BY 0. Chanute.

Special Webmaster's Note

HAVING for a number of years studied the physical principles underlying flight, and having passed in review the experiments of others in a series of articles which eventually swelled into a book, 1 I ultimately reached the conclusion that the contingent compassing of artificial flight by man involved the study of at least ten separate problems, or the devising of means for observing and mastering the conditions enumerated as follows:

  1. The resistance and supporting power of air.
  2. The motor, its character and its energy.
  3. The instrument for obtaining propulsion.
  4. The form and kind of the apparatus.
  5. The extent of the sustaining surfaces.
  6. The material and texture of the apparatus.
  7. The maintenance of the equilibrium.
  8. The guidance in any desired direction.
  9. The starting up under all conditions.
  10. The alighting safely anywhere.

It is probable that some of these problems can be solved in more ways than one, and these solutions must then be harmoniously combined in a design which shall deal with the general problem as a whole, before the best possible result is attained.

I further reached the conclusion that the seventh problem, the maintenance of the equilibrium under all circumstances, was by far the most important, and the first which should be solved; that until automatic stability, at all angles of flight and conditions of wind, was evolved, and safety thereby secured, it would be premature to seek to apply a motor or a propelling instrument in a full-sized machine, as these additions would introduce complications which might be avoided at the beginning.

I seriously doubted, at first, whether automatic stability could be secured with an artificial machine; whether such combinations could be devised, for an inanimate apparatus, as to perform the complicated functions of the life and instinct of the birds, who probably preserve their balance through almost unconscious reflex action of their nerves and muscles. Observation, however, indicated that this might be automatic, requiring no thought under ordinary conditions, and the final conclusion was reached that it might be possible to evolve an artificial apparatus which should afford automatic stability and safety most of the time; that the variations of the wind were the great difficulties to be encountered, that they must be met and overcome, and that perhaps they might be utilized in obtaining propulsion and support, as is daily done by the soaring birds.

I therefore published an article in the "Engineering Magazine" for April, 1896, in which I advised those seeking a solution of the problem of flight to turn their attention to experiments in soaring flight, with full-sized apparatus carrying a man, as the quickest, cheapest, and surest way of ascertaining the exact conditions which must be met in practical flight.

This mode of procedure doubtless involves a certain amount of personal danger of accident. It might be pointed out that the advice is easy to give, but hazardous to follow, and so I further determined to try such experiments myself, so far as my limited personal means would allow.

For this purpose I secured the services of Mr. A. M. Herring, who had tried some experiments of his own. He rebuilt for me his Lilienthal apparatus, with which he had made some gliding flights in 1894, and he also built another full-sized gliding apparatus after a design of my own.

These were completed in June, 1896, and on the 22d of that month we, a party of four persons, went into camp in the desert sand hills on the south shore of Lake Michigan, just north of the station of Miller, Ind., 30 miles east of Chicago.

These sand hills have been piled up by the wind blowing the sand from the beach. They gradually increase in altitude, from a point about 10 miles east of Chicago to the vicinity of St. Joseph, Mich., on the east shore of the lake, where they attain a height of 200 or 300 feet. They occupy a strip two to five miles wide around the south and south-eastern turns of Lake Michigan, and are bleak, bare, and deserted, being entirely incapable of cultivation. North of Miller, Ind., these hills rise about 70 feet above the lake. They are of soft yellow sand, almost bare of vegetation, and face in every direction of the compass, so that almost all directions of wind can be utilized in gliding experiments.

The method of carrying on these adventures is for the operator to place himself within and under the apparatus, which should, preferably, be light enough to be easily carried on the shoulders or by the hands, and to face the wind on a hillside. The operator should in no wise be attached to the machine. He may be suspended by his arms, or sit upon a seat, or stand on a dependent running board, but he must be able to disengage himself instantly from the machine should anything go wrong, and be able to come down upon his legs in landing.

Facing dead into the wind, and keeping the front edge of the supporting surfaces depressed, so that the wind shall blow upon their backs and press them downward, the operator first adjusts his apparatus and himself to the veering wind. He has to struggle to obtain a poise, and in a moment of relative steadiness he runs forward a few steps as fast as he may, and launches himself upon the breeze, by raising up the front edge of the sustaining surfaces, so as to receive the wind from beneath at a very small angle (2 to 4 degrees) of incidence. If the surfaces and I wind be adequate, he finds himself thoroughly sustained, and then sails forward on a descending or undulating course, under the combined effects of gravity and of the opposing wind. By shifting either his body or his wings, or both, he can direct his descent, either sideways or up or down, within certain limits; he can cause the apparatus to sweep upward so as to clear an obstacle, and he is not infrequently lifted up several feet by a swelling of the wind. The course of the glide eventually brings the apparatus within a few feet of the ground (6 to 10 feet), when the operator, by throwing his weight backward, or his wings forward if they be movable, causes the front of the supporting surfaces to tilt up to a greater angle of incidence, thus increasing the wind resistance, slowing the forward motion, and enabling him, by a slight oscillation, to drop to the ground as gently as if he had fallen only one or two feet.

