Share This Article

Long before the United States officially entered World War II, it was the world’s biggest exporter of fighter aircraft.

From early 1938 through the end of 1941, the United States, though still technically neutral in the conflict that was igniting around the world, sold more combat aircraft to foreign governments than any other nation. And fighters, because they could be built quickly and cheaply, sold in greater numbers—more than 5,600 ordered by December 1941—than any other type. The surge in demand for exports not only provided a needed boost to America’s depressed aircraft industries early in World War II, it also gave the nation’s military planners vital information on which to base requirements for future combat planes.

From 1939 onward, reports from Europe and the Far East indicated that American-made fighters were deficient in most respects— performance, firepower and armor protection—when compared to their enemy counterparts. Consider, for instance, what U.S. Marine Captain Phillip R. White had to say following a mission during the Battle of Midway on June 4, 1942, when 12 of 19 Brewster F2A-3 Buffalos of VMF-221 were shot down by Japanese fighters: “It is my belief that any commander who orders pilots out for combat in an F2A should consider the pilot as lost before leaving the ground.”

But for some beleaguered nations, American aircraft represented the best available defense against a seemingly overwhelming enemy. For example, Finland’s top fighter ace, Sergeant Eino Ilmari Juutilainen, scored 33 of his 941⁄6 aerial victories over Soviet aircraft while flying Brewster B-239 Buffalos sent to Finland by the U.S. in 1940. Despite the Buffalo’s poor reputation, Juutilainen wrote in his autobiography Double Fighter Knight: “Clearly, the best fighter arriving during the temporary peace [early 1940 to mid-1941] was the American designed Brewster 239 Buffalo….They had speed, agility and good weaponry too….[We] were happy to take them anywhere to take on any opponent.”

The following overview of American export fighters is limited to aircraft ordered prior to December 1941. In a number of instances, however, they were delivered later or taken into U.S. service. Since many of these airplanes never received U.S. serial numbers or failed to reach their destinations, the production and delivery numbers cited must be qualified as best estimates.


Model 239 and 339: These airplanes corresponded to U.S. Navy models F2A-1 (Model 239), F2A-2 (Models 339B, D and E) and F2A-3 (Model 339-23). The 44 Model 239s exported to Finland in early 1940 had originally been included in the Navy’s 1938 contract for 54 F2A-1s, so their tail hooks and other carrier-related equipment were removed prior to shipment. In contrast to the generally poor combat record of the type, the Finnish Brewsters, flown in action from mid-1941 to mid-1944, accounted for 477 aerial victories against the Soviet air force.

The Belgians ordered 40 Model 339Bs in late 1939, the first of which arrived in April 1940—only to be captured by the Germans. Six other 339Bs were diverted to Martinique; the remaining 33 were taken by Britain.

The first of the 170 Model 339Es purchased by Britain, which received the RAF designation Buffalo I, arrived in December 1940, but following evaluation they were deemed unsuitable for European combat. All British Buffalos were thereafter assigned to the Far East, equipping five squadrons (two RAF, two Royal Australian Air Force and one Royal New Zealand Air Force) during the defense of Singapore, Malaya and Burma in late 1941 and early 1942. Seventy-two Model 339Ds sold to the Netherlands for use in the Dutch East Indies (Java) were delivered between March and June 1941, and were in frontline service when the Japanese attacked.

In combat, the British Buffalos and Dutch 339Ds were no match for Mitsubishi A6M2 Zeros and Nakajima Ki.43 Oscars, and all but a few were destroyed. The final batch of 20 Model 339-23s, ordered by the Netherlands but not completed until after the fall of Java in early 1942, were transferred to RAAF units in Australia.


Model 14: The Model 14 was the export version of the U.S. Army Air Corps’ P-39 Airacobra. A contract for 170 airplanes, which had originally been ordered by France in April 1940, was subsequently taken over by Britain in June 1940 and simultaneously increased—to a total of 675 aircraft.

