Management and Conservation of Captive Tigers, Chapter 5

Reproduction and Propagation in Tigers
Adapted from U. Seal, D. Wildt, R. Tilson, A. Donoghue, N. Reindl, R. Taylor and other contributors

Editors' Note: The tiger's distribution ranges from the hot and humid tropics of Indonesia and Malaysia, to the seasonally dry but hot evergreen forests of India, Myanmar and Thailand, to the frozen oak and pine forests of northern China, Korea, and southeastern Russia. Historically they ranged as far south as 9 latitude and as far north as 39 latitude. This broad distribution over different climatic regimes has dramatic effects on seasonal patterns of reproduction. In addition, our knowledge of reproduction in captive tigers is more robust for Siberian tigers, less so for the Bengal and Sumatran subspecies, and virtually nonexistent for the other subspecies. Thus, here we attempted to identify the subspecies if there is reasonable doubt that differences exist; otherwise the statement is probably applicable to all subspecies.

The Siberian tiger appears to be a seasonally polyestrous breeder and an induced ovulator whose breeding season may be synchronized by photoperiod. Analysis of 1,239 captive births of Siberian tigers in collections throughout the Northern Hemisphere revealed a peak in April-June (Fig. 1), suggesting that Siberian tigers are seasonal breeders.

Siberian tigresses show behavioral estrous cycles and ovarian follicular phase cycles beginning in late January and cease in early June. The duration of anestrus is seven to eight months. These findings correlate with the observed pattern of births for tigers in the Northern Hemisphere.

One animal exposed to long day (16L:8D) photoperiods in the late fall exhibited a shortened anestrus interval (Seal et al. 1985), suggesting that seasonal cycles in these tigers may be synchronized by photoperiod. This might also account for some of the off-season births in zoo animals that are exposed to artificially extended photoperiods as a part of exhibit or management practices.

Fig. 1. Number of births by month and sex for registered Siberian tigers in Northern Hemisphere zoos. Data from 530 litters (1,239 young) reported in the International Tiger Studbooks 1976-82 (ISIS 1983, Seifert and Muller 1985).

Seal and colleagues (1987) conducted a study to develop a quantitative behavioral profile, based upon daily observations of female Siberian tigers during the breeding season, that might be used to identify the stage of estrus in individual animals as a prelude to ovulation induction and artificial insemination. Blood samples were collected and physical examinations conducted at least once a week to provide endocrine and physiological correlates of the estrous cycle for comparison with behavioral data.

Three of the females used for establishing the behavioral estrous cycle were observed daily for four months (February to May). The frequency of occurrence of a set of behaviors was recorded by the same person in five one-minute segments prior to the start of each day's routine maintenance activities. The behavioral indicators of estrus (Kleiman 1974) chosen for inclusion in this study were vocalizing (calling or moaning; Schaller 1967), prustening (a greeting call that sounds like air expelled softly through the nostrils), rubbing the cheek, forehead or flank against the walls/bars of the enclosure, and rolling over and writhing on the back and the exhi-bition of lordosis or semilordosis (postures assumed just prior to copulation; see Leyhausen 1956). Each of these behaviors was weighted equally, in contrast to Kleiman's (1974) method of establishing a symbolic score for estrous behavior by weighting particular patterns more than others (e.g., lordosis, rolling and flank rubbing were scored higher than cheek rubbing, prustening or calling). Other estrous behaviors noted by Kleiman for female Bengal tigers included urine-spraying, exhibiting flehmen (for definition see Leyhausen 1956) and pacing.

Although the frequencies of occurrence of some behaviors showed significant correlations with the endocrine profiles of the females (see Seal et al. 1985), total scores with no weighting of individual behaviors were most indicative of the females' estrous cycles. Like Kleiman's (1974) observations with Bengal tigers, urine-spraying among female Siberian tigers was negatively correlated with the occurrence of estrus.

Estradiol. Peaks of estradiol concentration occurred between February to June with low values from June through January under natural photoperiod. Baseline serum immunoreactive estradiol-17þ values (Figs. 2 and 3) ranged from <5 to 115 pg/ml in 520 weekly base-line samples collected over 42 months during natural cycles for three tigers, and 18 months for two tigers. Values greater than 20 pg/ml of immunoreactive estradiol were more than three standard deviations greater than the mean of the remaining values or the anestrous values and were considered indicative of a peak and of an active ovarian follicular phase. During anestrus, estradiol levels averaged 4.2 pg/ml, ranging from 0.5 to 9.3 pg/ml. Peak estradiol concentrations were 47.6 pg/ml, ranging from 21 to 115 pg/ml.

The duration of elevated estradiol values from 56 cycles in five tigers was 6-10 days. Excluding two outliers and the data from one female believed to be a spontaneous ovulator, the interval between peaks averaged 24.9 days.

Fig. 2. Baseline concentrations of estradiol and progesterone in weekly samples during two breeding seasons from female tiger 711. The photoperiod in the second year (1983) was extended to 16L:8D by use of a 250 watt floodlight mounted in the cieling of the indoor enclosure. The animal was shut into the enclosure at night to ensure exposure to the light.

