I. The Nature of Sociobiological Inquiry
Sociobiology is the study of social behavior and social evo-,< lution based on the theory of adaptation through selection. • As such it is primarily concerned with the adaptiveness of social behavior and the selective processes producing and maintaining adaptiveness. Understanding the selective processes involved includes studying the ecology, physiology, and behavior, as well as the population genetics of the species in question. Sociobiological investigation increasingly attempts to characterize the genetic breeding system, as well as the population structure, both of which importantly influence the effectiveness of evolutionary processes in molding the characteristics of species.
The sociobiological approach assumes that selection at the individual level is the predominant force producing adaptation. A proper understanding of social phenomena, therefore, needs an understanding of the benefits and costs that the individual derives from its interaction with the social environment. Explicitly, group selection is relegated to a secondary position, as in most circumstances, selection operates more strongly at the individual than at the group level because fertility, dispersal, and mortality events are more frequent and act much more forceful on individuals than on groups. Explanation at the level of the individual of social phenomena such as group formation, parental care, and mating systems form the core of sociobiological inquiry.
As the majority of sociobiological inquiry in the field of marine mammals has been done on whales and pinnipeds, these two groups form the focus of the following sections. Relevant information on sea otters (Enhydra lutris) and manatees (Trichechus spp.) is mentioned briefly in Section IV.
II. Group Formation
The most obvious phenomenon of social fife is group formation. Suitable feeding or breeding habitat may initially lead to an aggregation of individuals, thus setting the stage for selective processes moulding the evolution of elaborate social interactions. In contrast to aggregations, groups imply that individuals of a species come together to derive benefits from interactions that follow from diis proximity. Such a grouping may serve social foraging, predator avoidance, or defense against predators. Groups may also be established for mating and to share parental care. These kinds of advantages constitute the selective processes that promote group formation in a wide variety of animals. So-ciobiology tries to explain groupings from the advantages and disadvantages incurred by the individuals.
A. Whales
The open ocean habitat offers few options for hiding from predators. Consequently, predation by large sharks and killer whales (Orcimis orca), particularly on newborns, is one important factor selecting for group formation in whales and dolphins. Direct observational evidence for this hypothesis is scarce, but signs of scarring provide ample evidence of frequent encounters with predators. For example, about one-third of all humpback whale (Megaptera novaeangliae) calves carry tooth marks on their flukes when arriving in the foraging areas, presumably from encounters with killer whales or sharks during migration to the feeding grounds. The most spectacular groupings are found in open ocean species such as spotted dolphins (Stenella spp.), which benefit the most from the advantages of grouping as protection against predators, but these species may also benefit from group foraging, as may many others.
Several effects play a role in the protection offered an individual by a group. The “dilution effect” acts by reducing the chances of an individual to be attacked by a predator who has noticed the group (to 1/group size, if a predator takes only one individual out of the group). This effect thus dilutes the chances of an attack on a given individual dramatically. The “confusion effect,” many individuals rushing back and forth through the visual field of an attacking predator, makes it more complicated for the predator to concentrate on one prey to catch. Finally, animals in the middle of a group can in effect use other individuals around them as shields against predatory attack, as there will always be other individuals geometrically closer to the predator attacking from the edge of a group. This is die so-called “geometry of the selfish herd,” which allows an individual to use other group members as protection. These effects may also partly protect individuals in groups against parasites, such as the cookie-cutter shark, which takes little bites out of the side of its victim. The advantages of grouping using dilution, confusion, or cover against predation do not depend on individual recognition or social bonds among the animals joining such a group, but also function perfectly well in a totally anonymous group.
Eventually, the advantage of gaining protection in the group will be counteracted by increasing food competition among individuals in a large group, which will determine the upper size of groups. Because food competition depends on abundance, distribution, and behavior of food organisms, which are little known for whale prey, the importance of food competition is also largely unknown. However, signs of territorial behavior between killer whale groups provide at least some direct evidence for the role of food competition in limiting group size.