Very similar to the P-39D, the Model 14 differed in having a 20mm Hispano cannon in the nose in place of the 37mm, two .303-inch guns in the nose and four .303 guns in the wings. The production Model 14 (originally named Caribou but later changed to Airacobra I) began arriving in England during the summer of 1941 and equipped one Royal Air Force squadron for evaluation purposes. But after flying only one low-level operational sortie in late 1941, the British declared the type unsuitable for combat and withdrew it from frontline service. Of the original Airacobra I order, 212 were redirected to the USSR, 54 were lost at sea and the remaining 179 were taken into U.S. Army service as the P-400. Many Soviet pilots loved the Kobra, as they called it, using it for low-level aerial combat and close air support.


Model G-36: Though the F4F Wildcat is best remembered for holding the line during the early Pacific War, the very first production model to fly (in May 1940) was in fact a G-36A export version that had been ordered for the French navy. The G-36A differed from the F4F-3 in having six 7.5mm guns and a single-row Wright R-1820 engine in place of the twin-row Pratt & Whitney R-1830, but when the British afterward assumed the French contract for 81 planes, the aircrafts’ armament was respecified as six .50-caliber guns. De – spite the urgent need to reequip Navy and Marine fighter squadrons, production priority at Grumman was split with the British.

The first G-36As, which were taken into the Fleet Air Arm as the Martlet I, began to arrive in England in October 1940. Just two months later the type was used to score its first shoot-down, a Junkers Ju-88 off Scapa Flow.

Britain ordered 100 improved G-36B Martlet IIs in October 1940, and those airplanes, featuring R-1830 Wasp engines and folding wings, began reaching FAA squadrons in the spring of 1941. Martlett IIs were subsequently embarked aboard British carriers. A detachment flying from the escort carrier HMS Audacity destroyed five Focke Wulf Fw-200 Condors during the last three months of 1941. In April 1941, 30 F4F-3As (equipped with four .50-caliber wing guns) that were en route to Greece were seized by the British at Gibraltar and taken into service as Martlet IIIs.


Model NA-50 and NA-68: In mid-1938 the government of Peru contracted with North American to develop a scaled-down, more powerful derivative of its NA-26 (i.e., BC-1) two-place trainer. The resulting single-place NA-50—which was powered by an 840-hp Wright R-1820 engine and equipped with two .30-caliber machine guns in addition to bomb racks—flew in February 1939. The order for seven aircraft was completed by May 1939, and the NA-50s later saw action in Peru’s border war with Ecuador in 1941.

Thailand ordered an improved version of the fighter, the Model NA-68, featuring a revised cowling and reshaped rudder, in late 1939. The NA-68’s planned armament consisted of two .303 guns in the nose and two .303s in the wings, with the option of two 20mm cannons mounted in underwing pods. Although six unarmed examples of the NA-68 were completed in late 1940, they were commandeered by the Air Corps soon afterward and taken into U.S. service as P-64s, serving as advanced trainers.

Model NA-73: This has to rank as one of the most fortuitous developments in American aviation: In April 1940, when the British Purchasing Commission approached North American to license-build Curtiss H87As for the RAF, the company’s president, J.H. “Dutch” Kindelberger, counteroffered with a proposal to build an entirely new design around the same Allison V-1710 power plant. The British agreed—but with the proviso that a prototype of the new design must be available within 120 days. The NA-73, which made its first flight in October 1940, was destined to become what many consider the best all-around piston-engine fighter of WWII, the P-51 Mustang.

In May 1940, the British placed an order for 320 aircraft (later in – creased to 620) as the Mustang I, specifying armament of two .50- caliber machine guns in the nose plus two .50s and four .30s in the wings. When the U.S. government approved the sale to Britain, it specified that two of the airplanes be allocated to the Army as the XP-51, and the first example was flown to Wright Field in August 1941.

