Amphibia Fading.

The more we think we understand amphibian decline, the more mysterious it becomes-and the biggest mystery of all is whether we will act to stop it.

Why did it die out? The golden toad (Bufo periglenes) is by now perhaps the world's most famous amphibian, but it probably no longer exists it lived only in the Monteverde Cloud Forest Reserve in Costa Rica--a foggy that of dense upland forest wracked by heavy winds that blow off the Caribbean. The toad's main babitat was on one cold, wet ridge called Brillante where it emerged en masse in spring for five to ten days at a time, to mate in rainwater that pooled against the roots of the trees.

On the somber forest floor, these mating congregations were a spectacle of intense, almost hallucinatory color. The males were an improbable flaming orange and had eyes like black beads. They were only about 5 centimeters long they looked like Mayan treasure come to life. The females were slightly larger and colored so differently you wouldn't have thought they belonged to the same species; they were greenish black with bright red blotches edged in yellow. Photos of the male were used in the publicity campaign to establish the reserve. Its photos still adorn tourist posters. Eventually, it became the "poster road" for amphibian decline.

The last time the roads appeared en masse was in 1987. In 1988 only 10 toads were seen. A year later scientists found just one solitary male. In 1990 they found no toads at all. Initially, it was thought that perhaps the toads were just "hiding out"-- skipping a bad breeding season or two. The springs at Monteverde had been slightly warmer and drier than usual, and unfavorable spring weather is known to reduce the breeding populations of many other amphibians. But that generally does them no long term harm the creatures are usually back in force with the next good spring.

And over the very long term amphibians have indeed prospered. At some 350 million years of age, Amphibia is the world's oldest terrestrial vertebrate class. (A class is a taxonome group at the level of for example mammals or birds). Scientists have thus far identified nearly 5,000 species of frogs, toads, salamanders, newts and caecilians (legless largely subterranean creatures). There are more species of amphibians than there are of mammals. Amphibians' collective domain includes every continent except Antarctica, and probably most of the world's major islands. They achieve their greatest variety in tropical and warm temperate forests, but they also live in deserts, grasslands, northern bogs--even tundra, in the case of the wood frog (Rana sylvatica), one of four North American frogs that can "freeze solid" and survive--the only vertebrates known to have this ability.

Measured against the full breadth of this ancient and widespread class, the golden toad's plight didn't initially look very portentous. But it wasn't just a matter of the golden toad. By 1990, 19 other amphibian species had gone into serious decline or disappeared entirely from Monteverde. And it wasn't just a matter of Monteverde: as the 1990s wore on, reports of declines and disappearances emerged from most of the regions where amphibians were reasonably well monitored--in North America and parts of South America, in Europe and Australia.

Yet Monteverde cast a long shadow: many of these other declines unfolded in a way that bore an uncanny resemblance to events in Costa Rica. They were, in the first place, very rapid. They sometimes involved whole assemblages of species, rather than just one or two. And they were occurring not just in areas that were obviously disturbed, but in some of the world's most carefully protected parks. These were not the kinds of losses that could be readily predicted--or explained. The "Monteverde syndrome" suggested that something peculiar was happening to Amphibia--something bad enough to distinguish it from the broader tragedy we have come to know as the biodiversity crisis.

What Makes Them So Vulnerable?

"Amphibian" is a Greek construction meaning "double life"--a reference to the fact that the typical amphibian lifecycle is partly aquatic and partly terrestrial. That can make amphibians doubly vulnerable: disturbance of either water or land can affect them. In water, for example, some species have fairly narrow temperature requirements. Some do best in still water, others need flowing water. And many are particular about where they will breed. In southwestern California, the endangered arroyo toad (Bufo microscaphus californicus) does not reproduce well unless it lays its eggs on the sandy bottom of a slow moving stream. Some frogs and salamanders will lay eggs only in the shallow "vernal pools" that appear with the spring rains and disappear with the summer heat. This is a kind of evolutionary gamble with the weather: the young are safe from predatory fish in a vernal pool, but they must reach their terrestrial phase before the pool dries.

Given such preferences, it's not surprising that a primary ingredient of amphibian decline should be that standard form of environmental corrosion: habitat degradation. Many amphibians, for example, are forest animals and the world is currently losing about 14 million hectares of natural forest each year-that's an area larger than Greece. (It's true that tree plantation cover is expanding, but plantations do not generally provide an ecological substitute for natural forest; see "Paper Forests," WORLD WATCH, March/April 1998.) Even when the result is not outright deforestation, logging can devastate amphibian populations.

