PIPE. T HE KIND THAT YOU PUT IN YOUR MOUTH AND SMOKE.”

McMahon nodded.

“In a checked bag.” My voice registered my incredulity.

“An airline employee remembers telling this guy arriving at the last moment that his duffel was too large for the overhead bin and he would have to check it. The guy was sweaty and distracted, and pulled off his sport jacket and stuffed it into the duffel before giving it to a baggage handler. They're saying he left a hot pipe in the pocket of the jacket.”

“What about smoke detectors? Fire detectors?”

“Baggage compartments don't have them.”

Ryan, McMahon, and I were seated in folding chairs in a briefing room at NTSB central. I could see Larke Tyrell at the end of our row. The front of the room was filled with personnel of the response and investigative teams, the back crammed with journalists.

Magnus Jackson was making a statement, projecting visuals onto a screen behind him.

“Air TransSouth 228 was brought down by an unpredictable confluence of events resulting in fire, explosion, depressurization, and in-flight breakup. In that order. I'll take it step by step, take questions when I'm done.”

Jackson worked the keys of a laptop, bringing up a diagram of the passenger cabin.

“On October fourth, at approximately eleven forty-five A.M. passenger Walter Lindenbaum presented himself to Air TransSouth agent James Sartore for boarding of Flight 228. Agent Sartore had just announced last call for boarding and stated that Mr. Lindenbaum was extremely agitated, concerned that his late arrival had caused the forfeiture of his seat.

“Mr. Lindenbaum had two bags, a small one and a larger canvas duffel. Agent Sartore informed Mr. Lindenbaum that there was no overhead space left for the duffel and that it was too large to fit under the seat. He tagged the bag and told Lindenbaum to leave it on the jetway and the baggage handler would take care of it. Mr. Lindenbaum then removed a knitted fabric sport jacket, put it in the duffel, and boarded the aircraft.”

Jackson brought up a credit card receipt.

“Mr. Lindenbaum's credit card records reflect the purchase of a one-liter bottle of 151-proof Demerara rum on the evening prior to flight.”

More keystrokes, and the receipt was replaced by several views of a charred canvas bag.

“The Lindenbaum bag and its contents, and these objects alone, of all the artifacts recovered from the crash”— the phrase emphasized by a hard look to the audience—“manifest geometric burn patterns showing symmetry and more combustion inside than outside.”

He traced the patterns with his laser pointer.

“Interviews with family members have disclosed that Walter Lindenbaum was a pipe smoker. He was of the habit when entering a no-smoking area of slipping his pipe into his pocket and relighting it later. All evidence points to the presence of a smoldering pipe in the pocket of the Lindenbaum jacket when that jacket went into the cargo bay.”

A murmur spread through the back of the room. Hands shot up and questions were shouted. Jackson ignored them as he projected additional pictures of burned clothing, unfolded then folded.

“Inside the baggage compartment, fragments of smoldering tobacco and ash spilled from the pipe bowl and communicated incandescent combustion to surrounding fabrics in the bag, generating what we call a hot spot.”

More shots of burned canvas and clothing.

“Let me repeat. Geometric burn patterns have been found on no other items recovered from the wreckage. I'm not going to go into it here, but the press release explains how evidence of slow burning of folded clothes inside the bag cannot be explained by anything that occurred after a midair explosion.”

The next visual showed smoke-blackened fragments of glass.

“Mr. Lindenbaum's rum bottle. Inside the loosely packed duffel, smoke spread at a temperature consistent with that of the localized combustion, a temperature warmer than the bottle and its contents, which were not involved in the combustion process. The bottle remained intact, and smoke was deposited on it. These deposits, seen in this view, have been analyzed by our lab. The products of decomposition present in the smoke are consistent with the point of origin as I am describing it. Traces of tobacco smoke were positively identified on the bottle, among other traces, especially since forensic analysis also disposed of unburned tobacco strands in the pipe bowl as reference.”

Jackson switched to a diagram of the plane.

“In the Fokker-100, fuel lines run under the cabin floor, above the baggage compartments, from wing tanks to aft-mounted engines.”

He traced the route with his pointer, clicked to a close-up of a fuel line, then zoomed in on a fitting.

“Our structures team has found evidence of a fatigue crack in a fuel line fitting where it passes through the bulkhead at the rear of the baggage compartment. In all likelihood, this crack was generated by a flawed through-fitting acting as a stress riser.”

A magnified image of a hairline fracture filled the screen.

“Heat from the incandescent combustion in Mr. Lindenbaum's duffel aggravated the crack, allowing minute quantities of vaporized fuel to dissipate from the line into the hold.”

He brought up a dirty and discolored chunk of metal casting.

“Localized heat degradation, manifested in localized discoloration, is clearly recognizable on the fuel line at the point of failure due to heat exposure. I'll go to simulation now.”

