In August 1881, in the New Mexico Territory, a sergeant with the Ninth Cavalry reined in his horse at the edge of a dry creek bed and stared at the ground. The tracks he had been following for three days had simply stopped. Six Apache warriors had entered a stretch of rocky ground, and none had emerged on the other side. The sergeant, a skilled tracker for eleven years, knew they had not vanished into thin air.
He also knew that 29 of his men at Fort Stanton had already been killed by a warrior band that had been eluding the United States Army for six consecutive weeks. He turned his horse and rode back, because there was nothing else he could do. For the next forty years, cavalry officers across the American Southwest would ask the same question that sergeant faced in the New Mexico heat: How could a man simply disappear? This is not a legend or a work of fiction.
It is a collection of skills so effective that the United States military, at the height of its industrial power, spent nearly half a century trying to understand them—and mostly failed. In 1886, 5,000 American soldiers were chasing 38 Chiricahua Apache through the mountains of northern Mexico. It took them more than a year to finish the job. These were men on both sides: tired, cold, hungry, sometimes cruel, sometimes brave.
This is an account of the specific techniques the Apache used to hold a land that the Spanish crown, the Mexican Republic, and the United States government all insisted they could not hold. The first thing to understand is that an Apache warrior was made, not born. He was shaped by his grandfather, his uncle, and his older cousins, in a process that began before dawn and ended only on the day he died. The first test was at the water.
Every morning before sunrise, a young Apache trainee had to walk to the nearest river or spring and bathe. If the water was frozen, he broke the ice. Then he filled his mouth with water, not to swallow, but to carry. He would run up the nearest hill over rocks, through brush, sometimes for miles.
The water in his mouth forced him to breathe through his nose alone. If he opened his mouth, he swallowed the water and failed the test. His uncle would be waiting at the top of the hill, asking him to spit the water out at his feet as proof. Nasal breathing in high-altitude desert air did something to the body over time that mouth breathing could not.
It slowed moisture loss from the lungs, changed the rhythm of oxygen and carbon dioxide in the blood, and taught the body to work under pressure without panic. A grown man who had done this since he was old enough to walk could travel for hours in conditions that might exhaust a wool-clad soldier within a single morning. General George Crook, who fought the Apache longer than any other officer of that era, wrote of a fighting force in perfect physical condition. He was not a man given to embellishment.
He had faced every kind of enemy in the American West. He wrote plainly that a warrior could cover seventy or eighty miles of open country on foot without stopping, without pausing to drink, and still be able to fight upon arrival. Eighty miles on foot in the summer in the Sonoran Desert. A cavalry lieutenant leading twenty men and twenty horses understood the stakes: horses needed water every eight hours before they began to die, and men needed it every two hours.
A soldier carried rations for seven days. The warrior he was chasing left camp with only a slice of dried mescal, a water gourd, and a rifle. He needed no road and no supply train. The second secret ended the pursuit of the sergeant described at the start.
When an Apache war party knew it was being followed, and it always knew, one of its tactics was to search for what Chiricahua trackers called cover ground: a long stretch of exposed stone, a dry lava flow, a rocky slope, or a wash where the last flood had stripped away the sand and left bare granite. The warriors stepped onto the stone one by one, spread out, and changed their way of walking. They did not walk in a single file. They fanned out in every direction.
When the stone ended, they did not regroup on soft ground. Each man found his own separate exit, and they did not reunite until miles later. For any pursuing patrol, the result was exactly what that sergeant saw: sets of tracks entering rocky territory, none emerging on the other side, as if the earth had opened and swallowed them. The army had a word for it.
They called it melting. Major cavalry reports from the 1870s and 1880s used the term constantly. The Apache would raid a ranch, be chased at full gallop for an hour, and then at some point they would simply melt. Even Apache scouts recruited by the army could decode these trails only sometimes, because they had grown up walking on those same rocks.
It took hours, sometimes days, enough time for the warriors to be fifty miles away and doing something entirely different. The third secret was disputed by some historians, but the accounts are consistent enough to support its accuracy. Apache warriors sometimes reversed the direction of their footprints. The simplest method was walking backward for a certain distance.
