Chapter 7: Flame Throwers
Portable Flame Throwers
On 8 December 1942 near Buna Village, Papua, Corp. Wilbur G. Tirrell crawled through the underbrush to a spot some thirty feet from a Japanese emplacement. He stepped into the open and fired his flame thrower. The flaming oil dribbled fifteen feet or so, setting the grass on fire. Again and again Corporal Tirrell tried to reach the bunker, but the flame would not carry. Finally a Japanese bullet glanced off his helmet, knocking him unconscious. This was the first time an American flame thrower had ever been used in combat. It failed so miserably that Col. William A. Copthorne, Chief Chemical Officer, USAFFE, believed that infantrymen would never want to use flame throwers again. Yet the weapon had distinct advantages, and before the war was over the armed forces had first-rate models.1
Flame throwers were introduced by the German Army in 1915, and then adopted by other European armies. Americans, using European weapons as models, began to develop flame throwers in 1917, but the armistice was reached before they were completed. The American Expeditionary Forces was not enthusiastic about the flame thrower—General Fries called it “one of the greatest failures among the many promising devices tried out on a large scale in the war”—and after the conflict the Chemical Warfare Service dropped the weapon completely.2 So it was that in July 1940
when the Corps of Engineers asked the CWS for a portable flame thrower the service had to go back to the beginning and start all over again.3
The Kincaid Co. of New York manufactured a few of the first experimental models, designated as El, in the autumn of 1940. This and all subsequent models consisted of four main components: a storage system for fuel, a storage system for compressed gas, a flame gun, and an igniter. Each was an indispensable component and each required considerable development. The first fuel tank was a vertical cylinder having two compartments, the upper holding nitrogen under pressure and the lower containing five gallons of fuel—at that time diesel oil, fuel oil, or a blend of gasoline and oil. The filled weapon weighed seventy pounds. The fuel oil flowed through a flexible tube into the flame gun. The gun was a metal barrel to which was fastened an igniter consisting of a battery and a cylinder filled with compressed hydrogen. The gun had two triggers, one to release the fuel and the other for ignition as it issued from the nozzle. When the weapon was fired, compressed nitrogen blew the oil through the hose and gun at the rate of one-half gallon per second. At the nozzle an electric spark from the battery lit a small jet of hydrogen, which in turn set aflame the oil. The stream of burning oil had a range of fourteen to twenty-one yards.4
The Engineer Board tested the weapon and decided it was not ready for military operations. The combination fuel-compressed gas reservoir was impractical, a bend in the gun barrel made the flame thrower difficult to fire from a prone position, the gas pressure dropped steadily while the weapon was being fired, decreasing the range. The entire apparatus was cumbersome, heavy, and undependable. The CWS redesigned the flame thrower, and contracted with the Kincaid Co. for a better model, E1R1, which reached Edgewood in March 1941. Now the fuel and compressed nitrogen were stored in separate reservoirs, a feature retained in all future
Operator firing a portable flame thrower E1R1 at a concrete fortification during a test of the weapon.
models. The gun, the ignition system, and the valves were all improved. The weapon weighed 28 pounds empty and 57 pounds loaded.5
The E1R1 was far from perfect—parts were easily broken, valves were hard to reach unless the operator was a contortionist, and the weapon made an uncomfortably heavy load on the operator’s back—but the weapon held a range of 15 to 20 yards for 15 to 20 seconds and on the whole seemed suitable for use in special situations. Since the few that had been produced for the purpose of testing were the only practical American flame throwers in existence, the Army issued them to troops in training camps. Some troops actually carried these crude weapons overseas to the Pacific Islands and employed them in battle. Corporal Tirrell’s assault on a Japanese bunker was made with one of these, an E1R1.
The CWS in the meantime had been rushing an improved version of the E1R1 through final development. Suggestions from test boards had led to a slightly heavier, more rugged, longer range model standardized
as M1. In March 1942 the weapons began coming off production lines, some reaching the South Pacific Area by the end of the year.
The new flame thrower was employed for the first time on 15 January 1943 by marines and infantrymen on Guadalcanal. An infantry attack against a stubborn enemy pocket holding up the advance of the 2nd Battalion, 35th Infantry, did not succeed. But Marine engineers burned out three Japanese bunkers in a ravine, and thus helped rout enemy troops holding up a Marine advance.6
At that time the flame thrower was unfamiliar to the Army. Since most troops had never seen one before, they did not know what it could do. After its success on Guadalcanal the word spread around, and in later engagements in the Pacific, from New Guinea to the Ryukyus, foot soldiers always kept the flame thrower in mind when they had to overcome a stoutly defended Japanese position. This is not to say that the M1 was a perfect, reliable weapon. One specimen, just received from the States, might function properly and spurt a jet of flame the customary fifteen yards, but its twin might eject a harmless stream of nonburning oil a distance of five yards. Batteries in the ignition circuit deteriorated rapidly in the hot, humid, climate; inadequate waterproofing allowed moisture to corrode parts and to short-circuit the electrical system; and minute rust holes in tanks allowed compressed gas to escape and the pressure to drop. Chemical maintenance companies had their hands full inspecting, testing, repairing, and servicing flame throwers to keep them in proper working order for the troops.