Fig. 3. Baseline serum immunoreactive estradiol concentrations in female tiger 711 over three seasons.

Progesterone. Serum progesterone concentrations ranged from 0.5 to 12 ng/ml in the set of 440 weekly baseline samples (excluding the spontaneously ovulating female and experimentally induced ovulatory cycles). Progesterone values were less than 1 ng/ml in 145 of the baseline samples, and 18 values were between 1 and 2 ng/ml (Fig. 2). Excluding values greater than 2 ng/ml, serum progesterone concentration was 1.2 in the samples collected February through June (except for the spontaneous ovulating female). Values greater than 2 ng/ml were observed in 17 of the 56 baseline samples; ten from the most excitable female, three from another female (none above 3 ng/ml), and four from a third female (one value above 3 ng/ml). Eleven of 17 estradiol peaks were not associated with elevations of progesterone and none of the elevated progesterone levels persisted more than two weeks.

Spontaneous Ovulation. One female appeared to be a spontaneous ovulator. She gave no clear behavioral indications of estrus for several months after weekly blood collections began in January 1985, then an endocrine and behavioral estrus was followed by an increase in serum progesterone to >70 ng/ml. Serum progesterone values indicated about a five-week luteal phase ending in early June. She was not handled again until early 1986. Progesterone concentrations in weekly blood samples with no other manipulations indicated that she ovulated spontaneously three times in 1986. Each luteal phase appeared to be five or more weeks in duration.

Testosterone. Serum testosterone concentrations in the baseline samples (Fig. 4) ranged from 10 to 100 ng/dl. During anestrus, testosterone concentrations averaged 23.4 ng/dl. Peak testosterone levels averaged 73.9 ng/dl. All of the 54 estradiol peaks greater than 20 pg/ml were accompanied by testosterone peaks. Androstenedione concentrations were correlated with estradiol during the estrous season as were testosterone values. The correlation of serum testosterone and androstenedione levels with estradiol, their lack of correlation with progesterone, and their increase after PMSG treatment suggest an ovarian rather than adrenal origin for these hormones in these tigers.

Estrous behavioral profiles were significantly correlated (P <0.001) with the endocrine profiles of estradiol and testosterone concentrations. Clear peaks, showing a steady increase and subsequent decrease in expression, were apparent in each cycle the tigers exhibited. Also, the peak expression of each female's estrus profile was relatively constant, showing more variation between females than among the various cycles of any individual female.

The interval in days that female 711 showed estrous behavior increased from the first to fourth (or last) cycle she underwent for the season (Fig. 4). The first estrous cycle was about four days in length, the second and third were close to ten days in length and the last cycle was about 15 days in length.

On average, the interestrous interval for Siberian females was 25.0 + 1.3 days (N=10), an interval in agreement with the interestrous endocrine peaks (Seal et al. 1985). This 25 day interval has been confirmed from studies based upon both the behavioral and hormonal data derived from nine animals of different genetic lineages over four seasons of observations.

Fig. 4. Female tiger 711's behavioral profile and serum estradiol and testosterone levels during four cycles in the 1985 breeding season.

One behavioral feature of the interestrous period is the complete apathy females display during anestrus. In contrast to the constant vocalizing, pacing, rubbing and rolling that is indicative of estrus, anestrus is characterized by resting quietly, even though measurable (but low) concentrations of estradiol and testosterone are evident (Fig.4).

The value of using behavioral indicators of an estrous cycle is that they allow a relatively accurate prediction of when the next cycle will occur (approximately 25 days in Siberian tigers) and thus allow for more accurate timing for placing females with males for natural breeding or for assisted reproductive manipulations.

In summary, behavioral observation of 10 estrous cycles in three tigers yielded an estrous length of 5.3 + 0.2 days and an inter-estrous interval of 24.9 +1.3 days. The interestrous interval between estradiol peaks was 24.9 + 1.3 days with two outliers (42 days).

(adapted from D. Wildt, L. Phillips, L. Simmons, K. Goodrowe, J. Howard, J. Brown and M. Bush)

Considerable data now are available concerning ejaculate-endocrine characteristics of the male tiger. High quality semen can be collected by electroejaculation from most anesthetized males greater than 2.5 years of age. Details on electroejaculate characteristics are provided and are found to vary considerably among males, even in those with positive breeding histories. On the average, the tiger ejaculate contains approximately 40% pleiomorphic sperm forms, less than the proportion observed in other closely studied felids like the cheetah, puma and leopard (Wildt et. al., 1988). No morphometric correlates are indicative of reproductive potential; testicular size is not related to either electroejaculate volume, spermatozoal concentration or motility characteristics.

