The sound produced is musical tone, consisting physically of a series of like vibrations, and technically called voice. The rate per second of the vibration determines the key or musical note on which a sound is pronounced or sung. The shorter the cords and the higher the tension the higher the rate and the higher the note. The loudness is determined by the amplitude of the vibrations of the cords. The air waves produced by the cords are very feeble; but they are reinforced by the resonating effect of the mouth and nose cavity. It is estimated that the voice as heard is some 300 times as loud as it would be if not thus reinforced. The cords, like a violin string, vibrate not only as a whole but also in segments, the segmental vibra tions yielding overtones. For example, if the cord as a whole vibrates 100 times per second, the half cord will vibrate 200 times, the third 300 times, the fourth 400 times, the eighth 800 times, the twelfth 1,200 times and so on. The commingling of these overtones of varying loudness is what gives the main characteristic distinction to the voices of different persons.
The vowels are ((voice" with little or no audible commingling of other sounds. After much study it now seems to be established, that the difference between the various vowels is caused by the presence of certain characteristic tones. Some of these have been recently cal culated as follows: oo in moon ca. 225 vibrations per second o in mow, roam. " 460 aw in maw " 732 • a in ma " 10.50 id IV The following vowels have two character istics: et in med ca. 310 and ca. 3100 a in mate " 490 " " 2460 e in met " 690 " " 1950 a in mat ° 800 ° " 1840 A in ma has also been found with two characteristics, 950 and 1,240. According to the above definition we must regard the liquids and nasals 1, tn, n and r as vowels (even though they lack the clear, open quality of a, e, i, o, u), particularly in such words as apple, serum, oven (often pronounced ov'm) and over, in which the e and u are usually silent, the 1, in, n and r being the loudest sounds in their syllables.
In addition to musical tones the speech organs produce sounds which consist either of irregular vibrations or of regular vibrations, that do not produce on the ear the effect of musical tones. Such sounds are called noises. There are two classes of these noises: (1) Explosive noises, (2) frictional noises.
The former are made by the complete stop page of the air (effected by a closure at some point in the throat, mouth or larynx), and the subsequent compression of the air behind the closure, followed by the sudden opening of the closure by muscular action and the air pressure. The resulting consonantal sound is called a ((stop" or "explosive." The closure yielding p is made by bringing the lips into t by pressing the tongue against the teeth (French and German dental t) or the gums (alveolar t, as commonly made in England and America, though dental and alveolo-dental t is often found among us) or against the top of the mouth (cerebral or pre palatal t, as found in Sanscrit), k by pressing the rear part of the tongne against the back part of the hard palate (palatal k, as in kick) or against the velum . (velar k, as in koal or
coal) or against the back part of the throat below the velum (gutteral k, as in Arabic). In all of these sounds except Arabic k and the two now to be mentioned, the velum is also drawn upward, closing the nose passage. A k may also be made by pressing the epiglottis against the back of the throat and a glottal stop is produced by an explosion at the vocal cords (Greek A velar natal stop is made by closing the lips and velum and snapping open the latter. b, d, g are the same explosive noises accompanied by voice. Trilled r is produced by a succession of such stops made by the rapid vibration of the tongue or the uvula against the adjacent parts of the mouth. Frictional noises (spirant consonants or fricatives) are produced when the air pas sage is reduced by near closure to a very small channel and the air forced swiftly through it. Thus are produced voiceless s in see, sh in shall and th in think. To these correspond voiced z in zest, z in allure and th in these. All these are hisses. If the channel is larger and the breath driven more gently, the noise is a murmur as in l and American northern r (voiced or voiceless). Glottal spirants are English h (sometimes voiced) and the If immediately following an explosion the passage remains narrow for an instant the explosion is followed by a hiss. These sounds are called affricates: ch in change, j in jump. Y as in yet is a swiftly pronounced along" e; w in we is an evanescent oo as in boot.
In the production of most of the above sounds the velum may remain open, i.e., drawn forward and downward, adding the resonance cavity of the nose to that of the mouth. This greatly increases the quality (by reinforcing the higher overtones) and volume of the sound. Hence singers are especially trained to use nose resonance to the greatest possible extent. If the velum is closed, fully half the resonance effect is lost. By using more breath and over straining the muscles of the larynx one can partly make up the loudness, but the quality is irretrievably lost. A full, clear resonance can be secured only when the mouth opening is fairly wide. If it is completely stopped (as in pronouncing n), or made very small, we get a muffled nasal twang characteristic of certain French vowels and of pronunciation.