of greater frequency is the 3d harmonic of the slower. As the slower vibration, making the long tongues of flame, is f, the higher must be c" of the second octave above f. Each third vibration of this higher harmonic coincides with each vibration of f; hence each third tongue of flame is higher than the others.

EXPERIMENT 119.-In like manner the student must analyze Fig. 56 into its simple sonorous elements. Then he should, with the vibrating flame, examine the peculiarities of the various voices of his friends, and make neat and accurate drawings of the flames corresponding to them, so that he may analyze them at his leisure.

EXPERIMENT 120.-Blow your toy trumpet into the paper cone gently, and then strongly, and observe that the sound given by the trumpet is a complex one. Try if you cannot get a flame somewhat like the trumpet gives by singing ah, through your nose, into the cone.

The student will soon find that different persons, in singing the same note, as nearly alike as they can, will produce flames of very different forms. This is because the voices differ in the number and relative intensities of the simple sounds which form them.

Another cause of the different forms of flame obtained by different experimenters is due to the fact that they have used different lengths of tube leading from the cone to the membrane.

EXPERIMENT 121.-The fact can readily be proved by singing the same compound sound through different lengths of tube leading from the cone G to the membrane.

TERQUEM'S EXPERIMENT, IN WHICH KÖNIG'S FLAME IS USED INSTEAD OF THE EAR (AS IN EXPERIMENTS 68, 69, AND 70), AND THUS THE MOTIONS OF A VIBRATING DISK ARE MADE VISIBLE.

The method of analyzing the motions of a vibrating plate (as described in Experiments 68, 69, and 70), with the paper cone and tube applied to the ear, which has been used by us for a long time, has quite recently been adapted to M. König's flame by Professor Terquem, of Lille, who has thus made these motions visible to the student, and has given us a charming experiment.

EXPERIMENT 122.-If the rubber tube used in connection with the cone in Experiments 68, 69, and 70, is attached to the tube E of Fig. 51, instead of being placed in the ear, then König's flame will remain at rest when the cone is in position No. 1 of Experiment 68, or in position No. 3 of Experiment 70. In these positions of the cone you found that no sound was heard. But, when the mouth of the cone is placed in position No. 2 of Experiment 69, the flame becomes deeply serrated; and you found in Experiment 69 that in this position an intense sound was heard.

CHAPTER XVIII.

ON HOW WE SPEAK, AND ON THE TALKING MACHINES OF FABER AND EDISON.

HOW WE SPEAK.

THE little musical instrument with which we sing and speak is formed of two flexible membranes stretched side by side across a short tubular box placed on the top of the windpipe. This box is made of plates of cartilage, movable on each other, and bound together with muscles and membranes.

The top of the windpipe is formed of a large ring of cartilage, called the cricoid (ring-shaped) cartilage. Jointed to this is a broad plate of gristle, called the thyroid (shield-shaped) cartilage. This cartilage is bent into the shape of a V. The legs of this V straddle the cricoid and are jointed to its outer sides. The peak of the V stands up and points toward the front of your throat. You can feel it, as it is the "Adam's apple." On the back of the upper edge of the cricoid ring are jointed two small pointed cartilages, known as the arytenoid (funnelshaped) cartilages. Stretching from these to the inner sides of the legs of the V of the thyroid are two membranes, one to each leg. These are the vocal chords.

When the point of the thyroid is not pulled down, these membranes are lax, and the breath from the wind

pipe passes freely between them and does not make them vibrate. (See B of Fig. 57.)

Figs. A and B.-Views of the human larynx from above as actually seen by the aid 01 the instrument called the laryngoscope.

Fig. A.-In the condition when voice is being produced.

Fig. B.-At rest, when no voice is produced.

e. Epiglottis (foreshortened).

ev. The vocal cords.

cvs. The so-called false vocal cords, folds of mucous membrane lying above the real vocal cords.

a. Elevation caused by the arytenoid cartilages.

8, w. Elevations caused by small cartilages connected with the arytenoids. 7. Root of the tongue.

But when the peak of the thyroid is pulled down by its muscles the vocal cords are stretched. At the same time the arytenoid cartilages move nearer each other, and the thin, sharply cut edges of the vocal chords are brought parallel and quite close to each other, as is shown in A of Fig. 57. If the air is now forced through this narrow slit (called the glottis), the vocal chords vibrate just like the tongue in our toy trumpet, or like the reed in any reed-pipe. A puff of air passes between them; they separate; immediately afterward they come close together and the current of air is stopped. They again open, another puff goes into the cavity of the mouth, and then they close together again. Thus the glottis opens and closes with a frequency depending on the degree of stretch on the vocal chords.

Our experiments with König's flame have shown how composite are the sounds of the human voice. The quality of a voice depends on the number and relative intensities of the simple sounds which build it up.

SPEECH is voice modified and modulated by the movements of the parts of the cavity of the mouth, of the tongue and lips.

The oral cavity is made larger or smaller, longer or shorter, and thus, resounding to some lower or higher harmonic of the voice, makes the others feebly heard.

EXPERIMENT 123.-If you form your speaking organs to say o, and then take your vibrating A-fork and hold it before your lips, you will hear the cavity of the mouth resounding to this sound. On changing the vocal organs to say e the resonance ceases.

All the vowel sounds are formed by a steady voice modified by the resonance of the different sizes and shapes given to the oral cavity.

The consonants are made by obstructions placed at the beginning or end of the oral sounds, by the movements of the tongue and lips; but, as this is a book of experiments, I leave you to inform yourself by experiments as to these

matters.

EXPERIMENTS IN WHICH A TOY TRUMPET TALKS AND A SPEAKING MACHINE IS MADE.

EXPERIMENT 124.-Sing ah, and while doing so quickly open and shut your lips twice. These two sudden obstructions to the sound have made you say mama. If you will observe attentively the motion of your mouth you will see that for the last syllable of mama you open