1911 Encyclopædia Britannica/Voice

by John Gray McKendrick

VOICE (Fr. voix, from Lat. vox), the sound produced by the vibrations of the vocal cords, two ligaments or bands of fibrous elastic tissue situated in the larynx. It is to be distinguished from speech, which is the production of articulate sounds intended to express ideas. Many of the lower animals have voice, but none has the power of speech in the sense in which man possesses that faculty. There may be speech without voice, as in whispering, whilst in singing a scale of musical tones we have voice without speech. (See Song; and for speech see Phonetics; also the articles on the various letters of the alphabet.)

Fig. 1.		Fig. 2.
Fig. 1.—Cartilages and Ligaments of the Larynx, seen from the front; half nat. size. 1, epiglottis; 2, hyoid bone; 3, small cornu of hyoid bone; 4, middle thyro-hyoid ligament; 5, great cornu of hyoid bone; 6, small nodules of cartilage (cartilago triticea); 7, the lateral thyro-hyoid ligament; 8, left lamina or wing of thyroid cartilage; 9, cricoid cartilage; 10, lower cornu of thyroid cartilage; 11, part of cricoid united to the thyroid by the middle crico-thyroid ligament; 12, second ring of trachea. (From Krause.)
Fig. 2.—Cartilages and Ligament of Larynx, seen from behind; half nat. size. 1, epiglottis; 2, lesser cornu of hyoid bone; 3, greater cornu of hyoid; 4, lateral thyro-hyoid ligament; 5, cartilago triticea; 6, upper cornu of thyroid; 7, thyro-epiglottis ligament; 8, cartilages of Santorini; 9. arytenoid cartilages; 10, left lamina of thyroid; 11, muscular process of arytenoid cartilage; 12, inferior cornu of thyroid; 13, first ring of trachea; 14, posterior membranous wall of trachea; 15, lamina of cricoid cartilage. (From Krause.)

1. Physiological Anatomy.—The organ of voice, the larynx, is situated in man in the upper and fore part of the neck, where it forms a well-known prominence in the middle line (see details under Respiratory System). It opens below into the trachea or windpipe, and above into the cavity of the pharynx, and it consists of a framework of cartilages, connected by elastic membranes or ligaments, two of which constitute the true vocal cords. These cartilages are movable on each other by the action of various muscles, which thus regulate the position and the tension of the vocal cords. The trachea conveys the blast of air from the lungs during expiration, and the whole apparatus may be compared to an acoustical contrivance in which the lungs represent the wind chest and the trachea the tube passing from the wind chest to the sounding body contained in the larynx. Suppose two tight bands of any elastic membrane, such as thin sheet india-rubber, stretched over the end of a wide glass tube so that the margins of the bands touched each other, and that a powerful blast of air is driven through the lube by a bellows. The pressure would so distend the margins of the membrane as to open the aperture and allow the air to escape; this would cause a fall of pressure, and the edges of the membrane would spring back by their elasticity to their former position; again the pressure would increase, and again the edges of the membrane would be distended, and those actions would be so quickly repeated as to cause the edges of the membrane to vibrate with sufficient rapidity to produce a musical tone, the pitch of which would depend on the number of vibrations executed in a second of time. In other words, there would be a rapid succession of puffs of air. The condensation and rarefaction of the air thus produced are the chief cause of the tone, as H. von Helmholtz has pointed out, and in this way the larynx resembles the siren in its mode of producing tone. It is evident also that the intensity or loudness of the tone would be determined by the amplitude of the vibrations of the margins of the membrane, and that its pitch would be affected by any arrangements effecting an increase or decrease of the tension of the margins of the membrane. The pitch might also be raised by the strength of the current of air, because the great amplitude of the vibrations would increase the mean tension of the elastic membrane. With tones of medium pitch, the pressure of the air in the trachea is equal to that of a column of mercury of 160 mm.; with high pitch, 920 mm.; and with notes of very high pitch, 945 mm.; whilst in whispering it may fall as low as that represented by 30 mm. of water. Such is a general conception of the mechanism of voice.

