Biogeography is the study of the patterns of geographic distribution of organisms and the factors that determine those patterns. Although marine mammals are very mobile and there is an apparent lack of physical barriers in the world ocean, only the killer whale (Orcinus area), the sperm whale (Physeter macrocephalus). and perhaps some of the balaenopterids could arguably be considered to have cosmopolitan distributions. Other species have restricted distributions (e.g., coastal South America, Indo-West Pacific), reflecting their ecological requirements and their geographic centers of origin. Because related species tend to have similar ecological requirements and dispersal abilities, the distribution of higher taxa can also show distinct tendencies and restrictions, which reflect the cumulative distributions of their included species. For example, while delphinids, river dolphins, and sirenians have their highest diversity in tropical latitudes, the vast majority of pinniped, ziphiid, and phocoenid species occurs in temperate and polar regions. From a geographic perspective, specific regions can thus be characterized as centers of diversity for these higher taxa. and past global changes in the environment will have influenced their evolutionary history. For example, cooling of the world climates during the Tertiary may have contributed to the radiation of cold water-adapted pinnipeds and mysticetes.
I. Types of Distributions
At the species level, distribution patterns can be described at different spatial scales. Broadly speaking, individual species are usually limited to certain latitudinal zones, such as tropical, temperate, or polar regions. These descriptions can be refined further into subtropical, cold temperate, and so on and incorporated into patterns of ocean basin or hemisphere endemisin. For example, the Clymene dolphin (Stenella clymene) occurs only in the tropical Atlantic, the Steller sea lion (Eumetopias jubatus) in the cold temperate North Pacific, and the dugong (Dugong dugon) in the tropical Indo-West Pacific. On even smaller scales, species may be associated with specific physical features, such as nearshore coastal areas (e.g., the hump-backed dolphins, Sousa spp.) or the continental slope (e.g., Baird’s beaked whale, Berardius bairdii), or with oceanographic features, such as specific water masses or even bodies of freshwater (e.g., the baiji, Lipotes vexillifer, and the Baikal seal. Pusa sibirica).
A few species, notably some of the baleen whales, are highly migratory, summering at high latitudes and spending the winter breeding season at lower latitudes. Some of the migrating rorqual species occupy (at least seasonally) a wide range of latitudes in both hemispheres, although the movements of the Northern and Southern Hemisphere populations are seasonally offset so that they do not normally co-occur in the tropics. At the other end of the spectrum, there are some species (e.g., the vaquita, Phocoena sinus, and the Hawaiian monk seal, Monachus schauinslandi) that have highly restricted ranges. If a formerly wide-ranging species is now limited to a small area, its distribution is considered relict.
There are distributions that are described as pan-tropical (or pan-tropical/temperate), exhibited by some delphinids (many species), ziphiids (e.g., Blaineville’s beaked whale, Mesoplodon densirostris), kogiids (e.g., the pygmy sperm whale, Ko-gia breviceps), and balaenopterids (e.g.. Bryde’s whale, Balaenoptera edeni). A few species and species pairs occur at higher latitudes in both hemispheres but are absent from tropical waters, the so-called antitropical species and species pairs. These are seen in the families Delphinidae (e.g.. the right whale dolphins, Lissodelphis spp.), Ziphiidae (e.g., the bottlenose whales, Hyperoodon spp.), Phocoenidae (e.g., Phocoena sinus and Burmeister’s porpoise, P. spinipinnis), Phoci-dae (e.g., the elephant seals, Mirounga spp.). and Otariidae (e.g.. the Guadalupe fur seal, Arctocephalus townsendi, and Juan Fernandez fur seal, A. philippii).
II. Ecology and History Determine Distribution
Beyond these descriptive aspects of biogeography are the factors that determine a given species’ distribution. Generally, distributions are determined by the ecology and the history of the species. In some cases, distribution is limited because a species may not be adapted for living in certain environments.
For example, tropical delphinids may not range into higher latitudes due to limitations on their abilities to thermoregulate in colder water or find food in different habitats. Tied into this is competition, either from closely related species or from ecologically similar species, which may exclude a species from a particular region in which it could otherwise survive. In the case of South American manatees, it is reasonable to surmise that competition places at least one boundary on species’ ranges. Throughout most of its range, the Caribbean manatee (Trichechus manatus) occurs in both coastal and riverine habitats. However, it does not range into the Amazon River, where the exclusively freshwater Amazonian manatee (T. inunguis) occurs, although it occupies the coastal areas on either side of the river mouth. Here, the two species are parapatric, and competitive exclusion is likely at work.
