356 H. M. Vernon 



round the sides of the bell-jar, where of course the sand was exposed 

 to daylight, but it would seem to have also been due to the action 

 of baeteria or other oi'ganisms contaiued in the sand. Thus ou the 

 42"*^ day, when the rate of tìltration had agaiu become considerably 

 slower, and the purifying power of the fìlter somewhat less, the whole 

 apparatus was covered np with black paper, so as to prevent any 

 trace of light reachiug the sand. It was kept in this condition tili 

 the termiuation of the experiment, 25 days later. During this time 

 the rate of filtration gradually became slower and slower, tili at last 

 only about one litre per hour ran through. The purifying action of 

 the filter if anything increased however, though the amount of organic 

 ammonia removed never reached so high a level as when the filter 

 was exposed to light. That this purification was due to other than 

 Chlorophyll containing organisms, would necessarily follow froni the 

 absolute exclusion of light. It was proved also by a microscopical 

 examinatiou of the top layer of sand, which showed that these were 

 no longer present. A still further proof that these Chlorophyll con- 

 taining organisms took no part in the process is afiforded by the 

 series of experiments given in the right half of the above table. 

 From these it may be seen that in each of the three instances where 

 the rate of tìltration was so diniinished that a litre of water took 

 fifty miuutes or more to run through, the free ammonia in the water 

 was increased some three- or fourfold, though the organic ammonia 

 was but slightly affected. Now it may be seen in the above table, 

 that with the filter excluded from light, water running through at 

 the rate of a litre in 47 minutes had 94^ of its free ammonia 

 removed, and 19^ of its organic ammonia; and that when on the 

 (56*^^ day, by clamping the outflow tube, the water was made to filter 

 through at only a litre in 2 hrs. 50 min., the purification was still 

 nearly as great. Even the water drawn off at about the same rate 

 on the next day, immediately after the flow of water through the 

 filter had been entirely stopped for ten hours, was just as purified 

 in respect of its free ammonia, though none of its organic ammonia 

 had been absorbed. In fact, it would seem that within certain limits, 

 with the sand excluded from light, the slower the rate of filtration 

 the greater is the efficiency of the filter. On the other band, with 

 sand impregnated with Chlorophyll containing organisms, the quicker 

 the rate of filtration, the better the purification. Thus in the ex- 

 periment made on the lö*"" day, considerably less ammonia was 

 removed when the rate of filtration was diminished from a litre in 



