As Dr. Sid Gilman approached the stage, the hotel ballroom quieted with anticipation. It was July 29, 2008, and a thousand people had gathered in Chicago for the International Conference on Alzheimer's Disease. For decades, scientists had tried, and failed, to devise a cure for Alzheimer's. But in recent years two pharmaceutical companies, Elan and Wyeth, had worked together on an experimental drug called bapineuzumab, which had shown promise in halting the cognitive decay caused by the disease. Tests on mice had proved successful, and in an initial clinical trial a small number of human patients appeared to improve. A second phase of trials, involving two hundred and forty patients, was near completion. Gilman had chaired the safety-monitoring committee for the trials. Now he was going to announce the results of the second phase.

Alzheimer's affects roughly five million Americans, and it is projected that as the population ages the number of new cases will increase dramatically. This looming epidemic has added urgency to the scientific search for a cure. It has also come to the attention of investors, because there would be huge demand for a drug that diminishes the effects of Alzheimer's. As Elan and Wyeth spent hundreds of millions of dollars concocting and testing bapineuzumab, and issued hints about the possibility of a medical breakthrough, investors wondered whether bapi, as it became known, might be "the next Lipitor." Several months before the Chicago conference, Barron's published a cover story speculating that bapi could become "the biggest drug of all time."

One prominent investor was known to have made a very large bet on bapi. In the two years leading up to the conference, the billionaire hedge-fund manager Steven A. Cohen had accumulated hundreds of millions of dollars' worth of Elan and Wyeth stock. Cohen had started his own hedge fund, S.A.C. Capital Advisors, with twenty-five million dollars in 1992 and developed it into a fourteen-billion-dollar empire that employed a thousand people. The fund charged wealthy clients conspicuously high commissions and fees to manage their money, but even after the exorbitant surcharge investors saw average annual returns of more than thirty per cent. S.A.C. made investments in several thousand stocks, but by the summer of 2008 the firm's single largest position was in Wyeth, and its fifth-largest was in Elan. All told, Cohen had gambled about three-quarters of a billion dollars on bapi. He was famous for making trades based on "catalysts"—events that might help or hurt the value of a given stock. Sid Gilman's presentation of the clinical data in Chicago was a classic catalyst: if the results were promising, the stocks would soar, and Cohen would make a fortune.

Gilman had not wanted to make the presentation. Seventy-six years old and suffering from lymphoma, he had recently undergone chemotherapy, which left him completely bald—like the "evil scientist in an Indiana Jones movie," he joked. But Elan executives urged Gilman to participate. He was a revered figure in medical circles, the longtime chair of neurology at the University of Michigan's medical school. In Ann Arbor, a lecture series and a wing of the university hospital were named for him. His C.V. was forty-three pages long. As a steward for the fledgling drug, he conveyed a reassuring authority.

But soon after Gilman began his thirteen-minute presentation, accompanied by PowerPoint slides, it became clear that the bapi trials had not been an unqualified success. Bapi appeared to reduce symptoms in some patients but not in others. Gilman was optimistic about the results; the data "seemed so promising," he told a colleague. But the investment community was less sanguine about the drug's commercial prospects. One market analyst, summarizing the general feeling, pronounced the results "a disaster." The Chicago conference was indeed a catalyst, but not the type that investors had expected.

It appeared that Cohen had made an epic misjudgment. When the market closed the following day, Elan's stock had plummeted forty per cent. Wyeth's had dropped nearly twelve per cent.

By the time Gilman made his presentation, however, S.A.C. Capital no longer owned any stock in Elan or Wyeth. In the eight days preceding the conference, Cohen had liquidated his seven-hundred-million-dollar position in the two companies, and had then proceeded to "short" the stocks—to bet against them—making a two-hundred-and-seventy-five-million-dollar profit. In a week, Cohen had reversed his position on bapi by nearly a billion dollars.

Gilman and Cohen had never met. The details of the clinical trials had been a closely guarded secret, yet S.A.C. had brilliantly anticipated them. Cohen has suggested that his decisions about stocks are governed largely by "gut." He is said to have an uncanny ability to watch the numbers on a stock ticker and intuit where they will go. In the assessment of Chandler Bocklage, one of his longtime deputies, Cohen is "the greatest trader of all time."

But federal authorities had a different explanation for S.A.C.'s masterstroke. More than three years after the Chicago conference, in December, 2012, prosecutors in New York indicted a young man named Mathew Martoma, who had worked as a portfolio manager for Cohen. They accused him of using confidential information about bapi to engineer the most lucrative insider-trading scheme in history. According to the indictment, Martoma had been receiving secret details about the progress of the clinical trials for nearly two years and, ultimately, obtained an early warning about the disappointing results of the second phase. His source for this intelligence was Sid Gilman.

In 1977, after completing medical school at U.C.L.A. and teaching at Harvard and Columbia, Gilman was recruited to run the neurology department at the University of Michigan. He moved to Ann Arbor with his first wife, Linda, and their two sons. Gilman's marriage unravelled in the early eighties, and the older son, Jeff, developed psychological problems. Jeff committed suicide in 1983, overdosing on pills in a hotel room near campus. Gilman had experienced tragedy before: his father had walked out on the family when he was a boy, and his mother later committed suicide. After Jeff's death, Gilman seems to have dealt with his despair by throwing himself into his job. "The man worked himself to distraction," one of his many protégés, Anne Young, who went on to become the chief of neurology at Massachusetts General Hospital, told me. In 1984, Gilman married a psychoanalyst named Carol Barbour, but they never had children, and though his surviving son, Todd, attended the University of Michigan, they eventually became estranged, leaving him with no ties to his former family.

Over the years, however, Gilman became a father figure to dozens of medical residents and junior colleagues. "Helping younger people along—that was a constant," Kurt Fischbeck, a former colleague of Gilman's who now works at the National Institutes of Health, told me. Gilman was "incredibly supportive" of younger faculty, Young said. "He would go over grants with us, really putting an effort into it, which is something chairs rarely do."

One day in 2002, Gilman was contacted by a doctor named Edward Shin, who worked for a new company called the Gerson Lehrman Group. G.L.G., as it was known, served as a matchmaker between investors and experts in specialized industries who might answer their questions. "It was kind of ridiculous that the hedge-fund business got so much information by asking for favors . . . when it would certainly pay," the company's chief executive, Mark Gerson, told the Times.

Shin proposed that Gilman join G.L.G.'s network of experts, becoming a consultant who could earn as much as a thousand dollars an hour. Gilman was hardly alone in saying yes to such a proposal. A study published in the Journal of the American Medical Association found that, by 2005, nearly ten per cent of the physicians in the U.S. had established relationships with the investment industry—a seventy-five-fold increase since 1996. The article noted that the speed and the extent of this intertwining were "likely unprecedented in the history of professional-industrial relationships."

Gilman read the JAMA article, but disagreed that such arrangements were objectionable. In an e-mail to Shin, he explained that investors often offered him a fresh perspective on his own research: "Although remuneration provides an incentive, the most attractive feature to this relationship (at least for me) is the exchange."

"You're a little too niche."Buy or license »

Gilman's university salary was about three hundred and twenty thousand dollars a year, a sum that went a long way in Ann Arbor. As he took on more paid consultations, he began supplementing his income by hundreds of thousands of dollars a year. Acquaintances did not notice any abrupt change in his life style: Gilman wore elegant clothes, but otherwise he and his wife appeared to live relatively modestly. "He was not a flashy guy who revelled in expensive toys," Tim Greenamyre, a former student, who now runs the Pittsburgh Institute for Neurodegenerative Diseases, told me. Gilman counselled Greenamyre and other colleagues to avoid even the appearance of a conflict of interest in their professional dealings, and he made a point of telling people that he never invested in pharmaceutical stocks. The consulting, he later maintained, was simply "a diversion."