Meanwhile, the first production Mustang I reached England in the late fall. The new fighter quickly demonstrated its superiority over the Spitfire and Curtiss Kittyhawk in both speed and maneuverability at low altitudes. Mustangs Is began entering operational service with the RAF in April 1942; most were issued to squadrons in North Africa as replacements for Tomahawks.


Model 2PA: Developed in parallel with the Army P-35 and billed as the “convoy fighter,” the 2PA was a two-seat adaptation that could be armed with four fixed machine guns and one flexible gun in the rear cockpit, and was capable of carrying up to 500 pounds of bombs. In March 1938, two aircraft were delivered to the Soviets as the 2PA-A and 2PA-L, along with manufacturing rights, though none were ever produced there.

Twenty R-1820-powered examples armed with one flexible and two fixed .30-caliber machine guns were secretly sold to Japan in 1938 as 2A-PBs. They went on to serve with the Imperial Japanese Navy as the A8V1, known to the Allies by the code name “Dick.” Sweden ordered 52 planes as the R-1830-powered 2PA-BX, though only two had been delivered before the order was seized in late 1940 by the U.S. The remaining 50 2PA-BXs went into service with the Army as the AT-12 Guardsman advanced trainer.

Model AP-9: This was a military derivative of the AP-9 racer (flown by Frank Fuller and Jacqueline Cochran)—essentially a P-35 airframe with inward-retracting landing gear—apparently marketed to export customers along with the EP-1. No orders were received, and the sole example was sold to the Dominican Republic in 1939.

Model EP-1: Identical to the Army P-35 except for two .50-caliber guns added to the wings, the single EP-1 demonstrator, like the AP-9, was sold to the Dominican Republic after unsuccessful efforts to obtain other production orders. But soon afterward 120 planes listed as the EP-106 were purchased by Sweden under two separate contracts dated mid-1939 and early 1940. Delivery of the first 60 Swedish planes began in January 1940, and was completed by June. The aircraft type entered service with the Flygvapnet as the J9. In October 1940, however, while the second batch of EP-106s was still being manufactured, they were requisitioned by the U.S. government and taken into the Army as the P-35A.

In November 1941, 51 of the ex-Swedish P-35As were rushed to the defense of the Philippines, where they soon saw action against attacking Japanese forces. Within four days, all but 12 had been destroyed. A number of the P-35As not sent to the Philippines were eventually transferred to Ecuador.

Model AP-4A/P-43A-1: The last P-35 on the assembly line was lengthened and modified to accommodate a turbo-supercharger plus the inward-retracting landing gear of the AP-9. Following delivery to Wright Field in February 1939, it was designated the XP-41 (AP-4). Testing of the XP-41 led to a number of refinements—the cockpit area was lowered and the aft glazing replaced by a raised spine, and the supercharger intake was moved to the bottom of the cowling—and 13 service test examples were ordered as the YP-43.

Though the Air Corps was not overly impressed with the YP-43’s performance during trials conducted in late 1940, it nevertheless placed an order for 54 P-43s (with the R-1830-47, two .50-caliber and two .30-caliber guns), followed by 80 P-43As (with the R-1830-49 and four .50-caliber guns), simply to keep the struggling Republic company afloat while its XP-47 was under development (a wise move, as it turned out).

The Army contracts were augmented in mid-1941 by an order for 125 aircraft, which would be Lend-Leased to China as the P-43A-1. A total of 108 of those airplanes had arrived by March 1942. In limited action against the Japanese, Chinese P-43s reportedly suffered from poor maneuverability and climb plus leakage problems with their self-sealing tanks.


Hawk 75: An export version of the Air Corps’ P-36, the H75 was sold in both fixed- (H75H, M, N and O) and retractable-gear (H75A-1 through A-9) variants. Fixed-gear H75s were powered by 875-hp Wright GR-1820 Cyclone engines and initially armed with two synchronized machine guns in the nose, one .50-caliber and one .30-caliber. Curtiss-Wright delivered the first batch of 30 H75Ms to China in mid- 1938. The type’s poor showing in combat against similar Nakajima Ki.27 Nates and Mitsubishi A5M Claudes has been attributed more to inferior pilot training than to the quality of the aircraft itself.

