2 90 EARLE B. PHELPS 



the water. This tends to produce a difference in osmotic pressure 

 between the free solution and this surrounding layer, and it may be 

 that a certain definite concentration of the copper ions is necessary 

 before this film of "osmotic tension" can be pierced. An analogous 

 case is that of surface tension, which presents a certain resistance 

 to the entrance of a non- wetted body. 



Peptone. Peptone was the third substance studied. The results 

 of Experiment 7 show that a i per cent solution of peptone allows the 

 typhoid organism to multiply even in the presence of rather strong 

 copper- sulphate solution. Owing to the high electric conductivity 

 of the rather impure peptone, the results of conductivity determina- 

 tions are uncertain. There seems to be little doubt here, however, 

 that an actual combination has taken place between the copper and 

 the peptone. The addition of sufficient copper solution to the peptone 

 solution to give a distinct color resulted in the formation of a colloidal- 

 like solution, of a robin 's-egg blue color. 



These results have an important bearing upon many practical 

 points in connection with the use of copper sulphate for the destruc- 

 tion of the typhoid organism. Results obtained in the laboratory 

 in distilled water are not in the least indicative of what may be 

 expected under field conditions. Neither are actual field results 

 on one water reliable criteria for the undertaking of similar work 

 upon another. The impracticability of the internal use of copper 

 sulphate in the treatment of typhoid fever is also suggested by the 

 results obtained with peptone. 



Selection oj a resistant strain. Cultures A, B, and C were taken 

 from Bottle 2 in Experiment 5, and had lived for 48 hours in a solu- 

 tion of copper sulphate containing 0.63 parts of copper per million. 

 They were used in Experiments 5 and 6 in parallel with the original 

 strain in order to determine whether they possessed any increased resist- 

 ance to copper. The results are all negative, indicating that these 

 strains are not more resistant than is the parent strain. 



The time j actor in the germicidal effect. In Experiments 1,3, and 

 4, where determinations of the numbers of organisms were made 

 at various intervals during the experiment, it is seen that even in the 

 more dilute solution used there is a very rapid falling-off in numbers 

 during the earlier part of the test. In Experiment i, for instance, 



