(23) 
OBSEEVATIONS ON LIVING FISHES BEOUGrHT BY H.M.S. 
" CHALLENGE E " FEOM TEOPICAL EAST AFEICA TO 
CAPE WATEES. 
By J. D. F. GrILCHRIST. 
The general aspect of the fauna of the warm waters of the southern 
equatorial current and its branches is very different from that of the colder 
waters of the circumpolar Antarctic current. This difference is no doubt 
ultimately due to the differences in the character of these waters as regards 
temperature and salinity, nitrogen-contents, etc. Waters of such different 
characters do not readily mix with each other, and, as many animals seem 
highly sensitive to differences in salinity and temperature either directly or 
indirectly (through the food supply), an effective barrier to migration is in 
many cases established. We may therefore divide the marine fauna of the 
southern hemisphere into two great regions, one which may be called the 
Equatorial, or the Equatorial-Current Eegion, and the other the Antarctic, 
South-Polar, or Antarctic-Current Eegion. The Equatorial Eegion is sub- 
divided by the continents projecting into the southern seas, so that we may 
recognise as divisions of this region the Pacific, Atlantic and Indian sub- 
regions. These divisions are founded on fundamental physical differences 
which are clearly seen in the southern hemisphere, but which probably apply 
in a modified form to the northern hemisphere, where the Polar region is 
broken up by land masses, which are not found in the South-Polar regions. 
Three great regions may therefore be recognised, North Polar, Equatorial, 
and South Polar, more or less sharply divided from each other, and one of 
the most interesting problems of zoogeography is the relation of their 
respective fauna to each other. There is the difficult problem of the 
resemblance, in some cases identity, of species in the North-Polar and 
South-Polar regions, on which a diversity of views still exists, and there is 
the simpler one of the relation of the equatorial fauna to the adjacent polar 
fauna. These two questions are, however, not unconnected, and it is 
possible that the more easily ascertained facts in the case of the second may 
have some bearing on the first. 
The South African seas are well adapted for a study of the relation of 
the Equatorial to the South-Polar regions, and I have already recorded 
elsewhere some striking facts observed in South African waters in this 
