American Public jfijealtl) Association 
THE ETIOLOGY OF YELLOW FEVER.—A PRE- 
LIMINARY NOTE. 
By WALTER REED, M. D., Surgeon, U. S. A.; 
JAMES CARROLL, M. D.: A. AGRAMONTE, M. D., and 
JESSE W LAZEAR, M. D., 
Acting Assistant Surgeon, U. S. A 
Imprinted from the ’Proceedings of the Twenty-Eighth Annual Meeting, held at 
Indianapolis, Indiana, October 22, 22, 24, 25 and 26, 1900. 
COLUMBUS, OHIO: 
THE BERLIN PRINTING COMPANY. 
1901. American public fijealtl) Association 
THE ETIOLOGY OF YELLOW FEVER.—A PRE- 
LIMINARY NOTE. 
By WALTER REED, M. D., Surgeon, U. S. A.; 
JAMES CARROLL, M. D.: A. AGRAMONTE, M. D., and 
JESSE W. LAZEAR, M. D., 
Acting Assistant Surgeon, U. S. A. 
from the 'Proceedings of the Twenty-Eighth Annual Meeting, held at 
Indianapolis, Indiana, October 22, 23, 24, 25 and 26, 1900. 
THE BERLIN PRINTING COMPANY. 
1901. 
COLUMBUS, OHIO: THE ETIOLOGY OF YELLOW FEVER.—A PRELIMINARY 
NOTE. 
WALTER REED, M. D., Surgeon, U. S. A.; 
JAMES CARROLL, M. D.; A. AGRAMONTE, M. D„ and 
JESSE W. LAZEAR, M. D.,i 
Acting Assistant Surgeon, U. S. A. 
The writers, constituting a board of medical officers convened “for 
the purpose of pursuing scientific investigations with reference to the 
acute infectious diseases prevalent on the Island of Cuba,” arrived 
at our station, Columbia Barracks, Quemados, Cuba, on June 25th 
of the present year, and proceeded, under written instructions from 
the Surgeon General of the Army, to “give special attention to ques- 
tions relating to the etiology and prevention of yellow fever.” 
Two of its members (Agramonte and Lazear) were stationed on 
the Island of Cuba, the former in Havana, and the latter at Columbia 
Barracks, and were already pursuing investigations relating to the 
etiology of this disease. 
Fortunately for the purposes of this board, an epidemic of yellow 
fever was prevailing in the adjacent town of Quemados, Cuba, at the 
time of our arrival; thus furnishing us an opportunity for clinical ob- 
servations and for bacteriological and pathological work. The re- 
sults already obtained, we believe, warrant the publication at this 
time, of a preliminary note. A more detailed account of our obser- 
vations will be submitted to Surgeon General Sternberg in a future 
report. 
The first part of this preliminary note will deal with the results of 
blood cultures during life and of cultures taken from yellow fever 
cadavers; reserving for the second part a consideration of the mos- 
quito as instrumental in the propagation of yellow fever; with obser- 
vations based on the biting of non-immune human beings by mos- 
quitoes which had fed on patients sick with yellow fever, at various 
intervals prior to the biting. 
In prosecuting the first part of our work, we isolated a variety of 
bacteria, but of these we do not propose to speak at present. It will 
i Died of yellow fever at Columbia Barracks, Cuba, Sept. 25,1900. 4 
THE ETIOLOGY OF YELLOW FEVER. 
suffice for our purpose if we state the results as regards the finding of 
bacillus icteroides, leaving the mention of other bacteria to our de- 
tailed report. 
The cases studied during the Quemados epidemic had been diag- 
nosed by a board of physicians, selected largely by reason of their 
familiarity with yellow fever. This board consisted of Doctors 
Nicola Silverio, Manuel Herera, Eduardo Angles and Acting Assist- 
ant Surgeons Roger P. Ames and Jesse W. Lazear, U. S. Army. 
Those studied in Havana were patients in Las Animas Hospital 
and had been diagnosed as such by a board of distinguished practi- 
tioners of that city. 
An examination of Table I will show the character of the attacks. 
The milder cases studied, few in number, were attended by jaundice 
and albumen in the urine. 
I. 
Bacillus Icteroides (Sanarelli) as the Cause of Yellow Fever. 
The claim of Sanarelli for the specific character of B. icteroides as 
the causative agent in yellow fever, has excited such wide attention, 
since the publication of his observations, that it seemed to us of the 
first importance to give our undivided attention to the isolation of 
this micro-organism from the blood of those sick with yellow fever 
and from the blood and organs of yellow fever cadavers. 
(A) Cultures Taken from the Blood During Life. 
