Cross-cultural comparison of captive populations of PAN PANISCUS

Material Information

Cross-cultural comparison of captive populations of PAN PANISCUS
Thompson, Joan A.
Publication Date:
Physical Description:
xi, 94 leaves : ; 29 cm

Thesis/Dissertation Information

Master's ( Master of Arts)
Degree Grantor:
University of Colorado Denver
Degree Divisions:
Department of Anthropology, CU Denver
Degree Disciplines:


Subjects / Keywords:
Bonobo -- Behavior ( lcsh )
Bonobo -- Behavior ( fast )
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )


Includes bibliographical references.
General Note:
Submitted in partial fulfillment of the requirements for the degree, Master of Arts, Anthropology.
General Note:
Department of Anthropology
Statement of Responsibility:
by Joan A. Thompson.

Record Information

Source Institution:
University of Colorado Denver
Holding Location:
Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
28286315 ( OCLC )
LD1190.L43 1992m .T46 ( lcc )

Full Text
Joan A. Thompson
B.A., Wittenberg University, 1987
A thesis submitted to the
Faculty of the Graduate School of the
University of Colorado at Denver
in partial fulfillment
of the requirements for the degree of
Master of Arts

(c) 1992 by Joan A. Thompson
All rights reserved.

This thesis for the Master of Arts
degree by
Joan A. Thompson
has been approved for the
Department of
Duane Quiatt
Janis W. Driscoll
Michael J. Owren
'Y' /?72_

Thompson, Joan A. (M.A., Anthropology)
Cross-Cultural Comparison of Captive Populations of
Pan Paniscus
Thesis directed by Professor Duane Quiatt
Scientists know that chimpanzee groups in the wild
have different behavioral traditions in different parts
of Africa. Using a cross-cultural approach, which may
provide the necessary measurement for the true nature
of a species, and examining behavior of captive bonobos
(Pan paniscus) in the diversity of various environmental
settings, this study aims to demonstrate that populations
of bonobos also display behavioral traditions. The
comparison of different group traditions can clarify
understanding of the behavior of bonobos as a species.
During the summer of 1991, North American zoological
facilities collectively housed a total of 23 bonobos.
This study involves 21 of those apes. The individuals
included in the study were divided into five separate
social groups, in four different zoological parks. A
total of 155 hours of behavioral observations was logged,
in addition to time alloted to care giver/staff
interviews, identification of subjects, observer
orientation and other necessary diversions. The apes

were observed between 0800 and 1700 hours depending on
the times permitted by each facility. When possible,
behavioral observations were scheduled and conducted so
as not to alter the animals normal daily routines.
Comparative descriptions of each facility focused on
diet and food use, intragroup relatonships and
communication, locomotion, state of arousal, play,
sociosexual behavior, and pathologies and idiosyncrasies,
as well as time budgets. The prolonged maturation period
of this species has implications for cultural continuity
or the learning of group traditions. Therefore, the
social development method for each individual is
Remarkable differences were found in the behavioral
repertoire, the stereotypic behaviors, and the types and
rates of social interaction. These differences were
related to the environmental conditions of confinement
and the social history of the groups.
The form and content of this abstract are approved. I
recommend its publication.
Duane Quiatt

It is with heart felt sorrow at the loss of my
friend, Kole, that I dedicate this thesis to her and to
my husband and partner, Craig. Roles short and tragic
life touched the deepest parts of me. She taught me

I would like to express my appreciation to the
bonobo caregivers and administrators of The Zoological
Society of San Diegos Center for the Reproduction of
Endangered Species, The Senior Animal Staff Committee of
the Milwaukee County Zoo, The Columbus Zoological Park
Association, and The Cincinnati Zoo. My special thanks
to Nancy Harvey, Jan Rafert, Don Winstel, and Bob Lotshaw
for their responsiveness and cooperation.
I am grateful to Dr. Duane Quiatt for his continuing
support as my graduate work advisor. Dr. Quiatt
unconditionally encouraged me to pursue my goals. 1 also
appreciate the constructive criticism and encouragement
from Dr. Janis Driscoll and Dr. Michael Owren.
These acknowledgments would not be complete without
expressing my deepest gratitude and admiration to Dr.
Charles Southwick for his guidance during this project.
He is both a mentor and a friend.

1.1. Illustration of the bonobo 3
1.2. Historic range 9
1.3. Recent estimated range 10
2.1. Sample scoring sheet 25
2.2. Solitary arboreal stationary behavior category 33
2.3. Solitary arboreal movement behavior category 34
2.4. Solitary terrestrial stationary behavior category 35
2.5. Solitary terrestrial movement behavior category 36
2.6. Reclining behavior category 37
2.7. Feeding behavior category 39
2.8. Play behavior category 40
2.9. Grooming behavior category 41
2.10. Species-specific nongrooming interaction behavior category 42
2.11 . Grooming graph without juveniles behavior category 43
2.12. Species-specific nongrooming interaction without juveniles behavior category ... 44

2.1. Population membership ....................... 19
2.2. Research subjects and codes ................. 20
2.3. Environmental settings ...................... 22
2.4. Behavioral categories ....................... 26
2.5. ANOVA calculations .......................... 29
3.1. 1971 population subjects......................79
3.2. 1983 population subjects......................80
3.3. Offspring of original U.S. breeding pair . . 81

Figures .....................................
Tables ......................................
1. INTRODUCTION ................................
Contemporary Focus on Pan paniscus . .
Pan paniscus in situ ...................
Distribution and Status ...........
Field Research ....................
Pan paniscus ex situ ...................
Subjects ...............................
Zoo Settings ...........................
Procedure ..............................
Statistical Analysis ...................
Results ................................
Time Budgets ...........................
Discussion .............................
3. GENERAL DISCUSSION ..........................
Diet and Food Use ......................
Intragroup Relationships/Communication .
Locomotion .............................
State of Arousal .......................
. 1
. 4
. 7
. 8

Sociosexual Behavior ........................ 70
Pathologies and Idiosyncrasies .............. 74
A Suggestion Concerning Culture ............. 77
Summary and Conclusions ..................... 86
SELECTED BIBLIOGRAPHY .......................... 90

By examining the behavior of captive Pan paniscus in
different environmental settings this research presents a
comparative, descriptive study of these captive
behaviors. The influences of social development and
environmental differences are analyzed and discussed.
During the summer of 1991, North American zoological
facilities collectively housed a total of 23 Pan
paniscus. This research involves 21 of those apes.
Comparative inquiry at each facility focused on habitat,
food use, intragroup relationships, facial and gestural
communication methods, common methods of locomotion,
states of arousal, play behaviors, sociosexual behaviors,
behavioral pathologies and idiosyncrasies, as well as
activity patterns.
In 1925, Robert M. Yerkes pioneered what may have
been the first behavioral study comparing Pan paniscus
and Pan troglodytes. In fact, prior to the work
presented here, Pan paniscus had been studied only in
comparison to Pan troglodytes. Therefore, this study was
undertaken to compare captive behaviors between groups
within the species Pan paniscus.
Chimpanzee groups in the wild have different

behavioral traditions in different parts of Africa
(Goodall, 1986). So, to better understand a species, a
cross-cultural approach may provide the best measures for
a truer* nature of that species. The environment may
influence acquisition of new behavioral habits. Habits
thus acquired may be passed on from parent to offspring
or learning reinforced by imitation. Imitation is
characteristic of juvenile chimpanzee and bonobo behavior
and there may be a period of time during a young animals
life when it is especially sensitive to and more likely
to learn about particular features of the social
The prolonged maturation period of Pan paniscus thus
has implications for "cultural continuity or learning
group traditions. During particular stages of the
juvenile or infant apes life, the social environment may
have a profound influence upon later behavior. By
remaining with their biological progenitor during the
early stages of social development or by interacting with
human nursery caregivers, the future behavior of infants
can be greatly influenced by parents, whether human or
non-human. Therefore, the social development of group
members will be emphasized where applicable.
Pan paniscus can be recognized by its black face and
round eyes, small ears set close to a small spherical
head, reddish lips and hair fringing its face,


characteristically parted down the middle (see Figure
1.1.)* There continues to be an ongoing debate about the
correct generic name for Pan paniscus. Prior to the
official designation of species in 1933, this ape was
called a pygmy chimpanzee. The population was
incorrectly described as a smaller version or subspecies
of the chimpanzee {Pan troglodytes). Expanded knowledge
of the chimpanzee has identified statistical data which
indicate that, in fact, the chimpanzee subspecies Pan
troglodytes schweinfurthii is comparable in body weight
to Pan paniscus. Yet the bonobo has a smaller chest
girth and longer legs relative to body height than does
Pan troglodytes (Coolidge & Shea, 1982). Therefore, to
call the bonobo a pygmy chimpanzee perpetuates an
inaccurate perception. So, as do many scientists, this
thesis will identify this species as the bonobo and Pan
troglodytes as the chimpanzee.
Contemporary Focus on Pan Paniscus
Although there is no ideal or "best" model, in
recent years the bonobo has been proposed as the
fashionable model for understanding early hominid
behavior because of its close relationship to man (Bogart
& Benirschke, 1977). In the scientific search for that
common ancestor of humans and apes, the bonobo more than

any extant ape species is believed to best represent
traits which this "missing link" may have exhibited.
Modern biochemical analysis seems to imply, depending on
the interpretation, that the African apes speciated at
approximately the same time (Sarich, 1971). Some
scientists compare the bonobo to that extinct ancestor
who was common to the Pongidae and Hominidae families
(Zihlman, Cronin, Cramer, & Sarich, 1978). Using
analysis of the fossilized hominids, the arboreal
environment, and the anatomy of living apes, scientists
can consider the morphological, ecological, and
behavioral bonobo model.
The bonobo model emphasizes the "primitive"
features, those traits characteristic of being
phylogenetically older, of the limb bones. Zihlman and
Cramer (1978) speculate that the bonobo, as the only
extant African ape specialized for forest life, has
adapted fewer derived features than the other extant apes
and therefore more closely resembles the ancestral
condition. Using Australopithecus afarensis, one of the
oldest hominid fossil species known, in an anatomical
comparison to the bonobo, Zihlman attempts to reconstruct
the common ancestor. She argues that the similarity in
overall size, cranial capacity, and relative size of the
leg bones, suggests that early hominids might have
evolved from a small bonobo-like ancestor (Zihlman,

