The rapping of knuckles on wood brought Kevin Brown to his front door on the morning of January 9, 2014. The uniformed detectives standing on his doorstep introduced themselves as Michael Lambert and Lori Adams of the San Diego Police Department. Lambert looked familiar to Brown, who had worked for the department as a crime-lab analyst from 1982 until his retirement, in 2002. The detectives said they wanted to discuss an old case with Brown, and he invited them in.

Stepping into the living room, the two detectives sat down in chairs and faced Brown, who perched himself in the middle of a burgundy leather couch. At 61, he was tall, with wire-rimmed glasses and a retreating halo of grayish curly hair. They were briefly joined by Brown’s wife, Rebecca, a teacher at a Catholic high school. The detectives’ visit constituted more excitement than the Browns typically saw of late: Many of the couple’s activities centered on church or Rebecca’s work. Rebecca, who was vivacious and talkative, directed school plays; the more reserved Kevin liked to fish and garden. The two took ballroom-dance classes, traveled abroad, and hiked together. “We were just a standard, boring couple living in the suburbs, tending to our stray cats, roses, and each other,” Rebecca told me.

Casual and friendly, Lambert and Adams chatted with Brown for an hour about past cases and departmental gossip. Then Lambert asked whether Brown recognized the name Ronald Tatro and pulled out a photo showing a man with stony features and dead eyes, his dark hair slicked back.

No, Brown told the detectives, he didn’t know the man.

Claire Hough was 14 years old when her body was found at Torrey Pines beach, in San Diego, in August 1984. She had been strangled, sexually assaulted, and mutilated. (Courtesy of the Hough Family)

Lambert produced a second photo, which showed a pretty 14-year-old girl with feathered brown hair and heavy makeup, her eyes large and her lips parted. “Oh, I remember her,” Brown said. The girl’s name was Claire Hough, and she had been sexually assaulted, mutilated, and strangled to death on a San Diego beach in 1984. The notorious crime had never been solved. But in 2012, the SDPD crime lab had tested evidence recovered from the crime scene and discovered that DNA from bloodstains on Hough’s jeans matched that of the man in the first photo, Tatro, who was a convicted rapist.

The DNA analysis also indicated that a second suspect was involved. Lambert cleared his throat and looked at the retired criminalist.

We found your DNA among the evidence too, the detective told him.

Brown’s DNA had been identified on a vaginal swab, the police alleged in search-warrant affidavits that describe the SDPD investigation, including the January 9 visit to his house. This revelation was the latest development in a decades-long investigation involving hundreds of law-enforcement officers, from local cops to FBI agents. The murder had generated a ripple effect of incriminations, which in turn seems to have triggered a bizarre series of additional deaths. More recently, the case has resulted in an unfolding legal battle that pits the hallowed authority of DNA evidence against the underappreciated fact that it can sometimes be altogether too fallible.

Before all this, though, the story began with a summer vacation. In August 1984, Claire Hough and her best friend, Kim Jamer, traveled from Rhode Island to San Diego for a two-week visit with Hough’s grandparents. Hough, 14, was in a hard-rock phase, idolizing bands such as Kiss and AC/DC. But she wasn’t some aggressive rebel. The product of a free-spirited, intellectual family, Hough wrote poetry and stuck up for classmates who were bullied. She was warm, creative, and gregarious, Jamer says. “In a field of sunflowers, she was an orchid.”

Hough loved to sit by the ocean—“it was where she did her thinking,” her father told me—and she and Jamer spent most of their California days at the beach. Hough tried to convince Jamer that they should sneak out to the beach at night, too, but Jamer, who had been raised in a stricter household, wasn’t as impulsive as Hough. “She had no leash, and I was on a choke collar,” Jamer told me. But finally, when the trip was almost over, Jamer gave in. After going to their bedroom, the two girls snuck out of the house through a sliding glass door and walked 15 minutes to Torrey Pines State Beach.

Revelers clustered around campfires along the expanse of sand, the pinpricks of light quickly yielding to darkness. Hough headed for a secluded spot underneath a bridge that ran alongside the north end of the beach. She lit up a Marlboro and pushed play on her boom box. The music, wind, and crashing surf roared in the confined space, and Jamer soon felt panicky. “You couldn’t hear people until they were right up on you,” she says. When Jamer started to cry, Hough was sympathetic. “Claire took me home, wrapped me in a blanket, and promised she would never sneak out again,” Jamer says.

