VIII. On the Laws according to which Masses of Iron

influence Magnetic Needles.

By S. H. CHRISTIE, M. A. F. C. P.S.

OF TRINITY COLLEGE, CAMBRIDGE, AND OF THE ROYAL MILITARY ACADEMY,

WOOLWICH.

[Read May 15, 1820.]

SEVERAL of the phænomena which may be observed on placing a magnetic needle in the neighbourhood of a mass of iron, have been explained on the hypothesis that the iron, through the magnetic influence of the earth, becomes itself a magnet, having its north pole in the upper, and south pole in the lower part of the mass. I have for some time considered that many objections might be urged against this hypothesis, indeed that it was not adequate to the explanation of all the phænomena, and having adopted a new one, undertook a series of experiments as a criterion of its correctness. As the effects produced by the action of masses of iron on compass needles are, in some cases, of serious importance, and as I besides consider, that this subject is connected with the interesting question of the change in the variation and dip of the needle, I trust that this paper will not be unacceptable to the Society to which I have the honor to present it.

It is well known, that if a mass of iron be placed at a small distance from either end of a magnetic needle, suspended freely, that end will be attracted; but if the distance of the iron be

increased beyond a few inches, the end of the needle to which the iron is the nearest will, in some cases, be attracted, and in others repelled. If a bar of iron be held nearly vertically, and the upper end be presented to the north end, that is the south pole, of a needle, the end of the needle will move towards the iron if the lower end of the iron be presented to the north end of the needle, the needle will move from the iron: and this is accounted for, on the hypothesis of the iron becoming magnetic through the influence of the earth, by saying that the poles of the iron and needle, which are brought together, are of opposite kinds in the first instance, and consequently attract each other, but of the same kind in the second, which causes repulsion. The same phænomena, but in a reverse order, are observed at the south end of the needle, and to these a similar explanation is given.

According to the hypothesis in question, the magnetism of the earth is communicated to a mass of iron, in the direction which a magnetised bar takes, when freely suspended by its centre of gravity, that is in the direction of the dipping needle; and the magnetic influence is concentrated near the extremities of a line passing through the iron, in that direction, in two points, which are called the poles of the iron. If then the position of the mass of iron be changed, its poles will, according to this hypothesis, be immediately transferred to two other points; and however suddenly the change of position may take place, the shifting of the poles must be equally sudden, to account at all for the phænomena. To suppose that two points possessed of a quality, which they exert energetically, should be instantaneously deprived of that quality, and that it should be immediately transferred to two other points, by a simple change of position in the body, appears to me rather unphilosophical. But it has been said, that this instantaneous change takes place in consequence of the softness of the iron which is employed: it has however been found that the

same phænomena, which may be observed on suddenly changing the position of a bar of soft iron, equally take place with bars of the hardest steel, provided care has been taken that the steel should not acquire any magnetism in the manufacture. As, however, my object in the present paper is not to controvert this hypothesis, but to advance one which offers a ready explanation of the phænomena, and affords peculiar facilities in computing the effects produced, I shall for the present take leave of it, reserving some other arguments for a future period.

It appears to me that there is no necessity for supposing that any part of the mass of iron possesses the power of repulsion, and that all the phænomena which take place, may be clearly explained on the simple hypothesis that the particles of iron possess the single power of attraction, and exert it principally, if not wholly, on that which appears to be the cause of the direction of the needle, and but slightly, if at all, on the poles of the needle itself. I was first led to think that the iron acted in this manner on the needle, on being informed by Mr. Barlow, who has lately published some interesting experiments on this subject, that he had observed anomalies in the action of a mass of iron on a magnetic needle, for which he could not account; that there appeared to be a plane, passing through the centre of a sphere of iron, in which if the centre of the needle were placed, no effect would be produced in the direction of the needle. He did not at this time inform me what experiments had led him to this conclusion, nor of the angle which the plane made with the horizon; but, on consideration, it appeared to me, that if the iron were supposed not to act on the needle in its horizontal position, but in the same manner as if the needle were inclined to the horizon at an angle equal to the natural dip, then such a plane would necessarily exist, making an angle with the horizon equal to the complement of the dip, since then the perpendicular on this needle, by which the action of the iron

would be estimated, would fall on its centre. Another consequence of my supposition would be, that if the perpendicular from the centre of the sphere of iron fell on the upper, that is, the south branch, of this imaginary needle, then that end of the horizontal needle would deviate towards the sphere; and that the north end of the needle would deviate towards the sphere, when the same perpendicular fell on the lower branch of the imaginary needle.

Having this idea of the manner in which the iron acted, I witnessed some of Mr. Barlow's experiments, mentioning beforehand what I conceived would result from particular positions of the sphere of iron. In these experiments, the several deviations were of the nature I had predicted, and, as far as we could then judge, their magnitude was such as might be expected on the supposition which I had made. I now therefore considered in what manner this mode of action of the iron could be accounted for, and I was soon forcibly struck with the idea that the needle was guided in its horizontal direction by magnetic particles, passing through its centre in the direction of the natural dip, and that the iron acted principally, if not wholly, on these particles, causing, by their deviation towards it, a corresponding deviation of the horizontal needle. The results of the several experiments, which I afterwards made, perfectly coincided with this view of the subject.

The apparatus which was made use of, consisted of a cast iron ball 12.78 inches in diameter, suspended over the centre of a table in the construction of which iron was carefully excluded. In the middle of the table a circular hole was cut, 13.25 inches in diameter, so that the ball could be let down below the plane of the table, or raised above it, by means of a system of pullies. After the table was rendered perfectly steady and horizontal, the magnetic meridian was accurately ascertained, and being drawn, the table

was

divided at every 10°, reckoning from the meridian, by lines drawn from the centre to the circumference. The compass being placed on one of the divisions, so that its north and south line coincided exactly with that division, and its centre was at the distance of 12 inches from the centre of the table, the ball was raised until it appeared to have no influence on the needle: it was then lowered, inch by inch, and the deviations at every inch carefully noted, until the ball had descended so far below the table as to cease influencing the needle. This was done with two compasses, at every 10° from the north to the west, and from the south to the

cast.

I first considered, that, if the ball acted in the manner I have mentioned, namely, on magnetic particles passing through the centre of the needle in the direction of the dip, then when the centre of the ball was in the point where a line drawn from the centre of the needle, perpendicular to the direction of the dip, cut the vertical passing through the centre of the table, the ball, attracting equally the particles guiding the north end of the needle, and those guiding the south end, no deviation should take place; that when the centre of the ball was above this point, a preponderance would be given to the action of the particles guiding the south end, and these being attracted towards the ball, the south end of the needle would move in that direction, or the north end would deviate from the ball; and that the contrary would take place when the ball was below this point, namely, that the deviation of the north end of the needle would be towards the ball. The annexed diagram will illustrate this. (Fig. 21.) NESW represents the plane of the table, the centre of which is O: NOS is the magnetic meridian, and EOW at right angles to it, passing through the east and west points; C is the centre of the magnetic needle, and s Cn, in a vertical plane, parallel to SN, the direction of the dipping needle, in which I suppose