For more powerful animals such as killer and spenn whales (Physeter macrocephalus), grouping offers the additional option to defend smaller, more vulnerable individuals against predators by taking them into the middle of the group. This kind of group action has a completely different quality than the examples given earlier because it involves bonds among the individuals in the group, which are presumably well cemented by individual recognition. Individuals in these groups actively take some risk to defend others (mostly calves) against predators. This can be understood by the fact that the individuals involved in such cooperative behaviors are kin to each other or it may represent mutualistic cooperation. The evolution of such mutualism would be eased in kin groups through kin selection.
However, grouping can also be of advantage for a predator. Foraging groups have perhaps been best analyzed for killer whales. For the mammal-hunting so-called transient killer whales in British Columbia, groups of three individuals proved most efficient in terms of energy intake per unit time when hunting harbor seals (Phoca vitulina). Optimal efficiency of this group size results because a group of three cooperatively hunting whales seems to be better detecting and catching prey than single animals or duos. However, competition in such a small group is less than in larger group once it comes to sharing a harbor seal carcass. For more evasive prey such as Dall’s poipoise (Phocoenoides dalli), larger groups may be more successful because more animals are better at intercepting fleeing prey.
Dusky dolphins (Lagenorhijnchus obscurus) herd schools of fish cooperatively toward the surface and presumably thereby increase food intake. Such activities have also been observed in other dolphins that herded fish schools into bays or against fishing nets and thus may increase food intake for all individuals in the group.
Social systems in whales differ widely between mysticetes and odontocetes. Mysticetes often live solitarily and, except for the mother-calf association during the rearing period, show little evidence of long-term social bonds. This is most likely caused by the nature of their food resources and the fact that due to their body size they have largely escaped predation. They may, however, aggregate during feeding in particularly productive areas and during the mating season (see later).
In contrast, almost all odontocetes are quite social. The social groupings of sperm whales, killer whales, pilot whales (Glo-bicephala spp.), and bottlenose dolphins (Tursiops spp.) are best documented. In whales, groupings represent matrilines in which male offspring may (in killer and pilot whales) or may not (sperm whales) remain for life. In sperin whales, the grouping of kin serves protection of offspring. During foraging, these animals routinely dive to 400 m depth and stay at depth for 40-60 min. During this period, young would be unprotected, waiting at the surface. This can be and is avoided by adults in a group staggering their diving in such a way that one or more adults almost always attend the calves at the surface. Clearly, such behavior provides indirect fitness gains if the individuals in a group are relatives (such as mother, daughter, granddaughter). Kin selection effects may also explain the lack of dispersal of young in killer whales (in this case of both sexes). In resident, fish-eating groups, male offspring in a matrilineal group may offer protection to relatives as well as help in the defense of foraging areas. Females that stay with their mother may likewise gain fitness from cooperation with close relatives. The size of the group will be limited by food competition, and indeed large groups split up when they have grown too much into smaller units along matrilineal kinship fines. Limitation of group size by the potential for food competition is particularly evident, as the so-called transient, mammal-eating groups are much smaller (no more than three to five animals) than those of fish-eating residents, presumably because otherwise the disadvantage of food competition would offset the advantage of close cooperation with kin. Pilot whales show a similar social structure where male and female offspring stay with mothers. Advantages and disadvantages of this social structure are less understood for the pilot whale than for the other species mentioned previously.
Bottlenose dolphins and some other inshore odontocetes live in quite open fission-fusion societies in which females associate frequently with many different partners. These associations may be useful in foraging or vigilance and defense against predators, but this hypothesis seems less likely because females with calves tend to be less gregarious. Whether this is due to food competition is unclear. Females may also sometimes group to avoid harassement by males (see later). Male associations vary between study sites and seem to serve mating purposes.
