Animal Reproduction Science 124 (2011) 155–162

Contents lists available at ScienceDirect

Animal Reproduction Science

journal homepage: www.elsevier.com/locate/anireprosci

Female reproductive cycles of wild female felids夽

Janine L. Brown ∗
Center for Species Survival, Department of Reproductive Sciences, Smithsonian Conservation Biology Institute, Smithsonian’s National Zoological Park,
1500 Remount Road Front Royal, VA 22630, USA

a r t i c l e i n f o a b s t r a c t

Article history: Many felid species are endangered because of destructive human activities. As a result,
Available online 3 September 2010 zoos are being tasked with sustaining genetically healthy populations in case of catas-
trophic extinctions. Unfortunately, with the exception of a few species, most felids do not
Keywords: reproduce well in captivity. The ability to track reproductive activity via hormones is key
Gonadal steroids to developing successful ex situ breeding programs. Through the development of nonin-
Noninvasive monitoring
vasive fecal hormone monitoring techniques, a high degree of variability in estrous cycle
Fecal hormones
characteristics has been found to exist across the taxon, including the type of ovulation.
Ovarian cycle
Pregnancy For example, although all felids have induced ovulations, the occurrence of spontaneous
ovulations varies across species, and even between individuals within a species. Clouded
leopards, fishing cats and margays frequently have spontaneous ovulations, whereas these
are rarely observed in the cheetah, tigrina and ocelot. There are marked species differences
in the impact of season on reproductive function, with some being exquisitely sensitive
to photoperiod (e.g., Pallas’ cat), some moderately affected (tiger, clouded leopard, snow
leopard), and others that are not influenced at all (e.g., ocelot, tigrina, margay, lion, leopard,
fishing cat). One of the greatest challenges remaining is overcoming the problems asso-
ciated with highly variable ovarian responses to ovulation induction therapies used with
assisted reproductive procedures, like artificial insemination (AI). Success is relatively high
in the cheetah and ocelot, but few pregnancies have resulted after AI in clouded leopard,
fishing cat and tiger. Current knowledge of the reproductive physiology of nondomestic
felids, including aspects of the anatomy, behavior and ovarian cycles will be presented, and
how the rapidly growing endocrine database is aiding ex situ management efforts.
Published by Elsevier B.V.

1. Introduction species population declines in the wild are habitat loss

Twenty-three of the 36 known wild cat species are
threatened or endangered with extinction in at least some
part of their natural range (www.iucnredlist.org). Felids
are placed into small (<6.5 kg, n = 18 species), medium
(7–20 kg, 11 species) and large (35–135 kg, 7 species) size
categories, based on body weight. The major causes of
夽 This paper is part of the special issue entitled: Reproductive Cycles of
Animals, Guest Edited by Michael G. Diskin and Alexander Evans.
∗ Corresponding author.
E-mail address: brownjan@si.edu.
and fragmentation, human-animal conflict, and for some
species, poaching for pelts and traditional medicines. While
declines in wild felid numbers are significant, cats in gen-
eral have the advantage of being highly charismatic and
popular zoo exhibit animals. Yet, despite a long history in
captivity, a significant challenge continues to be determin-
ing how to manage these generally secretive animals in
environments that allow them to thrive and reproduce,
while at the same time satisfying the public’s viewing
needs. Even in relatively naturalistic enclosures, reproduc-
tion in captive felids remains poor, with the exception
of some of the Panthera species. Because reproduction is
key to species survival, and hormones drive this process,

0378-4320/$ – see front matter. Published by Elsevier B.V.

