Survey of the Reproductive Cyclicity
Status of Asian and African Elephants
in North America
Janine L. Brown,1n Deborah Olson,2 Mike Keele,3 and Elizabeth W. Freeman1
1Department of Reproductive Sciences, Conservation & Research Center, National
Zoological Park, Smithsonian Institution, Front Royal, Virginia
2Department of Conservation and Science, Indianapolis Zoo, Indianapolis, Indiana
3Oregon Zoo, Portland, Oregon
The Asian and African elephant populations in North America are not self-
sustaining, and reproductive rates remain low. One problem identi?ed from
routine progestagen analyses is that some elephant females do not exhibit normal
ovarian cycles. To better understand the extent of this problem, the Elephant
TAG/SSP conducted a survey to determine the reproductive status of the captive
population based on hormone and ultrasound evaluations. The survey response
rates for facilities with Asian and African elephants were 81% and 71%,
respectively, for the studbook populations, and nearly 100% for the SSP facilities.
Of the elephants surveyed, 49% of Asian and 62% of African elephant females
were being monitored for ovarian cyclicity via serum or urinary progestagen
analyses on a weekly basis. Of these, 14% of Asian and 29% of African elephants
either were not cycling at all or exhibited irregular cycles. For both species,
ovarian inactivity was more prevalent in the older age categories (>30 years);
however, acyclicity was found in all age groups of African elephants. Fewer
elephant females (B30%) had been examined by transrectal ultrasound to assess
reproductive-tract integrity, and corresponding hormonal data were available for
about three-quarters of these females. Within this subset, most (B75%) cycling
females had normal reproductive-tract morphologies, whereas at least 70% of
noncycling females exhibited some type of ovarian or uterine pathology. In
summary, the survey results suggest that ovarian inactivity is a signi?cant
reproductive problem for elephants held in zoos, especially African elephants. To
increase the fecundity of captive elephants, females should be bred at a young age,
before reproductive pathologies occur. However, a signi?cant number of older
Grant sponsor: Friends of the National Zoo; Grant sponsor: Morris Animal Foundation; Grant sponsor:
Smithsonian Scholarly Studies Program and Smithsonian Women?s Committee.
*Correspondence to: J.L. Brown, Conservation and Research Center, 1500 Remount Road, Front Royal,
VA 22630. E-mail: jbrown@crc.si.edu
Received for publication July 10, 2003; Accepted October 10, 2003.
DOI: 10.1002/zoo.20006
Published online in Wiley InterScience (www.interscience.wiley.com).
Zoo Biology 23:309?321 (2004)

c 2004 Wiley-Liss, Inc.
Asian elephants are still not being reproductively monitored. More signi?cantly,
many prime reproductive-age (10?30 years) African females are not being
monitored. This lack of information makes it dif?cult to determine what factors
affect the reproductive health of elephants, and to develop mitigating treatments to
reinitiate reproductive cyclicity. Zoo Biol 23:309?321, 2004. 
c 2004 Wiley-Liss, Inc.
Key words: ultrasound; ovarian cyclicity; reproductive dysfunction; hormones; progestagens
INTRODUCTION
The Asian (Elephas maximus) and African (Loxodonta africana) elephant
populations in North America are not self-sustaining, and it has been predicted that
demographic extinction will occur within three to ?ve decades unless reproductive
rates are increased severalfold [Wiese, 2000; Olson and Wiese, 2000]. Historically,
captive populations have been supplemented by elephants imported from range
countries, but this is not a preferred long-term, conservation-oriented strategy, and it
generates ?nancial, logistical, and ethical concerns. To address these management
issues, the Elephant Taxon Advisory Group (TAG)/Species Survival Plan (SSP)
works with advisors to identify problems and establish priorities for research and
collection planning. The challenges facing elephant managers are signi?cant. The
fecundity of elephants in captivity is low, in part because of limited numbers of bulls
and breeding facilities. However, even when introductions are made, breeding efforts
are not always successful.
The Endocrine Research Laboratory at the Conservation & Research Center
(CRC) currently assists the Elephant TAG/SSP by evaluating endocrine activity to
assess the fertility status of female elephants throughout North America. This is
done by measuring the concentration of progestagens in blood or urine samples
collected weekly. One ?nding from this work is that a number of elephant females
are not cycling [Brown, 2000]. In these animals progestagens remain at stable,
baseline concentrations, indicating a failure to initiate or sustain luteal activity. The
cause of this acyclicity is unknown, and there is concern that if the number of acyclic
females is high or growing, a signi?cant population decline is inevitable. Transrectal
ultrasound examinations have shown that aging females often have reproductive-
tract pathologies, such as uterine or ovarian cysts and tumors, that may prevent
conception [Hildebrandt et al., 2000].