Keys clicked, the screen went blank, then filled with an animation of an F-100 in flight. Time ticked in one-second increments at the top of the screen.

The Lindenbaum duffel could be seen high in the left rear of the baggage compartment, immediately below seats 23A and B. I watched it ooze from pink, to salmon, to red, a cold lump in the pit of my stomach.

“Incandescent combustion in the duffel,” Jackson narrated. “A first ignition sequence.”

Pale blue specks began to seep from the bag.

“Smoke.”

The particles formed a fine, transparent mist.

“The baggage compartment is pressurized the same as the passenger cabin, meaning it is supplied with air containing an adequate proportion of oxygen. The significance is that there is a lot of warm air moving around down there.”

The mist slowly dispersed. Red colored the ends of the Lindenbaum suitcase.

“Though it was contained at first, the smoke eventually spread from the duffel. The heat eventually pierced, and then there was a development to laminar flaming combustion outside the duffel, igniting the suitcases on each side and giving off dense smoke.”

Tiny black dots appeared at a fuel line running along the inner wall of the baggage compartment. I stared, mesmerized, as the dots multiplied and slowly descended, or were entrained in the ambient air movement.

“Then began the second ignition sequence. When fuel began to dissipate out of the pressurized line, the quantity was so minute it vaporized and mixed with the air. As the fuel expanded in a vapor state it sank, since fuel fumes are heavier than ambient air. At that point an odor would have been present and easily detected.”

Traces of blue appeared in the passenger cabin.

“Smoke seeped into the cabin through the ventilation, heating, and air-conditioning system, and eventually to the exterior via the pressurization outflow valve.”

I thought of Jean Bertrand. Had he noticed the odor? Seen the smoke?

There was a flash, red spread outward from the Lindenbaum suitcase, and a jagged hole appeared in the rear of the baggage compartment.

“Twenty minutes and twenty-one seconds into the flight, vaporized fuel crossed a wire bundle, which apparently contained some arcing wires, and ignited in a deafening detonation. This explosion can be heard on the cockpit voice recorder.”

I remembered Ryan's account of the pilot's last words, felt the same helplessness he had described.

“The circuit failed.”

I thought of the passengers. Had they felt the shock? Heard the explosion? Did they realize they were going to die?

“The initial explosion blew from the pressurized baggage compartment into the unpressurized fuselage behind, and air loads began tearing parts from the plane. At that point, more fuel escaped from the line and flaming fire ensued in the hold.”

Jackson identified items as they separated and fell to the ground.

“Skin from the aft fuselage. Speed brakes.”

The room was deathly quiet.

“Air loads then blew up through the vertical tail and dislodged the horizontal stabilizer and elevators.”

The plane in the animation pitched nose down and plunged toward the ground, the passenger cabin still intact. Jackson hit a key and the screen went blank.

No one seemed to breathe or move. Seconds passed. I heard a sob, or perhaps only a deep breath. A cough. Then the room exploded.

“Mr. Jackson—”

“Why weren't smoke detect—”

“Mr. Jack—”

“How long—”

“I'll take questions one at a time.”

Jackson pointed to a woman with Buddy Holly frames.

“How long would it have taken to raise the temperature in the duffel to the point of fire?”

“Let me clarify one thing. We're talking about incandescence, a glowing type of combustion generated when the little oxygen available comes in direct contact with a solid, like coals or embers. This is not flaming combustion. In a small volume like the bag's interior, incandescence could be quickly established and maintained at around five hundred to six hundred degrees Fahrenheit.”

His finger found another journalist.

“How could the rum bottle survive the fire in the bag?”

“Easy. On the other end of the temperature spectrum, incandescence can reach eleven hundred to twelve hundred degrees Fahrenheit, the temperature of a lit pipe or cigarette. That's hardly enough to alter a glass bottle containing liquid.”

“And the smoke deposits would remain on the bottle?”

“Yes. Unless it was subjected to a very intense and sustained fire, which was not the case, as it occurred inside the suitcase.”

The finger moved.

“The metal fatigue marks survived as well?”

“To melt steel you need temperatures of twenty-five hundred degrees Fahrenheit or more. Beach marks, your typical evidence of fatigue, generally survive fires of the intensity I'm describing.”

He pointed to a reporter from the Charlotte Observer.

“Did the passengers know what was happening?”

“Those seated close to the flash point would have felt the shock. Everyone would have heard the explosion.”

“What about smoke?”

“Smoke would have seeped into the passenger cabin via the heating and air-conditioning system.”

“Were the passengers conscious the whole time?”

“The type of combustion I've described can give off noxious gases which may affect people very quickly.”

“How quickly?”

“The old, the young, perhaps as fast as ninety seconds.”

“Could these gases have gotten into the passenger compartment?”