The weapon that replaced the M1 came about as a result of the invention of napalm, developed originally to thicken gasoline fillings in incendiary bombs.7 The CWS tested thickened gasoline in flame throwers and found that the range was greater than with ordinary gasoline. Furthermore, ordinary gasoline broke into a spray after it left the nozzle of the flame gun and burned itself out in a billow of fire while thickened fuel flew through the air in a compact stream that would ricochet into portholes and stick to flat surfaces.
Unfortunately, the M1 flame thrower could not get the greatest range out of the new thickened gasoline, for it was like running a 1910 model automobile on modern premium fuel. In August 1942 Col. William C.
Kabrich, chief of the CWS Technical Division, appointed a joint CWS-NDRC committee to modify the weapon. To keep from upsetting the procurement program, Colonel Kabrich asked the committee to make the changes in the M1 as few and as simple as possible. The committee modified the fuel system, pressure regulator, valves, and flame gun to permit the higher operating pressure necessary to obtain the maximum range with napalm, and at the same time improved the waterproofing.8 The CWS-MIT Laboratory tested napalm-gasoline mixtures to find the one giving the greatest range.9
The new flame thrower, M1A1, could expel thick fuel two or three times as far as the old model. In aiming at the port of a pillbox fifty yards away it could place half of the thickened fuel inside the structure. The old model could not reach a pillbox fifty yards away, and could place only about 10 percent of its charge in a pillbox twenty yards away. The improved electrical system was still not entirely reliable, but on the whole the new weapon was so superior to the M1 that the CWS decided to produce and issue it immediately.
Initial shipments of the fourteen thousand M1A1 flame throwers manufactured during World War II arrived in the Mediterranean theater in June 1943, the South Pacific Area in July, and the Southwest Pacific Area in August.10 This model gave good service throughout the war, but occasionally the electrical ignition system failed. When this happened the troops dropped the weapon or else lit it by some other means. In the landing on Leyte, 20 October 1944, a flame thrower failed to ignite when the operator fired it at a pillbox. Nearby Pfc. Frank B. Robinson, 19th Infantry, threw a handful of burning paper in front of the pillbox. The operator ignited the fuel by firing it through the flames from the paper.11 In an action at Azeville, France, 9 June 1944, Pvt. Ralph G. Riley ran up to a German blockhouse with his flame thrower, only to find that the ignition system would not work. He held a match near the nozzle and ignited the stream of fuel. Ammunition exploded inside of the
Attacking a Japanese bunker with an M1A1 portable flame thrower, Bougainville, March 1944.
fortification, the garrison of 169 men surrendered, and Private Riley received the Silver Star.12
The final portable flame thrower appeared in 1944. In developing this weapon the Chemical Warfare Service and the National Defense Research Committee set out on different roads to find the best method of ignition, and to design a more rugged, comfortable, dependable weapon. The NDRC contractor, Standard Oil Development Co., produced a model known as the E2. This featured a long-life battery, waterproof electrical system, lightweight aluminum fuel tank, and gasoline ignition system. The CWS turned out a weapon, E3, with a streamlined gun, improved valves, a comfortable pack similar to the standard Quartermaster pack-board used for carrying mortar shells and other ammunition, and a cartridge type of ignition. The latter was similar to a revolver. It held six cartridges, each filled with a pyrotechnic mixture. When the operator pressed the trigger a shower of sparks erupted from the cartridge and
Firing an M2-2 portable flame thrower into a wall opening, Manila, Luzon, February 1945.
ignited the fuel. Six cartridges allowed the operator to fire up to six bursts.13
In comparative tests the CWS and NDRC models showed approximately the same range, sixty yards with thickened fuel and one-third of this with ordinary fuel. The former model was slightly heavier and held less fuel than the latter. The Army preferred the rugged CWS model with pyrotechnic ignition to the light NDRC model with electrical ignition, and adopted it as the standard model, M2-2 in March 1944.14
American troops first employed the M2-2 in the Guam operation in July 1944. Manufacturers turned out more of this model than all earlier models combined—almost 25,000 as compared with 14,000—but production difficulties slowed down the issue of flame throwers to theaters of operation. March 1945 arrived before divisions in Italy received the weapon,
and some troops in the Pacific fought their last engagements of the war using the old M1A1.15
Although the M2-2 was better than its predecessor, it was not entirely satisfactory. It was too heavy, it did not hold enough fuel, and the fuel tanks were uncomfortable on the backs of the operators when the weapon had to be carried a considerable distance. It was, however, the most reliable flame thrower designed by any army up to that time. The CWS technicians continued to develop a light, large capacity weapon, but they did not reach their goal during the war.
The German Army, as the American, had portable-type flame throwers. The model in general use was Flammenwerfer 41. The fuel pack consisted of two steel cylinders, one containing approximately two gallons of fuel oil, and the other holding compressed nitrogen. A metal braided hose ran from the fuel pack to the flame gun. The original ignition system was similar in operation to the early CWS models; that is, an electrical spark from a small battery ignited a tiny jet of hydrogen. The Germans, too, had difficulty with this system and switched to a pyrotechnic cartridge. The filled flame thrower weighed approximately forty pounds, and expelled a jet of fuel from twenty-five to thirty-five yards.16
Japanese troops on occasion used flame throwers. Americans first faced them on Bataan in early February 1942, and shortly thereafter American soldiers captured two. The CWS laboratory staff at Fort Mills tested one flame thrower, and sent the ignition system to Edgewood where it was examined for usable features.17
Fuel packs on Japanese flame throwers were similar to the American type. Two cylindrical tanks held a mixture of oil and gasoline, while a small tank contained compressed nitrogen. A rubber hose connected the fuel pack and the flame gun. The ignition mechanism was a revolving cylinder holding ten pyrotechnic cartridges. Japanese weapons were lighter and held slightly less fuel than the American models. The range was somewhat more than the E1R1, M1, and MiAi models, but less than Model M2-2.
