H.F.u.f.207 (91)
Series A. No 90
Sequence:
110

ENGLISH TRANSLATION OF THE

THESIS

PRESENTED

TO THE FACULTY OF SCIENCES IN PARIS

TO OBTAIN

THE TITLE OF DOCTOR OF THE UNIVERSITY

the Right Reverend

J.I. WEDGWOOD

----------------

1ST THESIS ON THE PRODUCTION OF SOUND BY ORGAN PIPES

2ND THESIS PROPOSED BY THE FACULTY:
STUDIES ON THE SOLIDIFICATION OF METALLIC ALLOYS

----------------

Defended on January 1921 before the examination commission

MM. LIPPMAN President

M. GUICHARD \

GUILLET > Examiners

TOMBECK /

----------------

PARIS
EDITIONS "RHEA"
4, square Rapp, (7e)
---
1921

51.829

TABLE OF CONTENTS

----------------

The manufacturing of Organ Pipes (with two illustrations` )

CHAPTER I   -   Historical perspective

CHAPTER II   -   Critique of Helmholtz Theory

CHAPTER III   -   Study of the "air reed"

CHAPTER IV   -   Is the air reed a true reed?

   Summary and Conclusions.

----------------

On the production of Sound

by organ pipes

----------------

CONSTRUCTION OF THE ORGAN PIPES

ILLUSTRATIONS 1 & 2

The illustrations 1 & 2 show the organ pipes or open pipes, in their usual forms, in wood or in metal. Air under pressure (wind) is conducted to the base of the pipe (p). The flow of air, originally vertical, is deviated by the bevel which let the wind take the form of a thin blade through the narrow aperture called touch-hole (l) which is formed by the bevel and the underlip (l.s.), and induces a periodical movement in the air column which occupies the body of the pipe (c.).

The size of the pipe and the quality of the sound it should produce determine the air pressure, the width of the touch-hole, the angle the touch-hole is forcing the blade of air, and finally the height and shape of the upper lip.

The pipe is manufactured approximately to the determined length, and it is later tuned by variable cuts at the upper opening or by other similar means. The tuning of closed, or plugged pipes (which manufacturing procedures vary only by the fact that they are closed at the top aperture) is made by moving the plug.

----------------

CHAPTER I

THEORY OF THE MANUFACTURING OF ORGAN PIPES

HISTORICAL PERSPECTIVE

I. - Prevalent Theories

It is generally accepted that the sound produced by organ pipes is the result of the periodical movement of the air column they contain. But the initial cause of this periodical movement is not yet well established. The explanations on this topic provided by the books on acoustics are incomplete and often inaccurate; their learned authors only touch the surface while there are details about the process requiring a more elaborate study. Pietro Blaserna, for example, just says that air " partially penetrates inside the pipe, induces vibrations, and produces a clear and appreciable sound". (Theory of Sound in its relation to Music, pages 17-18 )

Privat Deschanel tell us that the upper lip "is itself capable of vibrating, in accordance with any note being placed at considerable distance, and the note is determined by the air column inside the pipe". A similar theory is exposed in the work of Schilmbach, Ueber die Erhaltung, Stimmung und Prüfung der Orgel (third edition, revised by Becker, Leipzig, 1843). This work declares (page 104) that the two lips are "violently projected in movement", and that this fact is proved by "the disappearance of the sound when a finger is placed on the lip". It happens that this last assertion is directly contrary to the truth. In the modern organs, the upper lip is often of substantial thickness, and even recovered with leather; the theory of a vibrating lip is entirely refuted by the fact that it can be maintained in a state of complete immobility, without degrading the sound of the pipe. Tyndall says that projecting the thin blade of air against the cutting of the lip "produces pulsations of a determined duration which is converted into musical sound by the resonance of the associated air column." (Sound, page 183, London, 1867 ) Speaking of such pulsation, he is much closer to the truth than the other Authors quoted above. The only critique one could make about his more precise assertion, is that it is incomplete.

It would be redundant to quote the works of other physicists, who holds similar opinions, and I am proposing to discuss several theories by Helmholtz, exposed in his work Théorie physiologique de la Musique (translated from the German by Mr. G. Guéroult, Paris, 1868). This book by Helmholtz is, since a long time, considered as the classic work on acoustic phenomenons. He goes much deeper in our topic than most of the other authors.

