Chapter #
CAPTIVE ORANGUTANS: FACTORS AFFECTING
MORTALITY, FERTILITY, AND WELL-BEING IN
RELATION TO SUBSPECIES DIFFERENCES
Leif Cocks
School of Environmental Biology, Curtin University of Technology, W.A. 6156 Australia
1.
INTRODUCTION
For most of the 20th century, apart from acknowledging the place of
origin of wild born orangutans, zoos did not generally attempt to separate
orangutans indigenous to different regions. Field biologists, however,
frequently separated Bornean and Sumatran orangutans into separate
subspecies of one species. MacKinnon (1974) first described the physical
differences between Sumatran and Bornean orangutans and a method of
differentiation using hair samples. Seuanez, et al. (1979) characterized the
genetic differences of Sumatran and Bornean orangutans. These and other
studies resulted in most zoos managing Bornean and Sumatran orangutans as
separate populations, but not before a sizeable number of hybrids had been
produced. More recently, it has been proposed that the Bornean and
Sumatran orangutans may be two distinct species, and that the Bornean
species may include between three and five subspecies (Groves 1999;
Rijksen and Meijaard, 1999). Further, Rijksen (1978) described two different
types of Sumatran orangutans based on differences in appearance and
behavior.
1
2
Chapter #
In my research on the factors affecting the well-being and survival of
captive orangutans, I noticed some unexpected statistical results with respect
to Bornean and Sumatran subspecies and subspecific hybrids. My results
provide additional insight into the still-debated question of orangutan
classification and resultant captive management strategies.
2.
MATERIALS AND METHODS
The data that I used for this analysis came from two sources: The
International Orangutan Studbook (Perkins 1994; Table 1) and a survey I
developed and sent to 214 institutions that now have, or once had,
orangutans (see Cocks, this volume, chapter X for more detailed
information).
Table 1. Studbook and survey data used in this analysis
Studbook Data
International Studbook Number
Facility Name
Location
Sire
Dam
Date of Birth
Date of Death (if applicable)
Sex
Subspecies
Wild or Captive Born
Survey Data
Adult Weight
Diet Type
Social Grouping
Enclosure Type
Occurrence of Infertility
Dam’s Rearing Type (human or
mother)
Dam’s Maternal Experience
Individual’s Rearing Type (human or
mother)
Individual’s Maternal Experience
Occurrence of Infant Rejection
Occurrence of Stereotypic Behavior
Occurrence of Sterilization
Subspecies
From the data, I created a series of variables that I related to captive
orangutan mortality, fertility, and well-being. Prognostic indicators have
indicative power and included: overall survival, infant survival (0-5 years),
infant rejection, stereotypic behavior, and maternal death and infertility.
Captive husbandry variables have a predicted relationship to the prognostic
indicator variables and included: sex, population, chronology, adult body
weight, diet type, social environment, enclosure type, occurrence of
sterilization, institution’s region, time from last sibling, dam’s rearing type,
dam’s maternal experience, individual’s rearing type, age of dam at
#. CAPTIVE ORANGUTANS: FACTORS AFFECTING
MORTALITY, FERTILITY, AND WELL-BEING IN RELATION TO
SUBSPECIES DIFFERENCES
3
individual’s birth, numbers of dam’s previous offspring, average interbirth
interval, and individual’s age at birth of first offspring.
I used SPSS (6.0) for Windows to analyze the studbook data. Since I had
censored and uncensored data, I used “Life Tables” for the analysis platform.
I used the Wilcoxon (Gehan) test statistic for survival variables. Categorical
data (nominal or ordinal) were treated in two ways: prognostic variables that
could not be ranked were tested for association using the Pearson Chi-Square
and Likelihood Ratio Analysis (Freeman, 1987). Prognostic variables that
could be ranked were tested for correlation using the Kendall and Spearman
correlation coefficients (Freeman, 1987). I set the significance level at 0.05
for all tests.
3.
RESULTS
The population of wild Bornean orangutans is estimated to be three times
larger than the Sumatran population (Wilkens, 1995). In captivity, the
Sumatran population in the period 1943 to 1994 has been 88% the size of the
Bornean population. In addition, a number of hybrid orangutans have been
produced (Table 2).
Table 2. Percentage distribution of populations
Population
Bornean
Sumatran
Hybrid
Unknown *
Number
736
652
290
297
Percentage
37.3
33.0
14.7
15.0
* These specimens often died before being identified and did not have offspring through
which they could be identified before karyotyping was available. As the orangutans of
unknown subspecies mostly died in infancy, very little data are available on these specimens,
and so I omitted them from my analyses.