The cartilages form the framework of the larynx. They consist of three single pieces (the thyroid, the cricoid and the cartilage of Fig. 3.—Right Half of the Larynx, from a vertical and slightly oblique section; abt. two-thirds nat. size. 1, epiglottis; 2, arytenoid cartilage; 3, processus vocalis of arytenoid; 4, cricoid cartilage; 5, capsular thyro-hyoid ligament; 6, lateral crico-thyroid ligament; 7, posterior crico-thyroid ligament; 8 inferior thyro-arytn oid ligament, or true vocal cord; 9, thyroid cartilage; 10, superior thyro-arytenoid ligament, or false vocal chord; 11, thyro-ary-epiglottideus muscle; 12, middle thyro-hyoid ligament; 13, hyo-epiglottic ligament; 14, body of hyoid bone; 15, smaller cornu of hyoid bone. (From Krause.) the epiglottis) and of three pairs (two arytenoids, two cornicula laryngis or cartilages of Santorini, and two cuneiform cartilages or cartilages of Wrisberg), see figs. 1 and 2. The epiglottis, the cornicula laryngis, the cuneiform cartilages and the apices of the arytenoids are composed of yellow or elastic fibro-cartilage, whilst the cartilage of all the others is of the hyaline variety, resembling that of the costal or rib cartilages. These cartilages are bound together by ligaments, some of which are seen in figs. 1 and 2, whilst the remainder are represented in fig. 3. The ligaments specially concerned in the production of voice are the inferior thyro-arytenoid ligaments, or true vocal cords. These are composed of fine elastic fibres attached behind to the anterior projection of the base of the arytenoid cartilages, processus vocalis, 3 in fig. 3, and in front to the middle of the angle between the wings or laminae of the thyroid cartilage. They are practically continuous with the lateral crico-thyroid ligaments, 6 in fig. 3.

The cavity of the larynx is divided into an upper and lower portion by the narrow aperture of the glottis or chink between the edges of the true vocal chords, the rima glottidis. Immediately above the true vocal cords, between these and the false vocal chords, there is on each side a recess or pouch termed the ventricle of Morgani, and opening from each ventricle there is a still smaller recess, the laryngeal pouch, which passes for the space of half an inch between the superior vocal chords inside and the thryoid cartilage outside, reaching as high as the upper border of that cartilage at the side of the epiglottis. The ventricles no doubt permit a free vibration of the true vocal cords. The upper aperture of the glottis is triangular, wide in front and narrow behind; and, when seen from above by means of the laryngoscope, it presents the view represented in fig. 4. The aperture is bounded in front by the epiglottis, e, behind by the summits of the arytenoid Fig. 4.—Laryngoscopic View of the Glottis. t, tongue; e, epiglottis; pe, pharyngo-epiglottis fold; g, pharyngo-laryngeal groove; ae, aryteno-epiglottis fold; c, cuneiform cartilage, or cartilage of Wrisberg; ar, arytenoid cartilage; r, inter-arytenoid fold; o, glottis; v, ventricle; ti, inferior or true vocal cord; ts, superior or false vocal cord. (From Mandl.) cartilages, ar, and on the sides by two folds of mucous membrane, the aryteno-epiglottis folds, ae. The rounded elevations corresponding to the cornicula laryngis and cuneiform cartilages, c, and also the cushion of the epiglottis, e, are readily seen in the laryngoscopic picture. The glottis, o, is seen in the form of a long narrow fissure, bounded by the true vocal cords, ti, whilst above them we have the false vocal cords, ts, and between the true and false cords the opening of the ventricle, v. The rima glottidis, between the true vocal cords, in the adult male measures about 23 mm., or nearly an inch from before backwards, and from 6 to 12 mm. across its widest part, according to the degree of dilatation. In females and in males before puberty the antero-posterior diameter is about 17 mm. and its transverse diameter about 4 mm. The vocal cords of the adult male are in length about 15 mm., and of the adult female about 11 mm. The larynx is lined with a layer of epithelium, which is closely adherent to underlying structures, more especially over the true vocal cords. The cells of the epithelium, in the greater portion of the larynx, are of the columnar ciliated variety, and by the vibratory action of the cilia mucus is driven upwards, but over the true vocal cords the epithelium is squamous. Patches of squamous epithelium are also found in the ciliated tract above the glottis, on the under surface of the epiglottis, on the inner surface of the arytenoid cartilages, and on the free border of the upper or false cords. Numerous mucous glands exist in the lining membrane of the larynx, more especially in the epiglottis. In each laryngeal pouch there are sixty to seventy such glands, surrounded by fat.

