EVOLUTION OF THE INFLORESOENCE, 519 
and “segregration’’ were used in connection with the formation of 
inflorescences. These words were used expressly to infer that inflorescences 
could be formed in two fundamentally different ways, in respect to the 
separation of the individual flowers composing the cluster from the foliage 
leaves. Let us take a hypothetical tree or shrub which has groups of foliage 
shoots each axis of which is terminated by a single flower. That is to say, 
the plant has solitary terminal flowers, but the leafy shoots bearing flowers 
tend to arrange themselves in groups. Such a group can hardly be considered 
an inflorescence, as the leaf-structures it contains are all foliar. Imagine the 
shortening of these shoots and the reduction of all the foliage leaves to bracts. 
The group of flowering shoots would now become an inflorescence. The 
author believes that flower-clusters have in several instances arisen by this 
method of segregation, e. g., intercalary inflorescences, already briefly referred 
to in the introduction and to be treated of more fully in a later section of 
this paper. 
But this does not seem to be the general way in which inflorescences have 
evolved. There is reason to think that the majority of inflorescences have 
arisen in a different manner, not by the segregation of a number of 
solitarily borne flowers, but by the production de novo of lateral flowers 
alongside the terminal one. The initial stage in the evolution of a flower- 
cluster from a single terminal flower may have been induced by a super- 
abundanee of reproductive material, more than the shoot could utilise in the 
formation of a single flower. One or more lateral flowers might thus be 
pushed out from the axils of the leaves below the terminal. Once this 
tendency for the production of laterals became fixed, then the increase in 
number of flowers composing the inflorescence might be considered to take 
place at the expense of the size and number of parts of the individual flower. 
An inflorescence of this type might thus be regarded as the equivalent of a 
single terminal flower. The facts suggesting this point of view will be dealt 
with in the next section treating of the “ solitary terminal flower." 
Another point in the study of the Inflorescence, requiring some clearness 
in thought, is the recognition of an essential difference between floral and 
vegetative branching. Failure to grasp this leads to false comparisons and 
reasonings. In Angiosperms the racemose manner of branching of the 
vegetative shoot is the simpler and would appear to be the primitive. The 
main (plumular) axis continues growing indefinitely and the lateral shoots arise 
from it in the ascending order. Cymose vegetative branching is derived 
from it by the main axis ceasing to grow and by one or more of the upper 
lateral buds taking on, as it were, the function of the original main axis. If 
only one of these buds continues the growth in height or length of the plant, 
the so-called sympodial system of branching results, which is of common 
occurrence. The apparently simple main axis of such a plant is really a 
sympodium of a number of axes of an ascending series. 
