H. A. Tempany 329 



A further approximate check on the vahdity of these deductions 

 is capable of being effected if the fine silt and clay fractions of soils of 

 known shrinkage are isolated, and separate determinations of the 

 shrinkage of the separated fine silt and clay fractions carried out, since 

 if the assumptions are correct and the approximate proportionahty 

 existing between the content of colloidal clay in the soil itself together 

 with the content of fine silt and clay is known then, since the bulk of the 

 colloidal material is separated in the fine silt and clay fractions, by 

 the employment of the factor in question it becomes possible to calculate 

 a theoretical value for the shrinkage which should be exhibited by the 

 separated fine silt and clay fractions and to compare the figure so 

 obtained with the result obtained by actual measurement. 



This has been done in the case of Examples A and B and the results 

 are given below: 



Actual shrinkage observed 

 in the case of the fine silt 



and clay fractions Calculated figure 



A. 20-0 per cent. 21-8 per cent. 



B. 18-2 „ 19-8 „ 



The agreement between the calculated and the theoretical values 

 in both these cases is sufficiently close to afford important confirmation 

 of the conclusion already arrived at especially when it is borne in mind 

 that the separation of the colloidal clay in the fine silt and clay 

 fractions is not complete inasmuch as some will remain in the silt 

 fraction. 



The deductions arrived at in the foregoing pages indicate that by 

 the employment of a factor of the value indicated it is possible to 

 calculate the approximate content of colloidal clay in soils from a 

 knowledge of the shrinkage. Such calculations are of course only 

 approximate but at the same time they would appear to be capable 

 of affording information of considerable value to the soil analyst apart 

 from data obtained by the ordinary methods of physical analysis which 

 as has been shown do not afford a complete insight into the character of 

 the soil, inasmuch as they give no expression for the amount of colloidal 

 material therein existing. 



It is hoped that in a subsequent paper it may be possible to trace 

 the relationship exhibited between the values for the content of colloidal 

 clay calculated from the shrinkage and data derived from a study of 

 the adsorption coefficient of different soil types. 



