ORIGIN OF ERUPTIVE AND PRIMARY ROCKS. 319 



regarding the decrease of gravity from the pole to the equator. 

 It appears however that Sartorius Von Waltershausen's estimate 

 of the average specific gravity of the constituents of the earth's 

 crust at its surface is too high, since it is well known that the 

 land only occupies one-fourth of the earth's surface, and that the 

 sea has sometimes a depth of more than 27,600 feet. It may 

 probably be assumed with some dearree of reason that the average 

 specific gravity of the first few thousand feet of the earth's crust 

 below the level of the sea, does not exceed 1.5. With regard to 

 the metals constituting the earth's centre, it will probably be 

 admitted that they exist there somewhat in the same proportion 

 as they occur on the surface, that consequently iron constitutes 

 by far the greater portion of the central mass. This supposition 

 seems confirmed by the fact that among the gaseous products 

 emitted by volcanoes, chloride of iron is very abundant, while 

 traces only of the chloride of lead and copper have been de- 

 tected. Since further, meteoric iron may be supposed to come 

 from bodies having a common origin with our earth, their com- 

 position might be supposed to afford a clue, however slight, to the 

 composition of the metallic centre of the earth. It would there- 

 fore seem not unreasonable to suppose that this centre is mainly 

 composed of metallic iron, combined with copper, cobalt, nickel, 

 lead, and perhaps silver, gold and platinum in comparatively 

 small quantity, and that its specific gravity may be estimated on 

 account of this admixture of heavier metals at 8. (Sp. gr. of 

 malleable iron 7. 78 ; cast iron, 7. 1 to 7. 5.) If we assume 1. 5 

 as the density of the earth's surface, and 8. as that of its centre, 

 we must also — since the average density of the earth is 5. 56 — 

 suppose the existence at the centre, of a globe of metallic matter 

 having a radius of 2245 English geographical miles. Assum- 

 ing further a gradual increase of density from the surface of the 

 earth to the surface of this metallic globe, we may calculate that 

 at a depth of 132 miles the density of trachytic lava is reached, 

 (2. 5), and at 202 miles the density of doleritic lava is slightly 

 exceeded (3. 0). According to this calculation therefore the 

 crust of the earth has a thickness of from 132 to 202 miles, 

 a result somewhat exceeding Naumann's estimate. Calculating 

 in the same way we further find that from a depth of 202 miles 

 to that of 352, molten rock would exist having a specific gravity of 

 from 3. to 4. 0, and containing much more basic matter and iron- 

 oxide than any rock now visible on the surface. At a depth of from 