These manoeuvres require considerable quickness and dexterity, yet they are easily learned in a few days, the principal rule to be learned being that the movements to be bodily made are the reverse of those instinctive motions which would occur to catch one's self from falling if walking on the ground. In point of fact, we found that a week's practice sufficed for a young, active man to become reasonably expert in these manoeuvres, and hundreds of glides were made with the several machines, experimented in 1896 under variable conditions of wind, without the slightest personal accident.

As before stated, we went into camp on the 22d of June, 1896. The party consisted of Mr. A. M. Herring, already mentioned, Mr. W. Avery, an electrician and carpenter, Mr. William Paul Butusov, a former sailor, and myself. The tent was large enough to shelter the machines, but we learned in a few days that this precaution was unnecessary, and that they could be safely left exposed to the wind, outside, by tying them down to pegs or to bushes, or even by loading them down with sand. There was a fishing station of two houses within a mile of the tent, from which outside aid might have been obtained in the improbable case of an accident. Miller Station was two miles inland, and, having come through that station with our suspicious baggage, we soon had more visitors than was altogether pleasant in preliminary experiments.

The Lilienthal machine was first set up. It is shown, poised for a flight, in Plate IV., Fig. I. The wings were 20 feet from tip to tip, 7 feet 6 inches in maximum breadth, and measured 168 square feet in surface, with a weight of 36 pounds. Mr. Herring, who had used it before, took the lead in gliding with it.

It was realized from the first that the machine was difficult to handle and to poise in the wind. The variable puffs pelted the apparatus; they occasionally lifted one wing more than the other, or rocked the machine fore and aft, so that a struggle was necessary before a poise could be obtained. Once under way the same action continued, and the operator was compelled to shift his weight constantly, like a tight-rope dancer without a pole, in order to bring the centre of gravity directly under the centre of pressure and to avoid being upset. This, in fact, is the principle of the Lilienthal apparatus. The equilibrium depends upon the constant readjustment of the weight, so as to coincide with the variable position of the centre of pressure due to the shifting direction and force of the wind. Lilienthal, who evolved this machine, so superior to any that had preceded it, was an expert in its use. He made thousands of flights without serious accident; but it is due to those who may desire to repeat such experiments to state here plainly that we found it cranky and uncertain in its action and requiring great practice. If strongly built it was not, however, nearly so hazardous to life and limb as the above statement would seem to imply. The radiating ribs forming the frame of the wings extend downward about as low as the waist of the operator when in flight, and whenever an awkward landing is made, by reason of the apparatus tilting to one side or the other, the ribs on that side are the first to strike the ground. Acting as springs, breaking or not as the case might be, they save the operator from bodily harm even in a descent of 20 feet. These breakages were easily repaired by wiring on wooden splints to the ribs, so that practice could be resumed in a few minutes.

About 100 glides were made with this machine, the longest being 116 feet, and the heights started from were 20 to 30 feet in winds of 12 to 17 miles per hour. Mr. Avery proved an apt pupil, and in the course of a week learned to manage the machine nearly as well as Mr. Herring. Mr. Butusov did not do so well and was upset, but not harmed. I did not venture myself, feeling that I was no longer young and active enough to perform such acrobatic exercises without breaking the apparatus. After it had been broken, mended, tried again, and overhauled a goodly number of times, it was finally decided, on the 29th of June, to discard it, and it was accordingly broken up.

This decision was most unfortunately justified on the 10th of the succeeding August, when Herr Lilienthal met his death while experimenting with a machine based on the same principle, but with two superposed sets of wings. This deplorable accident removed the man who has hitherto done most to show that human flight is probably possible, who was the first in modern times to endeavor to imitate the soaring birds with full-sized apparatus, and who was so well equipped in every way that he probably would have accomplished final success if he had lived.

Having discarded the Lilienthal machine, we next turned our attention to the apparatus after my own design. This was based upon just the reverse of the principle involved in the Lilienthal apparatus.2 Instead of the man moving about, to bring the centre of gravity under the centre of pressure, it was intended that the wings should move automatically so as to bring the movable centre of pressure back over the centre of gravity, which latter should remain fixed. That is to say, that the wings should move instead of the man.

The apparatus consisted in 12 wings, each 6 feet long by 3 feet wide, measuring 144 square feet in area, each pivoted at its root to a central frame, so that it could move fore and aft, this action being restrained by springs. The main frame was so constructed that the wings could be grouped in various ways, so as to ascertain the best arrangement for maximum support and for counterbalancing the effects of wind gusts, if possible. The total wing surface was 177 square feet, and the weight was 37 pounds. Fig. 2, Plate IV., shows the first grouping tested, which was found at once to be reasonably steady, but deficient in lifting power. It was recognized that the wings interfered with each other's efficiency; that the wind was deflected downward by the front wings, so that the middle and rear wings did not afford the same sustaining power as at the front. After making a few glides with this arrangement, a series of changes was tried to ascertain what was the best grouping and the best distance between the wings in order to obtain the maximum lift and the greatest steadiness. The paths of the wind currents in each arrangement of the wings were indicated by liberating bits of down in front of the machine, and, under their guidance, six permutations were made, each of which was found to produce an improvement in actual gliding flight over its predecessors.