In the light of these ?ndings, the Elephant TAG/SSP has endorsed research to
determine the extent of reproductive dysfunction in captive elephants, identify the
causes, and develop mitigating treatments [Keele and Ediger, 1997?2002]. Toward
that end, a reproductive survey was conducted in conjunction with the 2001?2002
Asian and African Elephant SSP Studbook updates to determine 1) how many
females in the captive population are being hormonally evaluated for estrous
cyclicity, 2) how many females are not cycling normally, 3) what types of
reproductive-tract pathologies exist, and 4) whether any pathologies are related to
acyclicity. The goal is to conduct periodic surveys to help the TAG/SSP continuously
monitor the reproductive health of individual elephants, as well as to assess changes
in the overall status of the population. Additionally, the ability to pinpoint the
precise time a female stops cycling would enable a better assessment of what factors
are associated with the development of reproductive problems (e.g., changes in social
status, environment, behavior, disease, nutrition, etc.).
310 Brown et al.
MATERIALS AND METHODS
In 2001?2002, reproductive surveys prepared by the Elephant TAG/SSP were
sent to all facilities listed in the Asian and African Elephant North American
Regional Studbooks. The survey questions related to this study asked 1) what
management system was used (free contact, protected contact, or other); 2) were
samples collected for reproductive monitoring (blood, urine, or none); 3) if samples
were collected, what was the collection frequency (weekly, biweekly, monthly,
elephant in training, or collected but not analyzed); 4) what was the estrous cyclicity
status of the elephants (cycling, not cycling, irregular cycles, or undetermined); and
5) was transrectal ultrasound performed, and if so, what were the results (normal,
ovarian cysts, uterine cysts/tumors, vaginal cysts, other pathologies, or in training)?
Because this was the ?rst survey, ??hormone cyclicity status?? referred to the results of
all hormone evaluations for an elephant at its respective institution. Future
reproductive data will be based on changes in cyclicity status that occur between
subsequent surveys. The results were summarized with the use of MS Excel. ??Animal
age?? refers to the elephant?s age in 2002 (the second year of the survey). Pregnant
cows were not included in the survey results or analyses. The mean data were
compared by Student?s t-tests. Differences in survey and ultrasound results between
species or cyclicity status within species were determined by Z-tests [SigmaStat v.
2.03, Jandel Scienti?c, 1997]. Percentage data for age categories within species were
tested by chi-square analysis. Data are presented as means7SEM.
RESULTS
Of the 66 facilities that maintained Asian elephants, 44 (67%) participated in
the Elephant SSP program. Of the 98 facilities that housed African elephants, 45
(46%) were part of the SSP. Twenty-two facilities held both Asian and African
elephants, and 14 were in the SSP. Of the 23 facilities that kept multiple Asian
females, 15 (65%) had all cycling females, seven (30%) had both cycling and
noncycling females, and one (5%) had two elephants that were both noncycling. Of
the 27 African elephant facilities, 12 (44%) had all cycling females, 14 (52%) had
both cycling and noncycling females, and one (4%) had two noncycling females. In
terms of the management systems used, of the 36 facilities with Asian elephants, 21
(58%; 145 elephants) employed a free-contact system, and 15 (42%; 48 elephants)
used protected contact. For African elephants, 20 facilities (43%; 74 females)
maintained African elephants under free contact, and 26 (57%; 66 females) used
protected contact.
The average age of captive female Asian elephants in 2002 was 33.770.9 years
(range? 2?77 years) for all elephants, and 31.371.2 years (2?59 years) for the SSP
population. The average age of captive female African elephants was 25.070.6 years
(2?52 years) for all animals, and 26.670.8 years (2?52 years) for those in the SSP.
Within species, there were no differences between the studbook and SSP populations
in average ages; however, African elephants were younger on average than Asian
elephants (Po0.05).
