Project Mohole and the Long Race to Drill Into Earth’s Mantle

In late June 2026, a ship with a 62-meter derrick sticking up from its middle sailed into a breaking yard on the Aegean coast of Turkey. Over the next few months, cutting torches took it apart. For almost four decades the JOIDES Resolution had been the busiest research ship in the earth sciences. It drilled thousands of holes in the ocean floor, pulled up hundreds of kilometers of mud and rock, and helped show that the Mediterranean once dried out, that the seafloor moves, and how Earth’s climate has swung over the past hundred million years. Its American funding ran out in 2024, and its owner sold it for scrap the next year.

Three months later and half a world away, a brand-new Chinese drillship called Meng Xiang, or “Dream,” came home to Guangzhou from its first trial voyage. It had lowered pipe through more than four kilometers of South China Sea water and carried a column of basalt from deep below the seabed back to port. Its builders say openly what it is for. Before 2030, they want to drill all the way through the ocean crust into the mantle.

That goal is older than the space program. It began in 1957 with an offhand suggestion at a government science panel, and it grew into Project Mohole, one of the most celebrated and most mocked science projects of the Cold War. Mohole never reached the mantle. It did invent the tools for a much bigger undertaking, scientific ocean drilling, which turned out to be one of the most productive projects in the history of geology. This is the story of that hole, the ones that came after it, and why the deepest rock on the planet is still out of reach.

A thin skin over a hot planet

Earth is built in layers. The crust under our feet is a thin, light skin. Beneath it lies the mantle, a dense layer of hot rock, about 2,900 kilometers thick, that holds most of the planet’s volume. It is solid, but over millions of years it creeps and flows like very stiff putty, and that slow movement drives everything from volcanoes to the drift of the continents.

The boundary between the two layers was found in 1909 by the Croatian seismologist Andrija Mohorovičić. While studying an earthquake near Zagreb, he noticed that some seismic waves arrived at distant stations sooner than they should have. They had dipped down into faster rock below and come back up. That sharp jump in wave speed marks the base of the crust, and geologists still call it the Mohorovičić discontinuity, or simply the Moho.

Nobody has ever seen a piece of mantle collected from where it sits. Bits of it come up in volcanic lava as green crystals of olivine, and in a few places mountain-building has shoved slabs of old seafloor and mantle onto land. Those samples have all been altered on the way up, though, and geologists have long wanted a clean, continuous column of rock running from the surface down through the Moho, the way an ice core runs down through the ages of a glacier.

On land the Moho is usually 30 to 50 kilometers down, far beyond the reach of any drill. The Soviet Union’s Kola Superdeep Borehole, the deepest hole ever drilled, stopped at 12,262 meters in 1989 when the rock grew too hot to work, and it got barely a third of the way through the continental crust. Under the deep ocean, though, the crust is only about six or seven kilometers thick. The catch is that the ocean floor itself sits four or five kilometers below the waves. To reach the mantle there, you have to drill from a ship floating on the surface, through miles of water, into some of the hardest rock on Earth.

A big idea at the American Miscellaneous Society

In March 1957, the National Science Foundation asked a panel of earth scientists to review proposals for funding. The oceanographer Walter Munk of the Scripps Institution of Oceanography was bothered that every proposal was a modest, safe project. What would really change geology, he suggested, was a hole through the crust to the mantle. The Princeton geologist Harry Hess, another panel member, agreed at once.

The two men took the idea to a group that called itself the American Miscellaneous Society, or AMSOC. It had been founded in 1952 by Gordon Lill and Carl Alexis of the Office of Naval Research, half in jest, as a home for ideas that did not fit anywhere else. It had no members list, no dues and no bylaws. Anyone could join whenever two members happened to be together. Munk and Hess were part of this loose circle, and the mantle hole became its most serious project by far.

The timing helped. The Soviet Union had just launched Sputnik, Washington was eager to fund big science, and there were rumors that Soviet scientists were thinking about their own deep hole. AMSOC got a $15,000 grant through the National Academy of Sciences in 1958 to study whether the job could be done, and named the hole after the boundary it was aiming for: the Mohole. The Academy’s own history describes it as the earth sciences’ answer to the space program.

To run the engineering, the committee hired Willard Bascom, an ocean engineer and writer who had studied waves and beaches in California. Bascom’s plan was cautious. Phase I would test the basic idea, drilling from a ship in deep water. A later phase would build a new purpose-built vessel to go all the way to the Moho.

