BACTERIAL DISEASE OF TOBACCO. 443. 
periments made later with pure cultures of bacteria 
obtained from diseased tobacco plants, where the reactions. 
of Bact. solanacearum were expected but were not 
obtained, led to the re-testing the Bacterium, No.1, obtained 
from potatoes after it had been kept in the laboratory for 
four months and re-subcultured at long intervals. The 
results were remarkable, the bacterium produced acid but 
no gas in media containing glucose and lactose, while the 
mannite broth was rendered alkaline. Milk was clotted 
with an acid reaction, and litmus milk became partially 
bleached in 17 days. 
These results show that for comparative results cultures. 
of bacteria must be freshly obtained. Hven when bacteria 
are subcultured for a length of time upon media that may 
be regarded as eminently suitable, they may change their 
characters. An example may be given in the case of 
Streptococcus lacticus. In my laboratory a ‘number of 
strains of these organisms are kept differing slightly in the 
rapidity with which they produce lactic acid in milk. Pure 
cultures of these organisms are supplied to butter factories 
to be used as ‘“‘starters’”’ in the ripening of cream. The 
organisms are subcultured every day upon milk. On one 
occasion the milk was found to be slimy and it was thought 
that contamination by another organism had taken place. 
Careful examination and subculturing from old tubes that 
had not been discarded revealed no such contamination. 
The same phenomenon has occurred several times since, 
and there seems no doubt but that after frequent subcul- 
turing, Streptococcus lacticus may cease to give its normal 
reactions with milk and cause it to be become slimy. 
Hasting™) has observed a similar phenomenon in America: 
“Many of the cultures gradually lose their fermentative proper- 
ties and do not form acid rapidly and in sufficient amounts to 
insure exhaustive churning and to produce the desired degree of 