The final arrangement to which this series of experiments led is shown on page 53. Five of the pairs of wings had gradually accumulated at the front, and the operator was directly under them, while the sixth pair of wings formed a tail at the rear, and being mounted so as to flex upward behind in flight, preserved the fore and aft balance. It was at once demonstrated that this apparatus was steady, safe, and manageable in winds up to 20 miles an hour. With it about 100 glides were made. The longest of these was 32 feet, in a descending course of about 1 in 4, against a wind of 13 miles an hour; the object constantly in view being not to make long glides, but to study the equilibrium of the machine and the principles which should govern in developing it further. These were found to be that the supporting surfaces should be concentrated at the front and the man placed directly under them; that the lowest wings should be at least 21 feet above the ground; that they should be about two-thirds of their breadth apart vertically, and not less than their breadth apart horizontally, being set so as to present an angle of incidence of 3 to 7 degrees above the horizon when in flight, and that the wings should be pivoted so as to move very easily, the friction upon this first set of pivots having been found entirely too great to permit the wings adjusting themselves easily to the variations of the wind, and the man having had to move his body.

Having ascertained these facts, the experiments were terminated on the 4th of July, and the equipment was sent back to Chicago in order to rebuild the machine.

It may safely be asserted that more was learned concerning the practical requirements of flight during the two weeks occupied by these experiments than I had gathered during many previous years of study of the principles involved, and of experiments with models. The latter are instructive, it is true, but they do not reveal all the causes for the vicissitudes which occur in the wind. They do not explain why models seldom pursue exactly the same course, why they swerve to the right or left, why they oscillate, or why they upset. When a man is riding on a machine, however, and his safety depends upon the observance of all the conditions, he keenly heeds what is happening to him, and he gets entirely new and more accurate conceptions of the character of the element which he is seeking to master.

The fact which most strongly impressed itself upon us was the inconstancy of the wind. It is incessantly changing in direction and in strength. This fact is not new, it has been well shown experimentally by Mr. A. F. Zahm, by Professor Langley, and probably by others, but its effects upon a man-ridden machine must be seen and felt to realize that this is the great obstacle to be overcome in compassing artificial flight. It cannot be avoided, it cannot be temporized with, and it must be coped with and conquered before we can hope to have a practical flying-machine.

One remarkable feature of the wind, however, struck us as hitherto unknown, or at least unmentioned. The wind gusts seem to come in as rolling waves, rotating at a higher speed than the general forward movement. The buffetings which the apparatus received from the wind, while the operator was endeavoring to steady it, preparatory to a flight, seemed to indicate that he was struggling with a rotary billow which produced the fluctuations. Professor Langley has termed these fluctuations "the internal work of the wind," and it is quite conceivable that they should be produced by a revolving motion, striking the surfaces with velocities varying with the distance from the centre of rotation, and producing all the pulsations which have been revealed by the instrumental measurements.

Mr. Herring first called my attention to this feature of the wind, and I have ever since been wondering how I could, for so many years, have been watching smoke curling away from chimneys, steam convolving from trains, or dust and leaves whirling in wind gusts, without realizing that the elastic tenuity of air must perforce produce rotary motions much more active than those which occur in water.

This observation, if confirmed by further investigation, promises to give us a better understanding of the forces to be mastered. There are indications that there is a certain synchronism about these air waves, and that arrangements can be devised, not only to encounter them, but to avail of them in securing propulsion and automatic stability.

Be this as it may, we returned to Chicago much encouraged by the result of these preliminary experiments, with much clearer ideas as to the difficulties to be surmounted, and with good hopes that by reconstructing the machine we could obtain still better performances.

The original twelve-winged machine was reconstructed by pivoting the wings upon ballbearings placed at the top and bottom of wooden uprights fastened to the main frame. The wings at the front were reduced to ten in number, in order to space them further apart without increasing their total height, but one pair was soon taken off, and the required supporting surface was restored by placing a concave aeroplane over the top of the wings. Two pairs of wings, superposed, were placed at the rear, but one pair was taken off after the first few trials, and the apparatus, provided with a rear keel or rudder, assumed the shape shown in Plate VI., Fig. I. The total supporting surface at the front was then 143.5 square feet, the wings at the back measured 29.5 square feet, and the weight was 33 1/2 pounds. The ball-bearings are at the level of the lower and of the third pair of wings from the bottom in the figure, and each set of moving wings, four in number, is connected rigidly by vertical wooden rods and diagonal wire ties so as to move together. Elastic rubber springs at front and rear connect them with the frame and restrain the movements produced by the fluctuations of the wind and relative speed. The detailed construction of the apparatus is shown on Plate VII. It had been originally intended to erect the machine with five pairs of superposed wings at the front, and they were in fact put on, but the first few trials in the wind showed that the height and leverage were too great for easy control, and the top pair was accordingly taken off.

