134 BACTERIA IN WATER. 
waters by sewage and the growing demand for water have led to 
development of methods of filtering such contaminated water in 
large quantities. This is done by passing it through layers of sand 
which are constructed in such a way as to remove most of the 
bacteria. These filters are in wide use to-day by cities that have 
to depend upon a contaminated supply. The bacteria are not 
wholly removed from the water, but so nearly that practically all 
dangers disappear. Experience has shown that the use of filters 
very greatly reduces the amount of typhoid fever in cities dependent 
upon a contaminated water-supply. It is also found that the 
purified water improves the general health of the community, quite 
apart from the decrease in typhoid fever. 
Ice. Ice, though not thought of as water, in summer months 
is put into drinking-water to cool it. The ice melts and whatever 
bacteria are in it are liberated and swallowed with the water. It 
has been a belief that freezing purifies water, so that many have 
been perfectly willing to use ice from ponds whose water they would 
not drink. It is a very wide practice to cut the year's ice supply 
from sewage-contaminated streams, and from places where no one 
would think of drinking the water: e.g., from the Hudson River, 
below Albany. It has become a matter of great importance to 
know whether freezing does purify such ice and render it safe. 
The subject has been most carefully investigated, with the following 
conclusion. Ice does in a measure purify itself in freezing, but not 
wholly. If typhoid bacilli are in the water, they may be found in 
the loose snow ice at the top of the frozen layer, but there are very 
few, if any, in the clear ice below. After the ice has been stored 
for a while the typhoid bacilli become less and less abundant, and 
after a few weeks they practically disappear. Even after months 
of freezing, however, a few may sometimes be found, so that no 
ice from contaminated water can be guaranteed as absolutely free 
from them even after six months' storage. But the number that 
resists this storage is so extremely small that the ice is as pure as 
filtered water. No cases of typhoid fever have been definitely 
traced to such a source, though one or two doubtful cases have 
been so attributed. In general, then, it appears that stored ice 
