Content uploaded by Marcus Rowcliffe
Author content
All content in this area was uploaded by Marcus Rowcliffe
Content may be subject to copyright.
237
© 2009 Universities Federation for Animal Welfare
The Old School, Brewhouse Hill, Wheathampstead,
Hertfordshire AL4 8AN, UK
Animal Welfare 2009, 18: 237-247
ISSN 0962-7286
Fecundity and population viability in female zoo elephants: problems and
possible solutions
R Clubb†, M Rowcliffe‡, P Lee§#, KU Mar¶, C Moss#and GJ Mason*¥
†Wildlife Department, RSPCA, Wilberforce Way, Southwater, West Sussex RH13 9RS, UK
‡Institute of Zoology, Zoological Society of London, London NW1 4RY, UK
§Department of Psychology, University of Stirling, Stirling FK9 4LA, UK
#Amboseli Trust for Elephants, PO Box 15135, 00509 Langata, Nairobi, Kenya
¶Department of Animal and Plant Sciences, University of Sheffield, Sheffield S10 2TN, UK
¥Animal Sciences Dept, University of Guelph, Guelph N1G 2M7, Canada
* Contact for correspondence and requests for reprints: gmason@uoguelph.ca
Abstract
We previously reported that African (Loxodonta africana) and Asian (Elephas maximus) female elephants in European zoos
have shorter adult lifespans than protected conspecifics in range countries. This effect was the cause of greatest concern in Asian
elephants, and risk factors within this species included being zoo-born, transferred between zoos, and possibly removed early from
the mother. Here, we investigate these risk factors further; assess fecundity and sustainability in European zoos; and propose
testable hypotheses as to the causes of these animals’ problems. Although imported wild-born Asian elephants live longer than
zoo-born conspecifics, being imported when juvenile or adult appears no more protective than being imported in infancy,
suggesting that the benefits of being wild- rather than zoo-born are conferred early in life. Zoo-born Asian neonates are signifi-
cantly heavier than those born to working animals in range countries, with a possible tendency to be fatter. In zoos, African
elephants have tended to be removed from their mothers at older ages than young Asians, and were also transferred between
zoos significantly less often: factors that could possibly underlie this species’ lower calf losses and improving adult survivorship in
Europe. Both species have low fecundity in European zoos compared to in situ populations, and are not self-sustaining, declining
at approximately 10% per annum if reliant on captive-bred females under historically prevailing conditions. Data from other
species suggest that stress and/or obesity are parsimonious explanations for the suite of problems seen. We recommend specific
screens for testing these hypotheses, and for potentially identifying vulnerable individuals within the extant zoo populations.
Keywords:African elephant, animal welfare, Asian elephant, fecundity, population viability, survivorship
Introduction
Zoo elephants experience a range of problems suggestive
of compromised welfare, such as elevated mortality and
reproductive failure (eg Kurt & Mar 1996; Taylor &
Poole 1998; Clubb & Mason 2002; Clubb et al 2008).
Here, we use data analysis to explore the reasons for this.
There are at least 2,114 elephants worldwide, currently
living in zoos, safari parks, sanctuaries, circuses and with
private keepers (Koehl 2008). European zoos and
circuses house approximately 300 African elephants
(EEG 2008) while North American zoos house approxi-
mately 220 (Olson 2008). In contrast, about 500,000
African elephants (Loxodonta africana) live in the wild;
these recently came off the IUCN ‘Endangered’ list
(AfESG 2004), having shown an annual increase of 4%
since 2002 (Blanc et al 2007). European zoos and
circuses house approximately 450 (EEG 2008), and
North American zoos, approximately 260 (Keele et al
2007) Asian elephants (Elephas maximus). This species
is still ‘Endangered’ (AsESG 1996), with around
60,000 living in range countries (Sukumar 2006; Hedges
2007). In North American zoos, neither species is self-
sustaining, requiring importation from range countries in
order to prevent population decline (Hutchins & Keele
2006; Wiese & Willis 2006): a practice criticised by the
IUCN and others (Waithaka et al 1998; AfESG 2003;
Hedges et al 2006). Despite considerable financial
outlay — elephants’ ex situ captive costs are an estimated
ten times higher per capita per year than in situ conserva-
tion costs (Balmford et al 1995, 1996) — zoo elephants
have apparently high mortality and low breeding rates.
This has led to concerns over their health and welfare,
with long-reported contributory problems (eg Taylor &
Poole 1998; Richman et al 1999; Brown et al 2004)
including Herpes virus, infertility and infanticide.
Recently, we have shown that adult females of both
species also have shorter lives in European zoos than in
protected in situ populations (Clubb et al 2008).
Universities Federation for Animal Welfare Science in the Service of Animal Welfare
238 Clubb et al
This effect was the cause of greatest concern in Asian
elephants: unlike Africans, overall lifespan in this species
did not appear to have improved in recent years, and infant
mortality was high. Risk factors included being born in a
zoo rather than in the wild, being moved between zoos, and
possibly being removed from the mother at a young age.
These findings are presented briefly elsewhere (Clubb et al
2008). Here, we take the opportunity to provide specifics
on the datasets used, expand on these findings, and discuss
the potential causes of — and, hence, potential solutions
to — these problems. First, we investigate the benefits to
Asian elephants of being wild-born by: i) investigating
whether a longer infancy spent in the wild prior to impor-
tation is associated with a longer lifespan, and ii) testing
the hypothesis that zoo-born infants are abnormally heavy
(cf Kurt & Mar 1996; Dale 2009), since large birth weights
can have lasting adverse effects on health (see Discussion).
Second, we investigate whether species differences in: i)
age at separation from the mother, and ii) transfers between
zoos, potentially explain the more worrying mortality
patterns of Asian elephants. Third, we assess fecundity and
long-term equilibrium population growth, to determine
whether zoo populations are self-sustaining.
Materials and methods
Datasets for elephants in zoos and reference populations
Zoo data came from European Endangered Species
Programme ‘studbooks’. These were accurate to 15 July
2005 (African elephants) and 12 June 2005 (Asian
elephants). Data were checked for internal consistency
using ‘Clean Up’ and ‘Data Validation’ functions in
SPARKS (Single Population Analysis and Records Keeping
System, Version 1 52); all errors found were corrected.
Supplemental information was added from the European
Elephant Group report (EEG 2002); this contained data for
additional elephants (108 African, 96 Asian) plus some key
missing dates. The African studbook contained few errors,
and was largely corroborated by EEG data (EEG 2002). The
Asian studbook initially contained numerous errors (eg
missing birth types, dams listed in different locations to
their calves at birth, etc), but most could be corrected via
internal checks and comparison with EEG data.
Zoo records cover many decades, but we used only data
from animals living in the population from 1st January
1960. Birth-dates for wild-born females were estimated to
the nearest year (cf Wiese 2000), and those imported when
under one-year old were assigned the maximum age at
import (one year) to avoid estimated lifespan of zero. We
excluded the few African forest elephants (Loxodonta
africana cyclotis), plus any individual with unknown
ages/dates of entry to (or exit from) the zoo population, or
ambiguous birth origins (eg unknown, or recorded as
‘timber camp’). This resulted in a dataset comprising infor-
mation on import/birth and death dates (where applicable)
for 1,055 elephants (402 African, 653 Asian). These came
from 236 zoos across Europe, Israel and the former Soviet
Union (which we call ‘European’ for brevity). Our analyses
focus on females (n = 786: 302 African, 484 Asian), since
the small population sizes for males and their relative
recency (few wild-caught elephants are male) means that
too few animals have died in each age class to allow valid
survivorship analyses (cf Wiese 2000).
