The History of Splenda the Best-Selling Artificial Sweetener in America

On that late-summer day in 1975, Phadnis was told to test the powder, but he misunderstood; he thought that he needed to taste it. And so, using a small spatula, he put a little of it on the tip of his tongue. It was sweet --- achingly sweet. “When I reported my findings to Les, he asked if I was crazy, “ Phadnis remembers. “ How could I taste compounds without knowing anything about their toxicity?” Before long, though, Hough was so delighted with the substance that he dubbed it Serendipitose and tried putting some in his coffee. “Oh forget it, “ he said, when Phadnis reminded him that it might be toxic. “We’ll survive!”

Over the next year, Hough and Phadnis worked with the British sugar company Tate & Lyle to make more than a hundred chlorinated sugars, finally settling on one that had three chlorine atoms and was about six hundred times as sweet as sugar. “It isn’t of any use as an insecticide,” Hough told me recently. “That was tested.” But it had proved useful as a food. In its pure form, it is known as sucralose. When mixed with fillers and sold in bright-yellow sachets, it’s known as Splenda, the best-selling artificial sweetener in America.

People will eat almost anything, it seems, as long as it’s sweet. And, until fairly recently, this mental programming served them just fine. When Columbus introduced cane to the New World, the anthropologist Sidney Mintz has noted, sugar was an exotic luxury. Most Europeans had never eaten sugar, but they quickly developed a taste for it. By 1700, the Americas had become a vast sugar mill and the English were eating four pounds per person per year. By 1800, they were eating eighteen pounds; by 1900, ninety pounds. But nowhere was the rise of sugar as dramatic as in the New World. Last year, the average American consumed about a hundred and forty pounds of cane sugar, corn syrup, and other natural sugars -- fifty percent more that the Germans or the French and nine times as much as the Chinese.

Artificial sweeteners are both a symptom of this craving and an attempt to curb it. Some two hundred million Americans now use them, but rarely with much enthusiasm. Like Splenda, the most popular products were all discovered by accident; none of them taste much like sugar; and there is no final verdict on their safety. Saccharin was found over dinner in 1879, by a chemist who was working with coal-tar derivatives and forgot to wash his hands properly. It’s used in Sweet’ N Low in the United States, where it was listed as a possible carcinogen until 2000, and is still banned as a food additive in Canada. Aspartame, which is used in Equal and most diet sodas, was found in 1965, by a chemist who was testing new drugs for gastric ulcers and licked his fingers before picking up a piece of paper. A recent study by the National Cancer Institute found no evidence that aspartame caused cancer in rats. Sucralose was declared safe by the Food and Drug Administration in 1998, but most taste researchers I talked to won’t eat it. “I look at that structure and I have an irrational fear of it,” one of them said. “I’ve seen the safety studies, and you feed it to rats and mice forever and nothing happens. But it just scares me.”

Inventing a sweetener with even a little of sugar’s appeal is one of the hardest tasks in food science. It’s less like imitating the taste of Coke or vanilla than like trying to imitate water--another simple but astonishingly versatile compound. Sugar’s sweetness only begins to explain our devotion to it. You can freeze it, cook it, candy it, and caramelize it. It adds bulk to baked goods and helps them to brown. Sugar is a powerful preservative. It triggers the taste buds almost instantly, fades quickly without aftertaste, and has a voluptuous mouth feel. Even its potency can’t easily be improved. Artificial sweeteners may be thousands of times as sweet by volume, but their flavor loses intensity with repeated tasting. Sugar stays sweet.

Of course, sugar can also make you fat, put diabetics into a coma, and make your children run screaming in circles. In the past twenty years, it has helped to double the number of obese Americans and rotted untold millions of teeth. Sugar may be the single most unhealthy part of the American diet, yet until recently there was little hope of finding a tastier, more wholesome substitute. Thousands of chemicals were known to be sweet, but none of them tasted or behaved like sugar. If all artificial sweeteners had been discovered by accident, the reason was simple: No one knew how to make them from scratch.

