CHANDLER—DEHERAINIA SMARAGDINA. 53 
These glands are sunk in the corolla lobes ; their stalks consist 
of but one cell. Inthe throat of the corolla, however, at the base 
of the petals, these glands become more numerous ; the basal 
cell divides to form a stalk of from two to three cells, and the head 
of the gland is raised some distance above the epidermis (Fig. 13a). 
Interspersed with the glands are unicellular hairs at intervals 
(Fig. 130). 
~The microscopical character of the stamen is perhaps the most 
interesting in the whole of the flower. The white mass to be seen 
on the upper side of the anther is shown under the microscope 
to be composed of particles of torn fibre, like torn, woody cells. 
The presence of this fibrous mass had already been remarked by 
Decaisne * in his description of the plant. eee 
The pollen sacs are on the under side. The cells of the anther 
are undifferentiated, except in the region round the pollen sacs 
themselves, where they are slightly thickened. Between the 
sacs is an open space, comparatively large, which is filled with 
acicular crystals in crystal dust (Fig. 140). What the purpose 
of these crystals are, it is difficult to determine. That they bear 
some relation to pollination and subsequent fertilisation is 
doubted, from their proximity to the pollen sacs. The crystals 
dissolved easily in HCl. The pollen grains are minute and round, 
and covered with a thick, brown exine, which is ‘minutely 
fringed. The cuticle of the epidermal cells of the connective is 
also fringed in the same manner—in appearance like the minute 
crystal grains on the sporangium of Mucor mucedo. The question 
of the purpose of the crystals in the anther, those on the pollen 
grain and on the cuticle of the connective, and the presence of 
the fragmentary fibrous cells on the upper side of the anther, form 
a part of the larger problem of the fertilisation of the flower. 
- The ovary in section (Figs. 15 and 16) shows a central placenta 
on which are situated the anatropous ovules. 
The seed and fruit are unknown in Europe, and wherever it 
has been naturalised. The plant is easily propagated by cuttings, 
roots being formed at the node without much callus formation. 
It has already been stated that in the young stage of the 
flower, the stamens are closely adpressed to the stigma, but at a 
later stage they fly apart, and stand with their filaments and 
anthers pressed back against the petals. The full significance 
of this fact has not yet been grasped, for nothing is known yet 
as to the fertilisation of Deherainia smaragdina. That the flower 
is insect-pollinated is evident by its foetid odour, by the presence 
of glands on the petals and the hairs lining the throat of the 
corolla. Various attempts were made to artificially fertilise the 
flower, Experiments were made by transferring the pollen grains 
+* Decaisne, in Ann, Sc. Nat., sér. 6, tom. iii. 
