EXPERIMENT STATION BULLETINS. 419 



were single salts or a mixture of different salts and acids. There were 

 evidences, however, of the formation of a series of salts in the alkali 

 titrations, analogous to the salt series formed by titrating H3PO4, with 

 KOH, although as previously stated, these salts might have belonged 

 to different series. A large portion of the active materials of soils must 

 be considered insoluble in water, but the solubility of these components 

 is sufficient to determine the reaction of a soil by means of its water 

 solution ; but the pH of a soil, or degree of acidity or alkalinity, depends 

 on the relative quantities of hydroge nions and hydroxyl ions in solution 

 which is controlled by the proportion of active bases to active acids, 

 and not by the total quantities of these components present. 



Calling attention again to Table 3, all the soils with pH above 7.00 

 contained carbonates, but no relationship was discovered between con- 

 tent of bases and pH in the four soils studied. Evidently concentra- 

 tion of hydroxyl ions in these soils depends on factors other than soil 

 class and quantity of carbonates. In this connection, it is believed that 

 CO, plays an important role. These pH determinations were made on 

 air dry soils and a longer time was required for equilibrium than in 

 case of the acid soils. 



The pH of acid soils was found to increase in magnitude on stand- 

 ing in contact with water. These results show that when the pH of 

 field soils is desired it is imperative that the determinations be made on 

 fresh, moist field samples. Results in Table 3 also show that acid soils 

 may have a greater neutralizing value for acids than carbonate soils; 

 and that the major cause for soils becoming acid is the loss of soil 

 bases. While the presence of carbonates may not be necessary for a 

 neutral or alkaline soil, they maintain a balance between silicate-bases 

 and acids when finely divided and well distributed through the soil 

 mass. 



The reason for using ALCle in this investigation was that it was 

 assumed that this reagent would hold back the dissociation of soil iron 

 and aluminum and react only with the alkali bases. That this effect 

 might be accomplished was shown by the fact that only slight traces 

 of iron was found in soil solutions after treatment with AloClg, by 

 qualitative tests, but because the hydroxide formed coated carbonate 

 and soil particles and tended to prevent further reaction, its use was 

 abandoned. As shown by the results, however, the effects produced by 

 added hydrogen ions in HCl were similar to those obtained by using 

 ALClo- There is a point here seemingly of practical importance, that 

 is, injury may result to plant growth if soils are treated in such a man- 

 ner as to precipitate hydroxides of iron or aluminum in them. 



The end point in the soil — Ca(0H)2 reaction is considered to be 

 slightly above pH 7.00. This being the case it may be expected that 

 the end point in a reaction between soils and stronger alkalis will be 

 still higher on the pH scale. Since Hoagland (4) has shown that 

 hydroxyl ion concentrations greater than pH 8.25 were distinctly in- 

 jurious to barley seedlings, and that hydroxyl ions were more toxic than 

 hydrogen ions for equal divergences from pH 7.00, alkalis might be in- 

 jurious when added to soils under conditions where the pH would be 

 raised above 8.25. The injurious effects of watering green house plants 

 with large quantities of hard waters are quite well known and may pos- 

 sibly be explained on this basis. 