1982). Thus the bonobo can serve as a prototype for the
common ancestor. Susman (1984) tempers his support for
the bonobo model but also maintains that the bonobo may
mirror the ancestral life and need for behavioral
flexibility. Herbert (1983) writes that the systematic
studies of the bonobo suggest that this isolated and
nearly extinct ape may provide the best model for
understanding the evolution of arboreal ape into
primitive man.
Although there continues to be an ongoing debate
about the bonobos position in understanding human
origins, there are other distinctive features of the
bonobo model to consider. The bonobo is unique compared
to other nonhuman primates in its frequency of
ventro-ventral copulation, tendency for copulation to be
initiated by either sex, females longer periods of
sexual receptivity during the estrus cycle and sooner
after the birth of an infant, propensity for food
sharing, predisposition for bipedal locomotion, strong
affiliative relationships between females, and a more
equal socionomic sex ratio in bonobo groups. All of
these characteristics have been compared to behaviors
previously thought to be exclusively human. Diamond
(1991) emphasized that humans have had only a short
history as a species distinct from other apes, especially
the genus Pan. His recent argument defends the plausible

reclassification of chimpanzees, bonobos, and humans as
Homo troglodytes, Homo paniscus, and Homo sapiens.
"Since our genus name Homo was proposed first it takes
priority, by the rules of zoological nomenclature, over
the genus name Pan (Diamond, 1991, p. 21).
Pan Paniscus in situ
Zaire is renowned for its biodiversity. It is the
only country in the world which is inhabited by all three
of the African great apes: the gorilla (including the
mountain gorilla, Gorilla gorilla beringei, and the
eastern lowland gorilla, Gorilla gorilla graueri), the
eastern or long-haired chimpanzee (Pan troglodytes
schweinfurthii), and the bonobo (Pan paniscus). The
bonobo, although living allopatrically from the other
species, is found only in Zaire. The Zaire River is a
geographic barrier which genetically isolates this
species. The bonobo is an ecological specialist adapted
to an arboreal environment from years of evolving in a
rainforest habitat (Kano, 1984). Some speculation has
focused on the ventrally situated external genitalia
(Savage-Rumbaugh and Wilkerson, 1978) as a morphological
indicator of an arboreal lifestyle and behaviors that are
obviously better suited for such an environment.

Distribution and Status
The bonobo's geographic distribution has
historically been limited to the comparatively small
region of forest on the left bank of the Zaire River (see
Figure 1.2.). In his ecological pilot study, Kano (1979)
defined its distribution as that area encompassed in the
northern "tip of an imaginary arc connecting the left
banks of the Zaire and Lualaba Rivers" (p. 123). The
southern range boundary has never clearly been defined,
but it is that boundary and the western area that is
shrinking. National Geographic (March 1992) illustrated
the estimated bonobo range, before September 1991, as a
reversed crescent moon shape limited to a small area in
the north eastern tip of that imaginary arc (see Figure
1.3. ) .
A rough preliminary survey to determine wild
population distribution was conducted in 1955 but no
further data were collected until the early 1970s.
Takayoshi Kano led the most recent major census in 1973
and estimated a population of approximately 50,000
bonobos. Concern about the perceived loss of wild
populations due to hunting and habitat loss prompted a
proposed intensive two year region-wide census survey in
1992 (T. Kano, 1990, personal communication). Susman et
al. (1981) report that the major threats to feral

Historic Range of the Bonobo
Figure 1.2.

Recent Estimated Range of the Bonobo
Figure 1.3.

populations are hunting, illegal trading, encroachment by
local slash-and-burn farmers, oil exploration, and
commercial logging. Field researchers speculate that as
much as one half of the 1973 estimated population had
been destroyed prior to 1990. From estimates based on
knowledge prior to the summer of 1991, National
Geographic (March, 1992) reports possibly as few as
10,000 bonobos exist in the forests of Zaire.
The current outlook of the bonobos status is bleak.
The politico-economic instability and continued civil
disorder in Zaire escalated in September of 1991 bringing
all scientific field investigations of bonobos to a halt.
Since scientists have not been able to focus on field
work they have come together to prepare a global
conservation action plan.
Field Research
Prior to September of 1991 there were but a few
research sites and a few researchers actively collecting
data pertaining to the bonobo. Relatively little
information has been recorded about this ape and even
less is available in published form. In 1972 scientists
in association with Kyoto University of Japan undertook
the establishment of the first long-term field site
dedicated to the study of the bonobo. This site is known

as The Wamba Research Station. Although controversial,
provisioning was instituted at the Wamba site in 1977
(Kano, 1982). Soon to follow, in 1979, was the formation
of The Lomako Pygmy Chimpanzee Research Station in
association with The State University of New York at
Stony Brook. To date observations continue to be
difficult at the Lomako site because those bonobos
persist in being unwilling to tolerate human presence if
the apes are on the ground (N. Thompson-Handler, 1992,
personal communication). Obviously this circumstance
will bias data toward an arboreal environment.
Therefore, much of the available information about the
bonobo results from observations of the wild provisioned
groups at the Wamba field site. These data may have a
terrestrial bias as well as possible group size and
interaction distortion.
Other locations concerned with studying the wild
bonobo include short-term observations near Lake Tumba
(Horn, 1980), an ecological study at Yalosidi of Ikela
zone (Kano, 1983; Uehara, 1988), and a Spanish research
team in the Lilungu region of the Zone of Ikela (in
press). It is apparent that few long-term studies,
critical for establishing evidence for culture, have been
conducted in the wild, primarily because few populations

have been identified and field work in Zaire is unusually
difficult, with bonobo groups being particularly
The bonobo is most frequently observed in parties of
both sexes and all age classes (White, 1988). Kano
(1982) reports that when the groups divide and merge,
their groups are larger than the chimpanzee groups.
Again it must be noted that the greater part of the
available data, which come from the Wamba site, is
influenced by provisioning of the subjects. In some
instances the observations were enhanced by baiting or
"mobile provisioning" (Kano, 1982) wherever the apes were
located. Kuroda (1979) reported the average group size at
Wamba to be 16.9 individuals. At Lomako the size of
groups or associations is smaller with the average being
7.9 (Badrian & Badrian, 1984). Kitamura (1983)
emphasized that individuals in a bonobo subgroup, who
move in association with each other for a relatively long
period of time, use the same range equally. This can be
illustrated by considering a core area, or island, which
a subgroup might utilize.
The bonobo seems to demonstrate a preference for
terrestrial dwelling. Badrian, Badrian, and Susman
(1981) report that the Lomako apes travel mainly along
the ground. Although the bonobos are primarily
frugivorous, they supplement their diet with the fibrous

foods most often associated with the gorilla diet and
found along the ground level of the forest. Although
Kano (1979) reported that the bonobo is a better climber
than the chimpanzee and that bonobos are quite agile in
the forest canopy, his long-term observations indicate
that bonobos are primarily terrestrial.
Pan Paniscus ex situ
Historically, captive life for the bonobo has not
been easy. The bonobo was "the last of the great apes to
be bred in captivity" (Schmidt, 1986, p. 117). On 22nd
January 1962, the first captive birth took place at the
Frankfurt Zoo (Kirchshofer, 1964). While this male
infant lived only two months, his short existence marked
a worldwide husbandry milestone for primatologists and
zoologists. On 22nd December 1963, the second captive
born bonobo was produced from the same parents at the
Frankfurt Zoo. This male has survived today and was
recently transferred to a social group at the Wuppertal
Zoo (R. Kirchshofer, 1992, personal communication) where
it is hoped that he will continue to successfully
Europe and the United States have formed
organizations whose mission plans are to "maintain the
bonobos in captivity in a way which preserves the natural

integrity of the species as much as possible (Reinartz,
SSP Master Plan 1991-1992, p. iii). The American
Association of Zoological Parks and Aquariums (AAZPA)
established the Bonobo Species Survival Plan (SSP) in
March of 1988. The European counterpart, which acts
cooperatively with the SSP, is known as the Europaisches
Erhaltungszuncht Programm or EEP. As reported in the 6th
November 1991 EEP Bonobo Workshop Minutes, the EEP
manages a captive European population of 39 bonobos. The
current U.S. population includes the captive management
of 41 individuals. At the time of this thesis research
the Yerkes Regional Primate Research Center housed three
bonobos at its language center, three at its main station
for breeding and two separate groups, with a total of
nine animals, at the field station for behavioral

The first goal of this normative study was to
consider the range of normal behavioral variations for
bonobos in captive environments. With some determination
about normal variation, relationships can be established
through a comparative analysis of differences seen in
different environments. Without attributing a causal
explanation, behavioral categories focus on terrestrial
and arboreal realms of activity, play, feeding time, and
interactive pursuits. Variations in activity are
reviewed with regard to a landscape immersion captive
habitat. Landscape immersion is a contemporary zoo
design where the animals are caged in naturalistic
looking environments. Groups B and C are housed in such
an environment. During workshop discussions of the
captive management of this species, some experts have
vehemently debated the merit of this landscape immersion
habitat. It has been argued that this exhibit design is
better for the apes. The issue is particularly relevant
to these facilities because at the time of this research
the Milwaukee, Columbus, and Cincinnati zoos were each
planning to provide new, more naturalistic environments

for their bonobos. Recent management of bonobos has
emphasized consideration of the group social behavior, as
well as the economics of visitor perceptions and
educating the public.
The research population included 21 bonobos housed
in five separate social groups, in four different
zoological parks: the San Diego Wild Animal Park in
California, the Milwaukee County Zoo in Wisconsin, the
Columbus Zoo in Ohio, and the Cincinnati Zoo in Ohio.
Their ages ranged from 6 months to 19 years old. Age
classifications were based on categories defined by Horn
(1980) from observations of captive bonobos: infant, 0-4
years; juvenile, 5-8 years; adolescent, 9-12 years;
adult, 13 years and older. Individual animals are listed
in Table 2.1. However, data were collected on 15 of the
21 animals. The 15 subjects are identified and coded in
Table 2.2. Their ages ranged from 6 years to 19 years
old. However, all but two focal animals were either
adults or adolescents.
Group A included an adult male, Lodi, and an adult
female, Maringa, who had been a successful breeding pair
for nearly a decade. These two animals had produced four
offspring, of which two were still living. They both

represent founder genes in the captive population and had
the benefit of early social development in their
respective wild natal groups. This collection also
included their juvenile male offspring, Lomako, and their
infant female offspring, Eliya. The adult female was in
lactational amenorrhea.
Group B encompassed an adult female, Loretta; an
adolescent female, Lana; a juvenile female, Lena; and
three infant females, Laverne, Marylin-Lori, and Lolita.
The adult male, Vernon, is a wild caught individual and
was this groups only animal not nursery-reared or raised
by a nursery-reared mother. Laverne was being
mother-reared by Lana, and Lena was mother-reared by
Loretta. Loretta, Lana, Marylin-Lori, and Lolita all
experienced social development in the nursery. Lolita,
recently out of species isolation, was not grouped with
her mother and had developed a strong dependence on Lana
for a surrogate relationship. In species isolation
Lolita was in the zoo nursery, at a different facility,
where she had no contact with other bonobos. Lana was
obviously menstruating and Loretta appeared to be in
midtumescence of her cycle.
Group C consisted of an adolescent male, Akili; an
adult female, Louise; an adolescent female, Lisa; and her
infant female, Lucy. The two adolescents were both
mother-reared at the Yerkes Regional Primate Center and

