Some Phases of the Disinfection Theory. 77 



(5) More economical utilization of the calories of food consumed 

 (as well as of the body material of the mother) in the rearing of 

 the young to weaning age. 



40 (1787) 



Some phases of the disinfection theory. 



By BARNETT COHEN. 



[From the Hygienic Laboratory, Washington, D. C] 



Bact. typhosum and Bad. coli (communis) were suspended in 

 distilled water, tap water, and M/500 buffer solutions, respectively, 

 at constant temperature levels (o°, io°, 20 0 , 30°C); and the 

 numbers of survivors were determined by means of decimal dilu- 

 tions upon agar plates. The conditions imposed (moderate H-ion 

 concentrations at moderate temperatures) permitted a closer study 

 of the disinfection process than has been usually possible. 



It was found, of course, that coli was relatively more resistant 

 than typhosum, but this greater resistance (at Ph 3-5) decreased as 

 the temperature level rose. At o° C, coli was 67 times more 

 resistant, and at 30 0 C. it was only 8 times more resistant than 

 typhosum. There was a high inconstancy in results between 

 duplicate tests carried out in tap or distilled water. This incon- 

 stancy could at times be correlated with comparatively insignifi- 

 cant fluctuations in P H of the water. When very dilute (M/500) 

 Clark and Lubs buffers were used, this variability disappeared very 

 largely. 



At 20 0 C, Bact. typhosum possesses the greatest tolerance 

 within a narrow zone delimited by Ph 5.0 and 6.4. A slight in- 

 crease in acidity beyond the zone results in conditions of maximum 

 mortality. For Bact. coli the zone is wider and centered about 

 neutrality. Cohen and Clark 1 found that the Ph optima for 

 growth and fermentation of bacteria may be different. It is now 

 found that the optimum for tolerance may also be distinct. 



The logarithmic decline in numbers of bacteria may be modified 

 by suitably chosen conditions. This applies also to some mono- 

 molecular chemical reactions. The logarithmic course in either 

 case is merely a statistical integration and gives no information 



1 Cohen, B., and Clark, W. M., Jour. Bad., 1919, iv, 409. 



