324 Journal of Agricultural Research voi. xv, no.6 



soils to slight variations in moisture. The sample of fully saturated 

 soil C, taken at nine months, lost some moisture before it was determined, 

 in which condition it had 27.2 per cent of water and 400 pounds' acidity 

 by the potassium-nitrate method. Three months later a sample of soil 

 from the same pot had an acidity by the sam.e method of 2,150 pounds 

 with 30.6 per cent of water. The soil from this pot showed but a slight 

 trace of iron in the potassium-nitrate extract with 27.2 per cent of water 

 and a very large amount with 30.6 per cent of water. This increase in 

 acidity and of soluble iron appears to be due to the fully satuated con- 

 dition rather than the longer time elapsed. 



The relative acidities of the various soils, high or low in organic matter, 

 gave quite wide variations with the different methods. In general, the 

 potassium-nitrate method measures mineral acidity, owing to acid-re- 

 acting silicates, and to a less degree it measures acid organic matter in 

 the soil. The calcium-acetate method, on the other hand, measures 

 acidity due to acid-reacting silicates, and in addition it responds readily 

 to acid organic matter. With soils high in organic matter the results 

 due to this method are influenced by organic matter more than by acid 

 silicates. Water-soluble acidity only slightly affects the results shown 

 by either the potassium-nitrate or the calcium-acetate methods. The 

 results shown by the ethyl-acetate test are very largely in proportion 

 to the strength of the water-soluble acidity of the soil. These results 

 would be affected by nitrates, sulphates, or chlorids of aluminium, iron, 

 and to a slight degree by manganese salts; also by any soluble acid re- 

 acting organic matter. Pure silicates which show a very high acidity by 

 other methods and which are not soluble in water do not affect ethyl 

 acetate at all (7). 



In titrating the potassium-nitrate acidity determinations, quite dis- 

 tinct differences were noted in the character of the precipitates formed. 

 In order to study this point, determinations were made of the metals 

 which had been dissolved in the reactions between the normal potassium- 

 nitrate solution and the soil. Table III gives the bases and soluble silica 

 found in 100 cc. of potassium-nitrate extract from both the wet and the 

 air-dried soils ; also the increases or decreases of soluble matter found on 

 air drying the soil samples. These results show that considerable iron 

 was made soluble when the soil was kept fully saturated. This soluble 

 iron was apparently all in the ferrous state. After the soils were air-dry* 

 the iron was very quickly and almost completely oxidized, as the air-dry 

 soil showed but little soluble iron. This chemical change in the con- 

 dition of the iron undoubtedly accounts for a large part of the decrease in 

 acidity caused by drying the fully saturated samples. Soluble iron is 

 seldom found in soil solutions in very large amounts, which is undoubtedly 

 due to the fact that the usual procedure in preparing soil samples for 

 analysis is first to air-dry them, allowing ample opportunity for the 

 oxidation of the iron. 



