258 
Journal of Agricultural Research 
Vol. XXXI, No. 3 
The action of silica gel in the simple medium of these experiments 
probably does not correspond closely to the action of the colloidal 
material which is present in a soil. The results do show that a sub¬ 
stance of the same general physical, but not chemical, nature as the 
soil colloidal material may markedly increase the efficiency of a 
phosphate fertilizer. Previous studies have shown that other sub¬ 
stances of the same general physical nature as the soil colloidal 
material (an iron gel and an alumina gel acting with a silica gel) may 
appreciably depress the efficiency of phosphate fertilizers. As pre¬ 
viously mentioned, the soil colloidal material is a mixed gel contain¬ 
ing many different elements but consisting chiefly of silica, iron, and 
alumina. It might be supposed that such a mixed gel would have an 
action on phosphates approaching that of a silica gel. Apparently 
this is not necessarily so. The mixed gel used in these experiments 
depressed the efficiency of rock phosphate to zero, which is all that 
would be expected of a pure iron gel. At first thought this seems 
rather surprising. It should be borne in mind, however, that while 
a pure iron or alumina gel has somewhat of an adsorptive power for 
electrolytes, it does not give an appreciable concentration of iron or 
alumina in solution when treated with a neutral salt. On the other 
hand, it was found by Anderson, of this laboratory, that the mixed 
gel which he had prepared gave a heavy concentration of iron and 
alumina in solution when treated with a fairly concentrated potas¬ 
sium chloride solution. The depressing action of the mixed gel in 
the writers 7 experiments was therefore probably due to the main¬ 
tenance of sufficient iron and alumina m solution to preclude the 
presence of phosphate ions. 
SUMMARY 
In an experiment with millet grown in sand culture, the addition 
of silica gel greatly increased the growth of plants supplied with 
rock phosphate and only slightly increased the growth of plants 
supplied with acid phosphate. The addition of a mixed gel contain¬ 
ing iron, alumina, and silica to a pot receiving rock-phosphate was 
marked by reduced growth as compared with the growth made in 
pots receiving no phosphoric acid. 
Growth made by the plants in the different pots was approxi¬ 
mately proportional to the quantities of phosphoric acid in the 
plants, but seemed to bear no relation to the quantities of silica 
taken up. 
The beneficial action of silica gel on the growth of plants supplied 
with rock phosphate is ascribed to the increasing of the availability 
of this phosphate through increasing the quantity of phosphoric acid 
in solution. Similarly, the deleterious effect of the mixed gel is 
ascribed to a diminishing of the quantity of phosphoric acid in solu¬ 
tion, owing to soluble iron and aluminum produced by base exchange. 
This conclusion is substantiated by the quantities of phosphoric 
acid found in solution on shaking up rock phosphate with silica gel 
and a nutrient solution such as was used in the pot experiment. 
Criticism is made of the idea that the increased growth of plants 
observed in similar experiments is due to an increased assimilation of 
silica which enables the plant to get along with less phosphoric acid. 
The relation of the action of silica gel on the efficiency of phos¬ 
phates to that of the soil colloidal material is discussed. 
