Among amphibians, Australia has (or had) at least two species of frogs that brood young in their stomachs. Instead of the colubrid snakes and vipers that are abundant elsewhere, Australia has had a radiation of elapid snakes (cobras and their relatives) into 75 species, all of them virulently poisonous. The largest Australian predators are the saltie crocodiles, which lurk along northern rivers and shorelines, and large monitor lizards related to the Komodo dragon of Indonesia. The monitors are ambush predators, the largest being 2 meters (over 6 feet) long. Smaller monitors dig for prey like the badgers of larger continents. In contrast, most Australian mammals are herbivores.
Extinct Australian reptiles include giant horned tortoises (Figure 19.1), a monitor lizard more than 5 meters (16 feet) long that competed with large terrestrial crocodiles, and a boa constrictor 7 meters (23 feet) long. Extinct Australian birds include Dromornis, the heaviest bird that has ever evolved.
Australia is the only continent with living monotremes. They have been in Australia since the Early Cretaceous (Chapter 14), and only one early Cenozoic monotreme tooth from southern Argentina shows that they once ranged more widely over Gondwanaland. The surviving monotremes are egg-laying mammals, including the duckbilled platypus and the echidna of Australia and New Guinea. Many aspects of monotreme biology are bizarre: for example, the platypus swims in muddy water with its eyes, ears, and nostrils tightly shut, searching for its crustacean prey with electrical sensors in its beak. Since monotremes have evolved to include the specialized platypus and ant-eating echidnas, it's likely that their fossil record will eventually show us many other surprises.
Marsupials now have been discovered in Eocene rocks in Antarctica and Australia, so it's likely they had reached Australia from South America when the south polar region was much warmer than it is now, well before the refrigeration of Antarctica in Oligocene times. It was surprising to find in 1992 that a mysterious primitive placental mammal had also reached Australia by Eocene times. Presumably, Tingamarra had also walked across Antarctica to reach Australia and drop one of its right upper molars into a billabong in Queensland, but that's all the evidence we have right now. Tingamarra's tooth is enough to show that marsupials did not evolve to dominate Australian mammalian faunas because they were somehow safe from placental competition there.
By sometime in the Late Cenozoic, marsupials had evolved to fill most of the ecological roles in Australia that are performed by placental mammals on other continents (Figure 19.2). Wallabies and kangaroos are grazers comparable with antelope and deer, wombats are large burrowing "rodents" rather like marmots, and koalas are slow-moving browsers like sloths. The cuscus is like a lemur, and the numbat is a marsupial anteater. There are marsupial cats, marsupial moles, and marsupial mice, and at least six gliding marsupials can be compared with flying squirrels. The honey possum Tarsipes is the only nonflying mammal that lives entirely on nectar and pollen, which it gathers with a furry tongue. The small marsupial Dactylopsila of New Guinea has evolved specialized teeth and a very long finger to become a marsupial woodpecker (Chapter 18). The Tasmanian wolf and the Tasmanian devil are marsupial carnivores comparable in size and ecology to wolf and wolverine. They once ranged over the main continent of Australia. The Tasmanian devil is now confined to Tasmania, and the Tasmanian wolf is probably extinct.
The fossil record of extinct Australian marsupials is even more impressive. Many forms were very large, including giant kangaroos, giant tree kangaroos, and giant wombats. Whole families of marsupials are now extinct. Thylacoleo was a Pleistocene carnivore whose name means the marsupial lion. It was indeed large, the size of a lioness, and had efficient stabbing and cutting teeth. It was better adapted for cutting off chunks of flesh than any living carnivore is (Figure 19.3). Diprotodonts were large quadrupedal Pleistocene marsupials about the size of tapirs and rhinoceroses (Figure 19.4). The largest diprotodont was the size of a small elephant, almost 3 meters (10 feet) long and 2 meters (over 6 feet) high at the shoulder. Recent discoveries of enormous numbers of bats and marsupials in Miocene rocks at Riversleigh, in Queensland, will soon allow a better description of the radiation of Australian mammals.
