Without a doubt, the Navy's strongest fighter, with it's distinctive inverted gull wing, was the Chance Vought F4U.

Designed as a powerful carrier fighter, the early F4us were restricted to land bases because they were difficult to fly aboard ship. But by the end of 1944 Navy and Marine squadrons were safely operating aboard carriers, and Corsairs remained active until after the Korean War. The Corsair was often used in ground-attack excelling in performance and payload.

It remained in production longer than any other U.S. fighter of the period and was credited with an 11:1 ratio of kills/losses against the Japanese. Development and pro-duction continued after the war for a total of 12,571 planes.

In production longer than any other US fighter of World War II, the Corsair had several claims to fame. It was credited with an 11 :1 ratio of kills to losses in action against Japanese aircraft and was the last pistonengined fighter in production for any of the US services. Its greatest attribute, though, was the excellence of its overall performance, making it certainly the finest carrier-based fighter of any used by the combatants in World War 11, and perhaps the best of any US fighters in that conflict.

Operational requirements for a new single-seat shipboard fighter were circulated to manufacturers by the US Navy early in 1938, and the Vought V-166B design, by a team led by Tex B. Beisel, was one of the proposals which resulted. To obtain the very high performance specified-matching that of contemporary land-based fighters-Beisel designed the smallest possible airframe around the most powerful available engine, the Pratt & Whitney XR-2800 Double Wasp. A characteristic feature of the Vought V-166B design was the inverted gull wing with the backward-retracting main legs of the landing gear located at the wing knuckles. This arrangement kept the legs short despite the height of the fuselage from the ground dictated by the large-diameter propeller. The wings folded upwards.

On June 30, 1938, the Vought company received a contract to build a single prototype of its Model V-1668, and this aircraft, designated XF4U-l, made its first flight on May 29, 1940. Powered by a 2,000 hp XR-2800-4 engine, the XF4U-I had a 030-in and a 050-in gun in the forward fuselage, one 050-in in each wing and compartments in the wings for 10 small bombs for use against bomber formations. Before the end of 1940 the XF4U-l had flown at 404 mph, faster than any US fighter then in the air, and on June 30, J 941, the Navy ordered production of 584 F4U-l s.

Deliveries of the F4U-ls began on October 3, 1942, four months after the first flight of the production Corsair, with the initial aircraft going to VF-12. Production aircraft had the R-2800-8 engine, two more guns in the wings with extra ammunition, self-sealing fuel tanks and armour protection: the cockpit was located 3 ft further aft to allow additional fuel to be carried in the fuselage. This last-mentioned change adversely affected the pilot's view, and carrier landing trials aboard the USS Sangamon in September 1942 cast doubts about the aircraft's suitability for the carrier role. Consequently, F4U-ls were issued primarily to land-based Marine units, starting with VMF-124. This unit took the Corsair into operation for the first time on February 13, 1943, at Bougainville, and seven more Marine units were flying the Corsair by August 1943. A month later, Navy Squadron VF-17, also land-based, became operational in New Georgia.

To speed production of the Corsair, contracts were placed with the Brewster and Goodyear companies for versions similar to the F4U-l, designated F3A-l and FG-l respectively and both incorporating a raised cockpit hood, which was introduced on the 689th F4U-l built by Vought; the FG-IA version had fixed wings. Some Corsairs were fitted with four 20 mm cannon in the wings in place of the machine guns and were designated F4U-IC, and others for use as fighter-bombers had fittings for a long-range tank under the fuselage and two 1,000 lb bombs or eight 5-in rockets under the wings, and the R-2800-8W engines with water injection: these were designated F4U-l D, FG-l D or F3A-l D according to source. Production of these initial Corsair versions totalled 4,120 F4U-ls by Vought (including 2,552 F4U-IC and F4U-ID), 735 F3A-ls and 3,808 FG-ls. Of these totals, 2,012 were supplied to Britain's Royal Navy under lend-lease, and another 370 went to the RNZAF. Operating from HMS Victorious, Corsair lis of No. 1834 Squadron of the Fleet Air Arm went into operation on April 3, 1944, during the attacks on the Tirpirz, this being the first time Corsairs had operated from aircraft carriers. The US Navy was still reluctant to commit its F4Us to carrier operation, but in April 1944 a further series of trials by VF-301 on the USS Gahier Bay showed that no serious problems remained and approval was finally given for the Navy squadrons to take their aircraft to sea.