- TABLE 2.1. -
Population Membership
Group Individual Sex Age Rearing
A Milwaukee Lodi M 17yr. wild
Maringa F 18yr. wild
Lomako M 7yr. mother
Eliya F 6 mo. mother
B San Diego Vernon M 19yr. wild
Loretta F 17yr. nursery
Lena F 6yr. mother
Lana F 12yr. nursery
Lave me F 3yr. mother
Marylin-Lori F 3yr. nursery
Lolita F lyr. nursery
C San Diego Akili M 1 lyr. mother
Lisa F lOyr. mother
Lucy F lyr. mother
Louise F 18yr. nursery
D Columbus Jimmy M 15yr. wild
Toby M 14yr. wild
Suzie F lOyr. wild
Lady F lOyr. wild
E Cincinnati Kevin M 15yr. nursery
Connie-Lenore F 9yr. mother
are half-siblings on their paternal side. The adult
female was nursery-reared. She had three offspring but
was able to rear only her first. Her offspring were not
in her social group. During this study, Louise was
clearly at maximum tumescence and Lisa was midtumescence
while still actively nursing Lucy.

Off exhibit, due to recent acquisition and
quarantine, Group D was made up of the newest additions
to the North American collective population. Two adult
males, Jimmy and Toby, and two adolescent females, Suzie
and Lady, all represent apparently unrelated founder
genes. As wild caught apes they each had the opportunity
for early social development with natal groups. None of
- TABLE to to l l
Name Research Code Subjects Age Class Sex
Lomako L juvenile M
Maringa M adult F
Lodi D adult M
Lana N adolescent F
Louise U adult F
Vernon R adult M
Akil i K adolescent M
Lena E juvenile F
Lisa I adolescent F
Jimmy J adult M
Toby T adult M
Suzie S adolescent F
Lady A adolescent F
Connie-Lenore C adolescent F
Kevin V adult M
these animals have produced offspring because, since
capture, they had been housed in single sex pairs or
groups. They had been caged as a mixed sex group for

only a few months and the group social dynamics were
still very much unfolding. Caregivers estimated both
females to be near maximum tumescence.
Finally, Group E was the smallest social group in
the study. The adult male, Kevin, is the full uncle of
the adolescent female, Connie-Lenore, on her maternal
side. Kevin is a product of nursery social development.
He had never successfully bred and was initially grouped
with Connie-Lenore in 1983 when their social group of
three animals included her mother, Louise, from Group C.
At that time Connie-Lenore was being mother-reared and
may have considered Kevin as a sibling. Connie-Lenore
appeared to be midtumescence.
Zoo Settings
Brief descriptions of facility settings are provided
in Table 2.3. Some distinctive features of each setting
were as follows. In Milwaukee one of only a few public
restrooms on the park grounds was located next to the
exhibit. This increased the volume of public traffic
tremendously. Groups B and C, in San Diego, were
exhibited outside on alternate days. The off-exhibit
group was confined to the bedroom area for that day. In
the event that rain was forecasted, both groups remained
inside. The two groups had vocal and auditory contact

- TABLE 2.3. -
Group Location Facility Description
A Milwaukee four inside concrete block rooms
with dimensions of lOXS, 12X8, 10X8, and
One room front window was covered for privacy.
Furnishings included: dead trees, wall shelves at
varying heights, corner water spigots, ropes, rubber
balls, rubber tubs, wood wool, hay, and in the
largest room there were full size rubber trash bins,
a log swing, a large rope airport cargo net hammock
and a plastic milk crate.
Skylights were covered by steel mesh.
The public was restricted by a metal fence-like
barrier approximately five feet from the glass plate
which formed the front exterior wall of the exhibit
B San Diego - 1.5 acre dry-moated island
originally designed for Drills (Papio leucophaeus).
Landscape immersion: large palm trees, foliage and
other ground level vegetation covering the island
Furnishings included: a rubber ball.
C San Diego - 1.5 acre dry-moated island
originally designed for Drills (Papio leucophaeus).
Landscape immersion: large palm trees, foliage and
other ground level vegetation covering the island
Furnishings included: a rubber ball.
D Columbus - three inside concrete block rooms
with dimensions of 10X12, 12X12,t and 12X12 and
two outside rooms each 12.5X12; two 5 foot long
access tunnels to outside, measured 2X3.5 each.
Furnishings included: rubber balls, rubber tubs,
climbing apparatus, bed platforms, and wood wool.
E Cincinnati - two inside, equal sized rooms with
dimensions of 12X12X15 each.
Furnishings included: bare trees, ropes, a large
plastic barrel, and hay.
The public was restricted by a metal fence-like
barrier approximately three feet from the front
exhibit window. The front wall in each room was
designed with small criss-crossing, one and one-half
inch block bars. There was a keeper access
walk-way, approximately five feet wide between the
bars and the front window.

during the day as well as visual and olfactory contact at
night. After approximately 1000 hour each day monorail
trains began to pass carrying patrons of the park.
During the remainder of the time the animals were
undisturbed by the public. The building which housed
Group D, in Columbus, served for many years as an animal
hospital and prior to that as an exotic feline house. The
group was alone in the building except for two red hawks
housed outside at the other end of the complex. Due to
the limited light and visual barriers (such as darkened
glass, steel mesh, and construction material), viewing
the activity of black apes was extremely difficult
inside. The outside rooms were only available for a
maximum of seven hours each day. During the days when
one of the females served as the focal animal, the group
was locked outside during the observation session. The
females tended to spend more time inside and therefore
were difficult to observe. This group had no immediate
exposure to the public but curious animal watchers could
look over and/or gesture over a six-foot-high privacy
fence when the animals were outside. If the apes wanted
to observe activity beyond the gate they could position
themselves at the height of the bars (see later
discussion). Group E, in Cincinnati, had visual access
directly into the caregivers preparation kitchen from

either room. Caregiver activity levels influenced the
apes activities. Groups A, D, and E were exposed to
inside artificial lighting 24 hours per day.
The 15 subjects were observed each day for periods
between 0800 and 1700 hours depending on times permitted
by each facility. However, data were primarily collected
between 0900 to 1500 hours. In other words, regardless
of what each facility would permit, observation always
occurred during those six hours. Sometimes the data
collection would begin earlier or last longer. For
instance, at San Diego observations began in the morning
when the group was released onto the island. At
Cincinnati the observations were not permitted to begin
prior to the general admittance of zoo patrons.
Behavioral observations were scheduled and conducted
so as not to alter the animals normal daily routines,
when possible (see comments above for group D). A total
of 155 hours of behavioral observations were conducted.
Data were collected on a different focal animal (Altmann,
1974) each day, in an effort to produce a representative
overview of the daily activities of the group. A scoring
sheet (Figure 2.1.) was used for one-minute sampling
periods of the focal individual and ten-minute group

Group ^ scan: Irt -r 11- S r-
terrestrial 5 stationary £
terrestrial c- movement
- arboreal stationary
arboreal movement
playing 1)
grooming 2 with who £
Interacting K with vho
Sample Scoring Sheet

scans. From this method the minute interval frequencies
(Dunbar, 1976) of the given behavior could be quantified.
Table 2.4 lists and defines the categories of behaviors
that were measured.
- TABLE 2.4. -
Behavioral Category
solitary terrestrial movement This category included
quadrupedal walking, running, pacing, and
(infrequently) bipedal walking,
solitary terrestrial stationary These postures
included sitting or standing in place,
solitary arboreal movement This category included
primarily solitary swinging and climbing. Also
Group D was observed to exhibit brachiation from one
point to another.
solitary arboreal stationary These postures included
sitting or hanging motionless,
play Typically this trifling activity occurred in an
arboreal setting, but the fluid activities would
quickly move between arboreal and terrestrial
realms. These behaviors were primarily seen as
solitary activity except in the infants of Group B
and between the individuals of Group D.
reclining This posture included resting or sleeping in
a recumbent position either on the back, side, or
feeding This category included primary activities of
foraging, eating, wadging, and drinking,
grooming Typically this behavior involved separating
the hair of another individual or self while picking
at the exposed skin. This behavior may also simply
have been manifested in relaxed tactile stimulation,
other Behaviors which do not fit into the categories
above are included here. Primarily these behaviors
included displaying, nest building, sociosexual
interacting, and nonadaptive acting.

