In the spring of 1931, in Black Canyon on the line between Nevada and Arizona, America started cutting four tunnels, each fifty-six feet across, to move the whole Colorado River around the spot where Hoover Dam was going to stand. A construction photograph from January 1932 looks out through the mouth of one of them, Tunnel Number Two. The opening is a round hole in dark volcanic rock, and the men standing in it are so small you have to hunt for them. You cannot pour a dam into a moving river.
So before any concrete, somebody had to take a river that flooded every spring and push it through a mountain. They had it inside the rock a year and a half after the first cut. This account is pieced together from the record that the Bureau of Reclamation’s own construction engineer published while the tunnels were being cut, and from the recorded memories of the people who lived at the bottom of that canyon. Two of those people carry the story.
Irma Godbay was a miner’s wife from Silverton, Colorado, living with her small children in two tents on the riverbank that first summer while her husband worked the canyon with a shovel. Marian Allen was a young man who walked into one of those tunnels in December 1931 without ever having seen the inside of a tunnel in his life. By the end, you’ll know how four round holes carried an entire river around a dry piece of canyon floor long enough for a dam to rise in the riverbed, what that first summer cost the people camped beside them, and why two of those holes still take the Colorado when it floods today. On the 11th of March 1931, the general superintendent for Six Companies, the group of contractors that had won the job, put his first men to work.
His name was Frank Crowe, and his first orders were plain ones: lay out a camp and cut a road down into Black Canyon. The man representing the government in that canyon was the Bureau’s construction engineer, Walker Young. Young later spelled out exactly where they were standing: 270 miles downstream from the place where the Bright Angel Trail reaches the river in the Grand Canyon. Needles, California, was another 100 miles on down the river.
What the two men had in front of them was a river squeezed between sheer walls of volcanic rock, stained almost black by centuries of weather. At low water, the canyon floor was only about 300 feet wide from wall to wall. The water ran brown and heavy. When the Bureau later pumped it up to the new government town on the hill, it had to be desilted, softened, and purified before anybody could drink it.
The Colorado kept a calendar. Every May and June, snow melting in the mountains far upstream sent a flood down the canyon, commonly more than 71,000 cubic feet of water a second. Some years it passed 100,000. In late summer, thunderstorms could dump a flash flood out of the side washes with almost no warning.
Only from fall into early spring did the river drop and settle. A generation earlier, farther downstream, the same river had broken out of its channel and poured into the California desert, and it had taken months of fighting to force it back. The biggest flow anyone had ever recorded through Black Canyon was 200,000 cubic feet a second, and whatever the men built had to survive a day like that. The contract gave the tunnels until the 1st of October 1933, with a fine for every day they ran late.
But the paper deadline wasn’t the real one. You can only turn a river when it’s low, in late fall or winter. So in practice, the tunnels had to be finished and ready for the low water at the end of 1932, or the whole job sat and waited out another spring flood. There was a second reason the river had to go, and you cannot see it from the rim.
The Bureau’s diamond drills had gone down through the riverbed looking for solid rock. In the middle of the channel, they did not hit bedrock until roughly 110 to 120 feet below the surface of the low water. Everything above that was river silt, sand, and loose rock. By Young’s own figures, the dam would stand more than 700 feet from its foundation to its crest, and it would be 650 feet thick at the bottom.
A dam that size cannot sit on gravel. It has to stand on rock. So the canyon floor had to be dry all the way down, a pit as deep as a ten-story building is tall, dug out in the exact spot where the Colorado was running. Every other piece of the job depended on those tunnels.
How do you cut a hole 56 feet across through solid rock? The answer in the Bureau’s plan is that you don’t cut it all at once. Picture a five-story building laid down flat inside the canyon wall, running back into the rock for about three-quarters of a mile. That’s the size of one tunnel.
