184 Prof. Davy on a Simple and Expeditious Method of 
determination, we ascertain the exact quantity of iron solution, 
and therefore of iron, necessary to produce this effect ; and from 
this we easily calculate the amount of phosphoric acid present 
in the manner before explained. 
Though the ferrocyanide and the sulphocyanide of potassium 
may be substituted for the gallic acid, using them in the way 
I have described for that acid, still I have found gallic acid to 
give more satisfactory results than either of them, and to be an 
exceedingly sensitive test to the presence of iron in solution. 
For I found, by direct experiment with 2 cubic centimetres of 
the iron solution containing 0:2 parts of a grain of iron, diluted 
with distilled water so as to make up 300 cubic centimetres, that 
one drop of this mixture let fall on the gallic acid paper, produced 
a faint purplish stain, and the smallest drop which could be 
taken up on the top of a glass rod gave with a solution of gallic 
acid a very decided effect ; so that this test is one of great deli- 
cacy in ascertaining the point when the slightest excess of iron 
has been added to the mixture in this method for the estimation 
of phosphoric acid and the phosphates. I may observe that 
when the phosphate under examination is very largely diluted 
with water, it will lead to more accurate results to reduce by 
evaporation the bulk of the liquid (having previously added a 
little hydrochloric acid in those cases where evaporation would 
cause the precipitation of any insoluble phosphates),—the effect 
of large dilution, by its rendering the gallic acid less sensitive to 
the point when iron is in excess, being to indicate a somewhat 
greater amount of phosphoric acid than is present. 
This source of inaccuracy might also in a great measure be 
obviated by seeing how much of the iron solution was necessary 
to give an indication of that metal when it was diluted with a 
bulk of fluid equal to that employed in the experiment; and 
this being deducted from the quantity of iron necessary in the 
determination, would give a very close approximation to the real 
amount required to combine with the phosphoric acid, where it 
might be inconvenient and occupy too much time to evaporate 
the liquid. 
Hitherto the estimation of phosphoric acid volumetrically by 
the use of a standard solution of iron has been thought by many 
to be impracticable, as the analyses of different chemists show that 
the composition of phosphate of iron is subject to great variation. 
This, however, may be accounted for by the different circum- 
stances under which it has been formed by those experimenters, 
which give rise to phosphates of different constitution. But I 
entertain no doubt that, by always placing the oxide of iron and 
the phosphoric acid under the same conditions, compounds of 
the same constitution would in every case be formed. 
