ANIMAL BEHAVIOUR, 2007, 73, 501e512

doi:10.1016/j.anbehav.2006.09.007

Core area quality is associated with variance in reproductive

success among female chimpanzees at Kibale National Park

M E L IS SA EM E R Y T H O MP S O N, S O NYA M . KA HLENBERG, I AN C. GI L BY & RIC HA RD W. WRANGH A M

Department of Anthropology, Harvard University

(Received 1 May 2006; initial acceptance 13 June 2006;

final acceptance 20 September 2006; published online 31 December 2006; MS. number: A10438R)

Female East African chimpanzees, Pan troglodytes schweinfurthii, tend to range apart from each other in dis-
persed core areas, and they have dominance interactions with each other so rarely that it is difficult for
observers to assess a dominance hierarchy. Nevertheless female chimpanzees can have high variance in fit-
ness. Here, we test the hypothesis that female chimpanzee fitness variance is associated with variation in
the foraging quality of their ranges. We studied range usage of 21 wild adult female chimpanzees within
the Kanyawara community, Kibale National Park, Uganda. Core areas of individuals remained stable over
a 9-year period and varied in their density of preferred foods. Females in neighbourhoods containing more
preferred foods had elevated ovarian hormone production, shorter birth intervals and higher infant survi-
vorship. Our results thus suggest that superior access to food may have enabled some community females
to reproduce more successfully than others. Although dominance interactions are less frequent among fe-
males than among males of this species, we propose that the intensity of selection on intrasexual compe-
tition may be similar between the sexes. We discuss potential applications to other fissionefusion species.

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

Keywords: chimpanzee; endocrinology; habitat ecology; intrasexual competition; Pan troglodytes schweinfurthii; ranging;
reproductive ecology; reproductive success

Sexual selection theory has traditionally pointed to high commonly associated with femaleefemale aggression
variance in male reproductive success as a strong selective and philopatry, and are maintained by alliances among fe-
pressure for sexual ornamentation and competitive ability, male kin (Wrangham 1980; Koenig 2002).
whereas covariates of female reproductive success are less Among chimpanzees, Pan troglodytes, by contrast, fe-
emphasized (Trivers 1972). For female mammals, socio- males show high fitness variance without consistent fe-
ecological models focus on the central role of resource ac- male aggression, philopatry or alliances. Dominant
cess as a limiter to reproductive success (Gadgil & Bossert females in one community at Gombe reproduced more
1970; Schneider & Wade 2000) and a major determinant quickly and had higher infant survival and faster matur-
of social relationships (Wrangham 1980; Sterck et al. ing daughters than did subordinates (Pusey et al. 1997).
1997; Koenig 2002). Yet, female dominance interactions were rare, as they
Resource competition among female primates fre- tend to be in this species compared with both conspecific
quently results in dominants outperforming subordinates males and females in other primates (de Waal 1982; Good-
(Gouzoules et al. 1982; Harcourt 1987; Bercovitch & Strum all 1986; Nishida 1989). Neither aggressive interactions
1993; Ellis 1995; Altmann & Alberts 2003b; Creel 2005), nor ritualized displays occur often enough to have a mean-
sometimes due to reproductive suppression of subordi- ingful impact on the social or feeding behaviour of fe-
nates (Abbott 1984; Epple & Katz 1984; Ziegler et al. males, and, even in the 22 years covered by the Gombe
1987; Abbott 1987; Barrett et al. 1990, 1993; Kuederling study, some female dyads were never observed to have
et al. 1995; Ziegler & Sousa 2002; Saltzman et al. 2004). a single dominance interaction. Females typically transfer
Rank and fitness differentials within social groups are out of their natal communities at adolescence and rarely
form alliances in their new communities. The observed
Correspondence: M. Emery Thompson, Department of Anthropology, variance in reproductive success, therefore, does not con-
Harvard University, 11 Divinity Avenue, Cambridge, MA 02138, U.S.A. form to expected patterns of social relationships, and has
(email: [email protected]). not been explained.
501
0003e 3472/07/$30.00/0 Ó 2006 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd. All rights reserved.
502 ANIMAL BEHAVIOUR, 73, 3

Although female chimpanzees rarely fight over food, recorded simultaneously by three to four observers (i.e.
there is evidence that energy availability influences their 2e3 Kibale Chimpanzee Project field staff and 1e2
reproductive rates, as expected. First, across populations, graduate students). Field staff collected group scan sam-
variation in interbirth intervals is correlated with differ- ples every 15 min, recording: (1) location of chimpanzees
ences in habitat quality (Knott 2001; Emery Thompson on a trail map, (2) party composition, (3) oestrous status
2005a). Second, timing of conception in chimpanzees co- of females and (4) food species and part(s) being con-
incides with the availability of high-quality foods, which sumed. This study incorporates results from 82 956 scan
show considerable temporal variability (Sherry 2002; Em- samples (20 649 h) collected between January 1996 and
ery Thompson 2005a). Food items also show considerable May 2004.
spatial heterogeneity within chimpanzee home ranges,
suggesting an additional dimension to variation in energy
availability. Chimpanzees live in a fissionefusion social Ranging Data Analysis
system in which a group of bonded males aggressively
In this analysis, we were concerned with the concen-
defends a home range, including the food resources and fe-
tration of ranging in particular locations within the larger
males within it (Nishida 1968; Sugiyama 1968; Wrangham
territory. To assess this we recorded the locations of
1975; Williams et al. 2004). While community members
chimpanzees on a 500  500-m grid superimposed over
maintain long-term affiliative ties, members of the com-
the trail system, and calculated the frequency that each
munity are typically dispersed across the home range in
chimpanzee was observed in each of 380 grid cells as a per-
smaller, temporary associations (parties) that may vary
centage of the total scans in which that chimpanzee was
in size in accordance with the presence and number of
observed. Because sexual receptivity probably influences
cycling females with sexual swellings, and the social affin-
female ranging patterns (e.g. Hasegawa 1990), we only
ities of community members (Wrangham et al. 1996;
considered nonoestrous observations. To test for site fidel-
Matsumoto-Oda 1999; Pepper et al. 1999; Emery Thomp-
ity, we divided the study into two time periods (1996e
son & Wrangham 2006). At Gombe, mothers share a com-
2000, 2001e2004) in which we included adult individuals
munity range but each female tends to maintain her own
and subadults who ranged independently of their mothers
small ‘core area’ where she spends the majority of her
and for whom reproductive information was available
time (Wrangham 1979). Core areas cluster into localized
(Table 1). Several females died in late 2000 or early 2001,
‘neighbourhoods’ within the community range (Williams
so this division represents an important demographic
et al. 2002). Therefore, it is possible that individual
change in the community.
females increase their fitness by accessing higher-quality
To compare range usage across females, we conducted
areas of the community range, rather than by competing
hierarchical cluster analyses of the grid cell usage data for
directly at individual food sites (Pusey et al. 1997;
each individual. This procedure yields a dissimilarity
Williams et al. 2002).
matrix by calculating the difference (squared Euclidean
To test this hypothesis, we collected data on patterns of
distance) between grid cell usage frequencies for each
range use, ovarian hormone levels and reproductive
chimpanzee dyad, and then produces a dendrogram de-
parameters of females in the Kanyawara community of
scribing their relative relationships (median clustering
chimpanzees in Kibale National Park, Uganda. We pre-
method) (Romesburg 2004). Based on this dendrogram,
dicted that (1) as in Gombe, Kanyawara females would
the cluster of the females with the most similar ranging
maintain differentiated core areas, and (2) females in
patterns were considered members of a single ‘neighbour-
higher-quality core areas would have a reproductive ad-
hood’, and the range of each other female was categorized
vantage, as indexed by ovarian function, birth rate and
in relation to this group by the evaluation of range histo-
offspring survival.
grams. Grid locations of males were also included in the
cluster analysis for comparison.
To determine the location of each female neighbour-
METHODS
hood within the home range, we calculated the mean
Study Site and Population percentage usage of each grid square by all females within
a particular classification based on the cluster analysis
The Kanyawara home range spans approximately (central, northern, or southern). Our observations of
32 km2 (Wilson 2001), consisting of approximately 60% chimpanzees were neither randomly nor evenly distrib-
moist deciduous forest, with small areas of swamp, grass- uted, so grid cell usage was expressed as a percentage more
land and colonizing forest (Chapman & Wrangham or less than expected from the distribution of all chim-
1993). The community consisted of 47 chimpanzees at panzee observations.
the beginning of the current study in 1996, and numbered
38 individuals (including 10 adult males and 19 subadult
and adult females) at the end of the study in May 2004. Neighbourhood Habitat Quality
The Kanyawara region can be divided into forest sectors
Data Collection based on logging history and general forest type. Skorupa
(1988) reported primate habitat quality statistics for
Chimpanzees were searched for daily or were followed Kanyawara according to these forest sectors, including
as they left their sleeping sites. Behaviour was typically the impact of logging outtake and densities of stems fruit
 EMERY THOMPSON ET AL.: RANGE AND REPRODUCTION OF CHIMPANZEES 503

