Primates (2009) 50:293–303
DOI 10.1007/s10329-009-0163-0
REVIEW ARTICLE
Infanticide and social flexibility in the genus Gorilla
Juichi Yamagiwa Æ John Kahekwa Æ
Augustin Kanyunyi Basabose
Received: 31 July 2008 / Accepted: 28 July 2009 / Published online: 18 August 2009
Ó Japan Monkey Centre and Springer 2009
Abstract Based on the cases of infanticide by male
mountain gorillas reported from the Virunga volcanic
region, the socioecological and life history features of
gorillas satisfy the conditions for which infanticide may be
expected. However, there are considerable variations in the
occurrence of infanticide between habitats. We analyze the
recent reports of infanticides that were directly observed or
are suspected based on field evidence in two populations of
eastern and western lowland gorillas (Kahuzi and Mbeli
Bai, respectively) along with previous reports on mountain
gorillas, and consider which social features are linked with
and which factors influence the occurrence of infanticide in
the gorilla populations. All victims were suckling infants
and most of them were killed by males who seemed
unrelated to them. Dependent infants are most vulnerable
to infanticide when the protector male (its putative father in
most cases) is absent, and so male protection ability seems
to be important in determining female transfer decisions.
Two cases observed in Kahuzi suggest that the infanticidal
male may discriminate between infants to accept and those
to kill according to his previous interactions with their
mothers. Mating for a prolonged period prior to parturition
J. Yamagiwa (&)
Laboratory of Human Evolution Studies,
Graduate School of Science, Kyoto University, Sakyo,
Kyoto 606-8502, Japan
e-mail: yamagiwa@jinrui.zool.kyoto-u.ac.jp
J. Kahekwa
Parc National de Kahuzi-Biega, Institut Congolais pour
Conservation de la Nature, B.P. 895, Bukavu,
Democratic Republic of Congo
A. K. Basabose
Centre de Recherche en Sciences Naturelles, Lwiro,
D.S. Bukavu, Democratic Republic of Congo
Special contributions to commemorate the 60th
anniversary of Japanese primatology
is necessary for immigrant females to avoid infanticide by
the new male of the group that they join. Infanticide was
usually associated with female transfer, and the patterns of
female association at transfer may shape variations in
social structure between populations. Female mountain
gorillas prefer large groups with multiple males and tend to
transfer alone in order to seek more protection against
infanticide in Virunga. By contrast, female eastern and
western lowland gorillas tend to transfer with other females
to small groups or solitary males, and maturing silverbacks
take females to establish new groups through group fission
in Kahuzi and Mbeli Bai. These differences may result in
more multi-male and larger groups in the Virungas than in
Kahuzi and Mbeli Bai. Rapid changes in density of gorilla
social units and their relations following drastic environ-
mental changes caused by recent human disturbances may
also increase the probability of infanticide.
Keywords Gorilla
Infanticide

Male reproductive tactics
Multi-female transfer

Social structure
Environmental changes
Introduction
Infanticide by males is witnessed or strongly suspected in
39 primate species (van Schaik 2000a). In most cases, the
victims are suckling infants, and the killers are males who
unlikely to have sired such infants. The loss of the infant
usually leads to cessation of lactation in its mother, and
thereby stimulates her return to sexual receptivity
(Struhsaker and Leland 1985; Sommer 1994; Agoramoorthy
and Rudran 1995). Infanticidal males get mating access
to the victims’ mothers after killing and are most likely to
sire the next offspring (Sommer 1994; Borris et al. 1999;
123
294
Palombit et al. 2001). These observations support the
sexual selection hypothesis that infanticide has evolved as
a male reproductive tactic (Hrdy 1979; Sommer 1994; van
Schaik 2000a). Infanticide is more common in folivorous
species than frugivorous species (Janson and van Scaik
2000), especially in the single male group structure
(Palombit 2003). It also occurs most commonly in primate
species that have long lactation periods relative to gestation
(van Schaik 2000a, b). Either prolonged consort with
paternity concentration in the particular male or promis-
cuous mating with paternity confusion are regarded as
female tactics against infanticide (Hasegawa 1989; Watts
1989; Altmann 1990; Steenbeek 1999; van Noordwijk and
van Schaik 2000). In order to avoid infanticide, the mating
history—such as the relative frequency of mating and its
timing—probably plays a decisive role (Crockett and
Sekulic 1982; Pope 1990; van Schaik 2000a).
Socioecological and life history features of gorillas
satisfy the conditions for which infanticides may be
expected. Gorillas usually form polygynous groups with
one male mating system in most populations (Schaller
1963; Harcourt 1978; Harcourt et al. 1981; Parnell 2002;
Yamagiwa et al. 2003; Stokes et al. 2003). Females have a
longer period (three years) of lactational anestrus than the
gestation period (8.5 months) (Stewart 1988). Females get
estrus for a few days at the time of ovulation during each
monthly sexual cycle and mate exclusively with the dom-
inant group male (Harcourt et al. 1981; Fossey 1982; Watts
1990). Unlike most group-living primate species with
female philopatry, female gorillas disperse from natal
groups and transfer between groups (Harcourt 1978;
Yamagiwa and Kahekwa 2001; Stokes et al. 2003). Unlike
other primate species in which infanticide usually occurs
after group takeover, extragroup gorilla males do not take
over breeding groups and oust the resident male (Watts
1989; Harcourt and Stewart 2007). Due to female dispersal,
infanticidal males may not always get access to victims’
mothers when they resume their ovulation cycles. Different
options for females to counteract infanticide, such as
remaining in the group or dispersing to search for a suitable
group to join, may shape the sizes and compositions of
gorilla social units (Watts 1996; Yamagiwa 1999).
So far, most episodes of infanticide (13 cases) have been
reported from the Virunga volcanic region, where long-
term research has been conducted on mountain gorillas
(Gorilla beringei beringei) since 1967 (Fossey 1984; Watts
1989). Although more than two groups of eastern lowland
gorillas (Gorilla beringei graueri) have been habituated
and monitored for 30 years at Kahuzi, three cases of
infanticide have been reported only recently (Yamagiwa
and Kahekwa 2001, 2004). Among western lowland
gorillas (Gorilla gorilla gorilla) that have been studied for
more than ten years at five or more sites, only two cases of
123
Primates (2009) 50:293–303
infanticide have been suspected after the reported death of
a leading silverback (Stokes et al. 2003). It is supposed that
there are considerable variations in the occurrence of
infanticide between gorilla populations. Nunn and van
Schaik (2000) suggested that social factors, including
infanticide, rather than ecology (food and predation), better
explain intraspecific variations in the nature of primate
societies. But which intraspecific social variations are
linked with the occurrence of infanticide? And which
factors influence the occurrence of infanticide in gorilla
populations?
In spite of large differences in dietary composition and
daily path length between habitats, no significant differ-
ences have been found in average group sizes of gorillas
(Doran and McNeilage 2001; Parnell 2002; Yamagiwa
et al. 2003). Distinct differences were found in the maxi-
mum group size and the proportion of multi-male groups in
the population. The maximum group size of western and
eastern gorillas inhabiting lowland tropical forests is
smaller (less than 30 individuals) than those (more than 40
individuals) of eastern gorillas in montane forests (Yam-
agiwa et al. 2003). About half of the groups include more
than two silverbacks in the two populations (Bwindi and
the Virungas) of mountain gorillas, while multi-male
groups accounted for less than 10% of the other popula-
tions (Yamagiwa et al. 1993; Parnell 2002; McNeilage
et al. 2001; Kalpers et al. 2003). The frugivorous diets of
gorillas in lowland tropical forests may result in the lower
upper limit on group size compared to that of folivorous
gorillas in montane forests, and infanticide is regarded as a
causational factor that promotes male philopatry through
the selection of multi-male groups by females seeking
more protection (Watts 1996, 2000; Yamagiwa et al. 2003;
Robbins 1995; Robbins and Robbins 2005). However, it is
still unknown whether environmental factors influence the
occurrence of infanticide, and how infanticide causes
changes in individual behavior and leads to social changes.
Yamagiwa and Kahekwa (2001) reported that female
eastern lowland gorillas frequently transferred with other
females and immatures during the period in which infan-
ticide was absent from Kahuzi. By contrast, female
mountain gorillas in the Virungas usually transfer alone,
probably to avoid infanticide and to get reliable protection
from the leading silverback in the new group that they join
(Watts 1996). Females traveled as all-female groups for a
prolonged period after the death of the leading silverback at
Kahuzi, while females dispersed to join other groups in the
same situation in the Virungas (Watts 1989; Yamagiwa and
Kahekwa 2001). These observations suggest that the
occurrence of infanticide may influence the association
patterns of females, which may, in turn, influence the
male’s decision to disperse. We have recently found three
cases of infanticide in Kahuzi (Yamagiwa and Kahekwa
Primates (2009) 50:293–303 295

