ZOOLOGICAL GEOGRAPHY. 201 
or representative genera, and sometimes even of the same or 
representative species, in both the north and south temperate 
provinces, which genera or species are altogether absent from the 
tropical zone. Very much the same thing is found with plants, 
and here we account for it by supposing that the temperate forms 
have migrated along the meridional chain of mountains which 
crosses the tropics in America; and it seems necessary to sup- 
pose that the marine forms which are found in both temperate 
zones have migrated across the tropics in the cold layers of 
water which underlie the warmer ones. It would be possible to 
pass from the north to the south hemisphere in water never ex- 
ceeding 60° F. without ever having to descend so low as 1000 
feet, at which depth seaweed is found, and consequently to which 
light can penetrate. But although this may possibly account for 
the distribution of the fishes, mollusks, and crustaceans, it seems 
hardly a sufficient explanation of the fact that the ancestors of 
the whales and seals which at present inhabit the Arctic and 
Antarctic Oceans must once have crossed the tropics ; for, being 
air-breathing animals, they would be compelled to come to the 
surface for respiration. A still greater difficulty is found in the 
distribution of some of the fresh-water fishes of S.E. Australia, 
Tasmania, New Zealand and Patagonia, which are closely allied ; 
in some cases, indeed, the very same species occurring in S. 
America, New Zealand, and Tasmania. It is true that all these 
fishes go to the sea for a short time every year to spawn, but 
they could not possibly cross the wide and deep oceans that now 
separate these countries, and in order to account for their distri- 
bution, we are compelled to suppose that the physical geography 
of the Southern Hemisphere was at one time very different from 
what it is now. | 
In taking a general view of the subject, we see that the land 
regions, with the exception of Arctogzea, run north and south, 
while the marine provinces run east and west. This is easily 
accounted for by the present distribution of land and water, com- | 
bined with the former submergence of the Isthmus of Panama, 
and, at an earlier date, with the existence of an ocean connecting 
the Mediterranean with Japan. The conclusion is that animals 
would arrange themselves in circumpolar zones, according to 
climate, were it not for the existence of barriers to migration. A 
good example of the effect of barriers is seen in the Gulf of Bothnia, 
which contains exclusively a Celtic fauna, while the White Sea 
and the coast of Norway in the same latitudes have Arctic faunas. 
Another conclusion we arrive at is that the present configura- 
tion of the different regions, both land and marine, dates, in its 
main outlines, back to the Miocene or perhaps even to the Eocene 
period, while during that time the sub-regions and districts 
have in many cases been considerably modified, because we see 
that migration has often taken place along routes that are at 
the present day impassable. Beyond the Eocene period we can- 
not go. There is no evidence that zoological distribution was 
the same during the Mesozoic era as at present, and none that 
were any distinct regions at all in the Paleozoic era. 
