Acoustic Communication

Although communication in cetaceans is by no means limited to the acoustic mode, it is the one most often mentioned, and is probably the primary sensory system in this group of animals. In fact (see Wood, 1973, for a good discussion of the subject), the large, well-developed brain of cetaceans is almost surely related more to the acoustic system (probably echolocation in particular) than to the high degree of "intelligence" often ascribed to these mammals. Also, acoustic data are gathered on a quantitative basis more easily than data concerning other sensory and communication systems. Thus, more is known about acoustic systems of cetaceans than about their other sensory systems such as touch and vision.

From the outset we want to make it clear that we do not believe that dolphins or any other cetaceans "talk." They certainly communicate via several sensory modalities, including sound, but despite a rather extensive popular literature to the contrary, they do not talk in the sense that humans talk (i.e., passing abstract information and ideas by the use of a language consisting of words that have specific meanings). There is good evidence that at least some cetacean sounds provide general information to members of a group, and they probably convey general states of emotion as well. There is also reason to believe that individuals recognize other individuals through the acoustic mode, but there is no evidence yet to support the concept of a true language used by dolphins or any other cetacean. As Norris (1974:196-97) noted, this human "method of acoustic communication is almost grotesquely clumsy and difficult" probably arising of necessity out of our use of tools.

Useful reviews of cetaceans known to produce sounds have been provided by Evans (1967) and Poulter (1968), and other shorter reports have added a number of species and even representatives of larger systematic groups to their lists. Sounds of the mysticetes include chirps, cries, moans, and clicks, and in addition to these the minke whale (Balaenoptera acutorostrata) may produce a whistle. These all have been ably reviewed to species recently (Winn and Perkins, 1976). Cummings (1971) has prepared a useful tape recording of undersea sounds that includes many emitted by both mysticetes and odontocetes. Species or representatives of larger groups that still have not been recorded undoubtedly will be found to produce some degree of sound as opportunities are found to make sound recordings. It is unlikely that any species of cetacean is mute. When cetaceans have been recorded, they have all been found to vocalize to some degree, and it is likely that at least some of these vocalizations are emitted in social situations and therefore have some communicative function. We know, for example, that the Atlantic bottlenosed dolphin is a very social species, and many of its sounds are clearly emitted in a social context —but other sounds are not.

Cetacean vocal emissions may be broadly divided into two categories: so-called pure-tone sounds, which are usually termed whistles or squeals (see Tavolga, 1968, for examples), and pulsed sounds. The latter in turn may be divided generally into (a) the trains of regular clicks emitted in exploratory or environmental search situations (echolocation; see Norris, 1969, for examples) and (b) those sounds emitted in an emotional context (usually termed burst-pulse sounds that are variously described as barks, yelps, squeaks, squawks, grunts, moans, and so on). (See M. C. Caldwell and D. K. Caldwell, 1967, for examples.) As Schevill (1964) pointed out, all odontocetes probably produce pulsed sounds and there is good evidence that mysticetes produce only this kind of sound. Whistles are known to be produced by dolphins and "whales" of the families Delphinidae, Monodontidae, Ziphiidae, and Steniidae (if one recognizes the family Steniidae as valid), and possibly by the minke whale (see above). The Phocoenidae and Platanistidae are not known to whistle although, like the others, they are known to produce pulsed sounds, which in many cases are complex and surely must have value for communication. The Physeteridae certainly produce pulsed sounds, but whether or not they whistle is still a matter of some controversy.

There is a considerable literature (see D. K. Caldwell and M. C. Caldwell, 1972a) in which attempts have been made to contrive some sort of complicated language from dolphin whistle contours. We have found instead that normally each individual dolphin has its own distinctive whistle contour, which we have termed its "signature" whistle (Caldwell and Caldwell, 1965). We have demonstrated the existence of this signature whistle primarily in the Atlantic bottlenosed dolphin (Caldwell, Caldwell, and Turner, 1970; Caldwell and Caldwell, unpubl.). There is evidence for it also in the saddleback dolphin (Caldwell and Caldwell, 1968), in the Atlantic spotted dolphin (Caldwell, Caldwell, and Miller, 1973), and to a lesser extent in the eastern Pacific white-sided dolphin (Caldwell and Caldwell, 1971), as well as other dolphin species. Although the rapidity, loudness, and/or duration of each whistle emission may vary, the basic contour for most animals remains much the same in a variety of situations in the life of the animal (after the whistle has developed; see Caldwell and Caldwell, in press). It is probably the variations in the signature whistle of an individual that carry a good share of the communication load, rather than the use of many different contours by the same individual. The individual contour, on the other hand, tells the rest of the community which animal is whistling. This identifying function may be even more important socially than a whistle language. We have found that on occasion a dolphin may emit only a portion of its signature whistle or may repeat it several times without stopping. Also, underwater sound recordings by hydrophone do not indicate which of several animals is vocalizing, and these tapes can indeed sound complex.

Excited dolphins tend to produce shorter whistles or an almost continuous whistling. The short ones usually occur when the animal hears other animals and stops to listen, or when it is closely attending to its environment. For instance, delayed playback (0.175 sec) of their own sounds to killer whales in the wild were found to inhibit rather than stimulate ongoing vocalizations (Spong, Bradford, and White, 1971). This pausing makes good biological sense because it is doubtful that dolphins can vocalize and listen attentively simultaneously. Our own experiments have shown, however, that a dolphin can distinguish between individuals' signature whistles by hearing only a small portion of one. Onehalf second is enough to identify one whistling dolphin to another.

Several investigators have examined the possibility that dolphins may transfer abstract information by electronic acoustic links or sound playback systems. Wood (1973:108ff.) reviewed the published literature in this area and concluded that the experiments did not prove that dolphins have any language ability but rather only that dolphins tend to respond to another's whistle and that they can be conditioned to respond to sound cues that are sometimes subtle.

