llt> 



leaving by the other outlet tube. V/hen the replacement 

 of the original atmosphere is considered sufficiently 

 complete, a gas bag half filled with the desired gas is 

 attached to one outlet tube and allowed to remain, the 

 other outlet tube being closed by a cork or by a pinch 

 cock on a rubber tube. If it be desired to change or re- 

 new the artificial soil atmosphere, as, for instance, for 

 the purpose of removing carbon dioxide or other gaseous 

 excreta from the roots, this process of passing gas 

 through the pot and attaching a bag full of fresh gas 

 may be repeated as often as desired. The use of rubber 

 tubes and pinch cocks on the outlet tubes of the pots 

 makes it possible to accomplish such changes or renewals 

 without danger of leakage from , or to, the general 

 atmosphere. 



There is a theoretical possibility of error 

 in this method owing to the ability of most gases to 

 diffuse through rubber membranes. This seems, however, 

 to be too slight to be important in the present experi- 

 ments. A gas bag half filled with nitrogen, closed and 

 exposed 4 days in the greenhouse contained, at the end 

 of that time, no oxygen determinable by absorption in 

 alkaline pyrogallol according to the usual gas-analysis 

 methods. That no substances deleterious to the plants 

 were given off by the rubber of the gas bags is shoi 

 by the fact tjiat even the plants most sensitive to dis- 

 turbances of the soil atmosphere grew satisfactorily with 



