Air University Review, January-February 1968

The Great Japanese Balloon Offensive

Master Sergeant Cornelius W. Conley

On 4 November 1944 a United States Navy patrol craft spotted something that looked like a large fragment of tattered cloth floating on the sea, 66 miles southwest of San Pedro, California. A sailor tried to drag the fabric on board the craft but discovered that a heavy undercarriage was attached to it. When the fabric was hauled on board, it was found to be a rubberized-silk balloon.

The apparatus connected to the undercarriage of the balloon consisted of a small radio transmitter with an on-off vibrator circuit to interrupt the transmitted signal periodically, giving a characteristic signal that could serve for identification purposes. The transmitter had a small variable condenser (actuated by an aneroid bellows), which caused the transmitted frequency to vary between 6.2 and 6.6 megacycles, depending upon the altitude of the balloon. Under reasonably favorable conditions, the transmitter would have a range of approximately 1000 miles.

There was also an aneroid bellows controlling the operation of a ratchet contact switch, which was designed to make contact successively with its seven switch points as the atmospheric pressure varied. The switch established contact when the external pressure went up and broke contact when the pressure went down. Contact was then made on the next successive point when the pressure rent up again.

The purpose of such a device was to change the transmitted signal for each cycle of the balloon's ascent and descent. In addition, the ratchet device was used to drop sand ballast automatically from the hopper type of undercarriage when the balloon fell to a pre-determined level.

All the equipment bore Japanese markings and indicated that something new and mysterious had been introduced into the struggle that was World War II.

After the finding of the first balloon, it was two more weeks before other fragments were salvaged from the ocean. At 1000 hours on 14 November 1944, the Coast Guard at Kailus, Hawaii, saw an airborne object descend into the sea, five miles out. The object was reported to be a paper balloon 30 feet in diameter. A fragment of the envelope and some of the attached gear were recovered from 20 feet of water.

Around 15 November, a balloon floated silently across Cape Flattery on Washington's rainswept northern coast. Finally, unseen, unheard, it buckled gently into a heap near Kalispell, Montana, 475 air miles east of the Pacific Coast. After its discovery by two puzzled loggers on 11 December 1944, men of the Federal Bureau of Investigation and the Army Air Forces investigated to determine its origin and purpose. The amount of snow on the balloon when discovered indicated that it had landed between 11 and 25 November. 

The balloon had presumably been launched from an offshore submarine, as had been the small seaplane, piloted by Nobuo Funita, which on 9 September 1942 had dropped about 165 pounds of fire bombs near Mt. Emily, Oregon, causing small forest fires. But why these unarmed balloons?

Though its appearance was quaint, the balloon was a practical and effective one. The FBI soon discovered that the Japanese had obligingly printed a good deal of information on the bag. It had been completed only a few weeks before, on 31 October 1944 to be exact, at a Japanese factory. Japanese characters also revealed the number of hours spent in its manufacture and data regarding work shifts.

The Rising Sun's V Weapon

The use of balloons as a weapon of war was conceived by the Japanese Military Scientific Laboratory in 1933, when study and research projects were started on the use of a 4-meter (13.12-foot) balloon with a flying range of 100 kilometers (62.14 miles). The study continued until 1935, when the research group of the lab started studying the theory of long-range balloon warfare, utilizing winds at altitudes of 3 to 6 km. They investigated methods of keeping balloons airborne for long periods of time and tried to determine if the west wind continued to blow the entire 10,000km distance across the Pacific Ocean. For some unknown reason the project was discontinued. Either the experiment was completed or, with no apparent need for this type of weapon at that time, the whole idea was shelved until some future date.

That date turned out to be 7 December 1941 when Japan entered World War II. At this time Major General Sueyoski Kusaba requested that the research group be given permission to conduct full-scale development of a long-range balloon. In addition, he requested that a 1000-km area be reserved for manufacture and test. But his pleas fell on deaf ears, and he found little if any support for his idea. The project remained a closed subject until a single event took place on 18 April 1942 and shook the Japanese military empire, dispelling once and for all the Japanese militarists' boast that their zone of inner defense was impregnable against air power.

The event was the Doolittle raid on Tokyo, by sixteen carrier-based B-25 medium bombers.

Apparently the attention given to the Japanese home defense had been secondary to that given to efficient operations in the war theaters. In spite of at least one radio message Tokyo received, the Japanese were unprepared for the attack.

In seeking reprisals for the Doolittle raid, the Japanese conceived the first transoceanic automatic-balloon campaign in history. It was to be their V-1 weapon. The Ninth Military Laboratory was immediately ordered to study various balloons capable of carrying bombs to the American continent.

