19 1] COW LES—VEGETATIVE CYCLES 169 
that these successions are and probably must remain the least 
understood of all. There are, perhaps, four great examples of 
extensive regional change, which may be accepted as demonstrated, 
namely: (1) the change from the Carboniferous to the Permian, 
which is made evident particularly through the replacement of the 
carboniferous ferns, fern allies, and primitive gymnosperms by 
the Glossopteris flora and later by the modern gymnosperms; 
(2) the subordination of the gymnosperms to the angiosperms 
in the Cretaceous; (3) the elimination of tropical forms in boreal 
regions in the late Tertiary; and (4) the postglacial invasion 
of southern forms into boreal regions accompanying and following 
the retreat of the glacial ice. Generally it is held that the dominat- 
ing factor in these vegetative successions is climatic change, and 
that this climatic change is chiefly one of temperature. Of this 
there can be no doubt in the case of the changes immediately before 
and after the Pleistocene ice invasion. The constant relation 
€etween glaciation and the development of the Glossopteris flora 
in the Permian makes it likely that the general vegetative changes 
of that epoch also were due primarily to temperature. 
On the whole, however, there has been a general tendency to 
overestimate the influence of temperature as an ecological factor. 
The trend of nearly all experiment has been to show that water is 
of vastly greater importance, and it well may be that the change 
from the atmospheric humidity which seems to have characterized 
the Carboniferous to the aridity which seems to have characterized 
the Permian had more to do than did the decreased Permian tem- 
peratures with the elimination of the carboniferous flora and with 
tts replacement by mesozoic forms. The most puzzling of the great 
vegetative transformations of the past was the sudden change from 
the dominantly gymnospermous forests of the Jurassic to the domi- 
nation of the world by angiosperms in the Cretaceous. We know 
that after the Permian there was a gradual climatic amelioration 
toward genial conditions similar to those which characterized the 
Carboniferous: this amelioration seems to have culminated in the 
Cretaceous, which, like the Carboniferous, was also a period of 
extensive base-leveling. Very probably the high temperatures 
and the great atmospheric humidity of the Cretaceous gave con- 