We know from behavioral observations and experimental work that a dolphin can determine with great precision the direction from which it hears most pure tones. (Caldwell et al., 1971; Renaud and Popper, 1975). Presumably it can do at least as well for pulsed sounds. Thus, in our experience with whistles in particular and with pulsed sounds from presumption, we can say, for example, that a dolphin whistling with frequent, loud, short-clipped emissions apparently conveys the following information to another dolphin: "Attention! I am a dolphin with which you are (or are not) familiar. I am located to your left, and I am to some degree excited." Contrary to the beliefs of other students of dolphin sounds, we believe that distress is indicated in this manner rather than by a special "distress whistle." Dolphins that we have recorded under all kinds of stress (distress), including capture, intensive medical examinations, and grounding, have produced no whistles other than excited variants such as vocal "quaver" or "breaks" in the signature whistle. When we have examined data presented by other researchers that is intended to demonstrate a distress whistle, we have found evidence only of the signature whistle of an individual that happened to be recorded in a very stressful situation.

It is apparent that the whistle sounds emitted by one dolphin have a noticeable effect on others within hearing range. Inasmuch as the whistle is more easily quantified than pulsed sounds, the effect of this type of sound emission has been studied more closely than the effects of pulsed sounds. We do believe, however, that both pulsed and whistle sounds are communicative in nature and, further, that they have such value from the day the dolphin is born.

We do not know if a cetacean calf is born with the ability to echolocate effectively, but we do now have hard data for several species that show that the animal is able to produce what appear to be good trains of echolocation clicks on the first day of its life. We have found that several species of cetaceans are capable of producing whistles on the first day of life as well, and although the whistles are subject to modification to the signature whistle for that individual (Caldwell and Caldwell, in press), they do appear to be recognized by the mother and of use to her in identifying and locating her calf if she should become separated from it.

After the sound production for a given animal is established in its own repertoire, there is still a question of the degree of communication provided, by the pulsed sounds in particular. We have already noted that we believe the whistle is used primarily to identify an individual, assist in locating it, and provide some indication of its emotional state. We do not know if there is some signature value also in the pulsed sounds produced by individuals. Pulsed sounds are so complex that it is hard to say at what mathematical point some of them become useless for obtaining environmental information (i.e., as echolocation signals) and must therefore be emotional. Nor do we yet know the limits of the acoustic analyzer of the animal in distinguishing the variety of pulsed sounds that are produced. We have recorded some twenty general kinds of pulsed sounds made by Atlantic bottlenosed dolphins that seem to be correlated with general kinds of behavior, but only one of these sounds (a socalled male sexual yelp) has shown an indication of being behavior specific (Puente and Dewsbury, 1976), and even then it certainly is not always emitted at times when an observer would expect it. Most pulsed sounds seem to intergrade when large numbers of them are considered, and it is difficult to make clear-cut statements about them. For example, the "squawk" produced during a play chase is difficult to distinguish from the squawk made during an aggressive fight chase. Likewise, it is difficult to say exactly when a play chase develops into a fight chase. Additional study might resolve a few of the problems regarding the specificity of pulsed sounds as they relate to behaviors, but we can add little to what we wrote about the matter some years ago, summarized here. Captive Atlantic bottlenosed dolphins can voluntarily shape both whistles and pulse-type sounds (M. C. Caldwell and D. K. Caldwell, 1967, 1972). This behavior can be conditioned or it can result from an extended isolation from other dolphins. We have found no evidence for it in either undisturbed captive communities or wild populations.

In recent years considerable field work has been conducted on wild populations of the humpback whale (see, for example, Payne and McVay, 1971; Winn, Perkins, and Poulter, 1971). When they are in the southern parts of their range (in Bermuda and the West Indies), these whales produce long series (lasting many minutes) of different pulsed sounds that have come to be known as "songs." The series of sounds may be repeated on a predictable basis over and over again, and apparently each individual has its own song, which resembles the signature whistle in dolphins. There may be geographical dialects as well (Winn, pers. comm., 1973), so that a listener may be able to distinguish not only an individual but also the shallow West Indian bank from which that whale comes. We have not been able to demonstrate such geographical dialects in dolphins, but they are possible.

Echolocation is primarily a sound system used for environmental exploration (Norris, 1968, 1969) that is utilized by a wide variety of cetaceans (Evans, 1973). We have also noted that it is not always clear just when echolocation grades into pulsed sounds that are strictly emotional. However, even when these sounds are being used for environmental exploration, they may still have some communicative value to other members of a cetacean herd. Extensive echolocation, for example, may indicate to another animal that there is a strange object or animal in the immediate vicinity, or that food may be present and is being actively hunted. Hunts are often accompanied by considerable echolocation even when the prey is visible.

There is still controversy regarding the site of vocalization in cetaceans, but a growing literature (see, for example, Norris et al., 1972; Hollien et al., 1976) demonstrates that both clicks (pulses) and whistles are produced in a series of air sacs that lie above the bony cranium in the soft-tissue region around the blowhole on the top of the head. There is also evidence to suggest that the sounds may be directed forward and out of the head through a "tunnel" formed by lipids of varying densities that make up the soft fatty melon (see, for example, Norris, 1968; Litchfield and Greenberg, 1974; Litchfield, Karol, and Greenberg, 1973; Norris and Harvey, 1974; Litchfield et al., 1975). This feature of the melon has sometimes been referred to as an "acoustical transducer" and a "sonic lens" (see, for example, Wood, 1961, and Norris, 1968). The top of the bony cranium may also serve as a sound reflector, and there seems to be a very generalized correlation between the shape of this bony surface and the ecological need of the species to echolocate. For example, the cranium is frequently more cupped in contour when a more precise echolocation beam would seem to be required in a murky-water habitat where vision would be less useful (Wood, 1964).

