Prof. J. Milne—Distribution of Voleanos. — 169 
(Fig. 1). If immediately beneath the crust of the earth we imagine 
a liquid material to exist, the pressure at H might be transmitted in 
the direction of h, and beneath this latter column we might expect a 
fracture to take place, and volcanic material to be thrown out. Under 
conditions like these, volcanos would be formed along a line parallel 
to a coast rather than on such coasts, the position where they actually 
occur. If, however, the rocks beneath the crust are solid, as we have 
supposed, this horizontal transmission would not take place, and 
therefore rather than along the line of weakest pressure, if fracture 
occurred it would be along the line of greatest pressure where the 
giving way would be experienced. On this supposition, the volcanos 
would be found upon the top of the slopes rather than at the bottom, 
and this, it is needless to remark, is where the greater number of 
them actually occur. . 
If we regard volcanic vents as having any connexion with the 
pressures exerted by the layers which form the crust of the earth, 
we see why they should rather be found at the upper end of a steep 
slope than near its foot. What I wish next to point out is the reason 
why volcanos are so often found at the upper end of a steep slope 
which rises from beneath the sea. 
To commence with, let us suppose that for-a long period, say for 
example 10.000 years, the hundred mile slopes of which I have 
spoken have had the same relative position to land and water as they 
have at present. This is almost equivalent to saying let us suppose 
that the Pacific Ocean and the continents surrounding it to have had 
pretty much the same relation to each other 10,000 years ago as they 
have now—a supposition which I do not think will meet with any 
serious objections. 
Dana, in his Geology, p. 749, says—‘ As planes of equal tempera- 
ture within the earth have a nearly uniform distance from the sur- 
face, the accumulation of sedimentary beds in the sinking trough 
would occasion, as Herschel long since urged, the corresponding 
rising of heat from below, so that with 40,000 feet of such accumu- 
lations a given isothermal plane would have been raised 40,000 feet.” 
The first portion of this argument I think is hardly just. 
From experiment we know something about the isothermal surfaces 
beneath the land on which we live, but about the isothermal surfaces 
in the land which lies beneath the ocean we can only speculate. 
Without entering into any calculations on the subject, it is not at 
all unlikely that, as in one case we have land cooling beneath an 
atmosphere and the compensating effects of a sun, whilst in the 
other case we have land cooling beneath water, which from all we 
know about deep-sea temperatures is usually very cold, we should 
find any given isotherm at a much greater depth beneath the rocks, 
which form the bed of an ocean, as compared with the depth at which 
we find it beneath the rocks which form the land. 
The consequence of this would be that the ground beneath the bed 
of an ocean would be more rigid than beneath the atmosphere, and 
also that the region of rock at a fusing temperature would be nearer 
to the surface under the land area than beneath a sea area. 
