November 1, 1912] 



SCIENCE 



573 



dye is taken up by filter paper and is then 

 completely washed out again by water. 

 The rational formula which has been de- 

 veloped by Arrhenius for this process 

 merits mention here. 



Another process commonly called ad- 

 sorption involves chemical change. An ex- 

 ample is seen in the taking up of fuchsin 

 from a watery solution by carbon: the 

 fuchsin can not be washed off again by 

 water, but is readily removed by alcohol. 

 In this case the dye has been changed from 

 a substance soluble in water to one in- 

 soluble in water but soluble in alcohol, 

 which indicates the formation of an iso- 

 meric substance. 



In other cases we may obtain the char- 

 acteristic adsorption curve and yet be able 

 to explain the phenomenon as a chemical 

 reaction which involves no adsorption. 



Numerous reactions take place in the 

 soil which give the adsorption curve. The 

 most important adsorbing substances are 

 colloidal materials. Although colloids play 

 a very important role in the soil, the rap- 

 idly accumulating discoveries in the field 

 of colloidal phenomena have as yet found 

 little application in soil studies. The use 

 of the ultra microscope and of ultra filters 

 capable of sorting out various grades of 

 particles too small to be seen with the 

 microscope, as well as the methods for 

 measuring the rate of diffusion, the osmotic 

 pressure, the viscosity and electrical prop- 

 erties of colloids, should find a place in the 

 study of soil colloids. And we may expect 

 important results from the application of 

 the principles derived from the study of 

 gel formation, of hysteresis, of the anal- 

 ogues of the Danysz effect, of temperature 

 coefficients, of reaction velocity and similar 

 fundamental matters which have proved 

 fertile in the study of colloids. To speak 

 of these in detail would take too much 

 time. I will therefore mention only one. 



According to the Gibbs-Thompson law 

 substances which lower the surface tension 

 tend to become more concentrated in the 

 surface and reactions which produce such 

 substances are favored at the surface. In 

 the soil we deal almost entirely with sur- 

 faces, and it will be seen that the opera- 

 tion of this principle must promote such 

 processes as oxidation and reduction where 

 they tend to reduce surface tension and to 

 retard them in the opposite case. 



In the production of colloidal substances 

 the organic materials in the soil play an 

 important part. The great value of these 

 substances is shown by the fact that good 

 soil can be made without sand or clay but 

 not without humus: fresh volcanic ash, 

 consisting of finely ground minerals ca- 

 pable of furnishing plant food, must never- 

 theless await the admixture of humus be- 

 fore it can support crops. 



Organic substances may combine chem- 

 ically or mechanically with plant food and 

 thus prevent it from being leached out of 

 the soil. 



Certain organic colloids are known to 

 have a remarkable power of keeping diffi- 

 cultly soluble salts in solution at concen- 

 trations far above those which are possible 

 in pure water. This consideration may 

 prove important in soil studies. 



The study of the organic substances in 

 the soil is in a very backward state. It is 

 known that some are beneficial and others 

 are toxic and that in some cases the tox- 

 icity is partly removed by simple oxida- 

 tion. 



This leads us naturally to a considera- 

 tion of oxidation in the soil. It is a stri- 

 king fact that while some plants take up 

 unoxidized substances and gain the energy 

 they need by oxidizing them the majority 

 of plants must have their food in a highly 

 oxidized form in order to make the best 

 use of it. Thus CO, is a food while CO 