A summary of the reproductive survey results for elephants in the studbook
and SSP populations is presented in Table 1. For both Asian and African elephants,
B55% of the studbook population was part of the SSP. Of the remainder, 44% of
Elephant Reproductive Status in North America 311
Asian and 29% of African elephants were privately owned (by private individuals,
circuses, corporations, etc.). Surveys were returned for over three-quarters of the
elephants in North America, and there was a higher (Po0.05) response rate for
Asian elephants compared to African elephants. For SSP participants, there was a
nearly 100% compliance. Of the animals represented in the survey responses, over
half of the Asian elephants in the studbook were bled weekly (54%); however,
samples collected from 10 females in the SSP were not analyzed, resulting in a lower-
percentage monitoring rate. Overall, the hormone monitoring rate for African
elephants exceeded that of Asian elephants (Po0.05) by over 10 percentage points
for both the studbook and SSP populations, with higher rates observed for animals
managed within the SSP (Po0.05). When data were combined for elephants that
were not being monitored (NM) and those for which no survey information was
available (??unknown,?? but assumed to be not hormonally monitored), the total
number of females with no reproductive hormone data was higher (Po0.05) for
Asian than for African elephants listed in the studbook, although the percentages
were similar (P40.05). However, within the SSP, both the numbers and percentages
of females not hormonally monitored were higher (Po0.05) for Asian than for
African elephants (Table 1).
Representative examples of normal-cycling, irregular-cycling, and noncycling
progestagen pro?les are shown in Fig. 1. The normal elephant estrous cycle ranges
Table 1. Overall response rates and cyclicity status results of the Elephant TAG/SSP 2001?2002
Reproductive Survey for captive female Asian and African elephants in North America. Data are
evaluated for total studbook (SB) and SSP populations for each species
Asian African
SB population 239 206
SSP population 131 114
Surveys returned (%SB) 193 (81%) 146 (71%)
Surveys returned (%SSP) 129 (98%) 111 (97%)
No. monitored1 (%SB responses) 95 (49%)5 91 (62%)
Unknown+NM4 (%SB) 144 (60%) 115 (56%)
No. monitored1 (%SSP responses) 75 (58%)5 77 (69%)
Unknown+NM4 (%SSP) 56 (43%) 37 (32%)
Cyclicity Status (% of SB total)2
Cycling 80 (84%)a 64 (70%)b
Irregular 2 (2%) 6 (7%)
Noncycling3 11 (12%)a,3 20 (22%)b,3
Cyclicity Status (% of SSP)2
Cycling 61 (81%)a 55 (71%)b
Irregular 2 (3%) 4 (5%)
Noncycling3 10 (13%)a,3 17 (22%)b,3
1Reproductive cyclicity status based on weekly or bi-weekly serum or urinary progestin
assessments.
2Numbers are based on those monitored hormonally.
3Number excludes femalesr10 years of age (2 Asian; 1 African).
4Unknown? no survey returned; NM? not hormonally monitored.
5Does not include 10 females in the SSP that were bled weekly, but hormones were not
analyzed.
a,bPercentage differences between cyclicity categories with different superscripts are signi?cant
(Po0.05).
312 Brown et al.
Fig. 1. Representative serum progestagen pro?les of normal-cycling (a), irregular-cycling (b),
and noncycling (c) African elephant females, from analysis of blood samples collected weekly.
Elephant Reproductive Status in North America 313
from 14 to 16 weeks in duration [Brown, 2000]. The designation of irregular cycling
was subjectively based on observations of extended nonluteal periods (49 weeks),
shortened luteal phases (o6 weeks), and/or the presence of marked, random
?uctuations in progestagen concentrations during nonluteal or luteal phases.
In the elephants that were hormonally evaluated, there were species differences
(Po0.05) in the percentage of cycling vs. noncycling females. A higher rate of
normal cyclicity was observed in Asian elephants, whereas the rate of acyclicity was
greater in African females (Po0.05). Within species, there were no differences
(P40.05) in cyclicity status between elephants managed within or outside of the SSP.
There also were no differences (P40.05) in cyclicity status between elephants
managed under free- or protected-contact systems. Six of 11 (55%) acyclic Asian
elephants were managed under free contact, whereas ?ve (45%) were managed under
protected contact. For African elephants, 11 (55%) of 20 noncycling females were
managed under free contact, and nine (45%) were managed under protected contact.
Tables 2 and 3 summarize the survey results based on age categories. In 2002,
the majority of females in the Asian population (67%) were 430 years of age,
whereas most African elephants (72%) were o30 years old. Similarly, in the SSP
population the majority of Asian elephants (64%) were 430 years old, and most
African elephants (67%) wereo30 years old. Excluding data for femalesr10 years
Table 2. Numbers and age distribution of Asian elephant females exhibiting normal estrous cycles
(cycling), no ovarian activity (noncycling) or irregular estrous cycles (irregular) in the total
studbook (SB) and SSP populations in North America based on an Elephant TAG/SSP 2001?