Multiple-wing Gliding Machine, Invented by O. Chanute, C.E.

There was built simultaneously another full-sized machine, based upon a different principle. Instead of having pivoted wings, this consisted of three superposed concave surfaces, stretching 16 feet across the line of motion, by a breadth of 4 feet 3 inches, these surfaces measuring an aggregate of 19l square feet. The lower surface was cut away at the centre to admit the body of the operator. The machine was provided with a combined horizontal and vertical rudder, and its total weight was 31 pounds. The first few trials developed the fact that the sustaining power was in excess, and that the bottom surface was too near the ground. It was removed, leaving the apparatus in the condition shown on Plate VI., Fig. 2. The sustaining surfaces and the rudder were connected by an automatic device, designed by Mr. Herring, for the purpose of securing stability. The curvature of the wings (versed sine) was about one-tenth of the chord. Estimates were made in advance of head resistance due to the framing and to the drift of this machine. It was computed that it required a relative speed of 22 miles an hour and an angle of incidence of 3 degrees for support, and that its angle of gliding descent would be 10 degrees, or 1 in 5.6, which computations were fully verified in the experiments, as will be seen hereafter.

Still a third full-sized machine was constructed at my expense at the same time. This was designed by Mr. William Paul Butusov, who has already been mentioned as being present at the preliminary trials in June, and who stated that he had already tested with success a similar construction some seven years previously. This closely resembled the apparatus experimented by Le Bris in 1855 and 1867. It consisted in a boat-like frame of ribs and stanchions, which might be covered with stout oilcloth and thus transformed into a boat. Above this were four longitudinal keels of balloon cloth, stretched on a frame, each 8 feet long and 3 feet deep. The central space was left open, but the two side spaces were roofed over. This occupied 8 feet in width, and immediately above were placed the wings, each 16 feet long, by a maximum width of 7 feet, tapering to the tips. The total spread was, therefore, 40 feet from tip to tip, and above this again a square aeroplane or kite was placed, hung on trunnions at its centre, so that its angle of incidence might be varied at will by lines carried to the hands of the operator. The latter stood upright in the boat on a running board 8 feet long, and might therefore shift his weight to that extent by walking forward or backward, and he might also shift it about 3 feet sideways by leaning to one side or the other. The whole arrangement is shown in Plate VIII., Fig. 1, except the rudder and tail, which are partly hidden by the man, and which are moved by light lines passing over pulleys and carried to his hands. In addition to this a pair of parallel bars (curtain-poles) were fastened to the frame, to which the man might cling or brace himself.

When finally completed the apparatus spread 266 square feet of sustaining surface and weighed 160 pounds. The various parts (wings, keel-roofs, top aeroplane, and tail) were then tested by suspending them inverted, and loading them with sand to the maximum load they might be called upon to carry, and as some of them showed signs of crippling, or did cripple, they were strengthened with additional material until they were safe to stand the strains. This brought the total weight up to 190 pounds.

These three machines being ready, we again went to the sand hills on the 20th of August, 1 896. Having on the previous occasion found the vicinity of Miller too accessible to the public, we went, this time, five miles further down the beach, where the hills were higher, the solitude greater, and the path more obscure to the railroad, which it reached at a sand-pit station consisting of a single house, and called Dune Park. The distance from our camp was about two miles, through a series of swamps, woods, and hills, so that intending visitors not infrequently got lost.

We went from Chicago by a sailing vessel in order to avoid arousing gossip at the railroad station, and in the afternoon of August 21 st we got the material unloaded and the tent pitched at the experimental hill. We hoped to begin setting up the machines on the morrow.

Unfortunately, that very night a fearful storm and whirling wind came up from the southwest at 3 A.M. It blew the tent to ribbons, blew away and wrecked such wings as were not boxed, while all of the party and the provisions were drenched, the camp equipage being moreover scattered and damaged.3 It became necessary to send at once to Chicago for another tent, which arrived at Dune Park by express in the afternoon of the twenty-second, but this disclosed our presence to the people at the sand pit, and some ten days later brought down the newspaper reporters to see what we were about.

Our party consisted of five persons, Mr. Herring Mr. Avery, Mr. Butusov, already mentioned, Dr. Ricketts,--a young surgeon who found that function such an entire sinecure that he could only exhibit to us his talents in cooking,--and myself. In addition to this there was, for a time, a carpenter to erect the trestle work from which to launch the Butusov machine. The hill selected faced the north and rose 100 feet above the lake, there being an intervening beach of about 350 feet between its base and the water. It was of soft yellow sand with many bare slopes, but with occasional clumps of trees and bushes. To the south it sloped to a bare wilderness of sand.

