Marine centres of origin as evolutionary engines

Authors


82-651 Skyview Lane, Indian Palms C. C., Indio, CA 92201, USA

The world ocean supports a dynamic system in which living organisms undergo constant movements. Although some would appear to be sedentary, all are capable of invading new territory at some stage in their life cycle. Underlying these comparatively rapid changes is a much slower evolutionary system whereby new species are formed and spread out. Depending on their place of origin and genetic resources, some of the new species may give rise to continuing phyletic lines. At the same pace, some older species approach extinction by continuing to lose territory. Over time, this evolutionary system appears to be no less dynamic than the contemporary one.

In recent years, considerable attention has been paid to the East Indies as a centre of origin for the marine tropics. While other centres of origin in the Antarctic and the North Pacific have been recognized, little attention has been paid to their external influence. Yet in the cooler waters of the oceans, they are as important to those areas as the East Indies is for the tropics. Evidence indicates that evolutionary flows from all three centres contribute to a dynamic system that extends throughout the world ocean. Each of the three centres and its salient features is discussed in turn, the information is summarized, and then stated in the form of a hypothesis.

The antarctic centre

Antarctica has a long history of harbouring a temperate marine fauna. That assemblage first become apparent in the Devonian when the Antarctic continent was still attached to the southern part of Gondwana. At that time, Antarctica belonged to the Malvinokaffric Realm that was distinguished by its lack of many common tropical animal groups (Boucot, 1988; Crame, 1994). This latitudinal position, suggesting the presence of relatively cool conditions, continued throughout the Mesozoic and into the early Tertiary. From the early Cretaceous to the Oligocene, a warm-temperate Weddellian Province extended from the tip of South America to Antarctica and to Australia/New Zealand. That Province disappeared with the development of colder temperature and the Antarctic Circumpolar Current (Crame, 1999).

A dramatic change in the temperature zonation of the earth began with the global climatic deterioration of the middle Eocene. A new cold-temperate zone, with winter temperatures from 12 °C down to 2 °C, began to form at the highest latitudes. As a consequence, the warm-temperate waters were forced away from the poles and into lower latitudes. At one time, there was a general belief that Antarctic ice sheets did not form until the early Miocene when there was a circulation of a deep current through Drake Passage between Antarctica and Australia. But there is now evidence that ice sheets probably first formed in the Eocene c. 42 Ma (Keller et al., 1992). At Seymour Island, the La Meseta Formation of the middle-late Eocene has yielded a rich fauna of marine invertebrates as well as fossils of fishes, penguins, whales and marsupials. Some of the bivalves and decapods represent first occurrences of taxa that appeared later on at lower latitudes (Crame, 1994).

At first, the Antarctic ice may have waxed and waned with minor climatic cycles, but by late Oligocene time, c. 25 Ma, it appears that a major icepack grounding event took place (Bartek et al., 1992). This is consistent with terrestrial evidence indicating the presence of glaciers and tundra by the early Oligocene (Janis, 1993). The presence of a major icepack in the Antarctic in the late Oligocene suggests that the modern, cold sea surface temperature regime (+2 °C to −2 °C) had become established. By that time, much of the rich Eocene fauna had disappeared, except for animals such as the penguins, whales, fishes, and several of the bivalves and gastropods that remained to become ancestral elements of the modern marine fauna. Late Oligocene–early Miocene fossils from other parts of Antarctica show surprisingly close resemblances to living taxa (Clarke & Crame, 1989).

The shelf waters of the Antarctic and Sub-Antarctic are now occupied by a highly distinctive fauna that owes its origin to four historical factors: (1) persistence of a small ancestral group of Mesozoic and early Cenozoic taxa, (2) extinction of many early Tertiary warm-temperate species, (3) geographical isolation produced by the opening of Drake Passage, and (4) invasions by cold-temperate species from the northern hemisphere. The Antarctic Biogeographical Region is divisible into three provinces. The South Polar Province includes the Antarctic continent itself plus the South Shetland, South Orkney and South Sandwich Islands. South Georgia Island and Bouvet Island have many endemic species and each is placed in a province of its own. The Sub-Antarctic is the one biogeographical region on the world that is entirely made up of small, oceanic islands. Within that Region, The Kerguelen Province includes not only Kerguelen Island itself, but also the McDonald, Heard, Prince Edward and Crozet Islands. Macquarie Island has a high degree of endemism indicating that a Macquarie Province should be recognized (Briggs, 1995).

The faunal distinctiveness of the Antarctic Region is almost incredible. In the major invertebrate groups, more than half the species are endemics (Knox, 1994). In the fishes, the endemic rate is c. 95% (Gon & Heemstra, 1990; Miller, 1993). Considering the 20+ Myr of isolation, this degree of species endemism might be expected. However, there are also large numbers of endemic genera; for fishes it is c. 70%. For one suborder of fishes, the Notothenioidei, the Antarctic is the site of a major evolutionary radiation. This group, consisting of five families, forty-six genera and 122 species (Nelson, 1994), dominates the ichthyofauna. Notothenioids often comprise 80–90% of the species in Antarctic fish catches. Some of them demonstrate remarkable adaptations to the coldest water (−1.9 °C) by having a glycoprotein in their blood that lowers the freezing point, a lack of red blood cells and aglomerular kidneys.

The shelf and upper slope of the Antarctic Region supports 213 fish species. Of these, ninety-six are notothenioids, sixty-seven liparidids (Liparididae) and twenty-three zoarcids (Zoarcidae). Together these three groups comprise 88% of the fish fauna (Eastman, 2000). Unlike the latter two groups, that apparently represent invasions from the North Pacific, the notothenioids probably arose in the Antarctic vicinity. A fossil skull originally identified as a gadiform fish (Eastman & Grande, 1989) has been found to be a notothenioid (Balushkin, 1994). This fossil, from the late Eocene of Seymour Island, Antarctic Peninsula, is the first known fossil of the suborder.

In the course of their evolution, the notothenioids adapted to a wide variety of ecological niches (Eastman & Grande, 1989). At the same time, they dispersed in a biogeographical sense. The most primitive (plesiomorphic) of the five families is generally considered to be the Bovichthyidae. Its distribution is almost entirely Sub-Antarctic with some species ranging as far north as New Zealand and Australia. One species inhabits freshwater streams in Australia and Tasmania. The Nototheniidae, the second most primitive family, is widely distributed in both the Antarctic and Sub-Antarctic. The remaining three families, usually considered the most advanced (apomorphic) (Iwami, 1985), are mainly confined to the Antarctic continent. This kind of phylogenetic pattern, in which the more primitive groups occupy the periphery, is a centre of origin characteristic.

Notothenioid phylogeny has also been investigated using mitochondrial DNA analysis (Bargelloni et al., 2000). The results confirmed repeated dispersals outward across the Antarctic Convergence, often followed by speciation. Within the notothenioid families Nototheniidae and Channichthyidae, the most primitive genera may be seen to have peripheral distributions (Andriashev, 1986). The Notothenioidei belongs to the order Perciformes and no definite perciform fossils have been found that are older than the Tertiary (Patterson, 1993). Most authors have correlated the radiation of the suborder with the cooling and subsequent isolation of Antarctica (Anderson, 1990). However, the discovery of the late Eocene notothenioid places it in a warm-temperate fauna along with many other warm water species. Considering the impressive amount of evolutionary change that has taken place, the notothenioids may have originated in the earliest Tertiary or even in the late Cretaceous.

