Having thus passed in review the-various attempts which have hitherto been made to compass artificial flight there remains the task of pointing out as briefly as possible whether and how the information gathered may be made to conduce to a possible solution of the problem of aviation.

It was thought more effective to bring out the various theories of flight, and my own views, while describing the experiments, rather than to present them in a series of abstract statements and propositions, the immediate bearing of which might not be so evident. The reader has probably reached deductions of his own; but he may also wish to know my own general conclusions, and in what manner if any the many failures which I have described can be made to subserve eventual success.

These failures have resulted from so many different causes that it is evident that many conditions must be observed. These conditions virtually each constitute a separate problem, which can probably be solved in more ways than one and these various solutions must then be harmoniously combined in a design which shall deal with the general problem as a whole. These various conditions. or problems, as I prefer to call them, may be 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.

Analyzed and viewed in this way, the reader may realize how complicated is the question and how formidable are the various difficulties which are to be surmounted. And yet the scrutiny which has been made of the various experiments attempted and of the progress accomplished in flying machines enables us to perceive that many of these problems have been approximately solved, more particularly since 1889 and that a better understanding of the difficulties to be overcome has been obtained concerning several others.

I. The first problem to be considered is that pertaining to the resistance and supporting power of air. By the use of currently accepted formulae it could not be figured out a few years ago how birds were supported in flight. Now that Professor Langley's experiments have confirmed many of those previously tried, we are enabled to say that the empirical formula of Duchemin (from which the table of Ïlift and" drift" herein given was calculated) is approximately correct, and to figure out the support and the resistance with some confidence of not going far wrong.

These calculations seem to indicate that artificial flight is possible, even with planes; that very flat angles of incidence, from 2° to 5°, hitherto considered inadmissible, will be the most advantageous, and that within certain limits of hull resistance high speeds will require less power than low speeds, because they admit of obtaining support from the air at a flatter angle.

We have seen that the "drift" diminishes as the angle of incidence becomes less, that the Ïhull resistance" (including car, framing, braces, etc.) increases as the square of the speed, and that the skin fraction is so small that it may for the present be disregarded, and we are enabled to calculate, approximately at least, the power required to obtain support in flight with planes, and to overcome the resistance, although we are not yet aware what limit will be imposed upon the size of artificial apparatus by the law that the weight will increase as the cube, while the sustaining surfaces will grow only as the square of the similar dimensions

Moreover, the formulae which give this promise of success were derived from experiments with plane surfaces, and we already know that concavo-convex surfaces will be still more effective, although the most favorable shapes are not yet ascertained. This statement indicates the direction in which scientific investigation and experiment should now proceed, and holds out the hope that this first problem is in a fair way of being solved.

2. The second problem--that concerning the motor to be employed--has justly been considered to be the most important and difficult of solution. It seemed hopeless to rival, with an artificial motor, the output of energy appertaining to the motor muscles of birds in proportion to their weight, which, as we have seen, there is good reason to believe develop work in ordinary flight at the rate of 1 H.P. to 20 lbs. of weight, and can for a brief period, in rising, give out energy at such rate as to represent an engine of only 5 or 6 lbs. of weight developing 1 H.P.

The writer has, on a former occasion,42 passed in review the comparative weights of various classes of engines. He found that the lightest engines in use in 1890 including the generator of power, weighed 60 lbs. per H.P. for steam, 88 lbs. per H.P. for gas engines, and 130 lbs. per H.P. for electric motors. He intended to discuss the subject further in this account of ÏProgress in Flying Machines," but recent achievements with steam-engines seem to make this unnecessary. Marine (yacht) engines have been reduced more than one half in weight, Mr. Hargrave has produced a steam-engine weighing 10.7 lbs. per H.P.; M. Maxim has created one weighing but 8 lbs. per H.P., including a condenser, and other experimenters are approximating closely to the same weights.

Steam-engines, therefore, seem to have been so much reduced in weight as to admit of their being employed as motors for flying machines. This may not be a final solution, for it may be that some form of gas or petroleum engine will prove to be still better adapted to aerial purposes. as indeed has been already hinted by M. Maxim but in any event, his steam-engine seems to be light enough to make a beginning of artificial flight, if the other problems pertaining thereto can also be solved.

But it is possible to utilize a still lighter power, for we have seen that the wind may be availed of under favorable circumstances, and that it will furnish an extraneous motor which costs nothing and imposes no weight upon the apparatus.

Just how much power can be thus utilized cannot well be told in advance of experiment; but we have calculated that under certain supposed conditions it may be as much as some 6 H P. for an aeroplane with 1000 sq. ft. of sustaining surface; and we have also seen that while but few experimenters have resorted to the wind as a motor, those few have accomplished remarkable results.