We are now in a position to understand the action of the muscles of the larynx by which the vocal cords, forming the rima glottidis, Fig. 5.— Muscles of the left side of the larynx, seen from within; abt. two-thirds nat. size. 1, hyo-epiglottic ligament, seen in profile; 2, epiglottis; 3, aryteno-epiglottis muscle; 4, Santorini's cartilage; 5, oblique arytenoid muscle; 6, transverse arytenoid muscle, seen in profile; 7, posterior crico-arytenoid; 8, lateral crico-arytenoid; 9, lower cornu of thyroid cartilage cut through; 10, insertion of posterior portion of crico-thyroid muscle; 11, left lamina of thyroid cartilage cut through; 12, long thyro-epiglottic muscle (a variety); 13, inferior thyro-arytenoid; 14, thyro-epiglottic; 15, superior thyro-arytenoid; 16, median thyro-hyoid ligament. (From Krause.) can be tightened or relaxed, and by which they can be approximated or separated. Besides certain extrinsic muscles—sterno-hyoid, omohyoid, sterno-thyroid and thyro-hyoid—which move the larynx as a whole, there are intrinsic muscles which move the cartilages on each other. Some of these are seen in fig. 5. These muscles are (a) the crico-thyroid, (b) the posterior crico-arytenoid, (c) the lateral crico-arytenoid, (d) the thyro-arytenoid, (e) the arytenoid, and (f) the aryteno-epiglottidean. Their actions will be readily understood with the aid of the diagrams in fig. 6. (1) The crico-thyroid is a short thick triangular muscle, its fibres passing from the cricoid cartilage obliquely upwards and outwards to be inserted into the lower border of the thyroid cartilage and to the outer border of its lower horn. When the muscle contracts, the cricoid and thy roid cartilages are approximated. In this action, however, it is not the thyroid that is depressed on the cricoid, as is generally stated, but, the thyroid being fixed in position by the action of the extrinsic muscles, the anterior border of the cricoid is drawn upwards, whilst its posterior border, in consequence of a revolution around the axis uniting the articulations between the lower cornua of the cricoid and the thyroid, is depressed, carrying the arytenoid cartilages along with it. Thus the vocal cords are stretched. (2) The thyro-arytenoid has been divided by anatomists into two parts—one, the internal, lying close to the true vocal cord, and the other, external, immediately within the ala of the thyroid cartilage. Many of the fibres of the anterior portion pass from the thyroid cartilage with a slight curve (concavity inwards) to the processus vocalis at the base of the arytenoid cartilage. They are thus parallel with the true vocal cord, and when they contract the arytenoids are drawn forwards, carrying with them the posterior part of the cricoid and relaxing the vocal cords. Thus the thyro-arytenoids are the antagonists of the crico-thyroids. K. F. W. Ludwig has pointed out that certain fibres (portioary-vocalis) arise from the side of the cord itself and pass obliquely back to the processus vocalis. These will tighten the parts of the cord in front and relax the parts behind their points of attachment. Some of the fibres of the outer portion run obliquely upwards from the side of the crico-thyroid membrane, pass through the antero-posterior fibres of the inner portion of the muscle, and finally end in the tissue of the false cord. These fibres have been supposed to render the edge of the cord more prominent. Other fibres inserted into the processus vocalis will rotate slightly the arytenoid outwards, whilst a few passing up into the aryteno-epiglottidean folds may assist in depressing the epiglottis (Quain). (3) The posterior and lateral crico-arytenoid muscles have antagonistic actions, and may be considered together. The posterior arise from the posterior surface of the cricoid cartilage, and passing upwards and outwards are attached to the outer angle of the base of the arytenoid. On the other hand, the lateral arise from the upper border of the cricoid as far back as the articular surface for the arytenoid, pass backwards and upwards, and are also inserted into the outer angle of the base of the arytenoid before the attachment of the posterior crico-arytenoid. Imagine the pyramidal form of the arytenoid cartilages. To the inner angle of the triangular base are attached, as already described, the true vocal cords; and to the outer angle the two muscles in question. The posterior crico-arytenoids draw the outer angles backwards and inwards, thus rotating the inner angles, or processus vocalis, outwards, and, when the two muscles act, widening the rima glottidis. This action is opposed by the lateral crico-thyroids, which draw the outer angle forwards and outwards, rotate the inner angles inwards, and thus approximate the cords. (4) The arytenoids pass from the one arytenoid cartilage to the other, and in action these cartilages will be approximated and slightly depressed. (5) The aryteno-epiglottidean muscles arise near the outer angles of the arytenoid; their fibres pass obliquely upwards, decussate and are inserted partly into the outer and upper border of the opposite cartilage, partly into the aryteno-epiglottis fold, and partly join the fibres of the thyro-arytenoids. In action they assist in bringing the arytenoids together, whilst they also draw down the epiglottis, and constrict the upper aperture of the larynx. The vocal cords will be also relaxed by the elasticity of the parts.

