly, if ecological factors caused disagreements with youngest ascendancy between sisters, the way in which disagreements with MRI arose seems to be different from that of wild troops since the rate of aggressive interactions was extremely high in the provisioned troop relative to the wild troop (HILL, 1999). Thus, the absence of youngest ascendancy between low-ranking sisters in this troop should not be regarded in the same light as that of wild troops even if, at first glance, it seems to be a similar phenomenon to that of wild troops. That is to say that, even if though we have found provisioned troops in which youngest ascendancy does not occur, this does not necessarily mean that youngest ascendancy is unrelated to provisioning. Therefore, we cannot unequivocally answer the question of whether youngest ascendancy is a by-product of provisioning or not. However, it is clear that disagreements with youngest ascendancy occur even when the mother is alive and even in provisioned groups of Japanese macaques.

Matrilineal Rank Inheritance in Japanese Macaques

333

Acknowledgements. I thank TOSHIKAZUHASEGAWAfor supervision and encouragement. I would not have been able to complete this work without help from DUNCANCASTLESwho corrected the English and provided critical comments on early versions of the manuscript. I also thank SACHIYOKAJIKAWA,FRANSB. M. DE WAAL, BERNARDCHAPAIS,YUKIOTAKAHATA,and one anonymous referee for critical comments, and lcHmo TANAKAfor advice on fieldwork. I especially thank EISHITOKIDA,SOGOHARA, HARUOTAKEFUSHI, TOSHIOHAGIWARA,and the other staff of Jigokudani Monkey Park for permitting this study and supporting every stage of my fieldwork. NK was supported by JSPS Research Fellowships for Young Scientists.

REFERENCES

ALTMANN,J. 1974. Observational study of behaviour: sampling methods. Behaviour, 49: 227-267. BERMAN, C. M. 1980. Early agonistic experience and rank acquisition among free-ranging infant rhesus monkeys. Int. J. Primatol., 1:153-170. BERNSTEIN, 1. S. 1969. Stability of the status hierarchy in a pigtail monkey group (Macaca nemestrina). Anita. Behav., 17: 452-458. BRAMBLETT,C. A.; BRAMBLETT,S. S.; BISHOP, D. A.; COELHO,JR. A. M. 1982. Longitudinal stability in adult status hierarchies among vervet monkeys (Cercopithecus aethiops). Amer. J. Primatol., 2: 43-51. CHAPAIS, B. 1985. An experimental analysis of a mother-daughter rank reversal in Japanese macaques (Macaca fuscata). Primates, 26: 407-423. CHAPAIS, B. 1988. Experimental matrilineal inheritance of rank in female Japanese macaques. Anim. Behav., 36: 1025-1037. CHAPAIS,B. 1992. The role of alliances in social inheritance of rank among female primates. In: Coalitions and Alliances in Humans and Other Animals, HARCOURT,A. H.; DE WAAL,E B. M. (eds.), Oxford Univ. Press, Oxford, pp. 29-59. CHAPA1S B.; GAUTHIER,C.; PRUD'HOMME,J. 1995. Dominance competition through affiliation and support in Japanese macaques: an experimental study. Int. J. Primatol., 16: 521-536. CHAPAIS, B.; GAUTHIER,C.; PRUD'HOMME,J.; VASEY, P. 1997. Relatedness threshold for nepotism in Japanese macaques. Anim. Behav., 53:1089-1101. CHAPAIS B.; GIRARD, M.; PRIMi, G. 1991. Non-kin alliances and the stability of matrilineal dominance relations in Japanese macaques. Anita. Behav., 41: 481-491. CHAPAIS B.; LECOMTE,M. 1995. Induction of matrilineal rank instability by the alpha male in a group of Japanese macaques. Ame~ J. Primatol., 36: 299-312. CHAPAIS B.; PRUD'HOMME,J.; TEIJEIRO, S. 1994. Dominance competition among siblings in Japanese macaques: constraints on nepotism. Anim. Behav., 48:1335-1347. CHAPAIS B.; ST-PIERRE,C. G. 1997. Kinship bonds are not necessary for maintaining matrilineal rank in captive Japanese macaques, lnt. J. Primatol., 18: 375-385. CHENEY, D. L. 1977. The acquisition of rank and the development of reciprocal alliances among free-ranging immature baboons. Behav. Ecol. Sociobiol., 2:303-318. CHENEY, D. L. 1983. Extra-familial alliances among vervet monkeys. In: Primates Social Relationships, HINDE, R. H. (ed.), Blackwell, London, pp. 278-286. CHENEY, D. L.; SEYFARTH,R. M. 1990. How Monkeys See the World : Inside the Mind of Another Species. Univ. of Chicago Press, Chicago. CHIKAZAWA,D.; GORDON,T. P.; BEAN,C. A.; BERNSTEIN,1. S. 1979. Mother-daughter dominance reversals in rhesus monkeys (Macaca mulatta). Primates, 20: 301-305. DATrA, S. 1988. The acquisition of dominance among free-ranging rhesus monkey siblings. Anita. Behav., 36: 754-772. DUNBAR,R. I. M. 1980. Determinants and evolutionary consequences of dominance among female gelada baboons. Behav. Ecol. Sociobiol., 7: 253-265. FURUICHI,T. 1983. Interindividual distance and influence of dominance on feeding in a natural Japanese macaque troop. Primates, 24: 445-455. GOUZOULES,H. 1975. Maternal rank and early social interactions of infant stumptailed macaques, Macaca arctoides. Primates, 16: 405-418. GOUZOULES,H. 1980. A description of genealogical rank changes in a troop of Japanese monkeys (Macaca fuscata). Primates, 21: 262-267.