Based on circulating concentrations of cortisol in electroejaculated and ACTH challenged males, the tiger produces a rather modest adrenal response compared to other wild felids (Wildt et. al.,1988). Cortisol levels also occasionally rise even in control males subjected to anesthesia only. Blood levels of LH and testosterone are comparable to values measured in the North Chinese leopard and puma but generally are greater than concentrations detected in the cheetah. Neither LH nor testosterone level is related to any electroejaculate trait measured. Significant fluctuations in both hormones suggest that a multiple blood sampling regimen is necessary to assess the endocrine status of any given male.

The in vitro viability of tiger spermatozoa is improved markedly by laboratory processing which involves dilution, centrifugation, resuspension and maintenance at ambient temperature. Spermatozoa of this species also survive pellet freezing on dry ice and storage in liquid nitrogen. The Tiger SSP has recommended that semen from significant male Siberian and Sumatran tigers be collected and cryopreserved. The most valuable males have been identified by the Tiger SSP and the freezing of multiple ejaculates from these individuals is in progress.

(from D. Wildt and A. Donoghue)

Editors' note: The AZA Tiger SSP has always recommended natural breeding of tigers to achieve its program goals. There is merit, however, in developing assisted reproductive techniques. These new techniques will provide avenues to more expeditiously and safely exchange genetic material between or among AZA zoos, regional programs of the world, captive and wild programs, and wild to wild programs. Another aspect of this technology is the development of the IUCN/SSC CBSG Genome Resource Bank.

At present, reproductive biotechnology is not being used for routinely managing the great cats. However, there is exciting progress and indications that some techniques will be available soon.

Artificial Insemination (AI). Although there have been several attempts, the use of the conventional (AI) approach in which sperm are placed in the female's vagina has never been successful in felids. In the tiger, vaginally deposited sperm are not transported through the reproductive tract to the site of fertilization (oviduct). This appears to be due to the need for anesthesia which quiets the uterus and reduces contractions that normally assist in sperm transport. One alternative being used at the National Zoo is the placement of sperm nearer to the site of fertilization using the technique of laparoscopy. A laparoscope is a fiber-optic telescope inserted through a small incision in the abdominal wall and used to view the reproductive organs. Laparoscopy can also be used to direct a needle through the abdominal wall and into the uterine horns. Tubing containing sperm can be threaded through the needle and into the uterus where the sperm are injected adjacent to the oviduct.

Liveborn offspring have resulted after intrauterine insemination in the domestic cat, puma, leopard cat, clouded leopard and cheetah (Howard et al, 1992a,b). The laparoscopic intrauterine AI technique has recently been adapted to the tiger (Donoghue et.al., 1992b). Eight females were given a single injection of pregnant mares' serum gonadotropin (PMSG) followed by human chorionic gonadotropin (hCG) 80 hours later. At 40-48 hours after hCG, electroejaculated/ processed spermatozoa were deposited trans-abdominally into the proximal aspect of the uterine horns of each female. The AI procedure was simple and rapid, generally requiring only 30 minutes after laparoscope insertion. One female maintained a pregnancy and gave birth to a single live cub after an 111-day gestation. Laparoscopic AI appears to have considerable potential as a tool for assisting captive propagation of the tiger. It appears the primary obstacle to the routine use of AI for helping to manage tigers is identifying the optimal hormone treatment to induce a "normal" ovulatory response. However, the birth of a healthy cub is encouraging, suggesting that with continued effort, laparoscopic AI will be a useful management tool in the propagation of this species.

In vitro Fertilization (IVF) and Embryo Transfer (ET). IVF is powerful because it eliminates concerns about behavioral incompatibility, and, unlike AI, does not require identifying the time of sexual receptivity (estrus) or ovulation. Like AI, IVF could be used to infuse new genes into genetically stagnant felid populations; sperm from free-living males could be used to inseminate eggs from captive females. In 1988, the National Zoo produced the first-ever carnivore offspring (domestic cat kittens) by IVF and ET. Females were injected with a hormone to synchronize and stimulate ovarian activity. Using a laparoscope, eggs from ovarian follicles were aspirated through the abdominal wall. Collected sperm were washed and cultured with the eggs in a laboratory incubator. Thirty hours later, the eggs were examined for embryo formation and transferred to surrogate females. Five of six cats became pregnant (Goodrowe et al. 1988). The Cincinnati Zoo reported the birth of an Asian wild cat kitten by similar procedures in 1989 while National Zoo scientists also produced leopard cat (Goodrowe et al. 1988), puma (Miller et al. 1989), the rare Florida panther (unpublished data), and cheetah (Donoghue et al. 1992a) embryos by IVF. The most consistent success has been with the tiger. IVF has been useful for developing an understanding of follicular development (ovulation induction treat ments), egg maturation, sperm function, gamete interaction and embryo development in this species (Donoghue et al. 1990, 1992c). Single injections of PMSG and hCG were as effective for provoking follicular development and intrafollicular oocyte maturation as a similar hormone treatment given to domestic cats, leopard cats, pumas and cheetahs. The National Zoo, in collaboration with Omaha's Henry Doorly Zoo and the Minnesota Zoo, has generated tiger embryos and the first-ever tiger offspring by in vitro fertilization and embryo transfer (Donoghue et al. 1990). Sixteen female tigers were subjected to the hormone therapy and produced more than 20 oocytes each. From a structural perspective, tiger oocytes appeared grossly similar to domestic cat, leopard cat, puma, and cheetah oocytes. Of the 358 mature oocytes inseminated with tiger sperm, 227 (63.4%) fertilized, a rate similar to that observed in the domestic cat (Johnston et al. 1991a,b). Eighty-six 2-4 cell embryos were transferred surgically into the oviducts of six females. A pregnancy occurred in one female and three live-born cubs were delivered by Caesarian section 107 days post-embryo transfer.

