The Wave That Moved at 35 Miles an Hour

On a January afternoon in Boston, a five-story tank of molasses split open and buried a neighborhood. The company insisted it had been bombed. The truth took five years and a Harvard rheology study nearly a century later to fully unravel.

At around half past noon on January 15, 1919, the workers of Boston’s North End were doing what people do at lunchtime: eating sandwiches, running errands, sitting on stoops to catch a few minutes of an unseasonably warm winter day. At the Engine 31 firehouse on Commercial Street, several firefighters had settled in for a card game. Nearby, a 50-foot-tall steel tank belonging to the Purity Distilling Company groaned, as it often did when full. Nobody thought much of it. The tank had been groaning and leaking for years.

Then, without further warning, it tore itself apart.

What happened next sounds like something from a fable, if fables involved building collapse and asphyxiation. A wave of molasses — thick, dark, and by some estimates as much as 40 feet high — burst from the ruptured tank and rolled down the street at an estimated 35 miles per hour. It splintered the support beams of an elevated railway. It knocked buildings clean off their foundations. It picked up a truck and hurled it into Boston Harbor. By the time it stopped moving, it had killed 21 people and injured about 150 more, and it had done it in a matter of seconds.

This is the story of the Great Molasses Flood — a disaster so strange that people still have trouble believing it happened, a legal battle that helped rewrite how buildings get inspected in America, and a piece of physics that nobody fully explained until a Harvard graduate student ran the numbers nearly a hundred years later.

A Tank Built in a Hurry

The tank belonged to the Purity Distilling Company, a subsidiary of the United States Industrial Alcohol Company (USIA). Molasses, a byproduct of sugar refining, wasn’t destined for pancakes here — it was the raw material for industrial ethanol, which USIA distilled into alcohol for both munitions and, more lucratively, liquor. With World War I driving demand for both, and Prohibition looming on the horizon (the 18th Amendment had just been ratified that same January, giving companies a year to stockpile before the ban on alcohol took effect), USIA wanted molasses storage, and it wanted it fast.

The tank went up in 1915: 50 feet tall, 90 feet in diameter, built to hold 2.3 million gallons. It was constructed under the supervision of a company official named Arthur Jell, a treasurer and clerk with no engineering background, who was under pressure to get the tank finished before a shipment of molasses arrived. According to later investigation, the plans were never properly reviewed by a structural engineer, and the steel used was thinner than it should have been for a vessel meant to hold that much weight. Nobody performed a basic test — filling the tank partway with water first to check for leaks — before putting it into service.

From the beginning, the tank leaked. Local residents, including children, would reportedly scrape drips of molasses from its outer walls with sticks. Rather than investigate the leaks, the company’s fix was cosmetic: they painted the tank brown, so the seeping molasses would be less visible.

On January 12, 1919, three days before the disaster, a fresh shipment of warm molasses arrived from the Caribbean and was pumped into the tank, which was already holding a large reserve. This detail matters more than it might seem — it’s central to understanding why the disaster unfolded the way it did, as researchers would only fully establish decades later.

The Wave

Boston in mid-January is usually cold enough to freeze the harbor, but the days before the disaster had brought a sudden warm spell, with temperatures climbing into the 40s Fahrenheit after a hard freeze. Whether the temperature swing stressed the tank’s welds and rivets, or simply happened to coincide with the moment the aging structure finally gave out, the effect was the same. Shortly after 12:30 p.m., witnesses heard a sound like machine-gun fire — the rivets popping — followed by a deep rumble as the tank’s walls failed outright.

Two and a quarter million gallons of molasses hit Commercial Street all at once. The wave was reported at 15 to 40 feet high (accounts vary, but even conservative estimates put it well over a person’s height) and about 160 feet wide. It moved fast enough to be genuinely lethal on its own — the widely cited figure of 35 miles per hour comes from period estimates of the initial surge — tearing an elevated train trestle from its supports, crushing several buildings, and sweeping horses, wagons, and people off their feet.

Firefighters playing cards at Engine 31 were trapped when the firehouse’s second floor came down onto the collapsed first floor, pinning several men in a pocket beneath the wreckage; rescuers spent hours cutting through the debris to reach them, and at least one man died before he could be freed. Horses caught in the flood thrashed and became stuck fast as the molasses cooled and thickened around them; over the following days, rescue crews had to shoot several animals that couldn’t be freed. Workers combed the wreckage for a week, recovering the last body found more than four months after the flood.

The dead ranged widely in age and circumstance — laborers, a truck driver, a young girl gathering firewood, several city employees. Survivors described a scene that resisted easy description: streets and harbor water stained brown for months, rescue workers and horses coated in syrup, the smell of molasses hanging over the neighborhood for years afterward. A persistent local legend holds that on hot summer days, the North End still smells faintly of molasses. It’s the kind of detail that’s impossible to verify and is best treated as folklore rather than fact — but it captures how deeply the disaster lodged itself in the neighborhood’s memory.

