ALKALI-FORMING BACTERIA FOUND IN MILK. 11 
white growth was its consistence, about 60 per cent of the cultures 
producing a slimy, viscous growth that was almost impossible to 
remove with a needle. In litmus milk these cultures produced a 
typical blue coloration which usually appeared in from 3 to 5 days 
when incubated at 30° C. Only 3 cultures out of the 68 were found 
to liquefy gelatin on incubation for 45 days at 20° C. The thermal 
death point of the alkali-forming organisms varied from 60° to 65.6° 
C, when heated in milk tubes for half an hour. 
SOURCES OF NITROGEN. 
From preliminary results of fermentation tests with the alkali- 
forming bacteria, it was evident that nitrogen should be supplied in 
some simple form, free from carbon. It was found that sodium- 
ammonium phosphate could be used as a source of nitrogen. This 
compound gave such satisfactory results that it was used through- 
out the work, carbon being supplied by other substances, as carbo- 
hydrates or organic acid salts. The alkali-forming bacteria were 
also able to utilize nitrogen from sodium nitrate and sodium nitrite, 
when a suitable source of carbon was present. With these two sources 
of nitrogen uniform results could not be obtained, for often a culture 
which would grow well on one inoculation would fail to develop at 
another time. The more complex nitrogen-containing substances 
were not suitable for fermentation tests, for reasons which will be 
•discussed. All the organisms studied grew readily with asp ar agin as 
a source of nitrogen. It was not satisfactory, however, because 
ammonia was liberated, which interfered with the measurement of 
the fermentation test substances. Similar objections also apply to 
the use of beef extract. This not only supplied nitrogen but 
carbon as well. All the cultures grew vigorously in a solution of 
0.2 per cent beef extract and 0.2 per cent sodium-dibasic phosphate 
and produced an alkaline reaction due probably to the formation of 
alkaline carbonates from the organic-acid salts present in the extract. 
It was, therefore, quite evident that beef extract was not a suitable 
source of nitrogen for use in fermentation tests when a definite fer- 
mentation was to be measured. 
Other nitrogen-containing substances, such as peptone, gelatin, 
casein, and urea were good sources of nitrogen when carbon was sup- 
plied in some other form, but as sources of both nitrogen and carbon 
they were unsatisfactory. In other words, they are good sources of 
nitrogen but poor sources of carbon. 
In studying the physiology of bacteria, it is believed that the 
sources of nitrogen should be given considerable thought, and when- 
ever possible a simple compound free from carbon should be selected. 
For the alkali-forming bacteria sodium-ammonium phosphate ful- 
filled this requirement. 