Table 1. Kanyawara chimpanzee subject females and sample sizes

1996e2000 2001e2004

Nonoestrous Nonoestrous Offspring born

Year of scans Grid cells scans Grid cells Urine (1988e)
Female birth* (ca. h/month)y usedz (ca. h/month)y usedz samplesx survived/died**

AL 1982 8984 (37) 94 9025 (55) 78 114 2/1

AR 1943 6629 (28) 65 4298 (26) 41 73 2/0
BL 1960 3342 (14) 53 6745 (46) 61 36 4/0
EK 1974 2177 (9) 34 2205 (13) 49 21 2/1
FG 1955 6987 (67) 81 n/a d 14 1/1
GO 1957 2123 (9) 42 41 (5) 4 n/a 1/0
HL 1975 83 (1) 8 n/a d n/a 0/0
JK 1992 n/a d 8276 (50) 75 47 n/a
JO 1960 2006 (8) 58 94 (12) 9 10 1/0
KL 1970 7745 (32) 83 n/a d 70 0/2
LP 1955 17410 (73) 98 9979 (61) 90 185 2/1
LR 1989 n/a n/a 7571 (46) 77 224 1/1
MU 1970 463 (2) 15 522 (3) 28 7 1/2
NG 1955 136 (2) 9 n/a d n/a 2/0
NL 1982 6438 (37) 80 8162 (500) 74 158 3/0
OU 1979 12648 (53) 83 12925 (79) 74 220 1/0
PE 1970 1198 (5) 40 n/a d 2 2/1
PU 1955 2535 (11) 48 1883 (20) 44 9 3/0
TG 1980 11183 (47) 71 11233 (68) 71 139 3/0
UM 1981 604 (3) 21 565 (3) 25 16 3/0

*Chimpanzees are a male philopatric species, so most female ages are estimates.
yScans of the subject females during periods of full anogenital swelling were excluded from analysis.
zNumber of unique grid cells in which each female had been observed.
xUrinary steroid analysis controlled for individual and longitudinal differences in sampling frequency, see Methods. Note that several individuals
died or disappeared before urine collections began.
**Surviving ¼ lived to age 4 or had not yet reached age 4 but was alive at the time of the study.

trees and primates. Where we could conclusively match at a time. So, we also calculated the proportion of
a forest sector to a chimpanzee neighbourhood, we report unique feeding locations that were found within each
these measures of general habitat quality (Table 2). neighbourhood. Third, as a measure of female access to
Kanyawara chimpanzees preferentially consume non- resources in each neighbourhood, we calculated the per-
fig fruits when available (Wrangham et al. 1996; Emery centage of preferred fruits, all fruits and fallback foods
Thompson 2005a), and three particularly preferred spe- (piths and leaves) constituting the diets of female parties
cies (Mimusops bagshawei, Pseudospondias microcarpa, in each grid cell. Finally, we determined the degree to
and Uvariopsis congensis) show strong relationships to re- which the distribution of events of non-fig fruit feeding
productive timing (Sherry 2002; Emery Thompson by chimpanzees predicted the ranging patterns of fe-
2005a), although they are sporadically available (Wrang- males in each neighbourhood. We calculated the regres-
ham et al. 1996; Emery Thompson 2005a). In any given sion of the percentage of non-fig fruit use occurring
month, these drupe fruits may constitute 0e83% of the within each cell against the mean percentage of observa-
diet (X ¼ 27:5, Emery Thompson, unpublished data). To tions that females of a particular neighbourhood spent
evaluate the quality of each female neighbourhood, we in those cells. Although we generally expected female
determined the locations of all episodes of feeding on ranging habits to correlate with feeding opportunities,
non-fig fruits by chimpanzees and compared these grid we also expected that females with the best quality
cells to the cells constituting each neighbourhood. We core areas and the best access to resources would have
then calculated four measures of non-fig fruit availability ranges that accorded most closely to the use of preferred
and feeding intensity by neighbourhood. First, we calcu- foods.
lated the number of observations of non-fig fruit feeding
that occurred within each neighbourhood and used
a chi-square analysis to evaluate whether this distribu- Urinary Steroid Analysis
tion differed from an even distribution across neighbour-
hoods. This measure may be biased by more intensive Field personnel have collected urine samples opportu-
observations in certain regions. However, we are confi- nistically from all chimpanzees since November 1997.
dent that chimpanzees would have been observed feed- Samples were collected on plastic sheets or pipetted
ing at least once in each location of their three most directly from vegetation (Knott 1997) and frozen until
preferred fruit species because these fruits tend to occur analysis at the Primate Reproductive Ecology Laboratory
in localized patches that are exploited for days or weeks at Harvard University. All samples were analysed for
504 ANIMAL BEHAVIOUR, 73, 3

oestrone conjugates (E1C) and pregnanediol-3-glucuro-

nide (PdG), metabolites of ovarian steroid hormones, us-
ing enzyme immunoassay reagents provided by the

reduction
Basal area

(%)yy

24.9
46.6
50.3
Clinical Endocrinology Laboratory at University of Cali-

d
fornia, Davis (C. J. Munro). Thorough assay procedures
Logging history

and validations are provided elsewhere (Emery Thomp-

son 2005b). To control for uneven sampling across fe-
males and expected hormonal fluctuations over time
1967e1969
Timber out-

