A major problem encountered in maintaining captive cetaceans is the integration of two or more animals in a confined area. Death due to lethal attacks on members of the same or different species constitute a large share of the conclusions as to cause of death on autopsy reports by any establishment maintaining at least several of these animals. In addition, there is no way of calculating how much of a contributing factor constant harassment is to other deaths.

A captive situation creates at best an abnormal social environment because of the inherent space limitations. Although practically nothing is known of the space requirements of marine mammals at sea, analogy drawn from other vertebrate groups including fishes, reptiles, birds and other mammals indicate definite individual distance requirements for practically every species. These requirements can be loosely classified under the word territory, although territory means many different things. According to species or individual, it can mean a certain fixed point in space, to another the distance between individuals, to others access to sexual partners or food. These requirements even within a species vary with sex, age, breeding season, time of day, exposure to outside threat, availability of physiological requirements, and the experience of each individual with each other individual.

Studies of social behavior in most mammalian societies demonstrate a sharp increase in territorial requirements in sexually maturing males. This brings with it the attendant aggressiveness necessary for the defense of territory, although many male mammals tolerate each other reasonably well except during the breeding season. In nature, most territorial conflicts are resolved at the territorial boundaries. Such settlements usually consist of elaborate threat displays, only occasionally resulting in a true flight wherein some damage may be inflicted. The usual result of such an encounter is a draw at the territorial boundary, or the vanquished occasionally flees the scene. Rarely is an animal pursued much beyond the territorial boundary of the victor, so the vanquished has little more than a few tooth marks to show for his trouble. Two large bull Tursiops truncatus at Marineland of Florida demonstrated a third solution to the ''fight or flight'' dilemma. During the day they were housed in adjacent tanks and spent much of their time threatening, posturing and jaw snapping at each other. When released together into a large swimming area at night, however, all hostility disappeared and they became old buddies. Whether this equilibrium was due to their being away from home base or their being an almost perfect physical match for each other is debatable.

Workable solutions are indicated in the wild in the toothed whales by the fact that stranded animals killed by their own species are not reported, although there are frequent reports of light tooth marks on newly-captured animals, the marks corresponding to the dentition of their own species. Such marks are found more frequently on males. Not all animals, even males, in a school are so marked, however. The males, and to a much lesser extent the females, of the Risso's dolphin or Grampus (Grampus griseus), however, are so heavily scarred by the tooth marks of their own kind that these scratches were considered by early writers to be typical pigmentation.

In captivity the animal which is offending a dominant male by his or her presence cannot leave the scene; hence an autopsy may be the next step unless the animals are separated.

A second unnatural social situation created by captivity is the sudden introduction of strange animals to each other. In nature, social groups are based on a relatively stable organization wherein each individual has the opportunity for long-term experience with other group members, and an animal is never introduced to the existing group without the opportunity to flee if not accepted. Dominance orders are frequently well established during the juvenile play fight period when no one gets hurt badly. When strange adults are suddenly confined together, dominance order must be established after the adult fighting weapons, as well as aggressiveness, are fully developed. Such fights may well be lethal. One animal importer learned that in one of the primate species it was useless to capture animals from separate social bands and ship them out as a group. They simply killed each other off at first exposure. Conversely, groups of this same species captured from a single herd, with previously-established early experience with the other individuals of the herd, got along harmoniously.

There are a few clues from behavior studies and observations at various aquaria that may help to prevent deaths from lethal attacks or stress-induced problems. One method that tends to diminish the problem of integration is to allow the animal that is expected to be low in the hierarchy to establish itself first in a territory. The little understood but very real psychological advantage of being the old resident frequently outweighs the major advantages of size or sex possessed by an animal secondarily introduced.

More than one mature bull in a community containing females is asking for certain trouble. Occasionally two or more bulls can live together without females, in a state of equilibrium, but this is an individual thing and the animals have to be watched for a time to be certain that the communal housing is going to be acceptable.

In most instances, a single bull in a tank with females and young appears to work out for about two years following onset of sexual Maturity in the male, and offspring will be born. After roughly two years of sexual maturity, presumably when the testosterone output is greater, attendants should be alert for indications of lethal attacks on other herd members from the bull. This is a particularly dangerous situation for the infants or juveniles of both sexes. At times these prime animals have been killed by bulls. We also have examples of rather brutal sexual attacks made by bulls on the young of both sexes. Another phenomenon occurring in the bottlenose dolphin (Tursiops truncatus) is the attraction of the other members of the tank, particularly the males, to the area in which an attempted copulation is ongoing. If a mature bull is interfered with at this time he may seriously attack the offender--of either sex.

These bulls in a community also routinely inflict damage on the mature females. Oddly enough, when housed with only one female, these attacks may not occur.

A bull also may interfere with the performance of the less dominant individuals in a trained act now so common in aquaria. For example, at Florida's Gulfarium, Ft. Walton Beach, the show was being disrupted by some unknown factor. A female normally closed the show with a magnificent 20-foot leap to take a fish from the trainer's hand. She began embarrassing the establishment by refusing to perform this act, although she was in good health and otherwise cooperative. J. B. Siebenaler, the curator and general manager of the Gulfarium, had the apparently inoffensive but dominant male removed. The show immediately proceeded normally for several days, with the female again making the high jump. Not wishing to leave the male in isolation for too long and hoping that the isolation treatment had left the male better disposed towards his former tank mates, Siebenaler replaced the male into the large show tank. The Caldwells were present with underwater recording gear at the time, and neither saw any aggressive action on the male's part as he remained away from the female and the jumping area; nor did they hear any unusual vocalizations.

