V74 = Arrhenius’ Theory of Physical Causes of Vulcanicity. 
‘probably iron. From the known data it is easy to calculate that about 
80 per cent. of the earth’s diameter must be occupied by gaseous iron, 
15 per cent. by gaseous rock-magma, and 4 per cent. by liquid 
‘magma ; the rest, less than 1 per cent., forms the solid crust. 
These theoretical conclusions are supported by the results of earth- 
quake observations. The seismograph record of a distant earthquake 
shock shows a clear division into two phases, viz. preliminary tremors 
and the principal shock. The explanation of this fact is, as Milne 
pointed out, that the principal shock is propagated along the crust, 
while the preliminary tremors travel along the chord of the are. 
Since these latter are much diminished in intensity, it follows that 
the internal friction of the material of the inner parts of the globe 
must be very great, which is in harmony with the assumption of 
a liquid or gaseous condition rather than with the solid state. 
(2) Volcanic Phenomena. 
It is unnecessary to prove in detail that in all volcanic eruptions 
‘great quantities of gases, and in particular water-vapour, are given 
off, and the pressure of this gaseous water at a temperature above its 
critical point is undoubtedly the chief motive-power of the eruption. 
Each explosion relieves the pressure for the time being, but the 
frequent recurrence of eruptions from the same vent shows that 
further supplies of water must by some means reach the magma, and 
this in spite of enormous counter-pressure. The ocean-floor, with its 
capillaries, may be supposed to act as a semi-permeable membrane, 
through which the water penetrates by osmotic pressure. Assuming 
‘the density of the magma to be about 2°7, it can easily be calculated 
‘that in order.to overcome the pressure of the overlying column of 
rock the osmotic pressure at a depth of 10,000 metres must be about 
1,700 atmospheres, and this in itself is not inconceivable. But since 
‘the dissclved constituents of the magma show this high osmotic’ 
pressure they must enter into chemical combination with the water. 
According to ‘homsen’s measurements water at the ordinary 
temperature is a very weak acid, about 100 times weaker than 
silicic acid, but on raising the temperature these relations no longer 
‘hold. While silicic acid, which has a heat of neutralization equal to 
‘those of nitrie and hydrochloric acids, does not change its strength 
notably, the case is very different with water. From the data with 
regard to the behaviour of water between 0° and 50°, it can be 
calculated by a well-known extrapolation formula that at 300° water 
and silicic acid are equally strong, at 1,000° water is 80 times, and at 
2,000° 300 times as strong as silicic acid. It is true that this result 
is obtained purely by extrapolation, but it is of no practical im- 
portance whether water is 300 times or only 100 times as strong as 
silicic acid, so long as it is distinctly the stronger, and this the 
calculation shows with certainty. 
(3) Zhe Chemical and Physical Action of Water on the Magiinat 
Let us now suppose water to be in contact through a semi- 
permeable membrane (the sea-floor), with a magma between 1,000° 
and 2,000° C. Water penetrates into the magma by osmosis and acts 
as a relatively powerful acid on the silicates, by which means silicic 
