Excerpted from The Kingdom of Rarities, by Eric Dinerstein. © January 2013, Island Press.


“Look for an overripe, black banana moving through the grass.” Edson Endrigo, our nature guide extraordinaire, was explaining his technique for spotting giant anteaters in Serra da Canastra National Park, just one of the rarities in this area. Obediently looking up on the hillside, I spotted a two-meter-long mobile banana. We jumped out of the van and circled behind a female anteater with a baby clinging to her back. My two companions, David Wilcove and John Morrison, and I closely tracked her progress.

If the greater one-horned rhino seems odd and prehistoric, the giant anteater offers good company as one of the most peculiar-looking mammals on the planet. Both are ranked as threatened on the IUCN Red List of Threatened Species. The tamandua-bandeira, or papa-formigas, as it is known in Brazil, cuts a comical figure, sporting an elongate, arching snout and bowlegged limbs, all ending in an immense shaggy tail. The rest of the body is shaggy, too, featuring a striking long pelage of dark bands on light. The female in front of us moved along like an animated throw rug.

An anteater walks on thickened pads on the outsides of its paws, as its digits are turned under its feet. An observer might think of this awkward creature, with its poor eyesight, bad hearing, and odd gait, as defenseless against secretive jaguars and pumas. That would be a miscalculation. With its acute sense of smell, the anteater can make up for its nearsightedness. If cornered, it will stand up on its hind legs and slash with its massive claws any human or feline predator foolish enough to tangle with it.

The mama anteater stopped and flicked her tongue in the dirt. Unlike the vast majority of mammals, the giant anteater lacks teeth. It has no real need for them because it inserts its long, narrow tongue into crevices, removes ants and termites with its sticky saliva, and swallows them whole. Crouching downwind, I inhaled deeply to catch its scent and wondered if consuming 30,000 ants a day gives this creature, or its flatulence, the odor of formic acid. I smelled nothing unusual.


Human activities sometimes bring species back from the brink of extinction. But more often they exacerbate rarity even to the point of disappearance, drive into rarity species once common, or further constrain those species that normally have narrow ranges or live at low densities. The most dramatic change happening today that is pushing already uncommon species toward even greater rarity is the conversion of rain forests and natural savannas into commodities production for industrialized agriculture. Big Ag, as it is now known, is largely mechanized, highly profitable, and controlled by multinational corporations. Some biologists and geographers describe extension of this trend as the future; increasingly, we live on a cultivated planet. The loss of natural habitats through nonagricultural use—that is, human settlements—and in nontropical areas is also high, but the conversion is greatest in the tropics and through big agriculture.

Few field biologists bother to check the daily price of soybeans or palm oil. This is an oversight because the market value of these commodities—along with that of beef, corn, sugar, and coffee—may over the coming decades define the future of rare species more profoundly than will any other driver of habitat loss. At present, nowhere is the conversion and fracturing of rain forests by industrialized agriculture in the world’s hotbeds of rarity more evident than in Southeast Asia and Brazil. In Kalimantan and Sumatra, Indonesia, expansion of oil palm and wood pulp plantations threatens the most species-rich rain forests in the world. In Brazil, vast areas of the Amazon are turning into cattle ranches and soybean farms. In addition to causing habitat loss, such rampant conversion imperils climate stability. Nearly 70 percent of the greenhouse gas emissions released annually from tropical forests originate from agriculture-driven forest conversion in just two places, Riau Province, Sumatra, and the state of Mato Grosso, at the edge of the Amazon in Brazil.

The plight of the Brazilian Amazon grabs headlines, but the status of its neighbor, the Cerrado, remains veiled in obscurity. This omission is a pity because the savannas of South America hold the key to reaching a balance between safeguarding rarities and growing the food we eat. Few environmental journalists are familiar with the Cerrado, which represents about 21 percent of Brazil and where both the total amount and annual rate of habitat conversion is higher than in the Amazon region. Over the past fifty years, more than 55 percent of the native habitat of the Cerrado has been cleared to make way for crops and livestock. Only about 2 percent of the region receives formal protection from the federal government, and Brazil’s Forest Code, at least on paper, requires protection of habitat on 20–30 percent (depending on the Brazilian state) of private lands.

The Cerrado borders the Amazon rain forest to the west and the green ribbon of the Atlantic Forest to the east. To the south lies the vast seasonal swamp known as the Pantanal. The Cerrado ranks as the world’s most diverse tropical savanna, even richer than the miombo, a similar habitat in southern Africa. The miombo’s infertile soils and tsetse fly infestations repel agriculturalists, whereas the Cerrado can be farmed for commercial crops after some soil modification. It has become the world’s largest producer of soybeans and beef and soon will be a major producer of sugarcane. The Cerrado has the dubious distinction—along with the previously mentioned Indonesian provinces of Sumatra and Kalimantan—of being among the most biologically diverse landscapes being converted most rapidly to agriculture.

This endangered tropical savanna features an unusual trio of rare mammals—the giant anteater, giant armadillo, and maned wolf. Very few tourists travel to Brazil explicitly to see them, even though the Cerrado offers the best chances of a sighting anywhere in the world. The ecotourism value of these species per hectare is far below the return that ranchers receive for beef cattle, soy, and sugarcane. So how do these rare species, along with jaguars, pumas, tapirs, and other wide-ranging Cerrado vertebrates, cope with massive land-use change driven by human economics?

