First Observations of Infanticides by a Silverback in Kahuzi-Biega
Three cases of infanticide have recently been recorded in a group of
eastern lowland gorillas inhabiting the montane forest of Kahuzi-Biega
National Park. This group is called the Chimanuka group, and they have
been habituated for tourism since 2003. The leading silverback, called
Chimanuka (means "appearance"), was born in a habituated group
(the Maheshe group) in 1986. Chimanuka left his natal group in 1998 when
he was 12 years old. In December 2002, he was observed as a leading silverback
with 2 females around another habituated group called the Mugaruka group.
Mugaruka (means "winner") was also born in a habituated group
(the Mushamuka group) in 1987. At the death of Mushamuka in 1997, Mugaruka
was 10 years old (a blackback male). An older male (a maturing silverback)
left the group, but Mugaruka remained with the females in the group, travelling
without any silverback. After large-scale hunting of gorillas in 1998
and 1999, Mugaruka was observed to travel with 3 females and 2 immatures
(Yamagiwa 2003). By 2003, he had acquired 11 females and lost 2 females.
A male baby (called Chubaka, Mugaruka's first offspring with a female
called Lushasha) was born in June, 2000 and another baby (called Maendeleo,
Mugaruka's second offspring with a female called Mwinja) was born in April
2003.
The first interaction between the Chimanuka group and the Mugaruka group
occurred in August 2003. After a fierce fight with physical contact, Mugaruka
suffered wounds on his shoulders. Two females transferred into the Chimanuka
group. The second inter-group interaction occurred in October, when 9
adult females transferred to the Chimanuka group. Mwinja (the mother of
Maendeleo) transferred, while Lushasha (the mother of Chubaka) did not.
At the transfer, Chimanuka stole Maendelo from the chest of its mother
and killed it by biting its face and chest. The dead body of Maendeleo
was abandoned on the ground. Mugaruka traveled with only Lushasha and
her baby Chubaka.
Two females who transferred into the Chimanuka group gave birth in November
and December 2003. A few days after their birth, Chimanuka killed the
two babies. Based on direct observations by park trackers, the victims'
mothers and the other females tried to prevent him from killing the babies
but failed to stop him. In the first case, Chimanuka suddenly rose up
during siesta and grabbed a baby being carried ventrally by its mother.
He carried it away to bite its shoulders. The baby started crying loudly,
and the rest of the group including its mother attacked Chimanuka, trying
to take it back. But Chimanuka ran away fast, carrying the baby, and finally
killed it. The dead baby was left on the ground. The mother picked it
up and continued to carry it for a week until the body became decomposed.
In the latter case, the victim's mother and the other females chased Chimanuka,
who carried the baby. The mother bit Chimanuka on the right foot to stop
the killing; but the baby was killed by Chimanuka's bite. The mother also
carried her dead baby's body until decomposition.
In December 2003, a female who had been with Chimanuka since 2002 gave
birth to a baby called Bonane (means "happy new year"). Chimanuka
did not attack the baby, who has survived without any wound. He was probably
the baby's father.
The third inter-group interaction occurred in January 2004. The last female
in the Mugaruka group transferred into the Chimanuka group, while Chubaka,
her 3-year-old son, did not transfer but remained in his natal group with
Mugaruka. This is the first case observed in Kahuzi-Biega National Park
that a silverback spent a long time together with only an immature male.
Recently several females and immatures immigrated to the Mugaruka group
from the Mishebere group, which previously consisted of 39 individuals
and whose leading male disappeared (probably killed by poachers) in 2003.
In September 2004 (at the time of writing this report), the Mugaruka group
consists of 10 individuals.
