1891.] on Electric and Magnetic Screening. 353 



that a broad enougli and long enough iron plate exercises a large 

 magneto-static screening influence ; which with a thick enough plate, 

 will be so nearly complete that comparatively little is left for aug- 

 mentation of the screening influence by alternations of greater and 

 greater frequency. 



A copper shell closed around an alternating magnet produces a 

 screening effect which on the principle stated above we may reckon 

 to be little short of perfection if the thickness be 2j cm. or more, 

 and the frequency of alternation 80 per second. 



Suppose now the alternation of the magnetic force to be produced 

 by the rotation of a magnet M about any axis. First, to find the 

 effect of the rotation, imagine the magnet to be represented by ideal 

 magnetic matter. Let (after the manner of Gauss in his treatment 

 of the secular perturbations of the solar system) the ideal magnetic 

 matter be uniformly distributed over the circles described by its 

 different points. For brevity call I the ideal magnet symmetrical 

 round the axis, which is thus constituted. The magnetic force 

 throughout the space around the rotating magnet will be the same 

 as that due to I, compounded with an alternating force of which the 

 component at any point in the direction of any fixed line varies from 

 zero in the two opposite directions in each period of the rotation. 

 If the copper shell is thick enough, and the angular velocity of the 

 rotation great enough, the alternating component is almost annulled 

 for external space, and only the steady force due to I is allowed to 

 act in the space outside the copper shell. 



Consider now, in the space outside the copper shell, a point. P 

 rotating with the magnet M. It will experience a force simply equal 

 to that due to M when there is no rotation, and, when M and P 

 rotate together, P will experience a force gradually altering as the 

 speed of rotation increases, until, when the speed becomes sufficiently 

 great, it becomes sensibly the same as the force due to the symme- 

 trical magnet I. Now superimpose upon the whole system of the 

 magnet, and the point P, and the copper shell, a rotation equal and 

 opposite to that of M and P. The statement just made with reference 

 to the magnetic force at P remains unaltered, and we have now a fixed 

 magnet M and a point P at rest, with reference to it, while the copper 

 shell rotates round the axis around which we first supposed M to 

 rotate. 



A little piece of apparatus, constructed to illustrate the result 

 experimentally, was submitted to the Royal Institution and shown in 

 action. The copper shell is a cylindric drum, 1 • 25 cm. thick, closed 

 at its two ends with circular discs 1 cm. thick. The magnet is sup- 

 ported on the inner end of a stiff wire passing through the centre of 

 a perforated fixed shaft which passes through a hole in one end of 

 the drum, and serves as one of the bearings ; the other bearing is a 

 rotating pivot fixed to the outside of the other end of the drum. The 

 accompanying sections, drawn to a scale of three-fourths full size, 

 explain the arrangement sufficiently. A magnetic needle outside, 



