Il6 CLINICAL BACTERIOLOGY AND HEMATOLOGY 



in which it occurs in malignant disease, and it would have been 

 overlooked on a casual examination. It is a bacillus of large size, 

 and has a tendency to grow into long threads, which are readily 

 visible under a |-inch lens (Plate IV., Fig. 5). In a wet specimen 

 these threads usually seem continuous, but on examining a 

 dried and stained specimen, they may be seen to be composed of 

 bacillary segments, much like the chains of the anthrax bacillus. 

 In a wet unstained specimen these threads are very easily 

 recognised, as they are highly retractile and may have a slow, 

 crawling motion, or may be non-motile : most usually the latter. 

 Another characteristic feature is the presence of an obtuse angle 

 in some portion of the length of many of the threads, whence the 

 name B. geniculatus is derived. The organism does not form 

 spores. It stains by Gram, and I have noted in several cases the 

 presence of a phenomenon which is very rare amongst bacilli, 

 that of longitudinal fission in a small number of the chains, 

 which thus come to form double rows of bacilli in close lateral 

 approximation. That this is not a mere effect of staining is 

 shown by the fact that occasionally the two bacilli at one end of 

 this double thread will turn away from one another, so that it 

 looks as if there were true branching at the extremity of the 

 filament. If this turns out to be the case in all specimens, it will 

 be a most important means of identifying the organism.* 



Cultures may readily be obtained on media rather highly 

 acidified with lactic acid, but are not requisite for the identifica- 

 tion of the bacillus. 



In addition to the Boas-Oppler bacillus, the vomit may contain 

 a few yeasts, distinguished by their large size, oval or spherical 

 shape, and by the way in which they reproduce themselves by 

 budding ; there may also be a few bacteria of other sorts. But 

 in many cases the bacilli are present in a state of almost absolute 

 purity, and in large numbers, so that they form tangled masses. 



In nearly all cases in which the bacillus is found the vomit 

 contains no hydrochloric acid, or only a trace. To test for it 

 filter some of the vomit and place a drop or two of the filtrate on 

 a white porcelain tile ; place a drop of a 0-5 per cent, alcoholic 

 solution of dimethyl-amido-azo-benzol close to it and let the two 

 gradually mix. If physiologically active HCl is present, even in 

 very small amount, a transient pink colour will be produced. The 



* Note to Third Edition. It is not, though I have since seen it two or 

 three times. 