In addition to the whistles produced by some cetaceans and the pulsed sounds produced by probably all of them, most cetaceans produce a variety of other sounds that have communicative value. They can slap their tails, flippers, heads, or entire bodies vigorously on the surface of the water. This results in a loud sonic report that can be heard both in and out of water. The context appears to be one in which the animal is disturbed or "angered" by some outside stimulus, or in the case of some captive animals by a trainer's tardiness in bringing the food bucket. Dolphins often snap their jaws together with such force that it makes a noticeable sound, and they may do this either out of the water or under water. This behavior, too, appears to indicate "displeasure."

Atlantic bottlenosed dolphins in captivity might squirt or splash water on a trainer and even strike out at him with the side or tip of the snout. These actions are often accompanied by the loud, raucous pulsed sound, made with the blowhole out of the water, that is sometimes called a "Bronx cheer."

In Atlantic bottlenosed dolphins, and probably other cetaceans as well, excitement may result in a loud, explosive exhalation of air through the blowhole out of water. Like the slapping of the water's surface, this sound is made when the animal is disturbed. Normal exhalation during regular breathing (blowing) is fairly quiet, and explosive exhalations appear to be communicative. A loud blow by one animal is frequently followed by the same type of exhalation by other animals nearby. In captivity blowhole responses may be elicited from animals in adjacent tanks which are neither in visual contact nor under the same stressful stimuli. Dolphins may also emit a large bubble of air from the blowhole under water without vocalizing. This makes an audible sound that behaviorally seems to suggest inquisitiveness or, in some cases, surprise. Amundin (1974) suggested that this release of air by the harbor porpoise may be displacement behavior.

Like those of land mammals, the digestive systems of cetaceans frequently rumble. We record these sounds most often when the animals have just been fed, and it is likely that other cetaceans recognize the rumbles as indicators of recent feeding and, therefore, the presence of food.

The very act of swimming may convey a message to other individuals in a herd, particularly if the swimming is rapid and accompanied by "porpoising" movements at or near the surface of the water. The sounds of rapid swimming might signal escape from danger, a chase for food, sexual activity, or merely play. Less-excited movements perhaps indicate that all is well and normal.

We have "recorded" many times yet another factor that surely has communication value to a group of cetaceans. When a strange animal or object enters the environment, an abrupt silence usually results, just as the forest goes silent when a hunter first enters. Vocalizations cease, breathing can hardly be heard, and swimming is almost noiseless. This sudden absence of sound can no doubt be an alert for danger as meaningful as a positive signal. Vocalization in the dolphin community resumes only gradually, usually with short echolocation bursts (apparently used in an exploratory manner) or with whistles so brief that they are only "chirps."

If we are to prove that sound plays a part in the communication process in cetaceans, we must clearly show that the animals are able to perceive the sounds at the levels at which they are being produced and to which behavioral responses (i.e., communication) should therefore be expected.

The potential methods by which cetaceans receive sound have been studied rather extensively from an anatomical point of view (Tavolga, 1965), but only recently has there been much experimental work on the subject (Norris, 1968, 1974; McCormick et al., 1970). The results of these experiments (often corroborated by simple behavioral observations) now suggest that the lower jaw of dolphins may function as the primary acoustic receiver for at least the sounds in the higher frequencies (50 kHz or more). Most of this work has been done by recording responses to sound stimuli directly from the auditory centers in the brain (Bullock et al., 1968; Bullock and Ridgway, 1972). The midbrain has loci responsive to echolocation-type click sounds; totally separated, the whistle-sensitive loci are primarily cortical.

Using behavioral techniques with free-swimming Atlantic bottlenosed dolphins, Johnson (1967) has shown that these animals are able to hear effectively frequencies from about 200 Hz up to about 150 kHz, or in the range where they produce the most sounds. Although most of the experimental work has been done with dolphins of this species, similar studies have used other odontocetes with similar (within variable species limits) general results. Findings indicate that cetaceans of a given species should be expected to hear the sounds that others of their kind produce and that many of the sounds might therefore be communicative.

It is important to remember that cetaceans are intensely aware of any sound they can perceive, and they are easily conditioned to it. Because they perceive and discriminate a wide range of sound (see, e.g., Herman and Arbeit, 1972), there is a strong probability that they use sound in the communication of a variety of emotions in addition to relaying such information as the presence of food or danger. Again, however, there is no evidence that with their repertoire of sounds cetaceans of any kind communicate in an abstract language, nor, as we are even sometimes asked, by mental telepathy. Norris (1974:197) introduced an idea worthy of consideration. He suggested that dolphins may offer considerable passive acoustic information to other dolphins. While an individual may be making no sounds of its own, another dolphin may be able to scan acoustically this individual. Thus, by using its highly refined echolocation system, the scanning dolphin may extract information from the body and nasal cavities of other dolphins that reveals their emotional state.

Visual Communication

It has long been suspected from behavioral observations, especially when they are made in captivity, where good visual acuity is required for the execution of many conditioned behaviors, that the Delphinidae have good vision. Only recently have experimental, detailed anatomical studies supported this theory (see, for example, Hall et al., 1972; Dawson, Birndorf, and Perez, 1972; Perez, Dawson, and Landau, 1972; Schusterman, 1973; Dawson and Perez, 1973; Dral, 1972, 1974). There is now good evidence that dolphins see well both in and out of water (Herman et al., 1975), and studies of their anatomy further suggest that they may have color vision as well as black and white. In our early research (D. K. Caldwell and M. C. Caldwell, 1972a, 1972b) we referred to "stereoscopic" vision in the Atlantic bottlenosed dolphin. Dawson, Birndorf, and Perez (1972) have used the more correct term "binocular" vision for this species (Fig. 1).

From both behavioral observations and anatomical data, then, there is good evidence that, except for the blind Ganges River dolphin, cetaceans use vision to some degree in communication. The visual stimuli to which they respond may be divided into two general groups: active and passive.

ACTIVE

We classify as active those visual stimuli that are produced under muscular control by one or more animals and that are in turn reacted to by one or more other animals in the social group.

