506 
THOMPSON YATES LABORATORIES REPORT 
have experienced their attacks at night.' By reducing the times of inoculation of these 
sixteen cases to noon and midnight and 6 o'clock morning and evening, and also con- 
sidering the hours about 6 o'clock as neutral we find that if the inoculation had been 
at : — 
6 a.m. the onset would have been during; the night in ten cases. 
Noon „ ,, „ ,, three cases. 
6 p.m. ,, ,, ,, ,, five cases. 
Midnight ,, ,, ,, ,, twelve cases. 
So far as this mode of argument is feasible it shews that the greatest number of onsets 
during the night would have been from night biting mosquitoes, infecting at midnight ; 
in Para this is incompatible with S. fasciata, but compatible with night mosquitoes as 
C. fatigans. The next largest figure is that for 6 a.m. infections, this is compatible 
with the bites of S. fasciata. 
Whilst it would not be profitable to enter into all the occupations in their liability 
to yellow fever danger, we were much struck by the number of bakers that we saw 
as yellow fever patients, thus out of five hundred and twenty-nine cases as many as 
thirty were in bakers ; the nature of their trade was suggestive of much exposure to 
mosquito bites. 
As an indication of the position of the belief that the fever is usually caught at 
night, the following quotation from de Azavedo Sodre and Couto may be made : — 
' As soon as an epidemic of yellow fever commences, many foreigners and unacclima- 
tized persons as well as all well-to-do families, withdraw from Rio de Janeiro and 
Santos to Petropolis and Sao Paulo respectively, thence they travel daily by early trains 
to the city, and return again in the evening. Now, although they remain in the 
pest-laden city from io a.m. to 4 p.m., and eat and drink there without any pre- 
cautions, they escape infection in all epidemics. Those, however, who for one reason 
or another have to remain overnight or for several nights in the city are often struck 
by the fever. This circumstance shews the necessity of avoiding spending the night 
in places which are infected with yellow fever.' 1 
The incubation period in gnat-infected cases (III) gave an average of nearly 
four days with the twelve cases (three days twenty-two hours), but the last case is of 
interest in having so short a period as two days twenty-two hours (provided that 
there was no previous accidental contamination). Compared with malaria this is 
extremely short, for in experimental cases a much longer incubation period has 
been established : Tertian, sixteen to twenty-five days ; Aestivoautumnal, twelve to 
sixteen days. 2 
In tsetse disease in rabbits, the first rise in temperature is generally about the 
eighth or ninth day ; certainly small animals (rats) will die as early as the sixth day 
1. Nothnagel, vol. IV, p. 302. 
2. Fearnside, Sci. mem. of med. officers of Indian army, XII, 1901. 
