QUALITATIVE BACTERIOLOGICAL ANALYSIS. 167 



After two or three days incubation the acid reaction of 

 B. coli in No. I. medium changes to an alkaline reaction. 



Most of the gas-producing varieties of B. coli conform to the 

 reaction described by Proskauer and Capaldi. 



The B. enteritidis of Gartner also behaves like a B. coli. 

 Varieties of B. coli which I isolated from water and which 

 strongly resembled the B. typhosus in nearly every test failed to 

 give the characteristic reaction in these media ; they all grew 

 in No. I. medium and rendered No. II. medium alkaline in 

 twenty hours. The B. typhosus simulans varieties (a, b, c, 

 and d) described by Houston were also, according to Horton- 

 Smith, at once differentiated from the typhoid bacillus by means 

 of these media. 



Radzievsky has lately tested Proskauer and Capaldi's reactions 

 with sixty-six varieties of the B. coli. All the microbes grew 

 in No. I. medium, and in No. II. medium the reaction was 

 neutral or slightly acid. 



Proskauer and Capaldi's tests are thus seen to be extremely 

 valuable, as by them the B. typhosus can at once be distinguished 

 from a large number of bacteria which strongly resemble it. 

 At the same time it cannot be said that a bacillus which does 

 not grow in No. I. medium and produces an acid reaction in 

 No. II. medium is necessarily the B. typhosus. I have quite 

 lately isolated from polluted water three varieties of B. coli 

 which produced gas in glucose media, and yet reacted in Pros- 

 kauer and Capaldi's media exactly like the true B. typhosus. 



(8) The Reactions with Anti-typhoid Serum. (a) The aggluti- 

 nation test. When serum from an animal immunised by injections 

 of B. typhosus is mixed with a broth culture of B. typhosus, or 

 an emulsion in broth of a twenty-four hours growth of the 

 typhoid bacillus on agar, it is found that the bacilli become 

 immobilised and heaped together in motionless masses, or in 

 other words agglutinated. The reaction &o produced was at 

 first considered specific, but further study soon showed that 

 the sera of healthy human beings and animals also possessed 

 the power of agglutinating the typhoid bacillus. WidaFs 

 experiments showed (1) that the serum from a horse, monkey, 

 or rabbit, diluted from 1-30 to 1-50, might agglutinate 

 the B. typhosus ; (2) that the serum of guinea-pigs had usually 



