The Norway rat, one of the few mammals as successful as humans, steps off the boat in Polynesia, circa 1767.

46

THREE

INTO THE PRESENT

Oxford is the odd twin of Cambridge, the slightly less illustrious sibling, a bit of a ne�er-do-well compared with its slightly more senior fraternal twin. Immensely illustrious certainly, old, rich, and smart. But not Cambridge. A geologist will note other differences immediately. While Cambridge sits upon the chalk of the Cretaceous, Oxford lies in the opposite direction from London, toward older rocks. Its buildings are made of yellow and tan Jurassic sandstone, limestone, and oolite (a delectable geologic term used to describe a particular grainy limestone). And for reasons unknown, it has bred or attracted a batch of evolutionists quite different from the Cambridge mix. Richard Dawkins and Robert May are there. But the most iconoclastic may be Norman Myers, a conservationist turned futurist who sees and fears the worst not only for the future of biodiversity, but for the future of evolution itself, in the upcoming years. Myers has been the most vocal prophet crying that the end of biodiversity � at least as we know it � is nigh.
Has the Earth indeed entered a new mass extinction event, or is such an event nearly over? The first of these two contentions was radical even as late as the 1980s, but in the earliest part of the twenty-first century it seems accepted as fact. (The second, that the most consequential phase of this extinction, at least for large animals, is over, is still new scientific territory.) Numerous articles and a succession of books have all treated this subject in detail. Yet Myers was there first � even arguing that the loss of the Pleistocene megamammals is connected to the modern-day biodiversity crisis. According to this hypothesis, the extinction of so many large

47

animals (mainly mammals and birds) over the past 50,000 years was only the beginning of a larger wave of extinction continuing into the present and for some unknown period into the future as well.
Myers believes that a new phase of this mass extinction � a broadscale reduction in biodiversity � has been under way since about 1950, when a major increase in human encroachment into wildlife environments began. At that time approximately 1.7 billion people lived in the so-called developing countries, located in large tropical and semitropical regions characterized by vast forests and other undisturbed habitats for wildlife. The human population of these regions was approaching 5 billion people by the turn of the millennium. Myers maintains that the forces producing deforestation, desertification, soil erosion and loss, inefficient agriculture, poor land use, inadequate technology, and above all, grinding poverty are driving habitat destruction and, ultimately, extinctions of species, and that these forces are most pronounced in developing equatorial countries. He estimates that 50% of the world�s species will go extinct over the next several centuries at most. While this estimate may sound drastic, it is in line with those of other biodiversity experts, including E. O. Wilson�s 1992 estimate that 20% of all species will go extinct before 2020 and another 30% or more thereafter; Peter Raven�s 1990 calculation that 50% of all species on Earth will be extinct by the year 2100, and Paul and Anne Ehrlich�s 1992 estimate that 50% of all species will be extinct by 2050.
All of the aforementioned seers take the position that the majority of the modern mass extinction is soon to occur (but has not yet happened). But how accurate is this view? Where are the figures on current extinction rates to support this claim?

Measuring Species Diversity

Determining rates of species loss seems straightforward: tabulate the number of species living at a given period of time, and compare that number to the number living at other time intervals. Yet there are numerous problems with this seemingly simple methodology. To arrive at extinction numbers, we need an accurate census of the living. Such a global census of biodiversity at the species level is still lacking. No one disputes that the activities of humankind have caused extinctions in the recent and not so recent past. The phrase �dead as a dodo� is not pure whimsy. But there is currently great debate about the extent of anthropogenic extinctions, and even more about the prospects for such extinctions in the future. Ultimately, the entire issue devolves into numbers. But the numbers we need are very difficult to

48

One hundred years ago the vast Amazon rainforest was a virtually pristine ecosystem. Today there is almost no portion of it that has not been touched by humans.