The role that history plays in biogeographic patterns should not be overlooked, but it is not always evident from contemporary distributions. The dispersal abilities of organisms may partly explain why species occur in some areas and not in others. For example, the lack of otariids in the North Atlantic is probably not due to the lack of suitable habitat, but rather lies in the inability of any North Pacific or South Atlantic species to get there. Of course, one could also tie this into their ecological requirements, in that dispersal to the North Atlantic would be more likely if North Pacific species ranged far enough north for animals to disperse via the Arctic Ocean across northern North America or Eurasia. For some species that have widely separated allopatric populations (e.g., Conunerson’s dolphin, Cephalorhynchus commersonii), dispersal from one region to the other is a likely explanation for their distribution. In other cases, vicariance events can explain allopatric distributions. For example, the two subspecies of the Indian river dolphin (Platanista gangetica) occur in different river systems: the Indus and Ganges-Brahmaputra River. Although the two forms are not presently in contact, these rivers were all part of a single system until the late Pliocene and probably had sporadic connections through stream capture even until historical times. Therefore, the geographic separation of the populations is from a rather recent vicariance event.
Large-scale changes in the environment can have dramatic influences on species’ distributions. For example, in times of global cooling, cold boundary currents in the ocean basins may have extended further toward the equator. This, in turn, could have enabled temperate species to disperse across the equator to similar habitats in a different hemisphere, giving rise to the antitropical species. Among the antitropical species and species pairs, some tendencies in their distributions are apparent. For example, for only one group (Eubalaena spp.) does the northern counterpart occur in both the North Atlantic and the North Pacific. Although the long-finned pilot whale, Globicephala melas, has only been recorded from the North Atlantic and the Southern Hemisphere, 1000+-year-old skulls of this species have been unearthed in Japan. For the rest of the eight or so recognized antitropical species and species pairs or trios, all except Htjperoodon have their northern members limited to the North Pacific. Perhaps the oceanographic and climatic conditions that allow transequatorial dispersal for temperate species occur more frequently or become more developed in the Pacific basin than in the Atlantic. These comparisons do not include the latitudinal migrant species, such as many of the species of balaenopterids. For these, their seasonal occurrence at low latitudes greatly facilitates transequatorial dispersal and would not likely require any significant change in oceanographic or climatic conditions.
Beyond the consideration of the underlying mechanisms of a single species’ distribution, it is possible to make inferences about the origins of entire ecological communities. Vicariance biogeographers look for congruence between the phylogenetic relationships among species and their geographical distributions. Species distributions can be superimposed on phylogenetic trees to create what are called area cladograms (Fig. 1). If the area cladograms of several unrelated but geographically similar higher taxa are congruent, it is good evidence that a specific sequence of vicariance events operated on all of those taxa as speciation mechanisms. Furthermore, it may allow the researcher to make inferences about the centers of origin for the higher taxa being considered.
Finally, one should try to incorporate the fossil and geologic record when inferring historical mechanisms in biogeography, especially among distantly related taxa. A case in point can be seen in the river dolphins. Among the river dolphins, the Amazon Inia geoffrensis and the franciscana Pontoporia blainvillei appear to be the closest (albeit very distant) relatives among the extant species, with the former occupying several South American rivers that flow into the Atlantic and the latter occurring along the Atlantic coast of South America. However, the closest (albeit even more distant) living relative of this pair is probably the baiji (Lipotes vexillifer), which is only found in the Yangtze River in China. Considering their freshwater and nearshore habits, it is not obvious from their contemporary distributions how they came to occupy areas a world apart. However, the fossil record has yielded intermediate species from various localities across the North Pacific. Furthermore, the geologic record shows that in the late Pliocene, the major river system in northern South America flowed westward, into what is now the Gulf of Guayaquil in Ecuador. It is thought that the ancestors of Inia entered this system from the Pacific. With the uplift of the Andes, much of the river reversed direction and flowed eastward, eventually becoming rivers such as the Amazon and Orinoco. With the North Pacific intermediate forms dying off and the Inia lineage splitting to give rise to the ancestors of Pontoporia along the coast, one can see how the present-day species distributions came to be. It should be kept in mind, however, that determining the distributions of fossil taxa is notoriously difficult, especially for offshore species that are poorly represented in accessible deposits.