Consider the logging boom in the U.S. southeast, where the forests shelter the world's richest assemblage of salamanders. About 60 percent of all salamanders belong to a lineage called the Plethodontidae, which lack lungs. These creatures breathe through their skin, which must remain moist at all times to facilitate gas exchange, or they'll suffocate. Plethodontids are consequently extremely sensitive to changes in temperature and humidity. Even selective logging is likely to reduce a population, because it opens up the canopy and dries out the floor. Clearcutting a population's habitat is a death sentence. In the U.S. southeast, the logging of mature hardwood forest, the primary salamander habitat, is expected to overtake the hardwood growth rate by 2010. More and more of the region's rich salamander diversity is likely to end up sharing the plight of the red hills salamander (see page 18).

Deforestation- induced losses are almost certainly far greater in the tropics, although we generally know far less about them. The extreme case appears to be Sri Lanka. As recently as 1993, the island's amphibian fauna was thought to comprise only 38 species, but a recent five-year survey of the remaining rainforest turned up more than 200 additional amphibian species, which are apparently endemic to Sri Lanka (that is, they occur nowhere else). Today Sri Lanka is believed to have the world's highest amphibian diversity, in terms of the number of species per unit area. And yet that diversity is probably just a shadow of what it once was. Over the past 150 years or so, the island has lost 96 percent of its original rainforest cover. When survey researchers checked the records of naturalists who were exploring Sri Lanka before 1900, they found that more than half of the amphibians mentioned by their predecessors were no longer present. Most of Sri Lanka's surviving natural forests are legally protected, but th ey continue to dwindle in the face of illegal logging, primarily for fuelwood.

Habitat loss is thought to be the leading cause of amphibian decline, but it obviously cannot account for the "Monteverde syndrome." Places like Monteverde would seem to be about as close as it's possible to get to pristine--their habitats are intact. And yet the amphibians in these places are apparently reacting to dramatic changes. But these are changes that most of us either don't see, or that we just don't read as "unnatural."

Toxics are the usual suspects in cases of invisible damage, and there's no question that amphibians are highly vulnerable to them. Amphibians have thin, permeable skin that readily absorbs contaminants; their eggs lack protective shells and are highly permeable as well. So it's hardly surprising that in heavily industrialized areas, pollution is frequently invoked as a cause for local declines. In some centers of heavy industry, the pollution is so intense and pervasive that it's a wonder there are any amphibians left to study. Here, for example, are the types of pollution that appear to be injuring amphibians in Ukraine: heavy metals, pesticides, aromatic hydrocarbons, acid rain, and radioactive waste.

But pollution is taking a toll in healthy-looking landscapes as well. In Britain, the acidification of ponds is a major factor in the endangerment of the Natterjack toad (Bufo calamita). The toad is now nearly extinct in British lowland heath, a habitat that used to support about half the species' population in that country. (The toad is faring poorly in Scandinavia too, but it seems to be in better condition farther south.) California's Sierra Nevada range is losing many of its amphibians, and some recent studies suggest that pesticide contamination may be a factor. Pesticides have been detected in precipitation as high as 2,200 meters. The chemicals are presumably drifting up from the state's heavily farmed lowlands.

Pesticides have also been invoked to explain the recent rash of amphibian deformities (although there is no clear relationship between the deformities and the declines). In Minnesota and in the St. Lawrence River Valley in Quebec, researchers have turned up many frogs with missing or extra legs. Some scientists think there is a link between these malformations and certain pesticides. On the other hand, deformed frogs have been found at various California sites where there are no signs of pesticide contamination. The problem in these places appears to be infection with a kind of parasitic flatworm called a trematode. Trematodes are also being blamed for deformities elsewhere in the United States, and that invites additional questions. Is there an "epidemic" of trematode-induced deformities, or are scientists finding lots of afflicted frogs simply because they are now looking for them? If there is a trematode epidemic, is that because something has upset the relationship between the parasite and its host, or i s it because the trematode is moving into new areas? The deformities may be no easier to explain than the declines.