2004). These incidents occurred a few years after large-
scale hunting of gorillas during the civil war. A drastic
change in population may constitute a causational factor
for infanticide in Kahuzi. The occurrence of infanticide
may have resulted in the decisions of other females to
transfer. We will add these new cases to the previous
reports of infanticide in the Virungas and elucidate the
general features of infanticide in order to reconsider its
effects on social changes. We will also discuss the envi-
ronmental factors that influence the occurrence of infanti-
cide in these gorilla populations.
Source of data and background history
Infanticide has been directly observed or strongly sus-
pected following inspection of the victim’s wounds in the
two populations of Gorilla beringei, those in the Virungas
(G. b. beringei) and Kahuzi (G. b. graueri). Both habitats
are located on the Albertine lift and are covered with
montane forests (Fig. 1). Kahuzi gorillas have a more
frugivorous diet and have longer daily path lengths than the
more folivorous mountain gorillas (Yamagiwa et al. 2003),
while both populations have similar annual range sizes and
show no territoriality between neighboring groups (Schal-
ler 1963; Watts 1998; Yamagiwa et al. 1996; Yamagiwa
and Basabose 2006). The population size fluctuated
between 242 and 380 gorillas in the Virungas and between
130 and 258 gorillas in Kahuzi for these three decades
(Kalpers et al. 2003; Inogwabini et al. 2000). During these
periods, several groups of gorillas have been habituated
and monitored regularly in both sites (Fossey 1983; Stewart
et al. 2001; Yamagiwa and Kahekwa 2001).
From the nine cases of infanticide reported by Fossey
(1984), Watts (1989) selected six cases with reliable evi-
dence and reported six other cases that had been newly
observed in the Virungas. We added three cases that were
recently witnessed in Kahuzi to those in order to elucidate
the general features of and intraspecific variations in
infanticide by gorillas (Table 1). Acceptable evidence
means that (1) the incident was witnessed and confirmed
with the death of a victim, and that (2) a dead infant was
found after a known encounter between social units
without any sign of carnivore activity (Watts 1989). An
attempted case of infanticide and another suspected case
reported by Watts (1989) were excluded from our analysis.

Fig. 1 Location of the Kahuzi-

Biega National Park Kisangani

Maiko N.P. Virunga National Park

Maiko R. 0

Mountain gorilla
Edward L. Gorilla beringei beringei
Oso R.

Bwindi
Luka R.
N.P.

Lowland
Highland Virunga N.P.

Ulindi R. Kivu L. 2

RWANDA
Bukavu
Kahuzi-Biega
N.P.
Elila R.

BURUNDI
Tanganyika L.

Itombwe Reserve 4

Luoma R.

Kahuzi-Biega National Park

DEMOCRATIC
REPUBLIC OF Eastern lowland gorilla
CONGO
Gorilla beringei graueri
6

0 100 200 Km

26 28 30

123
296 Primates (2009) 50:293–303

Table 1 Infanticide incidents observed in the Virungas and Kahuzi

EG encounter with another group, ES encounter with a solitary male, WI incident within group, SB silverback, BB blackback, GS silverback in
mother’s group, GB blackback in mother’s group, EGS extra-group silverback, SS solitary silverback, BG gave birth in the present group, JG
joined the present group with the victim, JNG joined the present group without the victim, SV separated from the victim, RG remained in the
group, DG dispersed from the group, R reproduced an infant with the infanticidal male, M mated with the infanticidal male, J joined the
infanticidal male, S separated from the infanticidal male
a
The victim’s mother remained and reproduced in the group, which included two silverbacks, and the infanticidal male was unknown
b
The victim’s mother joined the present group a few months before parturition

