ORIGIN OP ERUPTIVE AND PRIMARY ROCKS. 313 



Naumann quietly remarks that the theory of the two separate 

 reservoirs is surely not yet sufficiently proved, and characterises Von 

 Waltershausen's theory of theconstitution of the interior of the earth 

 as more natural, and more in accordance with our knowledge re- 

 garding the probable condition of. the earth's centre.* This theory, 

 which the author first promulgated in the work from which we 

 have just quoted, deserves to be better known. It is principally 

 founded upon certain reasonings deducible from the density of 

 the earth, and for this reason a recapitulation of what is known 

 concerning this point may not be inappropriate here. 



In 1776 Hutton and Maskelyne determined the density of the 

 earth from the attraction exerted on the plumb line by the mass of 

 the mountain Schiehallion in Perthshire. Assuming the mean of 

 specific gravities of the three principal rocks, of which it consists, 

 viz., mica slate, limestones, and quartzite, to be the density of the 

 whole mass, they calculated from their experiments the density of 

 the earth to be 4.713. 



The density of the earth has also been determined from observa- 

 tions on the oscillations of the pendulum on high mountains. In 

 this way Carlini found from experiments on Mount Cenis the den- 

 sity of the earth to be equal to 4.37, which value was however 

 raised by Schmidt to 4.837, by correcting an- error in Carlini's 

 calculations. 



The most exact method however yet applied towards deter- 

 mining the density of the earth is that by means of the torsion 

 balance invented by the Rev. John Mitchell, and used after his 

 death by Cavendish. In 1798 this philosopher communicated to 

 the Royal Society the result of his experiments with this appara- 

 tus. From seventeen sets of experiments he deduced twenty- 

 three results, from the mean of which he computed the density of 

 the earth to be equal to 5.48. Bailly, correcting an error in 

 Cavendish's calculation, makes it 5.45. Schmidt, likewise, after a 

 revision of Cavendish's computations, alters the result of these to 

 6.52. In 1837, Reich of Freiberg performed a series of experi- 

 ments with the same apparatus, much improved in various par- 

 ticulars. Fifty-seven experiments were made in all, from which 

 fourteen results were deduced, the mean of which makes the density 

 of the earth equal to 5.44. In 1848 Baily, at the request of the 

 Astronomical Society ,undertook to repeat Cavendish's experiments. 



* Lehrbuch, ii, 1101. 



