hile the humbler Molluscs
survive, are those which--to judge from the nautilus and octopus--love
warm seas, the impression is further confirmed. And when we finally
reflect that the most distinctive phenomenon of the period is the rapid
spread of deciduous trees, it would seem that there is only one possible
interpretation of the Cretaceous Revolution.

This interpretation--that cold was the selecting agency--is a familiar
idea in geological literature, but, as I said, there are recent writers
who profess reserve in regard to it, and it is proper to glance at, or
at least look for, the alternatives.

Before doing so let us be quite clear that here we have nothing to
do with theories of the origin of the earth. The Permian cold--which,
however, is universally admitted--is more or less entangled in that
controversy; the Cretaceous cold has no connection with it. Whatever
excess of carbon-dioxide there may have been in the early atmosphere
was cleared by the Coal-forests. We must set aside all these theories in
explaining the present facts.

It is also useful to note that the fact that there have been great
changes in the climate of the earth in past time is beyond dispute.
There is no denying the fact that the climate of the earth was warm from
the Arctic to the Antarctic in the Devonian and Carboniferous periods:
that it fell considerably in the Permian: that it again became at least
"warm temperate" (Chamberlin) from the Arctic to the Antarctic in the
Jurassic, and again in the Eocene: that some millions of square miles
of Europe and North America were covered with ice and snow in the
Pleistocene, so that the reindeer wandered where palms had previously
flourished and the vine flourishes to-day; and that the pronounced zones
of climate which we find today have no counterpart in any earlier
age. In view of these great and admitted fluctuations of the earth's
temperature one does not see any reason for hesitating to admit a fall
of temperature in the Cretaceous, if the facts point to it.

On the other hand, the alternative suggestions are not very convincing.
We have noticed one of these suggestions in connection with the origin
of the Angiosperms. It hints that this may be related to developments
of the insect world. Most probably the development of the characteristic
flowers of the Angiosperms is connected with an increasing relation
to insects, but what we want to understand especially is the deciduous
character of