Introduction

The importance of captive breeding programs for bottlenose dolphins has been increasingly recognized during the last two decades. (Ridgeway and Benirschke, 1977; Cornell and Asper, 1978; Cornell et al, 1982) As the capture of wild cetaceans for research and display is becoming a controversial issue, the demand for captive born animals to fill these needs can be expected to increase. The establishment of new breeding colonies and the expansion of present colonies requires substantial investment in time and money.(Cornell et al, 1982) One strategy to meet the increasing demand without increasing the breeding population would involve optimizing the survival of those calves that are being born. In a marine mammal census update by Cornell, et.al. (1982), the rate for stillbirth and early mortality in bottlenose dolphins was reported to be about 45% for the years 1976-1979, a rate similar to that reported at the Tursiops Breeding Workshop (Sweeney, 1977) for the years prior to 1975. This is a preliminary report on an ongoing study that is attempting to identify variables that might have an effect on early infant mortality (EIM) and survival. Data from only one facility, the Dolphin Research Center(DRC), are included in this report. Some preliminary analyses including data from the Miami Seaquarium may be available by the time this report is presented at the 1990 IAAAM Conference.

Description of Facilities

The first facility to participate in this study was the Dolphin Research Center (DRC), previously known as Flipper's Sea School and the Institute for Delphinid Research. DRC is located on Grassy Key in the Florida Keys. The dolphins are kept in the Gulf of Mexico in natural seawater pens which they share with other Gulf sea life including snappers, lobsters and shrimp. The area is subject to tides and to wave action, which may be turbulent in the winter. The water temperature averages 84°F in the summer and 70°F in the winter with maximum and minimum recorded temperatures of 90.5°F and 59°F. The largest pen is 79.9m by 42.1 m and has an average low tide depth of 8.2m. The smallest pen is 28.6m by 71.3m with a depth graded from 2.4m to 7.6m. Variable sub-divisions of these pens occurred during the 20 year interval. It was not always possible to determine how many dolphins had access to which pens when each birth occurred. The minimum space per dolphin was estimated to be 340,000 liters.

Records were available dating from 1970, but the data contained in them were highly variable. This was due in part to management changes over the 20 year span of time. DRC has provided care for dolphins belonging to other facilities, and the complete records for these dolphins were not available. The breeding population included 10 females that were in residence for variable periods of time. Three dolphins were resident from the early 70's to 1990. Another dolphin died in 1980 of septic metritis. Six dolphins have entered the population since 1984; two of these were transient. The exact age was known for 2 captive born females. Estimated ages at capture were used for 7 cows. The criteria for those estimations weren't available, and their accuracy may be questionable. The age estimated from a tooth pulled at necropsy in January, 1990, were used for the last cow. This estimate was 5 years greater than her estimated age at capture. The male population was variable, and, in most instances, paternity was not considered in this study.

Criteria for Selection of Cases

The records of all births (n=31) that occurred at DRC from January, 1970 through January, 1990, were reviewed. Birth peaks, calving intervals and individual cow's reproductive success were determined. Prenatal, perinatal and post-natal factors that might have affected survivability were identified by the examination of daily logs, medical records, necropsy reports and any other material available. The amount of data available for each birth was varied. Thus, not all births were included in the analysis of each factor considered. The two factors that were consistently recorded were the length of time that the calf survived and the estimated age at capture or date of birth of the cow. For other factors, only the cases with fairly definitive data pertaining to that factor were included. Organization of the data was done post hoc.

The births were divided into three categories that have been given labels for convenience of referral. A birth was considered successful if the calf survived weaning and was at least two years old. The youngest calf at DRC was included in this category since he will be two years old at the time this paper is presented. This category will be called Survivors. The second category including those calves that survived less than three months will be called Early Infant Mortalities (EIM). The third category including those calves that lived longer than the EIM interval but did not survive weaning will be called Later Infant Mortalities (LIM).

Results

Of the 31 births that have occurred over the last 20 years at DRC, 12 were Survivors, 4 were in the LIM category and 15 in the EIM category. The high rate of EIM may have been related to the large number of primiparous females in the breeding population and to one female, LB, with a very low reproductive success. These two segments of the population accounted for 11 of the 15 EIM (73%). First and second pregnancies accounted for 11 out of 15 EIM(l 1/15, 73%). Births occurred in every month except February and June, with a birth peak occurring in from August through December. A 30 day interval without births in mid-Fall facilitated the division of the peak into Fall and Winter to correlate the effect of season on survival. A much smaller Spring peak occurred from March through May.

Table 1.