It's not very surprising that pesticides should be implicated in both the deformities and the declines, since pesticides, after all, are designed to be toxic. But fertilizers, which are used in far greater quantities than pesticides, may be creating problems we are even less prepared to counter. Some amphibians are very sensitive to the nitrogen compounds that typically leach out of artificially fertilized fields. For example, researchers have discovered that tadpoles of the Oregon spotted frog (Rana pretiosa) are poisoned by water with nitrate and nitrite levels low enough to pass the drinking water standards set by the U.S. Environmental Protection Agency. (Nitrate and nitrite are compounds that soil microorganisms make from fertilizer.) Many water supplies in the United States contain levels of nitrate that violate EPA standards. If these standards aren't taken seriously as a matter of public health, it seems unlikely--to say the least--that more stringent standards will be mandated for the welfare of fro gs. The Oregon spotted frog has largely disappeared from its historical range in the heavily farmed Willamette River Valley.

Pollution, like habitat loss, is clearly a major factor in amphibian decline. But even when you take both stresses into account, many declines remain unexplained. Not far from the Willamette Valley, in the Cascades Range of Oregon, the Cascades frog (Rana cascadae) and the Western toad (Bufo boreas) are disappearing, even though their habitat has not been significantly disturbed or polluted. Oregon State University biologist Andrew Blaustein has shown that these species are the victims of another stress: increased exposure to ultraviolet (UV) light, a consequence of the weakening of the ozone layer, which filters much of the UV out of incoming sunlight. UV light can damage DNA and even kill cells. Amphibians, with their naked skins and eggs, are good candidate victims. The Cascade species may be losing their eggs to the extra UV.

It's likely that increased UV levels are injuring other amphibians as well, particularly those at higher latitudes, where the ozone layer tends to be weaker. And unfortunately, the seasonal fluctuation of the layer probably increases amphibian vulnerability: in either hemisphere, the layer tends to be at its weakest during winter and spring, a period that overlaps with the egg-laying season for most species. Amphibians at higher elevations could be especially susceptible as well, since the higher you go, the less atmosphere there is to filter out the UV. But researchers have found that not all amphibians are especially sensitive to UV light, and not all are exposed to appreciable quantities of it. Tropical amphibians are living beneath a thicker ozone layer, so their UV exposure has probably not risen much. Even in the temperate zones, forest amphibians would generally be protected by the forest canopy (although a deciduous canopy wouldn't offer much protection in early spring).

Amphibians face another major pressure, which is invisible in a different way. UV light and most types of pollution cannot literally be seen. Introduced, non-native species, on the other hand, are often in plain view but because they usually look perfectly "natural," it can be hard to see them as a threat. Yet non-native species frequently prey on amphibians or out-compete them for food. In the Yosemite region of California's Sierra Nevada Mountains, for example, biologists Charles Drost and Gary Fellers argue that introduced trout have played a role in the disappearance or severe decline of five of the region's seven native amphibians. Yosemite waters above 1500 meters had no native fish, so the local amphibians were apparently not adapted to cope with big aquatic predators. Intensive stocking of trout began in the 1920s, and today, the presence of trout correlates strongly with several declines and disappearances. But the correlation is not exact, since some species did not decline until long after the tro ut were introduced, and some declines occurred in waters where trout were never introduced at all. Some of these cases likely involve trout plus drought: the trout reduced the amphibians' ranges to isolated patches, and that increased their vulnerability to Yosemite's 1987-1992 dry spell. The troutless cases, however, remain a mystery.

One invader that frequently injures amphibians is itself an amphibian: the bullfrog (Rana catesbeiana), native to the eastern United States. An aggressive, fast-growing species that can reach a length of 15 centimeters, the bullfrog has been introduced into many ponds and marshes around the world for food and fishbait. The bullfrog is not particular about its habitat and it has a voracious appetite. It will try to swallow almost anything it can fit in its mouth. After it was introduced into California in the early 1900s, several populations of the red-legged frog (Rana aurora) and the foothill yellow-legged frog (R. boylii) vanished. Perhaps the bullfrog out-competed them for prey; perhaps it swallowed them. In South Korea, where it was imported for food in the early 1970s, the consequent loss of native frogs and other small creatures inspired an official anti-bullfrog campaign, with hunting contests and bounty prizes.

Epidemic Losses

There's another organism that is lurking in many forests and swamps, and that may be killing far more amphibians than bullfrogs and trout. In 1992, Karen Lips, a herpetologist then working in Las Tablas, Costa Rica, discovered several dead and dying frogs at her research site--a rare find, since dead frogs are usually snapped up quickly by scavengers, but the significance only emerged in retrospect.

