POTOMAC RIVER WATER SUPPLY RELATIVE TO TYPHOID FEVER. 259 
THE WASHINGTON FILTERS. 
There are 29 filter beds designed upon the principle of slow sand 
filtration. Each bed has an effective filter area of 1 acre. The 
filtered water is collected in a well-protected reservoir of 14,000,000 
gallons capacity. The filters and reservoirs for filtered water are 
built entirely of concrete masonry. They have floors of inverted 
groined arches on which rest the piers for supporting the groined 
arch vaulting. These filters contain, on an average, 40 inches of 
filter sand and 12 inches of gravel graded from coarse to fine, the 
lower and coarser layer acting as part of the underdrain system, 
while the upper and finer layer supports the filter sand. 
The effluent from one filter was turned into the filtered water reser- 
voir on August 18, 1905, and the other filters were put into service 
successively at intervals of about two days until October 5, 1905, 
when a sufficient number were in operation to furnish the entire water 
supply of Washington. The raw water was then shut off, and from 
that date to the present time the city has had an uninterrupted supply 
of filtered water. 
The filters are stoutly built and include several ingenious engineer- 
ing devices. Advantage has been taken of the experience gained by 
sand filtration in this country and abroad. There are two novel fea- 
tures of the plant, (1) the apparatus for handling the sand and (2) 
a double drainage S 3 ^stem so arranged that the water from each filter 
can either be wasted into the sewer or returned to the raw-water 
reservoir. 
Slow sand filtration was specified by the act of Congress providing 
for the plant. The use of a coagulant was prohibited. It was known 
before the filters were built that slow sand filtration alone would not at 
all times render the Potomac water clear. Experience has shown this 
to be correct. The recourse to further treatment, such as increased 
storage, preliminary filtration, or the occasional use of a coagulant, 
is still recognized as necessary in order to obtain a satisfactory efflu- 
ent at all times. 
It is too soon to speak of the work done by the filters. So far the 
results show that they are intelligently managed and have greatly 
clarified and purified the water. There is no doubt that continued 
experiments and experience with the special problem at hand will 
result in still further improvement, especially so far as turbidity is 
concerned. 
We have not made a critical study of the filtration plant and its 
operation in view of the brief space of time and the limited facilities at 
hand. The engineer officers of the Army in charge of the plant, and 
Messrs. Hardy and Longley, in immediate charge of the works, offered 
us every facility and placed their records at our disposal. To these 
gentlemen we are indebted for many courtesies. The bacteriological 
