20 
C. A. LUDWIG 
the gas was stored over water until required for use. Under these 
conditions the inhibitory effect of the gas was so sHght as to create 
doubt as to its purity. Accordingly a sample was analyzed^ and 
found to consist of a mixture of methane, hydrogen, oxygen, and 
nitrogen, of which the methane comprised about 50 percent and the 
nitrogen about 30 percent. 
In order to get more exact data a mercury seal gasometer was 
secured for storage of the gas and each quantity of gas used was 
sampled for analysis. By this means dilution of the gas due to its 
solution in the water of the gasometer and the giving up of nitrogen 
to it by the water were avoided. It was also possible by means of the 
analysis to get an accurate measure of the concentrations used. The 
upper limit for the concentrations thus secured was 65 percent of 
methane with a maximum oxygen content of about 10 percent. A 
sample analysis follows: 
CO2 2.0% CH4 65.0% 
CnU^n 0.5% H2 20.3% 
O, 10.3% N2 (difference) 1.9% 
The results of the later tests confirmed the previously indicated low 
toxic properties of this gas as affecting the organisms in question. 
In fact, some of the previous tests gave rather stronger reactions on 
the part of the organisms than did the last ones in spite of the fact 
that all conditions indicate that the gas in the former trials was more 
dilute. 
Effect on the Different Organisms 
In these reports the results given are based on both the earlier 
and later work, care being taken not to make the gas seem less toxic 
than the data warrant. 
Bacillus suhtilis. — In all of the concentrations of methane used the 
development of B. suhtilis was normal in character and good in amount. 
There was some checking of development in the higher concentrations 
which sometimes extended apparently as low as 25 percent (value un- 
corrected by analysis) ; but it was never great and was dissipated at 
the end of 3 days, except in a concentration of approximately 50 percent 
(corrected) or higher, where the slight retardation persisted in one 
or two cases for 6 days. 
2 The analyses of methane, carbon monoxide, and ethylene given in this paper 
were made by Prof. W. L. Badger and Mr. Philip W. Shepard, of the department of 
chemical engineering of the University of Michigan. 
