144 Journal of Agricultural Research Voi.xii. No. 3 



partial pressure of carbon dioxid in the manner suggested in this paper 

 it should be entirely practicable to obtain a measurement of the H-ion 

 concentration identical with that of the soil in the field. Russell and 

 Appleyard {10) have found that the carbon-dioxid content of the soil 

 atmosphere under different conditions varied between 0.02 and 2 per cent, 

 the general mean of arable soils being 0.25 per cent. If these percentages 

 of carbon dioxid are to be regarded as typical for field soils, then in view 

 of the present experiments the changes in carbon-dioxid content during 

 sampling and laboratory manipulation would not invalidate our inferences 

 with regard to the reaction of soils under natural conditions. 



The a priori considerations already presented in the first portion of this 

 paper with respect to the effect of carbon dioxid on the reaction of soils 

 are entirely substantiated by the experimental data, which lead to con- 

 clusions at variance with those of Maclntire {8) on this point. Even 

 saturating the soil suspension with carbon dioxid previous to measuring 

 the reaction did not decrease the H-ion concentration. Therefore it 

 follows that an acid soil would never present to the plant a soil solution 

 of alkaline reaction, notwithstanding any increase in the partial pressure 

 of carbon dioxid. It should be recalled that the criteria heretofore 

 used for judging the reaction of soils do not always permit of an accurate 

 distinction between soils of different reactions. Some soils may be 

 judged as acid from the standpoint of certain lime requirement methods, 

 when in reality their reaction may be alkaline. For this reason it may 

 be doubted whether some of the soils reported as yielding alkaline 

 extracts were in fact acid. Although in certain instances the application 

 of lime may be followed by an increased crop yield, this result may not 

 be dependent upon any change in the reaction of the soil due to the addi- 

 tion of lime. The more accurate interpretation of liming experiments 

 demands that an attempt be made to differentiate lime as a neutralizer 

 of acidity and the other directly or indirectly beneficial effects of lime 

 on the soil or the plant. 



The statement frequently met in the literature that the extracts from 

 soils considered acid have a neutral or alkaline reaction has led to the 

 conclusion that water-soluble acids are not found in acid soils. We 

 desire to emphasize again the point that extracts of acid soils, especially 

 those prepared with carbonated water, might become neutral or alkaline 

 after the loss of carbon dioxid from the extract. Moreover the reported 

 results on the extracts are likely to be misinterpreted, for they do not 

 take into account the H-ion concentration but are based upon titrations, 

 using indicators whose end points may be removed from neutrality by 

 several magnitudes. Indeed Gillespie {3) has found that soils determined 

 to be acid by the hydrogen electrode, yield extracts whose reactions, 

 colorimetrically estimated, were in close agreement with the hydrogen- 

 electrode measurements. 



