* 
.' 120 DEVELOPMENT OF THE FLOWER AND 
The ovula, when the perianth is 3 line long, are mere ex- 
tensions of the placenta of a rounded form, entirely cellular, 
and presenting no traces of coats. Pl. 40. C. Expansion—;. 
Perianth 14 line long. The funicle is tolerably long, and 
the ovula have commenced to become inverted, they are 
globular, with a conical obtuse apex, at the base of which 
there is an evident constriction. The same placenta gives 
instances of more advanced evolution—a second constriction 
but very faint, being visible near to, and below that mention- 
ed. These two constrictions mark out the future coats and 
parts of the ovula, the concealed apex becoming the nucleus, 
the part enclosed between the two ea the second 
‘coat, the remainder, the outer coat. Fig. 2. 
Perianth 12 line long. The second coat has begs: to 
extend upwards and to enclose the nucleus, but the outer 
scarcely commenced to do so. Fig. 3a. 
Perianth 2 lines long. The nucleus is now nearly } en- 
closed in the second coat, and this again is for the most part 
enclosed in the outer, but not by any upward growth of 
this, but apparently by an extension downwards of the inner 
or second coat. Fig. 4. 
erianth 3 lines long. "The nucleus is nearly enclosed, but 
the second coat is as much extended as it was before; the 
outer coat is considerably enlarged, but chiefly downwards, 
or towards its base. Fig. 6a. 
At the time of expansion, but before feecundation, the nu- . 
cleus is totally enclosed, it is ovate, of considerable size, and 
- hasa central discoloured spot, the site of the commencement ` 
of the future excavation: the second coat is lageniform. 
The mouth of the neck somewhat dilated, and projecting 
very slightly beyond the foramen. The raphe at this time 
has just entered the base of the outer coat: at 5 lines 
length of perianth it is not visible, at 8, it is nearly as 
long as the funicle. Fig. a, b, b, b, SE pansion but before 
foundation. 
I certainly agree with Mr. Brown, as to the nus out 
of the subsequent portions of the ovule by lines of constric- 
tion, and not as M. Mirbel supposes, by the growth upwards 
