SOIL REACTION 425 



A, allows of direct comparison of its Ph with that of a standard 

 solution, B. At certain points where the ranges of two or 

 more indicators overlap, the results can be checked by using 

 more than one indicator for the determination of the same Ph- 

 The method is not absolute ; it does not really measure the Ph 

 of a solution, but only shows that this function is identical with 

 that of a particular standard. Ultimately the Pnof the standards 

 must be determined by the more fundamental electrometric 

 method. If the standard solutions are strongly buffered 

 by the presence of reaction regulators they will maintain their 

 Ph unchanged for considerable periods, since small quantities 

 of impurities from the air, glass, or slight mould growth, etc., 

 have but little effect. Further, the standard buffer solutions 

 are generally easily prepared, and, once made up and their Ph 

 measured electrometrically, they can easily be renewed without 

 making fresh electrometric determinations. The method has 

 been applied with much success to biological fluids by Sorensen, 

 Palitzsch, Walpole, Clark and Lubs and others, and their 

 papers must be consulted for details of the method. Gillespie 

 and his co-workers in America applied both the electrometric 

 method and the colorimetric method to soil with excellent 

 agreement, in spite of the fact that they employed aqueous soil 

 extracts in the latter method, but mixtures of soil and water 

 in the former, soil extracts alone being unsuitable for use with 

 the electrometric method, as they are poor in buffer action. 

 Further, the NOg-ions were reduced to NHg by the hydrogen 

 of the electrode, thus rendering a constant potential impossible 

 of attainment, and in some cases even changing the reaction 

 of the whole fluid to indicators. This is the only systematic 

 comparison so far made between the two methods, as applied 

 to soils, and it deserves repetition on account of its fundamental 

 importance. The method certainly seems to present difficulties 

 in some cases and preliminary work by the writer ^ indicated that 

 the conditions of its application in soil investigation have not 

 yet been completely worked out. When the method has been 

 properly standardised it should, however, prove of the greatest 

 usefulness in soil investigation, and it is along such lines as 

 those indicated that soil reaction — and related problems, such 

 as lime requirements and the relation of hydrogen-ion concen- 

 tration of soils to plant distribution — can best be studied. 



* Journ. Agric. Sci., xi (1921), p. 45, 



28 