Over the next four years, Lips documented population collapses in five species formerly abundant at Las Tablas. Then in 1996 and 1997, she was surveying in the Reserva Forestal Fortuna in Panama, an area she had studied several years earlier. Five of the seven streams she checked were frogless; the other two contained only half the species encountered in her earlier visits. And at Fortuna, she found dead frogs too:

"I found most dead and dying animals 'frozen' in their normal calling positions, so it appeared as if they came to the stream the previous night and died in place. Many of the casualties still had a very lifelike appearance; most were found during morning surveys, still sitting in a perched position. Dying individuals were lethargic, had no righting response and exhibited convulsions and trembling of the limbs and head."

An amphibian epidemic was apparently moving through central America. That reminded scientists of an earlier series of extinctions along the east coast of Australia. In the area from Brisbane to the Cape York Peninsula, at least 14 rainforest frogs had gone extinct or declined by more than 90 percent since the late 1970s. In both regions, the victims were stream dwellers and had succumbed rapidly--traits suggestive of a virulent, waterborne pathogen. Researchers comparing skin samples from Panamanian and Australian victims found them infected by the same type of organism: one or more fungi of the phylum Chytridiomycota. Chytrid fungi are common pathogens of plants and insects but had never before been known to attack vertebrates.

At the same time the Australia/Panama research team was closing in on the chytrid, other scientists were finding it in the United States. Don Nichols, a pathologist at the National Zoological Park in Washington, D.C., first noticed the disease in 1991, when it started killing captive arroyo toads in California, but it wasn't until 1996 that the toads' disease was identified as the chytrid fungus. The fungus has since been found in the wild in various places around the country. In Illinois and Maryland, it is an apparently benign infection of some frog populations: the infected animals seem perfectly healthy. But last year, wildlife officials found it in lowland leopard frogs (Rana yavapaiensis) outside the city of Phoenix, Arizona, and in boreal toads (Bufo boreas boreas) near Denver, Colorado. In both cases, officials encountered large numbers of dead and dying animals. Both species have been in sharp decline, and the fungus is now a prime suspect in these casualties as well.

Scientists are now wondering how many other declines the fungus has caused. Could it have been behind some of the older die-offs--events later marshaled as evidence for global amphibian decline? Did it, for instance, cause the massive declines of leopard frogs in the Colorado Rockies in 1974? Cynthia Carey, a biologist at the University of Colorado, and D. Earl Green, a pathologist with the U.S. Geological Survey, began opening up specimen bottles from the 1970s, and found the chytrid in some preserved leopard frogs from Colorado. So the leopard frogs may also have been victims of the chytrid. The fungus is now being investigated as a possible agent for many declines in the western United States.

A similar investigation is going on in Australia, where the chytrid is continuing to expand its range. By the summer of 1998, it had traveled some 6,000 kilometers from the northeast coast to the southwest coast, where it was discovered in a frog population near Perth. It has now been detected in 24 Australian species and linked directly to 11 declines. And Australian researchers have been opening their old specimen bottles too. A dainty tree frog (Litoria gracilenta) collected in southern Queensland in 1978 is the oldest infected specimen found thus fir.

The fungus has now been isolated from one type of frog and given a name: Batrachochytrium dendrobatidis. But that doesn't make it much more of a known quantity. Scientists don't yet know whether there are other species of the pathogen, or just this one. Nor do they know how it kills its victims. It may suffocate them by causing their skin to thicken (many frogs breathe partly through their skin), or it may produce toxins. And there are plenty of other questions as well. Where, for instance, did it come from? One theory holds that until the 1970s or thereabouts, the chytrid had a much more restricted range somewhere in the northern latitudes, presumably in North America. That might explain the unaffected populations in the U.S. east coast and midwest, since long exposure could have given them a chance to develop resistance. From there, perhaps, it was only recently introduced into other regions, where its victims would not have been adapted to it.

So perhaps the chytrid is a recent invader of Central America and Australia. That possibility raises an issue of special urgency: what might be moving it around? Some herpetologists think it may have been brought into new terrain on the boots of tourists. Others see a likely conduit in the trade in aquarium fish, and in amphibians themselves. In Australia, scientists suspect that the chytrid made its way across the continent in the skin of an infected frog that stowed itself away in a box of fruit. And once the fungus arrives in an area, there are all sorts of ways it could spread: it may be dispersed by local people, for example, or cattle. It can even be carried by birds and insects.

