NEW HEBRIDES ISLANDS. 
99 
Scarcely a ray of light can pierce this dense mass of branches and leaves joined together by a 
network of climbing plants ; and the rivers, whose bed must be the only practicable highway, 
find their way to the sea 
“ With mazy error under pendant shades.” 
The traveller, hitherto unaccustomed to such luxuriant vegetation, is struck with awe in 
presence of such a display of the immense productive forces of nature, and feels that man 
is only a part of “one stupendous whole.” The tropical forest excludes him from the land 
which he considers as his especial inheritance, and seems to say, “Here is no place for you!” 
On the same day we sighted the southern shores of the larger island of Malikolo, and 
then commenced our cruise across the sea which stretches from the New Hebrides to Torres 
Strait. This is a remarkable basin whose depths we were the first to explore, and we therefore 
proposed to call it the Melanesian Sea. The most interesting result of our soundings was 
the fact, on this occasion observed by us for the first time, that the temperature of the water, 
which, as a general rule, is found to decrease gradually from the surface down to the bottom 
—rapidly near the former, very slowly towards the latter—may, under certain conditions, be 
arrested at a certain depth, and remain stationary between that depth and the bottom. Thus 
our observations on the distribution of temperature in the Melanesian Sea showed that its 
temperature, from an average of 26° C. at the surface, falls to 5 0 C. at a depth of 400 fathoms, 
to 2 0 .5 C. at 900 fathoms, and to i°.8 C. between 1200 and 1400 fathoms from the surface, 
and then remains stationary, or almost stationary, down to the bed of the sea at a depth 
of over 2000 fathoms, forming a bottom-stratum of nearly uniform temperature about one 
mile in thickness. The explanation naturally suggested by this curious phenomenon is, that 
the depths of the Melanesian Sea must be cut off from the depths of the Pacific Ocean by a 
submarine ridge or area of elevation, which prevents the colder bottom-stratum of the latter 
from communicating with the former. A glance at the chart shows that the Melanesian Sea 
may be considered, hydrographically speaking, as an almost land-locked basin. Its length 
from east to west is about 1200 miles, with an average width of about 300 miles. It is bounded 
on the north by the chain of islands which extend from Papua to the Hebrides, and on the 
south by a labyrinth of coral reefs which connect the peninsula of Cape York with the island 
of New Caledonia. The soundings of H.M.S. “Challenger,” combined with those of the 
U.S.S. “ Tuscarora,” indicate the existence of a narrow channel, from 1500 to 2000 fathoms 
in depth, between New Caledonia and the New Hebrides, and a similar channel of about the 
same depth will probably be found extending along the west coast of New Caledonia. 
These are a north-easterly extension of the channel which separates the Kermadec 
Islands from New Zealand, and they constitute probably the only deep-sea communication 
between the Melanesian Sea and that portion of the Pacific where, at a depth of 2900 fathoms, 
we found a temperature of o°-5 C. The decrease of temperature in the Melanesian Sea is 
arrested at a depth varying between 1200 and 1400 fathoms, which is apparently the maximum 
depth of its communications with the Pacific. The same phenomenon we found subsequently 
exhibited in a much more striking manner in the nearly land-locked seas of the Indian 