Biologists who point out that a number of species can coexist in environmentally friendly cultivated zones have coined the term “countryside biogeography” (“matrix conservation” in Europe) to characterize the study of this phenomenon. This new discipline is essentially the study of which species persist in agricultural landscapes, assuming that interspersed with intensively used farmland are patches of natural habitat. To explore this issue and its relevance to the preservation of rarity, our next stops include Serra da Canastra and Emas National Parks in Brazil, at the edge of an expanding agricultural frontier that threatens to plow under rarity. Here, biologists are using a startling field technique that, along with the global positioning system collars worn by jaguars and pumas in the Peruvian Amazon, could revolutionize the study of rare vertebrates.


Edson shooed us back into the van to pursue the other goal of this outing: to search for the Cerrado’s rare endemic birds. Earlier that morning, he had led us to the Brazilian merganser, an incredibly rare duck that, like the Kirtland’s warbler and the greater one-horned rhinoceros, is an extreme habitat specialist, one that lives only on the fast-moving, clear streams of the upland Cerrado. Contamination from gold mining (now banned) caused the decline of this species. Edson guided us down a canyon to give us a fabulous view of a bird whose entire global population was probably no more than 250 individuals. Mergansers are elegant-looking ducks, but the Brazilian version has a startling profile, accented by its pointy head feathers. This solitary female had chicks perched on her back as she guarded them through their first week of life.

As we moved on, nothing could have prepared us for the next sighting. In a grassland to our left, Edson’s sharp eyes spotted a hovering bird. It was the most dashing of raptors, an aplomado falcon. The aerial predator was preoccupied, following something gliding through the tall grass. Then a head with pointed ears emerged. The maned wolf looked around for a second and moved on. The falcon persisted, perhaps planning to feast on the large insects or birds scared up during the terrestrial predator’s afternoon hunt.

Shadows fell over the rugged escarpment in the distance as the afternoon wore on. We drove out of the park and entered the agricultural zone—“the Ag,” for short. An hour later, our magical sightings of the wolf and the falcon, the merganser and giant anteater, began to feel like a dream. This had been only a first taste of the Cerrado’s wildlife. To learn more about how these animals navigated the last natural pockets embedded in a landscape of soy and cattle would require a longer stay, and for that I had decided to join an unlikely pair of long-term researchers.

March 2008. The sea of grasses undulated in the warm, dry breezes. A tall, blond woman dressed in khakis and field vest reached down to release her dog from its leash. “Okay, Mason, let’s go to work!” The dog dashed into the tall grass of Emas National Park. Every so often a grassy wave broke over the upright tail of the black Labrador retriever as he bounded through a large marshy area bordering a palm glade. The tail zigged and zagged through the wet pampa. Carly Vynne, then a PhD candidate at the University of Washington in Seattle, kept her eye on the dog. Within minutes, Mason returned with a look of great urgency. “What is it, Mason? Let’s go look.” Having grabbed her attention, Mason led us back through the muck and stood with his nose pointed toward the base of a grass clump.

At first we couldn’t see anything. Then we bent down and noticed a cylindrical dropping half submerged below the tussock. Bingo. In a vast expanse of grassland filled with thousands of smells, Mason had detected the scent of rarity: he had sniffed out the droppings of a giant anteater. Carly could barely contain her excitement. Gathering herself, she placed a sample of anteater dung in a vial of preservative to protect the scat sample and the precious strands of DNA it contained from further degradation. Those convinced that dogs are superior to humans praise their loyalty, good nature, and capacity for unconditional love. Scientists appreciate another canine advantage—dogs have an uncanny sense of smell, surpassed only by that of bears and, by coincidence, the giant anteater. The homely bloodhound, with the keenest nose of any dog, possesses a sense of smell 300 times more acute than that of its handlers. Bloodhounds can detect a scent nearly two weeks old, but they are a lot harder to train than Labradors.

Sniffer dogs have recently been recruited for biological field studies because they excel at locating the fecal tidings of rare mammals. Carly had invited me to join the last year of her fieldwork in her study of a group of rare South American mammals in the Cerrado. The maned wolf, jaguar, puma, giant anteater, and giant armadillo of the continent’s pampas and central savannas are vestiges of a rich Serengeti-like fauna that flourished in the Pleistocene epoch, 15,000 years ago. Today, their secretive behavior, low population densities, and ability to hide in the waist-high grass make sightings of these charismatic vertebrates quite rare. Their presence in the agricultural landscape remained an open question. The small size of the existing Cerrado parks and the wide-ranging nature of these species probably meant that some of them lurked out there in the ranchlands as well.

Field biologists who study the habitat use of rare mammals look for any sign: a scrape, a footprint, or the unexpectedly precious gift, a dropping. Miraculous advances in molecular biology have enabled researchers to extract strands of DNA and hormones from animal droppings, transforming the lowly fecal deposit into a gold mine of information. A scat sample can reveal the species of the depositor, individual identity, sex, reproductive status, diet, and health. Moreover, accumulated droppings from any single species, giant anteater or jaguar, yield the most prized data of all for rarities—density, home range, and population size.

The challenge, of course, is to find the fresh material from secretive animals that are often solitary and only part-time above-grounders, as is the giant armadillo. The first three years of Carly’s study, and other studies like it, had begun to show such promise that by 2010 human-dog research teams had gone global. Scent dogs are now used to study grizzly bears, Mexican wolves, wolverines, fishers, Javan rhinos, Indochinese tigers, Amur tigers, and other secretive mammals. The roots of this booming human-canid collaboration, however, trace back to an animal shelter outside Seattle, Washington.


