One chilly October morning, Beth Cheever hopped out of an aluminum boat. In rubber boots, a life jacket, and a knit hat pulled down over her ears, she walked the portage trail, beneath denuded alders and paper birches damp with the previous night’s rain, to the granite shoreline. She had never poisoned a lake before. Yet the thirty-two-year-old ecologist from New Hampshire had driven her Dodge Caravan twenty-two hours from Trent University in Peterborough, Ontario, to this corner of northwestern Ontario, just thirty minutes from the Manitoba border, with a plan to do just that. Spruces guarded the glassine pool’s edges like stoic sentries, and signs posted all around told wayward anglers to keep their lines out of the water. Lake 221 contained the beginnings of an experiment, a study of what could go wrong when the team laced an entire lake with antimicrobial compounds — deliberately, with the utmost precision.

Nanosilver kills microbial life, and, as the “nano-” in its name suggests, the antibacterial battle takes place in minutiae, each particle so small that a million of them could fit on the period at the end of this sentence. Nanoparticles have applications in technology, medicine, and agriculture. As Cheever’s post-doctoral supervisor, Maggie Xenopoulos, an aquatic biologist at Trent, said to me earlier, “They’re the future, and yet we have no idea if they’re affecting our environment.” In one laboratory study, scientists found cranio-facial deformities in minnows exposed as embryos to high concentrations for ninety-six hours. But lab studies only show so much; a beaker or bottle experiment does not necessarily reflect the complexities of an entire lake. If nanosilver killed off too many species or a key component in the web of life, the whole ecosystem might malfunction and collapse. To learn what happens in situ, Cheever’s team intended to spend two years sending an infinitesimal galaxy of particles into the lake.

Lake 221 lies within the bounds of the Experimental Lakes Area, a field site familiar to readers of the Proceedings of the National Academy of Sciences, Limnology and Oceanography and the Journal of Plankton Research. Since 1968, fifty-eight freshwater basins contained by 200 square kilometres of granite and boreal forest have functioned as real-world test tubes, untouched by human hands except by scientific design. Lakes 226 and 227 demonstrated that phosphorus led to algal pollution, which convinced politicians to mandate the reformulation of detergents. Sulphuric acid caused a dramatic shift in species in Lake 223, bringing about international emissions limits to address acid rain. Long-term data has shown the lakes to be early warning sentinels for climate change. More recently, in an experiment on Lake 658, scientists demonstrated how mercury accumulated in fish; and on Lake 260, scientist Karen Kidd identified a single chemical, the synthetic estrogen in birth control pills, as the cause of mass feminization in male fish and a cataclysmic population crash. Anywhere else, such massive die-offs might have resulted from confounding factors: human activity, industrial effluent, or any number of synthetic organic compounds found in pharmaceuticals and personal care products. But the studies at the lake pinpointed cause and effect more decisively. They have drawn generations of scientists and students to the boreal shield, like pilgrims to the holy waters of ecological research. Cheever made her first pilgrimage in 2012.

Earlier that morning, she had motored across six lakes with two students — Jenn Vincent and Kat Cetinic — in tow, changing boats four times, sometimes relying on a puttering outboard to reach the other side. The scientists had spent the summer planting leaf packs in eight lakes, including 221, to establish a baseline for the experiment. Cheever served as a scientist, teacher, and matter-of-fact wilderness guide. When asked about a woodpecker drumming against a tree, she said, “I don’t know. I just put a box around it and call it a bird. I’m an ecosystems ecologist.” She was looking for macroscopic changes, the massive fluxes and the rates of exchange, the patterns in an ecosystem.

Vincent sloshed through the water with a pair of wire cutters while Cheever stood on the shoreline. Cheever called out, “Where’s the leaf pack?” Vincent shrugged before exhuming a black mesh bag resembling a waterlogged mud sandwich, and dropped the wad of decomposing maple leaves into a zip-lock bag. Wet maple leaves provide an index for all sorts of activity — physical, chemical, and biological: the algae and fungi and microbes that cycle phosphorus and nitrogen, which feed fish scavenged by birds that, upon dying, become fodder for bacteria and worms, all living off the sun’s infernal energy 150 million kilometres away. Study the leaves, and the biological data would explain what Lake 221 looked like before the team tainted the water with bottles of lab-grade nanosilver — if they ever managed to taint it. Back at the field station, a Canadian flag flew at half-mast, and no one could say with certainty whether the experiments would continue in the spring, or whether the team would ever determine their effects.

