involved military training and theoretical instruction in aviation. For this purpose seven universities in this country were selected, where instructions were given as to airplane engines and other relevant subjects.

After the 27th of August, 1917, applicants for commissions in the Signal Officers Reserve Corps, Aviation Section, were required to have "at least a high school education."

Twenty-two thousand six hundred and eighty-nine cadets entered these schools during the war, of which 17,540 graduated. Five thousand one hundred and forty-nine were dismissed without graduating. One thousand seven hundred and ninety-one were sent directly overseas for training. Fifteen thousand six hundred and twenty-seven were sent to the United States flying fields, and 122 were discharged immediately upon graduation at the armistice.

In addition to the above, men were entered and graduated from three schools of military aeronautics, as follows: Eight hundred and fifty-two graduated as supply officers out of 963 entered; 726 graduated as engineer officers out of 964 entered; 789 graduated as adjutants out of 887 entered; and 206 graduated as balloonists out of 219 entered.

The ground schools were conducted by universities, under a contract with the United States, by which the universities furnished the necessary buildings and equipment, except of a special nature, such as airplanes, motors, machine guns, etc. The universities furnished the instructors. The Government paid a tuition for each man of $10 per week for the first four weeks, and $5 per week thereafter. The Ground Schools Branch operated the schools of military aeronautics (2910-XXII-XX5, 3880, 3561).

FLYING PERSONNEL

The total number of fliers in the Air Service at the beginning of the war, trained according to the standards at that time, was 139. On the 30th of June, 1918, there were 3,944 flying officers in the United States and 2,840 in the American Expeditionary Force, making a total of 6,784 trained in elementary flying. On the 11th of November, 1918, there were 7,118 flying officers in the United States and 4,307 in Europe, making a total of 11,425 at the time of the armistice (454). On the 11th of November, 1918, 1,674 had been completely trained as pilots and 851 had been completely trained as observers in the American Expeditionary Forces. The total of the flying personnel of all classes completely trained to November 29, 1918, was 2,768 (p. 183). There were 757 completely trained pilots and 481 completely trained observers at the front at the time of the armistice. Ten hundred and ninety-six of the fliers in Europe at the time of the armistice were capable of engaging in battle work. An additional 572 were observers and machine gunners qualified to engage in action. Twelve hundred and thirty-six of these men were engaged in actual combat in strictly American squadrons. The American specialized flying schools had graduated 213 pursuit pilots, 338 bombing pilots, 1,195 Army Corps pilots, 415 bombers, and 907 aerial observers up to the time of the armistice who had not gone overseas where they would receive short final courses of training.

Thirty-two squadrons, 810 pilots and observers, were equipped with French and British planes at the time of the armistice. Thirteen squadrons, 428 pilots and observers, were equipped with American-made planes (2495-2496)..

The total personnel of the Air Service at the beginning of the war was approximately 1,300.

At the armistice there were approximately 20,000 officers and 149,000 men in the Air Service. Of these, 7,726 officers and 70,769 men were overseas, 6,816 officers and 51,229 men being in France.

The Navy and Marine Corps aviation personnel increased during the war from 48 officers and a total of 239 men to 3,117 officers and 45,632 men at the date of the armistice. Of this number, 1,324 officers and 10,137 men and 825 pilots were abroad at the time of the armistice, and 1,793 officers, 35,495 men, and 826 pilots were in the United States at the time of the armistice (3351).

In preliminary training a mortality occured in proportion to hours flown as follows: British, 1,993; French, 2,680; Italians, 949; United States in Italy, 5,853; in France, 2,887; in England, 1,614; in America in 1918, 2,973; in America in 1919, 1,309 (556A).

In advanced training among student pilots a fatality occurred in proportion to hours flown as follows: British, 573; Italian, 747; United States in Italy, 2,644; United States in France, 1,101; United States in England, 384; American Expeditionary Forces, 936; United States in America in 1918, 2,802 (p. 556-a).

American student pilots in preliminary training flew 362,706 hours in 1918.