Two nights later—on August 23, 1984, after Jamer had returned to Rhode Island—Hough broke that promise. She snuck out again, this time alone. A beachcomber named Wallace Wheeler spotted her body at five the next morning. “I was in the habit of shining my flashlight over the beach to see if cans or anything else was in range of the light,” he would later recount. “It appeared someone was sleeping so I quickly turned my light off and moved toward her. I had not taken many steps when I realized I had seen her exposed left hip and back, and people, with rare exceptions, do not sleep on the beach that way … I was sure from the color of her flesh that she was dead.”

Hough lay on her right side next to her boom box, her right arm bent and her knees drawn up toward her torso. The white towel underneath her was stained crimson with blood, and a coroner determined that she had been strangled to death. Her left breast had been cut off, and her mouth was filled with sand. Her panties had been rolled down below her buttocks and her genitalia were lacerated, leading police to wonder whether she had been raped. At the time of the autopsy, however, the coroner’s report noted that no sperm was found in her body. When Jamer heard the terrifying news, she didn’t need to ask about the location. “I knew that her body would be found underneath that bridge,” she says.

A young and attractive victim, a gruesome death—the murder made headlines for all the typical reasons. But there was an additional element that propelled the Hough murder to lasting notoriety. In 1978, another girl, a 15-year-old named Barbara Nantais, had been murdered in a nearly identical fashion on the beach at Torrey Pines: strangled at night, breast mutilated, mouth filled with sand. The two bodies, though separated by six years, were found only a short distance apart.

After the SDPD asked the FBI to examine the two murders, the federal profilers concluded, “These cases are suspected to be relate[d] based on victim selection, location, and [c]ase characteristics,” according to a heavily redacted FBI document. Detectives, in short, should hunt for a possible serial killer.

Kevin Brown, the accused, worked as a crime-lab analyst for the San Diego Police Department from 1982 until his retirement, in 2002. (Howard Lipin / The San Diego Union Tribune / Zuma Wire)

For a promising suspect, the police didn’t have to look any further than Wheeler, the beachcomber who had found Hough’s body. He had served in the Army during World War II and subsequently worked as an insurance salesman. But by 1984, possibly suffering from a mental illness, Wheeler, 61, was spending much of his time searching Torrey Pines for discarded cans. When Hough’s parents, Samuel and Penelope, visited the crime scene, Wheeler immediately put them on edge. “He approached us on the beach and introduced himself, and said he was a psychic,” Samuel told me.

Wheeler began sending long letters to the Houghs once they returned to Rhode Island. In neat, dense cursive, he explained that he had foreseen finding a body on the beach months beforehand. Afterward, he claimed, Claire’s spirit came to visit him and the two began an intimate, clairvoyant relationship. “She let me see her smiling face and really her eyes were radiant,” Wheeler wrote. The dead girl was trying to tell him the identity of her killer. The assailant, Wheeler told the Houghs, was a man with long hair, a high forehead, and a missing ear who had snuck up on Claire beneath the bridge. “I believe he immediately attacked and strangled her,” Wheeler wrote. The beachcomber even claimed to have psychically witnessed what happened next: “He was crouched over her shoulders and I just don’t have a clear picture of what he was doing but would think it was when he mutilated her.”

SDPD detectives shared the Houghs’ mounting suspicion of Wheeler, and encouraged Samuel and Penelope to keep the correspondence going. “I was trying to play a game to get him to confess,” Samuel told me. He said the police even hid a microphone in a memorial to Hough that had been set up on the beach, in the hope that Wheeler would visit it and say something incriminating within earshot.

But for all his strange ramblings, Wheeler never confessed to anything. The police eventually told the Houghs that Wheeler had been definitively ruled out as a suspect, possibly with the help of DNA testing, says Bonnie Hough, Claire’s aunt. (The SDPD declined to comment on this and most other specific aspects of the Hough and Nantais investigations.) Wheeler’s mental health, meanwhile, continued to deteriorate. A couple of years after he found Claire, he attempted to kill himself, by driving off a cliff. And in 1988, he succeeded, jumping to his death from the 13th floor of the apartment building where he lived.

The police, who had pursued the wrong suspect, were now stuck. As Detective Lambert would later summarize in an affidavit, “No eyewitnesses were identified, few leads were developed, and the case went cold.”

