EVOLUTION OF THE INFLORESCENCE. ` 661 
By the total suppression of any tertiary branching, and by the complete 
establishment of the ascending order of expansion, a true raceme results. 
This line of progression in flower-grouping culminates in the capitulum or 
head. 
Hence it follows that racemose inflorescences have proceeded from cymose ones, 
the pleiochasium or panicle being the intermediate stage. 
(8) The second tendeney, when followed, leads primarily to the production 
of a compound or continuous dichasium. This can give rise, on the one hand, 
through internodal compression to umbellate and capitate cymes : and, on the ( 
other hand, through one-sided development to sympodial eymes. In con- 
tinuous dichasia the branching (forking) progresses usually, if not invariably, 
at a quicker rate on one side than on the other. If this inequality be accen- 
.. | tuated until the branching becomes one-sided from the commencement, a 
M | continuous monochasium results, and, further, by the flowers assuming a 
| 
| 
lateral position, a sympodium comes to be formed. In this way the so-called 
€ 
scorpioid cymes of the Boraginacez and other families have originated. 
Regarding solitary axillary flowers :— 
(9) These may arise in at least three different ways, all capable of being 
traced back to the solitary terminal flower. 
(i) The solitary axillary flower exhibited by certain arborescent plants. 
There is a tendency in many trees and shrubs to transfer the flowers 
from the upper and principal to the lower and lateral shoots. 
When such a subordinate shoot, terminated by a single flower, 
| loses its foliage and shortens its axis, it eomes into the cateeory of 
the solitary axillary flower, provided the leaf subtending it persists 
(Michelia, Illicium): if, however the leaf falls previously, then the 
flower will appear just above the leaf-scar (Asimina ; Chimon- 
anthus). 
(i. The type commonly, but not exclusively, exhibited by climbing and 
trailing plants, in which single flowers are emitted in acropetal suc- 
cession from the axils of foliage leaves. Such an arrangement can be 
derived from a pleiochasium, in which the leaves on the main axis 
have remained largely foliaceous (unreduced). As the climbing 
habit evolves, the flowers will tend to develop acropetally, rather 
than basipetally, and ultimately the terminal flower will never 
appear ; for the main axis will continue growing the whole season, 
throwing off lateral flowers successively from each leaf-axil. Such 
an evolution for certain solitary axillary flowers has been suggested 
largely by a study of species of Clematis and Convolvulus. 
Gii.) The solitary axillary flower which arises through the reduction of an 
axillary inflorescence to a single flower. Cases of solitary axillary 
2Q2 