Holding a ship still over the abyss

Oil companies were already drilling offshore by the late 1950s, but only in shallow water, from platforms resting on the bottom or ships anchored firmly in place. In water three or four kilometers deep, anchoring was out of the question. The drill string would hang from a ship that the wind, waves and currents were constantly pushing around. If the ship drifted too far, the pipe would bend and snap.

Bascom’s team leased a drilling barge called CUSS I, short for the four oil companies that had paid to convert it from a surplus Navy barge: Continental, Union, Superior and Shell. It was operated by a young California firm, Global Marine. To keep it in one spot without anchors, Bascom came up with what we now call dynamic positioning. Four large outboard motors were mounted around the hull, each able to swing in any direction. Around the drill site the crew laid a ring of buoys, tracked by radar and sonar, and an operator in a control room used a joystick to work the motors and keep the barge inside the circle. It was crude by modern standards, but it worked. The ship stayed within about 180 meters of the hole, close enough for the pipe to flex without breaking.

After trials off La Jolla, CUSS I sailed in March 1961 to deep water near Guadalupe Island, off Mexico’s Pacific coast. There the seafloor lay about 3,600 meters down. The chief scientist was William Riedel of Scripps, and Munk and the oceanographer Roger Revelle came along too.

So did John Steinbeck. Life magazine had hired the novelist to cover the expedition. He had a famously sharp eye and a soft spot for marine scientists, and his account, published that April, is still the liveliest description of the work. He was not impressed by the looks of the barge, which he said had “the sleek race lines of an outhouse standing on a garbage scow.” He was impressed by the people, who worked around the clock in rough seas to make a crazy idea work.

Over a few weeks the crew drilled five holes. The deepest went 183 meters into the seafloor. The first 170 meters or so was soft sediment, much of it the remains of tiny plankton that had drifted down from the surface, with the oldest layers dating to the Miocene epoch, millions of years ago. Then the drill hit something much harder. The last 13 meters of core was basalt, the dark volcanic rock that makes up the second layer of the ocean crust. No one had ever sampled it from the deep ocean floor before.

Steinbeck wanted a piece of that basalt as a souvenir. As he told it in Life, he was turned down, so he quietly pocketed a small fragment anyway. Then the chief scientist secretly gave him a piece, and the guilty writer slipped the stolen one back. He also came away convinced the expedition should “destroy the old and well loved error that doers and thinkers are different breeds.” The whole of Phase I cost about $1.7 million, under budget, and newspapers all over the country carried the story. Bascom wrote it up the same year in a book called A Hole in the Bottom of the Sea.

How a triumph turned into a punch line

Success brought money, and money brought trouble. AMSOC had been a loose club of enthusiasts. The next phase would cost tens of millions of dollars, far beyond anything a volunteer committee could manage, so the National Science Foundation took over in late 1961 and invited companies to bid to run it.

What followed has become a textbook case of how not to run a big science project. A review panel rated Socony Mobil’s bid the best by a wide margin. Instead, in early 1962, the NSF awarded the contract to Brown & Root, a Houston construction firm that had built dams and Navy bases but had never drilled an oil well. Critics pointed out that Brown & Root was a major contributor to Vice President Lyndon Johnson and that its home was close to the district of Representative Albert Thomas, the man who chaired the House subcommittee that controlled the NSF’s budget. No one ever proved anything improper, but the choice poisoned the project from the start.

Within two months, Bascom’s engineering firm quit. The scientists and the contractor then fell into a long fight over what Mohole was for. Many scientists wanted an intermediate program first, drilling lots of shallower holes all over the ocean to learn the basic geology. Brown & Root and the NSF wanted to go straight for the big hole. The plans grew grander and more expensive. A huge floating platform, standing on six columns above two submarine-like pontoons, was designed to hold steady in the open Pacific. A drill site north of Maui was chosen in 1965, and a shipyard in San Diego got the contract to build the platform.

By then the cost estimates had climbed past $100 million, roughly ten times the first guesses. Hollis Hedberg, a Gulf Oil geologist and Princeton professor who chaired AMSOC’s own Mohole committee, told Congress in November 1963 that the project could be one of the greatest scientific ventures ever, or a “foolish and unjustifiably expensive fiasco” if it was not run on a sound scientific plan. After being reprimanded for his testimony, he resigned. AMSOC itself disbanded in early 1964. The press, which had cheered three years earlier, now competed for nicknames. Newsweek called it “Project No Hole.” Fortune ran a story under the headline “How NSF Got Lost in Mohole.”

