Mr . Watt's Observations 
of one species of molecules, it would be resolved into an aggre- 
gation of crystals of the same substance ; and probably, by a 
still farther continuation of the process of arrangement, into one 
crystal, which, though it might not possess a regular external 
form, would be perfect in its internal structure. 
But, if the mass contains two distinct species of molecules, 
different results must take place, which will be modified by the 
proportional quantities of the components. As it has been demon- 
strated by Berthollet, that the attraction of masses of matter 
are relatively as their quantities, it follows, that unless a very 
potent counteracting cause be exerted, the most abundant ingre- 
dient in the mixture will be the first to crystallize. But this 
crystallization will not comprehend the whole of its molecules ; 
for, after a certain quantity of them are arranged, the propor- 
tions of the remaining fluid are altered ; that ingredient which 
was before the least, may now be equal, or even greatest, and it 
will exercise its attraction. As the first crystallization, by sub- 
tracting a large portion of the fluid particles, must have obliged 
the molecules of the less abundant substance to approach each 
other very closely, they may be able to collect themselves en- 
tirely in their first attempt to crystallize, or they may form 
alternate crystallizations with the remaining unarranged mole- 
cules of the more abundant substance. However various the 
species of molecules may be, they will be regulated by analogous 
laws, and only serve to diversify the generated substances. 
It by no means follows, that the crystals afterwards found to 
be most infusible would be first generated. Their formation 
does not altogether depend on their greater or less fusibility, 
but on the relative strength of the attraction which unites them 
to the matter they are immersed in, and of the polarity which 