People often talk of marsupials as primitive and inferior to placentals, and it's true that today they are outclassed in diversity and range by placentals. But marsupials do not always have inferior adaptations, as we have already seen in Chapter 14. For example, a kangaroo is rather clumsy as it hops slowly around on the ground, using its tail as an extra limb in what is really a five-footed movement. It does use more energy than a placental at this speed. But at high speed a kangaroo is not only very fast (up to 60 kph, or 40 mph), but its incredibly long leaps are much more efficient than the full stride of a four-footed runner of the same weight.
Marsupials must have reached Australia from North America, where they first evolved (Figure 19.5). There were many successful marsupials in South America throughout the Cenozoic, while there are no living or fossil marsupials in Asia and only one early fossil marsupial in North Africa that is probably a stray refugee from Europe. Marsupials most likely reached Australia by a southerly route via South America, presumably across Antarctica. If so, they must have crossed before the refrigeration of Antarctica, and the process must have excluded almost all placental mammals from Australia and all mammals from New Zealand (except bats, which can fly). The recent discovery of Eocene marsupials in Antarctica and Australia adds support to this idea. Placentals were rather limited in South American faunas at the time, which may help to explain why only one nonflying placental, Tingamarra, reached Australia.
Dromornithids (mihirung in aboriginal legend) are giant extinct Australian birds that must have evolved flightlessness and large body size (Figure 19.6). Dromornis was probably as large as Aepyornis, the elephant bird of Madagascar, and rivals it for the heaviest bird of all time. The living Australasian emu and cassowary are large ground-running rat-ites. Ratites have been much discussed because they are distributed on the southern continents that are remnants of Gondwanaland. We do not know yet whether the cassowary, emu, ostrich, and rhea are related in a clade, all descended from ancient ratites that lived in Gondwanaland, or whether they are descended from separate groups that each found their way south into the southern continents).
The isolated position of Australia has meant that only very mobile birds and placental mammals (bats and humans) have reached it. Humans brought with them a host of other invaders, such as rats, cats, dogs, sheep, cattle, rabbits, cactus, fish, and cane frogs, with serious results for the Australian ecosystem. Captain Cook's first reaction to a kangaroo was to set his dog on it! More recently, other bizarre introductions have helped to restore a little of the damage‹for example, the organism that causes the rabbit disease myxomatosis, and the dung beetles that keep Australian grasslands from being buried in cattle dung. The biogeographic story of Australia is still in an active phase.
The dominant prehistoric creatures of New Zealand were birds. The kiwis survive as nocturnal insectivores, but the major vegetarians were very large ratites, at least a dozen species of moas. The largest moa was 3.5 meters (11 feet) in height (Figure 13.24). Moas coevolved with New Zealand plants so that 10% of the native woody plants have a peculiar branching pattern called divarication‹they branch at a high angle to form a densely growing plant with interlaced branches that are difficult to pull out or break. There are few leaves on the outside, and the largest, most succulent leaves are on the inside. But nine species of divaricating plants that grow more than 3 meters (10 feet) tall look more like normal trees once they reach that height, and other divaricating species grow more normally on small offshore islands. The only reasonable explanation of divarication is that it evolved as a defense against browsing moas, the largest of which was about 3 meters tall.
Other vegetarian guilds that were filled by small mammals on other land masses were partly occupied by moas and other birds and partly by huge flightless insects‹enormous weevils and wetas (giant grasshoppers). It's not easy to identify the major prehistoric predators, but they were present. The largest surviving New Zealand birds‹the kiwi, for example‹are well camouflaged, although there is no obvious surviving predator on them. But extinct New Zealand raptors include a bird that was the largest goshawk that ever evolved (3 kg or 7 pounds in weight) and a huge extinct eagle that weighed about 13 kg (30 pounds).
Apart from these brief periods of immigration, Cenozoic evolution in South America took place in isolation for over 60 m.y. The strange South American mammals in particular are well known, and they divided up available ecological roles in the usual way. Charles Darwin noticed some peculiar fossil mammals in Argentina during his voyage on the Beagle, and later expeditions to Argentina have found hundreds of beautifully preserved Cenozoic fossils.
From Early Cenozoic times South America had quite a number of marsupials, and these took on the roles of small insectivores (and still do). There is a living aquatic marsupial with webbed feet and a watertight pouch. Argyrolagus was a rabbit-sized marsupial that looked like a giant kangaroo rat. It hopped and had ever-growing molars for grazing coarse vegetation. The arrival of later placental rodents did not affect these small marsupials. One of the most successful marsupials in the world, even in the face of intense competition from placentals, is the small omnivorous opossum, Didelphis.