It was first intended that the balloons would be released from submarines off the West Coast of the United States, and in March 1943 a 6-meter (19.6-foot) balloon with a desired range of 300 km (186.4 mi) was developed which flew 1000 km (621.4 mi) between the west and east coasts of "Japan proper." Later it was found that this model could stay in the air for more than 30 hours at an altitude of 8 km (4.97 mi).

By this time, however, the Japanese Navy was so depleted that ships and submarines necessary to carry on such an attack were no longer available, and therefore further investigations were necessary to invent a balloon capable of traversing the expanse of ocean between Japan and North America.

So the Ninth Lab was ordered to develop a balloon with a range of 10,000 km, to be released from the Japanese home islands. The research was started in August, with the emphasis on two objectives: maintaining the balloon aloft for a long period of time and determining whether the west wind was continuous for 6200 miles across the Pacific. They began by studying what were termed the "A" and "B" types of balloons.

The materials of the "A" type balloon consisted of handmade and handpatched integumentary paper. The raw, handmade mulberry paper had a standard weight of 15 g/sq m. With four pieces of paper pasted together lengthwise and breadthwise alternately, the balloon section began to take shape. The next step was to soften the paper panels by first dipping them into a solution of soda ash, then washing them with water, and finally dipping them into a solution of glycerine.

After the panels were dried, the edges were trimmed and the panels were pasted together on a spindle form, the top part first, then the lower part. After the relief valve was installed, the suspending band was attached to the two hemispheres, and they were pasted together.

Then the balloon was filled with gas for test purposes and coated with a protective varnish. The earlier paper balloons were made in factories, but when the demand reached its peak, the factories merely processed the paper and made the majority of the panels. The panels were then sent to subcontractors, who assembled the panels into the finished product. Some of the industrial firms connected with the operation were the Mitsubishi Saishi (Paper Factory), the Nippon Kakokin Company, and the Kokuka Rubber Company. It is interesting to note that the major share of assembly work was performed by Japanese school girls working in large theaters and sumo wrestling arenas in the Tokyo, Osaka, and Kyoto areas.

The type "B" balloon, which was a Japanese Navy project, was made of habutai silk with gum coating. The Fujikur Industrial Company impregnated the silk with the rubber type of gum. The balloon had a standard weight of approximately 170 kg.

These balloons consisted of four sheets of gum-coated habutai silk pasted together to form panels for the bottom half and five layers for the top half. After the two halves had been formed, they were put together in spherical shape. The connecting parts were sewed together with a wide, overlapping seam, and cotton tape sealed the seams. Then the balloon was inflated and tested, and any leakage was stopped by a gum concentrate with benzol as the solvent.

General Kusaba had many other obstacles to overcome before the harmless-looking balloon could be turned into an effective weapon. The variability of atmospheric phenomena was at first regarded as the chief problem. After several meteorological observations, however, it was concluded that as long as the weather was not too bad the atmospheric pressure should remain fairly constant across the ocean at a given altitude. It was recognized that there were some ascending and descending air currents, though, even at 10 km (32,000 ft), as a result of weather conditions on the earth.

Probably the greatest problem was in developing a radiosonde that could operate at length under the varying stratospheric conditions. The responsibility for development of such a system went to the Japanese Fifth Army Technical Laboratory. The primary purpose of the radio equipment would be to test the balloon's flying course and its altitude, to indicate the balloon's inside pressure. In addition, it would provide data on the balloon's ascending and descending flight and indicate when the ballast was being dropped. The first tests were conducted from the Chiba prefecture area.

At the time, there was no known radiosonde that could operate for the desired length of time under stratospheric conditions. The most trouble was encountered in the development of a power source of adequate durability and the selection of frequencies. In order to check the function of the radio equipment during these experimental flights, various models of the radiosonde apparatus were developed and suspended from the balloon instead of bombs.

After considerable effort, the lab finally succeeded in building a set. It was attached to a balloon, and the balloon was released on a free flight. For 80 continuous hours the set continued to operate, relaying valuable flight information. The radio fell silent when the balloon reached a point 130 degrees west longitude. Based on theoretical calculations, a balloon could cross the Pacific in three days during the winter period of November to March.

The radiosonde system developed for monitoring the balloon's flying Course was one that produced a continuous wave, moderated by a multivibrator. This piece of equipment had a power output of 2 watts with an A & B frequency which worked on an alternating cycle. "A" would operate 10 minutes and rest 10 minutes; while "A" was resting, "B" was operating, and vice versa.

In October 1944, 50 hours of flying records were obtained on a type "A" balloon and 80 hours on a type "B" balloon. Charts that were plotted on both balloons showed that the type "B" maintained a more constant and stable altitude.

From the results of the 200 experimental balloon flights conducted from February to March 1944, it was concluded that if more improvements were made in the future, the objective would be reached.

