ANALYSIS OF HEAVY MINERAL SOIL. 225 



scopic examination of the fractions showed the presence of par- 

 ticles of magnetite smaller than the smallest quartz grains. It 

 has been noted by Stadler* in advocating grading analyses by 

 elutriation of mill products that an error in such analyses is 

 occasioned by the presence in Rand ore of a small percentage of 

 pyrites. 



In Table I the mechanical analyses of some samples of the 

 magnetite soil are given, together with certain other data. Atten- 

 tion is directed chiefly to the differences between the percentages 

 by weight and by volume of fine gravel and sand taken together 

 in the case of those samples containing much magnetite. A method 

 for reducing these soils to a common basis was sought without 

 much success. The chief reason for this non-success lay in the 

 practical impossibility of effecting a rapid and reasonably accurate 

 separation of the soil into heavy and normal particles. 



The percentages by weight of particles greater than 0-2 mm. 

 may be corrected to agree fairly well with the percentages by 

 volume by calculating the quartz equivalent of the magnetite par- 

 ticles, taking the density of the latter as 4-6, or, by calculating 

 the quartz equivalent of the particles separated by Thoulet's 

 solution of density 3, taking the density of such particles as 3-5. 

 Such corrections are shown in Tablo II. 



The whole question is complicated by the very nature of the 

 heavy mineral present. The products of the weathering of the 

 magnetite, hydrated ferric oxide, have a very marked flocculating 

 effect on clay, thus introducing another texture-modifying factor. 

 Had the heavy mineral been one which could have influenced 

 texture in only one way, say, zircon, an attempt might have been 

 made at exact mechanical analyses by first effecting the separation 

 in the usual way, then separating each fraction into two parts by 

 means of a heavy solution and finally putting the heavy portions 

 through the sedimentation process suitably modified so as to obtain 

 fractions between the same size limits as those obtaining for normal 

 soils. Such an analysis would be terribly tedious and would be 

 of doubtful value as it is impossible to assign an exact quantitative 

 value to the degree in which any particular mechanical fraction 

 influences the soil texture. It is essential, however, in dealing 

 with soils which contain abnormal quantities of heavy minerals, 

 that the proportion of these be taken into consideration when 

 basing a judgment on the results of the mechanical analyses, in 

 much the same way as the percentages of calcium carbonate and 

 of organic matter are used to modify the picture drawn by the 

 mechanical analyses. In this connection it may be noted, that soil 

 scientists in the United States frequently omit all mention of such 

 modifying factors as calcium carbonate and organic matter, and 

 that their example has been followed in this country. 



•Stadler, H. (1913), " Grading "Analysis by Elntriation." Transact.. Inst- 

 Mining and Metall. XXII, p. 686. 