The method followed was that ordinarily used in an attempt to 
isolate bacteria from the circulating blood; viz., from a vein at the 
bend of the elbow, a sufficient quantity of blood was taken with an 
hypodermic syringe, made sterile by boiling, and after careful cleans- 
ing of the skin with soap and water, followed by equal parts of abso- 
lute alcohol and ether, and 1-2000 bichloride solution. 
Exceptionally, the blood withdrawn was plated on agar, but as a 
rule, it was immediately transferred to sterile bouillon tubes (10 c. 
cm.) in quantities of 0.5 c. cm. to each of several tubes. These were 
then incubated at 35° to 37° C. for a period of one week. They were 
examined daily and if growth was observed, plates in agar or gela- 
tine, or both, were made and the colonies carefully studied by trans- 
ference to ordinary laboratory media. 
Eighteen cases have thus been carefully studied; of these, eleven 
were designated as “severe” cases of yellow fever with four deaths; 
three as “well-marked” cases with no deaths, and four as “mild” cases 
with no deaths. 
From these eighteen cases, blood cultures were made, as shown in 
the following table: THE ETIOLOGY OF YELLOW FEVER. 
5 
TABLE I. 
Blood Cultures During Life. 
Day of 
Disease. 
Character of 
Attack. 
No. of 
Cultures. 
No. of Bouillon Tubes 
Inoculated. 
B. Icteroides. 
1st. 
Severe. 
3 
4 (3 agar plates.) 
Negative. 
4 4 
Well-marked. 
1 
4 
44 
a 
Mild. 
1 
3 
44 
2nd. 
Severe. 
6 
18 
44 
44 
Well-marked. 
1 
2 
44 
4 4 
Mild. 
1 
3 (6 agar plates.) 
4 < 
3rd. 
Severe. 
7 
18 (6 agar plates.) 
44 
44 
Mild. 
2 
4 
44 
4th. 
Severe. 
5 
14 
44 
4 4 
Well-marked. 
2 
6 
44 
44 
Mild. 
1 
1 
44 
5th. 
Severe. 
5 
12 (3 agar plates). 
44 
“ 
Well-marked. 
1 
3 
44 
44 
Mild. 
1 
1 
44 
6th. 
Severe. 
4 
6 
44 
“ 
Well-marked. 
1 
2 
44 
7th. 
Severe. 
1 
2 
44 
44 
Well-marked. 
1 
2 
44 
8th. 
Severe. 
2 
6 
44 
4 4 
Well-marked. 
1 
2 
44 
9th. 
Severe. 
1 
2 
44 
Number of cultures 
... 48 
“ “ bouillon tubes inoculated 
... 115 
“ “ agar plates 
... 15 
It will be seen that of 48 separate cultures made from the blood on 
various days of the disease and representing 115 bouillon inocula- 
tions and 18 agar plates, we failed to find bacillus icteroides in any 
of our tubes or plates. 
The results of cultures taken in 181 cases of unmistakable yellow 
fever, on various days of the disease and in some cases on every day 
from the onset to death or recovery, would seem to exclude the 
presence of bacillus icteroides in the blood of these cases during life. 
It will, therefore, be seen that while Wasdin and Geddings taking 
cultures from the ear-lobe (Report on the Cause of Yellow Fever, 
1899), record that “in the blood of yellow fever cases extracted dur- 
ing life bacillus icteroides has been found in thirteen of the fourteen 
cases, with one negative,” (92.85 per cent.) we, by withdrawing blood 
from the veins of 18 patients, have to record 100 per cent, of failures. 
l Cultures fr«m tbe blood during life had been taken by Dr. Lazear in three other case* 
of yellow fever, but owing- to the death of our colleague, the necessary data as to the day of 
the disease on which cultures had been taken cannot be ascertained. 
These cultures were negative as regards the finding of Sanarelli’s bacillus. 6 
THE ETIOLOGY OF YELLOW FEVER. 
We have already stated that we will reserve for a later report a de- 
scription of the bacteria isolated from the blood in these cases. We 
now remark that but few organisms were obtained and that, as a rule, 
our blood cultures gave no growth whatever. 
(B) Cultures from Yellow Fever Cadavers. 
We tried to obtain autopsies very soon after death and sometimes 
succeeded in doing so. Tubes containing about 10 c. cm. of flesh- 
peptone bouillon were generally used for the first inoculation direct 
from the blood and organs. As soon as the laboratory was reached, 
agar plates were made from these inoculated bouillon tubes, the 
former as well as the latter being then incubated at 35°-37° C. In 
nearly every case gelatine plates were, also, made from the recently 
inoculated bouillon tubes and kept at a temperature of 19°-20° C. 
If colonies were found in the agar or gelatine plates, on the fol- 
lowing days, the corresponding bouillon tubes were, also, plated on 
agar and gelatine. The bacteria thus found in our plates were care- 
fully isolated and studied upon the usual nutritive media, so as to 
enable us to identify them, if possible. We will here content ourselves 
with giving the results as regards the presence of B. icteroides only. 