The first cut was a small passage, 12 feet by 12 feet, driven along what would become the very top of the tunnel, like a hallway running along the roofline of that building. Young gave two reasons for starting with a hallway at the top. The small bore let air move through the rock ahead of the heavy blasting, and it let the engineers see what the rock was really like, the cracks and the soft seams, before anybody opened up the full face. If the rock had a bad seam in it, the engineers wanted to find it in a passage only 12 feet wide.
Once that hallway was through, crews blasted the rock away on both sides of it, the wings, until the top of the tunnel was open to its full width. Then they brought the whole upper section down to a ledge 15 feet above the floor. That left a rock face 41 feet high and 56 feet wide. Last came the bottom 15 feet, the part engineers call the invert, broken out like the final step of a staircase.
They dug to 56 feet for a reason. Three feet of concrete would go onto every wall, leaving a finished tube 50 feet across. And the four finished tubes together were sized to carry more than 200,000 cubic feet of water a second, which is over 1. 5 million gallons.
The water tank the Bureau built on the hill above its new town held 2,000,000 gallons. The four tunnels were built to swallow that entire tank in less than a second and a half. Two tunnels went into each canyon wall: an inner pair close to the river, and an outer pair set farther back in the rock. Each one began upstream of the dam site and came out downstream of it, a bypass the river would have to take whether it wanted to or not.
Laid end to end, the four would run nearly 16,000 feet, about three miles of tunnel. And the tunnels were only half of the trick. The plan only works if the river has nowhere else to go. Once the tunnels were open, the builders would dump a barrier of earth and rock across the old channel above the dam site, and a second one below it, so the river couldn’t back up into the work from downstream.
Engineers call those temporary dams coffer dams. Think of a highway crew closing a stretch of road for repairs: the tunnels were the detour, and the coffer dams were the barricades at each end of the closed stretch. Between them, the riverbed could be pumped out and dug down to rock. All of that was a plan on paper.
In the spring of 1931, the machine that would do the big cutting hadn’t been built yet. And when it finally showed up, it would be riding on the back of a truck. Before any tunnel was more than a scratch in the canyon wall, people started showing up at the river looking for work. By June of 1931, somewhere between 600 and 1,400 people were camped on the flats where Hemenway Wash runs down into the Colorado at the mouth of Black Canyon.
It was a spread of tents and crude shacks with no sewer and no electricity, and everybody called it Ragtown. A deputy marshal tried to bring some order to the place by laying out streets and a church, and for a while it had a more respectable name, Williamsville. The people who lived there kept calling it Ragtown. Irma’s family lived in two of those tents right down on the canyon floor.
She’d been born in the mining town of Silverton, Colorado, in 1905, and at eighteen she married Tom, a miner who took whatever work there was. By 1931, with the country deep in the Depression, they were one more family that had come to the river looking for work on the dam. Tom found a job fast. He went to work on the 4th of July with nothing but a shovel, on a crew building the service roads the dam builders were going to need.
Down on the flats, the store belonged to a riverman named Merl Emory, who’d been running a ferry across the Colorado since before the dam work began. Supplies came in faster than anyone could check the paperwork, so the store ran on the honor system. Customers were told to pay whatever they used to pay for the same thing back home. People laid their money on the counter and took what they needed off the shelf.
Irma had hauled an ironing board all the way from Colorado, along with an electric iron, and there wasn’t an outlet within miles. So she laid the board across two empty powder boxes, the wooden boxes the blasting explosives came in, and made herself a bench. In Ragtown, she said later, a new powder box meant a new piece of furniture. Heat got into everything.
She once kept a slab of bacon in a corner of the tent with a few potatoes underneath it. The heat melted the grease right out of the bacon and down onto the potatoes, and the potatoes fermented. The family had to dig a hole and bury them. Emory had one more job, and it tied Ragtown straight to the tunnels.
He held the contract to carry the work crews down the river by boat to the tunnels. The tunnels were being dug from both ends at once, and Irma remembered eight crews working around the clock, twenty-four crews a day, every one of them riding the river down to the job and back. Every morning, the men cutting the river’s detour rode to work on the river itself. There was only one place to get clean, and that was the river.