In the early nineteen-eighties, the NutraSweet Corporation launched what would become a twenty-year effort to find a better sweetener. The company’s first product, aspartame, had been introduced in 1981 and had quickly become America’s best-selling sugar substitute. Because it was composed of two common amino-acids, aspartame was advertised, at first, as an almost natural product--”If you’ve had bananas and milk, you’ve eaten what’s in NutraSweet.” But aspartame also had problems. It couldn’t be used in baking, because it broke down under high heat, and it tended to lose its sweetness over time. NutraSweet had higher hopes for its successor. “Sugar is in trouble,” the company’s president, Robert Shapiro, said. “You can’t do anything to improve the product.”

The search for Sweetener 2000, as it was later known, turned into a race of sorts between two groups on opposite sides of the Atlantic. In Mount Prospect, Illinois, outside Chicago, Shapiro had assembled a team of more than a hundred chemists, taste researchers, and support staff. The NutraSweet scientists knew that the tongue’s taste buds are made up of clumps of fifty to a hundred cells, layered like an onion and tipped by chemical receptors. But they weren’t sure how many kinds of receptors there were, how they functioned, or how they sent their signals to the brain. So the team worked by trial and error. They began by building computer models of existing sweetener molecules, modified them a few atoms at a time, and then fed the most promising candidates to mice. If the mice didn’t “drop dead,” as one member of the team put it recently, the compounds were sent to an outside lab for a series of sweetness tests, culminating in a fifteen-member taste panel.

In France, meanwhile, Claude Nofre and Jean-Marie Tinti, two chemists at Claude Bernard University, in Lyons, were searching for sweeteners in a more old-fashioned way. Nofre had come to NutraSweet’s attention when he invented a potent substance called super-aspartame. Although it didn’t seem to be toxic, super-aspartame had some unsettling similarities to cyanide, so NutraSweet gave Nofre a grant to try again. The funding wasn’t enough to pay for computer modeling or elaborate taste tests, so Nofre and Tinti relied on Tinker toy-type molecular models, on Nofre’s instincts for the inner workings of taste buds, and on the evidence of their own tongues. “We tasted all of the compounds, all of them!” Nofre told me. “NutraSweet thought we were crazy.” After seven years, the Mount Prospect team and identified around five hundred new sweeteners; Nofre and Tinti eventually found more than two thousand.

The winning compound, now called neotame, is a version of aspartame to which Nofre and Tinti attached a chain of carbon and hydrogen atoms. Neotame doesn’t lose its flavor or break down when you cook with it, and it’s about eight thousand times sweeter than sugar. In 2002, after more than a hundred human and animal studies, neotame was approved by the F.D.A. This year, it made its first appearance in American stores -- in Ice Breakers candies, SunnyD Reduced Sugar orange drink, Mr. Fizz sodas from Wal-Mart-- and will now be judged by the only standard that really matters: how it tastes.

NutraSweet’s test kitchen in Chicago, where its new products are developed, is on the ninth floor of a labyrinthine office building downtown. Known as the Sweet Spot, it’s staffed by a team of “beverage formulators” in white lab coats and looks like the set of an industrial video from the nineteen-fifties--fluorescent lights, vinyl floor, and Formica counters bristling with scales, homogenizers, refractometers, pH meters, and other electronic gizmos. When I went there in January, for a taste test of neotame, I was met by Craig Petray, the company’s chief executive, and Ihab Bishay, the director of research. Petray, who is forty-five, has a linebacker’s build, a clean-shaven head, and an almost soldierly faith in his sweeteners. While he cited statistics and sales strategies, Bishay-- a plump, genial Egyptian with a black goatee and rectangular glasses-- quietly laid out the scientific evidence.

He began by setting nine plastic cups in front of me, each one filled with water flavored with a different artificial sweetener. The cups were a miniature overview of the global sweetener market. If their sizes had corresponded to actual consumption, the one with saccharin in it would have been by far the largest, with about sixty percent of the total volume. The one with aspartame would have been next, with about twenty percent, and the rest would have been tiny. Sucralose, the sweetener in Splenda, accounts for less than five percent of worldwide consumption, although it has conquered the tabletop market in the United States. Neotame claims less than one percent, most of that in China. Sugar wasn’t in the lineup, but its cup would have dwarfed the rest: last year alone, almost three hundred billion pounds of sugar was consumed worldwide--about three times more than all other sweeteners combined. Clearly, neotame had a lot of convincing to do.