3. As regards the selection of the instrument through which propulsion is to be obtained, we have seen that experiment has shown that reaction jets, whether obtained from explosives, steam, or blasts of air; that wave action; that valvular, folding or feathering paddles or vanes have all proved inferior in practical application to screw propellers or to propelling wings, and that the two latter (if we are to judge from Mr. Hargrave's experiments) are about equally effective. It being understood, however, that this statement refers to wings only as propelling instruments and not as sustaining surfaces. We may conclude, therefore, that the third problem may now be solved either with screws or with waving wings, as best conforms to the rest of the design.

4. This brings us, therefore, to consider the solution of the fourth and important problem of what kind or form of apparatus should be selected for sustaining the weight-- whether flapping wings, screws, or aeroplanes. The best measure of comparison will be the weights or number of pounds which experiment shows may be sustained per H.P. with each form, considered in connection with the weight of the construction required to make that form abundantly strong against the resulting strains. The difference between the two weights will indicate the proportion of the whole which may be devoted to the motor. It is desirable, therefore, to consider each form or kind of apparatus separately.

We do not yet know accurately how many pounds per H.P. can be sustained in horizontal flight with a bird-like apparatus of flapping wings. The toy birds which have been described support only from 6 to 20 lbs. per H.P., but this inefficiency is largely due to the undue friction of the working parts and to the abnormal head resistance of the framing in such imperfect models. The writer has estimated that in the case of a flying pigeon about 77 lbs. are sustained per H.P., but as this is partly based on conjecture, it may be an underestimate.43

Upon the whole, the writer is inclined to admit that about 100 lbs. per H.P. may be sustained with flapping wings, this including the power required both to support the weight and to overcome the head resistance. He believes, moreover, that in an artificial machine of sufficient size to sustain one man, the strength required to resist the constant reversals of strains due to the alternating motion of the wings will involve such dimensions that the weight of the apparatus and man will amount to at least three-fourths of the whole, thus leaving but one-fourth of the total weight which can be devoted to the motor and its adjuncts, including the fuel and supplies for the journey.

Concerning aerial screws we have abundant experimental data. Nadar, Wenham and Freninges each obtained a sustaining effect of 33 lbs. per H.P.; Dienaide realized 26.4 lbs., and Dahlstrom and Lohman secured 37.6 and 55 lbs. per H.P., while Renard obtained from 17 to 48 lbs. thrust by screws rotating at various speeds, and Moy recorded 40 lbs. per H.P. sustained from a wind wheel with vanes of variable pitch.

These performances, however, included a certain amount of ascension, which absorbed part of the power, so that probably we shall be quite sale in assuming that in mere horizontal flight some 45 lbs. per H.P. can be sustained with screws.

As the strains in a rotating apparatus will be less destructive than those involving reversals of motion, it seems probable that screws may be constructed with a less weight of materials than flapping wings of the same sustaining power. It is judged that an apparatus can be constructed to sustain the weight of one man with rotating screws, in which only about two-thirds of the weight shall be absorbed by the framing, screws, car and man, thus leaving one-third of the whole weight for the motor and its various adjuncts. The practical result of this estimate will be elicited further on.

We have also a number of experimental data concerning aeroplanes. Professor Langley sustained a maximum of 209 lbs. per H.P. with planes at an angle of incidence of 2° and M. Maxim sustained 133 lbs. per H.P. at an inclination of 1 in 14. These data apply to the plane only. Neither of these performances included the head resistance due to the framing and car which are indispensable in an actual machine, so that we must derive our premises from complete models. With one of the latter Tatin sustained 110 lbs.; Phillips, 72 lbs., and Hargrave, 89 lbs., 76 lbs., and 79 lbs. per H.P. in horizontal flight. We may safely conclude, therefore, that 100 lbs. per H.P. can be sustained in horizontal flight with an aeroplane.

As the latter consists of fixed surfaces receiving no strains save the sustaining pressure of the air, it is believed that such class of apparatus can he constructed of sufficient size to sustain one man, so that about one-half of the whole weight shall be devoted to the apparatus and man, and the other half to the motor and its adjuncts.

These estimates of the proportion of the sustained total weight which can be spared for the motor, are necessarily mere estimates made in advance of actual testing, and (for reasons to be stated hereafter) upon the smallest size of apparatus practicable for actual man flight, yet they enable a comparison to be made between the various forms of apparatus which have been herein described. The result is as follows:

Kind of ApparatusPounds sustained
per H.P.
Proportion available
for motor
Resulting possible
weight of motor
per H.P.
Screws451/3 15 lbs.
Wings1001/4 25 lbs.
Aeroplanes1001/2 50 lbs.