334

N. KUTSUKAKE

HASEGAWA, T.; HIRAIWA,M. 1980. Social interactions of orphans observed in a free-ranging troop of Japanese monkeys. Folia Primatol., 33:129-t58. HAUSFATER,G.; ALTMANN,J.; ALTMANN,S. 1982. Long-term consistency of dominance relations among female baboons (Papio cynocephalus). Science, 217: 752-755. HILL, D. A. 1999. Effects of provisioning on the social behaviour of Japanese and rhesus macaques: implications for socioecology. Primates, 40: 187-198. HILL, D. A.; OV,AYASU,N. 1995. Absence of 'youngest ascendancy' in the dominance relations of sisters in wild Japanese macaques (Macacafuscata yakui). Behaviour., 132: 367-379. HILL, D. A.; Og,AYASU,N, 1996. Determinants of dominance among female macaques: nepotism, demography and danger. In: Evolution and Ecology of Macaque Societies, FA, J. E.; LINDBURG, D. G. (eds.), Cambrigde Univ. Press, Cambrigde, pp. 459-472. HORROCKS, J.; HUNTE, W. 1983. Maternal rank and offspring rank in vervet monkeys: an appraisal of the mechanism of rank acquisition. Anita. Behav., 31: 772-782. [TOIGAWA,N.; TANAKA,T.; UKAI, N.; FuJII, H.; KUROKAWA,T.; KOYAMA,T.; ANDO, A.; WATANABE,Y.; IMAKAWA,S. 1992. Demography and reproductive parameters of a free-ranging group of Japanese macaques (Macacafuscata) at Katsuyama. Primates, 33: 49-68. KAWA1,M. 1958. On the rank system in a natural group of Japanese monkeys, I: The basic and the dependent rank. Primates, 1: 111-130. (in Japanese) KAWAMURA,S. 1958. The matriarchal social order in the Minoo-B group: a study on the rank system of Japanese macaque. Primates, 1: 149-156. (in Japanese) KOYAMA, N. 1967. On dominance rank and kinship of a wild Japanese monkey troop in Arashiyama. Primates, 8:189-216. KOYAMA, N. 1970. Changes in dominance rank and division of a wild Japanese monkey troop in Arashiyama. Primates, 11 : 335-390. KOYAMA,N.; TAKAHATA,Y.; HUFFMAN,M. A.; NORIKOSHI,K.; SUZUKI, H. 1992. Reproductive parameters of female Japanese macaques: thirty years data from the Arashiyama troops, Japan. Primates, 33: 33-47. KURLAND, J. A. 1977. Contributions to Primatology, Vol. 12: Kin Selection in the Japanese Monkey. Karger, Basel. MARTIN, P.; BATESOn,P. 1990. Measuring Behavior Cambridge Univ. Press, Tokyo. MlSSAKIAN,E. A. 1972. Genealogical and cross-genealogical dominance relations in a group free-ranging rhesus monkeys (Macaca mulatta) on Cayo Santiago. Primates, 13: 169-180. Morl, A. 1977. lntra-troop spacing mechanism of the wild Japanese monkeys of the Koshima troop. Prhnates, 18:331-357. MORI, A.; WATANABE,K.; YAMAGUCHLN. 1989. Longitudinal changes of dominance rank among the females of the Koshima group of Japanese monkeys. Primates, 30: 147-173. NAKAMICHI,M.; 1TOIGAWA,N.; IMAKAWA,S.; MACHIDA,S. 1995. Dominance relations among adult females in a free-ranging group of Japanese monkeys at Katsuyama. Amer J. Primatol., 37: 241-251. PAUL, A.; KUESTEr, J. 1987. Dominance, kinship and reproductive value in female Barbary macaques (Macaca sylvanus) at Affenberg Salem. Behav. Ecol. Sociobiol., 21: 323-331. PRUD'HOMME,J.; CHAPAIS,B. 1993a. Rank relations among sisters in semi-free-ranging Barbary macaques (Macaca sylvanus), hzt. J. Primatol., 14: 405-420. PRUD'HOMME, J.; CHAPAIS, B. 1993b. Aggressive interventions and matrilineal dominance relations in semifree-ranging Barbary macaques (Macaca sylvanus). Primates, 34:271-283. PUSEY, A. E.; PACKEr, C. 1987. Dispersal and philopatry. In: Primate Societies, SMUTS,B. B.; CHENEY, D. L.; SEYFArTH, R. M.; WRANGr~AM,R. W.; STRUHSAKEr,T. T. (eds.), The Univ. of Chicago Press, Chicago, pp. 250-266. SILK, J. B. 1982. Altruism among female Macaca radiata: explanations and analysis of patterns of grooming and coalition formation. Behaviour, 79: 162-188. SILK, J. B.; SAMUELS,A.; RODMAN,P. S. 1981. Hierarchical organization of female Macaca radiata in captivity. Primates, 22: 84-95. SILK, J. B.; SE'ffArtn, R. M.; CnEnEu D. L. 1999. The structure of social relationships among female savanna baboons in Moremi Reserve, Botswana. Behaviour, 136: 679-703. SOUMAn, A. G.; YOKOTA,N. 1991. Female rank and feeding strategies in a free-ranging provisioned troop of Japanese macaques. Folia Primatol., 57:191-200. TAKAnA'rA, Y. 1982. Social relations between adult males and females of Japanese monkeys in the Arashiyama B troop. Primates, 23: 1-23.

