546 Critical Periods in the History of the Earth. (September, 
the Tertiary. It is probable that the upheaval which destroyed 
the Cretaceous sea went much beyond the condition of things 
afterwards; that just at this interval the land was higher and 
rger than in the Tertiary; that, in short, this was again a 
continental period, and probably a period of greater cold than the 
subsequent Tertiary. 
e change in physical geography, then, was immense, but in 
most places by bodily upheaval, not by crumpling of the strata; 
and therefore the usual sign of such change, namely, unconform- 
ity, is often wanting. The change of climate all over the Amer- 
ican continent was no doubt very great, and the change in or- 
ganic forms correspondingly great everywhere and in all depart- 
ments ; but this was especially true of all water-inhabiting species 
_ in the region of the old Cretaceous interior sea, for here there 
was a transition, not only in climate but from salt to fresh water 
through the intermediate condition of brackish water. The 
Cretaceous marine species rapidly disappeared, partly by exter- 
mination and partly by transmutation into fresh-water species, 
as has been observed, recently, to take place in some crustaceans 
under this change of conditions.1 The Tertiary fresh-water 
species quickly appeared, partly by transmutation from the pre- 
vious marine species and partly by transportation in various 
ways from other fresh-water lakes. But all this occurred in 
some places without the slightest break in the continuity of the 
strata. 
The great change of climate and other physical conditions per- 
haps sufficiently explain the change in invertebrate species, but 1t 
is impossible to account for the somewhat sudden appearance of 
mammals in the lowest Tertiary, except by migration from other 
regions where they had existed in late Cretaceous times, having 
originated there by derivation in the usual way. That marsupi- 
als existed somewhere in Cretaceous times (though possibly not 
in America or Europe) there can be no doubt ; for they lived, 
we know, in the preceding Jurassic and the following Tertiary, 
and they exist now. It is from these rather than from Ora 
ceous reptiles that Tertiary mammals were doubtless- derived ; 
and this derivation took place probably at rapid rate in the ae 
est Cretaceous or during the lost interval, in some unknown ag . 
ity, whence they migrated into the Tertiary lake region of : 
United States during the interval. This migration came weet 
probably from Northern Asia, for it must be remembered th , 
1 Arch. des Sciences, November, 1875, p. 284. 
