the River Severn at Worcester. 



265 



I now give a Table which expresses in grains the amount of salts 

 contained in the imperial gallon, or 70,000 grains, of the various 

 waters : — 



Severn Water of 1866-67. 



Table of assumed Salts calculated as grains in the imperial 

 gallon. 





Water from imme- 

 diately above the 

 waterworks. 



Water 



from 



Bailey's 



Ferry. 



Water from 

 above Haw- 

 ford Brook. 



Water from 



Bailey's 



Ferry. 



April 12, 

 1866. 



Julv 23, 

 1866. 



Oct. 30, 

 1866. 



February 



14, 1867. 



Chloride of sodium . . . 

 Chloride of magnesium. 



Sulphate of lime 



Nitrate of ammonia . . . 

 Nitrate of lime 



6-944 



4-172 



traces 



1-435 



traces 



•623 



•378 



4-389 



2-436 



•280 



•714 



2-471 



9-730 



4-865 



traces 



•287 



traces 



•539 



•168 



4-823 



3052 



•140 



•560 



1-736 



4-816 



4970 



traces 

 •868 



traces 

 •756 

 •952 



4-844 



2-716 

 •210 

 •910 



1-687 



1-428 



•147 



2142 



"•581 



3-164 



1176 



•343 



•539 



•770 



2-534 



2-093 



•686 

 •070 

 3325 

 1393 

 •574 

 •574 

 •875 



Phosphate of lirae 



Carbonate of potassa... 



Carbonate of soda 



Carbonate of lirae 



Carbonate of magnesia. 



Silica 









23-842 



25-900 



22-729 



10-290 



12124 



I have, moreover, calculated the composition of 100 parts of 

 the dry mineral residue of each of these waters (Table, p. 266). 

 In so doing I have excluded the organic matter, and ought per- 

 haps to have excluded the nitric anhydride as being also, as it 

 were, an accidental constituent. The chlorine I have represented 

 in combination ; but the other constituents I have left uncom- 

 bined. The carbonic anhydride has been introduced in quantity 

 sufficient to combine with the free bases, and the silica is there- 

 fore considered to be free. 



I have alluded to the rainfall along the course of the river as 

 a point likely be of considerable interest in connexion with the 

 constituents of the water. In my endeavour to obtain informa- 

 tion on this head, I have been guided by Mr. Symons's annual 

 rainfall report to a number of observers who have most cour- 

 teously given me the information which I sought. It is difficult 

 to ascertain what length of time the water from the different 

 sources would take to reach Worcester ; but as an average I am 

 told that three days would suffice for the Welsh water, and about 

 two or less for the water from the nearer sources. My infor- 



Phil Mag. S. 4. Vol. 34. No. 230. Oct. 1867. T 



