130 BACTERIOLOGY OF WATER, AIR, MILK, ETC. 



cloaca groups liquefy gelatin; consequently, if the proportion of liquefy- 

 ing to nonliquefying be greater than as i to 10, the water is considered 

 suspicious. The test is not considered by American authorities as of 

 any particular value. 



The American Public Health Association recognizes the importance 

 of the information obtained from a comparison of the number of organ- 

 isms developing at 38 C. and those developing at 20 C. Bacteria 

 whose normal habitat is the intestinal canal naturally develop well at 

 body temperature, while normal water bacteria prefer the average 

 temperature of the water in rivers and lakes. Consequently when the 

 number of organisms developing at 38 C. at all approximates the num- 

 ber developing at 20 C., there is a strong suspicion that sewage or- 

 ganisms may be present. Normal waters give proportions of i to 25 

 or i to 50, while in sewage contaminated waters the proportion may be 

 as i to 4 or less. 



In addition, the appearance of pink colonies on the lactose litmus 

 agar is of great assistance in judging of a water. Both sewage strepto- 

 cocci and the colon bacillus give pink colonies 'those of the streptococci 

 are smaller and more vermilion in color. Microscopic examination 

 will differentiate the cocci from the bacilli. It is well to bear in mind 

 that the pink colonies after twenty-four hours may turn blue in forty- 

 eight hours from the development of ammonia and amines. Conse- 

 quently the lactose litmus agar plates should be studied after twenty- 

 four hours. 



A good water supply will rarely show a pink colony, while in a sew- 

 age-contaminated one the pink colonies will probably predominate. 



The diagnostic characteristics considered important by the Ameri-' 

 can authorities in reporting the colon bacillus (Recently designated 

 excretal colon bacillus) are: 



1. Typical morphology, nonsporing bacillus, relatively small and often quite 

 thick. 



2. Motility in young broth cultures. (This is at times unsatisfactory, as some 

 strains of the colon bacillus do not show it even in young bouillon cultures). 



3. Gas formula in dextrose broth. Of about 50% of gas produced, one- third 

 should be absorbed by a 2% solution of sodium hydrate (CC^). The remaining gas 

 is hydrogen. (Later views indicate that the gas formula is exceedingly variable and 

 should not be depended upon. To carry out this test one fills the bulb of a fermen- 

 tation tube with the caustic soda solution, holding the thumb over the opening 

 or witih a rubber stopper, the bouillon culture and the soda solution are mixed 

 by tilting the fermentation tube to and fro. The total amount of gas is first re- 



