On the morning of April 10, 1963, USS Skylark was sitting on the Atlantic above a submarine it was supposed to be able to help.

That was part of the reason Skylark existed. She was a submarine rescue ship, one of the descendants of the vessels that had accompanied earlier American submarines on dangerous test dives. Her equipment included a rescue bell that could be lowered through the water and attached to a disabled submarine so trapped sailors could be brought up a few at a time.

Skylark’s rescue chamber was rated to about 850 feet. Thresher’s deep-dive trial was taking the submarine toward roughly 1,300 feet. The rescue ship was therefore accompanying a submarine whose planned test depth already exceeded the reach of its principal rescue chamber.

Twenty-four years earlier, a version of that rescue technology had done something extraordinary. When USS Squalus sank in 240 feet of water off New Hampshire, a rescue chamber descended repeatedly and brought thirty-three men back alive—the story behind Final Vestige’s The Rescue Machine Nobody Had Ever Needed.

By 1963, the rescue machine was no longer deep enough for the submarines.

That is not why the men aboard Thresher died. The submarine almost certainly descended beyond the depth at which her pressure hull could survive. By the time there was anything to rescue, there may have been no intact submarine left to reach.

But the fact that Skylark was sitting above her with equipment already overtaken by submarine design tells us something about the world the Navy had entered. The boats had become very deep. Everything around them had not caught up.

The new submarine

Thresher was not old machinery being pushed beyond its time. She was the opposite.

Commissioned in August 1961, she was the lead ship of a new class of nuclear attack submarine built for speed, stealth, deeper operation and increasingly sophisticated sonar and weapons work. She spent much of her short service life testing those systems.

The photographs make her look almost unnervingly simple: a dark teardrop hull, a sail and very little else. That shape concealed a machine whose survival depended on a dense network of seawater piping, electrical systems, reactor machinery, valves, high-pressure air and thousands of individual decisions made during design, fabrication, inspection and maintenance.

In July 1962, Thresher returned to Portsmouth Naval Shipyard for overhaul. The work lasted longer than expected, and the boat finally went back to sea in April 1963. Seventeen civilian shipyard technicians accompanied her crew for the post-overhaul trials.

On April 9, she left Kittery, Maine, and rendezvoused with Skylark offshore. The following morning the deep-dive test began.

Every hundred feet

The descent was deliberate. Navy reconstructions describe Thresher circling beneath Skylark and pausing roughly every hundred feet so systems could be checked before the boat went deeper. Communication passed through an underwater telephone, but the link was poor enough that messages reaching Skylark were distorted and later accounts differ slightly on their wording.

At about the halfway point to test depth, things were apparently normal.

Then the messages changed.

One partial communication heard aboard Skylark was interpreted along the lines of “minor difficulties,” followed by references to a positive up-angle and an attempt to blow ballast. Another garbled transmission included what listeners understood as “900.” The precise words remain imperfect because the transmission itself was imperfect.

Then nothing useful came back.

On Skylark, the rescue problem was immediate: a submarine carrying 129 people had stopped answering beneath them, and the chamber aboard the rescue ship could not reach the depth Thresher was testing. The crew kept calling and then used signal grenades to send simple acoustic signals through the water. None brought back a response.

The silence below

By the end of the day, bits of material and an oil slick were being found at the surface. Families were informed that the submarine was overdue and presumed missing.

All 129 people aboard—112 Navy personnel and 17 civilian shipyard technicians—were dead.

The wreck eventually came to rest about 8,400 feet beneath the Atlantic. At that depth the older rescue-bell problem became academic. The surviving pressure hull had failed long before reaching the bottom. Current Navy accounts describe Thresher descending beyond crush depth and imploding under the pressure of the sea.

What happened immediately before that remains less certain than it is sometimes presented.

The Navy’s Court of Inquiry favored a sequence beginning with failure somewhere in a saltwater piping system. Flooding could have led to an automatic reactor shutdown. Without propulsion, the submarine would have lost one of its principal ways of arresting the descent. An emergency main-ballast-tank blow was apparently attempted, but later testing suggested that moisture in the high-pressure air system could freeze as the air expanded, restricting flow when it was needed most.

That remains a reconstruction, not a recording of events inside the control room.

Naval History and Heritage Command still describes the exact sequence as debated. Acoustic researchers have raised questions about whether a major piping rupture would have produced sounds that apparently do not appear in surviving recordings. The Navy’s broader conclusion is safer: an uncontrolled casualty developed during the deep dive, Thresher could not recover, and the boat descended beyond the survival depth of the hull.

The evidence supports the broad sequence more strongly than any single initiating mechanism: Thresher developed an uncontrolled casualty during the dive, failed to recover, and descended beyond the pressure hull’s survival depth. The exact chain inside the submarine remains debated.

