THE NATURE AND INTERPRETATION OE GEODETIC EVIDENCE. 15 



tion is a convenient one. This centre of compensation must be 

 clearly understood as something entirely different from the centre 

 of gravity of the defect of mass by which compensation is pro- 

 duced, the two are not coincident in position, and the divergence, 

 which will not be great in the case of distant topography, may 

 or may not become important in the vicinity of the station, 

 according as the distribution of the defect of density is concen- 

 trated in a layer of small, or distributed through one of great, 

 thickness. 



The calculation of the depth of the centre of compensation does 

 not, therefore, give ;.ny direct information regarding the nature 

 of the compensation, but an investigation of the effect of varvino- 

 the assumed depth of the centre of compensation affords a ready 

 means of seeing in what direction we may best look for an explana- 

 tion of the departure of the observed from the calculated deflec- 

 tion of the plumb-line. 



The general principle of this investigation can easily be deter- 

 mined. In Fig. 1 let A represent the centre of attraction of an ele- 





r m 



^ 



c 



; 



1 



i 



1 







i / 





1 







\ / 





1 







i / 





; 











i 







\ / 





i 











• 







i/ 





i 







r i 





i 



C- 







v 



t 





1 





f 











■ 







1 





i 







I 





% 







\ 



1 





i 







\ 





\ 







» 



< 





% 







1 





t 



* 







1 





Pig. 1. 



vated mass, whose compensation is distributed in an unknown 

 manner, so that the centre of compensation lies at the point C, 

 then, if the divergence of the line AC from the vertical is 



[ 103 ] 



C2 