2002 Reproductive Survey
Age of females in 2002 (years)
r10 11?20 21?30 31?40 >40
SB population1 21 19 39 93 67
SSP population1 14 12 21 55 29
Responses (Studbook) 20 17 33 72 51
Responses (SSP) 14 12 21 54 28
No. monitored (SB) 7 10 20 42 16
No. monitored (SSP) 6 10 13 31 15
Cycling (SB) 5 10 18 35 12
Cycling (SSP) 4 10 11 25 11
Noncycling (SB) 2 0 1 6 4
Noncycling (SSP) 2 0 1 5 4
Irregular (SB) 0 0 1 1 0
Irregular (SSP) 0 0 1 1 0
% noncycling (SB)2 28.6 0a 5.0b 14.3c 25.0d
% noncycling (SSP)2 33.3 0a 7.6b 16.1c 26.6d
Unknown+NM3 (SB) 14 9 19 51 51
Unknown+NM3 (SSP) 8 2 8 24 14
% Unknown+NM (SB) 66.7 47.4 48.7 54.8 76.1
% Unknown+NM (SSP) 57.1 16.7 38.1 43.6 48.3
1Number of elephants in each population.
2Percentage based on elephants being hormonally monitored.
3Unknown? no survey returned; NM? not hormonally monitored.
a,b,c,dPercentages of noncycling females between each age category with different superscripts
are signi?cantly different (Po0.05). Analysis does not include females r10 years of age.
314 Brown et al.
of age (i.e., potentially prepubertal), the average age was greater (Po0.05) in
noncycling than in cycling females for both Asian (31.971.1 years, range? 11?59
years vs. 37.271.7 years, range? 27?53 years, respectively) and African (24.470.8
years, range? 16?52 years vs. 28.971.7 years, range? 14?42 years, respectively)
elephants. There were only two irregular-cycling Asian elephants (27 and 32 years
old), and these were not were not statistically compared with the cycling and
noncycling Asian elephants. The average age of irregular-cycling African elephants
(20.871.2 years, range? 16?24 years) was less (Po0.05) than that of both cycling
and noncycling females. Within age categories, there was also a greater percentage of
noncycling females observed in the older age groups (Po0.05) (Tables 2 and 3).
However, there were also larger proportions of unmonitored and unsurveyed
females in these categories. Overall, the reproductive status of over half of Asian and
African elephants 430 years of age, and over two-thirds of elephants 440 years of
age was unknown (Tables 2 and 3). For African elephants, a high percentage of
unmonitored/unknown elephants (most of which were managed outside of the SSP)
was also observed in the 11?20-year-old category (Table 3).
Ultrasound examinations were conducted on 70 Asian and 50 African
elephants, and corresponding hormonal data were available for 53 of the Asian
and 45 of the African females (Table 4). The average age of females diagnosed with
reproductive-tract pathologies was greater (Po0.05) than that of elephants with
Table 3. Numbers and age distribution of African elephant females exhibiting normal estrous
cycles (cycling), no ovarian activity (noncycling) or irregular estrous cycles (irregular) in the total
studbook (SB) and SSP populations in North America based on an Elephant TAG/SSP 2001?
2002 Reproductive Survey
Age of females in 2002 (years)
r10 11?20 21?30 31?40 440
SB population1 3 72 74 45 12
SSP population1 3 22 52 26 11
Responses (SB) 3 33 58 41 11
Responses (SSP) 3 22 50 25 11
No. monitored (SB) 1 24 42 20 4
No. monitored (SSP) 1 20 37 15 4
Cycling (SB) 0 19 31 12 2
Cycling (SSP) 0 16 29 8 2
Noncycling (SB) 1 2 8 8 2
Noncycling (SSP) 1 2 7 6 2
Irregular (SB) 0 3 3 0 0
Irregular (SSP) 0 3 1 0 0
% noncycling (SB)2 100 8.3a 19.0a,b 40.0b 50.0b
% noncycling (SSP)2 100 10.0a 18.9a,b 40.0b 50.0b
Unknown+NM3 (SB) 2 48 32 25 8
Unknown+NM3 (SSP) 2 2 15 11 7
% Unknown+NM (SB) 66.7 66.7 42.2 55.6 66.7
% Unknown+NM (SSP) 66.7 9.1 28.8 42.3 48.3
1Number of elephants in each population.
2Percentage of elephants being hormonally monitored.
3Unknown? no survey returned; NM? not hormonally monitored.
a,bPercentages of noncycling females between each age category with different superscripts are
signi?cantly different (Po0.05). Analysis does not include females r10 years of age.
