5 18 THE AMERICAN NATURALIST. [Vor. XXXIX. 
The succession exhibited in this table brings into prominence 
the idea that such families as the Leguminosz, Vitacez, Eri- 
cacex, Magnoliaceze, Convolvulacez, etc., are inferior to the 
Salicaeze which is, in turn greatly inferior to the Cyperacex 
from an evolutionary point of view ‚when considered solely in 
the light of their geological succession ; but the sequence which 
is thus established is found to be utterly at variance with that 
usually adopted on the basis of morphological evidence which 
for obvious reasons must be regarded as the more trustworthy 
guide. Such discrepancies between the morphological and the 
geological succession, in our opinion, should not be so great as 
seem to be suggested by the available data, and no doubt they 
will diminish as our knowledge of the fossil representatives 
becomes more nearly complete, but that such discrepancies may 
be anticipated, and that they are to some extent to be regarded 
as a normal result of rapid evolution along many diverse lines of 
development, is a reasonable supposition. The Mesozoic age 
was a period of very great diversification among plants in con- 
sequence of the profound physical changes which had taken 
place in the atmosphere as well as in the configuration of the 
surface of the earth, and the fact that as a product of previous 
development, plants were in a condition to be widely and pro- 
foundly influenced by comparatively slight modifications of exter- 
nal conditions. The very great number of families and genera 
and even species of angiosperms which abruptly appeared in the 
Cretaceous, presenting as they did types of vegetation wholly 
unlike those of previous geological periods, indicates that the 
parental forms must have had their origin in the early Meso- 
zoic and possibly as far back as the Permian, since our accept- 
ance of the general principles involved in De Vries' mutation 
theory does not permit us to consider as possible, such phe- 
nomenal transitions as would be involved in the application of 
that theory to the case under consideration ; while on the other 
hand all paleontological evidence, as well as the evidence derived 
from existing types, confirms us in the belief that such highly 
organized forms could have arisen in the main, only through a 
long series of transitional forms occupying great periods of time. 
For the same reasons also, we are led to believe that the great 
