Germomics

The 1925 serum run, drawn in a single continuous line.

We like to start with a game. Name the things that are timeless, and the answers come fast: red lipstick, a plain white tee, blue jeans. Name the things that are emphatically not timeless, and the list is just as quick: micro bangs, mullets, low-rise jeans, Crocs, that brief and regrettable season of wearing a dress over jeans. Somewhere on that second list, if you are us, you will also find diphtheria.

The joke has a real question folded inside it. How does a pair of blue jeans connect to a bacterial disease that once killed children by the thousands, and to a team of sled dogs running through an Alaskan winter? The path runs from cloth woven in the south of France to a tailor’s shop in Nevada, out to a frozen harbor on Norton Sound, and down to a microbe that makes diphtheria toxin only after a phage brings the tox gene.

Before there was a brand, there was a fabric, and the fabric carried its birthplace in its name. A sturdy twill woven in Nîmes, in southern France, was sold as serge de Nîmes, and English speakers shortened “de Nîmes” into the word we still use: denim. The other half of the vocabulary comes from Italy. Genoa shipped a similar hard-wearing blue cloth across Europe, and the French name for the city, Gênes, is where English got “jean” and then “jeans.” The cloth and the garment were old and well traveled long before the name most of us attach to them ever showed up.

That name belonged to Levi Strauss, a Bavarian immigrant who arrived in the United States as a poor young man and went to work in his family’s dry-goods trade in New York. In 1853, at twenty-four, he opened a West Coast branch of the family business in San Francisco and stocked it with everything a boomtown could want.

The invention that made his name famous, though, was not his. In the early 1870s a tailor named Jacob Davis, working in Reno, Nevada, took an order from a laborer’s wife who wanted work pants that could survive real punishment. Looking at the rivets he used to attach straps to horse blankets, Davis tried hammering copper rivets into the spots where trousers always blow out first, the corners of the pockets and the base of the fly. They held. The riveted pants were far tougher than anything else on the market, and orders piled up faster than one tailor could fill them. What Davis lacked was the money to file a patent, so he wrote to the man who supplied his cloth, Levi Strauss, and proposed they hold the patent together. Strauss agreed. On May 20, 1873, the two men were granted U.S. Patent No. 139,121, for an “Improvement in Fastening Pocket-Openings,” and the riveted blue jean was born. Davis moved to San Francisco to run the manufacturing while Strauss ran the business. More than a century later, one company is still a synonym for blue jeans, and blue jeans are still a synonym for a certain idea of America. To understand why Strauss was in San Francisco to receive that letter at all, you have to follow the gold.

Strauss did not sail west for the scenery. He came for the largest sales opportunity in the country. In January 1848, a carpenter named James W. Marshall spotted flakes of gold in the tailrace of a sawmill he was building for John Sutter, near Coloma in the Sierra Nevada foothills. He and Sutter tried to keep it quiet, and failed spectacularly. Word reached the East Coast, the president confirmed it before Congress, and roughly 300,000 people from across the country and around the world set out for the California goldfields. Within a few years, the little port of San Francisco became a frontier metropolis. Even in that crush it already held around 120 dry-goods stores, and Strauss had an unglamorous advantage: his family’s firm back east kept his shelves full when other merchants ran dry.

The earliest gold was almost embarrassingly easy to reach. It lay concentrated in gravel creek beds, so the first prospectors could pluck flakes and nuggets out by hand or swirl them from a pan. Over the whole rush, miners pulled more than 750,000 pounds of it from California, worth tens of billions of dollars today.

That sudden weight of people and money landed hardest on the people who were already there. On the eve of the rush, California held roughly 6,500 Californios of Spanish and Mexican descent, some 700 other foreigners, and about 150,000 Native Americans. During the rush, an estimated 120,000 of those Native Americans, roughly four in five, died from disease, starvation, and outright killing. The gold that built a state was a catastrophe for the people who had lived on that land first, and it is worth saying so plainly before we move on.

The rush also dragged California to statehood at startling speed. In 1850, only two years after Marshall’s discovery and just after the Mexican-American War handed the territory to the United States, California entered the Union as part of the Compromise of 1850, a bundle of five bills meant to hold off the fight over slavery that would later become the Civil War. One admitted California as a free state, upsetting the Senate’s balance between free and slave states; another settled the borders of Texas, which was already a slave state, roughly where they sit on the map today. A pair of work pants, it turns out, shares an origin story with the shape of a state line.

Gold has a way of moving people, and it kept moving them north. Nearly fifty years after California, in 1896, a strike on a tributary of the Klondike River near the Alaska-Canada border set off the last great gold rush of the century. The discovery is usually credited to a party that included the American prospector George Carmack and a Tagish man named Keish, better known as Skookum Jim; the claim was filed in Carmack’s name. Word reached Seattle and San Francisco the next summer, and something like 100,000 people set out for the Yukon, among them a young Jack London, who mined more stories than gold and turned the experience into The Call of the Wild.

The Klondike was far less forgiving than California. The two main routes in, the Chilkoot and White Pass trails, were steep enough and deadly enough that the Canadian authorities refused to let anyone cross without a year of provisions, close to a ton of goods that each stampeder hauled up the passes load by load, trip after trip. Photographs of the Chilkoot show an unbroken line of climbers strung up the slope like ants on a staircase. Of the roughly 100,000 who started, only about 30,000 reached the goldfields at all.

The Klondike faded almost as fast as it flared, partly because easier gold turned up elsewhere, including the beaches of Nome, on Alaska’s western coast, where it could be scooped from the sand without staking a claim. Nome swelled toward 20,000 people around 1900, a tent city running for miles along the waterline, and it collected at least one famous resident in Wyatt Earp, late of the gunfight at the O.K. Corral, who ran a two-story saloon. (One of us, through no fault of his own, has seen the film Tombstone enough times to quote it on command; “I’ll be your huckleberry” is lodged in there permanently.) The boom collapsed the way booms do. By 1910 the census counted about 2,600 people, and Nome survives today as a town of fewer than four thousand, which is remarkable when you consider where it sits: just below the Arctic Circle, with long, dark, brutally cold winters and a harbor that freezes solid for much of the year. That frozen harbor is where the story stops being an adventure and turns into an emergency.

By the winter of 1925, the ice had sealed Nome off from the sea, as it did every year from about November to July. The town had one doctor, Curtis Welch, and four nurses. After the last ship of the season left, Welch began seeing small children with sore throats and what looked like ordinary tonsillitis. Several of them died. By January he saw the thing he must have been dreading: a grayish membrane spreading across the back of a child’s throat, the classic sign of diphtheria.

His problem was almost unspeakable. The diphtheria antitoxin in his office had expired, and the fresh supply he had ordered months earlier never arrived before the port closed. It is worth being precise about what that antitoxin was, because it was not a vaccine. A vaccine trains your own immune system to defend you before you are ever exposed. Antitoxin is a treatment given after infection: antibodies raised by injecting a horse with the toxin, then harvested from the horse’s blood and given to a sick person to neutralize the poison already loose in the body. Nome needed that treatment badly and had almost none of it.

The town understood exactly what it was facing. Only six years earlier, the 1918 influenza pandemic had swept the Seward Peninsula; in the hardest-hit Native villages, somewhere between half and nine in ten residents died. With that memory fresh, Welch and the town leaders feared diphtheria could approach near-total mortality among the unprotected, and they imposed a quarantine. Then Welch sent a telegram out across the territory, in the clipped grammar of the wire, warning that an epidemic of diphtheria was almost inevitable, that he urgently needed a million units of antitoxin, and that the mail, carried by dog team, was the only transportation left.

Getting that antitoxin to Nome was the problem that became legend. There were three airplanes in all of Alaska in 1925, none built for winter, all of them dismantled for the winter, and no one willing to risk flying one that far in that cold. That left the dogs. The plan was to run the serum 674 miles across Alaska in the dead of winter by relay, picking it up where the railroad ended at Nenana and handing it from team to team toward Nome. Welch calculated the antitoxin would survive only about six days in such cold, which meant the relay had to cover, in roughly five and a half days, a route that normally took a month and had been run, once, in nine. A smaller cache of 300,000 units turned up in an Anchorage hospital, not enough to stop an epidemic but enough to treat the first patients and slow it, and it was packed into a cylinder and rushed to the railhead while more than a million more units started up the coast by ship.

Leonhard Seppala with his team, including Togo.

On January 27, 1925, the first musher set out from Nenana into temperatures between roughly fifty and sixty-two degrees below zero, blizzard whiteouts, and hurricane-force winds. Within hours of that first leg, parts of the lead driver’s face had blackened with frostbite, and at the first roadhouse he had to leave behind dogs that were already failing. Back in Nome, the case count kept climbing, and another child died.

The longest and most dangerous stretch fell to Leonhard Seppala and his lead dog, Togo. Seppala had a personal stake, a young daughter at risk in town, but he was chosen for a plainer reason: at forty-seven, with a twelve-year-old Siberian husky in the lead, he was simply the best driver in the region. He ran farther than anyone else, about 261 miles in all once you count the trip out to meet the serum and the run back, and to save time he took a route no one ran on a good afternoon, straight across the frozen surface of Norton Sound. Sea ice there flexes and cracks in the wind, and a team that lands its claws wrong can drop through into water that kills in minutes. Seppala crossed it in a gale, with Togo reading the ice in the dark and choosing the line. By one account the stretch he had just traversed broke apart and drifted out to sea only a few hours behind him.

On February 2, the final musher, Gunnar Kaasen, drove his lead dog Balto down Front Street with every glass ampule of serum intact. The relay had taken about 127 and a half hours, just over five days, run by some twenty teams and more than 150 dogs, several of which died on the trail. The antitoxin worked, and the epidemic broke. The recorded death toll was small, five to seven, but that figure reflects mostly the town’s documented cases. Welch himself guessed there were at least a hundred more cases in the Native camps outside town, and families often buried children without reporting the deaths.

History, for a while, remembered the wrong dog. Balto, who ran the last leg into town and stood for the photographers, got the statue in Central Park and, decades later, an animated movie. Togo, who ran the longest and most lethal distance, was overlooked for years before the record was set straight; he has his own statue now, and a 2019 film at last told his side. Both dogs earned their place, and the run itself, the Great Race of Mercy, lives on every year in the Iditarod sled dog race, run on some of the same trails.