doi:10.1016/j.anireprosci.2010.08.024
156 J.L. Brown / Animal Reproduction Science 124 (2011) 155–162
understanding the basic reproductive endocrinology of
felids is vital for their conservation and management.
The recognition over two decades ago that steroid
metabolites could be measured in urine and feces revolu-
tionized our ability to track hormone status in wild species.
For most, the decision to measure fecal or urinary hor-
mones is driven by which material is easiest to collect.
However, for felids, nearly all gonadal steroids are excreted
in feces, so urinary analyses are not a reliable option (see
review, Brown, 2006). In the 15 years since fecal steroid
techniques were validated for wild felids, over 200 papers
have been published that describe gonadal cycle patterns
for just over half of the species. Thus, while an impressive
endocrine database now exists, and hormone monitoring
has played a major role in the breeding management of
many felid species, there obviously is still much left to do.
2. Anatomy of the reproductive tract
The ovaries are caudal to the kidneys and attached prox-
imally by the suspensory ligament and dorsally by the
mesovaria. The mesosalphinx covers the oviducts and lat-
eral aspect of the ovaries to form an ovarian bursa. Each
oviduct tracks cranially in the medial aspect of the ovarian
bursa, then caudally to the lateral aspect before terminat-
ing at the uterotubal junction. The uterus is bicornuate and
horns are suspended dorsally by the mesometrium. The
uterine body is partially divided internally by a septum,
and the cervix is short, opening at an angle at the vagi-
nal orifice. The lips of the vulva are located just below the
anus. After fertilization, blastocysts become evenly spaced
throughout the uterine horns for litter bearing felids, with
efficient transuterine migration. Cats have an endothelio-
chorial type, zonary placenta.
3. Noninvasive steroid monitoring
Injection of radiolabelled estradiol-17␤ and proges-
terone into domestic cats revealed that 95% of metabolites
are excreted in feces within 1–2 days (see review, Brown,
2006). High performance liquid chromatography, gas
chromatography/mass spectrometry and immunoreactiv-
ity analyses have determined that: (1) fecal estrogens
consist primarily of unconjugated and conjugated (3-
sulfate, 17␣-sulfate and 17␤-sulfate) estradiol, with minor
amounts of estrone and estriol; and (2) progesterone is
metabolized and excreted mostly as polar compounds
(>75%), probably conjugates, and some unconjugated preg-
nanes. Extracting steroid metabolites from feces generally
involves boiling, vortexing or shaking samples in combina-
tions of organic solvents (e.g., ethanol, methanol), usually
containing at least 10% water or buffer to assist in recovery
of conjugated steroids. There is no measure for indexing
fecal data as there is for urine (i.e., creatinine); however,
sample variability can be reduced by drying feces and mix-
ing the powder before extraction. Reasonable data can
be obtained, however, using well-mixed wet fecal sam-
ples if animal diets and health remain fairly constant.
Finally, it is important to keep in mind that the follow-
ing variables can significantly affect steroid metabolite
concentrations: age of sample after defecation, environ-
mental exposure, sample volume and consistency, and
sample homogeneity (Wielebnowski and Watters, 2007).
These effects may not be consistent across species or hor-
mones; thus, each assay and fecal handling technique must
be properly validated using both laboratory and biological
tests.
4. Estrous cycle
There are four phases of the cat estrous cycle: proestrus,
estrus, diestrus and anestrus (or interestrus) (see review,
Wildt et al., 1998). Proestrus usually lasts less than a
day, and is associated with the presence of ovarian fol-
licles, increasing circulating estrogens and occasionally
male interest excluding copulation. Estrus accompanies
advanced follicular development and peak concentrations
of estradiol and is characterized by male mounting and
coitus, and depending on the species, behaviors such
as vocalization, lordosis, rolling, rubbing and foot tread-
ing. Multiple intromissions, often over several days, are
believed necessary in most felids to stimulate the release of
gonadotropin-releasing hormone (GnRH) from the medial
basal hypothalamus and subsequent surges of luteinizing
hormone (LH) from the anterior pituitary gland (Shille et
al., 1983; Johnson and Gay, 1981). This cascade of events
causes final follicular and oocyte maturation and results in
ovulation after mating. During diestrus, one or more cor-
pora lutea (CL) produce progesterone that stays elevated
for varying lengths of time regardless of whether concep-
tion occurs or not. Anestrus is the period in between waves
of follicular development, when circulating estrogens are
at nadir.
Reproductive ovarian steroid cycle patterns now have
been published for about half of the nondomestic felid
species, with fecal steroid analyses being used in the vast
majority. Surges of estrogens distinguish estrous from
interestrus periods, with cycles ranging from 2 to 4 weeks
and estrus lasting 3–10 days (e.g., Fig. 1). Cats have his-
torically been categorized as induced ovulators – requiring
mating to stimulate ovulation. However, we now know that
felids exhibit a range of ovulatory patterns, from almost
exclusively induced to varied combinations of induced
and spontaneous. These differences occur not only across
species, but also between individuals within a species (see
Brown, 2006; Göritz et al., 2009). Indeed, spontaneous
increases in progestagens after estrogen surges are nonex-
istent or rare in the tiger (Panthera tigris), puma (Felis
concolor), snow leopard (Panthera uncia), cheetah (Acinonyx
jubatus), tigrina (Leopardus tigrinus), ocelot (Leopardus
pardalis) and lynx (Lynx pardinus, Lynx canadensi, Lynx
lynx), but occur, at least occasionally, in the lion (Pan-
thera leo), leopard (Panther. pardus), Pallas’ cat (Otocolobus
manul), fishing cat (Prionailurus viverrinus), and regularly
in the clouded leopard (Neofelis nebulosa), margay (Leopar-
dus wiedii) and domestic cat. In some species, non-mating
induced ovulations are more prevalent when females are
housed together, whereas in others they also can occur in
singletons. Thus, within this taxon ovulatory mechanisms
vary, regulated to a greater or lesser degree by species and
perhaps individual-specific responses to physical and/or
psychosocial stimuli.
 J.L. Brown / Animal Reproduction Science 124 (2011) 155–162 157

4000 Ocelot 100

Fecal Progestagens (μg/g)

Fecal Estrogens (ng/g)
3000 75

2000 50

1000 25

0 0
0 20 40 60 80 100 120 140

2000 Tigrina 100

Fecal Progestagens (μg/g)

Fecal Estrogens (ng/g)

1500 75

1000 50

500 25

0 0
0 20 40 60 80 100 120 140

3000 Margay 400

Fecal Progestagens (μg/g)

Fecal Estrogens (ng/g)

2000 300

200
1000

100
500

0 0
0 20 40 60 80 100 120 140

Days of Sampling

Fig. 1. Longitudinal profiles of fecal estrogens (open triangle) and progestagens (closed circle) in a singleton ocelot (top panel), tigrina (middle panel) and
margay (bottom panel). Ocelot and tigrina females exhibit regular follicular cycles indicative of induced ovulators. In the ocelot, fecal progestagens covary
with estrogens during the follicular phase. Successive nonpregnant luteal phases are observed in the margay after estrogen surges, which are indicative of
spontaneous ovulations.

