380 APPLIED BACTERIOLOGY 



these threads, usually after drying, into solutions of the 

 disinfectant. Defries and others have found that the 

 results of this method are not necessarily the same as those 

 obtained by the method of mixtures, even when the opera- 

 tions are conducted simultaneously on the same cultures 

 and solutions. This method is also unsatisfactory, in that 

 it affords no means of maintaining a standard relation 

 between the number of organisms and the quantity of dis- 

 infectant, because the penetration of a disinfectant into the 

 depths of the thread has been found to be notably irregular, 

 and because traces of the disinfectant are extremely liable 

 to adhere to the thread. Endeavours have been made to 

 avoid the difficulty by washing the threads in sterile water 

 until it is to be presumed that the disinfectant has dis- 

 appeared. The presumption, however, is very apt to fail. 

 With some disinfectants it is possible to form by the 

 addition of a substance possessing no disinfectant capacity 

 an inert insoluble compound ; and attempts have been 

 made to avoid the transfer of disinfectant with the organ- 

 isms on the thread by applying such treatment. For 

 example, it has been usual to wash the threads exposed to 

 perchloride of mercury with ammonium sulphide, and 

 thereafter preferably also with water, so as to avoid such 

 antiseptic effect as the compound resulting from the 

 previous treatment may possess. Irregular penetration, 

 however, of the disinfectant makes it quite possible that 

 some portions may not be reached by this process. These 

 objections to the method are to a considerable degree 

 removed when threads of slag-wool are used instead of silk. 

 This substitution does not, however, affect the liability of 

 the threads to be contaminated in one or other of the 

 successive manipulations ; and a new difficulty is introduced 

 in the liability of the organisms to be mechanically washed 

 off the smooth surface of the glass. Defries found this to 



