III. THE ACTIVITY OF VIRUS PARTICLES 21 



in general, exhibited their action easier in a broth than on an agar plate. 



Since the ratio can vary in such a wide range, it seems quite im- 

 possible to calculate the absolute number of active phage particles ; we 

 can only know the relative number of particles which exhibit their 

 action under a definite condition. It is a well known fact that even 

 the number of plaques that may appear on an agar plate varies wi<"h 

 the property of the agar plate. There may exist phage particles which 

 can exhibit their action neither in a broth nor on an agar plate. 



This fact must be of the utmost importance, giving warning to 

 virus researchers to be very prudent in estimating the activity of some 

 virus samples. If all the phage particles, the ratio of which is 3,000 as in 

 a sample above cited, can act as the active phage on an agar plate and 

 the activity of such particles are measured in a broth, then it will 

 appear as if only one of 3,000 particles had the activity ; one would 

 commit a great mistake, if one concluded from such a result that only 

 0.03 per cent of the particles were the true phage, the remaining 99.97 

 per cent being "impurities". 



2. The Effect of the Difference in Bacterial Strains on 

 Phage Activity 



The writer discussed above the difference in the phage activity due 

 to the conditions under which a certain strain of bacteria are affected 

 by phage. If similar observations were made with various bacterial 

 strains, much greater differences in the phage activity would be revealed. 



On endeavouring to isolate phage preparations possessing the ac- 

 tivity as high as possible, we succeeded in obtaining the phage sample 

 whose minimal quantity required for producing a plaque was 6.8 XlO"-'^ 

 g. ; a certain strain of E. colt was used in this experiment (2). On the 

 other hand, the average dry weight of a single particle of this pre- 

 paration was estimated to be about 1/2x10"^^ g.; hence almost all the 

 produced particles could be considered as possessing the phage activity. 

 The average dry weight was estimated by the method already described. 

 This indicates that all the particles obtained by our method can reveal, 

 under a proper condition, the phage action, a conclusion which is of a 

 profound significance from various points of view as will be mentioned 

 later. 



The experimental result from which this remarkable conclusion 

 was derived was obtained by using a certain strain of E. colt, which 

 was highly sensitive to the phage ; but if other strains of bacteria were 

 used, utterly different results would be obtained. For axample, the 

 phage strain used in this experiment could infect a strain of typhoid 