Fig. 1. Juvenile male Atlantic bottlenosed dolphin (MLF 232). The dark parallel marks on the forehead (melon) were inflicted by another juvenile male when the two were first placed together about one and a half years before this picture was made. At this writing, some five years later, the scars remain. The light gray markings on the head and between the eye and flipper are typical pigmentation for this species, but the degree and intensity may vary between individuals and thus provide cues for visual individual recognition. The light line on the upper jaw at the tip of the arrow marks the row of hair pits mentioned in the text that possibly have a sensory function. Even from this angle, the binocular vision in the species is evident. (Photograph courtesy Marineland of Florida.)

Dolphins do not have the mechanics for facial expressions found in many other mammals like primates, dogs, and cats. Nor do they have available to them the variety of moveable body extremities, such as hands, ears, and tails, utilized by some mammal groups in communication. Visual signals are nonetheless evident among cetaceans.

These signals may take the form of openmouth threats or body postures, which probably are most useful in the maintenance of dominance hierarchies, although it has been suggested that some body postures are related to sexual signaling (see below). An Atlantic bottlenosed dolphin may indicate threat by facing another dolphin and opening its mouth, thereby exposing its teeth, or by arching its back slightly and holding its head downward. Submission for dolphins, as for many mammals, may take the form of a display that is the opposite of the threat; in this case the mouth is kept closed and the side of the body is turned to the threatening dolphin. It is usually the dominant animal, most often the largest and oldest of either sex, that displays such threats in a dolphin community. We should note, however, that established dominants can control a group using only the slightest gesture, to which other members seem to be very sensitive.

Sexual solicitation, which may be displayed by an animal of any age, including infants, often involves the soliciting dolphin swimming ahead of the other animal, looking back, and rolling onto its side or back to display the genital region. Tavolga and Essapian (1957:12-13, Fig. 1) have suggested that a soliciting male may also assume a particular S-shaped posture. We have not observed this to any great extent, nor did Puente and Dewsbury (1976) find a correlation between the posture and copulation. Vision may also aid in special positioning between individuals in a group—again, behavior related to social hierarchies. And finally, although it is not true vision, some dolphins (and probably most, if not all, cetaceans) certainly "see" by means of their ability to echolocate, and this too may be used in spatial positioning, at least in murky waters. It is even possible that the configuration of the body, such as an open or a closed mouth, might be "seen" in this manner.

Speed and direction of motion are almost certainly perceived visually. Even a low-intensity movement of the flukes by a lead animal serves as a perfectly adequate intention movement for maintaining synchronous swimming (Caldwell and Caldwell, 1964).

PASSIVE

We classify as passive those visual stimuli that are not under the muscular control of the animal producing them, although they may well be received by another animal to the same degree that active signals are.

As noted by Norris (1967), sexual dimorphism in relative size occurs in a few cetacean species, and the development of the teeth varies conspicuously in some (with the teeth of the male much more developed and prominent). Although they are probably not primarily visual stimuli, these dimorphisms are surely visually perceived by other cetaceans. For most cetacean species, however, the only visible difference in sexes is in the genital region. When a new animal is introduced to a group, this area receives detailed examination by the others (see illustration in D. K. Caldwell and M. C. Caldwell, 1972b:50). The recipient of the scrutiny passively accepts examination even though it is surrounded by investigating animals. Although other sensory systems may be activated during this procedure (echolocation, tactile, and possibly even gustatory), visual information probably provides additional if not primary sensory input.

The young of some cetaceans are marked in a manner quite different from the adults (compare, for example, young and adult pigmentation in the Atlantic spotted dolphin as illustrated by Caldwell and Caldwell, 1966; the adult is heavily spotted while the young is not spotted at all). Such differences certainly help visually distinguish a young animal from an adult, which in turn induces a different behavioral response. A juvenile spotted dolphin, for example, may be just getting its spotted pigmentation even though it is nearly as large as some well-spotted adults. Norris (1967) noted that sexual and pattern dimorphisms occur in odontocetes. In our own experience there appears to be a tendency for Atlantic bottlenosed dolphin males to develop a very pink belly during certain times of the year, and this change in coloration may be related to breeding activity (Caldwell, 1960). If this is true and if Atlantic bottlenosed dolphins have the capability for color vision (see Perez, Dawson, and Landau, 1972), it is likely that other members of the school would perceive that the animal is in breeding condition rather than in the ever-ready sexual behavior pattern typical of most male dolphins.

Evans and Bastian (1969) also pointed out that many delphinids have areas of white or very light pigmentation which are vivid and visible at great distances under water, and that these visible displays may be important to the social interaction of these dolphins. Although it has never been tested, we concur with this hypothesis in general. There must also be some evolutionary basis for some of the bold markings on cetaceans like the killer whale and the Dall porpoise, and it seems quite likely that their markings are related to visual communication. Conspecific identification in these animals seems likely; individual recognition is potentially possible when, for example, animals have the individual markings noted in the Dall porpoise by Norris and Prescott (1961) and in the Atlantic bottlenosed dolphin by D. K. Caldwell and M. C. Caldwell (1972b). There are many other reports of unusually marked individuals of both odon-tocete and mysticete cetaceans, and all of them must provide visual identification (and communication) signals to other members of the herd.

Atlantic bottlenosed dolphins and sperm whales, for example, are known to have complex social organizations, and recent field studies have shown that many baleen whales do too. We can therefore presume from even limited studies that cetaceans also have complex social signals, many of which appear to be visual. Visual signals are much more difficult to study and analyze than acoustic, and scientists may never recognize more than a few of the more obvious and pronounced ones. From our own human experience we know that many of our visual signals are obscure and ambiguous. One individual must know another intimately before subtle nuances are recognized, and similar signals from another person may not mean the same thing at all, especially when the cultures of the two humans are different. We believe, therefore, that there must be many more visual signals in cetacean communication than we recognize. We suggest that while vision is probably of less importance than the use of sound, it is perhaps of more importance than recent literature suggests.

