166 
Transactions of the 
["Monthly Microscopical 
L Journal, March 1, 1869. 
that the leaflets in the latter have contracted union at their edges, 
it is difficult, if not impossible, to deny this kind of change to the 
analogous portion of the leaf of the former. The plan of the two 
seems identical. 
(3) . By analysis; that is, by assuming the lobed leaf itself to 
have been disintegrated in those parts where the natural seam or 
suture shows itself, and comparing the detached portions with the 
leaflets of a compound leaf. That lobed leaves will bear this test, is 
best shown by trying the experiment. In those cases where the 
leaflets have coalesced a short distance only from their origin (the 
deeply parted leaves of the botanists), the leaflets may be detached 
almost entire, and resemble all but perfect leaves. Witness, for 
example, the lesii of Passion-flower, or of the Oriental Plane-tree, 
amongst the radiating, or the leaf of the Artichoke {Gynara seolymus) 
amongst the pinnate varieties. In other instances, where the recesses 
are less deep, the leaflets can still be detached, leaving a sufficient 
portion of the normal venation to mark their identity. Take the 
Oak leaf; here the recesses are shallow, yet a number of elliptical 
netted leaflets can be excised from either side of its midrib ? (petiole) 
without difficulty. In some, again, the process of disintegration is 
more difficult, as in Tropseolum majus (Garden Nasturtium), where 
the primary vein first given off from one of two adjacent leaflets, is 
reciprocally common to the integrity of them both. 
(4) . By analogy. All analogy may be said to be in favour of 
our hypothesis. By common consent, the whole of the plant — 
bract, calyx, petal, stamen, pistil — is now considered to consist of 
modified leaves. The truth of this theory is universally adopted 
by all philosophical botanists. If, now, in order to bring the 
structure of the leaf into conformity with the theory, that by its 
modification it might become converted into an ovary, it was 
necessary to assume such leaf as curled round until its lateral 
margins met, and then united into a carpel ; and if the structure of 
the ovary, based on such an hypothesis, is considered as much a fact 
as any other fact in botany, the theory of one leaf lying flat and in 
perfect apposition with another, and both then contracting union 
into one, can scarcely be held untenable. There is nothing in the 
analogy of the case to prevent such union ; and it is as much within 
the range of probability that five leaflets may coalesce into one 
polyphyllous leaf {lobed leaf), as that an ovary may be composed of 
five syncarpous pistils. 
Of Classification. 
A classification based on the prevalence of some typical form per- 
vading a whole series must needs be a desideratum. The table seen 
in Plate Y. is constructed on this principle, and scarcely requires 
explanation, saving that all reticulated leaves are thus, for the most 
