PROCEEDINGS OF GEOLOGICAL SOCIETIES. 
14^9 
ammonia to it ; it then forms a dense white precipitate. In that state it 
is combined with water. It is perfectly soluble in caustic potash, which 
combines with it, and forms a clear solution. 
As far as we know at present, it does not appear that the process of 
dialysis can have played a very important part witli regard to the forma- 
tion of alumina in nature, or rather in the separation of alumina. Pro- 
fessor Graham has spoken of the dialysis of certain salts of alumina. For 
instance, one of these is basic acetate of alumina, which may be decom- 
posed by dialysis. The chloride of ahiminiuin is another. There is, how- 
ever, a salt which he has not examined at present ; that is the sulphate, 
and that is the most likely of all to occur in nature. Therefore, for the 
present, we have no reason to suppose, from the experiments reported, 
that this process has any applicability to the formation of alurnuia in 
nature. 
iS^ow, let us look at crystalline alumina. This is a subject which has a 
bearing upon geological phenomena. A minute crystal found in a rock 
shall be a clear exponent of the formation of that rock. However small 
it may be, it shall tell a most elo(|uent and indisputable story with refer- 
ence to its origin. He thought that what he said was not overstated with 
regard to the formation of some of the mineral constituents, though 
occurring in small quantitv in some of our so-called " igneous " rocks. 
The late distinguished, most excellent Professor Ebelmcn made nume- 
rous experiments upon the subject of artificially-crystallized alumina. 
He exhibited specimens received from him years ago. They are exceed- 
ingly small, but rcmarkabl}' characteristic. Here are some small scales of 
alumina crystallized by the operation of fire. The process consists of 
heating alumina, mixed with about three or four times its weight of borax, 
in a small platinum crucible, under certain precautions, in a porcelain 
furnace — a porcelain furnace where we obtam hard porcelain, and not the 
BO-called porcelain which we make in this country, and which is made at 
a low temperature. It must be the true China porcelain furnace, such as 
was in operation at Sevres, where he experimented. He exposed the 
borax and alumina at a high temperature in this furnace. The borax dis- 
solved the alumina, and after a time oil' went the boracic acid and the 
sodium. The crystallized alumina remained at last in brilliant colourless 
scales. This is exceedingly similar to what is obtained at ordinary tempe- 
ratures with liquid solvents. It is only just to the memory of poor old 
Berthier, who died not long ago, to slate that the lecturer found, in looking 
over his papers, that he really was the first man to adopt this mode of 
forming nuneral substances, and obtaining them in a crystallized state. 
He did not use borax, but he used an equivalent, namely, comnion salt. 
He formed crystallized silicate of lime and magnesia by exposing them 
with salt on precisely the same ])rinciple as when borax is used. We are 
indebted to JJevillo for that which is now the commonest process, and it is 
a very beautiful one. The operation is performed in a common crucible. 
He lines that with charcoal, which is a common process, — what we call 
brasquing, — thus making an inner lining of charcoal all the way ; then 
above he puts a little capsule or cupel, made of charcoal, — a little cup- 
shaped piece of charcoal ; he then covers the whole with a charcoal top, 
and afterwards places on the cover, and lutes it down. 
In the lower part of the crucible he puts fluoride of aluminnim ; m 
the upper part he puts fused boracic acid. He exposes the whole to a 
high temperature, and the two bodies vaporize. The two vapours come 
into contact, and they actually decompose each other; and the result 