To act as reference populations, we used two well-docu-
mented in situ populations judged to yield demographic
benchmarks that zoos should reasonably meet or exceed
(Clubb et al 2008). For African elephants, we used a popu-
lation continuously monitored in and around Amboseli
National Park, Kenya, from 1972 to the present (see Moss
2001). Data came from 2,173 individually-recognised
elephants of both sexes. Analyses here were based on births
up to the end of 2004, for a total of 1,093 females. Ages
were known with a maximum error ± 6 months for most
females under the age of 35 (n = 799). Ages were estimated
either to the nearest year or ± 2.5 years for females aged
between 35 and 45 (n = 125), and estimated to a maximum
of ± 5 years for animals over 45 years (n = 179). Animals
whose ages were estimated at first sighting were reassessed
at death (from tooth ages), or by changes in size and shape
during maturation compared to known-age reference
animals (Moss 2001). Human-caused deaths (for details, see
Clubb et al 2008) were treated as right-censored in
survivorship analyses to create ‘Natural mortality’ datasets
conforming more closely to fully protected populations.
Compared to similar protected in situ populations, our
African reference population seems representative, even
conservative. The Amboseli population has a long-term
average growth of 3.75% (Moss 2001), and rates in other
African reserves are similar or greater, eg 4.6% per annum
in Samburu, Kenya (Wittemeyer et al 2005), 6% in Addo,
South Africa (Whitehouse & Kerley 2002), and over 8%
elsewhere in South African (Slotow et al 2005; van Aarde &
Jackson 2007). Furthermore, calf mortality in other reserves
is similar to or lower than Amboseli, eg just 10.5% over
five years in Samburu, Kenya (Wittemeyer et al 2005)
compared to 10% in the first year for Amboseli (Clubb et al
2008; Moss & Lee, in press). Finally, deaths caused by
humans were almost certainly random with respect to
elephant quality (ie weak animals were not targeted): thus
treating them as right-censored should not increase the
apparent fitness of this population.
For Asian elephants, our reference was captive elephants
working for the Government-run Myanma Timber
Enterprise ([MTE]; Mar & Win 1997; Sukumar 2003; Mar
2007; Saragusty et al 2008). MTE studbooks document
animals’ histories to inform company veterinarians, forest
officials and mahouts, recording over 8,000 elephants living
in approximately 260 timber camps throughout Burma’s
forested regions since the early 20th century. For analysis,
we excluded males and animals with unknown ages, sexes,
birth origins or dates of entry to (or exit from) the popula-
tion, and records with obvious errors (eg birth dates later
than death dates). At the time of this analysis, valid data
were available on 5,213 animals of both sexes, 2,905 of
them female. The earliest useable captive birth record was
in 1925, the earliest useable wild-capture record was in
© 2009 Universities Federation for Animal Welfare
Viability of zoo elephants 239
1951, and the last year of follow-up was 2000. It is worth
noting that the data have been subject to a long and contin-
uing process of checking and validation over which the
number of useable records has fluctuated to a small degree;
the total number of individuals used here may therefore
differ slightly from that reported in earlier or subsequent
publications based on the same data set. For captive-born
animals, ages were known precisely. For wild-born animals
(largely from before a capture-ban in 1994), age at capture
was estimated by experienced elephant handlers, based on
shoulder height and other indicators, such as skin character-
istics (Kurt & Kumarasinghe 1998; Sukumar 2003) and
folding patterns of the dorsal aspect of the ear flap
(Goswami et al 2007). The error in these estimates is likely
to be within one year for young animals (under 20), which
form the majority (68%) of those captured from the wild
(n = 1,344). Again, human-caused deaths (see Clubb et al
2008) were treated as right-censored in analyses to create
‘Natural mortality’ datasets.
Just as for the Amboseli population, human-caused deaths
were almost certainly random with respect to elephant
age/health — thus, again, treating them as right-censored
should not bias increase this population’s apparent fitness.
While there have been suggestions that MTE records under-
report infant mortality, this was not a serious problem for our
analyses (see also Saragusty et al 2008). For MTE infant
mortality to be as high as that seen in European zoos (Clubb
et al 2008), some 161 deaths of live-born female calves (thus
approximately 320 calves in total, assuming similar male
losses) would need to be unreported: this would represent
approximately 20% of overall calf production, and experi-
ence of MTE recording procedures suggests this is implau-
sible. Furthermore, the MTE population is potentially
self-sustaining (based on modelling equilibrium growth rate
using the demographic rates of captive-born elephants; Mar
2007), and comparisons with other populations further
validate this dataset. Other, well-managed, self-sustaining
captive populations in southern India and Sri Lanka
resemble MTE (see Mar 2007; Clubb et al 2008) in having
infant mortality rates under 15%, median complete female
lifespan of approximately 45 years, and annual fecundities of
over 0.05 (Sukumar et al 1997; Taylor & Poole 1998). We
used captive working animals, rather than wild individuals
as our reference because good, valid data were readily
available. However, conditions for working elephants in
range countries are not necessarily ideal (eg Lair 1997;
Hedges et al 2006, Ramanthan & Mallapur 2008; Saragusty
et al 2008) particularly if wild-caught (see Mar 2007; Clubb
et al 2008). The MTE population thus probably underesti-
mates how well Asian elephants fare if optimally protected;
for instance, data from wild Asian elephants suggest even
better calf survivorship, with just 5% infant mortality per
annum between 0 and 5 years (Sukumar 1989).
Age at import
We focused on wild-born, zoo Asian elephants because ‘birth
origin’ affected this species (see Introduction), and because
sample sizes were too small for African elephants. The age at
which each elephant entered captivity (‘Import age’) was included
as covariate in an adult survivorship analysis (see below).
Birth weight
Birth weight data for zoo and in situ elephants were gleaned
from a data and literature search. Again, this focused on
Asian elephants due to their birth origin effect plus the
paucity of data for African elephants. Information came
from European zoo studbooks, along with Asdell’s Patterns
of Mammalian Reproduction (Hayssen et al 1993), Kurt and
Mar (1996), and the International Species Inventory System
accessed via the Elephant Care International website
(http://www.elephantcare.org). Where possible, data were
used to calculate two estimates of body fat: Body Mass
Index (BMI: weight/frame size2) and Ponderal Index (PI:
weight/frame size3).
Age at separation from mother
The age at which zoo-born, female calves were separated
from their mothers, defined as calf transfer to another zoo,
was extracted from studbooks. One calf was separated by
the dam moving zoo; this data point was excluded.
Transfer rates
For each individual, the number of times she was trans-
ferred between zoos was totalled, excluding initial importa-
tion for wild-born animals, as was the total number of years
she spent in captive facilities over her entire life. The ratio
of these two figures was our index of transfer rate.
