120 



The Philippine Journal of Science 



1916 



fication, is not only preferable, but even necessary, to provide 

 adequate protection to people using the pool. The fractional 

 addition has the further advantage that the objectionable fea- 

 tures of high dosage (odor, irritation of mucosae, and the like) 

 are largely eliminated. 



Obviously the water in a swimming pool should be as clear as 

 possible, not only because clear water makes a pool more attrac- 

 tive and lessens the danger of accidental drowning, but also 

 because it is more susceptible than turbid water to the disinfect- 

 ing action of chlorine. Therefore water should be subjected to 

 filtration, with or without coagulation, wherever practicable. 

 Aside from its coagulating effect, the action of alum is beneficial 

 in that it reacts with bicarbonated waters in such a way that the 

 action of chloride of lime or copper sulphate is interfered with 

 as little as possible. 



The basic carbonates, which would otherwise form precipi- 

 tates with copper sulphate or chloride of lime, preventing ger- 

 micidal action in the case of the former by removing copper 

 from the sphere of influence, and causing cloudiness of the water 

 in both cases, are removed. The addition to tank III of 1 part 

 of alum per million parts of water had the effect shown in 

 Table XIV. 



Table XIY.— Effect of alum. 



[Results expressed as parts per million.] 



■ 



Tur- 

 bidity. 



Total 

 solids. 



Oxygen 



con- 

 sumed. 



Before addition of alum 



14 

 10 

 28 



164 

 112 

 81 



1.51 

 1.04 

 31 



Twelve hours after addition of alum _ _- _ 



Reduction in percentage _ - - ____- 





An attempt was made to correlate the analytical chemical data 

 with the biological, with a view toward establishing chemical 

 methods of determining the state of purity of water in a pool. It 

 must be admitted that little success was achieved. The oxygen 

 consumption showed no traceable relation either to bacterial 

 content or to chlorine-consuming power, and even the variation 

 in chloride content failed to give reliable data on contamination. 

 It may be that the intermittent settling and stirring up of the sus- 

 pended matter which causes the high turbidity greatly influenced 

 the analytical results ; at any rate, differences noted were neither 

 uniform nor great enough to justify conclusions. That a rela- 

 tion does exist is shown by the comparison between the number 

 of bathers a week and the oxygen-consuming capacity at the end 



