DEVELOPMENT OF THE FLOWER—ADDITIONAL NOTE. 251 
In true bicarpellary structures the undulations are only 
two, or at any rate two are subordinate; and there are only 
two lines of confluence. 
If Cruciferæ were bicarpellary, the lines would be either 
atc.oratd,d. 
a. Inner Sepal. 
b. Outer Sepal. 
J.—Represents a very early stage; the sepals are complete 
as to form and imbrication. — 
g.—The petals are a posterior development, and are ex- 
tremely small when the anthers are perfectly marked out at 
a stage a little antecedent to that of figure 4, and then also 
the four lines of confluence reach to.the bottom nearly of the 
ovary, and its quaternary appearance is even more distinct : in 
this instance, the flat face of the ovary was opposite the outer 
sepals, so it was in another instance, and so it is always? 
À.—In this instance there did not appear to be any petals ; 
the staminal lobes seemed to be two outer, and four inner ; 
and the ovary was represented by an elevation with an entire 
ime flattened in the direction of the outer sepals. Outer 
sepal, A. 
indat ;,th, the ovarial apex presents traces of its lobes ; 
it is Ae and fungiform. 
i, i,— Very early, sepals open : centre canine of a convex, 
rugose, undivided, cellular disc. 
The first change in this occurs in the marking out of 
the stamina, but I do not know whether there is any differ- 
ence in the period or degree. | 
k.—Stigma much more developed. "The upper ovula are 
most developed, and the lower are without the annulus ; the 
upper with it, and beginning to turn on themselves. 
Vessels four, about equal, if bicarpellary, these should be 
six. 
This will however, require strict comparison with numerous 
other instances ; it is not the first time in which I have seen 
the rim of the ovarial orifice entire. Compare also the size 
of the vessels, and compare in every respect. 