Corynebacterium diphtheriae, Gram-stained, the crystal violet that gives Germomics its color.

The agent behind all of this is a bacterium called Corynebacterium diphtheriae, first seen by Edwin Klebs in 1883 and grown in pure culture a year later by Friedrich Löffler; for that reason it is still sometimes called the Klebs-Löffler bacillus. It is a gram-positive rod, meaning it wraps itself in a thick outer wall, and it is pleomorphic, able to shift its shape with its surroundings. Several subspecies exist, any of which can cause the disease. Here is the hinge: whether a given strain can kill you depends on whether it is carrying a particular virus.

The poison that does the damage, diphtheria toxin, is not written into the bacterium’s own core genome. It is delivered by a bacteriophage, a virus that infects bacteria, which can stitch itself into the bacterial chromosome and sit there quietly as what microbiologists call a prophage. The gene for the toxin rides along inside that viral DNA. A C. diphtheriae cell without the phage cannot cause classic diphtheria, because it makes no toxin. It can still cause milder throat and skin infections, and rarely worse. The same cell, once infected by the right phage, becomes a factory for diphtheria toxin. The bacterium causes toxin disease only once it has been infected itself.

A schematic of lysogeny and of the toxin’s effect on a cell. This is not the 1925 serum run, and it is not a vaccination schedule. Cholera toxin is carried by a phage too, in Confessions of a Cholera Girl.

Sources: Freeman, Journal of Bacteriology (1951); Holmes, Journal of Infectious Diseases (2000); Naglich, Metherall, Russell and Eidels, Cell (1992); Collier, Toxicon (2001).

The toxin kills by stopping protein synthesis. It slips into a human cell and disables eukaryotic elongation factor 2, which the cell needs to build every new protein. A cell that cannot make protein dies, and a single molecule of toxin is enough to shut one down. In the throat, where the bacteria usually settle, the wreckage becomes visible. Dead cells and fibrin congeal into a tough sheet, the pseudomembrane, that creeps across the back of the throat and can turn gray-green, or black where it has bled. The disease takes its very name from this membrane: diphtheria comes from the Greek for leather. Left to grow, alongside the swollen tissue and lymph nodes that give advanced cases their so-called bull neck, the membrane can close the airway outright. And if the toxin escapes into the bloodstream it travels, inflaming the heart muscle until the rhythm falters, attacking nerves until muscles give way, and injuring the kidneys. Diphtheria can suffocate you or stop your heart, and for most of human history a doctor at the bedside could do little but watch.

We do not hold our breath over a child’s sore throat the way families once did, and the reason is the thing the 1925 serum could never provide: a vaccine. Toxoid vaccines arrived in the 1920s, mass immunization of American children followed in the 1930s and 1940s, and protection now travels bundled with tetanus and whooping cough in the DTaP and Tdap shots calibrated to age. A disease that once arrived in deadly waves now turns up so rarely that the United States reported only 14 cases from 1996 through 2018. The horse-derived antitoxin still exists for those rare cases, though it is no longer a licensed commercial product; the Centers for Disease Control releases it as an investigational drug when a case is confirmed or strongly suspected, a direct descendant of the serum that crossed Norton Sound behind a team of dogs.

A tailor’s copper rivets, a frozen telegram, a horse’s antibodies, and a phage-borne tox gene meet in that Alaskan winter. Blue jeans are still everyday cloth. Diphtheria, thanks to vaccination and that serum run, is mostly a historical and clinical rarity in the United States. The gap between the garment that stayed and the disease that did not is the connection this episode traces.

The episode audio states fourteen things that are wrong. The corrections:

Carnival de Brazil

Leptospira

New Orleans during Mardi Gras is one long party, the streets a blur of parades and floats and beads and king cakes in purple, green, and gold. It is also a city whose rats carry a long list of pathogens, including the one behind this episode: leptospirosis. The biggest celebrations tend to happen in places this disease favors, warm, crowded, and prone to flooding, and Rio’s Carnaval sits in the middle of just such a city. Leptospirosis is, by most reckonings, the most widespread animal-borne disease in the world, and it is easy to miss: reservoirs shed it without looking sick, and people often never report the mild cases. The way into a disease that hides in kidneys and floodwater is to watch the animals that carry it.

In Brazil in the late 2000s, the forests went quiet. Howler monkeys, normally among the loudest animals in the Americas, began dying in large numbers, and the silence was itself the alarm. Testing the dead monkeys revealed yellow fever virus, and because New World monkeys sicken and die fast when infected, their die-offs work as an early-warning system; Brazil runs a national surveillance program built on exactly this, using monkey deaths to trigger human vaccination ahead of an outbreak. During the 2008-2009 yellow fever outbreak in southern and southeastern Brazil, those monkey deaths preceded the human cases and helped drive a vaccination campaign of millions of doses.

This is the old idea of the canary in the coal mine made literal. For most of the twentieth century, miners carried caged canaries underground, watching for the bird to falter as a sign of toxic gas; Britain announced the end of the practice in 1986. The principle keeps proving itself. In 1999, residents of New York City noticed an unusual number of dead crows around the same time that an unexplained cluster of human encephalitis appeared, and the two turned out to share a cause: West Nile virus, making its first-ever appearance in the Western Hemisphere. The crows were the sentinels. A 2009 Lancet Infectious Diseases column with a deliberately silly title argued that epidemiologists still under-use animal populations this way, because so many dangerous pathogens live in an animal reservoir, a host that can carry and shed a microbe without being sick itself, long before they reach people. A reservoir can be an animal, an environment like water or the soil that harbors anthrax, or even a person, the asymptomatic human carrier of whom Typhoid Mary is the most famous. For leptospirosis, the reservoir and the sentinel are often the same humble creature: the rat.

The disease has a second name, Weil’s disease, after the nineteenth-century German physician who described its severe form, which is why you sometimes hear it said with a softer, V-like opening. It is caused by Leptospira, a spirochete, a bacterium coiled like a tight little spring with hooked ends, propelled by internal flagella that let it corkscrew through tissue. It is, by most reckonings, the most common zoonotic infection on the planet. The trick to its success is where it lives: Leptospira colonizes the kidneys of animals, rats and other rodents above all but also dogs, cattle, pigs, and wildlife, and those animals shed it in their urine, often for life, without falling ill. From there it pours into soil and standing water, where it can survive for weeks to months.

People catch it not from a bite but from the water and mud, when the bacteria slip in through a cut, a scrape, or the soft tissue of the eyes, nose, or mouth. That is why the classic victims are people who get wet in the wrong places: kayakers, swimmers, rowers, farmers, veterinarians, sewer workers, and, above all, the residents of crowded tropical cities after a heavy rain, when floodwater stirs the rats’ world into everyone’s. A 2015 global estimate puts it at about a million cases and nearly fifty-nine thousand deaths a year, concentrated in exactly those warm, flood-prone, densely populated places. In the United States, by contrast, there are only about a hundred to a hundred and fifty cases a year, most of them in Puerto Rico and Hawaii.

A schematic of the maintenance cycle and of human exposure, not a risk score. How long the bacteria remain outside a host depends on the conditions, which this figure does not measure. Not every exposure infects, and not every infection is severe. This is not advice about a flood, a wound, or treatment.

Sources: Levett, Clinical Microbiology Reviews (2001); Haake and Levett, Current Topics in Microbiology and Immunology (2014); Centers for Disease Control and Prevention, About leptospirosis.

What makes leptospirosis so easy to miss is that most of the time it barely announces itself. Most infections are mild, a flu-like stretch of fever, headache, and aching muscles that people often never report, assuming they simply caught a bug. The illness can run a biphasic course, an early phase when the bacteria are in the blood, a brief lull, and then, in a minority, a second and far nastier phase. Perhaps 5 to 10 percent of cases turn severe, with abrupt high fever, vomiting, and the jaundice, a yellowing of the eyes and skin, that signals the liver is in trouble, often alongside kidney failure. The worst version is the one Weil put his name to: severe leptospirosis with multi-organ involvement, where kidney failure, liver failure, bleeding into the lungs, and meningitis can converge into a high-mortality crisis.

Doctors confirm it by amplifying the bacterium’s DNA with PCR, by testing for antibodies, or by culturing the organism from a sample like urine. Caught in time it is treatable, with oral antibiotics such as doxycycline or amoxicillin for milder cases and intravenous antibiotics for severe ones. There are vaccines for animals and pets, which is part of how you protect people, but no widely approved human vaccine in the United States. A disease this common and this often-silent is precisely the kind you would want an early-warning system for, which brings the whole thread back around.

Carnaval and leptospirosis share a stage: warm, crowded, rain-soaked cities where floodwater can move rat urine into human skin and mucosa. The rains that soak a celebration are the same rains that flush Leptospira into the places people wade. The early signs are often still in the animals first, the howler monkeys going silent ahead of yellow fever, the crows dropping ahead of West Nile, the rats carrying a passenger in their kidneys through a flooded gutter. Weil’s disease is the severe end of a common zoonosis that spends most of its time looking like nothing much. Looking where the reservoir lives is how you catch it before the hospital census does.

The episode audio states eight things that are wrong. The corrections:

Edgar Allan Poe

Rabies Virus

A storm knocks out the power, the internet dies, and you find yourself playing the little dinosaur game that appears when a browser has nothing to connect to. It is nearly Halloween. Somewhere in that ordinary evening is a path to one of the oldest and most terrifying diseases on Earth, because the road from a no-signal dinosaur to rabies runs through the literature of fear. Once upon a midnight dreary, as the most famous opening in American gothic has it, while the narrator nods nearly napping, something comes tapping at the chamber door. Edgar Allan Poe wrote that, and Poe died young and strangely, found delirious in borrowed clothes on a Baltimore street in 1849, and among the many theories about what killed him is rabies. That is fitting for a Halloween episode whose subject is the disease under the monsters. Rabies is older than the werewolf and the vampire stories that still echo its symptoms, and almost uniquely among those horrors, it is one we learned how to interrupt.

Trace the family tree of scary stories and Poe sits near the root. He gave us the modern horror tale and, in the same stroke, invented detective fiction. The gothic writers, Poe and Mary Shelley and Bram Stoker, fed directly into the great Universal monster movies of the 1920s through the 1940s, a run of films that gave Depression-era audiences a cheap hour of escape and gave us the silhouettes we still put on porches: Dracula, Frankenstein’s creature, the Mummy, and, most to the point, the Wolf Man. The lineage continues through Hitchcock and Kubrick and lands on Stephen King, the most adapted horror writer alive, whose early novel Cujo turns a beloved family Saint Bernard into a slavering killer that traps a mother and child in a broiling car. That dog is not possessed or supernatural. The dog has rabies.

