381 



analyses by chemists are multiplied on the one hand, and 

 observations by engineers as to the actual results with waters 

 of known composition on the other, some approximation to 

 such a standard will be arrived at. At present the engineer, 

 when not acquainted with chemistry, probably depends more 

 on the opinion appended by the chemist to his numerical state- ' 

 ment, than on that statement itself. It may give a better idea 

 than in any other way, what character I attach to certain num- 

 bers, if I copy a few analyses, with the report upon each : — 

 No. 1 contained 



Carbonate of lime 24^ grains per gallon. 



Muriate of magnesia 12 j „ 



Sulphate of soda 35 „ 



Total 72 



When such water as the above is evaporated, the carbo- 

 nate of lime separates, and as the quantity is large, the 

 water is very unfit for the purpose it is applied to. 



No. 2, river water : — 



Sulphate of lime 2 grains, nearly. 



Muriate of lime H „ 



Carbonate of lime 5 „ 



Dissolved vegetable matter 2 „ 



Total lOj „ or 2 J oz. 



per 100 gallons. 



The quantity of suspended earthy and vegetable matter 

 was 9^ grains per gallon when dried. The carbonate of 

 lime actually separated on boiling was 4.8 per gallon,^ or 

 about 1 oz. per 100 gallons. Hence this water, if free 

 from mud, would be very fit for boilers ; but taken directly 

 from the river, in its present turbid state, some of the mud 

 will adhere to the metal, producing the same evil as if dis- 

 solved. Such muddy water is also liable to the serious 

 defect of " priming." It should, therefore, be rendered 

 clear by subsidence or filtering. 



