REPORT ON THE COMPOSITION OF OCEAN- WATER. 
Ill 
Exp. 
1 . 
2. 
3. 
4. 
5. 
6. 
Water used, 
Quantity of hydrochlo- 
ric acid added,* 
Solution of barium 
chloride added, 
250 
0 
50 c.c.f 
250 
0 
60 c.c. 
200 
30 
0 
200 
0 
60 c.c. 
200 
30 
0 
200 c.c. 
0 
60 c.c. 
Milligrammes of Carbonic Acid obtained per litre of Water. 
Carbonic acid per litre), 657 32'3 99 - 6 32 - 6 10P5 367. 
Mean of (3) and (5) = 100*55, instead of 100*4 as calculated from the alkalinity. 
In experiments (2), (4), and (6) (where the chloride of barium was in excess of the 
sulphuric acid), only 65, 65, 73 per cent, respectively, of even the loose carbonic acid 
was set free, while in experiment (l),t where the sulphates predominated slightly over the 
baryta-salt, the whole of the loose carbonic acid, and besides about x %ths of the 
E 2 C0 3 acid were eliminated. An explanation of this result is afforded by the experiments 
of Torn0e, who found that in boiling down plain sea- water to near dryness, the bulk of 
the carbonic acid (if not the whole of it) is evolved with formation of a precipitate of 
magnesia free from carbonic acid. In other words, sea- water, on protracted distillation at 
least, behaves as if its carbonic acid were present, substantially, partly as magnesium 
carbonate and partly as free carbonic acid (MgC0 3 + ;rH 2 C0 3 ). In experiment (1) the 
influence of the baryta was practically eliminated, and the bicarbonate of magnesia 
suffered a considerable decomposition by the action of the water; in experiments (2), (4), 
and (6), the part of the bicarbonate present as MgC0 3 was, by the action of the chloride 
of barium, converted into barium carbonate, which, as we know, is proof against the action 
of even hot water ; and only the part which is considered to be free carbonic acid had a 
chance of being liberated. But why was it not all liberated ? As I could find no 
satisfactory answer to this question, I thought I had better make sure at least of the 
facts in the manner to be immediately described. 
Final Series of Experiments. 
About four litres of a mixture similar to the one used for the above experiments were 
prepared, and the alkalinity determined. This was found to be equivalent to, by synthesis, 
* Expressed in multiples of i— mgrms. 
t 1 c.c. = mgrms. ; 50 c.c. are short of the calculated minimum (through a slip in the planning of the experi- 
ment) ; 60 c.c. are an excess. 