Work on a night-fighter version of the Corsair had begun as early as January 1942 when a prototype XF4U-2 was ordered. Design features included Al radar on the starboard wingtip and an autopilot. The original XF4U-2 was not completed but the NAF at Philadelphia modified 12 F4U-ls as XF4U-2 night fighters with the same features, and six each were issued to VFN-75 at Munda, New Guinea, and VFN-10l which operated in turn from the USS Essex, Hornet and intrepid. Another service modification was designated F4U-IP, with cameras for reconnaissance duties.

Use of a turbosupercharged version of the Double Wasp engine was projected during 1941, and in March 1942 Vought received a contract for three XF4U-3s, including one F4U-l conversion. This work proceeded on low priority throughout the war, however, and the first XF4U-3 did not appear until 1946, with an R-2800-16 engine. Twenty-seven similar FG-3s were ordered from Goodyear, but only 13 were completed. The second major production version of the Corsair, therefore, was the F4U-4, the prototype of which first flew on April 19, 1944, with a 2,100 hp R-2800-18W engine. The additional power from this engine increased the maximum speed of the F4U-4 to 446 mph, and other small changes were made to improve the operational characteristics of this version. Production variants, in addition to the basic F4U-4, included the cannon-armed F4U-4C; the F4U-4N night fighter with APS-5 radar, and the F4U-4E with APS-4. Goodyear built 200 of the equivalent FG-4, while Vought production of these variants totalled 2,356. Some cameraequipped aircraft in this series were designated F4U-4P.

Despite large-scale cancellation of contracts following V-J Day, production of the F4U-4 by Vought continued to 1947.

In 1946 Vought produced a new Corsair variant, the XF4U-5, by fitting a 2,300 hp two-stage R-2800-32W engine and four 20 mm wing guns in an F4U-4. To meet immediate requirements for a carrier-based fighterbomber and night fighter, the Navy purchased 223 F4U-5s, 315 F4U-5Ns and 30 F4U-5Ps during 1947 and 1948. Then came the low altitude XF4U-6 with a single-stage R-2800-83W engine, additional armour protection and increased underwing load-carrying ability. Redesignated AU-I, this type went into production for use by Marine squadrons operating in Korea, 110 being built.

The Corsair line ended with 90 F4U-7s, similar to the AU-I but with the R-2800-18W engine more than a decade of F4U production.

The V-166B, as the company had designated the preliminary design, had been, in essence, the smallest practical airframe that could be built around the largest and most powerful air-cooled radial engine then under development. Promising almost twice the power of any then used by the US Navy, this engine had demanded a fuselage ground angle that would have made the Fieseler Storch look squat in order to achieve the requisite 18 in ground clearance for the propeller. Rex Biesel and his team had come up with a novel and extremely neat solution to this problem; the inverted gull wing that was to be the distinctive hallmark of the Chance Vought fighter thenceforth. Not only had this resolved the propeller clearance problem, it had simultaneously resolved the difficulty of keeping the undercarriage to a manageable length and, as a bonus, had minimised drag by offering the optimum right angle junction between wing and fuselage, eliminating the wetted area of a large fairing that would otherwise have been needed.

Determined to effect a major advance in the shipboard fighter-state-of-art, Chance Vought's Engineering Department had not been content with mating an innovatory airframe design with the world's first 2,000 hp engine and the largest propeller ever used by a fighter; it had elected to compound these advances with an entirely new spot-welding skin attachment technique to achieve a smoother surface finish than had been possible previously. The sum total of all these innovations was a very radical aeroplane, but a carrier-based fighter. .

On 11 June 1938, a contract had been issued for a prototype designated XF4U- 1; the mock-up had been inspected by the US Navy's Bureau of Aeronautics between 8-11 February 1939, and on 29 May 1940, after several hours of taxying trials, the innovatory fighter had been flown by Chance Voughts Chief of Flight Test, Lyman A Bullard Jnr. The combination of an entirely new and untried power plant with an equally new airframe embodying a number of untried features had certainly tempted providence, and an incident during the brief initial flight test was a portent of problems ahead.

Powered by an XR-2800-4 Double Wasp engine and weighing in at 9,357 lb (4 245 kg), the XF4U-1 had just attained a 200 knot cruise when the elevator tabs set up a high-frequency oscillation and then tore loose. Bullard had landed the prototype safely despite the exceedingly heavy control forces, but during the fifth flight, with Boone T Guyton, who had taken over as project pilot, at the controls, the XF4U-1 had forced-landed on a golf course, flipping over, losing a wing and writing off the vertical tail surfaces and propeller. Undeterred by this setback, Chance Vought had rebuilt the aircraft, which, on 1 October 1940, had demonstrated a ground speed of 404 mph in a flight between Stratford and Hartford.

