•)•) 



Tilt Afiiios/t/im of till' Siiil 



It is clearly not the only factor for the parallelism is not complete: 

 a rise in temperature in spring is more potent to increase the output 

 of CO2 than a similar rise later on. Thus the values for temperature 

 and CO2 in May and June no longer show the agreement obtained 

 earlier: the CO, maximum in May being above that in June while the 

 temperature maxima fall the other way. These differences in detail 

 indicate that other factors are operating, but they do not weaken the 

 main conclusion that from November to Maij (he temjierature determines 

 the rate of CO, product ion in the soil^. 



The dunged plots and the wilderness show the same general relation- 

 ships, but again there are differences in detail, the CO, and temperature 

 curves parting company earlier in the summer than on the unmanured 

 plot. The main obvious difference between the plots is that the crop 

 is larger on the dunged plot and the wilderness than on the unmanured 

 plot, and the bearing of this factor will become evident later on. 



From June to November, however, the temperature is not the main 

 factor for the curves show no kind of similarity. 



Effect of Moisture. A comparison of moisture content and CO2 

 content is made in Fig. 8. The moisture determinations only began 

 in June 1913, so that the curve does not run as long as that for tempera- 

 ture but it shows no connection with the COg curves except during a 

 few months in summer. The moisture is low during June, July and 

 August of 1913 when the CO, is falling: it rises in September and 

 October when the CO, first falls and then rises, it is steadily high from 

 November to March 1914 during which the CO^ first falls and then 

 rises ; it falls in April while the COj rises and falls low during summer 

 when the CO, also is low. 



Thus moisture does not have nearly so marked an effect as tempera- 

 ture, and it only shows any relationship to the COg during the summer 

 months July to September. 



The extreme case of water logged soil is dealt with ou p. 32. 



' The failure to find on some of the plots a maximum COo content in May 1914 of 

 tlio same order as the value obtained in 1013 may be attributed to the fact that quite 

 unwittingly wo allowed a favourable temperature period to pass without taking any 

 .samples. We made determinations on May 15 and again on May 25, but during the 

 interval there came a rise in temperature whicli we missed. 



May 1914 



15th 16th 17th 18th 19th . 20th 21st 22nd , 23rd 24th 



Soil tem])erature at i 15'5 

 depths of 6 in. °C. I 



1.51 161 



17-3 



16-9 



17-9 190 200 



161 14-6 