In the summer of 2006, Gilman received a call from Mathew Martoma, who explained that he had recently joined S.A.C. and was focussing on health-care stocks. They spoke about Alzheimer's remedies, and specifically about bapineuzumab. Although Martoma had no medical background, he was attuned to the scientific intricacies at play. His mother and his wife, Rosemary, were physicians, and he had a long-standing interest in Alzheimer's, dating back to his childhood, in Florida, when he volunteered as a candy striper at a local hospital. He and Gilman talked for more than two hours. Afterward, Martoma asked G.L.G. to schedule another consultation.

S.A.C. was a notoriously intense place to work. Its headquarters, on a spit of land in Stamford, Connecticut, overlooking the Long Island Sound, are decorated with art from Cohen's personal collection, including "Self," a refrigerated glass cube, by Marc Quinn, containing a disembodied head sculpted from the artist's frozen blood. It was nearly as frigid on the twenty-thousand-square-foot trading floor, which Cohen kept fiercely air-conditioned—employees were issued fleece jackets with the S.A.C. monogram, for keeping warm. The atmosphere was hushed, with telephones programmed to blink rather than ring, but a curious soundtrack could be heard throughout the building. As Cohen sat at his sprawling desk, before a flotilla of flat-screen monitors, and barked orders for his personal trades, a camera—the "Steve cam"—was trained on him, broadcasting his staccato patter to his subordinates. Cohen is not a physically imposing man: he is pale and gnomish, with a crooked, gap-toothed smile. But on the Steve cam he was Oz.

When S.A.C. first approached Martoma about a job, he was ambivalent. He was living in Boston, working happily at a small hedge fund called Sirios Capital Management. He knew that careers at S.A.C. followed a starkly binary narrative. Portfolio managers were given a pot of money. If their investments were consistently profitable, they became very rich very quickly. If their investments lost money, they were out of a job. Contracts at S.A.C. contained a "down and out" clause, so it was prosper or die.

Cohen likened his traders to élite athletes; for many years, he paid a psychiatrist who had worked with Olympic competitors to spend several days a week at S.A.C., counselling employees about mastering their fears. He hired high achievers who were accustomed to gruelling pressure. Martoma had studied bioethics at Duke, graduating summa cum laude. After a year working at the National Institutes of Health, where he co-authored a paper, "Alzheimer Testing at Silver Years," in the Cambridge Quarterly of Healthcare Ethics, he was admitted to Harvard Law School. He departed a year later, during the dot-com boom, and launched a startup. Next, he obtained an M.B.A. from Stanford. S.A.C. was another brand-name institution, a strong allure for someone like Martoma. After visiting the office in Stamford and spending a day shadowing Cohen on the trading floor, he accepted the job.

S.A.C. relied on portfolio managers to devise novel investment ideas. In a marketplace crowded with hedge funds, it had become "hard to find ideas that aren't picked over," Cohen complained to the Wall Street Journal, in 2006. In the business, a subtle but crucial informational advantage was called "edge." Richard Holwell, a former federal judge in New York who presided over high-profile securities-fraud cases, told me that, in order to evaluate a technology stock, hedge funds sent "people to China to sit in front of a factory and see whether it was doing one shift or two." He added, "An edge is the goal of every portfolio manager." When Cohen was asked about edge during a deposition in 2011, he said, "I hate that word." But S.A.C.'s promotional materials boasted about the firm's "edge," and Cohen provided his employees with every research tool that might offer a boost over the competition. The eat-what-you-kill incentive structure at S.A.C. put a damper on collegiality. Employees with edge had no motivation to share it with one another. But every good idea was shared with Cohen. Each Sunday, portfolio managers sent a memo to an e-mail address known as "Steve ideas," in which they spelled out their most promising leads, weighted by their level of conviction.

Martoma had always been avid about research, and he was impressed by S.A.C.'s resources. At his disposal was a boutique firm full of former C.I.A. officers who could monitor the public statements of corporate executives and evaluate whether they were hiding something; S.A.C. also had a "buffet plan" with the Gerson Lehrman Group, giving Martoma unrestricted access to thousands of experts. From his first days in Stamford, he was interested in the investment potential of bapi. He contacted G.L.G. with a list of twenty-two doctors he hoped to consult, all of whom were involved in the clinical trials of the drug. Most declined, citing a conflict of interest; clinical investigators had to sign confidentiality agreements that constrained their ability to talk about the progress of the trials. But Sid Gilman accepted, noting, in his response to G.L.G., that he would "share only information that is openly available." On the Sunday after the initial conversation with Gilman, Martoma sent an e-mail to Steven Cohen, suggesting that S.A.C. buy 4.5 million shares of Elan stock, and noting that his conviction level was "High."

Martoma was born Ajai Mathew Thomas in 1974, and grew up in Merritt Island, Florida. His parents had emigrated from Kerala, in southern India, during the sixties. They were Christian; the name Martoma, which the family adopted around the turn of the millennium, is a tribute to the Mar Thoma Syrian Church, an Orthodox denomination that is based in Kerala. Mathew's father, Bobby, was a stern man with a sharp nose and a clipped mustache. He owned a dry-cleaning business, and placed enormous pressure on his son to succeed. Mathew obliged, excelling in school and starting a lawnmowing operation in which he outsourced the actual mowing to other kids. The oldest of three brothers, he seems to have taken naturally to the role of family standard-bearer. Childhood photographs show him grinning, with his hair neatly parted, in a tiny three-piece suit.

When Martoma's father first came to America, he was admitted to M.I.T., but he could not afford to attend. He retained a fascination with Cambridge, however, and prayed daily that his oldest son would go to Harvard. Martoma graduated from high school as co-valedictorian, but he ended up going to Duke. Shortly after Mathew's eighteenth birthday, Bobby presented him with a plaque inscribed with the words "Son Who Shattered His Father's Dream."

During college, Martoma volunteered in the Alzheimer's wing of the Duke Medical Center, and developed an interest in medical ethics. Bruce Payne, who taught Martoma in a course on ethics and policymaking, remembers him as "creased and pressed—very pre-professional." Payne wrote a recommendation letter for Martoma's application to business school at Stanford, praising his subtle readings of Sissela Bok's "Lying" and Albert Camus's "The Plague." Martoma was unusually adept at cultivating mentors. "He was ambitious—he wanted to make something of his life," Ronald Green, who supervised Martoma during his year at N.I.H. and is now a professor at Dartmouth, told me. "To some extent, I felt like Mathew was an adopted son."

At Stanford, Martoma was introduced to a young pediatrician from New Zealand named Rosemary Kurian. Strikingly beautiful, she was studying for her medical boards so that she could practice in the United States. She had grown up in a sheltered family and had never dated before. But she felt an immediate bond with Mathew: her parents were also from Kerala, and she, too, felt both very Indian and very Western. "I was just enamored with how lovely he was," she told me recently. "And he seemed to be very respectful of my parents." Her mother and father endorsed the relationship, and in 2003 Mathew and Rosemary were married, in an Eastern Orthodox cathedral in Coral Gables, Florida.

"It was your first day—why not give it another twelve years?"Buy or license »

By the time they moved to Connecticut, they had one child and Rosemary was pregnant with a second. She stopped working, but she was very involved in advancing Mathew's career. "Mathew didn't just do that job by himself," she told me, with a smile. He worked perpetually. "It was heads-down, tails-up, twenty-four-seven kind of work." Martoma rose at 4 a.m. to keep up with the European health-care markets, then worked until the market in New York closed. After spending a few hours with the children, he put in another shift, sitting in bed with his laptop while Rosemary fell asleep beside him. He had numerous investment prospects, but bapi was the most promising, and it became an obsession. "As a portfolio manager, you live by your ups and downs," Rosemary said. "These stocks, they're your babies, and you're following them and you're nurturing them." The fixation became a running joke, and her conversations with him were often punctuated by the word "Bapsolutely!"

