298 MOSQUITOES OF NORTH AMERICA 
greatest number of coffee trees have been planted. Now suppose for a moment 
that there was a complete disappearance of all historical documents proving the 
Asiatic origin of the coffee plant and of its introduction into Brazil via Cayenne, 
who would imagine that the original country of this tree was Arabia? And is 
it not an absolute fact that this very historical remembrance is becoming weaker 
every day in popular knowledge here in Brazil with the increasing interval of 
time separating us from the moment of introduction which nevertheless was 
scarcely 200 years ago?” 
One of the authors (Knab) has already pointed out some of the weaknesses in 
Goeldi’s argument. He showed that Goeldi is in error where he denies the 
existence of large cities in America at the time of its discovery. The vast 
centers of population found by the Spanish conquerors in Mexico and Yucatan 
are too familiar to need discussion. Even the Amazons themselves, if we can 
believe the early explorers, had large towns upon their shores and supported a 
considerable population. Moreover it must be remembered that Africa is for 
the most part a dry, open country in which a mosquito with the habits of Aédes 
calopus could hardly originate and find opportunities to adapt itself to man. 
On the other hand in tropical America the ideal conditions exist which would 
bring about the adjustment of such a mosquito to man. We find extensive 
forests and plenty of moisture to support tree-hole breeding mosquitoes and 
human settlements within these forests to furnish the blood supply. In fact 
we find in this region a great variety of mosquitoes which have the tree-hole 
breeding habit; among these are some which clearly show a predilection for the 
proximity of man and may thus be said to be in progress of developing do- 
mesticity in the manner it must have been done by calopus. 
Had yellow fever existed in Africa prior to the discovery of America we would 
surely have had some indication of it through the earlier explorers, but nothing 
exists to warrant such a belief. The distribution of yellow fever in Africa, as has 
been recently set forth anew by Boyce, is in itself significant. It has there been 
confined to spots in a restricted area on the west coast, the part nearest America 
and formerly most in communication with it through the slave trade. This is 
well brought out by Prof. W. J. Simpson, in his discussion of the paper just 
mentioned, by Boyce. It is hardly conceivable, had yellow fever been endemic 
in Africa for a long time, that it could have failed to spread, not only over the 
hotter parts of the continent, but even to the Kast Indies. Finally the experi- 
ments of Marchoux and Simond, given on a previous page, which indicate a 
certain repugnance to the black race, while not in themselves conclusive, in con- 
nection with the foregoing, are at least significant. 
DISTRIBUTION BY ARTIFICIAL MEANS. 
It seems worth while to give some practical account, in the light of our present 
knowledge, of the way in which the yellow-fever mosquito is distributed by the 
presentation of specific cases. To illustrate shipboard carriage the story of the 
case of the steamer Luzor and its journey in the spring of 1906 may be de- 
scribed as follows: 