Elephant Reproductive Status in North America 315
normal tract morphology (Asian: 37.071.8 years vs. 27.071.5 years, respectively;
African: 26.371.6 years vs. 21.870.7 years, respectively). A comparison of
elephants for which both ultrasound and hormonal data were available revealed
that a greater proportion (Po0.05) of cycling females had normal reproductive-tract
morphology compared to noncycling elephants of both species. Overall, nearly 80%
of cycling females had normal reproductive-tract morphologies, whereas B70% of
noncycling females exhibited some type of ovarian or uterine pathology.
DISCUSSION
Based on this ?rst Elephant TAG/SSP reproductive survey, 49% of Asian and
62% of African elephant females in North America were being hormonally
monitored for estrous cyclicity. If one assumes that the facilities that did not return
surveys were not hormonally monitoring females, the hormonal monitoring
rate drops to 40% and 44% for Asian and African elephants, respectively.
The total numbers of hormonally assessed females in the different age categories
re?ected the age distribution of the population: more Asian elephants
in the 31?40-year age group were being monitored, and more African females in
the 21?30-year age group were being monitored.
These results con?rm earlier observations that many females in the captive
population are not cycling normally [Brown, 2000]. The problem is particularly
signi?cant for African elephants: 29% of hormonally-monitored females exhibited
some form of ovarian dysfunction, and many of these (18%) were of reproductive
age (i.e., o30 years old). Given the historically low reproductive rates of captive
elephants, and the need to increase fecundity to prevent demographic extinction, the
Elephant TAG/SSP now recommends that all elephant-holding facilities should
begin an active program of monitoring through routine hormone and ultrasound
assessments of both males and females [Keele and Ediger, 1997?2002]. Currently, the
Table 4. Numbers of elephants with transrectal reproductive tract ultrasound data from the
Elephant TAG/SSP 2001?2002 Reproductive Survey for captive Asian and African elephants in
North America
Asian African
Category1 Cycling
(n? 47)
Noncycling
(n? 6)
NM2
(n? 17)
Cycling
(n? 35)
Noncycling
(n? 11)
NM2
(n? 5)
Normal 35a 1b 13 27a 4b 3
Ovarian cysts 1 0 0 0 3 0
UT cysts/tumors 5 2 1 2 4 1
UT/VA cysts/tumors 2 2 0 3 0 0
UT cysts/other 1 0 1 0 0 0
Other pathologies3 3 1 2 2 0 1
1Some females had pathologies in more than one category; thus, some categories include
multiple items. A female is represented only once in a category.
2NM? not hormonally monitored.
3Unspeci?ed pathologies of reproductive origin.
Terms: UT? uterus; VA? vagina.
a,bDifferences in numbers within species for normal pathology between cycling and noncycling
females with different superscripts are signi?cant (Po0.05).
316 Brown et al.
TAG/SSP makes breeding recommendations only for females that have undergone
an ultrasound examination and a year of hormonal evaluations. However, even if
there are no immediate plans for breeding, or a female is considered to be
postreproductive, the TAG/SSP recognizes that lifelong assessments will be critical
to identify the causes of reproductive dysfunction. Until recently, it was believed that
elephants exhibited either a cyclic or a noncyclic progestagen pro?le. However, as
more elephants have been evaluated for longer periods of time, it appears that some
alternate between cyclic and noncyclic periods [Schulte et al., 2000], or exhibit
erratic progestagen secretion. Therefore, evaluations of reproductive data collected
through periodic TAG/SSP surveys could serve as valuable tools to help managers
identify and understand the factors that impact reproductive health [Keele and
Brown, 1998].
The urogenital tract pathologies identi?ed in the survey animals included cysts,
polyps, and tumors of the uterus, vagina, vestibule, and ovary, and (as reported in
previous studies) were more prevalent in older individuals. For example, in
nulliparous cows 430 years of age, vaginal cysts can become so extensive that
they ?ll the lumen [Hildebrandt et al., 2000]. Older African elephants are prone to
develop endometrial hyperplasia [Hildebrandt et al., 1997, 2000], whereas Asian
elephants develop multiple benign uterine tumors (leiomyomas) in the myometrium
[Hildebrandt and Go?ritz, 1995]. Follicular cysts on the ovary tend to occur more
frequently in African (B15%) than in Asian (B5%) elephants ex situ, but rarely
occur in free-ranging females (o1%) [Hildebrandt and Go?ritz, 1995; Hildebrandt et
al., 2000]. It is not known whether any of these pathologies have a direct impact on
fertility or cyclicity status, or whether they are a consequence of other intrinsic or
extrinsic factors that inhibit ovarian activity. Based on the age distribution of the
females examined to date, the types of pathologies identi?ed, and the historical
calving records of elephants in captivity, there appears to be a window of 10?15 years
from the onset of estrous cyclicity until a decline in reproductive ?tness is observed?
particularly in nulliparous cows [Hildebrandt et al., 2000]. This suggests that the
continuous cyclicity of nonbred females may have a negative and cumulative effect
on reproductive health. In the wild, females are either pregnant or lactating, and
therefore they experience comparatively few reproductive cycles in their lifetime.