Fecundity and population growth rates
Age-specific fecundity, the number of live-born, female
offspring born, Bx, per female alive, fx, between the ages of
xand x+ 1: Bx/fx(Caughley 1977), was calculated for zoo
and reference populations. Results were plotted for visual
comparison. The future population viability (long-term
equilibrium growth rate) of zoo elephants was also
modelled to identify whether European populations are
expected to be self-sustaining. Parameter values were taken
from fecundities calculated here, and previous survivorship
analyses (Clubb et al 2008). We used captive-born animals’
characteristics where they differed from wild-born, to
model the situation without further importation to zoos.
We then re-ran all matrices using the demographic param-
eters of our reference populations in order to model better
possible future scenarios for European zoo elephants and
identify the respective benefits of improved fecundity and
improved survivorship. Using their age-specific fecundi-
ties represents scenarios in which all zoo females have
ready access to males, high conception rates, and low
stillbirth rates. Using the ‘natural mortality’ survivorship
of our reference populations (Clubb et al 2008), instead
provides scenarios in which zoo elephant mortality rates
are reduced to the low levels experienced in the wild, thus
assessing whether improving survival alone would render
zoo populations self-sustaining.
Statistical analyses
All statistical tests are two-tailed, with an alpha of 0.05. Cox
Proportional Hazards Regressions were used to investigate
the effects of import age on elephant survivorship since
these can assess the impact of both categorical and contin-
Animal Welfare 2009, 18: 237-247
240 Clubb et al
uous independent variables on age-specific survivorship (cf,
Raggi et al 2004; Del Giudice et al 2006; Anthony et al
2007; Clubb et al 2008). Animals were included in survival
analyses from the point they entered the population (eg
through importation, capture or birth) and an ‘event’ (eg
death from natural causes, or censorship through loss to
follow-up or being alive at the end of the dataset).
Premature and stillborn calves were excluded. Survivorship
analyses were also used to explore species differences in the
age at separation from the dam, this being treated as an
‘event’. The proportionality assumption of the Cox model
was always tested using scaled Schoenfeld residuals
(Grambsch & Therneau 1994); all reported results met this
assumption. Analyses were all run using ‘R’ (version 1 9 1).
Body Mass and Ponderal Indices for zoo elephants and
in situ conspecifics were compared where possible using
Mann-Whitney U- or t-tests. Transfer rates for African and
Asian elephants were compared using a t-test (Minitab v 13).
Population growth rate analyses used a female-only Leslie
matrix formulation (Caswell 2000) with 70 annual age
classes, and a transition matrix of the form:
where survival sxis the probability that an individual aged x
survives to age x+ 1, and mxis the per capita production of
female calves by females in the year group to age x. The rate
of population change of this matrix at stable structure, λ,
gives the equilibrium population multiplication rate. Raw
age-specific fecundity estimates were used in the matrix up
to year group z, in which there were at least 20 potential
mothers in the population (z= 49 in Asian elephants, z = 53
in African); in older females, fecundity was assumed to be
constant at the average for this age group:
m>z=B>z/f>z. Survivorship values, σ, from Kaplan-
Meier curves were smoothed using Siler’s (1979) model of
mortality under competing risks:
where a, b, c, f and gare parameters to be estimated. From
this, survival probability for age class xis given by:
Curve parameters were estimated using a Gauss-Newton
non-linear fitting procedure, again implemented in R.
Results
Exploring the effects of being zoo-born
Import ages for wild-born, Asian females were extremely
wide-ranging: from 7 months to 54 years. However, import
age had no significant effect on adult survivorship (z= –0.88,
P= ns). Calf birthweight data did, in contrast, yield effects:
Asian calves born in zoos were significantly heavier than
those born in timber camps (Table 1); they also tended to
have greater Ponderal Indices — a trend we suggest needs
verifying and investigating in future work (see Discussion).
Exploring how husbandry differs between the species
Median ages of separation from the dam were 8.3 years for
zoo-born Asian females, and 16.3 years for zoo-born
African females: this showed a trend towards significance
(z= 1.65, P < 0.10). Inter-zoo transfer rates averaged
0.140 transfers per year for Asian females (n = 386 individ-
uals), and 0.026 transfers per year for African females
(n = 216 individuals) (t= 3.91, P< 0.0001, df = 393).
Fecundity and population growth rates of both species
Fecundity is much lower in zoos than reference populations,
and breeding ceases at least a decade earlier (Figure 1).
Note that sample sizes for Amboseli females over 50 were
small (fewer than 10 females per age class), and this
probably accounts for the apparent increases in fecundity
towards the end of life. Some females carry on giving birth
into their 60s (see Moss 2001), which is why the plot fails
to return to zero even at these older ages. Small sample
sizes precluded investigating birth origin effects on repro-
ductive rates in zoos.
Both species are projected to decline at a rate of approx-
imately 10% per year, if reliant on captive-bred animals
and kept under historically-prevailing conditions
(Figure 2). Improving fecundity to reference population
levels would have a much greater effect than would
improving survival (Figure 2); indeed, predicted popula-
tion growth rates approach stability for African
elephants if just fecundity alone is improved. However,
for both species it is still necessary to improve survival
too, this being especially crucial for Asian elephants.
Discussion
Here, we summarise our analyses of fecundity and popu-
lation viability in European zoos, and compare these
results, alongside our previous survivorship findings, with
North American zoo populations. We look in detail at risk
factors for Asian elephant survivorship, and potential
species differences in husbandry. Finally, we identify
potential causes of zoo elephants’ problems, and suggest
how to test these hypotheses.
The fecundity and population viability of female
elephants in European zoos
For both species, fecundity is much lower in European zoos
than in reference populations. To some extent, this is simply
caused by logistics, particularly restricted breeding opportu-
nities: zoo females, unlike Amboseli and MTE females, have
limited male access. However, there are clearly zoo-specific
physiological problems too: while their in situ conspecifics
breed well at least into their late 40s, zoo females show
minimal breeding after their 30s. This combines with
prevalent acyclity (see Brown et al 2004; Freeman et al
2004; Hermes et al 2004), and, in Asian females, high still-
birth rates (Clubb et al 2008). One potential explanation for
reproductive senescence involves not breeding while young
(eg Hermes et al 2004), but cause and effect have yet to be
© 2009 Universities Federation for Animal Welfare
Viability of zoo elephants 241
demonstrated here (Mason & Veasey 2009, in press), and
there are alternative explanations for both this phenomenon
and poor cyclity/high stillbirths (see below).
We project population declines in European zoos, if kept
under historically prevailing conditions with no imports, of
11% a year for African elephants and 10% a year for Asian
elephants. With no improvement, and assuming stable age
structures and no imports, European African elephant popu-
lations would be expected to fall to 10 females within
around 29 years, and Asian elephants to fall to 10 in around
37 years. For African zoo elephants, solely improving
fecundity to Amboseli levels would almost achieve a self-
sustaining population. For Asian elephants, in contrast, both
fecundity and survivorship need improving to at least MTE
levels if the European zoo population is to become self-
sustaining. This apparent species difference could reflect
differences between the two zoo populations (eg higher calf
losses in zoo Asians), and/or differences between reference
populations. MTE fecundity is lower than Amboseli,
suggesting that MTE values are not optimal (see Materials
and methods); thus, if data from well-protected wild, Asian
populations became available, it would be advisable to re-
analyse the relative importance of achieving reference
levels of fecundity versus survivorship.