Rosemary never met Sid Gilman, but throughout the fall of 2006 Martoma arranged frequent consultations with him about bapi. Much later, in court, Gilman recounted this phase in their relationship as a kind of intellectual seduction. They spoke for hours about the trials for various Alzheimer's drugs. "Every time I told him about a clinical trial, he seemed to know a good deal about it," Gilman testified. "The more I told him about each of the trials, the more he wanted to know." Gilman found himself wishing that his students in Ann Arbor were as bright and curious as Martoma.

That October, Gilman had plans to visit New York on other business, and Martoma arranged to meet with him at S.A.C.'s offices in Manhattan. In an e-mail to G.L.G., Martoma specified that he wanted the meeting "to be with just me and dr. gilman alone." The appointment was at lunchtime, and when Gilman was shown into the room he was pleased by a small courtesy—an array of sandwiches. Martoma walked in, broad-shouldered and genial, with close-cropped black hair and long eyelashes that gave his face a feline aspect. He was "very, very friendly," Gilman recalled. Martoma complimented him on "the previous consultations we had."

According to G.L.G.'s records, Gilman and Martoma had forty-two formal consultations over two years. Gilman consulted with many other investors during this time, and Martoma spoke to many other doctors, but neither spoke to anyone else with nearly the same frequency that they did with each other. It seemed to Gilman that Martoma shared his passion for Alzheimer's research, and regarded the efforts to create an effective drug as much more than a matter of financial interest. In e-mails, Martoma had a tendency to slip into the first-person plural, using "we" when discussing how medical professionals treated people with the disease.

Gilman also got the impression that Martoma wanted to be friends. Martoma proposed that they have coffee after meetings of the American Academy of Neurology. He talked to Gilman about his family's emigration from India, and about how he and Rosemary had had their children in rapid succession. In e-mails, he sent his best wishes to Gilman's "better half." Gilman called Martoma "Mat," but, even when they were speaking almost daily, Martoma always addressed him as "Dr. Gilman." Once, when Gilman was travelling in Istanbul, he forgot about a scheduled consultation. Unable to reach him, Martoma had his assistant make multiple calls to try to track the doctor down. Eventually, a hotel employee discovered Gilman by himself, reading, and alerted him to the calls. "I was in a foreign country, and he couldn't find me," Gilman testified. "It was touching."

Later, Gilman had trouble pinpointing just when his relationship with Martoma crossed into illegality. But he recalled a moment when Martoma asked, repeatedly, about the side effects that one might expect to see from bapi. "I didn't quite recognize it for what I think it was, which was an attempt to find confidential information," Gilman said. Initially, he offered theoretical responses, but Martoma "persisted in wanting to know what really happened," and finally the answers "slipped out." Gilman told him how many patients and how many placebo cases had experienced each adverse effect. While he was talking, Martoma periodically asked him to slow down, so that he could transcribe the numbers.

In 1942, lawyers in the Boston office of the Securities and Exchange Commission learned that the president of a local company was issuing a pessimistic forecast to shareholders and then offering to purchase their shares. What the president knew, and the shareholders didn't, was that earnings were on track to quadruple in the coming year. He had edge, which allowed him to dupe his own shareholders into selling him the stock at far below its real value. Later that year, the S.E.C. established Rule 10b-5 of the Securities Exchange Act, making insider trading a federal crime. At the time, one of the commissioners remarked, "Well, we're against fraud, aren't we?"

In the ensuing decades, however, enforcement of this prohibition has been inconsistent. Some academics have suggested that insider trading is effectively a victimless crime, and should not be aggressively prosecuted. At least privately, many in the financial industry agree. But in 2009, when Preet Bharara took over as United States Attorney for the Southern District of New York, with jurisdiction over Wall Street, he made it a priority to curb this type of securities fraud. The problem had become "rampant" in the hedge-fund industry, Bharara told me, in part because of a prevailing sense that the rewards for insider trading were potentially astronomical—and the penalty if you were caught was relatively slight. "These are people who are in the business of assessing risk, because that's what trading is, and they were thinking, The greatest consequence I will face is paying some fines," Bharara said. His strategy for changing their behavior was to throw a new variable into the cost-benefit equation: prison. Agents from the F.B.I. and the S.E.C. began asking investment professionals to identify the biggest malefactors. Peter Grupe, who supervised the investigations at the F.B.I., told me that all the informants were "pointing in the same direction—Stamford, Connecticut."

Rumors about insider trading had circulated around Steven Cohen since his first years in the business. As a young trader at a small investment bank called Gruntal & Company, he was deposed by the S.E.C. in 1986 about suspicious trades surrounding General Electric's acquisition of R.C.A. Cohen asserted the Fifth Amendment and was never indicted, but, during the nineties, as his fund became extraordinarily profitable, observers and rivals speculated that he must be doing something untoward. Like Bernard Madoff's investment firm, S.A.C. enjoyed a level of success that could seem suspicious on its face. "A lot of people assumed for years that S.A.C. was cheating, because it was generating returns that didn't seem sustainable if you were playing the same game as everyone else," the manager of another hedge fund told me.

When Cohen was growing up, as one of eight children in a middle-class family in Great Neck, New York, his father, who owned a garment factory in the Bronx, brought home the New York Post every evening. Cohen read the sports pages, but noticed that there were also "these other pages filled with numbers." In an interview for Jack Schwager's book "Stock Market Wizards" (2001), he recalled:

I was fascinated when I found out that these numbers were prices, which were changing every day. I started hanging out at the local brokerage office, watching the stock quotes. When I was in high school, I took a summer job at a clothing store, located just down the block from a brokerage office, so that I could run in and watch the tape during my lunch hour. In those days, the tape was so slow that you could follow it. You could see volume coming into a stock and get the sense that it was going higher. You can't do that nowadays; the tape is far too fast. But everything I do today has its roots in those early tape-reading experiences.

Cohen was never a "value investor"—someone who makes sustained commitments to companies that he believes in. He moved in and out of stocks quickly, making big bets on short-term fluctuations in their price. "Steve has no emotion in this stuff," one of his portfolio managers said in a deposition last year. "Stocks mean nothing to him. They're just ideas, they're not even his ideas. . . . He's a trader, he's not an analyst. And he trades constantly. That's what he loves to do."

"I discovered coffee!"Buy or license »

The business model at S.A.C., though, was based not on instinct but on the aggressive accumulation of information and analysis. In fact, as federal agents pursued multiple overlapping investigations into insider trading at hedge funds, it began to appear that the culture at S.A.C. not only tolerated but encouraged the use of inside information. In the recent trial of Michael Steinberg, one of Cohen's longtime portfolio managers, a witness named Jon Horvath, who had worked as a research analyst at S.A.C., recalled Steinberg telling him, "I can day-trade these stocks and make money by myself. I don't need your help to do that. What I need you to do is go out and get me edgy, proprietary information." Horvath took this to mean illegal, nonpublic information—and he felt that he'd be fired if he didn't get it.

When Cohen interviewed job applicants, he liked to say, "Tell me some of the riskiest things you've ever done in your life." In 2009, a portfolio manager named Richard Lee applied for a job. Cohen received a warning from another hedge fund that Lee had been part of an "insider-trading group." S.A.C.'s legal department warned that hiring Lee would be a mistake, but Cohen overruled them. (Lee subsequently pleaded guilty to insider trading.)

White-collar criminals tend to make soft targets for law enforcement. "The success rate at getting people to coöperate was phenomenal," Peter Grupe told me. Most of the suspects in insider-trading investigations have never been arrested, nor have they contemplated the prospect of serious jail time. When Michael Steinberg was waiting for the jury in his trial to pronounce a verdict, he fainted in open court. So the authorities approached hedge-fund employees, one by one, confronting them with evidence of their crimes and asking them what else they knew. Because the suspects weren't anticipating being under surveillance, the F.B.I. could tail them for weeks. Then one day, as a suspect headed into a coffee shop and prepared to place his usual order, an agent would sidle up and place the order for him.

