POINTS AND SIGNALS IN COMBINATION.

Let us suppose that a train is approaching the junction shown in Figs. 98 and 99 from the left. It is not enough that the driver should know that the tracks are clear. He must also be assured that the track, main or branch, as the case may be, along which he has to go, is open; and on the other hand, if he were approaching from the right, he would want to be certain that no train on the other line was converging on his. Danger is avoided and assurance given by interlocking the points and signals. To the left of the junction the home and distant signals are doubled, there being two semaphore arms on each post. These are interlocked with the points in such a manner that the signals referring to either line can be pulled off only when the points are set to open the way to that line. Moreover, before any shifting of points can be made, the signals behind must be put to danger. The convergence of trains is prevented by interlocking, which renders it impossible to have both sets of distant and home signals at "All right" simultaneously.

WORKING OF BLOCK SYSTEM.

We may now pass to the working of the block system of signalling trains from station to station on one line of a double track. Each signal-box (except, of course, those at termini) has electric communication with the next box in both directions. The instruments used vary on different systems, but the principle is the same; so we will concentrate our attention on those most commonly employed on the Great Western Railway. They are:—(1.) Two tapper-bell instruments, connected with similar instruments in the adjacent boxes on both sides. Each of these rings one beat in the corresponding box every time its key is depressed. (2.) Two Spagnoletti disc instruments—one, having two keys, communicating with the box in the rear; and the other, in connection with the forward box, having no keys. Their respective functions are to give signals and receive them. In the centre of the face of each is a square opening, behind which moves a disc carrying two "flags"—"Train on line" in white letters on red ground, and "Line clear" in black letters on a white ground. The keyed instrument has a red and a white key. When the red key is depressed, "Train on line" appears at the opening; also in that of a keyless disc at the adjacent signal-box. A depression of the white key similarly gives "Line clear." A piece of wire with the ends turned over and passed through two eyes slides over the keys, and can be made to hold either down. In addition to these, telephonic and telegraphic instruments are provided to enable the signalmen to converse.

SERIES OF SIGNALLING OPERATIONS.



We may now watch the doings of signalmen in four successive boxes, A, B, C, and D, during the passage of an express train. Signalman A calls signalman B's attention by one beat on the tapper-bell. B answers by repeating it to show that he is attending. A asks, "Is line clear for passenger express?"—four beats on the bell. B, seeing that the line is clear to his clearing point, sends back four beats, and pins down the white key of his instrument. "Line clear" appears on the opening, and also at that of A's keyless disc. A lowers starting signal. Train moves off. A gives two beats on the tapper = "Train entering section." B pins indicator at "Train on line," which also appears on A's instrument. A places signals at danger. B asks C, "Is line clear?" C repeats the bell code, and pins indicator at "Line clear," shown on B's keyless disc also. B lowers all signals. Train passes. B signals to C, "Train entering section." B signals to A, "Train out of section," and releases indicator, which returns to normal position with half of each flag showing at the window. B signals to C, "Train on line," and sets all his signals to danger. C pins indicator to "Train on line." C asks, "Is line clear?" But there is a train at station D, and signalman D therefore gives no reply, which is equivalent to a negative. The driver, on approaching C's distant, sees it at danger, and slows down, stopping at the home. C lowers home, and allows train to proceed to his starting signal. D, when the line is clear to his clearing point, signals "Line clear," and pins indicator at "Line clear." C lowers starting signals, and train proceeds. C signals to D, "Train entering section," and D pins indicator at "Train on line." C signals to B, "Train out of section," sets indicator at normal, and puts signals at danger. And so the process is repeated from station to station. Where, however, sections are short, the signalman is advised one section ahead of the approach of a train by an additional signal signifying, "Fast train approaching." The block indicator reminds the signalman of the whereabouts of the train. Unless his keyless indicator is at normal, he may not ask, "Is line clear?" And until he signals back "Line clear" to the box behind, a train is not allowed to enter his section. In this way a section of line with a full complement of signals is always interposed between any two trains.

THE WORKING OF SINGLE LINES.

We have dealt with the signalling arrangements pertaining to double lines of railway, showing that a system of signals is necessary to prevent a train running into the back of its predecessor. Where trains in both directions pass over a single line, not only has this element of danger to be dealt with, but also the possibility of a train being allowed to enter a section of line from each end at the same time. This is effected in several ways, the essence of each being that the engine-driver shall have in his possession visible evidence of the permission accorded him by the signalman to enter a section of single line.

A SINGLE TRAIN STAFF.

The simplest form of working is to allocate to the length of line a "train staff"—a piece of wood about 14 inches long, bearing the names of the stations at either end. This is adopted where only one engine is used for working a section, such as a short branch line. In a case like this there is obviously no danger of two trains meeting, and the train staff is merely the authority to the driver to start a journey. No telegraphic communication is necessary with such a system, and signals are placed only at the ends of the line.

TRAIN STAFF AND TICKET.

On long lengths of single line where more than one train has to be considered, the line is divided into blocks in the way already described for double lines, and a staff is assigned to each, the staffs for the various blocks differing from each other in shape and colour. The usual signals are provided at each station, and block telegraph instruments are employed, the only difference being that one disc, of the key pattern, is used for trains in both directions. On such a line it is, of course, possible that two or more trains may require to follow each other without any travelling intermediately in the opposite direction. This would be impossible if the staff passed uniformly to and fro in the block section; but it is arranged by the introduction of a train staff ticket used in conjunction with the staff.

No train is permitted to leave a staff station unless the staff for the section of line to be traversed is at the station; and the driver has the strictest possible instructions that he must see the staff. If a second train is required to follow, the staff is shown to the driver, and a train staff ticket handed him as his authority to proceed. If, however, the next train over the section will enter from the opposite end, the staff is handed to the driver.

To render this system as safe as possible, train staff tickets are of the same colour and shape as the staff for the section to which they apply, and are kept in a special box at the stations, the key being attached to the staff and the lock so arranged that the key cannot be withdrawn unless the box has been locked.

ELECTRIC TRAIN STAFF AND TABLET SYSTEMS.

These systems of working are developments of the last mentioned, by which are secured greater safety and ease in working the line. On some sections of single line circumstances often necessitate the running of several trains in one direction without a return train. For such cases the train staff ticket was introduced; but even on the best regulated lines it is not always possible to secure that the staff shall be at the station where it is required at the right time, and cases have arisen where, no train being available at the station where the staff was, it had to be taken to the other station by a man on foot, causing much delay to traffic. The electric train staff and tablet systems overcome this difficulty. Both work on much the same principle, and we will therefore describe the former.



At each end of a block section a train staff instrument (Fig. 101) is provided. In the base of these instruments are a number of train staffs, any one of which would be accepted by an engine-driver as permission to travel over the single line. The instruments are electrically connected, the mechanism securing that a staff can be withdrawn only by the co-operation of the signalman at each end of the section; that, when all the staffs are in the instruments, a staff may be withdrawn at either end; that, when a staff has been withdrawn, another cannot be obtained until the one out has been restored to one or other of the instruments. The safety of such a system is obvious, as also the assistance to the working by having a staff available for a train no matter from which end it is to enter the section.

The mechanism of the instruments is quite simple. A double-poled electro-magnet is energized by the depression of a key by the signalman at the further end of the block into which the train is to run, and by the turning of a handle by the signalman who requires to withdraw a staff. The magnet, being energized, is able to lift a mechanical lock, and permits the withdrawal of a staff. In its passage through the instrument the staff revolves a number of iron discs, which in turn raise or lower a switch controlling the electrical connections. This causes the electric currents actuating the electro-magnet to oppose each other, the magnetism to cease, and the lock to fall back, preventing another staff being withdrawn. It will naturally be asked, "How is the electrical system restored?" We have said that there were a number of staffs in each instrument—in other words, a given number of staffs, usually twenty, is assigned to the section. Assume that there are ten in each instrument, and that the switch in each is in its lower position. Now withdraw a staff, and one instrument has an odd, the other an even, number of staffs, and similarly one switch is raised while the other remains lowered, therefore the electrical circuit is "out of phase"—that is, the currents in the magnets of each staff instrument are opposed to one another, and cannot release the lock. The staff travels through the section and is placed in the instrument at the other end, bringing the number of staffs to eleven—an odd number, and, what is more important, raising the switch. Both switches are now raised, consequently the electric currents will support each other, so that a staff may be withdrawn. Briefly, then, when there is an odd number of staffs in one instrument and an even number in the other, as when a staff is in use, the signalmen are unable to obtain a staff, and consequently cannot give authority for a train to enter the section; but when there is either an odd or an even number of staffs in each instrument a staff may be withdrawn at either end on the co-operation of the signalmen.

We may add that, where two instruments are in the same signal-box, one for working to the box in advance, the other to the rear, it is arranged that the staffs pertaining to one section shall not fit the instrument for the other, and must be of different colours. This prevents the driver accidentally accepting a staff belonging to one section as authority to travel over the other.

INTERLOCKING.

The remarks made on the interlocking of points and signals on double lines apply also to the working of single lines, with the addition that not only are the distant, home, and starting signals interlocked with each other, but with the signals and points governing the approach of a train from the opposite direction—in other words, the signals for the approach of a train to a station from one direction cannot be lowered unless those for the approach to the station of a train from the opposite direction are at danger, and the points correctly set.

SIGNALLING OPERATIONS.