(Emery Thompson 2005a), we calculated an average E1C

(m3/ha)

14.4
20.9
17.0
take

and PdG value for each unique combination of female,

0

month and reproductive state. We controlled for repro-

ductive state by converting these monthly averages into
z scores relative to the mean of all Kanyawara females
in each reproductive state (Emery Thompson 2005b),
primate
index**
Total

82.5
73.2
45.9
35.1

and then considered each female’s grand mean relative

Table 2. Habitat characteristics and timber impact in forest sectors comprising central and northern neighbourhood areas (from Skorupa 1988)

to others.
Ovarian hormone levels are a powerful tool for evalu-
ating fecundity at a proximate level (Ellison 1995), be-
frugivore
richness,
Species

treesx

cause they reflect the size and function of the ovarian

9.5
6.6
6.9
11.1

follicle and corpus luteum and are associated with in-

creased probabilities of ovulation, fertilization and im-
plantation (Eissa et al. 1986; Yoshimura & Wallach 1987;
Basal area,
frugivore

Lenton et al. 1988; Akman et al. 2002). Thus, higher ovar-

(m2/ha)
dietsz

21.7
21.1
13.6
10.9

ian hormone levels predict increased conception probabil-

ity in chimpanzees (Emery Thompson 2005b), as well as
in humans (Lipson & Ellison 1996), gorillas (Nadler & Col-
Habitat quality measures

lins 1991) and baboons (Wasser 1996).

Canopy
15 m
(%)

72
50
32
21

Reproductive Data
Interbirth intervals were calculated using KaplaneMeier
(stems/
density

survival analyses on all complete and incomplete inter-

Ficus

ha)

4.1
6.5
0.6
0.5

vals. Twenty-four of 37 infants were first encountered as

**Average proportion of maximum abundance achieved by Kibale primate species in each plot.

newborns (<2 weeks old), and the birthdates of the

remainder were estimated based on the size of the infant
Large stem

and the time since the mother was last observed, with all
(stems/
density

zTrees in diets of frugivorous monkeys, Lophocebus albigena and Cercopithecus mitis.

25.2
12.0

10.5
8.0
ha)y

error estimates less than 6 months (1 month, 7 infants;

2 months, 4 infants; 6 months, 2 infants). If a female
switched neighbourhoods during the study, we catego-
rized her in the neighbourhood in which she spent the
Total stem

majority of the birth interval. Because two southern

(stems/
density

yAt least 2.1 m in circumference at breast height or above buttress.

255.9
267.2
125.1
127.6
ha)*

females would have contributed only three birth intervals

xSpecies/100 stems, diets of frugivorous monkeys (see above).

to these data, they were excluded from this analysis.

Although it is feasible that infants born to rarely encoun-
tered females could have been missed if infants died soon
Area (ha)

after, the oestrous cycle monitoring, ovarian steroid

300
390
360
1000

*At least 35 cm in circumference at breast height.

yyRelative to undistributed Parinari forest (K-30).

analysis and frequent pregnancy testing (Aimstick, Craig

Medical, Vista, California, U.S.A.) make it unlikely that
a pregnancy would have gone undetected.
Offspring survivorship was assessed with a Kaplane
neighbourhood
Chimpanzee

Northern
Northern

Meier survival analysis for 41 offspring born after 1988

location

Central
Central

whose mothers’ neighbourhoods could be assigned. These

are understood to be underestimates of life expectancy at
birth, since individuals born during observation could
only have reached an age of 17. We considered infants
that were orphaned during infancy or juvenile females
Forest area

K-12/13/

that were believed to have emigrated as incomplete

intervals (i.e. alive) terminating on the death of the
K-30
K-14
K-15

17

mother or the date of transfer, respectively. Infant survival

data were available up to 15 May 2005.
 EMERY THOMPSON ET AL.: RANGE AND REPRODUCTION OF CHIMPANZEES 505

RESULTS neighbourhood extended further south into previously

underexploited areas on the eastern edge of the home
Female Spatial Distribution range (Fig. 2).
During both periods, cluster analysis revealed that
Hierarchical cluster analyses of ranging distributions for
central females showed tight similarity in ranging pat-
both periods revealed a distinct cluster of females whose
terns, whereas northern females, who were all clearly
ranges were most similar (Fig. 1). The range of these fe-
north of the central cluster, had more differentiated
males was located centrally to other females whose activ-
ranging patterns within their neighbourhood. That is,
ities were concentrated near the northern and southern
while northern females tended to use the same grid cells
borders (Fig. 2). In both periods, the range use of all-
in the northern area of the home range, individuals used
male parties was indistinguishable from the central female
these cells differently.
cluster.
In the first period, 1996e2000, the central neighbour-
hood contained eight of 18 females. The eldest female in the
Core Area Quality
community (AR) fell just outside of the main cluster,
showing foci of range use in both central and southern To compare habitat quality between northern and
areas. Two females (GO, BL) ranged considerably south of central neighbourhoods, we reviewed forest surveys con-
the central group and formed a second spatial cluster. The ducted by Skorupa (1988). Two forest sectors surveyed by
remaining eight females ranged to the north, with the Skorupa, comprising 13.6 sq km, fell unambiguously
majority forming a third cluster with similar ranging habits. within the range of the northern females, while two addi-
In the second period, 2001e2004, nine of 15 females tional sectors, comprising 6.9 sq km, were contained
composed the central cluster. This group included all within the central female range. These surveys showed
living members of the original central neighbourhood, large differences in forest quality. The northern neigh-
suggesting strong site fidelity. Indeed, females in the bourhood had suffered considerably greater logging im-
northern and southern neighbourhoods during this pe- pact (1967e1969). Thus, the central neighbourhood had
riod had also occupied the same neighbourhoods in roughly double the density of total trees, including both
2001e2004. However, a number of changes in commu- large trees and fig trees (Ficus spp.), as well as a higher den-
nity demography occurred in early 2001, affecting range sity and diversity of primates.
distributions. One of the southern females (GO) died of The data on density of trees, figs and frugivorous
respiratory illness, and the only remaining southern primates suggest that chimpanzees in the central neigh-
female (BL) began to be observed more frequently in bourhood had more abundant or higher-quality food than
central areas. A female from the northern neighbourhood those in the northern neighbourhood. Feeding data
(JO) also died following an illness, after which her support this. During the period 1996e2000, 53% of
adolescent daughter (JK) moved into the central neigh- observations of non-fig fruit feeding occurred within
bourhood. Thus, in about 2001, the southern neighbour- grid cells assigned to the central neighbourhood, 55%
hood completely disappeared. Specific areas of the home occurred within the southern neighbourhood cells, and
range used by central and northern females were similar to only 19% occurred in northern neighbourhood cells
those used in 1996e2000, although the northern (chi-square test: c22 ¼ 1596:69, P < 0.0001 based on

