wax is such a poor conductor that an inappreciable amount of its charge is removed. It is because the two can be brought so close together that the inductive action is so strong and that so large a charge is obtained on the plate.

There is a question that is often asked at this point. Some form of it may have already occurred to the student. The conducting plate, when charged, certainly has energy; for the spark produced on the discharge produces some heat and noise. For example, the spark is at a high enough temperature to ignite a gas jet. If this plate does not take any charge away from the charged sealing-wax, where does it get its energy? A similar question might have been asked about the conductor B of Fig. 227; for after B is discharged, A could be brought near again and B charged as before, and the process be repeated over and over again without causing A to lose any of its charge. The answer is that the energy is supplied by the operator, who does work when the plate is removed from the sealing-wax or, in the other case, when A and B are separated. The negative charge on one attracts the positive charge on the other, and work must be done in separating them. It is a very small amount of work, usually not detected, yet it is enough to give the results described.

Several types of revolving electrostatic machines have been designed which use the principle of the electrophorus. Although producing rather long sparks and attracting much attention when in operation, the electrical power, or rate of developing electrical energy, is usually less than that which can be obtained from an ordinary dry cell. As the complete theory of these machines is rather complicated, the student is referred to larger treatises for a discussion of the details. Since these machines all depend on the principle of charging by induction, or, as it is sometimes called, influence, they are often called induction, or influence, machines.

384. Electric charges from dry cells. The ordinary electroscope is not sensitive enough to detect the charges on a single dry cell. For advanced work most laboratories have electroscopes, such as the quadrant electrometer, that are far more sensitive than the common type. In some laboratories one of these may be available for use on the demonstration table. If so, the instructor can show that the two terminals of a dry cell are charged. The central, or carbon, terminal has a positive charge, while the zinc terminal has a negative charge. In the case of a gravity cell the copper terminal is positively charged, and the zinc terminal negatively charged.

If a battery is made of seventy-five or one hundred dry cells connected in series (that is, the zinc terminal of one is connected to the carbon of the next, and so on), it is very easy to show with a good leaf electroscope the charges on the terminals of this battery. In performing this experiment the terminal to be tested is metallically connected to the leaves, while the other terminal of the battery should be connected to the case, as shown in a conventional manner in Fig. 234.*

+

385. Electric charges from a power circuit. If the electroscope E and a suitable lamp L are connected to a power circuit (such as the ordinary lighting

Battery

circuit) by the wires AB and CD, as shown in Fig. 235, the lamp will be lighted and the leaves of the electro

FIG. 234

scope will show a slight divergence.

This experiment suggests three noteworthy facts:

1. Although the wires are connected together by the lamp, they are kept charged by the power supply to which they are connected. Touching with the hand does not discharge them.

2. The divergence obtained is small compared with that obtained by using a hard-rubber rod or electrophorus. This is often a great surprise to one testing it for the first time. It will be referred to later, when the whole matter will be cleared up.

3. If the power circuit is a direct-current circuit, the wire connected to what is called the positive terminal is found to have the same kind of electricity as that on hard glass which has been rubbed with silk; and the one connected to the negative terminal has the same charge as that on hard rubber which has been rubbed

with wool. In other words, the terms "positive" and "negative," as applied to the terminals of cells and power circuits, are obtained from the kind of static charges these terminals carry.

386. The distribution of charges on conductors. A deep tin can is insulated by standing it on a block of paraffin. While there it can be electrified in a number of ways. A simple way is to introduce into the can a charged hard-rubber rod, care being taken not to touch the can with the rod. While the rod is there, the can is touched with the finger. When the hand is taken away and the rod removed, the can will be found to be charged. If the student cannot readily explain this process and the position of the charges during each step of the process, he should re-read section 378. After the can is charged, a small metal ball on a silk thread is lowered into it until it touches the bottom. On removing the ball and testing it with an electroscope, no charge will be found on it; but if the ball is touched

+

A

OL

B

E

C

D

FIG. 235

to the outside of the can, it will become charged. This experiment shows that the charge on the can stays on the outside. This is only an instance of the following general rule: The charge on a conductor tends to stay on the outside surface. When we recall that like charges repel each other, we can see why it is that each part of the charge, being repelled by other parts, gets as far from the center of the body as it can.