Fig. 6.—Diagrams explaining the action of the muscles of the larynx. The dotted lines show the positions taken by the cartilages and the true vocal cords by the action of the muscle, and the arrows show the general direction in which the muscular fibres act. A, Action of crico-thyroid: 1, cricoid cartilage; 2, arytenoid cartilage; 3, thyroid cartilage; 4, true vocal cord; 5, thyroid cartilage, new position; 6, true vocal cord, new position. B, Action of arytenoid: 1, section of thyroid; 2, arytenoid; 3, posterior border of epiglottis; 4, true vocal cord; 5, direction of muscular fibres; 6, arytenoid, new position; 7, true vocal cord, new position. C, Action of lateral crico-arytenoid; same description as for A and B; 8, posterior border of epiglottis, new position; 9, arytenoid in new position. D, Action of posterior crico-arytenoid; same description. (From Beaunis and Bouchard.)

2. Physiology of Voice Production.—The vocal cords are tightened by the action of the crico-thyroid, or, as it might Muscular mechanisms. be more appropriately termed, the thyro-cricoid muscle. It stretches the thyro-arytenoid ligaments, the free edges of which, covered by mucous membrane, form the vocal cords. The adductors of the cords are the lateral crico-arytenoids, while the posterior crico-arytenoids are the abductors. The arytenoid muscle brings the cords together. Many of the fibres of the thyro-arytenoid are inserted obliquely into the sides of the cord, and in contraction they tighten the cord by pulling on the edge and making it curved instead of straight. Some such action is indicated by the elliptical shape of the rima glottidis in passing from the chest register to the middle register. Other fibres, however, running parallel with the cord may tend to relax it in certain circumstances. All the muscles except the thyro-cricoid (which is innervated by the superior laryngeal) receive nerve filaments from the inferior laryngeal branch of the vagus, the fibres being derived from the accessory roots. Both the abductor and adductor nerves come therefore from the inferior laryngeal. When an animal is deeply anesthetized stimulation of the inferior laryngeal nerve causes abduction of the cord, but if the anaesthesia is slight, then we have adduction. The tonic contraction of the abductors is stronger than that of the adductors, so in a state of rest the glottis is slightly open. The centre of innervation is in the medulla oblongata, and this is dominated by a centre in the Rolandic region of the cerebral cortex.

The intensity or loudness of voice depends on the amplitude of the movement of the vocal cords. Pitch depends on the number of vibrations per second; and the length, size and degree of tension of the cords will determine the number of vibrations. The more tense the cords the higher the pitch, and the greater the length of the cords the lower will be the pitch. The range of the human voice is about three octaves—that is, from fa₁ (87 vibrations per second) to sol₄ (768 vibrations). In men, by the development of the larynx, the cords become more elongated than in women, in the ratio of General physiological characters. 3 to 2, so that the male voice is of lower pitch and is usually stronger. At the age of puberty the larynx grows rapidly, and the voice of a boy “breaks” in consequence of the lengthening of the cords, generally falling an octave in pitch. A similar change, but very much less in amount, occurs at the same period in the female. At puberty in the female there is an increase of about one-third in the size of the glottis, but it is nearly doubled in the male, and the adult male larynx is about one-third greater than that of the female. In advanced life the upper notes of the register are gradually weakened and ultimately disappear, whilst the character of the voice also changes, owing to loss of elasticity caused by ossification, which first begins about middle life in the thyroid cartilage, then appears in the cricoid, and much later in the arytenoid. Eunuchs retain the voices of childhood; and by careful training it is possible in normal persons to arrest the development of the larynx so that an adult male can still sing the soprano parts sometimes used in cathedral choirs. The ranges of the different varieties of voice are shown in the following diagram, where the dotted lines give the range of certain remarkable voices, and the figures represent vibrations per second, taking the middle C of the piano as 256 vibrations per second.

A basso named Gaspard Forster passed from fa₋₁ to la₃; the younger of the sisters Sessi had a contralto voice from do₂ to fa₅; the voice of Catalani ranged three and a half octaves; a eunuch singer, Farinelli, passed from la₁ to re₅; Nilsson, in Il Flauto Magico, could take fa₅; and Mozart states that he heard in Parma in 1770 a singer, Lucrezia Ajugari, range from sol₂ to do₆, which she gave purely, whilst she could execute trills on re₅. The latter is the most highly pitched voice referred to in musical literature, an octave and a half above the highest ordinary soprano. It will be observed that the lowest note of Gaspard Forster's voice is not much above the pitch at which the perception of musical tone begins, and that from this note to the upper note of Lucrezia Ajugari there is a range of nearly six octaves, whilst the extreme range of ordinary voices, from the lowest bass to the highest soprano, is a little over three octaves. It is also interesting to observe that the range of the human ear for the perception of musical tone is from do₋₁ to do₁₀, or from about 32 to 32,768 vibrations per second—eleven octaves.