The tactics echoed the approach the F.B.I. had used to dismantle the New York Mob. The plan was to arrest low-level soldiers, threaten them with lengthy jail terms, and then flip them, gathering information that could lead to arrests farther up the criminal hierarchy. Over time, agents produced an organizational chart with names and faces, just as they had with La Cosa Nostra. At the top of the pyramid was Steven Cohen.

In 2010, F.B.I. agents approached a young man named Noah Freeman, who had been fired by S.A.C. and was teaching at a girls' school in Boston. Freeman became a key witness. Asked in court how often he had attempted to obtain illegal edge, he replied, "Multiple times per day." According to an F.B.I. memo, "Freeman and others at S.A.C. Capital understood that providing Cohen with your best trading ideas involved providing Cohen with inside information."

When Martoma first came to S.A.C., his due-diligence report had noted his "industry contacts" and his personal "network of doctors in the field." Through the fall of 2007, he acquired more and more Elan and Wyeth stock, and Cohen followed his lead, supplementing the money that Martoma was investing from his own portfolio with funds from Cohen's personal account. That October, Martoma e-mailed Cohen that bapi was on track to start Phase III trials soon, and that they would make up "the MOST COMPREHENSIVE ALZHEIMER'S PROGRAM to date."

S.A.C. had a proprietary computer system, known as Panorama, which allowed employees to monitor the company's holdings in real time. Employees checked Panorama incessantly, and many noticed the scale of the bet that Martoma, a relatively junior portfolio manager, was making, and the fact that Cohen was backing him. Because of the open plan in the Stamford office and the simulcast from Cohen's desk, people could watch as Martoma approached the boss and murmured recommendations. A portfolio manager named David Munno, who had a Ph.D. in neuroscience, was skeptical about bapi's prospects. He didn't like Martoma, and didn't understand the source of his conviction. At one point, he wrote to Cohen, wondering whether Martoma actually knew something about the bapi trial or simply had "a very strong feeling."

"Tough one," Cohen replied. "I think Mat is the closest to it."

It's impossible to know exactly how Martoma buttressed Cohen's confidence in bapi. Portfolio managers at S.A.C. often wrote detailed explanations to support trading recommendations, but, when it came to bapi, Cohen and Martoma preferred to talk. Martoma's e-mails to his boss often consisted of a single line: "Do you have a sec to talk?" "Do you have a moment to speak when you get in?" Whenever Munno pressed Cohen on how Martoma knew so much about bapi, Cohen responded cryptically. "Mat thinks this will be a huge drug," he wrote to Munno at one point. On another occasion, he explained simply that "Mat has a lot of good relationships in this area."

A second portfolio manager, Benjamin Slate, shared Munno's concerns, suggesting, in one e-mail, that it was "totally unacceptable to bet ½ billion dollars on alzheimers without a real discussion." In a message to Slate a month before the Chicago conference, Munno complained that Martoma was telling people he had "black edge." In subsequent legal filings, S.A.C. has claimed that Munno and Slate coined the term "black edge," as "humorous commentary." But, according to subsequent filings by the Department of Justice, "black edge" was "a phrase meaning inside information."

Initially, Gilman may have "slipped" when he divulged secret details to Martoma, but as their friendship continued the malfeasance became more systematic. Whenever Gilman learned about a meeting of the safety-monitoring committee, Martoma scheduled a consultation immediately afterward, so that Gilman could share whatever new information he had obtained. Apart from consultation fees, Gilman did not receive any additional remuneration from Martoma, yet he slid into ethical breaches with an ease verging on enthusiasm. At one point, Gilman proposed outright deception, suggesting to Martoma that they supply the Gerson Lehrman Group with fraudulent pretexts for meetings, in order to deflect suspicion.

On June 25, 2008, Gilman sent an e-mail to Martoma with the subject line "Some news." Elan and Wyeth had appointed him to present the results of the Phase II clinical trials at the International Conference on Alzheimer's Disease, in July. Martoma scheduled a consultation, informing G.L.G., inaccurately, that he and Gilman would be discussing therapies for multiple sclerosis. Up to this point, Gilman had been given access to the safety results of the trials, but he had been "blinded" to the all-important efficacy results. Now, in order to present the findings, Gilman would be "unblinded."

Two weeks later, Elan arranged for a private jet to fly him from Detroit to San Francisco, where the company had offices. He spent two days with company executives, crafting his conference presentation. When he returned to Michigan, an Elan executive sent Gilman an e-mail titled "Confidential, Do Not Distribute." It contained an updated version of the twenty-four-slide PowerPoint presentation that would accompany his remarks. After downloading the slide show, Gilman received a call from Martoma. They spoke for an hour and forty-five minutes, during which, Gilman later admitted, he relayed the contents of the presentation. But the material was complicated—too complicated, perhaps, to convey over the phone. Martoma announced that he happened to be flying to Michigan that weekend; a relative had died, but he had been too busy to attend the funeral, so he was going belatedly to pay his respects. Could he swing by? "Sure, you can drop in," Gilman replied. Two days later, Martoma flew from J.F.K. to Detroit, took a taxi to Ann Arbor, and met with Gilman for an hour in his office on campus. He flew back to New York that evening, without having visited his family. Rosemary picked him up at the airport.

The next morning, Sunday, Martoma e-mailed Cohen, "Is there a good time to catch up with you this morning? It's important." Cohen e-mailed Martoma a phone number, and at 9:45 a.m. Martoma called him at home. According to phone records introduced in court, they spoke for twenty minutes.

When the market opened on Monday, Cohen and Martoma instructed Phil Villhauer, Cohen's head trader at S.A.C., to begin quietly selling Elan and Wyeth shares. Villhauer unloaded them using "dark pools"—an anonymous electronic exchange for stocks—and other techniques that made the trades difficult to detect. Over the next several days, S.A.C. sold off its entire position in Elan and Wyeth so discreetly that only a few people at the firm were aware it was happening. On July 21st, Villhauer wrote to Martoma, "No one knows except me you and Steve."

"I think I see your deadline approaching."Buy or license »

Martoma said nothing to Gilman about the selloff, and a week later he flew to Chicago for the conference, bringing along Rosemary and their children, as he often did when he travelled. Gilman also did not know that Martoma had cultivated a second source connected to the clinical trials—Joel Ross, a New Jersey doctor who had been involved in the efficacy tests. Ross had plans to attend a dinner the night before Gilman's presentation, at which he and other principal investigators would be shown the full data from the trials. Martoma met Ross in the lobby of the hotel immediately after the dinner. But Ross was mystified by their interaction. He was still moderately optimistic about bapi, having seen real improvements in the patients he was supervising, but Martoma was more skeptical. "He was always very detail-oriented," Ross later testified. And, indeed, Martoma knew every detail of the results that Ross had learned at the dinner, moments earlier. Ross was unnerved: it was as if Martoma had been "in the room."

As Gilman made his presentation, the next evening, word of the ambiguous results hit the news wires. Tim Jandovitz, a young trader who worked for Martoma, watched in dismay as the news appeared on his Bloomberg terminal in Stamford. He checked Panorama, which showed that S.A.C. still held huge positions in Elan and Wyeth. Jandovitz believed that both he and Martoma had just lost more than a hundred million dollars of Steven Cohen's money—and, along with it, their jobs. The next morning, he braced himself and went to the office. But when he consulted Panorama he saw that the Elan and Wyeth shares had vanished. Some time later, Martoma informed Jandovitz that S.A.C. no longer owned the stock. The two men had worked closely together, and Jandovitz was hurt that he had been left out of the loop. Martoma explained that the decision to sell had been kept secret on "instructions from Steve Cohen."

People outside the firm were equally startled to learn that S.A.C. had turned a potential disaster into a windfall. "TELL ME MARTOMA GOT OUT OF ELAN," a friend of Jandovitz's, who worked at J. P. Morgan, said in an instant message. Jandovitz replied, "w/out getting into detail, wed and this week have been GREAT for us."

"I LOVE IT," his friend wrote.

Jandovitz agreed: "Stuff that legends are made of."

