Lockheed was invited, along with Boeing, Consolidated, Curtiss, Douglas, and Vultee to take part in USAAC's Design Competition X-608 for a twin-engine, high-altitude interceptor. Specification called for a top speed of at least 360 mph at 20,000' and 290 mph at sea level, an endurance at full throttle of one hour at 20,000', and the ability to take-off and land over a 50' obstacle within 2,200'.

    The Lockheed design staff was headed by Hall Hibbard. Working with Hibbard was the soon-to-be famous Clarence "Kelly" Johnson. After studying many different designs, the pair settled on a twin-boom design with the pilot in a central nacelle. The booms would each house a 12-cylinder 1150hp Allison V-1710C with an exhaust-driven turbo-supercharger&#!51;the liquid-cooled Allison had just completed a 150-hour type-approval test at 1000 hp.

    The nacelle contained a forward-firing 20mm cannon and four .50 machine guns, armament quite heavy for its time. One advantage of the twin-boom design was a nacelle, unhampered by synchronizing mechanisms, that could deliver parallel streams of fire up to 3000'. It was estimated that the twin-tail assembly would increase the effective aspect ratio of the tailplane by the endplate effect, providing stability over a large c/g range. Trailing-edge Fowler flaps were inboard of the ailerons and between the booms.

As their Model 22-64-01, Lockheed promised a top speed of over 400 mph. Although the AAC was somewhat skeptical about so radical a design, Model 22 won Design Competition X-608 and, on June 23, 1937, Lockheed received a contract for one unarmed XP-38 prototype [37-457]. Construction began in July 1938 and proceeded rather rapidly despite the radical features that it embodied. Few problems were presented by the installation of the Allison V-1710-11/15 (C9)s, each with a General Electric B-1 turbo-supercharger. To combat torque, the propellers rotated inwardly in opposite directions—a special version of the Allison had a left-hand rotating propeller shaft.

    The XP-38 was completed in Dec 1938 and, on the last day of the year was stripped down, covered with canvas, and loaded onto three trucks. In great secrecy the convoy was escorted by police to March Field, where AAC Project Officer Lt Benjamin Kelsey was to do flight testing. However, on the very first ground run, wheel brakes failed and the plane ended up in a ditch. Kelsey finally got airborne on Jan 27, 1939. Early test flights disclosed some problems with the brakes and vibrations of the flaps, requiring modifications be made to the prototype. Maximum speed was 413 mph at 20,000', an altitude reached in 6.5 minutes—service ceiling was 38,000'. Empty weight was 11,507#, gross weight 13,964#, max take-off weight 15,416#.

    Reaction was highly favorable in spite of the problems encountered on its first few flights, and it was decided to try for a record transcontinental flight before delivering the ship to Wright Field OH. At daybreak on Feb 11, 1939, Kelsey departed for Mitchel Field NY, with refuelling stops at Amarillo and Wright Field, but as Kelsey was coming in for a landing at Mitchell, the XP-38 lost powerl Field and crashed on a golf course just short of the runway. Kelsey was unhurt, but the XP-38 was a total loss.

Even before YP-38s had been built, on Sep 20, 1939 the Army placed an order for 66 P-38 fighters. 29 of those were delivered as P-38 (company Model 222-62-02) [40-744/761, 763/773] with the same powerplants as YP-38, but armament was changed to one 37mm cannon and four .50 machine guns. Armor plate and bulletproof glass was added for pilot protection, and fluorescent instrument lighting for night flying.

    P-38 [40-744] was later modified to study the effects on flight crews of asymmetric cockpit locations. Superchargers were removed, and the space in the port boom housed a cockpit for a flight surgeon.

The first YP-38 flew on Sep 16, 1940 with Marshall Headle at the controls. In Mar 1940 the Army received its first one for service trials, but production lagged seriously behind schedule and all 13 were not completed until June 1941. Maximum speed was 405 mph at 10,000', climb to 20,000' took only six minutes. Normal range was 650 miles, and weights were 11,171# empty, 13,500# gross, and 14,348# maximum take-off.

    During trials the YP-38s suffered severe tail buffeting in high-speed dives, making pull-out difficult. On Nov 4, 1941 the tail booms of [39-689] came off during a dive, and test pilot Ralph Virden was killed. It was initially falsely diagnosed as elevator flutter, and a set of external mass balances were added above and below the elevator. The problem was later solved by adding large wing-root fillets at the points where the wings joined the fuselage, which had to be done very carefully—a loose fit would severely impair flight characteristics.

At one time, early on, the AAF considered naming the P-38 "Atlanta." However, P-38D and subsequent versions were officially christened "Lightning."

    Changes were taking place at such a rapid rate that even the those introduced on the P-38D did not really make it combat-ready. In 1942 the P-38 and P-38D were redesignated RP-38 and RP-38D, the prefix meaning "Restricted to non-combat roles"—they were to be used only as combat trainers.

    P-38s, RP-38s, and some P-38Ds from Selfridge Field were participants in the Sept 1941 Army-Navy joint maneuvers in Louisiana, but few of them actually had guns. By the time of Pearl Harbor, there were only 69 P-38 and P-38D fighters on strength.

    POP: 36 P-38D [40-774/809].