Despite the limited flight testing that had been conducted with the XF4U-l and the handling problems that had revealed themselves from the outset and remained unresolved, time had by now begun to be considered of the essence, and on 28 November 1940, the US Navy had requested proposals for productionising the design; a process that was to result in changes that were to enhance some characteristics and exacerbate some problems, and in the latter category were the Corsair's manifest shortcomings as a shipboard aeroplane.Engineering for production release had begun on 30 December, and flight test tempo had been accelerated in order that the prototype could be demonstrated to the US Navy at Anacostia at the earliest possible date. It now became apparent that the standard US Navy specification requirements would have to be rewritten to accommodate the Corsair and aircraft of its generation. One requirement was a vertical dive maintained for 10,000 ft and during attempts to demonstrate this capability, a speed of 515 mph had been clocked, but after landing it had been discovered that the fabric was stripping from the rudder and elevators, the starboard flotation hatch had partly torn away and an access door had buckled. On 28 January 1941, in a final test dive from 20,000 ft a speed of nearly 500 mph had been reached when negative g in the hydraulic system allowed the propeller to overspeed, leading to a complete engine failure and a deadstick landing. This incident had rendered it obvious that the US Navy's dive requirement was no longer feasible; if the dive was initiated at 20,000 ft unacceptably high g forces were encountered during recovery, while a dive begun from a higher altitude took the aircraft into compressibility effects. Another requirement that had to be waived was a 10-turn spin when it was ascertained that the control loads demanded for recovery exceeded human limits - during this test the pilot saved the XF4U-1 only by recourse to an anti-spin 'chute.

No fewer than 96 changes had to be made to the aileron control system alone during 110 successive flights early in 1941, but the XF4U-1 had still been decidedly immature on 21 April, when the Navy demonstration hadbeen completed, and US Navy pilots that flew the prototype, while evincing enthusiasm over the XF4U-1's speed and climb rate, had mostly voiced reservations over its low speed characteristics, the lack of visibility over the nose during the landing approach, the landing performance itself and the unnerving predilection of the aircraft to drop a wing just before touching down. Had it not been for the extreme urgency that was by this time attached to the service introduction of a really high performance fighter, there can be little doubt that a production go-ahead would have been delayed pending a solution to these problems; as it was, on 30 June 1941, the first contract was confirmed for 584 F4U-ls, and the initial production aircraft was to fly five days short of a year later.

The production F4U-1 had embodied many changes and improvements, resulting both from prototype trials and changes in US Navy requirements; regrettably, none of these improvements had extended to the less glamorous attributes so vital to any effective deck-landing aircraft. In the light of European combat experience, it had become obvious that the gun armament originally specified a single 05-in (12,7mm) gun in each wing plus a similar calibre weapon paired with a 03-in (7,62-mm) gun in the nose - was no longer adequate. The wing structure had accordingly been redesigned to accommodate three 05-in (12,7-mni) Brownings in each outer panel - although a few early production examples had mounted an interim armament of four wing guns - which had the incidental advantage of enabling the heavy synchronisation gear to be dispensed with.

Unfortunately, the integral wing leading edge tanks had been displaced by the revised armament and fuel capacity had had to be made up by installing a 237 US gal (897 I) self-sealing tank in the centre fuselage. This tank had of necessity to be as close as possible to the aircraft cc in order to minimise trim changes as fuel was consumed, but it so happened that this situation was already occupied by the cockpit! The design team had had no recourse but to relocate the cockpit 3 ft (0,92 m) farther aft. The forward view for take-off and landing offered by the prototype had come in for much criticism and that from the production model could only be described as appalling - some palliative measures were subsequently to be taken, but the fault was fundamental to the design and was to remain a major point of criticism throughout the Corsair's career. An early change had been the reintroduction of wing leading edge tanks, unprotected and situated outboard of the gun bays, each having a capacity of 62 US gal (235 I), these and the full armament taking the gross weight up to 12,694 lb (5 758 kg) from the 9,357 lb (4 244 kg) of the XF4U-l and wing loading rising dramatically from 298 to 404 lb/sq ft (145,5 to 197,2 kg/m~). Understandably, the higher weights had an adverse effect on manoeuvrability, but a reduction in aileron span had served to improve roll rate.