(a) 1996–2000 (b) 2001–2004

Rescaled distance cluster combine Rescaled distance cluster combine
0 5 10 15 20 25 0 5 10 15 20 25
ID ID
OU OU
TG TG
KL NL
LP AL
FG LP
Males BL
AL JK
NL LR
AR* Males
BL AR
GO EK
JO PU
PU MU
PE UM
EK JO
UM GO
MU
HL
NG
Figure 1. Dendrogram resulting from hierarchical cluster analysis of female chimpanzee range usage at Kanyawara, median method.
Bold ¼ central; italics ¼ strongly northern; underline ¼ strongly southern; * ¼ central, tending south of main cluster. (a) 1996e2000, (b)
2001e2004.
506 ANIMAL BEHAVIOUR, 73, 3

1996−2000 2001−2004

N=8 N=5

15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
21
32
33

15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
21
32
33
B B
C C
D D
E E
F F
G G
H H
I I
J J
K K
Northern L L
M M
N N
O O
P P
Q Q
R R
S S
T T
U U
N=8 N=9

15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
21
32
33
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
21
32
33
B B
C C
D D
E E
F F
G G
H H
I I
J J
Central K K
L L
M M
N N
O O
P P
Q Q
R R
S S
T T
U U
N=2
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
21
32
33

B
C
D
E
F Grid cell use vs expected
G
H Usage >5x expected
I 2.1–5.0x expected
J
Southern K 1.6–2.0x expected
L 1.1–1.5x expected
M <1.1x expected
N
O Excluded, location of <0.01% of
P total obsevations
Q
R
S
T
U

Figure 2. Range concentrations of female Kanyawara chimpanzees within each spatial cluster: (a) 1996e2000, (b) 2001e2004. Shaded areas
illustrate the observed/expected percentage of utilization of each grid square (500  500 m) by that group of females, with expected distri-
butions based on the frequency of all chimpanzee observations per grid cell. Blank cells outside of border were excluded from analysis; they
were not used by chimpanzees or were used in less than 0.01% of total observations.

observed versus expected frequencies). Similarly, during Males were also observed more often in the central
2001e2004, 45% of observations of non-fig fruit feeding neighbourhood than in the northern neighbourhood
occurred in the central neighbourhood, while only 28% (Fig. 1), as expected if the central neighbourhood had
occurred in the northern neighbourhood (c21 ¼ 308:49, more fruit available. This raises the possibility of biased
P < 0.0001). sampling if groups of males were more likely to be
 EMERY THOMPSON ET AL.: RANGE AND REPRODUCTION OF CHIMPANZEES 507

observed or followed by researchers. We therefore calcu- 90

% Diet, all-female parties

1996–2000 2001–2004
lated the number of unique feeding locations within 80
each neighbourhood, without respect to intensity of use. 70
Again, feeding locations for the three preferred fruit 60
species were strongly concentrated within the southern 50
and central neighbourhoods (1996e2000: c22 ¼ 21:65, 40
30
P < 0.0001; 2001e2004: c21 ¼ 7:37, P < 0.01; Table 3).
20
Northern areas contained very few preferred fruit trees.
10
In addition to the evidence that preferred foods may 0
have been more available and more frequently used in the Preferred Total Fallback Preferred Total Fallback
southern and central versus northern neighbourhoods, we fruits fruits foods foods fruits foods
found evidence that the dietary composition of females, Figure 3. Dietary composition of all-female parties ranging in
specifically, varied in the same manner according to range. grid cells constituting each neighbourhood. Bars represent the
Each neighbourhood contained at least some cells in mean  standard error across cells. ,: northern neighbourhood;
which preferred (non-fig) fruits were the only item -: central neighbourhood; : southern neighbourhood.
consumed, and each neighbourhood contained at least
some cells in which preferred fruits were never consumed, In summary, data on tree density, frugivore density, and
making significance testing impractical. However, in both locations and patterns of chimpanzee fruit-eating consis-
study periods, the average consumption of preferred fruits tently indicated higher food quality in the central neigh-
by female parties ranging in the southern or central grid bourhood than in the northern neighbourhood.
cells was approximately twice as frequent as in the
northern areas (Fig. 3). From 1996 to 2000, total fruit con-
sumption (including figs) in northern grid cells was com- Reproductive Consequences of Core Area
parable to that in the other neighbourhoods; however, Usage
from 2001 to 2004, the majority of the diet in northern
grid cells (72%) consisted of leaves and piths. Females in the central area had significantly higher
As expected, the distribution of episodes of non-fig fruit levels of urinary oestrogen and progesterone conjugates
feeding was a significant predictor of grid cell usage by all than did northern females after adjustment for reproduc-
neighbourhoods, although the strength of this relation- tive state (Student’s t test: E1C: t18 ¼ 2.195, P ¼ 0.04; PdG:
ship varied. In 1996e2000, frequency of non-fig fruit t18 ¼ 2.161, P ¼ 0.04; Fig. 4). Although some females were
consumption in each cell strongly predicted the range represented by relatively few urine samples, we found that
usage of southern females (R2 ¼ 0.590, N ¼ 153 grid cells, the variance in hormonal titers between females signifi-
P < 0.0001). The majority of the variance in central range cantly exceeded the variance in multiple samples from
use was also predicted by non-fig fruit consumption the same individual over the study period (ANOVA: E1C:
(R2 ¼ 0.539, N ¼ 153, P < 0.001), although the slope of P < 0.001; PdG: P < 0.03 for all reproductive states). En-
the relationship was significantly weaker than that for hanced ovarian function was accompanied by a large dif-
southern females (t302 ¼ 3.544, P ¼ 0.0005). Northern fe- ference in interbirth intervals. The mean  SE interbirth
males’ ranges showed only a very loose relationship with interval for the community at large was 39.8  7.3
this index of habitat quality (R2 ¼ 0.065, N ¼ 153, months (N ¼ 6) following the death of an infant, and
P ¼ 0.001), significantly weaker than that for ranges of 81.8  16.2 months (N ¼ 15 complete, 16 incomplete) if
southern (t302 ¼ 4.942, P < 0.0001) or central (t302 ¼ the previous infant survived to 4 years. The mean inter-
2.756, P ¼ 0.006) females. In 2001e2004, non-fig fruit birth interval for central females was 31.8  8.2 months
consumption was a significantly better predictor of central (N ¼ 4) following infant death and 70.3  6.5 months
female range (R2 ¼ 0.660, N ¼ 111, P < 0.0001) than it (N ¼ 10 complete, 10 incomplete) after infant survival,
was of northern female range (R2 ¼ 0.119, N ¼ 111, while the mean for the northern females was 55.9  0.0
P ¼ 0.0002; t218 ¼ 3.959, P ¼ 0.0001). (N ¼ 2) after infant death and 114.1  25.7 months

Table 3. Female neighbourhood quality assessed by feeding locations of three preferred fruit species
Percentage of feeding locations within neighbourhood range
Number of grid
Neighbourhood cells in neighbourhood Mimusops bagshawei Uvariopsis congensis Pseudospondias microcarpa

1996e2000
Northern 33 18 10 8
Central 31 39 42 42
Southern 34 46 58 100
2001e2004
Northern 34 37 16 0
Central 28 48 42 73

Neighbourhoods had a small degree of overlap and did not encompass the entirety of the chimpanzee range.
508 ANIMAL BEHAVIOUR, 73, 3

2 censored), while northern infants only survived for

Mean female z score across

Central N=14 a mean of 7.6  1.5 years (N ¼ 11, 6 censored).

