The north wall of room 11 was warm to the touch. Dr. Oswin Prace stood in the empty room at close to midnight, his hand against the baseboard, and he understood that the warmth was coming from something below. The building sat on a bench of land above the small North Carolina town of Grist Falls, three stories of timber on a foundation of local granite, and beneath that foundation, buried in the earth of the crawl space, was a wall of pale rose-colored sandstone that no man in the Blue Ridge had quarried.

The mortar between its stones had been warm the day his construction foreman uncovered it, warm the way a body is warm, and the foreman had asked him what he wanted to do about it. Prace had told him to continue with the foundation. That had been three years before. Now his wife was dead and buried, and he was standing in the darkness of room 11 listening to the sound of her breathing coming from the wall, and the breathing had not stopped when she had.

He had locked the door of the room himself that afternoon after the body had been removed. No one was in it. The sound had been clear and distinct through the ceiling of his consulting office below, the shallow rapid intake, the held moment, the slow rattling exhale, the pattern of a consumptive patient in the final stage. It had stopped when he crossed the threshold.

He checked the bed, the washstand, the desk, the chair, the window, latched and closed. He put his ear against the walls one by one. The north wall, low down near the baseboard, was warm. He had written only six words about the buried wall in his private notebook in the three years since it was covered, in a hand tighter than his usual script: “The mortar was warm on purpose.

” He did not yet know what that meant. He would spend the next three years and four months finding out. He did not sleep that night. He sat in his consulting office with the door open and listened to his wife’s breathing pattern continue from the room upstairs until near five in the morning, when it faded, not stopped, faded the way a sound heard through water fades as the listener draws away.

The second entry in the notebook is dated the 12th of June 1881: “My wife’s breathing has continued after her death and is coming from the wall above the buried stone. I do not know what this is. I will not discuss it with anyone until I understand it. ”

—

The town below the hill had been called Grist Falls since before the county was dissolved in 1917 and its territory absorbed by Yancey and Mitchell counties.

In 1880 its population was about 400 souls, the rest of Meridian County’s 1,911 scattered across the ridges in hollows and along creek roads. The county was known for two things: a modest vein of mica worked along the eastern boundary since the 1850s, and the quality of the air. The higher hollows stood at almost 3,000 feet, and the dryness, the coolness, the pine resin, were exactly what the medical men of the last quarter of the nineteenth century were beginning to recognize as beneficial for the illness they called consumption, which we now call pulmonary tuberculosis. Dr.

Oswin Prace was one of those medical men. Born in Boston in 1832 to a family of some means, he had taken his degree at Harvard in 1858, served as a Union Army surgeon at Antietam and three other engagements in the Eastern Theater, and come home in 1865 with a diagnosis of his own consumption and the quiet knowledge that Boston winters would kill him within a decade. He had held a small practice in Cambridge for eleven years, watching his own condition slowly worsen and his wife Eliza develop the same cough in 1875, reading everything he could find on what the German physician Hermann Brehmer had begun to call the sanatorium method. In the winter of 1876, at the age of forty-four, he sold his practice and his house, packed his wife and their two trunks of clothes, instruments, and books onto a series of trains, and moved to western North Carolina to save what was left of both their lives.

He wanted elevation above 2,500 feet, a south-facing slope taking morning and afternoon sun, proximity to a settlement large enough to supply food and coal and mail but small enough that land could be bought at a price a moderate means could bear. He wanted most particularly a spring on the property, because he had been convinced by his reading that clean cold water at the source was one of the elements Brehmer had not sufficiently emphasized. He rode north through Yancey County in March of 1877 and stopped overnight in Grist Falls. In the morning he spoke with a man named Vardis Karns, who owned 147 acres on the hill directly north of town, running from the county road up to the ridge above.

The tract met every one of Prace’s conditions: between 2,700 and 3,200 feet, facing almost directly south, with a cold steady spring rising from a limestone outcrop halfway up the slope. It was for sale because Karns had lost his wife the previous September to what the Grist Falls physician had called a wasting fever, and he did not want to keep the ridge where he had buried her. Prace bought it for $1,100. He walked the whole property with his wife in April and picked the site for the sanatorium on a natural bench about a third of the way up the slope, near the spring, sheltered on the north by the ridge above and open on the south to the view down the North Toe valley.

A small preliminary house was built by fall, and Prace moved Eliza into it in October. Ground was broken for the sanatorium proper in April of 1878. That is where the first of the anomalies entered the record, though no one at the time thought of it as one. The site Prace had chosen was a level bench running east to west across the slope, about forty yards long by twenty deep, with a slight rise at its center.

When the digging crew came in to prepare the foundation trenches, they found about three feet below the surface at the center of the bench, dressed stone. Not field stone, not the rough gathered stone any farmer would have used for a chimney or a springhouse. Dressed stone, cut on rectilinear faces, laid in courses, mortared with something the crew foreman, a man from Asheville named Otway Bream, could not identify. The stone itself was not local: a pale rose-colored sandstone with a faint tan cast, fine-grained, and in twenty-two years of building in the Blue Ridge, Bream had seen nothing that resembled it.

They had exposed a section of wall about six feet long and a foot deep. Prace came up the slope and looked at it. He asked how far it extended. Bream had probed with a rod east and west along the same course line and found the wall continuing in both directions at least twelve feet each way; he had also probed downward and found it continuing below the frost line to a depth he had not yet measured.

He asked Prace what he wanted to do. Prace stood at the edge of the trench for a long time and did not speak. Then he asked if the wall would interfere with the foundation. Bream said no, the wall was below the frost line, the foundation would sit on the shale bedrock two feet above it, and no one would ever see it again once construction was complete.

