ABSOEPTIOX BY SOIL CONSTITUENTS. 3 
Hundeshagen (12) and Dittler (7) tested, in a qualitative manner 
only, the absorption of various minerals for different dyes. 
Ammon (i). and later Von Dobeneck (8), determined the absorp- 
tive powers of a niunber of finely powdered substances, including 
quartz, calcite, and humus for water vapor, ammonia, and other 
gases, and found that quartz and calcite showed a small but appreci- 
ble absorption while humus absorbed to a marked degree. Using 
quartz of different degrees of fineness these investigators found ab- 
sorption higher in the finer fractions, but the increase due to fineness 
was less than the proportional increase in surface. 
Stremme and Aarnio (24) showed that rock powders absorb but a 
small amount of dye and water as compared with the amount ab- 
sorbed by natural clays or artificial gels. The exact fineness of the 
rock powders was not given, so the data throw but little light on our 
problem. 
DISTINCTION BETWEEN COLLOID AND NONCOLLOID. 
Usually colloids and noncoUoids are distinguished by the size of 
the particles. But this distinction is not all-sufficient, especially 
when dealing with soil constituents, since an important part of the 
colloidal material in soils may be present in such a condition that it is 
not readily dispersable. 
In the work reported here, mineral particles larger than 1 micron 
in diameter are assmned as noncolloids and all particles smaller than 
this as colloids. All the organic matter of the soil is also classed as 
colloid, irrespective of whether or not it happens to exist in a state of 
subdivision less than 1 micron. The insoluble part of more or less 
decomposed plant remains is of course largely made up of substances 
such as the celluloses, proteins, and their derivatives, which are uni- 
versally classified as colloids, although in the dried state those sub- 
stances may not be readily dispersable. 
Certain particles, or more accurately aggregates of particles, larger 
than 1 micron in diameter, are also classed as colloids. These parti- 
cles under the ultra microscope show no evidence of being crystalline 
and appear to be merely aggregates of particles which are less than 1 
micron in diameter. They are presumably either diflB.cultly reversible 
or irreversible colloids. 
The use of 1 micron as the dividing line between colloidal and non- 
collodal mineral particles permits good microscopical control; since 
the form of particles 1 micron in size can usually be readily distin- 
guished and such particles need not be confused with points of light 
emanating from very much smaller particles. 
Although 1 micron is an arbitrary limit for the upper diameter of 
colloidal mineral particles, there is some justification for its selection, 
aside from mere convenience of microscopical observation. Mineral 
particles which are 1 micron in diameter show a distinct Brownian 
movement, while larger particles show practically no such activity. 
The international commission in 1913 decided on 2 microns as the 
upper limit for colloidal clay {23, p. 189). 
P OUTLINE OF INVESTIGATION. 
The most obvious way of obtaining data on the relative absorptions 
of the colloidal and noncolloidal soil material would be to separate 
