Journal of Proceedings. 
lxxxix 
which come after will be lateral or axillary. If the terminal bud be 
abortive, or if it continue to grow in prolongation of the stem, all the 
flowers will be lateral in their relation to the main axis, but may be 
either terminal or lateral in their relation to the branch which bears 
them. Let us look at the results of these contingencies. 
In the King-cup (Caltha palustris) we have a stem whose terminal 
bud becomes a flower, and a bud in the axil of the uppermost leaf or bract 
is also a flower bud. This, the lateral, axillary or sub-terminal flower, 
does not expand so soon as the terminal flower. It is on a simple pedicel, 
which bears no bract nor any modification of a leaf below or outside the 
calyx, nor has it any capacity for branching. It is well to observe these 
particulars as they serve to distinguish this mode of inflorescence properly 
definite from any which can fairly be called centrifugal. From the axil 
of a lower leaf grows another branch, which, like the main axis, bears a 
flower at its top with a leaf or bract below it, in the axil of which is a 
sub-terminal flower. 
It may appear that in such cases as the above, where there is a stem 
with many branches, each of which is capable of bearing several flowers, 
every such branch ought to be considered as a separate individual, and 
that the relation of such branches to each other and to the main stem 
which bears them ought to be regarded as a matter quite distinct from 
the arrangement of the flowers upon each peduncle. Looking at the 
subject from this point of view, I arrived at the conclusion that the 
arrangement of individual flowers on a peduncle or flowering branch is of 
more importance as to natural affinity than the mode in which peduncles 
or flowering branches are placed on the main stem. 
The inflorescence of the King-cup is definite, but not centrifugal, for 
(1) the pedicel of the uppermost lateral or sub-terminal flower is simple, 
and with no capacity for branching ; (2) the same uppermost lateral or 
sub-terminal flower does not commonly open so soon as that which 
terminates the branch immediately below it. In both of these particulars 
the inflorescence of the King-cup is in striking contrast to that of the 
Stitchwort ( Stellaria ) or the Campion (Lychnis), which is strictly 
centrifugal. 
That the annual flowering stem of the King-cup, with all the branches 
growing on it, should be regarded as forming collectively a single system, 
all parts of which are related to each other, may be inferred from the 
facts of an apparent order existing in the expansion of their several 
flowers, and of the greater capacity for branching observed in those 
branches which spring from the lower part of the stem in comparison 
with those from its upper part- 
The flowering stem of the Columbine (Aquilegia vulgaris ), with its 
branches, forms an inflorescence more compact than that of the King¬ 
cup, but with a general resemblance to it; differing, however, in that the 
uppermost lateral flower-stalk is not a simple pedicel but a peduncle 
capable of branching, usually bearing two bracts below the flower, and 
often with flowers in their axils on stalks which may be properly called 
tertiary axes. In that the flower on the top of the main stem is the first 
to open, and that which terminates the secondary axis comes before that 
on the tertiary axis, this plant resembles those whose inflorescence is 
properly centrifugal; but differs, inasmuch as of several branches or 
secondary axes that which is lowest is often longest, bears most flowers, 
and is forwarder in their expansion than any of the branches of the same 
rank which grow above it. A source of error exists, however, in that 
while three or more branches rise to a level with the main axis, opening 
their flowers in a' centripetal order as compared with each other, another 
branch or two may sometimes be found lower down which flower later, 
as following a centrifugal order. 
