28 
On the other hand, the following instances are given of experi- 
menters who report a relatively short vitality for the tubercle bacillus 
dried in sputum not exposed to direct sunlight: Cadeac found it 
dead in ten days and sometimes less; Hill, sixteen days; Ransom 
and Delepine, nineteen days; and Galtier, after thirty days. 
The question at issue from a practical standpoint is not how long 
a few T tubercle bacilli may occasionally survive in dried sputum, but 
how soon they usually die or become harmless. In order to determine 
this point much more data will be required than are now available. 
All authorities who have worked upon the subject seem agreed that 
the tubercle bacillus loses its virulence before it finally dies. This is 
also the case with other pathogenic nonspore-bearing bacteria. 
It seems that the more carefully the tubercle bacillus is studied the 
nearer its viabilit}^ is seen to approach that of other bacteria of its 
class. 
CRITERION OF DEATH. 
It is difficult to determine just when a microorganism dies. This 
difficulty is increased in the case of the tubercle bacillus. The reluc- 
tance with which this organism grows upon artificial media makes 
cultural methods totally inapplicable; it is necessary to resort to 
animal inoculations. Here again we meet an unusual difficulty, for 
the dead tubercle bacillus may produce lesions indistinguishable from 
those which are caused by live cultures. 
A study of the literature shows that much of the work upon the 
viability of the tubercle bacillus has little value for lack of a clear 
criterion of death. Many experiments will be found recorded in 
which the tubercle bacillus is exposed to various influences, such as 
drying, heat, chemicals, sunlight, putrefaction, etc., and after a 
certain period of time the dried material is inoculated into rabbits 
or guinea pigs. In about two months after the inoculation the 
animal is usually killed, and if caseous abscesses and other lesions 
about the site of inoculation are found and the tubercle bacillus is 
rediscovered on stained smears made from the local lesions, the 
conclusion is drawn that the tubercle bacillus is alive, and the results 
are recorded as positive. 
It is now well known that the dead tubercle bacilli may cause 
lesions characterized by giant cells, caseous necrosis, etc. I have 
found that moderately large amounts of tubercle cultures surely 
killed by heat (100° for one hour) may cause very extensive lesions 
when injected into the peritoneal cavities of guinea pigs. So closely 
do these lesions sometimes resemble those produced by the living 
bacteria that unless the portions of the spleen, liver, lungs, glands, 
and local processes are inoculated into another animal it is not pos- 
sible to know whether we are dealing with the effects of living or of 
dead bacteria. When the lesions are plainly generalized through- 
