252 
Journal of Agricultural Research 
Vol. XXXI, No. 3 
Table I. — Influence of colloidal silica on efficiency of acid phosphate and rock 
phosphate in sand , as shown by growth of millet 
Pot 
Nos. 
Differential treatment of pots 
Weight of plants above ground 
Kind of phosphate 
applied 
Phos¬ 
phoric 
acid 
(P 2 O 5 ) 
applied 
per 
pot 
Colloidal silica 
applied per 
pot 
Oven-dry weight 
of plants 
per pot 
Average 
oven-dry 
weight of 
plants 
per pot 
Relative 
efficiency 
of the 
phosphor¬ 
ic acid 
(phosphor¬ 
ic acid 
of acid 
phosphate 
without 
silica 
gel=100) 
1.2.3.. ... 
4,5,6. 
7.8.9.. ... 
10 . 11 . 12 .. 
13.14.15.. 
16,17,18- 
19.20.21.. 
22.23.24.. 
25 _ 
26 _ 
No phosphate. 
Acid phosphate. 
_do. 
_do.. 
Rock phosphate. 
No phosphate... 
Acid phosphate. 
Rock phosphate. 
_do__ 
_do... 
Gram 
0.03 
.06 
.12 
Grams 
None.. 
.do..... 
_do.. 
Grams 
0.07 
1.35 
4.61 
9.17 
.24 
_do.. 
50 silica gel_ 
4.30 
.32 
.06 
.24 
.24 
_do_ 
_do_ 
150 silica gel... 
4.40 
9.79 
9.10 
.24 
45 “mixed” 
gel. 
.11 
Grams 
Grams 
0.06 
0.09 
1.09 
1.27 
4.20 
4.48 
9.56 
8.53 
2.56 
4.64 
.34 
.15 
5.55 
6.12 
8.61 
9.99 
Grams 
0.07 
1.24 
4.43 
9.09 
3.83 
100 
100 
100 
23 
.27 
5.36 
9.46 
9.10 
.11 
119 
51 
49 
0 
The increase in growth with increasing quantities of P 2 0 5 from 
acid phosphate without silica gel shows the extent to which growth 
was dependent upon the supply of phosphoric acid. Since the 
increased growth was practically a straight-line function of the 
quantity of acid phosphate applied, it is possible, by interpolating on 
the curve, to calculate, with very slight error, the growth that 
would have been produced by any quantity of P 2 0 5 from acid phos¬ 
phate below 0.12 gm. The efficiency of the P 2 0 5 of rock phosphate 
with and without silica gel and of acid phosphate with silica gel can 
therefore be calculated against that of acid phosphate alone on the 
basis of the quantities of phosphoric acid required to produce 
oqual increases m growth as previously suggested (7). 
It is apparent mat 50 grams of silica gel per pot increased the 
efficiency of the P 2 0 5 in acid phosphate slightly, raising it from 100 
to 119, while the gel more than doubled the efficiency of the P 2 0 5 in 
rock phosphate, increasing it from 23 to 51 as compared with 100 
for the P 2 0 5 of acid phosphate. The 150-gram application of silica 
gel gave practically the same increase in efficiency of rock phosphate 
as the 50-gram application, within experimental error. The mixed 
gel containing iron and alumina completely nullified the efficiency 
of rock phosphate. Where no phosphate was applied the silica gel 
increased growth by 0.20 gram. This increase is equivalent to what 
would have been produced by 0.0045 gram of P 2 0 5 from acid phos¬ 
phate; it may reasonably be ascribed to the gel rendering the phos¬ 
phorus impurities in the quartz sand, in the nutrient salts, and in 
the containers used more available to the plants. 
The results just given indicate that colloidal silica, in a simple 
medium of quartz sand and nutrient salts, increases the efficiency of 
acid phosphate slightly and that of rock phosphate greatly. It is 
believed that the beneficial action of the silica gel lies in increasing 
the quantity of P 2 0 5 in solution or the rate at which it goes into 
solution. A quite different explanation of the action of colloidal 
silica in increasing the growth of certain crops in soil and sand 