The end came quickly. Albert Thomas died in February 1966, and the project lost its most powerful protector just as the Vietnam War began squeezing the federal budget. That summer, Congress refused any more money. About $57 million had been spent, and not a single additional meter had been drilled. Decades later, Munk said that the triumph of Phase I had in a way doomed the rest of the project. It had made the job look easier than it was, and it attracted people who were more interested in a showpiece than in the science.

The consolation prize that changed geology

Even while Mohole was falling apart, the scientists who had wanted the “intermediate” program were quietly getting it. In 1964, four leading oceanographic institutions, Scripps, Woods Hole, Lamont at Columbia University and the University of Miami, formed a partnership called the Joint Oceanographic Institutions for Deep Earth Sampling, or JOIDES. In 1965 they drilled a series of test holes off Florida. In June 1966, just as Congress was killing Mohole, the NSF signed a contract with Scripps for a far more modest project: the Deep Sea Drilling Project.

The idea was not to drill one heroic hole, but hundreds of ordinary ones, spread across every ocean, to sample the sediments and the top of the crust. For a ship, Scripps turned to Global Marine, the company that had run CUSS I. In 1968, it launched the Glomar Challenger, named after HMS Challenger, whose voyage in the 1870s had founded the science of oceanography. It was the first ship built from the keel up for scientific drilling, with a 43-meter derrick, computer-controlled thrusters that grew out of Bascom’s outboard motors, and a laboratory where scientists could study the cores as soon as they came up.

Global Marine went on to help create an even stranger ship. A few years later, it designed and operated a vessel for the CIA in Project Azorian, the secret attempt to raise a sunken Soviet submarine from five kilometers down. That ship was the Hughes Glomar Explorer. The deep-water know-how that came out of Mohole had uses that its founders never imagined.

Proving the seafloor moves

When the Glomar Challenger went to sea, geology was in the middle of a revolution. Harry Hess, the same Princeton geologist who had cheered Munk’s idea in 1957, had proposed in the early 1960s that new ocean floor forms at the long volcanic ridges running down the middle of the oceans and slowly spreads outward from them, like two conveyor belts moving apart. Magnetic stripes on the seafloor strongly supported the idea, but many geologists were still skeptical. No one had actually dated the rocks.

The ship’s third voyage settled the question. Between December 1968 and January 1969, it sailed from Dakar to Rio de Janeiro, drilling 17 holes at 10 sites in a line across the Mid-Atlantic Ridge. At each site, the scientists dated the oldest sediment sitting directly on the basalt, using the fossils of tiny plankton. The pattern was unmistakable. Next to the ridge the bottom sediment was very young. Moving away from it, the bottom sediment got steadily older, in step with the magnetic stripes. The seafloor was spreading at about two centimeters a year on each side, roughly the rate your fingernails grow.

The results, published in Science in 1970 by Arthur Maxwell, Richard Von Herzen and their colleagues, are often cited as the drilling evidence that sealed the case for seafloor spreading and plate tectonics. The same ridge runs up through Iceland, where the Laki fissure tore open in 1783, one of the few places where you can stand on dry land and watch two plates pull apart.

The summer the Mediterranean ran dry

In the summer of 1970, the Glomar Challenger drilled in the Mediterranean, with the Swiss geologist Kenneth Hsü and the American William Ryan as co-chief scientists. The seismic surveys had shown a strange, hard layer a few hundred meters under the sea floor. When the drill reached it, it brought up gypsum, rock salt and other minerals that form when salty water evaporates. Mixed in were pebbles of the same minerals that looked like they had been rolled around in shallow water, and fossils of algae that could only have lived in sunlight.

The conclusion seemed outrageous. About six million years ago, the Mediterranean had been cut off from the Atlantic and had mostly evaporated, leaving a basin up to three kilometers deep, floored with salt flats and briny lakes. Rivers like the Nile carved deep canyons to reach its shrinking shores. Geologists now call it the Messinian salinity crisis, and they date it from about 5.96 to 5.33 million years ago. It ended when the Atlantic broke through at Gibraltar and refilled the basin, possibly in a catastrophic flood that took only a few years. Hsü told the story for a general audience in his book The Mediterranean Was a Desert.

