EEPOET ON THE COMPOSITION OE OCEAN- WATEE. 
115 
To sum up : after n treatments with 5 volumes of air, the carbonic acid per litre 
= c was found to be as follows : — 
Eor 71 — 0 3 6 9 
c=104'5 99'3 94 - 3 90'7 rogrms. 
The ultimate loss of carbonic acid (which of course does not represent the greatest possible 
loss) was 104‘5 — 90'6 = 13'9 mgrrns., or about 27 per cent, of the loose carbonic acid 
originally present. 
By way of appendix to the critical trials on Buchanan’s method, 1 will insert here an 
experiment which I made, in order to see to what extent a sea- water can be deprived of 
its carbonic acid by mere boiling in a current of air and without evaporation. The water 
which I utilised was part of a supply of a surface-water which Messrs. Burns & Co. had 
had the kindness to collect for me near the Ailsa Craig, in the Irish Channel, and which 
I kept in stock for trial of methods. 250 c.c. of this water when treated in my 
apparatus with hydrochloric acid, as usual, gave 97 ‘72 mgrms. of carbonic acid per litre. 
The same water, when simply boiled in the same apparatus (in a current of air) 
without addition of acid, gave 82 T 6 mgrms. The residue left, on addition of acid, gave 
18'56 mgrms. The sum of the two instalments of carbonic acid obtained, 82T6 + 18'56, 
comes up to 100 '72 instead of 97 '7, which is a fair approximation, because 
the corresponding difference of the absolute weights of carbonic acid determined is 
only Off 5 mgrm. About 84 per cent, of the total carbonic acid had been eliminated by 
mere boiling. (Compare Exp. (1) in table on page 111.) 
The experiments reported on in this chapter, so far, do not exhibit the degree 
of regularity I should wish them to possess ; but they are sufficient to prove, — 1st, 
that distillation with chloride of barium extracts from sea-water only a fraction even 
of the loose carbonic acid (i.e., of what is present over and above that existing in the 
form of normal carbonates), and this fraction, even under apparently similar conditions, 
has no constant value ; and 2nd, that, supposing a sea-water which contains its carbonic 
acid as bicarbonate, associated or not with free carbonic acid, to be exposed to the air even 
at ordinary temperatures, such a water will soon lose, not only its free, but also part at 
least of the loose carbonic acid of the bicarbonate. Hence, I did not consider it expedient 
to determine the carbonic acid in any large number of the samples of sea-water which 
had been placed at my disposal. 
