61 
The following table represents the above in combination : 
Ferrous Sulphate 
I. 
4090*4 
II. 
5454*3 
Aluminium Sulphate 
Calcium Sulphate 
1358-9 
922-3 
700*7 
376-0 
Magnesium Sulphate 
670-3 
225-7 
Potassium Sulphate , , 
70-3 
Sodium Sulphate 
49*9 
47*0 
Lead Sulphate 
1*25 
-- 
Calcium Chloride 
16-8 
Sodium Chloride 
19-4 
Sodium Nitrate 
4-8 
— - 
Ammonium Nitrate " 
7-2 
Aluminium Phosphate 
3-2 
— - - 
Manganese... 
trace 
trace 
Silica 
157-0 
149*0 
Albumenoid Ammonia 
0-34 
Bases for which there is not sufficient Acid 
15-5 
1*4 Loss 
7370-79 
6970-9 
With reference to this analysis the following observa- 
tions are to be made : — 
(1.) The determination of the total residue was first made 
at 180°C., as recommended by FreseniuS;^ and the result 
obtained corresponded to 8,100 parts per 1,000,000; it was 
found however that this was much too high, the reason being 
that ferrous sulphate, though it looses six molecules of water 
at 114°C., yet retains the seventh even at 280f. In order to 
drive off this remaining molecule, the residue from lOOcc of 
water was heated in an air bath to 300° — 310° and weighed; 
after repeated heating two successive weighings did not 
differ by more than a milligramme. In heating to so high 
a temperature, however, there is a danger of a little sul- 
phuric acid volatilising by decomposition of the sulphate of 
iron, but by careful heating this may be avoided ; a loss of 
ammonia will, nevertheless, have been incurred, but as this, 
together with the trace of organic matter, did not amount 
to more than 8 to 10 parts per 1,000,000, it was not of very 
much consequence. 
* Fresenius. Quantitative Analysis, 4th Edition, p. 560. 
t Watts. Dictionary of Chemistry. Vol. 5, p. 597. 
