BACTERIA IN THE INDUSTRIES 1 69 



believed that the soil humus and soil colloids are one and the same, but 

 this is not in accord with the facts, as may be gathered from the above 

 introduction. Humus is in part colloidal and all humus has colloidal 

 properties. The same may be said of the mineral constituents of the 

 soil no matter what their composition. The agriculturally active soil 

 colloids are represented by a long Ust of organic and inorganic compounds 

 (in 'very minute, mostly microscopic and ultra-microscopic molecular 

 aggregates) combined with more or less water, the water being the chief 

 dispersing medium for the colloidal particles. These soil colloids may be 

 roughly grouped as follows: 



1. SiUcic acid and the soluble siUcates. 



2. Aluminum hydroxid and its compounds with silicic acid, represented 

 by the clays. 



3. Iron hydroxid and its compounds. 



4. Humus. Represented by humus acids, organic matter generally, 

 including the soil bacteria and other microorganisms of the soil. 



A fairly good idea as to the nature of soil colloids may be obtained in 

 the following manner. Place a teaspoonful of rich soil in a tumblerful 

 of water and stir well, let stand for a few minutes and pour the superna- 

 tant murky liquid into a second tumbler and let this stand for thirty min- 

 utes 'and again pour the supernatant liquid into a third tumbler and let 

 stand for a few hours, decant very carefully; let stand for 24 hours and 

 again decant. You will now have five grades of soil particles. In tumblers 

 I and 2 you will find the coarse gravel and coarser sand particles, with a 

 certain amount of very fine matter. Tumblers 3 and 4 will contain the finer 

 sand particles and most of the precipitable colloids, while tumbler five will 

 retain the very finest sand particles and the majority of colloids in 

 solution and in suspension. The contents of tumbler 5 will appear quite 

 murky and will not clear itself entirely even if allowed to stand undis- 

 turbed for days and weeks. If this liquid should be entirely clarified by 

 filtering repeatedly through a fine filter paper, it would nevertheless 

 contain much colloidal matter, organic as well as inorganic, as might be 

 demonstrated by dialysis, chemical precipitation, etc. The enormous river 

 deltas (of the Nile and the Mississippi) are colloidal precipitates the result 

 of contact of the salt water with the fresh river waters carrying the colloids 

 brought down from the rain washed mountain sides and the valley lands. 



It has been stated that the fertility of the soils is directly proportional 

 to the percentage of colloids present. This is certainly not true. An 

 excess of colloids is as harmful as is a deficiency of colloids. It is therefore 

 correct to state that the fertility of the soils is proportional to the increase 

 of coUoids up to the optimum, not the maximum. It has been known for 

 centuries that the winter freezing of very rich soils results in increased 