The Japanese flame throwers were well made. Colonel Hamilton, chem-
American and German portable flame throwers. U.S. flame thrower M2-2, left; German 1942 model, right.
ical officer in the Philippines, stated his opinion in early 1942 that the “United States principle and basis of flame thrower research and development is somewhat inferior and less practical than the Japanese principle.”18 The Japanese weapon was of course not foolproof. The ignition cartridges were susceptible to moisture, and at times would not throw off a shower of sparks. In many instances in the Pacific Islands American tanks and troops who were fired on escaped with only a harmless shower of oil.
The One-Shot Flame Throwers
One disadvantage of the portable flame thrower was its weight, about 70 pounds including fuel. In December 1942 the chemical officer, Fifth Army, recommended the development of a single-shot flame thrower light enough for a man to carry long distances, and inexpensive enough to be discarded after firing. In addition to its light weight and low cost
the weapon could be shipped from the United States ready for use, thus saving the time of maintenance crews.19
The first crude model was cylindrical in shape, thirty-four inches long, and five inches in diameter. Inside at the butt end was a pressure bottle taken from a carbon dioxide fire extinguisher to provide gas for expelling the fuel. Seated against the pressure bottle was a piston. Between the piston and the nozzle were two gallons of fuel. The weapon was simple to operate. The soldier held it tightly against his body as he might the nozzle of a fire hose, and then pulled the firing pin. This released a steel spike which pierced the gas bottle. Escaping carbon dioxide pushed the piston forward, squirting the fuel out through the nozzle. A railroad fusee ignited the charge. The effective range of the flame thrower was about thirty yards.20
The Army was at first not particularly interested in the one-shot flame thrower and CWS engineers worked on the device only when they could spare time from other jobs. After the portable model proved valuable in jungle fighting, the service hastened development of the one-shot. The NDRC assisted by letting a contract with the Firestone Tire and Rubber Co. Firestone in the summer of 1944 designed a model containing compressed gas stored in a long, steel tube coiled around the outside. The maximum effective range was forty yards, the weapon emptying in four seconds.21
A new method of propelling fuel had meanwhile been suggested—gas from a combustible powder. Powder would eliminate carbon dioxide coils, it would simplify charging the weapon, and it would ease the problem of supply. The NDRC produced a model in which powder gave off gas at a pressure of 1,000 pounds per square inch. This pressure was great enough to shoot unthickened gasoline twenty-five yards. When the war ended the one-shot had reached the stage where the CWS was ready to produce some for battle testing.22
The American Army was not alone in seeking a light flame thrower. Toward the end of the war the German Army produced Einstossflammenwerfer 46, a single-shot close combat weapon for assault troops and paratroopers. The barrel was two feet long and three inches in diameter. It was lighter than the CWS weapon—it weighed only six pounds filled-
but in reducing the weight the designers reduced the fuel capacity to one and one-half quarts. The pressure to expel the fuel was provided by gas from an exploding cartridge. The range averaged thirty yards. Germany seems to have been alone among the enemy countries in producing a one-shot flame thrower; neither Italy nor Japan had this type of weapon.23
Mediumweight Flame Throwers
A disadvantage of the portable flame thrower was the limited capacity of its fuel tanks. During battle they often ran dry, leaving operators with the choice of going to the rear for more fuel or abandoning the weapons. To get around this objection the CWS attempted to increase the capacity.
For the use of Engineer troops in assault operations, the CWS developed a two-man flame thrower. With the exception of the fuel tanks, pressure tank, and minor parts, the design was the same as for the M1 portable. The fuel tanks held twenty-two gallons of fuel, and were fastened on a two-wheeled frame. One operator grasped the handles of the frame and pushed it along like a wheelbarrow, while the other operator carried the flame gun. The fuel charge lasted about thirty seconds, and the range with a mixture of gasoline and oil was around twenty-five yards. The CWS did not produce this flame thrower because the Engineer Board considered it too bulky and heavy for troops.24
Another large weapon for assault and mopping-up operations was the cart-mounted flame thrower. The CWS took a fuel pressure unit designed for flame throwing tanks, mounted it on a chemical mortar cart and connected it to a portable flame thrower gun by means of two hundred feet of high pressure hose. The weapon, filled with twenty-five gallons of fuel, weighed more than six hundred pounds. Two men were needed to pull the cart—more if the terrain was rugged. One man remained at the cart while another dragged the flame gun to the firing point. In tests conducted by the Infantry Board the weapon proved efficient and reliable, but too heavy for foot troops in rough country.25
The CWS designed a manifold portable flame thrower especially for use in the jungle. By means of a manifold assembly the pressure tanks
and fuel tanks of two or more portable flame throwers could be coupled together. Two hundred feet of hose connected the manifold to a special flame gun. The operator would crawl through the underbrush dragging the gun and hose along with him. With this device one operator could fire as much fuel as six men with portable flame throwers. The CWS developed the manifold flame thrower rather late in the war and it was not produced.
