E. J. EUSSELL AND A. APPLEYARD 



21 



The Causes of Fluctuations of Composition of Soil Air. 

 A. The variations due to season. 



These fluctuations consist in a rise to a maximum COj content in 

 late spring, a fall to a minimum in summer, a rise to a second maximum 

 in late autumn and a fall to a minimum in winter. The oxygen content 

 varies in the inverse sense, reaching minimum values in spring and 

 autumn and maximum values in summer and winter. 



All the curves show the same general shape when plotted over the 

 year; proving that the effect of season completely overrides the effect 

 of various soil treatments. Field experiments alone do not enable us 

 to disentangle all the factors, but we took measurements for the purpose 

 of discussing the effect of temperature and moisture content. 



Ejfect of femperntiire. This can be studied from Fig. 7 where the 

 mean soil temperatures taken from the continuous recording soil 

 thermometer are plotted along with the CO2 in the soil air from the 

 Broadbalk unmanured plot. 



0-8 



0-7 



»0-6 



I 0-5 



s 



^0-3 

 0-2 

 0-1 

 00 



I8°C 



IB 3 



a 



s 



14 I 



I 



12 E 



-2 

 -0°C 



Fig. 7. Curves showing percentage of CO., in air of Bioadbalk unmanured ijlot and 

 mean soU temperature (at 6" depth) for 24 liours previous to sampling. 



Beginning with the middle of April, 1913, when soil temperatures 

 were first taken, it is seen that the temperature curve runs closely 

 with the CO2 curve up to the early part of May, they then part 

 company and show no more resemblance till November. From that 

 time, however, up to early May, there is a close general resemblance 

 but this ceases from then onwards. Thus we can infer that the'tempera- 

 ture is the dominating factor in determining the amounts of CO.^ 

 production from November to May. 