The Antarctic is an evolutionary centre for penguins (Spheniscidae). Fossils have been found from the Eocene, Oligocene, Miocene, Pliocene and Pleistocene and all finds have been made in the circum-Antarctic area where penguins still occur (Simpson, 1974; Carroll, 1988). At times in the past, there were many more species but most of them have become extinct. The modern fauna consists of six genera and eighteen species. From their diversity centre in the Antarctic, penguins have spread to northern Chile and the Galapagos Islands, southern Africa, Amsterdam Island in the southern Indian Ocean and New Zealand.

Antarctica, because of fossil discoveries in its peninsula region, has been steadily assuming greater importance as a centre of origin for Southern Ocean invertebrates. Crame (1996) published a list of living molluscan genera found in present day New Zealand, Australia, or southern South America, that have their earliest fossil records in the La Meseta Formation. The list includes ten genera in six families. In addition, several additional genera from the same formation were noted to have become even more widespread. In the same work, Crame included an extensive list of bipolar mollusca. However, almost all the examples involved relationships between the cold-temperate zones of each hemisphere rather than strictly Antarctic vs. Arctic. Two species of benthic sea-weeds of probable Antarctic origin also exist in the Arctic (van Oppen et al., 1993), but they are also broadly distributed in the cold-temperate waters of both hemispheres.

The continual sinking of cold, saline water adjacent to the Antarctic continent and its subsequent movement northward at abyssal depths has had important biogeographical consequences. Historic dispersals from the Antarctic to the major ocean basins and trenches have been proposed for a large number of invertebrate families and genera (references in Vinogradova, 1997). For example, the holothurian family Elpidiidae apparently originated in the Antarctic, then dispersed northward through the Atlantic to the Arctic Ocean and also through the Pacific to the Bering Sea (Gebruk, 1990). The marine isopods in the deep sea may have been introduced to that environment via the Antarctic (Kussakin, 1973). As cold water was conducted into the deep sea from the late Eocene through the Oligocene, it caused significant extinctions. But, at the same time, it created a habitat suitable for the introduction of cold-acclimated organisms from the Antarctic. As a centre of origin, the Antarctic may have had its greatest effect on the abyssal and hadal zones.

In addition to its production of species that have invaded other areas, the Antarctic has provided a refuge for phylogenetic relicts. The most generalized gadiform family is probably the Muraenolepididae. Its four species are found on the shelf and slopes of the Antarctic and Sub-Antarctic, thus inhabiting the southern periphery of the order (Andriashev, 1988; Howe, 1990). The lantern fishes of the family Myctophidae originated in the tropical waters of the Tethys Sea but the three most primitive genera are now confined to the Antarctic and Sub-Antarctic (Andriashev, 1988). For the invertebrates, Antarctica has been described as one of the last strongholds of the brachiopods, together with hexactinellid sponges and certain bivalve genera (Crame, 1994). Gastropod relicts include some genera of volutes and marginellids (Powell, 1951).

The Antarctic regions have also provided a refuge for groups that have obviously invaded from the far north. Because such immigrants are not usually found in the intervening tropics, their distribution patterns are usually called bipolar. However, in most cases the term bipolar is misnomer, for the majority of the invaders appear to have originated in the northern cold-temperate seas rather than in the Arctic. Among the fishes, the dominant Antarctic groups, aside from the notothenioids, are the families Liparididae, Zoarcidae and Rajidae. The family Liparididae is of north Pacific origin (Andriashev, 1986). Members of the genus Paraliparis, a deep-water group, made their way to the Antarctic via the south American west coast.

The family Zoarcidae is also of north Pacific origin and also reached Antarctic waters by moving down the west coast route and across the Scotia Ridge (Anderson, 1988). In his analysis of the history of the Rajidae, Long (1994) determined its origin to have been in the western Tethys and in the boreal seas of western Europe. It apparently dispersed to Antarctica in the early to middle Eocene along the western margin of the Atlantic. There are seven endemic rajid (skate) species in Antarctic waters. Like the notothenioids, they are represented by ancestral fossils from the late Eocene of Seymour Island. Yet the rajids have changed very little in the last 40 Myr. The three fish families and most of the molluscs that have invaded from northern seas probably did so my means of isothermic submergence. But this seems unlikely for shallow water molluscs such as the littorinids and certain buccinids (Crame, 1996).

Five species of seals belonging to the family Phocidae are found in Antarctic waters. They are commonly known as the Southern elephant, Crabeater, Ross, Leopard and Weddell seals. The family originated along the shores of the North Atlantic, apparently in the Miocene. They must have reached the Antarctic early in their evolution for each of the species is placed in its own genus (Jefferson et al., 1993). One of the four southern fur seals of the family Otariidae, the Antarctic fur seal, occurs on the Antarctic Peninsula. The others occur in the Sub-Antarctic or farther north. The family originated in the north Pacific (Carroll, 1988) and, as all of the southern species belong to the same genus, their dispersal to the southern hemisphere probably occurred more recently than that of the phocids.

The north pacific centre

The westward convergence of North America on Eurasia, because of the opening of the North Atlantic, apparently resulted in the formation of a terrestrial connection in the late Cretaceous c. 80 Ma (Zonenshain & Napatov, 1989). Once formed, this Bering land bridge, often called ‘Beringia’, separated the warm-temperate marine biota of the North Pacific from that of the Arctic–North Atlantic. It has been generally thought that Beringia continued to act as a barrier until the opening of a seaway in the Pliocene, c. 3.5 Ma (Herman & Hopkins, 1980). But recent work on bivalves (Kafanov, 1999; Marinkovitch & Gladenkov, 1999) and Cancer crabs (Harrison & Crespi, 1999) indicate that the first Cenozoic opening took place earlier, perhaps as much as 6–12 Ma.

After the mid-Eocene, a series of temperature declines in the circumpolar region resulted in the southward movement of the warm-temperate zone and its replacement by a cold-temperate regime. By c. 14 Ma, fossils of numerous boreal (cold-temperate) species and genera could be found in the North Pacific (Golikov & Scarlato, 1989). But, judging from its systematic distinctiveness, that biota probably began its development in the mid-Eocene c. 40 Ma. At the same time, similar temperatures probably existed in the Arctic–North Atlantic. When the Bering Strait first opened, it may have been shallow and provided only limited passage but, by c. 3.5 Ma, it apparently allowed an unrestricted mingling of biotas that had been separated for more than 70 Myr.

The event of 3.5+ Ma has been called the Great Trans-Arctic Biotic Interchange (Briggs, 1995). Its biogeographical consequences have been evaluated, with an emphasis on the molluscan faunas, by Vermeij (1991). He identified 295 molluscan species that either took part in the interchange or had descended from taxa that did. Of these, 261 were determined to be of Pacific origin compared with thirty-four of Arctic–Atlantic origin. This gives a ratio of almost 8 : 1 in favour of the Pacific. The modern molluscan species diversity in the North Pacific is approximately twice as great as that of the Arctic–Atlantic. Although many of the molluscan species seemed to have had early Pleistocene origins, Vermeij determined that the vast majority arose by anagenesis (without lineage splitting) so that it was reasonable to suppose that there has been little diversification since the early Pliocene. This means that the asymmetry of the invasions cannot be accounted for by the 2 : 1 ratio in species diversity.

There are two viable hypotheses that might account for the predominate success of the invaders from the Pacific: (1) the Pacific species, having come from a more diverse ecosystem, are competitively superior, or (2) an extinction event eliminated much of the Arctic–Atlantic fauna so that it was easy for the Pacific species to occupy the vacated niches. Vermeij (1991) emphasized the importance of the latter cause which he called an ‘Hypothesis of Ecological Opportunity’. This is the same idea as ‘Incumbent Replacement Model’ presented by Rosenzweig & McCord (1991). Both are dependent on extinction to free ecological niches so that they can be occupied by an invader. Neither can explain how replacement occurs without the help of extinctions.