In the decades after Wheeler’s death, SDPD detectives periodically reexamined the Hough and Nantais files. According to Bonnie Hough, they tested the evidence for traces of DNA, but didn’t get any hits. As Lieutenant Paul Rorrison, the current head of the department’s cold-case unit, explained to me, “These cases, they don’t go away.” But investigators didn’t make significant headway until 2012, when the SDPD ran another DNA test on the Hough evidence, using new analytical technology that was far more sensitive. A blood sample found on Hough’s clothing yielded a DNA match for Ronald Tatro. And although no sperm had been detected on the vaginal swabs collected and tested back in 1984, the new analysis turned up apparent traces of DNA from Kevin Brown’s sperm.

Ronald Tatro, the convicted rapist whose DNA was detected in bloodstains found on Hough’s clothing (San Diego Police Department)

Of the two suspects, Tatro was by far the more obvious, having been previously imprisoned for sexual assault and other violent crimes. He had been convicted in 1975 of first-degree rape in Arkansas, after he kidnapped a woman, stuffed her in the trunk of his car, and drove her out to a remote country road, where he sexually assaulted her while holding a knife in her mouth. Prior to his trial, he told a psychiatrist that he had no control over his urge to assault women and that he’d “had a similar problem for many, many years, dating back to his early childhood.” After being paroled in 1982, Tatro moved to San Diego, where three years later he lured a teenage girl into his van and tried to subdue her with a stun gun. But she escaped, and Tatro was subsequently convicted of attempted rape.

I spoke with Sharon Kaye Villines-Tatro, who had been married to Tatro in the 1970s, and she sounded sad but unsurprised to hear that her former husband had been connected to another violent crime. Tatro used to tie her up and rape her, she said. He also told her of other violent crimes for which he hadn’t been caught. “I have been expecting this to come up, but not in San Diego,” Villines-Tatro said. “Ron told me that he had raped a girl on his way home from Fort Lewis, Washington … He thought he had killed her and left her for dead.” At first she didn’t believe him, but later she contacted law enforcement.

The discovery of Tatro’s DNA was a triumphant, case-solving break for the SDPD detectives, who pointed to the suspect’s violent background, his presence in San Diego in 1984, and the traces of his DNA as evidence of his culpability for Hough’s murder. But the new DNA revelations also caused two major problems for the department. The first was that the genetic evidence required detectives to retreat from the long-held theory that Hough and Nantais had both been murdered by the same person: Tatro had still been in jail for rape in 1978, when Nantais was killed. (A records custodian for the Arkansas Department of Corrections confirmed for me that Tatro had not escaped or been furloughed.) Brown, too, seemed an improbable perpetrator for the Nantais killing: In 1978, he was a college student living in Sacramento, more than 500 miles away. “We have no evidence at this point that says these cases are related at all,” Lieutenant Rorrison told me.

The detectives’ previous theory that the murders were connected made perfect sense until it was contradicted by the DNA findings. “The likelihood of two separate individuals, independent of each other, killing the same type of woman and cutting her breasts off is very low,” says Louis B. Schlesinger, an expert on sex-linked homicides at John Jay College of Criminal Justice, in New York City. Schlesinger speculates that Hough’s killer simply may have heard about the Nantais murder and been inspired to replicate it.

Ray Surette, a criminal-justice professor at the University of Central Florida, agrees that the Hough killing could be a classic example of a copycat crime, a phenomenon he has studied extensively. The other possibility, he says, is that there is no connection and the similarities are merely coincidental. “Sometimes the extraordinary similarity between crimes simply means people came up with unique ways to commit crimes independently of each other,” Surette says. And indeed, even a cursory Google search reveals that homicides in which one or both of a woman’s breasts are mutilated are not as rare as one would dearly hope.

Barbara Nantais, whose 1978 murder at Torrey Pines State Beach was for a time believed to be connected to Hough’s (Courtesy of Lorraine Thall)

The other major complication caused by the results of the 2012 DNA tests was that they pointed toward two possible suspects, not just one. Who was responsible for Hough’s murder, Tatro or Brown? Both were, the detectives decided: The criminal and the crime-lab employee had somehow committed the act together. Despite their efforts, though, police have never been able to find any evidence connecting the two men. From their very first interrogation of Brown, on January 9 of last year, detectives repeatedly tried to get him to admit that he knew Tatro. But Brown always maintained that the two had never met.

