LOGARITHMIC ORDER OF DEATH 



53 



pendently, resulted in survivor curves which were con- 

 cave downwards. Charlton and Levine (1937) suspend- 

 ed the dried spores of a Bacillus (either laterosporus or 

 r^uminatus) —which, were found by microscopic inspec- 

 tion to be mostly single — in water with varying amounts 

 of chlorine, and counted the survivors by plating. Mall- 

 mann and Ardrey (1940) used a suspension of Bacterium 

 coli in water which had been chlorinated 30 minutes pre- 

 viously; their results are represented in Fig. 15. 



It is remarkable that chlorine differs so much in this 

 respect from most other disinfectants. No comparison 

 with iodine could be made since the order of death by 

 iodine does not seem to have been investigated. It is con- 

 ceivable that chlorine, being such a very active chemical 

 reagent, kills bacteria by destroying the membrane, or 

 the protoplasm, or the enzymes before attacking the 

 mechanism of reproduction. This would result in the 

 observed order of death. 



Fig. 15. Semi-logarithmic plot of survivor curves of Bacterium coli 

 treated with chlorine. Abscissa: time of exposure; ordinate: per cent 

 of survivors, on logarithmic scale. (From data of Mallmann and 

 Audrey, 1940.) 



