GRANTIA. 225 
then rises to the surface where all the subsequent changes 
are carried on. 
The seed at this stage looks asif it had a comparatively 
large fungous growth from its apex. 
When it has reached a considerable size relatively to the 
seed itself, it becomes split along its upper margin by the 
growth of the plumula, but the cap always adheres to one, 
and generally to the lower margin. 
'The plumula which is always frondiform continues to grow, 
it does not separate from the embryo for a long time. The 
embryo never loses its attachment with the seed, not indeed 
until it has nearly reached its full growth. Long before this, 
the young frond has commenced to develope another from its 
included base, and thus the plant goes or increasing. 
All the specimens I have had did not appear to produce 
more than one frond, for having completed this they be- 
came dissolved in the water in which they had previously 
grown. 
I have not been able to ascertain whether there is a slit 
for the emission of the plumula, but I think not. 
The nature of the parts is obscure, judging of them by their 
situation, the fungous growth is a growth of the radicle, and 
the included part is the cotyledon, judging from appearances, 
and analogy with the formation of the mature frond, the in- 
included part is radicle, the fungous growth, cotyledon. 
The whole evidently performs the office of the cotyledon ; 
originally it is filled with granules, which subsequently dis- 
appear as the plumule becomes more independent. 
The marked points of difference between this and Lemna 
consist in its cellularity and want of stomata, the absence of 
a spatha, the axial exsertion of the flower, the unilocularity 
of the anther, and the antitropism of the ovulum. 
In addition its pollen is smooth, the ovulum solitary, and 
the radicle has no covering, being a production of the sub- 
stance as well as of the surface. 
The species may be thus characterised. 