But chytrid infection outside eastern North America is not an automatic death sentence; the dainty tree frog, for example, is still a common Australian species. That doesn't invalidate the "invasion theory" outright, but it could be evidence that the fungus was in Australia long before 1978. That suggests another theory. Perhaps the fungus is a well-established pathogen in many parts of world, and something is upsetting amphibian immune responses. "It doesn't do a parasite any good to kill its host," Don Nichols has noted. "Other factors may be tipping the balance."

This theory might find support in the behavior of the various other pathogens implicated in amphibian declines. A group of viruses called iridoviruses may have caused the tiger salamander (Ambystoma tigrinum) die-offs in Arizona, Utah, Maine, and Saskatchewan. Iridoviruses have also triggered declines of the common frog (Rana temporaria) and the common toad (Bufo bufo) in Britain. Boreal toads in Colorado have been attacked not just by the chytrid, but also by a bacterium, Aeromonas hydrophila. And a fungus of the genus Basidiobolus is killing Wyoming toads (Bufo hemiophrys baxteri) in Colorado, Wyoming, and Minnesota. Why so many epidemics in so many places, in so short a time? It's possible that scientists are just finding more disease because they're getting better at looking for it. Or it might be evidence that some widespread stresses are throwing amphibian immune systems Out of kilter.

By the late 1990s, evidence for one such stress turned up in Costa Rica. Working on the initial hunch that the golden toad's disappearance had something to do with the weather, a group of climatologists and biologists led by Alan Pounds, head of Monteverde's Golden Toad Laboratory for Conservation, found strong evidence that the reserve's cloud forest is losing its clouds. According to their findings, published in the April 15, 1999 issue of Nature, local sea surface temperatures have risen since the mid-1970s, and that has tended to push the cloud bank higher. The mountain tops are bathed in the clouds less frequently, so the forest is now somewhat drier. This drying might account for the amphibian losses. The theory is corroborated by bird observations: some lowland, "cloud-forest-intolerant" birds have moved upslope, into areas they had never occupied before. But the golden toad lived on top of the range and had nowhere to go.

Did the golden toad die of climate change? One appealing aspect of this theory is that it might apply to die-offs in other regions with a similar topography. The mountain forests of Puerto Rico, for example, have seen 12 of their 18 endemic frogs decline over the past 20 years, and three of them may now be extinct. There is as yet no satisfying explanation for these losses. Elsewhere, other forms of climate change could threaten amphibians--drought, for example, or rising water temperatures.

But many scientists see another, more ominous possibility in the team's findings. Climate change could be "overlapping" with disease: the stresses of a changing climate could make amphibians more susceptible to infection. Scientists have not yet looked for the chytrid at Monteverde, but perhaps the fungus haunts that forest as well. Perhaps the Monteverde declines are the result of a kind synergism between the pathogen and the warming seas.

Climate change and spreading disease: both of these forces have a global reach and they could overlap in a number of ways--either simultaneously or in sequence. A change in the moisture regime, as at Monteverde, or a change in water temperature might weaken amphibian immune systems. Warmer water might also affect a pathogen's virulence, or its capacity to move from one animal to another. Warmer air might increase the range of insects that carry it.

Infections are likely to combine with other types of stress as well. Excess UV exposure, like climate stress, could suppress amphibian immune systems. So could some forms of pollution. And some diseases appear to have been spread through the introduction of infected game fish like trout; in such cases, a new predator overlaps with a new disease.

There are many overlaps besides those that involve disease. Consider non-native species. Successful invaders are often capable of tolerating very disturbed conditions. If some kind of disturbance injures the native amphibians but does an invader no harm, then the latter may gain a level of dominance it might not otherwise have achieved. In the Ural Mountains of Russia, this mechanism has apparently allowed the introduced lake frog (Rana ridibunda) to displace some of the native frogs. The lake frog apparently tolerates industrial pollution much better than the natives do, so the natives may have succumbed to an invasion-pollution overlap.