People waded into the Colorado in whatever they had on and washed as best they could with the current pulling at their legs. One day Irma was in the water with her baby daughter, holding the little girl by one leg to keep her close. Irma had never learned to swim. Then the current took hold of the child and started carrying her off down the river.
Irma yelled for Tom. He swam like a fish and had the baby out of the current almost before she’d finished screaming. Fifty-four years later, when Irma told that story at a panel the University of Nevada recorded for the 50th anniversary of the dam, she stopped to point out that the same daughter was sitting right there in the audience. The river was the problem on every front of the first summer’s work, and there were three of them.
The first front was getting anything heavy down to the water. Tom’s shovel crew was part of it. Six Companies had to carve roads down the canyon walls, and one of the photographs in Young’s article shows what that meant: a road called the Black Canyon Highway running along a shelf blasted into the side of a cliff, with a retaining wall holding up its outer edge. Up on the rim, the job needed things the desert had never had.
Young listed them. A power line was strung 222 miles from San Bernardino, California, to a substation on the rim of Black Canyon. It carried 88,000 volts, and it would run the electric locomotives and shovels down in the tunnels. The Union Pacific built a railroad from its main line south of Las Vegas to the new government town, with a yard there big enough for 300 cars, and a second government line ran from there out to the edge of the canyon.
The state of Nevada paved a highway 23 miles long from Las Vegas to the town, and the government carried the road on from there to the canyon rim. All of that got men and freight to the top of the canyon. Getting them from the rim to the river was the hard part. Then there was Arizona.
Two of the four tunnels had to be driven into the Arizona wall, so everything they needed, the men, the trucks, the powder, and later the machines, had to cross the Colorado first. The contractors strung suspension bridges across the canyon so loaded trucks could drive over the river. Young’s article counts two of them. And when the big drilling machines finally had to go across, they rode a barge that was held against the current by cables.
Every one of those crossings was a bet on the river’s mood. In late summer, a thunderstorm miles away could send a flash flood down a side wash with almost no warning, and the people camped on the flats knew it. The panel that recorded Irma’s memories listed strong winds, cloud bursts, and flooding right alongside the heat as the canyon’s natural conditions. Once the bridges were up, the trucks could reach both walls.
The harder question was what they’d find once the crews went inside the rock. That was the second front. In May of 1931, crews started driving small tunnels called adits, straight into the canyon walls, each about 10 feet by 12. Every adit was aimed at one precise spot deep in the rock: the top of one of the big tunnel lines.
The Bureau’s history dates the first of them, on the Arizona side, to the 12th of May. On the Nevada side, crews also went straight in at the downstream ends, the lower portals, where two of the tunnels would one day come out of the wall. Where an adit reached its tunnel line, the crew turned and drove the roofline hallway in both directions at once, upstream and downstream. Young called it standard tunneling, and it was.
Small mucking machines scooped up the broken rock, and electric locomotives hauled it out on rails, the same way hard rock miners had been working for years. That’s why Irma remembered crews working the tunnels from both ends. A bore attacked from several faces at once goes through much faster, but only if the surveyors are good enough that the pieces actually meet in the dark. These surveyors had to aim small tunnels at points they couldn’t see inside a canyon wall, and hit them.
By midsummer, the small bores were pushing through the rock on both sides of the river. But every foot they advanced made the third problem bigger, because every foot came out as broken rock that had to go somewhere. The third front was the river’s edge. The tunnel mouths sat low on the canyon walls, and the walls above them had to be cleared of loose rock before anyone worked underneath.
Scaling crews pried it loose. One of the photographs in Young’s article shows the outlet of the Nevada inner tunnel with a scaling rig standing beside it, a frame men worked from to pull loose rock off the tunnel roof on the high canyon walls. The men who did that job, the high scalers, hung on ropes with jackhammers and bars. Six Companies had its own river camp about a mile upstream from the Nevada tunnels, and a photograph dated the 9th of August that year shows the signs its insurance department put up there.