These results demonstrate successful in vitro fertilization, embryo culture and production of offspring after IVF and embryo transfer in a Panthera species. Female tigers responded to a PMSG/hCG stimulus by producing large numbers of developing follicles and a high percentage of mature oocytes that were capable of being fertilized and developing in vitro and in vivo. The numbers of mature oocytes being fertilized and cleaving in vitro may have resulted from the successful gonadotropic stimulation of females, optimal in vitro culture conditions or the naturally high incidence of structurally normal spermatozoa in electroejaculates. Tiger embryos allowed to culture in vitro developed readily to the late morula stage but appeared to experience a partial developmental block to becoming blastocysts. Most importantly, these embryos were biologically competent as demonstrated by the birth of live young.

Oocyte Rescue / In vitro Maturation. To ensure that all genetically valuable animals contribute to the breeding population, every avenue must be explored. One frustration is the loss of genetic potential due to age, terminal illness, or unexpected death. Until recently, there were no methods for salvaging the genetic material of such animals. Recently, however, immature ovarian eggs have been "rescued" from animals that died unexpectedly. These eggs can be cultured or "matured" in the laboratory, subjected to IVF, and the embryos transferred to surrogate mothers. Generating life from recently dying animals has resulted in live births in several livestock and laboratory animal species. Scientists have begun salvaging eggs from cat species. In the domestic cat, about one-half of the eggs can be matured and about one-third of these form embryos after IVF. Studies have begun with other felid species, and a North American network is in place involving the cooperation of more than 30 zoos. Ovaries from animals which die unexpectedly are shipped to the National Zoo where the eggs are harvested, studied and used to produce embryos. Oocytes have been recovered from 13 felid species representing 35 individuals (Johnston et al. 1991c). To date, seven pairs of ovaries have been collected from tigers, a total of 167 oocytes recovered and eight of 88 oocytes fertilized after in vitro maturation and IVF (Johnston et al. 1991c).

Gamete and Embryo Cryopreservation. Leopard cats have been produced following laparoscopic AI of frozen-thawed spermatozoa. However, there have been no reports of offspring from cryopreserved tiger spermatozoa, oocytes or embryos. A recently completed study provides data on the viability of frozen-thawed tiger spermatozoa using the in vitro fertilization system (Donoghue et al. 1992c). Tiger electro-ejaculates were divided, and half were used fresh to inseminate tiger and hypertonically salt-stored domestic cat oocytes. The remainder was pelleted-frozen in a solution of 20% egg yolk, 11% lactose and 4% glycerol, thawed and cultured with tiger and domestic cat oocytes. The motility index [(sperm % motility) + (status rating x 20)]/2 for thawed spermatozoa was approximately 86% of that in fresh aliquots. Of the 49 tiger oocytes inseminated in vitro with fresh spermatozoa, 34 (69.4%) cleaved, compared with 33 of 47 oocytes (70.2%) cultured with thawed spermatozoa. Embryos generated by either treatment were capable of developing in vitro to the 16-cell or morula stage. Fresh and thawed tiger spermatozoa were equally capable of binding and penetrating the outer and inner zona pellucida of domestic cat oocytes. These results demonstrate the ability of frozen-thawed tiger spermatozoa to penetrate homologous and heterologous oocytes and result in conspecific, advanced development of preimplantation embryos in vitro . The potential for producing offspring in tigers following assisted reproduction using frozen sperm is great.

Reproductive Research Priorities. From a reproductive biologist's perspective, there are three high priority research areas:

