1884.] 



On a New Reflecting Galvanometer, tyc. 



293 



magnet be placed nearer to say the lower ends than the upper ends 

 of the needles, and the polarity of the end turned towards the needles 

 be of the same name as that of the nearer ends of the needles, they 

 will have a position of stable equilibrium when no current is flowing, 

 with a horizontal line joining a pole of each needle at right angles to 

 the direction of the magnet. The accurate law of variation of deflec- 

 tion with current is, however, in this case more complicated, and the 

 instrument in some cases might have to be graduated by experiments 

 with known currents of different amounts. Any change also of the 

 magnetic distribution of the controlling magnet would affect the 

 indications of the instrument. 



We propose in some cases to hang the needle system in a uniform 

 field, and to use a small needle rigidly connected with it, but 

 placed so as not to be perceptibly affected by the coils, to give direc- 

 tive force to the magnetic system. This arrangement, of course, 

 would not be astatic, but would give great sensibility on account of 

 the leverage of the horseshoe needles as arranged. 



Thus if m denote the magnetic moment of the small needle, H the 

 horizontal component of the earth's magnetic force, h a constant 

 depending on the coils, the strength of pole of each of the horse- 

 shoes (supposed of equal strength), and d the distance of these poles 

 from the suspension thread, we have, since the deflection is small, for 

 the turning couple exerted by the coils 4C&0c£, and for the return 

 couple mH0, and therefore — 



q m6 



4Jc<pd 



Of course this arrangement is applicable whether like or unlike 

 poles are turned in similar directions. It has the disadvantage that 

 any change of m or or of both would affect the constant of the 

 instrument.* 



The sensibility of any of these arrangements might also be increased 

 by bringing out a very light arm, say from the middle of the cross 

 bar connecting the horseshoes, or from any other convenient point, 

 and hanging the mirror by means of a bifilar, one thread of which 



* Note added March 20, 1884. — We propose to mount this small needle in such 

 a way that it can be turned round a horizontal axis at right angles to its length, and 

 also round a vertical axis, so that by means of this needle alone both the sensibility 

 and the zero of the instrument can be adjusted. When the galvanometer is not 

 intended for ballistic experiments, the frame on which the small needle is mounted 

 may conveniently be immersed in a liquid and made to act as a vane for bringing 

 the needle system quickly to rest. 



It is to be observed that, in consequence of the horseshoe needles being placed in 

 these instruments at a considerable distance from the axis of suspension, a verj 

 small value of I— I' is sufficient to give the needle system such a directive force 

 as to prevent any great error due to the rigidity or the viscosity of the suspending 

 fibre. 