The first machine which was repaired and set up after the tornado was the aerocurve, with three superposed fixed surfaces and automatic tail attachment. It was first tested on the 29th of August, with tentative glides from a height of 15 to 20 feet above the bottom of the hill, but it was found to rock so that the lower surface struck the ground, hard to manage, and to lift more than required. The lower aerocurve was therefore taken off, thus reducing the sustaining surface to 135 square feet, and the weight to 23 pounds. This was thereafter found ample to sustain an aggregate weight of 178 pounds (23 pounds of machine and 155 pounds of operator), and all the subsequent experiments were made with this arrangement. During the next 14 days scores and scores of glides were made with this machine, whenever the wind served. It was found steady, easy to handle before starting, and under good control when under way,-- motion of the operator's body of not over 2 inches proving as effective as a motion of 5 or more inches in the Lilienthal machine. It was experimented in all sorts of winds, from lo to 3 1 miles an hour, the latter being believed to be a higher wind than any gliding machine had been tried in theretofore, and yet the equilibrium was not compromised, the machine gliding steadily at speeds of about 17 miles per hour with reference to the ground, and of about 20 to 40 miles an hour with reference to the air, or relative wind. On one occasion a relative speed of 52 miles an hour was acquired in a descent. Some of the best glides made were as follows:

in feet.
Time in
Angle of
fallen, feet.
Speed, feet
per second
Descent of
Avery1998.10°34.624.91 in 5.75
Herring2348.77 1/2°30.426.91 in 7.69
Avery253...10 1/2°46....1 in 5.50
Herring239...11°46.3...1 in 5.24
Avery25610.225.525.11 in 7.18
Herring35914.10°62.125.61 in 5.75

One of these flights is shown by Fig. 2, Plate VIII.

The varying flatness of the angle of descent was undoubtedly due to the varying strength of the wind, and also to its ascending trends as it struck the slope of the hill. The latter were exhibited by liberating bits of down at the foot of the hill, whence they would ascend parallel with the surface and pass over the top to the plain beyond. On many occasions the machine and man were raised higher than the starting point by increasing wind velocity, but this action was found to be much too irregular to be availed of as a source of power.

It was found that by moving the operator's body backward or forward, an undulatory course could be imparted to the apparatus. It could be made to rise several feet to clear an obstacle, or the flight might be prolonged, when approaching the ground, by causing the machine to rise somewhat steeply and then continuing the glide at a flatter angle. It was very interesting to see the aviator on the hillside adjust his machine and himself to the veering wind, then, when poised, take a few running steps forward, sometimes but one step, and raising slightly the front of his apparatus, sail off at once horizontally against the wind; to see him pass with steady motion and ample support 40 or 50 feet above the observer, and then, having struck the zone of comparative calm produced by eddies from the hill, gradually descend to land on the beach several hundred feet away.

A few hidden defects were gradually evolved, such as lack of adjustment in the automatic device, and occasional swerving out of the course in sudden gusts of wind; but safe landings were made in every case, by simply throwing the body back and causing the front edge of the aerocurve to rise so as to diminish the speed; and the machine was not once broken. It was kept out of doors moored to pegs driven in the sand, and was injured by storms on but three occasions. It was concluded, however; that a permanent machine of this kind should be arranged to fold up (as this was not) so as to admit of carrying it about and of sheltering it from the weather.

The movable winged machine (12 wings) was not set up till the 4th of September, 1896. Upon being tested, it was found at once that a mistake had been made in not providing entirely new wings for it. The old wings were so racked, twisted, and distorted by their prior service that they did not lift alike, and that it was difficult to poise the machine and to balance it in the wind. Nothing is so important in such experiments as to keep the sustaining surfaces in perfect shape and to prevent any racking when under strains. This is inculcated to us by the birds, who are constantly "pluming" themselves when on the perch. They pass each flying feather through their beaks, repair those barbs which have become separated, rearrange the lap of the feathers, and beat their wings up and down to limber up the muscles. I have reason to believe that it was in consequence of the failure to keep his apparatus in constant rigid good order that Herr Lilienthal so unhappily lost his life. A correspondent in Germany, who had witnessed his exercises two weeks before the fatal fall, wrote me that he had found that in the particular machine with which the accident occurred "the connections of the wings and of the steering arrangements were very bad and unreliable," that he had remonstrated with Herr Lilienthal very seriously, and the latter had promised that he would put the apparatus in order, but, with that contempt of danger which long familiarity and thousands of successful flights is sure to create, it is much to be feared that he did rot attend to it immediately, especially as he was about to discard that particular machine for a new one from which he expected great results.

