310 Prof. Dr. G. Pfeffer on the Mutual Relations 
either case the forms most capable of resisting cold, and 
therefore best suited to a cooler environment, would remain in 
their old labitat. 
Secondly, although the separation of the surface- and 
subsurface-water faunas in the warmer regions of the earth 
appears to be fairly distinct, the Mediterranean forms an 
exception. Quite half the molluscs of the western shores of 
Norway and fully a quarter of those of the coasts of arctic 
Norway occur in the Mediterranean; but it is quite out of the 
question that in the Mediterranean they live only in the deeper 
layers of constant temperature. It is of course possible that 
faunistic displacements occur according to the season, so that 
Mediterranean animals of northern and arctic character live in 
the surface-water only in winter; on comparatively steep 
shores the distance, for many at least, would not be too long. 
Unfortunately I know of no data on this last point so far 
as it effects the benthos animals. Nevertheless the state of 
affairs in the Mediterranean confirms our conclusion that the 
separation between surface- and subsurface-fauna, whether it 
be actual or only potential, is not of supreme importance. 
CIRCUMTROPICITY OF THE EARLIER TERTIARY FAUNA. 
And now that nothing more stands in the way of the 
recognition of our Early ‘Tertiary fauna as one of tropical 
habit, we come to the question of the development of its 
circumtropicity. The surface-water fauna of our tropics is 
circumtropical, and this holds true of by far the greater 
number of genera and even of many of the species. ‘I'he 
similarity of many species from the Indo-Pacific and West- 
Indian seas, and, on the other hand, from the eastern and 
western shores of Central America, proves to us that the 
modern separating conditions have not sufficed to efface 
circumtropicity, and that, if these separations were suddenly 
to disappear, the circumtropicity would be expressed through- 
out the whole region to a much more perfect degree. 
Thus the surface-water fauna of our present tropics is the 
remains of the Early Tertiary fauna shrunk back into the 
equatorial zone; it lives in approximately the same thermal 
conditions as the ancestral fauna enjoyed in our latitudes. 
On what possible grounds, then, can it be asserted that 
circumtropicity was less developed in the Karly Tertiary 
fauna than in the present surface-water fauna of the tropics ? 
No one doubts that the subsurface-fauna of the Early 
Tertiary, whether it was actually or only potentially deve- 
loped, may have been distributed over the whole area of 