III. Taxonomic Patterns
As mentioned earlier, species within higher taxa share characteristics of their distributions to some degree. It is therefore possible to characterize the distributions of the different groups of marine mammals. The sirenians are primarily a tropical group, with mostly allopatric species occurring in warm coastal waters and some rivers of the Indo-West Pacific and both sides of the Atlantic. The trichechids are represented by two species in the new world (Trichechus manatus and T. inunguis) and a single congener, the African manatee (T. senegalensis), in western Africa, indicating the occurrence of a past trans-Atlantic dispersal event within that lineage. The family Dugongidae, formerly more diverse and widespread, now has only one extant species, the dugong (Dugong dugon), that occurs in the Indian and west Pacific Oceans. One recently extinct species of dugongid, Steller’s sea cow (Hydrodamalis gigas), had a restricted range in the Commander Islands of the North Pacific, an anomalously cool habitat for a sirenian.
Figure 1 In vicariance biogeographtj, speciation patterns are determined by vicariance events. The analysis attempts to reconstruct the sequence of vicariance events using the pattern of evolutionary relationships within a group of related species with al-lopatric distributions, (a) Species “a,” “b,” and “c” occupy ranges 7, II, and III, respectively. (b) If a phylogenetic analysis determines that “b” and “c” are sister species to the exclusion of “a,” this pattern of relationships is applied to their respective geographic ranges in an area cladogram. (c) Under this scenario, the range of the ancestral species is first divided by a vicariance event into a northern and a southern half Pojndations in these two areas speciate into species “a” and “a1.” Species “a”‘ is the inferred immediate common ancestor to “b” and “c.” A later vicariance event divides the range of “a”‘ into eastern and western halves, giving rise to species “b” and “c.” If unrelated species groups occupying these areas shoiv congruent area cladograms, the support for this sequence of vicariance events is strengthened.
The majority of phocid species inhabit cold temperate and polar regions. Although no species occurs in both Northern and Southern Hemispheres, there are numerous species that are circumpolar either in the Arctic (e.g., the bearded seal, Erignathus barbatus) or in the Antarctic (e.g., the crabeater seal, Lobodon carcinophagus). In fact, all of the southern phocid species have veiy broad distributions; their range expansions were probably assisted by the oceanic currents that traverse all longitudes in the Southern Ocean. In the Northern Hemisphere, however, the habitats and ocean currents are much more fragmented by the continental land masses. In addition to the circumpolar species, there are northern species that have more restricted ranges, either endemic to a single ocean basin (e.g., the gray seal, Halichoerus gnjpus), or limited to landlocked bodies of water (e.g., the Caspian seal, Pusa caspica). In contrast to the rest of the family, the three recent (two extant) species of monk seals (Monachus spp.) inhabit(ed) warmer waters of the Mediterranean and eastern Atlantic, Caribbean, and Hawaii. The spread of monk seals to Hawaii must have occurred prior to the rising of the Isthmus of Panama, which has separated the Caribbean and Pacific basins for the past 3 million years.
As a group, the otariids are similar to die phocids in their distribution, although they are less well represented at very high latitudes (near the pack ice) and do not occur in the North Atlantic at all. Also, individual species tend to have more restricted ranges that are widely allopatric from their congeners. For example, the fur seal genus Arctocephalus is very widespread in the Southern Hemisphere, represented by six species (with an additional species endemic to the Galapagos Islands and another to the eastern North Pacific). However, there are only a handful of localities where more than one species occurs together; for the most part, the species are allopatric. It appears then that the dispersal abilities of fur seals have allowed them to colonize many areas in the Southern Hemisphere but have not prevented the resulting disjunct populations from speciating. Odobenids are represented by a single circumpolar Arctic species, the walrus (Odobenus rosmarus).
