34 The Atiii(>.^i>liere <>f tli< Soil 



effect is to increase it (p. 23, Fig. 9). Tlie following data serve as 

 illustrations : 



These observations confirni tlie older results of Fodor^. 



(1) Strong winds. On several occasions, e.g. February 3nl. ]\Iarcli 

 7th, 1913, samples were taken directly after a windy night but there 

 was nothing at all to indicate that the composition of the air had been 

 affected by the wind. A current of air passing rapidly over the soil 

 might have been expected to draw out the soil air, but apparently it 

 does not. I'robably the force is insufficient, the layer of air in contact 

 with the surface of the soil moves less quickly than the layers a few 

 inches above. Moreover any removal of air by this process from the 

 surface layers of the soil probably leads to an upward movement of 

 air rich in COg from the lower depths. 



(2) Change in barometric pressure. Fodor- found that the C'O,^ in soil 

 air rose with falling barometer at three stations out of four where in- 

 vestigations were made. In the only continuous experiment we made we 

 were fortunate in hajjpening upon a time when the barometer was rapidly 

 falling and we also obtained a rise in COg during the period. But when 

 the whole of our COg figures are plotted against barometic pressures 

 or even against changes in barometric pressure no consistent relationshi]) 

 can be observed such as is obtained with rainfall, temperature, etc., so 

 that the influence of barometric pressure appears to be only minor and 

 easily swamped by other factors. 



(3) Night and day. Fodor^ and Wollny ' thought they had evidence 

 that CO2 streams out from the soil air at night but we can find no 

 indication of any greater loss by night than by day. Samples drawn 

 from the same .') holes at consecutive 3-hour intervals over a period of 



' Josef Fodor. Ili/gienisclie Unlenuchungen HberLuft, Boden und Wasmr, Brauni;ehwpig, 

 ISSI. p. 130. 



- Fodor, ihid. p. 135. ' Fodor. ibid. p. 53. 



' Wollny, Fnr.vh. nnf don (Irliietr d.r .\ij>ik.-Ph\jsik, 1885, 8. 417. 



