368 
ALFRED L. LOOMIS. 
be. If a tube shows signs of admixture, it is discarded at once 
and another trial made. 
Negeli has proposed a “ dilution method ” for obtaining “ pure 
cultures.” He had some urine containing large micrococci which 
he wished to cultivate. The urine also contained numerous 
bacilli. A single drop of the urine, supposed to contain five 
hundred thousand bacteria, was mixed with about two ounces of 
pure water and shaken thoroughly. This dilutes the urine a 
thousand times. One drop of this is now mixed with another 
two ounces of water, and a millionth dilution was obtained, in 
which every drop must average to contain one bacterium. He 
inoculated ten tubes of gelatine each with one drop of this dilu¬ 
tion. After incubation it appeared that four tubes remained 
sterile, one contained bacilli, and five the desired cocci. 
You will find in our biological laboratory the modern appli¬ 
ances for sterilizing the fluids, vessels, etc., as well as several in¬ 
cubating ovens, the temperature of which may be maintained at 
a given point for months. The large incubator constructed by 
Dr. Miller will accommodate several thousand culture-tubes at 
once. 
We have, then, finished the second step in our work, that is, 
we have obtained our bacilli pure, and have cultivated them 
through several generations, and it is improbable that our culture- 
tubes contain any contamination from the original source of our 
supply of bacteria. 
It remains for ns to complete the work, that is, we must now 
reproduce the disease by inoculation with our pure culture. 
Now we are forced to resort to the lower animals for experiment. 
No one, as yet, has sufficient admiration for science to inoculate 
himself. Hence it is impossible to complete a perfect chain of 
evidence. Objection can be reasonably made to the employment 
of mice and guinea-pigs, ft is just possible that deductions from 
such experiments would not hold witli human beings. But there 
seems to be no alternative. We proceed to inoculate several guinea- 
pigs with our pure culture of tubercle bacilli. The following are 
the results as first obtained by Koch, and which are almost iden¬ 
tical with the phenomena obtained in our own work. The 
