586 CUCURBITACE.E, 
cite our surprise, although the pendulousness of the ovula 
from the apex of the ovarium, should. This therefore must 
be introduced into the character of Actinostemma. The 
lateral lobes of the placentze are alone ovuliferous, the ovula 
are fixed to their outer surface as regards the carpellary leaf, 
to the inner as regards the placenta: they are immersed. 
This is their normal situation, so that judging from the early 
state of the ovaria, there is nothing anomalous in their situa- 
tion. The articulation of the pedicels is not a constant 
character. 
The completion of the raphe appears a. general character, 
and this as in some others certainly does not appear to run 
in the testa or outer integument. 
In one instance an exceedingly fine membrane is inter- 
posed between the nucleus and inner integument; of the 
origin of this I am ignorant, not having had an opportunity 
of examining the ovula until sometime after fecundation. 
In this too, the tissue immediately surrounding each ovule, 
is separable with the ovulum, and might be mistaken for an 
additional integument. 
This arilliform integument completely envelopes the seeds, 
it is perforate alone at its base for the passage of the vessels 
of the funicle. 
In Coccinia there is no distinction of coats, nothing to in- 
dicate any distinction of parts in the structure of the tube of 
calyx, all the vessels too are on the same plane; at a very 
early age the cavity of the ovary will be found occupied by 
three wedge-shaped placenta, which even now have some- 
what the appearance of inverted carpellary leaves. These 
placenta correspond to as many stigmata, which are quite 
distinct from each other, and also have an appearance as if 
slighty folded outwards. : 
Yet there are grave objections to the adoption of this 
opinion, in the first place, it implies a simplicity of placenta 
in a compound ovarium, and it requires that the spaces inter- 
posed between the placentz, constituting the cavities 
a la T TETN 
