ESTIMATION OF COLLOIDAL MATERIAL IX SOILS. 6 
-elected as the upper limit for colloidal particles rather than the 
more conventional 0.1 micron, since a fairly sharp separation of par- 
ticles can be made at 1 micron by subsidence or by centrifuging and 
the separation can readily be checked by microscopical observation. 
At 0.1 micron the control of size is inaccurate when dealing with 
mixtures of colloidal material, and a sharp separation by subsidence 
or centrifuging is exceedingly difficult. Soil particles 1 micron in 
diameter show a distinct Brownian movement, which is usually 
considered a colloidal characteristic. Also, some earlier investigators 
have used 1 to 2 microns as the upper limit of size of discrete 
"primary" or "ultimate"'' particles. It seems advisable, however, 
to place the limit of size for unaltered soil particles as low as can be 
accurately estimated, i. e.. at 1 micron. 
The third kind of material classified as colloidal is that of certain 
aggregates or lumps of material which may occur in sizes up to 50 
microns or even more in diameter. With the ultramicroscope it can 
be seen that these lumps are merely aggregates of small particles 
which are less than a micron in diameter. The constituent particles 
were not dispersable, however, by the methods we employed for this 
purpose. These lumps doubtless exist as such in the soil, although 
it is possible that they have been formed in some cases by the 
mechanical treatment adopted for fractionation of the soil. 
Distinctly mineral particles which are larger than 1 micron in diam- 
eter are here treated as constituting the noncolloidal part of the soil. 
PREVIOUS METHODS OF ESTIMATING THE COLLOIDAL MATERIAL IN 
SOILS. 
The first to call particular attention to the colloidal matter in soils 
were Schloesing and Hilgard. Schloesing (31 , 32) rubbed up clays 
and agricultural soils with considerable quantities of water, separated 
the coarser sands by decantation, treated the washings containing 
the finer soil particles with acid to dissolve CaC0 3 , and then resus- 
pended the fine material in water containing a little ammonia. The 
fine material from 5 grams of soil which remained in suspension for 
24 hours in 2 liters of water he called "clay." He reported (32) 
16 to 20 per cent of such material in heavy soils. 
According to Schloesing (33, p. 476, and 34, p. 67), this clay frac- 
tion, as well as a natural clay, contained two classes of substances: 
Very fine sands devoid of cohesion, and an amorphous substance, 
which he called 'Targile colloidale," that cemented the sand grains 
and was responsible for the plasticity of soils. This "colloidal clay" 
was very coherent, had a horny appearance, and was made up of 
formless (possibly submicroscopic) particles which would remain in 
suspension indefinitely. The colloidal clay was separated from the 
other material in the clay fraction by long-continued subsidence 
lasting several weeks or several months, with occasional decant ations 
to get rid of noncolloidal clay which settled out. The separation of 
colloidal clay from the very fine sands was considered complete when, 
'Te microscope ne permet plus de decouvrir dans le liquide aucun 
element figure.' ' (34, p. 67.) He stated that clays or soils con- 
tained very rarely more than 1.5 per cent of l'argile colloidale. 
Schloesing's statement that the colloidal clay was free from other 
material when no shaped particles were visible microscopically 