Six hundred and twenty-eight American-built planes were sent to the front and used, or were in reserve ready for use, on the day of the armistice. Four hundred and seventeen American-built planes were used on the front during the war. The casualties in the use of these planes amounted to less than 15 per cent, including the wounded and prisoners, of the total aviation combat casualties of the United States. These machines were used for observation and bombing purposes, and protection from attack was afforded through formation flying and maneuverable gun defense.

TRAINING PLANES

Planes for the use of military forces naturally divided themselves into two great classes-training and service planes. Training planes are elementary and advanced. Service planes are divided into four classes-pursuit or combat, observation, day bombers, and night bombers. The experience of the United States prior to the war had been confined to the elementary training plane.

The training-plane problem was met in a satisfactory way. Within six months after the declaration of war we had sufficient elementary training planes to satisfactorily meet our needs (pp. 3556, 3536, 152, 153). Deliveries of training planes began in June 1917, in small numbers; 103 were delivered during the month of August, 193 in September, and 340 in October. Approximately 5,646 training planes were delivered prior to the armistice (pp. 516, 517).

These training planes were principally the SJ-1, of which a total of 1,600 were manufactured; the Penguin, of which only 300 were made; and the JN4-D, of which 3,746 were manufactured before

the armistice. The Penguin was a ground plane and its use was finally discontinued. The SJ-1 was not satisfactory. A production of 1,208 JN4-Ds during the year 1917, which proved to be very satisfactory training planes, led to the abandonment of the use of the SJ-1 a few months later. The unsatisfactory work of the A-7-A engine, with which the SJ-1 was equipped, was the principal reason of its abandonment. It was discarded because it was regarded as unnecessarily dangerous.

During the war we produced approximately 2,174 advanced training planes, 20 of which were delivered in December 1917, and 199 in February 1918. These planes were principally what is known as JN-4-H, JN-6-H, S-4-B, S-4-C, and E. I. (p. 517).

In addition to the training planes manufactured in America, the French Government prior to June 30, 1918, furnished 1,984 such planes to our Army, and prior to November 11, 1918, had furnished. a total of 2,605 training planes to the United States forces. The British Government furnished a total of 72 training planes to the United States during the war. The Italian Government furnished 19 in February 1918 (pp. 489, 490).

Service planes were used for advanced training purposes in Europe, the flier being trained upon the particular machine he was expected to use before being sent to the lines. Service planes not of the standard requisite for the line service were frequently used by students in advanced training. The DH-4 was also used for advanced training both in the United States and abroad (p. 218).

SERVICE PLANES

Outside of experimental planes, the United States manufactured 3,431 De Haviland 4's prior to the armistice and completed the manufacture of 4,587 De Haviland 4's up to the 31st of December 1918. The United States also fabricated and delivered in England the parts for 101 Handley-Paiges.

In addition to these American-made service planes, our Army procured from the French 417 service planes before June 30, 1918, and 2,186 prior to November 11, 1918. From the British we procured a total of 189 service planes (489).

The American Expeditionary Forces received from the United States during the war 1,213 service planes.

In some instances, individual fliers have complained that obsolescent combat planes were furnished American fliers. It is particularly claimed that Nieuports were furnished the Americans after a superior Spad came into use in 1918.

The plan of cooperation with the Allies provided that the American pilots were to be furnished the same types of machines as those used by our allies during the same time. Gen. Patrick states: "The machines that were used by all the American Air Service were practically the same as those that were used by their allies and were the equals of those used by their allies" (219). Gen. Mitchel concurs with Gen. Patrick in this view and states, "About six weeks after we went on the line" the French "began to give us their last word in ships. They were just coming out of their production factories at that time" (2650).

Necessarily, as soon as a new machine was placed on the front, all could not be provided at the same time. It was impossible that all fliers could have the same type of machine and immediately discard. the older types. The flier using the obsolete type was naturally dissatisfied under these conditions until he could be provided with the latest. The obsolescent machines were placed on the inactive fronts or else used for advanced training purposes, and the period for which complaint is made of the use of the Nieuport by the Americans when the French had the Spads was a brief one (3632, 2650). One competent flier expressed his preference for the Nieuport over the Spad machine (3686).