Tactile Communication

Behavioral observations quickly lead one to the conclusion that touch is one of the most important means by which cetaceans communicate. First observed in captive dolphins of many species, touch has also been demonstrated in some of the larger baleen whales, as shown by underwater studies in recent years.

All cetaceans that we have observed in captivity seek and are receptive to gentle body contact (for photographs see Evans and Bastian, 1969, and D. K. Caldwell and M. C. Caldwell, 1972a, 1972b). Students of wild behavior have observed and illustrated body contact in both odontocetes (see, for example, Evans and Bastian, 1969) and mysticetes (see, for example, Cousteau and Diole, 1972, for illustrations of this behavior in the humpback whale; and Saayman and Tayler, 1973, for the southern right whale).

Some of the body contact is obviously related to sexual activity. Copulation is most frequently preceded by gentle mouthing and nipping. Particularly in mature animals the precopulatory play may become progressively more violent, with the participants even leaping from the water and diving forcefully at each other, sometimes only glancing bodies but at other times ramming melons so hard that it causes loud reverberations in a tank (D. K. Caldwell and M. C. Caldwell, 1972b:52). One would think that such violence would fracture their skulls, but instead it seems to function as a sexual stimulus because it almost always ends in copulation.

Most tactile stimulation is of a more delicate nature, involving as little as simply touching flippers as the animals slowly swim together for long periods. At the experimental level Pepper and Beach (1972) found that mild tactile stimulation from humans served as a reward to a captive Atlantic bottlenosed dolphin, a finding suggesting that such stimulation by another dolphin might also be rewarding.

On the obverse side of the coin, tactile communication is not always used to promote closer interpersonal relationships. Dolphins follow the normal mammalian pattern in displaying aggression (see photographs of aggressive behavior by the Atlantic bottlenosed dolphin in D. K. Caldwell and M. C. Caldwell, 1967, 1972a, 1972b; M. C. Caldwell and D. K. Caldwell, 1967; Evans and Bastian, 1969). Fights may occur over objects, space, proximity to other individuals, food, or for no apparent reason at all. Fighting dolphins slash and bite with the teeth, slash and ram with the jaws, and strike with the flukes. The encounters sometimes result in injuries with permanent scarring (Fig. 1). Mothers may punish their young by holding them down (Fig. 2), biting them, or even holding them out of water.

Although almost any animal may be provoked into an attack, large males are the most aggressive. Immature males are the most frequent recipients of attack. Norris (1967) pointed out the abundance of scars in the urogenital area of immature males, an indication of the age-old struggle by maturing groups for sexual rights.

Fig. 2. Adult female Atlantic bottlenosed dolphin punishing her infant male calf by holding him down on the bottom. The calf usually vocalized loudly when such punishment took place. (Photograph by David K. Caldwell at Marineland of the Pacific.)

Sexual aggression by large males has been discussed as a husbandry problem (Caldwell, Caldwell, and Townsend, 1968). Homosexuality in dolphins has also been seen and discussed at some length by several writers. We have a single example of homosexuality used solely in the context of communicating and establishing dominance. The incident occurred between two large male Atlantic bottlenosed dolphins when they were first put together. Although the encounter began with the usual open-mouth threats, it ended with sexual pursuit and two successful intromissions by the victor, and a somewhat reluctant submission by the loser. The entire episode lasted one hour, terminated, and was not seen again. The loser, who had been the dominant animal in the community, never again established himself, nor was he seen to try. This behavioral pattern of rape as an expression of aggressive dominance behavior is one of the unfortunate consequences of human penal institutions; whether it occurs in wild dolphins, as it does in other human contexts, we cannot say. In this instance, however, it was a clear example of tactile aggression.

On the anatomical level, the skin of several odontocetes had been examined histologically. Most recently studied by Simpson and Gardner (1972), dolphin skin appears richly innervated, particularly in the regions of the jaw, flukes, vulva, and perineum, which suggests a greater sensitivity in these areas.

A few rudimentary or vestigial hairs are present in cetaceans. Located on the upper jaw, they are variously termed bristles in the mysticetes; bristles, hairs, or even vibrissae in the Amazon River dolphin; or pits (Fig. 1) in the Atlantic bottlenosed dolphin (which bear hairs in the newborn). The terminology depends on the level of evolutionary regression of the species, but they most likely retain a tactile function in all species (see Simpson and Gardner, 1972, and D. K. Caldwell and M. C. Caldwell, 1972a, for illustrations and interpretations).

Chemical Communication

In an earlier summary (D. K. Caldwell and M. C. Caldwell, 1972a) we included what was then known about the olfactory and gustatory senses in cetaceans. Behaviorally, it appears that dolphins may be able to taste, although there is no evidence from a behavioral or anatomical point of view that they can smell. There is some anatomical evidence that mysticetes may be able to do the latter, although it has not been demonstrated behaviorally. While there is behavioral evidence for taste in dolphins (they will eat one kind of fish and not another, for instance, even if the fish appears to be of the same texture and is cut into small pieces), the anatomical evidence is confusing (Jansen and Jansen, 1969; Morgane and Jacobs, 1972). In Atlantic bottlenosed dolphins we were unable to demonstrate taste receptors (D. K. Caldwell and M. C. Caldwell, 1972a:479), although Suchowskaja (1972) reported their presence in Tursiops truncatus and Delphinus delphis. Kruger (1959:177) suggested that the excellent development of the nucleus ventralis medialis in the Atlantic bottlenosed dolphin might indicate the presence of the ability to taste.

The work of Sokolov and Kuznetsov (1971) indicates behavioral conditioning by Black Sea dolphins to chemical stimuli. The levels of success achieved by the dolphin on the initial tests (74 percent and 78 percent) suggest that they can discriminate. The behavioral breakdown that followed leaves room for doubt. As the writers pointed out, the test design itself was one that animals have difficulty solving. Our own experimental dolphin has shown difficulty solving similarly designed problems.