At the time of the great interchange c. 3.5+ Ma, the Arctic Ocean was ice-free and boreal (cold-temperate) conditions still prevailed (Golikov & Scarlato, 1989). The final closure of the Panamanian Isthmus c. 3.1 Ma strengthened the Gulf Stream system and favoured the onset of glaciation on the northern continents (Barry, 1989). Raymo (1994) concluded that a major intensification of northern hemisphere glaciation took place between 2.9 and 2.4 Ma. As a result, most of the boreal species were eliminated and the modern Arctic fauna began to develop.

Once established, the colder temperature of the Arctic waters prevented further penetration by boreal species, with the exception of some eurythermic arctic–boreal forms. This means that Atlantic boreal species of Pacific origin were already in place at least several hundred thousand years prior to the mid-Pliocene cooling episode. Therefore, it is not possible to ascribe the success of the Pacific invaders to an extinction in the Arctic–North Atlantic. It is more likely that the Pacific species, the products of a more highly diverse ecosystem, were the better competitors. The competition need not have been behavioural, but could have involved such factors as reproductive rate, individual size, or vulnerability to predators or parasites.

One must bear in mind that some invading species can successfully establish themselves by insinuation. This strategy is applicable to species that have evolved so that the niche they occupy is unusual to the extent that the invader will not directly compete with a native species. Insinuation may be suspected when an invader succeeds in colonizing an area with a more diverse ecosystem. It might help explain the success of the thirty-four molluscan species that dispersed from the Arctic–Atlantic to the North Pacific. Although the molluscan movements are the best known, as a result of the detailed analysis by Vermeij (1991), almost all groups of North Atlantic macroinvertebrates and fishes possess some species of North Pacific ancestry.

Among the fishes, the families Salmonidae, Osmeridae, Zoarcidae, Hexagrammidae, Cottidae, Agonidae, Liparididae, Stichaeidae and Pholididae probably originated in the North Pacific but, during the trans-Arctic interchange, contributed one or more species to the North Atlantic. The cod family Gadidae, on the other hand, developed primarily in the North Atlantic and contributed two species to the North Pacific. Among the marine mammals, fur seals and sealions of the family Otariidae and the walruses of the family Odobenidae originated in the North Pacific, while the seals of the Phocidae are of Atlantic–Mediterranean origin (Carroll, 1988), the eelgrass Zostera (Hartog, 1970) the red alga Phycodrys (van Oppen et al., 1995) and the kelp genus Laminaria (Estes & Steinberg, 1988) originated in the North Pacific, then spread to the North Atlantic.

Molluscan research indicates that the trans-Arctic invaders in the Atlantic have generally broader ranges than do native species with pre-Pliocene Atlantic histories (Vermeij, 1991). The apparent evolutionary consequences are of interest. Of the trans-Arctic species that extend into the Atlantic, 48% are derived (speciated) forms. In the Pacific 29% are derived. This indicates a remarkably low level of speciation for the past 3.5 Myr. Vermeij suggested that speciation among marine organisms appears to be much less frequent than assumed by evolutionary biologists. However, there is a difference in species longevity between cold-temperate and tropical habitats. Data from studies of the biota on each side of the Panamanian isthmus, where the separation has been in effect for c. 3.1 Myr, indicate an exceedingly high level of speciation.

Molluscan fossils have shown that c. 20–40% of early Pliocene boreal species in the North Pacific have become extinct; but in the boreal North Atlantic more than 50% of the early Pliocene species have been lost (Vermeij, 1989). Cooling episodes associated with northern hemisphere glaciation are generally recognized as the probable cause. Vermeij has maintained that cooling does not fully explain why the North Atlantic extinctions were so much greater, and that reduction in primary productivity must have played a part. But there are reasons why the North Atlantic, during glacial periods, undergoes more severe temperature declines (and extinctions) than does the North Pacific. The North Atlantic is a smaller ocean with a correspondingly smaller heat budget, and it is wide open to the inflow of ice from the Arctic Basin. The North Pacific is protected from ice and cold-water inflow by the Bering land bridge that has always been in place during glacial periods.

Another important effect of the cooling of the northern oceans was the separation of the boreal biotas. We have noted that, prior to the first Pliocene cooling episode c. 2.4–2.9 Ma, a boreal biota existed throughout the North Pacific and the Arctic–North Atlantic. Boreal organisms, with the exception of some wide ranging arctic–boreal forms, were extirpated, by the temperature drop, from the Arctic Ocean as well as from the northern parts of the Pacific and the Atlantic. A result was the establishment of a new cold-water Arctic Biogeographical Region. In the Atlantic, an Arctic biota now extends southward to the Strait of Belle Isle in the west and to the Kola Fjord at the base of the Murmansk Peninsula in the east. Included are all of the waters around Greenland and the northern half of Iceland. In the Pacific, an Arctic biota extends southward to Cape Olyutorsky in the west and Nunivak Island in the east. In each ocean, these southern extensions meant that the original Pliocene boreal regions were divided into two, one to the east and the other to the west. Typical boreal species were no longer able to maintain amphipacific and amphiatlantic distributions and evolutionary change began to take place separately in each region. This is why we are now able to define a boreal region on each side of each ocean in terms of its endemic species.

Molecular technology has been useful in analysing the genetic divergence displayed by some of the invaders from the North Pacific to the North Atlantic. Four distinct groups have been recognized (Cunningham & Collins, 1998). One group had apparently invaded c. 3.5+ Ma, then subsequently speciated on each side of the boreal Atlantic, but another group of early invaders had not speciated on each side. A third group showed evidence of a very recent migration, while a fourth apparently made two migrations, one early and the other recent. True boreal species should not be able to migrate through the Arctic Region but there is a minor eurythermic group of arctic–boreal species. The third and fourth groups identified probably consist of such species.

Although the biota of the Arctic Region owes most of its origin to the boreal North Pacific and North Atlantic, it has developed an appreciable amount of endemism. About 24% of the echinoderm species are endemic (Anisimova, 1989), 14% of the bivalves (Fedyakov & Naumov, 1989), and 19% of the prosobranch gastropods (Golikov, 1989). Among the whales, there are two monotypic genera that belong to the plesiomorphic family Monodontidae. These are the white whale (Delphinapterus) and the narwhale (Monodon). The narwhale is strictly Arctic while the white whale ranges southward into the northern parts of the cold-temperate North Pacific and North Atlantic (Jefferson et al., 1993). In comparison, the Antarctic Region possesses many more endemic species and numerous endemic genera. The differences are attributable to the 25 Myr history of the cold Antarctic biota as opposed to less than 3.0 Myr for the Arctic.

In addition to having a great influence on the biota of the Arctic and the North Atlantic, the North Pacific also had far-reaching effects on the southern hemisphere, the Antarctic and the deep sea. The cold-temperate waters of the North Pacific extend from the Arctic boundaries in the Bering Sea to southern California on the east coast and to about Wenchou, China and northern Japan on the west coast. That huge area may be subdivided into two regions and five provinces (Briggs, 1995). The rich marine biota of the North Pacific probably developed over a period of c. 40 Myr. In an evolutionary sense, this biota was an inheritance from warm-temperate and eurythermic tropical species that managed to adapt to the colder temperatures. Its high level of species diversity, compared with that of the North Atlantic, is the result of its larger size and more moderate temperature fluctuations. It has a larger heat budget and during the glacial stages it was protected from the influence of the Arctic Ocean by the Bering land barrier.

