PROPERTIES OF THE COLLOIDAL SOIL MATERIAL 13 
by the juxtaposition of these particles. In such a system the number 
of capillary spaces of a given diameter would be determined largely 
by the number of particles of a given diameter. It is interesting 
to note, in connection with this theory, that the quantities of vapor 
adsorbed by different soil colloids are roughly proportional to the 
estimated number of particles per gram. 
SWELLING 
In 1895 Williams {56) called attention to the swelling of his finest 
mechanical separate of soils as one of the properties distinguishing 
it from the coarser soil fraction. Tempany (58), Hardy (£2), and 
others have investigated the swelling of different soils, and Geclroiz 
(15) has studied the swelling of the same soil subject to different 
treatments. Hardy gives no direct determinations of the swelling 
of isolated colloids, but he suggests that the colloids from different 
soils may vary in degrees of swelling. 
Swelling might be generally defined as an increase in volume at- 
tendant on the taking up of a liquid or vapor. In the taking up of 
liquids by substances which do not swell, probably essentially the 
same mechanism is involved as in the taking up of liquids by sub- 
stances which do swell. The difference in the two cases may lie in 
that the process does not involve a change in structure of the colloid 
in the first case, but does involve a progressive alteration in struc- 
ture in the second case. Freundlich (i-^), for instance, considers 
that there is no sharp distinction in the mechanism by which the 
liquid is taken up by elastic and inelectic gels, that is, by gels that 
do and do not swell. 
Ignited or aged silica gel is usually cited as a typical inelastic 
gel. A granule of this material may take up considerable watei 
with little, if any, change in apparent volume. Thus it does not 
swell according to the common conception of swelling. It is gen- 
erally considered that the gel mass is porous and that the liquid 
taken up is held in the pores, which do not change appreciably in 
volume during the process. The typical elastic gels on the other 
hand, of which gelatin is an example, increase in apparent volume 
as water is added. The dry gelatin gel is usually pictured as a re- 
ticular structure of gelatin with interstitial spaces filled with air or 
liquid. As liquid is taken up, the interstitial spaces expand. 
It is usually assumed that when elastic or inelastic gels take up 
large quantities of a liquid that the liquid occupies approximately 
its normal liquid volume in the gel. Some hold, however, that the 
lesser amounts of liquid, such as would be absorbed from a vapor 
below its saturation pressure, are held under such compression as tc 
quite appreciably diminish the normal liquid volume. It seems that 
the taking up of liquids by elastic and inelastic gels is very similar. 
Differences in the quantity of liquid taken up and the manner in 
which it is held may be of degree onty, and may be due chiefly to 
differences in the size of pores, to differences in rigidity of the gel 
framework, and to differences in the affinity for the liquid. 
In order fully to chara. terize a gel by its swelling, it is obviously 
important to know the quantities of liquids it holds under various 
conditions and the volume and energy changes accompanying the 
taking up of the liquids. Various studies by Van Bemmelen (6), 