Once you notice that, the same clinical picture keeps surfacing under older monsters: a familiar creature, a dog or a wolf or a person, that loses its mind, fears nothing, recoils from water, and passes its condition through a bite. The werewolf transformed by a bite, the vampire whose bite recruits new vampires, the loyal animal turned murderous, are folklore wearing symptoms that match a real virus. Rabies turns the bitten into biters. Horror writers were not inventing from nothing so much as remembering a pattern people had watched for centuries.

The word rabies is Latin for madness, which is about as accurate as a disease name gets. It takes two broad forms, and the one we picture is furious rabies, with its hallmark cluster of agitation, erratic and aggressive behavior, drenching salivation, and the eerie symptom that gave the disease its old name of hydrophobia: a fear of water, caused by violent throat spasms so painful that the victim cannot swallow and may panic at the mere sight of liquid. Some patients react the same way to a puff of air. The animals show a strange tell, looking drunken and slow one moment and striking with shocking speed the next.

It is not a historical curiosity. The World Health Organization estimates that rabies still kills on the order of forty-four thousand people a year, the overwhelming majority in Africa and Asia, and roughly four in ten of the dead are children. Around the world, dogs are responsible for some ninety-nine percent of human cases; in the United States, where dog rabies is controlled, the usual sources are raccoons, skunks, foxes, and bats. Small animals like rats and mice almost never carry rabies and have not been known to pass it to people. And here is the fact that makes rabies the most frightening entry in this whole series: once symptoms appear, it is virtually one hundred percent fatal. In all of recorded history only about three dozen people are known to have survived symptomatic rabies, most with severe lasting damage. There is no monster in any movie with a kill rate like that.

Under a microscope the virus looks the part, shaped like a blunt bullet. It belongs to the genus Lyssavirus. After a bite, the virus enters muscle and nerve endings, though which molecules it uses to attach are still debated, then slips into the nervous system and travels up the long fibers of the nerve cells toward the spinal cord and brain. It hijacks the cell’s own internal transport machinery and moves slowly, about two to four inches per day. When it reaches the brain and replicates there, clinical rabies begins. Then the virus travels back outward to the salivary glands, loading the mouth with infectious particles so that the host’s next bite delivers the disease to a new victim. The animal has been turned into a delivery system.

A schematic of that path. How long it takes varies, and which molecules the virus uses to enter a nerve are still argued, so no receptor is named. Once symptoms have begun, rabies is almost always fatal. This figure does not say what anyone should do after a bite.

Sources: Fooks, Cliquet, Finke and colleagues, Nature Reviews Disease Primers (2017); Davis, Rall and Schnell, Annual Review of Virology (2015); World Health Organization, Rabies fact sheet.

That slow crawl up the nerves is what makes rabies so hard to stop once symptoms start, and what leaves a usable gap beforehand. Because the virus can take weeks, sometimes longer, to travel from a bite on the leg to the brain, there is a window. If the body’s defenses are ready before the virus arrives at the central nervous system, post-exposure vaccination can still work. Almost everything that saves people in this story depends on that gap of time.

People have dreaded rabies for as long as there are records. There is a Mesopotamian legal code from around 1930 BC fining the owner of a dog that bit someone to death, and through the Middle Ages dogs were branded with a red-hot iron called St. Hubert’s Key, while desperate sufferers had their wounds cauterized in an attempt to burn the disease out. Nothing worked, and for almost all of human history a bite from a rabid animal could be a death sentence. Then, in 1885, Louis Pasteur changed it. Working with attenuated virus he had aged by drying out the spinal cords of infected rabbits, and famously not a physician himself, Pasteur treated a nine-year-old boy named Joseph Meister who had been bitten more than a dozen times by a rabid dog. Over about ten days the boy received a series of increasingly potent injections, and he lived. It was the birth of the Pasteur Institute and of an idea that still saves lives. Meister survived into adulthood and worked at the Institute for more than twenty years before his death in 1940.

What Pasteur discovered is that you can vaccinate someone after they have already been exposed, a strategy called post-exposure prophylaxis, and it works precisely because the virus moves so slowly. The modern version pairs an immediate dose of antibodies, which buy short-term protection, with a series of vaccine shots that train the body to make its own, and given promptly it is close to one hundred percent effective. The catch, which cannot be stressed enough, is that it has to happen before symptoms begin; once the virus reaches the brain, the window has closed. There are exceptions so rare they prove the rule, like the case of Jeanna Giese, a teenager who in 2004 became the first known unvaccinated person to survive symptomatic rabies after doctors placed her in an induced coma in a last-ditch experiment, an approach that has rarely worked since. Meanwhile, the most effective prevention happens before anyone is bitten at all: vaccinating the animals, including wild ones, sometimes by air-dropping millions of vaccine-laced baits over the woods.

From a midnight dinosaur game to a bullet-shaped Lyssavirus by way of Poe, the Wolf Man, and Cujo, the thread ends on a disease that still kills tens of thousands of people a year, about four in ten of them children, most of them in places with less access to post-exposure care. Unlike the monsters it helped inspire, rabies has a weakness: that slow nerve crawl. A person bitten in the dark can still get to a clinic before symptoms begin and walk back out. It took four thousand years of dread and Pasteur drying rabbit spines in a Paris lab, but the disease can be outrun a few inches of nerve at a time.

The episode audio states a number of things that are wrong. The corrections:

Influenza Virus

Christopher Columbus

There is a rhyme almost every American can still recite without thinking: in 1492, Columbus sailed the ocean blue. It surfaces around the same time each year, when the construction-paper crafts start coming home from school and the first cold snap arrives, which is also when flu season begins. The connection between Columbus and influenza is not a stretch. Influenza may have arrived in the Americas with the same voyages that opened the hemisphere, and more than five centuries later it still comes back every autumn under a shifted antigenic name, because antigenic change is how it keeps finding susceptible hosts. A virus we have named for a hundred years still kills thousands to tens of thousands of people a year in the United States alone, and the defense that matches a face that changes annually is a shot you renew on that same schedule.

Christopher Columbus was an Italian navigator financed by Spain, and the voyages he began in 1492 opened a permanent sea route between Europe and the Americas. What followed for the people already living there was catastrophe, partly by the sword and partly by something nobody on either side understood. The Taino who met his ships had been separated from the Old World’s crowd diseases for roughly twelve thousand years, and they carried no immunity to them. When Europeans, their livestock, and their microbes arrived, the result was what historians call a virgin-soil epidemic. Estimates vary and are necessarily rough, but many scholars put the loss at eighty to ninety percent of the Indigenous population of the Americas over the following century and a half, with disease the largest cause alongside war, enslavement, and forced labor.

Smallpox is the name most people attach to that collapse, and it did arrive, devastatingly, a few decades later. But the first recorded epidemic came earlier. During Columbus’s second voyage, in 1493, a sudden, highly contagious illness with fever and crushing malaise swept the new settlement of La Isabela on Hispaniola, sickening the Spanish, including Columbus himself, and killing the Taino in large numbers. Because smallpox would not be documented on the island for decades more, and because the fleet had brought pigs aboard, some historians who studied the accounts, beginning with Francisco Guerra, have argued the likeliest culprit was swine influenza, an animal flu that jumped to people. The earliest American epidemic on record may well have been the flu. To understand why a respiratory bug could do that kind of damage, and why it keeps returning, you have to understand what influenza actually is.

Influenza is not interchangeable with a common cold. Colds have many causes and are usually mild; influenza is caused by influenza viruses, comes on abruptly, and brings the body aches that mark it. It is an RNA virus and an obligate intracellular parasite, which means it can only reproduce by hijacking your cells and budding out through their membranes. There are four types, A, B, C, and D, the last of which was only identified around 2011 and given its own genus in 2016. Type A is the dangerous one. Its natural home is wild aquatic birds, ducks and geese and shorebirds, which carry a huge variety of it; from there it can spill into poultry, pigs, and people, and that is where pandemics come from.

Type A is named by two proteins on its surface, and those names are the H#N# labels you hear on the news. Hemagglutinin, the H, is the key that lets the virus unlock and enter a host cell. Neuraminidase, the N, is the pair of scissors that cuts newly made viruses free so they do not clump on the cell they just destroyed. There are eighteen known H subtypes and eleven N subtypes, and they reassort, especially when two different flu strains infect the same animal, a pig being a notorious mixing vessel, and swap whole gene segments. That reshuffling, called antigenic shift, can produce a virus the human immune system has never seen, as in the 1957 H2N2 and the 1968 H3N2, and the 2009 swine-flu H1N1 was another reshuffled strain. The 1918 H1N1 appears to have been an avian virus that adapted to people whole. Avian strains like H5N1 and H7N9 are watched closely for the same potential. Even between those big jumps, the virus is constantly accumulating smaller mutations, a slower process called antigenic drift. The upshot is a pathogen whose defining trait is that it is always slightly new, which is precisely why catching the flu one year buys you only partial protection the next.

Influenza spreads about as easily as a respiratory virus can. It rides aerosolized droplets from coughs and sneezes, passes by direct contact, and lingers on fomites, the doorknobs and keyboards and phones we share, staying viable for hours, and by some laboratory measures a day or two, on hard surfaces. It favors winter partly because low humidity and weak ultraviolet light let the virus in those droplets stay infectious longer. None of this is a historical footnote. In a typical recent year the United States alone sees on the order of tens of millions of flu illnesses, hundreds of thousands of hospitalizations, and tens of thousands of deaths. The 2018-2019 season was later estimated at roughly twenty-nine million illnesses and about twenty-eight thousand deaths, and that was the milder of two back-to-back seasons. The 2017-2018 season was classified as high severity and killed tens of thousands of Americans, including a record number of children for a non-pandemic season at the time. Later seasons passed that record, with 290 pediatric deaths in 2024-2025. Influenza still kills tens of thousands of Americans in a bad season, even though it no longer appears on CDC’s list of the ten leading causes of death.