B. Pinnipeds
As a group, pinnipeds are characterized by an amphibious lifestyle. They forage at sea but females need to return to a firm substrate, land or ice, for parturition. A strong selective force during this period of birth and subsequent pup rearing is to avoid predation on mother and pup. In land-breeding pinnipeds, mostly otariids but also elephant seals (Mirounga spp.), monk seals (Monachus spp.), and the largest populations of gray seals (Halichoerus gnjpus), predator avoidance has led to the choice of oceanic islands for breeding. In comparison to the ocean used for foraging, these islands are limited in area, automatically leading to high densities of females on land. Otariid females come into estrus shortly after parturition, which sets the stage for sexual selection on males trying to station themselves in these female aggregations to breed (see later).
Primarily, the concentration of females on land has all the characteristics of an aggregation, but females of most land-breeding species stay closer together than necessitated by available habitat. One reason for such active groupings may be to avoid harassment by ardent males trying to mate with females. In addition, subadult males of some species of otariid abduct and herd pups, molesting them to the extent that they may die. In this dangerous situation, grouped pups profit from the dilution effect, which might select for tighter grouping of females.
Thermoregulation may also further grouping: Sea lions and the South African and Australian fur seal (Arctocephaus pusil-lus) tend to clump together as soon as they leave the water, which may serve to keep the body shell warm at minimal metabolic cost to the individual. This is not necessary for a fur seal with its highly insulating, air-trapping fur, and the function of clumping in the South African and Australian fur seal is therefore doubtful.
In contrast, ice-breeding seals tend to be much wider dispersed, be it on pack ice or fast ice. In these species, different factors select for some gregariousness. Most important for many species is predation: in the Arctic by polar bears (Ursus maritimus) and in the Antarctic by leopard seals (Hydrurga leptonyx) and killer whales. Some kind of a grouping advantage is perhaps operating that might equally benefit animals resting between foraging excursions and animals breeding on the ice. Indeed, harp seals (Pagophilus groenlandicus) are well known for breeding—even if dispersed—in well-defined local concentrations. It is not certain whether these concentrations are true social groupings or could also be induced by the characteristics of the ice in combination with the food resources below it, i.e., are aggregations due to resource distribution. It may actually be a combination of both. Weddell seals (Leptonychotes wed-dellii) group around tide cracks in fast ice, which offer holes for entry into the water where the seals forage under the ice. Because suitable holes are limited, this leads to a concentration of animals around entry holes and importantly influences the mating system.
Principally, pinnipeds are solitary foragers. There is little evidence for social foraging except occasional observations of sea lions herding fish into bays and communally preying on the trapped schools of fish. Aggregations of foraging sea lions and fur seals occur quite frequently near large fish schools where pinnipeds, birds, and whales may gather in so-called feeding frenzies. True cooperativity in these aggregations has not been demonstrated. This does not exclude that loose groups of pinnipeds may be more likely to locate food and through signs of foraging activity attract other animals to the site, an advantage frequently exploited by group-foraging birds.
III. Parental Investment Strategies
There is no convincing evidence of paternal investment in the rearing of young in any whale or pinniped. Lack of paternal care is typical for the majority of mammals and reflects that females alone provide for offspring during pregnancy and lactation, which frees males of parental duties. Only in a few whale societies, e.g., the killer whale matrilineal groups, do males act as helpers in defense and perhaps feeding of young. Maternal strategies in whales and pinnipeds are characterized by massive energetic investment in young dirough the production of large precocial young and extremely lipid-rich milk. However, parental investment is measured not by energy expenditure, but rather by a reduction in future fitness, a cost incurred by the mother through a loss in subsequent fertility or an increase in mortality due to a expenditure that benefits the offspring by increasing its fitness. Evidence of parental investment, so defined, is rare in whales and pinnipeds.