Comparisons with North American zoos
African elephants in North American zoos are not self-
sustaining, and have infant mortality rates, including
stillbirths, of approximately 40% (Olson & Wiese 2000;
Faust 2005) — more than double female calf loss rates in
Europe (Clubb et al 2008). Population declines are
similarly projected for North American Asian elephants,
even in analyses combining zoo-born with wild-born
individuals (eg Wiese 2000; Faust et al 2006). Infant
mortality rates — at least prior to 2005 — were similar in
both continents for this species, with total first year
mortalities of approximately 40% (Clubb et al 2008;
Saragusty et al 2008). Analyses of more up-to-date
datasets, however, suggest an improvement in Europe in
recent years, but no similar upturn in North America
(Saragusty et al 2008). Furthermore, after infancy, the
median lifespan of those Asian female elephants which
survive their first year seems slightly longer in Europe, at
47.6 compared to 44.8 in North America (Wiese & Willis
2004). Thus, overall, the female African and Asian
elephants of Europe seem to be faring similarly to, or
even better than, those in North America. The problems
reported here are therefore not particular to Europe.
Indeed, since North American and European populations
combined represent the vast majority of ex situ zoo
elephants, these problems (and the potential solutions we
identify) probably apply globally to all zoo elephants.
Risk factors for Asian elephants in European zoos
Our previous investigations (see Introduction) revealed that
one risk factor affecting adult Asian elephant survivorship in
European zoos was being born into a zoo rather than
imported from the wild. Demographically, this birth origin
effect is likely to become increasingly important; the captive-
born proportion of the Asian, female zoo population has
already increased from 6.7% in 1960 to 18.6% in 2004.
Analyses here yielded more information. First, for wild-
caught animals, longer periods in (and/or perhaps en route
from) the wild before transfer to a zoo did not confer signifi-
cant additional benefits. This suggests that the benefits of
being wild-born are conferred early in life (via potential
mechanisms discussed below). Second, building on previous
Animal Welfare 2009, 18: 237-247
Table 1 Asian elephant neonatal bodyweights and body fat estimates.
Group/measure Population Population difference Data source/notes
In situ Zoo
Birth weight 89.5 (± 6.3) kg
(n = 5)
102.1 (± 9.6) kg
(n = 63)
F1,66 = 8.32, P = 0.005 Hayssen et al 1993
74.0 kg (n = 6) 105.6 kg (n = 40) Reported in paper as significant Kurt & Mar 1996 (sexes pooled)
–118.8 kg (n = 7) n/a ISIS 2002 (females only)
–Females, including
one stillbirth:
112.9 kg (n = 14)
Males, including 4
stillbirths: 117.2 kg
(n = 14)
n/a EEP studbook for Asian elephants 2005
PI Median: 0.114
(n = 5)
Median: 0.150
(n = 19)
Mann-Whitney W= 262.5, P< 0.10 Kurt & Mar 1996 (sexes pooled)
BMI Median: 11.16
(n = 5)
Median: 13.00
(n = 19)
Mann-Whitney W= 39, P= 0.10 Kurt & Mar 1996 (sexes pooled
Bodyweight (kg) and height (cm) used to calculate Ponderal Index (1000 × [kg/cm3]) and Body Mass Index (1000 × [kg/cm2]).
242 Clubb et al
analyses (Kurt & Mar 1996, reviewed by Dale 2009), we
found that one consequence of being zoo-born is being heavy
at birth, possibly also with a higher Ponderal Index. Our
analyses probably underestimate the magnitude of this effect
since more recent work puts the mean zoo Asian calf weight
at approximately 118 kg: heavier than the zoo values we
used, and perhaps reflecting a recent increase in Asian zoo
neonate weight (Dale 2009). This is informative because in
other species, excess peri-natal weight has harmful long-term
effects (reviewed below).
© 2009 Universities Federation for Animal Welfare
Figure 1
Fecundity curves for female (a) Asian and b) African elephants, expressed as female calves per female per year. Note that sample sizes
for the reference African population in (b) are small after 50 years of age (10 females or fewer per age class).
Viability of zoo elephants 243
Other risk factors for Asian female survivorship in European
zoos are inter-zoo transfer, and possibly also early maternal
separation (see Introduction). Here, we show that compared to
female African elephants in European zoos, their Asian coun-
terparts are transferred around five times more frequently.
Asian females were also separated from their mothers at half
the median age seen in African females (although this differ-
ence was just a trend). These husbandry differences could
explain why Asian elephants in European zoos have more
marked survivorship problems than African elephants, and
with no significant improvement in recent years.
What are the potential causes of zoo elephants’
problems? Two hypotheses and how to investigate
them in extant zoo elephant populations
Current knowledge (eg reviewed Clubb & Mason 2002;
Mason & Veasey 2008) suggests that the main causes of zoo
elephant infant mortality are dystocia, infanticide, maternal
neglect and, especially for Asian elephants, Herpes virus;
while adult deaths in zoos often reportedly stem from cardio-
vascular disease, but are frequently for reasons unknown.
Poor conception rates stem from acyclicity and premature
senescence, ovarian cysts, male fertility problems, and low
libido. There are numerous possible explanations for why
these problems are more serious or prevalent in zoos than
in situ populations, but here we propose two broad candi-
dates: obesity and stress (not mutually exclusive, since
chronic stress can facilitate fat deposition: eg Chrousos
2000; Tiley et al 2007). As we review below, these two
hypotheses are parsimonious, having the potential to explain
all observed effects; plausible given what is known about
elephant health and husbandry; testable (ie open to falsifica-
tion); and, we hope, useful: they could potentially enable the
identification of at-risk individuals before they die (allowing
intervention), and the objective evaluation of changes in zoo
husbandry and management.
In many species, including humans, obesity increases
morbidity, for example from cardiovascular disease,
diabetes, cancer and asthma, and shortens adult lifespan
(Bennett 1999; British Nutrition Foundation 1999; Kealy
et al 2002; Rodrigues-Artalejo et al 2002; Royal College of
Physicians 2004). It also reduces fertility and increases the
risk of stillbirths (Clark et al 1998; British Nutrition
Foundation 1999; Kristensen et al 2005; Baur et al 2006).
There have long been suggestions that adult elephants in
zoos are overweight (Kurt & Mar 1996; Ange et al 2001;
Animal Welfare 2009, 18: 237-247
Figure 2
Predicted long-term population growth rate, λ, given by the rate of change at stable age structure of a Leslie matrix model, under var-
ious scenarios. A self-sustaining population is reached when λ= 1, as marked by the dashed line. Survival and fecundity rates are assumed
to be either as measured in zoo-born elephants, or, to model improved future scenarios, as measured in the captive-born Asian MTE
reference population and the ‘natural mortality’ African reference population (Amboseli). Models that used the rates of fecundity found
in either the Zoo or Reference populations are indicated by their location on the x-axis. Red bars indicate scenarios where Zoo sur-
vival rates were used, and blue bars, where Reference survival rates were used. The combination of x-axis location and colour of bar
thus indicates which fecundity and survival values were used, respectively, in each analysis yielding a given λ.
