326 MODERN METHODS OF WATER PURIFICATION. 



cities instead of little. For instance, at New Orleans from 

 one day's water supply for the city of the Mississippi River 

 water there is deposited by precipitation, on a day when the 

 river is in flood, over 400 tons of mud. The Susquehanna on 

 the day on which I visited it — a showery day in November — 

 was just like the Modder River — except, of course, in point 

 of size — of similar turbid appearance in bulk and in glass. 



Due prominence is given by the experts in America to the 

 question of the degree of alkalinity of the water, and to that 

 of temperature, and their influence on sedimentation, an influ- 

 ence we have noticed here and have suffered from in winter, 

 when sedimentation does not go on as rapidly as in the summer 

 when the water is warmer. 



For coarse particles precipitation is less affected by tempera- 

 ture, but for the fine colloidal clay it makes so much difference 

 that in Hazen's opinion water treated at 32° F. takes twice as 

 long to precipitate as when the thermometer registers 74°, a 

 summer temperature often exceeded here. This he states is 

 also true of filters. This throws light on our troubles. We 

 have disturbed clarification when our coldest temperature 

 comes, and the efficiency of the filters is not only less by cold, 

 but they have more to do than when precipitation is going 

 on well, and hence an inferior quality of water results. At 

 Bloemfontein we do not use alum, although it is the precipi- 

 tant which I have favoured from the beginning in one or 

 another of its forms. I do not believe it has ever been 

 thoroughly tested under sufficiently varying conditions of 

 turbidity and temperature, and I still believe it will give, when 

 properly investigated, as good results as Stead's method oi 

 lime and permanganate, which has given such excellent re- 

 sults. Alum is more convenient to work with, readily soluble 

 and applied, and is more uniform than is the lime which we 

 have available at the waterworks, which at one time varied in 

 chemical value within very wide limits. The application of 

 lime and permanganate is more difficult to control and apply, 

 and the variations in quantity that should accompany varia- 

 tions in requirements are more difficult to regulate, especially 

 if the lime varies in chemical value. The greatest considera- 

 tion of all is that of efficiency, and until experiments definitely 

 prove the superiority of alum (in which I include alumino- 

 ferric) when used as it ought to be, for I am of opinion it has 

 never yet had a fair trial, I would not advise changing from 

 Mr. Stead's process, to whom, due to his presence temporarily 

 as Acting Government Analyst in 1906, we owe it that the 

 principle of chemical sedimentation was adopted so early and 

 so easily. 



One great point in favour of the lime-permanganate process 

 of Stead is that it was possible to assure people that in that 

 process nothing permanent was added to the water and only 

 impurity taken away. In the matter of alum there is a popular 

 prejudice against its use until people know better, and learn 

 that the alumina when adequately controlled and added in proper 

 proportions does its work and is so deposited that it does not 

 pass the filters, especially where there is alkalinity. One 