3. The Voice Registers.—The voice has been divided by writers into three registers—the lower or chest, the middle and the small or head register. In singing, the voice changes in volume and in quality in passing from one register into another. There is remarkable diversity of opinion as to what happens in the larynx in passing through the various registers. There has also been much discussion as to the production of falsetto tones. Lehfeldt and Johannes Müller held that a weak blast of air caused only a portion of the cords, as regards length, to vibrate; M. J. Ortel noticed that when a falsetto tone is produced nodal lines are formed in the cords parallel to their edges, an observation supporting the first contention; M. Garcia was of opinion that as the voice rose in pitch into falsetto only the ligamentous edges of the cords vibrated; and W. R. E. Hodgkinson showed, by dusting finely powdered indigo into the larynx and observing the blue specks with the laryngoscope, that “in the deeper note of the lower register the vibrating margin extended from the thyroid cartilage in front to a point behind the junction of the ligamentous and cartilaginous portions of the cord.” In singing falsetto tones these additional parts are not thrown into action. Some remarkable and instructive photographs obtained by French show that in proceeding from the lowest to the highest notes of the lower register the cords became lengthened by one-eighth of an inch in a contralto singer's larynx; the same singer, in passing into the middle register, showed a shortening of the cords by one-sixteenth of an inch, and another increase in length when the upper part of the middle register was reached.

4. Condition of the Larynx in the Various Registers.—In singing, one can readily observe that the tone may appear to come chiefly from the chest, from the throat or from the head, or it may show the peculiar quality of tone termed falsetto. Authorities differ much in the nomenclature applied to these varieties of the voice. Thus the old Italian music masters spoke of the voce di petto, voce di gola and voce di testa. Madame Seller describes five conditions, viz. the first series of tones of the chest register, the second series of tones of the chest register, the first series of tones of the falsetto register, the second series of tones of the falsetto register, and the head register. French writers usually refer to two registers only, the chest and the head; whilst Behnke gives three registers for male voices (lower thick, upper thick and upper thin) and five for the voices of women and children (lower thick, upper thick, lower thin, upper thin and small). These distinctions are of more importance practically than as implying any marked physiological differences in the mechanism of the larynx during the production of the tones in the different registers. By means of the laryngoscope it is possible to see the condition of the rima glottidis and the cords in passing through all the range of the voice.

In 1807 Bozzini first showed that it was possible to see into the dark cavities of the body by illumining them with a mirror, and in 1829 W. Babington first saw the glottis in this way. In 1854 Garcia investigated his own larynx and that of other singers, and three years later Türck, and especially J. N. Czermak, perfected the construction of the laryngoscope. In 1883 Lennox Browne and Emil Behnke obtained photographs of the glottis in the living man. The laryngoscope is a small mirror, about the diameter of a shilling, fixed to the end of a long handle at an angle of 125° to 130°. This mirror is gently pushed towards the back of the throat, and if sufficient light be thrown into the mouth from a lamp, and if the eye of the observer be in the proper position, by angling the small mirror it is not difficult to get a view of the glottis. The light from the lamp is reflected by the mirror down on the glottis; from this it is reflected back to the mirror, and then by the mirror it is finally reflected to the eye of the observer. Usually the observer has in front of his eye a mirror by which a powerful beam of light can be thrown from a lamp into the mouth and throat. In the centre of the mirror there is a small hole through which the eye of the observer sees the image in the small mirror at the back of the throat. By placing a second plane mirror in front of the face, an observer can easily study the mechanism of his own larynx.

Suppose the picture of the larynx to be examined in the small mirror at the back of the throat, an image will be seen as in fig. 4. During calm breathing, the glottis is lance-shaped, between the yellowish white cords. A deep inspiration causes the glottis to open widely, and in favourable circumstances one may look into the trachea. When a sound is to be made, the vocal cords are brought close together, either along their whole length, as in fig. 7, or only along the ligamentous portion, the space between the arytenoids being still open, as in fig. 8. Then when the sound begins the glottis opens (fig. 4), the form of the opening influencing the kind of voice, whilst the degree of tension of the cords will determine the pitch.

