Rose Brady walked alone between bus stops on a busy street in Baltimore County one evening in April 1987. She was 28, with long, curly brown hair and blue eyes — perfect prey for the predator local police had named the “Sunglass Rapist.”

She hoped so, anyway.

Brady worked for the Baltimore County Police Department and had just been promoted to corporal. When she signed on in 1977 at age 18, she had been one of only 15 female officers in a department of about 1,100. Within a year, though, she’d shown her value, going undercover to help take down a pimp operating out of massage parlors. She had met the suspected pimp with no gun, no cellphone, no wire. Supervisors told her and another female cadet working with her to throw a glass ashtray out the window if he started to make trouble.

On the street in 1987, Brady wore jeans and a light coat, and she briefly touched the outline of the handgun hidden beneath the jacket. She knew the Sunglass Rapist kidnapped women from behind with a gun or knife, covering their eyes with dark glasses before shoving them into his car. She had convinced reluctant supervisors to let her act as a decoy. Everyone worried the rapist was about to turn murderer. Brady had backup waiting to pounce if he struck.

The decoy operation, though, proved a bust, one with a brutally dispiriting twist. That night, the Sunglass Rapist ended up grabbing a 17-year-old two blocks from Brady. The girl escaped and gave police a vague description of his pickup truck. Officers scrambled to converge on him, but when they picked someone up, the girl couldn’t identify him. The attacker had covered her eyes with his signature welder’s glasses. The suspect was released and melted back into the Baltimore night.

You’ve got to be kidding me, Brady thought.

Almost two decades later, Brady was back trying to lock up rapists, this time as the supervising sergeant in the department’s special victims unit. She’d risen through the agency as a repeat trailblazer — the first woman in the motorcycle unit and the first female sergeant in the homicide squad. Now, after nine years investigating murders, she was the first female supervisor in a unit dedicated to tackling rape cases.

Unsolved rape cases were piling up in the county. The Sunglass Rapist was just one suspected serial attacker who had never been caught.

It didn’t take Brady long to see why so many predators remained on the street.

The SVU had always been a neglected squad. In the department’s shimmering multistory headquarters, the unit operated from a converted 10th-floor storage closet. In 2004, the year Brady took over, the team had seven times as many cases as the homicide squad, but only half as many detectives.

Trips down to the department’s evidence room only deepened Brady’s frustration. Most of the evidence in cold rape cases had disappeared, everything from blood samples to basic investigative records.

Then, in the fall of 2004, just months into her new SVU position, Brady happened to attend the annual Fraternal Order of Police “basket bingo” fundraiser. She and family friend Mary Beck went every year. They loved the handcrafted baskets.

On the way, Beck asked Brady whether she was still working homicides.

Brady told her she’d recently been moved to sex crimes and lamented how little evidence remained in the department’s property room.

Beck was a supervisor in the pathology department at the Greater Baltimore Medical Center, which was the main rape care center for Baltimore County. From behind the wheel, she asked Brady whether she was interested in Dr. Breitenecker’s slides.

“Who’s Dr. Breitenecker?” Brady asked.

Beck explained that Dr. Rudiger Breitenecker had for years created microscopic slides of forensic evidence taken during rape examinations at GBMC.

The doctor had begun preserving the potential evidence before the advent of standardized rape kits and a decade before the first breakthrough using DNA to solve crimes in the 1980s. He’d been a quiet crusader of sorts, committed to seeing women’s assailants captured and confident that advances in science and technology would one day make his makeshift DNA database invaluable.

Breitenecker had given some of the samples to police upon request over the years, but by 2004, seven years after he’d retired, most of the slides remained in hospital storage. “How far back do these slides go?” Brady asked.

“1977,” Beck answered. They used to go back to 1975, but already the first two years of slides had been destroyed in adherence to the hospital’s retention policy.

“Whatever you do,” Brady said, “don’t destroy these slides!”

Brady later said it was lucky she wasn’t driving because she likely would have crashed the car.

Rape has been a plague in Baltimore for what can feel like forever.

In the late 1970s, for instance, someone reported a rape to Baltimore city police more than once a day. In suburban Baltimore County, it was more than once every other day. In the mid-1980s, as many types of violent crime went into steep decline, rapes surged. As recently as 2019, both Baltimore jurisdictions hovered above the national average in rapes per capita.