Prace said he should continue. Bream had one further question. The mortar, he said, was warm. Not warm from the sun, because the sun had not been on it at any point since it was uncovered that morning, and the day was overcast.

Warm on its own, the way a body is warm, the way a stone that has been sitting near a stove is warm. Prace put his hand on the exposed section himself, held it there for the count of perhaps thirty seconds, and said, “Thank you, Mr. Bream. Please continue with the foundation.

”

That is the entire conversation about the buried wall that appears in the surviving construction record. The foundation was laid over the exposed section within the week. The wall was covered. Construction proceeded through 1878, 1879, and nearly the whole of 1880, and the buried wall was not mentioned in any correspondence between Prace and any other party for the next three years and seven months.

Not to the bank in Asheville that financed the balance of construction. Not to the medical school classmate in Boston with whom he corresponded on professional matters throughout the 1880s. Not most surprisingly, to Eliza’s sister in Cambridge, to whom they both wrote weekly and with whom Prace confided most matters concerning the property. He had told no one.

He had made only one entry about the wall in his own private notebook, in ink very different from the ink of his medical journal, in a hand tighter and more careful, six words: “The mortar was warm on purpose. ”

—

The sanatorium opened on the 14th of April 1881. Three stories of framed construction on the stone foundation of local blue-ridge granite, with a slate roof, a full-length north porch and a full-length south porch, and a dormered attic that housed the resident staff. The ground floor held reception, dining, kitchen, storeroom, Prace’s consulting office, a small library, and an isolation room.

The second and third floors held the patient rooms, fourteen in total arranged along interior corridors, seven per floor. Each room had a single window facing east, south, or west, an iron bed frame, a washstand, a small desk, and a straight-backed chair. The rooms were numbered 1 through 14, beginning at the south end of the second floor and running through both floors in a pattern Prace had laid out on the plan he sent to the county for the building permit. Room 11 was on the second floor, southeast corner, one window facing east.

It was neither the largest nor the smallest of the rooms. It was, in every visible respect, ordinary. The first patient was admitted on the 22nd of April: Alice Combe, thirty-seven, a widowed schoolteacher from Charlotte, diagnosed with pulmonary tuberculosis the previous October. Prace confirmed the diagnosis, assigned her to room 4 on the second floor, and began the standard course Brehmer’s method prescribed: rest, cold water, sponging in the mornings, exposure to fresh mountain air during specified hours, a careful diet of milk and eggs and beef broth, close monitoring of temperature and pulse and expectoration.

Nine more patients arrived through May. Prace’s medical journal entries for those weeks are unremarkable. He was pleased with the building. The kitchen was running well under a housekeeper named Vinnie Lat.

The two attending nurses from Baltimore were competent and steady. The patients were responding to treatment the ways his reading and experience had led him to expect. His wife Eliza, whose consumption had continued to worsen through the winter, was installed in a room adjacent to his own on the third floor and was under his personal daily care. She was the second patient placed in room 11, moved into it on the 18th of May at Prace’s own decision.

His notebook explains it in a single sentence: the east light in the mornings would benefit her particular case, and room 11 had the strongest east light on the second floor. The first patient, a Mr. Corvin Halt, who had entered on the 9th of May with early-stage consumption, had been moved to room 9 after four nights because he complained the east window let in too much sun. Prace transferred him without comment, cleaned the room, and moved Eliza into it that afternoon.

Eliza died in room 11 on the 11th of June 1881 at about a quarter past four in the morning. Prace was in the room with her. Her breathing in the final hours had taken on a particular pattern he had learned to recognize as the sound of a consumptive patient in the last stage: a shallow rapid intake, held for a moment, released in a slow rattling exhale, held again, then another intake. The pattern had been going on since perhaps midnight, and when it stopped at about a quarter past four, Prace listened for another breath, did not hear one, took her pulse at the throat and at the wrist, confirmed absence, and sat with her for the next hour before waking the attending nurse to help prepare the body.

She was buried on the 15th of June in the small cemetery Prace had laid out at the north edge of the property in the shadow of the ridge. Reverend Ambrose Fell rode over from the Methodist circuit and read the service. Prace stood at the graveside beside the housekeeper, the two nurses, the Grist Falls physician, and about a dozen townspeople who had come up to pay respects to a man they had grown fond of in the four years since he bought the Karns tract. Alice Combe came down from the second floor for the service, sat quietly at the back, and returned to her room afterward.

By five in the afternoon the last of the visitors had gone back down the county road toward town, and the sanatorium was quiet again. That night at about half past eleven, Prace was in his consulting office writing up his notes for the day when he heard, through the ceiling above his head, the sound of his wife’s terminal breathing pattern. It was clear and distinct, and it was coming from the second floor above him, from the southeast corner, from room 11, which was empty. Vinnie Lat had cleaned the room that afternoon after the body had been removed, and Prace had locked the door himself with the key he kept on his own person.

He sat at his desk for a long time listening. Then he went upstairs, unlocked the door, and went in. The sound stopped when he crossed the threshold. —

In the first stage of his investigation, which lasted from June through August of 1881, Prace simply tried to determine what the wall was.

