TYPHOID FEVEE IX DISTRICT OF COLUMBIA. 45 
the infection was contracted probably will be approached with a fair 
deo^ree of accuracy. 
o V/- 
The following table shows the progress of the disease based upon 
the date of onset: 
Cases. 
May 1 to 15 (incomplete) 6 
May 16 to 31 (incomplete) 16 
June 1 to 15 (complete) 32 
June 16 to 30 (complete) 32 
July 1 to 15 (complete) 101 
July 16 to 31 (complete) 131 
August 1 to 15 (complete) 139 
August 16 to 31 (complete) 70 
September 1 to 15 (complete) 59 
September 16 to 30 (complete) 41 
October 1 to 15 (complete) 98 
October 16 to 31 (incomplete) 22 
Total 747 
In considering the occurrence of cases by these periods, the striking 
features are the sudden rise in the early part of July, the continued 
high rate to about the middle of August, and the marked fall in the 
rate between the middle and last of August (see charts Xos. 1 and 2). 
This decrease in the rate continued through September. In the early 
part of October the rise was due largely to the milk outbreak among 
the customers of milk dealer Xo. 4. (See Chart Xo. 3.) 
If the first s^miptoms of the disease appear about two weeks after 
the infection is contracted and are followed by symptoms of defi- 
nite onset about a week later, then the causes which operated to 
effect the rate must have become greatest about June 10 and con- 
tinued until about July 25, when they suddenly and markedly 
decreased. This decrease in the causes, which occurred in the latter 
part of July and continued through August, is one of the most remark- 
able facts brought out in the course of our studies, and it is difficult 
to explain why a decrease in the causes should have taken place at 
this season, which seemed most favorable to many of the factors 
which spread the infection of typhoid fever, such as flies, contacts, etc. 
The possibility of a prolonged period of incubation in some cases 
must be borne in mind. It has occurred to us that perhaps a cer- 
tain number of persons contract the infection earlier in the year 
and c^rry the bacilli in their intestinal tract, but do not show symp- 
toms of the disease until subjected to the depressing influences of 
the hot weather, which usually occurs in Washington early in July. 
Maps Xos. 1, 2, 3, 4, 5, 6, and 7 show the distribution of cases 
according to place of residence when the disease was contracted. 
The general distribution of the cases throughout the District of 
Columbia during each of the periods is striking. 