It was also found that in spacing the wings of the twelve-winged machine further apart, it had been made too high. The top was 10 feet 6 inches above the ground, and the leverage of the wind made it difficult for the operator to control the machine. The top pair of wings was accordingly taken off, and the experiments thereafter made with the apparatus as cut down. In this shape it proved steady and manageable, the flights being over twice as long, with the same fall as with the original machine in June. The following are some of the glides made on the 11th of September against a wind of 22.3 miles per hour:

in feet.
in seconds.
Speed, feet
per second
Herring1487.21.1Angle not measured
Avery1747.622.9" " "
Herring1667.522.1" " "
Avery1837.923.1" " "
Herring1727.822" " "

The angles were approximately 10 or 11 degrees, or 1 in 5.

This machine had been provided with a swinging seat, consisting of network with a narrow board at its front, and with a pair of swinging bars and stirrups against which the legs could be braced, so as to move the wings fore and aft by means of light lines running through pulleys. The heights started from being only 30 to 35 feet above the base of the hill, and the glides being accordingly very brief, these attachments could not be brought into action, but their efficacy was tested by suspending the apparatus between two trees and facing the wind with a man in the seat. It was found, as was expected, that by thrusting the wings forward the machine was tossed up, and vice versa that by thrusting one wing forward the machine turned towards the opposite side, and that these would be effective ways of directing the apparatus when under flight, either up or down, or in a circling sweep. The automatic regulation, however, did not work as well as was hoped, perhaps in consequence of inaccurate adjustment of the springs. The man still had to move about one inch to preserve the equilibrium when under way. The machine made steady flights and easy landings, and was not once broken in action. It is certainly considered safer and more manageable than the Lilienthal apparatus which we tested. No photographs were taken of this machine in flight, as it was not tested nearly so often as would have been desirable, and whenever it was, something always interfered to prevent getting the camera.

It must be confessed that the results with this apparatus were rather disappointing, and yet the principle is believed to be sound. As the variations of the wind are constantly changing the position of the centre of pressure, it is necessary that either the wings or the weight shall move, or that the angle of incidence relative to the air shall be absolutely maintained in order to keep the centre of pressure and the centre of gravity upon the same vertical line. These are the two principles which are involved in the two machines which have herein been described. Which of the two shall hereafter prove to be most effective in practical use, or whether the two can be combined, cannot be determined at present, but it is my judgment that one or two more seasons should be devoted to perfecting the automatic equilibrium, to eliminating hidden defects, and to adjusting the strength of the springs and moving parts, before it will be prudent to apply a motor, or to try to imitate the soaring of the sailing birds.

Towards the last we gathered such confidence in the safety of the machines that we allowed anybody to try them who wanted to. A number of amateurs took short flights, awkwardly of course, but safely. One of them was raised about 40 feet vertically and came down again so gently that he felt no jar upon alighting. Others glided from 70 to 150 feet, and all agreed that the sensation of coasting on the air was delightful, although they were somewhat timid about tempting fate too many times. Any young, active man can become expert in a week with either of these machines.

We performed nothing like continuous soaring with any of the machines. The fluctuations of the wind were entirely too irregular to be availed of; for a wind gust, which tossed a machine up, was almost immediately succeeded by a lull which let it down again. If we had had a long, straight ridge, bare of trees at its summit, and a suitable wind blowing at right angles thereto, we would have attempted to have sailed horizontally along the top of the ridge, transversely to the resulting ascending current. This manoeuvre is frequently and easily performed by the soaring birds over the edge of a belt of trees. They ride across the face of the ascending aerial billow, decomposing its upward trend into propulsion as well as support. The feat should be performable by man, and should, in my judgment, be attempted before circling flight is tried. It requires, of course, that the equilibrium shall be first mastered, and also that the angle of flight shall be flatter than with our machines. This was, as has been seen, from 8 degrees to 11 degrees, or a descent of 1 in 7 or 1 in 5. Now, the soaring birds generally sail at angles of 4 degrees to 6 degrees, or a descent of 1 in 15 to 1 in 9, and hence they lose very much less elevation. This disadvantage in the machines resulted from the increased head resistance due to the framing and spars as compared with the wing edges of the birds, and especially from the fact that in order to give the man easy command over his movements and to let him land on his feet, he has to be in the natural erect position. This produces a body resistance due to about 5 square feet of surface, while it would be that due to only about 1 square foot if the man were placed horizontally, as is the body of the bird. It is probable, however, that the machines can be improved in this respect, and that flatter angles of flight will be obtained than those recorded herein.

The apparatus of Mr. Butusov, like that of Le Bris, had been inspired by watching the sailing of the albatross in southern latitudes. He stated that having begun by experimenting with the main wings, he had been led to add various adjuncts, such as the keels and the top aeroplane in order to improve the stability. It was no part of the original programme to test such a machine, but in view of the degree of success said to have been attained both by Le Bris and by Mr. Butusov, it was determined to give the apparatus a trial.

