SEALAB II — the Navy's house on the sea floor off La Jolla.

On August 28, 1965, a steel habitat settled onto a ledge of Scripps Canyon at a depth of 205 feet, off the coast of La Jolla. It was fifty-seven feet long, twelve feet across, and built to sustain human life in the Pacific's cold dark — bunks, a galley, a laboratory bench, and a wet room with a hatch that opened directly to the sea floor, through which aquanauts could descend into the canyon without ever ascending to decompress. For the next forty-three days, three rotating teams lived and worked from that ledge, two hundred feet below the last of the daylight, in the most ambitious attempt the United States had yet made to put human beings on the floor of the ocean.

The program that placed them there had been conceived in the early Cold War, when the deep seabed suddenly mattered. Undersea telephone cables needed maintenance. Salvage operations needed labor. The continental shelf held mineral resources the Navy wanted to understand and — if necessary — contest. The problem was time: a conventional scuba diver working at 200 feet had perhaps twenty useful minutes before decompression requirements forced a long, slow ascent. The Navy's Man-in-the-Sea program was its answer. SEALAB I, in the summer of 1964, had placed four aquanauts 192 feet down in the waters off Bermuda for eleven days before a weather system forced an early recovery. It proved the concept would not immediately kill anyone. SEALAB II was designed to find out what was actually possible.

The man most responsible for the program's existence was Navy captain and physician George Bond, who had spent a decade arguing for saturation diving against institutional inertia. Bond's central insight was physiological: once the body's tissues are fully saturated with dissolved gas at a given depth, additional time at that depth costs nothing in decompression terms. A diver who has been at 200 feet for an hour and one who has been there for two weeks face essentially the same decompression schedule on the way up. The practical implication was transformative — if you could keep people alive at depth long enough to work, you could keep them there for weeks at no additional physiological cost. To breathe at 205 feet, his aquanauts used heliox, a helium-oxygen mixture that sidestepped the nitrogen narcosis that would have rendered them incoherent at that pressure. The helium had a side effect: it accelerated vocal vibration enough that communications with the surface were often unintelligible on the first pass, the aquanauts' voices reduced to something between a whisper and a cartoon.

Inside the habitat, life was spare and relentless. Condensation dripped from every surface. The water outside the hatch held near forty-five degrees Fahrenheit, cold enough to kill an unprotected man in hours, and the aquanauts wore electrically heated suits to work. The canyon walls beyond the habitat's port lights dropped away into a darkness that no amount of dive lighting fully penetrated. The teams ran eight-hour shifts — conducting marine biology surveys, testing tools and work procedures, evaluating human performance under sustained pressure — and returned to bunks that were never more than a few feet from the open sea. They were sealed in a pressurized cylinder unreachable by any rescue system then in existence, with the full weight of the Pacific above them.

The most recognizable member of the program was not a diver by training. Scott Carpenter had been a Navy aviator and test pilot before NASA selected him for Project Mercury, and on May 24, 1962, he became the second American to orbit Earth, completing three circuits in Aurora 7 before splashing down in the Atlantic. A navigation error at reentry caused him to overshoot the intended recovery zone by 250 miles; the search for his capsule lasted nearly three hours. His space career did not continue after that mission. When SEALAB II needed a commander for its second team, Carpenter volunteered. He entered the habitat in mid-September 1965 and then stayed through the entirety of the third team's rotation, spending a cumulative thirty days on the canyon floor — longer than any other aquanaut in the program. He became, at that point, the only human being who had lived in both orbital space and the deep ocean.

The program's most unconventional asset worked not inside the habitat but in the water column between it and the surface. Tuffy was a bottlenose dolphin trained under the Navy's Marine Mammal Program at Naval Air Station Point Mugu. He had been conditioned to respond to an acoustic buzzer and to carry sealed messages and tools in a harness from the surface support vessel down to the habitat — 205 feet through water too cold and too dark for a swimmer to navigate safely. Tuffy also served as contingency for a real hazard: visibility outside the hatch could drop to near zero when work operations stirred up the canyon silt, and a disoriented diver who overshot the ledge could find himself descending into open water with nothing to navigate by. Tuffy was trained to home in on the buzzer signal and lead a lost diver back to the hatch. He completed dozens of runs during SEALAB II without incident, making the mail route between sea level and the canyon floor as reliably as any line or cable the engineers had devised.

The canyon where SEALAB II rested lies directly below the bluffs of La Jolla, where the Scripps Institution of Oceanography has conducted marine research since 1903. Scripps scientists participated in site selection for the habitat, and the institution's proximity allowed a working relationship between Navy operational goals and academic oceanography that shaped Cold War-era ocean science from San Diego outward. Across the bay, Naval Base Point Loma today houses the successor organizations to the programs that made SEALAB possible, including the command that still administers the Marine Mammal Program that trained Tuffy. The Navy's investment in the waters off this city — in what lived in them, what could be built in them, what could be sustained in them — runs from Bond's pressure chambers in the late 1950s through to the submarines that transit the harbor mouth on their way to the open Pacific today.

The saturation diving methods Bond and his aquanauts validated became standard practice for commercial and military dive operations worldwide — today, wherever a pipeline needs a weld or a wellhead needs inspection at depth, divers use the same physiological logic that Bond proved in a yellow cylinder off La Jolla. The specific ambition of SEALAB — permanent habitation, the aquanaut as the continental shelf's resident — never came back after SEALAB III was cancelled in 1969 following a fatal accident during preliminary operations. But the canyon is unchanged. The cold shelf drops away at 205 feet off La Jolla exactly where Tuffy swam his mail runs and Carpenter watched the darkness through a porthole. You could sail out over it tonight.