Most people who get the flu recover on their own, but there are warning signs that warrant real attention: trouble breathing, chest pain, confusion, a bluish color from lack of oxygen, dehydration, or symptoms that improve and then come roaring back with fever, which can signal a secondary bacterial pneumonia. There are antiviral drugs, but they work best within the first forty-eight hours, a window that is easy to miss once the disease has flattened you onto the couch. That timing is why prevention matters more than waiting for a prescription.

If influenza keeps changing, the matching defense is a vaccine we update on the same schedule. Each year its composition is reviewed and updated to match the strains expected to circulate, which is the answer to the shrug that someone already got one last year. Last year’s shot was aimed at last year’s strains. A few other objections are worth retiring. The standard flu shot is inactivated, made from killed virus or just viral proteins, so it cannot give you the flu; the nasal spray uses a weakened virus engineered to survive only in the cool nose and not the lungs, and cannot give it to you either. There are egg-free formulations for people with egg allergies. The vaccine is recommended during pregnancy, where it protects both the parent and the newborn for its first months of life. Getting it is not only about you: the people who die of influenza are disproportionately the very young, who cannot be vaccinated before six months, and the very old and immunocompromised, whom the vaccine protects less well, so a shot you barely feel is also cover for them.

From a kindergarten rhyme to La Isabela to a syringe at the pharmacy, the subject stays the same: a virus that keeps arriving antigenically new. It may have come ashore with Columbus as one more thing the New World had never met, and it still comes back every fall as a boy named Flu, renamed by drift and shift. We cannot make it stop changing. What we can do is change the introduction once a year, on its schedule: a standing appointment, not a cure.

The episode audio states ten things that are wrong. The corrections:

Theodore Roosevelt

Salmonella enterica Typhi

This episode begins with a confession of culinary cowardice. Taking a turn as the family grill master, one of us decided to start with steak, on the logic that you can serve it less than well done and nobody gets sick, and to stay far away from hamburgers, because an undercooked burger can drop a whole birthday party. It is a reasonable fear. Ground beef blends the surface of the meat all through it, so whatever was on the outside is now on the inside, which is exactly why a rare steak is fine and a rare burger is a gamble. That small kitchen anxiety is the doorway into something much larger. The path from a backyard burger runs, by way of a muckraking novel and a grieving president, to a bacterium that hides inside your own immune cells, and finally to the most quarantined woman in American history. Along the way it asks a hard question: when contamination is invisible, do we fix the system, or do we blame a person? We did both, and the difference is the whole story.

Nobody can quite agree on who invented the hamburger. A Danish immigrant in New Haven, a fair in Hamburg, New York, the city of Hamburg in Germany, all have claims, and the sandwich genuinely took off after it appeared at the 1904 World’s Fair in St. Louis. What matters more is what happened to the meat industry right behind it. In 1906, Upton Sinclair published The Jungle, a novel meant to expose the brutal exploitation of immigrant workers in Chicago’s meatpacking plants. Readers barely noticed the workers. What horrified them was the filth: the descriptions of what was going into American food. Sinclair later said he had aimed at the public’s heart and hit it in the stomach instead.

The outrage was enormous. President Theodore Roosevelt, initially skeptical and reportedly dismissive of Sinclair, sent investigators to the Chicago plants; the owners knew of the visit in advance, and the inspectors were still appalled. The report went to Congress, and on June 30, 1906, Roosevelt signed two landmark laws on the same day: the Meat Inspection Act, which put federal inspectors over meat processing, and the Pure Food and Drug Act, whose enforcement fell to the Bureau of Chemistry, the agency that later became the Food and Drug Administration. This is the systemic answer to invisible contamination, the one we are rightly proud of. Faced with a faceless, industrial danger in the food supply, the country built institutions to police it, and we have eaten more safely ever since.

The man who signed those laws is worth pausing on, because typhoid was not an abstraction to him. Theodore Roosevelt was a sickly, asthmatic boy who willed himself into a barrel-chested adult, a naturalist who at seven acquired a dead seal’s head from a market and founded, with his cousins, a little Roosevelt Museum of Natural History in the house. He grew into a Rough Rider, a conservationist who set aside vast public lands, a Nobel Peace Prize winner, and a president so stubborn he once finished a speech of well over an hour after being shot in the chest. But the event that bent his life happened on February 14, 1884. Two days after his wife Alice gave birth to their daughter, Roosevelt was called home from the legislature to a house where both women he loved were dying on separate floors. His mother, Mittie, died of typhoid fever around three in the morning. His wife died that afternoon of a kidney disease the pregnancy had hidden. In his diary that day he drew a heavy X and wrote a single line, that the light had gone out of his life.

Mittie most likely caught typhoid the way most people did in Gilded Age New York: from food or water touched by sewage, in a city that did not yet treat either. The disease that widowed and orphaned a future president, and helped turn him into a reformer of cities and food, is the same one this whole hamburger thread has been walking toward.

Typhoid fever is caused by Salmonella enterica serotype Typhi, a gram-negative, flagellated rod that lives nowhere in nature except inside human beings. There is no animal reservoir, no environmental hideout; it passes from person to person by the fecal-oral route, when bacteria shed in one person’s stool reach another person’s food or water. That detail is the hinge of everything that follows. Swallowed in a large enough dose to survive the stomach’s acid, the bacteria slip across the gut lining at the Peyer’s patches, clusters of immune tissue in the lower small intestine. Then comes the clever, sinister part. They are engulfed by macrophages, the immune cells whose entire job is to eat and destroy invaders, and instead of dying they survive inside, ride those cells through the lymphatic system, and replicate, turning the body’s own defense into transport and shelter. They spill into the bloodstream, seed the liver and spleen, and crucially settle in the gallbladder.

The illness that results gives off fever, headache, exhaustion, a scatter of pink rose spots, and, in the worst cases, perforated intestines and bleeding. The name itself comes from the Greek typhos, meaning smoke, for the mental fog and delirium of the severe disease. Worldwide there are still roughly nine million cases a year; in the United States an estimated 5,700 a year, mostly carried home by travelers. It is treatable with antibiotics and preventable with a vaccine. But the gallbladder is the catch. In a small fraction of people, perhaps 3 to 5 percent, the bacteria take up permanent residence there, often nestled in biofilms on gallstones, long after the person has recovered and feels completely well. They become a chronic carrier, shedding live typhoid in their stool for years, even decades, while showing no symptoms at all. A human reservoir who feels fine. Which is to say, the disease was waiting for someone exactly like Mary Mallon.

Mary Mallon emigrated from Ireland as a teenager and made her living as a cook for wealthy New York families. Between roughly 1900 and 1907, typhoid kept breaking out in the households she worked for, usually within weeks of her arrival. All told she was linked to about fifty infections and three deaths. What made her case historic is that she was the first healthy carrier identified in the United States: infected, contagious, and yet never sick a day herself. A sanitary engineer named George Soper traced the outbreaks to her and, in the prim early 1900s, approached her to ask for stool and urine samples. From her point of view this was outrageous. She felt perfectly healthy, she had done nothing wrong, and a stranger was accusing her of poisoning the people she cooked for. She refused, and she was eventually arrested and tested by force. The samples came back swarming with Salmonella Typhi.

She was quarantined on North Brother Island in the East River, where her stool samples tested positive again and again, and the newspapers seized on her, branding her Typhoid Mary and calling her the most dangerous woman in America. After about three years she was released on two conditions: never cook for others again, and report to the health department regularly. The city helped her find work as a laundress, which paid less, and she was shunned wherever she was recognized. Eventually she changed her name to Mary Brown and went back to the one trade she knew, taking a kitchen job at a hospital, where she touched off another outbreak that sickened twenty-five people and killed two. She was arrested a second time and spent the rest of her life, more than two decades, confined to the island until she died there in 1938. She once wrote that she had been made into a kind of peep show. And here is the part that should sit uncomfortably: during the years she was caged, public health workers identified many other healthy typhoid carriers, and almost none of them were locked away. She was the one with a name the papers could sell.

So the thread from a hamburger to Typhoid Mary is really a study of two responses to the same problem. Invisible contamination terrifies us, and we answer it in two very different registers. When the danger was industrial and faceless, the meat in the grinder, we did something genuinely admirable: we passed laws, built what became the FDA, and made the food supply safer for everyone, the reform a typhoid widower happened to sign. When the danger wore a human face, an immigrant cook who felt perfectly well and had asked for none of this, we built a spectacle and a prison instead, and we singled out the one carrier we could name while quietly letting the others walk. Mary Mallon was a real hazard; no one should pretend a chronic carrier in a kitchen is harmless. But she was also a person we failed, and the asymptomatic carrier remains the hardest case any outbreak presents, precisely because the threat looks exactly like an innocent person, because it is one. The hosts’ plea at the end of the episode is the right one and it is not soft: treat the infected as people, whatever is living inside them, because the alternative is that the next carrier, afraid of becoming the next Typhoid Mary, simply never gets tested at all. There is, it turns out, something about Typhoid Mary, and most of it is about us.

The episode audio states several things that are wrong or out of date. The corrections:

Dilophosaurus

Treponema pallidum

When Jurassic Park opened in 1993 it was less a movie than a weather system. It won three Academy Awards, it was sold as an adventure sixty-five million years in the making, and for a while you could not get a haircut without the person holding the scissors talking to you about DNA. It put genetic engineering into ordinary conversation, and it handed the culture two lines it never gave back. One was Jeff Goldblum’s mathematician insisting that life finds a way. The other was his warning that the park’s scientists had been so consumed by whether they could bring dinosaurs back that nobody stopped to ask whether they should. That gap, between can and ought, is the real subject of this episode. The trail from a dinosaur blockbuster to a corkscrew-shaped bacterium runs through a diseased Ice Age bear, a doctor destroyed for being right, and the darkest chapter in the history of American public health, and the whole way down it is asking the same uncomfortable question.

Start with the bear. In 1987, the paleopathologist Bruce Rothschild reported in Nature that the bones of a Pleistocene bear from Indiana, roughly eleven and a half thousand years old, carried lesions characteristic of treponemal disease and tested positive with the same antisera the CDC used to confirm syphilis. Another researcher contested the diagnosis. Rothschild allowed that the animal might instead have had syphilis’s close cousin yaws.

