ON THE FORMATION AND GROWTH OF CRYSTALS. 
351 
twenty-four hours. It is then removed from the vessel, and from 
the gut, in the one case, are removed all the octahedra, excepting 
those more perfectly formed together with the whole of those 
which may have attached themselves to the bottom of the vessel. 
The prisms in the other case, which form upon the flat bottom 
of the vessel, are in like manner examined — the more perfect se- 
lected, the remainder removed. The respective liquors are then 
returned ; at the bottom of the one are arranged the prisms, while 
in the other the gut is resuspended. 
After another interval they are again similarly examined, small im- 
perfect crystals'removed, which, from the beginning, are to be per- 
served, the positions of the prisms reversed, and both — returned to 
their respective liquors — again set aside. 
This removal of the small crystals is repeated daily, so long as 
any continue to be deposited, since by their growth, they would 
weaken the solution. 
When the prisms and octahedrons have ceased to increase, and 
none, or scarcely any are deposited at the bottom, &c, the liquor 
may be removed to a cooler place, — the lower part of the building 
for instance — which in consequence of the temperature being more 
constant, is, perhaps, on the whole, the most suitable for the subse- 
quent growth of the crystals. 
When the liquor is at length so far reduced in strength that no 
further depositions are made, and no perceptible increase in the 
elected crystals is observable, the accumulated small crystals are 
redissolved by small quantities at a time, and added to the mother- 
liquor as before explained. These being exhausted, the crystals 
may be increased at pleasure, by preparing from a highly concen- 
trated solution a quantity of the same compound, to be added 
gradually as before. 
It remains to be added that in consequence of the liquor be- 
coming denser in the lower than in the upper strata, and their con- 
sequent more rapid growth in the former, it is necessary not only 
to regulate the strength of the solution, in the manner referred to, 
but to reverse the position of the crystals themselves almost daily. 
The gut, therefore, which has hitherto pierced the octahedra, 
must be cut off as close as possible, and attached externally by a 
noose, which will facilitate its removal and attachment. 
It ought also to be added, that when a crystal is suspended, it 