Table 2 Population census in the highland sector of KBNP then three new groups have appeared in the home ranges of
the previously habituated groups. Several females and
1978a 1990b 1996c 2000d 2004d
immatures from the previously habituated groups were
Population size 223 258 247 130 163 seen to associate with the new groups. The park started to
Number of groups 14 25 25 13 15 habituate these three new groups for gorilla tourism. In
Mean group size 15.6 10.8 9.8 9.6 17.3 2002, a solitary male acquired two females to form his new
% Infants 17.0 8.4 12.7 9.3 15.9 group and came to visit these ranges. The five groups
Number of solitary males 5 9 2 4 2 frequently encountered each other, and the first infanticide
occurred when females transferred to this new group in
Sources: a Murnyak (1981), b Yamagiwa et al. (1993); c
Inogwabini
et al. (2000), d Amsini et al. (2008)
2003.
Infanticide was also suspected in a population of fru-
givorous western lowland gorillas at Mbeli Bai in the
Nouabalé-Ndoki National Park, Republic of Congo.
The population of gorillas in Kahuzi was stable until the
Demographic changes in a total of 20 groups were moni-
mid-1990s (Table 2; Murnyak 1981; Yamagiwa et al.
tored when they appeared in a large swampy clearing
1993; Inogwabini et al. 2000). However, after the Rwandan
between February 1995 and July 2001 (Stokes et al. 2003).
genocide in 1994, the influx of 450,000 refugees into the
Two cases of infant loss and one case of a dead infant were
Kahuzi region caused large environmental stresses (Hall
observed after their mothers transferred to new groups.
1994). Two large camps for refugees were constructed by
Although infanticide was the probable cause of each case,
UNHCR on the western border of the Park, and at least
these cases are not listed in Table 1 due to inconclusive
three large-scale forest fires occurred in 1996–1997 (Sato
evidence.
et al. 2000). The outbreaks of two civil wars in 1996 and
1998–1999 had also negative effects on the survival of
gorillas. The disarmament of the park guards and the influx
of refugees and militias into the park have drastically Infanticides observed in wild gorilla populations
increased the destruction of their habitats and poaching
(Yamagiwa 2003). By the 2000 census, about half of the In both the Virungas and Kahuzi, all victims were infants
population in the highland sector had disappeared; they that were less than three years old, and most of them were
were probably killed for bush meat (Table 2). still suckling (Table 1). No disparity in the sexes of the
During the second civil war in 1998–1999, large-scale victims was observed (six males and six females). Infan-
poaching occurred. All leading silverbacks in the five ticide occurred in 12 cases when the victims’ groups
habituated groups were killed by poachers, and four groups encountered other groups or solitary males. The killers
disappeared from their previous home ranges (Yamagiwa were mature males who seemed unrelated to the victims in
2003). A blackback male from one of these groups was most of the cases. In the Virungas, nine of the 12 cases
seen traveling with six females after the poaching. Since occurred just after the death of the leading silverback in the