He measured its warmth at different points and different times of day: constant, independent of ambient temperature, concentrated most strongly on a section of the north wall about six feet long by four feet high, centered directly above where Bream’s crew had exposed the sandstone. He listened at the wall repeatedly and found his wife’s pattern audible every night, most clearly between eleven at night and four or five in the morning, completely inaudible during daylight hours. He placed a stethoscope against the wall and found the sound clearer than through the ear directly, which he took to mean it was traveling as vibration through the wall’s material rather than as ordinary airborne sound. In late June he went down into the crawl space beneath the ground floor with a lantern and crawled to the point directly below room 11.

He probed the floor with an iron rod and confirmed the buried stone was still there, about eighteen inches below the surface. With a mason’s trowel he exposed a portion of the top of the wall, perhaps a foot square. The top face of the dressed stone had been carved, not with any script or language he recognized: a series of small horizontal marks at regular intervals running along the length of the exposed section, and above them a continuous band of more complex marks at what would have been chest height if the wall had been standing exposed rather than buried. He copied the marks into his notebook by lantern light, replaced the earth, and went back upstairs at about three in the morning.

That night, his wife’s breathing was clearer than it had been on any night since her death. Through the fall of 1881 he kept a night-by-night record, sitting in a chair in room 11 for at least an hour every night from about eleven to midnight. His wife’s pattern was constant. It did not change, did not fade further after that first fading, did not vary in rhythm or quality.

It was, as far as he could determine, an exact recording of the pattern she had been breathing at the moment of her death, played back on repeat with a silent interval of perhaps twelve seconds between repetitions. He counted the intervals with his pocket watch. His notebook for August of 1881 contains a table of thirty-one nights of measurements: on every one of those nights, the interval was between eleven and thirteen seconds. The third patient placed in room 11 was a Miss Elva Renick, thirty-four, widowed, admitted on the 29th of March with what Prace diagnosed as mild bronchitis complicated by suspected early-stage consumption.

She responded to treatment better than any patient the sanatorium had admitted. By the end of May she was walking the south porch for an hour each morning. By the end of June she had gained six pounds. Prace moved her to room 11 on the 2nd of July, three weeks after his wife had died in that room, recording the transfer in a single line: “Miss Renick to 11, better light for reading in the afternoons.

”

He knew what he was doing. A passage written some months later, reflecting on what he had done, admits it: he knew what the wall was doing. He wanted to see if it would begin to do the same thing with a living patient in the room that it was doing with the pattern of a dead one. Elva Renick stayed in room 11 for eight nights.

Prace listened at the wall each night at his usual hour and heard his wife’s pattern continuing unchanged; nothing new, nothing that could be identified as Miss Renick’s own breathing. On the tenth of July she told him she was feeling remarkably well and wished to return to Yancey County to see her sister the following week, and requested he examine her once more to confirm she was fit to travel. Her lungs were clear. Her weight had increased another two pounds.

Her temperature had been normal for eleven consecutive days. He told her she was as recovered as any patient he had seen at that stage of a consumptive process, and she could travel whenever she wished. She left on the 12th of July on a coach that would connect with the train at Marion. She never returned for a follow-up.

She lived until 1912 and died at sixty-five of what her attending physician recorded as heart failure; her lungs, according to the same physician, had been clear at every examination over the twenty-six years he had been her family doctor. Whatever she had come to Grist Falls with had left her, and it had left her cleanly. The fourth patient placed in room 11 was Miss Mabel Core, forty-one, a spinster who had been a milliner in Asheville until a bout of pleurisy in the spring of 1881 left her weakened and unable to continue her work. Prace’s admission examination found no evidence of consumption.

He diagnosed residual pleurisy complicated by general weakness and placed her in room 11. His notebook, in the entry made the night of her admission, contains a line that the historian Ren Vessel would come to describe as the moment at which Dr. Oswin Prace stopped being a physician and became something else: “She is not consumptive. She will be a clean subject.

We will see what the wall does with a body that has nothing to teach it. ”

Mabel Core slept in room 11 on the night of the 15th of July. Prace did not go to the wall that night. He went on the 16th.

What he heard behind his wife’s pattern, faintly, at a lower volume, was a second respiration. It was slower, fuller, quiet and even, and it had a small catch on the inhale that Prace had noted during Miss Renick’s admission examination as a residual sign of the bronchial episode she was recovering from. It was Elva Renick’s breathing pattern, beginning to be recorded two weeks after she had left the building. Prace sat at the wall for four hours that night.

What he wrote the following morning, in his tight private hand, was the first thing in any of the three notebooks that indicates he understood what he had discovered: “The wall does not record only the dying. It records the sleeping. It records what is above it, and it holds what it records for a period I cannot yet determine. What I must now discover is what use it makes of what it holds.

”

Mabel Core stayed in room 11 for four weeks. Her pleurisy improved for the first week. In the second week, her cough returned, mild at first, then persistent. In the third week the cough developed a particular character that Prace recognized: the small catch on the inhale.

It was Elva Renick’s cough. And Miss Core, whom Prace had examined on admission and found clear of consumption, was now expectorating small amounts of blood-streaked mucus, running a low fever in the evenings, and had lost four pounds. On the 9th of August 1881 he wrote in his medical journal, in his ordinary hand, for the ordinary record, that patient Core was showing signs of pulmonary involvement not present at admission, suspected primary tuberculosis presumably subclinical on entry and now developing under the strain of pleuritic recovery, course of treatment continued. In the private notebook that night he wrote something different: “The wall does not merely record.

It also administers. A pattern absorbed into the stone is subsequently imprinted upon whoever sleeps above it. My wife’s pattern is still audible and has not been transferred to Miss Core because my wife’s pattern is a terminal pattern and Miss Core is not near enough to death for the wall to work upon her with it. Miss Renick’s pattern, being the pattern of a living tuberculin in early stage, has been transferred.