5. Reproductive seasonality Reproduction is at least somewhat seasonal in many

We know from extensive studies that the domestic cat
is seasonally polyoestrous under natural photoperiods, but
will cycle year round when housed on a 12–14 h light
cycle (Shille et al., 1979). In general, ovarian cyclic activ-
ity is reduced under decreasing photoperiod and resumes
after exposure to increasing light (i.e., a long-day breeder).
Melatonin appears to regulate photoperiod-induced sea-
sonality in the cat, with concentrations being highest
during the dark phase (Leyva et al., 1989). Follicular devel-
opment and estrous cyclicity can effectively be suppressed
in cats by oral melatonin administered several hours before
lights-off during a stimulatory photoperiod; i.e., a regimen
that mimics short-day melatonin secretion (Graham et al.,
2004).
non-domestic felids like the tiger, clouded leopard, Pallas’
cat, lynx and snow leopard. By contrast, follicular activity
in lions, leopards, bobcats, pumas, margays, ocelots, tigri-
nas, jaguars and fishing cats is not influenced by season
(Brown, 2006). Like the domestic cat, photoperiod plays a
role in modulating reproduction in seasonal nondomestic
felids, which likely also involves melatonin control.
6. Pregnant and nonpregnant luteal phases
Progestagen concentrations during pregnant and non-
pregnant luteal phases are quantitatively similar in
nondomestic felids as in domestic cats (Wildt et al., 1998).
In general, length of the nonpregnant luteal phase is about
one-third to one-half that of pregnancy. Luteal secretion
158 J.L. Brown / Animal Reproduction Science 124 (2011) 155–162
of progesterone is dependent on LH and prolactin from the
anterior pituitary. The placenta may also produce progesta-
gens because in the domestic cat, removal of the ovaries
after mid-gestation does not terminate pregnancy. Estro-
gens increase after mid-gestation in some species (e.g.,
cheetah, Pallas’ cat, fishing cat, domestic cat), but not in
others (e.g., clouded leopard, tiger) (Brown, 2006). From
a reproductive management standpoint it is technically
possible to diagnose pregnancy based on fecal progesta-
gen concentrations that remain elevated past the normal
length of a non-pregnant luteal phase. Measurement of
prolactin and relaxin also can be used diagnostically
provided appropriate samples are collected. Circulating
prolactin increases at about Day 35 in the domestic cat
and again just before parturition (Banks et al., 1983). Ges-
tational prolactin is luteotropic as it has been shown that
administering the dopamine agonist, cabergoline, after
mid-gestation decreases progesterone and causes fetal
absorption or abortion (Tsutsui and Stabenfeldt, 1993;
Jochle and Jochle, 1993). Circulating relaxin from the feto-
placental unit increases about Day 20 and continues to
rise throughout gestation before declining around the time
of parturition (Stewart and Stabenfeldt, 1985). This hor-
mone is responsible for softening pelvic connective tissue
to facilitate parturition (Tsutsui and Stabenfeldt, 1993).
Relaxin and prolactin can be dignostic of pregnancy after
mid-gestation because high concentrations are not found
in nonpregnant cats. Though most effective using serum or
plasma, a commercial canine relaxin RIA (Synbiotics Corpo-
ration, San Diego, CA) has been used to diagnose pregnancy
during late pregnancy in the domestic cat and leopard
using urine (De Haas van Dorsser et al., 2006). Even more
recently, a simple strip test from the same company (Wit-
ness Relaxin® ) was moderately effective in diagnosing late
pregnancy using domestic cat urine, but only if diluted 1:1
in nonpregnant cat serum to reduce interference (De Haas
van Dorsser et al., 2007). Similarly, the Witness Relaxin®
test has shown some positive results (though not 100%
effective) in Iberian lynx urine samples that were concen-
trated by ultrafiltration and brought up in serum (Braun
et al., 2009). Cats do not appear to produce a chorionic
gonadotropin.
7. Ovarian control for assisted reproduction
Artificial insemination (AI) and in vitro fertiliza-
tion/embryo transfer (IVF/ET) are powerful tools for
managing rare populations (see review, Pelican et al.,
2006). Successful ovarian stimulation for AI and IVF/ET
has been achieved in at least one-third of all cat species,
but results have been mixed and pregnancy success rates
are <20% in most species. The main problem is the vari-
able responses and sensitivities to gonadotropin treatment,
especially in species that exhibit spontaneous ovulations.
The gonadotropins eCG and hCG are typically used to
stimulate follicular development and induce ovulation,
respectively. Most wild felids have ovulations 37–42 h after
hCG, and AI success is only possible if inseminations are
conducted post-ovulation due to interfering effects of anes-
thesia (Howard et al., 1992). It is not possible to extrapolate
gonadotropin doses from one species to another based
on body size only. For some, despite decades of work,
appropriate dose combinations that do not over or under
stimulate the ovaries have not been identified (Graham
et al., 2006; Pelican et al., 2006). Furthermore, current
gonadotropin regimens appear to perturb normal female
reproductive function and reduce fertility (Roth et al.,
1997). For example, there now is considerable evidence
that standard eCG/hCG protocols cause ovarian hyperstim-
ulation, resulting in concentrations of estrogen that are
several-fold greater than those observed during natural
estrus (Roth et al., 1997; Pelican et al., 2006). The etiol-