The above table, based as it is upon experimental data of weights actually sustained, indicates that aeroplanes are probably the best forms to experiment with, because they admit of a larger proportion of the whole weight being appropriated to the motor. It also indicates the possibility of success in artificial flight with motors weighing 10 or 15 lbs. per H.P., provided that the remaining problems be also solved; but it must not be overlooked that more power will be required in rising from the ground than in horizontal flight and that the actual proportion of the total weight available for the motor, although conservatively estimated from the best data available,44 is still a matter to be proved by experiment.

The common basis which has been here selected for comparison is that size of apparatus sufficient to support the weight of one man. This is the smallest which can be adopted, and it is theoretically the most favorable, for inasmuch as the weight of the framing will presumably increase as the cube of the dimensions, while the sustaining surfaces will increase as the square of these same dimensions, it is seen that the ratio of the total weight sustained which can be spared for the motor will not be constant, but that the larger the apparatus the more it will weigh in proportion to its surface and the less there will remain for the engine and its adjuncts. Flying machines, therefore, should preferably be designed as small as practicable, and experimenters will place themselves at a disadvantage if they construct large machines.

5. As regards the fifth problem--the amount of the sustaining surface required--it depends on the speed, and it is probable that, within certain limits, no particular extent (in ratio to the weight) can be said to be absolutely the best, because a large part of the resistance will consist in the Ïdrift," and the latter is independent of the area of the sustaining surface; a small area at high speed being able to sustain as much weight as a larger area at a corresponding lesser speed, as indeed is indicated by the formula already given for the drift: R = W tang. @, in which the element of surface disappears.

Practically, however, the weight of the necessary framing and the hull resistance will determine the ratio of surface to weight which will be most advantageous. We have seen that encouraging experiments have been made with surfaces varying from 0.75 sq. ft. per pound in the case of Herr Lilienthal to 7 sq. ft. per pound in the case of M. Hargrave. It seems probable that the latter is in excess, and that it would be preferable to confine the dimensions of artificial machines within the proportions which obtain with fast flying birds, as shown in the table heretofore given, this being from 3.62 sq. ft. per pound in the case of the swallow, to 0.44 sq. ft. per pound in the case of the male duck, with which areas, if we consider their wings as planes, and the angle of incidence to be 3° the swallow requires a speed of 23.12 miles per hour and the duck a velocity of 66.2 miles per hour to sustain their weight.

To come down safely, at the speed of the parachute, requires about the ratio of the swallow, while the proportions of the duck are more favorable to high speed. As the drift will increase only if the angle of incidence be increased, it would seem preferable to maintain this angle as uniform as possible, and to provide variable supporting surfaces to be folded or unfolded with variations of speed, if such a construction can be devised in connection with the concavo-convex surfaces which have already been mentioned as likely to give the most satisfactory results.

6. The sixth problem cannot be said to be solved, for there is considerable uncertainty concerning the best materials to be employed for the framing and for the moving parts; or what should be the texture of the sustaining surfaces in an actual flying machine. Hitherto the main question has been to construct a model which would fly at all; and experiments with models have not thrown much light on the question of materials. If a partial success be realized, this problem will assume greater importance.

It involves considering materials from a somewhat new point of view, or investigating their strength and stiffness per unit of weight, so as to secure a maximum of resistance with a minimum of weight. The quill of a bird's feather is stronger and more elastic than an equal weight of steel, and the texture of its barb is peculiar.

It now seems probable that bamboo, the lighter of the stiff woods, and some varieties of steel, will be found to be the preferable materials for the framing. Contrary to popular belief; aluminium is inferior to steel per unit of weight, particularly in compression, but it does not corrode and may be preferable on that account. It may be utilized for the sustaining surfaces, either as thin sheets or as wire gauze made smooth by some coating; but textile fabrics will probably be the first to be employed for full-sized apparatus. One important requirement, however, is that the surfaces shall not unduly change their shape under varying air pressures. They must be rigid, and, perhaps, elastic, and the fluttering of textile fabrics is likely to give trouble to experimenters. It may be, therefore, that thin wood, parchment, or pasteboard may prove preferable, the latter being corrugated lengthwise of the direction of motion in order to gain stiffness.

The barb of a feather is smooth in one direction and asperous in the other; and it is possible that a similar texture of surface may prove of advantage in flying machines, but this probably will not be determined until partial success has been achieved with an apparatus of sufficient size to sustain the weight of a man.