The diagnosis sits on top of a genuinely unsettled question. Treponema pallidum, the bacterium behind syphilis, comes in subspecies that cause yaws and other diseases too, and where venereal syphilis itself came from is still argued. The best-known idea, the Columbian theory, holds that Columbus’s crews carried it back from the New World in the 1490s, just before it exploded across a war-torn Europe. Recent ancient-DNA work on treponemal bacteria from the Americas points toward an American origin, though the debate is not closed. The disease even got its name from a poem: in 1530 the Italian physician Girolamo Fracastoro wrote of a shepherd named Syphilus, cursed by Apollo with the affliction. We can read disease in eleven-thousand-year-old bone and still not agree on where it started, which is a good reminder that detecting a disease and knowing what to do about it are entirely different problems. Nobody learned that more bitterly than the man who discovered handwashing.

In the 1840s, a Hungarian physician named Ignaz Semmelweis was working at the Vienna General Hospital, which ran two maternity clinics. One was staffed by doctors and medical students, the other by midwives, and the doctors’ clinic killed far more mothers, sometimes several times as many, with childbed fever, the bacterial infection that followed birth. Semmelweis was haunted by the discrepancy. He noticed that women who gave birth in the street on the way in fared better than those delivered by physicians, and that the midwives’ clinic was safer too. He stripped away every difference he could think of until one remained: the doctors and students went straight from performing autopsies to delivering babies. He concluded they were carrying something he called cadaverous particles from the dead to the living, and in 1847 he ordered them to scrub their hands in a chlorinated lime solution. The maternal death rate in his clinic collapsed, from an average of about ten percent to about two.

He was right, and it ruined him. This was before germ theory, so he could not say what the particles were, only that washing removed them, and the medical establishment already had a long list of fashionable explanations for childbed fever into which his idea did not fit. Worse, many physicians took the implication that their own unwashed hands were killing patients as a personal insult to gentlemen. Semmelweis did not help his case; as the years passed he grew bitter and combative, publishing open letters that called the doctors who ignored him murderers. His behavior deteriorated, and in 1865 his wife and friends had him committed to an asylum. He tried to leave, was beaten by the guards, and died about two weeks later at forty-seven of sepsis from an infected wound, the very kind of infection he had spent his life trying to prevent. There is grim irony layered on irony here, because one of the conditions later proposed to explain his unraveling was syphilis.

Syphilis was, for most of history, almost everywhere. By some nineteenth-century reckonings a startling share of the population carried it at some point in their lives, and the roster of famous people rumored to have had it is long, mostly unprovable, and a parlor game unto itself. The ragtime that scores this episode is part of that story: Scott Joplin, who wrote The Entertainer and Maple Leaf Rag, declined and died of complications generally attributed to the disease, and composers from Schubert to Beethoven turn up on the same speculative lists, with the emphasis on speculative. What is not in doubt is how common and how protean the disease was, which earned it the nickname the great imitator, because its symptoms mimic so many other illnesses.

The culprit is Treponema pallidum, a spirochete, a bacterium shaped like a corkscrew, and it moves like one too, its internal flagella spinning it forward so it can bore through the smallest micro-tears in skin. It usually spreads through sex, and a mother can pass it to her child in the womb, which is called congenital syphilis. Untreated, it unfolds in four acts. First, a few weeks after infection, usually three to four, a chancre appears at the entry point, a firm, round, painless ulcer. Then, weeks to months later, comes a rash that famously turns up on the palms and soles, along with fever and aches. Then a silent latent phase that can last years. And finally, in perhaps a third of untreated people and sometimes decades later, tertiary syphilis, which is no longer contagious but quietly wrecks the organs it has settled into: the brain in neurosyphilis, the heart and blood vessels, the bones. It could be disfiguring enough to collapse the bridge of the nose, which is the grim origin of the so-called no-nose clubs, which appear in satirical London writing and later retellings as dinners for the disfigured; how real they were is hard to say. And then, since the 1940s, the twist that makes the rest of this story unbearable: a few injections of penicillin cure it. We could fix it.

For forty years, the United States government chose not to. From 1932 to 1972, the U.S. Public Health Service ran what it called the Tuskegee Study of Untreated Syphilis in the Negro Male, in Macon County, Alabama. It enrolled about six hundred poor Black sharecroppers, three hundred ninety-nine of them already infected with latent syphilis and two hundred one as uninfected controls. The men were never told they had syphilis. They were told they were being treated for bad blood, a vague local term, and were given free meals, free exams, and burial insurance in exchange for letting doctors track them. The purpose was simply to watch untreated syphilis run its course to autopsy. When penicillin became the standard cure in the 1940s, the men were not given it. The study ran on for decades more, and it ended in 1972 only because a whistleblower took the story to the press. By then, of the original infected men, twenty-eight had died of syphilis outright, about a hundred more of related complications, forty of their wives had been infected, and nineteen of their children had been born with congenital syphilis.

It was not an isolated lapse. The same agency, with at least one of the same physicians involved, had in the 1940s deliberately infected prisoners, soldiers, and psychiatric patients in Guatemala with syphilis, sometimes by injection and sometimes by sending in infected sex workers, to study the disease and its treatment. These were not failures of scientific capability. The science was sound enough to identify the spirochete, chart its stages, and cure it. The failure was the one Goldblum’s character named in a fictional lab a half century later, made real and monstrous: people so fixed on what they could learn that they never stopped to ask whether they should.

So that is the road from a dinosaur movie to a spirochete, and it turns out the dinosaurs were never the point. The point was the warning bolted onto the side of the spectacle. We learned to read treponemal disease in the bones of an Ice Age bear, to name and describe the corkscrew bacterium, to end its four-act tragedy with a cheap injection. None of that was the hard part. The hard part was the question the science could not answer for us, the one about what we owe the people the science is done to. Tuskegee is the answer to why trust in medicine has to be earned rather than assumed, and why it is thinnest in exactly the communities that were once treated as material rather than patients; the distrust it seeded outlived the study by decades and shapes public health to this day. The most important line in a movie about cloning dinosaurs was never about dinosaurs at all. It was the question we keep forgetting to ask before we find out whether we can. It’s on the syphilis, and it was on us.

The episode audio states a few things that are wrong. The corrections:

Truckee River

Cerebellum with Kuru

This one opens with a confession: one of us nearly got thrown out of George Washington’s house. The crime was chewing gum. A guard at Mount Vernon spotted it, delivered the firm reminder that this is a museum, sir, and set off a quiet panic in which a whole tour group of fellow gum-chewers discreetly tucked their wads into their cheeks before anyone else could be singled out. It is a small and very human story, and it is also, improbably, the first step on a path that ends at the strangest infectious agent biology has ever found, one that carries no genes at all. The road there runs through a founding father’s sugar idealism, a doomed wagon train, and a funeral ritual on the far side of the world. The thread holding it together, the whole way, is what people choose to eat.

From Washington’s Mount Vernon it is a short hop to Thomas Jefferson’s Monticello, and to one of Jefferson’s lesser-known enthusiasms: maple sugar. Jefferson was close with Dr. Benjamin Rush, the Philadelphia physician, signer of the Declaration, and one of the more outspoken abolitionists among the founders. In the early 1790s Rush threw himself into an unusual campaign to get Americans to sweeten their tea with sugar tapped from maple trees instead of cane. The appeal was practical and moral at once. Maple was cheap and homegrown. Cane sugar was grown by enslaved people in the West Indies; maple sugar could be made by free farm families in a New England spring. Rush’s stated aim was to cut the demand for slave-grown sugar and, by doing so, to help strangle slavery itself. To prove the product, he staged a kind of scientific tea party, pouring cups sweetened with equal measures of cane and maple sugar for guests including Alexander Hamilton, who agreed the maple was every bit as sweet.

Jefferson signed on with gusto. He urged farmers to keep a stand of sugar maples the way they kept an apple orchard, and he put his own land to the test, buying sixty saplings from a Long Island nursery and having them planted in a grove at Monticello. Virginia, it turned out, is too warm for good sugaring, and the trees died, nearly all of them within a few years. The grand scheme to make the young United States a sugar power never materialized. But maple sugar caught on enough in the North to become a real provision, the kind of thing a family would pack for a long trip. And in the 1840s, a great many families were about to take the longest trip of their lives.

They packed it onto the Oregon Trail, the roughly two-thousand-mile overland route from the Missouri River to the valleys of Oregon, beaten in by fur traders and trappers and considered passable for wagons by about 1840. The provisioning lists were enormous: hundreds of pounds of flour and bacon per adult, beans, a keg of rendered fat, sometimes live cattle driven alongside, and yes, sugar, including maple sugar carried west from the world Jefferson and Rush had been promoting. Canned food existed but was new and expensive, so most families ate what they could carry and what they could forage, picking berries along the way.

As the traffic grew, so did a small industry of guidebooks promising the fastest way west. A guidebook is a wonderful thing until it is selling you a shortcut. That is the hinge this whole story turns on, because one such promoter talked one such party into one such shortcut, and the result is the most infamous tragedy of the westward migration.

In the spring of 1846, a group that would swell to about eighty-seven people, several families plus stragglers met along the way, set out from Springfield, Illinois, bound for California in a column of ox-drawn wagons. Near Fort Bridger they chose to leave the established trail for a new route called the Hastings Cutoff, named for a self-promoting guide named Lansford Hastings who had publicized it as a faster way to California before ever traveling it himself. It was not faster. It dragged them across the Wasatch Range and the punishing salt flats west of the Great Salt Lake and added something like a hundred and twenty-five miles, costing them weeks they did not have. By the time they reached the Sierra Nevada in late autumn, an early and ferocious snowfall had sealed the passes, and the party became trapped near Truckee Lake, now called Donner Lake, for the winter of 1846 to 1847.

What followed is the part everyone half-remembers. Supplies and livestock gave out. A small group strapped on snowshoes in mid-December and set off to find help; the first relief party did not reach the camps until the middle of February, more than three months after they were trapped. People later photographed the tree stumps the survivors left behind and found them tall, not because anyone climbed the trees but because that was the depth of the snow when the trees were cut. Of the roughly eighty-seven in the party, about forty-eight survived, most of them women and children. Some survived, in the end, by eating the bodies of those who had already died. The Truckee River and the lake that bears the party’s name are the quiet backdrop to that fact.

