29 
RELATION OF INCREASED TEMPERATURE TO INCREASED PREV- 
ALENCE OF TYPHOID FEVER. 
Chart No. 4 shows in curves the relation of temperature to the 
incidence of typhoid fever in the District of Columbia for the periods 
(May 1 to November 1) of the three years 1906, 1907, and 1908. 
The temperature curve is compiled from figures furnished us by 
the United States Weather Bureau and based on the mean daily 
temperature in five-day periods. 
It should be noted that in 1906 the' t^’phoid curve rose to reach its 
maximum about a month subsequent to, and roughly paralleled, the 
rise in the temperature curve. As the t;\q)hoid curve would have to 
be set back probably about three weeks in order for it to represent the 
cases by time of probable infection the relation of the rise in this curve 
to that of the temperature is striking. The typhoid curve, however, 
begins to decline quite sharply in the early part of August and the 
decline continues through the remainder of August and through 
September, while the temperature curve reaches its maximum early 
in July and from then continues at about the same height until the 
latter part of September, which was about six weeks after the marked 
decline in the t 5 q)hoid fever rate had begun. 
From this it appears that the increasing causation of Uq)hoid 
fever occurred coincidently vdth the increasing temperature; but 
there was a marked decrease in the causation of t;\q)hoid fever about 
ten weeks prior to the decline in temperature, and therefore inde- 
pendent of the continued high temperature. 
It should be borne in mind, however, that the fall in the Uq^hoid 
fever rate before that in the summer temperature may have been 
due to the majority of the susceptible persons in the community 
already having had the disease. 
In 1907 the marked increase in the incidence of cases occurred about 
a month subsequent to the marked rise in temperature. It is inter- 
esting to note that in this year the rise in both the temperature and 
the typhoid curve occurred about a month later than that of 1906 
and 1908. If the typhoid curve for the 1907 period were set back 
about a month, the rises and falls in the typhoid and temperature 
curves would present a rather striking parallehsm. 
If the same thing is done for the 1908 period, the parallelism in the 
rises and falls of the U’phoid and temperature curves is for that year 
also quite striking. 
The seasonal (summer) prevalence of typhoid fever observed here 
may be due to the summer temperature being followed by one or both 
of the folio vdng conditions : 
1. Increased distribution of the infection in the community (seed). 
2. Increased susceptibihty of persons to the disease (soil). 
