On November 28, 1942, a fire in the Cocoanut Grove nightclub in Boston killed 492 people by the most commonly cited count, which makes it one of the deadliest building fires in American history. The main entrance was a single revolving door, several other exits were locked or hard to find, and the interior was dressed in decorations that carried flame across the ceilings in minutes. The fire changed how codes treat places of assembly. It also changed how Massachusetts General Hospital treated burns and smoke-damaged lungs. This guide covers what the record supports, and where it does not settle the ignition or the exact toll, it says so.

November 28, 1942: the building, the crowd and the first minutes

The Cocoanut Grove stood on Piedmont Street near Park Square, in the part of Boston now usually called Bay Village. By 1942 it was one of the best known nightclubs in the city. Its owner, Barnett Welansky, was in the hospital that night, according to most accounts. The club had grown by addition over the years. The main floor held a dining room with a stage, a bar, and a newer Broadway Lounge. A basement cocktail room called the Melody Lounge was reached by a narrow stairway that came up into the foyer near the front door. The interior followed a tropical theme, with artificial palm trees, fabric hung across the ceilings, and a good deal of rattan and bamboo.

The fire fell on the Saturday after Thanksgiving, in the first winter of the war. That afternoon Holy Cross had beaten a heavily favored Boston College football team, and the city was full of people going out. Histories of the fire commonly put the crowd that night at more than 1,000 people, in a building whose legal capacity is usually given as about 460. Those two figures are the ones most often repeated. I have not seen either reconciled against a primary count, so the crowd figure in particular should be read as an estimate rather than a measurement.

The fire began in the Melody Lounge at a time commonly given as about 10:15 p.m. It spread through the basement decorations, came up the stairway into the foyer, and moved across the main floor into the dining room and lounges within minutes. Survivors and the investigations described flame running along the fabric ceilings ahead of people trying to leave, followed by dense smoke and hot gas. The lights failed early. Most of the dead were found inside, many of them at or near exits that would not let them out. The bodies piled behind the revolving door on Piedmont Street became the image most associated with the fire. I had no role in any of this, and everything here comes from the published record: the investigations, the criminal trial, and the medical literature written afterward.

492 and the figures beside it: why the toll is a narrow range

The figure of 492 dead is the one the National Fire Protection Association has used in its accounts of the fire, and it is the number most histories and memorials repeat. Some published accounts give 490 or 491. The difference is small but real. The count includes people who died in hospitals in the days and weeks after the fire, and it was assembled in 1942 from hospital, morgue and police lists. Small discrepancies between published totals of that kind are common in disasters of the era. I have not found a published reconciliation that settles which figure is correct. The honest way to cite the toll is 492 by the most commonly used count, with some accounts giving a figure one or two lower.

The number of injured survivors is less settled than the death toll. Different sources count it in different ways, often as several hundred, and I would not cite a single figure without naming the source. Two hospitals carried most of the load. Boston City Hospital received roughly 300 victims, by the figure usually given. Massachusetts General Hospital’s own accounts put the number it received at 114, of whom most were dead on arrival or died shortly afterward. About 39 were treated there as surviving patients. Anyone writing a training slide should take those hospital figures from MGH’s published histories and the 1943 medical literature rather than from secondary summaries, which round them in different directions.

For a planner, the shape of those numbers matters more than the last digit. A fast fire in a crowded enclosed space with blocked exits produced a large number of deaths before anyone reached a hospital. Of the people who did arrive, a large share were dead or beyond help. The survivors were a smaller group of critically injured patients, many with damage to the airway and lungs on top of their skin burns. A mass casualty plan that assumes a burn incident will arrive as a steady flow of treatable patients does not match what Boston saw that night, and the Grove is one of the clearest historical illustrations of the different pattern.

The busboy and the match: an ignition story the record does not settle

The story most people have heard holds that a 16-year-old busboy named Stanley Tomaszewski started the fire. That story is disputed and was never established as the cause. In the usual telling, a patron in the Melody Lounge had removed a light bulb from a fixture in an artificial palm to darken a corner. Tomaszewski was sent to put it back, lit a match to find the socket in the dark, and the decorations caught. Tomaszewski acknowledged lighting a match. He said he put it out before anyone saw flame, and he was never charged.