7. The problem of the maintenance of the equilibrium is now, in my judgment, the most important and difficult of those remaining to be solved. It has been seen, from this review of "Progress in Flying Machines," that almost every failure in practical experiments has resulted from lack of equilibrium. This is the first requisite thing to secure, for, as has already been said, safety is the most important element of success--safety in starting up, in sailing, and in coming down.

If a flying machine were only required to sail at one unvarying angle of incidence in calm air, the problem would be much easier of solution. The center of gravity would be so adjusted as to coincide with the center of pressure at the particular angle of flight desired, and the speed would be kept as regular as possible; but the flying machine, like the bird, must rise and must fall, and it must encounter whirls, eddies, and gusts from the wind. The bird meets these by constantly changing his center of gravity; he is an acrobat, and balances himself by instinct; but the problem is very much more difficult for an inanimate machine, and it requires an equipoise--automatic if possible--which shall be more stable than that of the bird.

We have seen from the experiments described that the transverse stability can be procured in two ways: (1) by placing the two halves of the sustaining surfaces at a diedral angle to each other, and (2) by adding a longitudinal keel to the apparatus, as in the case of Mr. Boynon's fin kites. The mode of action is practically the same for both, and consists in producing increased air pressure upon the side which tends to dip downward. The two may be employed conjointly, but the keel will produce less head resistance to forward motion than the diedral angle, which resistance, however, may be diminished by turning upward only the outer ends of the sustaining surfaces in a manner similar to the upbending primary feathers of the soaring birds.

Longitudinal stability may be promoted in three ways: (1) By additional surfaces at a slight angle to the main sustaining surface, (2) by placing several surfaces behind each other, (3) by causing the center of gravity always to coincide with the center of pressure. The first way corresponds to the method which has been mentioned as procuring transverse stability by means of surfaces at a diedral angle; it is illustrated by M. Maxim's aeroplane, in which two such surfaces are affixed, front and rear; and by M. PÈnaud's aeroplane, in which but one is affixed in the rear. The second way is illustrated by M. Brown's bi-planes and by M. Hargrave's cellular kites; and the third is the method universally employed by the birds.

For an artificial machine this last method is as yet an unsolved problem. Several inventors have proposed methods of shifting weights to change the position of the center of gravity as the apparatus changes its angle of incidence but none of these are automatic, and none have been tested practically.

8. The guidance in a vertical direction--i e., up or down, depends in a great degree upon success in the changing the center of gravity which has just been alluded to. It may be partly effected by changes in the speed or by horizontal rudders, but in such case the equilibrium will be disturbed. Guidance in a horizontal direction has been secured, as we have seen in several experiments, by vertical rudders; but there are probably other methods still more effective, although their merits cannot be tested until a practical apparatus is experimented with. Upon the whole, this problem may give trouble, but it does not seem unsolvable.

9. A really adequate practical flying machine will hardly be said to have come into existence until it possesses the power of starting up into the air under all conditions This problem is as yet unsolved, and may not be until the other problems have been worked out to a success. It is clear that in rising upward more power will be required than in horizontal flight; for to the force required to obtain horizontal support must be added that required to ascend, and the latter will vary with the rapidity of the upward motion. Three principal methods have been experimented with: (1) By acquiring speed and momentum on the ground; (2) by the reaction of rotating screws; (3) by utilizing the force of the wind. The first we have seen to require the use of special appliances, such as railway tracks, so that its application must be limited, and the third necessitates that the wind shall blow, and with sufficient force; either or both may be utilized with the earlier types of practical machines should one or more be hereafter developed but the writer believes that the second method--that of rising through the reaction of a screw --will eventually supersede the two others. It will involve the difficult design of a simple form of sustaining surfaces which can be alternately rotated as a screw or held as a fixed aeroplane when sailing, the change being effected while under motion in the air.

The writer does not believe that a bird-like machine can rise into the air, under all conditions, by flapping its artificial wings. It would need to be already up some distance to permit such action. Birds spring up three or four times their own height, or run against the wind to acquire speed, and with vigorous flaps of wing they rise at an angle seldom greater than 45°, but their initial action would be quite impracticable to a machine of sufficient size to sustain the weight of a man.

10. The alighting safely anywhere is also an unsolved problem, and one, as will readily be perceived without argument, of vital consequence. It has been slurred over by most of the designers of flying machines, and the best method which has been thus far proposed involves the selection of a smooth, soft piece of ground and the alighting thereon at an acute angle. When it is considered that the speed required for support will be somewhere from 20 to 40 miles an hour, it will be realized that the performance will be somewhat dangerous, and that it would be preferable, if the design of the apparatus will admit of it, to imitate the manoeuvre of the bird who stops his headway by opening his wings wide, tilting back his body, and obtaining the utmost possible pressure and retardation from the air before alighting upon the ground. This would require for an artificial machine a rapid change of the center of gravity so as to tilt the apparatus backward to the angle of maximum lift (about 36° by the table) and, immediately thereafter, a counter change of the center of gravity, so as to bring the apparatus back upon an even keel in order to alight at the diminished velocity.

