Bonnington Water f near Leitht Scotland, 311 



grs. 



Silica (a, £?, Z, o) 1*637 



Oxide of iron {b, e) 0*349 



Oxide of manganese [c^ h) .... 0*107 



Alumina (yj^) 0*197 



Carbonate of magnesia (/, ^, ja) . . . 38*985 



Carbonate of lime (w) 40*574' 



Carbonate of soda (w) 0*740 



Total .... 82*589* 



III. Analysis for the Carbonic Acid Gas and Ammonia. 



1. Carbonic acid gas. — Before the above ingredients are 

 computed to binary combinations, we have to ascertain the 

 amount of carbonic acid gas and of ammonia. It is obvious 

 that these two ingredients could not be ascertained from the 

 evaporated water, and therefore new quantities of the mineral 

 water had to be taken. For the analysis of the first 2880 

 grs. ( = 6 ounces troy) of the water were precipitated by chlo- 

 ride of barium, and the precipitate collected upon a small 

 filter ; this, while still wet and properly compressed, was in- 

 troduced into a graduated tube over mercury with an adequate 

 quantity of hydrochloric acid, previously saturated with car- 

 bonic acid gas. Having made a due reduction for the elas- 

 ticity of the aqueous vapour, &c., 0*300 gr. were obtained, 

 making for 16 ozs. 0*802 gr. of carbonic acid gas. 



2. Ammonia. — 3000 grs. of the mineral water were intro- 

 duced into an appropriate tubulated glass retort, and mixed 

 with half an ounce of caustic soda. The vapours were, du- 

 ring distillation, conducted under dilute hydrochloric acid, 

 and the operation finished, when only 2 ounces were left in 

 the retort. The distilled fluid previously mixed with chloride 

 of platinum was evaporated in a porcelain vessel. [I have to 

 notice, that without this addition of the chloride of platinum 

 no correct results can be obtained, even at a temperature below 

 212° Fahr., inasmuch as the chloride of ammonium rises with 

 the aqueous vapours. This is easily ascertained by covering 

 the vessel during evaporation with filtering paper, which when 

 lixiviated will clearly evidence the presence of chloride of am- 

 monium, a circumstance which Berzelius has already noticed 

 in his Manual of Chemistry, vol. x.] The dry residue was 

 lixiviated by a mixture of 2 volumes of alcohol and 1 volume 

 of aether, in which mixture the ammonio-chloride of platinum 

 is perfectly insoluble. The edulcoration was carried on until 

 all acid reaction upon litmus paper ceased, when the precipi- 



* An amount which comes near to 84'.301 grs., but of course free from 

 organic matter, which the latter amount includes. 



