COLON-AEROGENES GROUP OF BACTERIA 



277 



though this method is very crude, some interesting data were 

 obtained. 



Cultures of the B. coli type were seldom observed to attain a 

 gas volume greater than 40 per cent when the medium had 1 

 per cent of glucose, the amount of gas usually being between 20 

 and 40 per cent. More specifically, of 173 fecal strains of B. 

 coli cultures used only 8 gave a gas volume of 40 per cent, all of 

 the remaining 165 falling below this figure. On the other hand, 

 322 out of 447 aerogenes strains gave a percentage higher than 

 40, 74 produced 40, and only 51 gave less than 40 per cent. 



On account of the lack of proper facilities gas determination 

 by the exact method was not undertaken until toward the end 

 of the present study. Considerable time was spent in designing 

 a vacuum pump which was a modification of the Sprengel pump, 

 the plan of the Boltwood pump being followed to a large extent. 

 As the Boltwood pump is- not adapted for the collection of gases, 

 and as no provision is made for removing the minute air bubbles 

 which collect in the mercury, it had to be further modified. The 

 distinct features of the new design of vacuum pump are: (1) 

 It can be employed for exhaustion of the gases as well as their 

 collection, (2) It possesses a device for the removal of air bubbles 

 from the mercury, and (3) Its operation is automatic and con- 

 tinuous. As it is planned to give a complete description of this 

 apparatus in a separate , publication (Chen), further discussion 

 of its design and application is omitted here, except a brief 

 allusion to the following table which gives the representative 

 results of one of the few estimations which were hurriedly made 

 at the close of the present study, and which as far as it goes 

 confirms the gas ratio of Rogers, Clark, etc. 



Gas-production in vacuum bulbs 