In the working of single lines, as of double, the signalman at the station from which a train is to proceed has to obtain the consent of the signalman ahead, the series of questions to be signalled being very similar to those detailed for double lines. There is, however, one notable exception. On long lengths of single line it is necessary to make arrangements for trains to pass each other. This is done by providing loop lines at intervals, a second pair of rails being laid for the accommodation of one train while another in the opposite direction passes it. To secure that more than one train shall not be on a section of single line between two crossing-places it is laid down that, when a signalman at a non-crossing station is asked to allow a train to approach his station, he must not give permission until he has notified the signalman ahead of him, thus securing that he is not asking permission for trains to approach from both directions at the same time. Both for single and double line working a number of rules designed to deal with cases of emergency are laid down, the guiding principle being safety; but we have now dealt with all the conditions of everyday working, and must pass to the consideration of



"POWER" SIGNALLING.

In a power system of signalling the signalman is provided with some auxiliary means—electricity, compressed air, etc.—of moving the signals or points under his control. It is still necessary to have a locking-frame in the signal-box, with levers interlocked with each other, and connections between the box and the various points and signals. But the frame is much smaller than an ordinary manual frame, and but little force is needed to move the little levers which make or break an electric circuit, or open an air-valve, according to the power-agent used.

ELECTRIC SIGNALLING.

Fig. 102 represents the locking-frame of a cabin at Didcot, England, where an all-electric system has been installed. Wires lead from the cabin to motors situated at the points and signals, which they operate through worm gearing. When a lever is moved it closes a circuit and sets the current flowing through a motor, the direction of the flow (and consequently of the motor's revolution) depending on whether the lever has been moved forward or backward. Indicators arranged under the levers tell the signalman when the desired movements at the points and signals have been completed. If any motion is not carried through, owing to failure of the current or obstruction of the working parts, an electric lock prevents him continuing operations. Thus, suppose he has to open the main line to an express, he is obliged by the mechanical locking-frame to set all the points correctly before the signals can be lowered. He might move all the necessary levers in due order, yet one set of points might remain open, and, were the signals lowered, an accident would result. But this cannot happen, as the electric locks worked by the points in question block the signal levers, and until the failure has been set right, the signals must remain at "danger."

The point motors are connected direct to the points; but between a signal motor and its arm there is an "electric slot," consisting of a powerful electro-magnet which forms a link in the rod work. To lower a signal it is necessary that the motor shall revolve and a control current pass round the magnet to give it the requisite attractive force. If no control current flows, as would happen were any pair of points not in their proper position, the motor can have no effect on the signal arm to lower it, owing to the magnet letting go its grip. Furthermore, if the signal had been already lowered when the control current failed, it would rise to "danger" automatically, as all signals are weighted to assume the danger position by gravity. The signal control currents can be broken by the signalman moving a switch, so that in case of emergency all signals may be thrown simultaneously to danger.

PNEUMATIC SIGNALLING.

In England and the United States compressed air is also used to do the hard labour of the signalman for him. Instead of closing a circuit, the signalman, by moving a lever half-way over, admits air to a pipe running along the track to an air reservoir placed beside the points or signal to which the lever relates. The air opens a valve and puts the reservoir in connection with a piston operating the points or signal-arm, as the case may be. This movement having been performed, another valve in the reservoir is opened, and air passes back through a second pipe to the signal-box, where it opens a third valve controlling a piston which completes the movement of the lever, so showing the signalman that the operation is complete. With compressed air, as with electricity, a mechanical locking-frame is of course used.

AUTOMATIC SIGNALLING.

To reduce expense, and increase the running speed on lines where the sections are short, the train is sometimes made to act as its own signalman. The rails of each section are all bonded together so as to be in metallic contact, and each section is insulated from the two neighbouring sections. At the further end of a section is installed an electric battery, connected to the rails, which lead the current back to a magnet operating a signal stationed some distance back on the preceding section. As long as current flows the signal is held at "All right." When a train enters the section the wheels and axles short-circuit the current, so that it does not reach the signal magnet, and the signal rises to "danger," and stays there until the last pair of wheels has passed out of the section. Should the current fail or a vehicle break loose and remain on the section, the same thing would happen.

The human element can thus be practically eliminated from signalling. To make things absolutely safe, a train should have positive control over a train following, to prevent the driver overrunning the signals. On electric railways this has been effected by means of contacts working in combination with the signals, which either cut the current off from the section preceding that on which a train may be, or raise a trigger to strike an arm on the train following and apply its brakes.



Chapter XII.

OPTICS.

Lenses—The image cast by a convex lens—Focus—Relative position of object and lens—Correction of lenses for colour—Spherical aberration—Distortion of image—The human eye—The use of spectacles—The blind spot.

Light is a third form of that energy of which we have already treated two manifestations—heat and electricity. The distinguishing characteristic of ether light-waves is their extreme rapidity of vibration, which has been calculated to range from 700 billion movements per second for violet rays to 400 billion for red rays.

If a beam of white light be passed through a prism it is resolved into the seven visible colours of the spectrum—violet, indigo, blue, green, yellow, orange, and red—in this order. The human eye is most sensitive to the yellow-red rays, a photographic plate to the green-violet rays.

All bodies fall into one of two classes—(1) Luminous—that is, those which are a source of light, such as the sun, a candle flame, or a red-hot coal; and (2) non-luminous, which become visible only by virtue of light which they receive from other bodies and reflect to our eyes.

THE PROPAGATION OF LIGHT.

Light naturally travels in a straight line. It is deflected only when it passes from one transparent medium into another—for example, from air to water—and the mediums are of different densities. We may regard the surface of a visible object as made up of countless points, from each of which a diverging pencil of rays is sent off through the ether.

LENSES.

If a beam of light encounters a transparent glass body with non-parallel sides, the rays are deflected. The direction they take depends on the shape of the body, but it may be laid down as a rule that they are bent toward the thicker part of the glass. The common burning-glass is well known to us. We hold it up facing the sun to concentrate all the heat rays that fall upon it into one intensely brilliant spot, which speedily ignites any inflammable substance on which it may fall (Fig. 103). We may imagine that one ray passes from the centre of the sun through the centre of the glass. This is undeflected; but all the others are bent towards it, as they pass through the thinner parts of the lens.



It should be noted here that sunlight, as we call it, is accompanied by heat. A burning-glass is used to concentrate the heat rays, not the light rays, which, though they are collected too, have no igniting effect.

In photography we use a lens to concentrate light rays only. Such heat rays as may pass through the lens with them are not wanted, and as they have no practical effect are not taken any notice of. To be of real value, a lens must be quite symmetrical—that is, the curve from the centre to the circumference must be the same in all directions.

There are six forms of simple lenses, as given in Fig. 104. Nos. 1 and 2 have one flat and one spherical surface. Nos. 3, 4, 5, 6 have two spherical surfaces. When a lens is thicker at the middle than at the sides it is called a convex lens; when thinner, a concave lens. The names of the various shapes are as follows:—No. 1, plano-convex; No. 2, plano-concave; No. 3, double convex; No. 4, double concave; No. 5, meniscus; No. 6, concavo-convex. The thick-centre lenses, as we may term them (Nos. 1, 3, 5), concentrate a pencil of rays passing through them; while the thin-centre lenses (Nos. 2, 4, 6) scatter the rays (see Fig. 105).



THE CAMERA.



We said above that light is propagated in straight lines. To prove this is easy. Get a piece of cardboard and prick a hole in it. Set this up some distance away from a candle flame, and hold behind it a piece of tissue paper. You will at once perceive a faint, upside-down image of the flame on the tissue. Why is this? Turn for a moment to Fig. 106, which shows a "pinhole" camera in section. At the rear is a ground-glass screen, B, to catch the image. Suppose that A is the lowest point of the flame. A pencil of rays diverging from it strikes the front of the camera, which stops them all except the one which passes through the hole and makes a tiny luminous spot on B, above the centre of the screen, though A is below the axis of the camera. Similarly the tip of the flame (above the axis) would be represented by a dot on the screen below its centre. And so on for all the millions of points of the flame. If we were to enlarge the hole we should get a brighter image, but it would have less sharp outlines, because a number of rays from every point of the candle would reach the screen and be jumbled up with the rays of neighbouring pencils. Now, though a good, sharp photograph may be taken through a pinhole, the time required is so long that photography of this sort has little practical value. What we want is a large hole for the light to enter the camera by, and yet to secure a distinct image. If we place a lens in the hole we can fulfil our wish. Fig. 107 shows a lens in position, gathering up a number of rays from a point, A, and focussing them on a point, B. If the lens has 1,000 times the area of the pinhole, it will pass 1,000 times as many rays, and the image of A will be impressed on a sensitized photographic plate 1,000 times more quickly.



THE IMAGE CAST BY A CONVEX LENS.

Fig. 108 shows diagrammatically how a convex lens forms an image. From A and B, the extremities of the object, a simple ray is considered to pass through the centre of the lens. This is not deflected at all. Two other rays from the same points strike the lens above and below the centre respectively. These are bent inwards and meet the central rays, or come to a focus with them at A^1 and B^1. In reality a countless number of rays would be transmitted from every point of the object and collected to form the image.



FOCUS.

We must now take special notice of that word heard so often in photographic talk—"focus." What is meant by the focus or focal length of a lens? Well, it merely signifies the distance between the optical centre of the lens and the plane in which the image is formed.



We must here digress a moment to draw attention to the three simple diagrams of Fig. 109. The object, O, in each case is assumed to be to the right of the lens. In the topmost diagram the object is so far away from the lens that all rays coming from a single point in it are practically parallel. These converge to a focus at F. If the distance between F and the centre of the lens is six inches, we say that the lens has a six-inch focal length. The focal length of a lens is judged by the distance between lens and image when the object is far away. To avoid confusion, this focal length is known as the principal focus, and is denoted by the symbol f. In the middle diagram the object is quite near the lens, which has to deal with rays striking its nearer surface at an acuter angle than before (reckoning from the centre). As the lens can only deflect their path to a fixed degree, they will not, after passing the lens, come together until they have reached a point, F^1, further from the lens than F. The nearer we approach O to the lens, the further away on the other side is the focal point, until a distance equal to that of F from the lens is reached, when the rays emerge from the glass in a parallel pencil. The rays now come to a focus no longer, and there can be no image. If O be brought nearer than the focal distance, the rays would diverge after passing through the lens.