123
Primates (2009) 50:293–303
victim’s group. In these cases, the victims were attacked by
infanticidal males when the leading males were absent, or
when their mothers had joined the other groups along with
the victims. Although the victims’ mothers usually tried to
defend them, and in case 6 maturing males (blackback
males) probably joined them to attack the infanticidal
males, they were not successful in saving the victims
(Watts 1989). In two cases witnessed in Kahuzi, the vic-
tim’s mother and the other females chased the infanticidal
male and the mother bit his foot but could not stop the
killing (Yamagiwa and Kahekwa 1984). After the killing,
the victim’s mother emigrated from the group in which the
leading male had died and joined the infanticidal male (in
cases 4 and 7) or separated from him (in cases 5 and 9).
When the victims were killed after their mother had joined
the new group, the mothers usually remained with the
infanticidal males and reproduced their next infants with
them. The infanticidal males succeeded in siring offspring
with the victims’ mothers in at least seven (and probably
ten) out of the 15 cases.
Infanticide was also suspected for western lowland
gorillas (Gorilla gorilla gorilla) after the death of the
leading male. In four cases of mothers with dependent
infants transferring between groups that were observed
following the death of the silverback male at Mbeli Bai in
the Nouabale-Ndoki National Park, two mothers lost their
infants (5 and 28 months old, respectively) in the new
group, presumably due to infanticide (Stokes et al. 2003).
However, in the other two cases, the dependent infants (31
and 56 months) survived in the new groups. Stokes et al.
(2003) reported that another two mother/infant pairs
transferred to known groups and that the infants (estimated
to be 10 and 4 months old, respectively) survived. These
observations are consistent with those reported for eastern
lowland gorillas in Kahuzi (Yamagiwa and Kahekwa
2001), while they contrast with mountain gorillas in the
Virungas, where all mother/infant pairs faced infanticide
after transferring to the new group (Watts 1989).
The notable feature of the infanticides observed in
Kahuzi is that infants born after their mothers’ immigration
were killed. One month (in case 14) and two months (in
case 15) after their mothers’ immigration, these victims
were born and killed by the leading silverback in a few
days (Yamagiwa and Kahekwa 2004). This group was
visited by trackers every day for tourism, but no copulation
was witnessed between the silverback and these mothers
prior to infanticide. In the same month as the second case,
another female transferred to this group 12 months before
giving birth, but her baby survived without any wounds. In
the Virungas, after the death of the leading silverback, a
maturing male (blackback) killed an infant of a mother
with whom he had never been seen to copulate (in case 5),
but did not kill another infant of a mother with whom he
297
frequently copulated (Watts 1989). A similar case was
reported for western lowland gorillas in Mbeli Bai, where
the shredded remains of a newborn infant were found on
the back of a female two months after immigration (Stokes
et al. 2003). This female did not have an infant prior to
transfer and was assumed to have conceived the baby with
the previous group’s silverback. These observations sug-
gest that the infanticidal male gorillas may have discrimi-
nated the infants they accepted from those they killed,
probably based on past interactions with their mothers.
The mating history with the female may also play a
decisive role in such discrimination in other primate spe-
cies (van Schaik 2000a). Mating during pregnancy after
takeover were effective at preventing infanticidal attacks
up to about 80 days before parturition in red colobus
(Struhsaker and Leland 1985), about 60 days in Hanuman
langurs (Sommer 1994), about 45 days in blue monkeys
(Fairgrieve 1995), and about 40 days in sooty mangabeys
(Busse and Gordon 1983). Hasegawa (1989) also reported
that mating close to parturition after transfer failed to stop
infanticide in chimpanzees at Mahale. Female gorillas
copulate throughout pregnancy and usually give birth more
than one year after transfer (Harcourt et al. 1980; Watts
1989; Yamagiwa and Kahekwa 2001; pers. obs.). The
survival of an infant that was born about one year after its
mother’s transfer in Kahuzi suggests that a prolonged
period between the first mating and birth in association
with the same male is necessary to prevent infanticide.
These observations may support the sexual selection
hypothesis for the evolution of infanticide as a male tactic
to increase the probability of mating with currently lac-
tating and infertile females in gorillas. Dependent infants
are most vulnerable to infanticide when the protector male
(its putative father in most cases) is absent, and so male
protection ability seems to be important in determining
female transfer decisions (Watts 1989, 1996; Stokes 2004;
Harcourt and Stewart 2007). Mating for a prolonged period
prior to parturition is necessary for immigrant females to
avoid infanticide by the new male of the group to which
they transfer.
Effects of infanticide on female transfer and population
structure
In most of the cases observed in the three gorilla popula-
tions (the Virungas, Kahuzi and Mbeli Bai), infanticide
was associated with female transfer. In the Virungas,
females left the group in which their infants were killed by
the extragroup males in the presence of the leading male, or
their infants were killed in their new groups after they had
transferred following the deaths of the leading males
(Watts 1989). The latter was also suspected in Mbeli Bai
123
298
(Stokes et al. 2003). In Kahuzi, infants were killed when
the mothers transferred to new groups, or at the time of
birth within a few months after their mothers’ transfers
(Yamagiwa and Kahekwa 2004). When infanticide occur-
red in the presence of the leading silverback, the victim’s
mother left the silverback who failed to save her infant in
both the Virungas and Kahuzi. In most cases, the victim’s
mother joined the infanticidal male and reproduced another
infant with him.
However, there are considerable variations in the
occurrence of infanticide between gorilla populations.
Infanticide occurred in the Virungas more frequently than
in both Kahuzi and Mbeli Bai. The patterns of female
association at transfer and their choice of social unit to
transfer may influence these variations. Unlike female
mountain gorillas that usually transfer alone in the Virun-
gas (Harcourt 1978; Watts 1992, 1996), female eastern and
western lowland gorillas tend to transfer with other females
in Kahuzi and Mbeli Bai, even when the leading male is
alive (Yamagiwa and Kahekwa 2001; Stokes et al. 2003).
Group fissions also occurred in the latter populations, and
maturing silverbacks took females with immatures to form
new groups (Yamagiwa and Kahekwa 2001; Stokes 2004).
By contrast, solitary males obtained opportunities to mate
later and gained lower reproductive success than maturing
males who remained to reproduce in their natal groups in
the Virungas (Robbins 1999; Watts 2000; Robbins and
Robbins 2005). In Mbeli Bai, females tend to prefer
smaller groups to join at transfer (Stokes et al. 2003). The
tendencies for multi-female transfer and group fission in
Kahuzi and Mbeli Bai may have enabled male gorillas to
attain reproductive success relatively quickly, and may
have prevented them from using infanticide for the forcible
solicitation of mating.
The occurrence of infanticide may also change patterns
of female association. In Kahuzi, a female left a dependent