The tasting that followed felt a bit like “To Tell the Truth” -- the old game show where celebrity panelists had to identify the bona-fide contestant from among a group of impostors. One by one, the sugar substitutes gave themselves away with the equivalent of a thick foreign accent or a laughably inauthentic manner. Saccharin had sugar’s quick punch and lack of aftertaste but was accompanied by a mildly bitter or metallic edge. So were cyclamate and acesulfame-K, two sweeteners often used in place of saccharin. Aspartame and sucralose didn’t have any off flavors, but their sweetness came on too slowly and stuck around too long. Tagatose, a low-calorie carbohydrate made from lactose, tasted nearly identical to sugar --its chemical structure is quite similar -- but even in moderate quantities acted as a laxative. “Don’t drink too much,” Petray said. “The plane trip will not be comfortable,” Bishay added.

Worst of all were the sweeteners found in health-food stores. Stevia, made from a South American shrub of the same name, seemed to combine all the failings of its artificial cousins: slow onset, heavy aftertaste, bitterness, and other disagreeable flavors. (Monsanto, NutraSweet’s former owner, considered genetically modifying stevia in order to make it less bitter, but the project was self-defeating: stevia’s only real selling point is its natural quality.) Thaumatin, derived from the West African katemfe fruit, was said to be two thousand times as sweet as sugar, but it tasted like nothing at first. Then, slowly, like the opening chords of a Wagner overture, the flavor began to build: deep and faintly dissonant, with echoes of licorice and cough syrup. Chewing-gum makers often add thaumatin to round off flavors and make them last longer, but taken alone, at a high concentration, it was truly awful.

“This is the whole world of sweeteners,” Bishay said. “These are the primary candidates, and there’s no really good one. They are not the answer.” That left neotame. Its taste was strong, clean, and straightforward -- like aspartame, with a deeper bottom -- bit it took a while to register on the tongue and lingered forever. Its molecules seemed to lock onto the taste receptors so stubbornly that later arrivals had nowhere to bind; by the fourth or fifth sip, the water was nearly tasteless. If I’d been chewing stevia leaves or katemfe fruit all my life, neotame might have seemed like a great improvement. But I hadn’t. When I sipped some sugar water afterward, the taste came as a blessing; bright, vivid, quick-blossoming, with unexpected hints of fruit and flowers. It wasn’t hard to see why Europeans in the Middle Ages considered sugar not a staple but a spice.

“Pretty good sweetener, this sugar,” Petray said, with a resigned smile. NutraSweet had given up on finding a true replacement for it, he added: there would be no successor to neotame. Instead, the company was focusing on blending sweeteners in order to minimize their weaknesses, then mixing them with sugar to get the same taste with fewer calories. This was where neotame came into its own. It cost a tenth as much as sugar and half as much as sucralose. It had no off-putting flavors, and it heightened the tastes of other foods much as sugar did. “We think the sweetener world in the future is going to be a blend world,” Petray said.

He set three more cups in front of me, filled with orange soda. One or two were made with natural sugars; one or two were sweetened with a mixture of sugars, neotame, and acesulfame-K. My job was to pick the soda that tasted different from the two others. “Triangle tests” like this were harder than straight comparisons, Petray said, and this blend was one of NutraSweet’s best. Acesulfame-K’s quick bite offset neotame’s slow, lingering sweetness. Still, I had no trouble telling the sodas apart. Petray and Bishay tried again, with different blends, replacing as little as twenty percent of the sugars. They even brought out two cakes that Bishay’s wife had made -- one with sugar, one with a neotame blend. The cakes managed to fool me, but in every other case I easily spotted the outlier. “In a taste panel, this would pass every time,” Petray told me at one point, a bit exasperated. “As an average consumer, there is no way you would say it was different.” But to my taste, that day, there was still no replacement for sugar.

Humans are connoisseurs of sweetness. No other species is so particular. Cats can’t taste sugar; neither can many dogs. Most other animals can’t taste artificial sweeteners. (We know this, in part, thanks to an enterprising Swiss anthropologist named Dieter Glaser, who has offered them to fish, hedgehogs, tree shrews, primates, elephants, horses, cows, sheep, pigs, dogs, cats, mice, birds, reptiles, kangaroos, and swamp wallabies.) But after a million years of devoted omnivorous ness -- of climbing trees, swatting at bees, and scouring the landscape for any hint of sugar -- people can taste every sweetener, and they can tell them apart.