Marian Allen, the young man who would follow into the tunnels later, remembered what signs like that kept telling the men: Safety first. Don’t go into the tunnel without your hard hat. Falling rock was the constant danger, and the men came up with their own answer. Some of them dipped cloth hats in tar and let them harden.
When men wearing those homemade hats were hit hard enough to break their jaws and still walked away without skull injuries, the company ordered thousands of commercially made ones. The crews called them hard-boiled hats. Today everybody just calls them hard hats. Allen told a story about those hats at that same panel.
A cement finisher from the South came into the tunnels for his first day, and rocks kept dropping out of the roof onto the work around him. So he stopped at the tool shed on his way in and asked for a second hard hat. When they asked him why, he said he spent all day bent over, and he wanted one for the other end. Down at the water line, the lower ends of the tunnels were so close to the river that the contractor built curved concrete walls across the tunnel mouths, small arch-shaped dams to keep the Colorado from pouring into the work at high water.
At the upstream ends, where the river would eventually go in, the tunnel mouths had to stay walled off from the water too, right up until the day the engineers were ready to turn it. Then there was the rock. Young’s account spells out a rule the Bureau laid down, and it tells you how seriously they took the river. Every load of broken rock hauled out of the tunnels had to be dumped above the river’s high-water line, in a place where the next flood could not wash it back into the channel.
That meant hauling the rock up into side canyons and building roads to get it there. Young wrote that it took considerable ingenuity to find dumping grounds the Bureau would approve. If the Colorado picked that rock back up in a spring flood, it could carry it right back into the channel they were trying to empty. Measured against the river’s calendar, midsummer of 1931 looked like this.
The low water they needed would come at the end of the next year. The pilot bores were only the hallways. The big cutting hadn’t begun, and the machine that would do it still hadn’t been built. So by midsummer, the roads were going in, the bridges were up, the portals were walled off, and the rock was going uphill away from the water.
The one thing nobody could wall off was the heat. It was the Depression. Men had walked miles to reach that river, some of them going without food on the way, and every one of them wanted a place on those crews. The summer of 1931 was a hard one, even for Black Canyon.
The daytime high that summer averaged 119. 9 degrees down on the flats. Some afternoons reached 125. Inside the tunnel workings, it got as hot as 140, and the air was thick with truck exhaust and powder smoke.
Men were dropping on the job. On Tom’s road crew, so many were passing out from heat stroke that the hours were changed. The day shift started at 4:00 in the morning and quit at noon, and nobody worked from noon to 4:00, the worst of the day. A second crew came on at 4:00 in the afternoon and worked until midnight under searchlights.
Part of the trouble was something nobody down there understood yet. Men were sweating the salt out of their bodies all day long, and nobody knew to replace it. They’d work and sweat, and then they’d fall over where they stood. In the tents, there was no shade except what you rigged yourself.
Irma pinned blankets to a clothesline to throw a little shadow over her children while she cooked on an open fire. The sun went after everybody else too. Irma’s face got so burned by the sun, the wind, and the campfire all at once that the skin cracked and bled when she changed expression. At first she thought she’d caught something from the river, and she dabbed her face with straight Listerine, which only dried it out and burned it worse.
The doctor at the Six Companies camp told her he couldn’t even look at a woman or a child; he was there to give the men their physicals. So she had to go to Las Vegas. That summer there was still no highway, just a track up and down through the arroyos and the dust, and the trip needed its own preparation. Cars in those days rolled on wooden-spoked wheels, and in that dry heat the spokes shrank until they rattled loose.
So before you could drive to Las Vegas, you parked your car in the river and let the wheels soak for about three days, moving it a little now and then so it wouldn’t float off. The doctor in Las Vegas took one look at her and said it was the worst case of desert sunburn and windburn he’d ever seen. He gave her a zinc salve, and for years afterward she kept buying it and giving it away to men who worked without shirts until their backs blistered, and to parents whose babies ran around bare and came down with sun blisters that got infected. The worst of it came in one stretch of 32 days, from the 25th of June to the 26th of July.