II. - The Helmholtz Theory

Helmholtz presupposes in his theory that every pipe produces a series of specific harmonics determined by its shape. Furthermore, other sounds are produced by the touch-hole, and his opinion, with the opinion of other authors, is that if these sounds are sufficiently in harmony with the harmonics specific to the pipe, they are reinforced. We will later examine this theory in detail.

Here is what he says:

"As it is the case for the pipes of this kind, for flutes, the bottles previously described, the resonance case of a violin, it is possible, in general, to produce sounds with any cavity full of air, and equipped with an opening sufficiently narrow, by directing towards this orifice a narrow stream of air, in the shape of a ribbon; it is supposed that such orifice is equipped with edges sufficiently salient and sharp.

" It is the shock of the air on these edges which initiates the sound in all these instruments; the flow hits it and produces a hissing or whistling, which is heard alone when there is no pipe adapted to to it, or when a similar opening is made in a small plank. The narrower the opening, the higher is the whistling. A sound of this nature is to be considered as said above, as a mixture of a great number of discordant sounds, close to one another. The cavity of the pipe will reinforce, by resonance, the sounds of the mix which correspond to its own sounds; the latter sounds take then an intensity superior to all the other sounds, and cover them completely. So, in all the instruments of this nature, the hissing of the air is heard more or less sharply, accompanying the sound, and that gives something particular to the quality of the instrument" (Pages 123-124).

He continues:

" It is even possible to reinforce the sound of a tuning-fork by bringing it close to the opening of a pipe, when the pitch of the tuning-fork corresponds to the specific mass of air of the pipe, and by using a series of tuning-forks, one can easily find and determine with precision the sounds particular to the pipe.

"Furthermore, the musical character of the sound of these pipes depends essentially on the following fact: the harmonics of the sound produced by the air are or not closer to the corresponding sounds of the pipe to be reinforced as the fundamental sound.

" Only in narrow cylindrical open pipes, as flutes and the principal violin register (1) of the organ, are the specific high sounds of the pipe in exact correspondence with the harmonics of the principal sound. When blowing harder, which increases the pitch of the whistling of the air, one can speak of the only high pitch of the pipe... In cylindrical pipes (2) , besides the fundamental sound, a series of harmonics is reinforced by the resonance of the pipe, especially if the wind blows with strength to increase greatly the pitch of the sounds included in the whistling of the air". (ibid , page 124)

III. - Theory of the air reed.

in 1840, Mr. Aristide Cavaillé-Coll, Parisian organ manufacturer, who later enjoyed gained celebrity, presented (on the 24th day of February), an essay to the Académie des Sciences in Paris. The essay was only published in 1895 under the title Etudes expérimentales sur les tuyaux d'orgues . (Experimental Studies on Organ Pipes).

Cavaillé-Coll observed that when pipes longer than 16 feet were sounded, he believed to notice, when placing the hand before the aperture of the pipe, that the blade of air exiting the opening was vibrating inside and outside the pipe, as would do a metallic vibrating blade that would be placed in the stream of air, and pinched by one of its ends inside the edge of the opening. He named this movement "free air reed" or simply "air reed".

The air reed theory was independently conceived by Hermann Smith of London, who was the first to propose an explanation of such production (cf. a series of articles published in 1873-1874 in Nature ; and also the addendum XIX, pages 708-711, of the English translation of the work of Helmholtz, by Ellis; and also Modern Organ Tuning and The Making of Sound in the Organ and in the Orchestra , both works by Hermann Smith). The question was also debated by Shneebeli, in 1874 (Poggendorffs Annalen der Physik und Chemie , Band 158) and by F.W. Sonreck in 1876 (ibid , Band 158).

These authors have slightly different opinions what regards the details of the operation of the air reed. Is the initial phase of the process a condensation or a raréfaction? But the question doesn't change anything in the process, and these authors claim that the air reed is produced by alternating phases of condensation and rarefaction of the air included in the pipe, forcing the air blade to oscillate quickly before the upper edge. Let us suppose that the initial phase was a condensation towards the inside of the pipe, caused by the accelerated passage of air through the opening. The air column is pushed upwards, and the air blade is equally pushed out of the pipe. This movement is immediately followed by a rarefaction, causing the air blade to penetrate again inside the pipe to be again pushed out of the edge. And so, the phases are repeated alternatively.

