The Fermentation Tube 199 



This table shows that of ten samples of beef broth two 

 were manifestly free from glucose. Hence the advice of Dun- 

 bar" to use simply beef infusion (^Fleischwasser) to test the 

 gas producing power of bacteria would lead to conflicting re- 

 sults unless glucose were added. That the sugar contained 

 in muscular tissue is glucose as affirmed by physiologists 

 seems to be borne out by the fact that it is attacked by bacteria 

 which do not ferment lactose or saccharose. 



In order to eliminate the source of error introduced by the 

 muscle sugar I tried a solution of salts reccommended by 

 Fermi" and of the following composition : 



MgSOi 0.2 gram. 



HK.PO, I. 



(NH,)3P0, 10 



Glycerin 45 " 



Water 1000 cc. 



In this solution the bacteria experimented with failed to 

 multiply when peptone was added and the glycerin omitted. 

 When both were present the fluid in the open bulb became 

 fairly turbid but that in the closed branch remained practically 

 free from any growth. Evidently the glycerin could serve as 

 food only in presence of oxygen. When glucose was added 

 gas appeared, but much more slowly and in much smaller 

 quantity than in peptone bouillon with glucose. A compari- 

 son of results obtained with this artificial solution and pep- 

 tone bouillon was not possible and further trials with it were 

 abandoned. 



It next occurred to me that the sugar in bouillon might be 

 removed by allowing some gas-producing bacteria to multiply 

 in the latter for a time. The bouillon might then be resteril- 

 ized after a certain quantity of some sugar had been added 

 and the fluid reinocculated with the species to be studied. 

 This procedure was found successful so far as gas production 

 is concerned, but it went on more slowly and apparentl}^ in a 

 somewhat different way. Hence this method was given up. 



Dunham's solution (i per cent, peptone and )4 per cent. 