The tracks looked, at a quick glance, identical to tracks made walking in the opposite direction. A cavalry patrol arriving at a fresh trail could follow it in the wrong direction for miles before someone noticed that the toes pointed the way the men had come. A more complex version involved modifying the moccasin. The Apache moccasin was made of soft leather, sometimes with a hardened cowhide sole.
When necessary, a warrior could bind extra leather around the toe so the print left in sand was rounded at both ends. A tracker could not tell which end was the heel and which was the toe. Combined with walking on rock, this was the daily experience of any Apache scout attached to a cavalry battalion in the 1880s. The Apache did not consider any of this a trick.
To them, it was simply walking. The fourth secret was what made the army lose its composure. When pursuers got too close, the Apache sometimes killed their own horses. For a nineteenth-century soldier, this was nearly incomprehensible.
A cavalry horse in 1880 was worth more than a private’s entire annual pay, and losing one was considered a disaster. Deliberately shooting one was almost unbelievable. But the Apache warrior treated his horse as a tool. When the tool became a burden, he discarded it.
If pursuers cornered them at a mountain and the only escape was up a rock face no horse could climb, the warriors shot their horses at the base of the cliff. Sometimes they ate what they could first. Then they climbed the mountain on foot, using their hands, over terrain a cavalry officer would swear was impassable. Cavalry reports from the Sierra Madre campaigns describe this exact scene: officers approaching a canyon wall to find a small pile of dead horses below, and faint marks in the dust above, where men had climbed with their hands up rock that seemed to offer no foothold.
By the time anyone figured out how to follow, the warriors were over two ridges and gone. The fifth secret was a place, representing an entire way of thinking about the land. In the Dragoon Mountains of southern Arizona there is a set of hills known today as Cochise Stronghold, named for the Chiricahua leader who lived and led his people from that country for over a decade during the Apache Wars. To the Apache, a stronghold was not a fort or a wall.
It was complex, difficult terrain, full of hidden entrances and exits, where a small group that knew the land could hold out against a much larger force that did not. Such places shared certain features. They had water, but the water was hidden. They had vantage points that gave a view for thirty miles in every direction.
Their approaches were narrow enough that one warrior with a rifle could hold off a hundred men. And they always had exits, multiple exits. A place that could be entered by only one route was a trap, not a stronghold. In 1883, General Crook managed to enter the Sierra Madre stronghold using Apache scouts who had once lived there.
It was the first time an American force had succeeded in doing so. What he found astonished him: orchards, hidden lakes, and farmland where the tribes grew corn, a whole hidden mountain region, self-sufficient and invisible from below. The lesson was not merely that these places existed. The Apache had mapped the Southwest and northern Mexico in their heads with a level of detail no American map came close to matching, and they had chosen places no army could seize.
You cannot defeat a man in his own kitchen, and the Apache lived in a kitchen the size of Texas. The sixth secret was water, the central fact of that country. The Sonoran and Chihuahuan Deserts together cover an area larger than many European nations. In summer, temperatures on the plains surpass 110 degrees Fahrenheit.
The Apache knew where the water was. Not the obvious water, because every known spring, river crossing, and tank the army could find on a map was eventually fortified or patrolled. So the Apache kept a mental record of what Spanish records called ojitos, or “little eyes”: the small springs, some seasonal, some merely a damp clay patch at the base of rock, enough to fill a few men’s gourds but not enough for a herd. Some were tinajas, natural stone basins high in the mountains that collected and held rainwater for weeks or months.
A warrior approached these places at night, because approaching them in daylight could expose him to a lookout ten miles away. In the night, in a land where moon and stars gave enough light to travel, a small group could fill their gourds and move on before dawn without leaving a trace. There are cavalry accounts from West Texas campaigns of soldiers dying of thirst within a mile of water they did not know existed. Meanwhile, the Apache drank from those same wells and watched the soldiers below, knowing exactly how much water each soldier had left.
The seventh secret was small and specific: the Apache blackened their weapons. Cavalry rifles of the 1870s had brass fittings, polished stocks, and sometimes silver on the trigger guard. Under the desert sun, they glinted at a distance of a mile. The Apache, like the Comanche and Kiowa, understood this, but made it a special practice.
