IN FLOWERING PLANTS. 205- 
this, with the nature of the females, incline me to think that 
the nearest affinity is with Stilagineæ.* 
In such view it will be the connecting link between this or- 
der and Cupuliferee. If Wallich's Gymnobotrys belongs to this 
place, an additional point of resemblance is presented with Eu- 
phorbiacez by milky juice. With regard to Lindley's distinc- 
tive characters of Scepa, and Lepidostachys, the number of 
stamina is perhaps the most to be depended on, for my plant 
has certainly a placentary plate to each ovule, although there 
are only two stigmata, and in addition the ends of the ovula 
are immersed in hair. r 
I have not yet seen the perfectly mature fruit. 
All the integuments adhere somewhat together : the outer, 
Lindley's arillus, is baccate testa ; the tegmen or inner mem- 
brane, is Roxburgh's parchment-like middle membrane ; 
the inner membranaceous coat results from the remains 
of the nucleus; the embryonary sac is identified with the 
albumen. ) 
Roxburgh’s description of the seed is correct, and although 
it would be difficult to ascertain what a Botanist of the old 
school would mean by an arillus, I see nothing that should 
induce a firstrate Botanist to call the fleshy coat of this 
genus an arillus, unless it be the mention of three coats, 
a fact by no means incompatible: with the testoid nature of 
the outermost. 
NELUMBIUM. 
On certain points of structure in Nelumbium. 
The leaves of this plant are very remarkable for having the 
stomata: confined exclusively to the callous-looking oval dise 
that exists in the centre of the upper surface of the limb, 
and which is opposite to the petiole. The remainder of the 
vast limb, throughout which the green matter is deposited, is 
* The wood should be examined. 