244 Clubb et al
Hatt & Clauss 2006). Kurt and Mar (1996) also report that
overweight zoo dams have more stillbirths, while Dale
(2009) shows that stillborn Asian calves weigh more than
liveborn (see also Kurt & Mar 1996). Excess body fat could
thus explain the high stillbirth rates of Asian elephants,
large calf sizes, and low fecundities and shortened adult
lifespans of both species. Furthermore, in humans, over-
weight babies have elevated risks of obesity, cancer, cardio-
vascular disease and Type II diabetes in adulthood
(McCance et al 1994; Rasmussen & Johanssen 1998;
Bennett 1999; Samaras et al 2003; Singhal & Lucas 2004;
Jiminez-Challaron & Patti 2007). Excess neonatal or infant
body fat could also thus potentially explain why zoo-born
Asian elephants have premature adult deaths.
To investigate whether peri-natal and/or adult obesity does
indeed cause poor fecundity or survivorship in zoo
elephants, useful screens of the extant population include:
scores for overall body fat and deposition patterns, and
kidney fat deposit size post mortem; serum triglyceride and
cholesterol levels; and indices of insulin resistance such as
decreased glucose tolerance and fasting hyperglycaemia
(Albl 1971; British Nutrition Foundation 1999; Samaras
et al 2003; Kronfeld et al 2005). Leptin, which correlates
with adiposity in many species (eg Fors et al 1999; Banks
et al 2001; Buff et al 2002), should also be assessed.
Our second hypothesised cause of problems is stress.
Chronic and acute stress both reduce adult lifespan in
humans and other species (eg Kiecolt-Glaser et al 2002;
Cavigelli & McClintock 2003; Donaldson 2003; Vitetta
et al 2005); indeed, the post-capture deaths of wild-born
Burmese elephants caught for logging are likely examples
(Mar 2007; Clubb et al 2008). In addition, in a range of
species, chronic stress reduces fertility and elevates still-
birth rates (eg Janczak et al 2003; Wingfield & Sapolsky
2003), impairs maternal care and infant survivorship (eg
Bahr et al 1998; Clubb & Mason 2003, 2007; Janczak et al
2003), and even induces reproductive senescence (Kaplan
& Manuck 2004; Cavigelli et al 2006). Zoo elephants are
often subject to treatments known or likely to elevate stress,
such as chaining and translocation (reviewed eg Clubb &
Mason 2002; Mason & Veasey 2008; see also Harris et al
2008). We strongly suspect that stress underlies the harmful
effects of repeated transfers on survivorship, and possibly
early separation from mothers (for effects of breaking social
bonds in other species, see eg Moore et al 1994; Capitanio
et al 1998; Ha et al 1999). Furthermore, stress early in life
can have lasting deleterious effects: exposure to elevated
stress hormones in utero and/or inadequate parental care in
infancy may disrupt stress responses throughout life (eg
Francis et al 1999; Otten et al 2001; Cronje 2003; Pryce
et al 2005), elevate stress-related disease (Danese et al
2007) and shorten lifespan (Lewis et al 2000). If zoo-born
calves experience more early stress than wild-born calves
(perhaps because their own dams are stressed, or a lack of
competent allomothering), this provides an alternative
explanation for zoo Asian elephants’ birth origin effect.
To test the stress hypothesis, useful screens would include:
measures of corticosteroid, ACTH and catecholamine
outputs; assessments of immune (eg IL-6) and inflamma-
tory responses (and related diseases); wound-healing rates;
and adrenal and thymus weights, post mortem (eg Broom &
Johnson 1993; Kiecolt-Glaser et al 2002; Cronje 2003;
Kielcolt-Glaser et al 2003; Terio et al 2004; Danese et al
2007; Mason & Veasey 2009, in press).
Additional questions that could be addressed from
zoo records
We also recommend further analyses of zoo records. Re-
running analyses with pooled European and North American
studbook data could elucidate whether the lack of significant
findings in African cf Asian elephants was simply the result
of low power; identify further aspects of husbandry with an
effect on survivorship and fecundity; test whether our two
weaning age trends (the possible earlier separation of young
Asian elephants than young African elephants; and the
possible link in the former between early ‘weaning’ and
reduced survivorship) can be replicated; and also enable
studies of male survivorship. Analyses of more up-to-date
datasets than our own could also show whether apparent
recent improvements in calf survival in Europe (Saragusty
et al 2008), are statistically significant. We suggest analysing
historical data on infant birth-weights, to see if these predict
later health and lifespan in adulthood. If Ponderal Indices
can be calculated from such records, such data would also
test the robustness of our trend finding that zoo neonates
have higher Ponderal Indices than timber camp neonates.
Analysing health records could also reveal whether zoo-born
and wild-born elephants die for different reasons, and
identify what causes Asian elephant deaths in the years after
each inter-zoo transfer, to help test the hypotheses outlined
above. Lastly, it would be valuable to analyse records of
changes in zoo housing or husbandry to investigate why
survivorship in African, but not Asian, elephants has
improved in recent years in Europe; compare the enclosure
sizes, group sizes and group structures of Asian and African
elephants in European zoos; and compare European and
North American husbandry, to find reasons for the appar-
ently lower infant losses seen in European African elephants
and more recently also perhaps Asian elephants.
Animal welfare implications
The high mortality and low fecundity seen in Asian
elephants, and to a lesser extent in African elephants, raises
questions regarding their health and welfare. To date, many
zoos have responded to their declining elephant populations
by advocating importation from successful in situ popula-
tions (Hutchins & Keele 2006; Wiese & Willis 2006).
Substantial sums have also been invested in new enclo-
sures — but research is still needed into what housing or
husbandry methods are most beneficial (Mason & Veasey
2008) and truly provide ‘an environment well-adapted to
meet the physical, psychological and social needs of the
species’ as stated in the 2002 EC ‘Zoo’ Directive
1999/22/EC; Zoo Licensing Act 1981 (Amendment)
(England and Wales) Regulations. We recommend that new
tests sensitive to altered stress physiology and morbid
obesity are incorporated into zoo elephant health screen-
© 2009 Universities Federation for Animal Welfare
Viability of zoo elephants 245
ings. This would enable testing of the hypotheses that stress
and/or obesity underlie the observed effects, and potentially
allow at-risk individuals to be identified, allowing timely
intervention. Even in the absence of such data, however, the
effect of inter-zoo transfer, and possible effect of early sepa-
ration from mothers, on the survivorship of Asian female
elephants seem very likely to be stress-related. We suggest
therefore that avoiding inter-zoo transfer and early separa-
tion should be a welfare priority, until it is understood how
to mitigate their harmful effects.
Acknowledgements
The work presented here was unfunded, but only possible
thanks to previous support to all authors enabling them each
to build up their databases and requisite skills. We gratefully
acknowledge them here. RC thanks Ruth Ripley for statis-
tical advice; RC and GJM thank the RSPCA for past
funding, Amelia Terkel and Ton Dorrestyn for access to
studbook data, Miranda Stevenson and Olivia Walter for
help with analysis of studbook data; KUM thanks Prospect
Burma Foundation, Charles Wallace Burma Trust, Three
Oaks Foundation, Whitley-Laing Foundation (Rufford Small
Grants), Toyota Foundation, Fantham Memorial Research
Scholarship and University College London for funding, and
Thoung Nyunt for comments and help in compiling data.