Different group demographics across facilities are
also taken into account. Data on infant subjects was not
collected other than in narrative form during the group
scan as the infants influenced the focal subject and
group behavior. Data collection included maintaining a
narrative description of group interactions between
sampling periods on a hand-held tape-recorder which was
later transcribed. This narrative was included to
facilitate understanding behaviors in context. In order
to reduce subject reactivity, each group was observed
from the same site-specific place each day, there was no
observer interaction with the subjects, and observer
movement was minimal.
Statistical Analysis
Multi-factorial analysis of variance (ANOVA) was
used to analyze the variation in activity patterns to
determine if different environments significantly
affected activity levels. The independent variable was
the environment or facility and the dependent variable
was the amount of time in each behavioral category.
Various aspects of each facility clearly influenced
behavioral outcomes. At the Milwaukee facility, Group A,
the enclosure was divided into rooms which restricted
movement over any distance. Dead trees, shelves, and a

large rope hammock suspended from the ceiling allowed for
the option of arboreal placement. Food was presented in
a localized space on the floor. Heavy human traffic
resulted from the location of the restrooms. At San
Diego, Groups B and C, the animals experienced landscape
immersion. The tree types and the distances between
trees reflected a design for a savanna dwelling primate.
Weather was a factor in these observations, specifically
when unforecasted rain occurred and stimulated the group
to pace around the enclosure. The subjects spent a
substantial portion of the day sitting in the dry-moat
around the enclosure. The island was not supplemented
with object furniture to manipulate. Human spectators
were limited to those with scheduled monorail passes and
caregivers. The facility at Columbus, where Group D was
housed, had a cage ceiling made entirely of bars and
allowed for greater arboreal movement. Visual
stimulation required that the apes maintained a position
near the top of the cage to view over the privacy fence.
Human interactions were limited to zoo personnel. A
mister was erected to ensure sufficient access to water
and was used frequently by one individual as a play
stimulant. The environment for Group E, at Cincinnati,
was most austere. The subjects were confined to one room
or the other for the majority of each day. Interaction
with humans was the focus of much behavior because the

caregivers access doors opened up directly into the
preparation kitchen and did not present a visual barrier.
The apes were limited to one main feeding per day and
that food was cleared away when the cage was cleaned.
As data were available from four zoo settings and
15 subjects, for each variable all ANOVA comparisons
involve 3 and 11 degrees of freedom, respectively. For
the analysis of variance, the critical values for F3.11
are .05 = 3.59 and .01 = 6.22. Table 2.5 displays the
ANOVA calculations for each behavioral category measured.
- TABLE 2.5. -
critical values: .05 = 3.59; .01
Fa.11 Value
terrestrial movement ...... 2.867
terrestrial stationary ..... 1.907
arboreal movement ...... 0.460
arboreal stationary ...... 3.018
play . . . . . . . .0.411
reclining ....... 2.918
feeding ........ 1.413
grooming others but not self .... 3.450
all grooming ....... 4.031*
cross-species interaction .... 1.165
species-specific nongrooming interaction. . 8.482**
all nongrooming interactions .... 5.282*
* p<.05, ** p<.01

There was no more difference between groups than within
(than might be seen by chance alone) for arboreal and
terrestrial activities including reclining.
Between-group differences were no greater than the
within-group differences for all the behavioral
categories except all grooming and all nongrooming
interactions. But the differences were highly
significant for species-specific nongrooming contact
interactions. The variables which measured significant
or highly significant are decidedly interrelated. This
overlapping relationship emphasizes the significance.
Thus the occurrence of social interactions were
apparently correlated with differences in environment
more than locomotion was correlated with those same
environmental differences. However, it is not clear if
these data are a consequence of different social or
physical environments. The significant outcome in the
all grooming behavior category might reflect a
consequence of the contrasting number of adult females in
different groups. However, analysis of time budgets,
discussed later, suggests that this is not the case.

Time Budgets
Between feedings, as the food availability
diminished, social interaction also diminished. All
individuals showed an increase in solitary activity as
the expectation of feeding rose. Activity levels
increased shortly after food access, followed by resting
and grooming interactions. As the day proceeded and the
next feeding became imminent, the animals generally spent
more time moving independently about the enclosure space,
scanned the floor as if foraging, and/or monitored for
activity in the direction of the kitchen. Resting and
grooming interactions frequently occurred together.
Harcourt (1977) emphasizes that animals perform their
activities in predictable cycles throughout the day.
This type of data has not been made available by field
observations of bonobos due to the inaccessibility of the
apes throughout the day. This was also not a research
question in the studies cited concerning captive bonobos.
Figures 2.2 through 2.12 show the percent time spent
in each activity by each group member. There were seven
male and eight female focal animals (see Table 2.2.).
Although the adult and adolescent males D, R, K, J, T,
and V spent little or no time in the arboreal milieu,
there was little variation across groups (see Figures 2.2
and 2.3.). Time budget comparisons of terrestrial

activities across groups (see Figures 2.4 and 2.5.)
revealed that these bonobos spent the greater portion of
time on the ground. Examining facilities independently
showed that in Group D, Toby rarely moved off the ground.
Toby had a dwarfish stature and seemed physically limited
by this abnormality. His behavior must have greatly
influence the outcome of the group data. In Groups B and
C Lisa spent a greater portion of her time in terrestrial
movement than the others. This was probably due to the
fact that she spent her time responding to the activity
level of her independently locomoting infant. She
repeatedly pursued Lucy back and forth across the
enclosure. Also, on one day unforecasted rain stimulated
the group to terrestrially locomote around the island for
approximately 20 to 30 minutes. Still more striking was
the time spent in a reclining posture or resting (Figure
2.6.). If you combine the data for reclining and
terrestrial activity for Groups B and C without juvenile
data, it becomes evident why I perceived a greater
difference across groups, where these groups had limited
arboreal opportunities. Groups B and C spent an average
of 65% of their time in these categories, with a range of
75% by Vernon to 50% by Lisa.
The time budget for feeding behavior (Figure 2.7)
and play (Figure 2.8) analysis indicated very little
variation across groups. It is important to consider the

Solitary Arboreal Stationary Behavior Category
San Diego Columbus Cincinn

Solitary Arboreal Movement Behavior Category
San Diego Columbus
Cincinn ati

Solitary Terrestrial Stationary Behavior Category
San Diego Columbus
Cincinn ati

Solitary Terrestrial Movement Behavior Category
30 h
San Diego
Cincinn ati

Reclining Behavior Category
San Diego
Cincinn ati

feeding time budget related to feral behaviors. In the
wild, feeding behaviors are the primary daily activity.
However, across these captive environments the average
daily time spent in feeding behavior was 8% of the
observation period. Lena spent a markedly higher period
of time in the play category which resulted from her
interactions with the other infants in her group.
In the grooming and contact interaction categories,
where the outcomes measured significant differences
across groups, the most obvious variation (Figures 2.9
and 2.10.) appeared in Group E. As mentioned earlier,
Group E individuals had the most restrictive environment,
forced access to the caregivers activity, and the fewest
species-specific interactions. Group D, whose members
were involved in the developing social dynamics of
recently mixed sex grouping, demonstrated consistently
higher species-specific contact interactions. In these
two categories the juvenile male and female were involved
in activities disproportionately with respect to the
other members of their groups. Grooming frequency may be
influenced by the number of adult females in the group.
However, examination of Figure 2.9 suggests that other
variables may effect which individuals participated in
grooming behaviors.

Feeding Behavior Category
17.5 H
12.00 12.00
San Diego Columbus
Cincinn ati

Play Behavior Category

Grooming Behavior Category
San Diego
Columbus Cincinn

FIGURE 2.10.
Species-specific Nongrooming Interaction
Milwaukee San Diego
Cincinn al

FIGURE 2.11.
Grooming Behavior Category Without Juvenile Data
San Diego
Cincinn ati

FIGURE 2.12.
Species-specific Nongrooming Interaction Without Juvenile Data
San Diego
Cincinn ati

Figures 2.11 and 2.12 show the grooming and
species-specific interactions without the juvenile data.
Here it is clearer that Group D was more interactive and
Groups A and D participated in grooming bouts more than
the others.
The all grooming and all nongrooming
interactions were significantly different between
groups. The species-specific nongrooming contact
interactions were highly significant between groups.
However, there were no significant differences in the
other behavioral categories. What might be most
remarkable about these findings was the perception of the
difference prior to statistical analysis. It was
concluded that the effect of the landscape immersion
environment, which was not designed to provide arboreal
opportunities, would be reflected in the statistical
outcomes. These observations suggested that this savanna
environment was not better for an ape who evolved in an
arboreal environment. It appears that the landscape
immersion environment makes little difference either way.
Of greater importance, these data will have value to
those governing bodies at zoological parks who have
invested, or are planning to invest, millions of dollars

on updating traditional style facilities. If the zoo is
only concerned with visitor perceptions, Maple and Finlay
(1986) discuss at length the benefits of a "naturalistic"
habitat. However, if wildlife preservation is the
primary focus of environmental design, the economic
argument might be to simply work with what is there and
focus more on the social grouping of the species, when
appropriate. Markowitz and Spinelli (1986) describe the
move toward landscape immersion as adequate-appearing
environments focused on educational goals only. Yet, if
the animals are not prompted to realize naturalistic
behaviors, the educational goals are limited.
My personal concern is the social and behavioral
well-being of the apes. This state of well-being seems
to be advanced by the opportunity for the apes to choose
interactions with others of their own kind. This study
would then seem to support the movement toward more
"naturalistic" socionomic sex ratios (the economics of
the social group calculated by counting the number of
reproductively mature individuals) in the groups rather
than "naturalistic" habitats.
The landscape immersion of Groups B and C as well as
the fixed environment of Group E cannot explain the
apparent differences in grooming and species-specific
interactions. Groups A and D have no nursery-reared
members. Although there are behaviors distinct to

populations, the implication here may be that social
development is more important than environmental
differences for group behavior. This then is the logical
place to begin to consider culture.

The second goal (perhaps the most important) of this
descriptive study was to observe, record, and present
anecdotal behaviors of the captive bonobo. Research on
captive behaviors of bonobos has been limited to a few
significant studies (de Waal, 1988; Patterson, 1979). As
mentioned earlier, research on wild populations has also
been limited. Therefore, scientific knowledge of this
ape, the least studied great ape, is deficient.
Behaviors, similar and different across these captive
groups and in comparison to wild data, when available,
will be presented. Primarily, this section responds to
the need for empirical information, to expand and enhance
the small amount currently available. The potential for
a nonhuman, nonlanguage culture is considered.
Diet and Food Use
Food, in additional to its nutritional value, plays
a critical role in providing a source of entertainment or
occupation for the captive apes. In the wild, bonobos
spend the greater part of their day feeding and traveling
between food sources (Badrian, Badrian, & Susman, 1981).