To investigate the weather conditions, lab scientists tried to collect all available material and information from the Central Meteorological Observatory and from the Military Observation Bureau, but at the time there was little information or reference material concerning altitudes above 500 meters. They obtained some data about the upper regions Over Japan and the American mainland, but nothing about the Pacific Ocean area. Since they were unable to obtain information about the west wind in the stratosphere, they had to analyze the data available from the Tateno Aerological Observatory.

Average winds aloft were plotted for each month beginning in January 1943. These were used to forecast wind velocities and directions. By September of that year maps had been completed, but they indicated a sharp curve of the airstream on the west coast of the American continent. This new discovery became the subject of many interesting discussions. By November 1943 the theoretical calculation had been verified to a small degree by the radio observations of the experimental balloon flights. These flights showed that the calculated routes were accurate as high as 300 km.

It is interesting to note how the predicted route charts were drawn up. First by supposing that the decreasing ratio of the temperature effected by the altitude was fixed, then by calculating the air pressure of this particular altitude, they were able to draw an aerological isobar. By calculating the inclining degree between the isobar and the speed of the wind from the latitude charted, they could draw a line to a point on a chart.

On the basis of this conception, they analogized the flying course of the balloon, its speed, and its diffusion; and thereby they decided where to launch the balloons. It was noted that the upper airstream which reaches the American continent is a winding one; that the airstream in the American continent area tends to flow southward. The time required for a balloon to fly the complete course was estimated from 30 to 100 hours; the average time was 60 hours.

The Japanese also had a minor problem in determining a launch site. If Hokkaido had been the starting point, the balloons would have entered the Soviet Union; so Choshi in southeast Honshu was chosen as the best launch site. Another reason for selecting this area was its proximity to the hydrogen supply. 

In determining when to launch a balloon, the Japanese used a simple process. If a high atmospheric pressure front had just passed the area, then it was most suitable for balloon launching; but if a high pressure front was approaching or if a low pressure front had just passed, then it was unsuitable for a balloon launching.

Another equally difficult problem resulted from the changes in temperature which the balloon encountered during its flight. A sudden change of temperature from 20 degrees in the day to -50 degrees after sunset would cause the balloon to drop. The Japanese Eighth Technical Laboratory was assigned to help solve the temperature and contraction problems.

Since the "B" type balloon had a greater inside pressure (35 mm Hg), there was little difficulty with the temperature contraction problem. This was determined by analyzing the radiosonde equipment which recorded the amount of ballast dropped. The "B" balloon proved to be a better balloon, but due to a shortage of materials only 300 were made.

In regard to the "A" type balloon, the problem was to determine how much sand should be carried, how much to drop at one time, and how many feet and how often the balloon would fall at night. Inasmuch as the duration of a flight was limited by the quantity of sandbags aboard the balloon, it was estimated that 35 sandbags, each weighing approximately 3 kilograms, would be needed. This quantity could keep the balloon flying for four days if it dropped approximately 25 kg of ballast per day.

It took a year's research and experimenting before both forms of the balloon were ready for practical use. It was decided that the first armed balloons would be launched on 3 November 1944. The balloons were released by a crew usually during the calm periods of evening or early morning. When the wind velocities were greater than five meters per second, the balloons were launched by one of two methods.

When the wind velocity was two miles per hour or less, the inflated and loaded balloon was secured with doubled ropes passed through the loops in the catenary rope at the equator of the envelope. One end of each holding rope was released simultaneously, permitting the balloon to rise free. When the wind velocity was greater, up to ten miles per hour, a different method was used. First, the ballast-dropping apparatus and load were placed on a stand several feet above the ground. The envelope was then filled upwind from the stand and loaded equatorially with sand ballast in special containers designed to open when pulled from below. The balloon was then "walked" into position and attached to the ballast-dropping mechanism. The ballast release ropes then were pulled, allowing the balloon to rise. It is presumed that this method was used to minimize the shock and oscillation that would have occurred if the balloon had been released abruptly. Launching normally required a crew of 30 men and could be done in 30 minutes. On days with favorable weather conditions, as many as 150 balloons were released.

The Japanese maintained launching points at Nakoso, Fukushima prefecture, with two companies of launching crews and six launching stations. In addition there were the same number at Ichinomiya, Aichi prefecture, but at Otsu, Shiga prefecture, there were three companies and nine stations.

In order to track the flights of the radio carrying balloons, they had direction-finding stations at Shikoku, Furukawa Iwanai, and one near Ichinomiya.

For What purpose?