Miss Core is now developing Miss Renick’s disease. She will develop the disease as Miss Renick would have developed it if Miss Renick had not recovered, because the wall is not delivering the illness that would kill Miss Core on her own trajectory. It is delivering the illness that was owed by Miss Renick and that Miss Renick did not pay. That is what the wall is.

It is an instrument that keeps a running account of what is due in the language of respiration and enforces the account. ”

Mabel Core died in room 11 at about two in the morning on the 29th of August 1881. Her terminal breathing pattern in the final hour before her pulse ceased was in exact rhythm with the breathing pattern Prace, sitting outside the door in the corridor with his ear pressed to the panel, could still hear coming faintly from the north wall: Miss Renick’s pattern, delivered whole. He sat in the corridor and listened to the two rhythms rise and fall together in a manner that suggested rehearsal.

When the pulse at Miss Core’s wrist stopped at last, the pattern from the wall did not stop. It went on for what he later noted as thirty-seven minutes after Miss Core’s heart had ceased, and during those thirty-seven minutes her chest continued to rise and fall in the same rhythm, though there was no pulse and no color in her face and no other sign that any life remained. He sat with the body until sunrise. He wrote the death certificate in his ordinary hand, gave the cause as advanced pulmonary tuberculosis with acute pleuritic complication, and cleaned the room himself rather than leaving it to Vinnie Lat.

He opened the window to air it and locked the door. He wrote in the notebook that evening what may be the single most important sentence in the entire three-volume record: “I have two courses available to me. The first is to close room 11, admit no further patients to it, seal it, and report what I have learned to the medical authorities in Raleigh and the ecclesiastical authorities in Asheville. The second is to continue using room 11 in the ordinary course of the sanatorium’s operation, understanding that patients placed in it will develop the disease patterns held by the wall from previous patients, and that some of those patients will die of diseases they did not bring in with them.

The first course requires that I explain what the wall is doing, and I cannot explain it. I do not understand it. I understand only what it does. And if I close the room and cannot explain what I have closed it for, the wall will continue to hold what it holds, and it will begin, I now believe, to draw.

It is not passive. It records because it wants what it records. The pattern of my wife has been in the wall for eleven weeks and has not faded. That is because my wife is dead, and the wall cannot draw her.

Miss Renick’s pattern has faded already because Miss Renick was drawn through Miss Core, and the account is closed. The wall is hungry. If I leave it with only my wife’s pattern, held indefinitely because unclaimable, the wall will find some other way to feed itself. I do not know what that way will be, but I do not want to find out with my own body or with the body of any person on this hill who does not already have consumption.

The second course is the course I am going to take. I will place in room 11 only patients who are already dying. The wall will draw from them what it would otherwise draw from someone else. And I will bear the account.

”

Ren Vessel, in her manuscript, calls this Prace’s justification. She writes that it is the most articulate defense of a monstrous decision she has ever read in an American medical document, and she writes that she does not believe it. She believes Prace made the decision he made for a different reason, and that the reason is on the same page of the notebook, three lines below the sentence about bearing the account: “My wife’s breathing continues from the wall. It has not faded.

It will not fade. If I close the room, I will not hear her again. If I keep the room in use, I will hear her every night until the wall is either destroyed or fed enough to release her. I do not know which of those will come first.

”

—

He kept the room in use. Through the fall and winter of 1881 and 1882 he admitted patients to the sanatorium, and he placed in room 11, according to the medical journal and the private notebook read together, only those whose admission examination showed advanced consumption with a probable survival period of less than six months. There were four such patients between August and December of 1881. All four died in room 11.

Two died with the terminal breathing pattern of Eliza Prace. The other two died with patterns the wall had begun to hold from other patients on the second floor who, in the course of treatment elsewhere in the building, had occasionally spent time in room 11 visiting or convalescing during a room reassignment. By spring of 1882 the wall held, by Prace’s careful count, seven distinct breathing patterns: his wife’s terminal pattern; Miss Renick’s early-stage tubercular pattern, which had returned to the wall from Miss Core in the final hours before her death and had been faintly audible ever since; and five more absorbed from patients who had been in room 11 at any point during their stay. Some were fading, which Prace attributed to having been partially drawn through subsequent patients who had died with elements of the patterns present in their terminal breathing.

Others were holding steady. The medical journal for the same period presents an ordinary Appalachian tuberculosis sanatorium at about the industry average. Prace admitted twenty-three patients between April of 1881 and April of 1882. Fourteen died: slightly above average, but not enough to attract attention.

His letters to Boston discussed cases in general terms, the difficulties of running a remote sanatorium, the death of his wife in restrained and dignified terms. They did not mention the north wall of room 11. He had also, over three nights in early May, gone back down into the crawl space and exposed a larger section of the buried stone, about four feet long by three feet high. On this larger section he confirmed that the continuous band of complex marks at chest height was a system of some kind.

He copied every mark into his notebook by lantern light over three consecutive midnights. In the four-foot length of exposed wall he counted 143 distinct marks in the band, each different from the one before it. He estimated from the density that the full band on the full length of the wall would contain something over 400 distinct marks before it began to repeat, if it repeated at all. He did not know what the marks were, what alphabet they belonged to, who would have cut them.

He knew only that they were there and that the mortar between the stones was warm. —

Dr. Corvis Shaw came to the sanatorium in April of 1882. He was twenty-seven, a recent graduate of the Baltimore Medical School who had answered a small advertisement Prace had placed in the American Medical Journal seeking a junior physician to assist in the operation of a tuberculosis sanatorium in the North Carolina mountains.