Charles Zuker, a molecular biologist at the University of California at San Diego, thinks that a craving so subtle and so deep can’t be satisfied by trial and error. You can’t just take chemical potshots at the tongue, he says. You have to isolate its taste receptors, understand how they work, and find ways to trigger them. Like Petray, Zuker doesn’t think that sugar can be replaced. But the right chemical might do something even better, he says. It might make foods with less sugar taste just as sweet.

Zuker, whose name means “sugar” in Yiddish, give or take a consonant or two, jokes that he was destined to do this work. He was born in Chile, the grandson of Polish and Russian refugees from the Holocaust. He played with microscopes when most boys were playing soccer, and went to Jesuit school, although his family was Jewish. By the age of fifteen, he was in college, by nineteen he was attending graduate school at M.I.T., and by twenty-three he had earned his doctorate. Now forty-eight, he had dedicated his life to the senses, scientifically and otherwise. He owns a house on the cliffs above Del Mar, drives a Porsche Twin Turbo to work, and is married to his college sweetheart, a Spanish instructor at the university who resembles the actress Charo.

The first time I saw Zuker, he was giving the keynote address at a conference on the senses, in Washington, D.C. He was slouched at the podium in a suede jacket and weathered jeans, facing an audience of neuroscientists in suits. A pair of reading glasses was perched on his slender, balding head, and the Rolling Stones’ tongue-and-lips logo was projected on a screen behind him. Until quite recently, he told the audience, the prevailing view of how taste receptors work was “idiotic.” Most scientists believed that each cell in a taste bud carried receptors for all five basic flavors -- sweet, sour, salty, bitter, and umami, the savory taste of protein. When food or drink passed over them each cell sent an elaborately coded message to the brain, like a shortwave broadcast in five languages. “It made no sense,” Zuker told me later. “Sweet and bitter prompt fundamentally different behaviors. Sweet is to determine caloric content; bitter is to warn you against toxins. It’s the difference between life and death.” Why would the same cell send both signals?

Zuker’s laboratory in San Diego is stocked both with bottles of hazardous chemicals and with bags of exotic treats. On the day I visited, one of the tables was piled with Warheads: hard candies so sour that the package bore a picture of a puckered face with a mushroom cloud exploding from it. When I asked Zuker about them, he leaned back in his chair and curled one arm around his head. “The students bring in new sensory experiences every week,” he said. “We had spicy ginger gummies last time.” He told me to help myself to a bag of shriveled Chinese wolfberries, on the counter. The berries had a sweet, strangely meaty flavor. After I’d had a few dozen, I asked him if there was any danger in eating too many. “We’ll find out,” Zuker said, grinning.

As a rule, today’s students refuse to offer their tongues in the service of science, so Zuker keeps a large colony of mutant mice in a building across the street. Some of them can taste bitter but not sweet, other sweet but not bitter, and so on-- more than a thousand mice in all. In the past few years, Zuker and the geneticist Nicholas Ryba, at the National Institutes of Health, together with a succession of graduate and postdoctoral students, have used these animals to help identify the taste receptors. They began, in 1998, by scanning RNA sequences from the tongue and homing in on the most likely genes. They then bred mice that lacked the genes, to see how their tastes were affected. Within two years, Zuker’s team had located the entire family of twenty-six bitter receptors. By 2001, they’d found the receptors for sweet and umami as well. Zuker’s team wasn’t alone in making some of these discoveries -- biologists at Harvard, the Monell Chemical Senses Center, in Philadelphia, and other labs also found receptors -- but his mice provided the decisive evidence. It was an astonishing feat of genetic sleuthing. “I was stupefied,” Claude Nofre told me. “I thought they would be found in the year 300.” To celebrate, the journal Cell put a chocolate cake on its cover, flanked by two mice.

