August 25, 1922] 



SCIENCE 



217 



silica, alumina and iron is a colloidal solution 

 as distinguished from the true solution of the 

 so-called soluble elements. 



The methods devised for separating colloids 

 from soils for analysis and study, and the 

 methods for the detemiination of the amount 

 of colloids in soils have apparently opened up 

 a new and vast field of research, the results of 

 which will undoubtedly throw a broader light 

 upon the subject of chemical denudation. 



In my paper on "The Origin of Soil Colloids 

 and Eeason for the Existence of This State of 

 Matter," published in Science, December 30, 

 1921, I called attention to the fact that in the 

 mechanical analysis of soils under microscop- 

 ical control practically the smallest particle of 

 matter is .0001 mm. I advanced the opinion 

 that complex silicates of such small diameters 

 contain relatively so few molecules that when 

 brought into contact with water the bombard- 

 ment of the water molecules shatters the min- 

 eral particles, the electrolytes going into true 

 solution and the silica, alumina and iron going 

 into colloidal solution, the colloids thus formed 

 reabsorbing much or all of the electrolytes, 

 which puts these electrolytes themselves in 

 large part into a colloidal state. 



It is difficult to conceive how the electrolytes 

 can be dissolved out of silicates at ordinary 

 temperatures and pressures without conceiving 

 that the whole thing breaks down into a dis- 

 solved mass. In fact, direct evidence of this 

 is to be found in the dry grinding of silicates 

 in the ball mill. If this fine ground material 

 is brought into contact with water, soluble salts 

 go into solution as determined by conductivity 

 or chemical test, and at the same time there is 

 released a relatively large amount of colloidal 

 material. What the exact ratio is between the 

 potash, for example, which has been released 

 from the silicate rock, and the silica, alumina 

 and iron in the resulting colloids formed has 

 not been deteiTnined. 



Soil colloids in suspension of a gram to the 

 liter give the appearance under the ultra micro- 

 scope of minute droplets immiscible with the 

 surrounding water capable of indefinite sus- 

 pension. In larger concentrations the droplets 

 appear to coalesce, forming a jelly-like mass 

 which settles to the bottom of the vessel. These 



soil colloids have a distribution equilibrium, 

 however, between the colloid and the surround- 

 ing water, involving silica, potash and other 

 electrolytes to a notable extent, but affecting 

 the iron and alumina only to an inconsiderable 

 extent. 



We have a general idea of this distribution 

 equivalent in the case of analyses which have 

 been made of the truly soluble material thait 

 passes through the Pasteur-Chamberlain filter 

 in comparison with the colloidal material which 

 remains on the outside of the filter tubes. The 

 colloidal material of the Miami silty clay loam 

 soil collected on the outside of the tubes on a 

 strictly water-free basis carries 54.15 per cent. 

 SiOj, 28.03 per cent. A^Og and 13.39 per cent. 

 Fe^Og. These together add up 95.57 per cent, 

 of colloidal material. The content of KjO is 

 2.89 per cent., giving a ratio of potash to silica, 

 alumina and iron of 1 : 33. In ithe dissolved 

 salts which have passed through the filter tube 

 there is 10.4 per cent. SiOj, 2.9 per cent. Al^Oj 

 and Te^Og, making 13.3 per cent., while the 

 K,0 content is 8.1 per cent., giving a ratio of 

 1 : 1.6. The salts of the river waters of the 

 United States contain on the average 8.60 per 

 cent, of SiOg, .64 per cent, of Al^Oj and FCjOg, 

 making a total of 9.24 per cent., while the Kj*-^ 

 content is 2.13 per cent., giving a ratio of 

 1 : 4.4. 



The average analysis of igneous rocks as 

 given by Washington on a strictly water-free 

 basis shows 60.14 per cent, of SiO^, 16.32 per 

 cent, of Al^Og and 7.84 per cent, of Fe^Og and 

 FeO calculated to Fe^O,. This makes 84.30 

 per cent., while the K,0 content is 3.27 per 

 cent. This gives a ratio of potash to silica, 

 alumina and iron of 1 : 25.8, as compared 

 with the ratio of 1 : 33 in the case of the 

 Miami soil colloid as above. The composition 

 and ratio of potash to silica, alumina and iron 

 in shale rocks is nearly identical with the 

 igneous rocks. 



The average composition of deep marine 

 sediments ("red clay") on a water-free basis, 

 excluding sea salts and shells, shows according 

 to Steiger 54.48 per cent. SiO,, 17.15 per cent. 

 AI2O3 and 10.36 per cent, of Fe^Og and FeO 

 calculated to Fe^Og. This makes a total of 

 81.99 per cent., while the K^O content is 3.07 