I found a survival difference when comparing the three orangutan
populations (p=0.0007, df=2; Figure 1). Statistically, there was no difference
in the survival of the Bornean (n=729) and Sumatran (n=606) subspecies
(p=0.7385, df=1), but hybrids (n=289) had a lower survival rate when
compared to either Bornean (p=0.0002, df=1) or Sumatran orangutans
(p=0.0017, df=1).
4
Chapter #
Figure 1. The effect of population on survival
I found similar results in my consideration of infant survival rates (Figure
2). There was a significant difference in overall infant survival for the three
populations (p=0.0079, df=2). The survival of Bornean (n=196) and
Sumatran (n=181) infant orangutans did not differ (p=0.3633, df=1), but
infant survival in hybrids (n=108) was significantly lower when compared to
Bornean (p=0.0024, df=1) and Sumatran (p=0.0199, df=1) subspecies.
1.2
1.0
.8
Cum Survival
.6
.4
Subspecies
Hybrid
.2
Sumatran
Bornean
0.0
-1
0
1
2
3
4
5
Years Alive
Figure 2. Infant survival in three sample populations
6
#. CAPTIVE ORANGUTANS: FACTORS AFFECTING
MORTALITY, FERTILITY, AND WELL-BEING IN RELATION TO
SUBSPECIES DIFFERENCES
5
At 13%, the incidence of infertility was three times greater in the
Bornean subspecies than in the Sumatran subspecies (4%; p=0.02209, df=1)
or hybrid population (4%; p=0.02318, df=1; Figure 3).
140
120
100
108
100
80
60
54
Count
40
Infertility
20
0
no
15
Bornean
yes
Sumatran
Hybrid
Subspecies
Figure 3. The effect of population on infertility
In considering rates of maternal death, I found that the Bornean
subspecies had an occurrence of 25% compared to 23% for Sumatran and
15% for hybrid populations (Figure 4). Hybrids appear to have a lower rate
of maternal death, but the differences between hybrid and Bornean and
hybrid and Sumatran populations are not statistically significant (Pearson
Chi Square p=0.32239, df=2; Likelihood Ratio Analysis p=0.30069, df=1).
6
Chapter #
100
69
80
64
60
40
29
Count
26
20
19
Survived > 30 days
0
Bornean
Female Survival
5
Hybrid
Sumatran
Survived  30 days
Subspecies
Figure 4. The effect of population on maternal death
At 80%, hybrid female orangutans have a higher percentage of infant
rejection than do either Bornean (29%; p=0.01018, df=1) or Sumatran (36%;
p=0.01077, df=1) female orangutans (Figure 5). Bornean and Sumatran
female orangutans did not differ in their infant rejection rates (p=0.4620,
df=1).
60
50
14
16
40
30
34
28
20
Infant Rejection
10
8
0
Infant Rejection
No Rejection
Bornean
Sumatran
Hybrid
Subspecies
Figure 5. The effect of population on infant rejection
The occurrence of stereotyped behaviors was nearly evenly distributed
across the three orangutan populations (13.9% in Bornean, 8.6% in
#. CAPTIVE ORANGUTANS: FACTORS AFFECTING
MORTALITY, FERTILITY, AND WELL-BEING IN RELATION TO
SUBSPECIES DIFFERENCES
7
Sumatran, and 10.7% in hybrids; Figure 6). My pairwise comparisons of
each population are not statistically significant (Pearson Chi Ssquare
p=0.58762, df=3; Likelihood Ratio Analysis p=0.35126, df=3).
200
138
123
100
50
Stereotypic Behavior
Absent
Count
20
0
Bornean
13
Sumatran
Present
Hybrid
Subspecies
Figure 6. The effect of population on stereotypic behavior
4.
DISCUSSION
There are at least two recognized orangutan subspecies: Bornean and
Sumatran. In the wild, these two subspecies appear to have different
behaviors. For example, Rijksen (1978) attributes the Sumatran orangutan’s
heightened sociability to the more clumped distribution of its food resources
in the Sumatran rainforest. I hypothesized that behavioral, genetic, and other
differences of the two subspecies might influence the health and longevity of
captive individuals. Further, subspecific hybrids were frequently produced in
captivity, and I posited that this hybridization might have an effect on the
longevity and health of hybrids.
Between 1946 and 1994, 32% of captive orangutans were Bornean and
33% were Sumatran. In the wild, the numbers of Bornean orangutans are
estimated to be approximately three times greater than that of Sumatran
orangutans (Wilkens, 1995). It appears from the captive numbers that zoo
personnel prefer to keep the more endangered subspecies or that more
specimens have been imported from Sumatra.
8
Chapter #
Hybrid orangutans, all of which are captive bred, represent 15% of the
captive population. Most zoo staff now prevent these animals from
reproducing, so they have no direct role in the species’ conservation. From
this perspective, they occupy captive positions that could arguably be better
filled by animals of the “pure” subspecies.