A. Whales
The patterning of parental effort differs widely between mysticetes whales and odontocetes. Mysticete females gather the material and energy for pregnancy (lasting about 1 year) and lactation during a feeding season of about half a year when food is plentiful and starve for the other half-year. Extraordinary fast fetal growth rates (in blue whales, Balaenoptera mus-culus, 27 mm/day) permit even the largest mysticete whale— with the exception of the bowhead (Balaena mysticetus), sei (Balaenoptera borealis), and gray (Eschrichtius robustus) whale—to produce a calf within 1 year. By migrating to warmer oceans in the nonfeeding period, mysticete whale females seem to minimize the metabolic overhead for themselves and newborn and sucking calves. Females suckle their young for a short period of 6-8 months on very lipid-rich milk, which is produced from maternal body stores, and wean abruptly. After lactation, mothers need to replenish body reserves, which usually takes a year. Females, therefore, generally breed every other or every third year. Females of the tropical Bryde’s whale {Balaenoptera edeni) have similar gestation and lactation lengths but show much less of a seasonal breeding pattern. Despite the impressive energy flux involved in pregnancy and lactation in mysticetes, there is no evidence that this reproductive effort incurs clear fitness costs, i.e., really constitutes parental investment, for example, by systematically reducing the probability of a successful pregnancy in the year following lactation.
Large odontocetes have pregnancy periods lasting longer than 1 year and take 8-36 months to wean their young (but reports on 13-year-old sperm whales with milk in their stomachs also exist). All larger odontocetes need more than a year to wean. Females consequently need much longer than 2 years to complete one reproductive cycle. The long period of nursing, even in dolphins, allows young to profit from the milk supply while gradually developing independent hunting skills.
It is presently speculated that the difference in rearing strategy between the two groups depends largely on the difference in hunting strategy. Mysticetes prey on small schooling prey, which are supposedly easy to catch for recently weaned young, particularly since weaning seems to coincide with the annual peak abundance of prey. In contrast, odontocetes eat large prey singly, which forces them to use more complex prey-hunting tactics. They may need to learn a lot more about prey distribution and behavior before young can feed successfully themselves. This may even involve teaching by mothers as suspected for the technique of beaching themselves used by southern killer whales hunting pinnipeds, e.g., on the breeding beaches of Argentina.
Another peculiar feature of some odontocetes is the occurrence of menopause. This phenomenon was documented for short-finned pilot whales (Globicephala macrorhijnchus) and killer whales. One idea about the functional significance of menopause, which finds some support in studies on humans, is that menopausal females may help their last born and previous daughters in the group through prolonged maternal and allo-maternal care, respectively, and by representing a living memory of how to deal with scarce resources and rare ecological disturbances. In matrilineal societies, menopause could be selected through indirect effects on the fitness of kin if the benefit to kin was larger than the benefit an old female could gain through further reproduction.
Sperm whales, long-finned pilot whales (Globicephala melas), false killer whales (Pseudorca crassidens), and bottlenose dolphins show evidence of reduced fertility with age, perhaps caused by extended periods of lactation. This change in maternal strategy might be expected from the life history theory because old females have a lower reproductive value and may therefore invest more in current offspring than young females. Alternatively, prolonged lactation in older mothers might be caused by a reduction in condition through previous parental care episodes and could then be considered a cost of reproduction.
Despite marked sexual size dimorphism in many species, there is little evidence for a sex-biased investment in sons vs daughters. Data on short-finned pilot whales suggest that sons are suckled for longer than daughters based on milk in the stomachs of caught individuals. Males had milk in their stomachs up to an age of 15 years, but females only up to an age of 7 years. This fits with the observation that males grow faster than females and are consistently larger at weaning than female juveniles. Similar observations have been reported for sperm whales. If such a difference in effort expended on the two sexes were consistent in the population, one would predict from Fishers model of sex allocation that the sex ratio at birth or at weaning should be strongly biased toward females. There is as yet no evidence for such a sex ratio bias.
