504 
Journal of Agricultural Research 
Vol. XXVIII, No 6 
first method. This method'’was found to give much higher results than the 
first method. This was probably due to the fact that the colloidal material was 
morejthoroughly broken up and dispersed. However, the results were less uni¬ 
form and very difficult to duplicate due to the fact that the same degree of dis¬ 
persion was difficult to attain in two different samples of the same material. 
Another difficulty was encountered in-drying the material. At the surface it 
would become very dry and hard while the interior was still very moist, so that 
by the time the critical moisture content of the whole sample was reached, the 
moisture was very unevenly distributed through the sample. This excessive 
local drying also caused the formation of aggregates in a manner that could not 
be duplicated. 
Fig. 4.—Showing the effect of the amount of moisture on the breaking strength of the briquette 
In the third method, 100 gm. of soil was suspended in three liters of water, 
with every precaution*>eing taken to get the maximum amount of deflocculation. 
The greater part of the excess water was drawn off through Pasteur-Chamberlain 
filters and the material dried down as described for the second method. This 
method consumed a large amount of time and on the two samples tried did not 
give results very greatly different from the second method. No attempt was 
made to duplicate the results by this method. 
Table III shows the results obtained by the three different methods of treat¬ 
ment on two soils and by the first and second method on five soils. Since the 
first method could be duplicated with greater reliability in addition to requiring 
a shorter time to carry out, it was adopted as a standard. 