ogy of excessive estrogen production after gonadotropin
treatment appears to be due, in part, to the development
of ancillary ovarian follicles. Hyper-estrogenism creates an
abnormal endocrine environment that is detrimental to
fertilization, embryogenesis and implantation. One excep-
tion is the cheetah, where fecal estrogen concentrations
after gonadotropin ovulation induction are not different
from those associated with natural estrus (Brown et al.,
1996). It is in this species that pregnancy success after
AI is among the greatest for felids (∼50%). Other regi-
mens now are under study, including use of pFSH, pLH and
various GnRH analogs to induce follicular maturation and
ovulation. However, to date none have proven routinely
successful. Thus, there remains a clear need to develop bet-
ter hormonal protocols that will result in more predictable
ovarian responses in felids. One proposed approach that
is beginning to show promise, though no pregnancies yet,
is the use of synthetic progestins or GnRH analogs to
temporarily shut-down the ovary prior to gonadotropin
stimulation (Pelican et al., 2006).
8. Comparative reproductive profiles of selected
felid species
The following examples highlight the diversity of
female reproductive mechanisms exhibited by felids.
Endangered status is based on IUCN Redlist categories
(www.iucnredlist.org).
8.1. Domestic cat
The domestic cat is considered a small-sized felid.
Females reach puberty between 4 and 12 months of age
(average 6–9 months), at approximately 75% of their body
weight (Tsutsui and Stabenfeldt, 1993; Wildt et al., 1998).
Short-haired breeds tend to reach puberty at an earlier age
than long-haired cats. Pregnancy lasts 58–65 days with the
length of the nonpregnant luteal phase being 35 and 40
days. Domestic cats are seasonally polyestrus, cycling at
∼21-day intervals and breeding during longer days (Jan-
uary and September). Though traditionally classified as an
induced ovulator (Shille et al., 1983), it is now recognized
that many individuals have ovulations regularly without
mating. During gestation, both prolactin and relaxin are
elevated, distinguishing it from a false pregnancy (Banks
et al., 1983; Stewart and Stabenfeldt, 1985). Females are in
estrus for 3–16 days, during which time they display clear
mating behaviors, e.g., calling, rolling, urinating. Females
assume a lordosis posture, head and front legs down, rump
held high, during mating. Kittens are weaned after ∼8
 J.L. Brown / Animal Reproduction Science 124 (2011) 155–162
weeks, thus permitting a female to produce up to three lit-
ters per year. Domestic cats have long served as the model
species for developing assisted reproductive techniques in
wild felids (Howard et al., 1992, 1996).
8.2. Cheetah
The cheetah is a medium-size species living in Africa
and is listed as vulnerable with a decreasing population
trend. Historically, this species has been known for its poor
reproductive performance in captivity. Cheetahs reach sex-
ual maturity at 2–3 years of age. Gestation averages 94
days, with nonpregnant luteal phases lasting ∼53 days.
Cheetahs exhibit comparatively short estrous cycles, which
range from 7 to 21 days in duration. Estrus lasts 2–6
days. One of the major impediments to successful cap-
tive propagation in the cheetah is the inability to reliably
identify when females are in estrus for timing of breed-
ing introductions. An extensive study of behavior and fecal
estrogen analyses revealed that while no single behavior
was indicative of estrus, higher rates of rubbing, rolling,
object sniff, chirping and urine spraying often correlated
with estrogen surges, though on a highly individual basis
(Wielebnowski and Brown, 1998). Periodic bouts of ovarian
inactivity typically are observed in this species, which are
known to be related to socially induced reproductive sup-
pression (Wielebnowski et al., 2002). Cheetahs in the wild
are solitary, yet many facilities house females in groups
because intra-species aggression is relatively infrequent.
One study found that although serious altercations were
rare, subtle antagonistic behaviors often were observed
(growl, hiss, chase, attack) along with increased stereotypic
pacing and reduced auto-grooming. Concurrently, average
fecal estrogen concentrations were less in the paired chee-
tahs, especially in subordinants, with little cyclic activity.
Separation of the pairs resulted in a rapid reinitiation of
normal behaviors and ovarian activity (Wielebnowski et al.,
2002). One possible explanation for this finding is stress-
induced ovarian suppression. Jurke et al. (1997) found that
two of seven zoo-held female cheetahs had high average
fecal corticoid concentrations and also were acyclic. How-
ever, there was no relationship between ovarian acyclicity
and increased adrenal corticoid activity in the study of
Wielebnowski et al. (2002). From a management stand-
point, knowledge that estrous cyclicity in cheetahs can
be socially suppressed even when aggressive interactions
are relatively minor has resulted in management changes
such that breeding females today are generally housed sep-
arately. In pregnant cheetahs, fecal estrogen metabolites
increase after about mid-gestation, though the source is
unclear. Brief estrogen surges often occur during the luteal
phase in both pregnant and nonpregnant females, sug-
gesting there may be some follicular activity even in the
presence of elevated progestagens (Brown et al., 1996).
8.3. Pallas’ cat
The Pallas’ cat, a small-sized felid of Central Asia, is
listed as near threatened with a declining population.