Early placental ungulates had reached South America in the earliest Paleocene, and by the Miocene they had evolved into a bewildering variety of forms ranging from rhino-sized to rabbit-sized. Thoatherium and Diadiaphorus (Figure 19.8) had an uncanny resemblance to horses, with long faces, horselike front teeth, grinding molars, straight backs, and slender legs ending in one or three toes. Some of their relatives looked like camels. Large vegetarians such as Toxodon had large grinding molars that grew through most of the life of the animal (Figure 19.9).
The ancestors of armadillos, sloths, and anteaters also arrived in South America very early. Armadillos and their relatives evolved heavy body armor for protection and became highly successful opportunistic insectivores and scavengers. The Pleistocene armadillo Glyptodon was very large, probably a vegetarian, 1.5 meters (5 feet) long. It had a thick armored skullcap as well as body armor, and some glyptodont species had a spiked knob at the end of the tail (Figure 19.10). Glyptodonts were certainly too big to burrow like the smaller armadillos, and they had to be heavily armored and armed to survive out on the surface. Naturally, their skeleton was very strong to support all the weight of the armor.
Sloths now live in trees, eating leaves and moving with painful slowness. But remains of huge ground sloths have been found in South America, including one that must have been almost as large as an elephant. Anteaters evolved from the same group of ancestors but are now specialized to an amazing extent for eating termites, beginning by tearing apart their nests with tremendously powerful clawed forearms.
Perhaps the most impressive South American creatures were the larger carnivores. None of them were placental mammals, and most were marsupials. This is not surprising, considering how savage the surviving little marsupial insectivores are, but it is unusual compared with other continents. Borhyaenids were basically like wolves, but were generally larger. Proborhyaena was as big as a bear and probably had a similar way of life. Borhyaena itself was a wolf-sized Miocene marsupial with canine teeth adapted for stabbing and molars that had evolved into meat-slicing teeth (Figure 19.11). It was obviously a successful medium-sized carnivore, but it was the last of the large borhyaenid carnivores. They were replaced by invading placentals from the north and by gigantic predatory birds.
Thylacosmilids were most like large cats. Thylacosmilus was a marsupial sabertooth, but its savage stabbing canines were better designed than those of the placental sabertooth cats of North America. In Thylacosmilus the sabers were longer, slimmer, more securely anchored in huge, recessed tooth cavities extending far up the face; thus, they were better protected from damage than those of true cats (Figure 19.12). The sabers were ever-growing and self-sharpening, and they were backed by more powerful neck and head muscles. Presumably they were adapted to killing large (placental) ungulates by stabbing and slashing deep into the soft tissues of throat or belly. The cheek teeth were not as powerful as those of placental cats, however.
These amazing marsupials had unusual competitors for mastery of the carnivorous guild, the phorusrhacids: flightless, ostrich-sized birds equipped with very powerful tearing beaks as well as foot talons. It seems that the phorusrhacids eventually gained the upper hand over the carnivorous marsupials.
South America had its own group of crocodiles, the sebecids. They apparently evolved in Gondwanaland in the Cretaceous, survived the KT extinction, and radiated in the Early Cenozoic in South America to become powerful terrestrial predators. Unlike aquatic crocodiles, they had high, deep skulls and snouts. Other crocodilians in South America also evolved into unusual morphologies; for example, a duckbilled caiman is known from the Miocene of Colombia.
The South American ecosystem gained new immigrants in Oligocene times, around 25 Ma, with the arrival of rodents and primates, probably from Africa by way of islands in the widening Atlantic Ocean (Figure 19.7). Both groups radiated widely, the rodents into the various guinea pigs and their relatives, and the primates into the distinctive New World monkeys, evolving habits and characters in parallel with Old World gibbons and monkeys.
Other members of the Cenozoic South American fauna included more giant predators, the largest flying birds of all time, the teratorns (Chapter 13). The largest turtle of all time, Stupendemys, lived along the north coast.