Habitat loss, pollution, UV exposure, non-native species, disease, and climatic instability--those are the stresses that we know or suspect are killing off so many of the world's amphibians. Perhaps there are other, as yet unidentified stresses as well. But a simple inventory like this, dismal as it is, still doesn't adequately represent the threat, because it doesn't account for the overlap factor. And unfortunately, we cannot predict exactly what those overlaps will eventually do.

Beyond the Declines

The loss of amphibians demands our attention not just because we need to know why they are dying, but also because we need to know what their death will mean. Amphibian decline is itself a form of environmental degradation, since amphibians play critical roles in many ecosystems. Although their secretive, inconspicuous nature might suggest otherwise, in some temperate and tropical forests, amphibians account for more biomass than any other vertebrate group: if you could weigh all the frogs and toads in a forest, there's a good chance you would find more mass in them than in the forest's reptiles, birds, or mammals.

That's important because it means that a great deal of the nutrients and energy in these places normally passes through or resides in amphibians. In ways large and small, amphibians shape the ecosystems of which they form a part. In ponds, for example, tadpoles may keep algal growth in check. Remove the tadpoles and you may end up with an oxygen-depleting algal bloom. Adult frogs and toads often devour vast quantities of invertebrates, especially insects. Large frog and toad species also eat fish, birds, and even small mammals. In some wetlands, amphibians are the top predators, exerting enormous influence on the diversity and abundance of other organisms.

And of course, amphibians are themselves important prey for many other animals, including fish, birds, reptiles, and mammals. Some bats and snakes live exclusively on amphibians; their fate will mirror that of their prey. In California's Sierra Nevadas the decline of the mountain yellow-legged frog (Rana muscosa) apparently underlies a decline in a local garter snake species, which is one of the frog's main predators. There are probably many such casualties in the world's forests and wetlands.

Amphibian decline is an incipient social tragedy as well. There are, first of all, the practical effects of losing major insect predators. During the 1970s, for example, India was a major supplier of large frogs to the culinary markets of Europe, the United States, and Japan. But after the trade had cleared many marshes, the mosquito populations exploded, malaria infections rose, and the authorities responded by increasing insecticide applications. The trade was banned in 1979, but much of it has gone underground, as a smuggling network into Bangladesh.

But the social tragedy is not simply a matter of immediate effect; it's also a question of lost potential.

Amphibians are an incredibly diverse group of organisms and we know relatively little about them. We do know that they are living chemical factories; amphibians produce all sorts of powerful compounds, or they concentrate compounds found in their prey. This characteristic is frequently a form of defense. Because amphibian skin is thin and permeable, it offers little physical protection from attack or infection. So the protection is often chemical instead. Many species produce antibiotics and fungicides. Some produce powerful poisons, which they advertise to predators with their bright coloring.

These chemicals are a medical treasure, as many traditional cultures have long recognized. Pulverized toads, for example, have long been used in traditional Chinese medicine for a variety of ailments. And while traditional remedies doubtless vary greatly in their efficacy, modern chemistry is substantiating the power of many of the raw materials. In Ecuador, for example, indigenous peoples have long used a local frog's skin secretion as a painkiller; the secretion contains a chemical that is reportedly 200 times more powerful than morphine--and that lacks the side-effects of opiates. Abbott Laboratories, a U.S. pharmaceuticals company, is developing a drug modeled on the chemical.

Many other amphibians could contribute to our pharmacopoeia as well. A foamy secretion of the African clawed frog (Xenopus laevis) could become an important new antibiotic. Compounds in the skin of various South American dart poison frogs might be useful as anesthetics, muscle relaxants, and heart stimulants. A chemical produced by the South American bicolored tree frog (Phyllomedusa bicolor) could open up new possibilities for treating Alzheimer's disease and depression.

We have yet to learn even the most obvious lessons from some of these creatures. The wood frog (Rana sylvatica) is able to tolerate temperatures low enough to turn up to 65 percent of its body water to ice. It produces some sort of natural antifreeze to keep the remaining water liquid, but how does this system work? An Australian water-holding frog (a Cyclorana species) is able to absorb enough water to last for months or even years of drought. How does it do that? Australia's gastric brooding frog (Rheobatrachus silus) was the only animal known to incubate its eggs in its stomach--a feat it apparently accomplished by switching off its digestive enzymes. Unfortunately, that frog's secrets are now beyond our reach. The gastric brooding frog disappeared in 1981.

