103 
of siliceous fossilization the process goes on in something 
like the following way. The clear transparent rods first 
lose a portion of their water, becoming transformed into 
the opalescent or semi-gelatinous-looking form No. II, and in 
so doing they begin to lose some of their internal structure, 
that is, the layers adhere more closely and the rods become 
brittle and lose the glass-like flexibility they originally 
possessed. At this point any further loss of water causes 
crystallization to set in, the silica not being able with a 
lower proportion of water to remain in the “ organic” state, 
and if this crystallization sets in at a great many points the 
nodule soon becomes an opaque mass of crystalline silica as 
in variety No. Ill; but if crystallization sets in at compara- 
tively few places and spreads slowly out, it forms the kind 
of nodules drawn in Fig. 10. If at this point any solvent 
agent comes in contact with the rod it dissolves away the 
non-crystalline portion, leaving nodules as shown by Dr. 
Young and in Fig. 11 ; but if no solvent action goes on the 
rod ultimately becomes transformed into variety III, which 
is crystalline silica. 
The transformation from “organic 5 ' to “inorganic 55 silica 
is one of arrangement, and as the “ organic 55 arrangement 
requires a certain amount of water, whenever the propor- 
tion falls below this a simpler arrangement sets in, and we 
have crystalline silica. That the rods of Hyalonema 
Smithii are the same in constitution at least as those from 
modern sponges is I think proved from these experiments, 
and to show a further analogy I have drawn a fractured 
specimen of such in Figs. 4 and 5, the similarity of which 
is very striking. 
