View report page

98
free from such conditions. Prom this I conclude that sanitary
defects have not had any casual connection with the outbreak in
question.
Watercress.— Early in the outbreak, while engaged in making
enquiries respecting the possible relation of any of the above media
to the spread of the disease, I was struck by the large number of
watercress eaters amongst the patients affected, and this was
especially marked during the A outburst. This led me to think that
probably watercress may have been the agent by which the infection
had been conveyed. What at first was mere suspicion acquired
greater certainty as cases multiplied, until at last I was convinced
that I had to deal with an outbreak of enteric fever largely due to the
consumption of specifically infected watercress. The evidence upon
which this hypothesis is based is the following —
a. Statistics.— During the A period of the outbreak, 48 cases of
enteric fever were notified. Of these, I could obtain no information
as regards the food-stuffs eaten in three cases; but of the remaining
45 cases, definite information was forthcoming in regard to foods
consumed. I found that out of these 45 cases, no less than 29
patients had partaken of watercress, in most cases more than once,
during the three weeks prior to the onset of the illness. This is
equivalent to 64.4 per cent. of the cases inquired into.
During the B period there were 62 genuine cases of enteric fever
notified. Of these, I could obtain fairly reliable information respecting
food in 58 cases. Out of this number, 28 certainly partook of
watercress during the three weeks prior to the onset of the disease.
This is equivalent to 48.2 per cent. of those attacked. If the two
periods are taken together, the watercress consuming cases numbered
57 out of a total of 103 cases, which is equivalent to 55.3 per cent. of
the total.
The following Table sets out these facts :—