After the first preliminary studies had been made, it was believed in the United States that the paper balloons were weather balloons or antiaircraft barrage balloons used by Japan to combat B-29s that were attacking the Japanese home islands. It was surmised that if this were the case the balloons must have strayed to the United States by accident. The Japanese were apparently unaware their balloons had reached this country.

Then on 23 December 1944, a paper balloon was recovered 15 miles north of Marshall, Alaska. This balloon, like the Kalispell one, did not have any undercarriage, but two sandbags were recovered, and the relief valve had a large "26" inscribed on it with chalk.

Another paper balloon was discovered at Estacada, Oregon, on 31 December 1944. It was similar to both the Kalispell and Marshall balloons in its basic construction features.

These incidents were brought immediately to the attention of the Commanding General, Western Defense Command, who initiated staff studies to determine the origin of the balloons, the purpose for which they were being directed toward the United States, and their capabilities.

The War Department was also kept fully informed of each reported balloon incident, and on 4 January 1945 the Chief of Staff designated the Commanding General, Western Defense Command, coordinator for all balloon intelligence activities in the Seventh, Eighth, and Ninth Service Commands. Concurrently, the Commander, Western Sea Frontier, was designated coordinator for balloon intelligence activities over ocean water areas.

One of the first objectives which the air defense organization had to consider before it could establish an effective defense was the purpose behind the Japanese balloons. Japan's subjection to bombing attacks and her inability to retaliate against them was undoubtedly of deep concern to the Japanese government.

After extensive studies were made, it was concluded that "ranging shots" were one of the most probable uses for the balloons. The prevailing winds over the North Pacific were of such velocity and direction that it would be possible to send balloons from Japan to the United States at the most favorable altitudes, which were between 20,000 and 25,000 feet. Reliable meteorological texts reported that at an altitude of 25,000 feet there is an air current that crosses Japan from the west, veers southward toward Hawaii, then swings to the north crossing California and part of Montana where it again veers southward. Our B-29 bombers had verified the existence of these strong westerly winds at high altitudes over Japan. A balloon at 25,000 feet could be carried eastward from Japan at a speed of from 30 to 60 mph, perhaps even faster. January and February would be the most favorable months of the year for the launching of the balloons. In addition, the latitude of Tokyo would provide the most favorable release point. From here an average wind flow at 20,000 feet could carry the balloons across the United States between Vancouver and southern Oregon. At 30,000 feet, velocity would be higher by 15 to 20 percent, and a balloon launched from the same latitude could land in Canada.

Very few accurate tracks (or series of plots) of the point of origin of the signals had been obtained because of technical difficulties, but the tracks that had been obtained showed movements of the transmitters along the general Course of the upper winds and at speeds that might be expected from free balloons. A direction-finder net for the purpose of obtaining accurate fixes was organized.

The signals were generally of the same type, pulsed continuous-wave transmissions of varying frequency and with marked transient characteristics. The pulse rates usually ranged between 20 and 150 pulses per minute. The frequencies usually ranged from about 5000 kilo cycles to over 12,000 kilocycles, with individual transmissions varying in some cases by more than 250 kilocycles. In many cases the frequency fell and rose several times during the period of monitoring, as. if the balloon were losing altitude, dropping ballast, -and rising again. In a number of cases the frequency dropped quite rapidly, and then the signal stopped suddenly as if the balloon fell into the ocean.

As additional information became available, the Western Defense Command estimated that the balloons were being released from an airfield near Sendai, 180 miles north of Tokyo, on the island of Honshu. From this information, it had to be assumed that the balloon episodes to date were Japanese "ranging shots" in preparation for mass launchings of such balloons.

On the assumption that these "ranging shots" would provide the Japanese with the flight Course of the balloon and the point at which it landed, we had to do everything possible to prevent the Japanese from obtaining this valuable information. It was almost certain that any news, if published, would be picked up by the Japanese. Furthermore, it was not considered advisable to alarm the U.S. civilian population until such time as the balloons might become a menace.

On 4 January 1945 two men working in a field near Medford, Oregon, heard a loud explosion and saw a spurt of flames 20 to 30 feet high, followed by a cloud of yellow smoke in a nearby field. When they arrived on the scene, they found a hole in the ground about 6 inches in diameter and 12 inches deep. The sides of the hole appeared to have been baked. Further investigation revealed a burned metal casing, cylindrical in shape, and pieces of molten metal, which indicated that the object was very likely an incendiary bomb. A hook, identical to those used on the shroud lines of the balloons, was also found. No planes were overhead at the time of the incident, nor were any other objects seen in the air. It was presumed likely that the bomb dropped from a balloon. On the same day, the U.S. Office of Censorship requested that all news agencies refrain from publishing news of the Japanese balloon operations. After the request had been issued, the commanding generals of the various commands personally presented the intelligence aspects of the problem to the various news agencies in the United States, Canada, and Mexico.