Cheever slid the boat up onto the shoreline and all of a sudden lost her footing. “If we do anything next year,” she said, “we’ll build a dock.”

Around eleven o’clock, the team headed back to camp, descending a trail worn down by hundreds of feet over the past forty years. Cetinic held out a backpack full of leaf samples as if they contained some kind of poison. “Why don’t you wear it like a backpack?” asked Vincent.

Not because it’s poison, she said: “It’s leaking.”

They puttered back across Roddy Lake, heads tucked down in the chilly wind, for leftover pulled pork and baked potatoes in Hungry Hall, a wooden building at the centre of camp. Overnight temperatures had fallen close to zero, and the warm blanket of water nearest the surface, the epilimnion, would sink as the water in the lakes turned over. “People start to leave once the lakes turn over,” Cheever told me. “The winds pick up.”

Six months earlier on May 17, 2012, at 7:55 am, as Cheever’s colleagues headed out from the lab for a day in the field, the phone rang. Michelle Wheatley, regional director for the Department of Fisheries and Oceans, the Experimental Lakes Area’s federal overseer, called a mandatory emergency teleconference for the half-dozen full-time employees who happened to be at the field station that day (it employs seventeen staff in total). The government scientists dropped their packs and rain gear and gathered around a scratched wooden table inside the library of Hungry Hall. Wheatley told them the ELA would be closing within a year and their services would no longer be required. The meeting lasted forty-five minutes and ended with tears.

The department made no official announcement. In an apparent effort to avoid a paper trail, all communication about the closure took place verbally, over the phone or in face-to-face meetings. The shutdown had been precipitated by cuts proposed in Bill C-38, the Conservatives’ federal omnibus Budget Implementation Act. Buried within 400 pages of amendments and billions of dollars worth of cuts was a provision that effectively pulled the plug on the ELA’s $2 million in federal funding. In “affected” letters sent to federal employees to notify them that their positions might be eliminated, the department explained that the closure reflected “the government’s efforts to reduce the deficit, aimed to modernize government, to make it easier for Canadians to deal with government, and to right-size the costs of operations and program delivery.” The department’s research needs would be met by other facilities, wrote David Burden, the official who sent the letters. He did not specify which facilities; nor did he respond to a request for additional comment.

One scientist, who asked for anonymity after being instructed not to speak with the media without prior written consent, said, “People say these cuts were made at a high level outside of the department, which is likely true, but at some point they were offered up. People higher up would have no idea what the ELA.” The source wondered if the motivation could be political: “The bulk of the cuts to scientific research programs come in the Prairie and Arctic regions, which have the most industrial development; the new Ring of Fire, the oil sands, huge industrial projects. It doesn’t quite add up.”

Federal funding had kept the boats and the kitchen and the entire facility operational. It paid the full-time employees’ salaries, but an additional $300,000 to $1 million in annual grants underwrote the actual experiments. Because the monies came from university collaborators, federal agencies in Canada and the United States, and industry groups such as Manitoba Hydro and the Northern Ontario Aquaculture Association, they rarely covered long-term monitoring. Cinching off federal financial support all but guaranteed that over four decades’ worth of continuous data would be lost.

As news of the closure spread, an intercontinental chorus of scientists rose to speak in the ELA’s defence. The impassioned global response contrasted sharply with the government’s oblique rationale. Cynthia Gilmour, a scientist at the Smithsonian Environmental Research Center outside Washington, DC, spent ten summers working on Lake 658. “It’s crazy, crazy to shut down the ELA,” she said. “Such a small amount of money for a huge impact.” Another ecologist, Stephen Carpenter at the University of Wisconsin–Madison, said, “Politicians are thinking about the next election cycle or what’s going to happen on some brief time scale, but they’re making decisions about things that operate over very long time horizons. Monitoring ecosystems is cheap, compared to the rest of things societies do. To throw away a site with decades of value is incredibly short sighted.”

