312 APPLIED BACTERIOLOGY 



passed through the filter, and the filtrate was found to be 

 sterile. Further investigation showed that this ceased to be 

 the case when the filtration was continued for a few hours 

 or less instead of a few minutes. It was found that in such 

 case the filtrate contained the same organisms as the un- 

 filtered water ; and the sterility of the earlier filtrates was 

 accordingly due to the circumstance that they had been 

 examined before sufficient time had been allowed for the 

 organisms to be washed through the filter. It was also 

 found that the chemical matters arrested by the filter 

 temporarily arrested a portion of the organisms, and served 

 as a suitable culture-ground for such organisms, which 

 survived and multiplied for considerable periods in the filter 

 before being ultimately washed through. In consequence, 

 the number of organisms became after a short time much 

 larger in the filtrate than in the unfiltered water. Filters 

 once polluted with the cholera or typhoid bacillus were also 

 found to convey the bacillus to sterile water passed through 

 them at considerable periods — up to six weeks or more- 

 after pollution. This fact has been responsible for several 

 epidemics, such as that of Lucknow in 1894, in which, out 

 of 646 officers and men in the East Lancashire Eegiment, 

 143 were attacked by cholera, and 92 died. This epidemic 

 was conclusively traced to the infection of the barrack-room 

 filter by the cholera microbe. 



Sand-filtration. — The working of sand-filters on a large 

 scale depends on the facts described above. A sand filter- 

 bed consists of a layer of sand from 2 to 4 feet deep, 

 supported on gravel. The fineness of the grains of sand, 

 the depth of the filter, and the rate of filtration, all affect 

 the working of the filter in the removal of organisms. The 

 coefficients given for safe working are filtration through a 

 sand-layer not less than 30 centimetres thick, at a rate not 

 exceeding 100 millimetres per hour, and giving a filtrate 