Thus the defense of silence was under-taken to prevent the enemy from gaining the information he desired. From now on the Japanese balloons would be referred to only by the code name "paper."

The second most logical purpose of the balloons would be to carry and introduce various types of bacteriological warfare (BW) agents. The intense cold (–20 to –50 °C) at the altitude of the balloon flights would facilitate the transmission of bacteria, and disease germs affecting humans, animals, crops, and forests could be transported. It should be emphasized that before BW agents could be effectively dispersed, technical difficulties had to be overcome. The following speculation was based on the assumption that the enemy had overcome all or part of these difficulties.

Despite the lack of specific evidence to prove that the two balloons carried BW loads, it nevertheless was well within theoretical limits for such vehicles to carry and disperse bacterial agents. The type of agent chosen would depend upon the degree of accuracy with which the balloons could be sent. In the event that the landing was within city limits, effective agents would be either those epidemic in character (e.g., pneumonic plague) or nonepidemic agents easily transmissible via the respiratory tract (e.g., psittacosis). If inaccurate dispersal were attempted, insect-borne agents or those affecting livestock would be an obvious choice. In the insect-borne group the mission might be accomplished by distribution of properly prepared bait or by infection of animals upon which the mosquito feeds. It would be theoretically possible to infect the vast U.S. culicine (mosquito) population and establish a permanent endemic focus of an agent.

In the meantime, to combat the possibility that the balloons would be used to shower pestilence in the form of plant-disease spores, animal plagues, Or even human disease germs, the government enlisted the services of the state health officers, veterinarians, county agricultural agents, 4-H Clubs, and agricultural college authorities in the defense program.

Decontamination squads were trained; stocks of decontamination chemicals, suits, and masks were made ready at strategic points. Farmers and ranchers were urged to report the first signs of any strange disease in their cattle, sheep, or hogs.

In order to obtain material for study, the closest possible liaison had to be maintained with all agencies that might spot the balloons. The War Department specifically prohibited mentioning the possible use of the balloons for bacteriological warfare, not only to the public in general but to anyone other than those individuals whose duties actually required that they have knowledge of this type of warfare.*

*Author's note: When I was in Japan, I talked to Mr. Kusaba about this subject, and he confirmed the fact that there was no intention to use the balloons for the carrying or spreading of BW agents.

Other possible uses of the balloons could be for the transportation of enemy agents and for Japanese propaganda purposes. It was considered unlikely that the balloons could be used to transport personnel from Japan, since the long trip through the stratosphere would require extensive equipment. The fact that the balloons had an automatic ballast-dropping device suggested that they were not intended to carry personnel. However, a partially inflated balloon launched from a submarine could conceivably make a relatively short trip inland with a person aboard.

The first reference to the balloons in Japanese propaganda was heard on 17 February 1945 in a Domei News Agency broadcast beamed to the United States in English. The Japanese claimed that 500 casualties had been inflicted in the United States and that numerous fires had been started. The broadcast also announced that government authorities in the United States had found it necessary to issue general warnings against attacks by the Japanese balloons and thus had agitated the people. It was emphasized that these occurrences had shattered the American feeling of security from attack by the Japanese. This broadcast was the first of a series designed as a war of nerves against the United States. Subsequent Japanese broadcasts beamed to Europe, Southeast Asia, and China repeated this same theme and in one instance added that several million airborne troops would be landed in the United States in the near future.

In the meantime, information concerning the hazards of the Japanese balloons was prepared for public dissemination, but by word of mouth only. It was distributed only within the areas affected by the balloons and was not published or broadcast over radio. Release of additional information did begin on 14 May 1945 through state educational systems and civic organizations such as Rotary, Kiwanis, Lions, and Chamber of Commerce.

Then a press and radio release concerning the balloons was authorized for 22 May 1945.  This was preceded by a confidential note to all editors and broadcasters that a prepared public release for publication was forthcoming and would outline the policy desired. The editors were requested not to report specific balloon incidents of any type and to report balloon activities in very general terms, in order to deny the Japanese useful infonnation concerning the landing dates and places.

The first Japanese reaction to the American newspaper and radio publicity was a Domei radio broadcast, in Romanized Japanese, datelined Lisbon, Portugal, 22 May 1945. This broadcast was directed to Greater East Asia. The broadcast was practically a repetition of the U.S. press and radio release. Later a brief Japanese report, dated 23 May 1945, was also heard. It, too, was just a repetition. 

