So Long and Thanks for All (Your) Fish

Adela

I. BACKGROUND/ THE QUESTION

Dolphins (Delphinidae) are part of a group known as the Odontoceti, which forms a suborder of the cetaceans. As their name suggests, Odontocetes are characterised by having teeth (as opposed to baleen whales in the other suborder of cetaceans, Mysticeti). Toothed whales are active hunters, feeding mainly on fish and squid.

Odontocetes vary considerably in size and morphology. Delphinidae alone vary in size from 1.2 metres and 40 kilograms (Heaviside's Dolphin, Cephalorhynchus heavisidii) to 7 metres and 4.5 tonnes (Orca, Orcinus orca). Teeth may be numerous, with some dolphins bearing over 100 teeth in their jaws; while at the other extreme are the Narwhal (Monodon monoceros) with its single long tusk, and the almost toothless beaked whales (family Ziphiidae).

Odontocetes have a single blowhole on the top of the head, which is used to produce vocalizations. They maintain a broad variety of calls to communicate, but are also capable of using ultrasound for echolocation. As an adaptation for their echolocation, odontocete skulls have become asymmetric, and their brains have increased in size.

Dolphins generally live in groups of up to a dozen animals. These groups, called pods or schools, occasionally merge to form "superpods", aggregations of up to thousands of whales. The complex social relations and high potential for learning of dolphins has led to their being considered being among the most intelligent of animals.

However, dolphins have often been observed to display altruistic behaviour towards schoolmates and even members of other species of odontocetes. Is there sufficient evidence to claim reciprocal altruism (or is it just some form of nepotism at work)? Does such behaviour fit in with selection theory, i.e. is it an evolutionary adaptation? If yes, how so? What are the implications of these findings?

II.YOU SCRATCH MY BACK… / EVIDENCE OF ALTRUISTIC BEHAVIOUR

A. Cooperative feeding

The first and most obvious form of altruistic behaviour would be cooperative feeding. Certain species of dolphins have been observed to swim in certain patterns to herd their prey. For example, the dusky dolphins of Argentina form tight circles around anchovies, paralysing them (possibly by inducing a hypoxic state due to crowding, or by debilitation via intense ensonification) before allowing a member of the pod to rush in for the kill (Booth 1988). This is immensely useful since dolphin prey, such as mahimahi, are typically much swifter in burst speed than dolphins.

In addition, there have been sightings of false killer whales and roughtooth dolphins feeding, in which one individual held a large prey in its jaws and allowed other dolphins to pull off piece of flesh. The prey was passed around from dolphin to dolphin, allowing every individual equal chances of feeding (Brower and Curtsinger 1979). Such food-sharing is evidently not typical behaviour to maximise survival in food-scarce situations.

B. Care-giving

Secondly, dolphins exhibit extended care for each other throughout their lives. Booth (1988) reports that in bottle-nosed dolphins, parental care is often carried out over a lengthy period, with offspring remaining highly dependent on the mother for up to four years (until the calf is weaned). Mothers often form ‘playpens’, in which all their offspring swim protected from predators. They have also been known to ‘babysit’ for each other when they leave the pod for short periods (though in many cases, these females are related and could merely be practising kin selection to improve inclusive fitness).

Dolphins also assist in birthing. Severe teeth marks concentrated on the posterior body and tail flukes of stillborn calves indicate attempts to pull them from the mother (Connor and Norris 1982).

There also exist various other forms of mutual assistance toward injured or distressed animals, under the broad heading of epimeletic behaviour:(a) Standing by occurs when an animal or animals stays with another animal in distress without offering any apparent aid. It involves remaining in dangerous situations for far longer than would be the case with no distressed animal. Odontocetes, from dolphins to sperm whales, will sometimes ‘stand by’ captive dolphins, circling and swimming near the captive vessel (Connor and Norris 1982). I would call this the odontocete equivalent of ‘moral support’.(b) Excitement includes approaching an injured comrade and displaying violent or excited behaviour, such as pressing against and biting restraining lines, propelling injured animals away from captors, or attacking when young animals of the pod are disturbed. For instance, Norris and Prescott (1961) report a case where a shot pilot whale was drifting in rigor mortis toward the capture vessel when two other pilot whales submerged it and took it away. (c) Support involves an animal pressing another to the surface to help it breathe. It is most common intraspecifically, but has also been observed among members of different species. A most dramatic example of this would be that recorded by Siebenaler and Caldwell (1956), in which two bottlenose dolphins buoyed a third which had been stunned by a freak explosion at sea. They had to leave the injured party’s side to breathe but kept returning until it had recovered. In addition, the remainder of the dolphins’ pod remained ‘standing by’ as described above.

