60 On the Analysis of 
of them in the present state of chemical sei- 
ence. Weshould then drop in certain known 
quantities of each from a. dropping tube 
graduated into grains, till the required effect 
was produced; then from the quantity of the 
test required, the quantity of saline matter 
in the water might be determined without the 
trouble of collecting the precipitate; or if 
this was done the one method would be a check 
upon the other. 
I shall now close this imperfect ahbihi by 
a few observations and experiments which I 
have noticed in the course of the present 
week relative to the subject before us. 
I essayed the water supplied by the Man- 
chester water-works, and found it nearly as I 
expected; river water is most commonly softer 
than spring water, and harder than rain water. 
This is the case with the water in question. 
It contains a very little sulphate of lime and 
some carbonate; but only one half of. the 
earthy matter that the above mentioned pump 
water contains. It curdles a little with soap, 
but gives no precipitate with lime water... It 
contains about 1 grain of earthy salts in'4000 
of water. 
When spring water contains supercarbo- 
nate of lime, boiling it precipitates the 
greater part of the carbonate and expels the 