Cannibalism, it turns out, comes in categories. Exocannibalism is eating someone from outside your own group, the stuff of warfare and, in the modern imagination, of serial killers. Endocannibalism is eating someone from within your community or family, and one form of it is ritual or funerary cannibalism, eating the dead not to survive but as a rite, a way of honoring or mourning them. The Donner Party is survival cannibalism, a horror born of circumstance. But the case that turns this episode from history into microbiology is funerary, and it belongs to the Fore people of the eastern highlands of Papua New Guinea.

The Fore practiced a funerary cannibalism in which deceased relatives were mourned by being consumed, a way of returning the dead person to the community and helping free the spirit rather than leaving the body to rot in the ground. The work of preparing and eating the dead fell largely to the women and the children, who handled the corpses and ate the organs, including the brain. That detail, who ate the brain, is the whole story, though no one knew it yet. Through the middle of the twentieth century, Fore women and children began dying by the thousands of a mysterious affliction no outsider could explain.

They called it kuru, from a Fore word meaning to shake or tremble, because that is how it began: a creeping loss of coordination, an unsteady stance, a gait that wobbled and failed. The medical term is cerebellar ataxia, the breakdown of the brain’s coordination center, and footage of afflicted Fore shows people unable to stand without support or to bring their own fingertips together. It had a crueler nickname too, the laughing sickness, because in its course patients’ faces would contort into something between a smile and a grimace, bursts that looked like laughter wrung out of failing muscles. The disease moved through three stages, from the still-walking phase, to the inability to walk unaided, to a final stage of helplessness, difficulty swallowing, and wasting, with death usually following within roughly three months to two years of the first symptoms. It was, without exception, fatal, and there was no treatment then and none now.

The most unnerving feature was the delay. Kuru’s incubation period averaged around a dozen years and could stretch to fifty or more, so a meal eaten in childhood could kill someone in middle age. The generally accepted explanation is that the epidemic grew from a single seed: sometime around 1900, one person on the edge of the kuru region likely developed a spontaneous case of something like Creutzfeldt-Jakob disease, a rare brain disorder that arises on its own in roughly one or two people per million each year. That person died, was eaten according to custom, and the agent began to spread, concentrated in the brain and passed on most efficiently to the women and children who consumed it. The men, who largely abstained from the brain, were mostly spared.

In 1957, an American researcher named Daniel Carleton Gajdusek, working with the physician Vincent Zigas, brought kuru to the world’s attention, and Gajdusek went to live among the Fore to study it. Back in the United States he did the experiment that cracked it open: he inoculated chimpanzees with brain tissue from people who had died of kuru, and after a long wait, one of the chimps, named Daisy, developed the disease. That proved kuru was transmissible, an infection rather than a purely inherited or environmental condition, even though no one could find any virus, any bacterium, any DNA or RNA at all in the infectious material. For that work Gajdusek shared the Nobel Prize in 1976. A veterinarian named William Hadlow had already noticed, in 1959, that kuru looked uncannily like scrapie, a fatal brain-rotting disease of sheep that leaves the animals scraping themselves raw against fences.

The agent with no genes turned out to be a prion, a word coined by the neurologist Stanley Prusiner, who won his own Nobel in 1997 for the idea. A prion is simply a protein gone wrong. The normal protein, called PrP, sits on the surface of nerve cells in a tidy, coiled, alpha-helical shape. The dangerous version is the same protein folded into a flat, sheet-like form (the abbreviation for that misfolded form borrows from scrapie, the first such disease recognized). What makes it infectious is templating: the misfolded protein, on contact, coaxes a normal copy to refold into the bad shape, which then converts the next one, and the next. Picture a circle that touches a square and becomes a square. In many cases of kuru the misfolded proteins pile up into dense clumps called amyloid plaques, and the disease punches the brain full of microscopic holes until it looks like a sponge, which is why this whole family of illnesses is called transmissible spongiform encephalopathy.

A schematic of templating, not a measured structure of PrPSc. It does not show how the misfolded protein injures the brain.

Sources: Prusiner, Science (1982); Gajdusek, Gibbs and Alpers, Nature (1966); Alpers, Philosophical Transactions of the Royal Society B (2008); Prusiner, PNAS (1998).

If that sounds like an exotic problem confined to a remote highland, the modern world got its own lesson in the 1980s. British cattle began staggering and collapsing with a new prion disease, bovine spongiform encephalopathy, better known as mad cow disease. The cause was a kind of industrial cannibalism: cattle were being fed protein supplement rendered from the remains of other animals, including other cattle, recycling the misfolded protein through the herd. It then crossed into people who ate contaminated beef, surfacing in the mid-1990s as variant Creutzfeldt-Jakob disease and killing 178 people in the United Kingdom. One scholar called it the first man-made epidemic, since it took a human decision, feeding ground-up animals to herbivores, to move the pathogen into our food. The Fore had stumbled into the same trap with their funerals, just by hand.

From a gum-chewing scolding in George Washington’s house, through Jefferson’s maple scheme, the Hastings Cutoff, and a Fore funeral, the chain ends at a protein that copies a bad fold. The connective tissue is food and the customs built around it. Looking at the Fore as uniquely self-inflicted misses the point the show lands on: every culture has practices that quietly harm it, the oversized soda and the cigarette as much as the funeral feast.

When the Fore gave up cannibalism around 1960, the disease stopped finding new victims, and the last known death was in 2005. In the survivors, researchers found a protective variant of the prion gene that had spread under the pressure of the epidemic, natural selection caught in living memory. A prion is not alive. It does not breathe, eat, or carry a single gene. It is a shape, copying itself one fold at a time. One flew over the kuru nest, and infection had to stretch to cover a protein with no genome.

The episode audio states seven things that are wrong. The corrections:

Hurricane Matthew

Vibrio cholerae

This episode opens, as several of ours do, with one of us confessing to a food crime: a lifelong indifference to breakfast. Not bacon, which is fine, but eggs of any preparation, sausage, the whole genre. The complaint is that breakfast asks too much effort for too little flavor; in the time it takes to build an omelet you could have made something worth eating. Which is how we ended up talking about steak and eggs, and about Waffle House, the diner that, by the company’s own telling, has sold well over a hundred million T-bone steaks since it opened in 1955 and may move more of them than anyone. That is a strange enough fact on its own. Stranger still is where it leads. From a T-bone at a Waffle House the path runs through a diner that refuses to close, the storms that diner is used to measure, and out to John Snow’s water-borne cholera and one of the deadliest epidemics of our century. The link the whole way is water.

Waffle House is open all the time. Twenty-four hours, every day, holidays included, and famously through weather that closes everything around it. That stubbornness is not an accident; it is a disaster plan. The chain pre-positions backup generators and fuel, trains staff for storm conditions, keeps stripped-down menus ready for when the power or the water goes, and runs jump teams of employees from unaffected towns who can drive in and keep a location open so local workers can go home to their families. Stay open through enough hurricanes and you become a kind of instrument. By most accounts, in 2004, after Hurricane Charley, an emergency manager named Craig Fugate, who would later run FEMA, noticed that the first thing open after a storm was almost always a Waffle House, and he turned it into a rule of thumb now known as the Waffle House Index. It has three settings: green for the full menu, yellow for a limited one running on a generator or bottled water, and red for closed. Green means the area is basically fine. Red means it is very bad. As Fugate put it, if you get there and the Waffle House is shut, that is exactly where you go to work. A breakfast chain became a gauge of how hard a place got hit.

What the index is really tracking is infrastructure: power, fuel, and clean water. That is the heart of this story, because the danger from a hurricane is not only the wind. The category number, one through five, measures wind speed, but a great deal of the death and damage comes from the storm surge, the wall of seawater that pressure and wind shove ahead of the storm onto the coast. The deadliest natural disaster in American history was a surge: the 1900 hurricane that put up to fifteen feet of water over low-lying Galveston, Texas, and killed somewhere between six thousand and twelve thousand people. Galveston answered by building a seventeen-foot seawall and physically raising much of the city’s grade, which is why some of those old buildings now have a sealed-up former first floor. Storms like that are expected to get worse, not necessarily more frequent but more intense and wetter, as a warming ocean hands them more energy and rising seas give every surge a higher running start.

Hurricanes get blamed for spreading disease, and they do. The usual picture is floating corpses seeding plague in the floodwater. The driver of outbreaks after a disaster is more often the collapse of ordinary life: people displaced into crowded shelters, knocked away from clean water, working toilets, healthcare, and decent food. Take away safe water and functioning sanitation and you get the conditions fecal-oral diseases need.

A fecal-oral disease means you get sick by ingesting food or water contaminated with traces of human waste. A single unwashed hand can start a case; the outbreaks that remake public health come from scale, a community whose sewage ends up in the same river it drinks from, which is what a flood or a failed water system produces. The fecal-oral disease that shaped epidemiology is cholera.

Cholera is an acute diarrheal infection caused by the bacterium Vibrio cholerae, a gram-negative, curved little comma-shaped rod, a facultative anaerobe that does fine with or without oxygen. It is violently efficient: somewhere between twelve hours and five days after you swallow enough of it, it can produce profuse watery diarrhea, vomiting, cramping, and, untreated, death. Worldwide there are still on the order of one to four million cases a year and tens of thousands of deaths, rare in the United States but endemic in places without reliable clean water. The organism was first seen under a microscope in 1854 by an Italian anatomist named Filippo Pacini, whose work was published locally and promptly ignored by the wider world; international credit went instead to the German bacteriologist Robert Koch, who studied an outbreak in Egypt in 1883 and grew the bacterium in pure culture in India the next year. The most important figure in the cholera story never saw the bacterium at all.

In the summer of 1854, the same year Pacini was peering at the comma-shaped bacterium, a vicious cholera outbreak tore through the Soho district of London, around Broad Street and Golden Square, and killed more than six hundred people in a matter of weeks. The medical establishment of the day blamed miasma, bad air, a noxious vapor thought to rise from filth and carry disease. A local physician named John Snow did not buy it. He suspected the route was water, and he did something then unusual: he gathered data. He mapped the deaths house by house and watched them cluster, tightly, around a single public water pump on Broad Street. He persuaded the skeptical local authorities to remove the pump’s handle, and the outbreak, already fading, broke. In a separate and even more careful study, he compared London households served by two water companies, one drawing from a sewage-tainted stretch of the Thames and one from cleaner water upstream, and showed the cholera deaths tracked the dirty supply. He had identified how the disease spread decades before anyone could name the germ that caused it. His peers largely rejected him at the time. He is now considered a founder of epidemiology, and the Broad Street pump is its origin myth: the moment someone proved that the thing killing people was not the air but the water.