From these observations, we found new behavioral tendencies of gorillas
in Kahuzi-Biega National Park. Since the late 1960s when Adrian Deschryver,
the first warden of the park, started to habituate gorillas for tourism,
no infanticide had been observed until the three cases in 2003. We analyzed
the demographic data on 2-4 habituated groups monitored by the park on
a daily basis between 1972 and 1998 (Yamagiwa & Kahekwa 2001). Neither
infanticide nor intense aggression toward any infant by a silverback male
was observed during the 27-year period. Between 1983 and 1998, most of
the individuals in the habituated groups were identified and monitored
by the park. 10 females carried suckling infants and 4 females were accompanied
by weaned juveniles when they transferred between groups during this period,
but no immature was killed or wounded in any of the cases.
Infanticide did not follow either the death of the leading silverback,
or takeover by an extra-group male, in Kahuzi. After the death of Mushamuka,
a leading silverback of a habituated group, 3 females with infants immigrated
into the group. They were accepted by two young males, and no wounds were
observed on either immigrant female or their infants after joining.
After the death of Maheshe, the leading silverback of another habituated
group, most of the females and immatures remained together and traveled
without any silverback for 27 months. Then a young silverback (named Lambchop)
joined the group as the new leading silverback. One year after his joining,
2 females with infants and a female with a juvenile immigrated into the
group. Lambchop was very tolerant toward them, and no infanticide occurred.
Based on these results, we concluded that these aspects of new group formation
and female transfers with dependent infants and juveniles were linked
to the lack of infanticide in Kahuzi and were in pronounced contrast to
the situation in the Virungas, where many occurrences of infanticide have
been reported.
In the Virungas, located about 200 km north of Kahuzi, long-term demographic
data on several habituated groups of mountain gorillas have been collected
by the late Dian Fossey and the Karisoke Research Center since 1967, and
16 cases of infanticides were reported (Fossey 1984; Watts 1989). All
of the victims were suckling infants or old infants less than 3 years
of age, and the killers were extra-group males except for two cases (a
group male and probably females). Infants were killed by extra-group males
when their group encountered other groups or solitary males. When the
leading silverback was present, infants were rarely killed (in 2 cases),
but they were frequently attacked and occasionally killed by extra-group
males after the death of the leading silverback. Infanticide accounted
for 37% of infant mortality in the Virunga Mountain gorilla population.
Watts (1996) argued that infanticide had a great influence on choice of
group by females at the time of transfer and on multi-male group formation.
Female mountain gorillas tend to transfer alone between groups and to
join a large group containing more than 2 silverbacks. After the death
of the leading silverbacks, females did not associate together but dispersed
to join other groups. About half of the reproductive groups are multi-male
groups, and no female group without any silverback has been observed in
the Virungas. These features of female movements and group formation are
probably shaped by the occurrence of infanticide.
The evolutionary implication of infanticide has been explained by the
sexual selection hypothesis (Hrdy 1979). Infanticide by males has been
reported for about 30 species of primates, and it functions as a male
reproductive tactic to hasten the resumption of cycling by suckling females;
it is likely to occur in species with relatively long lactation relative
to gestation (van Schaik 2000). The life history of gorillas may provide
the conditions that favour infanticide by extra-group males. In order
to reduce infanticide risk, females have evolved counter strategies; one
of these is females' prolonged association with protector males (Sterck
et al. 1997). The large proportion of multi-male groups in the Virunga
population is explained as a female strategy against infanticide by choosing
the group offering the greatest male protection (Watts 1996).
We examined the possible reasons for the absence of infanticide in the
Kahuzi population and reached the conclusion that some unknown factors
suppressed the occurrence of infanticide as a potential reproductive tactic
by males in Kahuzi (Yamagiwa & Kahekwa 2001). The 3 infanticides by
Chimanuka observed in 2003 may add new perspectives to our assessment.