If popular belief grants cats nine lives, dogs most certainly deserve at least two or three. Heath Smith told me that he had no opinion on animal karma when he entered the local shelter in Enumclaw, Washington, on a gray day in March 2004. He was not on a mercy expedition. Once inside the pound, Smith began bouncing a tennis ball off the concrete floor. Some dogs wagged their tails and rose to lick the visitor’s hand but ignored the ball. A long-legged black Labrador pressed against the edge of his cage. He ignored the stranger but fixated on the delightful object Smith tossed in the air. The dog trembled with excitement and panted heavily. Here on display was precisely the kind of overwrought behavior bound to discourage even bighearted adopters. The shelter employee offered some details. “He was picked up running along the highway; must have gotten lost while hunting. You sure that’s the one you want?” Heath nodded after taking a minute to confer with his boss by cell phone. “Great,” answered the relieved keeper, “because he had twenty-four hours left.”

Outside the pound, Heath escorted Mason into the cab of his pickup for the drive home. He turned onto I-90 and made for the house of his employer-landlord, Sam Wasser, a renowned conservation biologist at the University of Washington in Seattle. Heath was an accomplished dog handler, and Mason was one of the first candidates to become a new breed of working canine. The dog seemed delighted to be out of the kennel and most interested in the tennis ball hidden in the pocket of the person he was now accompanying. So began an improbable journey, from incarcerated stray on death row to professional scent dog set loose in the Brazilian outback.

Carly was one of Sam’s students. She decided to try out scent dogs in her pilot study site, Emas National Park, in south-central Brazil, home to all of the Cerrado’s largest vertebrates. “Ema” is the Brazilian name for the greater rhea, an ostrich relative that is common there. So are many other species that could be dangerous to a domestic animal bounding across the wild Cerrado. Having been trained to heel was vital if a dog were to stumble upon white-lipped peccaries, for example, which could tear a dog to shreds. The Cerrado is also home to fifty-three species of snakes. Carly wore snake guards to protect herself from the poisonous ones. Vipers, rattlesnakes, and fer-de-lances are all capable of killing a canine in one deadly strike. She was worried about unexpected encounters with anacondas, too. So she carried a machete in case there was a confrontation.

Protected areas are the best places to sniff out giant anteaters and giant armadillos in the Cerrado. Most grassland and dry forest mammals of South America are essentially rain forest species that have adapted to residing in seasonal forests and savannas. Thus, their extinction in the Cerrado may not spell the end of the species because others of their kind may still lurk in numbers in the forest. The maned wolf is an exception. Biologists call it an “obligate” grassland species because it cannot survive in the forest, having adapted, like the lion, to hunt its prey in open country.

Parks such as Emas, which is 1,320 square kilometers in size, are established to protect rare species, and they often perform well for habitat specialists and global rarities that live at high densities in small areas. The wandering kind of rarities, such as the maned wolf and giant anteater, that live at low densities can be much harder to protect. Outside of the deep Amazon, their world is changing by the minute. Circumstantial evidence shows that many rare tropical carnivores disappear when they leave the safe confines of their reserves.

The purpose of Carly’s study was to determine how wide-ranging, low-density species such as maned wolves, jaguars, and pumas navigate the countryside and to learn how they circulate among the highly altered and fragmented habitats outside reserves. She also wanted to learn whether these wild species can shift from living in natural grasslands and adapt to feeding in soy fields and cattle ranches. Finally, using maned wolves as a test case, she wanted to learn if animals that live in the Ag are more stressed, less healthy, or less reproductively active than individuals that live inside the park. New techniques that Sam and Carly were developing back at the lab allowed them to measure health from hormones extracted from the dung. If Carly and Mason could find enough scats of this grassland wolf, this charismatic species might be the most revealing one to study.

Biologists use the terms “source” and “sink” to define the dynamic at work here in the Cerrado. Source sites are places where recruitment, or population increase, exceeds mortality. The  expectation is that maned wolves have greater breeding success inside Emas than in Ag land, but because the park is too small to hold all the individuals born in the reserve, breeding individuals must spill out into the surrounding landscape. Sinks present the opposite situation, in which maned wolves (or members of another species) emigrate from the source to a location where they die in higher numbers than are recruited into the local population. With not enough large, preferably linked sources, and too many sinks, the species eventually will die off.

Some biologists avoid following their species into farmland or altered habitats. But in order to answer her questions, Carly randomly selected sampling routes that wound through the park as well as an additional 4,000 square kilometers of soybean fields, cattle ranches, and forest fragments on adjacent private lands. Curious about this woman and her dog, the local ranchers agreed to allow her to roam freely.

Within weeks after arriving in Emas, Mason had made the Cerrado his home, adapting well from a stint in the bitter cold of the Canadian Rockies, where he had worked on wolf, caribou, and moose scats. Now he was in the dry oven of the Cerrado and was mastering Cerrado mammal spoor: puma, jaguar, giant anteater, giant armadillo, and maned wolf. To train Mason on tropical mammals, Carly had obtained sample scats from zoos in the United States and from Leandro Silveira, the dean of Brazilian carnivore biologists, and his biologist spouse, Anah Tereza de Almeida Jacomo, whose home base was Emas.

When Carly told her friends about her project, they all expressed admiration for the maned wolf. The maned wolf, though, it should be said, is not actually a wolf at all, bearing no relation to the gray wolf. The vivid red fur, trimmed in black and white, and the lovely mane of this Cerrado carnivore make it one of the most striking of mammals. The showstopper is to watch a maned wolf‘s aristocratic gait—a smooth trot on its strikingly long legs that matches the elegance of its coat. Nothing else on nature’s runway compares to this handsome wild canid.

