li TIk a f Unisphere nf t/ie Soil 



air is commonly not more than 10 to 20 per cent, of the volume of the 

 soil. The pores appear to be continuous and seem to maintain fairly 

 complete communication between the various layers of the soil ; in 

 some places the communication is made more effective by the presence 

 of cracks and burrows. 



The soil atmosphere is the air present in these pores. Its biological 

 si<;nificance lies in the fact that it is the air surrounding the soil 

 organism.s and the roots of plants, and is either in actual contact with 

 them or is separated from them only by a thin film of water or colloidal 

 matter. It is obviously part of the ordinary atmosphere but its com- 

 position is influenced by two causes: oxygen is absorbed and carbon 

 dioxide produced by the inhabitants of the soil; while on the other 

 hand, diffusion and other processes of gaseous intercliange are constantly 

 replacing it with ordinary atmospheric air, thus eliminating any 

 differences in composition brouglit about by biochemical or other 

 changes. As a net result the composition of the soil air at any moment 

 is determined by the difference of velocity with which these two 

 processes take place. 



Unfortunately the mechanism of gaseous interchange in the soil is 

 not sufificiently well known to enable us to ascertain the speed at which 

 it goes on and so to discover the rate of production of carbon dioxide, 

 a quantity of great importance in the study of the biochemical changes 

 in the soil, but we have obtained evidence that our curves are mainly 

 determined by the production and not by the loss of carbon dioxide 

 from the soil. In any case for our present purpose of discovering the 

 conditions under which life goes on in the soil it is mainly necessary 

 to know the resultant of the various actions concerned. 



Preliminary determinations showed that it is not difficult to draw 

 a sample of gas from the soil, that is fairly representative of the soil 

 air and is uncontaminated by atmospheric air. In our experiments 

 the depth selected has been 6 inches, this being right in the region 

 where the soil changes take place, besides being convenient for working. 

 But as a matter of fact no great difference in composition was found 

 on going somewhat deeper: thus the following results (Table I) were 

 obtained at 6 and 18 inches respectively. 



In general the soil air was found to be very similar in composition 

 to ordinary atmospheric air, especially as regards the percentages of 

 oxygen and of nitrogen. It commonly contains less oxygen and more 

 carbon dioxide, usually also more nitrogen, Init the differences are often 

 small and only detected with certainty by careful analyses (Table VI). 