The Nature of Science Itself

It has been more than 10 years since the prospect of global amphibian decline first attracted widespread scientific attention. There is now little doubt that the problem is real. And yet, even as the evidence piles up, the phenomenon itself seems to grow increasingly mysterious. It is clearly global, but we don't know its full extent. It is definitely the result of human activity, but it cannot be explained by just one or two causes. It is doubtless of deep ecological significance--but we have only a rather vague sense of what that significance is. When we peer into amphibian decline, we are looking into the depths of our own ignorance.

In part, our ignorance seems to be a by-product of specialization--of the compartmentalization of information and research. The declines cannot readily be contained in a single field of inquiry. They involve microscopic pathogens and global climate change; they are part of forestry economics and wildlife toxicology. Understanding them will require a much more interdisciplinary, integrative approach than is typical of conventional research. Some of the most successful recent work is already moving in this direction--the climate research at Monteverde, for example, or the international investigation of the chytrid fungus.

That idea is beginning to resonate within some major scientific institutions, such as the National Science Foundation, the U.S. government agency that is the chief source of U.S. federal funds for scientific research. In 1999, the NSF awarded $3 million to a team of 24 scientists from fields as diverse as veterinary epidemiology and evolutionary ecology to study host-pathogen relationships as an aspect of amphibian decline. Jim Collins, the Arizona State University biologist who heads the team, explained the challenge this way: "as we went through thinking about how to answer the questions, we really had to think about how we did the science. And how we did the science had to change--it couldn't be just an individual investigator laboring away in an isolated laboratory." Collins emphasizes the need for interaction not just between the different biological disciplines, but also with the social sciences and possibly even the humanities. "To understand this problem we have to do a better job of integrating huma ns into ecological and evolutionary theory," he argues. "The nature of science itself is going to have to change."

The science is going to have to change in another way as well: there's an enormous geographical mismatch between the research capacity and the creatures themselves. Canada, for example, has plenty of amphibian specialists, but not a single endemic amphibian. On the other hand, Mexico, like most tropical countries, has a vast amphibian fauna and very few herpetologists. We know a good deal about the amphibians of the United States, western Europe, Costa Rica, and Australia, but little about those of South America, Asia, or Africa. It's possible that the tropics could harbor thousands of amphibians not yet known to science. And among the tropical species already identified, many are in the literature only by virtue of their original descriptions; in such cases, virtually nothing is known of the animal's ecology or even whether it is still extant.

If you want to get a sense of how skewed the research is, look at the April 13, 2000 issue of Nature, which contains the most exhaustive overview of amphibian population trends to date. To produce it, a team of researchers led by Jeff Houlahan, a biologist at the University of Ottawa, combed through the scientific literature and coaxed additional data out of more than 200 scientists in 37 countries. Of the 936 populations the team managed to cover, 87 percent were in western Europe or North America; 5 percent were in Central and South America, 2 percent were in Asia, and less than 0.5 percent were in Africa and the Middle East.

Despite the sketchy science, declines are being reported in many tropical countries. In Ecuador, for example, a little frog called the jambato (Atelopus ignescens) used to be so common you could find it in the backyards of Quito, but the last time a jambato was seen alive was in 1988. Santiago Ron, a biologist at the Universidad Catolica del Ecuador, thinks the frog may have fallen victim to a fungus, possibly the chytrid, but there is no money to investigate.

In Latin America, amphibian research may be constrained by a lack of funds--but not generally by a lack of interest. Last year, the Declining Amphibian Populations Task Force helped organize a series of workshops in Mexico, Panama, and Ecuador. (For more on the DAPTF, see below.) The organizers hope the meetings will be the first in a series of efforts to coordinate and build amphibian research in the region. Given the level of participation--the workshops drew in 88 people from 13 countries--that seems like a reasonable expectation.

Elsewhere, however, the research suffers from a much more profound institutional poverty. Take Sri Lanka, for example. That country's recent amphibian inventory is a first-rate scientific achievement by any standard, and yet Sri Lanka's Department of Wildlife Conservation, which is the agency responsible for all government conservation activities, and which has a staff of about 800 people, is reported not to employ a single biologist, even at the B.S. level. It is said to have fewer than 10 people with any type of degree, and to discourage independent research, because researchers may "show up" the Department. (The amphibian inventory was not a Department project.)

