CONTAMINATION OF MILK 187 



with excreta. This is no doubt explained by the fact that 

 B. lactis aerogenes often multiplies more rapidly in milk than 

 the other coli group organisms. The writer has found this to 

 be the case in his work. Orr, however, did not find, as might 

 have been anticipated, this strain more numerous in the retailer 

 and consumer samples compared with the cowshed samples. 

 Orr's cowshed, retailer, and consumer results are strikingly alike. 



These findings, interesting as they are, fail to demon- 

 strate the value of these additional fermentation tests for 

 routine work. To take but one example, they do not show 

 that the B. coli communis type (strain C), which does not 

 ferment saccharose, is of any different significance from the 

 B. neapolitanus type (strain E), which does ferment saccharose. 

 From these results it is not possible to attach any additional 

 significance to the presence or absence of saccharose fermenta- 

 tion, and the same applies to most of the other sugar -alcohols 

 employed. 



The writer has carefully studied the investigations both of 

 MacConkey and Orr, and while their work is of the highest 

 value for scientific purposes, and a basis for future practical 

 work, in his opinion it cannot be said that they have estab- 

 lished that any particular variants are of greater significance as 

 an index of excreted pollution than other types. In other words, 

 the practical value of these tests has yet to be demonstrated. 



Bergey and Deehan l have also studied what they call the 

 colon-aerogenes group of bacteria, using the same differentiat- 

 ing tests as those employed by MacConkey. They isolated 

 and studied bacteria from 50 samples of milk and 8 samples 

 of sewage. As they point out, the 8 differentiating tests 

 employed (fermentation of saccharose, dulcite, adonite, iriulin, 

 motility, indol production, Voges and Proskauer's reaction, 

 gelatine liquefaction) represent 256 possible combinations. 

 Of these they found organisms representing only 43 varieties, 

 while those from the milk samples represented only 27 

 different species, less than half of which had been found by 

 MacConkey. Their results also fail to show any special utility 

 to be obtained from the use of these tests. 



In the same way the relative significance of the strains 

 which produce indol, and those which do not, has yet to be 



1 Journ. of Medical Research, 1908, xix. p. 175. 