The reason cholera kills so fast is cholera toxin. Once Vibrio cholerae reaches the small intestine, it secretes that toxin, and the gene for it is not even the bacterium’s own core genome; it is carried by a virus, a bacteriophage, that infected the microbe. The toxin is built in two parts: a ring of subunits that latches onto the lining of your gut, and an active piece that slips inside the cell and jams one specific switch. It chemically locks a signaling protein into the on position, which leaves an enzyme called adenylate cyclase stuck running, which floods the cell with a messenger molecule called cAMP, which throws open a channel in the cell membrane. Through that open gate the cell dumps chloride, sodium, and water out into the gut, and stops absorbing them back. The result is the hallmark of severe cholera: torrents of pale rice-water diarrhea, as much as a liter an hour, until the body is wrung dry. People do not die of cholera so much as of dehydration, sometimes within hours.

A schematic of that path, not a measurement from one patient or one water sample. Not every Vibrio cholerae strain makes this toxin. The toxin genes are carried by a filamentous phage. The Soho map is a separate count of deaths, on the cholera data page. Diphtheria toxin is carried by a phage too, in Antitoxin Togo, Please.

Sources: Holmgren, Lönnroth and Svennerholm, Infection and Immunity (1973); Gill and Meren, PNAS (1978); Field, Journal of Clinical Investigation (2003); Waldor and Mekalanos, Science (1996).

The toxin does not kill the intestinal cells. It hijacks one signaling switch and leaves a separate pathway intact, a doorway that pulls sodium and water back into the body when sugar is there to power it. That pathway is the basis of oral rehydration therapy: a carefully balanced mix of salt, sugar, and clean water that lets the gut reabsorb fluid as fast as the toxin pours it out. It costs pennies, it can be made at home, and it has saved tens of millions of lives. The treatment for the deadliest part of cholera is the right glass of water, which is why it is so costly when the water itself is the problem.

In January 2010, a magnitude-7 earthquake leveled much of Port-au-Prince and killed well over a hundred thousand people. Ten months later, cholera appeared in Haiti for the first time in about a century. Because it followed the quake, people understandably blamed the disaster. They were wrong, and the truth is harder to sit with. The strain was Vibrio cholerae O1, and genetic and epidemiological fingerprinting traced it to South Asia, specifically to a contingent of United Nations peacekeepers from Nepal, where cholera is endemic, whose base sanitation leaked into a tributary of the Artibonite River, the river that people living along it drink from and cook with. The disease that John Snow diagnosed in a Soho water pump had been introduced, a century and a half later, the same exact way: human waste in the drinking water. The epidemic became the deadliest in the world in recent memory, with an estimated eight hundred thousand or more cases and roughly ten thousand deaths through early 2019, when the last case was confirmed. Cholera returned to Haiti in October 2022. The United Nations did not apologize until 2016. And that same fall, Hurricane Matthew slammed into Haiti and the cases surged again, worst in the storm-battered south, the Waffle House Index logic playing out on an island with no Waffle Houses: the storm did not import the germ, it just smashed the water and sanitation that had been holding it back.

From a T-bone steak to a cholera ward, the road is water. A diner becomes a disaster gauge because staying open means the power and the clean water are back. A storm becomes an epidemic when it severs people from safe water and working toilets. In a Soho street in 1854, John Snow showed that the difference between a frightening illness and a mass grave can be sanitation; Haiti proved the same point again, expensively. One of us confessed at the top of the episode to a weakness for cholera, the bug, as a subject; consider this the confession of a cholera girl. We already know how to stop this. We worked it out before we could even see the culprit. The pump handle still comes off.

The episode audio states several things that are wrong. The corrections:

Yellowstone thermal spring

Thermus aquaticus

The fight this time was over a dead country singer. One of us holds that you cannot call yourself culturally literate without knowing Johnny Horton, the 1950s balladeer behind “North to Alaska” and “The Battle of New Orleans.” The other had never heard the name and guessed it might be a knockoff of Tim Hortons, the donut chain. We did not resolve this one either. One of Horton’s hits was a galloping little number called “Sink the Bismarck,” and that song is the thread we are going to pull. Follow it far enough and it runs from a doomed battleship through a captured code machine, a CIA drug program, and a chemist’s hallucination, all the way to a bacterium living in a Yellowstone hot spring, and from there to PCR, the polymerase chain reaction that labs still use to copy DNA.

The Bismarck and her sister ship Tirpitz were the largest battleships Germany ever built, roughly 800 feet of armor and modern guns, fast and very nearly unsinkable. Bismarck was commissioned in August 1940, and in May 1941 she steamed out on her first and only operation, code-named Rheinübung, alongside the heavy cruiser Prinz Eugen: the plan was to break into the Atlantic and strangle the convoys keeping Britain fed and armed. She was spotted almost at once. A neutral Swedish cruiser, the Gotland, saw the German ships pass through the Kattegat, the strait at the mouth of the Baltic, and reported it, and the word made its way to the British naval attache in Stockholm and on to the Admiralty.

On May 24, in the Denmark Strait between Greenland and Iceland, Bismarck met the pride of the Royal Navy, the battlecruiser HMS Hood. Less than ten minutes after the shooting started, a shell found Hood’s magazines; she exploded and sank in about three minutes, and all but three of her roughly 1,400 men died. The hunt that followed gave the song its name: sink the Bismarck. Dozens of Royal Navy ships gave chase as the damaged German battleship ran for occupied France. What finally caught her was almost comically obsolete, a flight of Fairey Swordfish biplane torpedo bombers from the carrier Ark Royal; one torpedo jammed her rudder and left her able only to steer in a slow circle. On the morning of May 27 the battleships Rodney and King George V and their cruisers closed in and pounded her while her own crew set scuttling charges. Of more than 2,200 men aboard, only about 110 were pulled from the sea. The wreck was not found until 1989, by Robert Ballard, nearly three miles down. And the Bismarck was only one ship in a far larger fight, the Battle of the Atlantic, the longest continuous campaign of the war, which before it ended would cost the Allies some 72,000 sailors and merchant seamen, 3,500 merchant ships, and 175 warships, and cost Germany about 30,000 men and 783 submarines.

One of those lost submarines mattered out of all proportion to its size. On February 12, 1940, the Royal Navy ship Gleaner crippled the U-33 with depth charges and forced its crew to abandon ship. The submarine was carrying rotor wheels for an Enigma machine, the encryption device Germany used to scramble its military orders. The crew were supposed to scatter the rotors into the sea; some did not, and the British took them from the survivors. It was exactly the piece the codebreakers had been missing.

Those codebreakers worked at Bletchley Park, the country house north of London that held Britain’s Government Code and Cypher School. They did not start from scratch. Poland’s Cipher Bureau, and above all the mathematician Marian Rejewski, had been reading Enigma through the 1930s and had even built an electromechanical aid they called the bomba; weeks before the war, the Poles handed their methods and a working Enigma replica to astonished British and French officers. At Bletchley, Alan Turing took that head start and ran with it. He led Hut 8, the section attacking the German navy’s Enigma, and with Gordon Welchman he designed the British Bombe, a machine inspired by the Polish bomba but built on a different method, which could grind through millions of possible settings far faster than any person. The intelligence all of this produced had its own code name: ULTRA. Reading the U-boats’ own messages let the Allies route their convoys around the waiting wolf packs, and it helped turn the Battle of the Atlantic. A handful of rotors off a sinking submarine had helped win an ocean.

After the war, the word ultra did not retire. It surfaced again in the 1950s as the tail end of a CIA cryptonym, MKULTRA. The link is mostly in the name, but the program that name labeled was real and notorious: a sprawling effort to find drugs and techniques for interrogation, brainwashing, and mind control. Its researchers dosed people with powerful drugs, very often without their knowledge or consent, including prisoners, hospital patients, drug users, and the agency’s own employees. The drug they kept coming back to was LSD. Frank Olson, an Army bacteriologist, was secretly given LSD at a 1953 gathering, spiraled into paranoia and depression over the following days, and fell to his death from a tenth-floor hotel window; the episode has since been told and retold, including in a documentary series. The CIA eventually decided the drug was too unpredictable to weaponize and gave up on it. By then LSD was loose in the world.

LSD, short for lysergic acid diethylamide, had been synthesized back in 1938 by the Swiss chemist Albert Hofmann at the Sandoz laboratories, where it sat ignored for five years. In 1943 Hofmann accidentally absorbed a trace of it through his skin, discovered its strange effects, and a few days later deliberately swallowed what he assumed was a tiny, cautious dose: 250 micrograms, which we now know is more than a typical recreational hit. It was, by his own notes, far stronger than he had bargained for. The thread we care about, though, is not the CIA’s interest in the drug but one of the people who eventually fell into its orbit, a brilliant and restless American chemist named Kary Mullis.

In the early 1980s Mullis worked at Cetus, a California biotech company, synthesizing oligonucleotides, the short custom strings of DNA that researchers use as tools. One night in 1983, driving up Highway 128 toward his cabin with the molecules turning over in his head, he had the idea that would win him a Nobel Prize. The idea was the polymerase chain reaction, PCR, a way to take a single stretch of DNA and copy it into millions or billions of copies in a matter of hours. The recipe is short. Heat the double-stranded DNA close to boiling so its two strands come apart. Add two short primers that bracket the exact segment you want, marking where to start and where to stop. Cool it so the primers latch on. Then let a DNA polymerase, the cell’s own copying enzyme, build a fresh matching strand out of loose nucleotides. Each cycle doubles what you have, and doubling over and over climbs into the billions astonishingly fast.

A schematic of that order, not a temperature protocol and not a count of product. A real reaction does not double the target at every cycle.

Sources: Saiki, Gelfand, Stoffel and colleagues, Science (1988); Chien, Edgar and Trela, Journal of Bacteriology (1976); Mullis and Faloona, Methods in Enzymology (1987).

There was one maddening flaw. To begin each new cycle you have to heat the sample back up to pull the strands apart, and that heat cooked the polymerase Mullis was using, which came from the gut bacterium E. coli. A cooked enzyme is a dead enzyme, so someone had to open every tube and pipette in fresh polymerase after every single cycle, by hand, for hours. Mullis showed that PCR worked, on December 16, 1983, and went on to share the 1993 Nobel Prize in Chemistry for it. He was a genuine eccentric: he credited LSD with loosening his thinking, and he also held positions far outside the evidence, doubting that HIV causes AIDS, rejecting human-caused climate change, and once recounting a late-night encounter with what he took to be a glowing raccoon. The idea he had on that drive, though, was sound. It just needed a better enzyme. And that enzyme was already sitting, undisturbed, in a hot spring a thousand miles away.