Fig. 7.		Fig. 8.
Fig. 7.—Arrangement of Glottis previous to Emission of a Sound. b, epiglottis; rs, false cord; ri, true vocal cord; ar, arytenoid cartilages. (From Mandl.)
Fig. 8.—Closure of the Ligamentous Portion of Glottis. b, epiglottis; rs, false cord; ri, true vocal cord; or, space between arytenoids; ar, arytenoid cartilages; c, cuneiform cartilages; rap, ary-epiglottic fold; ir, inter-arytenoid fold. (From Mandl.)

During inspiration the edges of the true vocal cords may occasionally be close together, as in sobbing, and during inspiration the false cords are easily separated, even when they touch, and during expiration, owing to dilatation of the ventricles, they come together and may readily close. Thus, from the plane of the cords, the true cords are most easily closed during inspiration and the false cords during expiration. J. Wyllie clearly showed in 1865 that the false vocal cords play the chief part in closure of the glottis during expiration. Lauder Brunton and Cash have confirmed J. Wyllie's results, and have shown further that the function of the false cords is to close the glottis and thus fix the thorax for muscular effort.

During the production of the chest voice, the space between the arytenoid cartilages is open, and between the vocal cords there is an ellipsoidal opening which gradually closes as the pitch of the sound rises (see figs. 9, 10, 11). During head voice, the opening between the arytenoids is completely closed; the portion between the vocal cords is open, but in place of being almost a narrow straight slit as in chest voice, it is wide open so as to allow an escape of more air (see fig. 12). Paralysis of the motor fibres causes aphonia, or loss of voice. If one cord is paralysed the voice may be lost or become falsetto in tone. Sometimes the cords may move in breathing or during coughing, but be motionless during an attempt at the production of voice. Rarely, incomplete unilateral paralysis of the recurrent nerve, or the existence of a tumour on each cord, thus making them unequal in length, may cause a double tone, or diphthongia. Hoarseness is caused by roughness or swelling of the cords.

Fig. 9.		Fig. 10.
Fig. 9.—Chest Voice, Deep Tone. b, epiglottis; or, glottis; rs, false vocal cord: ri, true vocal cord; rap, ary-epiglottidean fold; ar, arytenoid cartilages. (From Mandl.)
Fig. 10.—Chest Voice, Medium Tone. orl, ligamentous portion of glottis; orc, portion of glottis between arytenoids; remaining description as in fig. 7. (From Mandl.)

5. The quality of the human voice depends on the same laws that determine the quality, clang-tint or timbre of the tones produced by any musical instrument. Musical tones are formed by the vibrations of the true vocal cords. These tones may be either pure or mixed, and in both cases they are strengthened by the resonance of the air in the air-passages and in the pharyngeal and oral cavities. If mixed—that is, if the tone is compounded of a number of partials—one or more of these will be strengthened by the cavities above the cords acting as a resonator; and so strongly may these partials be thus reinforced that the fundamental one may be obscured, and a certain quality or timbre will be communicated to the ear. Further, Helmholtz has shown that special forms of the oral cavity reinforce in particular certain partials, and thus give a character to vowel tones,—indeed to such an extent that each vowel tone may be said to have a fixed pitch. This may be proved by putting the mouth in a certain form, keeping the lips open, and bringing various tuning forks sounding feebly in front of the opening. When a fork is found to which the resonant cavity of the mouth corresponds, then the tone of the fork is intensified, and by thus altering the form and capacity of the oral cavity its pitch in various conditions may be determined. Thus, according to Helmholtz, the pitch corresponding to the vowels may be expressed:—

Vowels	OU	O	A	AI	E	I	EU	U
Tone	fa2	si♭3	si♭4	sol5	si♭5	re6	do5	sol5
				or	or	or	or	or
				re4	fa3	fa2	fa3	fa2
No. of vibrations	170	470	940	1536	1920	2304	1024	1536
				or	or	or	or	or
				576	341	170	341	170

R. Koenig has fixed the pitch of the vowels differently, thus:

Vowels	OU	O	A	E	I
Tone	si♭2	si♭3	si♭4	si♭5	si♭6
No. of vibrations	235	470	940	1880	3760

F. C. Donders has given a third result, differing from each of the above; and there is little doubt that much will depend on the quality of tone peculiar to different nationalities. By means of Koenig's manometric flames with revolving mirror the varying quality of tone may be illustrated: with a pure tone, the teeth in the flame-picture are equal, like the serrations of a saw, whilst usually the tone is mixed with partials which show themselves by the unequal serrations. Thus quality of voice depends, not merely on the size, degree of elasticity and general mobility of the vocal cords, but also on the form of the resonating cavities above, and very slight differences in these may produce striking results.

