50 
TYPHOID FEVER IIS' DISTRICT OF COLE^MBIA. 
SHELLFISH AND FBESH-WATER FISH. 
The following table shows the number of patients who ate shellfish 
or fresh-water fish from time to time vuthin the thirty days prior to 
the onset of illness: 
Yes. 
No. 
Not 
stated. 
Oysters 
60 
655 
32 
Clams 
64 
647 
36 
Lobsters 
8 
708 
31 
Crabs 
190 
516 
41 
Fresh- water fish 
205 
504 
38 
For 42 cases positive statements were given that no shellfish nor 
fresh-water fish of any kind were eaten. 
These data indicate that shellfish could not have been the means of 
transmitting the infection to any considerable number of cases. 
The small number of patients who gave a history of having eaten 
oysters is accounted for by the fact that the cases studied occurred 
between May 1 and October 31, during which period comparatively 
few ovsters are eaten in the District of Columbia. 
WATER. 
Of the 747 cases 721, or 96.54 per cent, gave a definite histor}^ 
of having used unboiled Potomac water supplied through the regular 
city system as the sole, principal, or occasional source of water for 
drinking during the thirty days prior to onset of illness. 
Of the 26 remaining cases 7 drank no unboiled Potomac water, and 
of these 7 all but 2 drank unboiled water from wells, etc., outside of 
the District of Columbia ; of the 2 who drank boiled Potomac water 
solely 1 had been nursing a case of typhoid fever during the three 
weeks prior to the onset of illness and the infection was attributed to 
contact. Seven cases used no city water, boiled or unboiled, for 
drinking. Of these 4 used water from private wells solely for drink- 
ing, one water from a public city pump solely, which case was attrib- 
uted to milk infection, 1 bottled water solely, and 1 melted ice 
solely, which case was attributed to infected milk. For 12 cases 
positive information regarding the use of unboiled city water for 
drinking could not be obtained. Of these 5 occurred in an institution 
for children, where boiled water was provided for the use of the 
inmates, but free access was had to the unboiled tap water; these 
5 cases were attributed to milk infection. 
Considering as a group the 26 cases which occurred among persons 
of whom definite histories of having drunk unboiled Potomac water 
could not be obtained, 2 were attributed to infection by contact, 7 to 
infection by milk, 1 to infection by ice cream, and 5 spent part of the 