In that month, 16 people died of heat prostration, workers and people living along the river alike, while the four tunnels they’d come to cut were still small holes in the rock. PBS, in its history of the dam, calls the diversion tunnels the most dangerous phase of the whole project. And that summer came before the crews had even started the big blasting. The last day of that stretch came right to Irma’s tent.
Three women died around her that day down on the canyon floor, and half a century later she could still name the date. After that, Irma told Tom they had to move somewhere she could get the children to a doctor. They moved to a place called Cowboy Bill’s Camp out on Bonanza Road in Las Vegas. To collect the wages he was owed, Tom had to quit his road job, and when he came back, he found work with another contractor building the railroad out to the gravel pit on the Arizona side of the river.
Boulder City was supposed to be the answer to Ragtown. The Bureau had started building its government town on the desert above the canyon the previous December, with its water pumped nearly 7 miles up from the river in four lifts. But that first summer, the town wasn’t ready yet for the people already living at the river. Then the pay changed.
On the 7th of August, the company moved a group of tunnel workers to lower-paying jobs. The tunnel crews walked off the job within hours, and the walkout spread. Their demands read like a list of everything that summer had been missing: clean drinking water, flush toilets, ice water on the job, and for the company to follow the mining laws of Nevada and Arizona. Every item on that list comes back to the same thing: staying alive at 120 degrees.
The walkout came 12 days after the last of those heat deaths. On the 9th of August, the answer came back. The demands were refused. All work was stopped, and the entire workforce was laid off, with everyone told to be off the job by 5:00 that afternoon.
Tom happened to be in town that day, bringing Irma his paycheck. A gate went up across the road out by Railroad Pass. Nobody outside could get in, and nobody inside could get out. By that evening, Black Canyon had gone quiet.
The pilot bores sat empty in both walls. The men were on the wrong side of a locked gate, and the Colorado was running through the canyon exactly the way it always had, with the low water of 1932 a little over a year away. It didn’t stay quiet for long. Within days, the men voted to go back to work.
The pay cut stayed, but the company promised it would be the last one. Afterward, it strung more lights through the work, carried water out to more of the crews, and pushed the dormitories in Boulder City to finish faster. Tom didn’t get straight back in. He spent almost two months outside that gate, riding out every morning with a crew of electricians who had passes, and asking at the gate for a job.
Through the rest of that summer, the pilot bores kept going. A photograph dated the 26th of August shows one of them 12 feet square, and its caption says the hole would later be enlarged to 56 feet. Another from the 10th of September shows drillers standing inside the roofline bore of Tunnel Number Three in the Arizona wall. By the fall, the pilot bores had done their work, and the big cutting could start.
On the 3rd of October 1931, crews began taking the tunnels out to full size, and the machine they did it with was built on a ten-ton truck. It was called a drilling jumbo. Picture a truck with a steel framework built up on its bed, several decks high, with horizontal steel bars running across the frame and air drills clamped along the bars, so a whole crew could drill a tunnel face at the same time. Each jumbo carried somewhere between 24 and 30 drills, and eight of them were built for this job.
They’re usually credited to Woody Williams, the assistant superintendent over the tunnel crews. Young’s account walks through a round step by step. First, government surveyors painted the outline of the 56-foot circle on the rock face in white, so everybody could see exactly where the tunnel stopped and the mountain began. Then the jumbo backed up to one half of the face.
The crew jacked it up and blocked it solid, hooked up the water and air lines, and started drilling. At an ordinary setup, they drilled 48 holes, some of them 20 feet deep, in less than two hours. Then they rolled the jumbo across to the other half of the face and did it again. Men were already loading the finished holes with dynamite while the drilling went on beside them.
When the face was loaded, everybody pulled out, and the round was fired in 15 delays, one blast after another. The first shots broke down the wings. The rest followed from the top and center, working down and outward, until the last one cleared the floor. Each blast used less than 2,000 pounds of explosive and broke the rock about 16 feet deep.
