REPORT ON THE COMPOSITION OF OCEAN-WATER. 
239 
SUPPLEMENTARY NOTES TO THE CHAPTER ON BROMINE. 
Pages 89 et seq. 
Mr. John M‘ Arthur’s determinations of the ratios of bromine to chlorine in the three 
Challenger water mixtures and the Arran water left nothing to be desired in point of 
relative precision ; but from the synthetical experiments previously made it appeared 
that the results are all liable to a small constant positive correction. Unfortunately those 
synthetical trials are not equal to the analyses themselves in precision, so that it would 
hardly be allowable to calculate the correction from them. 
In order, therefore, to finish satisfactorily a piece of very troublesome work, I, some 
time ago, caused Mr. M ‘Arthur to carry out the following series of critical experiments. 
I will begin by enumerating the several reagents which were employed, and explaining 
how they were prepared. 
1 . Pure Chloride of Silver. — A quantity of pure hydrochloric acid, containing about 
[HC1]=36 , 5 grams of acid per litre, was added to its equivalent of an approximately 
deci-normal solution of nitrate of silver, the precipitate washed by decantation, then 
dried on a water-bath, and finally in an air-bath at 150° C. All the operations were 
conducted in dim gaslight, and the preparation preserved in a blackened bottle. 
2. Pure Bromide of Silver. — A quantity of pure hydrobromic acid (vide infra under 
3), representing very nearly 126’6 milligram equivalents (meaning 126'6 times 80 
milligrams) of bromine, was diluted to 1266 cubic centimetres, and mixed with 633 c.c. 
of deci-normal (acid) nitrate of silver solution, so as to bring down one-half of the 
bromine as silver-salt, and leave the chlorine, which might be present, in the mother-liquor. 
This operation was carried out twice in the same way, to produce about 24 grams of 
pure bromide of silver, which was washed, collected, dried (at 150° C.), and preserved with 
the same precautions as the chloride. 
3. Pure Aqueous Hydrobromic Acid. — Ordinary bromine (proved to be free from 
iodine) was dissolved in bromide of potassium solution, and then recovered by distilla- 
tion. The bromine thus purified was placed under a mass of water, and sulphurous acid 
passed into it to dissolve it as hydrobromic acid. The resulting mixture then was 
subjected to distillation, and the receiver changed when all the excess of sulphurous acid 
had been expelled. The distillation was stopped before the sulphuric acid had become 
strong enough to react on the hydrobromic acid left. The hydrobromic acid was re- 
distilled, to remove any trace of sulphuric acid from it. Its strength was determined by 