The incipient bounce during landing resulting from overly stiff oleo action had been carried over from the prototype and was found to be most disconcerting by pilots fresh to the F4U-l, but not so disconcerting as the virtually unheralded torque stall that occured all too frequently in landing condition. Compounded by a serious directional instability immediately after touch-down and coupled with the frightful visibility from the cockpit during the landing approach, it was small wonder that the US Navy concluded that the average carrier pilot was unlikely to possess the necessary skill to master these unpleasant idiosyncrasies.

Indeed, carrier qualification trials aboard the USS Sangamon In Chesapeake Bay on 25 September 1942 had been something of a fiasco, the pilot, Lt-Cdr S Porter, having to contend with windscreen contamination by oil from the hydraulically-operated cowl flaps and the valve push-rod mechanism, as well as the handling and visibility problems. Later trials were to be punctuated by a series of nose-overs, barrier-bounces and bent propellers.

A consequence of this lack of shipboard finesse on the part of the Corsair was the assignment of delivery priority in the fighter to the Marine Corps for shore-based operation, and the first Corsair-equipped combat squadron thus became VMF-124 formed on 7 September 1942, declared combat ready on 28 December and despatched post-haste to reinforce beleaguered US forces on the island of Guadalcanal. From here the first combat mission was flown on 11 February 1943 - escorting a PB2Y to Vella Lavella to pick up a downed pilot. On the following day, Corsairs flew escort to PB4Ys of VP-Si on a strike against enemy shipping in the Kahili area of Bougainville, being blooded during a similar mission two days later when intercepted by a large formation of Zero-Sen fighters. The Japanese force shot down a top cover of four P-38 Lightnings, two accompanying P-40 Warhawks, two of the PB4Ys and two of VMF-124's Corsairs, all for the loss of four Zero-Sens. This inauspicious opening of the Corsair's combat career was not, happily, an unfavourable augury for the future; with experience, the Marine Corps pilots learned to avoid mixing it with their more nimble Japanese opponents and take advantage of the Corsair's immensely superior level and dive speeds.

Within six months, all Marine Corps fighter squadrons in the South Pacific had been Corsair-equipped. Meanwhile, the US Navy had formed its first Corsair squadron, VF-12, which received its initial F4U-1 on 3 October 1942, but this served primarily as trials unit and although it eventually became deck-qualified in April 1943 aboard the USS Saratoga, it was to turn its aircraft over to the Marines and relinquish the honour of being the first US Navy squadron to take the Corsair on operations to VF- 17 formed on 19 April. This was the first squadron to receive F4U-ls fitted with full plexiglas canopies - retrospectively referred to as F4U- 1 As- which supplanted the confining 'bird cage" canopy, with its numerous metal reinforcing~strips against which a pilot's head could jar unpleasantly in turbulence owing to restricted width towards its apex. In the event, another year was to elapse before the US Navy was to clear the Corsair for shipboard operation, and VF-17 sailed aboard the USS Bu,~ker Hill to New Georgia, which was reached in September, subsequently achieving conspicuous success as a shore-based unit.

Production tempo of the F4U-1 built up extremely rapidly. Chance Vought was to roll out its 1,000th aircraft on 22 August 1943 - and the parent company had been joined in the production programme by Brewster and Goodyear. In consequence, it was possible to offer supplies of this fighter to the Royal Navy which gladly accepted 95 early series F4U- is as Corsair Mk Is, these being followed by a total of 510 "high-cabin" F4U-1 s as Corsair ITs. The Royal Navy set up a procedure for forming and working-up squadrons on the type at either Quonset or Brunswick in North America, and then shipping the complete squadrons with their pilots to the UK aboard escort carriers. The first unit so formed was No 1830 Squadron, which, commissioned at Quonset on 1 June 1943, was subsequently to be attached to HMS Illustrious, Three more squadrons, Nos 1831, 1833 and 1834, formed in July, followed by two in August (Nos 1835 and 1836) and one each in September and October (Nos 1837 and 1838 respectively). A further eight squadrons were destined to form ata steady rate through 1944, and a final two in 194540 give a total of 19 FAA Corsair squadrons.

Oddly enough, the Royal Navy was not quite so fastidious as the US Navy regarding, deck landing characteristics and cleared the Corsair for shipboard operation some nine months before its American counterpart. The obstacles to the Corsair's shipboard use were admittedly not insurmountable, but I can Only surmise that the apparently ready acceptance by their LOrdships of the Admiralty of the Chance Vought fighter for carrier operation must have been solely due to the exigencies of the times, for the landing behaviour of the Corsair really was bad, a fact to which I was able to attest after the briefest acquaintance with the aircraft.