TABLE II. 
No. of 
Case. 
Day of 
Disease. 
Time of 
Autopsy. 
Source of Culture. 
B. Icter- 
oides. 
1 
7th. 
2 hours after 
death. 
Blood, liver, spleen, kidney. 
Neg’tive 
2 
6th. 
13 hours after 
death. 
Blood, liver, spleen, kidney. 
a 
3 
4th. 
8 hours after 
death. 
Blood, liver, spleen, kidney. 
a 
4 
8th. 
4 hours after 
death. 
Abdominal cavity, blood, liver, 
spleen, kidney, bile, duodenum. 
a 
5 
4th. 
4 hours after 
death. 
Blood, liver, spleen, kidney, bile, 
duodenum. 
u 
6 
6th. 
6% hours 
after death. 
Abdominal cavity, blood, peri- 
cardial fluid, lung-, spleen, kid- 
ney, liver, bile, duodenum. 
a 
7 
6th. 
50 minutes 
after death. 
Blood, lung, liver, spleen, kid- 
ney, bile, jejunum. 
i i 
8 
6th. 
j4-hour after 
death. 
Blood, lung, liver, spleen, kidney, 
urine, small intestine. 
a 
9 
4th. 
2 hours after 
death. 
Liver, spleen, small intestine. 
u 
10 
5th. 
7 hours after 
death. 
Liver, kidney, spleen, small in- 
testine. 
a 
11 
3rd. 
after 
death. 
Liver, kidney, spleen. 
*4 
Our failure to isolate B. icteroides in these eleven autopsies of 
yellow fever patients was a result which we had not anticipated. One THE ETIOLOGY OF YELLOW FEVER. 
7 
of us (Agramonte) who, at Santiago, Cuba, during the epidemic of 
1898, succeeded in finding B. icteroides in 33 per cent, of his autop- 
sies, has been much surprised at the absence of this bacillus in cul- 
tures from cadavers sectioned in Havana, during the present year. 
In two of the eleven cases we had reason to believe that from the 
character of colonies seen in gelatine plates, we would be able to 
isolate B. icteroides. These colonies, however, when transferred to 
other media and carefully studied, did not prove to be this bacillus. 
We wonder whether other observers have occasionally relied upon 
the appearance of colonies in gelatine plates, without further study. 
We only mention this as a possible explanation of the large percent- 
age of positive results recorded by some observers. 
Portier, of New Orleans, La., only succeeded, however, in isolating 
B. icteroides in three out of fifty-one autopsies (Journal of American 
Medical Association, April 16, 1896). 
Lutz (Revista D’lgiene e Sanita Publica, XI, No. 13, July, 1900, 
pp. 474-475), says as the result of his extensive observations on yel- 
low fever that bacillus icteroides cannot be found by present labora- 
tory methods in more than half of the cases of yellow fever, and that 
when present, the colonies are few in number. 
It is possible that our future autopsies may give more favorable 
results as regards B. icteroides. 
II. 
The Mosquito as the Host of the Parasite of Yellow Fever. 
Having failed to isolate B. icteroides either from the blood during 
life or from the blood and organs of cadavers, two courses of pro- 
cedure in our further investigations appeared to be deserving of atten- 
tion, viz., first, a careful study of the intestinal flora in yellow fever 
in comparison with the bacteria that we might isolate from the intes- 
tinal canal of healthy individuals, in this vicinity, or of those sick with 
other diseases ; or, secondly, to give our attention to the theory of the 
propagation of yellow fever by means of the mosquito,—a theory 
first advanced and ingeniously discussed by Dr. Carlos J. Finlay, of 
Havana, in 1881, (Anales de la Real Academia, Vol. 18, 1881, pp. 
147-169). 
We were influenced to take up the second line of investigation by 
reason of the well known facts connected with the epidemiology of 
this disease and, of course, by the brilliant work of Ross and the 
Italian observers in connection with the theory of the propagation of 
malaria by the mosquito. 
We were, also, much impressed by the valuable observations made 
at Orwood and Taylor, Miss., during the year 1898, by Surgeon 8 
THE ETIOLOGY OF YELLOW FEVER. 
Henry R. Carter, U. S. Marine Hospital Service. A note on the in- 
terval between infecting and secondary cases of yellow fever, etc. 
(Reprint from New Orleans Medical Journal, May, 1900). We do not 
believe that sufficient importance has been accorded these painstak- 
ing and valuable data. We observe that the members of the yellow 
fever commission of the Liverpool School of Tropical Medicine, Drs. 
Durham and Meyers, to whom we had the pleasure of submitting 
Carter’s observations, have been equally impressed by their impor- 
tance (British Medical Journal, September 8, 1900, pp. 656-7). 