Tatro, unfortunately, couldn’t confirm the theorized partnership either. By the time the police got the hit on his DNA, he was dead, having drowned in a boating accident in Tennessee the previous year. Noting that Tatro’s wallet, hat, and glasses had all been carefully placed inside the boat, a report compiled by the Tennessee Wildlife Resources Agency suggests that Tatro didn’t tumble overboard by accident. “It appears Mr. Tatro prepared to go into the water,” the report states. If so, he was, after Wheeler, the second suspect in the case to have committed suicide.

The date of Tatro’s death, moreover, was either a remarkable coincidence or a coded confession. He drowned on August 24, 2011—27 years to the day after Hough’s body was discovered.

With tatro dead, the SDPD focused on Brown. And despite his career as a criminalist and his churchgoing retirement, Brown’s past provided plenty of reasons to suspect he might be capable of committing a sex-related crime.

Brown had arrived in San Diego in 1982 to start his job with the police. He was a bachelor, and the city was a libertine paradise rocking to hits like “Physical,” by Olivia Newton-John, and “Centerfold,” by the J. Geils Band. Women prowled the beach, looking for parties, beer, maybe more, says Tom Wick, who became Brown’s roommate shortly after Hough’s death. “These gals would walk up and down … underage gals, it doesn’t matter. Stuff that we consider taboo now was everywhere.”

Brown pumped iron and studied karate. He was handsome and educated, and he carried a badge. “It should have worked out really good for meeting women, but he was too shy,” says Kevin Sheldon, another former roommate. The outgoing Sheldon played wingman for Brown when they went out. But Brown’s relationships typically crashed on the shoals of his self-doubt. “As you got to know Kevin, you realized that he was just very meek,” Sheldon says.

Brown was nonetheless intensely interested in women. He liked to go to strip clubs. He liked to talk about going to strip clubs. His nickname in the SDPD crime lab was “Kinky,” and he offended some colleagues, especially a fellow criminalist named Annette Peer. She told Detective Lambert that Brown had once shown a sickening porn movie at work. Another time, when Peer and Brown were alone, he read aloud from a report that he said was funny—but that in fact described a brutal sexual assault. Peer says that she never felt comfortable being alone with him again.

Brown was a photographer, a hobby he’d begun as a teenager, and in San Diego he started going to events where amateurs got to practice boudoir photography. The organizer would hire female models, and anywhere from half a dozen to 100 photographers like Brown would show up to take pictures. But Brown began setting up his own shoots, too. According to a police affidavit, he told Lambert and Adams that the women performed sex acts with one another at some of the shoots.

The police portrayal of Brown as a pervert was disturbing enough. But when Lambert and Adams returned to Brown’s house for a second round of questioning on January 10, Brown made a hair-raising admission. He said that he now recalled meeting a woman in the 1980s who was possibly named Claire. She was visiting from out of town with a friend. Brown said he may have even had sex with this person, though he couldn’t be sure.

Brown continued to maintain that he had never killed anyone, and he repeatedly volunteered to take a polygraph test. And so that same day, Lambert and Adams drove him to a nearby police station, where an interrogation specialist wired him up. Asked whether he had ever known Tatro or had committed a crime with him, Brown said no. The lie-detector results were inconclusive. But when he replied no to questions regarding whether he had had sex with or killed Claire Hough, the polygraph readings were clear: Deception indicated.

After moving to San Diego, Brown pursued his interest in photography by attending boudoir shoots with female models. (Courtesy of Kevin Sheldon)

As bad as that looked for Brown, what followed was more damning still. After the lie-detector test, Lambert left the room, and Brown sat alone with Adams. They were silent for a few moments, and then Adams sympathetically told him, “I don’t believe for a second that you thought she was 14. I really don’t.”

“Well I didn’t,” Brown replied, according to an affidavit. “I had no idea.” It sounded like a confession.

What’s more, shortly after the police first showed up to question him, Brown had placed a phone call to his old friend Tom Wick. When the police learned about it, any lingering doubt regarding Brown’s guilt must have vanished. As the same affidavit put it, “Brown told Wick the girl who was murdered was the same girl he had just finished a beach photo shoot with and that she ended up dead on the beach.”