Twenty-five H75Ns delivered to Thailand during 1939-1940 differed in having two additional .30-caliber machine guns mounted in wing bays. Thai Hawks participated in action against the French in Indo china in early 1941, and briefly flew against the Japanese in December 1941. Curtiss sold 29 H75Os to Argentina in 1939 and 1940 with extra provision for two 23mm Madsen cannons under the wings. Argentina subsequently obtained a manufacturing license from Curtiss and reportedly completed another 20 of the type at its Cordoba plant.

The retractable-gear 75A Hawk was very similar to the P-36A/C, with variations in armament, instrumentation and power plants. The 1,050- hp Pratt & Whitney R-1830 Twin Wasp powered the H75A-1, -2, -3 and -6, and the 1,200-hp Wright GR-1820 geared Cyclone powered the A-4, -5, -7 and -9. In May 1938, France ordered 100 H75A-1s armed with four FN 7.5mm machine guns, with deliveries commencing later that year, followed in early 1939 by an order for 100 H75A-2s armed with two more 7.5mm guns. All 200 had reached France by mid-1939.

When hostilities commenced in September 1939, the A-1 and -2 could hold their own against the Messerschmitt Bf-109D—better, in fact, than contemporaneous French-made fighters such as the Morane Saulnier M.S.406 and Bloch MB.152—but was seriously outgunned by the cannon-armed Me-109E. France ordered 285 improved H75A-3s, with 1,200-hp R-1830-17 engines, in the fall of 1939, but only 168 had arrived before the armistice with Germany in June 1940. Britain assumed the balance of the contract and put the aircraft into service with the RAF as the Mohawk III. In late 1939, France had also contracted for 285 Cyclone-powered H75A-4s, but only six had been delivered by the time French forces surrendered. Twenty-three of the A-4s already en route were diverted to French possessions in Martinique and Guadeloupe, another 30 were lost at sea and the remainder were allocated to Britain as the Mohawk IV. The only RAF Mohawks to actually see combat served in northern India during 1942 and 1943.

The Cyclone-powered factory H75A demonstrator was shipped to China in early 1939 for evaluation. In addition to standard P-36A armament of one .50-caliber and one .30-caliber in the nose, it could carry a 23mm Madsen cannon mounted under each wing. The Chinese planned to license-build the type as the H75A-5 at their factory in Loiwing. One pattern aircraft built for that purpose did reach Burma in 1940; however, the factory was subsequently destroyed in a Japanese bombing raid, and the entire program, plus the pattern aircraft, was transferred to Hindustani Aircraft in India. Only five A-5s had been completed when the program was terminated in mid-1942.

In September 1939, the Norwegian government ordered 24 H75A-6s, identical to the Twin Wasp–powered A-1 except in small details, all of which had arrived in Norway by early 1940. But when Norway capitulated to the Germans in May 1940, at least 13 of these planes were captured and, ironically, resold to Finland in mid-1941. Altogether the Germans sold 44 H75As, mostly ex-French survivors, to the Finns, who flew them in combat against the Soviets between 1941 and 1944, ac – counting for 190 victories for only 24 losses.

Curtiss delivered 20 Cy – clone-powered H75A-7s, armed with two .50-caliber machine guns in the nose and two in the wings, to the Netherlands in mid-1940 for its East Indies air force. Thirty-six H75A-8s, ordered by Norway in 1940 but not completed until after the surrender, were seized by the U.S. and entered Air Corps service as the P-36G. Twenty-eight of these were later transferred to Peru under Lend-Lease; others were assigned to the Norwegian training establishment near Toronto, Canada, as fighter trainers.

The final export examples, 10 H75A-9s delivered to Iran in mid- 1941, were discovered by occupying British forces a year later still in their shipping crates. They were sent to Karachi, assembled and issued to RAF squadrons in northern India.

