1887.] PUBLIC DOCU^IENT — No. 33. 123 



known quantity of the sample to dryness and weighing the 

 residue. 



Hardness, a rather arbitrary term, signifies that quality of 

 water which prevents the ready formation of lather with soap. 

 It is usually due to the presence of salts of lime or magnesia, 

 which decompose the soap, forming new insoluble compounds. 

 As long as these reactions occur no lather will be formed, con- 

 sequently the quantity of a standard soap consumed before a 

 permanent lather is obtained indicates the amount of earthy 

 salts contained in the sample, or its relative degree of hard- 

 ness. Frankland's scale has been changed to Clark's, the 

 latter being a more popular one. Ten degrees of Frankland's 

 scale are equivalent to seven of Clark's. 



Mr. Wancklyn's interpretation of the results of his mode of 

 investigation are as follows : — 



1. Chlorine alone does not necessarily indicate the presence 

 of filthy water. 



2. Free and alliuminoid ammonia in water without chlorine 

 indicates a vegetable source of contamination. 



3. More than five grains per gallon of chlorine, accompanied 

 by more than 0.08 parts per million of free ammonia and more 

 than one-tenth part per million of albuminoid ammonia, is a 

 clear indication that the water is contaminated with sewage, 

 decaying animal matter, urine, etc., and should be condemned. 

 • 4. Eight hundredths part per million of free ammonia 

 and one-tenth part per million of albuminoid ammonia render 

 a water very suspicious, even without much chlorine. 



5. Albuminoid ammonia over 0.15 parts per million ought 

 to absolutely condemn the water which contains it. 



6. The total solids found in the water should not exceed forty 

 grains per gallon. 



An examination of the above results of analyses shows that 

 Nos. 4, 6 and 13 are of a suspicious character, and that Nos. 1, 

 2, 3, 5, 9 and 12 ought to be condemned. 



Parties sending on water for an analysis ought to be very 

 careful to use clean vessels, clean stoppers, etc. The sample 

 should be sent on without delay after collecting. One gallon 

 is desirable for the analysis. 