The North Pacific Centre has contributed a variety of organisms to the cold-temperate southern hemisphere. The migrations took place primarily via isothermic submergence, whereby the species concerned could maintain a suitable temperature by moving beneath the tropics at great depth. Most cases of interhemispheric, isothermic submersion involved populations that have become separated long enough to become distinct at the species or generic level. For example, the fish genus Sebastes (family Scorpaenidae) is extraordinarily diverse in the North Pacific, being represented there by almost 100 species. This genus is represented along the Chilean coast, Tierra del Fuego, the Falkland Islands, Tristan da Cunha and the tip of South Africa (Eschmeyer & Hureau, 1971). A recent molecular study has determined that a total of three species may now be recognized (Rocha-Olivares et al., 2001). All three apparently stem from a single migration that took place through the Eastern Pacific (EP) within the past 200 ka.

As noted in the Antarctic account, two other fish families of North Pacific origin are represented by numerous species around the Antarctic continent, the Liparididae and the Zoarcidae. Both are capable of penetrating deep water and probably migrated using isothermic submergence. The Liparididae has an interesting distributional history that has been worked out by Andriashev (1986). The shallow water liparidids of the North Atlantic evidently reached that area during the trans-Arctic biotic interchange. However, members of the genus Paraliparis, a deep water group, made their way to the Antarctic along the west coast of the Americas. From the Antarctic, the genus dispersed northward along the mid-Atlantic ridge and thence to the Arctic Basin. As a result, the liparidid fauna of the Arctic–North Atlantic owes its origin to two migratory groups, the shallow water genera come directly through the Arctic Ocean while the deep water paraliparids migrated all the way to the Antarctic via the EP, then reached the Arctic Basin via the Atlantic.

The Zoarcidae probably made its way south in the same manner as Paraliprais. It has speciated extensively in Antarctic and Sub-Antarctic waters (Anderson, 1988). A different tropical submersion route has been suggested for the fish family Cottidae. The southern hemisphere species of this family belong to a distinct genus (Antipodocottus), while the family itself is undoubtedly of North Pacific origin. It has been determined that Antipodocottus is most closely related to Atopocottus tribranchius, a Japanese species (Nelson, 1985). Nelson indicated agreement with Bolin (1952) that the invasion route was probably from Japan southward by way of the Philippines, New Guinea and the New Hebrides.

The large brown algae or kelps of the Order Laminariales belong to four families, all of them found in the North Pacific. Three of the families are large and each of these has representatives in the southern oceans. Estes & Steinberg (1988) concluded that the centre of origin for the Order was in the North Pacific and that the southern species must have reached their present ranges via dispersal. Four of the southern species belong to the genus Laminaria. This genus also exists in the North Atlantic and has been taken in deep water off Brazil. So its route south may have been through the Western Atlantic (WA). However, two other southern genera (Macrocystis and Ecklonia) do not exist in the North Atlantic, so they probably reached the southern hemisphere by migrating along the west coast of the Americas.

Among the invertebrates, the crab genus Cancer was determined to have originated in the North Pacific, spread to the North Atlantic, and then reached the southern hemisphere by tropical submersion (Nations, 1979). Later, a molecular study of the genus suggested that the trans-Arctic invasion took place 6–12 Ma and that the dispersal to New Zealand occurred c. 6 Ma (Harrison & Crespi, 1999). Smith (1970), who studied the cymatiid gastropods, concluded that the genera Fusitriton, Argobuccinum and Aforia dispersed southward via tropical submergence. In his study of the evolution of high latitude molluscan faunas, Crame (1996) identified twenty-seven cases of bipolar or bitemperate distributions. In several of the examples, the genera concerned had evidently originated in the North Pacific.

In general, it may be observed that the predominant interhemispheric dispersals in the Pacific Ocean have taken place from north to south. This, together with the strong North Pacific influence in the Arctic–Atlantic, indicates that the rich fauna and flora of the former has produced many dominant species that have been able to transgress biogeographical boundaries and establish themselves elsewhere. At the same time, the North Pacific biota has resisted penetration of species from other areas. This means that the North Pacific Ocean has been functioning as an important centre of origin. It has had a profound influence on the composition of the marine ecosystems in the Arctic, North Atlantic, temperate southern ocean and the Antarctic. It has also been a significant source of species that continue to inhabit the deep sea (Zezina, 1997).

The tropical centres

The tropical biota generally occupies the lower latitudes, but the shape of the Tropical Zone is affected by the flow of the major oceanic currents. On the western sides of the Pacific and Atlantic oceans, the north and south equatorial currents turn towards higher latitudes and bring with them warm water and tropical organisms. On the eastern sides of the two oceans, the major currents transport cool water towards the tropics. This allows the tropics to occupy a broad latitudinal area to the west but only a relatively narrow one to the east. As there is a positive relationship between species diversity and geographical area, it means that the diversity of the western regions will exceed that of the eastern ones. The northern Indian Ocean is entirely tropical, so when its diversity is added to that of the western Pacific, to recognize an Indo-West Pacific (IWP) Region, we find more species than the other three tropical regions combined.

The shallow marine biota on the continental shelves and in the upper 200 m of the pelagic environment also exhibits a greater diversity than the equivalent biota at higher latitudes. The difference is the result of the latitudinal gradient in diversity, generally attributed to the greater solar radiation received by the earth at lower latitudes. A contributing factor may be that the tropics of today have a longer history than the cooler zones. During the Mesozoic and early Cenozoic, the tropics occupied the greater part of the globe except for warm-temperate waters at the highest latitudes. The tropics did not become significantly reduced in area until the global temperature decline that began in the mid-Eocene.

Although diversity is of compelling interest to many, others find the widespread evolutionary effects of the tropics to be even more fascinating. Jablonski et al. (1983) published a significant article that drew attention to onshore–offshore patterns in the evolution of Phanerozoic shelf communities. As evidence, they presented data on faunal changes that took place in the Cambrian and Ordovician periods of the early Palaeozoic and the late Cretaceous of the Mesozoic. Their analysis indicated that major new community types appeared first in nearshore settings and then expanded into offshore settings. This occurred despite higher rates of species evolution in the offshore habitats. By the end of the Ordovician, the three major evolutionary faunas of the Phanerozoic were arrayed in distinct community associations across the continental shelf and slope. The remnants of the Cambrian fauna were on the slope, the Palaeozoic fauna was on the mid to outer shelf, and the early members of the modern fauna on the inner shelf. A parallel pattern of change was found for the late Cretaceous.

More evidence of an onshore to offshore replacement was provided by Sepkoski & Miller (1985) and Bambach (1986). Then Jablonski & Bottjer (1990a) utilized additional data on echinoderms to substantiate earlier conclusions about onshore origins and recognized high diversity tropical settings as sources of innovation. The same authors (Jablonski & Bottjer, 1990b) also provided results from their study of forty post-Palaeozoic benthic invertebrate groups. They found that, while the higher taxa (orders) tended to first appear onshore, the lower taxa (species, genera, families) were diversity-dependent. Their conclusion that ‘diversity begets diversity’ had an important evolutionary connotation. Additional evidence of onshore to offshore replacements was reviewed by Jackson & McKinney (1990) who observed that the long-term persistence of certain clades in shallow water has been accompanied by major shifts in morphology. Clades that did not undergo such changes became restricted to offshore or cryptic habitats.

