10 Tlic A/lHOs/i/ltfr of tli( Soil 



Miintz demoiistiated tliat these were the active agents, the proof being 

 that, in presence of chloroform, soil produces only a fraction of the 

 amount of carbon dioxide formed in untreated soil. This was con- 

 firmed by Deherain and Demoussy^. From tha-t time it has been 

 generally recognised tliat the curljon dioxide is mainly produced by the 

 organisms of the soil. 



The application of the I'etteiikoiVi' iiiotlioci had thus carried the 

 j)roblem a long way, and had given considerable information about the 

 origin and fluctuations of the carbon dioxide in the soil air, but it gave 

 no information at all about the oxygen, and the idea gradually became 

 fixed that the soil atmosphere was deficient in oxygen, a view that was 

 strengthened by the well-known benefits of "aerating"' the soil. 



Boussingault and L^wy had indeed shown that the percentage of 

 oxygen in the soil air was almost the same as that in the atmosphere, 

 but their results were overlooked. As a matter of fact they rather 

 contributed to the growth of the idea, for in their paper they laid chief 

 stress on the fact that soil air contained 22 times as much carbon dioxide 

 as ordinary air, and did not emphasise its close similarity in oxygen 

 content. 



With the introduction of improved methods of gas analysis it 

 became possible to obtain still further refinements in the study of the 

 soil atmosphere. Schloesing^'fe- was one of the first to apply the new 

 methods and although his investigation was not very extensive it 

 sufficed to demonstrate the incorrectness of the current conception that 

 the soil air was necessarily deficient in oxygen. 



In 1880 Hempel published his book describing a fairly accurate form 

 of gas analysis apparatus which is as easy to use as Pettenkofer's and 

 readily allows of the examination of large numbers of samples of air 

 taken from the soil. It was adopted by Erich Lau in a series of analyses 

 of the air from the soil at Rostock^, one sample a month being taken 

 from a sand, a loam, and a peat soil. The general result is that the 

 soil air closely resembles ordinary air in its oxygen content, but that 

 it contains about six times as nmch carbon dioxide; the actual 

 mean values obtained at a depth of 15 cm. were, in percentages by 

 volume : 



» Ann. Afjrnii. 22, :i05. 



' Th. Sohloesinf; yi/.«, 'Sur ratmo.splidri' eoiiKiiee dans le sol,' Compt. Rend. 1889, 109, 

 (!18-20, ()7.'!-76. 



^ KrUh r.aii, Beitrdge zur Kennlnis der Zusammtntetzung tier im Ackerboditi hefindliehen 

 Lufl. Iiiiuii.'. Oissertation, Unstofk, lOOfi. 



