EARTHQUAKES—BULLEN 329 
pressure experiments on rocks made by Adams, Williamson, Bridg- 
man, Birch, and others. This work indicates that the outer mantle 
probably consists of ultrabasic rock, an olivine composition being 
commonly assumed; other possibilities that have been considered are 
that the rocks are eclogite or peridotite. Birch has suggested that 
the lower mantle may consist of phases including silica, magnesia, and 
iron oxide. It is possible that an appreciable quantity of free iron 
also occurs in the mantle. It is not yet certain whether the changes 
in the outer mantle are essentially progressive composition changes 
or are due to change of crystal type or other physical transformation. 
There is as yet no widely accepted explanation of the character of the 
changes inside the lowest 100 miles of the mantle. 
The question of the composition of the central core has lately be- 
come interesting. Until a few years ago, it was widely accepted 
that the whole core consists predominantly of iron and nickel. This 
conclusion was based on observations of meteorite composition, and 
on the known relatively high density in the core. 
Then in 1941 two European geophysicists put forward the radically 
different theory that the core consists of compressed hydrogen. Over- 
whelming arguments against this theory were quickly stated, but the 
theory led to an important calculation in 1946 which showed that, at 
700,000 atmospheres (a pressure reached inside the earth’s lower 
mantle), the density of hydrogen would suddenly jump from about 
0.4 to 0.8 gm/cm’. This led to the idea, first advanced by W. H. 
Ramsey of Great Britain, that perhaps at the huge pressure of 114 
million atmospheres reached at the outer core boundary, the material 
there existing might suddenly jump in density from 514 to 914 
gm/cm?® as a direct consequence of the high pressure and not as a 
change of composition from rock to iron. Such modification of the 
rock (or other material) of the lower mantle would incidentally have 
all the physical properties of a metal and be electrically conducting. 
Thus theories which attribute the earth’s magnetism to currents in 
the outer core would not be affected. 
An interesting point is that, on the new theory, the density in the 
earth’s interior would be largely determined by pressure alone, which 
would not be the case if the mantle and core were of distinct chemical 
composition. Hence, knowing the pressure-density relation for the 
earth, it becomes possible on the new theory, given the mass of a 
planet of identical composition, to compute its diameter. Starting 
from the known masses of Venus and Mars, calculations made in this 
way give in fact fair agreement with the observed diameters of the 
two planets, and also with the observed ellipticity of figure of Mars. 
In contrast, Jeffreys has shown that Venus, the Earth, and Mars must 
have considerably different over-all compositions if the earth’s mantle 
and core are chemically distinct. Hence the Ramsey theory is closely 