However, groups A, B, C, and D had two main feeding
sessions scheduled each day and Group E was provided
food once daily, which was available until the room was
cleaned. Groups A and E received juice and vitamins or
milk, respectively, at other regular times each day. All
five groups received a variety of fruits and vegetables
every day and were provided browse items, which might
include raisins, seeds, nuts, or dry cereal. Unique
between groups was that Group E received much of their
diet in near whole or whole parts and thus could occupy
otherwise vacant time by preparing the food themselves,
and Group D received a much wider variety of foods
including some items which were only available once a
week or seasonally. Group D received coconuts which,
when available, provided a seemingly favorite taste as
well as play items after the pulp had been eaten.
Consistently these captive apes chose to eat the
available fruits first and then eat the greens or
fiberous foods at a leisurely pace, typical of wild
feeding patterns (R. Malenky, 1991, personal
communication). The bonobo1s continuous foraging in the
wild may explain why these groups regulated their food
intake after the initial few minutes of food access
either by simply taking relatively small amounts of food
over long periods of time or by manipulating food items
through play or wadging. Wadging may be described as the

behavior resulting from parts of the meal or other items
not swallowed but maintained behind the lower lip. After
chewing, the individual continues to express the juices
and saliva by pressing the skin, leaves, etc., between
the lower lip and teeth.
Yerkes and Yerkes (1929) noted that the captive
bonobo never used the lower lip as a temporary repository
for food and the lower lip of the bonobo was much less
conspicuous than in the chimpanzee. Patterson (1979)
wrote, concerning his study of a captive bonobo group,
that the storage of food in the lower lip had frequently
been described in Pan troglodytes but was not observed in
Pan paniscus. However, observations of wild bonobos made
by Badrian, Badrian, and Susman (1981) led them to report
that bonobos often rested and chewed wadges made up of
fruit skins, seeds, and leaves as they became satiated
and their feeding pace slackened. Wadging occurred in
all five groups in the present study in different degrees
of frequency and intensity. The two males in Group A,
Lodi and Lomako, prepared several wadges of paper, wood
wool, and food remains. Lodi lined a number of these
wadges up along the edge of the water sink to use
intermittently. Lodi and Lomako were observed to use a
wadge in every instance of water consumption by soaking
the wadge first. They were not observed to drink
directly from the water source.

Individuals in Groups B and C would leave the feed
area, where food items were regularly presented scattered
over an open area of well-trodden grass, with their lips
bulging with fruits and seeds. They would spend time at
their water source mixing fruits and seeds as they drank.
The enclosure island was designed with a shallow pond,
which was kept dry except for a small continuous dribble
of water to serve as the water source.
Although much of the feeding time of wild bonobos
occurs arboreally, food at all these captive settings was
presented in a localized, exposed, open terrestrial area.
All groups performed "lip foraging" except in cases where
food pieces were somewhat larger, then they would pick up
those pieces with their hands. I defined lip foraging as
a posture of either squatting or standing bent at the
waist to pick up food items using the lips only. The
limbs are held near the torso of the body with the hands
behind the back or relaxed at the side of the body. In
Group C, one individual seemed to prefer using her
fingers to rake items into a pile and then gather them up
with her lips. This position had not been reported in
the wild. This self-handicapping may be a residual
behavior from arboreal feeding where the limbs would be
busy suspending the individual. In this posture, with the
face down, the ape would be at a disadvantage if
predation was a concern.

During these observations, status was not observed
to have any effect on individual food access in any
group. Kano and Mulavwa (1984) reported that females in
feral groups, especially multiparious females, advanced
toward food unafraid of males and that males did not
monopolize the food. Consistent with this report, during
my research males, females, and dominant or subordinate
animals were interchangeably given first access to food
at any given time without regard for status. The food
was regularly made available at the same time as the
animals were released either into newly cleaned and
prepared rooms or were released outside. Increases in
activity levels could not clearly be associated with
receiving food. In fact, on some occasions, food was not
even visible when increased social stimulation resulted
in copulation and increased group interaction.
Food sharing has been reported as a behavior unique
to bonobos. Groups A, B, C, and D all demonstrated food
sharing. Each incident involved unsolicited sharing and
was not seen as a result of begging or intense peering
gestures. Badrian and Badrian (1984) and Kuroda (1984)
reported food sharing between wild females only, but
these captive incidents demonstrated food sharing between
adult females, mothers and infants, and males and

In each facility some animals massaged food pieces
on the outer edges of their lips and often left them
lying on the lower lip for brief periods of time.
Connie-Lenore, of Group E, was the most unique in her
solitary activity and these activities often involved
food. Her group showed the fewest social interactions
and their environment was the most fixed. She frequently
engaged in activities such as juggling grapes
hand-to-hand, banana peel mush wadging until her lips
bulged, projecting her lower lip to inspect the banana
mush, playing with fruit pieces in the water, lying on
her back with an arm extended straight above her body and
dropping single grapes the length of her arm into her
open mouth, and dragging unpeeled banana through her
teeth several times squeezing the pulp little by little
into her mouth.
Hediger (1964) reported that one problem for
zoological gardens is the difficulty of providing
suitable occupation for captive animals. In nature they
are constantly preoccupied with avoiding enemies and
seeking out food sources. The apes in this study, as in
most captive environments, had plenty of spare time in
which to create ways to use the environment to entertain
themselves. The rudimentary living quarters had little
or no environmental differentiation and external
stimulation did not change much from day to day. The

enrichment programs had become routine. Hediger pointed
out that the more complex an animal's social
organizational structure is the more it experiences
enforced lack of activity. The bonobo, like all great
apes, is known for its complex social organization. At
each facility efforts are made by the caregivers to
enhance the environment periodically, but these efforts
are time consuming and expensive. Also, the activities
lose novelty to the apes if they are continued with any
Intragroup Relationships/Gommunication
Mutual- and self-grooming took place mainly during
quiet times or states of rest. Grooming, when it
occurred, was acted out in any combination or with any
number of participants and was initiated by any group
member. White (1989) reported that affiliative
interactions between males in wild groups were
infrequent, but in the present study grooming between
males occurred daily in the two groups that had more than
one male. They used their lips to groom much of the
time, a technique that, like the lip foraging posture,
may be derived from an arboreal lifestyle. During

grooming bouts the nose seemed to be favored. This
concern about nasal passages may be a result of the high
potential for respiratory ailments.
Grooming bouts relieved tensions and seemed to
strengthen social relationships. Individuals in Group E
were observed to mutually groom on only three occasions,
the longest session lasting only ten minutes. Animals in
Group B, the largest group, engaged in large mutual
grooming bouts centered around the two oldest females.
The four infants in the group often visually inspected
the interactions of the others.
Grooming behaviors unique to the groups at the San
Diego facility were detailed in a partial ethogram
detailed by de Waal (1988). However, scientists must be
cautious because ethological research can be problematic
in that behaviors may be subject to different
interpretations by different observers. De Waal
described a behavior he called "duck face," which he
reported to be a bonobo communication pattern. This act
was observed during my research but not in communication.
The "duck face" was seen in the context of lip grooming,
particularly when the upper lip was parting hair and the
groomer pulled back in a poised position to survey the
area. De Waal also identified a clapping pattern
associated with the grooming bout. This behavior will be
detailed as evidence of culture later in this text.

If the idle time imposed by captivity provides an
opportunity to increase social pursuits, more complex
social behaviors should be expected in captive
environments than in the wild. Captive observations
provide research opportunities to observe in detail those
gestural and facial signals with particularly high
potential as communication patterns. These patterns
allow participation in communicative exchanges even for a
complete stranger. Having no choice in associations, as
in captivity, may require more highly developed
communication and an ability to predict the behavior of
others. In many instances the gestures of the present
research subjects were directed towards humans when an
animal sought attention. Savage-Rumbaugh (1984) reported
captive bonobos to be extremely affiliative and socially
responsive to human interaction and contact. Most often
these subjects considered humans whom they recognized to
be part of their social system. Each morning, as I
arrived to setup my observation seat, several animals
would respond to me with head bobbing, tapping on the
glass with the back of a wrist or the bottom of a foot,
parallel movement along the front of the enclosure, and
rump presents.
On several occasions caregivers brought fruit or
shared items from their lunch with animals in Groups B
and C by throwing the food onto the island. The apes,

primarily the adult and adolescent females who were
raised in the nursery environment, would move to the
island edge to beg. These begging gestures included
bipedal standing with the arm extended toward the
caregiver and the open palm up, clapping, head bobbing,
lip rubbing, and hand shaking from the wrist (the hand
was either empty or held a twig). The begging gestures
were reinforced by the acquisition of food from the
The male in Group E directed enthusiastic signals
towards caregivers in the kitchen. He often sat at the
kitchen door and wiggled side-to-side on his haunches,
patting or slapping the wall by the door with his hands
and feet and head bobbing. His gestures to humans
included clapping, shaking his hand from the wrist which
was extended toward them (de Waal, 1988, labeled this
behavior "wrist-shaking"), jumping and spinning, putting
his head down and looking into the kitchen upside down
between his legs, or lifting one foot with the leg
extended to wave it. Sometimes he was observed to stand
bipedally with an arm straight above his head and his
hand stretched out while his fingers, held upward, moved
in a bending or flexing bout. De Waals (1988) bonobo
behavioral ethogram defined this behavior as a
species-specific behavior called "finger-flexing." This
behavior was only seen to be performed by Kevin, the male

of Group E, during the present study. De Waal defined
the behavioral context as an invitation. However,
further review of de Waals data revealed that 94.6% of
these observed behaviors were performed by only one
individual in his research, young Kevin. In this study
the behavior was not observed in any other animal and may
be a result of Kevins unique nursery experience during
his social development.
The content or function of a communicative act
cannot be reliably assessed without reference to its
context (Bramblett, 1976). In this research, the head
bob (up-and-down movement of the head) and the head shake
(side-to-side movement of the head) seem to have
particularly high communication potential. The head bob
was not observed in a context which included play and
generally seemed to indicate an invitation to approach.
Primarily this gesture was directed at individuals who
were not immediate family members. The head shake,
however, was often accompanied by a playface and was
included in activities of play or in contexts in which it
seemed to calm a provoked approach. Patterson (1979)
reported vigorous shaking of the head back and forth as a
component of display patterns, such as charging, and by
itself during tense periods. Head shaking was also
observed in his work as a component of play. He

interpreted the meaning of this signal to depend on
accompanying behaviors. These two communicative
behaviors were performed in every group.
It was obvious that the apes adapted their
locomotion behavior to suit the confined spaces. A
"sloth-like" progression has been reported to be used
frequently for suspensory locomotion throughout the first
two years of life by wild Pan troglodytes (van
Lawick-Goodall, 1968). The sloth-like progression is
described as grasping of an overhead support with the
hands and feet and alternately advancing a foot and hand
while releasing only one support at a time. In this
position the individual is suspended under the support.
During Pattersons (1979) study of captive bonobos, this
form of suspensory locomotion was not used by Pan
paniscus. However, this sloth-like pattern of travel was
used extensively in Group D, although primarily by one
female. The frequent use of this locomotion behavior may
be a result of the environment because Group D was housed
in the only facility that provided bars as the ceiling of
the room. This method of locomotion can only be
expressed with a cage ceiling or environment which allows
for arboreal-like travel.

