170 MR GOODCHILD. 
the two being separated by the septum formed of the three 
first formed layers constituting the “Skin.” Osmosis continued 
to act, but instead of the osmotic pressure being comparatively 
feeble, as it was when the solutions inside and out were nearly 
alike in density, the internal pressure increased in direct pro- 
portion to the difference of density of the solutions on either 
side of the septum ; so that when the solution outside the agate 
had increased in concentration to double that of the desilicified 
fluid within, the pressure increased to double that which was 
exerted when the solutions were alike in degree of concentra- 
tion; and so on, always in the same proportion. It may be 
mentioned here that a rise of temperature of the solution within 
the agate causes an expansion in the proportion of 1/273 of its 
volume at 0° cent. for each degree ; and, if the volume be kept 
constant, the increment of pressure with change of temperature 
is also 1/273, in accordance with the same law (Charles’s Law) 
which governs the relation between the temperature and the 
pressure of gases. 
It will, therefore, be evident from these considerations that, 
arrived at a certain stage of growth, the osmotic pressure 
exerted upon the coagulated silica jelly within a nascent 
agate is liable to give rise to great pressure upon its walls. 
In some eases this is a result of continued osmosis, in others 
it is due to a rise of temperature consequent upon crystal- 
lisation ; but usually the former cause is the chief factor. 
As a consequence of this pressure some of the jelly begins 
to squeeze out through any weak points that may occur, in 
much the same manner as oil paint is squeezed out of its 
tube. The material gives way first at a zone nearest the 
bounding walls of the agate, and subsequently the zones, one 
after another, from without inwards, participate in the dis- 
turbance. This may go on until the pressure is relieved, 
which does not appear to have happened, in some instances, 
until perhaps as much as one-tenth of the original has been 
forced out. Several Tubes of Escape usually occur together. 
They may be at once recognised in an agate by the fact that 
the different layers adjoining the tube are attenuated parallel 
to its axis, and are also deflected in the direction of the 
“ Point of Escape.” Usually there is a more or less flask- 
shaped expansion of the tube at its outer extremity. I have 
1 