In mounting a defense for the guilty-seeming Kevin Brown, his wife and friends acknowledged that he had been no Boy Scout in the past. Kevin’s passion for strip clubs and kinky photo shoots? Yes, he had those proclivities, but they didn’t prove a thing. As Rebecca put it, “He liked to take pictures of pretty women. So what?”

In any case, his friends told me, the X-rated Kevin cleaned up to PG after he married Rebecca, in 1993. And everyone who knew him in the 1980s agrees that he was almost comically incapable of aggression. Rebecca describes her husband as “just a nice, gentle, dorky guy” and “my teddy bear.” Even people who were otherwise willing to criticize Kevin describe him as fundamentally nonviolent. “If you had told me that Brown had sex with an underage female, that wouldn’t have surprised me,” says Jim Stam, who was a supervising criminalist in the SDPD crime lab from 1982 to 2008 and sometimes served as Brown’s boss. “Killer? Yeah, that surprises the hell out of me.”

Brown’s onetime roommate Sheldon was also his martial-arts instructor, and he recalls that he could never cultivate a fighting instinct in his pupil. “Anytime we started to ramp it up, he would unravel,” he says. Jon Dahlberg, a friend since junior high school, says that Brown, far from being a rapist, bragged about having developed an evidence-collection kit for hospitals to use with rape victims. “Kevin was definitely against the bad guys,” Dahlberg says.

Brown’s supporters concede that “he has some things that need to be explained,” as Wick puts it. There was, for instance, the failed polygraph test. Not long after he took it, Brown went to lunch with another old friend, Ed Pierce, for reassurance. Pierce, who had been trained to administer polygraph tests when he was in the Army, told him that the test didn’t prove anything. “A lie detector is a simple device which can’t tell whether you are lying,” he recalls telling Brown. “It can only tell if you are anxious or not.” This assertion is supported by a 2003 National Research Council report that sharply criticizes the reliability of the tests.

By all accounts, Brown was an exceptionally nervous person. Testifying about crime-lab results in court, he was easily unnerved and would get tongue-tied, his former boss, Stam, recalls. Friends say it is easy to imagine Brown getting bullied by police interrogators to the point of wrongfully incriminating himself, and they feel he never should have spoken to the detectives without a lawyer present. “They had him so rattled, he didn’t even know his middle name,” Sheldon says.

Moreover, there are questions concerning two statements in an affidavit that, in addition to the DNA evidence, appear to be the core of the case against Brown. The first statement is his shocking admission that he had slept with a woman named Claire who was visiting from out of town with a friend. The person he described in the affidavit does not otherwise resemble Hough. Brown told police that she was staying at the Holiday Inn in downtown San Diego and that he had visited her in her hotel room. (Hough, by contrast, was staying with her grandparents.) The affidavit even contains this line: “Detective Adams told him it does not appear to be the same girl.”

Brown said he met this Claire through a friend, whom police later determined was named Mike Newquist. Newquist declined to speak with me for this article, but when I asked Wick whether he remembered anything about someone named Claire from the 1980s, he told me that he and Newquist had discussed this. They both recalled knowing a Claire who was not Claire Hough and who was closer to 30 years old, although they did not remember Brown meeting her. In any case, in the “gotcha” moment after the polygraph test, when Brown confessed to Adams that he had “no idea” Claire was 14, it seems plausible that he was recalling a different Claire who was considerably older.

The second statement in question is Brown’s seeming confession on the phone with Wick that a girl he’d photographed had ended up dead on the beach. That evidence alone might have sunk Brown. But Wick repeatedly insisted to me that Brown had never told him this. How could the police have gotten this vital fact wrong? Wick guesses that he speculated to the police about ways Brown might have met Hough—and that somehow, one of his pie-in-the-sky hypotheticals was rendered by the police as fact.

In December, hoping to discuss some of these uncertainties, I met with Lieutenant Rorrison in a small, windowless room. Brusquely cutting off my inquiries, he said the police were closing out the case. “I’m sure that you’ve heard all kinds of things, and I can assure you that we have heard them too,” Rorrison told me.

The most confounding question in the case is how the meek Brown and the hyperaggressive Tatro—the ultimate odd couple—might have gotten to know each other, let alone become accomplices in a grisly crime. Other than theorizing that they could have crossed paths at a strip club, the police hadn’t connected the two men, and after months of research, neither had I. In Rorrison’s view, discovering the connection between Tatro and Brown would be nice, but it wasn’t necessary. “Who knows?,” Rorrison told me. “At some point they were together, because we both know that their DNA was on our murder victim.”

