EFFECT OF TEMPERATURE ON BACTERIA. 59 
Milk kept at 37°.—The results were almost a repetition of 
those of the last experiment, inasmuch as at the 16th hour, at 
the time of curdling, nearly 98 per cent. of the bacteria growth 
was confined to Group III., 2. aerogenes, this group at the end 
constituting 98 per cent. of the whole. The other 2 per 
cent. consisted of a few of the neutral Group IV. and a few 
liquefiers. Here, as in experiment 3, the temperature of 37° 
stimulated particularly the development of Bacterium aerogenes, 
and this resulted in the practical disappearance of nearly all 
other species, even the very vigorous Group I. failing to 
develop in this sample of milk kept at 37°. The total number 
of bacteria present at curdling was far less than in the 20° 
sample. 
EXPERIMENT NO. 5. MAY IQ. 
The conditions of this experiment were as in the last three, 
but the weather was warm and dry. The milk kept at 37° 
curdled in 18 hours, that kept at 20° curdled in about 45 hours 
(in the night), and that kept at 10° curdled in 312 hours, 
although no analysis was made later than 260 hours. ‘The fol- 
lowing tables, 13, 14, and 15, show the details of the analysis, 
and Plate 4 shows graphically the growth of the chief types. 
The results of this experiment were so similar to the last 
that only a brief mention of the essential points is necessary. 
Milk kept at ro°.—All species of bacteria originally present 
developed uniformly, except Group IV., and all were present 
in a large percentage, even in the last test, made at the 266th 
hour. The chief growth was due to Group IV. and Group I. 
Group IV. developed to a phenomenally high number, the 
analysis at the 218th hour showing 3,500,000,000 per cubic 
centimeter, 94 per cent. of all the bacteria present at that time 
being of this type. In spite of these high numbers the milk 
did not curdle. After this period the lactic bacteria increased 
in numbers and percentage, while the neutral Group IV. de- 
creased in both numbers and percentage. At the time of curd- _ 
ling these two groups were present in about equal numbers, 
shown graphically in Plate 4. 
Group III. showed a constant development in numbers through 
the whole experiment until the last two analyses, when they 
noticeably declined in numbers. 