49

obtain: How many species are there on Earth? How many have there been at various times in the past? How many species have gone extinct in the last millennium, the last century, or even the last decade or year? And, most important of all, how many will be gone in the next century, or millennium, or million years? None of these numbers is directly obtainable; all have to be reached, if at all, by abstraction, inference, deduction, or just plain guesswork. In contrast to the estimates above, some scientists wonder whether the loss of species will approach even 10% of current world diversity, and suggest that such a small loss would hardly be noticed.
Why is there any controversy at all about how many species are presently on Earth? In this day and age, when modern science can detect planets among stars light-years away and can deduce the age of the universe from the movement and activity of subatomic particles, what could be simpler than counting up the number of species on Earth and then, over a twenty-year time period, for instance, observing how many are going extinct? Such an endeavor would require a large army of biologists, many more than the small handful actually engaged in this type of research. In reality, we have only the haziest idea of how many species currently exist on Earth, how many there have been in the past, and how many are going extinct at any given time. It is our lack of the most basic and necessary information � the current number of species presently living on Earth � that is the cause of the greatest dissention.
Of the Earth�s 1.6 million currently described creatures, about 750,000 are insects, 250,000 are plants, 123,000 are arthropods other than insects, 50,000 are mollusks, and 41,000 are vertebrates; the remainder is made up of various invertebrate animals, bacteria, protists, fungi, and viruses. The majority of organisms leave no fossil record.
The precise figure for the world�s biodiversity is not known. There is no central registry for the names of organisms, and because of this, many species have been named several times. Taxonomist Nigel Stork believes that the level of synonymy may approach 20%. For example, the common �ten-spotted ladybird� found in Europe has forty different scientific names, even though it represents but a single species. Such mistakes may seem easily avoidable, but many species exhibit a wide range of variation, and the more extreme examples of a given species are often mistakenly described as new or separate species.
Does this mean that the number of species on Earth today is less than the currently defined 1.6 million? Probably not. Most biologists studying biodiversity suspect that there are far more, but an intense debate rages about exactly how many more. The most extreme estimates are in the range of 30 to 50 million species,

50

meaning that taxonomists have named just over 3% of the species on Earth, and thus have barely begun their work in the 250 years or so since Linnaeus set out the task of describing every species. Other, more cautious souls posit much lower numbers, between 5 and 1 5 million species. Yet even with this lower number it is clear that the work of describing the Earth�s biota has a long way to go.
The accuracy of species counts varies from group to group. For some groups, such as birds and large vertebrates, our census is nearly complete; there will be very few new discoveries of new species. Yet for the majority of invertebrate groups, and for the legions of one -celled organisms such as protozoans, bacteria, and other microbes, there are surely millions of forms yet undescribed.
It is clear that scientists will never succeed in describing every species (however wonderful that would be). Nevertheless, there is a pressing need to establish a reasonable estimate of world biodiversity. Is it closer to 1.6 million or 50 million? How can a more reliable estimate be established without describing every living species?
There have been several ingenious attempts to arrive at a reasonable estimate of the number of species living on Earth. As far back as the 1800s British zoologists knew that insects are the single most diverse group of animals on Earth and tried to census world insect diversity, coming up with an estimate of 20,000 species. It is now known that at least that many insect species are found in Britain alone. How, then, to make a more accurate accounting of the world�s species? The favored method today is to use the ratio of known to unknown species in taxonomic groups that have been long studied and are considered essentially well known (such as birds and mammals) to estimate total world biodiversity. Botanist Peter Raven used this method in 1980 to suggest that world biodiversity is about 3 million species. Specialists on insect diversity have been particularly adept at coming up with new and clever ways of making such estimates. Nigel Stork and his colleague K. G. Gaston noted that of 22,000 insect species known in Britain, 67 are butterflies. Assuming that the ratio of butterflies to other insect species is the same in the rest of the world (an assumption utterly untested, but plausible), they arrived at a global biodiversity estimate of 4.9 to 6.6 million species of insects alone.
A second method of arriving at a global biodiversity estimate is to extrapolate from samples. Samples from a particular geographic area, rather than a taxonomic group, are scaled upward to encompass the entire biosphere. It was this method that yielded the most famous of all recent biodiversity estimates, that of Smithsonian entomologist Terry Erwin published in 1982, which posited that there are at

