ON THE DOUBLE SALTS OF IRON. 
97 
shops, this experiment will not be uniformly successful. 
From an experiment which will be detailed in a subsequent 
part of this paper, this combination of oxide of iron and 
citric acid will be inferred to be in the ratio of 80 to 140, or 
of one equivalent of oxide of iron to two equivalents of 
crystallized citric acid. 
In case 4, tartaric acid was substituted for citric, and sub- 
jected to the same treatment as described incase 1. As soon 
as the solution appeared turbid, one-half was removed, fil- 
tered, and dried at 212°, when one hundred grains were 
found to be impregnated with twenty-six grains of sesqui- 
oxide of iron. I did not repeat this experiment. To the 
remaining half, more oxide was added, and during the con- 
tinuance of this part of the operation, a bulky hydrated bi- 
tartrate of iron was thrown down, very nearly resembling 
hydrated sesquioxide of iron, but differing in being totally 
soluble in alkalies. This precipitate was well washed, and 
dried at a gentle heat, which was raised towards the close 
of the process to 212°, and there kept until it ceased to lose 
in weight. One hundred grains were then heated to red- 
ness; it first evolved a suffocating inflammable vapor, then 
charred, and finally left me thirty-seven and a half grains 
of residue ; the residue was dissolved in nitro-hydrochloric 
acid, precipitated with caustic ammonia, and washed. It 
was again heated to redness, when there were found to bo 
thirty-six grains nearly.* 
In case 5, acetic acid was digested at a common tempe- 
rature with the hydrated oxide ; union was readily effected, 
and by adding a little alcohol, and filtering, a beautiful so- 
lution was obtained, which, after the lapse of months, had 
only a very trifling deposit. A quantity of this solution was 
left to spontaneous evaporation in divided portions of say 
half an ounce each ; the great bulk suffered decomposition 
* It is stated in Hose's Manual of Analytical Chemistry, that ammonia 
completely precipitates sesquioxide of iron from its solution in aqua regia ; an 
impression is now gaining ground that such a statement is inaccurate. 
9* 