Further experiments in this area would be desirable. It would seem that cetaceans would be sorely disadvantaged by the loss of all chemoreception. The loss of smell was probably a necessary concomitant to the movement of the blowhole to the top of the head during the transition from land to water. On the other hand, we can think of no comparable anatomical reason for evolutionary pressure to have forced these animals to discard something so basic as the sense of taste.

If indeed cetaceans do have chemoreceptors, they are capable of chemical signaling by waste products and glandular secretions. Male Atlantic bottlenosed dolphins (and other dolphin species as well, we have observed) do have two small openings located just anterior to the anus (Fig. 3). These openings lead to glandular tissue via a large duct. The base of the gland extends into small tubules that have a single row of secretory epithelium. Although it is possible that the structures are nothing more than undeveloped mammary glands, biochemical or behavioral work on their secretions might prove useful.

Fig. 3. Posterior ventral surface of an adult male Atlantic bottlenosed dolphin (MLF 214) showing paired anal pores (P). The pores are posterior to the male genital slit (GS) and just anterior to the anus (A). (Photograph by David K. Caldwell.)

Sensory Coordination

As behaviorists, we rely heavily on simple sensory clues that appear to elicit a particular kind of behavior. On the other hand, as cetologists, we have learned that cetaceans simply do not react to seemingly all-or-none signals as birds and insects, for example, do. A particular stimulus, whether it is color, shape, or sound, does not automatically elicit the initiation of a behavior such as courtship or feeding. As we have said before (D. K. Caldwell and M. C. Caldwell 1972a:485), in cetaceans "a single cue may increase the probability of a behavior's occurring, but somewhere in the brain all simultaneously incoming stimuli are being processed and balanced one against the other. It is the summation of all internal and incoming stimuli, plus individual experience, that determines the final behavior. . . We have not changed in our thinking on this point. After our considerable studies of dolphins in recent years, we find that one dolphin may react to another in a variety of ways although the same communicatory cues were presented each time. But perhaps we cannot perceive the subtle differences in cues that the attending dolphin can.

References

Amundin, M., 1974. Some evidence for a displacement behaviour in the harbour porpoise, Phocoena phocoena (L.); a casual analysis of a sudden underwater expiration through the blow hole. Rev. Comp. Animal 8(l):39-45.

Bullock, T. H.; Grinnell, A. D.; Ikezono, E.; Kameda, K.; Katsuki, Y.; Nomoto, M.; Sato, O.; Suga, N.; and Yanagisawa, K.; 1968. Electrophysiological studies of central auditory mechanisms in cetaceans. Zeitschrift für vergleichende Physiologie, 59:117-56.

Bullock, T. H., and Ridgway, S. H., 1972. Evoked potentials in the central auditory system of alert porpoises to their own and artificial sounds, f. Neurobiology, 3(l):79-99.

Caldwell, D. K., 1960. Notes on the spotted dolphin in the Gulf of Mexico, f. Mammal., 41 (1): 134-36.

Caldwell, D. K., and Caldwell, M. C., 1966. Observations on the distribution, coloration, behavior and audible sound production of the spotted dolphin, Stenella plagiodon (Cope). Los Angeles County Mus., Cont. in Sei., 104:1-28.

Caldwell, D. K., and Caldwell, M. C., 1967. Dolphins, porpoises and behavior. Underwater Nat., 4(2): 14-19.

Caldwell, D. K., and Caldwell, M. C., 1972a. Senses and communication. In: Mammals of the Sea; Biology and Medicine, S. H. Ridgway, ed. Springfield, 111.: Charles C. Thomas, pp.466-502.

Caldwell, D. K., and Caldwell, M. C., 1972b. The World of the Bottlenosed Dolphin. Philadelphia: J. B. Lippincott.

Caldwell, D. K., and Caldwell, M. C., 1972c. Dolphins communicate-but they don't talk. Naval Res. Rev., 25(6-7):23-27, cover illus.

Caldwell, M. C., and Caldwell, D. K., 1964. Experimental studies on factors involved in care-giving behavior in three species of the cetacean family Delphinidae. Bull. So. Calif. Acad. Sei., 63(1): 1-20.

Caldwell, M. C., and Caldwell, D. K., 1965. Individualized whistle contours in bottlenosed dolphins (Tursiops truncatus). Nature, 207:434-35.

Caldwell, M. C., and Caldwell, D. K., 1967. Intraspecific transfer of information via the pulsed sound in captive odontocete cetaceans. In: Les Systemes Sonars Animaux, Biologie et Bionique, R.-G. Busnel, ed. Jouy-en-Josas, France: Laboratoire de Physiologie Acoustique, 2:879-936.

Caldwell, M. C., and Caldwell, D. K., 1968. Vocalization of naive captive dolphins in small groups. Science, 159:1121-23.

Caldwell, M. C., and Caldwell, D. K., 1971. Statistical evidence for individual signature whistles in Pacific whitesided dolphins, Lagenorhynchus obliquidens. Cetology, 3:1-9.

Caldwell, M. C., and Caldwell, D. K., 1972. Vocal mimicry in the whistle mode by an Atlantic bottlenosed dolphin. Cetology, 9:1-8.

Caldwell, M. C., and Caldwell, D. K., in press. The whistle of the Atlantic bottlenosed dolphin (Tursiops truncatus). 2. Ontogeny. In: Natural History of Whales, H. E. Winn and B. L. Olla, eds. New York: Plenum Publishing.

Caldwell, M. G, and Caldwell, D. K., unpubl. The whistle of the Atlantic bottlenosed dolphin (Tursiops truncatus). 1. General description for the species and statistical analysis of stereotypy in individuals.

Caldwell, M. G; Caldwell, D. K.; and Miller, J. F.; 1973. Statistical evidence for individual signature whistles in the spotted dolphin, Stenella plagiodon. Cetology, 16:1-21.

