ANTISEPTIC TREAMENT OF SOILS. 175 



the course of the present work. Possibly the definite 

 increases noticed by Pickering in the water soluble material 

 of treated soils consist mainly of increased silica, iron and 

 alumina rather than of fertilising constituents. 



Unsatisfactory as the results are, this work has served 

 the purpose of once more independently testing the various 

 methods of soil analysis as measures of fertility. 



The same soils were analysed by the hydrochloric, citric, 

 and N/5 nitric acid methods, and in none of the estimations 

 were excessive differences noticed between the treated 

 and untreated soils. 



Since the work of Russell and Hutchison has quite 

 established an increase in fertility due to such treatment, 

 it follows that neither the citric acid nor the N/5 nitric acid 

 method is a better guide to fertility than the hydrochloric 

 method. 



In other words while existing methods of soil analysis 

 give us some idea of the amount of mineral plant food in a 

 soil in different stages of solubility, they do not tell us 

 whether a soil is fertile or infertile, or what treatment 

 will make a soil more fertile. In fact, it is vain to expect 

 a soil analysis to indicate the state of present fertility, a 

 condition depending on meteorological, biological and 

 physical considerations to a much greater extent than on 

 chemical composition. 



In conclusion I wish to thank Mr. F. B. Guthrie for his 

 continual cooperation in this work which was undertaken 

 at his suggestion. Throughout the course of this research 

 I have been liberally assisted by Mr. Guthrie with advice 

 and references to literature, for which I desire to express 

 my cordial thanks. 