Across the whole job, the Bureau figured it took a ton of dynamite for every 14 feet of tunnel. A ton every 14 feet for three miles. Then the fans and blowers pulled the smoke out, and the clearing started. A small bulldozer came out of the old pilot bore and shoved the broken rock off the upper ledge onto the floor below.
A bigger tractor cleared a path. An electric shovel moved up to the face, and the trucks lined up behind it. Marian Allen walked into the middle of that in December. Crowe had known him since about 1912, and Allen’s father had already come to the canyon that spring.
Even so, he was warned about the new employment agent, a hard man named McAdams. You could get on the job, but once you were there, you looked after yourself and did your work, because McAdams never gave anybody a second chance. Allen went to work on the 13th of December 1931, and they sent him straight into the tunnels. He’d never been inside one.
He remembered the tunnel looking about as big as the building he was sitting in when he told the story. Three days later, the timekeeper quit. The walker on that tunnel, meaning the walking boss over its shifts, was a man everybody called Red McCabe. He had worked on a Bureau job with Allen’s father years before, so he gave Allen the job.
It paid the same, about 65 cents an hour as he remembered it, but it was easier on the body. Part of it, he found out, was seeing the powder loaded onto the trucks, about two tons at a time. For a while they loaded the primers, the charges that set off the dynamite, onto the same trucks. Allen said he didn’t know then that it was against the law.
After a short time, somebody put a stop to it. He remembered the smoke too. Men were anxious to get back to the face after a blast, and if they went in before the blowers had done their work, the powder fumes gave them headaches. Behind the drilling jumbos came another machine, the trimmimg jumbo.
A pair of steel trusses shaped to the curve of a 50-foot circle, set about 9 feet apart on car wheels and rails, with timber platforms between them. Men rode it along the tunnel and knocked off every point of rock that stuck inside the line, so the concrete would have a true circle to follow. In January of 1932, the big bores were being driven from eight headings at once. In the single week that ended on the 23rd of January, the crews dug nearly 1,700 feet of tunnel.
That’s about a third of a mile of full-size tunnel in seven days, and it moved around 116,000 cubic yards of rock. Off the job, the heat never really left. Allen’s father had managed to get a small two-room house in Boulder City, and it had a big porch, which is where the beds went. The hardest part, Allen said, was trying to sleep.
About the time the house finally cooled off around 6:00 or 7:00 in the morning, you had to be on the bus to work. A cake of ice, he remembered, was gone before you got it home. By the 2nd of April 1932, the big upper sections had holed through in all four tunnels. Then came the floor.
The jumbos were cut down to a single deck and widened so they could drill the whole bottom face at once. Each round there took 25 holes and about 750 pounds of explosive, and moved the floor forward another 16 feet. On the 23rd of May, the last of the invert came out. Young summed it up plainly: in about a year, more than three miles of 56-foot tunnel had been driven, and about 1.
5 million cubic yards of rock had come out of the mountain. If you piled that rock on a football field, it would stand roughly 700 feet high, close to the height of the dam that was still waiting to be built. The concrete had already started chasing the blasting. Lining began in March of 1932.
Steel forms curved to the shape of the roof rolled along on rails, and a machine called a sidewall jumbo, a huge curved steel form with dozens of shoots, fed concrete into the forms along the walls. Three feet went onto every surface, 300,000 cubic yards of concrete in all, and then a grouting jumbo on rails drilled holes back through the lining so the gaps behind it could be filled. Allen moved with the work. He wasn’t a miner and wanted out of the mining, so he walked up to the crews that had just started concreting at the upper portals and told them he was a cement finisher.
They laughed at him and put him to work anyway. He stayed a cement finisher on that job for about four and a half years. On the 15th of May, crews blasted out the curved concrete walls that had kept the river out of the lower end of Tunnel Number Two. And back in January, other crews had started something that looked strange at the time: small bores climbing up at a steep angle from the roofs of the two outer tunnels.
