IX 



lation test to be free from sugar. The figures prefixed 

 to H 2 S0 4 and NaOH in the tables signify the quantity 

 in Cc. of Vi normal H.J So 4 and of ] /i normal NaOH 

 added to 100 Cc. of neutral bouillon or agar. 



To prevent error from the use of sulphuric acid we 

 have also experimented with organic acids (acetic, citric) 

 and obtained practically identical results. 



As it soon became evident that the results depended 

 upon many and varied causes, a systematical analysis 

 was made with 6 typical strains of B. coli and 5 of 

 B. typhosus. The alterations in each strain were iden- 

 tical, according to the different conditions to which 

 they were subjected. Unless otherwise stated in the 

 tables, the same strain has been used throughout the 



