THE FERTILITY OF THE SOIL 15 



working in the Rothamsted Laboratory. In the first 

 place they found that the soil which had been put 

 through the treatment was chemically characterized by 

 an exceptional accumulation of ammonia, to an extent 

 that would account for the increased fertility of the soil . 

 At the same time it was found that the treatment did not 

 effect complete sterilization of the soil, though it caused 

 at the outset a great reduction in the numbers of bacteria 

 present. This reduction was only temporary, for as soon 

 as the soil was watered and left to itself, the bacteria 

 increased to a degree that is never attained in the soil 

 under normal conditions. For example, one of the 

 Rothamsted soils employed contains normally about 

 seven million bacteria per gramme, a number which 

 remains comparatively constant under ordinary condi- 

 tions. Heating reduced the numbers to 400 per gramme, 

 but four days later they had risen to six million, after 

 which they increased to over forty million per gramme. 

 When the soil was treated with toluene, a similar varia- 

 tion in the number of bacteria was observed. The 

 accumulation of ammonia in the treated soils was 

 accounted for by this increase in the number of 

 bacteria, because the two processes went on at about 

 the same rate. Some rearrangement was effected also 

 in the nature of the bacterial flora ; for example, the 

 group causing nitrification was eliminated, though no 

 substantial change was effected in the distribution of the 

 other types. The bacteria which remained were chiefly 

 of the class which split up organic nitrogen compounds 

 into ammonia, and as the nitrate-making organisms, 

 which normally transform ammonia in the soil as fast 

 as it is produced, had been killed off by the treatment, 

 it was possible for the ammonia to accumulate. The 



