eo 
- NATURE 
Dae 
= 
= 
Sas —4i2 
shes 3 — 
differences observed both by the Norwegian Expedition 
-- and by Jacobsen in the aération of water from different 
depths are so slight. Indeed, the Challenger analyses 
show exactly the same result for Antarctic waters. As 
the amount of nitrogen is probably not exposed to diminu- 
tion, it may be taken as an indication of the temperature 
at which the water was last exposed to the air, and may 
therefore be used as a check on the depth from which the 
water has been brought, more especially ia tropical 
regions, where the temperature at the surface is very 
different from what it is either at the bottom or interme- 
diate depths. For this purpose, however, we require more 
observations on the absorption of atmospheric gases by 
sea water, especially at low temperatures ; and, further, 
any conclusions drawn must be inaccurate, in so far as 
we do not know the barometric pressure of the atmo- 
sphere to which the water has been exposed. This is a 
very important element, for the water at the surface of 
the Antarctic Ocean is exposed to a much lower mean 
barometric pressure than at any other part of the globe, 
whether Arctic, Temperate, or Tropical, consequently a 
-carefully-made determination of the nitrogen in a bottom 
water would, when taken in connection with the tempera- 
ture, indicate whether it came from Arctic or Antarctic 
sources. ‘This difference would certainly amount to I cc. 
per litre, which could easily be determined with careful 
work, 
The great value of the results obtained by Dr. Tornoe 
in this department of his work make it all the more to 
be regretted that through mechanical mishaps so many 
samples, involving much time and work, should have 
been lost. 
The form in which the results are presented to the 
reader might be improved by the addition of one or two 
columns to the table. It includes the analyses of samples 
collected in the three summers, 1876, 1877, and 1878, and 
it would have been more useful to the reader to have 
found the date of collection in the first column than a 
series of consecutive numbers. The second column is 
the “Station No.,” and it is important as facilitating 
reference to other results obtained atthe same place. By 
its means the writer was enabled to refer to Prof. Mohn’s 
papers in Petermann’s Mitthetlungen, and from them 
to supply a column giving the depth of the sea at the 
station. The omission of this information from the table 
made it impossible to distinguish between bottom water 
and water from intermediate depths. Another column 
might also with advantage have been added, giving the 
volume of oxygen in cubic centimetres per litre. 
In the second chapter of the work Dr, Tornoe treats 
of the carbonic acid dissolved in sea-water, and here 
also he adds very materially to our knowledge. The 
first reliable information on the subject was obtained 
by Jacobsen on board the Pomerania. He rejected 
the gasometric method, having recognised the uncer- 
tainty which attached to the elimination of the carbonic 
acid from sea water by boiling under reduced pressure 
and adopted the method of determining the carbonic acid 
directly, as soon as the sample of water was brought on 
board, by boiling it down nearly to dryness, and drawing 
a current of air through it, which conveyed the steam and 
carbonic acid into a suitable receiver charged with baryta 
water. In the Challenger substantially the same method 
was employed, with this important addition, that an 
excess of a saturated solution of chloride of barium was 
added to the water before distilling. By precipitating 
the sulphates, their effect in reducing the tension of the 
carbonic acid was destroyed, and also the liquid was got 
into a condition in which it boiled calmly, without 
bumping, until almost quite dry. The object aimed at 
was the determination of the carbonic acid present in the 
water in the free or /al/-bound state, to the exclusion 
of that present as neutral carbonate. There is no doubt 
that this was successfully accomplished, and the experi- 
ments made by Dr. Tornoe furnish satisfactory evidence. = 
“Tn order to ascertain whether the decomposition by 
boiling of the neutral carbonates in sea-water also took 
place to a considerable extent when insoluble sulphates 
were present in that fluid, I made a few experiments by 
Buchanan’s process. From several samples of sea-water, 
which, examined by the method I adopted, were found to 
contain 96 mgr, of carbonic acid per litre, I succeeded, by 
evaporation to dryness, after adding a solution of chloride 
of barium, in liberating about 50 mgr. only, with a solitary — 
exception, when the amount exceeded 50 mgr. per litre. 
The proportion of carbonic acid expelled was accordingly 
not much greater than that determined by Buchanan in 
water from equatorial seas, and but a few milligrammes 
in excess of what the carbonic acid forming bicarbonates, 
according to trustworthy observations, should have been ; 
of the carbonates said to be present in the residue I failed 
to detect any trace.” This experiment shows that what 
was sought was really obtained, namely, the determina- 
tion of the carbonic acid wot present as neutral carbonate. 
The method finally adopted by Dr. Tornoe is an ex- 
ceedingly ingenious one, and has the great advantage of 
giving both the free and the bound carbonic acid. It con- 
sists in adding to the sample of water a measured quantity 
of acid of known strength, driving off the liberated carbonic 
acid by gentle heating, and collecting it in baryta water 
of known strength. When the operation is finished, the 
excess of acid in the boiling flask and the excess of alkali 
in the receiver are separately determined. The amount of 
baryta neutralised gives the total carbonic acid, while 
the amount ofacid neutralised gives the amount present 
as neutral carbonate. 
It does not seem to have occurred to Dr. Tornoe that 
his method of determining the carbonic acid might be 
combined with the boiling out of the oxygen and nitrogen. 
If to the sample from which the gases are to be extracted 
by boiling under reduced pressure be added sufficient 
acid to more than neutralise the carbonates, and the 
boiling be then continued as if for the elimination of the 
oxygen and nitrogen, the whole of the carbonic acid 
should be obtained along with these gases, while the 
excess of acid in the flask could be measured when the 
operation was finished. This process would have the 
advantage that oxygen, nitrogen, and carbonic acid would 
be collected in one operation. It would be necessary to 
make the tube in which the gases are to be preserved 
larger than is at present usual, but a volume of 100 cubic 
centimetres would suffice for a yolume of 800 or 900 cubic 
centimetres water. 
By this method the carbonic acid was determined in 
seventy-eight samples of water from different parts, It 
is somewhat of a pity that the determinations were not 
made on board when the samples were fresh, though 
there is no doubt that in the case of sea waters which 
contain only traces of organic matter, the amount of car- 
bonic acid is not sensibly affected by keeping. The 
results obtained are very uniform, and he gives the 
following average formule : 
52°78 + 0°83 mgr. per litre 
for the carbonic acid forming carbonates with a probable 
error in a single observation of + 0°662 per litre ; and 
43°64 + 0°16 mgr. per litre 
for the carbonic acid, forming bicarbonates with a 
probable error in a single observation of + 1°26 mgr. per 
litre.” 
Touching the uniformity of these results, it must be 
observed that the samples would probably be all at nearly 
the sime temperature when examined, while they would 
be collected at different and lower temperatures. Hence 
the fact of keeping would tend to produce uniformity in 
the results. Hence also there is no mention of tempera- 
ture in his average formule. Now although the law 
regulating the absorption of carbonic acid by sea water at 
