In the spring of 1883, Owen Price was working as a mucker 160 feet down in the Iron King mine at Kingston, New Mexico, loading rock from the morning blast into an ore car, when his shovel struck something hard and smooth at the back of the pile. He took it at first for a timber the blast had knocked down, and then for a piece of rail from the ore track, but neither made sense. Behind the pile, the blast had opened a crack into dolomite no one had ever cut, and the crack had no loose rock in it at all. He brought his candle down to it.

What showed was the end of a dark metal bar about as thick as a man’s thigh, reaching out of the crack. When he put his hand on the rock beside it, the stone was warm, though the rest of the drift was cool. The Apache already knew it was there. They said it had been left behind.

Six years later, and 1,200 miles away, an old woman who had never been inside a mine would say something about that bar, and one thing she said would match a line in the superintendent’s letter, a line he had boxed and never explained. She never read the letter, and he never learned her name. That is where this story begins. Before anyone can understand what was found, it helps to understand the mountain above Kingston.

A man going down into the Iron King climbed timber ladders into cool, still air that smelled of powder smoke and wet stone. Every man carried his own candle on an iron spike. They drilled by hand—one man holding the steel, another swinging the hammer—then they blasted, and the next shift shoveled. A few feet a day through gray dolomite that had been lying there for more than 400 million years.

In all that rock, no one had ever found anything but rock. That spring, the Iron King gave up something else. Above the mine sat a camp barely a year old. Silver had been found in these hills in 1880, but the name Kingston appears on no paper before August of 1882.

Among the men who worked the Iron King in those years was a young miner named Edward Doheny, who would be married in Kingston in August of 1883, and who would one day be one of the richest oil men in America. In the next ten years, almost 6 million ounces of silver came out of 27 mines. The silver of the Iron King lay along a fault that runs north and south, so its drift ran north into the mountain. North is a direction this story will keep returning to.

In 1884, the camp built a bank of stone on its main street. The camp also had a bell that called men to fires and told them when the mail had come. Both will return. The first record of what happened is a letter, and it begins with the man who wrote it.

His name was Amos Hartley. He was 41, a Missouri man who had run mines in Colorado before the owners in St. Louis sent him to run the Iron King. He had been hired to waste nothing—not powder, not candles, not words.

He trusted a tape more than he trusted any man in the camp. On the morning of April 5, 1883, he went down with a tape, a compass, and a thermometer. That evening, he wrote five pages to the owners. He copied every page into the company’s letter book.

The letter holds three lines that do not belong. The first is about the crack. Its walls were bare, with no drill hole in them, no pick mark, no soot, and not one broken stone on its floor. A crack in rock is never clean.

This one was. The second line is about the far end of the bar. It went two feet down into solid dolomite, and around it the rock was not split, not crushed, not even chipped. Hartley ran the blade of his knife around the joint between metal and stone, and he could not find a gap thicker than a sheet of paper.

Nothing can be driven into dolomite without breaking it. The third line is the shortest. The rock of the drift: 64°. The rock around the buried end: 81.

In the margin, one word: warm. Years later, someone who had never read that letter would use that same word about that same rock. Then Hartley drew the crack. It ran north from the face of the drift for 19 feet, a yard wide and 9 feet high, and it closed at its far end in solid rock.

Along its floor lay the tool. Its grooved end lay toward the drift, and its head stood upright at the far end. One arm buried in the floor, the other curving up toward the roof. At the bottom of the last page, Hartley laid his compass along the handle and took a bearing: north 4° west.

He drew a box around those four words. He did not write why. Sixty-nine years later, another man would come down to that crack with a thermometer. The second record is not something anyone wrote.

It is the tool itself, cut out of the rock over four days and measured on the fifth. The handle came free easily, but the head did not. Two men with hand steels drilled a ring of holes around the buried end and broke the stone away a chip at a time, because Hartley would not let them blast. On the evening of April 8, the head came loose, and for the first time anyone could see the whole of it.

A haft 14 feet long and eight-sided, more than twice the height of the men standing around it. A head 8 feet from end to end, two curved arms like the head of every miner’s pick in the camp. A pick. But neither end of that head had a point.

Both arms ended square and flat, each face about the size of a man’s palm, like the face of a hammer. The buried end had no mark on it at all—no scratch, no burr, no flattening, nothing to show it had ever been driven into anything. Then there was the handle. The last three feet of the haft were worn smooth and bright, and around them ran four grooves, each about three inches wide.

Hartley measured from the first groove to the last: 13 inches. On the hand of a grown man, the same four fingers cover about three. Something had held it. On the 9th, twelve men with ropes and a timber sled brought it up the shaft and laid it on the ore scale: 2,300 pounds, about what iron that size should weigh.

Then the blacksmith touched a magnet to the haft, and it fell away. He tried the head, then the grooves. The magnet would not hold anywhere. It was not iron.

Hartley sent every number to St. Louis. That May, the owners wrote back with their explanation. It was short and sensible.