This manoeuvre is not as difficult and dangerous as may at first sight appear; but it must be acknowledged that it would be preferable to utilize the reaction of a rotating screw to diminish the forward motion and to hover over the ground before alighting. This involves the same difficult design which has been alluded to as desirable for use in rising, for it does not seem practicable, within the requisite limits of weight, to provide two sets of sustaining surfaces, one set to he used in rising and in alighting, and the other to serve in horizontal flight. These last two problems--the rising and the alighting safely, without special preparation of the ground--seem very difficult of solution, and are probably the last which will be worked out.

The general problem having been thus decomposed into its several elements, and each element considered as a separate problem, it will be seen that the mechanical difficulties are very great; but it will be discerned also that none of them can now be said to be insuperable, and that material progress has recently been achieved toward their solution.

The resistance and supporting power of air are approximately known, the motor and the propelling instrument are probably sufficiently worked out to make a beginning; we know in a general way the kind of apparatus to adopt, its approximate extent and required texture of sustaining surfaces, and there remain to solve the problems of the maintenance of the equilibrium, the guidance, the starting up, and the alighting, as well as the final combination of these several solutions into one homogeneous design.

In spite, therefore, of the continued failures herein recorded, it is my own judgment, as the result of this investigation into the Ïprogress in flying machines," more particularly the progress of late years, and into the recent studies of the principles and problems which are involved, that, once the problem of equilibrium is solved, man may hope to navigate the air, and that this will probably be accomplished (perhaps at no very distant day), with some form of aeroplane provided with fixed concavo-convex surfaces, which will at first utilize the wind as a motive power, and eventually be provided with an artificial motor.

The conclusion that important progress may be achieved without an artificial motor was little expected when this investigation was begun; but the study of the various experiments which have been passed in review, the perception of the partial successes which have been accomplished with soaring devices, and the general consideration of the subject, have led to the conclusion that the first problem which it is needful to solve is that of the equilibrium, and that in working this out the wind may furnish an adequate motive power.

Preliminary experiments will, of course, be tried upon a small scale, but no experiment with a model can be deemed quite conclusive until the same principles have been extended to a full-sized apparatus capable of sustaining a man, and until this has been exposed to all the vicissitudes of actual flight. It will readily be discerned that a less achievement than this would not prove an adequate performance, and that no matter how well a model might behave in still air, there would still remain the questions as to how it would behave in a wind, and how it was to solve the problems of starting up and of alighting.

It would seem, therefore, that the first problem to solve is that of the maintenance of the equilibrium at all the angles of incidence required; in rising, in sailing, in encountering wind eddies, and in alighting.

For this purpose, it is now my opinion, based upon the performance of the soaring birds and upon the partial success of some soaring devices, that the problem of equilibrium can best be solved with an apparatus which shall utilize the wind as a motive power--i.e., with some form of aeroplane of sufficient size to sustain a man, with which the operator shall endeavor to perform the various manoeuvres required to meet the varying conditions of actual flight, and to preserve at all times his balance in the air. In other words, a flying machine to be successful must be at all times under intelligent control, and the skill to obtain that control may be acquired by utilizing the impulse of the wind, thus eliminating, for a time at least, the further complications incident to a motor.

But whether a soaring device be first experimented with, or whether the initial apparatus be provided with a motor, the next question pertains to the conditions under which a. machine carrying a man can be experimented with most safely.

Various methods have been suggested, and a few have been tried. The most obvious is to suspend the apparatus from a cable stretched between two tall masts or between two steep hills. This has been proposed many times, but we have seen by the experience of M. Sanderval that it does not afford sufficient length of suspending rope to permit of unimpeded manoeuvres, and that experience gained in that way would scarcely be available in free flight. A preferable plan has been proposed by M. Duryea (and probably by others), which consists in suspending the apparatus to be experimented with from a captive balloon, anchored by several divergent ropes so as to remain a half mile or thereabouts from the ground, as shown in fig. 83. By means of a rope passing through a pulley block attached to the balloon, and thence to a windlass on the ground. the machine to be experimented with may be drawn into the air to a sufficient height to clear the gusty air conditions found at or near the ground, and there, in comparative safety, the sky-cycler might manipulate his devices, ascertain the effect of various manoeuvres, and gradually gain control, skill, and confidence preparatory to trusting himself to actual flight.