Any weapon that might reflect sunlight was rubbed with charcoal, plant ash, or a mixture of soot and grease. The barrel, the chamber, the sights, all became dull. A carefully blackened rifle held in a man’s arms on a rocky slope was effectively invisible from more than fifty yards. Even arrowheads and knife blades were treated so they would not reflect light.
A cavalry column, by contrast, flashed signals in every direction: sabers at their waists, buckles on their saddles, buttons on their coats. From an observation post three miles away, an Apache lookout could count the men in that column and learn its direction and speed simply from the glints of their equipment. The army eventually realized this and began issuing dark saddles and unpolished brass, but by then four decades of officers had ridden into ambushes unaware they were being watched on the way in. The eighth secret was smoke signals, though almost everything most people have seen in movies about them is wrong.
The Apache smoke signal system was not a language but a code, prearranged by the warriors of a specific band before they left camp on a specific mission. The meanings changed, deliberately, so the enemy could not learn them. Records collected by the army itself in the 1880s describe the general form. One long, steady column of smoke meant attention.
Two columns meant something like “all is well, we are here. ” Three columns or more meant danger, the approach of an enemy, or a call to gather the band. Quick puffs of smoke, made by throwing green grass or damp brush on embers, could layer messages in ways only the receiving party understood. In clear desert air, a well-made smoke column could be seen from thirty or forty miles away.
A network of watchers could pass a message across a hundred miles of land in under an hour. No telegraph in 1870 could carry a message from a Chiricahua warrior in southern Arizona to a Chiricahua camp in the Sierra Madre faster than smoke. The army’s messengers were men on horseback, who could be intercepted, killed, lost, or simply too slow. By the time a cavalry patrol learned of a raid, the raiders often already knew the patrol had received the news and were three days away in a different direction.
In the final years of the wars, the United States Army built its own visual signaling network to try to catch up. General Nelson Miles established a chain of heliograph stations across Arizona and New Mexico in 1886, specifically to capture Geronimo. According to the army’s own records, it was the most extensive heliograph network ever deployed on the North American continent. The Apache had been doing something nearly identical with their own mirrors for a generation before the army arrived.
The ninth secret was the mirror itself. An Apache warrior often carried a small mirror in a pouch hung on a cord around his neck. Sometimes they were commercial mirrors from Mexican markets, sometimes just a piece of polished metal. In bright sunlight, the flash of one of these mirrors could be seen at distances a modern person might find hard to believe: fifteen, twenty, even thirty miles under the right conditions.
The signals worked the same way as the smoke, prearranged and coded, with meanings changing from mission to mission. Two warriors on opposite sides of a mountain range could talk to each other in nearly real time, count the men in a cavalry patrol riding up a valley, report its direction, and tell the rest of the band to move, hold, or vanish. By the time General Miles was congratulating himself on his heliograph chain, the warriors he was chasing had been relaying messages by flash across the same territory since before he was assigned to the West. There is an account of Miles meeting Geronimo after the 1886 surrender and proudly showing the Chiricahua leader how his solar telegraph worked.
Geronimo reportedly appeared suitably impressed, and the army officer who recorded the scene believed it. There is a strong likelihood Geronimo was pretending; he had been flashing sun messages himself for thirty years. The tenth secret is less a technique than a numerical fact: Apache war parties were always small, deliberately so. In the Geronimo campaign of 1885–86, the total number of Chiricahua warriors involved was fewer than seventy men, dropping to thirty-eight by the end.
They faced more than 5,000 American soldiers, several thousand Mexican troops on the other side of the border, a small army of Apache scouts working for the Americans, and armed civilian militias in nearly every town in southern Arizona. During the pursuit of Victorio in 1879–80, the Ninth Cavalry lost 27 men in one campaign season against a force that numbered no more than 150 warriors and usually fewer. Nana, after Victorio’s death, conducted his 1881 raid across a thousand miles of Texas, New Mexico, and northern Mexico with a group that started at about forty men. He fought seven or eight battles, killed perhaps thirty to forty Americans and at least as many Mexicans, captured about 200 horses, and withdrew safely into Mexico.
The Apache understood something the army was slow to grasp: in rough terrain, twenty men move faster than two hundred. Twenty can hide where two hundred cannot. Twenty can live off the land where two hundred starve. A twenty-man war band can be led by one man who knows each member by name and does not need to pass orders through three layers of subordinates.