1 Northern N=6
reproductive states

0
DISCUSSION
–1
Our results indicate that female chimpanzees in the
–2 Kanyawara community used the community range in
a differentiated manner, forming distinct northern, cen-
–3 P2=0.04 P2=0.04
tral and southern clusters. Females with core areas in the
–4 north had reduced access to preferred fruit trees, low
Urinary oestrogens Urinary progestins ovarian hormone levels and decreased reproductive suc-
Figure 4. Urinary oestrone conjugates (E1C) and pregnanediol-3- cess as indexed by interbirth intervals and infant
glucuronide (PdG) in central and northern female Kanyawara survivorship.
chimpanzees, standardized for reproductive state. N ¼ number of Lower habitat quality in the northern neighbourhood
females; P values from t test. Southern females were omitted because was reflected in general forest structure and locations of
of inadequate sampling. preferred chimpanzee foods. These habitat differences
may have been partly due to a greater intensity of logging
in the northern neighbourhood, but effects of soil type,
(N ¼ 3 complete, 5 incomplete) after infant survival. The
drainage and altitude may also be important. Differences
sample of birth intervals for northern females was small,
in tree density between the northern and central neigh-
particularly because so few of their infants survived. Nev-
bourhoods are also likely to account for differences in
ertheless, the difference between northern and central fe-
frugivore density (Struhsaker 1997; Chapman et al. 2000).
males in mean birth interval approached significance
The strong stability of female spatial relationships over
(following infant death: log rank ¼ 3.86, P ¼ 0.05; follow-
the 9-year span of this study conforms to Williams et al.’s
ing infant survival: log rank ¼ 3.26, P ¼ 0.07).
(2002) suggestion that female chimpanzees gain impor-
Infant survival statistics over the study period also
tant advantages from site fidelity, such as increased knowl-
showed profound differences (Fig. 5). With a limiting
edge of feeding locations. Northern females may also
age of 17 (due to study length), mean  SE infant survival
ameliorate the effects of lower food quality by adopting
was 11.7  1.2 years (N ¼ 41 infants, 29 censored (i.e. liv-
social strategies that reduce feeding competition. Ranging
ing)). Offspring of southern females had the highest sur-
patterns of individual northern females were more differ-
vival (15.4  1.4 years, N ¼ 7, 6 censored). Central
entiated from one another than those of either central
offspring survived for 11.1  1.6 years (N ¼ 23, 17
or southern females, suggesting less overlap in individual
core areas or less association between individuals. Indeed,
the northern females at Kanyawara are less gregarious
1.1 than the central females (Emery Thompson & Wrangham
2006).
Southern
1 Our limited data on the southern females suggest that
Southern-censored these females may have had the highest quality range and
the best reproductive outcomes. High-quality resources
0.9 in the south might also have made this region desirable to
the neighbouring chimpanzee community in the south,
Cumulative survival

however, raising the possibility that intercommunity

0.8
Central fighting contributed to the dissolution of the southern
Central-censored neighbourhood after 2001. The southern neighbourhood
0.7 at Kanyawara provides an important indication that
habitat quality may not be the only consideration in
determining female core area use. Female chimpanzees
0.6
seem to face a compromise among considerations of
habitat quality, scramble competition and threat from
Northern
0.5 neighbouring communities. The presence of only two
Northern-censored
females in the southern neighbourhood suggests that
other females may have avoided this border area despite
0.4 its food richness.
The southern females also provide an important con-
0.3 trast to northern females in terms of the consequences of
0 2 4 6 8 10 12 14 16 18 20 range location. Although each neighbourhood is situated
Offspring age (years) close to a border area, northern females suffered the
Figure 5. Proportion of infants expected to survive at ages 0e17 highest infant and juvenile mortality and the lowest
based on infant and juvenile mortality data for northern, central fecundity, whereas southern females appeared to fare
and southern females using KaplaneMeier survival analysis. the best. Unfortunately, we cannot identify the cause of
 EMERY THOMPSON ET AL.: RANGE AND REPRODUCTION OF CHIMPANZEES 509