C. Mating alliances

Connor and Norris (1982) noted that pairings of juvenile and adult bottlenose males existed within pods. Such pairings existed to protect each other from predation, and to cooperate in hunting and in sexual herding. Male dolphins were observed to operate in pairs and triplets to sequester and control the movements of females. They preferentially herded non-pregnant females likely to be in estrus. Each alliance associated preferentially with one or two other alliances. Occasionally, two alliances combined and took females from another alliance or defended females against such efforts (Connor et al. 1992). Bottlenose dolphins are one of only a few mammalian taxa where the males are known to cooperate within their social group in order to maintain mating access to single females against other males. Male bonds in bottlenose dolphins have been hypothesised as evolving through kinship and associated inclusive fitness effects (Krutzen et al. 2003).

III. THE REAL PLOT/ EXPLANATORY MODELS

Several models have been suggested to explain the unlikely behaviour of dolphins and other odontocetes described above. As Brader and Wursig (1994) put it, “Group size and structure are guided by an incompletely understood matrix of factors, including foraging type, need for predator detection and avoidance, social and sexual interactions, and the care and maintenance of developing young”. They propose that dolphins who coordinate activities in efficiently sized groups benefit by enhanced fitness to themselves and their offspring. Connor and Richards (2001), on the other hand, believe that reciprocal altruism is the agent at work within dolphin societies.

Trivers (1971) stated that the conditions necessary for the evolution of reciprocal altruism are identical to those for kin selection (except that the recipient of an altruistic act need not be related to the actor). We must thus consider the possibility that the entire range of altruistic behaviour is a product of kin selection and that reciprocity does not exist in dolphin societies. Kin selection, also known as kin-directed altruism or nepotism, can be discriminatory or non-discriminatory, depending on the cost of typical altruistic acts to the donor, the benefit to the receiver, and the relatedness of the two parties.

A. Non-discriminatory nepotism

Non-discriminatory nepotism is by far the most commonly offered explanation of dolphin altruism. This explanation attributes helpful behaviour amongst individuals to an “innate behavioral repertoire” of the species that manifests in stereotyped responses to the distress of companions. The innate tendency to rescue offspring and relatives contributes greatly to inclusive fitness and is likely to have been fixed in the species.

However, the model non-discriminatory nepotism has also met with much opposition. The counter-argument provided by Connor and Norris (1982) is that dolphin behaviour is governed much more by learning than by innate patterns. Dolphins are capable of deuterolearning (second order learning, i.e., understanding how different domains of communication operate) – for example, roughtooth dolphins (Steno bredanensis) in captivity quickly made the logical connection between novel jumps and food incentives (Pryor et al. 1969). Innate tendency is thus not a convincing argument for giving aid to others.

Furthermore, epimeletic behaviour is spontaneous and tailored to individual situations. In the rescue of the dead pilot whale (Part II. B. (b)), its body was not supported in a stereotypical fashion but rather taken away from would-be captors in a manner opposite to that of supportive behaviour.

Finally, the discriminatory nepotism model has been refuted by numerous reports of interspecific and even intergeneric assistance behaviour, even between species that are strikingly different morphologically, and thus could not be phenotypically misidentified as kin (precluding the possibility of kin selection).

B. Discriminatory nepotism

Under a system of discriminatory nepotism, an animal will give assistance according to how much it values the recipient’s fitness in relation to its own, given their relatedness in a certain situation. Discriminatory nepotism has already been clearly observed in infant pigtail macaques, with siblings never before seen (Silk 1992).

Smaller dolphins in general have extremely fluid social structures. In fact, perhaps the most significant finding from several studies on the social lives of dolphins is that they form “fusion-fission” societies, in which individuals may join temporary parties of varying sizes, instead of operating in a fixed group (Booth, 1988). If dolphin relations were regulated by discriminatory nepotism alone, we would expect low group fluidity, since a dolphin would consistently prefer to assist more closely related animals than more distantly related animals. Merely swimming with a group of more closely related individuals would represent an increase in inclusive fitness (Connor and Norris 1982). Thus, the model of discriminatory nepotism is not entirely convincing as an explanation for mutual assistance in dolphins either.

