18 ASTRONOMY. 



lien, in Scotland, give the mean density equal to 4.71, 

 that is about 4-| times the weight of an equal bulk of 

 water. But as the mean density of the rocks at the 

 surface does not exceed 2.5 or 2.7, it follows that the 

 density of the central parts must be much higher. 



If, for instance, the exterior rocks form a shell 

 500 miles in thickness, the parts within this shell 

 would require to have an average specific gravity of 

 5.8 to produce the mean density of 4.7 for the whole 

 mass. 



Laplace thinks that the change is not sudden, but 

 progressive, and that it is probably the effect of con- 

 centric and elliptical beds of increasing density. A 

 conjecture may be ventured that metals occupy the 

 centre of the earth, and that those found in the outer 

 Crust consist of minute portidns cast up from the 

 central mass. 



It is probable that the structure of the external 

 portion of our planet, exposed to observation by va- 

 rious circumstances, does not extend four or five 

 miles; the variation of the magnetic needle, how- 

 ever, would lead us to infer that it is not an inert 

 mass, but rather a well-constructed machine in which 

 regular processes are taking place, conducive to its 

 own stability and future renovation. 



Compression of the Poles. 



Capt. Sabine's calculation gives -^^i for the com- 

 pression of the Poles, a very remarkable result, in 

 consequence of its being the same that expresses the 

 ratio of the centrifugal to the gravitating force. 

 When we consider this we cannot help believing that 

 there is some connexion between the external figure 

 of the earth and its internal constitution. 



j-JUg differs from all prior results, but is entitled to 

 the highest confidence. Indeed the singular coinci- 

 dence that from time to time has been found between 