Over time, the onshore to offshore replacement sequence has had a cumulative effect in the deep sea. Although the cold-water centres in the Antarctic and North Pacific have made significant contributions to the slope and abyssal faunas, so have the tropics. There are many families of deep-sea animals that live only at low latitudes. It has been suggested that the faunistic centres in the Western Pacific, the Western Indian and the WA oceans may be contributing directly to the deep sea (Zezina, 1997).

For may years, there had been a continuous argument about the evolutionary significance of the tropics. Did the tropics have high origination rates permitting them to act as a diversity pump, or did they have low extinction rates allowing them to accumulate diversity? Jablonski (1993) re-examined the data on the geographical origin of the marine orders and found that the great majority had their earliest occurrence in waters that were tropical at the time of fossil deposition. In this way, he provided direct evidence that the tropical regions had been a major source of evolutionary novelty and not simply a refuge that accumulate diversity owing to low extinction rates.

It is also important to note that elements of the tropical biotas have been able to extend their ranges by means of latitudinal dispersal. For example, Flessa & Jablonski (1996) had determined that tropical bivalve faunas were significantly younger than non-tropical ones. In a more detailed study, Crame (2000) concluded that the steep latitudinal gradients of the youngest bivalve clades provided additional evidence that tropics have served as a major source of evolutionary diversification. As a result, the tropics were seen as a species pool (or a series of pools) supplying bivalve taxa to the high latitude and polar regions. It has been suggested (Flessa & Jablonski, 1996) that higher tropical diversity reflected a higher rate of tropical origination. But, when mean rates of radiation were compared between tropical and temperate bivalve clades (Crame & Clarke, 1997), no significant difference could be found.

The tropics can also give rise to extinction patterns that may result in the establishment of disjunct populations that are antitropical or antiequatorial. The presence of such disjuncts has been recognized as the first step in an extinction process that may have an extended history (Briggs, 1995). As antitropical species become older, the separated populations may evolve into distinct species or genera. Some are apparently able to escape extinction by adapting to the cooler temperatures of higher latitudes. In this way, bitemperate or even bipolar distributions may become established.

Newman & Foster (1987; see Briggs, 1995) examined fossil and recent evidence for the historical biogeography of barnacle genera. Some were shown to be antitropical and others to be southern hemisphere endemics. Two of the southern endemics were represented by fossils in the northern hemisphere, so they were considered to be former antitropicals that had become extinct in the north. The older three of the five recent antitropical genera were represented only in populations of the North Atlantic. The two younger genera had populations in both the North Atlantic and the North Pacific. This led Newman and Foster to suggest that a sequence might lead from a broad antitropical distribution to extinction in the North Pacific, followed by extinction in the North Atlantic. This would leave a southern hemisphere population heading towards eventual termination.

Such accumulations of plesiomorphic relicts in the temperate parts of the southern hemisphere are a notable phenomenon in both terrestrial and marine habitats. However, one may find an interesting contrast between different parts of the Southern Ocean. Vermeij (1986) has noted that the warm-temperate marine faunas of New Zealand and Australia contain many relict invertebrate taxa that were common in western Europe during the Eocene and the Oligocene. He also observed that many ancient tropical taxa had extended their ranges into the same area. Among the fishes, there has occurred a notable penetration of IWP tropical groups into the cold-temperate areas of Tasmania–Victoria and New Zealand. Included are the families Labridae, Syngnathidae, Tripterygiidae, Gobiidae and Gobiesocidae (Briggs, 1995).

In addition to the presence of large numbers of relicts, the distributional patterns in the Southern Ocean are unique in another sense. In the New Zealand–Australian area most of the relicts appear to be of tropical ancestry. But this is not true for the temperate waters of South America. Here, there is clear evidence of a large-scale invasion from the North Pacific and, to some extent, from the North Atlantic. As noted, many of the invaders have also become established in the Antarctic and Sub-Antarctic. In addition, there is a large suite of organisms that show patterns influenced by the circumglobal West Wind Drift. Some of the latter were apparently picked up from South America by the current and carried eastward.

Over the long-term, the world has apparently been dominated by phyletic lines that have originated in the tropics. But it would be a mistake to assume that this had continuously been the case. In Phanerozoic history there have been five, relatively brief times when the marine environment was subjected to major extinctions. These occurred at the ends of the Permian, Ordovician, Devonian, Triassic and Cretaceous. The most severe extinction occurred at the end of the Permian when about half of all marine families (Sepkoski, 1986) and more than 95% of the species (Raup, 1979) disappeared. The other four extinction events were not so destructive, but in each case, half or more of the marine species were lost. These great disasters to diversity are often said to have occurred suddenly but actually took place over periods of c. 1–10 Myr (Briggs, 1995).

The great extinctions have been attributed to various causes such as comet impacts, volcanic eruptions, acid rain, sea level transgressions and sea level regressions. It seems that the only readily identifiable common cause is the latter and it is perhaps significant that the greatest regression, at the end of the Permian, occurred at the time of the greatest extinction (Hallam, 1992). From an evolutionary standpoint, it is important to note that the great extinctions preferentially eliminated the tropical shelf and epipelagic organisms. As McGhee (1990) stated, ‘One of the most pervasive effects during times of biotic crisis is the massive disruption of tropical and low-latitude ecosystems, and the relative non-disturbance of high-latitude and polar ecosystems’.

The heavy tropical extinctions created an ecological vacuum that was gradually occupied by simpler, plesiomorphic organisms from higher latitudes, deep water and other refuges. This process of succession by older, more primitive forms amounted to a setback of the tropical evolutionary clock to an earlier time. Furthermore, the damage to the tropical centres of origin must have interrupted, for some time, the production of successful new species and genera. The time span of the interruptions can be estimated by reference to the time it has taken the environment to recover from the extinction events. In the end-Permian event, all of the existing reef communities completely collapsed. It took c. 8 Myr until such communities began to re-establish themselves (Fagerstrom, 1987). A review of evidence for the recovery times from all five of the major extinctions indicated a range of 2–9 Myr (Erwin, 1998).

Although palaeontological investigations have clearly shown that the tropics have been a major source of evolutionary innovation, the question of location within the tropics has remained. The tropical seas cover an enormous area. Have the various parts differed in terms of evolutionary importance? As the work of fossils progressed, marine biologists were working on the relationships of the contemporary marine biota. The superior diversity of the IWP, in numbers of species, genera and families attracted considerable attention. Was this region an important source of living marine organisms or were they being produced everywhere?

The east indies centre

The biogeographical patterns that are currently displayed in the IWP have recently been reviewed (Briggs, 1999a, 2000). Despite a basic homogeneity caused by the occurrence of many wide ranging species, there are great differences in species diversity in the various parts of the IWP. The majority of tropical marine families have their greatest concentration of species within a comparatively small triangle formed by the Philippines, the Malay Peninsula and New Guinea. As one leaves this East Indies Triangle, there is a notable decrease in species diversity that is correlated with distance. In general, the decrease westward through the Indian Ocean is less than that found for the other directions. To the north and south there is a gradual decrease towards the limits of the tropics. Moving eastward, one does not encounter significant local endemism until the extremely isolated islands of the central and Eastern Pacific.

Another pattern of interest within the IWP is that of generic age. Work on marine groups that have left good fossil records (corals, barnacles, crinoids and ostracods) showed that the average age of the genera increased with distance from the East Indies. A third pattern is that of dispersal tracks. By working out the evolutionary relationships within certain families, some authors have inferred historical routes from the East Indies. This has been done in order to illustrate the migrations of fishes and family of molluscs. A fourth pattern deals with phylogenetics. In the past, some researchers have maintained that, as a given phyletic line expands into new territory, relatively advanced (apomorphic) species will evolve at the periphery. Others were convinced that the more apomorphic species were characteristic of the centre and that their more primitive (plesiomorphic) relatives were displaced towards the periphery. In the IWP, the latter pattern has so far been consistently displayed.

