GAUSS AND WEBER OX TERRESTRIAI^ MAGNETISM. 35 



and in a prescribed position, deflecting the suspended bar from 

 the magnetic meridian. 



11. The torsion-bar and deflecting bar. — That the thread to 

 which the magnet bar is suspended is without torsion in the 

 mean position of the latter, is recognised thus : a brass bar of 

 equal length and breadth, and of nearly equal weight, as the sus- 

 pended magnet bar, having a small magnet inserted in it (iu 

 order somewhat to shorten the duration of the vibration due to 

 the elasticity of the thread) is placed in the stirrup instead 

 of the magnet bar. If the thread is without torsion, the 

 magnetic axis of the small magnet will be in the same line 

 as that of the larger bar was. In order to test this accu- 

 rately, the auxiliary bar must, like the principal bar, be pro- 

 vided with a mirror and a min-or-holder. For measurements of 

 intensity a second magnet bar of like dimensions to the principal 

 bar is required, which may also be placed in the stirrup instead 

 of the latter, in order to observe its vibrations, and to measure 

 its moment of inertia. The same bar, however, must also serve 

 as a deflecting bar, and for this purpose it is fitted into a small 

 wooden case, which is bounded exteriorly by even surfaces and 

 straight edges parallel to its magnetic axis, in order to give it its 

 place quickly and accurately on the measm-ing scales. 



12. The weights and weight-holders. — For measurements of in- 

 tensity it is requisite that the deflecting bar may also be \'ibrated, 

 and its moment of inertia thus deduced. For this j)urpose a 

 thin wooden rod is placed across the vibrating magnet bar, and 

 two equal weights are suspended, at various distances from each 

 other, successively on both sides of the magnet bar. In order to 

 mark the points of suspension-, and to determine accurately 

 their mutual distances, both weights, each of which amounts to 

 500 grammes, are provided with a small capsule. The cap- 

 sule is placed on a fine point, projecting from the wooden rod. 

 There must be several such projecting points at 50 miUime- 

 tres distance from one another, with the exception of the two 

 central ones, which are situated at 100 miUimetres fi-om each 

 other. These distances must be measured with microscopical 

 accuracy. 



13. The quieting bar. — In order to perform the observations 

 promptly and accurately, it is of importance to be able to mode^ 

 rate at pleasure the vibrations of the magnet bar ; for instance, 



c 2 