At this point, I feel it appropriate to mention the research of MM. Charles Lootens en van Tricht, published in 1877, in the Journal de Physique (Volume VI, page 53). This very short article does not give many details on their experiments. They claim that the air, when reaching the upper edge, is divided. To the flow moving outside, (by the way, they tell us that it is the major part of the air), they give the name of "principal current", and to the flow moving inside, the name of "derived current". They have determined the direction taken by the latter current using cigarette smoke, and also using a brass support holding screws or valves made of elderwood which indicated a rotary flow (named "cyclone"), moving away from the front side, then partially or completely expelled from the pipe tonguing. Sometimes, an other part of this current makes an inverted "cyclone", but higher in the pipe.

These observations are in accordance with the theory of the air reed I just mentioned. The only critique one may attempt to make, concerns the used terminology. My personal observations have shown that the major part of the current is attracted inside the edge, and should therefore be named "principal current". What regards the phenomenon itself, and its cause, I have presented some observations further in this thesis.

------

CHAPTER II

CRITIQUE OF HELMHOLTZ's THEORY

When Helmholtz' theory is scrutinized, a number of weak points are becoming immediately apparent.

Research made since the publication of Helmholtz' book have added more weight to another theory, the theory of the "air reed". Furthermore, as Helmholtz has probably noticed, there are problems in establishing a very precise theory based on the principles he has described. It is thus very possible that he has been quite vague on this topic. He has never explained if he considers the pipe to be the place where sounds are produced by the whistling of the air on the edge, and that he supposes to be selected and reinforced by the pipe acting as an resonator. However, it can be argued that this theory is not admissible.

I have taken several pipes, removed the body, and listened the whistling with care. In any of them did I found the presence in this whistling the presence of the fundamental sound, nor harmonics.

For example:

I took two open pipes, sounding the same note, but with mouthpieces of different dimensions. These pipes measured 368,3 mm in length and 41,27 mm in diameter and sounding the Fa (F) above Ut-3 (C-3). I selected this note because it is sufficiently low to be easily distinguished from the whistling produced by the mouthpiece, which is quite acute. In the first of these pipes, the mouthpiece was 23,81 mm wide and 9,52 mm high, the second 26,98 mm and 9,52. The first had unusual fine and little dispersed indentations, and produced a clear whistling, easy to recognize. In the other, with rather thick indentation, the whistling was difficult to notice.

I removed the body of each pipe. In the first, I found that the lowest note I could hear had a variance close to one seventeenth above the normal sound of the pipe. In the second, the variance was even higher. I may thus conclude with certainty that no lower sound was produced. The whistling did not sound the fundamental note, nor the three following harmonics which, according to Helmholtz' theory should have be sounded to be reinforced by the pipe acting as an resonator.

To obtain a visible demonstration, I have repeated the experiment in similar circumstances, using a series of Koenig resonators (eight in total). The lowest resonator was tuned to Utah (C-2) [256 vibrations/sex]. I made the pipe sound this note before the resonators, and a rotating mirror showed that the sensitive candle attached to each resonator was affected. I then cut the pipe above the mouthpiece, leaving a significant of the upper edge, so that the whithling could be produces in favorable conditions. None of the resonator was affected by the whithling.

I have repeated the experiment with air under increased and decreased pressure, but (as it was to be expected) the results were equally negative.

Helmholtz' theory is obviously not well founded.

After having established this point, it may not be out of place to add a few reflections of a theoretical and inductive nature that will serve to demonstrate that, in reality, the whistle is only an rather unimportant factor.