Cow Demographics - The time of conception, and thus the length of gestation, was known in two cases where access to a male was limited to two days. The length of gestation for Bee's 1977 calf was 365-367 days and her 1987 calf, 378-381 days. Access to males could not be verified for 2 females, Bee and Shan, during the interval from 1972 to 1976. Variations in the interval after the previous calf was weaned or died until access to males ranged from 0 to 6 months. No attempt was made to correct for this variation in figuring calving intervals or reproductive success. The average calving interval was 33.5 months for the whole population, 22.8 months for the cows with EIM, 32 months for LIM and 40.8 months for the cows with surviving calves. The longest interval, 7 years, was not used to figure the averages, since access to males for that cow could not be confirmed for most of those years. The range of calving intervals with a surviving calf was 34 to 48 months and following EIM was 12.8 to 25 months. Three cows, LB, T and Bee, had four or more calves spanning an interval of 15 to 20 years and were parturient at 26-7 years of age. The calving interval for 1 cow (Bee) decreased from 46 to 36 months for her last calf, and the other 2 were relatively consistent. T's first 2 calves were EIM, and her calving intervals were 12 and 13 months. The same cow(T) also conceived within 1-4 months of the death of her 2 LIM calves, demonstrating that conception may occur very soon after both EIM and LIM.

Reproductive success was calculated for cows that had 3 or more calves. Since the age at primiparity can be highly variable (Cornell et al, 1987), the interval from the cow's first conception to the end of the survey period was used as the years available for reproduction. The number of years available for reproduction was divided by the average successful calving interval of 3.4 years to get the maximum number of calves that cow could have produced. That figure was then divided by the number of surviving calves actually produced. LB had the poorest reproductive success (0.36) and was the only cow to have EIM occur after the second calf. Bee was the most successful cow with 4/4 calves surviving (0.95). Her first 5 years in captivity were not considered since access to males could not be confirmed.

Table 2.

Prenatal Factors

Age at Parturition - Three cows calved at 7 years of age. An exact age of 7 years, 3 months was known for a captive born cow. The 3 oldest parturient cows were 26-27 years old, and a 31 year old cow is currently in her third trimester of pregnancy. Sergeant et al(1973) estimated age at sexual maturity to be 12 years old. Data from a wild population in Sarasota bay suggest that calf survival rates in the wild are low for cows under 15 years old. (Wells, pending) Age at parturition was organized in two age divisions -cows 12 years old and under vs. cows over 12, and cows 15 years old and under vs cows over 15 years old. Table 2 lists the results. Of the 8 cows to give birth at DRC at 12 years old or younger, 7 were primiparous, and their calves were EIM. One of these cows had a second calf at 10 years of age that survived. Three multiparous cows produced 2 calves each between ages 13 to 15 years. Their second calves were all EIM, and 1 third calf was LIM. The other 2 survived. A primiparous female and a female who was pregnant when captured (est. age-1 2 yrs.) both had surviving calves. Calf survival rates for the over 12 yr the over 15 year old calves were similar.(48% vs.47%).

Primiparity - Two cows were 12 years old at capture and may not have been primiparous. The others were either born in captivity or captured at an estimated age of six or less. One cow aborted, and another's calf was found dead, leaving 6 primiparous births where mother-infant behavior was observed. Maternal incompetence, either as negligence or inability to defend her calf from conspecifics, was noted in 5 of these 6 births. The only primiparous cow over 12 years old exhibited some degree of maternal incompetence, but her calf did survive. One 7 year old cow appeared competent, but her calf died after 12 days. No records as to cause of death were available. The opportunity as a juvenile to observe and participate in the successful rearing of a calf was documented for the 2 captive born cows. One cow that was 7 years old proved to be incompetent. The other cow's first calf was still born. She was 13 years old when her second calf was born. She proved to be competent, but her calf succumbed to aspiration pneumonia.

Other Factors - Diet appeared adequate in all cases where information was available. Husbandry techniques were difficult to evaluate from the records available but appeared adequate. Pregnancy diagnosis prior to 1980 was primarily done by physical examination. Progesterone assays have been used since 1980. Medical histories revealed several factors that might have had an effect in isolated cases, but no trends were discovered. One cow aborted a 9 lb. fetus 3 weeks after being transported from DRC's Key West facility back to the Grassy Key maternity pool. Seven other dolphins appeared to carry to term, 3 to 6 months after the transport. Two of these were EIM, but no necropsy records were available to determine if any abnormalities were present that could have been due to transport. The third cow, Misty, also had chronic respiratory problems and was given antibiotic therapy in the 1st, 6th, and 9th months of pregnancy. She was also given a bronchodilator, aminophylline (200mg, TID) during the last trimester. Other than a chronic cough, her physical condition appeared to be good the time of parturition. Her calf was found dead, and the necropsy diagnosis was dystocia. Aphro was treated in the 1 st trimester with an antibiotic, Keflex(2g, BID, 10 days) for a high white blood cell count(WBC). There were no problems in the rest of her pregnancy. Her calf was stolen by another female, and the necropsy diagnosis was failure of passive transfer(FPT). Aphro was given the same therapy for the same problem in the 2nd trimester of her 2nd pregnancy, and the calf was a survivor.