The circumstances under which Carter worked were favorable for 
recording with considerable accuracy the interval between the time 
of arrival of infecting cases in isolated farm-houses and the occur- 
rence of secondary cases in these houses. According to Carter “the 
period from the first (infecting) case to the first group of cases in- 
fected, at these houses, is generally from two to three weeks.” 
The houses having now become infected, susceptible individuals 
thereafter visiting the houses for a few hours, fall sick with the dis- 
ease in the usual period of incubation one to seven days. 
Other observations made by us since our arrival confirmed Carter's 
conclusions, thus pointing as it seemed to us to the presence of an 
intermediate host, such as the mosquito, which having taken the 
parasite into its stomach, soon after the entrance of the patient into 
the non-infected house, was able after a certain interval to reconvey 
the infecting agent to other individuals, thereby converting a non- 
infected house into an “infected” house. This interval would appear 
to be from nine to sixteen days (allowing for the period of incuba- 
tion), which agrees fairly closely with the time required for the pass- 
age of the malarial parasite from the stomach of the mosquito to its 
salivary glands. 
In view of the foregoing observations we concluded to test the 
theory of Finlay on human beings. According to this author’s ob- 
servation of numerous inoculations in 90 individuals, the application 
of one or two contaminated mosquitoes is not dangerous, but fol- 
lowed in about 18 per cent, by an attack of what he considers to be 
very benign yellow fever at the most. 
We here desire to express our sincere thanks to Dr. Finlay, who 
accorded us a most courteous interview and has gladly placed at our 
disposal his several publications relating to yellow fever, during the 
past nineteen years; and also for ova of the species of mosquito with 
which he had made his several inoculations. An important observa- THE ETIOLOGY OF YELLOW FEVER. 
9 
tion to be here recorded is that, according to Finlay’s statement, 30 
days prior to our visit, these ova had been deposited by a female just 
at the edge of the water in a small basin, whose contents had been 
allowed to slightly evaporate; so that these ova were at the time of 
our visit entirely above contact with the water. Notwithstanding this 
long interval after deposition, they were promptly converted into the 
larval stage, after a short period, by raising the level of the water in 
the basin. 
With the mosquitoes thus obtained we have been able to conduct 
our experiments. Specimens of this mosquito forwarded to Mr. L. 
O. Howard, Entomologist, Department of Agriculture, Washington, 
D. C., were kindly identified as Culex fasciatus Fabr. 
In this Preliminary Note we have not space to refer, at length, to 
the various interesting and valuable contributions made by Finlay to 
the mosquito theory for the propagation of yellow fever. In addi- 
tion to the paper already quoted, his most valuable contributions to 
this important theory are to be found in the articles designated as 
follows: Estadistica de las Inoculaciones con mosquitos contami- 
nados, etc. Reprint, Havana, 1891: Fiebre Amarilla, Estudio Clin- 
ico Patologico y Etiologico. Reprint, Havana, 1895: and Yellow 
fever immunity,—Modes of Propagation: Mosquito theory, 8th 
Congress of International Hygiene and Demography, Budapest, 
1894. 
His present views on this subject may be stated in his own lan- 
guage : “First, reproduction of the disease, in a mild form, within 
five to twenty-five days after having applied contaminated mosqui- 
toes to susceptible subjects. Second, partial or complete immunity 
against yellow fever obtained when even no pathogenous manifesta- 
tion had followed those inoculations.” (Medical Record, Vol. 55, 
No. 21, May 27, 1899.) 
Without reviewing the cases regarded as mild forms by the author 
of this theory, we believe that he has not, as yet, succeeded in repro- 
ducing a well marked attack of yellow fever, within the usual period 
of incubation of the disease, attended by albumen and jaundice, and 
in which all other sources of infection could be excluded. 
The experiments made by us on eleven non-immune individuals 
are embraced in the following table which should be carefully 
studied. 
The mosquito used in all cases was Culex fasciatus Fabr.: 10 
THE ETIOLOGY OF YELLOW FEVER. 
No. of 
Case. 
Date of 
Inoculation. 
Character of 
Time of between in- 
No. of 
Ag-e. 
Nativity. 
attack and No. of 
Day of Disease. 
fection of mosquito 
Mosqui- 
Result. 
Remarks. 
patients bitten. 
and inoculation. 
toes. 
1 
u. s. 
August 11th. 
“ 11th. 
Mild 
.1 
7th. 
5 days. 
One. 
Negative. 
2 
u. s. 
Very mild 
i i 
.1 
5th. 
5 “ 
ii 
“ 
3 
24 
u. s. 
“ 12th. 
.1 
5th. 
6 “ 
ii 
ii 
4 
20 
u. s. 
“ 12th. 
ii 
..1 
5th. 
