THE LAST GREAT EXTINCTION

SLOTHS today, both species of them, are weird but modest creatures that hang upside down in trees in South America, eating leaves, descending to the ground at intervals of several weeks to defecate, and perfecting the arts of lethargy. But until about 10,000 years ago - a mere blink of the eye in the timescale of evolution - they had some fine and formidable relatives. Among the scores of species of extinct ground sloths, some were taller than an elephant and probably stood on their hind legs, balanced by muscular tails like a kangaroo, and slashed at the vegetation with wicked claws. Earlier this year scientists found evidence of yet another sloth, Thalassocnus, that lived on the coast of Peru. It was as big as an Irish wolfhound and apparently swam in the sea like a kind of xenarthran sea-beaver (Xenarthra being the group of mammals to which all sloths belong). The surviving sloths give hardly a hint of their glorious past.

What is true of sloths applies to South Amer-ican mammals as a whole, and indeed to animals worldwide. Huge numbers of large species have disappeared in the past few million years, especially over the past few millennia. It is a phenomenon on which Darwin felt moved to comment during his survey of the Argentine coast in 1833: ''It is impossible to reflect on the changed state of the American continent without the deepest astonishment. Formerly it must have swarmed with great monsters; now we find mere pygmies. What, then, has exterminated so many species and whole genera?''

Only now are biologists discovering the true extent of the losses in South American fauna and unravelling the probable reasons - reasons of enormous significance for modern conservation. They show that animals are even more vulnerable than they seem, and expose Rousseau's myth of the noble savage, a myth that primitive man was at one with nature. The all-destructiveness of modern humans which now threatens half the species in the world, merely continues the tradition of our hunting-gathering ancestors who appear to be responsible for much of the mass extinctions of prehistory.

The story of the sloths of South America has its roots in the splitting up of the ancient southern continent of Gondwana. This resulted in the creation of the South American land mass which remained an island for about 40 million years before colliding with the North American land mass. It was this 40 million years of isolation that resulted in the unique and varied animal species of South America, including the richly diverse group of sloths. It acquired a collection of monkeys before it broke from Africa, which are thriving still. It had its own marsupials, some like bears and some like sabretooth cats. It had four orders of hoofed animals unrelated to the horses, deer, camels and elephants of other continents, yet evolving along similar lines to produce such features as hooves and quasi-elephantine trunks. Uniquely, too, it had the xenarthrans which, besides the sloths, in-clude the anteaters, armadilloes and the extinct tortoise-like glyptodonts with a ''shell'' as round as a golf-ball and as big as a bread van. The discovery of a maritime sloth shows that the xen-arthrans were even more versatile than anyone had previously suspected.

These weird and wonderful native species of South America are almost all gone, as extinct as the dinosaurs which disappeared more than 60 million years previously. The great mammals of South America vanished, it seems, in two main waves. The first mass extinction followed soon after South America met up with North Amer-ica in the Pliocene epoch, around three- and-a-half million years ago. The second occurred after about 13,000 years ago, with the arrival of human beings in the Americas.

During the first wave of extinctions, most of the unique marsupials and hoofed animals were replaced by creatures that flooded in from the north, via the new-formed isthmus of Panama. Those northern newcomers included the immediate ancestors of the largest South American animals that survive today, like the llama and vicuna, the tapir, the jaguar and puma, and the spectacled bear. They also included a host of creatures that have become extinct in their turn: one-toed and three-toed horses, elephant- like mastodons and even true elephants, and the formidable Arctodus - a bear half as big again as a modern grizzly but with a short face for seizing and long legs for running; a kind of ursine Rottweiler.

The replacement of southern creatures by northerners was dramatic, and most biologists assumed the northerners were more agile, more versatile, more streetwise. But Elisabeth Vrba, a zoologist at Yale University, has formulated a powerful argument which supposes instead that the first mass extinctions of South American mammals was brought about simply by a change of climate and hence of landscape. She believes that, at the time the two continents collided, the world was growing cooler and drier - a prelude to the ice ages. So, maintains Dr Vrba, animals migrated towards heat. The equator, the hottest place, is at the level of the Amazon, so that is where they went. If the northern continent had not met the southern, some northerners might have died out in Panama. As things turned out, they could press on to Brazil.