Perinatal Factors

Factors Predictive of Parturition - Reliable data on changes in food consumption prior to parturition were available for 28 births. Decreased appetite for 1 to 4 days prior to parturition was present in 21/28 cases (75%). Behavioral changes, primarily lethargy, were present in 14/28 cases (50%). Physical changes, primarily a swollen slit, were present in 4/28 cases (14.3%). The figures for physical and behavioral changes may be low because changes present may not have been recognized and/or recorded.

Negative Social Interactions - Interactions in the first 24 hours that put the calf at risk, either directly through trauma or indirectly through interference with nursing leading to failure of passive transfer (FPT), were called negative social interactions (NSI). NSI occurred in 8/26 cases (30.8%) where the infant was observed alive, with 6 being EIM. Five involved primiparous cows, and the other three involved the cow, LB. Parturition was observed on 10 occasions, 4 of which involved NSI. NSI were still evident when the calf was first seen in the other 4 cases of NSI.

Composition of the Social Group in relation to NSI - Group size ranged from 2 adults to 8 adults and juveniles. NSI occurred in all group sizes. Only 1 case of NSI, involving a lactating cow, occurred out of 10 groups containing only juveniles and/or pregnant and/or lactating females. The animal that was observed to initiate the NSI or that was actively resisting the mother's attempts to retrieve the calf was considered to be the instigator of the NSI. Adult males and open cows, those neither pregnant nor lactating, appeared to be the instigators in the other 7 NSI. One or more adult males were part of the group for 7 births, 4 of which involved NSI. In 2 of these cases, LB was the only female. Bee also had a calf with only males present, but there were no NSI. The male was not the instigator in the other 2 cases although they did participate once the action began. In these 2 cases and in 3 other cases of NSI, open adult females appeared to be the instigators. Three of the instigating females had lost a calf within the past year. The remaining 6 groups included open females without the occurrence of NSI. In one case, the parturient cow's mother was the open female. In the other 5 groups, the parturient cow was over 15 years old and had had long term relationships with the other animals. The same was true for the 3 groups without NSI that included males. Table 4, at the end of this paper, is a summary of all births that have occurred at DRC. A highly abbreviated indication of social composition is given in that table.

Table 3.

Sex of Calf - Male mortality, 11/15 calves (73%), was much higher than female mortality, 4/12 calves(33%). The 4 LIM were all male calves. If only EIM are considered, the disparity between males and females was not as great (46.7% vs. 33%).

Environmental Conditions - The time of year could be related to mortality in only one instance. An unusually severe winter storm left a 13 day old calf stranded in the mangroves. The calf died of aspiration pneumonia 4 days later. There were 11 instances of severe weather during the first month, with eight calves surviving. The necropsy diagnoses for the other 2 calves that died, T2 and LB8, could not be related to weather.

Discussion

Because our increased knowledge both of dolphin physiology and of dolphins in the wild has led to many changes in husbandry over the past 20 years, a review of the changes in husbandry practices at DRC was not feasible. For example, male aggression towards infants has been well documented (Caldwell and Caldwell, 1972), and no males have been kept with pregnant females at DRC in the last 10 years. It is recognized that good husbandry is vital for successful births. Cornell et al (1987) provide an excellent review of the husbandry techniques used in Sea World's dolphin breeding program. This study has focused instead on 1) providing general information on the results of DRC's breeding program and 2) identifying factors that might have contributed to infant mortality and correlating their occurrence with survivorship and mortality.

Because of the high incidence of stillbirths and early infant mortality reported in bottlenose dolphins, it was expected that more deaths would have occurred in the first three months than in any subsequent three month interval. (Cornell and Asper, 1978; Cornell et al, 1982) In fact, 79% of the mortality occurred in the first three weeks. The LIM were all 6 months or older. A high proportion of EIM and all LIM were males. Excessive male mortality has been well documented in many mammalian species, particularly for juvenile and sub-adult males. (Daly and Wilson, 1983) The small sample size precludes making any inferences about excessive male mortality in male dolphin infants, but the fact that all LIM were males is worth further investigation.