RELATIVE POSITIONS OF OBJECT AND IMAGE.



From what has been said above we deduce two main conclusions—(1.) The nearer an object is brought to the lens, the further away from the lens will the image be. (2.) If the object approaches within the principal focal distance of the lens, no image will be cast by the lens. To make this plainer we append a diagram (Fig. 110), which shows five positions of an object and the relative positions of the image (in dotted lines). First, we note that the line A B, or A B^1, denotes the principal focal length of the lens, and A C, or A C^1, denotes twice the focal length. We will take the positions in order:—

Position I. Object further away than 2f. Inverted image smaller than object, at distance somewhat exceeding f.

Position II. Object at distance = 2f. Inverted image at distance = 2f, and of size equal to that of object.

Position III Object nearer than 2f. Inverted image further away than 2f; larger than the object.

Position IV. Object at distance = f. As rays are parallel after passing the lens no image is cast.

Position V. Object at distance less than f. No real image—that is, one that can be caught on a focussing screen—is now given by the lens, but a magnified, erect, virtual image exists on the same side of the lens as the object.

We shall refer to virtual images at greater length presently. It is hoped that any reader who practises photography will now understand why it is necessary to rack his camera out beyond the ordinary focal distance when taking objects at close quarters. From Fig. 110 he may gather one practically useful hint—namely, that to copy a diagram, etc., full size, both it and the plate must be exactly 2f from the optical centre of the lens. And it follows from this that the further he can rack his camera out beyond 2f the greater will be the possible enlargement of the original.

CORRECTION OF LENSES FOR COLOUR.

We have referred to the separation of the spectrum colours of white light by a prism. Now, a lens is one form of prism, and therefore sorts out the colours. In Fig. 111 we assume that two parallel red rays and two parallel violet rays from a distant object pass through a lens. A lens has most bending effect on violet rays and least on red, and the other colours of the spectrum are intermediately influenced. For the sake of simplicity we have taken the two extremes only. You observe that the point R, in which the red rays meet, is much further from the lens than is V, the meeting-point of the violet rays. A photographer very seldom has to take a subject in which there are not objects of several different colours, and it is obvious that if he used a simple lens like that in Fig. 111 and got his red objects in good focus, the blue and green portions of his picture would necessarily be more or less out of focus.



This defect can fortunately be corrected by the method shown in Fig. 112. A compound lens is needed, made up of a crown glass convex element, B, and a concave element, A, of flint glass. For the sake of illustration the two parts are shown separated; in practice they would be cemented together, forming one optical body, thicker in the centre than at the edges—a meniscus lens in fact, since A is not so concave as B is convex. Now, it was discovered by a Mr. Hall many years ago that if white light passed through two similar prisms, one of flint glass the other of crown glass, the former had the greater effect in separating the spectrum colours—that is, violet rays were bent aside more suddenly compared with the red rays than happened with the crown-glass prism. Look at Fig. 112. The red rays passing through the flint glass are but little deflected, while the violet rays turn suddenly outwards. This is just what is wanted, for it counteracts the unequal inward refraction by B, and both sets of rays come to a focus in the same plane. Such a lens is called achromatic, or colourless. If you hold a common reading-glass some distance away from large print you will see that the letters are edged with coloured bands, proving that the lens is not achromatic. A properly corrected photographic lens would not show these pretty edgings. Colour correction is necessary also for lenses used in telescopes and microscopes.

SPHERICAL ABERRATION.

A lens which has been corrected for colour is still imperfect. If rays pass through all parts of it, those which strike it near the edge will be refracted more than those near the centre, and a blurred focus results. This is termed spherical aberration. You will be able to understand the reason from Figs. 113 and 114. Two rays, A, are parallel to the axis and enter the lens near the centre (Fig. 113). These meet in one plane. Two other rays, B, strike the lens very obliquely near the edge, and on that account are both turned sharply upwards, coming to a focus in a plane nearer the lens than A. If this happened in a camera the results would be very bad. Either A or B would be out of focus. The trouble is minimized by placing in front of the lens a plate with a central circular opening in it (denoted by the thick, dark line in Fig. 114). The rays B of Fig. 113 are stopped by this plate, which is therefore called a stop. But other rays from the same point pass through the hole. These, however, strike the lens much more squarely above the centre, and are not unduly refracted, so that they are brought to a focus in the same plane as rays A.



DISTORTION OF IMAGE.



The lens we have been considering is a single meniscus, such as is used in landscape photography, mounted with the convex side turned towards the inside of the camera, and having the stop in front of it. If you possess a lens of this sort, try the following experiment with it. Draw a large square on a sheet of white paper and focus it on the screen. The sides instead of being straight bow outwards: this is called barrel distortion. Now turn the lens mount round so that the lens is outwards and the stop inwards. The sides of the square will appear to bow towards the centre: this is pin-cushion distortion. For a long time opticians were unable to find a remedy. Then Mr. George S. Cundell suggested that two meniscus lenses should be used in combination, one on either side of the stop, as in Fig 115. Each produces distortion, but it is counteracted by the opposite distortion of the other, and a square is represented as a square. Lenses of this kind are called rectilinear, or straight-line producing.

We have now reviewed the three chief defects of a lens—chromatic aberration, spherical aberration, and distortion—and have seen how they may be remedied. So we will now pass on to the most perfect of cameras,

THE HUMAN EYE.

The eye (Fig. 116) is nearly spherical in form, and is surrounded outside, except in front, by a hard, horny coat called the sclerotica (S). In front is the cornea (A), which bulges outwards, and acts as a transparent window to admit light to the lens of the eye (C). Inside the sclerotica, and next to it, comes the choroid coat; and inside that again is the retina, or curved focussing screen of the eye, which may best be described as a network of fibres ramifying from the optic nerve, which carries sight sensations to the brain. The hollow of the ball is full of a jelly-like substance called the vitreous humour; and the cavity between the lens and the cornea is full of water.

We have already seen that, in focussing, the distance between lens and image depends on the distance between object and lens. Now, the retina cannot be pushed nearer to or pulled further away from its lens, like the focussing screen of a camera. How, then, is the eye able to focus sharply objects at distances varying from a foot to many miles?



As a preliminary to the answer we must observe that the more convex a lens is, the shorter is its focus. We will suppose that we have a box camera with a lens of six-inch focus fixed rigidly in the position necessary for obtaining a sharp image of distant objects. It so happens that we want to take with it a portrait of a person only a few feet from the lens. If it were a bellows camera, we should rack out the back or front. But we cannot do this here. So we place in front of our lens a second convex lens which shortens its principal focus; so that in effect the box has been racked out sufficiently.

Nature, however, employs a much more perfect method than this. The eye lens is plastic, like a piece of india-rubber. Its edges are attached to ligaments (L L), which pull outwards and tend to flatten the curve of its surfaces. The normal focus is for distant objects. When we read a book the eye adapts itself to the work. The ligaments relax and the lens decreases in diameter while thickening at the centre, until its curvature is such as to focus all rays from the book sharply on the retina. If we suddenly look through the window at something outside, the ligaments pull on the lens envelope and flatten the curves.

This wonderful lens is achromatic, and free from spherical aberration and distortion of image. Nor must we forget that it is aided by an automatic "stop," the iris, the central hole of which is named the pupil. We say that a person has black, blue, or gray eyes according to the colour of the iris. Like the lens, the iris adapts itself to all conditions, contracting when the light is strong, and opening when the light is weak, so that as uniform an amount of light as conditions allow may be admitted to the eye. Most modern camera lenses are fitted with adjustable stops which can be made larger or smaller by twisting a ring on the mount, and are named "iris" stops. The image of anything seen is thrown on the retina upside down, and the brain reverses the position again, so that we get a correct impression of things.

THE USE OF SPECTACLES.



The reader will now be able to understand without much trouble the function of a pair of spectacles. A great many people of all ages suffer from short-sight. For one reason or another the distance between lens and retina becomes too great for a person to distinguish distant objects clearly. The lens, as shown in Fig 117a, is too convex—has its minimum focus too short—and the rays meet and cross before they reach the retina, causing general confusion of outline. This defect is simply remedied by placing in front of the eye (Fig. 117b) a concave lens, to disperse the rays somewhat before they enter the eye, so that they come to a focus on the retina. If a person's sight is thus corrected for distant objects, he can still see near objects quite plainly, as the lens will accommodate its convexity for them. The scientific term for short-sight is myopia. Long-sight, or hypermetropia, signifies that the eyeball is too short or the lens too flat. Fig. 118a represents the normal condition of a long-sighted eye. When looking at a distant object the eye thickens slightly and brings the focus forward into the retina. But its thickening power in such an eye is very limited, and consequently the rays from a near object focus behind the retina. It is therefore necessary for a long-sighted person to use convex spectacles for reading the newspaper. As seen in Fig. 118b, the spectacle lens concentrates the rays before they enter the eye, and so does part of the eye's work for it.

Returning for a moment to the diagram of the eye (Fig. 116), we notice a black patch on the retina near the optic nerve. This is the "yellow spot." Vision is most distinct when the image of the object looked at is formed on this part of the retina. The "blind spot" is that point at which the optic nerve enters the retina, being so called from the fact that it is quite insensitive to light. The finding of the blind spot is an interesting little experiment. On a card make a large and a small spot three inches apart, the one an eighth, the other half an inch in diameter. Bring the card near the face so that an eye is exactly opposite to each spot, and close the eye opposite to the smaller. Now direct the other eye to this spot and you will find, if the card be moved backwards and forwards, that at a certain distance the large spot, though many times larger than its fellow, has completely vanished, because the rays from it enter the open eye obliquely and fall on the "blind spot."