Bill C-38 became law on June 29, 2012. The following month, more than a thousand protesters, many wearing white lab coats, marched on Parliament Hill. It was the summer of scientific discontent. A few pallbearers carried a coffin. Signs announced the passing: The Death of Evidence. Another foreign observer, a Swedish limnologist named Ragnar Elmgren, was even more pointed: “This is the kind of act one expects from the Taliban in Afghanistan, not from the government of a civilized and educated nation.” But the Department of Fisheries and Oceans had already changed tack. Rather than close down the camp and all of the research, it now intended to find a new operator, wrote David Gillis, a department official in charge of negotiating a transfer. Information about such efforts was scant, fuelling doubts about the sincerity of the DFO’s intentions. In one letter to department officials, Jules Blais, president of the Society of Canadian Limnologists, wrote, “The inescapable conclusion is that your government has no real interest in transferring the site to an alternate operator, assertions to the contrary notwithstanding.”

For over a year, beginning several months before the cuts were first proposed, I had been asking to visit the lakes. Finally, department spokesperson (and sender of “Approved Messages”) Rachelle Smith responded that any visiting journalists would disrupt ongoing confidential negotiations with a party that neither she nor anyone in the department could name publicly. “We’re looking toward the future,” she said. “There’s work to prepare a camp for shutdown.”

The rote refusal followed a more general pattern. As an editorial in the March 1, 2012, issue of Nature reported, “Policy directives and e-mails obtained from the [Canadian] government through freedom of information reveal a confused and Byzantine approach to the press, prioritizing message control and showing little understanding of the importance of the free flow of scientific knowledge.” By January 2013, the Ottawa-based advocacy group Democracy Watch had launched an open government campaign, calling for an end to muzzling scientists, as the organization dubbed the increasingly widespread tactic: preventing government scientists from speaking freely to the public about their work.

When I flew to Winnipeg in late October, I knew just one thing for certain, that the national laboratory I had come to visit was off limits, with only three ways to get in: by helicopter (an expensive proposition); by canoe (a strenuous journey that took days); or with an official road travel permit, already formally denied to me. Fear, I was told, had permeated the lowest levels of government bureaucracy. Meanwhile, at the ELA headquarters an icy twinkle of frost covered the Freshwater Institute’s campus lawn; that did not stop a few maintenance workers from mowing the frozen grey-brown prairie. The facility’s double doors were locked, and I was only allowed in for a surreptitious look around. It was unclear what was being kept secret.

I continued two and a half hours to the east, toward Kenora, Ontario, where Cheever had agreed to meet me so I could travel under her road permit. As we drove the thirty kilometres into camp, around skittering turns and teeth-jarring inclines, I expected to see fences, wardens, hunters, or maybe even an expedition of oil and uranium prospectors, given all of the rumours flying about. Instead, I found myself in a muted evergreen forest, far removed from the Trans-Canada Highway and the smells of woodsmoke. The roadside maples, goldenrod, and asters had been drained of their Indian summer colours, and the pretty little lakes we passed — 442… 626… 373 — appeared, like any slice of ecological time, quite ordinary and meaningless when considered alone. Any experiment conducted there only made sense as a larger accumulation of small things. However remote the place seemed, the lakes’ fate was inextricably bound up in the pollution and politics of the outside world. The lakes contained universals, and the experiments conducted there forced us look at their results on a global scale.

The Experimental Lakes were formed nearly 10,000 years ago, on the shallow eastern shore of Lake Agassiz, a vast, irregular basin that once covered the middle of the continent. As the glacial waters receded, they left Lake Winnipeg and the Lake of the Woods; and the Red River Valley, which drains the broad, flat plains northward into Hudson Bay. Eons later, in 1887, far beyond what would have been the river’s southernmost banks, at a scientific meeting in Peoria, Illinois, an entomologist named Stephen Forbes became one of the first naturalists in North America to give a semi-coherent account of a living laboratory. A lake, he wrote, “forms a little world within itself — a microcosm within which all the elemental forces are at work and the play of life goes on in full, but on so small a scale as to bring it easily within the mental grasp.” The idea of lakes’ reflecting the outside world and being tethered to it foreshadowed the discipline of ecology, the study of relationships between things both living and not. Stephen Carpenter, the Stephen Alfred Forbes Professor of Zoology at the University of Wisconsin, explains the discipline’s many challenges this way: “Ecology is not rocket science,” he writes. “It is far more difficult.” To build a rocket entails a certain technical prowess and a society’s worth of resources. To understand an ecosystem requires a patience and persistence similar to that of raising a child: both are autonomous beings, evolving, adapting — and actively not doing what you want.