B. Pinnipeds
Pinniped females produce one pup per year (instances of twinning are exceptional). They have a postpartum estrus (otariids) or copulate around the time of or after the weaning of young (phocids). Implantation is delayed and followed, after 3-4 months, by an 8- to 9-inonth pregnancy. The young are reared on lipid-rich milk and are usually weaned at an age less than 1 year, thus allowing an annual cycle of reproduction. Pinnipeds follow one of two alternative pup-rearing strategies. Fe males may nurse pups for a short time (a few weeks) from body reserves and then wean abruptly. This is termed the fasting strategy and is typical for larger phocids. Smaller phocids and otariids combine nursing from body reserves for a short (in otariids, 1 week) phase after parturition with regular cycles of alternate foraging close to the colony and nursing ashore (the foraging cycle strategy of pup rearing). Phocids wean pups after a shorter duration of lactation (4-65 days) at a smaller mass relative to maternal mass (about 25-35%) the otariids (after 120 days to 3 years, at 35-55% of maternal mass).
Contrary to earlier beliefs, the rearing strategy depends more on female body mass than on phylogenetic grouping. The larger a female, the richer a food resource she needs to support pup rearing by foraging during lactation. This is because the energetic costs of traveling to and from the food resource and maintaining the metabolism of mother and pup during the foraging sojourn become too high for large pinnipeds to make foraging during lactation a feasible strategy. Therefore, large species separate foraging completely from lactation, store massive energy reserves during a long foraging period in rich feeding areas often far away from breeding sites, and then fuel lactation out of body reserves. Because phocid females are on average larger (median maternal mass for all species 140 kg) than otariid females (median 55 kg), this can explain why phocids usually follow a fasting strategy and otariids a foraging cycle strategy. The largest otariid, the Steller sea lion (Eume-topias jtibatus), with a female mass around 250 kg, needs to take its unweaned young close to the foraging areas to maintain lactation by alternate foraging and suckling. Similarly, walrus (Odobenns rosmarus) females take their young to foraging areas where they are suckled while the mother can forage nearby. Ecological constraints thus play an important role in shaping maternal strategy.
Evidence is mixed that the energetic effort expended on pup rearing induces fitness costs of reproduction. In the Northern elephant seal (Mirounga angustirostris), primiparous (giving birth for the first time) young females are less likely to bear a pup in the year following birth than older females, thus suffering reduced fertility as a consequence of early onset of reproduction. Also, survival seems to be reduced when females first reproduce at 3 rather than 4 years of age, implying a mortality cost. However, these results were obtained at one but not another site on the Californian islands and the interpretation is not entirely clear. In otariids, Galapagos fur seal (Arctocephalus galapagoensis) females incur fitness costs of reproduction in terms of reduced fertility because they frequently lose a newborn pup when still accompanied by their previous young, be it a yearling or a 2 year old, by the time the next pup is born. This arises because young Galapagos fur seals grow extremely slowly and therefore take unusually long (up to 3 years) to become independent of their mothers. They may suckle for a second or third year if environmental conditions are poor, and thus preclude rearing of another pup by their mother. Clear evidence for a fertility cost of reproduction was also found for Antarctic fur seals (A. gazella). Parturient females of all ages were significantly less likely to reproduce in the subsequent year than nonparturient females. In addition, females that reproduce in a given year are less likely to survive to the following year than nonreproducing females, a clear mortality cost of reproduction. In this species and the northern fur seal (Cal-lorhinus ursinus). females older than about 13 years appear to show reproductive senescence. These old females are less fertile and produce smaller newborns than prime-age females. Particularly for otariids, there is thus evidence that the high energetic effort expended by females on pup rearing indeed constitutes maternal investment because it produces fitness costs to the mother.
