. OF FERNS FROM THEIR SPORES. 125 
two, al, «2; the cell b expands upwards, and by the formation of a septum inclined 
exactly in the opposite way towards the horizon, to that first formed, cuts off cell à into 
6 and c; bis then divided by a radial septum into 51 and 4 2, while c, the last new cell, 
overlaps «1 and «2; another division of c leaves c 1 and ¢2 over a1 and «a2, and carries 
up d to form d1 and d2, above b1 and b 2, until the papilla reaches its full height, and 
appears composed of tiers of four quadrant-shaped cells, the first tier consisting of al and 
a2, bl and 52 of the series; the next tier of c1 and c 2, dl and d2, &c. (figs. 58-61). 
In the meantime, the embryo-sac at the base, which very early appeared filled with 
opakish protoplasm, comes to exhibit a central globular body (figs. 56, 57 c), which in all 
. probability is an isolated mass of protoplasm, destined to secrete a membrane around it, 
and to form the germinal vesicle, just as occurs in the embryo-sae of Phanerogamia (for 
example in Orchis). The next peculiarity is the appearance of a clavate cavity running 
up the centre, between the convergent inner angles of the four series of cells of the papilla, 
forming a canal evidently communicating, through a constricted neck, with the embryo-sac 
(figs. 62-70). In this canal is observed a slender clavate filament of slightly granular 
mucilage (fig. 70) running down into a point in the embryo-sac below, where I believe it 
is in contact with the germinal vesicle: I imagine this to be a portion of the protoplasm 
of the embryo-sac which has been protruded up into the intercellular canal of the papilla, 
after the absorption of the wall of the embryo-sac, where it at first closed the bottom of 
the canal at the constricted neck. After a time the canal becomes opened to the external 
medium, and then the clavate filament is often seen hanging out from the open mouth 
(figs. 63-66); it is generally double (figs. 63, 64) at the upper end, which I cannot explain. 
It appeared firmly attached in the canal, and bore the movements of the preparation, by 
pressing and sliding over the covering glass, without becoming detached from the papilla. 
. It is at this stage of development that I believe the impregnation to take place; but 
before proceeding to speak of this part of the subject, I may conclude the description of 
the archegonia, by stating that the greater part of them are abortive; very rarely more 
than one being fertilized on a prothallium, and very often none. When thus abortive the 
cellular papillæ continue growing to some extent, the central canal becomes widely opened, 
often into a large funnel-shaped orifice, and the walls of the intercellular canal and the 
embryo-sac acquire the same deep brown colour as the interior of the effete antheridia 
(figs. 72-74). | 
IV. DEVELOPMENT OF THE EMBRYO. 
My opinion with regard to the fertilization is, that the operation is effected by the con- 
‘tact of one or more spermatozoids with the mucilaginous filament contained in or hanging 
from the mouth of the canal of the archegonium. I have seen the spermatozoids swim- 
ming in numbers around the mouth of the archegonia, but never detected one inside, and 
I do not see any good reason for supposing such a process necessary. The pollen-tube of 
flowering plants only comes in contact with the outside of the embryo-sac, and the in- 
fluence is sometimes communicated through a long suspensor ; and there does ge "ES. to 
be any sufficient objection to the supposition, that the contact of the spermatozoid with the 
filament of mucilage which lies in the canal of the archegoniwm, suffices to convey the 
VOL. XXI. | x 