“Yes.”

“Have traces of smoke or noxious gases been found in the victims?”

“Yes. Dr. Tyrell is going to make a statement shortly.”

“With so much smoke, how can you be sure about the source of the deposits on the rum bottle?” The questioner looked about sixteen.

“Fragments of the Lindenbaum pipe were recovered, and reference studies were conducted using unburned strands of tobacco adhering to the inside of the bowl. The deposits on the bottle were the by-products of the combustion of that tobacco.”

“How could there have been a fuel leak?” Shouted from the back.

“When fire broke out in the hold, flame impingement affected only a segment of the fuel line. This pulled the wall of the line, or induced a stress that opened very slightly the seed failure.”

Jackson called on a reporter who looked and sounded like Dick Cavett.

“Are you telling us that the initial fire did not directly cause the explosion?”

“Yes.”

“What caused the explosion?” he persisted.

“An electrical failure. That's the second ignition sequence.”

“How sure can you be?”

“Reasonably certain. When electricity sparks an explosion, the electrical energy is not lost, it must ground. Damage due to electrical grounding has been identified on the same segment of fuel line. Such damage is normally seen on copper items and rather seldom on steel parts.”

“I can't believe that the fire in the suitcase didn't cause the explosion.” Cavett made little attempt to hide his skepticism. “Wouldn't that be more normal?”

“Your question makes sense. It's really what we thought at first, but you see, the fumes are not yet mixed enough with air at such short distance from the source of emission. The fumes must mix before ignition can occur, but when it does, the blast is deafening.”

Another hand.

“Was the analysis done by certified fire and explosion specialists?”

“Yes. Outside experts were brought in.”

Another questioner stood.

Eighty-eight people were dead because one man was preoccupied about losing his seat. The whole thing was a tragic mistake.

I looked at my watch. Crowe would be waiting.

Feeling numb, I slipped from the room. I had victims waiting whose deaths were not due to simple carelessness.

The reefer trucks were gone from the grounds of the Alarka Fire Department. The lot held only the company's displaced engines and the vehicles of those assisting me. A single deputy guarded the entrance.

Crowe was there when I arrived. Seeing me, she climbed from her cruiser, collected a small leather case, and waited. The sky was pewter, and a cold wind was tearing through the gorge. Gusts teased her hat brim, subtly reshaping it around her face.

I joined her, and we entered what was now a different type of incident morgue. Stan and Maggie worked at autopsy tables, arranging bones where crash victims recently had lain. Four tables held unopened cardboard boxes.

I greeted my team and hurried to the cubicle I was using as an office. As I exchanged my jacket for a lab coat, Crowe took the chair opposite my desk, zipped open the case, and withdrew several folders.

“Nineteen seventy-nine came up zilch. All MPs accounted for. There were two from 1972.”

She opened the first folder.

“Mary Francis Rafferty, white female, age eighty-one. Lived alone over in Dillsboro. Her daughter checked on her every Saturday. One week Rafferty wasn't in her home. Never seen again. It was presumed she wandered off and died of exposure.”

“How often have we heard that?”

She went to the next folder.

“Sarah Ellen Deaver, white female, age nineteen. Left home to go to her job at a convenience store on Highway 74. Never got there.”

“I doubt we've got Deaver out there. Anything from Tommy Albright?”

“George Adair's positive,” Crowe confirmed.

“Dental?” I asked.

“Yes.” Pause. “You know that first alcove burial was missing its left foot?”

“Albright phoned me.”

“Jeremiah Mitchell's daughter thought she recognized some of the clothing. We're getting blood from a sister.”

“Albright asked me to cut bone samples. Tyrell's promised to rush them through. Did you check the other dates?”

“Albert Odell's family provided the name of his dentist.”

“He's the apple farmer?” I asked.

“Odell's the only MP still out from eighty-six.”

“Many dentists don't keep records past ten years.”

“Dr. Welch didn't sound like the brightest bulb in the marquee. I'm driving over to Lauada this afternoon to see what he has.”

“What about the others?” I knew what her answer would be even as I asked the question.

“The others will be tough. It's been over fifty years for Adams and Farrell, over forty for Tramper.”

She withdrew three more folders and laid everything on my desk.

“Here's what I've managed to dig up.” She stood. “I'll let you know what I get from the dentist.”

When she'd gone I spent a few moments perusing the folders. The one for Tucker Adams contained only the press items I'd already seen.

Edna Farrell's record was a little better, and included handwritten notes taken at the time of her disappearance. There was a statement by Sandra Jane Farrell, giving an account of Edna's last days and a detailed physical description. Edna had fallen from a horse as a young woman, and Sandra described her mother's face as “lopsided.”

I snatched up a black-and-white snapshot with scalloped edges. Though the image was blurry, the facial asymmetry was obvious.

