The airplane carried a presidential name and a fragile pedigree. Registered N90773, the Douglas DC-6B had first flown in 1953, eight years before the crash - middle-aged for a piston airliner of its generation, with four Pratt & Whitney R-2800 radial engines hung from a graceful straight wing. Her owner, President Airlines, was a supplemental air carrier, the regulatory category that covered the hybrid charter-scheduled operators of the early jet age. In 1960 President Airlines had bought the operating certificate of California Eastern Airways, a carrier that flew mostly under contract for the U.S. government. The accident at Shannon would, in effect, end the company. President Airlines went out of business shortly after. That September night they were flying a non-scheduled passenger service from Düsseldorf to Chicago, with refuelling stops at Shannon and Gander. Seventy-seven passengers had boarded. Six crew members were on duty.

The takeoff from Shannon's runway 24 went as it should. Air traffic control gave the customary instruction for a right turn after departure - standard noise abatement, standard routing, a thousand pilots had done it before. The crew turned left. Then they kept turning. Investigators reconstructing the final seconds concluded the aircraft passed through a steep bank, then a steeper one, the wings rolling toward vertical. By the time the bank angle reached ninety degrees, lift was no longer holding them up. The DC-6B fell, still rolling, and struck the broad cold surface of the River Shannon roughly 5,000 feet from the end of the runway. It is the kind of distance a Douglas can fly in seconds. It was enough.

The subsequent inquiry could not deliver a single, clean cause. What it could do was offer a list of possibilities, each plausible, none provable in isolation. A malfunctioning attitude indicator may have shown the crew level flight when they were already rolling - a deadly illusion in night conditions over dark water with no horizon. A fault in the starboard ailerons may have rolled the aircraft against the crew's intentions. Poor weather, with low visibility and possible cloud, would have removed the outside cues that might have caught the error. Crew fatigue, after a long European leg and a short Shannon turnaround, sat in the background as a contributing factor. The accident shares this taxonomy with later disasters - Air India Flight 855 in 1978, lost in a similar attitude-indicator failure off Bombay - in which a faulty instrument in the dark became a death sentence.

What the numbers do not tell you is the rest. The passengers were travelers - workers, families, people in transit between the Old World and the New, making the kind of long, multi-stop crossing that was still ordinary for the era of piston-engined transatlantic flight. Among the crew were cabin staff who had served meals over the North Sea hours earlier. Local fishermen and boatmen at Shannon were among the first to reach the wreckage; what they found in the cold tidal water of the estuary that night stayed with them. One cabin crew member was pulled alive from the river - the only person to survive the impact - but died of injuries within hours. The Irish state mourned. The aviation community on both sides of the Atlantic asked the questions that follow every disaster of that scale, and the answers, here, came back ambiguous, multiple, and slow. The bodies came home. The airline did not survive. The river, indifferent, kept running to the sea.

Stand today at the end of runway 24 and the geography that killed Theodore Roosevelt is still visible. The flat estuary opens to the west, with no terrain to guide the eye, no lights at night except the airport's own. Shannon's air traffic procedures, like those at every major airport, have been rewritten and rewritten again in the decades since 1961. Attitude indicators became dual, then triple-redundant. Pilot training added crew resource management, fatigue rules, and explicit drills for the dark sensory void of an overwater takeoff. None of it can undo the deadliest accident in Ireland's aviation history. All of it exists, in part, because of nights like 10 September 1961.