In 1965, a complex ecological problem manifested itself at the very heart of the continent: Lake Erie was being choked by explosive algae growth, a tangle of phytoplankton that looked like seaweed and smelled of dead fish. An international commission set out to find the source of the pollution, plumbing the continent’s pipes and weighing the cumulative effects of every sanitized whoosh of toilet, every rinse of sparkling-clean glass, every asphyxiating whir of the garburator, every field covered in fertilizer. The Department of Fisheries (its policy then set by scientists at the Fisheries Research Board of Canada, a Crown corporation, not a government department) appointed scientists Jack Vallentyne and Wally Johnson to establish the Freshwater Institute in Winnipeg. Johnson had previously experimented on lakes in Wisconsin, and he believed studies across an entire ecosystem could provide an effective way to convince policy-makers responsible for the expensive decisions. The two pored over aerial maps from all over northern Manitoba and northern Ontario, searching for uninhabited headwaters, lakes at the farthest reaches up the pipes with limited logging, no agricultural runoff, and little groundwater. The lakes needed to have measurable watersheds, and they needed to be small, so the experiments would not be prohibitively expensive. Vallentyne and Johnson identified 1,004 lakes around Kenora, a suite of self-contained basins that could act as variables and “reference systems,” just hours from the institute’s headquarters.

In January of 1967, Vallentyne invited David Schindler, a twenty-seven-year-old junior scientist, to visit an empty cement building he called the Freshwater Institute. He offered Schindler a job, but Schindler said no. He had recently returned from the University of Oxford in England, where he had trained under Charles Elton, a kind of Che Guevara of ecology who had studied the ebb and flow of lynx and hare populations based on Hudson’s Bay Company fur catches. Schindler had also turned down jobs at Yale and the University of Michigan because of their proximity to bright city lights — “a bit much,” he explained, “for a hillbilly from Minnesota.” Instead, he chose to teach at Trent University.

Lured back to Winnipeg the following year, he found that Vallentyne’s new institute, funded by government and directed by scientists, had attracted “big fish from Europe,” including a well-known zooplankton specialist from Poland and an influential limnologist from Switzerland. Canadian science was becoming the envy of the world’s freshwater ecologists, with other well-funded federal labs in Burlington, Ontario, at the Canada Centre for Inland Waters; and in Saskatoon, at the National Hydrology Research Centre. Moreover, the new field station was attractively remote — his kind of place. That May, he and Vallentyne drove from Winnipeg to investigate the lakes earmarked for study. Beyond Hillock Lake, its waters so clear they could have been distilled, where timber was still cut and floated out each spring, they found six students with two broken-down cars awaiting their arrival. “The cook was literally chasing people around with a cleaver,” Schindler said. By summer’s end, his team had identified forty-six lakes as suitable for experimentation (another twelve would be added later), and the Ontario-Minnesota Pulp and Paper Company and the Dryden Pulp and Paper Company gave them permission to use the land. “We got so much timber around here,” one official said. That fall, the new staff sat around a cramped ATCO trailer to plan the first experiment, to test the Lange-Kuentzel-Kerr hypothesis that carbon, not phosphorus, was causing the massive algal blooms. This was a theory popular with soap companies: phosphorus, of course, was the miracle ingredient in cheap synthetic detergents, keeping dishes and laundry spotless without the need for animal fats. Paul F. Derr, a chemical executive in Philadelphia, even suggested that the cause for the blooms might be unknowable, as mysterious as life itself.

At first, the group considered adding phosphorus to Lake 240, immediately south of the camp, but Schindler argued that it should be left for recreation; he preferred Lake 227, although it involved a longer haul. He became the camp’s driving force, even giving the Experimental Lakes Area its utilitarian name. A field biologist of uncompromising devotion, he has a farmer’s hands and a wrestler’s physique — a necessity in the field, where he had to portage canoes and carry outboards. It was not all drudgery: whenever possible, he would reward himself with a late-night dinner of fresh lake trout, presumably from recreational, not experimental, waters. His hiring criteria, he told me, could be distilled into a single question: “Can you lift two eighty-pound carboys?”