The German Army, like the American, found the small fuel capacity of portable flame throwers a disadvantage, and it devised a larger weapon for its troops. Two men pulled the flame thrower, which weighed more than two hundred pounds and was mounted on a dolly fitted with two pneumatic tires. The fuel tank held approximately eight gallons, sufficient for twenty-five seconds of continuous firing. The range was about thirty yards. The Germans limited the distribution of this weapon, and it was not used to any great extent.26
Main Armament Mechanized Flame Throwers
Just as World War II proved the value of portable flame throwers, it likewise demonstrated the usefulness of mechanized flame throwers. The CWS had designed its first mechanized flame thrower in World War I but did not have time to carry the weapon beyond the experimental stage. Intended for installation in a tank, the gun could expel a stream of oil fifty yards. After the war the service put mechanized flame throwers aside and did not work on them again for twenty years.27
The revival of the mechanized flame thrower began in the United States in the late 1930s after newspapers, magazines, and newsreels reported that the Italian Army had outfitted some of its armored units with this type of weapon. The Chemical Warfare Technical Committee laid down tentative military characteristics and in the summer of 1940 engineers constructed the weapon. Tests with this flame thrower uncovered flaws which were corrected in a second model, installed in an M2 medium tank. In the tank the flame gun replaced the cannon (the main armament), and for this reason the flame thrower became known as the main armament type. Two sixty-gallon reservoirs on the floor of the tank held fuel, and three commercial steel cylinders held compressed nitrogen for expelling it. Ignition was provided by propane gas lit by sparks from a spark plug. The
operator elevated and traversed the gun with his left hand and fired with his right.28
With this tank, observers on the Armored Force Board were able to see a mechanized flame thrower in action for the first time. They were not impressed and would not recommend flame throwers for the armored forces.29 Their opinion, widely shared during the early part of the war, kept the development of mechanized flame throwers moving along at a slow pace. The CWS had to give priority to items that the troops did want, and engineers had difficulty getting tanks for further experiments. The opinion of the Armored Board was based on guesses that seemed reasonable at the time, not on experience. Later, as battle showed the value of flame warfare, opinions changed, but by then the CWS had lost irreplaceable development time.
By the summer of 1942 engineers had the third model, mounted in an M3 medium tank, ready for tests. This flame thrower was provided with a larger fuel supply than previous designs, and used pressure from a rotary pump to expel the fuel. The pump eliminated the need for supplying gas cylinders and removed a potential safety hazard. But the pump took too much power from the tank engine, and the whirling blades of the pump smashed the cellular structure of thickened fuel, thinning it out so that it gave the same range as unthickened fuel. Changing back to compressed air, engineers obtained a range almost twice that of earlier flame throwers.30
At the same time the CWS was developing the flame throwing unit for tanks, NDRC engineers had taken up the study of some of the fundamental factors involved in flame throwers such as the design of nozzles and the composition of fuels.31 In March 1942 the NDRC contracted with several firms for a large flame thrower similar to the Ronson type of the Canadian Army. Several models were made, but none passed beyond the experimental stage. During this period, however, the firms gained experience that later made possible the rapid development of model “Q” (for Quickie). Work on “Q” began in November in the plant of the Standard Oil Development Co. Tanks were so scarce that the designers had to
mount the weapon on an old truck. With napalm thickened fuel, “Q” had a range of more than 100 yards.32
By the beginning of 1943 there were thus two fairly satisfactory flame throwers, the CWS model and NDRC’s “Q.” Although theaters of operation had not asked for main armament flame throwers the Army Ground Forces had watched the development of the weapon with approval. In March 1943 CWS arranged a demonstration to allow the AGF to decide which of the two models it preferred. This turned out to be “Q.” The Army decided to place the flame thrower in light tanks, these being the only tanks available.33
Engineers mounted the flame gun, fuel reservoir, and compressed gas cylinders in a turret basket that was interchangeable with the regular turret basket of an M5A1 tank. The reservoir held 105 gallons of fuel that could be discharged at a rate of approximately two gallons a second. The range with ordinary fuel was 30 to 40 yards, with thick fuel 105 to 130 yards.34
Difficulty in getting tanks delayed the installation of flame throwers, and January 1944 arrived before the Armored Board received a weapon for test. By this time the M5A1 tank was no longer in use, forcing the CWS and NDRC to start over and design a flame thrower for the M4 medium tank. The work went forward slowly because the Army did not want to hold tanks in the United States, the CWS lacked engineers for the project, and considerable time was needed to perfect the complex mechanism. Finally representatives of CWS, AGF, ASF, and NDRC agreed that the fastest way to get a main armament flame thrower into action was to modify the earlier “Q.” This was done rapidly, but the M5-4, as the flame thrower was designated, could not be installed until the spring of 1945 because tanks were still scarce. The war ended before the flame throwing tank could be shipped overseas.35
The CWS in the meantime had on its hands four obsolete light tanks that it had rigged up with flame throwers for Armored Board tests. The service did not wish to abandon these tanks since they were serviceable as flame weapons. It had the turrets strengthened and shipped the tanks to Manila early in 1945. The Sixth Army attached them to the 27th Division for assault operations on Myoko Mountain, and to the 38th Infantry Division in the Ipo Dam area. They were the only main armament flame throwers produced in continental United States to see combat.36
Main Armament Flame Throwers Produced in Hawaii
Troops in the Pacific early discovered the value of flame throwers in overcoming fortified positions. By 1944 experience with flame warfare convinced them of the need for main armament flame throwing tanks. The weapons were not available for the reasons noted above and servicemen in the Central Pacific Area produced their own.