He arrived on the 14th of April, was met at the coach stop by Vinnie Lat with the sanatorium’s small buggy, and was installed in a room on the third floor by evening. Room 11 at that moment was occupied by a patient named Vardis Bree, thirty-three, admitted the previous month with advanced pulmonary tuberculosis and expected to survive perhaps another six weeks. Shaw met him, examined him with Prace at the bedside, and agreed with the prognosis. Shaw did not learn about the wall for four months.

He learned about it on the 17th of August 1882, the night Vardis Bree died in room 11, and Prace, who had been at the bedside with Shaw present, listened to Bree’s terminal breathing pattern take on, in the final ten minutes before the pulse ceased, the exact rhythm of the pattern audible from the north wall of the room. That pattern was the terminal pattern of a Miss Henrietta Gore, a Grist Falls schoolteacher of twenty-eight, who had been admitted the previous March, placed briefly in room 11 for two nights during a reassignment in April, had subsequently recovered fully and returned to teaching in the town below. Prace had confirmed her recovery in a follow-up examination in June. At the time of Vardis Bree’s death she was alive and well in Grist Falls, and her breathing pattern was in the wall, and it was being drawn out through Bree, and Bree was dying to Gore’s rhythm.

Shaw heard it. He listened for perhaps three minutes without moving. Then he asked Prace quietly what he was hearing. Prace said, “Come to my office when Mr.

Bree has finished. ”

Bree finished about twenty minutes later. They took the body out together, prepared it for burial, locked the door of room 11, and went down to the consulting office. Prace opened the false bottom of his desk and put the three notebooks on the table in front of Shaw, who read the first that night, the second the following night, and the third over the two nights after.

By the 21st of August he understood what Prace had been doing. Shaw’s letter to Josiah Prace, written the following spring after Prace’s suicide, describes those four nights of reading in detail. He says he almost left the sanatorium after the second night. He says he almost went to the sheriff’s office in Grist Falls the morning after the third.

He says that on the fourth night, after finishing the third notebook, he sat at his desk on the third floor with a bottle of laudanum in one hand and a razor in the other and considered which he would use on himself. He used neither. He went down to the consulting office at four in the morning, woke Prace, and told him he understood everything. He told Prace he considered what he had done a monstrous violation of every principle of medicine he had been taught to uphold.

And he told Prace that he would nevertheless stay at the sanatorium and continue to assist him. He would stay, he told Prace, for the same reason Prace was staying: because he had read what was in the notebooks, because he had understood the argument about the wall drawing regardless of whether the room was in use, because he did not believe closing the room would end the deaths but only relocate them, and because he could not, having read what he had read, in good conscience leave the operation of the sanatorium to Prace alone. He would stay to bear witness and to record what he witnessed. He would keep his own notes.

Shaw did keep his own notes. They do not survive. He burned them in the spring of 1885 after Prace’s suicide, before he wrote the letter to Josiah. The letter itself, seventeen pages in a careful clear hand, is what survives of Shaw’s contribution to the record.

It sat in a strong box in Josiah Prace’s Boston office until Josiah’s death in 1927, then in the manuscript collection of the Boston Medical Library until Ren Vessel requested it in 1972. —

Between August of 1882 and September of 1884, thirty more patients died in room 11 or in the two rooms adjacent to it on the second floor. Some died with Eliza Prace’s pattern. Some died with patterns absorbed from patients who had recovered and left.

Some died with patterns absorbed from patients who had died elsewhere in the sanatorium, whose respiratory sounds had been present in the building long enough for the wall to record them from a distance. Prace and Shaw together, in the last two years of the sanatorium’s operation, identified nineteen distinct breathing patterns held by the wall at various times. Some faded quickly, some held for months. One pattern, that of a patient named Avel Cull, who had died in room 12 rather than room 11 in November of 1882, was audible from the wall of room 11 for over a year and was drawn through three subsequent room 11 patients before it finally faded.

Prace’s notebook records the third of these transfers with a single sentence: “The wall has begun to work through the walls of the adjacent rooms. ” That is when they understood that the sanatorium was not sealable by closing a single room. The mortality figures for 1883 and 1884 were what finally attracted the attention of the county and the state. By the end of 1883 the sanatorium had operated for two years and eight months and recorded a cumulative mortality rate of approximately 62 percent of admitted patients, some fifteen percentage points above the industry average for comparable institutions.

By June of 1884 the rate had climbed above 70 percent. Families of deceased patients began writing letters. Two families in particular, those of Avel Cull and of a patient named Pearlina Vast, who had died in room 10 in April of 1884 with a breathing pattern both Prace and Shaw recognized as belonging to a woman who had left the sanatorium recovered six months earlier, wrote directly to the North Carolina State Board of Health in Raleigh requesting an investigation. Sheriff Cadmus Pellar of Meridian County had also been receiving inquiries, not from families, from the townspeople of Grist Falls.

There was, Pellar wrote in the opening paragraph of what would become his sixty-one-page investigative report, a growing sense in the town below the hill that something was wrong at the sanatorium that could not be accounted for by the usual difficulties of caring for tubercular patients. People spoke of the number of coffins that came down the county road. They spoke of the fact that patients who had entered with mild conditions had, in an unusual proportion of cases, subsequently died. And they spoke of a rumor that had begun to circulate that summer, whose source was Vinnie Lat, the housekeeper, who had told her sister that she sometimes heard, when cleaning the rooms on the second floor in the late afternoon, the sound of patients breathing in rooms that were empty.