If the maned wolf is not a wolf, and the giant anteater an evolutionary oddity, no mammal is as strange as a giant armadillo. Its armorlike plating and bullet-train shape make it the perfect inspiration for a futuristic subterranean vehicle. Its huge claws, designed for digging its way through the underworld, are also impressive. Few biologists had ever studied this species before, and those who did mostly focused on its burrowing behavior and diet. Few had even found one alive, with or without the help of scent dogs. George Powell, our jaguar tracker from Peru (chapter 3), told me that once his research staff had heard loud snoring sounds coming from a burrow. They crawled in and pulled out a sleeping giant armadillo. Leandro Silveira also had an anecdotal encounter. But that was the extent of giant armadillo natural history—a few paragraphs.

These three rare mammals can all be described as unusual looking. So one might wonder if odd body plan, at least to human observers, is a predictor or correlate of rarity. I asked Carly about the link between countenance and ecology of giant armadillos, giant anteaters, and maned wolves and what it might tell us about rarity. Her answers had as much to do with the energetic balance of these mammals as with their appearance.

“Even though these creatures eat different things, they specialize on food items that just don’t support high densities of big mammals,” she responded. This, not appearance, is key. Giant anteaters and giant armadillos are equipped to feed on ants and termites. While their food resource is ubiquitous, it is also of low nutritional quality (with some exceptions, such as fat-filled winged termites). The physiological and behavioral adaptations of these animals to their food resources may account for both their looks and their rarity in nature.

Giant anteaters have low metabolic rates relative to their body size. One consequence of their slo-mo lifestyle is that they produce only a single offspring at a time and only every other year. They occur at their highest-known densities in grasslands where their food resources are concentrated. But even here, anteater populations are severely constrained by the wildfires that regularly burn through. So even where their favorite foods—ants and termites—are highly concentrated, giant anteater populations are often knocked back by killer fires racing across the pampas.

Maned wolves, which are restricted to the grasslands of central South America, also occur at naturally low densities, half of our two-part condition of rarity. The maned wolf is the largest canid that does not hunt prey larger than itself. Its body mass may well be the limit at which a canid is able to survive on small prey—primarily rodents, birds, and armadillos, heavily supplemented with fruit. To meet their dietary demands, maned wolves traditionally forage across large home ranges of about 70 square kilometers, thus contributing to their natural rarity.

These unique adaptations—the elongated, toothless skull of the giant anteater that accommodates its extra-long ant-lapping tongue; the large claws of the burrowing giant armadillo, used for ripping open termite mounds; the fox-on-stilts appearance of the maned wolf—can be viewed in a new light. These striking features are all the evolutionary products of highly specialized feeding behaviors. Might they be unfavorable attributes in a changing world? An intense specialization on a highly patchy food source, such as termite mounds, works only if the species in question—in this case an anteater—can move effectively between patches. Thus, if the species is to thrive here in the Cerrado, surrounding ranchers would need to keep the termite mounds in their cattle pastures and anteaters would need corridors to reach these patches of rich termite concentrations. The long legs of the maned wolf might allow it to travel long distances easily and pounce effectively on its abundant prey. Being a medium-sized predator but basically subsisting on abundant small mammals such as rats means that you could find your principal food source in any open habitat. Natural selection may have no foresight, but it seems to have left the maned wolf a better chance of survival than the other members of the Cerrado trio.

Before heading out in the morning, Carly checked her data sheets and GPS unit and strapped on her snake guards. She counted the bottles of drinking water on hand for us and Mason, and we set out. When we reached her starting point on the transect, Carly released Mason with the command “Let’s go to work!” Nose and tail in the air—the posture of a skilled scent dog—Mason trotted off into the grass.

The scent dog weaved back and forth across the route, off leash but always within sight of Carly. Within minutes, he came running back to fetch us. Then he raced back into a grassy area under some trees and sat down with his nose pointing a few inches away from a pile of maned wolf dung. Despite being an excellent fieldworker, Carly admitted she would have walked right past this scat had she been on her own. She immediately praised the dog and set about collecting the scat and noting its location on her GPS unit. The Cerrado project marked only the second time scent dogs had been called to duty outside the United States and Canada and the first under the tropical sun. By the end of a six-week pilot study here in 2004, three dog-handler teams had collected more than 650 scat samples from pumas, jaguars, and maned wolves. Their initial success made them believe that this novel technique could work well in the hot Cerrado.

We stopped for a break so Carly could give Mason a drink of water. The bells on his collar jingled as he lapped up the water from his bowl. The bells were a holdover from his grizzly work, a safety precaution designed to warn grizzlies that the team and Mason were near. Here in the Cerrado, Mason had wakened sleeping giant anteaters, and the nearsighted creatures merely ignored him and moved on as Mason retreated. He had also roused tapirs dozing in the grasslands, and Mason gave them a wide berth. Once he came face-to-face with a coiled rattlesnake, but Carly gently called him back to her side and the curious dog left the snake alone.

Here, the main reason for the bells was to warn off herds of peccaries. Only two months earlier, Mason had had a potentially fatal encounter. He was plowing through the grass and ran into a large gang of fierce peccaries. One turned and attacked, slashing Mason across his rump. Fortunately, Carly was close by and rushed Mason to the vet. After a few weeks of rest, Mason was ready to return to action, more wary of peccaries than ever before.

