25 



arc to be found in the bulletins of the Inland Revenue Department 

 will be seen to contain chlorine in varying amounts from i to lo, or 

 more parts per million. We must not, however, forget that in many 

 parts of Canada salt is found in the soil, and in various deep seated 

 springs, and it is therefore absolutely essential that the location and 

 surroundings of the well should be known to the analyst before he pro- 

 nounces an opinion on the results of chlorine estimation. Many wells 

 in Winnipeg and other parts of Manitoba contain from 200 to 300 parts 

 of chlorine per million, and are yet free from sewage pollution. 



II. The only other feature in water analysis to which 1 need refer 

 is the estimation of dissolved organic matter essentially non-nitrogenous 

 in character, in other words, of vegetable origin. Such organic matter 

 is with difficulty destroyed by oxidation and requires the employment 

 of the most powerful oxidizing agencies we know to effect its decomposi- 

 tion. For the purpose we always use permanganic acid, a sample of 

 which I show you in solution. Observe its beautiful deep purple colour 

 and see how the addition of a very small quantity of water, impure fiom 

 decomposing organic matter in solution, serves at once, or at least in a 

 very short interval of time to cause the purple to become less and less 

 intense, and shortly to disappear altogether. Now, by using a solution 

 containing a known amount of permanganic acid, and adding it in 

 excess to a measured quantity of the water to be examined, we can 

 easily, at the end of, say four hours, estimate the excess of permanganic 

 acid by chemical means, which need not be here explained, and thus 

 obtain by difference the cjuantity used up in oxidizing the organic matter 

 present in the sample ot water. Since pemanganic acid gives up a 

 definite amount of its oxygen to this purpose, it is convenient to state 

 the results of the examination as so many parts by weight of oxygen to 

 the million parts by volume of water. The observation is usually made 

 for two periods, viz., intervals of 15 minutes and 4 hours; the more 

 easily oxidized organic matter being attacked in the shorter interval, 

 and this part always includes any animal or more objectionable matter 

 present. The following uumbers quoted from Bull, v of the Inland 

 Revenue Department will serve to give an idea of the indications 

 afforded by this test: — - riif*~^'>v, 



LI8RARYJ 30 