Vinnie Lat was known as a steady and unfanciful woman. Her sister had told two other women in town. Those two had told others, and by August of 1884 the rumor was general. Pellar went up the hill on the 14th of August.

He was met by Prace at the front porch and given, at Prace’s own initiative, a full tour of the building. Prace took him into every room, showed him the medical journal, showed him the admission and discharge records, and explained the sanatorium’s methods and mortality figures in the professional terms he had spent his career learning to use with laymen. Pellar was not entirely a layman. He had served eight years as sheriff, grown up on a farm in the hollows west of Grist Falls, and had a steady practical intelligence that had made him effective in his job.

He listened. He walked the rooms of the second floor. He stood for a long time in room 11 with his hand on the north wall, and at the end of the tour he asked Prace a single question: “Doctor, why is the north wall of room 11 warm? ”

Prace stood in the corridor for perhaps ten seconds before he answered.

His answer, as Pellar recorded it in the report, was that he did not know why the wall was warm, that he had noticed the warmth himself and been unable to account for it by any mechanism of heating or plumbing or geological feature he had been able to identify, and that he had chosen not to disturb the wall because disturbing it might have compromised the structural integrity of the building. Pellar wrote in the report that he did not believe Prace’s answer. Prace had answered without hesitation, with the professional composure of a man who had rehearsed the answer many times. And during the ten seconds before he spoke, Prace had looked once at the door of room 11 and once at the ceiling of the corridor, and neither look had appeared to Pellar to be the look of a man puzzled by an unanswered question.

—

The state inspector, Dr. Audry Piles of the North Carolina State Board of Health, arrived on the 12th of September 1884. He was fifty-one, had held his position eleven years, and had inspected sanatoria, asylums, and county infirmaries across most of the state. He was a competent physician, a thorough inspector, and a man known in Raleigh for producing reports that did not exceed their evidence.

He spent three full days at the sanatorium. He reviewed the medical journal in detail. He examined every current patient. He interviewed both nurses.

He interviewed Vinnie Lat. He interviewed Shaw. And on the third afternoon, in the consulting office with the door closed, he interviewed Prace at length. Piles’s four-page public report, filed on the 24th of September 1884, recommended that the sanatorium at Grist Falls be closed to new admissions immediately, that its remaining patients be transferred to other facilities as their conditions permitted, and that the building be inspected by a state engineer before any future use of the property for medical purposes be authorized.

The recommendation was accepted by the board on the 2nd of October. The sanatorium was formally closed on the 7th of October. The last three patients were transferred to Asheville by the 15th. The building sat empty by the 18th.

Piles’s one-line private note, filed in his personal papers rather than in the board record and not discovered until 1982 when a graduate student named Portia Kern was cataloging his papers at the North Carolina State Archive, reads: “There is something below the building that has learned room 11. Prace knows what it is. Prace will not say. I have no legal instrument that will make him.

”

—

The sealing of the chamber was Prace’s idea, not the state’s. The state had ordered the building closed. The state had not ordered the chamber sealed, because the state did not know the chamber was there. What Piles had suspected, and what his one-line note recorded, he had chosen not to include in the public report because he had no evidence to support it, and because, as he wrote in a letter to his wife the day he returned to Raleigh, he did not want to be the man who had told the world what the sanatorium at Grist Falls had been.

Prace went to see Reverend Ambrose Fell on the 18th of October, eleven days after the sanatorium had been formally closed. Fell was fifty-six, had served the Methodist circuit in the Grist Falls area for nineteen years, and had known Prace since he bought the Karns tract seven years earlier. He had buried Eliza Prace. He had ridden up to the sanatorium a dozen times over the years to sit with dying patients or read the service at the small cemetery.

They had come to something that resembled a friendship. Fell’s account of the meeting, written in the winter of 1885 and preserved in the North Carolina Methodist archive at Duke University, describes it as the strangest professional conversation of his life. Prace came to the parsonage on a Saturday afternoon and asked to speak privately. Fell took him into the study and closed the door.

Prace sat down and said, without preface: “Ambrose, I have discovered something on my property that must not be left in the ground uncovered. It has been in the ground since before the county was surveyed. It has been active during the operation of my sanatorium. It is directly responsible for the deaths of a number of my patients.

I have chosen not to reveal what it is to the state or to the medical community because I do not know how to explain it in the language they will accept, and I do not believe explaining it would result in the outcome I need. What I need is to cover it. Not to destroy it, which I do not believe can be done. Not to expose it, which I believe would be much worse than leaving it as it is.

To cover it, to pour a quantity of lime concrete over the top of it in such a way that it cannot be uncovered by ordinary means for as long as the building above stands. I need a witness who will help me do this and who will not speak of it afterward. I need you. ”

Fell wrote that he sat for a long time and did not answer.

He asked Prace a single question: whether Prace was certain that pouring concrete would end whatever had been happening. Prace answered that he was not certain, but that it was the best course available. Whatever the chamber was doing, it was doing it through contact with air and through vibration transmitted through stone. Concrete would block the air.

Concrete would dampen the vibration. It would not stop the chamber’s operation entirely. It would, Prace hoped, reduce it to a level at which it could no longer reach the rooms of the building above. He said he did not have a better answer.

He said he had thought about it for months. He said the alternatives were worse. Fell agreed. He wrote that he agreed for two reasons.

The first was that he had come to trust Prace’s judgment on medical matters and his character on human matters, and Prace was asking for something that touched on both. The second was that he had, riding up to the sanatorium three days before the closure to bring communion to one of the last patients, sat in the second-floor corridor for a few minutes while the patient rested and heard something coming from behind the door of room 11. The room was empty. Fell had checked.