1. Monitoring ovarian cycles. There is a need for a more systematic approach to identifying behavioral clues indicative of estrus while confirming observational data through physiological methods like hormonal analyses via serial blood sampling or fecal analyses. We need to monitor reproductive activity consistently in female tigers using traditional or, as yet, undiscovered indices.
2. Improving assisted reproductive technology. There is a need to continue exploring and improving assisted reproductive technology within the Tiger SSP program. No species should be strictly propagated in captivity using assisted reproductive technology alone, however, biotechnical approaches offer unlimited opportunities for improving our management capabilities. Presently, there are six felid species, including the tiger, in which offspring have been produced by artificial insemination (AI). Furthermore, IVF has been used by two independent laboratories to produce liveborn offspring in three felid species, including the tiger. Assisted reproductive technologies remain as one of the most powerful, potential tools available for enhancing management. In addition to permitting the immediate capture of all existing genes in the present population (through the cryopreservation of sperm and embryos), these approaches would allow the inter-institutional propagation of species and the infusion of new genes from the wild without translocating individual animals. Also, regardless of advancements in husbandry techniques, there always will be sexually incompatible or dysfunctional animals designated by the SSP as priority individuals for required breedings. AI or IVF in combination with the use of fresh or frozen gametes would be useful for circumventing these problems.
3. Developing a genetic resource banking program for the tiger. A bank of cryopreserved gametes and embryos from captive founders and representative free-living individuals provides unique opportunities for controlling genetic diversity. Current data indicate that tiger spermatozoa pose no special problems for cryopreservation and banking, provided that acceptable ejaculates are collected. Research should be initiated to examine the efficacy of current techniques for assisted reproduction of captive populations. When possible, semen should be collected from founders and other genetically important males for cryopreservation. A draft review of a Genome Resource Bank Action Plan for Tigers, under the auspices of the IUCN/SSC CBSG, is underway.

The AZA Tiger SSP Masterplan attempts to achieve its genetic and demographic goals within the constraints of available space for tigers among participating institutions. In general, 8-10 breeding recommendations for Siberian tigers are made each year. This number is based upon the facts that there are about three cubs per litter, neonatal mortality is about 36%, and there is about 30% failure rate of paired adults to actually breed. The net result is the production of about 12 cubs per year, which closely matches the loss of tigers through old age or disease.

Many of these breeding recommendations are to zoos which have never bred tigers or whose staff have no experience in breeding tigers. The introduction of tigers for breeding is typically characterized by moderate to extreme levels of excitable or violent behavior, depending on the tigers' "personalities" and the observers' experience. These initial introductions and resulting interactions usually prompt a telephone call requesting guidance and reassurance. There is no magic formula for introducing tigers, but there are some general guidelines that have proven to be successful for some institutions. To convey this information a series of most commonly asked questions regarding tiger breeding introductions was developed. These questions were then answered by a keeper and curator at the Minnesota Zoo, who between them have managed the breed-ing of 10 pairs of tigers resulting in 33 tigers registered in the International Tiger Studbook. Because most of this information is anecdotal, other managers may have their own way for introducing tigers for breeding and are welcome to submit their opinions for future revisions of this husbandry manual.