Unique to the environment of Groups B and C the
individuals traveled along well worn trails, moved
bipedally more frequently through the tall grasses, and
walked bipedally for farther distances. These
observations may be of interest to anthropologists
investigating the evolution of bipedalism, particularly
in reference to the human ancestors shift from the forest
to the savanna environment. However, the research
subjects did not have opportunities to brachiate from
tree-to-tree due to the distances between and types of
All the juveniles, adolescents, and adults in every
group regularly performed an "arboreal hold." I define
the arboreal hold as the behavior used when an individual
is in a stationary position, sitting or lying, and holds
onto whatever is available above it with one hand or
sometimes two. This too may be a residual posture from
an arboreal heritage. Groups B and C usually moved to
sit under trees after food gathering. This may have been
an attempt to stay out of the hot sun (although it even
occurred early in the morning when the temperature was
cool and damp), an attempt to be out of the open space,
or an effort to sit under tree limbs to have something
above them to hold onto.

De Waal (1988) described a method of locomotion he
called the "mount walk." He identified this posture when
one individual rests its head, arms, and chest on the
back of another, from behind, while both move forward.
This posture may actually be a result of unnatural peer
groups when several infants are grouped together.
Bonobos do not have seasonal birth cycles and frequently
live in small parties made up of individuals of both
sexes and all age classes (White, 1989). The three
infants of Group B were observed in a play bout in which
Lolita rode on Marylin-Loris back while Laverne tried to
pull her off. In play this train-like movement appeared
to be an incomplete attempt to ride dorsally on a
similarly sized animal. De Waal reported that more than
88% of his observed occurrences were in an exclusively
juvenile group.
State of Arousal
In order to realize the best fit between the usage
of a labeling term and the full range of actually
observed phenomenon, it was clearer to avoid the term
aggression. Arousal, as in the state of being provoked
or excited, seemed to fit better as the description of
the behaviors and postures which individuals exhibited.
When the males were in this state they frequently

displayed penile and pilo erections. Aggression, the
more common term, implies to me an offensive attack. In
the captive behaviors observed in this study, it was not
clear to me that the individuals were hostile. Most
often it was clear that the individuals were not hostile
and were simply manipulating their social environment.
That environment included the people outside the cage, as
well as the other bonobos, and their reactions to the
Patterson (1979) observed aroused encounters between
captive adult and subadult males, but none between
females. Kuroda (1980) also reported observation of
feral groups in which "aggressive" behavior occurred most
frequently among males. However, during my research, a
high state of arousal was observed on a few occasions to
be a result of female agonism. In Group A the adult
female responded on three occasions, when the public was
particularly harassing, by shoving the large trash can
across the room to hit the wall in consecutive charges as
she screamed, or by brachiating out and kicking the front
glass with her feet. In Group B, Lana was startled by
the sprinkler system coming on abruptly and she charged
across the front width of the enclosure dragging a branch
and carrying Laverne dorsally. In Group C the adult
female directed apparently unprovoked arousal towards the
adolescent female who was at least 15 feet away.

Carrying a large piece of tree limb, the adult female
bipedally charged toward the adolescent female, who was
carrying her infant ventrally, and thumped her soundly on
the side of her torso using her fist which held the limb.
On a separate occasion a similar incident unfolded
between the same two individuals but no contact occurred.
On other occasions displays of arousal were directed
at human spectators. Research on wild bonobos by Susman
(1984) reported that bonobos may urinate on an observer
as a scare tactic, in order to initiate the withdrawal of
the observer. In Groups A and D in my study the males
would climb to a higher vantage point and urinate at
human individuals while sometimes repeatedly thrusting
their pelvises. In Groups B, C, D, and E, females were
observed to throw dirt clumps, feces, hay, pieces of
cardboard, or sticks by using an underhand toss with
impressive aim, in the context of arousal toward humans.
One male in Group D would throw food items, feces, and/or
Intragroup excitement was part of the social
dynamics of Group D and a method of testing the newly
mixed sex social group membership. The group received
fresh cut whole branches every day. The branches
provided long term activity and were used as nesting
material or were stripped and cleaned of bark and leaves,
then used as stick weapons by the males. Incidents

occurred between the two males in connection with
conflict over possession of willow branches, attention by
a female, or possession of a ball, but were not observed
as a result of food related conflict, probably because
precautions were taken to feed the males separately.
Tratz and Heck (1954) reported that bonobos fight only by
kicking with their heels, and Kuroda (1980) added that
bonobo aggression is expressed by leaping down or rushing
up to the attacked, chasing, or kicking. In Group D, the
subordinate male initiated aroused interaction by poking,
screaming, stomping his feet, and grinning at the
dominant male. The dominant male then proceeded to
actively check routes of escape and take them if they
were available. Goodall (1986) suggested that captive
animals need "room" for social maneuvering because they
cannot walk away from aggression. Also, de Waal (1989)
reported that nonhuman primates are endowed with
particularly well developed checks on fighting. As
evidenced in my study when an interaction escalated, the
dominant male sometimes chased the subordinate to another
room to minimize the event. The antagonism usually
continued for more than 30 minutes and peaked several
times when the two males abruptly moved to swing from the
ceiling bars and kick at each other, creating a loud
slapping noise from the impact of the bottoms of their
feet as they smacked together. Most often the bout would

conclude when the females, particularly Suzie, joined in
to assist the dominant male. Those instances brought an
abrupt end to the interaction especially when Suzie
chased or slapped at the subordinate male, he screamed
loudly and crouched away from her. The dominant male
would respond by distracting Suzie into a grooming
session. Suzie seemed to play a significantly cohesive
role in the interactions of this group.
The other Groups, A, B, C, and E, exhibited much
less frequent displays of high states of arousal which
included kicking either glass walls, kicking steel access
doors, or display charges. Crowd size outside the
enclosure of Group A seemed to affect the arousal state
of the adult male.
The method used to enhance display charges was
dependent on the environment. The display charges seemed
to be a way to release energy or tension and was not a
bluff threat. In Groups B and C the animals were either
observed to drag or push branches or dry grass clumps in
front of them. In Groups A, D, and E all observed
display charges involved pushing available items in the
enclosure in front of the moving animal. This difference
is probably a result of the more fixed environments of
Groups A, D, and E, versus the highly differentiated
environment of Groups B and C.

In a captive environment, play becomes a very
important part of the artificial life cycle. Play
involves those energetic activity patterns exhibiting
frisky freedom of action, usually with exaggerated body
movements. Frequently play is associated with an
openmouthed, loose-lipped expression and auditory panting
laughter with a distinctive tonal quality. Through play
activities the animals may release sudden bursts of
energy, learn about social relationships, and form
affiliations to individual animals. The two adult
females who experienced social development in a sterile
nursery environment, Louise and Loretta, were not
observed to play, whereas the female who developed in a
wild natal group, Maringa, appeared to hold a strong
central role in play activities with her juvenile male
son. Play most obviously involved the adult males in
association with the infants or juveniles. Every group
exhibited play in behaviors such as swinging, twist
hanging from ropes or branches, kicking, grabbing,
slapping, wrestling, laughing, chasing, tickling, biting,
sliding on wet floors, and self adornment. Mutual play
could involve two or more animals at any time, solicited
by anyone and often revolved around the adult male. In
Group A, when the entire group was involved in play, the

adult male was observed to cup his hands around the
infant, while she was on her mothers back, possibly to
protect her from activities of her juvenile brother when
he played too roughly.
De Waal (1988) described an act of "mouth kissing,"
which he observed most often during play bouts but
classified as oral sexual behavior. This behavior was
observed during my research in Group C when the infant
was playing with the adult male. She was mouthing his
face and rubbing her lips against his. Also, in Group D,
this behavior was observed several times in the context
of play between a male and female. In these instances,
again, the mouthing behavior was a part of playing. De
Waals classification of "kissing" implies something
sexual, more than was observed in this work.
In many instances the genitalia were involved as
play items. When interacting with a black animal in the
dark canopy of the rainforest the focus of animated
interaction reasonably will center around bright pink,
easily visible, body parts. However, speculation has
concentrated on the sexual implications of the bonobos
attention to the genitalia (de Waal, 1987). Primarily
the incidents of play with genitalia occurred between the
juvenile male of Group A towards his mother and the four
individuals of Group D. Interactions included insertion
of a finger into the vagina while the female walked with

an exaggerated bounce, pinching and holding the scrotum,
bouncing-patting the labia, finger flicking the penis,
and others.
Patterson (1979) reported one captive female
observed to perform a "headstand." He described her as
standing quadrupedally with her hind quarters elevated
and her face one or two inches from the ground, while
either peering through her legs or seeming to closely
study the ground. She was not oriented toward other
individuals and no reaction from the others was ever
observed in relation to this behavior. Extensions of
this head balancing behavior seemed to be a favorite play
behavior of all the observed groups in this work. Often
the headstand, which was much more developed in that the
feet were raised above the head, was preceded by a
somersault. Sometimes a support was utilized, as with a
wall or tree trunk. On some occasions the behavior
occurred in solitary play, but other times it was
oriented toward another ape in mutual game playing. The
most impressive and amplified execution of a headstand
occurred when Kevin, in Group E, was very playfully
interacting with children in front of his cage. He
kicked his feet up into a headstand without support
except the triangle formed by his head and hands. This
free-stand behavior with his weight balanced on his head

continued in several attempts lasting approximately 10
minutes. Amazingly he held the balanced stance, in the
center of the room, for several seconds each time.
Patterson (1979) also reported that the most
striking position he observed in the captive group, was
hanging from the feet only. He stated that Pan paniscus
hang from their feet during individual and social play as
well as during other situations. This was also a well
utilized posture during my research and was visible in
every group, to some degree, depending on the
environment. This behavior happened both during solitary
play and mutual activities in all age groups and sexes.
In groups where the environment allowed more arboreal
activity and support of the individuals weight, the
animals demonstrated this posture frequently. Where the
environment was more differentiated, only the lighter
weight individuals performed this behavior.
Patterson described observations of back rubbing
which he associated with drying of the fur. In
circumstances with or without water, Groups A, C, D, and
E included animals who performed this behavior either on
the ground, on a shelf, or against the cage bars. It
often was accompanied by a play face and may actually be
a play behavior.

