← back to the index
FR-012 brittle fracture

Boston Molasses Tank — A Brittle Steel Tank Burst and Drowned a Neighborhood in Syrup

Death toll
21 dead; ~150 injured
Structure
USIA molasses storage tank, 50 ft × 90 ft, ~2.3M US gallons, 529 Commercial St, North End, Boston
Failed
15 January 1919, ~12:30
Status
Broke up

Summary

At about 12:30 in the afternoon of 15 January 1919, on Commercial Street in Boston's North End, a 50-foot-tall riveted steel tank holding roughly 2.3 million US gallons of molasses split apart and discharged its entire contents in a wave that killed 21 people and injured around 150. The cause was not fermentation pressure, not sabotage, and not an explosion. It was a brittle fracture of the tank's thin, low-manganese steel shell, initiated at the over-stressed rivet holes beside a manhole at the base, propagating at the speed of sound through plates that were never thick enough and never inspected.

The tank belonged to United States Industrial Alcohol (USIA), through its subsidiary Purity Distilling, and had been thrown up in 1915 to store the molasses that fed wartime demand for industrial alcohol and munitions. It was 50 feet high and 90 feet in diameter and held about 13,000 short tons of liquid when full. The man who ordered and oversaw its construction, USIA treasurer Arthur Jell, had no architectural or engineering training. He did not commission stress calculations, did not have the steelwork checked by an engineer, and skipped the standard practice of filling the completed tank with water to test it under load — running only a few inches of water into the bottom rather than the full hydrostatic test that would have revealed the leaks before they revealed themselves.

The shell leaked from the day it was filled. Molasses wept from the seams so persistently that USIA painted the tank brown to disguise the runs, and neighborhood children collected the drips. At least one employee warned management that the structure was unsound; the company's response was to re-caulk. The forensic verdict, settled across a three-year court audit at the time and confirmed by modern fracture-mechanics analysis decades later, was unambiguous: the wall plates were roughly half as thick as the load required, the steel was deficient in manganese and therefore brittle, and the failure began as cracks at the rivet holes where stress concentrated. On a January day the metal was well below its ductile-to-brittle transition temperature, and a structure already loaded near its limit failed without warning and without yielding.

The disaster ended in one of the longest civil proceedings in Massachusetts history. A court-appointed auditor found USIA liable after years of hearings, and the company paid out about $628,000 in settlements — on the order of $11 million in present money. The litigation also did what the absent engineer never had: it forced the lesson into law. Massachusetts and then jurisdictions across the country began requiring that major structures be designed, signed, and sealed by a licensed engineer or architect, the regulatory ancestor of the professional stamp.

Timeline

1915
Tank erected without engineering oversight
USIA, anticipating wartime demand for industrial alcohol, builds a 50 ft × 90 ft riveted steel molasses tank at 529 Commercial Street. Treasurer Arthur Jell, with no engineering or architectural training, oversees the project and lets the contract with no independent design review.
1915
The hydrostatic test is skipped
Rather than fill the completed tank with water to test it under full load, Jell admits only a few inches of water before putting it into molasses service, omitting the standard proof test that would have exposed the leaking, under-strength shell.
1915–1918
Chronic leaking and a cosmetic fix
The tank leaks molasses from its seams from first fill. USIA paints the tank brown to hide the running streaks rather than investigate the cause. Local residents draw off the leaked molasses.
1917
USIA absorbs Purity Distilling
Ownership and operation of the North End tank pass to United States Industrial Alcohol, which continues to run it at or near capacity to supply alcohol and munitions feedstock.
1918 (approx.)
A worker's warning is ignored
An employee reports that the tank is structurally unsound; the company's only action is further caulking. Groaning and rumbling noises accompany each fresh filling.
1919-01-13
A warm load is delivered into a near-full tank
A shipment of warm molasses is added to the tank, which was already holding colder, older stock and was filled close to its working maximum, raising internal pressure on the already overstressed shell.
1919-01-14 to 15
A sharp temperature swing
Ambient temperature rises from roughly 2 °F to about 41 °F over the preceding day, but the steel of the loaded tank remains cold and brittle, well below the temperature at which the manganese-poor plate would behave in a ductile manner.
1919-01-15 ~12:30
Brittle fracture at the base
Cracks at the rivet holes beside a manhole near the bottom of the tank reach critical length. The shell tears open and the riveted plates fail in fast brittle fracture, rivets shearing off with a sound witnesses likened to machine-gun fire.
1919-01-15 ~12:30
The wave
About 2.3 million gallons — some 13,000 tons of molasses — surge out in a wave reported at 25–35 ft high moving near 35 mph, knocking buildings off foundations, buckling the elevated railway, and engulfing the street. Twenty-one people die, many by drowning or suffocation in the cooling syrup; about 150 are injured.
1920–1925
The class-action audit
Roughly 119 plaintiffs sue USIA. A court-appointed auditor, Hugh W. Ogden, presides over years of hearings; USIA blames an anarchist bomb, but the auditor finds the company liable for the structural failure.
1925
Liability and settlement
USIA is found responsible and pays approximately $628,000 in damages, with families of the dead receiving on the order of $7,000 each. The verdict drives new requirements for licensed engineering oversight of structures.
2014–2015
Modern fracture mechanics confirms the mechanism
Engineer Ronald A. Mayville of Simpson, Gumpertz & Heger applies finite-element and fracture-mechanics analysis, confirming the plates were at least 50% too thin, the steel was brittle from low manganese content, and the failure initiated as brittle cracking at the highly stressed rivet holes.