Pattern number five is that created by extinctions. When many IWP disjunct patterns representing various shelf species are examined, it may be noted that in almost all cases a distributional gap appears in the East Indies. The most common type of disjunction demonstrates an antitropical or antiequatorial pattern where the relict populations exist to the north and south of the central tropics but not within them. But there are a few cases that show east–west disjunctions. Various theories have been proposed to explain these patterns. They have been examined with the result that a tropical extinction hypothesis seems to fit most of the evidence (Briggs, 1987, 1999b). The drama of extinction often makes its debut in the East Indies, gradually spreads outward and plays itself out among small populations far from their centre of origin.

The sixth pattern is that shown by the level of genetic diversity within individual species or clades. Data from mitochondrial DNA in sea urchins demonstrated a decreasing gradient in genetic diversity that paralleled the decline in species diversity across the Pacific (Palumbi, 1997). Studies of the mtDNA from clams, shrimps and sea turtles provided genetic indications that the general East Indies was the place of origin for their interoceanic radiations. It seems evident that these six biogeographical patterns provide support for the hypothesis that the East Indies has been operating as a centre of evolutionary radiation. There still remains the question of why the evolutionary centre should be located in the East Indies instead of some other part of the tropics.

Fortunately, there are palaeontological data that help to clarify the history of the East Indies centre. Kay (1990) found that the earliest cowries (Cypraeidae) occurred in the Palaeocene of Pakistan and India. Wilson & Rosen (1998) traced the development of reef corals and found very little development in the East Indies until the early Miocene. The earliest coral reef fish assemblage has been found in the Eocene rocks of Monte Bolca in Italy (Bellwood, 1996). In their study of fossil molluscs, Piccoli et al. (1987) noted that, in the Tethys Sea, the area of greatest species diversity extended from Europe and North Africa to India. Beginning in the mid-Eocene, there was a substantial increase in the molluscan diversity of Java. By the late Miocene, the diversity centre in the East Indies was evidently well developed.

It seems clear that the late Mesozoic and early Tertiary location of the tropical diversity centre was in that part of the Tethys Sea that extended from Europe and North Africa to India. Two principal events were probably responsible for the eastward migration of the diversity centre. First, in the mid-Eocene, the global temperature began to deteriorate. Secondly, the early Miocene collision between Africa and Eurasia eliminated the Tethys and established the Mediterranean. The tropical biota that was trapped in the Mediterranean gradually disappeared as the climate got colder. A contributing factor was probably the continuing fusion of India to Asia, with the formation of the Arabian Sea and the Bay of Bengal. The northern parts of these two seas, with mud and sand bottoms, do not support coral reef communities, although the infaunal diversities may be high.

Two factors, one involved with the speciation process itself and the other with its geographical location, help explain why the East Indies continues to play a major role in evolutionary innovation. First, it has been proposed (Briggs, 2000) that the centrifugal speciation model, published by Brown (1957), is applicable to centres of origin. It predicts that ultimately successful species are produced by large, central populations. In contrast, Mayr's (1954) peripatric theory would have the successful species arise from small, peripheral populations with the consequence that places of maximum species diversity would only represent accumulations in a favourable habitat. This should be considered a centripetal, as opposed to the centrifugal, process (Brown, 1987).

The importance of geographical location was emphasized in an early work on centres of origin (Briggs, 1966). It was observed that dominant, advanced species apparently came from certain favourable centres and because we know that speciation also takes place in areas peripheral to such centres, we should recognize that two kinds of initial evolutionary change are taking place – one that may be successful in terms of a phyletic future and another that is generally unsuccessful. Over the short term, successful evolution may be defined as the process that takes place when a new species expands from its place of origin and establishes itself in new territories. Unsuccessful evolution takes place when a new species remains more or less confined to its place of origin and apparently exists in an evolutionary trap. Each of the two factors would appear to be important. The production of successful wide ranging species indicates the presence of a centre of origin and the centrifugal process provides a mechanism of dispersal from that centre.

In regard to the IWP itself, one may conclude that the East Indies Triangle appears to be the place of origin for a series of dynamic systems that appear to extend across the entire ocean. These systems are apparently maintained by a continuous outflow from the East Indies. The best known of these is the diversity gradient, but others are equally important. There is a gradient in average generic age in which the age increases with the distance from the East Indies, dispersal tracks of individual animal groups extend outwards, some phylogenetic patterns indicate that the East Indian species tend to be the more advanced (apomorphic), extinction patterns appear to originate in the East Indies, and a gradient of lessening genetic diversity extends outwards.

A continuous flow of species replacement from the East Indies over the past 10 Myr would create and maintain the systems that have been identified. The fact that these six systems exist and operate as they do makes sense only within the context of the centre of origin hypothesis. Successful species that are produced in the centre must be able to dominate peripheral relatives that are less fit in a genetic sense.

One might also take the larger view and consider that the influence of the East Indies extends well beyond the bounds of the IWP. Two principal barriers separate the IWP from the other tropical shelf regions. These are the deep-water East Pacific Barrier that lies between Polynesia and the New World, and the Old World Land Barrier comprising of Africa and Eurasia. A variety of IWP species have crossed the former to establish themselves in the EP. These include fifty-five fishes (Leis, 1984) and sixty-one gastropods (Emerson, 1991). Conversely, there is almost no evidence of successful migrations in the opposite direction, despite the fact that the north equatorial current has been shown to carry larval stages from the EP towards the IWP (Scheltema, 1988).

Similarly, the other boundaries of the IWP appear to function essentially as one-way filters. A few tropical shore species have been able to migrate around the Cape of Good Hope to establish themselves in the Atlantic, but there seems to have been no traffic from the Atlantic to the IWP (Briggs, 1995). Since the opening of the Suez Canal in 1869, more than 200 species of Red Sea organisms have established themselves in the Mediterranean, but less than a dozen have taken the reverse course (Por, 1978). When the biota of the other three tropical regions [EP, Eastern Atlantic (EA) and WA] is examined, the relationship to the IWP seems obvious. Most of the families and a large fraction of the genera are shared among all four. And, in most cases, the families and genera are best developed in the IWP.

Of course, the families and some of the older genera predate the establishment of the East Indies as a centre of origin some 10 Ma. The beginnings of such older tropical taxa probably took place in the Tethys Sea that was formed in the early Cretaceous when the northern continents finally became completely separated from Gondwana. At first, the tethyan biota was fairly homogeneous but, by the late Cretaceous, the North Atlantic had become considerably wider and the western end of the Tethys began to develop a significant endemism. By Aptian time, c. 124 Ma, a distinct Caribbean Province became apparent. By c. 86 Ma, the West Central American and Antillean subprovinces developed (Kauffman, 1979). In the meantime, the remainder of the Tethys had become divided into Eastern and Western Mediterranean and North Indian Ocean subprovinces. As tectonic plates shifted and ocean currents responded, the various Tethys subdivisions became the four contemporary tropical regions.

The western atlantic centre

The WA Region evolved from the Antillean Subprovince. In the late Cretaceous, the biota of the Subprovince was separated from that of the West Central American Subprovince by a Central American isthmus that may have lasted until the early Palaeocene. The presence of this marine barrier is indicated by its separation of the two subprovinces (Kauffman, 1979). At the same time, the isthmus served as a terrestrial connection between North and South America allowing the dispersal of freshwater fishes, frogs, lizards, snakes, dinosaurs and early mammals (Briggs, 1995). The isthmus may have been created from a previous archipelago by the dramatic fall in sea level that took place near the end of the Cretaceous. The submersion of the early isthmus in the Palaeocene allowed the subsequent intermixing of the marine subprovincial biotas. A common New World tropical biota then developed to remain in place until divided by the present Panama Isthmus c. 3.1 Ma (Hallam, 1994).