The owners’ reply called the bar an ore crusher from some old Spanish working, left in a natural fissure and forgotten. The explanation deserves a fair hearing. The rock around Kingston is cut by old faults, and water moving through broken rock for thousands of years can open a fissure no one ever dug. And men were digging for metal in New Mexico long before Kingston had a name.

Spanish miners were taking copper out of Santa Rita, about twenty miles to the southwest, by 1800. A heavy bar left behind in a fissure would be one of the most ordinary things a miner could find. But the reply leaves out three things Hartley had already written down. It does not explain how a bar 14 feet long came to lie in a crack that was closed at both ends—the only way in was the opening the blast had made.

It does not explain how a flat end went two feet into dolomite without leaving a single crack in the stone. And it does not explain why a crushing bar would carry a handle worn into four finger grooves that span 13 inches. Hartley filed the reply. No one asked a second question.

What the owners could not know was that the second question would be answered six years later, 1,200 miles away, by someone who had never seen Kingston. The third record was not written in Kingston. It was written in Alabama, in a camp of prisoners of war. The Black Range had belonged to the Warm Springs Apache, the Chihenne, and the heart of their homeland was a warm spring north of the range, a place the Spanish had called Ojo Caliente.

In 1876, 916 people lived there on a reservation. In 1877, it was taken away. Some 450 people were moved west to San Carlos in Arizona. That September, Victorio and more than 300 others walked away, back toward home.

War followed, and the army fought him on Percha Creek, a few miles below the place where Kingston would stand. In 1886, the government sent the Warm Springs people east as prisoners with the rest of the Chiricahua. First to Florida, then to Mount Vernon Barracks, north of Mobile, Alabama. Many died there of fevers and lung sickness.

Their interpreter was George Ratten, who stayed with them for the rest of his life. And in 1889, one of the prisoners came to him about a mine she had never seen. She came to his door in June. Born near Ojo Caliente, she had walked to San Carlos in 1877.

She had never seen a shaft, a ladder, or a candle spike. She had heard that miners were digging into the mountain above the Percha, and she asked him to write to the man in charge of the Iron King. Ratten put her words into English and mailed them. The letter reached Kingston that July, and Hartley copied it into the same letter book.

She said there was a digging tool in the mountain above the Percha, older than the miners. She said its handle had a place for each finger, and that our hands were too small. She said its point was not sharp because it was not made to break the rock. It was made to open it.

She said that where it bit, the rock was warm and would stay warm. She said it had been left behind, and that it should be put back where it was. Hartley had asked her one more thing: which way it faced. She did not point.

She said five words: “It faces the warm water. ”

Then he asked who had left it. She said it had been left by the ones who worked the mountain before her people came. When he asked who they were, she said nothing more.

The knowledge belonged to the Chihenne. It was not something to be given to an outsider. A name could be misfiled in a county ledger. It could not be misfiled in the mouth of a living person.

Hartley never answered. But the direction she gave was already in his letter book, inside a box. These two records were never meant to be read together. One was written by a superintendent, the other by an interpreter for a woman who could not write English.

Set them side by side, and they line up. Hartley wrote: four grooves, 13 inches from first to last. She said: its handle had a place for each finger, and our hands were too small. Hartley wrote that both ends of the head were square and flat, and that one was buried two feet in stone that was not cracked.

She said: its point was not sharp because it was not made to break the rock. Hartley wrote: 81 degrees, and in the margin, warm. She said: where it bit, the rock was warm and would stay warm. But the last pair is the one with a number in it.

Hartley laid his compass along the haft and wrote: north, 4 degrees west. She said: “It faces the warm water. ”

In 1883, a compass at Kingston pointed about 12 and a half degrees east of true north. Correct Hartley’s bearing for that, and the haft runs 8 degrees east of north.

Draw that line on a map. It crosses 46 miles of mountains and arrives at a warm spring: Ojo Caliente, the heart of the Warm Springs homeland. The head pointed home. One more line in Hartley’s letter book comes from the night of the 8th.

Price went back down alone and sat in the empty crack with his candle. He laid his palm flat inside the hollow the buried end had left in the stone and kept it there. When he came up, he told Hartley it felt like a hearth an hour after the fire has gone out. Hartley wrote it down in quotation marks.

None of this says what the pick was. It says only that two people who never met, 1,200 miles apart, describe the same tool and turned it toward the same spring. And in 1952, someone went back down to measure that hollow again. The Iron King went on for ten more years.

In 1893, the price of silver collapsed. The mines around Kingston closed one after another, and the drift where Price had found the pick filled slowly with water. In 1952, the Iron King was opened again, this time for manganese. An engineer named Ray Dunning, pumping out the old workings, found the crack Hartley had drawn, with the hollow still in the floor at its far end.

The hollow was full of water. Dunning lowered a thermometer into it: 104°. In 1883, the dry rock of that hollow had read 81. The water standing in the drift outside the crack, Dunning wrote, was cold to the hand.

He wrote the two numbers side by side and did not write what had changed. His report has one more line. The walls of the hollow were no longer bare. A thin white crust had grown on them, the kind that forms where hot water rises through stone.