And all along, whether in the 1970s or last year, the numbers have only reflected what was reported to police. The true scope of the crime, experts have long conceded, is much bigger.

Rapes involving women who knew their attackers were notoriously hard to prove. In what are known as stranger rapes, detectives decades ago had a few forensic tools, such as fingerprints and blood typing, to solve the cases, but the process was laborious and involved finding suspects and then manually comparing the prints and blood. More often than not, the evidence detectives retrieved from the hospital or crime scene went into the dumpster within the next year or two.

When Brady first joined the force, one of her chores as a rookie was to transport evidence from the precincts to the Baltimore County police evidence room. The sergeant who ran the evidence room was a “dictator,” she recalled. His credo, Brady said, was, “If the evidence hadn’t already led to an arrest, destroy it.”

“In homicide, they never destroyed anything, no matter what,” Brady said. But in the SVU in the 1970s and 1980s, Brady said, “They were destroying everything just to get it out of the evidence room.”

Baltimore County police were hardly alone in their lack of regard for evidence in rape cases. Agencies across the country regularly have destroyed rape evidence. But Baltimore’s practice was curious in this: Maryland does not have a statute of limitations on rape. Rapists can be prosecuted for their crimes no matter how long ago they were committed. Old evidence can be critical, even decisive.

Capt. Brian Edwards, who led the SVU through several policy reforms in recent years, said police didn’t have Breitenecker’s foresight in the 1970s and 1980s. But he was “shocked” to learn that destruction of evidence continued long after.

“Sometimes, when you’re buried in the policy, you don’t see the bigger picture,” he said. Supervisors dealing with budget concerns and storage costs can get bogged down and lose sight of what they are actually trying to accomplish, he said.

Baltimore County police now save rape evidence for 75 years. “It doesn’t matter how much it costs or how much room it takes. If you need to keep it, you need to keep it,” Edwards said.

The sense of wasted opportunity, Brady said, had only deepened over the years as DNA technology revolutionized crime fighting and gave rape squads a powerful tool for using aging evidence to make new cases.

In 2001, the state legislature moved to try to address the problem by adopting a law requiring that police retain evidence for three years in cases in which someone was convicted of rape, murder or manslaughter. The following year, the law was expanded to cover the length of the convicted person’s incarceration. It was not until 2017 that legislation was enacted to bar the destruction of any rape kit for 20 years.

Brady’s discovery of Breitenecker’s stash meant that despite the department’s years of neglect and destruction, someone else had kept perhaps the most vital and powerful evidence: the genetic signature of the rapist.

Brady allowed herself to dream about what she could do with such evidence. Maybe she would end up with nothing, of course, but she was going to try. Why else be a sergeant?

Brady’s entire SVU squad included a detective team of just three women and three men to help her investigate the roughly 150 to 200 reported rapes every year. She also had a corporal who researched the investigative files and applied for subpoenas.

Budgets, though, were a constant issue in the SVU. One of her detectives, Joan Wheeler-Felts, had long wanted to do cold cases full time, but the position was never approved. She and others would have to fit their efforts in during lunches, after work, in their weekends and downtime. Because the doctor’s slides were covered with a special stain, Brady had to outsource the analysis work to a private lab. For a while, she had to carve out money from other sections of the budget to pay for the independent analysis work.

Brady and her unit, understaffed but newly motivated, got to work. The unit needed to formally subpoena Breitenecker’s slides, and to do that they needed to find cases in which the victims had been treated at Breitenecker’s hospital, GBMC.

Steve Duffey, Brady’s corporal, holed up in the basement of police headquarters, pulling and printing paperwork and any remaining forensic evidence on unsolved rape cases. Sometimes, he found, detectives had managed to spare a piece of clothing. Most of the evidence and the investigative files, which typically include detectives’ notes and interviews with neighbors, witnesses and suspects, had been thrown away, but the bare-bones police reports had been copied and stored on microfiche. They included the basic narrative of the crime — who reported it, what happened, when and where, as well as where the victim had been treated.

When Duffey could confirm GBMC as the hospital where the victim sought treatment, he applied for a grand jury subpoena to obtain Breitenecker’s DNA slides and drove each sample set from the hospital to the police laboratory for analysis.

The trickiest part of the process, Duffey said, was contacting the survivors. Letting them know that there was, years later, a chance that their case could be solved risked re-traumatizing the victims, who had both suffered and lost hope. Detectives started bringing a brochure with a psychologist’s contact information with them on these visits.

