INFLUENCE OF ILLUMINATING GAS ON BACTERIA AND FUNGI 7 
and set aside. The reason for introducing the gas while the pressure 
was low was to secure the vacuum as an aid to distribution in getting 
the gas through the plugs into the tubes in contact with the cultures. 
Some accompanying experiments with tobacco smoke and methyl 
iodide vapor, in which diffusion alone was relied on to get the gases 
into the tubes, gave results in the resulting cultures which showed 
that the gases did get into the tubes, at least in small amounts, very 
promptly. It seems not unreasonable, therefore, to think that the 
composition of the gas within the tubes approached pretty closely 
that in the rest of the chamber, although it was impossible, of course, 
to get absolute data on the point. The concentrations mentioned in 
all cases are to be considered not as exact values but merely close 
approximations. When pure gas of some kind was desired in contact 
with the cultures, one of two or three different plans was employed. 
In the case of illuminating gas, it was either allowed to pass through 
the vessel continuously during the experiment or it was passed through 
rapidly for one to two hours and then stopped. In the latter case it 
was usually renewed daily during the experiment. For other gases, 
which had to be manufactured for the purpose, the test tubes contain- 
ing the cultures were fitted with perforated rubber stoppers through 
which small glass tubes passed. These test tubes were arranged in a 
chain and the gas was passed directly through them. The stoppers 
were sealed to the glass with which they were in contact by means of 
sealing wax, and the rubber connections between tubes were carefully 
wired and paraffined. 
All results here reported, unless otherwise stated, were from at 
least two trials; and many of them were checked several times. 
EXPERIMENTAL 
I. Illuminating Gas 
Source and Composition of the Gas 
A large part of the work consisted of tests with illuminating gas. 
Such tests have the disadvantage, of course, that the gas is a mixture, 
and not a perfectly constant mixture at that ; but its ready availability 
and the fact that it is the substance which usually contaminates 
laboratory air made it seem worth while to use it. The gas used 
was taken from the gas taps in the laboratories and was the same as 
