present.
Extended researches of this kind also gave a negative result, which in
my final report I stated as follows:

The specific cause of yellow fever has not yet been demonstrated.

It is demonstrated that microorganisms, capable of development in
the culture media usually employed by the bacteriologists, are only
found in the blood and tissues of yellow fever cadavers in
exceptional cases, when cultures are made very soon after death.

Since this report was made, various investigators have attacked the
question of yellow fever etiology, and one of them has made very
positive claims to the discovery of the specific germ. I refer to the
Italian bacteriologist, Sanarelli. His researches were made in Brazil,
and, singularly enough, he found in the blood of the first case examined
by him a bacillus. It was present in large numbers, but this case
proved to be unique, for neither Sanarelli nor any one else has since;
found it in such abundance. It has been found in small numbers in the
blood and tissues of yellow fever cadavers in a certain number of the
cases examined. But carefully conducted researches by competent
bacteriologists have failed to demonstrate its presence in a
considerable proportion of the cases, and the recent researches of Reed,
Carroll, and Agramonte, to which I shall shortly refer, demonstrate
conclusively that the bacillus of Sanarelli has nothing to do with the
etiology of yellow fever.

So far as I am aware, Dr. Carlos Finlay, of Havana, Cuba, was the first
to suggest the transmission of yellow fever by mosquitoes. In a
communication made to the Academy of Sciences of Havana, in October,
1881, he gave an account of his first attempts to demonstrate the truth
of his theory. In a paper contributed to _The Edinburgh Medical Journal_
in 1894, Dr. Finlay gives a summary of his experimental inoculations up
to that date as follows:

A summary account of the experiments performed by myself (and some also
by my friend, Dr. Delgado), during the last twelve years, will enable
the reader to judge for himself. The experiment has consisted in first
applying a captive mosquito to a yellow fever patient, allowing it to
introduce its lance and to fill itself with blood; next, after the lapse
of two or more days, applying the same mosquito to the skin of a person
who is considered susceptible to yellow fever: and, finally, observing
the effects, not only during the first two weeks, but durin