Wheeler-Felts said the victims had a range of reactions. Some hadn’t ever told their families about what happened and were shocked when police appeared at their door. Others were excited about the prospect of finally capturing their assailants. Almost all spoke of the way their lives had been altered — the constant struggle to feel safe. They spoke of years of chronic insomnia and nightmares, of daily routines of double-checking doors and windows, changed walking routes, the need to always have someone else in their house. Some had abandoned jobs because a client or a colleague reminded them of their rapist.

Martin Fedric Czosnowski and Anthony Joseph Klanavitch were among the first to be arrested and later convicted by Brady and her squad. The men had offered two women a ride home from a bar in June 1986, but instead had taken them up the interstate to an isolated area near the border with Pennsylvania and raped them. The women escaped, hid in the woods until dawn, found a phone to call the police and went to GBMC for an exam.

They would be the patients numbered 824 and 825 in the hospital’s Rape Care Center, according to Breitenecker’s log.

At the time, the rap sheets of Klanavitch and Czosnowski, then 22 and 21, were clean. But both went on to commit additional rapes and a raft of other crimes. They occasionally would be caught and convicted, and do time in prison. But they’d never been connected to the 1986 attacks.

Two decades later, one of the women raped that night traveled from her new home in Texas to see Klanavitch and Czosnowski plead guilty. They would be sentenced to 20 or more years in prison.

Another one of the early cases was also one of the most brutal: A 58-year-old woman had been raped at knifepoint on June 3, 1981, in her home and then been left hanging in a closet, bound at the hands and feet. She managed to escape after the rapist left, but the rope had carved a neck wound so deep that her larynx was exposed.

By the time Brady’s team tested the slides, the victim had died. But the team found a second, living victim in Breitenecker’s files where the suspect had the same DNA profile; she had been raped by the same man in 1980, just a year earlier. Herman Lee Bolling Jr. had been convicted of raping another woman in 1981 and had served about 25 years for that crime, but he was back out on the street.

Brady’s team arrested him, the newly identified victim agreed to testify against him, and Bolling pleaded guilty to raping her. He was sentenced to life in prison.

“When we first heard that Dr. Breitenecker had this stash of cold-case slides, we were in utter amazement and shock with a little bit of glee,” said one of Brady’s detectives, Jessica Hummel.

But, she said, exciting as it was, they were never certain whether such old evidence could actually solve cases.

“Quite frankly, we were all shocked [the slides] worked so well,” said Lynnda Watson, a police forensic biologist who worked with Brady and her squad.

Hummel said there was a “minor little party” with each new identification made from Breitenecker’s samples. Then it was quickly back to work.

“You always had that next unsolved case,” Wheeler-Felts said.

Similar scenes would play out across the country as law enforcement agencies started to process thousands of rape kits. The revelations from those cases have reshaped the conventional wisdom about rapists and the consequences of their crimes.

One finding is that serial rapists change their modus operandi more often than previously thought. Researchers examining 20 years worth of untested sexual assault kits in Cuyahoga County, Ohio, found that a quarter of the serial rapists in their files had assaulted both strangers and people they knew, according to a 2017 study in the Journal of Criminal Justice.

Another finding is that serial offending is far more common than thought — and expensive to ignore. Prosecutors in Wayne County, Michigan, have found more than 800 suspected serial rapists so far after processing 11,137 untested sexual assault kits in recent years. They calculated the economic cost of just one of the unchecked serial rapists to be $1.5 million, as he assaulted four additional women during the time that the rape kit from his first victim lay untested. The estimates included expenses incurred by the criminal justice system and the toll on the lives of victims, such as health costs and lost wages.

Other research in Cuyahoga County found that suspected rapists often serially committed a myriad of other offenses. A look into the criminal histories of suspects who were identified by testing previously ignored kits revealed that they had an average of seven arrests apiece for felony and other serious crimes, including murder, robbery, assault and domestic violence.

Late in 2004, one of the cold cases Duffey was digging into required calling his boss on her vacation. It was a 1987 case, and Brady’s name was on the paperwork.

“My name’s in the report?” she asked.

“Yes, he did multiple rapes and was wearing welder’s glasses,” Duffey said.

“I’ll be in,” Brady said.

