132 
CLARKE. 
cubic miles are ocean and 1,633,000,000 are solid matter- 
The mass of the atmosphere is equivalent to that of 1,268,000 
cubic miles of water, the unit of density. For these data, 
which cover all the terrestrial matter accessible to us, I am 
indebted to Mr. R. S. Woodward of the U. S. Geological 
Survey, and from them the relative masses of solid crust, 
ocean, and atmosphere can be determined within narrow 
limits of uncertainty. To sea water we may assign a density 
of T03, which is a trifle too high, and to the solid rocks a 
specific gravity not lower in average than 2*5, nor much 
higher than 27. With these values we can get the follow¬ 
ing expression for the percentage composition of the known 
matter of the globe: 
density. density. 
Per cent. of crust 2-5. of crust 2-7. 
Atmosphere_._ *03 -03 
Ocean j_____ 7*08 6*58 
Solid crust_____- 92-89 93-39 
100-00 10000 
In short, we may regard the earth’s crust, to a depth of 
ten miles, as composed essentially of 93 per cent, solid and 
7 per cent, liquid matter; treating the atmosphere as a small 
correction to be applied later. More elaborate estimations 
are unnecessary. Since the known nitrogen of the earth is 
mainly in the atmosphere, its relative scarcity as an element 
is at once curiously manifest. It cannot possibly exceed 
25 thousandths of one per cent of the total. 
For the composition of the ocean, the data given by Ditt- 
mar in the Reports of the Challenger Expedition are perhaps 
the best available. The maximum salinity he puts at 37*37 
grammes of salts in the kilogramme of water, and by taking 
this figure instead of a lower value we can make an allow¬ 
ance for saline masses enclosed in the solid crust, which 
would not otherwise appear in the final averages. Combin¬ 
ing this datum with Dittmar’s statement of the average com¬ 
position of the ocean salts, we get the second of the subjoined 
columns. The traces of other elements, not named here, 