51

least 30 million species of beetles in the world�s tropical forests. This particular estimate has been so widely quoted (and misquoted) that it bears further scrutiny.
Erwin�s concern at the time was not global diversity, but simply that the upper reaches of the rainforest canopy were little known and rarely sampled by taxonomists. Most known organisms from rainforests came from the forest floor. The multistoried canopies of large trees were known to harbor a quite different fauna, yet because of the difficulties of sampling this environment, its inhabitants were poorly known. Erwin devised a new method of sampling the upper canopy regions. Over three field seasons he poisoned and collected all of the insects living in nineteen trees belonging to a single tree species (Luehea seemannii, a forest evergreen). He found that there were an average of 163 species of beetles specific to that particular tree species. So far, well and good. At this point, however, Erwin set out a number of assumptions to arrive at his famous 30 million beetle estimate. First, he assumed that each hectare of forest in his sample area contained, on average, 70 different tree species. He then assumed that each of these tree species also had its own 163 species of beetles � a veritable army of beetles specific to that tree living in its canopy. By multiplying these numbers, he arrived at an estimate of 11,140 tree-specific beetle species in one hectare of forest in Panama, and then added in another 1,038 species of beetles just passing through the trees, to arrive at a figure of 12,448 beetle species per hectare of forest. Next, he assumed that beetles made up 40% of the total arthropod fauna in the canopy, so that the entire biodiversity of insects in his one hectare was 31,120 species of arthropods. He then added another third of this total to his estimate to take into account the insects found on the forest floor, and arrived at a grand total of 41,389 arthropod species per hectare of Panamanian forest. The final step � scaling up from a hectare of Panamanian forest floor to the entire world � was accomplished in the following way. Erwin noted that there are 50,000 species of trees in the tropics. Assuming his figure of 163 host-specific beetle species per tree, he arrived at his much-repeated estimate of 30 million species of beetles in the world.
Erwin�s �thought experiment� was simple and elegant, but full of untested assumptions. However, since it was based (at least at the start) on real sampling in regions that up until that time were virtually unknown, it took on a life of its own and was treated quite seriously. It is still the basis for the larger biodiversity estimates cited today.
The Erwin estimates were widely publicized, and rightly so. They gave us a whole new view of global biodiversity. But because of the way in which they were

52

produced, they were immediately controversial. New surveys that attempted to confirm or falsify these new, higher estimates of global biodiversity soon followed. One of the most thorough was carried out in Indonesia.
Project Wallace was a yearlong collaboration between scientists of the Natural History Museum in London and the Indonesia Department of Science. Some two hundred net-waving, jar-toting entomologists descended on the island of Sulawesi. Many insects were collected, including more than 6,000 species of beetles and almost 1,700 different species of flying insects, more than 60% of which were new to science. These scientists, applying methods similar to Erwin�s, estimated a worldwide biodiversity of between 1.8 and 2.6 million insects. Since insects are only one part of total biodiversity (albeit the single most important one), these new estimates confirm that world biodiversity in indeed far higher than the 1.6 million species currently described. On the other hand, even with so many insect species, world biodiversity would still fall well short of the 30 million species predicted by the Erwin estimate.
Yet another type of estimate was derived by noted biologist Robert May, who in 1988 pointed out that the observed correlation between body size and species diversity could be used to arrive at a rough estimate of world biodiversity. Using such a method, May estimated that the Earth contains between 10 million and 50 million species, a figure that seems to support the Erwin estimates.