A column of 200 soldiers takes hours to break camp; a war party of twenty is on its horses in ten minutes. A column of 200 requires roads wide enough for a supply train; twenty warriors require no road at all. A column of 200 cannot ambush anything; twenty warriors are an ambush waiting to happen. The eleventh secret was the design of the Apache ambush.
It was not a random event but a plan governed by rules passed from generation to generation. The first rule was terrain. The warriors always chose ground where the enemy could not spread out: a narrow valley or rocky pass where a column had to string itself into a long line, where the rear could not easily help the front, and where no one could climb the sides. The second rule was patience.
The warriors sat motionless behind cover for hours, sometimes through the entire heat of the day, because intelligence indicated the patrol would pass that afternoon. The waiting itself was the task. The third rule was about the trigger. Unlike most European-trained soldiers, Apache warriors usually did not fire from a distance.
They waited until the enemy was very close, close enough for the first volley to decide the outcome. The old leader Mangas Coloradas reportedly told his warriors that a bullet fired at 300 yards was a warning shot, while a bullet fired at 30 yards was a decision. Some of these ambushes used bait, a small group of warriors appearing at a distance to lure the enemy into pursuit, straight into a position where the rest of the band was already waiting. One account from the 1860s describes a group of miners hearing gunfire on the horizon and rushing toward it, thinking another group was under attack and needed help.
They rode directly into a kill zone the Apache had prepared for them. The Apache considered this the correct way to fight. In a war between forty warriors and thousands of soldiers, one cannot trade shots evenly. Every shot had to count, and that was achieved by never firing until the outcome was already decided.
The twelfth secret was what happened after the ambush. Before the fight even began, the warriors decided how they would vanish the moment it was over. The system was called dispersal. When the fighting ended, the warriors did not withdraw together or in any recognizable formation.
They scattered in every direction, each man or pair of men taking a different course. They crossed ridges and uplands, waded streams and walked against the current, and disappeared under the rocks discussed earlier. They did not agree to meet a mile or two away, because that was still close enough for a determined pursuit to catch them. They agreed to meet at a specific place at a specific time, sometimes days later.
The place might be a spring twenty miles away; the time might be sunset three days later. Each warrior knew his destination and how to reach it without leaving a trace. When they arrived, the war band regrouped as if it had never been separated, ready for the next target. To the cavalry, this looked like magic.
Cavalry officer Britton Davis, who spent years chasing Chiricahua war bands, wrote in his memoirs that attempting to capture an Apache war party was like trying to catch a handful of quicksilver: you could see it and touch it, but the moment you closed your hand, it was somewhere else. The thirteenth secret was that the Apache moved at night. Not always, but when moving under pressure or trying to cover distances unseen, they moved after dark. In summer, when daytime temperatures made travel dangerously demanding even for men conditioned from childhood, and in winter, when moving in darkness meant moving unseen by watchers on the hills, the night was their time.
On terrain they knew, they could cover forty or fifty miles between sunset and sunrise and rest in the shadow of a mountain by the time a cavalry patrol two ridges away was breaking camp. The cavalry could barely move at night at all. A horse on unfamiliar ground in darkness was a hazard; with an exhausted, dehydrated, perhaps wounded rider, it was a disaster waiting to happen. Cavalry patrols slept from about 10:00 p.
m. until about 4:00 a. m. , six hours each night that belonged exclusively to the Apache.
Over a six-month campaign, that was more than a thousand hours when the army was not watching, more than forty full days of unbroken movement. The fourteenth secret was what the warriors ate. The army moved on its stomach, in Napoleon’s phrase, and a cavalry column needed a supply train, which needed roads or passable trails, guards, and water for the mules. Feeding 200 men and 400 horses in the middle of the Sonoran Desert was a nightmare that constrained every campaign the army conducted there.
The Apache solved this problem by not having it at all. A raiding warrior carried at most a small pouch of dried meat and a slice of mescal, roasted in a pit oven and pressed into cakes, enough for a week. When that ran out, he ate what the land offered. Mescal, a type of agave, grew wild across the entire Apache range; its heart, roasted in a pit for 24 hours, was sweet, nutritious, and lasted for months.