The Build — A Wartime Tank Ordered by a Treasurer

The structure that failed in 1919 was less a piece of engineering than a piece of expediency. In 1915 the demand for industrial alcohol — distilled from molasses and bound for munitions and the war economy — was rising fast, and United States Industrial Alcohol needed bulk storage on the Boston waterfront. The job of getting a tank built fell to Arthur Jell, the company's treasurer. Jell could read a ledger; he could not read a stress diagram. He had no training in architecture or engineering, commissioned no independent structural design, and arranged for no engineer to check the steelwork or the plate specification against the load it would carry. The tank rose at 529 Commercial Street as a 50-foot cylinder, 90 feet across, assembled from curved steel plates lapped and riveted together — a standard construction method, executed without the standard checks.

The single most important number was the wall thickness, and it was wrong. Modern analysis would establish that the plates were at least half as thick as the hoop stress of a full tank demanded, even by the relaxed standards of 1915. Compounding the thinness was the metallurgy. The plate was a low-manganese steel — the same broad family of brittle, manganese-poor steel implicated in the Titanic's hull — and manganese is precisely the element that gives structural steel its toughness and keeps its ductile-to-brittle transition temperature low. A steel short of manganese stays brittle to higher temperatures; overstressed in the cold, it cracks like glass rather than stretching like metal. The shell was therefore under-strength in two independent ways: too thin for the load, and made of a steel that could not tolerate a flaw. The riveted joints added the third ingredient. Every rivet hole is a stress raiser, a circular notch where the membrane stress crowds and multiplies, and the holes beside the base manhole were the most heavily loaded of all. The tank was, in effect, pre-cracked at hundreds of points, built of a material that could not arrest a crack, sized to a load it could not safely hold. When Jell skipped the hydrostatic proof test — admitting a few inches of water instead of the full water fill that would have stressed the shell to its working load on the ground, harmlessly — he discarded the one chance to discover all of this before molasses replaced water.

The Failure Sequence — Brittle Fracture at the Speed of Sound

The tank announced its condition from the first fill. Molasses seeped through the riveted seams continuously; USIA's answer was cosmetic, a coat of brown paint to make the leaks invisible against the tank's own color. Residents treated the weeping shell as a neighborhood molasses dispenser. The leaks were not a nuisance to be hidden — they were strain gauges, reporting that the joints were working loose under a load the structure was never sound enough to carry, and the groaning each time the tank was topped up carried the same message. None of it was read as warning, because no one with the training to read it was looking.

On 13 January 1919 a delivery of warm molasses was run into the tank, which was already standing near its working maximum and holding colder stock. Over the following day the air temperature climbed from near 2 °F to about 41 °F, but a 50-foot column of cold molasses and the cold steel containing it do not warm in a day; the plate stayed well below the temperature at which its manganese-starved metallurgy would let it yield rather than fracture. The shell was now loaded near its limit, brittle, and notched at every rivet. Shortly after 12:30 on 15 January, the cracks that had been growing at the rivet holes beside the base manhole reached critical length. There was no slow bulge, no progressive sag — the hallmark of brittle fracture is that it gives none. The shell unzipped, the fracture running through the plate near the speed of sound, and the rivets sheared and flew with reports that bystanders described as gunfire. In an instant the bottom of the tank was open.

What came out was not a spill but a flood. Roughly 2.3 million gallons — about 13,000 tons — of molasses released at once, forming a wave reported at 25 to 35 feet high and moving at perhaps 35 miles per hour. It tore a firehouse from its footings, snapped the supports of the elevated railway, and filled cellars and streets. Twenty-one people died, most drowned or suffocated as the molasses cooled and thickened around them; about 150 more were injured. The lethality was a compound of the fracture — which dumped the full inventory in seconds rather than letting it drain — and the fluid itself, which set into a viscous trap as it cooled.

The Reckoning — A Bomb Defense, an Auditor's Verdict, and a Stamp

The legal aftermath ran for years and turned on the cause of the rupture. United States Industrial Alcohol mounted a defense that the tank had been destroyed by a bomb planted by anarchists — a plausible-sounding claim in the era of the Galleanist bombings, and one that would have absolved the company of any structural negligence. Some 119 plaintiffs pressed a consolidated suit. The matter went before a court-appointed auditor, Colonel Hugh W. Ogden, who heard from scores of witnesses and competing experts across roughly three years of hearings. The physical evidence did not support sabotage: there was no blast signature, and the failure pattern was that of an overstressed, under-built shell. Ogden found USIA responsible for the disaster, and in 1925 the company was ordered to pay damages, settling for about $628,000 — on the order of $11 million today — with families of the dead receiving roughly $7,000 apiece.