Chapter XIII.

THE MICROSCOPE, THE TELESCOPE, AND THE MAGIC-LANTERN.

The simple microscope—Use of the simple microscope in the telescope—The terrestrial telescope—The Galilean telescope—The prismatic telescope—The reflecting telescope—The parabolic mirror—The compound microscope—The magic-lantern—The bioscope—The plane mirror.

In Fig. 119 is represented an eye looking at a vase, three inches high, situated at A, a foot away. If we were to place another vase, B, six inches high, at a distance of two feet; or C, nine inches high, at three feet; or D, a foot high, at four feet, the image on the retina would in every case be of the same size as that cast by A. We can therefore lay down the rule that the apparent size of an object depends on the angle that it subtends at the eye.



To see a thing more plainly, we go nearer to it; and if it be very small, we hold it close to the eye. There is, however, a limit to the nearness to which it can be brought with advantage. The normal eye is unable to adapt its focus to an object less than about ten inches away, termed the "least distance of distinct vision."

THE SIMPLE MICROSCOPE.



A magnifying glass comes in useful when we want to examine an object very closely. The glass is a lens of short focus, held at a distance somewhat less than its principal focal length, F (see Fig. 120), from the object. The rays from the head and tip of the pin which enter the eye are denoted by continuous lines. As they are deflected by the glass the eye gets the impression that a much longer pin is situated a considerable distance behind the real object in the plane in which the refracted rays would meet if produced backwards (shown by the dotted lines). The effect of the glass, practically, is to remove it (the object) to beyond the least distance of distinct vision, and at the same time to retain undiminished the angle it subtends at the eye, or, what amounts to the same thing, the actual size of the image formed on the retina.[22] It follows, therefore, that if a lens be of such short focus that it allows us to see an object clearly at a distance of two inches—that is, one-fifth of the least distance of distinct vision—we shall get an image on the retina five times larger in diameter than would be possible without the lens.

The two simple diagrams (Figs. 121 and 122) show why the image to be magnified should be nearer to the lens than the principal focus, F. We have already seen (Fig. 109) that rays coming from a point in the principal focal plane emerge as a parallel pencil. These the eye can bring to a focus, because it normally has a curvature for focussing parallel rays. But, owing to the power of "accommodation," it can also focus diverging rays (Fig. 121), the eye lens thickening the necessary amount, and we therefore put our magnifying glass a bit nearer than F to get full advantage of proximity. If we had the object outside the principal focus, as in Fig. 122, the rays from it would converge, and these could not be gathered to a sharp point by the eye lens, as it cannot flatten more than is required for focussing parallel rays.



USE OF THE SIMPLE MICROSCOPE IN THE TELESCOPE.



Let us now turn to Fig. 123. At A is a distant object, say, a hundred yards away. B is a double convex lens, which has a focal length of twenty inches. We may suppose that it is a lens in a camera. An inverted image of the object is cast by the lens at C. If the eye were placed at C, it would distinguish nothing. But if withdrawn to D, the least distance of distinct vision,[23] behind C, the image is seen clearly. That the image really is at C is proved by letting down the focussing screen, which at once catches it. Now, as the focus of the lens is twice d, the image will be twice as large as the object would appear if viewed directly without the lens. We may put this into a very simple formula:—

Magnification = focal length of lens —————————— d



In Fig. 124 we have interposed between the eye and the object a small magnifying glass of 2-1/2-inch focus, so that the eye can now clearly see the image when one-quarter d away from it. B already magnifies the image twice; the eye-piece again magnifies it four times; so that the total magnification is 2 x 4 = 8 times. This result is arrived at quickly by dividing the focus of B (which corresponds to the object-glass of a telescope) by the focus of the eye-piece, thus:—

20 _ = 8 2-1/2

The ordinary astronomical telescope has a very long focus object-glass at one end of the tube, and a very short focus eye-piece at the other. To see an object clearly one merely has to push in or pull out the eye-piece until its focus exactly corresponds with that of the object-glass.

THE TERRESTRIAL TELESCOPE.

An astronomical telescope inverts images. This inversion is inconvenient for other purposes. So the terrestrial telescope (such as is commonly used by sailors) has an eye-piece compounded of four convex lenses which erect as well as magnify the image. Fig. 125 shows the simplest form of compound erecting eye-piece.



THE GALILEAN TELESCOPE.



A third form of telescope is that invented by the great Italian astronomer, Galileo,[24] in 1609. Its principle is shown in Fig. 126. The rays transmitted by the object-glass are caught, before coming to a focus, on a concave lens which separates them so that they appear to meet in the paths of convergence denoted by the dotted lines. The image is erect. Opera-glasses are constructed on the Galilean principle.

THE PRISMATIC TELESCOPE.

In order to be able to use a long-focus object-glass without a long focussing-tube, a system of glass reflecting prisms is sometimes employed, as in Fig. 127. A ray passing through the object-glass is reflected from one posterior surface of prism A on to the other posterior surface, and by it out through the front on to a second prism arranged at right angles to it, which passes the ray on to the compound eye-piece. The distance between object-glass and eye-piece is thus practically trebled. The best-known prismatic telescopes are the Zeiss field-glasses.



THE REFLECTING TELESCOPE.

We must not omit reference to the reflecting telescope, so largely used by astronomers. The front end of the telescope is open, there being no object-glass. Rays from the object fall on a parabolic mirror situated in the rear end of the tube. This reflects them forwards to a focus. In the Newtonian reflector a plane mirror or prism is situated in the axis of the tube, at the focus, to reflect the rays through an eye-piece projecting through the side of the tube. Herschel's form of reflector has the mirror set at an angle to the axis, so that the rays are reflected direct into an eye-piece pointing through the side of the tube towards the mirror.

THE PARABOLIC MIRROR.

This mirror (Fig. 128) is of such a shape that all rays parallel to the axis are reflected to a common point. In the marine searchlight a powerful arc lamp is arranged with the arc at the focus of a parabolic reflector, which sends all reflected light forward in a pencil of parallel rays. The most powerful searchlight in existence gives a light equal to that of 350 million candles.



THE COMPOUND MICROSCOPE.

We have already observed (Fig. 110) that the nearer an object approaches a lens the further off behind it is the real image formed, until the object has reached the focal distance, when no image at all is cast, as it is an infinite distance behind the lens. We will assume that a certain lens has a focus of six inches. We place a lighted candle four feet in front of it, and find that a sharp diminished image is cast on a ground-glass screen held seven inches behind it. If we now exchange the positions of the candle and the screen, we shall get an enlarged image of the candle. This is a simple demonstration of the law of conjugate foci—namely, that the distance between the lens and an object on one side and that between the lens and the corresponding image on the other bear a definite relation to each other; and an object placed at either focus will cast an image at the other. Whether the image is larger or smaller than the object depends on which focus it occupies. In the case of the object-glass of a telescope the image was at what we may call the short focus.



Now, a compound microscope is practically a telescope with the object at the long focus, very close to a short-focus lens. A greatly enlarged image is thrown (see Fig. 129) at the conjugate focus, and this is caught and still further magnified by the eye-piece. We may add that the object-glass, or objective, of a microscope is usually compounded of several lenses, as is also the eye-piece.

THE MAGIC-LANTERN.

The most essential features of a magic-lantern are:—(1) The source of light; (2) the condenser for concentrating the light rays on to the slide; (3) the lens for projecting a magnified image on to a screen.

Fig. 130 shows these diagrammatically. The illuminant is most commonly an oil-lamp, or an acetylene gas jet, or a cylinder of lime heated to intense luminosity by an oxy-hydrogen flame. The natural combustion of hydrogen is attended by a great heat, and when the supply of oxygen is artificially increased the temperature of the flame rises enormously. The nozzle of an oxy-hydrogen jet has an interior pipe connected with the cylinder holding one gas, and an exterior, and somewhat larger, pipe leading from that containing the other, the two being arranged concentrically at the nozzle. By means of valves the proportions of the gases can be regulated to give the best results.



The condenser is set somewhat further from the illuminant than the principal focal length of the lenses, so that the rays falling on them are bent inwards, or to the slide.

The objective, or object lens, stands in front of the slide. Its position is adjustable by means of a rack and a draw-tube. The nearer it is brought to the slide the further away is the conjugate focus (see p. 239), and consequently the image. The exhibitor first sets up his screen and lantern, and then finds the conjugate foci of slide and image by racking the lens in or out.

If a very short focus objective be used, subjects of microscopic proportions can be projected on the screen enormously magnified. During the siege of Paris in 1870-71 the Parisians established a balloon and pigeon post to carry letters which had been copied in a minute size by photography. These copies could be enclosed in a quill and attached to a pigeon's wing. On receipt, the copies were placed in a special lantern and thrown as large writing on the screen. Micro-photography has since then made great strides, and is now widely used for scientific purposes, one of the most important being the study of the crystalline formations of metals under different conditions.

THE BIOSCOPE.

"Living pictures" are the most recent improvement in magic-lantern entertainments. The negatives from which the lantern films are printed are made by passing a ribbon of sensitized celluloid through a special form of camera, which feeds the ribbon past the lens in a series of jerks, an exposure being made automatically by a revolving shutter during each rest. The positive film is placed in a lantern, and the intermittent movement is repeated; but now the source of illumination is behind the film, and light passes outwards through the shutter to the screen. In the Urban bioscope the film travels at the rate of fifteen miles an hour, upwards of one hundred exposures being made every second.