The difference between this whistle and the true note inside the pipe is very noticeable in the middle size pipes - let us say four feet - but, when the pipes of the octave, as from thirty-two feet are examined, the contrast is obvious. I have listen to several models of this kind. Close to the pipe, the sound produced at the mouthpiece is so powerful that it cannot be called a whithling, it is a powerful blowing of wind that is not related to any note or any defined group of notes. A note sounded by a pipe of such size is very penetrating and keeps its intensity at a longer distance than an the ones produced by pipes emitting shorter audible vibrations. (In fact, such pipes are often placed at a distance of the body of the organ to avoid disgracing the architecture of a building). At such a distance, only the melodious and penetrating note of the pipe is heard; none of the air noise is perceived, because it is not carried through space. The wind noise heard in the large pipes is caused by the impossibility to obtain attack, that is to say to set such a large air column in motion immediately. It is the most audible at the instant preceding the sounding of the pipe. Such difficulty does not present itself in pipe of more manageable sizes.

In the harmonization of the pipes, the upper edge is split to such a height that the air blade is dispersed and looses much of its initial pressure before hitting the edge. It is true that, near the pipe, a light whithling is heard at the edge, which (as Helmholtz says) "gives something particular to the sound". But it can be said that this noise, manifested at the mouthpiece is a mechanical noise, and not a musical sound, and that it can even be practically eliminated in normal size pipes, by adjusting the tongue to the appropriate height. It can be compared to purely mechanical noise that can be heard when the bow touches the sting of a contrabass.

III

There are other considerations indicating that the voice of an organ pipe depends from other factors than the simple amplification of certain sounds produced by the edge.

We know that splitting the air with a walking stick makes a sound. But the movement of the stick has to be fast and produced with a particular force, if a sound with certain volume is to be obtained - especially if the stick is thick and doesn't produce any vibration by itself. The opposite proposition is equally true. It will produce a sound if a current of air is directed against a cutting edge as the one of an organ pipe. As the edge itself does not vibrate by itself, a violent currant of air should be needed to produce a loud sound. The jet of air coming from the aperture of a pipe does not appear to be sufficiently intense or voluminous to produce a sound having the desired height and volume. The note produced by a stick through the air is only a very little stronger, is it is stronger at all, when the corresponding note vibrates on a tuning fork. But, when the latter is amplified by resonance, the resulting sound has no relationship with the note of an organ pipe. The organ pipe gives a firms and full sound that has little to do with the soft whispering of resonance. The body of the sound is thus the primary source of the sound, and not the amplification of a sound having its origin elsewhere. Not more, furthermore, than in the similar case of the violin string, by the light scratching of its surface. There is, in the sound of an organ pipe, such energetic vibration that makes unacceptable the "reibungstheorie" of Helmholtz and other physicists.

The theory of the air reed, on the other hand, (with the exception of the false ideas that its name may suggest) is perfectly suitable to explain the phenomenon, and allows for to consider the column of air, when intensely activated, to be the primary source of the sound. In the first case, there is a theory stating that the air when projected against a mass, produces a noise wherefrom some produced sounds are selected and reinforced by the pipe; the latter being used as the amplifier. In the second, a theory by which a regular condensation and evacuation at the mouthpiece of the pipe - which easily explains the characteristic beat of this class of organ pipes.

------

CHAPTER III

STUDY OF THE "AIR REED"

The general functioning of this air reed may be observed by different methods, of which the simplest consists to glue a sheet of paper close to the aperture; the flow of air lifts up the sheet and makes it oscillate through the upper edge. A way to make a detailed study of the air reed consists in projecting on it the flame of a thin jet of gas, a current of smoke of ammonia salt, talc powder, or blowing cigarette smoke in the pipe. I have used all these methods in my experiments.

Its Creation.

First, it is important to note that the air reed is produced at the mouthpiece only if there is an air column enclosed in a sonic chamber. I have blown air into a pipe wherefrom the body had been removed as from the top of the upper edge.; the air moving inside would hit the edge, but did not produce an oscillating air reed. The flame of sheet of paper used in my experiment, was blown away by the air flow, trembled a little, but didn't begin to oscillate rhythmically with a noticeable movement.

What is the direction of the air reed?

There are diverging opinions what regards the direction the air blade takes in relationship with the upper edge to produce an air reed. As the operation of organ pipes is examined, I have thought that is was appropriate to study this question.