infant (31 months old) and transferred to a group in which
infanticide had occurred one month before (Yamagiwa and
Kahekwa 2004). She probably witnessed the first infanti-
cide when the victim’s mother transferred from her group
to the infanticidal male’s group. She transferred to the
same group three months later, after second and the third
infanticides, which she may have noticed due to the loud
noises generated by the incidents. This is the first case in
which a female left a dependent infant at the time of
transfer in Kahuzi. It seems possible that this female may
have learned from the occurrence of infanticide and
effectively responded to the potential infanticide by
changing the pattern of association at transfer. This inci-
dent suggests that the occurrence of infanticide has far-
reaching effects within a population. In Kahuzi, female
transfer with a dependent infant or the formation of an all-
female group have never been seen after the occurrence of
123
Primates (2009) 50:293–303
infanticide, although both phenomena have occasionally
been witnessed before (Yamagiwa and Kahekwa 2001).
In the Virungas, it is unknown when the rate of infan-
ticide increased in the population of mountain gorillas.
However, the occurrence of infanticide has changed
between periods. It occurred frequently in the 1960s, 1970s
and 1980s, but was not seen from late 1980s until recently.
Recent social changes may have reduced the probability of
infanticide. Since the 1980s, the mean group size and the
proportion of multi-male groups have increased in the
Virungas (Robbins 2001; Kalpers et al. 2003). A large
proportion of multi-male groups is also found in another
mountain gorilla population in Bwindi Forest, where
infanticide has never been reported (McNeilage et al.
2001). Groups with multiple males are larger than groups
with a single male in both populations (McNeilage et al.
2001; Kalpers et al. 2003). These trends may have been
caused by females choosing multi-male groups in order to
counteract infanticide, and by male philopatry facilitated
by increasing group size and the number of females per
group. A multi-male group composition may be effective at
protecting suckling infants against infanticide by outside
males (Watts 2000; Robbins 2003; Robbins and Robbins
2005). Females do not have to leave their groups after the
death of the leading silverback, and their infants are
effectively protected by other males within their groups
(Watts 2000; Robbins and Robbins 2005). Established
groups with aged silverbacks and their sons may not
actively seek females, and encounters between such groups
seem to be peaceful (Schaller 1963; Harcourt 1978;
Yamagiwa 1987a).
A recent genetic analysis of paternity in four multi-male
groups in the Virungas indicated that both dominant and
subordinate males enjoyed reproductive success, with the
dominant male siring an average of 85% of the group
offspring (Bradley et al. 2005). Nsubuga et al. (2008)
reported an influence of paternity on associations between
immature and adult males upon group fission of mountain
gorillas in Bwindi. Based on a DNA analysis of paternity,
they found that the eight offspring assigned to the dominant
silverback (and their mothers) remained with their father,
while the two offspring of unknown paternity ended up in
the small group headed by the formerly subordinate
silverback. The low probability of infanticide in Bwindi is
possibly caused by females and immatures selectively
associating with adult males according to their paternal
relationships. The average lifetime reproductive success for
philopatric males who adopt a strategy of queuing to
become the dominant male is twice as high as that for
emigrants who seek mates as solitaries (Robbins and
Robbins 2005). Not only father–son pairs but also siblings
and half-siblings coexist in multi-male groups in both the
Virungas and Bwindi (Bradley et al. 2005; Nsubuga et al.
Primates (2009) 50:293–303
2008). Unrelated male mountain gorillas occasionally form
all-male groups and some of them continue to coexist after
accepting fertile females (Yamagiwa 1987b; Robbins 1995,
2001). These social features may have been shaped by
female strategies directed against infanticide in recent
times. In gorillas, the reproductive strategies of both males
and females interact with each other to produce intraspe-
cific social variations, rather than ecological factors. The
occurrence of infanticide may activate such interactions in
reproductive strategies between sexes.
Environmental changes influencing the occurrence
of infanticide
Large variations in the occurrence of infanticide between
gorilla populations suggest that some environmental fac-
tors may increase the probability of infanticide. Recent
reductions in population and habitat may constitute cau-
sational factors of infanticide in both Virunga and Kahuzi
gorilla populations. In the 1960s, prior to the first obser-
vation of infanticide in the Virungas, the Rwandan Gov-
ernment converted 40% of the park into farm land.
Cultivation and cattle encroachment forced the gorillas to
shift their ranges into the central part of the park (Fossey
1983). Dian Fossey started the habituation of gorillas in
1967 and found eight cases of infanticide in the 1970s
(Fossey 1984). The population censuses conducted in the
1970s and early 1980s indicated a reduction in the gorilla
population by half. The reduced ranges and increased
encounters between unfamiliar groups induced by such
Fig. 2 History of habituated
1972
and semihabituated groups (Gr)
of eastern lowland gorillas in 1975
the Kahuzi-Biega National Park.
Each arrow indicates multi-
female transfers between 1980
groups. Group fission is the
separation of two silverbacks
1985
with females
1990
1995
2000
2005
Mufanzala
Gr
299
unnatural disturbances may have produced conditions
leading to infanticide.
In Kahuzi, the habituation of gorillas started in the early
1970s for the purpose of tourism, but no infanticide had
been reported until 2003. During this period of with no
infanticide, females frequently transferred between groups
with other females and immatures, including dependent
infants, who survived in the new groups (Yamagiwa and
Kahekwa 2001). After the death of the leading silverback
male, the females did not disperse but traveled in associ-
ations without mature males for months or years. These
patterns of female association may correspond to the
absence of infanticide in the Kahuzi population. Which
environmental factors caused infanticide in 2003 after a
long absence? The large-scale poaching of gorillas induced
by the civil war in 1998 and 1999 may have produced
conditions favoring infanticide. Thousands of refugees and
soldiers roamed the forests of Kahuzi and hunted large
mammals for bushmeat (Yamagiwa 2003). All leading
silverbacks in the five habituated groups were killed by
poachers, and group disintegrations and female transfers
occurred after their deaths, although females continued to
associate among themselves without silverback males for
several months in some groups (Fig. 2). By contrast, young
silverbacks who had been solitary males survived during
the war and joined these female groups as the new leading
male. Among eight groups newly habituated by the park
and us after large-scale poaching, seven groups were led by
young silverbacks that were estimated to be less than
20 years old from their morphological features (body size,
shape of face, and sagittal crest). These maturing
Maeshe Gr
Mushamuka Gr
Multi-male group
One-male group
All-female group
Death of leading SB
Mubalala Gr S Solitary male
Group fission
Group fission Ganyamurume Gr
Nindja Gr
Young SB Mishebere Gr
S
S
Mugaruka
Gr
Chimanuka Gr
Birindwa Gr
S
123
300 Primates (2009) 50:293–303

