Feb. 5,1917 
Measurement of Inactive Moisture in Soil 
207 
The foregoing data show that the amount of water which fails to freeze 
in the second freezing is practically the same as in the first in the case of 
sands and light sandy loams or noncolloidal soils, but in the silts, loams, 
clay loams, and clays, or complex and colloidal types it is considerably 
less in the second freezing than in the first. In other words, repeated 
freezings increase the amount of free water and decrease to a correspond¬ 
ing degree the quantity of unfree water in the case of complex and col¬ 
loidal types of soil but not in the simple and noncolloidal types of soil. 
As will be observed, in some of the loams and clays about 19 per cent more 
water froze in the second than in the first freezing. 
These results therefore demonstrate that successive freezings cause 
some of the unfree or inactive water to become free in some soils and not 
in others. They go to prove, therefore, the validity of the hypothesis 
already mentioned, that the diminution of the lowering of the freezing 
point of colloidal soils at low moisture content with successive freezings 
is due to the dilution of the initial soil solution brought about by the 
inactive water in the soil becoming free through the process of freezing. 
It must have been already realized, however, that it is not only the 
freezing that causes the unfree or inactive water to become free but also 
the thawing. If it were due only to freezing, then all the water that 
would become free would have done so during the first freezing. In¬ 
stead, a large amount of water becomes free after the soil is thawed and 
frozen a second time. 
Finally, it must be stated that if 10 c. c. of water are added to the 25, 
gm. of soil instead of 5 c. c., as was done above, the amount of water that 
fails to freeze in the second freezing is practically the same as that in the 
first, even in the complex colloidal types of soil. These results confirm 
those obtained with the freezing-point method, which show that the’degree 
of the lowering of the freezing point of all types of soil at very high mois¬ 
ture content remains quite constant with successive freezings. 
EFFECT OF DEGREE OF SUPERCOOLING UPON THE AMOUNT OF WATER THAT 
FAILS TO FREEZE 
The amount of water that fails to freeze is dependent to a considerable- 
extent upon the degree of supercooling. This is, however, true only in 
the complex and colloidal soils and not in the simple noncolloidal soils. 
The data bearing upon this phase of the investigation are presented in 
Table VI. These data were obtained at two degrees of supercooling, 1 0 
and 3 0 C., in a cooling mixture of — 4 0 . For each degree of supercooling 
a new sample of soil was used. 
From an inspection of the data in Table VI it becomes at once clearly 
evident that the amount of water that fails to freeze is considerably less 
when the soil is supercooled to 3 0 C. than when it is supercooled to i°. 
This, however, is true only in the case of the complex types of soil, the 
silts, loams, clay loams, and clays, but not in the simple types of soil*, 
