20 DR. J. D. HOOKER ON WELWITSCHIA. 
as does the mixed character of the foliage of Podocarpus, of which some species have 
uninerved and others many-nerved leaves. The numerous flower-buds along the 
periphery of the crown also further favour this view. The development of the vas- 
cular tissue in the early condition of the leaf of Welwitschia may be expected to throw 
some light upon this point. 
Considering the origin of the leaf, its permanent duration, and the mode of growth of 
the trunk to which it is attached, the strictly horizontal and parallel arrangement of its 
vascular bundles ceases to be suggestive of a monocotyledonous affinity. Here the main 
increase of the trunk is horizontally outwards (centrifugal) ; and as the leaf increases in 
width by the addition of cellular tissue to either side of its base, it follows that the 
vascular bundles which originate independently in that tissue must occupy the same 
plane as those previously formed, and be carried outwards parallel to these, by the longi- 
tudinal extension of the leaf. 
The nervation of the leaf of Welwitschia appears at first sight to be singularly unsuited 
to the requirements of the plant, promoting, as it evidently does, the early splitting-up 
of the blade into innumerable narrow withered-looking thongs—except, indeed, we suppose 
that this early splitting is a special contrivance for exposing the cellular tissue surround- 
ing the vascular bundles to the action of the atmosphere when damp, and thus supple- 
menting the function of the stomata in a climate where rain is all but unknown. 
Inflorescence. 
The floriferous dise, or receptacle of the peduncles, is, as already explained, the lobed 
crown of the plant, the exact relation of which toa plumule on the one hand, and to a 
mass of axillary coalesced flower-buds on the other, is not clear, and has been already 
alluded to. 
The flower-buds are developed in deep ovoid cavities in the periphery of the crown, 
about {~§ inch deep, just above the insertion of the leaf. These cavities open by a 
vertical slit (Plate XI. fig. 1), with prominent lips, and disclose an ovoid bud, enclosed in 
the insertion of the leaf,—a very remarkable fact, analogous to the occurrence of shoots 
proceeding from the caulicle of various plants, and described by Bernardi in Linaria 
arenaria (Linnea, vii. p. 561, tab. xiv. fig. 1),# : | 
When fully developed, each branch of the inflorescence is essentially a dichotomous 
cyme, with persistent opposite bracts at the nodes ; this cyme is, however, much inter- 
rupted by the irregular stippression of some of the internodes, so that the nodes often 
* I am indebted to my friend Dr. Maxwell Masters for indicating this analogy, and for the reference to it. The 
same botanist also informs me that he has observed similar shoot: : . - 
S the caulicl E rbi 
on that of Anagallis arvensis. Son the eaulicle of Euphorbia Peplus, and frequently 
