2 BULLETIN 1122; U. S. DEPAETMENT OF AGRICULTURE. 
had similar absorptive capacities. In 1912 Mitscherlich {16, p. 476) 
made a general statement that the hygroscopicity of noncoUoidal 
soil particles was almost nothing, but that the colloids in different 
soils must have different capacities for absorbing water vapor. Evi- 
dence recently accumulated 'in this laboratory also seems to show 
that practicall}' all the absorptive capacity of the soil is located in 
the colloidal material. For instance, samples of the colloidal mat- 
ter, or ^' ultra clays, "^ extracted from a large number of different 
soils by use of a high-power centrifuge,* show absorptive capacities 
for malachite green, ammonia gas, and water vapor which are two 
to twenty times higher than those of the original soils. 
Moore, Fry, and Middle ton, (17) of this bureau, have suggested 
that the colloidal content of the soil may be measured by comparing 
the absorptive power of the soil for malachite green or ammonia gas 
with the absorptive power of the mixed colloids, or ultra clay, ex- 
tracted from the soil, if soil absorption is due almost entirely to the 
colloids. While the various ultra clays absorb widely different 
amounts of malachite green and of ammonia gas, they have a much 
more constant absorptive capacity for water vaf)or. It was, there- 
fore, suggested by Robinson {21) that the colloidal content of the 
soil might be estimated by dividing its water-absorption value by 
a factor which is an average of the specific absorptions of a large 
number of different ultra claj^s, if the noncoUoidal particles do not 
absorb appreciably^. This latter method would obviate the neces- 
sity of extracting a sample of colloid from the soil. 
Although there is considerable evidence that by far the greater 
part of the soil's absorptive capacity is due to colloids, it is obviously 
important to obtain quantitative data on the relative absorptions 
^f ihe colloidal and noncoUoidal constituents, if absorptive methods 
are to be used for determining the amount of colloids in soils. This 
paper deals chiefly with the absorptive capacities of the noncoUoidal 
;Soil particles formalachite green,water, and ammonia. The absorption 
of such particles is compared with the average absorptive capacities 
of the ultra clays extracted from different soils. Various colloidal 
mineral powders and synthetic gels were also tested for absorption 
in order to throw light on the nature of the soil colloids. 
PREVIOUS WORK. 
Several investigators, including Konig and Hasenbaumer {14), Ogg 
and Hendrick {18), and Atterberg {4) have fractionated soils into va- 
rious sized particles and determined the absorptive power of the 
different fractions for substances, such as water, dyes, gases, or 
salts. The results on the whole show that the absorption by the 
different sized particles increases with the fineness, although not 
in a regular manner, and that the finer soil fractions above colloidal 
size absorb appreciably but much less than the colloidal material. 
These determinations of the noncoUoidal absorption can not be ac- 
cepted without question, however, since the various soil fractions 
were probably not entirely free from colloidal material. 
» The term "ultra clay" has been suggested to denote the mixture of colloidal constituents present in 
* Details of the process are given in the paper by Moore, Fry, and Middleton (/7). 