Then a broadcast in English from Singapore (radio station Nampo Domei) was recorded on 4 June 1945. The broadcast was based on statements made at a press conference by Lieutenant Colonel Nakajima Shozo, a spokesman for the Imperial Forces in the Southern Area, who had been associated with press and propaganda activities since August 1943. He claimed that the balloons were causing havoc in the U.S., even though thus far they had only been released on an experimental scale. He also predicted that when the experimental period was over, "large-scale attacks with death-defying Japanese airmen manning the balloons would be launched."

Japanese propaganda then started asserting that such attacks by "death-defying Japanese airmen" would be launched soon. It was believed that the 32-foot balloon would be incapable of carrying a man from Japan to this continent with the necessary survival equipment at 30,000 feet. However, a balloon 62 feet in diameter, believed to be the largest practicable size, could carry a useful load of about 2600 pounds at 30,000 feet. According to the Army Air Force Materiel Division, such a balloon would require 1350 pounds of ballast for a 100-hour trip. The minimum weight of a man with survival equipment (pressurized gondola, oxygen equipment, food, water, and clothing) was estimated to be 640 pounds. On the basis of this weight, it was believed possible to transport an agent from Japan to this continent by a 62-foot balloon.

Japanese propagandists continued their efforts to inspire terror and divert forces in the United States. They also tried to convince their Japanese audience and others that the United States mainland had been successfully attacked with a new and ingenious weapon.

As a follow-up to the propaganda, Shimuzu Rikuro, a former press attachè of the Japanese Embassy in Argentina and manager of the Domei News Agency in that country, stated that the bombing attack by balloons on the United States had produced more damage than the Americans had admitted. He said that the balloons were a "prelude to something big."

Was this "something big" to be one-way suicidal attacks by long-range Japanese bombers? There were certain remote references to this possibility in some of the Japanese propaganda broadcasts. Yet the statements did not suggest that the balloon activity was a prelude to one-way attacks by long-range bombers; in fact, such attacks could be carried out whenever sufficient information was obtained on wind currents. The Japanese were known to have planes which, loaded only with the necessary gasoline and bombs and flying with favorable winds, could make such a flight over the shortest route—from Paramushiro to Seattle, or even to San Francisco.

No evidence indicated that enemy agents were being transported by balloons, and longrange bombers seemed impractical. Both were only remote possibilities.

The last possible use of the balloons which had to be considered was their ability to transport incendiary and antipersonnel bombs. These bombs could be dropped during flight by means of the ballast-release device designed to operate whenever the balloon descended below certain altitudes. Widespread dispersal of incendiary bombs in heavily forested areas, where most of the balloons were discovered, would have serious effects during the dry season. Incendiary bombs dropped during the wet season might have a delayed-action device that would ignite only after the surrounding ground had dried.

Gas bombs could also be dropped from the balloons, but this method would be an extremely ineffective one for gas warfare, since it lends itself neither to accuracy nor to building up the requisite lethal concentrations. It was unbelievable that the enemy would resort to such a random and futile method.

Incendiaries, on the other hand, presented a different picture. They required neither accuracy nor a large concentration to be effective. A single incendiary bomb might do as much damage as a dozen if it struck an area of combustible dry materials. Thus fires in forests and grain fields could readily be started over widely scattered areas during periods of dry weather by means of the balloons. This use of incendiaries appeared most logical, since the payload of the balloons was limited. Incendiaries would probably do more damage per unit of bomb-weight than any other type of light bomb, particularly against an indiscriminate target area. Antipersonnel bombs would probably be employed as a harassing device and would also have some incendiary effect.

Intelligence considered it highly improbable that the Japanese would attempt to employ incendiary bombs to start forest fires during the winter months, when the forests were covered with snow. It was necessary, however, to revise this opinion when information was received that a bomb had landed from a balloon near Minton, Saskatchewan, 12 January 1945. It was reported that the balloon had descended 6 miles north of the U.S.-Canada border and released a 20-pound bomb and two flares or incendiaries. One flare or incendiary exploded; the other and the bomb did not. The balloon then rose and disappeared.

So far no evidence had been obtained to disprove the theory that the balloons were being used to carry bombs and incendiaries, although the unusual radio signals heard from the Pacific area indicated that ranging might still be the primary purpose.

U.S. Defensive Measures

The unsolved mystery of the purposes of the balloons only hastened the preparation of defense plans. The uncertainty of the enemy's intentions and the possibility that the balloons might eventually become a real menace probably stimulated the development of countermeasures. On 17 January 1945, a conference was called to evaluate the situation, present to the various representatives the evidence that had been collected, and discuss the nature of the balloon threat and possible defensive measures.

