118 Mr Fisher, Densities of the Earth's Crust, etc. 



deflection of the plumb-line seawards in geodetic surveys points to 

 the same conclusion. 



Having decided that (8 — Cj) must be positive, all the five 

 equal products must be positive. 



Hence, by examining these successively, we see that 



from (1), p!>fi and 8>c 1} 



from (2), p 1 > p 2 and c^ + c 2 > 8 + Te lt 



from (3), a > p 2 and c x + c 2 > $ + k x + k 2 , 



from (4), cr > a , 



from (5), <f> 2 > <£j and 8 + h x > c 1} 



from (6), a- > (p 2 . 



The original assumptions were, that the density of the ocean 

 is less than that of rock, that the density of the upper layer of 

 the continental crust is less than that of the lower layer, and that 

 the density of the substratum of the continental crust is greater 

 than that of the lower layer. The results we have obtained are, 

 (1) that the thickness of the upper layer of the continental crust 

 is less than the depth of the ocean at any place where gravity has 

 its average value. From (2) that the density of the upper layer 

 of the suboceanic crust is greater than that of the lower layer, and 

 since from (5) we see that k ± is less than (8 — Cj) we gather that 

 this dense upper layer must be very thin. From (3) we learn that 

 the total thickness (cj + c 2 ) of the continental crust is greater than 

 the depth of the ocean and the thickness of the suboceanic crust 

 together, viz. than (8 + k x + k 2 ). From (4) we learn that a is 

 greater than cr' , i.e. that the density of the substratum beneath the 

 continent is greater than that beneath the ocean. 



The bearing of these results upon questions of geology appears 

 to be, that, assuming a liquid substratum beneath the crust of 

 the earth, the solid crust is thinner beneath the oceans than 

 beneath the land, but is at the same time more dense. The 

 substratum being thus less dense beneath the suboceanic crust 

 will be affected by upward convection currents, which must neces- 

 sarily flow horizontally towards the continents, and descend there. 

 The continental crust dips more deeply into the substratum, and 

 consequently these horizontal currents, impinging on it, will tend 

 to compress it, and cause corrugations in it. The upper layer of 

 the suboceanic crust, being very thin and very dense, probably 

 consists of basic lava flows from suboceanic volcanoes, which by 

 oxidation would produce the red clay covering the abysmal areas. 

 The upper layer of the continental crust is thinner than the 

 depth of the true ocean, and of small density, which would agree 

 with its sedimentary origin. 