6 “ 
ii 
ii 
S 
24 
u. s. 
“ 14th. 
t C 
..1 
5th. 
8 “ 
ii 
ii 
6 
34 
u. s. 
“ 16th. 
“ 
.1 
5th. 
10 “ 
ii 
ii 
7 
22 
u. s. 
“ 18th. 
Severe 
..1 
2nd. 
3 “ 
i ( 
i i 
8 
20 
u. s. 
“ 19th. 
Very mild 
..1 
Sth. 
13 “ 
- Two. 
a 
Severe.... 
.1 
1st. 
3 “ 
9 
28 
u. s. 
“ 25th. 
Fatal. ... 
.1 
2nd. 
6 “ 
Mild 
.1 
1st. 
4 “ 
k One. 
a 
Severe.... 
.1 
2nd. 
2 “ 
10 
46 
England. 
“ 27th. 
“ 
.1 
2nd. 
12 “ 
] 
11 
24 
U. S. 
“ 31st. 
Mild 
Severe.... 
Mild 
Fatal 
.1 
.1 
.1 
.1 
1st. 
2nd. 
2nd. 
2nd. 
6 “ 
4 “ 
2 “ 
12 “ 
J 
• One. 
• 
Positive. 
Severe at- 
tack of yel- 
low fever. 
Mild 
.2 
2nd. 
4 and 10 days. 
One. 
Severe.... 
.2 
2nd and 9th. 
2 and 8 “ 
i C 
.3 
1st, 2nd and 2nd. 
2, 8 and 16 “ 
1 
One. 
One. 
Mild 
Fatal 
Severe .... 
Mild 
.2 
.1 
.3 
1st and 2nd. 
2nd. 
1st. 
1st, 2nd and 2nd. 
6 and 10 “ 
12 “ 
2 “ 
4, 6 and 10 “ 
i 
• Positive. 
Well 
marked at- 
tack of yel- 
low fever. 
Severe.... 
..3 
All on 1st. 
2, 4 and 8 “ 
One. 
Mild 
.1 
2nd. 
6 “ 
J 
TABLE III. 
INOCULATION OF NON-IMMUNK INDIVIDUALS THROUGH THE BITE OF MOSQUITOES (C. FASCIATED). THE ETIOLOGY OF YELLOW FEVER. 
11 
It will be seen that we record nine negative and two positive re- 
sults. It is, we think, important to observe that of the nine failures 
to infect, the time elapsing between the biting of the mosquito and 
the inoculation of the healthy subject varied in 7 cases from 2 to 8 
days (Nos. 1, 2, 3, 4, 5, 7 and 9) and in the remaining two from 10 to 
13 days (Nos. 6 and 8). 
Five individuals out of the nine who failed to show any result 
(Nos. 2, 3, 4, 5 and 6) were inoculated by mosquitoes that had bitten 
very mild cases of yellow fever on the fifth day of the disease, and 
one individual by a mosquito that had bitten a mild case of yellow 
fever on the seventh day of the disease. (This latter patient was 
discharged from hospital three days later.) To this fact may pos- 
sibly be attributed the negative results. Of the remaining three 
negative cases (Nos. 7, 8 and 9), and which had been inoculated by 
mosquitoes that had bitten severe cases of the disease, the interval 
between the bite and the inoculation varied from 2 to 6 days. 
In the two cases (Nos. 6 and 8) where the interval was respectively 
10 and 13 days, the inoculations had been made with mosquitoes that 
had bitten very mild cases of yellow fever on the fifth day of the 
attack. No. 8 was also bitten by a mosquito which had been infected 
by a severe case of yellow fever 3 days before. 
We refrain from commenting further at this time upon the nine 
negative cases, preferring to record the results obtained rather than 
to indulge in speculation. 
Of the two cases which we have recorded as positive in Table III, 
we now propose to speak at greater length. 
Case io. Dr. James Carroll, Acting Assistant Surgeon, U. S. Army, a 
member of this board, was bitten at 2 p. m., August 27, 1900, by Culex fasciatus. 
This particular mosquito had bitten a severe case of yellow fever on the 
second day of the disease 12 days before; a mild case of yellow fever, on the 
first day of the attack, 6 days preceding; a severe case of yellow fever, on the 
second day of the attack, 4 days before, a mild case of yellow fever, on the 
second day of attack, 2 days before inoculation. 
Dr. Carroll remained well until the afternoon of the 29th, when he states 
that he felt tired and for this reason, when on a visit to Las Animas Hospital, 
the same afternoon (29th), sometime between 4 and 6 p. m., after visiting a 
few patients, he left the wards and waited outside on the porch, while his 
companions remained in the wards. 
August 30th. During the afternoon, although not feeling well, Dr. C. visited 
La Playa, distant about one and a half miles from Columbia Barracks, and 
took a sea-bath. 
