THE PIllNCIPLES OF BIOLOGY. 
19 
the shade near the stem, the terminal leaflet is largest, in 
Helosciadium, with less shading, the proximal leaflets are largest. 
This is evidently quite akin to the varying thickness of leaves 
on one and the same tree, according to whether they are 
growing on the outside, exposed to full sun-light, or in the 
interior in comparative shade. In the beech, for example, 
the former can be as much as three times the thickness of the 
latter. 
Before quitting this subject of leaves, we ought, while 
clearly acknowledging that they furnish many of the most 
beautiful illustrations of relations to external forces, never¬ 
theless to draw attention to the manifold dangers of inductive 
biology founded upon such variable members as leaves are. 
In many cases their forms may be due, not to present, but to 
ancestral conditions of life, and the plant may not be equi- 
plastic in morphology and in physiology. Thus, while we 
acknowledge that radial (peltate) leaves have erect leaf-stalks, 
and are, in general, radial, it is not true that radial leaves 
tend to be peltate. The great majority of erect growing leaves 
are sword-shaped or grass-like. 
The consideration of the evolution of flowers opens up a 
series of most interesting and important facts. Referable as 
floral parts are to the leaf-type, they have deviated as a 
rule widely from it, though the deviation of the bract of 
Helleborus fcetidm from the form of the radical leaves of the 
same plant is at least as great as the flower shows. The 
plasticity of the flower is about equivalent to that of the leaf. 
Floral branching, however, commonly differs widely from 
stem-branching, and the former may be present where the 
latter is not. 
The fundamental arrangement of leaves is no doubt 
primitively spiral, and the primitive arrangement of floral 
parts followed the same type, and produced the cone. 
The radial, or cyclical, arrangement for leaves, excepting on 
a very restricted scale, is rare ; the cyclical arrangement of 
floral parts is, for some part or other of the flower, well-nigh 
universal, at least in external appearance. The study of the 
organogenesis of the flower would suggest, however, that the 
cyclical state is an after state even in the individual flower, 
and that the parts, even when they are joined together, may 
be, in a limited sense, of different ages. But if equivalence 
comes about at all there is a tendency for it to be produced 
at progressively earlier periods, and hence fundamental spirals 
can ultimately give origin to true cyclism. 
No doubt an erect growing flower would tend to be radial, 
but Pinguicula and Viola , both fundamentally erect-growing 