The main armament flame thrower turned out in Hawaii was based on the Ronson flame thrower. The Ronson had been developed in Great Britain in 1941, rejected by the British Army because of its short range and low hitting power, and then accepted by the Canadian Army and produced in Canada. The Canadians sent twenty to the Pacific at the request of the V Marine Amphibious Corps. In January 1944 a team of men from the 43rd Chemical Laboratory Company, the Royal Canadian Army, and the V Amphibious Corps installed a Ronson in an LVT, and on 3 February demonstrated it at Koko Head before high-ranking Army, Navy, and Marine Corps officers. The weapon satisfied the officers, who decided that it should be developed further. Shortly thereafter the Marine Corps stepped aside, leaving Col. George H. Unmacht, CWS, responsible for future work.37
The medium tank, rather than the LVT, was the most suitable mount for the Ronson, but as in other instances, medium tanks were needed for their high explosive firepower and were scarce. Only the obsolescent M3A1 light tank could be obtained from Ordnance. The 43rd Chemical
Laboratory Company removed the 37-mm. cannon from a tank and replaced it with a flame gun, protected by a howitzer-like shroud. To lessen the possibility of mechanical failure and to permit easier operation, the company redesigned the entire electrical system. The Honolulu Iron Works fabricated the fuel reservoirs. Satan, as the finished tank was called, used compressed carbon dioxide as the propellant gas, had a fuel capacity of 170 gallons, and a range of 40 to 60 yards with oily fuel or of 60 to 80 yards with thickened fuel. It was demonstrated on 15 April before Army, Navy, and Marine officers, and the following day Lt. Gen. Holland M. Smith asked for twenty-four Satans.
In June on Saipan, marines employed the Hawaiian flame throwers for mopping up and, after the operators gained experience, in assaults. The tanks attacked dugouts, cane fields, buildings, and caves. Marines then shipped the tanks to Tinian and used them in the assault waves and in overcoming Japanese strongholds.38
These two islands served as proving grounds for the tank mounted flame thrower, and brought out faults in the design. The scarcity of napalm restricted the weapons to oily fuel, which gave a shorter range than marines wanted. Before the weapons were employed again, Colonel Unmacht’s group overcame some of the flaws, and the supply channels opened to the flow of napalm.
In September the Tenth Army, planning an attack—later canceled—on Formosa, requested that large capacity flame throwers be installed in fifty-four M4 medium tanks. In the first model the company installed a Ronson gun like that used on Satan. The Tenth Army pointed out that the silhouette was different from the 75-mm. gun of the regular M4 tank, and this would permit the enemy to spot flame tanks. The problem then arose of obtaining 75-mm. gun tubes to enclose the flame gun. Ordnance did not wish to sacrifice new tubes for the manufacture of flame throwers, and only a few salvaged tubes were available. Seabees machined these items to hold the flame gun, and in the meantime the ASF shipped fourteen salvaged tubes under high priority. Later, Colonel Unmacht obtained authority to use forty-two serviceable tubes for the purpose.
The new flame thrower tank, designated POA-CWS “75” H-1 (POA for Pacific Ocean Areas, H for Hawaii), was demonstrated to the Tenth
Army about 1 November. The weapon used compressed carbon dioxide gas to propel the fuel, had a fuel capacity of 290 gallons, a range of 40 yards with oily fuel and 60 to 80 yards with thickened fuel. Eight tanks were sent to the Fleet Marine Force, Pacific, for the Iwo Jima operation and 54 were supplied to the 713th Provisional Flame Thrower Tank Battalion for the Ryukyus operation.
On Iwo Jima, marines found the flame tanks particularly helpful in the later stages of the operation when they had to take a network of caves. By the time the marines had reached the northern end of the island, flame tanks had proven so useful that demands for them exceeded the supply. On Okinawa the operations took place on the hilly southern portion of the island where Japanese troops had defenses in cliffs, hills, and escarpments. The tank battalion carried out more than six hundred attacks, and fired almost 200,000 gallons of napalm thickened fuel.39
Troops on Okinawa employed an ingenious hose extension against caves that were out of range of tanks. The Navy donated fifty-foot lengths of fire hose which the men coupled together to form a hose four hundred feet long. They fastened one end of the hose to the fuel reservoir of the tank, and attached an M2-2 portable flame gun to the other end. In action the tank parked as close as possible to the target, the operators dragged the hose to a position within range, the tank pumped fuel through the hose, and the nozzleman ignited the fuel and directed the flame at the target. The extension was used with good results on a number of occasions.40
Troops objected to main armament flame throwers chiefly because they replaced the tank’s cannon. To meet this criticism, Colonel Unmacht’s staff drew up plans for mounting flame throwers alongside the cannon instead of replacing it. Work began in late 1944 when Fleet Marine Force, Pacific, anticipating an invasion of Japan, asked for at least seventy-two main armament flame throwers. Most of the tanks provided by the marines carried 75-mm. guns. The remainder carried 105-mm. howitzers. By judicious planning, designers arranged the interior of the vehicles to allow
M4 medium flame-thrower tank in action on Okinawa, May 1945
the storage of forty rounds of 75-mm. or twenty rounds of 105-mm. shells without decreasing the quantity of flame-thrower fuel. Handicapped by a scarcity of parts and a dearth of machinists and other craftsmen, the work proceeded slowly. During the battle on Okinawa the Tenth Army asked for eighteen of these tanks. They were on their way to that island by the time the battle ended, and were rerouted, instead, to the Marianas to equip the Marine division rehabilitating there. Seventy tanks were ready for the invasion of Japan when the operation was called off and the war came to an end.