August 31st, A. M. Dr. C. realized that he was sick and that he had fever, 
although he refrained from taking his temperature, but did visit the labora- 
tory, distant about one hundred and forty yards, for the purpose of examining 
his blood for the malarial parasite. The examination was negative. During 
the afternoon he was compelled to take to his bed. At 7 p. m. temperature 12 
THE ETIOLOGY OF YELLOW FEVER. 
was 102° F. No headache nor backache; only a sense of great lassitude. 
Eyes injected and face suffused. 
September 1st, 7 A. M. T. 102° F. Blood again carefully examined by Dr. 
Lazear with negative result. 11 a. m., T. 102°. 
The case having been diagnosed as one of yellow fever, Dr. C. was at noon 
removed to the yellow fever wards. 
9 p. m., T. 102.8°, pulse 90, 12 o’clock, midnight, T. 103.4°, pulse 84. 
September 2d, 3 A. M. T. 103.6, pulse 80. A trace of albumen was now 
found in the urine. The subsequent history of the case was one of severe 
yellow fever. Jaundice appeared on September 3d. 
The accompanying chart No. 1 contains all of the necessary data. 
The question of diagnosis having been clearly and easily estab- 
lished, it now becomes important to follow Dr. C.’s movements for a 
period of ten days preceding the mosquito inoculation, and during 
the period elapsing from the bite of the insect until the commence- 
ment of the attack. 
On August 21st, 22d and 23d, Dr. C. was at Columbia Barracks, 
outside of the epidemic zone. On August 24th he visited the au- 
topsy-room of Military Hospital No. 1, which is situated on Principe 
hill overlooking the city of Havana. He was present in this autopsy- 
room while an autopsy was made by Dr. Agramonte on a case of 
pernicious malarial fever. Dr. C. only took cultures from the blood 
and organs as the section proceeded. He was there about half an 
hour and then returned to Columbia Barracks. Subsequent micro- 
scopic study of sections of the liver and spleen showed that the case 
autopsied on the 29th was really a case of pernicious malarial fever. 
It should be stated that although cases of yellow fever are not 
admitted to Military Hospital No. 1, an English sea-captain had been 
admitted to its wards a few days before, whose case developed into 
one of yellow fever with fatal result, and the body had been autop- 
sied by Dr. Agramonte in this dead-room on the day preceding Dr. 
C.’s visit to it. 
According to Dr. C. the room was, by no means, in a cleanly con- 
dition. As Dr. C.’s visit to this room was made on August 24th, 
and as he began to complain on August 29th, about the average 
period of incubation of yellow fever, there is a possible chance for 
infection in this way. We must call attention, however, to the fact 
that Dr. Agramonte, whenever he performs an autopsy in this room, 
is always attended by a young soldier of the Hospital Corps, U. S. 
Army, who is detailed for that purpose, and whose duty it is to assist 
and to afterwards tend to the cleaning of the autopsy table. This 
soldier, a non-immune American, was present when Dr. Carroll was 
there and remained afterwards to attend to his duties. He has not 
contracted yellow fever by his duties in this room, from time to time. Case/.. //e/Zore Ceoer /o/Zoa'/pp a////a //e s/saa/ 
per/or/ o/ inca/af/o//, Z/e /a/e 0/ aa /afee/ec/ 
mospa/fo, (CulejcJasclutursJ 14 
THE ETIOLOGY OF YELLOW FEVER. 
Our own experience would seem to accord with that of others, viz., 
that attendance upon autopsies and the handling of portions of 
organs of yellow fever cases removed to the laboratory is unattended 
with danger. Certainly the three non-immune members of this 
board, up to the time of these mosquito inoculations, had during the 
past three months come in close contact with the dead bodies and 
organs of yellow fever cases, freely handling and examining these 
organs, including the small intestine, even kept at thermostat tem- 
perature for 24 hours, without contracting the disease. We have, of 
course, never neglected to cleanse our hands with disinfectants. 
Dr. C., upon his visit to the before-mentioned dead-room, only 
used the platinum loop for taking cultures and did not come in con- 
tact with the autopsy table. 
The only other opportunity for infection in his case would appear 
to have been during his visit to Las Animas Hospital, situated in the 
suburbs of Havana, as here yellow fever patients are admitted in 
large numbers. We have already pointed out that Dr. C. was com- 
plaining of lassitude at the hour of his visit, which was about fifty 
hours after his inoculation with the contaminated mosquito. We 
have also called attention to the fact that he remained, for the greater 
part of his visit outside of the hospital, on the piazza. This would 
appear to cast doubt upon his visit to Las Animas as the source of his 
infection. 