C. Reciprocal altruism

Reciprocal altruism is a phenomenon in which one organism provides benefit to another in the expectation of future reciprocation. It usually evolves in the presence of a mechanism to identify and punish "cheaters". However, under this ‘two-party’ model of reciprocal altruism, while there is ample time for dolphins to discover the cheating tendencies of others, consistent ousting of ‘cheaters’ has not been observed; furthermore, it is difficult to account for all cases of interspecific and intergeneric aid, since it requires that aid be given in anticipation of reciprocity specifically from the assisted individual.

The best explanation of epimeletic behaviour in dolphins appears to be a multiparty model of reciprocal altruism, in which altruistic acts are dispensed freely and not only to animals that are expected to reciprocate. Since reciprocal altruism is mediated by learning (i.e., functions independently of genetic relatedness), it explains interspecific epimeletic behaviour, as well as its unpredictable occurrence in dolphins (Connor and Norris 1982).

This model is also consistent with the group fluidity found within some species of dolphins. Group fluidity is based upon the ability to recognise individuals. The more individuals an animal can recognise outside one level of kin association, the higher the potential for altruistic relationships, since animals can then efficiently aggregate appropriate numbers for different activities such as travel, feeding, and socialising. As we already know, dolphins employ a system of calls (blowhole sounds) and imitation to communicate. Playback experiments (Sayigh et al. 1998) have proved dolphins’ capacity for individual recognition (via distinctive signature whistles) and motivation to respond, providing evidence for reciprocal altruism.

Recent studies (e.g., Möller 2001) have found that associations and alliance membership in male dolphins are not associated with either maternal kinship or genetic relatedness; the majority of male pairs within alliances are randomly related, although high relatedness values exist between males of different alliances in the resident population. These findings indicate that mechanisms other than kin selection may be foremost in the development and maintenance of cooperation between male bottlenose dolphins.

Furthermore, herding can explain why males travel in pairs or trios, but if they are competing for females, relationships between the males should be extremely hostile. Connor (2001) points out that on the contrary, alliance members associate with one another consistently even when not herding females, and a great deal of play and stroking occurs in these alliances, indicating that these alliances were not purely utilitarian.

IV. TAKING OVER THE WORLD/ QUESTIONS & IMPLICATIONS

Despite the strong evidence and arguments for reciprocal altruism, there remain several ambiguities and questions to be answered regarding the nature and findings of the research.

Firstly, it is notoriously difficult to observe cetacean activity. Since most observations were carried out on boats, researchers were able to report only events that happened near the surface, and even those were clear only when the water was not choppy (Wursig 1989). Research on dolphin vocalizations is also hindered due to the difficulties of localising sound underwater (Connor 2001).

Secondly, it is possible that small, possibly stable relations based on kinship could be undiscovered due to the small number of animals marked in studies (Connor and Norris 1982). The degree of relatedness between two individuals is also affected by migration, making it difficult to weigh factors against each other and determine if kin selection is operating or not (Hamilton, 1975). Of course, some argue that the scattered and often anecdotal literature on dolphins still adds up to considerable evidence for reciprocal altruism.

One area of curiosity is that seals and sea lions (pinnipeds), which like the odontocetes are carnivorous marine animals, exhibit much less flexible behaviour. Connor and Norris (1982) postulate that the birth and early nurture of dolphins at sea (as opposed to land births in pinnipeds) exposes them to higher predation and accelerates evolution of strategies for sheltering young.

There is also difficulty explaining the phenomenon of mass stranding in larger, polygynous species such as the false killer whale. In a case reported by Caldwell (1970), only one whale out of fifteen was seriously injured, and the others were sufficiently submerged that they could swim away, but they did not. Such behaviour does not confer higher fitness onto any party, nor does it fit any of the models examined so far. The explanation commonly given by proponents of the reciprocal altruism model is that polygyny creates such strong social bonds that a “one-for-all, all-for-one” mentality develops. Other parties such as Morimitsu et al. (1986) proposed that a parasite was responsible for causing a form of cranial nerve damage akin to meningitis that forced them to (fatally) enter shallow waters to respire more easily. However this theory has been met with skepticism as it seemed highly unlikely that the parasite would cause beaching numbering in the hundreds at a time. Larger sample sizes of brain autopsies need to be taken before any definitive conclusions may be reached about air sinus parasitism and its degree of pathology on dolphins.

