8 Till Atiii(»<i>li( >■( of til) Soil 



that liad not recently heen manured liaving tlie following mean 

 composition : 



Carbon dioxide 0-9 per cent, by volume 



Oxygen 19-() 



Nitrogen 7!)-5 ,, ,, ., 



It is clear that the method gives ratlier high results for carbon 

 dioxide because atmospheric air was found to contain 0-04 per cent, 

 instead of 0-03 per cent. The air from a recently manured soil contained 

 much more carbon dioxide up to 10 per cent. — while the oxygen fell 

 as low as 10 per cent.'; but as these are the only two out of the 36 

 thov have been omitted from the general mean. 



Houssingauit and Lewv did not continue their analvses over anv 

 prolonged period, nor did they study the effect of conditions such as 

 temperature, moisture content, etc., on the composition of the soil 

 atmosphere. These pro))lems were investigated in (Tcrmany and the 

 work was the outcome of the discovery by Pettenkofer- of a simple 

 and rapid method of estimating carbon dioxide which he successfully 

 applied in determining the amount of carbon dioxide in the air of the 

 Munich soils^. This new method was much more rapid than the older 

 one of Boussingault, enabling many determinations to be made and not 

 rec[uiring great skill in manipulation. Hence a number of workers took 

 it up and a succession of papers on the subject appeared in Wnllny's 

 Journal* also published from Munich. 



It is unnecessary to review all the papers in detail : especially as 

 this has already been done by Fodor^, Wollnv*, and Letts and Wake'. 

 Moreover, later work has shown that the results are about 30 per cent, 

 too high*. l*'or comparative purposes, however, the method serves 



' We cannot liilp tliinking there must have been some mistake here ; in our experience 

 the oxygen falls very low only in waterlogged soils (p. 32). 



* Letts and Blake (Proc. Roy. Soc. Dublin, I'MO, 9, 1 10) have shown that the principle 

 of the method had already been used by Dalton and liis pupils, but this work seems to 

 have been unknown to Pettenkofer. 



' M. von Pettenkofer, 'Ueber den Kohlensiiuregehalt der Gnnidluft ini GcroUboden von 

 .Miinohen in versohiedenen Tiefin und zu veisobiedenen Zeiten,' ZcUkcIi.J. Binlnyie, 1871, 

 7, 395-417; and 1873, 9, 250-257. 



' ForschxiiKjen uvf dem Oebicle der Agrik-ultur-Physik; 1878-1898. 



' J. Fodor, Jlygienische Unlersuchuvgen iibcr Lu/I, Bodcii und Wa,ucr, IJraunsohweig, 

 1881. 



* E. WoUny, Die Zersetzunrj der organischen Sloffe, 1897. 



' E. A. Letts and R. F. Blake, • The carbonic anhydride of the atmosphere,' Proc. 

 Roy. Soc. Dtibltn 1900, 9, 107-270. ospe<ially pp. 214 ei ««/. 



* Caldwell, in Letts and Blake's paper, Proc liny. Sor. Dnhlin I'.HKI, 9. 219-229. 