The Marine Corps was not alone in using mechanized flame throwers. The Navy, after seeing a demonstration of model “Q” in December 1943, recognized the effectiveness of flame throwers in amphibious assaults. In January 1944 the Bureau of Ordnance ordered a model mounted on an LCVP or LCM. The complete unit, designated the Navy Mark I flame thrower, had a fuel capacity of 300 gallons and a maximum range of approximately 110 yards with napalm thickened gasoline. The unit proved too heavy for an LCVP and the Chemical section at Pearl Harbor tried mounting it in an M3A5 tank. But the excessive weight (which reduced the mobility of the tank), the limited flame gun traverse of 60°, and other
Lt. Gen. Wilhelm D. Styer standing in front of a flame thrower mounted on a medium tank chassis at Schofield Barracks, Hawaii. To the right of General Styer is Col. Jay C. Whitehair, Howard C. Peterson, and Col. George F. Unmacht.
difficulties ended the experiment. Finally an LVT-4 lightly armored amphibious tractor was found to be a fairly satisfactory mount. Six of these were used in the mopping-up phase of the Peleliu operation, particularly around the caves of Umurbrogol mountain. The LVT-4 was not the ideal carrier and the crews were kept busy making repairs but the flame thrower itself proved rugged and dependable.41
Under Colonel Unmacht the composite group of Army, Navy, and Marine personnel produced more main armament flame throwing tanks than did engineers in the United States, but the problems which they faced were much simpler. The Army, Navy, and Marine Corps in the Pacific were anxious to get the weapons and gave the Chemical section all necessary assistance and supplies, whereas Edgewood could not get the Army to set up a requirement for a main armament flame thrower and had great
difficulty obtaining tanks for conversion into flame throwers. American forces on one side of the world wanted the weapon, gave technicians full support, and got it. Forces on the other side of the world were indifferent toward the weapon, gave technicians little support, and did not receive it.
Enemy armies had mechanized flame throwers, but used them infrequently. Indeed, there is no record of Italians or Japanese using flame tanks against United States troops. The Italians had had tank mounted weapons since the Ethiopian War, 1935–1936. The tank was quite light and mounted the flame gun coaxially with a machine gun. Fuel was carried in a trailer with a capacity of one hundred gallons. The range was rather low, from thirty to forty-five yards. The reasons why the Italians did not employ their tanks are not known. Among them may have been the lack of opportunity, unfamiliarity with flame thrower tactics, and conservatism of tank commanders.42
American troops did not see Japanese flame throwing vehicles until they captured eight on Luzon in 1945. The weapons were placed on amphibious tractors, similar to American DUKW’s. The Japanese did not have fuel thickeners comparable to American napalm, and had to use mixtures of crude oil, gasoline, and kerosene. Since Japanese troops employed portable flame throwers against Americans from early 1942 onward, it is difficult to explain why they did not use mechanized flame throwers. American troops learned by trial and error of the value of flame tanks, and perhaps the Japanese never threw off their conservatism sufficiently to give the tanks full-scale battle tests.43
The German Army was equipped with a Pz. Kw. III tank carrying a flame gun disguised as a cannon. The fuel capacity was more than 200 gallons, the range about 40 yards. In addition the Germans had armored half-track vehicles carrying 2 large flame throwers on the sides and a small flame thrower at the rear. Each large flame thrower had a fuel reservoir of 95 gallons, and a range of 40 yards. The Germans also developed a trailer-mounted flame thrower that could be hauled behind a truck or tank. The 2-wheeled trailer carried a 40-gallon fuel reservoir, and was protected by armor plate. Only a few of these were ever manufactured. German flame fuels did not have particularly good thickening agents. The best was aluminum alcoholace, but this was not as good as napalm. Generally German
thickened fuels had little more, if any, range than American unthickened fuels.44
Auxiliary Mechanized Flame Throwers
The auxiliary mechanized flame thrower was smaller than the main armament type, and at the most replaced only a machine gun in the tank. Since the tank retained its cannon, tank commanders did not raise as strenuous objections to the auxiliary flame thrower as they did to the main armament type. Warfare on the Pacific islands provided the impetus for the development of the auxiliary flame thrower. After the Battle of Guadalcanal infantrymen and marines tried to devise ways of putting portable flame throwers in tanks so as to protect the operator from Japanese rifle fire. One group of soldiers drilled a hole in the armor of the tank near the bow machine gun and shoved through a flame gun. They fired the flame thrower, and then ignited the fuel by means of tracers. The 3d Marine Division modified a flame gun to fit the bow machine gun ball-and-socket joint. For the invasion of Kwajalein the 7th Infantry Division mounted flame throwers on LVTs and on light tanks. None of the improvised mountings were successful until February 1944 at Bougainville when tanks carrying portable flame throwers supported an infantry assault against Japanese positions along the Torokina River.45
The idea of installing portable flame throwers in tanks was sound. The difficulty lay in the short range, small fuel capacity, and delicacy of the weapon which had not been designed to withstand the jarring vibration of tanks. Because troops constantly tried to improvise auxiliary flame throwers, the Army Service Forces in October 1943 requested the CWS to develop a standard model. Service engineers took the portable flame thrower and modified it to fit the bow machine gun ball-mount. Fuel reservoirs were designed for M3, M4, and M5 tanks. It took troops in the field several hours to install fuel reservoirs, but thereafter the operator could remove the machine gun and insert the flame gun in a minute or two. The flame thrower could fire a gallon of fuel per second to an effective range of 25 to 30 yards with oily fuel, 50 to 60 yards with thickened
M3A1 light tank equipped with flame gun firing during a demonstration, New Zealand, October 1943.
