AND POLLEN IN MELIA. 116 
Series K.—Pollen grains nearly filled with granules: these 
are developed very irregularly from the inner surface; in 
some cases half of the cavity is filled when the other half 
has only commenced developing. Is this the case in nature ? 
Series L.— Perfect pollen grains in water; they are sphe- 
_ rical, and are rendered opaque by the immense quantity of 
amylaceous granules they contain, presenting here and there 
pores, which are chiefly evident from the bulging out of the 
inner membrane, This is of extreme tenuity, and is scarcely 
demonstrable except by the production of mamilliform 
boyaux. When acted upon by Iodine it is so thin as not to 
present an uncoloured edge. The grains burst in water, 
discharging the greater part of the granules; when empty, 
they (its granules) appear flaccid cells with sulci. 
he development in this instance presents, I believe, no 
singularity ; it agrees step by step with that of Pardanthus.* 
Many points have been left undetermined, but the stamina 
are too minute to be very favourable for investigation. I 
know nothing as to how the 2 loculi are formed, nothing of 
how the 4 are formed; from some appearances, I suspect 
that the two locelli of each are continuous at the upper end. 
The development of the inner membrane has quite escap- 
ed my means of observation. 
It presents proofs of original continuity of tissue. 
It presents proofs of the separation of parts of that tissue, 
from, or in which the parent cells appear. 
It presents proofs of the production of pollen em by the 
division of the nuclei of the parent cells. 
The directions of the development are striking enough. 
- In the first place, we have a grumous mass appearing in 
a cellular one. 
Then we have a cellular mass appearing in the grumous 
one, or at its expense, in its place. 
Then each of these cellular divisions produce, or present 
a grumous mass. 
Then these grumous masses become changed into 3, 4, 
cells, of this there is no doubt. 
Journ. As. Soc. Beng. 1836, Vol. V, p. 732. ‘ 
