NEW! Canadian Military History Trivia Challenge

Canadian Military History Trivia Challenge

Take the quiz and Win a Trivia Challenge prize pack!

Canadian Military History Trivia Challenge

Take the quiz and Win a Trivia Challenge prize pack!

The West Coast Balloon Attack: Air Force, Part 34

Debris from a Japanese incendiary balloon. [PHOTO: LIBRARY AND ARCHIVES CANADA—PA203222]

Debris from a Japanese incendiary balloon.

Commencing Nov. 3, 1944, and continuing to mid-April 1945, Japan launched between 9,000 and 10,000 incendiary balloons from their home islands in an attempt to set North America’s forests alight from Alaska to California.

American and Canadian authorities imposed a news blackout on balloon arrivals to deny the enemy any intelligence about the effectiveness or non-effectiveness of the devices. This proved so successful that only one report of balloon landings filtered back to Japan, even though thousands of persons witnessed balloons passing overhead or saw balloon debris lodged in fence lines or treetops.

On May 5, 1945, six Oregon picnickers were killed by a balloon-delivered bomb they had discovered. On May 22, 1945, the news blackout was lifted to warn of hazards posed by the devices.

Balloons delivering bombs date from 1848 when Austria attempted to bombard rebellious Venice using such ordnance. Japanese design and testing of incendiary balloons (known as Fu-go weapons) began in 1933. Development of them picked up following the first direct American attack on the home islands (the famous Doolittle Raid of April 18, 1942) and the crippling of the Japanese carrier fleet at Midway in June 1942. Lacking any long-range bombers, Japanese authorities saw the balloon bombs as a chance to strike back, even create havoc in North America. Their meteorologists were aware of the high-altitude winds that swept across the Pacific, and these winds would be used to deliver the balloons.

Tangled remains of a balloon found near Delburne, Alta. [PHOTO: LIBRARY AND ARCHIVES CANADA—PA203216]

Tangled remains of a balloon found near Delburne, Alta.

The weapons were of simple design and construction. The balloon was made of paper treated to make it leak-proof (some were made of silk). It was 32 feet in diameter, and held approximately 19,000 cubic feet of hydrogen. Slung beneath the balloon or “envelope” was an elaborate array of equipment attached to a platform which became known as the “chandelier.” This included 32 bags filled with sand and a barometric control device. If the balloon descended within the first 40 hours after launch, the barometer would trigger two or four sandbags to drop, lightening the balloon which would rise again until it approached North America. Once over the target area the Fu-go was to begin dropping its ordnance at intervals. This usually consisted of four 4.5-kilogram incendiaries and a 15-kilogram high explosive bomb. A flash bomb attached to the envelope was to destroy the last vestiges of the device.

The Japanese knew only a fraction of the balloons would make it across the Pacific, but the subsequent news blackout deprived them of any meaningful intelligence about arrivals. Precisely how many Fu-go weapons reached North America is impossible to calculate. As of August 1945 a total of 285 confirmed sightings, including recoveries, had been recorded by American and Canadian authorities, and recoveries continued for years afterwards. In 1981, two Japanese military writers estimated that 10 per cent (900) arrived, most of which were rotting and rusting into oblivion in wilderness areas. In 1998, however, historian Michael Unsworth noted that the weapons had an Achilles heel. The ballast-dropping mechanism was powered by a battery which in turn was encased in a protective plastic box filled with antifreeze. The antifreeze solution was too weak; most batteries froze at high altitude and the greatest number of balloons simply descended into the Pacific with their ballast bags dragging them down.

On Nov. 4, 1944, an American naval vessel recovered a Fu-go weapon 66 miles southwest of San Pedro, Calif. A total of eight had been identified by the end of the year; the first Canadian incident was on Jan. 1, 1945, when fragments of a balloon envelope were found near Stoney Rapids, Sask. Once authorities had been alerted to their existence—and knew what to look for—the number of Fu-go incidents increased rapidly. The peak month was March 1945 when 114 were identified, 40 of them in Canada.

Intelligence officers, trained to im­agine the worst possible scenarios, presented several theories as to what the balloons were to accomplish, and these ranged from the fantastic—“carrying of espionage or sabotage agents”—to the horrific—“use in connection with bacteriological warfare.” This latter theory was especially worrisome, given even sparse knowledge of Japanese experiments in this field. Other possible goals included “study of air currents to procure data for future activities” and “hazards to air navigation.”

Balloon contents found near Minton, Sask. [PHOTO: LIBRARY AND ARCHIVES CANADA—PA203215]

Balloon contents found near Minton, Sask.

The intelligence problem was enhanced by the fact that most balloon “incidents” involved distant sightings or partial recoveries of balloon or bomb fragments. Nevertheless, fairly complete examples were found at widely scattered places, including Kalispell, Mont., on Dec. 11, 1944; Sebastopol, Calif., on Jan. 4, 1945; Minton, Sask., on Jan. 12; Fort Simpson, N.W.T., on Jan. 21; Provost, Alta., on Feb. 7 and Parkberg, Sask., near Moose Jaw on Feb. 8.

