159 
the thirty days previous to onset of illness but also of that used occa- 
sionally. Frequently it will be found that a family receiving its 
regular milk supply from a certain dairy will on occasions, when the 
regular supply is not sufficient for the needs of the day, obtain milk 
from some other dairy, directly or through the grocery store. The 
milk obtained on one of these occasions may be infected and so respon- 
sible for the case. The source of milk used at places other than the 
regular one for taking meals also should be ascertained if possible. 
Cases resulting from infection in the milk are by no means confined 
to persons who use milk as a beverage. The cream or milk used on 
cereals, fruits, or even in coffee may convey the infection. In the 
summer of 1906 in Washington there were six cases in one family of 
eight persons, all of which were attributed quite definitely to infected 
milk. None of the members of this family drank milk, but they all 
used cream on fruits and cereals. Of course the chances of contract- 
ing the infection from milk is greater among persons who use milk 
freely. 
George Newman a sums up the characteristics of milk-borne epi- 
demics as follows: 
(a) There is a special incidence of disease upon the track of the implicated 
milk supply. It is localized to such area. 
(&) Better-class houses and persons generally suffer most. 
(c) Milk drinkers are chiefly affected and they suffer most who are large 
consumers of raw milk. 
(d) Women and children suffer most, and frequently adults suffer propor- 
tionately more than children. 
(e) Incubation periods are shortened. 
(/) There is a sudden onset and rapid decline. 
( g ) Multiple cases in one house occur simultaneously. 
( h ) Clinically the attacks of the disease are often mild. Contact infectivity 
is reduced and the mortality rate is lower than usual. 
In the different outbreaks due to infected milk it is interesting to 
note how greatly the proportion of persons affected among the users 
of the milk varies. In some outbreaks the proportion is as great as 
25 per cent ; for instance, in the epidemic at Palo Alto, Cal., in 1903, 
which was traced to infected milk by Fish, Mosher, and Snow. Of 
the 900 persons who used milk from the infected supply, 232 had 
typhoid fever. In other outbreaks the proportion is as low as 1 or 2 
per cent. Several conditions no doubt influence the proportion of 
persons affected, the most important of which probably is the amount 
of infection in the milk. In the Palo Alto epidemic it was deter- 
mined that the milk became infected through the water used for 
washing the cans and also at times for diluting the milk. This water 
was obtained from a creek which received the drainage from several 
houses in which there were patients with typhoid fever. The water 
a George Newman, Bacteriology and the Public Health, 1904, 
