The instrument with which Dr Robison's experiments were made, is represented in Figs. 5. and 6. Fig. 6. being a perspective view of the whole instrument. A polished black ball A, s of an inch in diameter, is fixed on the point of a very slender needle, three inches in length, the other or the needle passes through a ball of amber or glass F. about three-fourths of an inch in diameter ; but though the ball is completely perforat ed, the end of the needle must not extend quit( to the surface. At light angles to the needle, a slender glass rod is fixed into the ball F ; it is bent at right angles at E, and is continued to L, exactly ano% e the centre of the ball A. The length of this rod is six inches from I' to L. At the extremity L is fixed a piece of amber C, with an opening in the middle to admit the rod DCB. The rod DB is formed of a silk thread coated with seal mg-wax ; the silk thread must be very strong and very dry, and, when completely penetrated by the melted wax, it must be made straight and smooth, so as to be sufficiently stiff and free of all asperities. This rod passes through a small cube of amber, which has fine holes drilled in two of its opposite sides; and by means of two round pins passing through the piece of amber C, the rod DB has a free motion round these pins as an axis. The branch CB of this rod is about three inches long, and carries at its extremity a ball B a quarter of an inch in diameter, which may be either made of metal, or of cork gilt and burnished. The other branch CD, which is of the same length, passes through a small cork ball. When the instrument is thus constructed, the lower ball 13 must just touch the ball A, w hen the arm FE is in a vertical position. The ball F, Fig. 5. must now be fixed at the extremity of a glass rod Fl, which passes at right angles through the centre of a divided circle GI-10, and has a handle of boxwood at its other extremity I. This rod, which is perpendicular to AF, is supported by the bead of a pillar HK, in which it turns with some diffi culty. An index Nil, fastened to the rod F1, is always set parallel to the line LA, drawn through the centre of the fixed ball. This index will point out the angle which LA forms with the vertical line. It will also be COIIVC 111CIlt to have another index, which turns stiffly on FI as an axis, and stretches a considerable way beyond the graduated circle. The graduated circle is divided into 360 degrees; the zero of the division is placed upper most, and 90 on the right hand.
In order to use this instrument, take hold of the han dle I, and set the index to 90. The lines LA and CB will now have a horizontal position, and the ball B will rest on A. Let the balls be now electrified, and let the index be brought back to 0 on the scale, by the handle I ; during this motion the balls will be seen to separate in some particular position of the index; repeat this ope ration till the exact position is ascertained at which the separation takes place. This will show the repulsive force when in contact, or at a distance equal to the sum of the radii. Let the instrument be now turned still more towards the vertical position, and the balls will be observed to separate more and more. An assistant must turn the long index, so as to make it parallel, by making the one appear to coincide with the other. In order to shew that this instrument will give an absolute measure of the repulsive force of the two balls, we have only to balance the rod BD in a horizontal position by loading the ball D with some greater weight ; then, by comput ing for the proportional lengths of BC and DC, we shall find the exact number of grains with which the balls must repel each other, in order merely to separate, when the rod BD has a horizontal position. When this rod has any oblique position, a very simple computation will give us the number of grains of repulsion, with which they will then separate; and a third computation, founded on the resolution of forces, will give us the repulsive force in grains, when AL has an oblique position, and when the rod BD forms any angle.
In order to make this instrument measure the attrac tions of two balls, one of which is positively and the other negatively electrified, we must incline the instru ment to the left, and by electrifying the balls A, 13, in an opposite manner, when at a great distance from each other, they will approach by their mutual attraction; and by means of the deviation of BC from a vertical line towards A, we shall be able to determine the magnitude of the attractive force. In making experiments with this instrument, the greatest care should be taken to provide as much as can be done against any dissipation or waste of the electric matter, by having all angles and asperities removed, by varnishing all its parts, and by making the experiments in a harm room, and when the atmosphere is particularly dry. Dr Robison recommends, that the dissipation per minute, under various circum stances, should be determined by a previous set of ex periments.
By means of different instruments of this kind, of dif ferent sizes, and some of them of balls of an inch in dia meter, and radii of 18 inches, Dr Robison performed more than an hundred experiments, from which he con cluded that the mutual repulsion of two spheres, electri fied in the same manner, varied as d, a' representing the distance of their centres.
Several years after these experiments of Dr Robison's were made, and read before a public society, the atten tion of Coulomb was drawn to the same topic, by the prize on the subject of the magnetic needle, which he had the good fortune to share with Van Swindell. His first object was to contrive an apparatus for measuring very small quantities of electricity. This apparatus, which he employed in all his experiments, and which he calls a torsion balance, is represented in Plate CCXLIV. Fig. 7. where ABCD is a glass cylinder, 12 inches high and 12 inches broad, on which is placed a plate of glass AC, 13 inches in diameter. This plate, which covers the whole of the cylinder, is pierced with two holes f,tn, both of which are 20 lines in diameter. Into the hole f is cemented a tube of glass f h, two feet high, and at the upper end of the tube at h is placed a torsion microme ter, which is seen on an enlarged scale in Fig. 8. The upper part of this micrometer consists of a button nz, an index i a, and a pair of pincers g. This upper part is let down into the hole G, in the centre of the circle a, b, which is divided upon its edge into 360 degrees. This plate is soldered upon a tube of copper X, which enters into the tube H, Fig. 9.; and this tube is cemented into the upper extremity of the long glass tube f h. The pincers 9 have nearly the shape of the extremity of a so lid port-crayon, which can be opened and shut by means of a ring r. These pincers lay hold of the extremity of a very slender silver wire, the other end of which is held at P, Fig. 10. by the pincers of an iron or copper cylinder P o, about a line in diameter. These pincers, like the former, are opened and shut by a ring Q. This cylinder is perforated at C, to receive the needle a g. The weight of this part of the apparatus must he suffi ciently great to stretch the silver wire without breaking it. The needle a g, which must be suspended horizon tally about six inches from the bottom of the glass cy linder ABCD, is formed of a silk thread, or a straw covered with sealing-wax, and has the part q a, about 18 lines in length, formed of a cylinder of gum lac. At this extremity is a small ball a of the pith of elder, about two or three lines in diameter; and at the other extremity g, is a small vertical plane of paper, covered with turpentine, which serves as a counterpoise to the ball a.