The Boston Fire Department investigation under Fire Commissioner William Arthur Reilly did not identify a cause of ignition. As I read the commissioner’s report and the histories that summarize it, the origin was left undetermined, while the report dealt at length with the building conditions that turned a small fire into a mass casualty event. In the decades since, researchers and writers have advanced other hypotheses. These include faults in electrical work done at the club and a leak of a flammable refrigerant gas from the cooling system. Each is a hypothesis put forward by people who had no new physical evidence to test against the scene, and none of them is a finding. A reader who wants the primary record should start with the commissioner’s report and the Massachusetts court record, both of which are held by Boston archives and discussed in the major histories of the fire.

The ignition question is worth getting right for two reasons. The first is fairness, because a teenager carried public blame for decades on the strength of a story the investigation did not adopt. The second matters more for anyone in fire prevention. The ignition source had very little to do with the death toll. A match, an electrical fault or a refrigerant leak could each have started a small fire in that basement. What killed hundreds of people was what happened next: fuel on the walls and ceilings, a stairway that carried the fire to the main floor, and exits that would not let a crowd out. Training that dwells on the busboy teaches the wrong lesson, because occupancy hazards and egress are what an inspector can actually control.

A revolving door and locked side doors: how the exits failed

The main entrance on Piedmont Street was a revolving door, and most patrons had come in through it. When the fire reached the foyer, the crowd pushed toward the door they knew. A revolving door moves people only when they push in turns, one compartment at a time. A crowd pressing from inside jammed it, and bodies piled behind it. Firefighters later found people stacked at that entrance. Accounts of the fire describe a swinging door near the revolving door that was not usable, though descriptions of it vary between sources.

The other exits did little better. According to the investigation and trial record as the histories summarize it, some doors were locked or bolted, apparently to stop patrons from leaving without paying, and some were concealed by decor or poorly marked. The door from the Broadway Lounge opened inward. When a crowd presses against an inward-opening door, the crowd itself holds it shut. Some people escaped through the kitchen and service areas, and some through windows. Firefighters broke through openings to reach people inside. The behavior was consistent with what fire researchers have since described in many crowd fires: people head for the way they came in, even when other exits exist, and those other exits help only if they are visible, unlocked, and open in the direction of travel.

The criminal case focused on these conditions rather than on the ignition. Welansky was convicted of involuntary manslaughter, and in 1944 the Massachusetts Supreme Judicial Court upheld the conviction in Commonwealth v. Welansky. The court’s reasoning on wanton or reckless conduct, based largely on the exit conditions, is still cited in Massachusetts law. Welansky received a sentence usually given as 12 to 15 years and was released after serving part of it because of illness. According to accounts of the investigation, a Boston Fire Department inspection shortly before the fire had found the club satisfactory. That detail is worth holding onto when this guide turns to enforcement, because the gap between an inspection on a weekday afternoon and the building at 10:00 on a Saturday night is where the Grove’s exits failed.

Cloth ceilings and artificial palms: interior finish as fuel

Much of the Cocoanut Grove’s interior was combustible, and some of it was hung where it would do the most harm. Fabric stretched across the ceilings put fuel overhead, where hot gases collect, so flame could travel the length of a room above the heads of people who had not yet seen fire. The artificial palms, rattan and bamboo added fuel at the level where people were standing and sitting. The narrow stairway from the Melody Lounge acted as a flue, carrying heat and smoke straight into the foyer by the main entrance. The club’s side claimed that decorations had been treated to resist flame. Histories of the investigation record serious doubt about whether any treatment had been applied effectively or had lasted, and I would not state either side of that as settled.

The speed of the spread is one of the reasons the ignition debate persists. Some later writers argue that ordinary decorations could not account for how fast the fire crossed the main floor, and they offer a flammable gas as the explanation. Others argue that a fabric ceiling over a crowded room, fed by a stairway acting as a chimney, explains the speed without help. That disagreement is a dispute among people working from the historical record without new evidence, and it remains unresolved. Either way, the lesson for code purposes is the same. Combustible decorative material on ceilings and walls in an assembly occupancy can carry fire faster than a crowd can move, and the codes that followed were written on that basis.