Schneebeli (Poggendorffs Annalen, 1874 ), has defended the idea that the air must hit the edge, while Hermann Smith claims that the air is not directed to the edge, and that "if it was the case, the pipe would not sound" (Appendix of the French translation of Helmholtz, by A.J. Ellis, London 1875, page 709). "Harmonization is an art consisting in directing the flow of air in such a manner that it avoids to hit the edge, but to touch it lightly without making any movement on the edge." (The Making of Sound in the Organ and in the Orchestra, op. cit. page 102). Dr, G.A. Audsley, in his important work The Art of Organ Building (Dodd, Mead and Co., New York, 1904, 2 vol.). writes about him as "the most prominent authority of all times, on the Organ Pipes Acoustic." He seems to share Hermann Smith's opinion, as he writes: "The pipe will not sound at all if the air blade is directed to the inside of it, or in such a manner that it would be split on the edge, regardless if the latter is thick or thin". (page 379).

The opinions of Hermann Smith and Audsley seem erroneous to me. It is not true that the organ builders direct the air blade, without exception, towards the outside of the pipe. They angle it in various directions, de pending on the type of pipe, and the particular sound they want to obtain. None of these writers has taken this fact in consideration. With the flutes and the Stoppel Diapason, old English Model, they usually direct the flow towards the inside of the pipe. With the others, the direction varies widely with a number of them pointing to the outside.

It is not difficult to resolve this issue decisively by well known experiments as Schneebeli has described them, an to this end I have used two test pipes, 20 cm long. The reed and the upper edge were tuned in such a manner that in the first case, the air blade would pass outside the upper edge, and in the second inside. This was carefully checked with cigarette smoke. The first of the two pipes would only sound when a continuous flow of air was directed through the mouthpiece, so that the blade was deviated towards the edge, otherwise only a light whistling was heard. The second pipe would only sound when a similar flow of air was directed through the top of the pipe, and also moved the air blade to the edge. Furthermore, I have found the way to adjust a third pipe with such a precision that a temporary current of air, coming from outside, would let it sound; and it would continue to sound until a new current of air would be directed from the opening at the top.
It is thus obvious that the air blade must as it ware, hit the upper edge.

III - (Continued )

Why is it necessary that the air blade in one way or another hits the edge? Schneebeli demonstrates that it is the case, but does not go deeper in the matter. Let us examine it briefly.

While observing several times using a gas burner (and other means) the direction taken by the air blade, I discovered a fact that the several authorities I have quoted did not mentioned. Cavaillé-Coll and Hermann Smith in some illustrations, have both shown us the air reed in a position it never occupies. They show the air blade rising in a vertical direction from the opening and then slowly dispersing. One could thus expect that a flame placed near the base of the air blade would be submitted to oscillations, and be violently pushed away in the opening of the pipe. We see, on the contrary, that air flowing the inside of the pipe attracts the flame through the entrance of the mouth. When a train travels through space at a great speed, the movement compresses the air before the train, and leaves an emptiness behind it. Similarly, the blade of air when moving towards the upper edge, pushes the air before it, and created a partial emptiness behind, taking some air within itself in its flow, on each side. At the outside of the pipe, such flow may be taking from the unlimited supply of air of the atmosphere, and a flame placed close to the lower edge, shows how the air blade , coming from the opening, aspires air from the outside. But, inside the pipe, the air blade goes through a close area limited on one side by the layer of air between the opening and the upper edge, and on the other by the pipe itself. A partial emptiness is immediately produced (that is proven by the fact that a flame placed at the base of the bevel is quickly aspired). So the balance of the air blade is destroyed, forcing it to move towards the inside of the pipe. The already described process is thus initiated.
If we drill a hole in the pipe, immediately above the bevel (or a large enough hole in the bevel itself) such partial emptiness cannot occur, and the sound of the pipe is destroyed. The same is true if the air blade is not making contact with the upper edge, because it is directed too far outside or inside allowing the atmospheric air to enter.

