RESEARCH METHODS IN STUDY OF FOREST ENVIRONMENT. 109 
ground and pressed before the preliminary freezing many serious changes may 
take place; enzymes may be liberated, many new chemical reactions may be 
brought about, and the solutions may change in various physical ways. After 
thawing, the material should be stirred with a glass or suitable wooden rod 
to expel all air bubbles. Thawing may be completed within 15 or 20 minutes 
at most, and the possibility of chemical change is thus very greatly reduced. 
When thawing is complete, the thermometer is inserted, the tube is placed in 
the freezing mixture, and the material allowed to reach a temperature about 
1° below its freezing point. Solidification is then brought about, as has been 
stated, by turning the thermometer backward and forward a few times to 
create a slight disturbance in the pulp. It has been found in practice that 
much more satisfactory results are obtained if the material is thus allowed to 
undercool about 1° than when solidification is brought about with less under- 
cooling. In the latter case the mercury rises to the freezing point much more 
slowly and the determination of this point is consequently more difficult. 
Osmotic pressure pi soils. 
Although it is possible to remove the soil solution from the soil 
and to determine its osmotic pressure by the freezing-point method, 
this will fall far short of the desired end, which is to determine how 
the water behaves in the presence of the capillary forces and ad- 
sorption tendencies of the soil particles and colloids. As has been 
suggested in the introductory paragraphs to this chapter, these in- 
fluences may run parallel with the influences of dissolved salts in 
the soil water. 
The freezing-point determinations for moist soils are so similar in 
method to those for plant pulps that it seems unnecessary to describe 
them here in detail. The reader is especially referred to the descrip- 
tion given by Bouyoucos (107) . It would seem that the fundamental 
consideration in testing a given soil at various moisture contents is 
to have samples very evenly wetted. This is accomplished by placing 
the sample in a moisture-tight jar and, after thorough shaking, allow- 
ing it to stand one or more days, so that the moisture is evenly dis- 
tributed and has ample opportunity to be adsorbed. While it is 
possible to use measured amounts of water in wetting the soil, it is 
probably safer procedure to take moisture samples at the same time 
that samples are taken from the jar for freezing tests. 
Vapor transfer in soils. — The vapor transfer method is the only 
other method of osmotic determination which appears feasible for 
soils, and where time is not an important element, it is believed to be 
preferable to the freezing-point method because of the possibility 
of treating soils in their natural states. It should be understood, how T - 
ever, that the value of the method is as yet theoretical rather than 
proven. 
The work of Alway (103) and Hilgard (125) on the hygroscopic 
coefficient of soils has already been mentioned, with the suggestion 
that since the moisture boxes used could not completely prevent the 