Genetic Losses

One of the great surprises of the mid-1960s to 1970s was the discovery that species � virtually all species � are characterized by far higher amounts of genetic variability than previously supposed. The then-nascent techniques of gel electrophoresis and DNA sequencing allowed geneticists to evaluate just how different individuals of the same species were. While everyone knew that genomes � the number and type of genes � varied tremendously from species to species, no one foresaw the great genetic variability that characterizes virtually every living species. Every organism carries a large number of genes: a bacterium typically carries about 1,000 genes, a mushroom about 10,000, and typical higher plants and animals as many as 50,000 to 400,000. It is variability among these genes that differentiates the various species on Earth, today as in the past. But there remains a great deal of variability within each species as well, which creates the various �races�, sub-races, and populations that make up a species. This variability appears to be of utmost service to species, for it provides a hedge against sudden changes in the environment: in highly variable populations, there will probably be at least a few

53

individuals that are �preadapted� for whatever new conditions come along, thus allowing the survival of the race. Any reduction in genetic variability is thus dangerous to a species � and appears to be a sure indicator of a species sliding toward extinction. Prior to the final extinction of a species, we see a dying off of its populations caused by a reduction in the overall genetic variation.
In the mid-1970s, a study of twenty-four proteins extracted from North American elephant seals revealed that this rare and highly endangered species shows essentially no genetic variation. This particular species had been hunted nearly to extinction, and even though there has been a rebound in the population since it was protected from further hunting, its overall genetic makeup was severely affected. The elephant seal population is said to have passed through a �population bottleneck�. At its low point, each seal seeking a mate had only a very small number of other seals to choose from, resulting in severe inbreeding and a loss of genetic variation. Such unions between close relatives are often characterized by high rates of birth defects, retardation, and reduced sperm counts. Inbreeding is so deleterious that humans of every culture have produced laws against it.
An extreme example of such genetic loss is the Florida panther. This subspecies of the American cougar has been reduced to fewer than thirty individuals in the wild. Low sperm counts and damaged sperm characterize the remaining males. Genetic studies show that this subspecies has the lowest genetic variation of any of the extant cougar populations.

Estimates of Current Extinction Rates

It can be argued that the current mass extinction is far less calamitous than either the end-Paleozoic or end-Mesozoic events because a lower percentage of families and genera are going extinct now than in the past. The severity of a given extinction event is commonly tabulated as the percentage of existing taxonomic units, be they families, genera, or species, that go extinct. Using this measure, it has been argued that the extinctions that have occurred since the onset of the Ice Age have been trivial compared with the great extinctions of the Paleozoic and Mesozoic eras because the percentage of taxa that have gone extinct is but a tiny fraction of the total diversity of the Earth. What is being overlooked, however, is the fact that the absolute � not relative � number of species (or other category) that have already gone extinct in the last million years may be substantial. For instance, biologists Stors Olson and Helen James have published data suggesting that as many as a thousand species of birds have disappeared from the Earth in the last two to five millennia.

54

This represents perhaps 20% of the total bird biota on the planet. These are species that have left a fossil record, and thus species we know about. How many more have gone extinct without leaving a trace?
Extinction is the ultimate fate of every species. Just as an individual is born, lives out a time on Earth, and then dies, a species comes into existence through a speciation process, exists for a given span of years (usually counted in the millions), and then eventually becomes extinct. Thus, extinctions of species happen all the time, not just during mass extinction events. University of Chicago paleontologist David Raup calls this concept background extinction. The fossil record can be used to tabulate the rate of such �random� extinctions taking place throughout time, and that rate turns out to be remarkably low. Raup has calculated that the background extinction rate during the last 500 million years has been about one species every four to five years. In contrast, Norman Myers has estimated that four species per day have been going extinct in Brazil alone over the past thirty-five years. Biologist Paul Ehrlich has suggested that by the end of the twentieth century, extinction rates were measurable in species per hour.
If attaining a reliable estimate of global species diversity has caused problems, estimates of current extinction rates have been no less controversial. While many different people disagree strongly on the number of species on Earth, and on the rate at which these species are currently declining in number, on one issue there is no disagreement: the vast majority of species currently living on Earth are found in the tropics, mainly in rainforests.
Tropical rainforests are characterized by a high canopy, often 30-40 meters above the ground with emergent trees towering to 50 meters, and two or three separate understories of vegetation. They are complex, layered communities with enormously varied and changing environments and microclimates.
Tropical rainforests today are found in three principal regions. The most extensive is the American, or Neotropical, rainforest region, centered in the Amazon Basin but extending up the Caribbean slope of Central America to southern Mexico. The Neotropical rainforest comprises about half the global areal total, and about one-sixth of the area of all broad-leaved forests in the world. The second large block occurs in the eastern tropics and is centered in the Malay Peninsula. The third is in central Africa.
Norman Myers estimates that between 76,000 and 92,000 square kilometers of tropical forest are lost each year to logging and field clearing, and that an additional 100,000 kilometers are grossly disrupted. This means that about 1% of the