Females reach puberty at about a year of age, pregnancy
lasts 66–77 days and the non-pregnant luteal phase is
159
∼50 days. Estrous cycles in the Pallas’ cats are somewhat
difficult to distinguish because the species is highly sea-
sonal with females exhibiting ovarian activity for only ∼3
months of the year (January–March) (Brown et al., 2002).
Based on fecal estrogens, cycles averaged ∼14 days, but
the pattern is unusual in that inter-estrous baseline estro-
gen concentrations also are elevated during the breeding
season, perhaps as a result of overlapping follicular waves.
Based on fecal progestagen analysis and laparoscopic ovar-
ian evaluations, Pallas’ cats occasionally have ovulations
without mating. During gestation, both fecal progestagens
and estrogens are elevated, although the temporal profiles
differ (Brown et al., 2002). Progestagens increase within
days after breeding and peak about mid-term, whereas
estrogen excretion increases after mid-term and remains
elevated until just before parturition. The biphasic pro-
file of estrogen excretion, temporally distinct from that
of progestagens, is observed only during pregnancy and
not during the nonpregnant luteal phase, lending sup-
port to the possibility of a placental origin. Pallas’ cats are
exquisitely sensitive to photoperiod. A serendipidous dis-
covery was the observation of stimulated ovarian activity
in the nonbreeding season by a sudden transition from
‘short days’ to ‘long days’ during a month-long Festival
of Lights. Although animals eventually cycled during the
breeding season, they did not reproduce. It was not until
cats were moved to a different part of the zoo, away from
the Festival lights, that conception occurred (Brown et al.,
2002). To more closely examine the effect of photoperiod
on gonadal function, one Pallas’ cat pair was subjected to an
artificial short-day photoperiod (9.5 h of light/day) during
the summer non-breeding season. After allowing a 1.5-
month acclimatization period, light was increased 0.5 h per
week until it reached 12 h per day. After reaching the 12 h
photoperiod, the female cycled based on endocrinological
changes. Several attempts at using AI in Pallas’ cats have
resulted in no pregnancies, despite apparent normal ovar-
ian responses to eCG/hCG (Brown et al., 2002; Swanson,
2006).
8.4. Clouded leopard
The clouded leopard is a medium-sized felid that lives in
tropical Asia, and is listed as vulnerable with a decreasing
population trend. Females reach sexual maturity at about
2 years of age and exhibit a comparatively long estrous
cycle at 15–40 days, with a 3–6-day estrus. Gestation lasts
∼90 days, with nonpregnant luteal phases lasting 45–50
days. Breeding management of clouded leopards is chal-
lenging as most successful pairs result when individuals
are introduced before puberty. Otherwise, a male will often
attack or even kill the smaller female, even when she is
in estrus. Behavioral changes as females approach estrus
include marking, lordosis and producing a low intensity
chuffing sound called prusten. Clouded leopards exhibit
reproductive seasonality in western zoos; those exposed
to temperate photoperiods have brief periods of anestrous
in the late summer to early fall (Brown et al., 1995). How-
ever, if housed indoors under a long-day light cycle, ovarian
activity continues year round. Clouded leopards are one
of the cat species that exhibit spontaneous ovulations,
160 J.L. Brown / Animal Reproduction Science 124 (2011) 155–162
some females doing so quite regularly. In a study of 14
clouded leopards, 40% exhibited spontaneous ovulation in
the absence of mating (Brown et al., 1995). Because there
is no way to predict when a female will have spontaneous
ovulations, developing ovulation induction protocols for
use with AI has proven very difficult for this species, result-
ing in only one pregnancy despite dozens of attempts
(Howard et al., 1996). Unlike the cheetah, Pallas’ cat, fishing
cat and domestic cat, increased estrogen production dur-
ing the latter part of pregnancy in the clouded leopard is
not observed, indicating potential differences in placental
steroidogenic function (Brown, 2006).
8.5. Tiger
The tiger, a large-sized species living in tropical Asia,
is one of the most beautiful and charismatic of the felid
species. It is listed as endangered, with some subspecies
on the brink of extinction (e.g., Siberian – P. tigris altaica;
South China tiger – P. tigris amoyensis). In addition to habi-
tat destruction, tigers suffer at the hands of man through
poaching of bones and penises for traditional medicines.
Befitting their larger size, tigers reach puberty at 3–4 years
of age. Female cycles can be quite variable, occurring at
18–40-day intervals with estrus lasting about a week (Seal
et al., 1985; Wildt et al., 1998; Graham et al., 2006). Siberian
tigers have been reported to be seasonally polyestrus in
captivity, with peak activity from February through June
(Seal et al., 1985). Graham et al. (2006), however, found
more variation in fecal endocrine patterns among indi-
viduals (Siberian and Sumatran), with some unexplained
periods of acyclicity not related to season. Behavioral signs
of estrus correlate with increased estrogen production and
include prustening, rolling, rubbing and lordosis (Seal et al.,
1987). Gestation is ∼108 days, and the nonpregnant luteal
phase lasts ∼35 days. Like the clouded leopard, an increase
in estrogen excretion is not observed during pregnancy.
In contrast to most other felid species, tiger reproduce
well in captivity. By contrast, use of assisted reproduc-
