130 DISCUSSION
How could they explain the local limitation of cutaneous
inoculations in diseases such as variola, anthrax, tuber-
culosis and kala-azar ? It had been suggested that the
absence of tendency to generalization might be due to the
special structure of the skin or to its lower temperature,
and experiments had proved that adipose tissue and dense
connective tissue were unfavourable to the propagation
of pathogenic organisms. To these reasons others might
be added. Thus it was quite likely that, in the case of
pathogenic agents which had a wide zoological distribu-
tion, a different zoological source might account for dif-
ferences in virulence and location. That belief wa's
founded not only on the observation of natural facts, but
also on the result of experimental investigation. Bacte-
riologists were constantly either exalting or attenuating
the virulence of disease-producing microbes by successive
passages through the organisms of different animals, and
they had noticed again and again that microbes obtained
from a diseased organ, when introduced into other
animals, usually localized themselves in that part which
had been attacked in the first subject.
In the case of Leishmaniasis, they knew that the infec- tion had been noticed in horses and dogs. In the dog, the cutaneous form had been early recognized in India, Persia, and Syria ; it presented itself almost invariably on the bare tip of the nose. The general infection had been demonstrated only quite recently by Nicolle in Tunisian dogs. Nicolle had not only found the " Leishraan- Donovan bodies " in the spleen and bone-marrow of stray local dogs, but had succeeded in reproducing the infection in other dogs by means of the inoculation of infected splenic blood into the liver and peritoneal cavity of normal puppies. Nicolle's experiments presented the objection of being carried out in a place in which dogs
