which the actual measured declination is the same.

Here the effects

of local disturbances have not been eliminated by lumping the stations together in groups so as to eliminate the effects of these disturbances, at any rate when they do not affect any large area.

From the amount of the differences between the values of the elements as obtained from the smooth curves and the actual measured values, an idea of the position and extent of the magnetic masses which cause these differences has been made by Professor Rücker, so that the value of the magnetic field on the surface of the earth is employed to get an idea of the geology of the portions of the earth's crust below the actual surface layers.

434. Continuous Magnetic Records. In a certain number of observatories a continuous record of the values of the magnetic elements is kept by means of self-recording instruments. The records are obtained by means of the trace left by a spot of light reflected from a mirror attached to a magnet on a sheet of photographic paper, which is kept in uniform movement by means of clockwork. In the case of the declination, the mirror is simply attached to a magnet which is suspended by a long fine thread, so that it can turn freely about a vertical axis, and so, by always setting itself in the magnetic meridian, gives a record of the changes that take place in the position of this meridian, that is, shows the changes in the declination.

The changes in the horizontal force are recorded by means of a magnet which is suspended by a bifilar suspension (§ 119). The top of the bifilar is turned till the magnet sets itself at right angles to the magnetic meridian, under which circumstances the earth's field exerts a turning couple on the magnet equal to MH (§ 425), this couple being balanced by the couple due to the bifilar. If the value of the horizontal force Halters, the couple due to the magnetic forces alters also in the same proportion, and the magnet turns about a vertical axis till the couple due to the bifilar becomes equal to the new couple due to the magnetic forces. Changes in the declination will, however, not affect the position of the magnet, since it is at right angles to the magnetic meridian.

Since no satisfactory way of recording the changes that take place in the dip has been devised, it is usual to record the changes in the vertical force. For this purpose a magnet is balanced on knife edges in such a way that it is in an approximately horizontal position. If, say, the vertical force decreases, then the downward force acting on the north pole and the upward force acting on the south pole both decrease, and hence the north pole of the balanced magnet rises and the south pole falls, just as when, in a balance, the load of one pan is increased and that of the other is decreased. The motions of such a balanced magnet will therefore indicate the changes that take place in the vertical force, and since the magnet with the bifilar suspension gives the changes that

DECLINATION

take place in the horizontal force, the changes in dip and in the total force can be immediately calculated from the records given by the two instruments.

435. Diurnal Range.-As a result of a study of the records of the magnetographs, as the self-recording magnetic instruments are called, it is at once evident that the values of the magnetic elements undergo small daily changes in value, the magnitude of this diurnal range depending on the position of the place and the time of year. The form +5'

AND DIP

-5'

A.M.

P.M.

FIG. 427.

of the diurnal range curves for Kew for the summer months are shown in Fig. 427, which gives the variation of each element from its mean value for the whole twenty-four hours. The fact that the curve is above the zero line means that the corresponding element is greater than its mean value.

436. Annual and Secular Change. In addition to the diurnal range, the magnetic elements undergo a periodic change of which the period is a year, which is called the annual range.

Slow changes of which, if they are periodic, the period must be many centuries, also take place in the values of the elements, and these are called secular changes. In Fig. 428 the change in the value of the declination at London during the last three hundred years is shown by means of a curve. It will be seen that the declination attained a maximum westerly value in 1810, while in 1660 the declination was zero, so that in that year the agonic line passed through London.

A very elegant method of showing the changes due to the secular variation has been introduced by L. A. Bauer. If we suppose a magnet

HORIZONTAL FORCE

suspended in such a way that it is free to set itself parallel to the lines of force of the earth's field, then, owing to secular change in the declination and in the dip, the north pole of the magnet would describe a curve in space. The form of the curve in the case of a magnet in London is shown 30

in Fig. 429. From this curve, and similar ones drawn for other places, Bauer was able to show that the north end of such a freely suspended needle describes a curve such that, to an observer situated at the centre of the needle, the curve is described in the same direction as that in which the hands of a watch move. The form of the curve given in Fig. 429

seems also to indicate that the curve described by the pole of the needle will be closed, the time taken for the needle to complete a whole cycle being about 470 years.

437. Magnetic Storms.-In addition to the regular changes in the magnetic elements which we have been considering, sudden disturbances of these elements sometimes occur, which are often, especially when the

phenomenon called the aurora borealis is seen, of considerable magnitude. The character of such magnetic storms, as they are called, is shown by the copy of the photographic trace of the self-recording declination magnetograph of Greenwich Observatory during a magnetic storm and

also during an ordinary quiet day, reproduced in Fig. 430. The cause of these magnetic storms has not yet been discovered, although there seems to be some connection between them and the condition of the sun, for whenever there are a large number of spots on the sun there always seem to be a number of magnetic disturbances.

PART II-ELECTRO-STATICS

CHAPTER III

ELECTRO-STATIC ATTRACTION AND REPULSION—
COULOMB'S LAW

438. Fundamental Experiment.-Thales, who lived about 600 B.C., discovered that amber when rubbed acquires the property of attracting light bodies, such as pieces of pith or cork. Towards the end of the sixteenth century Gilbert showed that this property was also possessed by other bodies, such as wax, sulphur, and glass. All such phenomena are studied in the science of electricity, the name being derived from the Greek name for amber.

A body which has acquired this property of attracting other bodies, the attraction considered being of course different from the gravitational attraction which all bodies exert one on the other, is said to be electrified, or to possess electrification. Electrification, unlike mass, is not a fundamental property of matter, since under ordinary circumstances matter is unelectrified, and it is only after the electrification has been produced by certain causes, which we shall examine in detail later on, that it becomes electrified.

The most usual manner of causing the electrification of a body is that referred to above, namely, friction with a suitable rubber. Thus a stick of sealing-wax, when rubbed with a dry piece of flannel, becomes electrified, as also does a rod of glass when rubbed with silk.

439. Conductors and Non-Conductors.-All substances may be roughly divided into two classes, called conductors and non-conductors. In a conductor the electrification spreads all over the body, so that if one point of the body is by any means electrified, this electrification immediately spreads all over the body. In the case of a non-conductor, or insulator, as such bodies are also called, the electrification does not spread in this way, but remains in the neighbourhood of the point where the electrification took place.

The best conductors are the metals and solutions of most salts in water, while the best non-conductors are ebonite, glass, shellac, sulphur, paraffin, sealing-wax, and silk. There is, however, no hard and fast line of demarcation between the two classes, for such bodies as dry wood