FIG. 83. -- DURYEA'S PROPOSAL -- 1893

This method is understood to have been employed by M. C. E. Myers, the aeronautical engineer, in experimenting with parachutes, and to have given promise of satisfactory results within certain limits. It is well worth testing as a preliminary trial of a flying apparatus, but it should be remembered that a machine suspended from a rope, however long, will not be under quite the same circumstances as in free flight. Even if it rises upon the wind and is wholly supported thereon it will still be hampered by the rope, and perhaps restrained from some action which it is important to understand in order to maintain the equilibrium, so that the operator will never be quite certain that he has gained complete control over his apparatus.

Other methods have been proposed by various writers. M. Ch. Weyher for instance, in the AÈronaute for July, 1884 suggested the construction of a circular railway of 600 to 1000 ft. diameter, upon which a large platform car, covered with a soft mattress, should carry the apparatus to be tested, attached with restraining ropes about breast high. This car to be towed at varying speeds by a locomotive, so as to afford a sustaining effect and to encounter the wind at various angles, until the operator shall master the necessary manoeuvres.

M. A. Goupil, on the other hand, proposes a circular elevated railway consisting of a single central girder suspended by wire ropes between two rows of posts, and serving to carry a truck to which the apparatus to be experimented with may be suspended. In this case the machine might be provided with its own motive power, or towed by a wire rope, or driven by an electric motor, but in either case there would still remain the restraint of the safety suspending rope, which, as previously suggested, might vitiate the various air reactions which it is important for the operator to experience practically.

These, and other devices which may be suggested, may doubtless prove useful in making the preliminary trials with various forms of apparatus, thus testing their behavior when restrained, but there will always come a time when such apparatus, if apparently adequate, must be tried at full liberty and encounter all the contingencies of free flight. It seems clear that after the preliminary trials with models have been made, time may be saved in ascertaining the full merits of a device and in improving it, if experiments with the full-size apparatus be made at entire liberty instead of under restraint, provided adequate precautions be taken to avoid serious injury in alighting.

Referring to the various experiments which have been made with full-sized apparatus, more particularly those of Dante, Le Bris, Mouillard, Lilienthal, and Montgomery, it is seen that Dante adopted the more rational plan of all by experimenting over a sheet of water, although the exact method he pursued is not known.

Upon the whole, the best mode of procedure is probably that proposed by Le Bris, which want of means prevented him from adopting--that is to say, to start from the deck of a steam vessel under way, so as to obtain initial velocity, as well as to face the wind from whatever direction it may blow, and to be quickly picked up after alighting. If the machine be provided with a light buoy and line, and the operator be encased in a cork jacket or life preserver, he may thus quickly put to the test the merits and the deficiencies of his apparatus with but little danger to himself, and ascertain whether it can be brought under control. The machine may experience breakages, the operator will doubtless suffer many duckings, he may even be stunned at times, but he is not likely to lose his life or to break a limb, as he might do were he to experiment over land.

It is believed that salt water is preferable to fresh water, over which to carry on such experiments, not only because of the greater buoyant power of the water, but especially because sea breezes are more regular and less gusty than land breezes. It is evident that it would be preferable to operate over a genial or a tepid sea, in trade-wind regions if possible, and in locations where steady sea breezes of no great intensity may be relied upon to blow almost daily. It would be desirable to select the vicinity of some projecting tongue of land or of some isthmus, where captive preliminary tests may be made, and also that there should be a cliff in the neighborhood whence models and perhaps the apparatus itself might be floated off. There are many such spots to be found within proximity of machine shops, in the Mediterranean, in the Gulf of Mexico, and on the coast of Southern California, and the attention of designers of flying machines, who may want to test the merits of their devices upon a really adequate scale, is particularly directed to the vicinity of San Diego, Cal., where all the circumstances which have been alluded to are to be found combined, even to a local railroad along the beach, on which the tests proposed by M. Weyher might be carried on.

All this presupposes that the preliminary experiments with small models have resulted satisfactorily, and that the designer wishes further to test the merits of his apparatus upon a practical working scale, with a machine capable of carrying a man and provided with the requisite devices to bring it under control while in the air, and thus to work out the problem of equilibrium. The expense will doubtless be considerable, and the mishaps not infrequent, but there seems to be no surer way of ascertaining whether a full-sized apparatus will preserve its balance in the air, while the risk of serious injury will be small. If such experiments finally succeed in solving the equilibrium problem, in securing safety in rising, in sailing, and in coming down, with a machine carrying its operator, an immense step forward will have been taken toward solving the other problems mentioned, and toward finally developing a safe flying machine, provided with a motor of its own and capable of being operated anywhere; for once safety has been secured under the various actual conditions of out-door performance, it ought to be a comparatively easy and short task to work out the other questions, save perhaps those pertaining to the starting up from and alighting upon the ground.