Data from Wells (pending) wild Sarasota population of bottlenose dolphins suggests that most wild dolphins under 15 years of age fail to raise calves successfully. Although the distributions of EIM, LIM and survivors were quite similar for the over 12 and over 15 year old dolphins at DRC, the distributions for dolphins 15 and under vs. over 15 indicate that Wells observations from wild dolphins might also apply to captive dolphins. Data from one cow with low reproductive success, LB, may have caused the distribution for cows over 15 to be skewed. Removing her 5 calves from the sample would have given a distribution of 20% EIM, 20% LIM and 60% survivors. A larger sample is needed to dilute the effects of data from unsuccessful cows like LB.

Primiparity and age at parturition are closely related since most primiparous females are also young. Estimated ages for sexual maturity in female dolphins range from 5 years (Harrison et al, 1969) to 12 years of age. (Sergeant et al, 1973). Cornell et al (1987) suggested immaturity as a cause for maternal incompetence in a 7 year old cow at Sea World. The difference for EIM between dolphins 12 and under vs. over 12 was substantial, and if only primiparous cows under 12 are considered, the rate for EIM would be 100%. From the DRC data, the calves of primiparous females 12 years old and under appear to be at greater risk for EIM. The rate of El M for second calves was also high, 5/7 or 71 %, while the rate for third calves was substantially lower, 3/4 or 25%. Again, a larger sample is needed.

The importance of practice in learning maternal behavior has been documented in some primate species, and maternal incompetence in captive populations has been related to lack of opportunity for practice. (Hrdy, 1976) The importance of allowing juvenile females the opportunity to observe and participate in mother-infant interactions has also been recognized in dolphins. (Cornell et al, 1987) For Anna, her earlier opportunities for practice appeared to be insufficient preparation for maternity at 7 years of age. Tursi was both older and had a stillborn calf. She appeared competent with her second calf, but a more experienced cow might have been able to protect its calf from the effects of the storm. The cow, Bee, had 12 years of experience as a young dolphin in the wild, then 5 years in captivity before her first calf was born with 3 males present and no females. It is interesting to note that Bee was perfectly in control and that the males kept their distance. The data from DRC suggest an interaction between age, immaturity, primiparity and lack of practice. DRC has not had a stable breeding colony such as that described for Marineland of Florida. (Tavolga and Essapian, 1957) Under these circumstances, the social composition of the group may be an extremely important factor in controlling the incidence of EIM in young cows. NSI in the first 24 hours may cause traumatic injuries to the infant and/or prevent nursing. Dolphins have epitheliochorial placentation (Slipjer, 1966), so the ingestion of colostrum in the first 24-48 hours is essential for passive transfer. In bovines, maximum intestinal uptake of antibodies occurs in the first 8-12 hours after birth and steadily decreases until pore closure at 36-48 hours. (Bainter, 1986) The critical period for maximum uptake in dolphins has not been identified, but if a similar pattern exists, infants with limited opportunities to nurse in the critical period will have limited immunity and increased susceptibility to infection. A calm social environment during that critical period would give the unpracticed mother the opportunity to gain the maternal skills necessary to allow nursing as quickly as possible.

In Well's Sarasota population long term associations between mothers and their daughters were common.(Wells, 1987) This association might afford both opportunity for practice by observing younger siblings and help from their mother in raising their own infant. Aunting has been documented in dolphins both for adults and juveniles.(Essapian,1963, Tavolga and Essapian, 1957). In some primate species where aunting has also been documented, aunts are more likely to be nulliparous daughters. It has also been noted that there may be a continuum between aunting behavior and kidnapping, with the other female's attraction to the infant vs. the mother's ability to retrieve the infant determining which occurs.(Hrdy, 1976) The mother might be less able to retrieve an infant from a more dominant female. The existence of dominance hierarchies in captive dolphins has been documented. (Caldwell and Caldwell, 1972; Corkeron, 1987) Again in primates, juveniles and sub-adults tend to be subordinate to adults unless they share in the dominance of an adult, such as their mother. (Bernstein, 1981) Young unpracticed dolphins without such an alliance may be unable to retrieve their calves from a more dominant adult female highly attracted to the infant. In 5 out of 6 primiparous births at DRC, the infant was kidnapped by an adult female, 4 open, one lactating. One of these involved a 7 year old captive born cow, Anna, whose mother was not present. Tursi's mother was present at her second calf's birth, and there were no problems. However, Tursi was both older (l 3 years) and had aborted her first calf. Only a 7 year old female was present at the birth of Aphro's second calf, the only survivor born to a cow under 12. This successful grouping gave minimal interference to the mother and afforded an excellent opportunity for practice to the younger female.