As it weighed 190 pounds, and the operator's own weight was 130 pounds, a total of 320 pounds, it was necessary to furnish special appliances for launching the machine. This was provided for by building an inclined trestle work, which consisted in a pair of tallowed guides or ways, 8 feet apart, descending at an angle of 23 degrees down the slope of the sand hill selected, the top being 94 feet and the bottom 67 feet above the lake. The last 10 feet of these launching ways was horizontal, and connected with the sloping portion by a curve of 5 feet radius. The ways stood about 11 feet above the side of the hill, the central space between them being entirely unobstructed, the supports being braced by raking posts and braces. The trestle faced due north, so as to avail of the north wind, which, blowing down the whole length of Lake Michigan, arrived with fewer of the whirls and eddies than prevailed with the winds coming from the south, south-east, or south-west. These had been disturbed by blowing over the sand hills, and it is a peculiarity well worthy of note by other experimenters that they will find it much preferable to avail of winds which have traversed across a sheet of water or a level plain than of those which have come over hills, trees, or other obstacles.

This fixed position of the launching ways, however, unfortunately required the waiting for a north wind to blow before experiments could be conducted with this apparatus. The prevailing winds in September were from the south, and there were many storms, so that the instances were rare indeed, during the three weeks which elapsed after the trestle and apparatus were completed, when the wind came from the right direction, and with just the velocity (18 to 25 miles per hour) which was desired. Hence the machine was not given that complete and thorough test which it would have received had the inventor accepted my proposal to launch from ways rigged up on a floating barge, which might have been anchored or towed against any wind of suitable velocity.

Before proceeding with the tests, the whole apparatus was carefully measured. It was ascertained that the whole sectional area of the framing, spars, wing edges, ribs, stanchions, guys, cords, etc., including 5 square feet for the body of the operator, was 44.92 square feet, reduced, however, by reason of the rounding of the parts to an equivalent of 33.28 square feet, which area, multiplied by the pressure, would give the head resistance; that the apparatus would require a relative speed of 25 miles an hour (3.06 pounds per square foot pressure) in order to float it at an angle of incidence of +2 degrees, and that, therefore, if Lilienthal's coefficients were used, the total resistances would be:

Head resistance,33.28 sq. ft. X 3.06 lbs. =101.83 lbs.
Tangential component,266 sq. ft. X 3.06 lbs. X 0.008 =6.51 lbs
Retarding component,320 Ibs. X (sin 2°=0.035) =11.20 lbs
Total119.54 lbs

So that the angle of descent might be expected to be:

320 lbs./120 lbs. = 1 in 2.67 or 22 degrees.

These calculations were closely verified, as in the case of the other two machines.

It was the 15th of September (1896) before a proper wind served. It then set in from the north about noon and blew 28 miles an hour. The apparatus was accordingly placed in the ways, tested as to fit by running it up and down restrained by head and tail ropes, and then it was placed upon the level portion of the ways facing the wind. Additional guy lines were fastened to the wings, and Mr. Butusov got into the machine. The guy lines were manned, and the apparatus was suffered to rise 2 1/2 to 3 feet above the ways, in order to test its balancing and the degree of control of the operator over its movements.

This appeared to be complete. A very slight step to the front or rear sufficed to depress or to raise the head of the machine, and the side motions were equally sensitive. The support was found to be ample from a 28-mile wind, and it was apparent that the great range of motion provided for the operator would give him command of the machine at all angles of incidence. The apparatus was then hauled down by the guy lines and settled back upon the ways squarely, resting thereon by means of four sliding shoes projecting from the machine on a line with the top of the boat-like body. It is shown in that position by Fig. 1, Plate IX.

It was desired next to launch it in ballast, and also to test it as a kite, and preparations were begun for that purpose; but a small rip having been discovered in one of the wing coverings, and a buckling in one of the braces, it was thought more prudent to repair these before proceeding further, and the machine was removed from the ways.

The wind changed to the south-west in the night, but on the 17th it again blew from the north, with a speed, however, of but 12 miles per hour. In the hope of its freshening, the machine was got into the ways, loaded with 130 pounds of sand in bags, and rigged as a kite, by fastening a bridle to the keel of the boat and leading therefrom a long rope passing through a pulley fastened to a post in the sand, 250 feet away on the beach. This rope was handled by four men, with instructions to run with it so as to take up the slack as soon as the apparatus left the ways. Four guy lines, hanging down from the front and rear, and from each wing of the machine, were likewise manned, in order to control the movements of the kite in case of need.

All being ready, the restraining line was cut and the machine slid down the tallowed ways and took the air fair and level. It went horizontally some 20 feet, but its motion was then checked by the friction on the sand of the kite line, which the crew, gazing open-mouthed at the sight, failed to haul in as the machine flew forward. This check was sufficient to overcome the initial velocity proper to the machine, and the wind (12 miles an hour) was insufficient to sustain it. The apparatus glided downward and landed squarely on its keel about 100 feet from the end of the ways, a descent of about 1 in 2. The tip of one wing struck the hillside, but no harm was done as it flexed. Some three or four of the stanchions of the boat frame were, however, broken. These were replaced in two hours, but the wind had fallen so light by that time that the experiment could not be repeated.