tive techniques has met with limited success. Despite early
accomplishments in the production of a single litter of cubs
through the use of IVF (Donoghue et al., 1990) and two lit-
ters by AI (Donoghue et al., 1993), poor ovarian responses to
eCG/hCG ovulation induction have prevented further suc-
cesses despite dozens of attempts (Graham et al., 2006).
Like the domestic cat, excessive and sustained elevations in
estrogens as a result of chorionic gonadotropin treatment
appear to disrupt oviductal embryo transport (Graham
et al., 2006). Furthermore, extensive trials with varying
eCG/hCG doses could not identify a combination that
resulted in adequate follicular development and ovulation
without causing hyperestrogen production. Trying another
approach, Crichton et al. (2003) found >93% homology
between amino acid sequences of Siberian tiger FSH and
LH and porcine gonadotropins, and so used pFSH and pLH
in ovarian stimulation trials. While ovarian responsiveness
was high, even after repeated treatments indicating no
inhibitory antibody production, and there was no evidence
of estrogen hyperstimulation, luteal progestagen excretion
was reduced compared with natural matings. Clearly, con-
siderably more research is needed to identify alternative
protocols to induce normal ovarian activity in conjunction
with the use of assisted reproduction techniques in tigers.
8.6. Leopardis species
Of the three small-sized species inhabiting the Amer-
icas, the margay is listed as threatened, the tigrina or
oncilla is vulnerable, and the ocelot is of least concern,
with suspected decreasing populations. Historically, none
of these small cats breed well in captivity. Puberty occurs
at about 1–2 years of age, estrous cycles are 10–20 days,
and estrus lasts 2–7 days (Moreira et al., 2001) (Fig. 1).
Duration of gestation is 75–85 days and mating-induced
nonpregnant luteal phases are about 40 days in length.
Females rarely exhibit regular overt signs of sexual recep-
tivity, with the exception of ocelots that occasionally
display behaviors such as rubbing, vocalizing, rolling, urine
spraying and sniffing. Comparatively, ocelots and tigrinas
rarely ovulate spontaneously (Fig. 1, top and middle panel),
whereas non-mating induced luteal phases of ∼40 days
in duration are observed frequently in margays (Fig. 1,
lower panel). Ovarian activity is not seasonal in these
species, although unexplained periods of acyclicity often
are observed. Unlike the cheetah, these anestrous periods
do not appear to be caused by social suppression as they
have been documented in females housed alone (Moreira
et al., 2001). In the ocelot, fecal progestagens often are
increased in conjunction with estrogen surges, correlat-
ing as high as r = 0.4 (Fig. 1, top panel). To date, only two
other species have shown this covariance between ovarian
steroids – the jaguar (Brown, unpublished) and lynx (see
below). The progestagens are believed to be of follicular
origin, because concentrations are only a fraction of those
observed after ovulation (Moreira et al., 2001). Pregnan-
cies have resulted after AI (ocelot and tigrina) and IVF/ET
(ocelot), but success has been limited, again due to incon-
sistencies in responses to ovulation induction protocols
(Swanson, 2006).
8.7. Lynx
Lynx are medium-sized cats, with the Iberian or Span-
ish lynx being critically endangered and considered the
world’s most endangered felid. Canadian and European
lynx both are of least concern with stable populations;
however, the U.S. Fish & Wildlife Service lists the Cana-
dian species as vulnerable. In general, females reach
puberty at 1–2 years of age. These are seasonally breed-
ing species, with births occurring in mid to late spring
(∼January–April). Estrus lasts less than a week, with a
gestation duration of 60–70 days. Ovarian follicular and
luteal function has been difficult to quantify in lynx using
standard immunoassay techniques (Fanson et al., 2009;
Pelican et al., 2009; Denhard et al., 2008). Estrous cycle
length remins largely undetermined because evaluations
of estrogens do not reliably correlate with estrous behav-
ior. Although it was possible to detect estrogenic (ovarian)
surges at the time of copulatory events with daily fecal
collections some of the time (Pelican et al., 2009), a con-
sistent cyclic pattern similar to other felid species was not
evident. Instead, there is a seasonal pattern of estrogen
 J.L. Brown / Animal Reproduction Science 124 (2011) 155–162
excretion, with baseline concentrations increasing dur-
ing the breeding season, regardless of whether mating
or pregnancy ensues. Fecal progestagen excretion also is
variable throughout the year and not diagnostic of luteal
activity, again very different to data in other felid species.
Further analyses of urinary progesterone and pregnane-
diol similarly failed to show clear elevations in pregnant
Iberian and Eurasian lynx females, despite finding signif-
icant amounts of native progesterone in HPLC-separated
urinary eluates (Göritz et al., 2009; Jewgenow et al., 2009;
Braun et al., 2009). Adding to this frustration, results
of a radio-metabolism study indicated a progesterone
immunoassay is capable of detecting relevant metabolites
in feces (Denhard et al., 2008), so it is unclear why these
noninvasive methods do not produce typical steroidogenic
patterns. Another lynx characteristic is a significant co-
excretion of estrogen and progestagen metabolites (up to
r = 0.8), and a post-partum lactational elevation in gonadal
steroid hormones (Denhard et al., 2008; Pelican et al., 2009;