Fig. 11.		Fig. 12.
Fig. 11.—Chest Voice, High Tone. Description same as for figs. 7 and 8. (From Mandl.)
Fig. 12.—Head Voice, Deep Tones. l, tongue; e, epiglottis; pe, pharyngo-epiglottidean folds; ae, ary-epiglottis folds; rs, false cords; ri, true vocal cords; g, pharyngo-laryngeal groove; ar, arytenoid cartilages; c, cuneiform cartilages; o, glottis; r, inter-arytenoid folds. (From Mandl.)

6. Vowel Tones.—A vowel is a musical tone produced by the vibrations of the vocal cords. The tone produced by the vocal cords is a mixed one, composed of a fundamental and partials, and certain of the partials are strengthened by the resonance of the air in the air-passages and in the pharyngeal and oral cavities. In this respect the quality of the human voice depends on the same laws as those determining the quality or timbre of the tones produced by any musical instrument. The pitch of the note of a musical instrument, however, depends on the pitch of the first or fundamental tone, while the partials are added with greater or less intensity so as to give a special character to the sound; and in the case of a vowel tone the pitch does not appear to depend on that of the fundamental tone but on the pitch of the resonance cavity, as adjusted for the sounding of any particular vowel. When we wish to pronounce or sing a vowel the oral cavity must be adjusted to a certain form, and it is only when it has that form that the vowel can be sounded. The nature of vowel tones has been investigated by means of the phonograph by Fleeming Jenkin and Ewing, L. Hermann, Pipping, Boeke, Lloyd, McKendrick and others. E. W. Scripture has worked with the gramophone. These observers may be ranged in two divisions—those who uphold the theory of relative as opposed to those who contend for the theory of fixed pitch. Assuming that a vowel is always a compound tone, composed of a fundamental and partials, those who uphold the relative pitch theory state that if the pitch of the fundamental is changed the pitch of the partials must undergo a relative change, while their opponents contend that whatever may be the pitch of the tone produced by the larynx, the pitch of the partials that gives quality or character to a vowel is always the same, or, in other words, vowel tones have a fixed pitch. Helmholtz held that all the partials in a vowel tone were harmonic to the fundamental tone, that is that their periods were simple multiples of the period of the fundamental tone. Hermann, however, has conclusively shown that many of the partials are inharmonic to the fundamental. This practically upsets the theory of Helmholtz. The methods by which this problem can be investigated are mainly two. The pitch of the oral cavity for a given vowel may be experimentally determined, or an analysis may be made of the curve-forms of vowels on the wax cylinder of the phonograph or the disk of the gramophone. By such an analysis, according to Fourier's theorem, the curve may be resolved into the partials that take part in its formation, and the intensity of those partials may be thus determined. The observations of Bonders, Helmholtz, König and others as to the pitch of the resonating cavities gave different results. Greater success has followed the attempts made by Hermann, Boeke, McKendrick, Lloyd and Marichelle to analyse the curves imprinted on the phonograph. (Examples of such phonograms are given by McKendrick in the article on “Vocal Sounds” in Schäfer's Physiology, ii. 1228; see also Phonograph.)

The following is an instructive analysis by Boeke of the curves representing the tones of a cornet, and it illustrates the laws that govern the production of quality in such an instrument:—

Note		1	2	3	4	5	6	7	8	9	10	Partials.
f′ = 170	vibs.	1	1.05	1.22	1.15	1.01	0.80	0.53	0.28	0.13	0.10	„
c′ = 256	„	1	0.92	0.81	0.53	0.39	0.20	0.07	0.04	0.06	0.04	„
g′ = 384	„	1	0.76	0.46	0.14	0.09	0.06	0.07	0.02	0.01	0.01	„
c′′ = 512	„	1	0.92	0.30	0.14	0.15	0.09	0.07	0.06	0.03	0.02	„

These figures represent the relative intensities of the partials entering into the formation of the note, and it will be observed that the intensity gradually diminishes. This analysis may be contrasted with that of the vowel āā sung by Boeke (aet. 50) on the notes f and c′, and the same vowel sung on the notes g′ and e′′ by his son (aet. 12).

Man, aet. 50, singing āā.

Pitch		1	2	3	4	5	6	7	8	9	10	Partials.
f′ = 170.6	vibs.	1	0.86	0.46	1.74	1.90	1.55	0.51	0.54	0.43	0.44	„
c′ = 256	„	1	0.49	1.96	1.25	0.60	0.56	0.23	0.05	0.06	0.10	„

Boy, aet. 12, singing āā.

