“We thought this killer was playing with us,” Adamson recalled, “when he put the skull right there. Some of the investigators thought that maybe it had been hidden and a coyote had carried it to where it was found.”

He called in Search and Rescue volunteers to help look for more remains. “We went a hundred yards down an extremely steep grade, searching for more bones or clothing. It was so steep that we had to use fire truck ladders so Search and Rescue could maneuver on the hill. I was worried about the people searching—some of them were kids. Once, someone dislodged a rock and it almost hit one of the kids in the head.

“But we found nothing at all. It was my choice not to go any farther down because it was just too dangerous. The hill went all the way down to Highway Eighteen.”

Although they found no more of Kimi-Kai’s remains, a forensic dentist was able to match the teeth in the skull to her dental charts. Her mother heard the news without shock. She had accepted a long time before that Kimi was gone. With great sadness, she whispered, “She’s not hurting now. She’s not cold. She’s not hungry. She’s no longer in any kind of pain. That’s been tormenting me for the last nine months.”

AS ADAMSON’S TASK FORCE moved forward, the investigators acknowledged that there were probably a lot more than a dozen victims of the Green River Killer. Adamson himself believed that. Had the killer’s favorite disposal site not been located with the finding of three bodies in the Green River in August of 1982, the man they sought probably would have left all of his victims there.

But the new commander had prior experience with bodies left in water; gases formed by decomposition are so strong that he’d seen a body in an earlier case pop to the surface even though it was weighted down with a concrete block connected to chains.

Adamson felt that both Wendy Coffield and Debra Bonner had originally been left in the secluded spot farther north up the Green River where the other three women were discovered. “But once we found them,” Adamson said, “he couldn’t go back there anymore, so he had to find new places to leave the bodies.”

IN EARLY 1984, the first order of business was to be sure that all the information the new task force had and all that continued to come in was organized. “You’re overwhelmed with information,” Adamson recalled, “and it’s not very well organized. We redesigned the case file books so we could find things easily. We got the physical evidence together. We had one room with case binders listing the Missing, the Homicides, and the Physical Evidence.”

In the early eighties, it was much more difficult to gather absolute physical evidence that will identify a murderer than it is now. Twenty-two years have made a tremendous difference. DNA matching was not a standard forensic tool then. Nor was the computerized Automatic Fingerprint Identification System (AFIS) in general use. The F.B.I.’s old fingerprint system had grown archaic, depending mainly on time-consuming manual methods to match suspect ridges and loops and whorls to the fingerprints in its vast files.

Before AFIS, it took two months for police departments to get reports back on prints submitted for matching in a normal request. Funds to complete the bureau’s goal of automation—thirteen years in the making—were not included in the national budget. The F.B.I. still needed $40 million to completely computerize its system, and Washington State was not yet set up to hook into it.

In the beginning, the Green River Task Force used the Alaska fingerprint identification system. Tests on blood or other body fluids were pre-DNA. All criminalists could accomplish absolutely was to differentiate samples submitted as being from humans or from animal species. They could tell if the specimen had come from a secretor or a nonsecretor (meaning that a small percentage of humans do not “leak” their blood types in their body fluids). For subjects who were secretors, criminalists could determine their blood types. That seemed like a lot two decades ago.

The one thing that was certain, and that had been certain since Dr. Edmond Locard of Lyon, France, directed the very first crime lab in the world, was his Exchange Principle. Formulated in the early years of the twentieth century, Locard’s theory became the basis for any investigation of any crime scene: Each criminal leaves something of himself at a crime scene, no matter how minute, and takes something of the crime scene away with him, even if it is infinitesimal. Every new generation of detectives—and television shows like CSI—works on Locard’s principle. But each decade brings with it more sophisticated tools to help the detectives find the minutiae that solves crimes.

As the Green River investigation moved through the eighties, there would be many advances in forensic science. The earliest, however, required a good portion of each substance to be tested—hair, blood, semen, chemicals, paint, and so on—because the various lab processes would destroy the samples submitted.

Even so, the second Green River Task Force was extremely optimistic in early 1984. Deluged with tips from the public, from prostitutes and street people, and from psychics, they investigated and cataloged all of them. However, they had to have a system to establish priority.

“We had ‘A,’ ‘B,’ and ‘C’ categories,” Adamson explained. “A’s had to be looked at as soon as possible, and they had to be eliminated conclusively before we moved on. B’s were people who might be good suspects, and we certainly needed to look at them, but we could take more time to get to them. C’s might possibly be suspects, but, with them, we didn’t have a lot to go on and we felt it wasn’t really important to get to them in the near future.”

One of the problems Adamson tried to monitor was the natural tendency of detectives to fixate on one suspect, and one suspect only. Homicide detectives are human like anyone else and they aren’t infallible, although their job requires that they produce irrefutable evidence that a prosecutor can present to a jury.

“I kept telling my staff that we had to remember not to get myopic,” Adamson said. “When a detective became too focused on a single suspect, I reminded him—or her—that all of us got ‘obsessed’ with certain people. One of us might be right, but all of us couldn’t be right.”

Dave Reichert was still convinced that Melvyn Foster was the Green River Killer. Another detective was positive the killer was a Kent attorney whose clientele included a disproportionate number of women in the “vice trade.” The lawyer lived very close to where the first five bodies had been left in the river, and this suspect had once spent time in a California town where there were several unsolved murders of women.

“I’d see that attorney, who was in his forties, occasionally,” Adamson said. “He was always immaculately dressed—not a hair out of place. I thought to myself, ‘No way is this guy gonna get in any river, get wet and dirty, and then show up in court.’ I’d been down to the river in my dress shoes, and a month later, you could still see the grime on them, and I didn’t get into the water. The real killer would have had to wade in there in the silt and mud.”

The suspicious-acting attorney was later murdered by a disgruntled tenant in one of the shabby apartment houses he owned.

All of the investigators felt deeply, and it was tough for them to let go of their beliefs. There were just too many truly bizarre people who came to their attention. Even Adamson would soon have his own “favorite” suspect, just as they all did. Rather, he would have three preferred suspects. Some of the public didn’t care who the Green River Killer was. They demanded only an early arrest.