238 
PROFESSOR FORCHHAMMER ON THE COMPOSITION 
had increased before. At 5400 feet it has a greater quantity of salt than any of the 
upper specimens has shown. In the series from 22° 37' S. lat. the surface has a high 
number, higher than any corresponding sample from the North Atlantic, it sinks a little 
at 900 feet, but rises at 1800 feet to a quantity of salt which does not occur in any 
other place in the whole Atlantic, not even the maximum of the Mediterranean, and 
we know only the Red Sea which exceeds it ; it is as if the water of the Red Sea were 
transported to this submarine current. I thought there might be a fault in the deter- 
mination of the chlorine, and repeated it; but the difference was very insignificant, 
being in the one case 23-187, in the other 23-191, the mean being 23-189. I thought 
that by some accident some salt might have come into the instrument by which the 
water was taken, and I made a complete analysis of the water, but the different sub- 
stances which were determined showed but slight differences from the normal propor- 
tions, viz. — 
Chlorine. 
Sulphuric acid. Lime. 
Potash. 
Magnesia. 
22° 37' S. lat., 1800 feet . 
. 100 
11-59 2-77 
2-14 
11-29 
South Atlantic .... 
. 100 
12-03 2-91 
— 
10-96 
might perhaps be owing 
to an evaporation in the bottle, 
but then the 
bottle was 
full, and cork and sealing-wax were sound, while about one-seventh of its whole con 
tents must have been evaporated to explain the difference. If there is any mistake in 
this curious observation, it must probably have been caused by a negligence which left 
the instrument for taking the water from the deeper part of the sea partly filled with 
sea-water, exposed to evaporation in tropical heat, and sent it down without being 
cleaned. I should hardly think that such a fault could have been committed, and we 
must hope that new experiments will confirm the fact. The series of observations from 
0° 15' S. lat. belong in fact to the same kind, by the alternation of stronger and weaker 
sea-water in different depths ; but the curious and surprising fact in the observation 
from 22° 37' S. lat. is, that in the whole Atlantic Ocean we do not know a single place 
where water with that quantity of salt occurs. The next specimen, from 22° 37' S. lat. 
and a depth of 2700 feet, is very nearly the same as that from 900 feet, and that from 
5400 feet very near that from the surface of the same place. 
It appears thus that the water of the North Atlantic Ocean, between the southernmost 
part of Greenland and the equator, decreases in salinity with the depth, but that this 
curious fact is observed only in the middle bed of the Atlantic, and disappears when 
we approach the shores on both sides of the ocean. As to the cause of this rather 
surprising state, I am still of the same opinion which I expressed when I first observed 
it, that it depends upon a polar under-current. The hypothesis has been published, 
that it depended upon fresh-water springs at the bottom of the ocean, and such an 
opinion might have some chance as long as we only had few observations ; but now we 
have such a number of observations spread over a vast extent of the ocean, that it 
appears to be quite impossible to explain it by springs of fresh water, which of course 
must be more frequent and more powerful near the land, from which they have their 
