Magnetometer

I above the magnet is the holder of the inertia bar, used in the special experiments for determining the moment of inertia of the magnet.

Measurement of Horizontal Force,

operations are involved. We first determine the time of vibration of the colli mator magnet. This enables us to calculate M H, where M is the magnetic moment of the magnet. We then use the same colli mator magnet to deflect an auxiliary magnet, and from the ob served deflection angle we can calculate M/H. Combining the two results, we obtain H and M as well. For the vibration observation the magnet is suspended as in fig. the shutters being attached. The time of vibration, deduced from several groups each of Ioo vibrations, is obtained from transits of the collimator magnet, taken to second by means of a chronometer, using the eye and ear method. The temperature is observed with the ther mometer C before and of ter taking the vibrations. The magnetic couple is aided during the vibrations by a small torsion couple from the suspension, which necessitates the determination of a torsion coefficient by observing the effect of a torsion of or on the pointing of the magnet.

In the deflection experiment the magnetometer is arranged as shown in fig. 2. The collimator magnet K on a carriage L, which slides on the deflection bar D, deflects a smaller auxiliary sus pended magnet. This carries a mirror, the normal to which is in the same vertical plane as the axis of the magnet. The auxiliary magnet is raised or lowered until its axis is level with that of K. Wooden panels which slide in vertical grooves are then inserted to protect it from draughts. When the unifilar has been turned until the deflection bar and the collimator K are approximately perpendicular to the magnetic axis of the auxiliary magnet, a reflection of the ivory scale B from the mirror becomes visible in the telescope A, suitably focussed. The position is further altered by a slow-motion screw until the centre of the scale coincides with the vertical wire in A, and the verniers are then read. The dif ference between this reading and that obtained when the magnet K is turned end for end is double the deflection angle 0 produced by K at that particular distance r. The value of M/H is obtained

from the equation ./1//H=1 sin 0 where P and Q are the two so-called "distribution constants." In reality deflections are taken with K at equal distances from the deflected magnet on the two arms of the bar. Also the arm opposite to that carrying the magnet carries a thermometer in a symmetri cal position to K, which is read immediately before or after read ing the verniers. With the thermometer is a counterpoise, the counterpoise and thermometer together equalling in weight the magnet and its carriage. The elimination of P and Q and the final calculation of M/H really involve the taking of deflections at three distances, e.g., 25, 3o and 4ocms. In practice P and Q are derived from the mean of a large number of observations. Theoretically it is possible to make either P or Q zero by suitably choosing the relative dimensions of the deflecting and deflected magnets. For example, assuming the two magnets exactly similar in every way, Q should vanish when the length of the deflected magnet is 0•467 times that of the deflecting magnet. If Q is zero, or even very small, and the deflection distances are large, deflec tion at two distances suffices. In an old magnet carefully handled M changes very slowly. When pressed for time the best course may be to omit the vibration experiment, and assume a value obtained for M by interpolation from previous and later com plete observations.

number of corrections are necessary, e.g., for temperature, temporary induction and even for the bending of the deflection bar. The practical use of the instrument postu lates the determination of various constants and the calculation of several tables. The most important "constant"—which may be expected to change in the process of time—is the moment of inertia of the collimator magnet.

Variations of

suspension, though supplying a satisfactorily small torsion, is affected by mixture and somewhat uncertain, and use is sometimes made of a fine metal suspension.