142 A^ TAXAKADATE. 



may naturally be expected to cause anomalies in the distribution 

 of magnetic force, as was already remarked by many. 



We may suppose with v. Bezold, Leyst and others, the 

 terrestrial magnetic force observed at a place to consist of the 

 average effect of all the magnetised parts superposed with ano- 

 malies of comparatively large extent, which again can be sub- 

 divided into mean anomalies of less extent superposed with 

 smaller. In this respect the differences above spoken of may 

 be called variatiowd anomalies, understanding thereby nothing 

 more than the results of numerical operations on the observed 

 data conducted as above. 



Disturbance due to a Simple Source. 



In order to see roughly what sort of disturbances in the 

 vertical variation of magnetic force is lilcely to be met with, 

 take as the disturbing source a simple positive pole of strength 

 ?/?, placed inside a sphere representing the earth. 



In Fig. 5, let C be its center, M the position of the source, 



P any point on the surface ; and put 



Fig. 5. 

 CM=r 



MPC = c=the zenith distance of the 

 direction of the force at P ; 

 and CP = ?,= the variable radius vector 

 through P (positive outward) which is to be made equal to the 

 mean radius R, after performing differentiation. As no restriction 

 is laid upon the value of r, the result can also be applied to 

 the case when the source is above by making r greater than R, 

 and paying due regard to the signs of the trigonometrical functions. 