I had joined the Royal Aircraft Establishment at Fainborough in January 1944, and one of the first tasks to which I was assigned was that of checking out the diving characteristics of the Corsair with undercarriage both retracted and extended. The aircraft with which I was to perform the tests was an early Lend-Lease Corsair Mk I (JTilS) and our encounter was certainly not a case of love at first sight. On the contrary, during my acquaintance with this impressively large and aesthetically unappealing fighter, which was to spread over several years, I was never to achieve any Sort of rapport.

The Corsair's inordinately large proboscis was, I suppose, its most outstanding feature - in the USA in later years I was to hear this fighter referred to by the sobriquet 'Old Hog Nose' - and coupled with its fairly acute and most distinctive ground stance, it imparted an impression of rugged strength rather than aerodynamic refinement. The cockpit was, I discovered, inordinately spacious, and I was reminded of the suggestion of an RAF pilot on first seeing the P-47 Thunderbolt that its occupant might take evasive action by releasing his harness and dodging about the cockpit! It appeared to have been tailored for an extremely tall pilot- I subsequently learned that the principal Corsair project pilot was 6 ft 4 in (1,93 m) in height - and one of more modest stature such as myself inevitably experienced some discomfort keeping one's feet on the rudder with the seat adjusted to a height from which what little forward view existed could be gained. The layout of the cockpit was poor and on the ground the only reasonable view was upward!

The immense Pratt & Whitney R-2800-8 Double Wasp was turned over by hand four or five times, the fuel booster pump was switched ON, the priming switch was flicked several times, the ignition switch activated and the starter cartridge fired. The Double Wasp, which has take-off rating of 2,000 blip at 2,700 rpm, usually burst into life immediately and with the firing switch depressed and the mixture control moved slowly to AUTO RICH was soon purring with all the smoothness and reliability so characteristic of this family of engines. The Double Wasp was opened up to 1,000 rpm to warm up, the usual temperature, pressure and magneto checks performed, the flaps lowered and raised, and the revs increased to 1,400, the operahon of the two-speed supercharger being checked by moving the control from NEUTRAL to LOW and, after a pause of a few seconds to ~nGH. With the propeller control fully down, the throttle was opened and take-off boost and static rpm checked, the stick being held hard back to contain a strong tendency for the tail to lift.

The feeling of not being at one with the aircraft was emphasised during taxying, when the totally inadequate forward view necessitated swinging that great nose from side to side, but to do this the tailwheel had to be unlocked and the Corsair was then very unstable directionally, necessitating constant use of the brakes with the danger of nosing over in the event of too harsh application. There could be no doubt that some practice was necessary to achieve satisfactory ground handling. For take-off, the trimmers had to be set six deg right rudder with six deg down on the right aileron and one deg nose up on the elevator. With mixture in AUTO RiCH, fuel cock on RESERVE, booster pump ON, intercooler shutters closed, the supercharger lever in NEUTRAL, and the tailwheel locked, the engine was opened up to 54 in boost and 2,700 rpm. If trimming was correct, the Corsair demonstrated no tendency to swing and unstick was rapid. With the application of 30 deg of flap, such as would be employed for a carrier take-off, and about two-thirds normal fuel at a take-off weight of about 11,150 lb (5058 kg), the Corsair would take-off within 185 yards (169 m) without wind and about 120 yards (110 m) into a 15 knot (28 km/h) wind.

The speed for maximum climb rate was 125 knots (232 km/h) from sea level up to 21,000 ft (6400 m) and the intercooler shutters had to be opened fully, but the cowl gills were only half opened otherwise there was some buffet. Climb was certainly impressive, with that immense 13 ft 4 in (4,06 m) diameter Hamilton Standard propeller pulling the aircraft up like a high-speed lift, 10,000 ft (3 050 m) being passed in 46 minutes and 20,000 ft (6095 m) in 96 minutes. Above 21,000 ft (6400 m) climb speed was reduced three knots (5,5 km/h) per 2,000 ft (610 in), but the two-stage two-speed supercharger ensured good climbing capability well above 30,000 ft (9 145 in). Once in level flight, the Corsair could be trimmed to a very stable hands-off flying condition. Its stability was positive at all times in the cruise and at high speeds. It imparted a feeling of solidity; to compare it with, say, the Seafire, was like comparing a shire horse with a poio pony, and this feeling of solidness was combined with an impression of immense power.

