324 



SOILS. 



may be brought to bear, must essentially influence the ade- 

 quacy of the plant-food thus supplied. Moreover, the greater 

 or less extent to which these sources may have been drawn 



J 



upon previously in the course of cultivation, will similarly in- 

 fluence that adequacy, on account of the diminution of the 

 readily available supply. 



Water-soluble and Acid-soluble Portions most Important. 

 It thus seems that while the undecomposed rock minerals are 

 indicative of the nature of the soil, but not directlv concerned 



J 



in plant nutrition, the most direct interest attaches to the water- 

 soluble portion, and the acid-soluble reserve. Both of these 

 can, of course, be withdrawn from the soil by treatment with 

 acids of greater or less strength ; and it would seem that if 

 we knew just what is the kind and strength of the acid solvent 

 employed by each plant, we could so imitate their action as 

 to determine definitely whether or not the soil contains an 

 adequate or deficient supply of actually available food for the 

 coming crop. 



We Cannot Imitate Plant-root Action. In this, however, 

 we encounter serious difficulties. The acids secreted by the 

 plant roots are not the only solvents active in the dissolution of 

 plant-food ; as yet we know the nature of only a few ; and even 

 these, instead of acting for a long time (season) on a relatively 

 small number of soil particles touched by the root-hairs, can 

 in our laboratories only be allowed to act for a short time on 

 the entire soil-mass. Clearly, the results thus obtained can- 

 not be a direct measure of the amount of plant-food which a 

 plant may take up in a given time; we can only gain com- 

 parative figures. These, however, can be utilized by com- 

 parison with actual cultural experience obtained in similar 

 cases. 



Cultural experience must, of course, be the final test in all 

 these questions; and it is generally more fruitful to investi- 

 gate the causes underlying such actual practical experience, 

 than to attempt to supply, artificially, the supposed conditions 

 of plant growth. The latter are so complex and so difficult 

 of control, that the results obtained by synthetic, small-scale 

 experiments are constantly liable to the suspicion that they 