% females with infant # female/group

90 Large-scale 14
80 hunting
12
70
s s 10
s 60
50 8
s
s 40 6
s Unhabituated 30
4
Habituated s s
20
2
s solitary 10
s 0 0
s
N 83 85 90 95 00 05
s
s
s Fig. 4 Annual changes in the number of females per habituated
0 10 km
1990 2000 group (black circle) and the proportion of infants per female (column)
in Kahuzi

Fig. 3 The location of each gorilla group and solitary males during 1983-1997 1998-2004
the census in 1990 (left) and that in 2000 (right) % infant/female % infant/female
100 100
silverbacks are strongly motivated to attract females in 80 80
order to establish their own groups (Harcourt 1978; Yam- 60 60
40 40
agiwa 1986), and they show a greater potential to be an
20 20
infanticidal male than the group males in the Virungas
0 0
(Watts 1989). The infanticidal male was one of these sol- 0 2 4 6 8 10 12 0 2 4 6 8 10 12
itary males during the civil wars in Kahuzi (Yamagiwa and # female/group # female/group
Kahekwa 2004). The increased number of young silver- R2=0.087, p=0.286 R2=0.628, p<0.05
backs relative to the number of mature silverbacks and the
Fig. 5 Correlation between the proportion of suckling infants per
increased competition among them possibly constituted female and the number of females per group in the habituated groups
causational factors of infanticide in Kahuzi. of gorillas in Kahuzi before (1983–1997) and after (1998–2004)
During the civil war, gorillas tended to shift their ranges large-scale hunting. Each dot indicates the annual score
into the central sector of the park, where the park staff
frequently patrolled (Fig. 3). This situation may have
stimulated unfamiliar groups to overlap their ranges and census (1990) also suggest that group fissions and new
thus frequently encounter each other. The infanticidal male group formation may have occurred widely in the Kahuzi
and another silverback who lost females to him were born population during this period. No correlation was found
in different groups (in 1986 and 1987, respectively) and between the number of females per group and the pro-
were unfamiliar with each other until their encounter in portion of females with infants between 1983 and 1997
2003. The increased poaching may have increased (Fig. 5, R2 = 0.087, NS). However, after the large-scale
encounters between unfamiliar groups and stimulated poaching during the second civil war in 1998–1999, the
young males to use aggressive strategies for reproduction, two parameters (the number of females per group, and the
such as infanticide. proportion of females with infants) markedly decreased,
The number of females per group and the proportion of but has increased again since 2004. A significant positive
females with infants fluctuated a great deal in the habitu- correlation was found between these two parameters for
ated groups during the study period (Fig. 4). In the 1980s, 1998 and 2004 (R2 = 0.628, p \ 0.05). This suggests that
most of the adult females had infants. However, the pro- the marked reduction in the number of groups led to fre-
portion of females with infants drastically decreased in the quent female transfer and reduced both the fecundity and
1990s, probably because young silverbacks established the number of females per group. Between 2000 and 2004,
their groups through group fission (Mubalala Group and at least two solitary males established new groups, and
Nindja Group) and accepted many females from neigh- females frequently moved between groups (Table 2,
boring groups (Fig. 2). The prominent increase in the Fig. 2). These situations may have increased competition
number of groups and reduction in the mean group size between males and the probability of infanticide in the
between the first population census (1978) and the second Kahuzi population.