The recent discovery of multi-level alliances amongst male dolphins (Connor, 2003) serves to complicate the model of reciprocal altruism. Apparently, males in stable first-order or derived second-order alliances are often strongly related so they gain inclusive fitness from alliance membership; while members of ‘superalliances’ (groups of about 14 individuals with highly labile pairs and trios) were related purely by chance. Furthermore, the strength of the association of alliance partners within the super-alliance, as measured by an index of joint participation in consorting a female, was not correlated with genetic relatedness. Thus, within one population and one sex, it appears that there may be simultaneous operation of more than one mode of group formation.

In any case, if dolphins do employ reciprocal altruism, they will be following in the footsteps of other ‘higher mammals’ such as humans and chimpanzees. This agrees with Connor and Norris’ assertion that selection for features of sociality (as exemplified by altruistic behaviour) has major evolutionary linkage with reciprocal altruism. Heavy predation has, perhaps, produced strong selection pressure to practice mutual assistance with higher sophistication, leading to the development of complicated social systems, with ethics not necessarily based on true kinship, but subject to cultural evolution.


V. REFERENCES

Richard C. Connor; Kenneth S. Norris. Are Dolphins Reciprocal Altruists? The American Naturalist, Vol. 119, No. 3. (Mar., 1982), pp. 358-374.

William Booth. The Social Lives of Dolphins. Science, New Series, Vol. 240, No. 4857. (Jun. 3, 1988), pp. 1273-1274.

J. B. Siebenaler; David K. Caldwell. Cooperation among Adult Dolphins. Journal of Mammalogy, Vol. 37, No. 1. (Feb., 1956), pp. 126-128.

Richard C. Connor; Rachel A. Smolker; Andrew F. Richards. Two Levels of Alliance Formation Among Male Bottlenose Dolphins (Tursiops sp.). Proceedings of the National Academy of Sciences of the United States of America, Vol. 89, No. 3. (Feb. 1, 1992), pp. 987-990.

Connor, R.C. Social relationships in a big - brained aquatic mammal. Model Systems in Behavioral Ecology (ed. L.A. Dugatkin). Princeton University Press, Princeton, 2001.

Luciana M. Möller; Luciano B. Beheregaray; Robert G. Harcourt; Michael Krützen. Alliance membership and kinship in wild male bottlenose dolphins (Tursiops aduncus) of southeastern Australia. Proceedings of the Royal Society B: Biological Sciences, Volume 268, Number 1479 / September 22, 2001.

Michael Krützen; William B. Sherwin; Richard C. Connor; Lynne M. Barré; Tom Van de Casteele; Janet Mann; Robert Brooks. Contrasting relatedness patterns in bottlenose dolphins (Tursiops sp.) with different alliance strategies. Proceedings of the Royal Society B: Biological Sciences, Volume 270, Number 1514 / March 07, 2003, pp. 497 – 502.

Stefan Bräger; Bernd Würsig; Alejandro Acevedo; Thomas Henningsen. Association Patterns of Bottlenose Dolphins (Tursiops truncatus) in Galveston Bay, Texas. Journal of Mammalogy, Vol. 75, No. 2. (May, 1994), pp. 431-437.

Laela S. Sayigh; Peter L. Tyack; Randall S. Wells; Andrew R. Solow; Michael D. Scott; A. B. Irvine. Individual recognition in wild bottlenose dolphins: a field test using playback experiments. Animal Behaviour, 1998, 57, pp. 41–50.

David K. Caldwell; Melba C. Caldwell; Cecil M. Walker, Jr. Mass and Individual Strandings of False Killer Whales, Pseudorca crassidens, in Florida. Journal of Mammalogy, Vol. 51, No. 3. (Aug., 1970), pp. 634-636.

Tamotsu Morimitsu; Tomoyuki Nagai; Minoru Ide; Akira Ishii; Masashi Koono. Parasitogenic Octavus Neuropathy as a Cause of Mass Stranding of Odontoceti. The Journal of Parasitology, Vol. 72, No. 3. (Jun., 1986), pp. 469-472.

Joan B. Silk. The Patterning of Intervention among Male Bonnet Macaques: Reciprocity, Revenge, and Loyalty. Current Anthropology, Vol. 33, No. 3. (Jun., 1992), pp. 318-325.

http://www.jengajam.com/r/dolphins-so-long-thank