55

world�s tropical forests are disappearing each year, a rate that will lead to the complete disappearance of all tropical forests in one century, if current practices continue. Biologist E. O. Wilson, in The Diversity of Life, estimated the rate of tropical forest loss in 1989 to be 1.8% per year. The Food and Agricultural Organization (FAO) of the United Nations officially placed the deforestation rate at 0.5% per year in the late 1980s.
Daniel Simberloff of Florida State University analyzed all available information regarding the rate of forest destruction � data mainly derived from satellite imagery and remote sensing. He found that the tropical forests of Asia are already virtually gone. There are currently about 92,000 recognized plant species (and an unknown number of plant species waiting to be described by science) in the New World tropical rainforests, and 704 species of birds. Simberloff has calculated that between 1950 and 2000, almost 14,000 plant species (15% of the total) and 86 bird species (12% of the total) became extinct in this region. If the tropical forests of the New World become restricted to current and planned reserves and national parks, Simberloff predicts that the extinction of over 60,000 plant species (66%) and 487 bird species (69%) will occur between 2050 and 2100 A.D. Simberloff concludes that �the imminent catastrophe in tropical forests is commensurate with all the great mass extinctions except for that at the end of the Permian�.

Recent Losses and Causes

Since 1600, a minimum of 113 species of birds and 83 species of mammals are known to have gone extinct. But these animals are large vertebrates, which through time have had a far lower background extinction rate than 5 per year. About three-quarters of these extinctions took place on oceanic islands. Historical records also suggest that, since 1600, extinction rates for these two groups have increased by a factor of 4, to produce the current extinction rates of around 0.5% of extant birds per century and 1% of mammals per century. Extinction rates in other groups of organisms have only begun to be tracked, but they are significantly higher than the historical average. In the United States, there were twice as many species of fish (350) classified as endangered in the 1990s than there were a decade earlier.
The major factor driving species to extinction in North America (and elsewhere in the world) appears to be changes in habitat, such as those that occur through climate change, desertification, or deforestation. Habitat perturbation often causes rapid extinction of species: the drying of a freshwater lake or the final

56

submergence of an eroding island obviously causes the immediate extinction of many species that once dwelled there. Others will die off later in time.
The number of individuals in any population of organisms is always fluctuating. There may be long-term trends toward increase or decrease, or even toward constancy, but these longer-term trends are themselves made up of shorter-term fluctuations. The fluctuations themselves have traditionally been thought to be related to environmental factors: changes in food supply, increased or decreased predation or competition; physical environmental changes such as long-term temperature change or habitat change. To understand these changes, ecologists have developed a series of equations that describe how birth and death rates � the ultimate determinants of population size � are affected by the external environment.
Just how important unpredictable fluctuations are to populations was perhaps first appreciated by Robert May. In the 1970s, May showed that population fluctuations in many species of animals and plants are not necessarily random, but instead may be an aspect of chaos, the relatively newly described phenomenon in which apparent randomness isn�t random after all. Although governed by precise mathematical rules, the behavior of a chaotic system is virtually impossible to predict. It may be that some populations of organisms show wild fluctuations that are caused not by external conditions, such as climate change, but by deeply rooted and complex dynamics within the ecosystems in which they reside. May also showed that the geographic distribution of organisms may be related to factors other than the external environment. May and his colleagues showed that population fluctuations within a patchy (or irregular) distribution may not be related simply to the favorability of each patch, but might be far more complex.
All of these findings have profound implications for conservation biology � and for the understanding of mass extinctions. In their 1996 book, The Sixth Extinction, Richard Leakey and Roger Lewin point out that