123
Primates (2009) 50:293–303
Infanticide and gorilla population structure
We still do not know what the appropriate population
structure or interunit relationship is for gorillas. Gorillas
form stable and cohesive groups with no territoriality
(Schaller 1963; Yamagiwa 1983; Tutin 1996). Both females
and males leave their natal groups and only females transfer
to other groups or to solitary males (Harcourt 1978; Yam-
agiwa and Kahekwa 2001; Stokes et al. 2003). These social
features are common in all habitats and species/subspecies.
Small variations in gorilla population density and home-
range size irrespective of large dietary differences between
habitats suggest that social factors may have a greater
influence on their population structure and stability than
ecological factors (Yamagiwa 1999). In the population of
Virunga mountain gorillas, male mating tactics influence
female residence choices, which in part reflect their ranging
strategies, and which in turn influence male dispersal
decisions (Harcourt and Stewart 2007). However, infanti-
cide may not promote male philopatry in the population of
Kahuzi gorillas. Although large groups including more than
40 individuals were observed in Kahuzi, they did not
include multiple silverbacks (Yamagiwa 1983; Yamagiwa
et al. 2003). The multi-male group composition appeared
for only a short period when males grew up at silverback
age and coexisted with their putative fathers temporarily in
their natal groups (Fig. 2). Most groups of western lowland
gorillas include only one adult male in Mbeli Bai (Parnell
2002; Stokes et al. 2003). Different demographic factors
and female choices from those of Virunga mountain gorillas
may prevent them from shifting to a multi-male group
structure. Patterns of female association during transfer or
after the death of the leading silverback may constitute
these factors.
Dispersal patterns of male and female gorillas may play
a decisive role in population structure. Recent genetic
analyses using DNA markers reveal that females, rather
than males, are restricted to small areas in the Congo
(Melanie et al. 2007). Patterns of individual dispersal may
vary according to habitat conditions. Ecological factors
also influence group size and composition, and thus within-
and between-group interactions. For frugivorous western
gorillas, within-group feeding competition sets an upper
limit on group size and stimulates larger groups to travel
farther than smaller groups (Dunbar 1988; Janson and
Goldsmith 1995; Goldsmith 1999). The lack of multi-male
groups in the population of western lowland gorillas at
Mbeli Bai is explained by the greater competitive costs of
large group size, which outweigh the potential reproductive
advantages of a large multi-male group (Stokes et al.
2003). The frugivorous diet of western gorillas and its
different ranging strategy from that of folivorous eastern
gorillas may result in different behavioral features that
301
influence the male’s mating tactics. Harcourt and Green-
berg (2001) examined the probability of infanticide when
female gorillas travel alone and when they associate per-
manently with males, using a gas molecule equation with
density of males, speed of travel, and duration of fertile and
nursing periods, and concluded that infanticide may occur
at a threefold higher rate in solitary females. This also
implies that the risk of infanticide may fluctuate if these
parameters change significantly, influencing female
movements. Our study shows that rapid changes in the
density of gorilla social units and their relations following
drastic environmental changes may increase the probability
of infanticide, and gorillas have the ability to respond to
such changes with high behavioral and social flexibility. In
order to determine healthy gorilla population structures for
different types of habitat and to implement effective con-
servation measures, we urgently need more information on
the ecological and social features of gorillas in these dif-
ferent habitats.
Acknowledgments This study was financed in part by Grants-in-
Aid for Scientific Research from the Ministry of Education, Culture,
Sports, Science, and Technology, Japan (No. 162550080 and No.
19107007 to J. Yamagiwa), the Global Environmental Research Fund
from the Japanese Ministry of the Environment (F-061 to T. Nishida,
Japan Monkey Centre), and the Kyoto University Global COE Pro-
gram ‘‘Formation of a Strategic Base for Biodiversity and Evolu-
tionary Research.’’ It was conducted in cooperation with the Centre de
Recherches en Sciences Naturelles, Democratic Republic of Congo.
We thank the Institut Congolais pour Conservation de la Nature of the
Congolese government for supporting our research project. We are
also greatly indebted to all field assistants in Kahuzi-Biega National
Park and the people living in the villages next to the parks for their
kind support and hospitality throughout the fieldwork.
References
Agoramoorthy G, Rudran R (1995) Infanticide by adult and subadult
males in free-ranging red howler monkeys of Venezuela.
Ethology 99:75–88
Altmann J (1990) Primate males go where the females are. Anim
Behav 39:193–195
Amsini F, Ilambu O, Liengola I, Kujirakwinja D, Hart J, Grossman F,
Plumptre AJ (2008) The impact of civil war on the Kahuzi-Biega
National Park: results of surveys between 2000–2008
Borris C, Launhardt K, Epplen C, Epplen JT, Winkler P (1999) Males
as infant protectors in Hanuman langurs (Presbytis entellus)
living in multimale groups: defense pattern, paternity, and sexual
behavior. Behav Ecol Sociobiol 46:350–356
Bradley Bj, Robbins MM, Williamson EA, Steklis DH, Steklis NG,
Eckhardt N, Boesch C, Vigilant L (2005) Mountain gorilla tug-
of-war: silverbacks have limited control over reproduction in
multimale groups. Proc Natl Acad Sci USA 102:9418–9423
Busse CD, Gordon TP (1983) Attacks on neonates by a male
mangabey (Cercocebus atys). Am J Primatol 6:133–141
Crockett CM, Sekulic R (1982) Gestation length in red howler
monkeys. Am J Primatol 3:291–294
Doran DM, McNeilage A (2001) Subspecific variation in gorilla
behavior: the influence of ecological and social factors. In:
123
302
Robbins MM, Sicotte P, Stewart KJ (eds) Mountain gorillas:
three decades of research at Karisoke. Cambridge University
Press, London, pp 123–149
Dunbar RIM (1988) Primate social system. Groom Helm, London
Fairgrieve C (1995) Infanticide and infant eating in the blue monkey
(Cercopithecus mitis stuhlmanni) in the Budongo Forest
Reserve, Uganda. Folia Primatol 64:69–72
Fossey D (1982) Reproduction among free-living mountain gorillas.
Am J Primatol 1:97–104
Fossey D (1983) Gorillas in the mist. Houghton Mifflin, Boston
Fossey D (1984) Infanticide in mountain gorillas (Gorilla gorilla
beringei) with comparative notes on chimpanzees. In: Hausfater
G, Hrdy SB (eds) Infanticide: comparative and evolutionary
perspectives. Aldine, Hawthorne, pp 217–236
Goldsmith ML (1999) Ecological constraints on the foraging efforts
of western gorillas (Gorilla gorilla gorilla) at Bai Hokou,
Central African Republic. Int J Primatol 20:1–24
Hall JS (1994) Counting gorillas in Zaire’s Kahuzi-Biega NP. African
Wildlife Update Nov–Dec:6
Harcourt AH (1978) Strategies of emigration and transfer by
primates, with particular reference to gorillas. Z Tierpsychol
48:401–420
Harcourt AH, Greenberg J (2001) Do gorilla females join males to
avoid infanticide? A quantitative model. Anim Behav 62:905–
915
Harcourt AH, Stewart KJ (2007) Gorilla society: conflict, compro-
mise and cooperation between the sexes. The University of
Chicago Press, Chicago
Harcourt AH, Stewart KJ, Fossey D, Watts DP (1980) Reproduction
in wild gorillas and some comparisons with chimpanzees.
J Reprod Fert 28:59–70
Harcourt AH, Stewart KJ, Fossey D (1981) Gorilla reproduction in
the wild. In: Graham CE (ed) Reproductive biology of the great
apes. Academic, New York, pp 265–279
Hasegawa T (1989) Sexual behavior of immigrant and resident
female chimpanzees at Mahale. In: Heltne PG, Marquardt LA
(eds) Understanding chimpanzees. Harvard University Press,
Cambridge, pp 90–103
Hrdy SB (1979) Infanticide among mammals: a review, classification,
and examination of the implications for the reproductive
strategies of females. Ethol Sociobiol 1:13–40
Inogwabini B, Hall JS, Vedder A, Curran B, Yamagiwa J, Basabose K
(2000) Status of large mammals in the mountain sector of
Kahuzi-Biega National Park, Democratic Republic of Congo, in
1996. Afr J Ecol 38:269–276
Janson CH, Goldsmith ML (1995) Predicting group size in primates:
foraging costs and predation risks. Behav Ecol 6:326–336
Janson CH, van Schaik CP (2000) The behavioral ecology of
infanticide by males. In: van Schaik CP, Janson CH (eds)