GJM thanks NSERC for salary support, and Anna Kate
Shovellor and Brian McBride for advice on obesity research
plus Niel Karrow for suggestions on immunological
screening. PL and CM thank the many donors to AERP who
have helped fund the long-term monitoring, and to S
Sayialel, N Njiraini and K Sayialel for invaluable contribu-
tions to data collection. We also thank Joseph Saragusty as
well as the referees for their useful comments.
References
African Elephant Specialist Group 2003 Statement from the
African Elephant Specialist Group of the IUCN Species Survival
Commission on the removal of African Elephants for Captive Use.
IUCN/SSC African Elephant Specialist Group: Mokuti Lodge,
Namibia, Africa
African Elephant Specialist Group 2004 Loxodonta africana.
2007 IUCN Red List of Threatened Species. www.iucnredlist.org.
(Accessed 8 July 2008)
Albl P 1971 Studies on assessment of physical condition in
African elephants. Biological Conservation 3: 134-140
Ange K, Crissey SD, Doyle C, Lance K and Hintz H 2001
A survey of African (Loxodonta africana) and Asian (Elephas max-
imus) Elephant diets and measured body dimensions compared to
their estimated nutrient requirements. Proceedings of the Nutrition
Advisory Group Conference 4: 5-14
Anthony KRN, Connolly SR and Hoegh-Guldberg O 2007
Bleaching, energetics, and coral mortality risk: Effects of temperature,
light, and sediment regime. Limnology and Oceanography 52: 716-726
Asian Elephant Specialist Group 1996 Elephas maximus.
2007 IUCN Red List of Threatened Species. www.iucnredlist.org.
(Accessed 8 July 2008)
Bahr NI, Pryce CR, Dobeli M and Martin RD 1998 Evidence
from urinary cortisol that maternal behaviour is related to stress
in gorillas. Physiology and Behavior 64: 439-437
Balmford A, Leader-Williams N and Green MJB 1995
Parks or arks — where to conserve threatened mammals?
Biodiversity and Conservation 4: 595-607
Balmford A, Mace G and Leader-Williams N 1996
Designing the ark: Setting priorities for captive breeding.
Conservation Biology 10: 719-727
Banks WA, Phillips-Conroy JE, Jolly CE and Morley JE
2001 Serum leptin levels in wild and captive populations of
baboons (Papio spp): Implications for the ancestral role of leptin.
Journal of Clinical Endocrinology & Metabolism 86: 4315-4320
Baur JA, Pearson KJ, Price NL, Jamieson HA, Lerin C,
Kalra A, Prabhu W, Allard JS, Lopez-Lluch G, Lewis K,
Pistell PJ, Poosala S, Becker KG, Boss O, Gwinn D, Wang
M, Ramaswamy S, Fishbein KW, Spencer RG, Lakatta
EG, Le Couteur D, Shaw RJ, Navas P, Puigserver P,
Ingram DK, de Cabo R and Sinclair DA 2006 Resveratrol
improves health and survival of mice on a high-calorie diet. Nature
444: 337-342
Bennett PH 1999 Type 2 diabetes among the Pima Indians of
Arizona: An epidemic attributable to environmental change?
Nutrition Reviews 57: S51-S54
Blanc JJ, Barnes RFW, Craig GC, Dublin HT, Thouless
CR, Douglas-Hamilton I and Hart JA 2007 African Elephant
Status Report 2007: An Update from the African Elephant Database.
Occasional Paper of the IUCN Species Survival Commission No. 33.
IUCN: Gland, Switzerland
British Nutrition Foundation 1999 Obesity — the Report of the
British Nutrition Foundation Task Force. Blackwell Science: Oxford, UK
Broom DM and Johnson KG 1993 Stress and Animal Welfare.
Chapman & Hall: London, UK
Brown JL, Olson D, Keele M and Freeman EW 2004 Survey
of the reproductive cyclicity status of Asian and African elephants
in North America. Zoo Biology 23: 309-321
Buff PR, Dodds AC, Morrison CD, Whitley NC, McFadin
EL, Daniel JA, Djiane J and Keisler DH 2002 Leptin in hors-
es: Tissue localization and relationship between peripheral con-
centrations of leptin and body composition. Journal of Animal
Science 80: 2942-2948
Capitanio JP, Mendoza SP, Lerche NW and Mason AW
1998 Social stress results in altered glucocorticoid regulation and
shorter survival in simian acquired immune deficiency syndrome.
Proceedings of the National Academy of Sciences, USA 95: 4714-4719
Caswell H 2000 Matrix Population Models: Construction, Analysis and
Interpretation, 2nd Edition. Sinauer Associates: Sunderland, MA, USA
Caughley G 1977 Analysis of Vertebrate Populations. Wiley: New
York, USA
Cavigelli SA and McClintock MK 2003 Fear of novelty in infant
rats predicts adult corticosterone dynamics and an early death.
Proceedings of the National Academy of Science 100: 16131-16136
Cavigelli SA, Yee JR and McClintock MK 2006 Infant tem-
perament predicts life span in female rats that develop sponta-
neous tumors. Hormones and Behavior 50: 454-462
Chrousos GP 2000 The role of stress and the hypothalamic-
pituitary-adrenal axis in the pathogenesis of the metabolic syn-
drome: neuron-endocrine and target tissue-related causes.
International Journal of Obesity 24: S50-S55
Clark AM, Thornley B, Tomlinson L, Galletley C and
Norman RJ 1998 Weight loss in obese infertile women results
in improvement in reproductive outcome for all forms of fertility
treatment. Human Reproduction 13: 1502-1505
Clubb R and Mason G 2002 A Review of the Welfare of Elephants
in European Zoos. RSPCA: Horsham, UK
Clubb R and Mason G 2003 Captivity effects on wide-ranging
carnivores. Nature 410: 35-36
Clubb R and Mason G 2007 Natural behavioural biology as a
risk factor in carnivore welfare: how understanding species differ-
ences could help zoos redesign enclosures. Applied Animal
Behaviour Science 102: 303-328
Animal Welfare 2009, 18: 237-247
246 Clubb et al
Clubb R, Lee P, Mar KU, Moss C, Rowcliffe M and Mason GJ
2008 Compromised survivorship in zoo elephants. Science 322: 1649
Cronje PB 2003 Foetal programming of immune competence.