fuel. The CWS procured 1,784 model M3-4-3 bow flame throwers with fuel capacity of 50 gallons for M4 tanks, and 300 model E5R2-M3 with a capacity of 10 gallons for M3 or M5 tanks. Many bow-type flame throwers saw action in the European Theater, in the Marianas operation, on Peleliu, Luzon, and other islands.46
Tank commanders were not all in favor of the bow flame thrower because it deprived them of an important machine gun. An alternative was to mount the flame gun in the turret alongside the periscope. The CWS and its contractors produced several periscope models, one of which (M3-4-E6R3) went into production in 1945 too late for war use.47
The Italian and Japanese Armies did not equip tanks with auxiliary flame throwers, but the German Army had two flame throwers that might be placed in this classification. For the Pz. Kw. II tank it developed a
Flame gun to fit .30-cal. machine gun mount of tanks is demonstrated to Seventh Army soldiers, France, February 1945.
small flame thrower mounted in an individual turret. Tanks carried two of these turrets on the bow. Behind each turret was a small cylinder of acetylene for ignition and a large reservoir for thirty-five gallons of fuel. Electrical controls for traversing the turrets and firing the flame throwers were in the tank. The weapons could fire eighty bursts, each lasting two to three seconds, a distance of thirty-five yards. For the Pz. Jg. 38 tank the Germans provided a projector flame gun similar to the portable Flammenwerfer 41. The fuel was held in three reservoirs with a capacity of 154 gallons, and was pumped to the gun by a gasoline engine. The weapon could fire twenty-four bursts, which was the number of cartridges carried in the ignition magazine. The German Army seems to have made little use of auxiliary flame throwers.48
Auxiliary Flame Throwers Made in Hawaii
Marines and infantrymen in the Pacific had hastened the development of auxiliary flame throwers and they were anxious to obtain the weapons.
Finally in 1944 the Tenth Army could wait no longer for weapons to come from the States. It asked the Chemical Section, USAFPOA, then engaged in producing main armament flame throwers, to provide at least eighty-eight periscope flame weapons. Colonel Unmacht’s composite group constructed a model based on an experimental periscope flame thrower gun manufactured by the Armored Board at Fort Knox, which in turn had been based on the M1A1 portable flame thrower gun. The gun was constructed from a piece of metal tubing, and passed through a 1½-inch hole in the periscope ring. The fuel reservoir developed by technicians at Edge-wood for the bow auxiliary flame thrower was used when it was available, but when it was not similar containers were manufactured at Pearl Harbor. Compressed air or nitrogen was used as the propellant. With unthickened fuel the range was twenty yards; with thickened fuel, twice that distance. The weapon could be mounted in any of the M4 medium tanks.49
One hundred and seventy-six of these were completed for the Tenth Army. They were on hand for the Okinawa and Iwo Jima campaigns, but they were not used extensively because Hawaiian-produced main armament flame throwers, preferred for their large fuel capacity and greater range, were also available for these battles.50
The demand for auxiliary flame throwers waxed and waned during the war. When the Army first used them in 1944, troops in the Pacific liked them because they had a longer range and carried more fuel than the portable flame thrower, and because armor allowed the flame thrower operator to approach Japanese positions without fear of rifle fire. But when Hawaiian main armament flame throwers became available and troops had an opportunity to compare the auxiliary and the main armament in battle, their choice swung toward the latter type with its larger fuel capacity, greater range, and all-around mobility. By the end of the war the CWS and NDRC were concentrating their efforts on the development and production of large main armament weapons rather than small auxiliary models.