Jewgenow et al., 2009). An ultrasonographic study found
active CL in Iberian and Eurasian lynx after parturition,
which could account for these equivocal data during the
lactational anestrus period (Göritz et al., 2009; Denhard
et al., 2008). Authors theorize these CL may act to sup-
port lactation, prevent new follicular cycle activity and thus
restrict the breeding season to midwinter. Altered HPLC
elution profiles of radio-labelled progestagens, and a sig-
nificant shift in the proportion of 17␤-estradiol to estrone
before pregnancy to after parturition supports the possibil-
ity of different hormonal sources (CL, placenta, adrenals)
(Denhard et al., 2008). The strong correlations between
estrogen and progestagen excretion further suggests at
least some estrogen immunoreactivity may, in fact, reflect
CL activity (Denhard et al., 2008). If so, these patterns would
be unique among the felid species.
9. Conclusions
It is clear that felids express marked variations in repro-
ductive mechanisms among species. Identifying the type
of ovulation (induced versus spontaneous) and effect of
seasonality on reproduction for each species is important
because these two characteristics impact both natural and
assisted breeding efforts. The ability to easily and safely
assess reproductive status through noninvasive means is
vital to our understanding the causes of poor fertility of cats
in natural breeding situations and in response to assisted
reproductive procedures. A high priority is developing
ovulation induction protocols that result in consistent
responses, without ovarian hyper-stimulation, to provide
an optimal maternal environment for fertilization and
embryo development. Along with this effort is the need to
control the reproductive cycle, including down-regulating
endogenous ovarian activity and synchronizing time of
estrus (Pelican et al., 2006; Swanson, 2006). We also need
a quick and reliable test for diagnosing pregnancy, prefer-
ably one that is noninvasive. Particularly important will
be systematic assessments of zoo habitats and determin-
ing how they impact animal health. Long-term evaluations
of adrenal activity in the context of exhibits and man-
agement strategies should allow identifying the optimal
161
captive environment compatible with welfare needs and
maximal reproductive potential. Ultimately, by utilizing a
variety of endocrine techniques in conjunction with phys-
iological and behavioral assessments, zoo managers will
be in a better position to determine which environmen-
tal factors are harmful, benign or stimulating to individual
animals. If we are to continue maintaining wild cats ex
situ, we no longer can ignore our obligation to assess the
adequacy of environmental and husbandry conditions for
optimal behavior, health, reproduction and wellbeing.
Conflict of interest statement
The author has no conflict of interest in publishing this
review.
References
Banks, D.H., Paape, S.R., Stabenfeldt, G.H., 1983. Prolactin in the cat. I.
Pseudopregnancy, pregnancy and lactation. Biol. Reprod. 28, 923–932.
Braun, B.C., Frank, A., Denhard, M., Voigt, C.C., Vargas, A., Göritz, F.,
Jewgenow, K., 2009. Pregnancy diagnosis in urine of Iberian lynx (Lynx
pardinus). Theriogenology 71, 754–761.
Brown, J.L., Wildt, D.E., Graham, L.H., Byers, A.P., Barrett, S., Howard, J.G.,
1995. Comparison of natural versus chorionic gonadotropin-induced
ovarian responses in the clouded leopard (Neofelis nebulosa) assessed
by fecal steroids. Biol. Reprod. 53, 93–102.
Brown, J.L., Wildt, D.E., Wielebnowski, N., Goodrowe, K.L., Graham, L.H.,
Wells, S., Howard, J.G., 1996. Reproductive activity in captive female
cheetahs (Acinonyx jubatus) assessed by faecal steroids. J. Reprod. Fer-
til. 106, 337–346.
Brown, J.L., Graham, L.H., Wu, J., Collins, D., Swanson, W.F., 2002.
Reproductive endocrine responses to photoperiod and exogenous
gonadotropins in the Pallas’ cat (Otocolobus manul). Zoo Biol. 21,
347–364.
Brown, J.L., 2006. Comparative endocrinology of domestic and nondomes-
tic felids. Theriogenology 66, 25–36.
Crichton, E.G., Bedows, E., Miller-Lindholm, A.K., Baldwin, D.M., Arm-
strong, D.L., Graham, L.H., Ford, J.J., Gjorret, J.O., Hyttel, P., Pope, C.E.,
Vajta, G., Loskutoff, N.M., 2003. Efficacy of porcine gonadotropins for
repeated stimulation of ovarian activity for oocyte retrieval and in
vitro embryo production and cryopreservation in Siberian tigers (Pan-
thera tigris altaica). Biol. Reprod. 68, 105–113.
De Haas van Dorsser, F.J., Swanson, W.F., Lasano, S., Steinetz, B.G.,
2006. Development, validation, and amplification of a urinary relaxin
radioimmunoassay for the diagnosis and monitoring of pregnancy in
felids. Biol. Reprod. 74, 1090–1095.
De Haas van Dorsser, F.J., Lasano, S., Steinetz, B.G., 2007. Pregnancy diag-
nosis in cats using a rapid, bench-top kit to detect relaxin in urine.
Reprod. Dom. Anim. 42, 111–112.
Denhard, M., Naidenko, S., Frank, A., Braun, B., Göritz, F., Jewgenow, K.,
2008. Non-invasive monitoring of hormones: a tool to improve repro-
duction in captive breeding of the Eurasian lynx. Reprod. Dom. Anim.
43, 74–82.
Donoghue, A.M., Johnston, L.A., Armstrong, D.L., Simmons, L.G., Wildt, D.E.,
1993. Birth of Siberian tiger cub resulting from laparoscopic interuter-
ine artificial insemination. J. Zoo Wildl. Med. 24, 185–189.
Donoghue, A.M., Johnston, L.A., Seal, U.S., Armstrong, D.L., Tilson, R.L.,
Wolff, P.L., Petrini, K.R., Simmons, L.G., Gross, T., Wildt, D.E., 1990. In
vitro fertilization and embryo development in vitro and in vitro in the
tiger. Biol. Reprod. 43, 733–744.