Those would become spillway tunnels, and they’re the reason the story of these holes didn’t end when the dam was finished. By late summer, the four tunnels were almost ready for the river. Then in September of 1932, the river gave them one more warning. A flood swept down through Black Canyon.
Work was abandoned, and the whole site had to be evacuated. The flood passed, and the crews went back in. But September was also the month the river’s calendar started to turn their way. The low water they’d been racing toward for a year and a half was coming, and the engineers didn’t wait for it.
That same September, they started the upper coffer dam, the barricade above the dam site. Even though the river still hadn’t been moved, before a single load of fill could go down, they had to dig out 250,000 cubic yards of river silt to find firm footing. That same month, Young’s article came out, and in it he laid out what would happen next in plain print. The upper and lower coffer dams would be built, diverting the river through the tunnels.
He wrote it as a plan. Two months later, it was a date. On the 13th of November 1932, the temporary dams that had kept the river out of the two Arizona tunnels were blasted away. The Colorado found the openings and started pouring into the mountain for the first time.
At the same time, crews started closing the old channel. Earth and rock were dumped into the river from a trestle, building a barrier across the riverbed that shoved more and more of the water toward the tunnel mouths. A river that had been cutting Black Canyon for longer than anyone could picture was being pushed into a hole in the wall that a few thousand men had spent a year and a half getting ready. By the 14th of November, the whole Colorado River was running through the two Arizona tunnels.
The Bureau’s own pages give both dates, the 13th for when the river first went in and the 14th for when it was diverted. Both are right. One is the moment the door opened. The other is the moment the river gave up its old bed.
But the river was in those tunnels only because it was low. Now the clock ran the other way. The spring flood would come in May or June, and when it did, the coffer dams had to be standing tall enough and thick enough that the river couldn’t climb over them and pour back into the work. So through the winter, the trucks ran day and night.
On the busiest days, as many as 4,000 truckloads of earth and rock went into the coffer dams. That’s one load every 22 seconds around the clock. The upper coffer dam rose about 600 feet downstream from the tunnel inlets. Finished, it stood 98 feet high, about 30 feet above the tops of the tunnels.
At its base it was 750 feet thick, two and a half football fields of packed earth and rock from front to back, and its upstream face was paved with a skin of concrete six inches thick. Some of the rock in it had come out of the tunnels themselves, along with fill hauled in from Hemenway Wash, the same wash where Ragtown had stood. It was built with a margin. Even with a flood as big as any on record running through the tunnels, the water behind it would still sit 13 feet below its crest.
Below the dam site, a lower coffer dam 66 feet high, and a rock barrier beyond it that stood 54 feet, kept the river from backing up into the work from downstream. By the middle of April 1933, both coffer dams were finished. Spring came anyway. The snow melted in the mountains upstream, and the Colorado rose the way it had risen every spring for as long as anyone knew.
The Bureau’s own history says the engineers watched nervously to see whether the dams would hold. They held, and the tunnels carried the flood easily. For the first time, the river’s spring rise went around the job, and the canyon floor was open to the shovels. Between the coffer dams, the power shovels went down into the old riverbed after the rock the Bureau’s diamond drills had found.
They worked through the silt, the sand, and the gravel, down past where the river’s bottom had been. And by June of 1933, the foundation excavation was done. On the 6th of June 1933, the first concrete went into Hoover Dam, a year and a half ahead of schedule, poured onto bare rock at the bottom of the canyon, right where the Colorado had been running the autumn before. Not everyone who worked in that canyon lived to see the job finished.
By the Bureau’s count, 96 men died at the dam site before it was done. Most of what those crews built is still out there, and you can follow it in a single drive. Start in Boulder City, up on the desert above the canyon. That’s the government town the Bureau was still building during that first summer.
Allen eventually rented a one-room house down on 7th Street for $15 a month. There was no water heater, and the sand blew in off the desert until it lay half an inch deep on the floor. His wife would sweep it out, and by the time she finished it was back. Water cost $2 a month, and you weren’t allowed to water a yard, not even a single green plant.
