28 
Janies Small. 
these two petals grow at the same rate as the others, but the organic 
connection between them ceases at an early stage, as in Taraxacum 
(Pig. 9, 31, 32, 33), the mature floret showing a cross-section as in 
Fig. 9, 21, with the free margins of these two petals closely adpressed. 
In the bilabiate corolla these two petals usually fuse at an early 
stage and then cease to grow, as in Calendula (Fig. 9, 30, 29, 28) the 
mature floret showing a cross-section as in Fig. 9, 20, with the free 
margins of the two lateral petals overlapping. In this type, however, 
the many variations in the degree of fusion and degree of develop¬ 
ment of these two petals give the various modifications described in 
Section B. 
The primordial meristems are free at first in all cases giving 
five protuberances from the top of the primordium of the flower ; 
these meristems then fuse to give a zone which is complete in the 
tubular corolla, interrupted regularly between the two posterior 
petals in the ligulate corolla, and interrupted regularly or irregularly 
in the bilabiate types. The development of the filiform corolla with 
2-4 teeth at the apex has yet to be elucidated. 
In ontogeny, therefore, as in phylogeny, the five-lobed tubular 
corolla is the primitive type, and from it arises two forms more 
distinct in their development than in their form when mature. 
Causal Morphology —Experiments have been carried out along 
the lines of causal morphology as suggested by Lang (47) and 
although more or less unsuccessful may be of interest. 
Apart from an internal modification of the constitution of the 
organism there are two factors which may influence the production 
of ray florets, pressure due to crowding of the florets in the bud and 
variation in food supply. 
Taking pressure first, this factor is not a very probable one as 
in the development of the capitulum the rays are differentiated from 
the disc before the primordia have increased sufficiently to exert a 
pressure upon one another (see also Church, Pt. II, p. 114). They 
always mature late ; the primordia of the petals in the ray are 
undeveloped when the first row of disc florets have the corolla, 
stamens and gynoecium differentiated. 
Experiments on the influence of pressure were made by tying 
twine around capitula or placing elastic bands around them so that 
as they expanded in the later stages of development they were 
subject to a pressure from within outwards in the case of the twine 
and partly from without towards the centre of the capitulum where 