Matrilineal Rank Inheritance in Japanese Macaques

335

TAKAHATA,Y. 1991. Diachronic changes in the dominance relations of adult female Japanese monkeys of the Arashiyama B group. In: The Monkeys ofArashiyama, FEDIGAN,L. M.; ASQUITH,P. J. (eds.), State Univ. of New York Press, Albany, pp. 123-139. THIErRY, B. 1990. Feedback loop between kinship and dominance: the macaque model. J. Theor. Biol., 145: 511-521. DE VRIES, H. 1995. An improved test of linearity in dominance hierarchies containing unknown or tied relationships. Anita. Behav. , 50:1375-1389. DE WAAL,E B. M. 1977. The organization of agonistic relations within two captive groups of Java-monkeys (Macaca fascicularis). Z. Tierpsychol., 44: 225-282. DE WAAL,E B. M.; LUTTRJ~LL,L. M. 1985. The formal hierarchy of rhesus macaques: an investigation of bared-teeth display. Amen J. Primatol., 9: 73-85. DE WAHL, E B. M.; LUTTrELL, L. M. 1986. The similarity principle underlying social bonding among female rhesus monkeys. Folia Primatol., 46: 215-234. DE WAAL,E B. M.; LUTTRELL,L. M. 1989. Toward a comparative socioecology of the genus Macaca: different dominance styles in rhesus and stumptail monkeys. Amer. J. Primatol.,19: 83-109. WADA, K.; ICHIKI,Y. 1980. Seasonal home range use by Japanese monkeys in the snowy Shiga Heights. Primates, 21: 468-483. WALTERS,J. 1980. Interventions and the development of dominance relationships in female baboons. Folia Primatol., 34: 61-89. WATANABE,K.; MORI, A.; KAWAI,M. 1992. Characteristic features of the reproduction of Koshima monkeys, Macacafuscatafuscata: a summary of thirty-four years of observation. Primates, 33: 1-32.

- -

Received: January 27, 1999; Accepted: May 24, 2000

Author's Name and Address: NOBUYUK1KUTSUKAKE,Department gf Cognitive and Behavioral Science, Graduate School (ff Arts and Sciences, The University ()f"Tokyo, 3-8-1 Komaba, Meguro-ku, Tokyo 153-8902, Japan. e-mail: [email protected]

Matrilineal rank Inheritance varies with absolute rank in ...

sals and noted the importance of alliances with kin (CHAPAIS, 1988) and ...... On the rank system in a natural group of Japanese monkeys, I: The basic and the ...

919KB Sizes 2 Downloads 517 Views

Recommend Documents

Learning to Rank with Ties
works which combine ties and preference data using statis- ... learning ranking functions from preference data have been .... Figure 1: Illustration of ties. 4.