Running a hose all night so the water sprayed over the roof and down across burlap was fine. That was how people cooled their houses. Irma’s family ended up in Boulder City too. They were among the families who bought land from the government there, and they built their house from freehand sketches, with doors a church had given away.
She spent years after that raising money and pushing for the town’s school, its hospital, its swimming pool, its library, and its cemetery. From Boulder City, US Highway 93 drops east toward the canyon. For most of the dam’s life, the highway crossed right over the top of it. Since October of 2010, it crosses Black Canyon on the bypass bridge instead.
That bridge sits 900 feet above the river, about 1,700 feet downstream of the dam, and it has a walkway for people on foot. Stand on that walkway and look upstream, and you’re looking at the result of everything in this story. The dam fills the canyon where the river used to run, and the river comes out at its foot, far below you, heading down Black Canyon the way it always did. Drive down to the dam itself and walk the crest from Nevada on one end to Arizona on the other.
On each side, one on the Nevada wall and one on the Arizona wall, there’s a spillway, a giant channel cut into the rock about 650 feet long and 150 feet wide. It’s 170 feet deep, about the height of a 17-story building. They’re there for the day Lake Mead gets too full, and each one on its own is built to pass as much water as all four tunnels were built to carry. Here’s where the old tunnels come back.
Water that goes over a spillway drops into a tunnel 50 feet across, inclined at a steep angle and about 600 feet long. At the bottom, it runs straight into one of the two outer diversion tunnels, the same holes those crews cut in 1931 and ’32. A construction photograph from September of 1932 already shows the Nevada spillway tunnel angling down into Tunnel Number One. The water in those spillway tunnels can reach about 120 miles per hour.
That has happened only twice. The Arizona spillway was first put into operation on the 6th of August 1941. The water tore a hole in the tunnel about 115 feet long and 30 feet wide. At its deepest point, the hole went down 45 feet, about as deep as a four-story building is tall.
It was repaired, and the spillways sat dry for more than 40 years. Then came 1983. On the 24th of July that year, Lake Mead reached the highest level ever recorded, with more than two feet of water spilling over the raised spillway gates. The high water did damage across the whole project, and the Colorado went right back into the tunnels that were cut for it half a century earlier.
The other two tunnels, the inner pair, have a different life. Young had spelled out the order in his article: the river sent through the outer pair, while the plugs went into the inner pair. The inner tunnels were plugged with concrete about a third of the way down from their inlets, and the outer pair were plugged about halfway. Young’s plan put steel pipes inside the inner tunnels, leading down from the intake towers to plugs fitted with big valves, and that was how those two tunnels were meant to carry water past the dam.
Carl Merrill helped pour one of those plugs. He came to the project in January of 1933, spent the better part of two years carrying dishes in a dining hall, and then went down into the work as a mucker and a puddler, working wet concrete with his feet in hip boots. When they poured the plug in Tunnel Number Two on the Arizona side, he was the hook tender, hooking and unhooking the buckets of concrete as they rode a monorail back into the tunnel. It paid $4.
50 a day. The last door on the detour closed on the 1st of February 1935. A steel gate weighing more than a thousand tons was lowered over the entrance to Tunnel Number Four, and the river began backing up behind the dam. By the end of that year, the new lake had covered the flats where Ragtown stood: the tents, the shacks, and the riverbank where Irma nearly lost her daughter.
Below the dam, the canyon is still there. The National Park Service runs about 30 miles of the river below Hoover Dam as the Black Canyon National Water Trail. Paddlers launch at the foot of the dam with a permit, or at Willow Beach, Arizona, 14 miles south of the dam off the same highway, and float out through the same canyon walls the tunnels were cut into. Go back to where this started: the picture taken from inside Tunnel Number Two that January, looking out at the canyon.
When it was taken, the Colorado was still running in its old bed just outside that opening, and the men standing in it had about ten months left to get the whole river inside the mountain. They made it with most of a year to spare on the contract. And everything that came after, the dam, the power plant, and the lake behind them, started in the dry space those four holes made.