Cetacean species exhibit a wide range of distribution patterns. The family Balaenidae includes one circumpolar Arctic species (B. mysticetus) and a trio of very closely related species distributed in the North Atlantic, North Pacific, and Southern Hemisphere (Eubalaena spp.); until recently, these were thought to comprise a single species. The gray whale (Eschrichtius robustus) and the various species of balaenopterids are mostly latitudinal migrants in both hemispheres, although two species (the Bryde’s whales, Balaenoptera brydei and B. edeni) are restricted to tropical and warm temperate waters, and some primarily migratory species include isolated populations that may be nonmigratory (e.g., the humpback whale, Megaptera novaeangliae, in the northern Indian Ocean). In addition to the widespread species of minke whale (Balaenoptera acutorostrata), the Southern Hemisphere also contains an endemic species of minke whale (B. bonaerensis). Similarly, the Southern Hemisphere is also home to two distinct forms (considered subspecies at present) of blue whale (B. musculus). In both of these cases, it is not known if the two southern forms represent divergent lineages that arose within the southern ocean or if they were the result of independent dispersal events across the equator.
Sperm whales are virtually cosmopolitan, and the kogiids (Kogia sima and K breviceps) are worldwide in tropical and warm temperate waters. Beaked whales show a variety of distribution patterns, including pan-tropical species (e.g., Mesoplodon densirostris), antitropical species pairs (Berardius spp.), and ocean basin endemics (e.g., Sowerby’s beaked whale, M. bidens). Some (e.g., the pygmy beaked whale, M. pemvianus) are only known from a few strandings within limited geographic areas. For most species of sperm whales and beaked whales, so little is known about their habits and ecological needs that it is difficult to hypothesize about the mechanisms that have led to their present distributions.
Three of the four species of river dolphins (Inia geoffrensis, Lipotes vexillifer, and Platanista gangetica) are strictly freshwater in specific tropical river systems, with the fourth species (Pontoporia blainvillei) having a restricted marine coastal range. The two species of monodontids (the narwhal, Monodon mo-noceros, and the beluga, Delphinapterus leucas) are circumpo-lar in the north and are among the few resident polar cetaceans, although fossil species of this family occurred as far south as San Diego, California.
Apart from a single Indo-West Pacific coastal species that also ranges into freshwater (the finless porpoise, Neophocaena phocaenoides), the phocoenids are strictly marine and cold temperate to warm temperate in distribution, some with very restricted ranges (e.g., Phocoena sinus). Only one phocoenid, the harbor porpoise (P. phocoena), has invaded the North Atlantic, becoming very widespread in both oceans of the Northern Hemisphere and even establishing isolated populations in the Black Sea and off West Africa.
The most speciose family of marine mammals, the delphinids, shows a wide variety of distributions, from pan-tropical species (e.g., the pantropical spotted dolphin, Stenella attenuata) to ocean basin endemics (e.g., the white-beaked dolphin, Lagenorhynchus albirostris) to species with wide-ranging but disjunct populations (e.g., the long-beaked common dolphin, Delphinus capensis). Many delphinids are pelagic, although some inhabit coastal waters (e.g., Cephalorhynchus spp.) and some even invade freshwater (e.g., the tucuxi, Sotalia fluvi-atilis). Only one, Orcinus orca, seems to regularly range to the pack ice in the far north and south. For the many pan-tropical/ warm temperate species, the continental land masses effectively separate the populations inhabiting the Indian and Pacific Oceans from those inhabiting the Atlantic Ocean, raising the question of how they came to inhabit all the ocean basins. It has been hypothesized that during warm climatic periods, warm water extended far enough south to allow interchange and range expansion around the Cape of Good Hope. This would enable some species to become pan-tropical in their distribution, and the subsequent retreat of the warm weather and isolation of populations could provide a speciation mechanism for the establishment of the tropical species endemic to the Atlantic Ocean (die Adantic spotted dolphin, S. frontalis, and S. clymene).
IV. Conclusion
Why do species five where they do? Answering such a simple question requires the examination of clues from the past as well as the present. Biogeography involves such diverse disciplines as geology, paleontology, ecology, physiology, behavior, and system-atics. For marine mammals, studying biogeographic patterns presents real challenges. There is a paucity of information about past distributions and habitats, gaps in our knowledge of contemporary and recent distributions, uncertainties about evolutionary relationships, and a tremendous amount to learn about the basic ecology and physiology of many marine mammals.