The report Duffey told her about was indeed the decoy operation from April 1987 when Brady was doing her first stakeout as a new corporal, hoping to catch the infamous Sunglass Rapist. Less than a month before the decoy operation, the man had raped someone else, forcing her to wear dark sunglasses. That woman had been taken to GBMC.

Breitenecker’s slides and the woman’s panties, which police had managed to hold on to, were both processed for DNA, according to police records. The resulting profile was entered into the federal DNA database, first created in 1994, and it matched a 53-year-old convicted sexual offender in Baltimore. The man had worked at a local General Electric plant applying enamel to appliances. It was a task that likely required protective glasses, a G.E. human resources contact confirmed to detectives.

Police finally had the guy Brady had been so upset about not catching in 1987. He’d represented everything that was so frustrating about combating serial rapists back then. There were close calls, eye witnesses, even partial fingerprints, but no evidence that would stand up in court and get him off the street. It turned out he assaulted at least two others after that failed decoy operation, including a little girl who was a relative of his. While he had been caught and sent to prison for assaulting the child, he was back out on the street.

The man, Thaddeus S. Clemons, ultimately was convicted in 2005 of five sexual assaults committed between 1985 and 1988.

For Brady, there was perhaps no cold case she wanted to solve more than the 1978 rape of a colleague on the force. Her attacker had hidden in the bushes until she returned home, then forced his way in as she opened her door. Two other women in the same apartment complex had been raped by a stranger over two months. One was grabbed as she took her trash to the dumpsters behind her building. The other opened the door for a man who posed as a taxi driver and asked if she’d called a cab before he broke in.

Brady, who was a rookie at the time, didn’t know there had been a failed stakeout for that rapist, too. Her colleague was traumatized by the attack, Brady said. Eventually the two lost touch, but Brady remained fixated on finding the attacker, telling her cold-case squad, decades later, “I’m not retiring until I solve this case.”

Detectives described Brady as an “Energizer Bunny” fueled by marshmallow donuts and an extra-large sweet tea from Chick-fil-A that she sipped on throughout the day. They gave her boxes of the donuts and gallons of the tea as gifts, especially if they needed to soften her up for something.

When Brady wasn’t at police headquarters, her idea of relaxation was working the mule farm she shared with her daughter and late husband. She and her husband, also a cop, rode their motorcycles on country roads and drove their mules on covered wagon rides along Maryland’s borders with Pennsylvania and West Virginia.

Even riding the Appalachian trail with her cowboy hat on, Brady was available to jump into cases 24/7, taking calls in her covered wagon or leaving in the middle of the night to drive to a major crime scene.

Brady had no idea whether her friend had been treated at GBMC in 1978, but she had reason to hope.

Initially, a police department logbook showed the case had been cleared, meaning solved or otherwise abandoned. She knew it hadn’t been. Brady then learned that sometimes, when a suspect was arrested, the case was marked “cleared” even when police didn’t have evidence strong enough to support charges and had to release the suspect. Frequently, the status in the logbook was never corrected.

Steve Duffey copied the case number from the logbook and rode the elevator down to the fifth-floor central records room. There, he opened a long drawer in a cabinet and thumbed to where the case file should have been.

The investigative report and any remaining evidence were gone, so he pulled a microfiche file, loaded it into a machine, scrolled through until he found the right report, and printed it out. And there, in the hard-to-decipher copy, he saw that Brady’s friend and colleague had been taken to GBMC. So had the two other women from her complex who’d been attacked.

On Monday, Feb. 7, 2005, Duffey presented the three reports to a Baltimore County Circuit Court grand jury, and they agreed to issue a subpoena to have GBMC hand over the slides that Breitenecker had collected from the three women. That afternoon, Duffey drove to pick up the slides, which had each been placed in a plastic tube for safe transport, and drove them back to police headquarters, where he handed them over to Baltimore County’s police forensic lab. Technicians slipped the slides below a microscope to confirm they had enough spermatozoa to test and then shipped them off to a private lab to create a DNA profile.

Although Baltimore County’s police lab by then had its own DNA analysis facilities, it did not have the capacity to analyze Breitenecker’s slides because of the special stains on top of the spermatozoa. Police thus also obtained a “victim standard” cheek swab from the three survivors to confirm the slides had been taken from them and to isolate their DNA strand from the suspect’s strand.

After the DNA profiles were obtained, the Maryland State Police uploaded the suspect’s DNA into the FBI’s database to see if there was a match with anyone already in the system.

