UNITED STATES DEPARTMENT OF AGRICULTURE 
DEPARTMENT BULLETIN No. 1311 
Washington, D. C. 
October, 1924 
THE CHEMICAL COMPOSITION OF SOIL COLLOIDS 
By W. O. Robinson and R. S. Holmes, Scientists in Soil Laboratory Investiga- 
tions, Bureau of Soils 
Contents 
Introduction 
Review of earlier work 
Ultimate chemical composition- 
Compounds present 
Soils selected for a study of the 
colloidal material 
Field description 
Chemical composition 
Isolation and description of the col- 
loidal material 
Ultimate chemical composition of the 
colloidal material 
Compositions of the colloidal 
materials of different soils — 
Difference in composition of col- 
loidal material, whole soil, 
and coarser particles 
Relation of sample analyzed to 
total colloidal matter in the 
soil 
Tage 
8 
S 
11 
13 
14 
14 
17 
18 I 
Page 
Ultimate chemical composition of the 
colloidal material — Continued. 
Cause of variation in composi- 
tion of the colloidal material- 20 
Compounds present in colloidal ma- 
terial 25 
Fragments of soil-forming min- 
erals 25 
Attempted fractionation of col- 
loidal material 27 
Action of solvents on colloidal 
material 28 
Compounds indicated hy stoicbio- 
raetrical calculation 34 
Conclusions regarding compounds 
present 37 
Summary 38 
Literature cited 39 
INTRODUCTION 
In the past conceptions regarding the chemical composition of 
soil colloids have been largely based on inference. During the early 
days of soil investigation it was assumed that the finest particles of 
the soil, those of colloidal dimensions, were of the same nature as the 
larger particles, differing only in size. This view, however, did not 
satisfactorily explain the reactivity with inorganic salts, particu- 
larly the property of base exchange. To account for this it was as- 
sumed that the finest particles of the soil consisted chiefly of zeolites, 
which are especially reactive minerals. This view was discarded 
later when it became evident from mineralogical investigations that 
definite crystalline zeolites could rarely, if ever, be identified in the 
soil. It is now coming to be quite generally believed that the inor- 
ganic colloidal material of the soil is mainly of a gel-like nature, 
much like the artificial gels of silica, iron, and alumina. 
The earlier conceptions of the composition of soil colloids were 
for the most part based upon analyses of, or experiments with, the 
whole soil. An examination of the colloidal material itself is ob- 
10875—24 1 