For decades the engineering verdict remained essentially that of the auditor: a structurally inadequate tank that failed under its own load. In 2014 and 2015, engineer Ronald A. Mayville of Simpson, Gumpertz & Heger reopened the case with modern tools, applying finite-element stress analysis and fracture mechanics to the surviving record. His conclusion sharpened the original finding into a precise mechanism. The wall plates were at least 50 percent too thin for the load. The steel was brittle because it was low in manganese. The stresses at the rivet holes were too high, and that is where cracks first formed; the fracture most likely originated at the manhole near the base, where a directly overhead rivet hole was severely overstressed. On a cold January day the brittle steel was below its transition temperature, and the structure failed by fast brittle fracture rather than by any ductile, gradual mode that might have leaked and warned before it broke. The bomb was never real. The tank had been failing in plain sight for four years.

Contributing Factors

01
A major structure built with no engineering oversight
The tank was specified, contracted, and accepted under a company treasurer with no engineering or architectural training, and no independent engineer ever checked the design or the steel. Bulk-storage structures concentrate enormous energy; their adequacy cannot be assumed from the fact that they stand up empty. A pressure-bearing or load-bearing structure designed without a qualified engineer's calculation and sign-off has not been engineered at all, only assembled.
02
Wall plates under-thickness for the hoop stress
The shell plates were roughly half the thickness the full-tank load required, even by 1915 standards, leaving the steel chronically overstressed in service. Thickness in a cylindrical tank is not a margin to be trimmed for cost — it is the direct resistance to the hoop stress that the contents impose every hour. Sizing a pressure boundary below its governing load guarantees that some flaw, somewhere, will eventually find the strength it was denied.
03
A brittle, low-manganese steel operated below its transition temperature
The plate was deficient in manganese, the element that confers toughness and lowers the ductile-to-brittle transition temperature; on a cold day the metal could not yield and instead fractured. Material toughness is not a detail separate from strength — a steel that is strong but brittle will shatter at a crack a tough steel would tolerate. Specifying a structural steel without controlling its toughness and its service-temperature behavior leaves the structure one cold day away from catastrophic fracture.
04
Rivet holes as unaccounted stress concentrations
The fracture initiated at the rivet holes beside the base manhole, where the membrane stress crowded around the bores to local peaks far above the nominal wall stress. Every hole, notch, and discontinuity in a loaded plate multiplies the stress beside it; in a brittle material those peaks become crack starters. A design that checks only the average wall stress and ignores the concentrations at its own connections has computed the wrong number.
05
Visible warnings dismissed as cosmetic
The tank leaked from the seams for four years and groaned when filled; the response was brown paint, re-caulking, and an ignored worker's report. Persistent leakage and noise from a loaded structure are the structure reporting distress, not a maintenance annoyance to be concealed. Treating the symptoms of overload as a cosmetic problem suppresses the only warning a brittle structure will give before it fails without warning.

Aftermath

The toll — 21 dead and about 150 injured, a North End neighborhood flooded and an elevated railway buckled — made the Great Molasses Flood one of the defining industrial disasters of early-twentieth-century America. Its largest legacy was regulatory. The three-year audit and the 1925 liability finding established, in a way no prior failure had, that a structure of consequence could not be safely left to an untrained owner and an unchecked builder. In its wake Massachusetts and then jurisdictions across the United States tightened construction law to require that significant structures be designed and certified by a licensed, registered engineer or architect — the regime that produced the professional engineer's seal, the stamp now affixed to structural drawings as the legal signature of a qualified person who has done and stands behind the calculations. The forensic record, reopened by modern fracture-mechanics analysis nearly a century later, was confirmed in every particular: thin plate, brittle low-manganese steel, overstressed rivet holes, brittle fracture in the cold. In engineering memory 'the Boston molasses flood' is the byword for the catastrophe that follows when a load-bearing structure is built with no engineer, no toughness, and no proof test — and for the reason a stamp is now required before steel is allowed to hold anything that can drown a street.

Lessons

  1. Require a qualified engineer's design and sign-off for anything that stores energy: never let an untrained owner or treasurer specify, accept, or proof a load-bearing or pressure-bearing structure — the seal exists because someone has to be competent and accountable for the calculation.
  2. Size the pressure boundary to its governing load, not its budget: set wall and plate thickness from the hoop stress of a full structure with proper margin, and treat any proposal to trim that thickness as a proposal to operate above the steel's strength.
  3. Specify toughness and transition temperature, not just strength: choose and control structural steel for its behavior at the coldest service temperature, because a strong but brittle material will fracture at a flaw a tough one would shrug off.
  4. Design for the stress concentrations you build in: check the peak stress at every rivet hole, notch, and connection — not the average wall stress — because a brittle structure will crack where the stress crowds, and those are the points your design created.
  5. Read leaks and noise as structural reports, never cosmetics: when a loaded structure seeps, groans, or is reported unsound, investigate the cause to failure — painting over the symptom removes the only warning a brittle structure gives before it breaks all at once.

References