The Pliocene cooling event(s) of 2.4–2.9 Ma, that eliminated the cold-temperate fauna of the Arctic Region, also had a detrimental effect on the molluscan fauna of the tropical Atlantic. Stanley (1986) concluded that all of the endemic tropical species along the coast of the south-eastern Untied States were lost, leaving only eurythermic forms at those latitudes. The molluscan data indicates a general extinction of a WA tropical biota that once extended farther to the north. The Pliocene cooling was followed by Pleistocene glaciations that also had a severe effect in the Atlantic, more so than in the Pacific. However, the temperature fluctuations also appeared to stimulate speciation; so many of the losses, at least in the WA, were replaced.

It is the EA Region that has the poorest tropical biota. Its species diversity is far less than might be expected judging from its geographical size. A count of the species of fishes and of some of the invertebrate groups (echinoderms, molluscs, stomatopods, brachyurans, reef coral) indicated only about half the number found in the EP Region (Briggs, 1985). The EA shelf appears to occupy c. 400,000 km while the EP shelf covers c. 380,000 km. So theoretically, the two regions should have diversities that are close to being equal.

Another unusual aspect of the EA Region is its relationship with two of the other regions, particularly the tropical WA. An examination of the EA inshore fish fauna has revealed large numbers of trans-Atlantic species or forms that are closely related to widespread WA species. Most such species belong to well-developed WA genera; none belong to typically EA genera (Briggs, 1995). Works on West African invertebrate groups also indicate appreciable numbers of trans-Atlantic species. Finally, there is a group of about twenty-four species of inshore fishes that have apparently invaded the EA Region by rounding the Cape of Good Hope. Such relationships suggest that the EA Region is still in the process of recovering from an extinction. The relatively narrow latitudinal area of the tropical EA may have been considerably reduced during the Pleistocene glacial stages. If so, the reduction of space could have resulted in multiple extinctions.

Considering that many species, having evolved in the WA, have proved capable of migrating eastward across the Mid-Atlantic Barrier to colonize the EA, and that very few species originating in the EA are apparently capable of successfully invading the western side, it appears that the WA has been operating as a centre of origin. Its effectiveness as a centre has been limited to the Atlantic Ocean and, compared with the other three centres, it has been in existence for a brief period of time. If we consider the East Indies to be the primary centre of origin for the marine tropics, then the WA should be designated a secondary centre of origin.

Summary

Antarctic

An early warm-temperate Antarctic zone probably became cold-temperate about the time of the initial appearance of the first continental ice sheets, some 42 Ma. The temperature decline continued until c. 25 Ma when cold temperatures of c. +2 to −2 °C were reached. The new low temperatures, plus the tectonic drift of the Australian continent, produced an isolation that has resulted in the world's most distinctive marine biota.

The external influence of the Antarctic Region, with respect to the surface waters, has been generally limited. Many species produced in the region have managed to migrate to the surrounding cold-temperate waters of the Sub-Antarctic. Some of these are plesiomorphic, when compared with their Antarctic relatives, indicating that they have been displaced outward. A few species such as some penguins, fishes and molluscs have extended their ranges much farther. Two seaweed species have migrated all the way to the Arctic.

The greatest external influence has been in the deep sea. Ice pack formation leaves behind cold, saline oxygenated water that sinks around the edges of the continent. This vertical flow is converted to deep horizontal currents that reach the major ocean basins and trenches. There is a voluminous literature (mainly in Russian) attesting to the fact that a variety of deep benthic organisms were introduced to that habitat via thermohaline currents originating around Antarctica.

The highly productive waters around Antarctica have proved to be attractive to a variety of carnivorous birds, mammals, fishes and invertebrates. A few, such as some brachiopods, gastropods and fishes, are phylogenetic relicts that have found a safe place for survival. Others represent apomorphic groups that have invaded from northern seas, especially the North Pacific. So the Antarctic clearly serves a dual role. It is a centre of origin that donates to other areas but, at the same time, it is a refuge for relicts and opportunistic invaders.

North Pacific

In the late Cretaceous, c. 80 Ma, connection between North America and Asia took place that separated the warm-temperate biotas of the North Pacific and the Arctic–Atlantic. Beginning in about the mid-Miocene, the sea surface temperatures of the high latitudes declined, the North Pacific warm-temperate organisms were extinguished or were forced southward and a new cold-temperate (boreal) biota began to develop. After some 40 Myr, there are indications that the land barrier to the Arctic Ocean started to become inundated. This may have occurred as early as 12 Ma. At first, the Bering Strait may have been very shallow and intermittent. By c. 3.5 Ma, it appears that the passage was unrestricted and the Great Trans-Arctic Biotic Interchange took place.

The opening of the Strait eventually permitted a continuous biotic flow between the North Pacific and the Arctic–North Atlantic. However, the North Pacific biota, that was more diverse to begin with, had a much greater success as invaders than did the species from the opposite direction. The result was a profound change in the composition of the ecosystem of the Arctic–North Atlantic but very little alteration in the north Pacific. Between 2.9 and 2.4 Ma, a substantial cooling, accompanied by a major glaciation, took place. The cold-temperate biotas were eliminated or forced southward from the Arctic and from the northern parts of the two adjoining oceans. In the North Pacific and in the North Atlantic, new cold-temperate regions were established, one on each side of each ocean.

A new cold water biota evolved in the Arctic Region. It may now be distinguished by a significant endemism at the species level. There have also arisen a number of eurythermic arctic–boreal species that exist in both cold and cold-temperate waters. Aside from their influence in the Arctic–North Atlantic, North Pacific families have exported considerable numbers of species to the southern hemisphere. Apparently, most of the species that extend far to the south bypassed the tropics by means of isothermic submergence. Many of them emerged from shallower depths in southern cold-temperate waters but others remain in the deep sea. Some evolved to become important components of the Antarctic and Sub-Antarctic ecosystems.

Tropics

Tropical seas, with their characteristic biotas, have been a continuous presence since the early Palaeozoic. During glacial stages, when global icehouse conditions prevailed, the tropics were squeezed into lower latitudes, somewhat more than they are today. But for most of the Phanerozoic, they occupied the major part of the world ocean. There are two important ways in which the tropics have controlled the global evolution and distribution of marine organisms. First, and perhaps most important, the tropics have been a major source of evolutionary novelty. New faunas, identified by the appearance of new orders of marine animals, appear in onshore habitats, become established for millions of years and are then displaced to offshore habitats.

A result of the historic onshore to offshore competitive replacements, in addition to extinctions, is the survival of many ancient, phylogenetic relicts on the lower slopes and the upper part of the abyssal zone. These include primitive molluscs, crinoids, starfish, crabs, sponges and many other groups whose ancestors once inhabited the continental shelves in Mesozoic and Palaeozoic times. It has also been observed that the long-term persistence of certain clades in shallow water has been accompanied by major shifts in morphology. Clades that did not undergo such changes became restricted to offshore or cryptic habitats.

