Galloping Gertie: The Bridge That Twisted Itself Apart — And Rewrote How Engineers Build With Wind

A Bridge That Danced From the Day It Opened

The Tacoma Narrows Bridge was supposed to be an achievement, not a warning. When it opened on July 1, 1940, it was the third-longest suspension span in the world, its 2,800-foot main deck hanging between two towers 425 feet tall over the tidal strait separating Tacoma from the Kitsap Peninsula. It cut a two-and-a-half-hour ferry-and-drive commute down to an eleven-minute crossing, and it was $1.6 million cheaper than the original design because chief engineer Leon Moisseiff — a respected consultant who had also worked on the Golden Gate and Manhattan Bridges — proposed replacing the deep, wind-permeable truss originally planned for the deck with two slender, solid steel plate girders, just 8 feet deep supporting a deck only 39 feet wide. It made the bridge dramatically lighter, more elegant, and cheaper to build. It also made it, structurally, one of the most flexible long-span bridges ever constructed.

Workers noticed the problem before a single car had crossed. In anything more than a light breeze, the deck rippled visibly, rolling in slow, rhythmic waves several feet high — enough that drivers ahead would disappear and reappear as the road undulated beneath them. Construction crews nicknamed it "Galloping Gertie" almost immediately. Rather than alarming the public, the motion became a tourist attraction; thrill-seekers drove out specifically to feel the bridge sway, and engineers, confident the vertical bouncing posed no real danger, installed hydraulic dampers and monitoring equipment to study — not stop — the motion. For four months, Gertie galloped, and nobody outside a small circle of engineers thought it was a structural emergency.

That last number — the ratio of girder depth to span, 1-to-350 — is the whole story in a single fraction. It was, at the time, identical to the ratio used on the George Washington Bridge, a span widely considered one of the great engineering successes of the era. On paper, Gertie wasn't reckless. It followed the same "deflection theory" Moisseiff had used successfully elsewhere: let a bridge flex with the wind instead of rigidly resisting it, and use the tension in the cables themselves to restore the deck to shape. What nobody had tested — because nobody yet had the tools to test it — was what happened when that same slender, solid-faced deck stopped just bouncing, and started to twist.

It Wasn't Resonance. It Was Something the Textbooks Hadn't Named Yet

For decades afterward, physics classrooms taught the Tacoma Narrows collapse as a textbook case of mechanical resonance — wind gusts happening to match the bridge's natural frequency, pumping in more and more energy with each cycle, the way pushing a swing at exactly the right moment sends it higher and higher. It's a tidy explanation. It's also not quite what happened, and modern structural engineers are careful to draw the distinction, because the real mechanism is more unsettling: self-excited aeroelastic flutter.

Simple resonance needs a matching external rhythm — a wind that gusts in time with the bridge. Gertie didn't have that. Instead, the deck's own motion changed the airflow around it, and that changed airflow fed energy directly back into the motion, regardless of whether the wind itself was steady or gusting. Once a sustained wind above roughly 35 mph got the deck twisting even slightly, the twisting motion itself began extracting more energy from the passing air than the bridge's structure could dissipate — a negative damping effect, where instead of a disturbance fading out over time, it fed itself and grew. There was no upper limit built into the physics. The oscillation simply kept amplifying until something tore.

The Last Four Hours of Galloping Gertie

The morning of November 7, 1940 started unremarkably: a stiff but unexceptional wind blowing broadside into the bridge's solid deck. Around 7:30 a.m., with wind measured at 38 mph, Gertie began its familiar vertical gallop — waves two to five feet high, the ordinary motion the bridge had displayed for months. That motion held steady for roughly three hours. Then, close to 10:00 a.m., something changed: a mid-span cable connection slipped, throwing the load out of balance, and the up-and-down bouncing shifted abruptly into the violent torsional twisting nobody had seen before. Professor F. B. Farquharson, an engineering researcher from the University of Washington who had been studying the bridge's motion for months, was on site that morning and captured what followed on film.

Coatsworth's car was the last vehicle on the span when the twisting began in earnest; he escaped on foot, and Tubby did not. Just after 11:00 a.m., with wind now sustained at 42 mph, sections of the roadway were tilting up to 45 degrees from horizontal. Concrete began cracking loose from the deck and dropping into the water below. Suspender cables, never designed for this kind of load, started snapping. At 11:02 a.m., a 600-foot section of the center span tore free and fell into Puget Sound. Radio reporters who rushed to the scene captured the reaction live; one Washington state senator watching from the shore called it "the most astounding sight I have ever witnessed in my lifetime." Nobody was hurt in the final collapse — Coatsworth's dog was the only fatality of the entire disaster.

The Bridge That Learned to Let the Wind Through

The collapse triggered something that had never really existed before: a systematic, experimental science of how wind interacts with flexible structures. An Advisory Board on the Investigation of Suspension Bridges convened leading American bridge engineers from 1942 to 1954, and Farquharson built a dedicated wind-tunnel laboratory at the University of Washington large enough to test scale models of an entire proposed bridge, up to 100 feet long, alongside smaller section models of individual deck designs. Every candidate replacement was built as a model first and subjected to controlled airflow before a single piece of full-scale steel was ordered — a discipline that, before 1940, essentially did not exist in American bridge engineering.

The replacement Tacoma Narrows Bridge opened on October 14, 1950, reusing the original towers and cable anchorages but replacing the solid plate girders with a deep, open steel truss that let wind pass through the deck instead of piling up against a flat face — visible proof of the diagram above, built at full scale. Locals nicknamed it "Sturdy Gertie." It was, by contemporary calculation, 58 times more rigid in torsion than its predecessor, and it has carried traffic safely for over seventy years, joined by a parallel companion span in 2007 to handle modern traffic volumes.

The effects reached far beyond Tacoma. Existing suspension bridges around the country were re-inspected specifically for aerodynamic vulnerability; New York's Bing­hamton-area and Whitestone Bridge, for example, had open trusses and deck slots retrofitted to break up the same kind of wind-induced motion. Wind-tunnel testing of scale models became, within a couple of decades, a standard and eventually mandatory part of designing any major bridge — a practice every landmark long-span bridge built since, from the Severn Bridge's pioneering aerodynamic box-girder deck in the 1960s to Japan's Akashi Kaikyō Bridge and France's Millau Viaduct, still relies on directly. The failure of one bridge in 1940 is, in a very literal sense, why every long-span bridge opened since has been tested against the wind before it was ever built.

Why a 1940 Bridge Collapse Still Matters for a House Foundation Today

Almost nobody planning a construction project on Estima's estimator is designing a 2,800-foot suspension span. But the mistake sitting underneath Galloping Gertie's collapse is a scaled-up version of one that shows up in ordinary residential and commercial construction constantly: engineering for the structural load you expect, without testing what actually happens when a structure moves under that load in the real world. Roof trusses, wide-span carports, exposed decks, and lightweight metal roofing all interact with wind in ways that simple static calculations can understate — which is exactly why building codes today specify wind-uplift ratings, bracing requirements, and connection hardware that have nothing to do with a structure's strength standing still, and everything to do with how it behaves once it starts to move.

The deeper lesson isn't really about bridges, or even about wind specifically — it's that the most dangerous gap in any project is the one between "this meets the calculation" and "this has actually been tested against how it will really behave." Gertie's engineers weren't reckless; they followed an established, previously successful design theory and hit a genuine, unmapped edge of what was then known about aerodynamics. The honest response to that kind of uncertainty, at any scale of construction, is the same one Farquharson's wind tunnel eventually forced onto the entire bridge-building industry: build in the margin, and verify it, before the wind ever gets to test it for you.