PROPERTIES OF THE BACTERIOPHAGE 277 



These results indicate, then, that for some undetermined reason, the 

 size of the corpuscles diminishes with age. 



This variation in filtrability can not be a question of the union of the 

 young bacteriophage corpuscles into agglutinates for in this case com- 

 parative countings of the plaques on agar, made from a recent filtrate 

 and from an old filtrate, would reveal this change, since each agglutinate 

 would yield only one plaque. But this is not the case, for comparable 

 spreadings on agar show rather a diminution in the number of cor- 

 puscles in the old filtrates, even if they are aged but a few days. 



Diffusibility 



Jotten"^ has implied that the bacteriophage diffuses into a 3 per cent 

 agar. With reference to such experiments it should be stated that tests 

 of diffusibility offer many opportunities for error. 



Prausnitz and Firle^^'^ carried out a series of studies for the purpose 

 of determining whether or not such a diffusion of the bacteriophage 

 corpuscles takes place. All of their tests gave results indicating that 

 this question must be answered in the negative. They were unable to 

 obtain a filtration through a layer of 3 per cent agar deposited on a 

 Berkefeld candle, even' when the pressure amounted to 26 cm. of 

 mercury and the test was prolonged for 3 or 4 hours. 



Flu^^^ reached the same conclusion, after a series of experiments, all 

 giving negative results. In his tests the bacteriophage did not diffuse 

 through a centimeter of 3 per cent agar. 



As a matter of fact, the uniformly accepted conclusion, that the 

 bacteriophage will not pass through tight collodion membranes renders 

 it highly improbable that it would diffuse through a 3 per cent agar. 

 For we know that the interstices found in a gel are much smaller than 

 those of membranes. Thus, if there were a diffusion it would necessa- 

 rily have followed that the passage of the material took place by virtue 

 of some power of motion resident within the corpuscle. Manifestly, 

 such a condition is possible but it would be strange in view of the 

 dimensions of the corpuscles. 



Volatility 



It is also because of an imperfect technic, an error, that Olsen and 

 Yasaki^^^ have been led to believe that the bacteriophage is volatile. 

 Apparently they had forgotten the experiments of Fliigge who showed a 

 long time ago that bacteria as large as B. tuberculosis, may be carried 

 over by the droplets of the liquid which leave a fluid boiling under 

 reduced pressure. 