It has been claimed repeatedly that female pinnipeds of highly polygynous species invest differentially in male and female offspring. Following sociobiological arguments, this would be expected if an increased investment in males (the larger sex showing greater variance in reproductive success) would lead to a greater expected reproductive success of such males. In many polygynous pinnipeds, males are born heavier and grow faster than females. This was taken as evidence for greater maternal investment in sons. However, this is no proof of greater investment in male offspring because male pups of some otariids store less fat and produce more lean body mass than lemale pups. Fat has a higher energy density than lean body mass, and consequently smaller female pups may have taken as much en-erg)’ from their mothers as the larger, leaner male pups. Also, the most important growth spurt determining later male adult size and probably reproductive success occurs generally after weaning, thus making it less likely that male offspring derive direct benefits for their future reproductive success from increased maternal investment.
IV. Mating Systems
Females are producers that are limited, in mammalian reproduction. by the time and resources needed for pregnancy and lactation, as well as by the recovery of condition after a reproductive cycle. This constitutes strong selection to optimize acquisition and efficiency of use of resources for reproduction. Because the maximal reproductive rate of female mammals is necessarily much lower than that of males, which in the extreme need only the time of one copulation to produce offspring, females become a limiting resource for the reproduction of males. This leads to selection on males for an increased ability to get access to and breed with as many females as possible, which is the essence of sexual selection. In mammals such as whales and pinnipeds, where males do not contribute to the care of offspring, males are expected to conform to the description of “ardent” males, eagerly searching for females and even harassing them for copulation. Females distribute themselves in relation to the distribution of resources (food and adequate habitat for reproduction), and males map onto this distribution of females. This difference in the selection on reproductive strategies of males and females leads to the phenomenon that quite often the sexes behave and morphologically look as if they belonged to different, competing species.
Sociobiological reasoning therefore leads to the expectation that observed mating systems represent the compromise emanating from the conflict of the sexes. Competition among males for access to females can take the form of aggressive competition. but can also occur via sperm competition when several males compulate with one female, as demonstrated for northern elephant seals. Such sperm competition, if occurring regularly, is expected to lead to the evolution of large testis mass, as larger testes produce more sperm and thus increase a male’s chances to fertilize the ova of females in competition with sperm of other males. Such an increase in relative testis mass was documented in other mammalian orders where species in which multiple copulation is frequent have larger testes than species where only one male copulates with a given female, whether the social system is monogamous or polygynous.
A. Whales
Whale mating systems are still largely unknown partly due to the problem that copulations are hard to observe. Except for a few particularly observable species, this leaves only indirect methods of investigation, such as genetic analyses, to determine the mating pattern in the more elusive species.
Among mysticete whales, much is known about the humpback whale, so well known for its spectacular songs. During the mating season, males station themselves well spaced out and advertise their position. This is very similar to the lek structures observed in many birds. This song may attract females and keep competing males away, but there is presently no firm evidence for this inference. Alternatively, males follow females, and several males may be doing this simultaneously, leading to competition for proximity to the female. Apparently these males compete over mating access to a female. Because humpback whales have small testes for their size, it is unlikely that females will copulate with several males, thus inducing sperm competition. Such sperm competition is likely to occur regularly in right whales (Eubalaena spp.), which—weighing about 50 tons—have testes weighing 1 ton, in strong contrast to blue whales, which weigh about 100 tons but have a testis mass of only about 70 kg. Copulation is observed frequently in right whales but has never been described convincingly in humpbacks, despite much more study of the latter.
Mating patterns in odontocetes are somewhat better known from a few species. Male strategies vary from singly roving males in the sperm whale via mating groups in killer and pilot whales to males that cooperate to herd and perhaps force females into copulation in the bottlenose dolphins. Sperm whales show a pattern similar to elephant mating patterns in which single fully adult, highly aggressive males rove among female groups in search of receptive females. They stay only briefly with each one of the female groups and then go on. Presumably, the long intervals between estrus in females make it unprofitable for these males to stay with any one group of females waiting for one of the females to get into estrous condition. Only fully adult males appear able to compete in this system, and subadult males as well as nonroving males stay at higher latitudes, often in small schools, feeding and maximizing energy intake in this way to grow to a competitive size. It is unclear why these males might stay in small groups as bachelors because they are certainly not endangered by predation and it is unknown which foraging advantages they might derive from grouping.