When we returned to camp, Carly took out Mason’s bowl and fed him his ration. Rather than put him back in his holding crate for his afternoon siesta, she left him on the porch, tethered to a post by a long chain. Out of the nearby forest came a female black-and-white curassow. This heavy-bodied, sharp-clawed ground dweller was a favorite of local hunters but not a bird to mess with. She made her way over to where Mason was resting, but rather than claw at him as a possible predator, she snuggled close. According to Carly, this had become a daily routine. The male black-and-white curassow is all black, like Mason, and she may have seen him as a larger version of a possible mate. A female curassow in love, but one with a mean streak. This same female had run down one of Leandro’s roosters and killed it.

The next morning we left Mason and his curassow flame snuggling on the porch while we toured Emas by car. We drove a long way to the northern border of the park but saw no anteaters or rheas. Their absence surprised me because we came upon huge numbers of termite mounds sticking up in odd funnel-like shapes. The mystery of the missing anteaters had a gruesome explanation. A catastrophic grassland fire had swept across Emas in 2005. The long fur of the anteaters had turned them into panicked torches, and five years later the population had yet to recover, in part because of their slow breeding pattern.

A central tenet of park design is to create reserves large enough to allow wildlife populations and natural processes, such as fire, to fluctuate naturally, with little or no human intervention. In this case, Emas would need to be several times larger than the area burned by the worst grassland fire of the century. An alternative design scenario would require that Emas be well connected by habitat corridors to other anteater reserves, to maintain a resupply route if a population inside a reserve is decimated by fire, poaching, or disease.

Somewhat discouraged by the lack of wildlife, we headed for lunch at a canteen attached to a run-down pool hall on the park’s outskirts. Over a meal of rice, platanos, and beefsteak, I mentioned that my Serra da Canastra birding trip, with its added sightings of maned wolves and giant anteaters, had sparked the collective curiosity of the van passengers. Together with Wes Sechrest, then head of the Global Mammal Assessment project of the International Union for Conservation of Nature and Natural Resources, John Morrison, David Wilcove, and I wondered: How many places on Earth still support the same roster of large mammals that were present there 500 years ago? Is Emas one of them? The question has an important relationship to rarity because 39 percent of large mammals with body mass greater than 20 kilograms (a maned wolf weighs in at around 23 kilograms) are considered threatened with extinction, as compared with 25 percent of mammals overall. Theory and lots of empirical data tell us that bigger mammals tend to be more wide-ranging than smaller mammals, and as George Powell had shown in Peru, most parks are too small to support them. With hunting of large mammals common almost everywhere today, the answer to our question, we speculated, could well be zero or at best very few places left on Earth with intact large-mammal faunas.

But the results of our research on the topic were more optimistic than we expected. The article we published in the Journal of Mammalogy in 2007 reported that 130 places on Earth serve as large-mammal refuges and that they support the full roster of big mammals that lived there 500 years ago. These refuges fell into two categories. One category included the most inhospitable places on Earth, places that were too cold, too damp, too dry, too humid, or too remote for humans to develop. This group featured vast regions of Siberia, northern Canada, Alaska, and the Amazon and Congo basins. The other group included a much smaller set of places, including Emas—remnants still afloat in human-dominated landscapes. We made no claims about whether the species survived at carrying capacity in these places—often they were present at much lower numbers or densities than in the past. It was clear, though, that places such as Emas would need intensive management to maintain the rich large-mammal fauna still present.

After lunch and a short drive, we arrived at a soy plantation that bordered the park. We stepped out of the car and into a completely different environment. Walking among the neat rows of soy, I asked Carly how many planters grew soybeans on farms adjacent to Emas. Gauging by my experience with rice cultivation in Nepal, I expected the answer to be in the hundreds. I was off by two orders of magnitude. Carly held up one finger. “This 40,000-hectare ranch is owned by one person. There probably aren’t more than a few landowning entities in this entire area.”

Big soy. It was my first encounter with such a vast expanse of agriculture. The lucrative plants covered the entire landscape. Corn grows here also, and in some parts of the Cerrado cotton is added to the rotation. The soybean is a recent addition to the farming economy of the region. Glycine max, as it is known to science, might have been more aptly named Glycine “min” until a few years ago, when crop and soil scientists figured out that the addition of lime to reduce soil acidity enabled the conversion of pasture and cattle ranching to soybean cultivation. As a consequence, the nitrogen-fixing legume began to prosper in areas where it could not grow before.

Brazilian farming practices changed almost overnight. Brazil has become the second-largest exporter of soy, after the United States. Much of it is exported as soybean oil, the most widely used cooking oil, or soybean meal, which has many uses but largely is fed to cattle. Producing high yields requires extensive use of fungicides, which are applied every few days. The corn grown in the next pasture was subjected to heavy doses of Roundup to control weeds. Pesticide use was rampant. White-tailed hawks and some kestrels flew by, but raptors and other birds were few. The agrotoxins may have already thinned their numbers. Perhaps a rumored decline in the bird fauna here, however anecdotal, is an early warning signal as to what lies ahead for surviving Cerrado wildlife.

That night at dinner with Leandro Silveira, Anah, and their team of researchers, the conversation focused on big mammals, beginning with rhinos and elephants in Asia and Africa.  “Our biggest herbivore in Brazil is now the tapir,” Leandro noted, “small by megaherbivore standards.” A tapir is about the size of a large pony, with a pig-shaped body. Leandro was pointing out one of the great anomalies of nature. Even though there are more mammal species in South America than anywhere else, large mammals are even rarer here than in other parts of the world. In truth, most South American mammals can fit inside a shoe box. Of course, there are important exceptions that are considerably bigger—the larger primates, peccaries, capybaras and other large rodents, deer, and larger predators—all of which are well documented as having a major impact on their surroundings.