The Glomar Challenger kept going for 15 years. Its crews learned to replace a worn drill bit and find the same hole again in water four kilometers deep, by lowering a funnel-shaped “re-entry cone” to the seafloor and guiding the pipe back into it with sonar. They invented a hydraulic piston corer that could punch into soft mud and bring it up undisturbed, which allowed the first detailed records of past climate from the deep sea. From 1975 the project went international, with the Soviet Union, West Germany, France, Britain and Japan joining in. By the time the ship retired in 1983, it had made 96 voyages and recovered more than 97 kilometers of core. It was scrapped soon after, but some of its equipment ended up in museums, including the Smithsonian.

Libraries of mud and rock

The next ship was bigger. The JOIDES Resolution started life in 1978 as an oil exploration ship. In 1984 it was converted for science, with seven stories of laboratories, and it began work for the new Ocean Drilling Program in 1985. Over the next four decades, as the program changed names and partners, it became the backbone of scientific ocean drilling. In the most recent phase, the International Ocean Discovery Program, which ran from 2013 to 2024, the JOIDES Resolution carried out most of the expeditions and recovered almost all of the core.

All that core has to go somewhere. Each section is cut in half lengthwise. One half is studied and sampled, and the other is kept untouched as an archive, so that scientists decades later can come back with new questions and new instruments. Texas A&M University’s Gulf Coast Repository stores some 151 kilometers of it, and there are matching libraries in Bremen, Germany, and Kochi, Japan. It works a lot like the ice cores pulled from Greenland, where the Camp Century core is still revealing surprises more than half a century after it was drilled.

Those cores have shown, among much else, how the Antarctic ice sheet formed, how the Indian monsoon got started, how Earth warmed during the hot spell 56 million years ago, and how the dinosaur-killing asteroid impact rearranged the planet. In 2016, an expedition drilled into the peak ring of the Chicxulub crater off Mexico’s Yucatán Peninsula. Drilling has also revealed microbes living in rock more than two kilometers below the seabed, part of a “deep biosphere” that no one knew existed when Mohole was planned. By 2025, Texas A&M counted more than 14,800 scientific articles based on the program’s work.

Still trying for the Moho

The original goal never went away. In 1979, drillers began a hole called 504B in the eastern Pacific off Costa Rica. Over seven expeditions and 14 years, they deepened it to 2,111 meters below the seafloor, through the lava flows and into the sheeted dikes, the vertical sheets of rock that once fed magma to the surface. Then trouble in the hole ended the effort. They never reached the gabbro, the coarse-grained rock that crystallizes deep in the crust from the magma chambers under the ridge.

Another hole, 1256D, in a stretch of very fast-spreading Pacific seafloor where the lower crust should be shallower, finally reached gabbro in 2005 and was later deepened to more than 1,500 meters. That was a milestone, but it still left four or five kilometers of rock between the bottom of the hole and the Moho.

The obvious answer was a bigger drill. Japan built one. The Chikyu, which means “Earth,” was launched in 2002 and entered service in 2005. It is a 210-meter monster with a derrick 70 meters tall and something no earlier research ship had, a riser. That is a wide outer pipe surrounding the drill string, like the systems oil companies use, which carries the drilling mud back up to the ship and stops the hole from caving in. It was designed to drill seven kilometers below the seabed, and reaching the mantle was one of the reasons it was built.

The Chikyu has had a rough career. The 2011 tsunami hit it while it was in port at Hachinohe, punching a hole in its hull and damaging a thruster. It has spent much of its time on other jobs, including drilling into the fault that caused that earthquake. In September 2024, it set a Guinness world record when it lowered 7,906 meters of drill pipe to sample the plate boundary in the Japan Trench. In early 2026, it tested lifting rare-earth-rich mud from 6,000 meters down near the remote island of Minamitorishima. A mantle hole has been planned for years, with sites studied off Hawaii, Costa Rica and Baja California, but the cost, which would run to many hundreds of millions of dollars, has kept it on the drawing board.

A shortcut on the Mid-Atlantic Ridge

In the end, the closest anyone has come to a mantle core took a shortcut. In some places along the slow-spreading Mid-Atlantic Ridge, faults have pulled mantle rock up to the seafloor, so there is no crust to drill through at all. One such place is a mountain called the Atlantis Massif, about 30 degrees north of the equator. On its flank sits the Lost City hydrothermal field, where white chimneys of carbonate rock, some as tall as a 20-story building, vent warm, alkaline water rich in hydrogen and methane.

Those vents are powered not by magma but by chemistry. When seawater seeps into the mantle rock, it reacts with olivine to form a greenish mineral called serpentine, a process called serpentinization, which gives off heat and hydrogen. Many scientists suspect that reactions like these, in places like Lost City, helped set the stage for the first life on Earth.

