34 The Influence of Potslx nJ>^ on Nitrification 



The jars were kept almost wholly embedded in the earth and the soils 

 therein were practically exposed to the same temperature conditions as 

 the soil in the field. Arrangements were provided to prevent the access 

 of rain water into the bottles which collected the percolates. At first, the 

 jars were periodically treated with measured volumes of water. Later on, 

 this method of artificial irrigation was discontinued. Rain was allowed 

 to fall on the jars, and the resulting natural 2)ercolates examined. 



The jars were filled up on the 7th May. They were irrigated once with 

 well water on the 15th May and the resulting percolates analysed. The 

 following results were obtained. 



Table I. 

 Percolates from the Jars, I5th May, 1916. 



nitrates per million. 



The influence of potsherds was manifest in two ways. Comparing jars 

 filled with the same depth of soil mixture it was seen that in every case 

 the concentration, as well as the total amount of nitrates, was greater 

 in the case of percolates from the jars which contained thirty per cent, 

 of potsherds in the soil. The figures about the dissolved oxygen were also 

 interesting. The presence of potsherds accelerated the oxidation of 

 organic matter. The percolates from jars containing no added potsherds 

 were coloured and low in dissolved ox3-gen. That these were richer in 

 dissolved organic matter as compared with the clear percolates from 

 other jars was confirmed by determining the amounts of potassium per- 