Dna evidence is the ultimate tool in criminal identification, making older forensic sciences such as fingerprint analysis look like dowsing by comparison. Examining cells left behind at a crime scene or on a victim, analysts check at least 13 key locations in the strands of DNA to create a numerical profile. The profile is then entered into the FBI’s Combined DNA Index System, or CODIS, to see whether it matches the profile of anyone already in the system. In the best-case scenario, with genetic information from all 13 locations successfully identified, the chance of an incorrect, coincidental match is less than one in 200 trillion.

But DNA’s justified reputation for extreme, impartial precision is also what makes it dangerous. When people hear that genetic evidence links a suspect such as Kevin Brown to a crime, his innocence becomes almost impossible to contemplate. Problems do occur with DNA evidence, however—not because the underlying science is flawed, but because the people practicing it are. Contamination can take place when DNA from one case taints evidence from another, or when a crime-lab employee’s own DNA gets into case materials. Problems can occur when a criminalist sneezes, touches lab equipment, or even talks near an evidence sample. Shaking hands can transfer DNA from one criminalist to another. “With the sensitivity of these instruments, just a few cells are needed,” says Greg Hampikian, a professor of biology and criminal justice at Boise State University.

Because of the risk of contamination, lab employees register their profiles in databases—which is why Brown’s DNA was on file to produce the “hit” in the 2012 reexamination of the Hough case. “Everyone will admit that contamination occurs,” says William C. Thompson, a DNA-evidence expert at the University of California at Irvine. “But people vary in their opinions of how common it is.” In 2004, the Washington State Patrol crime lab admitted to 23 instances of contamination and other errors; in 2008, the Baltimore Police Department fired the head of the city crime lab after discovering a dozen instances in which employees’ DNA had tainted evidence. Similar problems have been documented in Pennsylvania, Minnesota, Nevada, and California, as well as in Australia, Italy, Taiwan, and the United Kingdom. “Every lab that I have ever reviewed, when I get their records of unintended events, there are always examples of analysts’ DNA ending up in casework,” says Hampikian.

Rebecca Brown continues to maintain her husband’s innocence. However he may have behaved back in the 1980s, she describes him as “just a nice, gentle, dorky guy” and “my teddy bear.” (Spencer Lowell)

Contamination can take place in pristine modern labs with meticulous protocols. The criminalists working at the SDPD crime lab in the 1980s—before the era of DNA evidence—were like surgeons operating in a barnyard. According to Stam, the former supervising criminalist, Brown worked in the same room where the Hough evidence was analyzed, a space the size of a two-car garage, where evidence swabs were sometimes dried in the open air. “Whether we wore gloves or not was up to the individual examiner,” Stam says. “We didn’t wear masks, for sure.” Although Brown wasn’t personally assigned to the case, he worked near the analyst who was, possibly even at the same table.

Stam, along with former San Diego County District Attorney Paul Pfingst, is among those who believe that Brown might have been falsely incriminated. But the allegation isn’t easily proved, especially because it was purportedly Brown’s semen that was discovered on a vaginal swab from Hough.

The accidental presence of semen is harder to explain than wayward skin cells or saliva. But DNA experts familiar with Brown’s case say that even with sperm, contamination is possible. To begin with, cell samples can be misidentified as spermatozoa, says Dan Krane, a DNA expert at Wright State University, in Ohio. This is especially apt to happen when the sample is old and small, as is the case with the Hough evidence. “There is an awful lot of stuff that looks like ‘maybe that’s sperm,’ ” Krane says. What’s more, he says that the process of “differential extraction,” which is supposed to isolate sperm cells from non-sperm cells for analysis, fails up to half of the time. If differential extraction failed in Brown’s case, his skin cells or saliva could have gotten onto the Hough vaginal swab, and nobody would have been the wiser. The technician might have believed he was analyzing DNA that came from semen.

And even if the sample was correctly identified as semen, that does not mean this wasn’t a case of contamination. The fact is that Brown’s semen was definitely present in the crime lab—and the reason has nothing to do with his old “kinky” reputation.

