and Figure of the Earth. 33 



ficult to demonstrate that it does not even increase less rapidly. 

 As far, however, as any conjecture can be formed from the loose 

 analogy of the elasticity of steam, compared with that of water 

 and ice, the elasticity of a solid might, perhaps, be expected to 

 vary in the sesquiplicate ratio of the density, but certainly not 

 in the duplicate. 



However this may be, M. Laplace's hypothesis is not cor- 

 rectly applicable to the internal structure of the earth ; since it 

 either makes the mean density too small in comparison with 

 that of the surface, or the compressibility at the surface too 

 great; and if this hypothesis actually represented the law of 

 nature, it would follow that the earth is not " chemically homo- 

 geneous," but that the specific gravity of the internal parts is 

 naturally greater than that of the external. In this respect the 

 simple analogy of elastic fluids will afford us a result more 

 conformable to observation. 



M. Laplace supposes the mean density of the earth to be .5|, 

 according to Mr. Cavendish's experiments, and the superficial 

 density 2^ only. Now there is absolutely no rock, either pri- 

 mitive or secondary, of which the specific gravity is less than about 

 2| , and the mean of a great number of rocks gives at least 2| : so 

 that, allowing for a moderate admixture of metallic substances, 

 we can only consider it as certain that the specific gravity must 

 be between 2| and 3 ; and taking 2f for Shehallien, the mean 

 density of the earth, according to Maskelyne's observations, and 

 Hutton's computations, ought to be 4.95. The determination 

 of Cavendish, however, is susceptible of greater accuracy : his 

 result is 5.48, and it will be safest to adopt 5.4, as the most 

 probable mean of the two series of experiments. 



The superficial compressibility, assumed by M. Laplace, is 

 much greater than can be admitted, according to the expe- 

 riments of Chladni on sound, and to those which have been made 

 in this country, as belonging to any solid mineral substance 

 whatever. A column, of the height of one millionth of the 

 earth's axis, is supposed to produce an increase of density 

 amounting to 5.5345 millionths. Now the modulus of elasticity 

 of glass, and of other compact mineral substances, is generally 



Vol.. IX. D 