What he had heard coming from the empty room was the sound of a woman breathing in the final stage of consumptive collapse. He had recognized the pattern. He had heard it many times over many bedsides in nineteen years of ministry in the mountains. He had not told anyone.

He had ridden back down the hill that afternoon and had not slept well that night, and in the small hours had asked God what he was supposed to do about what he had heard. —

The mason was Barnaby Roose of Marion, sixty-two years old, over forty years in the trade. Prace hired him on the 21st of October to pour a lime concrete pad in the crawl space at a location Prace would identify. The pad was to be six feet by four feet by ten inches deep.

Roose was to bring the lime and sand and one assistant; Prace would provide the water and the labor of one additional man, which would be Fell. The job was to be kept confidential. Prace would pay double the usual rate. Roose agreed.

The four men met at the sanatorium on the 24th of October at about three in the afternoon. Roose unloaded the lime and sand at the north porch. Prace, Shaw, Fell, Roose, and Roose’s assistant, a man named Merl Vane, went down into the crawl space with lanterns. Prace directed them to the location above the buried wall.

Roose examined the earth floor and confirmed it suitable to receive a pour if the loose earth was first tamped down and a layer of gravel laid over it. Prace had gravel on the property from the drive. Vane brought a wheelbarrow of it down. The crew tamped the earth, laid the gravel, framed the pad with boards, and began mixing the concrete at about five in the afternoon.

The pour began at six. It was completed at about nine. During the pour, and during the four hours after it, while the concrete was setting, the wall sang. That is Reverend Ambrose Fell’s word for it.

He uses it without qualification and without apology because he could not think of another word to describe what he heard. “The wall was singing when we poured the concrete,” he writes, “and it did not stop singing while we watched the concrete set. The pattern was not the terminal breathing of a consumptive patient. It was not any breathing pattern I had ever heard.

It was to my ear a sequence of tones, not tones a human voice would produce. The pitch was too low, the rhythm was too steady, but it was distinctly a series of pitched sounds coming from the mortar between the stones we were covering. And it began the moment the first shovel of concrete entered the pad frame and continued without pause until the last of us climbed out of the crawl space at about one in the morning. ”

Roose, in the account he gave his grandson forty-eight years later, recorded in the margin of the family Bible, said the same thing in different words.

He told his grandson he had poured concrete on a mouth that was still breathing. The mouth had been under the concrete, and the mouth had been making a sound, and the sound had continued the whole time he watched the concrete set. He had gone home to Marion the next day and had not spoken of it to anyone, including his wife, for forty-seven years. He told his grandson only because he was old and dying and did not want the memory to go with him unshared.

His grandson asked what the sound had been. Roose said it was a very old sound. It was a sound the wall had been making for a very long time before anyone had listened to it, and it was a sound that would continue after they were all in the ground themselves. Then he refused to say anymore.

Shaw’s letter to Josiah Prace confirms the sound. Shaw writes that he had been present in Prace’s consulting office for hundreds of hours, that he had listened at the north wall of room 11 on dozens of nights, that he had heard the breathing patterns of many dead and departed patients coming from that wall. The sound during the pour, he writes, was different in kind from all of those. The breathing patterns had been recordings.

The sound during the pour was the wall itself. It was, he writes, the first time in the sanatorium’s operation that he had heard the wall as itself rather than as an instrument playing back what it had absorbed. He says that Prace stood at the pad frame for the entire three hours the pour took, holding a lantern, not speaking, and that Prace’s face during that time was the face of a man watching the last event of his life. —

Prace and Fell and Shaw came up out of the crawl space at about one in the morning on the 25th of October.

Roose and Vane loaded the wagon and left before dawn. Fell rode back down to Grist Falls at first light. Shaw slept perhaps four hours. Prace did not sleep at all.

He sat in his consulting office and wrote in the third notebook the entry that would become the last long entry in that volume. It begins with a description of the pour. It continues with a description of the sound. It ends with this paragraph:

“I have covered what I found.

I have not destroyed it. I do not know that it can be destroyed. What I have done is placed a barrier between it and the rooms above. And I have done so knowing that the barrier will hold only as long as the building above holds and the concrete below holds, and that both of those will hold only for the ordinary span of such things.

Someday, in a year I will not live to see, the building will collapse or be dismantled, and the concrete will crack, and the wall below will be exposed again to air and to vibration, and to whatever passes above it. On that day, whoever is nearest to it will begin to be recorded. I do not know how quickly the wall will resume what it was doing. I do not know if it has been dormant only during the time it was in the ground undisturbed, or if it was already doing something in the ground before we built above it, and I have merely closed one channel it had opened.

I have written down what I know so that whoever comes to it next may know as much as I do. I have not sent this record to anyone. I have hidden it. I have hidden it because I do not want the wall to be found before it is time for it to be found, and I do not know when that time will be.

I am tired. I do not know how to live in this building any longer. I do not know how to leave it either. My wife’s pattern from the wall is faint now.

It may fade entirely in the weeks to come. When it fades, I will not have any reason to remain. ”

He remained through the winter. Shaw stayed with him.

The two men lived in the empty sanatorium through the winter of 1884 and 1885, keeping fires in the kitchen stove and in the consulting office fireplace, eating what Vinnie Lat continued to bring up from Grist Falls three times a week, walking the empty corridors, and listening. Shaw wrote in his letter that the breathing patterns from the north wall of room 11 grew fainter through November and December but did not entirely disappear. Eliza Prace’s pattern faded more slowly than the others. It was still faintly audible on the night of the 2nd of March 1885, which was the last night Shaw heard it.

