498 Journal of Agricultural Research voi. xir, no. s 



carbonate in increasing the number of soil bacteria. This variation can 

 possibly be explained by the difference in solubility of these compounds. 

 It was shown (30) that magnesium carbonate is more soluble in carbo- 

 nated water than is calcium carbonate, and that calcium carbonate is 

 more soluble than limestone. Dolomitic limestone, however, is less 

 soluble than the nondolomitic. This order of solubility is in agreement 

 with the order in which the bacteria of the soil responded to treatment 

 with these different compounds. 



It does not seem correct to say that the entire influence which these 

 compounds had on the soil bacteria was due to the neutralization of the 

 soil acids. When neutral soil was treated with magnesium carbonate 

 or with calcium carbonate, the number of bacteria was increased, espe- 

 cially in the case of magnesium carbonate. Both in neutral and in acid 

 soil, the latter doubly neutralized, magnesium carbonate increased the 

 number of Bacillus azotobacter. All this would indicate that carbonates 

 either serve in part as a stimulant or effect an indirect action on other 

 compounds which are in turn rendered more soluble. 



Since the magnesium and calcium carbonates increased the number 

 of bacteria in acid soil when added in small amounts, and since appli- 

 cation in acid soil of these compounds gave better results than when 

 applied to neutral soil, it appears that the greater part of this influence 

 on the bacteria was due to neutralization. 



It may have been that the magnesium carbonate when added to the 

 soil was partially converted into magnesium phosphate. Truog (36) 

 pointed out the fact that for the amount of phosphate used, magnesium 

 phosphate increased the phosphorus content of plants more than did 

 other forms of phosphates. If this be the case, and the formation of 

 magnesium phosphate takes place in soil when magnesium carbonate is 

 applied, then magnesium chlorid and magnesium phosphate should give 

 an increase in the number of bacteria. The results from the magnesium- 

 chlorid treatment did not prove beneficial, however, even in so small 

 amounts; the chlorin radical may have been toxic. Magnesium phos- 

 phate did not prove favorable to the reproduction of bacteria in acid soil 

 but did in neutral soil. In the latter soil, the effect of magnesium phos- 

 phate was more beneficial than the effect of calcium phosphate. From 

 this evidence it appears that magnesium phosphate in a soil favorable 

 for the development of bacteria is a stimulant to the growth of bacteria. 

 This action of magnesium phosphate on bacteria in soil may in part 

 account for the enormous influence which magnesium carbonate exerts 

 in increasing the number of bacteria in both neutral and acid soils. 



From all the data obtained in the various experiments performed, 

 magnesium carbonate appears to play an important part in the devel- 

 opment of soil bacteria, much more so than does calcium carbonate. 

 Magnesium phosphate, when applied to neutral soil, caused an increase 