It was not always so. Paleontologists tell us that we are simply several million years too late to witness the South American Serengeti. Back in the Pliocene epoch, about 5.3 to 2.6 million years before the present, a rich megafauna filled South American forests, savannas, and pampas. Giant ground sloths rose up on their hind legs like giraffes to browse tree branches. Swamp mammals the size of hippos and rhinos crashed through the canebrakes. Across the grasslands galloped camel-like creatures. Around the waterholes lurked long-fanged marsupials that shared a common ancestor with opossums but bore a remarkable resemblance to saber-toothed cats.

The large mammals were basically done for before humans arrived, likely as a result of climate change. During most of the Age of Mammals—when the class Mammalia first evolved about 60 million years ago, after the ebb of the dinosaurs—South America was a continental island, and its unusual mammalian fauna evolved in isolation from the fauna of other continents. The first and probably greatest wave of extinctions occurred during the Great American Interchange about 4 million years ago, when the Isthmus of Panama rose above sea level, and resulted in a massively biased colonization. North American mammals flooded into South America and are now major components of the fauna (the maned wolf among them). Many fewer South American mammals became established in North America, and today only three species—the Virginia opossum, the porcupine, and the nine-banded armadillo—still survive there.  Interestingly, two of the three species of our focus in this chapter—anteater and armadillo—are remnants of that native South American fauna.

When the renowned paleontologist George Gaylord Simpson came out with his 1980 classic Splendid Isolation: The Curious History of South American Mammals, he highlighted the divergent route South American mammals took in evolution because of physical barriers posed by the oceans and the flooded isthmus, as well as by the Andes, high deserts, and rain forests. These virtual fences prevented the mixing of the fauna. Thirty years after the publication of his work, intensive human activities including agricultural expansion, settlement, and transportation infrastructure, rather than mountain ranges or large rivers, have become the great isolators. The results, too, threaten millions of years of evolution. The IUCN Red List now includes many formerly common species that cannot persist in human-dominated landscapes, such as many antelope species and other large, hoofed mammals that used to migrate across large grasslands.

I asked Leandro and Carly to assess which of the mammals they were studying would survive in the face of expanding agriculture. I wondered if it was an ecological stretch for the grassland-loving mammals to occupy soybean fields, should landowners happen to allow it. Alternatively, I wondered whether some might be preadapted to colonize and persist in another open habitat, albeit one with a monoculture of soy, or on cattle ranches where the grass was cropped to the ground.

Carly’s scat data showed that the answer varied, depending on the species. Maned wolves preferred Emas’s grasslands and avoided closed-canopy forests inside and outside the park. Surprisingly, maned wolf scats were quite common out in the soy, however. She attributed the presence of the species there to the abundance of rodents, its dietary staple. Soy acted like natural vegetation and gave the rodents shelter. As long as they could go a-ratting, the wolves seemed to do fine in the croplands. Carly noted, “In spite of its shy nature, the maned wolf is adapting to expansion of agriculture.” Maned wolves are common in Brazil’s agriculture-dominated landscapes, at least those of soy, it seems, because they are not persecuted as is the gray wolf in northern climes. Maned wolves tend to avoid cattle ranchlands, though, as those lands are too bare even for rats.

In contrast, the giant armadillo had no use for soy plantations. Giant anteaters also avoided the agricultural fields, but they visited the cattle pastures because these appeared to have greater populations of ants and termites. Ranchers often left the termite mounds and ant nests intact, and the long-snouted opportunists showed no hesitance in excavating their dinner in man-made habitat. Giant anteaters were also more common where ranchlands abutted riparian forests. The shaggy creatures entered these forests and bathed in the water to cool down. Jaguars tended to stay inside the park but occasionally wandered into nearby fields. Pumas spent more time outside the park in areas with heavily forested groves lining rivers and streams. Tapirs liked to be near springs close to the remaining forest fragments outside the park.

Overall, the landscape matrix in this region was friendlier to rarity than most pessimistic biologists would have predicted. Some of the large mammals survived massive land-use change in the Cerrado because enough bits of natural habitat remained in the matrix to meet their needs for food and cover. How long these vital pieces would remain before becoming soy or sugarcane fields was an open question. Leandro was lobbying in favor of forever. He is one of the few muddy-boots biologists who has learned to rub elbows with lawmakers, and he spends as much time working with them as he does tracking jaguars. Leandro practices what the best conservation biologists preach, that conservation is 10 percent science and 90 percent negotiation. He had fought hard to enact the current federal law requiring landowners to keep a minimum of 20 to 30 percent of their property in natural vegetation, as well as to keep their hands off development of river- and streamside forests. Depending on topography and drainage systems, this measure potentially protects more than 35 percent of the Cerrado outside the reserves.

The Round Table on Responsible Soy Association, an effort to bring together all the big soy producers, conservation groups, and the Brazilian government, is trying to make cultivation more harmonious with nature conservation. Leandro’s first step was to convince the group that the best-practices goal of leaving 20 to 30 percent of land as natural habitat would help the maned wolf persist in the Cerrado. It would be ideal if the protected areas were in contiguous blocks connecting ranchlands in one area to those in another. Carly worried that all of the set-asides would be in forests because including grassland in the 20 to 30 percent mix of intact land would bite into the profits of the ranchers. Under the current system, the big producers would benefit most by converting to agriculture as much grassland habitat as allowed. And that is what is happening. Even with a law that sounds good on paper, legislation that is not based on strong science will have uneven results. The current law could conserve good jaguar and tapir habitat outside reserves but do little for the grassland-dependent maned wolf.

