120 ELECTROMETER a strong vertical support fastened to the brass lid of the jar MM (fig. 9), and passes through a slit in the hollow cylinder. This vertical piece is fitted on one side with two V notches, into which the hollow cylinder is pressed by a spring fastened to the lid and bearing half way between the Vs, and on the other side with a rect angular groove in which slides the vertical part of a knee-piece D, in rigid connection with the hollow cylinder. D prevents the cylinder from turning round, but allows it to move vertically ; it also carries a fiducial mark running opposite a graduation on one edge of the groove, by means of which whole turns of the screw are read off, fractions being estimated by means of a drum head. The nut AC is arranged in two parts, with a spring between them, to prevent "lost time "and secure steadiness (for details, see paper cited above.) The disc G is connected by a spiral of line platinum wire with the main electrode S, which is insulated from the lid of the box by a glass stem. The arrange ment of this electrode is worthy of notice, and will be understood from fig. 11. The dome T is called the um brella ; its use is obvious. A similar, only less perfect, device was noticed in Ca- vallo s electroscope. The vital parts of the instrument are all inside the coated jar, and therefore removed from dis turbing influences ; only it is necessary to remove some of the tinfoil opposite the hair in order to see it. The effect of this is counteracted by means of a screen of fine wire. The use and the theory of the instrument are very simple. The body whose potential is to be measured is connected with the um brella, which is raised in Fig. 11. order to insulate the main electrode from the case, the last being supposed to be in connection with the earth. Let v be the poten tial of the inner coating of the jar, the disc, and guard plate, V that of the body and G, and d the distance between G and H when the hair is in the sighted position. Then, since F may be regarded as forming part of in infinite plate, 1 if its surface be S its potential energy will be S<r(y V) (see ELECTRICITY, p. 34), i.e., Hence the attraction /on F will be given by , S( ?;-)* (I)- 1 Here /is a constant, depending on the torsion of the suspending wire of the aluminium balance; hence, A 2 standing for Sa/^S, i.e., A being a constant depending on the construction of the instrument, we have v-V^Ad (2). If we now depress the umbrella, so as to bring G to the potential of the earth, and work the screw till the hair is again in the sighted portion, we have, d befiig the new reading of the screw, v = Ad (3). Hence, from (2) and (3), V = A(<t -d) (4). We thus get V in terms of A and the difference of two screw read ings, so that uncertainties of zero reading are eliminated. The value of A must be got by comparison with a standard instrument, if absolute determinations be required. Absolute Thomson s absolute electrometer (fig. 12) is an adaptation of the electro- attracted disc principle for absolute determinations Ve give merely meter. an indication of its different parts, referring to Thomson s paper (I.e.) for details. B is an attracting disc, which can be moved parallel to itself by a screw of known step (J, in. or thereby). A is a guard plate, in the centre of which is a circular balance-disc of aluminium suspended on three springs, and connected by a spiral of light platinum wire with A. The disc can be raised or depressed into definite positions by means of a screw (the kinematical arrange ments in connection with these screws are similar to that in The portable electrometer). A hair on the disc, an object lens h, a fiducial mark, and an eye lens I enable the observer to tell when 1 Those who desire to know the degree of approximation here should consult Maxwell. Electricity and Alaynetism, vol. i. 217. this disc is in such a position that its lower surface is plane with lower surface of A. y y are the hafeves of a box which screens tho disc from electric disturbances. An idiostatic gu.tge (consisting of an aluminium lever with guard plate, hair, and lens, as in the portable electrometer), placed over a plate F in connection with the guard plate, enables the observer to tell when the guard plate and the inside coating of the instrument (which forms a Leyden jar as in the portable instrument) are at a certain definite potential. And finally, a small instrument called the " replenisher enables Fio. 12. Thomson s Absolute Electrometer. him to raise or lower the potential of A till this definite potential is reached. A short description of the replenisher will be in place here. It Repl. is represented pretty clearly at E (fig, 12). Two metal shields, isher in the form of cylindrical segments, are insulated from each other by a piece of ebonite ; the left hand one is in connection with the guard plate, the right hand one with the case of the instrument (and therefore with the outer coating of the jar). A vertical shaft, which can be spun round by means of a milled head, carries two metal flies on the ends of a horizontal arm of vulcanite. Two small platinum springs (the front one is seen at e) are arranged so as to touch the flies simultaneously in a certain position just clear of the shields. Let us suppose the left, shield along with A to be posi tively electrified, and the flies to be in contact with the springs : e. being close to the left shield, the front fly will be electrified - and the back fly 4-. Suppose the shaft to revolve against the hands of a watch lying face up on the cover of the electrometer. The front fly carries oft its- charge, and, when near the middle of the right shield, comes in contact with a spring connected with the shield. Being thus practically inside a hollow conductor, it gives up its - charge to the shield. At the same time the back fly gives up its f charge to the left shield. The result of one revolution therefore is to increase the + and- charges on the respective shields, or, in other words, to increase the difference of potential between them B} giving the machine a sufficient number of turns, the potential of A may be raised as much as we please ; and, by spinning in the opposite direction, the potential can be lowered ; so that, once A is charged, it is easy to adjust its potential till the hair of the gaMS is in the sighted position.
To work the instrument, the electrode n of the lower plate B is