Hawk 81A: This was the early export derivative of the P-40A/C series, powered by a liquid-cooled 1,040-hp Allison V-1710 engine. All the 140 H81A-1s originally ordered by France in late 1939 were taken over by Britain as the Tomahawk I and began reaching England in September 1940. The planes retained French instrumentation and were armed with two .50-caliber synchronized guns in the nose and two 7.5mm guns in the wings.

The A-1s were followed by 110 H81A-2 Tomahawk IIs, which started arriving in October 1940 and differed in having four .303 guns in the wings, pilot armor and self-sealing fuel tanks. A total of 558 Tomahawk variants were delivered to England before the end of 1940, including an additional 308 H81A-3 Tomahawk IIBs armed with six .303 guns in the wings, followed by another 528 Tomahawk IIBs, which reached England during 1941. One hundred H81A-3s withheld from the British contract were earmarked for China to be used by the American Volunteer Group (soon to be better known as the “Flying Tigers”), 90 of which had arrived overseas by December 1941.

Most of the RAF Tomahawks saw service with squadrons in the Desert Air Force (North Africa) or the Middle East, and a number of these were reportedly later transferred to the USSR and Turkey. The famous shark’s mouth markings commonly associated with P-40s serving in the AVG actually originated with No. 112 Squadron, RAF, for its Tomahawks in North Africa.

Hawk 87A: Yet another evolution of the outdated P-36 airframe, the H87A illustrates what can happen to an aircraft company when it is run by Wall Street financiers rather than aviation visionaries. The design, originally ordered by the British in May 1940 as the Kittyhawk I, involved installation of a 1,150-hp Allison V-1710-39 engine and modifications to the radiator and top intakes, which gave the nose a deeper contour. The canopy was enlarged to improve visibility, and more armor added to the cockpit area. Top speed was 10 mph better than the Tomahawk IIA/B, and the aircraft’s useful load and range were slightly increased.

The first 20 of 560 Kittyhawk Is, armed with four .50-caliber guns, were delivered in August 1941; the remainder, with six .50s, had arrived by the end of December. Most went directly to North Africa, where they replaced Tomahawks, but 72 were sent to Canada and a smaller number were reportedly shipped to Russia and Turkey.

Within a similar time frame, the type began entering service with the U.S. Army as the P-40D (four guns) and P-40E (six guns). A third British contract awarded in May 1941 resulted in the eventual delivery during 1942 of 1,500 more H87As as the Kittyhawk IA, with many allocated to the Commonwealth forces of Australia and New Zealand. These H87As shared the assembly line with P-40Es destined for the Army Air Forces and included 30 more H87As shipped to China in March 1942 to reinforce the AVG. Months later, all H81As and H87As remaining in China were absorbed into the U.S. Tenth Air Force as part of the newly formed China Air Task Force.

CW-21: Developed from the CW-19 advanced trainer, the CW-21 was conceived as a lightweight interceptor for the export market. The proto type, which featured landing gear that retracted into bulbous fairings on the bottom of the wings, flew in October 1938, and al – though powered by a similar version of the Cyclone engine, was 25 mph faster and possessed twice the climb rate climb of the contemporaneous fixed-gear H75.

The prototype was shipped to China for evaluation in early 1939, and in a move similar to the H75A-5 program, three pattern aircraft were ordered—with the plan that at least 30 more examples would be assembled from parts at the Loiwing plant. The three pattern aircraft reached Burma in 1940, only to be destroyed on the ferry flight to China. No further planes were produced.

In April 1940, the Netherlands ordered 24 improved CW-21Bs featuring landing gear that retracted inward, flush with the wing, and a Cyclone engine uprated to 1,000 hp. All the airplanes arrived for service in Java by the end of 1940, and at least 17 reportedly went into action against the Japanese in early 1942. Given their comparatively light armament (one .50- caliber and two .30-caliber guns) and lack of armor and self-sealing tanks, the Dutch CW-21Bs did not last long against superior Japanese firepower.


