on Basalt, See. 
289 
separation is always perfectly defined. But the state of aggrega- 
tion into which the substance has now entered, is so perfect as 
to overcome the operation of the causes which formerly induced 
the fibrous structure, and the mass remains compact. The only 
change that the substance afterwards undergoes, consists in the 
gradual accumulation of the crystalline molecules, and their 
arrangement, by their individual polarity, into regular solids. 
This depends on very different laws from those which consoli- 
dated the fluid glass, and aggregated its particles into a compact 
uniform stone.* 
The appearances that I have endeavoured to describe, seem 
deserving of consideration in several points of view. Few things 
can be more at variance with commonly received opinion, than 
the diversified succession of changes of structure which this glass 
exhibits in its passage to a crystallized state. The generation of 
the globules which unite to form the jaspideous substance, is 
what we might be prepared to expect, by observing the cooling 
* The case is considerably different, where crystals possessing regular forms are 
generated in glass. The molecules of which they are formed, have doubtless been 
only suspended in the vitreous medium ; and their union is determined by crystalline 
polarity, which appears to me perfectly distinct from the simple aggregation which 
changes a fluid into a solid, whether it be homogeneous or compound, which affects 
the internal arrangement of those bodies, but which never can separate their compo- 
nents into distinct masses, or form them into regular solids. Every molecule, at the 
moment of its formation, must necessarily be endowed with all the properties it after- 
wards possesses. The suspension of such molecules in a fluid medium, though it may 
conceal, cannot alter those properties; and the union of such molecules, to form a 
regular solid, in no respect alters their individual or aggregate qualities. Whether heat 
be evolved at the moment of this union, is a question not easily solved ; as the crys- 
tallizations with which we are familiar are from chemical solutions, in which some 
of the molecules are generated by the separation of a combined substance, at the 
moment when others are united by crystalline polarity. 
P p 2 