IV - (Continued )

Schneebeli adds an observation to his experience. By projecting the air blade mainly at the outside of the pipe, smoke introduced through the mouth, is rarely seen coming out of the top. Thus, the smoke is moved in another direction by the return of the air reed.
Repeating this experience, I have found that the air blade, when principally directed towards the outside of the edge, produces an air flow from the top to the bottom of the pipe, that lets the smoke come out of its mouth, when introduced at the top of the pipe. On the other hand, when the air blade is principally directed towards the inside of the pipe, the air flow is directed from the bottom to the top, taking with itself the smoke from the outside.
As it is quite difficult to observe clearly the traces of smoke, when mixed with air, I have used the fact that warm air increases immediately the pitch of the sound of a pipe. With the first method, a burning match or a flame held at the mouth of the pipe didn't change its pitch immediately, with the other method, the match changed the pitch instantly.

------

Chapter VI

THE AIR REED - IS IT A REED?

Mr. Cavaillé-Coll was the man who gave to the oscillating air blade located at the mouth of an organ pipe, the name of "air reed" or even the "free air reed". However, there are differences in their operations that are worth of study:

- In a pipe with a reed, the little tongue being "free" or "moving", it is always these tongues (with the associated blows of air) that are the source of sound. This tongue will also operate also when it is detached from the pipe that is attached to it as a resonator, to reinforced and qualify the created sound.
- The pitch of the sound is determined by the vibrating length of the tongue. The pipe, by itself, has a rather unimportant influence on the pitch, but on the other hand has a great influence on the sound quality. (To obtain a sound pure and of good quality, the vibrations of the resonator must be in harmony with the tongue).
- The reed pipes are different from the mouth pipes, by the fact that they don't interfere with the harmonics of the resonator. The tendency (with the exception of the free reeds) is to increase the pitch in direct relation with the increase in pressure, but this may be changed under certain conditions. The general ideas following seem admissible to me, due to the results of my experiences.
Free Reeds - The free reed has the particular quality that distinguishes it from all the other organ pipes. When the power of the air flow in increased, the power of the sound is increased, but the pitch remains unchanged. This it the reason for its use in variable pressure harmoniums and American organs.
A free reed, without a resonator, lowers its pitch by less than half a tone, when the wind pressure is significantly increase, then becomes silent or produces a very high note, more than five or six octaves higher. (Such note is probably produced to rotating vibrations that is not produced when the tongue is twisted, as it is customary in American organs).
When free reeds are used in organs, the resonators are usually very short; this doesn't influences the tongue significantly; a long resonator kills the note quicker when the pressure increases.
Moving Reeds - Without resonator, the pitch gradually increases with the increase of pressure.
A short resonator (as used in clarinet, hoboes, and vox humama ) doesn't has a significant influence on the resulting sound.
A resonator of the right length will lower the pitch by three or four half tones, because of the superimposed air column requires additional and heavier work by the tongue, which reduces the speed of vibration. When the pressure is increased, the pitch increases, then the note springs suddenly to the original note produced by the tongue. When speaking of this phenomenon, it is said that a reed pipe is "overblowing". Such note doesn't belong to the harmonics of the pipe.
A resonator that is, let us say, too long by one tone, will not decrease the pitch proportionally (but decreases significantly the quality of the sound). An increase in pressure increases the pitch unequally.

II.

Let us now see how a mouth pipe operates under the same conditions:

- As we have seen, the mouth pipes will only produce a small sound by the air flow at the level of the mouth, unless a sound body is added. The source of the sound is such body.
- The pitch is not produced by what we could name the length of the "free air reed", that is to say the pitch of at the mouth, but by the sound body. As a variation in the length of the tongue of a reed pipe, produces significant changes in the pitch, the variations in the pitch at the mouth will only produce small changes in the pitch of the pipe, which is in proportion with the changes in the length of sound body and is analog to the reduction of the sound occurring when one holds a hand before the mouth. But the pitch at the mouth influences considerably the quality of the sound. (If the edge is too low, the pipe will have a tendency to jump an octave - if it is too high, the quality of the sound is altered because of the hissing it creates. But a decrease in pressure may correct the mistake of a low edge. What seems to be necessary is that the air blade, when reaching the edge, has lost enough - but not too much - of its original pressure to avoid an action too strong on one side, and too weak on the other).
That the air reed is not submitted (as is the tongue) to a constant periodicity that could be determined by its length, it easy to be demonstrated by experience.
By gradually reducing the length of a pipe, that is increasing the pitch, the mouth remaining unchanged, I found by using light paper glued at the mouth, that the air reed movement was increasing in conformity. On the other hand, when closing the pipe, which reduces the pitch by more or less an octave, I discovered that the pulsations of the air reed were reduced proportionally. Furthermore, if the wind is forced to jump an octave in an open pipe, the speed of the movement of the air reed is increased.
We can see by this experience, that the air reed has a no periodicity that is particular to it, or that is determined by the mouth, but that it adopts the periodicity of the sounding body. Which to my opinion is nothing less than the periodicity of the air column.
- When a mouth pipe is made to "overblow", it always reaches the first harmonics (an octave for an open pipe and a twelfth for a closed pipe). Contrary to the case of a reed pipe, such note is determined by the periodicity of the pipe. The other harmonics vary according to the increase in pressure.