The result was disappointing; the DNA profiles did not get a hit with a known offender. However, they did confirm the suspicions the detectives had shared back in 1978: The DNA profiles developed in all three cases matched. It was a serial rapist.

Then Brady’s team received news from state police that gave them the chills.

The national DNA database didn’t only store profiles from convicted sex offenders, but also profiles of unknown assailants in unsolved cases. And the DNA profiles from the three women’s cases matched the profile of an unidentified rapist who had attacked a woman in 2000 in the city of Baltimore.

The grim calculus was clear to them all: They were looking at a predator who had raped women over at least 22 years in a relatively tiny corner of the county, and they had no idea who it was or how many women he had actually raped.

One thing seemed to be clear. It couldn’t be someone currently behind bars. Otherwise, they’d get a DNA hit.

Brady’s detectives soon came across old police reports from the 1970s and 1980s that showed the attacks on the three women had been part of a notorious rape spree that included a number of adjacent neighborhoods. Brady spoke to a former SVU detective who confirmed they had set up a decoy operation to try to catch the man. Newspaper accounts detailed how warnings had been issued to local community groups at the time. Detectives had gone door to door to alert residents that a rapist was coming through unlocked doors and open windows. The stories in the newspapers crept from the inside pages to the front pages. The police were under fire.

But no one was ever caught.

Brady’s squad eventually plastered a large Baltimore County map across a wall of the special victims office, put colored stars on the locations of similar rapes, culled old files, talked to old sources.

They wound up not just frustrated but frightened. The unidentified rapist’s DNA continued to match up to more unsolved cases in the state’s database. Two years into their effort, he’d been linked to at least seven rapes, including the three from the late 1970s.

“So, we’re looking at all of these cases and comparing them and looking at the M.O.s, looking at the locations, looking at your victims and thinking, ‘Good God, who is this monster?’” Hummel said. “Please let us catch him.”

Brady called the Maryland State Police contact responsible for loading profiles into the FBI’s DNA system after each new hit. The two were on a first-name basis by now: “Are you sure there’s no suspect?”

“Rose, nobody,” she was told.

Brady was beside herself. The detectives knew the additional hits on unsolved cases connected to this one rapist were just what they had pulled from one doctor at one hospital.

“By that point, you’re thinking you’re never going to find out who this guy is,” she said.

Brady went up the chain of command and was given an additional full-time detective.

That’s how Detective Evelyn Grant joined Brady’s team.

Grant was not the typical detective. She wore a silver skull necklace and a silver-studded black belt, and had straight, jet-black hair that fell to her lower back.

And she loved to draw pictures.

Grant was an art student in high school who began drawing funny caricatures of her fellow police officers when she joined the department as a cadet in 2000. Headquarters wasn’t thrilled about her hobby and told her to “stop drawing the good guys and start drawing the bad guys.” They sent her to train as a forensic artist creating criminal suspect profiles.

Inside the SVU office, Brady asked Grant to look for any unsolved cases starting since 1977 within the identified neighborhoods, what the team called the rapist’s “hunting ground.” Link them by the rapist’s M.O., not just his DNA.

Grant began pulling not just unsolved rape cases but also other serious sex offenses and recording each one on a spreadsheet detailing the date, survivor name, address, crime and whether the crimes were linked by M.O. or DNA.

Her count reached seven, then 10, then 15, then 30.

Grant sifted through the various reports looking for Identi-Kit suspect profiles. An Identi-Kit is an old-school facial composite that is created by fitting together the eyes, noses and other features that come the closest to matching the victim’s memories of the suspect. It was a far from perfect method and had been replaced years ago.

But as Grant read the police reports, she noticed nearly all of the cases lacked any physical description of the suspect. The man had attacked women from behind or while they were sleeping.

There was one case from January 1979, though. The victim had been treated at GBMC. She was patient No. 240 in Breitenecker’s files. She had briefly caught a glimpse of the perpetrator.

Grant took the Identi-Kit numbers listed in the microfiche report — HH136, BB01, NN14, LL20, EE65, FF02, CC07 — that represented types of hair, eyebrows, noses and other facial features, and began piecing them together.

She took the profile home with her, too, and began the improvised work of aging it nearly 30 years.

Her husband, a detective with the Baltimore city police, was fascinated. He watched nightly as Grant added years to the suspect’s profile and the face of the rapist responsible for two decades of harm began to take shape.