154 TRANSACTIONS OF THE [FEB. 12, 
first, because, in Botany at least, it has not been much consid- 
ered. Other assigned causes, environment, modifications within 
the germ-cell, the “‘ survival of the fittest,” have all in my opinion 
more or less to do with this most intricate problem. Their 
consideration, however, does not come within the scope of my 
present communication. ' 
From the conditions thus surrounding the philosophical con- 
sideration of species, it is by no means strange that widely differ- 
ent opinions as to the relationships of different groups should be 
maintained by different authors working from different material. 
We cannot hope to alter these conditions, nor would there be 
any gain in unifying scientific opinion, for then there would be 
little progress in knowledge. 
It seems probable that some advantage may be drawn from a 
method which, though empirical, has still a considerable scien- 
tific basis, and which has been more or less employed by many 
systematists. I refer to the known geographical distribution of 
closely related or “intergrading” species. It has been found 
quite possible to divide the surface of the earth into biological 
areas and subareas, variously designated, these areas being 
characterized by different climates and inhabited by floras and 
faunas of a different general facies, and their outlines can be so 
drawn as to include the natural geographical range of many 
species, which thus become characteristic of the areas. Now if 
we have two closely related organisms, each inhabiting its own 
area, the conclusion that they are distinct species is strength- 
ened by giving us an artificial aid in their delimitation. 
The explanation of the occurrence of identical or closely re- 
lated species in widely separated geographical areas affords 
another most interesting problem. It is, I believe, in many 
instances, to be solved by reference to their origin in remote 
geological epochs rather than by migration in comparatively 
recent time. The basis for this explanation lies not so much in 
the inherent improbability of extensive migrations of plants in 
a body, which would have to be invoked for instance to satisfy 
the fact of the occurrence of numerous circum-boreal species in 
the extreme southern part of South America or the presence of 
numerous Texano-Mexican types in Southern Brazil and Para- 
guay, or the similarity of the Eastern North American flora to 
that of Japan, or that of the East Indian and West Indian 
Archipelagos, as in the uniformity of the floras of ancient geo- 
logical epochs over the entire earth’s surface before the great 
differentiation into climatic areas took place in the Cretaceous 
and Tertiary. This general uniformity is clearly indicated by 
our present knowledge of paleeobotany, and has repeatedly been 
