The Great Molasses Flood

On 15 January 1919, just after midday in Boston’s North End, a 50-foot (about 15-metre) steel tank on Commercial Street tore open. It held about 2.3 million U.S. gallons of molasses, roughly 8.7 million litres, waiting to be turned into industrial alcohol. A dense wave rushed into the crowded waterfront, wrecked buildings, damaged the elevated railway, and shoved the Engine 31 firehouse off its foundation. Twenty-one people were killed and about 150 were injured.

The tank had been finished in a hurry at the end of 1915 for Purity Distilling Company, a part of the United States Industrial Alcohol Company (USIA). The company’s man in charge of the project, Arthur P. Jell, was a financial administrator with no engineering training. Later testimony established that he could not read a blueprint. Because the tank was treated as a receptacle rather than a building, the city required a permit for the foundation, not a certified engineer’s sign-off on the tank itself. To save time before a molasses shipment arrived, Jell had the finished tank tested with only about six inches (15 centimetres) of water, not a full-height water test. The steel plates supplied by Hammond Iron Works were thinner than the plans called for.

The tank leaked from the time it was first filled. Neighbours, including children, collected the drips in pails. When people complained that they could see molasses seeping from the seams, the company had the tank painted brown so the leaks would not show. A sudden January warm spell followed very cold weather. The tank groaned; rivets shot out with a sound witnesses compared to gunfire; then the structure failed.

USIA blamed a planted explosive. After a long civil case and 341 days of testimony, court-appointed auditor Hugh W. Ogden ruled in 1925 that the tank had failed structurally and that USIA was liable. He rejected the sabotage claim. Boston’s building department began requiring that engineers’ and architects’ calculations be filed with their plans, and that drawings be signed. That change, and the sealed-plan and inspection habits that spread with it, is part of the flood’s official afterlife in Boston.

For decades afterward, people in the neighbourhood said that on hot days the streets still smelled sweet.

The first lesson is about ignored warnings and missing independent review. The leaks were visible for years. Painting the tank brown hid the symptom and left the structure unchanged. Nobody with the training to read the plans had signed off on the tank’s design, the steel was thinner than specified, and the only leak test used a shallow layer of water that could not stand in for a tank filled with millions of litres of a heavy fluid.

The second lesson is about viscosity: how thick a fluid is, and how that thickness changes with temperature. Molasses is a non-Newtonian fluid, meaning its resistance to flow is not a simple constant the way it is for water. In the temperature range of that January day, temperature mattered far more than fancier flow effects. Harvard researchers Jordan Kennedy and Shmuel Rubinstein, working with science communicator Nicole Sharp, measured blackstrap molasses and found that cooling it from 10°C to 0°C (50°F to 32°F) roughly tripled its viscosity, with even sharper increases as it got colder. A fresh, still-warm shipment from the Caribbean had been pumped into the tank only a couple of days earlier. Once the tank split, that warmer molasses could move with surprising speed. As evening came and the syrup cooled in the winter air, it thickened, which made swimming, walking, and pulling people out much harder.

A wave of molasses is also much denser than a wave of water, so it can knock down walls and hold a person in place after it stops rushing.