A description of the balloon incidents that had occurred during the preceding three months and the status of the defense forces on the West Coast were outlined during the meeting. It was explained that since early 1944 the primary mission of the Fourth Air Force had been training, with little attention given to air defense readiness. Nearly all the early-warning radar stations had been placed on a caretaker basis; the Ground Observer Corps had been inactivated, and the Information and Filter Centers had been closed. With no aircraft warning service in operation, the Fourth Air Force was unable to guard the West Coast against a surprise air attack of any kind. No combat planes were kept on ground alert, and a defense force could be organized only after several hours' warning. Antiaircraft defense was likewise on a very limited basis.

During the course of the conference, the Fourth Air Force was assigned the following missions to counteract the balloon menace:

(1) To direct all its installations to observe and report all unidentified free balloons, or similar airborne objects, as well as any suspicious objects on the ground.

(2) To establish such measures as may be necessary to intercept unidentified balloons or similar airborne objects, so as to make ground recovery possible.

(3) To prepare plans and initiate action for the visual or electronic detection of unidentified objects in the Western Air Defense Zone.

(4) To coordinate with the Western Sea Frontier to secure and study meteorological data affecting free balloon flights in the north Pacific area.

(5) To request the Pacific and Alaskan wings of the Air Transport Command and the Western Flying Training Command to instruct their personnel to report balloon sightings.

(6) To request the Civil Aeronautics Authority to instruct personnel of commercial airlines and pilots of private planes to report balloon sightings.

Of these responsibilities, the two most important were the second and third: interception and detection of balloons.

The Fourth Air Force established a warning system based on the use of meteorological data to provide as much as three days' advance information of weather conditions favorable for a balloon attack.

As balloon incidents persisted and more information became available, two major conferences were called by the Ninth Service Command at the request of the Commanding General, Western Defense Command. The first conference was held 9 March 1945 at the Presidio of San Francisco for the purpose of fully acquainting representatives of the Western Defense Command, Ninth Service Command, Western Sea Frontier, Central Pacific Command, Alaskan Department, and the Dominion of Canada with the possibilities of bacteriological warfare. The second conference was held 23 March 1945 at Fort Douglas, Utah, to inform interested state and federal agencies of the balloon incidents. At the meeting Western Defense Command representatives took an active part in the proceedings. One of the major topics discussed was the relationship of Japanese balloons to forest fires. As a result of this meeting, the War Department authorized the Army to assist the Forestry Service in fire control. A plan, the "Firefly Project," was immediately and jointly initiated by the Western Defense Command, Fourth Air Force, and Ninth Service Command.

A complete, detailed plan known as the "Joint Western Sea Frontier-Western Defense Command Plan covering defense against Japanese Free Balloons, Short Title "BD-l’" was completed on 15 August 1945 but not published. All previous plans, with a few minor changes, were incorporated in BD-l as a final consolidated plan covering all phases of balloon defense.

In April 1945 an experimental program known as "Sunset Project" had been conducted by the Fourth Air Force under supervision of the Western Defense Command. The purpose was to determine the effectiveness of radar in detecting balloon arrivals and to study the possibilities of bringing the balloons down at sea or in open land areas with machine-gun fire from fighter planes using a new experimental ammunition known as the headlight tracer. If the balloons were effectively downed, their potential danger could be minimized. Six sites along the coast of Washington and in the vicinity of Seattle were selected for the installation of ten radar units. Their coverage extended along the coast from Cape Flattery in the north to the mouth of the Columbia River. 

All equipment required for the project, however, was not immediately accessible for the Fourth Air Force. The operations of the project were to include search for balloons by radar and interception by fighter planes guided by VHF ground equipment. Radar plots were to be reported to the Silver Lake Region Control Center, where it was expected that balloon targets could be recognized by their speed, altitude, and relation to known wind currents. P-38 and P-61 aircraft were to be kept on an alert status at Paine Field, Quillayute Naval Air Station, and Shelton Naval Air Station. Detailed weather information indicating the probable courses of the balloons was to be furnished the control group by the weather officer of the Fourth Air Force.

In the opinion of several radar experts, the balloon envelope would not be radar-visible. The steel relief valve would be visible, but its small size would probably allow it to pass unnoticed through a radar search zone. The chances of detection at 10,000 feet would be small and at higher altitudes nil. These conclusions were checked out by sending aloft a facsimile of the valve. Radar-visibility tests of the metallic structure of the balloon were made to determine at what ranges it would be possible to detect the balloons in flight.

It had now become apparent that the Japanese balloons were being directed toward the continent of North America, rather than arriving by accident. The increase in the number of recoveries and sightings indicated that the experimental phase was over and that the balloons were being launched for effect. Although only a relatively small number had been recovered, it was likely that for each one recovered or sighted a large number had landed unseen in remote, uninhabited areas. Furthermore, those that had been recovered were ones which had failed to function properly.