Substantially one year was required to design, test, and place a machine in quantity production. Thus, the latest type constantly replaced the earlier, and the newest type of to-day was continually approaching elimination, or partial elimination, by improved machines. This is illustrated by the fact that in February 1918, 63 per cent of the machines used by the French were classified as obsolete. In keeping her contract with the French, America had to meet the same changing situation as the French who made the machines (490, 3314).

PURSUIT PLANES

The United States manufactured no pursuit plane that was used on the front during the war.

In August 1918, the Allied Governments advised the United States to give its attention to the production of heavy machines, rather than single seater pursuit planes (218, 2910-LXXXI). At an Interallied meeting on the 4th of December, it was agreed that the manufacture of pursuit machines should be carried out by the European Allies on condition that they could agree to provide the American Army with the necessary pursuit planes (LXXIX). At that time the French had agreed to furnish planes to the United States (226). Thereafter 1,434 pursuit planes were supplied to the American Expeditionary Forces by our Allies. The French greatly increased their production largely from materials furnished by the United States for the purpose of meeting our needs (489).

However, the United States did not abandon its purpose to construct pursuit planes. Two types of a Spad were originally recommended. One was at that time in an experimental stage and subsequently proved impractical.

The Bristol Fighter as developed in England was accepted as a model for a pursuit and observation plane to be produced in America. This machine was used with satisfactory results by the English. It was recommended for use with a 200 horsepower Hispano engine which we did not have in production at that time, although foreign contracts had been placed with our factories prior to the war. America attempted to accommodate it to an 8-cylinder Liberty motor. After the manufacture of 27 planes, the plan was accepted as a failure, responsibility for which evidently rests with the designing engineers (3450), although at least one of them is recognized as being one of the most competent aeronautical engineers in America. The result of the Bristol effort is incongruous, but it is practically a duplicate of the Italian experience with the S. I. A. after Italy

had been in the war for a considerable longer period than ourselves (133).

THE SPAD CANCELLATION

In July 1917 the Bolling Commission recommended American manufacture of the Spad with a 200 horsepower Hispano engine. A sample single place Spad had been previously ordered to be sent to the United States, but it did not arrive until the 18th of September (2910-LVIII-LIX). Prior to that date the Spad was outclassed by machines at the front (2910-LVI-II, LII-LIV). Gen. Pershing approved of the Spad cancellation (3816). Col. Patrick concurred in that view (165, 166). It also became apparent that the Spad, if manufactured in America, could not reach the front before the summer of 1918 (2910-LIX-LXI), that amount of time, at least, having been demonstrated as essential to secure quantity production and shipment. This situation led the Allies, in agreement, to approve a cablegram on the 8th of November recommending that the United States produce single place machines "already actually under contract and started. Believe we Believe we can obtain here all this type required future." Pursuit planes become antedated more rapidly than any other machines. The fact that transportation from the United States and the assembling of the machine in Europe consumed on an average of three months, was a material consideration in the situation. At that time the contract had been let for the Spad in America, but manufacture had not started. Shortly following this, the interallied agreement was made in Paris for the French supply of single seater pursuit planes for American use. This action was based on the belief, probably well founded, that the two place pursuit plane with gunner and maneuverable_guns, would practically eliminate the single place machine with fixed guns only. The Allies failed to produce the biplane pursuit plane in sufficient quantities to accomplish that result before the war terminated, although anticipating the result one year in advance.

On the day of the armistice, outside of the pursuit planes on the line, the Americans had 192 in the immediate rear, of which 175 were Spads and 17 were SE-5s.

From a financial standpoint the United States lost nothing by the cancellation of the Spad order as production had not begun (3340, 3471, 3472).

The Spad of the type canceled in 1917 would have been an obsolete machine before it could have been manufactured and placed on the western front in 1918. It would, nevertheless, have been useful and a good machine for advanced-training purposes. The Spad, as canceled in 1917, was superseded by later types of the same machine placed on the front in June or July 1918, by the French. During the war about 12 different types of the Spad were placed on the front, the latest model constantly replacing the earlier, but the use of the machines thus made obsolete continued for a time as a matter of necessity.

The rapid changes and experimental features of airplane manufacture are indicated by the fact that during the war the British used 27 types of single-place machines, the French 31, the Italians 13, and the Germans 12.