ANALYSIS OF HEAVY MINERAL SOIL. 223 



In conclusion, I wish to express my sincere thanks to Messrs. 

 Jones and Attwood, Ltd., for the privilege of visiting all their 

 works and inspecting their experimental tanks while experiments 

 were proceeding, and also to Dr. Ardern, Chief Chemist, Rivers 

 Department, Manchester, who always placed all information at 

 my disposal. I hope I have given sufficient evidence in this paper 

 without postulating to prove the great possibilities of the Activated 

 Sludge Process of sewage purification, and that some town in 

 South Africa will soon have a plant installed and reap the benefit 

 of the experiments carried out in England. I think the process 

 is particularly adapted to large inland towns in tropical countries. 

 As I have stated, the cost of drying the sludge is a very big item 

 in England, and can very easily be done out here for at least 

 half the cost, which is a very great consideration, and thereby 

 convert your sewage scheme into a very profitable concern. 



ON THE MECHANICAL ANALYSIS OF SOIL CONTAINING 

 HEAVY MINERALS. 



BY 



B. de C. Marchand, B.A., D.Sc, 

 Division of Chemistry, Union Depart meat of Agriculture. 



Read July 13. 1921. 



The mechanical analysis of soils is chiefly of value as a basis 

 for the correlation of soils of known properties with others regard- 

 ing which information is sought. It is obviously essential that 

 the basis of comparison should be common to the soils compared. 

 For this reason the comparison of soils differing widely as regard 5 

 their content of such constituents, as calcium carbonate or organic 

 matter, which, apart from size of particles, modify the texture, 

 cannot be made unless the relative quantities of such modifying 

 constituents be taken into account. When the proportion of such 

 modifying constituents is large mechanical analysis is almost worth- 

 less as a basis of comparison. 



Now in the mechanical analysis of mineral soils, that is of 

 soils to which the method may safely be applied, an important 

 fundamental assumption is made; t^at is, that the relation 

 between the weight and the volume of the soil particles is constant 

 for all particles of whatever size and for all soils. In the great, 

 majority of cases this assumption is justified. We have, however, 

 recently examined a series of soil samples, all belonging to the 

 same type, which differ considerably in density from the average 

 for normal soils and also, to some extent, from one another. 



In the case of the coarser particles, which are usually separ- 

 ated by means of sieves, the limiting size is determined by the 



