8 
N. C. AGRICULTURAL EXPERIMENT STATION. 
j|. SOME EXPERIMENTS ON THE EFFECT OF WASHING THE MAGNESIUM- 
AMMONIUM-PHOSPHATE PRECIPITATE UPON THE 
PER CENT. OF P2O* 
By B. W. KILGORE, Assistant Chemist. 
In the analysis of high grade phosphates and mixed fertilizers the 
error from incomplete washing of the white precipitate with dilute 
ammonia water is more common, we think, than is generally recog¬ 
nized. This source of error may affect light as well as heavy precipi¬ 
tates, but it is more liable to do so, and be larger with high ones, as 
they are more difficult to wash thoroughly. During the past three 
years quite a number of experiments have been made in this labor¬ 
atory on this point, but most of the results have not been preserved. 
A few, made at different times, but not with a view of publication, 
are presented in the table below (table II), and will serve to illus¬ 
trate our point, though they do not show the full magnitude of the 
differences that may and do, we think, occur from incomplete wash¬ 
ing when the per cent, of P 2 O 5 reaches 20 per cent, or more. We 
have, in fact, had differences greater than the ones here presented to 
occur, and on phosphates no higher in phosphoric acid than the ones 
on which these results were obtained. Differences of .20 per cent, to 
.50 per cent, may easily occur in some precipitates between 10 or 12 and 
15 or 18 washings. It is hardly necessary to say that there is a very 
great difference in magnesium-ammonium phosphate precipitates, 
even from the same material and solution, as to the ease or difficulty 
with which they can be freed from chlorids, depending on the excess 
of “magnesia mixture,” conditions of precipitation, kind of precip¬ 
itate (crystalline or amorphous), and other things, for aught we know, 
but just where the greatest difficulty lies, and how to overcome it, 
we do not venture to say, but that such difference and difficulty 
exists we have no doubt. We have had two precipitates from the 
same material, precipitated at different times and under different 
conditions, one of which could be freed of chlorids with 12'or 15 
washings, while the other would require 20 or even 25; but where 
the precipitations are made at the same time and under the same 
conditions there is nearly always close agreement in the per cent, of 
P 2 O 5 with equal amounts of washing, even when not free of chlo¬ 
rids. When the P 2 O 5 is above 10 per cent, we do not feel safe with 
less than 18 or 20 good washings, stirring the precipitate well each 
time with the stream of wash solution, and then we test occasionally 
to see if the precipitates are clean. 
This difficulty would not occur where the Gooch crucible is used 
to catch the precipitate, certainly not to the same extent as where 
paper is used. 