Caldwell, M. C.; Caldwell, D. K.; and Townsend, B. G, Jr.; 1968. Social behavior as a husbandry factor in captive odontocete cetaceans. In: Proceedings of the Second Symposium on Diseases and Husbandry of Aquatic Mammals. St. Augustine, Fla.: Marineland Res. Lab., pp. 1-9.

Caldwell, M. C.; Caldwell, D. K.; and Turner, R. H.; 1970. Statistical analysis of the signature whistle of an Atlantic bottlenosed dolphin with correlations between vocal changes and level of arousal. Los Angeles County Mus. Nat. Hist. Found., Tech. Rept. 8, LACMNHF/MRL ONR contract N00014-67-C-0358, 40 pp. (processed).

Caldwell, M. C.; Hall, N. R.; Caldwell, D. K.; and Hall, H. I.; 1971. A preliminary investigation of the ability of an Atlantic bottlenosed dolphin to localize underwater sound sources. Marineland Research Laboratory (Florida), Tech. Rept. 6, ONR contract N00014-70-C-0178. 6pp. (processed).

Cousteau, J.-Y., and Diole, P., 1972. The Whale: Mighty Monarch of the Sea. Garden City, N.Y.: Doubleday.

Cummings, W. C., 1971. Sonar training tape: non target sounds. Naval Acoustic Sensor Training Aids Dept., Fleet ASW School, San Diego, and Applied Bioacoustics Branch, Naval Undersea Center, San Diego, Calif. Biological Tape no. 12-3522. 5TR-P015.

Dawson, W. W.; Birndorf, L. A.; and Perez, J. M.; 1972. Gross anatomy and optics of the dolphin eye (Tursiops truncatus). Cetology, 10:1-12.

Dawson, W. W., and Perez, J. M., 1973. Unusual retinal cells in the dolphin eye. Science, 181:747-49.

Dral, A. D. G., 1972. Aquatic and aerial vision in the bottle-nosed dolphin. Netherlands f. Sea. Res., 5(4): 510-13.

Dral, A. D. G., 1974. Problems in image-focusing and astigmatism in Cetacea-a state of affairs. Aquatic Mammals, 2(l):22-28.

Evans, W. E., 1967. Vocalization among marine mammals. In: Marine Bio-acoustics, W. N. Tavolga, ed. New York: Pergamon Press, vol. 2, pp. 159-86.

Evans, W. E., 1973. Echolocation by marine delphinids and one species of fresh-water dolphin. J. Acoustical Soc. A mer., 54 (1) : 191 -99.

Evans, W. E., and Bastian, J., 1969. Marine mammal communication: social and ecological factors. In: The Biology of Marine Mammals, H. T. Andersen, ed. New York: Academic Press, pp.425-75.

Hall, N. R.; Hall, H. I.; Caldwell, M. C.; and Caldwell, D. K.; 1972. Visual acuity under conditions of near darkness and discrimination learning in the Atlantic bottlenosed dolphin (Tursiops truncatus). Cetology, 12:1-7.

Herman, L. M., and Arbeit, W. R., 1972. Frequency difference limens in the bottlenose dolphin: 1-70 KC/S./ Auditory Res., 2:109-20.

Herman, L. M.; Peacock, M. F.; Yunker, M. P.; and Madsen, C. J.; 1975. Bottlenosed dolphin: doubleslit pupil yields equivalent aerial and underwater diurnal acuity. Science, 189:650-52.

Hollien, H.; Hollien, P.; Caldwell, D. K.; and Caldwell, M. C.; 1976. Sound production by the Atlantic bottlenosed dolphin Tursiops truncatus. Cetology, 26:1-8.

Jansen, J., and Jansen, J. K. S., 1969. The nervous system of Cetacea. In: The Biology of Marine Mammals, H. T. Andersen, ed. New York: Academic Press, pp. 175-252.

Johnson, C. S., 1967. Sound detection thresholds in marine mammals. In: Marine Bio-acoustics, W. N. Tavolga, ed. New York: Pergamon Press, vol. 2, pp.247-60.

Kinne, O., 1975. Orientation in space: animals: mammals. In: Marine Ecology, O. Kinne, ed. New York: Wiley, vol. 2, Physiological Mechanisms, part 2, pp.702-852.

Kruger, L., 1959. The thalamus of the dolphin (Tursiops truncatus) and comparison with other mammals. f. Comp. Neurology, 111 (1): 133-94.

Litchfield, C., and Greenberg, A. J., 1974. Comparative lipid patterns in the melon fats of dolphins, porpoises and toothed whales. Comp. Biochemistry and Physiology, 47B:401-7.

Litchfield, C.; Greenberg, A. J.; Caldwell, D. K.; Caldwell, M. C.; Sipos, J. C.; and Ackman, R. G.; 1975. Comparative lipid patterns in acoustical and nonacoustical fatty tissues of dolphins, porpoises and toothed whales. Comp. Biochemistry and Physiology, 50B:591-97.

Litchfield, C.; Karol, R.; and Greenberg, A. J.; 1973. Compositional topography of melon lipids in the Atlantic bottlenosed dolphin Tursiops truncatus: implications for echo-location. Marine Biology, 23:165-69.

McCormick, J. G.; Weaver, E. G.; Palin, J.; and Ridgway, S. H.; 1970. Sound conduction in the dolphin ear. f. Acoustical Soc. Amer., 48(6-2): 1418-28.

Morgane, P. J., and Jacobs, M. S., 1972. Comparative anatomy of the cetacean nervous system. In: Functional Anatomy of Marine Mammals, R.J. Harrison, ed. London: Academic Press, pp.117-244..

Norris, K. S., 1967. Aggressive behavior in Cetacea. In: Aggression and Defense: Neural Mechanisms and Social Patterns, C. D. Clemente and D. B. Lindsley, eds. Berkeley: University of California Press, UCLA Forum in Medical Sciences no. 7, pp.225-41.