There may be a minimum age at which cows may be expected to raise an infant. Including those at DRC, all 5 known births to 7 year old cows resulted in EIM. Even under what appeared to be ideal conditions, the calf of a 7 year old cow at Marineland of Florida was EIM. Extensive aunting behavior was observed for Spray, a captive born cow. Her mother was present at the birth and exhibited aunting behavior. No NSI occurred and nursing in the first 24 hours was observed. However, Spray exhibited maternal negligence, and the calf was malnourished when it died of respiratory arrest 9 days later. (Tavolga and Essapian, 1957) Caldwell and Caldwell (1977) suggested a hormonal deficiency might be implicated in maternal incompetence for Spray. Six out of 8 cows 8-12 years old have been successful breeders in similar stable colonies at Sea World.(Cornell et al, 1987).

A careful evaluation of the group composition considering reproductive status, kinship and long-term associations, age and dominance relative to the pregnant cow may help to prevent EIM that are due to trauma or FPT in young cows.

Bibliography

1.  Bainter, Karoly. Intestinal Absorption of Macromolecules and immune Transmission from Mother to Young. CRC Press. Boca Raton, Florida, 1986.

2.  Bernstein, Irwin S. Dominance: The baby and the bathwater. The Behavioral and Brain Sciences (1981) 4: 419-457, 1981.

3.  Caldwell, M.C. and D.K. Caldwell. Social interactions and reproduction in the Atlantic bottlenose dolphin. In: Ridgeway, S.H. and K. Benirschke, (eds.) Breeding Dolphins: Present Status, Suggestions for the Future. Final Report to the U.S.M.M.C. #MMC-76/07. Zoological Society of San Diego, San Diego, California. pp. 132-144, 1977.

4.  Corkeron, P. Social behavior. In: Harrison, R. and M.M. Bryden, eds. Whales, Dolphins and Porpoises. Intercontinental Publishing Corp. Hong Kong. Pp. 142-147, 1987.

5.  Cornell, L.H. and E.D. Asper. A census of captive marine mammals in North America. International Zoo Yearbook 18: 220-224, 1978.

6.  Cornell, L.H., E.D. Asper and D. A. Duffield. Census update: Captive marine mammals in North America. International Zoo Yearbook 22: 227-232, 1982.

7.  Cornell, L.H. et al. Progress Report: Results of a long-range captive breeding program for the bottlenose dolphin, Tursiops truncatus and Tursiops truncatus gilli. Zoo Biology 6: 41-53, 1987.

8.  Daly, Martin and Margo Wilson. Sex, Evolution, and Behavior. Willard Grant Press. Boston, Massachusetts. Pp. 92-96, 1983.

9.  Essapian, F.S. Observations of abnormalities of parturition in captive bottlenosed dolphins, Tursiops truncatus, and concurrent behavior of other porpoises. J. of Mammalogy 44: 405-414, 1963.

10. Hrdy, S.H. Care and exploitation of nonhuman primate infants by conspecifics other than the mother. In: Advances in the Study of Behavior, Vol.6. Academic Press, Inc. New York. Pp. 101-149, 1976.

11. Ridgeway, S.H. and K. Benirschke, eds. Breeding Dolphins: Present Status, Suggestions for the Future. Final Report to the U.S.M.M.C. #MMC-76/07. Zoological Society of San Diego, San Diego, California, 1977.

12. Slipjer, E.J. Functional morphology of the reproductive system in cetacea. In: Norris, K.S., ed. Whales, Dolphins and Porpoises. University of California Press. Berkeley and Los Angeles, California. Pp. 287-313, 1966.

13. Sweeney, J. Difficult births and neonatal health problems in small cetaceans. In Ridgeway, S.H. and K. Benirschke, eds. Breeding Dolphins: Present Status, Suggestions for the Future. Final Report to the U.S.M.M.C. #MMC-76/07. Zoological Society of San Diego, San Diego, California, 1977.

14. Tavolga, M.C. and F.S. Essapian. The behavior of the bottle-nosed dolphin (Tursiops truncatus): Mating, pregnancy, parturition and mother-infant behavior. Zoologica, Part 1, 42: 11-31, 1957.

15. Wells, R.S. Population structure of bottlenose dolphins: Behavioral studies along the central west coast of Florida. NMFS Contract 40-WCNF-00366, 1987.

16. Wells, R.S. and M.D. Scott. Parental investment patterns of free-ranging bottlenose dolphins near Sarasota, Florida. Pending.