30 



The A tnioHjtlK'rc of the Soil 



difference in araouut and composition of the organic matter present in 

 the soil. The closest comparison we can set up is between two of the 

 Broadbalk plots: an arable plot receiving 14 tons of dung annually 

 and carrying each year a good crop of wheat, and an adjacent plot 

 known as the wilderness which has remained undisturbed since 1882 

 and now carries a dense growth of grasses, clovers, weeds, etc., only 

 young trees and bushes being removed. The percentages of COg in the 

 soil air are plotted in Fig. 12. There is no great difference betweea the 

 two curves. In April and early May the dunged plot contains more 

 ("0-2, from September to early January it contains less, but during 

 these months it has been ploughed up and left loosely exposed to the 

 atmosphere for a time prior to seeding. But the differences rarely 



Fob. Mar..\pr. Mjiy.IiinfJiily.\iij!.Si'pl.(>ct. Nov. Dec. Jan. Feb. Mar.Apr. May June jHWAuc.SpDt. 



Fig. 12. Curves showing COj in soil air of Broadbalk dunged and wilderness plots. 



exceed 0-3 per cent. When therefore the soil conditions are comparable 

 both as to the state of packing and to the amount of organic matter 

 the difference between grass and arable land is less than might be 

 expected. The result is all the more significant when it is remembered 

 that the air of the unmanured ])lot is as rich in CO^ during summer 

 as the air of the dunged plot. 



Taking them as a whole, these observations indicate that a growing 

 crop fer se has no very marked effect in increasing the amount of CO2 

 in the soil air. Comparison is rendered difficult by the numerous 

 differences between cropped and fallow land or between grass and 

 arable land, which can only partially be eliminated; if an ordinary 

 grass field is compared with an ordinary arable field considerable 

 diiTeiences are found, but when the conditions are made more nearlv 



