DENSITY OF THE EARTH. 159 



with their times of rotation, shews to exist also in those two 

 large planets. Thus the knowledge of the external form of 

 planetary bodies affords a basis for conclusions respecting 

 their internal constitution. 



The northern and southern terrestrial hemispheres appear to 

 present nearly the same curvature under equal latitudes ( 135 ) j 

 but, as has been already remarked, pendulum experiments 

 and measurements of degrees give such different results for 

 different parts of the Earth's surface, that it is impossible 

 to assign any regular figure which shall satisfy all the results 

 obtained by these methods. The actual figure of the Earth 

 is probably to a regular figure, " as the uneven surface of 

 agitated water is to the even surface of water in repose." 



Having thus measured the Earth, it had to be weighed ; 

 and here also the vibrations of the pendulum, and the em- 

 ployment of the plumb line, have served to determine its 

 mean density. This has been attempted in three ways : 1st, 

 by combining astronomical and geodesical operations for 

 the purpose of ascertaining the deflection of the plumb line 

 caused by the vicinity of a mountain; 2d, by comparing the 

 length of the pendulum vibrating seconds in a plain, and 

 on the summit of a mountain ; 3d, by employing a balance 

 of torsion, which may be regarded as a horizontal pendulum, 

 as a measure of the relative density of neighbouring strata. 

 Of these three methods ( 136 ) the last is the most certain, 

 because it is independent of the very difficult determina- 

 tion of the density of the mineral masses of which the 

 mountain consists, near which the observations of the 

 other two methods must be made. The most recent 



experiments with the balance of torsion are those of 

 VOL. i. N 