The harmony of control was poor, the elevators being heavy but the ailerons being moderately light, enabling the Corsair to be rolled to its maximum rate even at fairly high diving speeds. This effective high-speed aileron control must certainly have been valuable in the South Pacific as the Corsair's principal opponent in that theatre, the Zero-Sen, had poor aileron control at high speeds at which it rolled sluggishly. I had heard that Marine Corps pilots used this high-speed roll capability as a standard evasive tactic, diving the Corsair and then rolling to port or starboard before recovery to shake off any pursuing enemy fighter. Acceleration was quite dramatic and a clean aeroplane with about two-thirds fuel in main tank only and 200 rounds for each of its six 'fifties could reach a maximum speed of 342 knots (634 km/h) at the Corsair's critical altitude of 24.000 ft (7,315 m) on normal maximum power, which, at that altitude was 1,550 hp at 2,550 rpm. At combat power of 1,650 hp at 2,700 rpm (limited to five minutes), maximum speed was 343 knots (636 km/h).

The flaps could be lowered 20 deg to assist manoeuvring at speeds tip to 200 knots (370 km/h) and recommended speeds for aerobatics included 350-360 knots (650-670 km/h) for an upward roll, 330 knots (610 km/h) for a climbing roll, 300 knots (556 km/h) for a roll off the top of a loop and 260-280 knots (480-520 km/h) for a loop, but it was not recommended that the Corsair be held inverted for more than three seconds. The stalling characteristics were very poor, with little warning other than that afforded by the stall warning light on the instrument panel which was operated by the breakdown of airflow over the centre section. At the stall, the right wing dropped sharply and an incipient spin developed if the control column was not moved smartly forward. If the Corsair stalled in a steep turn it would normally flick out, but recovery was rapid if control column pressure was relaxed quickly. At about 11,500 lb (5 216 kg) with engine off and all up, the Corsair would stall at 90 knots (167 km/h), and with flaps and undercarriage down at 76 knots (141 km/h), the warning light coming on at 80 knots (148 km/h).

My initial task with the Corsair was, as previously mentioned, to check out the diving characteristics. We had had reports of Marine Corps Corsairs losing the fabric from their elevators, and I could well imagine this because, due to the combination of relative aerodynamic cleanliness, high power and weight, the Corsair accelerated very rapidly in a dive with the aircraft clean, and relatively inexperienced pilots working off some exuberance by diving steeply from high altitudes without close regard to structural limitations could easily have found themselves in trouble. In clean condition, acceleration was rapid to 400 knots (740 km/h) below 10,000 ft (3 050 in), but with the undercarriage extended to serve as a dive brake it took about the same height loss to reach 350 knots (650 km/h). Any attempt to exceed these limiting speeds produced pronounced elevator buffeting. Lowering the undercarriage as a brake was done by using the dive brake control, the tailwheel remaining retracted otherwise damage to the tailwheel doors would have resulted. Lowering the mainwheels produced a strong nose down trim change, and the elevators heavied up in diving and a pull-out called for plenty of height in consequence. Buffeting of the elevators could occur during the recovery, dictating easing off the pull-out and reducingg. Prior to the dive, the supercharger was set to NEUTRAL, the mixture in AUTO RICH, the throttle was set one-third open, the cowling flaps, oil and intercooler shutters were shut and the rudder was trimmed six deg left with the elevator set one-and-a-half deg nose down. As no automatic boost control was fitted, care had to be exercised in avoiding overboosting.

I was well aware that the US Navy had found the Corsair's deck-landing characteristics so disappointing in trials that it had been assigned for shore duties while an attempt was being made to iron out the problems, and although the FAA rvas deck-landing the aircraft, I knew that, by consensus, it had been pronounced a brute and assumed that shipboard operations with the Corsair were something of a case of needs must when the devil drives. The fact that experienced US Navy pilots could deck-land the Corsair had been demonstrated a couple of months earlier, in November 1943, when VF-17, providing high cover for the carriers Esse.r and Bunker Hill, had run short of fuel after decimating an attacking torpedo-bomber force and had landed safely aboard the carriers. All in all, I wasmost anxious to discover for myself if the Corsair was the deck-4anding dog that it was reputed to be. It was!

