RESISTANCE OF BACTERIA 229 



of them are capable of regaining a normal, visible form under the 

 conditions of the experiment. Certain observations which I have made 

 incline me to accept this last hypothesis. 



However this may be, Hauduroy has studied the morphology and the 

 characters of the resistant bacteria derived from these ultrabacteria. 



In bacteriophage suspensions filtered through a candle the cultures 

 may assume one of two forms. In both cases the turbidity is at first 

 very slight, resembhng rather an opalescence similar to that of an egg- 

 white suspension, but in a certain number of cases the turbidity 

 gradually increases up to the appearance of the clouding which one 

 observes in normal cultures of bouillon. In other cases the medium 

 clears and one may observe the appearance of smaU and large 

 agglutinates. 



I have proved that these two aspects correspond to a difference in 

 virulence and as a result to differences in the resistance of the bacteria 

 derived from the ultrabacteria. In the first case the bacteriophage is 

 always of relatively low virulence, whereas when agglutinates form, the 

 virulence of the bacteriophage is very high. 



Hauduroy has followed with the microscope the formation of these 

 cultures. His conclusions on this subject may be quoted. 



To these different macroscopic appearances there are corresponding micro- 

 scopic peculiarities. The tubes which have a homogeneous turbidity of the 

 "egg-white type" contain minute granules which do not take the Gram stain. 

 They are very abundant; isolated, or arranged as on a filament which may at 

 times be seen. There are occasional bacilli which do not take the Gram stain. 

 There are masses without definite structure of variable size, having a mucous 

 character and in which granules can be detected. In tubes containing the agglu- 

 tinates we may also find granules which do not stain with the Gram stain. They 

 are very fine, extremely numerous and are found in masses which appear to be 

 formed of a reticulum containing granules and very rare bacilli. 



These elements are living. They may, as a matter of fact, reproduce and give 

 birth to elements comparable to themselves or to typical dysentery bacilli. 



The cultural characters and the biochemical reactions of these elements are 

 ordinarily different from those of the Shiga bacilli from which they are derived. 

 Several more or less distinctive types can be differentiated, which might be 

 characterized as follows: 1. Elements, which, as a rule, may be readily trans- 

 planted serially into the usual liquid media and yield there either homogeneous 

 cultures, or agglutinated cultures, or a sedimented culture which can be removed 

 from the bottom of the tube only by shaking. At times these cultures do not 

 appear until after several days' incubation at 37°C. 2. These elements can be 

 cultivated only exceptionally on solid media under aerobic conditions. After 

 many attempts I obtained once a growth on ordinary agar slants, and once a 

 culture on lactose agar. 3. The fermentation reactions of these elements are, 

 as a rule, different from those of the normal Shiga bacillus. They may have a 