Assuming all this to be possible--and while the mechanical difficulties are doubtless great, they do not seem to be insuperable--the final working out of the general problem is likely to take place through a process of evolution. The first apparatus to achieve a notable success will necessarily be somewhat crude and imperfect. It will probably need to be modified, reconstructed, and readventured many times before it is developed into practical shape.

The inventor will doubtless have to construct the first models and perhaps the first full-sized machine at his own expense, in order to demonstrate the soundness of his conception and the comparative safety of its operation; but after this much is accomplished further remodelling and experiment will still probably be required to develop the apparatus into commercial value.

This phase of the evolution is likely to require the aid of capital, because the expense may be quite considerable; and inasmuch as a financial venture to develop such a difficult and novel contrivance must be gone into as a hazard, with the acceptance of the possibility of total loss as an offset for the hope of drawing a prize, the parties advancing the capital will probably require that the invention (if invention there be) shall be fully protected by patents.

In view of this probable requirement, it may be questioned whether M. Hargrave is quite prudent in taking out no patents for his various devices, for he hints in his last paper that he is hampered in his experiments by having to perform them in public. The difficulty arises from the fact that the experiments, to be of practical value, have to be performed out of doors, and the writer knows of some designers who, unable, on the one hand, to secure a patent--in the United States at least--until they can demonstrate the practical performance which they hope for, and apprehensive, on the other hand, of being annoyed by spectators, have retired into a wilderness to make their experiments, thus placing themselves at serious disadvantage in case a mishap of any kind occurs.

Most of the patents heretofore granted for flying machines are quite impracticable, yet the claims cover, here and there, some feature which may eventually contribute to success. It will be judicious, therefore, for designers of projected flying machines to study prior patents, and an attempt has been made in these pages to indicate some of those which contain valuable suggestions. The novelty (if any) in future patents will probably largely consist in new combinations of features already patented.

There are probably a good many arrangements of sustaining surfaces which will prove available for aeroplanes; some will prove more effective and steadier than others, and this must be ascertained by experiment; but in any event success would be hastened by a working association of experimenters in this inchoate research, for the problems, as has been seen, are many, and no inventor is likely to be in possession of all the miscellaneous knowledge and variety of talent required to perfect so novel an undertaking.

To the possible inquiry as to the probable character of a successful flying machine, the writer would answer that in his judgment two types of such machines may eventually be evolved: one, which may be termed the soaring type, and which will carry but a single operator, and another, likely to be developed somewhat later, which may be termed the journeying type, to carry several passengers, and to be provided with a motor.

The soaring type may or may not be provided with a motor of its own. If it has one this must be a very simple machine, probably capable of exerting power for a short time only, in order to meet emergencies, particularly in starting up and in alighting. For most of the time this type will have to rely upon the power of the wind, just as the soaring birds do, and whoever has observed such birds will appreciate how continuously they can remain in the air with no visible exertion. The utility of artificial machines availing of the same mechanical principles as the soaring birds will principally be confined to those regions in which the wind blows with such regularity, such force, and such frequency as to allow of almost daily use. These are the sub-tropical and the trade-wind regions, and the best conditions are generally found in the vicinity of mountains or of the sea.

This is the type of machine which experimenters with soaring devices heretofore mentioned have been endeavoring to work out. If unprovided with a motor, an apparatus for one man need not weigh more than 40 or 50 lbs., nor cost more than twice as much as a first-class bicycle. Such machines therefore are likely to serve for sport and for reaching otherwise inaccessible places, rather than as a means of regular travel, although it is not impossible that in trade-wind latitudes extended journeys and explorations may be accomplished with them; but if we are to judge by the performance of the soaring birds, the average speeds are not likely to be more than 20 to 30 miles per hour.

The other, or journeying type of flying machines, must invariably be provided with a powerful and light motor, but they will also utilize the wind at times. They will probably be as small as the character of the intended journey will admit of, for inasmuch as the weights will increase as the cube of the dimensions, while the sustaining power only grows as the square of those dimensions, the larger the machine the greater the difficulties of light construction and of safe operation. It seems probable, therefore, that such machines will seldom be built to carry more than from three to 10 passengers, and will never compete for heavy freights, for the useful weights, those carried in addition to the weight of the machine itself, will be very small in proportion to the power required. Thus M. Maxim provides his colossal aeroplane (5,500 sq. ft. of surface) with 300 horse power, and he hopes that it will sustain an aggregate of 7 tons, about one-half of which consists in its own dead weight, while the same horse power, applied to existing modes of transportation, would easily impel--at lesser speed, it is true--from 350 to 700 tons of weight either by rail or by water.

