Aug. 1, 1925 
Colloidal Silica and Efficiency of Phosphates 
249 
Thus there is considerable evidence that the reactions taking place 
when a phosphate is added to a soil may increase or decrease the 
efficiency of different phosphatic fertilizers, and that these reactions 
vary in different soils. Little is definitely known about these reac¬ 
tions, although Liebscher {21, p. 208 ) and subsequent investigators 
have pointed out that soils high in iron and aluminum usually respond 
to phosphate fertilization, and many have shown that calcium car¬ 
bonate influences the efficiency of the phosphates {17, 33, 34, 28, 8 ). 
Consequently, although we have a general idea of some types of soils 
which respond poorly to certain phosphates, we can not predict 
with any certainty from a laboratory examination of a soil what its 
effect will be on any particular phosphate. A determination of the 
soil constituents and the soil conditions affecting the availability 
of the different phosphates is obviously an essential step toward 
scientific fertilizer practice. 
It may well be that the properties of a soil affecting the efficiency 
of phosphates are largely localized in the colloidal material. The 
absorptive and base-exchange properties of the soil, which are almost 
exclusively a function of the colloidal material, might be expected 
to influence the solubility or decomposition of the relatively insoluble 
phosphates. Although Gedroiz (4) has suggested that insoluble 
phosphates are more efficient in absorptively unsaturated soils, the 
influence of the soil colloidal material on the efficiency of phosphates 
has not been directly studied. 
The soil colloidal material is a complex substance or mixture of 
substances. It is quite variable in composition but is made up 
chiefly of silicon, aluminum, iron, and organic matter. It also con¬ 
tains small amounts of magnesium, calcium, potassium, sodium, man¬ 
ganese, titanium, phosphorus, chlorine, and sulphur {30 ). The condi¬ 
tion of the elements in the colloidal material has not been definitely 
determined, but the constituents may be regarded as forming a mixed 
gel similar in nature to the artificial gels of silica, iron, alumina, etc. 
Some results obtained in this laboratory indicate that a mixed 
gel such as occurs in soils may have certain properties not possessed 
by simpler gels. Nevertheless, it seemed advisable to determine the 
influence of relatively simple gels on the efficiency of phosphates pre¬ 
liminary to dealing with the colloidal material of different soils. 
The effects of an iron gel alone and of an alumina gel mixed with 
silica gel have been tested by others. Prianishnikov {27, p. 36) found 
that hydrous ferric oxide depressed the efficiency of bone meal 
markedly in a sand culture experiment with barley. In an experi¬ 
ment of Pfeiffer and Blanck {25), the addition of a mixture of alumina 
and silica gels to a sand culture depressed the quantity of phosphoric 
acid assimilated by yellow lupines from potassium acid phosphate. 
The influence of a silica gel on the efficiency of rock phosphate and 
acid phosphate is shown m the following experiments. 
INFLUENCE OF COLLOIDAL SILICA ON PHOSPHATE EFFICIENCY, 
AS SHOWN BY THE GROWTH OF MILLET 
It was desirable to determine the effect of colloidal silica on the 
efficiency of rock phosphate and acid phosphate under the simplest 
conditions possible. Since organic matter, other colloids, and re¬ 
active soil minerals might complicate the effect, the test was con¬ 
ducted in a fairly pure quartz sand (99.5 per cent silica). 
