20 SEASONAL NITRIFICATION AS INFLUENCED BY CROPS, ETC. 



mean temperature of the air did not reach nearly its maximum until 

 some time later than June 19; in fact, the maximum temperature 

 occurred in September. 



In figure 6, which gives the corresponding data, obtained on the 

 wheat plat, there is a closer relation between the moisture and the 

 nitrate curves, but as the growing crop has removed much of the 

 moisture and the nitrates from the soil it is apparent that no close 

 causal relationship could be insisted on in this case either. There 

 is, on the other hand, a close agreement between the moisture curve 

 and the weekly crop-increase curve. The hygroscopic coefficient for 

 this soil is very high (about 12 per cent), and judging from the gen- 

 eral appearance of the crop and of the soil at the time it is likely 

 that the moisture curve from the first part of July represents about 

 the minimum to which the wheat plants could reduce the moisture 

 content of the soil, and the same holds true in regard to the nitrate 



Fig. 6. — Diagram showing the average parts per million of water-soluble nitrates in dry soil to a depth of 

 2 feet in the spring-wheat plat; also soil-moisture, precipitation, and weekly crop-increase curves. 



curve, as mentioned earlier. This same removal of moisture would, 

 of course, also check nitrification, so both this and the plant feeding 

 would keep the nitrates from increasing. 



Figure 7 brings out more clearly than figures 1 and 2 the phenome- 

 non of the general parallelism in the maximum nitrate accumulation 

 in the individual soil layers in the wheat and fallow plats. This 

 figure does not show the absolute amounts of nitrates found in the 

 soil, but shows simply the period during which the respective 6-inch 

 soil layers in the fallow and wheat plats contained an amount of 

 nitrates greater than any other 6-inch layer. It will be noticed that 

 there is a very close agreement between the two plats in the depth 



173 



