DISTRIBUTION OF B. COLI COMMUNE 119 
fermentation tubes, and usually contains no other sugar but lactose. It should, 
however, always be tested with a culture of B. enteritidis (Gartner), for example, to see 
if traces of glucose are present. An organism which, while generally resembling 
B. coli, coagulates milk and ferments lactose may, 1 think, be considered to belong 
to the coli group. 
The fermentation of saccharose does not appear to be a constant quality in 
the group ; while Escherich's original B. coli commune, of which I tested a culture, 
gave no fermentation in a saccharose broth, many isolated from different sources 
showed production both of acid and gas, while otherwise indistinguishable from the 
original Escherich culture. For example, out of thirty-seven examples that were 
thus examined, seventeen fermented saccharose with production of acid and gas, two 
produced acid but no gas, while sixteen showed no change at all in the medium. 
The duration of the experiments was from eight to eleven days. 
I AIR 
The experiments with air, already described,* being somewhat inadequate, the 
following were undertaken with a view to making the investigation more complete. 
The original apparatus was discarded in favour of one which was both more 
convenient and more accurate. As before, a measured quantity of air was drawn 
through a wool plug, the plug well washed with water, and a portion of the latter 
plated, and so analysed. As a rule the plug consisted of two parts ; the air first 
went through a glass wool plug and then through one of cotton wool, the former 
being much easier to wash thoroughly, while the latter is a more efficient filter. The 
air was drawn through by means of a pump, which was worked by a hot air engine, 
and each revolution of the engine wheel was recorded by a cyclometer. The record 
of the cyclometer was calibrated by aspirating a known volume of air, and it was found 
to give concordant results ; the apparatus was usually left running for about twenty- 
four hours. I am much indebted for the use of the machine to Dr. E. E. Glynn, 
under whose direction the whole apparatus was put together. 
Experiment I — The air outside the laboratory was analysed on the night of 
April 30-31, 1900, after wet weather. B. coli was found absent in 240 litres ; a 
bacillus resembling B. foecalis alealigenes was, however, isolated on this occasion. 
The details of this and of all the other experiments are given in Table I. 
Experiment II — The air of the College grounds was analysed for five 
consecutive days, in dry, dusty weather, from May 14 to May 19, 1900. B. coli 
was found invariably to be absent. 
Experiment III — A dusty street corner was chosen as the place for analysis 
of the air, which was examined for seven consecutive days ; the air was drawn 
* r. T. Reports, vol. Ill, p. n. 