S.A.C., by dumping its shares in Elan and Wyeth and then shorting the stocks, made approximately two hundred and seventy-five million dollars in profit. That year, Martoma received a bonus of $9.3 million. The last time he saw Gilman—before the two men met again in court—was the day after the presentation, when Martoma invited Gilman to lunch at a Chicago hotel.

"Did you hear about what happened to Elan stock?" Martoma said, adding that it had plummeted. The market does not like a drug that helps only half the people who receive it, he explained.

Several months later, at the end of September, 2008, Gilman sent Martoma an e-mail with the subject heading "How are you?"

Hi Mat. I haven't heard from you in awhile and hope that all is well with you and your family. I hope that you have not been too terribly set back by the great turmoil in the markets plus the disappointing drop in Elan stock. . . . Anyway, no need to call, I have nothing new; I just wonder how you are faring.

Martoma never responded.

Regulators at the New York Stock Exchange monitor millions of transactions. Six weeks after the Alzheimer's conference, investigators flagged the huge reversal by S.A.C. before Gilman's presentation and alerted the Securities and Exchange Commission. In the summer of 2009, Charles Riely, an attorney at the S.E.C., and Neil Handelman, an investigator, began combing through hundreds of phone records, trying to identify a link between an insider at one of the drug companies and S.A.C. It took more than a year of investigation, but one day Riley and Handelman were looking through Gilman's phone records and came across the cell-phone number of Mathew Martoma. Sanjay Wadhwa, who oversaw the S.E.C. investigation, told me, "That's when we said, 'This is probably the guy.' "

By that time, federal authorities had been investigating Steven Cohen for years. But Cohen was a more elusive target than perhaps they had imagined. He described his firm as having a "hub and spokes" structure, with him at the center, pulling in information, while his specialized portfolio managers ran their accounts with a degree of autonomy. This meant that the authorities could arrest and flip low-level suspects who might describe the crooked culture of the place, but these employees would not necessarily be in a position to testify that Cohen knowingly traded on inside information.

In the summer of 2009, the F.B.I. obtained a wiretap on Cohen's home, a thirty-five-thousand-square-foot mansion in Greenwich, but the tap yielded no incriminating evidence. According to a person involved in the investigation, Cohen spent most of the month that it was operational in a house that he owned in the Hamptons. For a time, the agency wanted to place an informant in Cohen's company, and groomed a stock trader who had once worked at S.A.C. to seek employment there again. But Cohen rejected the overture, explaining, in a 2011 deposition, that "rumors from people on the street" indicated that the trader was wearing a wire.

In most white-collar cases, the authorities subpoena reams of internal communications, but this approach had limited utility with S.A.C., whose legal department warned employees not to "compose or send any electronic communication, or leave any voice mail message, if you wouldn't want it . . . read by regulators." On one occasion in July, 2009, a new portfolio manager sent Cohen an instant message saying that he was going to short Nokia on the basis of "recent research." He apologized for this oblique rationale but explained that he had just gone through S.A.C.'s compliance training—"so I won't be saying much." Any time a written exchange approached potentially incriminating territory, Cohen insisted on oral communication. "I am getting coffee on tues afternoon with the guy who runs north American generics business," a colleague once informed him. Cohen's reply: "Let's talk later."

Even when there appeared to be ironclad evidence that Cohen had received and acted upon inside information, his lawyers went to impressive rhetorical lengths to challenge it. One day in 2008, Jon Horvath, the analyst, sent an e-mail to two colleagues about an upcoming earnings report from Dell. His source, he wrote, was "a 2nd hand read from someone at the company." One of the colleagues forwarded the e-mail to Cohen's personal research trader, who forwarded it to Cohen—and then telephoned him. Two minutes after the call, Cohen began liquidating his position in Dell, which was worth ten million dollars. Yet when this trade became a focus in the Steinberg trial Cohen's lawyers argued that Cohen's decision to sell Dell was independent: although the "2nd hand read" e-mail was sent to his inbox, Cohen "likely never read" it. He received a thousand e-mails each day, the lawyers elaborated; he sat at a desk with seven monitors, and it was on the far left monitor that his Outlook inbox appeared. Moreover, the Outlook window was behind two other programs, and the window was minimized, allowing Cohen to see only five e-mails at a time: "Cohen would have had to turn to the far left of his seven screens, minimize one or two computer programs, scroll down his e-mails, double-click into the 'second-hand read' e-mail to open it, read down three chains of forwards, and digest the information." Steinberg was ultimately charged with insider trading in the Dell case and convicted; he was sentenced to three and a half years in prison, but, having posted a three-million-dollar bond, he remains free while pursuing an appeal*. Cohen was never charged. In theory, Steinberg could have testified against his boss, but he might not have been able to produce any additional evidence that Cohen had knowingly traded on inside information. Moreover, Steinberg was an old friend of Cohen's who had worked with him for more than a decade, and was therefore unlikely to betray him.

"I'll find you some candy, but first tell me how you got past the Secret Service."Buy or license »

Martoma had no such loyalty to Cohen. After receiving his enormous bonus in 2008, he lost money in 2009. In 2010—down and out—he was fired. In an e-mail, a former colleague disparaged him as a "one trick pony with Elan." Martoma and his family moved to Boca Raton, where he and Rosemary bought a large house in a waterfront community, for $1.9 million. Neither of them had a job, and they focussed instead on their kids (they had a third child in 2009) and on charity, establishing the Mathew and Rosemary Martoma Foundation and giving it an endowment of a million dollars. Martoma's best friend from Duke, Tariq Haddad, who is now a cardiologist in Virginia, told me that Mathew has always been passionate about philanthropy. "He's given ten per cent of his life savings away," he said. "Over a million dollars, he's donated."

On the evening of November 8, 2011, the Martomas returned home from running errands to discover two F.B.I. agents in their front yard. One of them, B. J. Kang, had been a key figure in the investigation of Steven Cohen. Kang has a buzz cut and a brusque demeanor, and he is known for carrying his service weapon—and several magazines of extra ammunition—with a regularity that may not be entirely necessary for an agent on the hedge-fund beat.

"Get inside the house," he told Rosemary. "This has nothing to do with you."

"I'm staying right here," she replied. "Whatever you have to say to Mathew you can say to me."

Kang turned to Martoma. "Do you want to tell her or should I?"

Martoma looked unsteady. Then he said, "You can go ahead and tell her if you like."

Rosemary was confused and terrified. She had no idea what this was about.

According to Rosemary, Kang then said, "We know what you did at Harvard." Martoma fainted.

When Martoma was accepted at Harvard Law School, his father was so happy that he insisted on driving his son in a U-Haul all the way from Florida to Massachusetts. Martoma, who at the time was still using his birth name, did well in his first year. He was an editor of the Journal of Law & Technology, and he co-founded the Society on Law and Ethics. In the fall of his second term, he sent applications for judicial clerkships to twenty-three judges. But when a clerk for one of the judges scrutinized Martoma's transcript, something looked off, and the clerk got in touch with the registrar at Harvard. On February 2, 1999, the registrar confronted Martoma. His transcript had apparently been doctored: two B's and a B-plus had all been changed to A's. (A remaining B-plus, an A, and an A-minus were left unchanged.) Martoma initially insisted that "it was all a joke." But the school referred the matter to Harvard's Administrative Board, which recommended expulsion.

He fought the decision vociferously, hiring a lawyer and taking two polygraph examinations. There had been a misunderstanding, Martoma explained: he had altered his transcript not for the judges but for his parents. He brought the faked transcript home over winter break, and they were ecstatic. (The panel evaluating his case noted that Martoma was "under extreme parental pressure to excel.") But, after showing his parents the transcript, Martoma continued, he had to leave town abruptly, so he asked one of his younger brothers to compile the clerkship applications that he had left out in his bedroom. Unwittingly, the brother picked up a copy of the forged transcript, and included it in the mailing for the judges. Martoma had discovered the mistake before being confronted by the registrar, he insisted, and had sent e-mails to the secretaries of two professors from whom he had sought recommendations, asking them not to send the letters, "as I am no longer looking for a clerkship."

