THE CLASSIFICATION OF PLANTS 219 
relationship. In grouping species together into genera, 
Linnezus recognizes this bond of union between species ; 
but in his wider groupings he abandoned the natural 
system, and separated the genera into divisions based 
upon the number of stamens in the flower. This gave 
him a convenient key for the identification of plants, 
but it was extremely artificial, and afforded no glimpse 
of their true affinities. A more natural system was out- 
lined afterwards by Linnzus himself, but the working out 
of a natural system of classification has been the work of 
his successors. In English floras the system of Bentham 
and Hooker is generally adopted; on the Continent and 
in many examinations that of Engler is followed. No 
classification can be said to be quite satisfactory, and 
none ever will be until the last word on plant-relation- 
ships has been said, and we are very far from that at 
present. 
The Unit of Classification. Every exact science is based 
on units. In physics and mathematics these are purely 
arbitrary. In botany, as in zoology, the species is re- 
garded as the unit of classification. Each species is sup- 
posed to represent a definite race of plants. Before the 
idea of evolution became prevalent, most people thought 
that the species originated by an act of special creation, 
and remained unchanged and unchangeable throughout 
all time ; others thought that the species was mutable, 
but that the genus was not. Linnzus, for example, 
looked upon the genus as a special creation, but not the 
species derived from it. We know now that the species 
is, at most, only a relatively constant thing. No two 
plants even of the same parent are exactly alike, and 
observation will show that considerable variation occurs 
among the members of the same species. When these 
variations are pronounced and apparently constant, we 
call the variation a variety. What, therefore, constitutes 
a variety and what a species? The question is a very 
difficult one to answer. What one person will call a 
variety another will call a true species. The Floras are 
full of instances of this kind. In fact, there are two 
schools of botanical systematists — those who lump 
together under the same specific name as many varietal 
forms as possible, and those who separate them out, 
wherever possible, as true and distinct species. Bentham 