Although it by no means follows that the aggregate cost of transportation through the air will be in proportion to the power required, the latter being but a portion of the expense, it does not now seem probable that flying machines will ever compete economically with existing modes of transportation. It is premature, in advance of any positive success, to speculate upon the possible commercial uses and value of such a novel mode of transit, but we can already discern that its utility will spring from its possible high speeds, and from its giving access to otherwise unreachable points.

It seems to the writer quite certain that flying machines can never carry even light and valuable freights at anything like the present rates of water or land transportation, so that those who may apprehend that such machines will, when successful, abolish frontiers and tariffs are probably mistaken. Neither are passengers likely to be carried with the cheapness and regularity of railways, for although the wind may be utilized at times and thus reduce the cost, it will introduce uncertainty in the time required for a journey. If the wind be favorable, a trip may be made very quickly; but if it be adverse, the journey may be slow or even impracticable.

The actual speeds through the air will probably be great. It seems not unreasonable to expect that they will be 40 to 60 miles per hour soon after success is accomplished with machines provided with motors, and eventually perhaps from 100 to 150 miles per hour. Almost every element of the problem seems to favor high speeds, and, as repeatedly pointed out, high speeds will be (within certain limits) more economical than moderate speeds. This will eventually afford an extended range of journey --not at first probably, because of the limited amount of specially prepared fuel which can he carried, but later on if the weight of motors is still further reduced. Of course in civilized regions the supply of fuel can easily be replenished, but in crossing seas or in explorations there will be no such resource.

It seems difficult, therefore, to forecast in advance the commercial results of a successful evolution of a flying machine. Nor is this necessary; for we may be sure that such an untrammeled mode of transit will develop a usefulness of its own, differing from and supplementing the existing modes of transportation. It certainly must advance civilization in many ways, through the resulting access to all portions of the earth, and through the rapid communications which it will afford.

It has been suggested that the first practical application of a successful flying machine would be to the art of war, and this is possibly true; but the results may be far different from those which are generally conjectured. In the opinion of the writer such machines are not likely to prove efficient in attacks upon hostile ships and fortifications. They cannot be relied upon to drop explosives with any accuracy, because the speed will be too great for effective aim when the exact distance and height from the object to be hit cannot be accurately known. Any one who may have attempted to shoot at a mark from a rapidly moving railway train will probably appreciate how uncertain the shot must be.

For reconnoitering the enemy's positions and for quickly conveying information such machines will undoubtedly be of great use, but they will be very vulnerable when attacked with similar machines, and when injured they may quickly crash down to disaster. There is little question, however, that they may add greatly to the horrors of battle by the promiscuous dropping of explosives from overhead, although their limited capacity to carry weight will not enable them to take up a large quantity, nor to employ any heavy guns with which to secure better aim.

Upon the whole, the writer is glad to believe that when man succeeds in flying through the air the ultimate effect will be to diminish greatly the frequency of wars and to substitute some more rational methods of settling international misunderstandings. This may come to pass not only because of the additional horrors which will result in battle, but because no part of the field will be safe, no matter how distant from the actual scene of conflict. The effect must be to produce great uncertainty as to the results of manoeuvres or of superior forces, by the removal of that comparative immunity from danger which is necessary to enable the commanding officers to carry out their plans, for a chance explosive dropped from a flying machine may destroy the chiefs, disorganize the plans, and bring confusion to the stronger or more skillfully led side. This uncertainty as to results must render nations and authorities still more unwilling to enter into contests than they are now, and perhaps in time make wars of extremely rare occurrence.

So may it be; let us hope that the advent of a successful flying machine, now only dimly foreseen and nevertheless thought to be possible, will bring nothing but good into the world; that it shall abridge distance, make all parts of the globe accessible, bring men into closer relation with each other, advance civilization, and hasten the promised era in which there shall be nothing but peace and good-will among all men.


42 Aerial Navigation. A lecture to the students of Sibley College, 1890. 43 M. Lilienthal raised 93 lbs. per H.P. in his experiments, but this was counterbalanced weight, and there was no head resistance of forward flight to be overcome. 44 Maxim's aeroplane and the soaring devices of Le Bris, Mouillard, Lilienthal and Montgomery.
List of Illustrations Table of Contents Index