The Administrative Board remained dubious, because the secretaries did not receive the e-mails until the night of February 2nd—hours after Martoma had been questioned by the registrar. The e-mails were time-stamped February 1st, and Martoma maintained that there had been some sort of server delay, because he had definitely sent them the previous day. His mother, father, and brother all testified before the board and backed his account. Martoma even turned over his laptop to a company called Computer Data Forensics, which produced a technical report for the Administrative Board analyzing the metadata of the e-mails in which he asked to withdraw the recommendations. The firm found that the e-mails had indeed been sent on February 1st. Nevertheless, Harvard finalized the expulsion.

While contesting his dismissal, Martoma had moved to an apartment complex in Framingham, Massachusetts, where he became friends with a young M.I.T. graduate named Stephen Chan. The two began eating dinner together and training in martial arts at a local gym. Eventually, they started a business. Martoma's parents took out a second mortgage to assist the enterprise, and Martoma and Chan hired several employees. Martoma told the employees that he was a Harvard-trained lawyer. The name of the company was Computer Data Forensics. Martoma had supplied Harvard with a forensic report issued by his own company.

The partnership between Martoma and Chan ended, acrimoniously, not long afterward, with Martoma taking out a restraining order against Chan, and Martoma's parents were forced to mediate with disgruntled employees (who had not been paid). Bobby Martoma, incensed with his son, called him "a complete liability."

Later that year, Martoma applied to business school at Stanford. Soon after being accepted, he stopped calling himself Ajai Mathew Thomas and legally adopted his current name. Stanford surely would not have accepted him had it known of his expulsion from Harvard, but because Stanford will not comment on the case it is impossible to know whether Martoma mischaracterized his year in Cambridge or left it out of his academic history altogether. To Ronald Green, his former supervisor at the N.I.H., he explained his departure from Harvard by pointing to the entrepreneurial opportunities available at that time. "The way I understood it, he dropped out to start a business, and it was booming," Green told me.

When I asked Rosemary Martoma when she learned about the expulsion, she said that Mathew had confided it to her early in their relationship. "I'm a full-disclosure person," she explained. But the incident was a source of humiliation for Martoma and for his family, and it became a closely held secret. Even his best friend, Tariq Haddad, always believed that Martoma dropped out of law school; he learned the truth only recently, after Martoma was indicted. Martoma always feared exposure of the Harvard incident, Rosemary said: "It was like a dagger that had been hanging over his head." (S.A.C. performed background checks on prospective employees, but it is not known whether the firm detected this blemish in Martoma's record. Of course, S.A.C. could have learned of it and hired him anyway; forging a law-school transcript and mailing it to twenty-three federal judges demonstrates impressive comfort with risk.)

When Martoma regained consciousness, Agent Kang told him that the F.B.I. knew about "the trade in 2008." Both Rosemary and Mathew immediately understood what he meant. The other agent, Matt Callahan, hung back, but Kang was aggressive. "Your whole life is going to be turned upside down," he said. "You're going to lose all your friends, and your children are going to grow up hating you, because you're going to live your years in a jail cell." According to Rosemary, Kang said that the government would "crush" Martoma unless he coöperated. "We want Steve Cohen," Kang said.

Martoma was not an ideal star witness: if Cohen's lawyers could find a path of escape through a minimized Outlook window, imagine what they might do to Martoma's credibility on the stand by bringing up his Harvard career. Then again, criminal kingpins are often convicted on the testimony of morally dubious underlings. The key witness who put away John Gotti was Sammy (the Bull) Gravano, who had confessed to nineteen murders. A rap sheet was practically a prerequisite to testify against Whitey Bulger. And Martoma clearly possessed a dogged instinct for self-preservation. His parents still called him by his birth name, Ajai, which in Hindi means "Undefeatable." But then something surprising happened. Martoma refused to coöperate.

Agent Kang had already paid a visit to Sid Gilman. At an initial meeting at the university, and in several subsequent conversations, investigators asked Gilman if he had supplied confidential information about bapineuzumab to Martoma. Gilman repeatedly lied to them. "I was intensely ashamed," he explained later. "I had betrayed my colleagues, myself, my university." Kang told Gilman that he was a minor player in this saga—a "grain of sand"—and that the person the authorities were really after was Steven Cohen. Eventually, Gilman agreed to tell the government everything, in exchange for a promise not to prosecute him.

"I'm not actually doing anything—this is just punishment for not flossing."Buy or license »

Would Martoma flip next? In criminal cases where coöperation is a possibility, a defendant's attorney goes to prosecutors with a "proffer," explaining what the client might offer in exchange for lenient treatment. But, despite warnings from Agent Kang that if Martoma went to trial the F.B.I. would "ruin his life," Martoma's attorneys never broached the notion of a plea deal. Here was a hedge funder who might finally deliver Cohen and, because of his enormous profits from the bapi trade, would face extensive jail time if he didn't. Yet Martoma was intransigent. Eventually, a team of F.B.I. agents returned to Boca Raton and, in front of the children, marched him out of his house in handcuffs.

In financial and law-enforcement circles, many wondered why Martoma accepted the role of fall guy. One explanation suggested to me by numerous people was that a numbered account had been set up for the Martomas in some tropical banking haven. But this scenario struck me as unlikely. Suppose that Cohen did seek to witness-tamper in this way. Even if he did so in the billionaire's fashion—through multiple intermediary layers of deniability—wouldn't he be handing Martoma the ammunition for a lifetime of blackmail? If someone promised Martoma ten million dollars not to testify about securities fraud, what would stop him from renegotiating on the spot, by demanding twenty million not to testify about obstruction of justice on top of securities fraud? Even so, Cohen's money was an inescapable factor in the case. After working briefly with a criminal-defense attorney named Charles Stillman, Martoma chose to retain Goodwin Procter, a major law firm with very high fees. But Martoma was not paying for his lawyers: S.A.C. was. So the attorneys advising Martoma on whether he should risk a jail sentence or testify against Cohen were sending their bills to Cohen's company.

After the U.S. Attorney's office announced an indictment of Martoma, Cohen convened a company-wide meeting at S.A.C. and said that he was furious about the behavior of "a handful of employees." Martoma was the eighth person who had worked for Cohen to be charged with insider trading—the largest number of individuals from any U.S. financial institution to be criminally charged in recent years. Even if Martoma didn't turn on Cohen, the company was clearly in trouble. In March, 2013, lawyers for S.A.C. agreed to pay six hundred and sixteen million dollars to the S.E.C. in order to settle civil insider-trading charges. Several months later, the S.E.C. launched a separate case against Cohen personally, charging him with "failure to supervise" subordinates, and alleging that he received "highly suspicious information that should have caused any reasonable hedge-fund manager . . . to take prompt action." (That case is still pending.) Last summer, the Department of Justice announced a criminal indictment of S.A.C.—though not of Cohen directly—alleging that the company had become a "magnet for market cheaters," and that Cohen had presided over insider trading "on a scale without known precedent in the hedge-fund industry." Not long afterward, the firm pleaded guilty to the criminal charges, agreeing to pay a historic $1.8-billion fine.

Cohen had always greeted allegations of impropriety at S.A.C. with bored disdain. When a lawyer asked him in a deposition in 2011 about Rule 10b-5—the federal regulation against insider trading—Cohen claimed not to know what it said. The lawyer pointed out that Cohen's own compliance manual at S.A.C. spelled out the rule. Cohen responded that he didn't know what the compliance manual said, either.

The lawyer was incredulous: "You don't know, sitting here today as the head of the firm, what your compliance manual says?"

"That's right," Cohen said. "I've read it. But if you're asking me what it says today, I don't remember."