I found little difference in survival between the two subspecies, but
hybrids had a significantly lower survival rate than either subspecies. I found
no evidence of a difference in infant mortality between the two subspecies.
However, hybrid infants have a significantly higher infant mortality than
either of the two subspecies.
My results could occur either because zoo personnel selectively neglect
hybrid animals or because hybrids are inherently weaker than are the two
subspecies. As early as 1982, de Boer and Seuanez warned of the possible
negative consequences of orangutan hybridization.
To determine which argument better explains my results, I repeated
analyses for species survival using orangutan data only from the North
American region, where zoos operate under the authority of the Orangutan
Species Survival Plan (SSP). By SSP mandate, hybrid orangutans have never
been kept off-exhibit or, with the exception of prohibitions on breeding,
otherwise managed differently compared to Bornean or Sumatran orangutans
(Perkins, 1995). Most likely due to pressure from animal rights groups,
North American zoo personnel have paid special attention to providing equal
treatment of hybrid and “pure” orangutans.
I found no difference in the survival of Bornean or Sumatran orangutans
in North America, but again the survival of hybrid orangutans was
significantly below that of the two subspecies (Cocks, 1999). I suspected that
there might be a relationship between the reduced survival of hybrids and the
geographic location of the zoo, as zoos in developing regions have been less
strict about preventing of hybridization of Bornean and Sumatran
orangutans. Zoos in developed regions have also sometimes sent unwanted
hybrids to zoos in developing regions. My investigation shows that only
34% of hybrid orangutans are located in zoos developing countries (Table
3). Hybrids in both developed and developing regions show reduced
survival, so there is scant indication of a link between the zoo’s geographic
location and hybrid survival rates (Cocks, 1999). Thus, my data suggest that
the lower survival rate of hybrid orangutans is due to reduced fitness
associated with hybridization.
Table 3. Distribution of orangutans in zoos in developed and developing regions of the world
Bornean
Sumatran
Hybrid
Developed Region
59% (n=434)
76% (n=480)
66% (n=191)
Developing Region
41% (n=302)
24% (n=152)
34% (n=99)
#. CAPTIVE ORANGUTANS: FACTORS AFFECTING
MORTALITY, FERTILITY, AND WELL-BEING IN RELATION TO
SUBSPECIES DIFFERENCES
9
The taxonomy of orangutans remains unresolved, yet how the various
populations are classified has profound implications for conservation plans
and for captive management. Genetic (Chemnick and Ryder, 1994),
dermatoglyphic (Boedtani and Smits, 1994), and other (Groves, 1999)
evidence support recognition of two distinct species (Bornean and
Sumatran), but other taxonomic viewpoints exist. Andrew (1999), for
example, advocates a monotypic orangutan species. Muir’s (1998 and this
volume) DNA analyses reveal as much difference between the various
Bornean populations as exists between Bornean and Sumatran orangutans.
He maintains that there were ten occasions over the past million years or so
when land bridges connected Sumatra and Borneo, therefore allowing the
transfer of genetic material between some populations and preventing the
simple division of orangutans into two species based on island of origin.
If Muir’s view is correct, then at least some of the Bornean orangutans
should be managed as genetically separate populations to avoid the negative
effects of hybridization. It is interesting to note that the only significant
difference that I observed in this study is the higher infertility of Bornean
orangutans when compared to both Sumatran and hybrid individuals.
Infertile offspring is one common result of crossbreeding incompatible
populations. Infertile Bornean orangutans could also possibly result from the
captive breeding of distinct populations of Bornean orangutans. This would
support Groves’ (1999) recommendation that the Sumatran and Bornean
orangutans should be separate species, with the Bornean species containing
three subspecies.
In addition to having reduced survival rates, I found that hybrids have a
significantly higher frequency of infant rejection (80%) than do Bornean
(29%) and Sumatran (36%) orangutans. As suggested by their shorter life
spans, hybrids may have a host of mental and physical ailments that
negatively impact on females’ abilities to raise their infants. This high
occurrence of maternal rejection could also be due to less attention by
caretakers placed on ensuring hybrid reproductive success.
Many “pure” orangutans were wild-born (32%), but all hybrid orangutans
are captive-born. Arguably, because hybrid animals are the product of the
captive environment, they might, as a group, display more of the problems
caused by captivity, including an inability to raise infants. I found, however,
no significant change in infant rejection over time, and most wild-born
animals enter captivity as partially hand-reared infants1. These facts appear
to point to higher infant rejection as a result of hybridization. I found no
1As
adults, hand-reared animals are more likely to reject their offspring than are motherreared animals.