Pitch		1	2	3	4	5	6	Partials.
g′ = 384	vibs.	1	1.22	2.67	0.45	0.17	0.06	„
e′′ = 640	„	1	8.09	1.45	0.53	..	..	„

It will be observed that in both these cases the intensity of the partials does not fade away gradually as we proceed from the lower to the higher partials, as with the cornet, but that certain partials are intensified more than others, namely, those printed in black. In other words, the form of the resonating cavity develops particular partials, and these modify the quality of the tone. If we multiply the vibrational number of the fundamental tone by the number of the partial we obtain the pitch of the resonance cavity; or if we take the mean of the partials reinforced we obtain the pitch of the mean resonance. Lloyd applies this method to the foregoing figures as follows:—

	Partials  Reinforced.	Mean  Partial.	Pitch in  Complete  Vibration.
Man’s āā.
f = 170.6 vibs.	4–6	4.96	846
c′ = 256 6 vibs.„	3–4	3.39	868
Boy’s āā.
g′ = 384 vibs.	2–4	2.82	1084
c″ = 640	1–3	2.04	1307

This analysis shows: (1) that the man’s resonance rises slightly (half semitone) in ascending seven semitones in the middle of his register, (2) that the boy’s resonance rises three semitones in ascending nine semitones in the upper half of his register; and (3) in the mid register the boy’s resonance is to the man’s as 5:4. Thus, as we sing a vowel in an ascending scale the pitch of the oral cavity slightly changes, or, in other words, the pitch of the resonating cavity for a given vowel may be slightly altered.

It would appear that both theories are partially true, they are not mutually exclusive. The view of Donders that each vowel has an oral cavity of unchangeable and fixed pitch is too exclusive, and, on the other hand, it cannot be denied that each vowel has a predominant partial or predominant partials which give it a definite character, and which must be produced by the oral cavity as a whole, or by the double resonance of portions of the cavity, as suggested by Lloyd. As we sing a vowel in an ascending scale the form of the resonance cavity may slightly change, but not sufficiently to alter the quality of the vowel. Thus we still detect the vowel tone. A singer almost instinctively chooses such vowels as best suit the resonating arrangements of his or her voice, and avoids vowels or words containing vowels that would lead to the production of notes of inferior quality.

Authorities.—Helmholtz, Sensations of Tone, trans. by Ellis (1875), p. 165. König, Comptes Rendus (1870), t. lxx p. 931; also Quelques expériences d’acoustique (1882), p. 47. Donders, De physiologue der spraakklanken (1870), s. 9; also “Ueber de Vokell,” Archiv f. d. holländ Beitr. 3. Nat. v. Heil. (Utrecht, 1857), Bd. i. s. 354. Donkin, Fourier’s theorem, Acoustics, p. 65; Fleeming, Jenkin and Ewing, Trans. Roy. Soc. Ed. vol. xxviii. p. 750, Lloyd, Proc. Roy. Soc. Ed. (1898); Phonetische Stud. (1890–92); Jl. of Anat. and Phys. (London), vol. xxxi. p. 23, ibid, vol xxxi. p. 240. Hermann, Phonophotographische Untersuch., Bd. i.–v.; Archiv f. d. ges. Physiol. (Bonn), Bd. xlv. s. 582; Bd. xlvii. s. 44, Bd. xlvii. s. 347; Bd. liii. s. 1; Bd. lviii. s. 255. Pipping, Zeitschr. f. Biol. (Munich), Bd. xxvii. s. 1; also Acta Societatis Scientiarum Fennicae, Bd. xx. part ii. Boeke, “Mikroskopische Phonogramstudien,” Archiv f. d. ges. Physiol. (Bonn), Bd. 1. s. 297; also Proc. Roy. Soc. Ed. (1896). McKendrick, Trans. Roy. Soc. Ed. vol. xxxviii. part ii.; Proc. Roy. Soc. Ed. (1896–97); Sound and Speech Waves as revealed by the Phonograph (London, 1897); Schäfer’s Text-book of Physiology, vol. ii. art. “Vowel Sounds”, and Nature (Dec. 26, 1901). (In the latter there is an account of the important researches of Dr Marage.) Marichelle, La Parole d'après la tracé du Phonographe (Paris, 1897). Marage, Théorie de la formation des voyelles. E. W. Scripture, Speech Curves (1906). See also Nature (February 1907). (J. G. M.)