A sensitive electroscope when placed inside a metal box or cage will not be able to detect the presence of charges that are on the outside of the box. For example, an electroscope will not be affected in any way when inside a wire cage, even when sparks from an electrostatic machine are jumping over to the cage. This can be explained by the fact that the charges on the outside of a conductor do not exert any forces on the inside of that conductor. This is strictly true, however, only when the charges on the outside are at rest. While in motion, they may produce magnetic fields in the inside, as explained in section 405.

As a protection against lightning, part of the efficiency of metal

R

roofs and lightning rods that are well connected to moist ground is due to this tendency of charges to stay on the outside.

If an elongated insulated conductor is charged, and then different parts of the surface are tested with a small conductor mounted on a nonconducting handle, it will be found that there are larger charges on the ends of the conductor. In general, it will be found that those parts of a conducting surface that have the greatest curvature are charged the most highly.

387. The action of points. In the last section it was shown that

the charges on conductors have a tendency to accumulate on the more pointed parts. When the end of the conductor is a sharp point, the effect of a large charge is so great that the air in the neighborhood of the point becomes a

FIG. 236

conductor* and then becomes charged with the same kind of electricity as that on the pointed conductor. Fig. 236 shows diagrammatically the pointed conductor positively charged and the positively charged air near it. Since like kinds of electricity repel, the charged air is repelled from the point. In this way the point is very quickly discharged. If the point is kept charged, say by an electrostatic machine, there will be a stream of charged air driven off from the point. This stream of air is often strong enough to blow out a candle flame. Conductors brought into this stream of air become charged; for example, an electroscope fifteen feet away from a point attached to an electrostatic machine can be charged so rapidly that one can see the leaves move.

If an uncharged conductor B, with a sharp point on it, is brought near a charged conductor A, the conductor B will become charged with the same kind of electricity as that on A. The reason for this can be understood by the aid of Fig. 237. The point becomes charged by induction with negative electricity, which is conducted away by the stream of air which flows from the charged point toward A. This leaves a positive charge on B; the

+

+

+

+

A

+

B

+

+

+

+

+

FIG. 237

negative going to A neutralizes part of the positive charge on A. No matter whether the point is on the charged or uncharged conductor, the effect of the point is to make the air between the two a conductor, and the final result is the same as if the two conductors were brought into contact. The electric whirl (Fig. 238) depends for its action on the fact that when the electrified air near a point is repelled there must be a reaction on the conductor. The whirl consists of a light cross of wire, with ends sharpened and bent at right angles so that they all point in the same sense. When this is mounted on a pivot at its

center and kept charged, the repulsion between the charged air and the points causes it to rotate in a direction opposite to that in which the points are bent.

The fact that points can make the air near them conducting is utilized in many ways. In electrostatic machines rows of metallic points are used to conduct the electricity off from moving plates to stationary conductors. Part of the efficiency of lightning rods is due to the fact that the pointed tips can produce a silent discharge between the rods and low-flying clouds. Sometimes a heavily charged cloud will induce an

:

*

FIG. 238

opposite charge on buildings and other objects lying below it. Sharp points will tend to send this induced charge up into the air toward the cloud and thus to discharge the cloud. However, this action is slow, and hence it is not always effective.

388. The lightning rod. In section 386 it was pointed out that on account of the tendency of charges to stay on the outside of conductors a well-grounded system of lightning rods on a house is of some protection to the interior. In the last section the reason for putting points on the rods was explained. There is, however, considerable difference of opinion about the advantage of installing lightning rods. It is usually a difficult matter to get them well connected to the earth, to wet soil, and it is also difficult to keep them so connected. Electricity in motion does not obey the same laws as when at rest; for example, a charge in motion may prefer to go to the interior of a conductor. We now know that very sudden rushes of charges obey special laws, and that sometimes the