38 Thr AtiiiDsplxyr of the Soil 



Unfortunately there was a break in the bacterial counts durinii the 

 winter months, but the other observations were made. In March, 191 -1, 

 there occurred a high rainfall, followed by a rise in CO., and somewhat 

 later by a rise in nitrate: in April the CO^ falls, but in May and June 

 there is a sharp increase in rainfall and in bacterial numbers, fdlhnvod 

 by an increase of CO2 and of nitrate. 



If we take the unmanured (Fio;. 1) instead of the dunired plot we 

 obtain similar but nuniprically smallci' losults. The wilderness (Fig. 3) 

 also shows the same general phenomena, but the spring rise in the nitrate 

 is considerably flattened down in consequence of the rapid absorption 

 by the plants; the autumn rise, however, is seen, and as before it 

 comes after the rise in COg and this in turn after the rise in bacteria. 

 Again, the Hoos wheat and fallow plots (Figs. 4 and 5) show like 

 similarity between bacterial numbers, CO2 and nitrates, especially 

 during the fallow period. The fluctuations are not great — the land 

 having received no manure for many years is very impoverished — and 

 it would be unsafe to attach too much importance to some of them, 

 but they all go in the same direction. During the time when the land 

 carries a crop of wheat (Fig. 5) the nitrate curve is flattened from 

 April to July; while on the other hand the loosening of tlie land during 

 the fallow period causes a flattening of the COj curve. 



The general conclusion is that the llucUiations in bacterial numbers, 

 in CO2 content and in nitrates in the soil are all of the same general 

 character, and this character is mainly impressed by seasonal factors; 

 other conditions such as manuring, cropping, etc., may pidl out or 

 flatten the curves but they do not alter their general shape. The 

 production both of nitrates and of COo attains a maximum in late 

 spring or early summer, a mininuim in summer, a maximum in late 

 autumn and a minimum in winter^ ; the numbers of bacteria fluctuate 

 in the same way in summer, autumn and winter. When the autumn 

 rains came after the dry summer conditions, the bacteria immediately 

 responded by rapid multiplication : then there came an increase in the 

 amount of CO2 in the soil air and finally an increase in the amount of 

 nitrates. This order seems to be pretty general. 



The spring and autumn periods of ma.ximuin biochemical activity 

 in the soil are clearly of great significance in soil managoment. 



' Similar seasonal tliictuations in nitratu content are rceordtti in the paper already 

 quoted in ./. Agric. Sri. lilll, 6, IH-.IT. 