Apparently, however, the simple presence of the male was enough to again intimidate the more submissive female into a refusal to perform. It was necessary to remove the male again, and the female immediately resumed her polished performance.

At Marineland of Florida, a large bull prevented the females and juvenile males From jumping by the much more overt behavior of jaw snapping, biting and chasing the others when they attempted to perform. Similar behavior has been exhibited during the jumping shows at Marineland by the two bull Risso's dolphins in the tank with the performing Tursiops.

Interference was also accomplished by a male Amazon freshwater dolphin (Inia geoffrensis) at the Fort Worth (Texas) Zoo. The male took over the entire show and refused to let the female perform, although they otherwise lived in perfect harmony.

It may be helpful at times to remove the bull and allow the other animals to establish territory for over a year, and even this will probably not hold for any great length of time. It may well hold long enough, however, for one or more conceptions to occur.

We have no firm data that indicate more aggressiveness at any one season of the year. Our records at Marineland of Florida show both sexual and fighting aggressiveness year round. More detailed studies may show some pattern to the contrary, however, as there is a slight tendency toward more problems in the Spring and Fall.

Whereas one female may be more dominant than another, or become more aggressive when she has a calf, we know of only one female that inflicted physical damage on another individual. In this case a mature female bottlenose dolphin (Tursiops truncatus) at Marineland of Florida would not accept newly introduced spotted dolphins (Stenella plagiodon) in her small holding tank. She was credited with killing two individuals, one each on different occasions, by ramming them so hard that severe internal injuries resulted in death. The victims were both females, and were in the tank one week or less. We do have records at Marineland of female eastern Pacific pilot whales, Globicephala scammoni harassing spotted dolphins placed in the holding tanks with them. In contrast, a strange female is usually accepted by, and swimming with, other females within a few days at most, after a brief period of careful examination which includes much probing of the urogenital region. Females also normally accept young of both sexes.

Live births in captivity usually pose no problems of maternal care, but there is one report of a mother's possibly intentionally drowning her infant by pushing it to the bottom of the tank (Tavolga and Essapian, 1957). At Marineland of the Pacific the Caldwells have seen two live births in which both mothers pushed their just-born infants to the tank floor and held them there forcibly for a few seconds. The behavior was exquisitely patterned, however. The timing was so regular (about every 45 minutes for a day or two following birth) that one could predict it. The duration of holding down was enough to cause the infant to struggle and emit a ragged whistle or squeal but not long enough to cause severe distress. Because of the regular patterning, we are forced to consider that some advantage to the infant or to the species is involved. A mother also has been reported to hold her juvenile down momentarily as a form of punishment (Caldwell and Caldwell, 1967), but here again the action lasted only a few seconds and no harm could have been inflicted.

Births may cause a husbandry problem in another way. Recently at Marineland of Florida an infant was born to one of the performing Tursiops truncatus in her holding tank in the Stadium. For several days the shows were disrupted when the other animals (both male and female Tursiops and female eastern Pacific pilot whales) delayed or interrupted their performance to come over and look at the new infant through the holding tank gate.

Much has been written regarding the value of permitting a sick or injured cetacean access to others which might aid it, and to not isolating sick animals when they may gain from the companionship of other healthy individuals of the same or different species. There is no doubt that many sick animals have been saved by this procedure, either directly by being physically lifted to the surface by others where it can breathe, or by the less impressive but just as real mechanism of social stimulation provided by the others. There is, however, another side to this coin. An animal may be well enough to eat if given a chance, but not well enough to compete for food. We have seen partially sick animals that would take food readily enough have the fish snatched from their mouths time and time again by tank mates. Their tank mates quickly learn that the sick animals are not able to feed properly and it appears in the Marineland of Florida community tank as though at least one other animal relentlessly stalks the sick one during an attempted feeding, although all save the stalker can be drawn away by food offered by other feeding attendants at a distance from the sick one. When the stalker moves away to feed, a new antagonist takes its place to stalk the sick individual. When the sick animal is removed from the community to a holding tank where it can get more individual and less disturbed attention, it again feeds readily and frequently can be nursed back to health to again take its place and compete for food in the community.

Again, in both the bottlenose dolphin and the Amazon freshwater dolphin, sick females have become the object of sexual harassment by any male in the tank, including the juveniles, to the point where the sick animal cannot rest or feed. Under these conditions, the animal has to be placed in different quarters, preferably with a female, but in isolation if necessary.

There is one other problem more directly related to the human element and the practical side of marine mammal husbandry. Healthy animals often stand Tavolga, Margaret, and Frank S. Essapian 1957. The behavior of the bottlenose dolphin (Tursiops truncatus): Mating, pregnancy, parturition and mother-infant behavior. Zoologica, 42(1): 11-31, 3 Pls.