The next morning, we were out early again with Mason. We stopped at a huge hole in the ground. Here were the signature diggings of a giant armadillo, one dedicated earthmover. “There must be an ant nest or termite mound nearby,” Carly said. Giant armadillos often excavate burrows to reach under the nests of their favorite prey. Few people have ever seen a giant armadillo aboveground or in a zoo, so it’s hard to picture one. A good start would be to multiply the size of the common nine-banded armadillo by ten. The nine-banded armadillo, the unofficial state mammal of Texas, is among the most common roadkill along highways in the state. If striking a poor nine-banded would feel like rolling over a speed bump, hitting a giant armadillo, which can weigh up to 60 kilograms, would be like smashing into a retaining wall. Fortunately, giant armadillos tend to stay clear of roads in the Cerrado, though some are run over by passing vehicles anyway. Unlike the nine-banded, which ranges across much of the southern and southeastern United States all the way to Argentina and parts of the Caribbean, the giant armadillo is limited to South America. There it ranges widely into the Amazon basin, where very little is known about its habits and needs. Most likely, though, giant armadillos, which are both rare and considered by the IUCN to be threatened, thrive in the drier portions of the range, in Venezuela, the Cerrado, and the Chaco region, which extends into Paraguay.

Carly and Mason found nearly sixty scats of this subterranean mammal. Together with the data from Leandro’s team, which was the first to radio-track the species, giant armadillo ecology moved beyond the anecdotal phase. Even though tracked individuals proved to be largely nocturnal, this armadillo turned out to be the most sensitive of the large mammals to human disturbance. Not surprisingly, it shuns life in the soy fields. Leandro and his coworkers found that the home ranges of five individuals were about 10 square kilometers and put the density of armadillos in Emas proper at about 3.4 animals per 100 square kilometers, more akin to the densities of a top predator such as a jaguar than of a species that eats more abundant ants and termites. “Just like the anteaters, giant armadillos also have low birthrates,” Carly mentioned. In the scientific literature, low reproductive rate is not itself viewed as an initial cause of rarity, but as we saw with the greater one-horned rhino, serious depletion of a population hampers rapid recovery and can leave the species more vulnerable to extinction.

At lunch, Carly and I discussed the peculiar habitats of the guild of ant- and termite-loving mammals. Mammals that subsist on the social insects are few in number and restricted to the tropics and subtropics, but their taxonomic spread is remarkable. Biologists call different groups that feed in a similar way an example of “convergence”; the various species that fill the termite-eating niche (or any analogous niche) are said to be “ecological equivalents.” In South America, anteaters and armadillos, of the orders Pilosa and Cingulata, respectively, are examples. Africa is home to the aardvark, the sole living member of the order Tubulidentata; the aardwolf, a hyena relative; and pangolins, or scaly anteaters, also in their own order, Pholidota. Asia also harbors pangolins and has the sloth bear, a true carnivore and a termite connoisseur. Australia and New Guinea have the spiny anteaters or echidnas, which represent the monotremes, and Australia’s bandicoots represent the marsupials (though the most prevalent and species-rich termite-eating animals in Australia are not mammals but lizards).

Little is known about the ecology of pangolins, aardvarks, aardwolves, and sloth bears, but evidence indicates that where there are termites, the density of termite eaters can be impressive, even if their mostly nocturnal behavior makes them hard to see. Some, such as pangolins and sloth bears, are hunted for body parts used in folk medicine, and pangolins and armadillos are hunted for meat. Some of these species, including the sloth bear and some pangolins, are considered threatened by the IUCN. Without human pressures, then, the ant- and termite-eating contingent of the mammalian fauna might be even more common.

Back at the guesthouse, I had left my sneakers near where Mason was resting on the porch with his besotted curassow. “Be careful not to leave your shoes out at night,” Carly warned. “The maned wolves steal them and chew them up for the salt.”

I needed my shoes for the drive Carly was organizing for one of my last nights on the Cerrado, a trip to spotlight tapirs, crab-eating foxes, and the most important target, the maned wolf. Carly often searched along secondary roads, a mode of travel used by maned wolves that frequented the croplands. Giant armadillos, giant anteaters, tapirs, and pumas, however, avoided roads as much as possible.

The drive through the park grasslands turned up very few animals, but once we crossed over into searching the soy, sightings increased dramatically. Crab-eating foxes were everywhere and quite approachable. Nightjars, a kind of nighthawk also known as goatsuckers, sat on the roads that had been baked by the day’s sun to stay warm in the chilly night. Carly said she had often spotlighted tapirs out in the Ag, but we saw none. More crab-eating foxes. These canids were certainly attractive, but we wanted the red fox on stilts, the maned wolf.

On the secondary road bordering the edge of Emas, Carly swung her spotlight in a wide arc. The driver slammed on the brakes, and all of us standing in the back of the open-air pickup lurched forward. A maned wolf stopped and then resumed sauntering along the road. For the next twenty minutes we motored along a road inside Emas, the wolf just outside on a parallel road bordering the soy. We were all surprised at how effortlessly the wolf kept up with the vehicle; we were moving at nearly 16 kilometers an hour, and the wolf seemed to maintain its graceful rapid walk. Finally, it slipped back into the soy. On the drive back to camp, the image of the dashing wolf stayed in my head, and it remains in my memory as vivid as ever.


