468 



inoculated with 0.25 7,, and 0.5% NaCl + blood, the surface did 

 not anj more appear to me quite clear and transparent, though I 

 did not see any colonies either. Next da}' little more was noticeable. 

 However one preparation made of it revealed diplobacilli. 1 now 

 took a piece of this somewhat turbid-looking agar and ground it 

 in broth. 



In the depths of the stabculture something similar was to be 

 seen; preparations of it contained the same diplobacilli. Of this agar 

 I also put some pieces in broth. 



Preparations made from these broth-tubes showed numerous 

 diplobacilli. I made a subculture of some of this broth upon sloped 

 agar. Only after 24 hours did I see a few colonies appear. Likewise 

 a few colonies were visible upon the agar inoculated with expression- 

 water, very much like the preceding. Repeated subcultures upon 

 agar resulted in more colonies; though the growth was rather poor, 

 it gradually seemed to progress. 



The colonies produced looked white with incident light, were 

 clumped and could be easily shifted bodily with a needle all over 

 the agar-surface. It seemed as if the colonies floated upon the agar. 

 With transmitted light they looked bluish and translucent. 



The bacilli are not all exactly alike as to shape. We found short 

 rods and longer ones. The short ones generally occur in paii'S 

 joined end to end, so that they might be called diplobacilli. Most 

 often the centre of the longer ones is of a lighter colour, which 

 renders the bacillus dumbbell- or biscuit-shaped. Maybe we have to 

 do here also with two bacilli joined end to end, or all of them are 

 single bacilli depressed in the middle? However this may be, it is 

 evident that the two parts are linked together. In some cases we 

 might even term them diplococci, the cocci being slightly elongated 

 (coccobacilli). See Fig. 1 and 2 of the plate. 



The bacilli (I will stick to that term) have no spontaneous 

 motility. They readily stain with the ordinary dyes, most distinctly 

 with carbol-gentian- violet, which, therefore, I made ample use of, 

 also in my investigations of the bacilli in the digestive and intestinal 

 tract of lice. With Löffler's methyleneblue we can see at the poles 

 more darkly stained granules, occasionally distributed all over the 

 body. Tliej are Gram-positive and non -acid proof. In ascites-glycerin- 

 bouillon, and in ordinary broth cultures, especially the older ones, 

 other bacilli are found among the Gram-positive bacilli, that look 

 exactly like them, but are decoloi-ized. 



The bacilli are clumped, which renders suspension difficult. 



The best temperature for the growth seems to be 37°. At room- 