First, the large-scale poaching of gorillas in 1998 and 1999 may have
produced conditions favouring infanticide. Thousands of refugees and soldiers
roamed in the forest of Kahuzi during the civil war and hunted large mammals,
frequently for bushmeat. Most of the elephants and about half of the gorillas
were killed during this period (Yamagiwa 2003). All leading silverbacks
in the 5 habituated groups were killed by poachers, and disintegrations
of groups and female transfers frequently occurred after the death of
the leading males. Gorillas tended to shift their range into the central
sector of the park, where the park staff frequently patrolled. These situations
may have stimulated unfamiliar groups to overlap their ranges and so frequently
encounter each other. Infanticides in the Virungas also occurred after
large-scale disturbance by human activities. In the 1960s, the Rwandan
Government took 40% of the park for farm land. Cultivation and cattle
encroachment forced the gorillas to shift their range into the central
part of the Virunga Volcanoes. Schaller (1963) had never observed infanticide
in 1959 and 1960. No infanticide has been reported in the 1990s or more
recently. Unstable relationships between groups and between individuals
within groups induced by large human disturbances may have produced conditions
leading to infanticide.
Recent DNA analysis suggested that the silverbacks of neighbouring groups
were related at Mondika, Central African Republic (Bradley et al. 2004).
Western lowland gorillas possibly form a patrilocal social structure in
which males remain in their natal region and potentially benefit from
kin associations. In Kahuzi, several new groups were formed following
group fissions caused by maturing silverbacks taking females out of their
natal groups, before the large-scale hunting (Yamagiwa & Kahekwa 2001).
The new groups overlapped their ranges with those of their natal groups,
and these new group formations resulted in a local concentration of several
groups in which the leading males were related to each other. Infanticide
is unlikely to occur in these groups because all infants are more or less
related to any adult male, even to extra-group males. Human disturbances
during the civil wars probably destroyed these peaceful relationships
among related males and thus increased encounters and hostility between
unfamiliar males.
The second factor favouring infanticide would have been the recent increase
in the proportion of young silverbacks, which may have increased agonistic
interactions between groups and so caused infanticide. In the Virungas,
most infanticidal males were solitary males, who were strongly motivated
to attract females to establish their own groups (Fossey 1984; Watts 1989).
After the large-scale hunting and disintegration of groups, young males
dispersed as solitary males in Kahuzi. Most of the leading silverbacks
were killed, and the proportion of young silverbacks in the population
may have increased. These unmated males constitute a greater infanticidal
threat than do breeding males. Chimanuka was observed as a solitary male
just before he appeared in the range of the Mugaruka group in 2002. Both
Chimanuka and Mugaruka were young (17 and 16 years old, respectively)
and would be strongly motivated to attract females. They were born in
different groups and were not related to each other. Although the details
of interactions between Chimanuka and the victims' mothers before the
infanticides were not known, Chimanuka's strong motivation for reproduction
may have resulted in his killing these infants of transferred females.
Thirdly, it is worth noting that the occurrences of infanticides in Kahuzi
were different from those in the Virungas. Chimanuka killed 2 newborn
babies just after birth in his group. On the other hand, he did not kill
a baby (Bonane) whose mother had remained for at least one year in his
group prior to delivery. It seems likely that Chimanuka discriminated
the infant he accepted from those he killed, probably based on his past
interactions with their mothers. Was his decision on whether to kill based
on the relatedness of the infants to him? It is difficult to answer this
question because the interactions between Chimanuka and the victims' mothers
remain unknown in detail. The absence of copulation or the short length
of these mothers' stay in his group possibly influenced his decision.
When Chubaka's mother transferred into the Chimanuka group, Chubaka did
not join his mother but remained with Mugaruka, his putative father. In
most of the previous female transfers, immatures transferred with their
mothers in Kahuzi. Chubaka's case suggests that he or his mother may have
learned from the occurrences of infanticide. His mother probably noticed
the risk of infanticide at her transfer and left her weaned infant, who
could travel independently with Mugaruka. If this is the case, female
gorillas may be able to quickly and effectively respond to the occurrence
of infanticide by changing the patterns of association.
