MECHANISM OF VOLCANIC ACTION—JOHNSTON-LAVIS. 307 
mental sheet of rubber, already referred to, we can quite understand 
how any fluid in the earth’s interior would tend to be squeezed out, 
but are met by two difficulties—(1) is there any fluid in the earth’s 
interior? and* (2) is the tensile strength of the contractile shells 
sufficient to have a squeezing power ? 
Three classes of views have been held as to the constitution of our 
globe. Some hold that it is like an egg, a solid shell with a fluid 
interior; others maintain that by the increase of gravity as the center 
is approached there is a solid nucleus which is potentially fluid were 
it not for this gravitational condensation, so that there would be a 
solid nucleus, a solid crust, and a stratum of liquid rock séparating 
them. Finally, there is a third school who holds that the highly 
heated nucleus, although potentially fluid, is really solid in conse- 
quence of pressure or, more correctly, gravitational condensation. 
No known rock that we are acquainted with gives the conception 
of having sufficient tensile strength to be capable of exerting any 
really contractile or squeezing power on fluid inclosed within it or 
surrounded by it. There will be a tendency as the inner shells 
contract to split by fissures. Such fissures would extend from 
within outward, and would be top-shaped in section, with the edge 
extending up to the neutral zone of no contraction, and their lower 
limit at the inner surface of the lowest shell (pl. 1, E, F.) Such 
a fissure might be simultaneously filled by the fluid rock paste beneath. 
How this filling will take place requires consideration. As there is 
reason to disbelieve in any considerable constricting power of the 
inner cooling shells, and that even if such constricting power did exist 
is would be annulled by the development of fissures within its mass, it 
is evident one must look to other causes. The welling up into the 
fissure of the fluid rock, if we admit a fluid nucleus or a stratum or 
shell of such fluid, might be due to the settling down by gravitation of 
the cooled blocks * of crust limited by the fissures. If, on the con- 
trary, we admit the immediate contact of the lowest cooling shell 
with a highly incandescent nucleus (P), solid by pressure, but poten- 
tially fluid when this pressure is removed, we can well see what would 
take place. As soon as the fissures and therefore fluid in the inner 
cooling shells begin to form, their location and their edges will repre- 
sent a site of diminished pressure. The subjacent and neighboring 
but potentially fluid rock will in consequence liquefy and expand and 
fill the fissure. As the fissure broadens and extends so will the 
expansion and liquefaction increase pari passu. 
Liquid rock may thus reach up to the neutral zone of no contraction, 
but its extension further must be a matter of chance. It is evident 
that if the shells of compression were in every part homogeneous and 
“These blocks are quite different to the blocks referred to by some recent 
writers on terrestrial mechanics, 