57

Birds are relatively large and highly conspicuous members of the planet�s biota, and thus are among the best-known groups in terms of both their current diversity and their history of recent extinctions. Because of this, they play a prominent role in our study and understanding of biodiversity loss. Birds also have the potential to leave a fossil record of themselves, so that diversity levels and losses in the past can be measured.
In 1997, David Steadman, curator of birds at the Florida Museum of Natural History, summarized losses of bird species since 1600 on both continents and islands. Steadman posited that humans have caused extinctions of birds in four main ways: direct predation (hunting, gathering eggs, or removing nestlings for captive breeding and pets), introduction of non-native species deleterious to bird survival, the spread of disease, and habitat degradation or loss. Of the approximately 10,000 species of birds on Earth today, about one-fourth have restricted breeding ranges (designated as 50,000 square kilometers or less). These are the species most susceptible to extinction.
Little is known about the prehistoric human impact on birds in most continental regions, but North America is one exception. Soon after humans arrived in North America, about 11,000-13,000 years ago, between twenty and forty species of birds went extinct. All of these birds may have been tied into ecosystems dependent on the large mammals that also went extinct at that time. It is likely, then, that the birds� extinctions were only indirectly tied to human causes. From 11,000 years ago until 500 years ago, only two additional bird species went extinct in North America. Since the arrival of Europeans approximately 500 years ago, an additional five to seven birds species have gone extinct, with five of these extinctions occurring in the last 200 years (the great auk, Labrador duck, passenger pigeon, Carolina parakeet, and ivory-billed woodpecker). Eight more species (the California condor, whooping crane, red-cockaded woodpecker, black-headed vireo, golden-cheeked warbler, and Kirtland�s warbler) are so close to extinction that only expensive, concerted captive breeding efforts (such as that taking place for the condor) will save them.
The rates of avian extinction in tropical continental regions outside of North America have been little studied. The paleontology of birds is far better known for many islands. The relatively small land areas of most islands result in smaller local populations of all organisms and, as a result, greater sensitivity to extinction. Of the 108 species of birds known to have gone extinct worldwide since 1600, 97% came from islands. Even more extinctions occurred in prehistoric times. Steadman estimated that at least 2,000 bird species, or about 20% of the total diversity of birds

58

on Earth, went extinct on islands after human, but before European, contact. In each case the extinctions postdated the first human contact with each island.
Predicting the future of birds is no easy feat. It may be that the most susceptible and delicate species have already gone, or will soon go, extinct. Perhaps the losses we have seen so recently will be the major losses. Yet many bird experts are not so sanguine, and see the ongoing cutting of the world�s forests � and their replacement with agricultural fields, an entirely different type of habitat � as a factor ensuring the continued loss of bird species.

Why the Modern Mass Extinction
May Not Be as Bad as Projected

One of the great dangers facing those who attempt to prophesy is that estimates, coming from the best of intentions, may become more catastrophic than the data warrant. Extinction is an emotional issue for many of us, even (or especially) scientists, and emotion can color judgment and distort objectivity. There is a very real possibility that estimates of current extinction rates are inflated. Few studies are able to pinpoint how real the threat of elevated extinction rates really is, or how prolonged it will be. There is a tendency by some working in this field to cry doom when a much more muted response may be justified.
It is clear that the planet is in a period of elevated extinction rates. But just how elevated, compared with the period prior to the population run-up of our own species, is the most pressing question, and one that is very difficult to answer. There is a possibility that most of the consequential extinction (i.e., among the megamammals) has already occurred, and that little further reduction in the Earth�s biota will accumulate over the next few centuries or millennia. Thus it may be that the estimation of losses through mass extinction is wildly overstated.