Norris, K. S., 1968. The evolution of acoustic mechanisms in odontocete cetaceans. In: Evolution and Environment, E. T. Drake, ed. New Haven, Conn.: Yale University Press, pp.297-324.

Norris, K. S., 1969. The echolocation of marine mammals. In: The Biology of Marine Mammals, H. T. Andersen, ed. New York: Academic Press, pp.391-423.

Norris, K. S., 1974. The Porpoise Watcher. New York: W. W. Norton.

Norris, K. S.; Dormer, K.J.; Pegg,J.; and Liese, G.J.; 1972. The mechanism of sound production and air recycling in porpoises: a preliminary report. In: Proceedings of the Eighth Annual Conference on Biological Sonar and Diving Mammals, 1971. Fremont, Calif.: Biological Sonar Laboratory, pp.113-29.

Norris, K. S., and Harvey, G. W., 1974. Sound transmission in the porpoise head./ Acoustical Soc. Amer., 56(2):659-64.

Norris, K. S., and Prescott, J. H., 1961. Observations on Pacific cetaceans of Californian and Mexican waters. Univ. Calif. Publ. in Zool., 63(4):291-402, plates 27-41.

Payne, R. S., and McVay, S., 1971. Songs of humpback whales. Science, 173:585-97.

Pepper, R. L., and Beach, F. A., Ill, 1972. Preliminary investigations of tactile reinforcement in the dolphin. Cetology, 7:1-8.

Perez, J. M.; Dawson, W. W.; and Landau, D.; 1972. Retinal anatomy of the bottlenosed dolphin (Tursiops truncatus). Cetology, 11:1-11.

Poulter, T. C., 1968. Marine mammals. In: Animal Communication, T. A. Sebeok, ed. Bloomington: Indiana University Press, pp.405-65.

Puente, A. E., and Dewsbury, D. A., 1976. Courtship and copulatory behavior of bottlenosed dolphins (Tursiops truncatus). Cetology, 21:1-9.

Renaud, D. L., and Popper, A. N., 1975. Sound localization by the bottlenose porpoise, Tursiops truncatus. J. Exper. Biol., 63:569-85.

Saayman, G. S.; Bower, D.; and Tayler, C. K.; 1972. Observations on inshore and pelagic dolphins on the south-eastern cape coast of South Africa. Koedoe, 15:1-24.

Saayman, G. S., and Tayler, C. K., 1973. Some behaviour patterns of the southern right whale Eubalaena australis. Zeitschrift für Saugetier künde, 38(3): 172-83.

Saayman, G. S.; Tayler, C. K.; and Bower, D.; 1973. Diurnal activity cycles in captive and free-ranging Indian Ocean bottlenose dolphins (Tursiops aduncus Ehrenburg). Behaviour, 44(3-4):212-33, pi. 6.

Schevill, W. E., 1964. Underwater sounds of cetaceans. In: Marine Bio-acoustics, W. N. Tavolga, ed. New York: Pergamon Press, vol. 1, pp.307-16.

Schusterman, R. J., 1973. A note comparing the visual acuity of dolphins with that of sea lions. Cetology, 15:1-2.

Simpson, J. G., and Gardner, M. B., 1972. Comparative microscopic anatomy of selected marine mammals. In: Mammals of the Sea; Biology and Medicine, S. H. Ridgway, ed. Springfield, 111.: Charles C. Thomas, pp.298-418.

Sokolov, V. E., and Kuznetsov, V. B., 1971. Chemoreception in the Black Sea dolphin (Tursiops truncatus Mont.). Doklady Akad. Nauk SSR, ser. biol., 201:998-1000.

Spong, P.; Bradford, J.; and White, D.; 1971. Field studies of the behaviour of the killer whale (Orcinus orca ). In: Proceedings of the Seventh Annual Conference on Biological Sonar and Diving Mammals, 1970. Menlo Park, Calif.: Stanford Research Institute, pp. 169-74.

Suchowskaja, L. I., 1972. The morphology of the taste organs in dolphins. Investigations on Cetacea, 4:201-4, 2 plates.

Tavolga, M. C., and Essapian, F. S., 1957. The behavior of the bottle-nosed dolphin (Tursiops truncatus): mating, pregnancy, parturition and mother-infant behavior. Zoologica, 42(1):11-31, plates 1-3.

Tavolga, W. N., 1965. Review of marine bio-acoustics; state of the art: 1964. Port Washington, N.Y.: U.S. Naval Training Device Center, Tech. Rept.: NAVTRADEVCEN 1212-1: i-v, 1-100.

Tavolga, W. N., 1968. Marine animal data atlas. Orlando, Fla.: U.S. Naval Training Device Center, Tech. Rept.: NAVTRADEVCEN 1212-2: i-x, 1-239.

Tayler, C. K., and Saayman, G. S., 1972. The social organisation and behaviour of dolphins (Tursiops aduncus) and baboons (Papio ursinus): some comparisons and assessments. Ann. Cape Provincial MILS. (Nat. Hist.), 9(2): 11-49.

Winn, H. E., and Perkins, P. J., 1976. Distribution and sounds of the minke whale, with a review of mysticete sounds. Cetology, 19:1-12.

Winn, H. E.; Perkins, P. J.; and Poulter, T. C.; 1971. Sounds of the humpback whale. In: Proceedings of the Seventh Annual Conference on Biological Sonar and Diving Mammals, 1970. Menlo Park, Calif.: Stanford Research Institute, pp.39-52.

Wood, F. G., 1961. (No title.) In: Man and Dolphin, J. C. Lilly, Garden City, N.Y.: Doubleday, p.236.

Wood, F. G., 1964. (No title, in General Discussion section.) In: Marine Bio-acoustics, W. N. Tavolga, ed. New York: Pergamon Press, vol. 1, p.395.

Wood, F. G., 1973. Marine Mammals and Man. Washington, D. C.: Robert B. Luce.