Dh'^INFECTAXTl^ 111 



we draw them through lines which represent ranges of 

 death times. 



The curve obtained must be a straight line as explained 

 on p. 72. Different slopes indicate different concentra- 

 tion exponents. Figure 23 shows a number of such curves 

 for common disinfectants. Phenol and permanganate 

 furnish slopes quite different from those of the other 

 disinfectants, thus indicating unusual concentration expo- 

 nents. 



Rectilinear interpolation and extrapolation give the 

 death times for any concentration, or the concentrations 

 necessary to kill the bacteria at any exposure time. 



These curves represent the effect of the disinfectant 

 upon bacteria in the absence of organic matter. The ef- 

 fect of organic matter can be measured quantitatively by a 

 parallel set of death time determinations, the disinfectant 

 being dissolved in water containing 10% skimmed milk. 

 The results are plotted in the same way as those with 

 water, and again, straight lines are obtained. 



The effect of organic matter is expressed by the ''per- 

 centage delay." This is calculated from the two death 

 times at the same concentration (which may eventually 

 be obtained by extrapolation). Death time in milk will 

 be longer than death time in water or salt solution, and 

 the percentage of increase in time tells to what extent or- 

 ganic matter interferes ; e.g., for hydrogen peroxide the 

 delay in skimmed milk is 30%, for formaldehyde 20-27%, 

 for H^CL many thousand per cent. 



The phenol coefficient can be obtained directly from the 

 graph by dividing the phenol concentration which kills 

 "between 5 and 10 minutes" by the corresponding con- 

 centration of the disinfectant under test. In most cases, 

 the coefficient measured at 5 minutes will be larger than 

 that measured at 10 minutes because, phenol having the 

 highest concentration exponent known, the slope of the 

 phenol disinfection curve is more nearly vertical than 



