118 MR. BENTHAM ON MALVACEE 
five leaves each, the outer one forming the calyx, the next the 
petals and stamens, and the inner one the carpels. But when we 
consider that in the whole group of Columnifere the petals are 
either perfectly distinct and sometimes distant from the staminal 
column, or, if they adhere to it near the base, the attachment is 
superficial only, the vascular systems remaining perfectly distinct, 
and that even this attachment is wanting in those genera where 
dédoublement is most relied upon, we must have something more 
than mere conjecture, some strong cases of intermediate structure, 
to counteract the evidence of our senses, and establish in theory 
that two totally disconnected organs are, in fact, branches of one 
organ. 
It is well known that the (homological) leaf is very ready to 
ramify laterally—in its own plane; but, as far as my experience 
extends, ramification in a direction at right angles to that plane, 
either by the production of excrescences from either surface, or by 
anything approaching to a splitting or separation of the two sur- 
faces, is confined to the three following categories :— 
1. The production of epidermal excrescences, such as hairs, 
prickles, &c., never converted into real organs. 
2. Prolification, the result of plethora or of some accidental de- 
termination of sap to particular points, resulting in abnormal 
foliaceous appendages, or adventitious roots and buds, which may 
become independent individuals, but never efficient organs of the 
mother-plant. 
3. The production of petiolar glands, which alone can have any 
bearing on our present ease. These glands, which I have called 
petiolar to distinguish them from several other bodies bearing 
usually the same name of glands, are not, however, strictly con- 
fined to the petiole. In most stem-leaves where they occur, there 
are two or a single one of them at or near the summit of the pe- 
tiole or the base of the limb ; but they are sometimes more nume- 
. rous, irregularly placed on the petiole, rarely on some of the prin- 
cipal veins or in their axils, but not unfrequently on the margin 
of the leaf at the extremity of the principal veins ; and they are 
usually disk-shaped, concave, or cup-shaped. In bracts they some- 
times attain a size very large in proportion to the rest of the bract. 
In the petal they are very apt to assume the form of an entire or 
two-lobed scale at the base of the lamina or on the claw, some- 
times as large as the rest of the petal, sometimes reduced to a 
mere concavity in the petal, or to a slight discoloration or altera- 
tion in the texture of its surface. In the stamen, according to 