10
Chapter #
significant difference in the occurrence of stereotypic behavior for the three
populations used in my analysis. This suggests that higher hybrid infant
rejection was more closely associated with physical rather than mental
weakness, although reduced viability apparently does not increase the
occurrence of maternal death in hybrids.
Bornean orangutans have a more solitary nature than do Sumatran
orangutans (MacKinnon, 1974). In captivity, this may mean that Bornean
orangutans suffer from higher levels of social stress more easily than their
Sumatran and hybrid counterparts. Severe social stress has been shown to
produce infertility in primates (Wildt, 1996). While I did find a higher
occurrence of infertility in Bornean orangutans, I found no difference in the
incidence of stereotyped behaviors for the three populations. This supports
my argument that higher infertility is Bornean orangutans is a consequence
of captive breeding between distinct populations.
5.
CONCLUSION
My data indicate that hybrid orangutans are genetically weaker and have
much lower survival rates than do “pure” animals. This weakness may also
contribute to higher rates of infant rejection among hybrids.
Bornean orangutans have more than three times the occurrence of
infertility than either Sumatran or hybrid orangutans. My results suggest this
is due to hybrid infertility between genetically distinct Bornean populations.
The data from my investigation support the position that Bornean and
Sumatran orangutans are separate species and that there are likely to be
further subspecies within the Bornean population.
REFERENCES
Andrew, P. (1999). Captive goals and taxon characteristics. Paper presented at XXIII
Australasian Primate Conference, Taronga Zoo.
Boedtani, A. N. and Smits, W. (1994). Dermatoglyphics of the Bornean and Sumatran
orangutan. Paper presented at IPS Congress at IPS Congress 15, Kuta, Bali, Indonesia.
Chemnick, L. and Ryder, O. (1994). Cytological and molecular divergence of orangutan
subspecies. In: Proceedings of the International Conference on Orangutans: The neglected
ape. (Eds.) Ogden, J.J., Perkins, L.A. Sheeran, L. Zoological Society of San Diego,San
Diego. Pp. 74-78.
Cocks, L., this volume.
Cocks, L. R. (1999). Investigation of the factors affecting the well-being and survival of
orangutans (Pongo pygmaeus) in captivity. Unpublished Masters thesis, Curtin University,
Western Australia.
#. CAPTIVE ORANGUTANS: FACTORS AFFECTING
MORTALITY, FERTILITY, AND WELL-BEING IN RELATION TO
SUBSPECIES DIFFERENCES
11
de Boer, L.E.M. and Seuanez, H. N. (1982). The chromosomes of the orangutan and their
relevance to the conservation of the species. In: The orangutan: Its biology and
conservation, (Ed.) de Boer, L.E.M., Dr. Junk Publishers, London. Pp. 135-170.
Freeman, D. H. (1987). Applied categorical data analysis. Marcel Dekker, New York.
Groves, C. P. (1999). The question of species and subspecies in orangutans. Paper presented
at XXIII Australasian Primate Conference, Taronga Zoo.
MacKinnon, J .R. (1974). The behaviour and ecology of wild orangutans. Animal
Behaviour, 22: 3-74.
Muir, C. C. (1998). The descendants of our ancestors: Investigating the population structure
of the orangutan (Pongo pygmaeus) using DNA sequence and paleomigration modeling.
Unpublished PhD. Thesis, Simon Fraser University, Vancouver.
Muir, C.C., Galdikas, B.M.F., and Beckenback, B.T., this volume.
Perkins, L.A. (1994). International studbook for the orangutan. Atlanta/Fulton County Zoo,
Inc.
Perkins, L.A. (1995). Zoo news on captive orangutan management. Pongo Quest, 7(1): 14-16.
Rijksen, H. D. (1978). A field study of Sumatran orangutans (Pongo pygmaeus abelii). In:
Ecology, behaviour and conservation. (Eds.) Veenman, H., and Zoren, U., Wageningen.
Pp.
Rijksen, H. D. and Meijaard, E. (1999). Our vanishing relative: The status of wild orang-utans
at the close of the twentieth century. Kluwer Academic Publishers, London. Pp. 32 -49.
Seuanez, H., Evans, H.J., Martin, D. E., (1979). An inversion in chromosome two that
distinguishes between Bornean and Sumatran orangutans. Cynogenetics and Cell Genetics
23: 137-140.
Wildt, D. E. (1996). Male reproduction: Assessment, management and control of fertility. In:
Wild mammals in captivity. (Eds.) Harris, H., Kleiman, D.G., Allen, M.E., Thompson,
K.V., Lumpkin, S., The University of Chicago Press, Chicago. Pp. 429-450.
Wilkens, J. (1995). Captive species management plan for orangutan Pongo pygmaeus.
Australian Species Management Program, ARAZPA, Sydney.