Modern codes handle this in two ways. Interior finish, the permanent surfaces of walls and ceilings, is classified by how fast flame spreads across it, and assembly occupancies are restricted to the better classes. Decorative materials such as curtains, drapes and hanging fabric are required to be flame resistant, usually by passing NFPA 701, the standard test for flame propagation of textiles and films. Topical flame retardant treatments can wash out, wear off, or fail with age. A certificate from the day the drapes were installed says nothing about the drapes five years and several cleanings later. The Station nightclub fire in West Warwick, Rhode Island, in 2003, which killed 100 people, began when pyrotechnics ignited foam on the walls around the stage. That is a reminder that interior finish has stayed a live problem in nightclubs long after 1942.

What changed in the codes, and the pattern of code following catastrophe

The Grove is usually taught alongside a short list of assembly fires that each moved the codes. The Iroquois Theater fire in Chicago in 1903 killed more than 600 people by the usual count and drove the spread of outward-opening exit doors and panic hardware. The Cocoanut Grove came next in the sequence. The Beverly Hills Supper Club fire in Southgate, Kentucky, in 1977 killed 165 people in a crowded club with egress problems that many writers compared to Boston’s. The Station fire in 2003 led the NFPA to expand sprinkler requirements for nightclubs. I am not claiming that each fire caused one specific change. The point is that the record of assembly occupancy regulation in the United States is largely a record of response to mass casualty fires, and the changes tended to come in the years right after each one.

After the Grove, Massachusetts and many other jurisdictions tightened their rules for places of assembly, and the NFPA revised its Building Exits Code, the document later renamed the Life Safety Code. The themes of the changes followed what had failed in Boston. They restricted revolving doors and required hinged doors beside them. They required exit doors to open in the direction of travel and to be operable from inside without a key. They called for exit signs and emergency lighting that would keep working when normal power failed, and for limits on combustible decorations. Posting the permitted occupant load became the normal practice for assembly spaces. Today’s model codes give a revolving door limited credit toward required exit capacity and require it to collapse or to have swinging doors beside it. The details vary by edition, so check the edition your jurisdiction actually adopted rather than relying on a summary.

The pattern has a mechanism that emergency managers should understand, because it explains why the lessons arrive slowly. A model code committee revises a document on a fixed cycle. A state or city then adopts an edition, often years later and sometimes with amendments, and existing buildings are frequently held to an existing-occupancy standard that is less demanding than the one for new construction. After the Grove, the political pressure was strong enough to move this quickly in Massachusetts. Elsewhere it moved at the usual pace. The interval between the Grove and Beverly Hills, 35 years, shows that a code change on paper does not by itself prevent a repeat in a building that crowds past its limit, decorates with combustibles, or locks its side doors after the inspector leaves.

Massachusetts General: soft dressings, fluids, damaged lungs and grief

Massachusetts General was in some respects ready. Like other large hospitals in coastal cities, it had prepared during 1942 for the possibility of air raid casualties, and that work included organizing staff, supplies and plasma for a sudden influx. The surgeons who led the burn care that night were Oliver Cope and Francis D. Moore. They and their colleagues published a series of papers on the care of the Grove victims in 1943, the best known as a symposium in the Annals of Surgery. Those papers are the primary source for the clinical lessons, and anyone citing a figure about MGH’s patients should take it from them.

The best known change was in how the burn wounds themselves were treated. The standard treatment at the time was tannic acid, applied to form a hard coagulated crust over the burn. Cope’s team instead used soft dressings of gauze with a petrolatum-based ointment, left largely undisturbed. They reported results they considered better, and their experience helped move burn care away from tannic acid. The team also paid close attention to fluid replacement, giving large volumes of plasma to patients who were losing fluid from extensive burns. Moore went on to work on calculating fluid requirements in burns, one of several lines of work that came before the resuscitation formulas clinicians learn now. Merck supplied penicillin to MGH for the Grove patients, an episode commonly described as one of the earliest substantial civilian uses of the drug. That description appears widely, but I have not verified it against a primary source.

The MGH papers also gave serious attention to damage to the airway and lungs from heat and smoke. Many patients who reached the hospital alive had injuries to the respiratory tract that shaped whether they survived, independent of the size of their skin burns. As a physician assistant, I think that is the clinical lesson from the Grove that has aged best. A patient pulled from an enclosed-space fire has to be assessed for airway and inhalation injury from the first contact, and a burn mass casualty plan has to account for patients who need airway management and ventilation as well as wound care.

