BACTERIOLOGY: E. O. JORDAN 
163 
SO that the results as recorded range from a mere trace to the maximum 
intensity observed in any culture. The tests were always carried out on 
the same lot of culture medium and were controlled by a number of cul- 
tures (usually about 20) from the same strains but of different cultural 
histories. Milk was always rendered acid but sometimes not coagulated 
in forty-eight hours. The milk used was certified milk always obtained 
from one dealer and treated in as uniform a manner as possible through- 
out. Nevertheless, differences in rapidity and completeness of coagu- 
lation were so great in different lots of milk as to make such changes 
of little value in attempts to study variation in the physiological proper- 
ties of the organisms themselves. 
The most fundamental change thus far observed is the acquisition of 
saccharose-fermenting powers by one of the pure line strains. This 
quaHty appeared in the seventh transfer on sodium chloride agar. It 
was not manifested by all the cells of the culture, but at the time of ex- 
amination the saccharose-fermenting cells were greatly in the majority. 
On continuing the transfers they became the sole type found in the cul- 
tures, the non-saccharose type disappearing altogether. The power of 
fermenting raffinose was also possessed by the saccharose-fermenting 
strains. Gas is produced in both saccharose and raflfinose solutions and 
over 4% of normal acid is formed. The newly acquired fermenting prop- 
erty has remained permanent throughout a series of over 500 test-tube 
generations (forty-eight-hour transfers), and is shown both by the cul- 
tures on sodium chloride media and by the strains transferred imme- 
diately on the acquisition of this property to ordinary nutrient agar and 
grown side by side with the two parent strains. 
This instance of bacterial mutation therefore seems to fulfil the require- 
ments (a) of appearing suddenly without intermediate stages, (b) of being 
irreversible, at least for three years and for some hundreds of test-tube 
generations, (c) of comprising change in two characters (saccharose and 
raffinose-fermenting power), and (d) of not involving all the cells of the 
parent strain. 
It may be remarked that a differentiation of B.coli into species is com- 
monly made on the basis of the power to ferment saccharose and most 
of the recent classifications of B.coli groups start with this as a funda- 
mental distinction. In the great majority of strains of B.coH that have 
been tested by various observers saccharose fermentation is correlated 
with raffinose fermentation. Saccharose-fermenting streptococci on the 
other hand are often devoid of power to ferment raffinose. 