To-test the apparatus properly a north wind of about 25 miles per hour was required. This did not set in again till just before the advancing season compelled the breaking up of camp and returning to the city. On the 19th of September the equinoctial storm set in and blew from the north-west 56 to 60 miles an hour. Another gale of 60 miles an hour blew on the 22d, accompanied by heavy rains. These were followed by southerly winds, so that it was the 26th before the machine could be tested again. A wind then set in from the north-east, with a speed of 18 miles an hour, and although this was quartering, instead of dead ahead as was desired, it was determined to launch the apparatus. This was first attempted with the operator in the machine, but as the quartering wind greatly increased the friction of the launching ways and diminished the required initial speed, the operator was replaced by 90 pounds of sand in bags, and a rope was fastened to the front of the machine in order to increase its velocity by pulling thereon. The apparatus went off, but as soon as it had fairly left the end of the ways, the quartering wind swerved the head of the machine around, and it took a descending north-westerly course, describing a curved path. The tip of the left wing then struck the top of a tree, swinging the machine around further, and then this same wing struck the hillside and was broken. The machine then fell to the ground, landing upon its keel about 75 feet from the end of the ways, and a number of ribs and stanchions were broken, so that the repairs, if made, would probably have occupied a day or two.

It was evident that the machine was moderately stable; that on neither this nor on the previous trial would the operator have been hurt if he had been in the machine; but it was also evident that the apparatus, as then proportioned, glided at too steep an angle to perform soaring flight; that it would lose so much altitude when going with the wind that the loss would not be recuperated when turned to face the wind. It was recognized that this, as well as the other two machines, could be modified so as to materially reduce the head resistance and thereby flatten the angle of descent, but the season was so far advanced, the weather so inclement, that it was decided to break up camp and to return to the city. This was done on the 27th of September.

Such were the experiments. They occupied an aggregate of seven or eight weeks in the field, they were carried on without the slightest accident to the operators, and they made manifest several important conclusions. The first is that it is reasonably safe to experiment with full-sized machines, if the methods and writings of Lilienthal be previously studied. The second is that experiments with full-sized machines, carrying a man, are likely to be more instructive and fruitful of eventual progress than experiments with models. The third is the inference that it is probably possible to evolve an apparatus with automatic stability in the wind, but that in order to do so, there must be some moving parts, apart from the man, in order to restore the balance as often as it is compromised. The fourth conclusion is that the problem of automatic stability will be most easily worked out with a light apparatus, so light as to enable the operator to carry it with ease, and so arranged as to enable him to use his legs in landing. The fifth conclusion is that it will require a good deal of experimenting to adjust the working parts, to regulate the springs, and to discover hidden defects, before it will be quite safe to try to perform soaring feats in the wind. The sixth is that the incessant fluctuations of the wind, which so very greatly complicate the problem of maintaining automatic stability, probably result from the rotary action of its billows, and future experimenters are urgently advised to study this action and to endeavor to meet it.

A word or two of caution may also be given. It is best to begin experimenting with a new machine in short and low gliding flights over bare and soft sand hills, but more ambitious flights and soaring feats should be attempted first over a sheet of water to mitigate the fall should anything go wrong. Experiments should not be tried in high or gusty winds, and the apparatus should be frequently examined and kept in constantly perfect order. Wire stays should be employed as sparingly as possible. Not only do they vibrate when the machine is under way, and so increase the resistance, but they get loose and allow the apparatus to become distorted. It is well to fly a model of a projected apparatus as a kite, but it does not follow that a satisfactory kite will make a good flying- machine, because the required angles of incidence are so different. A good kite will fly steadily at an angle of 20 or 30 degrees with the wind, but a good flying-machine needs to fly at an angle of 2 to 5 degrees to reduce the drift to the lowest possible.

I do not know how much further I shall carry on these experiments. They were made wholly at my own expense, in the hope of gaining scientific knowledge and without the expectation of pecuniary profit. I believe the latter to be still afar off, for it seems unlikely that a commercial machine will be perfected very soon. It will, in my judgment, be worked out by a process of evolution: one experimenter finding his way a certain distance into the labyrinth, the next penetrating further, and so on, until the very centre is reached and success is won. In the hope, therefore, of making the way easier to others, I have set down the relation of these experiments, perhaps at tedious length, so that other searchers may carry the work of exploration further.

1 "Progress in the Flying Machines," M. N. Forney, N.Y., Editor, 1894.
2 To establish priority of invention a patent has been applied for.
3 The frying-pan was blown 200 yards away.

Webmaster's note:This article originally appeared in 1897 in The Aeronautical Annual by James Means, pp. 30-53. Scan by Cory Kotowsky. HTML and photos by Gary Bradshaw.

Special Webmaster's Apology: As usual, I'm behind on photo scans. Sorry that only a few are in place at the moment. Stand by for later additions, though I'm really far behind at the moment, so don't hold your breath.