The killer and pilot whale mating system is the most surprising. least understood of the whale mating systems and has no parallel in terrestrial mammals. Genetic evidence shows that males who remain philopatrically in their maternal group never father offspring in their own group but apparently in other groups. Some evidence suggests that several related males of one group may mate with several females in another group, presumably during repeated encounters of these pods. This was concluded from the genetic observation that offspring in a group seem to be paternally related.
B. Pinnipeds
The pattern of mating interactions among individuals depends greatly on the dispersion of females during the breeding season, which in turn reflects the availability of a suitable habitat for pupping and foraging. In pinnipeds, pack ice, fast ice, and land-breeding species differ widely in this respect.
Phocids breeding on ice floes seem to have a mating system best described as serial monogamy in which a male stays with a female that has recently pupped until she comes in estrus. He then leaves after copulation to search for another female. The reproductive success for males in such a mating system depends more on dieir aptness to locate females ready to mate than on fighting abilities. In such species, sexual dimorphism tends to be small, slightly reversed (males smaller than females), or nonexistent.
Some fast-ice breeding species also show reversed sexual dimorphism, which is best analyzed in the mating system of the Weddell seal. Females gather around cracks in the fast ice where they dive for food from holes in the ice. During the reproductive season, they pup near these holes and males claim territories under the ice and defend the holes against other males. Under these conditions, maneuverability is considered more important than sheer size in male-male competition. This may be the reason for the observed reverse sexual size dimorphism. Alternatively, females may be selected for larger size, enabling the production of larger young or the storage of more massive fat reserves for lactation, and males may have remained smaller for lack of such selection. Copulation is underwater and consequently little is known about the reproductive success of males in this mating system.
Phocid seals, such as the harbor seal, which breed in the water close to land areas where females loosely aggregate frequently, seem to engage in fights for the best stations where females have to pass by, and such males are often wounded. Fighting males seem to have the highest reproductive success and, in some places, the mating system of this species may be similar in structure to a lek.
When female otariids or land-breeding phocids come together on predator-free islands, the resulting high female density permits males to station themselves among females and attempt to monopolize access to females. This competitive situation sets up sexual selection for an ever increasing male size, leading to impressive sexual size dimorphism in a few phocid and many otariid species, such as the northern fur seal in which males weigh six times as much as females. In addition, a larger size also enables males of these species to remain fasting on territory for long periods, thus increasing their chances to mate with females.
Otariid mating systems have been described as resource defense, female defense, or leks. The presence of resources important to females with pups, such as shade or access to tide pools, on male territories was demonstrated experimentally for a few species. However, resource distribution is not sufficient to predict the exact location of female aggregations nor does it explain female gregariousness, i.e., an active tendency of females to approach each other. Because high female density correlates with increased pup mortality in breeding colonies, there is a marked cost to female gregariousness, which must be compensated by comparable benefits. Bartholomew (1970) suggested that female choice of genetically superior males was responsible for female gregariousness, but little evidence supporting this view has come forward. New studies suggest a strong selection of female gregariousness through avoidance of interaction with large males, whether territorial or not. In many otariid species and in elephant seals, interactions with a much larger male can be dangerous or even deadly to a female. Females can minimize the probability of interaction with territorial males by aggregating into a “selfish herd.” Through this effect and by avoidance of dangerous and sometimes directly fatal harassment of females and pups by marginal males, females are selected to group much more closely than can be explained by resource distribution. Thus, the stationing of large adult males on clustered territories among parturient and es-trous females creates a resource “peace from marginal male harassment.” Except for this socially created resource, the system could also be described as one in which males are lekking on areas where females are forced to stay for a while because they spend the period between parturition and postpartum estrus near-stationary on land. Males may benefit from clustered territories by the reduced chances of losing females when disturbed by marginal subadult or adult males. Within this system, the male defense of access to females varies intra- and inter-specifically with habitat and female density from female defense to territorial site defense with larger or smaller territories reminiscent, respectively, of a resource defense or a lekking system.