The impression of continuous movement arises from the fact that the eye cannot get rid of a visual impression in less than one-tenth of a second. So that if a series of impressions follow one another more rapidly than the eye can rid itself of them the impressions will overlap, and give one of motion, if the position of some of the objects, or parts of the objects, varies slightly in each succeeding picture.[25]

THE PLANE MIRROR.



This chapter may conclude with a glance at the common looking-glass. Why do we see a reflection in it? The answer is given graphically by Fig. 131. Two rays, A b, A c, from a point A strike the mirror M at the points b and c. Lines b N, c O, drawn from these points perpendicular to the mirror are called their normals. The angles A b N, A c O are the angles of incidence of rays A b, A c. The paths which the rays take after reflection must make angles with b N and c O respectively equal to A b N, A c O. These are the angles of reflection. If the eye is so situated that the rays enter it as in our illustration, an image of the point A is seen at the point A^1, in which the lines D b, E c meet when produced backwards.



When the vertical mirror is replaced by a horizontal reflecting surface, such as a pond (Fig. 132), the same thing happens. The point at which the ray from the reflection of the spire's tip to the eye appears to pass through the surface of the water must be so situated that if a line were drawn perpendicular to it from the surface the angles made by lines drawn from the real spire tip and from the observer's eye to the base of the perpendicular would be equal.

[22] Glazebrook, "Light," p. 157.

[23] Glazebrook, "Light," p. 157.

[24] Galileo was severely censured and imprisoned for daring to maintain that the earth moved round the sun, and revolved on its axis.

[25] For a full account of Animated Pictures the reader might advantageously consult "The Romance of Modern Invention," pp. 166 foll.



Chapter XIV.

SOUND AND MUSICAL INSTRUMENTS.

Nature of sound—The ear—Musical instruments—The vibration of strings—The sounding-board and the frame of a piano—The strings—The striking mechanism—The quality of a note.

Sound differs from light, heat, and electricity in that it can be propagated through matter only. Sound-waves are matter-waves, not ether-waves. This can be proved by placing an electric bell under the bell-glass of an air-pump and exhausting all the air. Ether still remains inside the glass, but if the bell be set in motion no sound is audible. Admit air, and the clang of the gong is heard quite plainly.

Sound resembles light and heat, however, thus far, that it can be concentrated by means of suitable lenses and curved surfaces. An echo is a proof of its reflection from a surface.

Before dealing with the various appliances used for producing sound-waves of a definite character, let us examine that wonderful natural apparatus

THE EAR,

through which we receive those sensations which we call sound.



Fig. 133 is a purely diagrammatic section of the ear, showing the various parts distorted and out of proportion. Beginning at the left, we have the outer ear, the lobe, to gather in the sound-waves on to the membrane of the tympanum, or drum, to which is attached the first of a series of ossicles, or small bones. The last of these presses against an opening in the inner ear, a cavity surrounded by the bones of the head. Inside the inner ear is a watery fluid, P, called perilymph ("surrounding water"), immersed in which is a membranic envelope, M, containing endolymph ("inside water"), also full of fluid. Into this fluid project E E E, the terminations of the auditory nerve, leading to the brain.

When sound-waves strike the tympanum, they cause it to move inwards and outwards in a series of rapid movements. The ossicles operated by the tympanum press on the little opening O, covered by a membrane, and every time they push it in they slightly squeeze the perilymph, which in turn compresses the endolymph, which affects the nerve-ends, and telegraphs a sensation of sound to the brain.

In Fig. 134 we have a more developed sketch, giving in fuller detail, though still not in their actual proportions, the components of the ear. The ossicles M, I, and S are respectively the malleus (hammer), incus (anvil), and stapes (stirrup). Each is attached by ligaments to the walls of the middle ear. The tympanum moves the malleus, the malleus the incus, and the incus the stapes, the last pressing into the opening O of Fig. 133, which is scientifically known as the fenestra ovalis, or oval window. As liquids are practically incompressible, nature has made allowance for the squeezing in of the oval window membrane, by providing a second opening, the round window, also covered with a membrane. When the stapes pushes the oval membrane in, the round membrane bulges out, its elasticity sufficing to put a certain pressure on the perilymph (indicated by the dotted portion of the inner ear).



The inner ear consists of two main parts, the cochlea—so called from its resemblance in shape to a snail's shell—and the semicircular canals. Each portion has its perilymph and endolymph, and contains a number of the nerve-ends, which are, however, most numerous in the cochlea. We do not know for certain what the functions of the canals and the cochlea are; but it is probable that the former enables us to distinguish between the intensity or loudness of sounds and the direction from which they come, while the latter enables us to determine the pitch of a note. In the cochlea are about 2,800 tiny nerve-ends, called the rods of Corti. The normal ear has such a range as to give about 33 rods to the semitone. The great scientist Helmholtz has advanced the theory that these little rods are like tiny tuning-forks, each responding to a note of a certain pitch; so that when a string of a piano is sounded and the air vibrations are transmitted to the inner ear, they affect only one of these rods and the part of the brain which it serves, and we have the impression of one particular note. It has been proved by experiment that a very sensitive ear can distinguish between sounds varying in pitch by only 1/64th of a semitone, or but half the range of any one Corti fibre. This difficulty Helmholtz gets over by suggesting that in such an ear two adjacent fibres are affected, but one more than the other.

A person who has a "good ear" for music is presumably one whose Corti rods are very perfect. Unlucky people like the gentleman who could only recognize one tune, and that because people took off their hats when it commenced, are physically deficient. Their Corti rods cannot be properly developed.

What applies to one single note applies also to the elements of a musical chord. A dozen notes may sound simultaneously, but the ear is able to assimilate each and blend it with its fellows; yet it requires a very sensitive and well-trained ear to pick out any one part of a harmony and concentrate the brain's attention on that part.

The ear has a much larger range than the eye. "While the former ranges over eleven octaves, but little more than a single octave is possible to the latter. The quickest vibrations which strike the eye, as light, have only about twice the rapidity of the slowest; whereas the quickest vibrations which strike the ear, as a musical sound, have more than two thousand times the rapidity of the slowest."[26] To come to actual figures, the ordinary ear is sensitive to vibrations ranging from 16 to 38,000 per second. The bottom and top notes of a piano make respectively about 40 and 4,000 vibrations a second. Of course, some ears, like some eyes, cannot comprehend the whole scale. The squeak of bats and the chirrup of crickets are inaudible to some people; and dogs are able to hear sounds far too shrill to affect the human auditory apparatus.

Not the least interesting part of this wonderful organ is the tympanic membrane, which is provided with muscles for altering its tension automatically. If we are "straining our ears" to catch a shrill sound, we tighten the membrane; while if we are "getting ready" for a deep, loud report like that of a gun, we allow the drum to slacken.

The Eustachian tube (Fig. 134) communicates with the mouth. Its function is probably to keep the air-pressure equal on both sides of the drum. When one catches cold the tube is apt to become blocked by mucus, causing unequal pressure and consequent partial deafness.

Before leaving this subject, it will be well to remind our more youthful readers that the ear is delicately as well as wonderfully made, and must be treated with respect. Sudden shouting into the ear, or a playful blow, may have most serious effects, by bursting the tympanum or injuring the arrangement of the tiny bones putting it in communication with the inner ear.

MUSICAL INSTRUMENTS.

These are contrivances for producing sonorous shocks following each other rapidly at regular intervals. Musical sounds are distinguished from mere noises by their regularity. If we shake a number of nails in a tin box, we get only a series of superimposed and chaotic sensations. On the other hand, if we strike a tuning-fork, the air is agitated a certain number of times a second, with a pleasant result which we call a note.

We will begin our excursion into the region of musical instruments with an examination of that very familiar piece of furniture,

THE PIANOFORTE,

which means literally the "soft-strong." By many children the piano is regarded as a great nuisance, the swallower-up of time which could be much more agreeably occupied, and is accordingly shown much less respect than is given to a phonograph or a musical-box. Yet the modern piano is a very clever piece of work, admirably adapted for the production of sweet melody—if properly handled. The two forms of piano now generally used are the upright, with vertical sound-board and wires, and the grand, with horizontal sound-board.[27]

THE VIBRATION OF STRINGS.

As the pianoforte is a stringed instrument, some attention should be given to the subject of the vibration of strings. A string in a state of tension emits a note when plucked and allowed to vibrate freely. The pitch of the note depends on several conditions:—(1) The diameter of the string; (2) the tension of the string; (3) the length of the string; (4) the substance of the string. Taking them in order:—(1.) The number of vibrations per second is inversely proportional to the diameter of the string: thus, a string one-quarter of an inch in diameter would vibrate only half as often in a given time as a string one-eighth of an inch in diameter. (2.) The length remaining the same, the number of vibrations is directly proportional to the square root of the tension: thus, a string strained by a 16-lb. weight would vibrate four times as fast as it would if strained by a 1-lb. weight. (3.) The number of vibrations is inversely proportional to the length of the string: thus, a one-foot string would vibrate twice as fast as a two-foot string, strained to the same tension, and of equal diameter and weight. (4.) Other things being equal, the rate of vibration is inversely proportional to the square root of the density of the substance: so that a steel wire would vibrate more rapidly than a platinum wire of equal diameter, length, and tension. These facts are important to remember as the underlying principles of stringed instruments.

Now, if you hang a wire from a cord, and hang a heavy weight from the wire, the wire will be in a state of high tension, and yield a distinct note if struck. But the volume of sound will be very small, much too small for a practical instrument. The surface of the string itself is so limited that it sets up but feeble motions in the surrounding air. Now hang the wire from a large board and strike it again. The volume of sound has greatly increased, because the string has transmitted its vibrations to the large surface of the board.