An exclusive focus on the scientific agenda would, however, confuse the basic issue. It's certainly reasonable to argue for more money for research and conservation. It's reasonable to demand better conservation policies. But the painful fact is this: no amount of money or cleverness is going to make this problem go away--if it's directed only at this problem. We cannot protect amphibians simply by trying to protect amphibians. The survival of these creatures now depends on our willingness to confront the major, systemic environmental issues of our day: climate change, forest loss, pollution, the spread of invasive species, and the control of the human population. Amphibian decline is a fundamental challenge to the way we live. We may not understand all the biological particulars, but the ethical issue is now very clear.

Ashley Mattoon is a research associate at the Worldwatch Institute.

In Costa Rica's Monteverde Cloud Forest Reserve, a species of harlequin frog (Atelpus varius, left), the green-eyed frog (Rana vibicaria, right), and the golden toad (Bufo periglenes, far right) are among the 20 amphibians that have disappeared since 1987. Climate change may have been a factor.

In Venezuela's Cordillera de Merida, a local harlequin frog (Atelopus carbonerensis) is among the five amphibians known to have declined sharply since the 1970s. The harlequin frog may now be extinct in the area. Conversion of forest to pasture seems to have played a role.

In Alabama, in the U.S. Southeast, the red hills salamander (Phaeognathus hubrichti) is now officially listed as threatened because of habitat loss from logging.

In Southeast Queensland, Australia, the gastric brooding frog (Rheobactrachus silus) apparently went extinct in the early 1980s. It may have been an early victim of the chytrid fungus.

In Britain, the natterjack toad (Bufo calamita) is now endangered, partly due to acidification of the ponds in which it breeds.

Amphibian Decline at a Glance

Examples included in this map are representative of the best documented declines. Hundreds of other declines have been reported but the evidence is often inconclusive. The emptiness of Africa and Asia reflects a lack of solid data far those regions. The 12 countries having the largest number of endemic amphibians (species occurring nowhere else) are shaded. The same set has the largest total number of amphibian species if Madagascar is removed and Venezuela is added. The numbers below the country names give the total number of amphibians and the number of endemics (in parentheses). These numbers came from the most recent general data set available, but they are already outdated, since additional species are continually discovered. (The numbers do not, for example, reflect the more than 200 endemics recently discovered in Sri Lanka.] It's important to notice also that this set of "high diversity" countries is biased towards larger countries: if diversity is measured not by absolute numbers, but in terms of the number of species per unit area, many smaller countries would rival the ones shaded here.

Yosemite, California: 5 of the 7 native amphibians have disappeared or declined since the early 1900s. Theory: introduced trout and a recent drought (highly probable, but only a partial explanation).

Arizona and Colorado: leopard frogs and boreal toads have declined sharply since the late 1970s. Theory: chytrid fungus infection and in the case of the boreal toad, bacterial infection as well (uncertain).

Las Tablas, Costa Rica and Fortuna, Panama: a large but undefined number of species have declined substantially since the early 1990s. Theory: chytrid fungus infection, possibly combined with other factors (highly probable].

Monteverde, Costa Rica: 20 of the 50 native amphibians have disappeared since 1987. Theory: climate change, possibly combined with infection (uncertain].

Ecuador: several species have undergone sharp, recent declines. (No organized theory, but there is speculation about fungal infection.)

Mountains of Puerlo Rico: l2 of the 18 endemic amphibians have declined sharply since the early 1980s; 3 may be extinct (No organized theory but there is speculation about climate change.)

Cordillera de Merida Venezuela: 5 species have declined sharply since the 1970s. Theories: deforestation, road kills, pollution (all uncertain).

Reserva Atlantica Brazil: 8 of the 13 native frogs disappeared or declined in the 1980s. Theories: dry winters, pollution or a combination of the two (one uncertain).

United Kingdom: all 6 native amphibians have declined. Theory: loss of breeding ponds, acidification of surviving ponds, and various other factors (virtually certain).

In Sri Lanka, more than 200 new amphibian species were discovered in a recent survey of the island's remnant rainforests--an indication of how mysterious the world's tropical amphibian fauna still is.

Mountains of eastern Australia: 14 native species have declined sharply since the late 1970s; 4 are probably extinct. Theory: chytrid fungus infection (likely).

Source for the numbers: Russell A. Mittermeier and Cristino Goettsch Mittermeier, Megadivresity; Earth's Biologically Wealthiest Nations (Prado Norte, Mexico: Cemex/Washington, DC: Conservation International, 1997).