mortality in most cases. However, low mortality rates in human contacts) may be proposed to influence female fe-
the southern neighbourhood suggest that, even though cundity via variation in social stress. Our results suggest
these females were living in a high-quality border zone that this is not the case, however. While central females
and did not typically enjoy the protection of community have the most interaction with aggressive males and the
males, intercommunity aggression was not a major threat most contact with human observers, they have higher fe-
to infant and juvenile survivorship. Females in both cundity than northern females.
southern and northern communities were less gregarious Differential range use could have effects on reproductive
than central females and spent less time in the presence success besides access to food resources. Females ranging
of observers (Emery Thompson & Wrangham 2006). in peripheral areas may be more susceptible to infanticides
Thus, direct feeding competition and anthropogenic ill- by neighbouring communities (Wilson & Wrangham
nesses are unlikely sources of mortality. We also have no 2003), although this did not appear to be the case for
evidence of predation on chimpanzees at Kanyawara. southern Kanyawara females. Increased maternal sociality
Our available evidence therefore suggests that the best ex- has been linked to infant survivorship in some primates
planation for differences in survivorship is differing ma- (Silk et al. 2003), so range dispersion could also affect re-
ternal and juvenile nutrition and resultant somatic productive success by affecting gregariousness.
condition. The degree of female reproductive variance reported
Assuming no difference in female life span or age at here suggests that there ought to be competition over
maturity among neighbourhoods and a mean reproduc- where female chimpanzees settle within the home range.
tive span of 16 years (Hill et al. 2001), our data for central Results from other field studies suggest that immigrant
mothers predict a mean of 3.7 offspring born; 2.2 of these adolescent females do encounter resistance from resident
offspring can be expected to survive to maturity. Northern females (Pusey 1980, 1990; Nishida 1989); we are cur-
mothers, by contrast, should have produced a mean of 2.7 rently investigating rates of aggression during immigration
offspring, of whom only 1.0 can be expected to survive to events at Kanyawara, as well as the rank relationships of
maturity. This evidence for an approximate doubling of re- northern and central females when they join parties
productive success in one cluster compared to another together. We hypothesize that female chimpanzees estab-
suggests that, in chimpanzees, there can be intense selec- lish their dominance relationships in the context of core
tion pressures for females to occupy high-quality core area establishment, and that the resultant ranging het-
areas. erogeneity reduces the need for frequent, potentially
The differences between neighbourhoods that we found costly contest competition. Thus, rather than indicating
in fruit availability, fecundity and infant survival support a lack of differentiated social status, the relative rarity
the role of energy availability in the reproductive success of overt dominance interactions observed among
of females. Such differences are consistent with the role of female chimpanzees may result from stable dominance
nutrition in mediating reproductive function in humans relationships.
(Ellison et al. 1989, 1993; Ellison 1995, 2003; Bentley et al. These results may not be generalizable to all chimpan-
1998) and nonhuman primates (van Schaik & van Noord- zee communities. Notably, female chimpanzees in the Ta€ı
wijk 1985; Bercovitch 1987; Bercovitch & Strum 1993; community in Cote d’Ivoire (Pan troglodytes verus) show
Strier et al. 2001; Altmann & Alberts 2003b, 2003a; Knott increased gregariousness and wider ranging habits than
2005). Conception timing and ovarian hormone levels in do Kanyawara chimpanzees (Lehmann & Boesch 2003,
Kanyawara chimpanzees have both been correlated to the 2005); they also show more clearly defined rank relation-
utilization of preferred non-fig fruits (Sherry 2002; Emery ships in the context of feeding competition (Boesch
Thompson 2005a), supporting a causal link between lower 1997; Wittig & Boesch 2003), with higher-ranking females
habitat quality and poorer reproductive variables in north- ranging further and making more use of peripheral areas
ern females. (Lehmann & Boesch 2005). It is unclear what is responsi-
There are possible alternative or intermediary explana- ble for the differences in social structure between Ta€ı and
tions for the observed relationships. One possibility is that East African sites (Kibale, Gombe). Similarly, chimpanzees
females with higher fitness attributes, such as larger body in Budongo Forest, Uganda, live in a relatively small range
size or better competitive abilities, succeed in settling in of high resource density (Newton-Fisher 2002), and like-
higher-quality core areas. However, it is difficult to wise appear to have very little differentiation of female
imagine this effect acting independently of subsequent ranges (Emery Thompson et al. 2006). Thus, although
effects of differing dietary quality, given that adult our results do not necessarily apply to chimpanzees in
chimpanzee body weights can show significant seasonal general, they suggest that reproductive skew among fe-
weight fluctuations (Pusey et al. 2005). Indeed, if we as- male chimpanzees may be largely determined by local eco-
sume that core area settlement is nonrandom, we have logical factors that remain to be characterized.
all the more reason to believe that there are important Few species can be expected to show spatially based
consequences of core area location. In Kanyawara, increas- reproductive skew within groups, because within social
ing age appears to be a strong predictor of dominance rank communities, females normally travel together. However,
among females (Kahlenberg 2006). While both of the analogous fissionefusion patterns occur outside of the
southern females at Kanyawara were past prime, there primate order and may be expected to be associated with
were no systematic differences in the ages of central and similar ecological constraints, as well as similar reproduc-
northern females. Differences in the socioecology of tive variance. Within-group female range heterogeneity
neighbourhoods (e.g. presence of males, party sizes, has been observed in one fissionefusion species, the
510 ANIMAL BEHAVIOUR, 73, 3

spotted hyaena, Crocuta crocuta. Low-ranking female hy- Bentley, G. R., Harrigan, A. M. & Ellison, P. T. 1998. Dietary com-
aenas at Masai Mara ranged closer to territorial bound- position and ovarian function among Lese horticulturalist women
aries, particularly when prey were scarce, and had lower of the Ituri Forest, Democratic Republic of Congo. European Jour-
reproductive success than did high-ranking females (Boy- nal of Clinical Nutrition, 52, 261e270.
dston et al. 2003). How much of the fitness variance was Bercovitch, F. B. 1987. Female weight and reproductive condition
in a population of olive baboons (Papio anubis). American Journal
due to variation in food intake was unclear, however, since
of Primatology, 12, 189e195.
subordinate hyaenas also experience severe aggression
Bercovitch, F. B. & Strum, S. C. 1993. Dominance rank, resource
and infanticide of their litters (Hofer & East 1995; Frank availability, and reproductive maturation in female savanna
1996; Muller & Wrangham 2002). To our knowledge, baboons. Behavioral Ecology and Sociobiology, 33, 313e318.
therefore, our results for Kanyawara chimpanzees are the Boesch, C. 1997. Evidence for dominant wild female chimpanzees
first example of intragroup variation in female reproduc- investing more in sons. Animal Behaviour, 54, 811e815.
tive success correlated with habitat variation and space Boydston, E. E., Kapheim, K. M., Szykman, M. & Holekamp, K. E.
use. Attention to these variables warrants future attention 2003. Individual variation in space use by female spotted hyenas.
in a range of species. Journal of Mammalogy, 84, 1006e1018.
Chapman, C. A. & Wrangham, R. W. 1993. Range use of the forest
chimpanzees of Kibale: implications for the understanding of
Acknowledgments chimpanzee social organization. American Journal of Primatology,
31, 263e273.
Research at Kanyawara was supported by funding from
Chapman, C. A., Balcomb, S. R., Gillespie, T. R., Skorupa, J. P. &
National Science Foundation grant 0416126 to R. Wrang-
Struhsaker, T. T. 2000. Long-term effects of logging on African
ham. Hormonal analyses were supported by grants to primate communities: a 28-year comparison from Kibale National
M.E.T. from Harvard University, the L.S.B. Leakey Foun- Park, Uganda. Conservation Biology, 14, 207e217.
dation, and the Wenner-Gren Foundation for Anthropo- Creel, S. F. 2005. Dominance, aggression, and glucocorticoid levels
logical Research. Local research support and permissions in social carnivores. Journal of Mammalogy, 86, 255e264.
were obtained from the Uganda National Council for Eissa, M. K., Obhrai, M. S., Docker, M. F., Lynch, S. S., Sawers,
Science and Technology, the Uganda Wildlife Authority R. S. & Newton, R. R. 1986. Follicular growth and endocrine
and the Makerere University Biological Field Station. Daily profiles in spontaneous and induced conception cycles. Fertility
data collection was done by Francis Mugurusi, Christo- and Sterility, 53, 81e87.
pher Muruuli, Peter Tuhairwe, John Barwogeza, Christo- Ellis, L. 1995. Dominance and reproductive success among nonhu-
pher Katongole and the late Donor Muhangyi, with field man animals: a cross-species comparison. Ethology and Sociobiol-
management by Michael Wilson, Martin Muller, Kather- ogy, 16, 257e333.
ine Pieta, Carole Hooven and Kimberly Duffy. We thank Ellison, P. T. 1995. Understanding natural variation in ovarian
function. In: Human Reproductive Decisions: Biological and Social
Meg Crofoot, Zarin Machanda, Martin Muller, Rachel
Perspectives (Ed. by R. I. M. Dunbar), pp. 22e51. London:
Carmody, Peter Ellison and four anonymous referees for
St. Martin’s Press/Galton Institute.
helpful comments on the manuscript.
Ellison, P. T. 2003. Energetics and reproductive effort. American
Journal of Human Biology, 15, 342e351.
References Ellison, P. T., Peacock, N. R. & Lager, C. 1989. Ecology and ovarian
function among Lese women of the Ituri Forest, Zaire. American
Abbott, D. H. 1984. Behavioral and physiological suppression of fer- Journal of Physical Anthropology, 78, 519e526.
tility in subordinate marmoset monkeys. American Journal of Prima- Ellison, P. T., Panter-Brick, C., Lipson, S. F. & O’Rourke, M. T.
tology, 6, 169e186. 1993. The ecological context of human ovarian function. Human
Abbott, D. H. 1987. Behaviourally mediated suppression of repro- Reproduction, 8, 2248e2258.
duction in female primates. Journal of Zoology, 213, 455e470. Emery Thompson, M. 2005a. Endocrinology and ecology of wild fe-
Akman, M., Erden, H., Bener, F., Liu, J. & Bahceci, M. 2002. Can male chimpanzee reproduction. Ph. D. thesis, Harvard University.
luteal phase estradiol levels predict the pregnancy outcome in in Emery Thompson, M. 2005b. Reproductive endocrinology of wild
vitro fertilization cycles of good responders whose excess embyros female chimpanzees (Pan troglodytes schweinfurthii): methodolog-
yield blastocysts? Fertility and Sterility, 77, 638e639. ical considerations and the role of hormones in sex and concep-
Altmann, J. & Alberts, S. C. 2003a. Intraspecific variability in fertility tion. American Journal of Primatology, 67, 137e158.
and offspring survival in a non-human primate: behavioral control Emery Thompson, M. & Wrangham, R. W. 2006. Comparison of sex
of ecological and social sources. In: Offspring: Human Fertility Be- differences in gregariousness in fissionefusion species: reducing
havior in a Biodemographic Perspective (Ed. by K. Wachter), pp. bias by standardizing for party size. In: Primates of Western Uganda
140e169. Washington, D.C.: National Academy Press. (Ed. by N. E. Newton-Fisher, H. Notman, V. Reynolds & J. Paterson),
Altmann, J. & Alberts, S. C. 2003b. Variability in reproductive suc- pp. 209e226. New York: Springer.
cess viewed from a life-history perspective in baboons. American Emery Thompson, M., Wrangham, R. W. & Reynolds, V. 2006.
Journal of Human Biology, 15, 401e409. Urinary estrone conjugates and reproductive parameters in
Barrett, J., Abbott, D. H. & George, L. M. 1990. Extension of repro- Kibale (Kanyawara) and Budongo (Sonso) chimpanzees. In: Pri-
ductive suppression by pheromonal cues in subordinate female mates of Western Uganda (Ed. by N. E. Newton-Fisher, H. Notman,
marmoset monkeys, Callithrix jacchus. Journal of Reproduction V. Reynolds & J. Paterson), pp. 227e246. New York: Springer.
and Fertility, 90, 411e418. Epple, G. & Katz, Y. 1984. Social influences on estrogen excretion
Barrett, J., Abbott, D. H. & George, L. M. 1993. Sensory cues and and ovarian cyclicity in saddle back tamarins (Saguinus fuscicollis).
the suppression of reproduction in subordinate female marmoset American Journal of Primatology, 6, 215e227.
monkeys, Callithrix jacchus. Journal of Reproduction and Fertility, Frank, L. G. 1996. Female masculinization in the spotted hyena: endo-
97, 301e310. crinology, behavioral ecology, and evolution. In: Carnivore Behavior,
 EMERY THOMPSON ET AL.: RANGE AND REPRODUCTION OF CHIMPANZEES 511

