290 
of barytes formed to bromide and chloride of barium. 
These salts were dissolved in water, deprived by carbonic 
acid and boiling of free barytes ; again evaporated to dryness, 
ignited, and accurately weighed. Water being poured on, 
they were re-dissolved, and converted, by the addition of 
nitrate of silver, into a mixture of the bromide and chloride 
of silver, which was also accurately weighed. Now, if w be 
the weight of the mixed bromide and chloride of barium, w’ 
that of the mixed bromide and chloride of silver, x the chlo- 
rine, y the bromine, m the ratio of the atomic weight of chlo- 
ride of barium to chlorine, that of the atomic weight of 
bromide of barium to bromine, m’ the ratio of the atomic 
weight of chloride of silver to chlorine, and 7’ that of the 
atomic weight of bromide of silver to bromine, we will have 
the following equations: 
Mx +- ny — Ww, 
from which we get 
and mx + nym w’, 
, , 
_ mw —-mw * 
mn —mn 
Having thus obtained the bromine, the chloride was easily 
inferred from the mixed chloride and bromide of silver, al- 
ready got by adding nitrate of silver to the water. The 
following are the final results. 
Chloride of calcium, 2.438) | 
Chloride of magnesium, 4 
Bromide of magnesium, 0.201 
Chloride of potassium, 0.852$= 18.78. 
Chloride of sodium, 7.839 . 
Chloride ofmanganese, 0.005 
Sulphate of lime, 0.075| . 
Water, 81.220 ; 
‘ 
100. 
* This method answers well, the atomic weight of bromine being so much 
greater than that ofchlorine, and is susceptible of far greater accuracy than any 
more direct one, having for its object the insulation of the bromine. 