The taste cells that Zuker found were much simpler than biologists had imagined. Instead of bristling with every kind of receptor, each cell was tuned to a single frequency: some cells detected sweet, others bitter, still others umami. (The receptors for salty and sour have yet to be found.) There seemed to be no elaborate signals to encode and decode, no danger that the brain might misread that little part about arsenic in a lengthy molecular report about mangoes and bananas. The tongue, like any good electrical system, was wired with well-insulated, well-labeled lines. All the brain had to do was follow instructions.

To demonstrate, Zuker led me to a small, tiled room with two cages full of mice. One set had white fur, red eyes, and untampered genes; the others were brown-haired, black-eyed mutants. The mice had not had water for a while, so they were extremely thirsty. We were going to offer them three bottles filled with different liquids, Zuker said, and he invited me to take a taste. The first was just water, the second was sugar water, and the third -- “Don’t slurp it!” Zuker said-- was the most unpleasant thing I’d ever had. The water had been dosed with the world’s bitterest known substance, denatonium benzoate, a “freak molecule” that’s often put into pesticides and household cleaners to prevent accidental poisonings. After a minute or so, it showed no signs of releasing it’s fierce grip on my tongue. I popped a Warhead in my mouth to try to blast it off. This was a mistake. The denatonium combined with the acids in the candy to trigger something like a mushroom cloud inside my head. “Let that teach you a lesson about the biology of taste,” Zuker said.

The white mice didn’t like the denatonium either. When Zuker gave them the two other liquids, they lapped them up so quickly that their tongues were a blur, but they could stand only a lick or two of denatonium before running to the other side of the cage. The mutants, though, had had their bitter receptors knocked out genetically, so the denatonium was tasteless to them. Zuker’s team had also engineered mutants that reacted to bitter as if it were sweet. They’d even taken a mouse’s sweet taste cell and inserted a receptor for a taste-less artificial compound that the mouse then guzzled as if it were sugar water. “The animal kingdom sees the world as a binary choice,” Zuker concluded. “Something is sweet not because it tastes sweet but because it activates cells in your brain that say, ‘This is good.’” Bitter foods activate cells that signal, “This is bad.” “It is an absolutely gorgeous example of Darwinian evolution,” Zuker said. “Otherwise, you eat it and you die.”

Zuker’s theory strikes some researchers as simplistic; isolated taste cells seem to respond to a number of tastes, not just those from a single receptor. But few doubt the practical value of the receptors he had found. Pharmaceutical firms have long used certain receptors to search for new compounds and to create targeted drugs with fewer side effects. The same technology can now be used to search for new sweeteners.

Zuker has little interest in doing this work. “I am pure basic scientist,” he told me. “I’m trying to figure out how the brain works, not how to make chemicals taste better.” But he isn’t averse to letting others do it for him. In 1998, Zuker and a small group of other scientists and businessmen founded Senomyx, a biotech firm devoted to taste. Senomyx now has a patent on the use of the sweet receptor, patents pending on the umami and bitter receptors, and partnerships with Kraft, Coca-Cola, Nestle, Campbell’s, and Cadbury-Schweppes. In the next few years, Senomyx and its partners hope to reinvent the flavors in our food without anyone really noticing.
The Senomyx laboratories are about a ten-minute drive from Zuker’s lab, in a low-slung stucco building in Torrey Pines. Inside, about seventy-five scientists pursue what is known as "high-throughput screening" : a modern, hyper-accelerated version of Nofre and Tinti's taste test. They start by creating what they call artificial taste buds: human cells, with a single taste receptor, engineered with a fluorescent dye that lights up only when the receptor is triggered. The cells are placed in clear plastic trays divided into three hundred and eighty-four wells, each a couple of millimeters wide. A robotic arm with three hundred and eighty-four nozzles squirts a different compound into every well. Whenever a cell lights up -- about a one-in-a-thousand occurrence -- it is registered by a fluorometric sensor and tallied by a computer. In Nofre and Tinti's days, testing this many samples would have taken months. At Senomyx, it takes less than five minutes.

"When you think about how many things have been tasted, it's not that many," Mark Zoller, the company's head of research, told me. "Usually what people do is create derivatives of what they already know: if you have aspartame, you create neotame. We can go in completely from left field, with no preconceptions about what can be sweet. We can throw it all at the receptor and let the results speak for themselves." In the past four years, Senomyx has tested more than twenty million samples. Its sweetener jprogram has identified the three most promising classes of chemicals, whittled those down to two candidates, and tinkered with them in the lab, adding some atoms for stability, some for potency. The final product won't be a new sugar substitute -- "How many of those do we need?" Zucker says. It may not even have any taste. All it will do is amplify the taste of sugar.