More noncycling elephants exhibited reproductive-tract pathologies compared
to the normal cycling cohorts, and again these tended to be in the older age
categories. However, a number of normal-cycling females also presented urogenital
tract pathologies. Conversely, a few acyclic females were categorized as having
normal ultrasound results. In other species, urogenital cysts and tumors are not
always associated with acyclicity, although ovarian follicular cysts often result in
infertility [Meirow et al., 1993; Hamilton et al., 1995; Garverick, 1997]. Several
noncycling elephants in this survey had ovarian cysts. Treatment options for
conditions such as ovarian cysts do exist for other species. For example, previous
studies [Seguin et al., 1976; Kesler et al., 1981] have shown that 80% of cystic cows
(Bos taurus) respond to exogenous gonadotropin-releasing hormone (GnRH) or
human chorionic gonadotropin (hCG) treatment, with a resumption of normal cyclic
activity. The administration of hCG and GnRH to an African elephant with an
ovarian cyst failed to initiate reproductive cyclicity [Brown et al., 1999]; however,
more work is needed to determine the potential ef?cacy of these therapies for
elephants.
Elephant Reproductive Status in North America 317
The dif?culty of interpreting current ultrasound results as they relate to fertility
is that the data are still limited. Only a third of surveyed females had undergone an
ultrasound examination, and corresponding hormone data were not available for all
of them. Even fewer females with known reproductive-tract pathologies have been
exposed to bulls for breeding, which makes it dif?cult to assess the impact of these
pathologies on conception or gestation. However, a recent ?nding of multiple
endometrial cysts in an Asian elephant that produced a calf after arti?cial
insemination demonstrates that not all pathologies prevent successful pregnancy
[Brown et al., in press].
The ?nding that reproductive problems are at least in part age-related is
particularly relevant given current population age structures, especially for Asian
elephants. In the wild, elephants can reproduce into their 50s. Reproduction in
elephants generally begins soon after puberty and continues throughout most of
their life, with an average interbirth interval of 4?6 years [Sukumar, 1989]. However,
in captivity few elephants are given the opportunity to breed at regular intervals.
Consequently, historical studbook data suggest that animals that are not bred until
they are >30 years of age experience reduced fecundity and higher incidences of
stillbirths and dystocias [Doyle et al., 1999]. At present, only B30% of Asian
elephant females are o30 years of age [Keele, 2000]. The age structure of African
elephants is better, with B70% between 10 and 30 years of age [Olson, 2001].
However, only about one-third of reproductive-age African females are managed by
the SSP and thus likely to be included in any national breeding effort. Given these
demographics, the Elephant TAG/SSP has stepped up efforts to breed all
reproductively normal females (based on hormone and ultrasound examinations)
between 13 and 25 years of age. Future management plans will necessitate the
breeding of females regularly beginning at a young age before pathologies develop. It
is estimated that about six female offspring will have to be produced annually per
species to sustain current levels. Even so, the need to import replacement females
from range countries likely cannot be avoided, at least for the near future.
Ultimately, more physiological studies are needed to determine cause-and-
effect relationships between reproductive pathologies and fertility. These might
include analyses of the nutritional, disease, and health status of individual elephants,
in addition to investigations of other potential endocrine abnormalities as they relate
to ultrasound and cyclicity results. However, it also is possible that not all
reproductive problems have an organic cause?at least directly. Thus, additional
research should focus on the extent to which reproductive acyclicity is associated
with behavioral or husbandry factors related to captivity. One possibility is that the
social structure of captive ??herds?? leads to reproductive suppression. This is a
common strategy used by many social mammalian species (e.g., primates [Abbott,
1984; Epple and Katz, 1984], wolves [Packard et al., 1985], mongooses [Creel et al.,
1993], and naked mole-rats [Faulkes et al., 1990; Smith et al., 1997]) to ensure
reproduction by speci?c individuals. Reproductive inhibition may involve the delay
or prevention of ovulation or conception by hormonal disruptions of the
hypothalamo-pituitary-gonadal axis through behavioral or chemical signals [Abbott,
1984; Epple and Katz, 1984; Faulkes et al., 1990; Creel et al., 1993]. It can also
involve behavioral interactions, sometimes aggressive, of dominants over subordi-
nates by either sex to limit breeding opportunities [Abbott, 1984; Creel et al., 1993].
