Training underwater for extravehicular activity (EVA)—popularly known as spacewalking—is now critical for preparing astronauts to work in weightlessness. But when cosmonauts and astronauts first ventured outside their spacecraft 50 years ago, in 1965 and 1966, they had no such training. Spacewalking did not appear difficult, nor did space program officials think that underwater work was needed.

In the United States, it took Eugene Cernan’s June 1966 Gemini IX EVA to change attitudes. Fighting against his pressurized suit, while trying to do work without adequate handholds and footholds, Cernan quickly became exhausted and overheated. Only afterward did NASA Manned Spacecraft Center in Houston reach out to a tiny company outside Baltimore: Environmental Research Associates, Inc. (ERA). Funded by another agency center, it had been experimenting with EVA simulation in a rented school pool on nights, holidays, and weekends. That project became the foundation for Houston’s first underwater training facility. In parallel, NASA Marshall Space Flight Center in Huntsville, Alabama, had begun doing its own experiments. Together they effectively co-invented “neutral buoyancy training,” so-called because astronauts in weighted, pressurized suits have to be made neutrally buoyant to simulate zero gravity.

Medical researchers had been aware for some time that water immersion could be a useful simulation of weightlessness. A handful had been experimenting with it to understand the effects on the human body. But when human spaceflight began in 1961, with the Yuri Gagarin and Alan Shepard flights, EVA concepts were fairly primitive. In the space advocacy and science fiction literature of the 1950s and earlier, one can find magnetic shoes, jetpacks, and “bottlesuits” (miniature spacecraft), but no systematic attempt to understand work in space. Little thought had been given to the implications of Newton’s third law of motion—for every action, there is an equal and opposite reaction—even though it was the foundation of rocket motion. For an individual in zero-G, however, a simple act like turning a bolt would be become difficult if not anchored in place: twisting a wrench would result in one’s body rotating in the opposite direction. But working outside the spacecraft was not an urgent problem when it was challenging simply to orbit a man (and in those days only a man). It was not until the Soviet Union and the United States began working on their second human spacecraft—Voskhod and Gemini—that EVA became something to train for. But both sides already had zero-G airplanes, in which weightlessness could be created for about 30 seconds by flying parabolas. Underwater training seemed like too much trouble.

In the U.S., neutral buoyancy experimentation began in 1963 and 1964 at aerospace companies and at NASA Langley Research Center in Hampton, Virginia. Space station studies were the primary motivation. These stations, both civilian and military, were supposed to be launched in the late 1960s or 1970s, meaning that work inside and outside in zero-G would become essential. Boeing in Seattle was the first company to build its own facility, followed by General Electric and others. Yet corporate work petered out in the late 1960s with the decline in the government space budget, leaving little legacy.

The Langley project came out of a study for a small NASA station. ERA founders Samuel Mattingly and Harry Loats had been working on a contract for the Virginia center and were briefed on it. Mattingly suggested that the station lacked an airlock for getting in and out in a vacuum without losing its entire atmosphere. That led to a discussion of how big the airlock should be and what the mechanics of entering, exiting, and turning around in it were. Their Langley contract monitor, Otto Trout, had a clear plastic airlock mockup built for testing, but it was immediately obvious that one-G simulations had limited realism. Mattingly and Loats borrowed some Navy pressure suits and one day in spring 1964 they sunk the mockup in the officers’ club pool on the adjacent Air Force base. But effective work was hardly possible, so Trout let them take the mockup back to Baltimore. Mattingly quickly made a rental arrangement with the McDonogh School, a private military boys’ school in the northwest suburbs, near Owings Mills, Maryland. In July, the ERA project began with time-and-motion studies. At the only dive shop in the area, Mattingly and Loats recruited college-aged scuba enthusiasts to serve as test subjects and safety divers. Scuba tank and regulator technology were only 20 years old and had led to a fast-growing hobby. It became an enabling technology for inventing neutral buoyancy training.

Because of the ERA project’s effectiveness and low cost, Otto Trout managed to extend the contract several times through mid-1966. But Mattingly had had no luck interesting NASA’s Houston center. Before and after Edward White’s first U.S. spacewalk in June 1965, officials there told him that they just did not need neutral buoyancy. They had the zero-G airplane and a frictionless air-bearing surface at the center, which provided a two-dimensional analogue to weightlessness. White’s EVA had been easy, except for getting back in his seat and closing the hatch over his head, but he had done no real work other than that. Aleksei Leonov’s dangerous crisis three months earlier, when he tried to get back in his airlock with an overinflated suit, had been concealed from the world. The next U.S. spacewalk, by David Scott on Gemini VIII in March 1966, was canceled because the spacecraft made an emergency return to Earth. So when Cernan had frightening problems on June 5, Houston officials were caught off guard.

Coincidentally, Mattingly and Loats staged a demonstration at the McDonogh pool 10 days later in the hope of keeping their project afloat. Representatives came from both Houston and Huntsville (Marshall had been conducting its own experiments since fall 1965). Don Jacobs from Houston saw neutral buoyancy’s value in preparing for the rapidly approaching Gemini X, XI and XII missions, all of which had spacewalks. But ERA’s work on procedures came too late in the rushed schedule to be of help to the first two crews. But there was time to train Buzz Aldrin at McDonogh, the EVA astronaut on the last Gemini flight. His November 1966 spacewalks demonstrated that careful procedures, plus the extensive use of tethers, handholds, and footholds, could manage the workload of working in a rigid, pressurized spacesuit in zero-G. It gave a clear message that the EVA problem could be mastered. (The other critical technology was the water-cooling undergarment introduced in Apollo suits.

Back in late July, Manned Spacecraft Center Director Robert Gilruth had already decided that Houston needed its own pool. A tank used for water egress training was moved and retrofitted by summer 1967. That pretty much was the death knell for ERA’s project, and eventually for the company, which had served its purpose in inventing the appropriate tools and procedures for neutral buoyancy. In parallel, NASA Marshall had gathered its own experience and had seen what ERA had accomplished. In late 1968 it finished the construction of a much larger tank that became essential to the Skylab and Shuttle programs. Houston went on to build two more facilities. The last, the gigantic Neutral Buoyancy Laboratory opened in 1997, has made the construction and maintenance of the International Space Station feasible. Yet it all went back to a handful of people working in a school pool outside Baltimore.