It had grown nowhere else in the crack. A few weeks later, his crew walled the crack off. Why the company built that wall, the last record will show. But the Iron King was not the first place in these mountains to be worked and then left behind.

The Mimbres River rises on the western slopes of the Black Range, on the far side of the mountain from Kingston. Along it, from about the year 1000 until about 1130, stood villages of stone rooms built by the people archaeologists call the Mimbres. From 1924 to 1927, a Silver City couple, Harriet and Burt Cosgrove, dug one of those villages for Harvard’s Peabody Museum. Swartz ruin lies about 17 miles southwest of Kingston.

Harriet drew more than 700 of its painted bowls by hand. Then the villages emptied, and no one has settled why. Forty miles northwest of Kingston, in the Gila, five caves in a canyon wall hold 46 rooms. Their timbers were cut between 1276 and 1287.

By the early 1300s, the people who built them were gone. In 1878, a prospector named Henry Ailman rode into that canyon and found the rooms empty. Within a few miles of those rooms, hot water still comes up out of the valley floor along the faults at about 100°. Long before Kingston, this country had been lived in, worked, and left more than once.

And the warm ground was already there. None of it says who made the pick. But the heaviest record in this file is not a ruin or a letter. It is a page of numbers kept by a pump crew.

The last record is not a letter. It is a trail, and it ends in two pieces of paper. It begins in the summer of 1883. Santa Fe was preparing a great fair, the Tertio Millennial Exposition, which opened on July 2, and the owners wanted the pick in it.

In the last days of June, it left the Iron King on a mule-drawn freight wagon down the canyon of the Percha to the railroad at Nutt, and north to Santa Fe. The exposition closed on August 3. The last line about it in Hartley’s book is a freight receipt dated August 11, 1883: one crated implement, 2,300 pounds, consigned to the owners in St. Louis.

Beneath it, in a different ink, someone later wrote two words: “Not received. ”

The second paper belongs to Dunning’s wall. The order that built it gives one reason: water. His pump log shows why.

For eleven days, his crew pumped the hollow dry every evening, and every morning it was full again at 104°. The hollow held about 40 gallons. The crack around it stayed dry, and its far end was closed in solid rock. Forty gallons of hot water a night were coming up out of stone that had no way in.

No one asked where. Nobody hid anything. The records were not suppressed, simply not followed up. The mine closed, the owners moved on, and the papers went wherever the papers of a closed mine go.

There is a reading of this case in which everything resolves into comfortable categories. In that version, the crack is natural—the range is built from the remains of ancient volcanoes and cut by old faults, and water and pressure can open clean passages that look as if they were made. The pick is old work—Spanish miners were digging in southern New Mexico by 1800, and a heavy bar left in a fissure could be forgotten for a lifetime. The magnet proves less than it seems—steel made with manganese will not hold a magnet, and the Iron King itself later gave up manganese ore.

The grooves could be wear from ropes or chains, left by whoever dragged the bar into the fissure. Even the one number that changed has an answer: 30 miles east, the town once called Hot Springs, now Truth or Consequences, sits on water that comes up out of the ground at a little over 100°, and when an old mine floods, hot water can rise along the same faults. The strongest objection is the last one. The old woman may simply have heard.

Chiricahua men served the army as scouts in those years. Freighters carried news along every road, and six years is long enough for word to cross a continent. Skeptics reasonably point out every one of these things, and every one of them is fair. But that version has to explain each record by itself.

It cannot explain them together. A flat end in uncracked stone, a grip spaced for fingers far wider than any man’s, a metal no magnet would hold, water rising out of closed rock, and a woman in Alabama naming the direction of a line she could not see. That is not a weakness in the record. It is the record.

So what is in those hills now? Kingston never quite emptied. About 50 people live there today along the same creek. The bank the camp built of stone in 1884, the Percha Bank, still stands, and it is a museum now.

The town still keeps an old fire bell, the kind that told a camp the mail had come. In July of 1889, it would have rung for Ratten’s letter. But the last place in this file that still holds an answer is not on the main street. The Iron King gave up manganese until 1959, and was then left to the water.

Somewhere in the hill north of town, behind Dunning’s wall, the hollow should still be in the rock. Forty-six miles north, the warm spring the Chihenne called the heart of their homeland still runs. The Warm Springs people were held as prisoners for 27 years and set free in 1913. Their descendants in the Fort Sill Apache tribe never stopped asking to come home, and in 2011 they won back a small piece of New Mexico—about 30 acres near Deming.

If the pick ever left New Mexico, no one has seen it since. The mines are real. The silver, the fair at Santa Fe, and the railroad at Nutt are real. The removal of 1877, the years in Florida and Alabama, and the long road home are real.

The unanswered questions begin only after that point. In the spring of 1883, 160 feet down, Owen Price laid his hand on warm stone and thought he had found something another miner had lost. 81° in 1883, 104 in 1952—no one has measured it since.

If it was never made to break the rock, what was it made to open?