To get the full sound-value of the vibrations of a string, we evidently ought to so mount the string that it may influence a large sounding surface. In a violin this is effected by straining the strings over a "bridge" resting on a hollow box made of perfectly elastic wood. Draw the bow across a string. The loud sound heard proceeds not from the string only, but also from the whole surface of the box.

THE SOUNDING-BOARD AND FRAME OF A PIANO.

A piano has its strings strained across a frame of wood or steel, from a row of hooks in the top of the frame to a row of tapering square-ended pins in the bottom, the wires passing over sharp edges near both ends. The tuner is able, on turning a pin, to tension its strings till it gives any desired note. Readers may be interested to learn that the average tension of a string is 275 lbs., so that the total strain on the frame of a grand piano is anything between 20 and 30 tons.

To the back of the frame is attached the sounding-board, made of spruce fir (the familiar Christmas tree). This is obtained from Central and Eastern Europe, where it is carefully selected and prepared, as it is essential that the timber should be sawn in such a way that the grain of the wood runs in the proper direction.

THE STRINGS.

These are made of extremely strong steel wire of the best quality. If you examine the wires of your piano, you will see that they vary in thickness, the thinnest being at the treble end of the frame. It is found impracticable to use wires of the same gauge and the same tension throughout. The makers therefore use highly-tensioned thick wires for the bass, and finer, shorter wires for the treble, taking advantage of the three factors—weight, tension, and length—which we have noticed above. The wires for the deepest notes are wrapped round with fine copper wire to add to their weight without increasing their diameter at the tuning-pins. There are about 600 yards (roughly one-third of a mile) of wire in a grand piano.

THE STRIKING MECHANISM.

We now pass to the apparatus for putting the strings in a state of vibration. The grand piano mechanism shown in Fig. 135 may be taken as typical of the latest improvements. The essentials of an effective mechanism are:—(1) That the blow delivered shall be sharp and certain; (2) that the string shall be immediately "damped," or have its vibration checked if required, so as not to interfere with the succeeding notes of other strings; (3) that the hammer shall be able to repeat the blows in quick succession. The hammer has a head of mahogany covered with felt, the thickness of which tapers gradually and regularly from an inch and a quarter at the bass end to three-sixteenths of an inch at the extreme treble notes. The entire eighty-five hammers for the piano are covered all together in one piece, and then they are cut apart from each other. The consistency of the covering is very important. If too hard, it yields a harsh note, and must be reduced to the right degree by pricking with a needle. In the diagram the felt is indicated by the dotted part.



The action carriage which operates the hammer is somewhat complicated. When the key is depressed, the left end rises, and pushes up the whole carriage, which is pivoted at one end. The hammer shank is raised by the jack B pressing upon a knob, N, called the notch, attached to the under side of the shank. When the jack has risen to a certain point, its arm, B^1, catches against the button C and jerks it from under the notch at the very moment when the hammer strikes, so that it may not be blocked against the string. As it rebounds, the hammer is caught on the repetition lever R, which lifts it to allow of perfect repetition.

The check catches the tail of the hammer head during its descent when the key is raised, and prevents it coming back violently on the carriage and rest. The tail is curved so as to wedge against the check without jamming in any way. The moment the carriage begins to rise, the rear end of the key lifts a lever connected with the damper by a vertical wire, and raises the damper of the string. If the key is held down, the vibrations continue for a long time after the blow; but if released at once, the damper stifles them as the hammer regains its seat. A bar, L, passing along under all the damper lifters, is raised by depressing the loud pedal. The soft pedal slides the whole keyboard along such a distance that the hammers strike two only out of the three strings allotted to all except the bass notes, which have only one string apiece, or two, according to their depth or length. In some pianos the soft pedal presses a special damper against the strings; and a third kind of device moves the hammers nearer the strings so that they deliver a lighter blow. These two methods of damping are confined to upright pianos.

A high-class piano is the result of very careful workmanship. The mechanism of each note must be accurately regulated by its tiny screws to a minute fraction of an inch. It must be ensured that every hammer strikes its blow at exactly the right place on the string, since on this depends the musical value of the note. The adjustment of the dampers requires equal care, and the whole work calls for a sensitive ear combined with skilled mechanical knowledge, so that the instrument may have a light touch, strength, and certainty of action throughout the whole keyboard.

THE QUALITY OF A NOTE.

If two strings, alike in all respects and equally tensioned, are plucked, both will give the same note, but both will not necessarily have the same quality of tone. The quality, or timbre, as musicians call it, is influenced by the presence of overtones, or harmonics, in combination with the fundamental, or deepest, tone of the string. The fact is, that while a vibrating string vibrates as a whole, it also vibrates in parts. There are, as it were, small waves superimposed on the big fundamental waves. Points of least motion, called nodes, form on the string, dividing it into two, three, four, five, etc., parts, which may be further divided by subsidiary nodes. The string, considered as halved by one node, gives the first overtone, or octave of the fundamental. It may also vibrate as three parts, and give the second overtone, or twelfth of the fundamental;[28] and as four parts, and give the third overtone, the double octave.

Now, if a string be struck at a point corresponding to a node, the overtones which require that point for a node will be killed, on account of the excessive motion imparted to the string at that spot. Thus to hit it at the middle kills the octave, the double octave, etc.; while to hit it at a point one-third of the length from one end stifles the twelfth and all its sub-multiples.

A fundamental note robbed of all its harmonics is hard to obtain, which is not a matter for regret, as it is a most uninteresting sound. To get a rich tone we must keep as many useful harmonics as possible, and therefore a piano hammer is so placed as to strike the string at a point which does not interfere with the best harmonics, but kills those which are objectionable. Pianoforte makers have discovered by experiment that the most pleasing tone is excited when the point against which the hammer strikes is one-seventh to one-ninth of the length of the wire from one end.

The nature of the material which does the actual striking is also of importance. The harder the substance, and the sharper the blow, the more prominent do the harmonics become; so that the worker has to regulate carefully both the duration of the blow and the hardness of the hammer covering.

[26] Tyndall, "On Sound," p. 75.

[27] A Broadwood "grand" is made up of 10,700 separate pieces, and in its manufacture forty separate trades are concerned.

[28] Twelve notes higher up the scale.



Chapter XV.

WIND INSTRUMENTS.

Longitudinal vibration—Columns of air—Resonance of columns of air—Length and tone—The open pipe—The overtones of an open pipe—Where overtones are used—The arrangement of the pipes and pedals—Separate sound-boards—Varieties of stops—Tuning pipes and reeds—The bellows—Electric and pneumatic actions—The largest organ in the world—Human reeds.

LONGITUDINAL VIBRATION.

In stringed instruments we are concerned only with the transverse vibrations of a string—that is, its movements in a direction at right angles to the axis of the string. A string can also vibrate longitudinally—that is, in the direction of its axis—as may be proved by drawing a piece of resined leather along a violin string. In this case the harmonics "step up" at the same rate as when the movements were transverse.

Let us substitute for a wire a stout bar of metal fixed at one end only. The longitudinal vibrations of this rod contain overtones of a different ratio. The first harmonic is not an octave, but a twelfth. While a tensioned string is divided by nodes into two, three, four, five, six, etc., parts, a rod fixed at one end only is capable of producing only those harmonics which correspond to division into three, five, seven, nine, etc., parts. Therefore a free-end rod and a wire of the same fundamental note would not have the same timbre, or quality, owing to the difference in the harmonics.

COLUMNS OF AIR.

In wind instruments we employ, instead of rods or wires, columns of air as the vibrating medium. The note of the column depends on its length. In the "penny whistle," flute, clarionet, and piccolo the length of the column is altered by closing or opening apertures in the substance encircling the column.

RESONANCE OF COLUMNS OF AIR.

Why does a tube closed at one end, such as the shank of a key, emit a note when we blow across the open end? The act of blowing drives a thin sheet of air against the edge of the tube and causes it to vibrate. The vibrations are confused, some "pulses" occurring more frequently than others. If we blew against the edge of a knife or a piece of wood, we should hear nothing but a hiss. But when, as in the case which we are considering, there is a partly-enclosed column of air close to the pulses, this selects those pulses which correspond to its natural period of vibration, and augments them to a sustained and very audible musical sound.



In Fig. 136, 1 is a pipe, closed at the bottom and open at the top. A tuning-fork of the same note as the pipe is struck and held over it so that the prongs vibrate upwards and downwards. At the commencement of an outward movement of the prongs the air in front of them is compressed. This impulse, imparted to the air in the pipe, runs down the column, strikes the bottom, and returns. Just as it reaches the top the prong is beginning to move inwards, causing a rarefaction of the air behind it. This effect also travels down and back up the column of air in the pipe, reaching the prong just as it arrives at the furthest point of the inward motion. The process is repeated, and the column of air in the pipe, striking on the surrounding atmosphere at regular intervals, greatly increases the volume of sound. We must observe that if the tuning-fork were of too high or too low a note for the column of air to move in perfect sympathy with it, this increase of sound would not result. Now, when we blow across the end, we present, as it were, a number of vibrating tuning-forks to the pipe, which picks out those air-pulses with which it sympathizes.

LENGTH AND TONE.

The rate of vibration is found to be inversely proportional to the length of the pipe. Thus, the vibrations of a two-foot pipe are twice as rapid as those of a four-foot pipe, and the note emitted by the former is an octave higher than that of the latter. A one-foot pipe gives a note an octave higher still. We are here speaking of the fundamental tones of the pipes. With them, as in the case of strings, are associated the overtones, or harmonics, which can be brought into prominence by increasing the pressure of the blast at the top of the pipe. Blow very hard on your key, and the note suddenly changes to one much shriller. It is the twelfth of the fundamental, of which it has completely got the upper hand.