In elephants, there is evidence from wild populations that higher-ranking females
318 Brown et al.
can suppress reproduction in less dominant individuals [Dublin, 1983]; however, the
mechanism is not clear. Temporary cessation of estrous cyclicity in elephants may be
a reproductive strategy to ensure that females in poorer condition have lower fertility
rates during periods of limited natural resources. Alternatively, conception may be
delayed in order to prevent con?ict with dominant individuals or to enable an animal
to assist in allomothering [Dublin, 1983; Schulte et al., 2000]. Given how different
the captive environment is socially and physically from that in the wild, studies are
needed to assess how all aspects of management impact elephant health and,
ultimately, reproduction.
To summarize, enough noncycling elephant females have now been identi?ed
to warrant concern about the reproductive health of the captive population. The fact
that many of these females are of reproductive age adds to this concern. However, it
is proving dif?cult to objectively evaluate the problem because many captive
elephant females are not being adequately monitored through routine progestagen
and ultrasound analyses. The importance of initiating a continual reproductive
monitoring program, preferably throughout an elephant?s lifespan, cannot be
overemphasized. It is also important to conduct periodic surveys to determine how
(or whether) the reproductive status of the captive population is changing. The goal
now is to conduct these surveys each time the studbooks are updated (about bi-
yearly). Alterations in ovarian cyclicity or reproductive-tract pathologies could then
be based on differences in results between subsequent surveys, providing a
mechanism by which one could study longitudinal patterns in individual elephants
as well as the population as a whole. This effort is absolutely critical if we expect to
determine the prevalence of reproductive problems, identify causes, and assess
treatment effectiveness. Future studies will attempt to identify causes of reproductive
problems in captive elephants by investigating the impact of the captive
environment, including social and management factors, on physiology and behavior.
It is hoped that we will learn what factors are related to reproductive dysfunction in
captive elephant females in time to undertake mitigating actions to prevent the
extinction of these animals in captivity.
CONCLUSIONS
1. In a 2001?2002 survey, about 49% of Asian and 62% of African elephants in
North America were being monitored for estrous cyclicity by the use of weekly
serum or urinary progestagen analyses.
2. Of the elephants that were hormonally monitored, 14% of Asian and 29% of
African elephant females either were not cycling normally or exhibited irregular or
no ovarian activity.
3. Reproductive-tract transrectal ultrasound examinations revealed that
pathologies were more prevalent in the noncycling females; however, the impact of
these pathologies on fertility is not clear.
4. Reproductive problems were more prevalent in the older age categories,
although a number of acyclic African females were in the younger age groups.
Consequently, elephant females should be bred at a young age, before pathologies
develop.
Elephant Reproductive Status in North America 319
5. Identifying the causes of reproductive problems in captive elephants will
require continual investigations into the impact of the captive environment,
including social and management factors, on physiology and behavior.
ACKNOWLEDGMENTS
We thank all participants of the Elephant TAG/SSP for their support and
active assistance with this project. This study would not have been possible without
the cooperation of all the zoological institutions that completed their surveys. We are
equally grateful to institutions that continually monitor the reproductive health
status of the elephants in their collections. Endocrine data for many of the surveyed
elephants were obtained through the efforts of the excellent technical staff (Sue
Walker, Astrid Bellem, David Kersey, Rachel Moreland, Nicole Abbondanza,
Jordana Meyer, and Abby Mathewson) at the Conservation & Research Center. We
also thank Norie Ediger (Oregon Zoo), Tanya Moeller (CRC), and Noel Freeman
for their help in collecting, collating, and summarizing the survey results. The
ultrasound examinations were performed by Dennis Schmitt (Southwest Missouri
State University) and Thomas Hildebrandt (Institute for Zoo Biology and Wildlife
Research, Berlin).
REFERENCES
Abbott DH. 1984. Behavioural and physiological
suppression of fertility in subordinate marmoset
monkeys. Am J Primatol 6:169?86.
Brown JL. 2000. Reproductive endocrine monitor-
ing of elephants: an essential tool for assisting
captive management. Zoo Biol 19:347?68.
Brown JL, Goeritz F, Pratt-Hawkes N, Hermes R,
Galloway M, Graham LH, Gray C, Walker SL,
Gomez A, Moreland R, Murray S, Schmitt DL,
Howard JG, Lehnhardt J, Beck B, Bellem A,
Montali R, Hildebrandt TB. Successful arti?cial
insemination of an Asian elephant at the
National Zoological Park. Zoo Biol (in press).
