H. A. Tempanv 823 



If the above figures are plotted it will be seen that the departure 

 from the linear relationship is not marked. 



When we go on to consider the relation between the actual weight 

 of water lost and the corresponding cubical contraction, we are met 

 with the difficulty that in the results adduced, some variation will be 

 experienced owing to the fact that the density of the wet bricks will 

 not be an absolutely constant quantity, owing to variations in the 

 density of the constituent soil particles and in the content of water. 



Actual measurements indicate that in the case of the soils examined 

 the density of the wet blocks ranged between 1-95 and 2-1 ; if we 

 assume as an approximation a mean density of 2-0 for the wet block 

 and compare the actual weight of water lost per unit of cubical con- 

 traction for a block which when wet measures 100 c.c, in the case of 

 Example A, over the range of values just examined, we arrive at the 

 following data : 



Taking into account the fact that in calculating the cubical contrac- 

 tion any errors inherent in the linear measurement are multiplied nearly 

 threefold, it appears reasonable to assume that, under the conditions of 

 the experiment the cubical contraction is equal to the volume of the 

 water lost by evaporation (when contraction is not influenced by internal 

 friction among soil particles), and that in the case of a pure colloidal 

 clay this linear relationship may be expected to hold good until shrink- 

 age ceases altogether. 



It follows from the foregoing that the relationship between linear 

 contraction and water loss is not strictly linear in character and as 

 the efEect observed grows in magnitude the departure from the straight 

 line relation w\\\ become more and more marked. 



Conclusions. 



1. In soils which have been moistened to the point of maximum 

 plasticity the whole of the water probably exists in union with the 

 colloidal material present in the soil in the form of a gel. 



2. The gel occupies the whole of the interstitial spaces of the soil 

 and as it loses water by evaporation draws together the soil particles 

 in its meshes. 