The southerners, by contrast, were already at the equator when the climate began to grow harsher. They had nowhere to run and nothing to gain by fleeing. Yet their native forest did not escape unscathed; it retreated and opened as the climate grew colder and drier. Northern immigrants were adapted to such open country, as it was the same as that of the north in former times. As the climate changed, they could stay in the kind of environment they were used to just by changing latitude; southerners were mostly obliged to stay put and died out as conditions changed.

Many southerners did survive the invasion, however, including some of the native hoofed animals: one shaped like a heavyweight camel and another like a hornless rhinoceros. Some even managed to migrate north, including giant ground sloths and glyptodonts which flourished for a time in the present United States, plus armadilloes and marsupial opossums which still abound. No signs of inferiority there.

But most large mammals that survived the Pliocene upsets - "large" means more than 100lbs - have disappeared in both Americas. Thus, until at least 100,000 years ago, late in the epoch known as the Pleistocene, North Amer- ica had no fewer than 45 genera of large mammals, each of which may include several or even dozens of related species. These included animals that no longer exist like mammoths and mastodonts, plus others that survive in Eurasia and Africa, like camels, horses, the yak and the cheetah. This impressive catalogue is now reduced to 12; including the world's largest deer, moose and wapiti, and bison, musk-ox and a few bears. In South America the destruction has been even greater. In the late Pleistocene, the native southern survivors plus the northern invaders between them spanned an astonishing 58 genera; almost as great a variety of large mammals as now exist over the entire Earth. Now they too are down to 12: the llama, tapir, jaguar, spectacled bear, and a few others.

What lay behind this second round of losses which occurred over the past few thousand years, long after the invasion of the northern species? The most likely cause, according to Professor Paul Martin of the University of Arizona, was the arrival of humans. Ancestors of the present-day native Americans crossed from Siberia to Alaska around 13,000 years ago and spread south during the next few thousand years. Species after species of large mammal vanished as they went. Prof Martin calls this time of devastation "the Pleistocene overkill".

Overall, the evidence for Pleistocene overkill is circumstantial. The direct fossil evidence for human butchering is patchy, but what else could have caused such a wipe-out? Climate has again been indicted, but if this were the cause we would expect small species to suffer more than large, for they are more sensitive. Yet in the late Pleistocene, the extinctions are of larger animals. But the most persuasive of past human destructiveness comes from fossils coming to light around the world. These show that, whenever human beings invaded new territory, great numbers of native animals disappeared soon afterwards - and always the larger species bore the brunt. The pattern is repeated in Australia, Madagascar, Mauritius, Hawaii, New Zealand and a host of smaller islands in all the oceans.

If Dr Vrba is right, then events in Pliocene America show how vital it is for animals to be able to move towards and away from the equator as the climate changes. Nowadays we confine wild animals to reserves, so when the climate changes next time they will die like rats in traps; no more mass excursions, north and south. If Prof Martin is right, then the events of the Americas and Australia and of islands worldwide show how easy it is to drive animals to extinction. You don't have to kill every one. You merely have to hunt the animals more quickly than they can reproduce, and big animals breed slowly. Blue whales, Asian eleph-ants, at least four of the five remaining species of rhino, pandas, tigers, several bears, orang-utans and many others - all may be snuffed out in passing, just like their Pleistocene forebears.

More broadly, it is clear that humans do not need modern technology to be destructive. We made our presence felt with fire and stone axes. Rousseau got it wrong partly because he did not know enough zoology. He was impressed by the faunas of exotic lands, but didn't appreciate what devastation had already taken place. Indeed, the true extent of the destruction - and its recency - is only now coming to light. !