8 BULLETIN 51, r. S. DEPARTMENT OF AGEICULTUBE. 



the relation between bacterial multiplication and the formation of 

 lactic acid in milk; Biirri and Kursteiner^ state that the lactic-acid 

 organisms of milk may increase to 100 millions per cubic centimeter 

 before there is a definite rise in acidity. The same observation has 

 been made by Pennington ' and associates when investigating the 

 decomposition of chicken flesh. In connection with this last study 

 it was found that a marked rise in the ammoniacal nitrogen content 

 of the flesh did not appear when the bacterial counts indicated a 

 few million organisms per gram, but at the next examination, when 

 the count was usually in the hundreds of millions, a rise was com- 

 monly found. The principles observed by the investigators cited 

 are corroborated by the results of the study of bacterial content 

 and chemical changes in eggs. 



Not until chemical analyses show an increase in the ammoniacal 

 nitrogen are the senses able to detect infected eggs readily. Studies 

 discussed in detail in another section of this bulletin (p. 73) showed 

 that sour eggs which in the early stages are, with difficulty, detected 

 by the sense of smell, and eggs with light green albumen, which 

 are recognized by careful scrutiny, contain bacteria in many millions. 



Eighty commercial samples which contained organisms capable of 

 producing gas from lactose in the presence of bile salt were examined 

 for B. coll, as described in tlie chapter on laboratory practice (p. 76). 



A portion from each of the higher dilution fermentation tubes, 

 which showed gas, was plated on Endo's medium, or on lactose litmus 

 agar, and from each plate having typical coli-like colonies several 

 were selected for examination. 



Organisms conforming strictly to the definition for B. coli com- 

 munis, in the 1905 report of the American Public Health Association 

 on Standard Methods of "Water Analj^sis, were isolated from 55 per 

 cent of the samples examined. Probably 15 per cent more contained 

 organisms which would be classed as typical B. coli by a majority of 

 observers because they differed very slightly from the definition just 

 given. In the remaining 30 per cent of samples, which were not found 

 to contain typical B. coli^ the predominating gas-producing organism 

 was B. {Lactis) aerogenes. Of the typical B. coli organisms ex- 

 amined 81 per cent produced gas from sucrose. Practically all of 

 the organisms examined would be classed as members of the Colon 

 CLerogenes group. If gas production from various sugars is sufficient 

 to distinguish between varieties of these organisms, a veiy large num- 

 ber of varieties were isolated. Some of them differed from any at 

 present described in the literature. 



iCentralbl. Bakt., liHl, 2. abt.^ 30:241. 



2 U. S. Dept. Agr., Bureau of Chemistry Cir. 70. 