In the spring of 2023, the JOIDES Resolution drilled a hole called U1601C about 800 meters from the Lost City vents. The plan was to go perhaps 200 meters into the mantle rock. Instead, the drill kept going and the core kept coming. By the time the expedition ended, the hole was 1,268 meters deep, and the crew had recovered about 71 percent of it, a column of serpentinized peridotite, mostly the mantle rocks harzburgite and dunite, longer than the Empire State Building is tall. The previous record for drilling into mantle rock was about 200 meters, with less than half of it recovered.

The first results, published in Science in August 2024 by Johan Lissenberg of Cardiff University and colleagues, showed how melt had once moved through the rock and how far seawater had reached into it. The core is not a perfect sample of the mantle. It has been brought up close to the surface and heavily altered by seawater. It is still the closest thing to a continuous section of the upper mantle that anyone has held, and it may take decades to study fully, which is exactly why the archive halves matter.

The ship that went for scrap

That expedition turned out to be one of the JOIDES Resolution‘s last. In 2023, the NSF announced that it would not renew its support for the ship, citing rising costs and a lack of firm commitments from international partners. The ship had a few years of useful life left. A congressional committee report in 2024 urged funding through 2028, but by then it was too late. The final expedition ended in 2024. In October 2025, a Norwegian company bought the ship for recycling, and in 2026 it made its last voyage to a shipbreaking yard near Izmir.

The program has not stopped, but its center of gravity has shifted. A European and Japanese successor program, run with the Chikyu and with platforms hired for individual expeditions, began work in 2025, and Australia and New Zealand have joined it. The Chikyu sailed in late 2025 to drill the young “petit-spot” volcanoes off Japan, where small amounts of magma leak up through old seafloor. Meanwhile, scientists in Texas and elsewhere have been going back to the shelves of old core to ask new questions.

The Dream ship

China, which had joined the international program as a partner, decided to build its own ship. The Meng Xiang was commissioned in Guangzhou in November 2024. It is about 180 meters long and is designed to reach as deep as 11,000 meters. A paper published in Nature Geoscience in 2025 by Chinese scientists laid out the plan to use it to drill the Moho beneath the Pacific or Indian Ocean before 2030.

The ship’s first trial expedition, in the South China Sea in water about 4,100 meters deep, ended in late September 2026. According to Chinese state media, it drilled 864.77 meters below the seafloor, reached a total operating depth of 4,929 meters and brought up a 4.27-meter core of basalt. China says that makes it the third country, after the United States and Japan, to recover hard rock from beneath a deep ocean basin. Another voyage is planned for next year.

Getting from there to the Moho is a far bigger step. It means drilling five or six kilometers into hard, hot rock, in water four kilometers deep, without the hole caving in and without the drill bit wearing out every few dozen meters. Temperatures at the bottom would be well above 150 degrees Celsius. Those are the same problems that defeated Mohole on paper in the 1960s and that Kola’s drillers met on land in the 1980s.

What the hole would tell us

Why keep trying? Most of the rock on Earth is in the mantle, and almost everything we know about it comes indirectly, from seismic waves, from lab experiments at high pressure, from lavas that have melted and changed on the way up, and from slabs that have been bent and weathered on land. A continuous core through the oceanic crust and across the Moho would show what the crust is actually made of, how it forms at the ridges, what the boundary really looks like and how deep life and seawater reach into the planet.

It would also settle questions that have been argued about since Mohorovičić. Is the Moho a sharp boundary between different kinds of rock, or a gradual change? How much water does the ocean crust carry down into the mantle when it sinks at subduction zones, feeding the kind of volcanoes that built North America’s largest eruptions? How much carbon is locked up in serpentinized rock?

Project Mohole was born in the same anxious, ambitious years as the race to the Moon, and in some ways it was its mirror image. NASA spent billions to send a few people a few hundred thousand kilometers into space, trusting a computer smaller than a suitcase to get them there. Mohole needed a few tens of millions to go a few kilometers down, and it failed for reasons that had little to do with rock. Yet its failure gave the world the Glomar Challenger, the proof of plate tectonics, the drowned Mediterranean, the climate archives in the cores and, almost by accident, a kilometer-long column of the mantle from the floor of the Atlantic.

The last ship of that line is now scrap metal in Turkey. The next one is flying a Chinese flag. Sixty-five years after John Steinbeck pocketed a chunk of basalt off Guadalupe Island, the Moho is still waiting, about six kilometers below the floor of the sea.

Further reading