FRUIT OF MELOCANNA BAMBUSOIDES. 4.09 
V. THE DEVELOPMENT OF THE FRUIT. 
When I read this paper in its original form before the Society, I remarked that I had 
not had an opportunity of examining young fruits and knew therefore nothing from direct 
observation of their stages of development prior to almost complete maturity. The 
material since received, although not so good and complete as might be wished, has filled 
the gaps sufficiently to enable me to give a more complete account of the development 
and structure of this remarkable bamboo fruit; but the early stages of the ovule still 
require close examination. ( 
The pistil (Pl. 45. figs. 8-10) of the hermaphrodite or female flowers is bottle-shaped, very 
slender and long-exserted from the spikelet, freely exposing three more or less unequal 
curved, feathery or, as Roxburgh describes them, woolly stigmas. The ovary is not dis- 
tinetly separated from the style, but passes imperceptibly into it; but if we estimate its 
length by that of the ovarial cavity, we find it to be several times shorter than the style. 
The pistil is more or less distinctly 5-grooved from the base upwards to or beyond the 
middle, the ridges separating the grooves being so placed, that one is ventral (fig. 2, 15, 
vr.), i. e. facing the palea, whilst the others are lateral, the ventral ridge being generally 
the least prominent of them. The body of the pistil consists mainly of small-celled 
parenchyma and is traversed longitudinally by 15 or more vascular (fig. 15) strands, 
the position of which is marked in cross-sections of the fruit by the presence of usually 
solitary tracheids. Their disposition is, however, apparently quite independent of the 
orientation of the ovule. 
The ovule (Pl. 45. figs. 12-15) is in the position usual in Graminew. In the mature pistil 
it is 1-1-5 mm. long and very slender, and on the whole small in comparison with the 
ovary, its diameter being only 4-1 of that of the ovary. It is attached to the pericarp 
with a very broad back, with the exception of the short, free, somewhat oblique and 
basiscope apex, and fills the ovarial cavity so completely that the latter appears in cross- 
sections made from spirit-material as a very slender crescent. The ovule-wall thins out 
towards the front, but there is no differentiation into integuments and nucellus, and 
accordingly also no micropyle. The ovules are therefore actually naked. Still there is 
this to be observed, the cells of the two outermost layers of the ovary-wall (fig. 14) are ` 
slightly larger than those of the following inner tissue, and they stain with Schultz's 
solution yellow, whilst the cell-walls of the remainder of the ovule stain blue, a reaction 
which indicates the first stage in the disorganisation which the peripheral portion of the 
ovule-wall undergoes, as I shall have to show presently. The small-celled tissue which 
surrounds the very large embryo-sac (Pl. 45. figs. 13-15, es.) in the front half of the ovule 
= continues also over the back, where it passes gradually into cells of greater diameter, 
such as build up the bulk of the pistil. How the pollen-tubes reach the egg-cell, 
I cannot say, having never seen a trace of them in the ovarial cavity, nor was there 
an indication of a stigmatic channel or a conducting-tissue. I was also unable to trace 
the early stages of the development of the embryo, as the smallest I came across was 
3 mm. long and already distinctly differentiated (Pl. 45. figs. 22, 23). 
SECOND SERIES.—BOTANY, VOL. VI. 9N 
