168 MR GOODCHILD. 
cations of the original lie of the rocks in which they have 
been formed, as it must be obvious that any inclination 
those layers now possess must mark the amount of tilt the 
rocks have undergone since they were formed. There seems 
to be no well-authenticated instance of onyx bands found 
after the rocks have undergone any disturbance. This, 
again, tends to confirm the conclusion that agates date from 
an early period in the history of the parent rock. 
It is generally considered that the process of filling the 
cavity with the complex materials which go to form an 
agate is largely due to the peculiar physical process known 
as Osmosis. For this to act with its fullest effect a porous 
septum or partition must have, on its opposite sides, two 
fluids, each of which must be capable of exercising some 
physical action upon that septum different from that of the 
other in both degree and kind. The result is that an inter- 
change arises, the substance in solution on either side passing 
through the partition to the side opposite. Where the 
chemical conditions are favourable the osmotic transfer may 
go on through a partition or septum which to all appear- 
ance is perfectly solid and compact. The general result is, 
however, liable to considerable modifications arising from 
variations in both the composition and the density of the 
solutions (especially upon the percentage of the alkaline 
carbonates present), upon the nature of the septum, upon the 
pressure, and, lastly, upon the temperature existing there 
while these actions are in progress. 
There can be no doubt that it is to Osmosis that we have 
to look for an explanation of the mode of entry of the 
different substances which form an agate. But it must be 
confessed that the subject is beset with many difficulties. 
One of these relates to the fact that Chaleedony undoubtedly 
consists of a mixture of both the colloid and the crystalloid 
forms of silica. Perhaps the explanation of the anomaly 
may lie in the fact that the solutions varied somewhat in 
chemical composition at different times, and that sometimes, 
perhaps, a small relative increase or decrease in the quantity 
of alkaline carbonate present may have favoured the diffusion 
of a larger or a smaller quantity than usual of silica in the 
colloidal form. On the other hand, silica which had passed 