Frequently Asked Questions Regarding Breeding Tigers (asked of R. Taylor and N. Reindl, Minnesota Zoo, by R. Tilson)
	When should tigers be shipped to the new facility in order to breed them according to SSP recommendation?
	Move the tigers as soon as the recommendation is made, avoiding extreme weather conditions of summer and winter.
	What if the facility is under construction and the zoo wants to wait as long as possible before receiving tigers?
	Because of interindividual differences among tigers, a calm tiger can settle into a new facility and probably be adapted for breeding at a minimum of two months; a more excitable tiger may be adaptable in three months, and some may take even longer than that. Caution: Care should be taken not to move a tiger from one facility to the next within the zoo just prior to breeding. Past experience has shown some tigers may begin their estrus on environmental cues, and when moved to a new facility they may shut down their reproductive response and go into anestrus.
	What do you do if a tiger does not eat when transferred?
	Some tigers just won't eat because of these changes. At the Minnesota Zoo, one female went 19 days with no food, yet still survived. One male who came to the Minnesota Zoo from Moscow Zoo via Leipzig Zoo never did completely switch to a commercially prepared feline diet, and 20-25% of his food was supplemented with horse meat up to the day he died.
	How can staff determine when the female tiger is ready to be introduced to the male?
	Most cues come from the female. She will begin to react in a more positive fashion to staff by increased rates in greeting, vocalizing with the "prusten", rubbing up against bars, becoming more active, etc. The female's appetite will either fall off slightly, dramatically or she may even stop eating altogether for a day or two. The male will not show a similar decrease in appetite. [Editors' note: In general, tigers that are hand-reared tigers always react positively to zoo staff, and it is difficult to distinguish when they are in estrus versus when they are in anestrus. Refer to the study by Seal et al. (1987) regarding correlations between endocrine events and behavioral expression of estrus for a more detailed overview.]
	How long should the prepatory introduction sequence between enclosures last?
	A metal grid should be designed so that there can be no physical contact between the two tigers but allow continual visual access and include olfactory cues. The grid should be in place just as soon as the two tigers are put in adjoining enclosures, and used throughout the breeding introductions. The tigers to be bred should be placed in adjoining enclosures as soon as possible and kept together long before the expected reproductive season begins. Again, the shift door with the screen partition should be in place during this entire period. The shift door with screen should be open for 24 hours except if the female is noticeably disturbed by the male's presence, in which case it should be closed at least during the evening hours.
	How many staff, and which ones, should be involved in the introduction?
	Choice of staff is critical. Primary responsibility should be given one staff person, with one or possibly two back-up keepers to fill in. At any one time, only one person should be in charge and present during the introduction process.
	What precautionary measures should be taken? (what do you do if there is serious aggression)?
	A keeper familiar with the tigers can probably anticipate a problem by their behavior, and separate them by making loud noises and by closing the shift door. Water hoses and anything that makes a loud noise (air horns) may also help separate aggressive tigers.
	How large should the enclosures where the cats will be breeding be?
	The initial introduction should take place in an area where the animals can be more easily controlled. Once it has been determined that the introduction is going well, it is possible to give the animals access to a larger area so that the male has space to back away after copulation; a minor confrontation can more easily escalate into a full blown fight in a restricted space with fewer opportunities for escape (D. Richardson).  At the Minnesota Zoo, eight of nine breedings occurred in enclosures 3.5m x 4m. However, adjacent enclosures were connected in a series so that up to four interconnected enclosures were available to the tigers, providing adequate space for breeding introduction. Tigers can be put together in outdoor enclosures, but it is more difficult to separate them if they get violent, it is difficult to tell whether they are relating well to each other, and it is particularly difficult to determine if they have actually copulated. Other problems include greater probability of distraction either by staff or other events, and particularly multiple scents from different tigers, possibly even different males, that can confuse or disrupt the introduction process. Under no circumstances should breeding tigers be left together without staff supervision. When cleaning the mixing enclosures it is also important not to scrub them too thoroughly or use disinfectants too liberally. Urine particularly carries important olfactory cues about the status of the female's reproductive cycle that the male is capable of perceiving. By leaving some of these odors in the enclosure it helps the tigers respond appropriately to each other.
	How long should the tigers be kept together during the introductory sequences?
	If the female is in estrus, she should be mixed with the male for at least one hour in the morning and another hour in the afternoon. These short durations are better than longer sequences because of the impact on staff time and because, in our experience, tigers breed well during these brief periods.
	How many days should these introductions last?
	As long as the female is reacting positively to the presence of the male, as long as the two are copulating or not fighting, the process should continue. The intensity builds up over a seven-day period, reaching a maximum at days 3, 4, 5 and 6 and then tapering off to no interactions. See Seal et al. (1987) for details of endocrine and behavioral correlates at this time.
	When do you know to separate the tigers?
	The female simply stops showing any interest in the male and ignores him completely. This implies she is probably in anestrus, and they should be separated until the next cycle begins or she is considered pregnant. Tigers sometimes bond, and if the pair is getting along together well, and the male is not needed for other breedings, it is possible to leave them together.
	When would you reschedule a second introduction and how would you know?
	The female should begin recycling within 30 days and if she does, a second reintroduction should begin. If the female does not recycle and continues not showing interest up through 50 days, she is either pregnant or going through a false pregnancy. Her behavior will be the same as during the first estrus. If the female is pregnant, she will certainly start showing an increase appetite as she proceeds in her pregnancy.
	How do you calculate the expected date of birth?
	Take the midpoint between the first and last day of copulation and project out 102 days for the date of birth. [Editors' note: Gestation in Siberian tigers is considered to be 104 days and in Bengal tigers it is 102 days.]
	How do you prepare the enclosure for impending birth?
	The female should have quiet, secure place to retreat to; it should be provided with a nest box 1.5m x 2.5 x 1m (a door at each end) and preferably fitted with a video screen to monitor the female's behavior. Note: some tigers do not like nest boxes and they may have to be removed if they disturb the female too much. The nest box should be cleaned only when necessary and not scrubbed with disinfectants or washed out and totally cleansed of all odors. This reinforces the female's sense of security and increases the likelihood of her rearing the cubs herself. The nest box should be introduced at the earliest indication she may be pregnant rather than waiting the one to two weeks before she is ready to parturate.

(adapted from Keeper Daily Log, Northern Trail, Minnesota Zoo)

[Editors' note: In order to impart some sense of how the introduction of tigers for breeding occurs, a day-to-day account as entered in a keeper's log is presented below.]

Friday, 15 Jan 1982:
0.1 Tiger 180 continues in estrus.
1.1 Tigers 2832, 180 introduced via expanded metal "intro" door, visual contact only. Considerable positive signs, such as lengthy greeting (prusten), 180 voluntarily remaining close to "intro" door, occasionally rolling and vocalizing.
1.1 Tigers 2832, 180 mixed at 2:15 PM. First time compatibility appeared good, considerable prusten, flank-rubbing, rolling (180), 180 presented herself once, 2832 somewhat hesitant, but appeared interested (made a move to grab neck before 180 pulled away). Occasional pawing by both cats. Mixed for 20 minutes with no serious defensive or aggressive interaction.

Saturday, 16 Jan 1982:
1.1 Tigers 2832, 180 mixed for second day (10:45 AM). Given access to three inside enclosures and one outside run. Fourteen (14) mounts observed, one successful (intromission) mount on the 11th try (11:28 AM). Mounts took place in both inside and outside runs. Pair separated after a somewhat more aggressive response from 180. Upright pawing took place after each attempt.

