How America’s military aviation arm rebuilt itself from surplus stock to cutting-edge aircraft in less than two decades
At the end of World War One in November 1918, the Air Service held roughly 7,900 aircraft, fewer than 1,000 of which had seen action in France. Most were sold as surplus, factory contracts were cancelled, and the service was left to rebuild from scratch.
Reorganization followed. In June 1920, the Air Service became a recognized combat arm of the US Army, and in July 1926, it was redesignated the US Army Air Corps, with an authorized fleet of 1,800 aircraft.
Building a bomber force
Impressed by wartime European multi-engine bombers, the Air Service arranged to have a modified Italian three-engine Caproni and a British twin-engine Handley Page assembled domestically. Only five Caproni units entered post-war service. The Glenn L. Martin Company was also contracted in late 1918 to produce the MB-1 twin-engine biplane bomber, but none reached France in time, and production stopped after ten units.
A follow-on order in 1921 produced twenty units of the updated MB-2, later known as the NBS-1 (Night Bomber, Short Range)—a slow aircraft considered vulnerable to enemy fighters in daylight. A subsequent War Department order for 110 additional units was competitively bid, and Martin was undercut by rivals.
Designation system
From 1920, US military aircraft were assigned alphanumeric designations: letter prefixes identified type and size, trailing numbers indicated acquisition sequence, and suffix letters denoted variants. The prefix “X” meant experimental; “Y1” denoted service test aircraft.
Pursuit planes
Post-war fighter development — officially “pursuit planes” until the USAF was established in 1947 — began with two designs commissioned in the final year of the war: the Orenco Model D and the Thomas-Morse MB-3. Both were competitively retendered, with Curtiss winning the rights to the Model D and Boeing to the MB-3.
Through the 1920s, biplane fighters evolved incrementally. Boeing’s P-12 series, ordered from 1928, eventually totaled 365 units across multiple variants. Despite their numbers, these aircraft still relied on wooden-and-fabric wings, open cockpits, and fixed landing gear—with only metal fuselages representing any advance on WWI-era design.
The shift to all-metal monoplane pursuit aircraft came with the Boeing P-26, 136 units of which were ordered in early 1933. Yet even the P-26 retained fixed landing gear, open cockpits, and externally braced wings. Its higher landing speed did introduce landing flaps—a British invention—as a concession to monoplane aerodynamics.
By the late 1930s, the P-26 was obsolete. It was replaced by the Curtiss P-36A and the Seversky P-35, the first Air Corps fighters with retractable landing gear and enclosed cockpits, though both were already outclassed when they entered service. The P-36A saw combat at Pearl Harbor before being withdrawn; the P-35 had already been pulled from the front line.
The bomber generation gap
The Keystone series of twin-engine biplane bombers, which served through the late 1920s and into the early 1930s, were essentially World War I designs in wood and fabric—open cockpits, fixed gear, and bombs carried under the wings.
The first genuinely modern bomber tested by the Air Corps was the Boeing Y1B-9 in 1931: an all-metal monoplane with retractable undercarriage, though it still lacked an internal bomb bay and retained an open cockpit. Only seven were built.
The Martin B-10, entering operational service in 1935, marked a decisive leap forward. Fully all-metal and featuring an internal bomb bay, retractable landing gear, an enclosed cockpit, and a gun turret for self-defense, it prompted the largest bomber order the Army’s aviation arm had placed to that point: 121 units. Its primacy was short-lived. The Douglas B-18—derived from the DC-2 commercial transport—replaced it following an Air Corps order in January 1936 for 133 units, quickly followed by 217 units of the modified B-18A.
The B-17
Many in the Air Corps considered the B-18 already obsolete against Boeing’s four-engine Model 299 prototype, the forerunner of the B-17. The 299 had been designed in response to a 1934 requirement for a bomber capable of reaching Hawaii, Alaska, and Panama. The Air Corps chose the cheaper B-18 instead, and Boeing’s prototype was destroyed in an accident before the competition concluded.
Nevertheless, the Air Corps scraped together funding to order thirteen pre-production Y1B-17s for testing in 1935. Twelve formed an operational squadron in 1937, were redesignated B-17 in 1936, and officially named the Flying Fortress in 1938. The thirteenth became a test-bed and was later redesignated B-17A.
The first production batch, the B-17B, was delivered from June 1939, three months before World War Two began. Design issues meant all B-17Bs were reassigned to non-combat roles by October 1942.
Ground-attack aircraft
In the early post-war years, insufficient engine power prevented development of an effective dedicated ground-attack aircraft, and modified WWI-era DH-4s filled the gap. The Curtiss A-3 biplane, of which 143 were taken into service from 1927, provided a stopgap; it was armed with machine guns and had provision for underwing bombs.
The first purpose-built ground-attack monoplane, the Curtiss YA-8—later redesignated A-8—entered service testing in 1932. Its improved successor, the A-12, served from 1933 to 1941. Alongside it, the Air Corps ordered five units of the Consolidated A-11. Both were followed by the Northrop A-17, a single-engine all-metal monoplane with an enclosed cockpit; 110 units of the original version were built, with a follow-on A-17A featuring retractable undercarriage delivered in 1937–38. In 1938, the Air Corps decided it wanted only multi-engine attack aircraft, rendering the entire A-17 series surplus; most were transferred to other countries.
This article is from the book “Combat Aircraft of the United States Air Force” by Michael Green, available here.