Incendiary Projector for Airplanes
In 1941 CWS engineers made some experiments to see if a flame thrower could be mounted on an airplane. They constructed a model and tested it on the ground. The turbulent blast of air from the propeller extinguished the flame and the CWS dropped the project. Late in 1943 the Army Air Forces requested the service to resume work. It was reasoned that a rain of burning thickened fuel dropping from the sky might be used to set fire to enemy-held jungle, grassy areas, and supply dumps. At this time the British were working on an airplane flame thrower and the CWS profited from their experience. CWS and NDRC engineers designed a fuel container of the size and shape of the 4,000-pound bomb, fitting it with a discharge pipe and nozzle similar to that used on toxic spray tanks. Rocket motors provided gas that expelled the 210-gallon load of thickened gasoline in one spectacular burst. Technicians tested the projector, aeroflame, as the weapon was called, on a wooden tower and in actual flight on B-25H airplanes. Although the aeroflame developed into a workable weapon, the CWS decided that droppable gasoline tanks were more effective in starting fires, and in 1944 it stopped the project for good.51
Emplaced Flame Throwers
After Dunkirk, Britain’s eastern coast lay vulnerable to invasion. As part of the defenses against German landings the Petroleum Warfare Department installed flame throwers along the Channel beaches. From underground storage tanks holding forty-five tons of fuel—a mixture of gasoline, kerosene, and diesel oil—pipes led outward beneath the sea. Valves on the ends of the pipes could be opened by remote control, releasing oil which floated to the surface. Ignited by naval flares, the pools of oil would burn fiercely for hours. The British also emplaced flame throwers along roads to burn enemy vehicles and troops that might get ashore.52
Taking their cue from the British, the CWS developed a remote controlled flame thrower to be used to guard U.S. airfields and beaches. The steel fuel tank holding 120 gallons of oil could be placed in a pit along with two cylinders of compressed nitrogen. Pipes led above ground to 2 flame thrower nozzles. The nozzles pointed in opposite directions, and threw streams of burning oil 40 yards. A wall of flame 80 yards long,
lasting about a minute and a half, could be set up. The CWS standardized the device in 1941, but the Army found no use for it and discarded it at the end of the war.53
The first recorded use of emplaced flame throwers was by the Russians in the defense of Moscow. A line of flame throwers was set up in the outer defense zone that the Soviets had built in front of their capital to block the German Army. The body of the flame thrower was a cylindrical tank twenty-one inches high, twelve inches in diameter, and holding eight gallons of oil. Pressure was furnished by gas given off from burning powder. These flame throwers were dug in at intervals of about thirty yards and covered with stones, earth, or other natural materials as camouflage. Detonated electrically the weapon threw out a stream of flame lasting a second or two. When the Germans went on the defensive they employed emplaced flame throwers similar to the Russian type. The Japanese in Burma set up some field expedients, but these were of little value.54
The British flame weapons emplaced along the channel might have been useful barriers to an invasion, but the Russian and German weapons of this type probably had only a temporary psychological effect since they suffered from a limited range and fuel capacity. Furthermore, there was considerable labor in digging them in, the aim could not be changed once they were emplaced, they could only be fired once, the enemy might bypass them, and their control wires could be easily damaged by shell fire. On the whole, World War II experience showed that the emplaced flame thrower had only slight value in warfare.
Servicing Flame Throwers
The job of the CWS did not end with the production of flame throwers. Weapons had to be serviced, fuels had to be mixed with thickeners, and compressed gas or air had to be supplied. The CWS rigged up an air compressor and mixing vat carried on a truck or trailer. With this equipment troops could mix the hundreds of gallons of thickened fuel used by flame tanks, and compress the large volume of air needed as a propellant.55
Portable flame throwers did not require as much fuel and propellant as mechanized types, but nevertheless problems existed. While napalm and gasoline could be carried in cans and mixed when needed, a machine weighing 800 pounds was required to furnish compressed air. Such a compressor could be carried only on a jeep or truck. The CWS decided to develop a small, light, gas producer. With cooperation of the NDRC, it turned out a generator weighing sixty pounds and containing lithium hydride. When water was allowed to drip on the compound, hydrogen under a pressure of 2,000 pounds per square inch was evolved. This hydrogen could be piped into flame thrower gas tanks. The service started producing these generators in the spring of 1945, and a few reached the theaters before the end of the war.56
Toxicology of Flame Attack
A year after the initial operations in the Pacific the CWS began studies to discover the predominating characteristics and cause of death by flame and to learn what defensive measures might be devised against the use of flame by the enemy. The agencies engaged in various aspects of flame attack research with the CWS included NDRC units at the Massachusetts Institute of Technology, the Standard Oil Development Co., New York University, Harvard and Johns Hopkins Medical Schools; units of the Bureaus of Ordnance and Medicine of the Navy Department, the Armored Medical Research Laboratory; and the Experiment Station at Suffield, Canada.57
In studying the toxicology of flame attack in poorly ventilated enclosed spaces like those found in Japanese bunkers and similar fortifications, researchers determined that three important changes occurred within them at the moment of flame attack, quite aside from the penetration of the flaming fuel itself: there was a sudden jump in temperature, lethal concentrations of carbon monoxide were built up in the bunker, and there was a dangerous lowering of oxygen content. They learned that 70 percent carbon monoxide in the blood resulted in unconsciousness and frequently in death and that this accumulation was obtainable in flame attack within two minutes. Furthermore, only one-tenth of one percent
carbon monoxide in the air was sufficient to maintain this lethal blood level, and it was present in bunkers for seven to ten minutes after flame attack. They also learned that for intervals up to fifteen seconds there was almost complete absence of oxygen in a bunker under attack, and that unconsciousness would likely be almost instantaneous in such an event. Any one of these factors or any combination of them, therefore, meant certain death, quite aside from the effects of direct contact with the flame.58
Work on flame defense led to the construction of a hood-type mask built to withstand 1,000° F. for one minute, to the development of a steel sliding door for pillbox apertures, to experimental fireproof clothing and water fog, and to spray extinguishers, all of which proved unsatisfactory. The CWS finally concluded that no positive defense could be devised against flame attack.59
Flame throwers were not major weapons in the same sense as cannon, rifles, and bombs. Rather they were weapons that proved valuable in certain tactical situations. The men in the Pacific, the locale of most of these situations, did much to bring about the improvement of flame throwers. Americans started work on them later than Europeans and Japanese, but while enemy armies did not push the development of the weapon, Americans, particularly in the Pacific, called for it more and more frequently as the war progressed. Despite the fact that the American achievement in flame thrower development and production does not look impressive, it surpassed that of the enemy during the same period.