Dr. Oswin Prace killed himself in the consulting office on the ground floor of the sanatorium at approximately eleven in the morning on the 3rd of March 1885. He used a service revolver he had kept from the war. Shaw found him.

There was no note. There did not need to be a note. The three notebooks were in the false bottom of the desk. Shaw did not open the false bottom.

He did not know it existed. He locked the desk, wrote the death certificate, sent word down to Grist Falls, and by the following week had begun the business of settling Prace’s estate and closing the property. Shaw left Grist Falls at the end of March. He returned to Baltimore, wrote the seventeen-page letter to Josiah Prace, sent it by post, and never spoke of the sanatorium again in any letter or paper that survives.

He practiced medicine in Baltimore for the rest of his career, married in 1890, had three children, and died of ordinary heart failure in 1931 at the age of seventy-six. His obituary in the Baltimore Sun mentioned only that he had assisted in the operation of a tuberculosis sanatorium in North Carolina in the mid-1880s before returning north to establish his own practice. Josiah Prace received the letter in April of 1885, read it once, put it in a strong box in his office, and did not open the strong box again for the rest of his life. He wrote no reply to Shaw.

He made no attempt to communicate with the state authorities in North Carolina. He did not visit the property. He arranged through a Boston attorney for the sale of the sanatorium building and surrounding acreage to a family named Vollard from Marion, who purchased it in 1887 for less than a third of what it had cost Oswin to build. The Vollards used the building briefly as a summer boarding house, closed it after two seasons, and let it stand empty.

It changed hands three more times before 1960. It was empty for most of the decades between. —

Wilder Cance bought the building in 1961 from a family named Boag who had inherited it and could not afford the taxes. Cance was fifty-three, recently retired from a career in the Norfolk shipyards, looking for a mountain property to fix up and live in.

He got it at auction for a very small sum. He moved in that summer and began the slow work of restoring the ground floor for use as living quarters. In July of 1962, prying up a warped floorboard in what had been the consulting office, he found the false bottom of Prace’s desk and the three leather-bound notebooks wrapped in oilcloth, with a small brass key on a length of ribbon. The oilcloth was old but not brittle.

The ribbon was faded but had not rotted. When he opened the first notebook, the ink on the page was still crisp, still black, and could have been written the week before. He read the first entry twice, sat with it, and turned the page. He read the first entrance of the first notebook that afternoon and the last entry of the third notebook by midnight.

He did not sleep in the building that night. He drove into Grist Falls, stayed at the motel on the courthouse square, and the next morning telephoned a professor in the history department at the University of North Carolina at Chapel Hill, whose name he had gotten from the librarian at the Yancey County Public Library. The professor’s name was Ren Vessel. Cance said he had found some notebooks and thought they might be of interest.

Vessel drove up from Chapel Hill the following weekend. She photographed every page of every notebook that weekend on a Kodak she had brought with her. Cance kept the originals. The paper trail that would run to 190 pages of unpublished manuscript, restricted at UNC Chapel Hill until the second decade of the current century, began that Saturday morning at the kitchen table in the former sanatorium building.

Cance also twice in the autumn of 1962 heard the breathing pattern from the wall. He was sleeping in room 3 on the second floor at the time, having done partial restoration work on that particular space. Room 11 he had left untouched. Both times he heard the pattern, he went into the corridor and stood outside room 11 and listened.

Both times the pattern was audible, faint, but distinct. And both times it was a slow, measured breathing with no obvious quality of illness in it. Cance moved out of the building the following spring. He sold it at a loss to a dealer in Asheville, who resold it to a family named Trell.

He kept the notebooks in his own possession until his death in 1979, at which point they passed to Vessel under the terms of his will. Vessel had by then spent seventeen years on the case. She had found the Pellar sheriff’s report in 1968 in the Meridian County records transferred to UNC Chapel Hill after the county’s dissolution in 1917. She had found the Shaw letter at the Boston Medical Library in 1972.

She had found the Fell account at Duke University in 1974, and the Roose family Bible margin note the same year, through a great-grandson of Barnaby Roose still living in Marion. She had found Piles’s four-page public report in the North Carolina State Archive by 1970. Piles’s one-line private note was found in 1982 by the graduate student Portia Kern, who sent Vessel a photocopy. Vessel added it to her manuscript in the final chapter.

She died in 1991. Access to the manuscript opened on schedule. A researcher named Melanie Proud found the papers in 2011 and visited the building in the summer of 2013 with a small digital recorder. The building had been empty for nine years at the time.

She stood in room 11 with the recorder set on the iron bed frame, the door closed, the window closed, no other person in the building, and no source of sound she could account for by any ordinary means. The recorder captured forty-three seconds of what appears to be a slow rhythmic breathing coming from below the floor of room 11. The pattern on the recording is steady. It is even.

It is the breathing of a body at rest. Proud published a short article in a small Appalachian folklore magazine in 2016 and declined further site visits. Lynwood Catt bought the building in 2019. He sleeps in room 3.

Room 11 is locked. He has not gone in. He has told the man he bought the building from that he does not intend to go in. He has told his brother the same.

He has told no one else. The concrete Prace, Fell, and Roose poured in October of 1884 sits where they left it. The wall below has not yet been exposed.

Vessel wrote in her closing chapter that the wall was learning until the moment they sealed it, and that what it knows now, and whether it is still using what it knows, is not a question the surface record can answer.