ON THE MANUFACTURE OP PRUSSIATE OP POTASH. 175 
tion of ammonia has been attributed to the decomposition 
of cyanide, which has not taken up its proper dose of iron, 
and it has heretofore been recommended to add a proto-salt 
of iron to the liquor formed by dissolving the cake. We 
have not found this addition of any service in preventing the 
formation of ammonia, and as iron in a finely divided state 
is always in excess in the cake, it is scarcely probable that 
there is any considerable quantity of uncombined cyanide 
in the liquor. 
We are more disposed to attribute the ammonia to the 
presence of cyanide, resulting from the oxidation of the 
cyanide in the furnace, when the flame beats too directly 
on the mass. Cyanate of potassa is decomposed on boiling 
into ammonia and bicarbonate of potassa. A more accurate 
investigation of the cause of this formation of ammonia is 
desired by the manufacturer of prussiate, as every equiva- 
lent of ammonia disengaged, indicates the loss of an equi- 
valent of cyanide. 
A modification of the process has been proposed and 
patented in England, by which the animal material is de- 
composed in iron retorts, heated to redness ; the gas disen- 
gaged being conveyed by curved iron pipes under the sur- 
face of a fused mixture of potash and carbon contained in 
another retort. This method is stated by the patentees to 
yield a larger product, and a greater proportion of animal 
material to the potash can be employed, as the carbon re- 
sulting from the decomposition of the animal material being 
retained in the first retort, does not mix with and thicken 
the cake. We have seen no account of the practical v/ork- 
ing of this method, but doubt whether the increased com- 
plicity and expense of apparatus will not counterbalance its 
advantages. 
Other sources of nitrogen than the material usually em- 
ployed, have also been resorted to, as for instance, the crude 
ammoniacal salts from the gas works, the gases resulting 