Infanticide by males and its implications. Cambridge University
Press, Cambridge, pp 469–494
Kalpers J, Williamson EA, Robbins MM, McNeilage A, Nzamuram-
baho A, Lola N, Mugiri G (2003) Gorillas in the crossfire:
assessment of population dynamics of the Virunga mountain
gorillas over the past three decades. Oryx 37:326–337
McNeilage A, Plumptre AJ, Brock-Doyle A, Vedder A (2001) Bwindi
Impenetrable National Park, Uganda: Gorilla census 1997. Oryx
35:39–47
Melanie ID, Gatti S, Levrero F, Duhamel G, Bermejo M, Vallet D,
Menard N, Petit E (2007) Sex-biased dispersal in western lowland
gorillas (Gorilla gorilla gorilla). Mol Ecol 16:2247–2259
Murnyak FD (1981) Censusing the gorillas in Kahuzi-Biega National
Park. Biol Conserv 21:163–176
Nsubuga AM, Robbins MM, Boesch C, Vigilant L (2008) Patterns of
paternity and group fission in wild multimale mountain gorilla
groups. Am J Phys Anthropol 135:263–274
123
Primates (2009) 50:293–303
Nunn CL, van Schaik CP (2000) Social evolution in primates: the
relative roles of ecology and intersexual conflict. In: van Schaik
CP, Janson CH (eds) Infanticide by males and its implications.
Cambridge University Press, Cambridge, pp 388–419
Palombit RA (2003) Male infanticide in savanna baboons: adaptive
significance and intraspecific variation. In: Jones CB (ed) Sexual
selection and reproductive competition in primates: new per-
spectives and directions. American Society of Primatologists,
Norman, pp 367–411
Palombit RA, Cheney DL, Seyfarth RM (2001) Female-female
competition for male ‘‘friends’’ in wild chacma baboons (Papio
cynocephalus ursinus). Anim Behav 61:1159–1171
Parnell RJ (2002) Group size and structure in western lowland gorillas
(Gorilla gorilla gorilla) at Mbeli Bai, Republic of Congo. Am J
Primatol 56:193–206
Pope TR (1990) The reproductive consequences of male cooperation
in the red howler monkey: paternity exclusion in multi-male and
single-male troops using genetic markers. Behav Ecol Sociobiol
27:439–446
Robbins MM (2001) Variation in the social system of mountain
gorillas: the male perspective. In: Robbins MM, Sicotte P,
Stewart KJ (eds) Mountain gorillas: three decades of research at
Karisoke. Cambridge University Press, Cambridge, pp 29–58
Robbins MM (1995) A demographic analysis of male life history and
social structure of mountain gorillas. Behaviour 132:21–47
Robbins MM (1999) Male mating patterns in wild multimale
mountain gorilla groups. Anim Behav 57:1013–1020
Robbins MM (2003) Behavioral aspects of sexual selection in
mountain gorillas. In: Jones CB (ed) Sexual selection and
reproductive competition in primates: new perspectives and
directions. American Society of Primatologists, Norman, pp
477–501
Robbins AM, Robbins MM (2005) Fitness consequences of dispersal
decisions for male mountain gorillas (Gorilla beringei beringei).
Behav Ecol Sociobiol 58:295–309
Sato H, Yasui K, Byamana K, Mugunda B, Murhagane M (2000)
Follow-up survey of environmental impacts of the Rwandan
refugees on eastern D.R. Congo. Ambio 29(2):122–123
Schaller GB (1963) The Mountain Gorilla: ecology and behavior.
University of Chicago Press, Chicago
Sommer V (1994) Infanticide among the langurs of Jodhpur: testing
the sexual selection hypothesis with a long-term record. In:
Parmigiani S, vom Saal FS (eds) Infanticide and parental care.
Harwood Academic Publishers, Chur, pp 155–198
Steenbeek R (1999) Tenure related changes in wild Thomas’s langurs:
I: between-group interactions. Behaviour 136:595–650
Stewart KJ (1988) Suckling and lactational anoestrus in wild gorillas
(Gorilla gorilla). J Reprod Fert 83:627–634
Stewart KJ, Sicotte P, Robbins MM (2001) Mountain gorillas of the
Virungas: a short history. In: Robbins MM, Sicotte P, Stewart KJ
(eds) Mountain gorillas: three decades of research at Karisoke.
Cambridge University Press, London, pp 1–26
Stokes EJ (2004) Within-group social relationships among females
and adult males in wild western lowland gorillas (Gorilla gorilla
gorilla). Am J Primatol 64:233–246
Stokes EJ, Parnell RJ, Olejniczak C (2003) Female dispersal and
reproductive success in wild western lowland gorillas (Gorilla
gorilla gorilla). Behav Ecol Sociobiol 54:329–339
Struhsaker TT, Leland L (1985) Infanticide in a patrilineal society of
red colobus monkeys. Z Tierpsychol 69:89–132
Tutin CEG (1996) Ranging and social structure of lowland gorillas in
the Lopé Reserve, Gabon. In: McGrew WC, Marchant LF,
Nishida T (eds) Great ape societies. Cambridge University Press,
Cambridge, pp 58–70
Van Noordwijk MA, van Schaik CP (2000) Reproductive patterns in
eutherian mammals: adaptations against infanticide? In: van
Primates (2009) 50:293–303
Schaik CP, Janson CH (eds) Infanticide by males and its
implications. Cambridge University Press, Cambridge, pp 322–
360
van Schaik CP (2000a) Infanticide by male primates: the sexual
selection hypothesis revisited. In: van Schaik CP, Janson CH
(eds) Infanticide by males and its implications. Cambridge
University Press, Cambridge, pp 27–60
van Schaik CP (2000b) Social counterstrategies against infanticide by
males in primates and other mammals. In: Kappeler PM (ed)
Primate males. Cambridge University Press, Cambridge, pp 34–52
Watts DP (1984) Composition and variability of mountain gorilla
diets in the Central Virungas. Am J Primatol 7:323–356
Watts DP (1989) Infanticide in mountain gorillas: new cases and a
reconsideration of the evidence. Ethology 81:1–18
Watts DP (1990) Mountain gorilla life histories, reproductive
competition, and socio-sexual behavior and some implication
for captive husbandry. Zoo Biol 9:185–200
Watts DP (1992) Social relationships of immigrant and resident
female mountain gorillas. 1: male–female relationship. Am J
Primatol 28:159–181
Watts DP (1996) Comparative socio-ecology of gorillas. In: McGrew
WC, Marchant LF, Nishida T (eds) Great ape societies.
Cambridge University Press, Cambridge, pp 16–28
Watts DP (1998) Long-term habitat use by mountain gorillas (Gorilla
gorilla beringei). 1. Consistency, variation, and home range size
and stability. Int J Primatol 19:651–680
Watts DP (2000) Causes and consequences of variation in male
mountain gorilla life histories and group membership. In:
Kappeler P (ed) Primate males. Cambridge University Press,
Cambridge, pp 169–179
Yamagiwa J (1983) Diachronic changes in two eastern lowland
gorilla groups (Gorilla gorilla graueri) in the Mt. Kahuzi region,
Zaire. Primates 24:174–183
Yamagiwa J (1986) Activity rhythm and ranging of a solitary male
mountain gorilla (Gorilla gorilla beringei). Primates 27:273–282
Yamagiwa J (1987a) Male life history and the social structure of wild
mountain gorillas (Gorilla gorilla beringei). In: Kawano S,
303
Connell JH, Hidata T (eds) Evolution and coadaptation in biotic
communities. University of Tokyo Press, Tokyo, pp 31–51
Yamagiwa J (1987b) Intra- and inter-group interactions of an all-male
group of Virunga mountain gorillas (Gorilla gorilla beringei).
Primates 28:1–30
Yamagiwa J (1999) Socioecological factors influencing population
structure of gorillas and chimpanzees. Primates 40:87–104
Yamagiwa J (2003) Bushmeat poaching and the conservation crisis in
Kahuzi-Biega National Park, Democratic Republic of the Congo.
In: Price SV (ed) War and tropical forests: conservation in areas
of armed conflicts. Food Products Press, New York, pp 115–135
Yamagiwa J, Basabose AK (2006) Effects of fruit scarcity on
foraging strategies of sympatric gorillas and chimpanzees. In:
Hohmann G, Robbins MM, Boesch C (eds) Feeding ecology in
apes and other primates. Cambridge University Press, Cam-
bridge, pp 73–96
Yamagiwa J, Kahekwa J (2001) Dispersal patterns, group structure
and reproductive parameters of eastern lowland gorillas at
Kahuzi in the absence of infanticide. In: Robbins MM, Sicotte P,
Stewart KJ (eds) Mountain gorillas: three decades of research at
Karisoke. Cambridge University Press, London, pp 89–122
Yamagiwa J, Kahekwa J (2004) First observations of infanticides by a
silverback in Kahuzi-Biega. Gorilla J 29:6–9
Yamagiwa J, Mwanza N, Spangenberg A, Maruhashi T, Yumoto T,
Fischer A, Steinhauer BB (1993) A census of the eastern lowland
gorillas Gorilla gorilla graueri in Kahuzi-Biega National Park
with reference to mountain gorillas G. g. beringei in the Virunga
region, Zaire. Biol Conserv 64:83–89
Yamagiwa J, Maruhashi T, Yumoto T, Mwanza N (1996) Dietary and
ranging overlap in sympatric gorillas and chimpanzees in
Kahuzi-Biega National Park, Zaire. In: McGrew WC, Marchant
LF, Nishida T (eds) Great ape societies. Cambridge University
Press, Cambridge, pp 82–98
Yamagiwa J, Kahekwa J, Basabose AK (2003) Intra-specific variation
in social organization of gorillas: implications for their social
evolution. Primates 44:359–369
123