To verify that evidence tests are set up correctly, criminalists analyze reference samples, also known as standards. Many of these standards are purchased from laboratory-supply companies. But in the 1980s, analysts frequently turned to a fresher, cheaper source. “The semen standards that we used were from ourselves,” Stam says bluntly. Criminalists’ using their own semen was a convenient, common shortcut, but one that, unbeknownst to them at the time, presented a high risk of contamination. “They had Brown’s pure, amplified DNA around,” Hampikian says.

Knowing that Brown’s semen was present in the lab, Stam offered a guess as to how it could have gotten into the Hough evidence. “If someone cut the [Hough] swab with the same pair of scissors that they used to cut Kevin’s swab for use as a standard, who knows?” he says. Contamination via laboratory equipment has been documented before. Brian Kelly, a Scottish police officer jailed for rape in 1989, had his conviction reversed in 2004 after scientists discovered that samples of blood and hair he had voluntarily provided to investigators might have contaminated a semen stain on a dress after both sets of evidence were tested next to each other. The retrial of Carlton Gary, Georgia’s alleged “Stocking Strangler,” was complicated in 2013 when prosecutors similarly revealed that an evidence sample of semen had been contaminated with a laboratory standard.

The SDPD response to all of this is: no way. “We have absolute confidence that there is no basis for any cross-contamination,” Rorrison told the press in October 2014. But the DNA experts I spoke with all agreed that contamination can’t be ruled out. What happened to Brown, Hampikian says, “is the nightmare scenario for any analyst.”

On october 20 of last year, a ranger at Cuyamaca Rancho State Park, located in the arid mountains east of San Diego, noticed a black Ford pickup truck parked at a trailhead. The truck was still there when the ranger drove by the next morning, so he and another park employee hiked down the trail, where they found Kevin Brown. He was hanging from a tree with a white plastic chair kicked out from beneath him.

Brown was now the second suspect in the Hough murder to have killed himself—the third if Tatro’s death was in fact a suicide too.

To believers in Brown’s guilt, his suicide could be seen as a confession. “We were close to arresting him,” Rorrison told me. “He knew it. And he decided to do what he did.”

Brown’s supporters, though, believe that overzealous investigators drove an innocent man to his death. Anxious, depressed, and unable to prove that he hadn’t killed Hough, Brown had been suicidal for weeks, his wife told the police. Her brother had confiscated three guns from the Browns’ house; Rebecca had taken pills away from Kevin and discovered a bullet on the floor. Three weeks before his death she had found a draft suicide note in which Brown thanked friends for their prayers and Rebecca for their marriage.

Channeling her grief into outrage, Rebecca held a press conference in December to announce that she had filed a claim against the City of San Diego for investigative misconduct and wrongful death. The claim, in turn, paved the way for a lawsuit, filed on July 16, against the city and individuals involved in the investigation. The suit could enable her attorneys to obtain detailed SDPD crime-lab records and have them analyzed by independent DNA experts. Evidence of contamination, if it took place, may or may not show up in the laboratory logs. But reviewing them is Rebecca’s only hope to prove that Brown was innocent. “My husband should not be dead,” Rebecca tearfully told me. “He should be sitting here in retirement with me.” (As of press time, all but one of the named defendants have filed a motion to dismiss the lawsuit.)

The night after the press conference, I drove to the house where Hough’s grandparents had lived in 1984 and parked nearby. Walking through the darkness, I passed the convenience store where Hough had bought cigarettes on her final night and then, several minutes later, I reached the beach, which was shrouded in fog.

The dank space beneath the bridge reverberated with echoes of the crashing surf. Some investigators have speculated that Hough came here to meet someone she already knew, but her friend Kim Jamer doesn’t buy it. During their entire trip, she and Hough never exchanged more than a few words with boys at the beach, she told me. “My theory is that she snuck out, that she was sitting there listening to music, and someone saw a crime of opportunity,” Jamer says. “And Claire didn’t hear them until it was too late.”

That someone, I felt sure, was the serial predator Ronald Tatro. On that much, the police and I agreed. I couldn’t say for certain that Kevin Brown wasn’t with him. But based on everything I had learned, it seemed terribly unlikely. Turning inland, I squinted into the darkness under the bridge. Someone was walking toward me, but even as the figure passed by, coming within 10 feet of me, I couldn’t make out his face. I lingered a minute more, then zipped up my coat and left the beach.