REPOET ON 
THE COMPOSITION OF 
OCEAN-WATER. 
247 
No. of the water analysed, 
I. 
II. 
III. 
IY. 
“ Chlorine ” per kilogram, 
Weight of water taken for the bromine deter- 
53-800 
14-880 
25-053 
13-972 
mination, grams, 
* 
138-80 
1299-01 
771-56 
999-46 
Bromine found in milligrams, — 
1st precipitate, 
23-04 
60-58 
61-40 
44-37 
2nd precipitate, . 
1-98 
3-60 
4-14 
2-88 
J Total, 
25-02 
64-18 
65-54 
47-25 
' Probable uncertainty, 
±0-5 
±0-5 
±0-5 
±0-5 
( Bromine per 100 of “ chlorine,” 
0-3351 
0-3321 
0-3390 
0-3384 
* Probable uncertainty, . 
±0-0067 
±0-0026 
±0-0026 
±0-0036 
Deviation from 0 - 3398, 
-0-0047 
-0-0077 
- 0-0008 
-0-0014 
Sulphuric acid, So 3 , per 100 of “ 
chlorine,” 
11-64 
11-77 
11-60 
11-90 
Deviations from average ll - 58, given page 138, 
+ 0-06 
+ 0-19 
+ 0-02 
+ 0-32 
The number 0‘3398 is the quantity of bromine per 100 
of chlorine 
which follows 
from the four analyses reported on page 99. The number CP3414 which is there given is 
the result of a slip in a calculation, which, however, is of no moment, as the substitution 
of the correct number would diminish the grand average given on page 101 as 0'3402, 
by only 0'0004. Considering that half a milligram of absolute error in the total bromine 
obtained in an analysis corresponds to only 0‘25 milligram of cumulative effect of the 
errors in the four weighings involved, we may well take our numbers as being com- 
patible with the assumption that the partial freezing of a sea-water involves no change 
in the ratio of bromine to chlorine. 
The quantities of sulphuric acid found (all except No. 1, the result of single analyses) 
do not agree so well with the general mean of 1 1 ’57 6, which was deduced from the seventy- 
seven analyses, as I should wish ; but their evidence, like that afforded by the bromine 
determinations, tends to show that the partial freezing of a sea- water (which constantly 
occurs as a natural phenomenon in the polar circles) does not involve the formation of 
any cryohydrate. Sea- water ice, it would appear, is just ice enclosing drops of highly 
saline mother-liquor. 
