342 TlMEHRI. 
perfectly free from iron, and pure lime were employed, 
no effect whatever would be produced on the super- 
phosphate. 
Experiments V and VI. — Hydrate of iron was pre- 
pared by adding ammonia to a boiling solution of ferric 
chloride. The resulting precipilate was collected, well 
washed with boiling water and dried at ioo — uo°c. It 
was then in the form of dark-red hard lumps, which 
were reduced to fine powder in an agate mortar. The 
powder dissolved only very slowly in cold hydrochloric 
acid, and was almost unacted upon by dilute acetic acid. 
Five parts of it were mixed with 250 parts of kaolin, 
and placed in a flask together with 1000 parts of super- 
phosphate solution. The mixture was shaken at inter- 
vals for five days and filtered. It was then found 
that the liquid contained only "45 of phosphoric acid 
per 1000, or 55 per cent, had become insoluble. This 
experiment was repeated, and the mixture allowed to 
stand for 30 days with occasional shaking. The phos- 
phoric acid in solution was thus reduced to '38 per 1000, 
or 62 per cent, had become fixed. 
Experiments VII and VIII. — Hydrate of alumina 
that had been precipitated from an aluminum salt was 
tried next. This substance had not been heated, but had 
become dry spontaneously by exposure to air for several 
months : it was in the form of very hard compact 
masses. After being reduced to fine powder, 5 parts 
were well mixed with 250 part of kaolin, and 1000 
parts of superphosphate solution were added. After 5 
days the amount of phosphoric acid remaining in the 
liquid was '38 per cent, per 1000, or 62 per cent, had 
been removed. The experiment repeated, and the mix- 