Australian Journal of Experimental Agriculture 43: 1427-1430
Dale RHI 2009 Birth statistics for African (Loxodonta africana)
and Asian (Elephas maximus) elephants in human care: history and
implications for elephant welfare. Zoo Biology, doi:
10.1002/zoo.20234
Danese A, Pariante CM, Caspi A, Taylor A and Poulton R
2007 Childhood maltreatment predicts adult inflammation in a life
course study. Proceedings of the National Academy of Sciences 104:
1319-1324
Del Giudice GD, Fieberg J, Riggs MR, Carstensen Powell M
and Pan W 2006 A long-term age-specific survival analysis of female
white-tailed deer. Journal of Wildlife Management 70: 1556-1568
Donaldson R 2003 Experiences of Older Burglary Victims. Home
Office Findings No 198. Home Office: London, UK
European Commission 2002 EC Zoo Directive 1999/22/EC. Zoo
Licensing Act Amendment Regulations European Commission:
Brussels, Belgium
European Elephant Group (EEG) 2002 Elefanten in zoos und
safariparks Europa. Schling Verlag: Munster, Germany
European Elephant Group 2008 http://www.elefanten-schutz-
europa.de/index.html
Faust L 2005 Technical Report on Demographic Analyses and
Modeling of the North American African Elephant Population. African
Elephant Modelling Technical Report, AZA. Association of Zoos and
Aquariums: Silver Springs, Maryland, USA
Faust LJ, Thompson SD and Earnhardt JM 2006 Is reversing
the decline of Asian elephants in North American zoos possible?
An individual-based modeling approach. Zoo Biology 25: 201-218
Fors H, Matsuoka H, Bosaeus I, Rosberg S, Wikland KA
and Bjarnson R 1999 Serum leptin levels correlate with growth
hormone secretion and body fat in children. Journal of Clinical
Endocrinology & Metabolism 84: 3586-3590
Francis D, Diorio J, Liu D and Meaney J 1999 Nongenomic
transmission across generations of maternal behavior and stress
responses in the rat. Science 286: 1155-1158
Freeman EW, Weiss E and Brown JL 2004 Examination of the
interrelationships of behavior, dominance status, and ovarian activ-
ity in captive Asian and African elephants. Zoo Biology 23: 431-448
Goswami VR, Madhusudan MD and Karanth KU 2007
Application of photographic capture-recapture modelling to esti-
mate demographic parameters for male Asian elephants. Animal
Conservation 10: 391-399
Grambsch PM and Therneau TM 1994 Proportional hazards tests
and diagnostics based on weighted residuals. Biometrika 81: 515-526
Ha J, Robinette RL and Sackett GP 1999 Social housing and
pregnancy outcome in captive pigtailed Macaques. American Journal
of Primatology 47: 153-163
Harris M, Sherwin S and Harris S 2008 The welfare, housing and
husbandry of elephants in UK zoos: final report. University of Bristol,
DEFRA Science and Research Project WC05007, Bristol, UK
Hatt JM and Clauss M 2006 Feeding Asian and African ele-
phants (Elephas maximus and Loxodonta africana) in captivity.
International Zoology Yearbook 40: 88-95
Hayssen V, van Tienhoven A and van Tienhoven A 1993
Asdell’s Patterns of Mammalian Reproduction. Cornell University
Press: Ithaca, USA
Hedges S, Tyson MJ, Sitompul AF and Hammatt H 2006
Why inter-country loans will not help Sumatra’s elephants. Zoo
Biology 25: 235-246
Hedges S 2007 Conservation. In: Fowler ME and Mikota SK
(eds) Biology, Medicine, and Surgery of Elephants pp 475-490.
Blackwell Publishing: UK
Hermes R, Hildebrandt TB and Göritz F 2004 Reproductive
problems directly attributable to long-term captivity: asymmetric
reproductive aging. Animal Reproduction Science 82-83: 49-60
Hutchins M and Keele M 2006 Elephants importation from
range countries: Ethical and practical considerations for accredit-
ed zoos. Zoo Biology 25: 219-233
International Species Inventory System (ISIS) 2002
Elephas maximus, Asian Elephant. Reference Ranges for Physiological
Data Values (average values calculated for: females only).
http://www.elephantcare.org/protodoc_files/Isis/isis2.htm
(accessed May 2009)
Janczak AM, Pedersen LJ, Rydhmer L and Bakken M 2003
Relation between early fear and anxiety-related behaviour and
maternal ability in sows. Applied Animal Behavior Science 82: 121-135
Jimenez-Chillaron JC and Patti ME 2007 To catch up or not
to catch up: is this the question? Lessons from animal models.
Current Opinions in Endocrinology, Diabetes and Obesity 14: 23-29
Kaplan JR and Manuck SB 2004 Ovarian dysfunction, stress
and disease: A primate continuum. ILAR Journal 45: 89-115
Kealy RD, Lawler BF, Ballam JM, Mantz SL, Biery DN,
Greeley EH, Lust G, Segre M, Smith GK and Stowe HD 2002
Effects of diet restriction on life span and age-related changes in dogs.
Journal of American Veterinary Medical Association 220: 1315-1320
Keele M, Lewis K and Owens T 2007 Asian Elephant (Elephus
maximus) North American Regional Studbook Update 1 May 2005-16
July 2007. Association of Zoos and Aquariums: Silver Springs,
Maryland, USA
Kiecolt-Glaser JK, McGuire L, Robles TF and Glaser R
2002 Emotions, morbidity, and mortality: new perspectives from
psychoneuroimmunology. Annual Review of Psychology 53: 83-107
Kiecolt-Glaser JK, Preacher KJ, MacCallum RC, Atkinson
C, Malarkey WB and Glaser R 2003 Chronic stress and age-
related increases in the proinflammatory cytokine IL-6. Proceedings
of the National Academy of Science 100: 9090-9095
Koehl D 2008 Elephant Database http://www.elephant.se/ele-
phant_database.php?open=Elephant%20database
Kristensen J, Vestergaard M, Wisborg K, Kesmodel U,
and Jørgen Secher N 2005 Pre-pregnancy weight and the risk
of still birth and neonatal death. BJOG: an International Journal of
Obstetrics and Gynaecology 112: 403-408
Kronfeld DS, Treiber KH and Geor RJ 2005 Comparison of
nonspecific indications and quantitative methods for the assess-
ment of insulin resistance in horses and ponies. Journal of American
Veterinary Medical Association 226: 712-719
Kurt F and Mar DK 1996 Neonate mortality in captive Asian ele-
phants (Elephas maximus). Zeitschrift für Säugetierkunde 61: 155-164
Kurt F and Kumarasinghe JC 1998 Remarks on body growth
and phenotypes in Asian elephant Elephas maximus. Acta
Theriologica Supplement 5: 135-153
Lair RC 1997 Gone Astray: The Care and Management of the Asian
Elephant in Domesticity. FAO/RAP Publication 1997/16. FAO Regional
Office for Asia and the Pacific: Thailand
Lewis MH, Gluck JP, Petitto JM, Hensley LL and Ozerd H
2000 Early social deprivation in nonhuman primates: Long-term
effects on survival and cell-mediated immunity. Biological
Pshychiatry 47: 119-126
Mar KU and Win S 1997 Working elephants for the timber
industry. Draught Animal News 26: 6-13
Mar KU 2007 The demography and life history strategy of timber ele-
phants in Myanmar. PhD Thesis, University College London,
London, UK. http://www.zoo.cam.ac.uk.theses.htm
Mason G and Veasey J 2008 How should the psychological wel-