Ecology, and Evolution (Ed. by J. L. Gittleman), pp. 78e131. Ithaca, (Ed. by C. S. Harcourt & B. Sherwood), pp. 287e308. Westbury
New York: Cornell University Press. Academic & Scientific.
Gadgil, M. & Bossert, W. H. 1970. Life historical consequences of Nishida, T. 1968. The social group of wild chimpanzees in the Ma-
natural selection. American Naturalist, 104, 1e24. hale mountains. Primates, 19, 167e224.
Goodall, J. 1986. The Chimpanzees of Gombe: Patterns of Behavior. Nishida, T. 1989. Social interactions between resident and immi-
Cambridge, Massachusetts: Belknap Press. grant female chimpanzees. In: Understanding Chimpanzees (Ed.
Gouzoules, H., Gouzoules, S. & Fedigan, L. M. 1982. Behavioural by P. G. Heltne & L. A. Marquardt), pp. 68e89. Cambridge, Mas-
dominance and reproductive success in female Japanese monkeys sachusetts: Harvard University Press.
(Macaca fuscata). Animal Behaviour, 30, 1138e1150. Pepper, J. W., Mitani, J. C. & Watts, D. P. 1999. General gregari-
Harcourt, A. H. 1987. Dominance and fertility among female pri- ousness and specific social preferences among wild chimpanzees.
mates. Journal of Zoology, 213, 471e487. International Journal of Primatology, 20, 613e632.
Hasegawa, T. 1990. Sex differences in ranging patterns. In: The Chim- Pusey, A. E. 1980. Inbreeding avoidance in chimpanzees. Animal
panzees of the Mahale Mountains: Sexual and Life History Strategies Behaviour, 28, 543e552.
(Ed. by T. Nishida), pp. 99e114. Tokyo: University of Tokyo Press. Pusey, A. E. 1990. Behavioural changes at adolescence in chimpan-
Hill, K., Boesch, C., Goodall, J., Pusey, A. E., Williams, J. & Wrang- zees. Behaviour, 115, 203e245.
ham, R. W. 2001. Mortality rates among wild chimpanzees. Jour- Pusey, A. E., Williams, J. & Goodall, J. 1997. The influence of dom-
nal of Human Evolution, 40, 437e450. inance rank on the reproductive success of female chimpanzees.
Hofer, H. & East, M. 1995. Population dynamics, population size, Science, 277, 828e831.
and the commuting system of Serengeti spotted hyenas. In: Seren- Pusey, A. E., Oehlert, G. W., Williams, J. & Goodall, J. 2005. Influ-
geti II: Dynamics, Management, and Conservation of an Ecosystem ence of ecological and social factors on body mass of wild chim-
(Ed. by A. R. E. Sinclair & P. Arcese), pp. 332e365. Chicago: Uni- panzees. International Journal of Primatology, 26, 3e31.
versity of Chicago Press. Romesburg, H. C. 2004. Cluster Analysis for Researchers. Moorisville,
Kahlenberg, S. M. 2006. Femaleefemale competition and male sexual North Carolina: Lulu Press.
coercion in Kanyawara chimpanzees. Ph. D. thesis, Harvard University. Saltzman, W., Prudom, S. L., Schultz-Darken, N. J., Wittwer,
Knott, C. D. 1997. Field collection and preservation of urine in D. J. & Abbott, D. H. 2004. Social suppression of cortisol in fe-
orangutans and chimpanzees. Tropical Biodiversity, 4, 95e102. male marmoset monkeys: role of circulating ACTH levels and
Knott, C. D. 2001. Ape models of female reproductive ecology. In: glucocorticoid negative feedback. Psychoneuroendocrinology, 29,
Reproductive Ecology and Human Evolution (Ed. by P. T. Ellison), 141e161.
pp. 429e463. Chicago: Aldine. van Schaik, C. P. & van Noordwijk, M. A. 1985. Interannual
Knott, C. D. 2005. Energetic responses to food availability in the variability in fruit abundance and the reproductive seasonality in
great apes: implications for hominid evolution. In: Seasonality in Sumatran long-tailed macaques (Macaca fascicularis). Journal of
Primates: Studies of Living and Extinct Human and Non-human Pri- Zoology, 206, 533e549.
mates (Ed. by D. K. Brockman & C. P. van Schaik), pp. 351e Schneider, J. E. & Wade, G. N. 2000. Inhibition of reproduction in ser-
378. Cambridge: Cambridge University Press. vice of energy balance. In: Reproduction in Context: Social and Environ-
Koenig, A. 2002. Competition for resources and its behavioral con- mental Influences on Reproduction (Ed. by K. Wallen & J. E. Schneider),
sequences among female primates. International Journal of Prima- pp. 35e82. Cambridge, Massachusetts: MIT Press.
tology, 23, 759e783. Sherry, D. S. 2002. Reproductive seasonality in chimpanzees and
Kuederling, I., Evans, C. S., Abbott, D. H., Pryce, C. R. & Epple, G. humans: ultimate and proximate factors. Ph.D. thesis, Harvard
1995. Differential excretion of urinary oestrogen by breeding fe- University.
males and daughters in the red-bellied tamarin (Saguinus labiatus). Silk, J. B., Alberts, S. C. & Altmann, J. 2003. Social bonds of
Folia Primatologica, 64, 140e145. female baboons enhance infant survival. Science, 302,
Lehmann, J. & Boesch, C. 2003. Social influences on ranging pat- 1231e1234.
terns among chimpanzees (Pan trolodytes verus) in the Tai National Skorupa, J. P. 1988. The effects of selective timber harvesting on
Park, Cote d’Ivoire. Behavioral Ecology, 14, 642e649. rain-forest primates in Kibale Forest, Uganda. Ph.D. thesis, Univer-
Lehmann, J. & Boesch, C. 2005. Bisexually bonded ranging in chim- sity of California, Davis.
panzees (Pan troglodytes verus). Behavioral Ecology and Sociobiol- Sterck, E. H. M., Watts, D. P. & van Schaik, C. P. 1997. The evolu-
ogy, 57, 525e535. tion of female social relationships in nonhuman primates. Behav-
Lenton, E. A., Gelsthorp, C. H. & Harper, R. 1988. Measurement of ioral Ecology and Sociobiology, 41, 291e309.
progesterone in saliva: assessment of the normal fertile range using Strier, K. B., Mendes, S. L. & Santos, R. R. 2001. Timing of births in
spontaneous conception cycles. Clinical Endocrinology, 38, 637e646. sympatric brown howler monkeys (Alouatta fusca clamitans) and
Lipson, S. F. & Ellison, P. T. 1996. Comparison of salivary steroid northern muriquis (Brachyteles arachnoides hypoxanthus). American
profiles in naturally occurring conception and non-conception cy- Journal of Primatology, 55, 87e100.
cles. Human Reproduction, 11, 2090e2096. Struhsaker, T. T. 1997. Ecology of an African Rain Forest: Logging in
Matsumoto-Oda, A. 1999. Mahale chimpanzees: grouping patterns Kibale and the Conflict between Conservation and Exploitation. Gain-
and cycling females. American Journal of Primatology, 47, 197e esville: University Press of Florida.
207. Sugiyama, Y. 1968. Social organization of chimpanzees in the Bu-
Muller, M. N. & Wrangham, R. W. 2002. Sexual mimicry in hyenas. dongo Forest, Uganda. Primates, 9, 225e258.
Quarterly Review of Biology, 77, 3e16. Trivers, R. L. 1972. Parental investment and sexual selection. In: Sex-
Nadler, R. D. & Collins, D. C. 1991. Copulatory frequency, urinary ual Selection and the Descent of Man 1871e1971 (Ed. by B. Camp-
pregnanediol, and fertility in great apes. American Journal of Prima- bell), pp. 136e179. Chicago: Aldine.
tology, 24, 167e179. de Waal, F. B. M. 1982. Chimpanzee Politics: Power and Sex Among
Newton-Fisher, N. E. 2002. Ranging patterns of male chimpanzees Apes. Baltimore, Maryland: Johns Hopkins University Press.
in the Budongo Forest, Uganda: range structure and individual dif- Wasser, S. K. 1996. Reproductive control in wild baboons measured
ferences. In: New Perspectives on Primate Evolution and Behaviour by fecal steroids. Biology of Reproduction, 55, 393e399.
512 ANIMAL BEHAVIOUR, 73, 3