The fire also produced a foundational study in psychiatry. Erich Lindemann, a psychiatrist at MGH, studied survivors and bereaved relatives, among other groups. His 1944 paper on the symptoms and management of acute grief drew in part on that work and became one of the founding texts of crisis intervention. The responder behavioral health practice that agencies now take for granted traces part of its lineage to his work with the families of people who died at the Grove.

The code on paper and the door on a Saturday night: enforcement today

The Grove’s exits failed under the conditions of that particular night, with a crowd far above the posted capacity, side doors locked against walkouts, and decorations hung for the season. An inspection done days earlier had not seen those conditions, or had not recorded them as violations. The same gap exists in most jurisdictions today. A fire inspector visiting a bar or event hall during business hours on a weekday sees an empty room with its doors unlocked because nobody is there to walk out without paying. The building that matters is the one at 11:00 p.m. on a Saturday, with a band, a line at the door and a manager who has chained a side door because of theft.

Some fire marshal offices run night inspection programs aimed at exactly this, visiting assembly occupancies during operating hours to count occupants against the posted load, check that exits are unlocked and unobstructed, and look at decorations and pyrotechnics. The authority to do this, and the authority to order an overcrowded room reduced on the spot, comes from the fire code your jurisdiction adopted and from state law. It varies, so confirm it with your fire marshal and your jurisdiction’s attorney before building a program around it. An enforcement program that exists only during business hours inspects a building that does not match the one that burns.

There is also a response side. Dispatch and first-arriving crews can know in advance which buildings in their area hold large crowds at night, what the posted occupant load is, and where the exits are, if that information is in the premise records attached to the address. During my years in fire rescue communications and quality assurance, premise information was one of the things that most often turned out to be stale when it was finally needed. A premise record that lists a club’s occupant load and secondary exits, and was verified recently, gives the first officer something to plan with during the minutes when the crowd is still coming out the front.

What to do at your agency

• Have your fire marshal pull the list of assembly occupancies with a posted occupant load over 100 and confirm, for each one, the date of its last inspection and whether that inspection took place during operating hours.
• Have your fire marshal or a designated inspector conduct one after-hours walk-through this month at a single nightclub or event venue, counting occupants against the posted load and recording the condition of every exit door, and write the results up for the fire chief.
• Have your training officer add a 20-minute Cocoanut Grove segment to the next scheduled fire prevention or company inspection class, presenting the ignition as undetermined and spending the time on revolving doors, inward-swinging doors and combustible ceiling decorations.
• Have your EMS medical director or emergency manager ask the regional burn center, in writing, for its realistic surge capacity for critically burned adults on a weekend night and its overflow destination, and enter the answer and date into the mass casualty annex.
• Have your communications center supervisor check the premise records for your five largest night-time assembly occupancies, confirming that each lists the posted occupant load and the locations of secondary exits, and correct any record that has not been verified in the past year.
• Have your fire marshal confirm with your building official which edition of the fire and life safety codes your jurisdiction has adopted and what local amendments apply to revolving doors and decorative materials, and put that answer in writing for your inspectors.

Takeaways

• The Cocoanut Grove fire in Boston on November 28, 1942, killed 492 people by the count the NFPA and most histories use, and some accounts give 490 or 491, a discrepancy no published reconciliation has settled.
• The story that a busboy’s match started the fire is disputed: the Boston Fire Department investigation did not determine the cause, and later theories about wiring or refrigerant gas are unproven hypotheses.
• Most deaths came from building conditions rather than the ignition source, chiefly a jammed revolving main entrance, locked or concealed side exits, inward-opening doors and combustible fabric ceilings.
• The Massachusetts Supreme Judicial Court upheld the owner’s involuntary manslaughter conviction in Commonwealth v. Welansky in 1944, and the case is still cited on wanton or reckless conduct.
• After the fire, Massachusetts and other jurisdictions restricted revolving doors and tightened requirements for exit doors, exit signs, emergency lighting and decorations, and the NFPA revised the code later renamed the Life Safety Code.
• At Massachusetts General, Oliver Cope and Francis Moore replaced tannic acid with soft petrolatum dressings, emphasized plasma fluid replacement, and documented the importance of airway and lung injury, with their results published in 1943.
• Erich Lindemann’s study of survivors and bereaved families at MGH contributed to his 1944 work on acute grief and to the foundations of crisis intervention.
• Codes have tended to follow mass casualty fires, but a code change prevents nothing in a building that is inspected only when it is empty, so after-hours enforcement and current premise information still matter.