DEFORMED GOODYERA. 131 
points and parts. No change likewise occurred in the 
ovarium. 
The column was remarkably altered, the whole of its 
posticous (or in the bud anticous) face being completely 
distinct, and existing in the form of a flat filament. The 
anther preserved its usual form. The stigma was complete- 
ly reversed in situation, and occupied the inner surface of 
the cavity, in which the cells of the anther are- immersed. 
It was completely ringent. From the base of this a very 
distinct stigmatic canal led into the ovarium. 
The anterior face of the column was furnished with the 
usual processes, but there was no attempt at the formation 
of a rostellum, usually so very much developed. 
Corresponding to this, there was no attempt at the forma- 
tion of a gland. The pollinia had pretty nearly the real 
form, but the transition from the scale-like masses, all arched 
to the caudicula (?) throughout its whole extent, was per- 
haps more gradual 
This instance is valuable, as indicating the extent and 
site of the filament, and as proving that the gland is essenti- 
ally a portion of, or production from, the stigma; perhaps 
limited to the existence of a rostellum. I take this to be 
quite corroborative of Mr. Brown's ideas on this point. 
But the most remarkable circumstance perhaps is the 
transposition of parts to adapt them to a certain end. 
And this is effected by a process apparently much more 
simple and direct than that usually resorted to. For here 
we see that the most decidedly pollen-looking part of the 
pollinizm, their bases, are separated from the stigma by the 
corresponding part of the anticous faces of the cells of 
the anther only, and hence mere dehiscence is quite suffi- 
cient, or would appear to be, for securing impregnation. In 
the other and usual form, not only is the actual removal of 
the pollinia. from the cells of the anther necessary, but the 
still greater obstacle exists in the extreme smallness of 
surface presented by the stigma, and which probably de- 
pends upon the intimate cohesion of part of the base of the 
labellum with it; and if this view be correct, the two pro- 
