27S ON THE TARTRATE OP IRON AND AMMONIA. 
When 30 grains of the hydrous salt was carefully incine- 
rated, a residue of 8.70 grains of sesqui-oxide of iron was ob- 
tained; 30 grains of the hydrous tartrate is composed of 25.29 
grains of dry salt, and 4.71 of water; then, theoretically, the 
quantity of sesqui-oxide in 25.29 grains of the anhydrous salt is 
8.83 grains, nearly that obtained by experiment. 
An aqueous solution, containing a known weight of the 
double tartrate, was precipitated by alcohol 40° Baume. The 
precipitate, when collected and dried, was found to have lost 
weight nearly in proportion to that dried in vacuo. A portion 
of this was incinerated, and the quantity of sesqui-oxide ob- 
tained was found to be in the proportion of that yielded by 
the salt when dried in vacuo. 
The fact that one of the bases, as well as the acid in this 
salt is volatile, presents a difficulty in its analysis which is not 
found in that of tartrate of iron and potassa. While the esti- 
mation of the iron is comparatively easy, that of the acid is 
proportionably difficult. Attempts were made to estimate the 
quantity of tartaric acid, with the chlorides of calcium and ba- 
rium, and the bisulphate of potassa, but in each instance, the 
precipitate obtained did not contain half the acid that theory 
required; the object was attained, however, by the employment 
of nitrate of lead. 
To 30 grains of tartrate of iron and ammonia in solution, an 
excess of nitrate of lead was added, the mixture heated, and 
when cold, the precipitate, washed, dried, and weighed, 
amounted to 44 grains, and had a yellowish color. 
When 30 grains of this precipitate was carefully incinerated, 
20 grains of residue was obtained, and the loss (tartaric acid) 
10 grains. Consequently, if 30 : 10 :: 44 : 14.66 grains, or 
the amount of acid in 30 grains of the hydrous salt, by the re- 
sult ot this experiment. Theoretically, as 273 (the equivalent 
of the hydrous salt) is to 133, (the quantity of acid it contains) 
as 30 (hydrous salt) is to 14.61, (the quantity of acid it should 
contain.) Thus it is evident, that the theoretical and practical 
results are sufficiently near to infer the presence of two atoms 