Model 322-61: In April 1940, the Anglo-French Purchasing Com – mittee placed an order for 667 Model 322-61 aircraft, the RAF portion of which were to be designated Lightning Is. Though similar to the Army P-38D, the export version called for 1,090- hp Allison V-1710-C15 engines lacking turbo-superchargers, with right-hand rotation of both propellers. At the time, based on combat experience thus far, French and British military officials believed the air war would be fought at medium altitudes (9,000-14,000 feet), and wanted to standardize the power plants between this airplane and the Curtiss 81A (P-40A/C).

Ironically, when two Lightning Is ultimately reached England in late 1941, the RAF rated the type as unsatisfactory for air combat, which by then was being fought at altitudes above 20,000 feet. The balance of the order was canceled.

The undelivered British aircraft were subsequently taken into the Army Air Forces as the P-322, and used only for training and testing. A follow-up British order for 524 Lightning IIs (with turbos and contrarotating props) was absorbed into the Army’s P-38F/G contract and thus was never delivered.


Model 48: In 1938 Vultee Aircraft conceived a design scheme in which a series of aircraft—fighter, advanced trainer and basic trainer—could utilize the same tooling for the wings, aft fuselage and tail group. Only one of these planes, the basic trainer design that went into service as the Army BT-13/15 and Navy SNV, achieved any measurable degree of success.

The Model 48 fighter, powered by a 1,200-hp Pratt & Whitney R-1830-S4, made its first flight in September 1939 under the factory name Vanguard. To reduce drag, the plane was initially built with a propeller shaft extension housed within an elongated cowl and pointed spinner, but persistent cooling problems led to the adoption of a more orthodox cowling arrangement in the Model 48X, which flew in early 1940. Further factory testing resulted in an enlargement of the tail surfaces and the addition of compound wing dihedral.

Sweden placed an order in February 1940 for 144 aircraft as the Model 48C, and an armed prototype, fitted with two .50-caliber guns in the nose and four .30-caliber guns in the wings, flew in September. But in December 1940, before any of the 48Cs could be delivered, the State Department barred exports to Sweden; the planes were instead allocated to the RAF training program in Canada as the Vanguard I. Then during the spring of 1941, in an interesting twist of events, Britain released production of the 48Cs back to the U.S. so they could be redirected to the Nationalist Chinese government under Lend-Lease.

According to Lend-Lease practice, the fighters were thereafter as – signed Army serial numbers and given the military designation P-66. In December 1941, about 15 of these planes were hastily requisitioned by the Army and sent to the 14th Pursuit Group at March Field to aid in the defense of California. The other 129 P-66s began reaching Karachi, India, during the summer of 1942 and were subsequently dispatched to China, where they replaced the obsolete Russian-made Polikarpov fighters being flown by the Nationalists.

The P-66’s combat record in China is spotty. Many were reportedly caught and destroyed on the ground during Japanese attacks, and a few are said to have been shot down by friendly fire—due to their resemblance to the Ki.43 Oscar.


In retrospect, it is amazing that although the United States was not fully prepared for war when Japan attacked Pearl Harbor in December 1941, American manufacturers produced 324,750 air- craft by the end of 1945, of which 99,950 were fighters. Insights gleaned from the combat experience of pre-1942 exports helped in the development of new or improved fighter designs—such as the Lockheed P-38, Republic P-47, North American P-51, Vought F4U and Grumman F6F—that would ultimately enable American forces to achieve air superiority in every theater of the war.


U.S. Navy veteran E.R. Johnson, who writes from Arkansas, is a past president of the Arkansas Aviation Historical Society and a major in the Arkansas Wing of the Civil Air Patrol. The author of American Attack Aircraft Since 1926, he also recommends for additional reading War Planes of the Second World War—Fighters, by William Green, and Ray Wagner’s American Combat Planes.

Originally published in the January 2010 issue of Aviation History. To subscribe, click here