They investigated before they found the submarine

The Court of Inquiry opened the day after the sinking. It eventually heard approximately 120 witnesses, reviewed about 1,700 pages of material and examined hundreds of exhibits. When it finished on June 5, investigators still had not located Thresher itself.

That is a difficult circumstance in which to explain a machine failure. The thing being investigated was lying more than a mile and a half underwater.

Search ships photographed debris in May. The bathyscaphe Trieste, brought into the search, began making dives later that summer. In August its crew found pieces that could finally be tied conclusively to Thresher.

A sonar-dome section. A watertight door. Insulation. Draft markings. Pieces of piping bearing the boat’s identification.

The photographs are strange because the objects are recognizable without looking like a submarine anymore. A door is lying by itself on the ocean floor. Numbers painted to show waterline depth are still visible on broken wreckage thousands of feet below any waterline they could possibly measure.

The wreck had become a parts inventory.

The pipe

One recovered piece of brass pipe carried markings identifying it with boat 593.

It is exactly the kind of object historians are tempted to turn into a verdict.

I would not.

The investigation uncovered serious concerns about silver-brazed joints used in seawater piping, and later Navy accounts describe failure of such a joint as the probable initiating casualty. But a recovered pipe is not a recording of the moment the submarine was lost, and even the Navy’s own historical writing continues to acknowledge uncertainty about the exact sequence.

What matters more is what the Navy did once it stopped asking only which component failed and began asking a much harder question.

How do you know every critical component will work before a submarine dives?

That question produced SUBSAFE.

SUBSAFE made material traceability part of the submarine

SUBSAFE was created within months of the disaster to provide maximum reasonable assurance that a submarine’s hull will remain watertight and that it can recover from unexpected flooding. The program follows safety-critical systems through design, material selection, fabrication, installation, testing, maintenance, and certification. NAVSEA describes traceability reaching back to the original manufacture and processing of critical material, with records retained for the life of the submarine. The documentation is not administrative residue; it is part of the evidence required to certify that the submarine can safely dive.

The Columbia Accident Investigation Board later identified SUBSAFE as one of the safety programs worth studying when NASA was examining how high-risk organizations might control catastrophic failure. NASA and the Navy also conducted a formal benchmarking exchange on submarine safety-assurance practices.

The second problem

SUBSAFE addressed the question of how to keep the submarine from sinking.

Thresher exposed another problem at the same time. What happens if it sinks anyway?

The old rescue system had been built for earlier boats operating at shallower depths. The 1939 Squalus rescue remains one of the great technical successes in submarine history, but its survivors had been sitting 240 feet down. By 1963, Thresher could operate far below the 850-foot reach of the existing rescue chamber.

The Navy formed the Deep Submergence Review Group after the disaster to examine search, rescue and recovery. Its recommendations helped produce the Deep Submergence Systems Project in 1964 and a much broader investment in deep-ocean capability.

The search for Thresher itself accelerated that work. Deep-ocean vehicles and camera systems made visual inspections thousands of feet below the surface, and Navy accounts explicitly credit the search with improving deep-water detection and inspection techniques.

One submarine since

There is a statistic around SUBSAFE that deserves careful wording.

Since SUBSAFE was applied across the fleet, the Navy has lost one additional nuclear submarine, USS Scorpion, in 1968. Scorpion was not SUBSAFE-certified, and her loss remains unresolved. NAVSEA states that no SUBSAFE-certified U.S. submarine has been lost at sea.

That does not prove that submarines became invulnerable. They did not.

It says something narrower, and probably more valuable: a safety system born from one catastrophic failure has accumulated more than half a century without losing a certified boat to the sea.

The ship on the surface

The photograph I would put near the end is not one of the wreck. It is USS Skylark.

She is an ordinary-looking Navy auxiliary compared with the submarine beneath her. There is nothing futuristic about her silhouette. She was there because somebody had understood that a submarine doing dangerous work should have rescue capability nearby.

That idea was correct. The capability was simply from an earlier depth.

In 1939, a rescue bell had descended through 240 feet of water and brought thirty-three men home from Squalus.

In 1963, Skylark listened while a nuclear submarine worked down toward 1,300 feet and sent back broken pieces of a message.

Then the messages stopped.

The men on the surface tried to call again. They dropped signal grenades into the Atlantic.

Nothing answered.

---

Sources & Further Reading

• Naval Sea Systems Command, “USS Thresher: A Loss. A Legacy.”. The strongest current Navy overview of the final trials, SUBSAFE and lifecycle certification.
• Naval History and Heritage Command, USS Thresher (SSN-593). Official ship history and Navy probable-cause assessment.
• National Archives, “A Modern Submarine on Eternal Patrol”. Deck-log chronology, rescue-bell limitation and investigation history.
• Naval History and Heritage Command, “H-019-3: Navy Non-Combat Submarine Losses”. Important for uncertainty around the precise technical failure sequence.
• NHHC, history of deep-submergence rescue and recovery. The post-Thresher expansion of deep-ocean capability.