THE CHEMICAL COMPOSITION OF SOIL COLLOIDS 
19 
ing several colloidal fractions successively extracted from a single 
soil. For this purpose four soils were extracted as completely as pos- 
sible of their colloidal matter. The details of the extraction are 
given in a previous bulletin (17). The results of the analyses are 
given in Table o. 
Table 5. — Composition of successively extracted fractions of colloidal material 
from certain soils 
No. 
Source and fraction of colloidal 
material 
Part of 
whole 
soil 
Huntington loam, subsoil: 
First colloidal fraction . 
Second colloidal fraction 
Third colloidal fraction 
Weighted average of all frac- 
tions I 
Fraction between 0.3 micron 
and 1 micron 
Marshall silt loam, surface: 
First colloidal fraction 
Second colloidal fraction 
Third colloidal fraction 
Fourth colloidal fraction 
Weighted average of all frac- 
tions 
Sharkey clay: 
First colloidal fraction 
Second colloidal fraction 
Third colloidal fraction 
Fourth colloidal fraction 
Weighted average of all frac- 
tions 
Vega Baja, surface: 
First colloidal fraction 
Second colloidal fraction 
Third colloidal fraction 
Weighted average of all frac- 
tions 
Per cent 
2.10 
6.06 
2.13 
10.29 
.33 
.64 
4.50 
.97 
.90 
7.01 
12.10 
11.90 
12.30 
2.20 
Si0 2 
TiO- i AI2O3 
0.51 ; 29.62 
.56 ! 28.98 
. 59 28. 00 ' 14. 18 
Per cent Per cent Per cent 
39.15 
38.84 
45.81 
40.34 
45.14 
45.92 
45.09 
45.98 
50.28 
45.96 
51.34 
49.67 
50.17 
43.58 
50.00 
Fe 2 3 
CaO 
Per cent 
14.11 
14.77 
12.57 
Igni- 
MgO tion 
loss 
16. 60 i 36. 26 
12. 20 1 35. 45 
1. 10 ' 35. 93 
30. 40 i 35. 91 
29.81 
.45 I 22.48 
. 44 I 22. 54 
23.08 
19.61 
44 : 22. 23 
.51 
.55 
.68 
1.00 
.61 
.65 
22.61 i 
22. 73 
19. 17 
20.37 I 
21. 42 I 
32. 85 
32. 86 
29. 69 
05 I 
9.11 
8.00 
7.38 
8.72 
8.79 
9.32 
8.28 
9.81 
8.85 
12.44 
14.37 
11.68 
79 i 32. 74 13. 22 
Per cent 
0.51 
.67 
.65 
.03 
.37 
1.23 
1.15 
1.26 
1.21 
1.18 j 
1. 41 
L40 ! 
1.12 
1.17 
Ptr cent Per cent 
1. 33 12. 73 
1.39 
1.19 
1.36 
.94 
1.67 
1.80 
1.92 
1.60 
12.97 
12.65 
12.95 
9.71 
17.45 
17.44 
17.21 
1. 78 17. 37 
1.30 
.44 
.42 I 
.26 
2.47 
2.09 
2.00 
.18 
.24 
.44 
, 42 . 21 
10.93 
12.46 
13.36 
18.66 
12.62 
16.88 
16.21 
18.52 
16.66 
1 Not determined. 
The fact that there is comparatively little variation in the compo- 
sition of colloidal matter extracted from the same soil shows that 
soil colloids are very difficult to separate into fractions of different 
composition and that if two or more colloids are present the mix- 
ture is very intimate and hard to break up. 
Although a small part of the colloid extracted from a soil appears 
to be fairly representative of the total colloid that it is possible to 
extract, we have no data showing the relation between the com- 
position of the extracted colloid and the composition of colloidal 
matter that is not extracted b} T our methods. It is probable that the 
colloid not extracted would approach more nearly the composition 
of the last fraction of colloidal matter extracted and would there- 
fore vary in composition from the total extractable colloid. The 
colloid not extracted by our methods may be as much as half of 
the total quantity of colloidal matter in the soil, as determined 
by both absorption and microscopical methods. 
The colloids whose compositions are given in Table 3 were graded 
to about 0.3 micron as the size diameter of the largest particles. 
One micron has, however, been used as the dividing line between 
colloids and noncolloids in other investigations of the bureau (i, 17). 
A comparison of the composition of the dispersible colloidal matter 
between 0.3 and 1 micron and the extracted colloid below 0.3 micron 
