The Bomb Full of Sleeping Bats That Burned Down Its Own Air Base

On a May afternoon in 1943, a group of very sleepy bats woke up earlier than anyone expected. Each one had a tiny napalm capsule glued to its chest and a timer already running. They did what bats do: they looked for somewhere dark and quiet to hang. They found the underside of a fuel tank, some outbuildings, and — according to every account of the day — a general's car. The United States Army had just firebombed its own airfield with a weapon it was still trying to invent.

2026-07-29 · 5 min read · Filed under Engineering

The Bomb Full of Sleeping Bats That Burned Down Its Own Air Base

What happened?

The idea started with a dentist. Lytle S. Adams, a dental surgeon from Irwin, Pennsylvania, was driving home from Carlsbad Caverns in December 1941 — having just watched millions of Mexican free-tailed bats pour out of the cave mouth at dusk — when he heard about the attack on Pearl Harbor. Those two facts collided in his head, and in January 1942 he mailed the result to the White House.

His pitch: Japanese cities were built largely of wood and paper. Bats roost before dawn, in eaves and attics, in exactly the cracks a bomb can’t reach. Attach a time-delayed incendiary to each bat, drop a canister of them over a city at night, and thousands of fires would start at once — in places no fire watcher would think to look.

Adams knew Eleanor Roosevelt. The letter reached the President, who forwarded it to William Donovan, soon to run the Office of Strategic Services, with a cover note that has outlived everything else about the project: “This man is not a nut. It sounds like a perfectly wild idea but is worth your time looking into.”

It got looked into. Harvard zoologist Donald Griffin, who had recently proven how bats navigate in the dark, assessed the plan and advised that it could work. With his endorsement the project was approved and handed to the Army Air Forces and the Chemical Warfare Service.

The engineering that followed was, by any measure, real engineering. Adams’ team — which at various points included a mammalogist, the Western film actor Tim Holt, and a young naturalist named Jack Couffer who later wrote the definitive account — settled on the Mexican free-tailed bat, a roughly 14-gram animal that could carry 15 to 18 grams of payload. Chemist Louis Fieser, who had just invented napalm, built a celluloid capsule two and a half inches long, filled it with napalm gel, and fitted it with a delay fuse. To make the bats manageable, the team chilled them into hibernation — sometimes, early on, in ice cube trays. The delivery device was a five-foot sheet-metal tube stacked with twenty-six circular trays holding 1,040 bats, designed to fall from a bomber, deploy a parachute, and split open at altitude.

Then came the field tests. On 15 May 1943, armed bats got loose on the ground. Accounts differ on the setting — most place it at the Carlsbad Army Airfield Auxiliary Air Base in New Mexico, while the 1982 American Heritage account describes the accident at the California test site, and the U.S. Army’s own retelling says the loose bats were part of a display being filmed for a training reel. Every version agrees on the ending. The bats roosted. The timers ran out. Hangars and outbuildings burned, and so did the general’s automobile.

The Army lost its enthusiasm. The Navy picked the project up in August 1943 and renamed it Project X-Ray; the Marine Corps took it that December, and Adams was squeezed out of the program he’d invented after colleagues concluded he couldn’t manage it.

Under new management, it worked. At Dugway Proving Ground in Utah, where the Chemical Warfare Service had built a mock Japanese residential block for incendiary testing, the bats performed. “It was concluded that X-Ray is an effective weapon,” an NDRC observer wrote. Chemists calculated that a planeload of bat bombs would start between 3,625 and 4,748 fires, against 167 to 400 for a planeload of conventional incendiaries.

Production of a million units was scheduled for May 1944. It never began. Fleet Admiral Ernest King cancelled the program on learning it likely wouldn’t be combat-ready until mid-1945. Roughly

Why was it strange?

The strangeness isn’t that someone proposed weaponizing bats. Wartime produces strange proposals constantly, and most die in a filing cabinet. The strange part is that this one kept passing review — clearing the President, a Harvard zoologist, the National Defense Research Committee, the Army, the Navy, and the Marine Corps — and that at the end of the gauntlet the official verdict was that it worked better than the standard bomb.

It’s also a rare case where the accident and the proof of concept were the same event. The 1943 fire is remembered as a fiasco, and it was. It was also, unambiguously, the weapon functioning as designed on the wrong target.

What did scientists learn?

Mostly, they learned how much precise animal biology a plan like this quietly depends on. The team had to answer questions nobody had needed answered before: how much can a 14-gram bat carry, at what temperature does it enter torpor — the state of lowered body temperature and metabolism that lets a bat shut down for days — how long can it stay there safely, and how fast does it wake once warmed.

Those answers weren’t free. Bats died in transport, broke wings on descent, and failed to rouse in time. The knowledge was real, and it was paid for by the animals. The same honesty applies to the goal: this was a device designed to burn civilian neighborhoods, and its inventor’s claim that it would spare lives was optimism, not analysis.

How does it affect us today?

The most durable legacy isn’t the bomb. It’s napalm, which Fieser developed for the NDRC and which went on to be used at devastating scale in later conflicts — a far heavier inheritance than anything the bats left behind.

The bat side of the story left a gentler mark. The wartime scramble poured funding into Mexican free-tailed bat biology at a moment when almost nobody was studying it, and several of the naturalists involved spent the rest of their careers on wildlife rather than weapons; Couffer became a nature-film cinematographer. Project X-Ray also sits in a small, instructive category alongside Project Pigeon and the anti-tank dog: animal-guided weapons that were technically demonstrable and operationally hopeless, and that mostly taught militaries a living creature will always have its own plans.

Fun fact

Donald Griffin, whose endorsement helped get the bat bomb approved, is the same researcher who — with Robert Galambos, a few years earlier — proved that bats navigate by bouncing high-frequency sound off their surroundings. Griffin coined the word for it in 1944: echolocation. The man who taught the world how bats find their way in the dark also helped convince the U.S. government to hand them incendiaries.

Sources

U.S. Army, “Bats Over Tokyo” — the Army’s own account of Project X-Ray, including FDR’s “not a nut” memo and the Carlsbad fire.

Air Force Test Center, “May 2, 1943: A Team Headed by Dr. Lytle Adams Arrived to Conduct Air Drop Tests on Canisters for the ‘Bat Bomb’ Project” — official test-history entry with canister specifications.

Joe Michael Feist, “Bats Away!”, American Heritage, April/May 1982 — archival detail on Griffin’s experiments, Fieser’s incendiary, the NDRC correspondence, and the program’s cancellation.

Jack Couffer, Bat Bomb: World War II’s Other Secret Weapon (University of Texas Press, 1992) — full text at the Internet Archive; the firsthand account by a member of the project team.

Alexis C. Madrigal, “Old, Weird Tech: The Bat Bombs of World War II”, The Atlantic, 14 April 2011 — includes reproductions of project photographs and documents.