212 



mugnetk inclinutiun, and its changes, however refined the 

 construction of the instrument. The sources of error seem, 

 in fact, to be inherent in every direct process of determining 

 the third element ; and it is only by an indirect method that 

 we can hope to evade them.* Of this character is the method 

 now proposed. 



If a soft iron bar, perfectly devoid of magnetic polarity, 

 be held in a vertical position, it immediately becomes a tem- 

 porary magnet under the inducing action of the earth's mag- 

 netic force, the lower extremity becoming a north pole, and 

 the upper a south pole. Accordingly, if a freely-suspended 

 horizontal magnet, whose dimensions are small in comparison 

 with those of the bar, be situated near, in a plane passing 

 through one of these poles, it will be deflected from the 

 magnetic meridian. The deflecting force is the induced 

 force of the bar, which may be regarded as proportional to 

 the energy of the inducing cause, i. e. to the vertical compo- 

 nent of the earth's force ; while the counteracting force is 

 the horizontal component of the same force, acting directly 

 on the magnet itself, to bring it back to the magnetic meri- 

 dian. Thus the magnet will take up a position of equili- 

 brium, under the action of these opposing forces ; and this 

 position will serve to determine the ratio which subsists be- 

 tween them. When the right line connecting the centre of 

 the horizontal magnet, and the acting pole of the bar, is 

 perpendicular to the magnetic meridian, the tangent of the 

 angle of deflection will measure the ratio of the two forces, 

 and will therefore be proportional to the tangent oj" the mag- 

 netic inclination. Accordingly, by observing the changes of 



* Two such indirect methods of determining the inclination have been pro- 

 posed in Germany, one by Professors Gauss and "Weber, the other by Dr. Sar- 

 torius von Walterhausen. That now suggested beai-s a close analogy, in principle, 

 to the former of these : it differs from it, however, not only in the means em- 

 ployed, but also in the end in view, — the main object of the present method 

 being the determination of the inclination -changes. 