We must now put on our thinking-caps and try to understand how this comes about. First, let us note that the vibration of a body (in this case a column of air) means a motion from a point of rest to a point of rest, or from node to node. In the air-column in Fig. 136, 1, there is only one point of rest for an impulse—namely, at the bottom of the pipe. So that to pass from node to node the impulse must pass up the pipe and down again. The distance from node to node in a vibrating body is called a ventral segment. Remember this term. Therefore the pipe represents a semi-ventral segment when the fundamental note is sounding.

When the first overtone is sounded the column divides itself into two vibrating parts. Where will the node between them be? We might naturally say, "Half-way up." But this cannot be so; for if the node were so situated, an impulse going down the pipe would only have to travel to the bottom to find another node, while an impulse going up would have to travel to the top and back again—that is, go twice as far. So the node forms itself one-third of the distance down the pipe. From B to A (Fig. 136, 2) and back is now equal to from B to C. When the second overtone is blown (Fig. 136, 3) a third node forms. The pipe is now divided into five semi-ventral segments. And with each succeeding overtone another node and ventral segment are added.

The law of vibration of a column of air is that the number of vibrations is directly proportional to the number of semi-ventral segments into which the column of air inside the pipe is divided.[29] If the fundamental tone gives 100 vibrations per second, the first overtone in a closed pipe must give 300, and the second 500 vibrations.

THE OPEN PIPE.

A pipe open at both ends is capable of emitting a note. But we shall find, if we experiment, that the note of a stopped pipe is an octave lower than that of an open pipe of equal length. This is explained by Fig. 137, 1. The air-column in the pipe (of the same length as that in Fig. 136) divides itself, when an end is blown across, into two equal portions at the node B, the natural point to obtain equilibrium. A pulse will pass from A or A^1 to B and back again in half the time required to pass from A to B and back in Fig. 136, 1; therefore the note is an octave higher.



THE OVERTONES OF AN OPEN PIPE.

The first overtone results when nodes form as in Fig. 137, 2, at points one-quarter of the length of the pipe from the ends, giving one complete ventral segment and two semi-ventral segments. The vibrations now are twice as rapid as before. The second overtone requires three nodes, as in Fig. 137, 3. The rate has now trebled. So that, while the overtones of a closed pipe rise in the ratio 1, 3, 5, 7, etc., those of an open pipe rise in the proportion 1, 2, 3, 4, etc.

WHERE OVERTONES ARE USED.

In the flute, piccolo, and clarionet, as well as in the horn class of instrument, the overtones are as important as the fundamental notes. By artificially altering the length of the column of air, the fundamental notes are also altered, while the harmonics of each fundamental are produced at will by varying the blowing pressure; so that a continuous chromatic, or semitonal, scale is possible throughout the compass of the instrument.

THE ORGAN.

From the theory of acoustics[30] we pass to the practical application, and concentrate our attention upon the grandest of all wind instruments, the pipe organ. This mechanism has a separate pipe for every note, properly proportioned. A section of an ordinary wooden pipe is given in Fig. 138. Wind rushes up through the foot of the pipe into a little chamber, closed by a block of wood or a plate except for a narrow slit, which directs it against the sharp lip A, and causes a fluttering, the proper pulse of which is converted by the air-column above into a musical sound.



In even the smallest organs more than one pipe is actuated by one key on the keyboard, for not only do pipes of different shapes give different qualities of tone, but it is found desirable to have ranks of pipes with their bottom note of different pitches. The length of an open pipe is measured from the edge of the lip to the top of the pipe; of a stopped pipe, from the lip to the top and back again. When we speak of a 16 or 8 foot rank, or stop, we mean one of which the lowest note in the rank is that produced by a 16 or 8 foot open pipe, or their stopped equivalents (8 or 4 foot). In a big organ we find 32, 16, 8, 4, and 2 foot stops, and some of these repeated a number of times in pipes of different shape and construction.

THE ARRANGEMENT OF THE PIPES.

We will now study briefly the mechanism of a very simple single-keyboard organ, with five ranks of pipes, or stops.



It is necessary to arrange matters so that the pressing down of one key may make all five of the pipes belonging to it speak, or only four, three, two, or one, as we may desire. The pipes are mounted in rows on a sound-board, which is built up in several layers. At the top is the upper board; below it come the sliders, one for each stop; and underneath that the table. In Fig. 139 we see part of the table from below. Across the under side are fastened parallel bars with spaces (shown black) left between them. Two other bars are fastened across the ends, so that each groove is enclosed by wood at the top and on all sides. The under side of the table has sheets of leather glued or otherwise attached to it in such a manner that no air can leak from one groove to the next. Upper board, sliders, and table are pierced with rows of holes, to permit the passage of wind from the grooves to the pipes. The grooves under the big pipes are wider than those under the small pipes, as they have to pass more air. The bars between the grooves also vary in width according to the weight of the pipes which they have to carry. The sliders can be moved in and out a short distance in the direction of the axis of the rows of pipes. There is one slider under each row. When a slider is in, the holes in it do not correspond with those in the table and upper board, so that no wind can get from the grooves to the rank over that particular slider. Fig. 140 shows the manner in which the sliders are operated by the little knobs (also called stops) projecting from the casing of the organ within convenient reach of the performer's hands. One stop is in, the other drawn out.



In Fig. 141 we see the table, etc., in cross section, with a slider out, putting the pipes of its rank in communication with the grooves. The same diagram shows us in section the little triangular pallets which admit air from the wind-chest to the grooves; and Fig. 142 gives us an end section of table, sliders, and wind-chest, together with the rods, etc., connecting the key to its pallet. When the key is depressed, the sticker (a slight wooden rod) is pushed up. This rocks a backfall, or pivoted lever, to which is attached the pulldown, a wire penetrating the bottom of the wind-chest to the pallet. As soon as the pallet opens, wind rushes into the groove above through the aperture in the leather bottom, and thence to any one of the pipes of which the slider has been drawn out. (The sliders in Fig. 142 are solid black.) It is evident that if the sound-board is sufficiently deep from back to front, any number of rows of pipes may be placed on it.



PEDALS.

The organ pedals are connected to the pallets by an action similar to that of the keys. The pedal stops are generally of deep tone, 32-foot and 16-foot, as they have to sustain the bass part of the musical harmonies. By means of couplers one or more of the keyboard stops may be linked to the pedals.

SEPARATE SOUND-BOARDS.

The keyboard of a very large organ has as many as five manuals, or rows of keys. Each manual operates what is practically a separate organ mounted on its own sound-board.



The manuals are arranged in steps, each slightly overhanging that below. Taken in order from the top, they are:—(1.) Echo organ, of stops of small scale and very soft tone, enclosed in a "swell-box." (2.) Solo organ, of stops imitating orchestral instruments. The wonderful "vox humana" stop also belongs to this manual. (3.) Swell organ, contained in a swell-box, the front and sides of which have shutters which can be opened and closed by the pressure of the foot on a lever, so as to regulate the amount of sound proceeding from the pipes inside. (4.) Great organ, including pipes of powerful tone. (5.) Choir organ, of soft, mellow stops, often enclosed in a swell-box. We may add to these the pedal organ, which can be coupled to any but the echo manual.

VARIETIES OF STOPS.

We have already remarked that the quality of a stop depends on the shape and construction of the pipe. Some pipes are of wood, others of metal. Some are rectangular, others circular. Some have parallel sides, others taper or expand towards the top. Some are open, others stopped.

The two main classes into which organ pipes may be divided are:—(1.) Flue pipes, in which the wind is directed against a lip, as in Fig. 138. (2.) Reed pipes—that is, pipes used in combination with a simple device for admitting air into the bottom of the pipe in a series of gusts. Fig. 144 shows a striking reed, such as is found in the ordinary motor horn. The elastic metal tongue when at rest stands a very short distance away from the orifice in the reed. When wind is blown through the reed the tongue is sucked against the reed, blocks the current, and springs away again. A free reed has a tongue which vibrates in a slot without actually touching the sides. Harmonium and concertina reeds are of this type. In the organ the reed admits air to a pipe of the correct length to sympathize with the rate of the puffs of air which the reed passes. Reed pipes expand towards the top.

TUNING PIPES AND REEDS.



Pipes are tuned by adjusting their length. The plug at the top of a stopped pipe is pulled out or pushed in a trifle to flatten or sharpen the note respectively. An open pipe, if large, has a tongue cut in the side at the top, which can be pressed inwards or outwards for the purpose of correcting the tone. Small metal pipes are flattened by contracting the tops inwards with a metal cone like a candle-extinguisher placed over the top and tapped; and sharpened by having the top splayed by a cone pushed in point downwards. Reeds of the striking variety (see Fig. 144) have a tuning-wire pressing on the tongue near the fixed end. The end of this wire projects through the casing. By moving it, the length of the vibrating part of the tongue is adjusted to correctness.

BELLOWS.

Different stops require different wind-pressures, ranging from 1/10 lb. to 1 lb. to the square inch, the reeds taking the heaviest pressures. There must therefore be as many sets of bellows and wind-chests as there are different pressures wanted. A very large organ consumes immense quantities of air when all the stops are out, and the pumping has to be done by a powerful gas, water, or electric engine. Every bellows has a reservoir (see Fig. 143) above it. The top of this is weighted to give the pressure required. A valve in the top opens automatically as soon as the reservoir has expanded to a certain fixed limit, so that there is no possibility of bursting the leather sides.



ELECTRIC AND PNEUMATIC ACTIONS.