Williams, J. M., Pusey, A. E., Carlis, J. V., Farm, B. P. & Goodall, J. Wrangham, R. W. 1979. Sex differences in chimpanzee dispersion.
2002. Female competition and male territorial behaviour influence In: The Great Apes (Ed. by D. A. Hamburg & E. R. McCown), pp.
female chimpanzees’ ranging patterns. Animal Behaviour, 63, 481e489. San Franciso: Benjamin/Cummings.
347e360. Wrangham, R. W. 1980. An ecological model of female-bonded pri-
Williams, J., Oehlert, G. W., Carlis, J. V. & Pusey, A. E. 2004. Why mate groups. Behaviour, 75, 262e299.
do male chimpanzees defend a group range? Animal Behaviour, Wrangham, R. W., Chapman, C. A., Clark-Arcadi, A. P. & Isabirye-
68, 523e532. Basuta, G. 1996. Social ecology of Kanyawara chimpanzees: im-
Wilson, M. L. 2001. Imbalances of power: how chimpanzees re- plications for understanding the costs of great ape groups. In:
spond to the threat of intergroup aggression. Ph.D. thesis, Harvard Great Ape Societies (Ed. by T. Nishida), pp. 45e57. Cambridge:
University, Cambridge, Massachusetts. Cambridge University Press.
Wilson, M. L. & Wrangham, R. W. 2003. Intergroup relations in Yoshimura, Y. & Wallach, E. E. 1987. Studies of the mechanism(s)
chimpanzees. Annual Review of Anthropology, 32, 363e392. of mammalian ovulation. Fertility and Sterility, 47, 22e34.
Wittig, R. M. & Boesch, C. 2003. Food competition and linear Ziegler, T. E. & Sousa, M. B. C. 2002. Parentedaughter relation-
dominance hierarchy among female chimpanzees of the Tai ships and social controls on fertility in female common marmosets,
National Park. International Journal of Primatology, 24, 847e Callithrix jacchus. Hormones and Behavior, 42, 356e367.
867. Ziegler, T. E., Savage, A., Scheffler, G. & Snowdon, C. T. 1987.
Wrangham, R. W. 1975. The behavioural ecology of chimpanzees The endocrinology of puberty and reproductive functioning in fe-
in Gombe National Park, Tanzania. Ph.D. thesis, University of male cotton-top tamarins (Saguinus oedipus) under varying social
Cambridge. conditions. Biology of Reproduction, 37, 618e627.