In the deck-landing configuration with approach power, the Corsair could demonstrate a very nasty incipient torque stall with dangerously little warning, the starboard wing usually dropping sharply. With the large flaps fully extended the descent rate was rapid, and a simulated deck-landing at 80 knots (148 km/h) gave very poor view and sluggish aileron and elevator control. A curved approach was very necessary if the pilot was to have any chance of seeing the carrier, let alone the batsman! When the throttle was cut, the nose dropped so that the aircraft bounced on its mainwheels, and once the tailwheel made contact, the aircraft proved very unstable directionally, despite the tailwheel lock, swinging either to port or starboard, and this swing had to be checked immediately with the brakes. On one approach, I tried a baulked landing and discovered that the sudden opening of the throttle at 80 knots (148 km/h) also produced the previously-mentioned torque stall, but this time the port wing dropped. I needed no more convincing of the wisdom of the US Navy in withholding the Corsair from shipboard operation! Oh yes, the Corsair could be landed on a deck without undue difficulty by an experienced pilot in ideal conditions, but with pilots of average capability, really pitching decks and marginal weather conditions, attrition simply had to be of serious proportions.

Changes were being continuously applied to the Corsair on the assembly lines, among the first that could be seen externally being the introduction of the higher, single-piece canopy as distinct from the original 'bird cage" hood. This vastly improved canopy was accompanied by the elimination of the cut-outs for aft-vision behind the headrest and the raising of the seat by about seven inches (18 cm), and the combination did marginally improve forward view. Although these modifications did not apparently warrant any change in US Navy designation at that time, their application produced the designation Corsair Mk II in the Royal Navy. Oddly enough, somewhat more significant changes in later batches of aircraft did not result in mark number changes, despite the addition of at least a suffix letter to the US Navy designation. For example: all Royal Navy Corsairs could carry the 142 Imp gal (646 I) centreline tank and with the 361st Corsair II this could be augmented by a 137 Imp gal (623 I) tank under the starboard wing only (in lieu ofabomb), but with the 441st aircraft, the unprotected integral wing leading edge tanks were discarded and a 137 Imp gal (623 I) tank could be carried under each wing. Thus, the last 150 Corsair Mk us were basically F4U-lDs, but unlike their US Navy counterparts they carried no distinctive designation. To make nonsense of the whole mark number system, identical aircraft produced by Brewster and Goodyear were assigned the designations Corsair Mk Ill and Mk IV respectively.

Other changes included the provision of longer-stroke oleos to take some of the bounce out of the landing and, peculiar to FAA Corsairs, the clipping of eight inches (20 cm) from each wingtip to reduce the folded height for stowage in the cramped below-decks hangars of escort carriers, this latter, somewhat rudimentary modification incidentally improving lateral control. The Corsair II was, of course, a heavy aircraft for escort carrier work and reports soon began to reach us at Farnborough of a disturbing accident rate and of excessive wear on arrester wires. It was not just the sheer bulk of the Corsair, but the fact that the pilots were throwing it onto the deck too fast, the lack of aerodynamic stall warning making for a tendency to approach at too high a speed. The real trouble, of course, lay in the fact that the modifications characterised by the Corsair Mk II had made little positive improvement in its deck-landing qualities except that there was less tendency to bounce over the wires after the mainwheels first deck contact.

In general, the arrester wires were only surviving about four landings before having to be replaced. Inevitably, a Corsair II came to the RAE to be tried out in the arrester gear fitted on the short runway. These trials included a series with the aircraft carrying two 1,000-lb (453,6-kg) bombs which gave an all-up weight of 14,000 lb (6 350 kg). At 24g and a deliberate 15 ft (4,57 m) off centre contact, the arrester wire was being destroyed at each landing. These were extreme conditions, of course, but arrester wire wear and tear remained excessive and the number of accidents escalated in proportion to the number of Corsairs deployed. Later, with a Corsair IV, we were to perform a number of tests with arrester wires fitted with heavy 1-in (2,54-cm) centre spans (ie, the piece contacted by the arrester hook). The very first landing that I made into this heavy span whipped the tailwheel olco clean off the aircraft, so this was obviously no solution to the problem. Indeed, no solution was to be found by the time the problem disappeared with the disembarkation of the last FAA Corsair squadron in August 1946.