“Way to go, Edna.”

There were photos of Charlie Wayne Tramper, and his disappearance and death were reported in several newspaper articles. Otherwise, there was little in the way of written information.

The following days were like the first I'd spent at the Alarka Fire Department, living with the dead from dawn until dusk. Hour after hour I sorted and arranged bones, determined sex and race, estimated age and height. I searched for indicators of old injury, past illness, congenital peculiarity, or repetitive movement. For each skeleton I built as complete a profile as was possible working from remains devoid of living tissue.

In a way, it was like processing a crash, where names are known from the passenger roster. Based on Veckhoff's diary, I was convinced I had a limited population because the dates entered in his lists matched precisely the disappearance dates of seniors from Swain and adjoining counties: 1943, Tucker Adams; 1949, Edna Farrell; 1959, Charlie Wayne Tramper; 1986, Albert Odell.

Believing them to be the earliest in time, we started with the four tunnel burials. While Stan and Maggie cleaned, sorted, numbered, photographed, and X-rayed, I studied bones.

I found Edna Farrell early. Skeleton number four was that of an elderly female whose right cheekbone and jawbone deviated sharply from the midline due to fractures that had healed without proper intervention.

Skeleton number five was incomplete, lacking portions of the rib cage, arms, and lower legs. Animal damage was extensive. Pelvic features told me the individual was male and old. A globular skull, flaring cheekbones, and shoveling on the front teeth suggested Native-American ancestry. Statistical analysis placed the skull squarely in the Mongoloid camp. Charlie Wayne Tramper?

Number six, the most deteriorated of the skeletons, was that of an elderly Caucasoid male who had been toothless at the time of his death. Save for a height estimate of over six feet, I found no unique markers on the bones. Tucker Adams?

Skeleton number three was that of an elderly male with healed fractures of the nose, maxilla, third, fourth, and fifth ribs, and right fibula. A long, narrow skull, Quonset hut nasal bridge, smooth nasal border, and anterior projection of the lower face suggested the man was black. So did the Fordisc 2.0 program. I suspected he was the 1979 victim.

Next, I examined the skeletons found in the alcove with Mitchell and Adair.

Skeleton number two was that of an elderly white male. Arthritic changes in the right shoulder and arm bones suggested repeated extension of the hand above the head. Apple picking? Based on the state of preservation, I guessed this individual had died more recently than those buried in the tunnel graves. The apple farmer, Albert Odell?

Skeleton number one was that of an elderly white female with advanced arthritis and only seven teeth. Mary Francis Rafferty, the woman from Dillsboro whose daughter had found her mother's house empty in 1972?

By late afternoon Saturday, I felt confident I had matched the bones with their proper names. Lucy Crowe helped by finding Albert Odell's dental records, the Reverend Luke Bowman by remembering Tucker Adams's height. Six foot three.

And I had a pretty good idea as to manner of death.

The hyoid is a small, horseshoe-shaped bone embedded in the soft tissue of the neck, high up behind the lower jaw. In the elderly, whose bones are often brittle, the hyoid fractures when its wings are compressed. The most common source of compressive force is strangulation.

Tommy Albright phoned as I was preparing to close up.

“Find any more hyoid fractures?”

“Five out of the six.”

“Mitchell, too. He must have put up a helluva fight. When they couldn't strangle him, they smashed his head in.”

“Adair?”

“No. But there's petechial hemorrhage.”

Petechiae are minute blood clots that appear as dots in the eyes and throat, and are strong indicators of asphyxiation.

“Who the hell would want to strangle old people?”

I did not answer. I'd seen other trauma on the skeletons. Trauma I found puzzling. Trauma I would not mention until I understood more.

When he hung up, I went to burial four, picked up the thighbones, and brought them to the magnifier light.

Yes. It was there. It was real.

I collected the femora from every skeleton, and took the bones to a dissecting scope.

Tiny grooves circled each right proximal shaft and ran the length of each linea aspera, the roughened ridge for muscle attachment on the back side of the bone. Other gashes ran horizontally, above and below the joint surfaces. Though the number of marks varied, their distribution was consistent from victim to victim.

I cranked the magnification as high as it would go.

When I focused, the grooves crystallized into sharp-edged crevices, V-shaped in cross section.

Cut marks. But how could that be? I'd seen cut marks on bone, but only in cases of dismemberment. Except for Charlie Wayne Tramper and Jeremiah Mitchell, these individuals had been buried whole.

Then why? And why only the right femora? Was it only the right femora?

I was about to begin a reexamination of every bone when Andrew Ryan burst through the door.

Maggie, Stan, and I looked up, startled.

“Have you been listening to the news?” Ryan asked, flushed and perspiring despite the coolness.

We shook our heads.

“Parker Davenport was found dead about three hours ago.”