Taste potentiators, as they're called, are not entirely new to the food industry. The ingredient list on a can of soup or a hunk of processed cheese sometimes includes a substance called IMP, a few entries below MSG, or monosodium glutamate. MSG is to umami what sugar is to sweet: the taste in its purest, most familiar form. IMP's singular virtue is its synergy with MSG. Like the sweet receptor, the umami receptor has multiple binding sites. IMP attaches to one spot, MSG to another; together, they fit so snugly that their effect is multiplied. Add a little IMP to a soup with MSG in it, and the umami taste will increase roughly ten-fold. "It's like a hearing aid," Holler told me. It turns up the volume.

Senomyx has found four new umami potentiators in the course of its chemical trawling, all of them more effective than IMP, all recently declared safe by the F.D.A. (The first products containing them should appear later this year.) The company's two sweet potentiators aren't quite as far along. The best one is known as Substance 951. If you add only a few parts per million of it to a soda, you can take out forty percent of the sugar and the soda will taste as sweet. But Senomyx is still working on making it stronger and on improving or eliminating its taste. (I wasn't allowed to try it.) Zoller says that the compound should be on the market by next year, but most consumers won't be aware of it. Like the new umami potentiators, Substance 951 will be used in such tiny quantities that it won't have to be listed on labels. Instead, it will join all the other "natural and artificial flavors" that float through our foods, ignored by all but the most obsessive ingredient-watchers, and quietly do the work that sugar once did.

Walking through the labs at Senomyx, watching taste cells turn on and off in their little plastic wells, I was reminded of a wooden display case that I'd seen at NutraSweet. The case was fitted with three glass vials, all with different sweeteners measured in proportions of equal strength. The first vial held forty grams of sugar and was nearly full. The second had a thin layer of aspartame--about a fifth of a gram. The last was labeled "neotame" and looked empty. I had to hold it up to the light to see the faint glimmer of powder inside. You could call this progress. The sweeter the chemical, the fewer of its molecules will wind up in our bloodstreams. And, if that chemical can also help curb obesity and diabetes, so much the better. "We consume too many calories and we don't have to, " Craig Petray told me. "If products can taste the same and have twenty-five or thirty percent less sugar, that's a start."

The same argument, of course, has always been made for artificial sweeteners. Like the dream of the paperless office or the superhighway that will untangle traffic for good, it presumes that there is a natural limit to our needs-- that humanity's sweet tooth can be satisfied. Yet our sweet receptors evolved in environments with so little sugar that they may not have a shutoff point. Elizabeth Cashdan, an anthropologist at the University of Utah, has seen African bushmen pick fruit apart for the barest trace of pulp. "And honey! What they will go through for a taste of honey is just incredible," she says.

A number of biologists have tried to gauge the depth of our appetite for sugar over the years. Newborns, they've found, are already fixated on sweetness. If you put some sugar on a latex nipple, an infant will suck it longer and harder than a plain nipple. Give her a drop of sweet water when she's crying and her heartbeat will slow, her face will relax, and her brain activity will fall into a "hedonically positive" pattern. (Hugs and pacifiers have a similar effect, but not as lasting.) According to the biologist Julie Mennella, at the Monell Chemical Senses Center, sugar seems to trigger the release of opiates in the brain, both bringing pleasure and blocking pain. (When Mennella asked children to stick their hands in icy water, those with some sugar water in their mouths kept their hands in longer.) Adults who are offered drinks of different sugar concentrations tend to reach a "bliss point" at about nine teaspoons per cup--fifty percent sweeter than the average soft drink. Children prefer eleven teaspoons per cup, and they'll take it even stronger. "For babies, the fundamental rule is: the sweeter the better," Monell's director, Gary Beachamp, told me. "There is nothing that is too sweet."