Robust Low-Rank Subspace Segmentation with Semidefinite ...
dimensional structural data such as those (approximately) lying on subspaces2 or ... left unsolved: the spectrum property of the learned affinity matrix cannot be ...

Learning to Rank with Selection Bias in ... - Research at Google
As a result, there has been a great deal of research on extracting reliable signals from click-through data [10]. Previous work .... Bayesian network click model [8], click chain model [17], ses- sion utility ..... bels available in our corpus such a

Automatic Rank Determination in Projective ...
(PNMF), introduced in [6–8], approximates a data matrix by its nonnegative subspace ... is especially desired for exploratory analysis of the data structure.

Reverse Split Rank - Semantic Scholar
there is no bound for the split rank of all rational polytopes in R3. Furthermore, ... We say that K is (relatively) lattice-free if there are no integer ..... Given a rational polyhedron P ⊆ Rn, we call relaxation of P a rational polyhe- dron Q âŠ

Rank 2016.pdf
... Channing School 9 3 ↑6 Magdalen College School 48 33 ↑15 St Catherine's, Bramley 4 4 - North London Collegiate School 131 34 ↑97 Immanuel College.

rank-def MRI.pdf
This requires minimizing the rank of the image matrix subject to data constraints,. which is unfortunately a nondeterministic polynomial time (NP) hard problem.

Online Rank Aggregation
We consider an online learning framework where the task is to predict a ... a loss defined as Kendall tau distance between the predicted permutation and the.

Rank Advancement Dashboard_FAQs.pdf
The Sales Month Calendar will appear through a pop-up. It ... Qualified – shows the order details of the sales that you personally made or by your retail sales.

Supervised Rank Aggregation
The optimization for. Markov Chain based methods is not a convex optimization .... reasonable to treat the results of the search engines equally. To deal with the ...

Reverse Split Rank - Semantic Scholar
Watson Research Center, Yorktown Heights, NY, USA. 3 DISOPT, Institut de ..... Given a rational polyhedron P ⊆ Rn, we call relaxation of P a rational polyhe- .... operator (hence, also of the split closure operator) applied to Qi are sufficient to.

Online Rank Aggregation
Then the online rank aggregation problem consists of the following .... The first idea is that we regard a permutation as a N(= n(n − 1)/2) dimensional comparison.

A PLANAR DIFFUSION WITH RANK-BASED ...
well-posed? If so, what is the probabilistic structure of the resulting two-dimensional diffusion process? What are its transition probabilities? How does it look like when time is reversed? Questions like these arise in the context of systems of dif

Low-Rank Spectral Learning with Weighted Loss ... - EECS @ Michigan
domains. 1 INTRODUCTION. Predictive state representations (PSRs) are compact models of ...... Low-Rank Spectral Learning with Weighted Loss Functions. 0. 50. 100. 10−5. 100. |H| .... Dimension-free concentration bounds on hankel ma-.

Learning to Rank with Joint Word-Image ... - Research at Google
notation that can scale to learn from such data. This includes: (i) .... tors, which is expensive for large Y . .... computing fi(x) for each i ∈ Y as the WARP loss does.

Discriminative Dictionary Learning with Low-Rank ...
matrix recovery theory and apply a low-rank regularization on the dictionary. ..... thesis, Massachusetts Institute of Technology, 2006. [8] M. Aharon, M. Elad, and ...

Learning to Rank Recommendations with the k ... - Research at Google
optimize precision at k than the WARP loss, which WARP was designed for. Secondly, it can optimize novel metrics like maximum mean rank which we believe ...

Learning to Rank with Joint Word-Image Embeddings
like a fast algorithm that fits on a laptop, at least at ..... Precision at 1 and 10, Sibling Precision at 10, and Mean ..... IEEE Conference on Computer Vision and.

A PLANAR DIFFUSION WITH RANK-BASED ...
[1] Banner, A., Fernholz, E.R. & Karatzas, I. (2005) Atlas models of equity markets. Annals of Applied Probability 15, 2296-2330. [2] Barlow, M.T. (1988) Skew ...

Contextual Decision Processes with Low Bellman Rank ...
any function of history that expresses a good policy & value function. ○ e.g., last 4 frames of images in Atari games. ○ e.g., (state, time-step) for finite-horizon tabular MDPs. ○ An episode: ... among the surviving functions. ○ Repeat until

otto rank will therapy pdf
There was a problem previewing this document. Retrying... Download. Connect more apps... Try one of the apps below to open or edit this item. otto rank will ...

Merging Rank Lists from Multiple Sources in Video ... - Semantic Scholar
School of Computer Science. Carnegie ... preserve rank before and after mapping. If a video ... Raw Score The degree of confidence that a classifier assigns to a ...