156 
In 1907 and 1908 the high typhoid fever rate continued until the 
cessation of warm weather; in 1906, however, there was a very 
marked decline in the typhoid fever rate eight weeks before the 
marked decline in the warm weather, indicating that in this year 
some factor or factors independent of warm weather per se operated 
to a relatively greater extent than in 1907 and 1908. 
Geographical distribution. — During the three years there has been a 
general and fairly uniform distribution of cases throughout the city. 
Sanitary conditions. — The majority of the cases occur among per- 
sons who live in houses of good or fairly good sanitary condition. In 
1907 and 1908 the proportion of cases among persons living under the 
best sanitary conditions was much smaller than in 1906. 
Sewerage system. — There is little difference in the prevalence of the 
disease in the sewered and in the nonsewered districts. There has 
been no especial grouping of cases, nor an excessive prevalence of 
the disease in the sections of the city in which a number of privies 
remain. 
Flies. — In a well-sewered city, as Washington now is, flies would 
not be expected to play much of a part; and the lack of correlation 
between the seasonal curve of fly abundance and that of typhoid 
fever seems to corroborate this view. 
Servants. — The majority of the cases occur in households without 
servants. There is no evidence to support the supposition that 
day servants frequently convey infection to the households of their 
employers. 
Imported cases. — ^With the decreased prevalence of the disease in 
1907 and 1908 the relative percentages of imported cases have 
increased, but the actual number has remained practically the same 
during the three years. 
Contact. — Our studies again show that ‘‘contact’’ is one of the 
major factors in the spread of the disease. 
Bacillus carriers. — The results of the examination of the excreta 
of about a thousand apparently healthy persons indicate that the 
typhoid bacillus is more commonl}^ distributed among persons than 
the actual number of clinically recognized cases of the disease 
suggests. 
Milk. — About 10 per cent of the cases in 1908, as in 1906 and 
1907, were definitely attributed to infected milk. In the 1908 out- 
break the milk was infected by a bacillus carrier. Our studies for 
the three years indicate that if all the market milk of Washington 
were pasteurized under official supervision the amount of typhoid 
fever here would be materially reduced. 
Shellfish. — It is now evident that oysters and other shellfish do 
not play much part. 
