148 DISINFECTION 



of time. Nor does it give us any information about the toxicity of the disinfectant 

 to the tissues, its anti-leucocytic power, its activity in the presence of protein or 

 lipins, or its selective action on different bacteria. Any attempt to express by 

 a single index a number of dissociated properties is clearly illogical. The phenol 

 coefficient, as given by the Rideal- Walker method, is at best a grossly over-simplified 

 answer to a very difficult problem, and at worst little short of bacteriological 

 prostitution. To scientific workers, and to those interested in human and animal 

 therapy, the result is meaningless. Were it not for the big financial interests 

 concerned in the commercial production of disinfectants, the test would doubtless 

 die of disuse. 



What is the remedy ? Several have been proposed, but none is altogether 

 satisfactoTy. Phelps (1911) suggests that a method of standardization should 

 comprise the determination of k, n, and d for each disinfectant. By the formula 



KC"« = log ? 

 b 



we can calculate the reaction velocity at any concentration, and by the formula 



Kt.=:K2ooX0'T-2O«) 



where Kj represents the velocity to be calculated at the temperature desired 

 and Kgo" the velocity actually determined at a temperature of 20° C. (or any 

 other convenient temperature), we can determine the value of the temperature 

 coefficient 6. 



One objection to this method is that the reaction velocity often varies during 

 the progress of disinfection. To meet this difficulty, Hobbs and Wilson (1942) 

 suggested that the value of k should be taken in the middle stage of the reaction. 

 Withell (194:2a, b) has improved on this by proposing that the time taken to destroy 

 50 per cent, of the organisms — Lt 50 — should be selected as a comparative measure 

 of bactericidal efficiency, and has pointed out that this value can be most easily 

 determined by the use of probit logarithm of time graphs (see Bliss 1938, 1941). 

 Withell (19426) has also drawn attention to the danger of attempting to compare 

 two bactericidal agents that yield different types of time-survivor curve. Many of 

 these are at variance with that yielded by phenol, and a phenol coefficient is therefore 

 inapplicable. 



If Phelps' method is adopted, then it is advisable to test the disinfectant against 

 two or three different organisms, as suggested by Eisenberg (1919), since it is known 

 that some germicidal agents have a strongly selective action on certain bacteria. 

 The information derived from the use of this method is considerably greater and 

 more valuable than that supplied by the Rideal-Walker coefficient. Nevertheless, 

 it still leaves us ignorant of a number of important properties of the disinfectant, 

 particularly its toxicity to body tissues and its activity in the presence of cellular 

 material. For this reason, the behaviour of the disinfectant in tissue suspensions 

 or cultures can be studied (see Bronfenbrenner et al. 1939, Salle et al. 1939, Welch 

 and Brewer 1942). 



In conclusion, we may say that work of recent years, particularly on the 

 sulphonamides, has shown how inadequate any laboratory test of germicidla 

 activity is to indicate the behaviour of a given substance in the animal body. 

 Attention, in the future, is therefore likely to be devoted far more than in the past 

 to a study of disinfectant agents in the presence of Uving tissues. 



