REPORT ON THE COMPOSITION OF OCEAN- WATEE. 
95 
(No. 23.) Bromine taken, 63 ‘83 mgrms. The weight of bromine in the successive 
silver precipitates was b v b 2 , b 3 , See., as shown in the following table : — 
I. Precip., lOOc.c. 
of silver solution, = 
53-96 
II. „ 
50 c.c. 
’> ^2 = 
5-98 
HI. „ 
50 c.c. 
2-16 
IV. „ 
25 c.c. 
■■ 6 4 = 
0-36 
v. „ 
25 c.c. 
» ^5 = 
- 0-18 
Total bromine, excluding number V., = 62'46 
Error = -1-37 
(No. 24.) Bromine taken, 64 , 02 mgrms. 
I. Precip., 100 c.c. of silver solution, 5 1 = 53'80 
II. „ 
50 c.c. 
1 J 
7-34 
III „ 
50 c.c. 
• J 
1-80 
IV. „ 
faile 1 ; 
taking 
0-36 (?) 
We have for total bromine, 63 - 3 
Error = — 07 
(No. 25.) Bromine taken, 63‘85 mgrms. 
T. Precip., 100 c.c. of silver solution, & 1 = 53'81 
(Not continued.) 
Seeing that a complete precipitation could not be effected by less than some 250 c.c. 
of decinormal silver, which means an inconveniently large silver precipitate, I tried to 
improve upon the method by effecting a fractional precipitation in a neutral solution, 
and adding nitric acid only after some standing to render the precipitate amenable to 
decantation. What I hoped for was that the milkiness of the haloid precipitate produced 
in the absence of acid would favour the exchange of precipitated chlorine for dissolved 
bromine, so that the latter could be got down by means of less silver. The result 
(Nos. 26 and 27) was not as expected. 
(No. 26.) Bromine taken, 61 ‘2 mgrms. 
I. Precip., 100 c.c. of silver solution, & 1 = 51'57 
(No. 27.) Bromine taken, 61‘24 mgrms. 
I. Precip., 150 c.c. of silver solution, 5j = 53'41 
In the second case the result for b v on account of an accident in the chlorination, is 
not quite safe ; but the decantate was proved, by qualitative testing, to contain 
abundance of dissolved bromide. In the following experiments acid silver solution was 
employed as usual : — 