That winter, he and another scientist, Gregg Brunskill, drove snowmobiles over two frozen lakes, pulling homemade sleds loaded down with fertilizer. If carbon had caused Lake Erie’s algal asphyxiation, then pouring nitrogen and phosphorus into Lake 227 would do nothing, because it contained little carbon. Within weeks, it had turned into “a teeming green soup,” and Schindler went on to publish his findings in the journal Science. The air around the lake still smells of sulphur and geosmin, the earthy, metallic scent characteristic of warm rain; and in a wooden supply shack, a mirrored disco ball still hangs there, as if to celebrate the textbook experiment.

Schindler’s team soon began a follow-up study on Lake 226, stringing an impermeable nylon curtain across a narrow isthmus to divide the hourglass-shaped lake into two basins. They added carbon and nitrogen to both sides, but this time they dumped phosphorus into just one of the basins; a few months later, it had clouded over with an opaque green layer of cyanobacteria—a perfect demonstration of how phosphorus runs the game. Just as one egg will limit how much cake you can bake, no matter how large your stockpile of sugar, butter, flour, and salt, phosphorus is a limiting factor. If everyone held back on using the element — in lawn chemicals, in treated sewage, in detergents — this would, in theory, limit the explosion of algae. The lake experiments presented a clear public relations victory: aerial photographs made the results obvious. By 1973, new legislation forced detergent companies to reformulate their products.

In the ’50s and ’60s, first in Scandinavia and later in New Hampshire’s Hubbard Brook Experimental Forest, scientists had also noticed that lakes and streams were acidifying to such an extent that they could no longer support life. Evidence pointed to sulphuric acid billowing out of coal-fired power plants, but no direct causal link had been proven, leaving room for obfuscation by politicians; and smear campaigns by coal companies, which worked hard to discredit scientists like Gene Likens, who conducted a seminal study on acid rain at Hubbard Brook.

In the early ’70s, Schindler wanted to test the acid rain hypothesis at the Experimental Lakes Area. First, though, he had to convince his new bosses at Fisheries and Environment Canada; the ELA, formerly operated by the Fisheries Research Board of Canada, changed managerial hands in 1973. That development was “unquestionably one of the biggest blunders in the history of Canadian environmental science,” he wrote in his personal history of the lakes. “Instead of answering to a panel of the country’s most eminent scientists, we now reported to politicians and bureaucrats.” Jack Vallentyne had had enough and told Schindler he was retiring. “This public interaction stuff is important,” he said, “but you have to do it now.” Public relations was not Schindler’s strong suit, and his proposal to study acid rain did not go over well with the bureaucrats. Perhaps it seemed like a desperate ploy to save the field station, now that its initial purpose, to prove the effect of phosphorus, had run its course; or perhaps the department’s chief concerns, to manage the collapsing cod fisheries, were salty. So he applied to the Alberta Oil Sands Environmental Research Program for funding, and in 1976 began lugging carboys full of sulphuric acid through the forest.

The experiment ran for eight years. To the outside world, it remained a mystery and a concern, and the apparent destruction alarmed at least one local politician, who launched an inquiry into allegations that government scientists clad in boxy neoprene suits and goggles were dumping acid into a lake. John Shearer, a biologist who worked with Schindler, recalled, “Locals who love to fish and hunt got their noses out of joint a bit. Someone would find a dead fish on Lake of the Woods, and they would say, ‘Oh, it must be the scientists back in the bush who are doing that.'”

Schindler’s acid rain study, which cost hundreds of thousands of dollars (far less than the expensive and inconclusive laboratory studies being run by the US Environmental Protection Agency), demonstrated that a drop in pH — a slight acidification, at far lower concentrations than power companies admitted were a problem — could turn plump, healthy lake trout into twisted, eel-shaped creatures. Sulphuric acid did not kill fish outright, but it altered the food supply and essentially starved them to near-death. Two of Schindler’s fellow scientists, John Rudd and Carol Kelly, also discovered that bacteria found in most lakes generated alkalinity. Lakes did not need costly additions of lime to reverse acidification; to a point, they could cure themselves.

