216.]

MAGNETIC INDUCTION.

447

FIG. 162.

n

S

been calculated. The polar character of magnetic induction may be seen by suspending two pieces of soft iron wire over one of the poles of a magnet, s (fig. 162); the lower ends of the wires n, n, repel each other, but are both drawn towards the magnet, and the upper extremities, s, s, also repel each other. It is this mutual repulsion of the corresponding ends of the pieces of iron which causes the iron filings (fig. 160) to distribute themselves in curves around the magnet; for in this experiment each particle of iron becomes for the time a magnet with opposite poles. It is likewise in consequence of this polarity that a number of pieces of fine iron wire under induction form a continuous chain. A bar of soft iron placed on a magnet of equal dimensions neutralizes its action for the time: by connecting the two extremities of the magnet, it diverts the induction from surrounding bodies, and concentrates it upon itself. On the other hand the induc tion is much strengthened if the magnetic circle be completed, as in fig. 163, by uniting the pieces of iron suspended from either pole by the connecting piece, a b. This induction is maintained across the greater number of bodies, such as atmospheric air, glass, wood, and the metals. It is, however, modified by the interposition of iron, cobalt, and nickel, which are themselves powerfully susceptible of magnetism.

N

FIG, 163.

S

n

n

S

n

b

Magnetic induction differs essentially from electric induction (228) in this particular--viz., that it is not possible to insulate either kind of magnetism from the other. For instance, if one end of the two united pieces of iron, s n, s n (fig. 163), exhibit the properties of a north magnetic pole, the other end will exhibit those of a south magnetic pole; but if the two pieces of iron, whilst still under the influence of induction, be separated from each other, and then the magnet be withdrawn, both pieces of iron will have lost their magnetism. Again, if a magnet be broken in the middle, it will not be separated into one piece with a north and another with a south pole; each fragment will still possess two poles, turned in the same direction as those of the original bar (fig. 164); and each fragment may again be subdivided into an indefinite number of smaller

448

DISTRIBUTION OF MAGNETISM.

[216.

fragments, each of which will still possess a north and a south pole.

These phenomena may be explained by supposing that a magnet consists of a collection of particles, each of which is magnetic and endued with both kinds of magnetism. In the unmagnetized condition of the bar, these two kinds of magnetism are mutually combined, and exactly neutralize each other; but when the mass becomes magnetized, the two kinds of magnetism are separated from each other, though without quitting the particle with which they were originally associated. The two halves of each particle assume an opposite magnetic condition. All the north poles are disposed in one direction; whilst all the south poles are disposed in the opposite direction. Each particle thus acquires a polar condition, and adds its induction to that of all the others: as a necessary consequence of such an arrangement, the opposite kinds of magnetism become accumulated at the opposite extremities of the bar. If in fig. 165 the small circles be taken to represent the ultimate magnetic particles, the portions in shadow would indicate the distribution of south magnetism, while the unshaded half of the particles would show the distribution of magnetism of the opposite kind. This hypothesis is supported by the fact that a magnet whilst producing induction loses none of its strength, but on the contrary suffers temporary increase of strength, owing to the reaction of the induced magnetism of the soft iron upon it.

SOC

FIG. 165.

N

(217) Preparation of Magnets.-Pure soft iron loses its magnetism as soon as it is withdrawn from the inductive influence; but the presence of certain foreign bodies in combination with the iron, particularly of oxygen, as in the natural loadstone, and of carbon, as in steel, enables the body to retain the magnetization permanently. Hardened steel is always the material employed in the preparation of permanent magnets; it is not susceptible of so intense a degree of magnetization as soft iron, but when induction has once been produced within it, the effect is retained for an indefinite length of time. The development of this power in steel is much facilitated by friction; and the amount of force developed by this means is greatly dependent upon the direction in which the friction is performed. A simple method of magnetizing a bar consists in placing the bar on its side and bringing down upon one of its extremities either of the ends of a bar magnet. If the north end be brought down on the steel bar, it must be drawn

217.]

PREPARATION OF MAGNETS-MAGNETIC BATTERIES.

449

FIG. 166.

n

slowly along towards that end of the bar which it is intended shall possess south polarity: this operation must be repeated three or four times in the same direction. A more effectual plan is to bring down upon the centre of the bar the two ends of a powerful horseshoe magnet as represented in fig. 166; the south pole being directed towards the end of the bar that is intended to possess the northern polarity, and vice versa. It is then moved along the surface from the middle, alternately towards either end, taking care not to carry the horseshoe beyond the ends of the bar, and to withdraw the horseshoe from the bar when at its centre, c. The bar is then turned over and the process repeated on the opposite side, but in the same direction, for an equal number of times. When two bars are to be magnetized, they may be placed parallel to each other, the ends being connected by pieces of soft iron. Both the poles of the horseshoe are brought down upon the centre of one of the steel bars, and it is carried round the parallelogram always in the same direction, taking care, as before, to withdraw it when over the centre of one of the bars. In the last arrangement, the induction of one bar acts upon and increases the intensity of the magnetism excited in the other. For this reason, the opposite poles of magnets, when not in use, should be connected by pieces of soft iron, so that the continued induction shall maintain the magnetism of each.