We do not wish to be understood as unnecessarily seeking to lay 
too much emphasis upon the exclusion in this case of other sources 
of infection than the mosquito, as we fully appreciate that Dr. C. had 
been, on two occasions, within the epidemic zone during the week 
preceding his attack of yellow fever. His movements on these occa- 
sions we have already given. 
We will again refer to Dr. C.’s case, after we have given the historv 
of Case No. 11, which we have designated as our second positive 
result. 
Case n. X. Y., white, American, a resident of the military reservation of 
Columbia Barracks, was bitten during the forenoon of the 31st day of August, 
1900, by the same mosquito that had bitten Case 10 (Dr. C.) four days before, 
and which in the meanwhile had bitten a mild case of yellow fever (first day) 
two days before being applied to X. Y. 
X. Y. was, also, bitten by a second mosquito that had been applied to a 
fatal case of yellow fever (second day) 12 days before, and to two mild 
cases (second day) 4 and 10 days previously; also, by a third mosquito that 
had bitten a fatal case of yellow fever (second day) 12 days before; a severe 
case (first day) 2 days before and three mild cases (first, second and third 
day) 4, 6 and 10 days before finally by a fourth mosquito that had bitten three 
severe cases of yellow fever (all on the first day) 2, 4 and 8 days previously, 
and one mild case (second day) 6 days before. (Vid. Table III.) THE ETIOLOGY OF YELLOW FEVER. 
15 
It will thus be seen that X. Y. was bitten by four mosquitoes, two of which 
had bitten severe (fatal) cases of yellow fever 12 days previously; one of which 
had bitten a severe case (second day) 16 days before and one which had bitten 
a severe case (first day) 8 days before. 
September 25th, X. Y. began to experience a sense of dizziness and disin- 
clination to work. This was just five days from the time of the mosquito 
inoculation. 
Twenty-four hours later, still dizzy and light-headed in attempting to move 
about. During the afternoon (sixth day after inoculation), chilly sensations, 
followed by fever and restlesness during the night. 
On the following day (seventh day after inoculation), 8 a. m., T. 102.8° 
F., eyes slightly injected, face suffused. Patient removed to the yellow fever 
wards; 9 a. m., T. 103° F., pulse 66. A trace of albumen was found in the urine 
during the afternoon (third day of the attack). This increased during the 
following days. Conjunctive slightly jaundiced on the fourth day of disease, 
which was more distinct and could be plainly seen on anterior aspect of 
chest on the fifth and following day. Bleeding from the gums was noticed on 
the third and subsequent days after admission. Repeated examinations of 
the blood failed to show any malarial parasites. 
The course of the fever, the appearance of albumen in the urine, 
with jaundice and hemorrhage from the gums, together with the slow 
pulse, all pointed distinctly to the diagnosis of yellow fever. His 
attending physician, Dr. Roger P. Ames, U. S. A., an expert in the 
diagnosis and treatment of this disease, did not hesitate to diagnose 
X. Y.’s attack as one of “well pronounced yellow fever.” Dr. A. 
was not cognizant of the method of inoculation in this case. (Vide, 
Chart II.) 
The diagnosis, therefore, not being in doubt, we must follow this 
patient’s movements during the ten days preceding the bite of the 
mosquitoes and from this time until five days later, when the attack 
began. It so happens that we can follow X. Y.’s movements for a 
much longer period. Fifty-seven days prior to his inoculation, he 
spent a day and night in the city of Havana. Sixteen days before 
the inoculation, he rode on horseback with six other non-immunes a 
distance of about one and a half miles towards the seashore and re- 
turned to his dwelling, without in the meantime dismounting from 
his horse. From this time until his complete convalescence was 
established, he had remained within the immediate vicinity of his 
home. So that it may be positively stated that X. Y. had not ab- 
sented himself from the Military Reservation of Columbia Barracks 
during a period of fifty-seven days prior to his inoculation (with the 
exception above stated), nor between the date of his inoculation and 
the establishment of convalescence. 
Let us now inquire whether the military reservation of Columbia 
Barracks is outside of the epidemic zone of yellow fever. To this 
we answer that since the commencement of the present epidemic fyfarfJZ,ffe//o2/jfeoerfo/foiO//2y, ir/f/i/a t//e asuaf 
per tod of i/icafaffor, ffe fife of a/2 2/2fee fed 
mosgru/fo, fCufejcJasclufutsJ 17 
THE ETIOLOGY OF YELLOW FEVER. 
of yellow fever in Havana, dating from May, 1900, the average 
monthly population of this Station, including civilian employes, has 
been 1400, nearly all of whom are young non-immunes. 