Sunday, 17 Jan 1982:
Primary keeper's day off. Back-up keeper's report:
1.1 Tigers 2832 and180 breeding. Twelve (12) breedings between 9:30 - 11:00 AM. Ten (10) breedings between 12:45 - 2:15 PM.

Monday, 18 Jan 1982:
Primary keeper's day off. Back-up keeper's report:
1.1 Tigers 180 and 2832 breeding 19 times.

Tuesday, 19 Jan 1982:
1.1 Tigers 2832, 180 mixed for breeding, two hours intermittently. Nineteen (19) mounts and fourteen (14) intromissions observed (average interval six minutes). Nothing but typical aggression observed from female.

Wednesday, 20 Jan 1982:
1.1 Tigers 2832, 180 mixed for breeding (10:50 AM). Seven (7) intromissions observed out of eight (8) attempted mounts. No serious aggression observed.

Thursday, 21 Jan 1982:
1.1 Tigers 2832, 180 mixed briefly, no signs, indications of 180 being "in heat" after initial greeting (prusten), very little positive interaction. Disinterest by 2832 and occasional offensive pawing by 180.

Full estrus appeared to be of a five-day duration with possible induced ovulation of the third day, extrapolating term (cubbing date) to be 4 May or 106 days with a mid-term (50 day) date of 9 March.

Friday, 22 Jan 1982:
1.1 Tigers 2832, 180 mixed very briefly. 180 definitely out of "heat" although 180 did some genital sniffing.

[Editors' note: Male 2832 (SB #1306) and female 180 (SB #1295) produced 2.2 (2 males, 2 females) cubs on 4 May 1982.]

Minnesota Zoo Breeding Log, January 1996

Fig. 8. Daily behavioral profile of tiger breeding.

(from M. Bush, L Phillips and R Montali)
Reversible control of reproduction has been accomplished through use of oral and intramuscular contraceptives (Rotterdam Zoo), oral and injectable progestogens for short-term contraception (EEP), and contraceptive implants placed subcutaneously in female tigers (SSP). The implant is a medical grade silastic compound impregnated with melengesterol acetate (Seal 1975, 1976). The slow continuous release of this progestogen-like chemical effectively prevents estrus and suppresses follicular activity on the ovaries. Each implant is usually effective for a two-year period and can be replaced if desired.

In most cases, removing the implant will cause a return to estrous cyclicity within weeks if cycling was normal beforehand. Animals that have been implanted may have increased risk of acquiring severe cystic endometrial hyperplasia, endometrial polyps and endometrial and mammary gland cancer. Therefore, these implants are not the method of choice for permanent sterilization and only should be used intermittently for temporary contraception. For the latter purposes, this is a simple, dependable and inexpensive contraceptive (see below).

(from L. Munson)
For temporary contraception of genetically valuable tigers: Physical separation is optimal. Tigers that are in long-term breeding "hold" (4-6 years) may be moved to facilities with more space. Melengestrol implant may be used as a contraceptive for females for a maximum of two years. Based on available data to date, short-term use of melengestrol should have no detrimental effects on the fertility and health of tigers. These recommendations may be modified if further studies indicate otherwise.

For surplus tigers: Ovariohysterectomy (OVH) is recommended for females as the optimal procedure for the long-term health. Ovariectomy by laparoscopy is recommended if OVH is impossible. If OVH and ovariectomy are impossible, tubal ligation by laparoscopy is recommended to replace long-term use of melengestrol implants. Removed reproductive tracts would be of value to the Tiger SSP for ongoing reproductive studies. For males, castration is the method of choice to prevent the production of sperm and to alter breeding behavior. This decision is based on the premise that uncastrated males with cycling females will induce ovulation by copulation which will increase the risk of females deve-loping uterine and mammary gland disease from exposure to endogenous progesterone. Vasectomy is recommended if castration is impossible. An exception would be made for males with females that will produce litters periodically, because pregnancy usually negates the negative effects of chronic exposure to endogenous steroids.

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. In vitro fertilization and embryo development in vitro and in vitro in the tiger. BIOLOGY OF REPRODUCTION 43:733-44, 1990.

Donoghue, A.M.; Howard, J.G.; Byers, A.P.; Goodrowe, K.L.; Bush, M.; Blumer, E.; Lukas, J.; Stover, J.; Snodgrass, K.; Wildt, D.E. Correlation of sperm viability with gamete interaction and fertilization in vitro in the cheetah (Acinonyx jubatus). BIOLOGY OF REPRODUCTION 46:1047-56, 1992a.

Donoghue, A.M.; Johnston, L.A.; Armstrong, D.L.; Simmons, L.G.; Wildt, D.E. Birth of Siberian tiger cub resulting from laparoscopic interuterine artificial insemination. JOURNAL OF ZOO WILDLIFE MEDICINE, 1992b.

