112 BULLETIN 1059, U. S. DEPARTMENT OF AGRICULTURE. 
such that vapor molecules remaining on said surfaces do not crowd 
one another; (3) on the size of the particles, the smaller particles 
having less gravitational power ; and (4) on the extent to which the 
substances in the soil act as solids or crystals with only exterior sur- 
faces available, as spongy masses capable of absorption, or as indi- 
vidual molecules each of which may attract and retain under partial 
control a sufficient number of molecules, so that in the aggregate the 
conditions are those of a liquid. 
The factors that affect the rate of absorption are far less impor- 
tant, but might be briefly mentioned, as follows: (1) The size of 
soil spaces as affecting free passage of vapor molecules; (2) the 
number of chances for a given molecule in motion through a given 
space to encounter an attractive force too strong for it (this has to 
do with the number of particles per unit of Volume, as well as size 
of air spaces) ; (3) the density of the soil particles; (4) the density 
of the vapor molecules as affected by temperature and whole pressure ; 
(5) the conductivity of the soil, governing the rate at which the 
heat of condensation can be eliminated from the soil mass. 
This examination of established facts and theory regarding vapor 
condensation in soils leads to the recent efforts by Bates (105) to 
show that the moisture of soils does exhibit a definite partial vapor 
pressure corresponding to that of a solution, and that the vapor 
transfer method has many latent possibilities. It should be stated 
that these investigations are not yet complete or convincing, but in 
some respects they have gone farther than any others and are worth 
mentioning at least as suggestions for further effort. 
The vapor -transfer method of Bates. — In its simplest form the 
vapor transfer method is similar to the plasmolytic method in at- 
tempting to find a point of osmotic equilibrium between the soil 
moisture and solutions of known concentration. In this case equi- 
librium must be shown by the cessation of transfer of water, through 
vapor, from the soil to the solution, or the reverse. It would be 
possible to take a number of samples of a given soil at a known-mois- 
ture content and place them in relatively small chambers, each with 
a solution of different concentration from the others, and in the 
case in which there was no transfer from the soil to the solution or 
vice versa equilibrium would exist and the osmotic pressure of the 
soil water could be directly calculated from the known concentration 
of the solution. 
In practice, however, it is far more feasible to place the soil 
sample of known moisture content in the vapor of a solution of 
approximately the same osmotic pressure, let the two come into 
equilibrium through vapor transfers, then compute the moisture 
content of the soil, the osmotic pressure of the solution, and the 
approximate original osmotic pressure of the soil on the assumption 
