1883. | Development of a Dandelion Flower. 1215 
of the style. It is at this stage that all the parts begin to elongate 
rapidly, and the swelling appears at the bottom of the ovary-cell 
which is to become the ovule, and then also the calyx appears in 
the form of minute scales, which develop into the long hairs 
known as pappus. Thus the apparent sequence in the develop- 
ment of the four floral organs is corolla, androecium, gynoecium, 
calyx; but of necessity the calyx is the oldest, though the part 
called pappus is the last to appear. It was attempted in vain to 
detect in the primitive ring, or later in the wall of the ovary, any 
evidence of the blending of two or more distinct parts. No such 
indications could be found, and the inference that all four floral 
ergans are represented in the wall of this inferior ovary rests, not 
so much upon the structure of the wall as upon the order of suc- 
cession in the appearance of the floral organs. The idea that this 
primitive ring really belongs to the receptacle, and that the node, 
so-called for convenience, is in reality a node, would be tenable in 
this case but for two reasons, viz., the late appearance of the calyx 
and the fact that the corolla-lobes appear, not after the ring but 
with it, indicating that it in reality belongs to the floral organs. 
It remains yet to speak of the ovule and the support it furnishes 
to any of the existing theories. 
The ovule appears not exactly at the bottom of the ovary-cell, 
but a little to one side. By carefully tracing the fibro-vascular 
bundles, it was found that the axial bundle belonging to the pedi- 
cel of the flower ended abruptly at the bottom of the cavity of 
the ovary, sometimes rising into it as a small convexity, repre- 
senting the real punctum vegetationis of the flower bud, the check- 
ing of whose growth determined the character of an inferior 
Ovary. Just beneath this terminal point two lateral fibro-vascular 
bundles arise and run up each side of the carpellary wall. From 
one of these lateral bundles, very close to its origin, a branch 
arises which enters the funiculus (Fig. 6). 
In this Case, therefore, the fibro-vascular 
bundle which reaches the ovule is a branch 
ansing from a lateral outgrowth from the 
axial bundle. An attempt was made to 
determine whether the nucleus of the ovule 
ko a terminal or lateral growth on the 
mculus. Both Grigorieff and Sachs, in re- 
fo ches on Compositæ, and Cramer in other 
tms, claim that the nucleus is a lateral growth on the funiculus, 