Brown JL, Hildebrandt T, Theison W, Neiffer DL.
1999. Endocrine and ultrasound evaluation of a
non-cycling African elephant: identi?cation of a
follicular ovarian cyst. Zoo Biol 18:223?32.
Doyle C, York B, Whitely A. 1999. A survey
of Asian elephant births from 1962?1998. J
Elephant Managers Assoc 10:146?8.
Dublin HT. 1983. Cooperation and reproductive
competition among female African elephants. In:
Wasser SK, editor. Social behavior of female
vertebrates. New York: Academic Press. p 291?313.
Epple G, Katz Y. 1984. Social in?uences on
estrogen excretion and ovarian cyclicity in saddle
back tamarins (Saguinus fuscicollis). Am J
Primatol 6:215?27.
Faulkes CG, Abbott DH, Jarvis JUM. 1990.
Social suppression of ovarian cyclicity in captive
and wild colonies of naked mole-rats, Hetero-
cephalus glaber. J Reprod Fertil 88:559?68.
Garverick HA. 1997. Ovarian follicular cysts in
dairy cows. J Dairy Sci 80:995?1004.
Hamilton SA, Garverick HA, Keisler DH, Xu ZZ,
Loos K, Youngquist RS, Salfen BE. 1995.
Characterization of ovarian follicular cysts and
associated endocrine pro?les in dairy cows. Biol
Reprod 53:890?8.
Hildebrandt TB, Go?ritz F. 1995. Transrectal
ultrasonography for ovary and pregnancy in
Indian elephant. Verh Erkrg Zoo 37:261?8.
Hildebrandt TB, Go?ritz F, Pratt NC, Schmitt DL,
Lehnhardt J, Hermes R, Quandt S, Raath J,
West G, Montali RJ. 1997. Assessment of health
and reproductive status in African and Asian
elephants by transrectal ultrasonography. Proc
Am Assoc Zoo Vet 207?12.
Hildebrandt TB, Go?ritz F, Pratt NC, Brown JL,
Montali RJ, Schmitt DL, Fritsch G, Hermes R.
2000. Ultrasonography of the urogenital tract in
elephants (Loxodonta africana and Elephas max-
imus): an important tool for assessing female
reproductive function. Zoo Biol 19:321?32.
Keele M, editor. 2000. Asian elephant North Amer-
ican regional studbook. Portland, OR: Oregon Zoo.
Keele M, Brown J. 1998. A letter to elephant
institutional representatives: urgent management
need. J Elephant Managers Assoc 9:101?2.
Keele M, Ediger ND-E, editors. 1997?2002. AZA
elephant masterplan. AZA publication number
1999 10285.zooe. Portland, OR: Oregon Zoo.
Kesler DJ, Elmore RG, Brown EM, Garverick
HA. 1981. Gonadotropin releasing hormone
320 Brown et al.
treatment of dairy cows with ovarian cysts. I.
Gross ovarian morphology and endocrinology.
Theriogenology 16:207?13.
Meirow D, Laufer N, Schenker JG. 1993. Ovula-
tion induction in polycystic ovarian syndrome: a
review of conservative and new treatment mod-
alities. Eur J Obstet Gynecol Reprod Biol
50:123?31.
Olson D, editor. 2001. African elephant North
American regional studbook. Indianapolis, IN:
Indianapolis Zoo.
Olson D, Wiese RJ. 2000. State of the North
American African elephant population and
predictions for the future. Zoo Biol 19:
311?20.
Packard JM, Seal US, Mech LD, Plotka ED. 1985.
Causes of reproductive failure in two family
groups of wolves. Z Tierpsychol 68:24?40.
Schulte BA, Feldman E, Lambert R, Oliver R,
Hess DL. 2000. Temporary ovarian inactivity in
elephants: relationship to status and time out-
side. Physiol Behav 71:123?31.
Seguin BE, Convey EM, Oxender WD. 1976.
Effect of gonadotropin-releasing hormone and
human chorionic gonadotropin in cows with
ovarian follicular cysts. Am J Vet Res 37:153?60.
Smith TE, Schaffner CM, French JA. 1997. Social
and developmental in?uences on reproductive
function in female Wied?s black tufted-ear
marmosets (Callithrix kuhli). Horm Behav
31:159?68.
Sukumar R. 1989. The Asian elephant: ecology
and management. Sukumar R, editor. Cam-
bridge, UK: Cambridge University Press. 251 p.
Wiese RJ. 2000. Asian elephants are not self-
sustaining in North America. Zoo Biol 19:299?310.
Elephant Reproductive Status in North America 321