We have mentioned in connection with railway signalling that the signalman is sometimes relieved of the hard manual labour of moving signals and points by the employment of electric and pneumatic auxiliaries. The same is true of organs and organists. The touch of the keys has been greatly lightened by making the keys open air-valves or complete electric circuits which actuate the mechanism for pulling down the pallets. The stops, pedals, and couplers also employ "power." Not only are the performer's muscles spared a lot of heavy work when compressed air and electricity aid him, but he is able to have the console, or keyboard, far away from the pipes. "From the console, the player, sitting with the singers, or in any desirable part of the choir or chancel, would be able to command the working of the whole of the largest organ situated afar at the western end of the nave; would draw each stop in complete reliance on the sliders and the sound-board fulfilling their office; ... and—marvel of it all—the player, using the swell pedal in his ordinary manner, would obtain crescendo and diminuendo with a more perfect effect than by the old way."[31]

In cathedrals it is no uncommon thing for the different sound-boards to be placed in positions far apart, so that to the uninitiated there may appear to be several independent organs scattered about. Yet all are absolutely under the control of a man who is sitting away from them all, but connected with them by a number of tubes or wires.

The largest organ in the world is that in the Town Hall, Sydney. It has a hundred and twenty-six speaking stops, five manuals, fourteen couplers, and forty-six combination studs. The pipes, about 8,000 in number, range from the enormous 64-foot contra-trombone to some only a fraction of an inch in length. The organ occupies a space 85 feet long and 26 feet deep.

HUMAN REEDS.

The most wonderful of all musical reeds is found in the human throat, in the anatomical part called the larynx, situated at the top of the trachea, or windpipe.

Slip a piece of rubber tubing over the end of a pipe, allowing an inch or so to project. Take the free part of the tube by two opposite points between the first fingers and thumbs and pull it until the edges are stretched tight. Now blow through it. The wind, forcing its way between the two rubber edges, causes them and the air inside the tube to vibrate, and a musical note results. The more you strain the rubber the higher is the note.

The larynx works on this principle. The windpipe takes the place of the glass pipe; the two vocal cords represent the rubber edges; and the arytenoid muscles stand instead of the hands. When contracted, these muscles bring the edges of the cords nearer to one another, stretch the cords, and shorten the cords. A person gifted with a "very good ear" can, it has been calculated, adjust the length of the vocal cords to 1/17000th of an inch!

Simultaneously with the adjustment of the cords is effected the adjustment of the length of the windpipe, so that the column of air in it may be of the right length to vibrate in unison. Here again is seen a wonderful provision of nature.

The resonance of the mouth cavity is also of great importance. By altering the shape of the mouth the various harmonics of any fundamental note produced by the larynx are rendered prominent, and so we get the different vocal sounds. Helmholtz has shown that the fundamental tone of any note is represented by the sound oo. If the mouth is adjusted to bring out the octave of the fundamental, o results. a is produced by accentuating the second harmonic, the twelfth; ee by developing the second and fourth harmonics; while for ah the fifth and seventh must be prominent.

When we whistle we transform the lips into a reed and the mouth into a pipe. The tension of the lips and the shape of the mouth cavity decide the note. The lips are also used as a reed for blowing the flute, piccolo, and all the brass band instruments of the cornet order. In blowing a coach-horn the various harmonics of the fundamental note are brought out by altering the lip tension and the wind pressure. A cornet is practically a coach-horn rolled up into a convenient shape and furnished with three keys, the depression of which puts extra lengths of tubing in connection with the main tube—in fact, makes it longer. One key lowers the fundamental note of the horn half a tone; the second, a full tone; the third, a tone and a half. If the first and third are pressed down together, the note sinks two tones; if the second and third, two and a half tones; and simultaneous depression of all three gives a drop of three tones. The performer thus has seven possible fundamental notes, and several harmonics of each of these at his command; so that by a proper manipulation of the keys he can run up the chromatic scale.

We should add that the cornet tube is an "open" pipe. So is that of the flute. The clarionet is a "stopped" pipe.

[29] It is obvious that in Fig. 136, 2, a pulse will pass from A to B and back in one-third the time required for it to pass from A to B and back in Fig. 136, 1.

[30] The science of hearing; from the Greek verb, [Greek: akouein], "to hear."

[31] "Organs and Tuning," p. 245.



Chapter XVI.

TALKING-MACHINES.

The phonograph—The recorder—The reproducer—The gramophone—The making of records—Cylinder records—Gramophone records.

In the Patent Office Museum at South Kensington is a curious little piece of machinery—a metal cylinder mounted on a long axle, which has at one end a screw thread chased along it. The screw end rotates in a socket with a thread of equal pitch cut in it. To the other end is attached a handle. On an upright near the cylinder is mounted a sort of drum. The membrane of the drum carries a needle, which, when the membrane is agitated by the air-waves set up by human speech, digs into a sheet of tinfoil wrapped round the cylinder, pressing it into a helical groove turned on the cylinder from end to end. This construction is the first phonograph ever made. Thomas Edison, the "wizard of the West," devised it in 1876; and from this rude parent have descended the beautiful machines which record and reproduce human speech and musical sounds with startling accuracy.



We do not propose to trace here the development of the talking-machine; nor will it be necessary to describe in detail its mechanism, which is probably well known to most readers, or could be mastered in a very short time on personal examination. We will content ourselves with saying that the wax cylinder of the phonograph, or the ebonite disc of the gramophone, is generally rotated by clockwork concealed in the body of the machine. The speed of rotation has to be very carefully governed, in order that the record may revolve under the reproducing point at a uniform speed. The principle of the governor commonly used appears in Fig. 146. The last pinion of the clockwork train is mounted on a shaft carrying two triangular plates, A and C, to which are attached three short lengths of flat steel spring with a heavy ball attached to the centre of each. A is fixed; C moves up the shaft as the balls fly out, and pulls with it the disc D, which rubs against the pad P (on the end of a spring) and sets up sufficient friction to slow the clockwork. The limit rate is regulated by screw S.

THE PHONOGRAPH.

Though the recording and reproducing apparatus of a phonograph gives very wonderful results, its construction is quite simple. At the same time, it must be borne in mind that an immense amount of experimenting has been devoted to finding out the most suitable materials and forms for the parts.



The recorder (Fig. 147) is a little circular box about one and a half inches in diameter.[32] From the top a tube leads to the horn. The bottom is a circular plate, C C, hinged at one side. This plate supports a glass disc, D, about 1/150th of an inch thick, to which is attached the cutting stylus—a tiny sapphire rod with a cup-shaped end having very sharp edges. Sound-waves enter the box through the horn tube; but instead of being allowed to fill the whole box, they are concentrated by the shifting nozzle N on to the centre of the glass disc through the hole in C C. You will notice that N has a ball end, and C C a socket to fit N exactly, so that, though C C and N move up and down very rapidly, they still make perfect contact. The disc is vibrated by the sound-impulses, and drives the cutting point down into the surface of the wax cylinder, turning below it in a clockwork direction. The only dead weight pressing on S is that of N, C C, and the glass diaphragm.



As the cylinder revolves, the recorder is shifted continuously along by a leading screw having one hundred or more threads to the inch cut on it, so that it traces a continuous helical groove from one end of the wax cylinder to the other. This groove is really a series of very minute indentations, not exceeding 1/1000th of an inch in depth.[33] Seen under a microscope, the surface of the record is a succession of hills and valleys, some much larger than others (Fig. 151, a). A loud sound causes the stylus to give a vigorous dig, while low sounds scarcely move it at all. The wonderful thing about this sound-recording is, that not only are the fundamental tones of musical notes impressed, but also the harmonics, which enable us to decide at once whether the record is one of a cornet, violin, or banjo performance. Furthermore, if several instruments are playing simultaneously near the recorder's horn, the stylus catches all the different shades of tone of every note of a chord. There are, so to speak, minor hills and valleys cut in the slopes of the main hills and valleys.



The reproducer (Fig. 149) is somewhat more complicated than the recorder. As before, we have a circular box communicating with the horn of the instrument. A thin glass disc forms a bottom to the box. It is held in position between rubber rings, R R, by a screw collar, C. To the centre is attached a little eye, from which hangs a link, L. Pivoted at P from one edge of the box is a floating weight, having a circular opening immediately under the eye. The link passes through this to the left end of a tiny lever, which rocks on a pivot projecting from the weight. To the right end of the lever is affixed a sapphire bar, or stylus, with a ball end of a diameter equal to that of the cutting point of the recorder. The floating weight presses the stylus against the record, and also keeps the link between the rocking lever of the glass diaphragm in a state of tension. Every blow given to the stylus is therefore transmitted by the link to the diaphragm, which vibrates and sends an air-impulse into the horn. As the impulses are given at the same rate as those which agitated the diaphragm of the recorder, the sounds which they represent are accurately reproduced, even to the harmonics of a musical note.

THE GRAMOPHONE.

This effects the same purpose as the phonograph, but in a somewhat different manner. The phonograph recorder digs vertically downwards into the surface of the record, whereas the stylus of the gramophone wags from side to side and describes a snaky course (Fig. 151b). It makes no difference in talking-machines whether the reproducing stylus be moved sideways or vertically by the record, provided that motion is imparted by it to the diaphragm.



In Fig. 151c the construction of the gramophone reproducer is shown in section. A is the cover which screws on to the bottom B, and confines the diaphragm D between itself and a rubber ring. The portion B is elongated into a tubular shape for connection with the horn, an arm of which slides over the tube and presses against the rubber ring C to make an air-tight joint. The needle-carrier N is attached at its upper end to the centre of the diaphragm. At a point indicated by the white dot a pin passes through it and the cover. The lower end is tubular to accommodate the steel points, which have to be replaced after passing once over a record. A screw, S, working in a socket projecting from the carrier, holds the point fast. The record moves horizontally under the point in a plane perpendicular to the page. The groove being zigzag, the needle vibrates right and left, and rotating the carrier a minute fraction of an inch on the pivot, shakes the glass diaphragm and sends waves of air into the horn.