Not long after S.A.C. announced its settlement with the S.E.C., the news broke that Cohen had bought Picasso's "La Rêve," for a hundred and fifty-five million dollars—the second-highest price in history for a painting. While he was at it, he bought a new house in East Hampton, a waterfront property worth sixty million.

The trial of Mathew Martoma began in January, 2014, and lasted a month. Blizzards had deposited huge snowbanks around the federal courthouse in downtown Manhattan, and every morning Mathew and Rosemary Martoma arrived in a chauffeured car and clambered, with their lawyers, over cordons of dirty snow. They had brought the children with them to New York and were staying at a midtown hotel. Mathew's mother and father had come from Florida for the trial, and they sat in the front row, bundled in winter coats and scarves and looking solemn. Rosemary's parents sat beside them.

"Ladies and gentlemen, the case is not about scientific testing and it is not about trading," the government's lead lawyer, Arlo Devlin-Brown, told the jury. "The case is about cheating." Martoma, wearing a dark suit, watched impassively as a parade of former S.A.C. colleagues testified; Rosemary smiled when she agreed with a witness and flared her nostrils when she didn't. She wore eye-catching clothes, becoming an attraction for the tabloid photographers who clustered at the base of the courthouse steps. An article in Bloomberg Businessweek remarked on her poise in the courtroom and the defiant smile she maintained when she and Mathew walked in and out of the courthouse, hand in hand, "as if she's walking a red carpet."

The government presented dozens of e-mails that Martoma sent to Cohen and other colleagues, and called on Joel Ross, the doctor from New Jersey, to recount how he shared inside information with Martoma. But the heart of the case was the testimony of Sid Gilman, who in the second week made his way, slowly, to the stand. Gilman had resigned from the University of Michigan, and administrators had scrubbed all traces of him from the institution: the wing of the hospital, the lecture series, the university's Web site. His federal grant support disappeared, his former colleagues wanted nothing to do with him, and he was banned from the campus. He had lately been advising patients at a free clinic.

"I had given a great deal to that university, and I am suddenly ending my career in disgrace," he said. Gilman still dressed elegantly, his shirt and tie cinched tightly around his neck, accentuating his large, round head. But he was eighty-one and visibly frail. During five days of testimony, he looked marooned in the witness box, a shipwrecked man.

Several lawyers suggested to me that Martoma's attorneys should never have let the case go to trial, because the evidence against him was so conclusive that he didn't stand a chance. But his defense team, a pair of lean and intense litigators, Richard Strassberg and Roberto Braceras, relentlessly attacked Gilman's credibility as a witness. He had apparently told prosecutors that he had e-mailed a copy of the PowerPoint presentation to Martoma. But, the defense team pointed out, the prosecutors had failed to find any trace of that e-mail. At times, Gilman just seemed confused. Asked about the population of Ann Arbor, he said that it was fifteen hundred. (The population exceeds a hundred thousand.) Some of Gilman's colleagues speculated that, after decades of studying neurodegenerative disorders, he was now succumbing to cognitive decline himself. Of course, this may be a generous interpretation of actions that many who knew him found inexplicable. "Nobody could believe it," Anne Young told me. "To jeopardize his career for a hundred thousand dollars or so is insane."

Martoma, Strassberg told the jury, was "the quintessential American success story," whereas Gilman was a confused old man who had been coached by the government. When speaking to Gilman, Strassberg combined the elevated volume you might use to address a deaf person with the patronizing tone you might employ with a seven-year-old. If this was a strategy, it backfired. Every time Strassberg asked whether Gilman hadn't heard or understood something he said, Gilman bristled. "You're slurring your words," he snapped at one point.

November 8, 2004"I'm self-published."Buy or license »

Martoma's lawyers suggested that the information Gilman shared with Martoma was already publicly available. "There is nothing nefarious or improper about trying to get edge," Braceras argued. "That was the job." The lawyers challenged the government's narrative of a special relationship between Gilman and Martoma, observing that Gilman had consultations with scores of other investors. But former colleagues of Gilman told me that the government's story was plausible. "Sid was a mentor to so many people, and enjoyed that role, and was good at it," Tim Greenamyre said. "I could certainly see how, if someone was cunning and perceptive, they could pick up on that, and take advantage of it." As Gilman answered questions on the stand, day after day, he looked, above all, lonely. His son Todd lived nearby, in New Haven, but they had hardly spoken for years. On his final day of testimony, Gilman was asked what set Martoma apart from the other investors he had dealt with. "He was personable," Gilman replied. After a pause, he said, "And he, unfortunately, reminded me of my first son. In his inquisitiveness. His brightness. And, sadly, my first son was very bright also, and committed suicide."

One matter that was not illuminated at trial was the substance of the twenty-minute phone call that Martoma had with Cohen on the Sunday morning after his trip to Michigan. If Martoma took the stand, prosecutors would attack his credibility by introducing evidence of his expulsion from Harvard Law School, so he elected not to testify in his own defense. Steven Cohen wasn't called to testify, either. In 2012, he had been asked about the phone call during his deposition with the S.E.C. He said only that Martoma was "getting uncomfortable with the Elan position." Asked whether he inquired why Martoma had grown uncomfortable, Cohen said that he remembered having done so—but that he could not recall Martoma's answer.

A second theory about why Martoma didn't flip on Cohen was that any conversation the two of them had that day would have been deliberately opaque. Cohen would never be so foolish as to sit and listen while a subordinate laid out the full provenance of an illegal tip. At some firms, Judge Holwell told me, there is an unwritten "don't ask, don't tell" policy, where the fact that a piece of information came from an insider would be conveyed not in so many words but with a facial expression, a tone of voice, or coded language (say, a conviction level of nine). The sociologist Diego Gambetta, in his book "Codes of the Underworld," explains that people engaged in criminal conduct often evolve an elaborate semiotics to communicate with one another, because they cannot speak openly about their plans. One federal official who has investigated S.A.C. told me, "In the Mob, sometimes it's just an expression. One expression means 'Kill him.' Another expression means 'Don't kill him.' How do you bring that to a jury?"

After deliberating for three days, the jury convicted Martoma of two counts of securities fraud and one count of conspiracy. Rosemary wept as the verdict was read. The guidelines for his sentence would be based not just on the $9.3-million bonus he had received from S.A.C. in 2008 but also on the two-hundred-and-seventy-five-million-dollar profit that S.A.C. had made on the bapi trades. Yet Cohen was not charged with those trades, or even named as an unindicted co-conspirator. The judge, Paul Gardephe, went so far as to ask the attorneys to avoid discussing Cohen altogether, because he had not been charged with any crime. "General questions about how Steve Cohen conducted his trading, I think, are very dangerous," he told them. "They represent a risk of opening the door to a broader examination of how Steve Cohen did business. . . . And I think we all agree that that is not a path we want to go down." (In a subsequent ruling, Gardephe left little doubt about his own views, concluding that Cohen's trades in July, 2008, "were based on inside information that Martoma had supplied.")

During the trial, Cohen was photographed at a Knicks game, sitting courtside with the art dealer Larry Gagosian. According to a recent article in New York, Cohen told his children that he felt betrayed by his subordinates. "People in the company have done things that are wrong, and they're going to pay for what they did," he said. "I didn't do anything wrong."

Before Judge Gardephe delivered his sentence, Martoma's family sent him a hundred and forty-three letters from friends and supporters, pleading for leniency. "We pressed him to excel until he maxed out," Bobby Martoma wrote. "As a father, I wonder . . . whether I was wrong to dream as I did."

On September 8th, Gardephe sentenced Martoma to nine years in federal prison. Delivering the sentence, he invoked the deception at Harvard and suggested that there was a "common thread" between that transgression and this case: an "unwillingness to accept anything other than the top grade, the best school, the highest bonus—and the willingness to do anything to achieve that result."