For all of the extinctions currently thought to be under way, actual case histories of extinctions are rather few. Those species that have gone extinct, ranging from the dodo to the passenger pigeon, may be species that for any number of reasons were extremely susceptible to extinction to begin with. Extinction requires the

59

death of every living individual of a given species. All species are the result of a long period of evolution. They do not just go away; something must eventually kill them all off, and that cause must be sufficient to end a history in most cases counted in millions of years.

Worldwide conservation efforts have brought to light the plight of many endangered species. Virtually every country on Earth now practices some form of conservation, be it by preserving large national parks or by protecting individual species or given habitats. These efforts have occurred only in the last two to three decades on a worldwide basis. Yet they have already registered a number of remarkable successes, notably in the restoration of whale and large bird species. Bans on dangerous chemicals such as DDT have vastly aided this process. These efforts alone may be sufficient to reverse the course of the oncoming and ongoing mass extinction.

As we have seen, one of the most maddening aspects of biodiversity studies is our very poor knowledge of the most basic baseline figure, the actual number of species on Earth, and the corollary to that figure, the reduction of species numbers among various taxonomic groups and specific habitats. In very few other avenues of science are the error ranges quite so great: an order of magnitude separates the high and low figures. It may be that there are a very large number of species on Earth, and that a relatively low percentage of them have recently undergone extinction, or will do so in the near future.

Why the Modern Mass Extinction
May Be Worse than Projected

Several factors could adversely affect the biological diversity of the Earth and serve to amplify the current rate of extinction. If we accept that the current levels of extinction are related to the activities of humankind � most importantly, the conversion of previously undisturbed habitat such as rainforest or native grassland into agricultural areas � then anything causing an increase in such conversion should adversely affect the biodiversity baseline. Any reduction in the land currently available for agriculture would be likely to spur more conversion. This could happen in several ways:

60

Several types of climate change could reduce the current area of farmland and hence create pressure for further habitat conversion. Global warming due to the greenhouse effect could cause the tropical regions to increase in size. This in turn would cause an expansion of the desert areas at higher latitudes, producing an adverse effect on the grain belts located there. If grain regions migrate to higher latitudes in turn, they will suffer shorter and harsher growing seasons, and thus reduced yields.
A second and opposite effect would be a return to a new glacial interval. The current warm period is but an interglacial interval in a long pattern of glacial cycles that has been operating for more than 2 million years. If past patterns are any guide, some thousands of years from now ice sheets will again begin to grow and cover vast regions of the Earth in some of the most productive agricultural regions.

Disruption of agriculture could also come from a rise in sea level. Even small rises in global sea level will result in significant land reductions in agricultural regions, and such small-scale rises will come about if current global warming patterns continue. River deltas, for example, are among the richest of all agricultural regions, and the first to be inundated by any rise in sea level. New evidence gathered from a study of Antarctic glaciers in 2001 indicates that the rate of sea level rise may be three or four times faster than the worst-case scenario of the late 1980s and early 1990s. There may be a 20-foot sea level rise in the next two centuries.

As we will see in greater detail in a subsequent chapter, the number of humans on Earth greatly affects the rest of its biota, and surely extinction rates as well. If the human population reaches some of the more extreme estimates over the next few centuries � over 50 billion people, for instance � there will certainly be greatly elevated extinction rates.
The Earth currently has more species than at any time during previous geologic epochs. This general pattern of increase in diversity over time may not continue, however. How it might change is described in the next chapter.

61

CONTENTS