Sociosexual Behavior
The amount of time spent in sexual activities was
surprisingly small in view of all that has been suggested
about bonobo sociosexual behavior (de Waal, 1987;
Savage-Rumbaugh & Wilkerson, 1978). Bonobos have often
been called the sexiest (sexually active) primate second
only to humans, but during my research entire observation
sessions could go by without any observations of sexual
activity. Additionally, genital stimulation and contact
seems to serve as a social behavior. Only on a few
occasions was I able to ascertain that sexual contact had
the potential for reproduction. These instances were
marked by intromission and ejaculation. Frequently even
intromission was not a part of these sociosexual
behaviors. When sex acts did occur they could happen in
different positions, solicited by any participant, with
various partners, and with or without apparent
reproductive implications.
The socionomic sex ratio, the number of males to
females in the adult and adolescent age classes of a
given group, were as follows: Group A 1:1, Group B -
1:2, Group C 1:2, Group D 2:2, and Group E 1:1. My
classification of socionomic sex ratios includes
adolescent individuals because in captivity both males
and females become sexually mature during this age range

and can contribute to the reproductive status of the
population. These ratios seem typical of wild bonobo
groups (Nishida & Hiraiwa-Hasegawa, 1987). However,
Nishida and Hiraiwa-Hasegawa (1987) report that the
male-to-female sex ratio, when all members are counted,
of wild bonobo groups is 1:1. These captive groups have
been artifically manipulated such that the sex ratios
were as follows: Group A 2:2, Group B 1:6, Group C -
1:3, Group D 2:2, and Group E 1:1.
Some individuals, male and female, who had not yet
produced offspring, performed fetish behaviors toward
inanimate objects, particularly rubber balls. The adult
male housed with a female in lactational amenorrhea also
participated in fetish activities. Stimulation with the
rubber ball sometimes resulted in a penile erection for
the adult male only.
Male-male interactions included "penis presents" in
the context of social stimulation. A penis present is
the posture which consists of squatting with the legs
spread apart so that the erect penis is conspicuously
displayed toward the receiver. The bonobo penis is
relatively larger in proportion to body size than the
other apes or humans (Izor, Walchuk, &. Wilkins, 1981).
The penis present was also used in the context of
solicitation of a female.

The adult male in Group B, Vernon, performed pseudo
copulation (copulation which did not include
intromission) with the adolescent and infant females
ventro-ventrally. The infant females in this group also
attempted genito-genital rubbing. GG rubbing is a
behavior of genital stimulation in which two females hold
each other face-to-face and swing their hips laterally
while keeping the front tips of the vulvae, where the
clitorises protrude, in touch with each other (Kuroda,
1980). Their immature attempts were manifested by one
individual on top, ventro-dorsally, bouncing up and down
or one individual lying face up and the other standing
quadrupedally backed into her and bouncing her genitalia.
GG rubbing was observed only in Groups B and C, where the
adult females were at maximum tumescence. GG rubbing was
observed on 16 occasions.
Kano (1980) reports that GG rubbing is typically
performed in silence. In my study however, on a number
of occasions, the dominant female was heard to emit a
vocalization which escalated into a scream while engaging
in this behavior.
In addition to the estrous state of the
participating individuals, proximity of the male may
influence performance of this behavior. In these groups
where membership included more than one adult female and
GG rubbing was observed, zoo management limited male

representation to one individual. Occasionally the
soliciting female, the subordinate one, visually located
the male prior to mounting the dominant female. If the
male was out of view, the female moved up to the crest of
the nearest knoll to scan until she sighted him. In less
differentiated environments where the male is in view
most of the time, GG rubbing may be inhibited. In one
instance of GG rubbing the subordinate female took a tree
limb from the dominant female and was not challenged.
All other instances occurred without gain by either
female other than the tactile stimulation. The dominant
female was also free to ignore the solicitation and did
in fact make that choice on three additional occasions.
Additionally when the male presented to the female in
Groups C, D, and E, she would most often move to him to
copulate. However, females did, on occasion, ignore male
solicitation presents.
Masturbation, or the sexual stimulation of ones own
genitals, was observed when a female in Groups B, C, D,
and E, most often lying on her back, would rub or pat her
labia either with a foot, an open hand, or the knuckles
of her hand. Both the male and female in Group E and a
female in Group B were observed to manually stimulate
either one or both of their own nipples. The juvenile
male in Group A was observed to manipulate his flaccid
penis on one occasion.

Pathologies and Idiosyncrasies
This category is presented with qualifications.
Labeling behaviors as pathological does not make them so.
This label is based on my perception of these behaviors
as maladaptive. In no way has it been proven (or studied
for that matter) that these behaviors, displacement
activities for the most part, either jeopardize the
reproductive success of those who exhibit them or are a
result of some internal dysfunction. But, as will be
discussed later, these behavior patterns are infrequently
displayed when considering the majority of the population
(Suomi, 1982 ) .
Displacement activities are behaviors which have no
functional relationship to the context except for the
possible release of tension. One response of anthropoids
to mild stress is excessive self-scratching, an act that
has no effect on the animals situation (Bramblett,
1976). All five social groups performed different
degrees of self-scratching, displacement stress yawning,
and fingernail chewing. The stress yawning sometimes
expanded into a motion of pumping the jaw quickly up and
down several times with the teeth exposed or panting with
an open mouth. All members of Group A participated in
plucking hair from each other or themselves. Individuals

in Groups B, C, and E were observed to peel paint from
the moat wallst room walls, or room floor to wadge. The
two adult females in Groups B and C stereotypically
rubbed their upper lip with an index finger and
self-induced vomiting by sticking their fingers or hand
down the throat until they regurgitated their stomach
contents, which they then ate. The adolescent female in
Group C was also observed on one occasion to self-induce
vomiting and reingest it. Group D had one female who
actively moved into a position which enabled her to drink
from another animal's urine stream or to lick urine drops
from a flaccid penis.
The most distinct aberrant behavior was exhibited as
possibly a learned response during social development.
This displacement act may have been a result of nursery
rearing, the trauma of separation from the natural and
natal groups, or behaviors learned from an individuals
mother. Louise, in Group C, was observed to "head hold"
when stressed. The head hold results when the outer side
of her hand was pressed into the side of her head near
the ear so that the wrist, was bent and the hand was
backward over the shoulder. She performed this gesture
4% of the time. The daughter she reared, Connie-Lenore
of Group E, responded to a more rigid environment by

performing this behavior so much so that it seemed almost
continuous but was actually more often than 10X of the
At times of stress, Connie-Lenore would even hold
both sides of her head and sometimes lie in a "catatonic
capsule. The catatonic capsule was a posture only
observed in this individual. This posture is described
as a fetal-like ball with the feet clasped together and
holding both sides of her head. The head-holding bouts in
conjunction with the catatonic capsule posture are not
included in the 10X statistic and so the actual
performance of this gesture was much more frequent. The
male of Group E, who was placed in the social group with
Connie-Lenore and her mother after his nursery
development, also exhibited this one-sided hold behavior,
but only 8X of the time.
Another maladaptive behavior observed exclusively in
Group E is the "anal pump." The anal pump is a behavior
where the individual sits alone and uses one or two
fingers to pump several times into the anus and draw out
feces. This activity occurred in bouts lasting up to 20
minutes or more. Sometimes the feces were wiped on the
shelf or wall. Additionally the female in this group
maintained an open sore on her knuckle by sucking or
chewing on it so that it stayed bloody."

A Suggestion Concerning Culture
Culture is an elusive term and must be defined
before it can be investigated. In anthropological
vocabulary, culture seems to be one of the most difficult
words to define. As a concept, culture may be used to
label different levels of abstraction. I have
participated in many debates with my peers in which we
tried unsuccessfully to agree on a definition of culture.
There seems to be no standard, commonly accepted
definition of culture. So when I went out looking for
evidence of this intangible "thing," I had to start out
by first defining my criteria. In this analysis I
defined cultural behavior as those patterned behaviors
1) are distinct between populations and are shared
by many members of the group,
2) are spread cumulatively across generations,
3) are socially determined or learned, and not
simply a result of social shaping,
4) are used in the day-to-day social communication,
5) are not a result of environmental conditions
specific to one enclosure and not others, and
6) are not behaviors which can be conventionalized
from the "natural" repertoire or are genetically

These criteria would eliminate behaviors that could be
perceived as resulting from conditions where the ability
to perform the behavior is fixed by the environment, are
idiosyncratic or random chance actions, result from the
"natural" tendency or propensity of the species to
manipulate objects, or are behaviors shaped by the
reinforcement of food acquisition.
Prior to this comparative study, previous studies of
the U.S. bonobo population had focused only on social
groups housed at the San Diego Zoo/Wild Animal Park.
Combining these historic works with mine made it possible
to analyze changes over the longer term for individuals
in Groups B, C, and E. Group E are individuals at the
Cincinnati Zoo on loan from the San Diego Zoo.
During the summer of 1971, Patterson (1979)
conducted one of the earliest documented observations of
behavior patterns for a group of captive bonobos. Even
at the time of Pattersons work, he emphasized that the
limited wild population density and distribution of this
rare ape accentuated the need for behavioral study. As
noted earlier, those factors have become much more
critical in the last twenty years. At that time the U.S.
bonobo population consisted of only four individuals.
Pattersons subjects were not consistently housed
together over the duration of his study. The social
groups were altered to facilitate reintroduction of the

youngest infant into the group. The population included
Kakowet, Linda, Linette, and Laura (Table 3.1.). The
data were then used in a comparative analysis with
chimpanzee behavior. Vineberg (1977) conducted her
thesis work shortly thereafter on the same, but expanded
During the winter of 1983 de Waal (1988) conducted a
study of the communicative repertoire of individuals, the
second published in-depth study of this group of bonobos.
Here too, group demographics were changed and manipulated
- TABLE 3.1. -
1971 Population
Name Age Sex Social Development
Kakowet 14yrs. M wild caught
Linda 17yrs. F wild caught
Linette 5yrs. F nursery reared
Laura 3yrs. F nursery reared
over the course of the study. The subjects of this
research were Vernon (on loan from a European Zoo),
Louise, Loretta, Kevin, Kalind, Leslie, Lana, Akili (on
loan from the Yerkes facility), Kakowet II, and
Connie-Lenore (Table 3.2.). Connie-Lenore was the first
mother reared infant at this facility. Again, the data
were used in comparison with chimpanzee behavior. In