III.

Most of the authors have naturally study the phenomenon of the "air reed" from the point of vue of the air blade. By following the path of the air blade from the opening to the top, it was easy to compare it with a vibrating reed. So Cavaillé-Coll speaks of it as the "first engine of sound in the mouthpiece of the flute pipes, an engine we associate with a free air reed (op. cit. p. 6). But, it may be more accurate to look it from the point of vue of the pipe. The air blade is nothing else than a process to have the air column move at regular intervals inside the pipe. In this case, the air column ceases the air blade when it reached a sufficient degree of dispersion and makes it oscillating from one end to the other. The oscillation of the air blade is more the result of the cause of the periodical movement of the air column. The first cause of the sound is the attack performed on the air column by the air blade when attracted to the inside, and that (as demonstrated by Lootens and van Tricht) is animated by a rotary movement. This aspect of the phenomenon resembles to the action of a bow on a string, which causes the rotary movement demonstrated by Cornu. The air reed, that is to say the oscillation of the air blade, is the result of the movement of the air column, and represents the air blade in its quality of vibrator which, by its regular intervals, allows the air column to move in a periodic phase.
After all, the choice of the expression "air reed" is not adequate, as such air blade has little in common with a reed, and doesn't operate at all in the same manner.

------

SUMMARY AND CONCLUSIONS

- In his Théorie physiologique de la Musique , Helmholtz has developed an hypothesis by which the whistling produced at the mouthpiece of such pipe must be considered as a mix of a great number of discordant sounds, close to one another, wherefrom the pipe, depending on its dimensions and shape, selects and reinforces some of them.
By examining several pipes, I have found that the fundamental sound, nor the harmonics were present in the whistling, and I have demonstrated this fact with a commonly used test pipe, by using Koenig resonators. Consequently, this theory by Helmholtz doesn't seem to have foundation.
Furthermore, the quality of the sound produced by a mouth organ pipe is absolutely different from the sound produced by a pipe simply serving as resonator.
- Cavaillé-Coll and Hermann Smith (and other authors after them) have demonstrated that the air blade, at the mouthpiece of the pipe is brought in an oscillating state while the pipe is sounding. They have given the name of "air reed" to this oscillating air blade. I have examined this "air reed" from different points of vue.
- Schneebeli has demonstrated that a pipe is not sounding as long as the air blade is not hitting, so to speak, the upper edge of the pipe in one way or another. Contrary assertions have been made since by Hermann Smith and Dr. G.A. Audsley (author of the largest and most important treatise on organ construction). They are of the opinion that the pipe will not sound if the blade is divided by the edge. I have verified, with great care, the accuracy of Schneebeli's conclusions, and attempted to explain why the wind must hit the edge.
- By repeating the experience by Schneebeli, I have found that the direction taken by the air flow inside the pipe - from top to bottom or from bottom to top - depends on the position of the air blade mainly outside or inside the pipe.
Finally, I have permitted myself to criticize the name "air reed", given to the oscillating air blade by Cavaillé-Coll and other following authors. I have demonstrated that such oscillating blade has little in common with a reed, and doesn't at all behave like one. Cavaillé-Coll considers such "free air reed" (as he calls it) as the "first engine of sound". I do not think of this phenomenon in the same manner, and consider it more like the effect caused of the vibratory movement of the air column. Its periodicity is not determined by the height of the upper edge (which would have to correspond with the vibrating length of the "air reed"), but is entirely determined by the vibration of the air column.
-------