“Canadian bureaucrats didn’t like what I had to say about acid rain,” said Schindler, who was unhappy with their proposed solutions. So he flew to Washington, DC, to testify at EPA hearings. When the Department of External Affairs demanded to screen his presentations, he consulted a lawyer, who advised him to take a day’s holiday when he testified; that way, “they can’t touch you.” Schindler appeared to relish the publicity, and told reporters that when he received reprimands he would photocopy them for all of his colleagues. By 1985, for the first time in decades, a sitting American president, Ronald Reagan, travelled to Quebec to discuss the initial step in an acid rain treaty, which was eventually signed by Canada and the US in 1991.

In 1998, scientists in North America began investigating how mercury moves through the environment. Industrial activity had painted the world with chronic, low-level background doses that showed up in even the remotest lakes. As microbes transformed mercury into methylmercury, the element became a toxin that accumulated in fish; notoriously difficult to diagnose, mercury poisoning in humans comes almost entirely from fish, and is a serious public health problem that affects millions of people around the world. Reducing mercury emissions, though, could cost billions. Corporations do not spend billions on a guess.

Distinguishing the newly added mercury from ambient doses proved all but impossible, and a team of about forty scientists identified only one site — politically and logistically — where a known neurotoxin could be administered to an entire ecosystem for decades to test their hypothesis. Formally known as METAALICUS (Mercury Experiment to Assess Atmospheric Loading in Canada and the United States), the experiment was conducted in and around Lake 658.

But John Rudd, one of two chief government scientists who conceived the study, faced a problem. To trace the added mercury, he needed to procure it in three different forms, or isotopes, each with its own unique molecular weight — an expensive proposition. When an industry group in California called the Electric Power Research Institute offered up the funding, he found a lab outside of Moscow, staffed by Russian scientists who were no longer employed in plutonium manufacturing for Cold War–era nuclear warheads and appeared eager for foreign cash. Before committing millions of dollars to purchase the precision-centrifuged isotopes, though, Rudd needed permission from a review board of local residents and government officials to temporarily contaminate a virgin lake with the stuff. It was not an easy sell; the deliberations took months. “People kept saying it was impossible,” he told me. “‘You’ll never get permission to do this.’ Then one day, a letter landed on my desk saying, ‘Go ahead.’ The final permission we had to get was from the Ontario government. They were really dragging their heels.”

What finally convinced them was just how little mercury was being added: not even a half a teaspoon over five years, less than what could be found in an average lake in the northeastern United States. In June of 2001, scientists finally began applying one form of mercury, known as mercury-202, to the lake surface at dusk. The massive project also involved the experimental lake’s watershed. Scientists from the United States Geological Survey, wearing white protective suits, sprayed mercury-198 over a nearby wetland. Art Robinson, a pilot with the Canadian Forest Service, flew regular sorties in a Cessna 188 out of Vermilion Bay to mist the treetops with mercury-200, each application mimicking deposits of mercury over the landscape. The toxin was handled with the utmost care; drift would confound the results, and “if you dropped it,” said Rudd, “you couldn’t get any more. It would be hundreds of thousands of dollars’ worth of isotope.”

Almost immediately, young yellow perch, a small-bodied fish, began testing positive for mercury-202, the isotope added to the surface of the lake. By fall, predator fish had also showed signs of it. The other isotopes never reached the lake. An editorial accompanying the 2007 study, published in the Proceedings of the National Academy of Sciences, announced that the research provided unequivocal proof for the need to reduce emissions. Less than a teaspoon of mercury could have a profound effect. In 2011, after years of delays (collateral damage from what is referred to in some circles as the George W. Bush Administration’s war on science), the EPA finally enacted new limits; in early 2013, the ELA research helped convince more than 140 countries to sign a global treaty to control and reduce mercury levels.

The mercury deposited over the forest and wetland remains bound up in the soil. Cynthia Gilmour thinks it might take decades or centuries for those isotopes to reach the lake’s perch. “We don’t have a full handle on the response on the terrestrial end,” she said, “and we’re going to lose that if we lose access.” The original agreement to lease the land stipulated that the lakes had to be returned to their natural state, which was mostly interpreted as hauling cinder blocks out of the wilderness. Rudd had told me the fish were probably safe to eat again, but taking anything from Lake 658 would skew the results. Today little evidence marks the experiment’s location: a pink square, the footprint of a building that once stood on the bedrock.