This unique ecosystem suffered four tremendous shocks in ten million years and has almost completely disappeared. First, Antarctica froze up, with the result that the Humboldt Current, flowing most of the way up the west coast of South America, became much colder and stronger. Second, tectonic activity along the Pacific coast raised the Andes as a major mountain chain. Together, these two events drastically lowered rainfall over most of the continent, and much of the area turned from forest and well-watered plain to dry steppe. This led, in the later Miocene, to the extinction of many animals, including the terrestrial crocodiles and especially the large-bodied savanna herbivores.
Third, South America drifted northward towards Central and North America (Figure 19.7). By about 6 Ma, the gap was small enough to allow a few animals to cross it, more or less by accident. North American raccoons and some mice and rats crossed to the south, while two kinds of sloths crossed to the north. The effect of the competition was seen immediately. Many borhyaenids were replaced by raccoons, and the largest of them, the bearlike Proborhyaena, was replaced by a bear-sized raccoon. Finally, at about 3 Ma, the last important sea barrier was bridged, and animals could walk from one continent to the other.
Ecological principles suggest what should happen when an exchange of animals takes place. A larger continent such as North America should have a larger diversity of animals than its smaller counterpart, and the fossil record confirms that this was true just before the exchange. Therefore, if the same proportion of animals from each continent migrated to the other, one would expect more North American animals to go south than the reverse. If a continent can hold only so many families or genera of animals, then one would predict extinctions on each continent, but more in South America than in North America. The effect would be accentuated because North America was at least intermittently connected with Eurasia, and altogether this huge northern area of temperate open country held a great variety of savanna animals. In contrast, the area of savanna in South America was not as large as one would think, because the continent is widest in equatorial latitudes and narrows significantly to the north and south. South American savanna faunas might have been very vulnerable to invasion from the north.
The major exchange happened after 3 Ma. Camels, elephants, bears, deer, peccaries, horses, tapirs, skunks, rabbits, cats, dogs, kangaroo rats, and shrews entered South America. Monkeys, opossums, anteaters, sloths, armadillos, capybaras, toxodonts, porcupines, and glyptodonts migrated north, with the giant birds‹a phorusrhacid and a few teratorns (Figure 19.13).
The South American immigrants to North America flourished there, and so did the successful North American immigrants that moved south. Overall, however, there was a net major extinction of South American groups (Figure 19.14). The large, native marsupial carnivores and most of the phorusrhacids seem to have been outcompeted by the cats and dogs from the North, and the remaining savanna browsers and grazers were almost all wiped out, perhaps outcompeted by the northern horses and camels, perhaps hunted out by the new predators. The sabertooth marsupial Thylacosmilus was replaced by a real sabertooth cat. Even earlier invaders suffered: the bear-sized raccoon was replaced by a true bear.
The geographical changes that had permitted the interchange also changed the climate of the Atlantic Ocean, and this in turn caused drastic changes in the land ecology of North and South America as the northern ice ages began in earnest around 2.5 Ma. Ice advances and retreats also caused great swings of vegetation patterns on time scales of 100,000 years, stressing the faunas in a way that had not happened before. Overall, the result was nearly as expected, and can be explained by current ecological theory, with the South American animals coming out much the losers. North American invaders survive in strength today in South America, including all the South American cats, the llamas, and dozens of rodents.
South American faunas suffered a fourth catastrophic extinction in the late Pleistocene. This time, similar extinctions took place in North America too, and we shall examine this in Chapter 23.
Diamond, J. M. 1990. Biological effects of ghosts. Nature 345: 769770. The prehistoric ecology of New Zealand.
Griffiths, M. 1988. The platypus. Scientific American 258 (5): 8491.
Marshall, L. G. 1988. Land mammals and the Great American Interchange. American Scientist 76: 380-388.
Marshall, L. G. 1994. The terror birds of South America. Scientific American 270 (2): 9095.
Rich, P. V., and G. F. van Tets (eds.). 1985. Kadimakara. Victoria, Australia: Pioneer Design Studio. The extinct animals of Australia.
Stehli, F. G., and S. D. Webb (eds.). 1985. The Great American Biotic Interchange. New York: Plenum.
Webb, S. D. 1991. Ecogeography and the Great American Interchange. Paleobiology 17: 266280.