Development of the Anther id iuin in Ferns. 239 



perfectly simultaneous. By it the opercular cell is separated 

 from the liollow cylindrical enveloping cell (ring cell). 



In each of the four cells of which the antheridium is com- 

 posed in this state of development, a nucleus is distinctly re- 

 cognizable. In the opercular cell it is applied to the lower 

 septum, and is surrounded by numerous chlorophyll-grains ; 

 in the ring cell it clings to the inner Avail on one side ; in the 

 central cell it occupies exactly a middle position, and, on ac- 

 count of the abundance of chlorophyll and protoplasm, appears 

 only as a lighter spot. 



The central and ring cells grow predominantly in lengtli 

 and less in circumference. At the same time the inclination 

 of the septum which separates the latter from the opercular 

 cell becomes somewhat less. Whilst all the other cells remain 

 undivided, the central cell is broken up, by a number of suc- 

 cessive divisions, into the special mother cells of tlie spcrmu- 

 tozoids. The ])osition of the septa with regard to the longi- 

 tudinal axis of the organ and to each other is now rather 

 irregular, as may be seen from figs. 3 & 4. 



The cells of the last generation round themselves otf from 

 each other, in the manner characteristic of the special mother 

 cells, until they become completely isolated. Within the deli- 

 cate cellulose membrane there is first a layer of hyaline proto- 

 plasm ; towards the middle numerous granules are imbedded 

 in the plasma. The evacuation of the special mother cells 

 always takes place through an irregular rupture of the oper- 

 cular cell. The torn fragments of the membrane of this shrink 

 together, and soon become unrecognizable. The gradual ap- 

 pearance of the cellular contents is accompanied by a consider- 

 able extension of the basal cell and ring cell (PI. VI. fig. 6). 

 This renders it probable that the opening of the antheridium 

 is effected chiefly by the turgescence of these two cells. In 

 the membrane of the ring-cell, wliich at the same time becomes 

 much shortened, folds are formed in larger or smaller number, 

 which, when seen from above, do not usually extend beyond 

 half the thickness of the ring (fig. 7), but in a side view some- 

 times present a deceptive resemblance to true septa. I suppose 

 that these have played a great part in the erroneous repre- 

 sentations of the structure and development of the antheridium 

 of ferns. That the ring cell is not, as supposed by several of 

 the observers above mentioned, produced by the amalgamation 

 of four or more originally separate peripheral cells, but is a 

 ring cell from its first foundation, is perfectly evident from 

 the constant presence of only one nucleus. Even after evacua- 

 tion has taken place, this remains fV)r some time distinctly 

 recognizable (fig. 6). 