fare of zoo elephants be investigated? In: Forthman DL, Kane LF
and Waldau PF (eds) An Elephant in the Room: the Science and Well
Being of Elephants in Captivity pp 154-174. Tufts University
Cummings School of Veterinary Medicine’s Center for Animals
and Public Policy: North Grafton, MA, USA
© 2009 Universities Federation for Animal Welfare
Viability of zoo elephants 247
Mason G and Veasey J 2009 How should the psychological well-
being of zoo elephants be objectively investigated? Zoo Biology, in press
McCance DR, Pettitt DJ, Hanson RL, Jacobsson LTH,
Knowler WC and Bennett PH 1994 Birth-weight and non-
insulin-dependent diabetes – thrifty genotype, thrifty phenotype or
surviving small baby genotype. British Medical Journal 308: 942-945
Moore AS, Gonyou HW, Stookey JM and McLaren DG
1994 Effect of group composition and pen size on behavior, pro-
ductivity and immune response of growing pigs. Applied Animal
Behaviour Science 40: 13-30
Moss CJ 2001 The demography of an Africa elephant
(Loxodonta africana) population in Amboseli, Kenya. Journal of
Zoology 255: 145-156
Moss CJ and Lee PC Female reproductive strategies: individual
life histories. In: Moss CJ and Croze H (eds) The Amboseli Elephant:
a Long-term Perspective on a Long-lived Species. University of
Chicago Press: Chicago, IL, USA, in press
Olson D and Wiese RJ 2000 State of the North American
African elephant population and projections for the future. Zoo
Biology 19: 311-320
Olson D 2008 North American Regional Studbook for the African
Elephant (Loxodonta africana). Association of Zoos and
Aquariums: Silver Springs, Maryland, USA
Otten W, Kanitz M, Tuchscherer M and Numberg G 2001
Effects of prenatal restraint stress on hypothalamic-pituitary-
adrenocortical and sympatho-adrenomedullary axis in neonatal
pigs. Animal Science 73: 279-287
Pryce CR, Ruedi-Bettschen D, Dettling AC, Weston A,
Russig H, Ferger B and Feldon J 2005 Long-term effects of
early-life environmental manipulations in rodents and primates:
Potential animal models in depression research. Neuroscience and
Biobehavioral Reviews 29: 649-674
Raggi P, Shaw LJ, Berman DS and Callister TQ 2004
Prognostic value of coronary artery calcium screening in subjects
with and without diabetes. Journal of the American College of
Cardiology 43: 1663-1669
Ramanthan A and Mallapur A 2008 A visual health assess-
ment of captive Asian elephants (Elephas maximus) housed in
India. Journal of Zoo & Wildlife Medicine 39: 148-154
Rasmussen F and Johanssen M 1998 The relation of weight,
length and Ponderal index at birth to body mass index and over-
weight among 18-year-old males in Sweden. European Journal of
Epidemiology 14: 373-380
Richman LK, Montali RJ, Garber RL, Kennedy MA,
Lehnhardt J, Hildebrandt T, Schmitt D, Hardy D,
Alcendor DJ and Hayward GS 1999 Novel endotheliotropic
herpesviruses fatal for Asian and African elephants. Science 283:
1171-1176
Rodríguez-Artalejo F, Garcés C, Gorgojo L, López García
E, Martín-Moreno JM, Benavente M, del Barrio JL, Rubio
R, Oretga H, Fernandez O and de Oya M 2002 Dietary pat-
terns among children aged 6-7 in four Spanish cities with widely
differing cardiovascular mortality. European Journal of Clinical
Nutrition 56: 141-148
Royal College of Physicians 2004 Storing Up Problems: The Medical
Case for a Slimmer Nation. Royal College of Physicians: London, UK
Samaras TT, Eirick H and Storms LH 2003 Birthweight,
rapid growth cancer and longevity: a review. Journal of the National
Medical Association 95: 1170-1183
Saragusty J, Hermes R, Göritz F, Schmitt DL and
Hildebrandt TB 2008 Skewed birth sex ratio and premature
mortality in elephants. Animal Reproduction Science. Oct 31
doi:10.1016/j.anireprosci.2008.10.019
Siler W 1979 Competing risk model for animal mortality. Ecology
60: 750-757
Singhal A and Lucas A 2004 Early origins of cardiovascular dis-
ease: is there a unifying hypothesis? The Lancet 363: 1642-1664
Slotow R, Garaï ME, Reilly B, Page B and Carr RD 2005
Population dynamics of elephants re-introduced to small fenced
reserves in South Africa. South African Wildlife Research 35: 23-32
Sukumar R 1989 The Asian Elephant: Ecology and Management.
Cambridge University Press: Cambridge, UK
Sukumar R, Krishnamurthy V, Wemmer C and Rodden
M1997 Demography of captive Asian elephants (Elephas max-
imus) in Southern India. Zoo Biology 16: 263-272
Sukumar R 2003 The Living Elephants: Evolutionary Ecology,
Behaviour, and Conservation. Oxford University Press: Oxford, UK
Sukumar R 2006 A brief review of the status, distribution and biol-
ogy of wild Asian elephants. International Zoology Yearbook 40: 1-8
Taylor VJ and Poole TB 1998 Captive breeding and infant mor-
tality in Asian elephants: a comparison between twenty western
zoos and three eastern elephant centers. Zoo Biology 17: 311-332
Terio KA, Marker L and Munson L 2004 Evidence for chron-
ic stress in captive but not free-ranging cheetahs (Acinonyx jubatus)
based on adrenal morphology and function. Journal of Wildlife
Diseases 40: 259-266
Tiley HA, Geor RJ and McCutcheon LJ 2007 Effects of dex-
amethasone on glucose dynamics and insulin sensitivity in healthy
horses. American Journal of Veterinary Research 68: 753-759
van Aarde RJ and Jackson TP 2007 Megaparks for metapopu-
lations: Addressing the causes of locally high elephant numbers in
Southern Africa. Biological Conservation 134: 289-297
Vitetta L, Anton B, Cortizo F and Sali A 2005 Mind-Body
Medicine: Stress and its impact on overall health and longevity.
Annals of the New York Academy of Science 1057: 492-505
Waithaka J, Osborne L and Saiwana L 1998 Captive breed-
ing of elephants. Pachyderm 25: 100-101
Whitehouse AM and Kerley GIH 2002 Retrospective assess-
ment of long term conservation management of elephants in
Addo Elephant National Park, South Africa. Oryx 36: 243-248
Wiese RJ 2000 Asian elephants are not self sustaining in North
America. Zoo Biology 19: 299-309
Wiese RJ and Willis K 2004 Calculation of longevity and
life expectancy in captive elephants. International Zoo
Yearbook 23: 365-373
Wiese RJ and Willis K 2006 Population management of zoo
elephants. Calculation of longevity and life expectancy in captive
elephants. International Zoo Yearbook 40: 80-87
Wingfield JC and Sapolsky RM 2003 Reproduction and
resistance to stress: when and how. Journal of
Neuroendocrinology 15: 711-724
Wittemyer G, Daballen D, Rasmussen H, Kahindi P and
Douglas-Hamilton I 2005 Demographic status of elephants in
the Sambusu and Buffalo Springs National Reserves, Kenya. African
Journal of Ecology 43: 44-47
Zoo Licensing Act 1981 Statutory Instrument 2002, No 3080
Amendment (England and Wales). http://www.opsi.gov.uk/
si/si2002/20023080.htm
Animal Welfare 2009, 18: 237-247