The Red River meanders through St. Vital, a suburb of Winnipeg. On the eastern bank stands a log and plaster home of poteau sur sole construction, where, for two days in 1885, the body of Louis Riel lay in state. Federal budget cuts forced Parks Canada to close interpretive services at that historic site last year. Around the corner lives John Shearer, the retired biologist who worked with Schindler and took over the camp’s management when Schindler left in 1989 (he now works as a professor at the University of Alberta in Edmonton).

Shearer is a tall, balding man with bushy eyebrows, and when I arrived at his house he was standing in a tangle of prairie grasses in his front yard, which may also be Riel land. He raised a hand in greeting. He has a habit of looking altogether unsurprised — an affect that makes his response to everything that has happened in the past year seem all the more understated.

He disappeared into his home office and emerged with a stapled manila envelope containing the Journal of the Fisheries Research Board of Canada, volume 28, number 2, from 1971, then unfolded a map the size of a large-screen television. It shows the Experimental Lakes Area as a mint green wilderness, two spindly varicose veins of red roadway woven through with filaments of grey-blue, with each lake numbered as if by some random toss from the cartographer’s hand.

Shearer recalled how he became involved with the lakes when he enrolled in an elementary statistics class at Trent University and his teacher, Schindler, invited him to take core samples through the ice. Not long afterward, Shearer joined the ELA staff and bought a Sportspal canoe. “Two things got to me,” he said. “One was the water. The second was the bedrock. I can think of nothing more peaceful than sitting on a rock that’s three billion years old.” He worked at the ELA for the next thirty-eight years. “It’s like being in medicine and working at the Mayo Clinic,” he said. “You’re part of an elite, in terms of what you are able to do.”

He recalled a career’s worth of stories: A fire that leaped through the forest, coming within an eyelash of camp, and the moment when everything turned green again. The dozens of scientists who took off at dusk to wash mercury into Lake 658. He had seen die-offs and population recoveries. All but one lake had returned to its natural state, but the same could not be said for science. Our understanding of the world, as seen through the prism of the ELA’s ongoing data collecting, might never be the same again. Shearer stared off at a wall. “If they ever really needed a facility like that again,” he said, “it would take far more political will than I’ve seen in recent years.”

This past April, six months after I visited Shearer and soon after the federal government barred scientists from entering the camp, the International Institute for Sustainable Development, a non-profit based in Winnipeg, and the government of Ontario announced a reprieve: they would reopen the ELA, at least for this season. The testing would go on — even if the feds no longer deemed the experiments a national priority.

Analogues exist to make complex systems visible and scalable. The entire universe cannot be condensed inside a sprawling underground particle accelerator in Switzerland. Likewise, cinder block colonies of laboratory mice can only begin to mimic the vagaries of our own bodies. A lake’s ecosystem cannot be stuffed inside a test tube. Many of the scientists I spoke with said the ELA had come to represent another kind of analogue. One, who spoke on condition of anonymity, said the effort to slash support for basic research on fresh water represented something else. “I don’t know if it’s a stake through the heart of Canadian identity,” the source said. “I’m not sure I know what that is anymore. But it’s a stake through my heart.”

Back at the ELA chemistry lab, in front of windows that look out into the forest, Cheever opened a fridge and pulled out a foil-wrapped bottle of NanoComposix, a fifty-nanometre solution, which she opened with her bare hands. The liquid looked like water and smelled of linden blossoms mixed with fresh newsprint. “I wouldn’t jump into a pool of the stuff,” she said. “I wouldn’t drink it.” She screwed the cap back on tightly.

The antimicrobial compound presented just one unknown in a sea of unknowns. Without the suite of lakes, even fewer places existed to test the waters, to determine the effects of nanomaterials or cadmium or mercury or polycyclic aromatic hydrocarbons, or any one of the tens of thousands of novel contaminants washing into the world’s largest supply of fresh water. As with so many potential contaminants, the real-life experiments were already up and running, trickling into streams, lakes, and oceans with few to no controls. Outside of the fifty-eight little granite-lined basins in the middle of nowhere, these experiments might one day go by another name: tragic accidents.

This article was reported in partnership with The Investigative Fund at The Nation Institute, now known as Type Investigations.