of the Arctic and the Antarctic Faunas. 311 
sufficiently cool water, since this quite agrees with the now 
prevailing conditions of the subsurface-water fauna. 
We must accordingly expect to find in the oldest and 
earlier middle Tertiary a large number of species identically 
occurring in the northern and southern hemisphere—for in- 
stance, in our own region and in South Australia. If we 
simply compare the lists that have been drawn up, this 
certainly does not seem to be quite the case; but if we take 
account also of the remarks made by the authors, we find 
that there is a large number of species closely allied to and 
difficult to distinguish from those of the Antipodes of a similar 
age. When, further, we recall that the paleontologists of 
different countries have very often named their species with 
little or no reference to the work of their colleagues, we 
have to admit that the circumtropicity of the earlier ‘Vertiar 
faunas was so strongly marked that it extended not only 
to the great majority of genera, but, in a great many cases 
(whose number future studies will probably increase), even 
to species. And thus it is certain that the Harly ‘Tertiary 
fauna had an approximately similar uniform expression or 
representation throughout the whole region of its distribution. 
“¢ UNIVERSAL’? FAUNAS, 
The paleontologists of the newer school are for the most 
part strongly opposed to the theory of faunas of cosmopolitan 
or universal character. Of course, if by a fauna of universally 
or uniformly similar character any person means one which 
exhibits in every locality throughout its region a similar 
combination of genera and species, he is asking more from 
Nature than it is reasonable to expect, and neither zoologist 
nor paleontologist can agree with him. We have, however, 
on the earth at the present day two universal or uniformly 
differentiated surface-water faunas which we know thoroughly 
—the arctic and the tropical—and through these we can best 
learn to recognize the characters of a uniformly differentiated 
or “universal” fauna. In the arctic fauna circumpolarity 
is exhibited by a large percentage of species, and we get the 
impression that it has hitherto been prevented in a consider- 
able percentage more by some hindrance or other, and that 
if all distribution-barriers were swept away circumpolarity, 
and therefore universality, would reach the highest possible 
degree of development within the arctic zone. Even in the 
arctic fauna, notwithstanding its pronounced circumpolarity, 
local differentiations have developed, and also local varieties 
and races; but it seems quite certain that, if the local causes 