A few days after the sentencing, I took an elevator to the twenty-sixth floor of a skyscraper on Forty-second Street to meet with Rosemary and Mathew Martoma. I walked into a glass-walled conference room that seemed to hover over midtown. Martoma was there, wearing a neat V-neck sweater. He shook my hand, smiled warmly, and thanked me for coming. But he did not want to talk. Rosemary explained that she would speak on his behalf. She was wearing a cream-colored blouse, tan slacks, and a tiny gold crucifix around her neck, and after Mathew left the room we talked for nearly four hours.

Things were looking dire for the Martomas. The government would likely take possession of their house in Boca Raton, and Judge Gardephe had ordered them to forfeit the millions of dollars that they have spread across several bank accounts, which would clear out their savings while still falling well short of the nine million dollars that they have been ordered to pay the U.S. government. When I consulted the tax returns of the Mathew and Rosemary Martoma Foundation, I discovered that the couple had not, in fact, given a million dollars to charity. Instead, after parking that sum in their tax-exempt nonprofit, they had given away smaller amounts to various charities. In 2011, they gave away only three thousand dollars; this included a check to the Florida chapter of the American Alzheimer's Association, in the amount of two hundred and ten dollars. All the remaining money in the foundation will now go to the government.

When I asked Rosemary why Mathew didn't flip on Cohen, her answer did not match any of the prevailing theories. "He's innocent," she said. Martoma could not plead guilty to a crime he did not commit. Outside the courthouse, after the sentencing, Bobby Martoma had told me much the same thing, invoking the Ten Commandments and bellowing, "Thou shalt not bear false witness!"

The government's case was a fiction, Rosemary assured me. "At S.A.C., there's an expectation that you're using the resources to formulate a hypothesis, and that's what he did," she said. But Gilman had admitted to violating his own confidentiality agreement, I pointed out. He may have had "little mini brain infarcts, where he was slipping on things he shouldn't have," Rosemary said. But these were "irrelevant to Mathew's trading."

Gilman had lost everything. Why would he lie on the stand about having committed these crimes? Because, Rosemary explained, when he was initially interviewed by F.B.I. agents, he lied to them, and at that point they had him on obstruction of justice. So the prosecutors could make him say anything they wanted. "His story was coerced," she said.

She told me about her grandfather, a lawyer in India who had worked alongside Mahatma Gandhi in the struggle for independence. British authorities threw her grandfather into prison, where he contracted cholera and other ailments, from which he never fully recovered. Rosemary noted the "parallels" between her grandfather's martyrdom and her husband's. Rosemary's mother, in a letter to Judge Gardephe, elaborated: "Mathew has given Rosemary courage by reminding her of her grandfather's suffering for a noble principle, and that he too is standing for a noble principle, that is sticking to the truth."

While Rosemary and I spoke, Mathew retreated to an inner room in the law offices where we were meeting. Periodically, Rosemary left me alone in the conference room and went to confer with him. It was a discomfiting interview scenario, with Martoma lurking in the wings like Polonius. People who maintain their innocence after a criminal conviction are often desperate to get their stories out, and, each time Rosemary disappeared to debrief Mathew on our conversation, I half-expected him to walk back in with her, and tell me that he had been railroaded by the feds. But he never came. They had arranged for a single chicken-salad sandwich to be delivered, and it sat on a sideboard, wrapped in plastic. Eventually, alone in the conference room, I ate it.

April 25, 2011"He died after a long battle with me."Buy or license »

When Rosemary returned, she spoke at length about the duplicity of Sid Gilman. "He's a strange man, and he compromised his values to save himself," she said. The notion that Gilman and Martoma had a special relationship was "farfetched"—a fabrication of the prosecutors that Gilman had parroted. "There is no relationship outside of a cordial consulting relationship," she said, mocking the notion that Gilman was genuinely moved by Martoma having arranged lunch at their initial meeting, in New York. She looked at me pointedly and said, "I mean, were you touched when we served a sandwich to you?"

Throughout our conversations, Rosemary was quick, animated, and intelligent. But her account stood at odds with what I had witnessed during the monthlong trial and had encountered in my reporting. She stressed that, when Mathew visited Michigan that summer weekend before the 2008 Alzheimer's conference, it was indeed because a relative had died.

"Did he see Gilman while he was there?" I asked.

"I don't think he has a specific memory of it," she replied.

Early in our conversation, I had asked if Martoma felt vindicated by his acceptance at Harvard Law School, having been denied admission at Harvard College. After one of her visits with him, Rosemary returned to the room and said that she needed to correct one point: "Mathew did get into college at Harvard."

"As an undergrad?" I asked.

"Yeah," she said. "He was admitted and chose to go to Duke instead."

This struck me as hard to believe. I asked why Martoma, the very opposite of a rebellious kid, might defy his father's deepest wish. She responded, vaguely, that Duke "was Southern" and "felt a little bit more comfortable to him."

I wondered if, on this and other points, Rosemary was simply lying to me. But as our conversation progressed it became clear that she ardently believed in her husband. She reminisced about her medical residency in Boston, when she would be on call overnight and Mathew would sleep in the hospital with her so that she wouldn't be alone. She pointed to the many letters written to Gardephe as evidence of the degree to which Mathew remained a beloved friend and a role model for his extended family. "Every Indian parent I've known, they take the weight of their children on their shoulders," she said. "When you look into the eyes of all the four parents that are left behind, every single heart is broken." Mathew's mother had told him recently that she wished she could serve his prison sentence for him, Rosemary said. She added, "I've said that to him, too."

Within this close-knit family, it seemed crucial to maintain that Martoma was going to prison for a crime that he did not commit, and it occurred to me that there might be one final explanation for his unwillingness to accuse Cohen of criminality. In order to implicate Cohen in a conspiracy, Martoma would have had to plead guilty and admit to being part of that conspiracy himself. Could it be that Martoma was prepared to leave his wife and family and spend the better part of a decade in prison for the sake of preserving their illusion that he was an honorable man? I thought of Gilman on the stand, abandoned by his friends and colleagues, while the first few pews in the courtroom were filled by Martoma's extended family—by people who believed in him.

Martoma is scheduled to begin his sentence, at a federal prison in Miami, next month. When I asked how Rosemary and the children would manage, she said, "I'm not sure." The children are nine, seven, and five. "They understand Daddy's going to jail," she said. "I mean, as an adult, I'm having a hard time understanding it." Neither side of the family has any savings to give them, she said, adding, "There is not, and never was, and never will be, any discussion of Steve Cohen taking care of us."

In April, S.A.C. ceased to exist, and Cohen's company was rechristened Point72 Asset Management. Under an agreement with the government, it will be limited to investing Cohen's personal fortune of roughly nine billion dollars. Cohen has announced that he will institute more robust compliance measures to prevent insider trading, and he has hired the Silicon Valley security company Palantir Technologies to monitor his traders. He has also reportedly banned certain kinds of instant messaging at the firm. When I asked Preet Bharara about the ultimate failure of his multi-year effort to catch Cohen, he responded, through a spokesman, that his office brings charges against "those for whom there is sufficient proof." The S.E.C.'s lawsuit charging Cohen with "failure to supervise" will progress, as will a class-action lawsuit brought on behalf of hundreds of Elan and Wyeth shareholders who lost money when Cohen shorted the stocks. Insider trading may not be a victimless crime after all—at any rate, not when the victims sue you. In August, the litigants amended their suit to include a racketeering provision, which alleges that Cohen, like a Mob boss, sat on top of a criminal enterprise.

After Martoma's conviction, Stanford Business School rescinded its original offer of admission, effectively stripping him of his degree. "What to make of the early interest in ethics?" his professor from Duke, Bruce Payne, asked. "A hugely ambitious guy wanting to know the exact contours of the boundaries that might limit him? Or an anchor to the windward for self-protection by someone already willing to break the rules to his own advantage? If it was the latter, I was conned, and conned quite effectively."

When I asked Rosemary about the future, she cried. "I don't have the answers, but you know it is my goal to find them," she said. "And I do pray that America will give us a chance to survive. And to thrive." 

*An earlier version of this article incorrectly stated that Michael Steinberg was serving a three-and-a-half-year sentence.