The Minton recovery was particularly important for the information it provided, not only about the balloon’s construction but also its tactical use and potential hazards. It had been first spotted crossing from Montana into Canada. It touched the ground briefly and released a bomb which did not explode, rose again and drifted northeasterly where it touched ground once more, rose again, and then drifted out of sight. Two children found the unexploded bomb and wisely reported it to the RCMP.

A joint Royal Canadian Air Force/ Canadian Army/RCMP team commenced a search, tracking it by civilian sightings and damaged fences before it finally snagged on its third barbed wire fence. Not only were examples of both explosive and incendiary bombs discovered, but also ballast bags filled with sand. Experts at Canada’s National Research Council and the United States Geological Survey eventually pinpointed the origins of the sand (and hence the launch sites) through tiny marine fossils. This helped lower the fear that they were being launched from submarines off shore.

The greatest number of sightings was over the more densely populated areas of the United States; five alone were spotted at 10,000 feet near Inyokern, Calif., on Feb. 9, 1945. Canadian settlement patterns hindered detection. Even so, Western Air Command reports included multiple daily sightings. On Feb. 18, for example, WAC’s diary mentioned five sightings and three fruitless fighter aircraft scrambles to intercept. On March 10, the diary described six sightings that generated interception attempts. The diary also mentioned several unconfirmed sighting reports. Given the known tracks, the number of balloon “fly-overs” and landings in Canada has probably been underestimated.

Balloon contents found near Fort Simpson, N.W.T. [PHOTO: LIBRARY AND ARCHIVES CANADA—PA203213]

Balloon contents found near Fort Simpson, N.W.T.

Once the balloons had been identified and their purpose deduced, the question became one of what to do. The natural instinct was to track them and shoot them down, but would that simply deliver them more promptly to the forests they were intended to destroy? Detection itself posed problems. The balloons were almost invisible to radar at more than 12 miles. A civilian organization, the Aircraft Detection Corps, had been disbanded before the Fu-go assault began, but consideration was given to reviving it. This was not done, but defensive measures were taken, including more widespread deployment of fighters.

But interception by aircraft was easier said than done. Sighting reports were often hours old before aircraft could be “scrambled.” Timing was important, because Western Air Command had no night interception ability. Balloons approaching at 30,000 feet were virtually unassailable. On March 12, 1945, No. 133 (Fighter) Squadron experimented with their Kittyhawk aircraft. Pushed to its limits, a Kittyhawk with six machine-guns could reach 29,800 feet; stripped down to two machine-guns the same machine could struggle to 32,000 feet.

American and Canadian pilots nevertheless shot down at least 19 balloons, including 10 in one day over the Aleutian Islands on April 13, 1945. Canadian records are maddeningly brief, but three balloons can definitely be credited to RCAF pilots. The first was destroyed over Sumas, B.C., on Feb. 21, 1945, by Flying Officer Edward E. Maxwell in a No. 133 Sqdn. Kittyhawk while on detached operations to Port Hardy. The squadron diary noted only that he was scrambled after “paper” was airborne from 1620 hours to 1805 hours and that Maxwell located his target at 25,000 feet and shot it down.

The second RCAF “paper kill” was achieved by another Kittyhawk pilot, Flying Officer James G. Patten, also of No. 133 Sqdn. Again, the unit diary is much too brief, stating only that on March 10, 1945, he intercepted a balloon at 13,500 feet near Ganges Harbour, Saltspring Island, and shot it down in a sortie lasting 55 minutes. Happily, the diary of Western Air Command is more detailed, stating that Patten and another pilot had attacked with ball ammunition. It exploded in a reddish flash; the appendages landed in the water. The pilots searched for 20 minutes to pinpoint whatever remained of the device.

The final “paper kill” was on March 12, 1945. A balloon was reported by ground personnel near Coal Harbour, B.C., at a very low altitude. Flight Lieutenant Russell L. Moodie, piloting a Canso of No. 6 (Bomber Reconnaissance) Sqdn., was returning from an anti-submarine patrol when he sighted the device, which was then descending. He used his slipstream to accelerate its descent and it became entangled in trees near Rupert Arm where it was salvaged intact.

Hopes were placed in De Havilland Mosquito aircraft to catch the high-level balloons. No. 133 Sqdn. received some “Mossies” on Feb. 25, 1945, which were declared operational on March 8. They were scrambled on at least three occasions, but failed to catch any Fu-go. However, on June 11, 1945, a section of Kittyhawks and a Mosquito shot down a balloon over Victoria. It proved to be an American one being used to test radar efficiency.