Donoghue, A.M.; Johnston, L.A.; Seal, U.S.; Armstrong, D.L.; Simmons, L.G.; Gross, T.; Tilson, R.L.; Wildt, D.E. Ability of thawed tiger (Panthera tigris) spermatozoa to fertilize conspecific eggs and bind and penetrate domestic cat eggs in vitro. JOURNAL OF REPRODUCTIVE FERTILITY 96: 555-64, 1992c.

Goodrowe, K.L.; Wall, R.J.; O'Brien, S.J.; Schmidt, P.M.; Wildt, D.E. Developmental competence of domestic cat follicular oocytes after fertilization in vitro. BIOLOGY OF REPRODUCTION 39:355-72, 1988.

Goodrowe, K.L.; Miller (Donoghue), A.M.; Wildt, D.E. In vitro fertilization of gonadotropin stimulated leopard cat (Felis bengalensis) follicular oocytes. JOURNAL OF EXPERIMENTAL ZOOLOGY 252:89-95, 1989.

Howard, J.G.; Barone, M.A.; Donoghue, A.M.; Wildt, D.E. The effect of pre- ovulatory anesthesia on ovulation in laparoscopically-inseminated domestic cats. JOURNAL OF REPRODUCTIVE FERTILITY 96:175-86, 1992a.

Howard, J.G.; Donoghue, A.M.; Barone, M.A.; Goodrowe, K.L.; Blumer, E.S.; Snodgrass, K.; Starnes, D.; Tucker, M.; Bush, M.; Wildt, D.E. Successful induction of ovarian activity and laparoscopic intrauterine artificial insemination in the cheetah (Acinonyx jubatus). JOURNAL OF ZOO WILDLIFE MEDICINE 23:288-300, 1992b.

Johnston, L.A.; Donoghue, A.M.; O'Brien, S.J.; Wildt, D.E. Culture medium and protein supplementation influence in vitro fertilization and embryo development in the domestic cat. JOURNAL OF EXPERMENTAL ZOOLOGY 257:350-59,1991a.

Johnston, L.A.; Donoghue, A.M.; O'Brien, S.J.; Wildt, D.E. Influence of temperature and gas atmosphere on in vitro fertilization and embryo development in the domestic cat. JOURNAL OF REPRODUCTIVE FERTILITY 92:377-82,1991b.

Johnston, L.A.; Donoghue, A.M.; O'Brien, S.J.; Wildt, D.E. Rescue and maturation in vitro of follicular oocytes collected from nondomestic felid species. BIOLOGY OF REPRODUCTION 45:898-906, 1991c.

Kleiman, D.G. The estrous cycle in the tiger (Panthera tigris). In THE WORLD OF CATS: BIOLOGY, BEHAVIOR AND MANAGEMENT OF REPRODUCTION. R.L. Eaton, ed. Feline Research Group: Seattle, Pp. 60-75, 1974.

Leyhausen, P. Das verhalten der katzen. HANDBOOK OF ZOOL., 8:1-34, 1956. Schaller, G.B. THE DEER AND THE TIGER. University of Chicago Press: Chicago, 1967.

Seal, U.S. Long-term control of reproduction in female lion with implanted contra- ceptives. AMERICAN ASSOCIATION OF ZOO VETERINARIANS ANNUAL PROCEEDINGS, 1975.

Seal, U. Hormonal contraception in captive female lions. JOURNAL OF ZOO ANIMAL MEDICINE 7(4):12-20, 1976.

Seal, U.S.; Plotka, E.D.; Smith, J.D.; Wright, F.H.; Reindl, N.J.; Taylor, R.S.; Seal, M.F. Immunoreactive luteinizing hormone, estradiol, progesterone, testosterone, and androstenedione levels during the breeding season and anestrus in Siberian tigers. BIOLOGY OF REPRODUCTION 32:361-368, 1985.

Seal, U.S.; Tilson, R.L.; Plotka, E.D.; Reindl, N.J.; Seal, M.F. Behavioral indicators and endocrine correlates of estrus and anestrus in Siberian tigers. In: TIGERS OF THE WORLD, eds. R.L. Tilson and U.S. Seal, Pp. 244-54, Noyes Publications: Park Ridge, NJ, 1987.

Seifert, S.; Muller, P. INTERNATIONAL TIGER STUDBOOK. Zoologischen Garten Leipzig: Leipzig, 1985.

Wildt, D.E.; Phillips, L.G.; Simmons, L.G.; Chakrabort, P.K.; Brown, J.L.; Howard, J.G.; Teare, A.; and Bush, M. A comparative analysis of ejaculate and hormonal characteristics of the captive male cheetah, tiger, leopard and puma. BIOLOGY OF REPRODUCTION 38:245-255, 1988.

Wildt, D.E.; Phillips, L.G.; Simmons, L.G.; Goodrowe, K.L.; Howard, J.G.; Brown, J.L.; Bush, M. Seminal-endocrine characteristics of the tiger and the potential for artificial breeding. In TIGERS OF THE WORLD. R.L. Tilson and U.S. Seal, eds. Noyes Publications: Park Ridge, NJ, Pp. 255-279, 1987.