Ten months later, in May 2010, I had a chance to stop in Seattle and again see Carly, who was in the last stages of completing her dissertation, and Mason, now officially retired (at the age of six) from his scat-searching expeditions. Carly had just answered one of her research questions with some final lab work on maned wolf hormones: Were the maned wolves that lived in the soy fields less healthy than those that stayed within the boundaries of Emas National Park? Carly bubbled with excitement over her findings. “The thyroid hormones from wolves in croplands are actually higher than from those in the park.” High thyroid hormone levels were a good thing, and thyroid levels are useful indicators of nutritional status. The story emerging from Carly’s work was that the maned wolves were selecting croplands for hunting rodents, partly because the hunting was easy and efficient. But there was more. Carly had discovered that the cropland wolves had higher levels of cortisol, indicating that the Ag land was also more stressful for that species to live in. So the soy fields may be an ecological trade-off for this elegant animal. Abundant food awaits in the soy, but so does the stress of leaving its natural habitat and venturing into the human landscape, where perhaps it feels more exposed and vulnerable.

We compared notes about how natural landscapes were being transformed in the places we had been working. I mentioned the vast plantations of oil palm and acacia (cultivated for pulp) I had just seen on Borneo and Sumatra. “Remember the soy plantation we walked through looking for wolf scat?” Carly interjected. “In one year, it changed over to sugarcane.”

Does it matter to rare grassland mammals if the agricultural expansion involves sugarcane, soy, tapioca, or cacao? To the Cerrado trio, the answer is yes. The conversion of low-growing soy to sugarcane will shake things up ecologically, Carly explained. With the soy, there is a relatively welcoming landscape matrix for the wild animals in the region. “But when converted to sugarcane, it becomes inhospitable to an open vegetation–adapted species like my wolves.” Compared with soy, sugarcane is a longer-growing crop between harvest periods, and it forms tall, dark stands, which entails loss of the open-landscape qualities mimicked by soy fields that are conducive to species such as the maned wolf. Sugarcane also requires more manual cultivation, and factories may be built to handle the cane on-site, so it could become a magnet to draw in workers. Carly didn’t sound hopeful. “I see conflict as animals try to avoid humans, and as poachers seek out the wildlife that wanders into the Ag.”

In Sumatra, I had spoken to David McLaughlin, an agricultural expert for the World Wildlife Fund who had worked in tropical agriculture all over the world. I asked him how agribusiness can maintain, year after year, the commodities we use without tropical nature evolving innovative pathogens and leaf pests to attack the widespread cultivated plants. Every crop plant has legions of pests, some minor, some fatal if not controlled, he replied. “For your oil palm, it’s lethal spear rot, found in South America in the 1980s but not yet present in Southeast Asia. Cacao trees have all sorts of things that can hit them. And tropical soy needs constant vigilance against fungi and herbivores.”

“So,” I continued, “you mean if some careless biologist introduced a pathogen into a soy or oil palm plantation, it could wipe out the whole place and native vegetation would return?” David eyed me suspiciously. What Carly had observed around Emas foreshadowed his answer. “Remember, commodity production can turn on a dime. Even if a crop pest attacked soy, farmers could shift immediately to growing sugarcane for biofuels. And if the sugarcane succumbed, they might go back to cattle ranching or corn.” The point is that land once converted to agriculture will likely never go back to native vegetation but rather will be converted to the next big monoculture. The same holds true in the vast oil palm developments in Southeast Asia. Those that fall to some new pathogen will probably be replaced by acacia trees for pulpwood or return to rubber. The commodities are never stable, but the pressure applied by the growing human footprint is mounting. In the meantime, programs designed to enhance wildlife-friendly certification of major cash crops—whereby consumers know that production of the crop has avoided the conversion of natural habitat to produce it—is still years away.

The Cerrado story helps us put countryside biogeography, or matrix conservation—or whatever term we create to describe the landscape approach to conservation that considers species both in parks and in the human-dominated areas outside of formal protection—into a larger context. Many species that do fine in wild habitats can persist in adjacent farmland dotted with pockets of natural vegetation. This is especially the case where riparian zones are protected from development. But for how long? Perhaps our assumption that various species will persist in agricultural lands is an ecological mirage: it may be that they will merely hang on for another decade or two and then start to decline as a result of pesticide residues, stress, or other sources of mortality. Furthermore, species that survive in human-dominated landscapes tend to be generalists, not the specialists or the rarities that this book portrays. Carly believes that upholding the Forest Code of Brazil is key, that it is these scattered remnant habitats that enable species to use the landscape as a whole. The future challenge of agroecology is to identify the opportunities for mutual accommodation and its limits—in the Cerrado and elsewhere as well.

The scale at which the face of Earth is being converted from natural habitats to cultivation and the pace of it are truly staggering. Global projections are that with 9 billion people on board Earth, world food demand will double by 2050, and the Cerrado will play a key role. Yet most people are unaware of the enormity of agriculturalization and other habitat encroachment. Various small-scale local accommodations, even if increased in number, may not be enough to protect wild nature. A hidden aspect of this problem is the scale of buy-ups of land in the tropics and elsewhere by corporations and nations as they position themselves for the future of their food supplies.

So the problem grows larger: wholesale conversion of land not only threatens to make no small number of common species rare through human activity; it also threatens the very existence of what is now rare. One has to hope that the rare species of the Cerrado and other areas of intensive cultivation are more adaptable than we think and that efforts to enable coexistence on working lands, by habitat protection and then by best practices on and near agricultural lands, will enable persistence. It would be a shame for others to lose the chance to observe the remarkable silhouette of a giant anteater or the outline of a graceful maned wolf on a moonlit night in the Cerrado.