both instances the primary focus was on similarities and
differences identified between the two species of Pan.
But of more interest to my research were the temporal
differences identified between the Patterson, Vineberg,
and de Waal studies.
At first glance the memberships of the bonobo
populations over time seem to be completely unrelated and
independent of each other. However, to understand the
route of behavioral transmission between the groups, the
missing Linda, time must be the patriarch filled-in. Kakowet and his mate and matriarch of most of the U.S.
- TABLE 3. 2. -
1983 Population
Name Age Sex Social Development
Vernon 12yrs. M wild caught
Louise llyrs. F nursery reared
Loretta lOyrs. F nursery reared
Kevin 8yrs. M nursery reared
Kalind 7yrs. M nursery reared
Leslie 6yrs. F nursery reared
Lana 5yrs. F nursery reared
Akili 4yrs. M mother reared
Kakowet II 3yrs. M nursery reared
Connie- Lenore 2yrs. F mother reared
population, produced ten offspring (Table 3.3.). Those
infants were raised in the zoo nursery. Until recently,
the nursery allowed social exposure to many different
species at any given time. The zoo nursery care room was
designed as a great room of activity. The infants were

visually, olfactorily, and auditorily stimulated by many
different experiences. They were permitted to have
contact, including play, with other infants in the
nursery. After approximately one year, when they could
eat solid foods and maintain healthy weight levels, the
infants were placed in a moated ape enclosure at the
- TABLE 3.3. -
Offspring of Kakowet (M) and Linda (F)
Name Birth Year Sex Death Year
Linette 1966 F 1972
Laura 1967 F
Lorel 1969 F
Louise 1972 F
Loretta 1974 F
Kevin 1975 M
Kalind 1977 M 1992
Leslie 1978 F 1985
Lana 1979 F
Kakowet II 1980 M
Childrens Zoo where they sometimes had contact with
other young great apes, gorillas and orangutans. At
approximately 4 years of age each infant was reintroduced
to a non-age-stratified (adults and infants), mixed sex
(with only one adult male) social group.
Kakowet died in 1980 and Linda was loaned to the
Yerkes research facility in 1982. However, most of their
offspring had the opportunity to reside and interact for
some length of time in a biologically related social

group and yet, because of their lack of familiarity, they
may not have been aware that the other individuals were
biologically related. Between the years 1971-1983 the
population membership at this facility included Kakowet,
Linda, Linette, Laura, Vernon, Louise, Loretta, Kevin,
Kalind, Leslie, Lana, Akili, Kakowet II, and
Connie-Lenore in various group compositions designed to
maintain harmony, primarily by eliminating male-male
interaction. All these individuals had extended
opportunities to be exposed to and learn from each other.
De Waal (1988) reported an apparent group
idiosyncratic behavior he labeled "clapping." He defines
this behavior as occurring when "two hands, two feet, or
a hand and a foot are brought together one or several
times, often resulting in audible clapping" (de Waal,
1988, p. 222). During his study 85.5% of the clapping
occurred during grooming bouts with the groomee (the
individual on the receiving end of the act) clapping
rarely. De Waals data revealed that only the seven
first-generation bonobos, who were all raised in the
nursery, exhibited this behavior. Those individuals were
Louise, Loretta, Kevin, Kalind, Leslie, Lana, and Kakowet
II. Of the seven first-generation bonobos in de Waals
research, Loretta, Louise, Kevin, and Lana are in my
study. In my work, animals exhibiting the behavior

included Kevin, Lana, Loretta, Lena, Louise, and
Connie-Lenore. Lena and Connie-Lenore were mother-reared
by Loretta and Louise respectively. To date, they are
the only animals observed to have learned this group
This invariant contextual behavior is limited to
populations B, C, and E. In the other zoological
populations, whose apes had not had contact with these
individuals, it was never observed. Additionally, this
behavior has not been reported from wild bonobo group
observations. In fact, in the African apes, clapping has
been reported only for local groups of western lowland
gorillas (Gorilla gorilla gorilla) in southwestern
Central African Republic (Fay, 1989). In those gorilla
groups the hand clap was reported exclusively among
females and young individuals as a gestural communication
indicating nervous temperament or as a danger alert to
the silver back, the sexually mature, usually dominant,
male of a gorilla group. Therefore, clapping in the
context of grooming does not appear to be an item in the
natural repetoire of the bonobo, chimpanzee, or gorilla.
However, captive apes are frequently observed to
clap. In the collection of captive behavioral data, The
Jane Goodall Institute reports ("ChimpanZoo Observers
Guide," 1991) that chimpanzees clap to produce attention
attracting noises in the category of begging behaviors,

most often toward human spectators or caregivers. In my
pursuit of evidence of nonhuman, nonlanguage culture, it
is important to emphasize that this behavior is highly
contextual as it is incorporated in grooming. Clapping
in this context is not used to solicit nor as an
attention-getting signal. Individuals in groups B, C,
and E have also been observed to clap as an
attention-getting behavior directed toward humans. But
these claps are observed to alter the behavior of other
apes in the group typically by diverting their attention
to the human and frequently occur in spurts of more than
one or two claps.
To identify the origin of this grooming-related
behavior, I interviewed nursery caregivers at the San
Diego facility where all the clapping culture
participants were raised. The interviews involved
caregivers who were consistently present during the
crucial years of 1971-1983. By identifying the
individuals who perform the behavior, attempts were made
to trace this behavior back to its origin. This gesture
may have its origin in human models. Yet no instance of
clapping can be recalled between the infant apes and
nursery caregivers. Interestingly, this behavior had not
been observed to be practiced in play. Interaction with
the general public involved activities such as knocking
on the cage glass. Perhaps and most likely, if the

behavior is based on a human model, it may have been zoo
patrons who, while the infants were between the ages of
one to four years, stood around the Childrens Zoo moated
enclosure and interacted unobserved with the
impressionable young apes.
During the grooming bouts an individual, typically
the groomer, would clap once or twice during the
activity. The behavior did not appear to be directed at
anything or anyone but may denote pleasure or
self-gratification in the task. The groomee was not
observed to clap or alter its behavior or position in
response. The clapping was incorporated into the
day-to-day uniquely contextual, gestural communication
during grooming. This behavior was evident in group
interactions which involved as many as all of the group
and was not limited to dyadic interactions. In group E,
where interindividual interaction was infrequent, the
behavior was observed during autogrooming and
A remarkable issue is that the behavior had been
transmitted to other individuals, reproduced across
generations. As mentioned earlier, Connie-Lenore and
Lena were mother-reared by Louise and Loretta,
respectively. Louise and Loretta exhibited the behavior
in 1983, but Connie-Lenore, 2 years old at the time, was
not reported to clap. However, she has since mastered

the gesture. Lena was born in 1985. She too has since
learned to clap in context. The other apes in groups B,
C, or E were either housed in the population after their
infancy or were 3 years old or younger at the time of
data collection and were not observed to clap. So it
seems that this gesture is socially learned during the
first 4 years of life but is not exhibited until early
juvenile ages. The behavior was not identified in either
of the earlier studies (Patterson, 1979; Vineberg, 1977).
Summary and Conclusions
In some instances of captive group behavior, an
individual may discover a new way of doing something,
which is then copied by the others. These group
variations may be described as social transformations as
other members of the group imitate the behavior. When a
behavioral variant of this sort spreads within a group,
when it is passed on from generation to generation
through tradition, and when the group persistently
retains it but only as an acquired behavioral trait, then
we have culture in the ethological sense (Immelmann L
Beer, 1989). Captive populations of bonobos have
produced generations of apes in recent years who may
reside in different environments or be housed in
different social groups. Analysis of the clapping

behavior has demonstrated the potential for a nonhuman
nonlanguage culture. Only through long-term study of a
population will ethologists be able to determine to what
extent an acquired behavioral variant is passed on. As
evidenced in this study, behavioral traits do spread
through a social group and can then be continued in a new
environment or different social group. Although as much
as environment affects the expression and possibly the
origin of behaviors, the social development of
individuals (particularly in the crucial
infant-to-juvenile years) appears to have a far greater
influence on the transmission of social behavior.
Most available studies of captive bonobos have been
conducted on the collections at the San Diego facility.
Many students of bonobo behavior base their perception of
the species on behavioral ethograms derived from these
groups. The historic management approach of this facility
has emphasized propagation as the ultimate goal.
Therefore, for various reasons, many animals were raised
by humans in the social and environmental deprivation of
the nursery. After the nursery experience, these infants
were often housed in exclusively infant/young juvenile
groups. Although they were not in species isolation, it
may be expected that the absence in captivity of that
normal age stratification which exists in nature must
play an essential role in social development. De Waal

(1988) admitted that many of his defined bonobo behaviors
are attributable to idiosyncrasies in the group he
observed. As conscientious scientists we must continue to
conduct behavioral observations across groups within
species. It also seems important to use caution when
implying that characteristics are common to both captive
and feral groups due to the unnatural influences of
When analyzing behavioral differences specific to
groups, potential explanations can be considered. All
the bonobos share the same evolutionary (biological)
history. Most of the U.S. population is closely related
as offspring of Kakowet and Linda. Therefore, it is
difficult to consider cultural behaviors in these groups
without reviewing the genetic relationships. The
clapping behavior demonstrates a potential for culture
because of the strong evidence for social learning.
However, all the apes who participate in this culture are
first or second generation relatives.
A biological explanation may be derived from
population analysis. As explained earlier, this
behavior, in context, is not an item in the natural (wild
population) repetoire. However, further investigation is
required. Analysis of the behavior patterns of
individuals who are genetically unrelated to these
culture participants (as much as possible considering the

limited population available to study) and who
experienced relatively the same social development,
possibly mother reared by a surrogate bonobo mother who
is a culture participant, may shed light on this
behavioral tradition.
Biological preparedness or the inherited propensity
for functioning is important in affecting behavior. Yet,
the expression of genetic variation may also depend on
local conditions. Outside the laboratory in groups in
which mothers rear their infants, where both genes and
environment are shared, we cannot separate the effects of
biology, environment, and social learning.