Fourthly, we can expect social changes in the Kahuzi population in the
near future. If the female gorillas in Kahuzi can learn the proper strategy
against the risk of infanticide, they will seek more protection from their
partners, as did the female mountain gorillas in the Virungas. They will
transfer alone to avoid competition with other females over a male mate
and will choose a multi-male group to gain more protection. Maturing males
will remain in their natal group after maturity and will start reproduction
without spending time in solitary life. The proportion of multi-male groups
will increase in the Kahuzi population.
Association among related males may be a common social feature within
Homininae, including African great apes and humans. The genus Gorilla
may have evolved two types of association among related males. One is
association within a group, and the other is tolerance between males of
neighbouring groups. The occurrence of infanticide may promote the former,
and its absence may promote the latter. Populations of gorillas may have
such flexibility between these two types of social organization.
Another population of mountain gorillas in the Bwindi Forest contains
a large proportion of multi-male groups as observed in the Virungas, but
no infanticide has been reported so far (Robbins 2001). This suggests
that the Bwindi population may have already experienced infanticide in
the past and may have succeeded in preventing it by shaping multi-male
social structure. Future intensive research on eastern and western lowland
gorillas and DNA analysis will examine this hypothesis.
We suspect, however, that the Kahuzi population is too small to sustain
large social changes. In order to support them, we should realize the
present conditions influencing them and make a wise conservation plan
based on new knowledge about them. We hope this report will contribute
to these urgent actions.
Juichi Yamagiwa and John Kahekwa
Photos: Carlos Schuler
References
Bradley, B. J. et al. (2004) Dispersed male networks in western gorillas.
Current Biology 14: 510-513
Fossey, D. (1984) Infanticide in mountain gorillas (Gorilla gorilla beringei)
with comparative notes on chimpanzees. In: Infanticide: Comparative and
Evolutionary Perspectives. Hausfater, G. & Hrdy, S. (eds.). Hawthorne,
NY (Aldine), pp. 217-236
Hrdy, S. B. (1979) Infanticide among animals. Ethology and Sociobiology
1: 13-40
Robbins, M. M. (2001) Variation in the social system of mountain gorillas:
the male perspective. In: Mountain gorillas. Robbins, M. M. et al. (eds).
Cambridge (Cambridge University Press), pp. 29-58
Schaller, G. B. (1963) The Mountain Gorilla. Chicago (University of Chicago
Press)
Sterck, E. H. M. et al. (1997) The evolution of female social relationships
in nonhuman primates. Behavioral Ecology and Sociobiology 41: 291-309
van Schaik, C. P. (2000) Infanticide by male primates: the sexual selection
hypothesis revisited. In: Infanticide by Males. van Schaik, C. P. &
Janson, C. H. (eds.). Cambridge (Cambridge University Press), pp. 27-71
Watts, D. P. (1989) Infanticide in mountain gorillas: new cases and a
reconsideration of the evidence. Ethology 81: 1-18
Watts, D. P. (1996) Comparative socio-ecology of gorillas. In: Great Ape
Societies. McGrew, W. C. et al. (eds.). Cambridge (Cambridge University
Press), pp. 16-28
Yamagiwa, J. (2003) Bushmeat poaching and the conservation crisis in Kahuzi-Biega
National Park, Democratic Republic of Congo. J. Sustainable Forestry 16:
115-135
Yamagiwa, J. & Kahekwa, J. (2001) Dispersal patterns, group structure
and reproductive parameters of eastern lowland gorillas at Kahuzi in the
absence of infanticide. In: Mountain gorillas. Robbins, M. M. et al. (eds.).
Cambridge (Cambridge University Press), pp. 89-122
Prof. Dr. Juichi Yamagiwa has studied eastern gorillas
since 1978. He promoted a cooperative research project on gorillas and
chimpanzees with CRSN and ICCN in Tshibati.
John Kahekwa has been working at the Kahuzi-Biega National Park
since 1983. He founded the Pole Pole Foundation that has been cooperating
with the population since 1992.