Yellowstone’s hot springs are ringed with bands of color, greens and oranges and reds, and those colors are alive: mats of microbes thriving in water hot enough to scald. In the 1960s a microbiologist named Thomas Brock went looking, with no particular application in mind, at what could possibly survive that heat. From Mushroom Spring he and his student Hudson Freeze pulled a rod-shaped bacterium they named Thermus aquaticus, the hot-water bug, and described it in 1969, more than a decade before Mullis took his drive. Because the bacterium lives at temperatures around 160 °F (70 °C), all of its machinery has to keep working there, including its DNA polymerase. That enzyme, nicknamed Taq after the organism, is thermostable: heat does not cook it.

Drop Taq into PCR and the heat problem is gone. You can heat the reaction to split the strands cycle after cycle, and the polymerase stays active through the next round, with no one opening the tubes. Pair it with a machine called a thermocycler, which just heats and cools the samples on a programmed schedule, and PCR becomes automatic, cheap, and everywhere. It is how a lab confirms a viral or bacterial infection by fishing for a gene unique to the pathogen. It is how a crime lab matches DNA from a scene to a suspect. It is how we run genetic tests, and how we sequence whole genomes by copying overlapping pieces and stitching them back together.

None of the people in this chain set out to invent PCR. The sailors who failed to ditch the rotors, the Poles and Turing breaking Enigma, Hofmann misjudging a dose, Brock asking what lives in hot water: none of them could have told you their work would one day let a doctor catch an infection from a swab, or let a lab read a virus letter by letter. The enzyme at the center came from a bacterium found by a scientist who wanted to know what survives in hot springs. A war ballad got us to a hot spring, and the hot spring gave PCR an enzyme that does not cook. We landed on the right Taq. Now, about that game of Battleship you still insist you won.

The episode audio states several things that are wrong. The corrections:

Coronavirus

Asian palm civet

One of us refuses to drink iced coffee. Not “tried it and disliked it,” but flatly will not, on principle, having never once let a cold cup near his mouth. The other finds this indefensible. You can write off a food you have never tasted in some cases, the argument goes, but coffee is not one of those things; you have to actually try it before you get an opinion. We have never settled it, which is fine, because the fight was never really about coffee. It was about where a cup of coffee can take you. Follow this one far enough and it runs from a goat in the Ethiopian highlands to the most expensive coffee on earth, to a small caged animal in an Asian market, and from there into the first new plague of the twenty-first century. Here is how a luxury beverage and a deadly virus end up sharing the same animal.

No one knows exactly where coffee began, but the plant is native to the highland forests of Ethiopia, and the legend everyone tells starts there. A goatherd named Kaldi, the story goes, noticed his goats growing strangely lively after eating the bright cherries of a certain shrub, tried the berries himself, and felt the same lift. The tale is almost certainly embroidered, but the Ethiopian origin is not: nearly every cultivated arabica plant traces back to those forests. What is solid history is what happened next. By the fifteenth century coffee was being cultivated and traded across the Red Sea in Yemen, brewed in Sufi communities to keep worshippers awake through long religious rituals, and within a hundred years it had spread through Egypt, Persia, Syria, and the Ottoman world, served in public houses called qahveh khaneh.

Europe got its first real taste through trade with the Ottomans, and the first European coffeehouses opened in Venice around 1645. The drink moved north, and the colonial powers, wanting supplies of their own, went looking for live plants. The Dutch got there first, obtaining seeds in the Yemeni port of Mocha and planting them on Java, the island that lent coffee one of its nicknames. From there cuttings traveled to the Caribbean, to Vietnam, and across the Americas, until coffee grew on nearly every tropical shore. The coffeehouse became its own institution along the way, a place to argue, trade gossip, and read the news. One London coffeehouse run by a man named Edward Lloyd became such a dependable spot for shipping talk that it grew, in time, into the insurance market still known as Lloyd’s of London. Coffee, in other words, was always less a drink than a network.

Goats are not the only animals that go for coffee cherries, and one of them produces the strangest luxury in the trade. The Asian palm civet, Paradoxurus hermaphroditus, is a small nocturnal mammal of tropical Asia, about the size of a house cat with a face somewhere between a cat and a young bear. Left to itself, it climbs through coffee plantations at night and eats only the ripest cherries. The beans pass through its gut, where digestive enzymes and bacteria break down some of their proteins and blunt the coffee’s bitterness, and come out the other end largely intact, bundled in droppings that, we are sorry to report, look a great deal like a candy bar studded with beans. Collected, cleaned, and roasted, those beans become kopi luwak, from the Indonesian words for coffee and for the civet, and a single pound of it can sell for more than a hundred dollars.

For a while this was good news for the civet. Then the money got ahead of the animal. Producers began trapping civets, packing them into cages, and force-feeding them whatever cherries came to hand, which ruined the very thing that made the coffee special, the animal’s wild instinct for the perfect berry. The caged trade was cruel and it was sloppy. Civets were also sold in the live-animal markets of southern China, where they sat stacked among other wild species in close, stressful, unsanitary quarters. Hold that picture of the caged civet in the crowded market. It is where the second half of this story begins.

In November 2002, in Guangdong province in southern China, a patient turned up at a hospital in Foshan with a severe atypical pneumonia, the kind not caused by the usual bacterial suspects. Within a few months there were hundreds of cases and dozens of deaths across the province’s hospitals, many of them among the nurses and doctors doing the treating; one early spreader, a seafood seller admitted to a Guangzhou hospital, infected some thirty staff by himself. Chinese authorities were slow to share what was happening, delaying reports to the World Health Organization and withholding details from the rest of the country, a silence that almost certainly cost lives and let the outbreak build.

It went global through a single hotel floor. In February 2003 a Guangdong doctor named Liu Jianlun, who had been treating these patients and was feeling unwell, traveled to Hong Kong for a family wedding and checked into room 911 of the Metropole Hotel. In one night he infected other guests on the ninth floor, and they carried the disease out into the world before he was hospitalized and died. The earliest and most severe outbreaks, in Hong Kong, Toronto, Hanoi and Singapore, trace back to that one floor. An elderly woman flew home to Toronto and seeded a Canadian cluster that would put thousands into quarantine; a young Singaporean named Esther Mok, also a Metropole guest, survived but started the outbreak there; and a Chinese American businessman named Johnny Chen fell ill after flying to Hanoi and was admitted to the French Hospital, where the staff began dropping one after another.

The doctor who examined Chen was Carlo Urbani, an Italian infectious disease specialist working for the WHO. He recognized within days that this was something new and dangerously contagious, alerted the WHO, and pressed the Vietnamese authorities to isolate patients and screen travelers, buying the world precious time. Then he boarded a flight to a conference in Bangkok, fell ill somewhere over the journey, and was hospitalized as soon as he landed. He died on March 29, 2003. The disease he had flagged for the world would be named Severe Acute Respiratory Syndrome, SARS. Before it was contained that summer, it reached more than 8,000 people in 29 countries and regions and killed 774, a fatality rate of roughly one in ten.

When investigators went looking for where the virus had come from, they went back to the markets. In the spring of 2003, samples taken from masked palm civets sold as food in Guangdong turned up coronaviruses almost identical to the one making people sick. The masked palm civet is not the same animal as the coffee-making Asian palm civet, but it is a close cousin in the same family, roughly the way humans are related to the other great apes. Authorities responded by culling more than a thousand civets in Guangdong in early 2004 and restricting their trade, and for a time the civet looked like the source.

The fuller picture turned out to sit one step further back. The civet strains did not carry enough genetic variety to be the original reservoir; they looked like a virus that had only recently arrived from somewhere else. That somewhere was bats. Cave-dwelling horseshoe bats were found to harbor a rich diversity of closely related SARS-like coronaviruses, and the consensus that emerged is that the virus lived quietly in bats, spilled into civets, and there, in the amplifying crush of the market, picked up the changes that let it infect and pass between people. The animal that bridged bat and human was a civet, the same sort of creature prized, some two thousand miles away, for making the world’s most expensive cup of coffee.

SARS is caused by a coronavirus, and the name is really a description. Studded across the surface of each viral particle, each virion, are spike proteins that stand out like the points of a crown or the rays of a halo. Those spikes are not decoration; they are the key. To get into a human cell the spike has to fit a specific lock on the cell’s surface, a receptor called ACE2, short for angiotensin-converting enzyme 2, which the body normally uses to help regulate blood pressure. The fit between that viral key and that human lock is the whole game: change the spike too much and it no longer turns the lock, which is exactly the barrier the bat and civet versions had to clear before they could open human cells.

The rest of the virus is built for travel. It is an enveloped virus, wrapped in a fatty membrane much like the one around our own cells, a trait shared by many of the viruses that spread well in coughs and sneezes. Its genome is a single strand of RNA only about thirty thousand letters long, a rounding error next to the three billion in ours, and it mutates at a moderate clip, enough now and then to change the shape of a protein or what that protein can grab. Coronaviruses are not exotic; milder ones cause a good share of ordinary colds. There was never a cure for SARS, only supportive care to keep patients breathing while their bodies fought it off, which is why the outbreak was beaten not in a laboratory but in the field, through quarantine, isolation, and the unglamorous machinery of public health.

What made SARS new was not really the virus. It was the speed. Coffee took centuries to cross from Ethiopia to the Americas, carried by ship a few months at a time. SARS crossed the same oceans in a day, because a sick traveler who could not tell a new plague from a bad cold could carry it from a hotel in Hong Kong to Hanoi, Toronto, and Singapore before the fever even set in. A disease that once would have burned itself out in one isolated place now boarded a plane. A sky full of SARS.

So the thread holds, all the way back to a petty argument about iced coffee. A luxury brewed from the droppings of one civet and a virus carried to us by another are not two stories but one, joined at a cage in a crowded market. The next time someone offers you the most expensive coffee in the world, you will at least know the whole strange family it comes from. We still cannot agree on iced coffee. But that, it turns out, was never really the point.

The episode audio states several things that are wrong. The corrections: