 
When the United States entered the war against Germany in 1917 there was no phase of her forthcoming industrial effort from which so much was expected as from the building of airplanes and equipment for aerial warfare; yet there was no phase of the immense undertaking in which the United States was so utterly unprepared. In many other branches of the work of providing matÉriel for a modern army, however inadequately acquainted America might be with the developments which had gone on in Europe since 1914, yet she had splendid resources of skill and equipment which could quickly turn from the pursuits of peace to the arts attending warfare. But there was no large existing industry in the United States which could turn easily to the production of airplanes, since such airplanes as were known in Europe in 1917 had never been built in the United States. It seems difficult now for us to realize how utterly unlearned we were, both in official and technical quarters, in the design, the production, or the use of aeronautical equipment in those early days of 1917. Here in America mechanical flight had been born; but we had lived to see other nations develop the invention into an industry and a science that was a closed book to our people. In the three years of warfare before American participation, the airplane had been forced through a whole generation of normal mechanical evolution. Of this progress we were aware only as nontechnical and distant observers. Such military study of the progress as we had conducted was casual. It had, in fact, brought to America scarcely a single basic fact on which we could build our contemplated industry. When the United States became a belligerent no American-built airplane had ever mounted a machine gun or carried any other than the simplest of necessary instruments. Such things as oxygen apparatus, electrically heated clothing for aviators, radio-communication with airplanes, landing and bombing flares, electric lighting systems for planes, bomb-dropping devices, suitable compasses, instruments for measuring height and speed, and the like—in short, all the modern paraphernalia that completes the efficiency of combat airplanes—these were almost entirely unknown to us. The best of the prewar activities of America in this line had produced some useful airplane engines and a few planes which the countries then at war were willing to use only in training of aviators. Within the Army itself there was small nucleus of skill around which could be built an organization expert and sophisticated. We had in the official files no adequate information as to sizes, capacities, and types of planes or engines, or character of ordnance, armament, or aeronautical appliances demanded by the exacting service in which our young birdmen were soon to engage. Even the airplanes on order in April, 1917 (over 350 of them), proved to be of such antiquated design that the manufacturers of them, in the light of their increased knowledge of war requirements a few months later, asked to be released from their contracts. Nor was there in the United States any industry so closely allied to airplane manufacture that its engineers and designers could turn from one to the other and take their places at once abreast of the progress in Europe. There was little or no engineering talent in the United States competent to design fully equipped military aircraft which could compete with Europe. Our aircraft producers must first go to France and England and Italy, and ground themselves in the principles of a new science before they could attempt to produce their own designs or even before they could be safe in selecting European designs for reproduction in this country. The first consideration of the whole program by the Joint Army and Navy Technical Board indicated a figure of 22,000 as the number of airplanes, including both training and battle types, which should be furnished for the use of the Army during the 12 months following July 1, 1917. This figure represented the determination of America to play a major part in aerial warfare. It was not possible for the board to realize at that time all of the problems which would be encountered, and the figures indicated confidence in the ability of the industrial organizations of the United States to meet a difficult situation, rather than an exact plan under which such production might be developed. It is probable that it was not fully realized that the production of this program, with the proper proportion of spare parts required for military operations, meant the manufacture of the equivalent of about 40,000 airplanes. Without an industry then, and with little knowledge or understanding of the problems of military aerial equipment, we faced the task of securing the equivalent of 40,000 airplanes in 12 brief months beginning July, 1917. In one respect we were in a degree prepared in professional skill and mechanical equipment to go ahead on broad lines. This was in the matter of producing engines. The production of aviation engines in America had, indeed, been comparatively slight, but in the automobile industry had been developed a vast engine-building capacity. The detail equipment of automobile shops was not entirely suited to aviation engines, but, nevertheless, it furnished the basis for the future successful production of the Liberty engine and the other engines called for by the air program. 250 AIRPLANES IN THE AIR AT ONE TIME OVER SAN DIEGO, CALIF. TWO ROWS OF AIRPLANES LINED UP AT A TEXAS FLYING FIELD. America succeeded, once the requirements were known, in producing the various accessories of aerial warfare. It was necessary first to learn from foreign sources what these accessories were and how they should be built; but as a rule it was possible to adapt American production resources to the problem, and the difficulties experienced were rather those of determining requirements and the exact adaptation of the various articles to specific airplanes. The achievements of America in aircraft production during the war period may be summarized as follows: In our 19 months of warfare we outdid any one of the belligerent nations in Europe in the production of airplanes in its first 19 months of intensive production. In our second year of war we nearly equaled the record of England in her third. At the end of the effort, after our designers had saturated themselves in the science and were abreast of the developments of Europe, they produced several typical American airplanes which gave promise of being superior to any that Europe was turning out. We created one of the three or four best airplane engines, if not the best of all, that the world had seen, and produced it in great quantities. We took a standard but complicated aero-engine from Europe and not only duplicated it in quantity here but turned out a finer product than the original French makers had been able to obtain with their careful and more leisurely methods. In the steel cylinders of all the aero-engines we built was a capacity for producing some seven or eight million horsepower, an energy equivalent to one-fifth of the commercially practicable water power of the United States. The Liberty engines built could alone do the work of the entire flood of Niagara and have a million horsepower to spare. In three years of warfare the allies had been able to develop only a single machine gun that could be successfully synchronized to fire through a revolving airplane propeller. In 12 months of actual effort America produced two others as good, both susceptible of factory quantity production. We developed new airplane cameras. We carried to new stages the science of clothing aviators. We developed in quantity the wireless airplane telephone that stilled in the ears of the pilot the bedlam of wind and machine guns and engine exhaust and placed him within easy speaking radius of his ground station and his commander in the air. We built balloons at a rate to supply more than our own needs. When the shortage of linen threatened the entire airplane output of the nations opposing Germany, we developed cotton wing fabric that not only substituted for linen, but proved to be better; and in producing a liquid filler to make this fabric wind-tight we established on a large scale an entirely new chemical industry in the United States. Such were the high points in the history of America's aircraft production for war. The details of the developments which led to these results are set forth on the following pages. VIEW OF AIRPLANE IN FLIGHT TAKEN FROM ANOTHER MACHINE. PARACHUTE JUMP FROM AN ALTITUDE OF 7,900 FEET, NOVEMBER 12, 1918. Picture taken at Love Field, Tex., from an airplane. Sketchy and incomplete as was our knowledge of airplane construction in the early days of 1917, it was no more hazy than our notion of how many planes to build. What would constitute overwhelming superiority in the air? As an indication of the rapidity with which history has moved, it may be stated that in January and February of 1917 the Signal Corps discussed the feasibility of building 1,000 planes in a year of construction. This seems now to us a ridiculously low figure to propose as representative of American resources, but in the early weeks of 1917 the construction of a thousand airplanes appeared to be a formidable undertaking. In March, when war was inevitable, we raised this number to 2,500 planes within 12 months; in April, when war was declared, we raised it again to 3,700. But once we were in the war, through the exchange of military missions our designers were taken into the confidence of the aviation branches of the French, British, and Italian Armies and shown then for the first time a comprehensive view of the development of the war plane, both what had been done in the past and what might be expected in the future. As a result our Joint Army and Navy Technical Board in the last week of May and the early part of June, 1917, recommended to the Secretaries of War and the Navy that a building program be started at once to produce the stupendous total of 19,775 planes for our own use and 3,000 additional ones, if we were to train foreign aviators, or approximately 22,000 in all. This was a program worthy of America's industrial greatness. Of these proposed planes, 7,050 were for training our flyers, 725 for the defense of the United States and insular possessions, and 12,000 for active service in France. Such was the task assigned to an industry that in the previous 12 months had manufactured less than 800 airplanes, and those consisting principally of training planes for foreign governments. The expanding national ambition for an aircraft industry was also shown by the mounting money grants. On May 12 Congress voted $10,800,000 for military aeronautics. On June 15 an appropriation of $43,450,000 was voted for the same purpose. Finally on July 24, 1917, the President signed the bill appropriating $640,000,000 for The figure 22,000, however, scarcely indicates the size of this undertaking, as we were to realize before long. We little understood the infinite complications of fully equipping battle planes. Lacking that invaluable experience which Europe had attained in three years of production, we had no practical realization of the fact that for each 100 airplanes an equivalent of 80 additional airplanes must be provided in spare parts. In other words, an effective fighting plane delivered in France is not one plane, but it is one plane and eight-tenths of another; which means that the program adopted in June, 1917, called for the production in 12 months of not 22,000 airplanes but rather the equivalent of 40,000 airplanes. Let us set down the inventory of the Government's own resources for handling this project. The American Air Service, which was then part of the Signal Corps, had had a struggling and meager existence, working with the old pusher type of planes until in 1914 an appropriation of $250,000 was made available for the purchase of new airplanes and equipment. Shortly after this appropriation was granted, five officers were sent to the Massachusetts Institute of Technology for a course in aeronautics. When the war broke out in Europe in August, 1914, these men constituted the entire technically trained personnel of the Air Service of the United States. By April 6, 1917, we had 65 officers in the Air Service, an enlisted and civilian personnel of 1,330, two flying fields, and a few serviceable planes of the training type. This equipment may be compared with that of Germany, France, and England at the time they went to war. Germany is believed to have had nearly 1,000 airplanes in August, 1914; France had about 300; and England barely 250. America's 224, delivered up to April 6, 1917, were nearly all obsolete in type when compared with the machines then in effective service in France. No sooner had the United States embarked upon the war than the agents of the European manufacturers of airplanes descended upon the Aircraft Board in swarms. France and Italy had both adopted the policy of depending upon the private development of designs for their supplies of airplanes, with the result that the builders of each country had produced a number of successful types of flying machines and an even greater number of types of engines. On the assumption that the United States would adopt certain of these types and build them here, the agents for the Sopwiths, the Capronis, the Handley-Pages, and many others proceeded to demonstrate the particular excellences of their various articles. Out of this confusion of counsel stood one pertinent fact in relief—the United States would have to pay considerable royalties for the use of any of these European devices. As to the relative merits of types and designs, it was soon apparent that no intelligent decision could be reached in Washington or anywhere but in Europe. Because of our distance from the front and the length of time required to put the American industrial machine into operation on a large scale, it was necessary that in advance we understand types and tendencies in aircraft construction, so that we might anticipate aircraft development in such special designs as we might adopt. Otherwise, if we accepted the types of equipment then in use in Europe, by the time we had begun producing on a large scale a year or so later we would find our output obsolete and out of date, so rapidly was the aircraft art moving. Consequently, in June the United States sent to Europe a commission of six civilian and military experts, headed by Maj. R. C. Bolling, part of whose duties was to advise the American War Department as to what types of planes and engines and other air equipment we should prepare to manufacture. Also, in April the Chief of the Signal Corps had cables sent to England, France, and Italy, requesting that aviation experts be sent at once to this country; and shortly after this we dispatched to Europe more than 100 skilled mechanics to work in the foreign engine and airplane plants and acquire the training that would make them the nucleus of a large mechanical force for aircraft production in this country. But while these early educational activities were in progress, much could be done at home that need not await the forthcoming reports from the Bolling mission. We had, for instance, in this country several types of planes and engines that would be suitable for the training fields which were even then being established. The Signal Corps, therefore, bent its energies upon the manufacture of training equipment, leaving the development of battle aircraft to come after we should know more about that subject. It was evident that we could not equip an airplane industry and furnish machines to our fliers abroad before the summer of 1918; and so we arranged with France for this equipment by placing orders with French factories for 5,875 planes of regular French design. These were all to be delivered by July 1, 1918. In the arrangement with the French factories we agreed to supply from the United States a great deal of the raw materials for these machines, and the contract for furnishing these supplies was given to J. G. White & Co. of New York City. This concern did a creditable job, shipping about 5,000,000 feet of lumber, much necessary machinery, and a multitude of items required in the fabrication of airplanes, all to the value of $10,000,000. The total weight of the shipments on this contract was something like 23,000 tons, this figure including 7,500 tons of lumber. The other tonnage consisted of tubing of steel, brass, copper and alumi The orders for French planes were divided as follows: 725 Nieuport training planes, 150 Spad training planes, 1,500 Breguet service planes; 2,000 Spad service planes; and 1,500 New Spad or Nieuport service planes. The decision between the New Spad or Nieuport service planes was to be made as soon as the New Spad could be tested. These planes were to be delivered in specified monthly quantities increasing in number until the total of 1,360 planes should be placed in our hands during the month of March, 1918, alone. The contracts were to be concluded in June with the delivery of the final 1,115 planes. We also contracted for the manufacture of 8,500 service engines of the Renault, Hispano and Gnome makes, all of these to be delivered by the end of June. When the armistice ended the fighting, we had produced a total of 11,754 airplanes in America, together with most of the necessary spare parts for about one-third of them. While a large part of the American airplanes built in the war period were of the training type rather than the service, or battle, type, nevertheless it was necessary that we have a large equipment of training planes in order to prepare the swiftly expanding personnel of the Air Service for its future activity at the front. The nations associated with us in the war, however, had produced their training equipment prior to our participation as a belligerent, and at the time we entered the war the French, British, and Italians were producing only enough training planes to maintain their training equipment and were going in heavily with the rest of their airplane industries for the production of service planes. With these considerations in mind, the reader may make an interesting comparison of British and American plane production, the British figures being for both the British Army and the British Navy, whereas the American figures are for the American Army alone. In the following table of comparison the British figures are based on the Lockhart Report of November 1, 1918:
Broadly stated, and without reference to types of planes produced, these figures mean that the United States in her second year of the war produced for the American Army alone almost as many airplanes as Great Britain in her third year of the war built for both her army and navy. In October, 1918, factories in this country turned out 1,651 planes, which, without allowing for the monthly expansion in the production, was at the rate of 20,000 planes per year. Assuming no increase in the October rate of production, we would have attained the 22,000 airplanes in 23 months after July 1, 1917, the date on which the production effort may be said to have started. Our production of fighting planes in the war period was 3,328. On the day the armistice was signed we had received from all sources 16,952 planes. Of these 5,198 had been produced for us by the allies. We had 48 flying fields, 20,568 Air Service officers, and 174,456 enlisted men and civilian personnel. These figures do not mean that we had more than 17,000 planes on hand at that time, because the mortality in airplanes is high from accidents and ordinary wear and tear. THE PROBLEM OF MATERIAL.Once we had started out on this enterprise we soon discovered that the production of airplanes was something more than a mere manufacturing job. With almost any other article we might have made our designs, given orders to the factories, and rested in the security that in due time the articles would be forthcoming. But with airplanes we had to create the industry; and this meant not only the equipping of factories, but the procurement and sometimes the actual production of the raw materials. For instance, the ideal lubricant for the airplane motor is castor oil. When we discovered that the supply of castor oil was not nearly sufficient for our future needs, the Government itself secured from Asia a large quantity of castor beans, enough to seed more than 100,000 acres in this country and thus to provide for the future lubrication for our motors. This actual creation of raw materials was conducted on a much larger scale in the cases of certain other commodities used in airplane construction, particularly in the production of lumber and cotton and in the manufacture of the chemicals for the "dope" with which the airplane wings are covered and made air-tight. An airplane must have wings and an engine with a propeller to make it go; and, like a bird, it must have a tail to make it fly straight and a body (fuselage) to hold all together. Part of the tail (the rudder) moves sideways and steers the airplane from left to right; part moves up and down (the elevators) and makes the airplane go up or down, and parts of the wings (the ailerons) move up and down and make the airplane tip from side to side. All of these things must be connected to the controls in the hands of the pilot. The front All of the airplanes built for the United States during the war were tractor biplanes. In a plane of the tractor type the propeller is in front and pulls the machine. The biplane is so called because it has two planes or wings, one above the other. The biplane has been the most largely used of all types in war for two reasons: first, the struts and wires between the planes form a truss structure, and this gives the needed strength; and second, there is less danger of enemy bullets wrecking a biplane in the air because its wing support is greater than that of the monoplane or single-winged machine. Since the airplane can lift only a limited weight, every part of the mechanism must be as light as possible. An airplane engine weighs from 2 to 3 pounds per horsepower, whereas an automobile motor weighs from 8 to 10 pounds per horsepower. The skeleton of the airplane is made of wood, mostly spruce, with sheet-steel fittings to join the wood parts together, and steel wires and rods to make every part a truss. This skeleton is covered with cloth, and the cloth is stretched and made smooth by dope. Wood, sheet steel, wire, cloth, varnish—these are the principal components of an airplane. As raw materials they all seem easy to obtain in America. And so they are in peace times and for ordinary purposes. But never before had quality been so essential in an American industry, from the raw material up to the finished product—quality in the materials used, and quality in the workmanship which fashions the parts. But combined with this quality we were forced to produce in quantities, bounded only by our own physical limitations, and these quantities must include not only the materials for our own air program but also some of the principal raw materials used by the airplane builders in France and England, specifically, all of the spruce which the allies would require and, later, much of the wing fabric and dope for their machines. Quite early it was apparent to us that we had on our hands a problem in spruce production which the Government itself must solve, if the airplane undertaking were not to fail at the outset. When we entered the war linen was exclusively used for the covering of wings; and it developed almost immediately that the United Kingdom was practically the sole source of linen. But the Irish looms could not begin to furnish us with our needs for this commodity. Later on came up the question of supplying dope and castor oil. Finally, during the last months of the war, it became necessary for us to follow up the production of all classes of our raw material, particularly in working out a suitable supply of steel tubing. But our great creative efforts in raw materials were confined to spruce, fabric, and dope. The lumber problem involved vast questions of an industrial and technical character. We had to conduct a campaign of education in the knowledge of aircraft requirements that reached from the loggers themselves in the woods to the sawmill men, to the cut-up plants, and then followed through the processes of drying and sawing to the proper utilization of the lumber in the aircraft factories. In working out these problems, while we drew heavily upon the experience in Europe, yet we ourselves added our own technical skill to the solution. The Signal Corps was assisted by the forest products laboratory at Madison, Wis., and by the wood section of the inspection department of the Bureau of Aircraft Production. The United States Forest Service contributed its share of technical knowledge. At the end of the war we considered that our practice in the handling of aircraft lumber was superior to that of either France or England. THE SPRUCE PROBLEM.Each airplane uses two distinct sorts of wood—first, the spruce or similar lumber for the wing beams or other plane parts; and second, mahogany, walnut, or other hardwoods for propellers. In each case the Army production authorities were involved both in securing the lumber and in educating manufacturers to handle it properly. In an ordinary biplane there are two beams for each lateral wing, eight beams to the plane. These form the basis of strength for the wings. Because of the heavy stresses put upon the airplanes by battle conditions, only the most perfect and straight-grained wood is suitable for these beams. All cross-grained or spiral-grained material, or material too coarse in its structure, is useless. Spruce is the best of all woods for wing beams. Our problem was to supply lumber enough for the wing beams, disregarding the other parts, as all other wood used in the manufacture of planes could be secured from cuttings from the wing-beam stock. At the beginning we built each beam out of one piece of wood; and this meant that the lumber must be extra long, thick, and perfect. Until we learned how to cut the spruce economically we found that only a small portion of the lumber actually logged was satisfactory for airplanes. An average sized biplane uses less than 500 feet of lumber. In the hands of skilled cutters this quantity can be worked out of 1,000 feet of rough lumber. But in the earlier days of the undertaking as high as 5,000 feet of spruce per plane were actually used because of imperfections in the lumber, lack of proper inspection at the mills, and faulty handling in transit and in the factories. We also used certain species of fir in building training planes. This wood is, like spruce, light, tough, and strong. The only great source of supply of these woods was in the Pacific Northwest, although there was a modest quantity of suitable timber in West Virginia, North Carolina, and New England. While at first we expected to rely upon the unaided efforts of the lumber producers, labor difficulties almost immediately arose in the Northwest to hinder the production of lumber. The effort, too, was beset with difficulties of a physical nature, since the large virgin stands of spruce occurred only at intervals and often at long distances from the existing railroads. By the middle of October, 1917, it became evident that the northwestern lumber industry unaided could not deliver the spruce and fir; and the Chief of Staff of the Army formed a military organization to handle the situation. On November 6, 1917, Col. Brice P. Disque took command of the Spruce Production Division of the Signal Corps, this organization later being transferred to the Bureau of Aircraft Production. When Col. Disque went into the Northwest he found the industry in chaotic condition. The I. W. W. was demoralizing the labor forces. The mills did not have the machinery to cut the straight-grained lumber needed and their timber experts were not sufficiently skilled in the selection and judging of logs to secure the maximum footage. The whole industry was organized along lines of quantity production and desired to avoid all high quality requirements insisted upon by the Government. One of the first acts of the military organization was to organize a society called the Loyal Legion of Loggers and Lumbermen, the "L. L. L. L.," to offset the I. W. W. propaganda, its platform being, no strikes, fair wages, and the conscientious production of the Government's requirements. On March 1, 1918, 75,000 lumbermen and operators agreed without reservation to give Col. Disque power to decide all labor disputes. The specifications for logs were then standardized and modified as far as practicable to meet the manufacturers' needs. We arranged financial assistance that they might equip their mills with the proper machinery. We instituted a system of instruction for the personnel. Finally, the Government fixed a price for aircraft spruce that stabilized the industry and provided against delays from labor disputes. While these basic reforms were being instituted our organization had energetically taken up the physical problems relating to the work. We surveyed the existing stands of spruce timber, built railroads connecting them with the mills, and projected other railroads far into the future. We began and encouraged logging by farmers in small operations. By these and other methods employed, the efficiency of this production effort gradually increased. In all, we took 180,000,000 feet of aircraft lumber out of the northwestern forests. To the allies went 120,000,000 feet; to the United States Army and Navy, 60,000,000 feet. ASSEMBLING DE HAVILAND-4 WINGS AT THE DAYTON-WRIGHT PLANT. SEWING FABRIC ON AIRPLANE WINGS. APPLYING THE DOPE TO AIRPLANE WINGS. Yet when we had resolved the difficulties in the forests only part of the problem had been met. Next came the intricate industrial question of how to prepare this lumber for aircraft use. We possessed little knowledge as to the proper methods of seasoning this timber. The vast majority of woodworking plants in this country, such as furniture and piano makers, had always dried lumber to the end that it might keep its shape. We now were faced with the technical question of drying lumber so as to preserve its strength. The forest products laboratory worked out a scientific method for this sort of seasoning. Incidentally they discovered that ordinary commercial drying had seldom been carried on scientifically. The country will receive a lasting benefit from this instruction carried broadcast over the industry. In the progress of our wood studies we discovered a method of splicing short lengths of spruce to make wing beams and in the later months of the production used these spliced beams exclusively at a great saving in raw materials. The use of laminated beams would probably have become universal in another year of warfare. COTTON FABRIC AND DOPE.The flying surfaces of an airplane are made by stretching cloth over the frames. When we came into war it was supposed that linen was the only common fabric with sufficient strength for this use, and linen was almost exclusively used by the airplane builders, although Italian manufacturers were trying to develop a cotton fabric. Of the three principal sources of flax, Belgium had been cut off from the allies, Russia was isolated entirely after the revolution there, and Ireland was left as the sole available land from which flax for airplane linen could be obtained. As late as August, 1917, England assured us that she could supply all of the linen that would be needed. It rapidly became evident that England had underestimated our requirements. An average airplane requires 250 yards of fabric, while some of the large machines need more than 500 yards. And these requirements do not take into consideration the spare wings which must be supplied with each airplane. This meant a demand for millions of yards put upon the Irish supply, which had no such surplus above allied needs. For some time prior to April 6, 1917, the Bureau of Standards at Washington had been experimenting with cotton airplane cloths. Out of the large variety of fabrics tested several promising experimental cloths were produced. The chief objection to cotton was that the dope which gave satisfactory results on linen failed to work with uniformity on cotton. Therefore, it became apparent that if we were to use cotton fabric, we should have to invent a new dope. Two grades of cotton airplane cloth were finally evolved—A, which had a minimum strength of 80 pounds per inch, and B, with a minimum strength of 75 pounds per inch. Grade A was later universally We placed our first orders for cotton airplane fabric in September, 1917—orders for 20,000 yards—and from that time on the use of linen decreased. By March of 1918 the production of cotton airplane cloth had reached 400,000 yards per month. In May the production was about 900,000 yards; and when the war ended this material was being turned out at the rate of 1,200,000 yards per month. Starting with a few machines, our cotton mills had gradually brought 2,600 looms into the enterprise, each loom turning out about 120 yards of cloth in a week. A total of 10,248,355 yards of cotton fabric was woven and delivered to the Government—over 5,800 miles of it, nearly enough to reach from California to France. The use of cotton fabric so expanded that in August, 1918, we discontinued the importations of linen altogether. There was, however, danger that we would be limited in our output of cotton fabric if there were any curtailment in the supply of the long-staple sea-island and Egyptian cotton of which this cloth is made. To make sure that there would be no shortage of this material the Signal Corps in November, 1917, went into the market and purchased 15,000 bales of sea-island cotton. This at all times gave us an adequate reserve of raw material for the new fabric. Cotton proved to be not only an admirable substitute for linen but even a better fabric than the original cloth which had been used. No matter how abundant the supply of flax may be, it is unlikely that linen will ever again be used in large quantities for the manufacture of airplane wings. Thus, as the airplane situation was saved by the prompt action of the Signal Corps in organizing and training the spruce industry, so again its decision to produce cotton fabric and its prompt action in cornering the necessary cotton supply made possible the uninterrupted expansion of the allied aviation program. The wings of an airplane must not only be covered with fabric, but the fabric must be filled with dope, which is a sort of varnish. The function of the dope is to stretch the cloth tight and to create on it a smooth surface. After the dope is on the fabric the surface is protected further by a coat of ordinary spar varnish. We found in the market two sorts of dope which were being furnished to airplane builders of all countries by various chemical and varnish manufacturers. One of these dopes was nitrate in character and was made from nitrocellulose and certain wood-chemical solvents including alcohol. This produced a surface similar to that of a photographic film. The other kind of dope had an acetate base and was made from cellulose-acetate and such wood chemical solvents as acetone. The nitrate dope burned rapidly when ignited but the acetate type was slow burning. Thus in training planes not subject to attack by enemy incendiary bullets the nitrate dope would be fairly satisfactory, but in the fighting planes the slow-burning acetate dope was a vital necessity. Up to our participation in the war the dopes produced in the United States were principally nitrate in character. It was evident that we should make our new dope acetate in character to avoid the danger of fire. But for this we would require great quantities of acetone and acetate chemicals, and a careful canvass of the supply of such ingredients showed that it would be impossible for us to obtain these in anything like the quantities we should require without developing absolutely new sources of production. Already acetone and its kindred products were being absorbed in large quantities by the war production of the allies. The British Army was absolutely dependent upon cordite as a high explosive. Acetone is the chemical basis of cordite; and therefore the British Army looked with great concern upon the added demand which the American aviation program proposed to put upon the acetone supply. We estimated that in 1918 we would require 25,000 tons of acetone in our dope production. The British war mission in this country submitted figures showing that the war demands of the allies, together with their necessary domestic requirements, would in themselves be greater than the total world production of acetone. There was nothing then for us to do but to increase the source of supply of these necessary acetate compounds; and this was done by encouraging, financially and otherwise, the establishment of 10 large chemical plants. These were located in as many towns and cities, as follows: Collinwood, Tenn.; Tyrone, Pa.; Mechanicsville, N. Y.; Shawinigan Falls, Canada; Kingsport, Tenn.; Lyle, Tenn.; Freemont, Mo.; Sutton, W. Va.; Shelby, Ala.; and Terre Haute, Ind. But it was evident that before these plants could be completed the airplane builders would be needing dope; and therefore steps were taken to keep things going in all the principal countries fighting Germany until the acetate shortage could be relieved. In December, 1917, we commandeered all the existing American supply of acetate of lime, the base from which acetone and kindred products are made. Then we entered into a pool with the allied governments to ration these supplies of chemicals, pending the era of plenty. Our agency in this pool was the wood-chemical section of the War Industries Board, whereas the allies placed their demands in the hands of the British war mission. These two boards allocated the acetate chemicals among the different countries according to the urgency of their demands. Since it was evident there might be financial losses incurred as the result of the commandeering order or in the project of the new Government chemical plants, the British war mission agreed THE TRAINING PLANES.The actual building of the airplanes gave a striking example of the value of previous experience, either direct or of a kindred nature, in the quantity production of an article. What airplanes we had built in the United States—and the number was small, being less than 800 in the 12 months prior to April, 1917—had been entirely of the training type. These had been produced principally for foreign governments. But this slight manufacture gave us a nucleus of skill and equipment that we were able to expand to meet our own training needs almost as rapidly as fields could be equipped and student aviators enlisted. The training-plane program can be called a success, as the final production figures show. Of the 11,754 airplanes actually turned out by American factories, 8,567 were training machines. This was close to the 10,000 mark set as our ambition in June, 1917. There are two types of training planes—those used in the primary instruction of students and those in the advanced teaching, the latter approaching the service planes in type. The primary plane carries the student and the instructor. Each occupant of the fuselage has before him a full set of controls which are interconnected so that the instructor at will can do the flying himself, or correct the student's false moves, or allow the student to take complete charge of the machine. These primary planes fly at the relatively slow speed of 75 miles per hour on the average and require engines so reliable that they need little attention. For our training planes we adopted the Curtiss JN-4, with the Curtiss OX-5 engine, and as a supplementary equipment the Standard Aero Corporation's J-1 plane, with the Hall-Scott A-7-A engine. Both of these planes and both engines had been previously manufactured here. The Curtiss equipment, which was the standard at our training camps, gave complete satisfaction. The J-1 plane was later withdrawn from use, partly because the plane itself was not liked, partly because of the vibration resulting from this Hall-Scott engine, it having only four cylinders, and partly because of the uncertainty of the engine in cold weather. CURTISS JN4-D, USED AS A PRIMARY TRAINING MACHINE. ENGINE, CURTISS OX-5. This machine has a dual control and is used solely for training purposes. V. E. 7. EQUIPPED WITH 180-HORSEPOWER HISPANO-SUIZA ENGINE. An American designed training plane. It was evident that at the start we must turn our entire manufacturing capacity to the production of training planes, since we would need these first in any event, and we were not yet equipped with the knowledge to enable us to make intelligent selections of service types. In taking up the manufacturing problem the first step was to divide the existing responsible airplane plants between the Army and Navy, following the general rule that a single plant should confine its work to the needs of one Government department only. There were, of course, exceptions to this rule. This division gave to the Army the plants of the—
The factories which fell to the Navy were those of the—
Of these concerns, Curtiss, Standard, Burgess, L. W. F., Thomas-Morse, and Wright-Martin were the only ones which had ever built more than 10 machines. These factories were quite insufficient in themselves to carry out the enterprise. Therefore it became necessary to create other airplane plants. Two new factories thereupon sprang into existence under Government encouragement. The largest producer of automobile bodies was the Fisher Body Co., at Detroit, Mich. The manufacture of automobile bodies is akin to the manufacture of airplanes to the extent that each is a fabrication of accurate, interchangeable wood and sheet-steel parts. The Fisher organization brought into the enterprise not only machinery and buildings but a skilled organization trained in such production on a large scale. At Dayton, Ohio, the Dayton-Wright Airplane Corporation was created. With this company was associated Orville Wright, and its engineering force was built up around the old Wright organization. A number of immense buildings which had been recently erected for other purposes were at once utilized in this new undertaking. As an addition to these two large sources of supply, J. G. White & Co. and J. G. Brill & Co., the well-known builders of street cars, formed the Springfield Aircraft Corporation at Springfield, Mass. At this point in the development we were not aware of the great production of spare parts that would be necessary. Yet we did understand that there must be a considerable production of spares; and in order to take the burden of this manufacture from the regular airplane plants, and also to educate other factories up to the point where they might undertake the construction of complete airplanes, we placed many contracts for spare parts with widely scattered concerns. Among the principal producers of spares were the following:
For a long time the supply of spare parts was insufficient for the needs of the training fields. This was only partly due to the early lack of a proper realization of the quantity of spares that would be required. The production of spares on an adequate scale was hampered by numerous manufacturing difficulties incident to new industry of any sort in shops unacquainted with the work, and by a lack of proper drawings for the parts. As to the training planes themselves, with all factories in the country devoting themselves to this type exclusively at the start, the production soon attained great momentum. The Curtiss Co. particularly produced training planes at a pace far beyond anything previously obtained. The maximum production of JN-4 machines was reached in March, 1918, when 756 were turned out. Advanced training machines are faster, traveling at about 105 miles per hour; and they carry various types of equipment to train observers, gunners, photographers, and radio men. For this machine we adopted the Curtiss JN-4H, which was substantially the same as the primary training plane, except that it carried a 150-horsepower Hispano-Suiza engine. We also built a few "penguins," a kind of half airplane that never leaves the ground; but this French method of training with penguins we never really adopted. The finishing school for our aviators was in France, where the training was conducted in Nieuports and other fighting machines. In July, 1918, we reached the maximum production of the advanced training machines, the output being 427. As the supply of primary training planes met the demands of the fields, the production was The actual production of training planes by months was as follows:
THE SERVICE PLANES.It was not until we took up the production of fighting, or service, airplanes that we came into a full realization of the magnitude of the engineering and manufacturing problems involved. We had perhaps a dozen men in the United States who knew something about the designing of flying machines, but not one in touch with the development of the art in Europe or who was competent to design a complete fighting airplane. We had the necessary talent to produce designs and conduct the manufacture of training planes; but at the outset, at least, we were unwilling to attempt designs for service planes on our own initiative. At the beginning we were entirely guided by the Bolling mission in France as to types of fighting machines. In approaching this, the more difficult phase of the airplane problem, our first act was to take an inventory of the engineering plants in the United States available for our purposes. With the Curtiss Co. was Glenn Curtiss, a leader of airplane design, and several competent engineers. The Curtiss Co. had been the largest producers in the United States of training machines for the British and had had the benefit of assistance from British engineers, and so possessed more knowledge and experience to apply to the service-plane problem than any other company. For this reason we selected this plant to duplicate the French Spad plane, the story of which undertaking will be told further on. Orville Wright, the pioneer of flying, was not in the best of health, but was devoting his entire time to experimental work in Dayton. The Burgess factory at Marblehead, the Aeromarine plants at Nutley and Keyport, N. J., and the Boeing Airplane Co. at Seattle were to work exclusively for the Navy, according to the mutual agreement, taking their aeronautical engineers with them. This gave the Army the engineering resources of the Curtiss, Dayton-Wright, and Thomas-Morse companies. Quite early we decided to give precedence in this country to the observation type of service plane, eliminating the single-place fighter altogether and following the observation planes as soon as possible with production of two-place fighting machines. This decision was based on the fact, not always generally remembered, that the primary purpose of war flying is observation. The duels in the air that occurred in large numbers, especially during the earlier stages of the war, were primarily to protect the observation machines or to prevent observation by enemy machines. The first service plane which we put into production and which proved to be the main reliance of our service-plane program was the De Haviland-4, which is an observation two-place airplane propelled by a Liberty 12-cylinder engine. As soon as the Bolling mission began to recommend types of service machines, it sent samples of the planes thus recommended. The sample De Haviland was received in New York on July 18, 1917. After it had been studied by various officers it was sent to Dayton. It had reached us without engine, guns, armament, or many other accessories later recommended as essential to a fighting machine. Before we could begin any duplication the plane had to be redesigned to take our machine guns, our instruments, and our other accessories, as well as our Liberty engine. The preliminary designing was complete, and the first American-built De Haviland model was ready to fly on October 29, 1917. Figure 11 does not tell quite the complete story of De Haviland production, since in August and September, 204 De Haviland planes which had been built were shipped to France without engines and were there knocked down to provide spare parts for other De Havilands in service. These 204 machines, therefore, do not appear in the production total. Adding them to the figures above, we find that the total output of De Haviland airplanes up to the end of December, 1918, was in number 4,587. DE HAVILAND-4. USED FOR OBSERVATION, RECONNAISSANCE, COMBAT, DAY BOMBING, AND DEFENSIVE FIGHTING. Engine, Liberty 12-cylinder, 400-horsepower. Weight, empty, 2,391 pounds. Weight, full load, 3,582 pounds. Ground speed is 124.7 miles per hour. Speed at 10,000 feet, 117 miles per hour. Speed at 15,000 feet, 113 miles per hour. 10,000 feet is reached in 14 minutes with full load. Ceiling, 19,500 feet. UNITED STATES DE HAVILAND 9-A. This is the American development of the British DH-4.
The production of the model machine only served to show us some of the problems which must be overcome before we could secure a standard design that could go into quantity production. Experimental work on the De Haviland continued during December, 1917, and January and February, 1918. The struggle, for it was a struggle, to secure harmony between this English design and the American equipment which it must contain ended triumphantly on the 8th day of April, 1918, when the machine known as No. 31 was completely finished and established as the model for the future De Havilands. The characteristics of the standard American De Haviland-4 were as follows:
Endurance here means the length of time the fuel supply will last. The ceiling is the maximum altitude at which the plane can be maneuvered in actual service. Ground level means only far enough above the ground to be clear of obstructions. The first De Havilands arriving in France were immediately put together, such remediable imperfections as existed were corrected then and there, and the machines were flown to the training fields. The changing and increasing demands of the service indicated the advisability of certain changes of design. The foreign manufacturers had brought out a covering for the gasoline tanks, making them nearly leak-proof, even when perforated by a bullet. In the first De Havilands the location of the principal gas tanks between the pilot and the observer was not the best arrangement in that the men were too far apart from each other so that, if the machine went down, the pilot would be crushed by the gas tank. Also the radius of action was not considered to be great enough, even though the later machines of this type carried 88 gallons of gasoline. As a result the American aircraft designers brought out an improved De Haviland known as the 9-A. This carried a Liberty-12 engine; and the main differences between it and the De Haviland-4 were new locations for pilot and tanks, their positions being changed about, increased gasoline capacity, and increased wing surface. The machine was a cleaner, more finished design, showed slightly more speed, and had a greater radius of action than the De Haviland-4 which it was planned to succeed. We ordered 4,000 of these new machines from the Curtiss Co., but the armistice cut short this production. The difficulties in the way of producing new service planes on a great scale without previous experience in such construction is clearly shown in the attempts we made to duplicate other successful foreign planes. On September 12, 1917, we received from the aviation experts abroad a sample of the French Spad. We had previously been advised to go into a heavy production of this model and had made arrangements for the Curtiss Co. at Buffalo to undertake the work. This development was well under way when in December a cablegram was received from Gen. Pershing advising us to leave the production of all single-place fighters to Europe. As a result we canceled the Spad order, and after that we attempted to build no single-place pursuit planes. THE LEPERE CORPS OBSERVATION PLANE. THE LEPERE (CAMOUFLAGED). THE ENGINE IS A LIBERTY 12-CYLINDER, 400-HORSEPOWER. This plane was developed in the United States. At the time this course seemed to be justified. The day of the single seater seemed to be over. The lone occupant of the single seater can not keep his attention on all directions at once; and as the planes grew thicker in the air, the casualties among flyers increased. But the development of formation flying restored the single-place machine to favor. The formation had no blind spot, thus removing the principal objection to the single seater. The end of the war found the one-man airplane more useful than ever. Our concentration here, however, was upon two-place fighters. On August 25, 1917, we received from abroad a sample of the Bristol fighting plane, a two-seat machine. The Government engineers at once began redesigning this machine to take the Liberty-12 engine and the American ordnance and accessories. The engine which had been used in the Bristol plane developed 275 horsepower. We proposed to equip it with an engine developing 400 horsepower. The Bristol undertaking was not successful. The fact that later in the airplane program American designers successfully developed two-seater pursuit planes around the Liberty-12 engine shows that the engine decision was not the fault in the Bristol failure. There were repeated changes in the engineering management of the Bristol job. First the Government engineers alone undertook it; then the Government engineers combined with the drafting force of the airplane factory; finally the Government placed on the factory the entire responsibility for the job, without, however, permitting the manufacturer to correct any of the basic principles involved. All in all, the development of an American Bristol was most unsatisfactory, and the whole project was definitely abandoned in June, 1918. The fundamental difficulty in all of these attempts was that we were trying to fit an American engine to a foreign airplane instead of building an American airplane around an American engine. It was inevitable that this difficulty should arise. We had skill to produce a great engine and did so, but for our earliest models of planes for this engine we relied upon the foreign models until we were sufficiently advanced in the art to design for ourselves. We were successful in making the adaptation only in the case of the De Haviland and then only after great delay. But eventually we were to see some brilliantly successful efforts to design a two-place fighter around the Liberty-12. We had need of such a mechanism to supplement the De Haviland observation-plane production and make a complete service-plane program. On January 4, 1918, Capt. Lepere, a French aeronautical engineer, who had formerly been with the French Government at St. Cyr, began experimental work on a new plane at the factory of the Packard Motor Car Co. By May 18 his work had advanced to a stage where The performance of the Lepere plane in the air is indicated by the following figures:
Here at last was a machine that performed brilliantly in the air and contained great possibilities for quantity production, because it was designed from the start to fit American manufacturing methods. We placed orders for 3,525 Lepere machines. None of the factories, however, had come into production with the Lepere on November 11, 1918. Seven sample machines had been turned out and put through every test. It was the belief of those in authority that at last the training and technique of the best aeronautical engineers of France had been combined with the Liberty, probably the best of all aerial engines; and it was believed that the spring of 1919 would see the Yankee fliers equipped with American fighting machines that would be superior to anything they would be required to meet. Nor were these expectations without justification. The weeks and months following the declaration of the armistice and extending through to the spring of 1919 were to witness the birth of a whole brood of new typically American designs of airplanes of which the Lepere was the forerunner. In short, when the armistice brought the great aviation enterprise to an abrupt end, the American industry had fairly caught that of Europe, and America designers were ready to match their skill against that of the master builders of France, Great Britain, Italy, and the central powers. The Lepere 2-seated fighter was quickly followed by two other Lepere models—one of them, known as the Lepere C-21, being armored, and driven by a Bugatti engine, and the other a triplane driven by two Liberty engines and designed to be a day bomber. Then the first American designed single-seat pursuit planes began making their appearance—the Thomas-Morse pursuit plane, its 164 miles an hour at ground level, making it the fastest airplane ever tested by our Government, if it were not the speediest plane ever built; the Ordnance Engineering Corporation's Scout, an advanced training plane; and several others. In two-seater fighting planes there was the Loening monoplane, an extremely swift and advanced type. There were several other new two-seaters designed experimentally in some instances and some of them giving brilliant promise. THE LOENING MONOPLANE. This is one of the new distinctively American planes. THE LOENING TWO-PLACE PURSUIT PLANE. Perhaps the severest and most exacting critic of aviation material is the aviator who has to fly the plane and fight with the equipment at the front. Brig. Gen. William Mitchell, then a colonel, was sent to France in 1917. He became in succession chief of the air service of the First Army Corps, chief of the air service of the First Army, and finally chief of the air service of the American group of armies in France. He commanded the aerial operations at the reduction of the St. Mihiel salient, where he gained the distinction of having commanded more airplanes in action than were ever assembled before under a single command. At St. Mihiel there were 1,200 allied planes in action, including, with our own, French, English, and Italian planes. Gen. Mitchell, therefore, is a high authority as to the relative merits of air equipment from the airman's standpoint. In the spring of 1919, after a thorough investigation of the latest types of American planes and aerial equipment at the Wilbur Wright Field at Dayton, he sent to the Director of Air Service, Washington, D. C., the following telegram under date of April 20, 1919: I recommend the following airplanes in the numbers given be purchased at once: 100 Lepere 2-place corps observation, 50 Loening 2-place pursuit, 100 Ordnance Engineering Corporation 1-place pursuit, 100 Thomas-Morse 1-place pursuit, 50 USD9-A day bombardment, 700 additional Hispano-Suiza 300-horsepower engines, 2,000 parachutes. All of the above types are the equal of or better than anything in Europe. Mitchell. Now, let us see some of the specifications and performances of these new models. The USD9-A, being the redesigned and improved De Haviland 4, may be given a place as a latest model. It is a two-place bombing plane of the tractor biplane type, equipped with a Liberty 12 engine and weighing 4,872 pounds, loaded with fuel, oil, guns, and bombs, and with its crew aboard. With this weight its performance record in the official tests at Wilbur Wright Field in Dayton was as follows:
The Lepere C-11, a tractor biplane equipped with a Liberty 12 engine, Packard make, weighing with its load aboard 3,655 pounds, performed as follows in the tests at the Wilbur Wright Field:
The Lepere carries two Marlin guns synchronized with the propeller and operated by the pilot and two Lewis guns operated by the observer. A total of 1,720 rounds of ammunition is carried. The Loening monoplane, a tractor airplane equipped with an Hispano-Suiza 300-horsepower engine and representing, loaded, a gross weight of 2,680 pounds, its military load including two Marlin and two Lewis machine guns, performed as follows at the Wilbur Wright Field:
The Ordnance Scout with a Le Rhone 80-horsepower engine, weighing, loaded, 1,117 pounds, is an advanced training plane. In its official test at Wilbur Wright Field it performed as follows:
The Thomas-Morse MB-3 pursuit plane, a tractor biplane equipped with an Hispano-Suiza 300-horsepower engine, weighing, including its crew but without military load, 1,880 pounds, in unofficial tests at Wilbur Wright Field, performed as follows:
THE THOMAS-MORSE PURSUIT PLANE. S. E. 5. EQUIPPED WITH 180-HORSEPOWER HISPANO-SUIZA ENGINE. The Thomas-Morse pursuit plane is armed with two Browning machine guns synchronized with the propeller and carries 1,500 rounds of ammunition. Uncertain as we were originally as to types of pursuit and observation planes to produce in this country, we were still more uncertain as to designs of night-bombing machines. These relatively slow weight-carrying planes were big and required the motive power of two or three engines, with the complications attendant upon double or triple power plants. They really presented the most difficult manufacturing problem which we encountered. Until the summer of 1918 there were only two machines of this type which we could adopt, the Handley-Page and the Caproni. We put the Handley-Page into production, not because it was necessarily as perfect as the Caproni, but because we could get the drawings for this machine and could not get the drawings for the Caproni, owing to complications in the negotiations for the right to construct the Italian airplane. We were not entirely satisfied with the decision to build Handley-Pages, because the ceiling, or maximum working altitude which could be attained by this machine, was low; and, 12 months later, when we were in production, we might find the Handley-Pages of doubtful value because of the ever-increasing ranges of antiaircraft guns. We secured a set of drawings, supposed to be complete, for the Handley-Page in August, 1917; but twice during the following winter new sets of drawings were sent from England, and few, if any, of the parts as designed in the original drawings escaped alteration. The Handley-Page has a wing spread of over 100 feet. Therefore, it was evident from the start that such machines could not have the fuselage, wings, and other large parts assembled in this country for shipment complete to Europe. We decided to manufacture the parts in this country and assemble the machines in England, the British air ministry in London having entered into a contract for the creation of an assembling factory at Oldham, England, in the Lancashire district. When it is realized that each Handley-Page involves 100,000 separate parts, the magnitude of the manufacturing job alone may be somewhat understood. But after they were manufactured, these parts, particularly the delicate members made of wood, had to be carefully packed so as to reach England in good condition. The packing of the parts was in itself a problem. We proposed to drive the American Handley-Pages with two Liberty 12 engines in each machine. The fittings, which were extremely intricate pieces of pressed steel work, were practically all to be produced by the Mullins Steel Boat Co. at Salem, Ohio. Contracts for the other parts were placed with the Grand Rapids Air All of the parts were to be brought together previously to ocean shipment in a warehouse built for the purpose at the plant of the Standard Aero Corporation at Elizabeth, N. J. The Standard Aero Corporation was engaged under contract to set up 10 per cent of the Handley-Page machines complete in this country. These were to be used at our training fields. Again, in the case of the Handley-Page, the engineering details proved to be a serious cause of delay. We found it difficult to install the Liberty engines in this foreign plane. When the armistice cut short operations, 100 complete sets of parts had been shipped to England, and seven complete machines had been assembled in this country. None of the American-built Handley-Page machines saw service in France. There had been great delay in the construction of the assembling plant in England, and the work of setting up the machines had only started when the armistice was signed. The performance table of the Handley-Page shows its characteristics as follows:
On its tests 390 gallons of gasoline, 20 gallons of oil, and 7 men were carried, but no guns, ammunition, nor bombs. After a long delay, about January 1, 1918, tentative arrangements had been made with the Caproni interests looking toward the production of Caproni biplanes in this country. These machines had a higher ceiling and a greater speed than the Handley-Page. Capt. d'Annunzio with 14 expert Italian workmen, bringing with him designs and samples, came to this country and initiated the redesigning of the Caproni machine to accommodate three Liberty engines. The actual production of Caproni planes in this country was limited to a few samples which were being tested when the armistice was signed. The factories had tooled up for the production, however, and in a few months Capronis would doubtless have been produced in liberal quantities. The performance of the sample planes in two tests is shown by the following figures:
ONE OF THE SMALL THOMAS-MORSE SCOUTS BESIDE A GIANT HANDLEY-PAGE MACHINE. ARMORED GERMAN AIRPLANE SHOT DOWN ON THE WESTERN FRONT. NIEUPORT SCOUT BESIDE A LOENING MONOPLANE. As we had produced fighting planes built around the Liberty motor, so, too, in the night-bombing class American invention, with the experience of several months of actual production behind it, was able to bring out an American bombing plane that promised to supersede all other types in existence. This machine was designed by Glen L. Martin in the fall of 1918. It was a night-bomber equipped with two Liberty 12-cylinder engines. The Martin spread of 75 feet gave it a carrying capacity comparable with that of the Handley-Page. Its speed of 118 miles an hour at ground level far exceeded that of either the Caproni or Handley-Page, and it was evident that its ceiling would be higher than that of the Caproni, the estimated ceiling of the Martin being 18,000 feet. The machine never reached the state of actual quantity production, but several experimental models were built and tested. Being built around its engine it reflected clean-cut principles of design, and its performances in the air were truly remarkable for a machine of its type. The following table shows the results of the preliminary tests of the Martin bomber:
The total delivery of airplanes to the United States during the period of the war was 16,952. These came from the following sources: United States contractors, 11,754; France, 4,881; England, 258; Italy, 59.
Estimates of aircraft strength on the front were always uncertain, due to variations in the estimates of the number of planes in a squadron. The standing of the United States in aeroplanes at the
These figures represent fighting planes equipped ready for service, but do not include replacement machines at the front or in depots or training machines in France.
The actual strength of the central powers in the air is at this time not definitely known to us. Such figures as we have are viewed with suspicion because of the two methods of observation in reporting an enemy squadron. This may be 24 planes to a squadron, that number representing the planes in active service in the air. But each squadron had a complement of replacement planes equalling the number of active planes, so that the squadron could be listed with 48 planes. However, as some indication of the relative air strengths of the central powers we have a report from the chief of the Air Service of the American Expeditionary Forces showing that on July 30, 1918, Germany had 2,592 planes on the front and Austria 717. THE GLENN MARTIN BOMBER. The gross weight of this machine is 9,663 pounds. It can be equipped with five Lewis machine guns. Its ground speed is 113 miles an hour and its service ceiling is 12,800 feet. It climbs to a height of 6,500 feet in 10 minutes 45 seconds and to 10,000 feet in 21 minutes 20 seconds. THE CAPRONI, EQUIPPED WITH THREE LIBERTY 12-CYLINDER ENGINES. The Liberty engine was America's distinctive contribution to the war in the air, and her chief one. The engine was developed in those first chaotic weeks of preparation of 1917, when our knowledge of planes, instruments, and armament as then known in Europe was still a thing of the future. The manufacture of engines for any aeronautical purpose was one which we might approach with confidence. We possessed in the United States motor engineering talent at least as great as any in Europe, while in facilities for manufacture—in plants which had built our millions of automobile engines—no other part of the world could compare with us. Therefore, while waiting word from Europe as to the best type of wings, fuselages, instruments and the like, we went ahead to produce for ourselves a new, typically American engine which would uphold the prestige of America in actual battle. Many Americans have doubtless wondered why we built our own engine instead of adopting one or more of the highly developed European engines then at hand; and no doubt our course in this vital matter has sometimes been set down to mere pride in our ability and to an unwillingness to follow the lead of other nations in a science in which we ourselves were preeminent—the science of building light internal-combustion engines. But national pride, aside from giving us confidence that our efforts in this direction would be successful, had little other weight in the decision. There were other reasons, and paramount ones, reasons leading directly from the necessity for the United States to arrive at her maximum aerial effort in a minimum of time, that irresistibly compelled the aircraft production organization to design a standard American engine. Let us examine some of these considerations. If there was anything to be observed from this side of the Atlantic with respect to the tendencies of aircraft evolution in Europe it was that the horsepowers of the engines were continually increasing, these expansions coming almost from month to month as newer and newer types and sizes of engines were brought out by the European inventors. It was evident to us that there was not a single foreign engine then in use on the western front that was likely to survive the test of time. Each might be expected to have its brief day of supremacy, only to be superseded by something more modern and more powerful. Yet time was an element to which in this country we must give grave consideration. To produce in quantities such as we were capable of producing would ordinarily require a year of maximum industrial effort to equip our manufacturing plants with the machines, tools, and skilled workmen necessary for the production of parts. The finished articles would under normal circumstances begin coming in quantity during the second year of our program. It would have been fatal to "tool up" our plants for the manufacture of equipment that would be out of date by the time we began producing it a year later. The obvious course for the United States to adopt, not only with engines but with all sorts of aeronautical equipment, was to come into the manufacturing competition not abreast with European progress but several strides ahead of it, so that when we appeared on the field it would be with equipment a little in advance in type and efficiency of anything the rest of the world had to offer. This factor of time was a strong element in the decision to produce a standard American engine, since with the possible exception of the Rolls-Royce there was no engine in Europe of sufficient horsepower and proved reliability to guarantee that it would retain its serviceability for the necessary two years upon which we must reckon. There was no other course that we could safely adopt. But there were other conditions that influenced our conclusion. We believed that we could design and produce an engine much more quickly and with much better results than we could copy and produce any approved foreign model. This proved to be true in actual experience. Along with the production of Liberty engines we went into the quantity manufacture of a number of European engines in this country; and the experience of our engineers and factory executives in this work was anything but pleasant. Among others we produced in American factories the Gnome, the Hispano-Suiza, Le Rhone, and the Bugatti engines. Now European manufacture of mechanical appliances differs from ours largely in the degree to which the human equation is allowed to enter the shop. In continental practice much of the metallurgical specifications and also of the details of mechanical measurements, limits of requisite accuracy, variations which can be allowed, etc., are not put on paper in detail for the guidance of operators, but are confided to the recollections of the individual workmen. A machine comes in its parts to the assembly room of a foreign factory, and after that it is subject to adjustments on the part of the skilled workmen before its operation is successful. It must be tinkered with before it will go, so to speak. Nothing of the sort is known in an American factory. When standard parts come together for assembly the calibrations must have been so exact that the machine will Thirteen months were required to adapt the Hispano-Suiza 150-horsepower engine to our factory methods and to get the first engine from production tools, while eight months were similarly spent in producing the Le Rhone 80-horsepower engines. Both of these engines had been in production in European factories for a long time, and we had the advantage of all the assistance which the foreign manufacturers could give us. These experiences merely confirmed the opinions of American manufacturers that the preparations for the production of any aviation engine of foreign design—if any such suitable and adequate engine could be found—would require at least as much time as to design and tool up for the production of an American engine. When to this was added the necessity of waiting for several weeks or months for a decision on the part of our aviation authorities, either in the United States or in Europe, as to which of the many types of engines then in use by the allies should be put into production here, procuring and shipping to this country suitable samples, drawings, and specifications, negotiating with foreign owners for rights to manufacture, etc., there was but one answer to be made on this score, and that was to design and build an all-American engine. Another factor in the decision was that of our distance from France, a fact making it necessary for us to simplify as much as possible the problem of furnishing repair parts. At the time we entered the war the British air service was using or developing 37 different makes of engines, while France had 46. Should we be lured into any such situation it might have disastrous results, if only because of the difficulties of ocean transportation. Germany was practically concentrating upon not more than 8 engines. The obvious thing for us to do was to produce as few types of engines as possible, thus making simpler the problem of manufacturing repair parts and shipping them to the front. With these considerations in mind, the Equipment Division of the Signal Corps in May of 1917 determined to go ahead with the design and production of a standard engine for the fighting forces of the aviation branch of the Army. In the engineering field two men stood out who combined in themselves experience in designing internal-combustion engines which approached nearest to combat engines, with experience also in large quantity production. J. G. Vincent, with the engineering staff of the Packard Motor Car Co., had for approximately two years been engaged in research work, developing several types of 12-cylinder aviation engines of approxi E. J. Hall, of the Hall-Scott Motor Car Co., for eight years had been developing and latterly producing several types of aeronautical engines, which he had delivered into the service of several foreign governments, including Russia, Norway, China, Japan, Australia, Canada, and England. He had also completed and tested a 12-cylinder engine of 300 horsepower, which, however, was of too great weight per horsepower to be suitable in its form at that time for military purposes. He had thus acquired a large experience and fund of information covering the proper areas and materials for engine parts, and proper methods of tests to be applied to such engines, and in addition he had general experience in quantity production. All of this information and experience was of invaluable assistance not only in designing the new engine, but in determining its essential metallurgical and manufacturing specifications. These two men were thus qualified in talent and in practice to lay down on paper the lines and dimensions of the proposed engine, an engine that would meet the Army's requirements and still be readily capable of prompt quantity production. They had in their hands the power to draw freely upon the past experience and achievement of practically the entire world for any features they might decide to install in the model power plant to be produced. And this applied not only to the patented features of American motors, but also of foreign engines; for each man had exhaustively studied the leading European engines, including the Mercedes upon which Germany largely pinned her faith up to the end of the war. With respect to American motor patents, an interesting situation had arisen in the automobile industry. The leading producers of motor cars were in an association which had adopted an arrangement known as the cross-licensing agreement. Under this agreement all patents taken out by the various producers (with a few exceptions) were thrown into a pool upon which any producer at will was permitted to draw without payment of royalties. A similar arrangement was adopted with respect to the Liberty engine, except that the Government pledged itself to pay an agreed royalty for the use of patents. Thus the engineers designing the engine might reach out and take what they pleased regardless of patent rights. The result was likely to be a composite type embracing The ideal aviation engine should produce a maximum of power with a minimum of weight; it must run at its maximum power during a large proportion of its operating time, a thing that an automobile motor seldom, if ever, does for more than a few minutes at a time; and it should consume oil and fuel economically to conserve space and weight on the airplane. Such was the problem, the design of an engine to meet these requirements, that confronted these two engineers when they were called to Washington and asked to undertake the work. There have been so many versions of the story of how the Liberty engine was designed and produced in its experimental models that it is fitting that the exact history of those memorable weeks should be set down here. The engine was put on paper in the rooms occupied by Col. E. A. Deeds at the Willard Hotel in Washington. Col. Deeds had been the man of broad vision who, by taking into consideration the elements of the problems enumerated above, determined that America could best make her contribution to the aviation program by producing her typically own engine. He had proposed the plan to his associate, Col. S. D. Waldon, who had thereupon studied the matter and agreed entirely with the plan. The two officers persuaded Messrs. Hall and Vincent to forego further efforts on their individual developments and to devote their combined skill and experience to the creation of an all-American engine. The project was further taken up with the European authorities in Washington, and it was supported unanimously. In these conferences it was decided to design two lines of combat engines. Each should have a cylinder diameter of 5 inches and a piston stroke 7 inches long; but one type should have 8 cylinders and the other 12. The 8-cylinder engine should develop 225 horsepower, as all the experts believed then, in May, 1917, that such a motor would anticipate the power requirements as of the spring of 1918, while the 12-cylinder engine should develop 330 horsepower, as it was believed that this would be the equal of any other engine developed through 1919 and 1920. Every foreign representative in Washington with aeronautical experience agreed that the 8-cylinder 225-horsepower engine would be the peer of anything in use in the spring of 1918; yet, so rapidly was aviation history moving that inside of 90 days it became equally clear that it was the 12-cylinder engine of 330 horsepower, and not the 8-cylinder engine, upon which we should concentrate for the spring of 1918. With these considerations in mind Messrs. Hall and Vincent set to work to lay out the designs on paper. With them were Col. Deeds On May 29, 1917, Messrs. Vincent and Hall set to work. Within two or three days they had outlined the important characteristics of the engine sufficiently to secure—on June 4—the approval of the Aircraft Production Board and of the Joint Army and Navy Technical Board to build five experimental models each of the 8 cylinder and the 12 cylinder sizes. The detail and manufacturing drawings of the two engines were made partly by the staff of the Packard Motor Car Co., under Mr. O. E. Hunt, and partly by an organization recruited from various automobile factories and put to work under Mr. Vincent at the Bureau of Standards at Washington. Due credit must here be given to Dr. S. W. Stratton, the director of that important governmental scientific bureau. The Liberty engine pioneers woke him up at midnight and told him of their needs. He promptly tendered all the facilities of the Bureau of Standards, turning over to the work an entire building for use the following morning. Thereafter Dr. Stratton gave the closest cooperation of himself and his assistants to the work. While the detail drawings were being made, the parts for the 10 engines were at once started through the tool rooms and experimental shops of various motor car companies. This work centered in the plant of the Packard Co., which gave to it its entire energy and wonderful faculties. Every feature in the design of these engines was based on thoroughly proven practice of the past. That the engine was a composite is shown by the origin of its various parts: Cylinders: The Liberty engine derived its type of cylinders from the German Mercedes, the English Rolls-Royce, the French Lorraine-Dietrich, and others produced both before and during the war. The cylinders were steel inner shells surrounded by pressed-steel water jackets. The Packard Co. had developed a practical production method of welding together the several parts of a steel cylinder. Cam shafts and valve mechanism above cylinder heads: The design of these was based on the general arrangement of the Mercedes and Rolls-Royce, and improved by the Packard Motor Car Co. for automatic lubrication without wasting oil. Cam-shaft drive: The general type as used on the Hall-Scott, Mercedes, Hispano-Suiza, Rolls-Royce, Renault, Fiat, and others. WELDING JACKET ON CYLINDER FOR LIBERTY ENGINE. CADILLAC MOTOR CAR CO., DETROIT, MICH. DRILLING CYLINDER FLANGES WITH MULTIPLE DRILL. PACKARD MOTOR CAR CO. MACHINING THE CONNECTING RODS FOR THE LIBERTY ENGINE. CADILLAC MOTOR CO. GAUGING VALVES AND PISTONS FOR THE LIBERTY ENGINE. LINCOLN MOTOR CO. Angle between cylinders: In the Liberty the included angle between the cylinders is 45°. This angle was adopted to save head resistance, to give greater strength to the crank case, and to reduce periodic vibration. This decision was based on the experience of the Renault and Packard engines. Electric generator and ignition: The Delco system was adopted, but specially designed for the Liberty to provide a reliable double ignition. Pistons: The die-cast aluminum-alloy pistons of the Liberty were based on development work by the Hall-Scott Co. under service conditions. Connecting rods: These were of the forked or straddle type as used on the DeDion and Cadillac automobile motors and also on the Hispano-Suiza and other aviation engines. Crank shaft: A design of standard practice, every crank pin operating between two main bearings, as in the Mercedes, Rolls-Royce, Hall-Scott, Curtiss, and Renault. Crank case: A box section carrying the shaft in bearings clamped between the top and bottom halves by means of long through bolts, as in the Mercedes and Hispano-Suiza. Lubrication: The system of lubrication was changed, this being the only change of design made in the Liberty after it was first put down on paper. The original system combined the features of a dry crank case, such as in the Rolls-Royce, with pressure feed to the main crank-shaft bearings and scupper feed to the crank-pin bearings, as in the Hall-Scott and certain foreign engines. The system subsequently adopted added pressure-feed to the crank-pin bearings, as in the Rolls-Royce, Hispano-Suiza, and other engines. Propeller hub: Designed after the practice followed by such well-known engines as the Hispano-Suiza and Mercedes. Water pump: The conventional centrifugal type was adapted to the Liberty. Carburetor: The Zenith type was adapted to the engine. As the detailed and manufacturing drawings were completed in Washington and Detroit they were taken to various factories where the parts for the first engine were built. The General Aluminum & Brass Manufacturing Co., of Detroit, made the bronze-back, babbitt-lined bearings. The Cadillac Motor Car Co., of Detroit, made the connecting rods, the connecting-rod upper-end bushings, the connecting-rod bolts, and the rocker-arm assemblies. The L. O. Gordon Manufacturing Co., of Muskegon, Mich., made the cam shafts. The Park Drop Forge Co., of Cleveland, made the crank-shaft forgings. These forgings, completely heat treated, were turned out The Packard Motor Car Co. machined the crank shafts and all parts not furnished or finished elsewhere. The Hall-Scott Motor Car Co., of Berkeley, Calif., made all the bevel gears. The Hess-Bright Manufacturing Co., of Philadelphia, made the ball bearings. The Burd High-Compression Ring Co., of Rockford, Ill., made the piston rings. The Aluminum Castings Co., of Cleveland, made the die-cast alloy pistons and machined them up to grinding. The Rich Tool Co., of Chicago, made the valves. The Gibson Co., of Muskegon, Mich., made the springs. The Packard Co. made all the patterns for the aluminum castings, which were produced by the General Aluminum & Brass Manufacturing Co., of Detroit. The Packard Motor Car Co. used many of its own dies in order to obtain suitable drop forgings speedily, and also made all necessary new dies not made elsewhere. As these various parts were turned out they were hurried to the tool room of the Packard Co., where the assembling of the model engines was in progress. Before the models were built, however, extraordinary precautions had been taken to insure that the mechanism would be as perfect as American engineering skill could make it. The plans as developed were submitted to H. M. Crane, the engineer of the Simplex Motor Car Co. and of the Wright-Martin Aircraft Corporation, who had made a special study of aviation engines in Europe, and who for upward of a year had been working on the production of the Hispano-Suiza 150-horsepower engine in this country. He looked the plans over, and so did David Fergusson, chief engineer of the Pierce-Arrow Motor Car Co. Many other of the best experts in the country in the production of internal-combustion motors constructively criticized the plans, these including such men as Henry M. Leland and George H. Layng, of the Cadillac Motor Car Co., and F. F. Beall and Edward Roberts, of the Packard Car Co. When the engineers were through, the practical production men were given their turn. The plane and engine builders examined the plans to make sure that each minute part was so designed as to make it most adaptable to quantity production. The scrutiny of the Liberty plans went back in the production scale even farther than this; for the actual builders of machine tools were called in to examine the specifications and to suggest modifications, if necessary, that would make the production of parts most feasible in machine tools either of existing types or of easiest manufacture. Thus scrutinized and criticized, the plans of the engine were the best from every point of view which American industrial genius could produce in the time which was available. It was due to this exhaustive preliminary study that no radical changes were ever made in the original design. The Liberty engine was not the materialization of magic nor the product of any single individual or company, but it was a well-considered and carefully prepared design based on large practical aviation-engine experience. On July 4, 1917, the first 8-cylinder liberty engine was delivered in Washington. This was less than six weeks after Messrs. Hall and Vincent drew the first line of their plans. The same procedure was even then being repeated in the case of the 12-cylinder engine. By the 25th day of August the model 12-cylinder liberty had successfully passed its 50-hour test. In this test its power ranged from 301 to 320 horsepower. As an achievement in speed in the development of a successful new engine this performance has never been equaled in the motor history of any country. No successful American automobile motor was ever put in production in anything under a year of trial and experimentation. We may well believe that in the third year of war the European aviation designers were working at top speed to improve the motive power of airplanes; yet in 1917 the British war cabinet report contains the following language: Experience shows that as a rule, from the date of the conception and design of an aero engine, to the delivery of the first engine in series by the manufacturer, more than a year elapses. But America designed and produced experimentally a good engine in six weeks and a great one in three months, and began delivering it in series in five months. This was due to the fact that we could employ our best engineering talent without stint, to the further fact that there were no restrictions upon our use of designs and patents proved successful by actual experience, and to the fact that the original engine design produced under such conditions stood every expert criticism and test that could be put upon it and emerged from the trial without substantial modification. As soon as the first Liberty models had passed their official tests plans were at once made to put them in manufacture. The members of the Aircraft Production Board chose for the chief of the engine production department Harold H. Emmons, an attorney and manufacturer of Detroit, Mich., who, as a lieutenant in the Naval Reserve Force, was just being called by the Navy Department into active service. The production of all aviation engines, for both Army and Navy, was in his hands throughout the rest of the war. He placed orders for 100,993 aviation engines of all types, which involved In August, 1917, it was intended to manufacture both engines, the 8-cylinder and the 12-cylinder, and an agreement was reached with the Ford Motor Co. of Detroit to produce 8-cylinder Liberty engines to the number of 10,000. But before this contract could be signed the increasing powers of the newest European air engines indicated to our commission abroad that we should concentrate our manufacturing efforts upon the 12 alone, that being the engine of a power then distinctly in advance in the rapid evolution of aviation engines. The engine production department, therefore, entered into contracts for the construction of 22,500 of the 12-cylinder Liberties, and the first of these contracts was signed in August, a few days after the endurance tests had demonstrated that the 12-cylinder engine was a success. Of this number of Liberty engines the Packard Motor Car Co. contracted to build 6,000; the Lincoln Motor Co., 6,000; the Ford Motor Co., 5,000; Nordyke & Marmon, 3,000; the General Motors Corporation (Buick and Cadillac plants), 2,000; while an additional contract of 500 engines was let to the Trego Motors Corporation. Early in the liberty engine project it became apparent that one of the great stumbling blocks to volume production would be the steel cylinder, if it were necessary to machine it out of a solid or partially pierced forging such as is used for shell making. This problem was laid before Henry Ford and the engineering organization of the Ford Motor Co., at Detroit, and they developed the unique method of making the cylinders out of steel tubing. One end of the tube was cut obliquely, heated, and in successive operations closed over and then expanded into the shape of the combustion chamber, with all bosses in place on the dome. The lower end was then heated and upset in a bulldozer until the holding-down flange had been extruded from the barrel at the right place. By this method a production of 2,000 rough cylinders a day was reached. The final forging was so near to the shape desired that millions of pounds of scrap were saved over other methods, to say nothing of an enormous amount of labor thus done away with. The development of this cylinder-making method was one of the important contributions to the quantity production of Liberty engines. EXPERIMENTAL WORK ON NEW IDEAS FOR LIBERTY ENGINE. TEST CYLINDERS FOR THE LIBERTY ENGINE AT THE PLANT OF NORDYKE & MARMON, INDIANAPOLIS, IND. CRANK-SHAFT DIE FOR THE LIBERTY ENGINE. BUICK ENGINE CO., DETROIT. ACCEPTED LIBERTY ENGINES BEING BOXED FOR SHIPMENT. It was evident that in the actual production of the Liberty engine there would continually arise practical questions of manufacturing policy that might entail modifications of the manufacturing methods, while our aviation authorities in Europe could be expected to advance suggestions from time to time that might need to be embodied in the mechanism. Consequently it was necessary to create a permanent development and standardization administration for the Liberty engine. Nor could this supervision be located in Washington, because of the extreme need for haste, but it must exist in the vicinity of the plants doing the manufacturing. For this reason the production of the Liberty engine was centered in the Detroit manufacturing district, since in this district was located the principal motor manufacturing plant capacity of the United States. James G. Heaslet, formerly general manager of the Studebaker Corporation and an engineer and manufacturer of wide experience, was installed as district manager. The problems incident to the inspection and production of the Liberty engine were placed in charge of a committee consisting of Maj. Heaslet (chairman); Lieut. Col. Hall, one of the designers of the engine; Henry M. Leland; C. Harold Wills, of the Ford Motor Co.; and Messrs F. F. Beall and Edward Roberts, of the Packard Motor Car Co. With them were also associated D. McCall White, the engineer of the Cadillac Motor Co., and Walter Chrysler, of the Buick Co. The creation of this committee virtually made a single manufacturing concern of the several, previously rival, motor companies engaged in producing the Liberty engine. To these meetings the experts without reservation brought the trade secrets and shop processes developed in their own establishments during the preceding years of competition. Such cooperation was without parallel in the history of American industry, and only a great emergency such as the war with Germany could have brought it about. But the circumstance aided wonderfully in the development and production of the Liberty engine. Moreover, the Government drew heavily upon the talent of these great manufacturing organizations for meeting the special problems presented by the necessity of filling in the briefest possible time the largest aviation engine order ever known. Short-cuts that these firms might have applied effectively to their own private advantage were devised for the Liberty engine and freely turned over to the Government. The Packard Co. gave a great share of its equipment and personnel to the development. The most conspicuous success in the science of quantity production in the world was the Ford Motor Co., which devoted its organization to the task of speeding up the output of Liberty engines. In addition to the unique and wonderfully efficient method of making rough engine cylinders out of steel tubing, the Ford organization also perfected for the Liberty a new method of producing more durable and satisfactory bearings. Messrs. H. M. and W. C. Leland, whose names were indissolubly linked with the Cadillac automobile, organized and erected the enormous plant of Balanced against these advantages brought by highly trained technical skill and unselfish cooperation were handicaps such as perhaps no other great American industrial venture had ever known. In the first place, an internal-combustion engine with cylinders of a 5-inch bore and pistons of a 7-inch stroke—the Liberty measurements—was larger than the automobile engines then in use in this country. This meant that while we apparently had an enormous plant—the combined American automobile factories—ready for the production of Liberty engines, actually the machinery in these plants was not large enough for the new work, so that new machinery therefore must be built to handle this particular work. In some cases machinery had to be designed anew for the special purpose. To produce every part of one Liberty engine something between 2,500 and 3,000 small jigs, tools, and fixtures are employed. For large outputs much of this equipment must be duplicated over and over again. To provide the whole joint workshop with this equipment was one of the unseen jobs incidental to the construction of Liberty engines—unseen by the general public, that is—yet it required the United States to commandeer the capacity of all available tool shops east of the Mississippi River and devote it to the production of jigs and tools for the Liberty engine factories. Then there was the question of mechanical skill in the factories. It soon developed that an automobile motor is a simple mechanism compared with an intricate aviation engine. The machinists in ordinary automobile plants did not have the skill to produce the Liberty engine parts successfully. Consequently it became necessary to educate thousands of mechanics, men and women alike, to do this new work. It was surprising to what extent unfriendly influence in the United States, much of it probably of a pro-German character, cut a figure in the situation. This was particularly true in the supply factories furnishing tools to the Liberty engine plants. Approximately 85 per cent of the tools first delivered for this work were found to be inaccurate and incorrect. These had to be remade before they could be used. Such tools as were delivered to the Liberty plants would mysteriously disappear, or vital equipment would be injured in unusual ways; in several instances cans of explosives were found in the coal at power plants; fire-extinguishing apparatus was discovered to be rendered useless by acts of depredation; and from numerous other evidences the builders of Liberty engines were aware that the enemy had his agents in their plants. Difficulty was also experienced in the production of metals for the new engines. The materials demanded were frequently of a much higher grade than the corresponding materials used in ordinary automobile motors. Here was another unseen phase of development which had to be worked out patiently by the producers of raw materials. LIBERTY ENGINE READY FOR TEST AT THE LINCOLN MOTOR CO., DETROIT, MICH. LIBERTY ENGINE ON INSTRUCTION STAND, WILBUR WRIGHT FIELD. VIEW SHOWING LIBERTY ENGINE WITH PROPELLER HUB ATTACHED. TEST SHED AND STAND WITH LIBERTY ENGINE MOUNTED WITH TEST PROPELLER. PACKARD MOTOR CAR CO. Difficulties in transportation during the winter of 1917-18 added their share to the perplexing problems of the engine builders, while at times the scarcity of coal threatened the complete shutdown of some of the plants. Under such obstacles the engine-production department forced the manufacture of the Liberty engine at a speed never before known in the automotive industry. In December, 1917, the Government received the first 22 Liberty engines of the 12-cylinder type, durable and dependable, a standardized, concrete product, only seven months after the Liberty engine existed merely as an idea in the brains of two engineers. These first engines developed a strength of approximately 330 horsepower, and this was true also of the first 300 Liberty engines delivered, these deliveries being completed in the early spring of 1918. When the Liberty engine was designed our aviation experts believed that 330 horsepower was so far in advance of the development of aero engines in Europe that we could safely go ahead with the production of this type on a quantity basis. But again we reckoned without an accurate prophetic knowledge of the course of engine development abroad. We were building the first 300 Liberty engines at 330 horsepower when our aviation reports informed us from overseas that an even higher horsepower would be desirable. Therefore our engineers "stepped up" the power of the Liberty 12-cylinder engine to 375 horsepower. Several hundred motors of this power were in process of completion when again our observers in France advised us that we could add another 25 horsepower to the Liberty, making it 400 horsepower in strength, and be sure of leading all of the combatant nations in size and power of aviation engines during 1918 and 1919. This last step, we were assured, was the final, definite one. But to anticipate possible extraordinary development of engines by other nations, our engineers went even further than the mark advised by our overseas observers and raised the power of the Liberty engine to something in excess of 400 horsepower. This enormous increase over the original power of the Liberty engine required changes in the construction, notably in increasing the strength of practically all of the working parts, including the crank shaft, the connecting rods, and the bearings. The change also resulted in making scrap iron of a large quantity of the jigs and special tools employed in making the lighter engines. A still further change had to come in the character of some of the steel used in some of the parts, and this went back to the smelting plants, where new and better methods of producing steel and aluminum for the Liberty engine had to be developed. Thus while there were no fundamental changes in the design of the engine, the increase of its power required a considerable readjustment in the engine plants. Yet so rapidly were these changes made that on the first anniversary of the day when the design of the Liberty engine was begun—May 29, 1918—the Signal Corps had received 1,243 Liberty engines. In this achievement motor history was written in this country as it had never been written before. From a popular standpoint it may seem that the Liberty engine was radically changed after its inception, but such an assertion is entirely unwarranted; for in the fundamental thing, the design, there was but one change made after the engine was laid down on paper in May, 1917, namely, in the oiling system. The original Liberty engine was partially fed with oil by the so-called scupper system, whereas this later was changed to a forced feed under compression. The scupper feed worked successfully, but the forced feed is foolproof and was therefore installed upon the advice of the preponderance of expert criticism. It is also true that in working out certain practical manufacturing processes some of the original measurements were altered. But this is a common experience in the manufacture of any internal-combustion engine, and alterations made for factory expediency are not regarded as design changes, nor are they important. The delivery of 22 motors in December of 1917 was followed by the completion of 40 in January, 1918. In February the delivery was 70. In March this jumped to 122; then a leap in April to 415; while in May deliveries amounted to 620. The quantity production of Liberties may be said to have started in June, 1918, one year after the engine's conception in Washington. In that month 1,102 motors of the most powerful type were delivered to the service. In July the figure was 1,589; in August, 2,297; in September, 2,362. Then in October came an enormous increase to the total of 3,878 Liberty engines. During the month before the armistice was signed the engine factories were producing 150 engines a day. In all, up to November 29, 1918, 15,572 Liberty engines were produced in the United States. In the disposal of them the American Navy received 3,742 for its seaplanes; the plants manufacturing airplanes in this country took 5,323 of them; 907 were sent to various aviation fields for training purposes; to the American Expeditionary Forces in France, in addition to the engines which went over installed in their planes, we sent 4,511 Liberty engines; while 1,089 went to the British, French, and Italian air services. Some of the earliest Liberties were sent to Europe. In January, 1918, we shipped 3 to our own forces in France. In March we sent 10 to the British, 6 to the French, and 5 to the Italians. By June 7 the English tests had convinced the British air minister that the Liberty engine was in the first line of high powered aviation engines and a most valuable contribution to the allied aviation program. The British air minister so cabled to Lord Reading, the British ambassador in Washington. Again on September 26 the British air ministry reported that in identical airplanes the Liberty engine performed at least as well as the Rolls-Royce engine. Birkight, who designed the Hispano-Suiza engine in France, declared that the Liberty engine was superior to any high-powered aviation engine then developed on the Continent of Europe. INSTALLING LIBERTY ENGINE IN THE LEPERE FUSELAGE AT PACKARD PLANT, SHOWING PROGRESSIVE ASSEMBLY.
A more concrete evidence of the esteem in which this American creation was held by the European expert lies in the size of the orders which the various allied Governments placed with the United States This increased demand for the engine had not been anticipated in our original plans, as we had no idea that the allied Governments would turn from their own highly developed engines to ask for Liberty engines in such quantities. The original program of 22,500 engines was only sufficient for our own Army and Navy requirements. As soon as the foreign Governments, however, came in with their demands we immediately increased the orders placed with all the existing Liberty engine builders, and in addition contracted to take the entire manufacturing facilities of the Willys-Overland Co. at its plants in Toledo and Elyria, Ohio, and Elmira, N. Y. We also engaged the entire capacity of the Olds Motor plant at Lansing, Mich. In addition we had subsequently contracted for the production of 8,000 of the 8-cylinder engines. Thus the number of engines which would have been delivered under contract, if peace had not cut short the production, would have been 56,100 engines of the 12-cylinder type and 8,000 of the 8's. The foreign Governments associated with us in the war against Germany showered their demands upon us for great numbers of the American engines, not only altogether because of the excellence of the Liberty, but because partially their plane production exceeded their output of engines. Mr. John D. Ryan, Director of Aircraft Production, verbally agreed to deliver to the French 1,500 Liberty engines by December 31, and further agreed to deliver motors to the French at the rate of 750 per month during the first six months of 1919. The British had already received 1,000 Liberty motors, and this order was increased with Mr. Ryan personally by several thousand additional engines to be delivered in the early part of 1919. When the armistice was signed the Liberty engine was being produced at a rate which promised to make it the dominant motive power of the war in the air before many months had passed. The engine was originally named the "United States Standard 12-cylinder Aviation Engine." In view of the service which it promised to render to the cause of civilization, Admiral D. W. Taylor, the chief construction officer of the Navy, suggested during the early part of the period of production that the original prosaic name be discarded and that the engine be rechristened the "Liberty." Under this name the engine has taken its place in the history of the war as one of the most efficient agencies which was developed and employed by this country. The production of the Liberty engine so captured popular attention that the public never fairly understood nor appreciated the extent of another production enterprise on the part of those providing motive power for our war airplanes. This was the supplementary manufacture of aero engines other than those which bore the proud appellation of "Liberty." Let the production figures speak for themselves. In those 19 months, starting with nothing, we turned out complete and ready for service 32,420 aero engines. Of these thousands of engines less than one-half—the exact figure being 15,572—were Liberty engines. The rest were Hispano-Suizas, Le Rhones, Gnomes, Curtisses, Hall-Scotts, and some others, a total of 16,848 in all—built largely for the training of our army of the air. This production would have been even more notable had the war continued, for at the date of the signing of the armistice the United States had contracted for the construction of 100,993 aircraft engines. Of these 64,100 were to be Liberty engines, so that the total plan of construction of engines other than the Liberty would have produced about 37,000 of them. The total cost of carrying through the combined engine project would have been in the neighborhood of $450,000,000. While at the outbreak of the war American knowledge of military aviation may have been meager, still it was evident from the start that we would be able to go ahead with certain phases of production on a huge scale without waiting for the precise knowledge of requirements that would come only from an exhaustive study of the subject in Europe. In the first place we knew that we must train our aviators. For this purpose there was at the start no particular need of the highly-developed machinery then in use on the western front. The first aircraft requirement of the early training program was for safe planes, regardless of their type, and motive power to drive them. Later on, when we were better prepared, would come the training that would afford our aviators experience with the fighting equipment. So at the start there was no reason why we should not proceed at once with the construction of such training machines as we knew how to build. An aviation program for war falls into these two divisions—the equipment required for training and that required for combat. While our organization, particularly through the Bolling commission which we had sent to Europe, was making a study of our combat requirements and while we were pushing forward the design and production of the Liberty engine, we forthwith developed on an ambitious scale the manufacture of training planes and engines in this country. The training of battle aviators, on the other hand, also separates into two parts, the elementary training and the advanced training. The elementary training merely teaches the cadet the new art of maintaining himself in the air. Later, when he has mastered the rudiments of mechanical flight, he goes into the advanced training, the training in his fighting plane, where he requires equipment more nearly of the type used at the front. For the elementary training we had some good native material to start with. The Curtiss Airplane Co. had been building training planes and engines both for the English and Canadian air authorities. This was evidently the most available American airplane for our first needs. The Curtiss plane was known as the "JN-4" and it was driven by a 90-horsepower engine called the Curtiss "OX." In the production of this equipment on the scale planned by the Signal Corps, the embarrassing feature, the choke point, was evidently to be the manufacture of the engine. The Curtiss plant at Buffalo for the manufacture of planes could be quickly expanded to meet the Government demands; but the Curtiss engine plant at Hammondsport, N. Y., could not develop the production of "OX" engines up to our needs and at the same time complete the orders which it was filling for the English and Canadian air services. Consequently, contracts were awarded to the Curtiss Co. for its capacity in the production of "OX" engines, and then the American aviation authorities came to an agreement with the Willys-Morrow plant at Elmira, N. Y., for an additional 5,000 of these motors. Ordinarily it would require from five to six months to equip a plant with the large machine tools and the smaller mechanical appliances necessary for such a contract as this. But the Willys-Morrow plant tooled up in three months and was ready to start on the "OX" manufacturing job. CURTISS ENGINE, MODEL OX-5. HALL-SCOTT ENGINES BEING INSTALLED IN AIRPLANE FUSELAGES. TWO VIEWS OF LE RHONE 80-HORSEPOWER ENGINE. This is one of the successful rotary engines. If speed in production was required at any point in the aviation development it was here in the manufacture of the elementary training planes and engines. Without training material, no matter how many aviation fields we set in order nor how many student aviators we enlisted, the movement of our flying forces toward the front could not even begin. And here entered an interesting engineering and executive problem that had to be worked out quickly by those in charge of our aircraft construction. If it were plotted on paper, the curve of requirements for aircraft training material would climb swiftly to its peak during the first six or eight months of the war and then decline with almost equal swiftness until it reached a low level. In other words, we must produce the great number of training machines in the shortest time possible in order to put our thousands of student aviators into the air at once over the training fields; but when this training equipment had been brought up to initial requirements, thereafter our needs in this direction could be met by only a small production, since the rate of wastage of such material is relatively low. Once our fields were fully equipped, the same apparatus could be used over and over again as the war went on, with little regard to the improvements of the type of battle planes, so that the ultimate manufacture need be large enough only to keep this equipment in condition. It soon became evident that the production of Curtiss planes and engines, even under the heavy contracts immediately placed, would not be sufficient to take care of our elementary training needs; and the aviation administration began looking around for other types of aircraft that would fit into our plans. The experts in all branches of war flying which the principal allied nations had sent to the United States, warned us against the temptation to adopt many types of material in order to secure a quick early production. If the training equipment were not closely standardized in types, it would result in confusion and delay, both in training the aviator to fly and in preparing him for actual combat. Such had been the experience in Europe; and we were now given the benefit of this experience, so that we might avoid the mistakes which others had made. We were advised to adopt a single type of equipment for each class of training; but if that were not consistent with the demands for speed in getting our service in the air, then at the most we should not have more than two types either of planes or engines. In the elementary training program it was evident that we could not equip ourselves with a single type of plane, except at considerable expense in time. Consequently we went ahead to develop another. We found a training airplane being produced by the Standard Aero Corporation and known as the "Standard-J." The company had been developing this machine for approximately a year, and its plant could be expanded readily to meet a large contract. For the engine to drive this plane we adopted the Hall-Scott "A7A." This was a four-cylinder engine. It had the fault of vibration common to any four-cylinder engine, but it was regarded otherwise by experts as a rugged and dependable piece of machinery. The Hall-Scott Co. was equipped to produce this motor on an extensive scale, since at the time this concern was probably the largest manufacturer of The Government placed contracts with the Hall-Scott Co. for 1,250 engines, its capacity. But, since a large additional number would be required, a supplementary contract for 1,000 A7A's was given to the Nordyke & Marmon Co. The Hall-Scott Co. cooperated with this latter concern by furnishing complete drawings, tools, and other production necessities. When it came to the advanced training for our aviators, more highly developed mechanical equipment was required. There must be two sorts of this equipment. The advanced student must become acquainted with rotary engines such as were used by the French and others to drive the small, speedy chassÉ planes, while he must also come to be familiar with the operation of fixed cylinder engines, possessing upwards of 100 horsepower. These latter were the engines in commonest use on observation and bombing planes. For each type, the rotary and fixed, we were permitted by our policy to have two sorts of engines in order to get into production as quickly as possible, but not more than two. Here again we had to survey the field of engine manufacture and select closely, at the same time making in point of speed approximately as good a showing as if we had adopted every engine with claims for our consideration and had told manufacturers of them to produce as many as they could. In this case of rotary engines, our aviation representatives in Europe advised the production here of Gnome and Le Rhone motors. There were two models of the Gnome engine, one developing 110 horsepower and the other 150. The Le Rhone engine produced 80 horsepower. The Bolling commission had recommended that the Gnome 150 be used in some of our combat planes. In the spring of 1917 we were producing a few Gnome 110 horsepower engines in this country. The General Vehicle Co. at some time previously had taken a foreign order for these engines. But neither the Gnome 150 nor the Le Rhone 80 had been built in the United States, both of these having been developed and used exclusively in France. The first recommendations from our observers in France advised us to produce 5,000 of the more powerful Gnome 150's and 2,500 Le Rhone 80's. The production of Gnome engines in this country forms a good illustration of the manner in which aircraft requirements at the So with the production of the heavy 150-horsepower Gnome engine. Our European advisors were first of the opinion that we should go heavily into this production. Consequently the equipment end of the Signal Corps projected a program of 5,000 of the large Gnome engines. Such a contract was entirely beyond the capacity of the General Vehicle Co., which had been building the lighter Gnomes. So the Government entered into negotiations with the General Motors Co. to assume the greater burden of this undertaking. Under the pilotage of the aircraft authorities, an agreement was reached for the industrial combination of the General Motors Co. and the General Vehicle Co. The former concern brought its vast resources and numerous factories into the consolidation; while the latter furnished the only skilled knowledge and experience there was in the United States in the art of making rotary engines. This seemed to be a great step in our progress and an achievement in itself; but just as the undertaking of the construction of large Gnome engines was about to be started, events in Europe had caused our observers there to revise their first judgment, and we received cabled instructions recommending that we discontinue the development of the Gnome 150. The entire program for Gnome 150's was canceled, and thereafter the General Vehicle Co., with its relatively small capacity, was called upon to produce as many of the small Gnome 110's as it could. As a matter of record the production of these engines amounted to 280 in number. The Signal Corps found it difficult to induce manufacturers in this country to undertake the construction of foreign designed engines at all. The plans and specifications of mechanical appliances furnished by foreign engineers and manufacturers are so different from ours that trouble is invariably experienced in attempts to use them here. Successful concerns in this country naturally hesitated to pick up contracts on which they might fail and thus tarnish their reputations. Our advisors in Europe were insistent that we should produce Le Rhone engines in quantity in the United States, yet it was hard to find any manufacturing concern willing to undertake such a develop It was only after strenuous efforts on our part that the Union Switch & Signal Co., of Swissvale, Pa., a member of the Westinghouse chain of factories, was induced to take up the Le Rhone contract. This project called for the production of 2,500 rotary Le Rhones of 80 horsepower each. Let us see how the manufacturers took this totally unfamiliar machine and went about it to reproduce it in this country. One might think that it would be necessary only to take the French drawings, change the metric system measurements to our own scale of feet and inches, and proceed to turn out the mechanism. But it was not so simple as that. We did receive the drawings, the specifications, the metallurgical instructions and the like, but these we found to be unreliable and unsatisfactory from our point of view. For instance, according to the French instructions the metallurgical requirements for the engine crank-shaft called for mild steel. This was obviously incorrect; and if an error had crept into this part of the plans there was no telling how faulty the rest of them might be. So from the metallurgical standpoint alone this became a laboratory job of analysis and investigation. A sample engine had been sent to us from France. Every piece of metal in this engine was examined by the chemists to determine its proper constituents, and from this original investigation new specifications were made for the steel producers. The drawings of the engine were quite unsatisfactory from the point of view of American mechanics. They were found to be incorrect, and there were not enough of them. Consequently this required another study on the part of engineers and a new set of drawings to be made up. All of this fundamental work monopolized the time of a large force of draughtsmen and engineers for several months, working under the direction of E. J. Hall and Frank M. Hawley. The engine could not be successfully built without this preliminary study, yet this is a part of manufacture of which the uninitiated have little knowledge. THREE VIEWS OF BUGATTI 410-HORSEPOWER ENGINE. THREE VIEWS OF THE HISPANO-SUIZA ENGINE. The production of the Le Rhone engine might have been materially delayed by these difficulties, except for the organizing ability of the executives handling the contract. While the metallurgists were specifying the steel of the engine parts and the engineers were drafting correct plans, the factory officials, with the assistance of the engine production division of the Air Service, were procuring machinery and tooling up the plant for the forthcoming effort. By the time this equipment was installed the plans were ready, the steel mills were producing the proper qualities of metal, and all was ready for the effort. The Gnome-Le Rhone factories in France sent one of their best engineers, M. Georges Guillot, and he assisted in the work at the Union Switch & Signal Co. So rapidly was the whole development carried out that the first American Le Rhones were delivered to the Government in May, 1918, considerably less than a year after the project was assumed by the Union Switch & Signal Co., which concern had not received the plans of the engine until September, 1917. By the time the armistice was signed the company had delivered 1,057 Le Rhone engines. Subsequent contracts had increased the original order to 3,900 Le Rhones, all of which would have been delivered before the summer of 1919, had the coming of peace not terminated the manufacture. Although France is the home of the rotary aviation engine, M. Guillot has certified to the Aircraft Board that these American Le Rhones were the best rotary engines ever built. When it came to the selection of fixed cylinder engines for our advanced training program, all of the indications pointed to a single one, the Hispano-Suiza engine of 150 horsepower. This was a tried and true engine of the war, tested by a wealth of experience and found dependable. France had used the engine extensively in both its training and combat planes. In 1916 it had been brought to the United States for production for the allies, and when we entered the war the Wright-Martin Aircraft Corporation was producing Hispano-Suizas in small quantity. By the early summer of 1917, however, the motor had fallen behind in the development of combat engines because of the increasing horsepowers demanded by the fighting aces on the front, but it was still a desirable training engine and could, if necessary, be used to a limited extent in planes at the front. The plane adopted by the American aircraft authorities for this type of advanced training was known as the Curtiss "JN 4H." It was readily adapted for the use of the Hispano-Suiza 150-horsepower engine. Contracts for several thousand of these engines were placed with the Wright-Martin Aircraft Corporation, and up to the signing of the armistice 3,435 engines were delivered. Before we could start the production of this engine it was necessary for the Government to arrange with the Hispano-Suiza Co. for the American rights to build it, this arrangement including the payment of royalties. Incidentally it is interesting to note that royalty was the chief beneficiary of the royalties paid by the American Government, King Alfonso of Spain being the heaviest stockholder of the Hispano-Suiza Co. Although our policy permitted us to produce a second training engine of the fixed cylinder type, no engine other than the Hispano-Suiza was taken up by us. A number presented their claims for To sum it up, our training program was built around the above named engines—the Curtiss "OX" and the Hall-Scott "A7A" for the elementary training machines; the Gnome and Le Rhone, for the rotary engine types of planes in the advanced training; and the Hispano-Suiza 150-horsepower, for the advanced training in fixed-cylinder-engine machines. Between the dates of September 1, 1917, and December 19, 1918, we sent to 27 fields 13,250 cadets and 9,075 students for advanced training. They flew a total of 888,405 hours and suffered 304 fatalities, or an average of 1 fatality for every 2,922.38 flying hours. At one field the training fliers were in the air 19,484 hours before there was 1 fatality; another field increased this record to 20,269 hours; while a third made the extraordinary record of 1 casualty in 30,982 flying hours. Although we do not possess the actual statistics, the best unofficial figures show that the British averaged 1 fatality for each 1,000 flying hours at their training camps, the French 1 for each 900 flying hours, while the Italian training killed 1 student for each 700 flying hours. These figures are significant, although varying conditions in the types of training programs may account to some extent for the wide differences in numbers of casualties at American as compared with allied training camps. But while we were producing engines for the training airplanes, both elementary and advanced, we were not staking our whole combat program on the Liberty engine alone, although we expected that engine to be our main reliance in our battle machines. Our organization, both at home and abroad, was on the alert continually for other engines that might be produced in Europe or the United States and which would be so far in advance of anything in use by the air fighters in Europe in 1917 as to justify our production of them on a considerable scale. One of these motors which seemed to promise great results for the future was the Rolls-Royce, which had even then, in 1917, taken its place at the head of the British airplane engines. Considerable difficulty was experienced in reaching a satisfactory arrangement with the Rolls-Royce Co. We expected to duplicate this engine at the plant of the Pierce-Arrow Motor Car Co., at Buffalo, N. Y., but the British concern objected to this arrangement on the ground that the Pierce-Arrow people were commercial competitors. It was several months before we could agree on a factory and arrive at a contract satisfactory to both sides. Meanwhile the Liberty engine had scored its great success, and the expected enormous production of Liberties tended to cool the enthusiasm of our aircraft authorities for the Rolls-Royce, as it was evident that the Liberty itself would be as serviceable and as advanced in type as the British product. The Rolls-Royce Co. wished to manufacture here its "190," an engine developing from 250 to 270 horsepower; and for this effort it was prepared to send to the United States at once a complete set of jigs, gauges, and all other necessary tooling of a Rolls-Royce plant. With this equipment ready at hand the company expected to produce about 500 American-built Rolls-Royce engines before the 1st of July, 1918. But so rapidly was the evolution of aircraft engines going ahead that even during the time of these negotiations it became evident that something more than 250 horsepower would soon be needed in the fighting planes on the Western front. We therefore abandoned the Rolls-Royce model 190 and started negotiations for the 270-horsepower engine, the latest and most powerful one produced by the Rolls-Royce Co. But for this engine the British concern could not furnish the tooling, which would have to be made new in this country, and this would reduce the schedule of deliveries. As a result no American-built Rolls-Royce engine was ever made. Another disappointing experience in attempting to produce a foreign designed motor in this country was the project to bring the manufacture of Bugatti engines to the United States. When our European aircraft commission arrived in France, the first experimental Bugatti engine had just made its appearance. It was apparently a long step in advance of any other motor that had been produced. This French mechanism was a geared 16-cylinder engine. It weighed approximately 1,100 pounds and was expected to develop 510 horsepower. It seemed to be the motor to supplement our own Liberty engine construction. Although heavier than a Liberty, it was much more powerful. The first Bugatti engine built in France was purchased by the Bolling commission and hurried to the United States with the urgent recommendation that we put it into production immediately and push its manufacture as energetically as we were pushing that of the Liberty engine. The Signal Corps acted immediately upon this advice and prepared to proceed with the Bugatti on a scale that promised to make its development as spectacular as that of the Liberty. The Duesenberg Motor Corporation, of Elizabeth, N. J., was even then tooling up for the production of Liberty engines. We took this concern from its Liberty work and directed it to assume leadership in the production of Bugattis. The Liberty engine construction had been centered in the Detroit district. We now prepared to establish a new aviation engine district in the East, associating in it such concerns as the Fiat Plant at Schenectady, N. Y., the Herschell-Spillman Co., of North Tonawanda, N. Y., and several others. For a time the expectation for the Bugatti production ran almost as high as the enthusiasm for the Liberty engine, but the whole undertaking ended virtually in failure, a failure again due to the tremendous difficulty in adapting foreign engineering plans to American factory production. This was the story of it. In due time the sample Bugatti engine arrived, and with it were several French engineers and expert mechanics. But, once set up, the Bugatti motor would not function, nor was it in condition to run; for, as we discovered, during its test in France a soldier had been struck by its flying propeller. His body had been thrown twice to the roof of the testing shed, and the shocks had bent the engine's crank shaft. Then, too, we learned for the first time that the design and development of this engine had not been carried through to completion and that a great deal of work would be required before the device could be put into manufacture. The tests in France had developed that such a fundamental feature as the oiling system needed complete readjustment, and this was only indicative of the amount of work yet to be done on the engineering side of the production. We did our best with this engine; but to redesign it and develop it so that it could pass the severe 50-hour test demanded by our Joint Army and Navy Technical Board was the work of months, and after that the tooling up of plants had to be accomplished. The American Bugatti was just getting into production when the armistice was signed, a total of only 11 having been delivered. As we have seen, we were already building several hundred Hispano-Suiza 150-horsepower engines for our training planes. Soon after the arrival of our aircraft commission in France we were advised to go into the additional manufacture of the latest Hispano-Suiza geared engine of 220-horsepower. Consequently the Washington office at once arranged with the Wright-Martin Aircraft Corporation, which was building the smaller Hispano-Suizas, to undertake the production of this newer model also. The preparations for this manufacture had gone on in the Wright-Martin plant for a considerable period of time when further advice from Europe informed us Along in the summer of 1918 the Hispano-Suiza designers in Europe brought out a 300-horsepower engine. By this date the development of military flying had made it apparent that engines of such great horsepower could be used advantageously on the smaller planes. However, the engine plants of the allied countries were already taxed to their capacities by their existing contracts, and the demands of these countries for high-powered engines could not be supplied unless we in America could increase our manufacturing facilities even further. In following out this ambition, we placed contracts for the production of 10,000 Hispano-Suiza 300-horsepower engines. Of these, 5,000 were to be built by the Wright-Martin Aircraft Corporation. To enable this company to fulfill the new contract we leased to it the plant owned by the Government in Long Island City which had formerly been owned by the General Vehicle Co. The other 5,000 of these engines were to be built by the Pierce-Arrow Motor Car Co. at Buffalo. We also contracted for the entire manufacturing facilities of the H. H. Franklin Co., of Syracuse, N. Y., to aid both the Wright-Martin Corporation and the Pierce-Arrow Co., in this contract. The first of these high-powered Hispano-Suiza engines were expected to be delivered in January, 1919, but this project, of course, was interrupted by the armistice. To summarize the complete engine program of the aviation development, the total contracts for engines provided for the delivery of 100,993 engines. These were divided as follows:
The delivery of aviation engines of all types to the United States Government, engines produced as part of our war program, were as follows, by months:
The production by types was as follows to November 29, 1918:
At the signing of the armistice the United States had produced about one-third of the engines projected in its complete aviation program. Of the output of training engines to November 29, 1918, the various airplane plants took 9,069 for installation in planes, 325 (all of these being Le Rhone rotaries) went to the American Expeditionary Forces in France, 515 (all of which were Hispano-Suizas) were taken by the Navy, a single A7A model was sent to one of the allied countries, while 6,376 engines were sent directly to the training fields. Of the combat engines produced to November 29, 1918 (which classification includes all of the Liberties, the two more powerful types of the Hispano-Suiza, and the Bugatti engine), 5,327 went to the various airplane plants for installation in planes, 5,030 of them were sent directly to the American Expeditionary Forces, 3,746 were turned over to the Navy, 1,090 went to the several allied nations, and 941 were taken by the training fields. The shipment of aviation engines to Europe, however, does not imply the immediate use of them by our airplane squadrons at the front. In this report shipment to the American Expeditionary Forces means the shipment of engines from the American factories producing them. As a matter of fact several months usually elapsed from the dispatch of an engine from an American shop until it actually reached the Air Service in France, and even then another month might be required to put the engine into actual service. As a result, of the 5,000 and more aviation engines sent to France by the American engine producers, outside of those installed in their planes, less than 3,000 are recorded in the annals of the American Expeditionary Forces as having been received by them up to the end of December, 1918, the missing 2,000 being in that period either somewhere in transit or in warehouses on the route to their destination. It is of interest to note what makes of foreign engines were used by our airmen in the war operations. An appended table shows the list of those received, their names, their rated powers, the numbers received month by month, and the totals. The records of the American Expeditionary Forces show that the squadrons in all received from all sources 4,715 aviation engines up to the end of the year 1918, but it should be borne in mind that this figure does not include more than 2,000 engines, principally Liberties, recorded on this side of the Atlantic as having been shipped to the Army abroad. Of the 4,715 engines noted as received, 2,710 were Liberties. None of the foreign engines used by our pilots even approached the Liberty in power. The nearest in power were a Renault and an Hispano-Suiza, both rated at 300 horsepower.
On one of the early days in the great war a Russian aviator, aloft in one of the primitive airplanes of that time, was engaged in locating the positions of the enemy when he chanced upon a German birdman engaged in a similar mission. In those ancient times—for they seem ancient to us now, although less than five years have elapsed—actual fighting in the air was unknown. The aviators had no equipment for battle; indeed, it was doubtful if the thought had occurred to either side to keep down the enemy's aircraft by the use of armed force borne upon wings. In the first months of aviation in the great war the fliers of both sides recognized a sort of noblesse oblige of the air, which, if it did not make for actual friendship or fraternizing between the rival air services, at least amounted to a respect for each other often evidenced by an innocuous waving of hands as hostile flying machines passed each other. But now the wounds of war had begun to smart; and when the Russian saw the German flier going unhindered upon a work that might bring death to thousands of soldiers in the Czar's army, a sudden rage filled his heart, and he determined to bring down his adversary, even at the cost of his own life. Maneuvering his craft, presently he was flying directly beneath the German and in the same direction and was but a short distance below his enemy's plane. Then, with a pull on his control lever, the Russian shot his machine sharply upward, hoping to upset the German and to escape himself. The result was that the machines collided, and both crashed to the ground. This was probably the first aerial combat of the war. It seems strange to us to-day that the highly complicated and standardized art of fighting with airplanes was developed entirely during the great war and, indeed, was only started after the war had been in progress for several months. Yet such was the case. At the beginning of the war there was no such thing as armament in aircraft, either of the offensive or defensive sort. It is true that a small amount of experimentation in this direction had occurred prior to the war and also in the early months of fighting, but it was not until the summer of 1915 that air fighting, as it is so well known to the entire world to-day, was begun. In this country we had successfully fired a machine gun from an airplane in 1912, while at the beginning of the war the French had a few heavy airplanes equipped to carry machine guns. Yet in August, 1915, Maj. Eric T. Bradley, of the United States Air Service, but then a flight sublieutenant in the Royal Flying Corps, frequently flew over the lines hunting for Germans; and his offensive armament consisted of a Lee-Enfield rifle or sometimes a 12-gauge double-barreled shotgun. The aviators in those pioneer days usually carried automatic pistols, but the danger to one side or the other from such weapons was slight, owing to the great difficulty of hitting an object moving as swiftly as an airplane travels. The earlier planes also packed a supply of trench grenades for dropping upon bodies of troops. Another pioneer offensive weapon for the airplane was the steel dart, which was dropped in quantities upon the enemy's trenches. Great numbers of these darts were manufactured in the United States for the allies, but the weapon proved to be so ineffective that it had but a brief existence. It is said that before the pilots carried any weapons at all the first war aviators used to shoot at each other with Very pistols, which projected Roman candle balls. The start of air fighting may be said to have come when the Lewis machine guns were brought out for use in the trenches. Presently these ground guns were taken into the planes and fired from the observers' shoulders. Then for the first time war flying began to be a hazardous occupation so far as the enemy's attentions were concerned. It was soon discovered that the machine gun was the most effective weapon of all for use on an airplane, because only with rapid firers could one hope to hunt successfully such swiftly moving prey as airplanes. It had become patent to the strategists that it was of supreme importance to keep the enemy's aircraft on the ground. Hence invention began adapting the machine gun to airplane use. The swiftest planes of all were those of the single-seater pursuit type. It was obviously impossible for the lone pilot of one of these to drop his controls and fire a machine gun from his shoulder. This necessitated a fixed gun that could be operated while the pilot maintained complete control of his machine, and such necessity was the mother of the invention known as the synchronizing gear. This ingenious contrivance, however, did not come at once. Most of the war planes were of the tractor type; that is, that they had the engine and propeller in front, this arrangement giving them better maneuvering and defensive powers in the air than those possessed by planes with the rear, pushing propellers. The first fixed machine gun was carried on the upper plane of the biplane so as to So the fixed gun was brought down into the fuselage and made to fire through the whirling propeller. At first the aviators took their chances of hitting the propeller blades, and sometimes the blades were armored at the point of fire, being sheathed in steel of a shape calculated to cause the bullets to glance off. This system was not satisfactory. Then, since a single bullet striking an unprotected propeller blade would often shatter it to fragments, attempts were made to wrap the butts of the blades in linen fabric to prevent this splintering, and this protection actually allowed several shots to pierce the propeller without breaking it. This was the state of affairs on both sides early in 1915. The French Nieuports had their fixed guns literally shooting through the propellers, the bullets perforating the blades, if they did not wreck them. As late as February, 1917, Maj. Bradley, who was by that time a flight commander in the British service, worked a Lewis gun over the Bulgarian lines with the plane propellers protected only by cloth wrappings. All of this makeshift operation of fixed machine guns was changed by the invention of the synchronizing device. This is an appliance for controlling the fire of the fixed gun so that the bullets miss the blades of the flying propeller and pass on in the infinitesimal spaces of time when the line of fire ahead of the gun is clear of obstruction. The term "synchronizing" is not accurate, since that word implies that the gun fires after each passage of a propeller blade across the trajectory. Such is not the truth. The propeller revolves much more rapidly than the gun fires. The device is also called an "interrupter," another inexact term, since the fire of the gun is not interrupted, but only caused at the proper moments. Technicians prefer the name "gun control" for this mechanism. Who first invented the synchronizer is a matter of dispute, but all observers agree that the Germans in the Fokker monoplanes of 1915 were the first to use it extensively. Not until some time after this did the allies generally install similar devices. Some have attributed the original invention to the famous French flier, Roland Garros. Two types of synchronizers were developed, one known as the hydraulic type and the other as the mechanical. In operation they are somewhat similar. In each case there is a cam mounted on the engine shaft so that each impulse of the piston actuates a plunger. In April, 1917, we knew practically nothing about the use or manufacture of aircraft guns. We had used airplanes at the Mexican border, but not one of them carried a machine gun. The Lewis gun, which is a flexible type of aircraft weapon pointed on a universal pivot by the observer in a two-place plane, was being manufactured by the Savage Arms Corporation for the British Government; but we had never made a gun of the fixed type in this country, nor did we know anything about the construction or manufacture of synchronizers. One special requirement of the aircraft machine gun is that it must be reliable in the extreme. It is bad enough to have a gun jam on the ground, but in the air it may be fatal, for little can be done there to repair the weapon. A jam leaves the gunner to the mercy of his adversary, so in the production of aircraft armament there must be not only special care in the manufacture of the guns, but the ammunition, too, must be as perfect as human accuracy can make it. The cartridges must be either hand-picked and specially selected from the run of service ammunition, or else manufactured slowly and expressly for the purpose, with minute gauging from start to finish of the process. Another requirement for the aircraft gun is that it must function perfectly in any position. On the ground a machine gun is fired essentially in a horizontal position, but the airman dives and leaps in his maneuvering and must be able to shoot at any instant. Aircraft guns are subject to extreme variations of temperature, and so they must be certain to function perfectly in the zero cold of the high altitudes, regardless of the contraction of their metal parts. Then, too, such guns must be able to fire at a much greater rate than those of the ground service. Five hundred shots per minute is regarded as sufficient for a ground gun, but aircraft guns have been brought up to a rate of fire as high as 950 to 1,000 shots per minute. The Browning aircraft gun, never used by us, but in process of development when the armistice was signed, had been speeded up to 1,300 shots per minute, with all shots synchronized to miss the blades of the propeller. The rate of fire in the air can not be made too swift. Suppose an airplane were flying past a long, stationary target, such as a billboard, at the relatively slow speed of 100 miles an hour. Assume on The Lewis gun, invented by Col. Lewis, of the United States Army, was the weapon most generally used by the allies as the flexible gun for their airplanes, operated on a universal mount which permitted it to be pointed in any direction. The Lewis aircraft gun was the ground gun modified principally by stripping it of the cooling radiator and by the addition of a gas check to reduce the recoil. The Lewis was fed by a drum magazine, a more desirable feed for flexible guns than any belt system. The German flexible gun, the Parabellum, had the unsatisfactory belt feed. The Vickers gun was the only successful weapon of the fixed type developed in the war before we became a belligerent. We were manufacturing Vickers guns in the United States prior to April, 1917; but when the Signal Corps faced the machine-gun problem, in September, 1917, it found that the Infantry branches of the Army had contracted for the entire Vickers production in this country. Accordingly, the equipment division of the Signal Corps, in the face of marked opposition, took up the development of the Marlin gun as an aircraft gun of the fixed type. This gun, however, proved to be extraordinarily successful and was regarded by our Flying Service and by the aviators of the allies to be the equal of the Vickers in efficiency. Because of this development, when there came the need of tank guns, in June, 1918, the Aircraft Board, which had succeeded the Signal Corps as the director of aerial activities, was able to supply 7,220 Marlin machine guns within two weeks for this purpose. The first order for Marlin guns was placed on September 25, 1917; and over 37,500 of them had been produced before December, 1918. The Marlin-Rockwell factory began producing 2,000 guns per month in January, 1918, and increased this rapidly until as many as 7,000 guns were built in one month. The Marlin gun shoots at the rate of 600 to 650 shots per minute and is fed by a belt of the disintegrating metal-link type. As to Lewis guns, which we adopted as our flexible weapon, more than 35,000 of them were delivered to the Air Service up to December, 1918. In February, 1918, the Savage Arms Corporation built 1,500 of them, increasing their monthly deliveries until in the In our De Haviland-4 planes we installed two Marlin fixed guns, each firing at the rate of 650 shots per minute, equipping the weapons with Constantinisco controls to give the plane a maximum fire of 1,300 shots per minute through the blades of a propeller whirling at a rate as high as 1,600 revolutions per minute. Four fixed guns have also been successfully fitted to one plane and timed so that none of the bullets struck the propeller blades. At the time the armistice was signed the rate of production of special aircraft ammunition, a classification including tracer bullets, incendiary bullets, and armor-piercing bullets, exceeded 10,000,000 rounds per month. The original estimate for the quantity of ammunition our Flying Service should have was later greatly increased because the squadrons at the front began installing as many as four guns on a single observation plane. Although different aviators had their own notions about the loading of ammunition belts, certain sequences in the use of the three types of special ammunition were usually observed. First usually came the tracer cartridge, which assists the gunner in directing his aim; then two or three armor-piercing cartridges, relied upon to injure the hostile engine or tap the gasoline tank; and finally one or two incendiary cartridges to ignite the enemy's gasoline as it escaped, sending him down in flames. Such a sequence would be repeated throughout the ammunition belt or magazine container. The belts for the fixed guns carry a maximum of 500 rounds of cartridges. The belt which we furnished to our fliers at the front was made of small metallic links fastened together by the cartridges themselves. As the gun was fired and the cartridges ejected, the links fell apart and cleared the machine through special chutes. The total production of such belting in this country amounted to 59,044,755 links. Although the links are extremely simple in design, the great accuracy required in their finish made production of them a difficult manufacturing undertaking. The production and inspection of each link involved over 36 separate operations. It actually cost more to inspect belt links than to manufacture them. We produced 12,621 British unit sights for airplane guns and sent 1,550 of them overseas. We also bought an adequate number of small electric heaters to keep the gun oil from congealing in the cold of high altitudes. A novel undertaking for our photographic manufacturers was the production of the so-called gun cameras which are used to train One of these gun cameras, invented by Thornton Pickard, of Altringham, England, imitated in design a Marlin aircraft machine gun; and in order to make a picture with it, the gunner must go through the same movements that he would employ in firing a Marlin gun. Thus, if the gun were pointed directly on the target, the target would appear squarely in the center of the picture taken; and this showed the gunner's accuracy as well as if he had fired cartridges from the actual weapon. These gun cameras were of two sorts. One type took a single picture each time the trigger was pulled. Those of the other sort took a number of pictures automatically at a speed approximately that of the firing of a machine gun. This latter type was much the same as a moving picture camera, the resulting film being a string of silhouettes of the target, each exposure showing whether the aim of the gunner was exact at the instant the picture was taken. In September, 1917, the Eastman Kodak Co. began the development of a camera gun of the "burst" or automatic moving-picture type. After our authorities had seen the model, the Navy ordered a number of them, while the Air Service placed increasing orders for these instruments until 1,057 had been produced and delivered to the Government by November, 1918. This camera was not used in the fixed airplane guns, but was designed to train the operators of the flexible Lewis gun. The camera exactly replaced the ammunition magazine on a Lewis gun. Of the single-shot gun cameras 150 were delivered during the hostilities. This design was obtained from Canada and duplicated here. The use of the so-called Bromotype paper in gun cameras was one of the interesting phases of this development. As everyone acquainted with photography knows, a picture is made ordinarily by exposing a sensitized plate or film, developing the latter to make a negative, then exposing sensitive print paper to the light that comes through the negative, thus reversing the lights and shadows and creating a positive in the exact semblance of the subject photographed. A concern in Cleveland, Ohio, the Positype Co., produced Bromotype paper which could be exposed directly in the camera, coming out of the developing process as a positive without the intervention of a film or plate negative. Bromotype paper is much more highly sensitized than ordinary print paper, so that it may be adequately exposed in an instantaneous, Under this system, of course, only one finished print of each exposure can be made; but the airplane gunners needed only one print to show their aim. Positype paper was thus admirably adapted for use in the airplane gun cameras; and because of its cheapness and the simplicity and rapidity of its use, it was rapidly supplanting film at the training camps in this country when the armistice was signed. AIRPLANE BOMBS.The American production of bombs to be dropped from airplanes was not started so soon as production in some of the other branches of ordnance development, due to numerous difficulties encountered in working up the design of this new matÉriel. Although aerial bombing was steadily increasing in effectiveness and magnitude when hostilities ended, yet this kind of fighting was a development that came relatively late in the war; and the lack of perfected standards at the time this country became a belligerent helped to impede our program. Some of the bombs first designed and put into production were later rejected by our forces in France, as they had become obsolete before being shipped overseas. We managed to manufacture a great quantity of unloaded bombs by the time the armistice was signed, enough, in fact, to provide for the Army's needs during another year of warfare. These had to be loaded with explosives before they were ready for use. We lacked adequate facilities for loading bombs with explosives, although these facilities were being provided rapidly when the war ended. The result was that the thousands of completed American bombs remained unloaded, while practically all the bombs used by our fliers in France were of foreign manufacture. Military science had had some small experience with aerial bombing prior to the great war. Italian aviators had dropped bombs of an ineffective sort during Italy's war in Africa. When Mexico was having a civil war in 1914 American air-sailors of fortune on one side or the other dropped bombs on troops from their planes. In the great war the first nation to attempt bombing on any systematic scale was Germany, who sent her Zeppelins over London and Paris early in the conflict and released bombs upon the heads of the helpless civilians. Yet this early and impressive effort was, in its difficulties, out of all proportion to the actual damage done to the city of London, largely due to the fact that Germany had not yet produced effective aerial bombs. The frightful scenes and noises of a bomb raid probably did more to reduce the morale in these early days than the destruction caused by the exploding missiles. It is an exceedingly difficult trick to drop a bomb from any considerable altitude and hit what you are aiming at. The speed of the airplane, its height above the ground, the shape of the bomb itself, and the currents of air acting on the falling missile influence its line of flight. The aviator approaching an enemy target drops the bomb long before his airplane is directly above the object aimed at. The line of the bomb's flight is a parabolic curve. The speed at which the airplane travels at first propels the bomb forward, almost as if it had been shot from a stationary gun. As the downward velocity of the bomb increases very rapidly, it soon becomes so great in proportion to velocity forward that the course of the missile bends sharply downward until, as it nears the ground, it is falling nearly in a vertical line. Hence, it becomes evident that accurate bomb dropping is an art attained only by much practice on the part of the aviator. The latest bombing machines were equipped with sights which enabled the birdman to drop these deadly objects with greater accuracy than had been possible earlier in the war. While some of the expert European bombers scorned the new inventions in sights and preferred to continue the use of makeshift sights which they themselves had invented and installed on their planes, the average accuracy of bomb dropping was considerably greater after bomb sights came into general use. These sights were adjusted to height, air speed, and strength of wind. When these adjustments had been made, the two sighting points were in such position that, if the bomb were dropped when the target was in line with them, an accurate hit would be registered. We adopted a British sight, tested and found satisfactory by the Royal Flying Corps, and known as the High Altitude Wimperis, and in the United States as the Bomb Sight Mark I-A. On November 11, 1918, American factories, working on contracts placed by the Ordnance Department, had produced 8,500 of them. The job of turning out this intricate mechanism was turned over to Frederick Pearce & Co., of New York City, in January, 1918. Later in the year additional contracts were given to the Edison Phonograph Works and to the Gorham Manufacturing Co. These contracts called for 15,000 sights. By December 12, 1918, these concerns had completed a total of 12,700 of them. A 250-POUND DEMOLITION BOMB CARRYING 125 POUNDS OF EXPLOSIVE AND HAVING HEAVY CAST-STEEL NOSE AND PRESSED SHEET STEEL REAR BODY. A 25-POUND FRAGMENTATION BOMB CARRYING 3 POUNDS OF EXPLOSIVES, DESIGNED FOR USE AGAINST TROOPS. A 40-POUND INCENDIARY BOMB OF THE INTENSIVE TYPE, WITH STEEL NOSE AND FUSIBLE ZINC REAR CASING. AIRPLANE FLARE. MARK I, HIGH CAPACITY DROP BOMB. A 105-POUND DEMOLITION BOMB, CARRYING 55 POUNDS OF EXPLOSIVE. MARK II, HIGH CAPACITY DROP BOMB, NOW OBSOLETE, HAVING BEEN FOUND TOO SMALL FOR DEMOLITION PURPOSES. MARK II—A FRAGMENTATION DROP BOMB. A 20-pound fragmentation bomb, made from a converted 3-inch artillery shell, carries 1½ pounds of explosives to be used against troops. Projection at nose causes burst to take place above ground. DROP BOMB, MARK III. Airplane bombs are shaped so as to offer the least possible resistance to the air. They have fins on their tails to steady them lest they tumble over and over. On the smaller types of bombing planes, such as the De Haviland-4, the bombs were usually carried underneath the lower wings or under the fuselage, hanging horizontally by hooks or fastened by bands around the bodies of the bombs, according to their type. The bombs were dropped by a quick-release mechanism operated by a small lever within the fuselage. The production of these release mechanisms, of which several types were made, was one of the troublesome jobs in connection with the airplane bombing. All bombs are carried on the planes either suspended under the wings or fuselage of the plane or in a compartment in the fuselage. The manner of carrying and the design of the release mechanism is determined by the type of plane used. Since the weight-carrying capacity of the planes is limited, release mechanisms must be designed with a view to lightness as well as safety. These mechanisms are so designed that the observer can release any desired number of bombs either as a salvo or in a "trail fire," and the order of releasing must be so arranged that the balance of the plane will be disturbed as little as possible; that is, if bombs are carried under the wings they should be released alternately from each wing. All bombs are fitted with a safety mechanism which enables the observer to drop them either "armed" or "safe," i. e., so that they will explode or not as desired. An occasion might develop where the aviator would have to get rid of his bombs over his own lines. These various points are all taken care of in the design of the release mechanism and are controlled by the observer with an operating-control handle placed in the observer's cockpit. All of the bombs used by our fliers and by the fliers of the other nations at war were of three distinctive types—demolition bombs, fragmentation bombs, and incendiary bombs. Our Ordnance Department built demolition bombs in five different weights: 50 pounds, 100 pounds, 250 pounds, 500 pounds, and, finally, the enormous bomb weighing 1,000 pounds—half a ton. The most frequently used demolition bombs, however, were those of the 100-pound and 250-pound sizes. The demolition bombs were for use against ammunition dumps, railways, roads, buildings, and all sorts of heavy structures where a high-explosive charge is desired. These bombs had a shell of light steel which was filled with trinitrotoluol—T. N. T., as it is more commonly known—or some other explosive of great destructive power. The charge was set off by a detonator held apart from the dangerous contents of the bomb by a pin. As the The first contract let for drop bombs of any type was given to the Marlin-Rockwell Corporation of Philadelphia in June, 1917. This contract was for the construction of 5,000 heavy drop bombs of the design known as the Barlow, and also for 250 sets of release mechanisms for this bomb. We were able to go ahead with the production of this bomb at this early date since it was the only one of which we had completed designs and working drawings when we entered the war. In November, 1917, this order was increased to 13,000, and in April, 1918, to 28,000. The Barlow bomb, however, was destined never to cut any figure in our fighting in France. The production was slow, due to the necessity of constant experimentation to simplify a firing mechanism which was regarded as too complicated by the experts of the War Department. Finally, in June, 1918, when 9,000 of these bombs and 250 sets of release mechanisms had been produced, a cablegram came from the American Expeditionary Forces canceling the entire contract. Meanwhile, the final type of demolition bomb, known variously as the Mark I, II, III, IV, V, or VI, depending upon its size, had been developed here. In December, 1917, a contract for 70,000 of the size known as Mark II, weighing 25 pounds, was given to the Marlin-Rockwell Corporation. But in June the American Expeditionary Forces informed us that this bomb would not be of value to the Air Service abroad because of its small explosive charge, and the contract was cut down to 40,000 bombs, which number the Army could use in training its aviators. By the end of November, 1918, bomb bodies of the Mark II size to the number of 36,840 had been completed. By the end of March, 1918, we had developed here a series of demolition bombs that promised to meet every need of our Air Service abroad in projectiles of their class. We let contracts for the manufacture of 300,000 of the 50-pound Mark III size, these contracts being reduced later to a total of 220,000. The manufacturers were the A. O. Smith Corporation, an automobile parts concern of Milwaukee, Wis.; the Edward G. Budd Manufacturing Co. of Philadelphia; and Hale & Kilburn of Philadelphia. Six months later the A. O. Smith Corporation had reached a production of 1,200 of these bombs a day, and completed their contract in October. Both the other concerns also completed their contracts in the autumn of 1918. TWO OF THE LARGEST DEMOLITION DROP BOMBS. The larger of these two bombs weighs 1,000 pounds and carries 570 pounds of explosive. The smaller weighs 550 pounds and carries 280 pounds of explosive. They are both made with a heavy cast-steel nose and pressed metal rear body. MARK II BOMB RELEASE MECHANISM FOR HANDLEY-PAGE MACHINE, SHOWING MARK I AND MARK IV BOMBS IN PLACE. MARK IX-A RELEASE MECHANISM AS ATTACHED TO MARK II RELEASE FOR HANDLEY-PAGE PLANE. The A. O. Smith Corporation had tooled up their factory so as to become one of our largest producers of airplane bombs. In addition to the contract already mentioned, during 1918 this concern received orders for approximately 300,000 demolition bombs of the 100-pound (Mark I) size. By November 11, 1918, they had turned out 153,000 of these and had developed a capacity for building 7,000 drop bombs daily. Another large manufacturer of drop bombs was McCord & Co., of Chicago, a concern which in 1918 received orders for nearly 100,000 bombs of the 250-pound, 550-pound, and 1000-pound sizes. By the day the armistice was signed this concern had produced 39,400 completed bombs. These bombs were the heaviest and largest ones intended for use by our service abroad. The fragmentation bombs differ from the demolition bombs in that they have thick metal walls and consequently smaller charges of explosive. They throw showers of fragments like those of high-explosive artillery shell. The demolition bombs contain, on the other hand, the maximum possible amount of explosive and produce destruction by the force of explosion. Fragmentation bombs always have instantaneous firing mechanisms, while demolition bombs are usually provided with delayed fuses, allowing them to penetrate the target before explosion. The fragmentation bombs produced by the Ordnance Bureau were smaller than the demolition type, the size most commonly used weighing 24 pounds. These bombs had thick cases and were constructed so that they would explode a few inches above the ground. As the bombs reach a velocity downward of over 500 feet per second, the mechanism had to operate to an accuracy of less than one-thousandth of a second. They were designed for use against bodies of troops. The fragmentation bombs were a late development in this class of work. The timing device to explode the bomb at the proper distance from the ground was undertaken by three concerns. The contracts for approximately 600,000 of these devices were let in July, 1918. The John Thomson Press Co. of New York City completed its contract for 100,000 mechanisms by the end of October, 1918. The National Tool & Manufacturing Co. of St. Louis completed its contract for 100,000 shortly after the armistice was signed. The Yale & Towne Manufacturing Co., Stamford, Conn., which had contracted to build approximately 400,000 of these devices, had turned out 150,000 by the end of November, 1918. Other concerns which manufactured various parts for the fragmentation bombs were the American Seating Co. of Grand Rapids, Mich., makers of school desks and seats, and the Dail Steel Products Co. of Lansing, Mich. Some idea of the quantity of fragmentation bombs in our program may be gained from the fact that the contract for the Cordeau-Bickford fuse used in the fragmentation bomb, let to the Ensign-Bickford Co. of Simsbury, Conn., called for the manufacture of 550,000 linear feet of fuse, or more than 100 miles of it. The con The Government discovered that 3-inch shell rejected for various reasons could be re-machined and used to make these airplane fragmentation bombs. The various arsenals had a large supply of them in storage. In August and September, 1918, contracts were let to large numbers of concerns to convert over 500,000 of these shell into fragmentation bombs, and by November 30, nearly 21,000 of the new bombs had been delivered. These bombs, made from the 3-inch shell, as far as the machining of the bodies is concerned were turned out in various quantities by the following firms:
The nose-firing mechanism for these bombs was produced by the Yale & Towne Manufacturing Co., Stamford, Conn.; the National Tool & Manufacturing Co., St. Louis, Mo.; and the John Thomson Press Co., New York City; while the rear cap stabilizer assemblies were produced by the Dail Steel Products Co., Lansing, Mich., and the American Seating Co., Grand Rapids, Mich. The last item on the bomb program to come into production was the fragmentation bomb Mark II-B, which was an exact copy of the British Cooper bomb, the most effective bomb of this type in use by the allied nations. Contracts for this bomb were not let until August 17, 1918, to the Lycoming Foundry & Machine Co., of Williamsport, Pa., and the Paige-Detroit Motor Car Co., of Detroit, Mich. The former company by December 1 was producing these bombs at the rate of 500 per day and the latter was just coming into quantity production the first week in December. TWO VIEWS, MARK V RELEASE TRAP (RIGHT HAND) WITH UNIVERSAL NOSE AND TAIL BEAM, MOUNTED ON T-RAILS UNDER RIGHT WING OF DH-4 PLANE. Upper—Front view, showing operating tube connected to alternating cam in fuselage. Two Mark III demolition drop bombs (150 pounds) held by supporting straps; one bomb released, showing free supporting strap. Lower—Rear view, showing method of retaining stabilizer by tail clip with three Mark III demolition drop bombs. TWO VIEWS OF MARK X RELEASE TRAP ON PLANES. Shows Mark X release trap (Cooper) mounted upon T-rails under wing of DH-4 plane. Bowden control wire and casing connected to fuselage. Two Mark II-B fragmentation bombs suspended—one arming vane retained, the other free. When the United States entered the war no satisfactory incendiary bombs had yet been produced by any country, and consequently a long period had to be given over to experimentation before quantity production could be attained. We produced two types of incendiary bombs, the first being of the scatter type, designed for use against light structures, grain fields, and the like, and the second of the intensive type, for use against large structures. Later on in our program we abandoned the manufacture of the scatter type incendiary bombs on cable instructions from abroad, as it was found that the wet climate made a bomb of this type of little value. The American intensive bomb, while it had not yet come up to our ideal and was in process of evolution during its manufacture, nevertheless was regarded by our officers as more effective than any other bomb of its type in existence, since it produced a larger and hotter flame. Our intensive incendiary bombs weighed about 40 pounds each and contained charges of oil emulsion, thermit, and metallic sodium, a combination of chemicals that burns with intense heat. These bombs were used against ammunition depots or any structures of an inflammable nature. The sodium in the charge was designed to have a discouraging effect upon anyone who attempted to put out the fire of the burning charge, since metallic sodium explodes with great violence if water is poured upon it. Of the scatter bombs we built 45,000 before abandoning the manufacture, an action taken in September, 1918. When hostilities ceased we had out contracts for 122,886 of the intensive bombs and about 86,000 of them had been delivered ready for loading. One of the large manufacturers of incendiary bombs was the Conron-McNeal Co., of Kokomo, Ind., manufacturers of skates. The company had to equip its plant with new machinery especially for handling this novel manufacturing enterprise. In all, they produced 50,000 bombs and were turning them out at the rate of 400 per day when the armistice was signed. This concern was the pioneer in the manufacture, the subsequent contractors profiting by the experience of the Conron-McNeal Co., and consequently being able to obtain quantity production more quickly than the Kokomo plant had been able to reach it. The Globe Machine & Stamping Co., of Cleveland, Ohio, built 30,000 bombs and 36,400 firing mechanisms before hostilities closed, and eventually reached a production rate of 500 bombs and 1,000 firing mechanisms per day. Parrish & Bingham, also of Cleveland, produced 13,000, and were turning them out at the rate of 400 daily when the production was stopped. The C. R. Wilson Body Co., of Detroit, built 42,562 of the intensive bombs and reached a daily production of 500. The New Home Sewing Machine Co., of Orange, Mass., manufactured 20,000 firing mechanisms for the scatter-type bombs. One of the interesting phases of the bomb manufacturing program grew out of the necessity for target practice for our aviators. For this work we built dummy bombs of terra cotta, costing about a dollar apiece. Instead of loading these bombs with explosive, we placed in each a small charge of phosphorus and a loaded paper shotgun shell, so that the bomb would eject a puff of smoke when it hit its object. The aviators could see the smoke puffs and thereby determine the accuracy of their aim. The Gathmann Ammunition Co. of Texas, Md., was the first contractor for dummy bombs, building 10,000, which were delivered in the spring of 1918. In the spring and summer of 1918, the Atlantic Terra Cotta Co., the New Jersey Terra Cotta Co., both of Perth Amboy, N. J., and the Federal Terra Cotta Co. of Woodbridge, N. J., each built 25,000 of these bombs. In September additional contracts for 50,000 dummy bombs were given to each of these three concerns, while another contract for 25,000 went to the Northwestern Terra Cotta Co. of Chicago. By the end of November these concerns had delivered nearly 34,000 of the 175,000 bombs contracted for, and were turning them out at the rate of 1,300 per day. The Essex Specialty Co. manufactured 10,000 phosphorus rolls for dummy bombs, and the Remington Arms-U. M. C. Co. supplied 10,000 shotgun shells for the first bombs produced. Later the Remington Arms Co. produced 100,000 shotgun shells for dummy bombs. AIRPLANE PHOTOGRAPHIC SUPPLIES.In four days of the final drive of the Yankee troops in the Argonne district the American photographic sections of the Air Service made and delivered 100,000 prints from negatives freshly taken from the air above the battle lines. This circumstance is indicative of the progress made by military intelligence from the days when a commander secured information of the enemy's positions only by sending out patrols, or from spies. The coming of the airplane destroyed practically all possibility for the concealment by day of moving bodies of men or of military works. Mere observation by the unaided eye of the airmen, however, soon proved inadequate to utilize properly the vantage point of the plane. The insufficient and often crude and inaccurate drawings brought in by the airplane observer were early succeeded by the almost daily photographing of the entire enemy terrain by cameras, which recorded each minute feature far more accurately than the human eye could possibly do. The airplane, to quote the common saying, had become the eye of the Army, but the camera was the eye of the airplane. This development in military information-getting from start to finish was entirely the product and an evolution of the great war. When the war broke out in 1914 there were no precedents for the military photographer to go by, nor had any specialized apparatus ever been designed by either side for this purpose. As a result the first crude makeshifts were rapidly succeeded by more and more highly developed equipment. At the outset of the war, before antiaircraft guns were brought to efficiency, it was possible for the observation planes of the British, the French, and the Germans to fly at low altitudes and take satisfactory pictures with such photographic appliances as were then in common use. But as the "Archies" forced the planes to go higher in the air, special equipment had to be designed for longer distance work under the adverse conditions of vibration and speed, such as exist on airplanes. It is a tribute to the photographic technicians of the world that they were able to produce at all times equipment to meet these increasing demands. TYPE DR-4, DE RAM CAMERA. TYPE A-3, HAND-HELD AIRPLANE CAMERA. TYPE L, 4 x 5 PLATE CAMERA. MOBILE FIELD PHOTOGRAPHIC OUTFIT, USED FOR AIR SERVICE. It includes a dark room, printing lantern, and light-generating plant. As the airplanes moved into higher altitudes, longer focus lenses had to be employed, special dry plates developed, and special color filters provided to overcome the haze created by humidity in the long spaces between the cameras and the ground. When the war ended, cameras were in common use taking photographs at an altitude of 4 miles with such microscopic fidelity as to show even where a single soldier had recently walked across a field. The American Army came into the war almost innocent of any information at all on the subject of war photography. Such technical information as the allied nations had developed during the war had been most carefully guarded from us and all other neutral countries, with the result that what information we had was of a meager and conflicting sort. Although in the early months of our participation in the war the Signal Corps, which then had charge of all phases of aerial warfare, made large purchases of motion-picture cameras, hand cameras, and view cameras, it was not until the end of 1917 that our officers were able to begin their real development of aerial photography. By this time we had received much valuable information from the foreign high commissions and samples of their earlier apparatus. Aerial photography had become one of the leading activities of the air service. Thus in April, 1917, the British service made 280,000 pictures at the front, and a great part of all flying was done to secure photographs. Moreover, the art was advancing at such a pace that practices in approved use one week at the front appeared likely to become obsolete the next, as new methods and new equipment superseded the old. For years America had been second to none as a photographic country, and it was to be expected that this country would make notable contributions to the new science. It may indeed be wondered why, with the experimental laboratories and the skilled technicians at our command, we did not start at once to develop our own aerial designs and equipment. Our officers, however, felt that such a course would be likely to duplicate much of the work already done by the allied countries, who stood ready then to furnish to us the results of their experiences. While original research work might result in the invention here of certain equipment of superlative merit, yet we would be sure, in the course of such an undertaking, to adopt methods which had been tried and discarded by the allies and which we ourselves would have to discard when experience had proven them to be without value. The information in our hands in December, 1917, showed that the British system of air photography differed radically from that of the French. The French cameras made a relatively large negative, 18 by 24 centimeters in dimension, on a glass plate. The magazines of the French cameras held 12 plates, and extra magazines were carried in the plane. These cameras were fitted with lenses of relatively long focus—20 inches. Three operations were necessary to make an exposure. The photographer must change the plate, set the focal plane shutter, and press the release. When the negatives were developed, fixed, washed, and dried, prints were made by contact. The British used a smaller-sized plate, 4 by 5 inches in size. Their cameras were equipped with the only lenses available in England in the early part of the war—lenses of relatively short focus, ranging from 8 to 12 inches in this respect. Instead of making contact prints from these plates, the British made enlargements, measuring 6½ by 8½ inches. In the earlier period of our development of aerial photographic apparatus, we were in the same position as the British as regards lenses. We had no adequate supply of long-focus lenses. Consequently we followed the British designs of cameras and adopted the British system almost explicitly in the training of aerial photographers. It had been our first thought to use films to a great extent on the front, since America was the country which had perfected the photographic film, and was therefore, presumably, best equipped in skill to adapt it to war uses. But plates had been used practically exclusively by the British, the French, and the Italians; and it appeared wisest to follow their experience at first, though all agreed that film, with its small bulk and weight, would be greatly superior for airplane use. The Photographic Experimental Department of the Air Service, which was organized in January, 1918, had as its major problems the design and test of aerial cameras and all their parts and accessories. Equally important with this problem was that of sensitive plates, papers, color filters, and photographic chemicals. The corps of photographic and optical experts, into whose hands these matters were placed, early secured the active cooperation of the chief manufacturers of photographic apparatus and materials in this country. In the laboratories in Washington, D. C, Langley Field, Va., and Rochester, N. Y., comprehensive development work was inaugurated, leading ultimately to perfection of new designs of cameras and the development of plates and other photographic materials equal or superior to any available abroad. The first airplane camera which it was decided to put into production in America was a close copy of the British type "L," which At the same time it was generally agreed that we should plan to follow the French practice as soon as lenses of greater focal length could be manufactured in this country. Increase in focal length was becoming imperative, because aerial photographers were being compelled to make exposures from much greater heights than in the earlier part of the war. For the sake of those unacquainted with photography it may be stated here that lenses of short focal lengths will not record the details of objects a great distance away from the camera, the longer-focus, rarer, and more expensive lenses being required for distance work. As a basis for the design of cameras of longer focus a sample of the 20-inch focus camera used by the French had been sent to this country by the American Expeditionary Forces. The first camera authorized of this focal length was similar in general character to this French camera. It was constructed on the unit system, each part—shutter, camera body, lens cone, and magazine—being of standardized dimensions. It was understood that these standard dimensions were to be followed in all subsequent cameras both in this country and in the countries of the allies. The idea constantly put before all designers of aerial cameras has been that of the automatic type, in the use of which the observer or pilot will have a minimum of work. Late in 1917 the Photographic Section of the Air Service, American Expeditionary Forces, secured the rights for the manufacture of an ingenious design of automatic plate camera invented by Lieut. DeRam, of the French Army, and requested that this be put in production. In this camera the magazine, which carries 50 plates, 18 by 24 centimeters in size, rotates between each exposure, while the exposed plate is removed from the front of the pile and carried to the back. After some study here of the incomplete model, this camera was redesigned in such form as to fit it for methods of American manufacture. It was made semiautomatic in operation; that is, the work of the observer or pilot consisted Meanwhile experiments were actively pushed in the matter of the utilization of film. Various difficulties and problems had to be solved before film could be considered practical. Considerable time was consumed in overcoming the peculiar static electrical discharges which occur on film in cold, dry regions, such as in high mountains or the upper atmosphere, and fog the sensitive surface by their light. The film camera finally decided upon was based on a fundamental design by the Folmer & Schwing organization of the Eastman Kodak Co. This camera, known as the "K" type, carries a film on which 100 exposures, 18 by 24 centimeters in dimension, can be made at one loading. The film is held flat by an ingenious device. The film strip passes over a flat perforated sheet behind which a partial vacuum is set up by a suction, or "Venturi," tube extending outside the body of the airplane. The camera is entirely automatic, and is driven either by a wind turbine of adjustable aperture or, in war planes, by electric current from the heating and lighting circuit. The observer in the airplane needs only to start the camera and regulate its speed according to the speed with which the airplane is passing over the ground below, and the camera thereafter will, of itself, take pictures at such intervals as to map completely the terrain under observation. In conjunction with the use of film in cameras came the question of handling the film in the dark-room; that is, the ordinary manipulations of developing, fixing, washing, and drying—a serious problem when the large dimensions of the film, its length, and difficult characteristics in handling are taken into consideration. This problem was attacked and a film developing, handling, and drying machine was produced. Some 200 of these automatic film cameras were on order at the close of the war. Altogether over 1,100 airplane cameras of all types had been and were about to be delivered when the armistice came. These were built by the Eastman Kodak Co., Rochester; the Burke & James Co., Chicago; the G. E. M. Engineering Co., of Philadelphia; and Arthur Brock, jr., of Philadelphia. One of the most serious problems in aerial photography is the proper mounting of the camera in the plane. Not only does the plane travel at great speed, which makes necessary exceedingly short exposures and therefore highly sensitive photographic materials, but the motor causes a continuous vibration which, communicated to the camera itself, would be fatal to obtaining sharp pictures. The experimenters of the Air Service carried out a long, extensive, and most interesting investigation at Langley Field to make clear the whole question of preventing the vibration of the airplane camera. The plan adopted was to send up a camera thus mounted on an airplane, focus it on a light on the ground below, open the shutter, and take a time exposure from the swiftly-flying plane. The result, of course, was a streak, or trail, written on the plate by the point of light below, the jagged or wavy character of this trail indicating the vibrations of the camera and determining the proper principles of a suitable mounting. The first thought was to do this work at night, as the British had done, when the light below would pierce the darkness distinctly. But night flying is hazardous, and a better plan was called for. Nor would the proposal to use an extremely strong light in broad daylight do, because, while the light would indeed be photographed continuously across the plate, so also would the surrounding ground, and the general result would be a fogging or blurring of the outlines of the streak. Finally the problem was solved by conducting the aerial experimental work over woodland in the late afternoon. A strong, reddish light was placed in the woods so as to be visible from above. The surrounding green foliage supplied a frame of sufficient contrast to the light to make its impression distinct on the plate. To emphasize the contrast, the camera lens was covered with a reddish colored ray filter, and this brought out sharply the outline of the streak. These tests resulted in the design and production of new and unique camera mountings which successfully stopped all vibrations of the camera. A problem on which it was necessary to have the closest cooperation of the plane designers was that of installing the large 20-inch focus cameras in the airplane. There is little room at best in a plane, and the demands for armament, wireless, and bombing space all had to receive attention. In the American service a distinct advance was made in the design of a special plane intended primarily for photographic reconnaissance. Several of these planes, which were the most completely equipped for photographic purposes of any designed during the war, were built and would have been put into quantity production in the late fall of 1918. Parallel with this development of apparatus went studies of the sensitive materials and methods of photography from the air. Because of the swift motion of the plane extremely short exposures are imperative. Consequently, the most advanced technique of instantaneous photography had to be applied. The cooperation of various plate manufacturers was obtained, who brought out especially for the Government several new plates which showed on As an airplane rises higher and higher in the sky, the moisture of the intervening atmosphere between the machine and the ground creates a haze which makes aerial photography above a certain height unsatisfactory and even impossible with the naked lenses as used on the ground. The problem of finding the best means for piercing aerial haze occupied the attention of a corps of experts working both in the laboratory and in the field. The solution lay in the use of special color filters of general yellow hue which obscured the bluish light characteristic of haze. Filters of new materials specially adapted to airplane use were made available as a result of this study. Field equipment of quite new and special design for performing photographic operations had to be designed and built. Among the most interesting of these developments was the photographic truck or mobile photographic laboratory. This consisted of a specially designed truck and trailer containing all the equipment necessary for the rapid production of prints in the field. The truck body was equipped with a dynamo for furnishing the electrical current required for lights and drying fans, while each unit was provided with an acetylene generator for emergency use, if the electrical apparatus should break down. The mobile dark room carried on the trailer of each unit was equipped with tanks, enlarging camera, printing boxes, and other necessary apparatus. In all, some 75 of these field laboratories were constructed. While the development of apparatus and new materials was from a popular standpoint in many ways the most interesting phase of the work of the photographic scientists, nevertheless it should be remembered that the great problem in this, as in all other fields of American endeavor, was to produce the supplies in tremendous quantities. In October, 1918, we shipped overseas 1,500,000 sheets of photographic printing paper, 300,000 dry plates and 20,000 rolls of film. We also sent 20 tons of photographic chemicals. These were merely the principal items in the consignment. Besides paper, plates and chemicals, the field force required developing tents, trays, printing machines, stereoscopes, and travelling dark rooms, to name only some of the principal items. Much of the material already on the market was not suitable for the purpose, a fact requiring the production of specially manufactured supplies. THE FIREWORKS OF FLYING.It is interesting to consider that without fireworks, and particularly some of the familiar forms of them used to celebrate the Fourth of July, war flying would have lost much of its efficiency. Night flying would have been well-nigh impossible, while day flying would have had to invent substitutes for fireworks had the latter not been available. MARLIN MACHINE GUN WITH FIXED MOUNTING, ON A JN-4 FUSELAGE. TWO LEWIS MACHINE GUNS WITH MOVABLE MOUNTING, IN THE OBSERVER'S COCKPIT OF A DE HAVILAND-4. AIRPLANE FLARE. HOLT WING-TIP FLARE HOLDER. The squadron fields near the front were kept as dark as possible at night for obvious reasons. The first inkling that a squadron commander might have of the approach of one of his aviators at night would be the sudden appearance high in the air of a green or red or white Roman-candle ball. This would be the signal inquiring if the landing field were clear. A pyrotechnic star of a predetermined color, shot from the ground, would answer the homing birdman; and, if the signal were in the affirmative, he would descend through the sheer blackness, unable to see clearly, yet confident that he would make his landing safely. As the plane neared the ground suddenly under one of the wings a flare of dazzling power would commence to burn, for a few seconds flooding the field with light. In that brief space of time the plane would have made its landing, and soon field and quarters would again be obscured under the protecting blanket of darkness. Every service airplane at the front was equipped with one or more signaling pistols. In appearance these weapons were more murderous than the "gat" carried by a desperado of the movies, but, like the prize bulldog with the undershot jaw, they were more deadly in looks than in deeds. Their formidable-appearing cartridges were larger than the shells used in shotguns, resembling the latter almost identically in appearance; but every one of these shells contained only a Roman-candle ball and a sufficient charge of powder to eject the star a good distance into the air. The sound of the discharge was a mere whisper of the shattering roar that might be expected from such a redoubtable piece of ordnance. These aviation pistols were similar to the Very signal pistols used in the trenches. The stars shot were three colors, red, green and white, and the color of a cartridge's star was painted on the end of the shell. This base was also ridged with a different pattern for each color, so that the aviator at night could feel with his fingers and tell the color of the cartridge without seeing it. Codes of numerous messages were worked out in different combinations of these three colors. The stars were quite visible in broad daylight, too, and were used for many signaling purposes. They indicated the position of enemy troops or the presence of hostile aircraft, they called for help from other airplanes, and they signaled squadron orders when the machines were flying in formation. But the signal pistol had a more sinister use. If the pilot were driven down in enemy territory, it became his duty to destroy his machine. In some cases the signal pistol was used effectively to set airplanes on fire under such conditions. The pilot had only to open his gasoline tank and fire a Roman candle ball into the escaping fluid. In other cases when the aviator landed amid enemy troops While we manufactured Very pistols in this country, all of those actually used by our fliers in France were purchased abroad. Night-flying is one of the most hazardous duties of the aviator, the chief danger being in landing. The fields well back of the front were usually brightly illuminated by flood lights at night, but those nearer the enemy were left in darkness, as a rule, to protect them from the attacks of hostile aircraft. The aviator at night can usually see the ground faintly, but he is unable to make an accurate judgment of the distance of his machine above the ground. This danger was greatly alleviated when the wing-tip flares were invented. The wing-tip flare consisted of a small cylinder of magnesium material in a metallic holder, one flare being fitted under each lower wing of the plane. Each flare was controlled by a push button in the pilot's cockpit. Pressure on the button sent an electric spark into the magnesium and touched it off. When the descending pilot at night judged that he was near the ground he pushed one of the buttons. Immediately the flare ignited and burned for about 50 seconds with the brilliant light of 20,000 candle power. Being hidden by the wing, this light did not dazzle the eyes of the aviator, but the reflection from the under surface of the wing lighted up the field for an adequate distance in all directions. Another important use of pyrotechnics occurred in those enterprises known as night-bombing raids. Since both sides kept their vulnerable ammunition dumps and their important buildings completely unlighted at night, even though the night raider knew he was in the general vicinity of his objective, hits from bombs dropped from aloft were almost accidental. To enable the night bomber to see his target the interesting piece of pyrotechnics known as the airplane flare was invented. This was a great charge of magnesium light held in a cylindrical sheet-iron case nearly four feet long and half a foot in diameter, the exact dimensions being 46 inches by 5 inches. The flare weighed 32 pounds. Within the cylinder was not only the magnesium stick but also a silk parachute 20 feet in diameter. The entire cartridge was attached to the airplane by a release mechanism similar to those holding the drop-bombs. When over his objective at night the pilot or observer touched a button and the entire cartridge, iron case and all, dropped from the plane. A pin wheel on the lower end of the case was instantly spun by the rush of air, and the resulting power not only ignited the magnesium but at the same time detonated a charge of black powder sufficient in force to eject from the case the flare and its tightly rolled parachute. The parachute immediately opened; and the burning flare descended slowly, flooding a large area of the ground below with a light of 320,000 candlepower, this light burning for about 10 minutes. Such a light not only enabled the bomber to drop his destructive missiles accurately, but it was found by experience that it dazzled the eyes of antiaircraft gunners below and made their aim inaccurate. The light of this flare was so strong that it was possible for the airplane above to obtain photographs of good detail on the darkest of nights. We were just starting to produce these flares when the war ended. In fact the actual production of pyrotechnic supplies in this country was small, the American Expeditionary Forces depending almost exclusively for these supplies upon French and British sources. KEEPING OUR FLIERS WARM.When the commander of an airplane squadron sends an aviator into the high altitudes, he sends him into climate that much of the year is colder and more severe than any known on earth, even at the North Pole. Not only is the temperature of the air likely to be many degrees below zero at the heights which war planes attained, but the flier must face this bitter cold in the gale of wind that is never blowing less than 100 miles per hour. Consequently when we trained a corps of aviators to fly at altitudes of 18,000 to 20,000 feet above the western front, it was necessary for us to design and manufacture for them the warmest clothing ever made. They were dressed more warmly than any Polar exploration party that ever set forth, more warmly in fact than any other class of men in the world. For we not only gave them the protection of all the fine wool, leather, and fur that they could wear without hindering their movements, but in addition we literally wrapped them in flexible electric heaters. The first purchases of aviators' flying clothes were made by the coordinated action of the Council of National Defense and the Quartermaster's Department. It was soon apparent that the design of such clothing was a special matter which the aviation authorities themselves should control, and purchases thereafter were all made by the Bureau of Aircraft Production. There were no standard styles at the time, so it became necessary for us to develop our own equipment. This development resulted in an output for the flier that became standard. In moderate weather the flier wore upon his head a woolen hood, or helmet, extending well down over the forehead to the eyes, and around the neck to the shoulders. In cold weather, or for high-flight work, this headgear was augmented by a silk helmet of double thickness, having between its layers an electrically heated pad connected by copper wire to the electric generator on the plane's engine. Outside of this was worn a soft leather helmet lined with fur, extending down over the back of the head, covering the ears and cheeks, and fastening under the chin. Then the face was entirely covered In addition to this equipment the aviator who went up to the great heights wore the oxygen mask. This was of rubber, and, besides supplying oxygen, it contained a transmitter, allowing him to speak as well as to hear by wireless. Over the body was worn a one-piece flying suit extending from the feet to the throat, belted and buttoned tight at the ankles and wrists. The outer material of this suit was waterproof, and when it was buttoned on there were no gaps through which the air might penetrate. This suit was lined throughout with fur. It was a considerable problem to find a fur of extreme warmth with a pelt strong enough to withstand rough usage and still not be too great in bulk, and purchasable at a price not too extravagant. After the furs of many beasts had been examined and tested, it was determined that the hide and fur of a Chinese Nuchwang dog met these requirements better than any other. We were making so many of these suits that we required all of the dogskins we could get, not only in this country, but in China. Merely the final purchase of these pelts before the armistice was signed was for nearly 500,000 of them, and that many dogs in an interior Chinese province gave up their lives that the American aviation warfare might succeed. With its waterproof outer surface and its furry lining, it might seem that such a garment would be warm enough for any work. But the aircraft authorities of the United States were not content until they had installed between the fur and the outer covering thin, flexible, electric-heat units connected by silk-covered wire with the dynamo on the engine. Similar heating pads were placed in the gloves and moccasins of the fliers. On their hands, besides the electrically heated gloves, the fliers wore gauntlets of muskrat fur, these extending well up the arms and being of special design which allowed the fingers of each glove to remain in a fur-lined pocket or to be withdrawn from the pocket without removing the gloves from the hand. Over the electrically heated moccasins were worn leather moccasins extending well up the calf of the leg and lined with heavy sheep wool. These were fastened with straps and buckles. Thus clad, our aviators were acknowledged generally to be the most warmly and efficiently equipped of any at the front. Besides these special garments for warmth, the fliers required many other items of clothing, such as sweaters, leather coats, fur-lined coats, helmets, and many styles of goggles. The total cost of air clothing, provided or in course of manufacture on November 11, 1918, was over $5,000,000. Some of the major items in round numbers were 50,000 fur-lined flying suits (at $36.25), 100,000 leather helmets, an equal number of leather coats, costing anywhere from $10 to $30 each, and over 80,000 goggles at $3.50 apiece. PROTECTION IN HIGH ALTITUDE FLYING.Even to-day the veteran of the air squadron scoffs at the newfangled outfits of oxygen masks and tanks carried in an experimental way on some of the high-flying planes at the western front when hostilities ceased. Nevertheless, had the war continued a few months longer, it is probably true that the oxygen apparatus would have been included in the indispensable equipment of every airplane in the front areas. Such a development, had it occurred, would have been due largely to the efforts of the American Aircraft Service. Many aviators who have gone into high altitudes, fought there, and lived to tell about it, doubt the necessity of oxygen-supplying apparatus, since they themselves returned safely without it. Nevertheless the experiments conducted by the Bureau of Aircraft Production demonstrated conclusively that the flyer artificially supplied with oxygen in the high altitudes is much more efficient than one who is without it. These experiments were conducted in a room which duplicated the conditions at high altitudes. At 19,000 feet the pressure of the atmosphere is one-half the atmospheric pressure at sea level. The lack of pressure in itself causes no appreciable physical or mental reaction; but the reduced pressure at 19,000 feet means that in a given amount of air there is only one-half the oxygen that there is in a similar amount at sea level. The lack of oxygen is serious. Experienced aviators were placed in an air-tight chamber under the observation of Government scientists. The air in this chamber was then exhausted until it corresponded to the atmosphere at the 19,000 feet level. The subjects were then set at small mechanical tests, such as the pushing of certain buttons when different colored lights were turned on, these tasks requiring a degree of mental concentration. In this and similar tests it was discovered that not only do the subjects lose accuracy in the attenuated air, but their movements become conspicuously slower. In the parlance of the pilot they become "dopey." More than one returning aviator has confessed to this feeling when at a high altitude. When the British analyzed their air casualties during the first year of the war they found that 2 of each 100 fliers in the casualty list were killed or hurt by the enemy, 8 of them owed their misfortune to defects in the planes, while the other 90 came to the hospital or the grave because of themselves, their carelessness or recklessness, their physical failings, and all other things which may be summed up in the human equation. A thorough study on the part of the British disclosed the fact that practically all of the flying personnel was suffering from what became known as oxygen fatigue, caused by flying so many hours each day in altitudes where there was not enough oxygen to feed the body properly. Before the war broke out the aviation record was 26,246 feet above sea level. In January, 1919, this record had been lifted nearly a mile, the high point being an altitude of 30,500 feet. Early in the war pilots at the 7,000 feet level could laugh at antiaircraft fire, and few machines ever went above 10,000 feet. Thus with the first equipment the "ceiling"—that is, the average high level to which every day flying goes—was about 12,000 feet. When the war closed, a pilot was not safe under the 15,000 feet level, due to the development of antiaircraft guns, and the safest machine had become that which could fly highest. The aviators were demanding a working ceiling of 18,000 feet, and were obtaining it, too, from the latest type of planes. It was evident that the reduced oxygen at this ceiling was responsible for casualties among the fliers, and we could expect the ceiling to be pushed even higher as antiaircraft guns became more powerful. The need of oxygen equipment was plainly indicated. Even at 18,000 feet the aviator relying upon the normal oxygen supply at that altitude, while he may feel perfectly fit, is actually slow to judge distances, to aim his guns, to fire them, and to maneuver his plane. The first oxygen apparatus was designed for the British Air Service and was made at the plant of de Lestang in Paris. The demand for the apparatus was so great that an automobile was constantly kept waiting at the factory that as soon as each set was finished it could be rushed straight to the front. The first British squadron which used oxygen equipment reported that its men gave six times the service of any other British squadron. Our Air Service adopted the Dreyer oxygen apparatus, which was the original device produced by the British. We found it to be a hand-made appliance, but under our direction we adapted it to American methods of manufacture. The British apparatus was built to supply oxygen to one man only. We changed it to take care of two men. The model received was too heavy; we reduced the weight. Finally we added improvements to make it more efficient and reliable and redesigned it to meet American factory methods. GUNNER IN COCKPIT EQUIPPED WITH OXYGEN HELMET AND TELEPHONE RECEIVER, OPERATING MOVABLE MACHINE GUN. AVIATOR'S OXYGEN HELMET EQUIPPED WITH TELEPHONE RECEIVER. OXYGEN APPARATUS FOR BREATHING AT HIGH ALTITUDES. Such an equipment has to be entirely automatic in its operation and as reliable as human ingenuity can make it. The Dreyer device embodies several instruments all of which must work perfectly under widely varying conditions. In use its tanks will contain oxygen under pressure ranging from 100 pounds to 2,250 pounds per square inch, yet the mechanism must deliver the oxygen to the aviator at a constant rate regardless of its tank pressure. Then the whole apparatus is subjected to temperatures that may be as high as 80° above zero or as low as 30° below. It must function evenly in the atmospheric pressure at any altitude up to 30,000 feet, delivering more oxygen as the atmosphere thins. Such was the problem of manufacture. Yet, taking up the work in January, 1918, we turned out six complete equipments by May 3, 1918, sending them overseas by special messenger for actual test on the front. Twenty-eight days later we shipped 200 sets. By the end of the war we had built 5,000 complete oxygen equipments. Of this number 3,600 had been sent to ports of embarkation awaiting shipment, and over 2,300 of these had been shipped overseas. In October we had reached a production rate of 1,000 sets per month. Some of the difficulties of this production may be read in the description of the complicated character of the apparatus. The equipment consists of a small tank or tanks, the pressure apparatus, the tube leading from the reservoir, and finally the face mask covering the mouth and nose. The mask has combined with it either the interphone, a mechanism which cuts off the roar of the engine from the ears of the passengers and allows the pilot and observer to talk freely with each other, or in certain cases the receiver of the radio telephone or telegraph. The flow-regulating apparatus consists of five parts. In front of the pilot is a high-pressure gauge to indicate the supply of oxygen in the tank. In the tank there is a high-pressure valve with an upper chamber which compensates for the temperature. There is also a shut-off valve, hand operated, which can be set to provide a flow of oxygen to one man, to two men, or to none at all. Then there is a regulating valve operated by an aneroid barometer which adjusts the oxygen flow to the altitude, the flow increasing as the machine goes higher. Finally in the pilot's view there is a flow indicator consisting of a small fan wheel which tells the aviator that the oxygen is actually flowing. The mask presented a difficult problem, as it must be big enough to contain the radio receivers and still enable the aviator to see and work. Yet the mask must keep its adjustment in a gale of wind at least 100 miles per hour in velocity. The actual use of the equipment on the front was just starting when the armistice was signed. We sent across to France a special TWO VIEWS OF BOMB SIGHTS USED ON AIRPLANES. Upper picture shows bomb sight on De Haviland 4. Lower picture shows high-altitude bomb sight. Set from readings of instruments showing altitude and air speed. It indicates to the bomber the precise instant for release of the bomb in order to reach the target. AVIATORS EQUIPPED WITH TELEPHONE TRANSMITTERS AND HEAD SETS TO COMMUNICATE WITH EACH OTHER. Electrical science was called upon to furnish marvels and prodigies indeed during the recent war as aids to the American arms, but in no respect did it respond in more successful and spectacular fashion than it did when asked to produce a wireless telephone system that would make possible the transmission of human speech to and from moving airplanes. It is doubtful if any other branch of science enlisted for war work produced any instrument or mechanism so far in advance of what was known before the war as the airplane wireless phone was in its class. To be sure, we had the radio telephone some time before America entered the war or even before the war broke out in Europe in 1914. Ever since the scientists began experimenting with wireless electricity it has been axiomatic that, at least theoretically, whatever you can do with wires you can do without wires. And so following the development of the wireless telegraph came the production of the wireless telephone, and the invention had been so perfected in 1915 and 1916 that in the United States Navy's official test at the Arlington Station, across the Potomac River from Washington, human speech sent out by the transmitters there was heard simultaneously at the Eiffel Tower in Paris and at the Government's own wireless station in Hawaii. But there is a vast difference between using the wireless telephone in the quiet of the radio rooms aboard ship or in the shore stations and using it amid the roar of the powerful engine propelling an airplane. The equipment, too, that had been used on the ground was altogether too cumbersome to go into the fuselage of an airplane. As early as August, 1910, American genius had successfully accomplished wireless telegraph transmission from airplane to ground, and in October of the same year the idea of aerial fleet command by telephone was conceived and plans for its development discussed by Army officers on duty at the International Aviation Tournament at Belmont Park, Long Island. In 1911 a message was successfully transmitted from an Army airplane over a distance of 2 miles. In 1912 the Signal Corps had increased the distance to 50 miles. Two years later, in the Philippine Islands, a message had been successfully received on an airplane in flight over a distance of 6 miles. In 1915 the Aviation Section entered upon a definite plan of development of aircraft wireless at the Signal Corps Aviation School, San Diego, Calif. This plan was based upon the Belmont Park idea and discussions, with the voice-commanded tactical air fleet as the ultimate goal. The airplane had changed from the pusher to the tractor type, with the noise of the motor of the latter driven back by the blast of the propeller into the face of the aviator. The airplane wireless problem was thus quite completely changed. Under these new conditions, however, the development was entered upon, and it has since been continuous. In October a spring-driven dictaphone was taken into the air and a record of speech made in the noise of the motor. This was contemporaneous with the successful long-range experiments in radio telephony at Arlington, referred to above. A study of this dictaphone record convinced the aviation officers that the idea of the radio telephone for airplanes was entirely practicable. Experiments during the fall and winter with various means of driving the wireless power plant resulted in a decision to develop the air fan as a source of power rather than the gear or belt system. This development continuing through 1916, transmission by telegraph from airplane was accomplished up to 140 miles, means for receiving in the noise of the motor were worked out, and a message successfully telegraphed between airplanes in flight. The radio telephone was under construction, and in February, 1917, the voice was first transmitted by telephone from airplane to ground. Like Alexander Graham Bell's first wire telephone, the apparatus was crude. But the door was unlocked and ready to be opened upon the new field of development. When on May 22, 1917, Gen. Squier, the Chief Signal Officer of the Army, called upon the scientists to develop at once an airplane telephone, he was not only introducing them into what was to many of them a new field, but he was asking them to produce what the science of Europe had been unable to create in nearly three full years of acquaintance with the successful ground system, although the needs of airplane fighting demanded this invention as they demanded almost nothing else. It will thus be seen that when we began this development as a war measure we had a considerable basis of experience to work upon. The Army had established the foundation of operation on the airplane, made a study of the tactical requirements, and knew what it wanted. The Western Electric Co. in 1914 and 1915 had conducted extensive experiments with the radio wireless telephone at a ground station at Montauk, Long Island, and had played an important part in the long-range experiments at the Arlington station. There had been wireless voice communication before this At the conference with Gen. Squier in May was Col. Rees of the Royal Air Force of Great Britain; Col. C. C. Culver, United States Army, then a captain; and F. B. Jewett and E. B. Craft, respectively the chief engineer and the assistant chief engineer of the Western Electric Co. At this meeting Gen. Squier outlined the future of the part the airplane was to play in the war, and pointed out how invaluable would be a successful means of communication between battle planes when flying in squadron formation. Mr. Jewett had received his commission as a major in the Signal Corps, and he was ordered to take charge of the work of developing radio communication for aircraft. Capt. Culver had taken part in the 1910 experiments and discussions, and since 1915 had been conducting the Army development of airplane wireless at the aviation school at San Diego, Calif. He was detailed to work with Maj. Jewett and his engineers, bringing to their assistance the result of his experience and the point of view of the trained military man and the aviator. The first development was carried on in the laboratories of the Western Electric Co. on West Street, in New York. Men and materials were drafted from every department of the company, and the laboratories were soon seething with activity. In a few weeks the first makeshift apparatus was assembled, and the first practical test of a radio phone on an airplane was made at Langley Field at Hampton, Va., less than six weeks after the Signal Corps had given the go-ahead. Three employees of the Western Electric Co. on that day established telephone communication between an airplane in flight and the ground. A few days later the first apparatus produced successful communication between planes in the air. It is not possible here to go into a technical description of the wireless telephone. The most vital part of the apparatus, however, and the essential factor in airplane wireless telephone communication is a vacuum tube containing an incandescent filament, a wire mesh or grid, and a metal plate. By means of electrical current the wire filament is heated to incandescence. The tube has the property of receiving the energy of the direct current of a dynamo and, through the medium of the wireless antennÆ, of throwing it out into space as a high-frequency alternating current. Such is the sending tube. A modification of the same tube picks up from the antennÆ the high-frequency alternating vibrations from another sending apparatus and transforms them into direct current, carrying the sound waves of the human voice along with them. The design of the radio apparatus itself was relatively simple for the experts who had undertaken the work, for the company had already developed some highly successful forms of vacuum tubes, and it was an easy matter for these technicians to assemble tubes with the necessary coils, condensers, and other apparatus of the transmitting and receiving elements and produce a system of such small compass that it could be carried on an airplane. But working this apparatus under ordinary conditions in the quiet laboratories and in a swift-moving and tremendously noisy airplane were two different propositions. One of the first problems was to design a comfortable head set which would exclude all undesirable noises and admit only the telephone talk. A form of helmet was finally devised with telephone receivers inserted to fit the ears of the pilot or observer. Cushions and pads adjusted the receiver to the ears, and the helmet fitted close to the face so as to prevent as far as possible the transmission of undesirable sounds either through the ear passages or through the bony structure of the head, these bones acting as a sort of sounding board. The designers finally developed a helmet that solved this portion of the problem. Not only was it necessary to exclude the roar of the engine and the rattle of the machine gun from the ears of the men receiving the radio communication, but it was also necessary to filter out these sounds from the telephone transmitter. Every person who has ever shouted into a telephone knows how sensitive the ordinary telephone transmitter is to extraneous noises. It requires no wide stretch of the imagination to hear in fancy how an ordinary transmitter would behave when beside the exhaust of a 400-horsepower Liberty engine. A brilliant line of experimentation conducted by one of the scientists at the laboratory resulted in a telephone transmitter or microphone which possessed the extraordinary quality of being insensitive to engine and wind noises and at the same time highly responsive to the tones of the human voice. With the receiver and the transmitter perfected the scientists thought that the problem of airplane telephoning was solved; but nevertheless three months of hard work were required before the entire system could be adjusted and put in such shape that it might be considered a practical device for everyday use. The question of weight was of utmost importance, and a structure that would adequately house and protect the delicate parts of the mechanism from the vibration and jars of flying and landing and at the same time not be too heavy for practical use on the plane was a difficult problem in mechanical design. Day after day the inventors took the mechanism up in flying machines and brought it back night after night for more work in the laboratory. INTERIOR VIEW. AIRPLANE RADIO TELEPHONE SET BOX. EXTERIOR VIEW OF SAME. This was a period, however, of rapid progress. Officials appearing on Langley Field from time to time witnessed informal demonstrations of this development. In August Mr. Baker, Secretary of War, and Gen. Scott, Chief of Staff, listened to a conversation being carried on in the air, and some six weeks later Brig. Gen. Foulois witnessed a similar demonstration and from the ground directed the movements of the airplane in flight. The experimental apparatus had reached such a state of efficiency that on October 16, at Langley Field, communication by voice was carried on between airplanes in flight 25 miles apart and from airplane to ground over a distance of 45 miles. By September cables had been sent abroad telling of the progress made in this country on the development of this apparatus. Our officers abroad were skeptical and could not believe that this country could outdistance the scientists of the allies who had had three years of war experience to draw upon. By October the designers had brought the system to a perfection where they were willing to risk its use in actual war flying, and Col. Culver took to the American Expeditionary Forces in France several trunk-loads of the apparatus to acquaint those abroad with what had been done and to test the apparatus under service conditions. Meanwhile the development work continued in this country. Early in December the operation of the apparatus was exhibited in an official test at the Morraine Flying Field at Dayton, Ohio. A large number of military and civilian officials not only of our own country but of the allies had been invited to witness this test. It must be remembered that at this time even those who had heard about the progress being made were skeptical of the possibilities of the successful adaptation of the radio telephone to airplane work. The designers of aircraft never look with favor upon additional equipment which may clutter up the machine with trailing wires and the like and possibly compel alterations in standard lines. The pilots, also, do not usually give a friendly reception to new equipment for their planes. The exhibitors at Dayton planned to have two planes in the air at once, so that the officials might listen in on their conversation at a ground station located on the top of a hill near the flying field. By hard work the inventors got their equipment installed, and just at dark on the evening before the day of the trial one machine equipped with wireless went up into the air and held successful communication with the ground. The next morning when the official party arrived the members viewed the apparatus in the planes while the inventors explained what it was expected to do. The visitors were then conducted to the station on the hill, where those who were putting on the show had rigged up a megaphone attached to the wireless receiver so that everyone could hear without putting on a head set. The attitude of some of the officials, particularly those from the foreign nations who had had experience in war flying, was skeptical, if not bored. The planes left the ground, and when the machines had gone up so high that they were but specks in the sky the receiver began emitting the premonitory noises that indicated that the men in the planes were getting ready to perform. Suddenly out of the horn of the loud-speaking receiver came the words: "Hello, ground station! This is plane No. 1 speaking. Do you get me all right?" Looks of amazement came over the faces of all those who had never heard the wireless phone in operation before. Soon came the signal from plane No. 2, and then the demonstration was on. Under command from the ground the planes were maneuvered over much of that part of the country. They were sent on scouting expeditions and reported what they saw as they traveled through the air. Continuous conversation was carried on, and finally, upon command, the planes came back out of space and landed as directed. From that moment there was nothing but enthusiasm in all quarters for the radiophone upon airplanes. It was no longer a question whether the device would work or was any good, but a question of how soon the company could start manufacture and in what quantities the device could be produced. The demonstrations Col. Culver had been conducting in France began, too, to bear fruit. Both the British and the French had developed experimental apparatus by this time and this was examined and tested. Then cablegrams began to arrive from abroad requisitioning the American apparatus in large quantities—convincing evidence that it had greater promise than any other. But still difficulties were ahead, for at this stage the wireless telephone consisted of a few experimental parts built by hand. It remained a heavy task to standardize the equipment and perfect the multitude of designs and drawings that must be in existence before quantity manufacture could begin. All sorts of mechanical details slighted in the experimenting and taken care of by makeshift devices had to be worked out as practical manufacturing undertakings. It was another case of day-and-night work to put the mechanism into condition for production. The factory of the Western Electric Co. is in Chicago but its drafting rooms and laboratories are in New York. As soon as any detail was finally worked out the drawings were taken by messengers and rushed to Chicago where the work of producing the manufacturing tools had begun. Only the fastest passenger trains between New York and Chicago were patronized in this part of the development. As every detail was perfected it had to be checked by actual test in the field, so that the company's engineers were almost constantly in the air. One of these experts made 302 flights himself; and a total Immediately after the official trial in December the Government ordered thousands of sets of the radio telephone. In spite of the enormous detail involved in making ready for production, the first systems were turned out early in 1918, well ahead of the delivery of the airplanes in which they were to be used. All through this development the designers had to confine their activities within limits set by the producers of the aircraft. This in itself created some puzzling problems. For instance, a constant current of electricity must be supplied to heat the filaments of the vacuum tubes and to operate the transmitter. A simple way to provide this current would seem to be to connect a dynamo with the driving shaft of the airplane engine, but the airplane constructors would not allow any such connection with the engine. Current could be supplied from storage batteries, but the planes were already loaded down with all the gear they could carry, and the use of heavy batteries was out of the question. Therefore it was the task of the phone designers to supply a dynamo plant that would not add appreciably to the weight of the plane. This was done by installing on the outside of the plane a wind propeller, which was driven by the rushing air and had power enough to turn the dynamo. The dynamo must deliver a constant and unvarying voltage to the radio phone, if its operation is to be possible, yet a wind propeller on the airplane would be driven by air rushing by at speeds varying from 90 or 100 to 160 miles per hour, the latter figure being the speed of a diving plane. This meant that the wind propeller, and hence the armature of the dynamo, would revolve at a speed varying from 4,000 to 14,000 revolutions per minute. It would seem to be impossible to procure current at a constant rate from a dynamo varying so widely in its speed of operation; yet one of the experts engaged in this enterprise solved the problem, and the dynamo thereafter performed always in a most steady going and dependable manner. Incidentally as a sort of by-product of the undertaking the special transmitter and helmet may be employed as a means of communication between the pilot and observer in a two-seated machine. When the helmet is used for this purpose, the wireless is not employed at all, but the head sets are connected by wires so that notwithstanding the fact that one can not hear himself talk because of the noise on the plane the pilot and observer can converse over the telephone with ease. Then at any time by throwing a switch they can connect themselves with the radio apparatus and talk with the men in another plane 3 or 4 miles away or to their squadron headquarters on the ground. One good result of the airplane telephone was to speed up the training of aviators in this country and to make that training safer. But the primary object of the wireless phone was to make it possible for the leader of an air squadron at the front to control the movements of his men in the air. For this purpose extra-long range was not required, and the distance over which the machines could talk was purposely limited to 2 or 3 miles so that the enemy could not overhear the conversation except when the planes were actually engaged in fighting each other. The Navy made use of the wireless telephone sets in the seaplanes, and here the range of the equipment was made greater. The Navy also adopted a modified form of the set for the 110-foot submarine chasers. The subchasers hunted the submarines in packs, and by means of the radio telephone the commanders of the boats kept in constant touch with each other, thereby greatly increasing the effectiveness of their operations. Altogether there were produced for the Army airplanes about 3,000 combined transmitting and receiving sets of the radio telephone and about 6,500 receiving sets alone. When Stephen Montgolfier and his brother Joseph, in November, 1782, sent a sheep, a rooster, and a duck into the sky, lifted by a paper bag inflated with hot air, these Columbuses of ballooning could scarcely foresee the importance that their invention was to have in the great war 135 years later. To the humble observation balloon in France rather than to his dashing hero of a cousin, the airplane, must go the chief credit for that marvelous accuracy which long-range artillery attained during the great struggle. The balloon itself was spectacular enough once its true character was known. The fact that the American production of observation balloons during our 19 months as a belligerent was a complete and unqualified success makes the story of ballooning in France of particular interest to the American reader. After the animals of the Montgolfier barnyard had made their ascent, two friends of the brothers, M. Pilatre de Rozier and Girond de Villette, essayed to be the first human beings to take an aerial flight, ascending to a height of 300 feet and returning to earth sound of limb and body. Thereafter and until the great war in Europe the balloon remained the awe of the circus and country fair grounds and the delight of the handful of sportsmen who took up the adventurous pursuit; but, except for a limited use of captive balloons in our Civil War and in the siege of Paris, in 1870 and 1871, the balloon had no important military use. The hot-air balloon never could have become of great value to armies. In the first place, it would descend when the balloon cooled off. This defect was overcome by the use of lighter-than-air gas. Moreover, the free balloon was subject to the whims of the breezes. To overcome this characteristic the balloon must be fastened by a cable or propelled by a portable engine. It was obvious, however, to military experts that a stationary observation post anchored thousands of feet in the air would be ideal in war operations; yet for all of this obvious need, until the great war military science had perfected nothing better than the spherical balloon. The spherical, anchored to a cable, bobbed aloft in the gales and zephyrs as a cork does on the ocean waves. Although there had been some experimentation with kite balloons before 1914, it was not until the great war had been in progress for some months that the The term "kite balloon" effectively describes the captive observation balloon as we knew it in the war. It rides the air on the end of its cable much in the manner of an ordinary kite, and some of the early "sausages" even flaunted steadying tails such as kites carry. These principles applied to the captive balloon gave to its observation basket a stability unknown by the pioneer aeronauts under their spherical bags. In the first stages of the war the Artillery relied principally upon airplanes for firing directions. But, while the airplane observers could locate the targets fairly well, they frequently lost touch with their batteries because of the difficulty of sending and receiving wireless or visual signals upon their swiftly moving craft. This disadvantage brought the captive balloon into use, gradually at first, but before the end of the war on a scale which had practically displaced the airplane as a director of gun fire. The balloon came to be the very eye of the Artillery, which, thanks to the development of this apparatus, reciprocated with an efficiency beyond anything known before in the history of warfare. Sitting comfortably aloft, the observer in the kite balloon basket had the whole panorama of his particular station spread before him. His powerful glasses could note accurately everything transpiring in a radius of 10 miles or more. He was constantly in touch with his batteries by telephone and not only could give by coordinated maps the exact location of the target and the effect of the bursting shell, but could and often did supply most valuable information of enemy troop movements, airplane attacks, and the like. He was a sentinel of the sky with the keen, long-range vision of the hawk. He played a part less spectacular than the scout airplane with its free and dazzling flights, but his duties were not less important. Nor did he suffer from ennui during his period aloft. When a kite balloon went up it became the subject of alert attention by the enemy, because it was up there on hostile and damaging business. Long-range high-velocity guns turned their muzzles on it, and planes swooped down upon it from dizzy heights, seeking to pass through the barrier of shell from antiaircraft guns and get an incendiary bullet through the fabric of the gas bag, an eventuality which meant the ignition of the highly inflammable hydrogen gas, the quick destruction of the balloon and perhaps of the luckless occupants of the basket as well, unless they could get away in their parachutes. Only quick work could save the men in the basket in such a case. From the time the gas leaped into flame until the explosion and fall of the balloon there was an interval of rarely over 15 or 20 seconds. While the artillery on both sides paid considerable attention to the observation balloons, the fact was that few of them were brought down by direct shell hits. The diving airplane with its incendiary bullets was a far more deadly enemy to the balloon than the ground artillery. Certain pilots in all the air services made a specialty of hunting sausages, the nickname given to kite balloons because of their shape. In the 17 days between September 26 and November 11, 1918, our Army lost 21 balloons, of which 15 were destroyed by enemy planes and 6 by enemy shell. But it may be noted that our aviators and artillery exacted a toll of 50 German balloons in the same period and on the same front. Of 100 balloons lost at the front, an average of 65 were destroyed by enemy attacks and 35 by natural wear and tear. The German general staff so strongly appreciated the work of the allied kite balloons that in its system of rating aviators it ranked a balloon brought down as the equal of one and one-half planes. The average life of a kite balloon on an active sector of the western front was estimated to be about 15 days. Some of them lived only a few minutes. One American balloon passed unscathed through the whole period of American activity on a busy sector. While ordinarily five or six months of nonwar service will deteriorate the balloon fabric, there are many cases of useful service longer than this. When the war broke out Germany is said to have had about 100 balloons of the kite type. France and England had few of them. The German balloon was known as the Drachen. Its gas cylinder of rubberized cotton cloth was approximately 65 feet long and 27 feet in diameter, the ends being rounded. To give it a kite-like stability in the air a lobe, which was a tube of rubberized fabric, of a diameter approximately one-third of the diameter of the main balloon, was attached to the underbody of the gas bag as a sort of rudder, which curved up around the end of the balloon. This lobe was not filled with gas, but the forward end of it was open so that when the balloon rose the breeze filled the lobe with air. The inflated rudder then held the Drachen in line. The lobe automatically met the emergency. In calm, windless weather the balloon needed no steadying and the lobe was limp. Let the gale blow, and the lobe The tail-cup was made of rubberized fabric, circular in shape, about 4 feet in diameter, and about 2 feet deep when inflated by the breeze. It looked like an inverted umbrella, and was attached to the tail end of the balloon for exactly the same purpose and with the same effect as the tail attached to a kite. The Drachen type of balloon was still in the experimental stage here and in France and England when the Germans swept over Belgium. The Drachen balloon was clumsy and relatively unstable in high winds, yet its importance to the Artillery could not be ignored by the allies. The results of its work daily became more apparent. The first effort of the allies was to improve the Drachen to give it greater stability and permit it to go to higher altitudes. While this work was going on, Capt. Caquot, of the French Army, produced a kite balloon so superior that it quickly superseded what had been in use. Germany clung to the Drachen for a time, but finally abandoned it for the Caquot principles of design. The earlier balloons of the sausage type had been merely cylinders with hemispherical ends. Now for the first time, in the Caquot model, appeared a captive that was sharply stream lined. Stream lines are lines so curved as to offer the least possible resistance to the medium through which a mobile object, such as a yacht, an automobile, or an airship, moves. The Caquot gas bag was 93 feet long, as compared with the Drachen's 65 feet of length, yet its largest diameter was only 28 feet, being but a foot thicker than the pioneer German type. The Caquot, as all balloons developed in the war, was made of rubberized cotton cloth. Its capacity of 37,500 cubic feet of hydrogen gas lifted the mooring cable, the basket, two observers, and the mass of necessary equipment, and in good weather the balloon could ascend to a maximum altitude of over 5,000 feet. The principal innovation in the design of the Caquot balloon was the location of the balloonette or air chamber within the main body of the gas envelope. This chamber was in the forward instead of the rear part of the bag and along the bottom of the envelope. It was separated from the gas chamber by a diaphragm of rubberized cotton cloth, which was sewn, cemented, and taped to the inner envelope somewhat below the "equator" or median line from the nose to the tail of the gas bag. CAQUOT, TYPE R, CAPTIVE OBSERVATION BALLOON. This balloon is 93 feet long and 28 feet in diameter. Its gross lifting power is 2,600 pounds. BALLOON CONTROLLED BY WINDLASS ON A MOTOR TRUCK. WINDLASS FOR CAQUOT BALLOON MADE BY JAMES CUNNINGHAM & SONS. When a balloon of the Caquot type is fully inflated, the diaphragm rests upon the underbody of the gas envelope, and there is no air in the balloonette. Then, as the balloon begins to ascend, at the higher levels the surrounding air pressure is reduced and the gas in the balloon expands. This expansion would normally burst the envelope when the balloon is at a high altitude, except for a safety valve which pops at the danger point and relieves the pressure. Also, when the balloon is anchored it gradually loses gas, since no fabric can be made entirely gas-tight. A flabby balloon in a gale of wind is dangerous to the men in the basket. This flabbiness might be expected to increase, too, as the balloon was hauled down into the heavier air pressures. It was to overcome this flabbiness that the interior balloonette was first invented, but the new location not only accomplished this end but increased the stability, lessened the tension on the cable and allowed an almost horizontal position of the balloon itself. As the balloon rises the wind blows into the balloonette through a simple scoop placed under the nose of the balloon. This forces up the diaphragm and compensates for any loss of gas from the envelope above. If the day is calm and no air is driven into the balloonette, there is no danger from a flabby balloon anyhow, and hence no need for the air chamber. The thing is automatic. The Caquot was equipped with lobes of rubberized fabric to act as rudders. These lobes, which were spaced equidistantly around the circumference of the rear third of the balloon, filled with wind when wind was blowing and there was need of rudders. In calm weather the lobes, particularly the two upper ones, hung loosely, resembling elephant ears. On account of this characteristic the Caquots were nicknamed "elephants" by the soldiers. The Caquot maintained its stability without tailcups, and its construction caused it to ride nearly horizontally and directly above its mooring, regardless of winds. In this position it put much less strain on the anchoring cable than the old-fashioned sausage. This balloon has been operated successfully in winds as high as 70 miles an hour, so that apparently no gale could keep it on the ground. When we went into the war both our Army and Navy were practically without observation balloons, and we knew little about their construction, although we had been watching the developments in Europe. One local National Guard organization had taken to the Mexican border a locally designed captive balloon, the gift of the Goodyear Tire & Rubber Co., of Akron, Ohio. In April, 1917, the total production capacity of the United States was for only two or three military observation balloons in a month. But when the emergency came the various concerns whose plants were adaptable to this class of manufacture—the list including the Goodyear and Goodrich organizations at Akron, the United States Rubber Co., the Firestone Tire & Rubber Co., the Connecticut Aircraft Co., and the Knabenshue Manufacturing Co.—all joined wholeheartedly with the Signal Corps to solve our balloon problems. One of these problems was the production of balloon cloth, for which there had never been any commercial call in this country. Such cloth obviously must be of cotton, for in cotton we had our largest supply of textile raw material. The cloth must be closely woven, smooth, and strong, to serve as a base for the rubberizing process. The standard balloon cloth should have a weave of approximately 140 threads to the inch both ways. In our vast cotton industry only a few mills had ever made such a cloth, and then only in small quantities. In fact we found only a few looms in existence capable of weaving such cloth, which must be from 38 to 45 inches wide. A single loom could turn out only an average of ten yards of this cloth in a day. Our balloon program was to call for millions of yards of high-count cloth, and this meant the construction of thousands of new looms, as well as the training of hundreds of weavers. Naturally our cotton manufacturers were reluctant to undertake such a production, and their fears were justified when we found that the earliest deliveries of balloon cloth were frequently as high as 67 per cent imperfect. By the middle of 1918, however, the mills had so perfected their methods that the wastage amounted to only 10 per cent of the cloth woven. This wastage was largely caused by "slubs," knots, and other imperfections which prevented an even surface for rubberizing. Because of the lives which depended upon having perfect balloon cloth, the fabric was literally inspected inch by inch, and hundreds of men and women had to be educated especially in this inspection work. The development of the new art of weaving balloon cloth was an achievement of no mean degree. In April, 1917, all of our cotton mills put together could produce only enough cloth to build two balloons a week. In November, 1918, our looms were turning out cloth sufficient for 10 balloons a day, an expansion in the industry amounting to 3,000 per cent in 19 months. This expansion proceeded at a rate that always kept us a little ahead of the military schedule. To produce 10 balloons a day the cotton mills had to turn out 600,000 yards of special cloth a month. In addition to the small army of weavers, this production called into service 3,200 looms. Had the war continued another year, we would have reached our goal of 15 complete new kite balloons produced every day. Our complete project of balloons and dirigibles of all types called for a total output of 20,000,000 yards of balloon cloth. Had we reached the quantity production planned, we would have been able to supply not only our own needs but also all of the balloon needs of the allies in Europe. America had the raw materials necessary for the whole anti-German balloon program. CUTTING AND CEMENTING BALLOON PANELS IN THE GOODRICH PLANT AT AKRON. SPREADER ROOM AT THE U. S. RUBBER CO. FACTORY, SHOWING MACHINES THAT RUBBERIZE THE CLOTH. FINAL BALLOON ASSEMBLY ROOM AT GOODRICH FACTORY. As it was, we supplied to France and England a considerable number of balloons when the materials shortage in those countries was becoming acute. The foreign users of this American made equipment reported that it was equal to the best European product. It should have been. No war material was ever manufactured more conscientiously than this. In addition to the painstaking care of the producers, from start to finish a large force of inspectors watched every step in the construction of each balloon, and when America sent a balloon to the front it was right for the work it had to perform. The weaving of the cloth was but the first step in the production of the balloon fabric. The fabric of the balloon envelope resembles a sandwich in its construction, there being a thin film of specially compounded rubber between two plies of the cotton cloth. The outer ply of the cloth is cut on the bias. This method prevents any long straight tear down the grain of the fabric. The threads of the inner ply are set at an angle of 45° to those of the outer ply, thus distributing strain sufficiently to stop a "snag" practically where it starts. The cotton cloth alone can not resist the seepage of gas, and, therefore, it is necessary to rubberize it, the rubber film being really the gas-resisting envelope. In this rubberizing process the cloth must be run through the spreading machine 30 to 35 times in order to build up the thin rubber film without a flaw in it of any kind. The outside ply of the balloon fabric is "spread," that is, painted with a rubber compound containing a coloring matter. This compound makes the fabric waterproof; it gives also protective coloring to the balloon when in the air, making it less visible to the enemy; and, finally and most important, this coloring absorbs the actinic rays of the sun which are so fatal to the life of rubber. In some of the fabric the rubber film itself was colored to withstand both the heat and ultra-violet rays, thus both protecting the rubber and reflecting the heat which would otherwise expand the gas in the balloon. While in general we adopted the European standards of construction, we had to develop our own rubber compounds and cures as well as our various fabrication processes. The latest reports we received from the front stated that the American fabric not only was successful, but that it had an added characteristic which was a direct means of saving life. It was discovered that the American fabric burned more slowly than the European balloon fabric, giving the men in the observation basket more time to get away in the parachutes when the balloons were destroyed by hostile attack. When we went into the war we had never built a windlass for a kite balloon. The ability of the American manufacturer solved this problem as it did almost every other problem in the development of The best known type of gasoline windlass was that having two motors, one to turn the cable drum controlling the balloon's ascent and descent, and one for moving the windlass itself along the road. A record pull-down speed of 1,600 feet a minute, or more than three times the speed of the fastest passenger elevator, has been attained by the gasoline windlass. The electric windlass, while pulling down the balloon at the slower rate of 1,200 feet per minute, was smoother in operation. The mobile windlass would move on a road under its own power at 20 miles an hour and could tow the balloon in the air at the rate of 5 miles an hour, or even faster if necessity demanded. To play on the safe side at the start, we adopted a satisfactory windlass that had been developed in France. It was difficult to manufacture this entirely French machine with American materials and methods; yet James Cunningham, Sons & Co., of Rochester, N. Y., succeeded in obtaining a delivery of four complete windlasses per week. In addition to this windlass we designed two of our own. One of these was the product of the United States Army Balloon School and was manufactured by the McKeen Motor Car Co., of Omaha, Nebr.; the other windlass was designed and manufactured by the N. C. L. Engineering Co., of Providence, R. I. Both were put into quantity production, assuring us a sufficient number of the best windlasses ever manufactured. The first cable used to hold the balloon captive was approximately a quarter-inch in diameter, weighed 1 pound for each 8 feet of length had a breaking strength of 6,900 pounds, and was made of seven twisted strands of plow-steel wire, containing in all 133 separate wires. This cable, while it accomplished the original purpose, was early seen to have fine possibilities of development. The observers in the basket must be kept in constant communication with the Artillery and their own windlass and this communication could best and most efficiently be obtained by means of the telephone. The balloon telephone, as first used, was an entirely individual unit with its own separate cable from the basket to the ground. In this way communication was indeed established, but only at the cost of a decrease in possible altitude, increased cable resistance, and the necessity of an extra windlass for winding and unwinding the telephone cable. BALLOONISTS READY TO ASCEND. The picture shows balloonist with telephone equipment, also a parachute on side of basket. BALLOON PACKED FOR OVERSEAS SHIPMENT. BALLOON IN GOODRICH FACTORY INFLATED TO BE SUBJECTED TO AIR TEST. NURSE BALLOON CONTAINING 5,000 CUBIC FEET OF GAS. It is used in the field to replenish kite balloons with hydrogen. Previous to the entrance of the United States in the war, preliminary experiments in France were being made with the view of putting the telephone wires in the center of the main cable, thus doing away entirely with the second cable and windlass. But there had never been developed a satisfactory cable of this construction. American inventiveness at the John A. Roebling Sons Co. and the American Steel & Wire Co. was set to work on this problem with the result that not only was a satisfactory cable developed but a steady production was attained, 50,000 feet per week being delivered regularly by the John A. Roebling Sons Co. alone. This new cable consisted of 114 separate wires of special steel besides the telephone center of 3 copper wires properly insulated and armored. The specifications demanded a breaking strength of 7,200 pounds while the actual test of the finished Roebling cable showed 8,250 pounds. Another of the balloon problems was the supply of hydrogen gas. Before the war only a little hydrogen was used in this country, the element being a by-product in the manufacture of commercial oxygen. We met the additional demand for millions of cubic feet of hydrogen for our balloons by establishing Government gas plants and expanding privately owned plants already in existence. There were two methods of supplying hydrogen to our balloon units at home and abroad. One of these was by furnishing portable plants which would generate hydrogen at the place where it was to be used. The other was to take the hydrogen from the stationary plants, condense it by pressure in steel cylinders, and ship it to points of demand. By far the greater part of the gas which we used not only in this country but in France was produced at the permanent supply stations and shipped in cylinders. Each cylinder held about 191 cubic feet of gas under a pressure of 2,000 pounds per square inch at 68° F. temperature. When the war ended we had placed orders for 172,800 of these cylinders, of which 89,225 had been delivered and were in use. We developed a manifold filler which would take the gas from 12 to 24 cylinders at the same time and quickly inflate a kite balloon, a speed of 23 minutes for a complete inflation having been reported from one training camp. In the production of portable hydrogen generators we had to produce not only the machine but the chemicals required in the process. We adopted the ferrosilicon and caustic soda process by which it was possible to produce 10,000 cubic feet of hydrogen per hour in a field generator. There was plenty of caustic soda to be had, but high grade ferrosilicon, a production of large electrolytic furnaces, was scarce, because of its heavy consumption in the steel industry. We procured, however, 2,482 tons of it for our generators, of which over 2,360 tons were supplied by the Electro-Metallurgical Sales Corporation alone. An interesting feature of the gas supply in the field was the use of "nurse balloons." The nurse balloon was simply a large rubberized-fabric bag with a capacity of 5,000 cubic feet of gas. It was used for storage of gas, and the observation balloons were fed from it. We have not received the exact figures of the quantity of gas used by Hydrogen itself, while the lightest of cheap gases, and therefore the one universally used in balloons, has the grave fault of being dangerous to the balloonist. When mixed with the air it is highly explosive, if touched off by a spark of fire or electricity. For years balloonists have dreamed of a gas light enough to have great lifting power, but which would not burn nor explode. There was such a gas known to chemistry, and this was helium, discovered first in spectroscope examinations of the corona of the sun, but later found by chemists to exist rather freely in the atmospheric envelope of the earth. Although one of every 100 parts of air is pure helium, it was not until comparatively recent years that this light nonexplosive gas was discovered in our atmosphere. Now helium was rare and expensive, and until the United States entered the war no one had considered its production as a commercial possibility. Up to two years ago the total world production of helium since its discovery had not been more than 100 cubic feet in all, and the gas cost about $1,700 per cubic foot. It had been discovered that certain natural gases issuing from the ground in the United States contained limited quantities of helium. The question was whether we could extract this helium in sufficient quantities to make its use practical. The Signal Corps, the Navy, and the Bureau of Mines combined in a cooperative plan to develop a practical helium production. By adopting a method of obtaining the helium from liquefied gas produced in the processes of the Linde Air Products Co. and the Air Reduction Co., and also by the Norton process, we attained astonishing success in this enterprise. On the day the armistice was signed we had at the docks ready for loading on board ships 147,000 cubic feet of helium. At its prewar value this gas would have been worth about $250,000,000. On November 11, 1918, we were building plants which would produce helium at the rate of 50,000 cubic feet per day, and the cost of obtaining it had dropped from $1,700 per cubic foot to approximately 10 cents. None of this gas was actually used in the war, but its production by our chemists was hailed as the greatest step ever taken in the development of ballooning. It now seems to have opened a new era in lighter-than-air navigation. In war helium will nullify the incendiary bullet which destroyed so many balloons and airships. In peace it brings the possibilities of new types of construction of dirigible The Army and Navy cooperated in the production of balloons. The Army furnished the balloon cloth to the Navy. Navy balloons had two automatic safety valves for the expanding gas, one on each side of the balloon a third of the way back from the nose and just above the equator; while the Army held to the French and British idea of a single valve in the nose itself. The Navy adopted a Caquot-type balloon which rides at an angle of about 25° to the horizontal and is somewhat smaller than the Army model. The Navy used these balloons as spotters for submarines and mines. They were towed on cables from the decks of war ships, and were connected with the ships by telephone. The use of parachutes with balloons is a comparatively recent development, the man who first successfully descended to earth in a parachute being not only still active and enthusiastic over aerial development, but being in fact the chief inspector of all United States Army balloons and parachutes. This is Maj. Thomas S. Baldwin, known the world over as Capt. "Tom" Baldwin, hero of innumerable aerial exploits of all kinds under all conditions and in all parts of the world, and at present chief of the United States Army balloon inspection. The Yankee balloon observer in France went up to his observation post in the security of knowing that the equipment on which his life depended had been O. K.'d by men who knew the business from beginning to end. The parachute as it is known at the county fair and the parachute used in the recent war were far apart in type, the latter embodying all the improvements that the world's aeronautical experts could add to it. The need for parachutes developed when hostile aviators began shooting down the sausages. At first the one-man parachute was used exclusively, the men in the basket leaping overboard the instant their balloon was fired over their heads. Any delay on their part would be fatal, since the entire bag would be consumed in 15 or 20 seconds and the observer would then be unable to leap out of the falling basket. When the individual parachutes were used, the maps and records in the balloon basket were usually lost. To overcome these difficulties, the designers invented the basket parachute. This was considerably larger than the individual parachute, and to operate it the balloonists pulled a cord which cut the basket away from the balloon entirely. The spreading parachute overhead then floated the basket, with the men themselves and all else it contained, safely and quickly to the ground. Although hundreds and even thousands of parachute jumps occurred during the war, there were few fatalities from this cause. As to the safety of our parachute equipment, the only complaint from the Yankee balloonists at the front was that they were too safe. The man who is escaping from a German airplane nose-diving at him with a machine gun spitting fire is in a hurry and does not wish to be detained by a parachute which floats him too slowly to the earth. In the rigging of each kite balloon there are about 2,000 feet of rope of different sorts. There was a shortage of proper cordage in the United States at first, and the French thought they could furnish this rigging to us. But this attempt proved to be unsuccessful, and we were forced to develop a cordage manufacture in this country of high quality and great quantity. We did this so swiftly that there was no serious delay to the balloon production. Up to November 11, 1918, we produced over 1,000 balloons of all kinds, 642 of these being of the final Caquot type which we adopted. This production included many propaganda balloons for carrying printed matter over the lines into the enemy's country. We supplied several target balloons for gun practice on our aviation fields. We developed new types of parachutes and built acres of canvas hangars for balloons. We produced 1,221,582 feet of steel mooring cable. These are only the major items in the balloon enterprise, and do not include hundreds of others of less importance. The balloon production was one of the most important and successful of all our war projects. Although we had a limited knowledge of the subject in the beginning, our balloons stood the hard test of actual service and could bear comparison in every way with the best balloons of Europe, where the art of balloon building had been in existence for many years. Once our production actually started, we never had any shortage of balloons for our own Army; and soon we would have been in a position to produce the observation balloons for all of the armies fighting Germany, if called upon to do so.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
