4 Mr. GnirrrrR on the Development of the Ovulum in Avicennia. 
have called the short central prolongation. But what has been recorded of 
Santalacee (and the whole of my observations on Avicennia) is opposed to 
this; for in all the instances observed, the posterior prolongation is a pro- 
longation of the posterior end of the sac itself, which obviously would not be 
the case if the ordinary relations of embryo-sacs to their nuclei existed in 
Avicennia. 
Another non-analogous instance may be observed in the gradual protrusion 
outwards of the young albumen, which is assumable as being at one period 
entirely interior to the nucleus or ovulum. In all the really analogous in- 
stances in which the albumen is exterior to the ovulum, it is always exterior, 
that part of the embryo-sac in which it is developed being protruded long 
before any albuminous tissue has been developed, which indeed is almost 
always subsequent to fecundation properly speaking, viz. the completion of 
certain relations between the anterior end of the pollen-tube and the embryo- 
sac. 
A third non-analogous instance seems to me presented by the exsertion or 
protrusion of the cotyledons. Protrusion of the radicular end of the embryo 
is not, perhaps, uncommon ; but in these cases it may be difficult to ascertain 
to what extent the protrusion may be due to germination. 
In Cryptocoryne ciliata (Ambrosinia ciliata, Roxb.) however the protrusion 
takes place long before the cotyledon has acquired its full growth, up to 
which period moreover it retains its firm fleshy substance. In a Malacca 
subgeneric form of Cryptocoryne, in which the margins of the spatha cohere 
into a tube to a great extent, although the plumula is still of considerable size, 
no protrusion whatever takes place. By the peculiar way in which this is per- 
formed the embryo becomes almost entirely naked, without however changing 
the direction it would have had, had it been developed, as it so generally is, 
within the body of the seed. It is curious that the obliquity in the direction 
takes place at a very 
ne of the axis of the 