“It Was Anarchists”

USIA’s public response was immediate and, it turned out, cynical. The company’s lawyers claimed the tank had been bombed — sabotage, they said, by anarchists targeting an industrial giant. It was not an outlandish accusation on its face: 1919 was a year of real anarchist bombings across the United States, part of the unrest that would culminate that same year in the bombing of U.S. Attorney General A. Mitchell Palmer’s home. Boston, with a large Italian immigrant population and simmering anti-immigrant sentiment, was fertile ground for the theory, and USIA’s lawyers leaned into it.

But the sabotage claim didn’t survive scrutiny. Victims’ families filed suit against USIA, and the cases were eventually consolidated into a single class action — at the time, the largest in Massachusetts history, with 119 individual claims rolled into one proceeding. A court-appointed auditor, Colonel Hugh Ogden, spent nearly six years hearing testimony from more than 3,000 witnesses and reviewing volumes of engineering evidence.

In 1925, Ogden issued his report. There had been no bomb. The tank had failed because it was badly designed and carelessly built: the steel was too thin, the factor of safety inadequate, the rivets and seams poorly engineered for the loads involved, and the whole structure had never been properly tested before it went into use. USIA was found liable, and the company ultimately paid out roughly $600,000 to settle the claims — worth well over $10 million today.

The Regulatory Aftermath

The Boston molasses disaster is now taught, half-jokingly, as a foundational case study in engineering ethics and structural safety — and the joke undersells how much actually changed because of it. In the wake of the disaster and the lawsuit, Massachusetts and other jurisdictions moved to require that engineering plans for large structures be reviewed and signed off by a licensed professional engineer, and that calculations for stress and safety margins be independently verifiable rather than taken on a company’s word. It’s one of the more direct throughlines from a specific disaster to a specific regulatory fix in American industrial history: a company cut corners building a tank, people died, and the response was to make “an unqualified clerk signed off on the math” structurally impossible to repeat, at least on paper.

What Took a Century to Explain

The auditor’s report settled the legal question of what went wrong: bad steel, bad design, bad oversight. But it didn’t fully explain something that had puzzled people for decades — why the disaster was quite so deadly, and why victims seemed to become trapped rather than simply swept along and eventually able to extract themselves, the way a person might struggle out of floodwater.

The answer, it turned out, was in the physics of the molasses itself, and it took a 2016 study — led by fluid dynamicist Nicole Sharp together with Harvard graduate student Jordan Kennedy and Harvard physicist Shmuel Rubinstein — using historical weather data and laboratory tests on real molasses to pin down. Molasses is a non-Newtonian fluid whose viscosity — its resistance to flow — changes dramatically with temperature. The batch that filled the tank in the days before the collapse had arrived warm from the Caribbean and hadn’t fully cooled to the ambient temperature of a Boston January. When the tank burst, that relatively warm, relatively fluid molasses surged out fast — fast enough to generate the initial high-speed wave that did so much structural damage in the first instants of the disaster.

But as the molasses spread out and lost heat to the cold January air and the ground, especially as temperatures fell after sunset, its viscosity increased sharply — by a factor of roughly four or more, according to the study’s modeling. Syrup that behaved almost like a fast-moving liquid in the first seconds of the flood became something closer to tar within minutes and hours. People and horses trapped in it weren’t just slowed down; they were caught in a substance actively thickening around them as rescuers worked, race against a molasses that was, quite literally, hardening in real time. The researchers concluded that if the same volume of molasses had failed in warmer weather, it likely would have spread farther but flowed thinner and drained away faster — behaving more like honey than tar — and probably would have killed and trapped significantly fewer people, even given an identical structural failure.

It’s a rare case where a legitimate scientific study, a century after the fact, adds real explanatory weight to a historical disaster rather than just novelty. The 1925 auditor could tell you the tank was badly built. It took fluid dynamics to explain why bad luck with the weather made that structural failure so much more lethal than it might otherwise have been.

A Marker, Finally

For decades, the site of the disaster carried no formal marker beyond word of mouth and local memory. A small plaque was eventually placed near the site — now occupied by a Boston Parks Department facility and a Little League field — commemorating the disaster, and it remains one of the more understated monuments to a mass-casualty event in American history. There’s no museum, no annual ceremony that draws national attention, just a plaque, a persisting urban legend about the smell of molasses on a hot day, and a case study that shows up in engineering ethics courses to this day.

It’s also, on reflection, a genuinely strange story to sit with: a company cut corners to beat a regulatory deadline (Prohibition), building a vessel to hold a substance nobody thinks of as dangerous, and the resulting failure killed more people than most building collapses of the era — not despite the material being molasses, but in some sense because of it. The cold turned a burst tank into a trap. Nobody designing that tank in 1915 was thinking about rheology. The people of the North End paid for that gap between engineering and physics, and it took the better part of a century for anyone to fully connect the two.