In the act of magnetization, the horseshoe loses nothing of its power, but the north and south magnetism, which are supposed to exist in every particle of steel and iron, and which in the unmagnetized condition are so combined as exactly to neutralize each other, appear from the effect of the induction to which they have been subjected, to be permanently disturbed in their equilibrium in the newly-magnetized bars. The more intense the power of the horseshoe, the greater is this disturbance, and the more powerful are the magnets which are produced.

By uniting together several bar magnets, taking care that the corresponding poles of each are in the same direction, magnetic batteries of great power may be obtained. The magnets should be all as nearly as possible of the same strength; because if one of the bars be weaker than the others, it materially diminishes the power of the whole, and acts in the same manner as a bar of soft iron would do, although to a more limited extent. M. Jamin has recently constructed magnets of great power by uniting into one

450

HORSESHOE MAGNETS-EFFECTS OF HEAT.

[217.

horseshoe a considerable number of thin magnetized steel strips. As a matter of convenience, the bar magnet is often bent into the form of a horseshoe, so that the induction and attraction of both poles may be simultaneously exerted on the same piece of iron: the effect is in this manner much increased, and the weight sustained by the two poles united is much greater than the sum of the two weights which would be supported by each pole separately. For this reason, the soft iron armatures N, s, of a loadstone (fig. 161) add greatly to its strength, and by facilitating the application of the keeper, or piece of soft iron which connects the two poles when not in use, prevent the loss of the magnetic power.

(218) Influence of Molecular Motions on Magnetism.—It has been mentioned that the friction of a steel bar, whilst under induction, facilitates its magnetization. The same effect is occasioned by percussion of the bar, or by any other mode of producing vibration in it whilst it is under magnetic induction. On the other hand, if a bar has been fully magnetized, its strength is reduced by the application of a sudden blow; even the simple act of scratching the surface with sand-paper, or with a file, may seriously impair the strength of a good magnet.

The influence of heat on magnetism is remarkable. If a steel bar be ignited and placed under induction, and whilst still in this condition it be suddenly quenched, it will be found to be powerfully magnetic. Again, if a steel magnet be ignited, and allowed to cool slowly; all its acquired magnetism will have disappeared. Elevation of temperature, therefore, evidently favours the transfer of magnetic polarity within its particles. Further, if the temperature of a piece of iron be raised to redness, about 980° (527° C.), it will become indifferent to the presence of a magnetic needle, although on again cooling it will be as active as before. A similar effect is produced upon cobalt at the temperature of melting copper.* Nickel at a much lower temperature loses its action upon the magnet, as at 600° (316° C.) it exerts scarcely any attraction on the needle. So great is the influence of temperature upon a magnetic bar, that at the boiling-point of water, the diminution of its strength is perceptible by the rudest tests. If the temperature do not exceed the boiling-point of water, the magnet regains its strength on cooling. On the other hand,

* Faraday has, however, shown that in the case of cobalt its magnetic power increases as the temperature rises until it reaches about 300° (149° C.), beyond which it slowly diminishes, and at length becomes nearly evanescent. (Phil. Trans. 1856, 179.)

220.]

MEASUREMENT OF STRENGTH OF MAGNETS-DIP.

451

by cooling a magnet artificially, its strength is for the time exalted.

(219) Measurement of Magnetic Strength of a Bar.-The simplest method of ascertaining the strength of a magnet, consists in attaching to its armature a scale-pan, and ascertaining the weight which it will support; but it is obvious that this plan is not susceptible of any high degree of accuracy; it is, moreover, in many cases, quite inapplicable. A still easier, and more generally useful, because far more accurate, method, consists in suspending the magnet horizontally at its centre, by means of a few fibres of silk, and allowing it to take a fixed direction under the influence of a standard bar magnet, sufficiently long to be considered as acting by a single pole only upon the magnet under experiment: it is then displaced from its position of equilibrium, and the number of oscillations which it describes in a given time is counted. The relative magnitude of the strength of two or more equal bars, which may thus be compared, is proportionate to the square of the number of vibrations performed in equal intervals of time, when the two bars are placed at the same distance from the standard magnet. For estimating low degrees of power, the torsion of a glass thread, as employed in Coulomb's electrometer (226), may be used.

FIG. 167.

(220) Magnetism of the Earth-The Dip.-The remarkable fact of the pointing of the needle towards the north pole of the earth has been explained upon the hypothesis that the globe of the earth itself is a magnet, the poles of which are situated nearly in the line of the axis of rotation; the magnetism of the earth's north pole being of the same kind as that of the unmarked end of the magnet. If a small magnetized needle, s n, be freely suspended horizontally by a thread over the equator of a sphere (fig. 167) 9 or 10 inches (about 25 centimetres) in diameter, in the axis of

which a small magnetic bar, N s, at right angles to the axis of suspension of the globe, is placed, the needle will, when the magnetic bar is horizontal, as in No. 1, assume a direction parallel to the magnetic bar, and will point towards N and s, preserving its horizontal position; for it is equally attracted by the north and south polarities of the bar; but if one of the ends of the magnetic bar be made gradually to approach the needle as at 2,