59 
The filtrate showed no growth in bouillon, and yet when injected 
into nonimmunes produced yellow fever. The blood from the latter 
was also shown to be capable of producing yellow fever when injected 
into a third subject. 
That the men inoculated with the filtrate suffered from yellow fever 
induced by a morphologic entity which passed the filter, and not from 
a toxemia, was shown not only by their rather long periods of incu¬ 
bation, but was conclusively shown by carrying their experiment to 
the third degree. 
The following experiments were planned in order to determine 
among other things whether the organism of yellow fever, as it exists 
in the blood serum, is capable of passing the pores of the Pasteur- 
Chamberland B filter: 
An investigation of the literature of the other filterable viruses 
shows that the South African horse sickness is the only one which has 
yet been reported as having passed the Chamberland B filter. 
In the filters of the Pasteur-Chamberland system those marked 
44 B ” are finer, more compact, with thicker walls, and consequently 
less porous than those marked F. We have been informed by 
Assistant Surgeon-General PI. D. Geddings, who has recently inquired 
about this in Paris, that only two grades—B and F—are now being 
made. 
The subjects used for our experimentation were all volunteers, non¬ 
immunes, and carefully selected from among the native Mexicans at 
Jalapa and the adjacent mountainous country, taken by train to Vera 
Cruz, and immediately placed within the screened wards of our hos¬ 
pital. All cases recovered. 
Jalapa is a town having an elevation of about 4,000 feet, where yel¬ 
low fever has never been known to spread and has not existed, except 
for the cases occasionally imported from the coast (tierra caliente). 
In order that the case from which we drew the blood for filtration 
should be one in which there was the highest degree of confidence as 
to the diagnosis of yellow fever, we decided to produce the disease 
through the bites of infected mosquitoes rather than to select a case 
by clinical evidence alone from the yellow-fever wards. 
Mosquitos which had been allowed to feed upon typical cases of 
yellow fever in San Sebastian Hospital, Vera Cruz^ were applied in 
succession to the hands of four persons whom we had selected as 
being nonimmunes. The first three failed to become infected, but 
the fourth took sick with what proved typical yellow fever. The 
histories of the three negative cases are here given in brief: 
G. M ., age Mexican .—On August 13 he was taken to Vera Cruz 
and placed in our screened ward. August 15, at 3.20 p. m., he was 
bitten by two mosquitoes which had fed twelve days previously, at 
9.30 a. m., on J. R., a fatal case of yellow fever. Nothing unusual 
13046—05 m- 3 
