204 DEVELOPMENT OF THE OVULUM 
16. Ditto lateral. 
17. Ditto laid open along the back of a carpellum ; seed 
removed, its plate and the abortive ovulum remain- 
ing. 
18. Similar one viewed laterally, seed somewhat re- 
flexed. 
19. Outer view of seed. 
20. Ditto inner ; a. raphe. : 
21. Long section of nearly ripe seed; a. testa which 
is baceate ; b. tegmen now somewhat cartilaginous ; 
c. nucleary membrane; d. albumen developed in 
embryonary sac; e. embryo; f. raphe; g. belongs 
to the second integument which is thick at the base. 
22. Embryo. 
23. Ditto, one cotyledon with its half of the radicle. 
The process of the placenta never enlarges, it is merely 
developed to assist in fecundation, and I will venture to say, 
that if a species be found with the foramen directed inwards 
towards the placenta, that it will not exist. It is present 
before fecundation. It must be borne in mind, that a process 
of the placenta connected with fecundation is totally different 
from any process subsequently becoming arillus. And particu- 
larly so in the present instance, because if the production 
were merely an excess of growth and arilloid, it would origi- 
nate from the funicle. Dr. Lindley has not borne this in mind, 
and reasoning on the idea that it subsequently forms the | 
arillus, he states that it does not cover the foramen before im- 
pregnation. This process is to be found in the ripe seed, 
with which, however, it does not separate. 
The albumen is as usual developed in the embryonary sac, 
which is pendulous, but establishes a communication by 
approximation with the chalaza. 
The cotyledons are always parallel with the true faces of 
the seed. | 
Dr. Lindley says it is at once obvious that these albuminous 
seeds, and dehiscent four-valved, four-ovulate fruits, cannot 
be combined with Amentales. But although the structure of 
the males is so decidedly amentaceous, the habit is not so, and 
