GREEN FLUORESCENT BACTERIA FROM WATER 71 



weeks, the period of time recommended on the Society's chart. 

 Seven strains began to liquefy after three to five months. The 

 time for complete liquefaction was not determined, but at the 

 end of eight and one-half months the tubes when melted and 

 poured into plates, showed growth in pure culture. The group 

 number is made up on the assumption that all strains liquefied 

 gelatin. No spore formers were found in any of the strains. 



EXPERIMENTAL 



Method of study 



The descriptive chart of the Society of American Bacteriolo- 

 gists was used for recording the data in this investigation of the 

 fluorescent bacteria. The group number was determined for 

 each stxain under as uniform conditions as possible. One objec- 

 tion to this group number is the lack of uniform and satisfactory 

 methods for the determination of some of the digits. Individual 

 workers may use their own methods, and it is possible that 

 group numbers determined in different laboratories ought not 

 to be compared. It is understood that a code of standard 

 methods is being prepared which ought to allow a wider appli- 

 cation of the group number in systematic studies of bacteria. 



Media and technique 



The media and technique used in this study conformed to 

 Standard Methods for the Examination of Water and Sewage 

 of the American Public Health Association, 1912, except that 

 in place of meat infusion, Liebig's meat extract was used. It 

 was believed that in this way more uniform media were secured. 



In those instances where special media were used the com- 

 position will be mentioned under the special heading. 



Sources of strains 



All the strains of fluorescent bacteria were isolated from 

 samples of water which were collected at different places in the 



