C.—GEOLOGY 713 
placed Angiosperms in all their details.’ Thus the Upper Cretaceous- 
Tertiary flora must have been preceded by one in which highly developed 
angiosperms were not uncommon, and the suddenness of the institution 
of the later flora must have been due to some factor that allowed the 
angiosperms scope, and inhibited the other elements in the Lower Cre- 
taceous flora. What the factor was we do not know, but the suggestion 
that it was the ‘Cenomanian transgression ’ is one of great value. Berry 44 
thinks it ‘ futile to speculate about the problem at the present time,’ and 
advocates a more intense study of Mesozoic floras, especially in the 
tropics, and the upland rather then the swamp flora. There is much in 
this criticism, and I would further add that the volcanic ashes of these 
days should be searched for traces of this upland flora. 
Nevertheless the suggestion that the ‘ Cenomanian transgression ’ was 
responsible for the alteration of conditions that gave the angiosperms the 
* boost ’ which initiated their present dominance is, I think, valuable and 
worth exploring. It happens to coincide with an idea I have held since 
I first stood on the edge of the Colorado River, at the bottom of the canyon, 
and looked at the sand and mud that was being swept along. One of my 
difficulties, at the time, was to understand the conditions of deposit of the 
Greensand, Gault and Chalk formations in England. Prof. E. B. Bailey *® 
had just published his suggestion of the Chalk being produced off a desert 
coast, and the desert conditions persisting after the beginning of Tertiary 
times. I found it difficult to reconcile the abundant plant remains of the 
upper beds of the Lower Greensand, and even of the Gault itself, with 
desert shores. But the geography of the Colorado River seemed to 
explain the situation. The river passed through areas of barren, desert 
country, and areas with abundant vegetation, carrying sand from the 
one, plants from the other, down to its estuary. Such a stream might 
be expected to produce a kind of deposit like the Greensand with smooth- 
grained, current-bedded sands, and abundant plant remains. A depres- 
sion of the drainage area up to the desert zone would widen its estuary, 
prevent sand from being swept into the areas once reached, and permit 
the accumulation of only fine clays or only pelagic deposits in these areas. 
The succession of plant-bearing, current-bedded sandstones would be 
followed by clays and calcareous oozes comparable with the succession 
of the Upper Cretaceous rocks in S.E. England ; and the shores might 
be deserts during the production of the calcareous oozes, and so permit 
the latter to accumulate close in-shore. After elevation, the desert 
conditions might still persist. 
But suppose the land prior to depression was clothed, in its non-desert 
track, with a vegetation consisting of an older flora holding i in check, by 
competitive power, a younger race; or, to put it into actual fact, a Mesozoic 
flora, similar to that found elsewhere without angiospermous associations, 
but here competing successfully or, at any rate, holding a balance with 
its angiospermous units. 
The depression of such a land would cause the flora to migrate, and 
4 “Revision of the Lower Eocene Wilcox Flora,’ U.S.G.S. Prof. Paper 156, 
1930, p. II. 
45 Geol. Mag., 1924, p. 102. 