Fanson, K., Wielebnowski, N., Lucas, J., 2009. Reproductive physiology
of Canada lynx (Lynx canadensis). In: Vargas, A., Breitenmoser, C.,
Breitenmoser, U. (Eds.), Iberian Lynx Ex-situ Conservation: An Inter-
disciplinary Approach. Fundación Biodiversidad, Madrid, Spain, pp.
390–399.
Göritz, F., Denhard, M., Hildebrandt, T.B., Naidenko, S.V., Vargas, A., Mar-
tinez, F., Palomares, F., Jewgenow, K., 2009. Non cat-like ovarian cycle
in the Eurasian and the Iberian lynx – ultrasonographic and endocrino-
logical analysis. Reprod. Dom. Anim. 44 (Suppl 2), 87–91.
Graham, L.H., Byers, A.P., Armstrong, D.L., Loskutoff, N.M., Swanson, W.F.,
Wildt, D.E., Brown, J.L., 2006. Natural and gonadotropin-induced ovar-
ian activity in tigers (Panthera tigris) assessed by fecal steroid analysis.
Gen. Comp. Endocrinol. 147, 362–370.
162 J.L. Brown / Animal Reproduction Science 124 (2011) 155–162
Graham, L.H., Swanson, W.F., Wildt, D.E., Brown, J.L., 2004. Influence of oral
melatonin on natural and gonadotrophin-induced ovarian function in
the domestic cat. Theriogenology 61, 1061–1076.
Howard, J.G., Barone, M.A., Donoghue, A.M., Wildt, D.E., 1992. The effect
of preovulatory anaesthesia on ovulation in laparoscopically insemi-
nated domestic cats. J. Reprod. Fertil. 96, 175–186.
Howard, J.G., Byers, A.P., Brown, J.L., Schwartz, R.J., Evans, M.Z., Barrett, S.J.,
1996. Successful ovulation induction and laparoscopic intrauterine
artificial insemination in the clouded leopard (Neofelis nebulosa). Zoo
Biol. 15, 55–69.
Jewgenow, K., Göritz, F., Varga, A., Denhard, M., 2009. Seasonal profiles of
ovarian activity in Iberian lynx (Lynx pardinus) based on urinary hor-
mone metabolite analyses. Reprod. Dom. Anim. 44 (Suppl. 2), 92–97.
Jochle, W., Jochle, M., 1993. Reproduction in a feral cat population and
its control with a prolactin inhibitor, cabergoline. J. Reprod. Fertil. 47,
419–424.
Johnson, L.M., Gay, V.L., 1981. Luteinizing hormone in the cat. I. Tonic
secretion. Endocrinology 109, 240–246.
Jurke, H.M., Czekala, N.M., Lindburg, D.G., Millard, S.E., 1997. Fecal corti-
coid metabolite measurement in the cheetah (Acinonyx jubatus). Zoo
Biol. 16, 133–147.
Leyva, H., Madley, T., Stabenfeldt, G.H., 1989. Effect of light manipula-
tion on ovarian activity and melatonin and prolactin secretion in the
domestic cat. J. Reprod. Fertil. 39, 125–133.
Moreira, N., Monteiro-Filho, E.L.A., Moraes, W., Swanson, W.F., Graham,
L.H., Pasquali, O.L., Gomes, M.L.F., Morais, R.N., Wildt, D.E., Brown, J.L.,
2001. Reproductive steroid hormones and ovarian activity in felids of
the Leopardus genus. Zoo Biol. 20, 103–116.
Pelican, K.M., Abaigar, T., Vargas, A., Rodríguez, J.M., Bergara, J., Rivas, A.,
López, J., Brown, J., Wildt, D.E., 2009. Unusual gonadal hormone pro-
files in the Iberian as determined by fecal monitoring. In: Vargas, A.,
Breitenmoser, C., Breitenmoser, U. (Eds.), Iberian Lynx Ex-situ Con-
servation: An Interdisciplinary Approach. Fundación Biodiversidad,
Madrid, Spain, pp. 341–352.
Pelican, K.M., Wildt, D.E., Pukazhenthi, B., Howard, J.G., 2006. Ovarian
control for assisted reproduction in the domestic cat and wild felids.
Theriogenology 66, 37–48.
Roth, T.L., Wolfe, B.A., Long, J.A., Howard, J.G., Wildt, D.E., 1997. Effects of
equine chorionic gonadotropin, human chorionic gonadotropin, and
laparoscopic artificial insemination on embryo, endocrine, and luteal
characteristics in the domestic cat. Biol. Reprod. 57, 165–171.
Seal, U.S., Plotka, E.D., Smith, J.D., Wright, F.H., Reindl, N., Taylor, R.S., Seal,
M.F., 1985. Immunoreactive luteinizing hormone, estradiol, proges-
terone, testosterone, and androstenedione levels during the breeding
season and anestrus in Siberian tigers. Biol. Reprod. 32, 361–368.
Seal, U.S., Tilson, R.L., Plotka, E.D., Reindl, N.J., Seal, M.F., 1987. Behav-
ioral indicators and endocrine correlates of estrus and anestrus in
Siberian tigers. In: Tilson, R.L., Seal, U.S. (Eds.), Tigers of the World.
Noyes Publications, Park Ridge, NJ, pp. 244–254.
Shille, V.M., Lundstrom, K.E., Stabenfeldt, G.H., 1979. Follicular function
in the domestic cat as determined by estradiol-17␤ concentrations in
plasma: relation to estrous behavior and cornification of exfoliated
vaginal epithelium. Biol. Reprod. 21, 953–963.
Shille, V.M., Munro, C., Farmer, S.W., Papkoff, H., 1983. Ovarian and
endocrine responses in the cat after coitus. J. Reprod. Fertil. 69, 29–39.
Stewart, D.R., Stabenfeldt, G.H., 1985. Relaxin activity in the pregnant cat.
Biol. Reprod. 32, 848–854.
Swanson, W.F., 2006. Application of assisted reproduction for population
management in felids: the potential and reality for conservation of
small cats. Theriogenology 66, 49–58.
Tsutsui, T., Stabenfeldt, G.H., 1993. Biology of ovarian cycles, pregnancy
and pseudopregnancy in the domestic cat. J. Reprod. Fertil. 47, 29–35.
Wielebnowski, N., Brown, J., 1998. Behavioral correlates of physiological
estrus in cheetahs. Zoo Biol. 17, 193–209.
Wielebnowski, N.C., Watters, J., 2007. Applying fecal endocrine monitor-
ing to conservation and behavior studies of wild mammals: important
considerations and preliminary tests. Isreal J. Ecol. Evol. 53, 439–460.
Wielebnowski, N.C., Ziegler, K., Wildt, D.E., Lukas, J., Brown, J.L.,
2002. Impact of social management on reproductive, adrenal and
behavioural activity in the cheetah (Acinonyx jubatus). Anim. Conserv.
5, 291–301.
Wildt, D.E., Brown, J.L., Swanson, W.F., 1998. Reproduction in cats. In: Kno-
bil, E., Neill, J. (Eds.), Encyclopedia of Reproduction, 1st ed. Academic
Press, New York, pp. 497–510.