In the meantime, the FAA had taken its Corsairs into action, their operational debut in the European theatre having taken place on 3 April 1944 when No 1834 Squadron aboard HMS Victo,-ious flew fighter cover in concert with Hellcat, Wildcat and Seafire squadrons for Barracudas attacking Tirpitz lying in Kaafiord, northern Norway. In further attacks on Tirpitz in July and August, the Corsairs of Nos 1841 and 1842 squadrons from HMS Formidable were also in action. April also witnessed the operational debut of the FAA Corsair in the Pacific area, when, on the 19th of that month. Nos 1830 and 1833 squadrons escorted Barracudas attacking Sabang.

Despite the fact that the FAA was flying the Corsair from carriers in several theatres, the original design aim of shipboard operation was not achieved by the US services until 28 December 1944, and then it was a Marine Corps squadron that first flew combat sorties from a carrier. The F4U-i had been finally cleared for shipboard service with the US Navy in the previous April, after VF-103 had completed 113 uneventful landings in succession aboard the USS Gambier Bay, its Corsairs having been fitted with the longer-stroke oleos mentioned earlier. The Marine Corps squadron that initiated shipboard Corsair operations. VMF-124, flew from the USS Essex, the first USMC unit to be deployed aboard a fast carrier specifically to meet the increasing threat of Japanese Kamikaze attacks, and by the end of the Okinawa campaign, nearly every US Navy carrier was equipped with Corsairs.

There can be no doubt that the Corsair was one of the fastest naval aircraft of WW 11 and few of its pilots criticised it from the performance standpoint. It had a good range, adequate firepower, an extremely reliable engine and it could absorb a lot of punishment. However, in my view it left much to be desired as a fighter from the viewpoint of manoeuvrability and this same shortcoming was apparent in the dive bombing r6le in which it saw widespread use. Finally, it had a very dreary track record as a deck-landing aircraft; many were the pilots that lauded its high-speed performance but decried its lack of affinity with a carrier deck.

It was with this unenthusiastic regard for the Corsair lingering in my memory that I found myself faced with the same aircraft in slightly different guise a number of years later, in 1951, when I was posted on attachment as a test pilot to the US Naval Air Test Center at Patuxent River. On the inventory of Flight Test was the AU-i, a dedicated low-level ground support derivative of the original Corsair evolved specifically for Marine Corps use in the Korean War. Fitted with a single-stage two-speed R-2800-83WA version of the Double Wasp with a sea level combat rating of 2,800 hp, a lot more armour in the cockpit and engine sections and a built-in armament of four 20-mm Mk 6 cannon with 231 rpg, the AU-i had taken on a lot of weight by comparison with the WW II Corsairs with which I was familiar. With a pair of 150 US gal (5681) drop tanks, a 1,000-lb (453,6-kg) bomb and six 500-lb (226,8-kg) bombs, the AU-1 weighed in at no less than 19,398 lb (8,799 kg).

Understandably, performance had suffered, the maximum speed attainable when laden in the fashion described being 207 knots (384 km/h) at 8,800 ft (2,680 in), but combat range was 715 nm (1 325 kin) cruising at 165 knots (306 km/h) at 15,000 ft (4570 in). It wasthe handlingof the AU-i that served to heighten my distaste for the Corsair, however, for if its ancestor had proffered some unendearing characteristics, they had been multiplied in the descendant. The AU-I had developed some highly undesirable directional stability and control characteristics, such as requiring almost full right rudder on a deck-landing approach, thus rendering baulked landing the most hazardous of operations. It also displayed a directional oscillation in diving with external stores, thus setting up wing rocking and seriously affecting the aiming accuracy. The manoeuvring forces were high - 13 lb (5,9 kg) per g at 200 knots 370 km/h) at mid CG - and aileron overbalance occurred above 265 knots (491 km/h). These shortcomings were compounded by the old fault of inadequate stall warning and, if anything, the forward view appeared to have worsened. Whether any of these defects were eventually remedied I know not, but unless they were, I sympathise with any pilots that operated the AU-I from a carrier.

The Corsair is a difficult aircraft to assess for, as I have already said, it was operationally successful in spite of itself. Designed essentially as a shipboard air superiority weapon, it was only a qualified success when operated from carriers, but when operated from shore bases by the US Marine Corps it built up an excellent record. The Corsair has, of course, a sentimental place in the hearts of FAA pilots because it gave us our only fighter VC, Lt Robert Hampton Gray, RCanNVR, who, leading a force of Corsairs attacking Japanese naval vessels in Onawaga Bay on 9 August 1945, attacked a destroyer despite his aircraft having suffered damage from anti-aircraft fire, and scored a direct hit before his aircraft was struck once more and burst into flames.