STIGMATA, VASCULAR FASCICLES, CARPELLARY LEAVES. 97 
to the axes of each carpellary leaf, and which is of a late 
development. 
In the earlier stages, the vascular fascicles are limited tothe 
line of union of the carpellary leaves supplying the placente, 
subsequently the axis of each leaf presents a smaller one. 
I believe myself that the fascicles of the axes of the carpel- 
lary leaves are recurrent, for I find that those of the axes 
of the placenta on arriving at the base of the stigmata become 
a good deal diminished, and that branches pass off which 
form the middle vein, the direction of the few branches of 
which, is downwards. I have not ascertained satisfactorily its 
termination, but similar peculiarities of a recurrent vein occur 
in Composite. In all cases of venation the first process is the 
lining out of the direction, this takes place some time before 
any extension of the vessels into the transparent tissue. I 
find in the young ovary that they are really recurrent veins. 
I consider this as a proof of my position, that there is 
nothing anomalous in the structure and relation of the stig- 
mata to the placente in Papaver. 
Nigella, In this the carpellary leaves are reduced almost 
to their simplest form; they are very leafy, the stigmatic 
surface is a mere humid line, the continuation of the ventral 
suture; it terminates just within the apex of the style, and 
is spiral from the twisting of the style. Pl. 34.—-Fig. 21. 
The carpellary leaf is excessively vascular ; independently 
of the main dorsal vein, the placenta is freely supplied, and 
so is the style with several, indeed many fascicles ; the pla- 
cental and dorsal fascicles are connected by the transverse 
veins. Are the placental vessels recurrent here? Why should | 
not the cellular bodies or the pistilla in this grow, they are 
like gemme ? 
"The androphore, opposite to the exsertion of the nectaria 
produces no stamina, these originate in simple oblique series 
of five or six filaments. - 
Although the nectaria are reducible barely to the type 
of a single anther, the curious disposition of the filaments 
asks, are they not compound ? I find there are 3 approxi- 
mated vascular fascicles, of which the central is the smaller. 
o 