Dominating males in these land-based breeding systems gain most copulations and can reach quite extreme reproductive success, sometimes up to 100 copulations in one breeding season. However, genetic studies have shown that observed number of copulations does not always correlate well with actual paternity, suggesting that peripheral, apparently “excluded” males may also gain some reproductive success by keeping close to female groups. There is no evidence of female choice beyond the mechanism that females in dense groups attract the strongest males in the best condition because only these are competitive enough to station themselves in the middle of female groups. Female elephant seals protest when they are attacked and forced into copulation by subadult or peripheral males, thereby attracting the attention of the dominant male who will often chase off the smaller male and copulate with the female. In this way females indirectly choose dominant males as copulation partners.
Gray seal mating systems on land strongly resemble the otariid system. Whether female choice plays a more important role in gray seal colonies remains to be seen. Genetic analysis of subsequent offspring of individual females suggests that females copulate year after year with the same male, even though they may stay in the territories of different males. This would suggest that they actively choose a particular male or, in case of multiple copulations, have mechanisms to choose among sperm of several males. In more dispersed breeding species, such as harbor seals, males have no chance to defend access to females as these are too mobile and not available continually in the same areas, depending on tide level and sea conditions. In situations in which females breed on sandbanks, harbor seal males were observed to station themselves in areas where females are likely to pass by and make themselves obvious through vocal display. Whether this and similar observations on walrus males that station themselves near females and produce bell-like sounds can be considered a lek display need further investigation.
V. Sirenians and Sea Otters
Much less is known about the social life of sirenians and sea otters and. therefore, these two groups are only treated briefly here.
The only clearly recognizable social structure in sirenians is the mother-offspring bond, which may last for up to 3 or 4 years. Other than that, it appears that the dispersion of most sirenians relates directly to the distribution pattern of food, aquatic macrophytes, fresh water for drinking (in the Florida manatee, Trichechus manatus), and, particularly in winter, warm water areas. Animals may migrate for large distances between such resources. However, it seems possible that underlying the apparent asocial pattern may be a subtle pattern of individualized relationships. This might be hypothesized from “greeting” displays exchanged between individuals that meet only occasionally at widely distant sites.
Cows in estrus seem to induce male scramble competition. In Florida, manatee herds of up to 20 males may follow an estrous female and compete by pushing to get into a favorable position for mating. In dugong males, competition may take a more aggressive lorm in which males may wound each other with their tusks. For West Australian dugongs. mating competition may lead to a form of lekking. However, the evidence is largely circumstantial.
Sea otter spatial dispersion also related to the need to live close to the coast where they forage relatively shallowly for macroinvertebrates. Females claim often overlapping foraging territories year-round along the coastline. They sometimes aggregate in small groups, so-called “rafts.” Young males, and fully adult males outside the reproductive season, also frequently form rafts close to areas of rich feeding resources. Such rafting is presumably related to the reduction of predation risk.
During the reproductive period, fully adult males establish territories that may overlap more than one female territory. This presumably provides males with a chance for a mild form of polygyny, but hard evidence for paternity of such males is at present missing.
VI. Concluding Remarks
Much of the sociobiological interpretation of observations on marine mammals is still in a speculative stage. This situation reflects our lack of detailed knowledge about the marine environment and in particular the macro- and microdistribution of resources vital to marine mammals. Clearly, more observation, more comparative studies, and especially more experimental work are urgently needed to understand the sociobiology of these magnificent animals. Obviously, experimental work will be particularly challenging and can only be successful if built on the thorough knowledge of marine mammal natural history. However, a well-founded functional understanding of the social behavior of marine mammals cannot be achieved without experimental tests of our many assumptions. Ingenious instrumentation and molecular’ genetic tools, developed during the last decade, should prove most helpful in making this summary of marine mammal sociobiology soon outdated.