Beauchamp has also tried to study the opposite tendency: the less sugar people eat, the less of a taste for it they have. He had to abandon the experiment, though, because his subjects couldn't stick to their sugar-free diets. (They were much better at abstaining from salt; and he did find that their appetite for it diminished.) The human palate is nothing if not adaptable, but it's hard to lose your craving for sugar when it's found in everything from wheat bread to spaghetti sauce to macaroni and cheese. Artificial sweeteners, far from diminishing that appetite, often seem to reinforce it. Americans ate about twenty-four pounds of sugar substitutes per person last year, nearly double what they did in 1982, yet sugar consumption rose about twenty-five percent in the same period. The trend is strongest among blacks and Hispanics--they like their food with about ten percent more sugar than whites do, studies by Susan Schiffman, a medical psychologist at Duke, have shown--and weakest among Asians. As Schiffman puts it, our taste for sweeteners is being "upregulated."

The closest analogy may be what has happened to out sense of pitch. In 1740, when Handel rang his tuning fork, an A above middle C had a frequency of four hundred and twenty-two hertz. Throughout the nineteenth century, orchestras where tuning it higher, straining to fill larger and larger halls and make their sound just a little more brilliant. These days, when Lorin Maazel rings his tuning fork, that same A gives a steady pitch of four hundred and forty hertz, but some conductors in Germany and Austria have gone up another five hertz. In music, if you go too high, stings snap and voices crack. In the matter of sweets, the only real limit is exhaustion: when Zuker offers sugar water to his mice, they keep on drinking until their tongues can hardly move.

One afternoon, Zuker drove me to an Asian strip mall a few miles from the university. A Chinese graduate student had recommended a place there called Spicy City. The restaurant had a bright-red carpet, devil masks and chili peppers on the walls, and a menu of daunting authenticity. Zuker glanced over the choices with a kind of sadomasochistic glee. "Host spicy pork blood with black sea cucumber, squid, and golden mushroom," he said. "Is it kosher?" He finally settled on a dish called Husband and Wife, made of cold sliced beef and tripe drenched in chili oil. He liked to experience "very distinct, acute sensory events," he explained. "The best is to eat something spicy, naked."

Nothing in the biology of taste could really explain the appeal of Spicy City. Our tongues are wired for yes and no, good and bad, not for "It tastes like it's rotting but I can't stop eating it" or "It's incinerating my flesh and I find this oddly pleasurable." Any mouse knows to shun bitter and spicy foods as poisonous, but Zuker is no mouse. Like all of us, he is part rationalist and part sensualist --though perhaps he pushes both sides to an extreme. He has taken driving lessons at a racecar track, keeps a cellar full of Chilean wines, and built a swimming pool on the cliffs beside his house with a vanishing edge that seems to tumble into the void. The best part of being human, he knows, is ignoring what your body tells you from time to time.

The rise of sugar since Columbus sometimes seems destined to turn us all into lab animals, dutifully gorging on sweets. But Columbus did more than bring sugarcane to the New World. He also brought a few things home. Cocoa was popular long before it was sweetened, and chilies are now eaten by a quarter of the world's adults every day. You can explain this in pharmacological terms (cocoa contains caffeine), in hygienic terms (chilies kill bacteria0, or as a function of peer pressure. But the best explanation may be what the psychologist Paul Rozin, at the University of Pennsylvania, call "benign masochism." We eat chilies, Warheads, and bitter greens, and drink bitter tonics and bitter coffee, for the same reason that we ride roller coasters and watch horror films: to fool the body into thinking it's in danger, and then enjoy the adrenal ride. Our taste buds may tell us that nothing is as good as sugar, but our minds can be taught to know better.

"We like to experience the edge, to push our sensory systems to the limit," Zuker told me later, in his car. "Whether it's tasting things or driving very fast cars, we like to enjoy things we should not enjoy." He took a winding road down the coast, past a cluster of surfers in wetsuits paddling into the Pacific, and a pair of hang gliders getting ready to throw themselves from the cliffs of La Jolla. When the road peeled away from the shore, he shifted into third gear and accelerated to a hundred and fifteen, hurtling past a red truck that nearly turned into our lane. He glanced at me, backed deep into my seat with my hands clutching the armrests, and laughed. "You're a pussy!" he shouted. Then he jammed the stick shift forward and threw the car into the next turn.