635 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1907. 
two weeks, when the changes described in the early parts of this 
address manifest themselves. After a few weeks general inflamma- 
tion of the structures of the eye develops, the inoculation wound 
becomes cheesy and the eye is more or less completely destroyed. 
The disease, however, remains usually localized in the eye for many 
months, and may remain there permanently, depending upon the 
virulence and number of bacilli injected. 
Tn the vaccinated animals, on the contrary, the introduction of the 
mammalian bacilli at once gives rise to a marked degree of irritation. 
From the second to the fifth day the vessels of the conjunctiva become 
engorged, and evidences of marked inflammation appear in the an- 
terior chamber and on the iris (reaction of immunity). However, 
at the end of the second to the third week, when the eyes of the con- 
trols begin to show progressive and steadily increasing evidence of 
inflammatory reaction, the irritation in those of the vaccinated ani- 
mals begins slowly to subside and the eyes to mend. In from six 
to twelve weeks, in the successful cases, all irritation has disappeared 
and the eyes present only the evidences of traumatism and inflamma- 
tion. This experiment leaves no doubt of the protective influence 
exerted by the first inoculations of the avian bacilli and clearly estab- 
lishes that related cultures of tubercle bacilli of moderate virulence 
for an animal species can afford protection to subsequent inoculation 
with special and more pathogenic strains of the bacillus. Notwith- 
standing the fact that, as Trudeau records, some of the protected 
animals slowly relapse and the disease resumes its progress, although 
by almost imperceptible stages, the experiment still shows that pro- 
tection, not absolute immunity, from tuberculosis may be obtained in 
rabbits by a species of vaccination. 
De Schweinitz in 1894 reported certain experiments which he made 
on guinea pigs and cattle. He inoculated the former with a culture 
of tubercle bacilli of human origin cultivated for about twenty gen- 
erations in broth. This culture was of a low grade of virulence for 
these animals, but it served to protect them to such an extent that when 
they were afterwards inoculated with tuberculous material from a 
cow they remained healthy, while control pigs injected with the same 
material became tuberculous and succumbed in about seven weeks. 
De Schweinitz injected large quantities of human tubercle bacilli into 
cattle—beneath the skin, into the peritoneal cavity and into the circu- 
lation—without injury. 
I may, at this time, digress for a moment and leave the more 
strictly chronological method of presentation to allude to the set of 
experiments on the protection of guinea pigs from tuberculosis which 
Trudeau reported to the National Tuberculosis Association at its last 
meeting. The special merit of this experiment is that it shows the ex- 
istence of a connection between virulence and infectivity in the germ 
