339 



Let S represent the present mean depth of the sea; L, and W 

 the areas of dry land and water, as determined by observation ; 

 then as a - «i is evidently the mean depth of the fluid covering 

 the sphere supposed at rest, 

 a -ax W 



y, or g = (a-ai) (l+-^)> 



g L+ W 



2g(5Z)-3) 



cos 01 = / 

 By observation, 



5qDa[\^^^ 



1 L 266 



D~5.5, ^-289' ^y- 734- 



" If, as is generally supposed, the mean depth of the sea 

 be proportional to the mean height of the land above its sur- 

 face in the relation of their respective areas, the greatest value 



g 1 



which can be attributed to - would be -77-7-^, or a mile nearly. 



a 4000 



Then 



24^2_89_>^4X ^ 

 27.5x2000000; ^ 



h = ^/(i 



In this case, therefore, the area of each circumpolar continent 

 would be a little more than a sixth of the area of the entire 



g 1 



surface. If - = -zTTTr-irp, or if the mean depth of the ocean 



a 066.95 



were 7.055 miles, no circumpolar continents would exist. All 

 authorities, however, appear to concur in thinking that so 

 great a mean depth cannot be attributed to the ocean, but, on 

 the contrary, that it must be, at most, some small fraction of 

 the earth's ellipticity. It follows that if the earth were origi- 

 nally spherical, two great circumpolar continents, with an in- 

 termediate equatorial ocean, should necessarily exist. If, in 

 accordance with the assumptions of the theory, the forces 

 tending to transport water towards the equator were more 

 effective than those tending to transport matter towards the 