The only known fatalities on the United States mainland from enemy attack during World War II came on 5 May 1945 ten miles northeast of Bly, Oregon, when five children and one woman were killed from the blast of a bomb that had been carried by a balloon. The cause of the tragedy was verified by forest men who said it appeared that the victims had clustered around the balloon and someone curiously tugged it enough to detonate one of the bombs carried underneath. The only publicity permitted on the incident at the time was that an unidentified object had exploded, killing six people.

Investigations of free balloon incidents disclosed three types of explosive ordnance. The first type is the standard Japanese Army 15-kilogram, Type 92, antipersonnel bomb. The danger radius for this bomb (weighing 33 pounds) on open terrain is 150 feet; casualties may result within a radius of 300 feet.

The second type was a new 12-kg thermite incendiary, fuzed for instantaneous function but crudely constructed. None of those examined contained explosive charges.

The third type was the 4.46-kg incendiary bomb. Recovered bombs of this type varied in weight from 10.5 to 11.1 pounds.

At the end of April 1945, the balloon barrage ceased. Had the Japanese called it off as a failure? Or was this a deceptive lull before a greater barrage? Weeks and months passed with no resumption.

The reason why was not solved until after the war. In 1947 Brigadier General W. H. Wilbur (now retired) visited Japan and conferred with General Kusaba. According to this foremost authority, the Japanese had figured that at least ten percent of the balloons should have reached the United States and Canada. Word of the initial landing in Montana was the only information that the Japanese received until they picked up the report of the six deaths near Bly, Oregon. With only one reported landing on the American continent, the Japanese General Staff began to doubt General Kusaba and his project. More than once he was reminded that he was wasting the fast-dwindling resources of the country. Finally, toward the end of April, General Kusaba was told to cease all operations. The paper balloons cost originally about 10,000 yen each, roughly $2300 at the prewar rate of exchange, but the cost was reduced somewhat by mass production. More than 9000 balloons were built, and over 6000 were launched.

The project in a sense failed because a wall of silence was formed by the American people. The success of the security measures is indicated by an Associated Press release of 2 October 1945 which contained the following comments: "The Japanese listened eagerly to radio reports, hoping to hear of the bombs' effectiveness. But American editors voluntarily kept the information to themselves and so discouraged the Japanese that they abandoned the project."

The campaign of silence was abandoned by the War and Navy Departments following the tragic incident on 5 May 1945. On 22 May the War and Navy Departments issued a joint statement describing the nature of the balloon bombs and warning all persons not to tamper with any such objects they might find. The balloon weapon was said to constitute no serious military threat to the United States because the attacks were "so scattered and aimless." The statement continued: ". . . the possible saving of even one American life through precautionary measures would more than offset any military gain accruing to the enemy from the mere knowledge that some of his balloons actually have arrived on this side of the Pacific."

The balloon project was a complete failure as far as damage to military installations was concerned. Nearly all the balloons that landed on the North American Continent did so in open country or in wooded, mountainous areas, far from centers of population and military establishments. It is possible that some bombs caused forest and brush fires, but no evidence was ever discovered to indicate that any of the balloons were armed for bacteriological warfare. Balloons were reported over an area stretching from the island of Attu to the state of Michigan and from northern Alaska to northern Mexico, in all some 285 of them. This fact points to the greatest weakness in the free balloon as a military weapon: it could not be controlled.

The balloon attacks ceased before the radar stations of Sunset Project had an opportunity to detect any balloons. Numerous reports of balloon sightings reached the Seattle control group, and 68 interceptions were actually attempted. However, none of the sightings were verified, and practically all were identified as weather balloons, blimps, or the planet Venus, which was often mistaken for a balloon. It became increasingly evident that Japan had ceased balloon launchings, and the lack of activity stateside led Continental Air Forces to terminate Sunset Project on 1 August 1945.

Thus in the same month that saw the end of World War II came the finale of Japan's balloon offensive.

1989th Communications Squadron


Master Sergeant Cornelius W. Conley (B.A., University of the Philippines) is assigned to the 1989th Communications Squadron, Torrejon AB, Spain. He enlisted in the Air Force in 1948 and was in the aircraft maintenance career field until 1961, when he retrained into his present career field. He has completed the following Air Force courses: Basic Aircraft and Engine (1948-49), Engine Specialist (1950), Aircraft Maintenance Technician (1954), Academic Instructor (1958), Management Instructors Institute (1963), Development and Management of Training Materials (1964), Instructional Programmer (1965), and Senior NCO Academy, Military Airlift Command (1965).


The conclusions and opinions expressed in this document are those of the author cultivated in the freedom of expression, academic environment of Air University. They do not reflect the official position of the U.S. Government, Department of Defense, the United States Air Force or the Air University.

[ Home Page | Feedback? Email the Editor ]