io Mr. Davy's Experiments on the 
glass tube, kept very cool; but I found that no moisture 
whatever was separated in the process. I mixed a few grains 
of potassium with red oxide of mercury, and ignited the mix- 
ture in contact with boracic acid, but no elastic product, except 
mercury, was evolved. 
I made some potash by the combustion of potassium in 
a glass tube, and ignition of the peroxide, I added to it dry 
boracic acid, and heated the mixture to redness. Sub-borate 
of potash was formed, and there was not the slightest indica- 
tions of the presence of moisture.* 
It is evident from this chain of facts, that common potash 
* These processes must not however be considered as shewing that boracic acid 
that has been heated to whiteness is entirely free from water; they merely prove that 
such an acid gives off no water by combination with pure potash at a red heat. I 
have found that boracic acid in perfect fusion, and that has been long exposed to the 
blast of a forge, and that has long ceased to effervesce, gives globules of hydrogene ; 
when dry iron filings are made to act upon it. I added to 54 grains of boracic acid in 
complete fusion, in a crucible of platina, 75 grains of flint glass that had been pre- 
viously heated to whiteness, and immediately reduced into powder in a hot iron mor- 
tar; by raising the heat so as to produce combination, a copious effervescence was 
produced ; and after intense ignition for half an hour, the mixture was found to have 
lost three grains and a quarter. 
The combinations of boracic acid with potash and soda, that have been heated to 
redness, I find lose weight when their temperature is raised to a much higher 
degree. Thus, in an experiment made in the laboratory of my friend John George 
Children, Esc^. and in which Mr. Children was so kind as to co-operate, 71 
grains of hydrat of potash, mixed with 96 of boracic acid that had been heated as 
strongly as possible in a blast furnace, lost by fusion together in a red heat 1 1 grains, 
but on raising the temperature to whiteness the loss increased to above 1 3 grains. 
55,5 grains of hydrat of soda, mixed with 80 of boracic acid, examined at intervals 
in a process of this kind, continued to lose weight for half an hour, during which time 
they were frequently heated to whiteness ; at the end of this period the whole loss was 
14 grains, of which at least one grain and a half may be referred to the acid. 95 
grains of soda, ignited to whiteness in a platina crucible, with J40 of dry flint glass, 
lost 22,2 grains ; 80 grains of boracic glass were added to this mixture ; a fresh effer- 