Mesquite beans could be ground into flour or eaten from the branch. There was prickly pear, pine nuts in the high country, yucca fruit, sotol, wild game, duck in the Rio Grande marshes, deer in the mountains, and rabbits, which a warrior could kill with a stone if necessary. A cavalry lieutenant looked at the same land and saw wasteland. An Apache warrior saw a pantry.
A cavalry column that ran out of supplies had to turn back. An Apache war band never ran out, because its supplies grew from the ground. There was no way to starve them because there was nothing to starve. There were no supply lines to cut because there were no supplies.
No siege could work because they never stayed in one place long enough for a siege to matter. The fifteenth secret was a woman named Lozen, from the last generation of the Warm Springs Apache. She was the sister of Victorio. She rode with him, and after his death she rode with Nana, and after that she rode with Geronimo.
She was not the only female warrior in Apache history, but her story is the most enduring, and her particular ability is one of the most remarkable records in the entire history of the Apache Wars. Men who fought beside her later told soldiers and anthropologists that Lozen possessed what the Apache call power. In her case, the specific power was finding the enemy. According to accounts gathered years later from Chiricahua men who saw her do it, Lozen would stand in an open place and raise her arms in a gesture the Apache described as a prayer, though the word does not precisely describe it.
She turned slowly in a full circle, and as she turned, her palms, according to the men who watched her, began to tingle and change color when they pointed toward the direction of an approaching enemy. The tingle grew stronger when she faced directly toward them and faded when she turned away. One may believe whatever one wishes about this; a scientific mind might call it coincidence, or acute observation clothed in ritual, or reading of subtle signs in the landscape. But the historical fact is that Victorio’s band trusted her guidance more than any other form of reconnaissance, and they held out for years against numbers that should have destroyed them in weeks.
Victorio reportedly called her his right arm. It is said, preserved in the memory of Chiricahua survivors, that she was worth more than a shield in battle and more than a scout on the trail. When she was with the band, they were never caught by surprise. When she was away on other errands, they were vulnerable.
In October 1880, Lozen was on another mission, accompanying a woman and her newborn child to the Mescalero reservation, and had not yet returned to Victorio’s camp. That was the month Victorio was trapped at Tres Castillos in Chihuahua by Mexican forces under Colonel Joaquin Terrazas. Sixty-two warriors died there, including Victorio himself. Nana survived with a small group and continued fighting.
Lozen returned to find her brother dead and her band destroyed. She rode with Nana on his 1881 raid and accompanied Geronimo to the end. When Geronimo surrendered in 1886, she was one of the warriors who surrendered with him. She was sent to Florida with the rest of the Chiricahua prisoners of war and died there of tuberculosis in a military barracks a few years later.
So how did the wars end? The answer is not heroic, and it is not comforting. The wars ended because the United States government eventually gave up trying to defeat the Apache in the field and began defeating them in every other way. It dismantled the reservations, cut off rations to families that had come in peacefully, forced formerly hostile tribes to live in the same small patch of land, arrested leaders during peace negotiations, and sent Apache scouts, men from other tribes, to hunt down those still resisting, because no one but Apache scouts could track Apache trails.
In September 1886, after Geronimo surrendered at Skeleton Canyon, Arizona, the government took not only the warriors who had fought but also the scouts who had helped capture them, and the peaceful bands who had never carried a rifle, and shipped them all east to prisons in Florida. They spent 27 years as prisoners of war. Some died in Florida, some in Alabama, some in Oklahoma, and when the survivors were finally allowed to make their own decisions in 1913, most had spent their entire adult lives behind wire. These fifteen techniques are still studied today, at West Point and Fort Benning, and by armies around the world that need to understand how a small force on familiar ground can withstand a much larger force from outside.
The Apache never wrote any of this down. They lived it. Their enemies wrote it down after the fact, in reports that read like the reports of men who had lost even when they claimed they had won. There is a saying the old ones used to describe the last Chiricahua bands still resisting in the Sierra Madre in the 1880s: the Apache were not fighting to win.
They were fighting to be left alone. Every technique described here served that single idea: leave us alone. They were not left alone.
But for forty years, in a land that was theirs before it was anyone else’s, they made the price of not leaving them alone very, very high.