Inland, steps were taken to have aircraft on the Prairies track or destroy balloons. On March 1, 1945, Air Commodore B.F. Johnson reported that No. 2 Air Command (Winnipeg) planned to remove five Hawker Hurricanes from stored reserve at Moose Jaw, and deploy them for interception duties—one at Moose Jaw, two at Saskatoon and two at Yorkton, Sask. In addition, two bombers at Suffield, Alta., would be available to follow Fu-go weapons, while a Spitfire and a Mosquito at Rivers, Man., would track any balloons drifting that far east.

The balloon bombs have generally been dismissed as futile gestures. The Japanese had chosen to launch the weapons at a season when high-level winds were most favourable to their reaching the Pacific coast forests, yet those were the least dangerous months in terms of fire hazards. Nevertheless, Dr. D.C. Rose, Scientific Adviser to the Chief of the General Staff, wrote on March 10, 1945, that continuation of the launchings might well increase the number of forest fires in Western Canada. He noted that over a 10-year period (1927-1936) there had been an average of 90 forest fires in British Columbia in April, but 235 in May, 200 in June and 490 in July. The figures for Alberta were 25 fires in April but 75 in May. Fortunately, the Japanese suspended the Fu-go campaign in April 1945, virtually on the eve of when it might have had some impact.

There were also incidents showing that the balloons were not harmless. At about 2 p.m. on Feb. 12, 1945, three bombs fell at Riverdale, 20 miles southwest of Great Falls, Mont. Two fishermen heard a hissing sound, saw the bombs dropping—although not the balloon itself—and watched as they exploded, starting a grass fire which they extinguished.

Although only one incident—in Oregon on May 5, 1945—resulted in deaths, there were several occasions when people were undoubtedly and unwittingly at risk. The children at Minton, Sask., could easily have fallen victim to the bomb they discovered. On Feb. 7, 1945, a balloon was sighted by natives near Alert Bay, B.C. They attempted to recover it but a strong wind blew it out of range. On March 22, 1945, fishermen tried to haul a Fu-go aboard their vessel near Texada Island, B.C.

Perhaps the greatest hazards were those faced by military personnel charged with recovery of any material associated with the weapons. In a 1993 personal memoir titled White Paper: Japanese Balloons of World War Two, Captain Charles East of the Royal Canadian Engineers recalled trips by jeep and snowshoes into bush to perform these tasks. He discovered one balloon in a tree with two bombs and 12 ballast bags intact. He studied them through binoculars, trying to determine if the bombs were still operable. While walking around the tree the heel of one of his snowshoes snagged something. He looked back and discovered a black object projecting from the snow. He dug around it and discovered the tail fins of another bomb—apparently an anti-personnel type. A cold chill ran up his back when he realized what he had almost tripped over.

Apart from the operational problems posed by the Fu-go campaign, Canadian military files reveal curious misconceptions about related matters. On April 12, 1945, Group Captain Hugh R. Stewart, Director of Intelligence, Air Force Headquarters, wrote to the RCMP suggesting that Japanese-Canadians should be interrogated as to whether they had been approached by Japanese agents seeking information about the balloons. Deputy Commissioner F.J. Read replied on April 17, 1945. Stewart’s idea had been “based on the assumption that the Japanese maintain a widespread network of espionage in Canada.” Read was almost contemptuous of the idea. “Such an assumption has no foundation and cannot be entertained in the light of our experience.” The police, in short, had a greater grasp of the espionage threat (non-existent) than the RCAF.

Spurious reports of high altitude balloons continued into July 1945. Western Air Command dismissed many of these as being sightings of the planet Venus. Fu-go dropped from operational notice until December 1952 when an RCAF staff officer suggested that incendiary balloons launched from Russian submarines and merchant ships might pose a serious threat to forests as far east as Ontario. He presented a weak case, especially as he blamed Japanese balloons for some 1944 fires that had occurred months before the Fu-go campaign began. Squadron Leader G.R.M. Hunt, writing on Dec. 16, 1952, shredded this paranoid projection, partly with facts about the ineffectiveness of the Japanese weapons, but also with sarcasm. Of the submarine threat he wrote, “Soviet submarines will have plenty to do trying to sink our shipping in the Atlantic, not wasting their time in the Pacific.” As to merchant ships as balloon launching platforms, Hunt asked, “What is the U.S. Navy doing letting ocean-going vessels steam up and down our West Coast?”

Overall, the Fu-go campaign may be regarded as an interesting subject from the vantage points of Japan and North America. It was, nevertheless, a doomed enterprise—a feeble offensive that largely frustrated the defenders but ultimately failed through its own operational limit­ations and technical shortcomings.

Email the writer at:

Email a letter to the editor at:


Sign up today for a FREE download of Canada’s War Stories

Free e-book

An informative primer on Canada’s crucial role in the Normandy landing, June 6, 1944.