The second way in which the tropics have controlled the development of the marine world is by the establishment of regions that possess high levels of species, generic and family diversity. As has been pointed out, the origin of species, genera and families is evidently diversity-dependent, i.e. diversity begets diversity. Currently, there is only one major centre of diversity for these taxonomic levels and that is the East Indies Triangle! As noted, the Triangle is a centre of origin that produces successful species. It is those species that survive, as the fossil record indicates, to give rise to phyletic lines leading to new genera and families. As one can tell, from the close biotic relationships of the IWP to the other tropical regions, it is relatively easy for species of probable East Indies origin to penetrate other tropical waters. In fact, many such species have achieved circumtropical distributions.

The third method of tropical influence is by means of latitudinal dispersal. Everywhere in the world the tropics are bordered by a warm-temperate biota. The warm-temperate regions and provinces are defined by their many endemic species, but also have numerous species called eurythermic tropicals. The latter are not only wide ranging in the tropics but also occupy the warm-temperate regions. The tropical relationships of most of the warm-temperate endemic species are obvious because they usually belong to tropical genera and families. In some places, such as southern Japan and the north-eastern United States, the northward migration of tropical species in the summer tends to obscure the presence of the resident warm-temperate forms.

In most parts of the world there is a distinct contrast between the warm-temperate and cold-temperate biotas. The cold-temperate organisms usually belong to high latitude families that are likely to be older and more plesiomorphic than their warm water counterparts. This general observation is consistent with palaeontological results that show considerable increases in evolutionary age from the tropics towards the polar regions. Also many of the cold-temperate families and genera show phylogenetic patterns that suggest origins from high latitude sources. This is especially true for the northern hemisphere where the influence of the North Pacific Centre is very strong. The cold-temperate and the warm-temperate waters of the southern hemisphere are of considerable evolutionary interest because they harbour many relict families and genera that were once widespread in the tropics. Some of the relicts are also found in the north temperate zones, thus having antitropcial distributions.

In the cold-temperate areas of Tasmania–Victoria and New Zealand there has occurred a notable penetration of IWP tropical groups including many invertebrate and fish families. The temperate species and genera belonging to these families appear to be phylogenetic relicts that have accumulated over a long period of time. On the other hand, in southern South America tropical families are scarce and those of North Pacific origin are prominent. It appears that the cold-temperate fauna of the southern hemisphere has accumulated from three principal sources: (1) the Antarctic centre of origin, (2) invasion of boreal forms into southern South America and Antarctica-Sub Antarctica, and (3) invasion of tropical groups into Australia–New Zealand. Many of the invasive species show patterns that were subsequently influenced by the circumglobal West Wind Drift.

Conclusion

The hypothesis states that living organisms in the sea are mobile parts of a dynamic system that is being powered by evolutionary engines called centres of origin. Successful species are continually moving out from such centres. As they do so, they replace other species that are less competitive or less fit in a genetic sense. The replacement process is generally slow and, as a given species expands its range, it is likely to evolve. Species, as they become older, may give rise to distinct genera and the genera themselves will age as they disperse. Over time, these changes will result in gradients of increasing age and other significant patterns that extend outward from centres of origin.

The oldest and most predominant of the centres is the one that has, for the past 10 Myr, been located in the tropical East Indies. During that time, it has been the source of a series of biogeographical patterns that have been created and maintained in the IWP by a continuous flow moving outward from the centre. The individual patterns are illustrated by geographical changes in species diversity, generic age, dispersal tracks, phylogenetic progression, extinction occurrences and genetic flow. Beyond the bounds of the IWP, dominant species, apparently from the East Indies, have crossed the deep water East Pacific Barrier to the New World, rounded the Cape of Good Hope to reach the Atlantic and passed through the Suez Canal to colonize the Mediterranean.

Although it has been easy for dominant species from the IWP to invade other tropical regions, they have also moved north and south into the adjoining warm-temperate zones. That is why those zones demonstrate a strong tropical relationship. In a few places, mainly in the southern hemisphere, members of some tropical families have been able to occupy cold-temperate waters. These longitudinal and latitudinal invasions are apparently current and have probably been going on for the past 10 Myr. The migratory traffic through the Suez Canal has occurred only since 1869. Before the Miocene, and since the early Mesozoic, the tropical centre of origin was located in the Tethys Sea. Apparently, taxonomic orders were continually being formed in tropical onshore habitats and lower taxa were evolving in places of maximum diversity, both onshore and offshore. When the original groups were succeeded by more apomorphic ones, the earlier biotas became extinct or were displaced horizontally or vertically to deeper waters.

In Palaeozoic times, other tropical centres of origin probably served as evolutionary engines just as the East Indies does today. That far in the past, tropical centres may have been even more important, for the tropics usually occupied a greater portion of the globe, and, at times, there may have been more than one tropical centre. Now, we have one major one (East Indies) and one minor one (WA).

The North Pacific Ocean is the location of the single centre of origin that supplies most of the cold-temperate world. It was formed c. 40 Ma but did not achieve its present influence until the opening of the Bering Strait permitted the Great Trans-Arctic Biotic Interchange. The invasion into the Arctic–North Atlantic produced enormous changes in the ecosystems of that area. The relatively few invaders in the North Pacific produced only minor changes. The North Pacific centre has also exerted a strong influence on the cold-temperate and polar waters of the southern hemisphere. Most of the southward invasions were performed by organisms that were capable of bypassing the tropics by means of submersion to deeper and colder waters.

The Antarctic presents a curious combination of evolutionary centre, refuge for phylogenetic relicts and a haven for invaders from the North Pacific. As a centre of origin, it supplies species to the Sub-Antarctic and adjacent cold-temperate habitats. But its greatest external influence is in the deep sea. Thermohaline currents originating from the shelf provide a conduit to the abyssal and hadal (trench) areas in various parts of the world.

The engines that keep the global marine system in motion are the three major centres of origin: the East Indies, the North Pacific and the Antarctic. Their power apparently comes from the constant production of successful species. As such species spread out and evolve, they leave tracks that can be recognized as gradations in diversity, age, phylogeny, genetic variations and extinction.

As many recent systematic works have demonstrated, widespread geographical patterns, consisting of a mosaic of closely related species or subspecies, can often be traced to the previous existence of a broadly distributed ancestral species. The presence of such a mosaic, with its evidence of many allopatric speciation events, has led many investigators to conclude that this mode of speciation has been of predominant evolutionary importance. Yet such conclusions do not usually take into consideration the place of origin of the ancestor or the reasons for its widespread success. In centres of origin (and elsewhere), the sympatric and parapatric modes are apt to be of considerable importance.

The evolutionary engine hypothesis does not predict that most speciation is concentrated in centres of origin. In fact, the enormous areas of lower diversity that exist outside the centres certainly produce more species than do the centres. Also, speciation in small, isolated populations is likely to proceed at a faster than normal rate. The difference is that successful species tend to be produced in the centres rather than elsewhere. All newly formed species do not have equal potentials. Those that are produced in the diversity centres by large populations with high levels of genetic variation are the ones that disperse outward to keep the system going.

Acknowledgments

I am indebted to E. J. Reitz, Georgia Natural History Museum, University of Georgia, for facility support.

Biosketch

John C. Briggs retired from the University of South Florida in 1990. Since that time he has been an Adjunct Professor in the Georgia Museum of Natural History at the University of Georgia. His interests are in evolutionary and historical biogeography and in the systematics of fishes. Professor Briggs' biogeographic books are: Marine Zoogeography (McGraw-Hill, 1974), Biogeography and Plate Tectonics (Elsevier, 1987) and Global Biogeography (Elsevier, 1995). He is currently involved, along with other biogeographers, in the writing of a book to be entitled Foundations of Biogeography. It will be published by the University of Chicago Press.

Ancillary

Advertisement