There have occurred amongst this non-immune population from 
May 1 to October 13, 1900, sixteen cases of yellow fever, all of which 
have been easily and readily traced to a visit to within the boundaries 
of the epidemic zone, except Cases 10 and 11 of Table III, and one 
other case of which we shall presently speak. These cases have been 
distributed as follows: 
May 24th 1 
June 10th 1 
“ 17th 1 
“ 19th 2 
“ 21st 1 
“ 29th 1 
July 9th 1 
“ 26th 1 
“ 29th 1 
August 11th 1 
“ 12th 1 
“ 16th 1 
“ 31st 1 
Sept’r 7th 1 
“ 19th 1 
Total 16 cases. 
Ten of these cases have occurred amongst an average monthly 
military population of 1295 men, and six cases in an average civilian 
population of 105. 
Whenever these cases have occurred, as soon as the patient has 
been removed to hospital most careful measures of disinfection have 
been immediately carried out by a trained sanitary squad, under the 
personal supervision of a medical officer. These have consisted of 
destruction by fire of mattresses, the disinfection of bedding and 
clothing with 1 to 500 bichloride solution, and the application of the 
same solution freely to the ceiling, walls and floors by means of a 
force pump. 
We repeat that no case has ever been connected with a preceding 
case, but that the source of infection has been readily shown to have 
occurred during the individual’s visit to Havana, 6 miles distant, or 
to some other nearer Cuban settlement. 
We now invite attention to the fact that from August 17th to Oc- 
tober 13th, a period of fifty-seven days, only three cases of yellow 
fever have occurred amongst this population of 1400 non-immune 18 
THE ETIOLOGY OF YELLOW FEVER. 
Americans, and we consider it very important to note that two of 
these had been bitten, within five days of the commencement of their at- 
tacks, by contaminated mosquitoes. 
Taken in connection with Case 11, in which we have been unable 
to find any other source of infection than the bite of an infected mos- 
quito, five days preceding the attack, the case of Dr. C. (Case 10, 
Table III) becomes strongly confirmatory of the same origin. 
We will now briefly give the history of the third case of yellow 
fever that has occurred at Columbia Barracks during the period 
August 17 to October 13, 1900. 
In the light of Cases 10 and 11, we consider this case of sufficient 
importance to be here included, especially as it is one that might be 
possibly designated as a case of accidental infection by a mosquito. 
Case. Dr. Jesse W. Lazear, Acting Assistant Surgeon, U. S. Army, a 
member of this board, was bitten on August 16, 1900, (Case 6, Table III), 
by a mosquito (Culex fasciatus) which ten days previously had been contami- 
nated by biting a very mild case of yellow fever (fifth day). No appreciable 
disturbance of health followed this inoculation. 
September 13, 1900, (forenoon), Dr. Lazear, while on a visit to Las Animas 
Hospital and while collecting blood from yellow fever patients for study, was 
bitten by a Culex mosquito (species undetermined). As Dr. L. had been 
previously bitten by a contaminated insect without after-effects, he deliber- 
ately allowed this particular mosquito, which had settled on the back of his 
hand, to remain until it had satisfied its hunger. 
On the evening of September 18th, 5 days after the bite, Dr. L. complained 
of feeling “out of sorts,” and had a chill at 8 p. m. 
September 19th, 12 o’clock noon, T. 102.4°, pulse 112. Eyes injected, face 
suffused; 3 p. m., T. 103.4°, pulse 104; 6 p. m., T. 103.8°, pulse 106. Albumen 
appeared in the urine. Jaundice appeared on the third day. The subsequent 
history of this case was one of progressive and fatal yellow fever, the death of 
our much lamented colleague having occurred on the evening of September 
25, 1900. 
As Dr. L. was bitten by a mosquito while present in the wards of a 
yellow fever hospital, one must, at least, admit the possibility of this 
insect’s contamination by a previous bite of a yellow fever patient. 
This case of accidental infection therefore cannot fail to be of in- 
terest, taken in connection with Cases 10 and 11. 
For ourselves, we have been profoundly impressed with the mode 
of infection and with the results that followed the bite of the mos- 
quito in these three cases. Our results would appear to throw new 
light on Carter’s observations in Mississippi, as to the period re- 
quired between the introduction of the first (infecting) case and the 
occurrence of secondary cases of yellow fever. 
Since we here, for the first time, record a case in which a typical 
attack of yellow fever has followed the bite of an infected mosquito. THE ETIOLOGY OF YELLOW FEVER. 
19 
within the usual period of incubation of the disease, and in which 
other sources of infection can be excluded, we feel confident that the 
publication of these observations must excite renewed interest in the 
mosquito-theory of the propagation of yellow fever, as first proposed 
by Finlay. 
From our study thus far of yellow fever, we draw the following 
conclusions: 
1. Bacillus icteroides (Sanarelli) stands in no causative relation to yel- 
low fever, but, when present, shoidd be considered as a secondary invader 
in this disease. 
2. The mosquito serves as the intermediate host for the parasite of yel- 
low fever.