346 MODERN METHODS OF WATER PURIFICATION. 



chemical precipitation to deal successfully with Modder River 

 water, because the filters get rapidly mud-clogged right 

 through to the bottom of the sand, and then the clay particles 

 stream through. It may be said, and it has been said, that 

 turbidity, although unpleasant in appearance, is not danger- 

 ous, but I am sceptical about this conclusion, and know I 

 am on sure ground that with us high turbidity and dangerous 

 bacterial impurity have usually gone hand, in hand, and that 

 they pass the sand filter together. But clear of the bacterial 

 aspect only, I belive that the clay particles of a highly turbid 

 water can act as irritants to the stomach and bowels, caus- 

 ing catarrh or diarrhoea, which would predispose the affected 

 intestines to attack by typhoid bacilli. 



Another consideration which is one of the strongest points 

 in favour of our present method of chemical treatment is that 

 the permanganate of potassium has, in my opinion, a dis- 

 tinctly inhibiting or destroying action on the germs which 

 are the causes of or associated with intestinal attacks, e.g., 

 typhoid, dysentery, and coli bacilli, even in the very dilute 

 solution which is used. I pointed out this in 1906, when I 

 recommended the use of permanganate alone to deal with 

 our water. If this action of the permanganate is more 

 directly or specifically destructive to these germs than akmi 

 in the various forms and strengths commonly used, then that 

 constitutes a strong argument in favour of the permanganate 

 as coagulant or precipitant. But whether we use alum or 

 permanganate, such coagulant should be regulated accord- 

 ing to requirements, as a hard and fast amount is liable to be 

 sometimes too much and at others times too little in a water 

 which varies so much. The effect should be controlled by 

 careful supervision. 



One point in comparing these modern methods of filtration 

 calls for notice. With the Candy filter no coagulant is used 

 in the working. Without such precipitation I am doubtful 

 if the Candy filter could deal satisfactorily with the Modder 

 River water, and to combine chemical precipitation with the 

 Candy is a new conjunction, although there is no reason, that 

 I am aware of, against it. 



In connection with all the makes of mechanical filters that 

 were described or shown to me by the makers or their agents, 

 I do not think any underestimate of the filtering powers was 

 made, and I doubt if some of the claims could be fulfilled in 

 actual working. That is just where the value of getting the 

 actual results from the heads of installations instead of the 

 makers comes in. 



I was not able to obtain in connection with some of the 

 English modern filters any accounts of special investigations 

 or complete exhaustive reports from the standpoint of bac- 

 terial efficiency, such as are available in the United States and 

 in Germany, and specially with the American filters at Alex- 

 andria by Professors Gottschlich and Bitter. The only 

 thorough work of that sort I found in England was Hous- 

 ton's on the London sand-filtered water. I have seen some 



